21st Century Astronomy PDF

This page intentionally left blank
21st Century Astronomy’s text, art, and

media work together to help students see
the universe through the eyes of a scientist.
Both text and

narrative art help
students to
understand physical
concepts . . .
AstroTour animations help

. . . while AstroTours
students visualize physical
use animation and
processes. interactivity to help
students visualize key
processes . . .
StudySpace is a free and

open website that provides
study plans that tie together ani-
mations, interactive simulations,
diagnostic quizzes, vocabulary
flash cards, news feeds, and more!
Visit www.wwnorton.com/
studyspace. . . . and SmartWork
online homework
extends the text’s
emphasis on
SmartWork online home- reasoning and
work provides students critical thinking.
with feedback right when they
need it, and it offers links to the
ebook version of the text. To log in
or purchase a registration code,
visit smartwork.wwnorton.com.
THIRD EDITION
21 ST CENTURY
ASTRONOMY
THIRD EDITION
21 ST CENTURY
ASTRONOMY
Jeff Hester • A R I Z O N A S TAT E U N I V E R S I T Y
Brad Smith • U N I V E R S I T Y O F H AWA I I — M A N O A
George Blumenthal • U N I V E R S I T Y O F C A L I F O R N I A — S A N TA C R U Z
Laura Kay • B A R N A R D COLLEGE
Howard Voss • A R I Z O N A S TAT E U N I V E R S I T Y
B W. W. N O R T O N & C O M PA N Y • N E W Y O R K • L O N D O N
W. W. Norton & Company has been independent since its founding in 1923, when William Warder
Norton and Mary D. Herter Norton first published lectures delivered at the People’s Institute, the
adult education division of New York City’s Cooper Union. The fi rm soon expanded its program
beyond the Institute, publishing books by celebrated academics from America and abroad. By
midcentury, the two major pillars of Norton’s publishing program—trade books and college
texts—were fi rmly established. In the 1950s, the Norton family transferred control of the company
to its employees, and today—with a staff of 400 and a comparable number of trade, college, and
professional titles published each year—W. W. Norton & Company stands as the largest and oldest
publishing house owned wholly by its employees.
Copyright © 2010 by Jeff Hester, Bradford Smith, George Blumenthal, Laura Kay, and Howard G. Voss.
Copyright © 2007 by Jeff Hester, David Burstein, George Blumenthal, Ronald Greeley, Bradford Smith,
and Howard G. Voss. Copyright © 2002 by Jeff Hester, David Burstein, George Blumenthal, Ronald
Greeley, Bradford Smith, Howard G. Voss, and Gary Wegner.
Since this page cannot accommodate all the copyright notices, the Credits pages, starting on p. C1,
constitute an extension of the copyright page.
All rights reserved

Printed in the United States of America
Third Edition
Editor: Rob Bellinger

Senior Editor, Physical Science: Erik Fahlgren
Editorial Assistant: Mary Lynch
Managing Editor, College: Marian Johnson
Associate Managing Editor, College: Kim Yi
Copy Editor: Stephanie Hiebert
Marketing Manager: Kelsey Volker
Developmental Editors: Patricia Longoria, Andrew Sobel
Science Media Editor: Rob Bellinger
Assistant Editor, Electronic Media: Laura Musich
Supplements Editor: Matthew A. Freeman
Senior Production Manager: Jane Searle
Art Director and Designer: Rubina Yeh
Photo Researchers: Junenoire Mitchell, Ramón Rivera Moret
Illustrations: Penumbra Design
Layout: Carole Desnoes
Compositor: Progressive Information Technologies
Manufacturing: R. R. Donnelley & Sons—Willard, Ohio
Library of Congress Cataloging-in-Publication Data

21st century astronomy / Jeff Hester . . . [et al.].—3rd ed.
p. cm.
Includes index.
ISBN: 978-0-393-93198-3 (pbk.)
1. Astronomy. I. Hester, John Jeffrey. I I. Title: Twenty-first century astronomy.
QB45.2.A14 2010
520—dc22 2009045134
ISBN: 978-0-393-11664-9 (pdf.)
W. W. Norton & Company, Inc., 500 Fifth Avenue, New York, N.Y. 10110
www.wwnorton.com
W. W. Norton & Company Ltd., Castle House, 75/76 Wells Street, London W1T 3QT
1 2 3 4 5 6 7 89 0
George Blumenthal gratefully thanks his wife, Kelly Weisberg, and
his children, Aaron and Sarah Blumenthal, for their support during
this project.
Jeff Hester thanks the students over the years who allowed him
the pleasure and the privilege of sharing with them something of
the extraordinary universe in which we live. He also acknowledges
his wife Vicki, his family, and the friends who offered much needed
support through trying times.
Laura Kay thanks her partner, M.P.M.
Brad Smith dedicates this Third Edition to his patient and

understanding wife, Diane McGregor.
Howard Voss dedicates this book to his wife, Helen Ann, who
cheerfully sacrificed much so that he could attend to his teaching
and other professional work. Helen Ann passed away at the time of
completion of the Second Edition. She is greatly missed by Howard,
their family, and her many friends.
Brief Contents
PART I Introduction to Astronomy

Chapter 1 Why Learn Astronomy?
Chapter 2 Patterns in the Sky—Motions of Earth
Chapter 3 Gravity and Orbits—A Celestial Ballet
Chapter 4 Light
Chapter 5 The Tools of the Astronomer
PART II The Solar System

Chapter 6 The Birth and Evolution of Planetary Systems
Chapter 7 The Terrestrial Planets and Earth’s Moon
Chapter 8 Atmospheres of the Terrestrial Planets
Chapter 9 Worlds of Gas and Liquid—The Giant Planets
Chapter 10 Gravity Is More than Kepler’s Laws
Chapter 11 Planetary Adornments—Moons and Rings
Chapter 12 Dwarf Planets and Small Solar System Bodies
PART III Stars and Stellar Evolution

Chapter 13 Taking the Measure of Stars
Chapter 14 A Run-of-the-Mill G-Type Star—Our Sun
Chapter 15 Star Formation and the Interstellar Medium
Chapter 16 Stars in the Slow Lane—Low-Mass Stellar Evolution
Chapter 17 Live Fast, Die Young—High-Mass Stellar Evolution
PART IV Galaxies, the Universe, and Cosmology

Chapter 18 Our Expanding Universe
Chapter 19 Galaxies
Chapter 20 The Milky Way—A Normal Spiral Galaxy
Chapter 21 Modern Cosmology
Chapter 22 The Origin of Structure
Chapter 23 Life
EPILOGUE The Universe and You

viii
Contents
Preface xxiii
About the Authors xxxv
PART I Introduction to Astronomy
CHAPTER 1 Why Learn Astronomy?

1.1 Getting a Feel for the Neighborhood 3
Glimpsing Our Place in the Universe 3
The Scale of the Universe 5
The Origin and Evolution of the Universe 5
1.2 Science Involves Exploration and Discovery 7
1.3 Science Is a Way of Viewing the World 9

The Scientific Method and Scientific Principles 9
Science as a Way of Knowing 10
Scientific Revolutions 10
Challenges to Science 10
1.4 Patterns Make Our Lives and Science Possible 14
Math Tools 1.1 Mathematical Tools 16
1.5 Thinking like an Astronomer 17
Seeing the Forest for the Trees 19

Summary 19
Student Questions 19
CHAPTER 2 Patterns in the Sky—Motions of Earth

2.1 A View from Long Ago 23
2.2 Earth Spins on Its Axis 25

The Celestial Sphere Is a Useful Fiction 26
The View from the Poles 27
Away from the Poles, the Part of the Sky We See
Is Constantly Changing 28
2.3 Revolution about the Sun Leads to Changes during the Year 32
ix
x Contents
Math Tools 2.1 How to Estimate the Size of Earth 33
Tiny Deviations in the Direction of Starlight Provide a Measure

of Earth’s Motion through Space 33
Seasons Are Due to the Tilt of Earth’s Axis 36
Four Special Days Mark the Passage of the Seasons 39
Earth’s Axis Wobbles, and the Seasons Shift through the Year 41
Excursions 2.1 Why Is It Surprising That A.D. 2000 Was a Leap Year? 42
2.4 The Motions and Phases of the Moon 42

We Always See the Same Face of the Moon 43
The Changing Phases of the Moon 43
2.5 Eclipses: Passing through a Shadow 46

Varieties of Eclipses 46
Eclipse Seasons Occur Roughly Twice Every 11 Months 50

Summary 53
CHAPTER 3 Gravity and Orbits—A Celestial Ballet

3.1 Gravity! 57
3.2 An Empirical Beginning: Kepler Describes the Observed Motions

of the Planets 59
Excursions 3.1 The Motions of the Planets in the Sky 60
Kepler’s First Law: Planets Move on Elliptical Orbits

with the Sun at One Focus 62
Kepler’s Second Law: Planets Sweep Out Equal Areas
in Equal Times 63
Kepler’s Third Law: Planetary Orbits Reveal a Harmony
of the Worlds 64
3.3 The Rise of Scientific Theory: Newton’s Laws Govern

the Motion of All Objects 65
Galileo Describes Objects in Motion 66
Newton’s First Law: Objects at Rest Stay at Rest;
Objects in Motion Stay in Motion 67
Newton’s Second Law: Motion Is Changed
by Unbalanced Forces 68
Newton’s Third Law: Whatever Is Pushed, Pushes Back 70
Foundations 3.1 Proportionality 71
3.4 Gravity Is a Force between Any Two Objects

Due to Their Masses 72
Where Do Theories Come From? Newton Reasons His Way
to a Law of Gravity 73
Connections 3.1 Inverse Square Laws 76
Putting the Pieces Together: A Universal Law for Gravitation 76
Math Tools 3.1 Playing with Newton’s Laws of Motion and Gravitation 78
3.5 Orbits Are One Body “Falling around” Another 79

Newton Fires a Shot around the World 79
How Fast Must Newton’s Cannonball Fly? 80
Contents xi
Planets Are Just like Newton’s Cannonball 81
Math Tools 3.2 Achieving Circular Velocity 82
Most Orbits in the Real World Are Not Circles 82
Math Tools 3.3 Calculating Escape Velocities 84
Newton’s Theory Is a Powerful Tool for Measuring Mass 85

Summary 87
CHAPTER 4 Light
4.1 Let There Be Light 91
4.2 How Our Understanding of Light Evolved 92

Light Is an Electromagnetic Wave 93
Waves Are Characterized by Wavelength, Speed, Frequency,
and Amplitude 96
Math Tools 4.1 Working with Electromagnetic Radiation 97
A Wide Range of Wavelengths Makes Up the Electromagnetic

Spectrum 97
Light Is a Wave, but It Is Also a Particle 99
4.3 The Quantum View of Matter 100

Atoms Have Discrete Energy Levels 101
The Energy Levels of an Atom Determine the Wavelengths of Light
That It Can Emit and Absorb 102
Emission and Absorption Lines Are the Spectral Fingerprints
of Atoms 106
How Are Atoms Excited, and Why Do They Decay? 107
4.4 The Doppler Effect—Is It Moving toward Us or Away from

Us? 108
Math Tools 4.2 Making Use of the Doppler Effect 109
Foundations 4.1 Equilibrium Means Balance 110
4.5 Why Mercury Is Hot and Pluto Is Not 111

Temperature Is a Measure of How Energetically Particles
Are Moving About 112
Hotter Means More Luminous and Bluer 113
The Stefan-Boltzmann Law Says That Hotter Means Much More
Luminous 115
Math Tools 4.3 Working with the Stefan-Boltzmann and Wien’s Laws 115
Wien’s Law Says That Hotter Means Bluer 116
4.6 Twice as Far Means One-Fourth as Bright 116
4.7 Radiation Laws Enable Us to Calculate the Equilibrium

Temperatures of the Planets 117

Summary 121
xii Contents
CHAPTER 5 The Tools of the Astronomer

5.1 The Optical Telescope—An Extension of Our Eyes 125
Refractors and Reflectors 129
Foundations 5.1 When Light Doesn’t Go Straight 130
Optical and Atmospheric Limitations 134
Math Tools 5.1 Diffraction Limit 134
5.2 Optical Detectors and Instruments 137

Photographic Plates 138
Charge-Coupled Devices 139
Spectrographs 141
5.3 Radio Telescopes 141
5.4 Getting above Earth’s Atmosphere:

Airborne and Orbiting Observatories 144
5.5 Getting Up Close with Planetary Spacecraft 144

Flybys and Orbiters 145
Landers, Rovers, and Atmospheric Probes 146
Sample Returns 147
5.6 Other Earth-Based Astronomical Tools 147

Summary 151
PART II The Solar System

CHAPTER 6 The Birth and Evolution of Planetary Systems
6.1 Stars Form and Planets Are Born 155
6.2 In the Beginning Was a Disk 158

The Collapsing Cloud Rotates 158
Math Tools 6.1 Angular Momentum 160
An Accretion Disk Forms 160

Small Objects Stick Together to Become Large Objects 162
6.3 The Inner Disk Is Hot, but the Outer Disk Is Cold 164
Foundations 6.1 Thinking about Energy in Different Ways 165
Rock, Metal, and Ice 166

Solid Planets Gather Atmospheres 167
6.4 A Tale of Eight Planets 167
6.5 There Is Nothing Special about Our Solar System 169

The Search for Extrasolar Planets 170
Planetary Systems Seem to Be Commonplace 172

Summary 176
Contents xiii
CHAPTER 7 The Terrestrial Planets and Earth’s Moon

7.1 How Are Planets the Same, and How Are They Different? 179
7.2 Four Main Processes Shape Our Planet 180
7.3 Impacts Help Shape the Evolution of the Planets 182
Excursions 7.1 Where Have All the Dinosaurs Gone? 184
Calibrating a Cosmic Clock 186
Foundations 7.1 Determining the Ages of Rocks 187
Math Tools 7.1 How to Read Cosmic Clocks 187
7.4 The Interiors of the Terrestrial Planets Tell Their Own Tales 188
We Can Probe the Interior of Earth 188
Building a Model of Earth’s Interior 191
The Moon Was Born from Earth 191
The Evolution of Planetary Interiors Depends on Heating and
Cooling 192
Math Tools 7.2 How Planets Cool Off 194
Most Planets Generate Their Own Magnetic Fields 194
7.5 Tectonism—How Planetary Surfaces Evolve 196

Continents Drift Apart and Come Back Together 196
Plate Tectonics Is Driven by Convection 197
Evidence of Tectonism on Other Planets 200
7.6 Igneous Activity: A Sign of a Geologically Active Planet 202

Terrestrial Volcanism Is Related to Tectonism 202
Volcanism Also Occurs Elsewhere in the Solar System 203
7.7 Erosion: Wearing Down the High Spots and Filling In the
Low 206

Summary 212
CHAPTER 8 Atmospheres of the Terrestrial Planets

8.1 Atmospheres Are Oceans of Air 215
Some Atmospheres Appeared Very Early 215
Some Atmospheres Developed Later 216
8.2 A Tale of Three Planets—The Evolution of Secondary

Atmospheres 218
Similarities and Differences 218
How Mass Affects a Planet’s Atmosphere 218
The Atmospheric Greenhouse Effect 219
8.3 Earth’s Atmosphere—The One We Know Best 221
Foundations 8.1 What Is a Gas? 222
Life Controls the Composition of Earth’s Atmosphere 222
Excursions 8.1 On Atmospheres and Life 224
Earth’s Atmosphere Is Layered like an Onion 225

Why the Winds Blow 229
xiv Contents
Foundations 8.2 The Coriolis Effect 231
When Convection Runs Amok 232

Is Earth Getting Warmer? 237
8.4 Venus Has a Hot, Dense Atmosphere 238
8.5 Mars Has a Cold, Thin Atmosphere 239
Excursions 8.2 Blue Skies, White Clouds, and Red Sunsets 240
8.6 Mercury and the Moon Have Almost No Atmosphere 242

Summary 244
CHAPTER 9 Worlds of Gas and Liquid—The Giant Planets

9.1 The Giant Planets—Distant Worlds, Different Worlds 247
Two Were Known to Antiquity, and Two Were Telescopic
Discoveries 249
9.2 How Giant Planets Differ from Terrestrial Planets 250

The Giant Planets Are Large and Massive 250
We See Only Atmospheres, Not Surfaces 252
Jupiter’s Chemical Composition Is More like the Sun’s
Than Earth’s Is 252
Short Days and Nights on the Giant Planets 252
9.3 A View of the Cloud Tops 254
9.4 A Journey into the Clouds 257
9.5 Winds and Storms—Violent Weather on the Giant Planets 259
Math Tools 9.1 How Wind Speeds on Distant Planets Are Measured 260
9.6 Internal Thermal Energy Drives Weather Patterns 262
Excursions 9.1 Jupiter—A Star That Failed? 263
9.7 The Interiors of the Giant Planets Are Hot and Dense 263
Uranus and Neptune Are Different from Jupiter and Saturn 265
Excursions 9.2 Strange Behavior in the Realm of Ultrahigh Pressure 265

Differences Are Clues to Origins 266
9.8 The Giant Planets Are Magnetic Powerhouses 266

Giant Planets Have Giant Magnetospheres 266
Magnetospheres Produce Synchrotron Radiation 268
Radiation Belts and Auroras 268
Foundations 9.1 Synchrotron Radiation—From Planets to Quasars 269

Summary 272
Contents xv
CHAPTER 10 Gravity Is More than Kepler’s Laws

10.1 Gravity Once Again 275
10.2 Gravity Differs from Place to Place within an Object 276
Foundations 10.1 The Power of Symmetry 278
10.3 Tides Are Due to Differences in the Gravitational Pull from

External Objects 279
10.4 Tidal Effects on Solid Bodies 282
Connections 10.1 Lunar Tides and Life 283
The Moon Is Tidally Locked to Earth 284
Connections 10.2 Earth, the Moon, and Conservation of Angular Momentum 285
Tides and Noncircular Orbits: An Imperfect Match 285

Tides Can Be Destructive 287
10.5 More Than Two Objects Can Join the Dance 287
Connections 10.3 Tides on Many Scales 288
Gaps in the Asteroid Belt 289

Gravity and Ring Systems 290
Gravity Affects the Orbits of Three Bodies 292
10.6 Such Wondrous Complexity Comes from Such a Simple

Force 293
Excursions 10.1 A World of Chaos 294

Summary 296
CHAPTER 11 Planetary Adornments—Moons and Rings

11.1 Galileo’s Moons 299
11.2 Moons as Small Worlds 302

Geologically Active Moons: Io, Enceladus, and Triton 302
Possibly Active Moons: Europa and Titan 306
Connections 11.1 Extreme Environments and an Organic Deep Freeze 307
Formerly Active Moons: Ganymede and Some Moons of Saturn and

Uranus 310
Geologically Dead Moons 312
11.3 Rings Surround the Giant Planets 313

Saturn’s Magnificent Rings—A Closer Look 314
Other Planets, Other Rings 316
Excursions 11.1 The Backlighting Phenomenon 317
The Composition of Ring Material 319

Moons Maintain Order and Create Gaps 320
Strange Things among the Rings 320
Planetary Rings Are Ephemeral 322

Summary 324
xvi Contents
CHAPTER 12 Dwarf Planets and Small Solar System Bodies

12.1 Leftover Material: From the Small to the Tiniest 327
12.2 Dwarf Planets: Pluto and Others 328
12.3 Ghostly Apparitions and Rocks Falling from the Sky 331
12.4 Asteroids and Comets: Pieces of the Past 332
12.5 Meteorites: A Chip Off the Old Asteroid Block 332
Excursions 12.1 Meteors, Meteoroids, Meteorites, and Comets 333
12.6 Asteroids Are Fractured Rock 336
12.7 Asteroids Viewed Up Close 337
12.8 The Comets: Clumps of Ice 339

The Abode of the Comets 340
Anatomy of an Active Comet 341
A Visit to Comet Halley 343
Visits to Comets Borrelly, Wild 2, and Tempel 1 345
Comets Have Two Types of Tails 346
The Orbits of Comets 346
12.9 Collisions Still Happen Today 349
Excursions 12.2 Comet Shoemaker-Levy 9 350
Connections 12.1 Comets, Asteroids, and Life 352
12.10 Solar System Debris 353

Meteoroids and Meteors 353
Meteor Showers and Comets 354
Zodiacal Dust 355

Summary 358
PART III Stars and Stellar Evolution
CHAPTER 13 Taking the Measure of Stars

13.1 Twinkle, Twinkle, Little Star, How I Wonder What You Are 361
13.2 The First Step Is Measuring the Distance, Brightness,

and Luminosity of Stars 362
We Use Parallax to Measure Distances to Nearby Stars 363
Math Tools 13.1 The Magnitude System 365
Once We Know Distance and Brightness, We Can Calculate

Luminosity 366
13.3 Radiation Tells Us the Temperature, Size, and Composition

of Stars 367
Wien’s Law Revisited: The Color and Surface Temperature
of Stars 367
Contents xvii
Stars Are Classified According to Their Surface Temperature 368

Stars Consist Mostly of Hydrogen and Helium 370
The Stefan-Boltzmann Law Revisited: Finding the Sizes
of Stars 371
13.4 We Measure Stellar Masses by Analyzing Binary Star

Orbits 373
Binary Stars Orbit a Common Center of Mass 373
Kepler’s Third Law Gives the Total Mass of a Binary System 375
13.5 The H-R Diagram Is the Key to Understanding Stars 376
Math Tools 13.2 Using a Celestial Bathroom Scale to Measure the Masses
of an Eclipsing Binary Pair 377
The H-R Diagram Is Worth Studying 378
The Main Sequence Is a Grand Pattern in Stellar Properties 380
Not All Stars Are Main-Sequence Stars 383

Summary 385
CHAPTER 14 A Run-of-the-Mill G-Type Star—Our Sun

14.1 The Sun Is More Than Just a Light in the Sky 389
14.2 The Structure of the Sun Is a Matter of Balance 390

The Sun Is Powered by Nuclear Fusion 391
Math Tools 14.1 The Source of the Sun’s Energy 392
Energy Produced in the Sun’s Core Must Find Its Way

to the Surface 393
Foundations 14.1 The Proton-Proton Chain 394
What If the Sun Were Different? 396
14.3 The Standard Model of the Sun Is Well Tested 398

We Use Neutrinos to Observe the Heart of the Sun 398
Helioseismology Enables Us to Observe Echoes
from the Sun’s Interior 399
Excursions 14.1 Neutrino Astronomy 400
14.4 The Sun Can Be Studied Up Close and Personal 402

The Solar Spectrum Is Complex 402
The Sun’s Outer Atmosphere: Chromosphere and Corona 404
Solar Activity Is Caused by Magnetic Effects 405
Math Tools 14.2 Sunspots and Temperature 407
Solar Activity Affects Earth 411

Summary 413
xviii Contents
CHAPTER 15 Star Formation and the Interstellar Medium

15.1 Whence Stars? 417
15.2 The Interstellar Medium 418

The Interstellar Medium Is Dusty 418
Interstellar Gas Has Different Temperatures and Densities 421
Regions of Cool, Dense Gas Are Called Clouds 424
15.3 Molecular Clouds Are the Cradles of Star Formation 426

Molecular Clouds Fragment As They Collapse 427
15.4 The Protostar Becomes a Star 428

A Shifting Balance: The Evolving Protostar 429
Evolving Stars and Protostars Follow “Evolutionary Tracks”
on the H-R Diagram 433
Not All Stars Are Created Equal 435

Summary 439
CHAPTER 16 Stars in the Slow Lane—

Low-Mass Stellar Evolution
16.1 This, Too, Shall Pass Away 443
16.2 The Life and Times of a Main-Sequence Star 444

Higher Mass Means a Shorter Lifetime 445
The Structure of a Star Changes As It Uses Its Fuel 445
Math Tools 16.1 Estimating Main-Sequence Lifetimes 446
Helium Ash Builds Up in the Center of the Star 447
16.3 A Star Runs Out of Hydrogen and Leaves the Main Sequence 447
The Helium Core Is Degenerate 448
Hydrogen Burns in a Shell Surrounding a Core of Helium Ash 448
Tracking the Evolution of the Star on the H-R Diagram 449
16.4 Helium Begins to Burn in the Degenerate Core 452

The Atomic Nuclei in the Core Form a “Gas within a Gas” 452
Helium Burning and the Triple-Alpha Process 452
The Helium Core Ignites in a Helium Flash 453
16.5 The Low-Mass Star Enters the Last Stages

of Its Evolution 455
The Star Moves Up the Asymptotic Giant Branch 455
Giant Stars Lose Mass 455
The Post-AGB Star May Cause a Planetary Nebula to Glow 456
Connections 16.1 Seeding the Universe with New Chemical Elements 459
The Star Becomes a White Dwarf 459
16.6 Many Stars Evolve as Pairs 461

Mass Flows from an Evolving Star onto Its Companion 461
Evolution of a Binary System 461
Fireworks Occur When the Second Star Evolves 462
A Stellar Cataclysm May Await the White Dwarf in a Binary
System 464

Contents xix
Summary 465
CHAPTER 17 Live Fast, Die Young—

High-Mass Stellar Evolution
17.1 Here Be Dragons 469
17.2 High-Mass Stars Follow Their Own Path 470

The High-Mass Star Leaves the Main Sequence 471
Excursions 17.1 Each Star Is Unique 472
Not All Stars Are Stable 473
17.3 High-Mass Stars Go Out with a Bang 475

The Final Days in the Life of a Massive Star 476
The Core Collapses and the Star Explodes 477
17.4 The Spectacle and Legacy of Supernovae 479

The Energetic and Chemical Legacy of Supernovae 479
Neutron Stars and Pulsars 481
The Crab Nebula—Remains of a Stellar Cataclysm 482
17.5 Star Clusters Are Snapshots of Stellar Evolution 484
17.6 Beyond Newtonian Physics 487

The Speed of Light Is a Very Special Value 488
Time Is a Relative Thing 490
The Implications of Relativity Are Far-Ranging 491
17.7 Gravity Is a Distortion of Spacetime 494

Free Fall Is the Same as Free Float 494
Spacetime as a Rubber Sheet 495
The Observable Consequences of General Relativity 497
Connections 17.1 When One Physical Law Supplants Another 498
Connections 17.2 General Relativity and the Structure of the Universe 501
17.8 Back to Black Holes 502

“Seeing” Black Holes 503

Summary 505
PART IV Galaxies, the Universe,

and Cosmology
CHAPTER 18 Our Expanding Universe
18.1 Twentieth Century Astronomers Discovered the Universe
of Galaxies 509
18.2 The Cosmological Principle Shapes Our View of the Universe 511
18.3 We Live in an Expanding Universe 513

All Observers See the Same Hubble Expansion 514
xx Contents
Math Tools 18.1 Redshift: Calculating the Recession Velocity and Distance
of Galaxies 515
We Must Build a Distance Ladder to Measure

the Hubble Constant 515
Hubble’s Law Maps the Universe in Space and Time 518
18.4 The Universe Began in the Big Bang 518

Galaxies Are Not Flying Apart through Space 519
Math Tools 18.2 Expansion and the Age of the Universe 520
Expansion Is Described with a Scale Factor 522

Redshift Is Due to the Changing Scale Factor of the Universe 523
Foundations 18.1 When Redshift Exceeds 1 524
18.5 The Major Predictions of the Big Bang Theory

Are Resoundingly Confirmed 525
We See Radiation Left Over from the Big Bang 525
Satellite Data Confirmed That the Cosmic
Background Radiation Is Real 528
The CBR Measures Earth’s Motion Relative to
the Universe Itself 528
The Big Bang Theory Correctly Predicts the Abundance
of the Least Massive Elements 529

Summary 531
CHAPTER 19 Galaxies
19.1 Galaxies Come in Many Flavors 535
Galaxies Are Classified According to Their Appearance 535
Stellar Motions Give Galaxies Their Shapes 537
Other Differences among Galaxies 539
19.2 Stars Form in the Spiral Arms of a Galaxy’s Disk 542
19.3 Galaxies Are Mostly Dark Matter 546

19.4 There Is a Beast at the Heart of Most Galaxies 549
AGNs Are the Size of the Solar System 551
Supermassive Black Holes and Accretion Disks Run Amok 552
What AGN We See Depends on Our Perspective 554
Normal Galaxies and AGNs—A Question of Feeding the Beast 556
Mergers and Interactions Make the Difference 557

Summary 559
CHAPTER 20 The Milky Way—A Normal Spiral Galaxy

20.1 We Look Up and See Our Galaxy 563
20.2 Measuring the Milky Way 565

Globular Clusters and the Size of the Milky Way Galaxy 565
Excursions 20.1 Nightfall 566
We Use 21-cm Radiation to Measure Rotation of the Galaxy 567

Contents xxi
The Milky Way Is Mostly Dark Matter 569
20.3 Studying the Milky Way Galaxy Up Close and Personal 571
We See Stars of Different Age and Different Chemical
Composition 571
Looking at a Cross Section through the Disk 575
The Halo Is More Than Globular Clusters 576
Magnetic Fields and Cosmic Rays Fill the Galaxy 576
20.4 The Milky Way Hosts a Supermassive Black Hole 577
20.5 The Milky Way Offers Clues about How Galaxies Form 579

Summary 582
CHAPTER 21 Modern Cosmology

21.1 The Universe Has a Destiny and a Shape 585
Reviving Einstein’s “Biggest Blunder” 586
The Age of the Universe 590
The Universe Has a Shape 591
21.2 Inflation 592

The Universe Is Much Too Flat 592
The Cosmic Background Radiation Is Much Too Smooth 593
Inflation Solves the Problems 594
21.3 The Earliest Moments 595

May the Forces Be with You 596
A Universe of Particles and Antiparticles 597
The Frontiers of Physics 597
Toward a Theory of Everything 600
Order “Froze Out” of the Cooling Universe 601

Summary 603
CHAPTER 22 The Origin of Structure

22.1 Whence Structure? 607
22.2 Galaxies Form Groups, Clusters, and Larger Structures 608
22.3 Gravity Forms Large-Scale Structure 610

Galaxies Formed Because of Dark Matter 611
There Are Two Classes of Dark Matter 613
A Galaxy Forms within a Collapsing Clump of Dark Matter 613
Connections 22.1 Parallels between Galaxy and Star Formation 615
Searching for Signs of Dark Matter 615
22.4 Observations Help Fill Gaps in the Models 617

Mergers Play a Large Role in Galaxy Formation 617
Star Formation Leaves a Gap in the Models 618
Peculiar Velocities Trace the Distribution of Mass 621
22.5 Multiple Multiverses 623
22.6 The Deep Future 625

xxii Contents

Summary 628
CHAPTER 23 Life
23.1 Life Is a Form of Structure 631
Excursions 23.1 Forever in a Day 632
23.2 Life on Earth 632
FPO A Definition of Life 633

The First Life 633
Life Becomes More Complex 635
Evolution Is a Means of Change and Advancement 637
23.3 Life beyond Earth 637

The Chemistry of Life 638
Connections 23.1 Life, the Universe, and Everything 639
Life within Our Solar System 640

Habitable Zones around Parent Stars 641
23.4 The Search for Signs of Intelligent Life 643

The Drake Equation 644
Technologically Advanced Civilizations 645
23.5 The Future 647

Eventually, the Sun Must Die 647
The Fate of Life on Earth 648
Life in a Future Universe 648

Summary 650
EPILOGUE The Universe and You

The Long and Winding Road E-1
We Are Stardust in Human Form E-2
The Future Arrives Every Day E-4
APPENDIX 1 Mathematical Tools A-1

APPENDIX 2 Physical Constants and Units A-7
APPENDIX 3 Periodic Table of the Elements A-9
APPENDIX 4 Properties of Planets, Dwarf Planets, and Moons A-10
APPENDIX 5 Nearest and Brightest Stars A-13
APPENDIX 6 Observing the Sky A-16
APPENDIX 7 Uniform Circular Motion and Circular Orbits A-25
APPENDIX 8 IAU 2006 Resolutions: Definition of a Planet in the Solar System,
and Pluto A-27
Glossary G-1
Credits C-1
Index I-1
Preface
Astronomy is one of the most rapidly advancing fields in modern science. Its subject mat-
ter intrigues many students and simultaneously challenges textbook authors to keep up
with the most recent discoveries, interpretations, and defi nitions. We built this textbook
and its supporting media package with one goal in mind: to help students understand the
world around them through the eyes of a scientist.
Students who study introductory astronomy typically come from diverse backgrounds;
often their introductory astronomy course is the only formal exposure to science they will
have in their lives! As instructors, everything we do should turn on the answer to basic
questions such as these: What should students carry away from this course that will still be
with them 20 years from now? How should this course change the students who take it? And
how can we, as scientists and educators, facilitate those changes within our students?
A thoughtful response to these questions does not lead to a course that is primarily
about memorization of facts. An introduction to science should be an introduction to what
it means to think deeply about the universe, fi nding patterns and relationships within
the world that go beyond the specifics of a particular object or setting, and applying those
patterns and relationships broadly. If an introductory science course does nothing else, it
should encourage students to think of their brains as muscles that grow strong only with
exercise and training.
An introductory astronomy course should also show students something of the episte-
mology of science: knowledge is arrived at as we constantly challenge and test our ideas.
How can it be that in science we never really prove things to be true, but instead only fail
to show that they are false? It is precisely because science acknowledges lack of certainty
that the pinnacle of human knowledge is a well-tested, well-corroborated scientific theory.
We will have succeeded if a student comes to recognize the difference between science
and pseudoscience, and if 20 years from now a former student still chuckles inwardly at
the naïveté of the phrase “it’s only a theory.”
There is a strong practical side to what we teach about the process of science. Think-
ing about the world as a scientist means learning to sort information from opinion or hap-
penstance. Physical scientists tend not to buy lottery tickets or call psychic help lines;
our students are served well if we pass on a healthy dose of our practical skepticism. A
democratic society, which in the long run can fare no better than its citizens can think,
is served also.
Yet the real reason it matters that students come to see the world through the eyes of
a scientist is the sheer beauty of the view. The majesty of the universe in which we live is
staggering. Simply staring at a deep image of distant galaxies or the glow of a newly formed
star or a picture of Jupiter’s swirling Great Red Spot can be a mind-boggling experience. Add
a dose of quantum mechanics or the idea of the tortured fabric of spacetime around a black
hole, and we fi nd ourselves in territory that would awe Aristotle. Science—this marvelous
expression of human curiosity and passion and reason—has shattered the mental cage in
which we have lived since our ancestors fi rst looked to the sky and wondered about what
they saw there. It is extraordinary to face these vistas surrounded not by speculation but
by the meticulously constructed edifice of science. Finally, to live during that moment in
history when we fi rst recognized our own origins in the Big Bang that occurred some 14
billion years ago—that is something to write home about!
xxiii
xxiv Preface
Science is a vital, exciting, ongoing expression of our humanity. While an introductory

astronomy book should teach students about what science is, showing them what science
has revealed and giving them powerful tools for thinking about the world, it should also
share with them something of the passion scientists feel for the endeavor to which we have
devoted our lives.
In order to meet our goals, our book is very different from many textbooks now on
the market. Our book tells a story. On a small scale, it tells the story of specific ideas.
Understanding comes from thinking carefully and critically about things and how they
work. Helping a student understand a concept as a scientist means guiding that student
through the concept, making heavy use of examples and analogies, and tying the con-
cept back to everyday phenomena and experiences to which the student can relate. That
is what the learning package we have created strives to do. The supporting media, from
tutorial animations to online homework, extend the text’s emphases to the realm of the
interactive.
On a larger scale, 21st Century Astronomy tells a web of different stories. Some of these
stories have to do with what science is and how it is done. Why did Newton choose the
form that he did for his universal law of gravitation? What are the fundamental differ-
ences between Kepler’s empirical “laws” and Newton’s theoretical derivation of the same
relationships? And if Einstein was “right,” why wasn’t Newton “wrong”? Other stories
explore the human side of science, such as how science has led us to think beyond the
box that evolution built for us. Representing science fairly and honestly even means tell-
ing those stories that might make some students uncomfortable, like the story of why our
understanding of the evolving universe and our place in it is science, while “creation sci-
ence” is not.
We tell the story of motion, the story of light, the story of matter and energy, the
story of Earth and our planetary system, the story of the Sun and stars, the story of
life, and the story of the universe as narratives, with one idea flowing into the next
and each idea connected to the whole. Knowledge and understanding are interwoven,
with each idea and insight given meaning by how it fits into the whole. None of the
stories in 21st Century Astronomy exist in isolation. The stories are threads in a grand
intellectual tapestry, woven from the recurring themes of the physics of matter, energy,
radiation, and motion.
Our writing style for 21st Century Astronomy fits this vision of science. The writing
you will fi nd here is far less formal than that found in most textbooks. We allow ourselves
to be conversational, to pursue the occasional digression, to be irreverent or provocative
or idiosyncratic, and to allow the sense of passion and wonder that we feel for the subject
to show through from time to time. Our hope is to catch the feel of a pleasant, stimulating
conversation with a student during office hours. For example, rather than stating that “the
altitude of the north celestial pole above the horizon equals the observer’s latitude” and
leaving it at that, we talk about what the sky looks like over the course of the day as viewed
from the North Pole and why. Then we invite the student to follow along as we head south,
seeing how things change along the way.
We hope that 21st Century Astronomy will appeal to instructors who, like us, have found
that a serious nonscience student in today’s colleges and universities is capable of thinking
more deeply about more conceptually challenging material than is often assumed. Frankly,
with the story that 21st Century Astronomy has to tell—a story that deals head-on with
some of the most fundamental, fascinating, and far-reaching questions humans have ever
asked themselves—few students can resist being drawn in if approached in the right way.
We kept this goal in mind as we created the Third Edition, reconsidering every sentence
in the book from a student’s perspective. We think the result is, as one peer reviewer put
it, “conceptually and analytically enchanting.”
One significant tension in the book is between a traditional organization of topics and an
organization that more accurately represents the way scientists think and work today. True
to our purpose, we chose an approach that reflects the state of our science. For example, we
organized the Solar System section around a theme of comparative planetology. We pulled
Preface xxv
the discussion of tidal interactions, orbital resonances, chaos, and similar phenomena into
a separate chapter (Chapter 10) titled “Gravity Is More than Kepler’s Laws.” We posit the
existence of a rotating, collapsing interstellar cloud and a rotating protoplanetary disk and
discuss the formation of the Solar System in Chapter 6, before discussing the planets them-
selves. This approach enables us to look at the planets from the outset within the context
of how they formed. Enough examples—you get the idea.
Even so, there is flexibility in the text. For example, we feel that the flow of material
through the streamlined early chapters discussing physical principles works well. We
tried to include enough discussion of the human and social aspects of science to break
up the treatment of conceptually difficult material. However, instructors who want to
show their students a bit more astronomy and planetary science before hitting the physics
might want to start with Chapter 6, on the formation of the Solar System, and then pull
in material about orbits and radiation as it is needed in what follows. Another approach
is to start with stars in Chapter 13, pull in radiation and orbits as needed to understand
the properties of stars, and then insert the Solar System as an extended excursion into the
topic of the formation of low-mass stars. Although we wrote the book to tell a complete,
well-integrated story, we tried to allow for different paths tailored to the needs and tastes
of individual instructors.
Our knowledge of the universe is constantly changing and expanding. For example, the
recent and ongoing discoveries by the Cassini spacecraft change our perception of Saturn,
its moons, and its rings on an almost daily basis. The tireless robots Spirit and Opportu-
nity continue to roam the surface of Mars, giving us a better understanding of a planet that
humans may visit and explore before the end of the 21st century. Another example con-
cerns the dark matter and dark energy that dominate the density of the universe. There is a
good chance that the dark matter particle will be discovered at the Large Hadron Collider
as this book is going to press or soon after. In the case of dark energy, 20 years ago few sci-
entists would have believed that it even exists. Today we understand that dark energy is
the major component of our universe.
In closing, we should say a few words about who we are. Among the authors of this
text you will find accomplished and well-known scientists who have been at the forefront
of many exciting and significant events in astronomy and planetary science in the latter
20th and early 21st centuries. The author team includes members of scientific teams that
built three of the cameras that have flown on the Hubble Space Telescope; team members
and leaders of many of the major planetary missions; the chancellor of a major research
university, who has done fundamental work on dark matter and galaxy formation; an extra-
galactic astronomer who also teaches about women in science; and a former president of
the American Association of Physics Teachers. We have been involved in significant fun-
damental research on topics ranging from planetary geology, to the origin and evolution
of the Solar System, to the formation and evolution of stars, to the structure and dynamics
of the interstellar medium, to the nature of galaxies, to the origin of structure in the uni-
verse. There are remarkably few fields discussed in this book to which one or more of the
authors have not contributed in some way over the years. For us, 21st century astronomy is
not a textbook; rather it is the life we live. We hope you find value in our attempt to share
with you what we see when we look up at the sky at night.
Acknowledgments
There are many at W. W. Norton & Company without whom this project would not have
come together. The authors would like to thank our editors: Stephen Mosberg, who fi rst
looked across the table at us and said, “We will find a way”; John Byram, who stepped
into the First Edition midstream and held it together through difficult times; Leo Wieg-
man, who spearheaded the Second Edition; and Rob Bellinger, who guided us through the
many improvements that characterize this Third Edition and developed its supporting
xxvi Preface
media package. Roby Harrington gave us enough rope to hang ourselves, and then helped
us out of trouble when we tried to do just that. We also thank Erik Fahlgren, for providing
editorial guidance; Kelsey Volker, for helping bring the Third Edition to the market; Mat-
thew Freeman, for his work in developing the print ancillary program; and Jane Searle and
Kim Yi, for their diligent work toward helping us publish this book on schedule. Among
those in the trenches were Patricia Longoria and Andrew Sobel, developmental editors;
Mary Lynch, editorial assistant; Stephanie Hiebert, copy editor; Laura Musich, assistant
editor, electronic media; Trish Marx, manager of photo permissions; Junenoire Mitchell
and Ramón Rivera Moret, photo researchers; Stacey Palen, Kevin Marshall, and Trina Van
Ausdal, accuracy checkers; Marian Johnson, managing editor, college books; Rubina Yeh,
designer; Carole Desnoes, layout artist; Second Avenue Software for our new AstroTour
animations and SmartWork online homework; and our ancillary authors, Steven Desch,
Gregory Mack, Ann Schmiedekamp, Tammy Smecker-Hane, Ben Sugerman, and Donald
Terndrup. Finally, we would like to thank Kelly Paralis Keenan of Penumbra Design, who
became a true collaborator on the project rather than simply an artist drawing to spec.
Jeff Hester
Brad Smith
George Blumenthal
Laura Kay
Howard Voss
The Features of 21st Century Astronomy

The overall themes of 21st Century Astronomy are emphasizing foundational physical sci-
ence concepts and stressing scientific literacy for nonmajors by showing the process of
scientific discovery. The following features support students as they start chapters, while
they read, and as they fi nish and prepare for the next topic.
Key Concepts boxes at the beginning of each chapter call attention to the main issues
and topics that will be discussed in the chapter.
Trailmarker statements—brief, one-sentence summaries—appear throughout each chap-
ter to draw students’ attention to fundamental concepts as they read. Redesigned for
this edition, these trailmarkers are also a useful scanning tool for review.
Math Tools boxes help show students how mathematics is the language of science.
Though we rarely use more than basic algebra, we have listened to our reviewers
and separated some of the more difficult concepts from the running text using these
boxes. This format allows instructors more flexibility in choosing which mathe-
matical concepts they cover and what reading they assign. We make every effort to
supply the equations with real-world values, to make the math and the underlying
concepts easier to understand.
Annotated figures throughout the book present clear, accurate science in visually com-
pelling ways.
Seeing the Forest for the Trees essays conclude each chapter with a thematic synthesis
and a glimpse ahead to future chapters. These essays also help students make con-
nections between astronomy and the liberal arts.
Summaries provide an outline of the key concepts as they were applied in the
chapter.
Double the number of end-of-chapter questions help instructors assess student progress.
Many of these questions have been adapted for use in SmartWork, Norton’s online
homework system. To make creating assignments easier for instructors, we have
Preface xxvii
marked questions that we consider to be of intermediate difficulty with one asterisk

(*) and those that are especially challenging with two asterisks (**).
Thinking about the Concepts questions provide an opportunity for review with
emphasis on the more important concepts.
Applying the Concepts questions involve problem solving and critical thinking. StudySpace
Each chapter ends with a reminder of additional resources available on the StudySpace
website for further chapter review. For more on StudySpace, see page xxix.
In the Third Edition, we have made an effort to reduce the number of boxes in the text,
though we retain three types of boxes from the Second Edition. In addition to the new
Math Tools boxes mentioned above, these boxes touch on material either because it is of
special importance and needs to be highlighted or because it is somewhat out of the main
stream of our journey.
Foundations boxes discuss the basic science that is central to our physical understand-
ing of the universe.
Connections boxes draw attention to recurring themes—bridges between different parts
of our journey.
Excursions boxes are short but interesting field trips that highlight how scientists apply
what they learn to solving problems.
What’s New in the Third Edition?

We were gratified that so many instructors found the First and Second Editions of 21st
Century Astronomy useful teaching tools. For the Third Edition, we called upon the feed-
back of our adopters and reviewers to build an edition that has evolved with the needs of
teachers and students. We have kept the “big story” approach to astronomy that was so well
received the fi rst time around, but we place more emphasis on keeping students focused
on the big picture. We thoroughly reconsidered every sentence in the text and revised for
clarity and style wherever it was necessary. And, as described already, we’ve introduced
a new approach to math and reduced the number of boxes overall.
Changes to Content
We have also introduced important new content—such as the new Chapter 23 on
astrobiology—and reorganized our introduction to the process of science. Some of the
more important content changes and additions include the following:
Chapter 1 brings together all of our introductory coverage on the process of science.
Chapter 3, on the work of Newton and Kepler, is more accessible, thanks to our new
approach to math.
Chapter 6 includes a new section on and expanded coverage of extrasolar planets.
Chapter 11 contains new discoveries about Saturn’s rings, its large moon Titan, and the
cryovolcanically active moon Enceladus from the Cassini mission.
Chapter 12 includes the properties of the five dwarf planets and new discoveries of the
properties of Kuiper Belt objects.
Chapter 17 provides new material on gravity waves and includes all treatment of the
xxviii Preface
An accretion disk is
(a) like a traffic circle with theories of relativity, including special relativity, which was formerly introduced
no exits.
in Chapter 4.
Chapters 18, 19, and 20 have been reorganized to bring our introduction to galaxies
into parallel alignment with our planet and star chapters by starting with the big
picture (“Our Expanding Universe”) and then going into details of galaxies and
the Milky Way. New results from the Wilkinson Microwave Anisotropy Probe
on cosmic background radiation are also discussed.
Chapters 21 and 22, on cosmology, have been greatly updated to illustrate the advance
of observationally based cosmology and new models and theories. We have also
included a discussion of the relative speculative topic of multiverses.
Chapter 23 brings together the discussion about life on Earth and the search for life
elsewhere—the science of astrobiology—in a single new chapter.
1 Material rains down

(b)
from the collapsing,
gdiVi^c\cloud. New Pedagogical Features
We have strived to make the new edition more student-friendly. The following in-text
features are designed to make the textbook even easier for students to use.
AstroTour AstroTour icons point out when an animation is available to help

a student master an important concept. These icons become hyperlinks in the ebook
2 The vertical motion of material version of the text.
from above cancels the vertical
motion from below…
VISUAL ANALOGY Students really like the analogies we make between astronomical
concepts and everyday phenomena, such as when we compare a forming accretion
disk to cars piling up in a traffic circle with no exits. New Visual Analogy icons help
students make immediate connections between the written analogies and the images
that illustrate them.
3 …but the rotational motion

Math Tools boxes help communicate to students how mathematics is the language of the
remains. The material joins
the rotating accretion disk.
scientist. These boxes also separate some of the more difficult mathematical concepts
from the running text, allowing instructors flexibility in assigning reading.
VISUAL ANALOGY FIGURE 6 .7 (a) Traffic piles up in a
traffic circle with entrances but no exits. (b) Similarly,
gas from a rotating cloud falls inward from opposite
sides, piling up onto a rotating disk.
M AT H T O O L S 7.1
How to Read Cosmic Clocks

As Foundations 7.1 explains, a geologist can calculate the age of isotope of the element lead (206Pb, its final daughter). The half-life
a mineral by measuring the relative amounts of a parent radio- of 238U is 4.5 billion years. This means that in 4.5 billion years, a
isotope and its daughter product. The time interval over which a sample that originally contained the uranium isotope (the parent)
radioactive isotope decays to half its original amount is called its but no lead (its final daughter) would be found instead to contain
half-life. With every half-life that passes, the remaining amount equal amounts of uranium and lead. If we were to find such a
of the radioisotope will decrease by a factor of 2. For example, mineral, we would know that half the uranium atoms had turned
over three half-lives, the remaining amount of a parent radioiso- to lead and that the mineral formed 4.5 billion years ago.
tope will be 1⁄2 × 1⁄2 × 1⁄2 = 1⁄8 of its original amount. We can make Let’s look at another example, this time with a different iso-
this concept more general with a simple relationship: tope of uranium (235U) that decays to a different lead isotope (207Pb)
with a half-life of 700 million years. Suppose that a lunar min-
PF
___
PO (2)
1 n
= __ , eral brought back by astronauts has 15 times as much 207Pb (the
daughter product) as 235U (the parent radioisotope). This means
where PO and P F are the original and fi nal amounts, respec- that 15 ⁄16 of the parent radioisotope (235U) has decayed to the daugh-
tively, of a parent radioisotope, and n is the number of half- ter product (207Pb), leaving only 1⁄16 of the parent remaining in the
lives that have gone by. mineral sample. Noting that 1⁄16 is (1⁄2)4, we see that 4 half-lives
The most abundant isotope of the element uranium (238U, the have elapsed since the mineral was formed, and that this lunar
parent) decays through a series of intermediate daughters to an sample is therefore 4 × 700 million = 2.8 billion years old.
Preface xxix
Student Resources
StudySpace: www.wwnorton.com/studyspace
Ann Schmiedekamp, Pennsylvania State University–Abington
Norton’s free and open student website for the introductory astronomy course has been
revised and expanded to support the Third Edition. Study Plans for each chapter tie together
all the resources that are available:
Quiz+ diagnostic quizzes Vocabulary flash cards

30 AstroTour animations News feeds
Interactive simulations Links to ebook and SmartWork
StudySpace now includes free Quiz+ diagnostic quizzes. These gradable multiple-choice
quizzes generate Custom Study Plans for students, linking them to other site resources that
will help students where they need it.
AstroTour animations plus interactive simulations developed at the University of
Nebraska–Lincoln reinforce the text’s explanation of important physical concepts (see
below for details).
From StudySpace, students can also access premium content in the ebook and Smart-
Work online homework (see below for details).
AstroTour Animations
We commissioned 12 new animations to complement those we had created for the Second
Edition. Many new animations use photorealistic 3-D art to help students visualize important
physical concepts. Animations also make use of analogies and art from the printed text.
The Earth Spins and Revolves The H-R Diagram

The Celestial Sphere and the Ecliptic The Solar Core
The View from the Poles Star Formation
The Moon’s Orbit: Eclipses and Phases Hubble’s Law
Kepler’s Laws Big Bang Nucleosynthesis
Velocity, Acceleration, Inertia Active Galactic Nuclei
Newton’s Laws and Universal Galaxy Interactions and Mergers
Gravitation Dark Matter
Elliptical Orbits
Light as a Wave, Light as a Photon
Atomic Energy Levels and the Bohr
Model
Atomic Energy Levels and Light Emis-
sion and Absorption
Doppler Effect
Geometric Optics and Lenses
Solar System Formation
Visual Analogy: Traffic Circle
Processes That Shape the Planets
Continental Drift
Hot Spot Creating a Chain of Islands
Atmospheres: Formation and Escape
Greenhouse Effect
Tides and the Moon
Cometary Orbits
Stellar Spectrum
xxx Preface
AstroTour animations are available from the free StudySpace student website, and they are
also integrated into assignable SmartWork exercises. Offline versions of the animations for
classroom presentation are available from the Instructor’s Resource disc.
Interactive Simulations
In addition to the AstroTour animations, many of which
are interactive, Norton licensed interactive simulations
created at the University of Nebraska–Lincoln. These
enable students to manipulate variables and work toward
understanding physical concepts. All simulations are
integrated into Study Plans and Quiz+ feedback on the
StudySpace student website. They are also included on
the Instructor’s Resource disc.
SmartWork Online Homework

www.wwnorton.com/smartwork
SmartWork online homework allows instructors to assign
algorithmically generated homework that offers students
feedback at just the right teachable moment. Developed
by university educators, SmartWork is the most intuitive
tutorial and online homework system available.
SmartWork combines the flexibility of traditional paper homework with the adminis-
trative ease of computerized grading and the effectiveness of student-focused feedback.
Instructors can draw from Norton’s bank of high-quality, class-tested, ready-made ques-
tions. Or they can easily modify existing questions or write new ones. The system sup-
ports a wide variety of question types, including numeric entry, multiple choice, ranking
tasks, and short text entry. For the Third Edition, we have also added questions based
on the AstroTour animations.
Students may purchase access to Smart-
Work bundled with the printed book or
Though most instructors ebook, or they may purchase an access code
use SmartWork content at smartwork.wwnorton.com.
as-is, a full suite of
intuitive authoring tools
allows instructors to edit
existing problems or
create new ones. No
programming knowledge
is required.
SmartWork’s gradebook
allows you to view your
student’s scores in
several formats. Here,
the At-a-Glance view
shows student progress
through an assignment.
Preface xxxi
Every SmartWork
question includes hints
that suggest how to
approach the problem,
as well as answer-
specific feedback that
addresses common
errors.
Every question in
SmartWork provides a
reference link to the
complete ebook.
SmartWork offers a
wide variety of question
types, including new
Ranking Task Exercises.
These exercises ask
students to make
comparisons between
similar items and rank
them in a specific order.
SmartWork and ebook integration SmartWork is available as a stand-alone purchase

or with an integrated ebook version of 21st Century Astronomy. Links to the ebook make it
easy for students to consult the text while completing their homework assignments. Instruc-
tors may order any volume of the print text bundled with access to both SmartWork and
the ebook for no extra charge. Ask your local Norton representative for details.
Starry Night College CD-ROM and Workbook

Workbook by Steven Desch, Guilford Technical Community College,
and Donald Terndrup, The Ohio State University
The remarkably realistic and user-friendly Starry Night College planetarium software allows
students to explore stars and objects in our cosmic neighborhood and beyond. The accom-
xxxii Preface
panying workbook, developed exclusively for 21st Century Astronomy, has been expanded
to include 22 observation exercises that guide students’ virtual explorations of the night
sky and help them apply what they’ve learned from the text.
Both Starry Night software and Norton’s Starry Night Workbook contain references to
passages in the textbook. Adopters of the Third Edition may also download electronic desk
copies of Starry Night’s instructor materials.
Ebook
21st Century Astronomy is available in Norton ebook format, a convenient alternative that
retains all of the printed book’s content. The ebook offers a variety of tools for study and
review, including sticky notes, highlighters, embedded links to the AstroTour animations
and SmartWork online homework, and a search function.
New copies of the text may be bundled with SmartWork and the ebook version of the
text at no extra charge. The SmartWork/ebook package makes it easy for students to consult
the text when completing their homework assignments.
For students who prefer a downloadable ebook, a PDF version of the ebook is also avail-
able from Powell’s Books at powells.com. Go to nortonebooks.com for more information
on all ebook options.
Instructor Resources
Instructor’s Manual
Gregory D. Mack, Ohio Wesleyan University
Ben Sugerman, Goucher College
Steven Desch, Guilford Technical Community College
Revised and expanded for the Third Edition, this resource includes brief chapter overviews,
lecture outlines to accompany the PowerPoint lecture slides, a media guide to web and other
electronic resources available to instructors that includes notes on the animations contained
on the Instructor’s Resource disc and StudySpace student website, worked solutions to all of
the end-of-chapter problems, and instructor’s notes for the Starry Night Workbook activities.
Test Bank
Tammy Smecker-Hane, University of California–Irvine
Revised and expanded for the Third Edition, each chapter of the Test Bank consists of three
question types classified according to Norton’s taxonomy of educational objectives:
1. Factual questions (what?) test declarative knowledge, including textbook defi nitions
and relationships between two or more pieces of information.
2. Applied questions (how?) pose problems in a context different from the one in which a
particular concept was learned, requiring students to draw from their declarative and/
or procedural understanding of important concepts.
3. Conceptual questions (why?) ask students to draw from their prior experience and use
critical thinking skills to take part in qualitative reasoning about the real world.
Questions are further classified by section and difficulty, making it easy to construct tests
and quizzes that are meaningful and diagnostic. This approach enables instructors to accu-
rately judge students’ mastery of the material—what they know, what they don’t know, and
to what degree—on the basis of the assessment outcomes.
Preface xxxiii
The question types are short answer, multiple choice, and true-false. The Test Bank is
available in print, Word RTF, PDF, and ExamView formats.
PowerPoint Lecture Slides

Gregory D. Mack, Ohio Wesleyan University
These ready-made lecture slides integrate selected art from the text, “clicker” questions,
and offline, lecture-ready versions of the AstroTour animations. Designed to accompany
the lecture outlines found in the Instructor’s Manual, these lecture slides are fully edit-
able and are available in Microsoft PowerPoint format.
Instructor’s Resource Disc

This helpful resource includes the PowerPoint lecture outlines with “clicker” questions,
offline versions of the animations, plus selected photographs and all drawn art from the
text. The Instructor’s Resource disc also contains the interactive simulations used on the
StudySpace student website, as well as additional clicker questions developed at the Uni-
versity of Nebraska–Lincoln.
BlackBoard and WebCT Course Cartridges

Course cartridges for BlackBoard and WebCT include access to the animations, a Study
Plan for each chapter, multiple-choice tests, and links to ebook and SmartWork premium
content.
Reviewers
Production of a book like this is a far larger enterprise than any of us imagined when we
jumped on board. We would like to thank the teachers who reviewed portions of the manu-
script along the way and helped us build a stronger book:
Timothy Barker, Wheaton College Matthew Francis, Lambuth University

Edwin Bergin, University of Pittsburgh Kevin Gannon, College of Saint Rose
William Blass, University of Bill Gutsch, St. Peter’s College
Tennessee–Knoxville Karl Haisch, Utah Valley University
Steve Bloom, Hampden Sydney College Barry Hillard, Baldwin Wallace College
Daniel Boice, University of Texas–San Hal Hollingsworth, Florida International
Antonio University
Jack Brockway, Radford University Olencka Hubickyj-Cabot, San Jose State
Juan E. Cabanela, Minnesota State University
University–Moorhead Kevin M. Huffenberger, University of
Amy Campbell, Louisiana State Miami
University Steve Kawaler, Iowa State University
Robert Cicerone, Bridgewater State Kevin Lee, University of
College Nebraska–Lincoln
Eric M. Collins, California State Matt Lister, Purdue University
University–Northridge M. A. K. Lodhi, Texas Tech University
John Cowan, University of Kevin Marshall, Bucknell University
Oklahoma–Norman Chris McCarthy, San Francisco State
Robert Dick, Carleton University University
Tom English, Guilford Technical Chris Mihos, Case Western University
Community College Milan Mijic, California State University–
John Finley, Purdue University Los Angeles
xxxiv Preface
J. Scott Miller, University of Louisville Caroline Simpson, Florida International

Scott Miller, Sam Houston State University
University Ian Skilling, University of Pittsburgh
Andrew Morrison, Illinois Wesleyan Tammy Smecker-Hane, University of
University California–Irvine
Edward M. Murphy, University of Virginia Allyn Smith, Austin Peay State University
Kentaro Nagamine, University of Nevada– Ben Sugerman, Goucher College
Las Vegas Neal Sumerlin, Lynchburg College
Jascha Polet, California State Polytechnic Christopher Taylor, California State
University University–Sacramento
Daniel Proga, University of Nevada–Las Trina Van Ausdal, Salt Lake Community
Vegas College
Laurie Reed, Saginaw Valley State Karen Vanlandingham, West Chester
University University
Masao Sako, University of Pennsylvania Paul Voytas, Wittenberg University
Ann Schmiedekamp, Pennsylvania State Paul Wiita, Georgia State University
University David Wittman, University of California,
Jonathan Secaur, Kent State University Davis
We also thank the following colleagues who served as Second Edition reviewers:
Scott Atkins, University of South Dakota Laura Kay, Barnard College

Peter A. Becker, George Mason University Bill Keel, University of Alabama
Timothy C. Beers, Michigan State Leslie Looney, University of Illinois at
University Urbana–Champaign
Juan Cabanela, Minnesota State Norm Markworth, Stephen F. Austin State
University–Moorhead University
Judith Cohen, California Institute of Scott Miller, Pennsylvania State
Technology University
Paul Hintzen, California State University– Ata Sarajedini, University of Florida
Long Beach Paul P. Sipiera, William Rainey Harper
Paul Hodge, University of Washington College
William A. Hollerman, University of Donald Terndrup, Ohio State University
Louisiana at Lafayette Richard Williamon, Emory University
Adam Johnston, Weber State University
About the Authors
Jeff Hester received his PhD from Rice constitutes most of the mass in the universe, the origin of galaxies
University and is professor emeritus of and other large structures in the universe, the earliest moments in
physics and astronomy at Arizona State the universe, astrophysical radiation processes, and the structure
University. His research interests are the of active galactic nuclei such as quasars. Besides teaching and con-
interstellar medium in the Milky Way and ducting research, Chancellor Blumenthal has served as the chair of
external galaxies; structure of the diffuse the UC–Santa Cruz Astronomy and Astrophysics Department, has
interstellar medium; interstellar shock chaired the Academic Senate for both the UC–Santa Cruz campus
waves; supernova remnants; pulsar wind and the entire University of California system, and has served as
interactions; Herbig-Haro objects; and H II region structure. the faculty representative to the UC Board of Regents.
Brad Smith has served as an associ- Laura Kay is Ann Whitney Olin profes-
ate professor of astronomy at New Mexico sor of physics and astronomy at Barnard
State University, a professor of planetary College, where she has taught since 1991.
sciences and astronomy at the University She received a BS degree in physics and an
of Arizona, and a research astronomer AB degree in feminist studies from Stan-
at the University of Hawaii. Through his ford University, and MS and PhD degrees
interest in Solar System astronomy, he has in astronomy and astrophysics from the
participated as a team member or imag- University of California–Santa Cruz. As
ing team leader on several US and international space missions, a graduate student she spent 13 months at the Amundsen Scott
including Mars Mariners 6, 7, and 9; Viking; Voyager; and the station at the South Pole in Antarctica. She studies active galac-
Soviet Vega and Phobos missions. More recently, Smith’s interests tic nuclei, using groundbased and X-ray telescopes. She teaches
have turned to other planetary systems, working as a team member courses on astronomy, astrobiology, women and science, and polar
of the Hubble Space Telescope NICMOS experiment. Asteroid 8553 exploration. At Barnard she has served as chair of the Physics &
Bradsmith was named for Smith. He has four times been awarded Astronomy Department, chair of the Women’s Studies Department,
the NASA Medal for Exceptional Scientific Achievement. Smith is chair of Faculty Governance, and interim associate dean for Cur-
a member of the IAU Working Group for Planetary System Nomen- riculum and Governance.
clature and is chair of the Task Group for Mars Nomenclature. He
is now semi-retired but remains affi liated with the Institute for Howard Voss is professor emeritus of
Astronomy at the University of Hawaii, Manoa. physics at Arizona State University, where
he taught for over four decades, served as
George Blumenthal is chancellor at chair of the Department of Physics, and
the University of California–Santa Cruz, received the Distinguished Faculty Award
where he has been a professor of astronomy and the Dean’s Teaching Award. He was
and astrophysics since 1972. Chancellor awarded the Melba Phillips Medal by the
Blumenthal received his BS degree from American Association of Physics Teach-
the University of Wisconsin–Milwaukee ers, which he served as president, secretary, a member of the
and his PhD in physics from the University Executive Board, and in other offices. He also served the Ameri-
of California–San Diego. As a theoretical can Institute of Physics in several positions, including as chair
astrophysicist, Chancellor Blumenthal’s research encompasses of the Publishing Policy Committee and as a member of the Gov-
several broad areas, including the nature of the dark matter that erning Board.
xxxv
THIRD EDITION
21 ST CENTURY
ASTRONOMY
PA R T I Introduction to Astronomy
The most beautiful thing we can experience is the mysterious.

It is the source of all true art and all science.
He to whom this emotion is a stranger,
who can no longer pause to wonder and stand rapt in awe,
is as good as dead: his eyes are closed.
ALBERT EINS T EIN (1879 –1955)
The Coma cluster of galaxies at a distance of more than 300 million light-years.
CHAP TER 1
Why Learn Astronomy?

1.1 Getting a Feel for
K E Y CONCEP T S
the Neighborhood Before traveling through unfamiliar terrain, it helps to have
We chose the title of this book—21st Century Astronomy— some idea of where you are going, what you might see
to emphasize that this is the most fascinating time in his- along the way, and what you should pack for the journey.
tory to be studying this most ancient of sciences. Loosely We’ll begin this chapter by sketching out a rough map of the
translated, the word astronomy means “patterns among the universe and our place within it. And we’ll present some of
stars.” But modern astronomy—the astronomy we will talk the tools that you will need to take along as you look at the
about in this book—has become far more than looking at the wonders of the universe through the eyes of a scientist. In
sky and cataloging what is visible there. We are confident in this chapter’s overview of what is to come, we will find that
our answers to many of the questions that you might have
• The universe is vast beyond all human experience, yet
asked yourself as a child when you looked at the sky. What
it is governed by the same physical laws that shape our
are the Sun and Moon made of? How far away are they? What
are the stars? How do they work? Do they have anything to daily lives.
do with me? In addition, issues as seemingly metaphysical • We are a product of the universe; the very atoms of
as the origin and fate of the universe and the nature of space which we are made were formed in stars that died long
and time have become the subjects of rigorous scientific before the Sun and Earth were formed.
investigation. The answers we are finding to these questions
• There is nothing special about our place in the
are often far more wondrous than our predecessors could
universe; it is indistinguishable from any other place.
have dreamed. They are changing not only our view of the
cosmos, but our view of ourselves as well. • Science—like art, literature, and music—is a creative
human activity; it is also a remarkably powerful, successful,
and aesthetically beautiful way of viewing the world.
Glimpsing Our Place in the Universe • Understanding comes from thinking carefully and
Most of us have a permanent address—street number, street, deeply about patterns in the world, not simply from
city, state, country. It is where the mail carrier delivers our memorization of facts.
mail. But let’s expand our view for a moment. We also live • Like climbing a mountain, the journey into astronomy
somewhere within an enormously vast universe. What then requires effort, but the view from the top is amazing to
is our “cosmic address”? It might look something like this: behold.
planet, star, galaxy, galaxy group, galaxy cluster.
We all reside on a planet called Earth, which is orbiting
under the influence of gravity about a star called the Sun.
3
4 Chapter 1 Why Learn Astronomy?
FIGURE 1.1 The place where

Earth we reside in the universe—
our cosmic address: Earth,
Solar System, Milky Way
Sun Galaxy, Local Group, Virgo
Supercluster. We live on
Earth, a planet orbiting the
Solar System Sun in our Solar System,
which is a star in the Milky
Way Galaxy. The Milky Way
is a large galaxy within the
Local Group of galaxies, which
in turn is located in the Virgo
Supercluster.
Milky Way Galaxy
Local Group
Virgo Supercluster
The Sun is an ordinary, middle-aged star, more massive many of these stars are themselves surrounded by planets,
and luminous than some stars but less massive and lumi- suggesting that other planetary systems may be the rule
nous than others. The Sun is extraordinary only because rather than the exception.
of its importance to us within our own Solar System. Our The Milky Way, in turn, is a member of a small collection
Solar System consists of eight classical planets—Mercury, of a few dozen galaxies called the Local Group. The Milky
Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune. Way Galaxy and the Andromeda Galaxy are true giants
It also contains many smaller bodies, such as dwarf plan- within the Local Group. Most others are what astronomers
ets, asteroids, and comets. In August 2006 the International call “dwarf galaxies.” Looking farther outward, the Local
Astronomical Union (IAU) redefi ned what astronomers Group is part of a vastly larger collection of thousands of
mean when they call something a planet (see Appendix 8.) galaxies—a supercluster—called the Virgo Supercluster.
Under this new defi nition, Pluto, known to all of us as the We can now define our “cosmic address”—Earth, Solar
“ninth planet” since its discovery in 1930, was unceremo- System, Milky Way Galaxy, Local Group, Virgo Supercluster—
niously stripped of its planet status and demoted to dwarf as illustrated in Figure 1.1. Yet even this address is not com-
planet. We might defi ne the size of our Solar System by its plete, because the vast structure
most distant planetary body, a dwarf planet called “Eris,” we just described is only the We all have
although others could rightfully include the realm of small, local universe. The part of the a cosmic address.
icy bodies that extends far beyond the orbit of Eris. universe that we can see
The Sun is located about halfway out from the center of extends far beyond—a distance that light takes 13.7 billion
a flattened collection of stars, gas, and dust, referred to as years to traverse—and within this volume we estimate that
the Milky Way Galaxy. Our Sun is just one among approxi- there are several hundred billion galaxies, roughly as many
mately 300 billion stars scattered throughout the galaxy, and galaxies as there are stars in the Milky Way!
1.1 Getting a Feel for the Neighborhood 5
ancestors were on the scene. To reach the limits of the cur-

The Scale of the Universe rently observable universe takes 13.7 billion years—the age
One of the fi rst conceptual hurdles that we face as we begin of the universe, or about three times the age of Earth.
to think about the universe is its sheer size. If a hill is big, Look at that comparison again. The size of Earth is to the
then a mountain is really big. If a mountain is really big, vast expanse of the universe as a single snap of your fingers
then Earth is enormous. But where do we go from there? We is to three times the amount of time that has passed since
quickly run out of superlatives as the scale of what we are the Sun and Earth were formed! That’s something to ponder
talking about comes to dwarf our human experience. One the next time you look up at a star-filled summer sky.
technique that can help us develop a sense for the size of
things in the universe is to use a little sleight of hand and
move from discussing distance to talking instead about The Origin and Evolution
time. If you are driving down the highway at 60 miles per
hour (mph), a mile is how far you go in a minute. Sixty miles
of the Universe
is how far you go in an hour. Six hundred miles is how far While seeking knowledge about the universe and how it
you go in 10 hours. So to get a feeling for the difference in works, modern astronomy and physics have repeatedly
size between 600 miles and 1 mile, you can think about the come face-to-face with a number of age-old questions long
difference between 10 hours and a single minute. thought to be solely within the domain of philosophers.
We can play this same game in astronomy, but the speed Figure 1.3 depicts a traveler raising the veil of the heavens
of a car on the highway is far too slow to be useful. Instead to see what lies there. Like this fictional traveler, philoso-
we will use the greatest speed in the universe—the speed phers throughout most of history looked at the universe and
of light. Light travels at 300,000 kilometers per second (km/s). saw it as remote and different from Earth—disconnected
At that speed, light can circle Earth (a distance of 40,000 from our terrestrial existence. When modern astronomers
kilometers [km]) in just under 1⁄7 of a second—about the time look at the universe, they see instead a network of ongoing
it takes you to snap your fin- processes that we are a part of. Astronomy begins by look-
gers. Fix that comparison in The travel time of ing out at the universe, but increasingly that outward gaze
your mind. The size of Earth light helps in turns introspective as we come to appreciate that our very
is like—snap!—a snap of your understanding the existence is a consequence of those same processes.
fingers. Now follow along in scale of the universe. The study of the chemical evolution of the universe is
Figure 1.2 as we move out- such a case. Theory and observation tells us that the uni-
ward into the universe. (This figure is highlighted as a verse was created in a “Big Bang” some 13.7 billion years
"Visual Analogy." The "Visual Analogy" labels indicate draw- ago. As a result of both observation and theoretical work, we
ings that make analogies between astronomical phenomena now understand that the only chemical elements found in
and everyday objects more concrete.) We next encounter the abundance in the early universe were hydrogen and helium,
Moon, 384,000 km away, or a bit over 11⁄4 seconds when mov- plus tiny amounts of lithium, beryllium, and boron. Yet we
ing at the speed of light. So if the size of Earth is a snap of live on a planet with a core consisting mostly of iron and
your fingers, the distance to the Moon is about the time that nickel, surrounded by a mantle made up of rocks contain-
it takes to turn a page in this book. Continuing on, we find ing large amounts of silicon and various other elements.
that at this speed the Sun is 81⁄3 minutes away, or the length Our bodies are built of carbon, nitrogen, oxygen, sodium,
of a hurried lunch at the student union. Crossing from one phosphorus, and a host of other chemical elements. If these
side of the orbit of Neptune, the outermost classical planet elements that make up Earth and our bodies were not pres-
in our Solar System, to the other takes about 8.3 hours. Think ent in the early universe, where did they come from?
about that for a minute. Let it sink in. Comparing the size of The answer to this question lies within the stars. Nuclear
Neptune’s orbit to the circumference of Earth is like com- fusion reactions occurring deep within the interiors of stars
paring the time of a good night's sleep to a single snap of combine atoms such as hydrogen, forming more massive
your fingers. atoms, and thereby accomplishing the alchemist’s unattain-
In crossing Neptune’s orbit, however, we have only just able dream of transforming one element into another. When
begun our journey. Many steps remain. It takes us a bit over a star exhausts its nuclear fuel and nears the end of its life,
four years to cover the distance from Earth to the nearest it often loses much of its mass—including some of the new
star (other than the Sun), or as much time as you spent in atoms formed in its interior—
high school. At this point, even our analogy using the travel by blasting it back into inter- We are stardust.
time of light can no longer bring astronomical distance to a stellar space. We will talk
human scale. Light takes about 100,000 years to travel across later about the life and death of stars. For now it is enough
our galaxy—about the time that modern humans (Homo sapi- to note that our Sun and Solar System formed from a cloud
ens) have walked the surface of Earth. To reach the nearest of interstellar gas and dust that had been “polluted” by the
large galaxies beyond our own takes a few million years—the chemical effluent from earlier generations of stars. This
amount of time that has passed since our australopithecine chemical legacy supplies the building blocks for the inter-
(a)
Moving outward through the VISUAL ANALOGY FIGURE
universe at the speed of light, 1. 2 Thinking about the
going around Earth is like a time it takes for light to
snap of your fingers,…
Earth’s circumference travel between objects
1/ 7 second helps us to better
comprehend the vast
(b) distances in the universe.
1.25 seconds Moon …the Moon is a little more
Earth
than a second away,…
Times shown are
light-travel times.
(c)
8.3 minutes
Earth
Sun …the Sun’s distance
Because of the vast distances in the is like a quick meal,…
universe, we’re not showing objects
to scale here: they’d be much too small!
(d) Neptune
…and the diameter of the
Sun Solar System, based on the
orbit of the most distant planet,
Neptune, is a night’s sleep.
8.3 hours
(e)
4.2 years The distance to the nearest

Sun Proxima Centauri, star is like the time you
the closest star spend in high school,…
to our Sun
(f)
Earth’s Sun
…the diameter of the
Milky Way Galaxy
galaxy is like the age
of our species,…
100,000 years
(g)
…and the distance between

2.5 million years galaxies is like the time since
Milky Way Andromeda
Galaxy Galaxy earliest human ancestors
walked on Earth.
(h)
13.7 billion years

The size of the universe is like
three times the age of Earth.
Radius of the observable universe
Comparing the size of the universe to the size of Earth is like comparing
three times the age of the planet to a snap of your fingers.
1.2 Science Involves Exploration and Discovery 7
tific explanations.1 How and when did the universe begin?

What combination of events—some probable, others much
less likely—have led to our existence as sentient beings liv-
ing on a small rocky planet orbiting a typical middle-aged
star? Was this a unique happening, or are there others like
us scattered throughout the galaxy? We’ll look into the gen-
esis of terrestrial life, but we’ll also examine the possibili-
ties of life elsewhere in our Solar System and beyond—a
subject called astrobiology. For life to exist around other
stars in the galaxy, there must be life-sustaining planets
to support it. We will include the discovery of extrasolar
planets and how they compare with the planets of our own
Solar System.
1.2 Science Involves

FIGURE 1. 3 Throughout most of history, humans conceived of
the rest of the universe as a place apart from us. Here a traveler
Exploration and
raises the curtain of the firmament to get a glimpse of what lies
beyond Earth.
Discovery
As we look at the universe through the eyes of astronomers,
we will also learn something of how science works. It is
esting chemical processes that go on around us—chemical beyond the scope of this book for us to provide a detailed
processes such as life. Figure 1.4 symbolizes this intimate justification for all that we will say. However, we will try
relationship between the world around us and our heritage to offer some explanation of where an idea comes from and
in the stars. Look around you. The atoms that make up why we believe it to be valid. We will not present some-
everything you see were formed in the hearts of stars. Poets thing as fact unless there is a compelling reason to believe
sometimes say that “we are stardust,” but this is not just it. We will be honest when we are on uncertain, specula-
poetry. It is literal truth. tive ground, and we will admit it when the truth is that
As humans, we have long speculated about our begin- we really do not know. This book is not a compendium of
nings. Who or what is responsible for our existence? How revealed truth or a font of accepted wisdom. Rather, it is
were Heaven and Earth created? In the modern world, an introduction to a body of knowledge and understanding
primitive creation myths have largely given way to scien- that was painstakingly built (and sometimes torn down and
rebuilt) brick by brick.
It is almost impossible to overstate the importance of
science in our civilization. One obvious manifestation of
FIGURE 1. 4 You and everything around you, including beautiful
science is the technology that has enabled us to explore well
waterfalls, are composed of atoms that were forged in the
interior of stars that lived and died before the Sun and Earth
beyond our planet. Since the 1957 launch of Sputnik, the
were formed. The supermassive star Eta Carinae, shown on the fi rst human-made satellite, we have lived in an age of space
left, is currently ejecting a cloud of chemically enriched material. exploration. Five decades later, we have seen humans walk
on the Moon (Figure 1.5) and have sent unmanned probes to
visit all of the classical planets. Spacecraft have flown past
asteroids, comets, and even the Sun. Our inventions have
landed on Mars, Venus, Titan (Saturn’s largest moon), and
an asteroid, and have plunged into the atmosphere of Jupi-
ter. Most of what we know of the Solar System has resulted
from these past five decades of exploration.
Satellite observatories in orbit around Earth have also
given us many new perspectives on the universe. The very
atmosphere that shields us from harmful solar radiation
1
A few cultures still retain their creation myths. And even within
“enlightened” society, certain factions have ignored science in favor
of nonscientific “creationism” and “intelligent design.”
(a)
KEY Space Lunar and Historical
observatories planetary explorers
Apollo lunar rover (1971)
(b)
Galileo (1989–), first Jupiter

orbiter, atmospheric probe Chandra (1999–),
Advanced X-ray Astrophysics
Mariner 4 (1964–1965), Surveyor 1 (1966), Sputnik (1957–1958), Facility (AXAF)
first images of Mars lunar lander USSR, first human-made
satellite
Viking lander 1 (1975–1982), Hubble Space Telescope

first of two Mars landers (1990–), UV, visible,
infrared astronomy
FIGURE 1. 5 (a) Apollo 15 astronaut James B. Irwin stands by the lunar rover during an excursion to explore and collect
samples from the Moon. (b) Artificial satellites and space probes have grown more complex since the 1957 launch of Sputnik
1. These spacecraft are all shown to the same scale. Some are astronomical observatories that view space from Earth’s
orbit. Others are interplanetary explorers sent to investigate other worlds within our Solar System.
also blinds us to much of what is going on around us. Space at a computer screen than peering through the eyepiece of
astronomy continues to show us vistas hidden from the gaze a telescope. As astronomers, we use computers to collect
of groundbased telescopes by the protective but obscuring and analyze data from telescopes, calculate physical mod-
blanket of our atmosphere. Satellites capable of detecting els of astronomical objects, and prepare and disseminate
the full spectrum of radia- the results of our work.
tion—from the highest-energy Space exploration has
gamma rays and X- rays, expanded our view
through ultraviolet and infra- of the universe. FIGURE 1.6 In the 20th century, advances in telescope
red radiation, and to the technology opened new windows on the universe. This is the
lowest-energy microwaves—have brought surprising dis- sky as we would see it if our eyes were sensitive to radio
covery after surprising discovery. Each has forever altered waves, shown as a backdrop to the National Radio Astronomy
Observatory site in Green Bank, West Virginia.
our perception of the universe, further expanding the
domain of the human mind. Since the closing years of the
20th century we have witnessed a renewed vigor in astro-
nomical observations from the surface of Earth. The view
of the sky seen by radio telescopes, as shown in Figure 1.6,
illustrates the new perspectives that have been made pos-
sible by our growing technological prowess.
When we think of astronomy, telescopes—both on the
ground and in space —immediately come to mind. But
you may be surprised to learn that a great deal of frontline
astronomy is now carried out in large physics facilities
like the one shown in Figure 1.7. Today astronomers work
along with their colleagues in related fields, such as physics,
chemistry, geology, and planetary science, to sharpen our
understanding of the physical laws that govern the behav-
ior of matter and energy and to use this understanding to
make sense of our observations of the cosmos. Astronomy
has also benefited enormously from the computer revolution.
The 21st century astronomer spends far more time staring
with an idea that might explain a particular observation or

phenomenon. She presents the idea to her colleagues as a
hypothesis. Her colleagues then look for testable predictions
capable of disproving her hypothesis. This is an important
property of the scientific method, because any hypothesis
that is not falsifiable—in other words, disprovable—must
be based on faith alone. Faith serves its own purpose in
society, but it is not science.
As continuing tests fail to dis- The scientific method
prove a hypothesis, scientists involves trying to
will come to accept it as a the- falsify ideas.
ory, but never as an undis-
puted fact. Scientific theories are accepted only as long as
their predictions are borne out. A classic example is Ein-
stein’s theory of relativity, which has withstood more than
a century of scientific efforts to disprove its predictions.
Science is sometimes misunderstood because of the spe-
cial ways that scientists use everyday words. An example
is the word theory. In everyday language, theory may mean
something that is little more than a conjecture or a guess:
“Do you have a theory about who might have done it?” “My
theory is that a third party could win the next election.”
FIGURE 1.7 The Tevatron high-energy particle collider In everyday parlance a theory is something worthy of little
at Fermilab provides clues about the physical environment serious regard. “After all,” we say, “it is only a theory.”
during the birth of the universe. Laboratory astrophysics, In stark contrast, a scientific theory is a carefully con-
in which astronomers model important physical processes
structed proposition that takes into account all the relevant
under controlled conditions, has become an important part of
data and all our understanding of how the world works, and it
astronomy.
makes testable predictions about the outcome of future obser-
vations and experiments. A theory is a well-developed idea
that is ready to be confronted by nature. A well-corroborated
1.3 Science Is a Way of theory is a theory that has survived many such tests. Rather
than being simple speculation, scientific theories represent
Viewing the World and summarize bodies of knowledge and understanding that
provide our fundamental insights into the world around us.
Given the visible importance of technology in our lives, you A successful and well-corroborated theory is the pinnacle
might be tempted to say that science is technology. It is true of human knowledge about the world.
that science forms the basis of technology through a mutu- Theories fill a place in a loosely defi ned hierarchy of sci-
ally supportive relationship in which each enables advances entific knowledge. In science, the word idea has its everyday
in the other. Yet science is much more than technology. Sci- use: an idea is just a notion about how something might be.
ence can no more be reduced to its practical technological In science, a hypothesis is an idea that leads to testable pre-
application than the accomplishment of an Olympic ath- dictions. A hypothesis may be the forerunner of a scientific
lete can be reduced to utilization of the athlete’s Olympic theory, or it may be based on an existing theory, or both.
fame to market shoes and other products. Rather, we should When an idea has been thought about carefully enough, has
understand science as a fundamental way of viewing the been tied solidly to existing theoretical and experimental
world through the scientific method, a worldview that has knowledge, and makes testable predictions, then that idea
survived constant upheavals and challenges. has become a theory. Scientists build theoretical models that
are used to connect theories with the behavior of complex
systems. Ultimately, the basis for deciding among competing
The Scientific Method and theories is the success of their predictions. Some theories
become so well tested and are of such fundamental impor-
Scientific Principles tance that we come to refer to them as physical laws.
If science is more than technology, you might instead sug- A scientific principle is a general idea or sense about
gest that science is the scientific method. When you took how the universe is that guides our construction of new
science courses in high school, you probably had the scien- theories. Occam’s razor, for example, is a guiding principle
tific method drilled into you. So let’s quickly review how in science stating that when we are faced with two hypoth-
the scientific method works. Consider a scientist coming up eses that explain a particular phenomenon equally well,
we should adopt the simpler of the two. At the heart of Start with an
modern astronomy is another principle: the cosmological observation or idea
principle. Simply put, the cosmological principle states
that on a large scale the universe looks the same every-
where. That is, when we look Suggest a
out around us, what we see is There is nothing hypothesis
representative of what the special about our
universe is generally like. In place in the universe.
other words, there is nothing
Make a
special about our particular location. By extension, matter prediction
and energy obey the same physical laws throughout space
and time as they do today on Earth. This premise means
that the same physical laws that we learn about in terres-
trial laboratories can be used to understand what goes on Test does not support Perform a test, Test supports
hypothesis; revise experiment, or hypothesis; make
in the centers of stars or in the hearts of distant galaxies. hypothesis or additional additional predictions
Each new success that comes from applying the cosmologi- choose a new one observation and test them
cal principle to observations of the universe around us—
each new theory that succeeds in explaining or predicting
patterns and relationships among celestial objects—adds FIGURE 1.8 The scientific method. This is the path by which
to our confidence in the validity of this cornerstone of our an idea or observation leads to a falsifiable hypothesis that is
worldview. We will discuss the cosmological principle in either accepted as a tested theory or rejected, on the basis of
more detail in Chapter 18. observational or experimental tests of its predictions. Note that
the green loop goes on indefinitely, as scientists continue to test
the hypothesis.
Science as a Way of Knowing

The path to scientific knowledge is solidly based on the hear them using words such as insight, intuition, and cre-
scientific method. This concept is so important to your ativity. Scientists speak of a beautiful theory in the same
understanding of how science works that we should empha- way that an artist speaks of a beautiful painting or a musi-
size it once again. The scientific method consists of obser- cian speaks of a beautiful performance. Yet science is not
vation or idea, followed by hypothesis, followed by the same as art or music in one important respect: whereas
prediction, followed by further observation or experiments art and music are judged by a human jury alone, in science
to test the prediction, and ending as a tested theory, as illus- it is nature (through application of the scientific method)
trated in Figure 1.8. For all practical purposes, it defi nes that provides the fi nal decisions about which theories can
what we mean when we use be kept and which theories must be discarded. Nature is
the verb to know. It is some- All scientific completely unconcerned about what we want to be true. In
times said that the scientific knowledge is the history of science, many a beautiful and beloved theory
method is how scientists provisional. has been abandoned. At the same time, however, there is
prove things to be true, but an aesthetic to science that is as human and as profound
actually it is a way of proving things to be false. Before sci- as any found in the arts, as Figure 1.9 illustrates.
entists accept something as true, they work hard to show
that it is false. Only after repeated attempts to disprove an
idea have failed do scientists begin to accept its likely valid-
ity. For a theory to be given serious scientific consideration,
Scientific Revolutions
it must be falsifiable. That is, it must be capable of being As we have seen, science cannot be defined as either technol-
shown to be false. Scientific theories are accepted only as ogy or the scientific method, although it is related to both.
long as they are able to be tested and are not shown to be It is also incorrect to say that science is simply a body of
false. In this sense, all scientific knowledge is provisional. facts. If we limited our defi nition of science to the body of
Despite being tied so closely to the scientific method, existing facts, we would fail to convey the dynamic nature
science can no more be said to be the scientific method than of scientific inquiry. Scientists do not pretend to have all
music can be said to be the rules for writing down a musi- the answers, and we constantly have to refi ne our ideas in
cal score. The scientific method provides the rules for ask- response to new data and new insights. (This is, after all,
ing nature whether an idea is false, but it offers no insight what it means to learn.) The vulnerability of knowledge that
into where the idea came from in the fi rst place or how an is implicit in the scientific method may seem like a weak-
experiment was designed. If you were to listen to a group of ness at fi rst. “Gee, you really don’t know anything,” the
scientists discussing their work, you might be surprised to cynical student might say. But this vulnerability is actually
FIGURE 1.9 The scientific worldview is as aesthetically pleasing as that of art, music, or
literature. Unlike the arts, however, science relies on nature alone to determine which scientific
theories have lasting value.
science’s great strength. It is what keeps us honest. Once The Structure of Scientific Revolutions by Thomas Kuhn. In
an idea is declared to be “truth,” then all progress stops. In this work, Kuhn emphasizes the constant tension between
science, even our most cherished ideas about nature remain the scientist’s human need to construct a system of beliefs
fair game, subject to challenge by evidence according to the within which to interpret the world and the occasional
rules of the scientific method. Many of history’s best scien- (and likely painful) need to drastically overhaul that sys-
tists earned their place in the forward march of knowledge tem of beliefs.
by successfully goring a sacred cow. A scientific revolution is not a trivial thing. A new the-
Scientists spend most of their time working within an ory or way of viewing the world must be able to explain
established framework of understanding, extending and everything that the previous theory could, while extend-
refi ning that framework and testing its boundaries. Occa- ing this understanding to new territory into which the ear-
sionally, however, major shifts occur in the framework of an lier theory could not go. By the middle of the 19th century,
entire scientific field. Many books have been written about many physicists felt that our fundamental understanding
how science progresses, perhaps the most influential being of physical law was more or less complete. For over a cen-
is far more sublime and powerful than the earlier under-

standing that it subsumed.
As we continue on our journey, we will encounter many
other discoveries and successful ideas that forced scientists
either to abandon their treasured notions or be left behind,
hopelessly locked into a worldview that had ultimately
failed the test of observation and experiment. The point is
this: in physical science, we are not just paying lip service
to a hollow ideal when we say that the rigorous standards
of scientific knowledge respect no authorities. No theory,
no matter how central or how strongly held, is immune
from the rules.
For those of us who grew up in a world transformed
by science, the scientific worldview might seem anything
but revolutionary. Throughout much of history, however,
knowledge was sought in the pronouncements of “author-
FIGURE 1.10 Albert Einstein, perhaps the most famous ity” rather than through observation of nature. This authori-
scientist of the 20th century, and Time magazine’s selection tarian view slowed the advance of knowledge throughout
as Person of the Century. Einstein helped to usher in two western Europe for the millennium prior to the European
different scientific revolutions, one of which he himself was Renaissance, and it was largely the Chinese and Arab cul-
never able to accept. tures that kept the spark of inquiry alive during this time.
The greatest scientific revolution of all was the one that
overthrew “authority” and replaced it with rational inquiry
tury, the classical physics developed by Sir Isaac Newton and the scientific method.
had withstood the scrutiny of scientists the world over. It From the perspective of this great scientific revolution,
seemed that little remained but cleanup work—filling in the it is clear why science is more than simply a body of facts.
details. Some even went so far as to pronounce this period Perhaps more than anything, science is a way of think-
the “end of science.” Yet during the late 19th and early 20th ing about the world. It is a way of relating to nature. It is a
centuries, physics was rocked by a series of scientific revolu- search for the relationships that make our world what it is.
tions that shook the very foundations of our understanding It is a belief that nature is not capricious, but instead oper-
of the nature of reality. ates by consistent, explicable, inviolate rules. It is a collec-
If one face can be said to represent these scientific revo- tion of ideas about how the universe works, coupled with
lutions—and modern science—it is that of Albert Einstein an acceptance of the fact that what is known today may be
(Figure 1.10). Einstein’s special and general theories of superseded tomorrow. The scientist’s faith is that there is an
relativity replaced the 300-year-old edifice of Newtonian order in the universe and that the human mind is capable
mechanics, not by proving Newton wrong, but by show- of grasping the essence of the rules underlying that order—
ing that Newton’s mechanics was a special case of a far or at least of inventing ever-better approximations to those
more general and powerful set of physical laws. Einstein’s rules. The scientist’s creed is that nature, through observa-
new ideas unified our concepts of mass and energy and tion and experiment, is the fi nal arbiter of the only thing
destroyed our conventional notion of space and time as sep- worthy of the term objective truth. Science is an exquisite
arate things. Yet scientific revolutions are seldom comfort- blend of aesthetics and practicality. And, in the final analy-
able for those who live through them, and even the greatest sis, science has found such a central place in our civiliza-
of scientists can be left behind. Einstein actually helped to tion because science works.
start two scientific revolutions. He saw the fi rst of these—
relativity—through and embraced the world that it opened.
Yet Einstein was unable to accept the implications of the
second revolution he helped start—quantum mechanics—
Challenges to Science
and he went to his grave unwilling to embrace the view of Science has not achieved its prominence without criticism
the world it offered. and controversy. Philosophers and theologians of the 16th
Quantum mechanics forced us to abandon our every- century did not easily give up the long-held belief that
day understanding of “substance” and even to part with Earth is the center of all creation. In the early 20th century,
the notion that we live in a universe in which effect fol- astronomers were arguing bitterly among themselves over
lows cause in lockstep. Together, these revolutions led to the size of the universe. For centuries, conventional wisdom
the birth of what has come to be known as modern physics. had held that the Milky Way was all that there was—a vast
Although modern physics contains Newtonian physics, the collection of stars and “fuzzy” objects known as “nebulae.”
understanding of the universe offered by modern physics That belief collapsed when it was realized that some of
FIGURE 1.11 Until the early

(a) (b) 20th century, astronomers
believed that all “fuzzy” celestial
objects were located within
our Milky Way Galaxy. (a) A
planetary nebula approximately
10,000 light-years away
within our own galaxy. (b) A
spiral galaxy 110 million light-
years distant, more than
10,000 times farther than the
planetary nebula shown in (a).
these fuzzy objects were actually distant galaxies (“island ter of careful thought and effort. One of the most remark-
universes”)—similar to but far beyond the confines of our able aspects of scientific knowledge is its independence
own Milky Way (Figure 1.11). For many, it came as a rude from culture. Scientists are people, and politics and cul-
awakening that our Milky Way does not, in fact, constitute ture enter into the day-to-day
the entire universe—that it is only an infi nitesimal part of practice of science. But in Nature is the
a greatly larger universe! the end, scientific theories arbiter of science.
You might feel that science is sometimes arbitrary. For should be judged not by cul-
example, the decision as to whether Pluto should be con- tural norms but by whether their predictions are borne
sidered a planet or a dwarf planet may seem subjective. But out by observation and experiment. As long as the results
it is important to keep in mind that this controversial deci- of experiments are repeatable and do not depend on the
sion, unsupported by many astronomers, was really more a culture of the experimenter—that is, as long as there is
matter of semantics than of science. Pluto is still the same such a thing as objective physical reality—scientifi c
scientifically important Solar System body that it always knowledge cannot be called a cultural construct. Science
was. It is just that “science” has now put a new label on it. is not just one of many possible worldviews; science is
As you will see in discussions throughout this book, the the most successful worldview in the history of our spe-
path to knowledge embodied in science does not seem to be cies because the foundations of the scientific worldview
any more arbitrary than the logic it is built on. have withstood centuries of fi ne minds trying to prove
In recent years, some critics of science have drawn atten- them false.
tion to how science is influenced by culture. It is hard to It is significant that no philosopher critical of science
avoid the conclusion that political and cultural consider- has ever offered a viable alternative for obtaining reliable
ations strongly influence which scientific research projects knowledge of the workings of nature. Furthermore, no other
are funded. This choice of funding channels the direc- category of human knowledge is subject to standards as
tions in which scientific knowledge advances and can lead rigorous and unforgiving as those of science. For this rea-
to serious ethical issues. For example, moral judgments son, scientific knowledge is reliable in a way that no other
about nontraditional lifestyles greatly restricted the fund- form of knowledge can claim. Whether you want to design
ing available for AIDS research during the decade or so after a building that will not fall over, consider the most recent
its discovery. More recently, progress in stem cell research medical treatment for a disease, or calculate the orbit of a
and solutions to human-caused climate change have been spacecraft on its way to the Moon, you had better consult
hindered by widely distributed misinformation on the Web a scientist rather than a psychic—regardless of your cul-
and by constant political bickering. tural heritage.
Some critics even carry this view a step further, argu- Finally, we must admit that some disreputable scien-
ing that scientific knowledge itself is an arbitrary cultural tists purposefully try to influence results by inventing or
construct. Yet successful scientific theories are never ignoring data, often when claiming a "major breakthrough”
arbitrary. Scientific theories must be consistent with all or challenging a well-established scientific principle. For-
that we know of how nature works, and turning a clever tunately, attempts by others to repeat the experiment will
idea into a real theory with testable predictions is a mat- eventually expose such scientific misconduct.
1.4 Patterns Make the ground. The biochemistry of the food we eat remains
the same. The Sun rises, sets, and then rises again at pre-
Our Lives and dictable times. Spring turns into summer, summer turns
into autumn, autumn turns into winter, and winter turns
Science Possible into spring. The rhythms of nature produce patterns in
our lives, and we count on these patterns for our very
We experience patterns in our everyday life and we fi nd survival. If nature did not behave according to regular
comfort in them. But imagine what life would be like if patterns, then our lives—indeed, life itself—would not
sometimes when you let go of an object it fell up instead be possible.
of down. What if one day apples were essential nutrition, The patterns that make our lives possible also make sci-
but when you bit into an apple the next day you discov- ence possible. The goal of science is to identify and char-
ered that they were deadly? What if, unpredictably, one acterize these patterns and to use them to understand the
day the Sun rose at noon and set at 1:00 P.M., the next day world around us. Some of the most regular and easily iden-
it rose at 6:00 A.M. and set at 10:00 P.M., and the next day tified patterns in nature are the patterns that we see in the
the Sun did not rise at all? In fact, objects do fall toward sky. What in the sky will look different or the same a week
FIGURE 1.12 Since ancient times, our ancestors recognized that patterns in the sky change with
the seasons. These and other patterns shape our lives.
Changing patterns in the sky echo changing patterns

on Earth. In early evening on June 21, the constellations
enclosed by this line can be seen from 40° north latitude.
LUPUS
CENTAURUS LIBRA
HYDRA
SCORPIUS
VIRGO
SERPENS SERPENS
CAPUT CAUDA
OPHIUCHUS
CORVUS
BOOTES
March 21 COMA June 21
BERENICES
CRATER
HERCULES
CORONA
HYDRA CANES BOREALIS
VENATICI
SAGITTARIUS
LYRA
LEO AQUILA
MINOR
LEO SAGITTA
DRACO CAPRICORNUS
HYDRA URSA URSA
MAJOR MINOR
CYGNUS
CANCER DELPHINIUS
CEPHEUS
PYXIS Polaris
LYNX
CANIS CASSIOPEIA
PUPPIS LACERTA
MINOR
AURIGA PEGASUS
PERSEUS
ANDROMEDA PISCIS
GEMINI AUSTRINUS
MONOCEROS
TRIANGULUM AQUARIUS
CANIS PISCES
MAJOR PLEIADES ARIES
LEPUS SCULPTOR
ORION
ERIDANUS
CETUS
COLUMBA
FORNAX
December 22 September 23
1.4 Patterns Make Our Lives and Science Possible 15
FIGURE 1.13 Because mathematics helps us identify and analyze patterns in nature, it has
become the language of science.
from now? A month from now? A year from now? Most of or spoken word. It quite simply cannot be done, or at least
us probably lead an indoor and in-town existence, removed cannot be done meaningfully.
from an everyday awareness of the patterns in the sky. Away On the other hand, as the authors of this book we under-
from the smog and glare of our cities, however, the patterns stand (and as is humorously depicted in Figure 1.13) that for
and rhythms of the sky are as easy to see today as they were many of you, math is not a four-letter word—it is the four-letter
in ancient times. Patterns in the sky mark the changing of word. Many people decide early in their education that they
the seasons (Figure 1.12) and the planting and harvesting cannot “do” math, and from that day forward the mere men-
of the crops. Patterns in the sky share the rhythms of our tion of the word causes their eyes to glaze over and their
lives. It is no surprise that astronomy, which is the expres- palms to sweat. A distaste for mathematics is one of the most
sion of our human need to understand these patterns, is the common obstacles standing between a nonscientist and an
oldest of all sciences. appreciation of the beauty and elegance of the world as seen
One important tool that astronomers use to analyze these through the eyes of a scientist. To move beyond this obstacle,
patterns is mathematics. There are many branches of math- scientist and nonscientist need to find common ground.
ematics, most of which deal Part of the responsibility for moving beyond this obsta-
with more than just numbers. Mathematics is cle lies with us, the authors. It is our job to take on the role
Arithmetic is about counting the science and of translators, using words to express as many concepts
things. Algebra is about language of patterns. as possible, even when these concepts are more concisely
manipulating symbols and and accurately expressed mathematically. When we do use
the relationships between things. Geometry is about shapes. mathematics, we will explain in everyday language what the
Calculus is about change. Other types of mathematics equations mean and try to show you how equations express
include topology (the properties of surfaces) and statistics concepts that you can connect to the world. We will also
(groups of objects and their relationships). What do all these limit the mathematics to a few basic tools that all college
branches have in common? Why do we consider all of them students should have been exposed to.
to be part of a single discipline called “mathematics”? All These basic mathematical tools, as described in Math
share one thing: they deal with patterns. The best working Tools 1.1 and Appendixes 1 and 2, enable scientists to con-
defi nition of mathematics is that “mathematics is the lan- vey complex information. Scientific notation, for example,
guage and science of patterns.” enables astronomers to express in simple terms the vast
If patterns are the heart of science, and mathematics is range of sizes of astronomical objects. Scientists use a basic
the language of patterns, it should come as no surprise that set of units to distinguish among time, distance, mass, and
mathematics is the language of science. Trying to study sci- energy. Astronomers also fi nd that a bit of geometry is nec-
ence while avoiding mathematics is the practical equivalent essary for understanding the sizes, shapes, and volumes
of trying to study Shakespeare while avoiding the written of astronomical objects and the distances between them.
M AT H T O O L S 1 .1
Mathematical Tools
Mathematics gives scientists many of the tools that they need d = s × t,
to understand the patterns they see and to communicate that
where d is distance, s is speed, and t is time.
understanding to others. As the authors of this text, we are
Algebra also lets us combine these ideas with other
aware that mathematics is not a friend to many of you taking
ideas to arrive at new relationships.
this course, so we have worked to keep the math in this text
to a minimum. Even so, there are a few tools that we will need Proportionality. Often, understanding a concept amounts to
in our study of astronomy: understanding the sense of the relationships that it predicts
Scientific notation. Scientific notation is the way that sci- or describes. “If you have twice as far to go, it will take you
entists deal with numbers of vastly different sizes. Rather twice as long to get here.” “If you have half as much money,
than writing out 7,540,000,000,000,000,000,000 in standard you will be able to buy only half as much gas.” These are
notation, we express the same number in simpler form examples of proportionality.
as 7.54 × 1021. Rather than writing out 0.000000000005, A proportion relevant in astronomy is the relationship
we write 5 × 10 –12. For example, the distance to the Sun among speed, time, and distance. If you are traveling at a con-
is 149,580,000 km, but astronomers usually express it as stant speed, then time is proportional to distance. We write
1.4958 × 108 km. t ∝ d,
For a more detailed explanation of the use of scientific
notation and significant figures, see Appendix 1. where ∝ means “is proportional to.”
Proportionalities often involve quantities raised to a
Ratios. Ratios are the most common way that astronomers power. For example, a circle of radius r has an area A equal
use to compare things. A star may be “10 times as massive to πr 2, so we say that area is proportional to the square of
as the Sun” or “10,000 times as luminous as the Sun.” These the radius, and we write
expressions are ratios.
A ∝ r 2.
Geometry. To describe and understand objects in astron-
omy and physics, we use concepts such as distance, shape, This means that if you make the radius of a circle three
area, and volume. Apparent separations between objects in times as large, its area will grow by a factor of 32, or 9.
the sky are expressed as angles. Earth’s orbit is an ellipse
Units. In this book we use the metric system of units, more
with the Sun at one focus. The planets in the Solar System
properly known as the Système International d’Unités (SI).
lie close to a plane. Geometry provides the tools for work-
The United States remains one of very few countries in the
ing with these concepts.
world still using the English system of units, but gradual
Algebra. Algebra provides a way of using and manipulat- acceptance of the metric system can be seen in everyday
ing symbols that represent numbers or quantities. We will life. We buy milk in quart-sized containers, but our soft
use algebra to express relationships that are valid not just drinks come in liter-sized bottles. Bank signs often display
for a single case, but for many cases. Algebra lets us conve- temperature in both Fahrenheit (°F) and Celsius (°C), also
niently express ideas such as “the distance that you travel known as Centigrade. In some municipalities road signs
is equal to the speed at which you are moving multiplied show distances in both miles and kilometers.
by the length of time you go at that speed.” Written as an Conversions to the English units used in the United
algebraic expression, this idea is States can be found in Appendix 2.
Finally, some algebra—mostly a few ratios and propor- that we use. Do not concede defeat while still in the starting
tionalities—will provide a way of expressing the patterns blocks. It is likely that you know what it means to square a
that relate one physical quantity to another. Basic does not number, or to take its square root, or to raise it to the third
necessarily mean easy, but it does mean that we will use power. The mathematics in this book is on a par with what it
the most accessible tools that will make our journey of dis- takes to balance a checkbook, build a bookshelf that stands
covery as comfortable and informative as possible. up straight, check your gas mileage, estimate how long it
Your responsibility is to accept the challenge and make will take to drive to another city, figure your taxes, or buy
an honest effort to think through the mathematical concepts enough food to feed an extra guest or two at dinner.
1.5 Thinking like an Astronomer 17
1.5 Thinking like an simply considered astronomy to be the least of the avail-
able evils!) Whatever your expectations, the story in 21st
Astronomer Century Astronomy can fascinate you if you open your
mind to it.
Not everyone is fascinated by mathematics or science, but Our best suggestion for a successful journey through
almost everyone harbors a spark of interest in astronomy. 21st Century Astronomy is to nurture your spark of interest
Because you are reading this book, you probably share this in astronomy until it grows enough to draw you in. When,
spark as well. The spark may have been struck when you as Einstein phrased it in the opening quote, you “pause to
were a child looking at the sky and found yourself won- wonder and stand rapt in awe”—then you will have learned
dering about what you saw there. The prominence of the the secret to thinking like a scientist!
Sun, Moon, and stars in cave paintings and rock drawings Reading the book and taking the journey with us are
(such as those in Figure 1.14) dating back thousands of years two different things. This is only a guidebook. It can lead
tells us that these questions have long occupied the human you to the trailhead and tell you something of what you
imagination. Your initial spark of interest in astronomy may might fi nd along the way, but you have to walk the path! If
have grown over the years as you saw or read news reports you become an active participant in this adventure, rather
about spectacular discoveries made in your lifetime. Some than a passive spectator, then what you gain from the jour-
of these discoveries may have sounded so amazing that it ney will remain a part of you long after the fi nal exam is
was difficult to draw the line between science fact and sci- forgotten. And if along the way you find yourself applying
ence fiction. your understanding to new situations and new informa-
If you are like many other people, your conception of tion, and if you learn to combine your understandings and
astronomy so far may not have gone much beyond learning arrive at new insights that are greater than the sum of their
about the constellations and the names of the stars in them. parts, then you will have learned something far more than
But if you nurture your spark of interest in astronomy, you just astronomy.
may be surprised to fi nd that spark growing into a flame This book will likely ask you to think in ways that are
that will fuel your journey through the chapters that fol- different from the ways in which you are accustomed to
low. This book will take you to places you never imagined thinking, and to learn to view the world from new and unfa-
going and will lead you to insights and understandings miliar perspectives. Knowledge and understanding have
you never imagined having. To those of you who are read- nothing to do with shoving facts into short-term memory
ing this book for a course in astronomy at your college or so that they can be regurgitated on an exam, and such a
university, we have a special note: The authors of this text short-term strategy is certainly not how astronomy—or any
have taught introductory astronomy many times over the other field of study, for that matter—should be approached.
years. We recognize that you may be in this course pri- Changing the way that you think takes more effort. It also
marily because you need a science credit to graduate. As means studying in ways that may be different from your
you flipped through your course catalog, perhaps you were normal habits. Above all, remember that building under-
reminded of your interest in astronomy, and that led you standing is always an active process, never passive! Here
to choose astronomy over other options. (Or perhaps you are a few practical suggestions for how you might better
study this text:
FIGURE 1.14 Ancient petroglyphs often include depictions of

• Read the text actively. Think about each section after
the Sun, Moon, and stars. you have read it. What major concepts were discussed
in the section? How are they related to what you have
read so far? Have you run into similar concepts else-
where? Why are the contents of that section impor-
tant enough to be included in the book? In your class
notes, briefly summarize the section and your thoughts
about it.
• Visualize the concept. Many physical and mathematical
concepts are most easily understood if they are visual-
ized. Review the photos, diagrams, and charts in each
chapter to study key concepts. Draw a picture of the con-
cept. If you understand a concept well enough to draw a
picture that expresses it, then you probably understand
the concept fairly well. Trying to sketch a picture will
also help you better identify what things you understand
and what things you do not.
• Ask yourself “What if?” when trying to understand a work through the appendixes or explore StudySpace
concept. What if Earth were more massive? How would (www.norton.com/studyspace) for other aids available
that affect Earth’s gravity? What if the Sun were hotter? to you. Your instructor is also there to help.
How would that affect the color of the light from the
Sun or the amount of energy that the Sun radiates? If You will occasionally find material set aside in boxes.
you cannot “what if” a concept, then you probably do A topic has been boxed either because it is somewhat out
not really understand it yet. of the mainstream of our journey or because we wish to
highlight it. In particular,
• Try to teach it. It is often said that you do not really
understand something until you try to explain it to
• Foundations boxes address material that is central to our
someone else. At the end of a reading assignment, talk
understanding of the physical nature of the universe.
about the material with a friend or family member. Try
to fi nd a partner or a group in your class with whom • Math Tools boxes provide additional math information
you can meet, and take turns teaching each other the or worked examples of algebraic expressions.
material. Each of you should read the assigned text,
• Connections boxes draw attention to recurring themes—
and then the group should divide up responsibility for
bridges between different parts of our journey.
who will present which sections. Ask each other lots
of questions—the harder the better! (Make a game of • Excursions boxes are short but interesting side trips.
“stump the instructor.”) Even explaining a concept out
loud to yourself can help. Throughout each chapter you will find Trailmarkers,
one-sentence summaries of fundamental concepts. They are
• Share your ideas and insights with your discussion
inserted in the text, set off by light green horizontal bars.
group. If a particular concept really “clicks” for you—if
As you study the pages of this book—in concepts and
you really think it is cool—share both your understand-
images—you will see that we truly live in a golden age of
ing and your enthusiasm with your group.
exploration and discovery. We can be certain that future
• Be honest with yourself about what things you under- historians will remember ours as the time when human-
stand and what things you do not, and try not to avoid kind fi rst stepped beyond the world of our birth and began
concepts that you find difficult. Getting the most from to reach out with our minds and our science to touch the
this journey will come from facing and mastering these fabric of the universe itself. It is probably safe to say that few
challenging concepts. You will find that personal growth things will have a more lasting impact on our culture than
comes from this real accomplishment. this revolution in our understanding of the universe and
our place in it. What has yet to be determined is whether
• Focus your discussion on concepts, relationships, and
future historians will remember us for reaching out to touch
connections. You have to know the facts, but the facts
the universe and embracing what we found, or whether we
are the starting point, not the end. Use the key concepts,
will instead be remembered for a loss of spirit—for step-
study questions, and other study aids to identify and
ping back and turning away from the frontier of explora-
concentrate your effort on the most important ideas.
tion and discovery. The direction that we take from here is
• Do not let discomfort with math keep you from suc- a decision in which you—and your willingness to embrace
ceeding. Mathematical formulas are not magical incan- science—will play a part.
tations. Rather, they are expressions of logical ideas. The journey of discovery on which we are embarking is
We will always present a plain-English discussion of not always easy, but few worthwhile journeys are. A hike
the idea behind any mathematics that we use. Begin by in the mountains can at times be an easy stroll and at other
focusing on this discussion. After you grasp the idea, times a more strenuous climb; but when you arrive, the view
then look at the math and any equations presented in the from the top is hard to beat. In much the same way, this book
separate “Math Tools” boxes. Try to see how the relation- will ask you to exercise your mental muscles in different,
ships between the quantities in the equation embody the possibly unaccustomed ways. But, as with the hike in the
concept and try to follow along with the worked exam- mountains, we feel certain that you will fi nd the rewards
ples, when given. If you need help with basic math skills, worth the investment.
ing pace. This progress comes courtesy of a great many

Seeing the Forest for the Trees advances both in our technology and in the sophistica-
tion of our physical understanding of matter and energy
At the end of each chapter of 21st Century Astronomy, you and of space and time themselves. We have seen many
will fi nd a brief narrative about the content of that chap- fundamental questions about the origin and fate of the
ter. The purpose of “Seeing the Forest for the Trees” is universe—and about the threads that tie our existence
not to rehash the entire contents of the chapter, but rather to the cosmos—move from the realm of philosophical
to pick out a few of the high points and put them into a speculation into the realm of rigorous scientific inquiry.
broader context. Building on the analogy of a hike in the The insights that this age of exploration and discovery
mountains, we will spend a lot of time looking in detail have brought are often far more profound and startling
at the rocks and the trees, but every so often we need to than dreamed of even a few decades ago. There can be
step back and look around at the forest as a whole. little doubt that this time will be looked on as one of the
We live in a world that has been profoundly shaped by more significant moments in the intellectual and cultural
the scientific revolution that took hold of Western thought history of our species.
during the Renaissance. That revolution fundamentally Like a hike through the mountains, 21st Century
altered our way of thinking about the world, as well as Astronomy will not be an effortless journey. Many travel-
our view of the relationship between ourselves and the ers will fi nd that they have to flex a few mental muscles
universe of which we are a part. A new spirit of rational in ways they are not used to, and they may even have
inquiry was turned on the heavens, dislodging Earth to face an old adversary or two on the trail. But muscles
and humankind from the center of the cosmos. Obser- that are sore after the fi rst day of a hike grow comfort-
vation, experiment, and rigorously applied reason came able and strong with time, and adversaries can become
to replace dogma and authority as the arbiter of knowl- the best of friends.
edge. The heavens became a realm not of mysticism and In Chapter 2 we begin the journey in earnest and, as
magic, but instead of physical law—the same physical with most journeys, our starting point is home. What
law that governs the behavior of matter and energy in patterns do we see in the skies of our planet Earth, and
laboratories here on Earth. how do those patterns come to be? This is not an easy or
The closing years of the 20th century saw our knowl- gentle slope on which to begin our trek, but our vistas
edge of the universe charge ahead at an ever-accelerat- will change rapidly as we climb.
Summary SmartWork, Norton’s online homework system,

includes algorithmically generated versions of these
questions, plus additional conceptual exercises. If
• Our Solar System is but a tiny speck in a vast universe.
your instructor assigns questions in SmartWork, log in at
• There is nothing special about our particular place in smartwork.wwnorton.com.
the universe.
• We, and everything around us, are composed of atoms
that were formed in stars whose lives ended long ago.
• Space exploration has expanded our view of the Student Questions
universe.
THINK ING A BOUT THE CONCEP T S
• The scientific method is a way of trying to falsify, not
prove, ideas. 1. Suppose you lived on imaginary planet Zorg orbiting
Alpha Centauri, a nearby star. How would you write
• All scientific knowledge is provisional.
your cosmic address?
• The entire universe is governed by the same physical
2. Without looking back in the text,
laws that shape our lives here on Earth.
a. How many of the eight classical planets can you
• Mathematics is the science and language of patterns, name?
and thus it is the language of science. b. How many dwarf planets?
3. Assuming that the number of stars in all the dwarf gal- or website without regard to your own sign. How many
axies in our Local Group is negligible compared to the of them might fit the day that you had yesterday? Repeat
number of stars in our Milky Way Galaxy, estimate the the experiment every day for a week and keep records.
total number of stars in our Local Group. Was your horoscope sign consistently the best descrip-
tion of your experiences?
4. If the Sun suddenly exploded (it can’t), how soon after
would we know about it? 17. A scientist on television states that it is a known fact that
life does not exist beyond Earth. Would you consider
5. If a star exploded in the Andromeda Galaxy (they some-
this scientist reputable? Explain your answer.
times do), how long would it take that information to
reach Earth? 18. Some astrologers use elaborate mathematical formulas
and procedures to predict the future. Does this show
*6. It is said that we are made of stardust. Explain why this
that astrology is a science? Why or why not?
statement is true.
*19. Imagine yourself living on a planet orbiting a star in a
7. A friend tells you that the reason astronomers put tele-
very distant galaxy. What does the cosmological prin-
scopes in space is to get closer to the planets and stars.
ciple tell you about the way you would perceive the uni-
You know better. How do you explain the real reason
verse from this distant location?
to your friend?
20. You run across an old newspaper with the headline
8. Your doctor can look for broken bones using X-rays,
“EINSTEIN PROVES NEWTON WRONG!” Did the newspaper
but an astronomer working with a telescope at your
get this story right? Explain your answer.
nearby observatory cannot see X-rays coming from a
star. Explain why not. 21. List patterns in your own life that repeat regularly. How
do these patterns affect you? Which patterns are of your
9. List three nonastronomical scientific disciplines that have
own making, which are set by others, and which are
contributed to our modern knowledge of astronomy.
determined by nature?
10. Controversial author Erich von Däniken proposed the
22. Name some supermarket items that are sold by the liter
theory that Earth was visited by extraterrestrials in
instead of by the quart.
the remote past. Would you regard this as scientific or
pseudoscientific theory? Is the theory falsifiable? Can
you think of any tests that could support or refute the
theory?
A P P LY I N G T H E C O N C E P T S
11. The scientific method requires that scientific theories be
23. (a) If it takes about 8 minutes for light to travel from the
falsifiable. List some beliefs or views that you conclude
Sun to Earth, and Pluto is 40 times as far from us, how
are not falsifiable.
long does it take light to reach Earth from Pluto? (b) Radio
*12. Explain how the word theory differs in meaning when waves travel at the speed of light. What does this fact
used in common everyday language and when used by imply about the problems you would have if you tried
a scientist. to conduct a two-way conversation between Earth and
a spacecraft orbiting Pluto?
13. What is the difference between the terms hypothesis and
theory as used by scientists? *24. Astronomers fi nd it useful to use the speed of light and
the time it takes light to travel a given distance as the
14. The tabloid newspaper at your local supermarket theo-
basis for discussing astronomical distances. Use the
rizes that, compared to average children, children born
travel time of light and more familiar units, such as miles
under a full Moon become more intelligent students.
(you can approximate 1 mile = 1.6 km), to illustrate this
a. Is this theory falsifiable?
point when referring to the diameter of the Milky Way
b. If so, how could it be tested?
Galaxy and the distance to the Andromeda Galaxy.
15. A textbook published in 1945 stated that it takes 800,000
*25. Imagine the Sun to be the size of a grain of sand and
years for light to reach us from the Andromeda Galaxy.
Earth a speck of dust 83 millimeters (mm) away. (On this
In this book we say that it takes 2,500,000 years. What
scale, each light-minute of distance equals 10 mm.) Look
does this tell you about a scientific “fact” and how our
at Figure 1.2. What would be the distance from Earth
knowledge evolves with time?
to the Moon on this scale? From the Sun to Neptune?
16. Astrology makes testable predictions. For example, it From Earth to the nearest star? To the nearest large gal-
predicts that the horoscope for your star sign on any day axies? (Note that 1 meter = 103 mm and that 1 km = 103
should fit you better than do horoscopes for other star meters.) At what point do you lose your “feel” for these
signs. Read each of the daily horoscopes in a newspaper distances?
26. The average distance from Earth to the Moon is 384,000 example, the price of a bag of apples is proportional to the
km. How many days would it take, traveling at 800 weight of the bag of apples.) For each proportionality, iden-
kilometers per hour (km/h)—the typical speed of jet tify the constant of proportionality (such as the price per
aircraft—to reach the Moon? pound of apples). How are these constants determined?
27. The surface area of a sphere is proportional to the square 34. A pizzeria offers a 9-inch-diameter pizza for $12 and an
of its radius. The radius of the Moon is only about one- 18-inch-diameter pizza for $24.
quarter that of Earth. How does the surface area of the a. Are both offerings equally economical?
Moon compare with that of Earth? b. If not, which is the better deal? Show why you think
so.
28. Write 86,400 (the number of seconds in a day) and 0.0123
(the Moon’s mass compared to Earth’s) in scientific 35. The circumference of a circle is given by C = 2πr, where
notation. r is the radius of the circle.
a. Calculate the approximate circumference of Earth’s
29. Write 1.60934 × 103 (the number of meters in a mile) and
orbit around the Sun, assuming that the orbit is a
9.154 × 10 –3 (Earth’s diameter compared to the Sun’s) in
circle with a radius of 1.5 × 108 km. You can approxi-
standard notation.
mate π as being about equal to 3.
*30. The time (t) it takes for light to reach us from a distant b. Noting that there are 8,766 hours in a year, how fast,
galaxy is equal to the distance (d) of the galaxy divided in kilometers per hour, does Earth move in its orbit?
by the speed of light (c). Use algebra to describe this c. How far along in its orbit does Earth move in one
relationship more simply. day?
31. A remote Internet Web page may sometimes reach your 36. Gasoline is sold by the gallon in the United States and
computer by going through a geostationary satellite by the liter nearly everywhere else. There are approxi-
orbiting approximately 3.6 × 107 meters above Earth’s mately 3.8 liters in a gallon. If gas cost $4 per gallon,
surface. What is the minimum delay, in seconds, that how much would it cost per liter?
the Web page takes to reach your computer?
37. There are 4 quarts in a gallon. Using the algebraic expres-
*32. Some people have a theory that a tray of hot water will sion, 1 quart = C × 1 gallon. What is the constant of pro-
freeze more quickly than a tray of cold water when both portionality, C?
are placed in a freezer.
a. Does this theory make sense to you?
b. Is the theory falsifiable?
StudySpace is a free and open Web site that pro-
c. Do the experiment yourself. Note the results. Was
vides a Study Plan for each chapter of 21st Century
your intuition borne out?
Astronomy. Study Plans include animations, read-
33. If you understand proportionality, then you understand ing outlines, vocabulary flashcards, and multiple-choice
most of the math you need to follow this text. Make a list quizzes, plus links to premium content in SmartWork and
of five different proportionalities from your daily life. (For the ebook. Visit www.norton.com/studyspace.
. . . marking the conclave of all the night’s stars,
those potentates blazing in the heavens
that bring winter and summer to mortal men,
the constellations, when they wane, when they rise.
AESCH Y LUS (525 – 456 B . C .)
The Moon rising over the ruins of the Greek Temple of Poseidon.
CHAP TER 2
Patterns in the Sky—

Motions of Earth
2.1 A View from Long Ago
K E Y CONCEP T S
The herds have reached the high meadows where they spend In this chapter we begin our journey in earnest—
the warm season, and for a time the life of the tribe has
starting out as our ancestors did when they first gazed
settled in as well. The weather is comfortable and the days
at the Sun, Moon, and stars and tried to understand
are long, with none of the hardships that accompany the
time of cold and snow and long, dark nights. It is a time of what they saw. With the benefit of knowledge hard-won
plenty and a time for telling the age-old stories of the tribe over the centuries, we will look at patterns present
around a fire that guards against the chill of the gathering both in the sky and on Earth, and then look beyond
night. You feel a sense of contentment and are thankful to appearances to the underlying motions that cause
the gods for this time when life is good. As the embers die those patterns. Here we will discover
down, you gaze upward at the familiar canopy of stars over- • How the stars appear to move through the sky as
head and, as you often do in such moments, you wonder Earth rotates on its axis, and how those motions differ
about what you see.
when seen from different latitudes on Earth.
To survive, you must learn the subtle patterns of your
world. You must know the ways of the herds and recognize • The fundamental concept of a frame of reference, and
the gathering of clouds that heralds a coming storm. When how Earth’s rotating frame of reference affects our
you turn your keen eye toward the heavens, you find subtle perception of celestial motions.
and changing patterns there as well—patterns that some- • How Earth’s motion around the Sun and the tilt of
how echo those of your life. The spirits of the great animals Earth’s axis relative to the plane of its orbit combine
dwell in the sky; as a child, you learned to recognize their
to determine which stars we see at night and which
pictures there. Above you now are the stars that rule the
seasons we feel through the year.
time of the short nights. These are the stars that bring sum-
mer and lead the herds to this pleasant place. • The motion of the Moon in its orbit about Earth, and
Some of the spirits of the sky can be difficult to please. how that motion, together with the motion of Earth and
The mischievous planets wander from place to place, using the Moon around the Sun, shapes the phases of the
their fearsome powers to sow chaos through the heavens. The Moon and the spectacle of eclipses.
Moon sometimes turns blood red, and the Sun is consumed
by an ominous beast. But as long as the tribe remembers
them, the gods and spirits of the sky will continue to bring
the seasons and send the stars to guide the tribe.
23
24 Chapter 2 Patterns in the Sky—Motions of Earth
Egyptian—1275 B.C. Chinese—940 A.D. European—1540 A.D.
FIGURE 2 .1 The region of the sky around what we now call the Big Dipper (Ursa Major, or the
“Great Bear”) as viewed by three different civilizations. Constellations are groupings of stars whose
images, such as a dipper or bear, exist only in the human mind.
Our ancestors lived their lives attuned to the ebb and flow in Figure 2.1, there have been as many different sets of con-
of nature, and the patterns in the sky are a part of that ebb stellations and stories to go with them as there have been
and flow. The coming of night and day, the changing of the cultural traditions in human history. But where are these
seasons, the rising and falling of the tides, the movement winged horses, dragons, chained maidens, and other imagi-
of the herds—all of these march in lockstep with the changes nary images formed from patterns of stars? The answer may
that we see in the sky. The repeating patterns of the Sun, seem obvious: “The constellations are overhead in the sky,
Moon, and stars echo the for all to see.” Yet if you look at the sky, no clear pictures
rhythms that have defined the Patterns in the sky of these images emerge. Instead there is only the random
lives of humans since before have always been pattern of stars—about 5,000 of them visible to the naked
the beginning of recorded his- important to our eye—spread out across the sky.1 Constellations exist only
tory. By watching the patterns species. within the imagination of the human mind. They are the
in the sky, our ancestors found ideas and pictures that humans impose on the lights in the
that they could predict when the seasons would change and sky in an effort to connect our lives on Earth with the work-
the rains would come and the herds would move. Knowl- ings of the heavens.
edge of the sky offered knowledge of the world, and knowl- Modern constellations visible from the Northern
edge of the world was power. Hemisphere draw heavily from the list compiled 2,000
It was a small step from here to thinking of the unreach- years ago by the Alexandrian astronomer Ptolemy. Con-
able, untouchable stars not only as a reflection of the pat- stellations in the southern sky are drawn from the lists
terns in the world, but also as the cause of those patterns. put together by European explorers visiting the Southern
The stars found a special place in legend and mythology Hemisphere during the 17th and 18th centuries. Today
as the realm of gods and goddesses, holding sway over the astronomers use an officially sanctioned set of constel-
lives of humankind. As writing came to replace oral tra- lations as a kind of road map of the sky. The entire sky is
ditions and legends, mythologies of the sky became more broken up into 88 different constellations, much as conti-
elaborate as well. And as humans invented numbers and nental landmasses are divided into countries by invisible
mathematics to describe and predict and account for things lines. Every star in the sky lies within the borders of a
in the world, predictions of the motions of the stars and single constellation, and the names of constellations are
planets were among their greatest successes. Some of our used in naming the stars that lie within their boundar-
ancestors came to look upon the orderly and predictable ies. For example, Sirius, the brightest star in the sky, lies
patterns of the sky as the true patterns of the world, and our within the boundaries of the constellation Canis Major
own lives as imperfect reflections of this heavenly reality. (meaning the “big dog”). Sirius’s official name is therefore
They looked for ways to use their knowledge of the sky to Alpha Canis Majoris (Canis Majoris is the Latin genitive,
fi nd order in the seeming chaos of their everyday lives, and or “possessive,” form—see Appendix 6), indicating that
astrology was born. it is the brightest star in that constellation and earning it
Elements of this same basic history played themselves its nickname, the “Dog Star.” Appendix 5 contains a list
out many times over and in every part of our globe. From
Africa to Asia, from Europe to Central America, from North 1
Although stars may appear close together in the sky, most are not
America to the British Isles, the archaeological record holds close together in space. The night sky is fi lled both with nearby low-
evidence of early humans who projected ideas from their luminosity stars and with others that are much more luminous but
own cultures onto what they saw in the sky. As illustrated much farther away.
of the nearest and brightest stars. Appendix 6 provides airplane high overhead and are caught off guard by the
sky maps showing the constellations. blazing spectacle of the summer Milky Way and the thou-
The connection between the patterns in the sky and the sands of pinpoints of light, which seem so close that you
patterns in the lives of the ancients was simply too compel- can almost reach out and touch them. For a moment, the
ling to be missed. The idea of the sky as a realm of mysti- thousands of years separating you from a long-dead tribal
cism and magic is deeply rooted in the traditions and beliefs nomad vanish as you share the same sense of wonder and
and history of our species. There is no mystery about the awe that has always defined humankind’s experience of
currents of mind that led our ancestors to their belief in the universe.
astrology and other celestial mythologies. At a time when
the causes of things were unknown, and humans existed at It is here that we begin the journey of 21st Century
the seeming whim of forces they could not comprehend, the Astronomy—with the changing patterns in the sky that cap-
sky seemed to offer a window into a mystical and powerful tured the attention and imagination of that long-ago nomad
world of spirits, gods, devils, and angels. and that still serve as beacons overhead on dark, cloudless
What an unwelcome shock it must have been when a nights. Yet unlike that nomad, we look on those chang-
few remarkable individuals, with names such as Coperni- ing patterns with the perspective of centuries of hard-won
cus and Kepler and Galileo and Newton, tugged at the knowledge. We will find that patterns of change in the sky
threads of this comfortable and familiar tapestry, only to are often the understandable and even unavoidable con-
discover that it fell apart in their hands! The stars and other sequences of the daily rotation of Earth about its axis and
heavenly bodies do not rotate about Earth each day, as Earth’s annual trip around the Sun. The discovery of won-
humans had thought since they fi rst took notice of the sky. derful variety arising from simple and elegant underlying
Rather, it is Earth that spins causes is an example of science at its best. And just as hap-
on its axis, giving the stars, Science shattered pened over the history of our species, curiosity about the
planets, Sun, and Moon the ancient mystical views changing patterns in the sky will show us the way outward
appearance of following daily of the heavens. into a universe far more vast and awesome than our distant
paths through the heavens. ancestor could have imagined.
Nor is Earth at the center of all existence, as befits the home
of humankind, “the pinnacle of all creation.” Earth is just
one of eight major planets orbiting the Sun. The subtle com-
plexity of the changing patterns we see in the sky results
from the motions of planets and moons as they step through 2.2 Earth Spins on Its Axis
their gravitational dance with the Sun. Even the Sun itself,
whose radiant energy makes our world what it is, is only When our remote ancestors first noticed the sky with
one of countless stars, adrift in a universe whose full extent something approaching human awareness, it was doubt-
is unknown even today. less the daily motion of the Sun in the sky that drew
The magic of astrology properly belongs to a time long their attention. As civilizations became more aware of
dead, when in the minds of humans Earth rode on the back the often complex motions of the Sun, Moon, planets,
of a giant sea turtle, and with each passing month the Sun and stars, early astronomers developed models to explain
moved from one stellar “house” to the next. Today it is a what they saw. The ancient Greeks, for example, devised
matter of experimentally verifiable fact that the imaginary geocentric models of the universe in which heavenly
patterns seen in the stars hold no more influence over our bodies were embedded in a celestial sphere that revolved
lives than do the random patterns of leaves blowing down around a stationary Earth. The most successful was
the street on an autumn day. The astrologers’ quest for deep Ptolemy’s, which survived for nearly 1,500 years until
connections between our lives and the patterns in the sky finally overthrown by Copernicus in the early Renais-
was both understandable and well placed, but knowledge sance (see Chapter 3.)
of the true nature of those connections had to wait for the Despite the apocryphal stories you may have learned
birth of modern science. Today the sky has become a win- in grade school, Christopher Columbus did not discover
dow of knowledge on the physical world. This knowledge that the world is round. Long before his famous (or possi-
has proven worth the wait. bly infamous) journey to the New World, anyone who had
read Aristotle or other Greek philosophers (as had Colum-
Today you had to make a long, hard climb before bus) knew that Earth is a ball. Far more difficult to accept
reaching the meadow by the river where you now camp, was the idea that the changes occurring in the sky from day
and tomorrow’s trek promises to be just as demanding. to day and month to month are the result of the motions of
Even so, the evening is pleasant, and you are content. The Earth rather than the motion of the Sun and stars around
embers of your campfire have almost died away, when Earth. The most apparent among these motions is Earth’s
the distant sound of a jetliner interrupts your reverie. rotation on its axis, which sets the very rhythm of life on
Without really thinking, you look up to catch sight of the Earth—the passage of day and night.
from Earth’s surface, the path each celestial body

The Celestial Sphere Is a Useful Fiction makes across the sky is called its “apparent daily motion.”
One reason the ancients did not believe that Earth rotates AstroTour: The Earth Spins and Revolves
is that they could not perceive the spinning motion of Earth. To help visualize the apparent daily motions of the Sun
In fact, as a result of Earth’s rotation, the surface of Earth and stars, it is sometimes useful to think of the sky as if it
is moving along at a respectable speed—1,674 kilometers were a huge sphere with the stars on its surface and Earth
per hour (km/h) at the equator (which we calculate by divid- at its center. (As we have said, from ancient Greek times
ing the circumference of Earth by the period of its rotation). to the Renaissance, most people believed this to be true.)
Even so, we do not feel that motion any more than we would Astronomers refer to this imaginary sphere as the celestial
“feel” the speed of a car with a perfectly smooth ride cruis- sphere (Figure 2.3a).2 The celestial sphere is a useful concept
ing down a straight highway. Nor do we directly sense the because it is easy to draw and visualize, but never forget that
direction of Earth’s spin, although it is clearly revealed by it is imaginary! Each point on the celestial sphere actually
the hourly motion of the Sun, corresponds to a direction in space. The direction in which
Moon, and stars across the Earth’s rotation is Earth’s axis of rotation points, and about which the stars
sky. As viewed from above counterclockwise appear to revolve as Earth turns, is called the north celes-
Earth’s North Pole, Earth when viewed from tial pole (NCP). The direction in space that is at the zenith
rotates in a counterclockwise above the North Pole. at the South Pole, and about which everything appears to
direction (Figure 2.2), com- spin as Earth rotates, is the south celestial pole (SCP).
pleting one rotation in a 24-hour period. As the rotating We divide the celestial sphere into a northern half and
Earth carries us from west to east, objects in the sky appear a southern half with an imaginary circle called the celestial
to move in the other direction, from east to west. As seen equator. Just as the north celestial pole is the projection of
the direction of Earth’s North Pole into the sky, the celestial
equator is the projection of the plane of Earth’s equator into
FIGURE 2 . 2 The rotation of Earth and the Moon, the the sky. And just as Earth’s North Pole is 90° away from
revolution of Earth and the planets about the Sun, and the orbit Earth’s equator, the north celestial pole is always 90° away
of the Moon about Earth are counterclockwise as viewed from from the celestial equator. If
above Earth’s North Pole. (Not drawn to scale.) you point one arm toward a The celestial equator
point on the celestial equator is the projection
and one arm toward the north of Earth’s equator
Mercury celestial pole, your arms will into space.
always form a right angle. If
Sun Moon you are in the Southern Hemisphere, the same holds true
Venus Earth Mars there: the angle between the celestial equator and the south
celestial pole is 90° as well.
The path that the Sun takes along the celestial sphere is
called the ecliptic. This imaginary circle is inclined 23.5° to
the celestial equator (see Section 2.3). To see how the concept
When viewed from Earth’s North Pole, of the celestial sphere can be used, let’s consider the Sun
planetary orbits—as well as the rotation at noon and at midnight. From our perspective on Earth,
of most planets and the orbits of most and as modeled in the celestial sphere, the Sun appears to
moons—are counterclockwise.
move across the sky and reach its highest point in the sky at
noon. In more precise terms, astronomers defi ne true “local
Rotation
of Moon noon” as the time when the Sun at our location crosses the
meridian, an imaginary north–south, 360° great circle that
divides the sky evenly into eastern and western halves.
Rotation The meridian runs from a point due north on the horizon,
of Earth through a point directly overhead called the zenith, to a
point due south on the horizon, and then continues below
North Pole the horizon through the nadir, a point directly below and
opposite the zenith, and back up to its starting point due
north on the horizon (Figure 2.3b). True “local midnight”
Earth’s orbit Moon’s orbit occurs when the Sun again crosses the meridian at its low-
est point below the northern horizon. What is really hap-
2
You can fi nd a description of celestial coordinates used with the
celestial sphere in Figure A6.1 in Appendix 6.
The celestial poles Stars all appear to lie

(both north and south) on the celestial sphere,
are directly above but really they lie at
(a) (b)
Earth’s poles. different distances.
FIGURE 2 . 3 (a) The celestial
North celestial pole Zenith sphere is a useful fiction for

thinking about the appearance
and apparent motion of the
stars in the sky. (b) Principal
features of an observer’s
coordinate system projected
N
Meridian
onto the celestial sphere.
W E
ic
ipt
Ecl 23.5°
Hor
Celest izon
i al Equator S
South celestial pole Nadir
The ecliptic is the Sun’s The celestial equator

apparent annual path is a projection of Earth’s
around the celestial sphere. equator into space.
pening at noon is that our location on Earth has rotated to ing at the center of a rotating carousel. As Earth rotates,
face most directly toward the Sun. Half a day later, at mid- the spot directly above you seems to remain fi xed while
night, our spot on Earth rotates closest to facing directly everything else in the sky appears to revolve in a coun-
away from the Sun. terclockwise direction around this spot (Figure 2.4b). (If
We will continue to use the celestial sphere model you are having trouble visualizing this, fi nd a globe and,
throughout the chapter to show apparent motions of the as you spin it, imagine standing at the pole of the globe.)
Sun and stars from different points on the Earth. Notice that objects close to the pole appear to follow
small circles, while the largest circles are followed by
objects nearest to the horizon. AstroTour: The View
from the Poles
The View from the Poles The view from the North Pole is unique because from
The apparent daily motions of the stars and the Sun wit- there we always see the same half of the sky (Figure 2.4c).
nessed by ancient nomadic tribes would have depended on Nothing rises or sets as Earth turns beneath you. Of course,
where on the surface of the planet they happened to live. regardless of where you are on the surface of our planet,
The apparent daily motions of celestial objects in northern you can never see more than half of the sky at any one time.
Europe, for example, are quite different from the apparent The other half of the sky is blocked from view by Earth. The
daily motions seen from a tropical island. To understand boundary between the part of the sky you can see and the
how our location affects our perception of apparent daily part that is blocked by Earth
motions of celestial bodies, let’s look at the daily motions is called the horizon. From The same half of the
of the stars when viewed from a place where humans did most locations on Earth, the sky is always visible
not set foot until the early 20th century—Earth’s North half of the sky that we can see from the North Pole.
Pole. (In science we often start by working out the “easy” or above the horizon changes
“limiting” cases—the view of the stars from the poles, for constantly as Earth rotates. (The direction in space in which
example—and then use these to guide our thinking about our zenith points right now is different from what it was
what happens in more complicated situations.) 12 hours ago, or even 12 seconds ago.) In contrast, Earth’s
Imagine that you are standing on the North Pole watch- North Pole points in the same direction, hour after hour
ing the sky as shown in Figure 2.4a. (Ignore the Sun for and day after day. For this reason, if you look off toward
the moment and suppose that you can always see stars in the horizon, you will see that the objects visible there fol-
the sky.) You are standing where Earth’s axis of rotation low circular paths that keep them always the same distance
intersects its surface, which is much the same as stand- above the horizon.
This disk represents the horizon, the boundary

between the part of the sky you can see and
the part that is blocked from view by Earth.
(a) (b) (c)
North celestial pole (NCP) Always visible
North Pole
Earth
North celestial
pole at the zenith
Equator
South Pole Never visible
From the North Pole looking directly overhead, As Earth rotates, the stars appear to revolve in From the North Pole, you always see the same
the north celestial pole is at the zenith. a counterclockwise direction around the NCP. half of the sky.
FIGURE 2 . 4 As viewed from Earth’s North Pole (a), stars move throughout the night on
counterclockwise, circular paths about the zenith (b). (c) The same half of the sky is always visible
from the North Pole.
The view from Earth’s South Pole is much the same, but north or south we are on the face of Earth. Imagine a line
with two major differences. First, the South Pole is on the from the center of Earth to your location on the surface
opposite side of Earth from of the planet. Now imagine a second line from the center
the North Pole, so the half of From the South Pole, of Earth to the point on the equator closest to you. (Refer
the sky you see overhead at the other half of the to Figure 2.5 for help imagining these lines.) The angle
the South Pole is precisely the sky is visible, and between these two lines is your latitude. At the North
half that is hidden from the stars circle in a Pole, for example, these two imaginary lines form a 90°
North Pole. The second differ- clockwise direction. angle. The latitude of the North Pole is thus 90° north of
ence is that instead of appear- the equator, which serves as the 0° mark. The South Pole
ing to move counterclockwise around the sky, stars appear is at latitude 90° south.
to move clockwise around the south celestial pole. (To see Our latitude determines the part of the sky that we can
this, sit in a swivel chair and spin it around from right to see throughout the year. As we follow the curve of Earth
left. As you look at the ceiling, things appear to move in a south from the North Pole, our horizon tilts and our zenith
counterclockwise direction; but as you look at the floor, moves away from the north celestial pole. At a latitude of
they appear to be moving clockwise.) AstroTour: The 60° north (as shown in Figure 2.5b), our horizon is tilted
View from the Poles 60° from the north celestial pole. This equality between
north latitude and the height of the north celestial pole
above the northern horizon holds everywhere. In Figure
Away from the Poles, the Part of the 2.5d, we have reached Earth’s equator, at a latitude of 0°.
Notice that the north celestial pole is now sitting on the
Sky We See Is Constantly Changing northern horizon. At the same time, we get our fi rst look at
Now let’s imagine what we see in the sky as we leave the south celestial pole, which is sitting opposite the north
the North Pole and travel south to lower latitudes. As celestial pole on the southern horizon. Continuing into the
you may already know, latitude3 is a measure of how far Southern Hemisphere, the south celestial pole is now visible
above the southern horizon, while the north celestial pole
is hidden from view by the northern horizon. At a latitude
3
You can fi nd defi nitions of the terrestrial coordinates—latitude and of 45° south (Figure 2.5e), the south celestial pole lies 45°
longitude—in Appendix 6. above the southern horizon. At the South Pole (latitude 90°
(a) Latitude 90°N (North Pole) (b) Latitude 60°N (c) Latitude 30°N
NCP (north celestial pole) NCP
The angle between the horizon
and the NCP is the same as
the observer’s north latitude.
NCP
90° Horizon 60°
N
Celestial N N 30°
equator Celestial
equator
90°N 60°N 30°N

Celestial
equator
S S S
(d) Latitude 0° (Equator) (e) Latitude 45°S (f) Latitude 90°S (South Pole)
NCP
N NCP N N
Celestial
equator
Equator
45°S
90°S
S SCP S
(south celestial pole) 45° 90°
SCP SCP
SCP
FIGURE 2 . 5 Our perspective on the sky depends on our location on Earth. Here we see how the
locations of the celestial poles and celestial equator depend on an observer’s latitude.
south—Figure 2.5f), the south celestial pole is at the zenith, tial pole is not a location in space, hovering over Earth’s
90° above the horizon. North Pole. Instead, it is a direction in space—the direc-
Probably the best way to cement your understanding tion parallel to Earth’s axis of rotation.
of the view of the sky at different latitudes is to draw Now that we have shown how the horizon is oriented
pictures like those in Figure 2.5. If you can draw a pic- at different latitudes, let’s see how the apparent motions
ture like this for any latitude—fi lling in the values for of the stars about the celestial poles differ from latitude to
each of the angles in the drawing and imagining what latitude. Figure 2.7a shows our view if we are at latitude
the sky looks like from that location—then you will be 30° north. As Earth rotates, the part of the sky visible to us
well on your way to developing a working knowledge of is constantly changing. Of course, from this perspective it
the appearance of the sky. That knowledge will prove is the horizon that seems to remain fi xed, while the stars
useful later, when we discuss a variety of phenomena, appear to move past overhead. If we focus our attention
such as the changing of the seasons. When practicing on the north celestial pole, from this perspective we still
your sketches, however, take care not to make the com- see much the same thing we saw from Earth’s North Pole.
mon mistake illustrated in Figure 2.6. The north celes- The north celestial pole remains fi xed in the sky, and all of
FIGURE 2 .6 The celestial

(a) CORRECT (b) WRONG poles and the celestial equator
are directions in space, not
NCP NCP The north celestial NCP …not locations fi xed locations hanging above
pole and celestial hanging over Earth. Earth. (The red X means “avoid
equator are directions North
Pole this misconception.”)
North in space…
Pole
CE
Celestial Celestial FIGURE 2 .7 (a) As viewed

equator equator from latitude 30° north, the
north celestial pole is 30°
above the northern horizon.
Stars appear to move on
counterclockwise paths around
South this point. At this latitude some
Pole South
parts of the sky are always
Pole
SCP SCP visible, while others are never
visible. (b) From the equator,
the north and south celestial
poles are seen on the horizon,
and the entire sky is visible over
a period of 24 hours.
(a) View from 30°N Always visible Circumpolar

NCP NCP
30°
6 P.M. 6 A.M. Apparent motion
of stars viewed
from 30°N.
North celestial
30°N pole Sometimes
Equator visible
30°
Horizon
S
Never visible
(b) View from equator North celestial pole is Sometimes visible

NCP on the horizon
NCP
6 P.M. 6 A.M.
North celestial
pole
SCP Horizon
S
FIGURE 2 .8 Time exposures

From a location in the Canadian …but at lower latitudes the
woods, the north celestial pole north celestial pole appears of the sky showing the apparent
appears high in the sky… closer to the horizon. motions of stars through the
night. Note the difference in the
circumpolar portion of the sky
as seen from the two different
latitudes.
the stars appear to move throughout the night in counter- The points where the celestial equator intersects the horizon
clockwise, circular paths around that point. But because are always due east and due west. (The only exception is
the north celestial pole is no longer directly overhead as it at the poles, where the celestial equator is coincident with
was at the North Pole, the apparent circular paths of the the horizon.) An object on the celestial equator rises due
stars are now tipped relative to the horizon. (More cor- east and sets due west. Objects that are north of the celes-
rectly, our horizon is now tipped relative to the apparent tial equator rise north of east and set north of west. Objects
circular paths of the stars.) AstroTour: The Celestial that are south of the celestial equator rise south of east and
Sphere and the Ecliptic set south of west.
Stars located close enough to the north celestial pole Figure 2.9 also shows that regardless of where you are
are above the horizon 24 hours a day (even if we can’t see on Earth (again with the exception of the poles), half of
them in the daytime) as they complete their apparent paths the celestial equator is always visible above the horizon.
around the pole (see Figure 2.7a and Figure 2.8). This Because half of the celestial equator is always visible, it fol-
always-visible region of our sky is referred to as being cir- lows that you can see any object that lies in the direction of
cumpolar, which means the celestial equator half of the time. An object that is in the
“around the pole.” There is Circumpolar stars direction of the celestial equator rises due east, is above the
also a part of the sky that can are always above horizon for exactly 12 hours, and sets due west. This is not
never be seen from this lati- the horizon. true for objects that are not on the celestial equator. A look
tude. This is the part of the at Figure 2.9b shows that from the Northern Hemisphere,
sky near the south celestial pole that never rises above your you can see more than half of the apparent circular path
horizon. And between this region and the always-visible of any star that is north of the celestial equator. And if you
circumpolar region lies a portion of the sky that can be can see more than half of a star’s path, then the star is above
seen for part but not all of each day. Stars in this interme- the horizon for more than half of the time.
diate region appear to rise above and set below Earth’s As seen from the Northern Hemisphere, stars north of
shifting horizon as Earth turns. The only place on Earth the celestial equator remain above the horizon for more
where you can see the entire sky over the course of 24 hours than 12 hours each day. The farther north the star is, the
is the equator. From the equator (Figure 2.7b) the north and longer it stays up. The circumpolar stars near the north
south celestial poles sit on the northern and southern hori- celestial pole that we mentioned already are the extreme
zons, respectively, and the whole of the heavens passes example of this phenomenon; they are up 24 hours a day.
through the sky each day. In contrast, objects south of the celestial equator are above
Look at the location of the celestial equator in Figure 2.9. the horizon for less than 12 hours a day, and the farther
(a) North Pole NCP (b) 30° N Latitude

N 1 From any point on Earth, 3 From locations other than
half of the sky is visible at 6:00 P.M. the poles, the part of the sky
90° any given moment… we see is constantly changing.
30°
NCP
NCP 30°
90°
N
30°
Celestial
sphere NCP
6:00 A.M.
SCP
W
S 90° 30°
SCP S
N
E
Horizon Horizon
Celestial SCP
equator Celestial
equator
S
SCP 2 …while the other half 4 Stars “rise” and “set” as
is beneath the horizon, the part of the sky we can
blocked by Earth. see changes.
FIGURE 2 .9 The celestial sphere is shown here as viewed by observers at four different
latitudes. At all locations other than the poles, stars rise and set as the part of the celestial
sphere that we see changes during the day.
south you look, the less time a star is visible. Stars that are
located close to the south celestial pole never rise above
2.3 Revolution about the
our horizon. Sun Leads to Changes
If you were an observer in the Southern Hemisphere (see
Figure 2.9d), the reverse of the preceding discussion would during the Year
be true: Objects on the celestial equator would still be up for
12 hours a day, but now objects south of the celestial equator The second motion that we will discuss is the motion of Earth
would be up more than 12 hours, and objects north of the about the Sun. Earth revolves around the Sun in the same
celestial equator would be up less than 12 hours. direction that Earth spins about its axis—counterclockwise
For many centuries, travelers, including sailors at sea, have as viewed from above Earth’s North Pole. A year, by definition,
used the stars for navigation. Perhaps the simplest of the navi- is the time it takes for Earth to complete one revolution around
gator’s techniques is to use the equality between latitude and the Sun. The motion of Earth around the Sun is responsible
the altitude of the north (or south) celestial pole. We can find for many of the patterns of
the north or south celestial poles by recognizing the stars that change we see in the sky and Earth’s orbital motion
surround them. In the Northern Hemisphere, a moderately on Earth, including changes in is counterclockwise as
bright star happens by chance to be located within 0.7° of the which stars we see at night. As viewed from above
north celestial pole. This star Earth moves around the Sun, Earth’s North Pole.
is called Polaris, or more com- The star Polaris the stars we see overhead at
monly, the “North Star.” If you marks the north midnight change. Six months from now, Earth will be on the
can find Polaris in the sky and celestial pole. other side of the Sun, and the stars that we see overhead at
measure the angle between the midnight will be in nearly the opposite direction from the
north celestial pole and the horizon, then you know your lati- stars we see near overhead at midnight tonight. The stars that
tude. If you are in Phoenix, Arizona, for example (latitude were overhead at midnight six months ago are the same stars
33.5° north), you will find the north celestial pole 33.5° above that are overhead today at noon, but of course we cannot see
your northern horizon. On the other hand, if you are studying them today because of the glare of the Sun. AstroTour:
astronomy in Fairbanks, Alaska (latitude 64.6° north), the The Earth Spins and Revolves
north celestial pole sits much higher overhead, 64.6° above If you could note the position of the Sun relative to the
the horizon in the north. The location of the north celestial stars each day for a year, you would fi nd that it traces out
pole in the sky can also be used to measure the size of Earth a great circle against the background of the stars (Figure
(see Math Tools 2.1). 2.10 on page 34). On September 1, the Sun appears to be in
(c) Equator (d) 30° S Latitude

5 At the equator, all stars
rise and set each day. 6:00 P.M.
6:00 P.M.
30°
SCP
30°
30°
SCP NCP
SCP
6:00 A.M. NCP

W W
S S
30°
6:00 A.M. SCP NCP
N N
E E
Horizon Horizon
Celestial Celestial NCP
equator equator
the direction of the constellation of Leo. Six months later, the plane of Earth’s orbit about the Sun. AstroTour:
on March 1, Earth is on the other side of the Sun, and the The Celestial Sphere and the Ecliptic
Sun appears to be in the direction of the constellation of
Aquarius. The apparent path
that the Sun follows against The ecliptic is the
the background of the stars is Sun’s apparent yearly Tiny Deviations in the Direction
called the ecliptic. The 12 path against the of Starlight Provide a Measure of
constellations that lie along background of stars.
the ecliptic and through
Earth’s Motion through Space
which the Sun appears to move are called the constella- As difficult as it is to “feel” the effects of Earth’s rotation on
tions of the zodiac. This is why ancient astrologers assigned its axis, it is even harder to sense the motion of Earth around
special mystical significance to these stars. Actually, the the Sun. As we have seen, through most of the history of
constellations of the zodiac are nothing more than random our species humans believed that Earth remains stationary
patterns of distant stars that happen by chance to lie near while the Sun, the Moon, and the heavens revolve around
How to Estimate the Size of Earth

You can use the location of the north celestial pole in the the circumference of Earth must be about 144 × 290 km, or
sky to estimate the size of Earth. Suppose we start out in about 42,000 km. The actual circumference of Earth is just a
Phoenix, Arizona, where we observe the north celestial pole shade over 40,000 km, so our simple calculation was not too
to be 33.5° above the horizon. Now we head north, and by bad, given our sloppy measurements of angles and distances.
the time we reach the Grand Canyon, about 290 kilometers Recall from geometry class that the circumference of a cir-
(km) from Phoenix, we notice that the north celestial pole cle is equal to 2π times its radius. So, the radius of Earth is
has risen to about 36° above the horizon. This difference its circumference (40,000 km) divided by 2π, or about 6,400
(2.5°) is 1/144 of the way around a circle. (A circle is 360°, and km. It was in much this same way that the Greek astrono-
2.5°/360° = 1/144.) This means that we must have traveled 1/144 mer Eratosthenes (276–194 B.C .) made the fi rst accurate mea-
of the way around the circumference, C, of Earth. In other surements of the size of Earth, in about 230 B.C . (well before
words, 1/144 × C = 290 km. So, rearranging the expression, Columbus’s time).
ARIES PISCES
AQUA
RIUS
RUS
TAU
CA
PR
I CO
I RN
IN U
M
S
GE
November October September
December August
January Earth’s July

orbit
February June
March May
April
S
IU
CA
AR
N T
CE T
R GI
SA
LEO US
RPI
SCO
VIRGO LIBRA
By December 1 Earth has
On September 1 the Sun The apparent path that the Sun traveled far enough in its
is seen in the direction of follows against the background orbit that the Sun is seen
Leo as viewed from Earth. of the stars is called the ecliptic. in the direction of Scorpius.
FIGURE 2 .10 As Earth orbits about the Sun, the Sun’s apparent position against the background
of stars changes. The imaginary circle traced by the annual path of the Sun is called the ecliptic.
Constellations along the ecliptic form the zodiac.
us. The history of modern astronomy, and to some degree examples in this chapter alone. For example, even though
the story of the rise of modern science, can be told as the Earth is spinning on its axis and flying through space in its
story of how this view was overthrown during the 17th and orbit about the Sun, the resulting relative motions between
18th centuries. However, the fi rst direct measurement of the objects that are near each other on Earth are small. New-
effect of Earth’s motion was not made until the 18th century. ton’s realization that motions are meaningful only when tied
To understand how this measurement was made, it’s help- to the frame of reference of a particular observer is also at
ful fi rst to understand the concept of a frame of reference. the heart of Einstein’s theories of relativity. These theories,
Briefly, a frame of reference is a coordinate system within which we will return to later, wound up changing the way
which an observer measures positions and motions. we think about space and time.
Aside from looking out the window or feeling road vibra- We can use a common experience to illustrate what you
tions, there is no experiment that you could easily do to tell might observe within a specific frame of reference. Imag-
the difference between riding in a car down a straight sec- ine that you are sitting in a car in a windless rainstorm, as
tion of highway at constant shown in Figure 2.11. If the car is sitting still and the rain
speed and sitting in the car Relative motion is falling vertically, when you look out your side window
while it is parked in your is all that counts. you see raindrops falling straight down. That is, if you hold
driveway. Because everything a vertical tube out the window, raindrops will fall straight
in the car is moving together, the relative motions between through the tube. When the car is moving forward, how-
objects in the car are all that count. In fact, the only reason ever, the situation is different. Between the time a raindrop
you can feel the roughness of the road is that it slightly appears at the top of your window and the time it disappears
changes the motion of the car. You feel these brief accelera- beneath the bottom of your window, the car has moved for-
tions as the car’s vibration. ward. The raindrop disappears beneath the window behind
The idea that only relative motions count occurs again the point at which it appeared, which means the raindrop
and again in astronomy and physics. There are numerous looks as if it falls at an angle, even though in reality it is
falling straight down. For raindrops to fall directly through (a) From the vantage point of a person outside, the
the tube you are holding out the window now, you would rain falls vertically, even when the car is moving.
have to tilt the top of the tube forward. As you go faster, the
apparent front-to-back motion of the raindrops increases, Car stationary Car moving
and their apparent paths become more tilted. An observer
by the side of the road would say the raindrops are coming
from directly overhead, but to you in the moving car they
are coming from a direction in front of the car. You are
observing this apparent motion of raindrops from within
your own special frame of reference.
This same phenomenon occurs with starlight, as shown (b) From the frame of reference of a person inside the car…
in Figure 2.12. The light from a distant star arrives at Earth
from the direction to this star. Because Earth moves, how-
…the rain falls …but at an
ever, to an observer on Earth the starlight seems to be com- vertically if the car angle if the
ing from a slightly different direction, just as the raindrops is stationary… car is moving.
appeared to be coming from in front of the car in Figure
2.11.4 As the direction of
Earth’s motion around the Sun Earth’s orbital motion
continuously changes during around the Sun was
the year, the apparent position first measured from
of a star in the sky moves in a the aberration of
small loop. This shift in appar- starlight.
ent position is what we refer
FIGURE 2 .11 On a windless day the direction in which rain falls
depends on the frame of reference in which it is viewed. (a) From
4
Actually, in the frame of reference of Earth, the starlight is coming outside of the car, the rain is seen to fall vertically downward
from a different direction; and in the frame of reference of the moving whether the car is stationary or moving. (b) From inside the car,
car, the raindrops are coming from in front of the car. the rain is seen to fall vertically downward if the car is stationary;
but if the car is moving, the
rain is seen to fall at an angle
determined by the speed and
1 Light from a distant
direction of the car’s motion.
star is like the vertical
rain in Figure 2.11.
2 In the reference frame

of the moving Earth, the
starlight is tipped in the
direction of Earth’s motion.
VISUAL ANALOGY FIGURE

Apparent 2 .12 The apparent positions
motion of star
of stars are deflected slightly
toward the direction in which
Earth is moving. As Earth orbits
3 Over the year,
the direction of the Sun, stars appear to trace
incoming starlight out small loops in the sky. This
traces out a loop. effect is called aberration of
starlight.
to as the aberration of starlight, and it was first detected in To understand how the combination of Earth’s axial tilt
the 1720s by two English astronomers, Samuel Molyneux and its annual path around the Sun creates seasons, let’s
and James Bradley. Measurement of the aberration of star- look at a special case. If Earth’s spin axis were exactly per-
light shows that Earth is moving on a roughly (but not pendicular to the plane of Earth’s orbit (the ecliptic plane),
exactly) circular path about the Sun with an average speed then the Sun would always appear to lie on the celestial
of just under 30 kilometers per second (km/s). Because dis- equator. Because the position of the celestial equator is fi xed
tance equals speed multiplied by time, the distance around in our sky, the Sun would follow the same path through the
this near-circle—its circumference—is the speed of Earth sky day after day, rising due east each morning and setting
(29.8 km/s) multiplied by the length of one year (3.16 × 107 due west each evening. If the Sun were always on the celes-
seconds). The circumference of Earth’s orbit is then tial equator, it would be above the horizon for exactly half
the time, and days and nights would always be exactly 12
Distance = Speed × Time hours long. In short, if Earth’s axis were exactly perpen-
km dicular to the plane of Earth’s orbit, each day would be just
= (29.8 ___
s ) × (3.16 × 10 s)
7
like the last, and there would be no seasons.
= 9.42 × 108 km. However, Earth’s axis of rotation is not exactly perpen-
dicular to the plane of the ecliptic. Instead it is tilted by
The radius of Earth’s nearly circular orbit is this circum- 23.5° from the perpendicular.7 As Earth moves around the
ference divided by 2π, or 1.50 × 108 km (150 million kilo- Sun, its axis points in almost exactly the same direction
meters). Astronomers refer to throughout the year and from
this distance—the average The astronomical unit one year to the next.8 As a Seasons result from
distance between the center is the average result, sometimes Earth’s the 23.5° tilt of
of the Sun and the center of distance between the North Pole is tilted more Earth’s axis with
5
Earth —as one astronomical Sun and Earth. toward the Sun, and at other respect to a line
unit, abbreviated AU. The times it is pointed more away perpendicular to its
astronomical unit is a good unit for measuring distances from the Sun. When Earth’s orbital plane.
within the Solar System. North Pole is tilted toward the
Modern measurements of the size of the astronomical Sun, an observer on Earth sees the Sun as lying north of
unit are made in very different ways, such as bouncing radar the celestial equator. Six months later, when Earth’s North
signals off Venus. However, the aberration of starlight pro- Pole is tilted away from the Sun, the Sun is seen as lying
vided a simple and compelling demonstration that Earth south of the celestial equator. If we look at the circle of the
orbits about the Sun—and a pretty good value for the size Sun’s apparent path through the stars—the ecliptic—we see
of Earth’s orbit as well. that it is tilted by 23.5° with respect to the celestial equator.
AstroTour: The Earth Spins and Revolves
To understand the effect that this tilt has on Earth, begin
Seasons Are Due to the by looking at Figure 2.13a. This figure shows the situation
on June 21, the day that Earth’s North Pole is tilted most
Tilt of Earth’s Axis directly toward the Sun.9 Note fi rst that the Sun is north of
So far we have discussed the rotation of Earth on its axis the celestial equator. We found earlier in the chapter that,
and the revolution of Earth about the Sun. To understand from the perspective of an observer in the Northern Hemi-
the changing of the seasons, we need to consider the com- sphere, an object north of the celestial equator can be seen
bined effects of these two motions. If you ask most people above the horizon for more than half the time. This is true
why it is cold in the winter and warm in the summer, the for the Sun, as well as for any other celestial object. Saying
reason they are likely to give is that Earth is closer to the that the Sun is above the horizon for more than half the
Sun in the summer and farther away in the winter. This is a time is just another way of saying that the days are longer
common (and commonsense) idea that does have something
7
to do with the seasons on Mars, but it has virtually noth- Astronomers use the term obliquity to refer to the angle between a
ing to do with the seasons on Earth! Earth’s orbit around planet’s equatorial and orbital planes. See, for example, Tables 7.1
the Sun is an almost perfect circle centered on the Sun,6 so and 9.1.
8
the distance from the Sun changes little during the year. In We say almost exactly because, as we will soon learn, Earth’s axis
wobbles slowly, like that of a spinning top, taking about 26,000 years
fact, Earth is slightly closer to the Sun during the northern
to complete a single wobble.
winter than it is during the northern summer. 9
We are a bit sloppy with language here. Earth’s North Pole tilts in
the same direction year-round. On the fi rst day of the northern sum-
5
Distances between celestial objects are almost always taken as the mer, Earth is on the side of the Sun where the tilt of the North Pole
distances between their centers. is toward the Sun. On the fi rst day of the northern winter, Earth is
6
Planetary orbits are actually elliptical in shape, as will be discussed on the opposite side of the Sun, so the tilt of the North Pole is away
in Chapter 3. from the Sun.
First day of northern spring
Sun
First day of First day of
northern summer northern winter
First day of northern autumn
(a) June 21 December 22 (b)
North North
23.5° Pole 1 From the Arctic Circle to 23.5° Pole
the North Pole, there are
24 hours of daylight.
Arc 66. Arc 66.

tic 5°
Cir N
2 At the Tropic of Cancer, the Sun tic 5°
Cir N
cle is directly overhead at noon. cle
Tro Tro
p ic o 23 p ic o 23
f C .5°N f C .5°N
anc anc
er er
3 More than half of the
Equ Northern Hemisphere Equ
ato ato
Tro
p
r, 0
°
is in daylight. Tro
p
r, 0
°
ic o ic o
fC fC
apr apr
ic ic
23. orn 23. orn
5°S 5°S
Ant 4 Less than half of the Southern Ant
arc arc
tic
Cir Hemisphere is in daylight. tic
Cir
66. cle 66. cle
5°S 5°S
5 From the Antarctic Circle

South South 6 The first day of northern winter is
to the South Pole, there
Pole Pole 6 the first day of southern summer.
are 24 hours of night.
FIGURE 2 .13 (a) On the first day of the northern summer (June 21, the summer solstice), the
northern end of Earth’s axis is tilted most nearly toward the Sun, while the Southern Hemisphere is
tipped away. (b) Six months later, on the first day of the northern winter (December 22, the winter
solstice), the situation is reversed. Seasons are opposite in the Northern and Southern Hemispheres.
than 12 hours. You can see this directly in Figure 2.13a by exactly reversed from what is going on in the north. Look
noting that when Earth’s North Pole is tilted toward the again at Figure 2.13. On June 21, while the Northern Hemi-
Sun, over half of Earth’s Northern Hemisphere is illumi- sphere is enjoying long days and short nights of summer,
nated by sunlight. These are the long days of the northern Earth’s South Pole is tilted in the direction away from the
summer. Six months later the situation is very different. Sun. Less than half of the Southern Hemisphere is illumi-
On December 22 (Figure 2.13b), Earth’s North Pole is tilted nated by the Sun, and the winter days are shorter than 12
away from the Sun, so the Sun appears in the sky south of hours. Similarly, on December 22, Earth’s South Pole is tilted
the celestial equator. Someone in the Northern Hemisphere toward the Sun, and the southern summer days are long.
will see the Sun for less than 12 hours each day. Less than The differing length of days through the year is part of
half of the Northern Hemisphere is illuminated by the Sun. the explanation for the changing seasons, but we need to
It is winter in the north. consider another important effect: the Sun appears higher
Over the course of the year, the length of the day changes, in the sky during the summer than it does during the win-
courtesy of the 23.5° tilt of Earth’s axis. In the preceding ter, so sunlight strikes the ground more directly during the
paragraph we were careful to summer than during the winter. To see why this is impor-
specify the length of the day Seasons in the tant, hold a piece of cardboard toward the Sun and look at
in the Northern Hemisphere Southern Hemisphere the size of its shadow. If the cardboard is held so that it is
because things are very differ- are the reverse of directly face-on to the Sun, then its shadow is large. As you
ent in the Southern Hemi- those in the Northern turn the cardboard more edge-on, however, the size of its
sphere. In fact, things in the Hemisphere. shadow shrinks. The size of the cardboard’s shadow tells
Southern Hemisphere are you that the cardboard catches less energy from the Sun
(a) (b)
First day of summer First day of winter

at 40° N at 40° N
16.5°
63.5°
Direction
of sunlight
Direction
of sunlight
1 meter
1 meter
1.04 meters 2.24 meters

Sunlight striking the The same amount of sunlight strikes
ground nearly face-on the ground at a shallower angle and
is more concentrated. so is spread out over a larger area.
FIGURE 2.14 Local noon at latitude 40° north. (a) On the first day of northern summer, sunlight
strikes the ground almost face-on. (b) On the first day of northern winter, sunlight strikes the ground
more obliquely, and less than half as much sunlight falls on each square meter of ground each second.
each second when it is tilted relative to the Sun than it does the horizon and only 16.5° away from the zenith. Sunlight
when it is face-on to the Sun. This is exactly what happens strikes the ground almost face-on (Figure 2.14a). In contrast,
with the changing seasons. at noon on the first day of winter, the Sun is only 26.5° above
During the summer, Earth’s The angle of sunlight the horizon, or 63.5° from the zenith. Sunlight strikes the
surface is more nearly face-on to the ground is closer ground at a rather shallow angle (Figure 2.14b). As a result of
to the incoming sunlight, so to perpendicular in these differences, more than twice as much solar energy falls
more energy falls on each summer than in on each square meter of ground per second at noon on June
square meter of ground each winter, so there is 21 as falls there at noon on December 22. Together, these two
second. During the winter, the more heating per unit effects—the directness of sunlight and the differing length of
surface of Earth is more area in summer. the day—mean that during the summer there is more heating
inclined with respect to the from the Sun and during the winter there is less.
sunlight, so less energy falls on each square meter of the We do not have to wait for the seasons to change to see
ground each second. That is the main reason why it is hot- the effect that the height of the Sun in the sky has on ter-
ter in the summer and colder in the winter. restrial climate. We need only compare the climates found
To see an example of how this works, look at Figure 2.14, at different latitudes on Earth. Near the equator, the Sun
which shows the direction of incoming sunlight striking passes high overhead every day, regardless of the season. As
Earth at latitude 40° north, which stretches across middle a result, the climate is warm throughout the year. At high
America from Northern California to New Jersey. At noon on latitudes, however, the Sun is never high in the sky, and the
the first day of summer, the Sun is high in the sky—73.5° above climate can be cold and harsh even during the summer.
crosses the celestial equator moving from the south to the

Four Special Days Mark the north (point 1 in Figure 2.15b). That direction on the celes-
Passage of the Seasons tial sphere, located in the constellation Pisces, is called the
vernal equinox. The term vernal equinox also refers to the
As Earth travels around the Sun over the course of a year,
day—about March 21—when the Sun appears at this loca-
the Sun appears to trace a path along the ecliptic, a great
tion. When the Sun lies on the celestial equator on the ver-
circle that is tilted 23.5° with
nal equinox, days are 12 hours long. (The term equinox
respect to the celestial equa- The changing seasons
means literally “equal night”; everywhere on Earth, night
tor. Follow along in Figure are marked by
and day are the same length on the days of the equinoxes.)
2.15 as we note the four spe- equinoxes and
In the Northern Hemisphere, the vernal equinox is the fi rst
cial points on this path that solstices.
day of spring.
mark the passage of the sea-
As Earth continues its journey around the Sun, the sun-
sons. Begin with the point in March when Earth’s axis is
ward direction moves along toward closer alignment with
perpendicular to the direction to the Sun (point 1 in Figure
the tilt of the northern end of Earth’s axis, and the Sun
2.15a). Here, the Sun’s apparent motion along the ecliptic
climbs higher into the northern sky. The northern end of
Earth’s axis tilts toward the Sun about 3 months after the
vernal equinox. When this happens, the Sun reaches its
FIGURE 2 .15 The motion of Earth about the Sun as seen from northernmost point in the sky, located in the constellation
the frame of reference of (a) the Sun and (b) Earth. Taurus, near its border with Gemini. This day, which marks
the longest day of the year and the beginning of summer in
Motion of Earth around the Sun
the Northern Hemisphere, occurs around June 21. This is
(a) the summer solstice. (Solstice literally means “sun standing
4 Winter solstice
still”; the Sun’s north–south motion in the sky stops as it
1 Vernal equinox reverses its direction.) Note that on this same day, the south-
ern end of Earth’s axis is tipped directly away from the Sun.
This is the shortest day of the year in the Southern Hemi-
Earth’s orbit
Sun sphere, marking the beginning of the southern winter.
Earth’s orbit Three months later, around September 23, the Sun is
around the Sun and
again crossing the celestial equator. This point on the Sun’s
the tilt of its axis…
apparent path, located in the constellation Virgo, is called
3 Autumnal equinox the autumnal equinox. The term refers both to the Sun’s
2 Summer solstice location on the celestial sphere and the date when this
happens. Because the Sun is again on the celestial equator,
days and nights are exactly 12 hours long. It is the fi rst day
Apparent motion of the Sun seen from Earth of autumn in the Northern Hemisphere and the fi rst day of
(b)
spring in the Southern Hemisphere.
…lead to the Sun’s apparent Around December 22, the Sun reaches its southernmost
yearly motion along the ecliptic. point in the sky as its apparent path takes it through the
constellation Sagittarius. This day is called the northern
3 At the autumnal equinox 2 At the summer solstice winter solstice. Earth’s North Pole is tipped most directly
(September 23) the Sun is (June 21) the Sun is north away from the Sun on this day. In the Northern Hemisphere
again on the celestial equator. of the celestial equator.
this is the shortest day of the year—the fi rst day of winter.
As the Sun passes the winter solstice and moves on toward
the vernal equinox, the northern days begin growing lon-
ger again. Almost all cultural traditions in the Northern
Hemisphere include a major celebration of some sort in
late December (Figure 2.16). Christmas, for example, is cel-
Earth
23.5°
ebrated just 3 days after the winter solstice. These winter
festivals have many different meanings to their various
Celestial celebrants, but they all share one thing: they celebrate the
equator return of the source of Earth’s light and warmth. The days
Ecliptic
have stopped growing shorter and are beginning to get
4 At the winter solstice 1 At the vernal equinox longer. It is a “new year,” and once again the Sun will hold
(December 22) the Sun is (March 21) the Sun is sway over night. Spring will come again.
south of the celestial equator. on the celestial equator. Interestingly, there is more to what we feel during the
different seasons than the amount of energy we are receiving
FIGURE 2 .16 Most cultural traditions in the Northern

Hemisphere include a major celebration in late December, around
the time when days begin to grow longer.
from the Sun. Just as it takes time for a pot of water on a stove Antarctic regions pay for these long days, however, with an
to heat up when the burner is turned up and time for the pot equally long period surrounding the first day of winter when
to cool off when the burner is the Sun never rises and the nights are 24 hours long. The
turned down, it takes time for Seasonal Sun never rises high in the Arctic or Antarctic sky, which
Earth to respond to changes in temperatures lag means that sunlight is never very direct. This is why, even
heating from the Sun. The hot- behind changes in the with the long days at the height of summer, the Arctic and
test months of northern sum- directness of sunlight. Antarctic regions remain relatively cool.
mer are usually July and The seasons are also different near the equator. Recall
August, which come after the summer solstice, when the from Figure 2.9c that for an observer on the equator, all
days are growing shorter. Similarly, the coldest months of stars are above the horizon 12 hours a day, and the Sun is
northern winter are usually January and February, which no exception. On the equator, days and nights are 12 hours
occur after the winter solstice, when the days are growing long throughout the year. Changes in the directness of
longer. The climatic seasons on Earth lag behind changes in sunlight through the year are also different on the equa-
the amount of heating we receive from the Sun. tor. Here the Sun passes directly overhead on the first day
Our picture of the seasons must be modified somewhat of spring and the fi rst day of autumn because these are the
near Earth’s poles. At latitudes north of 66.5° north and days when the Sun is on the celestial equator. Sunlight is
south of 66.5° south, the Sun is circumpolar for a part of most direct at the equator on these days. At the summer sol-
the year surrounding the fi rst day of summer. These lines stice, the Sun is at its northernmost point along the ecliptic.
of latitude are called the Arctic Circle and the Antarctic It is on this day, and on the winter solstice, that the Sun
Circle, respectively. When the Sun is circumpolar, it is is farthest from the zenith at noon, and therefore sunlight
above the horizon 24 hours a day, earning the polar regions is least direct. Strictly speaking, the equator experiences
the nickname “land of the midnight Sun.” The Arctic and only two seasons: summer, when the Sun passes directly
(a)
Polaris (current Vega (pole 2 .17 (a) Earth’s axis of rotation
pole star) star in changes orientation in the same
Thuban (pole A.D. 14,000)
way that the axis of a spinning
star in 3,000 B.C.) 23.5° top changes orientation.
Equator (b) This precession causes
Earth’s axis the projection of the Earth’s
of rotation rotation axis to move in a
47°-diameter circle, centered
on the north ecliptic pole
(orange cross), with a period of
25,800 years. The red cross
Ecliptic plane shows the projection of Earth’s
axis on the sky in the early 21st
century.
(b)
ernmost point and in the constellation of Sagittarius at its

southernmost point. Why, then, is the Tropic of Cancer not
Deneb
called the Tropic of Taurus? Why is the Tropic of Capricorn
Cygnus
A.D. 8000 not called the Tropic of Sagittarius? The answer is historical.
Cassiopeia
When these latitude limits were named about 2,000 years
ago, the Sun’s northernmost point was in Cancer. Since
Cepheus
Lyra then, it has slowly drifted from Cancer to Gemini and is
Vega now in Taurus.10 Likewise, the Sun’s southernmost point
has drifted from Capricorn to Sagittarius. How could this
A.D. 15,000 Polaris
happen? We are about to fi nd out.
Draco
A.D. 1
Earth’s Axis Wobbles, and the
Ursa
Minor
Seasons Shift through the Year
Path of north
celestial pole
Thuban When the Alexandrian astronomer Ptolemy (Claudius Ptole-
5000 B.C.
maeus) and his associates were formalizing their knowledge
Corona
Borealis
of the positions and motions of objects in the sky 2,000
Ursa Major years ago, the Sun appeared in the constellation of Cancer
on the fi rst day of northern
summer and in the constella- Earth’s axis wobbles
tion of Capricorn on the fi rst like the axis of a
day of northern winter— spinning top.
overhead; and winter, when the Sun is at its northernmost hence, the names of the Trop-
and southernmost points on the ecliptic. However, summer ics. Which leads us to this question: Why have the
and winter are not very different. The Sun is up for 12 hours constellations in which solstices appear changed? The
a day year-round, and the Sun is always so close to being answer has to do with the fact that there are two motions
overhead at noon that the directness of sunlight changes by associated with Earth and its axis. Earth spins on its axis,
only 8 percent throughout the year. but its axis also wobbles like the axis of a spinning top
If you live between the latitudes of 23.5° south and 23.5° (Figure 2.17). The wobble is very slow, taking about 26,000
north—in Rio de Janeiro or Honolulu, for example—the Sun years to complete one cycle. During this time the north
will be directly overhead at noon twice during the year. The celestial pole makes one trip around a large circle centered
band between these two latitudes is called the Tropics. The on the north ecliptic pole. Polaris is a modern name for the
northern limit of this region is called the Tropic of Cancer; star we see near the north celestial pole. If you could travel
the southern limit is called the Tropic of Capricorn (see Fig-
ure 2.13). (As a challenge, think about what the seasons are 10
It was only recently, in 1990, that the location of the summer solstice
like at different locations within the Tropics.) passed from Gemini into Taurus. In about 600 years the vernal equinox
Wait a minute! We already pointed out that the Sun will move from Pisces into Aquarius, marking the true beginning of the
appears in the constellation of Taurus when at its north- “Age of Aquarius.”
several thousand years into the past or future, you would Today’s calendar (which was not adopted in Russia
fi nd that the point about which the northern sky appears until the 1917 Bolshevik Revolution) is known as the Gre-
to rotate is no longer near Polaris. gorian calendar and is based on the tropical year, which
Recall that the celestial equator is the set of directions is 365.242199 solar days long. A solar day is the 24-hour
in the sky that are perpendicular to Earth’s axis. As Earth’s period of Earth’s rotation that brings the Sun back to the
axis wobbles, then, the celes- same local meridian. The tropical year measures the time
tial equator must appear to tilt A 26,000-year wobble from one vernal equinox to the next—from the start of spring
with it. And as the celestial causes the position to the start of spring. Notice that the tropical year is not an
equator wobbles, the locations of the equinoxes to integral number of days long, but is approximately 1⁄4 day
where it crosses the ecliptic— shift gradually. longer. An elaborate system of leap years—years in which
the equinoxes—change as a 29th day is added to the month of February—is used in
well. During each 26,000-year wobble of Earth’s axis, the our calendar to make up for the extra fraction of a day, pre-
locations of the equinoxes make one complete circuit around venting the seasons from slowly “sliding through” the year
the celestial equator. Together, these shifts in position are (becoming increasingly out of sync with the months): winter
called the precession of the equinoxes. in December one year, in August in another. Excursions 2.1
The tendency of the seasons to shift through the year expands on this topic.
from century to century has played havoc on human efforts
to construct reliable calendars, and history is full of inter-
esting anecdotes related to this difficulty. For example, for
much of the 16th, 17th, and 18th centuries, the calendars 2.4 The Motions and
in Protestant Europe lagged behind the calendars in Cath-
olic Europe by fi rst 10, and later 11, days. Not until 1752 Phases of the Moon
did England and her colonies, including those in America,
drop 11 days from their calendars to bring them into line The most prominent object in the sky after the Sun is the
with the Catholic calendar. This conciliatory step was met Moon. Just as Earth orbits about the Sun, the Moon orbits
by riots among the people, who somehow felt that these 11 around Earth. (Actually, Earth and the Moon orbit around
days had been stolen from them. each other, and together they orbit the Sun, as we will see
E X C U R S I O N S 2 .1
Why Is It Surprising That A.D. 2000 Was a Leap Year?

Almost everyone knows that years that are divisible by 4 are day every four years, our calendar would slip by 1⁄4 day each
leap years, and A.D. 2000 was no exception to that rule. Yet A.D. year, which means that the seasons would shift by not quite
2000 was a special case. To understand why, we need to take a a month each century. If we did not correct for leap years,
look at the way our calendar is constructed, and the purpose the fi rst day of summer would come in July, then in August,
that leap years serve. then in September, and so on.
It takes Earth 365.242199 days to travel from one vernal Even after we add a leap year, the calendar still has prob-
equinox to the next. This tropical year is not exactly equal lems. A 365-day year is short by 0.242199 day, which is a bit
to the 365 days we normally think of as making up a year. less than 1⁄4. If we kept adding an extra day every 4 years, then
(And thanks to the precession of the equinoxes, it is also not over time we would accumulate an average of 0.007801 extra
equal to the time—365.256366 days—that it takes for Earth day per year. In 400 years the calendar would have slipped by
to complete one orbit about the Sun!) For convenience, we about 3 days. This difference is what caused the Protestant and
count years as starting at midnight on the morning of Janu- Catholic calendars to fall into disagreement by 11 days by the
ary 1, and ending 365 days later, at midnight on the night middle of the last millennium. To fi x this problem, it is neces-
of December 31. But what about that extra fraction of a day sary to get rid of 3 days every 400 years. We accomplish this
(0.24 plus a bit) that we have not accounted for? Here is where by making century years into common 365-day years, except
leap years come in. A true tropical year is 0.242199 day—or for those century years that are divisible by 400—such as A.D.
about 1⁄4 day—longer than a 365-day year, which means that 2000—which remain leap years. With one slight further revi-
after 4 years these extra parts have added up to make about sion—making years divisible by 4,000 into common 365-day
one extra day. So every 4 years we add the extra day back years—the modern Gregorian calendar now slips by about only
in, and February gets a 29th day. If we did not add this extra 1 day in 20,000 years.
in Chapter 3.) In some respects, the appearance of the Moon Moon does exactly the same thing, rotating on its axis once
is constantly changing, but we begin our discussion of the per revolution around Earth, always keeping the same face
motion of the Moon by talking about an aspect of the Moon’s toward Earth, as shown in Figure 2.18. This phenomenon
appearance that does not change. is referred to as the Moon’s synchronous rotation. (The
Moon’s synchronous rotation is not an accident. In Chap-
ter 10 we will find that its cause is related to why we have
We Always See the Same tides on Earth.) AstroTour: The Moon’s Orbit: Eclipses
and Phases
Face of the Moon
The Moon constantly changes its lighted shape and position
in the sky, but one thing that does not change is the face of
the Moon that we see. If we
The Changing Phases of the Moon
were to go outside next week The Moon rotates on Humans have long been fascinated by the Moon and its
or next month, or 20 years its axis once for each changing aspects. We speak of the “man in the Moon,” the
from now, or 20,000 centuries orbit around Earth. “harvest Moon,” and sometimes a “blue Moon.” In mythol-
from now, we would still see ogy, the Moon was the Roman goddess Diana and the Greek
the same side of the Moon that we see tonight. This fact is goddess Artemis. Among the Aztecs, it was the god
responsible for the common misconception that the Moon Tecciztecatl. The Moon has been the frequent subject of art,
does not rotate. The Moon does rotate on its axis—exactly literature, and music. Some-
once for each revolution that it makes about Earth. times the Moon appears as a The Moon shines by
Imagine walking around the Washington Monument circular disk in the sky. At reflected sunlight.
while keeping your face toward the monument at all times other times, it is nothing more
(a reasonable thing to do—you want to get a good look at it). than a thin sliver. At still others, its face appears dark.
By the time you complete one circle around the monument, Popular culture often refers to the side of the Moon away
your head has turned completely around once. (When you from Earth as the “dark side of the Moon.” This is even the
were south of the monument, you were facing north; when title of one of the most influential rock albums of the 20th
you were west of the monument, you were facing east; and century. In fact, however, there is no dark side of the Moon.
so on.) But someone looking at you from the monument At any given time, half of the Moon is in sunlight and half
would never have seen anything other than your face. The is in darkness—just as at any given time, half of Earth is in
sunlight and half is in darkness. The side of the Moon that
faces away from Earth, the “far side,” spends just as much
FIGURE 2 .18 The Moon rotates once on its axis for each orbit time in sunlight as the side of the Moon that faces toward
around Earth—an effect called synchronous rotation. Earth does. But “I’ll see you on the backside of the Moon”
might not have been as wildly successful as a song lyric.
Orbit of Moon Our fascination with the Moon can be explained by its
unique features and its constantly changing aspect. Unlike
the Sun, the Moon has no light source of its own. Like the
planets, including Earth, it shines by reflected sunlight.
Like Earth, half of the Moon is always in bright daylight,
and half is always in darkness. The different phases of the
Moon result from the fact that the illuminated portion of
Earth the Moon that we see is constantly changing. Sometimes
(during a new Moon) the side facing away from us is illumi-
nated, and sometimes (during a full Moon) the side facing
toward us is illuminated. The rest of the time, only part of
the illuminated portion can be seen from Earth.
To help you visualize the changing phases of the Moon,
go outside at night and have a friend hold up a soccer ball
so that it is illuminated from
one side by a nearby street- The phase of the
light (representing the Sun). Moon is determined by
Have your friend walk around how much of its bright
you in a circle, and watch the side we can see.
changes in the ball’s appear-
The Moon rotates once on its axis for each orbit around ance. When you are between the ball and the streetlight,
Earth, and so keeps the same face toward Earth at all times. the face of the ball that is toward you is fully illuminated.
FIGURE 2 .19 The inner

circle of images (connected
Crosses meridian near sunrise.
by blue arrows) shows the
Moon as seen Moon as it orbits Earth, as
from Earth seen by an observer far above
Earth’s North Pole. The outer
Waning crescent Waning gibbous ring of images shows the
corresponding phases of the
Third quarter Moon as seen from Earth.
Waning
phases
New Moon Sunrise Full Moon
Noon Midnight
Sunset
Crosses meridian Crosses meridian
at noon. at midnight.
Waxing
phases
First quarter
Waxing crescent Waxing gibbous
Crosses meridian near sunset.
The ball appears to be a bright, circular disk. As the ball shape is called a crescent, from the Latin crescere, meaning
moves around its circle, you will see a progression of lighted “to grow.” Because the Moon appears to be “filling out” from
shapes, depending on how much of the bright side and how night to night at this time, the full name for this phase of
much of the dark side of the ball you can see. This progres- the Moon is a “waxing crescent Moon.” (Waxing here means
sion of shapes exactly mimics the changing phases of the “growing in size and brilliance.”) From our perspective,
Moon. the Moon has also moved away from the Sun in the sky.
Figure 2.19 shows the changing phases of the Moon. Because the Moon travels around Earth in the same direc-
When the Moon is between Earth and the Sun, the illumi- tion in which Earth rotates, we now see the Moon located
nated side of the Moon faces away from us, and we see only to the east of the Sun. A waxing crescent Moon is visible
its dark side. This is called a new Moon. (Study Figure 2.19 in the western sky in the evening, near the setting Sun but
to understand that a new Moon can be “seen” only from the remaining above the horizon after the Sun sets. The “horns”
illuminated side of Earth.) It appears close to the Sun in the of the crescent always point directly away from the Sun.
sky, so it rises in the east at sunrise, crosses the meridian As the Moon moves farther along in its orbit, more and
near noon, and sets in the west near sunset. A new Moon more of its illuminated side becomes visible each night,
is never visible in the nighttime sky. AstroTour: The so the crescent continues to fill out. At the same time, the
Moon’s Orbit: Eclipses and Phases angular separation in the sky between the Moon and the
As the Moon continues on its orbit around Earth, a small Sun grows. After about a week the Moon has moved a quar-
part of its illuminated hemisphere becomes visible. This ter of the way around Earth. We now see half the Moon as
(a) (b)
3 After one synodic 2 After one sidereal month (27.32

month (29.53 days) the days) the Moon has gone around
Moon is once again full. Earth once but is not yet full.
Earth’s Note: When shown to

orbit scale, the Moon’s orbit
is always concave toward
Full Moon the Sun. The Moon’s
principal motion is its
orbit about the Sun.
Earth orbit
To Sun
Moon’s
Earth Moon orbit
orbit
FIGURE 2 . 20 (a) The Moon completes one sidereal orbit in Moon

27.32 days, but the synodic period (the period between phases
seen from Earth) from one full Moon to the next is 29.53 days.
(The horizontal orange line to the right of the Moon indicates a
fi xed direction in space.) (b) The orbits of Earth and the Moon are
shown here to scale.
illuminated and half the Moon as dark—a phase that we call

first quarter Moon. A look at Figure 2.19 shows that the fi rst
quarter Moon rises at noon, crosses the meridian at sunset,
and sets at midnight. Note that first quarter refers not to
how much of the face of the Moon that we see illuminated,
but rather to the fact that the Moon has completed the fi rst
quarter of its cycle from new Moon to new Moon.
As the Moon moves beyond fi rst quarter, we are able to
see more than half of its bright side. This phase is called a To Sun
“waxing gibbous Moon,” from the Latin gibbus, meaning
“hump.” The gibbous Moon continues nightly to “grow”
until finally Earth is between the Sun and the Moon and we 1 Begin with
see the entire bright side of the Moon—a full Moon. The Sun a full Moon.
and the Moon now appear opposite each other in the sky.
The full Moon rises as the Sun sets, crosses the meridian at
midnight, and sets in the morning as the Sun rises.
The second half of the Moon’s orbit proceeds just like a dark circle rising and setting with the Sun, and the cycle
the fi rst half, but in reverse. The Moon continues in its begins again.
orbit, again appearing gibbous but now becoming smaller It takes the Moon 27.32 days to complete one revolution
each night. This phase is called a “waning gibbous Moon.” about Earth; this is called its sidereal period. However,
(Waning means “becoming smaller.”) A third quarter Moon because of the changing relationship between Earth, the
occurs when we once again see half of the illuminated Moon, and the Sun due to Earth’s orbital motion, it takes
part of the Moon and half of the dark part of the Moon. A 29.53 days to go from one full Moon to the next; this is called
third quarter Moon rises at midnight, crosses the merid- its synodic period. Figure 2.20 shows how this works. You
ian near sunrise, and sets at noon. The Moon continues on can always tell a waxing Moon from a waning Moon because
its path, visible now as a “waning crescent Moon” in the the side that is illuminated is always the side facing the Sun.
morning sky, until the Moon again appears as nothing but When the Moon is waxing, it appears in the evening sky,
so its western side is illuminated (the right side as viewed

from the Northern Hemisphere). Conversely, when the Moon
is waning, the eastern side (the left side as viewed from the
Northern Hemisphere) appears bright.
Do not try to memorize all possible combinations of
where the Moon is in the sky at what phase and at what time
of day. You do not have to. Instead, work on understand-
ing the motion and phases of the Moon, and then use your
understanding to figure out the specifics of any given case.
As a way to study the phases of the Moon, draw a picture like
Figure 2.19, and use it to follow the Moon around its orbit.
From your drawing, figure out what phase you would see
and where it would appear in the sky at a given time of day. FIGURE 2 . 2 1 Stonehenge is an ancient artifact in the English
You might also enjoy thinking about what phase someone countryside, used 4,000 years ago to keep track of celestial
on the Moon would see when looking back at Earth. events.
we doubt their desire to exert some control over the terror

2.5 Eclipses: Passing of eclipses by learning a few of their secrets—and in the
process assuring themselves that an eclipse did not mean
through a Shadow that all was lost?
Put yourself in the place of our nomadic friend from the

opening section of this chapter. You are fi nely attuned to
the patterns of the sky, and you view these patterns not
Varieties of Eclipses
as the inexorable consequences of physical law but as vis- An eclipse in which Earth moves through the shadow of
ible signs from the gods. Can you imagine any celestial the Moon is called a solar eclipse. Three different types of
events that would strike more terror in your heart than to solar eclipses are possible: total, partial, and annular. To
look up and see the Sun, giver of light and warmth, being see why, begin by looking at the structure of the shadow of
eaten away as if by a giant dragon, or the full Moon turn- the Sun cast by a round object
ing the ominous color of blood? Archaeological evidence such as the Moon, as shown There are three types
suggests that our ancestors put great effort into trying to in Figure 2.22. An observer at of solar eclipses:
figure out the pattern of eclipses and thereby bring them point A would be unable to total, partial, and
into the orderly scheme of the heavens. For example, the see any part of the surface of annular.
ancient and massive stone artifact in the English country- the Sun. This darkest, inner
side called Stonehenge, pictured in Figure 2.21, may have part of the shadow is called the umbra. If a point on Earth
allowed its builders to predict when eclipses might occur. passes through the Moon’s umbra, the Sun’s light is totally
The lives and motivations of the builders of Stonehenge, blocked by the Moon. This is called a total solar eclipse
4,000 years dead, may be lost in antiquity. Yet how can (Figure 2.23). Now look instead at points B and C in Figure
FIGURE 2 . 22 Different
Penumbra parts of the Sun are blocked
at different places within the
Umbra
Moon’s shadow. An observer
Moon
B in the umbra (A) sees a total
Sun solar eclipse, observers in
A D
the penumbra (B and C) see
C a partially eclipsed Sun, and
observers in region D see an
annular solar eclipse.
Solar eclipses occur when
the Moon blocks the Sun.
Total eclipse Partial eclipse Annular eclipse

2.22. From these points an observer can see one side of the
disk of the Sun but not the other. This outer region, which
is only partially in shadow, is the penumbra. If a point on
the surface of Earth passes through the Moon’s penumbra,
the result is a partial solar eclipse, in which the disk of the
Moon blocks the light from a portion of the Sun’s disk.
In the third type of eclipse, called an annular solar
eclipse, the Sun appears as a bright ring surrounding the
dark disk of the Moon (Figure 2.24). An observer at point
D is far enough from the Moon that the Moon’s angular
diameter appears smaller than the Sun’s. You may be won-
dering how one eclipse can be total and another annular.
Two things make this possible. One is a fluke of nature:
The diameter of the Sun is about 400 times the diameter
of the Moon, and the Sun is about 400 times farther away
from Earth than the Moon is. As a result, the Moon and Sun
FIGURE 2 . 23 The full spectacle of a total eclipse of the Sun.
have almost exactly the same apparent size in the sky. The
other factor is that the Moon’s orbit is not a perfect circle.
So when the Moon and Earth are a bit closer together than
on average, the Moon appears larger in the sky than the
Sun. An eclipse occurring at that time will be total. When
the Moon and Earth are farther apart than on average, the
Moon appears smaller than the Sun, so eclipses occurring
during this time will be annular. Sequential pictures of
total and annular solar eclipses are shown in Figure 2.25.
As it turns out, among all solar eclipses, one-third are
total, one-third are annular, and one-third are seen only
as a partial eclipse.
Figure 2.26a illustrates the geometry of a solar eclipse,
with the Moon’s shadow falling on the surface of Earth.
Note that figures like this (or like Figures 2.18 and 2.19) are
FIGURE 2 . 2 4 An annular eclipse, in which the Moon does not seldom drawn to scale. Instead, they show Earth and the
quite cover the Sun. Note that Earth’s atmosphere distorts the Moon much closer together than they are in reality. The rea-
shape of the Sun when close to the horizon.
FIGURE 2 . 25 Time sequences of images of the Sun taken during a total solar eclipse (a) and
during an annular solar eclipse (b). The sunspots seen in (b) will be discussed in Chapter 14.
(a)
(b)
(a) Solar eclipse geometry (not to scale)
Total or annular eclipse Partial eclipse
Sun
Moon
Earth
(b) Solar eclipse to scale
Moon Earth
(c) Lunar eclipse geometry (not to scale)
Sun Moon
Earth
(d) Lunar eclipse to scale
1/ °
2
Moon
Earth 1/ °
2
FIGURE 2 . 26 (a, b) A solar eclipse occurs when the shadow of the Moon falls on the surface of
Earth. (c, d) A lunar eclipse occurs when the Moon passes through Earth’s shadow. Note that (b) and
(d) are drawn to proper scale.
son for distorting figures in this way is simple: there is not quite large. In fact, with a bit of thought and a pencil and
enough room on the page to draw them correctly and still paper, you can convince yourself that the Moon’s penum-
keep the smaller details visible. The 384,400-km distance bra, where it hits Earth, must have a diameter about twice
between Earth and the Moon is over 60 times the radius of the diameter of the Moon itself, or almost 7,000 km. This
Earth and over 220 times the radius of the Moon. The rela- part of the shadow is large enough to cover a substantial
tive sizes and distances between Earth and the Moon are fraction of Earth, so partial solar eclipses are often seen
roughly equivalent to the difference between a basketball from much of the planet. In contrast, the path along which
and a baseball placed 7 meters apart. Figure 2.26b shows a total solar eclipse can be seen (Figure 2.27) covers only
the geometry of a solar eclipse with Earth, the Moon, and a tiny fraction of the surface of Earth. Earth is so close to
the separation between them drawn to scale. Compare this the tip of the Moon’s umbra that even when the distance
drawing to Figure 2.26a and you will understand why artis- between Earth and the Moon is at a minimum, the umbra
tic license is normally taken in drawings of Earth and the is only 269 km wide at the surface of Earth. As the Moon
Moon. If the Sun were drawn to scale in Figure 2.26b, it moves along in its orbit, this tiny shadow sweeps across
would be 6⁄10 of a meter across and located almost 64 meters the face of Earth at breakneck speed. The Moon moves in
off the left side of the page. its orbit around Earth at a speed of about 3,400 km/h, and
From any particular location—say, your home—the its shadow sweeps across the disk of Earth at the same rate.
probability that you will see a partial solar eclipse is very Earth is also rotating on its axis with a velocity of 1,670
much greater than the likelihood of your seeing a total km/h at the equator (and less than that at other latitudes).
solar eclipse. The reason is that the Moon’s penumbra is The situation is further complicated by the fact that the
TOTAL AND ANNULAR SOLAR ECLIPSE PATHS: 2001–2020

Aug 1, 2008 FIGURE 2 . 2 7 The paths
May 31, 2003 of total and annular solar
Mar 20, 2015 eclipses predicted for the early
21st century. Solar eclipses
Aug 01, 2008 occurring in Earth’s polar
60°N
May 20, 2012 May 20, 2012
regions cover more territory
Aug 21, 2017 Oct 3, 2005 because the Moon’s shadow
Jun 21, 2020 Jun 10, 2002 hits the ground obliquely.
Jun 10, 2002
Jul 22, 2009
30°N Mar 29, 2006
Mar 9, 2016
Nov 3, 2013
Latitude
May 10, 2013 Dec 14, 2001 Jan 15, 2010 Dec 26, 2019
Mar 9, 2016
0° Apr 8, 2005
Jul 22, 2009
Jul 2, 2019 Sep 1, 2016 May 10, 2013
Sep 22, 2006
Jun 21, 2001 Feb 26, 2017
30°S Jul 11, 2010
Jan 26, 2009 Nov 13, 2012
Nov 13, 2012 Dec 4, 2002
Dec 14, 2020 Dec 4, 2002
Feb 26, 2017
60°S
Nov 23, 2003
Apr 29, 2014
Feb 7, 2008
Total Eclipse
Annular Eclipse
180°W 120°W 60°W 0° 60°E 120°E 180°E

Longitude
ANNULAR SOLAR ECLIPSES:2001–2025 TOTAL SOLAR ECLIPSES:2001–2025
Aug 1, 2008
May 31, 2003
Mar 20, 2015
Jun 10, 2021
Apr 8, 2024
May 20, 2012
Aug 21, 2017

Jun 10, 2002 Oct 14, 2023
0 500 1,000 1,500 0 500 1,000 1,500

Kilometers Dec 14, 2001 Kilometers
Apr 8, 2015
Moon’s shadow falls on the curved surface of Earth. You When all of these effects are considered, the result is
may have noticed that the image displayed by an old-style that a total solar eclipse can never last longer than 71⁄2 min-
overhead or modern LCD projector is distorted when the utes and is usually significantly shorter. Even so, it is one of
beam is not perpendicular to the screen. Similarly, the cur- the most amazing and awesome sights in nature. People all
vature of Earth often causes the region shaded by the Moon over the world flock to the most remote corners of Earth to
during a solar eclipse to be elongated by differing amounts. witness the fleeting spectacle of the bright disk of the Sun
The curvature can even cause an eclipse that started out as blotted out of the daytime sky, leaving behind the eerie glow
annular to become total. of the Sun’s outer atmosphere (see Figure 2.23).
times the diameter of the Moon—a penumbral eclipse is

(a)
noticeable only when the Moon passes within about 1,000
km of the umbra.
Total eclipses of the Moon are relatively common, and
many of us have seen at least one. As you looked up at that
copper-colored disk (Figure 2.28a), you may have wondered
how it was possible to even see the Moon when completely
immersed in Earth’s shadow. To visualize how, imagine
standing on the Moon during a total lunar eclipse. What
would you see? One thing you would not see is the Sun,
because the Sun would be hiding behind Earth’s disk. (If you
did see some of the Sun peeking out from behind Earth, you
would be standing on a partially eclipsed Moon, as shown
in Figure 2.28b). What you would see is the Earth’s dark
disk surrounded by a thin, bright, reddish ring. That ring
is Earth’s atmosphere, with sunlight coming from behind
Earth and being scattered by dust and other small atmo-
spheric particles. It is reddish for the same reason that the
(b)
Sun appears reddened during sunset. (In Chapter 8 we’ll dis-
cuss scattering and the reddening of sunlight.) Now safely
back on Earth, you realize it is that bright, reddish ring of
Earth’s atmosphere that is lighting up the Moon’s surface
during a total lunar eclipse.
If you understand the geometry of solar and lunar
eclipses, you will also understand why so many more peo-
ple have experienced a total lunar eclipse than have expe-
rienced a total solar eclipse. To see a total solar eclipse, you
must be located within that very narrow band of the Moon’s
shadow as it moves across Earth’s surface (see Figure 2.27).
On the other hand, when the Moon is immersed in Earth’s
shadow, anyone located in the hemisphere of Earth that is
facing the Moon can see it.
FIGURE 2 . 28 (a) A total lunar eclipse. (b) The progress of a Eclipse Seasons Occur Roughly
partial lunar eclipse. Note the size of Earth’s shadow compared to
the size of the Moon. Twice Every 11 Months
We all know from experience that we don’t see a lunar
Lunar eclipses (Figure 2.28) are very different in char- eclipse every time the Moon is full, nor do we observe a solar
acter from solar eclipses. The geometry of a lunar eclipse is eclipse every time the Moon is new. So this must be telling
shown in Figure 2.26c (and is drawn to scale in Figure us something about how the Moon’s orbit around Earth is
2.26d). Because Earth is much larger than the Moon, the oriented with respect to Earth’s orbit around the Sun. If the
dark umbra of Earth’s shadow Moon’s orbit were in exactly the same plane as the orbit of
at the distance of the Moon is Lunar eclipses last Earth (imagine Earth, the Moon, and the Sun all sitting on
about 9,200 km in diameter, much longer than the same flat tabletop), then the Moon would pass directly
or over 21⁄2 times the diameter solar eclipses. between Earth and the Sun at every new Moon. The Moon’s
of the Moon. A total lunar shadow would pass across the face of Earth, and we would
eclipse is a much more leisurely affair than a total solar see a solar eclipse. Similarly, Earth would pass directly
eclipse, with the Moon spending as long as 1 hour and 40 between the Sun and the Moon every synodic month, and
minutes in the umbra of Earth’s shadow. A penumbral lunar each full Moon would be marked by a lunar eclipse.
eclipse occurs when the Moon passes through the penum- Solar and lunar eclipses do not happen every month,
bra of Earth’s shadow (to review what the penumbra is, see because the Moon’s orbit does not lie in exactly the same
Figure 2.22). A penumbral eclipse can be unspectacular: its plane as the orbit of Earth. Look at Figure 2.29 to see how
appearance from Earth is nothing more than a fading in the this works. The plane of the Moon’s orbit about Earth is
brightness of the full Moon. Although the penumbra of Earth inclined by about 5.2° with respect to the plane of Earth’s
is 16,000 km across at the distance of the Moon—over four orbit about the Sun. The line along which the two orbital
Moon’s orbit Earth’s orbit

Line of nodes
Eclipses Eclipses
impossible possible
New Moon
Line of nodes
Missed solar
eclipse Line of nodes
Eclipses Eclipses
possible impossible
5.2°
Full Moon
Missed lunar
Line of nodes eclipse
1 The intersection of the plane of the Moon’s 2 When the line of nodes points near the 3 Here the Moon, Earth, and Sun do
tilted orbit and the plane of Earth’s orbit Sun, then the Moon, Earth, and Sun not line up, because of the tilt of the
is called the line of nodes. can line up. Eclipses are possible. Moon’s orbit. Eclipses are impossible.
FIGURE 2 . 29 Eclipses are possible only when the Sun, Moon, and Earth lie along (or very close
to) an imaginary line known as the line of nodes. When the Sun does not lie along the line of nodes,
Earth passes under or over the shadow of a new Moon, and a full Moon passes under or over the
shadow of Earth.
planes intersect is called the line of nodes. For part of the 18.6 years to complete one “wobble” of 360°, so we say that
year, the line of nodes passes close to the Sun. During these the line of nodes regresses at a rate of 360° every 18.6 years,
times, called eclipse seasons, a new Moon passes between or 19.4° per year. This amounts to about a 20-day regression
the Sun and Earth, casting its shadow on Earth’s surface each year. If January 1 marks the middle of an eclipse sea-
and causing a solar eclipse. Similarly, a full Moon occurring son, the next eclipse season will be centered around June
during an eclipse season passes through Earth’s shadow, 20, and the one after that around December 10.
and a lunar eclipse results. An eclipse season lasts only 38 We have come far since our nomadic ancestor looked
days. That is how long the Sun is close enough to the line at the sky and saw there a mystical reflection of the pat-
of nodes for eclipses to occur. Most of the time, the line of terns and events that marked the life of the tribe. Yet when
nodes points farther away from the Sun, and Earth, Moon, we look at the sky today, our sense of awe and majesty is
and Sun cannot line up closely enough for an eclipse to no less than that experienced by that long-ago tribesman.
occur. A solar eclipse cannot take place, because the shadow Our distant ancestors had to look for patterns in the world
of a new Moon passes “above” or “below” Earth. Similarly, to survive. This same human impulse to seek patterns led
no lunar eclipse can occur, because a full Moon passes astrologers to look for connections between ourselves and
“above” or “below” the shadow of Earth. the heavens, and later led people to seek patterns that gave
If the plane of the Moon’s orbit always had the same ori- birth to science. We now know that the patterns of the sky
entation, then eclipse seasons would occur twice a year, as are connected to us much more directly than is any mysti-
suggested by Figure 2.29. In actuality, eclipse seasons occur cal link invented by an astrologer. The patterns and changes
about every 5 months and 20 days. The roughly 10-day differ- that we see in the sky are caused by the same forces of
ence is due to the fact that the plane of the Moon’s orbit slowly nature that bind us to our planet and that cause the wind
wobbles, much like the wobble of a spinning plate balanced to blow and the rain to fall. They are the same forces that
on the end of a circus performer’s stick. As it does so, the line push the blood through our veins and carry the electric
of nodes changes direction. This wobble rotates in the direc- impulses of thought through our brains. So far in our look
tion opposite the direction of the motion of the Moon in its at the changing patterns of the sky, we have only glimpsed
orbit. (That is, the line of nodes moves clockwise as viewed these connections, so it is to these underlying causes that
from above Earth’s orbital plane.) It takes the Moon’s orbit we now turn our attention.
as we will see in Chapter 3, the development of a physi-

Seeing the Forest for the Trees cal understanding of the world around us. Much of who
and what we are as a thinking species has its origins in
Patterns in the sky change in lockstep with the cycles of our experience of the larger universe.
life on Earth. It is no surprise that our ancestors sought The errors in perception that shaped our views of
some link between the stars and the events in their lives, the universe throughout most of human history are both
but early astrologers were ultimately limited by their understandable and forgivable. It is remarkably difficult
misconceptions about the universe. Earth does not reside to directly sense the effects of Earth’s motion. As you read
at the “center of creation,” nor does the Sun move from this, you are probably (depending on your latitude) mov-
“house” to “house” along the ecliptic. The eerie spec- ing at more than 1,000 km/h on a circular path around
tacle of a solar eclipse blotting out the Sun is no more Earth’s axis, while Earth itself is moving at over 100,000
mystical than the shadow you cast on the sidewalk on a km/h in its orbit about the Sun. (And the Sun itself is
sunny afternoon. The heavens are a place not of magic moving through our galaxy at almost 1 million km/h.) Yet
but of physical law, knowable though observation and you feel none of this motion. Objects around you share
experiment and subject to test by the scientific method. your motion, so to you they are motionless. The telltale
The signs of the zodiac that grace the entertainment sec- signs of your motion are far too subtle to perceive directly.
tion of your local newspaper and the covers of supermar- As you watch the Sun and stars cross the sky, it certainly
ket tabloids are anachronisms—pictures born from the seems as if it is you about which the cosmos revolves. In
human imagination and painted onto the random splash fact, without the benefit of discussions such as ours, it
of stars across the night sky. would be difficult to avoid that impression.
Although astrologers were off the mark in their con- When we replace simple perception with careful
clusions, their quest was well motivated. The motions experiment and reason, however, this commonsense
of Earth and the properties of the Sun do, indeed, give notion evaporates before our eyes. Minute changes in
rise to the most basic of all the patterns faced by life on the direction of starlight through the year provide proof of
Earth. Earth’s rotation is responsible for the coming of Earth’s motion in its orbit, and they enable us to measure
night and day. Earth’s axial tilt and its passage around our path around the Sun. And as our physical understand-
the Sun bring the changing of the seasons. Changes in ing of Earth’s motion improves, even the blowing of the
the direction in which sunlight falls on Earth cause wind and the need for a fi re on a cold, dark winter night
dramatic differences in climate from the equator to the become side effects of Earth’s motions through space.
poles. These patterns set the stage for the evolution of life In our journey we will encounter again many of the
on Earth, and they remain with us today, buried deep motions discussed in this chapter, but from a rather
within the genetic code of our species. Even as humans different perspective. So far, we have concentrated on
begin to venture beyond Earth and into space, we carry describing the motion of Earth about the Sun and of the
these patterns with us in everything from the length of Moon about Earth. From here we will take the step that
our cycle of waking and sleeping, to the temperature we separates post-Renaissance science from all that came
prefer, to the amount of light we need in order to see. before, by taking up the question of why these motions
Much of human culture also has its roots in the appar- are as they are. The search for understanding will lead
ent motions of celestial objects. Many of our legends and us to a discovery that changed our perception of the uni-
traditions arose in the ancient view of the sky as a place verse and our place in it—that the same natural forces
of gods and spirits. At the same time, patterns of objects that dictate the path of a well-hit baseball also govern the
in the sky spurred the development of mathematics and, clockwork motions of Earth, the Moon, and planets.
Summary 9. We tend to associate certain constellations with certain

times of year. For example, we see the zodiacal con-
stellation Gemini in the Northern Hemisphere’s win-
• Earth’s rotation on its axis causes the Sun and stars to
ter (Southern Hemisphere’s summer) and the zodiacal
appear to move through the sky.
constellation Sagittarius in the Northern Hemisphere’s
• The specific stars we see at night depend on where we are summer. Why do we not see Sagittarius in the Northern
on Earth and where Earth is in its orbit around the Sun. Hemisphere’s winter (Southern Hemisphere’s summer)
or Gemini in the Northern Hemisphere’s summer?
• A frame of reference is a coordinate system within which
an observer measures positions and motions. 10. Assume that you are flying along in a jetliner.
a. Defi ne your frame of reference.
• The tilt of Earth’s axis determines the seasons.
b. What relative motions take place within your frame
• The motion of the Moon in its orbit around Earth shapes of reference?
the phases of the Moon.
11. The tilt of Jupiter’s rotational axis is 3°. If Earth’s axis
• The phase of the Moon is determined by how much of had this tilt, explain how it would affect our seasons.
its bright side is visible from Earth.
*12. Describe the Sun’s apparent motion on the celestial
• Special alignments of the Sun, Earth, and Moon result sphere
in solar and lunar eclipses. a. at the time of vernal equinox.
b. at the time of summer solstice in Earth’s Northern
Hemisphere.
SmartWork, Norton’s online homework system,
13. Why is winter solstice not the coldest time of year?
questions, plus additional conceptual exercises. If 14. There is only a certain region on Earth when at some
your instructor assigns questions in SmartWork, log in at time of year the Sun can appear precisely at the zenith.
smartwork.wwnorton.com. Describe this region.
15. Many cities have main streets laid out in east–west and
north–south alignments.
Student Questions a. Why are there frequent traffic jams on east–west
streets during both morning and evening rush hours
within a few weeks of the equinoxes?
1. Some of the mythologies created by early civilizations b. Considering your conclusion in (a), if you work in the
as they gazed at the sky persist today. Give at least one city during the day, would you rather live east or west
example. of the city?
*2. Explain why the celestial sphere is not really a sphere but *16. Earth spins on its axis but wobbles like a top.
is instead a useful fictional feature. a. How long does it take to complete one spin?
b. How long does it take to complete one wobble?
3. Earth has a North Pole, a South Pole, and an equator.
What are their equivalents on the celestial sphere? 17. Why do we always see the same face of the Moon?
4. Polaris was used for navigation by seafaring sailors such 18. What is the approximate time of day when you see the
as Columbus as they sailed from Europe to the New full Moon near the meridian? At what time is the first
World. When Magellan sailed the South Seas, he could quarter (waxing) Moon on the eastern horizon? Use a
not use Polaris for navigation. Explain why. sketch to help explain your answers.
5. If you were standing at Earth’s North Pole, where would *19. Assume that the Moon’s orbit is circular. Suppose you
you see the north celestial pole relative to your zenith? are standing on the side of the Moon that faces Earth.
a. How would Earth appear to move in the sky as the
6. If you were standing at Earth’s South Pole, which stars
Moon made one revolution around Earth?
would you see rising and setting?
b. How would the “phases of Earth” appear to you, as
7. Where on Earth can you stand and, over the course of a compared to the phases of the Moon as seen from
year, see the entire sky? Earth?
8. What do we call the group of constellations through *20. Sometimes artists paint the horns of the crescent Moon
which the Sun appears to move over the course of a pointing toward the horizon. Is this depiction realistic?
year? Explain.
21. From your own home, why are you more likely to witness 34. Let’s say you use a protractor to estimate an angle of 40°
a partial eclipse of the Sun rather than a total eclipse? between your zenith and Polaris. Are you in the conti-
nental United States or Canada?
22. Why do we not see a lunar eclipse each time the Moon
is full or witness a solar eclipse each time the Moon is 35. Suppose the tilt of Earth’s equator relative to its orbit
new? were 10° instead of 23.5°. At what latitudes would the
Arctic and Antarctic Circles and the two Tropics be
23. Why does the fully eclipsed Moon appear reddish?
located?
24. The true length of a year is not 365 days, but actually
36. Suppose you would like to witness the rarity known as
about 3651⁄4 days. How do we handle this extra quarter
“midnight Sun” (when the Sun appears just above the
day to keep our calendars from getting out of sync?
northern horizon at midnight), but you don’t want to
**25. Does the occurrence of solar and lunar eclipses disprove travel any farther north than necessary.
the notion that the Sun and the Moon both orbit around a. How far north (that is, to which latitude) would you
Earth? Explain your reasoning. have to go?
b. When would you make this trip?
**37. Assume that Earth is a perfect sphere with a radius
of 6,400 km. What distance at Earth’s equator corre-
sponds to 1° of longitude? What distance corresponds
26. Earth is spinning along at 1,674 km/h at the equator. to 1° of latitude? Estimate the distance correspond-
Do your own calculation to fi nd Earth’s equatorial ing to 1° of longitude at latitudes of 30° and 60°. What
diameter. would be the distance of 1° of latitude at these same
latitudes?
27. Assume that rain is falling at a speed of 5 meters per
second (m/s) and you are driving along in the rain at a 38. Eratosthenes estimated the size of Earth more than
leisurely 5 m/s (or 18 km/h). Estimate the angle from the 2,000 years ago (see Math Tools 2.1.) Repeat his calcu-
vertical at which the rain appears to be falling. lation. The latitude of Alexandria, his hometown, was
30.7° north. The latitude of Syene, 790 km to the south,
**28. Suppose you are an astronaut on the Moon, which now
was 23.5° north. Compare Eratosthenes’ calculation
becomes your inertial frame of reference. Explain the
with the modern value of 6,371 km for Earth’s average
relative motion of Earth as seen from your lunar frame
radius.
of reference.
39. The vernal equinox is now in the zodiacal constellation
29. The Moon’s orbit is tilted by about 5° relative to Earth’s
of Pisces. Wobbling of Earth’s axis will eventually cause
orbit around the Sun. What is the highest altitude in the
the vernal equinox to move into Aquarius, beginning the
sky that the Moon can reach, as seen in Philadelphia
legendary, long-awaited “Age of Aquarius.” How long, on
(latitude 40° north)?
average, does the vernal equinox spend in each of the
30. Refer to Appendix 4. On the basis of the tilt of its rota- 12 zodiacal constellations?
tion axis, which planet experiences the most extreme
*40. Stonehenge was erected roughly 4,000 years ago. Refer-
seasons? Explain your answer.
ring to the zodiacal constellations shown in Figure 2.10,
31. Imagine that you are standing on the South Pole at the identify the constellation in which these ancient build-
time of the southern summer solstice. ers saw the vernal equinox.
a. How far above the horizon will the Sun be at noon?
41. Referring to Figure 2.17a, estimate when Vega, the fi fth
b. How far above (or below) the horizon will the Sun be
brightest star in our sky, will once again be the northern
at midnight?
pole star.
*32. Find out the latitude where you live. Draw and label a
42. There are 29.53 days between full Moons (the synodic
diagram showing that your latitude is the same as (a)
period). On average, how long does it take for the Moon
the altitude of the north celestial pole and (b) the angle
to move from one zodiacal constellation to the next?
(along the meridian) between the celestial equator and
your local zenith. What is the noontime altitude of the 43. The apparent diameter of the Moon is approximately 1⁄2°.
Sun as seen from your home at the times of winter sol- About how long does it take the Moon to move its own
stice and summer solstice? diameter through the star field?
33. The southernmost star in the famous Southern Cross 44. The Moon has a radius of 1,737 km, with an average dis-
constellation lies approximately 65° south of the celes- tance of 3.780 × 105 km from Earth’s surface. The Sun
tial equator. Using an atlas, determine from which US has a radius of 696,000 km, with an average distance of
states is the entire Southern Cross visible? 1.496 × 108 km from Earth. Show why the apparent sizes
of the Moon and Sun in our sky are approximately the *48. Which, if either, occurs more often: (a) total eclipses of
same. the Moon seen from Earth, or (b) total eclipses of the
Sun seen from the Moon? Explain your answer.
45. Earth has an average radius of 6,371 km. If you were
standing on the Moon, how much larger would Earth 49. Suppose you would like to see a solar eclipse without
appear in the lunar sky than the Moon appears in our venturing outside the continental United States. Refer
sky? to Figure 2.27.
a. When will be the next opportunity for you to see a
46. If the Moon were in its same orbital plane, but twice as
total solar eclipse?
far from Earth, which of the following would happen?
b. When will be the next opportunity for you to see an
a. The phases that the Moon goes through would remain
annular solar eclipse?
unchanged.
b. Total eclipses of the Sun would not be possible.
c. Total eclipses of the Moon would not be possible.
*47. In what way would the length of the “eclipse season” Astronomy. Study Plans include animations, read-
change if the plane of the Moon’s orbit were inclined ing outlines, vocabulary flashcards, and multiple-choice
less than its current 5.2° to the plane of Earth’s orbit? quizzes, plus links to premium content in SmartWork and
Explain your answer. the ebook. Visit www.wwnorton.com/studyspace.
The Newtonian principle of gravitation is now more
firmly established, on the basis of reason, than it would
be were the government to step in, and to make it an
article of necessary faith. Reason and experiment have
been indulged, and error has fled before them.
THOMAS JEFFERSON (1743 –1826)
The International Space Station in orbit about Earth.

CHAP TER 3
Gravity and Orbits—

A Celestial Ballet
3.1 Gravity!
K E Y CONCEP T S
Today, most of us know that the planets, including Earth, In this chapter we follow the story of the birth of
orbit around the Sun and that gravity—one of the fun-
modern science as astronomers and physicists
damental forces of nature—holds them in orbit. Yet this
together discovered regular patterns in the motions of
was not always so. Only 500 years or so have passed since
a soft-spoken Polish monk named Nicolaus Copernicus the planets and then went on to explain those patterns
(1473–1543) (Figure 3.1) started a revolution when he revived with fundamental physical laws. Along the way we will
the idea, discarded by the Greeks 2,000 years earlier, that explore
the Sun rather than Earth lies at the center of all creation. • Empirical rules discovered by Johannes Kepler that
To those with a 16th century Western philosophical view, describe the elliptical orbits of planets around the Sun.
this suggestion seemed outlandish. To think that Coperni-
• Theoretical physical laws discovered by Isaac Newton
cus wanted us to believe that humankind—the pinnacle of
creation—resides anywhere but at the center of all things! and Galileo Galilei that govern the motion of all objects.
To imagine that we occupy but one of several planets cir- • How proportionality is used to describe patterns and
cling the Sun’s central fi re, that we are nothing but a pebble relationships in nature.
among all the pebbles on the beach—absurd! • How Newton’s laws of motion and gravitation combine
It would be easy from our “modern” perspective to
to explain planetary orbits.
chuckle at the naïveté of our ancestors and their geocen-
tric (Earth-centered) view of the universe, but to do so • How theories lead to new knowledge, such as the way
would be unfair. After all, a universe with Earth located at Newton’s derivation of Kepler’s third law is used to
its very center was a perfect fit to what our ancestors saw measure the masses of objects from observations of
in the world around them. The previous chapter showed orbital motions.
that hard evidence of the motions of Earth was remarkably
difficult to come by. To a scholar educated in Greek and
Roman philosophy, Copernicus’s view of the universe sim-
ply made no sense. Wishing to avoid the controversy that
his theory would certainly cause, Copernicus chose not to
publish his ideas until late in his life. In fact, his great work
De revolutionibus orbium coelestium (“On the Revolutions
of the Heavenly Spheres”) did not appear until the year of
his death. This work pointed the way toward our modern
cosmological principle.
57
58 Chapter 3 Gravity and Orbits—A Celestial Ballet
(lef t) FIGURE 3.1 Nicolaus

Copernicus rejected the ancient
Greek belief in an Earth-
centered universe and replaced
it with one centered on the Sun.
(right) FIGURE 3. 2
Johannes Kepler explained the
motions of the planets with
three empirically based laws.
Copernicus knew his ideas would be controversial, but other object in its vicinity. These motions range from the
he had no way of guessing the long-term consequences of almost circular orbits of some planets to the extremely elon-
his fresh view of the universe. In retrospect, we see that he gated orbits of comets. A comet’s orbit may carry it from a
knocked the bottom out from location tens of thousands of astronomical units1 from the
under the house of cards rep- Copernicus's Sun to a point somewhere inside the orbit of Mercury and
resenting Western philoso- heliocentric back out again. Within this grand symphony, subthemes
phers’ view of the world cosmological model arise. A chorus of particles orbiting the giant planets gives
around them, and the follow- knocked humankind rise to majestic systems of rings, which in turn play coun-
ing centuries would see that from the center terpoint to the gravitational ballet of the planets and their
house of cards slowly fall of the universe. moons. The analogy between the motions of objects in the
apart. The repercussions of Solar System and the patterns of music is not new. German
Copernicus’s insight not only would shape our understand- mathematician and astronomer Johannes Kepler (1571–1630)
ing of the universe around us, but would change the direc- (Figure 3.2), who will figure prominently in our story, titled
tion of the progress of human civilization itself. his great work of 1619 Harmonices mundi (“The Harmony
Even today we need to avoid too much complacency of the Worlds”).
about what we think we know. People often confuse know- As our grasp of the universe has expanded, we have
ing the name of something with actually understanding come to realize that our Solar System is but one gravita-
something. We happily talk about spacecraft in orbit about tional opus in a far larger opera. Gravity binds stars into the
Earth, or planets in orbit about the Sun, but remarkably few colossal groups that we call galaxies. It is gravity that holds
people understand what those words mean. When asked the planets and stars together and keeps the thin blanket of
why astronauts float about the cabin of a spacecraft, most air we breathe close to the surface of the planet we live on.
educated adults answer, “The spacecraft has escaped Earth’s It is gravity that caused a vast interstellar cloud of gas and
gravity.” Yet the gravitational force acting on a space shuttle dust to collapse 4.5 billion years ago to form our Sun and
orbiting Earth is only slightly weaker than when the shuttle Solar System. It is gravity that gives space and time their
is sitting on the launchpad. In fact, were it not for Earth’s very shape. As we continue our journey outward through
gravity, the spacecraft would not orbit Earth at all! the cosmos, we will come to each of these ideas in turn,
Copernicus himself knew nothing of gravity, but his and time and time again we will fi nd gravity at the center
ideas inevitably led to it. As physicists and astronomers have of our growing understanding.
come to a better understanding of gravity, they have realized
that in most respects it is gravity that holds the universe
together. Our Solar System is a gravitational symphony. The 1
Recall from Chapter 2 that an astronomical unit (abbreviated AU) is
Sun’s gravity shapes the motions of the planets and every
the average distance between the Sun and Earth.
3.2 An Empirical Beginning: Kepler Describes the Observed Motions of the Planets 59
3.2 An Empirical Beginning:

Kepler Describes the
Observed Motions
of the Planets
Now that we have extolled the wonders of gravity and the
role it plays in the universe, you might expect us immedi-
ately to discuss what gravity is and how it works. But the
great minds that brought us to our modern understanding
did not have the benefit of our 20/20 hindsight. All they
could do was watch the motions of the planets over the
course of months and years and puzzle over what they saw
(see Excursions 3.1). With no way even to judge the dis-
tances to the planets, it is no wonder that it took humans
thousands of years to begin to see the reality behind the
celestial patterns before their eyes. So, in this chapter we
will follow these scientists as each of their discoveries led
to our modern understanding of gravity.
In Chapter 1 we painted a picture of science as a world-
FIGURE 3.6 Tycho Brahe, known commonly as Tycho, was one
view in which nature is governed by physical laws and in
of the great astronomical observers of all time.
which mathematical descriptions of these physical laws are
used to explain natural phenomena. But how do scientists
go about discovering these physical laws? When facing phe-
nomena as complex and puzzling as the motions of the plan- for powering the majority of life on Earth. But you did not
ets in the heavens, where can we fi nd a toehold? Just as the start out to discover photosynthesis. You started out noting
wise sailor settles for any port in a storm, the wise scientist that plants have green leaves.
knows that when facing a complex and poorly understood The quest to first note and then accurately describe pat-
phenomenon, there may be little choice but to turn directly terns in nature is called empirical science. Empirical science
to the information our senses provide. We carefully observe often involves a great deal of creativity and not a small amount
the phenomenon under study, systematically recording as of pure (but educated) guesswork. Copernicus’s theory that
much information as we can as accurately as possible. As Earth and the planets move in circular orbits about the Sun is
our observations start piling up, we look for patterns in an example of empirical science. Copernicus did not under-
those observations and start trying to formulate rules that stand why the planets move about the Sun, but he did realize
seem to describe what we have seen. that his heliocentric (Sun-centered) picture provided a much
Imagine that you are a scientist from another world, simpler description of the observed motions of planets than
setting foot for the fi rst time on Earth. You notice right a model with Earth at its center did. Copernicus’s work was
away that many interesting structures are sticking out of revolutionary because he was able to see beyond the geocen-
the ground on this unexplored planet. As you record your tric prejudice of his time and to think the unthinkable—that
observations, you fi nd that some of these structures are perhaps Earth is “merely” one planet among many.
large, some are small, some spread out over the ground, Copernicus’s work paved the way for another great
some stick up in the air, and so on. But after a time you empiricist, Johannes Kepler. Science has often benefited
realize that almost all of these structures are covered with from unlikely and fortunate collaborations, and Kepler’s is
some kind of appendage, and those appendages are green. one such story. Kepler, a mathematician who had studied
So you form a descriptive rule about these objects (call the ideas of Copernicus, worked in 1600 as an assistant to
them “plants”): most have green appendages. You decide Tycho Brahe (1546–1601) (Figure 3.6), a Danish astronomer
that green appendages must be fundamental to the nature who was a fi rm believer in an Earth-centered universe.
of plants, so you begin to study what makes these append- Although Tycho, as he is commonly known, is described
ages green. After a time you discover that the green appear- as anything but a pleasant individual, he was also one of
ance always comes from the same chemical substance, and the greatest observational astronomers of all time. Toiling
when you study that chemical substance you fi nd that it away through long nights with primitive equipment, Tycho
can absorb light and turn water and carbon dioxide into amassed a wealth of remarkably accurate naked-eye obser-
more complex organic molecules. You have discovered vations of the positions of the planets over the course of
chlorophyll and photosynthesis, the process responsible decades. Kepler, using Tycho’s data, took the next major step
The Motions of the Planets in the Sky

Ancient peoples had long been aware of the five planets that these moving objects would occasionally turn about in
(“wandering stars”) that moved in a generally eastward direc- a contrary manner, move westward for a while, and then
tion among the “fi xed stars.” Ancient astronomers also knew return to their normal eastward travel. This odd behavior of
the five “naked eye” planets—Mercury, Venus, Mars, Jupiter,
and Saturn—created a puzzling problem for the geocentric
(Earth-centered) model of the Heavens as it was summa-
rized by the Alexandrian astronomer Ptolemy in A.D. 150.
Although a few ancient Greek astronomers did con-
sider the possibility that Earth revolved around the
Sun, other astronomers of the day were skeptical;
“if this were so, we would actually feel Earth’s
motion,” they claimed. Most astronomers there-
fore preferred the geocentric model, in which
the Sun, Moon, and planets all moved in per-
fect circles around a stationary Earth, as
shown in Figure 3.3a, with the “fi xed stars”
being located somewhere way out there
beyond the planets. Although a stationary
Earth may have “felt right” to astronomers
in those early times, the geocentric model
in its simplest form failed to explain that
curious planetary behavior that we now call
retrograde motion.
To account for retrograde motion of the plan-
ets, Ptolemy had to resort to an embellishment
called an “epicycle,” a small circle superposed on
each planet’s larger circle, as seen in Figure 3.3b.
While traveling along its larger circle, a planet would
at the same time be moving along its smaller circle. When
(a)
its motion along the smaller circle was in a direction oppo-
FIGURE 3. 3 (a) The

Ptolemaic view of the Heavens.
(b) Additional loops called
“epicycles” were added to each
planet’s circle around Earth to
explain its retrograde motion. Moon
(Diurnal motion refers to daily
motion.)
Diurnal
motion
Earth
Mercury
Venus
Sun
Mars
Jupiter
Saturn
(b)
site to the forward motion of the larger circle, its forward

motion would be reversed. For nearly 1,500 years,
this awkward model of the heavens was the accepted
paradigm in the Western world.
In 1543, a heliocentric (Sun-centered) model (Fig-
ure 3.4) proposed by Copernicus provided a simpler
explanation of retrograde motion. In this model, the
outer of the planets known at that time—Mars, Jupi-
ter, and Saturn—appear to interrupt their east-
ward prograde motion and move backward in a
westward direction when Earth overtakes them
in their orbits. Likewise, Mercury and Venus
appear to reverse direction in their inner orbits
when overtaking Earth. With the exception of
the Sun, all Solar System objects exhibit retro-
grade motion, although the magnitude of the
effect diminishes with increasing distance from
Earth. Figure 3.5 shows a time-lapse sequence of
Mars going through its retrograde “loop.”
Retrograde motion is only apparent, not real.
We have all experienced how relative motions can
fool us. If we are in a car or train and we pass a slower-
moving car or train, it can seem to us that the other
vehicle is moving backward. Without an external
frame of reference, it can be hard to tell which vehi-
cle is moving and in what way. Copernicus provided
that frame of reference for the Sun and its planets.
FIGURE 3. 4 The Copernican heliocentric view of the
Solar System (II through VII) and the fi xed stars (I).
FIGURE 3. 5 The apparent

retrograde motion of Mars
through the Taurus-Gemini star
field as Earth overtook the red
planet in 2007–08.
toward understanding the motions of the planets. Working becomes more and more elongated. Kepler found that the
fi rst primarily with Tycho’s observations of Mars, Kepler orbit of each planet is an ellipse with the Sun located at one
was able to deduce three empirical rules that elegantly and focus. This result is now
accurately describe the motions of the planets. These three known as Kepler’s first law of The Sun is at one
rules are now generally referred to as Kepler’s laws. planetary motion. (You might focus of a planet’s
ask, “If the Sun is located at elliptical orbit.
one focus, what is at the other
focus?” The answer is, nothing but empty space.)
Kepler’s First Law: Planets Figure 3.8 illustrates Kepler’s fi rst law and shows how
Move on Elliptical Orbits with the features of an ellipse elegantly match observed plan-
etary motions. Let’s look more closely at Figure 3.8a to see
the Sun at One Focus some of these features. The dashed lines represent the two
When Kepler used Copernicus’s model to calculate where in
the sky a planet should be at a particular time, he expected
Tycho’s data to confi rm circular orbits but instead found FIGURE 3.8 (a) Planets move on elliptical orbits with the
disturbing disagreement between his predictions and the Sun at one focus. (b) Ellipses range from circles to elongated
observations. He was not the fi rst to notice such discrepan- eccentric shapes.
cies. Rather than discarding Copernicus’s ideas, however,
(a)
Kepler played with Copernicus’s heliocentric model until
Kepler’s First Law
it matched Tycho’s observations.
Kepler discovered that if he replaced Copernicus’s cir- Planetary orbits are ellipses
cular orbits with elongated elliptical orbits, his predictions with the Sun at one focus.
fit Tycho’s observations almost perfectly. You probably think
Planet
of an ellipse as an oval shape, 1 The size of an ellipse is
but to make sense of Kepler’s Planetary orbits specified by its semimajor
discovery we need to be a bit are ellipses. axis, A.
Semiminor axis (B)
more precise about what an
ellipse is. The most concrete way to defi ne an ellipse is to
Semimajor axis (A)
call it the shape that results when you attach the two ends Focus
of a piece of string to a piece of paper, stretch the string tight
with the tip of a pencil, and then draw around those two Focus
eA
points keeping the string taut (Figure 3.7). Each of the points
at which the string is attached is called a focus (plural: foci)
of the ellipse. The closer the two foci are to each other, the
more nearly circular the ellipse is. In fact, a circle is just an 2 The shape of an ellipse is
specified by its eccentricity, e.
ellipse with the two foci at the same place. (To see this, just
think about the shape you would draw if the two ends of
the string were attached at the same spot. In this case, each
half of the string would become a radius of the circle.) As (b)
the two foci are moved farther apart, however, the ellipse 3 The greater the eccentricity,
the more elongated the ellipse.
e = 0.983
e = 0.958
Foci
FIGURE 3.7 We can draw
e = 0.745
an ellipse by attaching a length
Focus Focus of string to a piece of paper at
two points (called foci; singular:
focus) and then pulling the e=0
string around as shown.
of Earth and a circle centered on the Sun. By contrast, one of

(a)
the many characteristics that distinguish the dwarf planet
Pluto from its classical cousins is its highly eccentric orbit,
Earth’s orbit is only
slightly eccentric. as seen in Figure 3.9b. With an eccentricity of 0.244, the
1 AU
distance between the Sun and Pluto varies by 24.4 percent
from its average value. Pluto’s orbit is noticeably oblong, and
Sun its center is noticeably displaced from the Sun.
e = 0.017 Circular orbit with
same semimajor axis,
centered on Sun Kepler’s Second Law: Planets Sweep
Out Equal Areas in Equal Times
The next empirical rule that Kepler found has to do with
(b) Pluto’s eccentric orbit is how fast planets move at different places on their orbits. A
both noticeably elongated planet moves most rapidly when it is closest to the Sun and
and noticeably offset as is at its slowest when it is far-
compared with a circle
39.2 AU centered on the Sun. thest from the Sun. The aver- Planets move fastest
age speed of Earth in its orbit when they are closest
about the Sun is 29.8 kilome- to the Sun.
Sun
ters per second (km/s). When
e = 0.244
Circular orbit with Earth is closest to the Sun, it travels at 30.3 km/s. When it
same semimajor axis, is farthest from the Sun, it travels at 29.3 km/s.
centered on Sun Kepler found an elegant way to describe the changing
speed of a planet in its orbit about the Sun. Look at Figure
3.10, which shows a planet at six different points in its orbit.
FIGURE 3.9 The shapes of the orbits of Earth (a) and Pluto (b) Imagine a straight line connecting the Sun with this planet.
compared with circles centered on the Sun. We can think of this line as “sweeping out” an area as it moves
with the planet from one point to another. Area A (in red) is
swept out between times t1 and t2, area B (in blue) is swept
main axes of the ellipse. Half of the length of the long axis out between times t3 and t4, and area C (in green) is swept out
of the ellipse is called the semimajor axis of the ellipse, between times t5 and t6. When the planet is closest to the Sun
often denoted by the letter A. The semimajor axis of an orbit
turns out to be a handy way to describe the orbit because,
apart from being half the longer dimension of the ellipse, FIGURE 3.10 An imaginary line between a planet and the Sun
it is also the average distance between one focus and the sweeps out an area as the planet orbits. Kepler’s second law
ellipse itself. The average distance between the Sun and states that if the three intervals of time shown are equal, then
Earth, for example, equals the semimajor axis of Earth’s the three areas A, B, and C will be the same.
orbit. The same is true for the orbits of all the other planets. Kepler’s Second Law
AstroTour: Kepler’s Laws
In the case of a circular orbit, the semimajor axis is just
the radius of the circle. Some ellipses, on the other hand, t1
Planet on
are very elongated. When describing the shape of an ellipse, elliptical orbit
we speak of its eccentricity. We define the eccentricity of an
ellipse as the separation between the two foci divided by
the length of the long axis. A circle has an eccentricity of 0 Sun t6
because the two foci coincide at the center of the circle. The A C
more elongated the ellipse becomes, the closer its eccentric- t5
ity gets to 1 (Figure 3.8b). Most planets have nearly circu- 1 These three
lar orbits with eccentricities close to 0. The eccentricity of areas—A, B, and
C—will be equal…
Earth’s orbit, for example, is 0.017, which means that the
distance between the Sun and Earth departs from its aver- t2
B
age value by only 1.7 percent. You may recall from Chapter
2 that the times when Earth is closest or farthest from the t3 t4
Sun have almost nothing to do with our seasons, and you
can perhaps get a better feeling for this when you look at 2 …if these time
intervals are equal.
Figure 3.9a. It is hard to tell the difference between the orbit
(area A in Figure 3.10) it is moving rapidly, but the distance

between the planet and the Sun is small. Kepler realized that TA B L E 3 .1
changes in the distance between the Sun and a planet and
changes in the speed of a planet work together to produce a
surprising result: the area swept out by a planet in the same Kepler’s Third Law: P 2 = A3
amount of time is always the same, regardless of the location The Orbital Properties of the Classical and Dwarf Planets
of the planet in its orbit. In Figure 3.10, this means that if the
three time intervals are equal (that is, t1→t2 = t3→t4 = t5→t6), Period Semimajor P2
then the three areas A, B, and C will be equal as well. ⎯3
Planet P (years) Axis A (AU) A
This is Kepler’s second law, which is also referred to as
0.2412
Kepler’s law of equal areas. It states that the imaginary line Mercury 0.241 0.387 ⎯⎯⎯ = 1.00
0.3873
connecting a planet to the Sun
0.6152
sweeps out equal areas in equal A planet “sweeps Venus 0.615 0.723 ⎯⎯⎯3 = 1.00
0.723
times, regardless of where the out” equal areas in 1.0002
planet is in its orbit. Note that equal times. Earth 1.000 1.000 ⎯⎯⎯ = 1.00
1.0003
this law applies to only one
1.8812
planet at a time. The area swept out by Earth in a given time Mars 1.881 1.524 ⎯⎯⎯ = 1.00
1.5243
is always the same. Likewise, the area swept out by Mars in
1.5592
a given time is always the same. But the area swept out by Ceres 4.599 2.765 ⎯⎯⎯ = 1.00
2.7653
Earth and the area swept out by Mars in a given time are not
11.862
the same. AstroTour: Kepler's Laws Jupiter 11.86 5.204 ⎯⎯⎯ = 1.00
5.2043
29.462
Saturn 29.46 9.582 ⎯⎯⎯ = 0.99*
9.5823
Kepler’s Third Law: Planetary Orbits 84.012
Uranus 84.01 19.201 ⎯⎯⎯ = 1.00
Reveal a Harmony of the Worlds 19.2013
164.792
Kepler’s fi rst law describes the shapes of planetary orbits, Neptune 164.79 30.047 ⎯⎯⎯ = 1.00
30.0473
and Kepler’s second law describes how the speed of a planet
247.682
changes as it travels along its orbit. But neither of these laws Pluto 247.68 39.236 ⎯⎯⎯ = 1.02*
39.2363
tells us how long it takes a planet to complete one orbit
557.002
about the Sun (referred to as the period of the orbit). Nor Eris 557.00 67.696 ⎯⎯⎯ = 1.00
67.6963
do these laws tell us how this orbital period depends on
the distance between the Sun and a planet. Kepler surely *These ratios are not exactly 1.00, due to slight perturbations from
the gravity of other planets.
saw this lack of a relationship between orbital period and
semimajor axis as a major flaw in his understanding of
planetary motions. In such cases one looks for patterns and
mathematical relationships. the Sun, whizzes along its short orbit at an average speed
Planets that are closer to the Sun do not have as far to of 47.9 km/s, completing one revolution in only 88 days. At
go to complete one orbit as do planets that are farther from a distance of 30.1 AU from the Sun, Neptune lumbers along
the Sun. Jupiter, for example, has an average distance of 5.2 at an average speed of 5.48 km/s, taking 164.8 years to make
astronomical units (AU) from the Sun—5.2 times as far from it once around the Sun.
the Sun as Earth is. That Kepler discovered a sim-
means Jupiter must travel 5.2 Outer planets have ple mathematical relationship The square of a
times farther in its orbit about farther to go and between the period of a planet’s planet’s orbital period
the Sun than Earth does in its move more slowly orbit and its distance from the equals the cube of the
orbit. We might guess, then, in their orbits Sun. Kepler’s third law states orbit’s semimajor axis.
that if the two planets were around the Sun. that the square of the period of
traveling at the same speed, a planet’s orbit, measured in years, is equal to the cube of the
Jupiter would complete one orbit in 5.2 years. But such a semimajor axis of the planet’s orbit, measured in astronomi-
guess would be wrong. Jupiter takes almost 12 years to com- cal units. Written as an equation, the law says,
plete one orbit. Clearly, Jupiter not only has farther to go in
its orbit but must be moving more slowly than Earth as well. (Pyears)2 = (AAU)3,
This trend holds true for all the planets. As we go farther
out from the Sun, the circumferences of planetary orbits where Pyears represents the period of the orbit expressed
become greater, and the speeds at which the planets travel in Earth years, and AAU is the semimajor axis of the orbit
decrease. Mercury, at an average distance of 0.387 AU from expressed in astronomical units.
3.3 The Rise of Scientific Theory: Newton’s Laws Govern the Motion of All Objects 65
Kepler’s Third Law

106 Eris
Pluto
FIGURE 3.11 A plot of A3
Neptune 1 Outer planets move more
slowly and have farther to versus P 2 for the eight classical
104 Uranus
(A in astronomical units)3
go on their orbits than do and three dwarf planets in our

inner planets… Solar System shows that they
Saturn
obey Kepler’s third law. (Note
Jupiter that by plotting powers of 10
2 …leading to Kepler’s third
102 law, (Pyears )2 = (AAU)3. on each axis, we are able to fi t
Ceres both large and small values on
the same plot. We will do this
Mars frequently.)
Earth
100
Venus
Mercury
10–2
10–2 100 102 104 106
(P in Earth years) 2
This equation is a case in which astronomers use non- pursuing are only the fi rst step in the study of a phenom-
standard units as a matter of convenience. Years are handy enon. Such empirical laws describe a particular phenom-
units for measuring the periods of orbits, and astronomi- enon and are even useful in predicting what will happen
cal units are handy units for measuring the sizes of orbits. in the future, but they do little to explain that behavior.
When we use years and astronomical units as our units, we Taken at face value, empirical rules offer little insight into
get the simple relationship just shown. However, it is impor- the more fundamental laws describing nature. Kepler was
tant to realize that our choice of units in no way changes the able to characterize the orbits of planets as ellipses, but he
physical relationship we are studying. If we instead stayed did not understand why they should be so.
with standard metric units, this relationship would read Once the empirical rules that describe a particular phe-
(Pseconds)2 = 3 × 1019 (Ameters)3. nomenon have been discovered, a modern scientist will next
To judge for yourself how well Kepler’s third law works, try to understand those empirical rules in terms of more
look at Table 3.1. Here you can see the periods and semima- general physical principles or laws. Beginning with basic
jor axes of the orbits of the eight classical and three of the physical principles and using the tools of mathematics, the
dwarf planets, along with the values of the ratio P 2 divided scientist works to derive the empirically determined rules.
by A3. These data are also plotted in Figure 3.11. This rela- Or sometimes the scientist starts with physical laws and
tionship was so beautiful to Kepler that he referred to it as predicts relationships, which are then verified (or falsified)
his harmonic law or, more poetically, as the “Harmony of empirically. This technique is sometimes referred to as the
the Worlds.” AstroTour: Kepler's Laws “theoretical approach” to science, although in practice, if the
relevant physical laws are already understood, the process
is often circumvented. A scientist may make a theoretical
prediction about the behavior of a system and then compare
3.3 The Rise of Scientific the prediction with experimental data directly to see how
well they fit. In fact, today a great deal of science is done
Theory: Newton’s without any attempt to invent an empirical rule, because
the relevant physical laws are already known.
Laws Govern the When the relevant physical laws are not known—as was
the case for planetary motion—the empirical rules become
Motion of All Objects a way of discovering the physical laws themselves. Can we
invent hypothetical physical laws that will enable us to
When investigating a newly discovered or poorly under- derive the empirical rules? If so, what other predictions
stood phenomenon, an empirical approach is often the only might we make on the basis of these hypothetical laws? Are
available way to proceed, as we saw in Kepler’s case. The these predictions also borne out by experiment and obser-
development of Kepler’s laws of planetary motion was an vation? If so, then we may have discovered something more
intellectual accomplishment with few peers. Yet to a mod- fundamental about the way the universe works. This is how
ern scientist, the empirical rules that Kepler spent his life physical laws are discovered and tested.
(lef t) FIGURE 3.12 Sir

Isaac Newton’s three laws of
motion formed the basis for
classical mechanics.
(right) FIGURE 3.13 Galileo

Galilei, known commonly
as Galileo, was among the
first to make telescopic
astronomical observations and
laid the physical framework for
Newton’s laws.
One of the earliest great advances in theoretical science great insight formed the starting point for Newton’s tour de
was also arguably one of the greatest intellectual accom- force that was to come.
plishments in the history of our species. In many ways, the
work of Sir Isaac Newton
(1642–1727) (Figure 3.12) on Newton’s laws of Galileo Describes Objects in Motion
the nature of motion set the motion are the
standard for what we now basis of classical In a strange quirk of history, the physical law almost invari-
refer to as “scientific theory” mechanics. ably referred to today as Newton’s first law of motion did
and “physical law.” Building not originate with Newton at all. It was the brainchild of
on the work of Kepler and others, Newton proposed three a contemporary of Kepler by the name of Galileo Galilei
laws that he believed govern the motions of all objects in (1564–1642), known commonly as Galileo (Figure 3.13).
the heavens and on Earth. Today Newton’s laws remain the Galileo is probably best known to the general public as the
basis for what is known as classical mechanics. Newton’s fi rst person to use a telescope to make significant discover-
laws themselves are beautifully elegant, and the relation- ies about the heavens and to report those discoveries. He
ships they describe between such everyday concepts as force, was among the fi rst to see craters on the Moon3 and the
velocity, acceleration, and mass are accessible to all. fi rst to realize that the “nebulous” Milky Way was actually
You might reasonably ask why Newton’s laws are an made up of a myriad of individual stars. In the history of
important stop on our journey through 21st Century Astron- science, however, Galileo’s work on the motion of objects
omy. “After all,” you might say, “this is a book about astron- is at least as fundamental a contribution as his astronomi-
omy, not physics.” Yes, but Newton’s laws of motion are cal observations. For example, by carefully rolling balls
essential to our understanding of the motions of the planets down an inclined plane, Galileo demonstrated that gravity
and all other celestial bodies. In a very real sense, it is with accelerates all objects at the same rate, and he determined
Newton’s laws, published in 1687, that truly modern astron- the value of Earth’s gravitational acceleration, g. We will
omy got its start. It was these laws that enabled Newton to return to g later in the chapter. AstroTour: Velocity,
look at the motion of a shot fi red from a cannon and see as Acceleration, Inertia
well the motions of the planets on their orbits around the By Galileo’s day, Copernicus and others had begun to
Sun. It was through Newton’s laws that Earth’s true place turn toward the view that knowledge comes from observ-
in the universe was fi nally seen. ing nature rather than only from reading the works of clas-
The work of great scientists is always built on the foun-
dation of the great scientists who came before,2 and Galileo’s 3
An English astronomer named Thomas Harriot made drawings of the
Moon several months before Galileo began his astronomical observa-
2
Newton himself is credited with the famous quote, “If I have seen tions. However, Galileo continued his scientific use of the telescope
further [than you] it is by standing upon the shoulders of giants.” and published his observations, whereas Harriot did not.
sical Greek and Roman philosophers. Yet even in the 16th and the phases of Venus had convinced Galileo that the
and 17th centuries the works of one of the greatest of these Copernican model was correct and that the church’s views
philosophers, Aristotle (384–322 B.C.), who had lived almost were wrong. In 1632, Galileo published his great work,
2,000 years earlier, still carried the weight of authority. Aris- Dialogo sopra i due massimi sistemi del mondo (“Dialogue
totle believed that the natural state of all objects was to be Concerning the Two Chief World Systems”). In the Dialogo,
at rest and that an object in motion would tend toward this the champion of the Copernican (Sun-centered) view of the
natural state. From what Aristotle observed, this seemed universe is a brilliant philosopher named Salviati. The Dia-
to be a good empirical rule about how objects in the world logo’s defender of Aristotelian authority is named Simplicio
around him behaved. For example, we notice that a cart and is as much an ignorant buffoon as the name might imply.
rolling down the street coasts to a stop when it is no lon- When Galileo published the Dialogo, he actually thought
ger being pulled. A bouncing ball eventually settles to the he had the tacit approval of the Vatican, which held to the
ground. Even an arrow shot from a bow loses much of its Aristotelian view. But when he placed a number of the
speed before striking its target. pope’s arguments in the unflattering mouth of Simplicio, he
It is seldom safe to challenge the entrenched wisdom found that the Vatican’s tolerance had limits. Fortunately
of your day, and that was especially true in Galileo’s time, for Galileo, he had more friends in high places than Bruno
when intellectual and religious authority and political did, so he spent the closing years of his life under house
power resided in the same hands. Within Galileo’s lifetime, arrest rather than ending up burned at the stake.
the Italian priest and philosopher Giordano Bruno had fallen To escape a harsher sentence from the Inquisition, Gali-
victim to the Inquisition and had been burned at the stake leo was forced to publicly recant his belief in the Coper-
for his beliefs—including Bruno’s support of Copernicus, his nican theory that Earth moves around the Sun. In one of
belief that the universe is infinite, and his suggestion that the great apocryphal stories of the history of astronomy,
Earth is but one of many habitable planets. In his writings it is said that as he left the courtroom following his sen-
on motion, Galileo similarly challenged prevailing wisdom tencing, Galileo stamped his foot on the ground and mut-
by proposing that Aristotle’s notion of an object’s tendency tered, “And yet it moves!” Galileo’s fi nal years were spent
to come to rest was the result of illusion. On the basis of his compiling his research on inertia and other ideas into a
own observations and experimental results, Galileo argued book, Discorsi e dimostrazioni mathematiche intorno a due
that in all of the cases just mentioned—indeed in every nuove scienze attenenti alla meccanica (“Discourses and
such case—there are hidden reasons why objects come to Mathematical Demonstrations Relating to Two New Sci-
rest. There is friction as the axle of the cart rubs against its ences”), which was published in 1638 outside the Inquisi-
bearing, resisting the motion and eventually bringing it to tion’s jurisdiction.
a halt. Every time a ball bounces, its shape is distorted, and
what we might think of as “internal friction” within the ball
causes it to bounce lower each time. The resistance of air,
which the arrow must push out of the way and which drags
Newton’s First Law:
against the arrow’s shaft, slows the arrow’s progress. Objects at Rest Stay at Rest;
Galileo agreed with Aristotle that an object at rest
remains at rest unless something causes it to move. But in
Objects in Motion Stay in Motion
disagreement with Aristotle, Galileo asserted that, left on It is a tribute to Galileo that his law of inertia became the
its own, an object in motion cornerstone of physics as Newton’s first law of motion. We
will remain in motion. Specifi- Galileo found that an can understand inertia in terms of what we discovered dur-
cally, Galileo said that an object left in motion ing discussions of relative motion and frames of reference in
object in motion will continue remains in motion. Chapter 2. Recall that within a frame of reference, only the
moving along a straight line relative motions between objects have any meaning. This
with a constant speed until an unbalanced force acts on it is the same as saying that there is no perceptible difference
to change its state of motion. When we speak of an unbalanced between an object at rest and an object in uniform motion.
force—as we will frequently throughout this chapter—we What objects are at rest and what objects are in motion any-
mean the net force acting on an object.4 Galileo referred to way? The object at rest beside you on the front seat of your car
this resistance to change in an object’s state of motion from as you drive down the highway is moving at 60 miles per hour
an unbalanced force as inertia. (mph) according to a bystander along the side of the road—
Galileo developed the idea of inertia relatively early in but it is moving at 120 mph according to a car in oncoming
his career, but much of his later life was taken up by conflict traffic. These two perspectives are equally valid.
with the Catholic Church over his support of the Copernican The connection between inertia and the relative nature
system. His astronomical observations of Jupiter’s moons of motion is so fundamental that a frame of reference mov-
ing in a straight line at a constant speed is referred to as
4
The “net,” or “resultant,” force is a single force that represents a com- an inertial frame of reference. Motion is meaningful only
bination of all of the individual forces acting on an object. when measured relative to an inertial frame of reference,
and any inertial frame of reference is as good as any other. law of motion goes on to say that if there is an unbalanced
The realization that the laws of physics are the same in any force acting on an object, then
inertial frame of reference is one of the deepest insights the object’s motion does Unbalanced forces
ever made into the nature of the universe. When thought change. Even more, Newton’s cause changes
of in this way, of course an object moving in a straight line second law tells us how the in motion.
at a constant speed remains in motion. As illustrated in object’s motion changes in
Figure 3.14a, in the frame of reference of that object, it is response to that force.
already at rest. Before going any further, it would be wise to pause to
be sure that we are all together on this. In the preceding
paragraphs we spoke of “changes in an object’s motion,” but
Newton’s Second Law: Motion Is what does that phrase really mean? When you are in the
driver’s seat of a car, a number of controls are at your dis-
Changed by Unbalanced Forces posal. On the floor are a gas pedal and a brake pedal. You
Newton often gets credit for Galileo’s insight about inertia use these to make the car speed up or slow down. A change
because it was Newton who took the crucial next step. New- in speed is one way the motion of an object can change. But
ton’s fi rst law says that in the absence of an unbalanced also remember the steering wheel in your hands. When
force, an object’s motion does not change. Newton’s second you are moving down the road and you turn the wheel,
(a)
An object (in this case a cup of
Liquid is
coffee) moving in a straight line
at a constant speed is at rest in
level.
its own inertial frame of reference.
Constant velocity
(b)
1 Any change in either speed 2 “Speeding up” means 3 “Slowing down” means
or direction is an acceleration. acceleration is in the acceleration is opposite
direction of motion. the direction of motion.
Side view Side view
Speeding up Slowing down
Acceleration
Velocity Turning
4 “Turning” at a constant speed

means acceleration is perpendicular
to the direction of motion. FIGURE 3.14 (a) An object moving in a straight
line at a constant speed is at rest in its own inertial
frame of reference. (b) Any change in the velocity of
an object is an acceleration. When you are driving,
for example, any time your speed changes or you
follow a curve in the road, you are experiencing an
acceleration. (Throughout the text, velocity arrows
will be shown as red and acceleration arrows will be
Rear view shown as green.)
Newton’s Second Law:

Force (F )
Acceleration (a) =
Mass (m)
(a) (b) FIGURE 3.15 Newton’s

Pushing on an object (applying a force ) The harder you push something (more
causes the object to accelerate . force ), the greater the acceleration. second law of motion says that
the acceleration experienced
by an object is determined by
the force acting on the object,
divided by the object’s mass.
F a More force F a (Throughout the text, force
means more
acceleration . arrows will be shown as blue.)
(100 kg) (100 kg)
More mass
means less
acceleration .
How much velocity changes
Acceleration = ______________________________
How long the change takes to happen.
(c)
For example, if an object’s speed goes from 5 meters per
second (m/s) to 15 m/s, then the change in velocity is 10 m/s.
If that change happens over the course of 2 seconds, then the
F a
acceleration is 10 m/s divided by 2 seconds, which equals 5
meters per second per second. This is the same as saying “5
meters per second squared,” which is written as 5 m/s2 or 5
(200 kg)
m s–2. AstroTour: Velocity, Acceleration, Inertia
Because the gas pedal on a car is often called the “accel-
erator,” some people think acceleration means that an object
is speeding up. But we need to stress that, as used in phys-
ics, any change in motion is an acceleration. Figure 3.14b
illustrates the point. Slamming on your brakes and going
The more massive an object, the less it accelerates
in response to a given force , and vice versa. from 60 to 0 mph in 4 seconds is just as much acceleration
as going from 0 to 60 mph in 4 seconds. Similarly, the accel-
eration you experience as you go through a fast, tight turn at
a constant speed is every bit as real as the acceleration you
your speed does not necessarily change, but the direction feel when you slam your foot on the gas pedal or the brake
of your motion does. A change in direction is also a kind of pedal. Speeding up, slowing down, turning left, turning
change in motion. right—if you are not moving in a straight line at a constant
Together, the speed and direction of an object’s motion speed, you are experiencing an acceleration.
are called the object’s velocity. The rate at which the veloc- Newton’s second law of motion says that changes in
ity of an object changes is called acceleration. Acceleration motion—accelerations—are caused by unbalanced forces.
actually refers to how rapidly the change in velocity hap- The acceleration that an object experiences depends on
pens. For example, if you go two things, as shown in Fig-
from 0 to 60 mph in 4 seconds, Acceleration measures ure 3.15. First, it depends on Greater force means
you feel yourself being pushed how quickly a change the unbalanced force acting greater acceleration.
back into the seat, but it is in motion takes place. on the object to change its
really your seat back that is motion. When all the forces acting on an object balance
shoving your body forward, causing you to accelerate along each other—making the total force on the object zero—the
with the car. If you take 2 minutes to go from 0 to 60 mph, object is not accelerating. If the forces acting on the object
on the other hand, the acceleration is so slight that you do not add up to zero, then there is an unbalanced force
hardly notice it. To formalize this concept a bit, your accel- and the object accelerates (Figure 3.15a). This is a pretty
eration is determined by how much your velocity changes, commonsense idea. The stronger the unbalanced force, the
divided by how long it takes for that change to happen: greater the acceleration. In fact, the acceleration of an
object is proportional to the unbalanced force applied (Fig- Instead of spelling this out in words every time, we can
ure 3.15b). Push on something twice as hard and it experi- introduce a convenient bit of shorthand—a for acceleration,
ences twice as much acceleration. Push on something three F for force, and m for mass—so we get
times as hard and its acceleration is three times as great.
F
(The idea of proportionality, discussed in Foundations 3.1, a = __
m.
will be used over and over again throughout our journey.)
The resulting change in motion occurs in the direction in Note that Newton’s second law is often written as F = ma,
which the unbalanced force is imposed. Push something giving force as units of mass multiplied by units of accel-
forward and it speeds up. Push it to the left and it veers in eration, or kilograms times meters per second squared (kg
that direction. m/s2). These units of force are aptly named newtons, abbre-
The acceleration that an object experiences also depends viated N.
on the degree to which the object resists changes in motion. This equation is the succinct mathematical statement
Some objects—for example, a baseball—are easily shoved of Newton’s second law of motion. If you are comfortable
around by humans. A baseball is thrown at great velocity with mathematics, this elegant expression may speak to
by the pitcher, only to be hit with a bat and have its motion you clearly and directly. If not, when you see this equation
abruptly changed again. The remind yourself that Newton’s second law is nothing more
hard-hit line drive comes sud- Mass is the property than the embodiment of three commonsense ideas: (1) when
denly to a stop in the glove of of matter that resists you push on an object, that object accelerates in the direc-
the second-base player. A changes in motion. tion in which you are pushing; (2) the harder you push on
baseball resists changes in its an object, the more it accelerates; and (3) the more massive
motion; that is, it has inertia—but not too much inertia. the object is, the harder it is to change its state of motion.
Other objects are less obliging. A piece of solid iron the
size of a baseball would make a poor substitute in the game.
A pitcher would be very hard-pressed to throw such an iron Newton’s Third Law:
ball hard enough to get it over home plate (and if he did,
the catcher would have an even more dangerous job). A
Whatever Is Pushed, Pushes Back
baseball and a ball of iron may be the same size, but they Imagine that you are a child again, sitting in a wagon or
are quite different in the degree to which they resist changes standing on a skateboard and pushing yourself along with
in their motion. The property of an object that determines your foot. Each shove of your foot against the ground sends
its resistance to changes in motion—the measure of an you faster along your way. But why does this happen? Your
object’s inertia—is referred to as the object’s mass (Figure muscles flex and your foot exerts a force on the ground.
3.15c). In other words, the greater the mass, the greater the (Earth does not respond much to that force, because its great
resistance. mass gives it great inertia.) Yet this does not explain why
You probably knew this answer intuitively. The iron ball you experience an acceleration. The fact that you accelerate
is “heftier” and therefore harder to throw than a baseball. at the same time means that as you push on the ground, the
However, you may not have thought much about what we ground must be pushing back on you.
really mean when we say that a ball of iron “has more Part of Newton’s genius was his ability to see sublime
mass”—or “is more massive”—than a baseball. You might patterns in such mundane events. Newton realized that
say that the ball of iron is made up of “more stuff” than the every time one object exerts a force on another, a matching
baseball; but again, what is meant by “more stuff”? If you force is exerted by the second object on the first. That second
grapple with this question for a time, you may fi nd yourself force is exactly as strong as the fi rst force but is in exactly
chasing the question around in circles. When it comes right the opposite direction. The child pushes back on Earth, and
down to it, the property of Earth pushes the child forward. A canoe paddle pushes
matter that we refer to as Acceleration is force backward through the water, and the water pushes forward
“mass” is nothing more and divided by mass. on the paddle, sending the canoe along its way. A rocket
nothing less than the degree engine pushes hot gases out of its nozzle, and those hot gases
to which an object resists changes in its motion. So if we push back on the rocket, propelling it into space.
want to know how an object’s motion is changing, we need All of these are examples of Newton’s third law of
to know two things: What unbalanced force is acting on the motion, which says that forces always come in pairs, and
object, and what is the resistance of the object to that force? those pairs are always equal
We can put this into equation form as follows: in strength but opposite in For every force there
direction. The forces in these is an equal and
( )
The The force acting to change action-reaction pairs always opposite force.
the object’s motion
acceleration = ___________________ Force act on two different objects.
= _____ .
experienced The object’s resistance Mass Your weight pushes down on the floor, and the floor pushes
by an object to that change back on you with the same amount of force. For every force
F O U N D AT I O N S 3 .1
Proportionality
Often in this text we will say that one quantity is proportional C O N S TA N T S O F P R O P O R T I O N A L I T Y
to another. Proportionality is a way of getting the gist of how
something works—understanding the relationships between Sometimes it is enough to know that two quantities are pro-
things—without having to actually calculate the details of one portional to each other, but sometimes it is not. What if you
case after another. need to know how much one of those bags of apples will actu-
ally set you back? We know that the full relationship between
the cost and weight of a bag of apples is that the cost is equal
PROPORTIONALITY to the price per pound of apples, multiplied by the weight of
If two quantities are proportional to each other, then making the bag. We write
one of them larger means making the other quantity larger
by the same factor. In other words, the ratio between the two Cost = Price per pound × Weight.
remains constant. For example, think about the weight of a bag
of apples and how much the bag costs. Double the weight of the Compare this expression with the previous one. When we
bag of apples and you double the cost. Increase the weight of say that two quantities are proportional to each other, what we
the bag of apples by a factor of 5, and the cost goes up by a fac- mean is that one quantity equals some number multiplied by the
tor of 5 as well. The cost of a bag of apples is proportional to the other quantity. The number by which one quantity is multiplied
weight of the bag of apples. We write this relationship as to get the other number is called the constant of proportional-
ity. In our example, the constant of proportionality is just the
Cost ∝ Weight,
price per pound of apples.
where the symbol ∝ means “is proportional to.” This expres- Look at the difference between the two expressions. The
sion captures the essence of the relationship between the cost fact that the cost of a bag of apples is proportional to the
and the weight of apples. It tells us that the more apples we buy, weight of the apples is a statement about the relationship
the more we will pay. between things. It is a statement about how things work.
Mass and acceleration provide another example of propor- More apples do not cost less than fewer apples. More apples
tionality. Mass is measured in units of kilograms (kg). An object cost more than fewer apples. The constant of proportional-
with a mass of 2 kg is twice as hard to accelerate as an object ity—here, the price per pound of apples—means something
with a mass of 1 kg. An object with a mass of 9 kg is three times very different. Hidden within the price per pound of apples
as hard to accelerate as an object with a mass of 3 kg. is a great deal of information, such as the cost of growing
apples, the cost of transporting them from the orchard, and
INVERSE PROPORTIONALITY the profit margin that the grocer needs to stay in business.
The constant of proportionality carries information about
If two quantities are inversely proportional to each other, then these aspects of the world.
making one of them smaller means that the other quantity Very often, physical laws work in this same fashion. Pro-
becomes larger by the same factor. For example, think about a portionalities tell us about relationships—how two things vary
trip to your grandmother’s house. If your driving speed is cut with one another. They let us get a feel for the “how” in how
in half, the time it takes you to get there is doubled. The time something works. In this chapter, for example, we will fi nd
of travel is inversely proportional to the average travel speed. that gravitational force is proportional to an object’s mass.
We write this relationship as Constants of proportionality more precisely tell us about the
1 way the universe is. The universal gravitational constant (G)
Time for a journey ∝ ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ . is a constant of proportionality that tells us about the intrinsic
Average speed driven
strength of gravitational interactions and enables calculation of
This expression tells us that if we drive half as fast, it will take the numerical value of this force. Constants of proportionality
us twice as long to get there. And if we drive twice as fast, our are needed if we are to turn an understanding of relationships
time will be cut in half. into hard numbers.
Newton’s Third Law there is always an equal and opposite force. This is one of
For every force … the few times when we can say “always” and really mean
Woman pulls it. Figure 3.16 gives a few examples. There is a great game
on rope.
Rope pulls …there is an equal hiding in Newton’s third law. It is called “fi nd the force.”
and opposite force .
on cage.
cage Look around you at all the forces at work in the world, and
Rope pulls for each force fi nd its mate. It will always be there!
on woman. Cage pulls To see how Newton’s three laws of motion work together,
on rope. think about the situation shown in Figure 3.17. An astro-
naut is adrift in space, motionless with respect to the
Floor
pushes nearby space shuttle. With no tether to pull on, how can
on woman. the astronaut get back to the ship? The answer? Throw
something. Suppose the 100-kg astronaut throws a 1-kg
Woman pushes
on floor. wrench directly away from the shuttle at a speed of 10 m/s.
Newton’s second law says that in order to cause the motion
of the wrench to change, the astronaut has to apply a force
to it in the direction away from the shuttle. Newton’s third
Both forces are equal, but law says that the wrench must therefore push back on the
the more massive bat astronaut with as much force but in the opposite direc-
experiences less acceleration . tion. The force of the wrench on the astronaut causes the
astronaut to begin drifting toward the shuttle. How fast
Bat pushes Baseball pushes will the astronaut move? Turn to Newton’s second law
on baseball. on bat.
again. A force that causes the 1-kg wrench to accelerate
to 10 m/s will have much less effect on the 100-kg astro-
naut. Because acceleration equals force divided by mass,
the 100-kg astronaut will experience only 1/100 as much
acceleration as the 1-kg wrench. The astronaut will drift
toward the shuttle, but only at the leisurely rate of 1/100
Acceleration × 10 m/s, or 0.1 m/s.
Now that we have developed a fundamental understand-
ing of motion, it is time to turn to the concept of gravity.
Road pushes Tire pushes
The unbalanced force of forward backward
the road on the tire causes against tire. against road.
the car to accelerate.
3.4 Gravity Is a Force

The upward push of the
floor balances the down- between Any Two
ward pull of gravity. There
are no unbalanced forces,
so there is no acceleration .
Objects Due to
Their Masses
Earth’s gravity Floor pushes Drop a ball and the ball falls toward the ground, picking
pulls down on man. up on man.
up speed as it falls. It accelerates toward Earth. Newton’s
second law says that where there is acceleration, there is
force. But where is the force that causes the ball to acceler-
ate? Many forces that we see in everyday life involve “direct
Man’s gravity Man pushes contact” between objects.5 The cue ball slams into the eight
pulls up on Earth. down on floor.
5
Actually the “direct contact” between “solid” objects is also force at
a distance—electric force acting at a distance between the electrons
Moon pulls Earth pulls and protons in the atoms that the objects are made of.
on Earth. on Moon.
F
FIGURE 3.16 Newton’s third law states that for every force
a there is always an equal and opposite force. These opposing
forces always act on the two different objects in the same pair.
3.4 Gravity Is a Force between Any Two Objects Due to Their Masses 73
1 An astronaut adrift in space… 2 … pushes on a wrench, which, according to

Newton’s third law, pushes back on the astronaut.
Space
shuttle
3 The wrench and the astronaut experience 4 …and move in opposite directions at constant
accelerations in inverse proportion to their velocities , in accord with Newton’s first law.
masses , as dictated by Newton’s second law…
FIGURE 3.17 According to Newton’s laws, if an astronaut adrift in space throws a wrench, the two will move in opposite
directions at speeds that are inversely proportional to their masses. (Acceleration and velocity arrows are not drawn to scale.)
ball, knocking it into the pocket. The shoe of the child nothing could be further from the truth. Where did New-
pushing along in the wagon presses directly against the ton get his ideas about gravity? What guided him in his
surface of the pavement. In development of those ideas, and how did he turn them into
cases where there is physical Gravity is “force a theory with testable predictions? How did he confront
contact between two objects, at a distance.” that theory in the crucible of experiment and observation?
the source of the forces By answering these questions, rather than simply stating
between them is easy to see. But the ball falling toward Newton’s law of gravitation, we will gain some insight into
Earth is an example of a different kind of force, one that how science is done.
acts at a distance across the intervening void of space. The
ball falling toward Earth is accelerating in response to the
force of gravity. We began this chapter with a qualitative
discussion of the fundamental role that gravity plays in
Where Do Theories Come
the universe. Having explored both Kepler’s empirical From? Newton Reasons His
description of the motions of planets about the Sun and
Newton’s laws of motion, we return now to gravity, for it
Way to a Law of Gravity
is gravity that unites these two pillars of empirical and As with inertia, the story of gravity begins with the insight
theoretical science. and observation of Galileo. Galileo discovered experimen-
You probably will not be surprised to learn that once tally that all freely falling objects accelerate toward Earth
again it is Newton we turn to for a universal law of gravi- at the same rate, regardless of their mass. Drop a marble and
tation. At this point in an introductory textbook it is cus- a cannonball at the same time
tomary to simply present Newton’s law of gravitation as and from the same height, and All objects on Earth
a “done deal” and go straight to its application, but such they will hit the ground fall with the same
a leap misses one of the most interesting aspects of this together. (If proof was needed acceleration, g.
stretch of our journey. A common misconception about that this is not solely a prop-
how science works is the notion that new theories just erty of Earth’s gravity, it was provided by astronaut David
spring fully formed into the mind of a scientist as if by Scott on the lunar surface—Figure 3.18.) The gravitational
magic. This idea is certainly supported by the apocryphal acceleration near the surface of Earth, also measured experi-
grade-school story of the apple falling on Newton’s head, mentally by Galileo, is usually written as g and has a value
literally knocking the idea of gravity into his brain. One of 9.8 m/s2 on average.6 Whether you drop a marble or a can-
could almost get the idea that scientific theories are arbi-
trary—that Newton could have invented some other law 6
Earth’s gravitational acceleration, g, is approximately 9.8 m/s2, but the
of gravity that would have worked just as well. Although value varies with location because Earth’s poles are closer to Earth’s
it might seem at fi rst glance that his work was arbitrary, center than is its equator, and because centripetal forces act against
tional force. In other words, the gravitational force on an

object on Earth is, according to Newton’s second law, the
object’s mass multiplied by the acceleration due to grav-
ity, or Fgrav = mg. Make an object twice as massive and you
double the gravitational force acting on it. Make an object
three times as massive and you triple the gravitational
force acting on it.
In precise terms, the gravitational force acting on an
object is commonly referred to as the object’s weight. On
the surface of Earth, weight is
just mass multiplied by Earth’s Your mass is a
gravitational constant, g. property of you, but
Because of our everyday use your weight depends
of language, it is easy to see upon where you are.
why people confuse mass and
weight. We often say that an object with a mass of 2 kg
“weighs” 2 kg, but it is more correct to express a weight in
terms of newtons:
Fweight = m × g,
where Fweight is an object’s weight in newtons, m is the

object’s mass in kilograms, and g is Earth’s gravitational
acceleration.
On Earth, an object with a mass of 2 kg has a weight of
2 kg × 9.8 m/s2, or 19.6 N. On the Moon, where the gravita-
tional acceleration is 1.6 m/s2, the 2-kg mass would have a
weight of 2 kg × 1.6 m/s2, or 3.2 N (Figure 3.19). Although
FIGURE 3.18 Astronaut Alan Bean’s portrait of fellow your mass remains the same wherever you are, your weight
astronaut David Scott standing on the Moon and dropping a depends on where you are. On the Moon your weight is
hammer and a falcon feather together. The two objects reached about one-sixth of your weight on Earth.
the lunar surface simultaneously. (Their lunar module was
nicknamed “Falcon.”)
FIGURE 3.19 A mass of 2 kg has a different weight (displayed
nonball, after 1 second it will be falling at a speed of 9.8 in newtons) on the Moon than it has on Earth.
m/s, after 2 seconds at 19.6 m/s, and after 3 seconds at 29.4
Terrestrial landscape Lunar landscape
m/s. (These numbers assume that we can neglect air resis-
tance, which is reasonably negligible for relatively dense
objects moving at relatively slow speeds.)
After working out the laws governing the motion of
2 kg
objects, Newton saw something deeper in Galileo’s fi nd-
ings. Newton realized that if all objects fall with the same 2 kg
acceleration, then the gravitational force on an object must
be determined by the object’s mass. To see why, look back
at Newton’s second law (acceleration equals force divided
by mass, or a = F/m ). The only way gravitational accelera-
20 20
tion can be the same for all objects is if the value of the 18 2 18 2
force divided by the mass is the same for all objects. Since
16 4 16 4
force and mass are proportional (see Foundations 3.1), then
greater mass must be accompanied by a stronger gravita- 14 6 14 6
12 8 12 8
and effectively diminish the gravitational force. The value of g ranges 10 10
from 9.78 m/s2 at the equator to 9.83 m/s2 at the poles, having a mean
value of 9.80 m/s2.
Newton’s next great insight came from applying his the force between them, and determine the value of that
third law of motion to gravity. Recall that Newton’s third something directly. Yet Newton had no such instruments.
law states that for every force there is an equal and opposite He had to look elsewhere to continue his exploration of
force. Therefore, if Earth exerts a force of 19.6 N on a 2-kg gravity.
mass sitting on its surface, then that 2-kg mass must exert It turns out that Kepler had already thought about this
a force of 19.6 N on Earth as well. Drop a 20-kg cannonball question. He reasoned that because the Sun is the focal
and it falls toward Earth, but at the same time Earth falls point for planetary orbits, the Sun must be responsible
toward the 20-kg cannonball! The reason we do not notice for exerting an influence over the motions of the planets.
the motion of Earth is that Earth is very massive. It has a lot Kepler speculated that whatever this influence is, it must
of resistance to a change in its motion. In the time it takes grow weaker with distance from the Sun. (After all, it must
a 20-kg cannonball to fall to the ground from a height of 1 surely require a stronger influence to keep Mercury whip-
kilometer (km), Earth has “fallen” toward the cannonball ping around in its tight, fast orbit than it does to keep the
by about 3.4 × 10 –21 meter, which is only about 1/300 the outer planets lumbering along their paths around the Sun.)
size of an electron! Kepler’s speculation went even further. Although he did not
Newton reasoned that this relationship should work know about forces or inertia or gravity, he did know quite a
with either object. If doubling the mass of an object doubles lot about geometry, and geometry alone suggested how this
the gravitational force between the object and Earth, then solar “influence” might change for planets progressively
doubling the mass of Earth ought to do the same thing. In farther from the Sun.
short, the gravitational force between Earth and an object As Kepler himself may have speculated, imagine that
must be equal to the product of the two masses multiplied you have a certain amount of plaster to spread over the sur-
by something: face of a sphere. If the sphere is small, you will get a thick
coat of plaster. But if the sphere is larger, the plaster has to
Gravitational force = Something × Mass of Earth spread farther and you get a thinner coat. The surface area
× Mass of object. of a sphere depends on the square of the sphere’s radius.
Double the radius of a sphere, and the sphere’s surface
If the mass of the object were three times greater, then the becomes four times what it was. If you plaster this new,
force of gravity would be three times greater. Likewise, if the larger sphere, the plaster must cover four times as much
mass of Earth were three times what it is, the force of grav- area and the thickness of the plaster will be only a fourth
ity would have to be three times greater as well. If both the of what it was on the smaller sphere. Triple the radius of
mass of Earth and the mass of the object were three times the sphere and the sphere’s surface will be nine times as
greater, the gravitational force would increase by a factor of large, and the thickness of the coat of plaster will be only
3 × 3, or 9 times. Because objects fall toward the center of a ninth as thick.
Earth, we know that this force is an attractive force acting Kepler thought the influence that the Sun exerts over
along a line between the two masses. the planets might be like the plaster in this example. As
“And by the way,” reasoned Newton, “why are we the influence of the Sun extended farther and farther into
restricting our attention to Earth’s gravity?” If gravity is a space, it would have to spread
force that depends on mass, out to cover the surface of a Gravity is governed by
then there should be a gravi- The force of gravity is larger and larger imaginary an inverse square law.
tational force between any proportional to the sphere centered on the Sun.
two masses. Suppose we have product of two (We will learn later that light works in exactly this way.)
two masses—call them mass masses. If so, then, like the thickness of the plaster, the influence
1 and mass 2, or m1 and m2 for of the Sun should be proportional to 1 divided by the
short. The gravitational force between them is something square of the distance between the Sun and a planet. We
multiplied by the product of the masses: refer to this relationship as an inverse square law (see
Connections 3.1).
Gravitational force between two objects = Kepler had an interesting idea, but not a scientific theory
Something × m1 × m2. with testable predictions. What he lacked was a good idea
of the true source of this influence and the mathematical
Realize that we have gotten this far just by combining tools to calculate how an object would move under such an
Galileo’s observations of falling objects with (1) Newton’s influence. Newton had both. If gravity is a force between
laws of motion and (2) Newton’s belief that Earth is a mass any two objects, then there should be a gravitational force
just like any other mass. There has been no wiggle room— between the Sun and each of the planets. Might this gravi-
there is nothing arbitrary in what we have done. But what tational force be the same as Kepler’s “influence”? If so, then
about that “something” in the previous expression? Today the something in Newton’s expression for gravity might be
we have instruments sensitive enough that we can put a term that diminishes according to the square of the dis-
two masses close to each other in a laboratory, measure tance between two objects. In essence, gravity might behave
C O N N E C T I O N S 3 .1
Inverse Square Laws

In this chapter we discover that gravity obeys what is called and a planet, and the force of gravity declines by a factor of
an inverse square law. This means that the force of gravity is 2 × 2 = 4, to 1⁄4 of its original strength. Triple the distance and
“inversely proportional” (see Foundations 3.1) to the square of this influence declines by a factor of 32 = 9, becoming 1⁄ 9 of its
the distance between two objects, or initial strength.
1 Gravity is only one of several inverse square laws found
Fgrav ∝ __2 .
r in nature. The other important one that we will deal with in
If two objects are moved so that they are twice as far apart this book involves radiation. The intensity of radiation from
as they were originally, the force of gravity between them an object is also proportional to 1 divided by the square of the
becomes only 1⁄4 of what it was. If two objects are moved three distance between the objects. Our discussion of radiation in
times as far apart, the force of gravity drops to 1⁄9 of its origi- Chapter 4 will present a clear picture of why an inverse square
nal value. For example, double the distance between the Sun law applies in that case.
according to an inverse square law. Newton’s expression for 3. It decreases in proportion to 1 divided by the square of
gravity now came to look like this: the distance between the two objects:
1
Gravitational force between two objects = Fgrav ∝ __2 .
m1 × m2 r
Something × ______________________ .
(Distance between objects)2 Written as a mathematical formula, the universal law of
gravitation states that
There is still a “something” left in this expression, and
m1 × m2
we realize now that the something is a constant of propor- Fgrav = G × _______ ,
tionality (see Foundations 3.1). Newton guessed that the r2
constant was a measure of the intrinsic strength of gravity where F is the force of gravity between two objects, m1
between objects and that it would turn out to be the same and m2 are the masses of objects 1 and 2, r is the distance
for all objects. He named it the universal gravitational con- between the centers of mass of the two objects, and G is the
stant, written as G. The value of G is 6.673 × 10 –11 N m2/kg2 universal gravitational constant.
(or its equivalent, m3/kg s2). Anytime you run across a statement like this, get into
the habit of pulling it apart to be sure it makes sense.
We’ll start by analyzing the algebraic expression just pre-
Putting the Pieces Together: sented. First, gravity is an attractive force between two
masses that acts along the straight line between the two
A Universal Law for Gravitation masses.7 Regardless of where you stand on the surface of
Newton had good reasons every step of the way in his think- Earth, Earth’s gravity pulls you toward the center of Earth.
ing about gravity—reasons directly tied to observations of Second, the force of gravity depends on the product of
how things in the world behave. Newton’s chain of logic the two masses. If you make m1 twice as large, then the
and reason brought him to what has come to be known as gravitational force between m1 and m2 becomes twice as
Newton’s universal law of gravitation. This law, illustrated large. Again, this relationship should make sense. Dou-
in Figure 3.20, states that gravity is a force between any two bling the mass of an object also doubles its weight. A
objects having mass and has these properties: subtle—but in retrospect very important—point lurks in
this statement. The mass that appears in the universal law
1. It is an attractive force acting along a straight line of gravitation is the same mass that appears in Newton’s
between the two objects. laws of motion. The same property of an object that gives
2. It is proportional to the mass of one object multiplied by 7

This is not obvious from the equation alone. More properly, this
the mass of the other object: equation should be written using vector notation, a form of algebra
that includes information about both the magnitude and the direc-
Fgrav ∝ m1 × m2. tion of the force.
Newton’s Universal Law of Gravitation: m 1 m2

F =G
r2
Gravity is an attractive force
that acts along the line between The force is proportional to the FIGURE 3. 20 Gravity is an
two objects. More mass product of the two masses . attractive force between two
means more objects. The force of gravity
force . depends on the masses of
F F F F the objects and the distance
m1 m2 m1
m2 between them.
Greater separation
means less force .
F F
m1 m2
The force is inversely proportional
to the square of the distance
between the masses.
it inertia is the property of the object that makes it inter- tional to the square of the distance between two objects.
act gravitationally. (This equivalence between the effect Doubling the distance between two objects reduces the
of gravitation and the effect of inertia later became the strength of gravity to (1/2)2, or 1⁄4, of its original value.
basis for Einstein’s general theory of relativity, in which Tripling the distance between two objects reduces the
mass literally warps space and time. We will return to strength of gravity to (1/3)2, or 1⁄ 9, of its original value
this idea later in our journey, when we discuss space- (Figure 3.21). Gravity is only one of several laws we will
time in Chapter 17.) encounter in which the strength of a particular effect
The third part of Newton’s universal law of gravita- diminishes in proportion to the square of the distance
tion tells us that the force of gravity is inversely propor- (see Math Tools 3.1).
FIGURE 3. 2 1 As two objects move apart, the gravitational force between them decreases by the
inverse square of the distance between them.
1 Gravity obeys an F = 1 unit

inverse square law:
F 1/ r 2
r = 1 unit
Gravitational force between masses
1
F = 1/ 4 unit
(arbitrary units)
r = 2 units
2 Doubling the separation
between two objects reduces
the gravitational force to 1/ 4
its original value.
F = 1/ 9 unit
r = 3 units
1/ 4 3 Tripling the separation
between two objects reduces
the gravitational force to 1/ 9
1/ 9
its original value.
0
0 1 2 3 4
Separation between masses (arbitrary units)
Playing with Newton’s Laws of Motion and Gravitation

If you have ever watched a child play with blocks, you have face of Earth. The expression that we arrived at says that this
probably noticed that she puts the blocks together in different acceleration (g) is determined by the mass of Earth (M⊕) and
ways, seeing what she can build. Perhaps if you are an artist, by the radius of Earth (R⊕). But the mass of the object itself (m)
you play with colors and patterns of light and dark as you cre- appears nowhere in this expression. So, according to this equa-
ate new works. If you are a musician, you might play with tones tion, changing m has no effect on the gravitational acceleration
and rhythms as you compose. Writers play with combinations of experienced by an object on Earth. In other words, our play
words. All of these uses of the term play mean the same thing. with Newton’s laws has shown us that all objects experience
Play is very serious business because it is by playing that we the same gravitational acceleration, regardless of their mass.
explore the world. And in exactly the same sense, scientists This is just what Galileo found in his experiments with falling
often play with the equations describing natural laws as they objects! We already saw that Galileo’s work shaped Newton’s
seek new insights into the world around them. We can play a thinking about gravity. Here we fi nd that Galileo’s discoveries
bit with Newton’s laws of gravitation and motion and see what about gravity are contained within Newton’s laws of motion
interesting things turn up. and gravitation.
The universal gravitational constant G has a value of 6.67 × What else can we discover? If we rearrange that last equa-
10 –11 N m 2/kg2. This indicates that gravity is a very weak force. tion a bit, so that the mass of Earth is on the left and everything
The gravitational force between two 16-pound (7.26-kg) bowl- else is on the right, we get
ing balls sitting 0.3 meter (about a foot) apart is only
g(R⊕)2
M⊕ = _____ .
Nm2 7.26 kg × 7.26 kg
Fgrav = 6.67 × 10–11 ____ × _____________ G
kg2 (0.3 m)2
–8 Everything on the right side of this equation is known. Gali-
= 3.91 × 10 N,
leo measured a value for g, the acceleration due to gravity on
or 0.0000000391 N. This is about equal to the weight on Earth the surface of Earth, almost 400 years ago; and in about 235
of a single bacterium! Gravity is such an important force in our B.C. Eratosthenes measured the radius of Earth in the manner
everyday lives only because Earth is so very massive. described in Chapter 2. The universal gravitational constant G
There are two different ways to think about the gravitational is a bit tougher, but it, too, can be measured in the laboratory.
force that Earth exerts on an object with mass m. The fi rst is For example, we can determine the value of G by measuring
to look at gravitational force from the perspective of Newton’s the slight gravitational forces between two large metal spheres.
second law of motion: gravitational force equals mass multi- With everything on the right side now known, we can calcu-
plied by gravitational acceleration, or late the mass of Earth:
Fgrav = mg. g(R⊕)2

M⊕ = _____
G
The other way to think about the force is from the perspec-
tive of the universal law of gravitation, which says that
=
(______________________
9.80 __) × (6.37 × 10 m)
m
2
s
6 2
M ⊕m
Fgrav = G ____2 . m3
(R⊕) 6.67 × 10–11 ____2
kg s
(The symbol ⊕ signifies Earth. Here, M⊕ is the mass of Earth = 5.97 × 10 kg
24
and R⊕ is the radius of Earth.) Because the force of gravity on
an object is what it is, the two expressions describing this force You may have wondered how we know the mass of Earth.
must be equal to each other. Fgrav = Fgrav, so (After all, we cannot just pick up a planet and set it on a bath-
M⊕m room scale.) Now you know. By playing with theories and equa-
mg = G ____2 .
(R⊕) tions, much as a child plays with building blocks, scientists
discover new relationships between things in the universe, and
The mass m is on both sides of the equation, so we can cancel
from those new relationships comes new knowledge.8
it out. The equation then becomes
M⊕m
g = G ____ . 8
Newton actually turned this around. He guessed at the mass
(R⊕)2
of Earth by assuming it had about the same density as typical
This is interesting. We started out to calculate the gravitational rocks. Then he used this mass and the previous equation to get a
acceleration experienced by an object of mass m on the sur- rough idea of the value of G.
3.5 Orbits Are One Body (a)

A cannonball travels over the
“Falling around” Another ground as it falls toward Earth.
v
If you have been following our discussion closely, you may
be about ready to take us to task. Kepler may have speculated
about the dependence of the solar “influence” that holds the a
planets in their orbits, and Newton may have speculated that
this influence is gravity, but physical law is not a matter of (b)
speculation! Newton could not measure the gravitational If fired faster, it travels farther in the
force between two objects in the laboratory directly, so how time it takes to fall to the ground.
did he test his universal law of gravitation? Again it was v
Kepler who provided what Newton lacked.
Newton used his laws of motion and his proposed law
of gravity to calculate the paths that planets should follow a
as they move around the Sun. When he did so, his calcula-
tions predicted that planetary orbits should be ellipses with (c)
the Sun at one focus, that equal areas should be swept out
If fired fast enough, Earth’s surface curves
during equal times, and that the square of the period of a out from under the cannonball as it falls.
planet’s orbit should vary as the cube of the semimajor axis
v
of that ellipse. In short, Newton’s universal law of gravita-
tion predicted that planets should orbit the Sun in just the
way that Kepler’s empirical laws described. This was the a
moment when it all came together. By explaining Kepler’s
laws, Newton found important corroboration for his law of
gravitation. And in the process he moved the cosmological
principle out of the realm of interesting ideas and into the
realm of testable scientific theories. To see how this hap-
pened, we need to look below the surface of how scientists
go about connecting their theoretical ideas with events in (d)
the real world.
Newton’s laws tell us how an object’s motion changes
in response to forces and how objects interact with each
other through gravity. To go from statements about how an If fired at the circular
object’s motion is changing to more practical statements velocity and parallel to the
about where an object is, we have to carefully “add up” ground, the cannonball
falls around Earth in a a
the object’s motion over time.98 To see how we can do this, vcirc
circular orbit.
let’s begin with a “thought experiment”—the same thought
experiment that helped lead Newton to his understanding
of planetary motions.
Newton Fires a Shot around the World

FIGURE 3. 22 Newton realized that a cannonball fired at
the right speed would fall around Earth in a circle. Velocity (v) is
Drop a cannonball and it falls directly to the ground, just as
indicated by a red arrow and acceleration (a) by a green arrow.
any mass does. However, if instead we fire the cannonball
out of a cannon that is level with the ground, as shown in
Figure 3.22a, it behaves differently. The ball still falls to the also traveling over the ground, following a curved path that
ground in the same time as before, but while it is falling it is carries it some horizontal distance before it fi nally lands.
The faster the cannonball is fi red from the cannon (Fig-
9
8
Learning how to make the jump from laws of gravitation and motion
ure 3.22b), the farther it will go before hitting the ground.
to calculations of the paths of the planets about the Sun led Newton AstroTour: Newton’s Laws and Universal Gravitation
to become one of the two co-inventors of the branch of mathematics In the real world this experiment reaches a natural limit.
known as “calculus.” Fortunately, we do not need calculus to build a To travel through air the cannonball must push the air out of
conceptual understanding of such motions. its way—an effect we normally refer to as “air resistance”—
which slows it down. But because this is only a thought Like Newton’s cannonball, an
experiment, we can ignore such real-world complications. astronaut falls freely around Earth.
Instead imagine that, having inertia, the cannonball con-
tinues along its course until it runs into something. As the v
cannonball is fi red faster and faster, it goes farther and
farther before hitting the ground. If the cannonball fl ies
far enough, the curvature of Earth starts to matter. As the
a
cannonball falls toward Earth, Earth’s surface “curves out
from under it” (Figure 3.22c). Eventually we reach a point
where the cannonball is flying so fast that the surface of
Earth curves away from the cannonball at exactly the same
rate at which the cannonball is falling toward Earth. This
is the case shown in Figure 3.22d. At this point the can-
nonball, which always falls toward the center of Earth, is Because both shuttle and
literally “falling around the world.” astronaut experience the same
gravitational acceleration…
In 1957 the Soviet Union used a rocket to lift an object
about the size of a basketball high enough above Earth’s v
upper atmosphere that wind resistance ceased to be a con-
cern, and Newton’s thought experiment became a matter
of great practical importance.10 9
This object, called Sputnik
a
1, was moving so fast that it fell around Earth, just as the
cannonball did in Newton’s mind. Sputnik 1 was the fi rst
human-made object to orbit Earth. In fact, this is what an
orbit is: one object falling freely around another.
The concept of orbits also answers the question of why
astronauts float freely about the cabin of a spacecraft. It is
not because they have escaped Earth’s gravity; it is Earth’s …both are independent satellites
gravity that holds them in sharing the same orbit.
their orbit. Instead the answer Orbiting means “falling
lies in Galileo’s early observa- around a world.” v
tion that any object falls in v
just the same way, regardless of its mass. The astronauts
and the spacecraft are both moving in the same direction, a
at the same speed, and are experiencing the same gravita- a
tional acceleration, so they fall around Earth together. Fig-
ure 3.23 demonstrates this point. The astronaut is orbiting
Earth just as the spacecraft is orbiting Earth. On the sur-
face of Earth our bodies try to fall toward the center of
Earth, but the ground gets in the way. We experience our FIGURE 3. 23 A “weightless” astronaut has not escaped
weight when we are standing on Earth because the ground Earth’s gravity. Rather, an astronaut and a spacecraft share the
pushes on us hard enough to counteract the force of grav- same orbit as they fall around Earth together.
ity, which is trying to pull us down. In the spacecraft, how-
ever, nothing interrupts the astronaut’s fall, because the
spacecraft is falling around Earth in just the same orbit.
The astronaut is not truly weightless. Instead the astronaut imaginary cannonball is a satellite. Sputnik 1, the fi rst arti-
is in free fall. ficial satellite (sputnik means “satellite” in Russian), was
When one object is falling around another, much more the early forerunner to the spacecraft and the astronauts,
massive object, we say that the less massive object is a satel- which are independent satellites of Earth that conveniently
lite of the more massive object. Planets are satellites of the happen to share the same orbit.
Sun, and moons are natural satellites of planets. Newton’s
10
9
Actually, wind resistance did not totally cease to be a concern. How Fast Must Newton’s
Objects in orbit within a few hundred kilometers of Earth are mov- Cannonball Fly?
ing through the thin outer part of Earth’s atmosphere. Friction caused
by this thin atmosphere will oppose the object’s motion and cause its If fi red fast enough, Newton’s cannonball falls around the
orbit to decay. world; but just how fast is “fast enough”? Newton’s orbit-
ing cannonball moves along a circular path at constant (a)

speed. This type of motion, referred to as uniform circu- 1 If not for the centripetal force
lar motion, is discussed in provided by the string…
more depth in Appendix 7. Centripetal forces
You are probably familiar maintain circular
with other examples of uni- motion.
form circular motion. For
F
example, think about a ball whirling around your head v
on a string, as shown in Figure 3.24a. If you were to let
go of the string, the ball would fly off in a straight line in
whatever direction it was traveling at the time, just as 2 …a ball moving on a circular
path would fly off along a line
Newton’s fi rst law says. It is the string that keeps this from tangent to the circle.
happening. The string exerts a steady force on the ball,
causing it constantly to change the direction of its motion,
always bending its fl ight toward the center of the circle.
This central force is called a centripetal force. Using a
more massive ball, speeding up its motion, or making the
circle smaller so that the turn is tighter all increase the
force needed to keep the ball from being carried off in a
straight line by its inertia.
In the case of Newton’s cannonball (or a satellite), there
(b)
is no string to hold the ball in 3 If not for the centripetal force
its circular motion. Instead provided by gravity…
Gravity provides
the centripetal force is pro- the centripetal
vided by gravity, as illustrated force that holds a
in Figure 3.24b. For Newton’s satellite in its orbit.
thought experiment to work,
m1
the force of gravity must be just right to keep the cannonball
moving on its circular path. In this case we can say that r
( uniform circular motion ) ( ).

The force needed for The force provided m1m2
= Fgrav = G
by gravity r2
In Appendix 7 we derive an expression for the centrip-

etal force needed to keep an object moving in a circle at a
vcirc
steady speed. If we put that expression on the left side of
this equation and the universal law of gravitation on the 4 …the same would be
m2
true for a satellite.
right side (and then do some algebra), we arrive at
___ VISUAL ANALOGY FIGURE 3. 2 4 (a) A string provides the
vcirc = √r
GM
___ , centripetal force that keeps a ball moving in a circle. (We are
ignoring the smaller force of gravity that also acts on the ball.)
(b) Similarly, gravity provides the centripetal force that holds a
where M is the mass of the orbited object and r is the satellite in a circular orbit.
radius of the circular orbit. This value is called the circu-
lar velocity.
Here is the result we were looking for. If a satellite is in
a stable circular orbit, then it must be moving at a velocity Planets Are Just like
vcirc, where vcirc is given by this expression. If the satellite
were moving at any other velocity, it would not be moving in
Newton’s Cannonball
a circular orbit. Remember the cannonball. If the cannon- We can apply this same idea to the motion of Earth around
ball were moving too slowly, it would drop below the cir- the Sun. In the case of Earth’s orbit, we already know from
cular path and hit the ground. Similarly, if the cannonball our discussion of the aberra-
were moving too fast, its motion would carry it above the tion of starlight that Earth We use Earth’s
circular orbit. Only a cannonball moving at just the right travels at a speed of 2.98 × 104 orbit to calculate the
velocity—the circular velocity—will fall around Earth on m/s or (29.8 km/s) on its orbit Sun’s mass.
a circular path. The circular velocity at Earth’s surface is about the Sun. We also know
about 8 km/s (see Math Tools 3.2). that the radius of Earth’s orbit is 1.50 × 1011 meters. So we
M AT H T O O L S 3 . 2
Achieving Circular Velocity

It is interesting to put some values into the equation for circular Newton’s cannonball would have to be traveling about 8 km/s—
velocity to show how fast Newton’s cannonball would really over 28,000 kilometers per hour (km/h)—to stay in its circular
have to travel to stay in its circular orbit. The average radius of orbit. That’s well beyond the reach of a typical cannon, but just
Earth11 is 6.37 × 10 6 meters, the mass of Earth is 5.974 × 1024 what we routinely accomplish with rockets.
kg, and the gravitational constant is 6.67 × 10 –11 N m 2/kg2. (We Suppose we wanted to put a satellite into orbit just above
have seen how each of these values is measured.) Putting these the lunar surface. How fast would it have to fly? The radius of
values into the expression for vcirc gives the Moon is 1.737 × 106 meters and its mass is 7.35 × 1022 kg.
_____________________ These values give us
_____________________
vcirc = √
(6.67 × 10–11) × (5.97 × 1024)
_____________________ = 7,900 m/s
√
–11 22
6.37 × 106 (6.67 × 10 ) × (7.35 × 10 )
_____________________
vcirc = 6 = 1,680 m/s
1.737 × 10
11
Earth is not perfectly round. Rotation causes it to bulge somewhat
at the equator. Earth’s equatorial radius is 6,378 km, its polar radius This is actually the speed of the cannonballs (projectiles) fi red
is 6,357 km, and its mean (average) radius is 6,371 km. When refer- by Big Bertha, the cannon used by the Germans to bombard
ring to Earth’s radius in this book, we will generally use its mean Paris during World War I.
radius, often expressed as 6.37 × 106 meters.
know everything about the nearly circular orbit except for Some algebra gives us
the mass of the Sun. If we can skip a step or two, a little
4π2
algebra applied to the equation for vcirc gives the mass of the P 2 = ____ × r 3.
Sun (M⊙)102 as GM⊙
(vcirc)2 × r Once again, our play has gotten us somewhere really
M⊙ = ________
G interesting. The square of the period of an orbit is equal to
a constant (4π2/GM⊙) multiplied by the cube of the radius of
(2.98 × 104 __ s ) × (1.50 × 10 m)
m 2
__________________________
11
the orbit. This is just Kepler’s third law applied to circular
=
m3 orbits. This is how Newton showed, at least in the special
6.67 × 10–11 ___2
kg case of a circular orbit, that Kepler’s third law—his beautiful
= 1.99 × 1030 kg. “harmony of the worlds”—is a direct consequence of the way
objects move under the force of gravity. A more complete
Once again, a bit of play has taken us places we might not treatment of the problem—the problem for which Newton
have imagined. We began with Newton’s thought experi- invented calculus—shows that Newton’s laws of motion and
ment about a cannonball fi red around the world, and ended gravitation predict all of Kepler’s empirical laws of planetary
up knowing the mass of the star that our planet orbits. motion—for elliptical as well as circular orbits.
As long as we are at it, we can carry our game one step
further. Kepler’s third law talks about the time for a planet to
complete one orbit about the Sun, known as the period, P, of
a planet’s orbit. The time it takes an object to make one trip Most Orbits in the Real
around a circle is just the circumference of the circle (2πr)
divided by the object’s speed. (Time equals distance divided
World Are Not Circles
by speed.) If the object is a planet in a circular orbit about Some Earth satellites—like Newton’s orbiting cannonball—
the Sun, then its speed must be equal to the circular veloc- move along a circular path at constant speed. Just like the
ity that we calculated. Bringing all this together, we get ball on a string, satellites traveling at the circular velocity
remain the same distance from Earth at all times, neither
Circumference of orbit
____ .
2πr
Period = __________________ = ______ speeding up nor slowing down in orbit. But now let’s change
Circular velocity GM⊙
____
r √ the rules a bit. What if the satellite were in the same place
in its orbit and moving in the same direction, but travel-
1
12
0
The symbol ⊙ signifies the Sun. ing faster than the circular velocity? The pull of Earth is as
(a) (b)
1 A ball slows as it 1 As the planet moves

moves away from away from the Sun, Fgrav
Earth, its motion gravity causes it
opposed by the to slow.
force of gravity…
2 …then picks up 2 It then picks up

speed as it falls speed as it falls back vcirc
back toward Earth. toward the Sun under Fgrav
the influence of the
Sun’s gravity.
Sun
v vcirc
VISUAL ANALOGY FIGURE 3. 25 (a) A ball thrown into the air slows as it climbs away from Earth and then speeds up as it
heads back toward Earth. (b) A planet on an elliptical orbit around the Sun does the same thing. (Although no planet has an
orbit as eccentric as the one shown here, the orbits of comets can be far more eccentric.)
strong as ever, but because the satellite has greater speed, its Newton’s laws do more than explain Kepler’s laws. New-
path is not bent by Earth’s gravity sharply enough to hold it ton’s laws also predict different types of orbits that are
in a circle. So the satellite begins to climb above a circular beyond Kepler’s empirical
orbit. AstroTour: Elliptical Orbits experience. Figure 3.26a Ellipses are
As the distance between Earth and the satellite begins to shows a series of satellite bound orbits.
increase, an interesting thing starts to happen. Think about orbits, each with the same
a ball thrown into the air, as shown in Figure 3.25a. As the point of closest approach to Earth but with different veloci-
ball climbs higher, the pull of Earth’s gravity opposes its ties at that point, as indicated in Figure 3.26b. A look at the
motion, slowing the ball down. The ball climbs more and figure shows that the greater the speed a satellite has at its
more slowly until its vertical motion stops for an instant and closest approach to Earth, the farther the satellite is able to
then is reversed; the ball begins to fall back toward Earth, pull away from Earth, and the more eccentric its orbit
picking up speed along the way. Our satellite does exactly becomes. Yet no matter how eccentric it becomes, as long as
the same thing as the ball. As the satellite climbs above a it remains elliptical, an orbit will eventually bring a satellite
circular orbit and begins to move away from Earth, Earth’s back to the planet that it orbits or bring a planet back to the
gravity opposes the satellite’s outward motion, slowing the Sun. All such orbits are called bound orbits because the sat-
satellite down. The farther the satellite pulls away from ellite is gravitationally bound to the object it is orbiting.
Earth, the more slowly the satellite moves—just as happened You might imagine, though, that somewhere in this
with the ball thrown into the air. And just like the ball, the sequence of faster and faster satellites there comes a point
satellite reaches a maximum height on its curving path, of no return—a point when
then begins falling back toward Earth. Now as the satellite the satellite is moving so fast A satellite moving fast
falls back toward Earth, Earth’s gravity is pulling it along, that gravity is unable to enough will escape a
causing it to pick up more and more speed as it gets closer reverse its outward motion, so planet’s gravity.
and closer to Earth. The satellite’s orbit has changed from the satellite coasts away from
circular to elliptical. Earth, never to return. This indeed is possible. The lowest
What is true for a satellite orbiting Earth in such an speed at which this happens is called the escape velocity
elliptical orbit is also true for any object in an elliptical (see Math Tools 3.3).
orbit, including a planet orbiting the Sun. As we saw earlier, Once an object reaches escape velocity, it is in an
Kepler’s law of equal areas says that a planet moves fastest unbound orbit. That is, the object is no longer gravitation-
when it is closest to the Sun and slowest when it is farthest ally bound to the primary body that it is orbiting, such as
from the Sun. Now we know why. As shown in Figure 3.25b, a satellite orbiting a planet or a comet orbiting the Sun.
planets lose speed as they pull away from the Sun and then When Newton solved his equations of motion, he found that
gain that speed back as they fall inward toward the Sun. unbound orbits are shaped as hyperbolas or parabolas. As
(a) Representative orbits (b) Velocity at closest approach

v/vcirc
v < vesc Bound v = vesc Unbound 0 0.5 1.0 1.5 2.0 2.5 3.0
elliptical orbits parabolic orbit
v = vcirc Bound
Orbits represented in (a)

v = vesc = 2vcirc
Unbound
v > vesc Unbound
hyperbolic orbits v = 2vesc
0 0.5 1.0 1.5 2.0

¢ v/vesc
FIGURE 3. 26 (a) A range of different orbits that share the same point of closest approach but
differ in velocity at that point. (b) Closest-approach velocities for the orbits in (a). An object’s velocity
at closest approach determines the orbit shape and whether the orbit is bound or unbound.
Calculating Escape Velocities

The concept of escape velocity occurs frequently in astronomy. farther from the planet), vesc becomes smaller (in other words, it
For NASA to send a spacecraft to another planet, it must be is easier to escape from the planet’s gravitational pull). We can
launched with a velocity greater than Earth’s escape veloc- now calculate the escape velocity from Earth’s surface, noting
ity. And we will see in Chapter 8 that planets can and do lose that Earth has an average radius (R) of 6.37 × 106 meters and a
their atmospheres when some of their atmospheric atoms are mass (M) of 5.97 × 1024 kg.
bumped up to escape velocity.
________________________
√
If we were to work through the calculation, we would fi nd
_ 2 × (6.67 × 10–11) × (5.97 × 1024)
________________________
that the escape velocity is a factor of √2 , or approximately 1.414 vesc = =
6.37 × 106
times the circular velocity. This relation can be expressed as
____
_ 11,180 m/s = 11.18 km/s = 40,250 km/h.
vesc = √ 2GM
____
R
= √2 vcirc ,
We can also calculate the escape velocity from the surface
where G is the universal gravitational constant, 6.67 × 10 –11 of Ida, an asteroid that you will meet in Chapter 12. Ida has
N m 2/kg2. an average radius (R) of 15,700 meters and a mass (M) of 4.2
Look at this equation for a minute to be sure it makes sense × 1016 kg.
to you. The larger the mass (M) of a planet, the stronger its grav-
_______________________
√
ity, so it stands to reason that a more massive planet would be
2 × (6.67 × 10–11) × (4.2 × 1016)
_______________________
harder to escape from than a less massive planet. Indeed, the vesc = =
1.57 × 104
equation says that the more massive the planet, the greater
the required escape velocity. It also stands to reason that the 19 m/s = 0.019 km/s = 68 km/h.
closer we are to the planet, the harder it will be to escape from
its gravitational attraction. Again, the equation confi rms our So, a fastball thrown from its surface would easily escape
intuition. As the distance R becomes larger (that is, as we get Ida and fly off into interplanetary space.
you can see from Figure 3.26, an object with a velocity less well. We have already seen, for example, how Newton’s laws
than the escape velocity (vesc) will be on an elliptically can be used to measure the mass of Earth and the Sun. A
shaped orbit. Elliptical orbits calculation such as this could never be done with Kepler’s
“close” on themselves. That Unbound orbits are empirical rules. This fact is especially important when we
is, an object traveling in an hyperbolas or remember that Newton’s laws apply to all objects, not just
elliptical orbit will follow the parabolas. the Sun and Earth, and it will prove handy as we continue
same path over and over our journey.
again.13 A hyperbola does not close like an ellipse but instead Astronomers use Newton’s form of Kepler’s third law
keeps opening up, as shown in Figure 3.26a. For example, to calculate the masses of planets, stars, and other bodies.
a comet traveling on a “hyperbolic orbit” makes only a single In doing so, they often rearrange Newton’s form of the law
pass around the Sun and then is back off into deep space, to read
never to return. The third type of orbit is the borderline case 4π2 A3
in which the orbiting object moves at exactly the escape M = ___ × ___2 .
G P
velocity. If it had any less velocity it would be traveling in
a bound elliptical orbit; any more, and it would be moving Everything on the right side of this equation is either a con-
on an unbound hyperbolic orbit. An object moving with a stant (such as 4, π, and G) or a quantity that we can mea-
velocity equal to the escape velocity follows a “parabolic sure (like the semimajor axis A and period P of an orbit).
orbit” and, as in the hyperbolic orbit, passes by the primary The left side of the equation is the mass of the object at the
body only once. As the object traveling in a parabolic orbit focus of the ellipse.
moves away from the primary body, its velocity relative to It is important to note that we cut a couple of corners
the primary body gets closer and closer to, but never quite to get to this point. For one thing, we arrived at this rela-
reaches, zero. An object traveling in a hyperbolic orbit, by tionship by thinking about circular orbits and then simply
contrast, always has excess velocity relative to the primary asserting that it holds for elliptical orbits as well. Another
body, even when it has moved infi nitely far away. corner we cut was assuming that a low-mass object such as a
cannonball is orbiting a more massive object such as Earth.
Earth’s gravity has a strong influence on the cannonball; but
Newton’s Theory Is a Powerful as we have seen, the cannonball’s gravity has little effect on
Earth. For this reason we can imagine that Earth remains
Tool for Measuring Mass motionless while the cannonball follows its elliptical orbit.
As mentioned earlier, Kepler’s empirical laws describe the Similarly, it is a good approximation to say that the Sun
motion of the planets but do not explain them. On the basis remains motionless as the planets orbit about it.
of Kepler’s laws alone, we might imagine that angels carry This picture changes, though, when two objects are
the planets around in their orbits, just as many people closer to having the same mass. In this case both objects
believed during the 16th century! Newton’s derivation of experience significant accelerations in response to their
Kepler’s laws changed all of that. Newton showed that the mutual gravitational attraction. The mass M now refers to
same physical laws that describe the fl ight of a cannonball the sum of the masses of the
on Earth—or the fall of the apocryphal apple on his head— two objects, which are both Objects having similar
also describe the motions of the planets through the heav- orbiting about a common cen- mass orbit each other.
ens. In this way Newton shattered the prevailing concept ter of mass located between
of the heavens and Earth, and at the same time opened up them. We now must think of the two objects as falling
an entirely new way of investigating the universe. Coper- around each other, with each mass moving on its own ellip-
nicus may have dislodged Earth from the center of the uni- tical orbit around their mutual center of mass. So, if we can
verse and started us on the way toward the cosmological measure the size and period of an orbit—any orbit—then
principle, but it was Newton who moved the cosmological we can use this equation to calculate the mass of the orbit-
principle out of the realm of philosophy and into the realm ing objects. This is true not only for the masses of Earth and
of testable scientific theory. And it was through Newton’s the Sun, but also for the masses of other planets, distant
work that astrophysics was born. stars, our galaxy and distant galaxies, and vast clusters of
Not only is Newton’s method more philosophically sat- galaxies. In fact, it turns out that almost all of our knowl-
isfying than simple empiricism—it is far more powerful as edge about the masses of astronomical objects comes directly
from the application of this one equation. In a sense, this
1
13
1 single equation even allows us to tackle the question, What
Strictly speaking, this is true only for a single body orbiting a sec-
ond body. If other bodies are present, the ellipse may not quite close,
is the mass of the universe itself?
causing the orientation of the ellipse to swing around slowly in the You may well be wondering why we have taken such a
orbital plane. This effect is referred to as orbital precession. Relativ- long and sometimes strenuous excursion through the work
istic effects can also cause orbital precession, as we will see in the of Galileo, Kepler, and Newton. Sometimes when we are hik-
case of Mercury’s orbit in Chapter 17. ing in the mountains, it is hard to see the summit from the
perspective of the trail. Now that we have arrived at the top time, Newton’s laws of motion and gravitation provided a
of this particular pass, we can look around and appreciate powerful new understanding of why planets and satellites
what we have gained. When Newton carried out his calcu- move as they do. Theory and
lations, he found that his laws of motion and gravitation empirical observation work Newton’s laws
predicted elliptical orbits that agree exactly with Kepler’s together hand in hand, and provided a physical
empirical laws. This is how Newton tested his theory that our understanding of the uni- explanation
the planets obey the same laws of motion as cannonballs verse strides forward. The for Kepler’s
and how he confirmed that his law of gravitation is correct. work of Copernicus, Galileo, empirical results.
Had Newton’s predictions not been borne out by observa- Kepler, and Newton was not
tion, he would have had to throw them out and go back to just a scientific revolution; it was the scientific revolution,
the drawing board! which changed forever not only our view of the universe but
Kepler’s empirical rules for planetary motion pointed also our very notion of what it means “to know.”
the way for Newton and provided the crucial observational We promised we would show you how science works.
test for Newton’s laws of motion and gravitation. At the same Well, this is how it works.
matched by another force that is equal in magnitude but

Seeing the Forest for the Trees opposite in direction. Gravity is a force between any two
masses, proportional to the product of the two masses
The story of planetary motions is also the story of how and inversely proportional to the square of the distance
we as a species learned to do science. No one taught between them. Putting all of this together, we fi nd that
Copernicus or Galileo or Kepler or Newton about the objects “fall around” the Sun and Earth on elliptical,
differences between empiricism and theory, or about parabolic, or hyperbolic paths. Orbits are ultimately
how the scientific method could be used to test their given their shape by the gravitational attraction of the
ideas. They had to figure these lessons out on their own objects involved, which in turn is a reflection of the
as they went along. Yet their accomplishments remain masses of these objects. We now understand that it is
near the top of the all-time intellectual feats of human- gravity that holds the universe together, giving planets,
kind, and the trail they blazed pointed the way for all stars, and galaxies their very shapes, as well as con-
that was to come. Kepler built on Copernicus’s revolu- trolling their motions through space. Using Newton’s
tionary ideas about planetary motions and uncovered theoretical insight, we look backward along this chain,
three empirical rules that showed the orbits of all the turning observations of the motions of objects through-
planets to be reflections of the same underlying patterns. out the universe into measurements of the masses of
Galileo offered powerful insights into the nature of mat- objects that no human has ever or, in most cases, will
ter and gravity. Newton combined the two sets of ideas: ever visit.
he built Galileo’s insights into a powerful theoretical Thus we arrive at the next stage of our journey. If the
edifice describing the motions of all objects, and then ancients could have stepped off Earth and touched the
he tested this edifice against Kepler’s empirical reality. planets, they would never have fallen into the conceptual
The culmination of this intellectual campaign was far errors that muddled our thinking for millennia—but they
greater than the sum of its parts. The walls separating had no such luxury. Today we have sent robotic surro-
the heavens and Earth came down once and for all, and gates to all of the classical planets in the Solar System,
the science of astrophysics was born. and humans have walked on the surface of the Moon.
The model for science that emerged during this era Even so, we have made only the most cursory visits to
remains the template for how science is done to this our immediate neighborhood. Even the nearest stars
day. Careful empiricism uncovers patterns in nature in remain thousands of times more distant than the most
need of explanation. Scientific theory seeks to discover far-flung of our robotic planetary explorers. For the most
the fundamental truths underlying all things. And in part we, like the ancients, are left with nothing on which
the meeting of the two—the test of theory against the to base our knowledge of the universe but the signals
unforgiving and stalwart challenge of empirical fact— reaching us from across space. Far and away the most
new knowledge and understanding emerge. common of these signals is electromagnetic radiation,
Newton’s work became the cornerstone of what is which includes light (such as the light by which you are
often referred to as classical mechanics. All objects reading this book). Our ability to interpret these signals
have inertia. An object will continue to move in a depends on what we know about light. What is it? How
straight line at a constant speed unless an unbalanced does it originate? How does it interact with matter? What
force acts to change its motion. Mass is the property changes does it experience during its journey? In Chapter
of matter that resists changes in motion. Every force is 4 we turn to these questions.
Summary a. At what point in its orbit is the planet moving the

fastest?
b. At what point is it moving the slowest?
• Gravity, as one of the fundamental forces of nature, binds
7. The distance that Neptune has to travel in its orbit
the universe together.
around the Sun is approximately 30 times greater than
• On the basis of Tycho’s observations, Kepler devel- the distance that Earth must travel. Yet it takes nearly
oped empirical solutions to describe the motions of the 165 years for Neptune to complete one trip around the
planets. Sun. Explain why.
• Newton’s three laws govern the motion of all objects. *8. Kepler’s and Newton’s laws all tell us something about
the motion of the planets, but there is a fundamental dif-
• Unbalanced forces cause changes in motion.
ference between them. What is the difference?
• Proportionality describes patterns and relationships
9. Aristotle taught that the natural state of all objects is to
in nature.
be at rest. Even though this idea seems consistent with
• Mass is the property of matter that gives it resistance to what we observe around us, explain why Aristotle’s rea-
changes in motion. soning was wrong.
• Gravity is a force between any two objects due to their *10. Galileo came up with the concept of inertia. What do
masses. we mean by inertia?
• All objects near Earth’s surface fall with the same accel- 11. What is the difference between speed and acceler-
eration, written as g. ation?
• Planets orbit the Sun in bound, elliptical orbits. 12. Imagine a planet moving in a perfectly circular orbit
around the Sun. Because the orbit is circular, the planet
• Unbound orbits are shaped as parabolas or hyperbolas.
is moving at a constant speed. Is this planet experienc-
ing acceleration? Explain your answer.
13. When riding in a car, we can sense changes in speed
SmartWork, Norton’s online homework system, or direction through the forces that the car applies on
includes algorithmically generated versions of these us. Do we wear seat belts in cars and airplanes to pro-
questions, plus additional conceptual exercises. If tect us from speed or from acceleration? Explain your
your instructor assigns questions in SmartWork, log in at answer.
smartwork.wwnorton.com.
14. An astronaut standing on Earth could easily lift a wrench
having a mass of 1 kg, but not a scientific instrument
with a mass of 100 kg. In the International Space Station
she is quite capable of manipulating both, although the
Student Questions scientific instrument responds much more slowly than
the wrench. Explain why.
*15. In 1920, a New York Times editor refused to publish an
1. Copernicus and Kepler engaged in what is called empiri- article based on rocket pioneer Robert Goddard’s paper
cal science. What do we mean by empirical? that predicted spaceflight, saying that “rockets could not
work in outer space because they have nothing to push
*2. The orbits of the planets around the Sun and the orbits of
against” (that statement the Times did not retract until
satellites around these planets are always ellipses rather
July 20, 1969, the date of the Apollo 11 Moon landing).
than perfect circles. Why?
You, of course, know better. What was wrong with the
3. Each ellipse has two foci. The orbits of the planets have editor’s logic?
the Sun at one focus. What is at the other focus?
16. Explain the difference between weight and mass.
4. Ellipses contain two axes, major and minor. Half the
17. Weight on Earth is proportional to mass. On the Moon,
major axis is called the semimajor axis. What is espe-
weight is also proportional to mass, but the constant of
cially important about the semimajor axis of a planetary
proportionality is different on the Moon than it is on
orbit?
Earth. Why?
5. What is the eccentricity of a circular orbit?
*18. Picture a swinging pendulum. During a single swing,
6. The speed of a planet in its orbit varies in its journey the bob of the pendulum fi rst falls toward Earth and
around the Sun. then moves away until the gravitational force between
it and Earth fi nally stops its motion. Would a pendulum b. Relative to the other driver’s frame of reference, how
swing if it were in orbit? Explain your reasoning. fast are you approaching the other driver’s car?
19. Had Kepler lived on one of a group of planets orbit- 30. You are riding along on your bicycle at 20 km/h and eat-
ing a star three times as massive as our Sun, would he ing an apple. You pass a bystander.
have deduced the same empirical laws? Explain your a. How fast is the apple moving in your frame of
answer. reference?
b. How fast is the apple moving in the bystander’s frame
20. When sending a spacecraft to Mars, we begin by launch-
of reference?
ing it into an Earth orbit with circular velocity.
c. Whose perspective is more valid?
a. Describe the shape of this orbit.
b. What velocity must we give the spacecraft to send it 31. The average distance of Uranus from the Sun is about
on its way to Mars? 19 times Earth’s distance from the Sun. How much
stronger is the Sun’s gravitational grip on Earth than on
21. Describe the difference between a bound orbit and an
Uranus?
unbound orbit.
32. During the latter half of the 19th century, a few astrono-
22. Two comets are leaving the vicinity of the Sun, one trav-
mers thought there might be a planet circling the Sun
eling in an elliptical orbit and the other in a hyperbolic
inside Mercury’s orbit. They even gave it a name: Vul-
orbit. What can you say about the future of these two
can. We now know that Vulcan does not exist. If a planet
comets? Would you expect either of them to eventually
with an orbit a fourth the size of Mercury’s actually
return?
existed, what would be its orbital period relative to that
23. Suppose astronomers discovered a comet approaching of Mercury?
the Sun in a hyperbolic orbit. What would that say about
33. Earth speeds along at 29.8 km/s in its orbit. Neptune’s
the origin of the comet?
nearly circular orbit has a radius of 4.5 × 109 km, and
*24. What is the advantage of launching satellites from space the planet takes 164.8 years to make one trip around the
ports located near the equator? Why are satellites never Sun. Calculate the speed at which Neptune plods along
launched in a westerly direction? in its orbit.
**25. We speak of Earth and the other planets all orbiting 34. Assume that a planet just like Earth is orbiting the bright
about the Sun. Under what circumstances do we have star Vega at a distance of 1 AU. The mass of Vega is twice
to consider bodies orbiting about a “common center of that of the Sun.
mass”? a. How fast is the Earth-like planet traveling in its orbit
around Vega?
b. How long in Earth years will it take to complete one
orbit around Vega?
A P P LY I N G T H E C O N C E P T S 35. Venus’s circular velocity is 35.03 km/s, and its orbital
radius is 1.082 × 108 km. Calculate the mass of the Sun.
26. A sports car accelerates from a dead stop to 100 km/h
in 4 seconds. 36. At the surface of Earth, the escape velocity is 11.2 km/s.
a. What is its acceleration? What would be the escape velocity at the surface of a
b. If it goes from 100 km/h to a dead stop in 5 seconds, very small asteroid having a radius 10 –4 that of Earth’s
what is its acceleration? and a mass 10 –12 that of Earth’s? If you were standing on
the asteroid and threw a baseball with a strong pitch,
27. A train pulls out of a station accelerating at 0.1 m/s2.
what would happen to it?
What is its speed, in kilometers per hour, 21⁄2 minutes
after leaving the station? 37. How long does it take Newton’s cannonball, moving at
7.9 km/s just above Earth’s surface, to complete one orbit
28. Flybynite Airlines takes 3 hours to fly from Baltimore
around Earth?
to Denver at a speed of 800 km/h. Wanting to save the
cost of fuel, management orders its pilots to reduce their 38. Earth’s mean radius and mass are 6,371 km and 5.97 ×
speed to 600 km/h. How long will it now take passengers 1024 kg, respectively. Show that the acceleration of grav-
on this route to reach their destination? ity at the surface of Earth is 9.80 m/s2.
29. You are driving down a straight road at a speed of 90 39. Two balls, one of gold with a mass of 100 kg and one of
km/h and see another car approaching you at a speed wood with a mass of 1 kg, are suspended 1 meter apart.
of 110 km/h along the road. What is the attractive force, in newtons, of
a. Relative to your own frame of reference, how fast is a. the gold ball acting on the wooden ball?
the other car approaching you? b. the wooden ball acting on the gold ball?
*40. Using 6,371 km for Earth’s radius, compare the gravita- 46. Suppose a new dwarf planet is discovered orbiting the
tional force acting on NASA’s space shuttle when sitting Sun with a semimajor axis of 50 AU. What would be the
on its launchpad with the gravitational force acting on orbital period of this new dwarf planet?
it when orbiting 350 km above Earth’s surface.
*47. Planet Zork takes 7 Earth years to orbit the bright star
41. The International Space Station travels on a nearly cir- Achernar at a distance of 7 AU. In terms of solar masses,
cular orbit 350 km above Earth’s surface. What is its how massive is Achernar?
orbital speed?
48. The asteroid Ida (mass = 4.2 × 1016 kg) is attended by a
42. At some time in the future, scientists might want to tiny asteroidal moon, Dactyl, which orbits Ida at an aver-
launch an interplanetary spacecraft from an orbiting age distance of 108 km. Neglecting the mass of the tiny
platform traveling at a speed of 7.61 km/s 500 km above moon, what is Dactyl’s orbital period in hours?
Earth. What is the minimum speed at which the space-
49. A pair of asteroids, each with a mass of 1016 kg, orbit about
craft would have to leave the platform?
a common center of gravity in 20 hours. What is the
43. The radius and mass of the Sun are 696,000 km and 1.99 separation of these asteroids, in kilometers?
× 1030 kg, respectively. A comet on a parabolic orbit falls
50. The two stars in a binary star system (two stars orbiting
into the Sun. Just before it vaporizes at the Sun’s surface,
about a common center of mass) are separated by 109 km
how fast is it traveling?
and revolve around their common center of mass in 10
44. (a) What does an acceleration of 9.8 m/s2 mean? (b) If you years.
jumped from a stationary balloon, after 1 second you a. What is their combined mass, in kilograms?
would be falling at a speed of 9.8 m/s. After 2 seconds b. The mass of the Sun is 1.99 × 1030 kg. What is their
your speed would be 2 × 9.8 = 19.6 m/s, or slightly more combined mass, in solar masses?
than 70 km/h! In the absence of any air resistance, how
fast would you be falling after 20 seconds?
45. Weight refers to the force of gravity acting on a mass. vides a Study Plan for each chapter of 21st Century
We often calculate the weight of an object by multiply- Astronomy. Study Plans include animations, read-
ing its mass by the local acceleration due to gravity. The ing outlines, vocabulary flashcards, and multiple-choice
value of gravitational acceleration on the surface of Mars quizzes, plus links to premium content in SmartWork and
is 0.39 times that on Earth. If your mass is 85 kg, your the ebook. Visit www.wwnorton.com/studyspace.
weight on Earth is 830 N (m × g = 85 kg × 9.8 m/s2 = 830
N). What would be your mass and weight on Mars?
Light brings us the news of the Universe.
WILLIAM HENRY BRAGG (1862–1942)
The setting Sun colors clouds over the Australian bush.

CHAP TER 4
Light
4.1 Let There Be Light K E Y CONCEP T S

Light is a fundamental part of our experience of the world Unlike the physicist or the chemist, who has control
because it is through light that we most directly perceive over the conditions in a laboratory, the astronomer
the world beyond our physical grasp. (Those who have lost must try to glean the secrets of the universe from the
their sight face a challenge greater than most of us can light and other particles that reach us from distant
imagine.) Throughout our language and culture, light is a objects. On this leg of our journey we turn our attention
metaphor for knowledge and information. To understand to light, a most informative messenger, and find that
something is to “see it.” A “bright idea” is symbolized by
• Light is not only a messenger but also a means by
a lightbulb going off over our heads. As we leave our igno-
rance behind, we become “enlightened.” A symbol of hope which energy is transported throughout the universe.
is a “light at the end of the tunnel.” At the same time, light • Light is an electromagnetic wave with a spectrum
plays other, even more essential roles in our lives. Light extending far beyond the colors of the rainbow.
from the Sun warms Earth, drives the wind and rain, and
• Light is also a stream of particles called photons.
powers photosynthesis in plants, which lie at the bottom of
the terrestrial food chain. The energy you obtain from con- • Reconciling the two identities of light and matter—its
suming plants—either directly or indirectly—and expend wave and particle characteristics—challenges our
as you move through your day arrived on the planet in the everyday ideas about what is “real.”
form of light. • The wave-particle nature of light and matter gives
The role of light in astronomy closely parallels the different types of atoms unique spectral “fingerprints”
role of light in our everyday existence. Our knowledge of
that we can use to measure the composition and
the universe beyond Earth comes overwhelmingly from
properties of distant objects.
light given off or reflected by astronomical objects. For-
tunately, light is a very informative messenger. For any • Temperature measures the thermal energy of an object
given object, light carries information about that object’s and determines the amount and spectrum of light that
temperature, composition, speed, and even the nature of a dense object emits.
the material that the light passed through on its way to
Earth. Yet light plays a far larger role in astronomy than
just being a messenger. Light is one of the main ways in
which energy is transported throughout the universe.
Light carries energy generated in the heart of a star out-
ward through the star and off into space. Objects as
diverse as stars, planets, and vast clouds of gas and dust
91
92 Chapter 4 Light
fi lling interstellar space are heated as they absorb light the lost time and once again passed behind Jupiter at the
and cooled as they emit light. predicted times.
Although light allows us to see the world, we cannot It is often the case in science that a difference between
actually “see” light as an object. When we view a tree or the theoretical predictions and experimental results points the
stars at night, our eyes and brains are responding to light way to new knowledge, and
that is reflected from or emitted by these objects. Light is the Rømer’s work was no excep- Rømer used Jupiter’s
messenger but is itself invisible. It is hard to study something tion. Rømer correctly sur- moons to measure
that cannot be seen, so an understanding of light was a long mised that rather than a failure the speed of light.
time coming. The property of light that is easiest to try to of Kepler’s laws, he was seeing
measure is the speed at which it travels. As we continue our the first clear evidence that light travels at a finite speed. As
journey, we will find that light sets the standard for what shown in Figure 4.1, the moons appeared “late” when Earth
we mean by when, where, or how fast because nothing can was farther from Jupiter because of the time needed for light
travel faster than light. to travel the extra distance between the two planets. Over
the course of Earth’s yearly trip around the Sun, the distance
between Earth and Jupiter changes by 2 astronomical units
(AU), which is about 3 × 1011 meters. The speed of light equals
4.2 How Our Understanding this distance divided by Rømer’s 16.7-minute delay, or about
3 × 108 meters per second (m/s). The value that Rømer actu-
of Light Evolved ally announced in 1676 was a bit on the low side—2.25 ×
108 m/s—because the length of 1 AU was not well known.
Suppose that you are a scientist living in the 17th century. But Rømer’s result was more than adequate to make the point
How might you go about measuring the speed of light? One that the speed of light is very great indeed. The International
way would be to have a friend stand on a hilltop far away. Space Station moves around Earth at a dazzling speed of
You uncover a lantern, and the instant your friend sees the about 28,000 kilometers per hour (km/h), almost 8,000 m/s.
light from your lantern, your friend uncovers her own lan- Light travels almost 40,000 times faster than this. It could
tern. The time it takes from when you uncover your lantern circle Earth in only 1⁄7 of a second. No wonder Galileo’s
to when you see your friend’s light will be the light’s round- attempts to measure the speed of light failed!
trip travel time—plus, of course, your friend’s reaction time. A good deal of work has been done to improve on Rømer’s
Galileo tried this experiment but could not measure any original result. Modern measurements of the speed of light
delay. He concluded that the speed of light must be very made with the benefit of high-speed electronics and lasers
great indeed, possibly even infinite. give a value of 2.99792458 × 108 m/s in a vacuum. As of
If light travels so rapidly, then to measure its speed we October 1983, the length of a meter is now defined as the
will need either very large distances over which to measure distance traveled by light in a vacuum in 1/299,792,458 of
its flight or very good clocks. Galileo had neither at his dis- a second.
posal, but by the end of the 17th century astronomers had The speed of light in a vacuum, about 300,000 kilome-
both. The great distances were the distances between the ters per second (km/s), is one of nature’s fundamental con-
planets, whereas the good “clock” was provided by Kepler stants, usually written as c.
and Newton. According to Newton’s derivation of Kepler’s Keep in mind, however, that The speed of light
laws, orbital periods are constant, with each orbit taking this is true only in a vacuum. is 300,000 km/s
exactly as much time as the orbit before. This property The speed of light through in a vacuum.
applies to moons orbiting planets just as it applies to plan- any medium, such as air or
ets orbiting the Sun. glass, is always less than c. We refer to the ratio of light’s
In the 1670s, Danish astronomer Ole Rømer (1644–1710) speed in a vacuum to its speed, v, in a medium as the medi-
was studying the moons of Jupiter, measuring the times um’s index of refraction, n:
when each moon disappeared behind the planet. Much to
c
his amazement Rømer found that rather than maintain- n = __
v.
ing a regular schedule, the observed times of these events
would slowly drift in comparison with predictions. Some- For typical glass, n is approximately 1.5. Rearranging this
times the moons disappeared behind Jupiter too soon, equation, we fi nd that the speed of light in glass is
and at other times they went behind Jupiter later than
c 300,000 km/s
expected. Rømer realized that the difference depended v = __
n =
___________ = 200,000 km/s.
on where Earth was in its orbit. If he began tracking the 1.5
moons when Earth was closest to Jupiter, then by the We will come back to the index of refraction in Chapter 5
time Earth was farthest from Jupiter, the moons were a when we discuss refraction and refracting telescopes.
bit over 161⁄2 minutes “late.” But if he waited until Earth Recall that in Chapter 1 we spoke of distances expressed
was once again closest to Jupiter, the moons “made up” not in kilometers or miles, but in units of time. For example,
1 When a moon of Jupiter

disappears behind the planet…
2 …the light must travel 4.2 AU

3 …but an additional 2 AU
to reach Earth here…
to reach Earth when it is
Light from
Jupiter’s here. The event is seen 16
Jupiter’s moon
moon minutes 40 seconds “late.”
Earth
Sun
Jupiter
4.2 AU 2.0 AU
2 AU 4 The observed delay allowed Rømer

Speed of light = = 300,000 km/s
16m 40s to measure the speed of light.
FIGURE 4 .1 Danish astronomer Ole Rømer realized that apparent differences between the
predicted and observed orbital motions of Jupiter’s moons depend on the distance between Earth
and Jupiter. He used these observations to measure the speed of light. (The superscript letters in
the expression “16m40s” stand for minutes and seconds of time, respectively.)
the Moon’s distance is such that it takes light 11⁄4 seconds same physical phenomenon. An electric force is the push
to travel between Earth and Moon. In other words, we can or pull between electrically charged particles. Opposite
say that the Moon is 11⁄4 light-seconds from Earth. The Sun charges attract, and like charges repel. A magnetic force,
is 81⁄3 light-minutes away, and on the other hand, is a force between electrically charged
the next-nearest star is 41⁄3 A light-year is the particles arising from their motion.
light-years distant. The travel distance that light To describe the electric and magnetic forces, Maxwell
time of light is a convenient travels in one year. introduced the concepts of the electric field and the mag-
way of expressing cosmic dis- netic field. A charged particle creates an electric field that
tances, and the basic unit is the light-year. (Astronomers points away from the charge if the charge is positive, as
frequently use another yardstick to describe stellar and shown in Figure 4.2a, or toward it if the charge is nega-
galactic distances, the parsec, abbreviated pc. One parsec tive. Because the electric field points directly away from
is equal to 3.26 light-years.) A light-year is defi ned as the a positively charged particle (or directly toward a nega-
distance traveled by light in 1 year, or about 9.5 trillion tively charged particle), the force that a second charge feels
kilometers (km). Remember, a light-year is a measure of is either directly toward or directly away from the fi rst
distance. It is not a measure of time. (“I tell you, it’s been charged particle.
simply light-years since we last saw them” is a frequent The picture gets a bit more interesting if we quickly move
misuse of the term light-year.) the fi rst charged particle (q1) by some amount, as shown in
Figure 4.2b. We might expect the force on the second posi-
tively charged particle (q2) to change immediately so that it
points away from the new position of the fi rst particle. Yet
Light Is an Electromagnetic Wave experiments show that it does not. Immediately after the
Since the earliest investigations of light, there has been a fi rst charge moves, there is no change in the force felt by the
good deal of controversy over the question of whether light second charge. Only later does the second particle feel the
is composed of particles, as change in location of the fi rst (Figure 4.2c). The situation is
Newton believed, or is instead A wave is a something like what happens if you are holding one end of
a wave. (A wave is a distur- disturbance that a long piece of rope and a friend is holding the other end.
bance that travels from one travels away When you yank your end of the rope up and down, your
point to another.) This contro- from a source. friend does not feel the result immediately. Instead your
versy was seemingly put to yank starts a pulse—a wave—that travels along the rope.
rest once and for all in 1873 by the Scottish physicist James Your friend notices the yank only when this wave arrives
Clerk Maxwell (1831–1879). One of Maxwell’s many accom- at his end. Similarly, when you move a charged particle,
plishments was his introduction of the concept that elec- information about the change travels outward through space
tricity and magnetism are actually two components of the as a wave in the electric field. Other charged particles are
94 Chapter 4 Light
1 A charged particle 2 …that acts in a direct 3 If q1 is moved up, the change 4 The force on q2 changes
(q1) gives rise to an line on another charged in the electric field moves direction only when the change
electric field… particle (q2). outward at the speed of light. in the electric field reaches q2.
+ + q1 +q q1 + +q
q2 2 2
+q
1
Electric
field
(a) (b) (c)
FIGURE 4 . 2 Maxwell’s theory of electromagnetic radiation describes how electrically charged particles
move and interact. (a) A positively charged particle, q1, has an electric field (shown in orange) that acts
outward along the field lines. (b) When q1 accelerates, changes in the electric field move outward at the
speed of light. For simplicity, the charge is shown moving instantly from one place to another, but in reality
this could not happen. (c) Particle q2 does not respond until it feels the change in q1.
not affected by the fi rst particle’s movement until the wave wave (Figure 4.3). These electromagnetic waves, and the
reaches them. accelerating charges that generate them, are the sources of
Maxwell summarized the behavior of electric and mag- electromagnetic radiation. By contrast, an electric charge
netic fields in four elegant equations. Among other things, that is moving at a constant velocity is stationary in its iner-
these equations say that a tial frame of reference and so does not radiate.
changing electric field causes Changing electric and In addition to predicting that electromagnetic waves
a magnetic field, and that a magnetic fields lead to should exist, Maxwell’s equations predict how rapidly
changing magnetic field a self-sustaining the disturbance in the electric and magnetic fields should
causes an electric field. These electromagnetic wave. move. In short, Maxwell’s equations predict the speed at
changes “feed” on themselves. which an electromagnetic wave should travel. When Max-
A change in the motion of a charged particle causes a chang- well carried out this calculation, he discovered that elec-
ing electric field, which causes a changing magnetic field, tromagnetic waves should travel at 3 × 108 m/s—which
which causes a changing electric field, and so on. Once the is the speed of light! This agreement could not be simple
process starts, a self-sustaining procession of oscillating coincidence. Maxwell had shown that light is an electro-
electric and magnetic fields moves out in all directions magnetic wave.
through space. Instead of a purely electric wave, an accel- Maxwell’s wave description of light also gives us an
erating charged particle gives rise to an electromagnetic idea of how light originates and how it interacts with mat-
1 A varying electric field causes a varying

magnetic field, which in turn causes
a varying electric field, and so on.
FIGURE 4 . 3 Far from its
2 The result is a
self-sustaining source, an electromagnetic
Electric electromagnetic wave consists of oscillating
field wave. electric and magnetic fields that
are perpendicular both to each
other and to the direction in
Direction which the wave travels.
of wave
Magnetic propagation
field
c
(a) A bobbing cork makes waves (a)

that travel out across the water.
(b)
1 An oscillating electric
charge produces
electromagnetic waves… A passing ripple causes a
cork to bob up and down.
2 …that travel outward
through space at the
speed of light.
(b)
Electric-field Charged
oscillations particles
Electromagnetic
waves
Speed
of light
Arrows indicate the direction
of the wave’s electric field. The alternating electric field
of a passing electromagnetic
wave causes a charged
VISUAL ANALOGY FIGURE 4 . 4 (a) A fish pulling downward particle to oscillate.
on a hook connected to a fishing cork generates waves that
move outward across the water’s surface. (b) In similar VISUAL ANALOGY FIGURE 4 . 5 (a) When waves moving across
fashion, an oscillating (accelerated) electric charge generates the surface of water from one bobbing cork reach a nearby cork,
electromagnetic waves that move away at the speed of light. they cause the second cork to bob up and down. (b) Similarly,
a passing electromagnetic wave causes an electric charge to
wiggle in response to the wave.
ter. Imagine a fishing cork floating on a lake on a perfectly

calm day. The surface of the lake is as smooth and fl at as
a mirror until a fi sh tugs on
the hook and line dangling Accelerating Now imagine that a second cork is afloat on the lake
beneath the cork. The motion charges cause some distance from the fi rst, as in Figure 4.5a. The second
of the cork causes a distur- electromagnetic cork remains stationary until the ripple from the fi rst cork
bance that moves outward as waves. reaches it. As the ripple passes by, the rising and falling of
a ripple on the surface of the the water causes the second cork to rise and fall as well.
lake (Figure 4.4a). In much the same way, an oscillating Similarly, the oscillating electric field of an electromagnetic
(and therefore accelerating) electric charge (Figure 4.4b) wave causes an oscillating force on any charged particle
causes a disturbance that moves outward through space that the wave encounters, and this force causes the particle
as an electromagnetic wave. Note, however, that the rip- to move about as well (Figure 4.5b). It takes energy to pro-
ples on the lake result from mechanical distortions of the duce an electromagnetic wave, and that energy is carried
surface of the lake. Mechanical waves such as these through space by the wave. Matter far from the source of
involve distortions that are measured as distances and the wave can absorb this energy. In this way, some of the
media that have mass. Maxwell showed that light waves energy lost by the particles generating the electromagnetic
are a fundamentally different type of wave. Light waves wave is transferred to other charged particles. The emis-
result not from mechanical distortion of a medium, but sion and absorption of light by matter are the result of the
from periodic changes in the strength of the electric and interaction of electric and magnetic fields with electrically
magnetic fields. charged particles.
96 Chapter 4 Light
FIGURE 4 .6 A wave is
Longer wavelength (λ)
Longer period ( P) characterized by the distance
Wavelength (λ) Lower frequency (f ) over which the wave repeats
λ itself (wavelength, λ), the
Amplitude maximum excursion from its
undisturbed state (amplitude),
v and the speed (v) at which
the wave pattern travels. In
an electromagnetic wave, the
Shorter wavelength (λ)
Shorter period (P) amplitude is the maximum
Higher frequency (f ) strength of the electric field,
v
λ and the speed of light is written
as c.
has a frequency of 100 Hz. More generally, the frequency of

Waves Are Characterized a wave is just 1 divided by its period:
by Wavelength, Speed, 1 1
Frequency = ______ or f = __ .
Frequency, and Amplitude Period P
Along our journey we will encounter waves of different There is also a relationship between the period of a
kinds, ranging from electromagnetic waves crossing the vast wave and its wavelength. The period of a wave is the time
expanse of the universe to seismic waves traveling through between the arrival of one wave crest and the next. During
Earth. We generally characterize waves by four quantities— this time the wave travels a distance equal to the separation
amplitude, speed, frequency, and wavelength—as illustrated between the two wave crests, or one wavelength. So far, so
in Figure 4.6. The amplitude of a wave is the maximum good. Now add to the picture the fact that distance traveled
deviation from its undisturbed or relaxed position. (In the equals speed multiplied by time taken. Change “distance
case of light waves, the amplitude is an indication of the traveled” to one wavelength and change “time taken” to one
intensity or brightness of the radiation.) Waves travel at a period, and we fi nd that the wavelength of a wave equals
particular speed, which is usually written as v. (In the case the speed at which the wave is traveling multiplied by the
of light waves, we have already shown that the speed in a period of the wave:
vacuum is 300,000 km/s and is written as c.) The number of
wave crests passing a point in space each second is called Wavelength = Speed × Period.
the wave’s frequency, denoted by f. The unit of frequency
is cycles per second, which is generally referred to as hertz Physicists use the letter c to represent the speed of light, so
(abbreviated Hz) after the 19th century physicist Heinrich we can say that
Hertz (1857–1894), who was the fi rst to experimentally con-
firm Maxwell’s predictions about electromagnetic radiation. λ = c × P.
The time taken for one complete cycle is called the period,
P, which is measured in seconds. Using the relationship between period and frequency just
The distance that a wave travels during one complete given, we can also write
oscillation is called the wavelength. This is just the distance
Speed c
from one wave crest to the next, or the distance from one Wavelength = _________ or λ = __ .
wave trough to the next. The wavelength is usually denoted Frequency f
by the Greek letter λ (pronounced “lambda”). There is a So if we know the speed of a wave, then knowing one of the
clear relationship between the frequency of a wave and its three properties—its wavelength, period, or frequency—
period. If the period of a wave is 1⁄2 second—that is, if it takes tells us the other two (see Math Tools 4.1.)
1
⁄2 second for one wave to pass by, crest to crest—then two Look at this relationship more closely. The longer the
waves will go by in 1 second. So a wave with a period of 1⁄2 length of a wave, the longer you have to wait between wave
second per cycle has a frequency of 2 cycles per second (or crests, so the frequency of the wave will be lower. A shorter
2 Hz). Similarly, if a wave has a period of 1/100 second per wavelength means less distance between wave crests, which
cycle, then 100 waves will pass by each second. This wave means a shorter wait until the next wave comes along.
Working with Electromagnetic Radiation

Most of us listen to radios, both AM and FM. When you tune in This FM wavelength is about 10 feet. As you can see, FM wave-
to a station at, say, “770 on your AM dial,” you are receiving a lengths are much shorter than AM wavelengths.
signal that is broadcast at a frequency of 770 kilohertz (kHz), or The human eye is most sensitive to light in the green to yel-
7.7 × 105 Hz. We can use the relationship between wavelength low part of the spectrum. This light has a wavelength of about
and frequency to calculate the wavelength of the AM signal: 500–550 nanometers (nm). Green light with a wavelength of
520 nm has a frequency of
c
(
3 × 108 m/s
λ = _ = _________ )
= 390 m.
f 7.7 × 105/s
This AM wavelength is about four times the length of a foot-
c
(3.00 × 108 m/s
f = __ = ___________
λ –9
5.20 × 10 /m
)5.80 × 1014
= ________.
s
ball field! or
We can compare this wavelength with that of a typical FM
broadcast signal, “99.5 on your FM dial,” or 99.5 megahertz 5.80 × 1014 Hz.
(MHz), or 9.95 × 107 Hz:
This frequency corresponds to 580 trillion wave crests pass-
c
(
3 × 108 m/s
λ = _ = _________
f 9.95 × 107/s
)
= 3 m. ing by each second!
Therefore, a shorter wavelength means a higher frequency. μ is the Greek letter mu). Another commonly used unit is
A tremendous amount of information can be carried by the nanometer, abbreviated nm. A nanometer is one-billionth
waves—intelligible speech, (10 –9) of a meter.1 The wavelengths of the light we perceive
for example, or complex and A long wavelength as red fall between about 600 and 700 nm. The shortest-
beautiful music. As we con- means low frequency, wavelength violet light that our eyes can see has a wave-
tinue our study of the uni- and a short length of about 350 nm. Stretched out between the two,
verse, time and again we will wavelength means literally in a rainbow, is the rest of the visible spectrum (see
fi nd that the information we high frequency. Figure 4.7.) The colors in the visible spectrum, in order of
receive, whether about the decreasing wavelength, can be remembered as a name: “Roy
interior of Earth or about a distant star or galaxy, rides in G. Biv,” which stands for
on a wave.
Red Orange Yellow Green Blue Indigo Violet.
A Wide Range of Wavelengths Makes When we say visible light, what we mean is “the light
that the light-sensitive cells in our eyes respond to.” But
Up the Electromagnetic Spectrum this is not the whole range of possible wavelengths for elec-
You have almost certainly seen a rainbow like the one in tromagnetic radiation. Radia-
Figure 4.7 spread out across the sky, or sunlight split into tion can have wavelengths Visible light is only one
many different colors by a prism. This sorting of light by that are much shorter or much small segment of the
colors is really a sorting by wavelength. When we talk about longer than our eyes can per- electromagnetic
light spread out according to wavelength, we refer to the ceive. The whole range of dif- spectrum.
spectrum of the light. On the long-wavelength (and there- ferent wavelengths of light is
fore low-frequency) end of the collectively referred to as the electromagnetic spectrum.
visible spectrum is red light. The spectrum Follow along in Figure 4.8 as we take a tour of the electro-
At the other end of the visible of visible light is magnetic spectrum, beginning with visible light and work-
spectrum is violet light, which seen as the colors
is the bluest of blue light. A of the rainbow. 1
Astronomers conventionally use nanometers (nm) when referring
micrometer, or micron, is the to wavelengths at visible and shorter wavelengths, micrometers or
unit often used for measuring the wavelength of electro- “microns” (
m) for wavelengths in the infrared, and millimeters (mm),
magnetic radiation. A micrometer is one-millionth (10 –6) of centimeters (cm), and meters (m) for wavelengths in the microwave
a meter. The abbreviation used for a micrometer is m (where and radio regions of the electromagnetic spectrum.
98 Chapter 4 Light
ing our way next to the part of the spectrum with shorter
wavelengths and then to longer wavelengths. We start at
the blue end of the visible spectrum. Beyond this short-
wavelength, high-frequency side of the visible spectrum,
there is light that is “bluer than blue” or actually “more
violet than violet.” This light, with wavelengths between
40 and 350 nm, is called ultraviolet (UV) radiation. You
can remember what ultraviolet light is just by looking at
the name. The prefi x ultra- means “extreme,” so ultraviolet
light is light that is more “extremely” violet than violet. It is
important to remember that ultraviolet light is fundamen-
tally no different from visible light, any more than high C
on a piano is fundamentally different from middle C.
As we go to still shorter wavelengths of light (and so to
higher frequencies), we pass through the ultraviolet part of
the spectrum. At a wavelength shorter than 40 nm, or 4 × 10–8
meter, we stop calling radiation ultraviolet light and instead
start calling it X-rays. This distinction arose for historical rea-
sons. When X-rays were discovered in the late 19th century,
they were given the name “X” by their discoverer, German
physicist Wilhelm Conrad Roentgen (1845–1923), to indicate
they were “a new kind of ray.” As we continue to even shorter
wavelengths, we come to another somewhat arbitrary break.
Electromagnetic radiation with the very shortest wavelengths
(less than about 10 –10 meter) is referred to as gamma radia-
tion, or more commonly as gamma rays. Again, the reasons
are historical. Gamma rays (or γ-rays) were first discovered
as a type of radiation given off by radioactive material. Only
later did their true nature become known.
So far, we have been considering ever-shorter wave-
lengths and higher frequencies. In principle, there is no
FIGURE 4 .7 The visible part of the electromagnetic spectrum

is laid out in all its glory in the colors of this rainbow. FIGURE 4 .8 By convention, the electromagnetic spectrum is
broken into loosely defined regions ranging from gamma rays to
radio waves.
Frequency (Hz)
31020 31018 31016 31014 31012 31010 3108 3106
Gamma rays Ultraviolet Microwave Television
X-rays Infrared FM Radio AM
10–12 10–11 10–10 10–9 10–8 10–7 10–6 10–5 10–4 10–3 10–2 10–1 1 101 102 103
Wavelength (m)
0.001 nm 0.01 nm 0.1 nm 1 nm 10 nm 100 nm 1 μm 10 μm 100 μm 1 mm 1 cm 10 cm 1m 10 m 100 m 1 km
Visible
350 400 500 600 700

Wavelength (nm)
limit to this process. We can conceive of gamma rays of (phot- means “light,” as in photograph; and -on signifies a
arbitrarily short wavelengths and arbitrarily high fre- particle). As massless particles, photons always travel at the
quencies (even though practical considerations eventu- speed of light, and they carry energy.
ally come into play). We can also go in the other direction. Recognition of the particle theory of light, however, did
Just as there is light that is more violet than violet, there is not mean that the wave theory had been discarded. The
light that is “redder than red.” Such light, covering wave- particle description of light is
lengths longer than about 700 nm and shorter than 500 μm tied to the wave description The energy of a
(5 × 10 –4 meter), is referred to as infrared (IR) radiation. of light by a relationship photon is proportional
Again, the key to remembering what infrared light is comes between the energy of a pho- to its frequency.
from looking at the word itself. Infra- is a prefi x that means ton and the frequency or
“below.” Infrared light is light that has a frequency lower wavelength of the wave. Specifically, we write
than (below) that of red light. When the wavelength of light
hc
gets still longer than this, we start calling it microwave E = hf or E = ___ .
radiation. The longest-wavelength (and therefore lowest- λ
frequency) electromagnetic radiation, with wavelengths The h in this equation is called Planck’s constant and
longer than a few centimeters and ranging up to arbi- has the value 6.63 × 10 –34 joule-second. (Planck’s constant
trarily long wavelengths, is called radio waves. In Chap- is named after the German physicist Max Planck, 1858–
ter 5 we will discuss the various kinds of telescopes used 1947.) According to the particle description of light, the
by astronomers to capture and analyze the wide range of electromagnetic spectrum is a spectrum of photon ener-
electromagnetic radiation. gies. The higher the frequency of the electromagnetic wave,
the greater the energy carried by each photon. Photons of
shorter wavelength (higher frequency) carry more energy
Light Is a Wave, but It than do photons of longer wavelength (lower frequency). For
example, photons of blue light carry more energy than do
Is Also a Particle photons of longer-wavelength red light. Ultraviolet photons
By showing that light is an electromagnetic wave, Maxwell’s carry more energy than do photons of visible light, and X-ray
work seemed to put to rest the issue of whether light consists photons carry more energy than do ultraviolet photons. The
of waves or particles. Although the electromagnetic wave lowest-energy photons are radio wave photons.
theory of light has successfully described many phenomena, The intensity of light measures the total amount of
there are also many phenomena that it does not describe energy that a beam of the light carries. A beam of red light
well. These range from the presence of sharp bright and can be just as intense as a beam of blue light—that is, it
dark lines, called “spectral lines,” at specific wavelengths can carry just as much energy—but because the energy of a
in the light from some objects, to the shape of the continu- red photon is less than the energy of a blue photon, it takes
ous spectrum of light emitted by a lightbulb. more red photons to reach that intensity than it takes blue
Many of these difficulties with the wave model of light photons. This relationship is a lot like money. A hundred
have to do with the way in which light interacts with atoms dollars is a hundred dollars, but it takes a lot more pennies
and molecules. Scientists working in the late 19th and early (low-energy photons) than quarters (high-energy photons)
20th centuries discovered that many of the puzzling aspects to make up a hundred dollars.
of light could be better understood if light energy came in When physicists speak of the energy of light as broken
discrete packages. In 1905, Albert Einstein published a paper into discrete packets called photons, they say that the light
in which he argued that light consists of particles. He based energy is quantized. The word
his argument on the photoelectric effect, the emission of quantized, which has the same Photons are the
electrons from surfaces when the surfaces are illuminated by root as the word quantity, quantum mechanical
electromagnetic radiation greater than a certain frequency. means that something is sub- description of light.
Einstein showed that the rate at which electrons are ejected divided into discrete units. A
depends only on the intensity of the incident radiation, and photon is referred to as a quantum of light. The branch of
that the electron velocity depends only on the frequency of physics that deals with the quantization of energy and of
the incident radiation.2 Effectively, scientists were modify- other properties of matter is called quantum mechanics.
ing our understanding of light to show that, although it can Although the predictions of quantum mechanics have
sometimes be explained as acting like a wave, light can also been confi rmed over and over again by experiment, its
be explained as acting like a particle. In this model we think fundamental assumptions seem counterintuitive. The
about light as being made up of particles called photons wave-particle description of light confl icts with everyday,
commonsense ideas about the world. It is hard for us to
2
Interestingly, it was Einstein’s work on the photoelectric effect, imagine a single thing sharing the properties of a wave on
not special or general relativity, that earned him the Nobel Prize in the ocean and a beach ball, yet light does just that. You may
1921. find yourself scratching your head in confusion over exactly
100 Chapter 4 Light
what light really is. If you do, consider yourself in good com- changing their structure as you read are made of atoms.
pany. The scientists who developed the seemingly bizarre Atoms are incredibly tiny—so tiny that a single teaspoon
quantum description of nature had a great deal of trouble of water contains about 1023 of
thinking about light as well. The wave model of light is them. (There are as many Virtually all matter
clearly the correct description to use in many instances, just atoms in a single teaspoon of we encounter is
as Maxwell showed. At the same time, the particle descrip- water as there are stars in the composed of atoms.
tion of light is also clearly the correct description to use in observable universe.) When
other cases, as scientists like Planck and Einstein demon- we talk about the interaction of light with matter, what we
strated. But how can the same thing—light—be described are really talking about is the interaction of light with atoms
as both a wave and a particle? and their components.
Light is what light is. The trouble with quantum mechan- Atoms are made up of three types of subatomic particles:
ics lies not with the nature of reality, but with what our protons, neutrons, and electrons (Figure 4.9a). At the center
brains are used to thinking about. The brains of our ances- of the atom is the nucleus, which is composed of positively
tors did not have to deal with things moving at nearly the charged protons and electrically neutral neutrons. An atom
speed of light. Our brains deal well with objects that are may have many protons and neutrons in its nucleus. Sur-
sitting still or even moving as fast as a hard-hit fly ball. rounding the nucleus of the atom are negatively charged
Basically, our brains cope best with things moving at the electrons. For an atom to be electrically neutral, it must
speeds of things in nature that we might want to eat or that have the same number of electrons as protons. Electrons
might want to eat us. have much less mass than protons or neutrons have, so
Our trouble with thinking of light as both a wave and a almost all the mass of an atom is found in its nucleus. This
particle only hints at the puzzling and philosophically description naturally leads to a mental picture of an atom
troublesome world of quan- as a tiny “solar system,” with the massive nucleus sitting
tum mechanics. In the next Like light, all matter in the center and the smaller electrons orbiting about it,
section we will learn that can behave as waves much as planets orbit about the Sun (Figure 4.9b). We refer
light is not the only thing that and particles. to this as the Bohr model, after the Danish physicist Niels
shares wave and particle Bohr (1885–1962), who proposed it in 1913. AstroTour:
properties. In fact, all matter shares wave and particle prop- Atomic Energy Levels and the Bohr Model
erties. Sometimes a “particle” such as an electron behaves Unless you have thought about atoms a great deal, the
as if it were a wave, while at other times a “wave” of light Bohr model probably matches your concept of the structure
clearly exhibits the properties of a discrete particle. Early of an atom. It is much the same picture that scientists in
quantum physicists would sometimes joke that on Monday, the early 20th century held as well. But it has a fatal prob-
Wednesday, and Friday, light and matter were particles, but lem. In this view, an electron whizzing about in an atom
on Tuesday, Thursday, and Saturday, they were waves. (And is constantly undergoing an acceleration—the direction of
on Sunday, it was best just not to think about them at all!) its motion is constantly changing. The wave description of
It’s true that our brains can be bounded by a conceptual electromagnetic radiation says that any electrically charged
box that is defi ned by the experiences and circumstances particle that is accelerating must also be giving off electro-
that we and our ancestors had to cope with. But they don’t magnetic radiation. This electromagnetic radiation, in the
have to remain trapped inside that box. One exciting thing form of photons, should be carrying away the orbital energy
about modern physics and astronomy is that they enable us of the electron. (Imagine that electron as the wiggling elec-
to break down the walls of that conceptual box by giving us tric charge in Figure 4.4b.) If you calculate how much energy
the tools for understanding what lies beyond its boundaries. should be radiated away from the electron, you find that
AstroTour: Light as a Wave, Light as a Photon only a tiny fraction of a second should be needed for the
electrons in an atom to lose all their energy and fall into
the atom’s nucleus! Fortunately for us, this does not hap-
pen. Atoms exist for very long periods of time, and electrons
4.3 The Quantum never “fall into” the nuclei of atoms. So something must be
wrong with this concept of an atom.
View of Matter A way out of this difficulty came when scientists real-
ized that, just as waves of light have particle-like properties,
If we want to understand better how light interacts with so, too, do particles of matter have wavelike properties. With
matter, we need to start by pinning down exactly what we this realization, the “miniature solar system” model of the
mean by matter in the fi rst place. To a physicist, matter is atom was modified so that a positively charged nucleus is
anything that occupies space and has mass. Virtually all of surrounded not by planetlike electrons moving in well-
the matter we have direct experience with is composed of behaved orbits, but by electron “clouds” or electron “waves”
atoms. An atom is the smallest piece of any chemical ele- (Figure 4.9c). This is why we use a featureless cloud to rep-
ment that retains the properties of that element. This book resent electrons in orbit around an atomic nucleus. Accord-
is made of atoms, and the neurons in your brain that are ing to the Bohr model of the atom, the angular momentum
(a) Parts of an atom (b) “Solar system” model (c) Quantum mechanical FIGURE 4 .9 (a) An atom
model (in this case, helium) is made
up of a nucleus consisting of
Electron cloud positively charged protons and
electrically neutral neutrons
Neutron
(neutral) and surrounded by less massive
negatively charged electrons.
Nucleus Proton (+)
(b) Atoms are often drawn as
Nucleus miniature “solar systems,” but
this model is incorrect, as the
red X indicates. (c) Electrons
Electron (–)
are actually smeared out
around the nucleus in quantum
mechanical clouds of probability.
This is a helium atom (2 protons, Electrons do not move …but rather are waves “smeared out”
2 neutrons, and 2 electrons). in orbits like planets… in a cloud of probability held in place by
the electric attraction of the nucleus.
of the electron in an orbit has to be given exactly by an inte- where something is (4x approaching zero), the less you
ger multiplied by a constant. There is no room for any uncer- can know about how fast and in what direction it is moving
tainty. In the quantum model, (4p approaching infi nity). Conversely, the better you know
however, we cannot know The uncertainty the momentum of something, the less you know about its
precisely where the electron principle tells us we location. This is not a matter of scientists making inferior
is in its orbit. This uncertainty cannot know both the measurements. You simply cannot do better, no matter how
about the electron’s location precise momentum precisely you measure!
is expressed by the famous and position of a Another example involves a different form of the Heisen-
Heisenberg uncertainty prin- particle. berg uncertainty principle:4E × 4t ≈ h, where 4E is the
ciple, named for the German uncertainty of energy and 4t is the time over which the
physicist Werner Heisenberg (1901–1976). Because particles energy is measured. In the next subsection we make the
have wave characteristics, we cannot simultaneously pin point that atoms can occupy only certain discrete energy
down both their exact location and exact momentum (p), states, implying that their electrons are restricted to spe-
defi ned as the product of mass and velocity (p = m × v). cific well-defi ned excited states. But can there really be no
There will always be some uncertainty about either the par- uncertainty whatsoever in these energy states? The answer
ticle’s position or its momentum. becomes apparent if we rewrite the previous equation as
The bottom line here is that we can be absolutely certain 4t ≈ h/4E. If we say that 4E = 0 (no uncertainty in the
that there is uncertainty at the root of everything physical. energy state), then 4t becomes infi nite.3 If an electron were
If you are bothered by this concept, you are in good com- forced to stay in a particular excited state forever, it could
pany. It really bothered Einstein too. Keep in mind that in never drop to a lower level, and we would never see narrow
the world of the very small, nothing seems intuitive. We spectral emission (bright) lines. We do, of course, see these
have learned that all electromagnetic radiation behaves bright lines, so there must be a certain amount of uncer-
as both waves and particles. Possibly more surprising was tainty in the electron’s energy. Remember that wavelength
the discovery that things like electrons and protons, which is related to energy (E = hc/λ). A mathematical operation
you may have visualized as “solid” particles, also have using calculus, which is beyond the level of this book,
wave characteristics. There is, of course, a nice symmetry tells us that 4E ∝ 4λ. A narrow range in energy therefore
here. Waves have particle-like characteristics and particles represents a narrow range in wavelength. So the longer an
have wavelike characteristics. This is not just a curious electron resides in an elevated energy state (4t is large), the
observation. It has huge implications in both science and narrower will be the spectral emission line when it fi nally
technology. For instance, the wave-particle property is the drops to a lower level (4E, and therefore 4λ, is small).
principle by which electron microscopes work, as you will
see in the next chapter.
Remember that velocity includes both speed and direc-
tion. We can be more quantitative about this. The product of
Atoms Have Discrete Energy Levels
the uncertainty in a particle’s position (4x) and the uncer- Another aspect of the wave-particle nature of electrons is
tainty in its momentum (4p) is always equal to or greater that electron waves in an atom can assume only certain
than a particular constant, which is on the order of Planck’s specific forms. (Analogously, the strings on a guitar can
constant, h. We can express this as a simple equation:
4x × 4p ≈ h. In other words, the more you know about 3
If you divide any number by zero, the result will be infi nite.
102 Chapter 4 Light
(a) (b) VISUAL ANALOGY FIGURE

4 .10 (a) Energy states of an
E5
atom are analogous to shelves
WAR & PEACE E4 in a bookcase. You can move a
book from one shelf to another,
Energy
E3
but books can never be placed
between shelves. (b) Atoms
E2 exist in one allowed energy
state or another but never
Energy states of E1 Ground state
atoms are like shelves in between. There is no level
in a bookcase. below the ground state.
E5 Atom in E4 energy state
E4
Energy E3
WAR & PEACE E2
E1 Ground state
Atom in E2 energy state
You can find a book on We use energy level diagrams Analogously, atoms exist in one
one shelf or another, but to represent the allowed states allowed energy state or another,
not in between. of an atom. but never in between.
vibrate only at certain discrete frequencies, giving rise to lower state by getting rid of some of its extra energy. An
the distinct notes we hear.) The form that the electron waves important difference between the atom and the book on the
take depends on the possible energy states of atoms. We shelf, however, is that whereas a snapshot might catch the
can imagine the energy states of atoms as being a bookcase book falling between the two shelves, the atom will never
with a series of shelves, as be caught between two energy states. When the transition
shown in Figure 4.10a. The Atoms can have only from one state to another occurs, the difference in energy
energy of an atom might cor- certain discrete between the two states must be carried off all at once. A
respond to the energy of one energies, much as common way for an atom to do this is to give off a photon.
shelf or to the energy of the guitar strings can play But not just any photon will do. The photon emitted by the
next shelf; but the energy of only certain notes. atom must carry away exactly the amount of energy lost
the atom will never be found by that atom as it goes from the higher-energy state to the
between the two shelves. A given atom may have a tremen- lower-energy state. In a similar fashion, atoms moving from
dous number of different energy states available to it, but a lower-energy state to a higher-energy state can absorb
these states are discrete. An atom might have the energy of only certain specific energies. The specific energies that
one of these allowed states, or it might have the energy of can be absorbed correspond to the allowed waveforms of
the next allowed state, but it cannot have an energy some- their electron clouds.
where in between.
The lowest possible energy state of an atom—the
“floor”—is called the ground state of the atom (Figure
4.10b.) Allowed energy levels above the ground state are
The Energy Levels of an Atom
called excited states of the atom. When the atom is in its Determine the Wavelengths of
ground state, it has nowhere to go. An electron cannot “fall”
into the nucleus because there is no allowed state there
Light That It Can Emit and Absorb
with less energy for it to occupy. It cannot move up to a To better understand the relationship between the energy
higher-energy state without getting extra energy from some- levels of an atom and the radiation it can emit or absorb,
where. For this reason an atom will remain in its ground imagine a hypothetical atom that has only two available
state forever, unless something happens to knock it into energy states. Call the energy of the lower-energy state (the
an excited state. A book sitting on the floor has nowhere ground state) E1 and the energy of the higher-energy state
left to fall, and it cannot jump to one of the higher shelves (the excited state) E2. The energy levels of this atom can be
of its own accord. represented in an energy level diagram like those in Figure
An atom in an excited state is a very different matter, 4.10, but with only two levels, as shown in Figure 4.11a.
however. Just as a book on an upper shelf might fall to a To understand the process of emission, imagine that the
lower shelf, an atom in an excited state might decay to a atom begins in the upper state (E2) and then spontaneously
A hypothetical atom has An atom with energy

two allowed energy states E2 decays to the lower
with energies E1 and E2. state with energy E1…
(a) (b)
Higher-
energy E2 E2
state
E2 E2
Energy
Energy
Ephoton = E2 – E1
λ2 = hc
1 E 2 – E1
Lower-
energy E1 E1
state
E1 E1
…by emitting a photon
that carries off the extra
FIGURE 4 .11 (a) The energy levels of a hypothetical two-level atom. (b) A photon with energy energy, E2 – E1.
hf = E2 - E1 is emitted when an atom in the higher-energy state decays to the lower-energy state.
drops down to the lower-energy state (E1). This change is cloud of gas made up of our two-state atoms. Any atom that
illustrated in Figure 4.11b, where the downward arrow found itself in the upper energy state (E2) would quickly
indicates that the atom went from the upper state to the decay and emit a photon in a random direction, and an
lower state. The atom just lost an amount of energy equal enormous number of photons would come pouring out of
to the difference between the two states, or E2 - E1. However, the cloud of gas. But instead of containing photons of all
energy is never truly lost or created, so the energy lost by different energies (that is, light of all different colors, like
the atom has to show up some- sunlight), this light would instead contain only photons
where. In this case the energy When an atom drops with the specific energy E2 - E1 and wavelength λ2→1. In
shows up in the form of a pho- to a lower energy other words, all of the light coming from the cloud would
ton that is emitted by the state, the lost be the same color.
atom. The energy of the pho- energy is carried We have all seen what happens to sunlight when it passes
ton emitted must exactly away as a photon. through a prism. Sunlight contains photons of all different
match the energy lost by the colors, so when it passes through a prism it spreads out into
atom, so the energy of the photon must be Ephoton = E2 - E1. all colors of a rainbow. But if we were to pass the light from
We have already seen how the energy of a photon is our cloud of gas through a slit
related to the frequency or wavelength of electromagnetic and a prism, as in Figure 4.12, The spectrum of a
radiation. Using this relationship, we can say that the fre- the results would be very dif- cloud of glowing
quency of the photon emitted by a transition from E2 to E1, ferent. This time there would gas contains
which we will denote as f2→1, is just the energy difference be no rainbow. Instead, all of emission lines.
divided by Planck’s constant (h): the light from the cloud of gas
would show up on the screen as a single bright line. The
Ephoton E2 ‒ E1
f2→1 = _____ = ______ . process just described—the production of a photon when
h h an atom decays to a lower-energy state—is referred to as
Similarly, the wavelength of the photon is just λ = c/f, or emission. The bright, single-colored feature in the spectrum
c hc of the cloud of gas is referred to as an emission line.
λ2→1 = ___ = ______. Until now in this discussion, we have ignored an impor-
f2→1 E2 ‒ E1
tant question: How did the atom get to be in the excited state
The energy level structure of the atom therefore determines E2 in the fi rst place? An atom sitting in its ground state will
the wavelengths of the photons emitted by an atom—the remain in the ground state unless it is somehow given just
color of the light that the atom gives off. An atom can emit the right amount of energy to kick it up to an excited state.
photons with energies corresponding only to the difference Most of the time this extra energy comes in one of two pos-
between two of its allowed energy states. sible forms: (1) the atom absorbs the energy of a photon (we
Imagine what the light coming from a cloud of gas will talk about this possibility shortly); or (2) the atom col-
consisting of our hypothetical two-state atoms would be lides with another atom, or perhaps an unattached electron,
like. This case is illustrated in Figure 4.12, which shows a and the collision knocks the atom into an excited state.
104 Chapter 4 Light
1 A hot cloud containing 2 …will emit photons 3 When light from the cloud 4 …the light shows up as a
many two-state atoms with energy E2 – E1 is collected and passed bright emission line at the
with energy E2… in all directions. through a slit and a prism… wavelength corresponding to
a photon energy of E2 – E1.
E2
Emitted
photon
Energy
Wavelength
λ2 1
Prism
Telescope Slit
E1
λ2 1
Spectrum
FIGURE 4 .12 A cloud of gas containing atoms with two energy states, E1 and E2, emits photons
with an energy E = hf = E2 - E1, which appear in the spectrogram (right) as a single emission line.
This second possibility is how a neon sign works. When a by the atom. Once again, the energy that it takes to get from
neon sign is turned on, an alternating electric field is set up E1 to E2 is the difference in energy between the two states,
inside the glass tube that pushes electrons in the gas back or E2 - E1. For a photon to cause an atom to jump from E1 to
and forth through the neon gas inside the tube. Some of E2, it must provide just this much energy. Using the relation-
these electrons crash into atoms of the gas, knocking them ship that Ephoton = hf, or f = Ephoton /h, we find that the only
into excited states. The atoms then drop back down to their photons capable of exciting atoms from E1 to E2 are photons
ground states by emitting photons, causing the gas inside whose frequency and wavelength are, respectively,
the tube to glow.
So far, we have focused on the emission of photons by Ephoton E2 ‒ E1
f1→2 = _____ = ______
atoms in an excited state, but what about the opposite pro- h h
cess? An atom in a low-energy state can absorb the energy and
of a passing photon and jump up to a higher-energy state as c hc
λ1→2 = ___ = ______ .
shown in Figure 4.13, but not just any photon can be absorbed f1→2 E2 ‒ E1
These photons have exactly the same energy—the same

FIGURE 4 .13 An atom in a lower-energy state may absorb a color of light—that is emitted by the atoms when they decay
photon of energy hf = E2 - E1, leaving the atom in a higher-energy from E2 to E1. This is not a
state. coincidence. The energy dif- Atoms can absorb
ference between the two lev- photons that have the
els is the same whether the same energies as the
atom is emitting a photon or photons they emit.
E2 absorbing one, so the energy
E2 Ephoton = E2 – E1 of the photon involved will be the same in either case.
When we shine light from a lightbulb directly though a
Energy
Only a photon
glass prism, a rainbow of colors comes out, as shown in
with energy hc
E2 – E1… λ1 2 = Figure 4.14a. What might the spectrum of light look like
E2 – E1
when viewed through a cloud composed of our hypotheti-
cal gas of two-state atoms? If we shine photons of all differ-
E1
ent wavelengths (that is, light of all different colors) through
E1
the gas from one side, almost all of these photons will pass
…can be absorbed by an atom through the cloud of gas unscathed. There is only one excep-
Ephoton = E2 – E1 with energy E1, causing the atom
tion. Rather than passing through the gas, some of the pho-
to jump to the state with energy E2.
tons with just the right energy (E2 - E1) might instead be
1 White light contains 2 …so its spectrum

all wavelengths… contains all of the
(a) Wavelengths colors of the rainbow.
Wavelength
(λ)
Prism
Telescope Slit
Light source
(b)
3 When white light passes
through a cool gas containing
hypothetical two-state atoms
Spectrum
with energy E1…
E2
Photon
absorbed
Energy
Wavelength
(λ)
E1
4 …photons with
energy E2 – E1 λ1 2 absorption line
may be absorbed… λ1 2
5 …and are missing

in a spectrum
of the light. Spectrum
(c)
FIGURE 4 .14 (a) When passed through a prism, white light produces a spectrum containing all colors. (b) When light of all
colors passes through a cloud of hypothetical two-state atoms, photons with energy hf = E2 - E1 may be absorbed, leading to the
dark absorption line in the spectrogram. (c) Absorption lines in the spectrum of a star.
absorbed by atoms. As a result, these photons will be miss- that has the same energy as the photon it just absorbed.
ing in the light passing through the prism. Where the color If the atom reemits a photon just like the one it absorbed,
corresponding to each of these you might reasonably ask why the absorption really mat-
missing photons should be, When viewed through ters. After all, the photon that was taken out of the passing
we will see a sharp, dark line a cloud of gas, the light was replaced, was it not? The answer is yes and no.
instead (Figure 4.14b). The spectrum of a The photon was replaced, true enough, but whereas all of
process of atoms capturing lightbulb contains the absorbed photons were originally traveling in the same
the energy of passing photons absorption lines. direction, the photons that are reemitted travel off in ran-
is referred to as absorption, dom directions. In other words, some of the photons with
and the dark feature seen in the spectrum is called an energies equal to E2 - E1 are, in effect, diverted from their
absorption line. Figure 4.14c shows such absorption lines original paths by their interaction with atoms. If you look
in the spectrum of a star. at a lightbulb through the cloud, you will notice an absorp-
One final point is worth making before we leave the sub- tion line at a wavelength of λ1→2 ; but if you look at the cloud
ject of emission and absorption of radiation. When an atom from the side (looking perpendicular to the original beam),
absorbs a photon and jumps up to an excited energy state, you will see it as a glowing light with an emission line at
there is a good chance that the atom will quickly decay this wavelength. AstroTour: Atomic Energy Levels and
back down to the lower-energy state by emitting a photon Light Emission and Absorption
106 Chapter 4 Light
(a) Energy states of the hydrogen atom (b) Visible emission spectrum from hydrogen
Hydrogen
E6
E5
E4 350 400 450 500 550 600 650 700 750
E3 Wavelength, λ (nanometers)
(c) Hydrogen emission spectrum (intensity vs. wavelength plot)
——hc— —— hc— ——hc— ——hc—

E2 E6 – E2 E5 – E2 E4 – E2 E3 – E2
Intensity
1 Decays to E2 emit photons 3 A common way to show a
with visible wavelengths. spectrum is to plot intensity
versus wavelength.
Energy
350 400 450 500 550 600 650 700 750

4 Each type of atom has a
unique spectral fingerprint.
2 Decays to E1 emit invisible (d) Emission spectra for helium, argon, neon, and sodium
ultraviolet photons. Helium
Argon
Neon
Sodium
E1
350 400 450 500 550 600 650 700 750
Wavelength, λ (nm)
FIGURE 4 .15 (a) The energy states of the hydrogen atom. Decays to level E2 emit photons in the
visible part of the spectrum. (b) This spectrum is what you might see if you looked at the light from a
hydrogen lamp projected through a prism onto a screen. (c) This graph of the brightness (intensity)
of spectral lines versus their wavelength illustrates how spectra are traditionally plotted. (d) Emission
spectra from several other types of gases: helium, argon, neon, and sodium.
Emission and Absorption Lines Are ble of emitting and absorbing photons that have the same
wavelengths. Figure 4.15a shows the energy level diagram
the Spectral Fingerprints of Atoms
of hydrogen, along with the spectrum of emission lines
In the previous subsection we used a hypothetical atom with for hydrogen in the visible part of the spectrum (Figures
only two allowed energy states to help us think about the 4.15b and c).
emission and absorption of photons. Real atoms have many Every hydrogen atom has the same energy states avail-
more than just two possible energy states that they might able to it, so all hydrogen atoms are, in principle, capable
occupy, so a given type of atom will be capable of emitting of producing the same spectral lines. But the energy states
and absorbing photons at many different wavelengths. An of a hydrogen atom are dif-
atom with three energy states, for example, might jump from ferent from the energy states The wavelengths at
state 3 to state 2, or from state 3 to state 1, or from state 2 available to a helium atom, a which atoms emit
to state 1. The emission lines from such a gas might have lithium atom, or a boron and absorb radiation
wavelengths of hc/(E3 - E2), hc/(E3 - E1), and hc/(E2 - E1), atom, just as the energy states form unique spectral
respectively. of these kinds of atoms differ fi ngerprints for each
The allowed energy states of an atom are determined from each other. Each differ- type of atom.
by the complex quantum mechanical interactions among ent type of atom has a unique
the electrons and the nucleus that make up the atom. Every set of available energy states and therefore a unique set of
hydrogen atom consists of a nucleus containing one proton, wavelengths at which it can emit or absorb radiation. Fig-
plus a single electron in a cloud surrounding the nucleus. ure 4.15d shows the set of emission lines that are given
As a result, every hydrogen atom has the same energy states off by discharge tubes (like those in a neon sign) contain-
available to it. It follows that all hydrogen atoms are capa- ing different kinds of atoms. These unique sets of wave-
lengths serve as unmistakable spectral fi ngerprints for these atoms in the toy instead tend to live for many sec-
each type of atom. onds before they decay. Suppose, for example, that on aver-
Spectral fingerprints are of crucial importance to astron- age these atoms tend to remain in their excited state for 1
omers. They let us figure out what types of atoms (or mol- minute before decaying and emitting a photon. In other
ecules) are present in distant objects by doing nothing more words, suppose that if we wait 60 seconds, there is a 50-50
than looking at the spectrum of light from those objects. If chance that any particular atom will have decayed and a
we see the spectral lines of hydrogen, helium, carbon, oxy- 50-50 chance that the atom will remain in its excited state.
gen, or any other element in the light from a distant object, There are trillions upon trillions of such atoms in the toy.
then we know that some of that element is present in that Although it is impossible to say exactly which atoms will
object. The strength of a line is determined in part by how decay after a minute, we can say with certainty that about
many atoms of that type are present in the source. By mea- half of them will decay within 60 seconds. From the stand-
suring the strength of the lines from different types of atoms point of the glow that we see coming from the toy, it makes
in the spectrum of a distant object, astronomers can often little practical difference which half of the atoms decay
infer the relative amounts of different types of atoms that and which half do not. All we need to know is that if we
make up the object. But it gets even better. The fraction of wait 1 minute, half of the atoms will have decayed, and the
atoms of a given kind that are in a particular energy state brightness of the glow from the toy will have dropped to
(as opposed to a different energy state) is often determined half of what it was. If we wait another minute, half of the
by factors such as the temperature or the density of the gas. remaining excited atoms will decay, and the brightness of
By looking at the relative strength of different lines from the glow will be cut in half again. Each 60 seconds, half of
the same kind of atom, it is often possible to determine the the remaining excited atoms decay, and the glow from the
temperature, density, and pressure of the material as well. toy drops to half of what it was 60 seconds earlier. The glow
from the toy slowly fades away.
We have now come upon one of the most philosophically
How Are Atoms Excited, and troubling aspects of quantum mechanics. In deep space,
where atoms can remain undisturbed for long periods of
Why Do They Decay? time, there are certain excited states of atoms that live, on
Earlier in this discussion we sidestepped an aspect of the average, for tens of million years or even longer. Envision an
emission process that has troubled physicists and philoso- atom in such a state. It may have been in that excited state
phers alike since the earliest days of quantum mechanics. for a few seconds, a few hours, or 50 million years when,
To appreciate this question, return to the analogy between in an instant, it decays to the lower-level state without any-
the emission of a photon and a book falling off a shelf. If thing causing it to do so. Newton and virtually every other
we place a book on a level shelf and do not disturb it, the physicist who lived before the turn of the 20th century
book will sit there forever. Once the book is resting on envisioned a clockwork universe in which every effect had
the upper shelf, something must cause the book to fall a cause. They imagined that if we knew the exact properties
off the shelf. So what about the atom? Once an atom is in of every bit of the universe today, it would be just a matter
an excited state, what causes it to jump down to a lower- of turning the crank on the laws of physics to predict what
energy state and emit a photon? What triggers the event? the state of the universe will be tomorrow.
Sometimes an atom in an upper-energy state can be “tick- Then, quantum mechanics came along and turned this
led” into emitting a photon—a process called “stimulated view on its head. Instead of dealing with strict cause-and-
emission”—but under most circumstances the answer is effect relationships, physicists found themselves calculat-
that nothing causes the atom to jump to the lower-energy ing the probabilities of certain
state. There is no trigger. Instead the atom decays sponta- events taking place and facing Quantum mechanics
neously. And while we can say about how long the atom is fundamental limitations on undermines the
likely to remain in the excited state, the rules of quantum what can ever be known about orderly, causal
mechanics say (and experiment shows) that we cannot the state of the universe. We universe of
know exactly when a given atom will decay until after the have mentioned that although Newtonian physics.
decay has happened. The atom decays at a random time that Einstein helped start the sci-
is not influenced by anything in the universe and cannot entific revolution of quantum mechanics, in the end that
be known ahead of time. revolution left him behind. He could never shake his firm
You have seen many examples of this rather amazing belief in Newton’s clockwork, causal universe. “God does
phenomenon. For example, you may be familiar with toys not play dice with the universe!” he insisted emphatically.
that “glow in the dark.” Photons in sunlight or from a light- As more of the predictions of quantum mechanics were
bulb are absorbed by the certain “phosphorescent” atoms borne out by experiment, most physicists came to accept
in the toy, knocking those atoms into excited energy states. the implications of the strange new theory. Einstein, how-
Unlike many excited energy states of atoms that tend to ever, went to his grave looking unsuccessfully for a way to
decay in a small fraction of a second, the excited states of save his notion of order in the universe.
108 Chapter 4 Light
It is interesting to note that although Einstein refused can reveal a wealth of information about the physical state
to accept quantum mechanics, our understanding of the of material located tremendous distances away. The nature
quantum mechanical nature of reality owes him a great of light had forced physicists to abandon many of their most
debt. His was one of the greatest minds of all time, and as cherished ideas about the nature of matter and energy. Yet
he searched tirelessly for flaws in quantum mechanics, he we have only begun to explore what light can tell us. It is
presented challenge after challenge to those who were trying time to step back from the precipice of the philosophical
to work out the details of the new theory. It was in respond- implications of quantum mechanics and look instead at how
ing to Einstein’s objections that physicists were forced to light can be used to measure one of the most straightforward
confront the full implications of their own work. This strug- questions about a distant astronomical object: Is it moving
gle continues to this day. A few theoretical physicists are away from us or toward us, and at what speed?
still pursuing Einstein’s dream, trying to recast quantum Have you ever stood outside and listened as a fi re truck
mechanics in a way that recaptures the strict causality that sped by with sirens blaring? If so, you might have noticed
seemed irretrievably lost shortly after the beginning of the something funny about the way the siren sounded. As the
20th century. So far, they have had little success, and most fire truck came toward you, its siren had a certain high pitch;
physicists doubt that they ever will. Like Einstein’s before but as it passed by, the pitch of the siren dropped notice-
them, however, their healthy skepticism has led to ever- ably. If you were to close your eyes and listen, you would
deeper understanding of the implications and limitations have no trouble knowing when the fi re truck passed you,
of the theory of quantum mechanics. just from the change in pitch of its siren. You do not even
need a fi re engine to hear this effect. The sound of normal
traffic behaves in the same way. As a car drives past, the
pitch of the sound that it makes suddenly drops.
4.4 The Doppler Effect— The pitch of a sound is like the color of light. It is deter-
mined by the wavelength or, equivalently, the frequency of
Is It Moving toward Us the wave. What we perceive as higher pitch corresponds to
sound waves with higher frequencies and shorter wave-
or Away from Us? lengths. Sounds that we perceive as lower in pitch are waves
with lower frequencies and
We have begun to see that to an astronomer, light is far more longer wavelengths. When an The motion of a
than just the stuff that bounces off the page and lets you object is moving toward us, source toward or
read these words. Light is a tightly packed bundle of infor- the waves that it gives off away from us changes
mation that, when spread into its component wavelengths, “crowd together” in front of the wavelength of the
the object. You can see how waves reaching us.
this works by looking at Fig-
ure 4.16, which shows the locations of successive wave
FIGURE 4 .16 Motion of a light or sound source relative to an
observer may cause waves to be spread out (redshifted, or lower
crests given off by a moving object. If you are standing in
in pitch) or squeezed together (blueshifted, or higher in pitch). front of an object moving toward you, the waves that reach
A change in the wavelength of light or the frequency of sound is you have a shorter wavelength and therefore a higher fre-
called a Doppler shift. quency than the waves given off by the object when it is not
moving. In the case of sound waves, the sound reaching you
Waves that reach this observer Waves that reach this observer
are spread out to longer are squeezed to shorter
from the object has a higher pitch than the sound that would
“redshifted” wavelengths “blueshifted” wavelengths be given off by the object if it were stationary. Conversely,
(lower frequency). (higher frequency). if an object is moving away from you, the waves reaching
you from the object are spread out. (Again, refer to Figure
4.16.) In the case of sound, this means that the pitch of the
sound drops, in line with our experience with the fire truck.
AstroTour: The Doppler Effect
This same phenomenon, which is referred to as the
v Doppler effect and named after Austrian physicist Christian
Doppler (1803–1853),4 occurs with light as well. The wave-
length of the light as measured from the source’s frame of
Speed reference is called the rest wavelength (λ0), as shown in
of light Figure 4.17a. If an object is moving toward you, the light
c
Moving source This observer sees
4
The phenomenon is sometimes referred to as the “Doppler-Fizeau
of light no Doppler shift. effect” to honor the French physicist Hippolyte Fizeau, who discov-
ered the phenomenon independently.
4.4 The Doppler Effect—Is It Moving toward Us or Away from Us? 109
(a)
The amount by which the wavelength of light is shifted by
the Doppler effect is called the Doppler shift of the light.
As long as the speed of an object is much less than the
speed of light (which means we can ignore relativistic
effects), the observed wavelength of the Doppler-shifted
(b) light, λobs, is given by the equation
λobs = (1 + __
c ) λ0 .
vr
In this expression, λ0 is the rest wavelength of the light.

The velocity v r is the velocity at which the object is moving
(c) relative to you. To be more precise, v r is the rate at which
the distance between you and the object is changing. If v r
is positive, the object is getting farther away from you. If v r
is negative, the object is getting closer to you.
Take a moment to be sure this expression makes sense
to you. Our argument says that if an object is moving
FIGURE 4 .17 Spectral lines of astronomical objects will away from you, the wavelength of the light coming to you
be blueshifted or redshifted from their rest wavelength (a), from that object should be greater than it would be if the
depending on whether they are moving toward us (b) or away object were “at rest” (not moving relative to you). If the
from us (c). object is moving away, v r will be greater than 0. If v r is
greater than 0, then 1 + vr /c is greater than 1, and therefore
λobs = (1 + vr /c)λ0 will be larger than λ0, so we see a red-
reaching you from the object has a shorter wavelength than shift. This is as we expected. What about the opposite
its rest wavelength. In other words, the light that you see is case? If an object is moving toward us, v r is less than 0,
bluer than if the source were so (1 + vr /c) will be less than 1. Now we fi nd that λobs is
not moving toward you. We Light from shorter than λ0. The observed wavelength is shorter than
say that the light from an approaching objects the rest wavelength, so we see a blueshift. Once again,
object moving toward us is is blueshifted. Light this observation is in line with our expectations (see
blueshifted (Figure 4.17b.) In from receding objects Math Tools 4.2).
contrast, light from a source is redshifted. Remember that the Doppler shift provides information
that is moving away from you only about whether an object is moving toward you or away
is shifted to longer wavelengths. The light that you see is from you. That is what the subscript r in v r signifies. The
redder than if the source were not moving away from you, variable v r stands for radial velocity, which is the rate at
so we say that the light is redshifted, as seen in Figure 4.17c. which the distance between you and the object is changing.
Making Use of the Doppler Effect

A prominent spectral line of hydrogen atoms has a rest wave- Consider the case of our nearest stellar neighbor, Alpha Cen-
length, λ0, of 656.3 nm (see Figure 4.15b). Suppose that, using a tauri, which is moving toward us at a radial velocity of -21.6
telescope, you measure the wavelength of this line in the spec- km/s (-2.16 × 104 m/s.) What would be the observed wavelength,
trum of a distant object and fi nd that instead of seeing the line λobs, of a magnesium line in Alpha Centauri’s spectrum having
at 656.3 nm, you see the line at a wavelength, λobs, of 659.0 nm. a rest wavelength, λ0, of 517.27 nm?
( )
You could then infer that the object is moving at a velocity of
λobs - λ0
vr = _______
659.0 nm - 656.3 nm
c = ________________ × (3 × 108 m/s)
( v
c ) –2.16 × 104 m/s
λobs = 1 + __r λ0 = 1 + ____________ × 517.27 nm
3 × 108/s
λ0 656.3 nm
= 517.23 nm.
= 1.2 × 106 m/s.
Although the observed Doppler blueshift (517.23 - 517.27)
The object is moving away from you with a radial velocity of is only -0.04 nm, it is easily measurable with modern
1.2 × 106 m/s, or 1,200 km/s. instrumentation.
110 Chapter 4 Light
At the moment that the fire truck is passing you, it is neither Doppler shifts become especially useful when we are
getting closer to you nor getting farther away from you, so looking at an object that has emission or absorption lines
the pitch you hear is the same in its spectrum. These spectral lines enable astronomers to
as the pitch heard by the crew Radial velocity is determine how rapidly the object is moving toward or away
riding on the truck. You can measured from from us. To determine the speed, astronomers first identify
see this directly by looking at Doppler shifts the spectral line as being from a certain type of atom, which
Figure 4.16 or by referring to of emission or has a unique rest wavelength (λ0). They then measure the
the previous equation. If an absorption lines. wavelength (λobs) of the spectrum of the distant object. The dif-
object is moving perpendicu- ference between the rest wavelength and the observed wave-
lar to your line of sight, then vr = 0 and λobs = λ0. The observed length enables us to determine the object’s velocity. Turning
wavelength equals the rest wavelength. The light is neither the preceding expression around a bit, we can write
redshifted nor blueshifted. (In the case of light, we need to
λobs – λ0
stress again the caveat that this result holds true only if the vr = _______ c.
λ0
object’s velocity is much less than the speed of light. In
Chapter 17 you will learn more about how relativity influ- If you know λ0, just measure λobs, plug both values into this
ences an object moving at nearly the speed of light.) expression, and voilà!—you know what v r is.
Equilibrium Means Balance

Equilibrium is the term we use to refer to systems that are in position. This is an example of an unstable equilibrium. When
balance. Imagine two well-matched teams struggling in a tug- an unstable equilibrium is disturbed, it will move farther away
of-war contest. Each team pulls steadfastly on the rope, but the from equilibrium rather than back toward it.
force of one team’s pull is only enough to match but not over- Not all types of equilibrium are static. Equilibrium can also
come the force exerted by the other team. Muscles flex, but the be dynamic, in which the system is constantly changing. In
scene does not change. A picture taken now and another taken dynamic equilibrium, one source of change is exactly balanced
5 minutes from now would not differ in any significant way. by another source of change, so that the configuration of the
In this book we will frequently encounter this kind of static system remains the same. Figure 4.19 shows a simple example
equilibrium, represented by a tug-of-war in which opposing of dynamic equilibrium. A can with a hole cut in the bottom
forces just balance each other. The equilibrium between the has been placed under an open water faucet. The depth of the
downward force of gravity and the pressure that opposes it will water in the can determines how fast water pours out through
play a central role in the stories of planetary interiors, planetary the hole in the bottom of the can. Once the water reaches just
atmospheres, and the interiors of stars. Static equilibrium can the right depth, as in Figure 4.19a, water continues to pour out
be stable, unstable, or neutral, as shown in Figure 4.18. Consider of the bottom of the can at exactly the same rate it pours from
the image in Figure 4.18b or a book standing on its edge, unsup- the faucet into the top of the can. The water leaving the can
ported on either side by other objects or books. If you nudge the balances the water entering the can, and equilibrium is estab-
book, it will fall over rather than settling back into its original lished. If you were to take a picture now and another picture a
few minutes from now, little of the water in the can would be
the same. Even so, the pictures would be indistinguishable.
FIGURE 4 .18 Examples of stable (a), unstable (b), and If a system is not in equilibrium, its configuration will
neutral (c) equilibrium. Imagine what happens to the ball if, in change. Look at Figure 4.19b. Here the level of the water in the
each case, you give it a small nudge. can is too low, so water will not flow out of the bottom of the
can fast enough to balance the water flowing into the can, and
thus the water level will rise. A picture taken now and another
a short time later will not look the same. The configuration of
the system is changing. Conversely, if the water level in the can
is too high, as in Figure 4.19c, water will flow out of the can
faster than it flows into the can, and the water level will fall.
Once again, if the system is not in equilibrium, its configura-
(a) Stable (b) Unstable (c) Neutral
equilibrium equilibrium equilibrium tion will change.
Water passing through a can is an example of a stable equi-
ammonia remain permanently frozen? And closer to home,

4.5 Why Mercury Is Hot why is the surface of Earth hot enough for water to melt over
most of the planet, but cold enough for that water to remain
and Pluto Is Not as a liquid instead of turning into vapor?
The temperature of any object is determined by two
Spectral lines also enable us to determine a distant object’s things: what is trying to heat up the object and what is
temperature. If you asked an elementary-school student why trying to cool it down. If an
Mercury is hot and Pluto is cold, he or she would probably object’s temperature is con- If an object’s
look at you like you were hopelessly uninformed and then stant, then these two must be temperature is
patiently explain that Mercury is hot because it is close to in balance with each other. constant, then heating
the Sun, while Pluto is cold because it is much farther away. Your body, for instance, is and cooling must be in
This explanation is fine as far as it goes, but we should push heated by the release of chem- balance.
a bit harder. Why are the planets as hot as they actually ical energy from inside. It is
are? Why does the surface of Mercury reach temperatures also sometimes heated by energy from your surroundings.
that are hot enough to melt lead while the surface of Pluto If you are standing in sunshine on a hot day, the hot air
remains so cold that even substances such as methane and around you and the sunlight falling on you both are work-
FIGURE 4 .19 Water

1 Pressure determines the rate at flowing into and out of a can
which water flows out of a hole
determines the water level in
in a can. The higher the water
level, the faster the flow. the can. This is an example of
dynamic equilibrium.
(a) Equilibrium (b) Water level low (c) Water level high
2 When the water is at the 3 When the water level is 4 When the water level is
correct depth, flow out of the too low, water flows out too high, water flows out
can just balances flow into the more slowly than it flows faster than it flows in, and
can. Equilibrium is achieved. in, and the water level rises. the level in the can falls.
The depth of the water does
not change.
librium. When a stable equilibrium is disturbed (forced away equilibrium. If you take Figure 4.19 and replace “water flows
from its equilibrium configuration), it will tend to return to its in” with “sunlight is absorbed,” “water flows out” with “energy
equilibrium state. If the water level is too high or too low, it is radiated by the planet,” and “water level” with “the planet’s
will move back toward its equilibrium level. The equilibrium temperature,” then the stable equilibrium that sets the level of
that we discuss in this chapter, between sunlight falling on a the water in the can becomes the stable equilibrium that sets
planet and thermal energy radiated away into space, is a stable the temperature of a planet.
112 Chapter 4 Light
ing to heat you up. In response to this heating, your body

must have some way to cool itself off. This is where per-
Temperature Is a Measure of
spiration comes in. When you perspire, water seeps from How Energetically Particles
the pores in your skin and evaporates. It takes energy to
evaporate water, and much of this energy comes from your
Are Moving About
body. Thus, as the perspiration evaporates it carries away When we say that something is hot or cold, we know exactly
your body’s thermal energy, cooling your body down. For what we mean. In everyday life hot and cold are defi ned in
your body temperature to remain stable, the heating must terms of our subjective experiences. Something is hot when
be balanced by the cooling. Such a balance is referred to it feels hot or cold when it feels cold. But our perceptions of
as thermal equilibrium. If your body is out of thermal hot and cold are a layer of subjective experience that comes
equilibrium in one direction—if there is more heating than between us and a defi nable, quantifiable concept called
cooling—then your body temperature climbs. If your body temperature. When we talk about temperature, we speak
is out of thermal equilibrium in the other direction—if of degrees on a thermometer. The way we define a “degree”
there is more cooling than heating—then your body tem- is arbitrary—a matter of convention. If you grew up in the
perature falls. United States, for example, you probably think of tempera-
Planets also have a thermal equilibrium, and electro- tures in “degrees Fahrenheit” (°F), whereas if you grew up
magnetic radiation plays a crucial role in maintaining that anywhere else in the world, you think of temperatures in
equilibrium. The energy from “degrees Celsius” (°C). Both of these are perfectly reason-
sunlight heats the surface of Earth is heated by able scales for measuring temperatures. But what does the
Earth, driving its temperature sunlight and cooled by thermometer actually measure?
up. This is one side of the the radiation of energy What we refer to as temperature is actually a measure-
equilibrium. The other side of back into space. ment of how energetically the atoms that make up an object
Earth’s thermal equilibrium are moving about. The air around you is composed of vast
is also controlled by light energy. Earth radiates energy numbers of atoms and molecules. Those molecules are mov-
back into space, thereby cooling itself. Our eyes are not ing about every which way. Some move slowly; some move
sensitive to the wavelengths of light that Earth radiates, more rapidly. Similarly, the atoms that make up the chair
but that light is there nonetheless. Overall, Earth must radi- that you are sitting in or the floor that you are walking on
ate away just as much energy into space as it absorbs from (or the anatomical parts of your body that are involved in
the Sun. If there were more heating than cooling, the tem- those two activities) are constantly in motion. We can char-
perature of Earth would climb. (It would absorb more acterize these motions by talking about the average kinetic
energy from the Sun than it got rid of, and this imbalance energy (EK), the energy of motion:
would show up in increasing temperature.) If there were
1
more cooling than heating, the temperature would fall. For EK = __ mv 2,
2
Earth to remain at the same average temperature over
time—which it has done for many centuries5—the energy where m is the mass of a particle and v is its velocity.
that Earth radiates into space must exactly balance the The more energetically the atoms or molecules in some-
energy absorbed from the Sun. Thermal equilibrium must thing are bouncing about, the higher is that something’s
be maintained. Equilibrium is an important concept in sci- temperature. In fact, the random motions of atoms and
ence. There are many kinds of equilibrium besides thermal molecules are often referred to as their thermal motions,
equilibrium, some of which we will encounter later in the to emphasize the connection between these motions and
book. Foundations 4.1 describes the basic properties of temperature.
equilibrium (see pp. 110–111). This defi nition of temperature should make some intui-
We now have a qualitative understanding that is inter- tive sense. If something is hotter than you are, experience
esting but is not yet quantitatively predictive. We would says that thermal energy flows from that object into you. At
like to turn this intuitive idea of thermal equilibrium into the atomic level that means the object’s atoms are bounc-
a real prediction for the temperatures of the planets. To do ing more energetically than are the atoms in your body, so
this we need to fi nd out more about light and temperature if you touch the object, its atoms collide with your atoms,
and the relationship between the two. Before going too far causing the atoms in your body to move faster. Your body
down this path, however, we should start by improving our gets hotter as thermal energy flows from the object to you.
understanding of what we mean by temperature. (At the same time, these collisions rob the particles in the
object of some of their energy. Their motions slow down.
5
Over the past century, Earth’s average temperature has slightly
The hotter object becomes cooler.) In general, when we talk
increased. Many climatologists attribute this rise in temperature about heating, we mean processes that increase the average
to a small decrease in the amount of energy radiated back to space, thermal energy of an object’s particles, and when we talk
which is caused by an increase in the amount of “greenhouse” gases about cooling, we mean a process that decreases the average
in our atmosphere. thermal energy of those particles. The connection between
Temperature (K) Temperature (K)

FIGURE 4 . 20 Hotter gas
1 Gas contains atoms 300 2 Temperature is 300 3 Doubling the gas
temperatures correspond
and molecules related to the temperature in a fixed
moving about average speed of box quadruples the to faster motions of the
0 600 0 600
in all directions. the gas particles. average particle speed. constituent atoms. K stands
for kelvin, the unit in the Kelvin
scale of temperature, which
is commonly used in scientific
investigations.
the temperature and the motion of particles is illustrated Hotter Means More
in Figure 4.20.
The change in thermal energy associated with a change
Luminous and Bluer
of one unit, or degree, is arbitrary on any temperature scale. So far, we have focused our attention on the way discrete
On the Fahrenheit scale there atoms emit and absorb radiation. Our discussion led us to
are 180 degrees between the Absolute zero, the a useful understanding of emission lines and absorption
melting point (32°F) and the temperature at lines and how we might use these lines to learn about the
boiling point (212°F) of water which thermal motions physical state and motion of distant objects. But not all
at sea level. On the Celsius stop, is zero on the objects have spectra that are dominated by discrete spectral
scale there are 100 degrees Kelvin scale. lines. For instance, if you pass the light from an incandes-
between those two tempera- cent lightbulb through a prism, as we saw in Figure 4.14a,
tures. For these two scales, the temperature corresponding then instead of discrete bright and dark bands you will see
to “zero degrees” is also arbitrary. On the Celsius scale, 0°C light spread out smoothly
is chosen to be the temperature at which water freezes; on from the blue end of the spec- Objects like lightbulbs
the Fahrenheit scale, 0°F is placed at a temperature corre- trum to the red. Similarly, if emit continuous
sponding to -17.78°C. However, there is a lowest possible you look closely at the spec- radiation at all
physical temperature below which no object can fall. As trum of the Sun, you will see wavelengths.
the motions of the atoms in an object slow down, the tem- absorption lines, but mostly
perature drops lower and lower. When the motions of the you will see light smoothly spread out across all colors of
particles fi nally stop, things have gotten as cold as they can the spectrum—red through violet. What is the origin of such
get. This lowest possible temperature, where thermal continuous radiation, and what clues might this kind of
motions have come to a standstill, is called absolute zero. radiation carry about the objects that emit it?
Absolute zero corresponds to -273.15°C, or -459.57°F. We can think of a dense material as being composed of a
The preferred temperature scale for most scientists collection of charged particles that are being jostled as their
is the Kelvin scale. For convenience, the size of one unit thermal motions cause them to run into their neighbors. The
on the Kelvin scale, called a kelvin (abbreviated K) is the hotter the material is, the more violently its particles are
same as the Celsius degree. What makes the Kelvin scale being jostled. Anytime a charged particle is subjected to an
special is that 0 kelvins is set equal to that absolute lowest acceleration, it radiates, so the jostling of particles due to
temperature where thermal motions stop—absolute zero. their thermal motions causes them to give off a continuous
There are no negative temperatures on the Kelvin scale. spectrum of electromagnetic radiation. This is why any mate-
The importance of the Kelvin scale is that when tempera- rial that is sufficiently dense for its atoms to be jostled by
tures are measured in kelvins, we know that the average their neighbors will emit light simply because of its tempera-
thermal energy of particles is proportional to the measured ture. Radiation of this sort is called thermal radiation.
temperature. The average thermal energy of the atoms in We can guess how the radiation from an object changes
an object with a temperature of 200 K is twice the average as the object heats up or cools down. Start with the ques-
thermal energy of the atoms in an object with a tempera- tion of the power (energy per second, measured in watts)
ture of 100 K. radiated, which is referred to as the object’s luminosity. To
114 Chapter 4 Light
a physicist or an astronomer, luminosity is the amount of

light leaving a source. By contrast, the brightness of electro-
magnetic radiation is the amount of light that is arriving at
a particular location, such as the page of the book you are T = 6,000 K, λpeak = 0.48 μm
reading or the pupil of your eye. The hotter the object, the Making an object
more energetically the charged particles within it wiggle. hotter makes it
(Again, this is what it means to be hot.) The more energeti- more luminous…
cally the charged particles move, the more energy they emit
…and shifts the peak
in the form of electromagnetic radiation. So as an object
of its Planck spectrum
gets hotter, we expect the light that it emits to become more to shorter wavelengths.
intense. Here is our fi rst intuition about thermal radiation:
Intensity
T = 5,000 K, λpeak = 0.58 μm
hotter means more luminous.
Now we move to the question of what color light an object
T = 4,000 K, λpeak = 0.73 μm
emits. Again, as the object gets hotter, the thermal motions
of its particles become more T = 3,000 K, λpeak = 0.97 μm
energetic. These more ener- Making an object
hotter also makes its T = 2,000 K, λpeak = 1.45 μm
getic motions are capable of
producing more energetic thermal radiation
photons. So as an object gets bluer and more 0
hotter, we might expect the luminous.
average energy of the photons 0 0.5 1.0 1.5 2.0 2.5 3.0
Wavelength, λ (μm)
that it emits to increase. In other words, we might expect
the average wavelength of the emitted photons to get shorter. FIGURE 4 . 2 1 Planck spectra emitted by sources with
The light from the object gets bluer. Here is our second intu- temperatures of 2,000 K, 3,000 K, 4,000 K, 5,000 K, and
ition about thermal radiation: hotter means bluer. 6,000 K. At higher temperatures the peak of the spectrum shifts
Both of these intuitive predictions are borne out by a toward shorter wavelengths, and the amount of energy radiated
per second from each square meter of the source increases.
simple experiment that you can do. An incandescent light-
bulb is a good example of an object that emits thermal radia-
tion. The electric current in the lightbulb filament heats the
filament. (More precisely, electrons being pushed through bluer. The detailed answers to these questions were worked
the filament by electric fields collide with atoms in the fila- out around 1900 by Max Planck. Planck was thinking about
ment, increasing the thermal motions of those atoms.) The a special situation—a hollow,
hot filament then glows. Somewhere in your home, dorm, or totally enclosed cavity of A blackbody emits
classrooms you can probably fi nd a lightbulb with a rheo- material at a specific tempera- thermal radiation
stat (a dimmer). Turning the knob on the dimmer changes ture, T. The crucial point here that has a Planck
the amount of electric current in the filament in the light- is that, inside the cavity, all of spectrum.
bulb. Turning the dimmer up increases the current, which the radiation emitted by the
increases the number and strength of collisions between cavity walls is also absorbed by the walls. In this situation
electrons and atoms, which in turn increases the tempera- a balance is set up, with each bit of the wall emitting just
ture of the filament. The hotter filament is more luminous. as much thermal radiation as it is absorbing from its sur-
This demonstration confi rms the fi rst of our expectations: roundings. Physicists refer to such a special setup as a
hotter means more luminous. blackbody. Planck used this balance to calculate the spec-
What about the color of the emitted light? Look again at trum of the light inside such a cavity. The result of his cal-
the lightbulb as you turn up the dimmer. When the bulb first culation, which beautifully matches the results of
comes on, it glows a dull red; but as the current through the experiments, is called a Planck spectrum or a blackbody
bulb increases, driving up the temperature of the filament, spectrum. Figure 4.21 shows plots of the Planck spectra for
the perceived color of the light changes. When the dimmer objects at several different temperatures.
is turned all the way up, the light from the bulb has lost its You might reasonably ask what this hypothetical cav-
red tint. The hotter the lightbulb gets, the more the highly ity has to do with the light from the filament of a lightbulb.
energetic blue photons become mixed with the less energetic Surely the filament of a lightbulb is not a cavity of this sort!
red photons, and the light becomes whiter. The color of the But in a certain sense it is. The light emitted by charged
light shifts from red toward blue, confi rming our second particles within the filament is mostly absorbed by other
intuitive expectation: hotter means bluer. charged particles within the filament. This is exactly the
These observations offer an intuitive grasp of how the assumption that Planck made when calculating the shape
light given off by an object depends on the temperature of of the spectrum of a blackbody. As a result, we expect that
the object, but to be really useful they need to be quantified. the radiation existing inside the filament of the lightbulb
We need to know how much more luminous and how much will have a Planck spectrum. This radiation “leaks out”
Working with the Stefan-Boltzmann and Wien’s Laws

We can use the Stefan-Boltzmann law to figure out what the sur- the peak in the spectrum is, we can use Wien’s law to calculate
face area of the filament in a lightbulb must be. Suppose the fila- the temperature of the object. For example, we have no way of
ment in an incandescent bulb operates at a temperature of about dropping a thermometer into the Sun and directly measuring
2,500 K. The amount of energy radiated by the bulb is stamped its temperature, but we can observe the spectrum of the light
right on the face of the bulb—a 100-W bulb has a luminosity of coming from the Sun. When we do so, we fi nd that the peak in
100 W, which means that it radiates away 100 joules each second. the spectrum occurs at a wavelength of about 0.5
m. Wien’s
(A joule is a unit of energy, abbreviated J.) If the total amount of law can be written as
light from the filament is equal to the flux () multiplied by the
2,900 μm K
surface area of the filament (A), we can write T = _________.
λpeak
 = 100 W = A × σT 4. If we plug the observed peak of the spectrum of the Sun (λpeak
= 0.50
m) into this equation, we get
Solving this equation for the area, we get
2,900 μm K
100 W T = _________ = 5,800 K.
A = _____ 0.5 μm
σT 4
100 W This is how we know the temperature of the Sun.
A = ________________________ = 4.5 × 10-5 m2.
( W
5.67 × 10-8 ____
m2K4
)
× (2,500 K)4
Suppose we want to calculate the peak wavelength at which
Earth radiates. The average temperature of Earth is 288 K. If we
insert this temperature into Wien’s law, we get
Note that not much fi lament surface area is needed to provide
2,900 μm K
the light that turns night into day in our homes and cities. λ peak = _________ = 10.1
m.
288 K
Wien’s law (described in the text) will prove handy as we
continue our study of the universe. If we can measure the spec- or slightly more than 10
m. This result tells us that Earth radi-
trum of an object emitting thermal radiation and fi nd where ates in the infrared region of the spectrum.
of the filament, just as light might leak out of a small hole increase in luminosity is proportional to the fourth power
in the side of Planck’s cavity. As a result, the spectrum of of the temperature: Luminosity ∝ T 4. This result is known
radiation from the filament of a lightbulb is very close to a as the Stefan-Boltzmann law because it was discovered in
Planck spectrum. The light from stars such as the Sun and the laboratory by Austrian physicist Josef Stefan (1835–1893)
the thermal radiation from a planet also often come close and derived by his student, Ludwig Boltzmann (1844–1906),
to having a blackbody spectrum. before Planck’s theory came along to explain it.
What the Stefan-Boltzmann law actually says is that
the amount of energy radiated by each square meter of the
The Stefan-Boltzmann Law Says That surface of an object is given by the equation
Hotter Means Much More Luminous  = σT 4.
We now ask the question that scientists must always ask:
Does the theoretical prediction agree with observation and In this equation  is called the flux. It is a measurement
experiment? Do the Planck of the total amount of energy coming through each square
spectra in Figure 4.21 agree The luminosity of meter of the surface each second. The constant σ (pro-
with our intuitive ideas and a blackbody is nounced “sigma”) is called the Stefan-Boltzmann constant.
with our experiment with the proportional to T 4. The value of σ is the same for all cases and is given by
lightbulb and the dimmer? 5.67 × 10 –8 W/(m2 K4) (where W stands for watts). To fi nd the
Begin with luminosity. As the temperature of an object total amount of energy emitted by an object in the form of
increases, Planck’s theory says that the object gives off more electromagnetic radiation, multiply  by the surface area
radiation at every wavelength, so the luminosity of the object of the object (see Math Tools 4.3.)
should increase. In fact, it increases in a hurry. Adding up The Stefan-Boltzmann law says that an object rapidly
all of the energy in a Planck spectrum shows that the becomes more luminous as the temperature increases. If
116 Chapter 4 Light
the temperature of an object goes up by a factor of 2, the too sloppy for science. Luminosity refers to the amount of
amount of energy being radiated each second increases by light leaving a source. The concept of brightness is certainly
a factor of 24, or 16. If the tem- related to the concept of luminosity. For example, replac-
perature of an object goes up Slight changes in ing a lightbulb with a luminosity of 50 W with a 100-W
by a factor of 3, then the temperature mean bulb succeeds in making a room twice as bright because
energy being radiated by the large changes in it doubles the light reaching any point in the room. But
object each second goes up by brightness. brightness also depends on the distance from a source of
a factor of 34, or 81! A light- electromagnetic radiation. If you needed more light to read
bulb with a filament temperature of, say, 3,000 K radiates this book, you could replace the bulb in your lamp with a
16 times as much light as it would if the fi lament tempera- more luminous one, but it would probably be easier just to
ture were 1,500 K. Even modest changes in temperature move the book closer to the light. Conversely, if a light were
can result in large changes in the amount of power radi- too bright for you, you could move away from it. Our every-
ated by an object. day experience says that as we move away from a light, its
brightness decreases.
The particle description of light provides a convenient
Wien’s Law Says That way to think about the brightness of radiation and how
brightness depends on dis-
Hotter Means Bluer tance. Suppose you had a Brightness measures
Look again at Figure 4.21, but this time, instead of paying piece of cardboard that was 1 how much light
attention to how high each curve is, notice where the peak meter on an edge. Intuitively falls per square meter
of each curve falls along the you might imagine that mak- per second.
horizontal axis. As the tem- The peak wavelength ing the light that falls on the
perature increases, the peak of a blackbody is cardboard twice as bright would mean doubling the number
of the Planck spectrum shifts inversely proportional of photons that hit the cardboard each second. Tripling the
toward shorter wavelengths, to its temperature. brightness of the light would mean increasing the number
which means that the average of photons hitting the cardboard each second by a factor of
energy of the photons becomes greater. Just as we surmised, 3, and so on. Here is a beginning point for understanding
increasing the temperature causes the light from the object brightness. Brightness depends on the number of photons
to get bluer. The shift in the location of the Planck spec- falling on each square meter of a surface each second.
trum’s peak wavelength with increasing temperature is Working with this idea of brightness, now imagine a
given by the equation lightbulb sitting at the center of a spherical shell, as shown
in Figure 4.22. Photons from the bulb travel in all direc-
2,900
m K
λpeak = __________ . tions and land on the inside of the shell. To find the number
T
of photons landing on each square meter of the shell dur-
This result is referred to as Wien’s law, named for German ing each second (that is, to fi nd the brightness of the light),
physicist Wilhelm Wien (1864–1928). In this equation, λpeak take the total number of photons given off by the lightbulb
(pronounced “lambda peak”) is the wavelength where the each second and divide by the number of square meters
Planck spectrum is at its peak. It is the wavelength where over which those photons have to be spread. The surface
the electromagnetic radiation from an object is greatest. area of a sphere is given by the formula A = 4πr 2, where r is
Wien’s law says that the location of the peak in the spectrum the distance between the bulb and the surface of the sphere
is inversely proportional to the temperature of the object. (thus, r = the radius of the sphere). When this relationship
If you increase the temperature by a factor of 2, the peak is written as an equation, we fi nd that
wavelength becomes half of what it was. If you increase the
temperature by a factor of 3, the peak wavelength becomes Total number of photons
a third of what it was (see Math Tools 4.3.)
( Number of photons
)
striking one square =
meter each second
emitted per second
____________________
Number of square meters
the photons are spread over
4.6 Twice as Far Means Total number of photons

emitted each second
= __________________
One-Fourth as Bright 4πr2
You might have noticed that we have consistently spoken of The next step in building an understanding of bright-
the “luminosity” of objects, where in everyday language we ness is to change the size of the spherical shell while keep-
probably would have just said that one object is “brighter” ing the total number of photons given off by the lightbulb
than another. This is a case where everyday language is each second the same. As the shell becomes larger, the pho-
4.7 Radiation Laws Enable Us to Calculate the Equilibrium Temperatures of the Planets 117
1 Light must spread out more The luminosity of an object is the total number of pho-
to cover a larger sphere, tons given off by the object multiplied by the energy of each
and so will be fainter, farther photon. Instead of talking about how the number of photons
from a source. must spread out to cover the surface of a sphere (brightness),
r=3m we now talk about how the energy carried by the photons
must spread out to cover the surface of a sphere. When
speaking of brightness in this way, we mean the amount
of energy falling on a square meter in a second. If L is the
luminosity of the bulb, then the brightness of the light at a
r=2m distance r from the bulb is given by
Energy radiated per second

r=1m Brightness = ___________________________
Area over which energy is spread
L
= _____2 .
4πr
Usually, the only information that astronomers have to

Brightness = 1 unit
work with is the light from a distant object. For this reason
Brightness = 1/ 4 unit we will use our understanding of radiation over and over
Brightness = 1/ 9 unit again throughout our journey. Time spent now thinking
2 The area of a sphere = 4πr 2,
1/ r 2. carefully about the electromagnetic spectrum, the emission
so brightness
and absorption of photons, Planck radiation, and the inverse
FIGURE 4 . 22 Light obeys an inverse square law as it spreads square law for brightness will be a very good investment
away from a source. Twice as far means one-fourth as bright. for what is to come.
tons from the lightbulb must spread out to cover a larger

surface area. Each square meter of the shell receives fewer
4.7 Radiation Laws Enable
photons each second, so the brightness of the light decreases.
If the shell’s surface is moved twice as far from the light,
Us to Calculate
the area over which the light must spread increases by a the Equilibrium
factor of 22 = 2 × 2 = 4. The
photons from the bulb spread Like gravity, light Temperatures of
out over four times as much obeys an inverse
area, so the number of pho- square law. the Planets
tons falling on each square
meter each second becomes 1⁄4 of what it was. If the surface We began our discussion of thermal radiation by asking a
of the sphere is three times as far from the light, as illus- straightforward question: Why does a planet have the tem-
trated in Figure 4.22, the area over which the light must perature that it does? In a qualitative way, we said that the
spread increases by a factor of 3 × 3 = 32 = 9, and the num- temperature of a planet is determined by a balance between
ber of photons per second falling on each square meter the amount of sunlight being absorbed and the amount of
becomes 1⁄9 of what it was originally. We encountered just energy being radiated back into space. We now have the tools
this kind of relationship when we talked about gravity in we need to turn this qualitative idea into a real prediction
Chapter 3. Just like gravity, light obeys an inverse square of the temperatures of the planets.
law. The brightness of the light from an object is inversely Begin with the amount of sunlight being absorbed. The
proportional to the square of the distance from the object. amount of energy absorbed by a planet is just the area of the
Twice as far means one-fourth as bright. planet that is absorbing the energy multiplied by the bright-
It is helpful to think of brightness in terms of photons ness of sunlight at the planet’s distance from the Sun. When
streaming onto a surface from a light because this gives us we look at a planet, we see a circular disk with a radius equal
a good mental picture of the physical nature of brightness to the radius of the planet. The area of this circular disk is
and why brightness follows an inverse square law. In prac- πR2, where R is the radius of the planet. The brightness of
tice, however, it is usually more convenient to speak of the sunlight at a distance d from the Sun is equal to the lumi-
energy coming to a surface each second, rather than the nosity of the Sun (L ⊙, in watts) divided by 4πd2, as we saw
number of photons arriving. in Section 4.6 (d here is the same as r was in Section 4.6;
118 Chapter 4 Light
we use d in this case to avoid confusion with the planet’s The equilibrium temperature of a planet is
radius, R.) We must consider one additional factor. Not all analogous to the water level in Figure 4.19.
of the sunlight falling on a planet is absorbed by the planet.
1 At the planet’s equilibrium temperature, thermal
The fraction of the sunlight that is reflected from a planet energy radiated balances solar energy absorbed,
is called the albedo, a, of the planet. The corresponding so the temperature does not change.
fraction of the sunlight that is absorbed by the planet is 1 Equilibrium Absorbed sunlight is
minus the albedo. A planet with an albedo of 1 reflects all analogous to water
the light falling on it. A planet that absorbs 100 percent of flowing in.
the sunlight falling on it has an albedo of 0.
Temperature
Writing this relationship as an equation, we say that is analogous
to water level.
( ) ( ) × ( of sunlight )/
Energy absorbed Absorbing Fraction
by the planet
each second
= area of
planet
× ( Brightness
of sunlight )
absorbed
Thermal energy
radiated is analogous
to water flowing out
L through the hole.
⊙
____ 2 If the planet is too cold, it absorbs more
= πR2 × × (1 – a) energy than it radiates, and heats up.
4πd2
Too cold
Moving to the other piece of the equilibrium, the amount
of energy that the planet radiates away into space each sec-
ond is just the number of square meters of the planet’s total
surface area multiplied by the power radiated by each square
meter. The surface area for the planet is given by 4πR2. The
Stefan-Boltzmann law tells us that the power radiated by
each square meter is given by σT 4. So we can say that
3 If the planet is too hot, it radiates more
(planet
Energy radiated by
each second
) = ( Surface area
of planet
) × ( square
Energy radiated by each
meter each second
) energy than it absorbs, and cools down.
Too hot
= 4πR 2
× σT .4
If the planet’s temperature is to remain stable—if it is to

keep from heating up or cooling down—then it must be radi-
ating away just as much energy into space as it is absorbing
in the form of sunlight, as indicated in Figure 4.23. There-
fore, we can equate these two expressions. We can set the
quantity “Energy radiated by planet” equal to the quantity
FIGURE 4 . 23 Planets are heated by absorbing sunlight (and
“Energy absorbed by planet.” When we do this, we arrive
sometimes internal heat sources) and cooled by emitting thermal
at the expression
radiation into space. If there are no other sources of heating
( Energy
planet each second )
= ( Energy absorbed by )
radiated by or means of cooling, then the equilibrium between these two
planet each second processes determines the temperature of the planet.
or
L⊙
4πR2σT 4 = πR2 ____ (1 ‒ a).
4πd2 L ⊙/4πd2, tells how much energy is falling on the planet
each second. The fi nal expression, (1 - a), tells how much
Look at this equation for a moment. It may seem rather of that energy the planet actually absorbs. Put everything
complex, but when we consider the pieces of this equa- on the right side of the equation together and you get the
tion separately, it becomes more digestible. On the left amount of energy absorbed by the planet each second. The
side of the equation, 4πR2 tells how many square meters of equal sign says that the energy radiated away needs to
the planet’s surface are radiating energy back into space, balance the sunlight absorbed. There is no magic here. In
while σT 4 tells how much energy each one of those square fact, when broken down, this formidable equation embod-
meters radiates each second. Put them together and you ies little more than a few straightforward ideas—such as
get the total amount of energy radiated away by the planet “hotter means more luminous,” “twice as far means one-
each second. On the right side of the equation, πR2 is the fourth as bright,” and “heating and cooling must balance
area of the planet as seen from the Sun. That amount multi- each other.” The math just gives us a convenient way to
plied by the brightness of the sunlight reaching the planet, work with these concepts.
4.7 Radiation Laws Enable Us to Calculate the Equilibrium Temperatures of the Planets 119
800
= Range of measured temperatures

600
Temperature (K)
= Predicted temperature based on

equilibrium model
400
Earth
Venus Mars
200
Jupiter Saturn Neptune
Mercury Uranus Pluto
0
0.1 1.0 10.0 100.0
Distance from the Sun (AU)
FIGURE 4 . 2 4 Predicted temperatures for the classical planets and Pluto, based on the
equilibrium between absorbed sunlight and thermal radiation into space, are compared with ranges
of observed surface temperatures. Some predictions are correct. Interestingly, others are not.
We started down this path hoping to fi nd a way to pre- of sunlight, and its temperature would rise. Only at this
dict the temperatures of the planets, and a bit of algebra equilibrium temperature do the two balance.
gets us the rest of the way there. Rearranging the previous The equation tells us that as the distance from the Sun
equation to put T on one side and everything else on the increases—in other words, as d gets bigger—the tempera-
other gives ture of the planet decreases. No surprise there. But now we
know how much the temperature should decrease. The tem-
L⊙(1 ‒ a)
T 4 = _______ . perature should be inversely proportional to the square root
16σπd2 of the distance. Using this formula can turn our intuition
If we take the fourth root of each side, we get about why planets that are close to the Sun are hot into a
prediction of just how hot they should be.
[ ]
× _.
1
L⊙(1 ‒ a) ⁄4 ___
1
T = _______ Figure 4.24 plots the predicted temperatures of the plan-
16σπ √ d ets. The vertical bars show the range of temperatures found
Putting in the appropriate numbers, we wind up with on the surface of each planet (or, in the case of the giant
____ 1 planets, at the tops of their clouds). The black dots show
(
√1-a ⁄2
T = 279 × _____ ,
dAU ) our predictions using the preceding equation. As the figure
indicates, overall we are not too far off. That should give
us a sense of accomplishment; it says that our basic under-
where dAU is the distance of the planet from the Sun in astro- standing of why planets have the temperatures that they do
nomical units. For example, we can see that the blackbody is probably fairly close to the mark. The data for Mercury,
(a = 0) temperature for a planet at Earth’s distance from the Mars, and the dwarf planet Pluto agree particularly well.
Sun (1 AU) is 279 K. (The agreement for Mercury would improve if we took into
We have now produced a full-fledged physical model for account the huge difference in temperature between the
why the temperatures of planets are what they are. Restat- daytime and nighttime sides of the planet and recomputed
ing the meaning in words, T our equilibrium accordingly.)
is the temperature at which Balancing cooling In other cases, however, our predictions are wrong. For
the energy radiated by a planet and heating sets Earth and the giant planets the actual temperatures are
exactly balances the energy an equilibrium a bit higher than the predicted temperatures. In the case
absorbed by the planet. If the temperature. of Venus, the actual surface temperature is wildly higher
planet were hotter than this than our prediction. Rather than cause for despair, though,
equilibrium temperature, it would radiate energy away faster these discrepancies between theory and observation are
than it absorbed sunlight, and its temperature would fall. cause for excitement. As we built our physical model for the
If the planet were cooler than this temperature, it would equilibrium temperatures of planets, we made a number of
radiate away less energy than was falling on it in the form assumptions. For example, we assumed that the temperature
120 Chapter 4 Light
of the planet was the same everywhere. This is clearly not incorrect. In other words, the places where the predictions
true; for example, we might expect planets to be hotter on of our theory are not confi rmed by observation point to
the day side than on the night side. We also assumed that a areas where there is something still to be discovered and
planet’s only source of energy is the sunlight falling on it. understood. The question of why these planets are hotter
Finally, we assumed that a planet is able to radiate energy than the prediction will lead us to a number of new and
into space freely as a blackbody. interesting insights into how these planets work. Sci-
The discrepancies between our theory and the mea- entific theories sometimes succeed and sometimes fail,
sured temperatures of some of the planets tell us that for but even when they fail they can teach us a lot about the
these planets, some or all of these assumptions must be universe.
Light may be both an informative messenger and an

Seeing the Forest for the Trees important player in the ebb and flow of the universe, but
the very nature of light itself plays havoc with our com-
In our daily lives, light is the ideal messenger, faithfully monsense ideas about the world. When we explored the
telling us about the world around us. As with any good motions of Earth, the Moon, and the planets, we relied
courier, we usually concentrate on the information that heavily on our intuition about the world. The pull or
light carries, taking the messenger itself for granted. It is shove of one object on another and the force of grav-
easy to forget that our seemingly immediate visual per- ity that holds us tightly to the surface of Earth are well
ception of reality is actually a derived experience—the within the realm of our everyday experience. But when
result of a sophisticated interplay between our eyes, our we consider the properties of light, the boundaries of our
brains, and the flood of electromagnetic radiation that is experience and intuition are shattered. We are forced
emitted, absorbed, transmitted, and reflected by objects beyond the confines of the “box” within which our brains
in the world around us. evolved. Answering a simple question like “How fast does
The light that our eyes see is only a tiny portion of light travel?” demands that we abandon our most cher-
the full span of the electromagnetic spectrum. Electro- ished ideas about the nature of space, time, matter, and
magnetic radiation carries with it a wealth of informa- energy. We ask, “What is light?” and confront abstract
tion about the temperature, density, composition, state concepts like electric and magnetic fields while running
of motion, and other physical characteristics of the place headlong into the seemingly impossible question of how
of its origin and the material it interacts with en route. something can be both a wave and a particle. We delve
In place of the carefully controlled laboratory experi- into the interaction between light and matter and find
ments of many other sciences, the astronomer uses a that at the scale of atoms and photons, Newton’s clock-
combination of ingenuity and technology to “slice up” work universe crumbles. In its place we discover a world
the light reaching us and tease out the information it of random chance and uncertainty, governed not by the
carries about conditions throughout the universe. The strict march of cause and effect but by laws of probability
role of electromagnetic radiation in astronomy is far and statistics. We ask about the nature of light—and we
more than that of messenger. Radiation is also a partici- collide squarely with the shortcomings of our intuitive
pant in the processes that we study. For example, light ideas about the nature of reality itself.
carries energy from the Sun outward through the Solar If electromagnetic radiation is the messenger, how do
System, heating the planets; and light carries energy we hear and interpret the message? In the next chapter we’ll
away from each planet, allowing it to cool. The balance learn about the many and varied tools that astronomers use
between these two processes establishes the conditions to detect and decipher the meaning of these communica-
of our existence. tions that come to us from the Solar System and beyond.
Summary *9. Patterns of emission or absorption lines in spectra can

uniquely identify individual atomic elements, just as
DNA testing uniquely identifies individual human
• Light carries both information and energy throughout
beings. Explain how positive identification of atomic
the universe.
elements can be used as one way of testing the validity
• From gamma rays to visible light to radio waves, all of the cosmological principle.
radiation is an electromagnetic wave.
10. Name the three subatomic particles that make up an
• Light is simultaneously an electromagnetic wave and a atom.
stream of particles called photons.
*11. Why is it impossible to know the exact position and the
• The speed of light in a vacuum is 300,000 km/s, and exact velocity of an electron simultaneously?
nothing can travel faster.
12. Name the two components that together define
• Nearly all matter is composed of atoms, and light can momentum.
reveal its identity.
13. Explain the difference between speed and velocity.
• Atoms absorb and emit radiation at unique wavelengths
like spectral fi ngerprints. 14. Describe the ground state of an atom and what this
implies for any electron in that state.
• Light from receding objects is redshifted. Light from
approaching objects is blueshifted. 15. An atom in an excited state can drop to a lower energy
state by emitting a photon. Is it possible to predict exactly
• Temperature is a measure of the thermal energy of an how long the atom will remain in the higher energy state?
object. Explain your answer.
16. Spectra of astronomical objects show both bright and
dark lines. Describe two properties that each is telling
SmartWork, Norton’s online homework system, you about the atoms responsible for the spectral lines.
questions, plus additional conceptual exercises. If 17. Astronomers may describe certain celestial objects as
your instructor assigns questions in SmartWork, log in at being redshifted or blueshifted. What do these terms tell
smartwork.wwnorton.com. them about the objects?
18. An object somewhere near you is emitting a pure tone at
middle C on the octave scale (262 Hz). You, having per-
Student Questions fect pitch, hear the tone as A above middle C (440 Hz).
Describe the motion of this object relative to where you
THINK ING A BOUT THE CONCEP T S are standing.
1. We know that the speed of light in a vacuum is 3 × 108 19. During a popular art exhibition, the museum staff fi nds
m/s. Is it possible for light to travel at a lower speed? it necessary to protect the artwork by limiting the total
Explain your answer. number of viewers in the museum at any time. New
viewers are admitted at the same rate that others leave.
2. Is a light-year a measure of time or distance or both?
Is this an example of static equilibrium or of dynamic
3. Would you describe light as a wave or a particle or both? equilibrium? Explain.
Explain your answer.
*20. The temperature of an object has a very specific mean-
*4. In what way can a charged particle create a magnetic ing as it relates to the object’s atoms. Explain the
field? relationship.
5. Name the variables that defi ne a wave. 21. A favorite object for amateur astronomers is the double
star Albireo, with one of its components a golden yellow
6. Name the types of electromagnetic radiation that lie at
and the other a bright blue. What do these colors tell you
the extreme ends of the electromagnetic spectrum.
about the relative temperatures of the two stars?
7. If photons of blue light have more energy than photons
22. The difference between brightness and luminosity con-
of red light, how can the intensity of a beam of red light
fuses many people. How should you explain the differ-
carry as much energy as a beam of blue light?
ence to a family member or a friend who is not taking
8. What is a continuous spectrum? this class?
122 Chapter 4 Light
*23. The stars you see in the night sky cover a large range of 34. The spacecraft Voyager 1 is now about 115 AU from the
brightness. What does that tell you about the distances Sun and heading out of the Solar System. Compare the
of the various stars? Explain your answer. brightness of the Sun seen by Voyager 1 with that seen
from Earth.
**24. The physical properties of an object are sometimes pro-
portional to its temperature, raised to some power. For 35. On a dark night you notice that a distant lightbulb hap-
example, as we have seen in this chapter, the luminos- pens to have the same brightness as a fi refly that is
ity of a radiating body is proportional to T 4. Why must 5 meters away from you. If the lightbulb is a million
the temperature T be expressed in kelvins rather than times more luminous than the fi refly, how far away is
degrees Celsius or Fahrenheit? the lightbulb?
25. Consider two hypothetical planets with no atmosphere. 36. Two stars appear to have the same brightness, but one star
One orbits the Sun at an average distance of 5.0 AU; the is three times more distant than the other. How much
other, at an average distance of 10.0 AU. Yet both have more luminous is the more distant star?
the same average surface temperature. Explain how this
37. A panel with an area of 1 square meter (m2) is heated to
could be possible.
a temperature of 500 K. How many watts is it radiating
into its surroundings?
38. The Sun has a radius of 6.96 × 105 km and a blackbody
temperature of 5,780 K. Calculate the Sun’s luminosity.
26. The index of refraction, n, of a diamond is 2.4. What is (Hint: The area of a sphere is 4πR2.)
the speed of light within a diamond?
39. You are given a bar of tungsten, a metal with a melt-
27. You are tuned to 790 on AM radio. This station is broad- ing point of 3,640 K and a boiling point of 6,170 K. You
casting at a frequency of 790 kHz (7.90 × 105 Hz). What heat the bar until it starts to melt and plot a graph of its
is the wavelength of the radio signal? You switch to 98.3 emitted Planck energy distribution. You then continue
on FM radio. This station is broadcasting at a frequency to heat the tungsten until it begins to boil, and you plot
of 98.3 MHz (9.83 × 107 Hz). What is the wavelength of a similar graph.
this radio signal? a. Sketch and label these graphs on the same set of
axes. At what wavelength does the spectrum of each
28. Your microwave oven cooks by agitating water molecules
peak?
at a frequency of 2.45 gigahertz (GHz), or 2.45 × 109 Hz.
b. How many times more luminous is the boiling tung-
What is the wavelength of the microwave’s electromag-
sten than the melting tungsten?
netic radiation in centimeters?
40. Some of the hottest stars known have a blackbody tem-
29. You observe a spectral line of hydrogen at a wavelength
perature of 100,000 K.
of 502.3 nm in a distant galaxy. The rest wavelength of
a. What is the peak wavelength of their radiation?
this line is 486.1 nm. What is the radial velocity of this
b. Can this wavelength be observed from Earth’s surface?
galaxy?
30. Assume that an object emitting a pure tone of 440 Hz is
41. Your body, at a temperature of about 37°C (98.6°F), emits
on a vehicle approaching you at a speed of 25 m/s. If the
radiation in the infrared region of the spectrum.
speed of sound at this particular atmospheric tempera-
a. What is the peak wavelength, in micrometers, of your
ture and pressure is 340 m/s, what will be the frequency
emitted radiation?
of the sound that you hear? (Hint: Keep in mind that
b. Assuming an exposed body surface area of 0.25 m2,
frequency is inversely proportional to wavelength.)
how many watts of power do you radiate?
31. If half of the phosphorescent atoms in a “glow-in-the-
*42. A planet with no atmosphere at 1 AU from the Sun would
dark” toy give up a photon every 30 minutes, how bright
have an average blackbody surface temperature of 279 K
(relative to its original brightness) will the toy be after
if it absorbed all the Sun’s electromagnetic energy fall-
2 hours?
ing on it (albedo = 0).
32. Assume that the atoms in a gas remain in an excited a. What would be the average temperature on this
state for an average lifetime of 1 minute before decaying planet if its albedo were 0.1, typical of a rock-covered
to the ground state. How many minutes will pass until surface?
only 1⁄32 of the atoms remain in the excited state? b. What would be the average temperature if its albedo
were 0.9, typical of a snow-covered surface?
33. Imagine that you have been transported to Neptune, 30
AU from the Sun. How bright would the Sun appear, 43. Consider a hypothetical planet named Vulcan. If such a
compared to its brightness as seen from Earth? planet were in an orbit 1⁄4 the size of Mercury’s and had
the same albedo as Mercury, what would be the average albedo of 0.8, what is the average temperature of Eris
temperature on Vulcan’s surface? Assume that the aver- when it is farthest from the Sun?
age temperature on Mercury’s surface is 450 K.
**44. Earth’s average albedo is approximately 0.3.
a. Calculate Earth’s average temperature in degrees
Celsius.
ing outlines, vocabulary flashcards, and multiple-choice
b. Does this temperature meet your expectations?
quizzes, plus links to premium content in SmartWork and
Explain why or why not.
the ebook. Visit www.wwnorton.com/studyspace.
45. The orbit of Eris, a dwarf planet, carries it out to a maxi-
mum distance of 97.7 AU from the Sun. Assuming an
All truths are easy to understand once they are
discovered; the point is to discover them.
GALILEO GALILEI (1564 –1642)
Robotic rovers Spirit and Opportunity have roamed the surface of Mars.
CHAP TER 5
The Tools of the Astronomer

5.1 The Optical Telescope—
K E Y CONCEP T S
An Extension In the previous chapter we learned how our
of Our Eyes understanding of the physical and chemical properties
of distant planets, stars, and galaxies comes to us
Do you remember the fi rst time you had a close-up view of in the form of electromagnetic radiation. But this
the Moon’s surface? Perhaps your family or a neighbor had information must first be collected and processed
a backyard telescope like the one shown in Figure 5.1. Or before it can be analyzed and converted to useful
the Moon might have been the featured attraction during a knowledge. Here we will learn about the tools that
visit to your local planetarium. With that fi rst view came
astronomers use to capture and scrutinize that
information. We will find that
• Telescopes of various types collect radiation over the
FIGURE 5.1 Amateur astronomers with a modern reflecting
telescope. There are several hundred thousand amateur entire range of the electromagnetic spectrum—from
astronomers in the United States alone. Advances in telescope gamma rays to radio signals.
design and electronics have enabled amateur observers to • Optical telescopes come in two basic types: refractors
discover new celestial objects and help with long-term research
and reflectors. All of the larger astronomical
projects of professional astronomers.
telescopes are reflectors.
• Telescope resolution increases with aperture; image
size is proportional to a telescope’s focal length.
• Earth’s atmosphere distorts telescopic images and
prevents large parts of the electromagnetic spectrum
from reaching the ground. Telescopes in orbit
overcome this problem.
• The most effective way to study the planets and moons
of our Solar System is to go there.
125
126 Chapter 5 The Tools of the Astronomer
FIGURE 5. 2 A replica of Galileo’s refracting telescope. FIGURE 5. 3 Newton’s reflecting telescope.
recognition of what the Moon really is—a nearby planetary is Latin for “lentil.” Looking back, it’s rather remarkable that
world covered with craters and vast, lava-flooded basins. more than 300 years would pass before these lenses would
You might also have been treated to a breathtaking look at be employed for something other than spectacles!
the Orion Nebula, a giant Hans Lippershey (1570–1619) was a German-born spec-
assemblage of gas and dust The telescope is the tacle maker living in the Netherlands around the turn of
1,500 light-years away, or the astronomer’s most the 17th century. Legend has it that in 1608, children play-
larger and more remote important tool. ing with his lenses put two of them together and saw a dis-
Andromeda Galaxy. These are tant object magnified. Lippershey looked for himself and
but a few of many celestial wonders that come alive in the mounted the lenses together in a tube to produce a kijker
eyepiece of even a small telescope. As it has for so many (“looker”), as he called it. As you might imagine, news of
before you, the telescope can change the way you view the Lippershey’s invention spread rapidly, soon reaching the
heavens. From such experiences you might understand why Italian instrument maker Galileo Galilei. Galileo at once saw
the telescope is the astronomer’s most important instru- the potential of the “looker” for studying the heavens and
ment. Yet it is only within the past century and a half that constructed one of his own (Figure 5.2). By 1610, Galileo had
its capabilities have been fully exploited. become the fi rst to see the phases of Venus and the moons
As long ago as the late 13th century, craftsmen in Venice of Jupiter, and among the fi rst to see craters on the Moon.1
were making small disks of glass that could be mounted in He was also the fi rst to realize that the Milky Way is made
frames and worn over the eyes to improve vision. The glass
disks were convex on both sides, shaped something like len- 1
As we mentioned in Chapter 3, English astronomer Thomas Harriot
tils. And so they became known as lenses, from lens, which made drawings of the Moon several months before Galileo began his
(a) (b)
Yerkes: 1-meter-diameter lens Keck: 10-meter-diameter mirror

World’s largest refracting telescope One of the world’s largest
reflecting telescopes
FIGURE 5. 4 (a) The Yerkes 1-meter refractor uses a lens to collect light. (b) The twin Keck
10-meter reflectors are more compact and use mirrors to collect light.
up of countless numbers of individual stars. As the story fabricate. Not until the latter half of the 18th century did
goes, Galileo was demonstrating his instrument to guests large reflecting telescopes come into their own.
when one of them christened it the telescope, from the Greek Throughout the 19th century, both refracting and reflect-
meaning “farseeing,” and the name stuck. With its ability ing telescopes continued to grow in size. By 1897, though,
to see far beyond the range of the human eye, the refracting refracting telescopes had reached their limit, with the com-
telescope—one that uses lenses instead of mirrors—quickly pletion of the Yerkes 1-meter
revolutionized the science of astronomy. (40-inch) refractor (Figure All of the world’s
Unfortunately, all simple-lens telescopes (those using 5.4a), destined to become the largest telescopes are
a pair of single lenses) suffer from a serious problem world’s largest operational reflecting telescopes.
called chromatic aberration, refracting telescope. Located
which forms blue halos There are two types of in Williams Bay, Wisconsin, the Yerkes telescope carries a
around bright objects. As optical telescopes: 450-kilogram (kg) objective lens mounted at the end of a
light passes through a simple refractors and 19.2-meter tube. But herein lies a major problem with refrac-
lens, blue light is brought to reflectors. tors: to get the most light-gathering power and produce the
a shorter focus than the lon- largest images, a refractor must suspend a massive piece of
ger visible wavelengths, causing the out-of-focus blue glass at the end of a very long tube without sagging unduly
halo effect. Realizing this problem, Sir Isaac Newton in under the force of gravity. Reflecting telescopes, on the other
1668 designed a telescope using mirrors instead of lenses. hand, seem to have no such size limits. Today the 10-meter
He cast a 2-inch mirror made of speculum metal (basi- twin Keck telescopes located on 4-kilometer (km)-high
cally copper and tin) and polished it to spherical curva- Mauna Kea in Hawaii (Figure 5.4b) are among the world’s
ture. He then placed the primary mirror at the bottom of largest reflecting telescopes. Each of the two Keck telescopes
a tube with a secondary fl at mirror mounted above it at has 4 million times the light-gathering power of the human
a 45° angle, which directed the focused light to an eye- eye! And even larger reflecting telescopes are in the works.
piece on the outside of the tube (Figure 5.3). As it turned Several organizations are considering telescopes with aper-
out, others did not share Newton’s talent for instrument tures in the 30-meter range, and the European Southern
making, and the reflecting telescope remained a curios- Observatory’s (ESO) 42-meter European Extremely Large
ity for decades. The spherical mirror surface used by Telescope (E-ELT) may be in operation by 2017, at an esti-
Newton works only for small mirrors. Larger mirrors mated cost of nearly €1 billion (Figure 5.5). It seems that the
require a parabolic surface to produce a sharply focused only limitation on the size of reflecting telescopes is the
image, and parabolic surfaces are much more difficult to cost of fabricating them. Table 5.1 lists the world’s largest
optical telescopes. As you can see, all are reflecting
telescopes.
astronomical observations. However, Galileo published his observa- Here is a professional secret we can share with you.
tions, while Harriot did not. Today’s working astronomers never—well, hardly ever—
look through the eyepiece of a telescope. Why ignore such

an opportunity? The reason is that they cannot afford to
waste valuable observing time. Telescope time can be very
expensive: Operating a large telescope for just a single night
can cost tens of thousands of dollars! So although it might
be exhilarating to glimpse Saturn through the eyepiece of a
really big telescope, astronomers can learn much more and
make better use of precious observing time by permanently
recording the planet’s image at a variety of wavelengths
or seeing its light spread out into a revealing spectrum.
Long after the observing session is over, the wistful peek
through the eyepiece would be just a distant memory, but
the recorded data remain as a permanent quantitative record
for subsequent analysis.
As important as telescopes are, they are not the only
instruments in the astronomer’s bag of gear. The tools
FIGURE 5. 5 Artist’s rendering of the European Extremely
used by modern astronomers are many and remarkably
Large Telescope (E-ELT), a 42-meter reflecting telescope diverse, ranging from powerful supercomputers to high-
proposed by the European Southern Observatory (ESO).
TA B L E 5 .1
The World’s Largest Optical Telescopes
Mirror
Diameter Operational
(meters) Telescope Sponsor Location Date
11.0 South African Large South Africa, USA, UK, Germany, Poland, New Sutherland, South Africa 2005
Telescope (SALT) Zealand
10.4 Gran Telescopio Spain, Mexico, University of Florida Canary Islands 2007
CANARIAS (GTC)
10 Keck I Caltech, University of California, NASA Mauna Kea, Hawaii 1993
10 Keck II Caltech, University of California, NASA Mauna Kea, Hawaii 1996
9.2 Hobby-Eberly University of Texas, Penn State, Stanford, Germany Mount Fowlkes, Texas 1999
2 × 8.4 Large Binocular University of Arizona, Ohio State, Italy, Germany, Mt. Graham, Arizona 2008
Telescope (LBT) Arizona State, and others
8.3 Subaru Telescope Japan Mauna Kea, Hawaii 1999
4 × 8.2 Very Large Telescope European Southern Observatory Cerro Paranal, Chile 2000
(VLT)
8.1 Gemini North USA, UK, Canada, Chile, Brazil, Argentina, Mauna Kea, Hawaii 1999
Australia
8.1 Gemini South USA, UK, Canada, Chile, Brazil, Argentina, Cerro Pachón, Chile 2000
Australia
6.5 Magellan I Carnegie Institute, University of Arizona, La Serena, Chile 2000

Harvard, University of Michigan, MIT
6.5 Magellan II Carnegie Institute, University of Arizona, La Serena, Chile 2002

Harvard, University of Michigan, MIT
FIGURE 5.6 A schematic

view of the human eye.
energy colliders to robotic probes sent to cruise the Solar (Figure 5.6). Although the lens forms an upside-down image
System. Astronomy’s tools are not only diverse; they are of an object on the retina, your brain translates the signal
also changing rapidly. It seems that every few months from the optic nerve to show you the correct orientation
we are likely to see the commissioning of a new moun- of the object. Evolutionary biologists have concluded that
taintop telescope, the launch of a satellite observatory, eyes—as crucial sensory organs—evolved independently in
or the arrival of a spacecraft at a remote planetary des- as many as 65 different species during the history of terres-
tination. In fact, in the time it takes between the fi nal trial life! Although the human eye is wonderfully evolved
changes we make to this volume and its appearance on to meet our daily needs, it is not so well suited for doing
your bookshelf, much of what we might say about the lat- astronomy. For that we need large, powerful telescopes. Now
est astronomy tools will already have become yesterday’s let’s look more closely at how telescopes mimic the eye’s
news. Rather than give in to instant obsolescence, we will light-gathering function and how they help us overcome
instead make use of the technology of the 21st century the eye’s limitations.
to bring you a discussion of (what else?) the technology As we have seen, astronomical telescopes come in two
of the 21st century! Keep up to date by visiting the 21st basic types: refracting and reflecting. A refracting telescope
Century Astronomy Web site. has a simple convex lens,
called the objective lens, A telescope’s aperture
whose cu r ved su rfaces determines its
refract—changes the direc- light-gathering power
Refractors and Reflectors tion of—the light from a dis- and resolution.
Humanity’s use of telescopes dates back to far earlier than tant object (Figure 5.7).
the night Galileo fi rst turned his telescope skyward. Our (Foundations 5.1 explores the optics of refracted light.) This
earliest human ancestors had telescopes of their own—their refracted light forms an image on the telescope’s focal plane,
eyes. The human eye essentially operates as a refracting tele- which is perpendicular to the optical axis of the telescope.
scope: it uses a simple convex lens to bend the light passing (The “optical axis” is the path that light takes through the
through it, bringing that light to a sharp focus on the retina center of the lens, or mirror in the case of a reflecting
(a)
Light from star 2
FIGURE 5.7 (a) A refracting telescope uses
Focal plane
a lens to collect and focus light from two stars,
Lens
Image of star 1 forming images of the stars in its focal plane. (b)
Telescopes with longer focal length produce larger,
Image of star 2
more widely separated images.
Light from star 1
Focal length
(b) Focal plane
Focal length
When Light Doesn’t Go Straight

REFRACTION FIGURE 5.8 (a) Light
waves are refracted
As its name suggests, the refracting telescope focuses and (bent) when entering a
enhances the light from a distant object through the process of medium with a higher
refraction. Refraction is the change in direction of light enter- index of refraction. They
ing a new medium. When a light wave enters a new medium, are refracted again as
its speed changes. Remember from Chapter 4 that the speed they reenter the medium
of light is always less in any material medium than it is in a with the lower index of
refraction. (b) Light from
vacuum. If the incident light is perpendicular to the surface of
a green laser beam is
the new medium, the speed changes but the direction of travel
refracted as it enters and
does not. If the incident light encounters the surface at a dif-
exits a plastic block.
ferent angle, the direction of travel also changes. This change
in the direction of the incident light is called refraction. The
amount of refraction depends both on the angle of incidence
and the relative speeds in the two media, which are defi ned by
their respective indices of refraction (see Chapter 4). In Figure (a)
5.8a, light waves are depicted as coming in from the upper left
and hitting the surface of the new medium at a certain angle
to the perpendicular. The part of each wave that enters the
new medium fi rst is slowed before the other part of the wave
enters. As a result, the wave bends, taking up a new direction of
travel. If the speed of light in the new medium is less than that
in the initial medium (that is, if the new medium has a higher
refractive index), the light bends toward the perpendicular. If
the speed of light in the new medium is greater than that in the
initial medium (that is, if the new medium has a lower refractive
index), the light bends away from the perpendicular. Figure 5.8b
shows a green laser beam being refracted by a plastic block.
REFLECTION
Picture a beam of light striking a piece of glass, as illustrated
in Figure 5.9. When light encounters a different medium—in
this case going from air to glass—there will always2 be a certain
(b)
amount of reflection from the surface of the new medium. In
other words, some of the light will change its direction of travel.
The most common example occurs when light encounters an
ordinary flat mirror. In Figure 5.9a, an incoming or incident That’s what makes a flat mirror so convenient for admiring
ray, AB, reflects from the surface, becoming the reflected ray our appearance.
BC. The angle between AB and PB, the perpendicular to the Curved mirrors can also be very useful, especially in astro-
surface, is called the “angle of incidence” (i). The angle between nomical telescopes. The same rules of incidence and reflection
BC and PB is called the “angle of reflection” (r). In the case of hold here for each ray, but in this case the reflected rays do not
a flat mirror, the angles of incidence and reflection are always maintain the same angle with respect to each other as they do
equal. What reflects from the mirror is a good representation with a flat mirror. A mirror that is concave toward the incom-
of what falls on it, although left and right are interchanged. ing light and has a parabolic surface (Figure 5.10) will reflect
the rays so that they converge to form an image. If the incom-
2
We have to be careful here. In those rare cases in which the index ing rays are parallel, as from a distant source—think “star”—
of refraction (see Chapter 4) is exactly the same in both media, there the reflected rays cross at a distance from the mirror called the
will be no reflection or refraction at the surface. focal length of the mirror. Parallel rays from a distant source on
(a) A P C DISPERSION
i r Shine white light through a glass prism, as shown in Figure
5.11, and you’ll get a rainbowlike spectrum. This effect dem-
onstrates that the refraction, and therefore the speed of light
in glass, depends on wavelength. Like most other transpar-
ent materials, glass has a refractive index (see Chapter 4) that
increases with decreasing wavelength. This means that shorter
wavelengths (those toward the blue) are refracted more strongly
B
than longer wavelengths (those toward the red). This wave-
length-dependent difference in refraction, which spreads the
(b) white light out into its spectral colors, is what we call disper-
sion. Although dispersion is helpful in producing prism spec-
tra, it creates a serious problem called chromatic aberration
(Figure 5.12a). Chromatic aberration causes blue light to come
to a shorter focus than the longer visible wavelengths. You can
see this effect when you look at a bright object such as a distant
streetlight through an inexpensive telescope. (Low-priced tele-
scopes usually have only a simple convex objective lens.) The
streetlight will appear to be surrounded by a blue halo, because
the blue component of the light is out of focus. Manufacturers
of quality cameras and telescopes avoid the use of simple con-
FIGURE 5.9 (a) A ray of incident light (AB) shines on a
vex lenses in favor of the compound lens, like the one shown
flat surface. The surface changes the direction of light by in Figure 5.12b. By using two types of glass, a compound lens
refraction and reflection. The angle of incidence (i) equals the corrects for chromatic aberration.
angle of reflection (r). (b) Light from a laser beam is reflected
from a flat glass surface.
FIGURE 5.11 White light is dispersed into its component

colors as it passes through a glass prism.
Focal
plane
Focal
length
FIGURE 5.10 Parallel

rays of light incident on a
concave parabolic mirror
are brought to a focus in
the mirror’s focal plane.
the axis come together on the axis at a point called the focus.
The surface at which all parallel rays cross is called the focal
plane (see Figure 5.10).
Continued on next page

(a) light will be bright.3 We call this constructive interference. If,

on the other hand, one wave is at its maximum positive value
and the other is at its maximum negative value, the sum is zero
and the result will be darkness. We call this destructive inter-
ference. When we replace the two slits with a large number
Optical of very narrow, very closely spaced parallel slits, we call the
axis result a transmission grating. We can achieve the same effect
by engraving closely spaced lines on a mirror.
(b)
If we now substitute a multiwavelength source of light, we get a
Crown Flint similar pattern for each and every wavelength, as shown in Figure
5.13b for a reflection grating. For each wavelength, constructive
interference takes place at a different point on the screen. In other
words, the grating produces a spectrum. Modern spectrographs
use a grating to disperse incoming light into its constituent wave-
Optical lengths. You can see this effect for yourself. Look at light reflected
axis
from a CD or DVD. The closely spaced tracks act as a grating and
create a respectable spectrum (Figure 5.13c).
FIGURE 5.12 (a) Light of different wavelengths (different When light passes through a small opening (or near an
colors) comes to different foci along the optical axis of a opaque edge), the effects of interference come into play. This
simple lens, causing chromatic aberration. (b) A compound
is called diffraction. Figure 5.14a shows what happens when
lens using two types of glass (crown and flint) with different
monochromatic light from a distant source passes through a
indices of refraction can compensate for much of the
lens and is brought to a focus. The image pattern is not a single
chromatic aberration.
point as we might expect, but rather is smeared out into a series
of bright and dark concentric rings (as seen in Figures 5.14b
and c), representing constructive and destructive interference
from the edge of the aperture. If we make the aperture smaller,
INTERFERENCE AND DIFFRACTION the diffracted image grows larger, causing more blurring and
The intersection of two sets of electromagnetic waves can pro- limiting how close two images can be and still be resolved.
duce patterns of high and low intensity called interference The angular size of the diffraction pattern depends on the ratio
patterns. Say we have a pair of slits and an opaque screen. between the wavelength of light (λ) and the aperture of the hole
Figure 5.13a illustrates monochromatic light (light having a (D): λ/D. At any given wavelength, larger apertures produce
single wavelength or a very narrow range of wavelengths) going smaller diffraction patterns and therefore higher resolution.
through the slits. Each slit now becomes a source of wavefronts. And at any given aperture, shorter wavelengths produce smaller
Notice the regular pattern on the screen where the wavefronts diffraction patterns and higher resolution.
from the two slits intersect. If the intersection point occurs 3
Don’t be concerned about the negative value of the wave’s ampli-
where the amplitudes of both waves are at their maximum posi- tude. The intensity of light is actually equal to the square of the
tive or maximum negative value, the two add together and the amplitude, so the negative sign goes away.
telescope.) The diameter of the objective lens is known as ture determines a telescope’s light-collecting power, and
the telescope’s aperture, and it determines the light-gath- the focal length establishes the size of the image.
ering power of the lens. Note that the light-gathering power We have already seen that there are physical limits
of a telescope is proportional to the area of its aperture— on the size, and thus effectiveness, of refracting tele-
that is, to the square of its diameter. The distance between scopes. Early astronomers were also well aware of another
the telescope lens and the images formed is referred to as serious drawback to simple-lens refracting telescopes:
the focal length of the telescope. Note that longer focal chromatic aberration (see Foundations 5.1). To minimize
lengths increase the size and separation of objects in the chromatic aberration, refracting telescopes now use com-
focal plane (see Figure 5.7). Aperture and focal length are pound lenses, which leave only small, residual effects of
the two most important parameters of a telescope. The aper- the blue halo.
(a) (b) (c)
FIGURE 5.13 (a) Constructive and destructive interference patterns are created when
monochromatic light passes through a pair of narrow slits. (b) Spectral dispersion is produced by
interference when multiwavelength (white) light reflects from a grating. (c) A spectrum is created by
the reflection of light from the closely spaced tracks of a CD.
(a) Diffraction due to the wave nature (b) (c)

of light limits the sharpness of a
telescope’s images.
Lens
Light from star
Diffraction from
Wave crests a simple lens
FIGURE 5.14 (a) Light waves from a star are diffracted by the edges of a telescope’s lens or
mirror. (b) This diffraction causes the stellar image to be blurred, limiting a telescope’s ability to
resolve objects. (c) Diffraction from a circular aperture illuminated by green laser light.
Reflecting telescopes deal with chromatic aberration ary mirror (Figure 5.15). The secondary mirror allows a
and other limitations of refracting in different ways. (See significant reduction in the length and weight of the tele-
Foundations 5.1 to learn more about the optics of reflected scope. For example, the effective focal length of each of
light.) A reflecting telescope forms an image in its focal the 10-meter Keck telescopes can be as long as 250 meters,
plane when light is reflected from a specially curved mir- even though the overall length of each telescope is a mere
ror rather than from the refracting telescope’s convex lens. 25 meters.
We call this mirror the primary mirror to distinguish it Reflectors have a number of important advantages over
from the additional mirrors usually employed in modern refractors. Because the direction of a reflected ray does not
reflecting telescopes. We can fold the light path from the depend on the wavelength of light, chromatic aberration
primary mirror to the focal plane by introducing a second- is no longer a problem. Primary mirrors can be supported
Optical and Atmospheric Limitations

One of the eye’s major limitations as an astronomical tele-
scope is resolution. When we speak of resolution, we are
referring to how close two points of light can be to each other
before a telescope is no longer able to split the light into two
separate images. Unaided, the human eye can resolve objects
separated by an angular distance of 1 arcminute,24 or 1⁄30 the
Secondary
diameter of the full Moon.35 This angular size may seem small,
mirror
Primary and in our daily lives it is; yet when we look at the sky, thou-
mirror sands of stars and galaxies may hide within the smallest area
FIGURE 5.15 Reflecting telescopes use mirrors to collect and
that the unaided human eye can resolve. Figure 5.7a shows
focus light. Large telescopes typically use a secondary mirror the path followed by rays of light from two distant stars as
that directs the light back through a hole in the primary mirror to they pass through the lens of a refracting telescope. Com-
an accessible focal plane behind the primary mirror. parison with Figure 5.7b illustrates that the longer the focal
length is, the greater is the separation between the images.
from the back, and can be made thinner and therefore less 4
2
A description of angular units—radians, degrees, arcminutes, and
massive than objective lenses. The lesser mass of a mirror arcseconds—can be found in Chapter 13 (Section 13.2) and in Appen-
becomes a distinct advantage when we are dealing with dix 1.
astronomers’ insatiable appetite for ever-larger telescopes. 5
3
Only the sharpest of human eyes achieve a resolution of 1 arcminute.
AstroTour: Geometric Optics and Lenses Typical resolution for many of us would be more like 2 arcminutes.
Diffraction Limit
The ultimate limit on the angular resolution of a telescope, called Let’s look at the human eye. The size of the human pupil
the diffraction limit, is determined by the ratio of the wavelength (see Figure 5.6) can range from about 2 mm in bright light to 8
of light passing through it to the diameter of the lens: mm in the dark. A typical pupil size is about 4 mm, or 0.004
meter. If we substitute this value for D and again assume vis-
λ
()
θ = 2.06 × 105 __ arcseconds,
D
ible light, we have
where θ is the diffraction-limited angular resolution in arc-

seconds, λ is the wavelength of light, and D is the diameter,
(
5.5 × 10–7
)
θ = 2.06 × 105 ________ = 28.3 arcseconds,
0.004
or aperture, of the telescope.6 Both λ and D are expressed in or about 0.5 arcminute. But, as we learned earlier, the best reso-
the same units, usually meters. As we can see, the smaller the lution that the human eye can achieve is about 1 arcminute, and
ratio of λ/D, the better will be the resolution of the telescope. more typically it is 2 arcminutes. Why don’t we achieve the the-
We can apply this relationship to the Hubble Space Telescope oretical resolution with our eyes? The reason is that the optical
(HST) operating in the visible part of the spectrum. The HST’s properties of the lens and the physical properties of the retina
primary mirror has a diameter (D) of 2.4 meters. Visible (green) (see Figure 5.6) that nature has given us are not perfect.
light has a wavelength (λ) of 550 nanometers (nm), or 5.5 × 10 –7 Electron microscopes take advantage of this property of dif-
meters. Substituting into the previous equation, we have fraction to achieve very high resolution by using (what else?)
)
electrons instead of photons to illuminate the target. Recall the
(
5.5 × 10–7
θ = 2.06 × 105 ________ = 0.047 arcsecond,
2.4
dual wave-particle nature of electromagnetic radiation, which
we discussed in Chapter 4. Electrons can behave both as par-
or about 1,000 times better than the resolving power of the ticles and as waves, with wavelengths shorter than 0.1 nm. This
human eye. means that electron microscopes have more than five thousand
6
times better resolution than conventional microscopes, which
The constant of proportionality here is simply the number of arc-
use visible light ( λ ≈ 550 nm).
seconds in a radian.
The focal length of a human eye is typically about 20 mil- 5.14a. Light from a distant star arrives at the top of Earth’s
limeters (mm). In comparison, telescopes used by profes- atmosphere as a series of flat, parallel waves called a wave-
sional astronomers often have focal lengths of tens or even front. If Earth’s atmosphere were perfectly homogeneous,
hundreds of meters. Such telescopes make images that are the wavefront would remain flat as it reaches the objective
far larger than those formed by your eye, and consequently lens or primary mirror of a groundbased telescope. After
they contain far more detail. making its way through the telescope’s optical system, the
Focal length explains only one difference between the wavefront would produce a tiny diffraction disk in the focal
resolution of telescopes and the unaided eye. The other dif- plane, as shown in Figure 5.14b. But Earth’s atmosphere is
ference results from the wave nature of light. As light waves not homogeneous. It is filled with small bubbles of air that
pass through the lens of a telescope, they spread out from have slightly different temperatures than their surround-
the edges of the lens (see Figure 5.14). The distortion of the ings. Different temperatures mean different densities, and
wavefront as it passes the edge different densities mean different refractive properties.
of an opaque object is called Diffraction, or blurring These air bubbles act as weak lenses, and by the time the
diffraction (see Foundations of an image, depends wavefront reaches the telescope it is far from flat, as shown
5.1). Diffraction “diverts” on the ratio of in Figure 5.16. Instead of a tiny diffraction disk, the image
some of the light from its path, wavelength to
slightly blurring the image telescope aperture.
made by the telescope. The
degree of blurring depends on the wavelength of the light FIGURE 5.16 Bubbles of warmer or cooler air in Earth’s
atmosphere distort the wavefront of light from a distant object.
compared with the diameter of the telescope lens. The larger
the lens relative to the wavelength of the light it is focusing, Light from star Wave crests
the smaller the problem posed by diffraction. The limiting 1 As light approaches Earth,
it is traveling in waves with
resolution that a given telescope can achieve is known as parallel wave crests.
the diffraction limit (see Math Tools 5.1).
The diffraction limit tells us that larger telescopes get
better resolution. Theoretically, the 10-meter Keck telescopes
2 Upon entering Earth’s
have a diffraction-limited resolution of 0.0113 arcsecond in
atmosphere, it encounters
visible light, which would allow you to read newspaper bubbles of warm or cool air…
headlines 60 km away. But for telescopes with apertures
larger than about a meter, Earth’s atmosphere stands in the
way of better resolution. If you have ever looked out across
3 …which refract and distort
the desert on a summer day, you have seen the distant hori- the waves so that the crests
zon shimmer as light from that horizon is constantly bent are no longer parallel.
this way and that by turbulent bubbles of warm air rising
off the hot desert floor.
The problem is less pronounced when we look overhead,
4 Other bubbles create
but the twinkling of stars in the night sky tells us the phe- more distortion…
nomenon is still there. As
telescopes magnify the angu- Earth’s atmosphere
lar diameter of a planet, they distorts images.
also magnify the shimmering 5 …and so the wave
crests of light reaching
effects of the atmosphere. The limit on the resolution of a a telescope on Earth
telescope on the surface of Earth caused by this atmospheric are distorted.
distortion is called astronomical seeing. One advantage of
launching telescopes such as the Hubble Space Telescope
into orbit around Earth is that from their vantage point
above the atmosphere, telescopes get a much clearer view
of the universe, unhampered by the astronomical seeing
phenomenon. Does that mean that groundbased telescopes
are becoming obsolete? Not at all. Modern technology has
come to their rescue with computer-controlled adaptive
optics, which compensate for much of the atmosphere’s
distortion.
To better understand how adaptive optics work, we need
to look more closely at how Earth’s atmosphere smears out
an otherwise perfect stellar image. Look again at Figure Telescope
(a) (b)
FIGURE 5.17 Images of
Neptune and a star cluster
taken without (a) and with (b)
adaptive optics.
in the telescope’s focal plane is distorted and swollen, of groundbased telescopes competitive with those of the
degrading the resolution. Now suppose we could measure Hubble Space Telescope. But image distortion is not the only
the amount of distortion in the problem caused by Earth’s atmosphere. Large regions of the
wavefront and somehow flat- Adaptive optics can electromagnetic spectrum are partially or completely
ten it out. This is how adaptive correct for absorbed by various atmospheric molecules.
optics work. First, an optical atmospheric distortion Another limitation on the human eye is that it is sensi-
device within the telescope of telescopic images. tive only to light in the visible part of the electromagnetic
constantly samples the wave- spectrum. (That is, after all,
front, measuring its departure from flatness. Then, before why we call it the “visible” Earth’s atmosphere
reaching the telescope’s focal plane, light is reflected from part of the spectrum!) Even blocks much of the
yet another mirror that has a deformable surface. (Astrono- though visible light is only a electromagnetic
mers sometimes call this a “rubber” mirror, although it is small part of the electromag- spectrum.
actually made of glass.) A computer analyzes the wavefront netic spectrum, it is anything
distortion and sends a signal to mechanisms that bend the but happenstance that our eyes work in this range of wave-
deformable mirror’s surface so that it accurately corrects for lengths. We evolved on a planet with an atmosphere that is
the distortion of the wavefront. An example of an image cor- transparent to light in the visible part of the spectrum but
rected by adaptive optics is shown in Figure 5.17. The wide- is opaque to almost all other wavelengths. Nearly all of the
spread use of adaptive optics has now made the image quality X-ray, ultraviolet, and infrared light arriving at Earth is
Kitt Peak (visible) Groundbased observatories look through spectral windows in the atmosphere.
Satellite observatories study wavelengths Some wavelengths can be studied from high,
where Earth’s atmosphere is opaque. dry mountains like Mauna Kea (blue line).
Compton Chandra JCMT Green Bank

(γ-rays) (X-rays) Fuse (UV) Spitzer (infrared) (microwave) VLA (radio) (radio) Arecibo (radio)
Visible Infrared
Transparency (%)
window windows
100
Radio window
50 Atmosphere Atmosphere Atmosphere
is opaque is opaque is opaque
0
10–12 10–11 10–10 10–9 10–8 10–7 10–6 10–5 10–4 10–3 10–2 10–1 1 101 102 103
0.001 nm 0.01 nm 0.1 nm 1 nm 10 nm 100 nm 1 μm 10 μm 100 μm 1 mm 1 cm 10 cm 1m 10 m 100 m 1 km
Wavelength
FIGURE 5.18 Earth’s atmosphere blocks most electromagnetic radiation.
blocked by Earth’s atmosphere before it reaches the ground.47 we can see and record. As with our earlier approach, we
The visible part of the spectrum, which is not blocked by start with the most basic of telescopes, the human eye.
our atmosphere, is a fairly narrow range of wavelengths, The detector in the human eye is the retina (see Figure
or window, through which we can view the universe. There 5.6), and the individual receptor cells that respond to light
are a few other atmospheric windows in the spectrum as falling on the retina are called “rods” and “cones.” Cones are
well, as shown in Figure 5.18. located near the eye’s optical axis at the center of our vision.
Do not make the mistake of thinking that light that They provide the highest resolution and enable us to rec-
fails to reach the surface of Earth is uninteresting. There ognize color. The size and spacing of cones—not the 7-mm
are many things that we can learn only by observing the pupil—determine the 1-arcminute resolution of the human
universe outside the visible window. Although radio obser- eye. Rods, which are located away from the eye’s optical axis
vations are also possible from the ground, we owe a large and are responsible for our peripheral vision, provide the
fraction of what we know about the universe to a host of highest sensitivity to low light levels, but they have poorer
ultraviolet, X-ray, gamma-ray, and infrared telescopes that, resolution and cannot distinguish color. The photons to
beginning in the 1960s, were carried above Earth’s atmo- which the human eye is sensitive have wavelengths ranging
sphere by rockets. We’ll discuss space telescopes in more from about 400 nm (deep violet) to 700 nm (far red).
detail later in the chapter. As photons from a star enter the aperture or pupil of the
eye, they fall on and excite cones at the center of vision.
The cones then send a signal to the brain, which interprets
this message as “I see a star.” So what limits the faintest
5.2 Optical Detectors stars we can see with our unaided eyes, assuming a clear,
dark night and good eyesight? This limit is determined in
and Instruments part by two factors that are characteristic of all detectors:
integration time and quantum efficiency.
In reality, the sole purpose of an astronomical telescope is Integration time is the limited time interval in which
simply to collect light waves from the cosmos and bring the eye can add up photons. For the human eye, the inte-
them to a focus. We turn now to the various ways in which gration time is about 100 milliseconds (ms). If two images
astronomers capture these light waves and convert them to on a television or computer screen appear 30 ms apart, you
useful information. Detectors are devices placed in a tele- will see them as a single image because your eyes will sum
scope’s focal plane to transform light waves into images that whatever they see over an interval of 100 ms. If the images
occur, say, 200 ms apart, you will see them as separate
47
Ozone and molecular scattering are the principal atmospheric compo- images. So the signals the cones send to your brain include
nents that block electromagnetic radiation shorter than visible wave- only those photons that arrive within an interval of 100 ms.
lengths from reaching Earth’s surface. Water vapor and carbon dioxide This relatively brief integration time is the biggest factor
tend to block radiation longer than visible wavelengths. limiting our nighttime vision. Stars too faint for us to see
(a) nomical detector. Permanent records of astronomical obser-

vations were limited to what an experienced observer could
sketch on paper while working at the eyepiece of a telescope,
as illustrated in Figure 5.19a. Photography would eventu-
ally change all that.
Photographic Plates
Turn a telescope or binoculars toward a field of stars and
what do you see? As you might expect, the stars appear both
closer and brighter. Now look at a distant landscape. The
scene seems closer, but its surface is no brighter. What is
going on here? It turns out that only “point sources” such as
stars appear brighter in a telescope. Like the distant land-
scape, the Orion Nebula and other extended astronomical
objects look bigger in the eyepiece, but their surfaces are
no brighter than they appear to the unaided eye. A tele-
(b)
scope gives you a closer view of the Moon, but it does not
increase the Moon’s surface brightness. For more than two
centuries after the invention of the telescope, astronomers
struggled with this surface brightness problem. No matter
how big they built their telescopes, nebulae and galaxies
might appear larger but their faint details remained elusive.
The problem, of course, was not with the telescopes but with
the limitations of optics and the human eye. Only with the
invention of photography and the later development of elec-
tronic cameras were astronomers finally able to discern the
faint but intricate fabric of the cosmos.
In 1840, John W. Draper (1811–1882), a New York chem-
istry professor, created the earliest known astronomical
photograph, shown in Figure
5.19b. His subject was the Photography opened
Moon. Photography was not the door to modern
quick to catch on among astronomy.
astronomers, though, because
the early “daguerreotype” process was slow and very messy.
The relatively simple dry emulsion process finally came
FIGURE 5.19 (a) A drawing of the galaxy M51 made by William
along in the late 1870s, and with that innovation, astronomi-
Parsons (Lord Rosse, 1800–1867) in 1845. Compare this cal photography took off. Astronomers could now create
drawing with the modern photograph of M51 in Figure 5.20. (b) A permanent images of planets, nebulae, and galaxies with
photograph of the Moon taken by John W. Draper in 1840. ease. Thousands of photographic plates soon filled the “plate
vaults” of major observatories. Photography had created its
own astronomical revolution.
with our unaided eyes are those that produce too few pho- In the dry emulsion process, a layer of gelatin containing
tons for our eyes to process in 100 ms. tiny crystals of silver halide is coated onto glass plates or
Another effect, called quantum efficiency, also restricts film.58 During an exposure, photons landing on the emulsion
our nighttime vision. As the name implies, quantum effi- energize the silver halide crystals, creating what is called
ciency is the likelihood that a particular photon landing on a “latent” image. “Developing” the emulsion turns these
the retina will, in fact, produce a response. For the human eye, small crystals into black grains of metallic silver, forming
it takes about 10 photons landing on a cone to activate a single a permanent image. The highest density of silver grains
response. In other words, the quantum efficiency of our eyes occurs where the telescope’s image was the brightest. So,
is about 10 percent. Together, integration time and quantum bright becomes black, and the photographic image is nega-
efficiency determine the rate at which photons must arrive on
the retina before the brain says, “Aha, I see something.” 58
Glass plates, although far more expensive than fi lm, are generally
For more than two centuries after the invention of the
used for imaging and spectroscopy because of their greater geomet-
telescope, the retina of the human eye was the only astro-
ric stability.
they have excellent linearity, which means that their elec-

tronic output is directly proportional to the intensity of
light falling on the photometric detector. But they can
measure only one stellar image at a time, so observation is
truly labor-intensive. Other detectors, such as “vidicons,”
are electronic imaging devices with sensitivity far superior
to photographic emulsions. Vidicons unfortunately suffer
from an electronic instability that causes small geometric
distortions of the image. You may have seen pictures taken
by early spacecraft that had little plus-shaped marks super-
imposed on them. These “fiducial” marks were engraved
on the faceplates of the vidicons to help with the removal
of geometric distortions in the image. The search for a bet-
ter detector continued.
In 1969, scientists at Bell Laboratories were devel-
oping “picture phones”—telephones containing a small
camera and viewing screen that could display an image
FIGURE 5. 20 An image of Galaxy M51 on a photographic of the person at the other end of the conversation. As it
plate. turned out, public opinion declared Bell’s picture phones
an invasion of personal privacy and they were never com-
mercially produced, but the research led to the invention
tive, as illustrated in Figure 5.20. The quantum efficiency of a remarkable detector called a charge-coupled device, or
of most photographic emulsions used in astronomy is very CCD. Astronomers soon realized that this was the detector
low—typically 1–3 percent, which is even poorer than that they had been looking for. Shortly after it was fi rst applied
of the human eye. But unlike the eye, photographic emul- to astronomical imaging in the mid-1970s, the CCD became
sions can overcome poor quantum efficiency by integrating the detector of choice in almost all astronomical imaging
photons over intervals of many hours. Photography made it applications. Gone were the problems associated with
possible for astronomers to record and study objects much photographic emulsions, photoelectric photometers, and
fainter than the human eye can see. vidicon-type imagers. The CCD is a photometrically linear
Photography is not without its own problems. Very faint imaging device, able to perform precise photometry over
objects often require long exposures that can take up much large regions of sky. It responds over a wide spectral range
of an observing night. (Imagine how you would feel if your from 200 to 1,200 nm and has a high quantum efficiency,
10-hour exposure were spoiled because of a mishap.) In typically 80 percent or greater. The output from a CCD is a
addition, the spectral range of photographic emulsions is digital signal that can be sent directly from the telescope
hardly broader than that of the human eye. In fact, for many to image-processing software or stored electronically for
years photographic plates were sensitive to only violet and later analysis.
blue light. Another problem is the nonlinear response of A CCD consists of an ultrathin wafer of silicon, less than
photographic emulsions to light. This is just another way of the thickness of a human hair, which is divided into a two-
saying that the optical density (blackness) of the processed dimensional array of picture elements, or pixels, as seen
emulsion is not proportional to the intensity of light falling in Figure 5.21a. When a photon strikes a pixel, it creates a
on it. Finally, there is the nontrivial matter of economics. small electric charge within the silicon. As each CCD pixel
Each photographic plate can be used only once, and they is “read out,” the digital signal that flows to the computer is
are expensive. By the middle of the 20th century, the search almost precisely proportional to the accumulated charge.
was on for electronic detectors that would overcome many This is what we mean when we say that the CCD is a linear
of the deficiencies of photographic plates. device. Like many electronic detectors, CCDs are subject to
thermal noise—a false signal caused by the thermal agita-
tion of charge-carrying atoms within the silicon wafer. But
Charge-Coupled Devices we can minimize thermal noise by cooling the CCDs down
to liquid nitrogen temperatures (~80 K). The fi rst astro-
Throughout the latter half of the 20th century, astronomers nomical CCDs were small arrays containing no more than
employed various electronic detectors to overcome the a few hundred thousand pixels. The larger CCDs used in
sensitivity, spectral range, and nonlinearity problems of astronomy today—like the one seen in Figure 5.21b—may
photography. Some, such as “photoelectric photometers,” contain as many as 100 million pixels. Still-larger arrays are
are nonimaging devices that receive photons from an object under development, as ever-faster computing power keeps
and convert them to an electronic signal that is propor- up with image-processing demands.
tional to the brightness of the object. These devices work The impact that CCDs have had on astronomy cannot be
extremely well for precision stellar photometry because overstated! The use of conventional photography by ground-
(a) (b)
–
– e
– e
e
(c)
FIGURE 5. 2 1 (a) A simplified diagram of a charge-coupled device (CCD). Photons

from a star land on pixels (gray squares) and produce free electrons within the silicon.
The electron charges are electronically moved sequentially to the collecting register at
the bottom. Each row is then moved out to the right to an electronic amplifier, which
converts the electric charge of each pixel into a digital signal. (b) This very large CCD
contains 12,288 × 8,192 pixels. (c) CCDs have been used in many consumer electronic
devices, such as camera phones..
(a) (b)
FIGURE 5. 22 (a) A CCD image of Saturn taken in 2005 by an amateur astronomer with a
36-cm telescope. (b) A pre-CCD photograph of Saturn taken in 1974 with a 1.5-meter telescope at
the Catalina Observatory of the University of Arizona.
based observatories has steadily declined over the past two moelectrically cooled CCD imagers with impressive results.
decades, and it is seldom Figure 5.22a is an image of Saturn taken by an amateur astron-
employed today. As you surf The CCD is the omer with a 36-centimeter (cm) telescope. This image shows
the Internet, nearly every spec- astronomer’s more detail than the best professional photographs of Saturn
tacular astronomical image detector of choice. taken before CCDs became available to astronomers, an exam-
that pops up on your screen ple of which is shown in Figure 5.22b.
was produced by a CCD, whether from groundbased telescopes You may never have seen a CCD, but in recent years they
or from those in space. And professional astronomers are not have found their way into many devices that we now take
the only ones making spectacular photos with CCDs. Ama- for granted—digital cameras, digital video cameras, and
teur astronomers are now using commercially available, ther- camera phones, to name just a few (Figure 5.21c).
Spectroscopy is the study of an object’s electromagnetic

Spectrographs radiation in terms of its component wavelengths, and we’ll
Spectrographs—or spectrometers, as they are often called— encounter its many applications throughout the chapters
are another of the astronomer’s essential tools. As we learned to come as we continue in our journey through the Solar
in Chapter 4, we can probe the chemical and physical proper- System and beyond.
ties of distant objects by studying their spectra. Early spectro-
graphs used glass prisms to disperse the incoming light into
its component wavelengths (Figure 5.23), creating a spectrum
like the ones shown in Figures 4.14 and 5.11. A photographic
plate recorded the spectrum for precise measurement of the 5.3 Radio Telescopes
wavelengths of its spectral lines. One disadvantage of glass
prisms is that dispersion (see Foundations 5.1) is not uni- Photons of visible light are not the only messengers of
form with wavelength. Prism spectrographs produce spec- important information from the cosmos. Karl Jansky (1905–
tra with more dispersion at the shorter-wavelength (violet) 1950) was a young physicist working for Bell Laboratories
end of the spectrum than at the longer-wavelength (red) end. in the early 1930s when he was assigned the job of iden-
Another drawback is that glass is opaque to ultraviolet and tifying sources of static in transatlantic radiotelephone
long-wavelength infrared light. Thus, prism spectrographs service. Jansky built a pointable antenna and soon iden-
are more or less limited to the visible part of the spectrum. tified the major sources of static as nearby and distant
Most modern spectrographs use a “diffraction grating” to thunderstorms. But one source remained a mystery—a
disperse the light (see Foundations 5.1 and Figure 5.13b) and faint, steady hiss rose and fell once every 23 hours and
a CCD to record the spectrum. 56 minutes. Recall from Chapter 2 that this is the same
Spectrographs may be designed for either low or high period as Earth’s rotation as seen from the stars. What
dispersion. Astronomers use low-dispersion spectrographs Jansky had detected was the characteristic time interval
to identify the chemical components of feeble light sources of celestial objects far beyond our Solar System. In 1932,
such as nebulae. Low-dispersion spectrographs also mea- Jansky identified the mysterious source. It was located in
sure the reflected spectral energy distribution of faint Solar the Milky Way in the direction of the galactic center, in
System objects such as small or distant asteroids. Mea- the constellation of Sagittarius. Excited by his discovery,
surements of temperature and radial velocity, on the other he submitted a request to build a large dish antenna, or
hand, usually require very high dispersion. Today’s high- radio telescope, to study these signals in more detail. Bell
dispersion spectrographs have evolved into substantial Labs turned down the request. After all, Jansky had already
scientific instruments—hardly the sort of thing that one provided the information they needed. Nevertheless, Jan-
would transport from one telescope to another. The high- sky’s discovery marked the birth of radio astronomy. In
dispersion spectrographs now associated with most major his honor, the basic unit for the strength of a radio source
telescopes tend to be huge and weigh several metric tons— is called the jansky (Jy).
think SUV size. (For obvious reasons, they are not attached In 1937, Grote Reber (1911–2002), an American radio
directly to the telescope!) A system of mirrors feeds light engineer and ham radio operator, decided to build his own
from the telescope’s focal plane into the spectrograph, which radio telescope. It consisted of a parabolic sheet of metal,
is located nearby. 9 meters in diameter, with a radio receiver mounted at the
focus. With this instrument, Reber conducted the fi rst sur-
vey of the sky at radio frequencies, and he published the fi rst
radio frequency map of the galaxy in 1944. Reber was largely
FIGURE 5. 23 A diagram of a prism spectrograph. responsible for the rapid advancement in radio astronomy
that blossomed in the post–World War II era.
Slit Collimating Radio telescopes are yet another of astronomy’s indis-
lens
pensable tools. From our Solar System to the most distant
galaxies, the penetrating power of radio waves unlocks
secrets not possible with shorter-wavelength optical or
infrared telescopes. Look back at Figure 5.18 and notice the
Camera wide radio window in Earth’s atmosphere, covering wave-
lens
lengths ranging all the way from about a centimeter to 10
The telescope meters.69 This ability of radio waves to pass unattenuated
focuses light
from a star here. 69
Microwave astronomy is considered a branch of radio astronomy.
Microwaves are very high-frequency radio waves with wavelengths
Re een
ranging from about 1 mm to 10 cm at the short-wavelength end of the

Gr iolet
d
Photographic
V
radio spectrum. As seen in Figure 5.18, Earth’s atmosphere is only

plate or CCD partially transparent to microwaves.
(a) est radio telescope is the 305-

meter Arecibo dish built into Radio telescopes
a natural bowl-shaped depres- enable astronomers
sion in Puerto Rico (Figure to “see” through
5.24b). But there can be a price obscuring gas
to pay for size. As you might and dust.
guess from the picture, this
huge structure is too big to steer. Instead, pointing it requires
moving the radio receiver suspended in the focal plane
above the dish. Arecibo’s targets are therefore limited to
celestial sources that pass within 20° of the zenith as Earth’s
rotation carries them overhead.
As large as radio telescopes are, they have relatively poor
angular resolution. Recall our earlier discussion about dif-
fraction (see Foundations 5.1
and Math Tools 5.1). A tele- Single radio
(b) scope’s angular resolution is telescopes have
determined by the ratio λ/D, relatively poor
where λ is the wavelength of resolution . . .
electromagnetic radiation and
D is the telescope’s aperture. (Keep in mind that a larger
ratio means poorer resolution.) Radio telescopes have diam-
eters much larger than the apertures of most optical tele-
scopes, and that helps. But the wavelengths of radio waves
are typically several hundred thousand times greater than
the wavelengths of visible light, and that hurts. Radio tele-
scopes are thus hampered by the very long wavelengths
they are designed to receive. Consider the huge Arecibo
dish. Its resolution is typically about 1 arcminute, no better
than the unaided human eye! So radio astronomers have
had to develop their own bag of tricks, and one of the clev-
erest is the interferometer.
When we combine the signals from two radio telescopes
in a certain way, the separation between them—not the
diameters of the individual telescopes—determines the
angular resolution. For example, if two 10-meter telescopes
are located 1,000 meters apart, the D in λ/D is 1,000, not 10.
Such an arrangement is called
an interferometer because it . . . but
makes use of the wavelike interferometric
properties of electromagnetic arrays overcome this
radiation, in which signals problem.
from the individual telescopes
interfere with one another (see Foundations 5.1). Usually
several telescopes are employed, in an arrangement called
FIGURE 5. 2 4 (a) A large radio telescope in Australia. (b) The
Arecibo radio telescope is the world’s largest. The steerable an interferometric array. Through the use of very large
receiver suspended above the dish permits limited pointing arrays, radio astronomers can attain and exceed the angular
toward celestial targets as they pass close to the zenith. resolution enjoyed by their optical colleagues.
One of the larger radio interferometric arrays is the Very
Large Array (VLA) in New Mexico, shown in Figure 5.25.
The VLA is made up of 27 individual movable dishes spread
through our atmosphere is also the property that enables out in a Y-shaped configuration up to 36 km across. At a
us to peer through the vast amounts of gas and dust found wavelength of 10 cm, this array can achieve resolutions of
in many galaxies. Most radio telescopes are large, steer- less than 1 arcsecond. Not satisfied, radio astronomers have
able parabolic dishes, typically tens of meters in diameter, sought still larger arrays—and no one can accuse them of
such as the one shown in Figure 5.24a. The world’s larg- having limited imagination. The Very Long Baseline Array
FIGURE 5. 25 The Very Large Array (VLA) in New Mexico.
(VLBA) employs 10 radio telescopes spread out over more

FIGURE 5. 26 The Very Large Telescope (VLT), operated by than 8,000 km from the Virgin Islands in the Caribbean to
the European Southern Observatory in Chile. Movable auxiliary Hawaii in the Pacific. At a wavelength of 10 cm, this array
telescopes allow the four large telescopes to operate as an can reach resolutions better than 0.003 arcsecond. It might
optical interferometer. seem that Earth’s diameter would set the ultimate limit on
resolution for radio astronomers, but a radio telescope put
into near-Earth orbit as part of a Space Very Long Baseline
Interferometer (SVLBI) overcomes even this limit. Future
SVLBI projects would extend the baseline to as much as
100,000 km, yielding resolutions far exceeding those of any
existing optical telescope.
Before leaving our discussion of interferometers, we
should point out that radio astronomers are not the only
ones using the interferometer’s greater resolving power.
Optical telescopes can also be arrayed to yield resolu-
tions greater than those of single telescopes, although for
technical reasons the individual units cannot be spread
as far apart as radio telescopes. The Very Large Telescope
Interferometer (VLTI), operated by the European Southern
Observatory (ESO) in Chile, makes use of a combination
of the four VLT 8-meter telescopes (Figure 5.26) and four
movable 1.8-meter auxiliary telescopes. It has a baseline of above most of the water vapor in the atmosphere. But
up to 200 meters, yielding angular resolution in the milli- gaining full access to the complete electromagnetic spec-
arcsecond range. trum is yet another matter. This means getting completely
above Earth’s atmosphere.710 In the late 1940s, scientists put
ultraviolet and cosmic-ray instruments in the nose cones
of captured German V-2 rockets and launched them from
5.4 Getting above the White Sands Proving Ground in New Mexico to alti-
tudes greater than 100 km. Of course, such observations
Earth’s Atmosphere: had to be brief because, thanks to gravity, the rockets and
their scientific instruments invariably came back down.
Airborne and Orbiting The next step was to put astronomical instruments into
orbit. The fi rst astronomical satellite was the British Ariel
Observatories 1, launched in 1962 to study solar ultraviolet and X-ray
radiation and the energy spectrum of primary cosmic
Various components of Earth’s atmosphere block infra- rays. Today we have a multitude of orbiting astronomical
red photons from reaching astronomical telescopes on the telescopes covering the electromagnetic spectrum from
ground, and one of the greatest culprits is water vapor. gamma rays to microwaves, with many more in the plan-
Imagine strolling through a Hawaiian rain forest and tak- ning stage (see Table 5.2).
ing in the fragrance of the warm, humid tropical air. In Optical telescopes, such as the 2.4-meter Hubble Space
your bliss you might be unaware that 4 km above you on Telescope (HST), can operate successfully at modest alti-
the summit of Mauna Kea, astronomers are at war with the tudes in what is called low
same atmospheric water vapor that has so stimulated your Earth orbit (LEO), 600 km Orbiting observatories
senses. Water vapor is the enemy of the infrared astrono- above Ea r t h’s su r face. explore regions of the
mer. We have already seen that Earth’s atmosphere distorts Launched in 1990, HST has spectrum inaccessible
telescopic images and that certain molecules in Earth’s been the workhorse for UV, from the ground.
atmosphere, including water, block large parts of the elec- visible, and IR space astron-
tromagnetic spectrum from getting through to the ground. omy for nearly two decades. LEO is also the region where
It shouldn’t surprise us then to fi nd that astronomers have the International Space Station (ISS) and many scientific
put considerable effort into locating their instruments above satellites orbit. For others, 600 km is not nearly high enough.
as much of the atmosphere as possible. Look for an astro- The Chandra X-ray Observatory, NASA’s X-ray telescope,
nomical observatory and you’ll usually encounter it at the cannot tolerate even the tiniest traces of atmosphere and
summit of a tall mountain. Most of the world’s larger astro- therefore orbits more than 16,000 km above Earth’s surface.
nomical telescopes are located 2,000 meters or more above And even this is not distant enough for some telescopes.
sea level. Mauna Kea, a dormant volcano and home of the NASA’s Spitzer Space Telescope, an infrared telescope, is
Mauna Kea Observatories (MKO), rises 4,200 meters above so sensitive that it needs to be completely free from Earth’s
the Pacific Ocean. At this altitude the MKO telescopes sit own infrared radiation. The solution was to put it into a
above 40 percent of Earth’s atmosphere; but more important, solar orbit, trailing tens of millions of kilometers behind
90 percent of Earth’s atmospheric water vapor lies below. Earth. Many future space telescopes, including the James
Still, for the infrared astronomer the remaining 10 percent Webb Space Telescope, NASA’s replacement for HST, will
is troublesome. orbit free of Earth, bound only to the Sun.
One way to solve the water vapor problem is to make use
of high-flying aircraft. NASA’s Kuiper Airborne Observa-
tory (KAO), a modified C-141 cargo aircraft, carried a 90-cm
telescope and was among the fi rst of these flying observa- 5.5 Getting Up Close with
tories. It could cruise at an altitude of 14 km, above 98 per-
cent of Earth’s water vapor. NASA retired KAO in 1995 and Planetary Spacecraft
replaced it with the Stratospheric Observatory for Infrared
Astronomy (SOFIA), which is expected to become fully As we have stressed often in this book, we live in a remark-
operational in 2011. SOFIA will carry a 2.5-meter telescope able time of discovery, when our newfound technological
and work in the far-infrared region of the spectrum, from prowess has enabled us to begin the process of explor-
1 to 650 micrometers (
m). It will fly in the stratosphere at ing our local corner of space. The general strategy for
an altitude of about 12 km, above 99 percent of the water
vapor in Earth’s lower atmosphere. 710
In a sense, there is no defi nable upper limit to Earth’s atmosphere. As
Airborne observatories are able to overcome atmo- we will learn in Chapter 8, our atmosphere simply blends into outer
spheric absorption of infrared light by placing telescopes space at an altitude of about 10,000 km.
5.5 Getting Up Close with Planetary Spacecraft 145
by, instruments aboard these spacecraft briefly probed

TA B L E 5 . 2 the physical and chemical properties of their targets and
their environments.
Reconnaissance spacecraft employ remote sensing
Selected Present and Future instrumentation much like the remote sensing techniques
Space Observatories used by Earth-orbiting satellites to study our own planet.
These include tools such as cameras capable of taking
images in different wavelength ranges, radar for mapping
Launch surfaces hidden beneath
Telescope Sponsor(s) Description Year obscuring layers of clouds, Planetary spacecraft
Hubble Space NASA, Optical, infrared, 1990 and spectrometers that spread take our instruments
Telescope (HST) ESA ultraviolet out the target’s light into a directly to the planets.
observations diagnosable spectrum. Remote
sensing enables planetary scientists to map other worlds,
Chandra X-ray NASA X-ray imaging and 1999
measure the heights of mountains, identify geological fea-
Observatory spectroscopy
tures, learn about types of rocks present, watch weather
Wilkinson NASA Cosmic 2001 patterns develop, measure the composition of atmospheres,
Microwave background and in general get a feeling for the “lay of the land.” Still
Anisotropy Probe radiation other instruments make in situ measurements of the
(WMAP) extended atmospheres and space environment through
Spitzer Space NASA Infrared 2004 which they travel.
Telescope observations The study of our Solar System from space is a truly inter-
national collaboration involving NASA, the European Space
Galaxy Evolution NASA Ultraviolet 2003 Agency (ESA), the Russian Federal Space Agency (Roscos-
Explorer observations mos), the Japan Aerospace Exploration Agency (JAXA),
(GALEX)
the China National Space Administration (CNSA), and the
Swift Gamma- NASA Gamma-ray bursts 2004 Indian Space Research Organisation (ISRO). Other countries
Ray Burst may soon join the endeavor.
Mission
Fermi Gamma- NASA, Gamma-ray 2008

ray Space European imaging and Flybys and Orbiters
Telescope partners gamma-ray bursts
Since the dawn of history, no human had ever seen the far
Planck ESA Cosmic microwave 2009 side of the Moon. As we learned in Chapter 2, we always
background see the same face of the Moon from Earth because its orbital
Herschel Space ESA Far-infrared and 2009 and rotational periods are equal to one another. With one
Observatory submillimeter side of the Moon permanently facing Earth, the inappro-
observations priately named “dark side” is hidden from our view. Hid-
den, that is, until October 18, 1959. On that date the Soviet
Kepler NASA Planet fi nder 2009
flyby probe Luna 3 sent back humanity’s fi rst view of the
James Webb NASA Replacement for 2014 far side of our nearest celestial neighbor (Figure 5.27). The
Space Telescope HST curious features of the Moon’s other side, so different from
(JWST) its Earth-facing half, amazed viewers around the world. No
matter how powerful we make our groundbased or Earth-
orbiting telescopes, Luna 3 showed us there is nothing quite
like going there.
Flyby missions have several distinct advantages in the
exploring our Solar System begins with a reconnaissance reconnaissance phase of exploration. First, they are rela-
phase, using spacecraft that fly by or orbit a planet or tively inexpensive and the easiest missions to design and
other body. Now, in the early 21st century, we have con- execute. Second, flyby spacecraft such as Voyager, shown
ducted preliminary reconnaissance of much of the Solar in Figure 5.28a, may be able to visit several different worlds
System. We have sent spacecraft flying by all of the clas- during their travels. The downside of flyby missions is that,
sical planets, giving humanity its fi rst ever close-up views thanks to the physics of orbits, these spacecraft must move
of these distant worlds and their moons. We have even by very swiftly. They are limited to just a few hours or at
seen comets and asteroids at close range. As they sped most a few days in which to conduct close-up studies of
cult missions than flyby missions; but orbiters can linger,

looking in detail at more of the surface of the object they
are orbiting and studying things that change with time,
like planetary weather. Spacecraft have orbited the Moon,
Venus, Mars, Jupiter, Saturn, and even an asteroid. Figure
5.28b shows the Cassini spacecraft, which was launched in
1997 and as this book goes to print is still sending us data
from its orbit around Saturn.
Landers, Rovers, and

Atmospheric Probes
Reconnaissance spacecraft provide a wealth of informa-
tion about a planet, but there is no better way of obtaining
“ground truth” than to put our instruments where they can
get right to the heart of it—within a planet’s atmosphere or
on solid ground. We have landed spacecraft on the Moon,
Mars, Venus, Saturn’s large moon Titan, and the asteroid
Eros. One spacecraft even shot a massive bullet into a comet
nucleus to observe the huge splash that created a tempo-
rary cometary atmosphere. These spacecraft have returned
pictures of planetary surfaces, measured surface chemis-
FIGURE 5. 2 7 Humanity’s first view of the far side of the Moon, try, and conducted experiments to determine the physical
seen in this image sent back by the Soviet probe Luna 3 in 1959. properties of the surface rocks and soils.
One disadvantage of using landed spacecraft is that only
their targets. Yet flyby spacecraft give us our first intimate a few landings in limited areas are practical because of the
views of our planetary neighbors and provide the details expense, and the results may apply only to the small area
we need to plan follow-up studies. around the landing site. Imagine, for example, what a dif-
More detailed reconnaissance work uses spacecraft that ferent picture of Earth we might get from a spacecraft that
orbit around planets. These are intrinsically more diffi- landed in Antarctica, as opposed to a spacecraft that landed
FIGURE 5. 28 Explorers of the planets. (a) Voyager flew past Jupiter, Saturn, Uranus, and
Neptune. (b) Cassini orbits Saturn.
(a) (b)
in the caldera (the summit crater) of a volcano or the floor of cut. This is quite different from simply grabbing any old
a dry riverbed. Sites to be explored with landed spacecraft rock by the side of the road, because that rock could have
must be very carefully chosen on the basis of reconnaissance come from anywhere. In geo-
data if we are to know what to make of the information they logical sampling it is impor- Sample returns
provide. We can mitigate some of the limitations of landers tant to have samples from a provide
by putting their instruments on wheels and sending them source of known context. “ground truth.”
from place to place, exploring the vicinity of the landing Some have claimed that we do
site. Such remote-controlled vehicles, called rovers, were not need samples returned from Mars, because we already
used fi rst by the Soviet Union on the Moon four decades have the martian meteorites. The problem is that we do not
ago, and more recently by the United States on Mars. The know where on Mars those meteorites came from and, there-
opening photo for this chapter shows an artist’s view of one fore, how they fit into the planet’s global geology. Plans are
of two rovers still roaming the martian landscape. currently under way for unmanned “sample and return”
We have also sent probes into the atmospheres of Venus, missions to Mars.
Jupiter, and Titan. As they descend, atmospheric probes con- Of course, we could not collect specimens in a national
tinuously measure and send back data on physical proper- park without permission and a scientifically valid reason.
ties such as temperature, pressure, and wind speed, along Similarly, the return of extraterrestrial samples to Earth is
with other properties, such as chemical composition. Meteo- governed by international treaties and standards to ensure
rologists take measurements of our terrestrial atmosphere that these samples do not contaminate Earth. For example,
from the surface up by sending their instruments aloft before the lunar samples brought back by the Apollo mis-
in balloons. By contrast, planetary scientists must work sions could be studied, they (and the astronauts) had to be
from the top down by suspending their instruments from placed in quarantine and tested for alien life-forms. The
parachutes. The end result is much the same. Atmospheric same international standards apply to spacecraft landing
probes have survived all the way to the solid surfaces of on planets. The goal of these standards is to avoid forward
Venus and Titan, sending back streams of data during their contamination, or transporting life-forms from Earth to
descent. An atmospheric probe sent into Jupiter’s atmo- another planet. If there is life on other planets, then not
sphere never reached that planet’s surface because, as we only is there concern about introducing potential harm, but
will learn later, Jupiter does not have a solid surface in the from a scientific perspective we do not want to “discover”
same sense that terrestrial planets and moons do. After life that we, in fact, introduced.
sending back its data, the Jupiter probe eventually melted With numerous missions under way and others on the
and vaporized as it dropped into the hotter layers of the horizon, unmanned exploration of the Solar System is an
planet’s atmosphere. ongoing, dynamic activity. In our journey we will frequently
refer to space missions and the information they return, but
today’s hot results may be tomorrow’s old news in light of
other, even more exciting discoveries. We hope that you will
Sample Returns make use of the 21st Century Astronomy Web site as a gate-
If you pick up a rock from a road cut, there is a lot you way to the wealth of exciting results that the future holds.
might learn from the rock using tools that you could eas-
ily carry in your pocket. But the sophistication of the tools
you could carry with you would be limited. It would be
much better to pick up a few samples and carry them back 5.6 Other Earth-Based
to a laboratory equipped with a full range of state-of-the-
art instruments capable of measuring chemical composi- Astronomical Tools
tions, mineral types, radiometric ages (see Chapter 7), and
other information needed to reconstruct the story of their High-profile space missions have sent back stunning images
origin and evolution. So, too, is the case in Solar System and data from across the electromagnetic spectrum, but
exploration. One of the most powerful methods for inves- there are still other—and perhaps less obvious—tools that
tigating remote objects is to collect samples of the objects astronomers use right here on the ground, including par-
and bring them back to Earth for detailed study. So far, ticle accelerators, neutrino and gravitational wave detec-
only samples of the Moon, a comet, and the solar wind (a tors, and high-speed computers. Some are even buried deep
stream of charged particles from the Sun) have been col- underground.
lected and returned to Earth. Ever since the early years of the 20th century, physi-
As we will learn in Chapter 12, we do know of meteor- cists have been peering into the structure of the atom by
ites that are considered to be part of Mars, but there is a observing what happens when small particles collide. By
problem with putting them in their proper geological per- the 1930s, physicists had developed the means to accelerate
spective. We just talked about picking up a rock from a road charged subatomic particles such as protons to very high
speeds and then observe what happens when they slam

into a target. From such experiments (which are continu-
ing even today) physicists have discovered many kinds
of subatomic particles811 and learned about their physical
properties. High-energy particle colliders have proven to
be an essential tool for physicists studying the basic build-
ing blocks of matter.
Why, then, do we regard the high-energy particle col-
lider as a tool of the astronomer? How does the astronomer’s
interest in the largest objects
in the universe relate to what Colliders provide us
happens on the very smallest with the physics
scales? As we will see in we need to
Chapter 22, to understand the understand the
very largest structures we see early universe
in the universe—indeed, the and the formation
large-scale universe itself— of structure.
we need to understand the
physics that took place during the earliest moments in the
universe, when everything was unbelievably hot and dense.
Although we have not yet reached that level of comprehen-
sion, the high-energy particle colliders that physicists use
today are designed to lead us there. As we will find in Chap-
ters 21 and 22, this knowledge may help us understand such
issues as the nature of the dominant matter and energy in
the universe and whether there really is a beginning or an
end to our universe.
Two factors determine the effectiveness of particle
accelerators: the energy they can achieve and the number
of particles they can accelerate. Whereas the fi rst particle
accelerator, the “cyclotron,” attained an energy of about
FIGURE 5. 29 The high-energy linear particle accelerator at 10 –13 joule, modern particle colliders now reach much
the Fermi National Accelerator Laboratory in Batavia, Illinois.
higher energies. For example, the accelerator at the Fermi
National Accelerator Laboratory (Fermilab—Figure 5.29)
can accelerate protons up to 1.6 × 10 –7 joule, resulting in
collision energies of 3.2 × 10 –7 joule. This may seem like
FIGURE 5. 30 The ATLAS particle detector at CERN’s Large
a small number, but it corresponds to more than a thou-
Hadron Collider near Geneva, Switzerland. The enormous size
sand times the rest mass energy (mc 2) of the proton. To
of this instrument is evident from the figure standing near the
bottom center of the picture. put it in perspective, this amount is the energy of a fly-
ing mosquito all concentrated into a single tiny proton.
Yet this energy is dwarfed by the new Large Hadron Col-
lider near Geneva, Switzerland, built by CERN, the Euro-
pean Organization for Nuclear Research (Figure 5.30).
This facility can produce collisions with seven times
as much energy as the Fermi Tevatron Collider—about
2.2 × 10 –6 joule.
In Chapter 14’s discussion of the Sun, you will be intro-
duced to the neutrino, an elusive elementary particle that
plays a major role in the physics of stellar interiors. Of
811
In Chapter 4 we learned about three subatomic particles: the proton,
neutron, and electron. In this section we will introduce another, the
neutrino; and in Chapter 14 we will meet still another, the positron.
However, many other subatomic particles are now known to physi-
cists—in fact, too many to list here.
course, we cannot see beneath the Sun’s surface, but obser- million pixels, with each pixel displaying roughly 10,000
vations of neutrinos can give us important insight into what levels of brightness. That adds up to a trillion pieces of
is happening deep within. There’s only one problem, and information in each image! And that’s only one image.
it’s a big one: neutrinos are To analyze their data, astronomers typically do calcula-
extremely difficult to detect. Neutrino detectors tions on every single pixel of an image in order to remove
To study a neutrino you fi rst see the death of stars unwanted contributions from Earth’s atmosphere or to cor-
have to grab one, and they are and measure the rect for instrumental effects. From the astronomer’s point
nearly impossible to catch. In Sun's interior. of view, without high-speed computers the CCD would be
a sense, this is fortunate for just another electronic curiosity.
us. In less time than it takes you to read this sentence, a High-speed computers also play an essential role in
thousand trillion (1015) solar neutrinos from the Sun are generating and testing theoretical models of astronomical
passing through your body. It doesn’t matter a bit if you are objects. Even when we completely understand the underly-
reading this at night. Neutrinos are so nonreactive with ing physical laws that govern the behavior of a particular
matter that they can pass right through Earth (and you) as object, it is frequently the case that the object is so complex
though it (or you) weren’t there at all. In fact, half of the that it would be impossible to calculate its properties and
neutrinos produced by the Sun would make it through a behavior without the assistance of high-speed computers.
slab of lead a light-year thick. For us to detect a neutrino, it For example, as we learned in Chapter 3, we can use New-
has to interact with a detector. Several neutrino detectors ton’s laws to easily compute the orbits of two stars that are
are in operation today. All are buried deep underground to gravitationally bound to one another because their orbits
avoid false detection of cosmic background radiation. Neu- take the form of simple ellipses. However, it is not so easy to
trino detectors typically record only one out of every 1022 understand the orbits of the hundred billion (1011) stars that
(10 billion trillion) neutrinos passing though them, but that’s make up our Milky Way Galaxy, even though the underly-
enough to reveal processes deep within the Sun or witness ing physical laws remain the same.
the violent death of a star 160,000 light-years away. As if that were not complicated enough, consider the
Another elusive phenomenon is the gravitational wave. problem involving the collision of two such galaxies—and,
Gravitational waves are disturbances in a gravitational for good measure, throw in some gas in addition to the stars.
field, similar to the waves that spread out from the dis- We can see the result in Figure 5.31. When we apply even
turbance you create when you toss a pebble into the quiet the fastest available computers to this problem, the sequence
surface of a pond. There is strong, although indirect, obser- shown in Figure 5.31 is only an approximation of what we
vational evidence for their existence, as we will see in believe really happens. Modern computers have enough
Chapter 17 when we discuss “binary pulsars.” However, speed and memory to handle the behavior of a few million
gravitational waves are so elusive that we have not yet actu- stars at best, so we are forced to assume that a single star
ally detected them. Several facilities, including the Laser in the computer simulation really represents hundreds of
Interferometer Gravitational-Wave Observatory (LIGO), thousands of real stars.
have been constructed to detect gravitational waves. Sci- Similar modeling procedures have worked well in deter-
entists are eager to detect gravitational waves—not so mining the interior properties of stars and planets, includ-
much to confi rm their existence but to study the physical ing our own Earth. Although we cannot “see” beneath their
phenomena they are likely to reveal, such as the birth and surfaces, we have a surprisingly good understanding of
evolution of the universe, stellar evolution, or the very their interiors, as we will learn in later chapters. We begin
force of gravity itself. a model by assigning well-understood physical properties to
Finally, a survey of the astronomer’s tools would not tiny volumes within a planet or star. The computer assem-
be complete without a discussion of the essential role of bles an enormous number of these individual elements into
computers. Imagine your life without computers. From an overall representation of the complete body. When it is
laptops to the largest servers, we depend on computers all put together, we have a rather good picture of what the
to surf the Internet, process our data, and organize our interior of the star or planet is like.
daily lives. Tiny computer processors control every mod- We have seen how our view of the sky changes with
ern convenience from automobiles to washing machines. location and time, discovered the laws that govern the
And, as you might assume, computers are essential in the motions of celestial bodies, studied the properties of light,
world of science. Data gathering, analysis, and interpreta- and learned about the tools that astronomers use in their
tion are entirely dependent on computers—and the more continuing pursuit of scientific knowledge. With this back-
powerful, the better. Consider, for example, analyzing a ground in the fundamentals now behind us, it’s time to step
night’s worth of astronomical images recorded by a very away from Earth and look outward, fi rst to our Solar System
large CCD. A single image may contain as many as 100 and then to the greater universe beyond.
(a) (b)
(c) (d)
FIGURE 5. 31 Numerical simulation of the merging of two galaxies, including the gravitational
attraction of all forms of matter. (a) Just before the merge. (b) After the galaxies have passed
through one another. (c) As the cores orbit each other and merge. (d) When the central core begins
to settle down.
dered about those mysterious few that seemed to move

Seeing the Forest for the Trees freely among the others. With nothing more than their
eyes to fulfill their curiosity, our early ancestors could
From our perspective in the 21st century, we can only only observe and wonder. But wondering alone does not
imagine what went through the minds of our distant give rise to comprehension. Insight far beyond personal
ancestors as they gazed upward toward the heavens. experience was necessary, and it was a long time coming.
We might guess that they felt completely comfortable Somewhat more than two millennia ago, enlightenment
with the daily movements of the Sun and the Moon, yet of a sort came to the classical world. Greek philosophers
they likely experienced fear when confronted with an concluded that these “wandering stars,” or planets, were
eclipse or the appearance of an occasional comet. They unlike the distant canopy of fi xed stars—that they were
probably took for granted the tiny points of light that much closer—but stopped short of claiming to know
filled the nighttime sky, but they also must have won- what they were. And so these and other heavenly mys-
teries lingered until the turn of the 17th century, when have recently joined the astronomer’s repertoire of gear.
two historic events took place. Some are buried deep underground, quietly observing
It all started when an obscure German-Dutch spec- one of nature’s most elusive particles. Others sit on desk-
tacle maker put a pair of lenses together and saw that tops crunching numbers. All are devoted to decoding the
distant objects appeared closer. Although Hans Lipper- cryptic messages sent to us from the cosmos.
shey may have regarded his “looker” as an amusing toy, Having stopped briefly to examine the many tools
it was a Tuscan instrument maker who fi rst recognized used by astronomers, we are now ready to resume our
its potential for studying the heavens. It didn’t take Gali- outward journey, armed with a new appreciation of how
leo Galilei long to discover that the Moon was a nearby important it is to proceed carefully. Common sense is
world covered with craters, and that those “mysterious not enough. We must instead rely on our growing under-
moving points of light” were tiny disks of planets. Gali- standing of physical law and on rigorously tested predic-
leo’s observations marked a turning point. The telescope tions of carefully constructed theories. These also are
had changed forever our fundamental perception of the useful tools that will enable us to look beyond the surface
heavens. of spectacular vistas that otherwise would be devoid of
Throughout the four centuries since Galileo’s historic sense or meaning or connection. Kepler’s laws, Newton’s
discoveries, astronomers have continued to perfect the laws, Doppler shifts, Wien’s law, the Stefan-Boltzmann
astronomical telescope. Reaching far outside the vis- law, energy states, spectral lines, and the rest—these are
ible, telescopes now cover the entire electromagnetic the keys we will use to build an understanding of planets,
spectrum from gamma rays to radio waves. Some tele- stars, galaxies, and ultimately the universe itself. The fi rst
scopes reside on mountaintops; others make their home place in which we will bring these tools to bear will be
in Earth orbit or nearby space. Still others make the long our immediate neighborhood as we consider the nature
and arduous journey through interplanetary space to and origin of our Solar System. And to put our own Solar
observe Earth’s neighbors up close. But telescopes alone System in perspective, we’ll take a look at some of the
cannot provide all the answers. Tools of a different type many other planetary systems that lie far beyond.

• Optical telescopes come in two basic types: refractors questions, plus additional conceptual exercises. If
and reflectors. your instructor assigns questions in SmartWork, log in at
• All large astronomical telescopes are reflectors.
• Large telescopes collect more light and have greater
resolution.
Student Questions
• The CCD is today’s astronomical detector of choice. THINK ING A BOUT THE CONCEP T S
• Earth’s atmosphere blocks many spectral regions and 1. Galileo’s telescope used simple lenses. What is the pri-
distorts telescopic images. mary disadvantage of using a simple lens in a refracting
telescope?
• Putting telescopes in space solves problems created by
Earth’s atmosphere. 2. The largest astronomical refractor has an aperture of 1
meter. List several reasons why it would be impracti-
• Infrared and radio telescopes can see through vast clouds
cal to build a still larger refractor with, say, twice the
of cosmic gas and dust.
aperture.
• Radio and optical telescopes can be arrayed to greatly
3. Name and explain at least two advantages that reflect-
increase angular resolution.
ing telescopes have over refractors.
• Most of what we know about the classical planets comes
4. Your camera may have a zoom lens, ranging between
from spacecraft we have sent there.
wide angle (short focal length) and telephoto (long focal
• High-speed computers are essential to the acquisition, length). How would the size of an object in the camera’s
analysis, and interpretation of astronomical data. focal plane differ between wide angle and telephoto?
5. Optical telescopes reveal much about the nature of astro- planets. Explain the advantages and disadvantages of
nomical objects. Why do astronomers also need informa- each of these types of spacecraft.
tion provided by gamma-ray, X-ray, infrared, and radio
*23. If we have meteorites that are pieces of Mars, why is it
telescopes?
so important to go to Mars and bring back samples of
6. What optical property is responsible for refraction? the martian surface?
7. When reflecting from a flat surface, the angles of inci- 24. Humans had our fi rst look at the far side of the Moon
dence and reflection are the same. Is this also true for only as recently as 1959. Why had we not been able to
the curved surface of a reflecting telescope’s primary see it earlier—say, when Galileo fi rst observed the Moon
mirror? Explain with a sketch. with his telescope in 1610?
8. What causes a prism to break light into a spectrum? 25. Where are neutrino detectors located, and why are neu-
trinos so difficult to detect?
9. How do manufacturers of refracting telescopes and cam-
eras avoid the problem of chromatic aberration?
**10. Explain constructive and destructive interference.
11. Consider two optically perfect telescopes having differ- A P P LY I N G T H E C O N C E P T S
ent diameters but the same focal length. Is the image of
26. Many amateur astronomers start out with a 4-inch (aper-
a star larger or smaller in the focal plane of the larger
ture) telescope and then graduate to a 16-inch telescope.
telescope? Explain your answer.
How much fainter are the stars that can be seen in the
12. Name two ways in which Earth’s atmosphere interferes larger telescope?
with astronomical observations.
27. Compare the light-gathering power of a large astronomi-
13. “Twinkle, twinkle, little star. How I wonder what you cal telescope (aperture 10 meters) with that of the dark-
are.” Explain why stars twinkle. (In a later chapter we adapted human eye (aperture 7 mm).
will fi nd out what the stars are.)
28. Compare the angular resolution of the Hubble Space
14. Why do we not have groundbased gamma-ray and X-ray Telescope (aperture 2.4 meters) with that of a typical
telescopes? amateur telescope (aperture 20 cm).
*15. Explain adaptive optics and how they improve a tele- 29. Assume a telescope with an aperture of 1 meter. Compare
scope’s image quality. the telescope’s resolution when we are observing in the
near-infrared region of the spectrum (λ = 1,000 nm)
*16. Explain integration time and quantum efficiency and
with that when we are observing in the violet region of
how each contributes to the detection of faint astronomi-
the spectrum (λ = 400 nm).
cal objects.
*30. Assume that the maximum aperture of the human eye,
17. For more than a century, astronomers used photographic
D, is approximately 8 mm and the average wavelength
plates to record their observations. List at least two short-
of visible light, λ, is 5.5 × 10 –4 mm.
comings of the photographic process.
a. Calculate the diffraction limit of the human eye in
18. Charge-coupled devices are now the most commonly visible light.
used astronomical detector. List three advantages that b. How does the diffraction limit compare with the actual
CCDs have over photographic emulsions. resolution of 1–2 arcminutes (60–120 arcseconds)?
c. To what do you attribute the difference?
19. Explain a major advantage that radio telescopes have
over optical telescopes when we are observing the inner 31. The diameter of the full Moon’s image in the focal plane
parts of galaxies. of an average amateur’s telescope (focal length 1.5 meters)
is 13.8 mm. How big would the Moon’s image be in the
20. Why are the world’s largest telescopes located on high
focal plane of a very large astronomical telescope (focal
mountains?
length 250 meters)?
21. Some people believe that we put astronomical telescopes
*32. The dark-adapted aperture, integration time, and quan-
in orbit because doing so gets them closer to the objects
tum efficiency of the human eye are 7 mm, 100 ms, and
they are observing. As an enlightened student taking
10 percent, respectively. Compare the brightness of the
this class, you know better. Explain what is wrong with
faintest stars that can be seen with the naked eye with
this popular misconception.
those in a 1-hour-exposure photograph using a telescope
22. We have now sent various kinds of spacecraft—includ- having a 1-meter aperture and a photographic emulsion
ing flybys, orbiters, and landers—to all of the classical with a quantum efficiency of 2 percent.
33. One of the earliest astronomical CCDs had 160,000 pix- 38. Rovers Spirit and Opportunity can move across the
els, each recording 8 bits (256 levels of brightness). A new martian landscape at speeds of up to 5 centimeters per
generation of astronomical CCDs may contain a billion second (cm/s). In contrast, our typical walking speed is
pixels, each recording 15 bits (32,768 levels of bright- about 4 kilometers per hour (km/h).
ness). Compare the number of bits of data that each CCD a. How long would it take Spirit to move the length of a
type produces in a single image. football field (91.44 meters)?
b. How long would it take you to walk the same
34. Consider a CCD with a quantum efficiency of 80 percent
distance?
and a photographic plate with a quantum efficiency of
1 percent. If an exposure time of 1 hour is required to 39. Voyager 1 is now about 115 astronomical units (AU)
photograph a celestial object with a given telescope, how from Earth, continuing to record its environment as it
much observing time would we save by substituting a approaches the limits of our Solar System.
CCD for the photographic plate? a. What is the distance of Voyager 1, expressed in
kilometers?
*35. The VLBA employs an array of radio telescopes rang-
b. How long does it take observational data to get back
ing across 8,000 km of Earth’s surface from the Virgin
to us from Voyager 1?
Islands to Hawaii.
c. How does Voyager 1’s distance from Earth compare
a. Calculate the angular resolution of the array when
with that of the nearest star (other than the Sun)?
radio astronomers are observing interstellar water
molecules at a microwave wavelength of 1.35 cm. 40. As with light, the speed, wavelength, and frequency of
b. How does this resolution compare with the angular gravitational waves are related as c = λ × f. If we were
resolution of two large optical telescopes separated to observe a gravitational wave from a distant cosmic
by 100 meters and operating as an interferometer at event with a frequency of 10 hertz (Hz), what would be
a visible wavelength of 550 nm? the wavelength of the gravitational wave?
36. When operational, the SVLBI may have a baseline of
100,000 km. What will be the angular resolution when
we are studying an interstellar molecule emitting at a
wavelength of 17 mm from a distant galaxy?
37. The Mars Reconnaissance Orbiter (MRO) flies at an aver- ing outlines, vocabulary flashcards, and multiple-choice
age altitude of 280 km above the martian surface. If its quizzes, plus links to premium content in SmartWork and
cameras have an angular resolution of 0.2 arcsecond, the ebook. Visit www.wwnorton.com/studyspace.
describe the size of the smallest objects that the MRO
can detect on the martian surface?
PA R T II The Solar System
You are a child of the universe, no less than the trees

and the stars; you have a right to be here. And whether
or not it is clear to you, no doubt the universe is
unfolding as it should.
MA X EHRMANN (1872–1945), DESIDERATA
From clouds of gas and dust, solar systems such as ours are born.
CHAP TER 6
The Birth and Evolution

of Planetary Systems
6.1 Stars Form and
K E Y CONCEP T S
Planets Are Born By the early years of the 21st century, astronomers
In thinking about how it all began, a good place to start is had come to see our Solar System as an unmistakable
right here in our own small corner of the universe—a collec- by-product of the birth of the Sun. The discovery of
tion of planets,1 moons, and planetary systems surrounding other stars has shown
other smaller bodies surround- Our Solar System that there is nothing unique about our Solar System.
ing an ordinary star that we is tiny compared It is this understanding that brings order to what we
call the Sun. Astronomers to the universe. see around us today. We begin our investigation of the
refer to such a system of plan-
Sun’s family by introducing the story of the formation of
ets surrounding a star as a planetary system, and we call our
planetary systems—a story in which we will learn that
own planetary system the Solar System (Figure 6.1). But first,
a word of caution. Do not confuse our Solar System with “the • A star forms when a cloud of interstellar gas and dust
universe.” Planetary systems are infinitesimally small com- collapses under its own weight.
pared to the universe as a whole. For example, light takes • While a star forms it is surrounded by a flat, rotating
about 4 hours to travel here from Neptune, the outermost disk that provides the raw material from which a
classical planet in our Solar System. Light from the most planetary system might form.
distant galaxies takes nearly 14 billion years to reach us!
For more than a decade astronomers have realized that • Dust grains in the disk around a young star stick
our Solar System is but one of an enormous number of other together to form larger and larger solid objects.
planetary systems scattered throughout the galaxy. It would • Temperature differences within the disk determine the
make sense, then, to begin this chapter by learning how kinds of materials from which solid objects can form.
these planetary systems, including our own Solar System, • Planets orbit the Sun in a plane, and they revolve in the
were created. Later, in Part IV, we will learn about a much
same direction that the Sun rotates.
vaster universe, and how it came to be.
Until the last part of the 20th century, every discussion • Giant planets form when solid, planet-sized bodies
of the Solar System necessarily started with an account- capture gas from the surrounding disk.
ing of its pieces: “There are planets with such and such • The atmospheres of smaller terrestrial planets are
properties. There are moons, there are asteroids, and there gases released by volcanism and volatile materials that
arrived on comets.
1
The formal meaning of the word planet, as defined by the International
• Planetary systems around other stars are common.
Astronomical Union, appears in Appendix 8. Curiously, the IAU defi-
nition applies only to planets within our own Solar System.
155
156 Chapter 6 The Birth and Evolution of Planetary Systems
Sun Mercury Venus Earth Mars Jupiter
Ceres
Asteroid
Belt
Kuiper
Belt
Saturn Uranus Neptune Pluto
FIGURE 6.1 Our Solar System. There are three other dwarf planets besides Pluto and Ceres that are not shown here.
are comets. And so forth.” The origin of the Solar System, Modern theory of planetary system formation suggests
though, remained speculative. Yet, in the last few decades that, when conditions are right, clouds of interstellar gas
stellar astronomers studying the formation of stars and collapse under the force of
planetary scientists analyzing clues about the history of the their own self-gravity (see Young stars are
Solar System have found themselves arriving at the same Chapter 10) to form stars. In surrounded by
picture of our early Solar System—but from two very dif- support of the nebular hypoth- rotating disks.
ferent directions. This unified understanding provides the esis, disks of gas and dust
foundation for the way we now think about the Sun and the have been found surrounding young stellar objects like the
myriad objects that orbit it. ones shown in Figure 6.2. From this observational evidence,
The fi rst plausible theory for the formation of the Solar stellar astronomers have shown that, much like a spinning
System, known as the nebular hypothesis, was proposed as ball of pizza dough spreads out to form a flat crust, the cloud
early as 1734 by the German philosopher Immanuel Kant that produces a star—our Sun, for example—collapses fi rst
(1724–1804) and conceived independently a few years later into a rotating disk. Material in the disk eventually suffers
by the French astronomer Pierre-Simon Laplace (1749–1827). one of three fates: it travels inward onto the forming star
Kant and Laplace argued that a rotating cloud of interstellar at its center, remains in the disk itself to form the planets
gas, or nebula (Latin for “cloud”), gradually collapsed and and other objects, or is thrown back into interstellar space.
flattened to form a disk with the Sun at its center. Surround- AstroTour: Solar System Formation
ing the Sun were rings of material from which the planets During the same years that astronomers were beginning
formed. Although the nebular hypothesis remained popu- to ferret out the secrets of star formation, other groups of
lar throughout the 19th century, it did have serious prob- scientists with very different backgrounds—mainly geo-
lems, as we will soon see. Even so, the basic principles of chemists and geologists—were piecing together the history
the hypothesis are retained today in our modern theory of of our Solar System. Planetary scientists looking at the cur-
planetary system formation.2 rent structure of the Solar System inferred what some of its
early characteristics must have been. The orbits of all the
2
In the early 20th century a competing hypothesis, known as the “close planets in the Solar System lie very close to a single plane,
encounter theory,” claimed that planets formed when blobs of material which says that the early Solar System must have been flat.
were pulled off the Sun by a close encounter with a passing star. The fact that all the planets orbit the Sun in the same direc-
6.1 Stars Form and Planets Are Born 157
(a) (b)
FIGURE 6. 2 Hubble Space
Telescope images showing
disks around newly formed
Jet stars. The dark bands are the
shadows of the disks seen
more or less edge on. Bright
Starlight reflected regions are dust illuminated by
from disk starlight. Some disk material
may be expelled in a direction
perpendicular to the plane of
the disk in the form of violent
jets.
Silhouette Protostar is
of disk hidden by disk
(c) (d)
FIGURE 6. 3 Meteorites
are the surviving pieces of
young Solar System fragments
that land on the surfaces
of planets. It is clear from
this cross section that this
meteorite formed from many
smaller components that stuck
together.
tion says that the material from which the planets formed
must have been swirling about the Sun in the same direc-
tion as well. Other clues about what the early Solar System
was like were harder to puzzle out. Meteorites, for example,
include bits and pieces of material that are left over from
the Solar System’s youth. These fragments of the early Solar
System can be captured by Earth’s gravity and fall to the
ground, where they can be picked up and studied. Many
meteorites, such as the one in Figure 6.3, look something
like a piece of concrete in which pebbles and sand are mixed
with a much fi ner filler. This structure is surely telling us
something about how these pieces of interplanetary debris
formed, but what?
Beginning in the 1960s, a flood of information about
Earth and other objects in the retical calculations. Scientists working with this wealth
Solar System poured in from The Solar System of information began to see a pattern. What they were learn-
a host of sources, including formed from a ing made sense only if they assumed that the larger bodies
space probes, groundbased rotating disk of gas in the Solar System had grown from the aggregation of
telescopes, laboratory analy- and dust. smaller bodies. Following this chain of thought back in
sis of meteorites, and theo- time, they came to envision an early Solar System in which
the young Sun was surrounded by a flattened disk of both interstellar space. The Sun was not yet a star in the true
gaseous and solid material. This swirling disk of gas and sense of the word. It was still a protostar—a large, hot ball
dust provided the raw material from which the objects in of gas whose nuclear fi res had yet to ignite. (The often-used
our Solar System would later form. prefi x proto- means “early form” or “in the process of for-
While reading the previous paragraphs, you may have mation.”) As the cloud of interstellar gas collapsed to form
noticed a remarkable similarity between the disks that stel- the protostar, its gravitational energy was converted into
lar astronomers fi nd around young stars and the disk that thermal energy and radiation. Surrounding the protostel-
planetary scientists hypothesize as the cradle of the Solar lar Sun was a flat, rotating disk of gas and dust. Orbiting
System. This similarity is not a coincidence. As astronomers is perhaps a better word than rotating because each bit
and planetary scientists compared notes, they realized they of the material in this thin disk was orbiting around the
had arrived at the same picture of the early Solar System Sun according to the same laws of motion and gravita-
from two completely different directions. The rotating disk tion that govern the orbits of the planets today. The disk
from which the planets formed was none other than the around the Sun was much like the disks that astronomers
remains of the disk that had accompanied the formation see today surrounding protostars and newly formed stars
of the Sun. The planet we live on, along with all the other elsewhere in our galaxy. This disk is referred to as a pro-
orbiting bodies that make up our Solar System, evolved toplanetary disk, sometimes abbreviated as proplyd. It
from the remnants of the interstellar cloud that collapsed probably contained less than 1 percent as much mass as
to form our local star, the Sun. The connection between the the nascent star at its center, but this amount was more
formation of stars and the origin and subsequent evolution than enough to account for the bodies that make up the
of the Solar System has become one of the cornerstones of Solar System today.
both astronomy and planetary science—a central theme Okay, so we know the Solar System formed from a proto-
around which a great deal of our understanding of our Solar planetary disk and that disks are seen around newly formed
System revolves. stars. But why is this the case? What is it about the process
of star formation that leads not only to a star itself, but to a
flat, orbiting collection of gas and dust as well? The answer
to this question lies with a skater spinning on the ice and
6.2 In the Beginning something called angular momentum.
Was a Disk
Later in the book we will turn our attention beyond the
The Collapsing Cloud Rotates
boundaries of our Solar System to the process of star for- You have likely seen a figure-skater spinning on the ice
mation itself. For now it is enough to jump into this story (Figure 6.5). Like any rotating object or isolated group of
of star formation midstream. Begin by planting the picture objects, the spinning ice-skater has some amount of angu-
shown in Figure 6.4 fi rmly in your mind. It presents the lar momentum. The amount of angular momentum that an
newly formed Sun, roughly 5 billion years ago, adrift in object possesses depends on three things:
FIGURE 6. 4 When you consider the young Sun, think of it as being surrounded by a flat, rotating
disk of gas and dust that is flared at its outer edge.
Protostellar Sun Protoplanetary disk

VISUAL ANALOGY This statement is referred to as the law of conservation of

FIGURE 6. 5 A angular momentum. In the parlance of physics, if we say
figure-skater relies that something is “conserved,”
on the principle of we mean that the quantity Angular momentum
conservation of angular
does not change by itself. This is conserved.
momentum to change
idea might remind you of
the speed with which
she spins.
Newton’s fi rst law of motion, which says that in the absence
of an external force, an object continues to move in a straight
line at a constant speed. Indeed, both Newton’s first law and
the conservation of angular momentum are examples of
conservation laws. There are many other conservation laws
in physics, including the law of conservation of momentum,
the law of conservation of energy (discussed later in this
chapter), and the law of conservation of electric charge.
Both the ice-skater and the collapsing interstellar cloud
are affected by conservation of angular momentum. You
have probably noticed that an ice-skater can control how
rapidly she spins by doing nothing other than pulling in or
extending her arms or legs. A compact object must spin more
rapidly to have the same amount of angular momentum as
a more extended object with the same mass. As our skater
spins, her angular momentum does not change much. (The
slow decrease is due to friction, an external force.) When
her arms and leg are fully extended, she spins slowly; but
as she pulls her arms and leg in, she spins faster and faster.
When her arms are held tightly in front of her and one leg
is wrapped around the other, the skater’s spin becomes a
blur. She fi nishes with a flourish by throwing her arms and
leg out—an action that abruptly slows her spin. Despite
1. How fast the object is rotating. The faster an object is the dramatic effect, her angular momentum remains the
rotating, the more angular momentum it has. A top that same throughout. This impressive athletic spectacle comes
is spinning rapidly has more angular momentum than courtesy of the law of conservation of angular momentum
the same top does when it is spinning slowly. and from the difference between an extended object and a
compact object.
2. The mass of the object. Suppose you compare two spin-
Now we turn our attention to how conservation of angu-
ning tops. Both tops have the same size, shape, and rate
lar momentum affected our protostellar Sun. Remember that
of spin. They are the same except that one top is made
its cloud of interstellar gas is
of metal while the other top is made of wood. The more
collapsing under the force of Interstellar clouds
massive metal top has more angular momentum.
its own gravity. It might seem have far more angular
3. How the mass of the object is distributed—how “spread most natural for the cloud to momentum than the
out” the object is. For an object of a given mass and rate collapse directly into a ball— stars they form have.
of rotation, the more spread out the object is, the more and so it would, but for the
angular momentum it has. An object that is rotating cloud’s own angular momentum. Interstellar clouds are truly
slowly but is very spread out might have more angular vast objects, light-years in size. (Recall that a light-year is
momentum than a more rapidly rotating but more com- the distance traveled by light in 1 year, or about 9.5 trillion
pact object has. kilometers [km].) As interstellar clouds orbit about the gal-
axy’s center, they are constantly being pushed around by
It is quite a bit more complicated to determine the rotational stellar explosions or by collisions with other interstellar
or spin angular momentum of a single object, such as the clouds. This constant “stirring” guarantees that all interstel-
spinning top just referred to or a rotating planet or interstellar clouds will have some amount of rotation. As spread out
lar gas cloud. See Math Tools 6.1 for details. as an interstellar cloud is, even a tiny amount of rotation
What makes angular momentum such an important and corresponds to a huge amount of angular momentum. Imag-
useful idea in physics and astronomy is that the amount of ine our ice-skater now with arms that reach from here to the
angular momentum possessed by an object or isolated group other side of Earth. Even if she were rotating very slowly at
of objects does not change unless those objects are affected first, think how fast she would be spinning by the time she
in just the right way by something other than themselves. pulled those long arms to her sides!
Angular Momentum
We will encounter angular momentum in its various forms is far more complicated. Here we must add up the individual
many times during our journey. In its simplest form we fi nd angular momenta of every tiny mass element within the body.
that the angular momentum of a system is given by The mathematical procedure for doing this is beyond the level
used in this book. However, we can look at one example. In the
L = m × v × r, case of a uniform sphere, the spin angular momentum is pro-
portional to the square of its radius
where m is the mass, v is the speed with which the mass is
4πmR2
moving, and r represents how spread out the mass is. Ls = _______,
5P
As an example, we can apply this relationship to the orbital
angular momentum, L o, of the Moon in its orbit about Earth. where R is the radius of the sphere and P is the rotation period
Here, the mass (m) of the Moon is 7.35 × 1022 kilograms (kg), of its spin.
the speed of the Moon in orbit (v) is 1.023 × 103 meters per sec- Let’s compare the Moon’s spin angular momentum with
ond (m/s), and the radius of the Moon’s orbit (r) is 3.84 × 108 its orbital momentum about Earth. (Note that, in this and the
meters. Putting this together, we have following example, we make the unrealistic assumption that
the Moon and Earth are uniform spheres.) The radius of the
Lo = (7.35 × 1022)(1.023 × 103)(3.84 × 108) Moon is 1.737 × 10 6 meters and its rotation period is 27.3 days
= 2.89 × 1034 kg m2/s = 2.36 × 106 seconds, the same as its orbital period about Earth.
This gives us
We can compare this with the orbital angular momentum of
4π × (7.35 × 1022)(1.737 × 106)2
Earth in its orbit about the Sun. Earth’s mass is 5.97 × 1024 kg, Ls = ___________________________ = 2.36 × 1029 kg m2/s.
5 × (2.36 × 106)
its orbital speed is 2.98 × 104 m/s, and its orbital radius is 1.496
× 1011 meters. This gives us We can see that the Moon’s orbital angular momentum is about
100,000 times greater than its spin angular momentum. We can
Lo = (5.97 × 1024)(2.98 × 104)(1.496 × 1011) make the same comparison with Earth, where Earth’s average
= 2.66 × 1040 kg m2/s. radius is 6.37 × 106 meters and its rotation period is 86,400
seconds.
Earth in its orbit about the Sun has about a million times more
4π × (5.97 × 1024)(6.37 × 106)2
orbital angular momentum than the Moon has in its orbit Ls = __________________________ = 7.07 × 1033 kg m2/s.
5 × (8.64 × 104)
about Earth.
Calculating the spin angular momentum of a spinning Earth’s orbital angular momentum is nearly 4 million times its
object, such as a top, a planet, or an interstellar galactic cloud, spin angular momentum.
Just as the ice-skater speeds up when she pulls in her the Sun’s self-gravity would have to be almost 200 million
arms, the cloud of interstellar gas rotates faster and faster times stronger to hold the Sun together!
as it collapses. However, there is a puzzle here. Suppose, Recall our mentioning in the previous section that there
for example, that we tried to form the Sun from a uniform were serious problems with the Kant-Laplace hypothesis.
cloud that was about a light-year across—say, 1016 meters— What the conservation of angular momentum seems to be
and was rotating so slowly that it took a million years to saying is that the nebular hypothesis does not work—that
complete one rotation. As we can see in Math Tools 6.1, stars cannot form from collapsing interstellar clouds. Yet
for a collapsing sphere to conserve spin angular momen- there is no other way for stars to form.
tum (L s = constant), its rotation period (P) must be pro-
portional to the square of its radius (R), and its spin rate
(∝ 1/P) must therefore be inversely proportional to the
square of its radius. So, by the time such a cloud collapsed
An Accretion Disk Forms
to the size of our Sun—a mere 1.4 × 109 meters across, or When scientists fi nd what appears to be a contradiction—
only one 10-millionth the size of the original cloud—it like the apparent contradiction between the principle of
would be spinning 50 trillion times faster, completing a conservation of angular momentum and the fact that a star
rotation in only 0.6 second! This is over 3 million times has far less angular momentum than the cloud that formed
faster than our Sun actually spins. At this rate of rotation, it—there is cause for great excitement. Such seeming con-
The key to solving the riddle of angular momentum in

Axis of rotation a collapsing interstellar cloud lies in realizing that the direc-
1 As a slowly rotating tion of the collapse is impor-
interstellar cloud tant. The cloud’s rotation may The cloud collapses
collapses because of thwart the collapse of the into a disk rather than
gravity, it rotates faster.
cloud toward its axis of rota- directly into a star.
tion, but there is nothing to
prevent collapse taking place parallel to the axis of rotation.
Thus, rotation slows the collapse perpendicular to, but not
parallel to, the axis of rotation (Figure 6.6). Instead of col-
lapsing into a ball, the interstellar cloud flattens into a disk.
As the cloud collapses, the strength of gravity causing the
collapse becomes ever greater. Eventually the flattening
cloud reaches a point where the inner parts of the cloud
begin to fall freely inward, raining down on the growing
object at the center. As this happens, the outer portions of
the cloud lose the support of the collapsed inner portion,
and they start falling inward too. The whole cloud collapses
inward, much like a house of cards with the bottom layer
knocked out. As this material makes its final inward plunge,
2 Rotation slows collapse
perpendicular to but not it lands on a thin, rotating structure called an accretion
parallel to the axis, so disk. The accretion disk serves as a way station for mate-
the cloud flattens. rial on its way to becoming part of the star that is forming
at its center.
The formation of accretion disks is common in astron-
omy, so it is worth taking a moment to think carefully about
this process. We can use what we learned about orbits in
Chapter 3 to better understand what happens during this
fi nal stage of the collapse of an interstellar cloud. As the
material falls toward the forming star, it travels on curved,
almost always elliptical, paths, just as Kepler’s laws say
it should. These orbits would carry the material around
the forming star and back into interstellar space, except
3 Eventually the cloud for one problem: the path inward toward the forming star
collapses from the inside is a one-way street. When material nears the center of the
out, and an accretion cloud, it runs headlong into material that is falling in from
disk and protostar form.
the other side. Material falling onto the disk from oppo-
site directions comes together at the plane of the accre-
tion disk, which is the plane perpendicular to the cloud’s
axis of rotation.
Protostar Accretion disk
To understand how interstellar material collects on the
accretion disk, a visual analogy might be helpful. Imagine
a traffic engineer’s worst nightmare: a huge roundabout,
or traffic circle, with multiple entrances but with all exits
blocked by incoming traffic (Figure 6.7). AstroTour:
Traffic Circle. As traffic flows into the traffic circle, it has
nowhere else to go, resulting in a continuous, growing line
of traffic driving around and around in an ever-tightening
FIGURE 6.6 A rotating interstellar cloud collapses in
circle. Eventually, as more and more cars try to pack in,
a direction parallel to its axis of rotation, thus forming an
the traffic piles up. This situation is roughly analogous to
accretion disk.
an accretion disk. Of course, keep in mind that traffic in
a roundabout moves on a flat surface, whereas the accre-
tradictions do not mean that nature is breaking the rules. tion disk around a protostar forms from material coming in
(Nature never breaks its own rules!) Instead it means that from all directions in three-dimensional space. As mate-
our understanding of what is going on is incomplete or that rial falls onto the disk, its motion perpendicular to the disk
we have the wrong rule. It means that we have found a place stops abruptly, but its mass motion parallel to the surface of
where there are new things to be learned. the disk contributes to and adds to the disk’s total angular
An accretion disk is momentum. In this way, the angular momentum of the

(a) like a traffic circle with infalling material is transferred to the accretion disk. What
no exits. is most important here is that the rotating accretion disk
has a radius of hundreds of astronomical units and is thou-
sands of times greater than the radius of the star that will
eventually form at its center. Therefore, most of the angu-
lar momentum in the original interstellar cloud ends up in
the accretion disk rather than in the central protostar. The
problem of the superrotating Sun has been solved.
The next question concerns how material falling onto the
accretion disk finds its way inward onto the growing proto-
star. We will pick up this discussion in Chapter 15, when we
return to follow the story of the growth and evolution of the
star at the center of the disk. For now it is enough to know
that most of the matter that lands on the accretion disk either
ends up as part of the star or is ejected back into interstellar
space, sometimes in the form of violent jets, as seen in Figure
6.2a. However, a small amount of material is left behind in
the disk. It is this leftover disk—the dregs of the process of
star formation—to which we next turn our attention.
Before going on with our story, however, we should
stress that theoretical calculations by astronomers have
long predicted that accretion disks should be found around
1 Material rains down young stars, and illustrations like Figure 6.4 have been a
from the collapsing, mainstay in textbooks for years. In the 1990s, however,
(b)
rotating cloud.
these cartoon drawings had to make room for images of the
real thing. Figure 6.2 shows Hubble Space Telescope (HST)
images of edge-on accretion disks around young stars. The
dark bands are the shadows of the edge-on disks, the top
and the bottom of which are illuminated by light from the
forming star. It would be nice if we could go back 5 billion
years and watch as our own Sun formed from a cloud of
interstellar gas, but we do not have to. All we have to do is
look at objects like the ones in Figure 6.2 to know what we
2 The vertical motion of material would have seen.
from above cancels the vertical
motion from below…
Small Objects Stick Together

to Become Large Objects
The chain of events that connects the accretion disk around
a young star to a planetary system such as our own begins
with random motions of the
gas within the protoplanetary Gas motions push
disk. These motions push the small particles into
smaller grains of solid mate- larger particles.
3 …but the rotational motion rial back and forth past larger
remains. The material joins grains; and as this happens, the smaller grains stick to the
the rotating accretion disk.
larger grains. The “sticking” process among smaller grains
VISUAL ANALOGY FIGURE 6.7 (a) Traffic piles up in a is due to the same static electricity that causes dust and hair
traffic circle with entrances but no exits. (b) Similarly, to cling to plastic surfaces. Starting out at only a few microns
gas from a rotating cloud falls inward from opposite (micrometers) across, the slightly larger bits of dust grow to
sides, piling up onto a rotating disk. the size of pebbles and then to clumps the size of boulders,
which are less susceptible to being pushed around by gas
(Figure 6.8). Astronomers believe that when clumps grow
to about 100 meters across, their rate of growth slows down.
1 Gas motions in a protoplanetary The gravity of a growing planetesimal

disk blow small particles around draws in additional material.
more easily than large particles.
Craters
Gas motion
2 Small particles are blown

into larger particles…
FIGURE 6.9 The gravity of a planetesimal is strong enough to

attract surrounding material, which causes the planetesimal to
grow.
In order for two such clumps to stick together, they must

bump into each other gently—very gently. Otherwise the
energy of collision would cause the two colliding bodies to
3 …forming larger and fragment into many smaller pieces instead of forming a single
larger aggregations. larger one. Collision speeds cannot be much greater than 0.1
m/s for colliding boulders to stick together. If you were to
walk that slowly, it would take you 15 minutes to travel the
length of a football field. In a real accretion disk, collisions
more violent than this certainly happen on occasion, break-
ing these clumps back into smaller pieces. Likely there are
many reversals in this growth of clumps of boulders.
Up to this point, larger objects have grown mainly by
“sweeping up” smaller objects that run into them or that get
in their way. As the clumps reach the size of about a kilome-
FIGURE 6.8 Motions of gas in a protoplanetary disk blow ter, a different process becomes important. These kilometer-
smaller particles of dust into larger particles, making the larger sized objects, now called
particles larger still. This process continues, eventually creating planetesimals (literally “tiny Gravity helps
objects many meters in size.
planets”), are massive enough planetesimals grow
that their gravity begins to into planets.
dominate, exerting a signifi-
These large objects are so few and far between in the disk cant attraction on nearby bodies (Figure 6.9). No longer is
that chance collisions become less and less frequent. Even growth of the planetesimal fed only by chance collisions
so, the process of growth continues at a slower pace as 100- with other objects; the planetesimal’s gravity can now pull
meter clumps join together to produce still larger bodies. in and capture other smaller planetesimals that lie outside
In matters of social dynamics we sometimes hear that “the its direct path. The growth of planetesimals speeds up, with
rich get richer at the expense of the poor.” Although we can the larger planetesimals quickly consuming most of the
always debate this social axiom, the principle certainly remaining bodies in the vicinity of their orbits. The final
holds true when it comes to the dynamics within a proto- survivors of this process are now large enough to be called
planetary disk: the larger dust grains get larger at the planets. As with the major bodies in orbit about the Sun,
expense of the smaller grains. some of the planets may be small and others quite large.
be the rest of the room that was whizzing by.) The atoms and
6.3 The Inner Disk Is molecules in the gas falling toward the protostar are like
these marbles. They are picking up speed as they fall as a
Hot, but the Outer group toward the protostar, but the gas is still cold because
the random thermal velocities of atoms and molecules with
Disk Is Cold respect to each other are still low. Now imagine what hap-
pens when the marbles hit the rough floor. They bounce
The accretion disks surrounding young stars form from every which way. They are still moving rapidly, but they
interstellar material that may have a temperature of only a are no longer moving together. A change has taken place
few kelvins, but the disks themselves reach temperatures from the ordered motion of marbles falling together to the
of hundreds of kelvins or more. What is it that heats up random motions of marbles traveling in all directions. The
the disk around a forming star? The answer lies with our atoms and molecules in the gas falling toward the central
old friend gravity. Material from the collapsing interstellar star behave in the same fashion when they hit the accre-
cloud falls inward toward the protostar, but because of its tion disk. They are no longer moving as a group, but their
angular momentum it “misses” the protostar and instead random “thermal” velocities are now very large. The gas
falls onto the surface of the disk. When this material reaches is now hot.
the surface of the disk, its infalling motion comes to an Another way to think about why the gas that falls on
abrupt halt, and the velocity that the atoms and molecules the disk makes the disk hot is to apply another conserva-
in the gas had before hitting the disk is suddenly converted tion law, the concept of conservation of energy. The law of
into random thermal velocities instead. That is to say, the conservation of energy states that, unless energy is added
cold gas that was falling toward the disk gets very hot when to or taken away from a system from the outside, the total
it lands on the disk. amount of energy in the system must remain constant. But
To help you visualize this, imagine dumping a box of the form the energy takes can change.
marbles from the top of a tall ladder onto a rough, hard floor Imagine lifting a heavy object—say, a brick. It is hard to
below (Figure 6.10). The marbles fall, picking up speed as do because you are working against gravity. It takes energy
they go. Even though the marbles are speeding up, however, to lift the brick, and conservation of energy says that energy
they are all speeding up together. As far as one marble is is never lost. But where does that energy go? It is changed
concerned, the other marbles are not moving very fast at all. into a form called gravitational potential energy. In a sense,
(If you were riding on one of the marbles, the other marbles this energy is “stored” in a way that is reminiscent of how
would not appear to you to be moving very much; it would energy is stored in a battery. Potential energy is energy that
VISUAL ANALOGY FIGURE 6.10 (a) Marbles dropped as a group fall together until they hit a rough
floor, at which point their motions become randomized. (b) Similarly, atoms in a gas fall together until
they hit the accretion disk, at which point their motions become randomized, raising the temperature
of the gas.
(a) (b)
Marbles falling together are not A cloud of gas falls toward an accretion
moving much relative to each disk. The atoms are not moving much
other. The group is “cold.” relative to each other. The gas is cold.
Accretion disk
After bouncing on a rough surface the When the cloud hits the disk
marbles are moving rapidly relative to the motion of the atoms is
each other. The group is “hot.” randomized. The gas is now hot.
6.3 The Inner Disk Is Hot, but the Outer Disk Is Cold 165
“has potential”—it is waiting to show up in a more obvious gas falls toward the disk surrounding a protostar, gravita-
form. If you drop the brick it falls, and as it falls it speeds tional potential energy is converted fi rst to kinetic energy,
up. The gravitational potential energy that was stored is causing the gas to pick up speed. When the gas hits the disk
being converted instead to and stops suddenly, that kinetic energy is turned into ther-
energy of motion, which, as The gravitational mal energy. (Foundations 6.1 discusses why it can be useful
you may recall from Chapter energy of infalling to think about the same thing—in this case, energy—in dif-
4, is called kinetic energy. material turns into ferent ways.)
When the brick hits the floor, thermal energy. In this way, as material falls onto the accretion disk
it stops suddenly. The brick around a forming star, the disk heats up. But the amount of
loses its energy of motion, so what form does this energy heating depends on where the material hits the disk. Mate-
take now? If the brick cracks, part of the energy goes into rial hitting the inner part of the disk (which we will call
breaking the chemical bonds that hold it together. Some of the “inner disk”) has fallen
the energy is converted into the sound the brick makes when farther and picked up greater The inner disk
it hits the floor. Some goes into heating and distorting the speed within the gravitational is hotter than
floor. But most of the energy is converted into thermal field of the forming star than the outer disk.
energy. The atoms and molecules that make up the brick has material hitting the disk
are moving about within the brick a bit faster than they farther out. Like a rock dropped from a tall building, mate-
were before the brick hit; so the brick and its surroundings, rial hitting the inner disk is moving quite rapidly when it
including the floor, grow a tiny bit warmer. Similarly, as hits, so it heats the inner disk to high temperatures. In con-
Thinking about Energy in Different Ways

In Section 6.3 we discuss how the gas falling onto a protostellar bines in a hydroelectric generator, it turns the generator and
disk heats the disk. First we present the analogy of marbles fall- produces electric energy that is carried out over power lines
ing on a floor, noting that the motion of marbles is jumbled up and lights up the lights in your home. Where did the energy
when they hit, like the atoms in the gas hitting the disk. As the to light your lights or heat the water that pours from your tap
correlated motion of gas molecules becomes randomized, the come from? If you have an electric water heater and get your
temperature of the gas increases. We then approach the expla- power from a hydroelectric plant, it comes from the gravita-
nation from a different perspective, focusing on how energy tional potential energy of the water in a reservoir near you—
is conserved but changes its form from gravitational potential just as the energy to heat the newly formed Sun came from
energy to kinetic energy and fi nally to thermal energy. These the gravitational potential energy of the reservoir of gas from
may be two different ways of explaining why the disk gets hot, which the Sun formed.
but they are not two different reasons that the disk gets hot. Conservation of energy is a powerful way to think about the
The disk heats up as a result of the same physical process— heating of disks around young stars because it enables astrono-
the “reason” is the same regardless of the words we may use to mers to calculate how hot these disks get. If we know the mass
describe it. Instead, these two explanations offer two different of the protostar and the surrounding disk, and how far away
ways to think about why the disk gets hot. from the protostar the gas is falling from, we can calculate how
Both ways of thinking about the process are included in the much gravitational energy the gas started out with. According
discussion because sometimes students—or scientists—need to the law of conservation of energy, this gravitational energy
to look at the same thing from several different angles before must eventually be converted to thermal energy, allowing
they understand it. In this case, however, most scientists would us to calculate the expected temperature of the protostar. So
agree that the second way of thinking about the problem is without even determining how the disk formed, we can say
far more powerful than the fi rst. The idea of energy changing right off the bat how much thermal energy will be deposited
forms connects how disks are heated with a wide variety of onto the disk.
other phenomena. Properly stated, the heating of protostellar Scientists spend much of their time trying to come up with
disks is an example of one of the most far-reaching patterns in new ways of thinking about problems, looking for particularly
nature: the conservation of energy. powerful ways that point to new insights and discoveries. The
Once you understand the different forms of energy and how most powerful means of thinking about a problem usually tie
energy is conserved, you begin to see this pattern of nature the problem to ever-grander patterns in nature. Conservation of
everywhere. For example, when water falls through the tur- energy is one of the grandest and most useful patterns around.
P FIGURE 6.11 Differences

Inner disk Outer disk
S VEM J S U in temperature within a
protoplanetary disk determine
Protoplanetary disk the composition of dust
grains that then evolve into
planetesimals and planets.
Temperature decreases with distance from the protostar
Shown here are the protostar
Refractory materials (PS) and the orbits of Venus (V),
Earth (E), Mars (M), Jupiter (J),
Saturn (S), and Uranus (U).
Water ice
Highly volatile materials
trast, material falling onto the outer part of the disk (which only if their temperature is quite low. These less refractory
we will call the “outer disk”) is moving much more slowly substances are called volatile materials (or “volatiles” for
(envision a rock dropped from only a foot or so), leaving the short). Astronomers generally refer to the solid form of any
outermost parts of the disk at temperatures that may be not volatile material as an ice.3
much higher than the original interstellar cloud. Stated Differences in temperature from place to place within
another way, material falling onto the inner disk converts the disk have a significant effect on the makeup of the dust
more gravitational potential energy into thermal energy grains in the disk (Figure 6.11). In the hottest parts of the
than does material falling onto the outer disk. disk (closest to the protostar), only the most refractory sub-
The energy released as material falls onto the disk is stances can exist in solid form. In the inner disk, dust grains
not the only source of thermal energy in the disk. Even are composed of refractory
before the nuclear fi res that will one day power the new materials only.4 Somewhat Volatile ices survive
star have ignited, conversion of gravitational energy into farther out in the disk, some in the outer disk,
thermal energy drives the temperature at the surface of the hardier volatiles, such as but only refractory
protostar to several thousand kelvins, and it also drives the water ice and certain organic5 solids survive in
luminosity of the huge ball of glowing gas to many times substances, can survive in the inner disk.
the luminosity of the present-day Sun. For the same reasons solid form, adding to the mate-
that Mercury is hot while Pluto is not (see Section 4.5), the rials that make up dust grains. Highly volatile components
radiation streaming outward from the protostar at the center such as methane, ammonia, and carbon monoxide ices and
of the disk drives the temperature in the inner parts of the some organic molecules survive in solid form only in the
disk even higher, increasing the difference in temperature coldest, outermost parts of the accretion disk, far from the
between the inner and outer parts of the disk. central protostar. The differences in composition of dust
grains within the disk are reflected in the composition of
the planetesimals formed from that dust. Planets that form
closest to the central star tend to be made up mostly of
Rock, Metal, and Ice refractory materials such as rock and metals. Those that
We have all noticed that temperature affects which materi- form farthest from the central star also contain refractory
als can and cannot exist in a solid form. On a hot summer
day, ice melts and water
quickly evaporates, whereas Refractory materials
3
on a cold winter night even remain solid even at In general astronomical usage, the term ice is used to describe the
the water in our breath freezes high temperatures. volatile itself, regardless of whether it happens to be in solid, liquid,
or gaseous form.
into tiny ice crystals before 4
Some geochemists believe that certain volatiles, such as water, did
our eyes. Some materials, such as iron, silicates (minerals
survive in the hot inner disk because they were bound chemically
containing silicon and oxygen), and carbon—rocky materi-
to refractory materials and were thus able to withstand the high
als and metals—remain solid even at quite high tempera- temperatures.
tures. Such materials, which are capable of withstanding 5
The term organic does not mean “life” but instead refers to a large
high temperatures without melting or being vaporized, are class of chemical compounds containing the element carbon. All ter-
referred to as refractory materials. Other materials, such restrial life is organic, but not all organic compounds come from liv-
as water, ammonia, and methane, can remain in a solid form ing organisms.
6.4 A Tale of Eight Planets 167
materials, but in addition they contain large quantities of the mini accretion disk might stay behind to coalesce into
ices and organic materials. larger bodies in much the same way that particles of dust
As we turn to a study of our own Solar System, we will in the protoplanetary disk came together to form planets.
find that the trend in composition expected in a protoplane- The result is a mini “solar system”—a group of moons that
tary disk is closely echoed in the makeup of the solid bodies orbit about the planet.
orbiting the Sun. The inner planets are composed of rocky A less massive planet may also capture some gas from
material surrounding metallic cores of iron and nickel. In the protoplanetary disk, only to lose its prize. Here again,
contrast, objects in the outer Solar System, including moons, more massive planets have the advantage. As we will learn
giant planets, and comets, are composed largely of ices of in Chapter 8, the gravity of small planets may not be strong
various types. In the years to come, as astronomers learn enough to prevent less massive atoms and molecules such
more about planetary systems around other stars, it seems as hydrogen or helium from escaping back into space. Even
likely that this trend will prove to be quite common. In fact, if a small planet is able to gather some hydrogen and helium
our understanding of the way stars and planetary systems from its surroundings, this temporary primary atmosphere
form suggests that this change in composition with distance will be short-lived. The atmosphere that remains around a
from the central star would be almost unavoidable. small planet like our Earth is
Even so, chaotic encounters like those we will discuss a secondary atmosphere. A Less massive planets
in Chapter 10 can shuffle the deck, adding diversity to the secondary atmosphere forms lose their primary
organization of planetary systems. In a process called planet later in the life of a planet. atmospheres and then
migration, gravitational scattering can force some planets Carbon dioxide and other form secondary
to end up far from the place of their birth. For example, gases released from the plan- atmospheres.
many planetary scientists believe that Uranus and Nep- et’s interior by widespread
tune originally formed near the orbits of Jupiter and Saturn, volcanism are probably one important source of a planet’s
but were then driven outward to their current locations by secondary atmosphere. In addition, as we will see later,
gravitational encounters with Jupiter and Saturn. A planet volatile-rich comets that formed in the outer parts of the
can also migrate when it gives up some of its orbital angular disk continue to fall inward toward the new star long after
momentum to the disk material that surrounds it. Such a its planets have formed, and will sometimes collide with
loss of angular momentum causes the planet to slowly spi- planets. Comets possibly provide a significant source of
ral inward toward the central star. We will see examples water, organic compounds, and other volatile materials on
when we discuss hot Jupiters in Section 6.5. planets close to the central star.
Solid Planets Gather Atmospheres

Once a solid planet has formed, it may continue growing 6.4 A Tale of Eight Planets
by capturing gas from the protoplanetary disk. To do so,
though, it must act quickly. Young stars and protostars are We are now at a point in our discussion where we can take
known to be sources of strong stellar winds (see Chapters our general ideas about the evolution of the material in a
19 and 20) and intense radiation that can quickly disperse protoplanetary disk and apply them to our own Solar Sys-
the gaseous remains of the accretion disk. Gaseous planets tem. Only in the closing years of the 20th century did our
such as Jupiter probably have only about 10 million years or knowledge progress to the point where this story could be
so to form and to grab whatever gas they can. Tremendous told. It is still sketchy in places and doubtless wrong in some
mass is a great advantage in a planet’s ability to accumulate ways. But error and uncertainty are part of the advance of
and hold on to the hydrogen and helium gas that makes up science. In this section we bring together a wealth of infor-
the bulk of the disk. Because of their strong gravitational mation taken both from what we know of our local star
fields, more massive young planets are thought to create and planetary neighbors and from what we have learned
their own mini accretion disks as gas from their surround- about stars forming around us today. It is a synthesis of the
ings falls toward them. What follows is much like the for- painstaking efforts of hundreds of astronomers and plan-
mation of a star and protoplanetary disk, but on a smaller etary scientists over the course of decades. Many more
scale. Just as happened in the accretion disk around the scientists will devote their lives to unraveling this story
star, gas from a mini accretion disk moves inward and falls before its details are fully known. There are good reasons
onto the solid planet. to believe that the explanations we give in this section are
We refer to the gas that is captured by a planet at the time basically correct, although we know they are not yet com-
of its formation as the planet’s primary atmosphere. The plete. Returning to our early discussion of what science is
primary atmosphere of a large planet can become massive and how science works, the story we tell here is one that
enough to dominate the mass of the planet, as in the case of many people have tried to prove wrong but that has with-
giant planets such as Jupiter. Some of the solid material in stood all such tests . . . so far.
rain down on their surfaces. Much of this barrage may have

originated in the outer Solar System, where the gravitational
tug of the massive, newly formed outer planets acted like
slingshots shooting debris inward. Today we can still see
the scars of these postformation impacts on the cratered
surfaces of all the terrestrial planets, such as the surface of
Mercury shown in Figure 6.12. This rain of debris contin-
ues today, albeit at a much lower rate.
Before the proto-Sun emerged as a true star, gas in the
inner part of the protoplanetary disk was still plentiful. Dur-
ing this early period the two larger terrestrial planets—Earth
and Venus—may have held on to weak primary atmospheres
of hydrogen and helium. If so, these thin atmospheres were
soon lost to space. For the most part the terrestrial planets
were all born devoid of thick atmospheres and remained so
until the formation of the secondary atmospheres that now
surround Venus, Earth, and Mars. Mercury’s proximity to
the Sun and the Moon’s small mass must have prevented
these bodies from retaining significant secondary atmo-
spheres. They remain nearly airless today.
Farther out in the nascent Solar System, 5 astronomical
units (AU) from the Sun and beyond, planetesimals coalesced
FIGURE 6.12 Large impact craters on Mercury (and on solid to form a number of bodies
bodies throughout the Solar System) record the final days of the with masses about 5–10 times The giant planets
Solar System’s youth, when planets and planetesimals grew as
that of Earth. Why such large formed cores from
smaller planetesimals rained down on their surfaces.
bodies formed in the region planetesimals and
beyond the terrestrial planets then captured gaseous
remains an unanswered ques- hydrogen and helium
Nearly 5 billion years ago, our Sun was still a protostar tion. Located in a much colder onto their cores.
surrounded by a protoplanetary disk of gas and dust. Over part of the accretion disk,
the course of a few hundred these planet-sized objects formed from planetesimals con-
thousand years, much of the Rocky terrestrial taining volatile ices and organic compounds in addition to
dust in the disk had collected planets formed in the rock and metal. Four such massive bodies would later become
into planetesimals—clumps inner Solar System. the cores of the giant planets, which we know today as Jupi-
of rock and metal near the ter, Saturn, Uranus, and Neptune. In a process astronomers
emerging Sun and aggregates of rock, metal, ice, and call core accretion, mini accretion disks formed around these
organic materials in the parts of the disk that were more planetary cores, capturing massive amounts of hydrogen and
distant from the Sun. Within the inner few astronomical helium and funneling this material onto the planets.
units of the disk, several rock and metal planetesimals, Jupiter’s massive solid core was able to capture and retain
probably fewer than a half dozen, quickly grew in size to the most gas—a quantity roughly 300 times the mass of
become the dominant masses at their respective distances Earth, or 300 M⊕. (The symbol ⊕ signifies Earth.) The other
from the Sun. With their ever-strengthening gravitational planetary cores captured lesser amounts of hydrogen and
fields, they either captured most of the remaining plane- helium, perhaps because their cores were less massive or
tesimals or ejected them from the inner part of the disk. because there was less gas available to them. Saturn ended up
These dominant planetesimals had now become planet- with less than 100 M⊕ of gas, whereas Uranus and Neptune
sized bodies with masses ranging between that of Earth were able to grab only a few Earth masses’ worth of gas. As
and about 1⁄20 of that value. They were to become the ter- we pointed out in the previous section, some planetologists
restrial planets. Mercury, Venus, Earth, and Mars are the believe that all of the giant planets formed closer to where
surviving terrestrial planets. Planetary scientists think Jupiter is now, and that their mutual gravitational interac-
that one or two others may have formed in the young Solar tions caused them to migrate to their present orbits.
System but were later destroyed. (As we will see later in Before moving on, we should mention that some plan-
this section, one of them may have been responsible for etary scientists do not believe that our protoplanetary disk
the creation of our Moon.) could have survived long enough to form gas giants such as
For several hundred million years following the for- Jupiter through the general process of core accretion. The
mation of the four surviving terrestrial planets, leftover core accretion model indicates that it could take up to 10
pieces of debris still in orbit around the Sun continued to million years for a Jupiter-like planet to accumulate. How-
ever, all the gas in the protoplanetary disk likely dispersed Appendix 8.) This region, now referred to as the asteroid
in a little more than half that time, which would have cut belt, contains many planetesimals that remain from this
off Jupiter’s supply of hydrogen and helium. early time. In the outermost part of the Solar System as well,
This is not just a Solar System predicament. The appar- planetesimals persist to this
ent confl ict between the time needed for a Jupiter-type day. Born in a “deep freeze,” Asteroids and
planet to form and the availability of gases within that these objects retained most of comet nuclei are
time period applies as well to other protoplanetary disks the highly volatile materials planetesimals that
and to the formation of their massive planets (see Sec- found in the grains present at survive to this day.
tion 6.5). To get around this time dilemma, some scientists the formation of the proto-
have proposed a process called “disk instability,” in which planetary disk. Unlike conditions in the crowded inner part
the protoplanetary disk suddenly and quickly fragments of the disk, planetesimals in the outermost parts of the disk
into massive clumps equivalent to that of a large planet. were too sparsely distributed for large planets to grow. Icy
Although core accretion and disk instability appear to be planetesimals in the outer Solar System remain today as
competing processes, they may not be mutually exclusive. comet nuclei—relatively pristine samples of the material
It is possible that both played a role in the formation of our from which our planetary system formed. Frozen Pluto and
own and other planetary systems. the distant dwarf planet Eris are especially large examples
For the same reasons that a forming protostar is hot— of these denizens of the outer Solar System.
namely, conversion of gravitational energy into thermal The early Solar System must have been a remarkably
energy—the gas surrounding the cores of the giant planets violent and chaotic place. Many objects in the Solar Sys-
compressed under the force of gravity and became hotter. tem show evidence of cataclysmic impacts that reshaped
Proto-Jupiter and proto-Saturn probably became so hot that worlds. A dramatic difference in the terrain of the northern
they actually glowed a deep red color, similar to the heating and southern hemispheres on Mars, for example, has been
element on an electric stove. Their internal temperatures interpreted as the result of one or more colossal collisions.
may have reached as high as 50,000 kelvins (K). In a sense, Mercury has a crater on its surface from an impact so dev-
the more massive protoplanets were trying to become stars. astating that it caused the crust to buckle on the opposite
But for them, this was an unreachable goal. In Chapter 15 side of the planet. In the outer Solar System, one of Sat-
we will fi nd that a ball of gas must have a mass at least 0.08 urn’s moons, Mimas, sports a crater roughly one-third the
times that of the Sun (0.08 M⊙) for it to become a star. This diameter of the moon itself. Uranus suffered a collision that
minimum mass is some 80 times the mass of Jupiter. Science was violent enough to literally knock the planet on its side.
fiction films notwithstanding, Jupiter never had a chance. Today its axis of rotation is tilted at an almost right angle
Some of the material remaining in the mini accretion to its orbital plane.
disks surrounding the giant planets coalesced into small Not even our own Earth escaped devastation by these
bodies, which became moons. cataclysmic events. In addition to the four terrestrial planets
(A moon is any natural satel- Moons formed from that remain, there was at least one other terrestrial planet
lite in orbit about a planet or the mini accretion in the early Solar System—a planet about the same size and
asteroid.) The composition of disks around the mass as Mars. As the newly formed planets were settling
the moons of the giant planets giant planets. into their present-day orbits, this fifth planet suffered a graz-
followed the same trend as ing collision with Earth and was completely destroyed. The
the planets that formed around the Sun: the innermost remains of the planet, together with material knocked from
moons formed under the hottest conditions and therefore Earth’s outer layers, formed a huge cloud of debris encircling
contained the smallest amounts of volatile material. As very Earth. For a brief period Earth may have displayed a magnifi-
young moons, Io and Europa may have experienced nearby cent group of rings like those of Saturn. In time, this debris
Jupiter glowing so intensely that it rivaled the distant Sun. coalesced into the single body we know as our Moon.
The high temperatures created by the glowing planet would
have evaporated most of the volatile substances in the inner
part of its mini accretion disk. Io today contains no water
at all. However, water is relatively plentiful on Europa, 6.5 There Is Nothing Special
Ganymede, and Callisto because these moons formed far-
ther from hot Jupiter. about Our Solar System
Not all planetesimals in the protoplanetary disk went on
to become planets. Jupiter is a true giant of a planet. Its grav- We began this chapter by discussing the formation of a
ity kept the region of space between it and Mars so stirred generic planetary system around a generic star. Only later
up that most planetesimals there never coalesced into a did we turn to the specifics of our own Solar System. We
single planet. (The one exception is Ceres, once considered did this to make a very important point: according to our
to be the largest asteroid but now redefined, along with Pluto, current understanding, there is absolutely nothing special
as a dwarf planet under the new IAU planet defi nition; see about the conditions that led to the formation of our Sun
FIGURE 6.13 An edge-on

circumstellar dust disk is
seen extending outward
to 60 AU from the young
(12-million-year-old) star
AU Microscopii. The star
itself, whose brilliance would
otherwise overpower the disk,
is hidden behind an occulting
disk placed in the telescope’s
focal plane. Its position is
represented by the dot.
and our Solar System. When astronomers turn their tele- the position of the star to wobble back and forth ever so
scopes to young stars around us, they see just the sorts of slightly. By observing the Doppler shift of the star, we can
disks from which our own Solar System formed (Figure sometimes detect this wobble
6.13). From what we know, the physical processes that led and infer the properties of the Several techniques
to the formation of our Solar System should be common- planet—its mass and its dis- are being used to find
place wherever new stars are being born. tance from the star. (We will extrasolar planets.
In 1994, astronomers found their fi rst confi rmed extra- use an almost identical
solar planet,6 a Jupiter-sized body orbiting surprisingly method in Chapter 13 to measure the masses of pairs of stars
close to 51 Pegasi, a solar-type star. Today the number of that orbit each other.)
known extrasolar planets, sometimes called “exoplanets,” We can see how this would work by using our own Solar
has grown to more than 400, with new discoveries occur- System as an example. We’ll start with the simplest of cases
ring almost daily. by assuming that the Solar System consists only of the Sun
The discovery of extrasolar planets raises the question and Jupiter. In fact, this makes sense because Jupiter’s mass
of what we mean by the term planet. Within our own Solar is greater than the mass of all the other planets, asteroids,
System, we feel reasonably comfortable with our definition and comets combined! Imagine an alien astronomer point-
of a planet.7 But what about those extrasolar bodies? The ing her spectrograph toward the Sun. Both the Sun and
International Astronomical Union defi nes an extrasolar Jupiter orbit a common center of gravity that lies just out-
planet as a body that orbits a star and has a mass less than side the surface of the Sun, as shown in Figure 6.14. Our
13 Jupiters (13 MJ). As we will see in Chapter 15, objects alien astronomer would fi nd that the Sun’s radial velocity
more massive than this, but less than 0.08 M⊙, are known varies by ±12 m/s, with a period equal to Jupiter’s orbital
as brown dwarfs. period of 11.86 years. From this information, she would
rightly conclude that the Sun has at least one planet with
a mass comparable to Jupiter’s. Without greater precision,
she would be unaware of the other less massive major plan-
The Search for Extrasolar Planets ets. But, spurred on by the excitement of her discovery, she
How do astronomers search for extrasolar planets, and why would improve the sensitivity of her instruments. If she
were they not found earlier? The “why” is easy to answer. could measure radial velocities as small as 2.7 m/s, she
Until the closing years of the 20th century, astronomers would be able to detect Saturn, and if the precision of her
lacked the sophisticated instruments needed to detect these spectrograph extended to radial velocities as small as 0.09
elusive objects. We turn now to “how.” m/s, she would be able to detect Earth.
The most successful technique employed so far has been The technology of today limits the precision of our own
the “spectroscopic radial velocity method.” More than 370 radial velocity instruments to about 1 m/s. This technique
extrasolar planets have been detected with this technique. enables astronomers to detect giant planets around solar-
As a planet orbits about a star, the planet’s gravity causes type stars, but we are still far from being able to reveal
bodies with masses similar to those of our own terrestrial
6
planets.
In 1988, astronomers announced the fi rst discovery of an extrasolar
Another technique for finding extrasolar planets is the
planet. However, the announcement was questioned because of the
quality of the observational data. That object was fi nally confi rmed “transit method,” in which we observe the effect of a planet
in 2002. passing in front of its parent star. Try picturing the geom-
7
Not entirely comfortable, as it turns out. The argument among astrono- etry required to observe a transit, however, and you may
mers continues to rage over whether dwarf planets should be defi ned see a major limitation. For a planet to pass in front of a star,
as a kind of planet or put in an entirely different category. Earth must necessarily lie nearly in the orbital plane of
FIGURE 6.14 Both the

Sun and Jupiter orbit around
a common center of gravity,
which lies just outside the
Sun
13,000 m/s Sun’s surface. Spectroscopic
measurements made by an
+ Jupiter extrasolar astronomer would
Center of gravity reveal the Sun’s radial velocity
varying by ±12 m/s over
12 m/s
an interval of 11.86 years,
Jupiter’s orbital period.
the planet. When an extrasolar planet passes in front of its A fourth method is “direct imaging.” This is the most
parent star, the light from the star will diminish by a tiny difficult technique because it involves searching for a rela-
amount, as illustrated in Figure 6.15. Whereas the radial tively faint planet in the overpowering glare of a bright
velocity method yields the mass of the planet and its orbital star—a challenge far more difficult than looking for a fi re-
distance from a star, the transit method provides a measure fly in the dazzling brilliance of a searchlight. Even when
of the size of a planet. Current groundbased technology lim- an object is detected by direct imaging, there remains the
its the sensitivity of the transit method to about 0.1 percent major problem of determining whether or not the observed
of a star’s brightness, but that has been good enough for object is indeed a planet. Suppose we detect a faint object
this method to detect more than 60 extrasolar planets. (In
fact, amateur astronomers have confi rmed the existence of
several extrasolar planets by observing transits using CCD
FIGURE 6.15 As a planet passes in front of a star, it blocks
cameras mounted on telescopes with apertures as small as
some of the light coming from the star’s surface, causing the
20 centimeters [cm]!)
brightness of the star to decrease slightly. (The brightness
Getting back to our alien astronomer, she could infer
decrease has been exaggerated in this illustration.)
the existence of Earth if she were located somewhere in the
plane of Earth’s orbit (that is the only way she would be able
to see Earth pass in front of the Sun) and could detect a 0.009
percent drop in the Sun’s brightness. Our own astronomers
now have space instruments with almost that capability.
Using the transit method, the French CoRoT satellite has
already discovered a planet with a diameter only 1.7 times
that of Earth. In 2009, NASA launched a solar-orbiting8
telescope called Kepler with even greater capability. Its Star
photometric instruments should be able to detect transits
of Earth-sized planets.
Still another technique is the “microlensing method,” Planet
which involves a relativistic effect called gravitational lens-
ing9 (see Chapter 17). The gravitational field of an unseen
planet bends light from a distant star in such a way that it
causes the star to brighten temporarily while the planet is
passing in front of it. Microlensing, like the radial veloc-
ity method, provides an estimate of the mass of the planet.
So far, nine extrasolar planets have been found with this
technique. Like the transit method used in space, lensing
is also capable of detecting Earth-sized planets.
Relative brightness
8
To keep stray light from our own planet out of its field of view, Kepler
was placed in an Earth-trailing heliocentric orbit.
9
Gravitational lensing generally refers to the relativistic lensing of
distant quasars by intervening galaxies. When referring to the lens-
ing of stars by small bodies such as planets, astronomers use the term
microlensing.
near a bright star. Might it be a distant star that just hap- tric orbits, unlike the relatively circular planetary orbits in
pens to be in the line of sight? Future observations could our Solar System. Still others contain planets far more mas-
tell if it shares the bright star’s motion through space. Could sive than Jupiter. Does this mean that our Solar System is
it be a brown dwarf (see the next subsection) rather than a an oddball? Not necessarily—but maybe.
true planet? Again, further observations to determine the Let’s look fi rst at planetary systems containing hot Jupi-
object’s mass would be required. ters. A massive planet orbiting close to its parent star is the
As of this writing, astronomers have identified 11 pos- easiest kind to detect. If you understand from the previous
sible planets by direct imaging. Some recent discoveries subsection how extrasolar planets are detected, you may
appear to be promising as actual planets. Four of the recent have already guessed why. A very massive planet orbiting
candidates (three around a single star) were discovered by very close to its parent star will create large radial veloc-
large, groundbased telescopes operating in the infrared ity variations in the star. This means that hot Jupiters are
region of the spectrum (Figure 6.16). The fi rst visible-light relatively easy targets when we are using the spectroscopic
discovery was made from space while HST was observ- radial velocity method. In addition, large planets orbiting
ing Fomalhaut, a bright naked-eye star only 25 light-years close to their parent stars are more likely to move in front of
away. The planet, temporarily named Fomalhaut b, has a the star periodically and reveal themselves with the transit
mass about three times that of Jupiter and orbits within a method. Therefore, those hot Jupiter systems that we see may
dusty debris ring some 17 billion km from the central star not be representative of most planetary systems. Scientists
(Figure 6.17). refer to this bias as the “selection effect.”
The most exciting discoveries, however, could come in Knowing why we fi nd so many hot Jupiters is one thing.
about a decade, when NASA plans to launch its Terrestrial Knowing how they got there is another. Many astronomers
Planet Finder (TPF) and ESA puts its hopes on Darwin. As were surprised by the hot Jupiters because these giant,
designed, both observatories will not only detect Earth-like volatile-rich planets should not be able to form so close
planets around nearby stars; they will also measure the plan- to their parent stars. Recall from Section 6.3 that Jupiter-
ets’ physical and chemical characteristics. As we will learn type planets form in the more distant, cooler regions of
later, in Excursions 8.1, certain telltale gases in a terrestrial the protoplanetary disk, where the volatiles that make up
planet’s atmosphere would be a strong indicator of life. much of their composition are able to survive. It seems
clear, then, that hot Jupiters must have formed much far-
ther away from their parent star and subsequently migrated
Planetary Systems Seem inward. The mechanism by which a planet migrates so far
inward is not well understood, but it must involve an inter-
to Be Commonplace action with gas or planetesimals in which orbital angular
Our searches for extrasolar planets have been remarkably momentum is somehow transferred from the planet to its
successful. Since the fi rst was confi rmed in 1994, astrono- surroundings. Recalling from Section 6.2 that decreased
mers have added more than angular momentum means decreased radius, we can see
400 to their inventory, bring- More than that a loss of angular momentum would cause the planet
ing the total number of plan- 400 extrasolar to spiral inward.
etary systems to greater than planets have been The selection effect does not explain the large number
340, including over 40 with discovered so far. of extrasolar planets having highly eccentric orbits. Perhaps
multiple planets. A represen- chaotic gravitational scattering takes place within most plan-
tative few are shown in Figure 6.18. Keep in mind that few etary systems, knocking their planets out of the circular orbits
are terrestrial-type planets. The least massive found so far in which they formed. Why, then, should our Solar System
is estimated to have a mass equal to 1.9 M⊕, and this object seem so stable? This is a question we cannot yet answer.
orbits much too close to its parent star to support an The discovery of planetary systems containing super-
atmosphere. massive planets raises an interesting question. When is a
As the number of known planetary systems grows, we “supermassive planet” too large to be labeled a planet? In
are provided with an opportunity to see how our own Solar Chapter 15 we will introduce you to a curious object called
System compares with other a brown dwarf, an in-between body not massive enough to
planetary systems. One of the Most planetary be considered a star, yet too massive to be called a planet.
first things astronomers systems found In their searches for planets, astronomers have come across
noticed about these planetary to date do not many brown dwarfs—some orbiting stars, some alone in
systems is that most of them resemble our own. space. Although the distinction between the most massive
do not look like our own. In planets and the least massive brown dwarfs is somewhat
what ways do the differ? Many contain hot Jupiters, which arbitrary, the International Astronomical Union has placed
are Jupiter-type planets orbiting solar-type stars in tight the boundary at 13 Jupiter masses (13 MJ).
circular orbits that are closer to their parent stars than Mer- Our studies of planetary systems, many unlike are own,
cury is to our own Sun. Others have planets in highly eccen- challenge some aspects of our understanding of planet
(a) (b)
b Debris disk
Beta Pictoris b Beta Pictoris

(location of
the star)
FIGURE 6.16 Direct images of probable planets around two nearby stars. (a) An infrared image
shows three planets (labeled “b,” “c,” and “d”), each with a mass several times that of Jupiter, orbiting
the star HR 8799 (hidden behind a mask.) (b) Beta Pictoris b is seen orbiting within a dusty debris
disk that surrounds the bright naked-eye star Beta Pictoris. The planet’s estimated mass is eight
times that of Jupiter. The star is hidden behind an opaque mask, and the planet appears through a
semitransparent mask used to subdue the brightness of the dusty disk.
FIGURE 6.17 Fomalhaut b

is seen here moving in its orbit
around Fomalhaut, a nearby
star easily visible to the naked
eye. The parent star, hidden
by an obscuring mask, is about
a billion times brighter than
the planet, which is located
within a dusty debris ring that
surrounds the star.
2006
2004
Solar System
Sun Mercury Venus Earth Mars
Tau Boötis Extrasolar planets are usually

3.9 MJ found from the wobble they cause
in the motion of the star they orbit.
HD 75289
Tau Boötis
0.42 MJ
51 Peg
0.468 MJ
500
Ups Andromedae
Doppler velocity of star (m/s)

0.69 MJ 1.98 MJ 3.95 MJ
55 Cnc*
0.034 MJ 0.824 MJ 0.169 MJ 0.144 MJ 0
HD 195019
3.7 MJ
HD 192263
0.72 MJ
–500
HD 114762 Orbital period
11.02 MJ = 3.313 days
HD 37124*
0 1 2 3 4 5
1.13 MJ Time (days)
HD 134987
1.58 MJ
HD 210277
1.23 MJ
16 Cyg B
1.68 MJ
HD 10697
6.12 MJ
14 Her*
4.05 MJ
* Multiple planet system with other planets not shown due to size constraints.
55 Cnc: 3.835 MJ at 5.8 AU
HD 37124: 0.61 MJ at 0.53 AU, 0.68 MJ at 3.2 AU, and 0.6 MJ at 1.6 AU
14 Her: 2.1 MJ at 6.9 AU
FIGURE 6.18 Planetary systems have been discovered around hundreds of stars other than the
Sun, confirming what astronomers have long suspected—that planets are a natural and common
by-product of star formation. A few of these planetary systems are represented here, along with an
example of the radial velocity method. (Masses of planets are given in units of Jupiter masses, MJ..)
The inner Solar System is shown to scale.
formation. Yet the message conveyed by our discoveries ancestors viewed Earth and the heavens as fundamentally
is clear: the formation of planets frequently, and perhaps different regimes—forever apart, each with its own rules
always, accompanies the formation of stars. The implica- and reality. Today we know there is nothing unusual about
tions of this conclusion are profound. Planets are a common the conditions of our own existence. All we have to do to
by-product of star formation. So, in a galaxy of a 100 billion see our own history is to look at the stars that continue to
stars, and a universe of hundreds of billions of galaxies, form around us today.
how many planets (or even moons) with Earth-like condi- It is ironic that at a time when science has turned
tions might exist? And with all of these Earth-like worlds such a dazzling spotlight on our place in the universe,
in the universe, how many might play host to the particular many otherwise educated individuals continue to cling
category of chemical reactions that we refer to as “life”? In to outdated and fanciful notions about the heavens and
Chapter 23 we will explore the idea of what kind of planet Earth.10 For someone interested in truly mind-boggling
life might choose for its home. insights, the speculations and “revelations” of the mystic
In learning about the formation of planetary systems, pale beside the discoveries of the scientist. Through the
then about our Solar System, and finally about extrasolar sometimes stodgy, often painstaking, and always uncom-
planets, our journey has come promising standards of science, we have come to appre-
full circle. We began by look- Modern discoveries ciate that we are not apart from the universe. Rather we
ing outward in wonder at the about the formation of are a part of the universe. And the processes and events
lights in the night sky. Now the Sun and Solar that link us to this larger universe are fascinating and
that outward exploration System “complete” the wondrous indeed.
becomes instead a look inward Copernican revolution.
as we discover the processes 10
We cannot resist the temptation to quote the late, famous science fic-
and events that set the stage for our own existence. In a tion author Arthur C. Clarke at this juncture: “In one sense, of course,
sense, the insights we have gained in this chapter complete every age renews itself, as indeed it should. But the nitwits currently
the revolution that Copernicus started long ago when he parroting this slogan [“New Age”] seem unable to understand that
had the audacity to challenge authority and suggest that their ‘New Age’ is exactly the opposite, being about a thousand years
Earth was not the center of all things. Not that long ago, our past its sale date.”
our Solar System and the many other planetary systems

Seeing the Forest for the Trees formed were none other than the stellar accretion disks
that surrounded the young suns.
In this chapter we have seen two great lines of investiga- The joining of these two great rivers of thought, obser-
tion merge into a single picture that shapes our under- vation, and exploration—the link between the accretion
standing of the context of our own existence. Working disks that surround young stars and the local collection
fi rst from the perspective of the stellar astronomer, basic of planets that we call home—is the starting point for a
physical principles—in particular, the conservation of modern study of the Solar System. It also ties our journey
angular momentum—demand that when a star like our of understanding within the Solar System to the course
Sun forms, it will be surrounded by a thin, orbiting disk we will later chart outward into the larger universe.
of gas and dust. This conclusion went from hypothesis Stated more bluntly, all that we know about the Sun, plan-
to fact with the discovery of such disks around numer- ets, and the world of our birth makes sense only when
ous young stars. In the meantime, as stellar astronomers viewed within the context of the evolving universe as a
were trying to understand star formation, planetary sci- whole. We have had a fi rst glimpse of the power of this
entists were scrutinizing the worlds that make up our insight in the current chapter. As we move on to look
Solar System. at the planets, moons, asteroids, and comets that orbit
The pieces of the planetary scientists’ puzzle range the Sun, we will time and again see the fi ngerprints left
from specimens from space—meteorites collected in Ant- behind by our birth among the stars.
arctica and lunar samples brought back by Apollo astro- The journey that we are taking in 21st Century Astron-
nauts—to data sent back by spacecraft visiting remote omy is one of scientific discovery, but the philosophical
worlds. Those pieces fit together to form a clear picture implications of the connection between star formation
of a flat, swirling cloud of gas and dust from which Earth and the formation of our Solar System and other plan-
and its neighbors coalesced about 5 billion years ago. etary systems should not be overlooked. Our very exis-
When the stellar astronomers and the planetary scien- tence is set within the context of the larger universe. We
tists compared notes, they realized that they had arrived are the legacy of the processes we see at work all around
at exactly the same description. The disks from which us, even today.
Summary individual object can never lose or gain angular momen-

tum or energy? Explain your reasoning.
• Stars and their planetary systems form from collapsing 8. How does the law of conservation of angular momentum
interstellar clouds of gas and dust. control a figure-skater’s rate of spin?
• Planets are a common by-product of star formation, and *9. Explain why the law of conservation of angular momen-
many stars are surrounded by planetary systems. tum created problems for early versions of the nebular
hypothesis.
• Planets grew from a protoplanetary disk of gas and dust
that surrounded the forming Sun. 10. What is an accretion disk?
• Our Solar System formed about 5 billion years ago, 11. Describe the process by which tiny grains of dust grow
nearly 9 billion years after the birth of the universe. to become massive planets.
• Solid terrestrial planets formed in the inner disk, where 12. There are two reasons that the inner part of a protoplan-
temperatures were high. etary disk is hotter than the outer disk. What are they?
• Giant gaseous planets formed in the outer disk, where 13. During the process in which an interstellar cloud col-
temperatures were low. lapses to form an accretion disk, energy exists in three
forms. Name and describe them.
• Dwarf planets formed in the asteroid belt and in the
region beyond the orbit of Neptune. 14. Why do we fi nd rocky material everywhere in the Solar
System, but large amounts of volatile material only in
• Asteroids and comet nuclei remain today as leftover
the outer regions? Would you expect the same to be true
debris.
of other solar systems? Explain your answer.
• Hundreds of extrasolar planets have been found orbit-
15. When we think of ice, we tend to think of water ice, but
ing other stars within our galaxy.
among scientists this is a very restrictive defi nition.
Explain why.
16. Why were the four giant planets able to collect massive
gaseous atmospheres, whereas the terrestrial planets
could not?
your instructor assigns questions in SmartWork, log in at 17. What is the meaning of the term organic?
18. Explain the source of the atmospheres now surrounding
three of the terrestrial planets.
19. What happened to all the leftover Solar System debris
Student Questions after the last of the planets formed?
THINK ING A BOUT THE CONCEP T S 20. Name the terrestrial and the giant planets in our Solar
System.
1. Compare the size of our Solar System with the size of
the universe. 21. Describe four methods that astronomers use to search
for extrasolar planets.
2. What is the source of the material that now makes up
the Sun and the rest of the Solar System? 22. Why is the transit method so restrictive as a method for
fi nding extrasolar planets?
3. In a poetic sense, planets are sometimes referred to as
“children of the Sun.” A more accurate description, 23. Why has it been so difficult for astronomers to take a
though, would be “siblings of the Sun.” Explain why. picture of an extrasolar planet?
*4. Describe the different ways that stellar astronomers and 24. Many of the planets that astronomers have found orbit-
planetary scientists each came to the same conclusion ing other stars have been giant planets with masses more
about how planetary systems form. like that of Jupiter than of Earth, and with orbits located
very close to their parent stars. Does this prove that our
*5 Explain the nebular hypothesis.
Solar System is unusual? Explain your answer.
6. What is a protoplanetary disk?
25. Step outside and look at the nighttime sky. Depending on
7. Physicists describe certain properties, such as angular the darkness of the sky, you may see dozens or hundreds
momentum and energy, as being “conserved.” What does of stars. Would you expect many or very few of those
this mean? Do these conservation laws imply that an stars to be orbited by planets? Explain your answer.
A P P LY I N G T H E C O N C E P T S *32. In the text we give an example of an interstellar cloud

having a diameter of 1016 meters and a rotation period
26. Use information about the planets given in Appendix 4
of 106 years collapsing to a sphere the size of the Sun
to answer the following:
(1.4 × 109 meters in diameter). We point out that if all the
a. What is the total mass of all the planets in our Solar
cloud’s angular momentum went into that sphere, the
System, expressed in Earth masses (M⊕)?
sphere would have a rotation period of only 0.6 second.
b. What fraction of this total planetary mass does Jupiter
Do the calculation to confi rm this result.
represent?
c. What fraction does Earth represent? 33. The asteroid Vesta has a diameter of 530 km and a mass
of 2.7 × 1020 kg.
*27. Compare Jupiter’s orbital angular momentum with its
a. Calculate the density of Vesta.
spin angular momentum using the following values:
b. The density of water is 1,000 kilograms per cubic
m = 1.9 × 1027 kg, v = 13.1 kilometers per second (km/s),
meter (kg/m3), and that of rock is about 2,500 kg/m3.
r = 5.2 AU, R = 7.15 × 104 km, and P = 9 hours 56 min-
What does this difference tell you about the compo-
utes. Assume Jupiter to be a uniform body. What frac-
sition of this primitive body?
tion does each component (orbital and spin) contribute
to Jupiter’s total angular momentum? Refer to Math Tools 34. If an alien astronomer observed a plot of the light curve
6.1 for help. as Jupiter passed in front of the Sun, by how much would
the Sun’s brightness drop during the transit?
28. Assume that the Sun is a uniform sphere with a radius
of 700,000 km and a rotation period of 27 days. Near the 35. A hot Jupiter nicknamed “Osiris” has been found around
end of its life, the remnant of the Sun will be a white a solar-mass star, HD 209458. It orbits the star in only
dwarf with a radius of only 5,000 km. Assuming that the 3.525 days.
mass remains unchanged, what will be the Sun’s new a. What is the orbital radius of this extrasolar planet?
rotation period as a white dwarf? b. Compare its orbit with that of Mercury around our
own Sun. What environmental conditions must Osiris
**29. The mass of the Sun (M⊙) is 2 × 1030 kg. Using informa-
experience?
tion provided in the previous two questions, compare
Jupiter’s orbital angular momentum with the Sun’s spin *36. The extrasolar planet Osiris passes directly in front of
angular momentum. What does this comparison tell you its solar-type parent star, HD 209458 (diameter = 1.7 ×
about the distribution of angular momentum in the Solar 106 km), every 3.525 days, decreasing the brightness of
System? the star by about 1.7 percent (0.017).
a. What is the diameter of Osiris?
30. Venus has a radius 0.949 times that of Earth and a mass
b. Compare the diameter of this extrasolar planet with
0.815 times that of Earth. Its rotation period is 243 days.
that of Jupiter (mean diameter = 139,800 km).
What is the ratio of Venus’s spin angular momentum to
that of Earth? Assume that Venus and Earth are uni-
form spheres.
31. Jupiter has a mass equal to 318 times Earth’s mass, an Astronomy. Study Plans include animations, read-
orbital radius of 5.2 AU, and an orbital velocity of 13.1 ing outlines, vocabulary flashcards, and multiple-choice
km/s. Earth’s orbital velocity is 29.8 km/s. What is the ratio quizzes, plus links to premium content in SmartWork and
of Jupiter’s orbital angular momentum to that of Earth? the ebook. Visit www.wwnorton.com/studyspace.
We’ve sent a man to the Moon, and that’s [240,000]
miles away. The center of the Earth is only 4,000 miles
away. You could drive that in a week, but for some reason
nobody’s ever done it.
ANDY ROONE Y (1919 – )
Mercury, Venus, and the Moon rising over the Australia Telescope Compact Array in New South Wales.
CHAP TER 7
The Terrestrial Planets

and Earth’s Moon
7.1 How Are Planets the
K E Y CONCEP T S
Same, and How Are The objects that formed in the inner part of the
They Different? protoplanetary disk around the Sun were relatively small
rocky worlds, one of which we call home. Comparing
Most of what we know about our planetary neighbors we those worlds with one another teaches us lessons
have learned only in the most recent decades. The second about what shapes a planet’s fate. Among the lessons
half of the 20th century was a wonderfully exciting time we will learn in this chapter are
of exploration and discovery about Earth and its sibling
• That each terrestrial planet is shaped by impacts,
worlds. It was a time that saw unmanned probes visit every
classical planet and watched astronauts walk on the surface tectonism, volcanism, and erosion.
of the Moon. (See Chapter 5 for a discussion of the variety of • How impacts scarred planets early in the history of the
techniques used to explore the Solar System.) The informa- Solar System and still occur on occasion today.
tion returned from these missions has revolutionized our
• Why the concentration of craters on a planetary
understanding of our planetary system, offering us insights
surface tells us how old the surface is.
into the current state of each of our neighbors and clues
about their histories. • How radiometric dating tells us the ages of rocks.
The four innermost planets in our Solar System are • How radiometric dating of lunar rocks is used to
Mercury, Venus, Earth, and Mars, collectively known as calibrate the cratering clock.
the terrestrial planets (Table 7.1). Although technically the
• How we test predictions about Earth’s interior by using
Moon is known as Earth’s lone natural satellite, we include
it here in this chapter because of its close similarity to the seismic waves from earthquakes.
terrestrial planets. • That larger worlds remain geologically active longer
The vast quantity of information about the planets because smaller planets cool off sooner.
returned by space probes can be hard to digest. What infor-
• That tectonism takes different forms on the terrestrial
mation is truly fundamental
planets, and plate tectonics is unique to Earth.
and exciting, and what infor- We learn about
mation is flashy but less impor- planets by comparing • That among the volcanoes found on Mercury, Venus,
tant? The key to sorting them with each other. Earth, and Mars, the most colossal are on Mars.
through this information has • The many ways in which erosion modifies and wears
proven to be comparison among the different planets. The down the surface features that other processes form.
ways the planets are alike and how they differ draw our
attention to the most fundamental issues, helping us ask and
179
180 Chapter 7 The Terrestrial Planets and Earth’s Moon
TA B L E 7.1
Comparison of Physical Properties

of the Terrestrial Planets and the Moon
Mercury Venus Earth Mars Moon*
Orbital radius (AU) 0.387 0.723 1.000 1.524 384,000 km
Orbital period (years)† 0.241 0.615 1.000 1.881 27.32d
Orbital velocity (km/s) 47.9 35.0 29.8 24.1 1.02
Mass (M⊕ = 1) 0.055 0.815 1.000 0.107 0.012
Equatorial diameter (km) 4,880 12,104 12,756 6,794 3,476
Equatorial diameter (D⊕ = 1) 0.383 0.949 1.000 0.533 0.272
Density (water = 1) 5.43 5.24 5.52 3.93 3.34

†
Sidereal rotation period 58.65d 243.02d 23h 56m 24h 37m 27.32d
Obliquity (degrees)‡ 0.04 177.36 23.45 25.19 6.68
Surface gravity (m/s2) 3.70 8.87 9.78 3.71 1.62
Escape speed (km/s) 4.25 10.36 11.18 5.03 2.38
*The Moon’s orbital radius and orbital period are given in kilometers and days, respectively.
†
The superscript letters d, h, and m stand for days, hours, and minutes of time, respectively.
‡
An obliquity greater than 90° indicates that the planet rotates in a retrograde, or backward, direction.
answer the right questions. The correct explanation for a astronauts looking back at Earth from space (Figure 7.1). For
particular aspect of one planet must be consistent with what many, this was a philosophical turning point, compelling us
we know about the other planets. For example, when we to view the world for what it truly is: “a small blue marble,”
explain why the Moon is covered with craters, our reason- a tiny and fragile lifeboat adrift in the vastness of space—
ing must also allow for the fact that preserved craters on our home. As seen in a closer view from space (Figure 7.2),
Earth are rare. An explanation for why Venus has such a Earth is awash with color. White clouds drift in our atmo-
massive atmosphere should point to reasons that Earth and sphere, and white snow and ice cover the planet’s frozen
Mars do not. Such comparisons—an approach called com- poles. The blue of oceans, seas, lakes, and rivers of liquid
parative planetology—have provided the guideposts to plan- water—Earth’s hydrosphere—covers most of the planet.
etary scientists as they sift through mountains of data. When Brown shows us the outer rocky shell of Earth, referred to
making comparisons, we need a place to start. Earth is the as Earth’s lithosphere. And green is the telltale sign of veg-
planet we know best, so it is here at home that we begin our etation, part of Earth’s biosphere—the most extraordinary
appraisal of the worlds of the inner Solar System. among Earth’s many distinctions.
The psychological impact of these images is rivaled only
by the change in our scientific perspective. We see Earth
in the context of the larger universe, shaped by processes
7.2 Four Main Processes that happen throughout its vast expanse. Earth is a place of
change: geological processes constantly work to reshape our
Shape Our Planet planet. Some geological processes originate in the interior of
Earth, powered by the thermal energy generated there. Earth-
For most of human history we have looked upon Earth as quakes are sudden reminders of the ongoing deformation of
a vast, almost limitless expanse. This view of our planet Earth’s lithosphere, referred to as tectonism. Tectonism folds
changed forever with a single snapshot taken by Apollo 8 and breaks Earth’s crust, forming mountain ranges, valleys,
7.2 Four Main Processes Shape Our Planet 181
FIGURE 7.1 Our first view of Earth seen from deep space. In
December 1968, Apollo 8 astronauts photographed our planet
rising above the Moon’s limb.
FIGURE 7. 3 Volcanism, such as this eruption in Hawaii, spills

molten rock and other materials onto planetary surfaces.
that flash across the sky when meteoroids, or chunks of

material from outer space, hit our atmosphere; but for at
least one woman, the experience with celestial debris is
more intimate. In 1954 a meteorite (a meteoroid that reaches
a planet’s surface) crashed through the roof of the woman’s
home in Alabama, striking her hand and hip. Fortunately,
the roof of the house slowed the meteorite, so she was left
FIGURE 7. 2 Seen from space, the colors of Earth tell of the with only a bad bruise.2 When a very large object hits a
diversity of our planet’s features. planet, as still happens occasionally in the Solar System,
the resulting devastation can be global. Distinctive scars in
Earth’s crust tell of impacts in our past. Such catastrophic
and deep ocean trenches. Volcanism—another geological events have altered Earth’s climate and have changed the
process affecting Earth and the other terrestrial planets— history of life itself (see Chapter 23).
is a form of igneous activity in which gas and molten rock, These three processes—tectonism, volcanism, and
or magma, erupt at the surface. Volcanic eruptions, like impact cratering—affect Earth’s surface in their own char-
the one shown in Figure 7.3, can spill sheets of lava and ash acteristic ways. Together, they work to form topographic
over vast areas, forming mountains or plains in the process. relief—mountains, valleys, and ocean basins. Meanwhile,
AstroTour: Processes That Shape the Planets a fourth process, called erosion, works slowly but persis-
Another process that affects Earth is external. Colli- tently to level Earth’s surface. Erosion by running water,
sions involving planetary objects—a process referred to as wind, and other agents wears down hills, mountains, and
impact cratering—are extremely important in our planet’s continents. The eroded debris collects downslope, fi lling
history. Most of us have seen meteors,1 the bright streaks in valleys, lakes, and ocean basins. Left on its own, ero-
1 2
The terms meteor, meteoroid, meteorite, and even comet are confus- You should not lose much sleep worrying about meteorites landing
ing to many people. These terms are all explained more completely in your lap. Unlike being struck by lightning, being hit by a meteorite
in Chapter 12. is even less likely than winning the state lottery.
sion would eventually leave the surface of our planet

smoot h and featureless.
7.3 Impacts Help Shape the
Because Earth is a geologi-
cally and biologically active
Tectonism, volcanism,
and impact cratering
Evolution of the Planets
world, however, its surface is rough up planetary Although all terrestrial planets are subject to impact crater-
an ever-changing battle- surfaces; erosion ing, tectonism, volcanism, and erosion, the relative intensity
ground between processes smooths them. of these processes varies among the planets. Of these four
that build up topography and geological processes, however, large impacts account for
those that tear it down. by far the most concentrated and sudden release of energy.
Each of the four types of geological processes at work Planets and other objects orbiting the Sun move at very
shaping the surface of Earth leaves its own distinctive sig- high speeds. Earth moves at about 30 kilometers per second
nature. Planetary scientists have learned to read these sig- (km/s) in its orbit around the Sun, and meteoroids can enter
natures on the surfaces of other planets as well. Earth’s atmosphere at relative speeds in excess of 70 km/s.
Because the kinetic energy of an object is proportional to
the square of its speed (EK = 1⁄2mv 2), collisions between such
objects release huge amounts of energy. When an object hits
FIGURE 7. 4 (a) Stages in the formation of an impact crater.
(b) A lunar crater photographed by Apollo astronauts, showing
the crater wall and central peak, surrounded by ejected material (b)
(ejecta), rays, and secondary craters—all typical features
associated with impact craters. (c) A rayed crater on Mercury.
(a)
The impact of an object heats and
compresses the surface it hits.
Secondary
craters
Rays Crater wall
Meteoroid Central peak
Material is thrown from the site of Ejecta

the impact. Ejecta lands around
the crater, forming rays and
secondary craters. Ejecta
(c)
Underlying strata
deformed
Rebound of the deformed surface

may form a central peak in the
crater, while melted rock pools in
the crater floor, giving it a flat bottom. Rays Secondary
craters
Crater wall
Central peak
Underlying strata
rebound
(a) (b) (c)
(d) (e) (f)
VISUAL ANALOGY FIGURE 7. 5 A drop hitting a pool of milk

illustrates the formation of features in an impact crater,
including crater walls (b and c), secondary craters (d and e),
and a central peak (f).
a planet, its kinetic energy goes into heating and compres-

sing the surface that it strikes and throwing material far
from the resulting impact crater (Figure 7.4). Sometimes
material thrown from the crater, called ejecta, falls back
to the surface of the planet with enough energy to cause
secondary craters. The rebound of heated and compressed
material can also lead to the formation of a central peak or
a ring of mountains on the crater floor. These processes are
similar, on a much smaller scale, to what happens when a
drop lands in a glass of milk, as shown in Figure 7.5.
The energy of an impact can be great enough to melt or
even vaporize rock. The floors of some craters are the cooled
surfaces of pools of rock FIGURE 7.6 Meteor Crater (also known as Barringer Crater),
melted by the impact. The Impacts can melt and located in northern Arizona, is an impact crater 1.2 kilometers
energy released in an impact vaporize rock. (km) in diameter that formed some 50,000 years ago by a nickel-
can also lead to the formation iron meteoroid’s collision with Earth.
of new minerals. In fact, some minerals, such as shock-mod-
ified quartz, are known to form only during the momentary
fury of an impact. Geologists look for these distinctive min- grams (kg) that hit Earth traveling at 13 km/s about 50,000
erals as evidence of ancient impacts on Earth’s surface. years ago. Planetary scientists believe that approximately
Meteor Crater in Arizona is one of the best-preserved half of the original mass was vaporized in the atmosphere
impact structures on our planet (Figure 7.6). It is thought to before hitting the ground. Such a collision would have
be the result of the impact of a nickel-iron meteorite about released about 300 times as much total energy as the fi rst
50 meters in diameter with a mass of about 300 million kilo- atomic bomb, which was detonated in New Mexico in 1945.
E X C U R S I O N S 7.1
Where Have All the Dinosaurs Gone?

In 1994 a host of astronomical instruments watched as about is marked in Earth’s fossil record by the Cretaceous-Tertiary
two dozen pieces of Comet Shoemaker-Levy 9 slammed one boundary, or simply the “K-T boundary.”3 (The Cretaceous
after another into the clouds of Jupiter, leaving temporary scars Period lasted from 146 million years ago to 65 million years ago.
that were visible through backyard telescopes. The Jupiter The Tertiary Period started when the Cretaceous Period ended
comet crash provided graphic evidence that although impacts and lasted until 1.8 million years ago. We live in the Quater-
are not as frequent today as they were when the Solar System nary Period of Earth’s history, which started at the end of the
was young, they still do occur. The giant planets are not the Tertiary Period.) In older layers found below the K-T boundary,
only targets for such cosmic bombardment. fossils of dinosaurs and other now-extinct life-forms abound.
In 2004 a very small asteroid, whose rocky body measured In the newer rocks above the K-T boundary, well more than half
about 30 meters across, whizzed past Earth at nearly 25,000 of all previous species are absent, and in their place are found
kilometers per hour (km/h), missing our planet by only 43,000 fossils of newly evolving species. Big winners in the new order
km. A close call to be sure, but it was after all a relatively small were the mammals—our own distant ancestors—which moved
body that probably would have broken up or exploded in our into ecological niches vacated by extinct species.
atmosphere, causing only local damage had we been hit. That The K-T boundary is marked in the fossil record in many
same year, University of Hawaii astronomers discovered an areas by a layer of clay. Studies at more than 100 locations around
asteroid that was given the rather forgettable name 2004 MN4. the world have found that this layer contains large amounts of
Forgettable, that is, until further observations and refi ned cal- the element iridium, as well as traces of soot. Iridium is very
culations showed that 2004 MN4 (now named Apophis), a mass rare in Earth’s crust but is common in meteorites. The soot at the
of rock 270 meters in diameter, will come within 30,000 km of K-T boundary tells of a time when widespread fires burned the
Earth’s surface on April 13, 2029 (Friday, the 13th). Although the world over. The thickness of the layer of clay at the K-T boundary
chance of a collision with Earth is very unlikely, the energy of and the concentration of iridium increase as we move toward
such an impact would be 880 megatons—15 times more powerful what is today the Yucatán Peninsula in Mexico. Although the
than the most powerful hydrogen bomb ever detonated. When original crater has been erased by erosion, geophysical surveys
large impacts happen on Earth, they can have far-reaching con- and rocks from drill holes in this area show a highly deformed
sequences for Earth’s climate and for terrestrial life. subsurface rock structure, similar to that seen at known impact
Earth’s fossil record shows that, on occasion, large numbers sites. Together, these results provide compelling evidence that 65
of species vanish from the face of the planet in a geological blink million years ago an asteroid 10 km in diameter struck the area,
of an eye. The most famous of these extinctions occurred 65 mil- throwing great clouds of red-hot dust and other debris into the
lion years ago, when more than 70 percent of all living species,
3
including the dinosaurs, became extinct. This mass extinction The K comes from Kreide, German for “Cretaceous.”
Yet, at only 1,200 meters in diameter, Meteor Crater is tiny There is another reason for the shortage of small cra-
compared with impact craters seen on the Moon or ancient ters on Earth. Whereas the surface of the Moon is directly
impact scars on Earth. (Excursions 7.1 describes the con- exposed to this cosmic bombardment, the surface of Earth
sequences of an especially violent impact that took place is partly protected by the blanket of Earth’s atmosphere. For
65 million years ago.) example, rock samples from the Moon show craters smaller
One of the most obvious differences among the terres- than a pinhead, formed by micrometeoroids. In contrast,
trial planets is the visible extent of impact cratering. On most meteoroids smaller than 100 meters in diameter are
some planets the surfaces are covered by impact craters; on either burned up or broken up by friction in Earth’s atmo-
others (especially Earth and Venus), impact craters seem to sphere before they reach the surface. With an atmosphere
be rare. For example, the Moon has millions of craters of all far more massive than that of Earth, Venus is even better
different sizes, one on top of another (Figure 7.8). Nearly all protected. Objects even larger than 100 meters across may
of these craters are the result of impacts. By comparison, be burned up or broken apart by Venus’s atmosphere before
fewer than 200 impact craters, or scars from impacts, have reaching the surface, creating clusters of craters in a pattern
been identified on Earth; and about 1,000 on Venus. The similar to a shotgun blast, or sometimes leaving only a dark
primary culprits behind Earth’s crater shortage are plate “splotch” on the surface with no crater at all.
tectonics in Earth’s ocean basins and erosion on land. Most The characteristics of a crater also depend on the prop-
craters on land have been obliterated by wind and water. erties of the planetary surface. An impact in a deep ocean
FIGURE 7.7 This

artist’s rendition depicts
an asteroid or comet,
perhaps 10 km across,
striking Earth 65 million
years ago in what is now
the Yucatán Peninsula
in Mexico. The lasting
effects of the impact
killed off most forms of
terrestrial life, including
the dinosaurs.
atmosphere (Figure 7.7) and igniting a worldwide conflagration. temperature changes, and decreased food supplies could have
The energy of the impact is estimated to have been more than led to mass starvation that would have been especially hard
that released by 5 billion nuclear bombs. on large animals such as the dinosaurs.
An impact of this magnitude clearly would have had a Not all paleontologists believe that this mass extinction
devastating effect on terrestrial life. Could this cosmic impact was the result of an impact. They point out that the evolution
have been responsible for the sudden disappearance of forms of of species is a complex process and that simple answers are
life that had ruled Earth for 150 million years? Many scientists seldom complete. However, the evidence is compelling that
believe so. In addition to the nearly global fi restorm ignited by a great impact did occur at the end of the Cretaceous Period.
the impact, dust thrown into Earth’s upper atmosphere would The rock record also shows other instances of mass extinctions
have remained there for years, blocking out sunlight and plung- associated with colossal impacts, suggesting that impacts have
ing Earth into decades of cold and darkness. The fi restorms, played a central role in the saga of life on Earth.
on Earth might create an impressive wave but leave no last- like this, so the water or ice must have been concentrated
ing crater. In contrast, an impact scar formed in granite can in only some areas, and these icy locations might have
be preserved for billions of years. Planetary scientists can changed with time.
tell a lot about the surface of It is possible that these craters were formed at a time in
a planet by studying its cra- The forms of impact the past when there was liquid water on the surface of Mars.
ters. For example, craters on scars give clues Canyons and dry riverbeds seen on Mars today attest to this
the Moon’s pristine surface about the surfaces possibility. But there is another intriguing possibility. Today
are often surrounded by they are on. the surface of Mars is dry and mostly frozen, suggesting
strings of smaller secondary that water that might once have been on the surface has
craters formed from material thrown out by the impact, like soaked into the ground, much
those shown in Figure 7.4b. Some craters on Mars have a like the water frozen in the Mars was once wetter
very different appearance. These craters are surrounded by ground in Earth’s polar than it is today.
what appears to be flows of material, much like the pattern regions. The energy released
you might see if you threw a rock into mud (Figure 7.9). The by the impact of a meteoroid would melt this ice, possibly
apparent flows seem to indicate that, unlike the lunar case, turning the surface material into a slurry with a consistency
the martian surface rocks contained water or ice at the time much like wet concrete. When thrown from the crater by
of the impact. Not all martian crater ejecta deposits look the force of the impact, this slurry would have hit the sur-
rounding terrain and slid out across the surface, forming

the mudlike craters we see today.
Calibrating a Cosmic Clock

Every terrestrial planet experienced a similar bombard-
ment early in its history. Although many factors affect the
formation of craters, the big-
gest difference among planets The number of
is the rate at which craters are craters on a surface
destroyed. Earth experienced indicates the
an impact history similar to age of the surface.
that of the Moon and the
other terrestrial planets, and yet preserved impact craters
are rare on Earth, as previously noted. The combined action
of erosion, tectonic processes, and volcanism throughout
geological time has obliterated most of Earth’s impact scars.
Geologically active planets such as Earth, Mars, and Venus,
whose interior thermal energy fuels volcanism and tec-
tonic processes, bear the scars of only the more recent
impacts. By contrast, the Moon’s surface still preserves
the scars of craters dating from the early years of the Solar
System. With no atmosphere or surface water for erosion
and a cold, geologically dead interior, the lunar surface
has remained much the same for more than a billion years.
Mercury's well-preserved craters suggest that it, like the
FIGURE 7.8 The Moon’s terrain, shown here in a photograph Moon, has been geologically inactive for a long time. Plan-
taken during the Apollo 17 mission, has been heavily cratered by etary scientists use this cratering record to estimate the
impacts. ages and geological histories of planetary surfaces—exten-
sive cratering means an older planetary surface and mini-
mal geological activity.
We can use the amount of cratering as a clock to measure
the ages of surfaces, but fi rst we need to know how fast that
clock runs. We need to be able to say that a surface with
this many craters is this old, but a surface with that many
craters is that old. In other words, we need a way of “cali-
brating the cratering clock.”
The key to calibrating the cratering clock came mostly
from our exploration of the Moon. The surface of the Moon
is not uniform. Some parts of
the Moon are heavily cratered, Most of the surface of
with craters overlapping their the Moon is older than
neighbors. Other parts are 3.4 billion years.
much smoother, telling of
more recent geological activity. Between 1969 and 1976,
Apollo astronauts and Soviet unmanned probes visited the
Moon and brought back samples taken from nine different
locations on the lunar surface. By measuring relative amounts
of various radioactive elements and the elements into which
they decay, scientists were able to assign ages to these dif-
ferent lunar regions (see Foundations 7.1 and Math Tools
FIGURE 7.9 Some craters on Mars look like those formed by 7.1). The results of that work were surprising. Although
rocks thrown into mud, suggesting that material ejected from the smooth areas on the Moon were indeed found to be younger
crater contained large amounts of water. This crater is about 20 than heavily cratered areas, the differences in age were not
km across. great. The oldest, most heavily cratered regions on the Moon
F O U N D AT I O N S 7.1
Determining the Ages of Rocks

Look at a picture of the Grand Canyon (Figure 7.10). The rock
layers tell the story of the canyon’s geological history. Most
of the layers were laid down by a process called “sedimenta-
tion,” in which material carried by water or wind buries what
lies below. Volcanism also contributes to the layering as lava
flows over Earth’s surface. At the top of the stack are the latest
deposits, such as those found on the rim of the Grand Canyon.
Moving down through these layers takes us progressively fur-
ther and further back into Earth’s past.
To assign real dates to these different layers—or to rocks
from any location, including the Moon—scientists use a tech-
nique called radiometric dating. Radiometric dating makes use
of the steady decay of radioactive parent elements into more sta-
ble daughter products. Some minerals can contain radioactive
isotopes as part of their chemical structure. (Different isotopes
of a particular element have the same number of protons in their
nuclei but different numbers of neutrons.) Chemical analysis
of such a mineral immediately after its formation would reveal
the presence of the radioactive parents, but the daughter prod-
ucts of the radioactive decay would be absent. As radioactive
atoms decay over time, however, the amount of parent elements
decreases and the amount of trapped daughter products builds
up. Chemical analysis of an old sample of such a mineral would
reveal both remaining radioactive parent atoms and daughter
products trapped within the structure of the mineral. By com-
paring the relative amounts of radioactive parent and daughter FIGURE 7.10 Layers of rock within the Grand Canyon reveal
products, scientists can determine when the mineral was formed its geological history.
and hence the age of the rock (see Math Tools 7.1).
M AT H T O O L S 7.1
How to Read Cosmic Clocks

As Foundations 7.1 explains, a geologist can calculate the age of isotope of the element lead (206Pb, its final daughter). The half-life
a mineral by measuring the relative amounts of a parent radio- of 238U is 4.5 billion years. This means that in 4.5 billion years, a
isotope and its daughter product. The time interval over which a sample that originally contained the uranium isotope (the parent)
radioactive isotope decays to half its original amount is called its but no lead (its final daughter) would be found instead to contain
half-life. With every half-life that passes, the remaining amount equal amounts of uranium and lead. If we were to find such a
of the radioisotope will decrease by a factor of 2. For example, mineral, we would know that half the uranium atoms had turned
over three half-lives, the remaining amount of a parent radioiso- to lead and that the mineral formed 4.5 billion years ago.
tope will be 1⁄2 × 1⁄2 × 1⁄2 = 1⁄8 of its original amount. We can make Let’s look at another example, this time with a different iso-
this concept more general with a simple relationship: tope of uranium (235U) that decays to a different lead isotope (207Pb)
with a half-life of 700 million years. Suppose that a lunar min-
(2)
PF 1 n
___ = __ , eral brought back by astronauts has 15 times as much 207Pb (the
PO
daughter product) as 235U (the parent radioisotope). This means
where PO and P F are the original and fi nal amounts, respec- that 15 ⁄16 of the parent radioisotope (235U) has decayed to the daugh-
tively, of a parent radioisotope, and n is the number of half- ter product (207Pb), leaving only 1⁄16 of the parent remaining in the
lives that have gone by. mineral sample. Noting that 1⁄16 is (1⁄2)4, we see that 4 half-lives
The most abundant isotope of the element uranium (238U, the have elapsed since the mineral was formed, and that this lunar
parent) decays through a series of intermediate daughters to an sample is therefore 4 × 700 million = 2.8 billion years old.
FIGURE 7.11 Radiometric dating of
1 Ages are measured lunar samples returned from specific

by radioactive dating sites by Apollo astronauts is used to
of lunar samples. determine how the cratering rate has
changed over time. Cratering records
can then be used to tell us the age of
other parts of the lunar surface.
2 Older surfaces still bear the

scars of ancient craters.
Apollo 16
Cratering rate
3 Younger surfaces have

experienced little cratering,
and so are relatively smooth.
Apollo 11
Apollo 15
4 From these data we reconstruct the history of impacts
Age of Apollo 12 in the Solar System. Impacts were much more
Moon frequent billions of years ago than they are today.
5 4 3 2 1 Present
Billions of years before present
date back to about 4.4 billion years ago, whereas most of the sands of kilometers below our feet, and how do we know?
smoother parts of the lunar surface are typically 3.1 billion These are the questions we turn to now.
to 3.9 billion years old.34 The clear implication is that the vast
preponderance of the cratering in the Solar System took
place within the first billion years of the Solar System’s for-
mation (Figure 7.11). Heavily cratered surfaces such as those
We Can Probe the Interior of Earth
of Mercury and the Moon are ancient indeed. For all the effect that human activity has had on Earth, we
have literally only scratched the surface of our planet. The
deepest holes ever drilled
have gone only about 12 km The density of matter
7.4 The Interiors of the below the surface, and Earth’s
center is 6,350 km deeper!
inside terrestrial
planets increases
Terrestrial Planets Even so, scientists are confi- with depth.
dent about what the interior
Tell Their Own Tales of Earth is like. Information about Earth’s interior comes to
us in many ways. For example, the size of Earth and the
To understand the processes responsible for all but wiping strength of Earth’s gravity, together with Newton’s universal
away the visible record of impacts in Earth’s past and for law of gravitation, tell us the mass and hence the density
continually remaking the surface of our planet, we begin of Earth. From this information we know that the average
by looking below its surface. What lies hundreds and thou- density of Earth is 5,500 kilograms per cubic meter (kg/m3),
or five and a half times the density of water (which is 1 gram
4
3
Although they have not been radiometrically dated, some of the young- per cubic centimeter [g/cm3], or 1,000 kg/m3). Even though
est flows show very few impact craters and are therefore believed to we have no direct samples of material from deep within
be no more than 1 billion to 2 billion years old. Earth’s interior, this number alone tells us that the compo-
(a) Longitudinal wave: (b) Transverse wave:

Displacement parallel Displacement perpendicular
to wave motion Position of bead on to wave motion
undisturbed spring
Direction 1 A bead is pushed toward its Motion of
of travel undisturbed position… wave pattern
Undisturbed
position of string
2 …but coasts past

because of its inertia…
Time
Time
3 …until the opposing force of
the string brings it to a halt.
4 After one wave period, the

bead is back where it started.
FIGURE 7.12 Mechanical waves result from forces that try to even out disturbances. (a) A
longitudinal wave involves oscillations along the direction of travel of the wave. (b) A transverse
wave involves oscillations that are perpendicular to the direction in which the wave travels. Primary
seismic waves are longitudinal and secondary seismic waves are transverse.
sition of the interior of our planet has a density higher than The other type of wave travels through Earth, rather
that of the surface, which averages about 2,900 kg/m3. Other than along its surface, probing the interior of our planet.
clues about Earth’s interior come from studies of meteorites. Waves in this category include primary waves and second-
Because these fragments are left over from a time when the ary waves. Primary waves (P waves) result from alternating
Solar System was young and Earth was forming from simi- compression and decompression of a material. P waves are
lar materials, the overall composition of Earth should resem- longitudinal waves (Figure 7.12a) that distort the material
ble the composition of meteorite material, which includes they travel through, much as sound waves travel through
minerals with abundant amounts of iron, whose density is air or water or compression waves move along the length
nearly 8,000 kg/m3. of a spring. Secondary waves (S waves) are more like the
By far the most important source of information about motion that occurs when we pluck a guitar string. S waves
Earth’s interior comes from monitoring the vibrations from are transverse waves (Figure 7.12b) that result from the
earthquakes. When an earth- sideways motion of material. Unlike primary waves, which
quake occurs, vibrations Seismic waves travel through rock because rock rebounds after being com-
spread out through and across provide information pressed, secondary waves travel through rock because rock
the planet as seismic waves. about the interior springs back after being bent.
There are two different classes configuration of Earth. The progress of seismic waves through Earth’s interior
of seismic waves. As their depends on the characteristics of the material they are
name implies, surface waves travel across the surface of a moving through (Figure 7.13). Primary waves can travel
planet, much like waves on the ocean. If conditions are through either solids or liquids, but secondary waves can-
right, surface waves from earthquakes can be seen rolling not travel through liquids, because liquids do not “spring
across the countryside like ripples on water. These waves back” when they are “bent.” The speed of seismic waves
are responsible for much of the heaving of Earth’s surface provides additional information about Earth’s interior.
during an earthquake, causing damage such as the buckling Seismic waves travel at different speeds depending on the
of roadways. density and composition of rock. As a result, seismic waves
For clarity, each type of wave is

shown on one half of Earth.
Primary waves (P waves) Secondary waves (S waves)

Longitudinal waves result from Transverse waves are like the
alternating compression and waves that move along the length
decompression, like a spring. of a string. They cannot pass
They travel through solid and liquid. through liquids.
Epicenter
P waves S waves
S waves are also bent

as they pass through rock
of different densities.
P waves are bent

as they pass through
layers of different Lithosphere
densities…
Solid
inner
core
Liquid
outer
core
No
Mantle
dir
…resulting in
ec
a “shadow area.”
tP
Crust
wa
es
av
ve
s w
tS
ec Because S waves cannot
dir pass through liquid, nearly
No half of Earth opposite the
epicenter is a “shadow area.”
FIGURE 7.13 Primary and secondary seismic waves move through the interior of Earth in
distinctive ways. Measurements of when and where different types of seismic waves arrive after
an earthquake enable us to test predictions of detailed models of Earth’s interior. Note the “shadow
areas” caused by the refraction of primary waves (yellow) at the outer boundary of the liquid outer
core and the inability of secondary waves (blue) to pass through the liquid outer core.
moving through rocks of varying densities or composition waves to penetrate the liquid outer core create “shadows” of
are bent in much the same way that waves of light are bent the liquid core on the side of Earth opposite an earthquake’s
when they enter or leave glass.45 In fact, when a wave comes epicenter, as shown in Figure 7.13. Much of our understand-
to a place where the density of rock layers varies abruptly, ing of the properties of Earth’s liquid outer core is due to
the wave can be refracted (bent) or even reflected just as these shadow areas and the individual characteristics of P
light is refracted or reflected by a pane of glass (see Foun- and S waves.
dations 5.1). Scientists use instruments called seismometers to mea-
The refraction of primary waves at the outer edge of sure the distinctive patterns of seismic waves. For nearly a
Earth’s liquid outer core and the inability of secondary hundred years, thousands of seismometers scattered around
the globe have measured the vibrations from countless
5
4
The reason light does not take a curved path through glass is that the earthquakes and other seismic events, such as volcanic
index of refraction remains constant throughout. Seismic waves take eruptions and nuclear explosions. When seismic waves
a curved path through Earth's interior because its density is continu- arrive at a seismometer station, geologists ask many ques-
ously changing. tions, including these: What types of waves were measured?
How strong were they? When were they received at the sta- This method is how geologists arrived at our current
tion? Alone, a single seismometer can record ground motion picture of the interior of Earth, as shown in Figure 7.13.
at only one place on Earth. But when we combine such mea- The major subdivisions of Earth’s interior include a two-
surements with those of many other seismometers placed component core, a thick mantle, and the lithosphere. At
all over Earth, we can use the data to get a comprehensive the center, Earth’s core consists primarily of iron, nickel,
picture of the interior of our planet. and other dense metals. In contrast, the outer parts of Earth
are made of lower-density materials. The mantle, which
surrounds the core, is made up of medium-density mate-
rials. The lithosphere, which includes the uppermost part
Building a Model of Earth’s Interior of the mantle and the crust, consists of the lowest-density
How geologists go from raw seismic data to an understand- materials. The crust, which is the outermost part of the
ing of Earth’s interior is a very good example of the interplay lithosphere, comes in two components: the low-density,
between theory and observation in modern science. To con- silica-rich crust that forms the continents; and the higher-
struct a model of Earth’s interior, geologists begin with some density crust of the ocean floor. Common continental rocks
obvious clues from the seismic data. Are there liquid regions include granite, a coarse-grained volcanic rock rich in
where secondary waves cannot penetrate? Are there jumps silicon and oxygen; most oceanic crust, however, consists
in the density of the rock from which waves are reflected? of basalt, a heavy, dark volcanic rock that is rich in iron
How do waves bend, and what does that bending say about and magnesium.
the density profile of Earth? The model must also be consis- As Figure 7.13 illustrates, the interior structure of Earth
tent with Earth’s average density of 5,500 kg/m3. is far from uniform in composition. Geologists speak of
To go beyond this basic sketch, geologists turn to the Earth’s interior as being “dif-
laws of physics, combined with knowledge of the properties ferentiated” and the process Differentiation
of materials and how they behave at different temperatures of separating materials by of planets shows
and pressures. For example, scientists take into account the density as differentiation. that they once
fact that the pressure at any point in Earth’s interior must Differentiation of the interiors were molten.
be just high enough to balance the weight of all the mate- of Earth, other terrestrial
rial above it. To see why this is so, think about what would planets, and the Moon results from the fact that these plan-
happen if it were not. If the outward pressure at some point etary interiors were once molten. When rocks of different
within a planet were less than the weight per unit area of types are mixed together, they tend to stay mixed. Once this
the overlying material, then that material would fall inward, rock melts, however, the denser materials sink to the bot-
crushing what was underneath it. If the pressure at some tom and the less dense materials float to the top (just as less
point within a planet were greater than the weight per unit dense oil floats on denser vinegar in a bottle of salad dress-
area of the overlying material, then the material would be ing). Today, little of Earth’s interior is molten, but the dif-
able to expand and push outward, lifting the overlying mate- ferentiated structure of the planet tells of a time when Earth
rial. The situation is stable only when the weight of matter was much hotter and its interior was liquid throughout. The
above is just balanced by the pressure within the whole cross sections in Figure 7.14 show the differentiated struc-
interior of the planet. This balance between pressure and ture of each of the terrestrial planets and Earth’s Moon. As
weight is known as hydrostatic equilibrium. We will see we continue our study of the composition of objects in the
this balance repeatedly as we talk about the structure of Solar System, differentiation will be an important concept.
planetary interiors, planetary atmospheres, and the struc- For example, in Chapter 12 we will fi nd that by analyzing
ture and evolution of stars. the chemical composition of meteorite material, we can
With such considerations in mind, scientists then con- often determine whether it was once part of a body that was
struct a model of Earth’s interior that follows the rough chemically differentiated.
outline of their basic sketch but that is also physically con-
sistent. They next “set off” earthquakes in their model, cal-
culate how seismic waves would propagate through a model
Earth with that structure, and predict what those seismic
The Moon Was Born from Earth
waves would look like at seismometer stations around the As Figure 7.14 shows, the Moon has only a tiny core, and
globe. They then test their model by comparing these pre- its core is composed mostly of material that is similar to
dictions with actual observations of seismic waves from real that found in Earth’s mantle.
earthquakes. The extent to which the predictions agree with The best model explaining the The Moon was
observations points out both strengths and weaknesses of Moon’s composition involves probably formed by a
the model. The structure of the model is adjusted (always a catastrophic impact on collision between a
remaining consistent with the known physical properties Earth. Planetary scientists protoplanet and the
of materials) until a good match is found between predic- believe that when Earth was young Earth.
tion and observation. very young, a Mars-sized pro-
Crust
Mantle
Outer core
Earth (liquid)
Inner
core
(solid)
Venus
FIGURE 7.15 An artist's rendition of a Mars-size protoplanet

colliding with young Earth. Debris from the impact coalesced
around Earth and formed our Moon.
toplanet collided with Earth, blasting off and vaporizing

parts of Earth’s partly differentiated crust and mantle (Fig-
ure 7.15). This debris condensed into orbit around Earth,
evolving into our Moon. This model explains the similari-
ties in composition between the Moon and Earth’s mantle,
Mars and the model can account for the Moon’s lack of water and
other volatiles (see Chapter 6) while Earth and its closest
neighbors—Mars and Venus—are volatile-rich. According
to this model, during the vaporization stage of the collision,
most gases were lost to space, leaving only the nonvolatiles
to condense as the Moon. Earth, however, was large enough
to keep its volatiles, which continued to be released from
the interior. Because of the stronger gravity of Earth, these
Mercury gases were retained as part of our atmosphere.
KEY The Evolution of Planetary Interiors

Core
Mantle Depends on Heating and Cooling
Crust
Moon A general feature of planetary interiors is that the deeper
we go within the planet, the higher the temperature climbs.
2,000 km When we look below a planet’s surface, we fi nd that the
thermal energy in its interior drives the planet’s geologi-
FIGURE 7.14 A comparison of the interiors of the terrestrial cal activity. Thermal equilibrium, which we introduced
planets and Earth’s Moon. Some fractions of the cores of in Chapter 4, governs the complex interplay of heating and
Mercury, Venus, and Mars are probably liquid. cooling within a planet.
Part of the thermal energy in the interior of Earth is left a temperature of perhaps 6,000 kelvins (K), it is hotter than
over from when Earth formed. We know that the tremen- the surface of the Sun. Yet the center of Earth is solid! It
dous energy liberated by the collisions during the accretion is the outer core of Earth that is molten, even though it is
process responsible for the formation of Earth, together with many hundreds of kelvins cooler than the inner core. How
energy from short-lived radioactive elements, was enough can we explain this solid, superhot inner core coexisting
to melt the planet. The differ- with a cooler, molten outer core?
entiated structure of Earth is Collisions and It turns out that whether a material is solid or liquid
evidence of this fact. The sur- radioactive heating depends on pressure as well as on temperature. With most
face of Earth then cooled made the forming materials (water ice being a rare exception), the solid form of
rather rapidly by radiating Earth molten. As it the material is more compact than the liquid form. Putting
energy away into space, form- ages, however, Earth the material under higher pressure forces atoms and mol-
ing a solid crust on top of a is cooling off. ecules closer together and makes the material more likely
molten interior. Because a to become a solid. Moving toward the center of Earth, the
solid crust does not conduct thermal energy well, it served effects of temperature and pressure oppose each other: The
as an insulator—a “wool sweater,” if you will. (Anyone who higher temperature makes it more likely that material will
has watched a lava flow “crust over” and then walked across melt, but the higher pressure favors a solid form. Only in the
its surface has experienced this fact. Molten rock is still outer core of Earth does the high temperature win, allow-
there, possibly only centimeters beneath the adventurous ing the material to exist in a molten state. At the center of
walker’s feet.) But the crust is not a perfect insulator. Over Earth, even though the temperature is higher, the pressure
the eons, energy from the interior of the planet continued is so great that the inner core of Earth is solid. We will fi nd
to leak through the crust and radiate into space. The inte- even stranger physical properties of the cores of the giant
rior of the planet slowly cooled, and the mantle and the planets when we turn to them in Chapter 9.
inner core solidified. Like stars, planets lose their internal thermal energy
However, there must be more to the story than leftover through a combination of convection, conduction, and radi-
energy from the time of Earth’s formation. If this were the ation. We already discussed radiation of energy by plane-
only source of heating in Earth’s interior, calculations show tary bodies in Chapter 4. We will describe the mechanism
that the interior of Earth would be much cooler than it is of convection as it relates to plate tectonics in Section 7.5.
today, and Earth would have long ago solidified completely. Finally, we will cover thermal conduction when we talk
There must be additional sources of energy continuing to about energy transfer within the Sun’s interior in Chapter
heat the interior of Earth if we are to account for the high 14. For now, we’ll simply say that liquids transfer heat by
temperatures that persist there today. convection and solids transfer heat by conduction.
One source of heating Earth’s interior is friction gener- The internal temperature of a planet depends on the
ated by tidal effects of the Moon and Sun. When we discuss planet’s size. A planet’s volume—and thus the amount of
tides in Chapter 10, we will find that tidal heating is respon- radioactive material (“fuel”) it contains—determines the
sible for keeping the interior of Jupiter’s moon Io molten. amount of interior heating produced. On the other hand,
However, tidal heating fails by a wide margin to explain the planet’s ability to get rid of the thermal energy in its
the elevated temperature of Earth’s interior. Instead, most interior depends on the planet’s surface area because ther-
of the extra energy in Earth’s interior comes from long-lived mal energy has to escape through the planet’s surface. (By
radioactive elements. As these radioactive elements trapped analogy, when you want to
in the interior of Earth decay, they liberate energy, heating keep warm, you curl up into Generally, smaller
the planet’s interior. Today the temperature of Earth’s inte- a ball, reducing the amount terrestrial planets
rior is determined by dynamic equilibrium (see Founda- of exposed skin through cool faster than larger
tions 4.1) between the heating of the interior and the loss which thermal energy can terrestrial planets.
of energy that is radiated away into space. As radioactive escape. But when you want
“fuel” in Earth’s interior is consumed by decay, the amount to cool off, you spread out your arms and legs, exposing
of thermal energy generated declines, and Earth’s interior as much skin as possible and allowing it to get rid of ther-
becomes cooler as it ages. mal energy.)
Temperature plays an important role in a planet’s inte- A simple mathematical relationship describes the con-
rior structure, but it’s not the only influence at work. The nection between a planet’s size and its temperature (see
structure of Earth’s core results from an interesting interplay Math Tools 7.2). Smaller planets have less energy to lose in
between increasing temperature and pressure. In our every- relation to surface area and are thus cooler. Larger planets
day experience we notice that temperature alone determines have more energy to lose per square meter of surface and so
whether a material is solid or liquid. When something gets remain hotter. It should not be surprising to learn that the
hot enough, it melts and becomes a liquid. When something smaller objects (Mercury and the Moon) are thus geologi-
gets cold enough, it freezes and becomes a solid. The cen- cally inactive in comparison to the larger terrestrial planets
ter of Earth is the hottest location in Earth’s interior. With (Venus, Earth, and Mars).
M AT H T O O L S 7. 2
How Planets Cool Off

Let’s assume that all terrestrial planets formed with the same the surface area through which thermal energy can escape—is
percentage of radioactive materials in their bulk composition, proportional to R3/R2, or R.
and that these radioactive materials are their sole source of Of course, we have to remember that a planet’s ability to
internal thermal energy. A planet’s volume therefore would transfer internal energy from its hot core to its cooling surface
determine the total amount of the thermal energy–produc- depends on factors such as its own internal convection and
ing material it contains. A planet loses its internal energy conduction properties. Nevertheless, all things being equal,
by radiating it away at its surface. A planet’s cooling surface larger planets retain their internal energy longer than smaller
area therefore determines the rate at which it can get rid of its planets do. Or, to put in another way, smaller planets lose their
thermal energy. The energy-producing volume of a planet is internal thermal energy more quickly. For example, Mars has
proportional to the cube of the planet’s radius (volume ∝ R3), a radius about half that of Earth, so it has been losing its inter-
whereas the cooling surface area of the planet is proportional nal thermal energy to space about twice as fast as our own
to only the square of the radius (surface area ∝ R2). The ratio planet has. This is one reason why Mars is less geologically
of the two—the amount of energy there is to lose, divided by active than Earth.
we show it in Figure 7.16, with the north magnetic pole of

Most Planets Generate Their a compass attracted to Earth’s south magnetic pole.
Own Magnetic Fields Earth’s magnetic field actually originates deep within
the interior of the planet, and the processes responsible for
Earth’s magnetic field is not difficult to detect, but what generating Earth’s magnetism are not understood in detail.
causes it? As most schoolchildren know, you can detect But there’s one thing we know for certain: Earth’s magnetic
Earth’s magnetic field with a compass consisting of an field is not due to “permanent magnets” (naturally occur-
arrow-shaped needle attached to a small bar magnet that ring magnetic materials whose individual atoms are mag-
is allowed to swing about netically aligned) buried within the planet. Even though
freely. The compass needle Earth’s magnetic naturally occurring magnetic materials do exist, permanent
lines up with Earth’s mag- field behaves like a magnets cannot explain the fact that Earth’s magnetic field
netic field and points “north” giant bar magnet, but is constantly changing. Some changes in Earth’s magnetic
and “south.” But if we were the orientation of the field, such as a shift in the exact location of the magnetic
to map the orientation of com- poles changes poles, can occur over periods that are much shorter than
pass needles at every place with time. a human lifetime.65 In addition, the geological record shows
on Earth, the north-pointing that much more dramatic changes in the magnetic field have
arrows would all converge not at the geographic North occurred over the history of our planet.
Pole, but at a location in northern Canada. This is Earth’s The study of paleomagnetism—the fossil record of
north magnetic pole. An opposite magnetic pole (Earth’s Earth’s changing magnetic field—is an important part of
south magnetic pole) is located just off the coast of Ant- geology. When a magnetic material such as iron gets hot
arctica, 2,800 km from Earth’s geographic South Pole. How enough, it loses its magnetization. (This is another reason
do compasses work the way they do? Bar magnets attract why permanent magnets cannot be responsible for Earth’s
or repel one another, and Earth behaves as if it contained magnetic field. At the high temperatures found in Earth’s
a giant bar magnet that was slightly tilted with respect to interior, permanent magnets lose their magnetization.) As
Earth’s rotation axis and had its two endpoints near the the material cools, it again becomes magnetized by any mag-
two magnetic poles (Figure 7.16). netic field in which it is immersed. Iron-bearing minerals
Wait a minute! Did you look carefully at Figure 7.16? thus record the direction of the Earth’s magnetic field at the
Did you notice that Earth’s hypothetical bar magnet has its very instant in time that they cooled. In this way a memory
south pole in Earth’s Northern Hemisphere? Can you guess
why? The answer is quite simple when you think about it. 6
5
At the moment, the north magnetic pole is on the move, traveling
Bar magnets are attracted to each other’s opposite poles. several tens of kilometers per year toward the northwest. If this rate
Traditionally, we have defi ned the north-pointing end of a were to continue, the north magnetic pole could leave Canada and be
compass’s bar magnet as its north pole. Having made this in Siberia by the end of the century. The north and south magnetic
decision, we must in turn define Earth’s magnetic field as poles wander independently of each other.
(a) (b)
Direction of
magnetic field S
North Pole
N
North
magnetic pole
Inner N
N core
South
magnetic pole
N
VISUAL ANALOGY FIGURE 7.16 (a) Earth’s magnetic field can be visualized as though it were a giant bar
magnet tilted relative to Earth’s axis of rotation. Compass needles line up along magnetic field lines and
point toward Earth’s north magnetic pole. Note that because bar magnets are attracted to each other’s
opposite poles, Earth’s hypothetical bar magnet has its south pole coincident with Earth’s north magnetic
pole. (b) Iron filings sprinkled around a bar magnet help us visualize such a magnetic field.
of that magnetic field becomes “frozen” into the material. astronauts used surface “magnetometers” (devices for mea-
For example, lava extruded from a volcano carries a record suring magnetic fields) to make local measurements; and
of Earth’s magnetic field at the moment when it cooled. By on two missions small satellites were placed into orbit to
using the radiometric techniques discussed in Foundations search for global magnetism.
7.1 to date these materials, geologists obtain a record of how Results show that the Moon The Moon had
Earth’s magnetic field has changed over time. Although either lacks a magnetic field a magnetic field long
Earth’s magnetic field has probably existed for billions of today or, at the most, has a ago, but it lacks
years, the north-south polarity reverses from time to time. very weak field. The lack of a one today.
On average, these reversals in Earth’s magnetic field take lunar magnetic field can prob-
place about every 500,000 years. ably be understood because of the small size of the Moon
Although the details are not known, we have a general and its correspondingly cooler interior. It also has a very
idea of how Earth’s magnetic field originates in the motions small core. Overall, it would be difficult to make a lunar
of material in Earth’s interior. Magnetic fields result from dynamo work today. However, remnant magnetism is pre-
electric currents, which are moving electric charges. Earth’s served in lunar rocks from an earlier time when the Moon
magnetic field is thought to be a side effect of three factors: likely had a molten core and a magnetic field.
Earth’s rotation about its axis; an electrically conducting, When we study the other terrestrial planets, we encoun-
liquid outer core; and fluid motions including convection ter a few surprises. Other than Earth, Mercury is the only
within the outer core. The interior of Earth acts as if it were terrestrial planet with a significant magnetic field today.
a giant dynamo,67 converting mechanical energy into mag- The existence of Mercury’s magnetic field is understand-
netic energy. (Other objects in the Solar System, including able. The planet has the ingre-
the Sun and the giant planets, are also thought to act like dients to form a dynamo: The lack of magnetic
dynamos.) rotation and a large iron core, fields on Venus and
The Moon is the most magnetically surveyed astronomi- parts of which seem to be mol- Mars is a puzzle.
cal object other than Earth. During the Apollo program, ten and circulating. But the
lack of a detected magnetic field from Venus presents a
6
7
Although the term dynamo is frequently used to describe the creation puzzle. Venus should, like Earth, have an iron-rich core and
of magnetic fields in the interiors of planets and stars, it is not tech- partly molten interior. Its lack of a magnetic field might be
nically correct. A true dynamo generates an electric current from a attributed to its extremely slow rotation, which could make
moving magnetic field, not the other way around. its dynamo ineffective. On the other hand, Mercury also
rotates very slowly (once every 58.6 Earth days) but still has shown in Figure 7.17, where the roadway has been cut
a planetary magnetic field. through rock. These cuts show layers of rock that have
Like the Moon, Mars has a weak magnetic field, pre- been bent, broken, or fractured into pieces. Sometimes
sumably frozen in place when its dynamo stopped work- the force responsible for tectonism—the deformation of
ing many billions of years ago. This pronounced remnant Earth’s crust—is just gravity. For example, for more than
magnetic field was discovered by the Mars Global Surveyor 60 million years, rivers have dumped trillions of metric
orbiter in the late 1990s. The magnetic signature occurs tons of rock and sediment into the Gulf of Mexico. Lay-
only in the ancient crustal rocks, showing that early in ers of sediment over 25 km thick have built up, and some
the history of Mars, some sort of an internally generated have solidified into rock. This enormous mass has been
magnetic field must have existed. Geologically younger pulled downward by gravity, causing the rock layers to
rocks lack this residual magnetism, so the planet’s origi- bend or to break along faults (fractures in Earth’s crust).
nal magnetic field has long since disappeared. The lack of Faults and folds in these rocks form traps for the accu-
a strong magnetic field today on Mars might be the result mulation of petroleum, creating some of the richest oil
of its small core. However, given that it is expected to have fields in the world.
a partly molten interior and rotates rapidly, the lack of a
field is still surprising.
Continents Drift Apart and
Come Back Together
7.5 Tectonism— Although the weight of the crust is responsible for some of
the deformation of Earth’s crust, most of the faulting and
How Planetary buckling that we see at Earth’s surface originates instead
from forces deep within Earth’s interior. Early in the 20th
Surfaces Evolve century, some scientists recognized that Earth’s continents
could be fit together like pieces of a giant jigsaw puzzle.
With a better understanding of planetary interiors, we The fit was particularly striking between the Americas and
now turn to their surfaces. As you drive through moun- Africa-plus-Europe. Other evidence also suggested that
tainous or hilly terrain, take a look at places like that this fit was more than coincidence. For example, the lay-
ers in the rock on the east coast of South America and the
fossil records they hold match those on the west coast of
Africa. On the basis of evidence such as this, in the 1920s
FIGURE 7.17 Tectonic processes fold and warp Earth’s crust,
as seen in these rocks along a roadside in Israel.
the German scientist Alfred Wegener (1880–1930) proposed
that over millions of years the continents had shifted their
positions. This theory is popularly referred to as conti-
nental drift. Wegener proposed that the continents were
originally joined in one large landmass that subsequently
broke apart as the continents began to “drift” away from
each other. AstroTour: Continental Drift
Originally, the idea of continental drift was met with
great skepticism among geologists because they could not
conceive of a mechanism that could move such huge land-
masses. In the 1960s, however, paleomagnetic studies of the
ocean floor provided compelling evidence for continental
drift (Figure 7.18). These surveys showed surprising char-
acteristics in bands of basalt found on both sides of the
ocean rifts. Ocean floor rifts such as the Mid-Atlantic Ridge
are spreading centers. These are locations where convec-
tion causes hot material to rise toward Earth’s surface and
fill the gap between tectonic plates, thereby becoming new
ocean floor.
Remember that when hot material cools, it becomes mag-
netized along the direction of the local magnetic field. So
as new ocean floor is created, it cools and is carried away
from the spreading center, “remembering” the direction
of Earth’s magnetic field at the time it was formed. In this
way the spreading ocean floor carries with it a record of the
(a) (b)
1 Hot material rises Spreading center
to fill the gap in Plate 1 Plate 2
Oceanic crust
the ocean floor. Iceland
(basalt)
4 million years
ago (mya)
− −
Ri -
e
tic Mid
dg
lan f
At is o
2 New basalt shows a
Ax
magnetic field oriented Reversed
Normal
in one direction.
3 mya
+ −
− +
Atlantic Ocean
3 After Earth’s magnetic field
reverses, new basalt shows
reversed magnetic field. 5 Magnetization of the
− ocean floor is a “ticker tape”
2 mya +
− documenting millions of years
−
+ of plate tectonic motion.
−
4 After Earth’s magnetic field

reverses again, new basalt
shows original magnetic field.
− FIGURE 7.18 (a) As new seafloor is formed at
+
− a spreading center, the cooling rock becomes
Ocean crust today + +
− magnetized. The magnetized rock is then carried away
+ 5.0
− 3.3 by tectonic motions. (b) Maps like this one of banded
2.5
0.7 magnetic structure in the seafloor near Iceland
0
0.7 provide support for the theory of plate tectonics.
3.3 2.5
5.0
Age of ocean floor (millions of years)
changes in Earth’s magnetic field over time. Greater distance Today, geologists recognize that Earth’s outer shell is
from a spreading center indicates an older ocean floor and composed of a number of relatively brittle segments, or litho-
an earlier time, as reflected by its magnetization. The ocean spheric plates, and that motion of these plates is constantly
floor acts much like an old ticker-tape recorder for Earth’s changing the surface of Earth. This theory, which is perhaps
magnetic history. The ocean surveys also revealed that this the greatest advance in 20th
banded magnetic structure is often symmetrical about rifts. century geology, is referred to The theory explaining
If a change in the magnetization of the ocean floor is seen as plate tectonics. Plate tecton- motions of Earth’s
100 km on one side of a rift, then the same change will ics is ultimately responsible lithosphere is called
usually be seen about 100 km on the other side of the rift. for a wide variety of geological plate tectonics.
Combined with radiometric dates for the rocks, this mag- features on our planet.
netic record proved that the spreading of the seafloor and
the motions of the plates have continued over long geologi-
cal time spans.
Precise surveying techniques and satellite positioning
Plate Tectonics Is Driven by Convection
methods (such as the global positioning system, or GPS) The forces required to set lithospheric plates into motion
now allow locations on Earth to be determined to within a are immense. We now understand that these forces are the
few centimeters. These measurements confi rm that Earth’s result of thermal energy escaping from the interior of Earth
lithosphere is indeed moving. Some areas are being pulled through the process of convection. If you have ever noticed
apart by more than 15 centimeters (cm)—about the length of how water moves about in a heated pot on a stovetop, then
a pencil—each year. This rate is slower than a snail’s pace, you have seen convection at work (Figure 7.19a). Thermal
but over millions of years of geological time, such motions energy from the stove warms water at the bottom of the pot.
add up. Over 10 million years—not a long time by geological The warm water expands slightly, becoming less dense than
standards—15 cm/year becomes 1,500 km, by which time the cooler water above it, and the cooler water with higher
the maps defi nitely need to be redrawn! density sinks, displacing the warmer water upward. When
These plates are driven by convection cells in Earth’s man-

tle, carrying both continents and ocean crust along with
(a) them. As we pointed out earlier, convection also creates
new crust along rift zones in the ocean basins, where mantle
material rises up, cools, and slowly spreads out.
Figure 7.20 illustrates the process of plate tectonics and
some of its consequences. If you think about convection,
you will understand that if material is rising and spread-
ing out in one location, it must be converging and sinking
1 Convection 2 …upward to in another. Locations where plates converge and convection
carries thermal the surface
energy from where it is currents turn downward are called subduction zones. In
below… released. a subduction zone, one plate slides beneath the other and
convection drags the submerged lithospheric material back
down into the mantle. The Mariana Trench—the deepest
3 Convective motions in part of Earth’s ocean floor—is such a subduction zone.
the upper mantle drag Much of the ocean floor lies between spreading centers
plates along, powering
plate tectonics. and subduction zones, and as a result the ocean floor tends
(b) to remain the youngest portion of Earth’s crust. In fact,
the oldest seafloor rocks are
less than 200 million years Plates separate, or
old. (As mentioned earlier, spread apart, in some
this is one reason we do not regions and collide in
see evidence for very large other regions.
impact craters beneath the
oceans.) In some places, however, the plates are not sub-
1 Mantle is heated 2 Mantle cools near surface and ducted but collide with each other and are shoved upward.
from below, becomes sinks, displacing hot mantle For example, the highest mountains on Earth, the Hima-
buoyant, and rises. and pushing it upward. layas, arose when the Indo-Australian subcontinental
plate collided with the Eurasian Plate. The Indo-Austra-
VISUAL ANALOGY FIGURE 7.19 (a) Convection occurs when a
lian Plate is continuing to move northward, causing the
fluid is heated from below. (b) Convection in Earth’s mantle drives
Himalayas to rise at a rate of 0.5 meter per century. In still
plate tectonics.
other places, lithospheric plates meet at oblique angles
and slide along past each other. A type of fault called a
transform fault marks the actively slipping fracture zone
the lower-density water reaches the surface, it gives up between plate boundaries. One such zone is the San
part of its energy to the air, and in so doing cools, becomes Andreas Fault in California, where the Pacific Plate shears
denser, and sinks back toward the bottom of the pot. Water past the North American Plate.
cannot rise and fall in the same place, so convection cre- Locations where plates meet tend to be very active geo-
ates an organized circulating pattern in which warm water logically. In fact, one of the best ways to see the outline of
rises in some locations and cool water sinks in others. As Earth’s plates is to look at a
we will see in later chapters, convection plays an impor- map of where earthquakes Most volcanoes
tant role in planetary atmospheres and in the structure of and volcanism occur, like that and earthquakes
the Sun and stars. shown in Figure 7.21. At loca- occur along plate
In the case of Earth, thermal energy generated by radio- tions where plates run into boundaries.
active decay in the interior of the planet causes convection each other, enormous stresses
to take place in the mantle build up. Earthquakes result as the friction between the two
(Figure 7.19b). Earth’s mantle Earth’s outer plates fi nally gives way and the plates slip past each other,
is not molten (if it were, sec- rocky shell is divided relieving the stress. Volcanoes are created when friction
ondary seismic waves could into seven major between plates melts rock, which is then pushed up through
not travel through it), but it is plates and about six cracks to the surface. Lithospheric plates can be thousands
somewhat mobile. You can smaller ones. of kilometers across and range in thickness from about 5 to
think of the mantle as having 100 km. As they shift, some parts move more rapidly than
the plastic consistency of hot glass. This consistency allows others, causing the plates to stretch, buckle, or fracture.
convection to take place, albeit very slowly. Careful map- These effects are readily seen on the surface as folded and
ping shows that Earth’s lithosphere is divided into about faulted rocks. Mountain chains also are common near con-
seven major plates and about a half-dozen smaller ones. verging plate boundaries, where plates buckle and break.
Materials from the mantle rise Where one plate meets another,
via convection and fill the gap the denser oceanic plate subducts
between the spreading plates. under the continental plate.
The continental plate deforms
by compression, bending and
folding its rock layers.
South
American Compression
Lithosphere Plate
Pacific Plate
Nazca Plate
Indo-Australian
Plate Subduction
zone Spreading
center Subduction
zone
Where two continental plates
Spreading meet, the crust can push up
Himalayan Mountains into high mountains.
India
FIGURE 7. 20 Divergence
and collisions of tectonic plates Indo-Australian Plate
Fault
create a wide variety of geological Eurasian Plate
features. The continental crust can form
cracks or faults as it deforms.
FIGURE 7. 2 1 Major earthquakes and volcanic activity are

KEY Separating plates Volcano often concentrated along the boundaries of Earth’s principal
tectonic plates.
Converging plates Earthquake
Transform fault
Eurasian
Mid-Atlantic
Plate
60°N Ridge
North
Eurasian Plate American
Juan de Fuca Plate Arabian
Plate Plate
Caribbean
Philippine San Andreas
30°N Plate
Plate Fault
Cocos
Mariana Plate African Plate
Trench Hawaii
0°
Indo- Pacific Plate

Australian South
Plate Nazca Plate American
30°S Plate
Antarctic Plate Scotia Plate

60°S
90°E 135°E 180° 135°W 90°W 45°W 0° 45°E

tem, is Valles Marineris (Figure 7.24). Stretching along the

Evidence of Tectonism equatorial zone for nearly 4,000 km, this system, if occur-
on Other Planets ring on Earth, would link San Francisco with New York.
Earth’s Grand Canyon would
Somewhat surprisingly, evidence of plate tectonics is found be little more than a minor Mars has experienced
only on Earth. But although spreading centers and sub- spur on the side of this chasm. extensive tectonism.
duction zones appear to be unique to our planet, all of the Valles Marineris includes a
terrestrial planets and some of their moons show evidence series of massive cracks in the lithosphere of Mars that are
of tectonic disruptions. Fractures have cut the crust of the thought to have formed as local forces, perhaps related to
Moon in many areas, leaving fault valleys such as the one mantle convection, pushed it upward from below. Once
shown in Figure 7.22. Most of these features are the result formed, the cracks were eroded by wind, water, and land-
of large impacts that crack and distort the lunar crust. slides, resulting in the massive chasm that we see today.
Mercury has fractures and faults similar to those on the Other parts of Mars show faults similar to those on the Moon,
Moon. In addition, there are numerous cliffs on Mercury but cliffs similar to those seen on Mercury are absent.
that are hundreds of kilome- Venus is similar to Earth in many respects. Venus has a
ters long (Figure 7.23). These Mercury’s surface mass about 0.81 times that of Earth, a radius 0.95 times that
appear to be the result of com- shrank after it cooled of Earth, and a surface gravity
pression of Mercury’s crust. from a molten state. 0.91 times that of Earth. As a Most of the surface of
Like the other terrestrial plan- result, many scientists specu- Venus is less than
ets, Mercury was once molten. After the surface of the planet lated that Venus might also 1 billion years old.
cooled and the crust formed, the interior of the planet con- show evidence of plate tecton-
tinued to cool and shrink. As the planet shrank, Mercury’s ics. However, these speculations were not borne out by the
lithosphere cracked and buckled in much the same way that Magellan mission, which used radar to peer through the
a grape skin wrinkles as it shrinks to become a raisin. To thick, dense layers of clouds that enshroud the planet.
explain the faults seen on the planet’s surface, planetary sci- Magellan mapped about 98 percent of the surface of Venus,
entists estimate that the volume of Mercury must have shrunk providing the fi rst high-resolution views of the surface of
by about 5 percent after formation of the planet’s crust.
On Mars the most impressive tectonic feature, and pos-
sibly the most impressive tectonic feature in the Solar Sys-
FIGURE 7. 23 Thrust faults on Mercury create cliffs hundreds
of kilometers long, such as the one seen here running from upper
left to lower right and alongside a double-ringed crater. Such
FIGURE 7. 22 An Apollo 10 photograph of Rima Ariadaeus, a features are consistent with compression of Mercury’s crust as
2-km-wide valley between two tectonic faults on the Moon. the planet cooled and contracted in size.
(a) (b)
FIGURE 7. 2 4 (a) A mosaic of Viking Orbiter images shows Valles Marineris, the major tectonic feature on
Mars, stretching across the center of the image from left to right. This canyon system is more than 4,000 km
long. The dark spots on the left are huge shield volcanoes in the Tharsis region. (b) This close-up perspective
view of the canyon wall was photographed by the European Space Agency’s Mars Express spacecraft.
the planet. Magellan’s view of one face of Venus is shown

in Figure 7.25. Although Venus has a wealth of volcanic
features and tectonic fractures, there is no evidence of litho-
spheric plates or plate motion of the sort seen on Earth. Yet
the relative scarcity of impact craters on Venus suggests that
most of its surface is less than 1 billion years old.
The absence of plate tectonics on Venus is a puzzle. The
interior of Venus should be very much like the interior of
Earth, and convection should be occurring in its mantle.
However, this model presents a problem: how does the
thermal energy from the interior of Venus escape from the
planet? On Earth, mantle convection and plate tectonism
provide a means for thermal energy to escape from the
interior of the planet. Earth also has a few hot spots, where
upwellings of hot mantle material rise, releasing thermal
energy. The Hawaiian Islands are the result of one such hot
spot. On Venus, hot spots may be the principal way that
thermal energy escapes from the planet’s interior. Circular
fractures (called “coronae”) on the surface of Venus, rang-
ing from a few hundred kilometers to more than 1,000 km
across, may be the result of upwelling plumes of hot mantle
FIGURE 7. 25 The atmosphere of Venus blocks our view of
that have fractured Venus’s lithosphere.
the surface in visible light. This false-color view of Venus is a
radar image made by the Magellan spacecraft. Bright yellow and
Some planetary scientists have suggested that a radically
white areas are mostly fractures and ridges in the crust. Some different form of tectonism may be at work on Venus. They
circular features seen in the image may be regions of mantle believe that hot spots are not adequate to allow the thermal
upwelling, or hot spots. Most of the surface is formed by lava energy generated within the planet to escape, and that as a
flows, shown in orange. result energy may continue to build up in the interior until
large chunks of the lithosphere melt and overturn. This the lithosphere and between tectonic plates, and concen-
melting and overturning would suddenly release an enor- trations of radioactive elements that produce energy from
mous amount of energy, after which the surface of the planet radioactive decay.
would cool and resolidify. This idea remains highly contro- Because the thermal-energy sources are not uniformly
versial, but it could help explain the relatively young surface distributed inside our planet, volcanoes tend to be located
of Venus. It also drives home the point of how different the only in specific areas, most
geological histories of the various planets appear to be. notably (but not exclusively) Friction between
Why Venus and Earth over hot spots and along plate moving plates
should have such different Earth’s tectonic boundaries. Maps of geological generates thermal
styles of tectonism remains plates are unique in activity, such as the one shown energy and leads
an unsolved puzzle. The seg- the Solar System. in Figure 7.21, leave little doubt to volcanism.
mentation of Earth’s litho- that most terrestrial volcanism
sphere into moving plates seems to be unique among the is ultimately linked to the same forces responsible for plate
planets and moons of the Solar System. motions. There is a tremendous amount of friction as plates
slide under each other at a subduction zone. This friction
generates a great deal of thermal energy, raising the tempera-
ture and pushing rock toward its melting point.
7.6 Igneous Activity: A When we studied Earth’s core, we found the counterin-
tuitive result that even though the inner core of the planet is
Sign of a Geologically hotter than its surroundings, it is solid while its surround-
ings are liquid. The reason is that the inner core is under
Active Planet more pressure than its surroundings, and this higher pres-
sure forces the material to stay solid. A similar effect occurs
Like tectonism, igneous activity is an important process near Earth’s surface. Material at the base of a lithospheric
that shapes planetary surfaces. On Earth, violent igne- plate is under a great deal of pressure because of the weight
ous events—including volcanic eruptions and some earth- of the plate pushing down on it. This pressure drives up the
quakes—can also result in widespread human casualties. In melting point of the material, forcing it to remain solid even
December 2004 a strong earthquake off the coast of Suma- though its temperature is above its normal melting point on
tra triggered an enormous tsunami, which swept across Earth’s surface. But as this material is forced up through the
the Indian Ocean. More than 225,000 people died in this crust, its pressure drops; and as its pressure drops, so does
disaster, one of the deadliest in modern history. Like earth- its melting point. Because of this declining pressure, mate-
quakes, volcanic eruptions throughout history have created rial that started out solid at the base of a plate may become
natural disasters on a grand scale. The eruption of Mount molten as it nears the surface.
Vesuvius in A.D. 79 that buried Pompeii; the 1883 explosion An obvious place to look for volcanic activity is along
of Krakatoa in the western Pacific that led to the loss of spreading centers, where convection carries hot mantle
36,000 lives; the hot ash flows from Mount Pelée that demol- material toward the surface. As mentioned in the previous
ished the Caribbean port of Saint-Pierre in 1902, killing all paragraph, the decrease in pressure as the material nears
but two of its 30,000 inhabitants—these are only a few of a the surface may allow it to become molten. Spreading cen-
long list of examples of death and destruction brought on ters are indeed found to be frequent sites of eruptions. Ice-
by volcanic eruptions. land, which is one of the most volcanically active regions
in the world, sits astride one such spreading center—the
Mid-Atlantic Ridge (see Figures 7.18 and 7.21).
Terrestrial Volcanism Is Once lava reaches the surface of Earth, it can form many
types of structures (Figure 7.26). Flows from spreading cen-
Related to Tectonism ters often form vast sheets, especially if the eruptions come
How do volcanoes form, and why are they found in some from long fractures called fissures. If very fluid lava flows
regions and not in others? Is there evidence of volcanoes from a single “point source,” it can spread out over the sur-
on other planets? To answer these questions we must fi rst rounding terrain or ocean floor, forming what is known as a
understand how and where magma—the main component shield volcano (Figure 7.26a), so named because it resembles
of volcanism—originates. an upside-down warrior’s shield. Viscous, pasty lava flows,
Magma does not come to Earth’s surface from its mol- alternating with pyroclastic (explosively generated) rock
ten core, as is sometimes believed. We know from seismic deposits, can form a steep-sided structure called a com-
signals that magma originates in the lower crust and upper posite volcano (Figure 7.26b) or smaller circular mounds
mantle, where sources of thermal energy combine. These called “volcanic domes.”
thermal-energy sources include rising convection cells in A third setting for terrestrial volcanism is found where
the mantle, frictional heating generated by movement in convective plumes rise toward the surface in the interi-
(a) FIGURE 7. 26 Magma

9 km reaching Earth’s surface
commonly forms (a) shield
volcanoes, such as Mauna Loa,
which have gently sloped sides
150 km built up by fluid lava flows; and
(b) (b) composite volcanoes, such
3 km as Vesuvius, which have steeply
symmetrical sides built up by
viscous lava flows. (c) Hot spots
are convective plumes of lava
that form a successive series
15 km
of volcanoes as the plate above
them slides by.
(c)
Hawaii
Kauai Oahu Molokai Maui
Loihi
Pacific
Ocean
Pacific Plate Hot spot
Mantle
ors of lithospheric plates, creating local hot spots (Figure Volcanism Also Occurs Elsewhere
7.26c). Volcanism over hot spots works much like volcanism
at a spreading center, except that the convective upwell-
in the Solar System
ing occurs at a single spot rather than along the length of Although Earth is the only planet on which plate tectonics
a spreading center (see Figure 7.20). These hot spots can is an important process, evidence of volcanism is found
melt mantle and lithospheric material and force it toward throughout the Solar System,
the surface of Earth. AstroTour: Hot Spot Creating a including several moons of the The dark areas on the
Chain of Islands outer planets. Even before the Moon’s surface are
There are numerous hot spots on Earth, including Apollo astronauts set foot on ancient lava flows.
the region around Yellowstone Park and the Hawaiian the lunar surface, photographs
Islands. The Hawaiian Islands are a chain of shield vol- showed it to have what appeared to be flowlike features in
canoes that formed as the lithospheric plate on which the dark regions. Some of the first observers to use telescopes
they ride was dragged across a relatively stationary hot thought that these dark areas looked like seas—thus the name
spot. Volcanoes erupt over a hot spot, building an island. maria (singular: mare), Latin for “seas.” Their appearance
The island ceases to grow as the plate motion carries the suggested to planetary scientists that these are vast lava flows
island away from the hot spot, which is the source for the similar to basalts on Earth. Because the maria contain rela-
volcanic activity. The slower process of erosion, going on tively few craters, we know that these volcanic flows occurred
since the island’s inception, continues to wear the island after the period of heavy bombardment ceased.
away. In the meantime, a new island grows over the hot When the Apollo astronauts returned rock samples from
spot. Today the Hawaiian hot spot is located off the south- the lunar maria, the rocks were indeed found to be basalts.
east coast of the big island of Hawaii, where it continues Many of these Moon rocks contained gas bubbles typical
to power the active volcanoes. On top of the hot spot, the of volcanic materials (Figure 7.27). Experiments show that
newest Hawaiian volcano is already forming. This vol- when this lava flowed across the lunar surface, it must
cano, called Loihi, remains submerged under the surface have been extremely fluid—something like the consistency
of the Pacific Ocean. Viewed another way, however, Loihi of motor oil at room temperature. The fluidity of the lava,
is already a massive shield volcano, rising more than 3 due partly to its iron- and titanium-rich chemical compo-
km above the ocean floor. Geologists expect that it will sition, explains why lunar basalts form vast sheets that fill
eventually break the surface of the ocean and merge with low-lying areas such as impact basins. It also explains the
the big island of Hawaii—but not anytime soon. Loihi is Moon’s lack of classic volcanoes like Mount Rainier: motor
not expected to show itself above sea level for perhaps oil poured from a can does not pile up; it spreads out (Fig-
another 100,000 years! ure 7.28).
FIGURE 7. 2 7 This rock sample from the Moon, collected

by the Apollo 15 astronauts from a lunar lava flow, shows gas
bubbles typical of gas-rich volcanic materials. This rock is about 6
by 12 cm.
FIGURE 7. 28 The lava flowing across the surface of Mare FIGURE 7. 29 The largest volcano yet seen on Mercury (right
Imbrium on the Moon must have been extremely fluid to spread of center at the top) lies on the edge of the Caloris Basin. The
out for hundreds of kilometers in sheets that are only tens of volcano is about 50 km across.
meters thick.
The samples also showed that most of the lunar lava

flows are older than 3 billion years! Only in a few limited
areas of the Moon are younger lavas thought to exist; most
of these have not been sampled directly. Samples from the
heavily cratered terrain of the Moon also originated from
magma, which shows that the young Moon went through a
molten stage. These rocks cooled from a “magma ocean” and
Edges of lava flow are more than 4 billion years old, preserving the early his-
tory of the Solar System. Thus, most of the sources of heat-
ing and volcanic activity on the Moon must have shut down
some 3 billion years ago—unlike on Earth, where volcanism
continues. This conclusion is certainly consistent with our
earlier argument that smaller planets should cool more effi-
ciently and thus be less active than larger planets.
Like the Moon, Mercury also shows evidence for past
volcanism. Mariner 10 and Messenger missions have
revealed smooth plains similar in appearance to the lunar
maria. These sparsely cratered plains are the youngest
areas on Mercury and, like those on the Moon, are almost
certainly volcanic in origin, where fluid lavas flowed into
and fi lled huge impact basins. High-resolution imaging
by Messenger has also identified a number of volcanoes
(Figure 7.29).
More than half the surface of Mars is covered with
volcanic rocks. Plain-forming lavas covered huge regions
of Mars, flooding the older, cratered terrain. Few of the
vents or fi ssures for these flows are visible, suggesting
FIGURE 7. 30 A Mars crater

flooded by lava is seen in this
perspective view taken by ESA’s
Mars Express spacecraft.
FIGURE 7. 31 The largest known volcano in the Solar System, Mars’s Olympus Mons is a 27-km-high shield-type volcano, similar
to but much larger than Hawaii’s Mauna Loa. (a) A partial view of Olympus Mons taken by the Mars Global Surveyor. (b) This
oblique view was created from an overhead Viking image and topographic data provided by the Mars Orbiter laser altimeter.
(a) (b)
that most were buried under the lava that poured forth from them (Fig-
ure 7.30). Among the most impressive features on Mars are the enormous
shield volcanoes. These volcanoes are the largest mountains in the Solar
System. Olympus Mons, standing 27 km high at its peak and 550 km wide at
its base (Figure 7.31), would tower over Mount Everest and dwarf Hawaii’s
Mauna Loa. Standing a “mere” 9 km above the floor of the Pacific Ocean
and spreading out to cover an area 120 km across, Mauna Loa is the largest
mountain on Earth.
Despite the difference in size, most of the very large volcanoes of Mars
are shield volcanoes, just like their Hawaiian cousins. Olympus Mons and its
neighbors grew as the result of hundreds of thousands of individual eruptions
that sent lava flows running down their flanks. The difference in size between
Olympus Mons and Mauna Loa could be a result of the absence of plate tec-
tonics on Mars. As discussed earlier, the motion of the plate that Hawaii rides
on carries the Hawaiian volcanoes away from their hot spot after only a few
million years. The martian volcanoes, on the other hand, have remained over
their respective hot spots for billions of years, growing ever Lava flows and other volcanic landforms span nearly
taller and broader with each successive eruption. the entire history of Mars, estimated to extend from the
Although no samples have been returned directly from formation of crust some 4.4 billion years ago to geologically
Mars, analysis by instruments on landed spacecraft, remote recent times, and to cover more than half of the red planet’s
sensing data, and the shapes of the lava flows and volca- surface. But remember, “recent” in this sense could still be
noes all suggest that martian lavas are basalts much like more than 100 million years ago, or back to our own dino-
those found on Earth and the Moon. However, chemical saur age. Although some “fresh-appearing” lava flows were
analyses at the Pathfinder lander site suggest that the rocks identified on Mars, until rock samples are radiometrically
contain slightly more silica than typical basalts contain, dated, we will not know the age of these latest eruptions.
which could mean that the magma was partly differentiated Mars could, in principle, experience eruptions today.
chemically before it erupted. On the other hand, chemical Most of Venus is covered with volcanic materials or
analyses obtained in Gusev Crater by the Mars exploration tectonically disrupted rocks of presumed volcanic origin.
rover Spirit reveal basalts that are rich in iron, more like A geological timescale for Venus has not yet been devised,
those of the Moon. In Chapter 12 we will discuss evidence but from its relative lack of impact craters, most of the sur-
that certain meteorites found on Earth were probably blasted face is considered to be less than 1 billion years old. When
from the surface of Mars by impacts. Chemical analysis of volcanism began on Venus and whether volcanoes are still
these meteorites supports the view that volcanism on Mars active today remain unanswered questions.
involved basaltic lavas. Earth remains the champion for the diversity of volcanism.
Of all the terrestrial planets, Venus has the largest popu- Volcanic rocks are found throughout the rock record, while
lation of volcanoes. Radar images reveal a wide variety of compositions of magma span the spectrum of silica-rich to
volcanic landforms. These include highly fluid flood lavas super-iron-rich materials erupted directly from the mantle.
covering thousands of square kilometers, shield volcanoes
approaching those of Mars in size and complexity, dome
volcanoes, and lava channels thousands of kilometers long.
These lavas must have been extremely hot and fluid to flow
for such long distances. Some
7.7 Erosion: Wearing
of the volcanic eruptions on Venus has the largest Down the High Spots
Venus are thought to have population of
been associated with defor- volcanoes among the and Filling In the Low
mation of Venus’s lithosphere terrestrial planets.
above hot spots such as the Erosion is the great leveler of planetary surfaces. The term
circular fractures mentioned earlier. erosion covers a wide variety of processes that together
Although we know little about the composition of the serve to smooth out planetary terrain, wearing down the
volcanic rocks on Venus, the Soviet Venera landers did high spots and filling in the low. The fi rst step in the pro-
measure some surface compositions; for the most part, the cess of erosion is called “weathering,” in which rocks are
results suggest that lavas on Venus are basalts, much like broken into smaller pieces and may be chemically altered.
the lavas on Earth, the Moon, and Mars. It is presumed that For example, rocks on Earth are physically weathered along
the lavas on Mercury are basalts as well. shorelines, where they are broken into beach sand by pound-
In summary, what can we say about the volcanic his- ing waves; and along streambeds, where they are slammed
tories of the terrestrial planets and the Moon? After going together. Other weathering processes include chemical
through a molten state—a sort of “magma ocean” phase— reactions, such as the combining of oxygen in the air with
shortly after its birth, the Moon developed an ever-thicken- iron in rocks to form a type of rust. One of the most effi-
ing lithosphere overlying a mantle. Pockets of radioactive cient forms of weathering involves freeze-thaw cycles, dur-
materials in the Moon’s interior generated local reservoirs ing which liquid water runs into crevices and then freezes,
of magma. Some large impacts penetrated these reservoirs expanding and shattering the rock.
or otherwise triggered the release of magma to the surface After weathering occurs, the resulting debris can be car-
through fractures. Most of this volcanism ceased about 3 ried away by flowing water, glacial ice, or blowing wind and
billion years ago, although some minor eruptions could have deposited in other areas as
continued sporadically for another billion years. In all, less sediment. Where material is The actions of water
than 18 percent of the lunar surface is covered with volca- eroded, we can see features and wind produce the
nic rocks (excluding those cooled from the “magma ocean”). such as river valleys, wind- greatest amount of
Volcanism on Mercury probably mimicked that of the Moon. sculpted hills, or mountains erosion on Earth.
Many of the volcanic plains on Mercury are also associated carved by glaciers. Where
with impact scars. The ages of these plains are not known, eroded material is deposited, we see features such as river
but from superimposed impact craters we can conclude that deltas, sand dunes, or piles of rock at the bases of mountains
the plains are probably billions of years old. and cliffs. It is not surprising to fi nd that erosion is most
7.7 Erosion: Wearing Down the High Spots and Filling In the Low 207
FIGURE 7. 33 This high-resolution image taken by a camera

aboard the Mars Global Surveyor shows windblown, frost-covered
sand dunes in the north polar region of Mars.
FIGURE 7. 32 A view of the surface of Mars taken by the

Phoenix lander showing the rock-littered surface of the northern
plains. Fine-grained material between the rocks and pebbles is
wind-transported sand and dust. storms have been seen on Mars. These storms have been
known to blot out the visibility of the surface of the planet
for months on end.
Sand dunes are common on Earth and Mars, and some
efficient on planets where water and wind are present. On have been identified on Venus. They occur frequently wher-
Earth, where water and wind are so dominant, most impact ever moderately strong winds blow and there is a supply of
craters are worn down and filled in even before they are loose grains. The largest field of windblown sand on Mars
turned under by tectonic activity. If other processes were (Figure 7.33) is in the north polar area. Covering more than
not at work to form mountains, valleys, and other topo- 700,000 square kilometers (km2), it is comparable in size to
graphic relief, erosion would eventually wear planets like the largest fields of sand dunes on Earth. The most com-
Earth as smooth as billiard balls. mon wind-related features on Mars and Venus are wind
Even on the Moon and Mercury, which have no atmo- streaks (Figure 7.34). These surface patterns appear, disap-
spheres or running water, a type of erosion (albeit very slow) pear, and change in response to winds blowing sediments
is still at work. Radiation from the Sun and from deep space around hills, craters, and cliffs. They serve as local “wind
slowly decomposes some types of minerals, effectively weath- vanes,” telling planetary scientists about the direction of
ering the rock. Such effects are only “skin deep”—usually a local prevailing surface winds.
few millimeters at most—and could be considered a kind of Today Earth is the only planet where the temperature
rock “sunburn.” Impacts of micrometeoroids also chip away and atmospheric conditions allow extensive liquid surface
at rocks. In addition, landslides can occur wherever gravity water to exist. Water is an extremely powerful agent of ero-
and differences in elevation are present. Although landslide sion and dominates erosion on Earth. Every year, rivers
activity is enhanced by the lubricating effects of water, land- and streams on Earth deliver about 10 billion metric tons
slides are also seen on dry planets like Mercury and the of sediment into the oceans. Even though today there is no
Moon. Debris from landslides has even been seen on the tiny liquid water on the surface of Mars, at one time water flowed
moons of Mars and on asteroids (see Chapter 12). across its surface in such vast quantities as to make the
Earth, Mars, and Venus, Amazon River look like a backyard irrigation ditch. Huge,
on the other hand, do have The surfaces of Mars dry riverbeds such as those shown in Figure 7.35 attest to
atmospheres, and all three and Venus are also tremendous floods that poured across the martian surface.
planets show the effects of modified by winds. In addition, many regions on Mars show small networks
windstorms. Images of Mars of valleys that are thought to have been carved by flowing
(Figure 7.32) and Venus returned by spacecraft landers water. Some parts of Mars may even have once contained
show surfaces that have clearly been subjected to the forces oceans and glaciers.
of wind. Likewise, orbiting spacecraft have returned pic- The search for water is a primary quest in the exploration
tures showing sand dunes, wind-eroded hills, and surface of Mars. In 2004, NASA sent two instrument-equipped rov-
patterns called “wind streaks.” Planet-encompassing dust ing vehicles, Opportunity and Spirit, to search for evidence
(a) (b)
FIGURE 7. 34 (a) The bright patterns associated with these craters and hills on Mars resulted
from winds that redistribute sand and dust on the surface. The winds forming these streaks blew
from right to left. The area shown is about 160 by 185 km. (b) Bright patches of windblown dust
streak across the surface of Venus.
FIGURE 7. 35 Viking Orbiter images showing channels on Mars carved long ago by flowing water.
Liquid water cannot exist on the surface of Mars today. The area shown is about 100 km wide. Inset:
High-resolution Mars Global Surveyor image showing geologically recent “gullies” that may also have
been carved by water.
Water channels
7.7 Erosion: Wearing Down the High Spots and Filling In the Low 209
of water on Mars (see the opening photograph of Chapter (a)

5). The choice for Opportunity’s landing site was based on
orbital remote sensing data that suggested the presence of
hematite in an area near the martian equator. “Hematite” is
an iron-rich mineral that commonly forms in the presence of “Eagle” Crater
water, and the goal for this mission was to test the validity
of the remote sensing information. Both Mars rovers were Lander
surrounded by inflatable airbags to cushion their impact
on the martian surface. Upon landing, the whole Opportu- Rover track
nity package bounced across the surface, fi nally coming to
rest inside a small crater like a well-driven golf ball (Figure
7.36a). The real payoff came when the walls of the crater
revealed a treasure trove of information. For the fi rst time, Landing rocket
outcrops of martian rocks were observed and available for blast effects
study. “Outcrops” are places were exposed rocks remain in
the original order in which they were laid down. Previously, Backshell and
the only rocks that landers and rovers had come across were parachute
those that had been dislodged from their original settings
by either impacts or river floods.
(b)
The layered rocks at the Opportunity site (Figure 7.36b)
revealed several lines of evidence indicating that they had
once been soaked in or transported by water. First, the
form of the layers was typical of what we see here on Earth
in sandy sediments that have been laid down by gentle
currents of water.78 Then, rover instruments detected the
presence of a mineral commonly found in acid lakes and
thermal springs, with so much sulfur content that it had
almost certainly formed by precipitation from water. Finally,
magnified images of the rocks showed “blueberries,” small
spherical grains a few millimeters across that appeared to
have formed in place among the layered rocks in a manner
FIGURE 7. 36 (a) An orbital view of Opportunity’s landing site.
similar to terrestrial features that form by the percolation
After landing, the protected package bounced along the surface
of water through sediments. Analysis of the “blueberries”
and ended up in the “Eagle” crater—a landing that mission
revealed abundant hematite, confi rming the interpretations scientists called a “hole in one.” The white dot in the middle of
of the remote sensing data. Observations by the European the crater is the lander. Tracks leading out of the crater were
Space Agency’s Mars Express orbiter and NASA’s Mars Odys- made by the Mars rover Opportunity as it began its exploration of
sey show the hematite signature and the presence of sulfur- Mars. (b) Layered rock seen near the Opportunity landing site may
rich compounds in a vast area surrounding the Opportunity represent sediments deposited by water.
landing site. These observations suggest the existence of an
ancient martian sea larger than the combined area of the
Great Lakes and as much as 500 meters deep.
The floor of Gusev, a 170-km impact crater, was chosen water. As of this writing, Opportunity and Spirit continue
for Spirit because it showed signs of ancient flooding by a to explore the martian surface.
now dry river, and it was hoped that surface deposits would Where is the water and
provide evidence for that past liquid water. Surprise, and ice on Mars today? At least Water ice could exist
perhaps some disappointment, followed when Spirit revealed some ice is locked in the polar on the Moon
that the flat floor of Gusev consisted primarily of basaltic regions, just as the ice caps and Mercury as it
rock. Only when the rover ventured cross-country to some on Earth hold much of our does on Mars.
low hills located 2.5 km from the landing site did it fi nd planet’s water. But, unlike
evidence of water in the martian past. Here the basaltic our own polar caps, those on Mars are a mixture of frozen
rocks showed clear signs of chemical alteration by liquid carbon dioxide and frozen water, so water must be hiding
elsewhere on Mars. Small amounts of water can be found
8
7
Some geologists believe that this layering was caused by deposits of on the surface, as seen in Figure 7.37, and NASA’s Phoenix
volcanic ash rather than aqueous sediments. We fi nd here an example lander has found water ice just a centimeter or so beneath
of honest disagreement among scientists, all of whom are looking at surface soils at high northern latitudes (Figure 7.38). How-
the same data. ever, much if not most of the water on Mars appears to
FIGURE 7. 37 This image, taken by ESA’s Mars Express

mission, shows water ice in a crater in the north polar region
of Mars. The crater is about 35 km in diameter, and the ice is
estimated to be about 200 meters thick. The 300-meter-high
crater wall blocks much of the sunlight at this high latitude,
keeping the ice from vaporizing. FIGURE 7. 38 Water ice appears a few centimeters below the
surface in this trench dug by a robotic arm on the Phoenix lander.
The trench measures about 20 by 30 cm.
be trapped well below the surface of the planet. Recent

radar imaging by NASA’s Mars Reconnaissance Orbiter
indicates huge quantities of subsurface water ice, not only
in the polar areas as expected but also at lower latitudes
(Figure 7.39).
It is also intriguing that although Earth and Mars are
the only terrestrial planets that show evidence for liquid
water at any time in their histories, water ice could exist
on Mercury and the Moon today. Some deep craters in the
polar regions of both Mercury and the Moon have floors
that are in perpetual shadow. Because these planets lack
atmospheres, temperatures in these permanently shad-
owed areas remain below 180 K. For many years, plan-
etary scientists speculated that ice—perhaps implanted
by impact comets—could be found in these craters. In the
early 1990s, Earth-based radar measurements of Mercury’s
north pole and infrared measurements of the Moon’s polar
areas by the joint Department of Defense/NASA Clemen- FIGURE 7. 39 Radar observations by NASA’s Mars
tine mission returned information that seemed to support Reconnaissance Orbiter indicate huge quantities of water ice
this possibility. In 1998, NASA sent another spacecraft, underlying these mid-latitude martian craters.
Lunar Prospector, to take a new look. Again the results
suggested subsurface water ice in the polar regions. When
its primary mission was completed, NASA crashed Lunar
Prospector into a crater near the Moon’s south pole while Although these observations do not prove one way or
groundbased telescopes searched for evidence of water another whether ice exists in these regions, the possibil-
vapor above the impact site. None was seen. Perhaps it ity is exciting and important. The existence of ice on the
was the wrong place to look or the amount of water was Moon could make any future lunar colonization much more
too small to be detected. In 2009, NASA's Lunar Recon- practical than if all water had to be either brought up from
naissance Orbiter was put into lunar orbit to continue the Earth or synthesized from hydrogen and oxygen extracted
search for possible sources of subsurface water ice. from the lunar soil.
in Andy Rooney’s tongue-in-cheek quote that opened

Seeing the Forest for the Trees the chapter, we are creatures of the surface of Earth,
as far from direct exploration of the inner reaches of
In Chapter 6 we placed the formation of the Solar System our own world as we are from direct exploration of the
and our Earth into the broader context of the ongoing heart of the Sun. Even so, on this leg of our journey we
process of star formation that we see occurring around have seen how the methods of science have enabled us
us in the universe today. Earth is but one of several rocky to construct a detailed picture of the interior of Earth.
planets that coalesced from the dust surviving in the hot We have built up that picture by applying physical laws
inner parts of the disk that surrounded the young Sun. to construct models of Earth and then tested the pre-
Mercury, Venus, Earth, and Mars share this common dictions of those models by analyzing the echoes from
heritage and have been further shaped by the same fun- earthquakes. At fi rst glance this might seem an inse-
damental processes of impact cratering, tectonism, vol- cure foundation for knowledge, but that is far from the
canism, and erosion over the ensuing 4.5 billion years. case. We go through our lives building knowledge of the
Earth is our benchmark for understanding the nature world around us through light waves that reach our eyes
and effects of these processes. The cosmological prin- and sound waves that reach our ears, both interpreted
ciple not only suggests that we can apply physical laws by brains shaped by evolution to accomplish the task.
discovered in terrestrial laboratories throughout the Is it so different to say that we know about the interior
universe; it also guides us in our exploration of the Solar of Earth from the seismic waves that reach our instru-
System. We live in a remarkable age of exploration and ments, interpreted through computer models that are
discovery, when our species has launched probes to visit shaped for the job by our understanding of physical
almost all of the major bodies in the Solar System and law? So it will be throughout the rest of our journey as
many of the minor bodies as well. The resulting flood we use the techniques and tools of science (and a care-
of information is a testament to 20th and 21st century fully thought-out notion of what it means “to know”) to
technology; but it is our understanding of Earth that has “sense” the nature of the universe.
guided scientists as they have pieced together the story We have seen several ways in which the study of
told by these data. terrestrial planets has forced us to change the way we
The impression we have gleaned from the results of view our planet. Earth is but one of several rocky plan-
the last five decades of planetary exploration is one of ets, and our world is only one small corner of the range
remarkable diversity. The surfaces of the four terrestrial of the possible. Yet no result of Solar System explora-
planets and Earth’s moon have all been stressed and tion has more dramatically changed our perspective on
fractured over the eons. One of the most startling dis- our own history—and perhaps our own future—than
coveries of the last century was the fact that the outer our understanding of the role of impact cratering in the
shell of Earth itself consists of multiple plates that are in inner Solar System. The active Earth is very effective at
constant motion as they ride slow but inexorable convec- erasing its own memory. Plate tectonics, igneous activity,
tion currents within Earth’s mantle. Exploration of the and erosion are continually at work, wiping the record
inner Solar System has provided new perspectives on the of our history clean. It took the ancient surfaces of Mer-
drifting motion of our planet’s lithosphere by showing cury and the Moon, and even the less ancient surfaces of
us that other possibilities exist. Whereas Mercury’s sur- Mars and Venus, to show us that impacts of objects from
face is cracked and wrinkled as a result of compression, space have played a significant role in shaping Earth.
tectonic activity on Mars fractured the planet, forming a With this realization, the cosmological principle truly
vast canyon system. Even our sister planet, Venus, shows comes full circle.
hints of a particularly violent form of tectonism in which Earth may provide many of the clues that we need to
the entire surface of the planet may have overturned and understand the Solar System, but it is the Solar System
released bursts of pent-up thermal energy from the plan- that provides the immediate context for the existence of
et’s interior. Likewise, igneous activity of several kinds Earth. Life on our planet has had its course altered time
has played an important role in the history of our planet, and again by sudden and cataclysmic events when aster-
building such monuments to the power of volcanism as oids and comets have slammed into Earth. It seems very
the Hawaiian Islands and the Cascade Range. Yet when likely that we owe our existence to the luck of our remote
looking at our neighbors, we fi nd that Earth’s volcanoes ancestors—small rodentlike mammals—that could live
shrink into relative insignificance beside the towering amid the destruction following such an impact 65 mil-
heights of Olympus Mons on Mars. lion years ago. The time since that event is but a brief
Along the way we have again seen the process of moment in the history of our planet. The time until the
science at work. The realm of our species encompasses next such event will almost certainly be no more than
only the tiniest fraction of our planet. As alluded to another short span of geological time.
Summary *6. Can all rocks be dated with radiometric methods?

Explain.
• Comparative planetology is the key to understanding 7. Explain how we know that rocks at the bottom of Ari-
the planets. zona’s Grand Canyon are older than those found on the
rim.
• Impacts, volcanism, and tectonism create topography
on the terrestrial planets. 8. A current theory suggests that a mass extinction occurred
as a consequence of an enormous impact on Earth 65 mil-
• Relative concentrations of impact craters divulge the
lion years ago. What is the evidence for this theory?
ages of planetary surfaces.
*9. Describe the sources of heating that are responsible for
• Radiometric dating tells us the ages of rocks.
the generation of Earth’s magma.
• Seismic waves reveal Earth’s interior structure.
10. Explain why the Moon’s core is cooler than Earth’s.
• The Moon was created when a Mars-sized protoplanet
11. Name the three components that make up Earth’s
collided with Earth.
interior.
• Earth has a strong magnetic field, but Venus and Mars
12. What is meant by differentiation of a planet, and what
do not.
causes it?
• Smooth areas on the Moon and Mercury are ancient
*13. Explain the difference between longitudinal waves and
lava flows.
transverse waves.
• The largest mountains in the Solar System are volca-
14. How do we know that Earth’s core includes a liquid
noes on Mars.
zone?
• Erosion wears down the surfaces of Venus and Mars by
15. Explain what is meant by hydrostatic equilibrium.
wind and the surface of Earth by wind and water.
16. Compare and contrast tectonism on Venus, Earth, and
• Mars today has large amounts of subsurface water ice
Mercury.
and once had liquid water on its surface.
17. Explain plate tectonics and identify the only planet on
which this process has been observed.
18. Volcanoes have been found on all of the terrestrial plan-
ets. Where are the largest volcanoes in the inner Solar
System?
your instructor assigns questions in SmartWork, log in at *19. Describe and explain the criteria you would apply to
smartwork.wwnorton.com. images with adequate resolution in order to distinguish
between a crater formed by an impact and one formed
by a volcanic eruption.
Student Questions 20. What are the differences between a spreading center and
a subduction zone?
21. Describe and explain the evidence for reversals in the
1. Name the four terrestrial planets. polarity of Earth’s magnetic field.
2. In discussing the terrestrial planets, why do we include 22. What is meant by erosion? What processes contribute to
our Moon? erosion?
*3. What do we mean by comparative planetology, and why 23. What are the primary reasons that the surfaces of Venus,
is it important? Earth, and Mars are younger than those of Mercury and
the Moon?
4. Name the four forces that have shaped Earth.
24. Explain some of the evidence that Mars once had liquid
5. One region on the Moon is covered with craters, while
water on its surface.
another is a smooth volcanic plain. Which is older? How
do we know? 25. Why do we not fi nd liquid water on Mars today?
A P P LY I N G T H E C O N C E P T S *32. Suppose you find a piece of ancient pottery and take it

to the laboratory of a physicist friend. He fi nds that the
26. Assume that Earth and Mars are perfect spheres with
glaze contains radium, a radioactive element that decays
radii of 6,371 km and 3,390 km, respectively.
to radon and has a half-life of 1,620 years. He tells you
a. Calculate the surface area of Earth.
that there could not be any radon in the glaze when the
b. Calculate the surface area of Mars.
pottery was being fi red, but that it now contains three
c. If 0.72 (72 percent) of Earth’s surface is covered with
atoms of radon for each atom of radium. How old is the
water, compare the amount of Earth’s land area with
pottery?
the total surface area of Mars.
27. Compare the kinetic energy of a 1-gram piece of ice (about 33. Assume that the east coast of South America and the
half the mass of a dime) entering Earth’s atmosphere at a west coast of Africa are separated by an average dis-
speed of 50 km/s with that of a 2–metric ton SUV (mass tance of 4,500 km. Assume also that GPS measurements
= 2 × 103 kg) speeding down the highway at 90 km/h. indicate that these continents are now moving apart at a
rate of 3.75 cm/year. If this rate has been constant over
28. The object that created Arizona’s Meteor Crater was
geological time, how long ago were these two continents
estimated to have a diameter of 50 meters and a mass
joined together as part of a supercontinent?
of 300 million kg. Calculate the density of the impact-
ing object, and explain what that may tell you about its *34. Shield volcanoes are shaped something like flattened
composition. cones. The volume of a cone is equal to the area of its
29. Using the information in Table 7.1, determine the rela- base multiplied by one-third of its height. The largest
tive rates of internal energy loss experienced by Earth volcano on Mars, Olympus Mons, is 27 km high and has
and the Moon. a base diameter of 550 km. Compare its volume with
that of Earth’s largest volcano, Mauna Loa, which is 9
30. Although oceans cover 72 percent of Earth’s surface, km high and has a base diameter of 120 km.
they represent but a tiny fraction of our planet’s mass.
Earth’s mass is 6.0 × 1024 kg, and its oceans have a total *35. Earth’s core represents about one-third of Earth’s total
mass of 1.5 × 1021 kg. mass. Assume that the diameter of Earth’s core is 20 per-
a. What fraction of Earth’s total mass do our oceans cent of Earth’s diameter. What fraction of Earth’s total
represent? volume does the core represent, and what does that tell
b. Does surface and atmospheric water represent Earth’s you about the composition of Earth’s core?
total aqueous inventory? Explain.
31. Earth’s mean radius is 6,371 km, and its mass is 6.0 × StudySpace is a free and open Web site that pro-
1024 kg. vides a Study Plan for each chapter of 21st Century
a. Calculate Earth’s average density. Show your work. Astronomy. Study Plans include animations, read-
Do not look this value up. ing outlines, vocabulary flashcards, and multiple-choice
b. The average density of Earth’s crust is 2,600 kg/m3. quizzes, plus links to premium content in SmartWork and
What does this tell you about Earth’s interior? the ebook. Visit www.wwnorton.com/studyspace.
Woe to you wicked spirits! hope not
Ever to see the sky again. I come
To take you to the other shore across,
Into eternal darkness, there to dwell
In fierce heat and in ice.
DAN TE ALIGHIERI (1265 –1321), THE DIVINE COMEDY: INFERNO , CAN TO III
Lightning strikes the ground near Kitt Peak National Observatory in Arizona.
CHAP TER 8
Atmospheres of the
Terrestrial Planets
8.1 Atmospheres Are
K E Y CONCEP T S
Oceans of Air A thick blanket of atmosphere warms and sustains
Earth’s atmosphere surrounds us like an ocean of air. We Earth’s temperate climate. On Venus, by contrast,
see it in the blueness of the sky and feel it in the breezes that a thick, sulfurous atmosphere pushes the planet’s
tousle our hair. People who live in large cities can sometimes surface beyond medieval poet Dante Alighieri’s worst
even smell it. It can bring joy into our lives in the form of descriptions of hell. A thin atmosphere leaves Mars
a spectacular sunset, or apprehension with the approach frozen. Just as we did with their surfaces in Chapter
of a hurricane or tornado. It is responsible for all of our 7, we now compare the atmospheres of the terrestrial
weather, be it pleasant or stormy. Without our atmosphere planets. Among many interesting insights, on this leg of
there would be neither clouds nor rain; nor streams, lakes, our journey we will discover
or oceans. There would be no living creatures. Without an
atmosphere, Earth would look somewhat like the Moon. • That terrestrial planets owe their atmospheres to
And we, quite simply, would not exist. volcanism and to volatiles captured from comets.
Among the five terrestrial bodies that we discussed in • Why some planets hold on to their atmospheres while
Chapter 7, only Venus and Earth have dense atmospheres others do not.
(Figure 8.1). Mars has a very low-density atmosphere, and
• That differences among Earth, Venus, and Mars are
the atmospheres of Mercury and the Moon are so sparse that
they can hardly be detected. Why should some of the terres- due largely to the atmospheric greenhouse effect.
trial planets have dense atmospheres while others have little • How Earth’s atmosphere has been reshaped by life.
or essentially none? Are atmospheres created right along • That atmospheres are layered by convection and
with the planets they envelop, or do they appear later? For differences in how they are heated, while atmospheric
answers to these questions we need to look back nearly 5
pressure steadily falls at higher and higher altitudes.
billion years to the story told in Chapter 6—to a time when
the planets were just completing their growth. • That Earth’s magnetic field interrupts the flow of the
solar wind and traps charged particles in a huge
magnetosphere responsible for auroras.
Some Atmospheres • That Venus has a hot, dense atmosphere and Mars
Appeared Very Early has a cold, thin atmosphere.
Planetary atmospheres form in a series of phases, as illus- • The unsettling fact that we are living through an
trated in Figure 8.2. The young planets at that time were uncontrolled experiment in climate modification.
still enveloped by the remaining hydrogen and helium that
215
216 Chapter 8 Atmospheres of the Terrestrial Planets
(a) (b) (c)

FIGURE 8 .1 Global views
of the atmospheres of Venus
(a), Earth (b), and Mars (c).
Mercury and the Moon have no
atmospheres to speak of.
filled the protoplanetary disk surrounding the Sun, and type has the same average energy, then the less massive mol-
they were able to capture some of this surrounding gas. Gas ecules must be moving faster than the more massive ones.
capture must have continued For example, in a mixture of hydrogen and oxygen at room
until the gaseous disk ulti- Primary atmospheres temperature, hydrogen molecules will be rushing around
mately dissipated (soon after consist of at about 2,000 meters per second (m/s) on average, while
formation of the planets) and captured gas. the much more massive oxygen molecules will be poking
the supply of gas ran out. The along at a sluggish 500 m/s. Remember, though, that these
gaseous envelope collected by a newly formed planet is are the average speeds. A small fraction of the molecules
called its primary atmosphere. Although the giant planets will always be moving much slower than average, while a
still retain most of their original primary atmospheres, the few will be moving much faster than average.
terrestrial planets probably lost theirs soon after the proto- Deep within a planet’s atmosphere, these high-speed
planetary disk was blown away by the emerging Sun. Why molecules will almost certainly collide with other mol-
did only the terrestrial planets lose their primary atmo- ecules before they have a chance to escape. During the
spheres? The answer lies in their relatively small masses. collision process, there is an exchange of energy. The high-
The terrestrial planets, with their weak gravity, lack the speed molecule usually emerges with a lower speed, and
ability to hold light gases such as hydrogen and helium. the slower one tends to move faster. After a collision, then,
When the supply of gas in the both are likely to move with speeds closer to the average.
disk ran out, their primary A planet’s atmosphere There are fewer surrounding molecules near the top of the
atmospheres began leaking can sometimes atmosphere, and a high-speed molecule has a good chance
back into space.1 How can gas escape into space. of escaping before colliding with another molecule if it is
molecules escape from a heading more or less upward. Less massive molecules and
planet? All it takes for any object—from a molecule to a atoms, such as hydrogen and helium, move faster and are
spacecraft—to escape a planet is a speed greater than the more quickly lost to space than more massive molecules
escape velocity (see Chapter 3) and pointed in the right such as nitrogen or carbon dioxide.
direction. Intense radiation from the Sun—which is the Heated by the Sun and lacking a strong gravitational
primary source of thermal, or kinetic, energy in the atmo- grasp, the terrestrial planets soon lost the hydrogen and
spheres of the terrestrial planets—raises the temperature helium they had temporarily acquired from the protoplan-
and thus the kinetic energy of atmospheric atoms and mol- etary disk. Thus, the primary atmospheres were lost early
ecules enough that some may escape. The kinetic energy of in the evolutionary development of the terrestrial planets.
any object is determined by its mass and its speed. Hotter Born naked, they were naked once more. The giant planets,
molecules have higher kinetic energies than do cooler mol- on the other hand, are far more massive than their terres-
ecules and therefore move faster. trial cousins and are situated in the cooler environment of
Let’s look more closely at how molecules move within a the outer Solar System. Stronger gravity and lower tempera-
planetary atmosphere. When a volume of air contains mol- tures have enabled them to retain nearly all of their mas-
ecules with different masses, the average kinetic energy sive primary atmospheres. AstroTour: Atmospheres:
tends to be distributed equally among the different types. Formation and Escape
In other words, each type of molecule, from light to mas-
sive, will have the same average kinetic energy. But if each
1
Some Atmospheres Developed Later
Not all atmospheric loss comes from slow leakage. Impacts by large
planetesimals may have blasted away substantial amounts of the ter- Given that Earth’s primary atmosphere was lost early in its
restrial planets’ primary atmospheres. history, what is the source of the air we breathe today? Three
8.1 Atmospheres Are Oceans of Air 217
sources likely contributed to what is known as a secondary

1 Gases in a protoplanetary atmosphere: accretion, volcanism, and impacts. During the
disk, mostly hydrogen and helium…
planetary accretion process,
minerals containing water, Secondary
carbon dioxide, and other vol- atmospheres are
atile matter collected in the a result of accretion,
interiors of the terrestrial volcanism, and
Protoplanetary planets. Later, as the interiors comet impacts.
2 …were captured by young
disk planets, forming primary heated up, the higher temper-
atmospheres.
atures released these gases from the minerals that had held
them. Volcanism then brought the various gases to the sur-
Sun face, where they accumulated and created our secondary
atmosphere. Many planetary scientists now believe that
there was another important source of gas that formed the
3 Sunlight heated the
Young planets
atmospheres. Rapid thermal
secondary atmospheres of the terrestrial planets: impacts
with primary by huge numbers of comets, which had formed in the outer
motion of light atoms and
atmospheres
molecules caused the primary parts of the Solar System and were therefore rich in vola-
atmospheres to escape.
tiles (see Chapter 6).
Why did these icy bodies come into the inner Solar
Sun System? Their orbits were disrupted by the growth of the
giant planets. As the giant planets of the outer Solar System
grew to maturity, their gravitational perturbations stirred
up the entire population of icy planetesimals (comet nuclei)
4 Volcanoes released gases…
that orbited within their domain. Many of these icy bodies
were flung outward by the giant planets to form the parts
of our Solar System that we call the Kuiper Belt and the
Oort Cloud. (We will discuss these remote regions in Chap-
ter 12.) Other comets were scattered into the inner parts of
the Solar System, where they could rain down on the sur-
faces of the terrestrial planets. These comet nuclei brought
with them ices such as water, carbon monoxide, methane,
and ammonia. Cometary water mixed with the water that
had been released into the atmospheres by volcanism. On
Earth, and perhaps Mars as well, most of the water vapor
then condensed as rain and flowed into the lower areas to
form the earliest oceans.
Comet 5 …and comets contributed
nucleus water and other molecules to Other cometary gases were not able to survive in their
form secondary atmospheres. original form. Ultraviolet (UV) light from the Sun easily
fragments cometary molecules such as ammonia and meth-
ane. Ammonia, for example, is broken down into hydrogen
and nitrogen. When this happens, the lighter hydrogen
atoms quickly escape to space, leaving behind the much
Larger planet heavier nitrogen atoms. The nitrogen atoms then combine
to form more massive nitrogen molecules, making it even
Smaller planet less likely that these molecules will escape into space.
Decomposition of ammonia by sunlight became the primary
Secondary
source of molecular nitrogen in the atmospheres of the ter-
atmosphere
restrial planets and on one of Saturn’s moons, Titan. Molec-
6 Larger planets’ 7 Smaller planets’
gravity could retain gravity could not retain
ular nitrogen makes up the bulk of Earth’s and Titan’s
an atmosphere: an atmosphere: atmospheres.
•Venus •Mercury Among the terrestrial Venus, Earth, and
•Earth •Earth’s Moon
•Mars
planets, today only Venus, Mars have significant
Earth, and Mars have signifi- secondary
cant secondary atmospheres. atmospheres.
FIGURE 8. 2 Phases in the formation of the atmosphere of
a terrestrial planet.
What happened in the case of
Mercury and the Moon? Even if these two bodies had experi-
enced less volcanism than the other terrestrial planets, they

could hardly have escaped the early bombardment of comet TA B L E 8 .1
nuclei from the outer Solar System. Some carbon dioxide
and water must have accumulated during volcanic eruptions
and comet impacts. Where are these gases now? Atmospheres of the
It appears that Mercury’s relatively small mass and its Terrestrial Planets
proximity to the Sun caused it to lose all of its secondary
Physical Properties and Composition
atmosphere to space, just as it
had earlier lost its primary Mercury and the
atmosphere. More massive Moon have basically Planet
molecules, such as carbon no atmosphere.
Venus Earth Mars
dioxide, can escape from a
small planet if the temperature is high enough. Further-
Surface pressure (bars) 92 1.0 0.006
more, intense ultraviolet radiation from the Sun can break
molecules into less massive fragments, which are lost to Surface temperature (K) 737 288 210
space even more quickly. Because the Moon is much farther
Carbon dioxide (%) 96.5 0.039 95.3
from the Sun than Mercury is, the Moon is much cooler
than Mercury, but its mass is so small that even at relatively Nitrogen (%) 3.5 78.1 2.7
low temperatures molecules can easily escape. From the
Oxygen (%) 0.00 20.9 0.13
beginning, a combination of their small mass and relative
proximity to the Sun doomed both Mercury and the Moon Water (%) 0.002 0.1 to 3 0.02
to remain almost totally “airless.”
Argon (%) 0.007 0.93 1.6
Sulfur dioxide (%) 0.015 0.02 0.00
8.2 A Tale of Three

Planets—The position, but it has a mass only about a tenth that of Earth
or Venus. In Table 8.1 we see that the atmospheres of Venus
Evolution of Secondary and Mars today are nearly identical in composition: mostly
carbon dioxide, with much smaller amounts of nitrogen.
Atmospheres Now let’s explore their differences. Table 8.1 shows us
that despite the similarities in atmospheric composition,
So far, we’ve discussed how the terrestrial planets gained the surface pressures of Venus and Mars are very different.
and lost their primary atmospheres; how their second- Mars and Venus have vastly different amounts of atmo-
ary atmospheres formed; and how only Earth, Venus, and sphere. The atmospheric pressure on the surface of Venus is
Mars retained their secondary atmospheres. We’ll now com- nearly a hundred times greater than Earth’s. By contrast, the
pare and contrast these secondary atmospheres, which are average surface pressure on Mars is less than a hundredth
quite different from one another. Would we have expected of our own. Earth differs in another important respect in
such differences? To answer this question, we need to that, alone among the planets, its atmosphere is made up
examine the processes that affected the evolution of these primarily of nitrogen and oxygen, with only a trace of car-
atmospheres. bon dioxide. Although all of these planets must have started
out with atmospheres of similar composition and compa-
rable quantity, they ended up being very different from one
another. Why did they evolve so differently?
Similarities and Differences
Let’s look at the similarities fi rst. All three of these planets
either are volcanically active today or have been volcani- How Mass Affects a
cally active in their geological past, and all must have shared
the intense cometary showers of the distant past. Their
Planet’s Atmosphere
similar geological histories suggest that the early second- As mentioned earlier, Venus and Mars have similar atmo-
ary atmospheres of these planets might also have been quite spheric compositions—except for water abundance, which
similar. In addition, two of the terrestrial planets—Venus we will discuss later. Both planets experienced widespread
and Earth—are similar in both mass and composition, and volcanism at some time during their history. Evidence sug-
they have adjacent orbits that are less than 0.3 astronomical gests that Venus might still be volcanically active, and Mars
unit (AU) apart. The third—Mars—is also similar in com- has certainly been volcanically active in the recent past. Car-
8.2 A Tale of Three Planets—The Evolution of Secondary Atmospheres 219
bon dioxide and water vapor must have poured out into the tion against emitted thermal radiation, and found that Earth
emerging secondary atmospheres of both Venus and Mars in is somewhat warmer than expected, while Venus is very
the form of volcanic gases. Decomposed cometary ammonia much hotter than our simple model predicts. When the
was the likely source of nitrogen on both planets. predictions of the model fail, the implication is that we are
The major distinction between the two planets cannot leaving something out of the model. In this case the “some-
be explained in terms of planetary mass alone. Venus has thing” is the atmospheric greenhouse effect.
nearly eight times as much The atmospheric greenhouse effect in planetary atmo-
mass as Mars, so we can Venus retained spheres and the conventional greenhouse effect operate
assume that it probably had its atmosphere in somewhat different ways, although the end results are
about eight times as much car- more effectively much the same. Planetary atmospheres and the interiors of
bonate within its interior to than Mars did. greenhouses are both heated by trapping the Sun’s energy,
produce carbon dioxide, the but here the similarities end. A good example of the con-
principal secondary-atmosphere component of both planets. ventional greenhouse effect is what happens in a car on a
Even allowing for the differences in mass, however, Venus sunny day when you leave the windows closed. Sunlight
today has more than 2,500 times more atmospheric mass pours through the car’s windows, heating the interior and
than Mars. Why such a large difference? We can find the raising the internal air temperature. With the windows
answer by considering the relative strengths of their surface closed, convection is unable to carry the hot air away, and
gravity, which involves both the mass and radius of a planet. temperatures can climb to as high as 180°F. Heating by solar
Venus has the gravitational pull necessary to hang on to its radiation is most efficient if the enclosure is transparent,
atmosphere; Mars has less gravitational attraction to keep which is why the walls and roofs of real greenhouses are
its atmosphere. Furthermore, when a planet such as Mars made mostly of glass.
loses so much of its atmosphere to space, the process begins In the case of planetary atmospheres, convection oper-
to take on a runaway behavior. With less atmosphere, there ates freely, and it is not hot air that is trapped, but rather the
are fewer intervening molecules to keep breakaway mole- electromagnetic energy received from the Sun. The atmo-
cules from escaping, and the rate of escape increases. This spheric greenhouse effect is illustrated in Figure 8.3. Atmo-
process in turn leads to even less atmosphere and even spheric gases such as nitrogen, oxygen, carbon dioxide, and
greater escape rates. water vapor freely transmit visible solar energy, allowing the
Sun to warm the planet’s surface. The warmed surface now
tries to radiate the excess energy back into space according
to the temperature of that surface (see Chapter 4), which is
The Atmospheric Greenhouse Effect much lower than that of the Sun. At the typical temperatures
Differences in the present-day masses of the atmospheres of planetary surfaces, this energy is reradiated as infrared
of Venus, Earth, and Mars have a large effect on their sur- (IR) radiation. But carbon dioxide and water vapor—among
face temperatures, and thus on the evolution of their atmo- other kinds of atmospheric molecules—strongly absorb
spheres. Solar radiation can be trapped by the atmospheric IR radiation, converting it to thermal energy. These same
greenhouse effect. In Chapter 4, we calculated the expected molecules subsequently reradiate this thermal energy in
temperatures of the planets, balancing absorbed solar radia- all directions—some of the thermal energy continues into
1 Visible sunlight passes through

the atmosphere and heats FIGURE 8 . 3 Greenhouse
the ground. gases such as water and
carbon dioxide trap infrared
2 The warm ground 3 Some infrared radiation
radiation, increasing a planet’s
radiates in infrared. escapes to space…
temperature.
4 …but some is trapped by green- 5 The temperature climbs until the

house gases and is reradiated back escaping infrared radiation balances
to the ground, heating it further. the absorbed sunlight.
space, but much of it goes back to the ground. The surface house effect. Imagine a hypothetical situation in which large
is now receiving thermal energy both from the Sun and quantities of carbon dioxide or another greenhouse gas sud-
from the atmosphere. denly appeared in Earth’s atmosphere. The increased warm-
Molecules such as water vapor and carbon dioxide ing would raise surface temperatures, driving more carbon
that transmit visible radiation but absorb IR radiation are dioxide and water vapor into the atmosphere, which would
called greenhouse molecules. (Methane, nitrous oxide, in turn increase the strength of the atmospheric greenhouse
and chlorofluorocarbons [CFCs] are other greenhouse mol- effect. The result would be even greater warming and still
ecules found in Earth’s atmosphere.) Thus, the presence larger amounts of atmospheric water vapor, ultimately cre-
of greenhouse molecules in a planet’s atmosphere will ating a surface pressure at least 300 times as great as it is
cause its surface temperature to rise. This rise in tempera- now2 and a surface temperature that might exceed 800 K.
ture continues until the surface becomes sufficiently hot— Long before reaching this stage, our planet would have
and therefore radiates enough energy—that the fraction of become devoid of all life.
infrared radiation leaking out through the atmosphere is In reality, the process is more complicated. Increased
just enough to balance the absorbed sunlight. Convection cloud cover caused by increased water in the atmosphere
also helps, by transporting thermal energy to the top of the might decrease the amount of sunlight reaching Earth’s sur-
atmosphere, where it can be more easily radiated to space. face to a point where the runaway effect would be turned
In short, the temperature rises until equilibrium between off. There are other complicating factors. Ocean currents are
absorbed sunlight and thermal energy radiated away by the important in transporting energy from one part of Earth to
planet is reached, just as our earlier discussion in Chapter another, but how they would be affected by increased warm-
4 said it must be. Even though the mechanisms are some- ing is something we really do not know. In fact, the process
what different, the conventional greenhouse effect and the is so complicated, wherein tiny changes lead to very large
atmospheric greenhouse effect produce the same net result: results, that we still cannot predict the long-term outcome
the local environment is heated by trapped solar radiation. of the small changes that humans are now making in the
AstroTour: Greenhouse Effect composition of Earth’s atmosphere. In a real sense, we are
Let’s look more closely at how the atmospheric green- experimenting with the atmosphere of Earth. We are asking
house effect operates on Mars, Earth, and Venus. We already the question (whether we know it or not), “What happens to
know that solar radiation can be trapped by greenhouse Earth’s climate if we steadily increase the number of green-
molecules, but what really matters is the actual number of house molecules in its atmosphere?” We do not yet know
greenhouse molecules in a the answer to this question, but we will eventually know
planet’s atmosphere, not the Without the by seeing the results. Whether we will be happy with the
fraction they represent. For greenhouse effect, results is another matter. We will explore this issue again
example, even though the Earth would freeze. in Section 8.3.
atmosphere of Mars is com- It is relatively easy to understand the present-day dif-
posed almost entirely of carbon dioxide—an effective green- ferences between Venus and Mars, but why does the atmo-
house molecule—its tenuous atmosphere contains relatively spheric greenhouse effect make the composition of Earth’s
few greenhouse molecules compared to the atmospheres of atmosphere so different from that of the other two? We
Venus or Earth. As a result, the atmospheric greenhouse may fi nd the answer in Earth’s special location in the Solar
effect on Mars raises the mean surface temperature by only System. Consider early Earth and early Venus, each having
about 5 kelvins (K). Earth’s more massive atmosphere is about the same mass, but with Venus orbiting somewhat
more efficient. Temperatures on Earth are 35 K warmer than closer to the Sun than Earth does. Volcanism must have
they would be in the absence of an atmospheric greenhouse poured out large amounts of carbon dioxide and water vapor
effect, produced mainly by water vapor and carbon dioxide. to form early secondary atmospheres on both planets. Most
Without this greenhouse warming, the mean global tem- of Earth’s water quickly rained out of the atmosphere to fill
perature of Earth would be well below the freezing point, vast ocean basins. But Venus was closer to the Sun, and its
leaving us with a world of frozen oceans and ice-covered surface temperatures were higher than Earth’s. Most of the
continents! rainwater on Venus immediately re-evaporated, much as it
Nowhere in the Solar System, however, is the atmo- does today in Earth’s desert regions. Venus was left with
spheric greenhouse effect more dramatic than on Venus. a planetwide surface containing very little liquid water,
Its massive atmosphere of carbon dioxide and sulfur com- but with an atmosphere filled with water vapor. The con-
pounds raises its surface temperature by more than 400 K, tinuing buildup of both water vapor and carbon dioxide in
to about 737 K. At such high temperatures, any remaining the Venus atmosphere then led to a runaway atmospheric
water and most carbon dioxide locked up in surface rocks greenhouse effect that drove up the surface temperature of
would long ago have been driven into the atmosphere, fur-
ther enhancing the atmospheric greenhouse effect. 2
Note that this surface pressure is even higher than that of present-day
The conditions existing on Venus today could be created Venus—a consequence of the enormous amount of available terrestrial
on an Earth-like planet by a runaway atmospheric green- water that would be released into Earth’s atmosphere.
the planet. The surface of the planet became so hot that no Venus? One possibility is that water molecules high in its
liquid water could survive there. atmosphere were broken apart into hydrogen and oxygen by
This early difference between a watery Earth and an arid solar ultraviolet radiation. Hydrogen atoms, being of very
Venus forever changed the ways that their atmospheres and low mass, were quickly lost to space. Oxygen, however,
surfaces would evolve. On would eventually have migrated downward to the planet’s
Earth, water erosion caused by Earth’s atmosphere surface, where it would have been removed from the atmo-
rain and rivers continually evolved very differently sphere by oxidizing surface minerals.
exposed fresh minerals, which from that of Venus.
then reacted chemically with
atmospheric carbon dioxide to form solid carbonates. This
reaction removed some of the atmospheric carbon dioxide, 8.3 Earth’s Atmosphere—
burying it within Earth’s crust as a component of a rock called
“limestone.” Later, the development of life in Earth’s oceans The One We Know Best
accelerated the removal of atmospheric carbon dioxide. Tiny
sea creatures built their protective shells of carbonates, and Now that we understand some of the overall processes that
as they died they built up massive beds of limestone on the have influenced the evolution of the terrestrial planet atmo-
ocean floors. As a result of water erosion and the chemistry spheres, we will look in depth at each of them. We begin
of life, all but a trace of Earth’s total inventory of carbon diox- with Earth’s atmosphere, not only because we know it best,
ide is now tied up in limestone beds. Earth’s particular loca- but also because it provides an introduction to atmospheric
tion in the Solar System seems to have spared it from the structure and weather phenomena that will help us better
runaway atmospheric greenhouse effect. But what if Earth understand the atmospheres of Venus, Mars, and Titan, and
had formed a bit closer to the Sun? Look now at Table 8.2. If even the atmospheres of the giant planets.
all the carbon dioxide now in limestone beds had not been In simplest terms, we can describe Earth’s atmosphere as
locked up by these reactions, Earth’s atmospheric composi- a blanket of gas that is several hundred kilometers deep and
tion would resemble that of Venus or Mars. has a total mass of approxi-
Differences in the amount of water on Venus, Earth, mately 5 × 1018 kg (about 5,000 Earth’s atmosphere
and Mars are not so well understood. Geological evidence trillion metric tons.) As enor- has a surface
tells us that liquid water was once plentiful on the surface mous as this amount may pressure equal
of Mars (see Chapter 7), and the Mars Odyssey spacecraft seem, it represents less than to a 10-meter
has found evidence that significant amounts of water still one-millionth of Earth’s total layer of water.
exist in the form of subsurface ice—far more than the atmo- mass. Yet the weight of Earth’s
spheric abundance indicated in Table 8.1. Earth’s liquid and atmosphere creates a force of approximately 100,000 newtons
solid water supply is even greater: about 1021 kilograms (kg), (N) acting on each square meter of our planet’s surface. We
or 0.03 percent of its total mass. More than 97 percent of express this amount of pressure as a unit called a bar (from
Earth’s water is in the oceans, which have an average depth the Greek baros, meaning “weight” or “heavy”). Earth’s aver-
of about 4 kilometers (km). Earth today has 100,000 times age atmospheric pressure at sea level is approximately 1 bar.
more water than Venus. What happened to all the water on (Meteorologists frequently quote atmospheric pressures in
millibars [mb], which are thousandths of a bar.) One bar of
pressure is equivalent to what we might experience under-
water at a depth of 10 meters, or 33 feet. Yet we seem to be
TA B L E 8 . 2 largely unaware of Earth’s atmospheric pressure. How can
this be? The reason is that the very same pressure exists both
within and outside of our bodies. The two precisely balance
Terrestrial Planet Atmospheres one another—another example of the concept of hydrostatic
If All Available Carbon Dioxide Were Included equilibrium that we introduced in the previous chapter.
As we learned in Chapter 7, hydrostatic equilibrium tells
Planet us that the pressure at any point within a planet must be
great enough to balance the weight of the overlying layers.
Venus Earth Mars The same principle holds true in a planetary atmosphere.
The atmospheric pressure on a planet’s surface must be great
Carbon dioxide (%) 96.5 98.0 96.0 enough to hold up the weight of the overlying atmosphere. Of
Nitrogen (%) 3.5 1.6 2.7
course, different forms of matter provide the pressure within
a planet’s interior and in its atmosphere. In the interior of
Oxygen (%) 0.0 0.4 0.1 a solid planet, solid materials exert pressure as they resist
being compressed. In a planetary atmosphere, the motions
All other constituents (%) 0.0 0.0 1.2
of gas molecules exert sufficient pressure to support the
What Is a Gas?
As we continue our journey of discovery through the universe, tant properties of a gas is how hard it pushes on its surround-
we will fi nd that the gaseous state is the most common form ings. Imagine a box containing gas, as shown in Figure 8.4a.
of matter. More than those of any other form of matter, it is the Molecules are constantly bouncing off the walls of the box, push-
properties of gases that we will turn to as we seek to under- ing outward. This outward push, measured in units of force per
stand the workings of planets, stars, and galaxies. square meter of the surface of the box, is referred to as pressure.
Matter is composed of atoms and molecules, and different (Anytime you think about the pressure of a gas, this is the men-
forms of matter result from dissimilarities in how those atoms tal picture that you should bring to mind: atoms and molecules
and molecules interact. In a solid, molecules are packed tightly, slamming against the walls of a box, pushing outward.)
held in place by adjacent molecules like bricks in a wall. In Armed with nothing more than this mental picture, we can
a liquid, molecules are free to move about and are constantly draw some interesting conclusions about the pressure of a gas.
jostling one another. You can picture molecules in a liquid as First, if we increase the number of molecules in the box, more
people in a crowded subway station. The crowd is free to flow, molecules will hit the walls of the box each second and the pres-
but the individuals that make up the crowd are still limited in sure will increase. Doubling the density of the gas but keeping
their movement by the people around them. The molecules in the temperature the same (Figure 8.4b) doubles its pressure.
a gas, on the other hand, go their own way, traveling relatively Increasing the temperature of the gas increases the aver-
long distances without interacting with other molecules. When age speed at which molecules move, which also increases the
you think of a gas, picture a swarm of tiny atoms and molecules pressure of the gas. There are two reasons for this effect. First,
flying about, each with its own speed and direction. if the molecules in the box are moving faster, they will hit the
As we have already seen in our discussion of escape veloc- walls more frequently (Figure 8.4c). More molecules hitting the
ity, the temperature of a gas is a measure of the average kinetic walls of the box each second means that the pressure is higher.
energy of the individual molecules as they fly about. Two Second, faster-moving molecules hit the wall harder, exerting
things determine a molecule’s kinetic energy: the mass of the more force. Together, these two effects mean that doubling the
molecule and the speed at which it is moving. The average temperature of a gas doubles the pressure of the gas.
speed of a molecule in a gas is inversely proportional to the Pressure is proportional to density, and also to temperature.
square root of its mass. Oxygen molecules, for example, are We can combine these two relationships to get
16 times more massive than hydrogen molecules. In a gas con-
taining both hydrogen and oxygen, the hydrogen molecules Pressure ∝ Density × Temperature.
___
will therefore be moving four times (√16 ) as fast as the oxy-
gen molecules. As we have seen, this difference explains why This relationship of density, temperature, and pressure is called
Earth can hold onto the oxygen in its atmosphere but loses the the ideal gas law. The empirically discovered ideal gas law has
hydrogen to space. a special place in the history of science. It provided strong sup-
These motions of gas molecules create the pressure needed port for the atomic theory of matter, in much the same way that
to hold up a planet’s atmosphere. As we study gases in different Kepler’s empirical laws of planetary orbits provided support for
astronomical settings, we will find that one of the most impor- Newton’s theories of motion and gravitation. The fact that labora-
atmosphere. (The relationship among density, temperature,

and pressure of a gas is discussed in Foundations 8.1.) Life Controls the Composition
The composition of Earth’s atmosphere is relatively con- of Earth’s Atmosphere
stant on a global scale, but temperatures can vary widely.
Two principal gases make up our atmosphere. About four- As Table 8.1 shows, the composition of Earth’s atmosphere
fifths is nitrogen and one-fifth is oxygen, although there are is very different from that of Venus and Mars. We have
many important minor constituents, such as water vapor and already discussed the differ-
carbon dioxide. The amounts of atmospheric water vapor ences in carbon dioxide con- Only Earth’s
and carbon dioxide are somewhat variable, depending on tent, but what truly sets Earth atmosphere contains
global location and season. Atmospheric temperatures near apart from all other known abundant oxygen.
Earth’s surface can range from as high as 60°C in the deserts planets is its oxygen. Earth’s
to as low as -90°C in the polar regions. The mean global atmosphere contains abundant amounts of oxygen, while
temperature is about 15°C. other planets’ atmospheres do not. Why should this be so?
Pressure Temperature (K) Pressure Temperature (K) Pressure Temperature (K)

50 300 50 300 50 300
0 100 0 600 0 100 0 600 0 100 0 600
(a) (b) (c)
Gases consist of atoms and

molecules moving about in
all directions.
The pressure exerted on the

walls of a box of gas comes
from the atoms and molecules
hitting the walls of the box. Doubling the density of the gas
(but keeping the same temperature)
doubles the number of atoms
hitting the walls each second,
thus doubling the pressure.
Doubling the temperature of molecules increases the

frequency of collisions with the box and also increases
the “push” of each collision, doubling the pressure.
FIGURE 8. 4 (a) The pressure of a gas comes from the motions of atoms and molecules. Making a gas denser
(b) or hotter (c)—that is, increasing the speed of atoms or molecules—increases the pressure of the gas.
tory gases come very close to obeying the ideal gas law provided the propellant gas is expanding. An air conditioner works by
compelling early evidence that atoms and molecules are real. alternately compressing a gas to make it hot, letting that gas
We need to know about one other property of gases as our cool, and then allowing the gas to expand and get really cold.
journey continues. If you compress a gas, you increase not Over and over again on our journey, our understanding of
only its density, but also its temperature. Conversely, when a gases will be our guide—equally as valid when applied to the
gas is allowed to expand, it cools off. There are many every- hot cores of stars as to a gentle breeze on a summer day. Every
day examples of this behavior. Pump up a bicycle tire, and the time we talk about an object made of gas, think back to the pic-
pump becomes hot because the air is being compressed. Hold ture of molecules bouncing around in a box, and remember the
down the nozzle on an aerosol can, and it feels cold because commonsense ideas that explain the behavior of the gas.
Oxygen, it turns out, is a highly reactive gas. It chemically releasing oxygen into Earth’s atmosphere as a waste product
combines with, or “oxidizes,” almost any material it touches. of their metabolism. At first, this biologically generated oxy-
Witness the rust (iron oxide) that forms on steel, for example. gen combined readily with
For a planet to retain significant amounts of this reactive gas exposed metals and minerals Life is responsible for
in its atmosphere, there would need to be a means for replac- in surface rocks and soils and the oxygen in Earth’s
ing what is lost through oxidation. Such a means exists on so was removed from the atmosphere.
Earth—we call it “plant life” (see Excursions 8.1). atmosphere as quickly as it
The oxygen concentration in Earth’s atmosphere has formed. In this way, emerging life dramatically changed the
changed over the history of the planet, as shown in Figure very composition and appearance of Earth’s surface—the
8.5. When Earth’s secondary atmosphere first appeared about first of many such widespread modifications imposed on
4 billion years ago, it was almost totally free of oxygen. About our planet by living organisms. Ultimately, the explosive
2.8 billion years ago, an ancestral form of cyanobacteria growth of plant life accelerated the production of oxygen,
(single-celled organisms that contain chlorophyll) began building up atmospheric concentrations that approached
On Atmospheres and Life

Take a deep breath. The oxygen you inhale sustains you. of life on Earth. Living organisms are the only known source
Breathe out. The carbon dioxide you exhale is food for plant of significant amounts of molecular oxygen or ozone in a plan-
life. If these two gases were to disappear suddenly, most life on etary atmosphere. Methane is yet another gas that would be
Earth would perish.3 As we learned earlier in the chapter, vol- unexpected in the atmosphere of an Earth-like planet because
canism is a major source of carbon dioxide, so this gas occurs it is readily destroyed by photodissociation. The presence of
quite naturally in our atmosphere. Molecular oxygen, however, methane means that a source is needed to keep up with pho-
is an entirely different matter. We would not expect to fi nd it in todissociative loss. As with oxygen, a likely source would be
the atmosphere of a planet like Earth. Oxygen is so chemically biological. Certain microorganisms produce methane as a meta-
reactive that it quickly destroys itself by combining with surface bolic by-product. Those bubbles you see rising to the surface of
minerals to form oxides. The reddish surface of Mars is coated a stagnant pond—sometimes called “swamp gas”—contain bio-
with oxidized iron-bearing minerals, and this is one reason logically produced methane. Yet, certain geological processes
the martian atmosphere is almost completely free of oxygen. can also produce methane, so its presence in the atmosphere of
What, then, about Earth? As we learned earlier, the metabolism a terrestrial-type exoplanet is a weaker argument for life than
of ancient, tiny organisms produced oxygen as a by-product, is the presence of oxygen and ozone.
much as present-day animal life produces carbon dioxide as a So there we have it. Find molecular oxygen, ozone, or meth-
respiratory by-product. Today, plants maintain oxygen levels ane in the atmosphere of an Earth-like planet and you have a
at about 20 percent of our atmospheric total. We would do well possible candidate for extraterrestrial life. Is this a realistic
to remember that without plant life, Earth’s oxygen would dis- expectation? Yes, and it could happen as soon as we develop
appear and all animal life would vanish with it. the instrumentation to detect Earth-like planets around other
You may have already jumped ahead to an obvious conclu- stars. Suppose a future orbiting observatory isolates the light
sion: one of the possible ways of detecting life on a terrestrial- from an Earth-like planet and feeds that light into its onboard
type extrasolar planet is by fi nding atmospheric gases that, in spectrograph. Suppose further that the spectra reveal substan-
the absence of life, should not be there. We have already iden- tial amounts of molecular oxygen in the planet’s atmosphere.
tified molecular oxygen and ozone as among the telltale gases The most likely conclusion would be that the planet harbors
life. Of course, the observation would not tell us what kind of
3
Certain microorganisms, called “anaerobes,” live in the total life, or even provide positive proof of its existence, but it would
absence of oxygen. In fact, most anaerobes are poisoned by even certainly mark that planet as an object of considerable interest
small amounts of oxygen. for further study.
Invasion of land First flowering plants Present today’s levels only about 250 million years ago. All true
100 plants, from tiny green algae to giant redwoods, use the
Percentage of present-day oxygen levels
80 First vertebrates Age of energy of sunlight to build carbon compounds out of carbon
reptiles
50 Arthropods dioxide and produce oxygen as a metabolic waste product
30 Only in the past 250 million in a process called “photosynthesis.” Earth’s atmospheric
Multicellular years have oxygen levels oxygen content is held in a delicate balance primarily by
10 organisms approached those of today.
plants. If plant life on our planet were to disappear, so, too,
5 First eukaryotes (organisms would nearly all of Earth’s atmospheric oxygen.
3 with compartmented cells) Several minor gases in Earth’s atmosphere also affect
2 our daily lives. Over the current range of terrestrial tem-
First oxygen release
1 by cyanobacteria peratures, water is a volatile substance, so its atmospheric
abundance varies from time to time and from place to
Rocks absorb
oxygen place. In warm, moist climates, water vapor may account
for as much as 3 percent of the total atmospheric composi-
First life tion. In cold, arid climates, it may be less than 0.1 percent.
0
4 3 2 1 0.5 0.25 0.1 The continuous process of condensation and evapora-
Billions of years ago tion of water involves the exchange of thermal and other
FIGURE 8. 5 The amount of oxygen in Earth’s atmosphere has forms of energy, making water vapor a major contributor
built up over time as a result of plant life on the planet. to Earth’s weather.
Carbon dioxide is another variable component of Earth’s Why should we be so concerned about the loss of a minor
atmosphere, and a complex pattern of “sources” (places constituent from so high in our atmosphere? Even in tiny
where it originates) and “sinks” (places where it goes) deter- amounts, ozone filters out harmful solar ultraviolet radia-
mines how much of it will be present at any one time. tion, preventing it from reaching Earth’s surface. Scientists
Plants consume carbon dioxide in great quantities as part have predicted that the continuing removal of ozone from
of their metabolic process. Coral reefs are colonies of tiny the high atmosphere could spell trouble for terrestrial life as
ocean organisms that build their protective shells with more and more UV radiation reaches the ground. Measured
carbonates produced from dissolved carbon dioxide. Fires, increases in the levels of UV radiation appear to be related
decaying vegetation, and human burning of fossil fuels to increases in skin cancer in humans, and we do not yet
all release carbon dioxide back into the atmosphere. This understand the effects it may have on other life-forms to
balance between sources and sinks can and does change which we are inexorably linked. By the late 1980s, the pro-
with time. Meteorological records show that the amount of duction of CFCs and other ozone-depleting chemicals was
carbon dioxide in our atmosphere has been increasing for being phased out by international agreements; and by 2007,
almost two centuries—since the Industrial Revolution. As satellite measurements showed some recovery of ozone in
noted earlier, this increase in carbon dioxide in turn has the polar latitudes. In 2008, however, the ozone hole deep-
had a direct impact on global temperature because carbon ened again. Full recovery is not expected until the late 21st
dioxide is also a powerful greenhouse gas. It should come century at the earliest.
as no surprise to learn that Earth’s mean global tempera- Although the ozone in our upper atmosphere protects
ture seems to be increasing with the growing abundance life, it may not be so beneficial when found elsewhere.
of this gas. Ozone in the lower atmosphere occurs primarily in urban
Another minor constituent in our atmosphere is “ozone” and suburban environments. It is a human-made pollutant
(O3). This important molecule is formed when ultraviolet and a health hazard, raising the risk of asthma and heart
light from the Sun breaks problems.
molecular oxygen (O2) into its High-altitude ozone
individual atoms (O) in a pro- protects life. Our
cess called photodissociation. continuing survival Earth’s Atmosphere Is
These atoms can then recom- depends on it.
bine with other oxygen mole-
Layered like an Onion
cules to form ozone (O2 + O → O3)—a process called Our atmosphere is made up of several distinct layers. Earth’s
recombination. Most of Earth’s natural ozone is concentrated lowermost atmospheric layer, the one in which we live and
in the upper atmosphere at altitudes between 20 and 50 km. breathe, is called the tropo-
There it acts as a very strong absorber of ultraviolet sunlight. sphere (Figure 8.6). It contains Pressure and
Without the ozone layer, this lethal radiation would reach 90 percent of Earth’s atmo- temperature decrease
all the way to Earth’s surface, making it completely uninhab- spheric mass and is the source with altitude in the
itable for nearly all forms of life. of all of our weather. At Earth’s troposphere.
In the 1980s, scientists began noticing that the measured surface, usually referred to as
amount of ozone in our upper atmosphere had been decreas- “sea level,” the troposphere has an average temperature of
ing since the 1970s, primarily over the polar latitudes in 15°C (288 K) and an average pressure of 1.013 bars. Within
both the Northern and Southern hemispheres. They referred the troposphere, atmospheric pressure, density, and tem-
to these depleted regions as “ozone holes,”34 although they are perature all decrease with increasing altitude. For example,
more like depressions than real “holes” in the ozone layer. at an altitude of 5.5 km, a few thousand feet below the sum-
Ozone depletion appears to be caused by a seasonal buildup mit of Denali in Alaska, the atmospheric pressure and den-
of atmospheric “halogens”—mostly chlorine, fluorine, and sity are only 50 percent of their sea-level values and the
bromine, such as those found in industrial refrigerants. Hal- average temperature has dropped to –20°C. Still higher, at
ogens readily diffuse upward into the stratosphere, where an altitude of 12 km, where commercial jets cruise, the tem-
they destroy ozone without themselves being consumed. perature is a frigid –60°C and the density and pressure are
Such agents are called catalysts—materials that participate less than one-fifth what they are at sea level. Mountain
in and accelerate chemical reactions but are not themselves climbers and astronomers are very much aware of this
modified in the process. Because they are not used up, halo- behavior of Earth’s troposphere. At the Mauna Kea Obser-
gens may remain in Earth’s upper atmosphere for decades vatories in Hawaii, even the most dedicated astronomers,
or even centuries. surrounded by thin air and subfreezing temperatures, have
been known to gaze longingly at the sunny beaches some 4
4
3
A common misconception confuses the greenhouse effect with the km below, where it is a pleasant 30°C warmer.
so-called ozone hole. Both are caused by the buildup of certain atmo- Why, though, does the atmosphere get colder as we
spheric gases and both are a cause for concern, but the individual climb to higher elevations? From what we have already
causes and effects are very different. learned, we might guess correctly that it is warmer near
(a) EARTH
150 36
Record for balloon flight
34 34.7 km
32
30
Thermosphere 28
Blackbird reconnaissance jet
26 26 km
100 24 Stratosphere
22
Altitude (km)
Altitude (km)
20
Military jet
18
18 km
Mesosphere
16
Commercial jet
14 12–15 km
50 12 Tropopause
10 Mt. Everest Cirrus clouds
8,840 m
Stratosphere 8
6 Denali
Tropopause 6,189 m
4 Mauna Kea Troposphere
Troposphere 2 4,205 m
0 0
100 200 300 400 500 600 0 0.2 0.4 0.6 0.8 1.0
Temperature (K) Pressure (bars)
(b)
VENUS MARS
100
Hazy clouds Cirrus clouds

Altitude (km)
50
Wispy clouds
0
100 200 300 400 500 600 700 120 140 160 180 200 220
FIGURE 8.6 (a) Temperature and pressure are plotted for Earth’s atmospheric layers as a
function of height. (b) Atmospheric temperatures for Venus and Mars are shown for comparison.
When the air temperature decreases to the point at which

3 Warm air heats the
upper atmosphere
the air can no longer hold all its water vapor, it becomes
and cools off. Cold saturated. Water begins to condense out in the form of tiny
droplets, which in large numbers become visible to us as
clouds. When these droplets
coalesce to form large drops, Most atmospheric
convective updrafts can no water stays close to
longer support them, and they Earth’s surface.
2 Warm air expands, 4 Denser, cool air sinks, fall as rain. For this reason
becomes buoyant, displacing less dense most of the water vapor in Earth’s atmosphere stays within
and rises. warm air. 2 km of the surface. At an altitude of 4 km, the Mauna Kea
Observatories are higher than approximately one-third of
Earth’s atmosphere, but they lie above nine-tenths of the
atmospheric water vapor. This is important for astronomers
who observe the heavens in the infrared region of the spec-
trum, because water vapor strongly absorbs IR light (see
1 Air is heated Figure 5.18).
by ground. Warm
The top of the troposphere, called the tropopause,
is defi ned as the height at which temperature no longer
decreases with increasing altitude (see Figure 8.6a). This
FIGURE 8.7 Atmospheric convection carries thermal energy change in atmospheric behavior is caused by heating from
from the Sun-heated surface of Earth upward through the absorbed sunlight within the atmospheric layers that lie
atmosphere. above the tropopause. And because temperature no longer
decreases with altitude above the tropopause, atmospheric
convection also dies out. The tropopause varies between 10
Earth’s surface because the air is in contact with the sun- and 15 km above sea level, depending on latitude, and is
light-heated ground, which also warms the air by its infra- highest at the equator.
red radiation. It also makes sense that the atmosphere Above the tropopause and extending upward to an alti-
would be cooler at very high altitudes because there the tude of 50 km above sea level is the stratosphere. This is a
atmosphere can freely radiate region in which little convec-
its thermal energy into space. Convection carries tion takes place, because the The stratosphere,
In fact, it would get colder thermal energy temperature-altitude relation- mesosphere,
with increasing altitude even upward through the ship actually reverses at the and thermosphere
faster if not for convection. troposphere. tropopause and the tempera- lie above the
We fi rst encountered convec- ture begins to increase with troposphere.
tion in our Chapter 7 discussion of the motions of material altitude. The reason for this
in Earth’s mantle that drive plate tectonics. Figure 8.7 illus- reversal is the presence of ozone, which warms the strato-
trates how convection carries thermal energy upward sphere by absorbing sunlight.
through Earth’s atmosphere. At a given pressure, cold air The region above the stratosphere is called the “meso-
is more dense than warm air. So when cold air encounters sphere.” It extends from 50 km to an altitude of about 90 km.
warm air, the denser cold air slips under the less dense In the mesosphere there is no ozone to absorb sunlight, so
warm air, pushing the warm air upward. (Rather than temperatures once again decrease with altitude. The base
“warm air rises,” we should perhaps say “cold air sinks.”) of the stratosphere and the upper boundary of the meso-
This convection sets up air circulation between the lower sphere are two of the coldest levels in Earth’s atmosphere
and upper levels of the atmosphere, and such circulation (see Figure 8.6a).
tends to diminish the extremes caused by heating at the At altitudes above 90 km, solar ultraviolet radiation and
bottom and cooling at the top. high-energy particles from the solar wind “ionize” (strip
Convection also affects the vertical distribution of atmo- electrons from) atmospheric atoms and molecules, causing
spheric water vapor. The ability of air to hold water in the the temperature once again to increase with altitude. This
form of vapor depends very strongly on the air temperature. region is called the “thermosphere,” and it is the hottest
The warmer the air, the more water vapor it can hold. We part of the atmosphere. The temperature can reach 1,000
refer to the amount of water vapor in the air relative to what K near the top of the thermosphere, at an altitude of 600
the air could hold at a particular temperature as the rela- km. The gases within and beyond the thermosphere are
tive humidity. Air that is saturated with water vapor has a ionized by ultraviolet photons and high-energy particles
relative humidity of 100 percent. As air is convected from the Sun to form a plasma. (A plasma is any gas that
upward, it cools, limiting its capacity to hold water vapor. is made up largely of electrically charged particles rather
(a)–Uniform magnetic field (c)
e–
e–
Bow shock Particles are trapped in

the “magnetic bottle” of
Earth’s magnetic field.
Solar wind Direction of

magnetic field
Charged particles are free to move along
a magnetic field but not across it. Trapped
radiation belts
(b)–Magnetic bottle
Magnetic
field direction
e–
Solar wind material
reaches Earth
Trapped
plasma
Charged particles can be trapped in a

“bottle” between regions of pinched field.
FIGURE 8.8 (a) The motion of charged particles, in this case electrons, in a uniform magnetic
field. (b) When the field is pinched, charged particles can be trapped in a “magnetic bottle.” (c) Earth’s
magnetic field acts like a bundle of magnetic bottles, trapping particles in Earth’s magnetosphere.
than only neutral atoms and molecules.) This region of ion- a volume over 1,000 times the volume of the planet itself.
ized atmosphere is called the ionosphere. The ionosphere To appreciate Earth’s magnetosphere, we need to begin by
is important to us in part because it reflects certain fre- looking more carefully at magnetic fields and the force they
quencies of radio waves back to the ground. For example, apply to charged particles. Magnetic fields have no effect on
the frequencies used by AM radio bounce back and forth charged particles unless the particles are moving. Charged
between the ionosphere and the surface, enabling radio particles are free to move along the direction of the magnetic
receivers to pick up stations at great distances from the field, but if they try to move across the direction of the field
transmitters. Amateur radio operators are able to commu- they experience a force that is perpendicular both to their
nicate with each other around the world by bouncing their motion and to the magnetic-field direction. This force causes
signals off the ionosphere. them to loop around the direction of the magnetic field, as
Earth and its atmosphere are surrounded by a large illustrated in Figure 8.8a. It is almost as if charged particles
region filled with electrons, protons, and other charged par- were beads on a string, free to slide along the direction of
ticles from the Sun that have been captured by Earth’s mag- the magnetic field but unable to cross it.
netic field. This region, called Earth’s magnetosphere, has The picture of how charged particles move in a mag-
a radius approximately 10 times the radius of Earth, filling netic field gets even more interesting if the field is pinched
together at some point. As particles move into the pinch, energetic charged particles, called radiation belts, can be
they feel a magnetic force that (if conditions are right) very damaging to both electronic equipment and astronauts.
pushes them back along the direction from which they Yet we need not leave the surface of the planet to witness
came. If charged particles are located in a region in which beautiful and dramatic effects of the magnetosphere. Dis-
the field is pinched on both ends, as shown in Figure 8.8b, turbances in Earth’s magnetosphere can lead to changes in
then they may bounce back and forth many times. This Earth’s magnetic field that are large enough to trip power
magnetic-field configuration is called a “magnetic bottle.” grids, causing blackouts, and to wreak havoc with com-
Earth’s magnetic field is pinched together at the two mag- munications. Earth’s magnetic field also funnels energetic
netic poles and spreads out around the planet. This config- charged particles down into the ionosphere in two rings
uration is like taking many magnetic bottles and bending located around the magnetic poles. These charged particles
them over, attaching them to Earth at either end. (mostly electrons) collide with atoms such as oxygen, nitro-
Earth and its magnetic field are immersed in the solar gen, and hydrogen in the upper atmosphere, causing them
wind, a constant stream of charged particles from the Sun. to glow like the gas in a neon sign. These glowing rings,
When these particles first called auroras, can be seen from space (Figure 8.9a). When
encounter Earth’s magnetic Earth’s magnetic viewed from the ground (Figure 8.9b), they appear as eerie,
field, the smooth flow is inter- field traps charged shifting curtains of multicolored light. People living far
rupted and they drop sud- particles from from the equator are often treated to spectacular displays
denly from supersonic to the Sun. of the aurora borealis (“northern lights”) in the Northern
subsonic speed at a point Hemisphere or the aurora australis in the Southern Hemi-
called the bow shock. As they stream by, they are diverted sphere. Auroras have also been seen on Venus, Mars, the
by Earth’s magnetic field as a river is diverted around a giant planets, and some moons.
boulder. As they flow past, some of these charged particles Although our discussion has concentrated on the atmo-
become trapped by Earth’s magnetic field, where they bounce sphere of our own planet, it is important to know that the
back and forth between Earth’s magnetic poles as illustrated structure we have described here is not limited to Earth’s
in Figure 8.8c. atmosphere. The major vertical structural components—
An understanding of Earth’s magnetosphere is of great troposphere, tropopause, stratosphere, and ionosphere—
practical importance for space travel. Regions in the mag- also exist in the atmospheres of Venus and Mars, as well as
netosphere that contain especially strong concentrations of in the atmospheres of Titan and the giant planets. And, as
we will see in the following chapter, the magnetospheres
of the giant planets are among the largest structures in the
Solar System.
FIGURE 8.9 Auroras result when particles trapped in Earth’s
magnetosphere collide with molecules in the upper atmosphere.
(a) An auroral ring around Earth’s north magnetic pole, as seen
from space. (b) Aurora borealis—the “northern lights”—viewed Why the Winds Blow
from the ground.
Winds are the natural movement of air, both locally and
on a global scale. Variation in solar heating is the chief
reason for differences in the
ground-level temperature of Nonuniform solar
Earth’s atmosphere from place heating creates
to place at similar altitudes our weather.
and throughout the year. It is
usually warmer in the daytime than at night, warmer in
the summer than in winter, and warmer at the equator than
in the polar regions. Large bodies of water, such as oceans,
also affect atmospheric surface temperatures. As we dis-
(a) covered in Foundations 8.1, heating a gas increases its pres-
sure, which in turn causes it to push into its surroundings.
The horizontal component of these pressure differences is
the reason we have winds, and the strength of the winds
is closely related to the magnitude of the difference in tem-
perature and pressure from place to place.
Think about what happens as air in Earth’s equato-
rial regions, heated by the warm surface, begins to rise
because of convection. The warmed surface air displaces
the air above it, which then has no place to go but toward
(b)
(a) (b)
N N
Polar easterlies
If Earth did not rotate, Westerlies

global Hadley circulation The Coriolis effect breaks up
would be as shown. Hadley circulation and diverts it
into predominant east-west winds.
Trade winds
Equator Equator
Trade winds
Westerlies
Polar easterlies
S S
FIGURE 8.10 (a) The classic Hadley circulation. (b) Hadley flow often breaks up into smaller
circulation cells. The poleward-equatorward flow is diverted into zonal flow by the Coriolis effect.
the poles. Cooling and becoming denser as it moves pole- will be. We will encounter very strong zonal winds when
ward, the displaced air now descends in the polar regions. we discuss the atmospheres of the giant planets in Chapter
There it displaces the surface polar air, which is forced 9. Zonal winds are often confi ned to relatively narrow
back toward the equator, completing the circulation. As a bands of latitude. Between the equator and the poles in
result, the equatorial regions remain cooler and the polar most planetary atmospheres,
regions remain warmer than they otherwise would be. Such the zonal winds alternate Nonuniform heating,
planetwide flow of air between equator and poles is called between “easterly” (those together with the
Hadley circulation (Figure 8.10a). Global Hadley circula- blowing from the east and Coriolis effect, causes
tion, it turns out, seldom occurs in planetary atmospheres toward the west) and “west- east-west zonal winds.
because other factors break up the planetwide flow into a erly” (those blowing from the
series of smaller “Hadley cells.” Planet rotation is a major west and toward the east). Confusing? Very! This unfortu-
factor here. Most planets and their atmospheres are rotating nate terminology is a historical carryover from early ter-
rapidly, and the effects produced by this rotation strongly restrial meteorology, in which winds are labeled not by the
interfere with Hadley circulation by redirecting the hori- direction toward which they are blowing but by the direc-
zontal flow (Figure 8.10b). tion from which they come.
The effect of Earth’s rotation on winds—and on the In Earth’s atmosphere, several bands of alternating zonal
motion of any object—is called the Coriolis effect (see Foun- winds lie between the equator and the poles of both hemi-
dations 8.2). On a rapidly rotating planet, air is not free to spheres. This zonal pattern is called Earth’s global circu-
flow in just any direction. When a volume of air starts to lation because its extent is planetwide. The best-known
move directly toward or away from the poles, the Coriolis zonal currents are the subtropical trade winds—more or
effect diverts it into relative motion that is more or less less easterly winds that once carried sailing ships from
parallel to the planet’s equator. This change in motion cre- Europe westward to the Americas—and the midlatitude
ates winds that blow predominantly in an east-west direc- prevailing westerlies that carried them home again (see
tion (see Figure 8.10b). Meteorologists call these zonal Figure 8.10b).
winds. In general, the more rapid a planet’s rotation is, the Embedded within Earth’s global circulation pattern
stronger the Coriolis effect and the stronger its zonal winds are systems of winds associated with large high- and
F O U N D AT I O N S 8 . 2
The Coriolis Effect

Earth’s rotation inf luences
(a) Frame of reference: Viewer on the street
things as diverse as the motion
Slower car of weather patterns on Earth
and how an artillery gunner
must aim at a distant target.
1 A ball is thrown
directly at the
Any object sitting on the sur-
slower car from 3 …resulting in a total motion face of Earth follows a circle
the faster car. relative to the ground. each day as Earth rotates on its
axis. This circle is larger for
2 The ball shares the forward motion
of the car from which it is thrown… objects near Earth’s equator
and smaller for objects closer
to one of Earth’s poles. But
Faster car
because Earth is a solid body,
all objects must complete their
circular motion in exactly one
day. Because an object closer to the equator has farther to go
(b) Frame of reference: Viewer in faster car each day than an object nearer a pole, the object nearer the
Slower car
In the frame of reference of equator must be moving faster than the object at a higher lati-
the faster car, the ball misses tude. If an object starts out at one latitude and then moves to
because the slower car is another, its apparent motion over the surface of Earth is influ-
moving backward.
enced by this difference in speed.
Now recall our discussion of frame of reference in Chap-
ter 2 to understand this apparent motion. Imagine that two
cars are driving down the road at different speeds, as shown
in Figure 8.11. Ignoring for the moment any real-world com-
plications like wind resistance, if you were to throw a ball
from the faster-moving car directly at the slower-moving car
as the two cars passed, you would miss. The ball shares the
forward motion of the faster car, so the ball outruns the for-
Faster car
ward motion of the slower car. From your perspective in the
faster car (Figure 8.11b), the slower car lagged behind the ball.
(c) Frame of reference: Viewer in slower car
From the slower car’s perspective (Figure 8.11c), your car and
Slower car In the frame of reference
of the slower car, the ball the ball sped on ahead.
misses because the ball Now do the same experiment, but instead of two cars
and the faster car are
think about two locations at different latitudes. Suppose you
moving forward.
fi re a cannon directly north from a point in the Northern
Hemisphere, as shown in Figure 8.12b. Because the cannon
is located nearer to the equator than its target is, the cannon
itself is moving toward the east faster than its target. Even
though the cannonball is fi red toward the north, it shares the
eastward velocity of the cannon itself. This means that the can-
nonball is also moving toward the east faster than its target!
Recall how the ball thrown from the faster car outpaced the
Faster car slower-moving car in Figure 8.11. Similarly, as the cannonball
fl ies north, it fi nds itself moving toward the east faster than
the ground underneath it is. To an observer on the ground,
FIGURE 8.11 The motion of an object depends on the the cannonball appears to curve toward the east as it outruns
frame of reference of the observer.
the eastward motion of the ground it is crossing. The farther
north the cannonball flies, the greater is the difference between
Continued on next page

F O U N D AT I O N S 8 . 2
FIGURE 8.12 The Coriolis

(a)
effect causes objects to appear
The surface of Earth moves to be deflected as they move
faster at the equator than at
across the surface of Earth.
higher latitudes.
290 km/h
80°
835 km/h 60°
1,280 km/h
40° its eastward velocity and the
eastward velocity of the ground.
1,573 km/h
20° Thus, the cannonball follows a
path that appears to curve more
1,674 km/h
0° and more to the east the farther
1,573 km/h
north it goes. If you are located
20°
in the Northern Hemisphere and
1,280 km/h fi re a cannonball south toward
The ground near the equator is 40°
like the faster car in Figure 8.11. 835 km/h
the equator (Figure 8.12c), the
60°
The ground at higher latitudes 290 km/h opposite effect will occur. Now
is like the slower car. 80°
the cannon is moving toward
(b) (c) the east more slowly than its
80° 290 km/h 80° target. As the cannonball fl ies
Slower car 60° 60°
835 km/h toward the south, its eastward
40° 1,280 km/h
40° motion lags behind that of the
ground underneath it, and the
20° 20°
1,573 km/h cannonball appears to curve
toward the west.
0° 0°
1,674 km/h This effect of Earth’s rota-
Faster car
Ground speed 1,573 km/h tion is called the Coriolis effect.
20° 20°
In the Northern Hemisphere the
40°
1,280 km/h
40° Coriolis effect causes a cannon-
835 km/h ball fi red north to drift to the
60° 60°
80°
290 km/h
80° east as seen from the surface of
Earth. In other words, the can-
A cannonball fired away from A cannonball fired toward the nonball appears to curve to the
the equator outruns the ground equator lags behind the eastward right. A cannonball fi red south
it flies over, so its ground track motion of the ground, so its
curves to the east. ground track curves to the west.
appears to curve to the west,
low-pressure regions. A combination of a low-pressure regions rotate in a direction opposite to that of low-pressure
region and the Coriolis effect produces a circulating pat- regions. These high-pressure circulating systems experi-
tern called cyclonic motion. Cyclonic motion is associated ence anticyclonic motion and are generally associated
with stormy weather, including hurricanes.54 Similarly, with fair weather.
high-pressure systems are localized regions where the
air pressure is higher than average. We think of these
regions of greater-than-average air concentration as “moun-
tains” of air. Owing to the Coriolis effect, high-pressure
When Convection Runs Amok
It takes the absorption of thermal energy to turn liquid
water into vapor. Water in Earth’s oceans, lakes, and rivers
5
4
Cyclones are regions characterized by rising moist air and are there- is evaporated by thermal energy acquired from the absorp-
fore associated with stormy weather. Conversely, anticyclones are vast tion of sunlight. The water vapor then carries this thermal
regions of sinking dry air and thus tend to produce fair weather. energy along with it as it circulates throughout the atmo-
weather systems. As air is pushed from regions of higher pres-

sure toward regions of lower pressure, these motions are influ-
enced by the Coriolis effect. Think about a low-pressure region
in the Northern Hemisphere. When air is pushed toward this
L region of low pressure from the south, the Coriolis effect deflects
Air moving into a low-pressure region this flow of air toward the east. Similarly, air moving toward
from the north or south is diverted by
the Coriolis effect, resulting in circulation
the region of low pressure from the north is deflected to the
around the low-pressure region. west by the Coriolis effect. The net effect is that as air moves
toward a region of low pressure in the Northern Hemisphere,
the Coriolis effect deflects it into a counterclockwise circulation
(Figure 8.13). (Think about this carefully. The Coriolis effect
As a result, air moves in L deflects objects toward the right as you face north in the North-
a counterclockwise Deflection
direction around a low- to right
ern Hemisphere, causing weather patterns that rotate toward
pressure region in the the left.) The next time you see a television weather map with a
Northern Hemisphere… low-pressure region, look at the direction of the winds around
Deflection the region and you will see this counterclockwise flow. The
to left most spectacular example of this cyclonic motion is the swirl
L of wind and clouds around the deep low-pressure area at the
…and clockwise in the
Southern Hemisphere.
eye of a hurricane or typhoon. As the air moves in closer and
closer to the central region of low pressure, it rotates faster and
faster, giving rise to the winds of hundreds of kilometers per
hour that make hurricanes so destructive.
FIGURE 8.13 As a result of the Coriolis effect, air circulates The Coriolis effect is only one example of the type of effect
around regions of low pressure on the rotating Earth. associated with relative motions. However, the Coriolis effect
has nothing to do with the direction that water swirls in a toi-
let bowl, as many people imagine. The difference in the speed
which also gives it the appearance of curving to the right. In of Earth’s motion between the two sides of a toilet bowl is not
the Northern Hemisphere the Coriolis effect seems to deflect enough to matter much. Other effects, such as the direction in
things to the right. If you think through this example for the which the water flows into the bowl, are much more important.
Southern Hemisphere, you will see that south of the equator the However, the Coriolis effect is enough to deflect a fly ball hit
Coriolis effect seems to deflect things to the left. In between, north or south into deep left field in a stadium in the north-
at the equator itself, the Coriolis effect vanishes. ern United States by about a half a centimeter. At one time or
These differences in the Coriolis effect between the North- another the Coriolis effect has probably determined the out-
ern and Southern hemispheres are seen in the rotation of come of a ball game.
sphere. When the water vapor recondenses, it gives up its tiny water droplets that we call cumulus clouds. Cooling
thermal energy to its surroundings. This is the process that then causes the water vapor in the moist air to condense as
powers rainstorms, thunderstorms, hurricanes, and a host rain. As it condenses, the water vapor gives up its thermal
of other dramatic weather phenomena. energy to the surrounding air, warming it and thus increas-
ing the strength of the convection. For the most part this
Rainstorms and Thunderstorms A rainstorm begins is a gentle process. Water, falling back to the surface as
when Earth’s surface, heated by the Sun, warms moist air rain, eventually returns to the lakes and oceans, wearing
close to the ground (see Figure 8.15). The moist air is con- down mountains, eroding the soil, and nourishing life as
vected upward, cooling as it gains altitude. Those puffy it flows. From the oceans to the air and back again—this
white clouds that you see on a summer day are created is the “water cycle,” or hydrological cycle. (In Chapter 11
when this warm, moist air is convected upward to cooler we will see that methane plays the same role on Titan that
atmospheric levels a few kilometers above the surface. Here water plays on Earth in what some planetary scientists call
the moisture condenses out of the air to form the myriad a “methanological cycle.”)
FIGURE 8.14
Thunderstorms are powered
by convection and by thermal
energy released as water vapor
condenses to form droplets. The
“anvil” top of a thunderhead is
caused by stratospheric winds
shearing the top of the convective
system.
Convection can also be violent. With strong solar heat- For every parcel of air that rises within a thunderstorm,
ing and an adequate supply of moist air, this self-feeding another must come back down (Figure 8.15). Downdrafts,
process can grow within minutes to become a violent thun- including the rain and hail that descend with them, produce
derstorm. Summer is the peak time for thunderstorms and ferocious winds that can exceed 100 kilometers per hour
lightning. The process that (km/h) in the immediate vicinity of the storm. The winds and
creates thunderstorms is the Runaway atmospheric turbulence associated with thunderstorms become so great
same as that which forms convection can lead to that even commercial jet aircraft prefer to keep their distance.
cumulus clouds, but the violent weather. At its base, a thunderstorm may be several kilometers across
amount of energy involved is and travel for tens of kilometers across the landscape. Thun-
far greater. Thunderstorms tend to form in the afternoon derstorms cause billions of dollars in crop damage every year
when solar heating of the ground reaches its maximum. and are responsible for hundreds of deaths.
They begin as familiar cumulus clouds; but if the supply of
warm, moist air at the surface is great enough, the clouds Lightning Lightning is essentially a gigantic electric spark
will continue to grow vertically, and we have a case of con- that results from billions of volts of potential difference—a
vection run amok. As more and more moist air rises and huge example of the “static electricity” you sometimes gen-
the moisture condenses within the cloud, the heat released erate when walking on a carpet—and is usually associated
by condensation continues to warm the surrounding air, with thunderstorms and rain. However, lightning can also
forcing convection ever higher. be created by snow, sand and dust storms, volcanic erup-
We can easily recognize these cumulonimbus clouds— tions, earthquakes, and nuclear explosions. Ice formation in
known popularly as “thunderheads”—by their flat, anvil- cumulonimbus clouds is a key factor for starting the “elec-
shaped tops, as seen in Figure 8.14. This upper surface tric generator” that produces most lightning. Falling small
visibly marks the tropopause, the level in the atmosphere ice pellets in cumulonimbus clouds become negatively
where convection fi nally ceases (see Figure 8.6a). In the charged by friction as they move through the surrounding
midlatitudes of the continental United States, the tropo- air, while small supercooled cloud droplets that strike them
pause occurs at an altitude of about 12 km above sea level. bounce off the ice pellets and become positively charged.
Convection in the more violent storms can be so strong that The supercooled cloud droplets rise on updrafts to the top
the tops of the thunderheads punch right through the tropo- of the storm cloud while the ice pellets fall and melt in the
pause, carrying cloud-forming moisture up into the strato- lower regions of the cloud or, as often is the case, fall all
sphere to heights of 20 km or more. The anvil shape occurs the way to the ground.
when strong stratospheric winds pull ice crystals from the As a result, a difference in electric potential is created
top of the cloud and spread them out horizontally. between the top and bottom of the cloud that can exceed a
14 Stratospheric winds
Ice crystals
12
Snow and
ice crystals
10
Kilometers
Hail
and rain
6
FIGURE 8.16 Lightning bolts carry enormous electric
4 currents, killing or injuring hundreds of people each year in the

United States alone.
2
The thermal energy causing this rapid expansion of the
Downdrafts
Updrafts of with heavy air is converted into sonic energy, resulting in the familiar
warm moist air rain and hail boom of thunder.
0 Lightning also creates powerful electromagnetic waves,
FIGURE 8.15 Convection within a thunderstorm. some of which can be heard on AM radio receivers as “light-
ning static.” When the Pioneer
Venus spacecraft was orbiting Lightning has been
billion volts. At some point the potential difference between our sister planet during the detected in the
parts of the cloud becomes so great that the electrical resis- 1980s, its radio receiver picked atmospheres of
tance of the air breaks down56 (the molecules become ion- up many bursts of lightning Venus and all four
ized) and a series of lightning bolts flashes between the static—so many that Venus giant planets.
positively and negatively charged regions of the cloud. This appears to have more light-
is called “cloud-to-cloud lightning.” About a third of the ning activity than our own planet. On Venus, as on Earth,
time, lightning travels between the cloud and the ground. lightning is created in the clouds; but Venus’s clouds are so
It is here that lightning can be so dangerous. In the United high—typically 55 km above the surface of the planet—that
States lightning kills an average of 100 people every year the lightning bolts never hit the ground. All of this light-
and injures more than 300. Worldwide, lightning strikes ning is of the cloud-to-cloud type. The Pioneer Venus results
occur about 100 times every second. supported observations made by the Soviet lander Venera
Lightning bolts are typically 3–5 km long and can carry 9, whose optical spectrometer had detected flashes of light-
tens of thousands of amperes of current at speeds of 200 ning on Venus’s dark side 5 years earlier. As we will learn
kilometers per second (km/s) (Figure 8.16). On average, the in the next chapter, lightning has also been detected in the
energy of a single lightning bolt could keep a 100-watt (W) atmospheres of Jupiter, Saturn, Uranus, and Neptune.
lightbulb burning for several months. Although the bril-
liance of lightning bolts makes them appear huge, they are Hurricanes, Tornadoes, and Dust Devils Hurricanes
really no larger in diameter than a quarter or a half-dollar. are examples of Coriolis forces acting on air rushing into
As it travels through the air, lightning heats the air to tem- regions of low atmospheric
peratures as high as the surface of the Sun, causing the pressure, creating a huge Coriolis forces can
volume of air along the path of the bolt to expand rapidly. vortex (see Chapter 9.) But create powerful, and
these most powerful of storms often deadly,
56
Some atmospheric physicists now believe that all lightning is initi- are much more complicated atmospheric vortices.
ated by secondary cosmic-ray particles (see Chapter 20), which trig- than the simple systems
ger the electrical breakdown (ionization) of the air along the path of described in the preceding discussion. Hurricanes are huge
the lightning bolt. heat engines. They derive their enormous energy from a very
common physical phenomenon: the heat of vaporization of the heat of vaporization of water to feed it, the hurricane
water. The conditions for formation must be just right—warm eventually fades away.
tropical seawater, light winds, and a region of low pressure With what you now know about the Coriolis effect, you
in which air spirals inward. As warm seawater evaporates, can show that hurricanes in the Southern Hemisphere rotate
moisture-laden air rises and releases its heat of vaporization in the opposite direction from hurricanes in the Northern
as it condenses at cooler levels. (Remember, this is the same Hemisphere. Instead of curving to the right, air moving into
process that leads to cumulonimbus thunderstorms.) When a region of low pressure curves to the left, causing a clock-
the supply of warm seawater is sufficient, a complex of thun- wise rotation around the low-pressure region. In crossing
derstorms develops. Then, if the winds aloft are light, the from the Northern Hemisphere to the Southern Hemisphere,
complex remains intact and grows in size and strength. Con- the Coriolis effect disappears at the equator. The difference
vection ceases at the tropopause, located about 15 km above in the direction of rotation between the hemispheres and
sea level at tropical latitudes; but the number of individual the weakness of the Coriolis effect near the equator mean
storm cells in the complex continues to increase. The stage that a northern hurricane would literally collapse if it tried
is now set for the birth of a hurricane. to cross into the Southern Hemisphere. (Imagine throwing
As surface winds driven by the Coriolis effect rush your car into reverse while traveling down the road at 150
inward to replace the air rising upward in the cumulo- km/h.) This is why countries around Earth’s equator do not
nimbus complex, the hurricane grows in size and strength. experience the ravages of hurricanes.
Sustained winds near the center of the storm can reach Tornadoes generally last only a few dozen minutes, but
speeds greater than 300 km/h, causing widespread dam- because their energy is so concentrated they are extremely
age and fatalities when a hurricane moves ashore. In 1900, violent and among the most dangerous and destructive of
a hurricane in Galveston, Texas, took 8,000 human lives— storms. Like their larger hurricane cousins, tornadoes are
more than any other natural disaster in US history. Hur- also vortices. They are usually too small to be governed by
ricane Andrew caused $12 billion in damage when it hit the Coriolis effect; but the general atmospheric circulation
Florida in 1992. In 2005, Hurricane Katrina (pictured in in their vicinity causes many tornadoes, like hurricanes, to
Figure 8.17) devastated much of the Gulf Coast and the city rotate counterclockwise in the Northern Hemisphere and
of New Orleans, taking more than 1,000 lives and costing clockwise in the Southern Hemisphere.
more than $200 billion in damage. The eye of a hurricane, Tornadoes tend to form in the vicinity of hurricanes
however, is relatively calm and free of clouds. The eye is and violent thunderstorms where strong, thermally gener-
typically 40–50 km wide, whereas the hurricane itself may ated updrafts are present. When surrounding air rushes in
extend outward for more than 600 km. A hurricane may to replace the rising air in the updraft, it may start spin-
last weeks and travel thousands of kilometers, as long as it ning, creating a vortex. As the column of rising air extends
remains over open ocean. But what happens if the hurri- upward, its diameter shrinks. And because the circulating
cane moves ashore? Over land it loses its principal supply of air must obey the law of conservation of angular momen-
warm, moist air and therefore its source of energy. Without tum, the vortex spins ever faster as it continues to shrink.
(Recall the skater in Chapter 6.) Wind speeds in the most
severe tornadoes have been estimated to reach 800 km/h!
The base of an average tornado is about 400 meters in diam-
FIGURE 8.17 A satellite view of Hurricane Katrina in the Gulf eter. Atmospheric pressure at the base can be extremely
of Mexico approaching New Orleans and the Gulf Coast in 2005. low—more than 200 mb lower than the surrounding air.
This low pressure causes the tornado to act like a gigantic
vacuum cleaner, picking up dust, cars, and even buildings.
Over the past 50 years, tornado-related deaths in the United
States have averaged about 90 per year. The debris swept up
by tornadoes makes them highly visible and reveals their
characteristic funnel shape, as in Figure 8.18. When a tor-
nado passes over a lake or the ocean, it picks up water and
is called a “waterspout.”
Dust devils are similar in structure to tornadoes, but
they are generally smaller and less intense and they usually
occur in fair weather. Diam-
eters range from a few meters Dust devils occur on
to a few dozen meters, with both Earth and Mars.
average heights of several
hundred meters. As with tornadoes, the lifetime of a typi-
cal dust devil is brief, limited to a dozen or so minutes. Dust
devils form in areas of strong surface heating, usually at
the interface between different surface types, such as asphalt

and dirt or even irrigated fields and dirt roads. Typically
they occur under clear skies and light winds, when air close
to the ground is heated to temperatures much hotter than
the cooler air above it—a very unstable condition. Though
their wind speeds seldom exceed 25 m/s, dust devils can
still be destructive as they lift dust and other debris into
the air. Small structures can be damaged and even destroyed
if they have the misfortune of being located in the path of
a strong dust devil. Dust devils are a common sight in the
deserts of the American Southwest.
Dust devils have also been seen on the arid surface of
Mars, fi rst by the Viking orbiters in 1976. In 1997, Mars
Pathfinder sensed one passing right over it. More recently,
Mars Global Surveyor spotted a large number of dust devils,
made easily visible by the shadows they cast on the mar-
tian surface. Most dust devils leave dark meandering trails
behind them where they have vacuumed up bright surface
dust, revealing the dark surface rock that lies beneath the
dust (Figure 8.19). Dust devils on Mars—typically higher,
wider, and stronger than their Earth counterparts—reach
heights of up to 8 km and have diameters ranging from a
few dozen to a few hundred meters.
FIGURE 8.18 A large tornado made visible by the debris it

is sweeping up. For scale, note the power line structures at
Is Earth Getting Warmer?
lower right. The state of Earth’s atmosphere at any given time and place
is what we commonly call “weather.” Climate is the term
we use to defi ne the average state of Earth’s atmosphere,
including temperature, humidity, winds, and so on. Earth’s
climate appears to go through lengthy temperature cycles,
usually lasting hundreds of thousands of years and occa-
sionally producing shorter cold periods called “ice ages.”
These oscillations in the mean
global temperature are far Climate is the average
smaller than typical geo- state of Earth’s
graphic or seasonal tempera- atmosphere.
ture variations. But Earth’s
atmosphere is so sensitive to even small temperature changes
that it takes a drop of only a few degrees in the mean global
temperature to plunge our climate into an ice age. We still
do not understand all of the mechanisms for these climate-
changing temperature swings. An external influence, such
as small changes in the Sun’s energy output, may be the
cause. Changes in Earth’s orbit or the inclination of its rota-
tion axis also have been suggested. Or these temperature
changes may be triggered internally by volcanic eruptions
(which can produce global sunlight-blocking clouds or
hazes) or long-term interactions between Earth’s oceans and
its atmosphere.
Nearly all climatologists now believe that changes in
Earth’s climate have been accelerating recently and that
Earth is becoming warmer. Temperature measurements over
FIGURE 8.19 Meandering dark trails cross a martian crater the past century show a steady increase in the mean global
where numerous dust devils have removed bright surface dust. temperature. Does this trend represent the beginning of a
long-term change caused by the buildup of human-made tures for the two planets, we concluded that they should be
greenhouse gases (Figure 8.20)—as most computer models very close. But that was before we considered the green-
suggest—or merely a temporary short-term cycle? This is a house effect and the role of carbon dioxide in blocking the
question that some still debate. However, we know that our infrared radiation that a planetary surface typically emits.
atmosphere is a delicately bal- The atmospheric pressure at the surface of Venus is a crush-
anced mechanism. Tiny Increases in Earth’s ing 92 times greater than that at the surface of our own
changes can produce enor- greenhouse gases can planet—equal to the water
mous and often unexpected lead to disastrous pressure at an ocean depth of Venus has a massive
results. Earth’s climate is an climate change. 900 meters. (The pressure at carbon dioxide
example of the sort of com- this ocean depth would crush atmosphere, making it
plex, chaotic system we will discuss in Chapter 10. To add the hull of a World War II–era a “poster child” for
to the complexity, Earth’s climate is intimately tied to ocean submarine.) Most of this mas- the greenhouse effect.
temperatures and currents. We see examples of this connec- sive atmosphere (96 percent)
tion in the periodic El Niño and La Niña conditions, when is carbon dioxide, with a small amount (3.5 percent) of nitro-
small shifts in ocean temperature cause much larger global gen and still lesser amounts of other gases. This thick blan-
changes in air temperature and rainfall. Recent studies sug- ket of carbon dioxide effectively traps the infrared radiation
gest that changes in the flow of the Gulf Stream in the North from Venus, driving the temperature at the surface of the
Atlantic can have very large and unpredictable effects on planet to a sizzling 737 K, which is hot enough to melt lead.
the climates of North America and northern Europe—and Whereas Earth is a lush paradise, the runaway greenhouse
that these changes may take place not over centuries but effect has turned Venus into a convincing likeness of hell—
within a matter of decades! Are we unknowingly jeopardiz- an analogy made complete by the presence of choking
ing our future by meddling with our atmosphere? In a sense, amounts of sulfurous gases. Venus may be our “sister” planet
we are much like children playing with matches. in many respects, but it will be a very long time before
humans visit its surface, if ever.
As on Earth, the atmospheric temperature of Venus
decreases continuously throughout the planet’s troposphere,
8.4 Venus Has a Hot, dropping to a low of about 160 K at the tropopause. At an
altitude of approximately 50 km, the Venus atmosphere is
Dense Atmosphere similar to our own atmosphere at sea level in both pres-
sure and temperature. At altitudes between 50 and 80 km
Venus and Earth are similar in many ways—so similar that (see Figure 8.6b), the atmosphere is cool enough for sulfu-
they might be thought of as sister planets. Indeed, when we rous oxide vapors to react with water vapor to form dense
used the laws of radiation in Chapter 4 to predict tempera- clouds of concentrated sulfuric acid droplets (H2SO4). These
FIGURE 8. 20 Our climate

KEY is delicately balanced, and
380 Difference from average 0.8 carbon dioxide plays a vital
global temperature role in that balance. This graph
Atmospheric CO2 (parts per million)
(temperature anomaly) shows that global temperatures

370 Atmospheric CO2 0.6 are climbing along with
Temperature anomaly (°C)
concentrations of CO2. Annual

360 0.4 variations in atmospheric CO2
can be attributed to seasonal
variations in plant life and fossil
350 0.2 fuel use, while the steady climb
is due to human activities.
340 0
330 –0.2
320 –0.4
310 –0.6
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Year
to the stars, Venus spins once every 243 Earth days, but a
solar day on Venus—the time it takes for the Sun to return
to the same place in the sky (see Chapter 2)—is only 117
Earth days. Regardless, this slow rotation means that Corio-
lis effects on the atmosphere are small, resulting in a global
circulation that is quite close to a classic Hadley pattern
(see Figure 8.10a). Venus is the
only planet known to behave Surface temperatures
in this way. Its massive atmo- on Venus vary little
sphere is highly efficient in from pole to equator
transporting thermal energy or from day to night.
around the planet, so the polar
regions are only a few degrees cooler than equatorial regions,
and there is almost no temperature difference between day
and night. Because the Venus equator is nearly in the plane
of its orbit, seasonal effects are quite small, producing only
negligible changes in surface temperature. (Recall the dis-
cussion in Chapter 2 about how the seasons change because
of the tilt of Earth’s equator relative to the plane of its orbit
around the Sun.) Such small temperature variations also
mean that wind speeds near the surface of Venus are quite
low, typically about a meter per second. High in the rarefied
atmosphere, though, where temperature differences can be
FIGURE 8. 2 1 Thick clouds obscure our view of the surface of
larger, winds reach speeds of 110 m/s, circling the planet
Venus.
in only 4 days. The variation of this high-altitude wind
speed with latitude can be seen in the chevron, or V-shaped,
dense clouds completely block our view of the surface of cloud patterns in Figure 8.23.
Venus, as Figure 8.21 shows. In the 1960s, spacecraft with Large variations in the observed amounts of sulfurous
cloud-penetrating radar provided low-resolution views of compounds in the high atmosphere of Venus suggest to plan-
the surface of Venus. But it was not until 1975, when the etary scientists that the source of sulfur may be sporadic
Soviet Union succeeded in landing cameras there, that we episodes of volcanic activity. This evidence strengthens the
got a clear picture of the surface. Radar images taken by possibility that Venus remains a volcanically active planet.
the Magellan spacecraft in the early 1990s (see Figure 7.24)
produced a global map of the surface of Venus.
Imagine yourself standing (and surviving) on the sur-
face of Venus. Since sunlight cannot easily penetrate the 8.5 Mars Has a Cold,
dense clouds above you, noontime on the surface of Venus
is no brighter than a very cloudy day on Earth. High tem- Thin Atmosphere
peratures and very light winds keep the lower atmosphere
of Venus free of clouds and hazes. The local horizon can be Compared to Venus, the surface of Mars is almost hospitable.
seen clearly, but distant mountains are not as clear. Mole- For this reason we can confidently expect that humans will
cules in even a pure gas will scatter light, and the scattering eventually set foot on the red
efficiency increases sharply with decreasing wavelength. planet, quite likely before the The surface of
Strong scattering by molecules in the dense atmosphere end of the 21st century and Mars is cold and
of Venus would greatly soften any view you might have of possibly much sooner. What the air is thin.
distant scenes. Molecular scattering, always stronger at the they will see is a stark, water-
shorter wavelengths, as discussed in Excursions 8.2, causes less landscape, colored reddish by the oxidation of iron-
a loss of contrast and adds a bluish cast to distant terrain. bearing surface minerals. The sky will sometimes be a dark
(We see the same effect, but to a lesser extent, in our own blue but more often a pinkish color caused by windblown
atmosphere.) The high atmospheric temperatures on Venus dust (Figure 8.24). The lower density of the Mars atmosphere
also mean that neither liquid water nor liquid sulfurous makes it more responsive than Earth’s to heating and cool-
compounds can exist on its surface, leaving an extremely ing, so temperature extremes are greater. Near the equator
dry lower atmosphere with only 0.01 percent water and at noontime, future astronauts may experience a comfort-
sulfur dioxide vapor. able 20°C—about the same as a cool room temperature.
Unlike most other planets, Venus rotates on its axis in Nighttime temperatures typically drop to a frigid –100°C,
the opposite sense of its motion around the Sun. Relative and during the polar night the air temperature can reach
E XCURSIONS 8. 2
Blue Skies, White Clouds, and Red Sunsets

Have you ever noticed the beam from a fl ashlight or the head- fraction of the blue photons in the sunlight are scattered, and
lights of a car on a foggy night? As the beam of light shines when you look at the sky your eyes detect the blue photons that
through the cloud bank, some of the light bounces off tiny water were scattered in your direction. As illustrated in Figure 8.22,
droplets, and you see the light that happens to bounce in your that is why the sky is blue.
direction. In this way, you “see the beam of light.” This situation changes dramatically as evening approaches.
The bouncing of light off small particles in its path is called As the Sun drops lower in the sky, the light from the Sun must
scattering. If the particles that are scattering light are much larger pass through more air before it reaches you. As the Sun nears
than the wavelength of that light, as is the case for water drop- the horizon, sunlight passes through hundreds of times more
lets in a fog bank, then photons of all colors are equally likely air than it did when the Sun was high overhead. So much of the
to be scattered. As a result, the light scattered by a cloud is the blue light is scattered away that by the time the light reaches
same color as the light shining on the cloud. Sunlight is white, so you, the Sun looks orange. Tiny particles of dust and other
clouds that are illuminated by direct sunlight are also white. materials in the atmosphere (which are similar in size to the
Things get more interesting if the scattering particles are wavelength of light) scatter away additional blue light, leaving
smaller than the wavelength of the light they are interacting only a glorious red.
with. In such instances, shorter-wavelength photons are more The next time you are captivated by the beauty of a sun-
likely to be scattered than longer-wavelength photons. Mol- set, remember also to look up at the deepening blue of the sky
ecules in Earth’s atmosphere, for example, scatter blue light overhead. The red sunset is white sunlight minus the blue
(λ = 400 nanometers, or nm) about seven times more effectively light that was scattered away. The blue sky is scattered blue
than they scatter red light (λ = 650 nm). When the Sun is high light only. The red of sunset, the blue of the sky, and the white
overhead, sunlight follows a short path through the atmosphere of a billowing cloud are three facets of the same gem—a gem
and so is relatively unaffected by scattering. Even so, a small called scattering.
FIGURE
(a) 8 . 22 The blue of
a daytime sky (a)
and the red of a
sunset (b) are both
due to differences
in the way Earth’s
atmosphere
scatters light
of different
The blue of the sky is short-wavelength wavelengths.
light scattered from sunlight by Earth’s
atmosphere, but at midday most
sunlight reaches the ground.
(b)
The red of sunset is sunlight
after short-wavelength blue
light has been scattered away.
FIGURE 8. 23 This Pioneer Venus image, taken in ultraviolet

light, shows clouds of sulfur compounds in the atmosphere of FIGURE 8. 2 4 A view of the dust-filled pink sky of Mars as
Venus. Variations in wind speed with latitude cause the clouds to seen from the Viking lander. In the absence of dust, the sky’s thin
streak out into a “chevron,” or V-shaped, pattern. atmosphere would appear deep blue.
–150°C—cold enough to freeze carbon dioxide out of the air scientists over the possibility of extant life on Mars (see
in the form of a dry-ice frost. For human visitors, the low Excursions 8.1). Atmospheric methane is quickly destroyed
surface pressure will certainly be uncomfortable. The aver- by solar ultraviolet radiation, typically lasting only a few
age atmospheric surface pressure of Mars is equivalent to hundred years, so its very survival in the martian atmo-
the pressure at an altitude of 35 km on Earth, far higher sphere implies the existence of a continuous source to
than our highest mountain. There is, of course, no “sea level” replace the losses. Life is certainly one possibility. Methane
on Mars. Surface pressure varies from 11.5 mb (1.1 percent in Earth’s atmosphere is largely a product of microorgan-
of Earth’s pressure at sea level) in the lowest impact basins isms, and similar life-forms might exist in protected envi-
of Mars, to 0.3 mb at the summit of Olympus Mons. ronments far below the martian surface. On the other hand,
Like Earth, Mars does have some water vapor in its atmo- many scientists argue that geological processes remain as
sphere, but the low temperatures condense much of it out reasonable alternative possibilities. The conversion of iron
as clouds of ice crystals (see Figure 8.6b). Early-morning oxide into certain complex minerals releases methane as a
ice fog in the lowlands (Figure 8.25a) and clouds hanging by-product. Ancient methane trapped within ice crystals
over the mountains will give Earth visitors some reminders is another possible source. The likelihood of life existing
of their home planet. Although the cold, thin air might be on Mars today continues to be one of the great unanswered
endured, what is seriously lacking and what future astro- questions of our time.
nauts must carry with them is oxygen. Without plants, Mars The inclination of the Mars equator to its orbital plane
has only a tiny trace of this life-sustaining gas. Like Venus, is similar to Earth’s, so it has similar seasons (see Chapter
the atmosphere of Mars is composed almost entirely of car- 2). But the effects are larger for
bon dioxide (95 percent), with a lesser amount of nitrogen two reasons: Mars varies more Seasonal changes
(2.7 percent). The near absence of oxygen also means that in its annual orbital distance affect climate more on
Mars can have very little ozone, and this lack of ozone from the Sun than does Earth, Mars than on Earth.
allows solar ultraviolet radiation to reach all the way to the and the low density of the
surface. To survive on Mars, any surface life-forms would Mars atmosphere makes it more responsive to seasonal
have to develop protective layers that could shield against change. The large diurnal, seasonal, and latitudinal surface
the lethal ultraviolet rays. temperature differences on Mars often create locally strong
The recent discovery of significant amounts of methane winds, some estimated to be higher than 100 m/s. High winds
in the martian atmosphere has refueled arguments among can stir up huge quantities of dust (Figure 8.25b) and dis-
(a) (b)
FIGURE 8 . 25 (a) Patches of
early-morning water vapor fog
forming in canyons on Mars.
(b) A dust storm raging in the
canyon lands of Mars.
tribute it around the planet’s surface. For more than a cen- once flowed across its surface, as we saw in Chapter 7, but
tury, astronomers have watched the seasonal development its low gravity was responsible for the loss of much of this
of springtime dust storms on Mars. The stronger ones spread earlier atmosphere. This is a good example of a runaway
quickly and within a few weeks can envelop the entire planet atmosphere, as we discussed in Section 8.2.
in a shroud of dust (Figure 8.26). Such large amounts of
windblown dust can take many months to settle out of the
atmosphere. Seasonal movement of dust from one area to
another alternately exposes and covers large areas of dark,
rocky surface. This phenomenon led some astronomers of
8.6 Mercury and the
the late 19th and early 20th centuries to believe that they Moon Have Almost
were witnessing the seasonal growth and decay of vegeta-
tion on Mars. Public imagination carried the astronomers’ No Atmosphere
interpretations a step further—to stories about advanced
civilizations on Mars and invasions of Earth by warlike Our story would not be complete if we did not include the
Martians (a theme found in some movies). kinds of “atmosphere” that exist today around the remain-
Mars likely had a more massive atmosphere in the dis- ing two terrestrial objects, Mercury and the Moon. There
tant past, and geological evidence suggests that liquid water is actually little to say. The atmospheres of Mercury and
the Moon would hardly be noticed by a visi-
tor from Earth. They are less than a million-
billionth (10 –15) as dense as our own, and they
probably vary somewhat with the strength
of the solar wind and the atoms of hydro-
gen and helium they capture from it. Other
atoms, such as sodium, calcium, and even
water-related ions, were seen in Mercury’s
atmosphere by the Messenger spacecraft, and
may have been blasted loose from Mercury’s
surface by the solar wind or micrometeor-
oids. Such ultrathin atmospheres can have no
effect whatsoever on local surface tempera-
tures. Unless you are one of the few astrono-
mers interested in the interaction between
solar wind and solid bodies, you may com-
fortably ignore the atmospheres of Mercury
and the Moon.
FIGURE 8. 26 Hubble Space Telescope images showing the development of
a global dust storm that enshrouded Mars in September 2001
If we step outward from the Sun, we fi nd Mars. Like

Seeing the Forest for the Trees Venus, Mars is a planet that is not so different from
Earth. Although smaller and less massive, it still has
While writing these words, one of the authors is sitting in strong enough gravity to hold on to its atmosphere. But
his backyard on a pleasant summer morning. As birds fly whereas Venus was too hot, Mars when it was young
overhead, he is warmed by the rays of the Sun and cooled was too cold. It had a thick atmosphere, and rain fell in
by a gentle morning breeze. Such an idyllic setting might torrents. Images of the surface of the planet show flood
seem far from the path of our journey through astronomy, basins into which the Amazon would have been but a
but nothing could be further from the truth. Some of the minor tributary. Liquid water on the surface of Mars
atoms drifting past arrived at Earth as it formed, later to was too effective at scrubbing the planet’s atmosphere of
be expelled by volcanism into Earth’s forming second- carbon dioxide. The process that prevented Earth from
ary atmosphere. The rest, especially volatile materials, becoming a Venus-like hothouse got out of hand on Mars,
arrived in the rain of ice-laden comets that fell on the and the temperature fell until the planet’s atmosphere
young Earth. During the ensuing billions of years, a deli- nearly froze out.
cate chemical and physical dance has taken place. The We owe our lives to the thin blanket of atmosphere
temperature of the young Earth was such that water could that covers our planet. At one time we may have felt
condense and fall on the surface as rain. The presence of justified in taking this atmosphere for granted, but after
liquid water served to scrub our atmosphere, absorbing looking around us in the Solar System we have come to
carbon dioxide and locking it up in carbon-bearing rock appreciate that our world is maintained by the most deli-
such as limestone. cate of balances. Over billions of years, life has shaped
Liquid water also provided the bath in which organic the atmosphere of our planet, and today through the
chemistry could take place, leading first to molecules that activities of humans, life is reshaping our planet’s atmo-
had the useful property of being able to reproduce them- sphere once again.
selves at the expense of other molecules, and later to life Great political debate surrounds issues such as the
itself. Early life on Earth consumed molecules such as release of ozone-destroying chlorofluorocarbons and
carbon dioxide. The waste products formed by this life other halogen compounds into Earth’s atmosphere, as
included what, at the time, was a deadly poison— oxygen. well as the destruction of oxygen-producing plant life and
Over the eons the amount of oxygen in our planet’s atmo- the release of huge amounts of carbon dioxide through
sphere increased, and new forms of life evolved to take the burning of fossil fuels. In the rush for profit and con-
advantage of its presence. We are among these later life- venience, there are those who say that it is too early to
forms. To us, the reactive, corrosive gas called oxygen is worry about the effects that humans are having on the
the breath of life. planet. “Earth is too complex,” they say, “and our models
In contrast to Earth, Mercury and the Moon were too primitive to accurately predict the effect that human
simply too small to hold on to either their primary or activities will have on our atmosphere and climate.”
their secondary atmospheres, leaving them as airless This burden of proof is incorrectly placed, however.
rocks. Looking inward from Earth toward the Sun, we We know that factors such as the greenhouse effect made
fi nd a planet that would seem more like our own home. our planet what it is today. In the absence of the green-
Venus has much the same size and mass as Earth and house effect, Earth would have an average temperature
it is a similar distance from the Sun. We might have below the freezing point of water. We also know that
imagined Venus to be the ideal world for future human human activities are measurably changing the chemical
colonization. Yet its history has been different from balance of our atmosphere. We are undeniably playing
that of Earth. Slightly warmer than the young Earth, with the knobs that regulate our climate.
the young Venus was too hot for liquid water to pool The only intellectually honest way to look at the
on its surface, allowing the same greenhouse effect situation is to start with the hypothesis that unchecked
responsible for maintaining Earth’s balmy climate to human activity such as the destruction of rain forests
run away with itself. The result is an environment that and the burning of fossil fuels will have a significant,
is hellish beyond even the fevered imagination of Dante adverse effect on the future of our planet. We point out
Alighieri. that this hypothesis has not been disproved.
Summary which the Sun and planets formed. Where did Earth’s
nitrogen come from?
• Primary atmospheres consist mainly of hydrogen and 5. Some of Earth’s water was released aboveground by
helium captured from the protoplanetary disk. volcanism. What is another likely source of Earth’s
water?
• The terrestrial planets lost their primary atmospheres
to space soon after the planets formed. *6. The force of gravity holds objects tightly to the surfaces
of the terrestrial planets. Yet atmospheric molecules
• Secondary atmospheres are created by volcanic gases,
are constantly escaping into space. Explain how these
and from volatiles brought in by impacting comets.
molecules are able to overcome gravity’s grip. How does
• The atmospheric greenhouse effect keeps Earth from the mass of a molecule affect its ability to break free?
freezing, but it turns the atmosphere of Venus into an
*7. Explain what is meant by a runaway atmosphere and how
inferno.
it works.
• Plant life is responsible for the oxygen in Earth’s
8. We attribute the warming of a planet’s surface to the
atmosphere.
atmospheric greenhouse effect. How does this mech-
• Temperature and pressure decrease with altitude in the anism differ from the warming of a conventional
tropospheres of Venus, Earth, and Mars. greenhouse?
• Earth’s magnetosphere shields us from the solar wind. 9. Name at least two greenhouse molecules other than
carbon dioxide.
• Nonuniform absorption of solar energy is the cause of
all our weather. 10. In what way is the atmospheric greenhouse effect ben-
eficial to terrestrial life?
• The Coriolis effect causes hurricanes to rotate.
*11. Why is Venus very hot and Mars very cold if both of
• Venus has a massive, hot carbon dioxide atmosphere.
their atmospheres are dominated by carbon dioxide,
• Mars has a thin, cold carbon dioxide atmosphere. an effective “greenhouse” molecule?
• The atmospheres of the Moon and Mercury are almost 12. In what ways does plant life affect the composition of
nonexistent. Earth’s atmosphere?
• Human modification of Earth’s atmosphere may lead to 13. Ozone is often vilified as an urban pollutant, yet our
unintended consequences. very lives depend on it. Explain how ozone protects
terrestrial life.
14. Eventually astronomers will have the technical capabil-
SmartWork, Norton’s online homework system, ity to detect Earth-like planets around other stars. When
includes algorithmically generated versions of these we find such objects, what observations would indicate
questions, plus additional conceptual exercises. If with near certainty that they harbor some form of life
your instructor assigns questions in SmartWork, log in at as we know it?
*15. You check the barometric pressure and find that it is
reading only 920 mb. Two possible effects could be
responsible for this lower-than-average reading. What
are they?
Student Questions
16. What mechanism produces the aurora borealis (north-
THINK ING A BOUT THE CONCEP T S ern lights)?
17. What is the principal cause of winds in the atmospheres
1. Describe the origin and fate of primary atmospheres.
of the terrestrial planets?
2. Primary atmospheres of the terrestrial planets were com-
18. How does the solar wind affect Earth’s upper atmo-
posed almost entirely of hydrogen and helium. Explain
sphere, and what effects can it have on society?
why they contained only these gases and not others.
19. Global warming appears to be responsible for increased
3. How were the secondary atmospheres of the terrestrial
melting of the ice in Earth’s polar regions.
planets created?
a. Why does the melting of arctic ice, which floats on
4. Nitrogen, the principal gas in Earth’s atmosphere, was the surface of the Arctic Ocean, not affect the level
not a significant component of the protostellar disk from of the oceans?
b. How is the melting of glaciers in Greenland and Ant- c. What would you guess is the reason we do not see
arctica affecting the level of the oceans? more evidence of wind erosion on Earth?
20. Why are we unable to get a clear view of the surface of *29. Suppose you seal a rigid container that has been open
Venus, as we have so successfully done with Mars? to air at sea level when the temperature is 0°C (273 K).
The pressure inside the sealed container is now exactly
21. Assume you are somehow able to survive on the surface
equal to the outside air pressure: 105 newtons per square
of Venus. Describe your environment.
meter (N/m2).
22. Explain why surface temperatures on Venus hardly vary a. What would be the pressure inside the container if
between day and night and between the equator and the it were left sitting in the desert shade where the sur-
poles. rounding air temperature was 50°C (323 K)?
b. What would be the pressure inside the container if it
23. In 1975 the Soviet Union landed two camera-equipped
were left sitting out in an Antarctic night where the
spacecraft on Venus, giving planetary scientists their
surrounding air temperature was –70°C (203 K)?
fi rst (and only) close-up views of the planet’s surface.
c. What would you observe in each case if the walls of
Both cameras ceased to function after only an hour.
the container were not rigid?
What environmental conditions most likely led to their
demise? 30. Oxygen molecules (O2) are 16 times more massive than
hydrogen molecules (H2). Carbon dioxide molecules
24. Humans may eventually travel to the surface of Mars.
(CO2) are 22 times more massive than H2.
Describe the environment they will experience.
a. Compare the average speed of O2 and CO2 molecules
*25. Mars has seasons similar to those of Earth, but more in a volume of air.
extreme. Explain why. b. Does this ratio depend on air temperature?
26. Mercury and the Moon have extremely tenuous *31. Using the average density of air at sea level (1.225 kilo-
atmospheres. What are two possible sources of these grams per cubic meter, or kg/m3) and the average mass
atmospheres? of Earth’s atmosphere above sea level per square meter
(1.033 × 104 kg/m2), what would be the total depth of
Earth’s atmosphere (in kilometers) if its density were the
same at all altitudes? (This is called a scale height [H], a
A P P LY I N G T H E C O N C E P T S useful quantity for comparing Earth’s atmosphere with
the atmospheres of other planets.)
27. The total mass of Earth’s atmosphere is 5 × 1018 kg. Car-
bon dioxide (CO2) makes up about 0.06 percent of Earth’s 32. The average surface pressure on Mars is 6.4 mb. Using
atmospheric mass. Figure 8.6a, estimate how high you would have to go in
a. What is the mass of CO2 (in kilograms) in Earth’s Earth’s atmosphere to experience the same atmospheric
atmosphere? pressure that you would experience if you were stand-
b. The annual global production of CO2 is now estimated ing on Mars.
to be 2.6 × 1013 kg. What annual fractional increase
33. Water pressure in Earth’s oceans increases by 1 bar for
does this represent?
every 10 meters of depth. Compute how deep you would
28. The ability of wind to erode the surface of a planet is have to go to experience pressure equal to the atmo-
related in part to the wind’s kinetic energy. spheric surface pressure on Venus.
a. Compare the kinetic energy of a cubic meter of air at
sea level on Earth (mass 1.23 kg) moving at a speed StudySpace is a free and open Web site that pro-
of 10 m/s with a cubic meter of air at the surface of vides a Study Plan for each chapter of 21st Century
Venus (mass 64.8 kg) moving at 1 m/s. Astronomy. Study Plans include animations, read-
b. Compare the terrestrial case with a cubic meter of ing outlines, vocabulary flashcards, and multiple-choice
air at the surface of Mars (mass 0.015 kg) moving at a quizzes, plus links to premium content in SmartWork and
speed of 50 m/s. the ebook. Visit www.wwnorton.com/studyspace.
Before the starry threshold of Jove’s Court
My mansion is …
Above the smoke and stir of this dim spot
Which men call earth.
JOHN MILTON (1608 –1674)
Magnificent Saturn looms before the cameras of the Cassini spacecraft.

CHAP TER 9
Worlds of Gas and Liquid—

The Giant Planets
9.1 The Giant Planets—
K E Y CONCEP T S
Distant Worlds, Unlike the solid planets of the inner Solar System, four
Different Worlds worlds in the outer Solar System were able to capture
and retain gases and volatile materials from the Sun’s
The four largest planets in our Solar System are Jupiter, protoplanetary disk and swell to enormous size and
Saturn, Uranus, and Neptune, and all have characteristics mass. Examining these giant planets, we will discover
that clearly distinguish them from the terrestrial, or Earth-
• Two gas giants, composed primarily of hydrogen and
like, planets. The most obvious difference is that they are
helium; and two ice giants, composed primarily of
all enormous compared with their terrestrial cousins. Even
the smaller of them, Uranus and Neptune, are nearly four water and other volatile materials.
times the size of Earth. Another distinguishing feature is • Atmospheres and oceans, but no solid surfaces.
their very low density. All are composed almost entirely of • How changes in temperature and pressure with
light materials such as hydrogen, helium, and water, rather
increasing depth lead to changes in the chemical
than the rock and metal that make up the terrestrial planets.
composition of clouds in giant planet atmospheres.
Collectively, we call Jupiter, Saturn, Uranus, and Neptune
the giant planets (Table 9.1), although you may sometimes • Gravitational energy being converted into thermal
hear them referred to as the “Jovian planets,” after Jupiter, energy in the interiors of three giant planets, driving
the largest (“Jove” is another name for Jupiter, the highest- strong convection in their atmospheres.
ranking Roman deity). • How the Coriolis effect on these rapidly rotating worlds
Even though the giant planets share many characteris-
turns convective motions into powerful winds, huge
tics, there are significant differences among them. Jupiter
storms, and planet-spanning bands of multi-hued
and Saturn are similar to one
clouds.
another in size and are com- Jupiter and Saturn
posed primarily of hydrogen are gas giants. • Extreme conditions deep within the interiors of the
and helium. Uranus and Nep- Uranus and Neptune giant planets.
tune are also similar to one are ice giants. • Brilliant auroras and glowing doughnuts of gas
another in size, and both con-
associated with the huge magnetospheres and strong
tain much larger amounts of water and other ices than
magnetic fields of the giant planets.
do Jupiter or Saturn. These differences are sufficiently
large that many astronomers feel it makes sense to divide
the giant planets into two classes: Jupiter and Saturn as
247
248 Chapter 9 Worlds of Gas and Liquid—The Giant Planets
TA B L E 9 .1
Physical Properties of the Giant Planets
Jupiter Saturn Uranus Neptune
Orbital radius (AU) 5.2 9.6 19.2 30.0
Orbital period (years) 11.9 29.5 84.0 164.8
Orbital velocity (km/s) 13.1 9.7 6.8 5.4
Mass (M⊕ = 1) 318 95 14.5 17.1
Equatorial diameter (1,000 km) 143 120.5 51.1 49.5
Equatorial diameter (D⊕ = 1) 11.2 9.45 4.0 3.9
Oblateness 0.065 0.098 0.023 0.017
Density (water = 1) 1.3 0.7 1.3 1.6
Rotation period (hours) 9.9 10.7 17.2 16.1
Obliquity (degrees) 3.1 26.7 97.8 29.3
Surface gravity (m/s2) 25.1 10.5 9.0 11.2
Escape speed (km/s) 60.2 35.5 21.3 23.5
(a) (b)
FIGURE 9.1 The gas giants. (a) Jupiter imaged by

Cassini. (b) Saturn seen via the Hubble Space Telescope.
The giant planets orbit the Sun far beyond the orbits of
Earth and Mars. The closest to the Sun, and to us as well,
gas giants (Figure 9.1) and Uranus and Neptune as ice giants is Jupiter, and even Jupiter is
(Figure 9.2). Other differences between the gas giants and more than five times as far Giant planets orbit
the ice giants1 will become evident as we more closely exam- from the Sun as Earth is. Nep- much farther from the
ine their physical and chemical characteristics. tune, the most distant, is 41⁄2 Sun than do the
billion kilometers (km) away, terrestrial planets.
1
Remember from Chapter 6 that planetary scientists often refer to vola- or some 30 times farther from
tile substances as ices even when they are in their liquid or gaseous the Sun than we are. To put this distance into perspective,
form. Uranus and Neptune may be called ice giants, but that does not if you were traveling at the speed of a commercial jetliner
mean that they are solid, frozen worlds. it would take you more than 500 years to reach Neptune.
9.1 The Giant Planets—Distant Worlds, Different Worlds 249
(a)
Two Were Known to Antiquity, and
Two Were Telescopic Discoveries
Although very far away, the giant planets are so large that
all but Neptune can be seen with the unaided eye. Jupiter
and Saturn are a familiar sight in the nighttime sky, com-
parable to the brightest stars. Along with Mercury, Venus,
and Mars, they were among the five planets (a Greek word
meaning “wandering stars”) known to ancient cultures. In
contrast, Uranus appears only slightly brighter than the
faintest stars visible on a dark night, and Neptune cannot
be seen without the aid of binoculars.
Uranus was the first planet to be “discovered,” but it was
not found until late in the 18th century, more than 170 years
after Galileo made his fi rst astronomical observations with
a telescope. In 1781 William Herschel, a German-English
professional musician and amateur astronomer, came upon
(b)
it quite by accident. Herschel was producing a catalog of the
sky at his home in Bath, England, when he noticed a tiny
disk in the eyepiece of his six-inch telescope. At fi rst he
thought he had found a comet, but the object’s slow nightly
motion soon convinced him that it was a new planet beyond
the orbit of Saturn. Politically astute, Herschel proposed
calling his new planet Georgium Sidus (“George’s Star”)
after King George III of England. Obviously pleased, the
monarch rewarded Herschel with a handsome lifetime pen-
sion. The astronomical community, however, later rejected
Herschel’s suggestion, preferring the name Uranus instead.
For 65 years Uranus would remain the most distant known
planet.
Over the decades that followed Herschel’s discovery,
astronomers found to their dismay that Uranus was straying
from its predicted path in the sky. Mathematicians viewed
this aberrant behavior with grave concern. By then, New-
ton’s laws of motion had been fi rmly established and were
the basis for predicting the motion of the newly discovered
planet. Could something be wrong with the theory? As a rea-
sonable explanation, a few astronomers suggested that the
gravitational pull of an unknown planet might be respon-
FIGURE 9. 2 The ice giants. Uranus (a) and Neptune (b) imaged sible for this “unacceptable” behavior of Uranus. Using
in visible light by Voyager. measured positions of Uranus provided by astronomers,
two young mathematicians, Urbain-Jean-Joseph Le Verrier
(1811–1877) in France and John Couch Adams (1819–1892) in
England, independently predicted where the hypothetical
The Sun shines only dimly and provides very little planet should be. Although Adams was fi rst to compute his
warmth in these remote parts of the Solar System. If we were predictions, he was unable to gain the attention of England’s
to journey to Jupiter, we would see the Sun as but a tiny disk, Astronomer Royal, so the opportunity for England to gain
1
⁄27 as bright as it appears from Earth. At Neptune, the Sun credit for a second planetary discovery was lost. Meanwhile,
would no longer show a disk but would appear as a brilliant Le Verrier was having similar problems convincing French
starlike point of light about 500 times brighter than the full astronomers. It was a German who would fi nally triumph.
Moon in our own sky. Daytime on Neptune is equivalent to Armed with Le Verrier’s predictions, Johann Gottfried Galle
a perpetual twilight here on our own planet. With so little (1812–1910) began a search at the Berlin Observatory. He
sunlight available for warmth, daytime temperatures hover found the planet on his fi rst observing night, just where Le
around 123 kelvins (K) at the cloud tops on Jupiter, and they Verrier and Adams had predicted it would be. Galle’s “dis-
can dip to just 37 K on Neptune’s moon Triton. covery” of Neptune in 1846 became a triumph for math-
ematical prediction based on physical law— and for the decades of the 20th century. Modern instruments on both
subsequent confi rmation of theory by observation. groundbased telescopes and the Hubble Space Telescope
Neptune became the eighth known planet, and it remains (HST) have provided new insight into the composition and
the outermost classical planet in our Solar System. As we physical structure of the giant planets; but our greatest leaps
will learn in Chapter 12, erroneous discrepancies between of knowledge have come from close-up observations made
the predicted and observed positions of Uranus and Nep- possible by the planetary probes: Pioneer, Voyager, Galileo,
tune during the 19th century led to the fortuitous discov- and Cassini. Over the past several decades, one or another
ery of Pluto in 1930. Pluto was immediately declared to of these spacecraft has visited each of the four giant plan-
be the ninth planet. In 2006, however, the International ets, sending back a wealth of scientific data with a level of
Astronomical Union officially reclassified Pluto as a dwarf detail that cannot be obtained from the vicinity of Earth.
planet. Indeed, Pluto’s characteristics are much closer to Although groundbased telescopes and the HST have made
those of a huge comet nucleus than to either a giant or a significant contributions to our knowledge of the giant plan-
terrestrial planet. For this reason we will discuss Pluto in ets, much of what follows is based on what we have learned
Chapter 12 when we talk about the smaller bodies of the from planetary probes.
Solar System.
The Giant Planets Are

Large and Massive
9.2 How Giant Planets Figure 9.3 compares the true relative sizes of the giant plan-
Differ from ets with their relative sizes as seen from Earth. Jupiter is not
only the largest of the eight planets; it is more than one-tenth
Terrestrial Planets the size of the Sun itself and more than 11 times the size of
Earth. Saturn is only slightly smaller than Jupiter, with a
Historically, the vast distances to the giant planets have diameter of 9.5 Earths. Uranus and Neptune are each about
made them difficult objects for scientific study. All of 4 Earth diameters across, with Neptune being the slightly
this changed with the arrival of the space age and with smaller of the two. The most accurate measurements of
the simultaneous development of more powerful optical planet sizes have come from observing the length of time it
and electronic groundbased instruments during the latter takes a planet to eclipse, or “occult,” a star. We call these
FIGURE 9. 3 (a) Images of the giant planets, shown to the same physical scale. (b) The same
images, scaled according to how the planets would appear as seen from Earth.
0 50,000 100,000 150,000 200,000

kilometers
(a)
Jupiter Saturn Uranus Neptune
(b)
0 10 20 30 40 50
arcseconds
As Earth and a planet orbit the Sun, the planet FIGURE 9. 4 Occultations
may occult the light from a distant star. occur when a planet, moon,
or ring passes in front of a
Earth star. Careful measurements
Light from a distant star
Time 3 of changes in the starlight’s
Time 3 Planet brightness and the duration of
Telescope these changes give information
Time 2 Time 2 about the size and properties of
the occulting object.
Time 1
Time 1
View from Time 1 Time 2 Time 3

Earth
This is the
appearance of
the planet and
star from Earth.
The length of the occultation

depends on the known orbital
Brightness of star
motions and the size of the planet.
Time 1 Time 2 Time 3
events stellar occultations (Figure 9.4). For example, New- times as massive as Earth and some 31⁄2 times as massive as
ton’s laws may tell us that a particular planet is moving along Saturn, its closest rival. Jupiter, in fact, weighs in at more
at a speed of precisely 25 kilometers per second (km/s) rela- than twice the mass of all the other planets combined. Even
tive to Earth’s own motion, and we observe that this planet so, its mass is only about a thousandth that of the Sun.
takes exactly 2,000 seconds to Uranus and Neptune, the ice giants, are the lightweights
pass directly in front of a star. Stellar occultations among the giant planets, but each still has the equivalent
The planet then must have a let us determine a of approximately 15 Earth masses (M⊕).
diameter equal to the distance planet’s diameter. How do we measure the mass of a planet? In Chapter 3
it traveled during the 2,000 we learned how a planet’s gravitational attraction can affect
seconds, or 50,000 km. The center of the planet rarely passes the motion of a nearby small
directly in front of a star, of course; but observations of body—say, one of its moons or The motions of a
occultations from several widely separated observatories a passing spacecraft. We planet’s moons allow
provide the geometry necessary to calculate both the planet’s found that the motion of the us to determine the
size and its shape. Occultations of the radio signals trans- small body can be accurately planet’s mass.
mitted from orbiting spacecraft and images taken by the predicted if we know the
spacecraft cameras have also provided accurate measures planet’s mass and apply Newton’s law of gravitation and
of the sizes and shapes of planets and their moons. You may Kepler’s third law. Having seen how this works, we should
be surprised to learn that the giant planets are not perfectly now be able to invert the procedure and calculate the plan-
round. Later in the chapter we will see why. et’s mass by observing the motion of a small body. Prior to
The giant planets represent an overwhelming 99.5 per- the space age, we measured a planet’s mass by observing
cent of all the nonsolar mass in our Solar System. The vast the motions of its moons. This technique, of course, worked
multitude of other Solar System objects—terrestrial plan- only with planets that have moons. (We had to estimate the
ets, dwarf planets, moons, asteroids, and comets— are all masses of Mercury and Venus from their size.) The accuracy
included in the remaining 0.5 percent. Jupiter alone is 318 of those early calculations was limited by how precisely we
could measure the positions of the moons with our ground- Saturn would actually float with 70 percent of its volume
based telescopes. Planetary spacecraft have now made it submerged. Jupiter and Uranus have densities intermediate
possible to measure the masses of planets with unparalleled between those of Neptune and Saturn. As we will see later,
accuracy. As a spacecraft flies by, the planet’s gravity tugs however, the size of the giant planets is not determined by
on it and deflects its trajectory. By tracking and comparing density alone.
the spacecraft’s radio signals using several antennae here The chemical compositions of the giant planets are not
on Earth, we can detect minute changes in the spacecraft’s all the same. Jupiter’s chemical composition is quite similar
trajectory, thereby providing a highly accurate measure of to that of the Sun. Astrono-
the planet’s mass. mers take the relative abun- Jupiter’s chemistry is
dance of the elements in the similar to the Sun’s.
Sun as a standard reference,
We See Only Atmospheres, termed solar abundance. As illustrated in Figure 9.5, hydro-
gen (H) is the most abundant element, followed by helium
Not Surfaces (He). Jupiter has about a dozen hydrogen atoms for every
The giant planets are made up mostly or perhaps entirely atom of helium, which is typical of the Sun and the universe
of gases and liquids. Their characteristic structure is that as a whole. Only 2 percent of its mass is made up of heavy
of a relatively shallow atmosphere merging seamlessly into elements (atoms more massive than helium). Many of these
a deep liquid “ocean,” which in turn merges smoothly into massive elements have combined chemically with hydrogen
a denser liquid or solid core. Although shallow compared (H) because of its great abundance. Thus, atoms of oxygen
with the depth of the liquid layers below, the atmospheres (O), carbon (C), nitrogen (N), and sulfur (S) have combined
of the giant planets are still much thicker than those of the with hydrogen to form molecules of water (H2O), methane
terrestrial planets—thousands of kilometers rather than (CH4), ammonia (NH3), and hydrogen sulfide (H2S), respec-
hundreds. As with Venus, only the very highest levels of tively. More complex combinations produce materials such
their atmospheres are visible to us. In the case of Jupiter or as ammonium hydrosulfide (NH4HS). Helium— and certain
Saturn, we see the tops of a layer of thick clouds, the highest other gases, such as neon and argon— are what we call inert
of many others that lie below. Although a few thin clouds are gases, meaning they do not normally combine with other
visible on Uranus, we find ourselves looking mostly into a elements or with themselves. Most of the iron—the remains
clear, seemingly bottomless atmosphere. Atmospheric mod- of the original rocky planet around which the gas giant
els tell us that thick cloud layers must lie below, but strong grew— and even much of the water have ended up in Jupi-
scattering of sunlight by molecules (see Chapter 8) in the ter’s liquid core.
clear part of the atmosphere prevents us from seeing these Proportionally, Saturn has a somewhat larger inventory
lower cloud layers. Neptune displays a few high clouds with of massive elements than Jupiter. In Uranus and Neptune,
a deep, clear atmosphere showing between them. massive elements are so abundant that they are major com-
positional components of these two planets. The principal
compositional differences among the four giant planets lie
Jupiter’s Chemical Composition Is in the amounts of hydrogen and helium that each of them
contains. This turns out to be an important clue to the pro-
More like the Sun’s Than Earth’s Is cess by which the giant planets formed— a subject we will
In Chapter 7 we learned that the terrestrial planets are com- return to in Section 9.7.
posed mostly of rocky minerals, such as silicates, along with
various amounts of iron and other metals. It is true that the
atmospheres of the terrestrial planets contain lighter mate- Short Days and Nights on
rials, but the masses of these atmospheres— and even of
Earth’s oceans— are insignificant compared with the total
the Giant Planets
planetary mass. The terrestrial planets are thus very com- Still another distinguishing characteristic of the giant plan-
pact, with densities ranging from 3.9 (Mars) to 5.5 (Earth) ets is their rapid rotation, meaning that their days are short.
times that of water. They are the densest objects in the A day on Jupiter is just under 10 hours long, and Saturn’s
Solar System. is only a little longer. Neptune and Uranus have rotation
In contrast, the giant planets are composed almost periods of 16 and 17 hours, respectively, giving them days
entirely of lighter materials, such as hydrogen, helium, and that are intermediate in length between those of Jupiter
water. Consequently, their densities are much lower than and Earth.
the those of the rock-and-metal terrestrial planets. Nep- The rapid rotation of the giant planets distorts their
tune is the most compact among the giant planets, having shapes. If they did not rotate at all, these fluid bodies would
a density about 1.6 times that of water. Saturn is the least be perfectly spherical. In Chapter 6 we used the analogy of
compact— only 0.7 times the density of water. This means a spinning disk of pizza dough to see why a collapsing,
that, placed in a sufficiently large and deep body of water, rotating cloud must settle into a disk. The same principle
The Astronomer’s
Periodic Table
of the Elements
H
The area of the box for each element
is shown in proportion to the fraction
of the atoms of each type found in the
Sun. To the astronomer, the universe
is mostly hydrogen and helium. He
C N O Ne
Na Mg Al Si S Ar
Ca Fe Ni
Periodic Table of the Elements

1 2
1
H He
3 4 5 6 7 8 9 10
2
Li Be B C N O F Ne
11 12 13 14 15 16 17 18
3
Na Mg Al Si P S Cl Ar
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
4
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
The periodic table arranges 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
5
the chemical elements in Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
55 56 57 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
ascending order according 6
Cs Ba La Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
to the number of protons in 87 88 89 104 105 106 107 108 109 110 111 112 114 116 118
7
the nucleus of each element. Fr Ra Ac Rf Db Sg Bh Hs Mt
58 59 60 61 62 63 64 65 66 67 68 69 70 71
FIGURE 9. 5 The traditional periodic table of the elements Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
(lower right) shows the chemical elements laid out in order 90 91 92 93 94 95 96 97 98 99 100 101 102 103
Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
of the number of protons in the nucleus of each. But in
the “astronomer’s periodic table,” which shows the relative
abundances of the Sun’s elements in terms of size, hydrogen and helium appear as the most
abundant. See Appendix 3 for a full Periodic Table of the Elements.
is at work in the rapidly rotating giant planets as well, caus- A planet’s obliquity—the inclination of its equatorial
ing the planets to bulge at their equators and giving them plane to its orbital plane—is a major factor in determin-
an overall flattened appearing the prominence of its seasons. (Recall from Chapter 2
ance. We call this flattening Rapid rotation that Earth’s obliquity of 23.5° is responsible for our distinct
oblateness, and it is defi ned flattens the shapes of seasons.) The obliquities of the giant planets are shown in
as the difference between a the giant planets. Table 9.1. With an obliquity of only 3°, Jupiter has almost
planet’s equatorial radius (RE) no seasons at all. The obliquities of Saturn and Neptune
and its polar radius (RP), divided by its equatorial radius: are slightly larger than those of Earth or Mars, giving these
planets moderate but well-defi ned seasons. Curiously, Ura-
RE - RP nus spins on an axis that lies nearly in the plane of its orbit.
Oblateness = ______
RE Viewed from Earth, then, the planet appears to be either
spinning face-on to us or rolling along on its side (or some-
Saturn’s oblateness is especially noticeable because its equa- thing in between), depending on where Uranus happens to
torial diameter is almost 10 percent greater than its polar be in its orbit.
diameter (see Figure 9.3 and Table 9.1). In comparison, the Uranus is one of five major Solar System bodies with very
oblateness of Earth is only 0.3 percent. high obliquities. Its obliquity is 98°, and a value greater than
90° indicates that the planet rotates in a clockwise direc-

tion when seen from above its orbital plane (see Chapter
2). Venus, Pluto, Pluto’s moon Charon (see Chapter 12), and
Neptune’s moon Triton (see Chapter 11) are the only other
major bodies that behave this way. The 98° obliquity of Ura-
nus causes its seasons to be extreme, with each polar region
alternately experiencing 42 years of continuous sunshine
followed by 42 years of total darkness. If we calculate the
average amount of sunlight absorbed over an entire orbit, it
turns out that the poles are warmed more than the equator—
a situation quite different from that of any other planet. Why
does Uranus have an obliquity so different from the other
planets? As we learned in Chapter 6, many astronomers
believe the planet was “knocked over” by the impact of a
huge planetesimal near the end of its accretion phase.
9.3 A View of the

Cloud Tops
Even when viewed through small telescopes, Jupiter is per- FIGURE 9.6 This Hubble Space Telescope image shows a
haps the most colorful of all the planets (see Figure 9.1a). A storm on Saturn that began shortly after the launch of the
dozen or more parallel bands, ranging in hue from bluish telescope in 1990.
gray to various shades of orange, reddish brown, and pink,
stretch out across its large, pale yellow disk. Traditionally,
astronomers call the darker bands “belts” and the lighter by any other celestial object. The rings are what make it so
ones “zones.” Many small clouds— some dark and some spectacular. Farther away and somewhat smaller, the planet
white, some circular and others more oval in shape— appear itself appears less than half as large as Jupiter, as illustrated
along the edges of, or within, the belts. The most prominent in Figure 9.3b. Like Jupiter, Saturn displays atmospheric
of these is a large, often brick-red feature in Jupiter’s south- bands, but the bands tend to be wider and their colors and
ern hemisphere known as the Great Red Spot, seen at the contrasts much more subdued than those on Jupiter. Indi-
lower right in Figure 9.1a. Oval in shape, with a length of vidual clouds on Saturn are seen only rarely from Earth.
25,000 km and a width of 12,000 km, the Great Red Spot On these infrequent occasions, large, white, cloudlike fea-
could comfortably hold two Earths side by side within its tures may suddenly erupt in the tropics, spread out in lon-
boundaries (see Figure 9.12b). gitude, and then fade away over a period of a few months
No one really knows how long this huge feature has been (Figure 9.6).
circulating in Jupiter’s atmosphere, but it was fi rst spotted In even the largest telescopes, Uranus and Neptune
more than three centuries ago, appear visually as tiny, featureless disks with a pale bluish
shortly after the invention of The Great Red Spot is green color (similar to the Voyager view in Figure 9.2a). As
the telescope. Since then, the a giant anticyclone. we will see, however, infrared imaging reveals a number
Great Red Spot has varied of individual clouds and belts, giving these distant planets
unpredictably in size, shape, color, and motion as it drifts considerably more character.
among Jupiter’s clouds. Small clouds seen moving around Observed from close up, the giant planets suddenly
the periphery of the Great Red Spot show that it circulates appear as real and tangible worlds. The clouds of Jupiter
like a giant vortex. Its cloud pattern looks a lot like that of are a landscape of variegated color and intricate forma-
a terrestrial hurricane, but it rotates in the opposite directions (see Figure 9.1a). Time-lapse imaging shows a roiling,
tion, exhibiting anticyclonic rather than cyclonic flow. (You swirling giant with atmospheric currents and vortices so
may recall from Chapter 8 that anticyclonic flow is indica- complex that we still do not fully understand the details
tive of a high-pressure system.) Because of its colorful of how they interact with one another, even after decades
appearance and unpredictable changes, the Great Red Spot of analysis. The Great Red Spot alone displays more struc-
has long been a favorite among amateur astronomers. ture than was visible over all of Jupiter prior to the space
Our fi rst look at Saturn through a telescope is a moment age. Dynamically, it also reveals some rather bizarre behav-
not quickly forgotten (see Figure 9.1b). Adorned by its mag- ior, such as “cloud cannibalism.” In a series of time-lapse
nificent system of rings, Saturn provides a sight unmatched images, Voyager observed a number of Alaska-sized clouds
9.3 A View of the Cloud Tops 255
1 A large cloud in 2 …gets caught in the

Jupiter’s atmosphere… Great Red Spot…
3 …and is swallowed
by the giant storm.
FIGURE 9.7 This sequence of images, obtained by the Voyager spacecraft during its encounter
with Jupiter, shows the swirling, anticyclonic motion of Jupiter’s Great Red Spot.
(a)
being swept into the Great Red Spot vortex. Some of these
clouds were carried around the vortex a few times and then
ejected, while others were swallowed up and never seen
again (Figure 9.7). Other smaller clouds with structure and
behavior similar to that of the Great Red Spot are seen in
Jupiter’s middle latitudes.
Broad bands and individual clouds show prominently
in Voyager and Cassini images of Saturn’s atmosphere. A
relatively narrow, meandering
(b)
band in the midnorthern lati- Saturn has jet streams
Saturn jet stream
tudes encircles the planet in a similar to Earth’s. L
L L
manner similar to our own
terrestrial jet stream (Figure 9.8). The largest features are
H H
about the size of the continental United States, but many
that we see are smaller than terrestrial hurricanes. Close-up
FIGURE 9.8 (a) Voyager image of a jet stream in Saturn’s
views from the Cassini spacecraft (Figure 9.9) show immense
northern hemisphere, similar to jet streams in our terrestrial
lightning-producing storms in a region of Saturn’s southern
atmosphere. (b) There is a dynamic relationship between the jet
hemisphere known to mission scientists as “storm alley.” stream and vortices of low and high pressure. The jet stream
Uranus and Neptune present a rather bland appearance dips equatorward below regions of low pressure and is forced
at visible wavelengths. Yet in the near infrared, beyond the poleward above regions of high pressure.
spectral range of our eyes, these planets take on a very dif-
ferent appearance, showing atmospheric bands and small
clouds suggestive of those seen on Jupiter and Saturn, but observations made in different spectral regions can add sig-
more subdued (Figure 9.10). The atmospheres of Uranus nificantly to our understanding of astronomical objects.
and Neptune are rich in methane, and methane is a strong A number of bright cloud bands appear in HST images
absorber of infrared light. The strong absorption of reflected of Neptune’s atmosphere (Figure 9.10b). Located near the
sunlight by methane causes the atmospheres of Uranus and planet’s tropopause (see Chapter 8), these cloud bands cast
Neptune to appear dark in the near infrared, allowing the their shadows downward through the clear upper atmo-
highest clouds and bands to stand out in contrast against sphere onto a dense cloud layer 50 km below (Figure 9.11).
the dark background. Here we fi nd an illustration of how Some of Neptune’s clouds are dark. A large, dark, oval
(a)
(b)
FIGURE 9.9 Violent storms are known to erupt on Saturn. (a) An enhanced Cassini image of an
intense lightning-producing storm (left of center) located in Saturn’s “storm alley.” The inset shows a
similar storm on Saturn’s night side illuminated by sunlight reflecting off Saturn’s rings. (b) A near-
infrared Cassini image showing a storm nicknamed “the dragon.”
(a) (b)
FIGURE 9.10 Hubble Space Telescope images of Uranus (a) and Neptune (b), taken at a wavelength of light
that is strongly absorbed by methane. The visible clouds are high in the atmosphere. The rings of Uranus show
prominently because the brightness of the planet has been subdued by methane absorption. Seasonal changes
in cloud formation on Neptune are evident over a 6-year interval.
9.4 A Journey into the Clouds 257
9.4 A Journey into

the Clouds
Our visual impression of the giant planets is based on our
two-dimensional view of their cloud tops. Atmospheres,
though, have depth. They are three-dimensional structures
whose temperature, density, pressure, and even chemical
composition vary with height and over horizontal distances.
As a rule, atmospheric temperature, density, and pressure
all decrease with increasing altitude, although temperature
will sometimes reverse itself at very high altitudes. The
stratospheres (see Chapter 8) above the cloud tops appear
relatively clear, but closer inspection shows that they contain
layers of thin haze that show up best when seen in profile
above the limbs of the planets. The composition of the haze
particles remains unknown, but we believe that they are
photochemical—smoglike—products created when ultravio-
let sunlight acts on hydrocarbon gases such as methane.
FIGURE 9.11 Neptune’s clouds cast shadows through a clear What lies beneath the cloud tops of the giant planets?
atmosphere onto the cloud deck 50 km below in this Voyager 2 Imagine riding along in the Galileo probe as it descends
image. through Jupiter’s atmosphere.
At fi rst the only change we Different volatiles
note is the expected increase produce
feature in the southern hemisphere (Figure 9.12a) is remi- in the outside pressure and different clouds
niscent of Jupiter’s Great Red Spot (Figure 9.12b). Predict- temperature as our altitude at different heights.
ably, astronomers called it the Great Dark Spot. However, rapidly decreases. Suddenly
the Neptune feature was gray rather than red, changed its we fi nd ourselves passing through dense layers of cloud,
length and shape more rapidly than the Great Red Spot does, separated by regions of relatively clear atmosphere. Each of
and it lacked the permanence of the Great Red Spot. When these cloud layers is composed of a different chemical sub-
the HST trained its corrected optics on Neptune in 1994, stance. In Earth’s atmosphere water is the only substance
the Great Dark Spot had disappeared. It has not been seen that can condense into clouds, but the atmospheres of Jupi-
since. However, a different dark spot of comparable size had ter and the other giant planets contain a variety of volatiles
appeared in Neptune’s northern hemisphere. that condense at different temperatures and atmospheric
FIGURE 9.12 The Great Dark Spot on Neptune (a) and the Great Red Spot on Jupiter (b) reproduced
approximately to scale. Earth is shown for comparison. The Great Red Spot has persisted for centuries,
but the Great Dark Spot disappeared between the time Voyager flew by Neptune in 1989 and the time
HST images were obtained in 1994.
(a) Neptune’s Great Dark Spot Earth (b) Jupiter’s Great Red Spot
258 (a) Jupiter
Chapter 9 Worlds of Gas and Liquid—The Giant Planets (b) Saturn
100 200
0.01
50 Stratosphere 0.01 100 Stratosphere
0.1 0 0.1
0 Haze
Pressure (atm)
Pressure (atm)
Haze
Altitude (km)
Altitude (km)
Troposphere Troposphere
Ammonia ice
–50 1 –100 Ammonia ice

1
Ammonium 2
hydrosulfide ice
Water ice
Water/ammonia droplets
–100 –200 Ammonium
10 hydrosulfide ice
10
Gaseous hydrogen, helium,
methane, ammonia, and water
Water ice
Water/ammonia droplets
–150 –300
0 100 200 300 400 500 0 100 200 300 400 500
(c) Uranus (d) Neptune

100 100
0.01
0.01
50 Stratosphere 50 Stratosphere
Methane ice Methane ice

0 1 0 1
Pressure (atm)
Pressure (atm)
2
Altitude (km)
Altitude (km)
2
Hydrogen sulfide ice Ammonia ice Hydrogen sulfide ice
5
5 10
–50 –50
10 Ammonium hydrosulfide ice 20
Ammonium Water ice
20 50
hydrosulfide ice
Water ice
–100 –100 Water, ammonia, 100
50 hydrogen sulfide
Water, ammonia, droplets 200
hydrogen sulfide
droplets 100
500
200
–150 –150
0 100 200 300 400 500 0 100 200 300 400 500
FIGURE 9.13 Volatile materials condense at different levels in the atmospheres of the giant planets, leading to
chemically different types of clouds at different depths in the atmosphere. The red line in each diagram shows how
atmospheric temperature changes with height. The arbitrary zero points of altitude are at 0.1 atmosphere (atm)
for Jupiter (a) and Saturn (b) and 1.0 atm for Uranus (c) and Neptune (d). The value of 1.0 atm corresponds to the
atmospheric pressure at sea level on Earth (approximately 1 bar). Note that Saturn’s vertical scale is compressed to
show the layered structure better.
pressures (Figure 9.13). A descent through Jupiter’s cloud tochemical products produced close to the ground are rather
layers thus becomes a journey that explores the many minor obvious; we call them “smog.”
constituents of its atmosphere. Each kind of volatile, such The upper tropospheres of Uranus and Neptune, unlike
as water, condenses at a particular temperature and pres- those of Jupiter and Saturn, are relatively clear, with only a
sure, and each therefore forms clouds at a different altitude. few white clouds—probably methane ice— appearing here
Below the condensation cloud layer, each volatile is freely and there. Uranus and Neptune are not bluish green because
mixed as a gaseous atmospheric constituent. Above, it is of clouds. Instead they take on a bluish color for much the
highly depleted. We can see the reason for this difference: same reason that Earth’s oceans are blue. Methane gas is
As the planet’s atmosphere convects (a process explained much more abundant in the atmospheres of Uranus and
in Chapter 7 and discussed further in Chapter 8), volatile Neptune than in Jupiter and Saturn. Like water, methane gas
materials are carried upward along with all other atmo- tends to selectively absorb the longer wavelengths of light—
spheric gases. When a particular volatile reaches an altitude yellow, orange, and red. Absorption of the longer wave-
where the temperature is low enough, the condensation lengths leaves only the shorter wavelengths— green and
process removes most of it from the other gases, leaving it blue—to be scattered from the relatively cloud-free atmo-
depleted in the air above. spheres of Uranus and Neptune, giving them a characteristic
During our descent we find that ammonia has condensed bluish green color. Earlier we described the atmosphere of
near the top of Jupiter’s troposphere (see Chapter 8) at a tem- Uranus as appearing nearly “bottomless.” Molecular scat-
perature of about -140°C (133 K). Next we pass through a tering also contributes to the bluish color and is so strong
layer of ammonium hydrosulfide clouds at a temperature of in the clear Uranus atmosphere that it completely hides the
about -80°C (193 K). Not long after this, information from thick ammonia and water cloud layers that lie far below.
the probe ends. (While descending slowly via parachute to
an atmospheric pressure of 22 bars and a temperature of
about 100°C [373 K], the Galileo probe failed, presumably
because its transmitter got too hot.) What lies below this
level in Jupiter’s atmosphere must for now be left to our
9.5 Winds and Storms—
theories and atmospheric models. Violent Weather on
The distance of each planet from the Sun partly deter-
mines its tropospheric temperature. The farther the planet the Giant Planets
is from the Sun, the colder its troposphere will be. Distance
from the Sun thus determines the altitude at which a par- The rapid rotation and resulting strong Coriolis effects (see
ticular volatile, such as ammonia or water, will condense to Chapter 8) in the atmospheres of the giant planets create
form a cloud layer on each of the planets (see Figure 9.13). If much stronger zonal winds than we see in the atmospheres
temperatures are too high, some volatiles may not condense of the terrestrial planets, even though less thermal energy is
at all. The highest clouds in the frigid atmospheres of Ura- available. Figure 9.14 shows the zonal wind patterns on the
nus and Neptune are crystals of methane ice. The highest giant planets. On Jupiter the strongest winds are equatorial
clouds on Jupiter and Saturn are made up of ammonia ice. westerlies, which have been clocked at speeds of 550 kilo-
Methane exists only in gaseous form throughout the warmer meters per hour (km/h), as seen in Figure 9.14a. (Remember
atmospheres of Jupiter and Saturn. from Chapter 8 that westerly winds are those that blow from,
In their purest form, the ices that make up the clouds of not toward, the west.) At higher latitudes the winds alternate
the giant planets are all white, similar to snow on our own between easterly and westerly in a pattern that seems to be
planet. Why, then, are some related to Jupiter’s banded structure, but scientists are not
clouds so colorful, especially Clouds on Jupiter sure. Near a latitude of 20° south, the Great Red Spot appears
Jupiter’s? These tints and hues and Saturn are to be caught between a pair of easterly and westerly cur-
must come from impurities in colored by impurities. rents with opposing speeds of more than 200 km/h. If you
the ice crystals, similar to the think this fact might imply something about the relationship
way syrups color snow cones. Although the identities of between zonal flow and vortices, you are right.
these impurities remain unknown, our prime suspects The equatorial winds on Saturn are also westerly, but
include elemental sulfur and phosphorus and various stronger than those on Jupiter. In the early 1980s, Voyager
organic materials produced by the photochemical action of measured speeds as high as
sunlight on atmospheric hydrocarbons. Ultraviolet photons 1,690 km/h. Later, HST Equatorial winds are
from the Sun, absorbed by molecules of hydrocarbons such recorded maximum speeds of fast on Jupiter and
as methane, acetylene, and ethane, among others, have 990 km/h, and more recently even faster on Saturn
enough energy to break these molecules apart. The molecu- Cassini has found them to be and Neptune.
lar fragments can then recombine to form complex organic intermediate between the Voy-
compounds that condense into solid particles, many of ager and HST measurements. What can be happening here?
which are quite colorful. Photochemical reactions also As Math Tools 9.1 explains, winds on other planets are mea-
occur in our own terrestrial atmosphere. Some of the pho- sured by the motions of their clouds. Saturn’s winds appear
How Wind Speeds on Distant Planets Are Measured

How do we measure wind speeds on planets that are so far away? spot is then seen at longitude 153.0° west exactly 24 hours later.
It turns out to be surprisingly easy. If we can see individual clouds Neptune’s equatorial winds have carried the white spot 80.0°
in their atmospheres, we can measure their winds. As on Earth, in longitude in 24 hours (86,400 seconds).
clouds are typically carried right along with the local winds. By The equatorial radius of Neptune is 24,760 km. Multiplying
measuring the positions of individual clouds and noting how the radius by 2π gives the planet’s circumference (360° of lon-
much they move during an interval of a day or so, we are able to gitude) as 155,570 km. Thus, each degree of longitude on Nep-
calculate the local wind speed. We need to know one important tune’s equator is equal to 155,570/360 (approximately 432) km.
additional piece of information, though: how fast the planet itself So, Neptune’s equatorial winds have carried the spot 432 × 80
is rotating. The reason is that we want to measure the speed of = 34,560 km (3.456 × 107 meters) in one day (86,400 seconds).
the winds with respect to the planet’s rotating surface. Therefore, the wind speed is 400 meters per second (m/s), the
In the case of the giant planets, of course, there is no solid sur- fastest observed zonal wind speed in the Solar System. So we
face against which to measure the winds. We must instead assume know the equatorial winds are very strong, but in what direc-
a hypothetical surface— one that rotates as though it were some- tion are they blowing? To an external observer, Neptune’s west
how “connected” to the planet’s deep interior. How can we know longitude is seen to increase with time. In other words, west
how rapidly the invisible interior of a planet is rotating? Periodic longitude increases in a direction opposite the direction in
bursts of radio energy caused by the rotation of a planet’s magnetic which the planet is rotating. The fact that the white spot’s west
field tell us how fast the interior of the planet is rotating. longitude is increasing with time means that Neptune’s equa-
Let’s see an example of how this works using a small white torial winds are blowing in a direction opposite the planet’s
cloud in Neptune’s atmosphere. The cloud, on Neptune’s equa- rotation. As we learned in Chapter 8, such winds are called
tor, is observed to be at longitude 73.0° west on a given day. (This “easterly winds.” On Earth, their much slower equivalents are
longitude system is anchored in the planet’s deep interior.) The called “trade winds.”
to decrease with height, so the apparent time variability of Our knowledge of global winds on Uranus (Figure 9.14c)
Saturn’s equatorial winds may be nothing more than changes is poorer than that of the other giant planets because of the
in the height of the equatorial cloud tops. Alternating east- relatively few clouds we have been able to see and track (see
erly and westerly winds also occur at higher latitudes; but Math Tools 9.1). When Voyager 2 flew by Uranus in 1986,
unlike Jupiter’s case, this alternation seems to bear no clear the few clouds we saw were all in the southern hemisphere
association with Saturn’s atmospheric bands (Figure 9.14b). because the northern hemisphere was in complete darkness
This behavior is but one example of unexplained differences at the time. The strongest winds observed were 650-km/h
among the giant planets. westerlies in the middle to high southern latitudes, and no
As mentioned earlier, Saturn’s jet stream is a narrow easterly winds were detected on the part of the planet that
meandering river of air with alternating crests and troughs could be seen. More recently, as Uranus approached equinox
(see Figure 9.8a). This feature, located near latitude 45° season (see Chapter 2), terrestrial observers have reported
north, is similar to our own terrestrial jet streams, where wind speeds of up to 900 km/h.
high-speed winds blow generally from west to east but Because its peculiar orientation gives Uranus a “reversed”
with alternate wanderings toward and away from the temperature pattern wherein the poles are warmer than the
pole. Nested within the crests and troughs of Saturn’s jet equator, some astronomers had thought earlier that Uranus
stream are anticyclonic and cyclonic vortices. They appear might have a global wind system very different from that of
remarkably similar in both form and size to terrestrial the other giant planets. Even though the Sun was shining
high- and low-pressure systems, which bring us alternat- almost straight down on the south pole of Uranus when Voy-
ing periods of fair and stormy weather as they are carried ager 2 observed it, the dominant winds on Uranus turned out
along by our terrestrial jet streams. This similarity in jet to be zonal, just as they are on the other giant planets. The
streams on Saturn and Earth is a good illustration of the Coriolis effect is thus more influential than individual atmo-
many reasons we study other planets—for the ability to spheric temperature patterns in determining the fundamental
compare them to each other and to similar phenomena on structure of the global winds on all the giant planets.
Earth. From observing and analyzing similar atmospheric As Uranus has moved along in its orbit over the course of
systems on other planets, we often learn more about the our recent observations, regions previously unseen by mod-
way our own weather works. ern telescopes have become visible (Figure 9.15). In December
(a) Jupiter (b) Saturn (c) Uranus (d) Neptune
Wind speeds on Jupiter Equatorial winds on Saturn Even with its peculiar The strongest winds on
reach about 550 km/h. reach an extraordinary orientation, Uranus exhibits Neptune are easterly.
1,650 km/h, or almost half zonal winds.
a kilometer per second.
North pole North pole North pole North pole
Equator Equator Equator Equator
–100 –50 0 50 100 150 0 200 400 –200 0 200 400 –700 –350 0 350
Velocity (m/s) Velocity (m/s) Velocity (m/s) Velocity (m/s)
FIGURE 9.14 Strong winds blow the atmospheres of the outer planets, driven by powerful
convection and the Coriolis effect on these rapidly rotating worlds.
2007, Uranus arrived at equinox (see Chapter 2) and spring of about 1 week. On a rapidly rotating planet, winds gener-
had come to its northern hemisphere. Observations by HST ated by Coriolis effects can be very strong. Clouds circulat-
and groundbased telescopes showed bright cloud bands in ing around the circumference of the Great Red Spot have
the far north extending over 18,000 km in length and revealed been clocked at speeds of up to 1,000 km/h. Such weather
wind speeds of up to 900 km/h. As we approach northern systems dwarf our terrestrial storms in both size and inten-
summer solstice in the year 2027, we can expect to learn much sity. Comparable whirlpool-like behavior is observed in
more about the “unseen northern hemisphere” of Uranus. many of the smaller oval-shaped clouds found elsewhere in
As expected, the strongest winds on Neptune occur in Jupiter’s atmosphere, as well as in similar clouds observed
the tropics (Figure 9.14d). The surprise was that they are in the atmospheres of Saturn and Neptune.
easterly rather than westerly, with speeds in excess of 2,000
km/h. Westerly winds with speeds of 900 km/h and higher
FIGURE 9.15 Uranus is approaching equinox in this 2006 HST
were seen in the south polar regions. With wind speeds five image. Much of its northern hemisphere is becoming visible. The
times greater than those of the fiercest hurricanes on Earth, dark spot in the northern hemisphere (to the right), is similar to
Neptune and Saturn are the windiest planets we know of. but smaller than the Great Dark Spot seen on Neptune in 1989.
Summer solstice arrived in Neptune’s southern hemisphere
in 2005, so much of the north is still in darkness. We’ll have
to wait a while before we can get a good look at its northern
hemisphere. Each season on Neptune lasts 40 Earth years!
Chapter 8 taught us that vertical temperature differences
can create the localized type of atmospheric motion called
convection. On the giant plan-
ets the thermal energy that Thermal energy drives
drives convection comes both powerful convection
from the Sun and from the hot on the giant planets.
interiors of the planets them-
selves. As heating drives air up and down, the Coriolis effect
shapes that convection into atmospheric vortices, examples
of which are familiar to us on Earth as high-pressure sys-
tems, hurricanes, and thunderstorms. On the giant planets,
convective vortices are visible as isolated circular or oval
cloud structures. The Great Red Spot on Jupiter and the Great
Dark Spot on Neptune are classic examples. Observations of
small clouds distributed within the Great Red Spot show
that it is an enormous atmospheric whirlpool or eddy, swirl-
ing around in a counterclockwise direction with a period
FIGURE 9.16 Convection can

2 Material flowing away from 3 Together these
cause atmospheric gases to
regions of convective upwelling is convective vortices
deflected into powerful vortices by drive rapid zonal winds. rise, creating a region of high
the strong Coriolis effect on these pressure. The Coriolis effect
rapidly rotating worlds. then produces anticyclonic
flow around the high pressure
region. Such convective vortices
drive zonal winds on the giant
planets.
H H
1 Thermal energy from

within some giant planets
drives strong convection.
As the atmosphere ascends near the centers of the vorti- pairs of strong zonal winds flowing in opposite directions,
ces, it expands and cools. As we learned in Chapter 8, cool- as was Neptune’s Great Dark Spot. Is this by chance? Similar
ing condenses certain volatile materials into liquid droplets, relationships observed between other isolated vortices and
which then fall as rain. As they fall, the raindrops collide the zonal wind flow suggest that it is not. The enormous
with surrounding air molecules, stripping electrons from the wind energy developed within numerous vortices seems to
molecules and thereby developing tiny electric charges in drive the alternating easterly and westerly zonal winds that
the air. The cumulative effect of countless falling raindrops characterize the global circulation of the giant planets (Fig-
can generate an electric charge and resulting electric field so ure 9.16).
great that they break down a conductive path through the
atmosphere, creating a surge of current and a flash of light-
ning. A single observation of Jupiter’s night side by Voyager 1
revealed several dozen lightning bolts within an interval of
3 minutes. We estimate the strength of these bolts to be equal
9.6 Internal Thermal
to or greater than the “superbolts” that occur in the tops of Energy Drives
high convective clouds in the terrestrial tropics. Although
various constraints have prevented an imaging search for Weather Patterns
lightning on the other giant planets, radio receivers on Voy-
ager picked up lightning static on all of them. We saw in Chapter 8 how the uneven heating of a planet,
How are these atmospheric vortices connected to the along with the planet’s rotation, drives global atmospheric
global circulation of the giant planets? From a careful study circulation, and how temperature-related differences in
of the interaction between vor- pressure from place to place drive winds. In fact, virtually
tices and zonal winds, scien- Vortices created by all weather on every planet is driven by the interplays of
tists believe that the massive convection drive the thermal energy within the planet’s atmosphere. On Earth
vortices actually drive zonal strong zonal winds of and the other terrestrial planets, the source of this ther-
wind currents. We have noted the giant planets. mal energy is as clear as the Sun shining on a summer sky.
that the narrow zonal jet on Sunlight powers our climate. The Sun, however is not the
Saturn moves with a wavelike motion (see Figure 9.8a). primary source of heat for the giant planets.
Nestled in each of its crests and troughs are clockwise (anti- This is not a new insight. In Chapter 4 we learned about
cyclonic) and counterclockwise (cyclonic) features, strongly the equilibrium between the absorption of sunlight and the
suggesting a dynamic relationship between these systems radiation of infrared light into space, and in Chapter 8 we
and the zonal jet. Jupiter’s Great Red Spot is situated between saw how the resulting equilibrium temperature is modified
by the greenhouse effect on Venus, Earth, and Mars. Yet of gas collapses under the force of its own self-gravity (see
when we calculate this equilibrium for the giant planets, Chapter 10), it converts its gravitational potential energy to
as we did in Chapter 4, we thermal energy. The gaseous planets continue to contract
fi nd that something seems Jupiter has a large indefinitely, releasing their gravitational potential energy as
amiss. According to these cal- internal heat source, they shrink. This is the primary energy source for replacing
culations, the equilibrium as do Saturn and the internal energy that leaks out of the interior of Jupiter
temperature for Jupiter, for Neptune. (see Excursions 9.1) and is probably an important source
example, should be 109 K, but for the other giant planets as well.
when it is measured we find instead an average temperature The giant planets are still contracting and converting their
of about 124 K. A difference of 15 K might not seem like gravitational potential energy into thermal energy today as
much, but remember that according to the Stefan-Boltzmann they did when they first formed, but they are doing it more
law (see Chapter 4), the energy radiated by an object depends slowly. The annual amount of contraction necessary to sus-
on its temperature raised to the fourth power. Applying this tain their internal temperature is only a tiny fraction of their
relationship to Jupiter, we get radius. For Jupiter, this is only 1 millimeter (mm) or so— a
hundred-billionth of its radius—per year. If this same rate
(TT ) = (124
109 K)
K4 4
were to continue for the next billion years, Jupiter would
⎯⎯⎯⎯
true
= 1.67.
⎯⎯⎯
expected shrink by only 1,000 km, a little more than 1 percent of its
radius. In Saturn’s case, and perhaps Jupiter’s as well, there is
The implications of this result are somewhat startling: Jupi- an additional source of internal energy. Under the right condi-
ter is radiating roughly two-thirds more energy into space tions, liquid helium separates from a hydrogen-helium mix-
than it absorbs in the form of sunlight! Thus, almost half ture and “rains” downward toward the core. As the droplets
of the thermal energy powering Jupiter’s weather comes of liquid helium sink, they release their gravitational poten-
from somewhere other than the Sun. Similarly, the inter- tial energy as thermal energy. Planetary physicists believe
nal energy escaping from Saturn is observed to be about that most of Saturn’s internal energy and perhaps some of
1.8 times greater than the sunlight that it absorbs. Neptune Jupiter’s come from this separation of liquid helium.
emits 2.6 times as much energy as it absorbs from the Sun.
Strangely, whatever internal energy may be escaping from
Uranus is small compared with the absorbed solar energy.
With energy continually escaping from the interiors
of the giant planets, we can easily wonder how they have
9.7 The Interiors of the
maintained their high internal temperatures over the past Giant Planets Are
41⁄2 billion years. The short answer is that they are still
shrinking in size, converting gravitational potential energy Hot and Dense
into thermal energy. This continual production of thermal
energy is sufficient to replace the energy that is escaping At depths of a few thousand kilometers, the atmospheric
from their interiors. gases of Jupiter and Saturn are so compressed by the weight
We can think of the internal energy that lies deep within of the overlying atmosphere that they liquefy. When the
the giant planets as being primordial. In other words, it is interior pressure climbs to 2 megabars and the temperature
left over from their creation. Though they have much smaller reaches 10,000 K, hydrogen molecules are battered so vio-
masses than the Sun, the giant planets formed much like lently that their electrons are stripped free, and the hydrogen
collapsing protostars (see Chapter 6). Recall that as a mass becomes an electrical conductor like a liquid metal. This
Jupiter—A Star That Failed?

In the popular literature Jupiter is frequently and erroneously the temperatures necessary to initiate the self-sustaining ther-
labeled as “a star that failed”—probably because it is gaseous, monuclear reactions that occur in stars are in the tens of mil-
has a composition similar to the Sun, has a high internal tem- lions of degrees. To achieve such temperatures, Jupiter would
perature, and releases a large amount of primordial energy. require a total mass 80 times greater than it actually has. In
Although these statements are true, to imply that it is “almost other words, the least massive stars must still have masses that
a star” is most certainly an exaggeration. Jupiter’s central tem- are approximately 80 times greater than Jupiter’s. Jupiter is
perature is probably not much greater than 35,000 K, whereas merely a large planet and never came close to being a star.
happens at a depth of about 20,000 km in Jupiter’s atmosphere was already there in the original bodies that captured
and 30,000 km in Saturn’s. Uranus and Neptune are less hydrogen and helium from the Sun’s protoplanetary disk to
massive than Jupiter and Saturn, have lower interior pres- form what ultimately became giant planets. Differentiation,
sures, and contain a smaller however, has occurred and is still occurring in Saturn, and
f raction of hyd rogen— Hydrogen-helium perhaps Jupiter too. On Saturn, helium condenses out of the
conditions that do not favor oceans lie at depths hydrogen-helium oceans. (Helium can also be compressed
the formation of metallic of a few thousand to a metal, but it does not reach this metallic state under
hydrogen. Thus, their interiors kilometers on Jupiter the physical conditions existing in the interiors of the giant
probably contain only a small and Saturn. planets.) Because these droplets of helium are denser than
amount of liquid hydrogen, the hydrogen-helium liquid in which they condense, they
with little or none of it in a metallic state. This transition sink toward the center of the planet. This process tends to
from a gas to a liquid is so subtle as to be hardly noticeable. enrich helium concentration in the core while depleting
To put it another way, the physical difference between a liq- it in the upper layers. At the same time the process heats
uid and a highly compressed, very dense gas is something the planet by converting gravitational potential energy to
that could be appreciated only by—well— a physicist. Thus, thermal energy. In Jupiter’s interior, by contrast, the liquid
unlike the well-defined surface between Earth’s atmosphere helium is mostly dissolved together with the liquid hydro-
and its oceans, on Jupiter and Saturn there is no clear bound- gen. Within Saturn’s interior, temperatures are lower than
ary between the atmosphere and the “ocean” of liquid hydro- on Jupiter, making the helium less soluble. Those who cook
gen and helium that lies below. The depths of these know that you can easily dissolve large amounts of sugar
hydrogen-helium oceans are measured in tens of thousands in hot water but relatively little when the water is cold. So
of kilometers, making them the largest structures within the it is with helium and hydrogen.
interiors of any of the giant planets. Uranus and Neptune do In the cores of the gas giants, temperatures may be in
not have these oceans of liquid hydrogen and helium. the tens of thousands of degrees, with pressures of tens of
Figure 9.17 shows the interior structure of the giant megabars. As we learned in
planets. At the center of each giant planet is a dense, liquid Chapter 8, a pressure of 1 bar Jupiter’s core is liquid
core consisting of a very hot mixture of heavier materials corresponds closely to Earth’s water and rock at
such as water, rock, and metals. In Chapter 7 we learned atmospheric pressure at sea a temperature of
that denser materials, such as iron and other metals, sank to level. A megabar is a million about 35,000 K.
the centers of the terrestrial planets when they were in an times as great as 1 bar. For
earlier, more molten state. This process, which we call dif- comparison, when the submersible research vessel Alvin
ferentiation, deposited most of the metals in what became cruises Earth’s ocean bottoms 10,000 meters below the sur-
the cores of the terrestrial planets, leaving their mantles face, it experiences a surrounding pressure of about 1,000
and crusts relatively depleted of metals. bars, or 1/1,000 of a megabar. It may seem strange to be talk-
The cores of the giant planets did not form in the same ing about water that is still liquid at temperatures of tens of
manner, however. Much of the material now in their cores thousands of degrees, but there is really nothing peculiar
FIGURE 9.17 The central cores and outer liquid shells of the interiors of the giant planets. Only Jupiter and Saturn have significant
amounts of the molecular and metallic forms of liquid hydrogen surrounding their cores
Earth
0 km 50,000
Uranus Neptune
KEY
Molecular hydrogen
Metallic hydrogen
Jupiter Saturn “Ices”
Rock
E XCURSIONS 9. 2
Strange Behavior in the Realm of Ultrahigh Pressure

You might naturally assume that gaseous, Jovian-type planets compress the interior further. The decreased volume caused
with mass greater than Jupiter’s should also be larger, but to do by increased pressure more than makes up for the increased
so would be wrong. Jupiter’s mass is well more than three times volume of the additional hydrogen.
as great as Saturn’s, even though both have approximately the It turns out that, by chance, Jupiter is almost the largest that
same composition and are nearly the same size. This apparent any planet can be in this or any other solar system. Planets that
anomaly occurs because Jupiter’s massive self-gravity— caused are either less massive or more massive will be smaller! Does
by its greater mass— compresses its interior more than Saturn’s this mean that stars should also be smaller than Jupiter? They
can. For planets the size of Jupiter and Saturn, adding more would be except for one important difference. Stars have nuclear
hydrogen does not make them much bigger; it makes them reactions going on within their cores and thus have very much
denser instead. We now fi nd ourselves with a curious situation higher internal temperatures than any planet has. These ultra-
that goes against our intuition: if we were to add much more high temperatures create enormous internal pressures that better
hydrogen to Jupiter, its diameter would actually get smaller resist gravitational compression, keeping even the smallest “nor-
rather than larger. A planet with a mass, say, 10 times greater mal” stars much larger than the largest planets. We say “normal”
than Jupiter would be about 10 percent smaller. The additional stars because later, in Chapters 16 and 17, we will introduce you
overlying mass creates higher internal pressures, which in turn to dying stars that are even smaller than Earth.
about this. Like a super pressure cooker, the extremely high density of water, even less than that of Saturn. Their lower
pressures at the centers of the giant planets prevent the mass would not be as effective as Saturn’s in compressing
water from turning to steam. The temperature at Jupiter’s their hydrogen and helium. But such low average densities
center is thought to be as high as 35,000 K, and the pressure are not what we observe. Instead we fi nd Uranus and Nep-
may reach 45 megabars. Central temperatures and pressures tune to be about twice as dense as Saturn (see Table 9.1). We
of the other, less massive giant planets are correspondingly now have a clear indication that, unlike Jupiter and Saturn,
lower than those of Jupiter. Uranus and Neptune must have denser material dominat-
ing their chemical composition. Is this denser material water
or rock? Our observations should be able to tell us. Neptune,
Uranus and Neptune Are Different the densest of the giant planets, is about 11⁄2 times as dense
as uncompressed water and only about half as dense as
from Jupiter and Saturn uncompressed rock. Uranus is even less dense than Nep-
You might suppose that the average densities of the giant tune. These observations tell us that both Uranus and Nep-
planets would tell us how much heavy material they con- tune must contain more water than rock. But we must also
tain. In practice, it is not so simple. Although the actual keep in mind that the high pressures within the interiors
numbers are very uncertain, the heavy-element components of Uranus and Neptune cause both water and rock to be
of the cores of Jupiter and Saturn have masses of about 5–10 more dense than in their uncompressed states. Thus, water
M⊕. As Table 9.1 shows, Jupiter and Saturn have total masses and other low-density ices, such as ammonia and methane,
of 318 and 95 M⊕, respectively. The heavy materials in their must be the major compositional components of Uranus and
cores, then, contribute little to their average chemical com- Neptune, with lesser amounts of silicates and metals. The
position. This means we can think of both Jupiter and Saturn total amount of hydrogen and helium in these planets is
as having approximately the same composition as the Sun probably limited to no more than 1 or 2 M⊕, with most of
and the rest of the universe: about 98 percent hydrogen and these gases residing in their relatively shallow atmospheres.
helium and only 2 percent of everything else. Why, then, On the basis of density alone, as we suggested at the begin-
with nearly identical compositions, should Jupiter’s density ning of this chapter, neither Uranus nor Neptune very well
be nearly twice as great as Saturn’s? In Excursions 9.2 we fit the description of a gas giant. It is more appropriate to
fi nd that the internal pressure created by Jupiter’s much refer to them as ice giants.
greater mass compresses its hydrogen and helium and its The water that makes up
core to a higher average density than in the core of Saturn. so much of Uranus and Nep- Uranus and Neptune
What does all of this imply about Uranus and Neptune? tune is probably in the form of are ice giants with
If Uranus and Neptune were also of solar composition— deep oceans. Dissolved gases deep, salty oceans.
that is, if they were made primarily of hydrogen and and salts would make these
helium—their average densities would be less than half the oceans electrically conducting. All of the giant planets have
deep layers of a conducting fluid—metallic hydrogen in the opposite to that of Earth as defi ned by where the north end
case of Jupiter and Saturn, and a saltwater brine in the case of a compass would point. The total strength of Jupiter’s
of Uranus and Neptune. Currents in these conducting layers magnetic field is nearly 20,000
are likely the source of the giant planets’ intense magnetic times that of Earth’s. On the Jupiter’s magnetic
fields, which we will discuss in Section 9.8. other hand, Jupiter is huge field is 20,000 times
compared to Earth. By the as strong as Earth’s.
time Jupiter’s magnetic field
emerges from the cloud tops, the field has dropped to about
Differences Are Clues to Origins 4.3 gauss— only 15 times Earth’s surface field. The strength
That each of the giant planets formed around rocky-metal of a magnetic field is often measured in gauss, named for
cores but turned out so differently is an important clue the German mathematician and scientist Carl Friedrich
to their origins. Why do Jupiter and Saturn have so much Gauss (1777–1855).
hydrogen and helium compared with Uranus and Neptune? The bar magnet used to simulate Saturn’s magnetic
Why is hydrogen-rich Jupiter so much more massive than field is located almost precisely at the center of Saturn and
hydrogen-rich Saturn? The answers may lie both in the is almost perfectly aligned with the planet’s rotation axis
time that it took for these planets to form and in the distri- (Figure 9.18b). Saturn’s magnetic field is much weaker than
bution of material from which they formed. We think that Jupiter’s, but overall it is still more than 500 times stronger
all of the hydrogen and helium in the giant planets was than Earth’s. Because Saturn’s diameter is much greater than
captured from the protoplanetary disk by the strong gravi- Earth’s, the magnetic-field strength at its cloud tops is about
tational attraction of their massive cores.2 The much lower 0.5 gauss— similar to the strength at Earth’s surface. As on
hydrogen-helium content of Uranus and Neptune suggests Jupiter, a compass would point south on Saturn.
that their cores were smaller and formed much later than Voyager 2 found that the magnetic-field axis of Uranus is
those of Jupiter and Saturn, at a time when most of the gas in inclined nearly 60° to its rotation axis, and its center is dis-
the protoplanetary disk had been blown away by the emerg- placed by a third of a radius from the planet’s center (Figure
ing Sun. Why should the cores of Uranus and Neptune have 9.18c). Considering the strange spin orientation of Uranus,
formed so late? Probably because the icy planetesimals from this observation did not come as any great surprise to the
which they formed were more widely dispersed at their Voyager scientists. The total strength of the field averages 50
greater distances from the Sun. With more space between times Earth’s field, but the large displacement of Uranus’s
planetesimals, it would take longer to build up their cores. field from the planet’s center causes the field at the cloud
Saturn may have captured less gas than Jupiter both because tops to vary between 0.1 and 1.1 gauss.
its core formed somewhat later and because less gas was The really big surprise came when Voyager 2 reached
available at its greater distance from the Sun. Neptune. The orientation of Neptune’s rotation axis is
similar to that of Earth, Mars, and Saturn. But Neptune’s
magnetic-field axis is inclined 47° to its rotation axis, and
the center of this magnetic field is displaced from the plan-
9.8 The Giant Planets Are et’s center by more than half the radius— an offset even
greater than that of Uranus (Figure 9.18d). The displace-
Magnetic Powerhouses ment of the field is primarily toward Neptune’s southern
hemisphere, thereby creating a large asymmetry in the field
All of the giant planets have magnetic fields that are much strength at the cloud tops, with 1.2 gauss in the southern
stronger than Earth’s. Planetary magnetic fields are produced hemisphere and only 0.06 gauss in the north. The overall
by the motions of electrically conducting liquids deep within strength of Neptune’s magnetic field is only half as great
planet interiors. In Jupiter and Saturn, magnetic fields are gen- as that of Uranus.
erated in deeply buried layers of metallic hydrogen. In Ura- The reason for the unusual geometry of the magnetic
nus and Neptune, magnetic fields arise in salt brine oceans. fields of Uranus and Neptune remains unknown, but it is
Although their origins are complex, we can schematically clearly not related to the orientations of their rotation axes.
illustrate the geometry of the magnetic fields of the giant
planets as if they came from bar magnets (see Chapter 7) in
the interiors of the planets, as shown in Figure 9.18. Giant Planets Have Giant
Jupiter’s magnetic-field axis is inclined 10° to its rotation
axis— an orientation similar to Earth’s— but it is displaced
Magnetospheres
about a tenth of a radius from the planet’s center (Figure Just as Earth’s magnetic field traps energetic charged par-
9.18a). Note that the direction of Jupiter’s magnetic field is ticles to form Earth’s magnetosphere, the magnetic fields of
the giant planets also trap energetic particles to form mag-
2
Although many planetary scientists believe that the giant planets netospheres of their own. Although Earth’s magnetosphere
formed by core accretion, others have proposed a disk instability pro- may have a radius over 10 times that of our planet itself, our
cess, as we learned in Chapter 6. magnetosphere is tiny in comparison with the vast clouds
(a) Jupiter (b) Saturn

Rotation axis Magnetic axis
N
N
N Eq N
ua
tor
Equator S
S
Magnetic
equator
S
S
(c) Uranus (d) Neptune
N
Equator
S S Eq
ua
tor
N N
N
S
S
VISUAL ANALOGY FIGURE 9.18 The magnetic fields of the giant planets can be approximated by
the fields from bar magnets offset and tilted with respect to the planets’ axes. Compare these with
Earth’s magnetic field, shown in Figure 7.16a.
of plasma held together by the much more powerful mag- The size and shape of a planet’s magnetosphere can change
netic fields of the giant planets. By far the most colossal of a great deal depending on how the solar wind is blowing at
these is Jupiter’s magneto- any particular time. Planetary magnetic fields also divert the
sphere. Its radius is as much The magnetospheres solar wind, which flows around magnetospheres the way a
as 100 times that of the planet of the giant planets stream flows around boulders. Just as a rock in a river cre-
itself, or about 7 million km. are enormous. ates a wake that extends downstream, the magnetosphere of
That is roughly 10 times the a planet produces a wake that can extend downstream for
radius of the Sun! Although the magnetospheres of the other great distances. The wake of Jupiter’s magnetosphere (Figure
giant planets are much smaller, even the relatively weak 9.19) extends over 6 astronomical units (AU) outward from
magnetic fields of Uranus and Neptune form magnetospheres the planet—well past the orbit of Saturn. Jupiter’s magneto-
that are comparable in size to the Sun. sphere is the largest permanent “object” in the Solar System,
The solar wind (see Chapter 8) does more than supply surpassed in size only by the tail of an occasional comet.
some of the particles for a magnetosphere. The pressure of the The magnetic wakes of Uranus and Neptune have a curious
solar wind also pushes on and compresses a magnetosphere. structure. Because of the tilt and the large displacements of
their magnetic fields from the centers of these planets, their locked to the conducting liquid layers deep within the
magnetospheres wobble as the planets rotate. This wobble planet’s interior; the magnetic field thus rotates with exactly
causes the wakes of their magnetospheres to twist like cork- the same period as the deep
screws as they stretch away from the planets. interior of the planet. Given Radio signals
the fast and highly variable broadcast a planet’s
winds that push around true rotation period.
Magnetospheres Produce clouds in the atmospheres of
the giant planets, measurement of radio emission is the only
Synchrotron Radiation way we have to determine the true rotation periods of the
We need not send spacecraft to the outer Solar System, giant planets.
or even call on telescopes orbiting Earth, to see evidence
of the giant planets’ magnetospheres. Rapidly moving
electrons in planetary magnetospheres spiral around the
direction of the magnetic field, and as they do so they emit
Radiation Belts and Auroras
synchrotron radiation, as discussed in Foundations 9.1. If As a planet rotates on its axis, it drags its magnetosphere
our eyes were sensitive to radio waves, then the second around with it. In the enormous magnetosphere of a rap-
brightest object in the sky would be Jupiter’s magneto- idly rotating planet like Jupiter, charged particles are swept
sphere. The Sun would still be brighter, but it would not around at high speeds. These fast-moving charged particles
appear larger; even at a distance from Earth of 4.2–6.2 AU, slam into neutral atoms (which do not share the motion of
Jupiter’s magnetosphere would appear roughly twice the material in the magnetosphere), and the energy released
size of the Sun in the sky. Saturn’s magnetosphere would in the resulting high-speed collisions heats the plasma to
also be large enough to see, but it would be much fainter extreme temperatures. In 1979, while passing through Jupi-
than Jupiter’s. Even though Saturn has a strong magnetic ter’s magnetosphere, Voyager 1 encountered a region of tenu-
field, pieces of rock, ice, and dust in Saturn’s spectacular ous plasma with a temperature of over 300 million K; this is
rings act like sponges, soaking up magnetospheric par- 20 times the temperature at the center of the Sun! The den-
ticles. Magnetospheric particles typically collide with sity of the plasma (about 10,000 atoms per cubic meter) was
ring material soon after those particles enter the magne- much lower than that of the best vacuum we can produce
tosphere. With far fewer magnetospheric electrons, there on Earth, however, so the spacecraft was in no danger.
is much less radio emission from Saturn. Charged particles trapped in planetary magnetospheres
We can learn a great deal about planets by studying the are concentrated in certain regions called radiation belts.
synchrotron radiation from their magnetospheres. For exam- Although Earth’s radiation belts are so severe as to worry
ple, precise measurement of periodic variations in the radio astronauts, the radiation belts that surround Jupiter are
signals “broadcast” by the giant planets tells us the planets’ searing in comparison. In 1973 the Pioneer 11 spacecraft
true rotation periods. The magnetic field of each planet is passed through the radiation belts of Jupiter. During its brief
FIGURE 9.19 The solar wind compresses Jupiter’s (or any other) magnetosphere in the direction
toward the Sun and draws it out into a magnetic wake in the direction away from the Sun. Jupiter’s
wake stretches beyond the orbit of Saturn.
Note that this drawing is not to

scale. If Jupiter’s magnetosphere Saturn’s
were as drawn, Saturn would be orbit
3 meters to the right of the page.
Jupiter
Jupiter’s Magnetic
magnetosphere wake
Saturn’s
Saturn magnetosphere
Solar
wind
Bow shock
Synchrotron Radiation—From Planets to Quasars

In Chapter 4 we found that anytime a charged particle expe-
riences an acceleration, the particle emits electromagnetic
radiation. The accelerations resulting from forces exerted by
magnetic fields are no exception. As we have seen, charged
particles moving in a magnetic field experience a force that Synchrotron
radiation
is perpendicular both to the motion of the particle and to the
direction of the magnetic field. This force produces an accelera- Magnetic-field
tion, causing the particles to follow helical paths around the direction
direction of the magnetic field. The accelerations also cause
the particles to radiate. e–
If the particles are traveling at a significant fraction of the
speed of light, then relativistic effects cause the radiation they e–
emit to be beamed in the direction in which they are travel-
ing. The situation is illustrated in Figure 9.20. The resulting
e–
radiation is called synchrotron radiation, so named because
it was fi rst discovered in a type of particle accelerator called
a “synchrotron.”
The amount of radiation from a particle depends on the
amount of acceleration the particle experiences. For a given
Synchrotron radiation is beamed
amount of force, a less massive particle experiences more in the direction that an electron
acceleration. Because electrons are much less massive than is moving at the moment.
protons or any ion, it is the electrons in a magnetized plasma
that experience the greatest acceleration. Combining these FIGURE 9. 20 Rapidly moving charged particles loop around
ideas, we see that it must be the electrons in a magnetized the direction of a magnetic field, giving off electromagnetic
plasma that produce the overwhelming majority of its syn- radiation as a result of the acceleration they experience.
chrotron radiation. The radiation, called synchrotron radiation, is beamed by
The magnetospheres of the giant planets contain energetic relativistic effects in the direction of particle motion.
electrons moving in strong magnetic fields, so they satisfy the
requirements for synchrotron radiation. Synchrotron radiation from planetary magnetospheres is concentrated in the low-
is unlike thermal (Planck) radiation because instead of being energy radio part of the spectrum.
strongly peaked in one part of the electromagnetic spectrum, This is our fi rst encounter with synchrotron radiation, but
synchrotron radiation from a single source can range from it will not be our last. As we move outward into the universe
radio waves to X-rays. The spectrum of synchrotron radiation we will fi nd many objects, from quasars to the remnants of
is determined by the strength of the magnetic field and how supernovae, that emit synchrotron radiation throughout the
energetic the radiating particles are. Synchrotron radiation electromagnetic spectrum.
encounter, Pioneer 11 picked up a radiation dose of 400,000 intense radiation belt in the Solar System is a toroidal (that
rads, or about 1,000 times the is, torus- or doughnut-shaped) ring of plasma associated
lethal dose for humans. Sev- Jupiter has intense with Io, the innermost of Jupi-
eral of the instruments on radiation belts. ter’s four Galilean moons. As Io is a source of
board were permanently dam- you will learn in Chapter 11, magnetospheric
aged as a result, and the spacecraft itself barely survived to Io is the most volcanically particles.
continue its journey to Saturn. active body in the Solar Sys-
In addition to protons and electrons from the solar wind, tem. Because of its low surface gravity and the violence of
the magnetospheres of the giant planets contain large its volcanism, some of the gases erupting from its interior
amounts of various elements, including sodium, sulfur, can escape the moon and become part of Jupiter’s radiation
oxygen, nitrogen, and carbon. These elements come from belt. As charged particles are slammed into the moon by
several sources, including the planets’ extended atmo- the rotation of Jupiter’s magnetosphere, even more material
spheres and the moons that orbit within them. The most is knocked free of the surface and ejected into space. If the
FIGURE 9. 2 1 A faintly
glowing torus of plasma
surrounds Jupiter (center).
The torus is made up of
atoms knocked free from
the surface of Io by charged
particles. Io is the innermost
of Jupiter’s Galilean moons.
A semitransparent mask was
placed over the disk of Jupiter
to prevent it from overwhelming
Io’s plasma torus the much fainter torus.
dislodged atoms are electrically neutral, they will continue variable because the orbit of Titan is sometimes within and
to orbit the planet in nearly the same orbit as the moon from sometimes outside Saturn’s magnetosphere, depending on
which they escaped. Images of the region around Jupiter, the strength of the solar wind. When Titan is outside Sat-
taken in the light of emission lines from atoms of sulfur or urn’s magnetosphere, any nitrogen molecules lost from the
sodium, show a faintly glowing torus of plasma supplied moon’s atmosphere are carried away by the solar wind.
by the moon (Figure 9.21). Charged particles spiral along the magnetic-field lines of
Other moons also influence the magnetospheres of the the giant planets, bouncing back and forth between the two
planets they orbit. The atmosphere of Saturn’s largest moon, magnetic poles, just as they do around Earth. As with Earth,
Titan, is rich in nitrogen. Leakage of this gas into space is the these energetic particles collide with atoms and molecules
major source of a plasma torus that forms in Titan’s wake. in a planet’s atmosphere, knocking them into excited energy
The density of this rather remote radiation belt is highly states that decay and emit radiation. The results are bright
FIGURE 9. 22 Hubble Space Telescope images of auroral rings around the poles of Jupiter (a) and Saturn (b). The auroral
images were taken in ultraviolet light and then superimposed on visible-light images. (High-level haze obscures the ultraviolet
views of the underlying cloud layers, as the insets show.) The bright spot and trail outside the main ring of Jupiter’s auroras
are the footprint and wake of Io’s flux tube.
Auroral rings appear where the planet’s Flux tubes associated with moons
magnetic field channels energetic particles can form auroral footprints that move
into the planet’s atmosphere near the poles. through a planet’s atmosphere.
(a) JUPITER (b) SATURN
Footprint of
Io flux tube
FIGURE 9. 23 The geometry

of Io’s plasma torus and flux
tube.
Io
Jupiter
Io’s flux tube
Io’s plasma torus
Direction of
magnetic field
auroral rings (Figure 9.22) that surround the poles of the trillion watts (W) of power or about 1⁄6 of the total electric
giant planets—just as the aurora borealis and aurora austra- power produced by generating stations on Earth. Much of
lis ring the north and south magnetic poles of Earth. the power generated within the flux tube is radiated away
Jupiter’s auroras have an added twist that we do not see as radio energy. These radio signals are received at Earth
on Earth, however. As Jupiter’s magnetic field sweeps past as intense bursts of synchrotron radiation. However, a sub-
Io, it behaves like a dynamo, generating an electric potential stantial fraction of the energy of particles in the flux tube
of 400,000 volts (V). Electrons accelerated by this enormous is deposited into Jupiter’s atmosphere. At the very location
electric field spiral along the direction of Jupiter’s magnetic where Io’s flux tube intercepts Jupiter’s atmosphere, there
field. The result is a magnetic channel, called a flux tube, is a spot of intense auroral activity. As Jupiter rotates, this
that connects Io with Jupiter’s atmosphere near the planet’s spot leaves behind an auroral trail in Jupiter’s atmosphere.
magnetic poles (Figure 9.23). Io’s flux tube carries an elec- The footprint of Io’s plasma torus, along with its wake, can
tric current of 5 million amperes (amps), amounting to 2 be seen outside the main auroral ring in Figure 9.22a.
In exploring the giant planets, we encounter worlds

Seeing the Forest for the Trees with scales beyond human experience. In place of
Earth’s gentle trade winds, we fi nd vast atmospheric
In our discussion of the terrestrial planets, a major theme bands streaming around the giant planets at speeds
was diversity. We used Earth as the basis for understand- greater than 2,000 km/h. The most powerful terrestrial
ing our planetary neighbors, and at the same time we hurricanes would be but inconsequential eddies within
learned from those neighbors what it is about Earth that Jupiter’s Great Red Spot, a system so enormous that it
sets it apart. On this leg of our journey we have come to could swallow two Earths whole. On Earth we marvel
understand that even the range of conditions from the at the sight of a terrestrial thunderstorm towering over-
frozen poles of Mars to the hellish surface of Venus is head, lit up by powerful flashes of lightning. Can we
modest in comparison with what is possible. As the Solar begin to imagine the appearance of multi-hued clouds
System formed, planets in the cold outer reaches grew of water, ammonia, methane, and other compounds,
more massive than the planets closer to the Sun. They tinted by sulfur, phosphorus, and photochemical smog,
became massive enough to capture and hold on to the as they roil up from within the depths of a giant planet?
hydrogen and helium gas that was so abundant in the Can we envision the bolts of lightning that sear through
disk surrounding the young Sun. those clouds?
Of course, these analogs of our terrestrial experience the Sun and the physics and chemistry of our atmosphere
only scratch the surface of the differences between the and oceans, the fierce weather systems on the giant plan-
giant planets and Earth. In the case of the terrestrial planets are powered largely by the conversion of gravitational
ets, the distinction between planet and atmosphere is energy to thermal energy within their interiors.
clear-cut. On our Earth, clouds billow in a blue sky, adrift Although the bulk of the giant planets dwarfs that
over the palpable surface of land and ocean. As we descend of their terrestrial cousins, their influence does not end
into the giant planets, there is no sudden transition to solid here. Powerful magnetic fields trap energetic charged
or liquid. Instead there is only the steady and inexorable particles streaming outward from the Sun, leading to
increase in pressure, a smooth transition from the tops of the formation of giant magnetospheres—the largest per-
the clouds to a core that in the case of Jupiter may have an manent structures in the Solar System.
absolute temperature about six times that at the surface of In the following chapter, the time has come for a brief
the Sun, and a crushing pressure 45 million times that at digression. The systems of moons and rings that surround
the surface of Earth. The higher-than-expected tempera- the giant planets are gravitational playgrounds, rich with
tures of the giant planets that we found in Chapter 4 are phenomena that go far beyond Kepler’s laws. Before we
signposts of these internal differences. Whereas Earth’s take the next step outward, we will pause and return our
climate is defined by the interplay between energy from attention to our old friend gravity.
Summary Student Questions

• The giant planets are large and massive, and less dense THINK ING A BOUT THE CONCEP T S
than the terrestrial planets.
1. Identify and describe the two subclasses of giant plan-
• Jupiter and Saturn are made up mostly of hydrogen and ets, and indicate which planets fall into each subclass.
helium— a composition similar to that of the Sun.
2. Describe the ways in which the giant planets differ from
• Uranus and Neptune contain much larger amounts of the terrestrial planets.
“ices” such as water, ammonia, and methane.
3. What observations led Adams and Le Verrier to predict
• We see only atmospheres, not solid surfaces, on the the location of Neptune?
giant planets.
4. Jupiter’s chemical composition is more like the Sun’s
• Clouds on Jupiter and Saturn are composed of various than Earth’s is. Yet both planets formed from the same
kinds of ice crystals colored by impurities. protoplanetary disk. Explain the difference.
• Powerful convection and the Coriolis effect drive high- 5. What can we learn about a Solar System object when it
speed winds in the upper atmospheres of the giant “occults” a star?
planets.
6. Explain two methods used to determine the mass of a
• Jupiter, Saturn, and Neptune have large internal heat giant planet.
sources. Uranus seems to have little or none.
7. Why can we not see the surfaces of the giant planets?
• The interiors of the giant planets are very hot and very
8. Jupiter is sometimes described as “a star that failed.”
dense.
Explain why this is a gross exaggeration.
• The giant planets have enormous magnetospheres that
9. Astronomers take the unusual position of lumping
emit synchrotron radiation.
together all atomic elements other than hydrogen and
helium into a single category, which they call “heavy
elements.” Why is this a reasonable thing for astrono-
mers to do?
includes algorithmically generated versions of these 10. None of the giant planets is truly round. Explain why
questions, plus additional conceptual exercises. If they have a flattened appearance.
*11. Why do the individual cloud layers in the atmo-
spheres of the giant planets have different chemical
compositions?
12. What is the source of color in Jupiter’s clouds? Earth sees the star disappear for 37 minutes and 2 sec-
onds and notes that the center of Uranus passed directly
13. Uranus and Neptune, when viewed through a telescope,
in front of the star.
appear distinctly bluish green in color. What are the two
a. On the basis of these observations, what value would
reasons for their striking appearance?
the observer calculate for the diameter of Uranus?
14. Which of the giant planets have seasons similar to b. What could you conclude about the planet’s diam-
Earth’s, and which one experiences extreme seasons? eter if its center did not pass directly in front of the
star?
15. What creates metallic hydrogen in the interiors of Jupiter
and Saturn, and why do we call it “metallic?” 28. Jupiter’s equatorial radius is 71,500 km, and its oblate-
ness is 0.065. What is Jupiter’s polar radius?
*16. Jupiter’s core is thought to consist of rocky material and
ices, all in a liquid state at a temperature of 35,000 K. 29. Ammonium hydrosulfide (NH4HS) is a molecule in Jupi-
How can materials such as water be liquid at such high ter’s atmosphere responsible for many of its clouds.
temperatures? Using Figure 9.5, calculate the molecular weight of an
ammonium hydrosulfide molecule, where the atomic
17. Explain how astronomers measure wind speeds in the
weight of a hydrogen atom is 1. (Hint: The weight of a
atmospheres of the giant planets.
molecule is equal to the sum of the weights of its com-
18. The Great Red Spot is a long-lasting atmospheric vortex ponent atoms.)
in Jupiter’s southern hemisphere. Winds rotate coun-
30. Jupiter is an oblate planet with an average radius of
terclockwise around its center. Is the Great Red Spot
69,900 km, compared to Earth’s average radius of 6,370
cyclonic or anticyclonic? Is it a region of high or low
km.
pressure? Explain.
a. Remembering that volume is proportional to the cube
*19. What drives the zonal winds in the atmospheres of the of the radius, how many Earth volumes could fit inside
giant planets? Jupiter?
b. Jupiter is 318 times as massive as Earth. Show that
20. Lightning has been detected in the atmospheres of all
Jupiter’s average density is about 1⁄4 that of Earth’s.
of the giant planets. How is it detected?
31. The obliquity of Uranus is 98°. If you were located at one
*21. Jupiter, Saturn, and Neptune all radiate more energy into
of the planet’s poles, how far from the zenith would the
space than they receive from the Sun. Does this violate
Sun appear at the time of summer solstice?
the law of conservation of energy? What is the source of
the additional energy? **32. A small cloud in Jupiter’s equatorial region is observed
to be at a longitude of 122.0° west in a coordinate sys-
*22. Saturn has a source of internal thermal energy that Jupi-
tem that rotates at the same rate as the deep interior of
ter may not have. What is it and how does it work?
the planet. (West longitude is measured along a planet’s
23. Explain how we know that Uranus and Neptune are not equator toward the west.) Another observation made
gas giants like Jupiter and Saturn. exactly 10 Earth hours later finds the cloud at a longitude
of 118.0° west. Jupiter’s equatorial radius is 71,500 km.
24. When viewed by radio telescopes, Jupiter is the second
What is the observed equatorial wind speed in kilome-
brightest object in the sky. What is the source of this
ters per hour? Is this an easterly or a westerly wind?
radiation?
33. The equilibrium temperature for Saturn should be 82
25. What creates auroras in the polar regions of Jupiter and
K, but instead we fi nd an average temperature of 95 K.
Saturn?
How much more energy is Saturn radiating into space
than it absorbs from the Sun?
*34. Neptune radiates into space 2.6 times as much energy
as it absorbs from the Sun. Its equilibrium temperature
(see Chapter 4) is 47 K. What is its true temperature?
26. The Sun appears 400,000 times brighter than the full
Moon in our sky. How far from the Sun (in astronomi-
cal units) would you have to go for the Sun to appear
only as bright as the full Moon appears in our nighttime
sky? Compare your answer with the semi-major axis of
Neptune’s orbit.
27. Uranus occults a star at a time when the relative motion quizzes, plus links to premium content in SmartWork and
between Uranus and Earth is 23.0 km/s. An observer on the ebook. Visit www.wwnorton.com/studyspace.
Then there are the Tides, so useful to man . . .
We must be grateful for the Moon’s existence on that
account alone.
JAMES NASM Y T H (1808 –1890)
Jupiter’s moon Io, showing reddish fallout from an enormous erupting volcano.
CHAP TER 10
Gravity Is More than Kepler’s Laws

10.1 Gravity Once Again
K E Y CONCEP T S
Long before Kepler wrote down the laws describing the
motions of the planets about the Sun, or Newton explained Newton’s law of gravitation is simplicity itself. As we
and interpreted these motions in terms of the effects of grav- saw in Chapter 3, it helps explain Kepler’s laws of
ity, our ancestors knew that the Moon and the Sun have a planetary motion. Yet when applied to extended rather
direct influence on Earth. Those living near the ocean were than pointlike masses or when acting among three or
especially attuned to these effects. Twice each day—in har- more objects, this simple law gives rise to a surprising
mony with the changing position of the Moon in the sky— diversity of phenomena. In this chapter we apply
they saw the seawater rise and then recede once again. They Newton’s law beyond Kepler’s elegant laws of planetary
would have noted that this effect, Earth’s tides, is great-
motion and discover
est when the Sun and the Moon are either together in the
sky or at opposite extremes (that is, during a full Moon or • How symmetry enables us to say a great deal about
a new Moon) and is more subdued when the Sun and the the gravity of an object without actually calculating
Moon are separated in the sky by 90° (fi rst quarter Moon or anything.
third quarter Moon). Doubtless the unarguable association • That the gravity within a spherical object is determined
between the Moon and tides had a great deal to do with the only by the mass within a given radius.
development of superstitions about the power of the Moon
over our lives. • Tides on Earth result from the fact that gravity from
Today we understand that tides are the result of the the Moon and the Sun pull harder on one side of Earth
gravitational pull of the Moon and the Sun on Earth. More than the other.
to the point, tides are the result of differences between how • Tidal interactions between planets and moons
hard the Moon and Sun pull on one part of our planet in (including Earth’s Moon) that lock a moon’s rotation to
comparison with their pull on other parts of the planet. its orbit.
So far in our discussion of gravity we have focused on the
gravitational interaction between two whole objects. Such • Comets that have been shattered by tidal stresses,
interactions explain the motion of the planets about the and tortured moons alive with tide-powered volcanism.
Sun, the Moon about Earth, and a space shuttle about our • Orbital interactions that nudge asteroids from their
globe. Yet the concept of gravity can explain far more than orbits and sweep out gaps in planetary rings.
Newton’s derivation of Kepler’s laws. While concentrating on
the elliptical, parabolic, and hyperbolic orbits of one mass
about another, we have overlooked many other fascinating
and important manifestations of gravity. We now turn our
attention to some of these effects.
275
276 Chapter 10 Gravity Is More than Kepler’s Laws
10.2 Gravity Differs responding piece of Earth is pulling you backward. And,
because Earth is almost spherically symmetric, all of these
from Place to Place “sideways” forces cancel out, leaving behind an overall force
that points toward Earth’s center.
within an Object Understanding the size of the force is a bit trickier. Some
parts of Earth are closer to you while others are farther
A good place to begin a discussion of the additional effects away, but there must be an
of gravity is with the way an object interacts gravitationally “average” or “characteristic” Gravity from a sphere
with itself. You can think of distance between you and acts as though all of
Earth, for example, as a col- Self-gravity holds each of the small fragments of its mass were
lection of small masses, each Earth together. Earth that is pulling on you. concentrated at the
of which feels a gravitational Not surprisingly, this average sphere’s center.
attraction toward every other small part of Earth. This gravi- distance turns out to be the
tational attraction between different parts of Earth is what distance between you and the center of Earth. So, as illus-
holds our planet together. We call the mutual gravitational trated in Figure 10.1b, the overall pull that you feel is the
attraction that occurs between all parts of the same object same as if all the mass of Earth were concentrated at a single
self-gravity. AstroTour: Tides and the Moon point located at the very center of the planet. This relation-
In a certain sense you are one of these small pieces of ship is true for any spherically symmetric object. As far as
Earth. You are perhaps slightly more mobile than the aver- the rest of the universe is concerned, the gravity from such
age lump of clay; but as far as the gravitational makeup of an object behaves as if all the mass of that object were con-
our planet goes, you serve more or less the same purpose. centrated at a point at its center. We have already made
As you sit reading this book, you are exerting a gravitational extensive use of this result. For example, it was an implicit
attraction on every other fragment of Earth, and every other assumption in our discussion in Chapter 3 of orbits and
fragment of Earth is exerting a gravitational attraction on Kepler’s laws that the force acting on an object in orbit about
you, as illustrated in Figure 10.1a. Your gravitational inter- Earth is equal to GM⊕mobject/r2, where M⊕ is the mass of Earth
action is strongest with the parts of Earth closest to you. and r is the distance between the center of Earth and the
The parts of Earth that are on the other side of our planet orbiting object.
are much farther from you, so their pull on you is corre- We now have a way of accounting for the net gravita-
spondingly less. tional attraction of Earth on an object on or above the sur-
As you know from experience, the net effect of all these face of Earth, but what about
forces is to pull you toward the center of Earth. If you drop the effect of gravity within Net gravity is zero at
a hammer, it falls directly down toward the ground. We can Earth? What is the overall the center of Earth.
understand why this is so just by thinking about the shape gravitational attraction felt by
of Earth. Because Earth is nearly spherical, for every piece of an object buried deep inside our planet? To answer this
Earth pulling you toward your right a corresponding piece question, let’s begin by imagining a physically impossible
of Earth is pulling you toward your left with just as much but theoretically plausible scenario: you are in a room at
force. For every piece of Earth pulling you forward a cor- the center of Earth, as shown in Figure 10.2. (Ignore for a
(a) (b)
FIGURE 10.1 The net
gravitational force due to a
spherical mass is the same as
the gravitational force from
m m
the same mass concentrated
m at a point at the center of the
¢
R
Total force sphere.

Center of
Earth m mobserver F mobserver
M = M¢
m
m m M¢ mobserver
Total mass F=G
R¢2
= M¢
An object on the surface of a spherical mass The net force is the same as if we scooped
(such as Earth) feels a gravitational attraction up the mass of the entire sphere and
toward each small part of the sphere. concentrated it at a point at the center.
10.2 Gravity Differs from Place to Place within an Object 277
(a) (b)
In an imaginary (and unreachable) room
at the center of Earth the gravity from each
small piece of Earth would pull you outward. FIGURE 10. 2 In an imaginary
room at the center of Earth,
the gravitational pull from each
small piece of Earth would
cancel, leaving you weightless.
The pull of each small piece would cancel Because of Earth’s spherical
the pull of the corresponding piece at the symmetry all of these pulls would
opposite location in the sphere. cancel, leaving you weightless.
moment the crushing pressure or immense temperatures your mass, and r is the distance from the center of Earth
that would surround you!) Every small part of Earth would to your room.
be exerting a slight gravitational force on you, each of which So far so good; but what about the gravitational attrac-
would be outward, away from the center of Earth. Again, tion that you would feel from the rest of Earth—the shell
Earth’s spherical symmetry tells us that, overall, these gravi- of material that surrounds this “inner planet”? The parts
tational forces must sum to zero. For each part of Earth of the shell that are closest to you, and so individually pull
pulling you outward in one direction, a corresponding piece on you most strongly, are above you. The force from this
of Earth on the other side would pull you outward in exactly part of the planet would pull you away from the center of
the opposite direction, as shown in Figure 10.2a. These two Earth. However, most of the mass in the shell is on the side
forces would cancel each other so that the net force you where its gravitational attraction pulls you toward the cen-
would feel from these two small parts of the planet would ter of Earth. Although this material is farther away from
be zero (Figure 10.2b). The same holds true for each and you—which means that each small part of it pulls on you
every small part of Earth. The net effect of the sum of all less strongly—there is a good bit more of it than the mate-
the gravitational forces acting on an object at the center of rial immediately over your head. If we were to work it out
Earth is zero! If you were in a room at the center of Earth, in detail, we would fi nd that the gravitational attraction
you would hang there, surrounded by the enormous mass from the part of the shell pulling you away from the center
of a planet but truly weightless. (See Foundations 10.1 for of Earth is exactly canceled by the gravitational attraction
more discussion of symmetry.) of the part of the shell that is pulling you toward the center
Now imagine that your room is located partway out from of Earth. The net effect—the overall force—is zero. If you
the center of the planet. What gravitational forces would you were inside a huge spherical shell, the overall gravitational
feel at this new location? To tackle this question we need to attraction from the material in that shell would be zero.
think of Earth as being made up of two different pieces, as (This example is just a more general case of the “room at the
shown in Figure 10.3. The fi rst piece is just a spherical ball center of Earth” problem illustrated in Figure 10.3.)
containing all the parts of Earth that are closer to the cen- Let’s pull all of this information together. Regardless of
ter of Earth than your imaginary room is. You could think where we are within Earth, we feel a net gravitational force
of this inner ball as a “planet within a planet.” Your room only from the part of the planet that lies deeper within the
is sitting on the surface of this imaginary sphere. From the planet than we are, and that mass acts as if it were all con-
discussion in the preceding paragraphs, we know the net centrated at the center of Earth. If we want to know what
effect of the gravity of this inner sphere. It is equivalent to the net gravitational attraction is at any point within a
taking all the mass contained within that imaginary ball spherical object, all we need to do is determine how much
and placing it at the center of Earth. The force that you mass is closer to the center of the sphere than is our point
would feel from this inner planet is of interest. This mass then acts as if it were concentrated
at the center of the sphere. This general result is extremely
G × Minner × m
F = ____________ , important. We have already seen something of this effect
r2 in our discussion of the interiors of planets. It will become
where Minner is the mass of this inner ball (which depends even more important as we consider the forces responsible
on where within the planet your room is located), m is for the structure of stars.
F O U N D AT I O N S 1 0 .1
The Power of Symmetry

In our discussion of gravity we have managed to arrive at some it has “circular symmetry,” meaning that its shape is the same
very interesting results based on nothing but the way one part of regardless of how it rotates about the axis running through its
an object matches up with another—a property called symmetry. center. Your image in a mirror is fl ipped right for left and left
We were able to say that the overall gravitational attraction of for right from your actual appearance— an example of “reflec-
all parts of Earth must point along a line connecting us with tion symmetry.” You hardly notice the difference, however,
the center of Earth, simply by noting that each sideways pull because your body itself is nearly symmetric between its right
from a different part of Earth is balanced by a corresponding and left sides. A child’s cubical blocks can be stacked any of
pull in the other direction. We could make this claim because six ways because when you rotate them by 90° they still look
the distribution of mass to our right as we stand on Earth is the same— a special type of “rotational symmetry.” A soccer
just the same as the distribution of mass to our left, which is just ball can be kicked in any direction at any time, and it will roll
the same as the distribution of mass to our front or back. That any which way— both results of its “spherical symmetry.” An
is to say, Earth is symmetric. Similarly, we were able to argue American football makes for a lousy game of soccer but a fan-
that the net effect of gravity on an object at the center of Earth tastic spiraling forward pass because of its rotational symme-
must be zero because every small force from one part of Earth try about a single axis.
is balanced by a corresponding force from the corresponding Science progresses by fi nding and making use of patterns
part of Earth on the opposite side of us. Again, no calculation that exist in the universe. Symmetry turns out to be an espe-
was needed— only an appeal to the symmetry of the distribu- cially elegant and powerful type of pattern. Physicists and
tion of mass within Earth. astronomers can often learn a great deal about the properties
If you look around, you will fi nd symmetry everywhere; and behavior of an object or system without making a single
and when you fi nd it, you will often fi nd a clue to why some- calculation, simply by understanding the symmetry of the
thing works the way it does. The tire on a bicycle rolls because object or system that they are studying.
GRAVITY FROM INNER SPHERE

1 To understand gravity within
a sphere, think of the sphere
as an inner sphere and an Minner
outer shell. r inner
2 As in Figure 10.2, 3 …as if the entire mass Minner

gravity from the inner of the inner sphere were
r inner
sphere is the same… concentrated at its center.
GRAVITY FROM OUTER SHELL
7 Within a sphere, only

4 Gravity from this part gravity from the “inner”
of the outer shell… mass matters:
M m
F = G inner2 observer
r inner
5 …is balanced by
gravity from this part
of the outer shell… FIGURE 10. 3 The gravitational
force felt in a room in the interior
of a spherical body such as Earth
6 …leaving no net is equal to the force due to the
gravitational force mass that is closer to the center
from the outer shell. of the body than the room is.
10.3 Tides Are Due to Differences in the Gravitational Pull from External Objects 279
10.3 Tides Are Due to Moon. In contrast, the side of Earth facing away from the
Moon is farther than average from the Moon, so it feels a
Differences in the weaker-than-average attraction toward the Moon. Evaluating
these effects numerically, we find that the pull on the near
Gravitational Pull from side of Earth is about 7 percent larger than the pull on the
far side of Earth.
External Objects You can perhaps better visualize how this difference in
gravitational attraction works if you carry out a “thought
We have seen that each small part of an object feels a grav- experiment.” Imagine holding three rocks at different alti-
itational attraction toward every other small part of the tudes high above the surface of the Moon, as shown in Figure
object, and that this self-gravity differs from place to place. 10.4a. If you let them go at the same time, they would all fall
In addition, each small part of an object feels a gravitational toward the Moon. But rock 1, located closest to the Moon and
attraction toward every other mass in the universe, and thus subject to greater acceleration, would fall faster than
these external forces differ from place to place within the rocks 2 and 3. Rock 3, located farthest from the Moon, would
object as well. fall more slowly than either of the other two rocks. As the
The most notable local example of an external force rocks fell toward the Moon, the separation between them
involves the effect of the Moon’s gravity on Earth. Overall, would grow. A person falling along with rock 2 would see
the Moon’s gravity pulls on both rocks 1 and 3 moving away in opposite directions. So
Earth as if the mass of Earth The Moon’s gravity far so good. Now suppose the three rocks were connected
were concentrated at the plan- pulls harder on by springs, as in Figure 10.4b. As the rocks fell toward the
et’s center, as we just saw. the side of Earth Moon, the differences in the gravitational forces they felt
When astronomers calculate facing the Moon. would stretch the springs. To someone falling along with
the orbits of Earth and the rock 2, it would seem as if there were forces pulling rocks 1
Moon assuming that the mass of each is concentrated at the and 3 in opposite directions. Exactly the same thing happens
center point of each body, their calculations perfectly match with Earth. If we replace the three rocks with different parts
the observed orbits. Yet there is more going on, because of of Earth, as shown in Figure 10.4c, we see that differences
the finite size of Earth. The side of Earth that faces the Moon in the Moon’s gravitational attraction on different parts of
is closer to the Moon than is the rest of Earth, so it feels a Earth try to stretch Earth out along a line pointing toward
stronger-than-average gravitational attraction toward the the Moon. AstroTour: Tides and the Moon
FIGURE 10. 4 (a) As three objects fall toward the Moon, they experience different gravitational
accelerations. (b) If the three objects were connected by springs, the springs would be stretched
as if there were forces pulling outward on each end of the chain. (c) The difference in the Moon’s
gravitational attraction between the near and far sides of Earth is the cause of Earth’s tides.
(a) (b) (c)

Rock 3 Tidal “force” Tidal “force”
Farther, less
acceleration
Rock 2
Rock 1
Net
acceleration
Closer, more
acceleration Tidal “force” Tidal “force”
Moon
As three objects fall toward the If there were springs between As Earth as a whole falls
Moon, they spread out because the objects, they would be toward the Moon, tides stretch
of differences in the gravitational stretched, as if by forces Earth out along the line between
accelerations they experience. pulling in opposite directions. the two bodies.
1 Because it is closer, the The average force felt by all parts

Moon’s gravity pulls harder of Earth is responsible for Earth’s
on the near side of Earth… overall motion toward the Moon.
Earth
Moon Moon
FIGURE 10. 5 Tidal stresses

2 …than on the far stretch Earth along the line between The difference between the actual
side of Earth. force at each point and the average
Earth and the Moon but compress
force is called a tidal stress.
Earth perpendicular to this line.
We can also think about this situation by applying the Out of convenience we speak of tidal stresses1 as if there
idea of relative motion from our discussions of Newton’s were actually a force pulling the far side of Earth away from
laws (see Chapter 3). Earth as a whole is constantly falling Earth’s center. However, it is important to remember that the
toward the Moon. The acceleration that Earth experiences Moon is not pushing the far side of Earth away. Rather, it sim-
is not very great— only about 3.3 × 10 –5 meter per second ply is not pulling on the far side of Earth as hard as it is on
per second (m/s2). The gravitational acceleration that we the planet as a whole. The far side of Earth is “left behind” as
feel toward the center of Earth is 300,000 times greater. Yet the rest of the planet falls more rapidly toward the Moon.
the gravitational pull of the Moon on Earth is a substantial In Chapter 3 we learned that the strength of the gravi-
influence on our planet, causing it to shift more than 9,300 tational force between two bodies is proportional to their
kilometers (km) back and forth about their common center masses and inversely proportional to the square of the dis-
of gravity over the course of a month (see Figure 6.14 for a tance between. But the strength of tidal forces caused by
similar effect involving the Sun and Jupiter). However, we one body acting on another is far more complicated. Tidal
do not personally perceive any sensation of movement from influence is also proportional to the mass of the tide-rais-
the gravitational attraction of the Moon on Earth. Just as ing body, but it is inversely proportional to the cube of its
everything in a traveling car shares the motion of the car, distance. Consider, for example, the tidal force acting on
everything on Earth falls together toward the Moon. Earth by the Moon:
What is not exactly the same everywhere on Earth is the
2GM⊕MMoonD⊕
“leftover” acceleration—the difference between the gravi- Ftidal = ___________ ,
tational acceleration due to d3
the Moon at any particular Tidal stresses stretch where D⊕ is Earth’s diameter and d is the distance between
location and the average grav- out Earth in one Earth and the Moon.2 Billions of years ago, the Moon was
itational acceleration acting direction and squeeze much closer to Earth than it is today. Had the Moon’s dis-
on Earth as a whole. Figure it in the other. tance been half of what it is now, oceanic tides would have
10.5 shows the effect of these been eight (23) times as large.
leftover accelerations. On the side of Earth closer to the The tidal acceleration of a mass on either the near or the
Moon, the actual acceleration is greater than the average far side of Earth is only about one 10-millionth of the accel-
acceleration. The result is that 1 kilogram (kg) of material eration due to Earth’s gravity. This acceleration may not seem
on the side of Earth closer to the Moon is pulled toward the like much until we consider the enormous masses involved.
Moon with a force of 1.1 × 10 –6 newton (N) relative to the Because F = ma, a really huge
Moon’s attractive force on Earth as a whole. But on the side m multiplied by a small a Earth’s oceans
of Earth away from the Moon, the actual force is less than results in a large force. In addi- have a tidal bulge.
the average force; the difference between the actual force tion, the liquid water covering
and the average force points away from the Moon! Figure the majority of Earth’s surface is free to move in response to
10.5 shows that there is also a net force squeezing inward tidal forces. In the idealized case—in which the surface of
on Earth in the direction perpendicular to the line between
Earth and the Moon. Together, the stretching of tides along 1
Tidal stresses are forces caused by differences in the gravitational
the line between Earth and the Moon and the squeezing of attraction of one mass acting on different parts of another mass.
the tides perpendicular to this line distort the shape of Earth 2
The proof of this relationship involves the calculus derivative of the
like a rubber ball caught in the middle of a tug-of-war. gravitational inverse square law and is beyond the level of this text.
10.3 Tides Are Due to Differences in the Gravitational Pull from External Objects 281
Earth is perfectly smooth and covered with a uniform ocean, (a)

and Earth does not rotate—the lunar tides (tides on Earth The Moon’s tidal stresses stretch Earth and its
due to the gravitational pull of the Moon) would pull our oceans into an elongated shape. The departure
from spherical is called Earth’s tidal bulge.
oceans into an elongated tidal bulge like that in Figure 10.6a.
The water would be at its deepest on the side toward the Earth before
Moon and on the side away from the Moon, and at its shal- tidal distortion
lowest at points midway between. Of course, our Earth is
Earth’s oceans
not a perfectly smooth, nonrotating body covered with per-
Moon
fectly uniform oceans, and many effects complicate this
simple picture. One complicating effect is Earth’s rotation. Earth
As a point on Earth rotates through the ocean’s tidal bulges,
that point experiences the ebb and flow of the tides. In addi- Tidal Tidal
tion, friction between the spinning Earth and its tidal bulge bulge bulge
drags the oceanic tidal bulge around in the direction of
Earth’s rotation, as shown in Figure 10.6b. (b)
Follow along in Figure 10.6c as we ride our planet through Because of friction, Earth’s rotation
the course of a day. We begin as the rotating Earth carries drags its tidal bulge around, out of
us through the tidal bulge on perfect alignment with the Moon.
the Moonward side of the Tides rise and fall
planet. Because the tidal bulge twice each day.
is not exactly aligned with the
Moon, the Moon is not exactly overhead but is instead high
in the western sky. When we are at the high point in the
tidal bulge, the ocean around us is deeper than average— a
condition referred to as “high tide.” About 61⁄4 hours later,
probably somewhat after the Moon has settled beneath the
western horizon, the rotation of Earth carries us through a (c)
point where the ocean is shallowest. It is “low tide.” If we Ocean tides rise and fall as the
wait another 61⁄4 hours, it is again high tide. We are now pass- rotation of Earth carries us through
the ocean’s tidal bulges.
ing through the region where ocean water is “pulled” (rela-
tive to Earth as a whole) into the tidal bulge on the side of Low tide High tide
(the Moon has (the Moon has
Earth that is away from the Moon. The Moon, which is set in the west) crossed overhead
responsible for the tides we see, is itself hidden from view into the western
on the far side of Earth. About 61⁄4 hours later, probably some- part of the sky)
time after the Moon has risen above the eastern horizon, it
is low tide. About 25 hours after we started this journey—
the amount of time the Moon takes to return to the same
point in the sky from which it started—we again pass through
the tidal bulge on the near side of the planet. This is the age- High tide Low tide
old pattern by which mariners have lived their lives for mil- (the Moon (the Moon has
lennia: the twice-daily coming and going of high tide, is not seen) risen in the east)
shifting through the day in lockstep with the passing of the
FIGURE 10.6 (a) Tidal stresses pull Earth and its oceans into
Moon. AstroTour: Tides and the Moon a tidal bulge. (b) Earth’s rotation pulls its tidal bulge slightly out
Those who live near the ocean may have noticed that of alignment with the Moon. (c) As Earth’s rotation carries us
our description of daily tides is often simpler than what through these bulges, we experience the well-known ocean tides.
you actually observe. The shapes of Earth’s shorelines and
ocean basins complicate our simple picture of tides seen at
any particular point, even in midocean.3 As they respond to
the tidal stresses from the Moon, Earth’s oceans flow around small sizes and the narrow passages connecting these bod-
the various landmasses that break up the water covering ies of water with the larger ocean. In other places, the shape
our planet. Some places, like the Mediterranean Sea and of the land funnels the tidal surge from a large region of
the Baltic Sea, are protected from tides by their relatively ocean into a relatively small area, concentrating its effect.
The Bay of Fundy lies between the Canadian provinces of
3
High tide does not necessarily occur when the Moon is near its high- Nova Scotia and New Brunswick. This bay, along with the
est point above the horizon. The reason is that, in addition to local Gulf of Maine, forms a great basin in which water naturally
shoreline effects, there are oceanwide oscillations similar to water rocks back and forth with a period of about 13 hours. This
sloshing around in a basin. amount of time is close to the 121⁄2-hour period of the ris-
ing and falling of the tides. The characteristics of the basin the Moon. This stretching in perpendicular directions has
amplify the tides, sending the water sloshing back and forth the effect of making low tides higher and high tides lower.
like the water in a huge bathtub. Figure 10.7 shows a picture Such tides, when lunar tides are diminished by solar tides,
of one location on the Bay of Fundy at both high and low are called neap tides (Figure 10.8b). Neap tides are only
tides. The average difference between low and high tides about half as strong as average tides and only a third as
on the bay is about 14.5 meters, and under the right condi- strong as spring tides.
tions this difference can exceed 16.6 meters! The daily ebb and flow of the tides allows beachgoers to
The Sun also influences Earth’s tides. The side of Earth experience the wonders of tide pools, the natural aquari-
closer to the Sun is pulled toward the Sun more strongly ums teeming with marine life. As the tides roll in and out,
than is the side of Earth away from the Sun, just as the side they recharge the tide pools with fresh seawater while
of Earth closest to the Moon is pulled more strongly toward sometimes exchanging the pool’s inhabitants. As you gaze
the Moon. Although the absolute strength of the Sun’s pull into these microenvironments, think back a billion years
on Earth is nearly 200 times greater than the strength of or so: tide pools hold an ancient connection to all plants
the Moon’s pull on Earth, the Sun’s gravitational attraction and creatures now living on solid ground, as described in
does not change by much from one side of Earth to the other, Connections 10.1.
because the Sun is much farther away than the Moon. (Recall
that tidal effects are inversely proportional to the cube of the
distance to the tide-raising body.) As a result, solar tides—
tides on Earth due to differences in the gravitational pull of 10.4 Tidal Effects on
the Sun— are only about half as strong as lunar tides.
The solar and lunar tides interact as illustrated in Fig- Solid Bodies
ure 10.8. If the Moon and the Sun are lined up with Earth,
as occurs at either new Moon So far in our discussion we have focused on the some-
or full Moon, then the tides Solar tides may times dramatic movements of the liquid of Earth’s oceans in
on Earth due to the Sun are reinforce or diminish response to the tidal stresses from the Moon and Sun. But
in the same direction as the lunar tides. these tidal stresses also affect the solid body of Earth. As
tides on Earth due to the we pointed out earlier, Earth is somewhat elastic (like a rub-
Moon. At these times the solar tides reinforce the lunar ber ball), and tidal stresses cause a vertical displacement of
tides, so the tides are about 50 percent stronger than aver- about 30 centimeters (cm) between high tide and low tide, or
age. The especially strong tides near the time of new or full roughly a third of the displacement of the oceans. As Earth
Moon are referred to as spring tides (Figure 10.8a). Con- rotates through its tidal bulge, the solid body of our planet
versely, around the first and third quarter Moon, the stretch- is constantly being deformed. It takes energy to deform the
ing due to the solar tide is at right angles to the stretching shape of a solid object. (If you want a practical demonstra-
due to the lunar tide. The solar tides pull water into the tion of this fact, hold a rubber ball in your hand, and squeeze
dip in the lunar tides and away from the tidal bulge due to and release it a few dozen times. While you are shaking your
FIGURE 10.7 The world’s most extreme tides are found in the Bay of Fundy, located between
Nova Scotia and New Brunswick, Canada. The difference in water depth between low tide (a) and
high tide (b) is typically about 14.5 meters and may reach as much as 16.6 meters.
(a) (b)
(a) (b)
Solar tides
Lunar tides
Lunar
New tides
Moon
Solar tides
Spring tides occur when solar Neap tides occur when
1st quarter
and lunar tides add together, solar and lunar tides partially
resulting in above-average tides. cancel each other.
3rd quarter
Lunar tides
Full
Moon
Lunar
tides
Solar tides
Solar tides
FIGURE 10.8 Earth experiences solar tides that are about half as strong as tides due to the Moon. The interactions of solar and
lunar tides result in either (a) spring tides when they are added together or (b) neap tides when they partially cancel one another.
C O N N E C T I O N S 1 0 .1
Lunar Tides and Life

To ancient Romans and Greeks she was known as Diana or Today, all creatures of land and air, including us, may well
Artemis, goddess of the Moon. From the earliest times to the owe their existence to the tides created by the Moon’s grav-
present, the Moon has occupied a special place in the minds ity. Does this mean, as some have claimed, that advanced life
of romantics and scientists alike. Whether we look upon the can arise only on planets that have a large moon? Perhaps. No
Moon as an object of nighttime beauty or as our nearest plan- one really knows.
etary neighbor, few of us are aware that lunar tides played a The Moon’s gravitational grip on Earth has had another
crucial early role in the drama of life. Directly and indirectly, probable effect on terrestrial life, albeit one that is a bit more
the Moon imposed its influence on the emergence and evolu- difficult to quantify. The Moon’s gravity has held Earth’s axial
tion of advanced life-forms. Life’s fi rst home was the oceans, tilt, or obliquity, relatively steady between 21.5° and 24.5° over
and instant death was in store for any sea creature having the at least the past half billion years. This stability has maintained
misfortune to be tossed onto barren land by an errant wave. But a smooth and steady transition in climate between our frigid
there were places of relative safety where life could be exposed polar regions and the warmth of the tropics, allowing for a wide
briefly to air and land before retreating to the refuge of its natu- diversity of life. Not so for Mars, though, which has no stabiliz-
ral aquatic environment. These were the tide pools. ing companion. Gravitational tugs from the Sun and planets,
The twice-daily rise and fall of sea level swept traces of among other forces, cause its axial tilt to change chaotically
sea life into tide pools, where some of these creatures spent a from as little as 0° to as much as 60° over a few tens of mil-
few adventurous hours on solid ground. Many probably per- lions of years. Without restraint from our Moon, Earth might
ished, but a few organisms were successful in making the dif- also have gyrated widely, possibly resulting in an unstable
ficult transition to dry land. The fi rst to take this enormous environment for life.
step were the plants, establishing their dominance on land So the next time you gaze at the Moon, look upon it with
about 500 million years ago. Animal life followed soon after. respect. It may be one of the reasons you are here!
sore hand, imagine the energy that it must take to compress causing it to gradually slow. Earth’s internal friction adds to
Earth by a third of a meter twice a day!) This energy is con- the even greater slowing caused by friction between Earth
verted into thermal energy by friction in Earth’s interior. This and its oceans as the planet rotates through the oceans’ tidal
friction opposes and takes energy from the rotation of Earth, bulge. As a result, Earth’s days are currently lengthening by
about 0.0015 second every century. We’ll turn now to tidal The same side of the Moon
stresses on other solid bodies beyond Earth. always faces Earth, so the
Moon does not rotate
through its tidal bulge.
The Moon Is Tidally Locked to Earth

The Moon has no bodies of liquid to make tidal stresses obvi- Moon
ous, but it is distorted in the same manner as Earth. In fact,
because of Earth’s much greater mass and the Moon’s smaller
radius, the tidal effects of Earth on the Moon are about 20
times as great as the tidal effects of the Moon on Earth. Given
Earth
that the average tidal deformation of Earth is about 30 cm,
the average tidal deformation of the Moon should be about
6 meters. However, what we actually observe on the Moon
is a tidal bulge of about 20 meters! This unexpectedly large
displacement exists because the Moon’s tidal bulge was “fro-
zen” into its relatively rigid crust at an earlier time when
the Moon was closer to Earth and tidal stresses were much
stronger than they are today. Planetary scientists sometimes
refer to this deformation as the Moon’s “fossil tidal bulge.”
This extreme deformation is the reason that the Moon
always keeps the same face toward Earth. Early in its his- The Moon is permanently
tory, the period of the Moon’s deformed and tidally locked
rotation was almost certainly Tidal forces lock the to Earth.
different from its orbital Moon’s rotation to its
FIGURE 10.9 Tidal forces due to Earth’s gravity lock the
period. As the Moon rotated orbit around Earth. Moon’s rotation to its orbital period.
through its extreme tidal
bulge, however, friction within the Moon’s crust was tre-
mendous, rapidly slowing the Moon’s rotation. After a fairly
short time, the period of the Moon’s rotation equaled the is exactly what is happening. As we can see in Connections
period of its orbit. When its orbital and rotation periods 10.2, the angular momentum lost by Earth as its rotation
became equalized, the Moon no longer rotated with respect slows is exactly equal to the angular momentum gained by
to its tidal bulge. Instead, the Moon and its tidal bulge the Moon as it accelerates in its orbit.
rotated together in lockstep with the Moon’s orbit about The acceleration of the Moon in the direction of its
Earth. With the frictional forces within the Moon gone,4 the orbital motion causes the orbit of the Moon to grow larger.
Moon’s rotation no longer slowed but instead remained equal At present, the Moon is drifting away from Earth at a rate
to the period of the Moon’s orbit about Earth. This scenario of 3.83 cm per year. That’s a very small number when com-
is how things remain today as the tidally distorted Moon pared with the Moon’s distance, so how are we able to make
orbits Earth, always keeping the same face and the long axis such a precise measurement? Going to the Moon made it
of its tidal bulge toward Earth (Figure 10.9). The synchro- possible. In 1969, the Apollo astronauts placed an array of
nous rotation of the Moon discussed in Chapter 2 is a result special reflectors on the lunar surface. Laser beams pointed
of the tidal locking of the Moon to Earth. at the Moon from Earth are reflected by these arrays and
Tidal forces affect not only the rotations of the Moon directed back to Earth, and by precise timing of a laser
and Earth, but also their orbits. Because of its tidal bulge, pulse’s round-trip, astronomers can measure the distance
Earth is not a perfectly spherical body. Therefore, the mate- between the centers of Earth and the Moon with an accu-
rial in Earth’s tidal bulge on the side nearer the Moon pulls racy of better than one part in 10 billion!
on the Moon more strongly than does material in the tidal As the Moon grows more distant, the length of the lunar
bulge on the back side of Earth. Because the tidal bulge on month increases by about 0.014 second each century. At this
the Moonward side of Earth “leads” the Moon somewhat, rate, within slightly over a bil-
as shown in Figure 10.10, the gravitational attraction of the lion years the Moon will be far The Moon’s orbit
bulge causes the Moon to accelerate slightly along the direc- enough away from Earth that is growing larger
tion of its orbit about Earth. It is as if the rotation of Earth a total solar eclipse will no lon- and Earth’s rotation
were dragging the Moon along with it, and in a sense this ger be possible. If this were to is slowing.
continue long enough (about 50
4
A small amount of tidal friction remains in the Moon. It is caused billion years), Earth would become tidally locked to the Moon,
by the Moon’s noncircular orbit and libration, as discussed in the just as the Moon is now tidally locked to Earth. At that point
next subsection. the period of rotation of Earth, the period of rotation of the
1 As in Figure 10.6, rotation 2 The near side of Earth’s tidal

(a) pulls Earth’s tidal bulge out bulge pulls on the Moon harder (b)
of alignment with the Moon. than the far side of the bulge…
3 …so a slight component of
the net force is in the direction
Fnear of the Moon’s orbit.
Moon
Moon Earth
Earth
Ffar
4 This force causes the Moon to slowly

accelerate along its orbit, causing the
orbit of the Moon to grow.
FIGURE 10.10 Interaction between Earth’s tidal bulge and the Moon causes the Moon
to accelerate in its orbit and the Moon’s orbit to grow.
CONNECTIONS 10.2
Earth, the Moon, and Conservation of Angular Momentum

Earlier in this chapter we learned that the tidal interaction tum is conserved. Where does the angular momentum that Earth
between Earth and the Moon is causing Earth’s rotation to slow loses as tidal forces slow it down go? The answer is that angular
while it causes the Moon’s orbit to grow larger. Given enough momentum is transferred to the Moon. The angular momentum
time, Earth and the Moon will become tidally locked to one that the Moon picks up as its orbit grows balances exactly the
another, with each keeping the same face toward the other. angular momentum that Earth loses as its rotation about its axis
Looking at Figure 10.10, you might imagine that it would be slows down. This may seem like a grand conspiracy, but it is not.
rather difficult to calculate the exact size of the net force caus- The frictional forces in Earth slow its rotation, but they also cause
ing the Moon’s orbit to grow, or the net frictional forces within its tidal bulge to be pulled around out of alignment with the
Earth that slow its rotation. You would be correct. However, we Moon. This misalignment in turn causes the Moon’s orbit to grow.
can understand the relationship between these two effects, as In the end it must be the case that these effects work together to
well as their end result, without having to worry about any of balance Earth’s angular-momentum loss and the Moon’s angular-
those difficult details. momentum gain. Conservation of angular momentum is a grand
As Earth’s rotation slows, Earth loses angular momentum. pattern that always holds true, whether we are talking about the
How can this be? When we discussed the formation of stars and formation of stars, rotations and orbits of planets, spinning of a
planetary systems in Chapter 6, we found that angular momen- top, or pinwheel motion of the galaxy itself.
Moon, and the orbital period of the Moon would all be exactly are in circular orbits. Even though on average an object
the same—about 47 of our present days—and the Moon would in an elliptical orbit may rotate at exactly the same rate at
be about 43 percent farther from Earth than it is today. How- which it moves in its orbit, as the object follows its orbit, it
ever, this situation will never actually occur—at least not is constantly speeding up and slowing down. This is the
before the Sun itself has burned out. case with the Earth-Moon system, in which the Moon is in
an elliptical, not circular, orbit about Earth.
Like a spinning top, the Moon rotates on its axis at a
Tides and Noncircular Orbits: steady rate that equals the average rate of the Moon’s prog-
ress around Earth. When the
An Imperfect Match Moon is closest to Earth and Tidal stresses
The effects of tidal forces can be seen throughout the Solar moving fastest in its orbit, make orbits more
System. Most of the moons in the Solar System are tid- however, the Moon’s orbital circular in shape.
ally locked to their parent planets, and in the case of Pluto motion “gains” a little on its
and its largest moon, Charon, each is tidally locked to the rotation. Conversely, when the Moon is farthest from Earth
other. The tidal locking between a moon and a planet (or and moving most slowly on its orbit, the Moon’s orbital
between any two objects) can be perfect only if both objects motion lags behind its steady rotation. From our perspec-
FIGURE 10.11 Because the Moon’s orbit is

elliptical, an observer on Earth sees lunar libration,
while an observer on the Moon sees Earth moving
slightly back and forth in the sky.
1 Because of the Moon’s
elliptical orbit…
2 …it wobbles back and 3 The apparent size of the Moon

forth slightly relative to and the features we see change
Earth’s tidal forces. slightly through the month.
tive on Earth, the face of the Moon appears to be rocking moons of the giant planets, Io is tidally locked to Jupiter,
back and forth. This effect is referred to as lunar libration. rotating once on its axis in exactly the same amount of time
The motions responsible for lunar libration are illustrated that it takes Io to complete one orbit about the planet. Nor-
in Figure 10.11. One consequence of these motions is that mally, we would have expected Io long ago to have settled
the Moon’s frozen gravitational bulge is constantly swing- into a perfectly circular orbit. But the fact that its orbit is
ing back and forth with respect to the direction of Earth’s constantly being perturbed by the gravity of Jupiter’s other
gravitational pull. Friction within the Moon, working moons has prevented it from achieving a steady circular
together with the gravitational interaction between the orbit. As a result, Io experiences libration, just as the Moon
Moon’s tidal bulge and Earth’s gravity, is slowly causing the does, but with much more dramatic consequences. As Io’s
Moon’s orbit to become more circular. libration forces it to rock back and forth through the moon’s
The masses of the giant planets are far greater than the tremendous tidal bulge, there is enough frictional heating
mass of Earth, so the tidal stresses they exert on their moons to keep the interior of Io molten, powering the perpetually
are correspondingly stronger. The tidal effect of Jupiter on active volcanism seen on this small world (see the open-
its innermost moon, Io, is 250 times greater than the effect ing photograph for this chapter and Figure 10.12). In Chap-
of Earth’s tidal stresses on the Moon. As with most of the ter 11 we will learn that Io is the most volcanically active
stresses trying to tear an object apart become greater than

the self-gravity trying to hold the object together, the object
will begin to break into pieces.
Astronomers have a special name for the distance from
the center of a planet at which tidal stresses on a second,
Plumes from active smaller body just equal that
volcanic eruptions
body’s self-gravity. It is called Tidal stresses exceed
the planet’s Roche limit. For self-gravity inside
a body having the same den- the Roche limit.
sity as the planet and having
no internal mechanical strength, the Roche limit is about
2.45 times the planet’s radius. Such an object bound together
solely by its own gravity can remain intact when it is out-
side a planet’s Roche limit, but not when it is inside the
limit.6 Tidal disruption of small bodies is thought to be the
source of the particles that make up the rings of the giant
planets. Comparison of the major ring systems around the
FIGURE 10.12 Volcanoes are constantly erupting on Jupiter’s giant planets shows that most rings lie within their respec-
moon Io. The energy to power these volcanoes comes from tides tive planets’ Roche limits.
due to Jupiter’s gravity. In the mid-1990s, astronomers were treated to a rare
look at tidal disruption in action. Comet Shoemaker-Levy
9 was discovered in March
object in the Solar System. The powerhouse behind that 1993. As astronomers studied Comet Shoemaker-
activity is tides. the comet, they realized that Levy 9 was shattered
Tidal locking, in which an object’s rotation period it had been captured by Jupi- by Jupiter’s
exactly equals its orbital period, is only one example of ter and that in July 1992 the tidal force.
spin-orbit resonance.5 Other types of spin-orbit resonance comet had passed within 1.4
are also possible. For example, Mercury is in a very ellip- Jupiter radii of the planet’s center, well within the planet’s
tical orbit about the Sun. As with the Moon, tidal stresses Roche limit. The comet was broken into pieces by this
have coupled Mercury’s rotation to its orbit. Yet unlike the encounter, so pictures taken by the Hubble Space Telescope
Moon with its synchronous rotation, Mercury rotates in a showed not a single comet, but instead a chain of orbiting
3:2 spin-orbit resonance, spinning on its axis three times fragments (Figure 10.13a). In the summer of 1994, Comet
for every two trips around the Sun. The period of Mercury’s Shoemaker-Levy 9 treated the world to an even more
orbit—87.97 Earth days—is exactly 11⁄ 2 times the 58.64 days unusual spectacle: one after another, the pieces of the comet
that it takes Mercury to spin once on its axis. Thus, each slammed into Jupiter’s atmosphere, creating enormous fi re-
time Mercury comes to perihelion—the point in its orbit balls and scars larger than Earth. Figure 10.13b shows the
that is closest to the Sun—fi rst one hemisphere and then shattered remains of another comet, Comet West, which
the other faces the Sun. was broken apart by the Sun’s tidal stresses as it passed
close to the Sun in 1976.
We have seen that tidal forces are at work throughout
the Solar System. In Connections 10.3 we fi nd they are at
Tides Can Be Destructive work throughout the universe as well.
We normally think of the effects of tidal forces as small com-
pared with the force of gravity holding an object together, yet
this is not always the case. Under certain conditions, tidal
effects can be extremely destructive. Consider for a moment 10.5 More Than Two Objects
the fate of a small moon, asteroid, or comet that wanders
too close to a massive planet such as Jupiter or Saturn. All Can Join the Dance
objects in the Solar System larger than about a kilometer
are held together by their self-gravity. However, the self- We have seen that tidal interactions can lead to simple rela-
gravity of a small object such as an asteroid, a comet, or a tionships or resonances between the orbital period of an
small moon is feeble. In contrast, the tidal stresses close to object and the period of the object’s rotation. Resonances can
a massive object such as Jupiter can be fierce. If the tidal
6
Small objects generally have mechanical strength that is much stron-
5
In a broader sense, spin-orbit resonance is a relationship between ger than their own self-gravity. This is why the International Space
the spin and orbital periods of an object such that the ratio of the two Station and other Earth satellites are not torn apart even though they
periods can be expressed as simple integers. orbit well within Earth’s Roche limit.
CONNECTIONS 10.3
Tides on Many Scales

This section of the book concerns itself with the objects that huge amounts over minute distances. In Chapter 17 we will
make up our Solar System and the processes that influence them. fi nd that the tidal effects near the surface of a black hole can
Therefore, it is fitting that we have concentrated on the role that be billions of times stronger than the force of gravity holding
tidal forces play in the Solar System. On the other hand, we us to the surface of Earth. Such tidal stresses are enough to rip
could have discussed tides and their effects at almost any point normal matter apart atom by atom.
in the book. Tidal stresses result from differences in gravitational Tides continue to play a role at even larger scales. Tidal
forces from place to place, and such differences are a general effects can strip stars from clusters consisting of thousands
consequence of the inverse square law of gravity. Any time two of stars. Beginning in Chapter 18, we will also learn about
objects of significant size or two collections of objects interact galaxies—vast collections of hundreds of billions of stars that
gravitationally, the gravitational forces will differ from one place all orbit each other under the influence of gravity. Often two
to another within the objects, giving rise to tidal effects. galaxies pass close enough together to strongly interact gravi-
We might easily have included a discussion of tidal forces tationally. When this happens, as in Figure 10.14, both galaxies
in Chapter 13, where we will fi nd that stars are often members taking part in the interaction can be grossly distorted by tidal
of binary pairs in which two stars orbit each other. Strong effects. Tidal forces even play a role in shaping huge collections
tidal interactions between such stars can tidally lock the rota- of galaxies— the largest known structures in the universe.
tion of each star to its orbital period about its companion, and The next time you are at the beach “watching the tide roll
can even pull material from one star onto the other. When we in,” you might take a moment to think about tides and the role
study this process in Chapter 16, we will learn of the Roche they play throughout the universe. Vast collections of billions
lobes of a binary star. These are named after Édouard A. Roche of stars strewn about millions of light-years of space, mat-
(1820–1883), the French scientist who did pioneering work ter shredded on its way into a black hole, gas stripped from a
on the disruption of satellites and for whom the Roche limit burgeoning star, comets and moons pulled to pieces, planets
is named. Under some circumstances in the universe, tidal “fl ipped” like pancakes— all are the result of the same gravita-
stresses become unimaginably powerful. In the vicinity of very tional effect that is responsible for the gentle twice-daily rise
massive, very dense objects the pull of gravity can change by and fall of the waterline on an oceanside dock.
1 Tidal interactions between galaxies 2 …and distort the inner parts of

draw stars into long tails… the galaxies as well.
FIGURE 10.14 Tidal interactions distort the appearance of two galaxies. The tidal “tails” seen
here are characteristic of tidal interactions between galaxies.
FIGURE 10.13 (a) Comet

(a) Fragments of Comet
Shoemaker-Levy 9 was
Shoemaker-Levy 9
shattered into pieces as it
passed within Jupiter’s Roche
limit in July 1992. (b) Comet
West was shattered by solar
tidal stresses during its 1976
pass through the inner Solar
System.
(b) March 8 March 12 March 14 March 18 March 24
also exist between the orbital periods of two or more bodies nance. (Note that this is different from spin-orbit resonance,
orbiting about the Sun or a planet. We’ll now see how tidal which we discussed in Section 10.4.)
interactions affect the orbits of asteroids and the multitude The easiest way to understand the origin of the Kirkwood
of particles that make up planetary rings. gaps is to imagine what would happen to an asteroid that
started out with an orbital period exactly half that of Jupiter.
Follow along in Figure 10.16 as we watch the orbit of such
an asteroid, beginning when the asteroid and Jupiter are at
Gaps in the Asteroid Belt their closest to one another. Because the asteroid is closer
In Chapter 12 we will study the smaller objects that orbit to the Sun than Jupiter is, its orbital velocity is greater than
about the Sun, including the small rocky or metallic worlds that of Jupiter, so it leaves Jupiter behind as they continue
called asteroids. Most of the asteroids in the Solar System on their respective orbits. By the time the asteroid has com-
are located in a “belt” between the orbits of Mars and Jupi- pleted half of its orbit, Jupiter has completed only a fourth of
ter. You might imagine that asteroids would be distributed its orbit. By the time the asteroid has gone around its orbit
randomly throughout this region. However, if you look at once, Jupiter is halfway. When the asteroid has completed
the distribution of the sizes of the orbits` of the asteroids in one and one-half orbits, Jupiter has been through three-
the main belt, shown in Fig- quarters of its orbit. Only after the asteroid has made two
ure 10.15, you will see an Orbital resonances complete orbits and Jupiter has been around the Sun once
amazing thing. Rather than with Jupiter cause the do the two objects again line up. Thanks to the relationship
asteroids having orbits of all Kirkwood gaps. between the periods of their two orbits, when Jupiter and
sizes in this region, there are the asteroid line up they are in the same place where they
certain-sized orbits that are seldom occupied. These gaps started. As Jupiter and the asteroid continue in their orbits,
in the asteroid belt are referred to as the Kirkwood gaps they line up at this same location again and again, every
after Daniel Kirkwood (1814–1895), the American astrono- 11.86 years (the orbital period of Jupiter).
mer who fi rst recognized them. Astronomers studying the The gravitational force that Jupiter exerts on an asteroid
Kirkwood gaps found a simple relationship between the at its closest approach is tiny compared with the gravita-
“missing” orbits and the orbit of Jupiter. For example, none tional force of the Sun on the asteroid, which is over 360
of the asteroids orbiting the Sun have an orbital period equal times stronger. A single close pass between Jupiter and the
to half of Jupiter’s orbital period. In fact, all of the Kirkwood asteroid does very little to the asteroid’s orbit. For an asteroid
gaps in the asteroid belt correspond to orbits that are related that is not in orbital resonance with Jupiter, the tiny nudges
to the orbital period of Jupiter by the ratio of two small inte- from Jupiter come at a different place in its orbit each time.
gers. Such a simple relationship between the periods of the The effects of these random nudges average out, and as a
orbits of two or more objects is referred to as an orbital reso- result even multiple passes close to Jupiter have little over-
3:1 Gap 5:2 Gap of Jupiter. If we were to put an asteroid in such an orbit, it
An asteroid orbits 3 times An asteroid orbits 5 times would not stay there long. The same phenomenon works for
for each orbit of Jupiter. for each 2 orbits of Jupiter. other combinations of orbital periods as well. With a piece
of paper and a pencil, you ought to be able to convince your-
self that other orbital resonances, such as a 3:1 resonance
Orbital resonance 3:1 5:2 7:3 2:1
(in which an asteroid completes three orbits about the Sun
in the time that it takes Jupiter to complete one orbit), will
140 have a similar effect. The reason asteroids are not found in
the Kirkwood gaps is that their gravitational interaction
120
with Jupiter prevents them from staying there.
Number of asteroids
100
80 Gravity and Ring Systems

60 Asteroids are not the only objects in the Solar System that
are affected by orbital resonances. Exactly the same phe-
40 nomenon occurs in the ring systems around the giant plan-
ets. These ring systems consist of countless numbers of
20 small particles—varying in size from tiny grains to house-
sized boulders—that orbit individually around the giant
0 planets like so many tiny moons. Kepler’s laws dictate that
2 2.2 2.4 2.6 2.8 3 3.2 3.4
the speed and orbital periods
Semimajor axis (AU)
of all ring particles must vary Rings are swarms
FIGURE 10.15 This plot shows the number of asteroids in the with their distance from the of tiny moons
main belt with a given orbital period. The gaps in the distribution
planet, with the closest mov- orbiting according to
of asteroids, called Kirkwood gaps, are caused by orbital
ing the fastest and having the Kepler’s laws.
resonances with Jupiter.
shortest orbital periods. The
orbital periods of particles in Saturn’s bright rings, for
all effect on the asteroid’s orbit. As we just saw, however, example, range from 5 hours 45 minutes (5h45m) at their
for an asteroid that has a 2:1 orbital resonance with Jupiter inner edge to 14h20m at the outer one. In the densely packed
(in other words, it has an orbital period equal to half that rings of Saturn, the orbits of all of these particles are per-
of Jupiter), the nudge from Jupiter comes at the same place fectly circular and in precisely the same plane.
in its orbit every time. To understand why the orbits of particles in dense rings
Although no single tug has much influence on the aster- must be so orderly, imagine yourself riding along on a large
oid, the repeated tugs from Jupiter at the same place add particle in the middle of Saturn’s bright rings. Around
up, changing the asteroid’s orbit. This is why there are no you is a swarm of particles of all shapes and sizes, many
asteroids with orbital periods equal to half the orbital period of them close enough for you to reach out and touch. They
FIGURE 10.16 An asteroid with an orbital period exactly half that of Jupiter would line up with Jupiter at exactly the same place
in every second orbit. The repeated gravitational pull from Jupiter at the same point in the asteroid’s orbit would nudge the asteroid
enough to change its orbital period. No asteroids have a 2:1 orbital resonance with Jupiter.
1 An asteroid in an orbit with a 2 ...in the time it takes Jupiter 3 After the asteroid completes two
semimajor axis of 3.276 AU to complete one orbit. orbits and Jupiter completes one orbit,
completes two orbits... Jupiter and the asteroid line up in the
same places where they started.
Asteroid orbit 1 Asteroid orbit 2
Jupiter orbit
Jupiter Asteroid
Sun
are all moving along in almost perfect step with your own, small telescopes. The Cassini Division corresponds to a 2:1
like members of a well-disciplined marching band. We say orbital resonance with Mimas. That is to say, a ring particle
“almost” because the particles slightly closer to the planet located in the Cassini Division would have an orbital period
are moving just a bit faster than you are, and those a little about the planet equal to half the orbital period of Mimas.
farther out are moving just a bit slower, as Kepler’s laws Just as Jupiter’s gravity pulls asteroids out of resonant orbits,
demand. The differences in speed are very small. The par- leading to the Kirkwood gaps, Mimas pulls ring particles
ticle orbiting 1 meter inward from you is moving only a tenth out of the resonant orbits found within the Cassini Division.
of a millimeter per second faster than you are— a relative Such resonances with other moons are known to produce
speed difference slower than a snail’s pace. some of the gaps that appear in Saturn’s bright rings.
Now consider what would happen if that particle were Many planetary rings have sharp edges with no visible
moving in a slightly noncircular orbit, perhaps because it material appearing in the space beyond. For these rings,
was bumped by one of its other something is holding most of
neighbors. In response to its Particles in dense the particles in place, pre- Moons maintain sharp
orbital eccentricity, it would rings must move in venting the rings from quickly ring edges and create
drift alternately a bit outward circular orbits and dissipating. The key to the gaps in the rings.
and then inward as it com- travel in unison with sharp edges and stability of
pleted each orbit. But as it nearby particles. the rings lies with the same sort of orbital resonance that
moved outward, the particle causes the Cassini Division. As moons orbit a planet, their
would eventually nudge up against you and be unable to go gravity can nudge ring particles that are in resonant orbits,
any farther. With no room for maneuvering, the orbits of keeping them in line. Consider, for example, the abrupt outer
errant particles would (and do) quickly become circular. The edge of Saturn’s outermost bright ring. The ring particles at
same would be true for any particle in an orbit with even the the edge of this ring are in a 7:6 orbital resonance with Sat-
smallest inclination: During each orbit, the deviant ring frag- urn’s co-orbital moons Janus and Epimetheus, meaning that
ment would be carried alternately above and below the ring the ring particles make precisely seven orbits for every six
plane. As it approached your particle from, say, above, the orbits of its two co-orbital moons. (“Co-orbital” moons are
mild encounter would quickly remove its out-of-plane motion moons that occupy the same orbit.)
and bring it into a coplanar (in the same plane) orbit. Saturn’s Now picture a particle suffering a collision and being
densely packed rings have no place for nonconformists. Like forced a little outside the edge of the ring. Each time Janus
the members of a well-disciplined marching band, all par- passes the particle, once out of every six Janus orbits and
ticles must march along together in unison. every seven particle orbits, it does so at the same position
In addition to these interactions with each other, the in the particle’s orbit. And each time the moon passes, the
orbits of these small particles deep within a ring can be particle receives a small gravitational nudge from the moon
influenced by the planet’s larger moons. All that is neces- in the same direction, causing the particle to lose a bit of
sary is for the moon to be massive enough to create a sig- its orbital energy. The effects build up, and eventually the
nificant gravitational tug on the particles and be in orbital particle loses enough energy to fall back into the ring. In
resonance with them. Such is the case with Saturn’s moon short, the sharp edge of the ring itself is just like the sharp
Mimas, which causes the famous gap in the rings around edge of a resonance gap such as the Cassini Division.
Saturn called the Cassini Division (Figure 10.17). This gap is Each time a moon nudges a ring particle inward, the
so prominent that it can be seen easily from Earth, even with minute amount of energy lost by the particle is absorbed by
FIGURE 10.17 Saturn and

its rings. The Cassini Division is
easily seen here as the widest
gap in Saturn’s bright ring
system.
the more massive moon, causing it to move imperceptibly

farther from the ring and the planet. Over time, a moon may Gravity Affects the Orbits
move so far from the ring edge of Three Bodies
that it can no longer provide Orbital resonances
stability to the ring. The ring shape Saturn’s rings. One of the most interesting examples of orbital resonance
is then free to dissipate. In the occurs when two massive bodies (such as a planet and the
case of Saturn’s bright rings, the situation is even more com- Sun, or a moon and a planet)
plicated because Janus itself is also in resonance with other move about their common Lagrangian points are
moons of Saturn, meaning that all of these moons must be center of mass in circular or orbital resonances
pushed away before the rings can dissipate. Thus, the bright nearly circular orbits. In this formed by the gravity
rings of Saturn may be much more stable than most rings. situation, there are five loca- of two objects.
Other kinds of orbital resonances are possible in ring tions where the combined
systems. For example, most narrow rings are caught up in gravity of the two bodies adds up in such a way that a third,
a periodic gravitational tug- lower-mass object will orbit in lockstep with the other two.
of-war with nearby moons. Moons that keep These five locations, called Lagrangian equilibrium points
These are called shepherd narrow rings from or simply “Lagrangian points,” are named for the Italian-
moons in recognition of the spreading are called born French mathematician Joseph-Louis Lagrange (1736–
way they shepherd a flock of shepherd moons. 1813), who fi rst called attention to them.
ring particles. Shepherd The exact locations of the Lagrangian points depend
moons are usually small, are located close to a narrow ring, on the ratio of the masses of the two primary bodies. The
and often come in pairs, one orbiting just inside and the Lagrangian points for the Earth-Moon system are shown
other just outside the narrow ring. The shepherding mecha- in Figure 10.18a. (Remember that the pattern shown in the
nism is much like the resonances discussed earlier. A shep- figure rotates like a turntable as the two massive objects
herd moon just outside a ring will rob orbital energy from orbit each other.) Three of the Lagrangian points lie along
any particles that drift outward beyond the edge of the ring, the line between the two principal objects. These points
causing them to move back inward. A shepherd moon just are designated L1, L2, and L 3. Although they are equilib-
inside a ring will give up orbital energy to a ring particle rium points, these three points are unstable in the same
that has drifted too far in, nudging it back in line with the way that the top of a hill is unstable. A ball placed on the
rest of the ring. In some cases, narrow rings are trapped top of a hill will sit there if it is perfectly perched, but
between two shepherd moons in slightly different orbits. give it the slightest bump and it goes rolling down one
Planetary rings are a virtual laboratory for studying side. Similarly, an object displaced slightly from L1, L2,
such orbital resonances, as well as tidal stresses and similar or L 3 will move away from that point. Thus, although L1,
gravitational interactions. Tidal stresses, for example, are L2, and L 3 are equilibrium points, they do not capture and
thought to be largely responsible for much of the material hold objects in their vicinity. Even so, they can be use-
found in planetary rings. If a moon or other planetesimal ful. For example, as of this writing, the SOHO (Solar and
held together by gravity alone happens to be within the Heliospheric Observatory) and ACE (Advanced Composi-
Roche limit of a giant planet, it will be pulled apart by tidal tion Explorer) spacecraft sit near the L1 point of the Sun-
stresses. The location of Saturn’s bright rings inside the plan- Earth system, where tiny nudges from their onboard jets
et’s Roche limit, for example, suggests that the rings formed are enough to keep them orbiting directly between Earth
from the breakup of one or more moons or captured plan- and the Sun, giving a constant, unobscured view of the side
etesimals that were perturbed by larger moons into orbits of the Sun facing Earth.
that took them too close to the planet. If rings are made of The situation is different for the other two Lagrangian
material that originated on moons, we might expect to fi nd points: L 4 and L 5. These two points are located 60° in front
the composition of a planet’s rings mimicking the makeup of and 60° behind the less massive of the two main bodies
of the moons that surround the planet. Such is indeed the in its orbit about the center of mass of the system. These
case, as we will see in the next chapter. are stable equilibrium points, like the stable equilibrium
Jupiter’s ring system turns out to be largely the product that exists at the bottom of a bowl. If you bump a marble
of the strong gravity of the planet itself, combined with the sitting at the bottom of a bowl, it will roll around a bit but
presence of a handful of small, rocky moons close to the it will stay in the bowl. In like fashion, objects near L 4 or
planet. As interplanetary meteoroids are pulled toward L 5 follow elongated, tadpole-shaped paths relative to these
Jupiter by its strong gravity, a few of them strike the sur- gravitational “bowls.”
face of one of Jupiter’s four innermost moons. These moons Under the correct circumstances, L 4 and L 5 are able to
are so tiny that some of the dust from these impacts is capture and hold on to passing objects. Objects are often
kicked off with speeds exceeding the escape velocities of found orbiting about the stable L 4 and L 5 Lagrangian points
the moons. This dust provides a steady supply of material of two-body systems. For example, the stable Lagrangian
for the rings. points of the Sun-Jupiter system are home to a collection of
10.6 Such Wondrous Complexity Comes from Such a Simple Force 293
A less massive object in a Lagrangian

equilibrium point will orbit in lockstep
with the two more massive objects
(in this case, Earth and the Moon).
(a) Moon L2 (b) Jupiter (c)
Trojan
L1 asteroids Tethys
L4 L5
60° 60°
60° 60° 60° 60°
Tadpole
paths Earth Saturn
Sun Dione
60°
Lagrangian
moons
L3
Trojan asteroids occupy the stable Tiny moons occupy the L 4 and
L 4 and L 5 Lagrangian points L 5 points formed by Saturn and
formed by the Sun and Jupiter. its moons Tethys and Dione.
FIGURE 10.18 (a) The pattern of Lagrangian equilibrium points of a two-body system, in this case composed of Earth
and the Moon. This entire pattern of Lagrangian points rotates like a turntable. (b) Trojan asteroids share Jupiter’s orbit. (c)
Small moons occupy Lagrangian points formed by Saturn and Tethys and by Saturn and Dione.
planetesimals called the Trojan asteroids (Figure 10.18b). It is with regret that we leave behind the gravitational
There are also small moons in the L 4 and L 5 Lagrangian ballet, but move on we must. Before departing, however,
points of Tethys and Dione, two moons of Saturn (Figure it is worth pointing out just how broad and deep the gen-
10.18c). (A group that formed in the 1970s and advocated eral problem of gravitational interactions is. Although
human colonization of space called itself the “L5 Society” in the motion of two masses orbiting about each other was
recognition of the fact that the L 4 and L 5 Lagrangian points solved centuries ago by Isaac Newton, the general prob-
of the Earth-Moon system would make excellent locations lem of the motion of three bodies remains unsolved today!
for large orbital colonies.) More than a few PhD theses, scholarly journal articles,
and monographs in mathematics, astronomy, and phys-
ics have been fi lled with years of work on various aspects
of this seemingly simple problem. Even in this day of
10.6 Such Wondrous powerful computers, which can crank out brute-force
solutions to problems that seemed unapproachable a few
Complexity Comes from decades ago, we remain humble in the face of gravity. As
discussed in Excursions 10.1, even the gravitational inter-
Such a Simple Force actions between a few objects such as the planets in our
Solar System are chaotic. Unbelievably tiny differences
The tidal forces and resonances that we have discussed in in the initial positions and velocities of the planets—
this chapter are but a small sample of the tremendous wealth differences that are far too tiny to keep track of with the
and variety of phenomena best computers— can, over time, lead to dramatically dif-
that result from the gravita- Gravity causes ferent end results.
tional interactions between the precession of As we leave this interlude about gravity behind,
extended objects or collec- Earth’s axis. we can only marvel at the extraordinary complexity
tions of more than two objects. and diversity that arise in nature from this deceptively
We might have mentioned many other such phenomena. simple force. What an amazing array of implications,
The 26,000-year precession of Earth’s axis, for example, is insights, puzzles, and surprises there are within New-
a result of the Sun’s gravity causing Earth’s axis to wobble ton’s elegant statement that the force of gravity is pro-
like a top’s. How a gravitational tug in one direction can portional to the product of the masses of two objects
cause the axis of a planet (or a top) to move in another direc- and inversely proportional to the square of the distance
tion is a truly fascinating story. between them!
E X C U R S I O N S 1 0 .1
A World of Chaos
Imagine that you head for class but you leave 1 second later again and again. Pasiphae’s orbit is an example of chaos. Cha-
than you intended. That single second causes you to get caught otic orbits are complex, irregular motions in which extremely
at a traffic light, delaying you enough that you wind up fi rst in small differences in an object’s position or speed grow into
line waiting for a passing train. While you await the train, an large differences in the object’s subsequent motion.
accident occurs up ahead, and as you sit in the resulting traf- Capture of moons by planets is another example of chaos
fic jam, your car’s engine overheats. Eventually, while your at work. Figure 10.19b shows the path of a small planetesimal,
classmates are listening to an inspiring lecture about orbital such as a comet, as it approaches a planet like Jupiter. Very tiny
motion, you fi nd yourself sitting in a mechanic’s waiting room differences in the position of the comet lead to very different
learning the hard way about how much expensive damage you outcomes. In the illustration there is a gnat’s eyelash of differ-
can do to an engine by letting it get too hot. You do not real- ence between the starting point of a path in which the comet
ize it, but the only thing that separated you from a normal day impacts the planet, a path in which the comet swings around
in class was the extra sip of coffee that put you out the door a Jupiter and continues to orbit the Sun, and a path that winds
second later than you might have been. You could never have up with the planetesimal orbiting the planet. Each of the giant
predicted such a thing would happen. Welcome to the wonder- planets has a handful of moons that revolve about the planet
ful world of chaos. in the wrong direction, indicating that they must have been
Chaos is the technical term to describe interrelated systems captured in such chaotic events.
in which tiny differences at one point in time lead to large dif- Chaos also affects the orbits of planets themselves. Each
ferences at later times. Any system capable of exhibiting chaotic planet in our Solar System moves under the combined gravita-
behavior is referred to as a complex system. Complex systems tional influence not only of the Sun, but of all the other planets
need not be extremely complicated, like the example in the as well. Although these extra influences are small, they are not
previous paragraph. They can be seemingly very simple. New- negligible. Over millions of years they lead to significant dif-
ton’s law of gravity is simplicity itself, and as long as only two ferences in the locations of planets in their orbits. Although
objects are involved, the resulting motions are simple as well. analysis of our Solar System indicates that the orbits of all
When we think of a planet orbiting the Sun, we think of the eight classical planets are fairly stable, at least over times of a
regular, repeating elliptical orbits described by Kepler’s laws. few billion years, that is not always the case. In many possible
When more than two objects are involved, however, the result- planetary systems, chaotic interactions among planets might
ing motions can be anything but simple and regular. cause planets to dramatically change their orbits or even be
Figure 10.19a shows a calculation of the orbit of Jupiter’s ejected from the system entirely.
moon Pasiphae over an interval of approximately 38 years. In Chapter 6 we discussed the fact that planetary systems
Pasiphae orbits far enough from Jupiter that its orbit is strongly have been found around many other stars. Some of these sys-
perturbed by the Sun’s gravity. Rather than a simple ellipse, tems contain giant planets that seem much too close to their
Pasiphae’s orbit looks more like what a 5-year-old might pro- parent stars to have formed there. Models suggest that these
duce if handed a crayon and asked to quickly trace out a circle planets might have formed farther out in these systems and then
not only the simple elliptical orbits of planets about the

Seeing the Forest for the Trees Sun, but also an amazing wealth of different phenom-
ena. The coming and going of the tides, the synchronous
In Chapter 3 we learned about the birth of modern sci- rotation of the Moon, the exquisite structure of Saturn’s
ence. We saw how a few straightforward ideas about rings—all of these and more are the logical consequences
forces and the inverse square law of gravity opened a of the fact that gravity is one-fourth as strong when two
new window on the universe. When Newton used his objects are twice as far apart. The insights that we have
laws of motion and gravitation to explain the motions of arrived at in this chapter will have profound implica-
the planets about the Sun, he changed the face of science tions again and again as we continue our journey. For
and the course of history. example, when we follow the birth, evolution, and death
On this leg of our journey we have pursued the impli- of stars like our Sun, much of what we will see will be
cations of Newton’s insight even further. We have seen determined by the sizes and masses of the cores of those
how that same inverse square law of gravitation explains stars. Because of what we have learned in this chapter, we
10.6 Such Wondrous Complexity Comes from Such a Simple Force 295
moved inward toward the stars as a result of chaotic interac- the fi rst to recognize the existence of chaotic behavior, and in
tions. Indeed, events such as the collision between Earth and the last decades of the 20th century they began to develop the
a Mars-sized body that formed the Moon tell of a more chaotic mathematical tools needed to describe and even exploit chaos.
period early in the history of our own Solar System. These tools are fi nding more and more common applications
A key characteristic of chaotic systems is that although in our lives. The beating human heart, weather patterns, and
they are deterministic, they are not predictable. The law of traffic control are all examples of complex systems to which
gravity is well known, and there is in principle no problem these tools have been applied. Pacemakers use these techniques
with calculating the path that Earth will follow as a result of its to recognize when the regular beating of a heart is about to
ongoing interactions with other planets. It is a perfectly well- give way to the deadly, uncoordinated fluttering called “fibril-
determined situation. However, even Earth’s orbit is subject to lation.” A tiny nudge on the part of the pacemaker, slightly
chaos. An uncertainty of only 1 centimeter in our knowledge changing the timing of a handful of heartbeats before fibrilla-
of the position of Earth along its orbit today is enough to make tion begins, is enough to prevent disaster. Chaos— the physics
it completely impossible to predict where Earth will be in its of complexity—is but one of many examples in which studies
orbit 200 million years from now. of the universe have spilled over into better understanding of
Scientists studying the orbits of celestial bodies were among the immediate world in which we live.
Far from Jupiter, Pasiphae orbits Chaos means that tiny differences
chaotically under the influence at one time can lead to huge
of Jupiter and the Sun. differences later.
(a) Pasiphae (b)

Starting
points
Comet orbits
Jupiter
Jupiter Sun
0 5 10 15 20 25 FIGURE 10.19 (a) The chaotic orbit of Jupiter’s moon Pasiphae. (b) Tiny differences in the
Millions of km motion of a comet lead to dramatic differences in the comet’s fate.
will be able to say that the force of gravity at the surface ter with the single statement about the inverse square
of a stellar core is determined by the core alone, regard- law of gravitation, and went on to tell stories ranging
less of what the rest of the star is doing. from the ebb and flow of Earth’s tides to the shredding
Even more important, this leg of our journey has of distant galaxies.
shown us again the aesthetics of science. If you ask At the turn of the 21st century, physicists have come to
a physicist today about the ultimate goal of physics, believe that even the basic physical laws that we are using
you will hear about the desire to explain the way the on our journey—Newton’s laws, relativity, gravitation,
universe behaves in terms of the fewest and simplest and quantum mechanics— are themselves embodiments
possible physical laws. In Chapter 3 we discovered of patterns in nature that are more fundamental still.
Newton’s laws of motion. In a sense, once we know Toward the end of our journey we will learn that today’s
those laws, everything—from the fl ight of a bumblebee, frontiers of science are located at the two extremes. Theo-
to the swirl of a hurricane, to the motions of an Olym- retical physicists and cosmologists push ever backward
pic gymnast—is just arithmetic. We began this chap- in time toward the beginning of the universe in search
of the single theory—the “Theory of Everything”—from as the those describing chaos, to enable them to find the
which all else follows. Some scientists believe we may hand of simple physical law in all that we see. So the
be just a few decades (or perhaps only a few years) from next time you find yourself looking up at the Moon, gaz-
knowing the shape of such an all-encompassing theory. ing on the same face of our sister world that our ances-
We may even come to know whether it was possible for tors saw from time immemorial, think about gravity and
the universe to become other than it is. tides and the interplay between sister worlds. But at the
At the same time, physicists, astronomers, planetary same time, think about what they represent—the glori-
scientists, biologists, and others are pushing forward ous complexity of this fascinating universe in which we
with their understanding of the complex world around live, and the beautiful simplicity of physical law from
us. They are developing new tools and approaches, such which it all derives.
Summary ing in the International Space Station. Explain why this

statement is true.
• Planets and moons are held together by self-gravity. 4. What determines the strength of gravity at various radii
between Earth’s center and its surface?
• Both the Moon and the Sun create tides on Earth.
5. The best time to dig for clams along the seashore is when
• As Earth rotates, tides rise and fall twice each day.
the ocean tide is at its lowest. What phases of the Moon
• Tidal forces lock the Moon’s rotation to its orbit around and times of day would be best for clam digging?
Earth.
6. The Moon is on the meridian at your seaside home, but
• Tidal forces cause the Moon’s orbit to grow and Earth’s your tide calendar does not show that it is high tide.
rotation rate to slow. What might explain this apparent discrepancy?
• Tidal forces make orbits circular. **7. We may have an intuitive feeling for why lunar tides
raise sea level on the side of Earth facing the Moon, but
• Tidal stresses are stronger than self-gravity inside the
why is sea level also raised on the side facing away from
Roche limit.
the Moon?
• Tidal forces create gaps in the distribution of asteroid
8. Both the Sun and the Moon raise tides on Earth. Explain
orbits.
how the phase of the Moon affects the total strength of
• Tidal forces control the orbits of particles in planetary the tides.
rings.
9. Tides raise and lower the level of Earth’s oceans. Can
they do the same for Earth’s landmasses? Explain your
SmartWork, Norton’s online homework system, answer.
includes algorithmically generated versions of these 10. If lunar tides cannot raise the ocean surface more than
questions, plus additional conceptual exercises. If 1 meter above mean sea level, how can tides as large as
your instructor assigns questions in SmartWork, log in at 5–10 meters occur?
11. If the Moon spins on its axis once every 291⁄2 days rela-
tive to Earth, why are there not Earth tides on the Moon
rising and falling over this same interval?
Student Questions
12. Tidal friction slows Earth’s rotation and causes the Moon
THINK ING A BOUT THE CONCEP T S to orbit ever farther from Earth. Is this an example of
conservation of angular momentum or conservation of
1. Defi ne what we mean by self-gravity. orbital energy? Explain your answer.
2. In our daily lives we encounter many types of symmetry, 13. What is meant by tidal locking?
including rotational, reflection, and bilateral (the case
*14. In 1959, a spacecraft launched by the Soviet Union pho-
in which left and right sides are identical). List some
tographed the far side of the Moon, showing us a large
ordinary examples of each of these types of symmetry.
part of the Moon’s surface that had never before been
3. If you could live in a house at the center of Earth, you seen. Yet more than half (59 percent) of the Moon’s sur-
could float from room to room as though you were liv- face had already been mapped prior to the space age. If
the Moon always keeps the same face toward Earth, how Saturn than does Hyperion. The orbital period of Titan
was this possible? is approximately 16 days. What is the orbital period of
Hyperion?
*15. Could there be tidal heating effects on a moon located
in a perfectly circular orbit around a planet? Explain *30. Tidal influence is proportional to the mass of a disturb-
your answer. ing body and is inversely proportional to the cube of the
distance to that body. Some astrologers claim that your
16. How is tidal stress responsible for powering volcanic
destiny is determined by the “influence” of the planets
activity on Jupiter’s moon Io?
that are rising above the horizon at the moment of your
17. Small bodies orbiting large bodies within the Roche birth. Compare the tidal influence of Jupiter (mass 1.9 ×
limit can be torn apart. Explain why this happens. 1027 kg; distance 7.8 × 1011 meters) with that of the doc-
tor in attendance at your birth (mass 80 kg, distance 1
18. Most commercial satellites orbit Earth well inside the
meter).
Roche limit. Why are they not torn apart?
31. What is Earth’s Roche limit, expressed in kilometers, for
19. There is a region in the main asteroid belt (about 3.28
a body having the same density as Earth and no internal
astronomical units [AU] from the Sun) where the orbital
strength? (Earth’s mean radius is 6,371 km.) Compare
period is exactly half that of Jupiter’s. Explain why no
this distance with the orbital radius of the International
asteroids are found there.
Space Station, which circles Earth at an average alti-
20. Explain why all of the particles in Saturn’s bright rings tude of 350 km.
remain in circular orbits and in the same plane.
32. Saturn’s radius is approximately 60,000 km, and the
21. Explain why Saturn’s bright ring system ends abruptly outer radius of its bright ring system is approximately
at its outer edge. 137,000 km. Show that Saturn’s bright ring system falls
within Saturn’s Roche limit.
22. What are shepherd moons, and what is their relation-
ship to planetary rings? **33. The mass of the Sun is 30 million times greater than the
mass of the Moon, but the Sun is about 400 times farther
*23. Explain the significance of the five Lagrangian points.
away. Show why solar tides on Earth are only about half
24. Which of Earth’s five Lagrangian points can capture and as strong as lunar tides.
hold passing objects?
34. Calculate the orbital radius of the Kirkwood gap that is
**25. Even our most powerful computers have limited ability in a 3:1 orbital resonance with Jupiter. (Hint: Refer to
to predict the orbits and locations of the various bod- the discussion of Kepler’s third law in Chapter 3. If you
ies in our Solar System in the far distant future, and no cannot do the calculation, show how you would set it
matter how powerful we make our computers there will up.)
always be uncertainties. Why?
35. The radius of the outer edge of Saturn’s bright ring sys-
tem is 136,755 km, and the radius of the orbit of Janus
is 151,460 km. Using tools learned in Chapter 3, show
A P P LY I N G T H E C O N C E P T S that particles at the edge of the bright ring system are
in a 7:6 orbital resonance with Janus.
26. Assume for the moment that Earth is homogeneous (hav-
36. The L 4 and L 5 Lagrangian points form equilateral tri-
ing constant density throughout) and spherical. (We
angles with the Sun and Earth (see Figure 10.18). If we
know, of course, from Chapter 7 that neither of these
were ever to colonize L 5, how far would the space pio-
assumptions is true.) If you were in a deep well halfway
neers making a home at L 5 be living from their previous
to Earth’s center, how would your weight there compare
homes on Earth?
with your weight at Earth’s surface?
27. If the Moon had twice the mass that it does, how would
the strength of lunar tides change?
*28. At some time in the past, the Moon was only two-thirds
its present distance from Earth. How would lunar tides
at that time compare with those of today?
29. Saturn’s small moon Hyperion is in a 4:3 orbital reso- quizzes, plus links to premium content in SmartWork and
nance with its largest moon, Titan, which orbits closer to the ebook. Visit www.wwnorton.com/studyspace.
Although we are mere sojourners on the surface of the planet,
chained to a mere point in space, enduring but for a moment
in time, the human mind is not only enabled to number worlds
beyond the unassisted ken of mortal eye, but to trace the
events of indefinite ages before the creation of our race.
SIR CHARLES LY ELL (1797–1875)
CHAP TER 1
Complex structure in Saturn’s three bright rings, as seen by the Cassini spacecraft.
CHAP TER 11
Planetary Adornments—
Moons and Rings
11.1 Galileo’s Moons
K E Y CONCEP T S
In 1610 the Italian astronomer Galileo Galilei observed that
Jupiter was accompanied by four “stars” that changed their For centuries, celestial wonders such as Saturn’s rings
positions nightly. He quickly realized that, like Earth, Jupi- and the Galilean moons of Jupiter delighted those who
ter has moons of its own. We honor his discovery by call- looked through telescopes. But with the dawn of the
ing them the “Galilean moons” of Jupiter. Galileo showed space age, robotic explorers traveling through the Solar
that Jupiter and its moons resemble a miniature Copernican System have shown us much more than points of light,
planetary system: just as planets revolve around the Sun, revealing wondrous, diverse families of worlds orbiting
so do the many moons of Jupiter revolve around it. By the other planets. Among them we will find
end of the 17th century, astronomers had found five moons
orbiting around Saturn as well, and this discovery further • Scores of small worlds composed of rock and solid ice,
strengthened their belief in the Copernican system (see some of which formed with their planets and others of
Chapter 3). Today we realize that the Solar System abounds which were captured later.
with moons. As of August 2009, we count more than 170 • Geologically active moons freckled with volcanoes and
observed moons in orbit around the classical and dwarf geysers, and geologically dead moons covered with
planets, and there are probably many others— especially in impact craters.
the outer Solar System—that we have not yet found. Some
of these moons are unique worlds of their own, exhibiting • Moons that may harbor deep liquid oceans beneath
geological processes similar to those on the terrestrial plan- their ice-covered surfaces and may conceivably provide
ets, and some may even harbor forms of life. a home for extraterrestrial life.
The moons of our Solar System are not distributed • Exquisite, delicate structure in ring systems resulting
equally among the planets. In the inner part of the Solar from subtle gravitational interactions among planets,
System there are only three moons; Earth has one, Mars has moons, and ring particles.
two, and Mercury and Venus have none. Among the dwarf
planets, Pluto possesses three known moons, Haumea has • The fate of most planetary rings, short-lived compared
two, and Eris has one. All of the remaining moons belong to the planets they surround.
to the giant planets. We might ask why Mercury and Venus
failed to form or capture any moons of their own. But is
this really surprising? Mercury and Venus, after all, have
only one less moon than Earth has. Mars apparently cap-
tured two small asteroids, but Mars is situated adjacent to
the main asteroid belt. (We will return to the two martian
moons, Phobos and Deimos, when we discuss asteroids in
299
300 Chapter 11 Planetary Adornments—Moons and Rings
JUPITER
Io Europa Ganymede Callisto Earth’s Moon
SATURN MERCURY
Mimas Enceladus Tethys Dione Rhea Iapetus
Titan
URANUS NEPTUNE PLUTO
Miranda Ariel Umbriel Titania Oberon Triton Charon
2,000 km
FIGURE 11.1 Images of the major moons in the Solar System obtained by various spacecraft. The
images are shown to scale. The planet Mercury and dwarf planet Pluto are shown for comparison.
The martian moons, Phobos and Deimos, are too small to be shown. (The pictures of Pluto and
Charon are artists’ conceptions rendered from groundbased and space-based images.)
Chapter 12.) Finally, as we learned in Chapter 7, Earth would size to or even larger than Mercury, and the smallest known
be without a moon if not for a cataclysmic collision when moons would fit within the expanse of a large metropolitan
the planet was young. What seems important is that while airport. Although most moons are airless, one has an atmo-
the larger planets were forming, they had greater attracting sphere denser than Earth’s, and several have very low-
mass and greater amounts of debris around them, which is density atmospheres. Some of the larger moons appear to
what gave rise to their greater number of moons. have differentiated chemically, and three are known to
Figure 11.1 displays images of some of the major moons have active volcanoes or geysers. (Our own Moon is not
in the Solar System, all shown to scale. In many ways covered in this chapter; we discussed it in Chapter 7 because
moons resemble smaller ver- of its close similarity to the four terrestrial planets of the
sions of the terrestrial plan- Moons are made inner Solar System.)
ets. Some, such as our own, of rock, ice, or As discussed in Chapter 6, we believe that many moons
are made of rock. Others, mixtures of both. formed at the same time as the planets they orbit, and
especially in the outer Solar that they accreted in much the same way that the planets
System, are mixtures of rock and water ice, with densities themselves grew—from the accumulation of planetesimals
intermediate between the two. A few seem to be made orbiting the Sun. For example, Ganymede, Jupiter’s larg-
almost entirely of ice. Several moons are comparable in est moon, grew from the accretion of ice and dust grains
11.1 Galileo’s Moons 301
FIGURE 11. 2 The evolution

ACCRETION AND DIFFERENTIATION 500 MILLION YEARS LATER
of Ganymede’s interior. In
Water Ice the final stages of accretion,
Ganymede was covered with
Silicate Ice-silicate a global ocean of liquid water,
underlain by a zone of silicates
Ice-silicate Silicate and a mixed core of ice and
silicates. Five hundred million
years later the dense silicates
had sunk gravitationally to form
a core, and much of the ocean
had frozen.
that were orbiting in a disk surrounding the young, hot way that is spectacularly different from the rotation of any
proto-Jupiter. At the distance of Ganymede’s orbit from the other known moon in the Solar System. Hyperion’s rotation
glowing proto-Jupiter core, temperatures were low enough is chaotic (see Chapter 10), meaning that it tumbles in its
for grains of water ice to survive, and along with silicate orbit with a rotation period and a spin-axis orientation that
materials, they coalesced to form planetesimals. It took are constantly and unpredictably changing!
less than a half million years for these planetesimals to Some moons revolve in a
accrete and form the moon Ganymede. Heating generated direction that is opposite the Some captured,
by accretion melted parts of the moon to form an outer rotation of their planets, and or “irregular,”
water layer, an inner silicate zone, and an ice-silicate core. some are situated in distant, moons have
These layers, however, were not stable. As cooling took unstable orbits. These are retrograde orbits.
place, much of the outer water layer froze to form a dirty almost certainly bodies that
ice crust. Most of the denser silicate materials sank to the
center to form a core, leaving an intermediate ice-silicate
zone (Figure 11.2).
FIGURE 11.3 Saturn’s Hyperion, certainly among the
We call Ganymede and other moons that formed, together
strangest-looking moons, rotates chaotically, with its rotation
with their parent planets, regular moons. Regular moons
period and spin axis constantly changing. The 250-kilometer (km)
revolve around their planets moon’s low density and spongelike texture, seen in this Cassini
in the same direction as the Moons that formed image, suggest that its interior houses a vast system of caverns.
planets rotate and in orbits together with their
that lie nearly in the planets’ planets are called
equatorial planes. These char- regular moons.
acteristic orbits arose because
the debris from which the regular moons formed was orbit-
ing in the planets’ equatorial planes and in the same direc-
tion as the evolving planets were rotating. With few
exceptions, regular moons are tidally locked to their parent
planets. Recall from Chapter 10 that tidal locking causes a
body to rotate synchronously with respect to its orbit, as
does Earth’s Moon. A moon in synchronous rotation around
its planet has fi xed leading and trailing hemispheres: one
hemisphere permanently faces the direction the moon is
traveling in its orbit around the planet, and the other per-
manently looks backward. These two surfaces can appear
very different from one another, because the leading hemi-
sphere is always flying directly into any local debris sur-
rounding the planet.
An especially strange regular moon is Saturn’s Hyperion,
seen in Figure 11.3. Hyperion has an unusual appearance,
but that is not the only thing peculiar about it. It rotates in a
formed elsewhere and were later captured by the planets.

We call them irregular moons. The largest irregular moon is Geologically Active Moons:
Neptune’s Triton. It orbits Neptune in a retrograde, or “back- Io, Enceladus, and Triton
ward,” direction. Other moons, such as Saturn’s Phoebe,
also have retrograde orbits. Most of the recently discovered One of the more spectacular surprises in Solar System
moons of the outer planets are irregular, and many are only exploration was the discovery of active volcanoes on Io, the
a few kilometers across. innermost of the four large
Galilean moons of Jupiter. Yet Io is the
in one of those rare events Solar System’s
that happen in science, plan- most volcanically
etologists predicted Io’s vol- active body.
11.2 Moons as Small Worlds canism just 2 weeks before its
discovery, basing their conclusions on the very same tidal
There are several ways we could group the moons of our stresses discussed in Chapter 10. Did you ever take a piece
Solar System. Some schemes are based on the sequence of of metal and bend it back and forth, eventually breaking it
the moons in their orbits around the parent planets; others in half? Touch the crease line and you can burn your fin-
are based on the sizes or compositions of the moons. For gers! Just as bending metal in your hands creates heat, the
example, we have already noted that a few moons are pre- continual flexing of Io’s crust caused by the changing
dominantly rocky objects, some are mostly ices, and most strength and direction of the tides generates enough energy
appear to be mixtures of ice and rock. to melt parts of the crust. In this way, Jupiter’s gravitational
The scheme we will use here is based on the amount energy is converted into thermal energy, powering the most
and timing of the moons’ geological activity as expressed active volcanism in the Solar System.
by the features we see on their surfaces. As we learned when As Voyager approached Jupiter, images showed that Io’s
looking at the terrestrial planets and Earth’s Moon, surface surface is literally covered by volcanic features, including
features observed on planets and moons provide critical vast lava flows, volcanoes, and volcanic craters (see the
clues to their geological his- opening photograph for Chapter 10). Amazingly, however,
tory. For example, water ice is Brightness, pictures from Voyager (and later from Galileo) failed to show
a common surface material structure, and crater a single impact crater, making Io unique among all the solid
among the moons of the outer density of moon planetary bodies seen so far in planetary exploration. With
Solar System, and the fresh- terrains give clues to a surface so young, Io must be volcanically active indeed,
ness of that ice tells us some- geological activity. with lava flows and volcanic ash burying impact craters
thing about the ages of those as quickly as they form. Scientists working on the Voyager
surfaces. Meteorite dust darkens the icy surfaces of moons mission discovered just how volcanically active Io is in
just as dirt darkens snow late in the season in our own urban spectacular and undeniable fashion when postencounter
areas. In other words, a bright surface often (but not always) images looking back toward this moon showed explosive
means a fresh surface. (As we will see later in this section, volcanic eruptions sending debris hundreds of kilometers
the oldest surfaces on Jupiter’s Europa are bright, whereas above Io’s surface!
the youngest are dark.) As of this writing, Io has more than 300 known volcanic
As we discussed in Chapter 7, the size and number of vents, with more than 150 active volcanoes observed by the
impact craters gives us the relative timing of events such Voyager, Galileo, and New Horizons spacecraft. The most
as volcanism, and this timing allows us to gauge if and vigorous eruptions, with vent velocities of up to 1 kilometer
when a moon may have been active in the past. Terrains per second (km/s), spray sulfurous gases and solids as high
having a large number of craters are older than those hav- as 300 km above the surface. Ash and other particles rain
ing only a few. Observations of erupting volcanoes, as on onto the surface as far as 600 km from the vents, as can be
Io and Enceladus, are direct evidence that some moons are seen in the opening photograph in Chapter 10. The moon
geologically active today. is so active that several huge eruptions are often occurring
What happens when we apply these age-dating tech- at the same time. One look at an image such as Figure 11.4
niques to the moons in the Solar System? We find an immense leaves little doubt about the source of the material supply-
diversity: some moons have been frozen in time since their ing the plasma torus and Io flux tube that we discussed in
formation during the early development of the Solar System, Chapter 9.
while others are even more geologically active than Earth. Io’s surface, shown in Figure 11.5 and the opening pho-
In our classification scheme of moons, we include four cat- tograph for Chapter 10, displays a wide variety of colors—
egories of geological activity: (1) defi nitely active today, (2) pale shades of red, yellow, orange, and brown. Mixtures of
probably or possibly active today, (3) active in the past but sulfur, sulfur dioxide frost, and sulfurous salts of sodium
not today, and (4) apparently not active at any time since and potassium on the moon’s surface are the likely cause
their formation. of the colors. Bright patches may be fields of sulfur diox-
FIGURE 11. 4 When this

image of Io was obtained by
Galileo, two volcanic eruptions
could be seen at once. The
plume of Pillan Patera rises 140
km above the limb of the moon
on the left, while the shadow
of a 75-km-high plume can be
seen to the right of the vent of
Prometheus, near the moon’s
center.
Plume Shadow
ide snow. You may well wonder how snow can fall from
Io’s nearly nonexistent atmosphere. According to current
understanding, liquid sulfur dioxide must flow beneath
Io’s surface, held at high pressure by the weight of overly- FIGURE 11. 5 An image of Jupiter’s volcanically active moon Io,
ing material. Like water from an artesian spring, this pres- constructed of images obtained by Galileo.
surized sulfur dioxide can be pushed out though fractures
in the crust, producing sprays of sulfur dioxide snow crys-
tals that travel for up to hundreds of kilometers before set-
tling back to the moon’s surface. (To see a similar process
in action, just operate a carbon dioxide fi re extinguisher.
These fi re extinguishers contain liquid carbon dioxide at
high pressure that immediately turns to “dry ice” snow as
it leaves the nozzle.)
Voyager and Galileo images show parts of the surface of
Io at high resolution. They reveal a variety of plains, irregu-
lar craters, and flows, all related to eruption of mostly sili-
cate magmas onto the surface of the moon. They also show
high-standing mountains, some nearly twice the height of
Mount Everest, Earth’s tallest mountain. Huge structures,
some 65 km across, show multiple calderas and other com-
plex structures telling of a
long history of repeated erup- Silicate magmas
tions followed by collapse of dominate Io’s
the partially emptied magma volcanism.
chambers. Many of the floors
are very hot (Figure 11.6) and might still contain molten
material similar to magnesium-rich lavas that erupted on
Lava flow
FIGURE 11.6 A Galileo image of Io showing regions where lava FIGURE 11.7 Saturn’s moon Enceladus is the source of
has erupted. A molten lava flow is shown in false color to make it material in Saturn’s E Ring. Note the distortion in the distribution
more visible. of ring material in the immediate vicinity of the moon caused by
its gravitational influence on the orbits of ring particles. Other
bright objects in this Cassini image are stars.
Earth more than 1.5 billion years ago. Note that volcanoes
on Io are spread around the moon in a much more random
fashion than on Earth, where tectonic patterns influence could generate enough thermal energy to warm the ice and
the location of volcanism. drive it to the surface. This process, called cryovolcanism,
Io probably formed at about the same time as the other is similar to normal volca-
giant planets’ moons. Io’s current volcanic activity suggests nism but is driven by subsur- Water volcanoes on
that its entire mass has recy- face low-temperature liquids Enceladus are an
cled, or turned inside out, Volcanism may have rather than molten rock. example of low-
more than once in the past, turned Io inside out Cassini showed that active temperature
leading to chemical differen- several times. cryovolcanic plumes, like cryovolcanism.
tiation. Volatiles such as water those seen in Figure 11.8b, are
and carbon dioxide probably escaped into space long ago, energetic enough to overcome the moon’s low gravity, send-
with most heavier materials sinking to the interior to form ing tiny ice crystals into space to repopulate particles that
a core. Sulfur and various sulfur compounds, aided by sili- are being continually lost from Saturn’s E Ring.1 We might
cate magmas, are constantly being recycled, forming the logically ask why Enceladus is so active while Mimas, a
complex surface we see today. neighboring moon of about the same size and also subject
As we move farther out in the Solar System, we fi nd vol- to tidal heating, appears to be quite dead. The answer, unfor-
canism of a different type. Enceladus, one of Saturn’s icy tunately, remains a mystery.
moons, is only 500 km in diameter. Even so, it shows a wide Cryovolcanism also occurs in the Neptune system. Tri-
variety of ridges, faults, and smooth plains— evidence of ton, the largest moon in the Neptune system, is an irregular
tectonic processes unexpected for such a small body. Some moon that must have been captured by its planet. To achieve
impact craters, rather than showing crisp features, appear its present circular, synchronous orbit, it must have expe-
“softened,” as though they were “relaxed” by the viscous rienced extreme tidal stresses from Neptune following its
flow of warm ice. Parts of the moon show no craters at all, capture. Such stresses would be similar to those on Io and
indicating recent resurfacing. As we will see in the next sec- would have generated large amounts of thermal energy. The
tion, the orbit of Enceladus is located in the densest part of interior might even have melted, allowing Triton to become
Saturn’s E Ring, and astronomers had long suspected that chemically differentiated.
the moon was providing the ring with a continuous supply Like all other moons in the outer reaches of the Solar
of tiny particles, most probably ice crystals (Figure 11.7). System, Triton is a cold place. Its surface temperature is only
The joint NASA–European Space Agency Cassini mis- about 38 kelvins (K). Triton has a thin atmosphere, with a
sion confi rmed the existence of tectonic activity on Encela- surface composed mostly of ices and frosts of methane and
dus. Close-up images of terrain near the moon’s south pole
reveal a cracked and twisted “taffylike” appearance (Figure 1
Saturn’s diffuse G Ring also contains a tiny “moonlet” embedded in a
11.8a). Temperature measurements of these geologically ring arc. This moonlet appears to be the source of the G Ring particles,
young “cracks” show them to be warmer than their sur- although it is much too small to be geologically active. The small icy
roundings. A combination of tidal heating and the decay of particles that make up the ring are easily dislodged from the moon-
radioactive elements within the moon’s rocky component let's surface by impacts of micrometeorites.
(a) (b)
FIGURE 11.8 Cassini images of Enceladus. (a) The twisted and folded surface of deformed ice cracks near the moon’s south
pole (shown blue in false color) are warmer than surrounding terrain and were found to be the sources of cryovolcanism. (b)
Cryovolcanic plumes in the south polar region are seen spewing ice particles into space.
nitrogen. From the relative lack of craters, we know that slushy ice onto the surface from the interior. Veinlike fea-
the surface is geologically young. Part of Triton is covered tures include grooves and ridges that could result from the
with “cantaloupe terrain,” so named because it looks like extrusion of ice along fractures. The rest of Triton is covered
the skin of a cantaloupe (Figure 11.9). Irregular pits and with smooth plains of volcanic origin. Irregularly shaped
hills may represent surface deformation and extrusion of depressions as wide as 200 km (Figure 11.10) formed when
FIGURE 11.9 This Voyager

mosaic shows various
terrains on the Neptune-
facing hemisphere of Triton.
“Cantaloupe terrain” is visible
at the top; its lack of impact
craters indicates a geologically
younger age than the bright,
cratered terrain at the bottom.
Possibly Active Moons:

Europa and Titan
One of the more fascinating objects in the outer Solar Sys-
tem is Jupiter’s moon Europa, a rock world slightly smaller
than our Moon but with an outer shell of water. We know
that the surface of the water is frozen, but what lies beneath
the ice? Like Io, Europa experiences continually changing
tidal stresses from Jupiter that generate internal energy and
possibly volcanism. Some calculations show that the tidal
heating may be too small to produce volcanism, but large
enough to keep much of the subsurface water in a liquid
state. Beneath its cap of solid ice— estimated to be anywhere
from a few hundred meters to tens of kilometers thick—
Europa may possess a global ocean.
Some planetologists believe this subsurface sea could
be 100 km deep and contain more water than all of Earth’s
oceans! Such an ocean would have some interesting charac-
teristics. We might expect it to be salty because of dissolved
minerals. Supporting evidence for Europa’s salty sea came
when physicists took a close look at Galileo’s magnetome-
FIGURE 11.10 This irregular basin on Triton has been partly ter data and found that Europa’s magnetic field is variable,
filled with frozen water, forming a relatively smooth ice surface. indicating an internal electrically conducting fluid. It may
The state of New Jersey could just fi t within the basin’s boundary. also contain an abundance of organic material brought in
by impacting comet nuclei. In fact, what we may have on
Europa is an environment that is not so different from some
of our terrestrial ecological niches that support “extremo-
mixtures of water, methane, and nitrogen ice melted in philes” (see Connections 11.1). The universal conditions
the interior of Triton and erupted onto the surface, much necessary to create and support life—liquid water, heat, and
as rocky magmas erupted onto the lunar surface and filled organic material— could all be present in Europa’s oceans.
impact basins on Earth’s Moon. As such, Europa is a high-priority target in the search for
Voyager 2 found four active geyserlike cryovolcanoes on extraterrestrial life.
Triton. Each consisted of a plume of gas and dust as much Europa’s surface is young, with few impact craters, but
as a kilometer wide rising 8 km above the surface, where it is deformed by tectonic activity driven by heating from
the plume was caught by upper atmospheric winds and car- its interior. Regions of chaotic
ried for hundreds of kilome- terrain, as shown in Figure Europa is covered with
ters downwind. How do the Nitrogen propels 11.11, are places where the icy broken slabs of ice.
eruptions on Triton work? Triton’s geyserlike crust has been broken into
Their association with Tri- volcanism. slabs that have been rafted into new positions. These slabs
ton’s southern hemispheric could be fitted back together like pieces of a jigsaw puzzle.
ice cap may provide some clues to the process. We know In other areas the crust split apart and has been filled in
that nitrogen ice in its pure form is transparent and allows with new dark material forced to the surface from the inte-
easy passage of sunlight. Thus, clear nitrogen ice could cre- rior, in a process somewhat resembling seafloor spreading
ate a localized greenhouse effect (see Chapter 8), in which on Earth (see Chapter 7). With time, frosts cover the dark
solar energy trapped beneath the ice raises the temperature material, causing a general brightening of the surface. Of
at the base of the ice layer. A temperature increase of only the handful of large impact structures preserved on Europa,
4°C would be enough to vaporize the nitrogen ice. As gas is all are shallow features resembling the patterns formed
formed, the expanding vapor exerts very high pressures when a rock is dropped into stiff mud. The chaotic terrain,
beneath the ice cap. Eventually the ice ruptures and vents spreadinglike features, impact structures, and other surface
the gas explosively into the low-density atmosphere. Dark features all suggest that the icy crust of Europa consisted
material, perhaps silicate dust or radiation-darkened meth- of a thin brittle shell overlying either liquid water or warm,
ane ice grains, is carried along with the expanding vapor slushy ice at the time when the features formed. Although
into the atmosphere, from which it subsequently settles to we do not know whether these are the conditions on Europa
the surface, forming dark patches streaked out by local today, the geologically young surface holds open that
winds, as seen near the bottom of Figure 11.9. possibility.
The case for current geological activity on Saturn’s large

moon Titan appears to be stronger than for Jupiter’s Europa.
Titan is larger than Mercury
and has a density that sug- A dense, hazy
gests a composition of about atmosphere obscures
45 percent water and 55 per- Titan’s surface.
cent rocky material. What
makes Titan especially remarkable is a thick atmosphere,
which historically has obscured our view of its surface. As
we learned in Chapter 8, Mercury’s secondary atmosphere
has been lost to space, while Titan’s greater mass and dis-
tance from the Sun has allowed it to retain an atmosphere
FIGURE 11.11 A high-resolution Galileo image of Jupiter’s moon that is 30 percent denser than Earth’s. Like Earth, Titan has
Europa, showing where the icy crust has been broken into slabs an atmosphere that is mostly nitrogen, but views of Titan’s
that, in turn, have been rafted into new positions. These areas limb show atmospheric layers that are most likely photo-
of chaotic terrain are characteristic of a thin, brittle crust of ice chemical hazes, much like smog. Titan’s atmosphere is
floating atop a liquid or slushy ocean. reminiscent of the smog over Los Angeles on a bad day.
C O N N E C T I O N S 1 1 .1
Extreme Environments and an Organic Deep Freeze

During a visit in 1835 to the Galápagos Islands in the eastern As we have learned in this chapter, Jupiter’s moon Europa
Pacific Ocean, British naturalist Charles Darwin (1809–1882) may contain a shell of liquid water beneath a surface layer of
observed variations in animal and plant life that eventually water ice. Europa is not the only moon suspected of having a
led to his now legendary theory of evolution. Off the coast of subsurface ocean. Like Europa, Callisto also shows magnetic
the Galápagos, 2,500 meters beneath the ocean’s surface, is variability, the signature of a salty ocean. In our search for life
a form of life that Darwin could never have imagined. Here elsewhere in the Solar System, Jupiter’s large moons should be
the Nazca and Pacific plates (see Figure 7.19) grind furiously among our prime targets. So, too, should Enceladus, the moon
against one another, creating friction, high temperatures, and of Saturn that spews water-ice crystals.
seafloor volcanism. Mineral-rich superheated water pours out The presence of comet-borne organic material in the
of “hydrothermal vents.” The surrounding water contains very oceans of Europa and Callisto cannot yet be confi rmed, but
little dissolved oxygen. No sunlight reaches these depths. Yet there is one place where we can fi nd such materials—in Titan’s
in the total darkness of the ocean bottom, life abounds. From massive atmosphere. Here, organic gases reveal themselves
tiny bacteria to shrimp to giant clams and tube worms, sea life through the analytical eyes of the spectrograph. We know
not only survives but thrives in this severe environment. In the that Titan’s nitrogen atmosphere contains methane and traces
complete absence of sunlight, the small single-celled organ- of ethane, propane, ethylene, hydrogen cyanide, carbon mon-
isms at the bottom of the local food chain get their energy from oxide, acetylene, and other compounds of biological interest.
“chemosynthesis,” a process by which inorganic materials are For example, five molecules of hydrogen cyanide (HCN) will
converted into food through the use of chemical energy. spontaneously combine to form adenine (C5H3N4NH2), one of
Biologists call these life-forms “extremophiles.” Robust the four primary components of DNA and RNA. HCN is also
types of bacteria are found flourishing in the scalding waters a building block of amino acids, which in turn combine to
of Yellowstone’s hot springs; in the bone-dry oxidizing envi- form proteins.
ronment of Chile’s Atacama Desert; and in the Dead Sea, where Photodissociation and recombination of these various gases
salt concentrations run as high as 33 percent. Bacteria have produce complex organic molecules that then rain out onto
even been found in core samples of ancient ice 3,600 meters Titan’s surface as a frozen tarry sludge. Biochemists believe that
below the surface of the East Antarctic ice sheet. When it comes many of these substances are biological precursors, similar to
to harsh habitats, life is amazingly adaptable. If life can exist the organic molecules that preceded the development of life on
under such extreme conditions on Earth, then it might also Earth. For now, these possible clues to the origins of terrestrial
exist on those moons of the giant planets that have the neces- life remain locked up in Titan’s deep freeze, quietly awaiting
sary ingredients for life: liquid water, an energy source, and analysis by future space missions.
the presence of organic compounds.
As Titan heated up and differentiated chemically, vari- sphere) is now in darkness and not accessible to infrared
ous ices, including methane (CH4) and ammonia (NH3), imaging. Radar imaging, however, easily pierces the moon’s
emerged from the interior to form an early atmosphere. thick cloud cover and reveals its surface features without
Ultraviolet photons from the Sun have enough energy to the aid of solar illumination. Radar images have revealed
break apart ammonia and methane molecules— a process irregularly shaped features in Titan’s northern hemisphere
called photodissociation. Pho- that appear to be widespread lakes of methane, ethane, and
todissociation of ammonia is Titan’s atmosphere other hydrocarbons (Figure 11.13). However, the abundance
the likely source of Titan’s and surface are rich in of methane— both on the surface and in the atmosphere of
abundant supply of atmo- organic compounds. Titan—presents a problem. The photodissociative process
spheric nitrogen. Photodisso- by sunlight should have removed all atmospheric meth-
ciation of methane breaks the molecules into fragments, ane within a geologically brief period of 50 million years
which can recombine to form complex hydrocarbons such or so. Therefore, there must be some process for renewing
as ethane and other organic compounds. Organic com- the methane that is being destroyed by solar radiation. A
pounds tend to form in tiny particles, creating an organic likely candidate is cryovolcanism. As on Earth, radioac-
smog and giving Titan’s atmosphere its characteristic orange tive decay is an important source of internal heating, and
hue (Figure 11.12a). The Cassini spacecraft, which began Titan’s internal thermal energy could cause the release of
orbiting Saturn in 2004, gave astronomers their fi rst close- “new” methane from underground through cryovolcanism
up views of the moon’s surface (Figure 11.12b). Haze-pene- in much the same way that terrestrial volcanism releases
trating infrared imaging shows broad regions of dark and water vapor and carbon dioxide into Earth’s atmosphere
bright terrain (Figures 11.12c and d). (see Chapter 8). So far, there has been no direct evidence of
Titan’s northern hemisphere (currently its winter hemi- active cryovolcanism on Titan, but the presence of abundant
FIGURE 11.12 Cassini

(a) (b)
images of Saturn’s largest
moon, Titan. (a) Visible-light
imaging shows Titan’s orange
atmosphere, which is caused
by the presence of organic
smoglike particles. (b) This
infrared- and ultraviolet-
light combined image shows
surface features and a bluish
atmospheric haze caused by
scattering of ultraviolet sunlight
by small atmospheric particles.
Bright clouds in Titan’s lower
atmosphere are seen near the
moon’s south pole. (c) Infrared-
light imaging penetrates Titan’s
smoggy atmosphere and
reveals surface features. (d)
This infrared image covers the
same general area of Titan as
(c) (d)
seen in (b). The large dark area,
called Xanadu, is about the size
of Australia.
the production of terrestrial proteins. During its descent,

Huygens encountered 120-meter-per-second (m/s) winds
and temperatures as low as 88 K. As it reached the surface,
though, winds had died down to less than 1 m/s and the
temperature had warmed to 112 K. Images taken during the
descent (Figure 11.14a) show terrains reminiscent of those
on Earth, with networks of channels, ridges, hills, and flat
areas that may be dry lake basins. These terrains suggest
a sort of “hydrological cycle” in which methane rain falls
to the surface, washes the ridges free of the dark hydrocar-
bons, and then collects into drainage systems that empty
into low-lying, liquid methane pools.2 Stubby, dark channels
appear to be springs where liquid methane emerges from the
subsurface; bright, curving streaks could be water ice that
has oozed to the surface to feed glaciers. Although no liq-
uid methane rain or surface pools were seen in the Huygens
images, the near absence of impact craters on the surface
FIGURE 11.13 Radar imaging (false color) near Titan’s north
would point to recent—if not current—hydrological activ-
pole shows what appear to be lakes of liquid hydrocarbons
ity. Radar views also indicate an active surface, showing
covering 100,000 square kilometers (km2) of the moon’s surface.
features that resemble terrestrial sand dunes and channels,
Features commonly associated with terrestrial lakes, such as
islands, bays, and inlets, are clearly visible in many of these radar
but only an occasional impact crater (Figure 11.14b).
images. Once on Titan’s surface, Huygens continued to take
pictures and make physical and compositional measure-
ments. Huygens scientists described the surface as having
characteristics similar to wet or dry sand or lightly packed
atmospheric methane and methane lakes strongly suggests snow. The surface was wet with liquid methane, which
that Titan is indeed geologically active. evaporated as the probe—heated to 2,000 K during its pas-
Shortly after arriving at Saturn, the Cassini spacecraft
released a 320-kilogram (kg) probe, Huygens, which plunged 2
We are a bit loose with our terminology here. Technically, hydrologi-
through Titan’s atmosphere measuring composition, tem- cal refers to water, as in the terrestrial hydrological (water) cycle dis-
perature, pressure, and winds and taking pictures as it cussed in Chapter 8. On Titan, methane assumes the role that water
descended. Huygens confi rmed that cloud particles contain plays on Earth. Some planetologists prefer to use the term methano-
nitrogen-bearing organic compounds—key components in logical cycle when applied to Titan.
FIGURE 11.14 (a) The surface of Titan viewed from the Huygens probe during its descent to the
surface. The dark drainage patterns resemble river systems on Earth. (b) This radar image of Titan
taken from the Cassini orbiter shows a rare impact crater and its ejecta blanket.
(a) (b)
FIGURE 11.15 View areas on Earth’s Moon. The high number of impact craters
of the surface of Titan on the dark terrain reflects the period of intense bombard-
obtained from the Huygens ment during the early history of the Solar System. The most
probe, showing a relatively
extensive region of ancient dark terrain includes a semicir-
flat surface littered with
cular area more than 3,200 km across (about the size of
water-ice “rocks.” The two
Australia) on the leading hemisphere. Furrowlike depres-
rocks just below the center
of the image are about 85 sions occurring in many dark areas are among Ganymede’s
centimeters (cm) from the oldest surface features. They may represent surface defor-
camera and roughly 15 and mation from internal processes, or they may be relics of
4 cm across, respectively. impact-cratering processes.
The dark “soil” is probably Impact craters range up to hundreds of kilometers in
made up of hydrocarbon diameter, with the larger craters being proportionately more
ices. shallow. With time, the icy crater rims deform by viscous
(very slow) flow, as might a lump of soft clay, and they can
ultimately lose nearly all of their varying topography. Such
features are seen as flat circular patches of bright terrain
called “palimpsests,” which are characteristic of Gany-
mede’s icy lithosphere (Figure 11.16a). Palimpsests are
found principally in the dark terrain of Ganymede and are
believed to be scars left by early impacts onto a thin, icy
crust overlying water or slush (Figure 11.16b).
When fi rst viewed in Voyager images, the bright terrain
was thought to represent regions that had been flooded
sage through the atmosphere—landed in the frigid soil. The by water or slush erupted from the interior of Ganymede.
surface was also rich with other organic compounds, such Galileo images, however, failed to show any indications
as cyanogen (CN)2 and ethane (C2H6). As shown in Figure of such flooding, other than in a few local places. How,
11.15, the surface around the landing site is relatively flat then, did the bright terrain form? The answer seems to be
and littered with rounded “rocks” of water ice. The dark related to a style of surface formation by tectonic processes
“soil” is probably a mixture of water ice and of hydrocar- not previously considered. In Chapter 7 we discussed how
bon ices that have precipitated from the atmospheric haze planetary surfaces can be fractured by faults or folded by
seen in the background. compression resulting from movements such as those ini-
In many ways, Titan resembles a primordial Earth, albeit tiated in the mantle. On Ganymede, the tectonic processes
at much lower temperatures. The presence of organic com- have been so intense that the fracturing and faulting have
pounds that could be biological precursors in the right completely deformed the icy crust, destroying all signs of
environment makes Titan another high-priority target for older features, such as impact craters.
continued exploration (see Connections 11.1). We might ask why Ganymede is no longer geologically
active. The energy that powered Ganymede’s early activity
was liberated during a period of differentiation when the
moon was very young. Once the differentiation process
Formerly Active Moons: was complete, that source of internal energy ran out and
Ganymede and Some Moons geological activity ceased.
Many other moons show evidence of an early period of
of Saturn and Uranus geological activity that has resulted in a dazzling array of
Some moons show clear evidence of past ice volcanism and terrains. A 400-km impact crater scars Saturn’s moon Tethys,
tectonic deformation, but no current geological activity. covering 40 percent of the diameter of the moon itself, and
Ganymede is the largest moon an enormous canyonland wraps at least three-fourths of
in the Solar System, even Ganymede is the the way around the moon’s equator. Saturn’s moon Dione
larger than the planet Mer- Solar System’s shows bright ice cliffs up to several hundred meters high,
cury. The surface is composed largest moon. created by tectonic fracturing. The trailing hemisphere of
of two prominent terrains: a Saturn’s Iapetus is bright, reflecting half the light that falls
dark, heavily cratered (and therefore ancient) terrain, and on it, while much of the leading hemisphere is as black as
a bright terrain characterized by ridges and grooves. Gany- tar (Figure 11.17). These dark deposits appear only in the
mede’s low density (1.9 times that of water) indicates that leading hemisphere of Iapetus, suggesting that they might
its bulk composition is about half water and half rocky be debris that was blasted off small retrograde moons of Sat-
materials. Overall, the moon’s surface is relatively bright. urn by micrometeoritic impacts and swept up by Iapetus as
Even the so-called dark terrains are brighter than the bright it moved along in its prograde orbit around Saturn.
(a) (b)
1 Impact
Icy crus
t
Water o
r
slush
Solid m
2 The resulting crater is an tle
surrounded by fractured
crust, and begins to fill in.
Palimpsests
3 The crater fills
and the fractured
crust settles.
Palimpsest
FIGURE 11.16 (a) A Voyager image of impact scars called

palimpsests on Jupiter’s moon Ganymede. (b) Palimpsests form as
viscous flow smooths out structures left by impacts on icy surfaces.
FIGURE 11.17 Cassini images of Iapetus show the boundary between its bright and dark hemispheres.
FIGURE 11.18 (a) A Voyager

(a) (b) image of Saturn’s moon Mimas
and the crater Herschel. (b)
The Death Star from Star
Wars. (The movie was released
3 years before the image of
Mimas was taken.)
Saturn’s moon Mimas, no larger than Ohio, is heavily larger than the one that created Herschel. Some astronomers
cratered with deep, bowl-shaped depressions. The most believe that Mimas (and perhaps other small, icy moons as
striking feature on Mimas is a huge impact crater in the well) was hit many times in the past by objects so large as
leading hemisphere. Named “Herschel” after the German- to fragment the moon into many small pieces. Each time
English astronomer Sir William Herschel, who discovered this happened, the individual pieces still in Mimas’s orbit
many of Saturn’s moons, the crater is 130 km across, or a would coalesce to re-form the moon, perhaps in much the
third the size of Mimas itself. same way that our own Moon coalesced from fragments
(Figure 11.18 illustrates the Mimas may have that remained in Earth orbit after a large planetesimal
striking similarity between been broken apart impacted Earth early in its history (see Chapter 7).
Mimas and the “Death Star” and reassembled Areas on Uranus’s small moon Miranda have been
from the original Star Wars, many times. resurfaced by eruptions of icy slush or glacierlike flows
which was released in 1977, 3 (Figure 11.19). Other Uranus moons— Oberon, Titania,
years before the image of Mimas was taken.) It is doubtful and Ariel— are covered with faults and other signs of
that Mimas could have survived the impact of a body much early tectonism. On Ariel, in particular, very old, large
craters appear to be missing, perhaps obliterated by ear-
lier volcanism.
FIGURE 11.19 A Voyager image of fault zones, high cliffs,
and cratered terrain on Miranda, a 472-km-diameter moon of
Uranus.
Geologically Dead Moons
Geologically dead moons, including Jupiter’s Callisto, Ura-
nus’s Umbriel, and a large assortment of irregular moons,
are those for which there is little or no evidence of internal
activity having occurred at any time since their formation.
Their surfaces are heavily cratered and show no modifica-
tion other than the cumulative degradation caused by a long
history of impacts.
Callisto (Figure 11.20) is about the size of Mercury.
It is also the darkest of the Galilean moons, yet it is still
twice as reflective as Earth’s Moon. This brightness indi-
cates that Callisto is rich in water ice, but with a mixture
of dark, rocky materials. Except for terrains that experi-
enced large impact events, the surface is essentially uni-
form, consisting of relatively dark, heavily cratered terrain.
High-resolution images reveal that Callisto’s surface has
been modified by local landslides and places where the
small craters have been erased by an unknown process. Its
most prominent feature is a 2,000-km, multiringed struc-
ture of impact origin named Valhalla (the largest bright
feature visible on Callisto’s face in Figure 11.1). The impact
companions on both sides. Unlike the moons of Jupiter that

he had found a few months earlier, these features did not
move. Galileo was troubled by the lack of movement because
it was like nothing else he had observed. Two years later
the “companions” had vanished. Their unexpected disap-
pearance upset the Italian astronomer greatly; he feared his
earlier observations had been in error. A few years later, the
mysterious features reappeared.
For more than four decades astronomers puzzled over
Galileo’s discovery. In 1655 a 26-year-old Dutch instru-
ment maker, Christiaan Huygens (1629–1695), pointed a
superior telescope of his own design at Saturn and saw
what the astronomers of his day had failed to see. Saturn
is surrounded by an apparently continuous, flat ring, and
as Huygens correctly deduced, the variations in its vis-
ibility are caused by changes in the apparent tilt of the
ring as Saturn orbits the Sun. What was the nature of this
strange ring encircling Saturn? Astronomers assumed it
was a solid disk spinning around the planet, an interpreta-
tion that persisted for more than a century after Huygens’s
discovery. That notion began to weaken in 1675 when the
FIGURE 11. 20 A Galileo image of Jupiter’s second-largest great Italian-French astronomer Jean-Dominique Cassini
moon, Callisto. Its ancient surface is dominated by impact craters (1625–1712) found a gap in the planet’s seemingly solid ring.
and shows no sign of early internal activity. Saturn now appeared to have two rings rather than one, and
the gap that separated them became known as the Cassini
Division that we learned of in the previous chapter.
In 1850 a fainter ring, located just inside the two bright
may have occurred in a relatively thin, rigid crust overly- rings, was found independently by English and American
ing a fluidlike interior. The fluid mass then rapidly fi lled observers, giving Saturn a total of three known rings. For
the initial crater bowl, leaving only a trace of the impact illustrators and cartoonists, Saturn had become the iconic
scar. Geophysical measurements obtained from the Gali- image of a planet. Yet over the centuries following the
leo spacecraft suggest that Callisto is not differentiated, discovery of Saturn’s rings, exhaustive searches failed to
implying that it never went through a molten phase. On detect rings around any other planet. Papers published by
the other hand, the magnetometer aboard Galileo returned more than one distinguished astronomer explained why,
results suggesting that a liquid ocean could exist beneath theoretically, only Saturn could have rings. Then, in the
the heavily cratered surface, implying that some sort of dif- latter part of the 20th century, a new search technique
ferentiation has occurred. These observations simply point became available: observation of stellar occultations (see
out that many times in science we are faced with confl ict- Chapter 9). In 1977 a team of American astronomers using
ing ideas that usually can be resolved only with additional the occultation technique to study the atmosphere of Ura-
measurements or observations. nus saw brief, minute changes in the brightness of a star as
Umbriel, the darkest and third-largest of the Uranus it fi rst approached and then receded from the planet. The
moons, appears uniform in color, reflectivity, and general interpretation was immediately obvious: Uranus has rings!
surface features, indicative of an ancient surface. The real Over the next several years stellar occultations revealed
puzzle posed by Umbriel is why it has been geologically a total of nine rings surrounding the planet. In 1986 Voy-
dead, while the surrounding large moons of Uranus have ager 2 imaged two additional Uranus rings, and in 2005
been active, at least at some time in their past. the Hubble Space Telescope recorded two more, bringing
the total to 13.
Stellar occultations not only show the existence of rings;
they also tell us something about the rings themselves. The
11.3 Rings Surround the duration of an occultation event is a measure of the width
of the ring. The observed decrease in the brightness of a star
Giant Planets is an indication of the ring’s transparency and therefore of
the amount of material it contains. The very brief interrup-
If Galileo’s discovery of Jupiter’s moons was personally sat- tion of starlight as the Uranus rings passed in front of the
isfying, his other important discovery was decidedly less star showed that they are far too narrow to have been seen
so. When he first pointed his telescope at Saturn in 1610, the in the earlier, unsuccessful searches by more conventional
tiny disk of the planet seemed to be accompanied by smaller methods.
More ring discoveries were to follow from still another with Saturn gave serious consideration to flying the space-
technology: close-up studies by planetary probes. In 1979, craft through the Cassini Division to get as close to the
cameras on Voyager 1 recorded a faint ring around Jupiter; planet as possible. Had they carried out this plan, the Pio-
and in the same year, Pioneer 11 found a narrow ring just neer 11 mission would cer-
outside the bright rings of Saturn. tainly have ended right there. “Divisions” and “gaps”
For a while Neptune seemed to be the only giant planet Images taken the following in Saturn’s rings are
devoid of rings. Then, in the early to mid-1980s, occultation year by Voyager 1 show the not truly empty.
searches by teams of Ameri- Cassini Division to be fi lled
can and French astronomers One of Neptune’s with material. Why, then, did astronomers think it was
began yielding positive but rings contains higher- empty? The Cassini Division and many other smaller “gaps”
confusing results. Several density segments in the bright rings are simply regions with less ring mate-
occultation events t hat called ring arcs. rial. In contrast with their denser surroundings, they appear
appeared to be due to rings darker and thus empty.
were seen on only one side of the planet. The astronomers The B Ring is the brightest of Saturn’s rings. With a width
concluded that Neptune was surrounded not by complete of 25,500 km, two Earths could fit side by side between its
rings but rather by several arclike ring segments. Only when inner and outer edges. Strangely, the B Ring seems to have
Voyager 2 reached Neptune in 1989 did we learn that its no gaps at all, at least on the scale of those seen in the other
rings are indeed complete, and that the ring arcs are merely bright rings. The C Ring often fails to show up in normally
high-density segments within one of its narrow rings. All exposed photographs because of the limited ability to dis-
of Neptune’s rings are faint and, with the exception of the play a wide range of brightnesses. At the eyepiece of the
ring arcs, they contain too little material to be detected by telescope, though, this beautifully translucent ring appears
the stellar occultation technique. like delicate gossamer and hence is often called the Crepe
Ring. There is no known gap between the C Ring and either
of the adjacent rings. Only an abrupt change in brightness
Saturn’s Magnificent marks the boundary between them. What could cause such
a sharp change in the amount of ring material remains
Rings—A Closer Look an unanswered question. Too dim to be seen next to Sat-
All of the giant planets are now known to have ring systems, urn’s bright disk, the D Ring is a fourth wide ring that was
although each is unique (Figure 11.21). Moreover, it turns unknown until imaged by Voyager 1. It shows less structure
out that the giant planets are the only planets in our Solar than any of the bright rings, and it does not appear to have
System that have rings; none of the terrestrial planets do. a defi nable inner edge. The D Ring may extend all the way
The most complex ring system belongs to what was once down to the top of Saturn’s atmosphere, where its ring par-
thought to be the only ringed planet— spectacular Saturn. ticles would enter and burn up as meteors.
Huygens, with his mid–17th century understanding of phys- Saturn’s bright rings are far from homogeneous. The A
ics, believed Saturn’s ring to be a solid disk surrounding and C rings contain hundreds, and the B Ring thousands,
the planet. This view was challenged in later years, but it of individual ringlets, some
was not until the middle of the 19th century that the bril- only a few kilometers wide Saturn’s rings
liant Scottish mathematician James Clerk Maxwell showed (Figure 11.23). Each of these contain thousands
that solid rings would be unstable and would quickly break ringlets is a narrowly con- of individual narrow
apart. As we saw in Chapter 10, orbital resonances and fi ned concentration of ring rings called ringlets.
other gravitational interactions maintain the stability of particles bounded on both
these rings. sides by regions of relatively little material. Spacecraft
Figure 11.21b shows the individual components of Sat- images of Saturn’s rings turned up many other surprises,
urn’s bright ring system and its major divisions and gaps. as we will see later.
The most conspicuous features are its expansive bright Each time the plane of Saturn’s rings lines up with
rings, which dominate all photographs of Saturn (Figure Earth, as it does about every 15 years, the rings all but
11.22). Among the four giant planets, only Saturn has rings vanish for a day or so in even the largest telescopes. With
so wide and so bright. the glare of the rings temporarily gone, astronomers can
Photographs often show only the two outer and brighter search for undiscovered moons or other faint objects
rings, separated by the Cassini Division. The outermost close to Saturn. In 1966, American astronomer Walter A.
ring, the A Ring, is the narrowest of the three bright rings. Feibelman (1930–2004) was looking for moons when he
It has a sharp outer edge and contains several narrow gaps. found weak but compelling evidence for a faint ring near
The conspicuous Cassini Division is so wide (4,700 km) the orbit of Saturn’s moon Enceladus. In 1980, Voyager
that the planet Mercury would almost fit within it. Astrono- 1 confi rmed the existence of this faint ring, now called
mers once thought that it was completely empty. In fact, the E Ring, and found another closer one known as the
space scientists planning the 1979 encounter of Pioneer 11 G Ring. The E Ring and the G Ring are both examples of
(a) JUPITER
Main
Halo Gossamer Gossamer
Radius (inner) (outer)
of Jupiter
0 100,000 200,000 300,000 400,000 500,000
Cassini
(b) SATURN
Division
D C B A F G E
Radius
of Saturn
0 100,000 200,000 300,000 400,000 500,000

Titan Maxwell Encke Keeler
Ringlet Ringlet Gap Gap
(c) URANUS
6 α (Alpha)
5 β (Beta)
4 η (Eta)
γ (Gamma)
Radius of Uranus δ (Delta)
λ (Lambda)
ε (Epsilon)
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000
Distance from planet center (km)
0 100,000 200,000 300,000 400,000 500,000
(d) NEPTUNE
LeVerrier
Galle Lassell Arago

Radius of Neptune Adams
Unnamed
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000
0 100,000 200,000 300,000 400,000 500,000

FIGURE 11. 2 1 A comparison of the ring systems of the four giant planets.
what astronomers call a diffuse ring, a ring with no dis- edge of the A Ring—they are extremely thin. From our pre-
tinct boundaries. We’ll say more about diffuse rings in vious discussion, you might guess that they could be no
the next subsection. thicker than the diameter of the larger ring particles. But
Although Saturn’s bright rings are very wide—more than in these densely packed rings, there is simply not enough
62,000 km from the inner edge of the C Ring to the outer room to jam all of the particles into the same plane, so they
FIGURE 11. 22 A Hubble

Space Telescope image showing
Saturn and its A Ring (the
outermost bright ring), B Ring
(the middle bright ring), Cassini
Division (the wide, relatively
dark division between the A and
B rings), and Encke Gap (the
narrow division near the outer
edge of the A Ring)
A Ring B Ring Cassini Encke

Division Gap
the Cassini Division, you could paint a dark stripe 1.5 cm

wide and about 12 cm from the outer edge. After mount-
ing the paper ring around the basketball, you would have
a splendid model of Saturn and its rings. Unfortunately,
your model would not be completely to scale; the paper
rings would be more than a thousand times too thick! If
you wanted to make your rings from paper similar to that
used in this book, the planet would have to be a ball 250
meters in diameter, and the paper rings would have to
extend over the length of six football fields! The diameter
of Saturn’s bright ring system is 10 million times the thick-
ness of the rings.
Other Planets, Other Rings

FIGURE 11. 23 This Voyager 2 image of the outer B Ring of Ring structure among the other giant planets is not as
Saturn shows so many ringlets and minigaps that it looks like a diverse as Saturn’s. Most rings other than Saturn’s are quite
close-up of an old-fashioned phonograph record. The narrowest narrow, although a few are diffuse. Diffuse rings tend to be
features are only 10 km across, at the limit of resolution. Even sparsely populated and have characteristics quite different
finer structure was noted during stellar occultations by the rings, from Saturn’s tightly packed bright rings. The separations
as observed by the Voyager photometer and imaged later by between particles in diffuse rings are generally very large,
Cassini. The cause of most if this structure is unknown.
and the occasional collisions between particles can cause
their individual orbits to become eccentric, inclined, or
both. Because it is unlikely
settle down as close as they can get to the ring plane. Sat- that these disturbed orbits Rings around the
urn’s bright rings are thus no will experience a restoring other giant planets
more than a hundred meters Saturn’s bright rings collision, the particles tend to are mostly narrow.
and probably only a few tens are exquisitely thin. remain in their noncircular,
of meters from their lower to non-coplanar orbits. For this reason, diffuse rings spread
upper surfaces. The extremes between their width and their out horizontally and thicken vertically, sometimes without
thickness can be difficult to picture, but let’s try. any obvious boundaries. These characteristics make diffuse
Suppose you want to make a scale model of Saturn and rings difficult to detect at most viewing angles. When seen
its rings using a basketball to represent Saturn. The basket- under backlit conditions however, diffuse rings appear to
ball is about 20 cm in diameter. You could make the three light up because of the strong forward scattering of very
bright rings out of paper by cutting a circle 45 cm in diam- small ring particles by sunlight (see Excursions 8.2 and
eter, with a 25-cm hole cut from the center. To represent Excursions 11.1).
E X C U R S I O N S 1 1 .1
The Backlighting Phenomenon

Diffuse rings contain tiny particles that show up best when we FIGURE 11. 24
are looking backward toward the Sun. You can witness this Backlighting of a person’s
same effect by observing a familiar phenomenon here on Earth. hair creates a halo effect.
The light coming from
Pebbles and boulders are easiest to see when the light illuminat-
behind the subject is
ing them is coming from behind. Dust and other small particles,
scattered in a direction
however, stand out most strongly when you look into the light.
toward the photographer.
For example, notice how dust particles on your windshield This effect is caused by
appear brightest when you are driving toward the Sun. forward scattering of small
Photographers call this effect backlighting. A photographer particles. Typical human
will often place his subject in front of a bright light to highlight hair has a thickness only a
her hair (Figure 11.24). Under backlit conditions, individual few times the wavelength of
strands of hair shine brightly, creating a halo effect around the visible light.
person’s face. This effect happens when light falls on very small
objects—those with dimensions a few times to several dozen
times the wavelength of light. Human hair is near the upper ring particles scatter very little sunlight back toward the Sun
end of this range. Light falling on strands of hair is not scat- and the approaching spacecraft. When the spacecraft passes
tered uniformly in all directions; rather, it tends to continue in by the planet and looks backward in the general direction of
the direction away from the source of illumination. Very little the Sun, these same dusty rings suddenly appear as a circular
of the light is scattered off to the side, and almost none is scat- blaze of light, much like a halo surrounding the nighttime hemi-
tered back toward the source. sphere of the planet. Many planetary rings are best seen with
Some of the dustier planetary rings are filled with particles backlighting, and some have been observed only under these
that are just a few times larger than the wavelength of visible conditions. Backlighting has also been remarkably successful
light. To a spacecraft approaching from the direction of the Sun, in revealing tiny ice crystals erupting from cryovolcanoes on
such rings may be difficult or even impossible to see. These tiny Enceladus, as shown dramatically in Figure 11.8b.
When Voyager scientists looked at Jupiter’s ring system Ring, material in this ring may be spiraling into the top of
with the Sun behind the camera, all they saw was a nar- the Uranus atmosphere. When viewed under backlit con-
row, faint strand. But when they looked back toward the ditions by Voyager 2 (Figure 11.26b), the space between the
Sun while in the shadow of the planet, Jupiter’s rings sud- Uranus rings turned out to be filled with dust, much as in
denly blazed into prominence. Figure 11.25a shows a nearly Jupiter’s ring system.
edge-on, backlit view of Jupiter’s rings taken by the Galileo Four of Neptune’s six rings are very narrow, similar
spacecraft. Most of the material in Jupiter’s rings is made to the 11 narrow rings surrounding Uranus. The other
up of fi ne dust dislodged by meteoritic impacts on the sur- two have widths of a few thousand kilometers (see Figure
faces of Jupiter’s small inner moons, shown orbiting among 11.21d). Neptune’s rings are named for 19th century astrono-
the rings in Figure 11.25b. mers who made major contributions to Neptune’s discovery.
Nine of the 13 rings of Uranus are very narrow and Among Neptune’s six rings, it is Adams Ring that attracts
widely spaced relative to their widths. Most are only a few our greatest attention (Figure 11.27). Much of the material in
kilometers wide, but they are many hundreds of kilometers the Adams Ring is clumped together into several ring arcs.3
apart (see Figure 11.21c and Figure 11.26a). Two rings, dis- These ring segments extend over lengths of 4,000–10,000 km
covered by the Hubble Space Telescope in 2005, are much yet are only about 15 km wide. When first discovered, the
wider and more distant than the narrow rings. With the ring arcs were a puzzle, because mutual collisions among
outermost nearly twice as far from the planet as the outer- their particles should cause them to be spread more or less
most of the previously known rings, they constitute what uniformly around their orbit. Most astronomers now attri-
is being called an “outer ring system.” The most prominent bute this clumping to orbital resonances (see Chapter 10)
Uranus ring, the Epsilon Ring, is eccentric and the wid-
est of the planet’s inner narrow rings, varying in width 3
The names of the three most prominent ring arcs—Liberté, Égal-
between 20 and 100 km. The innermost ring is wide and ité, and Fraternité— are taken from the famous motto of the French
diffuse, with an undefi ned inner edge. As with Saturn’s D Revolution.
(a)
FIGURE 11. 25 (a) A backlit
Galileo view of Jupiter’s
rings. Note also the forward
scattering of sunlight by tiny
particles in the upper layers
of Jupiter’s atmosphere. (b)
A diagram of the Jupiter ring
system and the small moons
that form them.
(b)
Gossamer rings Halo

Main ring
Amalthea Adrastea
Metis
Thebe
(a) (b)
FIGURE 11. 26 The
appearance of rings depends
dramatically on lighting
conditions. Here the rings of
Uranus appear as narrow,
faint bands when viewed with
the Sun at our back (a) but
burst into dazzling brilliance
when illuminated from behind
(b). (Bright lines in the images
are stars “streaked out” by
the spacecraft motion during
exposures.)
with the moon, Galatea, that orbits just inside the Adams images taken by Voyager in 1989. One of the ring arcs, Lib-
Ring. Recent groundbased imaging suggests that some parts erté, may disappear entirely before the end of the century.
of the Neptune rings may be unstable. Images obtained early The Adams Ring is not unique. Uranus’s Lambda Ring and
in 2002–2003 show decay in the ring arcs when compared to Saturn’s G Ring also show ring arcs.
organic materials and ices that have been radiation-dark-

ened by high-energy, charged particles in the planets’ mag-
netospheres. (Radiation processing of organic ices, such as
methane, blackens the ice by
releasing carbon from the ice’s The rings of Uranus
molecular structure.) Jupiter’s and Neptune are
rings are neither as bright as darker than coal.
Saturn’s nor as dark as those
of Uranus and Neptune, suggesting that they may be rich
in dark silicate materials, like the innermost of Jupiter’s
small moons.
The jumble of fragments that make up Saturn’s rings
is easily understood as a product of tidal disruption of a
moon or planetesimal (see Chapter 10), but moons can con-
tribute material to rings in other ways as well, as we see in
the case of Jupiter’s ring system. The last view that Voyager
had at Jupiter’s rings carried hints of things unseen, and so
matters would remain for 20 years. It is ironic that for so
FIGURE 11. 27 A Voyager 2 image of the three brightest many years we knew less about Jupiter’s rings than about
arcs in Neptune’s Adams Ring. Neptune itself is very much any other ring system in the Solar System; yet as a result of
overexposed in this image. Galileo’s arrival at the planet, Jupiter’s ring system is now
among the best observed (see Figures 11.21a and 11.25).
The brightest of Jupiter’s rings is a narrow strand only
6,500 km across, consisting of material from Metis and
The Composition of Ring Material Adrastea (see Figure 11.25). These two moons orbit in
As we learned in Chapter 10, some planetary rings are Jupiter’s equatorial plane, and the ring they form is thin.
created when a moon or planetesimal comes within a Beyond the main ring, however, are the very different gos-
planet’s Roche limit and is torn apart by tidal stresses. samer rings, so called because they are extremely tenuous.
The fragmented pieces of the disrupted body are then dis- The gossamer rings are supplied by dust from the moons
tributed around the planet in the form of a ring. The com- Amalthea and Thebe. Unlike the main ring, the gossamer
position of the ring will, of rings are rather thick; the inner gossamer ring, associated
course, be the same as that of Saturn’s rings are with Amalthea, is actually located within the outer gos-
the fated body. Such is the composed primarily samer ring formed of material from Thebe. These rings are
case for Saturn’s bright ring of water ice. so thick because the orbits of the moons that supply the
system. Saturn’s bright rings ring material are slightly tilted with respect to Jupiter’s
appear bright because they reflect about 60 percent of the equatorial plane. The orbital planes of these moons wob-
sunlight falling on them. We might suspect from their ble, as does the orbital plane of Earth’s Moon; but instead
brightness alone that they are made of water ice, and spec- of taking almost 19 years to complete one wobble (as our
tral observations confi rm this suspicion, clearly showing Moon does), these moons complete a wobble in only a few
the distinct signature of water. A slight reddish tint to the months. The gossamer rings, made up of material from
rings tells us that they are not made of pure ice but must these wandering moons, are spread as far below and above
be contaminated with other materials, such as silicates. Jupiter’s equatorial plane as the orbits of the satellites that
The icy moons around Saturn or the frozen comets that form them.
prowl the outer Solar System could easily provide this The innermost ring in Jupiter’s system, called the Halo
material. Ring, consists mostly of material from the main ring. As the
Saturn’s bright rings are the brightest in the Solar Sys- dust particles in the main ring drift slowly inward toward
tem and are the only ones that we know are composed of the planet, they pick up electric charges and are pulled
water ice. In stark contrast, the rings of Uranus and Neptune into this rather thick torus, or doughnut-shaped ring, by
are among the darkest objects known in our Solar System. electromagnetic forces associated with Jupiter’s powerful
Only 2 percent of the sunlight falling on them is reflected magnetic field.
back into space, making the ring particles blacker than coal Finally, moons may contribute ring material through
or soot. No silicates or similar rocky materials are this dark.4 volcanism. Volcanoes on Jupiter’s moon Io continually eject
The rings of Uranus and Neptune are likely composed of sulfur particles into space, many of which drift inward
under the influence of pressure from sunlight and fi nd their
4
way into the Jupiter ring. The particles in Saturn’s E Ring
A number of carbon-rich meteorites are nearly this dark, as is the
are ice crystals ejected by cryovolcanism from the moon
organic-rich material that makes up the nucleus of Comet Halley.
Enceladus, which is located in the very densest part of the that of Mimas, Saturn’s small icy moon that is less than 400
E Ring (see Figure 11.7). km in diameter. The amount of material in the narrow rings
is, of course, much less. All of the particles in the largest
ring of Uranus, the Epsilon Ring, could be compressed into
Moons Maintain Order a single body no more than 20 km across. All of the mate-
rial in both the Neptune rings and the Jupiter ring could fit
and Create Gaps into single objects only a few kilometers in diameter.
In Chapter 10 we discussed the 2:1 orbital resonance between
Saturn’s moon Mimas and particles in Saturn’s rings that
creates the Cassini Division. Jupiter creates the Kirkwood
gaps in the asteroid belt by this same mechanism. Similarly,
Strange Things among the Rings
we know that one of the gaps in Saturn’s C Ring is caused Among the many strange rings imaged by Voyager, the arche-
by a 4:1 resonance between the ring particles and Mimas. type is clearly Saturn’s F Ring, shown in Figure 11.28. Images
Unfortunately, the cause of many—perhaps we should say of the ring taken by Pioneer 11 a year earlier had shown noth-
most— other gaps in Saturn’s rings remains unexplained. If ing out of the ordinary, but the first high-resolution images of
they also are produced by resonances, we have yet to iden- the F Ring made by Voyager had spacecraft scientists staring
tify the source. One possibility is that these gaps are the in disbelief. The ring was separated into several strands that
result of collisions between ring particles. Any collision appeared to be intertwined. Some media reporters quickly
between two ring particles will cause one of the particles claimed that the F Ring was “disobeying the laws of physics.”
to move to an orbit farther out, and the other to an orbit This, of course, was not the case; but ready explanations for
farther in. Over time, this process could sweep some areas this seemingly non-Keplerian behavior of the ring particles
clean of ring particles, forming gaps while piling those ring were not immediately forthcoming. And as if the multiple
particles up in narrow ringlets between the gaps. strands were not perplexing enough, the ring also displayed
There are also important gravitational interactions what appeared to be a number of knots and kinks.
between ring particles themselves. These interactions deter- Saturn’s F Ring is now understood to be a dramatic
mine the ring shapes at the edges of gaps. In fact, analysis example of the action of a pair of shepherd moons (see Chap-
of the shapes of ring edges enables us to estimate the masses ter 10). The F Ring is flanked
of the rings. Even though by Prometheus (a moon that Rings can be
planetary rings can be large The mass of all orbits 860 km inside the ring) distorted by the
and prominent, they account Saturn’s rings and Pandora, a second moon, gravitational influence
for only the tiniest fraction of combined is about the which orbits 1,490 km on the of nearby moons.
the mass of the material same as the mass of outside (Figure 11.29a). Both
around a giant planet. Sat- a small icy moon. moons are irregular in shape, with average diameters of 85
urn’s bright rings are by far and 80 km, respectively. Because of their relatively large
the most massive rings in the Solar System. In fact, they size and proximity, they exert significant gravitational
contain more material than all other planetary rings com- forces on nearby ring particles. The resulting tug-of-war
bined. Even so, their total mass is estimated to be less than between Prometheus pulling ring particles in its vicinity
FIGURE 11. 28 (a) A Voyager

(a) (b)
1 image of Saturn’s “twisted” F
Ring. (b) A Cassini view of the F
Ring and the 80-km-diameter
moon Pandora.
FIGURE 11. 29 (a) Cassini image of

(a) (b)
Saturn’s F Ring and its shepherd moons:
80-km-diameter Pandora (left) and 85-km
Prometheus (right). (b) Voyager 2 image
of Uranus’s Epsilon Ring with its 40-km
shepherds, Cordelia and Ophelia.
Ophelia
Cordelia
to larger orbits and Pandora drawing its neighbors into structure that must be related to a 20-km-diameter moon,
smaller orbits is the cause of the bizarre structure that origi- named Pan, that orbits within the gap. The rings in the
nally baffled scientists and reporters alike. Encke Gap are unusual but not unique, for they bear some
The F Ring is not an isolated case. The 360-km-wide resemblance to the arcs in Neptune’s Adams Ring.
Encke Gap in the outer part of Saturn’s A Ring contains two If shepherd moons, such as those shown in Figure
narrow rings that show knots and gaps (Figure 11.30a)—a 11.29b, are in eccentric or inclined orbits, they cause the
FIGURE 11. 30 (a) One of

(a) (b)
two discontinuous and knotted
rings within Saturn’s Encke
Gap. (b) A higher-resolution
view of the Encke Gap reveals
a scalloped pattern along the
division’s inner edge caused by
the small moon Pan. (c) The tiny
moon Daphnis disrupts both
edges of Saturn’s Keeler Gap.
(c)
FIGURE 11.31 Spokes in

(a) (b)
Saturn’s B Ring appear dark in
normal viewing (a) but bright
with backlighting (b), indicating
that the spoke particles are
very small.
confi ned ring to also be eccentric or inclined, as is the case B Ring particles and are absorbed. Such a model can nicely
for Uranus’s Epsilon Ring. Because shepherd moons can be explain what causes the spokes to disappear, but it does not
so small, they often escape detection. According to current tell us why they appear.
theories of ring dynamics, a number of still unknown shep- One hypothesis links the origin of the spokes with mete-
herd moons must be interspersed among the ring systems oroid impacts on large ring particles. As they strike these
of the outer Solar System. particles, meteoroids can collide with so much energy that
Small moons orbiting within ring gaps can also disturb they create an ionized cloud of tiny charged particles— a
ring particles along the edges of the gaps. Figure 11.30b plasma— that becomes briefly suspended above the ring
shows the scalloped pattern, caused by Pan, that is found plane. Before the cloud can return to the ring plane, Sat-
along the inner edge of the Encke Gap. Figure 11.30c catches urn’s magnetic field causes the charged particles to drift
the 7-km-diameter moon Daphnis in the act of disrupting outward, creating the spokelike features. Still, important
the inner and outer edges of Saturn’s 35-km-wide Keeler questions remain. For example, the spokes seem to form
Gap, located near the outer edge of the A Ring. primarily in the “morning” sector of the rings, which tends
Gravitational interactions do not hold a monopoly on to argue against the meteorite hypothesis. In addition, our
strange behavior among ring systems. One of the more puz- model has no explanation for the fact that Voyager and
zling discoveries made by Voyager 1 at Saturn was the Cassini have imaged spokes only in the outer part of the
appearance of dozens of dark spokelike features in the outer B Ring and not in the inner part of the ring or in either
part of the B Ring (Figure of the other two wide rings. When Voyager observed the
11.31). As they formed, the Saturn’s B Ring spokes in 1980–81, Saturn was near equinox. But when
spokes tended to grow in a has transient Cassini arrived at Saturn in 2004, 5 years prior to equi-
more or less radial direction, radial features nox, the spokes were at fi rst not visible. So there is even
but they were distorted by called spokes. some speculation that Saturn’s ring spokes may be a sea-
“Keplerian shear,” which sonal phenomenon. As of now, three decades after their
causes the outer tips to rotate more slowly than the inner discovery, the spokes in Saturn’s rings remain as myste-
tips. No single spoke was seen to last for more than half an rious as ever.
orbit. This half-orbit survival tells us that the particles in
the spokes must be suspended above the ring plane, prob-
ably by electrostatic forces. Why is this obvious? Any par-
ticle that is not in the ring plane must be in an inclined
Planetary Rings Are Ephemeral
orbit, and it thus has to pass through the plane twice dur- Planetary rings do not have the long-term stability of most
ing each orbit of the planet. As the spoke particles try to Solar System objects. Ring particles are constantly collid-
pass through the densely packed B Ring, they run into the ing with one another in their tightly packed environment,
either gaining or losing orbital energy as they do so. This Any number of comets or asteroids must have passed within
redistribution of orbital energy can cause particles at the Earth’s Roche limit5 and disintegrated catastrophically into
ring edges to leave the rings and drift away, aided by non- a swarm of small fragments, thereby creating a temporary
gravitational influences such ring. Yet Earth lacks shepherd moons to provide orbital sta-
as the pressure of sunlight. Rings don’t bility to rings. Interactions between ring particles would
Although various orbital reso- last forever. have caused such a ring to spread out and dissipate, and the
nances with moons may help inner parts of a ring around Earth would feel the drag of
guide the orbits of ring particles and delay the dissipation Earth’s extended atmosphere and spiral inward, creating
of the rings themselves, at best this can be only a temporary spectacular meteor displays as they fell. A similar absence
holding action. Most planetary rings eventually face their of small inner moons also prevents Venus and Mercury from
inevitable fate—total dissipation. At least one ring, however, keeping rings over geological timescales. We leave Mars off
seems immune from this eventual demise: as we learned the list for now. Although we know of no ring around Mars,
earlier, volcanic emissions from Saturn’s moon Enceladus its two tiny moons—Phobos and Deimos— could, in prin-
are constantly supplying icy particles to Saturn’s E Ring, ciple, serve to shepherd a collection of orbiting debris.
replacing those that drift away. The E Ring will survive for Planetary rings in the outer Solar System will continue
as long as Enceladus remains geologically active. to come and go over the eons ahead for as long as the giant
It is quite unlikely that any of the planetary rings we see planets maintain the small moons that provide temporary
today have existed in their ring stability. Whether any will rival the splendor of Saturn’s
current form since the Solar Earth’s lack of bright ring system that we see today is, of course, unknown.
System’s beginning. Indeed, it shepherd moons We may be living in a fortuitous time when Saturn is put-
is far more likely that many explains why it ting on its best face.
ring systems have come and has no rings.
gone over the history of the
Solar System. Even our own planet has probably had several 5
Earth’s Roche limit is about 25,000 km for rocky bodies and more
short-lived rings at various times during its long history. than twice that for icy bodies.
the outer reaches of the disk that surrounded the newly

Seeing the Forest for the Trees formed Sun nearly 4.6 billion years ago. As such, these
bodies offer important clues about the history of our
We normally think of the Solar System as consisting of Solar System, allowing us to test our ideas of what that
the planets we learned about in grade school. Yet there young Solar System must have been like. These worlds
are far more moons in our Solar System than there are may even give us clues about the history of life in the
planets, and each moon is a unique world in its own universe. The fractured ice floes that make up the surface
right. There was no way in 1610 for Galileo to know that of Europa may well cover a vast, deep ocean, warmed
the four points of light he had discovered circling giant and enriched by geothermal vents not unlike those that
Jupiter were the flagships of a vast armada of strange dot the floors of Earth’s oceans. When scientists fi nally
and amazing worlds. The diversity of the solid worlds realized that Mars and Venus were not as Earth-like as
of the inner Solar System, while remarkable, is noth- once imagined, prospects for finding life elsewhere in the
ing compared with the diversity of the worlds that sur- Solar System seemed dim. Now, in the opening decades
round Jupiter, Saturn, Uranus, and Neptune. Some of of the 21st century, we are turning our attention with
these worlds are frozen remnants of the time long ago fresh excitement and hope to these small worlds that
when the giant planets formed at the centers of their own circle far from the Sun.
swirling disks. Others are testament to chaotic events Systems of moons may offer a fascinating future for
in which passing objects were captured in gravitational geologists; but they also represent gravitational play-
interactions between planets and other moons. At the grounds in which the simple two-body interactions
other extreme are moons such as Io, a world so geologi- described by Kepler’s and Newton’s laws give way to
cally active that its surface is remaking itself through complex, chaotic interactions among many different
volcanism as we watch. objects. The tidal stresses we learned about in Chapter
These moons are not the uniform worlds, with iron- 10 also rip and pull at these worlds, locking them into
and-nickel cores and silicate mantles, found close to the synchronous orbits and in some cases churning their
Sun. Instead they are assembled from the diverse mix interiors into seething cauldrons of geological activity.
of metals, rocks, and ices that existed in solid form in Sometimes these tidal stresses are so great that a moon or
passing comet pays the ultimate price and is ripped into when we have more time. Planetary scientists also share
tiny fragments. Born from these violent events are the our wistful regret. Our glances from passing spacecraft
majestic rings that circle Saturn and the less spectacu- at the moons and rings surrounding the giant planets
lar but equally fascinating systems of rings surrounding are really only enough to whet our appetites. No doubt,
the other giant planets. The extraordinary complexity of lifetimes of wonder and surprise remain as we continue
these rings is also a testament to Newton’s remarkable to explore the smaller worlds of our Solar System.
inverse square law of gravitation as planets, moons, and There may be more moons than there are planets in the
rings do their gravitational dance. These interactions can Solar System, but planets, moons, and rings do not complete
even distort rings into structures so remarkable that they the accounting of objects adrift within the gravitational
seemed all but impossible until revealed by the eyes of realm of the Sun. Still remaining are the dwarf planets and
the Voyager and Cassini spacecraft. the uncounted swarms of asteroids and comets that orbit
It is with sadness that we leave this realm of moons within the domain of the planets and also stretch halfway
and rings so quickly; these worlds deserve much more to the nearest star. In the next stage of our journey we will
attention than we can give them on our journey. As we discover that, far from being insignificant, this flotsam and
might do on a package tour promising to show us Europe jetsam carries with it the most direct record of the history of
in 3 days, we look at the marvels through the window, our Solar System, providing essential pieces to the puzzle
glance at our guidebooks, and promise to return someday of the origin and history of life on Earth.

• The moons of the outer Solar System are composed of THINK ING A BOUT THE CONCEP T S
rock and ice.
1. Which among the planets in our Solar System do not
• Some moons were formed along with their parent plan- have moons?
ets; others were captured later.
2. Explain the difference between regular and irregular
• A few moons are geologically active; most are dead. moons.
• Jupiter’s moon Io is the most volcanically active body 3. Identify the three moons known to be geologically
in the Solar System. active.
• Saturn’s moon Enceladus is erupting water-ice crystals *4. Explain the process that drives volcanism on Jupiter’s
into space. moon Io.
• Jupiter’s moon Europa may contain an enormous sub- 5. Describe cryovolcanism and explain its similarities and
surface ocean. differences with respect to terrestrial volcanism.
• All four giant planets are surrounded by rings. 6. Which moons show evidence of cryovolcanism?
• Saturn’s ring system is the most complex. 7. Europa and Titan may both be geologically active. What
leads astronomers to suspect such activity?
• Some rings may be transient features held in place by
moons. 8. Discuss evidence supporting the idea that Europa might
have a subsurface ocean of liquid water.
• Planetary rings are ephemeral.
9. Titan contains abundant amounts of methane. What
process destroys methane in this moon’s atmosphere?
SmartWork, Norton’s online homework system, 10. In certain ways, Titan resembles a frigid version of Earth.
includes algorithmically generated versions of these Explain the similarities.
11. Some moons display signs of geological activity in the
past. Identify some of the evidence for past activity.
12. Name three geologically dead moons.
13. Name the planets of the Solar System known to have 27. Enceladus has a mass of 8.4 × 1019 kg and a radius of 250
rings. km. Calculate the minimum speed at which ice crystals
from the moon’s cryovolcanoes must be traveling in order
14. Describe a groundbased technique that has led to the
to escape to Saturn’s E Ring.
discovery of rings around the outer planets.
*28. Planetary scientists have estimated that Io’s extensive
15. What are ring arcs and where are they found?
volcanism could be covering the moon’s surface with
*16. Identify and explain two possible mechanisms that can lava and ash to an average depth of up to 3 millimeters
produce planetary ring material. (mm) per year.
a. Io’s radius is 1,815 km. If we model Io as a sphere,
17. Will the particles in Saturn’s bright rings eventually
what are its surface area and volume?
stick together to form one solid moon orbiting at the
b. What is the volume of volcanic material deposited on
mean distance of all the ring particles? Explain your
Io’s surface each year?
answer.
c. How many years would it take for volcanism to per-
18. How does the mass of a planet’s rings compare with the form the equivalent of depositing Io’s entire volume
mass of its individual moons? on its surface?
d. How many times might Io have “turned inside out”
*19. Explain two mechanisms that create gaps in Saturn’s
over the age of the Solar System?
bright ring system.
29. Particles at the very outer edge of Saturn’s A Ring are
20. Under what lighting conditions are the tiny dust par-
in a 7:6 orbital resonance with the moon Janus. If the
ticles found in some planetary rings best observed?
orbital period of Janus is 16 hours 41 minutes (16h41m),
21. Explain ways in which diffuse rings differ from other what is the orbital period of the outer edge of Ring A?
planetary rings.
*30. The inner and outer diameters of Saturn’s B Ring are
22. Describe and explain a mechanism that keeps planetary 184,000 and 235,000 km, respectively. If the average
rings from dissipating. thickness of the ring is 10 meters and the average den-
sity is 150 kilograms per cubic meter (kg/m3), what is
23. Astronomers believe that most planetary rings eventu-
the mass of Saturn’s B Ring?
ally dissipate. Explain why they do not last forever.
31. The mass of Saturn’s small icy moon Mimas is 3.8 ×
24. Name one ring that might continue to exist indefinitely,
1019 kg. How does this mass compare with the mass of
and explain why it could survive when others might not.
Saturn’s B Ring, as calculated in the previous question?
25. Why does Earth not have a ring? Why is this comparison meaningful?
26. Io has a mass of 8.9 × 1022 kg and a radius of 1,820 km. vides a Study Plan for each chapter of 21st Century
a. Using the formula provided in Chapter 3, calculate Astronomy. Study Plans include animations, read-
Io’s escape velocity. ing outlines, vocabulary flashcards, and multiple-choice
b. How does Io’s escape velocity compare with the 1-km/s quizzes, plus links to premium content in SmartWork and
vent velocities from the moon’s volcanoes? the ebook. Visit www.wwnorton.com/studyspace.
Almost in the center of it, above the Prechistenka Boulevard,
surrounded and sprinkled on all sides by stars … shone the
enormous and brilliant comet of 1812—the comet which was
said to portend all kinds of woes and the end of the world.
LEO TOLS TOY (1828 –1910), WAR AND PEACE
Comet Hyakutake approached close to Earth in 1996.

CHAP TER 12
Dwarf Planets and Small

Solar System Bodies
12.1 Leftover Material: From
K E Y CONCEP T S
the Small to the Tiniest
Dwarf planets are among the lesser worlds in the Solar
We began our discussion of the Solar System in Chapter System. Asteroids and comets are even smaller; yet
6 with the story of its history. We learned that very early these objects, and their fragments that fall to Earth as
on— at the same time our Sun was becoming a star—tiny meteorites, have told us much of what we know about
grains of primitive material were sticking together to pro- the early history of the Solar System. As we explore
duce swarms of small bodies called planetesimals. Those these pieces of interplanetary flotsam and jetsam, we
that formed in the hotter, inner part of the Solar System will discover
were composed mostly of rock and metal, while those in
the colder, outer parts were made up of ice, organic com- • The littlest of planetary bodies, which we call dwarfs.
pounds, and rock. We also learned what happened to many • Small, irregular worlds called asteroids that are made
of these planetesimals. Some were consumed during the era of rock and metal.
of large-body building to become planets and moons, and • Different types of meteorites that are fragments of
others were ejected from the Solar System by gravitational
these varieties of asteroids.
encounters with larger bodies. However, not all of these
primitive bodies suffered the same fate. Many of them are • That some asteroids are primitive while others are
still around, and they remain a small, but scientifically differentiated or are pieces of larger, differentiated
important, component of our present-day Solar System. As bodies.
we now conclude our discussion of the Solar System, we • That most asteroids orbit between Mars and Jupiter,
fi nd that we have come full circle. These remaining plan- but some have orbits that cross Earth’s.
etesimals, and the fragments they continually create, pro-
• Comet nuclei—pristine, icy planetesimals—adrift in the
vide planetologists with the opportunity to look back to the
physical and chemical conditions of the earliest moments frozen outer reaches of the Solar System.
in the history of the Solar System. • Spectacular active comets that are warmed by the Sun
Among the larger surviving planetesimals are five bod- as they dive through the inner Solar System.
ies known as dwarf planets (Table 12.1). Four of them— • World-jarring meteorite impacts that still occur today
the former planet Pluto and the more recently discovered and have played a vital role in shaping the history of life
Eris, Haumea, and Makemake— are an icy-organic-rocky
on Earth.
type of planetesimal. All four reside in the frigid region
beyond the orbit of Neptune. The fifth, the former asteroid • Meteor showers that occur when Earth passes
Ceres, orbits within the main asteroid belt between Mars through a comet’s trail of debris.
327
328 Chapter 12 Dwarf Planets and Small Solar System Bodies
TA B L E 1 2 .1
Physical Properties of the Dwarf Planets
Ceres Pluto Haumea Makemake Eris
Orbital radius (AU) 2.8 39.5 43.1 45.7 67.7
Orbital period (years) 4.6 248.1 283.3 309.9 557
Orbital velocity (km/s) 17.9 4.7 4.5 4.4 3.4
Mass (M⊕ = 1) 0.00016 0.0021 0.00067 ~0.0007 0.0028
Equatorial diameter (km) 975 2,306 ~1,400 ~1,500 ~2,400
Equatorial diameter (D⊕ = 1) 0.07 0.18 0.11 ~0.12 0.19
Density (water = 1) 2.1 2.0 ~3 ~2 ~2
Rotation period (hours) 9.1 153.3 3.9 7.8 >8?
Obliquity (degrees) ~3 119.6 ? ? ?

2
Surface gravity (m/s ) 0.27 0.81 0.44 ~0.5 ~0.6
Escape speed (km/s) 0.51 1.27 0.84 ~0.8 ~1.3
and Jupiter. As such, it is a member of the rock-metal group position. It became the Solar System’s ninth planet and
of planetesimals. was named Pluto for the Roman god of the underworld. In
Smaller than dwarf planets, yet respectable in size, are the years that followed, however, observational evidence
the rocky asteroids and the icy comet nuclei (the nuclei of began to indicate that the mass of Pluto was far too small
comets that orbit beyond Neptune and have yet to come into to have produced the presumed perturbations in the orbits
the inner Solar System). These remote icy planetesimals of Uranus and Neptune. When astronomers reanalyzed
are known as Kuiper Belt objects (KBOs) or sometimes as the 19th century observations, they found that the orbital
“trans-Neptunian objects” (TNOs.) “discrepancies” were in fact erroneous. Pluto’s discovery
At the very smallest scale are meteoroids and interplan- thus turned out to be a strange and improbable coincidence
etary dust—the widespread debris created by collisions based on faulty data.
among asteroids and the disintegration of comet nuclei by Pluto’s largest moon, Charon, was fi rst observed in
solar heating. 1978. By applying Kepler’s and Newton’s laws to observa-
tions of the moon’s motion
around Pluto, astronomers Pluto has a moon half
were able to accurately as big as it is.
12.2 Dwarf Planets: “weigh” the Pluto-Charon
system. Its total mass turned out to be only about 1/400
Pluto and Others that of Earth (Figure 12.1). Pluto is only two-thirds as large
as the Moon and only twice as big as its largest moon,
Pluto has been an enigma since its discovery. The story Charon. Because of the relative similarity in size between
begins with what appeared to be discrepancies between Pluto and Charon, we might think of the Pluto-Charon
the observed and predicted orbital positions of Uranus system as a “double planet.” Moreover, Pluto and Charon
and Neptune throughout the 19th century. In the early are a dynamically locked pair— the only known example
20th century, astronomers began a search for the unseen in the Solar System. The two are in synchronous rotation
body that they believed was responsible for perturbing with one another, so each has one hemisphere that always
the orbits of Uranus and Neptune. They called it Planet faces the other body and another that never sees the other
X and estimated that it had six times Earth’s mass and body (Figure 12.2). If immobile inhabitants were living
was located somewhere beyond Neptune’s orbit. Planet X on opposite sides of Pluto and Charon, they would never
was fi nally found by the American astronomer Clyde W. know of each other’s existence. In 2005, two smaller moons
Tombaugh (1906–1997) in 1930, not far from its predicted were found orbiting Pluto. All three of its moons appear
12.2 Dwarf Planets: Pluto and Others 329
to be in orbital resonance with one another. As with Ura-

(a)
nus, the plane of Pluto’s equator is tipped at almost right
angles to its orbital plane.
Pluto’s eccentric orbit periodically brings it inside Nep-
tune’s orbit, which is nearly circular, and from 1989 to 1999
Pluto was closer to the Sun
than Neptune was. More than Pluto is a mixture
248 Earth years are required of rock and ice and
for Pluto to complete one orbit. has a thin methane
Since its discovery, Pluto has atmosphere.
traveled less than a third of its
way around the Sun. With densities nearly twice that of
(b) water, both Pluto and Charon probably consist of a rocky
core that makes up about 70 percent of their mass, sur-
rounded by a water-ice mantle. Because of their great dis-
tance from Earth, Pluto and Charon were not included in
the great program of spacecraft exploration that took place
late in the 20th century. Therefore, we have limited infor-
mation about their surface properties, and we know nothing
about their geological history. Spectra obtained with ground-
based telescopes and the Hubble Space Telescope tell us
that Pluto’s surface contains an icy mixture of frozen water,
carbon dioxide, nitrogen, methane, and carbon monoxide.
FIGURE 12 .1 The dwarf planet Pluto and its moon Charon in a Unlike Pluto, Charon’s surface seems to be made up primar-
Hubble Space Telescope image (a) and an artist’s rendition based ily of dirty water ice. Pluto has a temporary, low-density
on analysis of HST images (b). Pluto and Charon are so similar in atmosphere composed mostly of nitrogen, methane, and
size they could be considered a “double planet.” carbon monoxide. As Pluto moves farther from the Sun and
temperatures drop, its atmosphere will gradually freeze and
fall to the ground as snow. We should soon know much more
about Pluto and Charon. In 2006, NASA launched a plan-
Pluto and Charon are etary spacecraft called New Horizons that will reach the
tidally locked to each other “double planet” in 2015.
in inclined, retrograde orbits. As more was learned about Pluto and other bodies beyond
Neptune’s orbit, some astronomers called into question Plu-
to’s status as a planet, and a heated debate ensued. The dis-
covery of a distant body, later named Eris, added fuel to the
already fiery argument. Eris (Figure 12.3) was found to be
even larger and more massive than Pluto, but it was similar
to Pluto in other respects. Each is attended by a relatively
Pluto large moon, although Pluto has two other smaller moons.
Both Pluto and Eris have frozen methane on their surfaces. At
this point, the inevitable question emerged: should astrono-
mers consider Eris to be the Solar System’s tenth planet— or
Center should neither Pluto nor Eris be called planets?
of mass
The International Astronomical Union (IAU) made its
decision in August 2006. “To be a planet, a body must be
massive enough to (1) pull itself into a round shape and (2)
be able to clear the neighborhood around it.” (Details of
the IAU resolution can be found in Appendix 8.) Pluto is
round like the classical planets, but it is not able to clear its
neighborhood. With that simple pronouncement, Pluto lost
Charon its status as the “ninth planet.” Although the IAU is consid-
Center-of-mass ered by professional astronomers to be the fi nal authority
orbit about the Sun
on such matters, not all were happy with its decision, and
FIGURE 12 . 2 A diagram showing the doubly synchronous the question of Pluto’s classification remains controversial.
rotation and revolution in the Pluto-Charon system. The two We should make it very clear, however, that regardless of
bodies permanently face one another. the label that the IAU has put on it, Pluto is in no way a less
present location, the average surface temperature on Eris is

about 43 kelvins (K), easily cold enough to freeze out any
atmospheric methane. As is now the case for Pluto, Eris will
probably develop a methane atmosphere when it comes
closest to the Sun, in the year 2257.
The most recent KBOs to be classified as dwarf planets
are Haumea and Makemake. Both are smaller than Pluto
and both orbit slightly farther from the Sun than does
Pluto (see Table 12.1). Haumea has two moons of its own—
Hi’iaka and Namaka— enabling astronomers to calculate
the system’s mass as they did for Pluto and Eris. Although
Haumea has sufficient mass to pull itself into a spherical
shape (one of the IAU requirements for classification as a
dwarf planet), it spins so rapidly on its axis that its shape
is distorted into a flattened oval shape, with an equatorial
radius that is approximately twice its polar radius. As we
FIGURE 12 . 3 The distant dwarf planet Eris (shown within the learned in Chapter 9, this difference between the equatorial
white circle) is slightly larger than Pluto and has similar physical and polar radii would give Haumea an oblateness of 0.5, by
characteristics. far the largest of any classical planet or dwarf planet. No
moons have been discovered orbiting around Makemake,
so we know less about this dwarf planet than about Pluto,
interesting body to the scientific community, as is evidenced Haumea, or Eris.
by NASA’s funding of the New Horizons mission. On New Year’s Day 1801, Sicilian astronomer Giuseppe
What, then, can we say about Eris, Haumea, and Piazzi found a bright object between the orbits of Mars and
Makemake? Observations by the Hubble Space Telescope Jupiter. Piazzi named the new
confi rmed earlier estimates that the diameter of Eris is object Ceres. When Piazzi dis- Dwarf planet Ceres
approximately 2,600 kilometers (km), making it about 13 covered Ceres, he thought he was previously the
percent larger than Pluto. In 2005, a year before the IAU might have found a hypotheti- largest asteroid.
decision, astronomers had discovered Dysnomia, Eris’s only cal “missing planet.” But as
known moon. As observations of Charon did for Pluto, obser-
vations of Dysnomia’s orbital motion around Eris yielded
an accurate mass for Eris, which turned out to be about 28 FIGURE 12 . 4 Ceres, once known as the largest asteroid, is
percent greater than Pluto’s mass. Eris thus has sufficient now called a dwarf planet. Note its spherical shape, one of the
mass to pull itself into a round shape, but not enough mass criteria that gives Ceres its dwarf planet status. The nature of
to clear its surroundings. Along with Pluto, Eris was des- the bright spot is unknown, but it reveals the rotation of Ceres as
ignated as a dwarf planet. seen in this set of four images taken by HST over an interval of 2
Its highly eccentric orbit carries Eris from 37.8 astro- hours and 20 minutes. Time increases from left to right and top
to bottom.
nomical units (AU)—its closest point to the Sun—out to
97.6 AU, with an orbital period
of 557 years. By chance, Eris Dwarf planet Eris is
was found near the most dis- currently the most
tant point in its orbit, making distant known Solar
it currently the most remote System object.
known object in the Solar Sys-
tem.1 Because of its great distance at this time, Eris now
appears about 100 times fainter than Pluto. When we com-
bine the observed brightness of Eris with its diameter, we
arrive at an albedo of 0.86. This is an astonishing result.
With the single exception of Saturn’s moon Enceladus, the
surface of Eris is more highly reflecting than that of any
other major Solar System body. Like Enceladus, Eris must
have a coating of pristine ice. The surface of Enceladus is
water ice, while Eris is covered with methane ice. At its
1
Eris’s remote status is temporary. Other Solar System bodies, such as
the Kuiper Belt object Sedna, have eccentric orbits that will eventually
carry them as far as a thousand astronomical units from the Sun.
12.3 Ghostly Apparitions and Rocks Falling from the Sky 331
more objects were discovered orbiting the region between Throughout human history, the sudden and unexpected
Mars and Jupiter, astronomers classified Ceres as a new kind appearance of a bright comet has elicited fear, wonder, and
of Solar System object, which they called asteroids. superstition. Early cultures viewed these spectacular visi-
With a diameter of about 975 km, Ceres (Figure 12.4) is tors as omens. Comets were often seen as dire warnings of
larger than most moons but smaller than any planet. It con- disease, destruction, and death, but sometimes as portents
tains about a third of the total mass in the asteroid belt, but of victory in battle or as heavenly messengers announcing
only about 4 percent of the mass of our Moon. Ceres rotates the impending birth of a great leader. The earliest records of
on its axis with a period of about 9 hours, typical of many comets date from as long ago as the 23rd century B.C. Medi-
asteroids. As a large planetesimal, Ceres seems to have sur- terranean and Far Eastern literatures are especially full of
vived largely intact, although there are indications that it references and popular superstitions about comets.
underwent differentiation at some point in its early history. Meteorites have also engendered fascination and awe for
Spectra show that the surface of Ceres contains hydrated as long as humankind has watched the heavens. Nearly all
minerals such as clays and carbonates, indicating the pres- ancient cultures venerated these rocks from the sky. Early
ence of significant amounts of water in its interior. Perhaps Egyptians preserved meteorites along with the remains of
as much as a quarter of its mass exists in the form of a water- their pharaohs, Japanese placed them in Shinto shrines,
ice mantle, which surrounds a rocky inner core. and ancient Greeks worshipped them. In 1492—the same
For more than two centuries, Ceres would be known year that Columbus voyaged to the New World—the towns-
as the largest asteroid—but no longer. Curiously, the same folk of Ensisheim (now in France) watched as a 172-kilo-
IAU resolution that demoted Pluto from planet to dwarf gram (kg) stone fell to Earth. A woodcut artist recorded the
planet has raised the status of Ceres from asteroid to dwarf famous event (Figure 12.5), and for a while the meteorite
planet— Ceres is round, but it is unable to clear its neigh- was enshrined in a local church. Despite numerous eye-
borhood. NASA’s Dawn mission, launched in 2007, will witness accounts of meteorite falls, however, many people
explore a large asteroid named Vesta in 2011–12 and then were slow to accept that these peculiar rocks actually came
go on to Ceres in 2015. from far beyond Earth. In 1807, two Yale professors inves-
tigating the meteorite that landed in Weston, Connecticut,
concluded that the object truly had dropped from the sky.
Thomas Jefferson, third president of the United States and
12.3 Ghostly Apparitions an enlightened scientist, is rumored to have responded,
“I would rather believe that two Yankee professors would
and Rocks Falling lie, than that stones fall from heaven.” But as the evidence
continued to mount, it became impossible to ignore. By the
from the Sky early 1800s, scientists had documented so many meteorite
falls that their true origin was indisputable. Today, hardly
Like an ocean filled with flotsam and jetsam, our Solar Sys- a year passes without a recorded meteorite fall, including
tem swarms with smaller members that might escape our some that have smashed into cars, houses, pets, and (occa-
attention until one of them washes ashore. It would be a sionally) people (see Chapter 7).
mistake to assume that this planetary debris is unimport- A third type of interplanetary debris was not discovered
ant, however. Comets, asteroids, and meteorites provide until the age of telescopes, when Piazzi found Ceres in 1801.
astronomers with some of their most important clues about Subsequently, a number of similar objects were discovered
how our Solar System formed and evolved. in the region between the orbits of Mars and Jupiter. Because
FIGURE 12 . 5 A woodcut of
the meteorite fall at Ensisheim
in 1492. More than three
centuries would pass before
scientists were convinced that
such objects come from beyond
Earth.
these new objects appeared in astronomers’ eyepieces as up into tiny clumps of dusty ice. When Earth passes through
nothing more than faint points of light, William and Caro- swarms of dusty ice from a
line Herschel (the brother-sister pair of English astronomers disintegrating comet nucleus Comet nuclei
who discovered Uranus) named them asteroids, a Greek or pebbles from the breakup are remnants of
word meaning “starlike.” As the years went by, more aster- of an asteroid, the particles, icy planetesimals
oids were discovered, and by the first decade of the 21st cen- called meteoroids, can burn that formed far
tury we have found nearly 450,000 of them, of which about up in our atmosphere, creat- from the Sun.
200,000 have well-determined orbits. Discoveries continue ing an atmospheric phenom-
at the rate of nearly 4,000 per month. enon we refer to as a meteor. Excursions 12.1 describes and
Apart from their role as fascinating curiosities and differentiates among meteoroids, meteors, meteorites, and
reminders that there is more to heaven and Earth than meets comets.
the eye, these types of inter- Planetesimals that became part of the planets or their
planetary debris play a very Comets, asteroids, moons were so severely modified by planetary processes that
significant role in our quest to and meteorites nearly all information about their original physical condi-
discover humankind’s cosmic provide scientific clues tion and chemical composition has been hopelessly lost. By
origins. We now know that to our past. contrast, asteroids and comet nuclei constitute an ancient
meteorites, asteroids, and and far more pristine record of what the early Solar Sys-
comets are fragments of our Solar System’s ancient past. tem was like. If you visit your local planetarium or science
Like archaeologists who assemble the history of civiliza- museum, you may find a meteorite on display that you can
tions from fossil bones and shards of pottery, planetary walk right up to and touch. Do so with respect. The mete-
scientists work to piece together the story of our Solar Sys- orite under your hand may be older than Earth itself. Some
tem from these relics of a time when the Sun, Earth, and meteorites contain tiny grains of material that pre-date the
planets were young. We begin by describing what the com- formation of our Solar System. These grains include dia-
pleted puzzle looks like and then talk about the pieces and monds and carbon compounds that originated as material
how they fit together. ejected from dying stars.
12.4 Asteroids and Comets: 12.5 Meteorites: A Chip Off

Pieces of the Past the Old Asteroid Block
As we learned in Chapter 6, when the planet-building pro- Most asteroids are so small compared with planets that they
cess fi nally ended, some of the original planetesimals appear as nothing but unresolved points of light through
remained in orbit around the telescopes on Earth. (They reveal themselves by their motion
Sun. Most asteroids are relics Asteroids are relics of among the fi xed stars.) Historically, their small size has
of rocky or metallic planetesi- rocky planetesimals made it difficult to learn much about them directly. Even
mals that originated in the that formed between determining gross properties such as size, shape, and rate
region between the orbits of the orbits of Mars of spin has required sophisti-
Mars and Jupiter. Although and Jupiter. cated analysis of telescopic, Meteorites are pieces
early collisions between these spacecraft, and radar observa- of asteroids. They tell
planetesimals created several bodies large enough to dif- tions. Despite this challenge, us about the bodies
ferentiate, Jupiter’s tidal disruption prevented them from we probably know more about they came from.
forming a single Moon-sized planet. As they orbit the Sun, the structure and composition
asteroids continue to collide with one another, producing of asteroids than about any other Solar System object besides
small fragments of rock and metal. Most meteorites are Earth! How is this possible? As asteroids orbit the Sun, they
pieces of these asteroidal fragments that have found their occasionally collide with each other, chipping off smaller
way to Earth and crashed to its surface. rocks and bits of dust. Sometimes one of these fragments is
Icy planetesimals surviving today form the solid cores, captured by Earth’s gravity and survives its fiery descent
or “nuclei,” of comets. Although asteroids can survive in through Earth’s atmosphere as a meteor, to be picked up
the inner Solar System, comet nuclei cannot. Today most and added to someone’s collection of meteorites.
comet nuclei are preserved in the frigid regions of space Thousands of meteorites reach the surface of Earth every
beyond the planets. Sometimes, however, something hap- day, but only a tiny fraction of these are ever found and iden-
pens to cause one of these icy bodies to change direction tified as such. Antarctica offers the best meteorite hunting in
and dive inward toward the domain of the inner planets. the world. Meteorites are no more likely to fall in Antarctica
When a comet nucleus approaches too close to the Sun, solar than anywhere else, but in Antarctica they are far easier to
heating unglues its surface structure, gradually breaking it distinguish from their surroundings. Antarctic snowfields
E X C U R S I O N S 1 2 .1
Meteors, Meteoroids, Meteorites, and Comets

Meteors are familiar to all of us. Just stand outside for a few A 1-gram meteoroid (about half the mass of a dime) entering
minutes on a starry night, away from bright city lights, and Earth’s atmosphere at 50 km/s has a kinetic energy comparable
you will almost certainly be rewarded by a glimpse of a “fall- to that of an automobile cruising along at the fastest highway
ing star,” a meteor. Most of us realize that we are not witness- speeds. Before plunging into our atmosphere, the meteoroid
ing the demise of a real star but are instead seeing a piece of may have been orbiting the Sun for millions of years after being
Solar System debris entering our atmosphere. Few are aware chipped from an asteroid or left behind by a disintegrating
that what we see is seldom the actual debris, but rather a trail comet nucleus. Most meteoroids are so small and fragile that
of heated atmospheric gas glowing from the friction caused by they burn up completely before reaching Earth’s surface.
a small unseen object entering at speeds of up to about 70 kilo- A meteoroid large enough to survive all the way to the
meters per second (km/s). A meteor, then, is an “atmospheric ground is called a meteorite. All meteoroids are pieces of Solar
phenomenon.” System debris. The larger pieces that survive the plunge through
The small2 solid body that creates the meteor is called a Earth’s atmosphere are usually fragments of asteroids. Meteor-
meteoroid. These small objects are of either cometary or aster- ites likely litter the ground of all solid planets and moons. Most
oidal origin. Cometary fragments are typically less than a cen- of the smaller pieces that burn up in the atmosphere before
timeter across and have about the same density as cigarette ash. reaching the ground are cometary fragments.
What they lack in size and mass, they make up for in speed. Comets appear as fuzzy patches of light in the nighttime
sky. They are bodies of ice, dust, and gas that shine by reflected
2
By defi nition, meteoroids are objects whose size lies within a sunlight. Comets frequently remain visible in our skies for
range of 100 micrometers (μm) to 100 meters. Particles smaller weeks at a time, with tails that can be 100 million km long
than this are considered “interplanetary dust”; larger objects are and stretch halfway across our sky. In contrast, a meteor may
called planetesimals. Some astronomers place a meteoroid’s upper streak across 100 km of our atmosphere and last at most a few
size limit at 10 meters. seconds.
serve as a huge net for collecting meteorites. Glaciers then silicate rocks. A stony meteorite can be recognized by the
carry the meteorites along, concentrating them in regions thin coating of melted rock that forms as it passes through
where the glacial ice is eroded by wind. The meteorites are the atmosphere. Many stony meteorites contain small round
left lying on the surface for collectors to pick up. The big spherules called chondrules (see Figure 12.6a), which range
advantage of hunting on Antarctic ice is that in many places in size from sand grains to marble-sized objects. Laboratory
the only stones to be found on the ice are meteorites. Antarc- experiments simulating formation show that chondrules
tic meteorites have also spent their time on Earth literally formed at higher temperatures than the fi ner-grained min-
in the deep freeze; many are thus relatively well preserved, erals that surround them.
showing little weathering or contamination from terrestrial Chondrules must once have Meteorites tell us
dust or organic compounds. been molten droplets that rap- the age of the
Meteorites are extremely valuable because they are sam- idly cooled to form the crys- Solar System.
ples of the same relatively pristine material from which tallized spheres we see today.
asteroids are made. In addition, unlike planets or asteroids Chondrules are all held together by a matrix of finer-grained
themselves, we can take meteorites into the laboratory and material, much as the gravel in concrete is held together by
study them using sophisticated equipment and techniques. a matrix of cement. Stony meteorites containing chondrules
Scientists compare them to rocks found on Earth and the are called chondrites (see Figure 12.6a). One type of chon-
Moon and contrast their structure and chemical makeup drite stands out with particular interest: carbonaceous
with those of rocks studied by spacecraft that have landed chondrites, chondrites that are rich in carbon, are thought
on Mars and Venus. Meteorites are also compared with to be the very building blocks of the Solar System. Although
asteroids and other objects on the basis of what colors of these meteorites cannot be dated directly, because they are
sunlight they reflect and absorb. composed of aggregates of many individual grains, indirect
Meteorites are normally grouped into three categories measurements suggest that they are about 4.56 billion years
based on the kinds of materials they are made of and the old— our best measurement of the time that has passed since
degree of differentiation they experienced within their par- the formation of the Solar System. By contrast, the stony
ent bodies. Over 90 percent of meteorites are stony meteor- meteorites that do not contain chondrules are known as
ites (Figures 12.6a and b), which are similar to terrestrial achondrites (see Figure 12.6b). Unlike chondrites, achon-
FIGURE 12 .6 Cross sections

(a) (b)
of several kinds of meteorites:
(a) a chondrite (a stony
Chondrules meteorite with chondrules),
(b) an achondrite (a stony
meteorite without chondrules),
(c) an iron meteorite, and (d) a
stony-iron meteorite.
(c) Stony (d)

material
Once-molten
iron and nickel
drites have crystals that appear to have formed in the same planetesimals accreted smaller objects. Additional thermal
place and at the same time. energy was released as radioactive elements inside the plan-
The second major category of meteorites, iron meteorites etesimals decayed. Some
(Figure 12.6c), is the easiest to recognize. The surface of an planetesimals never reached C-type asteroids never
iron meteorite has a melted and pitted appearance gener- the temperatures needed to differentiated.
ated by frictional heating as it streaked through the atmo- melt their interiors. These
sphere. (The meteorite in your local museum is probably an planetesimals— aggregations of the primordial material
iron meteorite.) Even so, many are never found, either from which they formed— simply cooled off, and they have
because they land in water or simply because no one hap- since remained pretty much as they were when they formed.
pens to notice and recognize them for what they are. As an These planetesimals are known as C-type asteroids, which
interesting side note, one of are composed of primitive material that we believe has
the more pleasant surprises There are three largely been unmodified since the origin of the Solar Sys-
from the Mars rover Opportu- classes of meteorites: tem, almost 4.6 billion years ago.
nity was the discovery of an stony (including In contrast to the C-type asteroids, some planetesimals
iron meteorite lying in plain chondrites and were heated enough by impacts and radioactive decay to
sight on the martian surface achondrites), iron, cause them to melt and differentiate, with denser matter
(Figure 12.7). Bot h its and stony-iron. such as iron sinking to the center of the planetesimal.
appearance— typical of iron Lower-density material—such
meteorites found on Earth— and its position on the smooth, as compounds of calcium, sil- S- and M-type
featureless plains made it instantly recognizable. Chemical icon, and oxygen— floated asteroids may be
analysis from the rover instruments showed it to be com- toward the surface and com- pieces of larger
posed mostly of iron and nickel. The final category is the bined to form a mantle and differentiated bodies.
stony-iron meteorites (Figure 12.6d), which consist of a crust of silicate rocks. S-type
mixture of rocky material and iron-nickel alloys. Stony-iron asteroids (and the stony achondritic meteorites that come
meteorites are relatively rare. from them) may be pieces of the mantles and crusts of such
Meteorites come from asteroids, which in turn derive differentiated planetesimals. They are chemically more
from rocky-iron planetesimals. Refer to Figure 12.8 as we similar to igneous rocks found on Earth than to primitive
follow the possible fates of a planetesimal. Recall from chondrites and C-type asteroids. That is, they were hot
Chapter 6 that, during the growth of the terrestrial planets, enough at some point to lose their carbon compounds and
large amounts of thermal energy were released as larger other volatile materials to space. The large interlocking
transition zone between the stony mantle and the metallic

core of such a planetesimal.
A few planetesimals in the region between the orbits
of Mars and Jupiter seem to have evolved toward becom-
ing tiny planets before being shattered by collisions. They
even became volcanically active, complete with eruption of
lava flows onto their surfaces. Vesta (now the largest known
asteroid, with a diameter of 525 km) seems to be a piece of
such a highly evolved planetesimal. Vesta does not fit well
into any of the main asteroid groups, but its spectrum is
strikingly similar to the reflection spectrum of a peculiar
FIGURE 12 .7 A basketball-sized iron meteorite lying on group of meteorites. These meteorites, which are prob-
the surface of Mars, imaged by the Mars exploration rover ably pieces of Vesta, look like rocks taken from iron-rich
Opportunity. lava flows on Earth and the Moon. The main difference is
that they are only 4.4 billion to 4.5 billion years old, a bit
younger than most meteorites. It is hard to escape the con-
crystals that make up achondrites are also like those seen clusion that early in the history of the Solar System, some
in rocks that slowly cooled from molten material. Similarly, planetesimals were large enough to become differentiated
M-type asteroids (from which iron meteorites come) are and to develop volcanism. But rather than going on to form
fragments of the iron- and nickel-rich cores of one or more planets, these planetesimals broke into pieces in collisions
differentiated planetesimals that shattered into small pieces with other planetesimals.
during collisions with other planetesimals. Indeed, slabs The story of how planetesimals, asteroids, and meteor-
cut from iron meteorites show large, interlocking crystals ites are related is one of the great successes of planetary
characteristic of iron that cooled very slowly from molten science. A vast wealth of information about this diverse
metal. The rare stony-iron meteorites may come from the collection of objects has come together into a single con-
FIGURE 12 .8 The fate of a rocky planetesimal in the young Solar System depends on whether
it gets large and hot enough to melt and differentiate, as well as on the impacts it experiences.
Different histories lead to the different types of asteroids and meteorites found today.
1 Thermal energy from 2 Some planetesimals

impacts and radioactive just cooled off in
decay heated a newly their original state…
formed rocky planetesimal.
C-type
asteroid
Carbonaceous
chondrite
4 Further impacts
shattered these 5 ...forming the different
planetesimals... classes of asteroids and
meteorites we see today.
Iron meteorite
Achondrite Stony-
Light materials Iron and nickel
iron meteorite
in mantle and crust
3 ...but others got hot

enough for their interiors
to melt and differentiate.
S-type M-type
asteroid asteroid
sistent story of planetesimals growing and possibly becom- asteroids be less dense than the rock that likely came from
ing differentiated, and then being shattered by subsequent them? Planetologists think that some of them are shattered
collisions. The story is even more satisfying because it, in heaps of rubble, with large voids between the fragments.
turn, fits so well with the even grander story of how most Once again, this is what we would expect of objects that
planetesimals were accreted into the planets and their were assembled from smaller objects and then suffered a
moons. history of violent collisions.
As we stress throughout this book, patterns are extremely Asteroids rotate just as planets and moons do, although
important in science—not only when they are followed, but the rotation of irregularly shaped asteroids is more of a
also when they are broken. tumble than a spin. A day on
Some types of meteorites fail To hold certain a typical asteroid is about 9 A day on a typical
to follow the patterns just dis- meteorites is to hold hours long, although the rota- asteroid is only
cussed. Whereas most achon- a piece of Mars. tion periods of some asteroids 9 hours long.
drites have ages in the range are as short as 2 hours, and
of 4.5 billion to 4.6 billion years, some members of one group others are longer than 40 Earth days. How can we measure
are less than 1.3 billion years old. Some are chemically and the rotation period of an object that appears starlike in a
physically similar to the soil and the atmospheric gases that telescope? Unless the asteroid is perfectly round, and we
our lander instruments have measured on Mars. The simi- know that very few are, we measure rotation periods by
larities are so strong that most planetary scientists believe watching changes in their brightness as they alternately
these meteorites are pieces of Mars that were knocked into present their broad and narrow faces to Earth.
space by asteroidal impacts.32 This means we have pieces of Asteroids are found throughout the Solar System. Most
another planet we can study in laboratories here on Earth. orbit the Sun in several distinct zones, with the majority
The martian meteorites support the general belief that much residing between the orbits of Mars and Jupiter in the main
of the surface of Mars is covered with iron-rich volcanic asteroid belt. The main belt contains at least 1,000 objects
materials. larger than 30 km in diameter (about the size of Washington
If pieces of Mars have reached Earth from its orbit almost DC), of which about 200 are larger than 100 km. We estimate
80 million km beyond our own, we might expect that pieces that there may be as many as 10 million asteroids larger than
of our companion Moon would fi nd their way to Earth as 1 km in the main asteroid belt. However, although there are
well. Indeed, meteorites of another group bear striking simi- a great number of asteroids, they account for only a tiny
larities to samples returned from the Moon. Like the mete- fraction of the matter in the Solar System. If all of the aster-
orites from Mars, these are chunks of the Moon that were oids were combined into a single body, it would be about a
blasted into space by impacts and later fell to Earth. third the size of Earth’s Moon. Recall from Chapter 10 that
there are several empty regions within the main asteroid
belt, known as Kirkwood gaps, that have been depleted by
perturbations from Jupiter.
12.6 Asteroids Are Figure 12.9 shows orbits of several asteroids that are
representative of their classes. In Chapter 10 we discussed
Fractured Rock the Trojan asteroids that occupy the L 4 and L 5 Lagrangian
points of the Sun-Jupiter system. Asteroids whose orbits
With a few exceptions, asteroids are too small for their self- bring them close to Earth’s orbit are called near-Earth aster-
gravity to have pulled them into a spherical shape. Some oids. Members of one group, called the Amors, cross the
asteroids imaged by spacecraft or by Earth-based radar have orbit of Mars but do not cross Earth’s orbit. Members of two
highly elongated shapes, suggesting objects that either are other groups, the Atens and the Apollos, have orbits that
fragments of larger bodies or were created haphazardly cross Earth’s orbit (see Figure 12.9). The difference between
from collisions between smaller bodies. Astronomers have these two populations is that the Apollos also cross the orbit
measured the masses of a number of asteroids by noting of Mars, while the Atens
the effect of their gravity on spacecraft passing nearby. remain inside Mars’s orbit. Apollo and Aten
Knowing the mass and the size of an asteroid enables Atens and Apollos, along with asteroids cross
us to determine its density. The densities of these aster- a few degenerate comet nuclei, Earth’s orbit
oids range between 1.3 and 3.5 times the density of water. are known collectively as and could collide
Those at the lower end of this range are considerably less near-Earth objects (NEOs) with us one day.
dense than the meteorite fragments they create. How can and occasionally collide with
Earth or the Moon. Astronomers estimate that there are
more than 3,500 Atens and Apollos with diameters larger
23
In 1996 a NASA research team announced that a particular martian than a kilometer. As we discovered in Chapter 7, collisions
meteorite (ALH84001) showed physical and chemical evidence of early between Earth or the Moon and such objects are geologi-
life on Mars. These conclusions have since been challenged by many cally important and some have dramatically altered the
scientists and are now generally discredited. history of life on Earth.
12.7 Asteroids Viewed Up Close 337
FIGURE 12 .9 Blue dots

(a) Face-on view
show the locations of known
Main-belt asteroids orbit Trojans share Jupiter’s asteroids at a single point
Jupiter
between Mars and Jupiter. orbital period. in time. Most families of
Achilles
asteroids take their names
Trojans Jupiter’s orbit
from prototype members
Sylvia of the family. For example,
Apollo-family asteroids cross
the orbits of Earth and Mars,
Trojans like the asteroid Apollo. Most
asteroids, such as Sylvia, are
main-belt asteroids. Achilles
Apollo was the first Trojan asteroid to
Apollos cross orbits
Aten of Earth and Mars. be discovered.
Mercury
Venus
Atens’ orbits cross

Amor Earth’s orbit but stay
inside Mars’s orbit.
Mars Earth
Ceres
Amors cross Mars’s
orbit but not Earth’s.
Main
asteroid belt
Asteroid orbits
Asteroids follow prograde

orbits that lie near the
(b) Edge-on view plane of the Solar Sytem.
Plane of
Solar System
12.7 Asteroids Viewed nomenon revealed that people’s ideas about it were wrong!)
Fortunately, as spacecraft began to return detailed informa-
Up Close tion about the asteroids they visited, the data they collected
placed our ideas about the history of the Solar System on
Until the space age, scientists had no good idea of what even more solid footing.
asteroids looked like. As the 1980s drew to a close, plan- In 1991 the Galileo spacecraft was following a circuitous
etary scientists were about to have their theories put to the path through the inner Solar System, gaining momentum
test. Would the fi rst close-up views of asteroids pull the rug from encounters with Earth and Venus to help send it on
out from under our carefully constructed story of asteroids, its journey to Jupiter. While flying through the main aster-
meteorites, and the early Solar System? (If so, it would not oid belt, it got close enough to the asteroid Gaspra to obtain
be the fi rst time a closer examination of a particular phe- dozens of good images. Gaspra is an S-type asteroid, about
9 × 10 × 20 km. The Galileo images show it to be cratered

and irregular in shape. Faint, groovelike patterns on its sur-
face may be fractures resulting from the same impact that Ida
chipped Gaspra from a larger planetesimal. Images taken
in several different wavelengths of light show distinctive
colors on Gaspra’s surface, implying that Gaspra is covered
with a variety of different types of rock.
Later in its mission, Galileo made another pass through
the main belt, and this time it sent back images of the aster-
oid Ida (Figure 12.10). Ida orbits in the outer part of the main
Dactyl
asteroid belt, and it, too, is an S-type asteroid. Galileo flew
so close to Ida that it could see details as small as 10 meters
across— about the size of a small house or cottage. Ida is
shaped like a croissant, 54 km long and ranging from 15 to FIGURE 12 .10 A Galileo spacecraft image of the asteroid Ida
24 km in diameter. Like Gaspra, Ida shows the scars of a long with its tiny moon, Dactyl (also shown enlarged in the inset).
history of impacts with smaller bodies. Just as planetary
scientists use the number and sizes of craters to estimate
the ages of the surfaces of planets and moons (as we saw in
Chapter 7), the cratering on Ida indicates that its surface is early 2000 it was placed into orbit around the asteroid Eros
a billion years old, twice the age estimated for Gaspra. Also to begin long-term observations of this S-type object (Fig-
like Gaspra, Ida contains fractures. The fractures seen in ure 12.11). As a near-Earth asteroid, Eros is one of more
the two asteroids indicate that they must be made of rela- than 5,000 known objects whose orbits bring them within
tively solid rock. (You can’t “crack” a loose pile of rubble.) 1.3 AU of the Sun. Eros measures about 34 × 11 × 11 km
This is an important confi rmation of the idea— crucial to and, like its counterparts Gaspra and Ida, shows a surface
our story of asteroids and meteorites—that some asteroids with grooves, rubble, and impact craters, including a cra-
must be pieces chipped from larger, solid objects. ter 8.5 km across. However, the scarcity of smaller craters
The Galileo images also revealed a tiny moon orbiting suggests that its surface could be younger than Ida’s. A
the asteroid. Ida’s moon, called Dactyl, is only 1.4 km across, year after being placed in orbit, the spacecraft was gently
and like nearly all Solar Sys- crash-landed onto the asteroid’s surface. Chemical analyses
tem objects with solid sur- Ida’s feeble gravity confi rmed the similarity in composition between Eros and
faces, it is cratered from holds a tiny moon primitive meteorites.
impacts. Although planetary called Dactyl. On its way to Eros, NEAR Shoemaker flew past the aster-
scientists had long speculated oid Mathilde and provided our fi rst information about a
that asteroids might themselves have moons, Dactyl was C-type body. Mathilde was found to measure 66 × 48 × 46
the fi rst such object to be discovered. But is Dactyl really a km and to have a surface only half as reflective as charcoal.
moon? To planetary scientists, any significant Solar System Its color properties suggest a composition of materials such
body orbiting a larger body other than the Sun is called a as carbonaceous chondrules. However, estimates of the
moon. Dactyl may not be the stuff of poetry, but by the plan- overall density of Mathilde are about 1,300 kilograms per
etary scientist’s defi nition of a moon, it counts. As it turns cubic meter, or kg/m3 (1.3 times the density of water), only
out, Dactyl is hardly unique. Moons have now been found half that of carbonaceous chondrite meteorites measured in
around nearly 100 asteroids, and at least three asteroids are the laboratory. This density estimate implies that Mathilde
known to have two moons. is porous inside, as we would expect if it is composed of
Since Galileo’s visits to Gaspra and Ida, spacecraft have chunks of rocky material stuck together by the accretion
visited four other asteroids: Mathilde, Eros, Braille, and process with open spaces between the chunks. The surface
Itokawa. Braille was seen only briefly, in 1999 during the of Mathilde is dominated by craters, the largest of which is
flyby of Deep Space 1. It was found to have a highly reflec- more than 33 km across. The great number of craters sug-
tive surface, similar to Vesta’s, leading to speculation that gests that Mathilde likely dates back to the very early his-
Braille could be a chunk knocked from Vesta by a collision tory of the Solar System.
sometime in the past. The spectrum of sunlight reflected by Mars’s tiny moons,
The NEAR Shoemaker34 spacecraft was designed to do Phobos and Deimos (Figure 12.12), is similar to that of C-type
more than make a brief encounter with an asteroid—in asteroids; and many scientists
believe these moons must be Phobos and Deimos
34
NEAR, the original name, stands for Near Earth Asteroid Rendezvous. asteroid-like objects that were may be asteroids
The spacecraft was renamed NEAR Shoemaker in March 2000 in honor somehow captured by Mars. captured by Mars.
of Gene Shoemaker, a pioneer of the study of comets and asteroids who But are Phobos and Deimos
died in an automobile accident in 1997 while studying impact craters really asteroids? Controversy about this question has never
in the Australian outback. been put to rest. Some scientists argue that it is unlikely
12.8 The Comets:

Clumps of Ice
Close observations and more refi ned analyses have rede-
fined our view of asteroids. Likewise, our concept of comets
changed markedly during the 16th and 17th centuries, when
we looked upon nature with an increasingly refined and
rational eye. In much earlier times comets were regarded
simply as ghostly apparitions, pale luminous patches or
streaks of light in the nighttime sky that would mysteriously
appear, remain for a few days or weeks, and then vanish.
Until the end of the Middle Ages, the Aristotelian view of
comets prevailed, regarding them as atmospheric phenom-
ena rather than astronomical objects. The popular beliefs
that comets were very nearby and that their tails contained
poisonous vapors contributed to the fear and panic that
FIGURE 12 .11 Asteroid Eros viewed by the NEAR Shoemaker comets frequently generated.
spacecraft. This image was produced by high-resolution scanning A major change in our view came in the 16th century,
of the asteroid’s surface by the spacecraft’s laser range finder.
thanks to Tycho Brahe. Tycho reasoned that if comets were
NEAR Shoemaker became the first spacecraft to land on an
atmospheric phenomena like
asteroid, when in 2001 the orbiter was gently crashed onto the
surface of Eros.
clouds, as Aristotle had sup- Comets are
posed, then their appearance astronomical objects,
and location in the sky should not atmospheric
be very different to observers phenomena.
Mars could have “captured” two asteroids and that Phobos located many miles away
and Deimos must somehow have evolved together with from each other. But when Tycho compared sightings of
Mars. A third possibility is that the bodies were parts of a comets made by observers at several different sites, he found
much larger body that was fragmented by a collision early no evidence of such differences, and he concluded that
in the history of Mars. comets must be at least as far away as the Moon. Lest we
Once called the “vermin of the skies,” asteroids have now become arrogant in our faith in humankind’s rationality,
achieved considerable respectability in scientific circles. it is worth noting that the proof that comets are far outside
Not only do they hold fascinating and unique clues about Earth’s atmosphere did not completely dispel the fear they
how the Solar System formed, but they also play a part in evoke. As recently as 1910, the news that Earth would pass
the continuing (albeit slow) evolution of the Solar System. through the tail of Comet Halley (popularly known as Hal-
And if an asteroid were to get too close to Earth, as happens ley’s Comet) was accompanied by widespread apprehension
from time to time, it would quickly become a very respect- throughout Europe and the United States. And during the
able topic for conversation! approach of Comets Kohoutek in 1973 and Hale-Bopp in
(a) (b)
FIGURE 12 .12 Mars
Reconnaissance orbiter and
Mars Express images of Mars’s
two tiny moons: (a) Deimos and
(b) Phobos.
1997,45 many otherwise rational people experienced similar Kuiper Belt objects Oort Cloud comets
anxieties, regrettably fueled by certain opportunistic writ- are shown in red. are shown in green.
ers who exploit human fear and superstition for power or
profit. 500 AU
Eris’s orbit Plane of Solar System
Sun
The Abode of the Comets Neptune’s orbit (diameter 60 AU)
When we think of a comet, we think of a spectacular dis- Zoom out by 20

play of the sort produced by Comet Hyakutake in 1996, as 10,000 AU
seen in the opening photo-
graph of this chapter. But for Swarms of comets The Kuiper Belt is
most of their lives, comets are spend most of an extension of the
merely icy planetesimals— their lives far beyond Solar System disk.
comet nuclei formed from pri- the planets.
Zoom out by 20
mordial material that managed
to escape the planet-building process— adrift in the frigid 200,000 AU
outer reaches of the Solar System. Drifting in these distant
regions, most are much too small and far away to be seen
and counted by telescopes on Earth, so no one really knows
their total number. Estimates range as high as a trillion (1012)
comet nuclei—more than all the stars in the Milky Way
Galaxy. Nevertheless, we have seen hardly more than a
thousand, perhaps only a billionth of the presumed total.
Comet nuclei put on a show only when they come deep
enough into the inner Solar System to suffer destructive
heating from the Sun. When they are close enough to show
the effects of solar heating, we call them active comets.
We know where comets reside by observing their orbits The Oort Cloud is a
as they pass through the inner Solar System. These studies larger spherical cloud
suggest that comets fall into of comet nuclei.
two distinct groups: the The Kuiper Belt and
“Kuiper Belt comets” and the the Oort Cloud are
“Oort Cloud comets.” These reservoirs of comets. Distance to nearest star Proxima Centauri
two populations of comet approximately 270,000 AU
nuclei are named for scientists Gerard Kuiper (1905–1973)
and Jan Oort (1900–1992), who fi rst proposed their existence FIGURE 12 .13 Most comet nuclei near the inner Solar System
in the mid–20th century.56 form an extension to the disk of the Solar System called the
The extent of the realm of comets is illustrated in Fig- Kuiper Belt. The spherical Oort Cloud is far larger and contains
ure 12.13. Comet nuclei from the Kuiper Belt orbit the Sun many more comets.
within a disk-shaped region that begins just beyond the
orbit of Neptune and extends outward to perhaps a thou-
sand astronomical units from the Sun. The innermost part
of the Kuiper Belt, which seems to end somewhat abruptly larger than Ceres and similar in size to Pluto.67 Some plan-
at about 50 AU, appears to contain tens of thousands of etologists would classify Pluto not as a dwarf planet but as
icy planetesimals, which we call Kuiper Belt objects. The a large member of the family of Kuiper Belt objects. Yet one
larger KBOs are well over 1,000 km across, making them characteristic sets Pluto aside from all the known KBOs and
makes it seem more planetlike: Pluto has an atmosphere.
Like the asteroids, some KBOs are attended by moons, and
45
at least one has as many as three moons. Astronomers have
Believing that Comet Hale-Bopp was accompanied by a UFO that
would carry their souls away to a “better place,” 39 members of the
67
cult Heaven’s Gate committed mass suicide in March 1997. With a few exceptions, the sizes of KBOs cannot be measured directly
56
Some planetary scientists refer to the Kuiper Belt as the Edgeworth- and must be estimated. Although brightness and approximate dis-
Kuiper Belt in honor of Kenneth Edgeworth (1880–1972), an Irish tance are known, their albedos are uncertain, making estimates of
engineer who predicted a reservoir of comets beyond the planets in their size correspondingly uncertain. Nevertheless, reasonable lim-
a paper published in 1943, several years before the predictions of its for the albedos of KBOs can set maximum and minimum values
Kuiper and Oort. for their size.
already found more than 1,000 KBOs, but many smaller of comet nuclei. The gravitational attraction from huge
ones must also be there, just beyond the reach of our larg- clouds of dense interstellar gas concentrated in the plane
est telescopes. Although they are too far from the Sun to of the Milky Way Galaxy78 might also stir up the Oort Cloud
be “active,” it seems likely that these planetesimals are as the Sun passes back and forth through the galactic plane
comet nuclei. during its 220-million-year orbital journey around the cen-
One of the larger known KBOs— Quaoar (pronounced ter of our galaxy.
“kwa-whar”)—is also one of the few whose size we have These disturbances from beyond the Solar System have
been able to measure. Hubble Space Telescope observations little effect on the orbits of the planets and other bodies
show a tiny disk with a diameter of about 1,250 km, slightly in the inner Solar System. Inner Solar System objects are
larger than Pluto’s Charon. By knowing its apparent bright- close enough to the Sun that external disturbances never
ness, distance, and size, we can easily calculate Quaoar’s exert more than a tiny fraction of the gravitational force
albedo. It turns out to be 0.12, which is more reflective than of the Sun. In the distant Oort Cloud, however, things are
the nuclei of those comets that have entered the inner Solar quite different. Oort Cloud comet nuclei are so far from the
System but far less reflective than Pluto. Quaoar’s remote Sun, and the Sun’s gravitational force on them is so feeble,
location and pristine condition have apparently allowed that they are barely bound to the Sun at all. The tug of a
some bright ices to survive on its surface. Quaoar orbits slowly passing star or interstellar cloud can compete with
between 42 and 45 AU from the Sun, and in this frigid region the Sun’s gravity, significantly changing the orbits of Oort
of the Solar System its surface temperature is only 50 K. Cloud objects. If the interaction adds to the orbital energy
Sedna is an interesting object that remains outside the of a comet nucleus, it may move outward to an even more
inner Kuiper Belt. Its highly elliptical orbit takes it from 75 distant orbit, or perhaps escape from the Sun completely to
AU out to 1,000 AU from the Sun. In such an extended orbit, begin an eons-long odyssey through interstellar space. On
Sedna requires more than 10,000 years to make a single the other hand, a comet nucleus that loses orbital energy
trip around the Sun. When discovered in 2003, Sedna was falls inward. Some of these comet nuclei come all the way
about 90 AU from the Sun and getting closer. It will reach into the inner Solar System, where they may appear briefly
its closest point in 2076. Because of its great distance, Sed- in our skies as active comets before returning once again
na’s size cannot be measured directly, even by the Hubble to the Oort Cloud.
Space Telescope. If its albedo were the same as Quaoar’s, Unlike comet nuclei in the Oort Cloud, those in the
Sedna would be the largest KBO. Kuiper Belt are packed together close enough to inter-
Right now our knowledge of the chemical and physical act gravitationally from time to time. In such events, one
properties of KBOs remains meager at best. Because of their nucleus gains energy while the other loses it. The “winner”
great distance, we know very little about their composition may gain enough energy to be sent into an orbit that reaches
and surface properties and absolutely nothing about their far beyond the boundary of the Kuiper Belt. It seems likely
geology. This lack of information may change in the not- that the Oort Cloud was populated in just this way (by comet
too-distant future, however. Following its encounter with nuclei ejected from the inner edge of the Kuiper Belt in what
Pluto in 2015, the New Horizons spacecraft is scheduled is now the Uranus-Neptune region). The experience could
to continue outward into the Kuiper Belt, where it will be prove fatal for the “loser” if enough orbital energy is lost for
maneuvered to fly close to one or more KBOs. it to fall inward toward the Sun.
Unlike the flat disk of the Kuiper Belt, the Oort Cloud is
a spherical distribution of comet nuclei that are much too
remote to be seen by even the most powerful telescopes. We
know the size and shape of the Oort Cloud because comet
Anatomy of an Active Comet
nuclei from the Oort Cloud approach the inner Solar Sys- Unlike asteroids, which have been through a host of chemi-
tem from seemingly random directions in the sky, following cal and physical changes as a result of collisions, heating,
orbits that bring them in from as far as 100,000 AU from the and differentiation, most comet nuclei have been preserved
Sun, or nearly halfway to the nearest stars! over the past 4.6 billion years by the “deep freeze” of the
We can think of the Kuiper Belt and Oort Cloud as enor- outer Solar System. Comet nuclei are made of the most
mous reservoirs of icy planetesimals that now and then fall nearly pristine material remaining from the formation of
into the inner Solar System. But why do they sometimes the Solar System.
leave their frigid haven and take what will probably be a The small object at the center of the comet—the icy plan-
fatal plunge inward toward the Sun? It seems likely that etesimal itself—is the comet nucleus. The nucleus is by far
gravitational perturbations by objects beyond the Solar the smallest component of a comet, but it is the source of all
System— such as stars and interstellar clouds— disturb the mass that we see stretched across the skies as the comet
the orbits of comet nuclei, kicking the nuclei in toward nears the Sun (Figure 12.14). Comet nuclei are anywhere from
the Sun. Astronomers estimate that every 5 million to 10 a few dozen meters to several hundred kilometers across. We
million years, one star or another, traveling in nearly the
same orbit around the Milky Way as does the Sun, passes 78
Interstellar clouds and the structure of the Milky Way Galaxy will
within about 100,000 AU of the Sun, perturbing the orbits be discussed in Chapters 15 and 20, respectively.
might formally describe comet nuclei as “planetesimals of comet form a nearly spherical atmospheric cloud around
modest size composed of a mixture of various ices of volatile the nucleus called the coma
compounds, organics, and (plu r a l comae — Fig u r e The coma is an
dust grains loosely packed An active comet has 12.14a). The nucleus and the extended cometary
together to form a porous con- a tiny nucleus inner part of the coma are atmosphere
glomerate”; but, mercifully, a at its heart. sometimes referred to collec- surrounding the
shorter description is possible. tively as the comet’s head. nucleus of a comet.
In the middle of the 20th century, astronomer Fred Whipple Pointing from the head of the
(1906–2004) offered an elegant way to sum all of this up in comet in a direction more or less away from the Sun are
just two words: comets are “dirty snowballs.” long streamers of dust, gas, and ions called the tail.
As a comet nucleus nears the Sun, sunlight heats its Most naked-eye comets seem to exhibit similar behavior—
surface, turning volatile ices into gases, which then stream developing fi rst a coma and then an extended tail as they
away from the nucleus, carrying embedded dust particles approach the inner Solar System. Comet McNaught in 2007
along with them. This process of conversion from solid to was such a comet (Figure 12.15a). But there are exceptions.
gas is called sublimation. An example much closer at hand Comet Holmes was a very faint telescopic object when it
illustrates this process. Dry ice (frozen carbon dioxide) reached its closest point to the Sun just beyond the orbit of
does not melt like water ice but instead turns directly into Mars on May 4, 2007. Then, several months later, on October
carbon dioxide gas. Water ice melts.89 Dry ice sublimates— 21, as it was well on its way outward toward Jupiter’s orbit,
that is why we call it “dry.” Set a piece of dry ice out in the the comet suddenly became a half-million times brighter
sun on a summer day, and you will get a pretty good idea in just 42 hours! This abrupt outburst made Comet Hol-
of what happens to a comet. mes a bright naked-eye comet that graced Northern Hemi-
The gases and dust driven from the nucleus of an active sphere skies for several months (Figure 12.15b). Astronomers
remain puzzled over the cause of this dramatic eruption.
89
Water ice also sublimates in certain environments. For example, an Explanations range from a meteoroid impact to a sudden
ice cube left in a frost-free freezer will eventually disappear as water (but unexplained) buildup of subsurface gas. Astro-
molecules slowly break free from its surface. Tour: Cometary Orbits
(a) (b)
Ion tail
Dust tail
Head
Sun
Comet nuclei The much larger coma and

are “dirty tail are material driven off
snowballs.” the nucleus by solar heating.
Nucleus
Coma
FIGURE 12 .14 (a) The principal components of a fully

developed active comet are the nucleus, the coma, and two types
of tails, called the dust tail and the ion tail. Together, the nucleus
and the coma are called the head. (b) Comet Hale-Bopp, the great
comet of 1997.
(a)
despite the fact that even relatively pristine comet nuclei are
frequent visitors to the inner Solar System, we seldom have
enough advance warning of a comet’s visit or sufficiently
detailed knowledge of its orbit to mount a successful mis-
sion to intercept it. But although planetary scientists knew
the orbit of Halley extremely well, sending probes to the
comet still turned out to be a difficult task as space mis-
sions go. The problem was that the orbit of Comet Halley
is retrograde—it goes around the Sun in a direction oppo-
site to that of Earth and the other planets. As a result, the
closing speed between an Earth-launched spacecraft and
the comet would be 68 km/s, or about 75 times the speed
of a bullet from a high-powered rifle. Observations close to
the nucleus would have to be made very quickly, and there
would be the always-present danger of high-speed collisions
with debris from the nucleus.
Despite the difficulties, working in cooperation with
each other, three different space agencies rose to the chal-
lenge of sending a five-spacecraft armada to meet Comet Hal-
(b) ley as it passed through the inner Solar System. The whole
operation was rather like driving down the highway at 70
miles per hour (mph) and trying to thread a needle held out
the window by someone in the oncoming lane, but in the
end the mission was a glorious success. Much of what we
know about comet nuclei and the innermost parts of the
coma was learned from data sent back by this armada of
Soviet, European, and Japanese spacecraft.
Images taken by these spacecraft showed the nucleus
of Halley to be an irregular peanut shell–shaped object
measuring 8 × 8 × 14 km (Figure 12.16a). As the Soviet
VEGA and the European Giotto spacecraft passed close by
the nucleus, they observed jets of gas and dust moving at
speeds of up to 1,000 meters per second (m/s) from its sur-
FIGURE 12 .15 (a) Comet McNaught, in 2007, was the face, far exceeding the ability of its feeble gravity to contain
brightest comet to appear in decades, but its true splendor was the material. Here was dramatic evidence that the gas and
visible only to observers in the Southern Hemisphere. (b) Months dust that make up a comet’s coma and tail come directly
after its closest approach to the Sun in 2007, Comet Holmes from the nucleus. The spacecraft images showed, however,
suddenly became a half-million times brighter within hours,
that instead of streaming out from the whole surface of the
turning into a naked-eye comet with an angular diameter larger
nucleus, material was coming from several hot spots, or
than that of the full Moon. Comet Holmes favored Northern
Hemisphere observers.
jets, that covered only about a tenth of the surface. Ninety
percent of the surface was “quiet” at the time of the obser-
vations. The escaping gas emerged from beneath the crust
through a number of small cracks or fissures.
A Visit to Comet Halley By observing the jets of material streaming away from
the nucleus of Halley, planetary scientists estimated that
Typical comet nuclei at typical distances appear so small that 20,000 kg of gas and 10,000 kg
even the most powerful telescopes see them only as points of of dust were being lost by the Comet Halley loses
light. Even when a rare comet happens to approach close to nucleus each second. During 100 billion kg of mass
Earth, we are still out of luck because the nucleus remains its 1986 passage around the each time it swings
hidden within the dusty coma that surrounds it. Although Sun, Comet Halley must have around the Sun.
astronomers have studied the gases that make up the comae lost at least 100 billion (1011)
and tails of comets with great interest for centuries, direct kg of material. But these jets also indirectly enabled plan-
observations of comet nuclei remained an elusive prize. etary scientists to measure another fundamental property
All this changed in 1986, when Comet Halley made its of this comet. Just like the engines on a jet airplane, the jets
“once in a lifetime” trip into the inner Solar System, proof material streaming away from the nucleus push on it,
viding a rare opportunity to send space probes to observe a subtly altering its orbit. By observing the jets flowing away
nucleus at close range. This opportunity was so rare because, from the nucleus, scientists could tell how hard the jets
FIGURE 12 .16 (a) The

(a) (b)
nucleus of Comet Halley as
Sunlight
imaged in false color by the
Jets of gas Giotto spacecraft in 1986. (b)
and dust
An HST image of the coma of
Comet Hyakutake taken in red
light. A jet of material is seen
at the upper left pointing in the
antisunward direction.
Nucleus
were pushing on the nucleus. And by observing the comet’s sphere surrounding the nucleus, consisting of gas and dust
orbit, they learned how the nucleus was responding to these that are driven from the nucleus by solar heating. Figure
shoves. Using these observations, scientists were then able 12.16b shows an image of the coma of Comet Hyakutake
to apply Newton’s laws of motion, which require the change obtained with the Hubble Space Telescope. As a comet
in the momentum of the nucleus to balance the change in nucleus falls from the outer Solar System inward toward the
momentum of the material streaming away. They thus esti- Sun, the coma begins to form at about the time the nucleus
mated the mass of the nucleus to be about 2.2 × 1014 kg. passes within the orbit of Jupiter. (Although from our van-
This mass, divided by the volume as measured from the tage point so close to the heart of the Solar System, Jupiter
images of the nucleus, tells us that the nucleus is only about may seem far removed from the Sun’s warmth, from the
a fourth as dense as water. This is more like a loosely packed comet’s perspective the solar warming at 5 AU is hundreds
powdery snow than normal ice. As suggested by Whipple to many millions of times greater than it is in the Kuiper Belt
years earlier, the nucleus of Comet Halley must be a very or the Oort Cloud.) As the nucleus approaches still closer
fragile object that is porous throughout. Even with a mass to the Sun, rapid sublimation of ices near its surface can
loss of 300 billion kg of material during each encounter, the cause the coma to swell to a diameter of a million kilome-
comet is losing only about 0.1 percent of its total mass, so ters, three times the distance between Earth and the Moon.
Comet Halley should still be entertaining terrestrial onlook- But although the coma is huge compared with the nucleus,
ers hundreds of generations from now. It will return to the it is still more tenuous than the very best vacuums we can
inner Solar System again in the summer of 2061. produce on Earth.
The surface of the nucleus is also much darker than VEGA and Giotto entered the coma of Comet Halley
expected. It reflects only 3 percent of the sunlight falling on when they were still nearly 300,000 km from the nucleus.
it, making it even blacker than The spacecraft measured the size and chemical composi-
coal or black velvet. The “dirty The nucleus of Comet tion of the dust particles in Halley’s coma and found that
snowball” at the heart of Halley is blacker the dust grains escaping from the nucleus were generally
Comet Halley is not only dirty; than coal. between 0.01 and 10 millimeters (mm) in size. This dust
it is among the darkest known is very fi ne—fi ner even than the dust responsible for that
objects in the Solar System. This low reflectivity means that annoying thin film that collects on your car and ruins its
comet nuclei, or at least the nucleus of Halley, must be rich shine. Experiments found that the dust from Comet Hal-
in organic matter that is far more complex than simple hydro- ley consists of a mixture of light organic substances (com-
carbons such as methane, propane, butane, or octane. Simple binations of hydrogen, carbon, nitrogen, and oxygen) and
organic compounds such as these form clear liquids and ices heavier rocky material (combinations of magnesium, silicon,
at low temperatures. In contrast, more complex hydrocar- iron, and oxygen). Instruments on the spacecraft designed
bons such as tar are usually very dark. The implication is to study the gases surrounding the nucleus found that it
that very complex organic matter was present as dust in the consisted of about 80 percent water (H2O) and 10 percent
disk around the young Sun, perhaps even in the interstellar carbon monoxide (CO), with smaller amounts of carbon
cloud from which the Solar System formed. dioxide (CO2), methane (CH4), ammonia (NH3), and hydro-
The coma of a comet is a tenuous and temporary atmo- gen cyanide (HCN).
Visits to Comets Borrelly,

Wild 2, and Tempel 1
The early years of the 21st century have witnessed dramatic
progress in our understanding of comet nuclei. NASA sent
three spacecraft to visit three Jupiter-family comets: Bor-
relly, Wild 2, and Tempel 1. Such comets are escapees from
the Kuiper Belt that have been captured by Jupiter into short-
period orbits. It may come as no surprise to learn that all
three comet nuclei appear vastly different when studied up
close. In 2001, NASA’s Deep Space 1 spacecraft flew within
2,200 km of Comet Borrelly’s nucleus. Comet Borrelly’s tar-
black surface is among the darkest seen on any Solar System
object and, much to the surprise of scientists, showed no
evidence of water ice. In 2004, NASA’s Stardust spacecraft FIGURE 12 .17 The nucleus of Comet Wild 2, imaged by the
flew within 235 km of the nucleus of Comet Wild 2 (pro- Stardust spacecraft.
nounced “vilt too”). Astronomers were especially interested
in this comet because it is a newcomer to the inner Solar
System. Comet Wild 2 had previously resided in the frigid
region between the orbits of Jupiter and Uranus, but a close
encounter with Jupiter in 1974 perturbed its orbit, bringing
this relatively pristine body closer to the Sun as it travels
back and forth in its new orbit between Jupiter and Earth.
At the time of Stardust’s encounter with Wild 2, the comet
had made only five trips around the Sun in its new orbit,
compared with more than 100 visits by Comet Halley.
Wild 2’s nucleus is about 5 km across, and unlike other
comets visited earlier by spacecraft (Comets Halley and
Borrelly), it has a more spherical shape. Images taken by
Stardust show at least 10 active gas jets. Some were as large
as 100 meters in diameter and carried within them surpris-
ingly large chunks of surface material. (A few particles as
FIGURE 12 .18 The surface of Comet Tempel 1 seen just
large as a bullet penetrated the outer layer of the space-
before impact by the Deep Impact projectile. The impact occurred
craft’s protective shield.) The surface of Wild 2 appears
between the two 370-meter-diameter craters located near the
covered with craterlike features (Figure 12.17). Although
bottom of the image. The smallest features appearing in this
these depressions do not have the normal impact charac- image are about 5 meters across.
teristics, they may in fact be impact craters that have been
modified by ice sublimation, small landslides, and erosion
by jetting gas. The larger depressions are more than 1.5 km
across. Some show flat floors, suggesting a relatively solid
interior beneath a porous surface layer.
How do we know what really lies beneath a comet’s sur-
face? To find the answer, astronomers came up with a novel
approach. They proposed firing a massive bullet at very high
speed into the nucleus of a comet and observing what hap-
pens. In 2005 a 370-kg impacting projectile, launched by
NASA’s Deep Impact spacecraft, hit the nucleus of Comet
Tempel 1 (Figure 12.18) at a speed of more than 10 km/s. The
kinetic energy of the impact was equivalent to 5 tons of TNT,
sending 10,000 tons of water and dust flying off into space
at speeds of 50 m/s (Figure 12.19). A camera mounted on the
projectile snapped photos of its target until both were vapor- FIGURE 12 .19 A 370-kg projectile impacts Comet Tempel 1
ized by the impact. Observations of the event were made at a speed of 10.2 km/s. This image, taken 67 seconds after
locally by the Deep Impact spacecraft and back at Earth by impact, shows subsurface cometary material being ejected into
a multitude of orbiting and groundbased telescopes. space by the force of the impact.
Spectra indicated the presence of water, carbon diox-

ide, hydrogen cyanide, iron-bearing minerals, and a host of Ion tails point directly Dust tail
away from the Sun.
complex organic molecules. Analysis of the debris showed
that the comet’s outer layer is composed of fi ne dust with a Ion tail
consistency of talcum powder. No large chunks were seen
flying from the impact site. Beneath the dust are layers
Comet
made up of water ice and organic materials. The nucleus
has a density only 0.6 times that of water and is perhaps
best described as a porous rubble pile weakly held together
by self-gravity rather than the mechanical strength of its Sun
constituent materials. One surprise for scientists was the
presence of well-formed impact craters (see Figure 12.18),
which had been absent in close-up images of Comets Bor-
relly and Wild 2. Why should some comet nuclei have fresh
impact craters and others none? This remains a question
that planetary scientists have yet to answer.
Dust tails curve along

Comets Have Two Types of Tails the comet’s orbit.
The tail, which is the largest and most spectacular part of a
comet, is also the “hair” for which comets are named. (Comet FIGURE 12 . 20 The orientation of the dust and ion tails at
comes from the Greek word kometes, which means “hairy several points in a comet’s orbit. The ion tail points directly away
one.”) Active comets have two different types of tails, as from the Sun while the dust tail curves along the comet’s orbit.
shown in Figure 12.14a. One type of tail is called the ion tail.
Many of the atoms and molecules that make up a comet’s
coma are ions. Because they are electrically charged, ions from one comet to another. Some comets, such as Comets
in the coma feel the effect of the solar wind, the stream of West and Halley, display both types of tails simultaneously;
charged particles that blows continually away from the Sun. while, for reasons that we do not understand, some comets
The solar wind pushes on these ions, rapidly accelerating produce no tails at all! A tail often forms as the comet crosses
them to speeds of more than 100 km/s—far greater than the the orbit of Mars, where the increase in solar heating drives
orbital velocity of the comet itself— and sweeps them out gas and dust away from the nucleus at a prodigious rate. In
into a long wispy structure. Because the particles that make October 1965, early-morning risers were held spellbound by
up the ion tail are so quickly picked up by the solar wind, the sight of Comet Ikeya-Seki’s glowing 150 million km-long
ion tails are usually very straight and point from the head tail (Figure 12.21) stretching across the sky— a length as
of the comet directly away from the Sun. great as the distance between Earth and the Sun. For a brief
Dust particles in the coma can also have a net electric few weeks, Comet Ikeya-Seki achieved temporary splendor
charge and feel the force of the solar wind. In addition, sun- as the longest object in the entire Solar System.
light itself exerts a force on cometary dust. But dust particles The gas in a comet’s tail is even more tenuous than the
are much more massive than individual ions, so they are gas in its coma, with densities of no more than a few hun-
accelerated more gently and do not reach such high rela- dred molecules per cubic centimeter. Compare this density
tive speeds as the ions do. As a result, the dust particles are with that of Earth’s atmosphere, which at sea level contains
unable to keep up with the comet, and the dust tail often more than 1019 molecules per cubic centimeter. Dust parti-
appears to curve gently away from the head of the comet cles in the tail are typically about a micrometer in diameter,
as the dust particles are gradually pushed from the comet’s about the size of smoke particles. One astronomer’s often-
orbit in the direction away from the Sun. repeated quip that “all of the material in a long comet tail
Figure 12.20 shows the tails of a comet at various points could be packed into a suitcase” is less an exaggeration than
in its orbit. Remember that both types of tails always point you might imagine. When Earth passed through the tail of
away from the Sun, regardless Comet Halley in 1910, there were no observable effects. In
of the direction in which the Comet tails always contrast to the widespread predictions of disaster, it was
comet is moving. As it point away from all “much ado about nothing.”
approaches the Sun, the com- the Sun.
et’s two tails trail behind its
nucleus. But like the flag on a sailboat sailing with the wind,
the comet’s tails extend ahead of the nucleus as it moves
The Orbits of Comets
outward from the Sun. How long a dying nucleus will survive depends on its orbital
Comet tails are fascinating structures that vary greatly period and perihelion (closest point to the Sun) distance—in
the Kuiper Belt. The surprise is that more than 100 short-
period comets remain relatively close to the Sun. How do
these denizens of the frigid hinterlands become resident
aliens in the realm of the planets?
The key to solving this mystery lies in an interesting
difference between the orbits of long- and short-period com-
ets. Figure 12.22 shows the orbits of a number of comets.
Whereas long-period comets come diving into the inner
Solar System from all directions, short-period comets have
orbits that are strongly concentrated in the ecliptic plane.
Because their orbits cross the orbits of many of the planets,
short-period comets frequently get close enough to a planet
for the planet’s gravity to change the comet’s orbit about the
Sun in the sorts of chaotic encounters discussed in Chap-
ter 10. Presumably, short-period comets originated in the
Kuiper Belt; but as they fell in toward the Sun, they were
forced into their present short-period orbits by gravitational
encounters with planets.
Short-Period Comets Figure 12.22a shows the orbits of

a number of short-period comets in the inner Solar System.
Nearly all short-period comets have prograde orbits. That
is, they orbit the Sun in the same direction as the planets.
(Comet Halley, with its retrograde orbit, is an exception.)
This tendency toward prograde orbits, along with the con-
centration of short-period comets in the plane of the Solar
System, is important evidence supporting our theory of the
origin of short-period comets and how they were captured.
The disklike Kuiper Belt appears to share the prograde
rotation of the rest of the Solar System (except the Oort
Cloud). Because Kuiper Belt comet nuclei start out on pro-
FIGURE 12 . 2 1 Comet Ikeya-Seki as it appeared in the predawn
grade orbits, they are more likely to be captured into short-
hours in October 1965. Soon afterward, just before it passed period prograde orbits as well. In addition, the encounter
behind the Sun, the head of the comet became so bright that between a planet and a comet nucleus on a prograde orbit
it was visible in broad daylight. When it reappeared a few days will last longer than the encounter between a planet and a
later, it had split into two separate components. comet nucleus on a retrograde orbit. This prolonged contact
means that a prograde comet’s orbit will be more strongly
influenced by such an encounter than will the orbit of a ret-
other words, how frequently it passes by the Sun and how rograde comet. Jupiter, the most massive of the planets, is
close it comes. Each passage takes its toll of ice and dust. By the one most likely to affect the orbit of a Kuiper Belt comet.
convention, comet orbits are generally referred to as short- There are about 270 known short-period comets, referred to
period or long-period orbits. The division between the two is as the Jupiter family, that all have their aphelion (farthest
somewhat arbitrarily set at a period of 200 Earth years. Com- point from the Sun) near Jupiter’s orbit and travel in pro-
ets with periods of less than 200 years are called short-period grade orbits around the Sun.
comets. The total number of short-period comets known Comet Halley is the brightest and most famous of the
today is about 400. Comets with orbital periods longer than short-period comets. It also has a very special place in the
200 years are termed long-period comets. The total number history of science because it was the first comet whose return
of long-period comets observed to date is about 3,000, with was predicted. In 1705, Edmund Halley was studying the
an average of six new ones being discovered each year. orbits of comets and employing the gravitational laws that
The orbits of nearly all comets are highly elliptical, with had recently been discovered by his friend and colleague
one end of the orbit close to Isaac Newton. (Halley was instrumental in encouraging
the Sun and the other in the The orbits of most Newton to publish his work on orbits and gravitation—work
distant parts of the Solar Sys- comets are very that became the foundation for the science of physics.) Hal-
tem. As a result, we might elliptical. ley noted that a bright comet that had appeared in 1682 had
expect all comets passing an orbit remarkably similar to those of comets seen in 1531
through the inner Solar System to have extremely long and 1607. He concluded that all three were actually one
orbital periods that carry them back to the Oort Cloud or and the same, and he made the daring prediction that this
(a) (b)
Inner Solar System Outer Solar System

Halley Saturn
1 Short-period comets in the 3 Long-period comets orbit far into

inner Solar System mostly the outer Solar System on both
have prograde orbits… prograde and retrograde orbits…
Jupiter
Halley
Encke
Saturn
Pluto
Hale-
Bopp
Face-on view Face-on view
2 …that lie close to the 4 …and dive through the Solar

plane of the Solar System. System from all directions.
Plane of
Solar System
Plane of orbits
Pluto’s orbit Solar System
Halley
Halley Hale-Bopp
Edge-on view Edge-on view
FIGURE 12 . 22 Orbits of a number of comets in face-on and edge-on views of the Solar System.
Populations of short-period comets (a) and long-period comets (b) have very different orbital
properties. Comet Halley, which appears in both diagrams for comparison, is a short-period comet.
comet would return in 1758. Unfortunately, Halley did not and Earth were on opposite sides of the Sun when the comet
live to see it. His comet reappeared on Christmas Eve in the put on its display— seeing the ghostly sight of Halley is an
very year he had predicted. Astronomers quickly named it experience we will long remember. Comet Halley will reach
Halley’s Comet and heralded it as a triumph for the genius aphelion in 2024 and then begin its long journey back to the
of both Newton and Halley. inner Solar System, becoming visible to the naked eye once
Comet Halley’s highly elongated orbit takes it from peri- again in the summer of 2061.
helion, about halfway between the orbits of Mercury and
Venus, out to aphelion beyond Long-Period Comets Of the known long-period comets,
the orbit of Neptune. Astrono- Halley’s Comet comes more than 600 have well-determined orbits. Some have
mers and historians have now close to Earth orbital periods of hundreds of
identified possible sightings every 76 years. thousands or even millions of Some long-period
of the comet that go back to years, and their nuclei spend comets come by
467 B.C. Comet Halley has an average period of 76 years, and almost all their time in the only once in a
many of us mark the “once in our lifetime” when we are frigid, outermost regions of million years.
fortunate enough to see it. Mark Twain is famous for saying the Solar System. Orbits of a
that he came in with the comet in 1834 and he would go out few long-period comets are shown in Figure 12.22b. Unlike
with the comet in 1910—a promise he kept. For the authors the mostly prograde, ecliptic-plane orbits of the short-period
of this book, our opportunity to see Halley’s Comet came comets, long-period comets split about evenly into prograde
in 1986. Although that appearance was not especially spec- and retrograde and fall into the inner Solar System from all
tacular compared to the one in 1910—in 1986 Comet Halley directions. These are the comets that tell us of the existence
of the Oort Cloud. Because of their very long orbital periods, and approaching the orbit of Neptune. Comet Hale-Bopp
these comets cannot make more than a single appearance will continue its outward journey for well over 1,000 years.
throughout the course of recorded history and, in most It will not return to the inner Solar System until sometime
cases, in all of human history. around the year 4530.
Long-period comets differ from short-period comets in The most spectacular comet yet seen in the 21st cen-
another way as well. With a few exceptions, Comet Halley tury was Comet McNaught. Known as the “Great Comet of
among them, the nuclei of short-period comets have been 2007,” it was the brightest to appear in over 40 years and
badly “worn out” by their repeated exposure to heating was perhaps the second brightest in nearly a century. The
by the Sun. As the volatile ices are driven from a nucleus, comet’s nucleus and coma were visible in broad daylight
some of the dust and organics are left behind on the surface. as its orbit carried it within 25 million km of the Sun.
The buildup of this covering slows down cometary activity. When Comet McNaught next appeared in the evening skies
(Envision how, as a pile of urban— and therefore dirty— to observers in Australia and elsewhere in the Southern
snow melts, the dirt left behind is concentrated on the Hemisphere, its tail had grown to a length of more than
surface of the snow.) That is why most short-period comets 160 million km and stretched 35° across the sky (see Fig-
create little excitement. In contrast, long-period comets are ure 12.15a). Comet McNaught came into the inner Solar
usually relatively pristine. More of their supply of volatile System from the Oort Cloud, but it left on a slightly hyper-
ices still remains close to the surface of the nucleus, and bolic orbit (see Chapter 3), meaning that this is one comet
they can produce a truly magnificent show. A half dozen or that will never return.
so long-period comets arrive each year. Most pass through When will the next bright comet like Hyakutake, Hale-
the inner Solar System at relatively large distances from Bopp, or McNaught come along? On average, a spectacular
Earth or the Sun and never become bright enough to attract comet appears about once per decade, but it is all a matter
much public attention. of chance. It might be many years from now— or it could
Comet Ikeya-Seki (see Figure 12.21) is a member of a happen tomorrow.
family of comets called sungrazers, comets whose perihelia
are located very close to the surface of the Sun. Many sun-
grazers fail to survive even a single pass by our local star.
Ikeya-Seki became so bright as it neared perihelion in 1965 12.9 Collisions Still
that it was visible in broad daylight, only two solar diam-
eters away from the noontime Sun; and when it reappeared Happen Today
from behind the Sun, it had been split into two pieces. Sun-
grazers generally come in groups, with successive comets As we studied the inner planets and the moons of the outer
following in nearly identical orbits. Each member of such planets, we learned that almost all hard-surfaced objects in
a group started as part of a single larger nucleus that broke the Solar System still bear the scars of a time when the Sun
into pieces during an earlier perihelion passage. and planets were young and tremendous impact events were
The closing years of the 20th century witnessed two common occurrences. The collision of Comet Shoemaker-
spectacular long-period comets sporting long, beautiful Levy 9 with Jupiter in 1994 (see Excursions 12.2) focused
tails: Hyakutake in 1996 (see the opening photograph of this attention on the fact that although such impacts are far less
chapter) and Hale-Bopp in 1997 (see Figure 12.14b). Both frequent today than they once
were widely seen by the viewing public, Comet Hale-Bopp were, they still happen.910 Hav- In 1994, much of the
being perhaps the most observed comet ever. It is a huge ing observed the effects of world watched as
comet, with a nucleus perhaps as large as 60 km in diameter. Shoemaker-Levy 9’s collision a comet crashed
It was bright enough to spot even in urban nighttime skies, with Jupiter, we can easily into Jupiter.
and it dazzled those who were fortunate enough to see it far imagine the global firestorms
from city lights. Comet Hale-Bopp is an especially impor- on Earth that accompanied the impact that occurred at the
tant scientific object for professional astronomers because Cretaceous-Tertiary boundary 65 million years ago (see
it was discovered far from the Sun near Jupiter’s orbit 2 Excursions 7.1). Connections 12.1 discusses several ways
years before its perihelion passage. This early discovery that such impacts affected the course of life on Earth. The
extended the total time available to study its development “Jupiter comet crash” was also a landmark event in the his-
as it approached the Sun and provided ample opportunity tory of the public’s access to fast-breaking scientific events.
to plan the important observations as it neared perihelion. Occurring when the Internet was growing in popularity, the
Warmed by the Sun, the nucleus produced large quantities impact’s latest images were downloaded daily by millions
of gas and dust, and as much as 300 tons of water per second, of people around the world, scientists and laypeople alike.
with lesser amounts of carbon monoxide, sulfur dioxide,
cyanogen, and other gases. Comet Hale-Bopp is still being 9
10
In 2009, observers saw evidence of yet another comet hitting Jupi-
observed with large telescopes. It remains active and con- ter, the sudden appearance of a dark atmospheric scar nearly as large
tinues to show a tail, even though it is now far from the Sun as Earth.
What we learned from the crash of Shoemaker Levy-9 region was blasted with the energy equivalent of 2,000
into Jupiter also helps explain an event that has puzzled times the atomic bomb dropped over Hiroshima. Figure
scientists for the past cen- 12.25 (on p. 353) shows a map of the region, along with a
tury. The Tunguska River In 1908 a small photo of the devastation caused by the blast. Eyewitness
f lows t h rough a remote asteroid or comet accounts detailed the destruction of dwellings, the incin-
region of western Siberia, smashed into Siberia. eration of reindeer (including one herd of 700), and the
inhabited mostly by reindeer deaths of at least five people. Although trees were burned
and a few reindeer herders. In the summer of 1908 the or fl attened for over 2,150 square kilometers (km 2)—an
E XCURSIONS 12 . 2
Comet Shoemaker-Levy 9
From a wealth of ground- and space-based observations, plan- subsequently spread out along more than 7 million km of its
etary scientists have pieced together the events leading to the orbit. The trajectory carried the fragments around for one more
collision of Comet Shoemaker-Levy 9 with Jupiter. Early in the 2-year orbit about the planet.
20th century this comet nucleus from the Kuiper Belt was per- Then, in July 1994, the entire string of fragments crashed
turbed from its path and sent on an orbital journey that carried into Jupiter. Over a week’s time, one after another of the frag-
it close to Jupiter. In 1992 it passed so close to the planet that ments, each traveling at 60 km/s, plunged through Jupiter’s
tidal stresses broke it into two dozen major fragments, which stratosphere. Even though the impacts occurred just behind
FIGURE 12 . 23 Events following the impact of a fragment of Comet Shoemaker-Levy 9 on Jupiter.
Impact Impact Impact Impact

+ 1 minute + 5 minutes + 20 minutes
1 A fragment of the 2 The comet slams through 3 A fireball of hot, 4 …and sails free 5 …that crashes back
comet approaches Jupiter’s atmosphere, high-pressure gas as a plume of into the atmosphere,
Jupiter at 60 km/s. heating a column of gas to erupts back along the material… leaving a visible scar.
thousands of kelvins. comet’s path…
1,000 1,000
800 800
600 600
Altitude (km)
10,000
400 400
200 200
5,000
0 0 Point of
impact
–200 –200
0 0
–400 –400
100 0 100 200 100 0 100 200 300 0 5,000 10,000 0 5,000 10,000
Distance from impact (km) Distance from impact (km) Distance from impact (km) Distance from impact (km)
area greater than metropolitan New York City— no crater Tunguska event was the result of a tremendous high-altitude
was left behind! explosion that occurred when a small asteroid or comet
For many years, scientists and nonscientists alike specu- nucleus hit Earth’s atmosphere, ripped apart, and formed a
lated about the cause of the Tunguska event. Ideas included fi reball before reaching Earth’s surface. This is just the sort
such fanciful notions as a collision with a mini–black hole of event seen on Jupiter (when Shoemaker-Levy 9 crashed
that passed through Earth, a collision with an object made into it) and hypothesized to account for the “splotches” seen
of antimatter, or an act of aggression by aliens (a favorite of on Venus. Recent expeditions to the Tunguska area have
the supermarket tabloids). It now seems fairly clear that the recovered resin from the trees blasted by the event. Chemi-
the limb of the planet, where they could not be observed from sphere, causing ripple effects like pebbles thrown into a pond.
Earth, the Galileo spacecraft was on its way to Jupiter and was Figure 12.23 shows the sequence of events that took place as
able to image some of the impacts. In addition, astronomers each fragment of the comet slammed into the giant planet.
using groundbased telescopes and the Hubble Space Telescope Sulfur and carbon compounds released by the impacts formed
could see immense plumes rising from the impacts to heights giant, Earth-sized scars in the atmosphere that persisted for
of more than 3,000 km above the cloud tops at the limb. The months (Figure 12.24) and were visible even through small
debris in these plumes then rained back onto Jupiter’s strato- amateur telescopes.
FIGURE 12 . 2 4 (a) Groundbased infrared images of Jupiter, showing the hot, glowing scars left by
fragments of Comet Shoemaker-Levy 9. (b) Although the fragments struck Jupiter on the back side,
these HST images show the fireball from one fragment rising above the limb of the planet. (c) HST
images of the evolution of the scar left by one fragment of the comet.
(a) (c)
Fireball from
Thermal infrared fragment G impact
radiation from
impact scars
1.5 hours
after impact
(b) Fireball from

fragment G impact
4 minutes 3 days after
after impact impact
9 minutes
after impact
12 minutes 5 days after

after impact impact
15 minutes
after impact
22 minutes
after impact
cal traces in the resin suggest that the impacting object may to see until they come very close to Earth. Recently, the US
have been a stony asteroid. government— along with the governments of several other
The Tunguska impact and the collision of Shoemaker- nations—became aware of the risk posed by NEOs. Although
Levy 9 with Jupiter are truly sobering events. We know of the probability of a collision between a small asteroid and
several impacts occurring in the Solar System within a Earth is quite small, the consequence can be catastrophic.
human lifetime, and more than one of them involved our In response to this recognition, NASA has been given a
planet!1011 We know enough about the distribution of relatively congressional mandate to catalog all NEOs and to scan the
large asteroids in the inner Solar System to say that it is skies for those that remain undiscovered.
highly improbable a populated area on Earth will experience Comets, however, present a more serious problem. Sev-
a major collision with an asteroid within our lifetimes (see eral previously unknown long-period comets enter the
Chapter 7). Comets and smaller asteroids, however, are less inner Solar System each year.
predictable. There may be as many as 10 million asteroids If one happens to be on a col- Comet or asteroid
larger than a kilometer across, but only about 130,000 have lision course with Earth, we impacts are infrequent
well-known orbits, and most of the unknowns are too small might not notice it until just but devastating.
a few weeks or months before
1
11
0
impact. For example, Comet Hyakutake was discovered
At 10:38 A.M. on February 12, 1947, yet another planetesimal struck
only 2 months before passing near Earth, and a potentially
Earth, this time in the Sikhote-Alin region of eastern Siberia. Com-
posed mostly of iron, it had an estimated diameter of about 100 meters destructive asteroid that just missed Earth in 2002 was
and broke into a number of fragments before hitting the ground, not discovered until 3 days after its closest approach!
leaving a cluster of craters and widespread devastation. Witnesses Although this has become a favorite theme of science fic-
reported a fi reball brighter than the Sun and sound that was heard tion disaster stories (Lucifer’s Hammer, by Larry Niven
300 km away. and Jerry Pournelle, being a favorite of some of the authors
C O N N E C T I O N S 1 2 .1
Comets, Asteroids, and Life

One of the great questions of modern astronomy involves under- asteroids with Earth have resulted in widespread devasta-
standing how life on Earth is connected with processes at work tion of Earth’s ecosystem and the extinction of many species.
in the greater cosmos. In a certain sense, this quest is not so Passing stars or the passage of the Sun through the “galactic
different from that pursued by many of the world’s great reli- plane” may send showers of comet nuclei into the inner Solar
gions and philosophies. At most of the stops along our journey System every 10 million to 100 million years— intervals simi-
through the universe, we fi nd threads of the cosmic tapestry of lar to those between mass extinctions found in fossil records
which terrestrial life is a part, and this stop is no exception. of life on our planet. Although such events certainly qualify
For example, were it not for impacts by comet nuclei, water as global disasters for the plants and animals alive at the time,
on Earth might have been less plentiful than it is today. We they also represent global opportunities for new life-forms to
think that some of Earth’s water was contributed during impacts evolve and fill the niches left by species that did not survive.
of icy planetesimals (in other words, comet nuclei) early in the Such a collision with an asteroid or comet probably played a
history of the Solar System. How could this have come about? central role in ending the 180-million-year reign of dinosaurs
The icy planetesimals likely condensed from the protoplane- and provided our ancient mammalian ancestors an opportu-
tary disk surrounding the young Sun and grew to their present nity for world dominance (see Excursions 7.1).
size near the orbits of the giant planets. These planetesimals In our study of comets, we may also have found a key to
subsequently suffered strong orbital disturbances from those the chemical origins of life on Earth. Comets turn out to be
same planets. In such interactions, about half of the planetesi- rich in complex organic material—the chemical basis for ter-
mals would be flung outward to form the Kuiper Belt and Oort restrial life— and cometary impacts on the young Earth may
Cloud and half inward toward the Sun. Some of the objects have played a role in chemically seeding the planet. In addi-
flung toward the Sun would likely have hit Earth, the largest tion, the fact that comets are believed to be pristine samples
planet in the inner Solar System. Because most of the mass in of the material from which the Sun and planets formed means
comet nuclei appears to be in the form of water ice, it is likely that organic material must be widely distributed throughout
that some of Earth’s current water supply came from this early interstellar space. This conclusion, which is supported by radio
bombardment. Water, as we know, is essential to life. telescope observations of vast interstellar clouds throughout
Yet the existence of comets can also threaten life on Earth. our galaxy, might have significant implications as we consider
It seems certain that occasional collisions of comet nuclei and the possibility of life elsewhere in the universe.
Bering Strait St
o ny Baykit Strelka Tunguska region
Tu
Explosion
ng
us
a epicenter r
k
River ve
Vanavara Ri
Barents Tetere na
Sea Le
Tunguska
impact site N Area over which
SIBERIA object was seen
al
Sikhote-Alin
ai k
(RUSSIA)
impact site
eB
Lake
ak
Baikal L
0 100 200 300 400 km
KA Irkutsk
Z AK
ea
HS
Caspian S
TAN
MONGOLIA
FIGURE 12 . 25 A large region of forest

near the Tunguska River in Siberia was
flattened by an atmospheric explosion in 1908
caused by the impact of a small asteroid or
comet.
of this text), Earth’s geological and historical record sug- tem. As Earth and other planets move along in their orbits,
gests that actual impacts by large bodies are infrequent they continually sweep up this fi ne debris. This debris is
events. Does this mean we need not lose any sleep over a the source of most of the meteoroids that Earth encounters.
possible collision with a large comet or asteroid? Probably, When they burn up in our atmosphere, these meteoroids
but remember that even though the probability may be become the meteors that you can see streaking across the
small, the consequences of such an event are enormous. sky on any clear, dark night. As discussed in Excursions
Just ask the dinosaurs. 12.1, only when they are large enough to reach Earth’s sur-
face do we call them meteorites.
12.10 Solar System Debris Meteoroids and Meteors

Some 100,000 kg of meteoritic debris is swept up by Earth
Although some comets and asteroids meet the spectacular every day, and what does not burn up (mostly particles
fate of Comet Shoemaker-Levy 9 or the Tunguska planetesi- smaller than 100 μm) eventu-
mal, most go out with more of ally settles to the ground as Meteors are
a fi zzle than a bang. Comet Comet nuclei and fi ne dust. We can measure atmospheric
nuclei that enter the inner asteroids are the meteor heights with radar phenomena.
Solar System generally disin- sources of Solar because radio waves bounce
tegrate within a few hundred System debris. off the ionized gas in meteor trails just as they bounce off
thousand years as a result of the metal in an airplane or an automobile. Using these tech-
their repeated passages close to the Sun. Asteroids have niques, we fi nd that the altitudes of meteors are between 50
much longer lives but still are slowly broken into pieces and 150 km. For comparison, commercial jet aircraft typi-
from occasional collisions with each other. Disintegration cally fly at heights of about 10 km.
of comet nuclei and asteroid collisions are the sources of Fragments of asteroids are much denser than cometary
most of the debris that fi lls the inner part of the Solar Sys- meteoroids. If an asteroid fragment is large enough—about
the size of your fist—it can survive all the way to the ground the result is a meteor shower. Because the meteoroids that
to become a meteorite. The fall of a 10-kg meteoroid can pro- are being swept up are all in similar orbits, they all enter
duce a fireball so bright that it lights up the night sky more our atmosphere moving in the
brilliantly than the full Moon. Such a large meteoroid, travel- same direction. As a result, Meteor showers
ing many times faster than the speed of sound, may create a the paths of all shower mete- occur when Earth
sonic boom heard hundreds of kilometers away. It may even ors are parallel to one another. passes through
explode into multiple fragments as it nears the end of its flight, But just as the parallel rails of cometary debris.
becoming a bolide. Some fireballs glow with a brilliant green a railroad track appear to van-
color, caused by metals in the meteoroid that created them. ish to a single point in the distance, as in Figure 12.26b,
from our perspective all the meteors appear to originate
from the same point in the sky. This point is called the
shower’s radiant.
Meteor Showers and Comets The Perseids are the result of Earth’s crossing the orbit of
Standing under a dark nighttime sky with a clear horizon, Comet Swift-Tuttle. Although spread out along the comet’s
you can expect to see about a dozen meteors per hour on any orbit, the debris is more concentrated in the vicinity of the
night of the year. These are called sporadic meteors, and comet itself. In 1992, Comet Swift-Tuttle returned to the
they occur as Earth sweeps up random bits of cometary and inner Solar System for the fi rst time since its discovery in
asteroidal debris in its annual path around the Sun. How- 1862. An exceptional Perseid meteor shower resulted, with
ever, if you happen to be meteor watching on the nights of counts of up to 500 meteors per hour.
August 11–13, for instance, you may see four to five times More than a dozen comets have orbits that come close
this number. If you pay close attention, you might also enough to Earth’s to produce annual meteor showers. Around
notice that nearly all of the meteors seem to be coming from November 16 each year, Earth passes almost directly through
the same region of the sky. This phenomenon is called a the orbit of Comet Tempel-Tuttle, a short-period comet with
meteor shower (Table 12.2). We call the particular meteor an orbital period of 33.2 years. We call the meteors respon-
shower that peaks in mid-August the Perseids because the sible for the meteor shower that Tempel-Tuttle produces the
trails left behind all point back to the constellation Perseus Leonids. In most years the Leonids fail to produce much of
(Figure 12.26a). a show because Comet Tempel-Tuttle distributes little of its
Meteor showers happen when Earth’s orbit crosses the debris around its orbit. In 1833 and 1866, however, Comet
orbit of a comet. Bits of dust and other debris released by a Tempel-Tuttle was not far away when Earth passed through
comet nucleus as it rounds the Sun remain in orbits of their its orbit. The Leonid showers in those 2 years were so intense
own that are similar to the orbit of the nucleus itself. When that meteors filled the sky with as many as 100,000 per hour
Earth passes through this concentration of cometary debris, (Figure 12.27). In 1900, one comet orbit later, nothing out of
TA B L E 1 2 . 2
Selected Meteor Showers
Shower Approximate Date Parent Object
Quadrantids January 3–4 ?
Lyrids April 21–22 Comet Thatcher
Eta Aquariids May 5–6 Comet Halley
Perseids August 12–13 Comet Swift-Tuttle
Draconids October 8–9 Comet Giacobini-Zinner
Orionids October 21–22 Comet Halley
Taurids November 5–6 Comet Encke
Leonids November 17–18 Comet Tempel-Tuttle
Geminids December 13–14 Asteroid Phaethon
Ursids December 22–23 Comet Tuttle

(a) CEPHEUS (b)

Meteors in a shower appear
to move away from a point
called the radiant.
CAMELOPARDALIS
LYNX
CASSIOPEIA
Radiant of
Perseids Actually the meteor paths
are parallel. The radiant
is a vanishing point.
ANDROMEDA
AURIGA PERSEUS
PISCES
TRIANGULUM
VISUAL ANALOGY FIGURE 12 . 26 Meteors appear to stream away from the radiant of the Perseid
meteor shower (a). These streaks are actually parallel paths that appear to emerge from a vanishing
point, like the railroad tracks shown in (b)
the ordinary happened. Again in 1933, the Leonids were disap-

pointing. Perturbations of Comet Tempel-Tuttle by Jupiter had
moved the orbit of the comet’s nucleus slightly away from Earth’s,
causing a sharp decrease in shower strength. What Jupiter took
away, though, it gave back. Further perturbations of the comet’s
orbit caused a spectacular Leonid shower in 1966—which one
of the authors of this text witnessed—that may have produced
as many as a half-million meteors per hour! The Leonid shower
put on less spectacular but still impressive shows between 1999
and 2003, when several thousand meteors per hour were seen
(Figure 12.28).
Zodiacal Dust
In the same way that we can “see” sunlight streaming through an
open window by observing its reflection from motes of dust drift-
ing in the air, we can see the sunlight reflected off tiny zodiacal
dust particles that fill the inner parts of the Solar System close
to the plane of the ecliptic. On a clear, moonless night, not long
after the western sky has grown dark, this dust is visible as a faint
column of light slanting upward from the western horizon along
the path of the ecliptic. This band, called the zodiacal light, can
also be seen in the eastern sky before dawn (Figure 12.29). With
good eyes and an especially dark night, you may be able to follow
the zodiacal dust band all the way across the sky. In its brightest
parts, the zodiacal light can be several times brighter than the
Milky Way, for which it is sometimes mistaken.
Like meteoroids, zodiacal dust is a mixture of cometary debris
and ground-up asteroidal material. The dust grains are roughly
FIGURE 12 . 2 7 An engraving of the 1833 Leonid shower a millionth of a meter in diameter—the size of smoke particles.
seen in France. In the vicinity of Earth there are only a few particles of zodiacal
FIGURE 12 . 28 Leonid meteors seen in 2001. This image is FIGURE 12 . 29 Zodiacal light shines in the eastern sky before
a summation of eight individual exposures with a total exposure dawn.
time of about an hour.
dust in each cubic kilometer of space. The total amount of

zodiacal dust in the entire Solar System is estimated to be
1016 kg, equivalent to a solid
body about 25 km across, or Zodiacal dust
roughly the size of a large is cometary debris
comet nucleus. Grains of zodi- and ground-up
acal dust are constantly being asteroidal material.
lost as they are swept up by
planets or pushed out of the Solar System by the pressure
of sunlight. Such interplanetary dust grains have been
recovered from Earth’s upper atmosphere by aircraft flying
very high (Figure 12.30). If not replaced by new dust from
comets, all zodiacal dust would be gone within a brief span
of 50,000 years.
In the infrared region of the spectrum, thermal emission
from the band of warm zodiacal dust makes it one of the
FIGURE 12 . 30 A 10-μm-diameter cluster of interplanetary brightest features in the sky. It is so bright that astronomers
dust grains collected in Earth’s stratosphere by a NASA U-2 wanting to observe faint infrared sources are frequently
airplane. hindered by its foreground glow.
at that time. Other asteroids instead carry clues about

Seeing the Forest for the Trees how planets formed. Iron and stony-iron meteorites,
for example, are pieces broken off from what were once
Long before their true nature was understood, comets larger, differentiated bodies. Pick up a stony-iron mete-
were granted great significance by humankind. These orite and you hold in your hand a snapshot, frozen in
spectacular celestial displays were taken as omens of time, of the processes that shaped the world.
great events and, more often than not, as harbingers of Comets and asteroids are far more than scientific
the end of the world. How ironic that they have turned curiosities; they have played a major role in shaping the
out instead to be messengers from the time when our history of life on our planet. Your body is largely water,
world was born. and it is quite likely that much of that water (along with
Anyone who has read a mystery novel knows that the water that covers the surface of our world) arrived
sometimes things that seem least significant at first glance on Earth billions of years ago as volatile-rich comets
turn out to hold the crucial clues to the biggest questions. slammed into the surface of our young world. Across
In our study of the Solar System, two questions rise above the ages, occasional impacts of comets and asteroids on
all others: how did the Solar System form, and what is its Earth have dramatically altered the planet’s climate for
history? Planets may dominate the environment around a time and redirected the course of life’s flow. Intelligent
the Sun, but planets keep imperfect records of the earli- life on Earth descended from mammals only because of
est days of the Solar System. The violence of the planet a cosmic fluke—the impact of a comet, 65 million years
formation process and eons of geological activity together ago, in the region that is now the Yucatán Peninsula. The
effectively erase most clues to their origin. The best sam- awe-inspiring fi reballs that accompanied the impact of
ples of the early Solar System that have survived to the Comet Shoemaker-Levy 9 on Jupiter and the devastation
present day are instead the smallest bodies: asteroids and of a remote corner of Siberia by a small piece of a comet or
especially comets, frozen denizens of the outer reaches of asteroid that hit Earth’s atmosphere in 1908 are remind-
the Sun’s influence. Buried within these dirty snowballs ers that there is a grain of truth in humankind’s deep-
are grains that pre-date even the Solar System itself. seated superstitions about these objects. Sometimes the
These tiny bits of solid material formed in the atmo- appearance of a comet does mean the end of the world as
spheres of stars and in material blasted into space in we know it. It has happened in the past, and unless we
tremendous stellar explosions, survived for a time in develop technology to prevent such events (not an easy
the vast reaches between the stars, participated in the task), it will happen again in the future.
collapse of the interstellar cloud that would become However, as we look to the future of human explora-
the Solar System, and wound up embedded within the tion and utilization of the Solar System, asteroids and
small bodies that are the most numerous citizens of our comets may play a vital and positive role as ready-made
planetary system. The discovery of such grains provides way stations and caches of raw materials that we would
a direct link between our existence and our origins in need if we were to expand beyond our home planet. The
the stars. How remarkable and fortuitous it is that these history and destiny of our kind, as well as the course of
very pieces of our Solar System’s past are delivered to our intellectual journey through the universe, are inex-
our doorstep, falling to Earth as meteorites to be picked tricably tied to this flotsam and jetsam adrift in inter-
up from a cornfield in the Midwest or a glacier in the planetary space.
Antarctic, and then deciphered using the most advanced So far on our journey, we have spent our time digging
tools modern science has to offer. in our own backyard. The Solar System may dwarf the
Grains that pre-date the Sun make up only the tiniest scales of our everyday lives, but it is vanishingly small
fraction of the material found in meteorites. Most of this compared with the universe. Just as we found that we
material was instead formed along with the Solar Sys- could understand our own planet and its history only
tem itself. This material was vaporized in the violence by putting it within the context of the Solar System, we
that accompanied the formation of the disk around the are unable to understand the Sun and the worlds of our
young Sun and then condensed again into solid form. Solar System without placing them within the context of
Some asteroids and all comets are formed directly from the broader universe. The next leg of our journey begins
this pristine material. Meteorites broken off from such as we gaze at the myriad stars that fill the night sky and
bodies provide a window into the conditions that existed wonder about what we see there.
Summary 4. If the dwarf planet Eris is larger than Pluto, why is it

nearly 100 times fainter?
• Pluto, Eris, Haumea, Makemake, and Ceres are classi- 5. How does the composition of an asteroid differ from that
fied as dwarf planets. of a comet nucleus?
• Asteroids are small Solar System bodies made of rock 6. Defi ne meteoroid, meteor, and meteorite.
and metal.
7. What are the differences between a comet and a meteor
• The orbits of most asteroids are located between the in terms of their size, distance, and how long they remain
orbits of Mars and Jupiter. visible?
• Some asteroids cross Earth’s orbit and are potentially *8. Most meteorites (pieces of S-type and M-type asteroids)
dangerous. are 4.54 billion years old. Carbonaceous chondrites
(pieces of C-type asteroids), however, are 20 million years
• Comets are small, icy planetesimals that reside in the
older. What determines the time of “birth” of these pieces
frigid regions beyond the planets.
of rock? What does this information tell you about the
• Comets that venture into the inner Solar System are history of their parent bodies?
warmed by the Sun, producing an atmospheric coma
9. Most meteorites are pieces of asteroids, but a few came
and a long tail.
from elsewhere. What is the origin of these nonasteroi-
• Very large asteroids or comets striking Earth create dal meteorites?
enormous explosions that can wipe out most of terres-
10. Most asteroids are found between the orbits of Mars and
trial life.
Jupiter, but astronomers are especially interested in the
• Meteoroids are small fragments of asteroids and relative few whose orbits cross that of Earth. Why?
comets.
*11. How could you and a friend, armed only with your cell
• When a meteoroid enters Earth’s atmosphere, frictional phones and a knowledge of the night sky, prove conclu-
heat causes the air to glow, producing a phenomenon sively that meteors are an atmospheric phenomenon?
called a meteor.
*12. Suppose you find a rock that has all of the characteris-
• Meteor showers occur when Earth passes through a trail tics of a meteorite. You take it to a physicist friend who
of cometary debris. confi rms that it is a meteorite but says that radioisotope
dating indicates an age of only a billion years. What
• A meteoroid that survives to a planet’s surface is called
might be the origin of this meteorite?
a meteorite.
13. Describe differences between the Kuiper Belt and the
Oort Cloud as sources of comets.
14. Name the three parts of a comet. Which part is the small-
SmartWork, Norton’s online homework system, est? Which is the most massive?
questions, plus additional conceptual exercises. If 15. Kuiper Belt objects (KBOs) are actually comet nuclei.
your instructor assigns questions in SmartWork, log in at Why do they not display comae and tails?
smartwork.wwnorton.com. 16. Comets contain substances closely associated with the
development of life, such as water (H2O), ammonia (NH3),
methane (CH4), carbon monoxide (CO), and hydrogen cya-
Student Questions nide (HCN). Which of these are organic compounds?
17. In 1910, Earth passed directly through the tail of Halley’s
Comet. Among the various gases in the tail was hydro-
1. This chapter deals with leftover planetesimals. What gen cyanide, deadly to humans. Yet nobody became ill
became of the others? from this event. Why?
2. Pluto’s designation as a dwarf planet continues to be a 18. Picture a comet leaving the vicinity of the Sun and rac-
source of controversy among astronomers. Discuss arguing toward the outer Solar System. Does its tail point
ments for and against its being listed as a planet. What backward toward the Sun or forward in the direction
is your opinion and why? the comet is moving? Explain your answer.
3. Describe ways in which Pluto differs significantly from 19. Comets have two types of tails. Describe them and explain
the classical Solar System planets. why they sometimes point in different directions.
20. Explain the orbital terms perihelion and aphelion. *30. Comet Halley has a mass of approximately 2.2 × 1014 kg.
It loses about 3 × 1011 kg each time it passes the Sun.
21. What is the ultimate fate of every short-period comet?
a. The fi rst confi rmed observation of the comet was
22. If collisions of comet nuclei and asteroids with Earth made in 230 B.C. Assuming a constant period of 76.4
are rare events, why should we be concerned about the years, how many times has it reappeared since that
possibility of such a collision? early sighting?
b. How much mass has the comet lost since 230 B.C.?
23. What is the source of meteors we see during a meteor
c. What percentage of its total mass does this amount
shower?
represent?
*24. During a meteor shower, all meteors trace back to a single
31. A cubic centimeter of the air you breathe contains about
region in the sky known as the radiant. Explain why.
1019 molecules. A cubic centimeter of a comet’s tail may
25. What is zodiacal light, and what is its source? typically contain 10 molecules. Calculate the size of a
cubic volume of comet tail material that would hold 1019
molecules.
A P P LY I N G T H E C O N C E P T S *32. The total number of asteroids larger than 1,000 meters
in diameter that cross Earth’s orbit (Aten and Apollo
26. Our Moon has a diameter of 3,474 km and orbits at an
asteroids) is currently estimated to be about 3,500, with
average distance of 384,400 km. At this distance it sub-
five times as many having diameters larger than 500
tends an angle just slightly larger than half a degree in our
meters. Assuming this progression of number versus
sky. Pluto’s moon Charon has a diameter of 1,186 km and
size remains constant for still smaller asteroids, how
orbits at a distance of 19,600 km from the dwarf planet.
many would there be with diameters larger than 125
a. Compare the appearance of Charon in Pluto’s skies
meters? (Note that the impact on Earth of any asteroid
with the Moon in our own skies.
larger than 100 meters in diameter could cause major,
b. Describe where in the sky Charon would appear as
widespread damage.)
seen from various locations on Pluto.
*33. A 1-megaton hydrogen bomb releases 4.2 × 1015 joules
*27. One recent estimate concludes that nearly 800 meteorites
(J) of energy. Compare this amount of energy with that
with mass greater than 100 grams (massive enough to
released by a 10-km-diameter comet nucleus (m = 5 ×
cause personal injury) strike the surface of Earth each
1014 kg) hitting Earth at a speed (v) of 20 km/s. Recall
day. Assuming that you present a target of 0.25 square
from Chapter 4 that EK = 1 ⁄2mv 2 (where EK is the kinetic
meter (m2) to a falling meteorite, what is the probabil-
energy in joules, m is the mass in kilograms, and v is
ity that you will be struck by a meteorite during your
the speed in meters per second).
100-year lifetime? (Note that the surface area of Earth
is approximately 5 × 1014 m2.) 34. The estimated amount of zodiacal dust in the Solar
System remains constant at approximately 1016 kg. Yet
28. Electra is a 182-km-diameter asteroid accompanied by a
zodiacal dust is constantly being swept up by planets
small moon orbiting at a distance of 1,350 km in a circu-
or removed by the pressure of sunlight.
lar orbit with a period of 3.92 days. Refer back to Chapter
a. If all the dust would disappear (at a constant rate)
3 to answer the following questions.
over a span of 30,000 years, what would the average
a. What is the mass of Electra?
production rate, in kilograms per second, have to be
b. What is Electra’s density?
to maintain the current content?
29. The orbital periods of Comets Encke, Halley, and b. Is this an example of static or dynamic equilibrium?
Hale-Bopp are 3.3 years, 76 years, and 2,530 years, Explain your answer.
respectively.
a. What are the semimajor axes (in astronomical units)
of the orbits of these comets?
b. Assuming negligible perihelion distances, what are StudySpace is a free and open Web site that pro-
the maximum distances from the Sun (in astronomi- vides a Study Plan for each chapter of 21st Century
cal units) reached by Comets Halley and Hale-Bopp Astronomy. Study Plans include animations, read-
in their respective orbits? ing outlines, vocabulary flashcards, and multiple-choice
c. Which would you guess is the most pristine comet quizzes, plus links to premium content in SmartWork and
among the three? Which is the least? Explain your the ebook. Visit www.wwnorton.com/studyspace.
reasoning.
PA R T III Stars and Stellar Evolution
To man, that was in th’ evening made,

Stars gave the first delight;
Admiring, in the gloomy shade,
Those little drops of light.
EDMUND WALLER (1606 –1687)
Stars on the Hertzsprung-Russell (H-R) diagram.

CHAP TER 13
Taking the Measure of Stars

13.1 Twinkle, Twinkle, Little
K E Y CONCEP T S
Star, How I Wonder To all but the most powerful of telescopes, a star is
What You Are just a point of light in the night sky. But by applying our
understanding of light, matter, and motion to what we
As children, we look at the sky, see the stars, and may won- see, we are able to build a remarkably detailed picture
der, What are those points of light? How far away are they? of the physical properties of stars. As we take this first
How bright are they? How hot is a star? How big is a star? step beyond our local Solar System, we will
How much does a star “weigh”? What are stars made of? How
• Apply a form of stereoscopic vision, extended by Earth’s
long does a star last? What makes a star shine? These ques-
tions have always been a part of the human experience. orbit, to measure distances to nearby stars.
For most of human history, stars, like so much of nature, • Use the brightness of stars and their distances from
have seemed mysterious and unknowable. Even so, their Earth to discover how luminous they are.
passage follows the daily and annual rhythms that shape
• Use our knowledge of thermal radiation to infer the
our lives. It is no wonder that in the absence of any real
temperatures and sizes of stars from their colors.
understanding of the stars, ancient humans turned their
imaginations loose and viewed the stars as the province of • Measure the composition of stars from spectra.
gods and magic, with power over our lives. Today we under- • Study the orbits of binary stars and use Kepler’s laws
stand that the patterns of the constellations have no more to calculate their masses.
mystical influence over the course of events than does the
• Classify stars, and organize this information on a
random toss of a coin (even though some people persist in
believing the ramblings of astrologers). plot of luminosity versus temperature called the
The story of how we know what we know about the stars Hertzsprung-Russell, or H-R, diagram.
is a wonderful tale. It starts with simple but ingenious obser- • Discover that 90 percent of the stars we see lie along
vations of the sky, borrows from lessons learned about the a well-defined “main sequence” in the H-R diagram.
behavior of matter on Earth and in the Solar System, and
• Discover that the mass and composition of a main-
ends with clear, straightforward answers to the very ques-
sequence star determine its luminosity, temperature,
tions that a child might ask while gazing at the sky.
Unlike our exploration of the Solar System, we cannot and size.
send space probes to a star to take close-up pictures or land • Explore the range of stellar properties, learning how
on its surface! Instead, we study the stars by observing their our Sun compares to other stars.
light, by using some of the laws of physics discussed in ear-
lier chapters, and by fi nding patterns in subgroups of stars
361
362 Chapter 13 Taking the Measure of Stars
that enable us to extrapolate to most other stars. A bit of distances. (Figure 13.1 shows that if each eye is shown a
geometry, a bit about radiation, a bit about orbits— all things different view, the brain can even be fooled into perceiving
that we studied in earlier chapters— and we begin to fi nd distance where none exists.)
ourselves with solid answers to age-old questions. Our stereoscopic vision allows us to judge the distances
of objects as far away as a few hundred meters, but beyond
that it is of little use. Your right eye’s view of a mountain
several kilometers away is indistinguishable from the view
13.2 The First Step Is seen by your left eye. Comparing the two views tells your
brain only that the mountain is too far away for you to judge
Measuring the its distance. (Evolution provided us with only enough depth
perception to judge distances to things that we might be
Distance, Brightness, trying to eat or that might be trying to eat us!) The distance
over which our stereoscopic vision works is limited by the
and Luminosity of Stars separation between our two eyes, which is only about 6
centimeters (cm). If you wanted to increase the differences
In a sense, one of the most amazing feats that a human can between the views your two eyes see, the obvious thing to
perform is to catch a fly ball. Here comes the ball, traveling do would be to somehow move them farther apart. If you
at speeds of 150 kilometers per hour (km/h) or more. To put could separate your eyes by several meters, their perspec-
a glove on the ball, the fielder must judge not only the direc- tives would be different enough to allow you to judge the
tion to the ball but also its distance. How does the fielder do distances to objects that were kilometers away.
it? Like most predators, we have two forward-looking eyes, Of course, we cannot literally take our eyes out of our
one on each side of the face. Our two eyes have somewhat heads and hold them apart at arm’s length, but we can com-
different views of the world. Exactly how different these pare pictures taken from two
perspectives are depends on the distance to the object you widely separated locations. We measure distances
are looking at. If you are looking at a house down the street The greatest separation we to nearby stars by
and you blink back and forth between your two eyes, your can get without leaving Earth comparing the view
view changes very little. But if you are looking at a nearby is to let Earth’s orbital motion from opposite sides of
object—perhaps a finger held up at arm’s length—the per- carry us from one side of the Earth’s orbit.
spectives of your two eyes differ quite a lot. This difference Sun to the other. If we take a
in our eyes’ perspectives on objects at different distances picture of the sky tonight and then wait 6 months and take
is the basis of our stereoscopic vision. Put another way, another picture, our point of view between the two pictures
stereoscopic vision is the major key to the way we perceive will have changed by the diameter of Earth’s orbit, or 2
FIGURE 13.1 Your brain uses the slightly different views offered by your two eyes to “see” the
distances and three-dimensional character of the world around you. This stereoscopic pair shows
the stars in the neighborhood of the Sun as viewed from the direction of the north celestial pole.
The field shown is 40 light-years on a side. The Sun is at the center, marked with a green cross.
The observer is 400 light-years away and has “eyes” separated by about 30 light-years. To get the
stereoscopic view, hold a card between the two images and look at them from about a foot and half
away. Relax and look straight at the page until the images merge into one.
Summer
solstice
Vernal
North equinox
celestial
pole
(toward
observer)
13.2 The First Step Is Measuring the Distance, Brightness, and Luminosity of Stars 363
(a)
A. Nearer Right-eye
pencil (pink) view
B. Farther Left-eye
pencil (blue) view
The brain judges distances to objects

by comparing the view from the left
eye with the view from the right eye.
The two views of nearby objects differ

more than views of distant objects.
Left eye Right eye
(b)
Earth today
1 AU
Star A Star B
Star B is twice as far away as star A…

Earth 6 months later
In the same way, astronomers use the …and appears to move

Star A half as much as star A
changing perspective of Earth through
the year to measure distances to stars. over the course of the year.
Star B
Nearby stars appear to change their

positions more than distant stars do. View today View 6 months later
FIGURE 13. 2 As we move around the Sun, the apparent positions of nearby stars change more than the apparent positions of more
distant stars. (The diagram is not to scale.) This is the starting point for measuring the distances to stars.
astronomical units (AU). With 2 AU separating our two shows Earth, the Sun, and three stars. Look first at star 1.
“eyes,” we should have very powerful stereoscopic vision When Earth, the Sun, and the star are in this position, they
indeed. Figure 13.2 shows how our view of a field of stars form a long, skinny right triangle. The short leg of the tri-
changes as our perspective changes during the year. This angle is the distance from Earth to the Sun, or 1 AU. The
change in perspective is what enables us to measure the long leg of the triangle is the distance from the Sun to the
distances to nearby stars. star. The small angle at the end of the triangle is called the
“parallactic angle,” or simply parallax, of the star. Over
the course of a year, the star’s position in the sky appears
We Use Parallax to Measure to shift back and forth, returning to its original position
1 year later. The amount of this shift—the angle between
Distances to Nearby Stars one extreme in the star’s apparent motion and the other—
The eye cannot detect the changes in position of a nearby is equal to twice the parallax.
star throughout the year, but telescopes can reveal these The more distant the star, the longer and skinnier the
small shifts relative to the background stars. Figure 13.3 triangle that it forms, and the smaller the star’s parallax.
The parallax of a star is one-half of the

angle through which the star appears
to move over the course of a year.
p1 = p2 = p3 =
1 arcsec 1/ 2 arcsec 1/ 3 arcsec
Earth Fixed direction in space
(e.g., a view toward a very distant star)
1 AU
1 pc 2 pc 3 pc
1 arcsec
1/ 2 arcsec 1/ 3 arcsec
Sun Star 1 Star 2 Star 3
1 parsec 2 parsecs 3 parsecs
5 light-years 10 light-years
A star with parallax of More distant stars have smaller The distance to a star in parsecs
1 arcsecond is at a distance parallaxes. The parallax, p, is inversely equals 1 divided by the parallax
of 1 parsec = 3.26 light-years. proportional to the distance: of a star in seconds of arc, or
1 1
p –d . d (parsecs) =
p (arcsecs) .
FIGURE 13. 3 The parallax of three stars at different distances. (The diagram is not to scale.)
Parallax is inversely proportional to distance.
Look again at Figure 13.3. Star 2 is twice as far away as star If the angle at the apex of a triangle is 1 arcsec, and the
1, and its parallax is only half as great. If you were to draw base of the triangle is 1 AU, then the length of the triangle
a number of such triangles for different stars, you would is 206,265 AU (see Appendix
fi nd that increasing the distance to the star always reduces 1). This distance, which cor- A parsec is defined as
the star’s parallax. Move a star 3 times farther away, as with responds to 3.09 × 1016 meters, the distance at which
star 3 in Figure 13.3, and you reduce its parallax to 1 ⁄3 of its or 3.26 light-years, is referred the parallax equals 1
original value. Move a star 10 times farther away, and you to as a parsec (abbreviated pc). arcsecond.
reduce its parallax to 1 ⁄10 of its original value. The parallax The relationship between dis-
of a star (p) is inversely proportional to its distance (d):1 tance measured in parsecs and parallax measured in arc-
seconds is illustrated in Figure 13.3. Using the fact that a
1 1
p ∝ __ or d ∝ __
p. star with a parallax of 1 arcsec is at a distance of 1 pc, we
d
can turn the inverse proportionality between distance and
The parallaxes of real stars are tiny. Rather than talk- parallax into an equation:
ing about parallaxes of 0.0000028° or 4.8 × 10 –8 radian
(1 radian = 57.3°), astronomers normally measure parallaxes ( Distance ) Parallax 1measured .
measured = ____________________
in units of seconds of arc. Just as an hour on the clock is
in parsecs
( in arcseconds )
divided into minutes and seconds of time, a degree can be
divided into minutes and seconds of arc. A minute of arc If you measure the parallax of a star to be 0.5 arcsec, then
(or arcminute, abbreviated arcmin) is 1 ⁄60 of a degree, and you know the star is located at a distance of 1/0.5 = 2 pc. A
a second of arc (or arcsecond, abbreviated arcsec) is 1 ⁄60 of star with a parallax of 0.01 arcsec is located at a distance
a minute of arc.2 That makes a second of arc 1/3,600 of a of 1/0.01 = 100 pc.
degree, or 1/1,296,000 of a complete circle. (The apparent In this book we will usually use units of light-years to
diameter of the full Moon in the sky varies from 29 to 34 indicate distances to stars and galaxies. One light-year is the
arcmin, averaging just over half a degree.) An arcsecond is distance that light travels in 1 year— about 9 trillion kilo-
about equal to the angle formed by the diameter of a golf meters (km). We use this unit because it is the unit you are
ball at the distance of 5 miles. most likely to see in a newspaper article or a popular book
about astronomy. When astronomers discuss distances to
1 stars and galaxies, however, the unit they often use is the
Note that parallax is inversely proportional to distance only if the
parallax is tiny, as it is for stars. For nearby objects, it is the tangent parsec. (You can always convert between the two units. One
of the parallax that is inversely proportional to the distance. parsec equals 3.26 light-years.)
2
Arcseconds are often denoted by the symbol " and arcminutes by '; The star closest to us (other than the Sun) is Proxima
one second of arc is written as 1", and one minute of arc is written as Centauri. Located at a distance of 4.22 light-years, or 1.3
1'. In this text, however, we will spell out these units. pc, Proxima Centauri is a faint member of a system of three
13.2 The First Step Is Measuring the Distance, Brightness, and Luminosity of Stars 365
stars called Alpha Centauri. This star has a parallax of only ume of about 18,000 cubic light-years, corresponding to a
about 3⁄4 arcsec. It is no wonder that ancient astronomers local density of 50 systems per 18,000 cubic light-years. That
were unable to detect the apparent motions of the stars over is about 0.003 star system per cubic light-year. Stated another
the course of a year! way, in the neighborhood of the Sun each system of stars
When astronomers began to apply this technique, they has, on average, about 360 cubic light-years of space (a vol-
discovered that stars are very distant objects indeed. The ume about 4.4 light-years in radius) all to itself.
fi rst successful measurement of the parallax of a star was Knowledge of our stellar neighborhood took a tremen-
made by F. W. Bessel, who in 1838 reported a parallax of dous step forward during the 1990s with the completion of
0.314 arcsec for the star 61 Cygni. This fi nding implied that the Hipparcos mission. The Hipparcos satellite measured
61 Cygni was 3.2 pc away, or 660,000 times as far away as the positions and parallaxes of 120,000 stars. These mea-
the Sun. With this one measurement, Bessel increased the surements, taken from a satellite well above Earth’s obscur-
known size of the universe 10,000-fold! Today we know of ing atmosphere, are better than the measurements that can
65 stars in 50 single-, double-, typically be made from telescopes located on the surface of
or triple-star systems within Stars are few and far Earth. But even this catalog has its limits. The accuracy of
5 pc (16.3 light-years) of the between in our any given Hipparcos parallax measurement is about ±0.002
Sun. A sphere with a radius neighborhood. arcsec. Because of this observational uncertainty, our mea-
of 16.3 light-years has a vol- surements of the distances to stars are not perfect. For exam-
M AT H T O O L S 1 3 .1
The Magnitude System

The magnitude system of brightness for celestial objects can be bers. For example, Sirius, the visually brightest star in the sky,
traced back 2,000 years to the Greek astronomer Hipparchus, has a magnitude of –1.46. Venus can be as bright as magnitude
who classified the brightest stars he could see as being “of the –4.4, or about 15 times brighter than Sirius and bright enough
fi rst magnitude” and the faintest as being “of the sixth magni- to cast a shadow. The magnitude of the full Moon is –12.6, and
tude.” Later, astronomers defi ned Hipparchus’s 1st magnitude that of the Sun is –26.7. Thus, the Sun is 14.1 magnitudes— or
stars as being exactly 100 times brighter than his 6th magnitude more than 400,000 times— brighter than the full Moon.
stars. With five steps between 1st and 6th magnitudes, each step The magnitude of a star, as we have discussed it, is called
is thus equal to the fifth root of 100, or 1001⁄5, which is a factor the star’s apparent magnitude because it reflects the brightness
of approximately 2.512. To put it another way, (2.512)5 = 100. of the star as it appears to us in our sky. As we know, however,
Let’s see how this works. The magnitude system says that stars are found at different distances from us, so a star’s apparent
5th magnitude stars are 2.512 times brighter than 6th mag- brightness does not provide a clue to its luminosity (see Chapter
nitude stars, and that 4th magnitude stars are 2.512 × 2.512 4). But suppose all stars were located at the same distance from
= 6.310 times brighter than 6th magnitude stars. Continuing us. The brightness of each star would then be representative of its
with this progression, we find that 3rd and 2nd magnitude stars luminosity. In principle, astronomers make exactly that assump-
are, respectively, (2.512)3 and (2.512)4 = 15.85 and 39.81 times tion. If the distance to a star is known, they assign a brightness
brighter than 6th magnitude stars. We can now put this progres- to each star as if it were located at a standard distance—in this
sion into a more useful expression. The brightness ratio between case, 10.0 pc (32.6 light-years). Astronomers refer to this funda-
any two stars is equal to (2.512)N, where N is the magnitude dif- mental property of a star as its absolute magnitude.
ference between them. Note that this is really a backward sys- The brightness of astronomical objects generally varies
tem, in which a larger magnitude refers to a fainter object. with spectral region (color), and astronomers use special sym-
Hipparchus must have had typical eyesight, because an bols to represent magnitudes in various colors. For example,
average person under dark skies can see stars only as faint as they use V and B, respectively, to represent magnitudes in
6th magnitude. How does this visual limit compare with that the visual (green) and blue regions of the spectrum. Recall 51
of our most powerful telescopes? Hubble Space Telescope can Pegasi from Chapter 6—the fi rst solar-type star found to have
detect stars as faint as 30th magnitude. Since N = 30 - 6 = 24, an orbiting planet. This star has a visual magnitude, V, of 5.49.
HST can detect stars that are (2.512)24 = 4 × 109, or 4 billion, Its blue magnitude, B, is 6.16 (a larger number), indicating that
times fainter than the naked eye can see! it is less bright in blue light than in visual (green) light. With
So far so good, but what do we do with stars that are brighter these numbers, we can assign the star’s color index, B-V, as
than 1st magnitude? The answer in this backward system is 6.16 - 5.49 = 0.67 magnitude (see Appendix 6 and the second
that we go fi rst to zero magnitude and then to negative num- footnote accompanying Table 13.3).
ple, a star with a parallax measured by Hipparcos of 0.004 dix 6.) But although the brightness of a star is directly mea-
arcsec might really have a parallax of anywhere between surable, it does not immediately tell us much about the star
0.002 and 0.006 arcsec. So instead of knowing that the dis- itself. As illustrated in Figure
tance to the star is exactly 250 pc (1 divided by 0.004 arcsec), 13.4, a bright star in the night Brightness depends
we know only that the star is probably between about 170 sky may in fact be a dim one, on the observer’s
pc (1/0.006 arcsec) and 500 pc (1/0.002 arcsec). With cur- appearing bright only because perspective;
rent technology, parallax becomes useless as a way of mea- it is nearby. Conversely, a faint luminosity does not.
suring stellar distances for stars more than a few hundred star may be a powerful bea-
parsecs away. If you are measuring stellar distances using con, still visible despite its tremendous distance.
parallax, and you need to know the distance to an accuracy To learn about the stars themselves, we need to know the
of 10 percent or better, you are restricted to stars that are total energy radiated by a star each second—the star’s lumi-
less than about 50 pc (160 light-years) away. nosity. In Chapter 4 we studied the relationships between
the brightness, luminosity, and distance of objects. Bor-
rowing from that earlier work, recall that the brightness of
an object that has a known luminosity and is located at a
Once We Know Distance distance d is given by the following equation:
and Brightness, We Can Total light emitted per second
Brightness = ________________________
Calculate Luminosity Area of a sphere of radius d
When we talk about the brightness of an object, we are mak-
Luminosity .
ing a statement about how that object appears to us. In Chap- = __________
ter 4 we saw that brightness corresponds to the amount of 4πd 2
energy falling on a square meter of area each second in the We can rearrange this equation, moving the quantities
form of electromagnetic radiation. (When astronomers talk we know how to measure (distance and brightness) to the
about the brightness of stars, they usually use a system right-hand side and the quantity we would like to know
called magnitude, discussed in Math Tools 13.1 and Appen- (luminosity) to the left, giving us
(a) 1 One star may appear

fainter than another… FIGURE 13. 4 The brightness
of a visible star depends on
2 …either because the star both its luminosity and its
is truly less luminous... distance. If brightness and
distance are measured,
luminosity can be calculated.
3 …or because it
is farther away.
(b)
1 pc
2 pc
3 pc
If brightness and distance
can be measured…
…then luminosity
can be calculated.
13.3 Radiation Tells Us the Temperature, Size, and Composition of Stars 367
Luminosity = 4πd 2 × Brightness.

13.3 Radiation Tells Us the
This equation answers the question, How much total light
must a star be giving off to appear as bright as it does at its
Temperature, Size, and
distance from us? In the process, we take two measurable Composition of Stars
quantities that depend on our particular perspective (dis-
tance and brightness) and from them obtain a quantity that Two everyday concepts— stereoscopic vision, and the fact
is a property of the star itself—namely, luminosity. that the closer an object is the brighter it appears—have
When measuring the luminosity of stars as just described, given us the tools we need to measure the distance and
we fi nd that stars vary tremendously in the amount of light luminosity of relatively nearby stars. In these two steps,
they give off. The Sun provides a convenient yardstick for stars have gone from being merely faint points of light in the
measuring the properties of night sky to being extraordinarily powerful beacons located
stars, including their lumi- Some stars are at distances almost impossible for the mind to comprehend.
nosity. The luminosity of the 10 billion times To continue on this journey of discovery, we again turn to
Sun is written as L ⊙. The most more luminous the laws of radiation that we studied in Chapter 4.
luminous stars can exceed than the least Stars are gaseous, but they are fairly dense— dense
6
1,000,000 L ⊙, or 10 L ⊙ (a mil- luminous stars. enough that the radiation from a star comes close to obey-
lion times the luminosity of ing the same laws as the radiation from objects like the
the Sun). The least luminous stars have luminosities less heating element on an electric stove or the filament in a
than 0.0001 L ⊙, or 10 –4 L ⊙(1/10,000 that of the Sun). The most lightbulb. That means we can use our understanding of
luminous stars are therefore over 10 billion (1010) times more Planck radiation— results such as the Stefan-Boltzmann
luminous than the least luminous stars. Only a very small law (hotter at same size means more luminous) and Wien’s
fraction of stars are near the law (hotter means bluer)—to understand the radiation from
upper end of this range of There are many stars. In particular, these two laws enable us to measure the
luminosities. The vast major- more low-luminosity temperatures and sizes of our stellar neighbors.
ity of stars are at the faint end stars than high-
of this distribution, less lumi- luminosity stars.
nous even than our Sun. Fig- Wien’s Law Revisited: The Color
ure 13.5 shows the relative number of stars compared to
their luminosity in solar units. (Distances for the nearest
and Surface Temperature of Stars
stars are obtained from their parallax; other methods—to The hotter the surface of an object, the bluer the light that
be discussed later— are used for the more distant stars.) it emits. Stars with especially
hot surfaces are blue, stars Measuring the color
FIGURE 13. 5 The distribution of the luminosities of stars. Both with especially cool surfaces of a star tells us its
axes are plotted logarithmically, as explained in Section 13.5. are red, and our Sun is a surface temperature.
middle-of-the-road yellow.
The vast majority of stars are
less luminous than the Sun. More formally, Wien’s law states that
10 2,900 μm K
T = __________,
Luminosity of Sun λpeak
1 £
where λpeak is the wavelength, in micrometers (μm, 10 –6
meter), at which the electromagnetic radiation from a star
10–1
Relative number of stars
is most intense. Obtain a spectrum of a star, measure the

wavelength at which the spectrum peaks, and Wien’s law
10–2 will tell you the temperature, T, of the star’s surface.3
In practice, it is usually not necessary to obtain a com-
10–3 plete spectrum of a star to determine its temperature.
Only a few are at the upper Instead, stellar temperatures are often measured in a way
10–4 range of luminosities. that is similar to the way that your eyes see color. Your eyes
and brain distinguish color by comparing how bright an
10–5
3
It is important to remember that the color of a star tells us only about
10–6
the temperature at the star’s surface, because the surface is the part
of the star giving off the radiation that we see. Star surfaces may be
10–4 10–2 1 102 104 106 hot, but later in our journey we will discover that star interiors are
Stellar luminosity (in units of solar luminosities) far hotter still.
A hot star appears brighter A cool star appears brighter transmits roughly the range of wavelengths to which our
through a blue filter than through through a visual filter than eyes are most sensitive.
a visual filter (bB /bV > 1). through a blue filter (bB /bV < 1). Figure 13.6 shows four Planck spectra with temperatures
of 2,500–20,000 kelvins (K), adjusted so that they are all the
4 same brightness at 550 nm (the wavelength of the center of
Light seen Light seen
through through
the range transmitted by the visual filter). A hot star, with a
Brightness relative to brightness
blue filter visual filter spectrum like the 20,000-K Planck spectrum shown, gives
3 off more light in the blue part of the spectrum than in the
visual part of the spectrum. If we were to divide the bright-
ness of the star as seen through the blue filter by the bright-
20
at 500 nm
00
,0
2 K K
ness of the star as seen through the visual filter, the blue/
0 visual brightness ratio (bB/bV) would be greater than 1. In
50
10,0 2,
00 K contrast, a cool star with a spectrum more like the 2,500-K
Planck spectrum shown is much fainter in the blue part of
1
the spectrum than in the visual part of the spectrum. The
K
5,000 ratio of blue light to visual light is less than 1 for the cool
star. This ratio of brightness between the blue and visual
0
filters is referred to as the color index of the star.5 By con-
350 400 450 500 550 600 650 700 vention, astronomers usually adjust the way they measure
Wavelength (nm) the blue and visual brightnesses of stars so that the star
FIGURE 13.6 A star’s color depends on its temperature. The
Vega (which has a surface temperature of about 10,000 K)
Planck spectra (blackbody curves) shown here are adjusted so has a defined color index of 1.00. On this scale, the Sun has
that they have the same brightness at 550 nm. A wider range of a color index of 0.56.
blackbody curves is shown in Figure 4.21. The fact that the color of a star depends on the star’s
temperature is an extremely handy tool for astronomers.
It means that a single pair of snapshots of a group of stars,
each taken through a different filter, provides a measure-
object is at one wavelength with how bright it is at another ment of the surface temperature of every star in our camera’s
wavelength. The human eye contains nearly 100 million field of view! This technique makes it possible to measure
light-sensitive cells, and among them are certain color- the temperatures of hundreds or even thousands of stars at
sensitive cells called cones. Cones come in three color- once. When we do, we fi nd that just as there are many more
discriminating types, with their peak responses in red, low-luminosity stars than high-luminosity stars, there are
green, and blue regions of the spectrum, respectively.4 If you also many more cool stars than hot stars. We also discover
look at a very red lightbulb, the red-sensitive cones in your that most stars have surface temperatures lower than that
eye report a bright light, and the blue- and green-sensitive of the Sun. AstroTour: Stellar Spectrum
cones see less light. If you look at a yellow light, the green-
and red-sensitive cones are strongly stimulated, but those
with blue sensitivity are not. White light stimulates all three Stars Are Classified According
kinds of color-sensitive cells. By combining the signals from
cells that are sensitive to different wavelengths of light, your
to Their Surface Temperature
brain can distinguish among fi ne shades of color. So far, we have concentrated on what we can learn about
Astronomers often measure the colors of stars in much stars by applying our understanding of thermal radiation,
the same way. The brightness of a star is usually measured an approach that seems from
through a filter— sometimes just a piece of colored glass— the previous section not to Absorption lines
that lets through only a certain range of wavelengths. Two have led us too far astray. form as light
of the most common filters used by astronomers are a blue However, the spectra of stars escapes through
filter that transmits light with wavelengths of about 440 are far from perfect Planck the atmosphere of
nanometers (nm), and a yellow-green filter that passes light spectra. Instead, when we the star.
with wavelengths of about 550 nm. The fi rst of these fil- look at the spectra of stars, we
ters is (sensibly enough) called a “blue” filter. By conven- see a wealth of absorption lines and occasionally emission
tion, however, the second of these filters is referred to as a lines. In Chapter 4 we found that absorption lines occur
“visual” filter rather than a “yellow-green” filter because it when light passes through a cloud of gas: the atoms and
4 5
Cones sensitive to red and green light have a rather broad spectral Rather than a blue/visual brightness ratio, astronomers generally refer
response and, in fact, partially overlap one another. This overlap is to color index as the difference between an object’s blue magnitude (B)
the cause of red-green color blindness in some people. and its visual magnitude (V)—that is, B-V (see Math Tools 13.1).
Planck
spectrum
Intensity
A continuous Planck spectrum emerges
from the hot interior of the star.
Wavelength, λ
As radiation passes through the star’s

atmosphere, atoms absorb photons
Absorbing of certain wavelengths, causing
atmosphere absorption lines in the stellar spectrum.
E2 E2
Photon
Energy
Intensity absorbed
Absorption
spectrum
E1 E1
Emitting Atoms in the hot outer atmosphere

of the star may also emit photons, E2 E2
hot gas
causing emission lines in the
spectra of some stars.
Energy
Photon
Intensity
emitted
Emission
spectrum
E1 E1
350 400 450 500 550 600 650 700

Wavelength (nm)
FIGURE 13.7 The formation of absorption and emission lines in the spectra of stars.
molecules of the gas absorb light of certain specific wave- properties, but on the appearance of the dark bands (now
lengths, characteristic of the kind of atom or molecule that known as absorption lines) seen in their spectra. The origi-
is doing the absorbing. Similarly, the atoms and molecules nal ordering of this classifica-
in diffuse hot gas will emit light of specific wavelengths as tion was arbitrarily based on Stars are classified
well. Both of these processes are at work in stars. the prominence of particular by the appearance
Although the hot “surface” of a star emits radiation with absorption lines known to be of their spectra.
a spectrum very close to a smooth Planck curve, this light associated with the element
must then escape through the outer layers of the star’s atmo- hydrogen. Stars with the strongest hydrogen lines were
sphere. The atoms and molecules in the cooler layers of the labeled “A stars,” stars with somewhat weaker hydrogen
star’s atmosphere leave their absorption-line fingerprints lines were labeled “B stars,” and so on.
in this light, as shown in Figure 13.7. The atoms and mol- This classification scheme was refined and turned into
ecules in the star’s atmosphere and any gas that might be a real sequence of stellar properties in the early 20th cen-
found in the vicinity of the star can also produce emission tury. The new system, originally proposed in 1901, was the
lines in stellar spectra. Although absorption and emission work of Annie Jump Cannon (1863–1941), who led an effort
lines complicate how we use the laws of Planck radiation to at the Harvard College Observatory to systematically exam-
interpret light from stars, spectral lines more than make up ine and classify the spectra of hundreds of thousands of
for this trouble by providing a wealth of information about stars. She dropped many of the earlier spectral types, keep-
the state of the gas in a star’s atmosphere. ing only seven, which she reordered into a sequence that
The spectra of stars were fi rst classified during the late was no longer arbitrary, but instead was based on surface
1800s, long before stars, atoms, or radiation were well under- temperatures. Spectra of stars of different types are shown
stood. Stars were classified not on the basis of their physical in Figure 13.8. The hottest stars, with surface temperatures
FIGURE 13.8 Spectra of

O5 44,500
stars with different spectral
B0 30,000 types, ranging from hot blue
B5 15,400 O stars to cool red M stars.
Hotter stars are brighter at
A0 9,520 shorter wavelengths. The dark
A5 8,200 lines are absorption lines.
Surface temperature (K)

F0 7,200
Spectral types
F5 6,440
G0 6,030
G2 5,860
G5 5,770
K0 5,250
K5 4,350
M0 3,850
M5 3,240
350 400 500 600 700

Near UV Wavelength (nm) Near IR
over 30,000 K, are labeled “O stars.” O stars show relatively Most absorption lines have an optimal temperature at
featureless spectra, with only weak absorption lines from which they are formed most strongly. For example, absorp-
hydrogen and helium. The coolest stars—“M stars”—have tion lines from hydrogen are most prominent at surface
temperatures as low as about 2,800 K. M stars show myriad temperatures of about 10,000 K, which is the surface tem-
lines from many different types of atoms and molecules. The perature of an A star. (This should be no surprise. Spectral-
complete sequence of spectral types of stars, from hottest to type A stars were so named because they are the stars with
coolest, is O, B, A, F, G, K, M. This sequence has undergone the strongest lines of hydrogen in their spectra.) At the very
several modifications over time. lowest stellar temperatures, atoms in the atmosphere of the
It should be stressed that the boundaries between spec- star begin to react with each other, forming molecules. Mol-
tral types are not precise. A hotter-than-average G star is ecules such as titanium oxide (TiO) are responsible for much
very similar to a cooler-than-average F star. Astronomers of the absorption in the atmospheres of cool M stars.
break down the main spectral types into a fi ner sequence Because different spectral lines are formed at different
of subclasses by adding numbers to the letter designations. temperatures, we can use these absorption lines to measure
For example, the hottest B stars are called B0 stars, slightly a star’s temperature directly. The surface temperatures of
cooler B stars are called B1 stars, and so on. The coolest B stars measured in this way agree extremely well with the
stars are B9 stars, which are only slightly hotter than A0 surface temperatures of stars measured using Wien’s law,
stars. The Sun is a G2 star. again confi rming a prediction of the cosmological prin-
Returning to Figure 13.8, we can see that not only are hot ciple: the physical laws that apply on Earth apply to stars
stars bluer than cool stars, but the absorption lines in their as well. AstroTour: Stellar Spectrum
spectra are quite different as well. The reason is that differ-
ences in the temperature of the gas in the atmosphere of a
star affect the state of the atoms in that gas, which in turn Stars Consist Mostly of
affects the energy level transitions available to absorb radia-
tion. At the hot end of the spectral classification scheme,
Hydrogen and Helium
the temperature in the atmospheres of O stars is so high The most obvious differences in the lines seen in stellar
that most atoms have had one or more electrons stripped spectra are due to tempera-
from them by energetic collisions within the gas. There are ture, but the details of the Spectral lines are
few transitions available in these ionized atoms that cause absorption and emission used to measure many
absorption lines in the visible part of the electromagnetic lines found in starlight carry properties of stars,
spectrum, so the visible spectrum of an O star is relatively a wealth of other information including chemical
featureless. At lower temperatures there are more atoms in as well. By applying our composition.
energy states that can absorb light in the visible part of the knowledge of the physics of
spectrum, so the visible spectra of cooler stars are far more atoms and molecules to the study of stellar absorption
complex than are the spectra of O stars. lines, we can accurately determine not only surface tem-
peratures of stars, but also pressures, chemical composi- elements, which are collectively referred to as heavy ele-
tions, magnetic-field strengths, and other physical ments or (more properly) as “massive elements,” are present
properties of stars. In addition, by making use of the Dop- in only trace amounts.6 (This would be a good time to go
pler shift of emission and absorption lines, we can measure back and review the “astronomer’s periodic table” in Fig-
rotation rates, motions of the atmosphere, expansion and ure 9.5.) Table 13.1 shows the chemical composition of the
contraction, “winds” driven away from stars, and other atmosphere of the Sun, which is fairly typical for stars in
dynamic properties of stars. our vicinity. On the other hand, chemical composition can
As we continue to learn about stars, we will come to vary tremendously from star to star. In particular, some stars
appreciate that one of the most interesting and important show even smaller amounts of elements other than hydrogen
things about a star is its chemical composition. We already and helium in their spectra. The existence of such stars, all
saw in Chapter 4 that if we look at a source of thermal radia- but devoid of more massive elements, provides important
tion through a cloud of gas, the strength of various absorp- clues about the origin of chemical elements and the chemi-
tion lines tells us what kinds of atoms are present in the gas cal evolution of the universe.
and in what abundance. Although we must take great care in
interpreting spectra to properly account for the temperature
and density of the gas in the atmosphere of a star, in most The Stefan-Boltzmann Law Revisited:
respects stars are ready-made laboratories for carrying out
just such an experiment.
Finding the Sizes of Stars
When analyzed in this way, most stars are found to have Once we know the temperature of a star, we also know how
atmospheres that consist predominantly of the least massive much radiation each square meter of the star is giving off
elements. Hydrogen typically makes up over 90 percent of each second. As shown in Figure 13.9b, each square meter
the atoms in the atmosphere of a star, with helium account- of the surface of a hot blue star gives off more radiation per
ing for most of what remains. All of the other chemical second than a square meter of the surface of a cool red star.
A hot star will be more luminous than a cool star of the same
size. A small hot star might even be more luminous than
a larger cool star. As noted in Figure 13.9c, we can use the
TA B L E 1 3 .1
relationship between temperature and luminosity of each
square meter of a surface to infer the sizes of stars.
According to the Stefan-Boltzmann law (see Chapter 4), the
The Chemical Composition of amount of energy radiated each second by each square meter
the Sun's Atmosphere* of the surface of a star is equal to the constant σ multiplied
by the surface temperature of the star raised to the fourth
power. Written as an equation, this relationship says
Percentage by Percentage by
Element Number† Mass††
Energy radiated each
= σT 4.
Hydrogen 92.5 74.5 second by 1 m2 of surface
Helium 7.4 23.7
To fi nd the total amount of light radiated each second by
Oxygen 0.064 0.82 the star, we need to multiply the radiation per second from
each square meter by the number of square meters of the
Carbon 0.039 0.37
star’s surface:
Neon 0.012 0.19
Total energy radiated Energy radiated Number of square
Nitrogen 0.008 0.09
by the star = by a square meter × meters of surface .
Silicon 0.004 0.09 each second each second of the star
Magnesium 0.003 0.06 The left-hand item in this equation—the total energy
Iron 0.003 0.16 emitted by the star per second (in units of joules per sec-
ond, abbreviated J/s)—is the star’s luminosity, L. The middle
Sulfur 0.001 0.04 item—the energy radiated by each square meter of the star in
Total of others 0.001 0.03 a second (in units of joules per square meter per second, or
J/m2/s)—can be replaced with the σT 4 factor from the Stefan-
*The relative amounts of different chemical elements in the atmo- Boltzmann law. The remaining item—the number of square
sphere of the Sun.
†
The percentage of atoms accounted for by the listed element.
6
††
The percentage of mass consisting of the listed element. Astronomers often (although incorrectly) refer to all elements more
massive than hydrogen and helium as metals.
FIGURE 13.9 We measure

(a) (b)
the temperature and size
of stars by applying our
Stefan-Boltzmann Law: understanding of Planck
A square meter of a hot …so a hot star is more
star’s surface radiates luminous than a cool radiation.
more energy than a star of the same size.
square meter of a cool
star’s surface…
Wien’s Law:
Blue stars are hot,
red stars are cool.
(c)
d
R
Measure luminosity
from distance
and brightness.
Calculate how big
the star must be.
peak
Intensity
Measure temperature
from color.
Wavelength
meters covering the surface of the star—is the surface area matical expression is really a shorthand way of writing
of a sphere, Asphere = 4πR2 (in units of square meters, or m2), our understanding of what is going on physically—in this
where R is the radius of the star. If we replace the words in case the commonsense result that larger, hotter stars are
the equation with the appropriate mathematical expressions more luminous than smaller, cooler stars.)
for the Stefan-Boltzmann law and the area of a sphere, our The previous equation answers the question, How lumi-
equation for the luminosity of a star looks like this: nous will a star of a given size and a given temperature be?
Now turn this question around and ask, How large does a
J J
Luminosity _ = ___
s m2s () ( )
× m2 star of a given temperature need to be to have a total lumi-
nosity of L? The star’s luminosity (L) and temperature (T)
are quantities that we can
= (σT 4) × (4πR2). measure, and the star’s radius If the temperature
(R) is what we want to know. and luminosity of a
Combining the right sides of these two equations, we get Rearrange the previous equa- star are known, its
tion, moving the things that size can be calculated.
L = 4πR2 σT 4, in units of J/s (watts). we know how to measure
(temperature and luminosity) to the right-hand side of the
It is worth thinking about this equation for a minute equation, and the thing that we would like to know (the
to see the sense behind it. Because the constants (4, π, and radius of the star) to the left side. After a couple of steps of
σ) do not change, the luminosity of a star is proportional algebra, we fi nd
only to R2T 4. The R2 makes sense: The surface area of a star ____
is proportional to the square of the radius of the star. Make
a star three times as large, and its surface area becomes
R= √ 4πσ
L
×
1
____
T
. ___
2
32 = 9 times as large. There is

nine times as much area to A larger and hotter Again, the right-hand side of the equation contains only
radiate, so there is nine times star is also more things that we know or can measure. The constants 4, π,
as much radiation. The effect luminous. and σ are always the same. We can fi nd L, the luminosity
of T 4 is understandable as of the star, by combining measurements of the star’s bright-
well: make a star twice as hot, and each square meter of ness and parallax (although only for those nearer stars with
the star’s surface radiates 24 = 16 times as much energy. known parallax). T is the surface temperature of the star,
(As we have seen before, an apparently complex mathe- which can be measured from its color. From the relation-
13.4 We Measure Stellar Masses by Analyzing Binary Star Orbits 373
ship of these measurements we now know something new:

the size of the star. We will refer to this last equation as the Binary Stars Orbit a Common
luminosity-temperature-radius relationship for stars. Center of Mass
The luminosity-temperature-radius relationship has
been used to measure the sizes of many thousands of stars. When discussing the motions of the planets, we usually say
When we talk about the sizes of stars, we again use our Sun that the planets orbit around the Sun and we don’t worry
as a yardstick. The radius of the Sun, written R⊙, is 696,000 about the effect of their gravity on the Sun’s motion. On the
km, or about 700,000 km. When we look at stars around us, other hand, if two objects— say, two stars— are closer to the
we fi nd that the smallest stars we see, called white dwarfs, same mass, we have to abandon this simple mental picture.
have radii that are only about 1 percent of the Sun’s radius Each object’s motion will be noticeably affected by the gravi-
(R = 0.01 R⊙). The largest stars that we see, called red super- tational force from the other, and we must worry about what
giants, can have radii more happens when the two objects orbit each other. (In fact, as
than 1,000 times that of the There are many we learned in Chapter 6, our knowledge of planets beyond
Sun. There are many more more small stars the Solar System comes primarily from the wobbles they
stars toward the small end of than large stars. cause in the motions of stars.)
this range, smaller than our Think about what happens if we set two unequal
Sun, than there are giant stars. masses—m1 and m2—adrift in space with no motion of one
mass relative to the other. As soon as we release the two
masses, gravity begins to pull them together. The force of
mass 1 on mass 2 equals the force of mass 2 on mass 1, and
13.4 We Measure Stellar each mass begins to fall toward the other. But even though
the forces on each mass are equal, the accelerations experi-
Masses by Analyzing enced by the two masses are not. Acceleration equals force
divided by mass, so the less massive object experiences the
Binary Star Orbits greater acceleration.
Suppose, for the moment, that m1 is three times as mas-
The most important property of stars that we have yet to sive as m2. That means that the acceleration of m2 will be
investigate is their mass. Determining the mass of an object three times as great as the acceleration of m1. At any given
can be a tricky business. We certainly cannot rely on the point in time, m2 will be moving toward m1 three times as
amount of light from an object or the object’s size as a mea- fast as m1 is moving toward m2. When the two objects col-
sure of its mass. Massive objects can be large or small, faint lide, m2 will have fallen three times as far as m1. The point
or luminous. The only thing where the two objects meet, called the center of mass of
that always goes with mass is To measure mass, the two objects, will be three times as far from the original
gravity. Mass is responsible astronomers look for position of the less massive m2 as from the original position
for gravity, and gravity in turn the effects of gravity. of the more massive m1. (If the two objects were sitting on
affects the way masses move. a seesaw in a gravitational field, the support of the seesaw
When astronomers are trying to determine the masses of would have to be directly under the center of mass for the
astronomical objects, they almost always wind up looking objects to balance, as shown in Figure 13.10.)
for the effects of gravity. If we give our two masses some motion perpendicular
When discussing the orbits of the planets in Chapter to the line between them, then
3, we found that Kepler’s laws of planetary motion are the instead of simply falling into Each star in a
result of gravity. We even went so far as to show that the each other they will orbit binary system follows
properties of the orbit of a planet about the Sun can be used around each other. When an elliptical orbit
to measure the mass of the Sun. If we could fi nd a planet Newton applied his laws of around the system’s
orbiting a distant star, we could apply the same technique motion to the problem of center of mass.
to determine the mass of that star. Unfortunately, today’s orbits, he found that two
telescopes are barely powerful enough to directly see plan- objects move in elliptical orbits with their common center
ets orbiting other stars, but we can watch as two stars orbit of mass at one focus of both of the ellipses, as shown in Fig-
about each other. About half of the higher-mass stars in the ure 13.11. The center of mass, which lies along the line
sky are actually multiple systems consisting of several stars between the two objects, remains stationary. The two objects
moving about under the influence of their mutual gravity. will always be found on exactly opposite sides of the center
Most of these are binary stars in which two stars orbit each of mass, and the elliptical orbit of the more massive object
other in elliptical orbits as predicted by Kepler’s laws. How- is just a smaller version of the elliptical orbit of the less
ever, most low-mass stars are single, and low-mass stars far massive object.
outnumber their higher-mass brethren. This means that Because the perimeter of the orbit of the less massive star
most stars are single. is larger than that of the more massive star’s orbit, the less
Two objects fall toward each other.

m1 = 3m2 m2
Center of mass
Star 1 Star 2
The forces on the
objects are the same…
Center of mass
…but the accelerations
are not. The less massive
object picks up speed faster.
The less massive
As the objects star moves faster
fall, the center on a larger orbit.
of mass is
Orbit of more
stationary.
massive star
If m1 = 3m2, then (m2 = 2m1)
m2 falls 3 times
faster than m1…
...and three times

d1 d2 = 3d1 The center of mass
as far as m1.
remains stationary
while the stars orbit.
The point where d2 m1 3
they hit is their = =
d1 m2 1
center of mass.
m1 = 3m2 m2
The center of mass is also where
the fulcrum of a balance must be.
FIGURE 13.10 The center of mass of two objects is the

“balance” point on a line joining the centers of two masses.
massive star must be moving FIGURE 13.11 In a binary star system, the two stars orbit
faster than the more massive The less massive star on elliptical paths about their common center of mass. In this
star. The less massive star has has a larger orbit and case, star 2 has twice the mass of star 1. The eccentricity of the
farther to go than the more so must move faster. orbits is 0.5. There are equal time steps between the frames.
massive star, but it must cover
that distance in the same amount of time. As a result, the
velocity (v) of a star in a binary system is inversely propor-
that the absorption lines from star 1 are redshifted relative
tional to its mass:
to the overall center-of-mass velocity, while the absorp-
v1 ___
__ m tion lines from star 2 are blueshifted. If you were to look
= 2.
v2 m1 again half an orbital period later, the situation would be
reversed: lines from star 2 would be redshifted and lines
This relationship between the velocities and the masses from star 1 would be Doppler-shifted to the blue. Because
of the stars in a binary system is a crucial part of how the two stars are always exactly opposite one another
binary stars are used to measure stellar masses. Imagine about their common center of mass, they are always mov-
that you are watching a binary star edge-on, as shown in ing in opposite directions. We can measure the ratio of
Figure 13.12. As one star is moving toward you, the other the masses of the two stars by comparing the size of the
star will be moving away from you; and vice versa. If you Doppler shift in the spectrum of star 1 with the size of
were to look at the spectra of the two stars, you might see the Doppler shift in the spectrum of star 2.
13.4 We Measure Stellar Masses by Analyzing Binary Star Orbits 375
Stars in a binary system

orbit their center of mass.
Time = 0 1/ 4 period 1/ 2 period 3/ 4 period 1 period
v2
Star 1
Star 2
v1
Center
Observations
of mass
show absorption
lines from
both stars
in the spectrum.
Spectrum
of binary
system
– 0 + – 0 + – 0 + – 0 + – 0 +
Doppler shift Doppler shift Doppler shift Doppler shift Doppler shift
Star 1 is moving toward the Star 2 is moving away from When star 1 is redshifted, When the stars are moving
viewer, so it is blueshifted the viewer, so it is redshifted star 2 is blueshifted; and neither away nor toward the
(shorter [–] wavelengths). (longer [+] wavelengths). vice versa. viewer, their lines coincide.
v2
Doppler shift of star 2 (less massive)
Doppler velocity
v1
0
0P 1/ 4 P 1/ 2 P 3/ 4 P P
–v1
Doppler shift of star 1 (more massive) The higher-mass star has
–v2 a small Doppler shift. Here
v2 = 2v1, so m1 = 2m2.
Time
FIGURE 13.12 The orbits of two stars in a binary system are shown here, along with the Doppler shifts of the absorption lines in
the spectrum of each star. Star 1 is twice as massive as star 2 and so has half the Doppler shift.
4π2 A3
Kepler’s Third Law Gives the P 2 = __________.
G (m1 + m2)
Total Mass of a Binary System Rearranging this a bit turns it into an expression for the
sum of the masses of the two objects:
In Chapter 3 we conveniently ignored all of this complex-
ity having to do with the motion of two objects about their 4π2 A 3
m1 + m2 = ___ × ___2 .
common center of mass. Now, however, it is this very com- G P
plexity that enables us to measure the masses of the two
stars in a binary system. In his derivation of Kepler’s laws, We can cleverly ignore the value of 4π2/G by using what we
Newton showed that if two objects with masses m1 and m2 know about Earth’s orbit around the Sun, and then express-
are in orbit about each other, then the period of the orbit, P, ing m, A, and P as ratios. If A = 1 AU and P = 1 year, then we
is related to the average distance between the two masses, know that m1 + m2 = M⊙ + M⊕ ≈ M⊙ (because M⊙ is much
A,7 by the equation larger than M⊕). So if we express the masses m1 and m2, A,
and P in that equation in terms of Solar System units— such
7
Here, A is equivalent to the semimajor axis, which we discussed in as m1 = m1/(1 M⊙), AAU = A/(1 AU), and Pyears = P/(1 year)—
Chapter 3. then this equation simplifies to
m1 ___
___ m (AAU)3 star gives its total orbital velocity. The period of the orbit
+ 2 = ______ . is given by the time it takes for a set of spectral lines to go
M⊙ M⊙ (Pyears)
2
from approaching to receding and back again. If we know

We now know all that we need to know to measure the orbital velocities of the stars and we know the period
the masses of two stars in a binary system. If we can mea- of the orbit, we also know the size of the orbit because dis-
sure the period of the binary tance equals velocity multiplied by time. (Observations of
system and the average sepa- Kepler’s third law the changing brightness of eclipsing binaries can provide
ration between the two stars, gives the total mass an additional bonus. Under certain conditions, we can mea-
then Kepler’s third law gives of a binary system. sure the size [radii] of the stars as well.)
us the total mass. Referring Historically, most stellar masses were measured for
back to the previous subsection, if we can measure the stars in eclipsing binary systems because all that we need
sizes of the orbits of the two stars independently, or inde- to determine the mass in this case is a series of spectra of
pendently measure their velocities, then we can deter- the system. Recently, however, new observational capabili-
mine the ratios of the masses of the two stars. If we know ties have greatly improved our ability to see the stars in a
the total mass of the system, as well as the ratio of the binary directly. Accurate measurements of masses have
two masses, we have all we need to determine the mass been obtained for several hundred binary stars, about half of
of each star separately. To express these relationships in which are eclipsing binaries. (Math Tools 13.2 gives a con-
symbols, if we know both m1/m2 and m1 + m2, a bit of crete example of how the masses of two stars in an eclips-
algebra yields separate values for m1 and m2, as can be ing binary are determined.) These measurements show that
seen in Math Tools 13.2. some stars are less massive than the Sun, whereas others
There are two ways to go about measuring the orbital are more massive.
properties that we need to determine the masses of stars. If The range of stellar masses is not nearly as great as the
a binary system is a visual binary— that is, if we can take range of stellar luminosities. The least massive stars have
pictures that show the two stars separately— then we can masses of about 0.08 M⊙; the most massive stars probably
watch over time as the stars orbit each other. From these have masses somewhere between 130 and 150 M⊙. You
observations we can measure the shapes and period of the might wonder why the mass of a star should have any lim-
orbits of the two stars. When combined with Doppler mea- its. These lower and upper limits are determined solely by
surements of the line-of-sight velocities of the stars, this the physical processes that go on deep in the interior of the
is all the information we need. In many binary systems, star. We will learn in the chapters ahead that a minimum
however, the two stars are so close together and so far away stellar mass is necessary to ignite the nuclear furnace that
from us that we cannot actually see the stars separately. We keeps a star shining, but the furnace can run out of control
know these stars belong to binary systems only because we if the stellar mass is too great.
see the spectral lines of the two stars as they are Doppler- Thus, although stellar luminosities change by a factor
shifted away from each other fi rst in one direction and of 1010, or 10 billion, from most luminous to least luminous,
then in the other; these are called spectroscopic binary the most massive stars are only about 103, or a thousand,
stars. Yet even those Doppler shifts alone can sometimes times more massive than the least massive stars.
be used to measure the masses of such stars.
If a binary system is an eclipsing binary, in which we see
a dip in brightness as one star passes in front of the other
(Figure 13.13), then we know we are looking at the system
nearly edge-on. In this case, the peak Doppler shift of each
13.5 The H-R Diagram
Is the Key to
FIGURE 13.13 The shape of the light curve of an eclipsing Understanding Stars
binary can reveal information about the relative size and surface
brightness of the two stars. We have come a long way in our effort to measure the
physical properties of stars. Table 13.2 summarizes the
techniques we have used for certain types of stars, such
as nearby stars and binary stars. However, just knowing
some of the basic properties of stars does not mean that
we understand stars. The next step in our journey involves
looking for patterns in the properties we have determined.
The fi rst astronomers to take this step were a Dane by the
Intensity
name of Ejnar Hertzsprung (1873–1967) and the American

astronomer Henry Norris Russell (1877–1957). In the early
part of the 20th century (from 1906 to 1913, to be precise),
Time Hertzsprung and Russell were independently studying the
M AT H T O O L S 1 3 . 2
Using a Celestial Bathroom Scale to Measure

the Masses of an Eclipsing Binary Pair
It is worth looking at an example to see how the relationships Next you apply Kepler’s third law, which says that
discussed in the text are actually put to work. Suppose you are 3
m
m1 ___
___ (AAU)
an astronomer studying a binary star system. After observing + 2 = ______2
.
M⊙ M⊙ (Pyears)
the star for several years, you accumulate the following infor-
mation about the binary system: where m1/M⊙ refers to the star’s mass in terms of solar masses
and A is the average distance between the two stars. Here
1. The star is an eclipsing binary.
you have A = 1.8 AU + 0.6 AU = 2.4 AU. Because you know
2. The period of the orbit is 2.63 years. You learned this by
A and the period P, you can calculate the total mass of the
observation.
two stars:
3. Star 1 has a Doppler velocity that varies between +20.4 and
-20.4 kilometers per second (km/s). m
m1 ___ (AAU) 3
(2.4) 3
___ + 2 = ______ = ______2 = 2.0,
4. Star 2 has a Doppler velocity that varies between +6.8 and M⊙ M⊙ 2
(Pyears) (2.63)
-6.8 km/s.
5. The stars are in circular orbits. You know this because the so you have learned that the combined mass of the two stars is
Doppler velocities about the star are symmetric. twice the mass of the Sun.
To sort out the individual masses of the stars, you use
These data are summarized in Figure 13.14. You begin your the fact that the more massive the star is, the less it moves in
analysis by noting that the star is an eclipsing binary, which response to the gravitational attraction of its companion:
tells you that the orbit of the star is edge-on to your line of
m2 __
___ v 20.4 km/s
sight. The Doppler velocities tell you the total orbital velocity
m1 = 1 = ________ = 3.0,
v2
of each star, and you determine the size of the orbits using the 6.8 km/s
relationship Distance = Speed × Time. In one orbital period, so star 2 is three times as massive as star 1. Algebra gets you
star 1 travels a distance of the rest of the way. Substituting m2 = 3 × m1 into the equation
m1 + m2 = 2.0 M⊙ gives you
( 20.4 ___
s ) × (2.63 years) × ( 3.156 × 10
km 7 s
____
year )
9
= 1.69 × 10 km. m1 + (3 × m1) = 4 × m1 = 2.0 M⊙,
This distance is 2π times the radius of the star’s orbit, so star 1 or m1 = 0.5 M⊙. Substituting m1 = 0.5 M⊙ into the equation
is following an orbit with a radius of 2.7 × 108 km, or 1.8 AU. m2 = 3 × m1 gives you m2 = 1.5 M⊙.
A similar analysis of star 2 shows that its orbit has a radius of Star 1 has a mass of 0.5 M⊙, and star 2 has a mass of 1.5 M⊙.
0.6 AU. You may tell your stars to step off the scale now.
20 Observations
of star 1 Masses are calculated from observations
of stars in a binary system.
Doppler velocity (km/s)
10 v1 P
A1 = = 1.8 AU ( AAU) 3
v1 = 20.4 2π m1 + m2 =
km/s v P ( Pyears) 2
v2 = 6.8 A2 = 2 = 0.6 AU
0 2π = 2 M£
km/s Orbits that
best fit the A = A1 + A2 = 2.4 AU
observations
–10 Observations
of star 2
m2 v1 m1 = 0.5 M£
m1 = v2 = 3 m2 = 1.5 M£
–20 Period = 2.63 years
0 0.5 1.0 1.5 2.0 2.5

Years
FIGURE 13.14 Doppler velocities of the stars in an eclipsing binary are used to measure the masses of the stars.
TA B L E 1 3 . 2
Taking the Measure of Stars*
Property Methods
Luminosity • For a star with a known distance, measure the brightness, then apply the
inverse square law of radiation:
Luminosity = 4π × Distance2 × Brightness
• For a star without a known distance, take a spectrum of the star to
determine its spectral and luminosity classes, plot them onto the H-R
diagram, and read the luminosity from the diagram.
Temperature • Measure the color index of the star using blue and visual filters. Use Wien’s
law to relate the color to a temperature.
• Take a spectrum of the star and estimate the temperature from its spectral
class by noting which spectral lines are present.
Distance • For relatively nearby stars (within a few hundred parsecs), measure the
parallax shift of the star over the course of the year.
• For more distant stars, fi nd the luminosity using the H-R diagram as noted
above, and then use the spectroscopic parallax method to relate luminosity,
distance, and brightness.
Size • For a few of the largest stars, measure the size directly or by the length of
eclipse in eclipsing binary stars.
• From the width of the star’s spectral lines, estimate the luminosity class
(supergiant, giant, or main-sequence).
• For a star with known luminosity and temperature, use the Stefan-
Boltzmann law to compute how large the star must be (the luminosity-
temperature-radius relationship).
Mass • Measure the motions of the stars in a binary system, and use these to
determine the orbits of the stars; then apply Newton’s form of Kepler’s laws.
• For nonbinary stars, use the mass-luminosity relation to estimate the mass.
Composition • Analyze the lines in the star’s spectrum to measure chemical composition.
*A brief summary of the methods used to measure basic properties of stars. Of the properties listed here, only
temperature, distance, and composition can be measured. Luminosity must be inferred from the H-R dia-
gram, and size and mass must be calculated. Other properties that can be measured include brightness, color
index, spectral type, and parallax shift.
properties of stars. Each scientist plotted the luminosities of single most used and useful
stars versus a particular measure of their surface tempera- diagram in astronomy. Every- The H-R diagram is
tures (such as color index or spectral type). The resulting thing about stars, from their the single most used
plot is referred to as the “Hertzsprung-Russell diagram” or formation to their old age and and useful diagram
simply the H-R diagram. The H-R diagram proved to be a eventual death, is discussed in astronomy.
key figure for understanding stars. using the H-R diagram.
Begin with the layout of the H-R diagram itself, shown
in Figure 13.15. Along the horizontal axis (the x-axis) we
plot the surface temperature of stars, but it is plotted back-
The H-R Diagram Is Worth Studying ward: temperature starts out hot on the left side of the
Before going further in the text, it would be very much diagram and decreases going to the right. Remember that.
worth your while to study the H-R diagram carefully until Hot blue stars are on the left side of the H-R diagram; cool
you understand what it can tell you. The H-R diagram is the red stars are on the right. This sequence is plotted loga-
Hot stars Cool stars

are blue. are red.
Hotter
10,000,000 107
FIGURE 13.15 The layout of the
Hertzsprung-Russell, or H-R, diagram
1,000,000 106 A cool star must 1,0 –10 used to plot the properties of stars.
be very large to 00
R More luminous stars are at the top of the
be so luminous. £
diagram. Hotter stars are on the left. Stars
100,000 105
of the same radius (R) lie along the dotted
10 lines moving from upper left to lower right.
0R
10,000 104 £
–5 Absolute magnitudes are discussed in Math
Tools 13.1 and Appendix 6.
Absolute visual magnitude

1,000 103 10
Luminosity (L£)
More luminous
£
100 102 All stars with a 0

radius of 1 R£ lie
along this line. 1R
£
10 101
0.1
1 1 R £
+5
£
Our Sun is at
0.1 10–1 0.0
1R
this point on
£
the diagram.
A hot star must

0.01 10–2 be very small to +10
0.0
01 be so dim.
R£
0.001 10–3 this case with each step along the axis
40,000 30,000 10,000 6,000 3,000 corresponding to a multiplicative factor
of 10 in the luminosity. To understand
H-R diagrams are sometimes
plotted with either spectral type…
why the plotting is done this way, recall
that the most luminous stars are 10 bil-
Spectral type
lion times more luminous than the least
O5 B0 B5 A0 F0 G0 K5 M5
luminous stars, yet all of these stars must
…or color index along the horizontal fit on the same plot.
axis in place of temperature.
It turns out that the H-R diagram
Color index tells us about more than just the surface
1.35 1.3 1.2 1.0 0.7 0.40 0.25 temperatures and luminosities of stars.
Earlier in the chapter we discovered that
if the temperature and the luminosity of
rithmically, which is to say that a step along the axis from a star are known, we can calculate its radius. Because each
a point representing a star with a surface temperature of point in the H-R diagram is specified by a surface tempera-
30,000 K to one with a surface temperature of 10,000 K ture and a luminosity, we can use the luminosity-temper-
(that is, a temperature change by a factor of 3) is the same ature-radius relationship to fi nd the size of a star at that
as a step between points representing a star with a tem- point as well. We can think through how this works using
perature of 9,000 K and a star with a temperature of 3,000 what we know about radiation. A star in the upper right
K (also a factor-of-3 change). The temperature axis can corner of the H-R diagram is very cool, which, according to
also be labeled with the spectral types or color indices of the Stefan-Boltzmann law, means that each square meter of
stars. Hot blue O stars with large color indices are on the its surface is radiating only a small amount of energy. But
left side of the diagram; cool red M stars with small color at the same time, this star is extremely luminous. Such a
indices are on the right. star must be huge to account for its high overall luminosity
Along the vertical axis (the y-axis) we plot the luminos- despite the feeble radiation coming from each square meter
ity of stars—the total amount of energy that a star radiates of its surface. Conversely, a star in the lower left corner
each second. This time the plot is the “right way” around: of the H-R diagram is very hot, which means that a large
more luminous stars are toward the top of the diagram, and amount of energy is coming from each square meter of its
less luminous stars are toward the bottom. As with the tem- surface. However, this star has a very low overall luminos-
perature axis, luminosities are plotted logarithmically, in ity. The conclusion is that its surface area cannot be very
large. Stars in the lower left corner of the H-R diagram are
small. AstroTour: H-R Diagram The Main Sequence Is a Grand
This result persists everywhere in the H-R diagram. Pattern in Stellar Properties
Moving up and to the right takes you to larger and larger
stars. Moving down and to the left takes you to smaller and Figure 13.16a shows the H-R diagram for 16,600 nearby
smaller stars. All stars of the same radius lie along lines stars based on observations obtained by the Hipparcos
running across the H-R diagram from the upper left to the satellite. A quick look at this
lower right, as shown in Figure 13.15. diagram immediately reveals Most stars lie along
Like an experienced surveyor who “sees” the lay of the a remarkable fact. Instead of the main sequence of
land when looking at a topographic map, or a classical musi- being strewn all about the the H-R diagram.
cian who “hears” the rhythms and harmonies when look- diagram, we fi nd instead that
ing at a piece of sheet music, you should get to the point about 90 percent of the stars in the sky lie along a well-
where you automatically “see” the properties of a star—its defi ned sequence running across the H-R diagram from
temperature, color, size, and luminosity—from a glance at lower right to upper left. This sequence of stars is called
its position on the H-R diagram. the main sequence. On the left end of the main sequence
FIGURE 13.16 (a) An H-R diagram for 16,600 stars obtained by the Hipparcos satellite. Most of the stars
lie in a band running from the upper left of the diagram toward the lower right, called the main sequence. (Dot
color represents number of stars, with blue indicating the fewest and red the most.) (b) An H-R diagram for two
different samples of stars. The red symbols show the H-R diagram for 46 stars that are especially close to the
Sun. The blue symbols show the 97 brightest stars in the sky. Note that because these are observational H-R
diagrams, they are plotted against an observed quantity: their spectral type.
(a) (b)
106 –10 106
105 105
104 The majority of stars fall 104

along the main sequence –5
of the H-R diagram.
103 103
102 102
Visual luminosity relative to Sun
Visual luminosity relative to Sun
0
MA
1 1 I N
10 10 SE
Q UE
M The Sun is less luminous NC
1 AI £ 1 than most of the brightest £ E
N +5
stars we see at night… Sun
Sun S
10–1 EQ 10–1
UE
NC
E
10–2 +10 10–2
…but more luminous than
10–3 10–3 most of its neighbors.
10–4 +15 10–4
10–5 10–5
10–6 10–6
O5 B5 A5 F5 G5 K5 M5 O5 B5 A5 F5 G5 K5 M5
Spectral type Spectral type
are the O stars: hotter, larger, and more luminous than High-mass main- Mass determines the
the Sun. On the right end of the main sequence are sequence stars are location of a star along
the M stars: cooler, smaller, and fainter than the Sun. 107 hot and luminous. the main sequence.
If you know where a star lies on the main sequence,
then you know its approximate luminosity, surface 1,0
106 00 –10
temperature, and size. R
In
£
cr
M
From a practical standpoint, the main sequence
ea
AI
105 SE
si
60 M£ g
n
is extremely handy. For example, you may have won- Q m 10
0R
20 M UE as
dered in our discussion of stellar parallax how we 4
£
NC s £
10 –5
measure the distances to stars that are farther away

E
than a few hundred light-years. The main sequence 10 M£
10
103 R
comes to the rescue. We can determine whether a star
Luminosity (L£)
£
is a main-sequence star by looking at the absorption 5 M £

lines in its spectrum. If a star is on the main sequence, 102 0
1R
then by virtue of its location on the H-R diagram we £
know its luminosity. If we know its luminosity and 101

3 M£
we measure its brightness, we can use the inverse 0.1
R 2 M£
square law of radiation to fi nd its distance. (How far 1 £ £ +5
1.5 M£
away must a star of that luminosity be to have the
brightness we measure?) This method of determin- 0.0 1 M£
10–1 1R
ing distances to stars is called spectroscopic paral- £ Low-mass main 0.8 M£
lax, and is discussed further in Appendix 6. Parallax sequence stars are 0.6 M£
10–2 0 cool and dim. +10
and spectroscopic parallax are the fi rst two steps . 0 01 0.2 M£
R
along a chain of reasoning that will let us build our £
10–3
knowledge of distances all the way to the edge of the 40,000 30,000 10,000 6,000 3,000
observable universe.
Figure 13.16b makes another important point for
astronomy, as well as for all other sciences. The red
O5 B0 B5 A0 F0 G0 K5 M5
symbols represent 46 of the nearest stars to Earth; the
blue symbols represent the 97 brightest stars as seen Spectral type
in our sky. There are many more cool, low-luminosity FIGURE 13.17 The main sequence of the H-R diagram is a sequence
stars in the sky than there are hot, high-luminosity of masses.
stars. If we restrict our attention to stars that are near
the Sun, these are all that we see. Yet if we look at the
brightest stars in the sky, we get a very different pic- main sequence are high-mass stars. Stated another way,
ture. Even though these stars are farther away, they are so if we know the mass of a main-sequence star, then we
luminous that they dominate what we see. Think about how know where on the main sequence the star is, which
different your impression of the properties of stars would means we know the star’s approximate temperature, size,
be if all you knew about were the nearest stars, or if all you and luminosity. If a main-sequence star is less massive
knew about were the brightest stars. Neither of these groups than the Sun, it will be
alone accurately represents what stars as a whole are like, smaller, cooler, redder, and Low-mass main-
but the nearest stars provide a much fairer sample because less luminous than the Sun; sequence stars are
luminous stars, which dominate the brightest-stars group, it will be located to the lower faint and cool.
are rare. Whether you are an astronomer studying the prop- right of the Sun on the main High-mass main-
erties of stars or a political pollster measuring the sense sequence. On the other hand, sequence stars are
of the people, the validity of your results depends on the if a star is more massive than hot and luminous.
care with which you choose the sample (of stars or people) the Sun, it will be larger, hot-
that you study. If astronomers were conducting a poll, they ter, bluer, and more luminous than the Sun; it will be
would be much better off interviewing the nearest stars located to the upper left of the Sun on the main sequence.
rather than the brightest. The mass of a star determines where on the main sequence
When we add mass to the picture, the main sequence the star will lie.
of the H-R diagram gets even more interesting. Stellar To see the relationship between a star’s mass and its
mass increases smoothly as we go from the lower right to position on the main sequence directly, look at Table 13.3
the upper left along the main sequence. The faint, cool and Figure 13.17, which show the properties of stars along
stars on the right-hand side of the main sequence have the main sequence. If, for example, a main-sequence star
low masses; the luminous, hot stars on the left side of the has a mass of 18 M⊙, it will be a B0 star. It will have a sur-
TA B L E 1 3 . 3
The Main Sequence of Stars*
Spectral Type Color Index† Temperature (K) Mass (M⊙) Radius (R ⊙) Luminosity (L ⊙)
O5 1.36 44,500 60 17.8 794,000
B0 1.32 30,000 18 9.3 52,500
B5 1.17 15,400 5.9 3.8 832
A0 1.02 9,520 2.9 2.5 54
A5 0.87 8,200 2.0 1.74 14
F0 0.76 7,200 1.6 1.35 6.5
F5 0.67 6,440 1.3 1.2 3.2
G0 0.59 6,030 1.05 1.05 1.5
G2 (Sun) 0.56 5,860 1.00 1.00 1.0
G5 0.53 5,770 0.92 0.93 0.8
K0 0.47 5,250 0.79 0.85 0.4
K5 0.35 4,350 0.67 0.74 0.15
M0 0.27 3,850 0.51 0.63 0.08
M5 0.22 3,240 0.21 0.32 0.011
M8 0.19 2,640 0.06 0.13 0.0012
*Approximately 90 percent of the stars in the sky fall on the main sequence of the H-R diagram. This table
gives the properties of stars along the main sequence.
†
Normally, astronomers refer to the color index (or blue/visual brightness ratio) of a star as the difference
between the B and V magnitudes of the star. (See Math Tools 13.1 and Appendix 6 for more information on cal-
culating the color index.)
More massive main-sequence

stars are more luminous… …larger… …and hotter.
106 100 105
104
Temperature (K)
Luminosity (L £)
Radius (R £)
10
102
104
1 £ £
1 £
10–2
10–4 0.1 103

0.01 0.1 1 10 100 0.01 0.1 1 10 100 0.01 0.1 1 10 100
Mass (M£) Mass (M£) Mass (M£)
FIGURE 13.18 Plots of luminosity, radius, and temperature versus mass for stars along the main sequence.
The mass (and chemical composition) of a main-sequence star determines all of its other properties.
face temperature of 30,000 K, a radius of over 9 R⊙, and a How can astronomers tell which stars are not members
luminosity about 50,000 times that of the Sun. If a main- of the main sequence? Stars that lie off the main sequence
sequence star instead has a mass of 0.21 M⊙, it will be an on the H-R diagram can be identified by their luminosi-
M5 star. It will have a surface temperature of 3,240 K, a ties (determined by their distance), or by slight differences
radius of about 0.32 R⊙, and a luminosity of about 0.01 L ⊙. within their spectral type. The width of a star’s spectral
A main-sequence star with a mass of 1 M⊙ will be a G2 star lines is an indicator of the density and “surface pressure” of
like the Sun and will have the same surface temperature, gas in a star’s atmosphere. Those puffed-up stars above the
size, and luminosity as the Sun. main sequence have atmospheres with lower density and
It is difficult to overemphasize the importance of this lower surface pressure compared to main-sequence stars.
result, so we state it again for clarity. For stars of similar It is important for astronomers to know where stars
chemical composition, the fall on the H-R diagram. For example, when using the H-R
mass of a main-sequence star The mass of a diagram to estimate the distance to a star by the spectro-
alone determines all of its main-sequence star scopic parallax method, they must know whether the star
other characteristics. Figure determines what its is on, is above, or falls below the main sequence in order
13.18 illustrates this point. fate will be. to get the star’s luminosity. Stars both on and off the main
Knowing the mass (and chem- sequence have a property called luminosity class, which
ical composition) of a main-sequence star tells us how large tells us the size of the star. Luminosity classes are defi ned
it is, what its surface temperature is, how bright it is, what as follows: supergiant stars are luminosity class I, bright
its internal structure is, how long it will live, how it will giants are class II, giants are class III, subgiants are class
evolve, and what its fi nal fate will be! IV, main-sequence stars are class V, and white dwarfs are
Upon reflection, this relationship is both sensible and class WD. Luminosity classes I though IV lie above the main
possibly the most important and fundamental relationship sequence; class WD falls below and to the left of the main
in all of astrophysics. If you have a certain amount and type sequence (Figure 13.19). Thus, the complete spectral clas-
of material to make a star, there is only one kind of star you sification of a star includes both its spectral type, which
can make. As we go on to discuss the physical processes tells us temperature and the color; and its luminosity class,
that give a star its structure, this statement will make even which indicates size.
more sense. We will fi nd that a star is a “battle” between Another point to consider is that some non-main-
gravity (which is trying to pull the star together) and the sequence stars vary in luminosity, and through the lumi-
energy released by nuclear reactions in the interior of
the star (which is trying to blow it apart). The mass of
the star determines the strength of its gravity, which FIGURE 13.19 Stellar luminosity classes.
in turn determines how much energy must be gener-
107
ated in its interior to prevent it from collapsing under
its own weight. The mass of a star determines where
106 –10
the balance is struck.
105 Ia
Not All Stars Are Main- Supergiants
104 –5
Sequence Stars Ib

Before leaving our discussion of the observed properties 103 II
Luminosity (L£)
of stars, we need to point out that although most stars III

in the sky are main-sequence stars, some are not. Some 102 0
stars are found in the upper right portion of the H-R dia- Giants
gram, well above the main sequence. From their posi-
101
tion we know that they must be bloated, luminous, cool Subgiants IV
giants, with radii hundreds or thousands of times the
1 £ +5
radius of the Sun. If the Sun were such a star, its atmo-
sphere would swallow the orbits of the inner planets, MAIN SEQUENCE
including Earth. At the other extreme are stars found 10–1
in the far lower left corner of the H-R diagram. These
stars must be tiny, with sizes comparable to the size of 10–2 +10
Earth. Their small surface areas explain why they have White dwarfs (WD)
such low luminosities despite having temperatures that 10–3
can rival or even exceed the surface temperature of the O5 B5 A5 F5 G5 K5 M5
hottest main-sequence O stars. Spectral type
nosity-temperature-radius relationship we see that their In the decades that followed the discovery of the main
temperature and size (radius) must be changing as well. sequence, few problems in astronomy attracted more atten-
We will return to these stars in later chapters as we look at tion than these questions. Their answers turned out to be
stellar evolution. as fundamental as stellar pioneers Annie Jump Cannon,
The existence of the main sequence, together with the Ejnar Hertzsprung, and Henry Norris Russell could ever
fact that the mass of a main-sequence star determines where have hoped. The existence of the main sequence holds the
on the main sequence it will lie, is a grand pattern that essential clue to what stars are and how they work. The
points to the possibility of a deep understanding of what properties of stars that are not on the main sequence point
stars are and what makes them tick. By the same token, to an understanding of how stars form, how they evolve,
the existence of stars that do not follow this grand pat- and how they die. Much of the rest of the text discussing
tern raises yet more questions. What is it about a star that stars will be spent on the lessons learned from patterns in
determines whether or not it is part of the main sequence? the H-R diagram.
in terrestrial laboratories. We might have found that

Seeing the Forest for the Trees the temperatures of stars inferred from their absorption
spectra disagreed with the temperatures of stars inferred
When you walk down a path, each step you take covers from their peaks in blackbody emission. We might have
only a short distance. But by persistently putting one foot found that measurements of the sizes of stars based on
in front of the other, you find after a time that you have the Stefan-Boltzmann law disagreed with measurements
gotten somewhere. The same is true of the road we trav- of the sizes of stars in eclipsing binaries. We might have
eled in this chapter. Each step along the way was relatively found that the motions of stars in binary systems failed
small and understandable—involving the application of a to follow the predictions of Newton’s physics. We might
physical principle that we see at work in the world around have found any or all of these things, but we did not! The
us, or the use of a tool from algebra or geometry. But when successes of this chapter give us strong reason to believe
we compare our understanding of stars reached by the that the same physical laws at work right here on Earth
end of the chapter with the understanding we started out and in our Solar System also describe the fundamental
with, it is amazing how far we have come. character and behavior of matter and energy throughout
Our path in this chapter followed the course of the tri- the rest of the universe.
umph of our physical understanding of the universe. The With an understanding of the basic physical prop-
properties of electromagnetic radiation, the structure of erties of stars in place, we are now ready to ask much
atoms and molecules, the gravitational attraction between more fundamental questions about stars. We are ready,
masses— all of these and more came into play as we built figuratively speaking, to “lift the hood” and see what lies
up our understanding of the physical nature of stars one within. How do stars work? How do they form? How do
piece at a time. We saw many occasions in this chapter they evolve? How do they die? We will begin to address
when the cosmological principle might have collapsed. these questions by investigating the star that serves as
We might have seen emission and absorption lines in the the standard by which we measure other stars: the star
spectra in stars that were different from those measured we know best, our Sun.
Summary 4. What would happen to our ability to measure stellar par-

allax if we were on the planet Mars? What about Venus
or Jupiter?
• The distances to nearby stars are measured stereoscopi-
cally by their parallax. 5. To know certain properties of a star, you must fi rst deter-
mine the star’s distance. For other properties, knowledge
• The nearest star (other than the Sun) is about 4 light-
of distance is not necessary. Into which of these two cat-
years away.
egories would you place each of the following properties:
• When stars are too remote for trigonometric parallax size, mass, temperature, color, spectral type, and chemi-
measurements, we must use their spectra and the H-R cal composition? In each case, state your reason(s).
diagram to estimate their distances.
*6. The light from stars passes through dust in our galaxy
• Radiation tells us the temperature, size, and composi- before it reaches us, making stars appear dimmer than
tion of stars. they actually are. How does this phenomenon affect
stellar trigonometric parallax? How does it affect spec-
• The color index of a star reveals its surface
troscopic parallax?
temperature.
7. Compare the temperature, luminosity, and radius of stars
• A star’s luminosity class and temperature tells us its
at the lower left and upper right of the H-R diagram.
size.
8. Albireo, a star in the constellation of Cygnus, is a visual
• Small, cool stars greatly outnumber large, hot stars.
binary system whose two components can be seen easily
• Binary stellar systems enable us to measure the masses with even a small amateur telescope. Viewers describe
of stars. the brighter star as “golden” and the fainter one as “sap-
phire blue.”
• The H-R diagram is the single most useful diagram in
a. What does this description tell you about the relative
astronomy, plotting stellar evolution and showing the
temperatures of the two stars?
relationship among the various physical properties of
b. What does it tell you about their respective sizes?
stars.
9. Very cool stars have temperatures around 2,500 K and
• The mass and composition of a main-sequence star deter-
emit Planck spectra with peak wavelengths in the red
mine its luminosity, temperature, and size.
part of the spectrum. Explain whether or not these stars
emit any blue light.
*10. It is possible for two stars to have different temperatures
SmartWork, Norton’s online homework system, and different sizes but the same luminosity. Explain
includes algorithmically generated versions of these how.
11. The stars Betelgeuse and Rigel are both in the constel-
lation Orion. Betelgeuse appears red in color and Rigel
is bluish white. To the eye, the two stars appear equally
bright. If you can compare the temperature, luminosity,
or size from just this information, do so. If not, explain
Student Questions why.
12. Capella (in the constellation Auriga) is the sixth bright-
est star in the sky. With a high-power telescope, Capella
1. Distances to stars can be measured in inches, miles, kilo- is actually two pairs of binary stars: the fi rst pair are
meters (km), astronomical units (AU), light-years, and G-type giants; the second pair are M-type main-sequence
parsecs (pc). Why do many stellar astronomers prefer stars. From this information, what color do you expect
to use parsecs? Capella to appear as, and why?
2. Even our best measurements always have experimen- 13. Two stars have the same luminosity but one is larger.
tal uncertainties. Using the measurement uncertainties Compare their temperatures. Now suppose that the two
given in the book, how accurately can we know the dis- stars have the same size but one is more luminous.
tance to a star using trigonometric parallax? Again, compare their temperatures.
3. The distances of nearby stars are determined by their *14. In this chapter, we learned that the spectra of stars
parallaxes. Why is the uncertainty in our knowledge of contain many absorption features, yet we treat them as
a star’s distance greater for stars that are farther from blackbodies when discussing their properties. Why is
Earth? this approach justified?
15. Explain why the stellar spectral types (O, B, A, F, G, K, M) Because the circumference of a circle is 2πr, there must
are not in alphabetical order. Also explain the sequence be 2πr/r or 2π radians in a circle.
of temperatures defi ned by these spectral types. a. How many degrees are there in 1 radian?
b. How many arcseconds are there in 1 radian? How
*16. Other than the Sun, the only stars whose mass we can
does this value compare to the number of astronomi-
measure directly are those in eclipsing or visual binary
cal units in 1 pc (206,265) that we cite in the text?
systems. Explain why.
c. What angle would a round object that has a diameter
17. In Math Tools 13.2, we show how one can calculate the of 1 AU make in our sky if we see it at a distance of 1
masses of both stars in an eclipsing binary system using pc? A distance of 10 pc?
their orbital details. Implicit in these calculations was d. What do your answers to parts (b) and (c) of this ques-
the assumption that the binary system is seen edge-on. tion tell you about how the actual size of an object
What would happen to our fi nal masses if the system is related to its angular size measured in units of
were tilted (or “inclined”) instead? radians?
18. Once we determine the mass of a certain spectral type 27. Our eyes are typically 6 cm apart. Suppose your eye
of star that is located in a binary system, we assume separation is average and you see an object jump from
that all other stars of the same spectral type and lumi- side to side by half a degree as you blink back and forth
nosity class also have the same mass. Why is this a safe between your eyes. How far away is that object?
assumption?
28. Sirius, the brightest star in the sky, has a parallax of 0.379
*19. How do we estimate the mass of stars that are not in arcsec. What is its distance in parsecs? In light-years?
eclipsing or visual binary systems?
**29. Sirius is actually a binary pair of two A-type stars. The
20. Although we tend to think of our Sun as an “average” brighter of the two stars is called the “Dog Star,” and
main-sequence star, it is actually hotter and more lumi- the fainter is called the “Pup Star” because Sirius is in
nous than average. Explain. the constellation Canis Major (meaning “Big Dog”). The
Pup Star appears about 6,800 times fainter than the Dog
21. Many times throughout this chapter, we have com-
Star. Compare the temperatures, luminosities, and sizes
mented that there are many more low-mass stars than
of both stars.
high-mass stars. Why do you think this is so? (We will
address the reasons in chapters to come). 30. Sirius and its companion orbit around a common center
of mass with a period of 50 years. The mass of Sirius is
22. Star masses range from 0.08 to about 100 M⊙.
2.35 times the mass of the Sun.
a. Why are there no stars with masses less than 0.08
a. If the orbital velocity of the companion is 2.35 times
M⊙?
greater than that of Sirius, what is the mass of the
b. Why are there no stars with masses much greater than
companion?
100 M⊙?
b. What is the semimajor axis of the orbit?
23. Very old stars often have very few heavy elements,
31. Sirius is 22 times more luminous than the Sun, and
while very young stars have much more. What does this
Polaris (the “North Pole Star”) is 2,350 times more lumi-
difference imply about the chemical evolution of the
nous than the Sun. Sirius appears 23 times brighter than
Universe?
Polaris. How much farther away from us is Polaris than
**24. Explain whether we would still be able to identify the Sirius? What is the distance of Polaris in light-years?
spectral types of stars if there were no elements other
32. Proxima Centauri, the star nearest to Earth other than
than hydrogen and helium in their atmospheres.
the Sun, has a parallax of 0.772 arcsec. How long does
25. Logarithmic (log) plots show major steps along an axis it take light to reach us from Proxima Centauri?
scaled to represent equal factors, most often factors of
*33. Betelgeuse (in Orion) has a parallax of 0.00763 ± 0.00164
10. Why, in astronomy, do we sometimes use a log plot
arcsec, as measured by the Hipparcos satellite. What is
instead of the more conventional linear plot?
the distance to Betelgeuse, and the uncertainty in that
measurement?
34. Rigel (also in Orion) has a Hipparcos parallax of 0.00412
arcsec. Given that the two stars appear equally bright in
26. Sketch a circle and mark an arc along the circumference the sky, which star is actually more luminous? Know-
equal to the radius of the circle. The size of the angle sub- ing that Betelgeuse appears reddish while Rigel appears
tended by the arc, measured in radians, is given by the bluish white, as stated in question 11, which star would
length of the arc divided by the radius (r) of the circle. you say is larger and why?
35. The Sun is about 16 trillion (1.6 × 1013) times brighter of the spectrum does this wavelength fall? What color
than the faintest stars visible to the naked eye. will this star appear to us?
a. How far away (in astronomical units) would an iden-
38. About 1,470 watts (W) of solar energy hits each square
tical solar-type star be if it were just barely visible to
meter of the Earth’s surface. Use this and our distance
the naked eye?
to the Sun to calculate its luminosity.
b. What would be its distance in light-years?
*36. Our galaxy (the Milky Way) contains over 100 billion
stars. If you assume that the average density of stars is
the same as in the solar neighborhood, how much vol-
ume does the Milky Way take up? If the galaxy were a
sphere, what would its radius be?
37. Find the peak wavelength of blackbody emission for a quizzes, plus links to premium content in SmartWork and
star with a temperature of about 10,000 K. In what region the ebook. Visit www.wwnorton.com/studyspace.
It is stern work, it is perilous work to thrust your hand in the sun
And pull out a spark of immortal flame to warm the hearts of men.
JOYCE K ILMER (1886 –1918)
Sunset in New York City.

CHAP TER 14
A Run-of-the-Mill G-Type
Star—Our Sun
14.1 The Sun Is More Than K E Y CONCEP T S

Just a Light in the Sky To most humans the Sun is the most important object
in the heavens. It lights our days, warms our planet, and
How wonderful, after a long cold night, to see the light and provides the energy for life. But to astronomers the Sun
feel the warmth of the rays of the Sun. Energy from the Sun is a typical main-sequence star, located conveniently
is responsible for daylight, for our weather and seasons, and nearby for detailed study. In this chapter, as we take a
for terrestrial life itself. No object in nature has been more closer look at our local star, we will learn about
revered or more worshipped than the Sun. In fact, the mod-
• The balances between pressure and gravity, and
ern symbol for the Sun, ⊙, is nothing other than the ancient
Egyptian hieroglyph for the Sun god Ra. Yet although the between energy generation and loss, that determine
Sun may have dominated human consciousness since the the structure of the Sun.
dawn of our species, the discussion in Chapter 13 offers • Fusion of hydrogen to helium, and how mass is
a very different perspective. To an astronomer of the 21st efficiently converted into energy in the Sun’s core.
century, the Sun is the prototype for main-sequence stars.
• The different ways that energy moves outward from
With a middle-of-the-road spectral type of G2, the Sun is
the Sun’s core toward its surface.
all but indistinguishable from billions of other stars in our
galaxy. It is also the star against which all other stars are • Physical models of the Sun’s interior, and how
measured. The mass of the Sun, the size of the Sun, the observations of solar neutrinos and seismic vibrations
luminosity of the Sun—these basic properties provide the on the surface of the Sun test these models.
yardsticks of modern astronomy.
• The structure of the Sun’s atmosphere, from its
The Sun may be run-of-the-mill as far as stars go, but
5,780-kelvin photosphere to its 1-million-kelvin corona.
that makes it no less awesome an object on a human scale.
The mass of the Sun, 1.99 × 1030 kilograms (kg), is over • The solar wind streaming away from the Sun.
300,000 times that of Earth. The Sun’s radius, 696,000 kilo- • Sunspots, flares, coronal mass ejections, and other
meters (km), is over 100 times that of Earth. At a luminos- consequences of magnetic activity on the Sun.
ity of 3.85 × 1026 watts (W), the Sun produces more energy
• Solar-activity cycles of 11 and 22 years.
in a second than all of the electric power plants on Earth
could generate in a half-million years. The Sun is also the • How solar activity affects Earth.
only star we can study at close range. Much of the detailed
information that we know about stars has come only from
studying our local star.
389
390 Chapter 14 A Run-of-the-Mill G-Type Star— Our Sun
In the previous leg of our journey, we looked at the gross

physical properties of distant stars, including their mass,
luminosity, size, temperature, and chemical composition. At each point within
Now, as we turn our attention toward our own local star, the Sun, the outward
push of pressure…
we face more fundamental questions: How does the Sun
work? Where does it get its energy? Why does it have the
size, temperature, and luminosity that it does? How has it …is balanced by the
inward pull of gravity.
been able to remain so constant over the billions of years
since the Solar System formed? In short, we now confront
the question, What is a star? Core
14.2 The Structure of

the Sun Is a Matter The energy radiated
from the surface of the
Sun balances the energy
of Balance produced in its interior.
In Chapter 7 we tackled the question of how we know about FIGURE 14 .1 The structure of the Sun is determined by the
the interior of Earth even though no machine has ever done balance between the forces of pressure and gravity, and the
more than scratch the surface of our planet. The answer was balance between the energy generated in its core and energy
a combination of physical understanding, detailed computer radiated from its surface.
models, and clever experiments that test the predictions of
those models. The task of exploring the interior of the Sun
is much the same. As with Earth, the structure of the Sun
is governed by a number of physical processes and relation- that the pressure at any point within the Sun’s interior must
ships. Using our understanding of physics, chemistry, and be just enough to hold up the weight of all the layers above
the properties of matter and radiation, we express these pro- that point. If the Sun were not in hydrostatic equilibrium,
cesses and relationships as mathematical equations. High- then forces within the Sun would not be in balance, so the
speed computers are then used to solve these equations and surface of the Sun would move. The Sun today is the same
arrive at a model of the Sun. One of the great successes of as it was yesterday and the day before. That is all the obser-
20th century astronomy was the successful construction vation we require to infer that the interior of the Sun is in
of a physical model of the Sun that agrees with our obser- hydrostatic equilibrium.
vations of the mass, composition, size, temperature, and Hydrostatic equilibrium becomes an even more powerful
luminosity of the real thing. concept when combined with what we know about the way
Our current model of the interior of the Sun is the cul- gases behave. As we move deeper into the interior of the Sun,
mination of decades of work by thousands of physicists and the weight of the material above us becomes greater, and
astronomers. Understanding the details that lie within this hence the pressure must increase. As we learned in Founda-
model is the work of lifetimes. Even so, the essential ideas tions 8.1, in a gas higher pressure means higher density and/
underlying our understanding of the structure of the Sun or higher temperature. Figure 14.2a shows how conditions
are found in a few key insights. In turn, these insights can vary as distance from the center of the Sun changes. As we
be summed up in a single statement: The structure of the go deeper into the Sun, the pressure climbs; and as it does,
Sun is a matter of balance. the density and temperature of the gas climb as well.
The fi rst key balance within the Sun is the balance A second fundamental balance within the Sun is a bal-
between pressure and gravity illustrated in Figure 14.1. The ance of energy (see Figure 14.1). Stars like the Sun are
Sun is a huge ball of hot gas. remarkably stable objects.
If gravity were stronger than The Sun is in Geological records show that Solar energy
pressure within the Sun, the hydrostatic the luminosity of the Sun has production must
Sun would collapse. Con- equilibrium. remained nearly constant for balance what is
versely, if pressure were stron- billions of years.1 In fact, the radiated away.
ger than gravity, the Sun would blow itself apart. We saw very existence of the main
this balance in Chapter 7 and gave it a name: hydrostatic sequence says that stars do not change much over the main
equilibrium. Hydrostatic equilibrium sets the pressure at
each point within a planet and determines the atmospheric 1
Astrophysicists believe that the Sun increases its luminosity about
pressure at Earth’s surface. Hydrostatic equilibrium says 5 percent every billion years.
1 The same data are 2 Plotting using powers 3 …while a linear plot shows
plotted two ways. of 10 shows the full how quantities rise sharply
range of the data… at the center of the Sun.
(a) Structure
250
Atmospheres 1011 200
109 150
105
107 Pressure 100 Pressure
107 (atmospheres) (billions of atmospheres)
50
105
0
kg/m3
109
200
2
10 1
105
10 150
103
103
1011 100
Energy- 104 10 Density Density
(kg/m3) (thousands of kg/m3)
5
producing core 10 50
–1
10
12
8 0
2
10 4 20
10
0
2 1010
–2 1 Temperature 15 Temperature
10 108 (K) (millions of K)
10
–4
106 K 10
106
5
104
0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Fraction of radius (R/R£) Fraction of radius (R/R£)
watts/m3
(b) Energy generation
300
102 250
100 200
Energy produced Energy produced
FIGURE 14 . 2 A cutaway figure showing the interior 150
10–2 (watts/m3) 100 (watts/m3)
structure of the Sun. (a) Temperature, density, and
10–4 50
pressure increase toward the center of the Sun. (b) Energy 0
is generated in the Sun’s core. 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Fraction of radius (R/R£) Fraction of radius (R/R£)
part of their lives. To remain in balance, the Sun must pro- consist of two protons and two neutrons. Most carbon nuclei
duce just enough energy in its interior each second to replace consist of six protons and six neutrons. Protons have a posi-
the energy that is radiated away by its surface each second. tive electric charge, and neu-
This is a new type of balance, one we have not dealt with trons have no net electric Atomic nuclei are held
before. Understanding the balance of energy within the Sun charge. Like charges repel, so together by the strong
requires thinking about how energy is generated in the inte- all of the protons in an atomic nuclear force.
rior of the Sun (Figure 14.2b), and how that energy fi nds its nucleus must be pushing away
way from the interior to the Sun’s surface, where it is radi- from each other with a tremendous force because they are
ated away. so close to each other. If electric forces were all there was
to it, the nuclei of atoms would rapidly fly apart—yet atomic
nuclei do hold together. We conclude that there must be
another force in nature, even stronger than the electric force,
The Sun Is Powered by Nuclear Fusion that “glues” the protons and neutrons in a nucleus together.
One of the most basic questions facing the pioneers of stellar That force, which acts only over extremely short distances,
astrophysics was where the Sun and stars get their energy. is called the strong nuclear force.
The answer to this question came not from astronomers’ The strong nuclear force is indeed a very powerful force.
telescopes, but from theoretical work and the laboratories It would take an enormous amount of energy to pull apart
of nuclear physicists. At the heart of the Sun lies a nuclear the nucleus of an atom such as helium into its constitu-
furnace capable of powering the star for billions of years. ent parts. The reverse of this process says that when you
The nucleus of most hydrogen atoms consists of a single assemble an atomic nucleus from its component parts, this
proton. Nuclei of all other atoms are built from a mixture same enormous amount of energy is released. The process
of protons and neutrons. Most helium nuclei, for example, of combining two less massive atomic nuclei into a single
more massive atomic nucleus is referred to as nuclear fusion. Whether a ball rolling downhill, a battery discharg-
Many nuclear processes are possible, and as we continue ing itself through a lightbulb, or an atom falling to a lower
our study of stars we will fi nd that a wide range of nuclear state by emitting a photon, most systems in nature tend to
reactions can occur in stars. In the Sun, as in all other main- seek the lowest-energy state available to them. Going from
sequence stars, the only significant process going on is the hydrogen to helium is a big ride downhill in energy, so we
fusion of hydrogen to form helium—a process often referred might imagine that hydrogen nuclei would naturally tend to
to as hydrogen burning (even though it has nothing to do fuse together to make helium. Fortunately for us, however,
with fi re in the usual sense of the word). a major roadblock stands in the way of nuclear fusion.2 The
Main-sequence stars get their energy by fusing hydrogen strong nuclear force responsible for binding atomic nuclei
atoms together to make helium—that is, by nuclear fusion. together can act only over very short distances—10 –15 meter
To judge the effectiveness of or so, or about a hundred-thousandth the size of an atom. To
this reaction, we can make Nuclear fusion is a get atomic nuclei to fuse, they must be brought close enough
use of one of the key results of very efficient source to each other for the strong nuclear force to assert itself, but
special relativity (which we of energy. this is hard to do. All atomic nuclei have positive electric
will discuss further in Chap- charges, which means that any two nuclei will repel each
ter 17): the equivalence between mass and energy. Mass can other. This electric repulsion, illustrated in Figure 14.3,
be converted to energy and energy can be converted to mass, serves as a barrier against nuclear fusion. Fusion cannot take
with Einstein’s famous equation E = mc2 providing the place unless this electric barrier is somehow overcome.
exchange rate between the two. Comparing the mass of the Energy in the Sun is produced in its innermost region,
products of a reaction with the mass of the reactants tells called the Sun’s core, as shown in Figure 14.2b. Conditions
us the fraction of the original mass that was turned into in the core are extreme. Matter at the center of the Sun has
energy in the process. The mass of four separate hydrogen
atoms is 1.007 times greater than the mass of a single helium
2
atom; so when hydrogen fuses to make helium, 0.7 percent One of today’s engineering goals is the development of controlled
of the mass of the hydrogen is converted to energy (see Math nuclear fusion. Such an achievement could produce a limitless, clean
Tools 14.1). source of energy.
M AT H T O O L S 1 4 .1
The Source of the Sun’s Energy

We have seen that the conversion of four hydrogen nuclei (pro- lion metric tons of matter converted to energy in the process),
tons) into a single helium nucleus results in a loss of mass. The and you have our Sun. The sunlight falling on Earth may be
mass of a single proton is 1.6726 × 10 –27 kg. So, four protons responsible for powering almost everything that happens on
add up to a mass of 6.6904 × 10 –27 kg. But the mass of a helium our planet, but it amounts to only about a hundred-billionth of
nucleus is 6.6465 × 10 –27 kg. Doing the subtraction, we see that the energy radiated by our local star.
this is 0.0439 × 10 –27 kg (or 4.39 × 10 –29 kg) less than the mass The Sun has been burning hydrogen at this prodigious
of the four protons— a loss of about 0.7 percent. Conversion of rate for 4.6 billion years, so how much of its available fuel has
0.7 percent of the mass of the hydrogen into energy might not already been spent? We estimate that the amount of hydrogen
seem very efficient—until we compare it with other sources of available for fusion in the young Sun’s core was about 10 per-
energy and discover that it is millions of times more efficient cent of the total solar mass, or 2 × 1029 kg. At a rate of 600 mil-
than even the most efficient chemical reactions. lion metric tons per second, the Sun consumes 6 × 1011 kg per
Using Einstein’s equation E = mc2, we can see that the second, multiplied by 3.16 × 107 seconds per year, or 1.90 ×
energy released by this mass-to-energy conversion is 1019 kg per year. Over the Sun’s age of 4.6 billion years, it has
consumed 1.9 × 1019 kg per year multiplied by 4.6 × 109 years,
E = (4.39 × 10–29 kg)(3.00 × 108 m/s)2 = 3.95 × 10–12 J. or a total of 8.7 × 1028 kg of hydrogen. This is about 44 percent
of the Sun’s original supply of available hydrogen fuel, con-
Fusing a single gram of hydrogen into helium releases about fi rming that Old Sol is now approaching middle age. A favor-
6 × 1011 joules (J) of energy, which is enough to boil all of the ite theme of science fiction is the fate that awaits Earth when
water in about 10 average backyard swimming pools. Scale this the Sun dies, but we need not worry anytime soon. The Sun
up to converting roughly 600 million metric tons (1 metric ton is only about halfway through its 10-billion-year lifetime as a
= 103 kg) of hydrogen into helium every second (with 4 mil- main-sequence star.
(a) Proton (p+) Hydrogen is the most abundant element in the universe, so
Slow-moving protons it offers the most abundant source of nuclear fuel at the
approach each other beginning of a star’s lifetime. Hydrogen burning is also the
but are pushed apart
by electric repulsion…
most efficient form of nuclear fusion, converting a larger
+
fraction of mass into energy
+ than does any other type of Hydrogen burns
reaction. But the most impor- mostly via the
tant reason why hydrogen proton-proton chain.
burning is the dominant pro-
…but the faster they cess in main-sequence stars is that hydrogen is also the
(b)
are going, the closer easiest type of atom to fuse. Hydrogen nuclei—protons—
+ together they can get. have an electric charge of +1. The electric barrier that must
be overcome to fuse protons is the repulsion of a single pro-
+ ton against another. Compare this to the force required, for
example, to get two carbon nuclei close enough to fuse. To
At high enough fuse carbon we must overcome the repulsion of six protons
temperatures, protons in one carbon nucleus pushing against the six protons in
may move rapidly another carbon nucleus. The resulting force is proportional
(c) enough to overcome
Positron the electric repulsion
to the product of the charges of the two atomic nuclei, mak-
(e+) and fuse… ing the repulsion between two carbon nuclei 36 times stron-
Neutron
ger than that between two protons. For this reason, hydrogen
fusion occurs at a much lower temperature than does any
other type of nuclear fusion. In the cores of low-mass stars
…releasing energy
Neutrino and leaving behind a
such as the Sun, hydrogen burns primarily through a pro-
deuterium nucleus, a cess called the proton-proton chain. The dominant branch
positron, and a neutrino. of the proton-proton chain is discussed in Foundations 14.1.
AstroTour: The Solar Core
FIGURE 14 . 3 Atomic nuclei are positively charged and so
repel each other. (a, b) If two nuclei are moving toward each
other, the faster they are going, the closer they will get before
veering away. (c) At the temperatures and densities found in the
Energy Produced in the Sun’s Core
centers of stars, thermal motions of nuclei are energetic enough Must Find Its Way to the Surface
to overcome this electric repulsion, so fusion takes place.
Some of the energy released by hydrogen burning in the
core of the Sun escapes directly into space in the form of
neutrinos (see Foundations 14.1), but most of the energy
a density about 150 times the density of water (which is goes instead into heating the solar interior. To understand
1,000 kilograms per cubic meter [kg/m3]), and the tempera- the structure of the Sun we must understand how thermal
ture at the center of the Sun is energy is able to move outward through the star. The nature
about 15 million kelvins (K). Hydrogen fuses of energy transport within a star is one of the key factors
The thermal motions of atomic to helium in the determining the star’s structure.
nuclei in the Sun’s core con- core of the Sun. Thermal energy can be transported from one place to
tain tens of thousands times another by a number of methods. Pick up a bucket of hot
more kinetic energy than the thermal motions of atoms at coals and carry it from one side of the room to the other,
room temperature. As illustrated in Figure 14.3c, under and you have transported thermal energy. A common way
these conditions atomic nuclei slam into each other hard in which energy is transported in our everyday lives is ther-
enough to overcome the electric repulsion between them mal conduction. For example, hold one end of a metal rod
and allow short-range nuclear forces to act. The hotter and while you put the other end into a fi re. Soon the end of the
denser a gas is, the more of these energetic collisions will rod that is in your hand becomes too hot to hold. Thermal
take place each second. For this reason, the rate at which conduction occurs as the energetic thermal vibrations of
nuclear fusion reactions occur is extremely sensitive to the atoms and molecules in the hot end of the rod cause their
temperature and the density of the gas. Half of the energy cooler neighbors to vibrate more rapidly as well. However,
produced by the Sun is generated within the inner 9 percent although thermal conduction is the most important way
of the Sun’s radius, or less than 0.1 percent of the volume energy is transported in solid matter, it is typically ineffec-
of the Sun (see Figure 14.2b). tive in a gas because the atoms and molecules are too far
There are several reasons why hydrogen burning is the apart to transmit vibrations to one another efficiently. Ther-
most important source of energy in main-sequence stars. mal conduction is unimportant in the transport of energy
The Proton-Proton Chain

In the Sun and in other low-mass stars, hydrogen burning takes in the proton-proton chain consists of a proton and a neutron.
place through a series of nuclear reactions called the proton- This is the nucleus of a heavy isotope of hydrogen called “deu-
proton chain. The proton-proton chain has three different terium,” or 2H. In the second step of the proton-proton chain,
branches. The most important of these, responsible for about another proton slams into the deuterium nucleus, fusing with it
85 percent of the energy generated in the Sun, consists of three to form the nucleus of a light isotope of helium, 3He, consisting
steps, illustrated in Figure 14.4. In the fi rst step, two hydrogen of two protons and a neutron. The energy released in this step
nuclei fuse. In the process, one of the protons is transformed is carried away as a gamma-ray photon. In the third and fi nal
into a neutron by emitting a positively charged particle called step of the proton-proton chain, two 3He nuclei collide and fuse,
a positron and another type of elementary particle called a neu- producing an ordinary 4He nucleus and ejecting two protons in
trino. The conversion of a proton into a neutron by the emission the process. The energy released in this step shows up as kinetic
of a positron and a neutrino is one variety of a process referred energy of the helium nucleus and two ejected protons.
to as beta decay. This dominant branch of the proton-proton chain can be
The positron is expelled at a great velocity, carrying away written symbolically as
some of the energy released in the reaction. Electrons and 1
H + 1H → 2H + e+ + ν,
positrons have opposite electric charges, so they attract each
other. As a result, our expelled positron will soon collide with followed by
one of the many electrons moving freely about in the center e+ + e– → γ + γ
of the Sun. But the positron is the antiparticle of the electron 2
H + 1H → 3He + γ
and, as we’ll learn in Chapter 21, when particle and antipar- 3
He + 3He → 4He + 1H + 1H
ticle collide they annihilate each other, with their total mass
being converted into energy. Thus, the annihilation of electrons Here the symbols are e+ for a positron, e – for an electron,
and positrons in the Sun’s core produces energy in the form ν (the Greek letter nu) for a neutrino, and γ (gamma) for a
of gamma-ray photons. These photons carry part of the energy gamma-ray photon.
released when the two protons fused, thereby helping to heat The rate at which the proton-proton chain reaction takes
the surrounding gas. The neutrino, on the other hand, is a very place depends on both temperature and pressure. At the tem-
elusive particle. Its interactions with matter are so feeble that perature and pressure that exist within the Sun’s core, the reac-
its most likely fate is to escape from the Sun without further tion rate is relatively slow—in fact, extremely slow compared
interactions with any other particles. to a nuclear bomb explosion. The Sun’s slow nuclear burning
Follow along in Figure 14.4 as we step through the proton- is fortunate for us. Otherwise the Sun would have exhausted
proton chain. The new atomic nucleus formed by the fi rst step its supply of hydrogen long ago.
Gamma rays (G
FIGURE 14 . 4 The Sun
e– and all other main-sequence
Positron (e+) G stars get their energy by
1 Neutrino fusing the nuclei of four
H
hydrogen atoms together
to make a single helium
1H
atom. In the Sun, about
1
Deuterium (2H) 1H H 85 percent of the energy
Helium-3 (3He) produced comes from the
1H
branch of the proton-proton
1H
Helium-4 chain shown here.
1
(4He) H
1
H
Neutrino G
– e+
e
1 In the first step, colliding 2 In the second step, protons 3 In the final step, helium-3
protons create deuterium (2H). collide with deuterium nuclei nuclei collide to create
to produce helium-3 (3He). helium-4 (4He).
1 Hotter regions are more crowded with

photons than cooler regions…
Hotter region Cooler region
2 ...making it more likely 3 As a result, radiation

that photons will move carries energy
from hotter to cooler Net energy transfer outward from the hot,
regions than in the energy-generating
reverse direction. core of the Sun.
FIGURE 14 . 5 Higher-temperature regions deep within the Sun produce more radiation than do
lower-temperature regions farther out. Although radiation flows in both directions, more radiation
flows from the hot regions to the cooler regions than from the cooler regions to the hot regions.
Therefore, radiation carries energy outward from the inner parts of the Sun.
from the core of the Sun to its surface. Thermal energy is rial depends on many things, including the density of the
instead carried outward from the center of the Sun by two material, its composition, its temperature, and the wave-
other mechanisms: radiation and convection. length of the photons moving through it.
In radiative transfer, energy is transported from hotter Radiative transfer is most efficient in regions where
to cooler regions by photons, which carry the energy with opacity is low. In the inner part of the Sun, where tempera-
them. Imagine a hotter region tures are high and atoms are ionized, opacity comes mostly
of the Sun sitting next to a Radiation carries from the interaction between photons and free electrons
cooler region, as shown in energy from hotter (electrons not attached to any atom). Here opacity is rela-
Figure 14.5. Recall from our regions to cooler tively low, and radiative transfer is capable of carrying the
study of radiation in Chapter regions. energy produced in the core outward through the star. The
4 that the hotter region will region in which radiative transfer is responsible for energy
contain more (and more energetic) photons than the cooler transport extends 71 percent of the way out toward the
region. (There will be a Planck spectrum of photons in both surface of the Sun. This region is referred to as the Sun’s
regions, so the total energy carried by photons will be pro- radiative zone (Figure 14.6). Even though the opacity of
portional to the fourth power of the temperature in each the radiative zone is relatively low, photons are still able
region.) More photons will move by chance from the hotter to travel only a short distance before interacting with mat-
(“more crowded”) region to the cooler (“less crowded”) ter. The path that a photon follows is so convoluted that, on
region than in the reverse direction. Thus, there is a net average, it takes the energy of a gamma-ray photon produced
transfer of photons and photon energy from the hotter region in the interior of the Sun about 100,000 years to find its way
to the cooler region, and in this way radiative transfer car- to the outer layers of the Sun. Opacity serves as a blanket,
ries energy outward from the Sun’s core. holding energy in the interior of the Sun and letting it seep
If the temperature differs by a large amount over a short away only slowly.
distance, then the concentration of photons will differ From a peak of 15 million K at the center of the Sun, the
sharply as well, favoring rapid temperature falls to about 100,000 K at the outer margin of
radiative energy transfer. The Opacity impedes the the radiative zone. At this temperature, atoms are no lon-
transfer of energy from one outward flow of ger completely ionized, and the opacity is therefore higher.
point to another by radiation radiation. As the opacity increases, radiation becomes less efficient
also depends on how freely in carrying energy from one place to another. The energy
radiation can move from one point to another within a star. that is flowing outward through the Sun “piles up.” The
The degree to which matter impedes the flow of photons physical sign that energy is piling up is that the tempera-
through it is referred to as opacity. The opacity of a mate- ture gradient—that is, how rapidly temperature drops with
FIGURE 14 .6 The interior

structure of the Sun is divided
into zones on the basis of
where energy is produced and
how it is transported outward.
1 Energy produced in 2 …is carried outward 3 …then by 4 …then away from
the Sun’s core… first by radiation… convection… the Sun by radiation.
Core
Radiative
zone
Convective
zone
Energy radiated
from the Sun’s surface
increasing distance from the center of the Sun— becomes convective zone (see Figure 14.6) extends from the outer
very steep. (Radiative transfer carries energy from hotter boundary of the radiative zone out to just below the visible
regions to cooler regions, smoothing out temperature dif- surface of the Sun.
ferences between them. As the opacity increases, radiation In the outermost layers of stars, radiation again takes
becomes less effective in smoothing out temperature differ- over as the primary mode of energy transport. (This must
ences, so temperature differences between one region and be the case. After all, it is radiation that transports energy
another become greater.) from the outermost layers of a star off into space.) Even so,
As we move farther toward the surface of the Sun, radia- the effects of convection can be seen as a perpetual roiling
tive transfer becomes so inefficient (and the temperature of the visible surface of the Sun.
gradient so steep) that a dif-
ferent way of transporting In the outer part of
energy takes over. Like a hot- the Sun, energy is
air balloon, cells (or packets) carried by convection.
What If the Sun Were Different?
of hot gas become buoyant and At this point we have seen the pieces that go into calculat-
rise up through the lower-temperature gas above them, car- ing a model of the interior of the Sun. To put these pieces
rying energy with them. Thus, convection begins. Just as together, we again stress that the key is balance. The tem-
convection carries energy from the interior of planets to perature and density in the core of the model Sun must be
their surfaces, or from the Sun-heated surface of Earth just right to produce the same amount of energy as would
upward through Earth’s atmosphere, convection also plays be radiated away by a star of the same size and surface
an important role in the transport of energy outward from temperature as the Sun. The temperature and density at
the interiors of many stars, including the Sun. The solar each point within the model Sun must be just right so that
transport of energy away from the core by radiation and

convection just balances the amount of energy produced by Imagine a hypothetical Sun
fusion in the core. The density, temperature, and pressure that has a mass of 1 M£ but
is larger than the Sun.
of the model Sun must vary from point to point in such a
way that the outward push of pressure is everywhere bal-
anced by the inward pull of gravity. Finally, the whole True Hypothetical
Sun “larger”
model must depend on only two things: the total mass of Sun
gas from which the star is made, and the chemical compo-
With more surface area
sition of that gas. to radiate, the larger Sun
Let us restate that last idea to drive the point home. In would be more luminous
than the true Sun…
order to be successful, our model of the Sun must start with
no more information than the known mass and chemical
composition of the real Sun, and from these it must cor-
rectly predict all the other observed properties of the real
Sun. The amazing thing is that our model does exactly that. Fgrav …but gravity would
Computer modeling shows that a ball of gas containing one be weaker inside the
solar mass with the same composition as our Sun can have Fgrav F = GMm larger Sun.
only one structure and still satisfy all the different balances r2
at the same time. Our model predicts what the size, tem-
perature, and luminosity of the resulting star should be—
predictions that agree remarkably well with the observed
properties of the real Sun.
To better understand why there is only one possible
structure that a 1-M⊙ star (a star with a mass equal to the
Sun’s mass) can have, we can “what if” the Sun. For example,
we might ask, “What if a star had the same mass, surface P
P Weaker gravity means that the
temperature, and composition as the Sun, but was somehow pressure in the interior of the
larger than the Sun? What properties would such a hypo- larger Sun would be lower.
Hydrostatic
thetical star have? Specifically, what would happen to the equilibrium
balance between the amount of energy generated within
such a hypothetical star and the amount of energy that it
Lower pressure means lower
radiates away into space?” Follow along in Figure 14.7 as we temperature and less energy
consider what would happen if the Sun were “too large.”3 generated in the larger Sun
We begin with the second part of this balance. Because than in the actual Sun.
our hypothetical star would have more surface area than the
Sun, it would be able to more effectively radiate its energy
into space. To keep a one-solar-mass star inflated to a size The larger Sun would not
larger than the Sun would require that the star be more produce enough energy to
replace that radiated away.
luminous than the Sun.
Now let’s consider what is going on in the interior of our
hypothetical star. Because the star is larger than the Sun but
contains the same amount of mass as the Sun, the force of
gravity at any point within our hypothetical star would be
less than the force of gravity at the corresponding location
within the Sun. (This is a result of the inverse square law
of gravitation: if the radius R is larger in our hypothetical
star, then 1/R2 must be smaller.) Weaker gravity means that
3
Our approach here is admittedly simplified. We have not taken into
consideration certain subtle factors, such as the Sun’s differential
As the out-of-balance
rotation. hypothetical Sun lost
energy, it would shrink
until the energy produced
again balanced the
FIGURE 14 .7 A star like the Sun can have energy radiated away.
only the structure that it has. Here we imagine
the fate of a Sun with too large a radius.
the weight of matter pushing down on the interior of our more than these properties. In particular, the standard model
hypothetical star would be less than in the Sun. Because of the Sun predicts exactly what nuclear reactions should
hydrostatic equilibrium means that the pressure at any point be occurring in the core of the
within a star is equal to the weight of overlying matter, the Sun, and at what rate. The Neutrinos escape
pressure at any point in the interior of our hypothetical nuclear reactions that make freely from the
star would be less than the pressure at the corresponding up the proton-proton chain core of the Sun.
point in the Sun. This reduction in pressure would affect produce copious quantities of
the amount of energy that our star produced. The proton- neutrinos. As we have noted, neutrinos are extremely elu-
proton chain runs faster at higher temperatures and densi- sive beasts— so elusive that almost all of the neutrinos pro-
ties, so the lower pressure in the interior of our hypothetical duced in the heart of the Sun travel freely through the outer
star means that less energy would be generated there than parts of the Sun and on into space as if the Sun were not
in the core of the Sun. there. The core of the Sun lies buried beneath 700,000 km
Wait a minute—there is a contradiction here. We found of dense, hot matter, seemingly buried forever away from
that our hypothetical star would have to be more luminous our view. Yet the Sun is transparent to neutrinos.
than the Sun, but at the same time it would be producing It may take thermal energy produced in the heart of the
less energy in its interior than the Sun does. This discrep- Sun 100,000 years to fi nd its way to the Sun’s surface, but
ancy violates the balance that must exist in any stable star the solar neutrinos streaming through you as you read these
between the amount of energy generated within the star and words were produced by nuclear reactions in the heart of the
the amount of energy radiated into space. Our hypotheti- Sun only 81 ⁄3 minutes ago. If we could fi nd a way to capture
cal star cannot exist! Stated another way, even if we could and analyze these neutrinos, think of what we might learn!
somehow magically pump up the Sun to a size larger than it In principle, neutrinos offer a direct window into the very
actually is, it would not remain that way. Less energy would heart of the Sun’s nuclear furnace.
be generated in its core, while more energy would be radi-
ated away at its surface. The Sun would be out of balance.
As a result, the Sun would lose energy, the pressure in the We Use Neutrinos to Observe
interior of the Sun would decline, and the Sun would shrink
back toward its original (true) size.
the Heart of the Sun
We could do the same thought experiment the other way Transforming the promise of neutrino astronomy into real-
around, asking what would happen if the Sun were smaller ity turns out to be a formidable technical challenge. The
than it actually is. With less surface area, it would radi- same property of neutrinos that makes them so exciting to
ate less energy. At the same time the Sun’s mass would be astronomers—the fact that their interaction with matter is
compacted into a smaller volume, driving up the strength so feeble that they can escape unscathed from the interior of
of gravity and therefore the pressure in its interior. Higher the Sun— also makes them notoriously difficult to observe.
pressure implies higher density and temperature, which Suppose we wanted to build a neutrino detector capable of
in turn would cause the proton-proton chain to run faster, stopping half of the neutrinos falling on it. Our hypothetical
increasing the rate of energy generation. Again there is a detector would need the stopping power of a piece of lead a
contradiction— an imbalance. This time, with more energy light-year thick! Yet despite the difficulties, neutrinos offer
being generated in the interior than is radiated away from a unique window into the Sun so powerful that they are
the surface, pressure in the Sun would build up, causing it worth going to great lengths to try to detect.
to expand toward its original (true) size. Fortunately, the Sun produces a truly enormous num-
In the end, there is only one possible Sun. If a star ber of neutrinos. As you lie in bed at night, about 400 tril-
contains one solar mass of material of solar composition, lion solar neutrinos pass through your body each second,
only one structure allows that having already passed through Earth. With this many neu-
star to maintain all of the If the Sun were any trinos about, a neutrino detector does not have to be very
balances that must be main- different, it would not efficient to be useful. Several methods have been devised
tained. Our stable, reliable be in balance. to measure neutrinos from the Sun and from other astro-
Sun is the result. nomical sources, such as supernova explosions, and a
number of such experiments are under way. These experi-
ments have successfully detected neutrinos from the Sun,
and in so doing they have provided crucial confi rmation
14.3 The Standard Model of that nuclear fusion reactions indeed are responsible for
powering the Sun.
the Sun Is Well Tested As with many good experiments, however, measure-
ments of solar neutrinos raised new questions while answer-
The standard model of the Sun correctly predicts such global ing others. After their initial joy at confi rming that the Sun
properties of the Sun as its size, temperature, and luminos- really is a nuclear furnace, astronomers became troubled
ity. This is a remarkable feat, but the model predicts much that there seemed to be only about a third to a half as many
Neutrino oscillation can change an electron neutrino νe into a muon neutrino νμ in two ways.
νμ νe
νe νμ νe
νe νμ
1 An electron neutrino 2 It begins to change 3 It is now a 4 A measurement here has a 5 The muon neutrino changes
is emitted in space. into a muon neutrino. muon neutrino. 50-50 chance of finding an back to an electron neutrino.
electron or muon neutrino.
6 An electron neutrino is
emitted in the Sun’s core.
7 At a particular matter density,

it converts to a muon neutrino
and stays in that form.
νe νμ νμ νμ
FIGURE 14 .8 If neutrinos have mass, they should oscillate among

the three types: electron, muon, and tau. In sequence (1)–(5), a neutrino
oscillates between electron and muon types. Changes from one type to
another can also take place in the presence of matter at a certain density
found within the Sun. In (6), an electron neutrino created in the Sun’s core
is converted to a muon neutrino (7), which then arrives at Earth. Early
neutrino detectors would not have recognized this muon neutrino.
solar neutrinos as predicted by solar models. The difference telescopes around the world is showing that neutrinos do
between the predicted and measured flux of solar neutrinos have a nonzero mass, and this work has uncovered evidence
was referred to as the solar neutrino problem. of neutrino oscillations.
Understanding the solar neutrino problem was an area Solving the solar neutrino problem is a good example
of very active research. One possible explanation was that of how science works—how a better model of the neutrino
our understanding of the structure of the Sun was some- showed that the solar neutrino problem was real and not
how wrong. This possibility seemed unlikely, however, merely an experimental mistake, and how a single set of
because of the many other successes of our solar model. anomalous observations was later confi rmed by other more
A second possibility was that our understanding of the sophisticated experiments. All of this has led to a better
neutrino itself was incomplete. The neutrino was long understanding of basic physics.
thought to have zero mass (like photons) and to travel at
the speed of light. But if neutrinos actually do have a tiny
amount of mass, then theories from particle physics pre-
dict that solar neutrinos should oscillate (alternate back
Helioseismology Enables Us
and forth) among three different kinds, or “fl avors”—the to Observe Echoes from
electron, muon, and tau neutrinos, as shown in Figure
14.8. Because only one of these, the electron neutrino,
the Sun’s Interior
could interact with the atoms in the earlier neutrino detec- In Chapter 7 we found that models of Earth’s interior pre-
tors (described in Excursions 14.1), neutrino oscillations dict how density and temperature change from place to
provided a convenient explanation for why we saw only place within our planet. These differences affect the way
about a third of the expected number of neutrinos. And, pressure waves travel through Earth, bending the paths
as seen in Figure 14.8, electron neutrinos should also of these waves. We test models of Earth’s interior by com-
change fl avor as they interact with solar material during paring measurements of seismic waves from earthquakes
their escape from the Sun. with model predictions of how seismic waves should travel
After several decades of work on the solar neutrino prob- through the planet.
lem, this last idea has won out. Work currently under way The same basic idea has now been applied to the Sun.
at high-energy physics labs, nuclear reactors, and neutrino Detailed observations of motions of material from place
E X C U R S I O N S 1 4 .1
Neutrino Astronomy
A neutrino telescope hardly fits anyone’s expectation of what Kamiokande is located in an active zinc mine 2,700 meters
a telescope should look like. The fi rst apparatus designed under Mount Ikena, near Kamioka, Japan (see Figure 21.14.) It
to detect solar neutrinos consisted of a cylindrical tank is a 50,000-ton tank of ultrapure water, surrounded by 13,000
fi lled with 100,000 gallons of dry-cleaning fluid— C2Cl4, or photomultiplier tubes capable of registering extremely faint
perchloroethylene—buried 1,500 meters deep within the Home- fl ashes of light. When a neutrino interacts with an atom in
stake Gold Mine in Lead, South Dakota. A tiny fraction of neu- the tank, a faint conical fl ash of blue light is produced. This
trinos passing through this fluid interact with chlorine atoms, fl ash is seen by some of the photomultipliers. Figure 14.9b
causing the reaction shows a map of the flash of light from a single neutrino. SNO
uses 1,000 tons of heavy water (D2O) contained in a 12-meter
37
Cl + ν → 37Ar + e– sphere surrounded by light detectors (Figure 14.9c) and bur-
ied 2,000 meters deep in an active nickel mine near Sudbury,
to take place. The 37Ar formed in the reaction is a radioactive Ontario. An objective of still newer neutrino telescopes is to
isotope of argon. The tank must be buried deep within Earth collect higher-energy neutrinos that originate from the most
to shield the detector from the many other types of radiation distant objects in space. These experiments use the oceans
capable of producing argon atoms. The argon is flushed out of or Antarctic ice as part of the detector. Figure 14.9d shows
the tank every few weeks and measured. the IceCube Neutrino Observatory detector under construc-
The Homestake detector (Figure 14.9a) began operations tion at the South Pole. IceCube’s optical sensors are buried
in 1965. Over the course of 2 days, roughly 1022 (10 billion tril- far beneath the surface, at depths of up to 2.5 km within the
lion) solar neutrinos pass through the Homestake detector. Antarctic ice.
Of these, on average only one neutrino interacts with a chlo- Neutrino telescopes observe neutrinos produced in the
rine atom to form an atom of argon. Even so, this interaction heart of the Sun, allowing us to directly observe the results of
produces a measurable signal. Since then, over a dozen and a the nuclear reactions going on there. Although these observa-
half neutrino detectors have been built, each using different tions have provided crucial confirmation that stars are powered
reactions to detect neutrinos of different energies. For exam- by nuclear reactions, they have also challenged our models
ple, the Soviet-American Gallium Experiment (SAGE) and the of the solar interior and led us to change our ideas about the
European Gallium Experiment (GALLEX) use reactions involv- nature of the neutrino itself (as is evident from the discussion
ing the conversion of gallium atoms into germanium atoms of the solar neutrino problem in the text). In addition to solar
(71Ga + ν → 71 Ge + e –) to detect solar neutrinos. SAGE uses 60 neutrinos, a number of experiments detected neutrinos from
tons of metallic gallium. GALLEX uses 30 tons of gallium in Supernova 1987A. As we will see in Chapter 17, this was an
the form of gallium chloride. These two experiments account explosion marking the end of the life of a massive star located
for most of the world’s current supply of gallium. 160,000 light-years away in a small galaxy called the Large
Among the more ambitious detectors capable of detect- Magellanic Cloud. Neutrino astronomy was one of the great
ing all three fl avors of neutrinos are the Super-Kamiokande innovations of 20th century astronomy, and it is certain to have
detector and the Sudbury Neutrino Observatory (SNO). Super- many applications in the 21st century.
to place across the surface of the Sun show that the Sun Observing the “ringing” motion of the surface of the
vibrates or “rings,” something like a struck bell. Compared Sun is a difficult business. To detect the disturbances of
to a well-tuned bell, however, the vibrations of the Sun are helioseismic waves on the surface of the Sun, astrono-
very complex, with many different frequencies of vibrations mers must use instruments capable of measuring Doppler
occurring simultaneously. These motions are echoes of what shifts of less than 0.1 meter per second (m/s) while detect-
lies below. Just as geologists use seismic waves from earth- ing changes in brightness of only a few parts per million
quakes to probe the interior of Earth, solar physicists use at any given location on the Sun. In addition, tens of mil-
the surface oscillations of the Sun to test our understanding lions of different wave motions are possible within the
of the solar interior. This science, new to the later years of Sun. Some waves travel around the circumference of the
the 20th century, is called helioseismology. Like neutrino Sun, providing information about the density of the upper
astronomy, helioseismology has created quite a stir among convection zone. Other waves travel through the interior
astronomers by letting us “see” into the invisible heart of of the Sun, revealing the density structure of the Sun close
the Sun (Figure 14.10). to its core. Still others travel inward toward the center of
(b)
(a) (c)
(d)
FIGURE 14 .9 Neutrino “telescopes” do not look much like visible-light telescopes. (a) The Homestake neutrino detector is
a 100,000-gallon tank of dry-cleaning fluid located deep in a mine in South Dakota. (b) A map of the flash of light from a single
neutrino caught by a neutrino detector. (c) The Sudbury Neutrino Observatory, buried 2 km deep in a Canadian nickel mine. (d)
The IceCube Neutrino Observatory at the South Pole.
the Sun, until they are bent by the changing solar den- a joint mission between NASA and the European Space
sity and return to the surface. All of these wave motions Agency. By orbiting at the L1 Lagrangian point of the Sun-
are going on at the same time. Sorting out this jumble Earth system (see Chapter 10), SOHO moves in lockstep
requires computer analysis of long, unbroken strings of with Earth at a location approximately 1,500,000 km (0.01
solar observations. astronomical unit, or AU) from Earth almost directly in line
Helioseismology studies of the Sun got a huge boost in between Earth and the Sun. SOHO carries a complement of
the closing years of the 20th century with two very suc- 12 scientific instruments designed to monitor the Sun and
cessful projects. The Global Oscillation Network Group, or measure the solar wind upstream of Earth. SOHO observa-
GONG, is a network of six solar observation stations spread tions have dramatically improved our detailed knowledge
around the world. With this network, solar astronomers of the Sun.
are able to observe the surface of the Sun approximately To interpret helioseismology data, scientists compare the
90 percent of the time. The other project is the Solar and strength, frequency, and wavelengths of observed vibrations
Heliospheric Observatory, or SOHO spacecraft, which is with predicted vibrations calculated from models of the
stand such a surface, imagine a fog bank. The surface of

the fog bank is a gradual thing— an illusion really. Imagine
watching some people walking into a fog bank. When they
disappeared from view, you would say that they were defi-
nitely inside the fog bank, even though they never passed
through a noticeable boundary. The apparent surface of
the Sun is defi ned by the same effect. Light from the Sun’s
surface can escape into space, so we can see it. Light from
below the Sun’s surface cannot escape directly into space,
so we cannot see it.
The Sun’s surface is referred to as the solar photosphere.
(Photo means “light”; the photosphere is the place light
comes from.) There is no
instant when you can say that The apparent surface
you have suddenly crossed of the Sun is called
the surface of a fog bank, and the photosphere.
by the same token there is no
instant when we suddenly cross the photosphere of the Sun.
The surface of the Sun—the photosphere—has an effective
FIGURE 14 .10 The interior of the Sun rings like a bell as temperature4 of 5,780 K and ranges from 6,600 K at the bot-
helioseismic waves move through it. This figure shows one tom to 4,400 K at the top. It is a zone about 500 km thick,
particular “mode” of the Sun’s vibration. Red shows regions where across which the density and opacity of the Sun increase
gas is traveling inward; blue, where gas is traveling outward. We
sharply. The reason the Sun appears to have a well-defi ned
can observe these motions by using Doppler shifts.
surface and a sharp outline when viewed from Earth (never
look at the Sun directly!) is that this zone is relatively shal-
low; 500 km does not look very thick when viewed from a
solar interior. This technique provides a powerful test of our distance of 150 million km.
understanding of the solar interior, and it has led both to Look at a photograph of the Sun such as Figure 14.11a,
some surprises and to improvements in our models. For and notice that the Sun appears to be fainter near its edges
example, some scientists had proposed that the solar neu- than near its center. This effect, called limb darkening, is
trino problem might be solved an artifact of the structure of the Sun’s photosphere. (The
if there were less helium in Helioseismology “limb” of a celestial body is the outer border of its visible
the Sun than generally imag- confirms the disk.) The cause of limb darkening is illustrated in Figure
ined—an explanation that was predictions of 14.11b. Near the edge of the Sun you are looking through
ruled out by analysis of the solar models. the photosphere at a steep angle. As a result, you do not see
waves that penetrate to the as deeply into the interior of the Sun as when you are look-
core of the Sun. Helioseismology also showed that the value ing directly down through the photosphere near the center
for opacity used in early solar models was too low. This real- of the Sun’s disk. The light you see coming from near the
ization led astronomers to recalculate the location of the limb of the Sun is from a layer in the Sun that is shallower
bottom of the convective zone. Both theory and observation and hence cooler and fainter.
now put the base of the convective zone at 71.3 percent of
the way out from the center of the Sun, with an uncertainty
in this number of less than half a percent! The amazing
agreement between the predictions of computer models of
The Solar Spectrum Is Complex
the Sun and the temperature and density structure within In one sense the surface of the Sun may be an illusion, but
the Sun measured by helioseismology is truly remarkable. in another sense it is not. The transition between “inside”
the Sun and “outside” the Sun is quite abrupt. In the out-
ermost part of the Sun, the density of the gas drops very
rapidly with increasing altitude. This is the region we refer
14.4 The Sun Can Be Studied to as the Sun’s atmosphere. Figure 14.12 shows how the
pressure and temperature change across the atmosphere
Up Close and Personal of the Sun. The Sun’s atmosphere is where all visible solar
phenomena take place.
The Sun is a large ball of gas, so it has no solid surface in
the sense that Earth does. Instead, it has the kind of surface 4
The effective temperature of an object is the temperature of a black-
that a cloud on Earth does, or a “surface” like that of one of body (Chapter 4) that would emit the same amount of electromagnetic
the Jovian planets in our Solar System. To help you under- radiation as the object.
The Sun is “limb-darkened.” …because near its edge we see

It is dimmer near its edge… the Sun at a steep angle and so do
FIGURE 14 .11 (a) When
(a) (b) not see deeply into its atmosphere.
viewed in visible light, the Sun
appears to have a sharp outline,
even though it has no true
surface. The center of the Sun
appears brighter, while the limb
of the Sun is darker— an effect
known as limb darkening. (b)
Looking at the middle of the Sun
allows us to see deeper into
the Sun’s interior than looking
at the edge of the Sun does.
Because higher temperature
means more luminous radiation,
the middle of the Sun appears
brighter than its limb.
Near its center we observe the surface

of the Sun face-on, so we see deeper
into hotter, brighter regions.
10–3
Corona
Photosphere
Convective
Chromosphere 10–5
zone
Density (kg/m3)
Prominence
10–7
Photosphere
Sunspot 10–9
Radiative Chromosphere Corona
zone
10–11
Core 106
Temperature (K)
105
104
103
0 1,000 2,000 3,000
Height (km)
FIGURE 14 .12 The components of the Sun’s atmosphere, along with plots of how the
temperature and density change with height at the base of the Sun’s atmosphere.
Very nearly all the radiation from below the Sun’s pho- leak out of the Sun’s interior has a spectrum very close to
tosphere is absorbed by matter and cannot escape—it is being a Planck (blackbody) spectrum. This is why, in Chap-
trapped. This is exactly our definition from Chapter 4 for ter 13, we were able to understand much about the physical
the conditions under which blackbody radiation is formed. properties of stars by applying our understanding of black-
We should not then be surprised that the radiation able to body radiation.
13 we discussed the presence of absorption lines in the

spectra of stars. Now we can take a closer look at how these
absorption lines form. As photospheric light travels upward
through the solar atmosphere, atoms in the solar atmosphere
absorb the light at discrete wavelengths. Because from our
perspective the Sun is so much brighter than any other star,
its spectrum can be studied in far more detail. Today, spe-
cially designed telescopes and high-resolution spectrome-
ters have been built specifically to study the light from the
Sun. The amazing structure present in the solar spectrum
can be seen in Figure 14.13. Absorption lines from over 70
elements have been identified. Analysis of these lines forms
the basis for much of our knowledge of the solar atmosphere,
including the composition of the Sun, and is the starting
point for our understanding of the atmospheres and spec-
FIGURE 14 .13 A high-resolution spectrum of the Sun, tra of other stars.
stretching from 400 nanometers (nm) (lower left corner) to 700
nm (upper right corner), showing a wealth of absorption lines.
The Sun’s Outer Atmosphere:
Chromosphere and Corona
As we examine the structure of the Sun in more detail, As we move upward through the Sun’s photosphere, the
however, this simple description of the spectra of stars temperature continues to fall, reaching a minimum of about
begins to go wrong. The fact that light from the solar pho- 4,400 K at the top of the photosphere. At this point the trend
tosphere must escape through the upper layers of the Sun’s reverses and the temperature slowly begins to climb, rising
atmosphere affects the spectrum that we see. In Chapter to about 6,000 K at a height of 1,500 km above the top of the
FIGURE 14 .14 (a) This image of the Sun, taken in H∝ light during a transit of Venus, shows structure in the Sun’s chromosphere. The
planet Venus is seen in silhouette against the disk of the Sun. (b) The chromosphere seen during a total eclipse. (c) This eclipse image
shows the Sun’s corona, consisting of million-kelvin gas that extends for millions of kilometers beyond the surface of the Sun.
(a) (c)
(b)
photosphere. This region above the photosphere is called

the chromosphere (Figure 14.14a). The chromosphere was
discovered in the 19th cen-
tury during observations of The Sun’s
total solar eclipses (Figure chromosphere lies
14.14b). The chromosphere is above the
seen most strongly as a source photosphere.
of emission lines, especially
the Hα line from hydrogen (the “hydrogen alpha line”). In
fact, the deep red color of the Hα line is what gives the chro-
mosphere (“the place where color comes from”) its name. It
was also from a spectrum of the chromosphere that the ele-
ment helium was discovered in 1868. Helium is named after
helios, the Greek word for “Sun.”5 The reason for the chro-
mosphere’s temperature reversal with increasing height is
not well understood, but it may be caused by magnetic waves
propagating through the region.
At the top of the chromosphere, across a transition FIGURE 14 .15 A close-up image of the Sun, showing the
region that is only about 100 km thick, the temperature tangled structure of coronal loops.
suddenly soars (see Figure
14.12). Above this transition The corona has a
lies the outermost region of temperature of Recall from Chapter 8 that “hot” atoms are able to escape
the Sun’s atmosphere, called millions of kelvins. from the tops of planetary atmospheres. Shouldn’t this hap-
the corona, in which temper- pen on the Sun as well? In this case, the answer is yes and
atures reach 1 million–2 million K. The corona is probably no. The corona is far too hot to be held in by the Sun’s grav-
heated by magnetic waves and magnetic fields in much the ity, but over most of the surface of the Sun coronal gas is
same way the chromosphere is, but why the temperature confi ned instead by magnetic loops with both ends fi rmly
changes so abruptly at the transition between the chromo- anchored deep within the Sun. The magnetic field in the
sphere and the corona is not at all clear. Since ancient times corona acts almost like a network of rubber bands that coro-
the Sun’s corona has been known—visible during total solar nal gas is free to slide along but cannot cross. In contrast,
eclipses as an eerie outer glow stretching a distance of sev- about 20 percent of the surface of the Sun is covered by an
eral solar radii beyond the Sun’s surface (Figure 14.14c). ever-shifting pattern of coronal holes. These are apparent
Because it is so hot, the solar corona is a strong source of in X-ray images of the Sun as dark regions (Figure 14.16),
X-rays. Atoms in the corona are also highly ionized. The indicating that they are cooler and lower in density than
spectrum of the corona shows emission lines from ionic their surroundings. Coronal holes are large regions where
species such as Fe13+ (iron atoms from which 13 electrons the magnetic field points outward, away from the Sun, and
have been stripped) and Ca14+ (calcium atoms from which where coronal material is free to stream away into inter-
14 electrons have been stripped). planetary space.
We have encountered this flow of coronal material away
from the Sun before. This is the same solar wind respon-
Solar Activity Is Caused sible for shaping the magne-
tospheres of planets (see A “wind” blows away
by Magnetic Effects Chapters 8 and 9) and for from the Sun.
Virtually all of the structure seen in the atmosphere of the blowing the tails of comets
Sun is imposed on the gas by the Sun’s magnetic field. High- away from the Sun (see Chapter 12). The relatively steady
resolution images of the Sun, such as Figure 14.15, show part of the solar wind consists of lower-speed flows, with
“coronal loops” that make the Sun look as though it were velocities of about 350 kilometers per second (km/s), and
covered with matted, tangled hair. This fibrous or rope- higher-speed flows, with velocities up to about 700 km/s.
like texture in the chromosphere is the result of magnetic The higher-speed flows originate in coronal holes. Depend-
structures called flux tubes, much like the flux tube that ing on their speed, particles in the solar wind take about
connects Io with Jupiter (see Chapter 9). Magnetic fields are 2–5 days to reach Earth. Frequently, 2–5 days after a coro-
responsible for much of the structure of the corona as well. nal hole passes across the center of the face of the Sun, there
is an increase in the speed and density of the solar wind
5
It is rather remarkable that helium, the second most common ele- reaching Earth. The solar wind drags the Sun’s magnetic
ment in the universe, was discovered in the Sun before it was identi- field along with it. The magnetic field in the solar wind gets
fied on Earth! “wound up” by the Sun’s rotation, as shown in Figure 14.17.
The magnetic field is “wound

up” by the Sun’s rotation like
water from a sprinkler.
Magnetic field
Earth’s Solar wind

orbit
Sun’s rotation
As the solar wind streams away

from the Sun, it acts counter to
the Sun’s rotation by dragging
the magnetic field with it.
FIGURE 14 .16 X-ray images of the Sun show a very different
FIGURE 14 .17 The solar wind streams away from active areas
picture of our star than do images taken in visible light. The
brightest X-ray emission comes from the base of the Sun’s and coronal holes on the Sun. As the Sun rotates, the solar wind
corona, where gas is heated to temperatures of millions of picks up a spiral structure, much like the spiral of water that
kelvins. This heating is most powerful above magnetically active streams away from a rotating lawn sprinkler.
regions of the Sun. The dark areas are coronal holes, regions
where the Sun’s corona is cooler and has lower plasma density
than in their surroundings.
FIGURE 14 .18 The solar

wind streams away from the
Sun for about 100 AU, until
it finally piles up against the
pressure of the interstellar
Magnetic medium through which the
The solar wind field Sun is traveling. The Voyager
blows a bubble… Interstellar 1 spacecraft is now passing
medium through this boundary and is
expected to cross over into the
Solar wind true interstellar medium within
the coming decade.
Jupiter
Saturn Sun’s movement
Uranus Sun 50 100 150
AU
Neptune
...that pushes interstellar

gas out of the path
of the Sun's motion.
Consequently, the solar wind has a spiral structure, some-

thing like a stream of water from a rotating lawn
sprinkler.
The effects of the solar wind are felt throughout the Solar
System. As we have seen, the solar wind causes the tails of
comets, shapes the magnetospheres of the planets, and pro-
vides the energetic particles that power Earth’s spectacular
auroral displays. Using space probes, we have been able to
observe the solar wind extending out to 100 AU from the
Sun. But the solar wind does not go on forever. The farther
it gets from the Sun, the more it has to spread out. Just like
radiation, the density of the solar wind follows an inverse
square law. At a distance of about 100 AU from the Sun,
the solar wind is assumed no longer to be powerful enough
to push the interstellar medium (the gas and dust that lie
between stars in a galaxy and that surround the Sun) out
of the way. There the solar wind will stop abruptly, “piling
up” against the pressure of the interstellar medium. Figure
14.18 shows this region of space over which the wind from
the Sun holds sway. Sometime within the next decade the
Voyager 1 spacecraft is expected to cross the very outer edge
of this boundary and begin sending back our fi rst “on the
FIGURE 14 .19 These SOHO images show a large sunspot
scene” measurements of true interstellar space.6
group. Sunspots are magnetically active regions that are cooler
The best-known features on the surface of the Sun are
than the surrounding surface of the Sun.
relatively dark blemishes in the solar photosphere, called
sunspots, that come and go over time. Sunspots appear
dark, but only in contrast to the brighter surface of the Sun
(see Math Tools 14.2). Early telescopic observations of sun-
spots made during the 17th century led to the discovery of
the Sun’s rotation, which has an average period of about 27
6
The truth is, we really don’t know just how far from the Sun this
boundary lies, but we should know it when we see it.
M AT H T O O L S 1 4 . 2
Sunspots and Temperature

Sunspots are about 1,500 K cooler than their surroundings.
What does this tell us? Think back to the Stefan-Boltzmann
FIGURE 14 . 20 A very high-resolution view of a sunspot,
law in Chapter 4. The flux from a blackbody is proportional
showing the dark umbra surrounded by the lighter penumbra. The
to the fourth power of the temperature. We write this rela-
solar surface around the sunspot bubbles with separate cells of
tionship as hot gas called “granules.” The smallest features are about 100
 ∝ T 4. km across.
Let’s take round numbers for the temperature of a typi-
cal sunspot and the surrounding photosphere: 4,500 and
days as seen from Earth and 25 days relative to the stars.
6,000 K, respectively. The Stefan-Boltzmann law informs
Observations of sunspots also show that the Sun, like Sat-
us that the surface brightness of a sunspot is about
urn, rotates more rapidly at its equator than it does at higher
latitudes. This effect, referred to as differential rotation, is
( 6,000
4,500
_____
) = 0.32
4
possible only because the Sun is a large ball of gas rather
than a solid object.
or about 1 ⁄3 the brightness of the surrounding photosphere. Figure 14.19 is a photograph of a large sunspot group,
and Figure 14.20 shows the remarkable structure of one of
Cardboard masks
Card with a
small hole
Binoculars
Tripod
Shadow
FIGURE 14 . 2 1 There are a number of ways of safely viewing sunspots on the Sun or observing a
partial solar eclipse. Two methods are illustrated here. Never look at the Sun directly (except during
the totality phase of a total solar eclipse)!
these blemishes on the surface of the Sun. Each sunspot come and go in about 2 days, and 90 percent are gone within
consists of an inner dark core called the umbra, which is 11 days. The amount of sunspot activity and the locations
surrounded by a less dark region called the penumbra. The on the Sun where sunspots are seen to change over time in
penumbra shows an intricate radial pattern, reminiscent of a pronounced 11-year pattern called the sunspot cycle (Fig-
the petals of a flower. Obser- ure 14.22a). At the beginning of a cycle, sunspots begin to
vations of sunspots show that Magnetic fields appear at solar latitudes of about 30° to the north and south
they are magnetic in origin, cause sunspots. of the solar equator. Over the
with magnetic fields thou- following years the regions Sunspots come and go
sands of times greater than the magnetic field at Earth’s where most sunspots are seen in an 11-year cycle.
surface. Sunspots occur in pairs that are connected by move toward the equator as
loops in the magnetic field, much like the shape of a horse- the number of sunspots increases and then declines. As the
shoe. Sunspots range in size from about 1,500 km across last few sunspots near the equator are seen, sunspots again
up to complex groups that may contain several dozen indi- begin appearing at middle latitudes and the next cycle
vidual spots and span 50,000 km or more. The largest sun- begins. A diagram showing the number of sunspots at a
spot groups are so large that they can be seen with the given latitude plotted against time has the appearance of a
naked eye and have been noted since antiquity. But do not series of opposing diagonal bands and is often referred to
look directly at the Sun! More than one casual observer of as the sunspot “butterfly diagram” (Figure 14.22b).
the Sun has paid the price of blindness for a glimpse of a Telescopic observations of sunspots date back almost
naked-eye sunspot. Figure 14.21 shows two ways to look 400 years, and there were naked-eye reports of sunspots
at the Sun in safety. Direct viewing through welder’s glass even before that. Figure 14.23 shows the historical record of
is also safe. sunspot activity. Although astronomers often speak of the
Individual sunspots are ephemeral. Although sunspots 11-year sunspot cycle, the cycle is neither perfectly periodic
occasionally last 100 days or longer, half of all sunspots nor especially reliable. The time between peaks in the num-
(a) Number of sunspots The number of sunspots Each sunspot peak is

200 varies in an 11-year cycle. called a solar maximum.
Sunspot number
100
0
(b) Location of sunspots
90° N
During each cycle sunspots first
appear far from the equator…
30° N
Latitude on Sun
…then later appear

0° close to the equator.
30° S
90° S
(c) Average magnetic field
90° N
Magnetic Magnetic
South North
30° N Magnetic
Latitude on Sun
The Sun’s magnetic field

field flips at each + 40 gauss
0° solar maximum.
+ 20 gauss
0 gauss
30° S
Magnetic Magnetic 20 gauss
North South
90° S 40 gauss
1975 1980 1985 1990 1995 2000
Date
(d) The Sun in ultraviolet light
Overall solar activity increases with

the approach of a solar maximum.
Early 1997 Mid-1998 Late 1999
FIGURE 14 . 22 (a) The number of sunspots versus time for the last few solar cycles. (b) The solar
butterfl y diagram, showing the fraction of the Sun covered at each latitude. (c) The Sun’s magnetic
field flips every 11 years. (d) The approach of a solar maximum is apparent in these SOHO images
taken in ultraviolet light.
FIGURE 14 . 23 Sunspots
Before 1645 there were From the mid–17th century to the early Since the early 18th century, a regular
observations of sunspots. 18th century—the Maunder Minimum— 11-year cycle is clearly visible. have been observed for
almost no sunspots were seen. hundreds of years. In this plot
the 11-year cycle in the number
200
of sunspots (half of the 22-
year solar magnetic cycle) is
Number of sunspots
150 clearly visible. Sunspot activity

Maunder varies greatly over time. The
Minimum
period from the middle of
100
the 17th century to the early
18th century, when almost no
50 sunspots were seen, is called
the Maunder Minimum.
1600 1700 1800 1900 2000

Year
ber of sunspots actually varies between about 9.7 and 11.8 pair tends to be a north magnetic pole, whereas the trail-
years. The number of spots seen during a given cycle fluctu- ing sunspot tends to be a south magnetic pole. In the next
ates as well. Some cycles are real monsters. There have also sunspot cycle this polarity is reversed: the leading spot in
been times when sunspot activity has disappeared almost each pair is a south magnetic pole. The transition between
entirely for extended periods. The most recent extended these two magnetic polarities occurs near the peak of each
lull in solar activity, called the Maunder Minimum, lasted sunspot cycle (Figure 14.22c). The predominant theory of
from 1645 to 1715. Normally there would be six peaks in what causes this magnetic cycle involves a dynamo in the
solar activity in 70 years, but virtually no sunspots were interior of the Sun, much like the dynamos that generate
seen during the Maunder Minimum. the magnetic fields of the planets.
In the early 20th century, American solar astronomer The effects of magnetic activity on the Sun are felt
George Ellery Hale (1868–1938) was the fi rst to show that throughout the Sun’s photosphere, chromosphere, and
the 11-year sunspot cycle was actually half of a 22-year corona. Sunspots are only one of a host of phenomena that
magnetic cycle during which the direction of the Sun’s follow the Sun’s 22-year cycle of magnetic activity. The
magnetic field reverses from one 11-year sunspot cycle to peaks of the cycle, referred to as solar maxima, are times
the next. In one sunspot cycle the leading sunspot in each of intense activity, as can be seen in the ultraviolet images
(a)
FIGURE 14 . 2 4 (a) Solar prominences are magnetically
supported arches of hot gas that rise high above active
regions on the Sun. (b) A close-up view at the base of a large
prominence. An image of Earth is shown to scale.
(b)
FIGURE 14 . 25 A large
(a) (b) sunspot group (a) emits a
powerful flare, shown (b) in
Sunspot group ultraviolet light.
Flare
of the Sun in Figure 14.22d. Although sunspots are darker- Figure 14.25 shows images of a solar flare erupting
than-average features, they are often accompanied by a from a sunspot group. Solar flares are the most energetic
brightening of the solar chromosphere that is seen most form of solar activity. Flares are violent eruptions in which
clearly in the light of emission lines such as Hα. The mag- enormous amounts of magnetic energy are released over
nificent loops seen arching through the solar corona are also the course of a few minutes to a few hours. Solar flares
anchored in solar active regions. These include solar promi- can heat gas to temperatures of 20 million K, and they are
nences such as those shown in Figure 14.24. Prominences the source of intense X-ray and gamma-ray radiation. Hot
are magnetic flux tubes of relatively cool (5,000–10,000 K) plasma (consisting of atoms stripped of some of their elec-
gas extending through the million-kelvin gas of the corona. trons) moves outward from flares at speeds that can reach
Although most prominences are relatively quiescent, oth- 1,500 km/s. Magnetic effects can then accelerate subatomic
ers can erupt out through the corona, towering a million particles to almost the speed of light. Such events, called
kilometers or more over the surface of the Sun and ejecting coronal mass ejections (Figure 14.26), send powerful bursts
material into the corona at velocities of 1,000 km/s. of energetic particles outward through the Solar System,
affecting us in many ways when they reach Earth (see the
next subsection). Coronal mass ejections occur about once
FIGURE 14 . 26 A SOHO image of a coronal mass ejection per week during the minimum of the solar activity cycle,
(upper right), with a simultaneously recorded ultraviolet image of
but they can reach bursts of two or more per day near the
the solar disk superimposed.
maximum of the cycle.
Solar Activity Affects Earth

The amount of solar radiation at Earth’s distance from the
Sun is, on average, 1.35 kilowatts per square meter (kW/m2).7
Satellite measurements of the amount of radiation coming
from the Sun (Figure 14.27) show that this value can vary
by as much as 0.2 percent over periods of a few weeks as
dark sunspots and bright spots in the chromosphere move
across the disk. Overall, however, the increased radiation
from active regions on the Sun more than makes up for the
reduction in radiation from sunspots. On average, the Sun
seems to be about 0.1 percent brighter during the peak of a
solar cycle than it is at its minimum.
Solar activity has many effects on Earth. Solar active
regions are the source of most of the Sun’s extreme ultra-
violet and X-ray emission. This energetic radiation heats
7
Photons from the Sun provide an important alternative source of
energy.
1 Near solar maxima the

Sun is slightly brighter
Solar irradiance (Watts/m2 at Earth)
1,368 1,400
and more variable…
1,200
1,367 2 …but the variation is a
1,000 tiny fraction of the Sun's
Watts per square

meter (on Earth)
total luminosity.
1,366
800
600
1,365
400
1,364
200
1,363 0
1980 1991
200
Year
3 The changes in solar
Sunspot number
luminosity and variability go

hand in hand with changes
100 in sunspot activity.
0
1980 1991
Year
FIGURE 14 . 2 7 Measurements taken by satellites above Earth’s atmosphere show that the amount of
light from the Sun changes slightly over time.
Earth’s upper atmosphere and, during periods of solar activ- disrupt electric power grids, causing blackouts across large
ity, causes Earth’s upper atmosphere to expand. When this regions. Coronal mass ejections also have adverse effects
happens, the swollen upper atmosphere can significantly on radio communication and navigation, and they can dam-
increase the atmospheric drag on spacecraft orbiting near age sensitive satellite electronics, including communication
Earth, causing their orbits to decay. (Prior to the launch of satellites. Energetic particles accelerated in solar flares pose
the Hubble Space Telescope, many people working on the one of the greatest dangers to human exploration of space.
project were concerned that, because it was to be launched Over the years, people have attempted to relate sunspot
into the teeth of a solar maximum, the telescope might not activity to phenomena ranging from the performance of
survive very long.) One reason for repeated shuttle visits to the stock market to the mating habits of Indian elephants.
the HST was to “reboost” it to its original orbit to make up Few of these supposed relationships have withstood seri-
for the slow decay in its orbit caused by drag in the rarefied ous scrutiny, however. An especially interesting idea is that
outer parts of Earth’s atmosphere. NASA gave a final shuttle variations in Earth’s climate might be related to solar activ-
“reboost” to the HST in 2009, keeping the space telescope ity. We have certainly seen that solar activity affects Earth’s
in operation until at least 2014, when it will be replaced by upper atmosphere, and it might not be too much of a reach
the larger James Webb Space Telescope. Earth’s atmospheric to imagine that it could affect weather patterns as well. It
drag will eventually bring the HST back to Earth— although has also been suggested that variations in the amount of
not in one piece. radiation from the Sun might be responsible for variations
Earth’s magnetosphere is the result of the interaction in Earth’s climate in the past. Although the causes of the
between Earth’s magnetic field and the solar wind. Increases current global climatic change remain controversial, mod-
in the solar wind accompany- els suggest that observed variations in the Sun’s luminos-
ing solar activity, especially Solar storms cause ity could account for only about 0.1-K differences in Earth’s
coronal mass eject ions auroras and disrupt average temperature—much less than the effects due to the
directed at Earth, can disrupt electric power ongoing buildup of carbon dioxide in Earth’s atmosphere.
Earth’s magnetosphere in grids on Earth. On the other hand, triggering the onset of an ice age may
ways that are obvious even to require a sustained drop in global temperatures of only
nonscientists. Spectacular auroras can accompany such about 0.2–0.5 K, so the quest for a definite link between solar
events, as can magnetic storms that have been known to variability and changes in Earth’s climate persists.
from the fusion of hydrogen to make helium in its core.

Seeing the Forest for the Trees That energy, carried from the Sun’s core to its surface by
radiation and convection, arrives at just the rate needed
The universe contains more stars than there are grains to replace the energy radiated away into space. All of
of sand on all of Earth’s beaches. Even “nearby” stars this takes place while a delicate standoff is maintained
are so distant that at best we see them as faint points of in the battle between the inward pull of gravity and the
light in the night sky. Only one star is close enough for outward push of pressure fueled by the nuclear fi res
its complex structure to be studied in great detail. That in the Sun’s interior. Models of the Sun make a host of
star is a maelstrom of fierce activity, fueled by nuclear detailed predictions, ranging from the density and tem-
fi res burning deep within its interior. Storms rage across perature at every point within the Sun’s interior to the
its surface. Giant flares burst forth, arching millions of nuclear reactions that take place in its core. Over the past
kilometers into space before collapsing back again. That decades the predictions of models of the Sun have been
star heats a huge and rarefied corona to millions of kel- tested in ever more sophisticated and detailed ways, and
vins and blows a bubble in interstellar space large enough those predictions have been confi rmed. One decade into
to swallow our entire Solar System. That star— a main- the 21st century, it is fair to say that we know what the
sequence G2 star, different in no way from main-sequence interior of the Sun is like.
G2 stars throughout our galaxy and the universe—is the Our knowledge of the Sun also forms the cornerstone
star we call our Sun. of our knowledge of all other stars everywhere. Starting
No machine of human manufacture has ever visited with a successful model of the Sun, we can now go on to
the interior of the Sun, just as no machine has more than ask a host of other questions: What happens if we increase
scratched the surface of Earth. Even so, our understand- the mass of the model star? What happens if we change its
ing of the interior of the Sun is remarkably complete. Like chemical abundances? What happens if we run the model
everything else in the universe, the Sun is governed by forward in time until all of the hydrogen at the star’s cen-
the laws of physics. Computer models based on those ter is gone? How do we assemble something that looks like
physical laws show that there is only one possible struc- the Sun in the first place? Our study of the Sun has set the
ture that a 1-M⊙ ball of gas with the chemical composi- stage for the next leg of our journey as we turn our atten-
tion of the Sun can have while remaining in balance. tion to how stars form, how they differ from one another,
According to those models, the Sun draws its energy how they change with time, and how they die.
Summary • Sunspots reveal the 11- and 22-year cycles in solar

activity.
• Nuclear reactions converting hydrogen to helium are • Solar storms can produce terrestrial auroras and dis-
the source of the Sun’s energy. rupt power grids.
• Energy created in the Sun’s core moves outward to the
surface, fi rst by radiation and then by convection.
• Neutrinos are elusive, almost massless particles that includes algorithmically generated versions of these
interact only very weakly with other matter. questions, plus additional conceptual exercises. If
• The apparent surface of the Sun is called the
photosphere.
• The temperature of the Sun’s atmosphere ranges from
about 6,000 K near the bottom to about 1 million K at
the top.
Student Questions
• Material streaming away from the Sun’s corona creates THINK ING A BOUT THE CONCEP T S
the solar wind.
1. The Sun’s stability depends on hydrostatic equilibrium
• Sunspots are photospheric regions that are cooler than and energy balance. Describe how both of these work.
their surroundings.
*2. Using the “what if” discussion in Section 14.2 as a guide,
• The Sun’s outermost region is called the corona. explain how hydrostatic equilibrium acts as a safety
valve to keep the Sun at its constant size, temperature, *17. Discuss the “solar neutrino problem” and how this prob-
and luminosity. lem was solved.
3. What is the strong nuclear force, and how does it provide 18. What technique do you find in common between how we
stability for atomic nuclei? probe the internal structure of the Sun and the internal
structure of Earth?
4. Explain nuclear fusion and how it relates to the Sun’s
source of energy. 19. The Sun’s visible “surface” is not a true surface, but a
feature we call the photosphere. Explain why the pho-
5. Two of the three atoms in a molecule of water (H2O) are
tosphere is not a true surface.
hydrogen. Why are Earth’s oceans not fusing hydrogen
into helium and setting Earth ablaze? 20. The second most abundant element in the universe was
not discovered here on Earth. What is it and how was it
6. Engineers and physicists dream of solving the world’s
discovered?
energy supply problem by constructing power plants that
would convert globally plentiful hydrogen to helium. 21. Describe the solar corona. Under what circumstances
Our Sun seems to have solved this problem. On Earth, can we see it without using special instruments?
what is a major obstacle to this potentially environmen-
22. The solar corona has a temperature of millions of degrees;
tally clean solution?
the photosphere has a temperature of only about 6,000
7. Name the part of the Sun where most of its energy is K. Why isn’t the corona much, much brighter than the
produced, and explain why the Sun’s primary source photosphere?
of energy is located there.
23. What is the solar wind?
*8. Explain the proton-proton chain process through which
24. Sunspots are dark splotches on the Sun. Why are they
the Sun generates energy by converting hydrogen to
dark?
helium.
25. What have sunspots told us about the Sun’s rotation?
9. On Earth, nuclear power plants use fission to generate
electricity. Here, a heavy element like uranium is broken 26. Explain the relationship between the 11-year sunspot
into many atoms, where the total mass of the fragments cycle and the 22-year magnetic cycle.
is less than the original atom. Explain why fission could
27. What is a solar flare?
not be powering the Sun today.
28. How is the fate of the Hubble Space Telescope tied to
10. In the proton-proton chain process, the mass of four pro-
solar activity?
tons is slightly greater than the mass of a helium nucleus.
Explain what happens to this difference in mass. 29. Solar flares and magnetic storms can have an adverse
effect on certain things important to our daily lives.
11. What experiences have you had with energy transmit-
Name some of the things that are affected.
ted to you by radiation alone? By convection alone? By
conduction alone? (Hint: You are exposed to all three
every day.)
12. Is gas an efficient conductor of thermal energy? Explain A P P LY I N G T H E C O N C E P T S
why or why not.
30. Assume that the Sun’s mass is about 300,000 Earth
*13. Suppose an abnormally large amount of hydrogen sud-
masses and that its radius is about 100 times that of
denly burned in the core of the Sun. What would hap-
Earth. The density of Earth is about 5,500 kg/m3.
pen to the rest of the Sun? Would we see anything as a
a. What is the density of the Sun?
result?
b. How does this compare with the density of water?
14. The Sun has a radius equal to about 2.3 light-seconds.
31. The Sun shines by converting mass into energy accord-
Explain why a gamma ray produced in the Sun’s core
ing to Einstein’s well-known relationship E = mc2. Show
does not emerge from the Sun’s surface 2.3 seconds
that if the Sun produces 3.85 × 1026 J of energy per sec-
later.
ond, it must convert 4.3 million metric tons (4.3 × 109
15. A high-energy photon and a neutrino are created in the kg) of mass per second into energy.
Sun’s core at the same instant. Which one will reach
32. Assume that the Sun has been producing energy at a
Earth fi rst? Explain your answer.
constant rate over its lifetime of 4.5 billion years (1.4 ×
16. Why are neutrinos so difficult to detect? 1017 seconds).
a. How much mass has it lost creating energy over its energy, how many 5-megaton bombs would it take to
lifetime? equal that energy?
b. The present mass of the Sun is 2 × 1030 kg. What frac-
37. Verify the claim made at the start of this chapter that the
tion of its present mass has been converted into energy
Sun produces more energy per second than all the elec-
over the lifetime of the Sun?
tric power plants on Earth could generate in a half-mil-
33. Imagine that the source of energy in the interior of the lion years. Estimate or look up how many power plants
Sun changed abruptly. there are on the planet, and how much energy an average
a. How long would it take before a neutrino telescope power plant produces. Be sure to account for different
detected the event? kinds of power—for example, coal, nuclear, wind.
b. When would a visible-light telescope see evidence of
*38. Let’s examine how we know that the Sun cannot power
the change?
itself by chemical reactions. Using Math Tools 14.1 and
34. On average, how long does it take particles in the solar the fact that an average chemical reaction between two
wind to reach Earth from the Sun if they are traveling atoms releases 1.6 × 10 –19 J (1 electron-volt, or eV), esti-
at an average speed of 400 km/s? mate how long the Sun could emit energy at its current
luminosity. Compare that estimate to the known age of
35. A sunspot appears only 70 percent as bright as the sur-
Earth.
rounding photosphere. The photosphere has a tempera-
ture of approximately 5,780 K. What is the temperature *39. How much distance must a photon actually cover in its
of the sunspot? convoluted path from the center of the Sun to the outer
layers?
36. The hydrogen bomb represents humankind’s efforts to
duplicate processes going on in the core of the Sun. The
energy released by a 5-megaton hydrogen bomb is 2 × StudySpace is a free and open Web site that pro-
1016 J. vides a Study Plan for each chapter of 21st Century
a. How much mass did Earth lose each time a 5-megaton Astronomy. Study Plans include animations, read-
hydrogen bomb was exploded? ing outlines, vocabulary flashcards, and multiple-choice
b. This textbook, 21st Century Astronomy, has a mass quizzes, plus links to premium content in SmartWork and
of about 1.6 kg. If we converted all of its mass into the ebook. Visit www.wwnorton.com/studyspace.
There are no whole truths; all truths are half truths. It is
trying to treat them as whole truths that plays the devil.
ALFRED NORT H W HIT EHE AD (1861–19 47)
A star-forming region 3,000 light-years away in the constellation Cepheus.

CHAP TER 15
Star Formation and the

Interstellar Medium
15.1 Whence Stars? K E Y CONCEP T S

To mortal humans, the stars seem fi xed and unchanging. When you look at the night sky it is stars that you
Indeed, our local star, the Sun, has remained much the same notice, yet stars fill only the tiniest fraction of the
for long enough to allow life on Earth to arise, evolve, and volume of space. On this leg of our journey we turn our
begin to contemplate its own origin and fate. But our per- attention to the tenuous medium between the stars,
spective is distorted by our brief existence. All of recorded and we discover that
human history amounts to less than a millionth of the age
• There are various phases of the interstellar medium,
of the Sun and Earth. In Chapter 6 we told the story of how
ranging from cold, relatively dense molecular clouds to
our Solar System formed some 4.6 billion years ago out of a
swirling disk of gas and dust that surrounded the protostel- hot, tenuous intercloud gas.
lar Sun. We are now ready to return to that story, but with a • Different phases of the interstellar medium emit
different focus. First we will step back and look at the inter- various types of radiation and can be observed at
stellar environment that gave birth to the Sun and the Solar different wavelengths, ranging from radio waves to
System, and all the other stars, both large and small. Then X-rays.
we will focus our attention on the protostar at the center of
• Dust in the interstellar medium blocks visible light but
the circumstellar disk and watch as it becomes a star.
The birth of a star—from a cloud of gas and dust to glows with infrared radiation.
nuclear burning—is a process that can happen within a • Gas in the interstellar medium is heated and ionized by
few tens of thousand of years for the most massive stars, or energy from stars and stellar explosions.
require hundreds of millions of years for the least massive.
• Stars form in clusters from dense cores buried within
Such evolutionary times are far too long for us to witness
giant molecular clouds.
changes in an individual protostar, but astronomers have
pulled the larger picture together by observing many dif- • Forming stars are seen by their infrared emission and
ferent stars at various stages of stellar development. by the effects they have on their surroundings.
The root of the word astronomy may mean “star,” but if • We can follow the progress of an evolving protostar by
we ask what fraction of space is filled with stars, we get a its track across the H-R diagram.
very different picture of their importance. The volume of our
Sun is about 1.3 million times the volume of Earth. Though • Protostars collapse, radiating away their gravitational
this volume might seem enormous to us, it is almost incom- energy until fusion starts in their cores and they settle
prehensibly tiny compared with the volume of space. Our onto the main sequence.
Sun is the only star in a region of space with a volume of
over 300 cubic light-years. If our Sun were the size of a golf
417
418 Chapter 15 Star Formation and the Interstellar Medium
ball sitting on an author’s office desk in Tempe, Arizona, its average density of interstellar gas—that is, for it to have the
nearest stellar neighbor would be a similar golf ball sitting same amount of mass per unit volume as the interstellar
on another author’s desk in Santa Cruz, California, 1,000 medium has—you would have to stretch it into a tube 25,000
kilometers (km) away. The rest of the volume of our galaxy light-years long! Stated another way, there is about as much
is filled with the interstellar medium. It is here that the material in a column of air between your eye and the floor
story of stars begins and ends. Stars are formed from the gas beside you as there is interstellar gas in a column with the
in the interstellar medium, they live their lives there, and same cross section stretching from the Solar System to the
when they die they bequeath some of their chemical and center of the galaxy.
energetic legacy back to the interstellar medium.
Before taking a look at the interstellar medium, how-
ever, we need to put it into another context as well. Remem-
ber that galaxies are crowded in comparison with the vast
The Interstellar Medium Is Dusty
expanses of intergalactic space. Returning to the golf ball About 1 percent of the material in the interstellar medium
analogy, if we ask the whereabouts of a star in the spiral is in the form of solid grains, referred to as interstellar dust.
galaxy nearest to our own, that star would be equivalent to “Interstellar soot” might be a
a golf ball somewhere on Jupiter. But that is a story for later better name. Ranging in size About 1 percent of
in our journey. For now, we turn our attention to our “local from little more than large the mass of the
neighborhood” and the interstellar medium that fills the molecules up to particles interstellar medium
volume of our own galaxy. about 300 nanometers (nm) is in grains.
across, these solid grains more
closely resemble the particles of soot from a candle flame
than they do the dust that collects on a windowsill. (It would
take several hundred “large” interstellar grains to span the
15.2 The Interstellar Medium thickness of a single human hair.) Interstellar dust begins
to form when refractory materials (see Chapter 6) such as
The Sun formed from the interstellar medium, so it is not iron, silicon, and carbon stick together to form grains in
too surprising that the chemical composition of the inter- dense, relatively cool environments like the outer atmo-
stellar medium in our region of the Milky Way is similar to spheres and “stellar winds” of cool, red giant stars; or in
the chemical composition of the Sun (see the “astronomer’s dense material thrown into space by stellar explosions.
periodic table” in Figure 9.5). In the interstellar medium (More about these topics later.) Once these grains are in the
about 90 percent of the atomic nuclei are hydrogen, and the interstellar medium, other atoms and molecules may stick
remaining 10 percent are almost all helium. More massive to them. This process is remarkably efficient: about half of
elements together account for only 0.1 percent of the atomic all interstellar matter more massive than helium (1 percent
nuclei, or about 2 percent of the mass in the interstellar of the total mass of the interstellar medium) is found in
medium. Roughly 99 percent of that interstellar matter is interstellar grains.
gaseous, consisting of individual atoms or molecules moving If there is only as much material between us and the
freely about, as do the molecules in the air around us. center of the galaxy as there is between your eye and the
However, interstellar gas is far more tenuous than the floor, then you might expect
thick soup that we breathe. Each cubic centimeter of the air our view of distant objects to Interstellar dust
around you contains about be as clear as your view of obscures our view.
2.5 × 1019 molecules. A good Interstellar gas is your own big toe. But nothing
terrestrial vacuum pump can extremely tenuous. could be further from the truth. It turns out that interstellar
reduce this density down to dust is extremely effective at blocking light. If the air around
about 1010 molecules per cubic centimeter (cm3)—about a you contained as much dust as a comparable mass of inter-
billionth as dense. By comparison, the interstellar medium stellar material, it would be so dirty that you would be hard-
has an average density of less than 1 atom/cm3—one ten- pressed to see your hand held up 10 cm in front of your face.
billionth as dense still! To understand this concept more Go out on a dark summer night in the Northern Hemisphere
clearly, imagine a square tube 1 centimeter (cm) on a side (or a dark winter night in the Southern Hemisphere) and
and a meter long, filled with air and sealed at both ends. look closely at the Milky Way, visible as a faint band of dif-
Also imagine that you can stretch the tube without chang- fuse light running through the constellation Sagittarius. You
ing its cross section. As you do so, the air in the tube has will see a dark “lane” running roughly down the middle of
to spread out to fill the larger volume and so is less dense. this bright band, splitting it in two. This dark band is the
When the tube is stretched to a length of 10 meters, the air result of vast expanses of interstellar dust blocking our view
inside will be a tenth as dense as the air around you. When of distant stars (see, for example, Figure 20.1).
the tube is stretched to 100 meters, the air inside will be a When interstellar dust gets in the way of radiation from
hundredth as dense. For the air in the tube to match the distant objects, the effect is called interstellar extinction.
(a) a speedboat came by.) Now imagine viewing these two

situations from the perspective of the wave. The wave is
hardly affected by the small boat, but it is strongly affected
by the large boat. The energy to drive the wild motions of
the boat comes from the wave, so the motion of the wave
is affected by the interaction.
The interaction of electromagnetic waves with matter is
more involved than that of a boat rocking on the ocean, but
the same basic idea often
applies. Tiny interstellar dust Long-wavelength
grains interact most strongly radiation penetrates
with ultraviolet and blue light, interstellar dust.
(b)
which have wavelengths com-
parable to the typical size of dust grains. For this reason,
ultraviolet and blue light are effectively blocked by inter-
stellar dust. Short wavelengths suffer heavily from inter-
stellar extinction. Infrared and radio radiation, on the other
hand, have wavelengths that are too long to interact strongly
with the tiny interstellar dust grains, so they can travel
largely unimpeded across great interstellar distances. To
sum up, at visible and ultraviolet wavelengths, most of the
galaxy is hidden from our view by dust. In the infrared and
radio portions of the spectrum, however, we get a far more
complete view.
FIGURE 15.1 (a) An all-sky picture of the Milky Way taken
in visible light. The dark splotches blocking our view are dusty In the visible part of the spectrum, short-wavelength
interstellar clouds. (b) The same field of view as seen in the near blue light suffers more from extinction than does longer-
infrared. Infrared radiation penetrates interstellar dust, giving us wavelength red light. As a result, as shown in Figure 15.2,
a clearer view of the stars in our galaxy. an object viewed through dust looks redder (actually less
blue) than it really is. This effect is called reddening. Cor-
recting for the fact that stars and other objects appear both
fainter and redder than they would in the absence of dust
Not all electromagnetic radiation suffers equally from can be one of the most difficult parts of interpreting astro-
interstellar extinction, however. Figure 15.1 shows two nomical observations, often adding to our uncertainty in
images of our galaxy: one taken in visible light, the other the measurement of an object’s properties.
taken in the infrared (IR). The dark clouds that block the Interstellar extinction may not be much of a concern at
shorter-wavelength visible light seem to have vanished in infrared wavelengths, but dust still plays an important role
the longer-wavelength IR image (Figure 15.1b), enabling in what we see when we look
us to see through the clouds to the center of our galaxy at the sky in the infrared. Like Dust glows
and beyond. any other solid object, grains in the infrared.
As a starting point for understanding why short-wave- of dust glow at wavelengths
length radiation is blocked by dust while long-wavelength determined by their temperature. In Chapter 4 we learned
radiation is not, think about a different kind of wave— that the equilibrium between absorbed sunlight and emit-
waves on the surface of the ocean. Imagine you are on ted thermal radiation determines the temperatures of the
the ocean in a strong swell. If you are floating in a small terrestrial planets (see Figure 4.23). A similar equilibrium
boat— one much smaller than the wavelength of ocean is at work in interstellar space, where dust is often heated
waves— the swell causes you to bob gently up and down. by starlight and by the gas in which it is immersed to tem-
But that is about all; there is no other interaction between peratures of tens to hundreds of kelvins (Figure 15.3). At a
the waves and the small boat. The story is quite different temperature of 100 K, Wien’s law says that dust will glow
if you are on a larger boat that is about half as long as the most strongly at a wavelength of 29 micrometers (μm):
wavelength of the ocean waves. Now the bow of the boat
2,900
m K 2,900
m K
may be on a wave crest while the stern of the boat is in a λmax = __________ = __________ = 29
m.
trough, or vice versa. The boat tips wildly back and forth T 100 K
as the waves go by. If the size of the boat and the wave- Cooler dust— say, at a temperature of 10 kelvins (K)—glows
length of the waves are a good match, even fairly modest most strongly at a wavelength of 290 μm. Both of these wave-
waves will rock the boat. (You might have noticed this if lengths are in the far-infrared part of the electromagnetic
you were ever in a canoe or rowboat when the wake from spectrum. Much of what we see when looking at the sky
(a)
1 A star emits radiation 2 Interstellar dust blocks short-wavelength UV 3 As a result, a star looks redder
at all wavelengths. and blue light but lets most long-wavelength and fainter when viewed
infrared and radio radiation pass. through interstellar dust.
Long-wavelength
visible light
IR radiation
(b) (c)
Spectrum of star Spectrum of star seen through dust
1 0.10
0.8 0.08
When viewed through
interstellar dust, the
0.6 0.06 spectrum is fainter at
all wavelengths but
Intensity
much fainter in the blue.

0.4 0.04
0.2 0.02
0.0 0.00
400 500 600 700 400 500 600 700
Wavelength, λ (nm) Note that this scale is 1/ 10 Wavelength, λ (nm)
that of the graph at left.
FIGURE 15. 2 (a) The wavelengths of ultraviolet and blue light

are close to the size of interstellar grains, so the grains effectively
Absorption of visible Warm dust
block this light. Grains are less effective at blocking longer-
and ultraviolet starlight gives off infrared
wavelength light. As a result, the spectrum of a star (b) when heats insterstellar dust. thermal radiation.
seen through an interstellar cloud (c) appears fainter and redder.
with infrared “eyes” is not really starlight at all, but ther-

mal radiation from dust. When NASA launched the Infrared
Astronomical Satellite (IRAS) in 1983, astronomers got their Dust grain
fi rst look at the sky at wavelengths out to 100 μm, and the
views (like Figure 15.4a) were unlike any seen before. IRAS
showed the Milky Way’s dark clouds glowing brilliantly in
infrared radiation from dust. More recently, the Japanese
infrared satellite AKARI provided a similar view (Figure
15.4b) at much shorter wavelengths (9 μm). T 100 K
Like planet temperatures

(see Chapter 4), dust
FIGURE 15. 3 The temperature of interstellar temperature is determined
grains is determined by the same type of equilibrium by the equilibrium between
absorbed and radiated energy.
between absorbed and emitted radiation that
determines the temperatures of planets.
(a) very rapidly; so when one of these atoms runs into you, it
hits you very hard. But the gas is also extremely tenuous.
Typically, you would have to search a liter (1,000 cm3) or
more of hot intercloud gas to fi nd a single atom. To gather
up 1 gram of hot intercloud gas— about the mass found in
a liter of air—you would have to collect all of the gas in
a cube over 8,000 km on an edge! There are so few atoms
in a given volume of hot intercloud gas that atoms would
very rarely run into you, so this million-kelvin gas would
do little to keep you warm. You would radiate energy away
and cool off much faster than the gas around you could
replace the lost energy.
(b) Extremely hot intercloud gas is thought to occupy about
half the volume of interstellar space. This gas is heated
mostly by the energy of tre-
mendous stellar explosions We live in a bubble of
called supernovae. (We will million-kelvin
return to the story of super- intercloud gas.
novae in Chapters 16 and 17.
For now, it is enough to know that these are very energetic
events!) Our Solar System is located inside a rarefied bubble
of hot intercloud gas that has a density of about 0.005
hydrogen atom per cubic centimeter and is at least 650
FIGURE 15. 4 (a) Far-infrared (100 μm) IRAS images of the sky light-years across. Some astronomers have suggested that
show clouds of warm, glowing interstellar dust. (The black bands this is the remnant of the hot bubble produced by a super-
here and in Figure 15.5 are due to missing data.) (b) This AKARI nova explosion 300,000 years ago. Like the million-kelvin
image shows the galactic plane at shorter infrared wavelengths gas in the corona of the Sun, hot intercloud gas glows
(9 μm).
faintly in the energetic X-ray portion of the electromag-
netic spectrum. X-rays cannot penetrate Earth’s atmo-
sphere, so it is necessary to get above our atmosphere to
Interstellar Gas Has Different observe the radiation coming from hot intercloud gas. For
rocket- and satellite-borne X-ray telescopes, however, the
Temperatures and Densities entire sky is aglow with faint X-rays coming from the bub-
The gas and dust that fill interstellar space within our gal- ble of million-kelvin hot gas in which our Solar System is
axy are not spread out evenly. About half of all interstellar immersed (Figure 15.5).
gas is concentrated into only Not all intercloud gas is as hot as our local bubble. Most
about 2 percent of the volume Most interstellar of the remaining intercloud gas has a temperature of about
of interstellar space. These material is 8,000 K and a density ranging from about 0.01 to 1 atom per
relatively dense regions are concentrated in
referred to as interstellar relatively dense
clouds. Interstellar clouds are clouds.
FIGURE 15. 5 The faint X-ray glow that fills the sky is due
not like terrestrial clouds,
largely to emission from the bubble of million-kelvin gas that
which are locations where water vapor has condensed to surrounds the Sun. Bright spots are more distant sources,
form liquid droplets. Rather, interstellar clouds are places including objects such as bubbles of very hot, high-pressure gas
where the interstellar gas is more concentrated than in sur- surrounding the sites of recent supernova explosions.
rounding regions. The other half of the interstellar gas is
spread out through the remaining 98 percent of the volume
of interstellar space. This is called intercloud gas, meaning
gas that is found between the clouds.
The properties of intercloud gas vary from place to
place. Some intercloud gas is extremely hot, with tem-
peratures in the millions of kelvins, close to those found
in the centers of stars. Even so, were you to fi nd yourself
adrift in an expanse of hot intercloud gas, your fi rst con-
cern would be freezing to death! The gas is hot, which
means that the atoms that make it up are moving about
FIGURE 15.6 Warm (about

8,000 K) interstellar gas glows
in the Hα line of hydrogen. This
image, showing the Hα emission
from much of the northern
sky, reveals how complex the
structure of the interstellar
medium is.
cubic centimeter. To gather up a gram of this warm inter- discussion of energy states of atoms in Chapter 4). The atom
cloud gas, we would need to collect all of the gas in a cube then drops down to lower and lower energy states, emitting
of interstellar space “only” 800 km on an edge. a photon at each step. Together
Interstellar space is awash with the light from stars. these photons carry away an Interstellar gas can
Some of this is ultraviolet light with wavelengths shorter amount of energy equal to the produce both emission
than 91.2 nm. Each photon of this ultraviolet light has ionization energy of hydrogen and absorption lines.
enough energy that, if absorbed by a hydrogen atom, it will plus whatever energy of
ionize the atom, kicking its electron free of the nucleus. If motion the electron and proton had before recombining.
enough of these energetic photons are around, then atoms The photons emitted in this process have wavelengths cor-
in the interstellar gas cannot remain neutral (because any responding to the energy differences between the allowed
neutral hydrogen atom will soon be ionized by starlight). energy states of hydrogen. In other words, warm ionized
In this way about half of the volume of warm intercloud interstellar gas glows in emission lines characteristic of
gas is kept ionized. But if there is a large enough expanse hydrogen (as well as other elements). Usually, the strongest
of warm intercloud gas, then the ionizing photons in the emission line given off by warm interstellar gas in the vis-
starlight get “used up.” About half of the volume of warm ible part of the spectrum is the Hα (hydrogen alpha) line,
intercloud gas is mostly neutral. This gas is usually sur- which is seen in the red part of the spectrum at a wavelength
rounded by regions of ionized intercloud gas that shield of 656.3 nm.
the neutral gas from starlight, much as Earth’s ozone layer Figure 15.6 shows an image of a portion of the sky taken
shields the surface of Earth from harmful ultraviolet radia- in the light of Hα emission. The faint diffuse emission in
tion from the Sun. the figure comes mostly from
One way to look for interstellar gas, including both warm, ionized intercloud gas. H II regions are
warm and hot interstellar gas, is to study the spectra of The especially bright spots, ionized by hot,
distant stars. As starlight passes through interstellar gas, on the other hand, are quite luminous O
absorption lines form. Just as absorption lines that formed different. These are regions and B stars.
in the atmosphere of a star tell us much about the star’s tem- where intense ultraviolet radi-
perature, density, and chemical composition, interstellar ation from massive, hot, luminous O and B stars is able to
absorption lines provide similar information about the gas ionize even relatively dense interstellar clouds. These are
the light has passed through. We can also study interstel- called H II (“H two”) regions, signifying that they are made
lar gas by looking for the radiation that it emits. In regions up of the second, or ionized, form of hydrogen. In the chap-
of warm ionized gas, protons and electrons are constantly ters to come we will address the question of how long stars
“fi nding each other” and recombining to make hydrogen live, and we will discover that O stars live only a few mil-
atoms. Because it takes energy to tear a hydrogen atom lion years. As a result, they usually do not move very far
apart, conservation of energy says that when a proton and from where they formed. The glowing clouds seen as H II
an electron combine to form a hydrogen atom, this same regions are the very clouds from which the stars were born.
amount of energy must be given up. This freed-up energy Thus, H II regions are signposts to regions where active star
must go somewhere, and that somewhere is into the form formation is taking place.
of electromagnetic radiation. One of the closest H II regions to the Sun is the Orion
Typically, when a proton and an electron combine, the Nebula, a nebula in the constellation Orion that was for-
resulting hydrogen atom is left in an excited state (see our merly known as the Great Nebula (Figure 15.7). In fact, the
FIGURE 15.7 (a) The Orion

(a) (b)
Nebula is seen here as a
glowing region of interstellar
gas surrounding a cluster of
young, hot stars. New stars are
still forming in the dense clouds
surrounding the nebula. (b) The
Orion Nebula (at bottom) is only
a small part of the larger Orion
star-forming region. The dark
Horsehead Nebula is seen at
the upper left of part (b) and in
Figure 15.10a.
Orion Nebula is tiny as H II regions go. It is “great” only

because it happens to be very close to us. Almost all of the
ultraviolet light that powers the nebula comes from a single
FIGURE 15.8 This European Southern Observatory image
hot star, and all told only a few hundred stars are forming shows the 30 Doradus Nebula, located in a small companion
in its immediate vicinity. In contrast, Figure 15.8 shows galaxy to the Milky Way, 160,000 light-years from Earth. The
a giant H II region called 30 Doradus (also known as the ultraviolet radiation coming from the dense cluster of thousands
Tarantula Nebula). The 30 Doradus H II region is located in of young, massive stars causes interstellar gas in this region of
the Large Magellanic Cloud, a small companion galaxy to intense star formation to glow.
the Milky Way, 160,000 light-years distant. This enormous
cloud of ionized gas is powered by a dense star cluster con-
taining thousands of hot, luminous stars. If 30 Doradus were
as close as the Orion Nebula, it would be bright enough in
the nighttime sky to cast shadows!
Warm, neutral hydrogen gas gives off radiation in a
different way than warm, ionized hydrogen does. Many
subatomic particles, including protons and electrons, are
magnetized. It is as if each of these particles had a bar mag-
net, with a north and a south pole, built into it. According
to the rules of quantum mechanics, a hydrogen atom can
exist in only two different configurations. Either the mag-
netic “poles” of the proton and electron point in the same
direction, or they point in opposite directions. The electron’s
“magnet” and the proton’s “magnet” push on each other, so
the way they line up affects the energy that the atom has.
When the two “magnets” point in opposite directions, the
atom has slightly more energy than when they point in the
same direction.
It takes energy to change a hydrogen atom from the
lower-energy, magnetically aligned state to the higher-
energy, magnetically unaligned state. This is not a problem,
interstellar gas is concentrated into much denser interstel-

lar clouds that occupy only 2 percent of the volume of the
galaxy. Most interstellar clouds are composed primarily of
neutral atomic hydrogen (that is, isolated neutral hydrogen
atoms). These clouds are much cooler and denser than the
warm intercloud gas. They have temperatures of about 100
K and densities in the range of about 1–100 atoms per cubic
centimeter. But even at a density of 100 atoms/cm3, you
would still have to gather up all of the gas in a box 180 km
on an edge to collect a single gram of this material.
On Earth it is uncommon to fi nd atoms in isolation—
FIGURE 15.9 This radio image of the sky taken at a most atoms are tied up in molecules. (For example, in Earth’s
wavelength of 21 cm shows clouds of neutral hydrogen atmosphere only nonreactive gases such as argon are typi-
gas throughout our galaxy. Because radio waves penetrate cally found in their atomic form.) Yet the interstellar gas
interstellar dust, 21-cm observations are a crucial probe of the that we have encountered so far consists of isolated atoms
structure of our galaxy.
rather than molecules. The reason is that throughout most
of interstellar space, including most interstellar clouds,
molecules do not survive for long. If interstellar gas is too
because in warm, neutral interstellar gas, interactions hot, then any molecules that do exist will soon collide with
between atoms easily supply enough energy to do the trick. other molecules or atoms that have enough energy to break
But once a hydrogen atom is in its higher-energy (magneti- the molecules back down into their constituent atoms.
cally unaligned) state, getting back into its lower-energy The temperature in a neutral hydrogen cloud may be low
(magnetically aligned) state requires nothing but time. If enough for some molecules to survive, but even here a dif-
left undisturbed long enough, ferent process will soon destroy any molecules that might
a hydrogen atom in the higher- Neutral hydrogen form. Photons of starlight with enough energy to break mol-
energy state will spontane- emits 21-cm radiation. ecules apart can penetrate neutral hydrogen clouds. Only in
ously jump to the lower-energy the hearts of the densest of interstellar clouds, where dust
state, emitting a photon in the process. The energy differ- effectively blocks even the relatively low-energy photons
ence between the two magnetic states of a hydrogen atom required to shatter molecules, does interstellar chemistry
is extremely small: less than half a millionth of the energy get its chance. For obvious reasons, these dark clouds are
needed to tear the electron away from the proton completely. referred to as molecular clouds.
The radiation given off by these low-energy transitions has In images such as those in Figure 15.10, molecular clouds
a wavelength of 21 cm, which lies in the radio portion of are evident from their silhouettes against a background of
the spectrum. stars. Inside such clouds it is dark and usually very cold,
The tendency for hydrogen atoms to emit 21-cm radiation with a typical temperature of only about 10 K—that is, 10
is extremely weak. On average, you would have to wait about kelvins above absolute zero. Most of these clouds have den-
11 million years for a hydrogen atom in the higher-energy sities of about 100–1,000 molecules/cm3, but densities as
state to spontaneously jump to the lower-energy state and high as 1010 molecules/cm3
give off a photon. Even so, there is a lot of hydrogen in the have been observed. Even at Molecular clouds are
universe. In Figure 15.9, the sky is aglow with 21-cm radia- 1010 molecules/cm3 this gas is dense and cold.
tion from neutral hydrogen. Because of its long wavelength, still less than a billionth as
21-cm radiation freely penetrates dust in the interstellar dense as the air around you, making it an extremely good
medium, enabling us to see neutral hydrogen throughout vacuum by terrestrial standards. In this cold, relatively
our galaxy, while measurements of the Doppler shift of the dense environment, atoms combine to form a wide variety
line indicate how fast the source of the radiation is moving. of molecules.
These two attributes make the 21-cm line of neutral hydro- By far the most common component of molecular clouds
gen perhaps the most important kind of radiation there is is molecular hydrogen. Molecular hydrogen (H2) consists of
for understanding the structure of our galaxy. two hydrogen atoms and is the smallest possible molecule.
Molecules radiate mostly in the radio and infrared portions
of the electromagnetic spectrum. Molecular emission lines
Regions of Cool, Dense are useful to astronomers in much the same way that the
spectral lines from atoms are useful: Each type of molecule
Gas Are Called Clouds is unique in its properties, and the energies of its allowed
As mentioned earlier, intercloud gas fills 98 percent of the states are unique as well. The wavelengths of emission lines
volume of interstellar space but accounts for only half of from molecules are an unmistakable fingerprint of the kind
the mass of interstellar gas. The remaining 50 percent of all of molecule responsible for them.
(a) (b)
(c)
In addition to molecular hydrogen, approximately 150

other molecules have been discovered in interstellar space.
These molecules range from very simple structures such
as carbon monoxide (CO), one of the more common mol-
ecules, to complex organic compounds such as methanol
(CH3OH) and acetone [(CH3)2CO], to carbon chains as large as
HC11N (cyanopolyyne). Very large carbon molecules, made
of hundreds of individual atoms, bridge the gap between
large interstellar molecules and small interstellar grains.
As mentioned earlier, visible light cannot escape from the
molecular clouds where interstellar molecules are con-
centrated. Fortunately, however, the radio waves emitted
by molecules are unaffected by interstellar dust, so they
escape easily from dark molecular clouds. Observations
of radio waves from molecules give us a remarkable look
at the innermost workings of the densest and most opaque
interstellar clouds. Among the more important species is
CO. Many astronomers believe that the ratio of CO to H2 is
relatively constant, and because molecular hydrogen is so
difficult to detect even by infrared and radio observations,
carbon monoxide is often used to reveal the amounts and
distribution of molecular hydrogen.
FIGURE 15.10 In visible light, interstellar molecular clouds Molecular clouds have masses ranging from a few times
are seen in silhouette against a background of stars and the mass of our Sun to 10 million solar masses. The smallest
glowing gas. (a) The Horsehead Nebula in Orion is one of the
molecular clouds may be less than half a light-year across,
most identifiable of dark nebulae because of its recognizable
whereas the largest molecular clouds may be over a thou-
shape. The circular halos around the bright stars in this image
are a photographic artifact. (b) Light from background stars
sand light-years in size. The largest molecular clouds qualify
is blocked by dust and gas in nearby Barnard 68, a dense, for the title giant molecular cloud. These behemoths typi-
dark molecular cloud. (c) Infrared wavelengths can penetrate cally have masses of a few hundred thousand times that of
much of this gas and dust, as seen in this false-color image of our Sun and on average are about 120 light-years across. Our
Barnard 68. galaxy contains about 4,000 giant molecular clouds and a
much larger number of smaller ones. Molecular clouds fill
only a tiny fraction of the volume of our galaxy—probably All of these same concepts can be applied to clouds in
only about 0.1 percent of interstellar space. These clouds the interstellar medium. As shown in Figure 15.11, inter-
may be rare, but they are extremely important to our story stellar clouds also have self-gravity: each part of the cloud
because they are the cradles of star formation. feels a gravitational attraction for every other part of the
Having looked at the many different phases of the intercloud. However, interstellar clouds are rarely in hydrostatic
stellar medium, ranging from vast expanses of tenuous equilibrium. In most interstellar clouds, internal pressure is
million-kelvin intercloud gas to the cold dense interiors of much stronger than self-gravity. (Because gravity follows an
molecular clouds, an obvious question to ask is, How does inverse square law, the more spread out an object’s mass is,
it all fit together? Discussing the structure of the interstel- the weaker is its self-gravity.) The outward push of pressure
lar medium is a lot like discussing the nature of weather on in these clouds would cause them to expand if not for the
Earth: it is extremely complex, often difficult to understand, opposing pressure of the surrounding intercloud medium.
and all but impossible to predict precisely. The intercloud medium is much less dense than interstel-
In the interstellar medium, the energy to power the lar clouds, but it is also much hotter, so its pressure is high
“weather” comes from stars. Warm gas heated by ultravio- enough to confi ne the clouds. (Pressure is proportional to
let radiation from massive, hot stars pushes outward into both density and temperature.)
its surroundings. Blast waves from supernova explosions Some molecular clouds, on the other hand, violate hydro-
sweep out vast, hot “cavities,” piling up everything in their static equilibrium in the other direction. These clouds are
path like snow in front of a snowplow. Interstellar clouds massive enough and dense
are destroyed under the onslaught of these violent events, enough— that is, there is Some molecular
but they are also formed by these same forces. Swept-up enough mass packed together clouds collapse under
intercloud gas becomes the next generation of clouds. Hot in a small enough volume of their own weight.
bubbles of high-pressure interstellar gas crush molecular space—that their self-gravity
clouds, driving up their densities and perhaps triggering becomes important. (More mass, along with smaller dis-
the formation of new generations of stars. The interstellar tances between different parts of the cloud, means that
medium is so well stirred that on average, interstellar clouds gravity is stronger.) Furthermore, these clouds are cool
are moving around with random velocities of approximately enough that their internal pressure is relatively low despite
20 kilometers per second (km/s). The fastest motions of inter- their high density. In such clouds, self-gravity is much
stellar material are those of very hot gas and are measured greater than pressure, so the clouds collapse under their
in thousands of kilometers per second. own weight.
The story of the wide range of phenomena that shape If self-gravity in a molecular cloud is much more signifi-
the interstellar medium is fascinating. We could easily stay cant than pressure, we might expect that gravity should win
here and look around for quite some time, but instead we outright, and that the cloud should rapidly collapse toward
must press on in our journey. The time has come to watch its center. In practice, the process goes very slowly because
as a molecular cloud begins the slow process of collapsing several other effects stand in the way of the collapse. One
under its own weight, beginning a chain of events that will
culminate with the ignition of nuclear fi res within a new
generation of stars. FIGURE 15.11 Self-gravity causes a molecular cloud to
collapse.
Parcels of gas within a molecular cloud

feel the gravitational attraction of all
15.3 Molecular Clouds other parts of the molecular cloud…
Are the Cradles of

Star Formation
As we have seen, objects such as planets and stars are held
together by gravity. The mutual gravitational attraction Center of
between each part of a planet or star and every other part of mass of cloud
the object results in a net inward force that pulls all parts of
the object toward its center. If an object is stable, then this
Net gravitational
inward force of the object’s self-gravity must be balanced by force
something— often the outward push of pressure within the
object. In our discussion of planets and stars we referred to …leading to a net gravitational
the balance between gravity and pressure in a stable object force toward the cloud’s center.
as hydrostatic equilibrium.
15.3 Molecular Clouds Are the Cradles of Star Formation 427
such effect that slows the collapse of a cloud is conservation the different parts of the cloud are, on average, only half as
of angular momentum. (We discussed the effects of angular far away from each other as they were when they started.
momentum and the flattening As a result, the gravitational attraction they feel toward
of a collapsing cloud in Chap- Angular momentum, each other will be four times stronger than it was when
ter 6.) Other properties of turbulence, and the cloud was 4 light-years across. When the cloud is 1 ⁄4
molecular clouds that prevent magnetic fields as large as when it started out (or 1 light-year across), the
them from collapsing rapidly slow the collapse force of gravity will be 16 times as strong. As a collapsing
are turbulence and the effects of a molecular cloud, cloud becomes smaller, gravity becomes stronger. As grav-
of magnetic fields. Even but gravity wins ity becomes stronger, the collapse picks up speed. And as
though these effects may slow in the end. the collapse picks up speed, gravity increases even more
the collapse of a molecular rapidly. In other words, the collapse “snowballs.”
cloud, in the end gravity will win. Magnetic fields in the Molecular clouds are never uniform. Some regions
cloud can slowly die away. One part of the cloud can lose within the cloud are invariably denser than others and so
angular momentum to another part of the cloud, allowing collapse within themselves
the part of the cloud with less angular momentum to col- more rapidly than do sur- Stars form in
lapse further. Turbulence ultimately fades away. The details rounding regions. As these molecular-cloud cores.
of these processes are complex and the subject of much cur- regions collapse, their self-
rent research, but the crucial point is this: the effects pre- gravity becomes stronger because they are more compact,
venting the collapse of a molecular cloud are temporary, and so they collapse even faster. Figure 15.12 shows the result
gravity is persistent. As the forces that oppose the cloud’s of this process. What started out as slight variations in the
self-gravity gradually fade away, the cloud slowly becomes density of the cloud grow to become very dense concentra-
smaller. AstroTour: Star Formation tions of gas. Instead of collapsing into a single object, the
molecular cloud fragments into a number of very dense
molecular-cloud cores. A single molecular cloud may form
Molecular Clouds Fragment hundreds or thousands of molecular-cloud cores. It is from
these dense cloud cores, typically a few light-months in
As They Collapse size, that stars may form.
Crucial to the process of star formation is the fact that as As a molecular-cloud core becomes smaller, the gravita-
a cloud becomes smaller, the gravitational forces trying to tional forces that are trying to crush the cloud grow stron-
pull it together grow stronger. This effect is a result of the ger. Eventually, gravity is able to overwhelm the forces due
inverse square law of gravitation discussed in Chapter 3. to pressure, magnetic fields, and turbulence that have been
Suppose a cloud starts out being 4 light-years across. By resisting it. This happens fi rst near the center of the cloud
the time the cloud has collapsed to 2 light-years across, because that’s where the cloud material is most strongly
FIGURE 15.12 As a molecular cloud collapses, denser regions within the cloud collapse more
rapidly than less dense regions. As this process continues, the cloud fragments into a number of
very dense molecular-cloud cores that are embedded within the large cloud. These cloud cores may
go on to form stars.
Molecular-cloud
cores
1 Molecular clouds are never uniform. 2 Slightly denser regions collapse 3 The collapsing cloud fragments
Some regions inside the cloud are faster than their surroundings, and into dense, star-forming cores.
more dense than others. become more pronounced.
1 A molecular-cloud core concentrated. The inner parts of the cloud core start to fall
collapses under its rapidly inward, starting a sequence of events much like a
own weight…
chain of dominoes in which one domino topples the next,
which topples the next. The pressure from the central part of
the cloud core had been supporting the weight of the layers
above it. When the center of the cloud collapses, this sup-
port is suddenly removed. Without this “pressure support,”
the next-outer layer begins to fall freely inward toward the
center as well. But what about the layers farther out? Now
there is nothing to hold them up either, so the process con-
tinues: each layer of the cloud core falls inward in turn,
thereby removing support from the layers still farther out.
2 …until the center of the
core begins falling inward As shown in Figure 15.13, the cloud core collapses like a
faster than the rest of the house of cards when the bottom layer is knocked out. The
cloud can follow. whole structure comes crashing down.
We have been here before. It was at this point in the story
of star formation that our discussion of the formation of the
Solar System in Chapter 6
began. Figure 15.14 shows an Our Solar System
overview of the process. Mate- began in a
rial from the collapsing molec- molecular cloud.
ular-cloud core falls inward.
Because of its angular momentum, this material lands on a
flat, rotating accretion disk. Figure 6.2 shows several such
3 The cloud continues to disks. The dark bands in these images are the disks seen
rapidly collapse from the
inside out… edge-on; the bright regions are starlight reflected from the
surfaces of the disks. Most of this material eventually fi nds
its way inward onto the growing protostar at the center of
the disk, but a small fraction of it remains to become the
stuff planets are made of. The last time we passed this way,
we followed the evolution of the gas and dust left behind
in the disk and saw how it leads naturally to a planetary
system with the properties of our Solar System. This time
through, we will instead follow the story of the protostar
as it becomes a star.
4 …with the falling material
feeding an accretion disk
and growing protostar at
its center.
15.4 The Protostar
Becomes a Star
We pick up the story of our protostar at a time when the
cloud is still collapsing, and more and more material is fall-
ing onto the disk. At this point
the surface of the protostar Protostars are large
has been heated to a tempera- and luminous.
ture of thousands of kelvins
as the gravitational energy of the original molecular cloud
FIGURE 15.13 When a molecular-cloud core gets very has been converted into thermal energy. The protostar is
dense, it collapses from the inside out. Conservation of also huge—hundreds of times larger than our Sun—which
angular momentum causes the infalling material to form an means that the surface of the protostar is tens of thousands
accretion disk that feeds the growing protostar. of times larger than the surface of the Sun. Each square
meter of that enormous surface is radiating energy away in
accordance with the Stefan-Boltzmann law. As a result, the
protostar is thousands of times more luminous than our
1 A giant molecular cloud 2 …and fragments 3 Here one of the molecular

begins to collapse under into small, dense cores continues to collapse
its own weight… molecular cores. under its own weight…
FIGURE 15.14 An overview of how stars like the Sun form,

beginning with the gravitational collapse of a molecular cloud 4 …until the bottom falls out
and ending with the ignition of a star sitting at the center of a and material rains down on
revolving system of planets. a forming accretion disk.
Sun, even though the nuclear reactions that will power it

through its life have yet to begin.
Although the protostar is extremely luminous at this
stage of its life, it probably cannot yet be seen by astrono-
mers in visible light. There are two reasons for this. First,
5 Protostellar winds and
the protostar is relatively cool as stars go, so most of its bipolar jets from the
radiation is in the infrared part of the spectrum. Even more protostar help disperse
important, the protostar begins its life buried deep in the the cloud core…
heart of a dense and dusty molecular cloud. The dust in the
cloud blocks any visible light, so our view of the protostar is
obscured. However, much of the longer-wavelength infrared
light from the protostar is able to escape from the star’s hid-
den cradle. In addition, the dust in the cloud core is heated
by radiation from the star, and this dust also glows in the
infrared. For this reason, protostars are studied mostly in
the infrared part of the spectrum.
The study of protostars and other related young stellar
objects was revolutionized during the 1980s and ’90s as sen-
sitive new instruments capable of “seeing” in the infrared
were trained on regions of known star formation. What we
discovered was that the dark clouds, long familiar to us but
impenetrable until then, are really entire clusters of dense 6 …revealing the new
T Tauri star and its
cloud cores, young stellar objects, and glowing dust when circumstellar disk.
viewed in the infrared. Figure 15.15 shows infrared pictures
of stars forming within columns of gas and dust in the Eagle
Nebula. AstroTour: Star Formation
7 As the protostar contracts,
temperature and pressure in
A Shifting Balance: The the core climb until fusion of
hydrogen begins…
Evolving Protostar
At any given moment, the protostar is in balance, with pres-
sure and gravity successfully opposing each other. However,
this balance is also changing. Material is continuing to fall 8 …while material in the accretion
disk may coalesce to form a
onto the protostar, adding to its mass and gravitational pull planetary system.
and therefore increasing the weight that underlying layers
of the protostar must support. The protostar is also slowly
losing its internal thermal energy by radiating it away. How
can an object be in perfect balance and yet be changing at
FIGURE 15.15 This HST

image of the Eagle Nebula
shows dense columns of
molecular gas and dust at the
edge of an H II region. Infrared
images of the same field, also
taken with the HST, show young
stars forming within these
columns.
HST image in visible

light shows columns of
molecular gas illuminated
by nearby stars. When viewing in infrared
radiation, we see into the dusty
columns where new stars are
forming from interstellar gas.
the same time? To answer this question, consider a more of pressure (Figure 15.16b). But its balance, too, is changing
everyday example. Figure 15.16a shows a simple spring bal- with time. Just like sand that has fallen onto the spring bal-
ance, which works on the principle that the more a spring is ance, material that has fallen onto the protostar compresses
compressed, the harder it pushes back. If an object is placed the protostar more and more, and the protostar evolves.
on the balance, it compresses the spring until the downward As material continues to fall onto the protostar, it adds
force of the weight of the object is exactly balanced by the weight to the outer layers of the protostar. This growing
upward force of the spring. We measure the weight of the weight compresses the material in the interior of the proto-
object by determining the point at which the pull of gravity star, and as it is compressed, the protostar’s interior grows
and the push of the spring are equal. The situation is analo- hotter. (Compressing a gas always causes it to heat up, and
gous to our protostar, but in the protostar the pressure of letting a gas expand always causes it to cool down. The cool-
the gas takes the place of the spring. (The analogy is a good ing systems in your air conditioner and refrigerator work
one because compressing the gas in the protostar causes its by compressing gas to make it hot, then letting this hot gas
pressure to increase, just as compressing the spring causes cool so that when it expands again it gets really cold.) If
the spring to push back harder.) the interior of the protostar is now denser and hotter, the
Let’s now slowly pour sand onto our spring balance. At pressure is also higher—just enough higher to balance the
any point the downward weight of the sand is balanced by increased weight of the material above it. Balance is always
the upward force of the spring. As the weight of the sand maintained.
increases, the spring is slowly compressed. The spring and Even after material stops falling on the protostar, the pro-
the weight of the sand are always in balance, but this bal- tostar keeps evolving in much the same way. Earlier we saw
ance is changing with time as more and more sand is added. that the protostar is radiating away many times as much
In just the same way, the protostar is always in balance, with energy as the Sun. The source of the energy being radiated
the inward force of gravity matched by the outward force away is the thermal energy trapped in the interior of the
protostar—the same thermal energy that is responsible for ing hotter and hotter, until the center of the star is finally hot
supporting the protostar against the forces of gravity. So, as enough for hydrogen to begin fusing into helium.
thermal energy leaks out of the protostar, gravity gains the Something strange is going on here. When an object
upper hand and the protostar radiates energy, it normally gets cooler. When you turn off
slowly contracts. As the proto- The protostar radiates the electric coil on your stove, the coil cools as it loses the
star shrinks, the forces of grav- away thermal energy thermal energy within it. Yet we just concluded that as a
ity become greater (because and shrinks. protostar radiates thermal energy away, it actually grows
the parts of the protostar are hotter. How can that be? Once again the answer lies with
closer together—remember the inverse square law of gravity). the concept of conservation of energy. As the protostar con-
Meanwhile, as the protostar shrinks, its interior is com- tracts, every part of the protostar is slowly falling toward
pressed, so it grows hotter. As density and temperature climb, its center, which means that the protostar is losing gravita-
so does the pressure—just enough to continue to counteract tional energy. This gravitational energy has to show up in
the growing force of gravity. Balance between gravity and another form, which in this case is thermal energy. Some
pressure is always maintained. This process continues, with of this thermal energy is radiated away, but not all of it— so
the star becoming smaller and smaller and its interior grow- the interior of the star grows hotter.
VISUAL ANALOGY FIGURE 15.16 (a) A spring balance comes to rest at the point where the downward
force of gravity is matched by the upward force of the compressed spring. As sand is added, the location
of this balance point shifts. (b) Similarly, the structure of a protostar is determined by a balance between
pressure and gravity. Like the spring balance, the structure of the protostar constantly shifts as the
protostar radiates energy away and additional material falls onto its surface.
(a) (b) 2 Likewise, the gravitational force

pulling material toward the center
of a protostar is exactly balanced
by the outward force of pressure.
Force of Sand
spring
Weight
Infalling
material
1 The position of the tray is set

by a balance between weight
and the force of the spring.
Force of
spring
Weight 4 Likewise, as more material falls on the

protostar, and as heat from its interior
radiates away, the protostar becomes more
compact. Pressure in the protostar increases.
3 As weight is added, the
balance position shifts as
the spring is compressed.
1 Thermal energy escapes from the interior

of a protostar and is radiated into space.
The protostar contracts.
Pressure
Gravity 2 Gravity is stronger in the
smaller protostar…
Pressure
Gravity
A protostar’s
luminosity comes
from gravitational 3 …pushing up pressure
collapse. and temperature.
8 The new star settles down into 4 The cycle continues. As

its main-sequence life, burning the protostar gets smaller
hydrogen in its core. and smaller…
Ignition!
7 …until the core
gets so hot that
hydrogen begins
fusing into helium.
Pressure
Gravity
6 In this way the 5 …its interior gets

protostar’s gravitational hotter and hotter.
energy is converted
into thermal energy…
FIGURE 15.17 As a protostar radiates away energy, it loses pressure support in its interior and contracts.
This contraction drives the interior pressure up. The counterintuitive result is that radiating energy away
causes the interior of the protostar to grow hotter and hotter until nuclear reactions begin in its interior.
Figure 15.17 shows this chain of events in a nutshell. The smallest known brown dwarf has a mass of about 8 MJ
The protostar radiates away thermal energy and, in the pro- and a temperature of about 1,350 K.
cess, loses pressure support. Since the fi rst brown dwarfs were identified in the mid-
Deprived of pressure support, Fusion begins, and 1990s, many hundreds have been found. The cooler among
the protostar contracts. As the the protostar them have methane and ammonia in their atmospheres,
protostar contracts, gravita- becomes a star. similar to what we find in the atmospheres of the giant plan-
tional energy is converted to ets of the Solar System. Winds can be very high, though,
thermal energy, driving up the temperature in the proto- producing weather far more violent than storms observed
star’s interior. If the protostar is massive enough, the tem- in the atmospheres of our own giant planets.
perature in its interior will eventually become hot enough
for nuclear fusion to begin. This is the point at which the
transition from protostar to star takes place. The distinction
between the two is that a protostar draws its energy from
Evolving Stars and Protostars
gravitational collapse, whereas a star draws its energy from Follow “Evolutionary Tracks”
thermonuclear reactions in its interior.
The protostar’s mass determines whether it will ever
on the H-R Diagram
become a star. As the protostar slowly collapses, the tem- Within the evolving protostar it is convection rather than
perature at its center gets radiation that carries energy outward, keeping the proto-
higher and higher. If the pro- At least 0.08 M⊙ is star’s interior well stirred. Interestingly, although the inte-
tostar is more massive than needed for a protostar rior of the protostar grows hotter and hotter as it contracts,
about 0.08 times the mass of to become a star. its surface stays at about the same temperature through most
the Sun (0.08 M⊙), the temper- of this phase of its evolution. This distinction is important.
ature in its core will eventually reach the 10-million-K mark, The surface temperature of a star or protostar is not the
and fusion of hydrogen into helium will begin. At this onset same as the temperature deep in its interior. For example,
of fusion, the newly born star will once again adjust its in Chapter 14 we found that the temperature of the surface
structure until it is radiating energy away from its surface of the Sun is about 5,780 K while the temperature of its
at just the rate that energy is being liberated in its interior. interior is millions of kelvins. As the protostar contracts,
As it does so, it “settles” onto the main sequence of the H-R the temperature deep within the star grows hotter and hot-
diagram, where it will spend the majority of its life. ter, but the temperature of the star’s surface remains nearly
If the mass of the protostar is less than 0.08 M⊙, it will constant.
never reach the point where nuclear burning takes place. In the 1960s the Japanese theoretician Chushiro Hayashi
Such a failed star is called a explained why this is so. Hayashi pointed out that the
brown dwarf. A brown dwarf Brown dwarfs are atmospheres of stars and protostars contain a natural ther-
is neither star nor planet, but failed stars. mostat: the H– ion. (An H–, or “H minus,” ion is a hydrogen
something in between—what atom that has acquired an extra electron and therefore has
we might call a “substellar object.” The International Astro- a negative charge.) The amount of H– in the atmosphere of
nomical Union (IAU) sets the boundary between a brown a protostar is highly sensitive to the temperature at the sur-
dwarf and a supermassive giant planet at 13 Jupiter masses face of the protostar. The cooler the atmosphere of a star,
(MJ), although there is still some debate as to whether this the more slowly atoms and electrons are moving, and the
is really the appropriate criterion. easier it is for a hydrogen atom to hold on to an extra elec-
A brown dwarf forms the same way a star forms, yet in tron. As a result, the cooler the atmosphere of the star, the
many respects it is like a giant Jupiter. Brown-dwarf can- more H– there is.
didates are seen as especially cool M stars, and stars that The H– ion, in turn, helps control how much energy is
are even cooler than spectral type M.1 A brown dwarf never radiated away by a star or protostar. The more H– there is
burns hydrogen in its core but instead glows by continually in the atmosphere of a star or
cannibalizing its own gravitational energy.2 As the years protostar, the more opaque the H– is a stellar
pass, a brown dwarf gets progressively smaller and fainter. atmosphere is, and the more thermostat.
effectively the thermal energy
1
of the protostar is trapped in its interior. Imagine that the
Relatively new spectral types, L and T, have been added to represent
surface of the protostar is too cool. “Too cool” means that
stars that are even cooler than M-type stars.
2 extra H– forms in the atmosphere, making the atmosphere
The cores of brown dwarfs can get hot enough to burn deuterium,
a heavy isotope of hydrogen; and those with a mass greater than 65
of the protostar more opaque. The atmosphere thus traps
Jupiters can burn lithium. But both of these energy sources are in more of the radiation that is trying to escape, and the trapped
very short supply, and after a brief period of deuterium or lithium energy heats up the star. As the temperature climbs, H– ions
fusion, brown dwarfs shine only by the energy of their own gravita- are destroyed (that is, changed to neutral H atoms). Now
tional contraction. imagine the other possibility—that the protostar is too hot.
Then, H– in its atmosphere will be destroyed, so the atmo- and age of the protostar. Because the temperature of the
sphere will become more transparent, allowing radiation star is not changing much, the amount of energy per unit
to more freely escape from the interior. Because the proto- time (power) radiated away by each square meter of the
star cannot hold on to enough of its energy to stay warm, surface of the star does not change much either. Recall the
the surface cools. In either case—too cold or too hot—H– is Stefan-Boltzmann law from Chapter 4, which says that the
formed or destroyed until the star’s atmosphere once again amount of power radiated by each square meter of the star’s
traps just the right amount of escaping radiation. surface is determined by its temperature. But as the star
The H– ion is basically doing the same thing that you shrinks, the area of its surface shrinks as well. There are
do with your bedcovers at night. If you get too cold, you fewer square meters of surface to radiate, so the luminos-
pile on extra covers to trap your body’s thermal energy and ity of the protostar drops. As viewed from the outside, the
keep you warm (more H– ions). If you get too hot, you kick protostar stays at nearly the same temperature and color
off some covers to cool off (fewer H– ions). It is a shame that but gradually gets fainter as it evolves toward its eventual
we have no such “automatic cover” to maintain our body life as a main-sequence star.
temperature at night as effectively as the H– ion controls the In Chapter 13 we introduced the H-R diagram and used
surface temperature of a protostar. it to begin to understand how the properties of stars dif-
The amount of H– in the atmosphere of a protostar fer. For the next several chapters we will also use the H-R
keeps the surface temperature of the protostar somewhere diagram to keep track of how stars change as they evolve
between about 3,000 and 5,000 K, depending on the mass through their lifetimes. The path across the H-R diagram
FIGURE 15.18 (a) The evolution of pre–main sequence stars can be followed on the H-R diagram.
Protostars in the upper right portion of the diagram are large and cool. (b) The roughly vertical,
constant-temperature parts of the evolutionary tracks of low-mass protostars are referred to as
Hayashi tracks.
107 Protostars of different masses follow different

paths on their way to the main sequence.
106 60 M£ 1,0 –10
00
20 M£ R
£
105 As a 1-M£ protostar
descends its Hayashi track,
10 M£ the H– thermostat keeps
104 M –5 its surface temperature
AI
N almost constant.
SE 5 M£
103 Q
UE
Luminosity (L£)
NC
E 3 M£
102 0
1R
£
101 2 M£
0.1
1 R £ +5
£
1 M£
10–1 0.0 0.8 M£

1R
£
10–2 +10
0.0
01
R£
10–3
40,000 30,000 10,000 6,000 3,000
MA
IN
O5 B0 B5 A0 F0 G0 K5 M5
Spectral type
SE Hydrogen
QU begins
EN to burn.
CE
that a star follows as it goes through the different stages to perhaps 100 times the mass of the Sun. What determines
of its life is called the star’s evolutionary track (Figure how massive a star will be? The answer to this question is
15.18a). The particular path that an evolving protostar unclear and is a topic of a great deal of ongoing research.
follows as it approaches the main sequence is called its One obvious possibility is that a forming star grows until it
Hayashi track (Figure 15.18b). The protostar is brighter uses up all of the gas around it; it becomes no larger simply
than it will be as a true star on the main sequence, so because it has run out of material. Although this explana-
a protostar’s Hayashi track is located above the main tion would be easy to understand, it does not match our
sequence on the H-R diagram. Because the surface tem- observations of how stars actually form. At this point in
perature of the protostar stays nearly constant as the pro- our story the contracting protostar is at the center of a col-
tostar contracts, the protostar’s Hayashi track prior to the lapsing molecular-cloud core, which in turn is a denser-
start of hydrogen burning is an almost vertical line on the than-average region inside a molecular cloud whose total
H-R diagram. Figure 15.18 shows the pre–main sequence mass may be hundreds of thousands of times greater than
evolutionary tracks of stars of several different masses. the mass of the Sun. Observations indicate that under most
Astronomers say that an evolving protostar “descends the circumstances star formation is a very inefficient process.
Hayashi track.” Only a small fraction of the material in a molecular cloud—
perhaps a few percent— ends up as part of the stars forming
within it. Something must prevent most of the material in a
molecular cloud from ever actually falling onto protostars.
Not All Stars Are Created Equal There are a number of ideas about what this something
In Chapter 13 we found that stars can have a wide range of might be. One intriguing possibility is that forming stars
masses, varying from less than 1 ⁄10 the mass of the Sun up control their own masses.
FIGURE 15.19 (a) Material falls onto an accretion disk around a protostar and then moves inward, eventually
falling onto the star. In the process, some of this material is driven away in powerful jets that stream perpendicular
to the disk. (b) This infrared Spitzer Space Telescope image shows jets streaming outward from a young,
developing star. Note the nearly edge-on, dark accretion disk surrounding the young star.
(a) (b)
Material moves in Bipolar jets and outflows

toward the protostar flow away from the young
in the accretion disk. star and disk.
Wind
Jets
This structure is seen in

images of forming stars.
(a) (b) (c)
Bipolar jets of material

stream outward
perpendicular to the disk
around a forming star.
(d)
When the jets hit

interstellar material,
they form glowing
bow shocks called
HH objects.
FIGURE 15. 20 (a) Artist’s view of jets from a protostar slamming into surrounding interstellar
gas, heating the gas and causing it to glow. (b) This HST image shows the bow shocks formed at the
ends of a bipolar jet from a protostar. Enlargements of the bow shock (c) and jet (d) are shown at
right. Only one side of the jet itself is visible because the other side is hidden behind the dark cloud in
which the star is embedded.
When we observe young stellar objects, we often fi nd material in these jets flows out into the interstellar medium,
clear signs that material is falling onto the central proto- where it heats, compresses, and pushes away surrounding
star and accretion disk, as discussed earlier. However, it is interstellar gas. Knots of glowing gas accelerated by jets are
also common to observe powerful flows of material moving referred to as Herbig-Haro objects (or HH objects for short),
away from forming stars at the same time that material is named after the two astronomers who fi rst identified them
being accreted onto the stars. How can this be? As Figure and associated them with star formation. Figure 15.20 shows
15.19a shows, material falls onto the accretion disk and a Hubble Space Telescope image of the fi rst HH objects dis-
moves inward toward the equator of the star, while at the covered: HH 1 and HH 2. These two HH objects are the two
same time other material is blown away from the protostar sides of the bipolar outflow from a single source.
and disk in the two opposite directions away from the plane The origin of outflows from protostars is not as well
of the disk. The resulting stream of material away from the understood as we would like, but current models suggest
protostar is called a bipolar outflow. that they are the result of magnetic interactions between
Some bipolar outflows from young stellar objects are the protostar and the disk. The interior of a protostar on its
slow and fairly disordered, but others produce remarkable Hayashi track is convective. Great cells of hot gas are rising
jets of material that move from the interior of the star, while other cells of cooler gas
away from the central proto- Protostars drive are falling toward the center. This convection, coupled with
star and disk at velocities of powerful bipolar the protostar’s rapid rotation, can lead to the formation of a
hundreds of kilometers per outflows. dynamo, similar to the dynamo that drives the Sun’s mag-
second (Figure 15.19b). The netic field. The dynamo in the center of a protostar would
(a)
FIGURE 15. 2 1 If you
have good eyes (or a pair
of binoculars), on a clear
winter night you can see the
brightest of the stars in this
cluster— a tight bunch in the
constellation Taurus called the
Pleiades or Seven Sisters. The
diffuse blue light around the
stars is starlight scattered by
interstellar dust. (b) A Spitzer
infrared image of the Pleiades
shows more detail in the
structure of the interstellar
dust through which the cluster
is traveling.
(b)
view of the protostar: the protostar is “revealed.” Once the

contracting protostar makes its appearance, it is referred
to as a T Tauri star. This name comes from the fi rst recog-
nized member of this class of objects, the star labeled T in
the constellation Taurus.
Astronomers have long known that stars are often found
together in closely knit collections called star clusters. Fig-
ure 15.21 shows one such star cluster: a group called the
Pleiades or Seven Sisters. Star clusters gave astronomers
their fi rst evidence that many stars of all different masses
can form together at the same place and at about the same
time. When we look around the galaxy at large, we see a
hodgepodge of stars— some very old and others very young.
If these were the only stars we had to study, it would be
extremely difficult to learn much about how stars evolve.
Star clusters, on the other hand, are large collections of
stars that all formed at the same time, in the same place,
and from the same material. They provide us with tailor-
made samples to study star formation.
Even though the few brightest and most massive stars
in a cluster dominate what we see, the vast majority of
stars in a cluster are much less massive, like our Sun.
be much more powerful than the Sun’s dynamo, however. In fact, some star-forming regions do not seem to form
The protostar’s resulting strong magnetic field might cause any especially massive stars at all. Astronomers are very
the protostar to begin blowing a powerful wind. It might interested in how and why molecular clouds subdivide
also act something like the blade in a blender, tearing at the themselves into low- and high-mass stars. The details of
inner edge of the accretion disk and flinging material off this division— specifically, what fraction of newly formed
into interstellar space. stars will be of what masses— are crucial if we are to use
Powerful protostellar winds, jets, and other outflows observations of the stars around us today to untangle the
could disrupt the cloud core and accretion disk from which history of star formation in our galaxy. Unfortunately,
the protostar formed, shutting down the flow of material onto we are still far from a detailed understanding of why
the protostar. Until the protostellar wind begins, the proto- some cloud cores become 1-M⊙ stars while others become
star is enshrouded in the dusty molecular-cloud core from 10-M⊙ stars.
which it was born. As the wind from the protostar disperses After a cloud core collapse, the evolution of a protostar
this obscuring envelope, we get our fi rst direct, visible-light is determined largely by its mass. Calculations suggest that
a star with the mass of our Sun probably takes about 10 mil- At what point during star formation is it determined that
lion years or so to descend its Hayashi track and become a a collapsing cloud core will form several stars instead of
star on the main sequence. If we look at the entire history, just one? Some models suggest that this split may happen
including the collapse and fragmentation of the molecular early in the process, during the fragmentation and collapse
cloud itself, the total time is of the molecular cloud. The
probably more like 30 million Star formation may advantage of these ideas is Collapsing cloud
years. More massive stars go take millions of years. that they provide a natural cores often form
through this process much way of dealing with much of multiple stars.
faster. A 10-M⊙ star might go from the stage of being a molec- the angular momentum of the
ular-cloud core to burning hydrogen in its interior in only cloud core: it goes into the orbital angular momentum of
100,000 years. A 100-M⊙ star might make the journey in less the stars about each other. Other models suggest that addi-
than 10,000 years. By comparison, a 0.1-M⊙ star might take tional stars may form from the accretion disk around an
100 million years to fi nally reach the main sequence. initially single protostar.
The 30 million years or so that it took for our Sun to The picture of star formation presented here is remark-
form is a long time, but it is a tiny fraction of the 10 billion ably complete, considering that we have never visited a
years during which the Sun will steadily fuse hydrogen protostar or watched a star form. Instead, astronomers have
into helium as a main-sequence star. It is no wonder that so observed many different stars at different stages in their
few among the many stars we see in the sky are such young formation and evolution, and then used their knowledge
objects. But every star that we see was young at one time, of physical laws to tie these observations together into a
including our own Sun. coherent, consistent description of how, why, and where
Not surprisingly, many questions about star formation stars form. This two-pronged attack—using observations to
remain. For example, how must we modify our story to see what things exist in the universe, and using physics to
account for the formation of binary stars or other multiple- understand how they work and the relationships between
star systems? When we observe the sky, we fi nd that about them—is how all astronomy (and in a certain sense all
half of the stars we see are part of multiple-star systems. physical science) works.
subtle ecosystem. It is one thing to catalog the inhabit-

Seeing the Forest for the Trees ants of an ecosystem, but it is quite another to under-
stand the interrelationships and dependencies among
We began our discussion of the Solar System in Chapter those organisms well enough to say that you know how
6 by describing its formation. Nearly 5 billion years ago the ecosystem works. In like fashion, astronomers at the
a vast interstellar cloud collapsed under its own weight turn of the 21st century have a fairly complete picture
to form a swirling disk of gas and dust. Within that disk, of what makes up the interstellar medium, but we do
small objects stuck together to become parts of larger not fully understand the complex interplay between
objects, culminating in the formation of the planets and the phases of the interstellar medium and the stars that
other solid bodies of our Solar System. We now can see are born and die there. Even so, we have learned some
that earlier story— and in some sense our entire discus- important lessons and have developed some important
sion of the Sun, Earth, and Solar System— as a “side- tools. For example, 21-cm radiation from ubiquitous
bar” to a larger story, the thread of which we picked up interstellar hydrogen— radiation that easily penetrates
again on this leg of our journey. The tenuous expanse the dust that obscures our view of the universe at vis-
of gas and dust that fi lls the vast reaches of interstellar ible wavelengths— gives us a tool to see out to the far
space sets the stage for the ongoing formation of genera- reaches of our galaxy and map out our home in the
tions upon generations of stars and planetary systems, universe.
of which ours is but one among thousands of billions of Fortunately, our understanding of stars is far more
billions in the universe. complete. In fact, the workings of stars represent such
The interstellar medium is a difficult topic, even a nicely posed physics problem that we would probably
for experts in the field. The time that we have for our include them on our journey even if this were a pure
journey is limited. To go beyond the basic description physics text rather than an astronomy text. Many of the
of the interstellar medium here would be a long and physical principles employed in our discussion of star
arduous excursion. It would also quickly move on to formation and the evolution of protostars in this chap-
more speculative ground. For now we suggest you think ter should have sounded familiar. The balance between
of the interstellar medium as you might a complex and gravity and pressure that gives a protostar its structure
is the same balance at work within the Sun. The chain within their hearts, and when they exhaust their fuel
of physical reasoning that we used to understand the they die. The changing balance of the protostar is only
collapse of a protostar (gravitational energy that is con- the fi rst chapter in a process of evolution that continues
verted to thermal energy and then radiated away) applies throughout the star’s life. We found it convenient to fol-
almost unmodified to understanding the ongoing (albeit low the changes taking place within an evolving proto-
slow) collapse of Jupiter. star by tracking its progress across the face of the H-R
As we go forward on our journey, the physical insight diagram. In doing so, we discovered the best way that
that we have developed will continue to serve us well. astronomers have found to draw a road map of a star’s
Stars are temporal objects. They are born from inter- life. That road map will be our guide for the next stretch
stellar gas, they shine by the light of nuclear fi res deep of our journey.

• The interstellar medium is a complex region, ranging THINK ING A BOUT THE CONCEP T S
from cold, relatively dense molecular clouds to hot,
1. The interstellar medium is approximately 99 percent gas
tenuous intercloud gas.
and 1 percent dust. Why is it the dust and not the gas that
• Dust and gas in the interstellar medium blocks visible blocks our visible-light view of the galactic center?
light but becomes more transparent at longer, infrared
*2. In Chapter 13 we learned that we can measure the tem-
wavelengths.
perature of a star by comparing its brightness in blue and
• Neutral hydrogen cannot be detected at visible and yellow light. Does reddening by interstellar dust affect
infrared wavelengths, but it is revealed by its 21-cm the temperature we measure of a star? If so, how?
microwave emission.
3. Using your own experience of walking or riding in a
• Stars form in clusters within fragmented collapsing dense fog, explain how the existence of large amounts
cores inside giant molecular clouds. of interstellar dust may have influenced the views held
by 19th century astronomers regarding the structure of
• Our Solar System began within a molecular-cloud
our galaxy.
core.
4. Explain how the development of infrared astronomy has
• Protostars glow with thermal energy released from con-
made it possible for astronomers to study the detailed
verted gravitational energy as they collapse.
processes of star formation.
• A protostar lands on the main sequence when nuclear
5. How does the material in interstellar clouds and inter-
reactions begin in its core.
cloud gas differ in density and distribution?
• A protostar must have at least 0.08 M⊙ to become a true
6. When a star forms inside a molecular cloud, what hap-
star.
pens to the cloud? Is it possible for the molecular cloud
• Brown dwarfs are neither stars nor planets, but some- to remain cold and dark with one or more stars inside
thing in between. it? Explain your answer.
• What we know about the evolution of the birth of stars 7. If you placed your hand in boiling water (100°C) for even
comes from observations of many protostars at various one second, you would get a very serious burn. If you
stages of their development. placed your hand in a hot oven (200°C) for a second or
two, you would hardly feel the heat. Explain this dif-
ference and how it relates to million-kelvin regions of
SmartWork, Norton’s online homework system, the interstellar medium.
8. What is the “local bubble” and how was it formed?
your instructor assigns questions in SmartWork, log in at 9. What is the primary source of heating of hot intercloud
smartwork.wwnorton.com. gas?
10. Interstellar gas typically cools by colliding with other ter? Would you classify a brown dwarf as a supergiant
gas atoms or grains of dust; during the collision, the atom planet? Explain your answer.
loses energy and hence its temperature is lowered. How
*22. The H– ion acts as a thermostat in controlling the surface
does this effect explain why tenuous gases are generally
temperature of a protostar. Explain how.
so hot, while dense gases tend to be so cold?
*23. The Hayashi track is a nearly vertical (constant-temper-
11. Stellar radiation can convert atomic hydrogen (H I) to
ature) evolutionary track on the H-R diagram. Why?
ionized hydrogen (H II).
a. Why does a B8 main-sequence star ionize far more 24. What do astronomers mean by bipolar outflow?
interstellar hydrogen in its vicinity than does a K0
25. Describe a Herbig-Haro object.
giant of the same luminosity?
b. What properties of a star are important in determin- 26. What distinguishes a T Tauri star from a protostar?
ing whether it can ionize large amounts of nearby
27. Why are star clusters so important to our understanding
interstellar hydrogen?
of how stars of different mass evolve?
12. When a hydrogen atom is ionized, a single particle
28. Of the many stars that astronomers see in the sky, why
becomes two particles.
are so few of them protostars?
a. Identify the two particles.
b. If both particles have the same kinetic energy, which 29. What is the single most important property of a star that
moves faster? will determine its evolution?
*13. What causes a hydrogen atom to radiate a photon of 21-cm
radio emission?
*14. Explain how the important discovery of 21-cm radio
emission has enabled us to detect interstellar clouds 30. Estimate the typical density of dust grains (grains per
of neutral hydrogen (H I), even when large amounts of cubic centimeter) in the interstellar medium. A typical
interstellar dust are in the way. grain has a mass of about 10 –17 kg. (Hint: You know the
typical density of gas, and what fraction of the interstel-
15. Molecular hydrogen is very difficult to detect from the
lar medium, by mass, is made of dust.)
ground, but we can easily detect carbon monoxide (CO)
by observing its 2.6-cm microwave emission. Describe 31. Using Figure 15.2, estimate the blackbody temperature of
how observations of CO might help astronomers infer the star as shown in part (b) (without dust) and part (c)
the amounts and distribution of molecular hydrogen (with dust). How significant are the effects of interstel-
within giant molecular clouds. lar dust when we are using observed data to determine
the properties of a star?
16. Our galaxy contains several thousand giant molecular
clouds. Describe a giant molecular cloud and the role it 32. A typical temperature of intercloud gas is 8,000 K. Using
plays in star formation. Wien’s law (see Chapter 4), calculate the wavelength at
which this gas would radiate.
17. What two forces establish hydrostatic equilibrium in an
evolving protostar? 33. The mass of a proton is 1,850 times the mass of an elec-
tron. If a proton and an electron have the same kinetic
18. As a cloud collapses to form a protostar, the forces
energy, EK = 1 ⁄2 mv2, how many times greater is the veloc-
of gravity felt by all parts of the cloud (which follow
ity of the electron than that of the proton?
an inverse square law) become stronger and stronger.
One might argue that under these conditions, the cloud 34. If a typical hydrogen atom in a collapsing molecular-
should keep collapsing until it has zero size (a black cloud core starts at a distance of 10,000 astronomical
hole). Why doesn’t this happen? units, or AU (1.5 × 1012 km) from the core’s center and
falls inward at an average velocity of 1.5 km/s, how many
*19. The internal structure of a protostar maintains hydro-
years does it take to reach the newly forming protostar?
static equilibrium even as more material is falling onto
Assume that a year is 3 × 107 seconds.
it. Explain how this can be.
35. From Table 13.1 we can see that the ratio of hydrogen
20. What determines whether a protostar will become a true
atoms (H) relative to carbon atoms (C) in the Sun’s atmo-
star or a brown dwarf?
sphere is approximately 2,400 to 1. It would be reason-
21. You can think of a brown dwarf as a failed star—that able to assume that this ratio also applies to molecular
is, one lacking sufficient mass for nuclear reactions to clouds. If 2.6-cm radio observations indicate 100 M⊙
begin. What similarities and differences do you see of carbon monoxide (CO) in a giant molecular cloud,
between a brown dwarf and a giant planet such as Jupi- what is the implied mass of molecular hydrogen (H2)
in the cloud? (Carbon represents 3 ⁄7 of the mass of a CO a. Compare the surface area of the protostellar Sun with
molecule.) today’s Sun.
b. Using the Stefan-Boltzmann law, compare the lumi-
36. Neutral hydrogen emits radiation at a radio wavelength
nosity of the protostellar Sun with today’s Sun. (Note:
of 21 cm when an atom drops from a higher-energy mag-
The surface temperature of the Sun is 5,780 K.)
netic state to a lower-energy magnetic state. On average,
each atom remains in the higher-energy state for 11 mil- *39. A protostar with the mass of the Sun starts out with a
lion years (3.5 × 1014 seconds). temperature of about 3,500 K and a luminosity about
a. What is the probability that any given atom will make 200 times larger than the Sun’s current value. Estimate
the transition in 1 second? this protostar’s size and compare it to the size of the Sun
b. If there are 6 × 1059 atoms of neutral hydrogen in a today.
500-M⊙ cloud, how many photons of 21-cm radiation
40. The star-forming region 30 Doradus is 160,000 light-
will the cloud emit each second?
years away in the Large Magellanic Cloud and appears
c. How does this number compare with the 1.8 × 1045
about 1 ⁄6 the brightness of the faintest stars visible to the
photons emitted each second by a solar-type star?
naked eye. If it were located at the distance of the Orion
37. The Sun took 30 million years to evolve from a collaps- Nebula (1,300 light-years) how much brighter than the
ing cloud core to a star, with 10 million of those years faintest visible stars would it appear?
spent on its Hayashi track. It will spend a total of 10 bil-
lion years on the main sequence. Suppose we were to
compress the Sun’s main-sequence lifetime into just a
single day.
a. How long would the total collapse phase last?
b. How long would the Sun spend on its Hayashi
track?
*38. Assume that the Sun, starting out as a protostar, had a
diameter 100 times that of the present-day Sun and a
surface temperature of 3,300 K.
It is said an Eastern monarch once charged his wise men to
invent him a sentence to be ever in view, and which should
be true and appropriate in all times and situations. They
presented him the words “And this, too, shall pass away.”
ABRAHAM LINCOLN (1809 –1865), SEP T EMBER 30, 1859
Shells of dust reflect successive outbursts of the supergiant star V838 Monocerotis.
CHAP TER 16
Stars in the Slow Lane—

Low-Mass Stellar Evolution
16.1 This, Too, Shall K E Y CONCEP T S

Pass Away Within its core, the Sun fuses over 4 billion kilograms of
hydrogen to helium each second; and although the Sun
If you go outside tonight and look up, you will not see the may seem immortal by human standards, eventually it
same sky that you would have seen a year ago. The stars will run out of fuel. When it does, some 5 billion years
themselves have not changed, but the way you see them has from now, the Sun’s time on the main sequence will
changed. A year ago the sky was probably full of points of come to an end. As we examine what happens when a
light, called stars, that were “somewhere off in space.” Now low-mass star like the Sun nears the end of its life, we
you see an expanse of distant suns, each farther away than
will find that
the mind can easily comprehend, and each shining with a
devastating brilliance powered by a nuclear inferno deep • The more massive a star is, the shorter is its lifetime.
in its heart. When you look at the dark splotches that mar • When the Sun runs out of fuel at its center, it will grow
the Milky Way’s eerie glow, you see vast clouds of interstel- into a larger, more luminous red giant star.
lar gas and dust. In your mind’s eye you look inside these
• We can follow the post–main sequence evolution of
clouds at the new suns and new solar systems being born
stars by tracing their paths on the H-R diagram.
there, and imagine the birth of our own planet and the star
we orbit, 4.6 billion years ago. The sky is like a page from a • Red giants, and some other evolved stars, are shaped
book. To those who cannot read, written words are so many by dense “degenerate” cores in which atoms have been
chicken scratches; but to those who can, thoughts and words crushed by gravity.
and ideas spring off the page to touch our hearts and stimu-
• Evolving low-mass stars go through a series of stages,
late our minds and imaginations. The sky is a book worth
eventually burning helium to carbon and building up a
reading, as is the rest of nature. At this stage in our journey
core of carbon ash.
we are well on our way to becoming literate.
So far on our journey, we have seen interstellar clouds • In the end, a low-mass star will eject its outer layers,
collapse under the force of gravity to form immense pro- possibly forming a planetary nebula, and leaving behind
tostars surrounded by swirling disks of gas and dust. We a tiny degenerate white dwarf.
have watched as dust left behind in the disk accumulates
• Low-mass stars in binary systems may experience
into planets and moons, asteroids and comets. We have
more exciting fates as novae or supernovae.
followed along as protostars continue their collapse until
the nuclear fuel in their cores ignites, and we have recog-
nized a grand pattern—the main sequence—in the stars
that those protostars go on to become. We have even looked
443
444 Chapter 16 Stars in the Slow Lane—Low-Mass Stellar Evolution
inside one of these stars, our Sun, and come to appreciate stars, which have masses less than about 3 M⊙, are referred
the battle between gravity and pressure that gives our Sun to as low-mass stars and are typified by our Sun. In this
its structure. Our understanding of that battle provides chapter we begin our discussion of stellar evolution by
our understanding of all stars along the main sequence. A examining the stages through which low-mass stars prog-
star’s mass determines the strength of its gravity, and the ress. What better place to start than by asking what fate
need to balance that gravity in turn determines the pres- awaits our own Sun?
sure in the star’s interior. The more massive the star, the
higher the pressure that is needed to hold it up, and the
more rapidly the star must burn its nuclear fuel to support
its own weight. From luminous O stars on the hot end of 16.2 The Life and Times of
the main sequence to faint M stars on the cool end, mass
is the fundamental and overriding property that makes a a Main-Sequence Star
main-sequence star what it is.
We have seen stars born and seen how they live. Now In Chapter 14 we learned that the structure of the Sun is deter-
the time has come to watch them die. Like all main-sequence mined by a balance between the inward force of gravity and
stars, the Sun gets its energy the outward force of pressure. The pressure within the Sun
by converting hydrogen to Stars eventually is, in turn, maintained by energy released by nuclear fusion
helium in its core. This is exhaust their in the heart of the star. If you were to add mass to the Sun,
what defines the main nuclear fuel. the weight of material pushing down on the inner regions of
sequence: being on the main the star would increase. Gravity would gain an advantage.
sequence means that a star is burning hydrogen in its core. The inner parts of the Sun would be compressed by the added
But a star cannot remain a main-sequence star forever. weight, driving up the temperature and density there. This
Hydrogen at the core of a star is a consumable resource. Any increase in temperature and density would in turn accelerate
star eventually exhausts this resource—it “runs out of the pace of the nuclear reactions occurring there.
gas”—and when it does, its structure begins to change dra- We have followed this chain of reasoning before, but it
matically. Just as the balance between pressure and gravity is so crucial to what is to come that it is worth reviewing
within a protostar constantly changes as it descends the here. Increasing the tempera-
Hayashi track toward the main sequence, new balances ture and pressure at the cen- Increases in
must also constantly be found as a star evolves beyond the ter of a star means several temperature and
main sequence, until at last no balance is possible at all. things. For one, it means that pressure speed up
The mass and composition of a star control the star’s life more atomic nuclei are nuclear burning.
on the main sequence, and they remain at center stage in packed together into a smaller
the closing acts of the star’s volume. You know that you are far more likely to bump
story as well. The evolution- Mass and into another person while strolling through a crowded
ary course followed by each composition determine shopping mall than through an empty park. Similarly,
of the hundred billion or so a star’s fate. packing atomic nuclei more tightly together increases the
stars that make up the Milky likelihood that they will run into each other and fuse.
Way Galaxy (and each of the stars in every other galaxy Higher density means more frequent collisions between
throughout the universe) is locked in place when the star atomic nuclei, and a higher number of collisions means
forms, determined foremost by the seemingly incidental faster burning. Higher temperature also drives up the rate
fact of the amount of mass incorporated into the star at the of nuclear reactions: higher temperature means that atomic
time of its birth, and secondarily by the chemical composi- nuclei are moving faster, so they are more likely to bump
tion of the material from which it formed. into each other. More important, higher temperature means
Each star is unique. Relatively minor differences in the that atomic nuclei collide with each other more violently,
masses and chemical compositions of two stars can some- making it more likely they will overcome the electric
times result in significant, repulsion that pushes the positively charged nuclei apart.
and possibly even dramatic, Low-mass and high- As a result of the combined effects of temperature and
differences in their fates. The mass stars evolve density, modest increases in pressure within a star can
course followed by a star with differently. sometimes lead to dramatic increases in the amount of
a mass of 1.1 times the mass energy released by nuclear burning.
of the Sun (1.1 M⊙) is not identical to the fate of a star with Here is the key to understanding why the main sequence
a mass of 0.9 M⊙. Nevertheless, stars can be divided roughly is primarily a sequence of masses, with low-mass stars on
into two broad categories whose members evolve in quali- the faint end and high-mass stars on the luminous end.
tatively different ways. Massive, luminous O and B stars More mass means stronger gravity, stronger gravity means
follow a course fundamentally different from that of the higher temperature and pressure in the star’s interior,
cooler, fainter, less massive stars found toward the lower higher temperature and pressure mean faster nuclear reac-
right end of the main sequence (see Figure 13.17). These tions, and faster nuclear reactions mean a more luminous
16.2 The Life and Times of a Main-Sequence Star 445
Amount of fuel (∝ mass of star)

star. If the Sun were more massive, it would necessarily Lifetime of star = ____________________________.
Rate fuel is used (∝ luminosity of star)
have a different balance between gravity and pressure—a
balance in which the Sun would burn its nuclear fuel more As Math Tools 16.1 shows us, this relationship among mass,
rapidly and thus would be more luminous. In other words, luminosity, and lifetime indicates that lower-mass stars will
if the Sun were more massive, it would be located at a differ- have much longer lives than their higher-mass cousins.
ent position on the H-R diagram: farther up and to the left The lifetimes of stars should be of more than passing
on the main sequence. Mass determines the structure of a interest to us. We don’t yet know if life exists on planets
star and its place on the main sequence. This, in a nutshell, orbiting other stars, but we might assume that life would
is the heart of our understanding of stellar structure. be unlikely to evolve on a planet orbiting a massive star
with a stable life of only a few million years (see Chapter
23). Likewise, it would be unlikely for life to develop on
Higher Mass Means a a planet surrounding a very low-mass star that had yet to
initiate the stable nuclear burning phase.
Shorter Lifetime
A main-sequence star can live only so long. But how long is
long? The question “How long can a main-sequence star con- The Structure of a Star
tinue to burn hydrogen in its core?” is much like the ques-
tion “How long can you drive your car before it runs out of
Changes As It Uses Its Fuel
gas?” In the case of the car, the answer depends in part on To say that a main-sequence star is stable is not to say that
how much gas your tank holds. The larger the gas tank, the it does not continually change. When the Sun formed, about
more fuel you have and the longer your motor might run. But 90 percent of its atoms were hydrogen atoms. Since then, the
the answer also depends on the size and efficiency of your Sun has produced its energy by converting hydrogen into
motor. A gas-guzzling eight-cylinder SUV drinks fuel a lot helium via the proton-proton chain. As the composition of
faster than a motor scooter. The amount of time your motor a star changes, so must its structure. When we discussed
will run is determined by a competition between these two the collapse of a protostar toward the main sequence in
effects. The larger motor might run out of gas fi rst, even if Chapter 15, we considered the idea of a changing balance
it is attached to a much larger tank. between gravity and pressure. The protostar was always in
The competition between these two effects—tank size balance; but as the star radiated away thermal energy, this
and motor size—is most readily expressed as a ratio. How balance was constantly changing, shifting toward that of a
long your motor runs is given by the amount of gas in the smaller and denser object.
tank, divided by how quickly the motor uses it: The same concept applies here. As a main-sequence star
uses the fuel in its core, its structure must continually shift
Amount of fuel
Lifetime of in response to the changing core composition. At any given
(gallons)
tank of gas = ___________________ . point in its lifetime, a main-sequence star like the Sun is in
(hours) Rate at which fuel is used
balance, but the balance in the Sun today is slightly different
(gallons/hour)
from the balance the Sun had 4 billion years ago, and slightly
If you have a 15-gallon tank and your motor is burning fuel different from the balance it will have 4 billion years from
at a rate of 3 gallons each hour, then your motor will use up now. Between the time the Sun was born and the time it will
all of the gas in just 5 hours. leave the main sequence, its luminosity will roughly double,
The same principle works for main-sequence stars. The with most of this change occurring during the last billion
amount of fuel is determined by the mass of the star. The years of its life on the main sequence. Stars evolve even as
more massive the star is, the they “sit” on the main sequence, although this evolution is
more hydrogen is available to How quickly a star slow and modest in comparison with the events that follow.
power nuclear burning. The runs out of fuel What happens to the
rate at which fuel is used is depends on its mass helium that is produced in Helium ash builds up
measured by the luminosity and luminosity. the core of a low-mass main- in the core of a main-
of the star. Main-sequence sequence star like our Sun? sequence star.
stars are “in balance,” so energy is radiated into space from We might imagine that it
the surface of the star at the same rate at which energy is begins to burn, with helium atoms fusing to form heavier
being generated in its core. (This balance between energy elements; but such is not the case. For fusion to occur, atomic
generation and luminosity remains true at almost every nuclei must be slammed together with enough energy to
stage of a star’s evolution.) If one main-sequence star has overcome the electric repulsion between them. For two
twice the luminosity of another, then it must be burning helium nuclei to get close enough to fuse, the two protons
hydrogen at twice the rate of the other star. in one nucleus must overcome the repulsive force of the
An expression for the main-sequence lifetime of the star two protons in the other nucleus—four times the repulsive
looks very similar to our previous expression for the time force of one hydrogen nucleus acting on another hydro-
it takes your car to run out of fuel: gen nucleus. At the temperature found at the center of a
M AT H T O O L S 1 6 .1
Estimating Main-Sequence Lifetimes

Astronomers can determine the lifetime of main-sequence stars As an example, let’s look at a main-sequence O5 star. Accord-
either observationally or by modeling the evolution of stars of ing to Table 16.1, which is based on computer models, an O5 star
a given composition. One simple method is to employ what has a mass that is about equal to 60 times that of the Sun:
we might call a “rule-of-thumb approach.” If we use what we
know about how much hydrogen must be converted into helium τO5 = (1.0 × 1010) × (60)–2.5 = 3.6 × 105 years.
each second to produce a given amount of energy, as well as
the fraction of its hydrogen that a star burns, we can come up Instead of the 10-billion-year life span of the Sun, an O5 star
with a relationship that says the main-sequence lifetime of a has a main-sequence lifetime of less than a half-million years!
star, τMS, can be expressed as Several generations of O stars have lived and died in the time
that hominids have walked the surface of Earth. Even though
M
τMS ∝ ____
MS
, the 60-M⊙ star starts out with 60 times as much fuel as the
LMS
Sun, it burns that fuel so much faster that it uses it up in less
where M is mass (amount of fuel) and L is luminosity (the rate than a ten-thousandth of the Sun’s lifetime. While low-mass
fuel is used). We can put this relationship in quantitative terms stars live their lives in the slow lane, high-mass stars live fast
by introducing the constant of proportionality, 1.0 × 1010, which and die young.
is the computed lifetime (in years) of a 1-M⊙ star:
MMS/M⊙
τMS = (1.0 × 1010) × _______ years.
LMS/L ⊙ FIGURE 16.1 The mass-luminosity relationship for main-
Now, let’s see how the lifetime of a more massive star com- sequence stars: L ∝ M 3.5. The exponent (3.5) is an average
value over the wide range of main-sequence star masses.
pares with that of the Sun. The relationship between the mass
Observational data show that the deviation of stars from the
and the luminosity of stars is very sensitive. Relatively small
average relationship depends on their mass.
differences in the masses of stars result in large differences in
their main-sequence luminosities. One method for estimating
luminosities of main-sequence stars is known as the mass-
luminosity relationship, L ∝ M 3.5, which is based on observed
luminosities of stars of known mass,1 as illustrated in Figure Actual luminosities
16.1. As in the preceding example, we can express this relation- vs. masses
ship relative to the Sun’s mass and luminosity:
L
( )
M 3.5
MS
___ MS
= ____ Straight-line
Luminosity
L⊙ M⊙ (exponential)
approximation
Substituting the mass-luminosity relationship into the lifetime L M 3.5
equation gives us
MMS/M⊙
( )
MMS –2.5
τMS = (1.0 × 1010) × __________ = (1.0 × 1010) × ____ years.
(MMS/M⊙)3.5 M⊙
1
The exponent 3.5 is not a hard-and-fast number. Observational
evidence shows that it can vary from 2.5 to 5.0, depending on the Mass
mass of the star (see Figure 16.1).
low-mass main-sequence star, collisions are not energetic the resulting helium collects there, building up like the
enough to overcome the electric repulsion between helium nonburning ash in the bottom of a fi replace.
nuclei.2 As hydrogen burns in the core of a low-mass star,
that the strong-nuclear-force interaction between the two 3He nuclei
2
You may be puzzled, remembering from Chapter 14 that one step of is much more powerful than the strong-nuclear-force interaction
the proton-proton chain is the fusion of two 3He nuclei, which have between two 4He nuclei, so two 3He nuclei do not have to get as close
as strong an electric repulsion as two 4He nuclei do. The answer is together as do two 4He nuclei in order to fuse.
TA B L E 1 6 .1
Main-Sequence Lifetimes*
Spectral Main-Sequence
Type Mass (M⊙) Luminosity (L⊙) Lifetime (Years)
O5 60 794,000 3.6 × 105
B0 17.5 52,500 1.0 × 107
B5 5.9 832 7.2 × 107
A0 2.9 54 3.9 × 108
A5 2.0 14 1.1 × 109
F0 1.6 6.5 2.1 × 109
F5 1.3 3.2 3.5 × 109
G0 1.05 1.5 8.3 × 109
G2 (our Sun) 1.0 1.0 1.0 × 1010
G5 0.92 0.8 1.5 × 1010
K0 0.79 0.4 3.7 × 1010
K5 0.67 0.15 5.3 × 1010
M0 0.51 0.077 6.6 × 1010*
M5 0.21 0.011 1.9 × 1011*
M8 0.06 0.0012 5.0 × 1012*
*Lifetimes marked with an asterisk (*) are based on the “rule of thumb” discussed in Math Tools 16.1.
The other lifetimes are based on computer models.
uniform composition throughout, with hydrogen account-

Helium Ash Builds Up in ing for about 70 percent of the mass in the Sun and helium
the Center of the Star accounting for most of the remaining 30 percent. With time,
the helium fraction in the center of the Sun climbed. Today,
Helium does not build up at the same rate throughout the roughly 5 billion years later, only about 35 percent of the
interior of a star. In Chapter 7 we found that the tempera- mass at the center of the Sun is hydrogen.
ture and pressure within Earth must be highest at the center
of the planet. No other configuration makes sense. Exactly
the same arguments apply equally well to stars. The fact
that the temperature and pressure are highest at the center
of a main-sequence star means that hydrogen burns most
16.3 A Star Runs Out of
rapidly there as well. As a result, nonburning helium “ash” Hydrogen and Leaves
accumulates most rapidly at the center of the star.
If we could cut a star open and watch as it evolves, we the Main Sequence
would see the chemical composition of the star changing
most rapidly at its center and less rapidly as we move out- Hydrogen burning in the core of a star cannot continue for-
ward through the star. Figure 16.2 shows how the chemical ever. Eventually—about 5 billion years from now in the case
composition inside a star like the Sun changes throughout of the Sun—a star exhausts all of the hydrogen fuel avail-
its main-sequence lifetime. When the Sun formed, it had a able at its center. At this point, the innermost core of the
(a) Sun when formed energetic collisions. (In other words, the gas is completely
100 ionized.) So the gas inside the Sun is a mixture of electrons
80 and atomic nuclei all flying about freely. Even so, the gas
Hydrogen that makes up the Sun is still mostly empty space, with
60 the electrons and atomic nuclei filling only a tiny fraction
40
of the volume.
When a low-mass star like the Sun exhausts the hydro-
20 gen at its center, the situation changes. As gravity begins to
Helium win its shoving match against
0 pressure, the helium core is The crushed helium
(b) Sun today
100 crushed to an ever-smaller core is electron-
Over time, helium ash size and an ever-greater den- degenerate.
Percentage of mass
80 collects in the core of a low- sity, but there is a limit to how

mass main-sequence star…
dense the core can get. The rules of quantum mechanics
60
Hydrogen (the same rules that say that atoms can have only certain
40 discrete amounts of energy and that light comes in packets
called photons) limit the number of electrons that can be
20
packed into a given volume of space at a given pressure.3
Helium
0 As the matter in the core of the star is compressed further
(c) Sun in 5 billion years and further, it finally bumps up against this limit. The space
100 occupied by the core of the star is no longer mostly empty,
…until no more hydrogen but is now effectively “fi lled” with electrons that are
80 remains at the center.
smashed tightly together. Matter at the center of the star is
60 now so dense that a single cubic centimeter of this material
Hydrogen can have a mass of a metric ton (1,000 kilograms [kg]) or
FIGURE 16. 2 Chemical 40
more. Matter that has been compressed to this point is said
composition of the Sun
is plotted here as a 20 to be electron-degenerate.
percentage of mass Helium
0
against distance from 0.0 0.2 0.4 0.6 0.8 1.0
the center of the Sun. (a) Center Surface Hydrogen Burns in a Shell
When the Sun formed 5 Fraction of radius (R/R£)
billion years ago, about 30
Surrounding a Core of Helium Ash
percent of its mass was helium and 70 percent was hydrogen Once a low-mass star exhausts the hydrogen at its center,
throughout. (b) Today the material at the center of the Sun is nuclear burning may end there, but the story is very differ-
about 65 percent helium and 35 percent hydrogen. (c) The Sun’s ent outside the core. The layers surrounding the degenerate
main-sequence life will end in about 5 billion years, when all of the
core still contain hydrogen, and this hydrogen continues to
hydrogen at the center of the Sun is gone.
burn. Astronomers speak of hydrogen shell burning because
the hydrogen is burning in a shell surrounding a core of
helium, like the flesh of a fruit around its pit.
star is composed entirely of helium ash. As thermal energy The electron-degenerate core of a star has a number
leaks out of the helium core into the surrounding layers of fascinating properties. For example, as more and more
of the star, no more energy is generated within the core to helium ash piles up on the degenerate core, the core shrinks
replace it. The balance that has maintained the structure in size. It does so because the added mass increases the
of the star throughout its life is now broken. The star’s life strength of gravity and therefore the weight bearing down
on the main sequence has come to an end. on the core, which means that the electrons can be smashed
together into a smaller volume. The presence of the degen-
erate core triggers a chain of events that will dominate the
evolution of our 1-M⊙ star for the next 50 million years. Fol-
The Helium Core Is Degenerate low along as we step through the chain of cause and effect
Throughout our lives, all of the matter we directly expe- that takes center stage in the post–main sequence evolution
rience is mostly empty space. Just as the Solar System is of a low-mass star:
mostly empty except for the tiny bit of space occupied by
the Sun and the planets, an atom is mostly empty except 3
If you have taken high school or college chemistry, you may remember
for the tiny bit of space occupied by the nucleus and the the “Pauli exclusion principle,” which limits the number of electrons
electrons. The same is true for the matter within the Sun. that can go into a single orbital in an atom. This principle also lim-
At the enormous temperatures within the Sun the electrons its the number of electrons that can be packed into the energy states
have almost all been stripped away from their atoms by available in the center of a star.
1. From our discussion in Chapter 10 (see Figure 10.3), we in Figure 16.3, the internal structure of the star is now fun-
know that just outside the star’s degenerate core, the damentally different from the structure when the star was
gravitational acceleration gcore is given by on the main sequence. The giant can grow to have a lumi-
nosity hundreds of times the luminosity of the Sun and a
GMcore radius of over 50 solar radii (50 R⊙). Yet at the same time the
gcore = _____ ,
r 2core core of the giant star is far more compact than that of the
Sun, with much of the star’s mass becoming concentrated
where Mcore is the mass of the helium core and rcore is its into a volume that is only a few times the size of Earth.
radius. As the helium core grows, its larger mass (bigger From our vantage point outside the star, we are not privy
Mcore) and its shrinking size (smaller rcore) both cause the to the changes taking place deep within its interior. All we
strength of gravity at the surface of the core to increase. can see directly is that the star becomes larger and more
As more helium is added to the core, the strength of grav- luminous, and perhaps surprisingly, cooler and redder as
ity at its surface increases dramatically. well. The enormous expanse of the star’s surface allows it
2. Increasing the strength of gravity around the core to cool very efficiently. Even though its interior grows hot-
increases the weight of the overlying material pushing ter and its luminosity increases, the surface temperature of
down on the hydrogen-burning shell surrounding the the star actually begins to drop.
core. Just as the H-R diagram helped us follow the changes in
a protostar on its way to the main sequence, it is a handy
3. This increase in weight must be balanced by an increase device for keeping track of the
in pressure in the inner parts of the star. In particu- changing luminosity and sur- An evolving low-mass
lar, the pressure in the hydrogen-burning shell must face temperature of the star as star moves up and
increase. it evolves away from the main to the right on the
4. Increasing the pressure in the hydrogen-burning shell sequence (Figure 16.4). As H-R diagram.
drives up the rate of the nuclear reactions occurring in soon as the star exhausts the
the shell. hydrogen in its core, it leaves the main sequence and begins
to move upward and to the right on the H-R diagram, grow-
5. Faster nuclear reactions release more energy, so the ing more luminous but cooler. We refer to such a star, which
luminosity of the star increases. is somewhat brighter and larger than it was on the main
sequence, as a subgiant star. As the subgiant continues to
This is a very counterintuitive result! We might have evolve, it grows larger and cooler, but after a time its prog-
imagined that when a star like the Sun used up the nuclear ress to the right on the H-R diagram hits a roadblock: the
fuel at its center, it would H– thermostat. When the surface temperature of the subgiant
grow fainter. Yet just the oppo- Running out of fuel star has dropped by about 1,000 kelvins (K) relative to its
site happens. A degenerate makes the star grow temperature on the main sequence, H– ions start to form in
core means stronger gravity, more luminous. great abundance in its atmosphere. We have encountered
stronger gravity means higher the formation of H– ions before. In our discussion of proto-
pressure, higher pressure means faster nuclear burning, stars in Chapter 15, it was the H– ion in the protostar’s atmo-
and faster nuclear burning means a more luminous star. sphere that acted as a thermostat, regulating the star’s
When the low-mass star “runs out of gas” at its center, it temperature. The H– ion serves exactly the same role here,
responds by getting more luminous! regulating how much radiation can escape from the star
and preventing it from becoming any cooler.
Because the star can cool no further, it begins to move
Tracking the Evolution of the almost vertically upward on the H-R diagram, growing
larger and more luminous but remaining about the same
Star on the H-R Diagram temperature. The star has become a red giant—an obvious
The changes that occur in the heart of a star with a degener- name for a star that is now both redder and larger than it
ate helium core are reflected in changes in the overall struc- was on the main sequence. We can think of the path that
ture of the star. With time, the a star follows on the H-R diagram as it leaves the main
mass of the degenerate helium A bloated luminous sequence as being a tree “branch” growing out of the “trunk”
core grows as more and more giant surrounds a tiny of the main sequence. Astronomers refer to these tracks as
hydrogen is converted into degenerate core. the subgiant branch and the red giant branch of the H-R
helium ash in the surrounding diagram. Interestingly, the path that a red giant follows on
shell. And as the mass of the degenerate helium core grows, the H-R diagram closely parallels the path that it followed
so does the rate of energy generation in the surrounding earlier as a collapsing protostar on its way toward the main
hydrogen-burning shell. This increase in energy generation sequence—except, of course, in reverse: this time the star
heats the overlying layers of the star, causing them to expand. is moving up that path rather than coming down it. This
The star becomes a bloated, luminous giant. As illustrated similarity is not a coincidence. The same physical processes
FIGURE 16. 3 The structure of a star near the top of the red
giant branch is compared with the structure of the Sun. Left
Nonburning
panels compare the size of the Sun with the size of the red giant. envelope
Right panels compare the size and structure of the Sun with the
core of the red giant. The panels at right are magnified about 50
times compared to the panels on the left.
1-M£ MAIN-SEQUENCE STAR
1 A luminous red giant star is

enormous compared to the Sun…
Hydrogen-
1-M£ RED GIANT STAR burning core
50 R£ = 3.5 10 km 7
1 R£ = 7 105 km
2 …but this luminosity comes from

hydrogen burning in a thin shell
around a tiny degenerate core.
Degenerate
helium core
Hydrogen-
burning shell
Nonburning
envelope
10 L ⊙ to almost 1,000 L ⊙. The evolution of the star, illustrated

in Figure 16.5, is reminiscent of the growth of a snowball
rolling downhill. The larger the snowball becomes, the
faster it grows; and the faster it grows, the larger it becomes.
(such as the H– thermostat) that give rise to the Hayashi “Growth” and “size” feed off each other, and what began
track followed by a collapsing protostar also control the as a bit of snow at the top of the mountain soon becomes a
relationship of luminosity, size, and surface temperature huge ball that could trigger an avalanche.
in an expanding red giant. The analogy between the evolution of a red giant star
As the star leaves the main sequence, the changes in and the growth of a snowball is actually not a bad one. The
its structure occur sluggishly at fi rst, but then they pick helium core of the star grows in mass (but not in radius!) as
up steam as the star moves up the red giant branch faster hydrogen is converted to helium in the hydrogen-burning
and faster. It takes 200 million years or so for a star like the shell. The increasing mass of the ever–more compact helium
Sun to go from the main sequence to the top of the red giant core drives up the force of gravity in the heart of the star.
branch. Roughly the fi rst half of this period is spent on the Stronger gravity means higher pressure, and higher pressure
subgiant branch as the star’s luminosity increases to about accelerates nuclear burning in the shell. But faster nuclear
10 times the luminosity of the Sun (L ⊙). During the sec- reactions in the shell convert hydrogen into helium more
ond half of this time the star’s luminosity skyrockets from quickly, so the part of the star’s mass that is in its core grows
Spectral FIGURE 16. 4

type The evolution of a
O5 B0 B5 A0 F0 G0 K5 M5
107 red giant star on the
RED GIANT STAR
H-R diagram. The
A 1-M£ star exhausts the
106 structure of a red
hydrogen at its center, leaves 1,0
the main sequence, and 00 giant star consists
R Nonburning
ascends the red giant branch. £ of a degenerate
105
M
degenerate He core core of helium ash

AI
SE
N
10 surrounded by a
Q 0R H-burning
104 UE hydrogen-burning
NC
£
shell
E shell. As the star
103 10 Nonburning moves up the red
Luminosity (L£)
R hydrogen envelope giant branch, it

£
102 The tiny core and shell

comes close to
are not drawn to scale in retracing the Hayashi
1R
£ this schematic drawing. track that it followed
101
when it was a
0.1 MAIN SEQUENCE STAR protostar collapsing
1 R £
toward the main
£
sequence.
10–1 0.0
1R
£
H-burning
10–2 core
0.0
01 Nonburning
R£
10–3 hydrogen envelope
40,000 30,000 10,000 6,000 3,000
The more compact and more

luminous the energy-producing
shell becomes, the more the
outer parts of the star swell.
Core growth and rapid hydrogen

burning feed off each other in
a red giant star, causing it to
move more and more rapidly
…stronger gravity, up the red giant branch.
which leads to…
…greater weight of overlying …a smaller, more massive degen-

layers, which leads to… erate He core, which leads to…
Red giant
branch
Luminosity (L£)
…higher pressure in H-burning

shell, which leads to…
…faster nuclear …faster growth of helium

burning, which leads to… Surface core mass, which leads to…
temperature (K)
FIGURE 16. 5 As a star moves up the red giant branch,

burning hydrogen to helium in a shell surrounding a
…faster conversion of H
degenerate helium core, its evolution feeds on itself. The
to He, which leads to…
luminosity of the star grows faster and faster.
more rapidly. We have come full circle in a cycle that feeds star, then surely the atomic nuclei are packed as tightly as
on itself. Increasing core mass leads to ever-faster burning possible into the core as well. After all, we normally think
in the shell; and the faster hydrogen burns in the shell, the of atomic nuclei as being larger than electrons.” But the
faster the core mass grows. As a result, the star’s luminos- world behaves in strange ways at the quantum mechani-
ity climbs at an ever-higher rate. cal level. The rules of quantum mechanics say that when
packed together tightly, less massive particles like elec-
trons effectively “take up more space” than do more mas-
sive particles like atomic nuclei. As the core of the star is
16.4 Helium Begins to Burn compressed, electrons become degenerate much sooner than
the atomic nuclei do.
in the Degenerate Core “Fine,” you might continue, “but how can the atomic
nuclei go moving freely about within a core that is wall-to-
The growth of the red giant cannot continue forever, and wall electrons?” Actually, this is no problem at all: The laws
we fi nd ourselves once again asking a crucial question for of quantum mechanics place few restrictions on electrons
understanding the evolution of stars: What will be the next and atomic nuclei occupying the same physical space. As far
thing to give? The answer lies in another unusual property as the atomic nuclei are concerned, the electron-degenerate
of the degenerate helium core, and we now turn our atten- core of the star is still mostly empty space. The nuclei are
tion there. free to go flying about through the sea of degenerate elec-
trons almost as if the electrons were not there. The negative
charges of the electrons are important because they balance
The Atomic Nuclei in the Core out the positive charges of the nuclei; but apart from that,
the atomic nuclei in the electron-degenerate core of a star
Form a “Gas within a Gas” are a perfectly normal “gas within a gas,” behaving just as
The core of the red giant star is electron-degenerate, which the (electrically neutral) atoms and molecules in the air
means that as many electrons are packed into that space as around you do.
the rules of quantum mechanics allow at that pressure. How-
ever, atomic nuclei in the core are still able to move freely
about, as shown in Figure 16.6, just as they are throughout Helium Burning and the
the rest of the star.
“Wait a minute!” you might say. “That last statement is
Triple-Alpha Process
nonsense for at least two different reasons. First, if the elec- As the star evolves up the red giant branch, its helium core
trons are packed as tightly as possible into the core of the grows not only smaller and more massive, but hotter as
FIGURE 16.6 In a red giant star the weight of the overlying layers is supported by electron
degeneracy pressure in the core arising from the fact that electrons are packed together as tightly
as quantum mechanics allows. Even so, atomic nuclei in the core are able to move freely about within
the sea of degenerate electrons, so they behave as a normal gas.
The weight of overlying …is supported by

layers of the star… degeneracy pressure
of the “sea” of tightly
packed electrons.
Red giant Electron- Helium

degenerate nuclei
helium core
Hydrogen-
burning shell
Nonburning
envelope
Degenerate
electron “sea”
Helium nuclei are free to move

about within the electron sea,
acting like a normal gas within the
degenerate electron gas.
16.4 Helium Begins to Burn in the Degenerate Core 453
well. This increase in temperature is partly because of the 1 The triple-alpha process begins
gravitational energy released as the core shrinks (just as the when two 4He nuclei fuse to
protostar’s core grew hotter as it collapsed) and partly due form an unstable 8Be nucleus.
to the energy released by the ever-faster pace of hydrogen 4
He
burning in the surrounding shell. The climbing temperature 8Be
of the core means that the thermal motions of the atomic 12C
nuclei in the core become more and more energetic. Even- 4
He
tually, at a temperature of about 108 K (a hundred million
kelvins), the collisions between helium nuclei in the core gamma-ray
become energetic enough to overcome the electric repul- 4
He
sion between them. Helium nuclei are slammed together
hard enough for the strong nuclear force to act, and helium 2 If this nucleus collides with 3 The energy released is
burning begins. another 4He nucleus before carried off both by the
Helium burns in a two-stage process, referred to as the it breaks apart, the two will motion of the 12C nucleus
fuse to form a nucleus and by a gamma ray.
triple-alpha process, illustrated in Figure 16.7. First, two of carbon-12 (12C).
helium-4 nuclei (4He) fuse to
form a beryllium-8 nucleus Helium burns via the FIGURE 16.7 The triple-alpha process. Two helium-4 (4He)
(8Be) consisting of four pro- triple-alpha process. nuclei fuse to form an unstable beryllium-8 (8Be) nucleus. If this
tons and four neutrons. The nucleus collides with another 4He nucleus before it breaks apart,
8
Be nucleus is extremely unstable. Left on its own, it would the two will fuse to form a stable nucleus of carbon-12 (12C).
break apart again after only about a trillionth (10 –12) of a The energy produced is carried off both by the motion of the 12C
nucleus and by a high-energy gamma ray emitted in the second
second. But if, in that short time, it collides with another
4 step of the process.
He nucleus, the two nuclei will fuse into a stable nucleus
of carbon-12 (12C) consisting of six protons and six neutrons.
The triple-alpha process takes its name from the fact that
it involves the fusion of three 4He nuclei, which are tradi- However, the degenerate core of a red giant is not a
tionally referred to as alpha particles. normal gas. In a certain sense the degenerate core behaves
more like a rock than like a normal gas. A rock is hard to
crush because of how closely the atoms within it are packed
The Helium Core Ignites together. Heating a rock does not cause it to swell up like
a hot-air balloon, and cooling a rock does not cause it to
in a Helium Flash deflate. Similarly, the pressure in a red giant’s degenerate
In the next phase of the star’s evolution, the helium in the core comes from how tightly the electrons in the core are
core begins burning, as Figure 16.8 illustrates. Degener- packed together. Heating the core does not change the num-
ate material is a very good conductor of thermal energy, so ber of electrons that can be packed into its volume, so the
any differences in temperature within the core are quickly core’s pressure does not respond to changes in temperature.
evened out. As a result, the degenerate core of a red giant And if the pressure does not increase, then there is nothing
star is at almost exactly the same temperature throughout. to cause the core to expand.
When helium burning begins at the center of the core, the Take another run at that to be sure it clicks. When helium
energy released quickly heats the entire core. Within a few begins to burn in the degenerate core, the temperature of
minutes the entire core is burning helium into carbon by the core goes up but the pres-
the triple-alpha process. sure does not. So the onset of The helium flash is
In a normal gas like the air around you, the pressure helium burning in the degen- a thermonuclear
of the gas comes from the random thermal motions of the erate core of a red giant does runaway—an explosion
atoms. Increasing the temperature of such a gas means that not cause the core to expand! within the star.
the motions of the atoms become more energetic, so the pres- Yet even though the higher
sure of the gas increases. If the helium core of a red giant temperature does not change the pressure, it does cause the
star were a normal gas, the increase in temperature that helium nuclei to be slammed together with more frequency
accompanies the onset of helium burning would lead to an and greater force, so the nuclear reactions become more
increase in pressure. The core of the star would expand; the vigorous. The process begins to snowball again. More vig-
temperature, density, and pressure would decrease; nuclear orous reactions mean higher temperature, and higher tem-
reactions would slow; and the star would settle down into perature means even more vigorous reactions. Thermonuclear
a new balance between gravity and pressure. These are burning in the degenerate core runs away with itself, wildly
exactly the sorts of changes that are steadily occurring out of control as increasing temperature and increasing
within the core of a main-sequence star like the Sun, as the reaction rates feed each other. This is the helium flash.
structure of the star steadily and smoothly shifts in response It is difficult to imagine the drama of the thermonuclear
to the changing composition in the star’s core. runaway that takes place during the helium flash. Helium
Spectral
type HELIUM FLASH
O5 B0 B5 A0 F0 G0 K5 M5
7
10
1 The degenerate helium

106 core of the red giant ignites 1,
00
in a helium flash… 0R
£
105
M
Runaway He burning:
AI
SE
N
10 The degenerate helium

Q 0R
104 UE £ core explodes within
NC the star.
E
103 10
R
Luminosity (L£)
2 …and then the star settles £
2
onto the horizontal branch.
10
1R
£
1
10 HORIZONTAL BRANCH STAR
Stable He-burning core
0.1
1 R £
£
Hydrogen-burning shell
10 –1 0.0
1R
£ Nonburning hydrogen
10–2 envelope
0.0
01
R£
–3
10
40,000 30,000 10,000 6,000 3,000
FIGURE 16.8 When the core temperature of a red giant reaches about 108 K, helium begins to burn
explosively in the degenerate core, leading to a helium flash. After a few hours the core of the star begins to
inflate, ending the helium flash. Over about the next 100,000 years (a relatively short time) the star settles
onto the horizontal branch, where it burns helium in its core and hydrogen in a surrounding shell.
burning begins at a temperature of about 100 million K. By longer degenerate) is much larger, so the acceleration due to
the time the temperature has climbed by a mere 10 percent, gravity within it and the surrounding shell is much smaller.
to 110 million K, the rate of helium burning has increased (Again, g = GM/r 2, so a larger core radius means smaller val-
to 40 times what it was at 100 million K. By the time the ues of g.) Weaker gravity means less weight pushing down
core’s temperature reaches 200 million K, the core is burning on the core and the shell, which means lower pressure.
helium 460 million times faster than it was at 100 million K! Lower pressure, in turn, slows the nuclear reactions. The
As the temperature in the core grows higher and higher, the net result is that following the helium flash, core helium
thermal motions of the electrons and nuclei become more burning keeps the core of the star puffed up, and the star
energetic, and the pressure due to these thermal motions becomes less luminous than it was as a red giant.
becomes greater and greater. Within seconds of ignition, the The star spends the next 100,000 years or so settling into
thermal pressure increases until it is no longer smaller than a stable structure in which helium burns to carbon in a
the degeneracy pressure, at which point the core literally normal, nondegenerate core
explodes. Because the explosion is contained within the while hydrogen burns to Horizontal branch
star, however, we do not see it. The energy released in this helium in a surrounding shell. stars burn helium
runaway thermonuclear explosion lifts the overlying layers The star is now about a hun- in the core and
of the star, and as the core expands, the electrons are able dredth as luminous as it was hydrogen in a shell.
to spread out. The drama is over within a few hours. The at the time of the helium flash.
expanded helium-burning core is no longer degenerate, and The lower luminosity means that the outer layers of the star
the star is on its way toward a new equilibrium. are not as puffed up as they were as a red giant. The star
Following the helium flash, the star once again does shrinks, and as it does so its surface temperature climbs.
something counterintuitive. You might imagine that helium (This is just the reverse of the sequence of events that caused
burning in the core would cause the star to grow more lumi- the red giant to become larger and redder as it grew more
nous, but it does not. The tremendous energy released dur- luminous.) At this point in their evolution, low-mass stars
ing the helium flash goes into fighting gravity and puffing with chemical compositions similar to that of the Sun will
up the core. After the helium flash, the core (which is no bunch up on the H-R diagram just to the left of the red giant
16.5 The Low-Mass Star Enters the Last Stages of Its Evolution 455
branch. Stars that contain much less iron than the Sun tend helium with carbon yields a pretty good description of how
to distribute themselves away from the red giant branch the star evolves as it leaves the horizontal branch.
along a nearly horizontal line on the H-R diagram. This The star’s life on the horizontal branch, however, is much
stage of stellar evolution takes its name from this horizontal shorter than its life on the main sequence. For one thing,
band. The star is now referred to as a horizontal branch star there is now less fuel to burn in its core. In addition, the
(see Figure 16.8). star is more luminous, so it must be consuming fuel more
rapidly. Finally, helium is a much less efficient nuclear fuel
than hydrogen. Even so, for 100 million years the horizontal
branch star remains stable, burning helium to carbon in its
16.5 The Low-Mass Star core and hydrogen to helium in a shell.
The temperature at the center of a horizontal branch star
Enters the Last Stages is not high enough for carbon to burn,4 so carbon ash builds
up in the heart of the star, just
of Its Evolution as helium ash accumulates at The star forms a
the center of a main-sequence degenerate carbon
The evolution of a solar-type star from the main sequence star. When the horizontal core as it leaves the
to the helium flash and horizontal branch is fairly well branch star has burned all of horizontal branch.
understood. Just as our understanding of the interior of the the helium at its core, gravity
Sun comes from computer models of the physical condi- once again begins to win. The nonburning carbon ash core
tions within our local star, our understanding of the evolu- is crushed by the weight of the layers of the star above it
tion of a red giant comes from computer models that look until once again the electrons in the core are packed together
at the changes in structure as the star’s degenerate helium as tightly as the laws of quantum mechanics allow at its
core grows. These models show that any star with a mass pressure. The carbon core is now electron-degenerate, with
of about 1 M⊙ will follow the march from main sequence physical properties much like those of the degenerate helium
to helium flash, and then drop down onto the horizontal core at the center of a red giant.
branch. But when we try to use computer models to push The small, dense electron-degenerate carbon core drives
our understanding to what happens next, the road that we up the strength of gravity in the inner parts of the star,
follow gets a bit trickier. We already noted that differences which in turn drives up the pressure, which speeds up the
in chemical composition between stars significantly affect nuclear reactions, which causes the degenerate core to grow
where they fall on the horizontal branch. From this point more rapidly—we have heard this story before. The internal
on, small changes in the properties of a star—mass, chemi- changes occurring within the star are similar to the changes
cal composition, strength of the star’s magnetic field, or even that took place at the end of the star’s main-sequence life-
the rate at which the star is rotating—can lead to qualita- time, and the path the star follows as it leaves the horizontal
tive differences in how the star evolves. The farther we go branch echoes that earlier phase of evolution as well. Just
in time beyond the helium flash, the more possible it is that as the star accelerated up the red giant branch as its degen-
divergent evolutionary paths are open to a low-mass star. erate helium core grew, the star now leaves the horizontal
With this caution in mind, we continue our story of the branch and once again begins to grow larger, redder, and
evolution of a 1-M⊙ star with solar composition, following more luminous as its degenerate carbon core grows. The
what we currently believe to be the most likely sequence path that the star follows on the H-R diagram (Figure 16.9)
of events awaiting our Sun. closely parallels the path it followed as a red giant, getting
closer to the red giant branch as the star grows more lumi-
nous. That is why this phase of evolution is referred to as
The Star Moves Up the the asymptotic giant branch (AGB) of the H-R diagram. An
AGB star burns helium and hydrogen in nested concentric
Asymptotic Giant Branch shells surrounding a degenerate carbon core.
The structure of a horizontal branch star is much like the
structure of a main-sequence star in many respects. The big-
gest difference is that now instead of burning hydrogen into
helium in a stable, nondegenerate core, the horizontal branch
Giant Stars Lose Mass
star is burning helium into carbon in a stable, nondegener- Building on our analogy between AGB stars and red giants,
ate core. (The other difference, of course, is that hydrogen you might imagine that the next step in the evolution of an
is continuing to burn in a shell surrounding the core.) The AGB star should be a “carbon flash,” when carbon burn-
behavior of a star on the horizontal branch is remarkably ing begins in the star’s degenerate core. Yet a carbon flash
similar to that of a star on the main sequence. This similar- never happens. Before the temperature in the carbon core
ity offers the key to understanding what comes next. In our
account of what happened as the star evolved off the main 4
You may recall from Chapter 14 that burn as used here refers not to
sequence, replacing the word hydrogen with helium and then fi re but to nuclear fusion.
Spectral
2 An AGB star leaving the horizontal branch is much like
type a red giant leaving the main sequence, but with helium
O5 B0 B5 A0 F0 G0 K5 M5
107 burning around a degenerate carbon core instead of
1 The star exhausts the helium hydrogen burning around a degenerate helium core.
at its center, leaves the horizontal
106 1,0
branch, and ascends the 00 ASYMPTOTIC GIANT BRANCH STAR
asymptotic giant branch. R
£
105 Nonburning
M
IN Asymptotic giant
A
SE 10 branch (AGB) degenerate

0R
104 Q £
carbon ash core
UE
N CE He-burning shell
103 10
R
Luminosity (L£)
£ H-burning shell
2
10
1R
£
101 Nonburning
hydrogen
0.1 envelope
1 R £
£
10–1 0.0
1R
£
FIGURE 16.9 An asymptotic giant branch star consists of a
10–2 degenerate carbon core surrounded by helium- and hydrogen-
0.0
01 burning shells. As the carbon core grows, the star brightens,
R£
–3 accelerating up the asymptotic giant branch just as it earlier
10
40,000 30,000 10,000 6,000 3,000 accelerated up the red giant branch while its degenerate helium
Surface temperature (K) core grew.
becomes high enough for carbon to burn, the star loses its the star ascends the asymptotic giant branch, it loses another
gravitational grip on itself and expels its outer layers into 20 percent or even more of its total mass. By the time it is
interstellar space. well up on this branch, a star that began as a 1-M⊙ star prob-
Red giant and AGB stars are huge objects. The AGB star ably has a mass less than about 0.7 M⊙, and it may even have
into which a 1-M⊙ main-sequence star evolves can grow to lost more than half of its original mass. Mass loss on the
a radius hundreds of times the radius of our Sun. When asymptotic giant branch can be spurred on by a lack of sta-
our Sun becomes an AGB star, its outer layers will swell to bility in the star’s interior. The extreme sensitivity of the
the point that they engulf the orbits of the inner planets, triple-alpha process to temperature can lead to episodes of
possibly including Earth. (Do not lose much sleep over this rapid energy release, which can provide the extra kick needed
eventuality, however. It will be 5 billion years before Earth to expel material from the star’s outer layers. It also means
is engulfed by the burgeoning Sun, and well before then that stars that are initially quite similar can behave very dif-
Earth will be well toasted by the thousandfold increase in ferently when they reach this stage in their evolution.
the Sun’s luminosity.) When a star expands to such a size,
its hold on its outer layers becomes tenuous indeed.
Once again, an understanding of what happens to the The Post-AGB Star May Cause
star rests with Newton’s universal law of gravitation. Recall-
ing that g = GM/r 2, we know
a Planetary Nebula to Glow
that the acceleration due to Red giants and AGB Toward the end of an AGB star’s life, mass loss itself becomes
gravity (g) at the surface of a stars lose mass from a runaway process. When a star loses a bit of its outer layers,
star with a mass of 1 M⊙ and a their outer layers. the weight pushing down on the underlying layers of the
radius of 100 R ⊙ is only star is reduced. Without this weight holding them down, the
1/10,000 as strong as the gravity at the surface of the Sun. It outer layers of the star puff up even larger than they were
takes little in the way of a kick from below to push material before. The star, which is now both less massive and larger,
near the surface of such a giant star over the edge, driving is even less tightly bound by gravity, so even less energy is
it completely away from the star. The process of stellar mass needed to push outer layers away from it. The situation is a
loss begins when the star is still on the red giant branch; by bit like taking the lid off a pressure cooker. Mass loss leads to
the time a 1-M⊙ main-sequence star reaches the horizontal weaker gravity, which leads to faster mass loss, which leads
branch, it may have lost 10–20 percent of its total mass. As to weaker gravity, and so on. When the end comes, much of
PLANETARY NEBULA
EJECTION
Degenerate carbon core
Post-AGB star
He-burning shell
Forming planetary H-burning shell
nebula (expanding
outer layers of star) Nonburning envelope
1 The bloated asymptotic
giant branch star begins
Ejected stellar material
107
losing its outer layers,
ejecting a planetary nebula…
106 1,0
00
R
£
105
M
10
AI
0R
SE
N
104 £
Q
2 …leaving behind UE
only the degenerate NC
103 carbon core, a tiny E 10
R
Luminosity (L£)
£
cooling white dwarf.
White dwarf
102
1R
£
101
Hot electron-
0.1 degenerate carbon
1 R £
£ FIGURE 16.10 At the end of the AGB
star’s life, it ejects most of its mass in a
10–1 0.0 planetary nebula, becoming a post-AGB
1R
£ star and finally leaving behind nothing
but the star’s degenerate core, which
10–2
0.0 appears on the H-R diagram as a cooling
01
R£ white dwarf star.
10–3
100,000 40,000 30,000 10,000 6,000 3,000
the remaining mass of the star is ejected into space, typically right to left across the top of the H-R diagram, as shown in
at speeds of 20–30 kilometers per second (km/s). Figure 16.10.
After ejection of its outer layers, all that is left of the The surface temperature of the star may eventually reach
low-mass star itself is a tiny, very hot electron-degener- 100,000 K or hotter. Wien’s law says that at such tempera-
ate carbon core, surrounded by a thin envelope in which tures, most of the light from the star is in the hard (high-
hydrogen and helium are still burning. This star is now energy) UV part of the spectrum:
somewhat less luminous than when it was at the top of the
2,900
m K
asymptotic giant branch, but it is still much more luminous λpeak = __________ = 0.029
m.
than a horizontal branch star. The remaining hydrogen and 100,000 K
helium in the star rapidly burn to carbon, and as more and The intense uv light from what remains of the star heats and
more of the mass of the star ends up in the carbon core, the ionizes the expanding shell of gas that was recently ejected
star itself shrinks and becomes hotter and hotter. Over the by the star. The ultraviolet light causes the shell of gas to
course of only 30,000 years or so following the beginning glow in the same way that UV light from an O star causes
of runaway mass loss, the star moves very rapidly from an H II region to glow.
FIGURE 16.11 At the end of its life, a low-

(a) (b)
mass star ejects its outer layers and may form
a “planetary nebula” consisting of an expanding
shell of gas surrounding the white-hot remnant
of the star. (a) This picture of a famous planetary
nebula called the Ring Nebula, as it appears
through a small amateur telescope, shows why
astronomers thought these objects looked like
planets. (b) However, an HST image of the ring
shows the remarkable and complex structure of
this expanding shell of gas.
If conditions are right,5 the mass ejected

by the AGB star will pile up in a dense,
expanding shell. If you were to look at such a
shell through a telescope, you would see it as
a round or perhaps oblong patch of light sur-
rounding the remains of the AGB star. When
these glowing shells were fi rst observed in
small telescopes, they looked like disks of
Central Ejected planets (Figure 16.11a), which is why such
star material objects were named planetary nebulae.
But there is
nothing truly Planetary nebulae may
“ pl a n e t a r y ” form around dying
about a plan- low-mass stars.
etary nebula,
as is apparent in Figure 16.11b. Instead, a
planetary nebula consists of the remaining
Spirograph Eskimo outer layers of a star, ejected into space as a
dying gasp at the end of the star’s ascent of
the asymptotic giant branch.
Rather than being simple spherical shells
of the sort we might naïvely predict would
surround a nice spherical star, planetary
nebulae show a dazzling range in appear-
ance, earning them names like Owl Nebula,
Clown Nebula, Cat’s Eye Nebula, and Dumb-
bell Nebula. This extraordinary menagerie,
illustrated in Figure 16.12, serves to rub
NGC 6751 Helix
astronomers’ noses in the complexity of the
late stages of a star’s evolution, telling of eras
when mass loss from the star was slower or
faster, and of times when mass was ejected
primarily from the star’s equator or its poles.
The gas in the planetary nebula also con-
tains chemical elements that were produced
by nuclear burning in the star, offering us
5
Not all stars form planetary nebulae. Massive stars
Cat’s Eye M2-9 go through the post-AGB stage too quickly. Stars
with insufficient mass take too long, so their enve-
FIGURE 16.12 Planetary nebulae are not all simple spherical shells around their lope evaporates before they can illuminate it. Some
parent stars. These HST images of planetary nebulae show a wealth of structure astronomers believe that our own Sun will not retain
resulting from the complex processes by which low-mass stars eject their outer enough mass during its post-AGB phase to form a
layers. planetary nebula.
C O N N E C T I O N S 1 6 .1
Seeding the Universe with New Chemical Elements

One of the grand themes of 20th and 21st century astronomy is of nuclear burning in their spectra. This extra carbon originated
the origin of the chemical elements. When we discuss the prop- in the star’s helium-burning shell and then was carried to its
erties of the early universe, we will see that the only chemical surface by convection.
elements that were formed in abundance when the universe was Mass loss from giant stars carries the chemical elements
young were hydrogen and helium, along with trace amounts of enriching the stars’ outer layers off into interstellar space. Plan-
lithium, beryllium, and boron. etary nebulae—the ejected atmospheres of low-mass stars—
Astronomers studying red giant and AGB stars see direct often show an overabundance of elements such as carbon,
evidence that more massive chemical elements formed deep nitrogen, and oxygen, which are by-products of nuclear burn-
within the interiors of low-mass stars fi nd their way to the ing. Once this chemically enriched material leaves the star, it
“outside” world. In Chapter 14 we learned that if the change in mixes with the interstellar gas there, enriching the chemical
temperature within a star becomes too steep, convection will diversity of the universe. Novae and Type I supernovae, dis-
begin. Like the boiling motion in a pot of water, convection is cussed at the end of this chapter, also do their part. The nuclear
a process of mixing. Material from the inner parts of the star is burning that occurs in these explosive events goes far beyond
carried upward and mixed with material in the outer parts of the formation of elements such as carbon. Novae and Type I
the star. As a star ascends fi rst the red giant and later the asymp- supernovae help seed the universe with much more massive
totic giant branches of the H-R diagram, the core of the star atoms, such as iron and nickel.
grows hotter and hotter, and the temperature gradient within As we continue to discuss the evolution of stars and the
the star grows steeper. Convection spreads through more and universe in the chapters to come, we will learn more about the
more of the star. Under certain circumstances, convection can origin of the chemical elements. Today, when scientists study
spread so deep into a star that it is able to dredge up chemical the patterns of the abundance of different chemical elements
elements formed by nuclear burning within the star and carry on Earth and in our Solar System, they recognize the patterns
them to the star’s surface. Many AGB stars, known as carbon they see. Those patterns are clear signatures of nuclear reac-
stars, show an overabundance of carbon and other by-products tions in stars.
our fi rst look at the processes responsible for the chemical cools, so it moves down and to the right on the H-R diagram,
evolution of the universe (see Connections 16.1). The plan- following a line of constant radius. Even though the white
etary nebula is visible for 50,000 years or so before the gas dwarf may remain very hot for 10 million years or so, its
ejected by the star disperses so far that the nebula is too luminosity may now be only 1/1,000 that of a main-sequence
faint to be seen. star like our Sun. Many white dwarfs are known, but all
are much too faint to be seen without the aid of a telescope.
Yet when you look at Sirius, you are also looking at a white
dwarf. The brightest star in Earth’s sky has a faint white
The Star Becomes a White Dwarf dwarf as a binary companion.
Within 50,000 years or so, a post-AGB star burns all of the Here we have the fate of our Sun, some 6 billion or so
fuel remaining on its surface, leaving nothing behind but years from now. It will become a white dwarf that, bereft
a cinder—a nonburning ball of all sources of nuclear fuel, will shine ever more feebly as
of carbon with a remaining A newly formed white it continues to cool, radiating its thermal energy away into
mass that is probably less than dwarf is hot but tiny. space. It is amazing to recall that this superdense ball—with
70 percent of the mass of the a density of a ton per teaspoonful—actually began its life
original star. In the process, the star plummets down the billions of years earlier as a cloud of interstellar gas billions
left side of the H-R diagram, becoming smaller and fainter. of times more tenuous than the vacuum in the best vacuum
Within a few thousand years the burned-out core shrinks chamber on Earth.
to about the size of Earth, at which point it has become fully Figure 16.13 recaps the evolution of a solar-type 1-M⊙
degenerate and so can shrink no further. The degenerate main-sequence star through to its fi nal existence as a 0.6-
stellar cinder is now referred to as a white dwarf. The white M⊙ white dwarf. We point out once again that what we have
dwarf continues to radiate energy away into space, and as described is representative of the fate of low-mass stars.
it does so it cools, just as the filament of a lightbulb cools Although all low-mass stars form white dwarfs at the end
when the switch is turned off. Because the white dwarf is points of their evolution, the exact path a low-mass star fol-
electron-degenerate, its size does not change much as it lows from core hydrogen burning on the main sequence to
PLANETARY ASYMPTOTIC GIANT

NEBULA EJECTION BRANCH STAR
H-burning shell
He-burning shell
Degenerate C core
WHITE
DWARF
Degenerate HORIZONTAL
C cinder BRANCH STAR
107 H-burning shell

He-burning core
106 1,0
Post-AGB stars 00
R
£ HELIUM
105 Planetary
FLASH
nebula Planetary nebula ejection
central 10
0R
104 stars M £
AI
ch
N
ran
SE Helium
tb
103 Q
UE
10 i an flash
R
Luminosity (L£)
ic g
NC £
E
t
pto
102
ym
1R
As
nts
£
101 Co Horizontal
Red gia
oli
ng branch
w hit 0.1
1 ed R £
Runaway He burning
wa
rfs
£
in degenerate core
0.0 RED GIANT

10–1 1R BRANCH STAR
£
H-burning envelope
10–2 0.0
01 Degenerate He core
R£
10–3 MAIN-SEQUENCE
100,000 40,000 30,000 10,000 6,000 3,000 STAR (SUN)
Surface temperature (K) H-burning core
FIGURE 16.13 This H-R diagram summarizes the stages in the post–main sequence evolution of a 1-M⊙ star.
white dwarf depends on many details particular to the star. In our discussion of stellar evolution we have focused
Some stars less massive than the Sun may become white on what happens after a star leaves the main sequence.
dwarfs composed largely of helium rather than carbon. It is important, however, to remind ourselves that the
Temperatures in the cores of evolved 2-M⊙ to 3-M⊙ stars spectacle of a red giant or AGB star is ephemeral. The
are high enough to allow additional nuclear reactions to Sun will travel the path from subgiant to white dwarf
occur, leading to the formation of somewhat more massive in less than one-tenth of the time that it spends on the
white dwarfs composed of materials such as oxygen, neon, main sequence steadily burning hydrogen to helium in
and magnesium. Differences in chemical composition of a its core. Stars spend most of their luminous lifetimes on
star can also lead to dramatic differences in its post–main the main sequence, which is why most of the stars that
sequence evolution. Finally, we note that studies of the evo- we see in the sky are main-sequence stars. (Were it oth-
lution of stars whose main-sequence masses are less than erwise, the “main sequence” would not be “main” at all!)
about 0.8 M⊙ or so are pure theory. These stars are able to In the end, though, white dwarfs will carry the day. Too
burn hydrogen in their cores for longer than the current faint to be noticed as we look at the sky on a clear sum-
age of the universe. No star with a mass lower than that mer evening, white dwarfs still constitute the fi nal rest-
has ever evolved beyond the main sequence except in the ing place for the vast majority of stars that have been or
virtual world of a computer’s calculations. ever will be formed.
You may recall the name Édouard A. Roche from Chap-

16.6 Many Stars Evolve ters 10 and 11, when we discussed how the tidal disruption
of moons and comets can lead to the formation of planetary
as Pairs rings. The Roche limit describes how close a small object
might come to a planet before being torn apart by the plan-
Possibly the most significant complication in our picture et’s tides. Similarly, mass transfer between stars in a binary
of the evolution of low-mass stars arises from the fact that can be thought of as the “tidal stripping” of one star by the
many stars are members of binary systems. Binary stars were other. The physical principles at work in the two situations
quite useful to us when we were uncovering the properties are much the same, and Roche is responsible for early cal-
of stars. We used them as a sort of celestial bathroom scale, culations of both.
applying Kepler’s laws to the orbits of two stars about each
other to determine their masses. Although both members of
a binary pair are on the main sequence, they usually have
little effect on each other. When one of the stars evolves
Evolution of a Binary System
off the main sequence, however, it can swell to hundreds The best way to understand how mass transfer affects the
of times its original size, and its outer layers can literally evolution of stars in a binary system is to apply what we
cross the line between what belongs to the evolved star and have learned from the evolution of single low-mass stars.
what belongs to its companion. Shortly we will see the con- Figure 16.14a shows a binary system consisting of two low-
sequences of this behavior. mass stars. We will rather unimaginatively call the more
massive of the two stars “star 1” and the less massive of the
two “star 2.” This is an ordinary binary system, and each
Mass Flows from an Evolving of these stars is an ordinary main-sequence star for most of
the system’s lifetime. When one of the stars begins to evolve,
Star onto Its Companion however, things start to get interesting.
Think for a moment about what would happen if you were The more massive a main-sequence star is, the more
to travel in a spacecraft from Earth toward the Moon. When rapidly it evolves. Therefore, star 1, which is more massive,
you are still near Earth, the force of Earth’s gravity is far will be the fi rst to use up the hydrogen at its center and
stronger than that of the Moon. There is no question that begin to evolve off the main sequence (Figure 16.14b). If the
you “belong” to Earth, just as there is no question that you two stars are close enough together, star 1 will eventually
“belong” to Earth as you sit in your chair reading this book. grow to fill its Roche lobe, and the overflow from star 1 will
Yet, as you move away from Earth and closer to the Moon, start falling onto star 2 (Figure 16.14c). A number of inter-
this terrestrial dominance becomes less secure: the gravi- esting things can happen at this point. For example, the
tational attraction of Earth weakens, and the gravitational transfer of mass between the
attraction of the Moon becomes stronger. At a certain point two stars can result in a sort The more massive
you reach an intermediate zone where neither body has the of “drag” that causes the orbits star evolves first.
upper hand. If you continue beyond this point, the lunar of the two stars to shrink,
gravity begins to successfully assert itself until you fi nd bringing the stars closer together and further enhancing
yourself fi rmly in the grip of the Moon. mass loss. The two stars can even reach the point where
Exactly the same situation exists between two stars. they are effectively two cores sharing the same extended
Gas near each star clearly “belongs” to that star. But what envelope of material.
happens if, as a star swells up, its outer layers cross that Despite these complexities, star 2 probably remains a basi-
gravitational dividing line separating the star from its cally normal main-sequence star throughout this process,
companion? Any material that crosses this line no longer burning hydrogen in its core. However, it does not remain the
belongs to the fi rst star, but instead can be pulled toward same main-sequence star that it started out to be. The mass
the companion. A star reaches of star 2 increases, thanks to the largesse of its companion.
this point when it fi lls up its An evolving star As it does so, the structure of star 2 must change to accom-
portion of an imaginary fig- loses mass to its modate its new status as a higher-mass star. If we plotted
ure eight–shaped volume of companion. star 2’s position on the H-R diagram during this period, we
space, as shown in Figure would see it move up and to the left along the main sequence,
16.14. These regions surrounding the two stars—their becoming larger, hotter, and more luminous.
gravitational domains—are referred to as the Roche lobes While star 2 gains from the interaction, star 1 suffers. Star
of the system. Once the fi rst star has expanded to fi ll its 1 never gets to experience the full glory of being an isolated
Roche lobe, material begins to pour through the “neck” of red giant or AGB star, because star 1 can never grow larger
the figure eight and fall toward the other star. Astronomers than its Roche lobe. The presence of star 2 prevents star 1
refer to this exchange of material between the two stars as from swelling to the size of the behemoths that populate the
mass transfer. top of the H-R diagram. Yet star 1 continues to evolve, spend-
ing its allotted time burning helium in its core on the hori-
1 Two low-mass main- zontal branch, proceeding through a stage of helium shell
sequence stars orbit (a) Star 1 Star 2
their center of mass. burning, and finally losing its
outer layers and leaving behind The first star becomes
a white dwarf. Figure 16.14d a white dwarf orbiting
shows the binary system after a main-sequence star.
star 1 has completed its evolu-
Roche lobes tion. All that remains of star 1 is a white dwarf, orbiting about
(b)
its bloated main-sequence companion, star 2.
2 The more massive
star 1 begins to evolve…
Fireworks Occur When the

3 …until it overfills its
Roche lobe and begins
Second Star Evolves
transferring mass onto (c)
its companion, star 2.
The stage is now set for a sequence of events that can lead
to one of nature’s most spectacular displays. Figure 16.14e
picks up the evolution of the binary system as star 2 begins
4 Star 2 gains mass, to evolve off the main sequence. Like star 1 before it, star
becoming a hotter,
more luminous main-
2 grows to fill its Roche lobe; as it does, material from star
sequence star. 2 begins to pour through the “neck” connecting the Roche
(d) lobes of the two stars. However, this time the mass is not
White dwarf being added to a normal star but is drawn toward the tiny
5 Eventually star 1
leaves behind a white dwarf left behind by star 1. Because the white dwarf
white dwarf orbiting is so small, the infalling material generally misses the
together with the now star. Instead of landing directly on the white dwarf, the
more massive main-
sequence star 2. infalling mass forms an accretion disk around the white
dwarf, similar in some ways to the accretion disk that
forms around a protostar. As in the process of star forma-
6 When star 2 evolves (e)
beyond the main
tion, the accretion disk serves as a way station for material
sequence, it too that is destined to fi nd its way onto the white dwarf but
overfills its Roche lobe that starts out with too much angular momentum to hit the
and begins transferring
mass onto its white
white dwarf directly.
dwarf companion. We have already seen that a white dwarf has a mass
comparable to that of the Sun, but a size comparable to that
of Earth. A large mass and a
7 Different possible fates
may await star 1, including (f)
small radius mean strong Material from the
recurrent eruptions of nova gravity. In Chapter 6 we saw second star gets hot
explosions and possibly how the gravitational energy as it falls onto the
complete disintegration
in a Type I supernova. of material falling toward a white dwarf.
forming protostar is converted
(g) (h) into thermal energy when the material hits the accretion
disk around the protostar. The same principle applies here,
but with a vengeance. A kilogram of material falling from
space onto the surface of a white dwarf releases a hundred
times more energy than a kilogram of material falling from
the outer Solar System onto the surface of the Sun. The
material streaming toward the white dwarf in the binary
system falls down into an incredibly deep gravitational
8 If star 1 survives, 9 …but if star 1 explodes as a
two white dwarfs are Type I supernova, star 2 remains “well.” All of this energy has to go somewhere, and that
eventually left behind… as an isolated giant evolving to somewhere is thermal energy. The spot where the stream
become a lone white dwarf. of material from star 2 hits the accretion disk can be heated
FIGURE 16.14 A compact binary system consisting of two
to millions of kelvins, where it glows in the far ultraviolet
low-mass stars passes through a sequence of stages as the and X-ray parts of the electromagnetic spectrum.
stars evolve and mass is transferred back and forth. The Eventually the infalling material accumulates on the sur-
evolution of the binary system can lead to novae or even to a face of the white dwarf (Figure 16.15a), where it is compressed
Type I supernova that destroys the initially more massive star 1. by the enormous gravitational pull of the white dwarf to a
density close to that of the white dwarf itself. As more and

more material builds up on the surface of the white dwarf,
the white dwarf shrinks (just as the core of the red giant
shrinks as it grows more massive). The density increases more
and more, and at the same time the release of gravitational
energy drives the temperature of the white dwarf higher and (a)
higher. Now recall that the infalling material is from the
1 Hydrogen collects on
outer, unburned layers of star 2, so it is composed mostly of the surface of a white
hydrogen. We have a dangerous situation here. Hydrogen, dwarf in a binary system.
Degenerate
the best nuclear fuel around, is being compressed to higher carbon white
and higher densities and heated to higher and higher tem- dwarf
peratures on the surface of the white dwarf. Hydrogen skin
The mental picture you should have at this point is of accreted from
gasoline pooling on the floor of a match factory. Once the binary companion
temperature at the base of the layer of hydrogen reaches
about 10 million K, hydrogen begins to burn. But this is not (b) NOVA (c) COLLAPSE
the contained hydrogen burn- 2 The temperature in the 5 If the white dwarf mass
ing that takes place in the cen- Runaway burning of degenerate hydrogen exceeds the Chandrasekhar
ter of the Sun. Rather, this is hydrogen on a white skin climbs… limit, it begins to collapse…
explosive hydrogen burning dwarf causes a nova.

in a degenerate gas. Energy
released by hydrogen burning drives up the temperature,
and the higher temperature drives up the rate of hydrogen
burning. This runaway thermonuclear reaction is much like
the runaway helium burning that takes place during the
helium flash, except now there are no overlying layers of a
star to keep things contained. The result is a tremendous
explosion—a nova (plural: novae)—that blows part of the
layer covering the white dwarf out into space at speeds of 3 …until the hydrogen burns 6 …pushing up the
thousands of kilometers per second (see Figures 16.14f and explosively in a nova… temperature until carbon
16.15b). ignites and burns explosively.
About 50 novae are thought to occur in the Milky Way
Galaxy each year, but because our view is restricted by
interstellar extinction, we can see only a few of these, TYPE I SUPERNOVA
typically two or three each year. Novae get bright in a
hurry—typically reaching their peak brightness in only a
few hours—and for a brief time they can be almost half a
million times more luminous than the Sun. Although the
brightness of a nova sharply declines in the weeks follow-
ing the outburst, it can sometimes still be seen for years.
During this time the glow from the expanding cloud of
material ejected by the explosion is powered by the decay
of radioactive isotopes created in the explosion.
The explosion of a nova does not destroy the under- 4 …leaving the white dwarf 7 The Type I supernova
lying white dwarf star. In fact, much of the material that and companion to possibly consumes the white
repeat the show. dwarf completely.
had built up on the white dwarf can remain behind after
the explosion. The nova leaves the binary system in much FIGURE 16.15 (a) In a binary system in which mass is
the same configuration it was in previously—the configu- transferred onto a white dwarf, a skin of hydrogen builds up on
ration shown in Figure 16.15a—with material from star 2 the surface of the degenerate white dwarf. (b) If hydrogen burning
still pouring onto the white dwarf. This cycle can repeat ignites on the surface of the white dwarf, the result is a nova. (c)
itself many times, with material building up and igniting If enough hydrogen accumulates to force the white dwarf to begin
over and over again on the surface of the white dwarf. In to collapse, carbon ignites and the result is a Type I supernova.
most cases, outbursts are separated by thousands of years,
so most novae have been seen only once in historical times.
Some novae, however, are known to be recurrent, erupting
anew every decade or so.
fuse them. Once again, the onset of nuclear burning in a

A Stellar Cataclysm May Await the degenerate gas leads to a thermonuclear runaway, with ever-
higher temperatures and ever-faster nuclear burning feeding
White Dwarf in a Binary System off each other. But whereas the thermonuclear runaway in
It is possible that eventually star 2 will simply go on to form a nova involves only a thin shell of material on the surface
a white dwarf, leaving behind a binary system consisting of the white dwarf, runaway carbon burning involves the
of two white dwarfs, as in Figure 16.14g. There is another entire white dwarf. Within about a second, the whole white
possibility, however. Novae may be spectacular events, but dwarf is consumed in the resulting conflagration. In this
they pale in comparison with an alternate fate that may single instant, 100 times more energy is liberated than will
await the white dwarf in a binary system. Through millions be given off by the Sun over its entire 10-billion-year lifetime
of years of mass transfer from star 2 onto the white dwarf, on the main sequence. Runaway fusion reactions convert
and possibly through countless nova outbursts, the mass of a large fraction of the mass of the star into elements such
the white dwarf slowly increases. As it does, the white dwarf as iron and nickel, and the explosion blasts the shards of
shrinks, and the grip of self- the white dwarf into space at top speeds in excess of 20,000
gravity on this already The mass of a km/s. The explosion completely destroys star 1, leaving star
extremely dense object gets white dwarf cannot 2 behind as a lone giant to continue its evolution toward
tighter and tighter. The mass exceed 1.4 M . becoming a white dwarf (Figure 16.14h).
of a white dwarf can be pushed Explosive carbon burning in a white dwarf is a leading
only so far. There comes a point when even the pressure theory used to explain colossal events called Type I super-
supplied by degenerate electrons is no longer enough to bal- novae. Type I supernovae 6
ance gravity. This precipice occurs at a mass of about 1.4 occur in a galaxy the size of The carbon in a white
M⊙, a value referred to as the Chandrasekhar limit, named the Milky Way about once a dwarf can ignite in a
for Subrahmanyan Chandrasekhar (1910–1995), the Indian- century. For a brief time they Type I supernova.
born American astrophysicist who derived it. It is impos- can shine with a luminosity
sible for a star to increase in mass above 1.4 M⊙ and remain 10 billion times that of our Sun, possibly outshining the
a white dwarf. Raise its mass above the Chandrasekhar galaxy itself.
limit, and it will begin to collapse into even more extreme
forms of matter (Figure 16.15c).
As the white dwarf in our binary system approaches this 6
Strictly speaking, we should refer here to Type Ia supernovae. Tech-
limit, it grows smaller and smaller and hotter and hotter. As
nically, a “Type I” supernova is a supernova that shows no emission
the star crosses the Chandrasekhar limit, it begins to col- lines from hydrogen. This category includes some variants of the
lapse and the conversion of gravitational energy into thermal supernovae from massive stars that will be discussed in the next chap-
energy quickly drives the temperature in the white dwarf ter. We hope the purists will forgive our simplification of referring to
past the 6 × 108 K needed to slam carbon nuclei together all supernovae from white dwarfs as Type I and all supernovae from
with enough force to overcome their electric repulsion and massive stars as Type II.
lived low-mass stars. Yet eventually, even these steady

Seeing the Forest for the Trees performers must exhaust the hydrogen at their centers.
In this chapter we have followed the evolution of
Stars are born of the tenuous matter of interstellar space, a low-mass star like our Sun as it evolves beyond the
and then settle down onto the main sequence as stable main sequence, passing through a fascinating series of
and steady cosmic citizens. But the universe is a place evolutionary stages. Some of these stages are dramatic
of constant change and evolution. Stars consume the and violent, such as the runaway thermonuclear reac-
fuel they were born with, and when that fuel is gone tions of the helium flash or the ejection of a planetary
they must die. nebula that carries the chemical legacy of nuclear burn-
In the world of humans, it is wealth and power and ing deep within the dying AGB star. Other phases are far
birthright that determine our lifestyle. In the lives of more quiescent, such as the star’s time on the horizontal
stars, birthright is spelled “mass.” Mass determines the branch burning helium to carbon in its core and hydro-
weight bearing down on the interior of a star, which in gen to helium in a surrounding shell. Eventually, even
turn controls the rate of the nuclear reactions taking place an undistinguished star like our Sun may reach a lumi-
there. High-mass stars are spectacular but ephemeral, nosity rivaling all but the most luminous main-sequence
burning out the fuel in their cores in only a few million O stars, and swell to a size that would swallow much of
years. Our Sun, however, is typical of the much longer- our inner Solar System. Curiously, such a behemoth is
supported by frenzied nuclear reactions in a region sur- dent of our knowledge of the internal structure of the Sun
rounding the tiniest of cores—a superdense inert cinder and Earth. Physics works, and the closing decades of the
crushed by the weight of the overlying star to a size not 20th century saw steady improvements in the quality of
much larger than Earth. At a density of a thousand kilo- our calculations of the structure and evolution of stars.
grams per cubic centimeter, the electron-degenerate core And although we cannot witness the life cycle of a single
defies our normal notions of matter. star, when we look at the sky we fi nd many examples of
All of these stages are signposts along the star’s inexo- red giants, AGB stars, white dwarfs, and the other mem-
rable march toward its eventual fate, in which the outer bers of this evolutionary menagerie. Later on our journey
layers of the star will have been ejected into space and we will also see how clusters of stars that form together
nothing will remain but a tiny cooling cinder—a degener- provide snapshots that enable us to test our theories of
ate white dwarf that will eventually fade from the notice stellar evolution in exquisite detail.
of the rest of the universe. Yet if it happens to orbit about What we have seen is representative of the fate await-
a binary companion, the white dwarf may live again. ing the vast majority of stars in the universe. In the next
Unburned nuclear fuel pouring onto the surface of the chapter we turn our attention to the evolution of the small
white dwarf may trigger the episodic nuclear outbursts fraction of stars that fall into the category of high-mass
of novae or even a cataclysmic Type I supernova. stars. There we will fi nd that the extreme and counter-
We have never witnessed the life cycle of a low-mass intuitive behavior of white dwarfs presages a whole zoo
star from birth to death, nor will we. Not even a hundred of bizarre and exotic objects. Where low-mass stars have
million human lifetimes would be enough to see the saga bent our notions about matter, high-mass stars will shat-
through from beginning to end. Yet we are confident of ter those notions, leading us to consider forms of matter
most of what we have reported here, just as we are confi- so extreme that they cease to be matter at all.

• All stars eventually exhaust their nuclear fuel. questions, plus additional conceptual exercises. If
• The more massive a star is, the shorter will be its smartwork.wwnorton.com.
lifetime.
• Helium accumulates like ash in the cores of main-
sequence stars.
• After exhausting its hydrogen, a low-mass star leaves the
Student Questions
main sequence and swells to become a red giant.
• A red giant burns helium via the triple-alpha process.
1. Why do most nearby stars turn out to be low-mass, low-
• In their dying stages, some stars form planetary luminosity stars?
nebulae.
2. The Sun is slowly becoming more luminous as it ages
• All low-mass stars eventually become white dwarfs. on the main sequence. How do you think this change
will affect terrestrial life?
• A white dwarf is very hot but very small, and cannot
exceed 1.4 M⊙. 3. Is it possible for a star to skip the main sequence and
immediately begin burning helium in its core? Explain
• White dwarfs cool to become cold, dark cinders—the
your answer.
stellar graveyard.
4. What is the primary reason that the most massive stars
• Transfer of mass within a binary system can lead to
have the shortest lifetimes? (Note: The answer can be
novae or supernovae.
expressed in just a few words.)
5. Consider a hypothetical star having the same mass as
our Sun being one of the fi rst stars that formed when
the universe was still very young. Would you expect it
to be surrounded by planets? Explain your answer.
6. Describe some possible ways in which a star might continue to evolve. Include how you think the star will
increase the temperature within its core while at the evolve on the H-R diagram.
same time lowering its density.
19. As an AGB star evolves into a white dwarf, it runs out of
7. Suppose a main-sequence star were to suddenly start nuclear fuel, and one might argue that the star should
burning hydrogen at a faster rate in its core. How would cool off and move to the right of the H-R diagram. Why
the star react? Discuss changes in size, temperature, and does the star move instead to the left?
luminosity.
20. Why must a white dwarf move down and to the right
8. We typically say that the mass of a newly formed star along the H-R diagram?
determines its destiny from birth to death. However,
*21. Why does fusion in degenerate material always lead to
there is a frequent environmental circumstance for
a runaway reaction?
which this statement is not true. Identify this circum-
stance and explain why the birth mass of a star might 22. The intersection of the Roche lobes in a binary system
not fully account for its destiny. is the equilibrium point between the two stars where
the gravitational attraction from both stars is equally
9. What physical effect or explanation do you think could
strong and opposite in direction. Is this an example of
be responsible for the “break” in slope in the mass-lumi-
stable or unstable equilibrium? Explain.
nosity relationship between low- and high-mass stars,
as shown in Figure 16.1? Justify your answer. 23. Suppose the more massive red giant star in a binary sys-
tem engulfs its less massive main-sequence companion,
**10. Is it fair to assume that stars do not change their struc-
and their nuclear cores combine. What structure do you
ture while on the main sequence? Why or why not?
think the new star will have? Where will the star lie
11. Suppose Jupiter were not a planet, but instead were a G5 along the H-R diagram?
main-sequence star with a mass of 0.8 M⊙.
24. In Latin, nova means “new.” Novae, as we now know,
a. How do you think terrestrial life would be affected,
are not new stars. Explain how novae might have gotten
if at all?
their name and why they are really not new stars.
b. How would the Sun be affected as it came to the end
of its life? 25. T Corona Borealis is a well-known recurrent nova.
a. Is it a single star or a binary system? Explain.
12. When a star runs out of nuclear fuel at the core, why
b. What mechanism causes a nova to flare up?
does it become more luminous? Why does the surface
c. How can a nova flare-up happen more than once?
temperature of the star cool down?
13. When a star leaves the main sequence, its luminosity
increases tremendously. What does this increase in
luminosity imply about the amount of time the star has
left to live, or the amount of time the star spends on any
subsequent part of its evolutionary path? 26. For most stars on the main sequence, luminosity scales
with mass as M 3.5 (see Math Tools 16.1). What luminos-
14. Why are the paths along the H-R diagram that a star
ity does this relationship predict for (a) 0.5-M⊙ stars, (b)
follows as it forms (from a protostar) and as it leaves
6-M⊙ stars, and (c) 60-M⊙ stars? Compare these numbers
the main sequence (climbing the red giant branch) so
to values given in Table 16.1.
similar?
27. Compute the main-sequence lifetimes for (a) 0.5-M⊙ stars,
**15. When compressed, ordinary gas heats up but degenerate
(b) 6-M⊙ stars, and (c) 60-M⊙ stars. Compare them to the
gas does not. Why, then, does a degenerate core heat up
values given in Table 16.1.
as the star continues shell burning around it?
28. Notice in Figure 16.1 that low-mass and high-mass stars
16. Suppose you were an astronomer making a survey of
have different mass-luminosity relationships with the
the observable stars in our galaxy. What would be your
average value of M 3.5. What does this fact imply about the
chances of seeing a star undergoing the helium flash?
actual main-sequence lifetimes of a 0.5- or 6-M⊙ star, as
compared to the predictions you made in question 27?
________
*17. Why is a horizontal branch star (which burns helium at a
29. The escape velocity (νesc = √ 2 GM⊙/r⊙ ) from the Sun’s
high temperature) less luminous than a red giant branch
surface today is 618 km/s.
star (which burns hydrogen at a lower temperature)?
a. What will the escape velocity be when the Sun
*18. Suppose a star is able to heat its core temperature high becomes a red giant with a radius 50 times greater
enough to begin fusing oxygen. Predict how the star will and a mass only 0.9 times that of today?
b. What will it be when the Sun becomes an AGB star 34. In typical binary systems, a red giant can transfer mass
with a radius 200 times greater and a mass only 0.7 onto a white dwarf at rates of about 10 –9 M⊙ per year.
times that of today? Roughly how long after mass transfer begins will the
c. How would these changes in escape velocity affect white dwarf undergo a Type I supernova? How does
mass loss from the surface of the Sun as a red giant, this length of time compare to the typical lifetime of a
and later as an AGB star? low-mass star? (Hint: Assume that a typical white dwarf
starts with a mass of 0.6 M⊙.)
30. Each form of energy generation in stars depends on
temperature. **35. How fast does material in an accretion disk orbit around
a. The rate of hydrogen fusion (proton-proton chain) near a white dwarf? (Hint: Use Kepler’s third law.)
107 K increases with temperature as T 5. If you raise
36. A white dwarf has a density of approximately 109 kilo-
the temperature of the hydrogen-burning core by 10
grams per cubic meter (kg/m3). Earth has an average den-
percent, how much does the hydrogen fusion energy
sity of 5,500 kg/m3 and a diameter of 12,700 kilometers
increase?
(km). If Earth were compressed to the same density as a
b. Helium fusion (the triple-alpha process) at 108 K
white dwarf, how large would it be?
increases with an increase in temperature at a rate of
T 38. If you raise the temperature of the helium-burning 37. What is the density of degenerate material? Calculate
core by 10 percent, how much does the helium fusion how large the Sun would be if all of its mass were
energy increase? degenerate.
31. After leaving the main sequence, low-mass stars “seed” 38. Recall from Chapter 4 that the luminosity of a spherical
the interstellar medium with heavy elements such as object at temperature T is given by L = 4πR 2 σT 4, where
carbon, nitrogen, and oxygen. Assume that the universe R is the object’s radius. If the Sun became a white dwarf
is approximately 1.4 × 1010 (14 billion) years old and that with a radius of 107 meters, what would be its luminos-
the Sun is 4.6 billion years old. Refer to Table 16.1 for ity at temperatures of (a) 108 K, (b) 106 K, (c) 104 K, and
the masses and main-sequence lifetimes of stars. How (d) 102 K?
many generations of each type of star preceded the birth
of our Sun and Solar System?
32. Assume that there are twice as many F0 stars as A0
stars, and that each type contributes heavy elements to
the interstellar medium in approximate proportion to its
main-sequence mass. What was the relative contribution
from the two types of stars before the Sun was born?
33. Roughly how large does a planetary nebula grow before
it disperses? Use an expansion rate of 20 km/s and a life-
time of 50,000 years.
It does not do to leave a live dragon out of your calculations, if
you live near him.
J. R. R. TOLK IEN (1892–1973), THE HOBBIT
The inner Crab Nebula seen in X-ray (blue) and visible (red) light.
CHAP TER 17
Live Fast, Die Young—

High-Mass Stellar Evolution
K E Y CONCEP T S
17.1 Here Be Dragons
Our Sun is but a pale ember, compared with the
So far in our discussion of the lives of stars, we have con- brilliance of more massive stars. These stars consume
centrated on what happens to low-mass stars like our Sun. their fuel in a cosmic blink of the eye, and then in their
These are stars that steadily burn hydrogen to helium in death they dazzle us with brilliant fireworks, seed the
their cores and produce a fairly constant output of energy universe with the elements of life, and force us to once
for billions of years—long enough for our Earth to cool from again alter our concept of space and time. As our
a molten ball of rock and for life to evolve to the point of exploration of stars draws to a close, we will discover
contemplating its place in the universe. High-mass stars—
stars with masses greater than about 8 solar masses (M⊙)— • How the structure of high-mass stars differs from that
are very different beasts. These are objects that burn with of low-mass stars.
luminosities thousands or even millions of times as great • The stages of nuclear burning that evolving high-mass
as the luminosity of our Sun and squander their generous stars experience as they form progressively more
allotments of nuclear fuel in less time than mammals have massive chemical elements.
walked on Earth.
• That massive stars die as Type II supernovae.
The differences between the ways in which high-mass
stars and low-mass stars evolve follow from the same rela- • The origin of the most massive chemical elements.
tionship among gravity, pressure, and the rate of nuclear • A menagerie of bizarre objects, including neutron
burning that determines the balance in a star like our Sun. stars, pulsars, and X-ray binary stars.
The litany should be familiar by now: More mass means
stronger gravity and therefore more weight bearing down • How H-R diagrams enable us to measure the ages of
on the inner parts of the star. The increase in weight means stars and test our theories of stellar evolution.
higher pressure; higher pressure means faster reaction rates; • That the speed of light is a universal constant, which
and faster reaction rates mean greater luminosity. There forces us to completely rethink classical physics.
are many differences in the evolution of low- and high-
• That gravity is a consequence of the way mass distorts
mass stars, but in the end they all trace back to the greater
the very shape of spacetime.
gravitational force bearing down on the interior of a high-
mass star. • The bottomless pits in spacetime that we call black
holes.
469
470 Chapter 17 Live Fast, Die Young—High-Mass Stellar Evolution
(a) CNO cycle FIGURE 17.1 In high-mass

Gamma rays () stars, carbon serves as a
Gamma radiation
catalyst for fusion of hydrogen
12C e– to helium. This process is called
Positron (e+) the carbon-nitrogen-oxygen (or
1
H
Neutrino () CNO) cycle.
13
N Gamma rays ()

13C e–
1H Positron (e+)
14N
Neutrino () 4He
1
(b) Net reaction H 15O
12
C is a catalyst for H burning. 15N
1
H
+ + + + +
12C
12C + 4 1H + 2 e– 12C + 4He + 2 + 7
17.2 High-Mass Stars Follow a lot of energy to get past the electric barrier that repels a
carbon nucleus, with its six protons, from a hydrogen
Their Own Path nucleus. But if this barrier
can be overcome, the strong- The CNO cycle
Recall from Chapter 14 that the fi rst step in the hydrogen- nuclea r-force interaction burns hydrogen in
burning proton-proton chain is the collision and fusion of between the carbon and the massive stars.
two protons. Protons have only a single positive charge, hydrogen is much more favor-
so they are the easiest atomic nuclei to force close enough able than the interaction between two hydrogen nuclei.
together to fuse. This is a great advantage for low-mass stars, This reaction, in which a carbon-12 (12C) nucleus and a
in which hydrogen burns at temperatures as low as a few proton combine to form a nitrogen-13 (13N) nucleus consist-
million kelvins. However, the proton-proton chain suffers ing of seven protons and six neutrons, is the fi rst reaction
from a large disadvantage as well. The “affi nity” between in a chain of events that is referred to as the carbon-nitro-
two protons is not very great. Even when two protons are gen-oxygen (CNO) cycle. The remaining steps in the CNO
slammed together hard enough for the strong nuclear force cycle are shown in Figure 17.1. Note that carbon is not
to act on them, the likelihood that they will fuse is low. The consumed by the CNO cycle but instead is a catalyst. The
12
lack of vigor of this fi rst step in the proton-proton chain C nucleus that started the cycle is there again at the end
limits how rapidly the entire process can run. of the cycle, but now instead of the four hydrogen nuclei
At the much higher temperatures at the center of a high- that took part in the chain of reactions, we have a helium
mass star, additional nuclear reactions become possible. nucleus. Figure 17.2 shows that the CNO cycle becomes
In particular, hydrogen nuclei are able to interact with the far more efficient than the proton-proton chain in stars
nuclei of more massive elements such as carbon. It takes more massive than 1.5 M⊙.
The CNO cycle produces FIGURE 17. 2 Plots of the

Mass of main-sequence star most energy in main-sequence rate of energy generation as
1010
Energy generation from H burning
with specified central temperature stars above 1.5 M£. a function of temperature for
the proton-proton chain and
0.5 M£
1.0 M£
1.5 M£
3 M£
7 M£
108 the CNO cycle. At the higher

16 M£
25 M£
40 M£
70 M£
150 M£
central temperatures of stars

more massive than 1.5 M⊙, the
(W/kg)
106 Proton-proton chain produces

most energy in main-sequence CNO cycle more efficiently fuses
stars below 1.5 M£. hydrogen into helium.
104
102
100
0 10 20 30 40
Central temperature (million K)
(a) t = 0 years
100 visible differences in its structure and evolution become far
80 more pronounced. As the high-mass star runs out of hydro-
Hydrogen gen in its core, the weight of the overlying star compresses
1 Convection in the core of a 60 the core, just as it does in a low-mass star. Yet long before the
massive main-sequence
star “mixes” material… 40 core of the high-mass star becomes electron-degenerate, the
pressure and temperature in the core reach the 108-kelvin
20 (K) point needed for helium burning to begin. There will
Helium be no growing degenerate core in the high-mass star. There
0
will be no accelerating ascent up the red giant and asymp-
(b) t = 4 million years
100 totic giant branches on the H-R diagram. Instead, the star
2 …so as hydrogen makes a fairly smooth transition from hydrogen burning to
Percentage of mass
80 burns, helium builds up

uniformly throughout helium burning. The overall structure of the star responds
60 the core. to the changes taking place in its interior, but its luminos-
Hydrogen ity will change relatively little.
40 When a low-mass star leaves the main sequence, the path
20 that it follows on the H-R diagram is largely vertical (see
Helium Figure 16.4), going to higher
0 and higher luminosities. As High-mass stars skip
(c) t = 7 million years the high-mass star leaves the the red giant phase.
100
main sequence, on the other
80 hand, it grows in size while its surface temperature falls,
so it moves mostly off to the right on the H-R diagram (Fig-
60
ure 17.4). The massive star now has the same structure as
Hydrogen a low-mass horizontal branch star, burning helium in its
40
20
Helium
0
0.0 0.2 0.4 0.6 0.8 1.0
Center Surface FIGURE 17. 4 When massive stars leave the main sequence,
Fraction of radius (R/R£) they move horizontally across the H-R diagram.
FIGURE 17. 3 Convection keeps the core of a high-mass
High-mass stars move As stars pass through
main-sequence star well mixed, so the composition remains
horizontally back and forth the “instability strip,”
uniform throughout the evolving core. (Evolution times are for 107 across the H-R diagram after they become pulsating
a 25-M⊙ star.) leaving the main sequence. variable stars.
106 –10
The different ways that hydrogen burning takes 105

place in high- and low-mass stars are reflected in
the different structures of their cores. The tempera- 104 –5

ture gradient in the
core of a high-mass Convection stirs 103
Luminosity (L£)
More luminous
star is so steep that the core of a
convection sets in high-mass star. 102 0
within the core itself, 1R
“stirring” the core like the water in a boiling pot. 101
£
Compare Figure 17.3 with Figure 16.2. Rather than

building up from the center outward, helium ash is 1 £ +5
spread uniformly throughout the core of a high-mass
star as the star consumes its hydrogen. Main
10–1
sequence
10–2 +10
The High-Mass Star Leaves
the Main Sequence 10–3
40,000 30,000 10,000 6,000 3,000
These distinctions between high- and low-mass Surface temperature (K)
stars might seem a bit esoteric, but as the high-mass Spectral
star’s life on the main sequence comes to an end, the type O5 B0 B5 A0 F0 G0 K5 M5
E X C U R S I O N S 1 7.1
Each Star Is Unique

The distinction made in this book between low-mass and high- massive stars do, burning helium in their cores immediately
mass stars is a useful, convenient way to think about stellar after their hydrogen is exhausted and skipping the red giant
evolution. After all, there is a fundamental and qualitative and helium fl ash phases of low-mass star evolution. When
difference between stars that end their lives quietly as dying helium burning in the core is complete, however, the tem-
cinders and those that die in spectacular explosions. Even so, perature at the center of an intermediate-mass star is too low
this distinction is an oversimplification. In reality, each star for carbon to burn. From this point on, the star evolves more
is unique, with its own individual mass and composition and like a low-mass star, ascending the asymptotic giant branch,
its own particular circumstances. burning helium and hydrogen in shells around a degenerate
Stars with masses between about 3 and 8 M ⊙ exist in core, and then ejecting its outer layers and leaving behind a
something of a gray area between the low-mass stars dis- white dwarf. Yet even at the end, the star carries the signa-
cussed in Chapter 16 and the high-mass stars discussed in ture of its early “high-mass” years. The chemical composi-
this chapter. Like massive stars, these intermediate-mass tions of planetary nebulae and white dwarfs left behind by
stars, despite being on the main sequence, burn hydrogen intermediate-mass stars can be quite distinct from those of
via the CNO cycle. They also leave the main sequence as truly low-mass stars.
core and hydrogen in a surrounding shell, but it has skipped massive elements, including sodium, neon, and magnesium.
the red giant phase of low-mass star evolution. (Some less The star now consists of a carbon-burning core surrounded
massive stars skip the red giant phase as well. See Excur- by a helium-burning shell, which in turn is surrounded by
sions 17.1 for a more nuanced view of the distinction we a hydrogen-burning shell. The sequence does not end here.
have drawn between low-mass and high-mass stars.) When carbon is exhausted as a nuclear fuel at the center of
The next stage in the evolution of a high-mass star has the star, neon burning picks up the slack; and when neon
no analog in low-mass stars. is exhausted, oxygen begins to burn. The structure of the
When the high-mass star Progressively more evolving high-mass star, shown in Figure 17.5, is reminis-
exhausts the helium in its massive elements cent of an onion, with layer surrounding layer surrounding
core, the core again begins to burn as massive layer. As we move inward toward the center of the star, we
collapse, but this time as the stars evolve. pass through a layer of hydrogen burning, then a layer of
core collapses it reaches tem- helium burning, then a layer of carbon burning, and so on.
peratures of 8 × 108 K or higher, and carbon begins to burn Finally, we reach the most advanced stages of nuclear burn-
(see Table 17.1). Carbon burning produces a number of more ing in the core of the star.
TA B L E 1 7.1
Burning Stages in High-Mass Stars
Typical Core
Core Burning Stage 9-M⊙ Star 25-M ⊙ Star Temperatures
Hydrogen (H) burning 20 million years 7 million years (3–10) × 107 K
Helium (He) burning 2 million years 700,000 years (1–7.5) × 108 K
Carbon (C) burning 380 years 160 years (0.8–1.4) × 109 K
Neon (Ne) burning 1.1 years 1 year (1.4–1.7) × 109 K
Oxygen (O) burning 8 months 6 months (1.8–2.8) × 109 K
Silicon (Si) burning 4 days 1 day (2.8–4) × 109 K

MASSIVE “what if” the Sun, considering the consequences of forcing

MAIN-SEQUENCE H burning to He
the Sun to be larger or smaller than it is, we find that changes
STAR in convective core
in the star would cause it to settle back into this unique bal-
Nonburning
1 A massive main-
envelope ance between the inward pull
sequence star
burns hydrogen in
of gravity and the outward Evolved stars pulsate
a convective core… push of pressure. This rela- while passing through
tionship remains true for high- the instability strip.
He burning to C
mass stars on the main
H burning to He
sequence as well. However, it is untrue for many evolved
Nonburning
envelope stars. As a star undergoes post–main sequence evolution, it
may make one or more passes through a region of the H-R
2 …then immediately diagram known as the instability strip (see Figure 17.4).
begins burning helium Rather than finding a steady balance, stars in the instability
in its core as it leaves
the main sequence. strip pulsate, alternately growing larger and smaller.
Stars lying within the instability strip of the H-R dia-
gram are heat engines, powering their pulsation by tapping
C burning to
Na, Ne, Mg into the thermal energy flowing outward through them.
He burning to C Changes in the ionization state of the gas within the star
H burning to He
alternately trap thermal energy, forcing the pressure in the
Nonburning
star up and causing it to expand, and then they release this
envelope energy, allowing the star to contract again. These changes
serve the role of the valves in an engine: they rob the star
of pressure support at one part of its cycle, allowing it to
shrink too far, and then pump the pressure in the star back
3 The ash of one reaction up at a different point in its cycle, puffi ng up the star to a
becomes the fuel for the size larger than it can support. The process is illustrated in
next as the star evolves
a layered structure.
Figure 17.6. Rather than settling at a constant radius, like
Degenerate Fe core
the Sun, the star alternately expands and shrinks, moving
S, Si burning to Fe
HIGHLY EVOLVED out and in like the piston of a steam engine. The pulsations
MASSIVE STAR O burning to S, Si in the outer parts of the star have very little effect on the
Ne burning to O, Mg nuclear burning in the star’s interior. However, the pulsa-
C burning to Na, tions do affect the light escaping from the star. Recall from
Ne, Mg
Chapter 13 the luminosity-temperature-radius relationship
He burning to C
for stars: both the luminosity and the color of the star change
H burning to He
as the star expands and shrinks. The star is at its brightest
Nonburning
envelope
and bluest while it expands through its equilibrium size,
and at its faintest and reddest while it falls back inward.
5 As in a low-mass
red giant, burning
Such stars are referred to as pulsating variable stars.
occurs in a tiny, The most massive and luminous pulsating variable stars
4 The end: A degenerate iron dense region at the are named Cepheid variables, after Delta Cephei, the fi rst
core sits within “onionlike” center of the
shells of progressive 0.01 R£ giant bloated star. recognized member of this class. A Cepheid variable com-
nuclear burning. pletes one cycle of its pulsation in anywhere from about 1
1,000 R£
to 100 days, depending on its luminosity. The more lumi-
FIGURE 17. 5 As a high-mass star evolves, it builds up a nous the star, the longer it takes it to complete its cycle. This
layered structure like that of an onion, with progressively more period-luminosity relationship for Cepheid variables, fi rst
advanced stages of nuclear burning found deeper and deeper discovered experimentally by Henrietta Leavitt (1868–1921)
within the star. Note the change in scale for the bottom image. in 1912, is the basis for the use of Cepheid variables as indi-
cators of the distances to galaxies beyond our own.
Cepheid variables are not the only type of variable star.
The horizontal branch of the evolutionary tracks of low-mass
stars (see Figure 16.8) may pass through the instability strip
Not All Stars Are Stable as well. These unstable horizontal branch stars are known
When we discussed the structure of the Sun in Chapter 14 as RR Lyrae variables after their prototype. They pulsate
and a low-mass star in Chapter 16, we asserted that a single by the same mechanism as Cepheid variables but are typi-
stable solution always exists. Given one solar mass with a cally hundreds of times less luminous. The instability strip
mix of 70 percent hydrogen and 30 percent helium, that mass also intersects the main sequence around spectral type A,
will form a stable star with the structure of our Sun. As we and many A stars do show significant variability (although
Velocity (red)
Net force due

to gravity and 1 The star falls inward. As
pressure (blue) the star contracts,…
8 …until helium has all recombined. 2 …thermal energy is used

The star coasts until it stops and to ionize helium, lowering
begins to fall back inward. the temperature…
7 …adding to the pressure, 3 …robbing the star of pressure

which pushes the star past support, allowing it to fall
its equilibrium size… through its equilibrium size…
6 As it expands, recombining 4 …until helium is all

helium releases energy, ionized and pressure
driving up the temperature… starts to build up…
5 …stopping the
Equilibrium contraction. The
radius star “bounces.”
FIGURE 17.6 In a pulsating Cepheid variable, ionization of atoms in the star’s interior acts like
the valves in a steam engine, alternately allowing the surface of the star to fall inward, and then
pushing it back out again. (Color changes shown here are greatly exaggerated.)
it is much less pronounced than the variability in Cepheid all wavelengths) of 5 million Suns. Currently, Eta Carinae
or RR Lyrae variables). A number of other kinds of variable is losing mass at a rate of about 10 –3 M⊙ per year (or 1 M⊙
stars are seen elsewhere in the H-R diagram, each driven every 1,000 years). However, during a 19th century erup-
by its own variety of “engine.” tion, when Eta Carinae became the second-brightest star in
High-mass stars also change their composition by vio- the sky, its mass loss must have reached the amazing rate of
lently expelling a significant percentage of their mass back 0.1 M⊙ per year, shedding 2 M⊙ of material over a mere 20
into space throughout their existence. Even while on the years. Eta Carinae is expected to explode as a supernova in
main sequence, massive O the astronomically near future.
and B stars drive away tenu- Massive stars lose
ous winds at velocities that mass as they generate
can reach 3,000 kilometers high-velocity winds.
per second (km/s). These 17.3 High-Mass Stars Go
winds are pushed outward by the pressure of the radiation
from the star. We do not normally think of light as “push- Out with a Bang
ing” on something, but the pressure of the intense radiation
at the surface of a massive star is enough to overcome the We have already seen how a low-mass star approaches the
star’s gravity and drive away material in its outermost lay- end of its life—relatively slowly and gently, ejecting its outer
ers. Main-sequence O and B stars lose mass at rates ranging parts into nearby space (some-
from about 10 –7 to 10 –5 M⊙ of material per year, with the times forming a planetary Iron is the most
greatest mass loss occurring in the most massive stars. nebula) and leaving behind a massive element
These numbers may sound tiny, but added up over mil- degenerate core. In stark con- formed by fusion.
lions of years they mean that mass loss plays a prominent trast, for a high-mass star the
role in the evolution of high-mass stars. O stars with masses end comes suddenly and amid considerable fury. An evolv-
of 20 M⊙ or more may lose about 20 percent of their mass ing high-mass star builds up its onionlike structure (see
while on the main sequence, and possibly more than 50 per- Figure 17.5) as nuclear burning in its interior proceeds to
cent of their mass over their entire lifetimes. Even an 8-M⊙ more and more advanced stages. Hydrogen burns to helium,
star may lose 5–10 percent of its mass. An extreme example helium burns to carbon and oxygen, carbon burns to mag-
of mass loss in a massive star is seen in Eta Carinae (Fig- nesium, oxygen burns to sulfur and silicon, and then silicon
ure 17.7), a 100-M⊙ star with a luminosity (summed over and sulfur burn to iron. Many different types of nuclear
FIGURE 17.7 This Hubble

Space Telescope image shows
an expanding cloud of dusty
material ejected by the luminous
blue variable star Eta Carinae.
The star itself, which is largely
hidden by the surrounding dust,
has a luminosity 5 million times
that of the Sun and a mass
probably in excess of 100 M⊙.
Dust is created when volatile
material ejected from the star
condenses.
9 The energy it would take to break up an atomic nucleus

Fe Zr into its constituent parts is called the binding energy of the
S Cu Ba
8 Si nucleus. The net energy
Mg Pb
O Ne released by a nuclear reaction Different nuclei have
U
Binding energy per kilogram (1014 J)
C
7 He N is the difference between the different binding
Fusing less massive elements to binding energy of the prod- energies.
6 more massive elements does not ucts and the binding energy
release energy. Iron (Fe) and more of the reactants:
5 massive elements do not burn.
4
Net energy = (Binding energy
of products
- ) (
Binding energy
of reactants
)
Fusing less massive elements Figure 17.8 shows a plot of the nuclear binding energy per
3 to more massive elements kilogram for a number of different atomic nuclei.
releases energy. Elements
like H and He burn. It is worth making a brief digression to see how this
2 works in practice. Using these values to work through
the previous example of the triple-alpha process, we fi nd
1 that the binding energy of a helium nucleus is 6.822 × 1014
H
joules (J) per kilogram of helium. The binding energy of a
0 12
C nucleus is 7.410 × 1014 J per kilogram of carbon. To find
1 10 100
Mass of nucleus (atomic mass units) the amount of energy available from fusing 1 kilogram (kg)
of helium nuclei into carbon, take the difference between
FIGURE 17.8 The nuclear binding energy of a kilogram of
the two numbers:
material is plotted against the mass of the atomic nuclei. This is
the energy that it would take to break a kilogram of the material
apart into protons and neutrons. Energy is released by nuclear
fusion only if the binding energy of the products is greater than
(
Net energy from
burning 1 kg He
=) (
Binding energy
of C formed
- ) (
Binding energy
of He burned
)
the binding energy of the reactants. = 7.410 × 1014 J - 6.822 × 1014 J
= 5.88 × 1013 J
reactions occur up to this point, forming almost all of the This release of net energy indicates that helium is a good
different stable isotopes of elements less massive than iron. nuclear fuel, as Figure 17.8 shows directly. Moving from
But the essential point is this: With iron, the chain of nuclear helium to carbon on the plot increases the binding energy
fusion stops. of each nucleon,1 so fusing helium to carbon releases energy.
Why does hydrogen burn while iron does not? More But what if we instead try to fuse iron to form more massive
generally, what must be true of a material for it to serve as elements? Because iron is at the peak of the binding-energy
fuel? Gasoline burns because when it chemically combines curve, the products of iron burning will have less binding
with oxygen, energy is released. This extra energy pushes energy than the reactants (going from iron to more massive
up the temperature, which then causes the chemical reac- elements takes us down on the binding-energy curve in Fig-
tion between gasoline and oxygen to go faster. The reaction ure 17.8), so the net energy in the reaction will be negative.
is “self-sustaining,” meaning that the reaction itself is the Rather than producing energy, fusion of iron uses energy.
source of thermal energy needed to cause the reaction to Iron does not burn.2
go. The same is true of nuclear fusion reactions in the inte-
riors of stars. When four hydrogen atoms combine to form
a helium atom, the resulting helium atom has less energy The Final Days in the Life
than the four hydrogen atoms had separately. This differ-
ence in energy is converted to thermal energy, which main-
of a Massive Star
tains the temperature of the gas at the high levels needed The nuclear reactions following hydrogen burning are ener-
to sustain the reaction. getically much less favorable than conversion of hydrogen to
How much energy is available from a nuclear reaction helium. A look at Figure 17.8 shows that conversion of 1 kg
of combining, say, three helium nuclei to form a 12C nucleus of helium to carbon produces less than 1⁄10 as much energy
(the triple-alpha process)? We can answer the question in
steps. First, how much energy would it take to break down 1
A “nucleon” is a constituent of an atomic nucleus—that is, a proton
each of the three helium nuclei into its constituent six neu- or a neutron.
trons and six protons? Next, how much energy would be 2
Iron does not burn in a nuclear sense, but iron does burn chemically.
released if these six protons and neutrons combined to form For example, combining iron and oxygen increases the chemical bind-
a 12C nucleus? The net energy produced by the reaction is ing energy of the atoms involved in the reaction. However, chemical
just the difference between these two amounts. burning plays no role in the interior of a star.
as conversion of 1 kg of hydrogen to helium. In order to star burning silicon is actually giving off about 200 million
support the star against gravity, this less efficient nuclear times more energy per second than its former self did!
fuel must be consumed more rapidly. Although conversion Following silicon burning, we come to the end of the
of hydrogen into helium can provide the energy needed to line. Once a star forms its iron core, no source of nuclear
support the high-mass star against the force of gravity for energy remains to replenish the energy that is being taken
millions of years, helium burning can support the star for away by escaping neutrinos. The high-mass star’s life as a
only a few hundred thousand years. balancing act between gravity and a controlled thermonu-
Following helium burning, the nature of the balance clear furnace is over. Gravity will have its way.
within the star becomes qualitatively different. There is
almost as much energy available from burning a kilogram
of carbon, neon, oxygen, or silicon as there is from burn- The Core Collapses and
ing a kilogram of helium. We might expect, then, that each
of these stages would last about as long as helium burning.
the Star Explodes
Yet when we look at Table 17.1, we see that the star proceeds Throughout the story of stellar evolution, we have seen that
from helium burning to the end of its life in a cosmic blink although gravity can be held off for a time—perhaps a very
of the eye. So where is the catch? long time—it is infi nitely patient, and in the end it gener-
You can think of the balance in a star as something like ally wins the battle. Bereft of support from thermonuclear
trying to keep a leaky balloon inflated. The larger the leak, fusion, the iron core of the massive star begins to collapse.
the more rapidly you have to pump air into the balloon. A Refer to Figure 17.9 as we follow the dramatic sequence of
star that is burning hydrogen or helium is like a balloon events that comes next.
with a slow leak. At the temperatures generated by hydro- The early stages of the collapse of the iron ash core of
gen or helium burning, energy leaks out of the interior of a an evolved massive star are much the same as in the col-
star primarily by radiation and convection. Neither of these lapse of a nonburning core in a low-mass star. As the core
processes is very efficient, because the outer layers of the collapses, its density and temperature skyrocket, and the
star act like a thick, warm blanket. Nuclear fuels need to force of gravity becomes even stronger. The gas in the core is
burn at only a relatively modest rate to support the weight once again compressed beyond the realm of normal matter,
of the outer layers of the star while keeping up with the becoming electron-degenerate when it reaches the approxi-
energy escaping outward. mate size of Earth. Unlike the electron-degenerate core of a
Beginning with carbon burning, this balance shifts in a low-mass red giant, however, the weight bearing down on
dramatic and fundamental way. Rather than being carried the interior of the iron ash core is too great to be held up
by radiation and convection, energy now begins to escape by electron degeneracy. Here, gravity is too strong even for
from the core primarily in the form of neutrinos produced the quantum mechanical rules that limit how tightly elec-
by the many nuclear reactions occurring there. When this trons can be packed together. As the collapse continues,
happens, the floodgates are opened. Like air pouring out the core reaches temperatures of 10 billion K (1010 K) and
through a huge rent in the side of a balloon, neutrinos pro- higher, while the density exceeds 10 metric tons per cubic
duced in the interior of the star stream through the overly- centimeter, or 10 times the density of an electron-degenerate
ing layers of the star as if they were not even there, carrying white dwarf.
the energy from the stellar interior out into space. Keeping These phenomenal temperatures and pressures trigger
the balloon inflated becomes difficult indeed. As thermal fundamental changes in the makeup of the core. The laws
energy pours out of the interior of the star, the outer layers describing thermal radiation say that at these temperatures
of the star push inward, driving up the density and tem- the nucleus of the star will be awash in extremely ener-
perature, and forcing nuclear reactions to run at the furious getic thermal radiation. This radiation is so energetic that
rate necessary to replace the energy escaping in the form thermal gamma-ray photons are produced with enough
of neutrinos. energy to break iron nuclei apart into helium nuclei. This
Once this process of neutrino cooling becomes signifi- process, known as “photodisintegration,” literally begins
cant, the star begins evolving much more rapidly. Carbon undoing the results of nuclear fusion—a task that uses up
burning is capable of supporting the star for only something a tremendous amount of the thermal energy of the core. At
less than about a thousand the same time, the density of the core is so great that elec-
years. Oxygen burning holds Stages of burning trons are squeezed into atomic nuclei, where they combine
the star up for only about a after helium burning with protons to produce neutron-rich isotopes in the core
year. Once a massive star are progressively of the star. This process uses up thermal energy as well,
reaches the point that it begins shorter-lived. robbing the core of even more of its pressure support. All
to burn silicon, it is within a the while, neutrinos continue to pour out of the core of
few days of the end. The luminosity of a silicon-burning the dying star. These events take place within a second!
star in electromagnetic radiation may be not much more The collapse of the core accelerates, reaching velocities of
than it was when the star was burning helium in its core. 70,000 km/s, or almost one-fourth the speed of light, on
But if our eyes could sense neutrinos, we would see that a its inward fall.
1 Not even electron degeneracy FIGURE 17.9 Shown here are the
pressure can stop the collapse stages that a high-mass star goes
of an iron ash core. through at the end of its life as its core
collapses and the star explodes as a
Type II supernova.
Iron nucleus
Neutrinos Iron core of evolved
streaming from massive star Gamma rays
collapsing core
2 As the core collapses, the core temperature

climbs so high that thermal gamma-ray photons
photodisintegrate iron…
Electron (–)
3 …and the core becomes so dense Neutron

that electrons are absorbed by protons
in atomic nuclei, forming neutrons
and releasing energetic neutrinos. Proton (+) Neutrino ()
4 Photodisintegration and
electron absorption rob
the core of pressure support.
The collapse accelerates…
Neutron star
5 …until nuclear forces suddenly become
repulsive. The overcompressed core
bounces, driving its outer layers
outward through the star.
9 …and leaving behind the

collapsed remains of the
core, a neutron star.
8 …blasting forth in a
Type II supernova…
Shock wave 7 The shock continues through

the outer layers of the star…
6 The expanding shock is strengthened

by the pressure of a hot bubble of
trapped neutrinos from the core.
The next hurdle standing in the way of the collapsing Under the extreme conditions in the center of the star,
core is the force that holds atomic nuclei together. As mate- neutrinos are being produced at an enormous rate. Over
rial in the collapsing core reaches and exceeds the density of the next second or so, almost a fifth of the mass of the mate-
an atomic nucleus, the strong nuclear force actually becomes rial in the core is converted into neutrinos and their energy.
repulsive. Computer models say that about half of the col- Most of these neutrinos pour outward through the star; but
lapsing core suddenly slows its inward fall. The remaining at the phenomenal densities found in the collapsing core
half slams into the innermost part of the star at a significant of the massive star, not even neutrinos pass with complete
fraction of the speed of light and “bounces,” sending a tre- freedom. A few tenths of a percent of the energy of the neu-
mendous shock wave back out through the star. trinos streaming out of the core of the dying star is trapped
by the dense material behind the expanding shock wave. phases of the interstellar medium and pushing around the
The energy of these trapped neutrinos drives the pressure clouds in the interstellar medium. Yet even this amount of
and temperature in this region ever higher, inflating a bub- energy is a pittance in comparison with the energy carried
ble of extremely hot gas and away from the supernova explosion by neutrinos—an amount
intense radiation around the The high-mass star of energy at least a hundred times larger still!
core of the star. The pressure explodes as a One of the most important astronomical events in the
of this bubble adds to the tremendous Type II last part of the 20th century was the explosion of a mas-
strength of the shock wave supernova. sive star in the small companion galaxy to the Milky Way
moving outward through the known as the Large Magellanic Cloud. Even at a distance
star. Within about a minute the shock wave has pushed its of 160,000 light-years, Supernova 1987A was so bright that
way out through the helium shell within the star. Within it dazzled sky gazers in the Southern Hemisphere (Figure
a few hours it reaches the surface of the star itself, heating 17.10). While astronomers working in all parts of the elec-
the stellar surface to 500,000 K and blasting material out- tromagnetic spectrum scrambled to point their telescopes
ward at velocities of up to about 30,000 km/s. Our evolved at the new supernova, solar astronomers had already qui-
massive star has exploded in an event referred to as a Type etly and unknowingly captured one of the true scientific
II supernova. prizes of SN 1987A. Solar neutrino telescopes recorded a
The death of a star as a Type I or Type II supernova should burst of neutrinos passing through Earth—neutrinos that
mean more to us than just another example of nature’s spec- originated not in the Sun but in the tremendous stellar
tacular fi reworks. As we will see in the following section, explosion that occurred beyond the bounds of our galaxy
supernovae leave behind a chemical legacy that was critical itself. The detection of neutrinos from SN 1987A provided
in the formation of next-generation stars like our Sun and its us with a rare and crucial glimpse of the very heart of a
Solar System. Were it not for supernovae, it is highly likely massive star at the moment of its death, confi rming a fun-
that either we or our planet would not exist. damental prediction of our theories about the collapse of
the core and its effects.
17.4 The Spectacle and The Energetic and Chemical

Legacy of Supernovae Legacy of Supernovae
Type II supernova explosions leave a rich and varied legacy
A Type II supernova is comparable in its spectacle to a Type to the universe. We have already mentioned their energetic
I supernova. For a brief time a Type II supernova can shine legacy: the energy from these explosions is responsible for
with the light of 100 billion suns, yet the energy that comes driving much of the dynamics of the interstellar medium.
out of the supernova in the form of light is but a tiny fraction When we look at the sky over a wide range of wavelengths,
of the energy of the explosion itself. About a hundred times we see huge expanding bubbles of million-kelvin gas, like
more energy comes out as the kinetic energy of the outer that in Figure 17.11a, aglow in X-rays and driving visible
parts of the star that are blasted into space by the explosion. shock waves (Figure 17.11b) into the surrounding interstellar
This ejected material contains about 1047 J of kinetic energy— medium. These bubbles are the still-powerful blast waves
enough energy to accelerate the mass of our Sun to a speed of of supernova explosions that took place thousands of years
10,000 km/s. The kinetic energy of the ejecta from both Type ago. In fact, in many cases supernova explosions are thought
I and Type II supernovae is responsible for heating the hottest to be responsible for compressing nearby clouds (like the
(a) (b)
FIGURE 17.10 SN 1987A
was a supernova that exploded
in a small companion galaxy of
the Milky Way called the Large
Magellanic Cloud (LMC). These
images show the LMC before
the explosion (a) and while the
supernova was near its peak
(b).
SN 1987A
(a) X-rays (b) Visible (c) HST visible
FIGURE 17.11 The Cygnus Loop is a supernova remnant—an expanding interstellar blast wave
caused by the explosion of a massive star. Gas in the interior with a temperature of millions of
kelvins glows in X-rays (a), while visible light comes from locations where the expanding blast wave
pushes through denser gas in the interstellar medium (b). (c) A Hubble Space Telescope image of a
location where the blast wave is hitting an interstellar cloud.
one in Figure 17.11c) enough to trigger their collapse toward normal circumstances, electric repulsion keeps positively
the formation of new generations of stars. charged atomic nuclei far apart. Extreme temperatures are
Perhaps even more important to us is the chemical legacy needed to slam nuclei together hard enough to overcome
left behind by supernova explosions. Later in our journey we this electric repulsion. Free neutrons, on the other hand,
will turn our attention to the earliest moments after the birth are not subject to these ground rules: they have no net elec-
of the universe; we will find that the only chemical elements tric charge, so there is no electric repulsion to prevent them
that formed at that time were the least massive ones: hydro- from simply running into an atomic nucleus, regardless of
gen, helium, and trace amounts of lithium, beryllium, and how many protons that nucleus contains. Under normal
boron. All of the rest of the chemical elements, including a circumstances in nature, free neutrons are very rare. In
large fraction of the atoms of which we are made, were formed the interiors of evolved stars, however, a number of nuclear
in the hearts of stars and then returned to the interstellar reactions produce free neutrons, and under some circum-
medium. This process, which is responsible for the progres- stances—including those shortly before and during a Type
sive chemical enrichment of II supernova—free neutrons can be produced in very large
the universe, is called nucleo- Supernovae eject numbers. Free neutrons are easily captured by atomic nuclei
synthesis. But although mass newly formed massive and later decay to become protons. In this way, elements
loss from both low- and high- elements into with higher and higher mass are formed.
mass stars enriches the inter- interstellar space. Calculations of nucleosynthesis in stars make clear pre-
stellar medium with massive dictions about which nuclei should be formed in abundance
elements formed in their interiors, Types I and II supernovae and which should not. These same patterns are found in mea-
are the true champions of nucleosynthesis. A look at Figure surements of the abundances of nuclei on Earth, in meteorite
17.5 should offer a clue as to why this is so. Whereas low- material, and in the atmospheres of stars. Although some of
mass stars are able to form elements only as massive as car- these patterns are subtle, we can appreciate others by com-
bon and oxygen, nuclear burning in high-mass stars produces paring what we have learned about nucleosynthesis with a
elements as massive as iron. Yet this is not the whole story. plot of the relative abundance of elements found in nature
A look at a table of elements that occur in nature (see Appen- (Figure 17.12). Notice, fi rst of all, that less massive elements
dix 3) shows that many elements, up through uranium (the are far more abundant than more massive elements—a con-
most massive among the naturally occurring elements), are sequence of the way more massive elements are progressively
far more massive than iron. If iron is the most massive ele- built up from less massive elements. An exception to this
ment that can be formed by nuclear burning, then where do pattern is the dip in the abundances of the light elements
these even more massive elements come from? lithium (Li), beryllium (Be), and boron (B). These light ele-
Answering this question takes us back to the reason that ments are easily destroyed by nuclear burning, but their pro-
high temperatures are necessary for fusion to occur. Under duction is mostly bypassed by the main reactions involved
FIGURE 17.12 The relative

1 H abundances of different elements
–1
on Earth are plotted against the
10 He C, N, and O element number of the nucleus.
are winners This pattern can be understood
10–2 in He burning.
as a result of the process of
Number of atoms relative to hydrogen
10–3 C O nucleosynthesis in stars. The

periodic table of elements in
10–4 N Fe, the most stable Appendix 3 identifies individual
Fe
element, is the end elements by their element
point of nuclear burning.
10–5 number (the number of protons).
10–6
10–7 More massive elements are

formed by neutron capture
B followed by beta decay.
10–8
Sawtooth pattern
10–9 Li
Li, Be, and B are
10–10 destroyed, not
formed, in stars.
10–11
Be U
10–12
10 20 30 40 50 60 70 80 90 100
Element number
in burning hydrogen (H) and helium (He). Conversely, car- that was left behind? Picking up our story where we left
bon (C), nitrogen (N), and oxygen (O) are big winners in the off, the matter at the center of the massive star has col-
CNO cycle of hydrogen burning and the triple-alpha process lapsed to the point where it has about the same density as
of helium burning, and their high abundances reflect this the nucleus of an atom. As
fact. The spike in the abundances of the “iron peak” ele- long as the mass of the core The Type II supernova
ments is evidence of the nucleosynthetic processes, includ- left behind by the explosion leaves behind a
ing those in Type I and Type II supernovae, that favor these is no more than about 3 M⊙, neutron-degenerate
most tightly bound of all nuclei. Even the sawtooth pattern this collapse will be halted core.
in the abundances of even- and odd-numbered elements can by quantum mechanical rules
be understood as a consequence of stellar nucleosynthesis. similar to those responsible for holding up a white dwarf.
(To complete our discussion of nucleosynthesis, we will But in this case, instead of electrons it is neutrons that are
need to consider the formation of elements that took place forced together as tightly as the rules of quantum mechan-
not in stars, but during the formation of the universe itself. ics allow. The neutron-degenerate core left behind by the
We will take up this discussion in Chapter 18.) explosion of a Type II supernova is referred to as a neutron
Once again we find connections and insights in places star. It has a radius of perhaps 10 kilometers (km), making
that, prior to our study of the universe, we might never have it roughly the size of a small city; but into that volume is
imagined looking. Our understanding of the processes at packed a mass from 1.4 to about 3 times that of our Sun.
work within the interiors of dying stars is being confirmed by At a density of about a billion metric tons per cubic cen-
analysis of the chemical composition of the very Earth under timeter, the neutron star is a billion times more dense than
our feet. At the same time, we can make a pretty good stab at a white dwarf and a thousand trillion (1015) times more
identifying the kinds of stars responsible for forming the atoms dense than water! That density is roughly what we would
that make up our own bodies. Our growing understanding of get by crushing the entire Earth down to an object the size
the chemical evolution of the universe and our connection to of a football stadium.
it represents one of the triumphs of modern astronomy. As if neutron stars were not extraordinary enough in
their own right, they also form the hearts of a number of
other exotic objects. If the
massive star responsible for X-ray binaries arise
Neutron Stars and Pulsars the formation of a neutron star from the accretion
We have now seen the fate of the outer parts of the star, is part of a binary system, of mass onto
which are blasted back into interstellar space by the explo- then the neutron star will be neutron stars.
sion of a Type II supernova. But what remains of the core left with a binary companion.
FIGURE 17.13 X-ray binaries

are systems consisting of a
normal evolving star and a white
dwarf, neutron star, or black hole.
As the evolving star overflows its
Roche lobe, mass falls toward the
Accretion
disk collapsed object. The gravitational
well of the collapsed object is so
deep that when the material hits
Neutron star the accretion disk, it is heated to
Evolving star such high temperatures that it
radiates away most of its energy
as X- rays.
X-rays
1 An evolving star overflows its
Roche lobe, pouring matter onto
its neutron star companion.
2 Infalling matter heats the 3 …and feeds relativistic

accretion disk to X-ray– jets from the rotating
emitting temperatures… neutron star.
Such a system might remind you of the white dwarf binary poles of the system. Conditions there produce intense elec-
systems responsible for novae and Type I supernovae dis- tromagnetic radiation, which is beamed away from the
cussed in Chapter 16. As the second star in such a binary magnetic poles of the neutron star as shown in Figure 17.14.
system evolves and overfills its Roche lobe, matter plum- As the neutron star rotates, these beams of radiation sweep
mets down the deep gravitational well of the tiny but mas- through space much like the rotating beams of a lighthouse.
sive neutron star. This matter slams into the accretion disk When we are located in the paths of these beams, we detect
around the neutron star with enough energy to heat the disk a pattern resembling what the beams from a lighthouse look
to temperatures of millions of kelvins, and the accretion like to sailors entering a harbor at night. The neutron star
disk glows brightly in X-rays. Such an object, illustrated in appears to flash on and off with a regular period equal to
Figure 17.13, is known as an X-ray binary. Many fascinat- the period of rotation of the star (or half the rotation period,
ing phenomena occur in X-ray binaries, including the for- if we see both beams).
mation of powerful jets that blast away from the neutron Rapidly pulsing objects Pulsars are rapidly
star in directions perpendicular to its accretion disk and at were first discovered by spinning, magnetized
speeds approaching the speed of light. observers working with radio neutron stars.
Besides phenomenal density, a neutron star has a number wavelengths in 1967. These
of other extraordinary properties. The same principle of con- objects, which blinked like regularly ticking clocks, puz-
servation of angular momentum that requires a collapsing zled astronomers. One of the early tongue-in-cheek names
molecular cloud to spin faster as it grows smaller also says given to these objects was “LGMs,” which stood for “little
that as the core of a massive star collapses, it must spin faster green men.” Today these objects are referred to by the less
as well. As a main-sequence O star, a massive star rotates per- flamboyant but more accurately descriptive term pulsar. As
haps once every few days. As a neutron star, it might instead of this writing, well over a thousand pulsars are known,
rotate tens or even hundreds of times each second! We also and more are being discovered all the time.
saw in our discussion of a collapsing interstellar cloud that
the magnetic field in the cloud is carried along and concen-
trated by the collapse. This phenomenon also occurs in the The Crab Nebula—Remains
collapse of a star, amplifying its magnetic field to values that
are trillions of times greater than the magnetic field at Earth’s
of a Stellar Cataclysm
surface. A neutron star has a magnetosphere, just as Earth In A.D. 1054, Chinese astronomers recorded the presence
and several other planets do, except that the neutron star’s of a “guest star” in the part of the sky that we call the con-
magnetosphere is unimaginably stronger and is whipped stellation Taurus. The new star was so bright that it could
around many times a second by the spinning star. be seen during the daytime for 3 weeks, and it did not fade
Energetic subatomic particles such as electrons and from visibility altogether for many months. On the basis
positrons move along the magnetic-field lines of the neutron of the Chinese description of the changing brightness and
star and are “funneled” by the field toward the magnetic color of the object, we can say today that the guest star of
1 Neutron stars have enormously 2 Electrons and positrons moving in the neutron star’s
strong magnetic fields. magnetic field produce radiation that is beamed away
from the poles of the neutron star.
3 As the neutron star

rotates, these beams
sweep around like the
beam of a lighthouse.
Neutron star
Magnetic
field
Lighthouse beam
4 As the beams sweep past an

observer, the neutron star appears
to pulse on and off, earning it
the name pulsar.
Rotation period
Brightness
Primary pulse Primary pulse

Secondary pulse Secondary pulse
Time
VISUAL ANALOGY FIGURE 17.14 As a highly magnetized neutron star rotates rapidly, light is given off,
much like the beams from a rotating lighthouse lamp. From our perspective, as these beams sweep
past us the star will appear to pulse on and off, earning it the name pulsar.
HST image
(a) Groundbased image (b) Visible-light image

FIGURE 17.15 The Crab
Nebula is the remnant of a
supernova explosion witnessed
by Chinese astronomers in
A.D. 1054. (a) The object we
see today is an expanding
cloud of “shrapnel” from that
earlier cataclysm. The spinning
pulsar at the heart of the Crab
Nebula sends off a “wind” of
(c) X-ray image electrons and positrons moving
at close to the speed of light.
Pulsar The synchrotron radiation from
these particles is shown in
Pulsar
visible light (b) and in X-rays (c).
Filaments of gas Synchrotron

ejected by the emission from
supernova explosion pulsar wind
1054 was a fairly typical Type II supernova. When we look shot, the rotating pulsar magnetosphere fl ings elementary
at this spot in the sky today, we see an expanding cloud of particles—probably mostly electrons and positrons—away
debris from this explosion—an extraordinary object called from the neutron star in a powerful wind moving at nearly
the Crab Nebula (Figure 17.15). the speed of light. Material from this wind fills the space
The Crab Nebula consists of several components. Images between the pulsar and the
of the Crab Nebula taken in the light of nebular emission expanding shell. The Crab A pulsar powers
lines show filaments of glowing gas. Doppler shift measure- Nebula is almost like a big the Crab’s eerie
ments of these filaments reveal a pattern much like that balloon; but instead of being synchrotron glow.
seen in planetary nebulae—the hallmark of an expand- filled with hot air, it is fi lled
ing shell. But whereas planetary nebulae are expanding at with a mix of relativistic particles and strong magnetic
20–30 km/s, the shell of the Crab is expanding at closer to fields—an environment more like what we might fi nd in a
1,500 km/s. Studies of the spectra of these filaments show physicist’s particle accelerator than what we normally
that they contain anomalously high abundances of helium think of as interstellar space. Images of the Crab Nebula
and other more massive chemical elements—the products (see Figures 17.15b and c) show this bizarre bubble as an
of the nucleosynthesis that took place in the supernova and eerie glow. This glow is synchrotron radiation from the
its progenitor star. relativistic electrons and positrons as they spiral around
There is a pulsar at the center of the Crab Nebula. This the magnetic field in the Crab.
was the fi rst pulsar to be seen at visible wavelengths as it
flashed on and off 30 times a second. Actually, the Crab
pulsar flashes 60 times a second: fi rst with a main pulse
associated with one of the “lighthouse” beams, then with
a fainter secondary pulse associated with the other beam.
17.5 Star Clusters
Today, the Crab Nebula has been observed in all parts of the Are Snapshots of
electromagnetic spectrum, from low-energy radio waves to
high-energy X-rays and even higher-energy gamma rays (see Stellar Evolution
the chapter opening photograph).
As the Crab pulsar spins 30 times a second, it whips its Over the course of this and the previous chapters we have
powerful magnetosphere around with it. At a distance from told a remarkable tale about the evolution of stars of differ-
the pulsar about equal to the radius of the Moon, material ent masses, presenting it as a well-corroborated theory. In
in the magnetosphere must move at almost the speed of other words, we have presented the story of stellar evolu-
light to keep up with this rotation. Like a tremendous sling- tion as fact. Yet even the most massive stars take hundreds
17.5 Star Clusters Are Snapshots of Stellar Evolution 485
(a) (b) (c)

ZERO-AGE MAIN SEQUENCE 4 MILLION YEARS 10 MILLION YEARS
107
106 1,0 1,0 1,0

00 00 00
105
60 M £ R R R
£ £ £
10 4 20 M £ 20 M
£
1 Massive stars in a 2 As the cluster ages, less
and less massive stars
Luminosity (L£)
cluster evolve off of the

103 10 M £ 10 10 M £ 10 M£
leave the main sequence.
R main sequence first.
£
5M 5M 5M
102 £ £ £
3M £ 3M
£ 3M
£
101 2M 2M 2M
0.1 £
0.1 £
0.1 £
1 R 1.5 M £ R 1.5 M £ R 1.5 M £
£ £ £ £ £ £
1M 1M 1M
10–1
£ £ £
0.8 M £ 0.8 M £ 0.8 M £
0.6 M 0.6 M 0.6 M

10–2
£ £ £
0.2 M £ 0.2 M £ 0.2 M £
10–3
60,000 30,000 10,000 5,000 2,000 60,000 30,000 10,000 5,000 2,000 60,000 30,000 10,000 5,000 2,000
Surface temperature (K) Surface temperature (K) Surface temperature (K)
(d) (e) (f)

100 MILLION YEARS 1 BILLION YEARS 10 BILLION YEARS
107
106 1,0 1,0 1,0

00 00 00
105 3 The cluster age is R 4 Phases of stellar R R
£ £ £
determined from the evolution are seen in
104 main-sequence turnoff. cluster H-R diagrams.
Luminosity (L£)
103 10 10 10
R R R
£ £ £
102
3M £
101 2M
0.1 £
0.1 0.1
1 R 1.5 M £ R 1.5 M £ R
£ £ £ £ £ £
1M 1M 1M
10–1
£ £ £
0.8 M £ 0.8 M £ 0.8 M £
0.6 M 0.6 M 0.6 M

10–2
£ £ £
0.2 M £ 0.2 M £ 0.2 M £
10–3
60,000 30,000 10,000 5,000 2,000 60,000 30,000 10,000 5,000 2,000 60,000 30,000 10,000 5,000 2,000
Surface temperature (K) Surface temperature (K) Surface temperature (K)
FIGURE 17.16 H-R diagrams of star clusters are snapshots of stellar evolution. These are H-R
diagrams of a simulated cluster of 40,000 stars of solar composition seen at different times
following the birth of the cluster. Note the progression of the main-sequence turnoff to lower and
lower masses.
of thousands of years to evolve, which is far longer than the tion. A look at a cluster that is 10 million years old shows us
handful of decades we have spent studying their ways. Upon what stars of all different masses evolve into during the fi rst
what testable predictions of our theories of stellar evolution 10 million years after they are formed. A look at a cluster 10
do we base such bold claims of knowledge? billion years after it formed shows us what becomes of stars
As we learned in Chapter 15, when an interstellar cloud of different masses after 10 billion years pass.
collapses, it fragments into pieces, forming not one star but This basic result—the fact that high-mass stars in a clus-
many stars of different masses. ter evolve more rapidly than low-mass stars that formed
We see many such star clus- Stars in clusters at the same time—provides the key to our knowledge of
ters around us today, contain- formed together at stellar evolution. Figure 17.16 shows the H-R diagram of
ing anywhere from a few dozen about the same time. a simulated cluster of 40,000 stars as it would appear at
to millions of stars. The fact several different ages. In Figure 17.16a stars of all masses
that all of the stars in a cluster formed together at nearly the are located on the zero-age main sequence, showing where
same time means that clusters are snapshots of stellar evolu- they begin their lives as main-sequence stars. The increas-
107 FIGURE 17.17 The observed H-R diagram of

1 The remarkable agreement stars (dots) in the cluster 47 Tucanae agrees
between observations of remarkably well with the theoretical calculation
106 stars in a cluster (dots)… –10 (solid line) of the H-R diagram of a 12-billion-
year-old cluster.
h
105
nc
ra
tb
i an
Horizontal
104
dg
branch –5

Re
103
More luminous
Luminosity (L£)
102 0
2 …and calculations of where

101 evolved stars should lie in
the H-R diagram (red line)…
1 +5
MA
3 …provides strong support

IN
for our understanding of

SE
10–1 Q
UE
stellar structure and evolution.
NC
E
+10
10–2
10–3
10,000 8,000 6,000 5,000 4,000
ing masses of stars along the main sequence are indicated. its original location along the main sequence. The location
We would never expect to see a cluster H-R diagram that of the most massive star that is still on the main sequence
looks like the one in Figure 17.16a, however, for the sim- is called the main-sequence turnoff. As the cluster ages, the
ple reason that the stars in a cluster do not all reach the main-sequence turnoff moves farther and farther down the
main sequence at exactly the same time. Star formation main sequence to stars of lower and lower mass.
in a molecular cloud is spread out over several million As a cluster ages (Figures 17.16d and e), we see more
years, and it takes considerable time for lower-mass stars than the movement of the main-sequence turnoff to lower
to contract to reach the main sequence. The H-R diagram and lower masses; we see the details of all stages of stellar
of a very young cluster normally shows many lower-mass evolution. By the time the star cluster is 10 billion years old
stars located well above the main sequence, still descend- (Figure 17.16f), stars with masses of only 1 M⊙ are beginning
ing their Hayashi tracks. to pull away from the main sequence. Stars slightly more
The more massive a star is, the shorter its life on the massive than this are seen as giant stars of various types. In
main sequence will be. After only 4 million years (Fig- this cluster the horizontal branch appears as a knot of stars
ure 17.16b), all stars with masses greater than about 20 M⊙ sometimes referred to as the “red clump.” Note how few
have evolved off the main sequence and are now spread out giant stars are present in any of the cluster H-R diagrams.
across the top of the H-R diagram. The most massive stars The giant, horizontal, and asymptotic giant branch phases
have already disappeared from the H-R diagram entirely, in the evolution of low- and intermediate-mass stars pass so
having vanished in supernovae. quickly in comparison with a star’s main-sequence lifetime
As time goes on, stars of lower and lower mass evolve that even though the cluster started with 40,000 stars, only a
off the main sequence, and the turnoff point moves toward handful of stars are seen in these phases of evolution at any
the bottom right in the H-R given time. Similarly, it takes a newly formed white dwarf
diagram. By the time the clus- The main-sequence only a few tens of millions of years to cool to the point that
ter is 10 million years old (Fig- turnoff shifts to lower- it disappears off the bottom of these figures. Even though the
ure 17.16c), only stars with mass stars as a majority of evolved stars in an old cluster are white dwarfs,
masses less than about 15 M⊙ cluster grows older. all but a few of these stars will have cooled and faded into
remain on the main sequence. obscurity at any given time.
The cluster H-R diagram looks as if we grabbed the top of The cluster H-R diagrams in Figure 17.16 are theoretical
the band of cluster stars and gradually peeled it away from calculations of what clusters of different ages should look
105 dances of massive elements in the atmospheres of cluster

NGC 2362 h + ␹ Persei stars, as well as the cluster age.)
We can apply our understanding of stellar evolution even
These clusters are when groups of stars are so far away that we cannot see each
104 very young. h + ␹ Persei
star individually. Although many fewer high-mass stars
form in a cluster than low-mass stars, higher-mass stars are
107 years
M3
far more luminous than their lower-mass siblings. Likewise,
3 Pleiades
10 giant, evolved low-mass stars are far more luminous than
M41
less massive stars that remain on the main sequence. As a
Luminosity (L£)
M41 M11 M11

result, the most massive, most luminous stars that are pres-
102 108 ent will dominate the light from a star cluster. If the cluster
Coma H+P M67
years Hyades
Praesepe
is young, then most of the light we see will usually come
from luminous hot, blue stars. If the cluster is old, then the
NGC 752 752
101 109 light from the cluster will have the color of red giants and
years relatively cool, low-mass stars.
These are old. As always, there are caveats to such general statements.
Young clusters may contain very luminous red supergiants,
100 £
1010 and stars with lower abun-
years
dances of massive elements in Colors reveal the ages
their atmospheres often look of stellar populations.
10–1 significantly bluer than their
O5 B5 A5 F5 G5 K5 M5 more chemically enriched counterparts. Even so, we can usu-
Spectral type ally get some idea about the properties of a group of stars by
looking at its overall color. We will put this knowledge to good
FIGURE 17.18 H-R diagrams for clusters having a range of
use in the chapters that follow, as we turn our attention to the
different ages. The ages associated with the different main-
sequence turnoffs are indicated. large collections of stars called galaxies. A group of stars with
similar ages and other characteristics is referred to as a stellar
population. An especially bluish color to a galaxy or a part of
like. The crucial point is that this is what H-R diagrams of a galaxy often signifies that the galaxy contains a young stel-
real clusters must look like if our theories of stellar evolution lar population that still includes hot, luminous, blue stars that
are correct. Fortunately, this is also what H-R diagrams of must have formed recently. In contrast, if a galaxy or part of a
real clusters do look like. Figure 17.17 shows the observed galaxy has a reddish color, we expect that it is composed pri-
H-R diagram for the cluster 47 Tucanae, along with a theo- marily of an old stellar population.
retical calculation of the H-R diagram for a 12-billion-year-
old cluster. The quality of the agreement speaks for itself.
The fact that observed star clusters have H-R diagrams that
agree so well with the predictions of models is strong sup- 17.6 Beyond Newtonian
port for our theories of stellar evolution.
Armed with confidence in the reliability of our ideas Physics
about stellar evolution, we can turn cluster evolution mod-
els into a powerful tool for studying the history of star for- Having discussed the evolution of clusters of stars of all
mation. When we observe a star cluster, the location of the masses, we now return to the stellar remnants left behind by
main-sequence turnoff immediately tells us the age of the the evolution of the most massive stars. You might imagine
cluster. If shown any of the cluster H-R diagrams in Figure that, surely, the neutron star represents the final extreme of
17.16, we should have no difficulty estimating the age of stellar evolution; but there is one more step a star can take.
the cluster. Figure 17.18 traces the observed H-R diagrams The physics of a neutron star is much like the physics of a
for several real star clusters. Once we know what to look white dwarf, except that it is neutrons rather than electrons
for, the difference between young and old clusters is obvi- that cause it to be degenerate. A white dwarf can have a mass
ous. NGC 2362 is clearly a young cluster. Its complement of no more than about 1.4 M⊙. This is the Chandrasekhar limit,
of massive, young stars shows it to be only a few million discussed in Chapter 16. If the mass of the object exceeds this
years old. In contrast, 47 Tucanae (see Figure 17.17) has a limit, then gravity will be able to overcome electron degen-
main-sequence turnoff of about 0.85 M⊙, signifying a clus- eracy, and the white dwarf will begin collapsing again.
ter age of about 12 billion years. (Note that the evolution Just as there is a Chandrasekhar limit for white dwarfs,
of stars depends on their chemical composition as well as there is a Chandrasekhar limit for the mass of neutron stars
their masses. Detailed comparisons between models and as well. If the mass of a neutron star exceeds about 3 M⊙,
observed cluster H-R diagrams must account for the abun- then gravity will begin to win out over matter once again.
The neutron star grows smaller, and gravity becomes stron- 100 mph that you gave it with your throw.) In the frame of
ger and stronger at an ever-accelerating pace. However, this reference of the oncoming green car (bottom panel of Figure
time there is no force in nature 17.19a) traveling at 50 mph, the ball is moving at 200 mph.
powerful enough to prevent If a neutron star’s (This is the 150 mph that the ball is moving relative to the
gravity’s final victory. The col- mass exceeds 3 M⊙, ground plus the 50-mph motion of the oncoming car.) In our
lapsing object quickly crosses it will collapse to a everyday experience, velocities simply add. This is also how
a threshold where the escape black hole. Newton’s laws say the universe should behave.
velocity from its surface Now do exactly the same thought experiment again,
exceeds the speed of light, and not even light can escape its but with two changes. Instead of cars traveling at 50 mph,
gravity. From this point on, nothing can escape from the imagine two spacecraft approaching each other, both trav-
collapsing object and fi nd its way back into the universe of eling at half the speed of light, or 0.5c, as shown in Figure
which it was once a part. The object is now a black hole. A 17.19b. You are in the yellow spaceship traveling at 0.5c
black hole will form if the stellar core left behind by a Type and, instead of throwing a baseball, you shine a beam of
II supernova exceeds about 3 M⊙. Alternatively, a neutron light forward. If we replace the 100-mph speed of the ball
star will collapse to become a black hole if it accretes enough in the previous paragraph with c, we think we know what
matter from a binary companion to push it over the 3-M⊙ to expect for all observers. In your frame of reference (top
limit. Regardless of how it formed, any collapsed object panel of Figure 17.19b) you are stationary, the oncoming
with a mass greater than 3 M⊙ must be a black hole. (blue) spacecraft would be approaching at 1.0c, an observer
At this point we must pause. What sort of object is this? on a nearby planet is moving toward you at 0.5c, and the
It turns out that black holes are so strange, so far from our beam of light is moving away from you at the speed of
common understanding of reality, that the laws of Newto- light, c. In the frame of reference of an observer on a nearby
nian physics (which we explored in Chapter 3) are inad- planet (middle panel of Figure 17.19b), the light should
equate to describe them. To do so, we must fi rst take a step travel by at a speed equal to the speed of your spacecraft
back and—as Einstein did—question our intuitive assump- plus the speed of light, or 1.5c. Similarly, in the reference
tions about the very nature of space and time. frame of the oncoming blue spacecraft (bottom panel of
Figure 17.19b) traveling at 0.5c, the light should appear to
travel at a speed of 2.0c. (This is the 1.5c that the light is
The Speed of Light Is a moving relative to the nearby planet plus the 0.5c motion
of the oncoming spacecraft.) But something is wrong here.
Very Special Value The beam of light appears to be traveling at 1.5c and 2.0c
Even though we have seen that light behaves like a wave, to, respectively, observers on the nearby planet and the
it turns out that there is a serious flaw in our Newtonian approaching blue spacecraft, and the two spacecraft appear
description of that wave. To understand this flaw, we need to be approaching one another at the speed of light. We have
to think back to Chapter 2, where we used a moving car as already learned in Chapter 4 that the speed of light has a
an example of a moving frame of reference. Imagine that constant value, c; and, as we will see later in this chapter,
you are sitting in a moving car and there is a ball on the seat objects cannot move at a speed equal to that of light. In the
beside you. In your frame of reference the ball is at rest. But relativistic world, it turns out, speeds do not simply add.
if the car is moving at 50 miles per hour (mph) down the The relative speed between the two spacecraft in the top
highway, someone standing by the road will say that the and bottom panels of Figure 17.19b (0.5c + 0.5c) adds to
ball is moving at 50 mph. To someone in oncoming traffic 0.8c, not 1.0c! At relativistic speeds, our everyday experi-
moving at 50 mph, the relative speed of both your car and ence no longer works for us.
the ball would be 100 mph. There really is no difference Although we would like to compare the speed of light
among these three perspectives. The laws of physics are the measured by observers in the spacecraft and by a stationary
same in any inertial frame of reference. observer, we do not have the
As a variant of the “ball in the car” experiment, imagine luxury of performing this Surprisingly,
two cars approaching one another, each traveling at 50 mph, experiment while traveling experiments showed
as in Figure 17.19a. As your car (the red car) moves down through space at half the speed that the speed of light
the highway at 50 mph, you pitch a fastball forward at 100 of light. However, physicists is the same for all
mph. In your frame of reference (top panel of Figure 17.19a), are ingenious folk. During the observers.
you are stationary, the oncoming (green) car is approaching closing years of the 19th cen-
you at 100 mph, an observer standing still by the side of the tury and the early years of the 20th century, physicists were
road is moving toward you at 50 mph, and the ball is moving conducting laboratory experiments that were the functional
at 100 mph. Now consider frame of reference of the observer equivalent of our thought experiment. What they found puz-
(middle panel of Figure 17.19a). Both cars are approaching zled them greatly. Rather than the speed of the beam of light
at 50 mph, and the ball is moving at 150 mph. (The ball has differing from one observer to the next, as expected on the
the original 50-mph speed of your car plus the additional basis of Newton’s physics and “common sense,” they found
instead that all observers measure exactly the same value for beam from your light is traveling at exactly c in her own
the speed of the beam of light, regardless of their motion! frame of reference. In fact, it turns out that every observer
As you ride in your spaceship, you measure the speed of always finds that light in a vacuum travels at exactly the
the beam of light to be c, or 3 × 108 meters per second (m/s). same speed, c, regardless of his or her own motion or the
That is as expected because you are holding the source of motion of the light source.
the light. But the observer on the planet also measures the If at this point you are feeling uneasy and saying to
speed of the passing beam of light to be 3 × 108 m/s. Even yourself, “This is very bizarre,” then you probably have fol-
the passenger in the oncoming spacecraft fi nds that the lowed what the last few paragraphs state. And if this discus-
FIGURE 17.19 The rules of motion that apply in our daily lives break down when speeds approach
the speed of light. The fact that light itself always travels at the same speed for any observer is the
basis of special relativity. (Note that relativity also affects the relative speeds of the two spacecraft.)
(a) In everyday experience velocities simply add… (b) …but as v nears c, things are different.
4 A moving spaceship fires a
1 A ball thrown at 100 mph laser. In the reference frame of
relative to a car moving the spaceship, the light travels
at 50 mph… at the speed of light, c.
v light = c
vball = 100 mph

0.8c
100 mph Stationary
Stationary
Reference frame 0.5c

Reference frame of the yellow
of the red car 50 mph spaceship
5 By analogy with the ball in the panel

2 …moves at 150 mph relative at left, we might expect that in a
to an observer by the side planetbound observer’s reference frame
of the road… the light’s velocity would be 1.5c…
v light = c
vball = 150 mph
0.5c 0.5c
50 mph 50 mph
Stationary 6 …but actually it is still c!

Reference frame of
Reference frame planetbound
of observer Stationary observer
7 Observers in any reference

3 …and at 200 mph in the frame always measure the
reference frame of an speed of light in a vacuum to
oncoming car. be c, regardless of their motion!
v light = c
vball = 200 mph

0.8c
100 mph
Stationary Stationary
0.5 c
Reference frame
Reference frame of the blue
of the green car 50 mph
spaceship
sion bothers you, imagine the reaction of those physicists! events were separated from each other in space by a mile.
Newton’s laws of motion had been the bedrock of science In this everyday, “Newtonian” view, the distance between
for 200 years, standing up to every experimental challenge two events depends on the motion of the observer, but the
that came their way. Now, suddenly that bedrock seemed time between the two events does not.
to turn to sand. How could light have the same speed for Einstein questioned why there was such a distinction
all observers, regardless of their own velocity? Preposter- between the way Newton treated space and the way Newton
ous! And yet that was the inescapable experimental result. treated time. Einstein realized that the only way the speed
Despite all its spectacular successes, Newtonian physics of light can be the same for all observers is if the passage
seemed to be in serious trouble. of time is different from one observer to the next! This is a
Enter a young German-born Swiss patent clerk named very counterintuitive idea, but it is so central to our modern
Albert Einstein. As a 16-year-old schoolboy Einstein had understanding of the universe that it is worth wrestling with
already realized that there was trouble afoot. Light travels a bit. Hang on to your hat while we reconstruct some of the
in a straight line at a constant speed. Einstein reasoned that reasoning that led Einstein to this remarkable conclusion.
according to Newton’s laws of motion, there should be a per- To measure time, the fi rst thing we need is a clock. The
fectly good inertial frame of reference that moves along with best way to build a clock is to base it on a value that every-
the light and in which the light is stationary. That is, you one can agree on—such as the speed of light. Figure 17.20a
should be able to “keep up” with light so that you are moving shows just such a clock as seen by observer 1, who is station-
right along with it. But if you could do that, the light would ary with respect to the clock. At time t1 a flashlamp gives
be an oscillating electric and magnetic wave that does not off a pulse of light. Call this “event 1.” The light bounces
move. This was impossible according to Maxwell’s equations off a mirror a distance l meters away, and then heads back
for electromagnetic waves. There was a contradiction here. toward its source. At time t2 the light arrives and is recorded
Either Maxwell was wrong in his understanding of electric- by a photodetector. Call this “event 2.” The time between
ity and magnetism, or Newtonian physics did not apply at events 1 and 2 is just the distance the light travels (2l meters),
very large velocities. As the experimental results rolled in divided by the speed of light: t2 - t1 = 2l/c.
on measurements of the speed of light, it became clear that So far so good, but now look at the clock from the per-
it was Newtonian physics that needed revision. spective of observer 2 in a frame of reference that is mov-
ing relative to the clock (Figure 17.20b). In this observer’s
frame of reference, he is stationary and it is the clock that is
moving at speed v. (Recall that because any inertial frame
Time Is a Relative Thing of reference is as good as any other, this observer’s perspec-
Einstein resolved the contradiction between Maxwell and tive is as valid as the fi rst observer’s perspective.) We see the
Newton and ushered in a scientific revolution with his spe- same two events as before: event 1 when the light leaves the
cial theory of relativity, which was published in 1905. Spe- flashlamp and event 2 when the light arrives at the detec-
cial relativity was Einstein’s answer to the question “What tor. There is a difference, however. In this frame of refer-
must the universe be like if every observer always measures ence the clock moves between the two events, so the light
the same value for the speed of light in a vacuum?” Einstein has farther to go. (If you do not see this right away, use a
focused his thinking on pairs of events. In relativity, an ruler to measure the total length of the light path in Figure
event is something that happens at a particular location in 17.20b and compare it with the total length of the light path
space at a particular time. Snapping your fingers is an event, in Figure 17.20a.) The time between the two events is still
because this action has both a time and a place. From every- the distance traveled divided by the speed of light, but now
day experience we know that the distance between any two that distance is longer than 2l meters. Because the speed
events depends on the frame of reference of the person of light is the same for all observers, the time between the
observing them. Suppose you are sitting in a car that is two events must be longer as well!
traveling down the highway in a straight line at a constant Let’s go over that again. The two events are the same
60 mph. You snap your fingers two events, regardless of the frame of reference from which
(event 1), and a minute later Special relativity they are observed. The question is, how much time passed
you snap your fi ngers again concerns the between the two events? Because the speed of light is the
(event 2). In your frame of ref- relationship between same for all observers, there must be more time between the
erence you are stationary and events in space two events when they are viewed from a frame of reference
the two events happened at and time. in which the clock is moving. It takes a moving clock more
exactly the same place. They time than a stationary clock to complete one “tick.” Moving
are separated by a minute in time, but there is no separation clocks must run slow, and the passage of time must depend
between the two events in space. This is very different from on an observer’s frame of reference.
what happens in the frame of reference of an observer sit- To Newton, and to us in our everyday lives, the march of
ting by the road. This observer agrees that the second snap time seems immutable and constant. But in reality the only
of your fi ngers (event 2) occurred a minute after the fi rst thing that is truly constant is the speed of light, and even
snap of your fingers (event 1), but to this observer the two time itself flows differently for different observers.
(a) Reference frame in which clock is stationary (b) Reference frame in which clock is moving
A light clock measures the time it 4 …so the light must
takes for light to travel a fixed path. go farther than 2l.
Mirror Mirror
1 In observer 1’s
c c c c 5 But because the speed
reference frame the
of light is the same for
clock is stationary.
all observers…
Observer 1 l
Event 1: Light Event 2: Light
leaves bulb. l reaches detector.
Event 1 Event 2
2 The light must travel
distance 2l at speed c,
v
so the travel time = 2l/c.
Flash lamp Detector
v Observer 2 is 3 But in observer 2’s Observer 2 6 …the time between the two
moving to left. reference frame the is stationary. events is greater than 2l/c.
Observer 2 clock is moving…
CLOCK IS STATIONARY. CLOCK IS MOVING.
FIGURE 17. 20 The “tick” of a light clock as seen in two different reference frames: stationary (a)
and moving (b). As Einstein’s thought experiment demonstrates, if the speed of light is the same for
every observer, then moving clocks must run slow.
Here, in a nutshell, is the heart of Einstein’s special differently from what Newtonian physics would lead us to
theory of relativity. In our everyday Newtonian view of the expect, we call this a relativistic effect.
world, we live in a three-dimensional space through which
time marches steadily onward. Events occur in space at a
certain time. By the time Einstein fi nished working out the The Implications of Relativity
implications of his insight, he
had reshaped this three- Space and time
Are Far-Ranging
dimensional universe into a together form a The story of special relativity is another case study of how
four-dimensional spacetime. four-dimensional science works. Newton’s laws had proven for a long time to
Events occur at specific loca- spacetime. be an extraordinarily powerful way of viewing the world.
tions within this four-dimen- But as science turned its attention to a different phenome-
sional spacetime, but how this spacetime translates into non—light—difficulties arose.
what we perceive as “space” and what we perceive as “time” Newton’s theory of motion, Conflicts between
depends on our frame of reference. Maxwell’s theory of electro- theory and
It is very important to state that Einstein did not throw magnetic radiation, and observation point
out Newtonian physics. We were not wasting our time in empirical measurements of the way to new
Chapter 3 when we studied Newton’s laws of motion. Instead, the speed of light met head- knowledge.
Einstein found that Newto- on. Such confl icts are what
nian physics is contained Newtonian physics is scientists live for; they point the way to new knowledge and
within special relativity. In contained within new understanding. Einstein was able to step in and recon-
our everyday experience we special relativity. cile this conflict, and in the process he changed the way we
never encounter speeds that think about the universe. Einstein’s ideas remained contro-
approach that of light. Even the breakneck speed of the versial well into the 20th century. But as one experiment
space shuttle is only about 0.000025c. When Einstein’s spe- after another confi rmed the strange and counterintuitive
cial relativity is restricted to cases in which velocities are predictions of relativity, scientists came to accept its valid-
much less than the speed of light, then Einstein’s equations ity. Today, special relativity is an integral and indispensable
become the very equations that describe Newtonian phys- part of all of physics, shaping our thinking about the motions
ics! In our everyday lives we experience a Newtonian world. of both the tiniest subatomic particles and the most distant
Only when relative velocities approach that of light do galaxies. Puzzling out relativity is time well spent, but here
things begin to depart from the predictions of Newtonian we will explore only a few of the essential insights that
physics. When great velocities cause something to turn out come from Einstein’s work:
1. What we think of as “mass” and what we think of as Velocity using Newton’s theory
“energy” are actually two manifestations of the same 1.0
thing. Usually we think of the energy of an object as
depending on its speed. The faster it moves, the more
0.8
energy it has. But Einstein’s famous equation E = mc2 Velocity of rocket ship
says that even a stationary object has an intrinsic “rest” experiencing acceleration g
energy that equals the mass (m) of the object multiplied using special relativity
v /c 0.6
by the speed of light (c) squared. The speed of light is a
very large number. This relationship between mass and
energy says that a single tablespoon of water has a rest 0.4
energy equal to the energy released in the explosion of
over 300,000 tons of TNT! All reactions that produce 0.2
energy do so by converting some of the mass of the
reactants into other forms of energy. But even the most
efficient chemical or nuclear reactions release only tiny 0
fractions of the total energy available. Exploding TNT, 0 1 2 3 4 5
for example, converts less than a trillionth of its mass Time (years)
into energy. Even the explosion of a hydrogen bomb FIGURE 17. 2 1 The speed of a rocket ship experiencing an
releases far less than 1 percent of the energy contained acceleration equal to Earth’s acceleration of gravity. The rocket
in the mass of the bomb. ship approaches the speed of light but never gets there.
The equivalence between mass and energy points
both ways. In Chapter 3 we defi ned mass as the property
of matter that resists changes in motion. Does the energy the speed of light. We can get the electron arbitrarily
of an object really increase its resistance to changes in close to that number—0.9999999999999999999999 . . .
motion? Yes. Even adding to the energy of motion of an × c is no problem, at least in principle—but there is no
object increases its inertia. For example, a proton in a getting over the hump. In Figure 17.21 we show how a
high-energy particle accelerator may approach the speed rocket ship, which experiences a constant acceleration
of light so closely that its total energy is 1,000 times equal to that of gravity on Earth (so that its occupants
greater than its rest energy. Such an energetic proton is, will feel at home), moves faster and faster but never
indeed, harder to “push around” (in other words, it has reaches the speed of light. Faster-than-light travel may
more inertia) than a proton at rest. be a mainstay of science fiction, but this is one of those
cases where wishing that something is physically pos-
2. The speed of light is the ultimate speed limit. There are sible does not necessarily make it so.
several ways to think about this. We already discussed
the insight that led Einstein to relativity in the fi rst place. 3. Time passes more slowly in a moving reference frame.
Were it possible to travel at the speed of light, then in This phenomenon is referred to as time dilation because
that frame of reference light would cease to be a trav- time is “stretched out” in the moving reference frame.
eling wave, and all of the laws of physics would come Were you to compare clocks with an observer moving
tumbling down around us. We can also think about this at 9 ⁄10 the speed of light (0.9c), you would fi nd that the
limit in terms of the equivalence of mass and energy other observer’s clock was running less than half as fast
just discussed. As the speed of an object gets closer and as your clock (about 0.44 times as fast).3 You might guess
closer to the speed of light, its energy, and therefore its that to the other observer your clock would be fast, but
mass, become greater and greater, so it becomes increas- actually the other observer would fi nd instead that it
ingly resistant to further changes in its motion. We can is your clock that was running slow! A bit of thought
continue to push on it all we like, making it go faster, shows why it must be this way. To you, the other observer
but we face diminishing returns. may be moving at 0.9c, but to the other observer, you
The situation is like trying to get from 0 to 1 by are moving. Either frame of reference is equally valid,
halving the remainder again and again. The resulting so it stands to reason that if a clock in a moving refer-
sequence—0, 1⁄2, 3⁄4, 7⁄8, 15⁄16, 31⁄32, 63⁄64, . . .—gets arbitrarily ence frame runs slow, you would each fi nd the other’s
close to 1 but never actually reaches it. In the same way, clock to be slow.
a continuous force applied to an object will cause its
velocity to get closer and closer to the speed of light, 3
The factor by which time is dilated and space is contracted is given by
but it will never actually reach the speed of light. You 1
_______
_____ .
just cannot get there. It would take an infinite amount of
energy to accelerate an object with a nonzero rest mass √ v2
1 – __2
c
to the speed of light. In short, all the energy in the entire This is called the Lorentz factor, named for Dutch physicist Hendrik
universe is inadequate to accelerate a single electron to Lorentz (1853–1928), and is often symbolized as γ.
These consequences of relativity can be combined in

Imagine that 1,000 μ μ μ μ what is often called the “twin paradox.” Suppose you head
muons are produced
at a height of 15 km.
off on a trip to the center of the Milky Way Galaxy, roughly
25,000 light-years distant.
Your spectacularly powerful The twin paradox
v = 0.9c v = 0.999c
star drive accelerates your illustrates many
v = 0.99c v = 0.9999c
ship up to 0.9999999992c. To aspects of relativity.
you, the surrounding galaxy
15 km
is moving by at this speed, so the distance of 25,000 light-
years to the center of the galaxy is compressed by a factor
of 25,000, to a distance of a single light-year (see number
5 in the previous list). At your speed, you cross this dis-
tance in a single year. You snap a picture of the gas swirl-
ing around the black hole at the center of the galaxy and
then turn around to head home and show it to your twin.
1 108 495 800
Again, the return trip takes only a year. So, in 2 years you
Number of muons reaching have traveled to the center of the galaxy and back again.
ground before decaying.
(Who says interstellar travel is such a big deal?) But when
you return, you fi nd that your twin died 50,000 years ago.
FIGURE 17. 22 Muons created by cosmic rays high in Earth’s
In the reference frame of Earth, your spacecraft crossed
atmosphere would decay long before reaching the ground if they
the 25,000–light-year distance to the center of the galaxy
were not traveling at nearly the speed of light. An observer on the
moving at just under the speed of light. The only reason
ground sees a longer lifetime because of time dilation. Here we
show what happens to 1,000 muons produced at an altitude of you survived the journey, according to an Earth-bound
15 km for a variety of speeds. observer, is that in your moving frames of reference, clocks
(including biological clocks) ran extremely slow (number
3 in our list). Each leg of the two-way journey took 25,000
A straightforward scientific observation demon- years to observers on Earth, and your twin just could not
strates this effect. As illustrated in Figure 17.22, fast wait that long.
particles called “cosmic-ray muons” are produced at You might puzzle over the twin paradox a bit. Both on
about the 15-km level in Earth’s atmosphere when high- the way out and on the way back, in your reference frame
energy primary cosmic rays strike atmospheric atoms it is the clocks on Earth that are running slowly, so you are
or molecules. Muons at rest decay very rapidly into aging faster than your twin. Yet when you return, more
other particles. Within 2.2 microseconds (
s), half of all time has passed for your twin than for you. How can this
muons will have changed their identity. This happens be? The answer is that, unlike your twin, you changed ref-
so quickly that, even if they could move at the speed erence frames during your trip. Event 1 is when you left
of light, virtually all muons would have decayed long Earth, and event 2 is when you returned to Earth. Your
before traveling the 15 km to reach Earth’s surface. How- twin went from one event to the other while riding along in
ever, the time dilation effect causes the muons’ clocks Earth’s frame of reference. You, on the other hand, changed
to run slower, so the particles live longer and can travel reference frames when you left Earth, changed again when
farther. That is why we are able to detect cosmic-ray you stopped at the center of the galaxy, changed a third
muons on the ground. time when you left the galactic center to return home, and
changed reference frames one fi nal time when you arrived
4. “At the same time” is a relative concept. Two events back at Earth. It happens that the path through spacetime
that occur at the same time for one observer may occur that you followed between the two events involved the pas-
at different times for a different observer. Hold out your sage of only 2 years of what you experienced as time, while
arms and snap the fingers on both hands at the same your twin’s path involved 50,000 years of what your twin
time. For you, the two snaps were simultaneous. But to experienced as time.
an observer moving by you from right to left at nearly Another way to view this is that the key difference
the speed of light, you snapped the fingers of your left between you and your twin is that you experienced accel-
hand fi rst and the fi ngers of your right hand later. eration during your trip, while your twin did not. When
two observers are in uniform motion relative to one another,
5. An object is shorter in motion than it is at rest. More neither of them can lay claim to being in a unique frame
specifically, moving objects are compressed in the direc- of reference. However, acceleration is a real phenomenon.
tion of their motion. A meter stick moving at 0.9c is only You feel acceleration when you are riding in a car, and you
43.6 centimeters (cm) long.4 would surely feel the acceleration of the spaceship in this
example. It is the fact that you experienced an acceleration
4
See footnote 3. that enabled you to “outlive” your twin.
(a) “Stationary”
17.7 Gravity Is a Distortion
of Spacetime
Our exploration of special relativity began with the obser-
vation that the speed of light is always the same (regardless
of the motion of an observer or the source), and ended by
shattering our everyday
notions of space and time. Mass warps the fabric
Now, as we confront the prop- of spacetime. A spacecraft that is “stationary” in deep space
erties of black holes—indeed, and a spacecraft that is moving at a constant
velocity both represent inertial reference
of all massive objects in the universe—our concepts of space frames. Both are floating freely in space.
and time will be pulled even further from the comfortable
absolutes of Newtonian physics. First we learned that what (b) Moving at constant velocity
we traditionally called (three-dimensional) space and time
are actually just a result of our particular, limited perspec-
tive on a four-dimensional spacetime that is different for
each observer. Now we shall discover that this four-dimen-
sional spacetime is warped and distorted by the masses it
v = 0.9999 c
contains. One of the consequences of this deformation is
the gravity that holds you to Earth. This realization, called
the general theory of relativity, is another of Einstein’s great
contributions to science.
A crucial clue to the fundamental connection between
gravity and spacetime has been with us since Chapter 3,
where we found that the iner-
tial mass of an object—the Inertial mass and
A spaceship falling freely in a gravitational
mass appearing in Newton’s gravitational mass field also represents an inertial reference
(c) Freely falling in a
F = ma—is exactly the same are the same. gravitational field frame, even though it is accelerating.
as the object’s gravitational
mass. Another clue is that, left on their own, any two objects
at the same location and moving with the same velocity will
follow the same path through spacetime, regardless of their v
masses. The space shuttle astronaut falls around Earth,
moving in lockstep with the space shuttle itself. A feather
dropped by an Apollo astronaut standing on the Moon falls a
toward the surface of the Moon at exactly the same rate as
a dropped hammer does. In some sense, rather than think- The equivalence between
ing of gravity as a “force” that “acts on” objects, it is more “free fall” and “free float” is
accurate to think of it as a consequence of the path through the basis of general relativity.
spacetime that objects will follow in the absence of other
forces. Gravitation is the result of the shape of the spacetime FIGURE 17. 23 Special relativity says that there is no
terrain through which objects move. difference between (a) a reference frame that is floating
“stationary” in space and (b) one that is moving through the
galaxy at constant velocity. General relativity adds that there is
no difference between these inertial reference frames and (c)
Free Fall Is the Same as Free Float an inertial reference frame that is falling freely in a gravitational
field. Free fall is the same as free float, as far as the laws of
The essence of special relativity is that any inertial reference physics are concerned.
frame is as good as any other. There is no experiment you can
do to distinguish between sitting in an enclosed spaceship
floating stationary in deep space (Figure 17.23a) and sitting
in an enclosed spaceship traveling through our galaxy at Each of these reference frames is an equally valid inertial ref-
0.99999 times the speed of light (Figure 17.23b). You cannot erence frame. As long as nothing acts to change the motion—
tell any difference between these two cases—they do not feel that is, nothing pushes on either spaceship—the laws of
any different—because there is no difference between them. physics are exactly the same inside both spacecraft.
General relativity begins by the application of this same shown in Figure 17.24, demonstrates the point. Imagine you
idea to an astronaut inside the space shuttle orbiting Earth, are in a box inside a rocket ship that is accelerating through
as shown in Figure 17.23c. As deep space at a rate of 9.8 meters per second per second
long as we restrict our atten- A freely falling object (m/s2) in the direction of the arrow shown in Figure 17.24b.
tion to a small enough volume defines an inertial The floor of the box exerts enough of a force on you to over-
of space and a short enough reference frame. come your inertia and cause you to accelerate at 9.8 m/s2,
period of time that we can so you feel as though you are being pushed into the floor
ignore changes in the strength and direction of gravity from of the box. Now imagine instead that you are sitting in a
place to place, our astronaut again has no way to tell the dif- closed box on the surface of Earth. Again the floor of the box
ference between being inside the space shuttle as it falls exerts enough upward force on you to overcome your iner-
around Earth and being inside a spaceship coasting through tia, causing you to accelerate at 9.8 m/s2. You feel as though
interstellar space. Close your eyes and jump off a diving you are being pushed into the floor of the box.
board. For the brief time that you are falling freely through According to the equivalence principle, the two cases
Earth’s gravitational field, the sensation you feel is exactly are identical. There is no difference between sitting in an
the same as the sensation that you would feel adrift in the armchair in a rocket ship traveling through deep space with
abyss of interstellar space! The implications of this result an acceleration of 9.8 m/s2 and sitting in an armchair on the
are somewhat startling. Even though its velocity is constantly surface of Earth reading this book. In the first case, the force
changing as it falls, the inside of a space shuttle orbiting of the rocket ship is pushing you away from your “floating”
Earth is as good an inertial frame of reference as that of an straight-line geodesic through spacetime. In the second case,
object drifting along a straight line through interstellar space. Earth’s surface is pushing you away from your curved “fall-
This principle—which can be simply stated as “free fall is ing” geodesic through a spacetime that has been distorted
the same as free float”—is called the equivalence by the mass of Earth. An acceleration is an acceleration,
principle. regardless of whether you are
The equivalence principle says that a falling object is being accelerated off a straight- Being stationary in a
simply following its “natural” path through spacetime—it line geodesic through deep gravitational field is
is going where its inertia carries it—every bit as much as space or being accelerated off the same as being in
an object that drifts along a straight line at a constant speed a “falling” geodesic in the an accelerated
through deep space. The natural path that an object will gravitational field of Earth. reference frame.
follow through spacetime in And in all cases, it is the same
the absence of other forces is Falling objects follow mass—the mass that gives an object inertia—that resists
referred to as the object’s curved paths through the change. Gravitational mass and inertial mass are the
“world line” or geodesic. In curved spacetime. same thing!
the absence of a gravitational There is an important caveat to the equivalence prin-
field, the geodesic of an object is a straight line—hence ciple. In an accelerated reference frame such as an accel-
Newton’s statement of inertia that an object, unless acted erating rocket ship, the same acceleration is experienced
on by an unbalanced external force, will move at a constant everywhere. In contrast, the curvature of space by a massive
speed in a constant direction. However, in the presence of object changes from place to place. Tides are one result of
mass the shape of spacetime becomes distorted, so an changes in the curvature of space from one place to another.
object’s geodesic becomes curved. A more careful statement of the equivalence principle is that
The equivalence principle gives us a way of understand- the effects of gravity and acceleration are indistinguish-
ing why gravitational mass and inertial mass are one and able locally—that is, as long as we restrict our attention to
the same thing. When we discussed Newton’s law, F = ma, small enough volumes of space that changes in gravity can
we found it useful to state it instead as a = F/m. When you be ignored.
apply a force F to an object, you move it away from its nat-
ural path and its inertial mass m tells you how strongly it
resists the change. General relativity says that when you
are standing still on the surface of Earth, your natural path
Spacetime as a Rubber Sheet
through spacetime—your geodesic—is actually a path fall- General relativity is a geometric theory. It describes how
ing inward toward the center of Earth. In the absence of mass distorts the geometry of spacetime. You can get a sense
any external forces, this is what you would do. Of course, for how mass distorts spacet-
the surface of Earth gets in your way. Put another way, the ime by imagining the surface Mass distorts
surface of Earth exerts an external force on your feet, and of a tightly stretched rubber the geometry
that force causes you to accelerate continuously away from sheet. The rubber sheet is flat. of spacetime.
your natural path through spacetime. If you roll a marble across the
This idea leads to a different, equally valid, way of stat- sheet, it will roll in a straight line. All of the Euclidean
ing the equivalence principle. Another thought experiment, geometry that you learned in high school applies on the
FIGURE 17. 2 4 According

(a)
to the equivalence principle,
an object falling freely in a
1 Free fall… 2 …is the same as free float.
gravitational field is in an inertial
reference frame (a), and an
g Gravity object at rest in a gravitational
field is in an accelerated frame
of reference (b).
Both are inertial reference

frames—natural geodesics
through spacetime.
4 …is the same as sitting

(b)
in an accelerating room
with no gravity.
g Gravity
3 Sitting still in a
gravitational field… a
Force F
Acceleration
F
Force
Both are accelerated reference frames
in which forces are pushing objects away
from their natural geodesics.
F = m gravitational g F = m inertial a
Inertial and gravitational mass both resist
these forces. They are the same thing.
surface of the sheet as well: The angles in a triangle on the you do this and then look at the sheet from above, you will
sheet add up to 180°. Right triangles obey the Pythagorean be amazed to see that rather than adding to 180°, the angles
theorem. If you draw a circle on the sheet, you will find that in this new triangle sum to more than 180°. The surface of
the circumference of the circle is equal to 2π times its radius. the sheet is no longer flat, and Euclid’s geometry (“plane
If you draw a line on this sheet and a point off to one side, geometry”) no longer applies.
there will be exactly one line that passes through that point Mass has an effect on the fabric of spacetime that is
but never intersects the fi rst line. analogous to the effect of the bowling ball on the fabric
Now, though, think about what happens if you place a of the rubber sheet. The bowling ball stretches the sheet,
bowling ball in the middle of the rubber sheet, as in Fig- changing the distances between any two points on the sur-
ure 17.25. The surface of the sheet will be stretched and face of the sheet. (Think of the deep depression in the rub-
distorted. If you roll a marble across the sheet, its path will ber sheet as a “well.” We will frequently use this language.)
dip and curve (Figure 17.25a). You might even fi nd that you Similarly, mass distorts the shape of spacetime, changing
can roll the marble so that it moves around and around the the “distance” between any two locations or events in that
bowling ball, like a planet orbiting about the Sun. Next spacetime.
you might revisit the relationship between the radius of a It is easy to understand how the two-dimensional sur-
circle and its circumference. If you draw a circle around face of the sheet is distorted by the bowling ball, because
the bowling ball, measure the distance around that circle, we can visualize how the sheet is stretched through a third
and then compare that distance with the distance from the spatial dimension. It is virtually impossible for us to “see” in
circle to its center along the surface of the sheet (Figure our mind’s eye what a curved four-dimensional spacetime
17.25b), you will fi nd that the circumference of the circle would “look like.” Once again we have run into a limitation
is less than 2πr. Finally, you might try to draw a triangle in how our brains are wired. Yet certain experiments, much
on the surface of the sheet, connecting three points around like those done on the surface of the rubber sheet, demon-
the bowling ball with the straightest and shortest lines you strate that the geometry of our four-dimensional spacetime
can draw on the surface of the sheet, as in Figure 17.25c. If is distorted much like the rubber sheet.
1 Mass warps spacetime 2 Like space, the rubber

much as a bowling ball sheet “looks” flat when
warps a rubber sheet. viewed face-on.

17. 25 Mass warps the
(a)
geometry of spacetime in
much the same way that a
bowling ball warps the surface
of a stretched rubber sheet.
This distortion of spacetime
has many consequences; for
example, (a) objects follow
3 Marbles roll on 4 …and objects follow curved curved paths or geodesics
curved paths across geodesics—orbits—about
through curved spacetime, (b)
the warped sheet... massive objects. Gravity is
the result of curved space. the circumference of a circle
around a massive object is
less than 2π times the radius
of the circle, and (c) angles
(b)
in triangles need not add to
rr c < 2πr
2pr exactly 180°.
5 On a warpedRsheet, 6 ...the circumference

as in spacetime of a circle is less
warped by mass,... than 2πr...
(c)
7 ...and triangles have

more than 180°.
Before going any further in our discussion of general equal to 2π times the radius of Earth’s orbit, just like a circle
relativity, it is important to point out that general relativity drawn on a flat piece of paper.5 If you could carry out this
does not mean that Newton’s law of gravitation is “wrong.” experiment, however, you would fi nd instead that the rope
See Connections 17.1 for a discussion of what happens when around the circumference of Earth’s orbit was shorter by 10
one physical law supplants another. km than 2π times the length of the rope stretched from Earth
to the center of the Sun—just as the circumference of the
circle on the stretched rubber
The Observable Consequences sheet is less than 2π times the The consequences of
radius of the circle. curved spacetime
of General Relativity It is not practical to stretch include precession of
Curved spacetime has many observable consequences. You a rope from Earth to the Sun. Mercury’s orbit.
can imagine, at least in principle, stretching a rope all the But we can do an experiment
way around the circumference of Earth’s orbit about the that is conceptually very similar. The long axis of Mercury’s
Sun, and then comparing the length of that rope with the elliptical orbit about the Sun is slowly “precessing”; that is,
length of a rope taken from the orbit of Earth to the center
of the Sun. Having studied geometry in high school, you 5
As we learned in Chapter 3, Earth’s orbit is actually an ellipse, but it
might expect to find that the circumference of Earth’s orbit is is close enough to a circle for purposes of this thought experiment.
C O N N E C T I O N S 1 7.1
When One Physical Law Supplants Another

If you have been reading this chapter closely, you might feel the kind of approximation we use when, despite the curvature
some justifi able annoyance with this business of gravity. of Earth, we navigate a city using a flat road map.
Throughout the book until this point, we have described grav- Similarly, we now know that Newton’s laws of motion
ity as a force that obeys Newton’s universal law of gravitation: are actually approximations to the more generally applicable
F = Gm1m2/r 2. Now we suddenly introduce the ideas of general rules of special relativity and quantum mechanics. In fact, we
relativity and in the process ask you to totally change the way can derive Newton’s laws from special relativity and quantum
you think about gravity. So which is the real deal? If general mechanics by making the “everyday” assumptions that speeds
relativity is right, does that not imply that Newton’s formula- of objects are much less than the speed of light and that the
tion of gravity is wrong? And if so, then why have we continued sizes of objects are much greater than the subatomic particles
to use Newton’s law? from which atoms are made. We use Newton’s laws of motion
The answers to these questions go to the heart of how sci- and gravitation most of the time because they are far easier
ence progresses and how our conception of the universe evolves. to apply than the relativistically and quantum mechanically
Under most circumstances there is virtually no difference “correct” laws, and because any inaccuracies we introduce by
between the predictions made using general relativity and the using Newtonian approximations are usually far too tiny to
predictions made using Newton’s law. As long as a gravitational matter. It is only in conditions very different from those of our
field is not too strong, Newton’s law of gravitation is a very close everyday lives (such as the behavior of an electron in an atom or
approximation to the results of a calculation using general rela- the gravitational field of a black hole), or in special cases when
tivity. The meaning of “too strong” in this context is relative. very high accuracy is needed (such as the precise timing used
For example, in most ways the enormous gravitational field by the global positioning system [GPS] satellite network), that
near the core of a massive main-sequence star would be con- the more general laws must be used.
sidered “weak.” Had we used a general relativistic formulation This is a general feature of new scientific theories. If a new
of gravity rather than Newton’s laws to compute the structure theory is to replace an earlier, highly successful scientific
of a main-sequence star, it would have made virtually no dif- theory, the new theory must normally “hold the old theory
ference in the results of the calculation. Similarly, even though within it”—it must be able to reproduce the successes of the
spacetime is curved by gravity, this curvature near Earth is earlier theory—just as special relativity contains Newton’s
very slight, so over small regions it can be ignored entirely. The laws of motion and general relativity holds within it the suc-
flat Euclidean geometry is a good “local” approximation even cessful Newtonian description of gravity that we have used
to curved spacetime—and is a lot easier to use. This is exactly throughout the book.
the axis of Mercury’s orbit is slowly changing its direction. time bends the path of light resembles the way the lenses
Even after allowing for the perturbations caused by the grav- bend light in a pair of eyeglasses.
ity of the other planets, this precession is not predicted by The first measurement of gravitational lensing came dur-
Newton’s inverse square law of gravity, working in a flat ing the solar eclipse of 1919. Prior to the eclipse, English
Euclidean space. However, it is just what is predicted for astrophysicist Sir Arthur Stanley Eddington (1882–1944)
the path of a planet that is moving alternately deeper and measured the positions of a number of stars in the part of
outward within the stretched-out non-Euclidean fabric of the sky where the eclipse would occur. Eddington then
spacetime that has been warped by the Sun. repeated his measurement during the solar eclipse and
The real-life equivalent of the triangle with more than found that the apparent positions of the stars had been
180° is probably easier to understand. A straight line in space deflected outward by the presence of the Sun. The light from
is defined by the path followed the stars followed a bent path through the curved spacetime
by a beam of light. This is the Gravitational lensing around the Sun, causing the stars to appear farther apart
shortest distance between any can displace in Eddington’s measurement, as illustrated in Figure 17.26.
two points. A beam of light and distort an During the eclipse, the triangle formed by Earth and the two
moving through the distorted object’s image. stars contained more than 180°—just like the triangle on
spacetime around a massive the surface of our rubber sheet. The results of Eddington’s
object is bent by gravity, just as the lines in Figure 17.25c are measurements were just as predicted by Einstein’s theory.
bent by the curvature of the sheet. This phenomenon is called Eddington’s result was the fi rst experimental test of a pre-
gravitational lensing because the way the curvature of space- diction of general relativity and is considered to be one of
FIGURE 17. 26
Apparent position Measurements obtained by
1 Starlight following a geodesic through of star 1 Sir Arthur Eddington during
spacetime curved by the Sun’s mass does the total solar eclipse of 1919
not follow a straight path through space. found that the gravity of the
Sun bends the light from distant
stars by the amount predicted
Deflection of starlight Star 1 by Einstein’s general theory of
by the Sun’s gravity
relativity. This is an example of
gravitational lensing. Note that
Earth the “triangle” formed by Earth
True position and the two stars contains
Sun
of stars more than 180°, just like the
triangle in Figure 17.25c.
Star 2
2 As a result, the apparent positions

of stars are deflected when the
Sun is between us and the stars. Apparent position
of star 2
the landmark experiments of 20th century physics. More objects deep within a gravitational well are shifted to longer
recently, gravitational lensing has been used to search for wavelengths. Gravitational redshift is similar in its effect
unseen massive objects adrift in space. These objects do to the Doppler redshift we saw earlier. In fact, there is no
not give off light, but their gravity can distort the light from way to tell the difference between light that is redshifted by
background stars that they happen to pass in front of. (See gravity and the Doppler-shifted light from an object mov-
the discussion of MACHOs in Chapter 20.) ing away from us. Astronomers often describe the gravi-
General relativity also affects spacetime in ways that tational redshift of an object as an “equivalent velocity.”
have no direct comparison with a rubber sheet because The gravitational redshift of lines formed on the surface
mass distorts not only the geometry of space, but the geom- of the Sun is equivalent to a Doppler shift of 0.6 km/s. The
etry of time as well. The deeper we descend into the gravi- gravitational redshift of light from the surface of a white
tational field of a massive object, the more slowly our clocks dwarf is equivalent to a Doppler shift of about 50 km/s. The
appear to run from the perspective of a distant observer. gravitational redshift from the surface of a neutron star is
This effect is called general relativistic time dilation. To equivalent to a Doppler shift of about a tenth the speed of
understand one consequence light. Sometimes astronomers get sloppy and talk about the
of general relativistic time Time runs more slowly gravitational redshift as if it truly were a Doppler shift. We
dilation, suppose a light is near massive objects. might say, for example, that the “gravitational redshift of
attached to a clock sitting on the surface of a particular white dwarf is 57.1 km/s.” How-
the surface of a neutron star. The light is timed so that it ever, this analogy does not mean that the surface of the
flashes once a second. Because time near the surface of the white dwarf is moving away from us at 57.1 km/s. It means
star is “stretched out,” however, an observer far from the that time is running so slowly on the surface of the white
neutron star perceives the light to be pulsing less frequently dwarf that the light reaching us from the white dwarf looks
than once a second. The frequency of the flashing is low- as though it were coming from an object moving away from
ered. Now suppose we have an emission line source on the us at 57.1 km/s.
surface of the neutron star. Because time is running slowly Bringing this phenomenon a bit closer to home, a clock
on the surface of the neutron star, at least from our distant on the top of Mount Everest gains about 80 nanoseconds
perspective, the light that reaches us will have a lower fre- (ns—80 billionths of a second) a day compared with a
quency as well. Remember that lower frequency means clock at sea level. The difference between an object on the
longer wavelength. So the light from the source will be surface of Earth and an object in orbit is even greater. A
seen at a longer, redder wavelength than the wavelength at global positioning system (GPS) receiver uses the results of
which it was emitted. sophisticated calculations of the effects of general relativ-
This phenomenon, shown in Figure 17.27, is called the istic time dilation to help you accurately find your position
gravitational redshift because the wavelengths of light from on the surface of Earth. Even after allowing for slowing due
(a) Identical flashlights Color of light seen by

Massive object observer at a particular location
1 Time proceeds slower

when deep in a
gravitational field…
2 …than it does when

the gravitational field 3 As a result, light coming from deep
is weaker. in a gravitational field is redshifted
toward lower frequencies.
(b)
Neutron star
Light from the surfaces of neutron

stars and white dwarfs is noticeably
White dwarf redshifted by their strong gravity.
Normal star
FIGURE 17. 2 7 Time passes more slowly near massive objects because of the curvature of
spacetime. As a result, to a distant observer light from near a massive object will have a lower
frequency and longer wavelength. The closer the source of radiation is to the object (a) or the more
massive and compact the object is (b), the greater the gravitational redshift will be.
to special relativity, the clocks on the satellites that make We could easily fi ll the rest of this book with fascinat-
up the GPS run faster than clocks on the surface of Earth. ing tales about general relativity. It is pretty heady stuff if
If the satellite clocks and your GPS receiver did not correct you think about it—discuss-
for this and other effects of general relativity, then the posi- ing the fabric of the universe Gravitational waves
tion your GPS receiver reported would be in error by up to itself as though it were a sub- travel through the
half a kilometer. The fact that the GPS works is actually a stance in a test tube to be fabric of spacetime.
strong experimental confi rmation of a number of predic- poked and prodded (see Con-
tions of general relativity, including general relativistic nections 17.2). One fi nal phenomenon that we should men-
time dilation. tion before moving on is the phenomenon of gravitational
C O N N E C T I O N S 1 7. 2
General Relativity and the Structure of the Universe

Our discussion of gravity as the warping of spacetime by the The grandest application of general relativity will come as
mass it contains was motivated by the events that accompany we consider the history and fate of the universe itself. We live
the collapse of a massive star. If general relativity applied only in an expanding universe—one that is the result of a singular
to black holes, it might not have been appropriate to devote so event that took place about 14 billion years ago. This event,
much effort in this chapter to a discussion of it. However, this is called the Big Bang, was not the result of mass coming into
not the last time we will run into general relativity. Many of the existence within the spacetime of the universe. Rather, it was
same physical processes we have seen at work here also shape the coming into existence of the spacetime of the universe itself.
events in the larger universe. In our discussion of galaxies, for When we say that the universe is expanding, we do not mean
example, we will fi nd that some galaxies contain supermassive that the stars and the galaxies in the universe are flying away
black holes at their centers—objects with masses millions of from each other through space. Instead, the spacetime of the
times that of our Sun, which grow by consuming entire stars universe itself is expanding with time. Just as the spacetime
and interstellar clouds. Material falling into such monsters is around a black hole has a shape, so does the spacetime of the
a leading candidate to explain the powerful radiation from universe, and that shape is determined at least in part by the
quasars—beacons at the edge of the observable universe— mass that our universe contains. What was the universe like
which we will explore in greater detail in Chapter 19. We will when it was very young? Will the universe expand forever,
also fi nd that just as the Sun bends the path of light passing or is there enough mass in the universe to warp spacetime
near it, as observed by Eddington during the total eclipse of into a closed shape that will eventually collapse back in on
1919, the curved spacetime around distant galaxies can act as itself? General relativity gives astronomers the tools they
a lens, magnifying and distorting the appearance of even more need to address such basic questions about the nature of
distant objects behind them. existence itself.
waves. If you thump the surface of our rubber sheet, waves and they found once again that the orbital energy loss was
will move away from where you thump it, something like consistent with the radiation of gravitational waves. Still,
ripples spreading out over the surface of a pond. Similarly, both systems provide only indirect evidence. These mea-
the equations of general relativity predict that if you surements very strongly suggest that gravitational waves
“thump” the fabric of spacetime (for example, with the exist, but they do not prove it.
catastrophic collapse of a high-mass star or the formation Let’s stop for a minute and think about this. Here we are
of a black hole), then ripples in spacetime will move out- talking about a phenomenon that has never been observed,
ward at the speed of light. These gravitational waves are at least not directly. At the moment, gravitational waves
like electromagnetic waves in some respects. Accelerating exist only as a scientific theory. Is this so different from
an electrically charged particle gives rise to an electromag- the pseudoscientific theory of “intelligent design,” which
netic wave. Accelerating a massive object gives rise to gravi- claims that life on Earth was deliberately designed by an
tational waves. intelligent agent? Yes, very much so. The prediction that
Gravitational waves have never been observed in a gravitational waves exist is falsifiable. By this we mean
laboratory or anywhere else, but there is strong circum- that the gravitational wave theory can be shown to be false.
stantial evidence for their existence. In 1974, astronomers We have already noted that the theory predicts that certain
discovered a binary system consisting of two neutron events, such as the catastrophic collapse of a high-mass star
stars, one of which is an observable pulsar. By using the or the formation of a black hole, will generate gravitational
pulsar as a precise clock, astronomers have been able waves. As we learned in Chapter 5, astrophysicists have a
to very accurately measure the orbits of both stars. The new kind of “telescope,” called the Laser Interferometer
stars themselves are 2.8 solar radii apart, but their orbits Gravitational-Wave Observatory (LIGO), that will be able
are gradually decaying, which means that they are los- to detect the predicted gravitational waves emanating from
ing energy somewhere. Calculations show that the energy such events. When this happens, either we will see them or
being lost by the system is just what general relativity we will not. The theory that predicts gravitational waves can
predicts the system should be losing in the form of gravi- be tested because it is falsifiable! Pseudoscientific theories,
tational waves. More recently, astronomers discovered a on the other hand, are not falsifiable; they cannot be put to
similar binary pair separated by only 1.0 solar radius (R⊙), a straightforward “yes or no” test.
17.8 Back to Black Holes 1 As an explorer approaches

a stellar black hole, tides rip
her apart…
We began our digression into relativity when we encoun-
tered black holes, and we return to the nature of black holes
now. When we placed an object on the surface of our rubber
sheet, it caused a funnel-shaped distortion that is analogous
to the distortion of spacetime by a mass. Now imagine such
a funnel-shaped distortion in the rubber sheet that is infi-
nitely deep—a funnel that keeps getting narrower and nar-
rower as we go deeper and deeper, but that has no bottom.
This is the rubber sheet ana-
log to a black hole. With black Black holes are
holes, t he mat hematics bottomless pits in
describing the shape of space- spacetime.
time fail in the same way that
the mathematical expression 1/x fails when x = 0. Such a
mathematical anomaly is called a singularity. Black holes
are singularities in spacetime (Figure 17.28).
A black hole has only three properties: mass, electric
charge, and angular momentum. The amount of mass that 2 …and her signals to us
falls into a black hole determines the extent of its distor- are redshifted by the
black hole’s gravity.
tion of spacetime. The electric charge of a black hole is the
net electric charge of the matter that fell into it. The angu-
lar momentum of a black hole FIGURE 17. 29 Journey into a black hole.
causes the spacetime around

the black hole to be twisted
around much like the water
around an eddy in a river. German physicist Karl Schwarzschild (1873–1916), and it
Apart from these three prop- is proportional to the mass of the black hole:
erties, all information about the
A black hole 2GM BH
is a bottomless material that fell into the black hole Rs = ______ ,
well in the fabric is lost. Nothing of its former composi- c2
of spacetime. tion, structure, or history survives. where RS is the Schwarzschild radius, G is the universal
We can never actually “see” the gravitational constant, MBH is the mass of the black hole, and
singularity at the center of a black c is the speed of light. A black hole with a mass of 1 M⊙ has
hole. The closer an object is to a black an event horizon of about 3 km. A black hole with a mass
hole, the greater is its escape velocity equivalent to that of Earth would have an event horizon of
(the speed at which it would have to only about a centimeter—the mass of our planet compressed
move to escape from the gravity of into a volume equal to that of a pecan!
the black hole). There is a radial dis- Let’s consider what would happen if an adventurer were
tance from the black hole at which the willing to journey into a black hole (Figure 17.29). From
escape velocity reaches the speed of our perspective outside the black hole, we would see our
light. This point of no return, beyond adventurer fall toward the
which even light is trapped by the event horizon; but as she did, The event horizon
black hole, is called its event horizon. her watch would run more is the boundary of
The radius of the event horizon of a and more slowly and her no return.
black hole that has neither angular progress toward the event
momentum nor charge is called the horizon would get slower and slower as well. Like Achil-
“Schwarzschild radius,” named for les in Zeno’s famous paradox, even though our adventurer
got closer and closer to the event horizon, she would never
FIGURE 17. 28 A black
quite make it, from our perspective. The event horizon is
hole is a singularity in the where the gravitational redshift becomes infi nite and
curvature of spacetime. where clocks stop altogether. Yet our adventurer’s own
It is a gravitational well with experience would be very different. From her standpoint,
no bottom. there would be nothing special about the event horizon at
17.8 Back to Black Holes 503
1 In a vacuum, virtual particle- 2 When a pair forms near the event

antiparticle pairs spontaneously horizon of a black hole, one of the
appear and then annihilate each other. pair may fall into the black hole… 3 …while the other escapes.
Particle Antiparticle
4 The stream of particles from the

black hole is called Hawking radiation.
FIGURE 17. 30 In the vacuum of empty space, particles and antiparticles are constantly being
created and then annihilating each other. Near the event horizon of a black hole, however, one
particle may cross the horizon before it recombines with its partner. The remaining particle leaves
the black hole as Hawking radiation.
all. She would fall past the event horizon and on, deeper In the ordinary vacuum of empty space, quantum theory
into the black hole’s gravitational well. However, she would says that particles and their antiparticle “mates” sponta-
now have entered a region of spacetime cut off from the neously spring into existence and then quickly annihilate
rest of the universe. The event horizon is like a one-way each other and disappear. These particle pairs typically
door: once our adventurer has passed through, she can live for less than about 10 –21 second, but their effects are
never again pass back into the larger universe to which seen in sensitive measurements of atomic transitions. If
she once belonged. such a pair of virtual particles comes into existence near
Actually, we have overlooked a rather crucial fact. Our the event horizon of a very small black hole, as shown in
intrepid explorer would have been torn to shreds long before Figure 17.30, then one of the particles might wind up fall-
she reached the black hole. Near the event horizon of a ing into the black hole while the other particle is able to
3-M⊙ black hole, the difference in gravitational acceleration escape. Some of the gravitational energy of the black hole
between our explorer’s feet and her head—the tidal “force” will have been used up in making one of the pair of vir-
pulling her apart—would be about a billion times her weight tual particles real. Taking into account all of the esoteric
on the surface of Earth. Obviously, this is not an experiment physics, Hawking was able to show that a black hole should
we would ever want to perform! Although scientific theories actually emit a Planck spectrum, and that the effective
must produce testable predictions, it is not required that all temperature of this spectrum would increase as the black
individual predictions be directly testable. hole became smaller. Although this phenomenon, called
Hawking radiation, is of considerable interest to physicists
and astronomers, in a practical sense it is usually negli-
gible. A 3-M⊙ black hole should emit radiation at a whop-
“Seeing” Black Holes ping temperature of only 2 × 10 –8 K, which means that the
In 1974 the British physicist Stephen Hawking (1942–) real- black hole should radiate with a power of 1.6 × 10 –29 watts
ized that black holes should actually be sources of radiation. (W)—very feeble indeed.
FIGURE 17. 31
An artist’s depiction of the
Cygnus X-1 binary system,
showing material from the
B0 supergiant being pulled
off and falling onto an
accretion disk surrounding
the black hole, thereby
producing X-ray emission.
Hawking radiation is hardly a useful way to “see” a black also discovered to be part of a binary system. The wave-
hole. For all intents and purposes, black holes remain true lengths of absorption lines in the spectrum of HD 226868
to their name. Nonetheless, by the end of the 20th century are Doppler-shifted back and forth with a period of 5.6
astronomers had found strong circumstantial evidence for days. Using the same techniques we used to measure the
black holes in two very different kinds of systems. In Chap- masses of stars in Chapter 13 (namely, analyzing the orbits
ter 19 we will find that there is strong evidence for super- of binaries), astronomers found that the mass of the unseen
massive black holes at the very centers of galaxies, but the compact companion of HD 226868 must be at least 6 M⊙.
fi rst and perhaps strongest evidence for black holes comes (Because the tilt of the orbit of the binary is not known,
from X-ray binary stars in our only a lower limit can be determined.) The companion to
own galaxy. In 1972, astrono- Black holes are found HD 226868 is too compact to be a normal star, yet it is much
mers did not yet have a good through the effects more massive than the Chandrasekhar limit for a white
understanding of the X-ray of their gravity. dwarf or a neutron star. According to our understanding of
emission from stars. In that the laws of physics, such an object can only be a black hole.
year, the Uhuru X-ray satellite made a puzzling discovery. Astronomers believe that the X-ray emission from Cygnus
The brightest X-ray source in the constellation Cygnus was X-1 arises when material from the B0 supergiant falls onto
found to be rapidly flickering. We now know that the bright- an accretion disk surrounding the black hole, as illustrated
ness of the X-ray emission from this object, called Cygnus in Figure 17.31.
X-1, can change in as little as 0.01 second. For reasons we Since 1972, a number of other good candidates for stellar-
will cover in more detail in our discussion of quasars in mass black holes have been discovered. One such object is a
Chapter 19, this means that the source of the X-rays must rapidly varying X-ray source in our companion galaxy, the
be smaller than the distance that light travels in 0.01 sec- Large Magellanic Cloud. Called LMC X-3, this X-ray source
ond, or 3,000 km. Thus, the source of X-rays in Cygnus X-1 orbits a B3 main-sequence star every 1.7 days, and the data
must be smaller than Earth! show that the compact source must have a mass of at least
When astronomers began to study this object in other 9 M⊙. Although the evidence that these systems contain
parts of the electromagnetic spectrum, Cygnus X-1 was black holes is circumstantial, the arguments that lead to
identified with both a radio star and with an already cata- this conclusion seem airtight. With dozens of compelling
loged optical star called HD 226868. The spectrum of HD examples of such objects on the books, the evidence is in.
226868 shows that it is a normal B0 supergiant star with a Black holes, once regarded as nothing more than a bizarre
mass of about 30 M⊙. Such a star is far too cool to explain quirk of the mathematics describing gravitation and space-
the X-ray emission from Cygnus X-1. But HD 226868 was time, exist in nature!
Summary 505
term for the bottomless pit in the fabric of the universe

Seeing the Forest for the Trees that we call a black hole.
This part of our journey has not only taken us into the
The story of low-mass stars like our Sun is by and large realm of tortured spacetime but has also shown us the
a story of longevity and stability. No one has ever seen a answers to immediate questions about our own existence.
star less massive than about 0.8 M⊙ evolve off the main We have watched as high-mass stars begin their lives by
sequence, because all the time in the universe has liter- tapping the same reservoir of nuclear fuel as their low-
ally not been enough for this to happen even once. What mass counterparts do, fusing hydrogen into helium and
a contrast on this leg of our journey to look instead at helium into elements such as carbon. But the alchemy
the high-mass end of the family of stars! Rather than sta- of high-mass stars does not end with a cinder of carbon
bility, these stars offer spectacle. Living for only a short and oxygen. Instead, it goes on, fusing massive elements
while, they blaze with the light of thousands or even into more massive elements, and those elements into
millions of Suns. Such stars are few and far between, more massive elements yet.
but the role they play in the life (and in our understand- Finally, we have seen the spectacle of a Type II super-
ing) of the universe is significant far beyond their num- nova. As gravity has its final victory, pulling the inner
bers. Indeed, when we move on to study galaxies and the parts of a massive star down into a neutron star or black
universe, massive stars such as Cepheid variables will hole, the outer layers of the star blaze forth, casting into
provide the signposts that we use to gauge the scale of the universe the seeds of future generations of planets
our universe. and, in all likelihood, life.
Contemplating the lives of massive stars has also We now come to the end of this leg of our journey,
taken us even further afield into the extremes to which having filled out the family album of stars, following
matter, space, and time can be molded. White dwarfs may them from their origins in the vast reaches of interstel-
have seemed incomprehensibly bizarre when we encoun- lar space to their fi nal places of rest. But we are not
tered them in our discussion of low-mass stars. Objects done with stars—far from it. All that we have seen has
with densities of a ton per teaspoonful are so far beyond happened on the stage of galaxies, and the universe is
our everyday experience as to be all but unimaginable. the grand hall in which that stage resides. Knowledge
Now we have encountered neutron stars—objects a bil- of stars and the interstellar medium is the astronomers’
lion times more dense yet. The matter in a neutron star is starting point as we endeavor to tease from galaxies and
to the electron-degenerate matter in a white dwarf as the the universe their secrets. At the same time, knowledge
matter in a white dwarf is to the air in a summer breeze. of the theater will give us further insight into the play-
But even the neutron star is but a way station on our jour- ers on the stage.
ney to the limits of matter and mind. As we contemplate Thus, we again head outward, on the last major seg-
massive stars, we also come face to face with the most ment of our journey: an exploration of our expanding uni-
extreme object of all—if object is even an appropriate verse and the galaxies and other structures it contains.
• Special relativity concerns the relationship between

Summary events in space and time.
• The CNO cycle burns hydrogen in massive stars. • Inertial mass and gravitational mass are the same.
• Iron is the most massive element formed by fusion. • Mass warps the fabric of spacetime so that objects move
on the shortest path in this warped geometry.
• More massive elements are created in successive burn-
ing stages. • Time runs more slowly near massive objects.
• Objects deep in a gravitational well appear redshifted.
• Massive stars eventually explode as supernovae, leaving
behind neutron stars. • Black holes are bottomless pits in spacetime.
• Supernovae eject newly formed massive elements into
interstellar space. SmartWork, Norton’s online homework system,
• Neutron stars contain from 1.4 to about 3 M⊙ packed into
a 10-km-diameter sphere.
• Pulsars are rapidly spinning magnetized neutron stars. smartwork.wwnorton.com.
Student Questions 15. An astronomer sees a redshift in the spectrum of an

object. With no other information available, can she
determine whether this is an extremely dense object
(gravitational redshift) or one that is receding from us
1. Explain the differences between the ways that hydrogen (Doppler redshift)? Explain your answer.
is converted to helium in a low-mass star (proton-proton
*16. Einstein’s special theory of relativity tells us that no object
chain) and in a high-mass star (CNO cycle). What is the
can travel faster than, or even at, the speed of light. We
catalyst in the CNO cycle, and how does it take part in
know that light is an electromagnetic wave, but we know
the reaction?
that it is also a particle called a photon. If it acts as a par-
2. How does a low-mass star begin burning helium in its ticle, how can a photon travel at the speed of light?
core? What about a high-mass star? How are these pro-
17. Imagine a future cosmonaut traveling in a spaceship at
cesses different or similar?
0.866 times the speed of light. Special relativity says that
*3. Why does the core of a high-mass star not become degen- the length of the spaceship along the direction of flight
erate, as in the cases of low-mass stars? is only half of what it was when it was at rest on Earth.
The cosmonaut checks this prediction with a meter
4. For what two reasons does each post-helium-burning
stick that he brought with him. Will his measurement
cycle for high-mass stars (carbon, neon, oxygen, silicon,
confi rm the contracted length of his spaceship? Explain
and sulfur) become shorter than the preceding cycle?
your answer.
5. Cepheids are highly luminous, variable stars in which
18. Of the four forces in nature (strong nuclear, electro-
the period of variability is directly related to luminosity.
magnetic, weak nuclear, and gravity), gravity is by far
Explain why Cepheids are good indicators for determin-
the weakest. Why, then, is gravity such a dominant
ing stellar distances that lie beyond the limits of accurate
force in stellar evolution? (Note: Although not explic-
parallax measurements.
itly discussed so far in this text, the weak nuclear
6. Identify and explain two important ways in which super- force is involved in certain decay processes within the
novae influence the formation and evolution of new nucleus.)
stars.
19. Suppose astronomers discover a 3-M⊙ black hole located
7. Is a Type II supernova really an explosion, or is it an a few light-years from Earth. Should we be concerned
implosion? Explain your answer. that its tremendous gravitational pull will lead to our
untimely demise?
8. Describe what we will observe on Earth when Eta Cari-
nae explodes. 20. If you could watch a star falling into a black hole, how
would the color of the star change as it approached the
9. Recordings show that SN 1987A was detected by neu-
event horizon?
trinos on February 23, 1987. About 3 hours later it was
detected in optical light. Why was there this time 21. Why don’t we detect the effects of special and general
delay? relativity in our everyday lives here on Earth?
10. Why can the accretion disk around a neutron star release 22. Many movies and TV programs (like Star Wars, Star
so much more energy than the accretion disk around a Trek, or Battlestar Galactica) rely on faster-than-light
white dwarf, even though both stars have approximately travel. How likely is it that we will ever develop such
the same mass? technology in the future?
11. In Section 17.3, we told you that Type II supernovae blast
material outward at 30,000 km/s, while the material in
the Crab Nebula (see Section 17.4) is expanding at only A P P LY I N G T H E C O N C E P T S
1,500 km/s. What explains the difference?
23. Our galaxy has about 50,000 stars of average mass (0.5
12. An experienced astronomer can take one look at the H-R M⊙) for every main-sequence star of 20 M⊙. But 20-M⊙
diagram of a star cluster and immediately estimate its stars are about 104 times more luminous than the Sun,
age. How is this possible? and 0.5-M⊙ stars are only 0.08 times as luminous as the
Sun.
*13. Can the main-sequence turnoff tell us anything besides
a. How much more luminous is a single massive star
the age of a cluster? If so, what?
than the total luminosity of the 50,000 less massive
14. What do we mean by the binding energy of an atomic stars?
nucleus? How does this quantity help us calculate the b. How much mass is in the lower-mass stars compared
energy given off in nuclear fusion reactions? to the single high-mass star?
c. Which stars—lower-mass or higher-mass stars— 30. We know that pulsars are rotating neutron stars. For a
contain more mass in the galaxy and which produce pulsar that rotates 30 times per second, at what radius in
more light? the pulsar’s equatorial plane would a co-rotating satel-
Explain your answers. lite (rotating about the pulsar 30 times per second) have
to be moving at the speed of light? Compare this to the
24. In 1841 the 100-M⊙ star Eta Carinae was losing mass at the
pulsar radius of 1 km.
rate of 0.1 M⊙ per year. Let’s put that into perspective.
a. The mass of the Sun is 2 × 1030 kg. How much mass 31. Verify the claim in Section 17.4 that Earth would be
(in kilograms) did Eta Carinae lose each minute? roughly the size of a football stadium if it were as dense
b. The mass of the Moon is 7.35 × 1022 kg. How does Eta as a neutron star.
Carinae’s mass loss per minute compare with the mass
32. The Moon has a mass equal to 3.74 × 10 –8 M⊙. Suppose
of the Moon?
the Moon suddenly collapsed into a black hole.
*25. Using values given in Section 17.2, verify that an O star a. What would be the radius of the event horizon (the
can lose 20 percent of its mass during its main-sequence “point of no return”) around the black-hole Moon?
lifetime. b. What effect would this collapse have on tides raised
by the Moon on Earth? Explain.
**26. The approximate relationship between the luminosity
c. Do you think this event would generate gravitational
and the period of Cepheid variables is L star (L ⊙ units) =
waves? Explain.
335 P (days). Delta Cephei has a cycle period of 5.4 days
and a parallax of 0.0033 arcseconds (arcsec). A more dis- 33. If a spaceship approaching us at 0.9 times the speed
tant Cepheid variable appears 1/1,000 as bright as Delta of light shines a laser beam at Earth, how fast will the
Cephei and has a period of 54 days. photons in the beam be moving when they arrive at
a. How far away (in parsecs) is the more distant Cepheid Earth?
variable?
34. Suppose we discover signals from an alien civilization
b. Could the distance of the more distant Cepheid vari-
coming from a star that is 2,500 light-years away, and
able be measured by parallax? Explain.
we send you to visit it using the same spaceship as dis-
27. If the Crab Nebula has been expanding at an average cussed in the twin paradox of Section 17.6.
velocity of 3,000 km/s since A.D. 1054, what was its aver- a. How long will it take you to reach that planet, accord-
age radius in the year 2006? (Note: There are approxi- ing to your clock? According to our clock on Earth?
mately 3 × 107 seconds in a year.) According to the aliens on the other planet?
b. How likely is it that someone you know will be there
28. Use Einstein’s famous mass-energy equivalence formula
to greet you when you return to Earth?
(E = mc2) to verify that 5.88 × 1013 J of energy is released
from fusing 1 kg of helium.
29. According to Einstein, mass and energy are equivalent.
So which weighs more on Earth: a cup of hot coffee or
a cup of iced coffee? Why? Do you think the difference
is measurable?
PA R T I V Galaxies, the Universe, and Cosmology
A man said to the universe: “Sir, I exist!” “However,”

replied the universe, “The fact has not created in me a
sense of obligation.”
STEPHEN CRANE (1871–1900)
A hemispheric sky map showing tiny fluctuations in the cosmic background radiation (CBR).
CHAP TER 18
Our Expanding Universe

18.1 Twentieth Century
K E Y CONCEP T S
Astronomers Just as stars are the building blocks of the structure of
Discovered the our own galaxy, so too are galaxies the building blocks
of the structure of the universe. Observations of the
Universe of Galaxies motions of galaxies led to one of the most remarkable
discoveries in the history of our species: the universe is
We have come a long way on our journey of discovery, expanding! As we investigate this discovery we will learn
building an understanding of stars and the planetary
• That the universe contains hundreds of billions of
systems that surround them. Yet cosmically speaking,
everything we have come across so far is in our own back- galaxies.
yard. We will now greatly expand our vision to regions • That the cosmological principle predicts isotropy and
far beyond our Solar System and Milky Way. Let’s look homogeneity of the observed distribution and velocity of
at a deep-space image of a piece of “dark” sky, such as galaxies.
the Hubble Space Telescope image shown in Figure 18.1.
• About the discovery of Hubble’s law, which relates the
It reveals myriad faint smudges of light fi lling the gaps
redshift of a galaxy to its distance.
among a sparse smattering of nearby stars. It has long
been known that the sky contains faint, misty patches of • How Hubble’s law is used to map the universe in space
light. These objects were originally referred to as nebu- and to look back in time.
lae (singular: nebula) because of their nebulous (fuzzy) • That galaxies are not flying apart through space
appearance. Prior to the 1780s, observers using telescopes
but rather “ride along” as space itself expands in
and naked-eye observations had found only about a hun-
accordance with the general theory of relativity.
dred of these smudges of light. In 1784 the French comet
hunter Charles Messier (1730–1817) published a catalog of • That spacetime itself was born approximately 13.7
103 nebulous objects, mostly as a warning to other comet billion years ago in an event called the Big Bang.
hunters not to waste their time on these objects. Yet 20 • That each of the major predictions of Big Bang theory,
years later, courtesy of the remarkable observations of the from the glow of the cosmic background radiation to
astronomers William Herschel and his sister Caroline Her-
the types of atoms that fill the universe, is confirmed by
schel, that number jumped to 2,500. From this time on,
observation.
astronomers were aware of systematic differences in the
appearance of nebulae. Although some of the Herschels’ • What observations of the cosmic background tell us
nebulae looked diffuse and amorphous, most were round about the properties of the young universe.
or elliptical or resembled spiraling whirlpools. These dis-
509
510 Chapter 18 Our Expanding Universe
FIGURE 18.1 A deep image of a

section of “dark” sky made with the
Hubble Space Telescope (HST). When we
look hard enough, the sky seems literally
to be covered with faint galaxies. Nearly
every object in this image is a galaxy. The
faintest smudges are images of galaxies
being formed more than 10 billion years
ago.
instead galaxies located far beyond the

bounds of our local galaxy. Each tiny
smudge in Figure 18.1 is such a galaxy.
The universe contains billions upon
billions of galaxies—perhaps even an
infinite number of galaxies—but cer-
tainly more galaxies than there are
stars in our own galaxy! Each galaxy
is a collection ranging from millions to
hundreds of billions of stars, thereby
rivaling our own cosmic home. Most
of these galaxies are located at such
astonishing distances from us that they
appear too small and faint to see with
any but the most powerful telescopes.
We will study distant galaxies and the
Milky Way in Chapters 19 and 20.
As has often been the case in
tinctions formed the original three categories—diffuse, astronomy, the nature of galaxies was, at its heart, a ques-
elliptical, and spiral—used to classify nebulae. tion about size and distance. Early attempts to understand
Discovering the existence of nebulae was one thing, but the size of the Milky Way were confounded by interstellar
uncovering their true nature was quite another. Specula- dust, which blocks the passage of visible light and thus lim-
tions about the nature of these objects abounded for the its our view. Early astronomers, not knowing of the existence
next 140 years. It was suggested that spiral nebulae might or consequences of this dust, assumed that what they could
be planetary systems in various stages of formation. The see in visible light was all there is. Unable to see past this
great 18th century German philosopher Immanuel Kant obscuring shroud, they concluded that we live in a system
(1724–1804) had a very different idea. He speculated that of stars some 6,000 light-years across. It was not until the
nebulae were instead “island universes”—distant realms beginning of the 20th century that American astronomer
of existence separate from our own Milky Way. Herschel Harlow Shapley (1885–1972), of the Harvard College Obser-
himself shared this belief but realized that no telescope of vatory, estimated that our galaxy is over 300,000 light-years
his day would ever be able to resolve the issue. At the time, in size.1 Shapley based his estimate on observations of glob-
many astronomers thought that the sum of existence—the ular clusters, and these observations greatly expanded the
universe—consisted solely of the swarm of stars to which previously held views about the extent of space.
our Sun belongs. It was not until the first third of the 20th In an interesting historical twist, the same insight that
century that the technological tools became available to turn brought Shapley to the correct conclusion about the approxi-
Kant’s philosophical musing into scientific knowledge. mate size of the Milky Way also led him to an erroneous
Today we know that Kant was correct. Our Milky Way conclusion about the nature of the so-called spiral and ellip-
is only one of Kant’s island universes, which were renamed tical nebulae. In 1920, Shapley and fellow astronomer Heber
galaxies (from the Greek gala, meaning “milky”) to reflect D. Curtis (1872–1942) from California’s Lick Observatory met
this change in understanding. (The term universe is now in Washington DC to publicly debate these issues. Historians
used to refer to the full expanse of space and all that it con- call this meeting astronomy’s “Great Debate.” When the Great
tains.) Whereas most diffuse nebulae are nearby clouds of
gas and dust in our own Milky Way Galaxy, what Messier 1
This is actually the size of the galactic halo (see Chapter 20). The
and others thought of as elliptical and spiral nebulae are galactic disk is smaller, about 100,000 light-years across.
18.2 The Cosmological Principle Shapes Our View of the Universe 511
FIGURE 18. 2 The

Andromeda Galaxy, our nearest
large galactic neighbor, is about
2.5 million light-years away.
Observations of variable stars
as standard candles enabled
Hubble to estimate its distance.
His measurement provided the
first observational evidence of
the vastness of the universe.
Debate was held, Curtis defended the earlier, smaller model and the nearby Magellanic Cloud galaxies, though his stars
of our galaxy, but the tide against that picture was already appeared much fainter. Using the period-luminosity rela-
turning. The question about tion for Cepheid variable stars discussed in Chapter 17,
the nature of the observed spi- The Great Debate Hubble turned his observations of these stars into measure-
ral nebulae, however, was still focused attention on ments of the distances to these objects. The results showed
wide open. In Shapley’s opin- the size and distance that the distances to these
ion, his far larger Milky Way of nebulae. nebulae are far greater than The nondebater
was ample enough to encom- even the galaxy size that Hubble settled the
pass everything in the universe. Having worked to show that Shapley measured. No doubt Great Debate by
the galaxy is 50 times larger than previously thought, Shap- remained: spiral and ellipti- measuring the
ley balked at the idea that the whole universe was hugely cal nebulae are really galaxies distance to galaxies.
larger still. But Curtis favored the idea that the spiral nebu- in their own right, similar in
lae were really galaxies separate from our own and that the size to our own galaxy but located at truly immense dis-
universe was, indeed, far larger than our own galaxy. tances. Hubble may not have shared the stage during the
Unlike questions of politics and law, scientific questions Great Debate, but when he spoke through his results, those
are not resolved by the rhetorical skills of partisans. Instead, earlier questions were answered once and for all. Shapley’s
they are settled by the results of well-crafted and carefully Milky Way, itself vast beyond comprehension, is but a speck
conducted experiments and observations. However, scientific adrift in a universe full of galaxies.
debates do help bring issues into sharper focus, leading sci-
entists to concentrate their attention and efforts on key ques-
tions. The reason we call that 1920 meeting the Great Debate is
that it clearly marked the final steps that would lead to a cor-
rect understanding of nebulae. The 1920 debate set the stage
18.2 The Cosmological
and pointed the direction for the subsequent work of Edwin Principle Shapes Our
P. Hubble (1889–1953), whose name was to become forever
entwined with our modern understanding of the universe. View of the Universe
Using the newly fi nished 100-inch telescope on Mount
Wilson, high above the then-small city of Los Angeles, Hubble’s discovery greatly expanded our concept of the uni-
Hubble was able to fi nd some variable stars in the large verse, and as we look back on our journey of the mind—and
neighboring galaxy of Andromeda (Figure 18.2). He recog- on the history of our species—it is remarkable how far we
nized that these stars were very similar to the Cepheid vari- have come. Beginning with Copernicus’s realization some
able stars studied by Henrietta Leavitt in the Milky Way 500 years ago that Earth is not the center of all things, we
HOMOGENEOUS ANDAND
(a) HOMOGENEOUS ISOTROPIC
ISOTROPIC NEITHER
(b) NEITHER
HOMGENEOUS
HOMOGENEOUS
NOR ISOTROPIC
NOR ISOTROPIC
Same in all More galaxies in one

directions: Isotropic direction: Not isotropic
1 1
2 Same as observer 1: 2 Sees more galaxies than

Homogeneous observer 1: Not homogeneous
(c) HOMOGENEOUS,
HOMOGENEOUS, NOTNOT ISOTROPIC
ISOTROPIC HOMOGENEOUS,
(d) HOMOGENEOUS,
NOTNOT
ISOTROPIC
ISOTROPIC
Gaps in galaxies Sees motion along

in this direction… this direction…
…but not in
1 this direction: 1
Not isotropic
…but not along this

direction: Not isotropic
2 Same as observer 1: 2 Same as observer 1:

Homogeneous Homogeneous
FIGURE 18. 3 Homogeneity and isotropy in four different theoretical models of a universe. Blue
arrows indicate the direction of view. (a) The distribution of galaxies is uniform, so this universe is
both homogeneous and isotropic. (b) The density of galaxies is decreasing in one direction, so this
universe is neither homogeneous nor isotropic. (c) The bands of galaxies lie along a unique axis,
making this universe anisotropic. (d) The distribution of galaxies is uniform, but galaxies move along
only one direction, so this universe also is not isotropic.
have shared the insights of Galileo and Newton and their cosmological principle is that the conclusions we reach about
intellectual heirs as they have torn down the conceptual our universe should be more or less the same, regardless of
barriers separating terrestrial existence from that of the whether we live in the Milky Way or in a galaxy billions of
heavens. Casting aside the aura of mysticism and magic, light-years away at the limits of the observable universe. In
we can now go out and look at the night sky and begin to other words, we predict that if the cosmological principle is
see it for what it is. We now know that Earth sits within an correct, then our universe will be homogeneous.
enormous spiral galaxy consisting of hundreds of billions According to one popular dictionary, the word homoge-
of stars. This galaxy, in turn, is but one of at least hundreds neous means “having at all points the same composition
of billions of galaxies that fill a universe vastly larger than and properties.” Clearly the
our ancestors might have imagined. Through it all, the cos- universe is not truly homoge- Observers everywhere
mological principle has been at the center of our conceptual neous in an absolute sense of should see the
understanding. No progress has been possible without the the word. The conditions we same universe.
enabling realization that the rules that apply to one part of encounter at the surface of
our universe apply everywhere. Earth are very different from those we would encounter in
The time has come to put the cosmological principle to deep space or in the heart of the Sun. (Even homogenized
work in its namesake field: cosmology. The science of cosmol- milk varies at the molecular level from place to place.) When
ogy is the study of the universe itself, including its structure, we speak as cosmologists of homogeneity of the universe,
history, origins, and fate. As we stressed in Chapter 1, the we mean instead that stars and galaxies in our part of the
cosmological principle is not an article of faith. Rather, it is universe are much the same, and behave in the same man-
a testable scientific theory. An important prediction of the ner, as stars and galaxies in remote corners of the universe.
We also mean that stars and galaxies everywhere are dis-

tributed in space in much the same way as they are in our
cosmic neighborhood, and that observers in those galaxies
see the same properties for our universe that we do. In other
words, when we speak of a homogeneous universe, we are
applying the term homogeneous with a very broad brush.
It is not easy to verify the prediction of homogeneity
directly. We do not have the luxury of traveling from our
galaxy to a galaxy in the remote universe to see whether con-
ditions are the same. However, we can observe light arriving
from the distant universe and see the ways in which features
look the same or different. For example, we can look at the
way galaxies are distributed in space (see Figure 18.1) and
ask whether that distribution is homogeneous.
In addition to predicting that the universe is homoge-
neous, the cosmological principle requires that all observ-
ers (including us) have the same impression of the universe,
regardless of the direction in which they are looking. If
something is the same in all directions, then it is isotro-
pic. This prediction of the cosmological principle is much
easier to test directly than is homogeneity. For example, if
galaxies were lined up in rows—a violation of the cosmo-
logical principle—we would get very different impressions,
depending on the direction in which we looked. In most
instances, isotropy goes hand in hand with homogeneity,
and the cosmological principle requires them both. Fig-
ure 18.3 shows examples of how the universe could have
FIGURE 18. 4 Edwin Hubble, shown on the cover of
violated the cosmological principle by being either inho-
the February 9, 1948, issue of Time magazine.
mogeneous or anisotropic, as well as examples of how the
universe might satisfy the cosmological principle.
The isotropy and homogeneity of the distribution of gal-
axies in the universe are predictions of the cosmological Hubble’s colleagues, was taking images of these galaxies at
principle that we can go to the telescope and test directly. Lowell Observatory in Flagstaff, Arizona, to obtain their
As you read this, you can ben- spectra. Unsurprisingly, Slipher’s galaxy spectra looked
efit from the experience of The distribution like the spectra of ensembles
decades of astronomers who of galaxies is isotropic of stars with a bit of glowing Light from distant
have used the world’s most and homogeneous, interstellar gas mixed in. The galaxies is redshifted
powerful telescopes to test as predicted. surprise, however, was that to longer wavelengths.
these predictions. These the emission and absorption
observations show that the properties of the universe are lines in the spectra of these galaxies were seldom seen at
basically the same, regardless of the direction in which we the same wavelengths in laboratory-generated spectra. The
look. If we look on very large scales, the universe appears lines were almost always shifted to longer wavelengths, as
homogeneous as well. Had these observations turned out shown in Figure 18.5.
otherwise, they would have proven the cosmological prin- Slipher characterized the observed shifts in galaxy spec-
ciple false, but the cosmological principle withstood the tra as redshifts. That is, the spectral lines in these galaxies
test and is now a bedrock of modern cosmology. were shifted to longer, or redder, wavelengths. Recall from
Chapter 4 that the Doppler shift causes the observed wave-
lengths of objects moving away from us to shift toward the
red end of the spectrum. The wavelength at which a line
18.3 We Live in an is observed in an object that is stationary relative to the
observer is called the rest wavelength of the line, written
Expanding Universe λrest. The redshift of a galaxy, written z, is defi ned as the
difference between the observed wavelength, λobserved, and
As we have seen, one of the great pioneers of cosmology was the rest wavelength, divided by the rest wavelength:
Edwin Hubble, pictured in Figure 18.4. In the 1920s, Hubble
and his coworkers were studying the properties of a large λobserved - λrest
z = ___________ .
collection of galaxies. Vesto Slipher (1875–1969), one of λrest
(a)
1 Absorption lines in a
A star in the Milky Way
nearby star appear at λrest = 656.3 nm
their rest wavelengths…
vr = z c 0.75
λ – λrest
= observed c
Hα λrest
(b) Hδ Hγ Hβ = 672.9 nm 656.3 nm c
A galaxy with 2 …but absorption lines in the 656.3 nm
Vr = 7,600 km/s spectrum of a distant galaxy = 0.0253c = 7,600 km/s
are redshifted by the galaxy’s
motion away from us. 3 Vr is calculated from
the measured redshift.
0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75

Wavelength (μm) λobserved = 672.9 nm
FIGURE 18. 5 (a) A star in our galaxy, shown with its spectrum. (b) A distant galaxy, shown with its spectrum at the
same scale as that of the star. Note that lines in the galaxy spectrum are redshifted to longer wavelengths.
Hubble’s insight was to interpret Slipher’s redshifts as

Doppler shifts, and he concluded that almost all of the gal- All Observers See the Same
axies in the universe are moving away from the Milky Way Hubble Expansion
(see Math Tools 18.1). When
he combined these measure- Hubble’s law says that Hubble’s law is a remarkable observation about the universe
ments of galaxy recession a galaxy’s recession that has far-reaching implications. For one thing, Hubble’s
velocities with his own esti- velocity is proportional law helps us test the prediction that the universe is homoge-
mates of the distances to these to its distance. neous and isotropic. When we look at galaxies in one direc-
galaxies, he made one of the tion in the sky, we find that they obey the same Hubble law
greatest discoveries in the history of astronomy. Hubble that galaxies observed in other directions in the sky obey.
found that distant galaxies are moving away from us more But although Hubble’s law cor-
rapidly than are nearby galaxies. Specifically, the velocity roborates the prediction that Hubble’s law says
at which a galaxy is moving away from us is proportional our view of the universe is iso- that the universe is
to the distance of that galaxy. This simple relationship tropic, the law appears at first expanding uniformly.
between distance and recession velocity has become known glance to contradict the pre-
as Hubble’s law. AstroTour: Hubble’s Law diction of the cosmological principle that the universe is
When talking about the distances of remote galaxies in homogeneous. Hubble’s law might seem to imply that we are
the universe, we need a suitable yardstick. The yardstick sitting in a very special place—at the center of a tremendous
we choose is the mega-light-year (Mly), equal to a million expansion of space, with everything else in the universe
light-years. According to Hubble’s law, a galaxy located 100 streaming away from us. This initial impression, however,
Mly away is, on average, moving away from us at twice the is incorrect. Hubble’s law actually says that we are sitting in
speed of a galaxy at a distance of 50 Mly. a uniformly expanding universe and that the expansion looks
We usually write Hubble’s law as the same, regardless of the location of the galaxy from which
we view it! To understand why this is the case, we now turn
vr = H0 × dG,
to a useful model that you can build for yourself with mate-
where dG is the distance to the galaxy and v r is the reces- rials you can probably find in your desk.
sion velocity of the galaxy. H0 (which astronomers pro- Figure 18.6 shows a long rubber band with paper clips
nounce as “H naught”) is a attached along its length. If you stretch the rubber band,
constant of proportionality Hubble’s constant the paper clips, which represent galaxies in an expanding
and is called the Hubble con- relates a galaxy’s universe, get farther and farther apart. Imagine what this
stant. When thinking of Hub- recession velocity to expansion would look like if you were an ant riding on paper
ble’s law, just remember that its distance. clip A. As the rubber band is stretched, you notice that all of
if you double the distance dG, the paper clips are moving away from you. When you look at
you also double the recession velocity vr . clip B, which is the next clip over, you see it moving away
M AT H T O O L S 1 8 .1
Redshift: Calculating the Recession

Velocity and Distance of Galaxies
In our discussion of an expanding universe we defi ned the Suppose we observe a hydrogen line, with a rest wavelength
redshift, z, of a galaxy as of 122 nanometers (nm), in the spectrum of a distant galaxy. If
the observed wavelength of the hydrogen line is 124 nm, then
λobserved - λ rest its redshift is
z = __________ ,
λ rest
λobserved - λ rest ________
124 - 122
z = __________ = = 0.016.
λ rest 122
where λobserved and λ rest are, respectively, the observed and rest
wavelengths of a spectral line. We can now calculate the recession velocity from this
Because lines from distant galaxies have wavelengths redshift:
shifted to the red, the galaxies must be moving away from us.
Note that the redshift of a galaxy is independent of the wave- vr = z × c = 0.016 × 300,000 km/s = 4,800 km/s.
length of the line used to measure it. Note also that this equa-
tion is identical to the Doppler formula (see Chapter 4) if we How far away, though, is our distant galaxy? This is where
replace z with vr/c, where v r is the speed of an object moving Hubble’s law and the Hubble constant (H0 = 22 kilometers per
away from us and c is the speed of light. So, with a little rear- second per mega-light-year, or km/s/Mly) come in. Hubble’s law
ranging, we can rewrite the preceding equation as relates a galaxy’s recession velocity to its distance (v r = H0 ×
dG), where dG is the distance to a galaxy measured in millions
λobserved - λ rest
vr = __________ ×c of light-years, or mega-light-years (Mly). Again, a little shifting
λ rest
of variables gives us
or
vr 4,800 km/s
vr = z × c dG = ___ = ___________ = 220 Mly.
H0 22 km/s/Mly
(Be aware, however, that this correspondence works only for From a simple measurement of the wavelength of a hydrogen
velocities much slower than the speed of light. See Founda- line, we have learned that the distant galaxy is approximately
tions 18.1.) 220 million light-years away.
slowly. When you look at clip C, located twice as far down The observation that nearby paper clips move away
the line from you as B, you see it moving away twice as fast slowly and distant paper clips move away more rapidly does
as B. By the time you work your way down the line to clip E not say that we are at the center of anything. Instead, it says
(located four times as far away as B), you see it moving away that the rubber band is being stretched uniformly along its
four times as fast as B. In other words, from the perspective length. In like fashion, Hubble’s law for galaxies does not
of an ant riding on clip A, all of the other paper clips on the mean that our galaxy is at the center of an expanding uni-
rubber band are moving away with a speed that is propor- verse. Hubble’s law means that the universe is expanding
tional to their distance. The paper clips located along the uniformly. Any observer viewing our universe from any
rubber band obey a Hubble-like law. galaxy will see nearby galaxies moving away slowly and
This demonstration of Hubble’s law is a handy result, more distant galaxies moving away more rapidly. Observers
but the key insight to our analogy comes from realizing anywhere else will fi nd that the same Hubble law applies
that there is nothing special about the perspective of paper from their vantage point as applies from our vantage point
clip A. If, instead, the ant were riding on clip E, clip D on Earth. The expansion of the universe is homogeneous.
would be the one moving away slowly and clip A would
be moving away four times as fast. Repeat this experi-
ment for any paper clip along the rubber band, and you We Must Build a Distance Ladder
will arrive at the same result: the speed at which other
clips are moving away from the ant is proportional to
to Measure the Hubble Constant
their distance. Bringing our cosmological terminology to The discovery of Hubble’s law for galaxies marked a fun-
bear, the stretching rubber band is “homogeneous.” The damental change in our perception of the universe. It also
same Hubble-like law applies, regardless of the paper clip marked the beginning of a quest that has taken center stage
selected as a vantage point. in astronomy for nearly a century. The form of Hubble’s
Time 1 A B C D E
Ants ride paper clip As the rubber band is stretched,
“galaxies” attached to the distances between the
a rubber band universe. paper clips increase.
Time 2
Time 3
The universe according to ant A The universe according to ant E
A B C D E A B C D E
Time 1 Time 1
Time 2 B D B D Time 2
Time 3 Time 3
Paper clip A is stationary. Paper clip D, located three Paper clip B, located three Paper clip E is stationary.
All other paper clips are times as far away as B, moves times as far away as D, moves All other paper clips are
moving away. away three times as fast as B. away three times as fast as D. moving away.
Both ants see themselves as stationary,

with all other paper clips moving away
at speeds proportional to their distance.
VISUAL ANALOGY FIGURE 18 .6 In this analogy of Hubble’s law, a rubber band with paper clips
evenly spaced along its length is stretched. As the rubber band stretches, an ant riding on clip A
will observe clip C moving away twice as fast as clip B. Similarly, an ant riding on clip E will see clip C
moving away twice as fast as clip D. Any ant will see itself as stationary, regardless of which paper
clip it is riding, and it will see the other clips moving away with speed proportional to distance.
law—as a proportionality between recession velocity and The difficulty in determining the Hubble constant comes
distance—tells us that our universe is expanding. But to from the fact that we must measure the distances not only
know the present rate of the expansion, we need a good to nearby galaxies, but to galaxies that are very far away.
value for the Hubble constant, H0. To see why, think about the motion of the water in a river.
How shall we go about obtaining the value for the Hub- All of the water in a flowing river moves downstream, but
ble constant? In principle, the answer is straightforward. even very uniform and steady rivers contain eddies and
If we can measure the redshifts and distances of a number crosscurrents that disturb the uniform flow. If you want to
of galaxies and plot velocity versus distance, then H0 will get a good overall picture of river flow, you need to look at
be the slope of the resulting line. With a large enough tele- a large portion of the river, not just the motion of a single
scope, measuring the redshifts of galaxies is easy; however, leaf or two drifting downstream.
measuring the actual distances to galaxies is much, much Similarly, there are eddies and crosscurrents in the
more difficult. motions of galaxies that make up our universe. The over-
109 ly
laxies
is t a nt ga
D
106 ly laxies
by ga
Near
103 ly Way
Milky
1 ly ars
by st
Near
10–3 ly
m
Syste White dwarf
Solar supernovae
Radar
Parallax
Main-Sequence Fitting
Cepheids
Distant Standards
FIGURE 18.7 The distance ladder, indicating how the

distances to remote objects are achieved through a series of
steps beginning with relatively nearby objects. Distances are
given in light-years (ly).
all motion of galaxies in accord with Hubble’s law is often The distance ladder begins with the size of the astronomi-
referred to as “Hubble flow.” Departures from a smooth Hub- cal unit, measured by radar and telemetry from space
ble flow, referred to as peculiar velocities, are the result of probes. It then steps outward with stellar parallax (see
gravitational attractions that cause galaxies to fall toward Chapter 13), which enables us to measure distances to
their neighbors or toward large concentrations of mass scat- nearby stars and thereby to build up the H-R diagram. For
tered throughout the universe. If we look only at nearby more distant stars, astrono-
galaxies, their motions due to Hubble’s law will be small, so mers use their spectral and Measuring H0 requires
most of the motion that we see will instead be due to their luminosity classifications to measuring distances
peculiar velocities. If we want to measure the Hubble flow determine their position on to remote galaxies.
itself to obtain a reliable value for H0, we need to study galax- the H-R diagram. Its position
ies that are far enough away that most of their velocity comes on the H-R diagram provides a star’s luminosity, which in
from the expansion or Hubble flow and relatively little of turn enables us to estimate its distance by comparing its
their observed motion is due to peculiar velocity or gravity. apparent brightness with its known luminosity, as described
Peculiar velocities of galaxies are typically a few hun- in Appendix 6. Moving farther out, we can measure the
dred kilometers per second, so to determine H0 we need to distance to relatively nearby galaxies by identifying rec-
measure accurately the distances to galaxies with Hubble ognizable luminous objects, such as O stars, globular clus-
velocities of several thousand kilometers per second or more ters, planetary nebulae, novae, and variable stars. (In so
(that is, galaxies with redshift z greater than about 0.01). The doing, we must assume that the luminosity of each object
distances to such galaxies—150 Mly—are far too great to is the same as in our galaxy.) For example, the Magellanic
measure using the same techniques that enable us to mea- Clouds (see Figure 20.20) are located 160,000 light-years
sure the distances of objects in our own galaxy. Instead, we away and contain large numbers of Cepheid variable stars
must find recognizable objects of known luminosity within (we learned about Cepheid variables in Chapter 17). Cep-
our own galaxy that are also luminous enough to be seen at heid variables take us another step up the distance ladder
greater distances. We refer to these objects as standard can- and enable us to accurately measure distances to galaxies
dles.2 We will come back to the use of standard candles when as far away as 100 Mly. Even this is not far enough to deter-
we turn to our own Milky Way Galaxy in Chapter 20. mine a reliable value for the Hubble constant, but within
Distances of remote objects are measured in a series of that volume of space are many galaxies that we can scour
steps, referred to as the distance ladder, which relates dis- for yet more powerful distance indicators. Among the best
tances on a variety of scales, as illustrated in Figure 18.7. of these are Type I supernovae.3
2 3
This term is borrowed from an old unit of light intensity that was Formally, a specialist in the field would refer to these as “Type Ia
based on actual candles. supernovae,” as noted in Chapter 16.
(a) (b)
Recall from Chapter 16 that Type I supernovae are thought BEFORE AFTER
to occur when gas flows from an evolved star onto its white
dwarf companion, pushing the white dwarf over the Chan-
drasekhar limit for the mass of an electron-degenerate object.
When this happens, the overburdened white dwarf begins
to collapse and then explodes.
Because all Type I supernovae Type I supernovae
occur in white dwarfs of the are very luminous
same mass, we might expect standard candles.
all such explosions to have
about the same luminosity. This prediction is borne out by Supernova
observations of Type I supernovae in galaxies with known
distances. With a peak luminosity that outshines a hundred
billion Suns (Figure 18.8), Type I supernovae can be seen
FIGURE 18.8 Galaxy NGC 3877 before (a) and after (b) the
and measured with modern telescopes that can reach almost
explosion of a Type I supernova. Type I supernovae are extremely
to the edge of the observable universe.
luminous standard candles.
It is often said that a chain is no stronger than its weak-
est link. In constructing the cosmic distance ladder, the
situation is even worse: the fi nal ladder is no better than years away; a galaxy with a redshift of 0.2 is twice that dis-
the combination of the weaknesses in each of its rungs. For tance away. In short, once we
the last few decades of the 20th century, astronomers work- know H0, Hubble’s law makes Redshift tells us a
ing to obtain a value for H0 the once-difficult task of mea- galaxy’s distance.
were largely split into two Current suring distances in the uni-
camps. One group favored a measurements put verse relatively easy, providing us with a tool to literally
Hubble constant of about 18 H0 at 22 km/s/Mly. map the structure of the universe (see Math Tools 18.1). We
km/s/Mly; a second camp will put this tool to good use in Chapter 22 when we turn
viewed the data as supporting a value of about 35 km/s/Mly. our attention to the large-scale structure of the universe.
A virtual war raged between these two groups for years, Hubble’s law does more than place galaxies in space. It
until HST observations in the mid-1990s began to converge also places galaxies in time. Light travels at a huge but finite
on an intermediate value of about 22 km/s/Mly.4 The most speed. Remember that when we look at the Sun, we see it
recent (2008) analysis of satellite data using other measures as it existed 8 minutes ago. When we look at Alpha Centauri,
of distance has yielded results consistent with that value. the nearest stellar system
Figure 18.9 plots the measured recession velocities of beyond the Sun, we see it as When observing the
galaxies against their measured distances. Notice how well it existed 4.3 years ago. When universe, elsewhere
the universe follows Hubble’s law. Today, most astronomers we look at the center of our is “elsewhen.”
believe that we have measured the Hubble constant to an galaxy, the picture we see is
accuracy of 10 percent—and perhaps even better than that. 27,000 years old. When looking at a distant object, we speak
Therefore, we can be confident that it lies between 20 and of its look-back time—the time it has taken for the light
25 km/s/Mly and that it is likely to be further refi ned in from that object to reach our telescope. As we look into the
the years to come. distant universe, look-back times become very great indeed.
The distance to a galaxy whose redshift z = 0.1 is 1.4 bil-
lion light-years (assuming H0 = 22 km/s/Mly), so the look-
Hubble’s Law Maps the back time to that galaxy is 1.4 billion years. The look-back
time to a galaxy where z = 0.2 is 2.7 billion years. As we
Universe in Space and Time look at objects with greater and greater redshifts, we are
Why all the fuss about Hubble’s law and the Hubble con- seeing increasingly younger versions of our universe.
stant? What do these tell us about the universe? First, Hub-
ble’s law gives us a practical tool for measuring distances
to remote objects. Once we know the value of H0, Hubble’s
law enables us to turn a straightforward measurement of 18.4 The Universe Began
the redshift of a galaxy into knowledge of its distance. For
example, a galaxy with a redshift of 0.1 is 1.4 billion light- in the Big Bang
4
Astronomers normally express H0 in units of kilometers per second Hubble’s law provides a very powerful and practical tool
per million parsecs (km/s/Mpc) rather than kilometers per second for mapping the distribution of galaxies throughout the
per mega-light-years. In these units, 22 km/s/Mly corresponds to 72 universe. Yet the most significant aspect of Hubble’s law is
km/s/Mpc. what it tells us about the structure and origin of the uni-
Hubble’s original proposal that

(a) the universe is expanding was
based on measurements of only
a few nearby galaxies. FIGURE 18.9 (a) Hubble’s
original figure illustrating that
more distant galaxies are receding
faster than less distant galaxies.
(b) Modern data on galaxies up
to 30 times farther away than
those studied by Hubble show that
recession velocity is proportional
to distance.
(b)
30,000
Modern measurements of
expansion of the universe use
Recession velocity of
galaxies 30 times as distant H0 = 22 km/s/Mly

20,000 as those used by Hubble.
galaxy (km/s)
10,000 The slope of this line is

Hubble’s constant H0.
0
0 Hubble’s 500 1,000 1,500
original plot Distance (Mly)
verse itself. We know that all galaxies in the universe are observed fact of Hubble expansion without resorting to the
moving away from each other, but if we could run the idea that the universe came into existence in an extraordi-
“movie” of this separation backward in time, we would find narily dense fi reball billions of years ago. However, as more
the galaxies getting closer and closer together. Using Hub- and more observations have come in and more discoveries
ble’s law, we find that about 6.8 billion years ago, when the about the structure of the universe have been made, the Big
universe was half its present Bang theory has only grown stronger. Today only a very
age, all of the galaxies in the Expansion started few serious workers in this field question whether the Big
universe were separated from with a Big Bang. Bang took place. Virtually all the major predictions of the
each other by half their pres- Big Bang theory (expansion of the universe being but one
ent distances. Twelve billion years ago, all of the galaxies of them) have proven to be correct. As we will see, the Big
in the universe must have been separated from each other Bang theory for the origin of our universe is now such a well-
by about a tenth of their present distances. Assuming that corroborated theory that most astronomers would probably
galaxies have been moving apart at the same speed that we say it has crossed into the realm of scientific fact.
see today, then 13.7 billion years ago (a time equal to 1/H0), The implications of Hubble’s law are striking. This sin-
all the stars and galaxies that make up today’s universe gle discovery forever changed our concept of the origin,
must have been concentrated together at the same location! history, and possible future of the universe in which we
(See Math Tools 18.2.) The value of 1 divided by the Hubble live. At the same time, Hubble’s law has pointed to many
constant is referred to as the Hubble time. If our line of rea- new questions about the universe. To address them, we
soning is correct, then today’s universe is hurtling outward next need to consider exactly what we mean by the term
from a tremendous expansion of space that took place expanding universe.
approximately 13 billion to 14 billion years ago. This colos-
sal event, which marked the beginning of our universe, is
referred to as the Big Bang (Figure 18.11). AstroTour: Galaxies Are Not Flying
Hubble’s Law
The idea of the Big Bang greatly troubled many astrono-
Apart through Space
mers in the early and middle years of the 20th century. Sev- The mental picture that you have of the expanding uni-
eral different suggestions were put forward to explain the verse at this point in our discussion is probably one of a
M AT H T O O L S 1 8 . 2
Expansion and the Age of the Universe

We can use Hubble’s law to estimate the age of the universe. Therefore,
Consider two galaxies located 100 Mly (dG = 9.5 × 1020 kilo-
meters [km]) away from each other (Figure 18.10). If these two 19 × 1020 km
Time = ___________ = 4.32 × 1017 s = 1.37 × 1010 yr.
galaxies are flying apart from each other, then at some time 4,400 km/s
in the past they must have been together in the same place at
the same time. According to Hubble’s law, and assuming that Again, we calculate time as distance divided by speed
H0 = 22 km/s/Mly, the distance between these two galaxies is (twice the distance divided by twice the speed) to fi nd that
increasing at the rate these galaxies also took 13.7 billion years to reach their cur-
rent locations. We could do this calculation again and again
vr = H0 × dG = 22 km/s/Mly × 100 Mly = 2,200 km/s. for any pair of galaxies in the universe today. The farther apart
the two galaxies are, the faster they are moving. The thing that
Knowing the speed at which they are traveling, we can calcu- is the same for all galaxies is the time that it took them to get
late the time it took for the two galaxies to become separated to where they are today.
by 100 Mly: A look at the following math makes this conclusion clear.
Velocity equals Hubble’s constant multiplied by distance, so
Distance 9.5 × 1020 km when we calculate a time as “distance divided by velocity,” the
Time = _______ = ___________ = 4.32 × 1017 s = 1.37 × 1010 yr.
Speed 2,200 km/s distance factors on top and bottom cancel out. Writing this out
as an equation, we get
In other words, if expansion of the universe has been constant,
two galaxies that today are 100 Mly apart started out at the same Distance Distance 1
Time = _______ = ____________ = ___.
place 13.7 billion years ago. Velocity H0 × Distance H0
Now let’s do the same calculation with two galaxies that are
twice as far apart: 200 Mly, or 19 × 1020 km (see Figure 18.10). We defi ne 1/H0 as look-back time, or Hubble time. It is one
These two galaxies are twice as far apart, but the distance way of estimating the age of the universe.
between them is increasing twice as rapidly:
vr = H0 × dG = 4,400 km/s. FIGURE 18 .10 Assuming that the velocities of galaxies

remain constant, the time needed for any two galaxies to
reach their current locations is the same, regardless of
their separation.
Hubble’s law says that galaxies At this speed the galaxies

100 Mly apart are moving away reach their present separation
from each other at 2,200 km/s. in 13.7 billion years.
100 Mly
v = 2,200 km/s t= = 13.7 billion years
100 Mly 2,200 km/s
200 Mly
200 Mly v = 4,400 km/s t= = 13.7 billion years
4,400 km/s
Because more distant galaxies also move apart

faster, Hubble’s law says that any two galaxies reach
their present separation in the same 13.7 billion years.
300 Mly 300 Mly

v = 6,600 km/s t= = 13.7 billion years
6,600 km/s
Time
FIGURE 18 .11 Looking backward in time,
the distance between any two galaxies is
Distance between galaxies

1 Hubble time 1/H0
= 13.7 billion years smaller and smaller, until all matter in the
3v universe is concentrated together at the same
point: the Big Bang.
2v If we run the clock backward,
Hubble’s law implies that
everything in the universe
v was at the same place 13.7
billion years ago.
Running the clock forward,

Hubble’s law implies that the universe
is expanding from a “Big Bang” that
The Big occurred 13.7 billion years ago.
Bang
13.7 billion
years ago 12 billion
years ago
7 billion years ago
Today
Time
cloud of debris from an explosion flying outward through the analogy of a ball placed on a stretched rubber sheet, show-
surrounding space. As we will see, however, this is not an ing how the ball distorted the surface of the sheet.
explosion in the usual sense of the word, and in fact there The surface of a rubber sheet can be distorted in other
is no surrounding space. ways as well. Imagine a number of coins placed on a rub-
One of the fi rst questions students usually ask about the ber sheet, as shown in Figure 18.12. Suppose we grab the
Big Bang is, “Where did it take place?” The answer to this edges of the sheet and begin pulling them outward. As the
question, amazing as it seems, rubber sheet stretches, each coin remains at the same loca-
is that the Big Bang took place The Big Bang tion on the surface of the sheet, but the distances between
everywhere. Wherever you are happened everywhere. the coins increase. Two coins sitting close to each other
in the universe today, you are move apart only slowly, while coins farther apart move
sitting at the site of the Big Bang. The reason is that galax- away from each other more rapidly. In other words, the dis-
ies are not flying apart through space at all. Rather, space tances and relative motions of the coins on the surface of
itself is expanding, carrying the stars and galaxies that a rubber sheet will obey a Hubble-like relationship as the
populate the universe along with it. sheet is stretched.
This may seem an incredible notion, but we have already Analogously, this movement is what is happening in the
dealt with the basic ideas that enable us to understand the universe, with galaxies taking the place of the coins and
expansion of space. In our dis- space itself taking the place of the rubber sheet. Obviously,
cussion of neutron stars and The expansion in the case of coins on a rubber sheet there is a limit to how
black holes in Chapter 17, we of space itself is far we can stretch the sheet before it breaks. With space and
encountered Einstein’s general similar to stretching a the real universe, there is no such limit. The fabric of space
theory of relativity. General two-dimensional can, in principle, go on expanding forever. Hubble’s law is
relativity says that space is dis- rubber sheet. the observational consequence of the fact that the space
torted by the presence of mass, making up the universe is expanding.
and that the consequence of this distortion is gravity. For How will this expansion of the universe behave in the
example, the mass of the Sun, like any other object, distorts future? In the next chapter we will learn that most of the
the geometry of spacetime around it; so Earth, coasting along mass in the universe consists of invisible dark matter, and
in its inertial frame of reference, follows a curved path around in Chapter 21 we will see that gravity due to this dark mat-
the Sun. We illustrated this phenomenon in Figure 17.25 with ter has the effect of slowing down the expansion of the

(a) Coins on a 18.12 (a) As a rubber sheet is
rubber sheet
stretched, coins on its surface
move farther apart, even
though they are not moving
with respect to the sheet itself.
Any coin on the surface of the
sheet will observe a Hubble-
like law in every direction. (b)
In analogous fashion, galaxies
in an expanding universe
As a rubber sheet stretches, coins
are not fl ying apart through
get farther apart, even though they
are not moving across the sheet. space. Rather, space itself is
stretching.
Similarly, galaxies do not move apart
through space. Rather, galaxies get
farther apart as space expands.
(b) Galaxies
in space
universe. But now it becomes more complicated. We will drew our ruler, each tick mark is separated from its neigh-
also learn that there is another unseen constituent of the bors by 1½ times the original distance, or 1.5 cm. If we
universe, called dark energy, and this constituent causes wanted to know the distance between two points, we could
the expansion of the universe instead to accelerate. At the still count the marks, but we would have to scale up the
current stage in the expansion of the universe, the acceler- distance between tick marks by 1.5 to fi nd the distance in
ating effect of dark energy is beginning to dominate over centimeters. The scale factor of the sheet is now 1.5. If the
the slowing effect of dark matter. sheet were twice the size that it was when we drew the
ruler (Figure 18.13b), each mark would correspond to 2 cm
of actual distance; the scale factor of the sheet would now
Expansion Is Described be 2. The scale factor tells us the size of the sheet relative
to its size at the time when we drew our ruler. The scale
with a Scale Factor factor also tells us how much the distance between points
When astronomers discuss the expansion of the universe, on the sheet has changed.
they talk in terms of the scale factor of the universe. To We can apply this same idea to the universe. Suppose
understand this concept, we return to our analogy of the rub- we choose today to lay out a “cosmic ruler” on the fabric of
ber sheet. Suppose we place a ruler on the surface of the sheet space, placing an imaginary
and draw a tick mark every centimeter, as in Figure 18.13a. tick mark every 10 Mly. We The scale factor, RU,
If we want to know the distance between two points on the defi ne the scale factor of the increases as the
sheet, all we need to do is count the marks between the two universe at this time to be 1. universe expands.
points and multiply by 1 centimeter (cm) per tick mark. In the past, when the universe
As the sheet is stretched, however, the distance between was smaller, distances between the points in space marked
the tick marks does not remain 1 cm. When the sheet is by our cosmic ruler would have been less than 10 Mly. The
stretched to 150 percent of the size that it had when we scale factor of that younger, smaller universe would have
(a) (b) Side of square = 2 cm VISUAL ANALOGY FIGURE

Scale factor RU = 2 18.13 (a) On a rubber sheet,
Rubber sheet
tick marks are drawn 1
cm apart. As the sheet is
stretched, the tick marks move
farther apart. (b) When the
spacing between the tick marks
is 2 cm, or twice the original
Side of square = 1 cm value, we say that the scale
Scale factor RU = 1 factor of the sheet, R U, has
doubled. A similar scale factor,
R U, is used to describe the
The stretching of a rubber sheet, or If the distance between two expansion of the universe.
the expansion of space, is measured points doubles, the scale
by changing the scale factor, RU. factor RU doubles as well.
been less than 1 compared to the scale factor today. In the are sitting. This is an important concept. If a particular
future, as the universe continues to expand, the distances point in today’s universe marked the site of the Big Bang,
between the tick marks on our cosmic ruler will grow to that would be a very special point indeed. But there is no
more than 10 Mly, and the scale factor of the universe will such point. The Big Bang happened everywhere. A Big Bang
be greater than 1. In this way we can use the scale factor, universe is homogeneous and isotropic, consistent with the
usually written as RU, to keep track of the changing scale cosmological principle.
of the universe.
It is important to remember, when thinking about the
expanding universe, that the laws of physics are themselves Redshift Is Due to the Changing
unchanged by the changing
scale factor. For example, when Expansion does not
Scale Factor of the Universe
we stretched out the rubber affect local physics— General relativity gives us a powerful tool for interpreting
sheet, we did not change the stars, atoms, or Hubble’s great discovery of the expanding universe. It also
properties of the coins on its anything else. forces us to rethink just what we mean when we talk about
surface. In like fashion, as the the redshift of distant galaxies. Although it is true that the
universe expands, the sizes and other physical properties of distance between galaxies is increasing as a result of the
atoms, stars, and galaxies also remain unchanged. expansion of the universe, and that we can use the equation
Let’s return now to the question of locating the center for Doppler shifts to measure the redshifts of galaxies, these
of expansion. Looking back in time, we see that the scale redshifts are not due to Doppler shifts at all! As light comes
factor of the universe gets smaller and smaller, approaching toward us from distant galaxies, the scale factor of the space
zero as we get closer and closer to the Big Bang. The fabric of through which the light travels is constantly increasing;
space that today spans billions of light-years spanned much and as it does so, the distance between adjacent wave crests
smaller distances when the universe was young. When the increases as well. The light is “stretched out” as the space
universe was only a day old, all of the space that we see it travels through expands. (See Foundations 18.1.)
today amounted to a region only a few times the size of our Let’s return to our rubber sheet analogy. If we draw a
Solar System. When the universe was a 50th of a second old, series of bands on the rubber sheet to represent the crests of
the vast expanse of space that makes up today’s observable an electromagnetic wave, as in Figure 18.15, we can watch
universe (and all the matter in it) occupied a volume only what happens to the wave as the sheet is stretched out. By
the size of today’s Earth. As we continue to approach the Big the time the sheet is stretched to twice its original size—
Bang itself, going backward in time, the space that makes up that is, by the time the scale factor of the sheet is 2—the
today’s observable universe becomes smaller and smaller— distance between wave crests has doubled. When the sheet
the size of a grapefruit, a marble, an atom, a proton. Every has been stretched to three times its original size (a scale
point in the fabric of space that makes up today’s universe factor of 3), the wavelength of the wave will be three times
was right there at the beginning, a part of that unimaginably what it was originally.
tiny, dense universe that emerged from the Big Bang. We apply this idea to light coming from a distant galaxy.
These points bear repetition: Where is the center of the When the light left the galaxy of its origin, the scale factor
Big Bang? There is no center. The Big Bang did not occur of the universe was smaller than it is today. The universe
at a specific point in space, because space itself came into expanded while the light was in transit, and as it did so,
existence with the Big Bang. Where did the Big Bang hap- the wavelength of the light grew longer in proportion to
pen? It happened everywhere, including right where you the increasing scale factor of the universe. The redshift of
When Redshift Exceeds 1

In our discussion of the Doppler shift in Chapter 4, we found see light from a galaxy with z = 2, we are seeing the universe
that (λobserved - λ rest) / λ rest is equal to the velocity of an object when it was one-third its current size. Nearby, this means that
away from us, divided by the speed of light. In this chapter we distance and look-back time are proportional to z. As we look
see how Edwin Hubble used this result to interpret the observed back closer and closer to the Big Bang, however, redshift climbs
redshifts of galaxies as evidence that galaxies throughout the more and more rapidly.
universe are moving away from us. Einstein’s special theory of Doppler’s original formula is essentially correct—as long
relativity says that nothing can move faster than the speed of as we are looking at shifts due to an object’s motion and the
light. Hubble’s initial assumption was that redshifts are due to velocities we measure are far less than the speed of light. In this
the Doppler effect. The resulting relation, z = vr /c, would then case, vr /c = (λobserved - λrest)/λrest. When we look at the motions of
seem to imply that no object can have a redshift (z) greater than orbiting binary stars or the peculiar velocities of galaxies rela-
1. Yet that is not the case. Astronomers routinely observe red- tive to the Hubble flow, this equation works just fi ne. But any
shifts significantly in excess of 1. As of this writing, the most time you have a redshift of 1 or greater, you should immediately
distant objects known have redshifts greater than 8! What is recognize that you are now in the realm of relativity.
wrong here? It is worth taking a brief digression to consider
how redshifts can exceed 1.
The fi rst thing to note is that, to arrive at the expression FIGURE 18 .14 Plot of the redshift (z) of an object versus
for the Doppler effect—vr /c = (λobserved - λ rest)/λ rest —we have its recession velocity (vr) as a fraction of the speed of light.
to assume that v r is much less than c. If that were not the According to special relativity, as vr approaches c, the redshift
becomes large without limit.
case—that is, if v r were close to c—we would have to take into
account more than just the fact that the waves from an object 10
are stretched out by the object’s motion away from us. We would
Doppler shift (z) gets arbitrarily
also have to consider relativistic effects, including the fact that 9 large as velocity approaches
moving clocks run slowly. When combining these effects, we the speed of light.
8
would fi nd that as the speed of an object approaches the speed
of light, its redshift becomes arbitrarily large (Figure 18.14).
7
A second source of redshift is the gravitational redshift
discussed in Chapter 17. As light escapes from deep within 6
a gravitational well, it loses energy, so photons are shifted
Redshift (z)
to longer and longer wavelengths. If the gravitational well 5

is deep enough, then the observed redshift of this radiation
can be boundlessly large. In fact, the event horizon of a black 4
hole—that is, the surface around the black hole from which (Relativistic)
3 Classical approximation to Doppler shift
not even light can escape—is where the gravitational redshift
Doppler shift, z = v r / c , works only
becomes infi nite. 2 at velocities much less than c.
Cosmological redshift, which is most relevant to this chap-
ter, results from the amount of “stretching” that space has 1
undergone during the time the light from its original source Classical approximation
has been en route to us. The amount of stretching that has 0 to Doppler shift z vr
occurred is given by the factor 1 + z. When we look at a dis-
tant galaxy whose redshift z = 1, we are seeing the universe 0 0.2 0.4 0.6 0.8 1.0
at a time when it was half the size that it is today. When we vr / c
light from distant galaxies is therefore a direct measure of When the light left its source, RU was equal to half of today’s
how much the universe has expanded since the time when value. If we see radiation with a redshift of 2, then the wave-
the radiation left its source. Redshift measures how much length of the radiation is three times its original wavelength.
the scale factor of the universe, RU, has changed since the The radiation was emitted when the scale factor of the uni-
light was emitted. verse was a third of what it is today. This wonderfully direct
If we see radiation with a redshift of 1, then its wave- relationship lets us use the observed redshift of a galaxy to
length is twice as long as when that radiation left its source. calculate the size of the universe at the look-back time to
18.5 The Major Predictions of the Big Bang Theory Are Resoundingly Confirmed 525
Scale, RU = 1 Electromagnetic
wave We See Radiation Left Over
Bands on a rubber sheet spread
out as the sheet is stretched.
from the Big Bang
The story of the single most important confirmation of the Big
Wave crests of Bang theory begins in the mid-1940s, when Russian-American
Scale, RU = 2
electromagnetic cosmologist George Gamow (1904–1968) and his student Ralph
waves spread out Alpher (1921–2007) were thinking about the implications of
as the universe
expands, redshifting Hubble expansion. Their logical reasoning about the expan-
the radiation. sion of the universe connected it to changes in temperature:
When a gas is compressed, it grows hotter. By contrast, as a gas
expands it becomes cooler. So, reasoned Gamow and Alpher,
since the universe is expanding, it must also be cooling. Con-
Scale, RU = 3
sequently, when the universe was very young and small, it
must have consisted of an extraordinarily hot, dense gas. As
with any hot, dense gas, this early universe would have been
awash in the same kind of radiation that we have encountered
so many times before on our journey of discovery: radiation
from a blackbody, which exhibits a Planck spectrum.
Gamow and Alpher took this idea a step further. As the
VISUAL ANALOGY FIGURE 18 .15 Bands drawn on a rubber sheet
universe expanded, they reasoned, this radiation would
represent the positions of the crests of an electromagnetic
have been redshifted to longer and longer wavelengths.
wave in space. As the rubber sheet is stretched—that is, as the
Recall Wien’s law in Chapter 4, which states that the tem-
universe expands—the wave crests get farther apart. The light is
redshifted.
perature associated with Planck radiation is inversely pro-
portional to the peak wavelength: T = (2,900 μm)/λpeak.
Shifting the wavelength of Planck radiation to longer and
longer wavelengths is therefore the equivalent of shifting
that galaxy. Written as an equation, the scale factor of the the temperature of the radiation to lower and lower values.
universe we see when looking at a distant galaxy is equal As illustrated in Figure 18.16, doubling the wavelength of
to 1, divided by 1 plus the redshift of the galaxy: the photons in a Planck spectrum by doubling the scale fac-
tor of the universe is equivalent to cutting the temperature
1
RU (z) = _____. of the Planck spectrum in half.
1+z On April Fools’ Day, 1948, Alpher and Gamow published
a paper asserting that this radiation should still be visible
today and should have a Planck spectrum with a tempera-
18.5 The Major Predictions ture in the neighborhood of 5–10 kelvins (K). The paper
actually listed as authors not only Alpher and Gamow, but
of the Big Bang Theory also Hans Bethe (1906–2005),
who—though an esteemed Alpher, Bethe, and
Are Resoundingly physicist in his own right— Gamow predicted the
did not actively participate in glow from a young hot
Confirmed this research. Gamow, being universe in 1948.
a renowned jokester, added
The questions we are grappling with when discussing ideas Bethe’s name between his and Alpher’s. “Alpher, Bethe, and
such as the origin of the universe are some of the most Gamow” is a play on the fi rst three letters of the Greek and
fundamental questions humankind can ask. Throughout Hebrew alphabets—an appropriate authorship for a paper
human history, answers to these same questions have been making predictions about the very early universe!
among the most prized goals of philosophers and theolo- This prediction languished until the early 1960s, when
gians. It is quite remarkable that we live in a time when two physicists at Bell Laboratories, Arno Penzias (1933–) and
we are fi nding real, testable answers to these questions by Robert Wilson (1936–), were trying to bounce radio signals
appealing not to divine inspiration or to the blind logic of off the newly launched Echo satellites. This hardly seems
philosophy, but rather to the empirical methods of science. much of a feat today, when we routinely use cell phones
It is essential, then, that we place extraordinary demands on and handheld GPS systems that communicate directly with
the quality of the evidence we rely on to support the theory satellites. At the time, however, the effort pushed radio
of the Big Bang. What evidence, apart from the observed technology to its limits. Penzias and Wilson needed a very
expansion of the universe itself, requires us to accept that sensitive microwave telescope for their work. Any spurious
the Big Bang actually took place? signals coming from the telescope itself might wash out the
FIGURE 18 .16 As the

Planck radiation in the
young universe is stretched universe expanded, Planck
by Hubble expansion… radiation left over from the hot
…into radiation that is still young universe was redshifted
a Planck spectrum, but to longer wavelengths.
at a lower temperature. Redshifting a Planck spectrum
ho is equivalent to lowering its
temperature.
Intensity
2ho
To
Intensity
Wavelength
To
2
Wavelength
faint signals bounced off a satellite. Pictured in Figure 18.17 along FIGURE 18 .17 Penzias and Wilson next to the Bell Labs
with their radio telescope, Penzias and Wilson worked tirelessly radio telescope antenna with which they discovered the
to eliminate all possible sources of interference originating from cosmic background radiation. This antenna is now a US
within their instrument. This work included such endless and National Historic Landmark.
menial tasks as keeping the telescope free of bird droppings and
other extraneous material. Even so, Penzias and Wilson found that
no matter how hard they tried to eliminate sources of extraneous
noise, they could still detect a faint microwave signal when they
pointed the telescope at the sky. Eventually, they came to accept
that the signal they were detecting was real. The sky faintly glows
in microwaves.
In the meantime, physicist Robert Dicke (1916–1997) and his
colleagues at Princeton University also predicted a hot early uni-
verse, arriving independently at the
same basic conclusions that Alpher Penzias and Wilson
and Gamow had reached two discovered the cosmic
decades earlier. When Dicke and background radiation.
colleagues heard of the signal that
Penzias and Wilson had found, they interpreted it as the radiation
left behind by the hot early universe. The strength of the detected
signal was consistent with the glow from a blackbody with a tem-
perature of about 3 K, very close to the predicted value. Their
results, published in 1965, reported the discovery of the glow left
behind by the Big Bang. Penzias and Wilson shared the 1979 Nobel
Prize in physics for their remarkable discovery.
It is worth noting that timing can be everything in science.
Alpher, who fi rst predicted the existence of a faint glow from the
Big Bang, had searched unsuccessfully for the signal 10 years
before Penzias and Wilson made their discovery. Unfortunately,
however, the technology of the late 1940s and early 1950s was
simply not up to the task.
This radiation left over from the early universe is called the
cosmic background radiation (CBR). Today, the cosmic background
radiation and the conditions in the early universe are much Hydrogen atoms are much less effective at blocking
better understood than they were in the early 1960s. The radiation than free electrons are; so when recombination
origin of the CBR is illustrated in Figure 18.18. When the occurred, the universe suddenly became transparent to
universe was young, it was hot radiation. Since that time, the radiation left behind from
enough that all of the atoms in The CBR is thermal the Big Bang has been able to travel largely unimpeded
the universe were ions. In our radiation that arose throughout the universe. At
discussion of the structure of when the universe was the time of recombination, Since recombination,
the Sun and stars, we found hot and ionized. when the temperature of the the CBR has traveled
that radiation does not travel universe was a few thousand freely and cooled by a
well through an ionized plasma. Free electrons in a plasma kelvins, the wavelength of factor of 1,000.
interact strongly with the radiation, blocking its progress. this radiation peaked at about
At this time in the early universe, the conditions within the 1 μm, according to Wien’s law (see Chapter 4). As the uni-
universe were much like the conditions within a star: the verse expanded, this radiation was redshifted to longer
universe was an opaque blackbody. and longer wavelengths. Today, the scale of the universe
As the universe expanded, the gas filling it cooled. By has increased a thousandfold since recombination, and
the time the universe was about a thousandth of its cur- the peak wavelength of the cosmic background radiation
rent size, the temperature had dropped to a few thousand has increased by a thousandfold as well, to a value close
kelvins, so protons and electrons were able to combine to to 1 mm. The spectrum of the CBR still has the shape of a
form hydrogen atoms. This event, called the recombination Planck spectrum, but with a characteristic temperature of
of the universe, occurred when the universe was several 2.73 K—only a thousandth what it was at the time of
hundred thousand years old. recombination.
1 In the ionized early universe,

light was trapped by free
electrons. Radiation had a
Planck spectrum.
(a) VISUAL ANALOGY FIGURE 18 .18 The

origin of the cosmic background
radiation. (a) Prior to recombination, the
universe was like a foggy day, except that
the “fog” was a sea of hydrogen atoms.
Radiation interacted strongly with free
electrons and so could not travel far. The
trapped radiation had a Planck spectrum.
(b) When the universe recombined, the
fog cleared and this radiation was free to
travel unimpeded.
KEY
Proton
Electron
Path of photon
(b)
2 When the universe recombined, 3 Recombination

it became transparent, and the was like the fog
Planck radiation went its own way. suddenly clearing.
Cosmic background radiation

Satellite Data Confirmed That the has a Planck spectrum.
0.0020
Cosmic Background Radiation Is Real
Planck spectrum,
Brightness (W/m2/steradian)
The presence of cosmic background radiation with a Planck T = 2.728 K
spectrum is a very strong prediction of the Big Bang theory. 0.0015
COBE measurements of the
Penzias and Wilson had confi rmed that a signal with the
cosmic background radiation
correct strength was there, but they could not say for cer-
tain whether the signal they saw had the spectral shape of
0.0010 The uncertainties in
a Planck spectrum. From the late 1960s to the 1980s, most
the measurements are
experiments at different wavelengths supported these same much less than the
conclusions. Yet it was not until the end of the 1980s that the thickness of the line.
predictions of Big Bang cosmology for the CBR were put to 0.0005
the ultimate test. The year 1989 saw the launch of a satellite
called the Cosmic Background Explorer, or COBE. Instru-
ments carried on COBE were capable of making extremely
precise measurements of the CBR at many wavelengths, 0 1 2 3 4 5
from a few micrometers out to 1 cm. In January 1990, hun- Wavelength (mm)
dreds of astronomers gathered in a large conference room
FIGURE 18 .19 The spectrum of the cosmic background
in Washington DC, at the winter meeting of the American radiation as measured by the Cosmic Background Explorer (COBE)
Astronomical Society, to hear the COBE team present its satellite (red dots). The uncertainty in the measurement at each
fi rst results. Security surrounding the new findings had wavelength is much less than the size of a dot. The line running
been tight, so the atmosphere in the room was electric. The through the data is a Planck spectrum with a temperature of
tension did not last long; presentation of a single viewgraph 2.728 K.
brought the room’s occupants to their feet in a spontane-
ous ovation.
The data shown on that viewgraph are reproduced in structure of the universe itself, but rather is the result of
Figure 18.19. The small dots in the figure are the COBE the motion of Earth with respect to the CBR.
measurements of the CBR at Time and again on our journey we have stressed that
different frequencies. The COBE unambiguously there is no preferred frame of
uncertainty in each measure- showed the CBR to reference. The laws of physics Our motion makes
ment is far less than the size have a Planck are the same in any inertial the CBR slightly
of each dot. The line in the spectrum at 2.73 K. reference frame, so none is hotter in the direction
figure, which runs perfectly better than any other. Yet in a we are moving toward
through the data points, is a Planck spectrum with a tem- certain sense there is a pre- and slightly cooler
perature of 2.73 K. The agreement between theoretical preferred frame of reference at behind us.
diction and observation is truly remarkable. The observed every point in the universe.
spectrum so perfectly matches the one predicted by Big This is the frame of reference that is at rest with respect to
Bang cosmology that there can be no real doubt we are see- the expansion of the universe and in which the CBR is iso-
ing the residual radiation left behind from the primordial tropic, or the same in all directions. The COBE map shows
fi reball of the early universe. that one side of the sky is slightly hotter than the other
because Earth and our Sun are moving at a velocity of 368
km/s in the direction of the constellation of Crater, relative
The CBR Measures Earth’s Motion to this cosmic reference frame. Radiation coming from the
direction in which we are moving is slightly blueshifted
Relative to the Universe Itself (shifted to a higher characteristic temperature) by our motion,
COBE provided us with much more than a measurement of whereas radiation coming from the opposite direction is
the spectrum of the cosmic background radiation. Figure Doppler-shifted toward the red (or cooler temperatures). Our
18.20a shows a map obtained by COBE of the CBR from the motion is due to a combination of factors, including the
entire sky. The different colors in the map correspond to motion of our Sun around the center of the Milky Way and
variations in the temperature of the CBR. The differences the motion of our galaxy relative to the CBR.
are not as extreme as the colors might suggest, however. The If we subtract from the COBE map this asymmetry in the
range in temperature in the map corresponds to a variation CBR caused by the motion of Earth, only slight variations
of only about 0.1 percent in the temperature of the CBR. Most in the CBR remain, as shown in Figure 18.20b. The slight
of this range of temperature is present because one side of variations seen in this map have an amplitude that is only
the sky looks slightly warmer than the opposite side of the about 1/100,000 the brightness of the CBR. This means that
sky. This difference has nothing to do with the large-scale the brighter parts of this image are only about 1.00001 times
(a) universe. These concentrations later gave rise to galaxies and

the rest of the structure that we see in the universe today.
Subsequent observations from Antarctica and from
instruments carried aloft by balloons support the COBE
fi ndings. Beginning in 2001, more precise measurements
of the variations of the CBR have been carried out by a sat-
ellite called the Wilkinson Microwave Anisotropy Probe,
or WMAP. Figure 18.20c shows the ripples measured by
WMAP with much higher resolution than could be detected
by COBE. The much higher-resolution maps obtained by
WMAP have profound implications for our understanding
(b) of the origin of structure in the universe and enable us to
determine several cosmological parameters. For example,
the value of the Hubble constant we use in this book is iden-
tical to the value inferred from the WMAP experiment.
The Big Bang Theory Correctly

Predicts the Abundance of the
Least Massive Elements
(c) The next confrontation between the predictions of the Big
Bang theory and observations of the universe came from a
very different direction. When the universe was only a few
minutes old, its temperature and density were high enough
for nuclear reactions to take place. Just as we can use our
knowledge of nuclear physics to calculate the nuclear reac-
tions occurring in the interiors of stars, we can use this
knowledge to calculate the nuclear reactions that took place
in the early universe. Collisions between protons in the early
universe built up low-mass nuclei, including deuterium
(heavy hydrogen) and isotopes of helium, lithium, beryl-
lium, and boron. The formation of new elements in this early
nuclear brew, called Big Bang nucleosynthesis, determined
the fi nal chemical composition of the matter that emerged
FIGURE 18. 20 (a) The COBE map of the cosmic background from the hot phase of the Big Bang. AstroTour: Big
radiation. The CBR is slightly hotter (by about 0.003 K) in one Bang Nucleosynthesis
direction in the sky than in the other direction. This difference
We have already discussed how differences in the abun-
is due to Earth’s motion relative to the CBR. (b) The COBE map
dances of the products of stellar nucleosynthesis help us
with Earth’s motion removed, showing tiny ripples remaining in
track the chemical evolution of the universe and the his-
the CBR. (c) WMAP has provided the highest resolution yet of
the CBR. The radiation seen in this image was emitted less than tory of star formation. These ideas will play an important
400,000 years after the Big Bang. role in our discussion of the Milky Way in Chapter 20, and
they will come to the fore again in our later discussion of
galaxy formation. Here we focus on the products of Big Bang
nucleosynthesis.
brighter than the fainter parts. The amounts of various elements that formed from Big
These slight variations might COBE found, and Bang nucleosynthesis depended in detail on the tempera-
not seem like much, but they WMAP further ture and density of normal
are actually of crucial impor- revealed, tiny matter in the early universe. Big Bang theory
tance in the history of the uni- variations in the CBR In turn, the density in the predicts a universe
verse. Recall from Chapter 17 resulting from the early universe at appropriate that is 24 percent
that gravity itself can create formation of structure nucleosynthesis temperatures helium by mass, which
a redshift. These tiny fluc- during early times. accounts for the density of observations confirm.
tuations in the cosmic back- normal matter found in the
ground radiation are the result of gravitational redshifts universe today. Figure 18.21 shows the calculated predic-
caused by concentrations of mass that existed in the early tions of element abundances from Big Bang nucleosynthesis,
FIGURE 18 . 2 1 Observed and

1 The amounts of different 2 If the density of normal
elements produced in the Big matter in today’s universe calculated abundances of the
Bang were determined by the lies in this range… products of Big Bang nucleosynthesis,
density of normal matter. plotted against the density of normal
0.25 matter in today’s universe. Big Bang
Observed 4He abundance nucleosynthesis correctly predicts the
mass fraction
0.24
amounts of these isotopes found in the
0.23 4He 3 …then the light elements universe today.
formed would match what
0.22
Deuterium (2H) we find in nature.
4He
10–4
3He
Observed deuterium abundance
Number relative to H
10–5 Observed 3He abundance
4 The amount of normal

10–9 Curves show the matter we actually observe
amount of an isotope (black line) lies within the
7
Li that would have been predicted range.
formed in the Big Bang.
Observed 7Li abundance
10–10
1 2 3 4 5
Density of normal matter in today’s universe (10–28 kg/m3)
plotted as a function of the present-day density of normal Figure 18.21. Once again, the agreement is remarkable. Turn-
(luminous) matter in the universe. The fi rst thing we see ing this around, we can begin with an observation of the
from this figure is that about 24 percent of the mass of the amount of normal matter in and around galaxies and then
normal matter formed in the early universe should have use our understanding of the Big Bang to calculate what the
ended up in the form of the very stable isotope 4He, regard- chemical composition emerging from the Big Bang should
less of exactly what the density of matter in the universe have been. When we do this, the answer we get agrees
was. Indeed, when we look about us in the universe today, remarkably well with the amounts of these elements we
we fi nd that about 24 percent of the mass of normal matter actually find in nature.
in the universe is in the form of 4He, in complete agreement Two other points are worth noting here. The fi rst is that
with the prediction of Big Bang nucleosynthesis. no elements more massive than boron could have been
Unlike helium, the abundances of most of the isotopes formed in the Big Bang. Reactions such as the triple-alpha
formed in the Big Bang depended sensitively on the density process, which forms carbon in the interiors of stars (see
of normal matter in the uni- Chapter 16) simply would not work under the conditions
verse. Beginning with the Abundances of existing in the early universe. That is how we know that
amounts of isotopes such as other low-mass all the more massive elements in the universe, including
deuterium (2H) and 3He found elements are also the atoms making up the bulk of our planet and ourselves,
in the universe (shown as consistent with the must have formed in subsequent generations of stars. The
roughly horizontal bands in Big Bang model. second point is that the agreement between the observed
Figure 18.21), we can ask what density of normal matter in the universe and the abun-
the density of normal matter in today’s universe must be dances of light elements also provides a powerful con-
for these isotopes to have been formed in the Big Bang. This straint on the nature of the dark matter dominating the
prediction is shown as the vertical band in the figure. We mass in the universe. As we will see in the next chapter,
have talked a great deal about the ways astronomers mea- dark matter cannot consist of normal matter made up of
sure the amount of normal matter in the universe. The best neutrons and protons; if it did, the density of neutrons
current measurements give a value of about 3.9 × 10 –28 kilo- and protons in the early universe would have been much
grams per cubic meter (kg/m3) for the average density of higher, and the resulting abundances of light elements in
normal matter in the universe today. This value lies well the universe would have been much different from what
within the range predicted by the observations shown in we actually observe.
Summary 531
see completely different surroundings as we look about.

Seeing the Forest for the Trees This, of course, is what we would expect. But suppose we
were transported suddenly to another galaxy billions of
You never know when you get up in the morning if this light-years away. Would the universe appear to be quite
will be one of those rare days when you happen upon different from the way we see it at home? The cosmologi-
an extraordinary experience that leaves you looking at cal principle tells us it would not. We have learned that
the world in a different way. So it was for Edwin Hubble. the universe is both homogeneous and isotropic, mean-
Hubble, who also was the fi rst to prove the true nature of ing that at the largest scales it appears the very same in
galaxies, was certainly at the right place at the right time every direction and from every location.
in his career. He was doing as scientists do, using the We could possibly take some comfort in knowing that
cutting-edge tools of his day and following his instincts no matter where we might find ourselves in the universe,
about what questions might turn out to have interest- everything would look and work pretty much the same as it
ing answers. His investigation into the basic properties does right here. But that doesn’t mean change isn’t happen-
of the newly recognized class of objects called galaxies ing. It is happening, and on a grand scale. No matter what
took an amazing and unexpected turn. With one graph, our location in space might be, we would find that galaxies
plotting the redshift of galaxies against his estimates of appear to be flying apart as space itself expands.
their distances, he forever changed our view of the uni- At one time, everything we see in this vast universe—
verse. What must it have felt like to be Edwin Hubble, including space itself—was contained in a volume that
looking at his data and realizing for the fi rst time what was as small compared to the nucleus of an atom as you
the implications of those data were? What would it mean are compared to today’s enormous universe. From a single
to be the fi rst human to glimpse the true history and moment 13.7 billion years ago, space has been growing
potential future of the universe? into a colossus so spread out that even light takes nearly
From the early discoveries of Hubble and of those who 30 billion years to cross it.
followed in his footsteps, we have come to marvel at how There is still much more to learn about the origin,
vast the universe truly is. Yet, from our vantage point structure, and destiny of the universe, but we will put
here in our home galaxy, much of what we see is not the the universe itself aside briefly while we take a close look
actual universe as it exists today; instead, what we see at the largest of its individual components. The next step
is how it appeared billions of years ago. How, then, do in our journey brings us to the galaxies themselves, these
we know what the entirety of today’s universe is really immense collections of stars, gas, and dust that were once
like? When we travel from the city to the countryside, we called “island universes.”
Summary
• Our observable universe contains hundreds of billions • We observe thermal radiation at 2.73 K, which is the
of galaxies, each with millions to hundreds of billions cooled remnant of the radiation present in the universe
of stars. when it was much smaller and hotter.
• Observations suggest that our universe is homoge- • The CBR enables us to measure our velocity with respect
neous and isotropic, in agreement with the cosmologi- to the background radiation, and it also shows evidence
cal principle. of the ripples that grew to become large-scale structure
in the universe.
• The expansion of the universe is governed by Hubble’s
law: a galaxy’s recession velocity is proportional to its • Nuclear reactions of normal matter in the hot early uni-
distance. verse produced all the helium we see today, as well as
trace amounts of other light elements.
• Expansion of the universe produces the observed red-
shifts, but it does not affect the local physics or struc-
ture of objects.
• The universe is expanding uniformly from a Big Bang,
which occurred nearly 14 billion years ago.
• Observed redshifts of distant galaxies are a result of the your instructor assigns questions in SmartWork, log in at
increasing scale factor of the universe. smartwork.wwnorton.com.
Student Questions *17. As the universe expands from the Big Bang, we know
that galaxies are not actually flying apart from one
another. What is really happening?
18. The scale factor can be used as a “cosmic ruler” in
1. Explain what astronomers mean by the term universe.
describing the expansion of the universe. Explain the
2. What was the subject of the Great Debate, and why was scale factor.
it important to our understanding of the scale of the
19. Knowing that you are studying astronomy, a curious
universe?
friend asks where the center of the universe is located.
3. How did Cepheid variable stars finally settle the Great You smile and answer, “Right here and everywhere.”
Debate? Explain in detail why you would give this answer.
4. Describe the cosmological principle in your own words. *20. The general relationship between radial velocity (v r)
and redshift (z) is v r = cz. This simple relationship fails,
5. We see individual stars and galaxies. What, then,
however, for very distant galaxies with large redshifts.
do astronomers mean when they say the universe is
Explain why.
“homogeneous”?
21. What is the origin of the cosmic background radiation
*6. The universe could be homogeneous without being iso-
(CBR)?
tropic. Explain what is meant by isotropic.
*22. Why is it significant that the CBR displays a Planck
7. Imagine that you are standing in the middle of a dense
spectrum?
fog.
a. Would you describe your environment as isotropic? 23. As the sensitivity of our instrumentation increases, we
b. Would you describe it as homogeneous? are able to look ever farther into space and, therefore,
ever further back in time. When we reach the era of
Explain your answers.
recombination, however, we run into a wall and can see
8. Early in the 20th century, astronomers discovered that no further back in time. Explain why.
most galaxies are moving away from our own (that is,
24. What is the significance of the tiny brightness variations
they are redshifted.) What was the significance of this
that are observed in the CBR?
discovery?
25. What important characteristics of the early universe are
9. Edwin Hubble later made an even more important dis-
revealed by today’s observed abundances of various iso-
covery: that the speed with which galaxies are receding
topes, such as 2H and 3He?
is proportional to their distance. Why was this among
the more important scientific discoveries of the 20th
century?
10. If you lived in one of the remote galaxies pictured in
Figure 18.1, would you observe distant galaxies reced- 26. Hubble time (1/H0) represents the age of a universe that
ing from your own galaxy, or would they appear to be has been expanding at a constant rate since the Big Bang.
approaching? Explain your answer. Assuming an H0 value of 22 km/s/Mly and a constant
rate of expansion, calculate the age of the universe in
11. Why is the Milky Way Galaxy not expanding together
years. (Note: 1 year = 3.16 × 107 seconds, and 1 light-
with the rest of the universe?
year = 9.46 × 1012 km.)
12. Why can we not use the measured radial velocities of
27. Throughout the latter half of the 20th century, estimates
nearby galaxies, such as the Andromeda Galaxy, to
of H0 ranged from 18 to 35 km/s/Mly. Calculate the age
evaluate the Hubble constant (H0)?
of the universe in years for each of these estimated val-
13. What are peculiar velocities, and how do they affect our ues of H0.
ability to measure H0?
28. Assume that the most distant observable galaxies have
14. Explain what astronomers mean by a distance ladder. a redshift z = 10. (None have yet been found at this
distance.)
15. As astronomers extend their distance ladder beyond
a. What was the scale factor (RU) of the universe at the
100 Mly, they change their measuring standard from
time light was leaving those hypothetical distant
Cepheid variable stars to Type I supernovae. Why is this
galaxies?
necessary?
b. How did the volume of the universe at that time com-
16. What is meant by Hubble time? pare with today’s volume?
29. One of the more distant galaxies has a redshift z = 5.82 b. Which of the two calculations would you consider to
and a recession velocity of 287,000 km/s (about 96 per- be more trustworthy? Why?
cent of the speed of light). c. Estimate the peculiar velocity of the closer galaxy.
a. If H0 = 22 km/s/Mly and if Hubble’s law remains valid d. If the more distant galaxy had this same peculiar
out to such a large distance, then how far away is this velocity, how would your calculated value of the
galaxy? Hubble constant change?
b. Assuming a Hubble time of 13.7 billion years, how old
35. The temperature of the CBR is 2.73 K. What is the peak
was the universe at the look-back time of this galaxy?
wavelength of its Planck spectrum expressed both in
c. What was the scale factor of the universe at that
micrometers and in millimeters?
time?
36. COBE observations show that our Solar System is mov-
30. The spectrum of a distant galaxy shows the Hα line of
ing in the direction of the constellation Crater at a speed
hydrogen (λrest = 656.28 nm) at a wavelength of 750 nm.
of 368 km/s relative to the cosmic reference frame. What
Assume that H0 = 22 km/s/Mly.
is the blueshift (negative value of z) associated with this
a. What is the redshift (z) of this galaxy?
motion?
b. What is its recession velocity in kilometers per
second? *37. If new observations suggested that the mass fraction of
c. What is the distance of the galaxy in mega-light- helium remains at 24 percent but the mass fraction of
years? deuterium (D) is really 10 –6, how would our estimates of
the density of normal matter in the universe be affected?
31. The rest wavelength of the Ly α line of hydrogen is in
Refer to Figure 18.21.
the extreme ultraviolet region of the spectrum at 121.6
nm. (“Ly” stands for Theodore Lyman [1874–1954], the 38. The average density of normal matter in the universe is
American physicist who discovered this far-ultraviolet 4 × 10 –28 kg/m3. The mass of a hydrogen atom is 1.66 ×
series of lines in the hydrogen spectrum.) 10 –27 kilograms (kg). On average, how many hydrogen
a. What would be the wavelength of this line in the atoms are there in each cubic meter in the universe?
spectrum of a galaxy with a redshift z = 6.4?
39. To get a feeling for the emptiness of the universe, compare
b. In what region of the spectrum would this redshifted
its density (4 × 10 –28 kg/m3) with that of Earth’s atmo-
line be located?
sphere at sea level (1.2 kg/m3). How much denser is our
32. One of the most distant known galaxies has a redshift atmosphere? Write this ratio using standard notation.
z = 7.6.
*40. Assume that the most distant galaxies have a redshift z
a. What would be the observed wavelength of the Hα
= 10. The average density of normal matter in the uni-
line (λrest = 656.28 nm)?
verse today is 4 × 10 –28 kg/m3. What was its density when
b. In what region of the spectrum would this line be
light was leaving those distant galaxies? (Hint: Keep in
located?
mind that volume is proportional to the cube of the scale
33. The emission of the Ly α line of hydrogen (121.6 nm) from factor.)
a distant galaxy is observed at the same wavelength as
the rest wavelength of the Hα line of hydrogen (656.28
nm). Calculate the redshift of this galaxy.
34. Suppose we observe two galaxies: one at a distance 35 Mly vides a Study Plan for each chapter of 21st Century
with a radial velocity of 580 km/s, and another at a dis- Astronomy. Study Plans include animations, read-
tance of 1,100 Mly with a radial velocity of 25,400 km/s. ing outlines, vocabulary flashcards, and multiple-choice
a. Calculate the Hubble constant (H0) for each of these quizzes, plus links to premium content in SmartWork and
two observations. the ebook. Visit www.wwnorton.com/studyspace.
It may not be amiss to point out some other very remarkable
Nebulae which cannot well be less, but are probably much
larger than our own system; and being also extended,
the inhabitants of the planets that attend the stars which
compose them must likewise perceive the same phenomena.
For which reason they may also be called milky ways.
SIR WILLIAM HERSCHEL (1738 –1822)
The large, barred spiral galaxy NGC 1300 is 70 million light-years away in the constellation Eridanus.
CHAP TER 19
Galaxies
K E Y CONCEP T S
19.1 Galaxies Come in
Stars are not spread uniformly through space but are
Many Flavors instead grouped into what Immanuel Kant referred to
as “island universes,” which today we call “galaxies.” As
It is still less than a century since we came to realize that we look beyond stars to the galaxies they belong to, we
the universe is filled with “spiral nebulae,” objects that will find that
closely resemble our own galaxy, the Milky Way. But just
as stars come in many flavors based on their mass or on • Galaxies are classified into different types according to
their stage of evolution, galaxies also come in many forms. their shapes, which are determined by the orbits of the
After realizing that galaxies are huge collections of stars stars they contain.
and dust, astronomers found ways to classify these galax- • The arms of spiral galaxies, which form whenever the
ies into various types and make sense of the differences disk of a spiral galaxy is disturbed, are sites of star
among them. One of their goals is to understand the origin formation.
of each flavor of galaxy, just as we interpret the origins of
the different types of stars. • Stars and gas account for only a small fraction of the
mass of a galaxy.
• Galaxies are composed mostly of dark matter, which
Galaxies Are Classified According interacts through gravity but does not emit or absorb
to Their Appearance noticeable amounts of electromagnetic radiation.
• Most—perhaps all—large galaxies have supermassive
Imagine what you would see if you took a handful of coins
black holes at their centers.
and threw them into the air as shown in Figure 19.1. You
know that all of these objects are very much the same: dimes, • When these supermassive black holes are “fed,”
pennies, nickels, quarters—all flat and circular. When you possibly during encounters between galaxies, they
look at the objects falling through the air, however, they do may blaze forth with the light of thousands of normal
not appear to be all the same. Some coins you see face on, galaxies.
and they appear circular. Some coins are instead seen edge
• The light emerging from active galactic nuclei comes
on and appear as nothing but thin lines. Most coins are seen
from a region no larger than the orbit of Neptune in
from an angle between these two extremes and appear with
our own Solar System.
various degrees of “ellipticity,” or flattening.
In principle, we can learn a lot about the properties of
coins by looking at a picture like Figure 19.1. If we began
535
536 Chapter 19 Galaxies
by assuming that coins have a particular three-dimensional

shape, we could predict what we should see in Figure 19.1.
We could then compare our prediction with what is actu-
ally observed. In this example we would probably have lit-
tle trouble convincing ourselves that coins must be disks.
As astronomers, we play exactly this game in our efforts to
discover the true three-dimensional shapes of galaxies. Fig-
ure 19.2 shows a set of galaxies seen from various viewing
angles, from face on to edge on. We can infer from images
of the sky that, just like the coins in Figure 19.1, galaxies
often have a disklike shape and are randomly oriented on
the sky.
A quick look at a group of galaxies imaged by the Hubble
Space Telescope (Figure 19.3), shows that galaxies come in a
wide range of apparent sizes and shapes. The fi rst advance
in understanding galaxies came from sorting these different
shapes into categories. The classifications we use today date
back to the 1930s, when Edwin Hubble devised a scheme
much like that shown in Figure 19.4. Hubble grouped all

19.1 A handful of coins thrown
in the air provides a helpful
analogy for the difficulties in
identifying the shapes of certain
types of galaxies. We see some
face on, some edge on, and
most somewhere in between.

19. 2 Disk-shaped galaxies
seen from various perspectives
or angles. The variety of
angles we see for galaxies
corresponds to the range of
perspectives for the coins in
Figure 19.1.
and placed them along the right-hand tine of the tuning fork,
as shown in Figure 19.4. Spirals that lack a barlike bulge lie
along the left-hand tine. Notice that Hubble positioned the
spiral galaxies vertically along the tines of the fork accord-
ing to the prominence of the central bulge and how tightly
the spiral arms are wound. For example, Sa and SBa galax-
ies have the largest bulges and display tightly wound and
smooth spiral arms. Sc and SBc galaxies have small cen-
tral bulges and more loosely woven spiral arms, often very
knotty in appearance.
Galaxies that fall into none of these classes are called
irregular galaxies (“Irr”). As their name implies, irregular
galaxies are often without symmetry in shape or structure,
and they do not fit neatly on Hubble’s tuning fork.
Originally, Hubble thought that his tuning fork diagram
might do for galaxies what the H-R diagram had done for
stars. This hope turned out to be incorrect, but his classi-
fication scheme did succeed in bringing order to the study
of these objects. Table 19.1 summarizes the criteria that
Hubble used to classify galaxies.
FIGURE 19. 3 This Hubble Space Telescope image of a small

group of galaxies called Hickson Compact Group 87 illustrates
Stellar Motions Give
something of the range of shapes and sizes found among Galaxies Their Shapes
galaxies.
A galaxy is not a solid object like a coin, but a collection of
stars, gas, and dust orbiting under the influence of its over-
all gravitational field. The
galaxies according to appearance and positioned them on a shapes of elliptical galaxies The collective orbits
diagram that resembles the tuning fork used in the tuning are determined from the of all its stars
of a musical instrument. almost random orbits of their give an elliptical
On the bottom (or “handle”) of this tuning fork diagram stars. For example, in any galaxy its shape.
sit oval-shaped objects called elliptical galaxies (labeled region in an elliptical galaxy,
“E” on the diagram); these galaxies can have either round or some stars are falling in while others are climbing out—in
elliptical shapes, and they show little evidence of the disk- fact, there are stars moving in all possible directions. Unlike
like structure that we see in other types of galaxies. planets, which move on simple elliptical orbits about the
On the two “tines” of the fork are the spiral galaxies, Sun, stars in an elliptical galaxy follow orbits with a wide
designated with an initial “S.” The defi ning feature of a spi- range of different shapes, as shown in Figure 19.5. These
ral galaxy is a flattened, rotating disk. Spiral galaxies also orbits are more complex than the orbits of planets because
contain the spiral arms that give these galaxies their name. the gravitational field within an elliptical galaxy does not
In addition to disks and arms, spiral galaxies have central come from a single central object.
bulges, which look like elliptical galaxies. Taken together, all of these stellar orbits are what give
Hubble recognized that the distinction between spiral an elliptical galaxy its shape. Orbital speeds are also a fac-
and elliptical galaxies is not always clear-cut. Some galax- tor. The faster the stars are moving, the more spread out
ies seem to be a cross between the two types, having stellar the galaxy is. (After all, if the stars were not moving at all,
disks but no spiral arms. Hubble called these intermedi- they would all clump together at the center of the galaxy.) If
ate types of galaxies S0 galaxies and placed them near the the stars in an elliptical galaxy are moving in truly random
junction of his tuning fork. Today, the distinction between directions, the galaxy will have a spherical shape. However,
elliptical and S0 galaxies is even more blurred. Modern tele- if stars tend to move faster in one direction than in others,
scopic observations have revealed that many, if not most, the galaxy will be more spread out in that direction, giving
elliptical galaxies contain small rotating disks at their cen- it an elongated shape. These differences in stellar orbits are
ters. Elliptical and S0 galaxies share another similarity: as responsible for the variety of shapes that Hubble noted—
of now, both have stopped producing any new stars. namely, that some elliptical galaxies (those on the bottom
Hubble noticed that the bulges of roughly half of the spi- of the tuning fork handle—see Figure 19.4) are round, while
ral galaxies are bar-shaped (see the opening photograph of others (those on the top of the handle) are elongated. One
this chapter). He called these galaxies barred spirals (“SB”) difficult problem faced by astronomers studying elliptical
The Hubble tuning fork is a way

of classifying galaxies but is not a
physical or evolutionary sequence.
Irregular
galaxies Irr
Sd SBd Irregular galaxy
Sc SBc
Arms become more

Sb open and knottier. SBb
Bulge becomes
Unbarred spiral galaxy less prominent.
Sa SBa
Unbarred Barred
spirals spirals
S0 SB0
Barred spiral galaxy
E5
Elliptical Galaxies become

E3 more elongated.
galaxies
E0
Elliptical galaxy
FIGURE 19. 4 Tuning fork diagram showing Edwin Hubble’s scheme for classifying galaxies according
to their appearance. Elliptical galaxies form the “handle” of this tuning fork. Unbarred and barred spiral
galaxies lie along the left and right tines of the fork, respectively. S0 galaxies lie along the bottom left
and right tines. Irregular galaxies are not placed on the tuning fork.
TA B L E 1 9 .1
The Hubble Sequence of Galaxies

A Morphological Classification Scheme Based on the Gross Properties of Galaxies
Sequence
Category/Criteria Abbreviation Range of Features
Ellipticals E0 Rounder
Mostly bulge E1
Old, red stellar population E2
Smooth-appearing E3
E4
E5
E6
E7 More elongated
S0 (unbarred/barred) S0/SB0 Smooth disk and bulge

Bulge and disk with no arms
Bulge and disk contain mostly old, red stars
Spirals (unbarred/barred) Sa/SBa More bulge Tightly wound arms Smooth arms
Bulge and disk with arms Sb/SBb
Bulge has old, red stars Sc/SBc
Disk has both old, red stars and young, Sd/SBd Little bulge Open arms Knotty arms
blue stars
Spirals (S) have roundish bulges
Barred spirals (SB) have elongated or
barred bulges
Irregulars Irr
No arms
No bulge
Some old stars, but mostly young stars,
giving a knotty appearance
galaxies, however, is that their appearance in the sky does in the same direction about a concentration of mass at the
not necessarily tell us their true shape. For example, a gal- center of the galaxy. But the stellar orbits in a spiral galaxy’s
axy might actually be shaped like an American football, central bulge are quite different from those in the galaxy’s
but if we happen to see it end on, it will look instead round disk. As with elliptical galaxies, the gravitational field
like a baseball. within the bulge does not come from a single object, and the
In contrast, the orbits of stars in the disks of spiral gal- stars therefore follow random orbits. The bulges of spiral
axies are quite different from those of stars in elliptical gal- galaxies are thus roughly spherical in shape.
axies. The components of a
spiral galaxy are shown in Spiral galaxies have a
Figure 19.6. The defining fea- rotating disk and a
ture of a spiral galaxy is that central bulge.
Other Differences among Galaxies
it has a flattened, rotating disk. In addition to the differences in their stellar orbits, there is
Like the planets of our Solar System, most of the stars in the another important distinction between spiral and elliptical
disk of a spiral galaxy follow nearly circular orbits and travel galaxies. Most spiral galaxies contain large amounts of dust
FIGURE 19. 5 Elliptical galaxies

take their shape from the orbits
Stars in elliptical galaxies
move in all directions… of the stars they contain. The
colored lines superimposed on
the galaxy represent the complex,
irregular orbits of its stars.
…on complex,
irregular orbits.
and cold, dense molecular gas concentrated in the mid- The colors of spiral and elliptical galaxies tell us a great
planes of their disks. Just as the dust in the disk of our own deal about their star formation histories. As we learned in
galaxy can be seen on a clear Chapter 14, stars form from
summer night as a dark band Ellipticals contain dense clouds of cold molecu- Stars are still forming
slicing the Milky Way in two mostly hot gas, lar gas. Because the gas we see in spiral galaxies but
(see Figure 15.1), the dust in and spirals contain in elliptical galaxies is very not in elliptical
an edge-on spiral galaxy mostly cold gas. hot, we would not expect star galaxies.
appears as a dark, obscuring formation to be taking place
band running down the midplane of the disk (Figure 19.7). today. The red colors of elliptical and S0 galaxies tell us
The cold molecular gas that accompanies the dust can also that we are indeed looking at old stellar populations and
be seen in radio observations of spiral galaxies. In contrast, that there has been little or no star formation for quite some
elliptical galaxies contain large amounts of very hot gas that time. The stars in these galaxies are an older population of
we see primarily by observing the X-rays it emits. lower-mass stars. The blue colors of the disks of spiral gal-
The difference in shape between elliptical and spiral axies, on the other hand, tell us that we are looking at
galaxies offers some insight into why the gas in ellipticals regions of ongoing star formation where massive, young hot
is hot, while spirals contain a large amount of cold, dense stars are being born in the cold molecular clouds contained
gas. Just as gas settles into a disk around a forming star, so, within the disk. Even though most of the stars in a spiral
too, does cold gas settle into the disk of a spiral galaxy as a disk are old, the massive, young stars are so luminous that
result of conservation of angular momentum. In contrast, their blue light dominates what we see. When it comes to
the only place in an elliptical galaxy where cold gas could star formation, most irregular galaxies are like spiral galax-
collect is at the center. However, the density of stars in ellip- ies. Some irregular galaxies are currently forming stars at
tical galaxies is so high that Type I supernovae continually prodigious rates, given their relatively small sizes.
reheat this gas, preventing most of it from cooling off and There is not a straightforward relationship between
forming cold clouds. luminosity and size among the different types of galaxies.
FIGURE 19.6 The components

1 Stars in the disks of spiral of a barred spiral galaxy. The
Disk Barlike galaxies move together… orbits of stars in the rotating
bulge
disk and the ellipse-like bulge are
indicated.
Spiral arm
2 …on nearly
circular orbits.
3 Bulges of spiral galaxies

are like elliptical galaxies:
stars move in irregular orbits
in all directions.
FIGURE 19.7 The dust in the

plane of the nearly edge-on spiral
galaxy M104 is seen as a dark,
obscuring band in the midplane of
the galaxy. Note the bright halo
made up of stars and globular
clusters. Compare this image with
Figure 15.1, which shows the dust
in the plane of the Milky Way.
Galaxies range in luminosity size difference between elliptical and spiral galaxies. About
from about a million up to a All types of galaxies half of both types of galaxies fall within a similar range of
million million solar lumi- come in a wide sizes. Although it is true that the most luminous elliptical
nosities (10 -10 L ⊙) and in
6 12 range of sizes. galaxies are more luminous than the most luminous spiral
size from about a thousand up galaxies, there is considerable overlap in the range of lumi-
to hundreds of thousands of light-years. There is no strict nosities among all Hubble types (Figure 19.8).
50% of galaxies fall Earlier in our journey we found that mass is the single
(a)
within these limits. most important parameter in determining the properties and
evolution of a star. In contrast, differences in mass and size
100 do not lead to such obvious differences in galaxies. Only
Radius (thousands of light-years)
subtle differences in color and concentration exist between

large and small galaxies, making it difficult to distinguish
which are large and which are small. Even when a smaller,
nearby spiral galaxy is seen next to a larger, distant spiral
30 (Figure 19.9), it can be hard to tell which is which. If you
were to look at a photo of a nearby child appearing to stand
next to a more distant adult, you could almost certainly
tell which one was the child. There would be obvious dif-
50% of galaxies may ferences in appearance. But that is not the case with spiral
10 be larger or smaller. galaxies. Still, astronomers prefer to call galaxies that have
relatively low luminosity (less than 1 billion L ⊙) dwarf gal-
axies and those that are more luminous than this giant gal-
(b) axies. Only elliptical and irregular galaxies come in both
types. In fact, among spiral and S0 galaxies, we fi nd only
giants. It is relatively easy to tell the difference between
1011
a dwarf elliptical galaxy and a giant elliptical galaxy (as
shown in Figure 19.10). Giant elliptical galaxies have a much
higher density of stars than dwarf ellipticals have.
Luminosity (L£)
1010
19.2 Stars Form in the Spiral

Spiral and elliptical galaxies
Arms of a Galaxy’s Disk
span similar ranges of luminosity.
109
Spiral galaxies take their name from the spiral arms
they contain, but what is a spiral arm? From pictures
of spiral galaxies outside of our own spiral Milky Way
Elliptical S0 Sa Sb Sc Sd Galaxy, we might have guessed that stars in the disk of
galaxies
Spiral galaxies a spiral galaxy are concentrated in the spiral arms. This
FIGURE 19.8 Galaxies span an enormous range in both size (a) turns out not to be the case. Figure 19.11 shows images
and luminosity (b). Note the great overlap in luminosity and size of the Andromeda Galaxy taken in ultraviolet and visible
among galaxies of different Hubble types. light. Notice that whereas the spiral arms are relatively
(a) (b)
FIGURE 19.9 The mass or
size of a spiral galaxy does not
determine its appearance. Even
though these galaxies appear to
be similar in size and luminosity,
the larger galaxy (a) is 4 times
more distant and 10 times more
luminous than the smaller galaxy
(b).
(a) (b)
FIGURE 19.10 Dwarf elliptical galaxies (a) differ in appearance from giant elliptical galaxies (b).
(a) (b)
FIGURE 19.11 The Andromeda
Galaxy in ultraviolet (a) and visible
(b) light. Note that the spiral arms
are most prominent in ultraviolet
light, which is dominated by young
hot stars, and in emission from
interstellar clouds that are ionized by
the radiation from young hot stars.
The spiral arms are less prominent
in visible light.
prominent in the UV image (Figure 19.11a), they are less quite smoothly as it decreases outward from the center
prominent when viewed in visible light (Figure 19.11b). of the disk to the edge of the galaxy.
If we carefully trace the actual numbers of stars rather Spiral arms look so prominent when viewed in blue or
than just their brightness, we fi nd that although stars are UV light because that is where we fi nd significant concen-
slightly concentrated in spiral arms, this concentration trations of young, massive, luminous stars. In other words,
is not strong—certainly not strong enough to account for what is strongly concentrated in the arms of spiral galaxies
the prominence of the spiral arms we see. In fact, the con- is ongoing star formation. H II regions, molecular clouds,
centration of stars in the disks of spiral galaxies varies associations of O and B stars, and other structures that we
FIGURE 19.12 Two images

(a) (b)
Spiral arm of a face-on spiral galaxy
showing the spiral arms. (a)
Dust lane This visible-light image also
shows dust absorption. (b) This
image of 21-cm emission shows
the distribution of neutral
interstellar hydrogen, CO
emission from cold molecular
clouds, and Hα emission from
ionized gas.
Spiral arms are traced by

concentrations of gas and …and emission from
dust, seen here as dark ionized, neutral, and
absorbing lanes… molecular gas.
have learned to associate with star formation are all found As Figure 19.13 shows, the way disk galaxies rotate
predominantly in the spiral arms of galaxies. implies that any disturbances, or “kicks,” to a galaxy’s disk
We can say a lot about what spiral arms must be like just will naturally lead to spiral structure in that disk. A spiral
by applying what we already know about star formation. galaxy can be kicked in several ways. One way is by gravi-
Stars form when dense interstellar clouds become so mas- tational interactions with other galaxies. (We will see much
sive and concentrated that they begin to collapse under the more of these interactions later in this chapter.) However,
force of their own gravity. If stars form in spiral arms, then just giving a galaxy disk a single kick (whether through
spiral arms must be places where clouds of interstellar gas gravitational interactions or star formation) will not produce
pile up and are compressed. Such is indeed the case. There a stable spiral-arm pattern. Spiral arms produced from a
are many ways to trace the presence of gas in the spiral arms “one-shot” kick will wind
of galaxies. Pictures of face-on themselves up completely in Gravitational
spiral galaxies, such as the Gas, dust, and two or three rotations of the interactions and
one featured in Figure 19.12a, young stars are disk and then disappear. In star formation are
show dark lanes where clouds concentrated in addition, some types of kicks processes that
of dust block starlight. These spiral arms. are repetitive, so they are kick disks.
lanes provide one of the best capable of sustaining spiral
tracers of spiral arms. Spiral arms also show up in other structure indefi nitely. Some kicks come from within a gal-
tracers of concentrations of gas, such as 21-centimeter (cm) axy itself. When the bulge in the center of a spiral galaxy
radiation from neutral hydrogen or radio emission from is not spherically symmetric (as indeed seems to be the case
carbon monoxide (Figure 19.12b). for most spiral galaxies), then the bulge produces a gravita-
Spiral arms are concentrations of gas where stars form, tional disturbance in the disk. As the disk rotates through
but why do spiral arms exist at all? Part of the answer is that this disturbance, it is subjected to repeated kicks that, in
any disturbance in the disk of a spiral galaxy will naturally turn, trigger star formation and the formation of spiral
be made into a spiral pattern by the disk’s rotation. Material structure.
that is closer to the center takes less time to complete a revo- Spiral structure can also be created by the process
lution around the galaxy than does material farther out in of star formation itself. Regions of star formation dump
the galaxy. Figure 19.13 illustrates the point. We begin with considerable energy into their surroundings in the form
a single linear arm through the center of a model galaxy and of UV radiation, stellar winds, and supernova explosions.
then watch what happens as the model galaxy rotates. In the All of this energy drives up the pressure in the region,
time it takes for objects in the compressing clouds of gas and triggering more star for-
inner part of the galaxy to Rotation in a disk mation. Many massive stars typically form in the same
complete several rotations, galaxy naturally region at about the same time; their combined mass out-
objects in the outer parts of produces spiral flows and supernova explosions occur one after another
the galaxy may not have com- structure. in the same region of space over the course of only a few
pleted even a single revolu- million years. The result can be large, expanding bub-
tion. In the process, the originally straight arms are slowly bles of hot gas that sweep out cavities in the interstellar
made into the spiral structure shown. medium and concentrate the swept-up gas into dense,
star-forming clouds, much like the snow that piles up in

t=0
1 Rotation in a spiral galaxy front of a snowplow. In this way, star formation can actu-
takes any structure—even ally propagate through the disk of a galaxy. The resulting
a straight line—…
strings of star-forming regions can then be swept into spi-
ral structures by rotation.
Many galaxies show clear evidence of a relationship
between the shapes of their bulges and the structure of their
spiral arms. Barred spirals, for example, have a characteristic
two-armed spiral pattern that is connected to the elongated
bulge, as seen in the opening photograph for this chapter.
Even the bulges of galaxies that are not obviously barred
may be nonspherical enough to contribute to the formation
t=1
2 …and wraps it into of two-armed spiral structure. Smaller galaxies in orbit about
a trailing spiral. larger galaxies can also give rise to a periodic gravitational
kick, triggering the same sort of two-armed structure.
Regular disturbances in the disks of spiral galaxies are
called spiral density waves. We call them “density” waves
because they are regions of
greater mass density and Regular disturbances
increased pressure in the gal- lead to two-armed
axy’s interstellar medium. spirals.
These waves move around a
disk in the pattern of a two-armed spiral. Spiral density
t=2
waves act like the spiral-shaped blade in a blender slicing
through the liquid. Models of how spiral density waves form
tell us that this spiral-shaped, two-armed wave pattern does
not necessarily rotate at the same rate as the rest of the gal-
axy. Consequently, as material in the disk orbits about, it
passes through the spiral density waves.
A spiral density wave has very little effect on the motions
of stars as they pass through it, but it does compress the gas
that flows through it. As an
analogy for this process, con- Spiral density waves
sider what happens when you compress gas,
t=3 turn on the tap in your kitchen triggering star
3 The longer
we wait… sink (Figure 19.14). The water formation.
hits the bottom of the sink and
spreads out in a thin, rapidly moving layer. A few inches
out, depending on the rate at which water is flowing, there
is a sudden increase in the depth of the water called a “hydro-
static jump.” Spiral arms in galaxies work in much the same
way. Gas flows into the spiral density wave and piles up like
4 …the tighter the water in a hydrostatic jump.
spiral becomes. Stars form in the resulting
compressed gas. Massive
t=4 t=5 t = 20 stars have such short lives
(typically 10 million years or
so) that they never get the
chance to drift far from the
spiral arms where they were
born, so that is where we see
them. Less massive stars, on
the other hand, have plenty
of time to move away from
FIGURE 19.13 The differential rotation of a spiral galaxy will naturally take even an originally their places of birth, forming
linear structure and wrap it into a progressively tighter spiral as time (t) goes by. a smooth underlying disk.
estimate based on the luminosity of starlight from our gal-

axy. Yet the planets would continue on their orbits, moving
under the influence of its gravity, showing that gravitational
attraction still exists, even if there is no light coming from
the center of the Solar System.
Fortunately, we have a method Astronomers use
for determining mass that Kepler’s laws to
does not involve luminosity. measure galaxy mass.
At every point in this book we
have relied on measuring gravity’s effect on an object’s
motion to determine its mass. We can use this method for
galaxies as well. The disks of spiral galaxies are rotating,
which means that the stars in those disks are following
orbits that are much like the Keplerian orbits of planets
around their parent stars and binary stars around each
other. To measure the mass of a spiral galaxy, all we need
to do is apply Kepler’s laws, just as we did for those other
systems.
If we are to apply Kepler’s laws, we should have some
FIGURE 19.14 Water from a tap flows in a thin layer along the idea of how the mass is distributed in a galaxy. We might
bottom of a kitchen sink. The sudden increase in the depth of the begin our study of the rotation of spiral galaxies with the
water is called a hydrostatic jump. hypothesis that the mass in a galaxy is distributed in the
same way that its light is distributed. That is, we could begin
by assuming that the luminous mass in these galaxies is all
the mass there is. On the basis of this hypothesis, we would
19.3 Galaxies Are Mostly make a prediction. The light of all galaxies, including spiral
galaxies, is highly concentrated toward their centers. We
Dark Matter could infer from this fact—and our initial assumption that
luminous matter is all there is—that all the mass of a spiral
We have seen the importance of stellar motions in defi ning galaxy is contained in its central concentration of stars, gas,
the shape of a galaxy, and we now turn to another important and dust. We might then apply our knowledge of the orbital
property of a galaxy: its mass. As we found earlier, mass velocities of the planets in our Solar System to predict that
plays the dominant role in determining the properties of we should see fast orbital velocities near the center of the
stars. However, mass does not play such a clear-cut role for spiral galaxy and slower orbital velocities farther out.
galaxies. Even so, efforts to measure the masses of galaxies As Figure 19.15 shows, orbital velocities of material in
during the last decades of the 20th century led to some of the disk of a spiral galaxy should change with distance from
the most remarkable and surprising findings in the history the center of the spiral, assuming that mass is distributed in
of astronomy. To understand these results, we fi rst need to the same way as starlight. To test this prediction, we use the
ask how astronomers go about measuring the mass of a gal- Doppler effect to measure orbital motions. There are several
axy. One way is to look at the amount of light it gives off. ways to do this. We can measure the velocities of stars from
We can use the spectrum of starlight from a galaxy to deter- observations of absorption lines in their spectra. In addi-
mine the types of stars the galaxy contains. We then use our tion, astronomers determine the velocities of interstellar gas
knowledge of stellar evolution to turn the luminosity of the using emission lines such as those produced by Hα emis-
galaxy into an estimate of the total mass in stars. Finally, sion or 21-cm emission from neutral hydrogen. Once we find
our knowledge of the physics of radiation from interstellar the velocities, we can create a graph that shows how orbital
gas at X-ray, infrared, and radio wavelengths enables us to velocity in a galaxy varies with distance from the galaxy’s
estimate the mass of these other components. Together, the center. This kind of graph is called a rotation curve.
stars, gas, and dust in a galaxy are called luminous matter Astronomers had to wait until the mid-1970s before tele-
(or simply normal matter) because this matter emits elec- scope instrumentation provided reliable measurements of
tromagnetic radiation. orbital velocities of stars and
However, looking at the light from a galaxy does not interstellar gas outside the Rotation curves of
allow us to determine a galaxy’s total mass. Imagine, for inner, bright regions of galax- spiral galaxies are
example, if we were to replace the Sun with a black hole of ies. After making these obser- remarkably flat.
the same mass. We would see no light coming from this vations, astronomers used the
black hole, and thus its mass would not be included in our data to create rotation curves that showed that, contrary to
19.3 Galaxies Are Mostly Dark Matter 547
(b) FIGURE 19.15 (a) The profile

1 If the mass in a spiral galaxy
were distributed like the light, of visible light in a typical spiral
Predicted density
the density of matter in the galaxy. (b) The mass density
galaxy would look like this… of stars and gas located at
a given distance from the
galaxy’s center. If stars and gas
accounted for all of the mass
of the galaxy, then the galaxy’s
rotation curve would be as
(c)
2 …and we would predict shown in (c). However, galaxies
Predicted velocity
rotation velocities like this. have observed rotation curves
more like the curve shown in
(d).
(a)
(d)
Observed velocity
Light
3 But actual rotation

velocities look like this.
Distance from galaxy center
Distance from galaxy center
prediction, the rotation velocities of spiral galaxies remain Had the dark matter been distributed just like the stars, we
about the same out to the most distant measured parts of the might not have realized that galaxies have these two major
galaxies. As shown in Figure 19.15, the rotation curves of components. In their centers, galaxies have a higher con-
spiral galaxies appear level, or “flat,” in their outer parts, centration of luminous matter than dark matter. The rota-
rather than sloping downward as predicted. (These flat curves tion curves of the inner parts
are quite logically referred to as flat rotation curves.) Obser- of spiral galaxies match fairly Most of the mass
vations of 21-cm radiation from neutral hydrogen show that well what would be predicted comprising
the rotation curves remain flat even well outside the extent by their luminous matter, spiral galaxies
of the visible disks. These discoveries came as a shock. They indicating that normal lumi- is dark matter.
meant that the long-held idea that mass in galaxies is distrib- nous matter accounts for most
uted in the same way as their visible light was wrong. of the matter in the inner parts of spiral galaxies. Within
The rotation curve of a spiral galaxy enables us to the entire visual image of a galaxy, the mix of dark and
directly determine how the mass in that galaxy is distrib- luminous matter is about half and half. However, rotation
uted. We can apply Kepler’s laws to these rotation curves curves measured with 21-cm radiation from neutral hydro-
and ask, “How much mass must be present inside a given gen show that the outer parts of spiral galaxies are mostly
radius to account for the orbital velocity we measure at that dark matter. Astronomers currently estimate that as much
radius?” Recall from Chapter 10 that only the mass inside as 95 percent of the total mass in some spiral galaxies con-
a given radius contributes to the net gravitational force felt sists of a greatly extended dark matter halo (Figure 19.16b),
by an object. Strictly speaking, this is true only for spheri- far larger than the visible spiral portion of the galaxy
cally symmetric objects, but a spiral galaxy is symmetric located at its center. This is a startling statement. A spiral
enough for this to be a good approximation. Figure 19.16a galaxy illuminates only the inner part of a much larger
shows the result of such a calculation. In addition to the distribution of mass that is dominated by some type of mat-
centrally concentrated luminous matter, these galaxies must ter we cannot see!
have a second component consisting of matter that does not Like spiral galaxies, elliptical galaxies contain a large
show up in our census of stars, gas, and dust. This material, fraction of dark matter. To calculate an elliptical galaxy’s
which reveals itself only by the influence of its gravity, is total mass, we begin by measuring its luminous matter.
called dark matter. AstroTour: Dark Matter Since elliptical galaxies do not rotate, however, we cannot
We “see” the presence of dark matter in spiral galaxies use the same approach to measuring their masses that we
only because it is distributed differently from the starlight. used for spiral galaxies. Our discussion of planetary atmo-
(a) 200 1 If the rotation curve (b)

of a spiral galaxy is
at ass
flat like this…
rk d m
r
te
da se
m
The observed
of nclo
rotation curve Normal luminous
E
matter galaxy
150
Mass enclosed within radius

2 …but there is only this
much normal luminous
Rotation velocity
matter enclosed within

a given radius… 3 …then this much dark
100 matter must be present
within a given radius to
provide the needed gravity. Dark matter halo
The luminous galaxies we
see sit within larger, more
massive dark matter halos.
50
tter
minous ma
ass of normal lu
Enclosed m
0
0 40 80 120 160
Radius (thousands of light-years)
FIGURE 19.16 (a) We can use the flat rotation curve of the spiral galaxy NGC 3198 to determine the total
mass within a given radius. Notice that the normal mass that can be accounted for by stars and gas accounts
for only part of the needed gravity. Extra dark matter is needed to explain the rotation curve. (b) In addition to
the matter we can see, galaxies must be surrounded by halos containing a large amount of dark matter.
FIGURE 19.17 Combined visible-light and X-ray images of
spheres in Chapter 8 provides just the tool we need. A plan- elliptical galaxy NGC 1132. The false-color blue/purple halo is
X-ray emission from hot gas surrounding the galaxy. The hot gas
et’s ability to hold on to its atmosphere depends on its mass.
extends well beyond the visible light from stars.
In like fashion, an elliptical
galaxy’s ability to hold on to The dark matter of
its hot, X-ray-emitting gas elliptical galaxies
depends on its mass. First we enables them to hold
infer the total amount of gas on to their hot gases.
from X-ray images, such as the
blue/purple halo seen in Figure 19.17. Next we calculate the
mass that is needed to hold on to the gas. We then compare
that mass with the amount of luminous matter in the center
of the galaxy (the whitish region of the galaxy pictured in
Figure 19.17). The amount of dark matter is the difference
between what is needed to hold on to the gas and the
observed luminous matter.
When we do this, we discover the same thing that we
found from the rotation curves of spirals. Some elliptical
galaxies contain up to 20 times as much mass as can be
accounted for by their stars and gas alone, so they also must
be dominated by dark matter. As with spirals, the luminous
matter in ellipticals is more centrally concentrated than
is the dark matter. The transition from the inner parts of
galaxies (where luminous matter dominates) to the outer
parts (which are dominated by dark matter) is remarkably astronomers called them “radio stars.” Obtaining spectra
smooth. As we will see in Chapter 22, this seamless com- of the fi rst two radio stars was a laborious task, requiring
bination of dark and luminous matter is an important piece 10-hour exposures with the only recording technique avail-
of evidence that any successful theory of galaxy formation able in those days: slow photographic plates (see Chapter 5).
must explain. Of course, some galaxies may contain less Hardly prepared for what they would see in these painstak-
dark matter than others, but it is probably safe to conclude ingly obtained spectra, astronomers were greatly puzzled
that about 90 percent of the total mass in a typical galaxy by the results. Rather than displaying the expected absorp-
is in the form of dark matter. tion lines that are characteristic of blue stellar objects, the
What, then, is this dark matter that makes up most of a spectra showed only a single pair of emission lines that
galaxy? We do not yet know. A number of suggestions have were broad—indicating very rapid motions within these
been made over the years: Jupiter-like objects, swarms of objects—and that did not seem to correspond to the lines
black holes, copious numbers of white dwarf stars, and of any known substances. Puzzling indeed!
exotic unknown elementary particles. (The last of these is For several years astronomers believed they had discov-
currently the favored explanation.) Given that galaxies are ered a new type of star. After all, they reasoned, “If it looks
formed mostly of dark matter, have we been wasting our like a star, it must be a star.” Finally, one astronomer, Maarten
time paying so much attention to the small fraction of mass Schmidt (1929–), realized what no one else had considered:
tied up in stars and the interstellar medium? No. For one that these broad spectral lines were, in fact, the highly red-
thing, stars and gas are the only parts of galaxies that we shifted lines of ordinary hydrogen. The implications were
can see directly. Stars are also of undeniable importance astounding! These “stars” were not stars. They were extraor-
from our human perspective. Stars formed the atoms in our dinarily luminous objects at enormous distances.
bodies and are the kernels around which planetary systems Other “quasars,” as they came to be known, were soon
form. The star we call our Sun supplies the energy that found by the same techniques. Many were relatively easy to
makes life on Earth possible. The attention we have paid to identify because of their unusual blue color. As still more
stars is well placed indeed! were found, astronomers did what astronomers always do:
they began cataloging them. (The young daughter of two
of the catalogers misunderstood what her parents termed
“quasi-stellar objects,” hearing it instead as “crazy-stellar
19.4 There Is a Beast at the objects”—a not altogether unfitting description of this
strange new phenomenon, and a source of much amuse-
Heart of Most Galaxies ment in that household!)
Quasars are phenomenally powerful, pouring forth the
Galaxies are remarkable objects, each shining with the light luminosity of a trillion to a thousand trillion (1012–1015) Suns!
of hundreds of billions of stars. Galaxies themselves change Quasars are objects of the distant universe. The nearest qua-
over time as the universe expands and as their stars evolve. sar to us is approximately 1 billion light-years away. Literally
In the past, the rates of star formation in galaxies were billions of galaxies are closer to us than the nearest quasar.
higher and the galaxies were closer together, increasing the Recall that the distance to an object also tells us the amount
chance of galaxy collisions. of time that has passed since the light from that object left its
We will return to these points Quasars are source. The fact that the nearest quasar is seen as it existed
and to the subject of dark mat- phenomenally about a billion years ago tells us that quasars are quite rare
ter in galaxies when we dis- luminous. in the universe today. Quasars were once much more com-
cuss galaxy formation in mon. The discovery that quasars existed in the distant and
Chapter 22. But galaxies pale in comparison with the most therefore earlier universe provided one of the fi rst pieces of
brilliant beacons of all: quasars. Quasar is short for “quasi- evidence that the universe has evolved over time.
stellar radio source,” so named because astronomers fi rst Quasars are not the isolated beacons they were once
observed these mysterious objects as unresolved points at thought to be. Instead, they are centers of violent activity in
radio wavelengths. The story of the discovery of quasars the hearts of large galaxies, as
provides an interesting insight into the discovery of new seen in Figure 19.18. Astrono- There are several
phenomena in astronomy and how conventional thinking mers now recognize that qua- types of active
can sometimes hinder progress. sars are only the most extreme galactic nuclei.
In the late 1950s, radio surveys had detected a number of form of activity that can occur
bright, compact objects that at fi rst seemed to have no opti- in the nuclei of galaxies. In the universe we see today, approxi-
cal counterparts. Eventually, astronomers found the opti- mately 3 percent of galaxies contain brilliant points of light
cal counterparts when improved radio positions revealed in their centers that may outshine all of the stars in the gal-
that the radio sources coincided with faint, very blue, stel- axies that host them. Together, quasars and their less lumi-
larlike objects. Unaware of the true nature of these objects, nous but still active cousins are called active galactic nuclei,
FIGURE 19.18 The environments around quasars, which are found in the centers of galaxies.
Those galaxies often show evidence of interactions with other galaxies.
or simply AGNs. Today, we know that there are several dis- typical Seyfert nucleus can be 10 billion to 100 billion L ⊙,
tinct types of active nuclei within various types of galaxies. comparable to the luminosity of the rest of the galaxy as a
Seyfert galaxies, named after American astronomer whole. The luminosities of AGNs found in elliptical galaxies
Carl Seyfert (1911–1960), who discovered them in 1943, are are similar to those of Seyfert nuclei (10 billion to 100 bil-
spiral galaxies whose centers contain AGNs discernible in lion L ⊙). Unlike Seyfert nuclei, however, AGNs in elliptical
visible light as a distinct bright spot. The luminosity of a galaxies are usually most prominent in the radio portion
FIGURE 19.19 Radio emission

from a double-lobed radio galaxy
(shown in red) is superimposed
over an image of visible starlight
from the galaxy (shown in
blue). The lobes, powered by a
relativistic jet streaming outward
from the nucleus of the galaxy,
extend to more than 3 million
light-years from the galaxy’s
center.
Radio Microwave IR Optical X-ray a-ray ies at speeds of thousands or even tens of thousands of
kilometers per second.
10–12
Brightness at each frequency (W/m2)
AGNs Are the Size of the Solar System

The enormous radiated power and mechanical energy of
active galactic nuclei are mind-boggling on their own, but
they are made even more spectacular by the fact that all of
10–14
this power emerges from a region that can be no larger than
a light-day or so across. That is comparable in size to our
own Solar System! How can we even make such a claim?
Quasars and other AGNs appear only as unresolved points
of light, even in our most powerful telescopes.1 What infor-
10–16
mation in the light that we receive from AGNs tells us they
are such compact objects? For an answer, we turn not to the
sky but to the halftime show at a local football game.
1010 1015 1020 1025 Figure 19.21 illustrates a problem that faces every
Frequency (Hz) marching-band director. When a band is all together in a
FIGURE 19. 20 Electromagnetic energy spectrum of the well-
tight formation at the center of the field, the notes you hear
studied and bright quasar 3C 273 from low-frequency radio to in the stand are clear and crisp; the band plays together
high-energy gamma rays. Most of the energy emitted by this beautifully. But as the band spreads out across the field, its
quasar lies between infrared and gamma-ray frequencies. sound begins to get mushy. This is not because the marchers
are poor musicians. Rather, it is a consequence of the fact
that sound travels at a fi nite speed. On a cold, dry, Decem-
of the electromagnetic spectrum, earning them the name ber day sound travels at a speed of about 330 meters per
radio galaxies. Radio galaxies, and their distant, extremely second (m/s). At this speed, it takes sound approximately
1
luminous cousins the quasars, are often the sources of slen- ⁄3 of a second to travel from one end of the football field to
der jets that extend outward millions of light-years from the other. Even if every musician on the field played a note
the galaxy, powering twin lobes of radio emission, such as at exactly the same instant in response to the director’s cue,
those seen in Figure 19.19. in the stands you would hear the instruments close to you
What could power such phenomenal cosmic beacons fi rst but would have to wait longer for the sound from the
as quasars, Seyfert galaxies, and radio galaxies? Clues lie far end of the field to arrive.
in the light they emit. Figure 19.20 shows the electromag- If the band is spread from one end of the field to the other,
netic spectrum, ranging from radio frequencies to high- then the beginning of a note will be smeared out over about
1
energy gamma rays, of the well-studied quasar 3C 273. This ⁄3 of a second, or the difference in sound travel time from
quasar is about 100 times as the near side of the field to the far side. If the band were
luminous as our entire Milky AGNs emit spread out over two football fields, it would take about 2 ⁄3
Way Galaxy, and it exhibits synchrotron radiation. of a second for the sound from the most distant musicians
much stronger radio and to arrive at your ear. If our marching band were spread out
gamma-ray emission than is typical for AGNs. Much of the over a kilometer, then it would take roughly 3 seconds—
light from AGNs is synchrotron radiation. This is the same the time it takes sound to travel a kilometer—for us to hear
type of radiation that we fi rst encountered coming from a crisply played note start and stop. Even with our eyes
Jupiter’s magnetosphere and later saw again in such extreme closed, it would be easy to tell whether the band was in a
environments as the Crab Nebula. Recall from Chapter 9 tight group or spread out across the field.
that synchrotron radiation comes from relativistic charged Exactly the same principle works for AGNs; but here we
particles spiraling around in the direction of a magnetic are working with the speed of light, not the speed of sound.
field (see Figure 9.20). The fact that AGNs accelerate large We observe that quasars and other AGNs can change their
amounts of material to nearly the speed of light indicates brightness dramatically over the course of only a day or
that they are very violent objects indeed. In addition to the two—and in some cases as briefly as a few hours! This rapid
continuous spectrum of synchrotron emission, the spectra
of many quasars and Seyfert nuclei also show emission 1
HST and large groundbased telescopes show faint “fuzz” around the
lines that are smeared out by the Doppler effect across a images of some quasars and AGNs. This fuzz is light from the sur-
wide range of wavelengths. This observation implies that rounding galaxy. The quasars and AGNs themselves, however, remain
gas in AGNs is swirling around the centers of these galax- as unresolved points of light.
Band
Musicians
4 Sound waves
from a band that
is spread out…
Direction
of sound
Direction
1 Sound waves from of sound
a compact band…
Sound
waves
2 …reach a listener at
Listener almost the same time…
5 …are spread out as
Duration = Size of band they reach a listener.
3 …so the note heard of sound Speed of sound
is crisp and clean. The sound is “mushy.”
Loudness
Loudness
Duration = Size of band

of sound Speed of sound
Similarly, a quasar that
changes brightness rapidly
Time must be very small.
Time
FIGURE 19. 2 1 A marching band spread out across a field cannot play a clean note. Similarly,
AGNs must be very small to explain their rapid variability (see the text).
variability sets an upper limit on the size of the AGN, just tions revealed the tiny sizes and incredible energy
as hearing clear music from the band tells us that the band densities of AGNs, only one answer seemed to make sense:
musicians are close together. The AGN powerhouse must violent accretion disks surrounding supermassive black
therefore be no more than a holes—black holes with masses from thousands to tens
light-day or so across because AGNs vary rapidly of billions of solar masses—power AGNs. We have run
if the powerhouse were larger, and so must be across accretion disks on a number of occasions during
what we see could not possi- relatively small. our journey. Accretion disks surround young stars, pro-
bly change in a day or two. viding the raw material for
Here is the image that should come to mind when you think solar systems. Accretion AGNs are powered
about active galactic nuclei: the light of 10,000 galaxies disks around white dwarfs, by accretion onto
pouring out of a region of space that would come close to fueled by material torn from supermassive
fitting within the orbit of Neptune! their bloated evolving com- black holes.
panions, lead to novae and
Type I supernovae. Accretion disks around neutron stars
Supermassive Black Holes and and star-sized black holes a few kilometers across are seen
as X-ray binary stars. Now take these examples and scale
Accretion Disks Run Amok them up to a black hole with a mass of a billion solar
When astronomers fi rst discovered AGNs, they put for- masses and a radius comparable in size to the orbit of
ward a variety of ideas to explain them. But as observa- Neptune. Furthermore, imagine an accretion disk fed by
(d) (c)
3 This observer sees 2 This observer also sees
relativistically beamed the accretion disk and
emission from the jet. surrounding gas.
UV radiation
NGC 1275
(a) Gas ionized M87

by UV from
central source
1 The torus blocks this observer’s

view of the central black hole and
accretion disk. Only the extended
Shadow Shadow region of emission is seen.
of torus Supermassive of torus (b)
black hole
Accretion disk
Obscuring
torus
NGC 7052
Radio jet
FIGURE 19. 22 (a) The unified model of active galactic nuclei. The appearance of the object
changes when the disk and torus are viewed from the edge (b), at higher inclination (c), and close to
face on (d). Astronomers believe that the viewing angle, the mass of the central black hole, and the
rate at which it is being fed determine the properties of any AGN that we see.
substantial fractions of entire galaxies rather than the ent components of the unified model accounts for different
small amounts of material being siphoned off a star. That observed properties of AGNs.
is an active galactic nucleus. In our discussion of star formation, we learned that
Astronomers have developed this basic picture of a gravitational energy is converted to thermal energy as mate-
supermassive black hole surrounded by an accretion disk rial moves inward toward the
into a more complete physical description called the uni- growing protostar. As mate- AGN accretion disks
fied model of AGNs. The unified model takes its name rial moves inward toward a are phenomenally
from the fact that it attempts to unify our understanding of supermassive black hole, con- efficient at converting
all AGNs—quasars, Seyfert galaxies, and radio galaxies— version of gravitational energy mass to energy.
within the same framework of understanding. heats the accretion disk to
Figure 19.22a shows the various components of the uni- hundreds of thousands of kelvins, causing it to glow brightly
fied model of AGNs. In this model, an accretion disk sur- in visible and ultraviolet light. Conversion of gravitational
rounds a supermassive black hole. Much farther out lies a energy to thermal energy as material falls onto the accre-
large torus (doughnut) of gas and dust consisting of mate- tion disk is also a source of X-rays, UV radiation, and other
rial that is feeding the central engine. Each of the differ- energetic emission. When we discussed the Sun, we mar-
veled at the efficiency of fusion, which converts 0.7 percent of the mass
50,000 of hydrogen into energy. In contrast, approximately 50 percent of the
Lobes of a radio galaxy
light-years hundreds of thousands mass of infalling material around a supermassive black hole is con-
of light-years in size…
verted to luminous energy. The rest of that mass is pulled into the black
hole itself, causing it to grow even more massive.
The interaction of the accretion disk with the black hole gives rise
to powerful radio jets—superpowerful analogs to the Herbig-Haro jets
formed by accretion disks around young stellar objects (see Chapter
15). Throughout, twisted magnetic fields accelerate charged particles
such as electrons and protons to relativistic speeds, accounting for the
observed synchrotron emission. Gas in the accretion disk or in nearby
clouds orbiting the central black hole at high speeds gives off emission
lines that are smeared out by the Doppler effect into the broad lines
seen in AGN spectra. This accretion disk surrounding a supermassive
5,000
light-years black hole is the “central engine” that leads to AGNs.
The outer torus plays a somewhat different role in the unified model.
Located far from the inner turmoil of the accretion disk, and far larger
than the central engine, some of the outer torus is ionized by UV light
from the AGN much as an H II region is ionized by the UV light from
O stars. This ionization provides one pos-
sible source for the H II region–like emis- The outer torus of
sion lines seen in many AGNs. The most an AGN determines
important conceptual role of the outer what we can see.
torus is that it allows a single model to
explain many of the differences observed among various AGNs. The
outer torus obscures our view of the central engine in different ways,
1,000 depending on the angle from which we happen to see it. By invoking
light-years
variations in the viewing angle, the mass of the black hole, and the rate
at which it is being fed, the unified model of AGNs can account for a
wide range of AGN properties, including the average spectra of AGNs,
an example of which is shown in Figure 19.20.
What AGN We See Depends

50 light-years
on Our Perspective
When we view the unified model edge on, we see emission lines from
the surrounding torus and other surrounding gas. We can also some-
FIGURE 19. 23 The visible jet from the galaxy M87 extends
over a hundred thousand light-years but originates in a tiny
volume at the heart of the galaxy (far lower right).
…originate in a central
engine perhaps no larger
than our Solar System.
10 light- 1 light- 0.1

years year light-year
FIGURE 19. 2 4 An HST

image of the nearby radio
galaxy M87. The high velocities
Redshifted observed provide direct
emission from evidence of rotation about a
this side… supermassive black hole at this
galaxy’s center.
Redshifted lines
Blueshifted lines
…and blueshifted
Intensity
emission from
this side…
…are the signature

of a rotating disk.
0.48 0.49 0.50 0.51

Wavelength (μm)
times see the torus in absorption against the background of Doppler effect called relativistic beaming. Matter traveling
the galaxy. Figure 19.22b is a Hubble Space Telescope image at close to the speed of light concentrates any radiation it
of the inner part of the galaxy NGC 7052, showing what emits into a tight beam pointed in the direction in which
appears to be just such a it is moving. As a result of relativistic beaming, an AGN jet
shadow. From this nearly Viewed edge on, an coming toward us will look much brighter than its twin
edge-on orientation, we can- AGN’s accretion disk moving away from us. The
not see the accretion disk is not visible. unified model clearly predicts Relativistic effects
itself, so we do not expect to that AGN jets should be two- influence what part of
see the Doppler-smeared lines that originate closer to the sided, but most of the time an AGN we see.
supermassive black hole. If jets are present in the AGN, what we actually see appear
however, we should be able to see these emerging from the to be one-sided jets. The reason is that we see only the por-
center of the galaxy. tion of the jet that is being beamed toward us. The emission
If we look at the inner accretion disk somewhat more from the other portion of the jet always exists. We come to
face on (Figure 19.22d), we can see over the edge of the torus, this conclusion because the radio lobes of radio galaxies
and thus we get a more direct look at the accretion disk and are always two-sided. We don’t see the jet that is moving
the location of the black hole. In this case, we should see away from us because it is beaming its radiation in the other
more of the synchrotron emission from the region around direction.
the black hole and the Dop- In rare instances we happen to see the accretion disk
pler-broadened lines produced When we view it more in a quasar or radio galaxy almost directly face on. In
in and around the accretion face on, we see the such cases, relativistic beaming dominates what we see.
disk. Figure 19.22c shows a AGN’s accretion disk. Rather than seeing emission lines and other light com-
Hubble Space Telescope image ing from hot gas in the accretion disk, we are blinded by
of one such object, called M87,2 at an intermediate inclina- the bright glare of jet emission beamed directly at us (see
tion. M87 is a source of powerful jets that continue outward Figure 19.22d).
for a hundred thousand light-years, but originate in the tiny Relativistic beaming is not the only thing that com-
engine at the heart of the galaxy (Figure 19.23). Spectra of plicates what we see when we look down the barrel of an
the disk at the center of this galaxy (Figure 19.24) show the AGN jet. The material in an AGN jet is moving so close to
rapid rotation of material around a central black hole with the speed of light that the radiation it emits is barely able
a mass of 3 billion solar masses (3 × 109 M⊙). to outrun its source. As observers, we see all of the light
The material in an AGN jet travels very close to the speed emitted by the jet over thousands of years arrive at our tele-
of light. As a result, what we see is strongly influenced by scopes over the course of only a few years. Time appears
relativistic effects. One of these is an extreme form of the to be compressed. From our perspective, the jet seems to
travel great distances in brief periods of time. In extreme
2
Object number 87 in Messier’s catalog (see Chapter 18). cases, such as the jet in M87 (Figure 19.25), features in the jet
unified model make, and how have those predictions fared

in the light of new observational tests?
The essential elements of the unified model are a central
An object moving almost engine (an accretion disk surrounding a supermassive black
Year directly toward us at close to hole) and a source of fuel (gas and stars flowing onto the
the speed of light will appear accretion disk). We commonly speak of the infall of material
1994 to move faster than light.
onto the accretion disk as “feeding the beast.” If we were to
shut off this infall—if we were to stop feeding the beast—
the supermassive black hole would remain, absent all the
1995 fi reworks. Without a source of matter falling onto the black
hole, an AGN would no longer be active. If we were to look
at such an object, we would see a normal (not active) galaxy
with a supermassive black hole sitting in its center.
1996 Should such objects be common? As we have noted,
only about 3 percent of present-day galaxies contain AGNs.
But when we look at more distant galaxies (and therefore
look back in time), the percentage is much larger. Our obser-
1997
vations show that when the
universe was younger, there The unified model
were many more AGNs than predicts that normal
there are today. If the unified galaxies contain
1998
model of AGNs is correct, supermassive
then all the supermassive black holes.
Position on sky black holes that powered
those dead AGNs should still be around. If we combine
Features in this radio jet …but that is an illusion. what we know of the number of AGNs in the past with ideas
appear to be moving at six Nothing moves faster
about how long a given galaxy remains in an active phase,
times the speed of light… than light.
we are led by the unified model to predict that many—
FIGURE 19. 25 Because the jet material in M87 is moving perhaps even most—normal galaxies today contain super-
toward us at relativistic speeds, we see an optical illusion: the massive black holes!
radio blobs appear to be moving apart at more than six times the This is a somewhat startling prediction—that quiescent
speed of light. collections of stars, gas, and dust like our own Milky Way
should have slumbering beasts at their centers. It is a bit like
suggesting that all of our dignified, soft-spoken grandfathers
were once members of biker gangs. Yet here is a prediction
appear to be moving across the sky faster than the speed of of the unified model that can be tested.
light. We stress the word appear because this phenomenon, If supermassive black holes are present in the centers of
referred to as superluminal motion, is an optical illusion. normal galaxies, they should reveal themselves in a number
Despite the name, nothing in these jets is actually travel- of ways. For one thing, such a concentration of mass at the
ing through space faster than the speed of light. Einstein’s center of a galaxy should draw
special theory of relativity remains safe. surrounding stars close to it. Supermassive black
It is worth noting that we have used the same galaxy, The central region of such a holes have been
M87, as an example of several of the phenomena dis- galaxy would be much brighter discovered in the
cussed in this chapter. The presence of many different than could be explained if nuclei of many
phenomena predicted by the unified model in the same stars alone were responsible nearby galaxies.
object is important support for the view that the model for the gravitational field in
truly is unified. the inner part of the galaxy. Stars feeling the gravitational
pull of a supermassive black hole in the center of a galaxy
should also orbit at very high velocities. We should there-
Normal Galaxies and AGNs— fore see large Doppler shifts in the light from stars near the
centers of normal galaxies. Astronomers have, in fact, found
A Question of Feeding the Beast evidence of this sort in every normal galaxy with a substan-
Astronomers began developing the unified model of AGNs tial bulge in which a careful search has been conducted.
in the 1980s, and it has been modified frequently to account The masses inferred for these black holes range from 10,000
for new and better data. What testable predictions does the M⊙ (for a “small” black hole) to 5 billion M⊙ (a “gargantuan”
FIGURE 19. 26 These tidally interacting galaxies show severe distortions, including stars and gas
drawn into long tidal tails.
one)! The mass of the supermassive black hole seems to be hole, forming an accretion disk and a surrounding torus.
related to the mass of the elliptical-galaxy or spiral-galaxy The predicted result of this process is that the nucleus of
bulge in which it is found. At the beginning of the 21st cen- this galaxy would change into an AGN.
tury we have come to accept that all large galaxies probably
contain supermassive black holes. These observations con-
fi rm the most fundamental prediction of the unified model. Mergers and Interactions
They also tell us something remarkable about the structure
and history of normal galaxies.
Make the Difference
Apparently the only difference between a normal galaxy Galaxies do not exist in isolation. Even our own Milky Way
and an active galaxy is whether the supermassive black hole has several neighbors. In Chapter 10 we discussed tidal
at its center is being fed at the time we see that galaxy. The interactions between galaxies. Figure 19.26 shows the havoc
fact that only 3 percent of present-day galaxies have AGNs that such interactions can cause, pulling interacting galax-
does not indicate which galaxies have the potential for AGN ies into distorted shapes in which stars and gas are drawn
activity. Rather, it indicates which galaxy centers are being out into sweeping arcs and tidal tails (see also Figures 5.31
lit up at the moment. If we were to drop a large amount of and 10.14). Sometimes when galaxies interact, they pass by
gas and dust directly into the center of any large galaxy, each other and go their separate ways. Such interactions
this material would fall inward toward the central black can trigger the formation of spiral structure, and they can
also slam clouds of interstellar gas together at high speeds, were forming, and large amounts of matter were constantly
triggering additional star formation. Tidally distorted or being drawn in by the gravity of newly formed galaxies.
interacting galaxies often contain regions of vigorous ongo- This process is still at work today. Galaxies that show evi-
ing star formation. Sometimes when two galaxies interact, dence of recent interactions with other galaxies are far more
they merge to form a single, larger galaxy. This turns out to likely to house AGNs in their centers.
be an important part of the process of galaxy formation, as There are still many puzzles. For example, the unified
we will learn in Chapter 22. AstroTour: Galaxy Inter- model has not been developed to the point that it predicts
actions and Mergers how long an outburst of AGN activity will last, or how often
To account for the many large galaxies that we see today, galaxies will undergo episodes of AGN activity. To answer
interactions and mergers must have been much more preva- these questions, we will have to combine the unified model
lent in the past. The preva- of AGNs with much better models of galaxy formation and
lence of interacting galaxies Galaxy-galaxy evolution than we currently have. In addition, some obser-
when the universe was interactions fuel vations are not yet explained by the unified model of AGNs.
younger explains the large AGN activity. For instance, the unified model does not account for why one
number of AGNs that existed quasar can be a very powerful radio source while another,
in the past. Computer models show that galaxy-galaxy inter- identical in all other respects, is radio-quiet, even when
actions can cause gas located tens of thousands of light-years observed with the most sensitive radio telescopes.
from the center of a galaxy to fall inward toward the center, Our understanding of AGNs is far from complete. Even
where it can provide fuel for an AGN. During mergers, a so, the unified model has had many successes—enough
significant fraction of a cannibalized galaxy might wind up for us to say with confidence that any large galaxy, includ-
being fed to the beast. HST images of quasars, such as those ing our own, might be only a chance encounter away from
in Figure 19.18, often show that quasar host galaxies are becoming an AGN. How different our own sky might be if,
tidally distorted or are surrounded by other visible matter a few tens of millions or hundreds of millions of years from
that is probably still falling into the galaxies. The most vio- now, our descendants look toward the center of our galaxy
lent forms of AGN activity were likely most common in the and see the brilliant light of a powerful Seyfert nucleus
early universe because that was the time when galaxies blazing forth.
The scale and distance of galaxies may have shat-

Seeing the Forest for the Trees tered preconceptions about the extent of the universe,
but galaxies themselves seemed at first to be composed of
A century is but a blink of an eye compared with the familiar objects. In essence, the zoo grew far larger but the
age of Earth or even the age of our species. Such is the animals themselves seemed the same. Stars in distant gal-
story of 20th century astronomy. A hundred years ago, axies shine in accordance with the same laws that govern
humankind knew virtually nothing of the true answers the Sun. Galaxies are held together by the same force of
to the most basic questions we might ask about the uni- gravity that set the course of the cannonball in Newton’s
verse. Virtually all that we know of these matters we have famous thought experiment. But as our understanding
learned within the past century. The discovery and study grew and new observations were made, observations of
of galaxies is one of the major threads running through- galaxies forced astronomers to push our knowledge of
out the story of that century. physics in extreme and unexpected ways.
From the time that Copernicus fi rst dislodged Earth When we fi rst learned of black holes, we were faced
from the center of creation, each new discovery has with the thought of several solar masses compressed into
pushed us further and further from the human-centered a region only a few kilometers across, and we had to con-
conceptions that shaped our view of the world for millen- front ideas of general relativity that turn our everyday
nia (and persist in many ways to this day). The discovery notions of space and time inside out. Now we discover
of the almost unthinkable size of our own Milky Way that such star-sized black holes are the merest of specks
and the realization that the universe contains countless compared with the monsters residing in the centers of
more galaxies comparable to our own were huge steps large galaxies. We have witnessed the consequences when
in humanity’s expanding awareness of the universe. these behemoths, with masses as great as billions of times
The work of Hubble and others did far more than merely that of the Sun, are fed with gas supplied by collisions
answer a few arcane and esoteric scientific questions. between galaxies. Radio observations of the sky show us
These scientists forever changed the way we must view relativistic jets stretching across millions of light-years of
everything, including ourselves. intergalactic space. Yet the source of such a jet is tiny. The
light of a quasar outshines a thousand galaxies—a beacon As we have seen, in the 20th century astronomers
that can be seen from the very edge of the universe—yet made dramatic progress toward building a physical
originates from a region no larger than our Solar System. understanding of the formation, evolution, and death of
As we continue our journey, we will find that such a mon- stars. We now can use the laws of physics to peel back
ster lurks at the heart of our own mundane Milky Way, the layers of a star, peer into the heart of a supernova,
waiting (perhaps forever) for its next meal. or run the clock forward and see our Sun’s ultimate
Active galaxies stir the imagination, but one of the fate. Astronomers cannot claim a similar understand-
most startling and fundamental results of our study of ing of galaxies, because the study of galaxies is still in
galaxies comes from the simple application of Newton’s its infancy. Remember how little time has passed since
derivation of Kepler’s laws to the motions of stars in gal- galaxies were even recognized for what they are. It is
axies. We have used this technique over and over on our unlikely that there will ever be a simple scheme that
journey to measure the mass of planets and stars. When does for galaxies what the H-R diagram did for stars.
we apply this comfortable tool to galaxies, however, the On the other hand, astronomers are beginning to better
results are shocking. The matter that we see in the stars, understand what the “ecology” within galaxies is like
gas, and dust is but the tip of a much larger iceberg, the and to piece together something of how they form and
rest of which is composed of a substance known only as evolve. In the next chapter we take a step down this road
dark matter. Most of the matter in the universe consists by looking in more detail at the galaxy we know best:
of we know not what. our own Milky Way.

• The shapes of galaxies and the types of orbits of their THINK ING A BOUT THE CONCEP T S
stars determine their Hubble classification.
1. Name and describe a common, everyday object that
• Currently, stars are forming in the disks of spiral galax- appears so dissimilar when viewed from different angles
ies but not in ellipticals or S0 galaxies. that you might think from these different views that it
is actually a different object.
• Spiral arms are regions of intense star formation, and
the arms are visible because of the concentration of 2. Name and describe three types of galaxies.
bright young stars.
3. What are the principal morphological (structural) dif-
• Most of the mass in galaxies does not reside in gas, dust, ferences between spiral and elliptical galaxies?
or stars; rather, about 90 percent of a galaxy’s mass is in
4. How do E0 and E7 elliptical galaxies differ in shape?
the form of dark matter, which does not emit or absorb
light to any significant degree. 5. What would be the shape of an elliptical galaxy whose
stars were all traveling in random directions in their
• Most—perhaps all—large galaxies have supermassive
orbits? Explain your answer.
black holes at their centers.
6. How does molecular-gas temperature differ between
• When gas accretes onto one of these supermassive black
elliptical and spiral galaxies?
holes, the center of the galaxy becomes an active galactic
nucleus, which can emit as much as a thousand times 7. Explain why star formation in spiral galaxies takes place
the light of the whole galaxy, all coming from a region mostly in the spiral arms.
the size of our Solar System.
8. Some galaxies have regions that are relatively blue in
color, while other regions appear redder. Aside from
color, what can you say about the differences between
SmartWork, Norton’s online homework system, these regions?
9. Describe the characteristics of irregular galaxies.
your instructor assigns questions in SmartWork, log in at 10. Galaxies come in a large range of both luminosity and
smartwork.wwnorton.com. size. Which of these two properties varies more among
galaxies?
*11. Describe the spiral arms in a galaxy and explain at least 27. Suppose the number density of galaxies in the universe
one of the mechanisms that create them. is, on average, 3 × 10 –68 galaxies per cubic meter (m3).
If astronomers could observe all galaxies out to a dis-
12. In describing galaxies, what do astronomers mean by
tance of 1010 light-years, how many galaxies would they
luminous (or normal) matter?
fi nd?
13. With regard to its interaction with electromagnetic radia-
28. The Keplerian speed of a distant star orbiting a galaxy
tion, how does dark matter differ from normal matter?
is twice the speed that the galaxy’s visible mass would
14. What evidence do we have that most galaxies are com- suggest. What is the ratio of dark matter to normal mat-
posed largely of dark matter? ter within the star’s orbit?
15. Contrast the rotation curve for a galaxy containing only 29. The nearest known quasar is 3C 273. It is located in the
normal matter with one containing dark matter. constellation of Virgo and is bright enough to be seen
in a medium-sized amateur telescope. With a redshift
16. How does a spiral galaxy’s dark matter halo differ from
of 0.158, what is the distance of 3C 273 in light-years?
its visible spiral component?
30. The quasar 3C 273 has a luminosity of 1012 L ⊙. Assum-
*17. Name some of the candidates for the composition of dark
ing that the total luminosity of a large galaxy, such as
matter.
the Andromeda Galaxy, is 10 billion times that of the
18. How would you explain a quasar to a relative or Sun, compare the luminosity of 3C 273 with that of the
friend? entire Andromeda Galaxy.
19. Which is more luminous: a quasar or a galaxy with a 31. Consider a hypothetical star orbiting 3C 273 at a distance
hundred billion solar-type stars? Explain you answer. of 100,000 astronomical units (AU) with a period of 1,080
years. What is the mass of 3C 273 in solar masses?
*20. The nearest quasar is about a billion light-years away.
Why do we not see any that are closer? 32. Stars at a distance of 50 light-years from the center of a
galaxy are orbiting the galactic center at a speed of 200
21. What distinguishes a “normal” galaxy from one we call
kilometers per second (km/s). Use Newton’s laws to cal-
“active”—that is, one that contains an AGN?
culate the mass of the supermassive black hole at the
22. Contrast the size of a typical AGN with the size of our galaxy’s center.
own Solar System. How do we know how big an AGN
33. A quasar has the same brightness as a foreground gal-
is?
axy that happens to be 6 million light-years distant. If
23. Describe what must be happening at the centers of gal- the quasar is 1 million times more luminous than the
axies that contain AGNs. galaxy, what is the distance of the quasar?
**24. The material in some AGN jets appears to be moving 34. You read in the newspaper that astronomers have dis-
faster than the speed of light (superluminal motion). covered a “new” cosmological object that appears to be
Why is this not possible, and what contributes to this flickering with a period of 83 minutes. Having read 21st
illusion? Century Astronomy, you are able to quickly estimate the
maximum size of this object. How large can it be?
25. It is likely that most galaxies contain supermassive black
holes, yet in many galaxies there is no obvious evidence 35. A solar-type star (M⊙ = 2 × 1030 kg), accompanied by its
for their existence. Why do some black holes reveal their retinue of planets, approaches a supermassive black hole.
presence while others do not? As it crosses the event horizon, half of its mass falls into
the black hole, while the other half is completely con-
verted to luminous energy. As it signals its demise in a
burst of electromagnetic radiation, how much energy (in
joules) does the dying solar system send out to the rest
26. Assume that there are 1 trillion (1012) galaxies in the of the universe?
universe, that the average galaxy has a mass equivalent
*36. A quasar has a luminosity of 1041 watts (W), or joules per
to 100 billion (1011) average stars, and that an average
second (J/s), and 108 M⊙ to feed it. Assuming constant
star has a mass of 1030 kilograms (kg).
luminosity and 50 percent conversion efficiency, what
a. Ignoring dark matter, how much mass (in kilograms)
would be your estimate of the quasar’s lifetime?
does the universe contain?
b. If the mass of an average particle of normal matter is 37. Assume that Earth, whose mass (M⊕) is 5.97 × 1024 kg,
10 –27 kg, how many particles are there in the entire fell into a supermassive black hole with a 50 percent
universe? energy conversion.
a. How much energy (in joules) would be radiated by b. Compare this with the angular size of the Moon.
the black hole?
40. A lobe in a visible jet from the galaxy M87 is observed at
b. Compare that with the energy radiated by the Sun
a distance of 5,000 light-years from the galaxy’s center
each second: 3.85 × 1026 joules (J).
and moving outward at a speed of 0.99 times the speed
38. If a luminous quasar has a luminosity of 2 × 1041 W (or of light (0.99c). Assuming constant speed, how long ago
J/s), how many solar masses (M⊙ = 2 × 1030 kg) per year was the lobe expelled from the supermassive black hole
does this quasar consume to maintain its average energy at the galaxy’s center?
output?
39. Material ejected from the supermassive black hole at the
center of the galaxy M87 extends outward from the gal- StudySpace is a free and open Web site that pro-
axy to a distance of approximately 100,000 light-years. vides a Study Plan for each chapter of 21st Century
M87 is approximately 50 million light-years away. Astronomy. Study Plans include animations, read-
a. If this material were visible to the naked eye, how ing outlines, vocabulary flashcards, and multiple-choice
large would it appear in our nighttime sky? Give your quizzes, plus links to premium content in SmartWork and
answer in degrees, recalling from Chapter 13 that 1 the ebook. Visit www.wwnorton.com/studyspace.
radian = 57.3°.
O Milky Way, sister in whiteness
To Canaan’s rivers and the bright
Bodies of lovers drowned,
Can we follow toilsomely
Your path to other nebulae?
GUILL AUME APOLLINAIRE (1880 –1918)
The star fields and dark dust clouds of the Milky Way Galaxy, with a Lyrid meteor and ancient
Bristlecone pines in the foreground.
CHAP TER 20
The Milky Way—

A Normal Spiral Galaxy
K E Y CONCEP T S
20.1 We Look Up and Of the hundreds of billions of galaxies in the universe,
See Our Galaxy the one that means the most to us is our cosmic home,
the Milky Way. The Milky Way may be just another
We live in a universe full of galaxies of many sizes and types, galaxy, but it is the only galaxy we can study at close
visible in our most powerful telescopes all the way to the range. As we focus our attention on our galaxy, we will
edge of the observable universe. Yet when we go outside at learn
night away from city lights and look up, it is not this universe
• How variable stars in globular clusters are used as
of galaxies that we see. Rather, the night sky is filled with a
single galaxy—our home, the standard candles, enabling us to measure the size of
galaxy we call the Milky Way. We live in a barred the Milky Way.
With what we know of galax- spiral galaxy called • How Doppler-shifted radio emission from gas
ies from the previous chapter, the Milky Way. throughout the rotating disk of the Milky Way enables
we can learn a great deal about us to map the galaxy’s structure.
our local galaxy by looking at the night sky. Figure 20.1a
• That the Milky Way is a typical giant barred spiral
shows what our Milky Way Galaxy looks like as seen in our
own skies; Figure 20.1b is an image of an edge-on spiral gal- galaxy, and like all such galaxies it is composed mostly
axy. Comparison of the two can leave little doubt that we live of dark matter.
in the disk of such a spiral. The flattened disk of the Milky • How the chemical composition of the Milky Way has
Way is obvious when we look at the sky, once we appreciate evolved with time.
what we are looking at. We can even see dark bands where
• What differences in the age and chemical composition
clouds of interstellar gas and dust obscure much of the cen-
of groups of stars tell us about the history of star
tral plane of our galaxy. There are many things that we know
about galaxies only from our experience with the Milky Way. formation in our galaxy.
We see it from a much closer perspective than we do any • About the environment within the disk of the Milky
other galaxy. We see it from the inside! Way, and the halo of stars, globular clusters, and dark
At the same time, there are real disadvantages to our matter that surrounds and permeates our galaxy.
perspective on the Milky Way Galaxy. Buried within the
• About the black hole at the center of our galaxy.
disk of the Milky Way, we lack a bird’s-eye view of our
galactic home. The situation is further complicated by the
dust in the surrounding interstellar medium, which also
limits our view. Figure 20.2a shows a model of the structure
563
564 Chapter 20 The Milky Way—A Normal Spiral Galaxy
FIGURE 20.1 (a) We see the

(a)
Milky Way as a luminous band
stretching across our night sky.
Note the prominent dark lanes
caused by interstellar dust that
obscures the light from more
distant stars. (b) The edge-on
spiral galaxy NGC 891, whose
disk greatly resembles the Milky
Way.
(b)
of the Milky Way proposed in the late 1700s by William Early models of our galaxy are of interest today mostly as
Herschel. Bearing little resemblance to our modern under- historical curiosities. Even so, today’s astronomers must cope
standing of the Milky Way, with the same suite of difficulties that stood in the way of
Herschel’s model was based Dust obscures our 19th and early 20th century astronomers. What type of spi-
simply on counting the num- view of much of our ral galaxy do we live in? Are the spiral arms prominent? Is
ber and brightness of stars own galaxy. the bulge barlike, and is it large or small relative to the disk?
seen in different directions. The questions that are easier to ask and answer for distant
Herschel did not know about the interstellar extinction of galaxies are far more difficult to address for our own galaxy.
starlight, the properties of stars, or the relationship between Yet at this stage in our journey, we have already learned far
our galaxy and the “spiral nebulae” that he studied through- more about our galaxy than about any other, as seen in Figure
out his life. 20.2b. Stars, planets, and the interstellar medium—almost
(a) (b)
FIGURE 20. 2 (a) An early
model proposed by William
Herschel in the late 1700s
in which he depicts the Milky
Way as a dense slab of
stars. (b) Infrared and radio
observations contribute to a
modern model of the Milky Way
Galaxy. The galaxy’s two major
arms (Scutum-Centaurus and
Perseus) are seen attached to
the ends of a thick central bar.
everything we know about them we have learned within

the context of our own galactic home. It is time to merge
the perspective of our look outward at the universe of other
galaxies with our knowledge of our own locality to better
understand our Milky Way as a spiral galaxy.
20.2 Measuring the

Milky Way
As noted in earlier chapters, one of the more difficult prac-
tical issues faced by astronomers is determining the dis-
tances to objects in the sky. Such distance measurements
are crucial to determining the sizes of objects. In Chapter 13
we learned how trigonometric parallax is used to measure
the distances to nearby stars, but this method got us out to
distances of only a few hundred light-years. In Chapter 18
we mentioned that Harlow Shapley determined the size of
the Milky Way and we discussed the effect that his mea-
surement had on astronomy. We also discussed Hubble’s
discovery of Cepheid variables in other galaxies, and then
we learned how distances to galaxies are estimated through FIGURE 20. 3 A Hubble Space Telescope image of the globular
the use of standard candles. We turn now to how some of cluster M80.
these standard candles were developed and how they have
been used in the study of the Milky Way.
perspective: we fi nd only about 50 stars within that same
distance from our own Sun. (You might imagine how our
Globular Clusters and the Size sky would appear if our Solar System were located at the
center of a globular cluster, as portrayed so dramatically in
of the Milky Way Galaxy Excursions 20.1.) Globular clusters are therefore much
The key to determining the size of the Milky Way Galaxy denser concentrations of stars than occur on average
turned out to be our finding a type of object, a standard can- throughout our galaxy. Still, they contain only a small per-
dle, whose distance we can measure throughout our galaxy. centage of the Milky Way’s stars; our galaxy has over 100,000
Astronomers found that globular clusters were just what times as many stars as a typical globular cluster has.
they needed. Recall from Chapter 17 that we observe many About one-fourth of the globular clusters in the Milky
stars in large groups called clusters. A globular cluster, such Way reside in or near the disk of our galaxy. The rest of
as the one in Figure 20.3, is a large spheroidal group of stars them occupy a large volume of space surrounding the disk
held together by gravity. Many clusters can be seen through and bulge, referred to as the halo of the Milky Way. Globu-
small telescopes. At fi rst glance they look something like lar clusters offer two distinct advantages that make them
small elliptical galaxies, and the analogy is not a bad one. relatively easy to study. First, they are very luminous and
The motions of stars within a globular cluster are much like so can be easily seen at great distances. Second, because
the motions of stars within an elliptical galaxy. However, many globular clusters lie outside the disk of the Milky
globular clusters are quite different from elliptical galaxies Way, we can see them at great distances without encounter-
in size and in concentration of stars. ing much absorption due to the obscuring dust within the
There are more than 150 cataloged globular clusters in disk. However, simply being able to see globular clusters
our galaxy (and there are likely many more—dust in the does not necessarily put us any closer to measuring their
disk of our galaxy may hide distances. To do that, we need to look at the properties of
them from view). The known Globular clusters are the stars they contain.
globular clusters have lumi- very luminous and This is not the fi rst time that we have been confronted
nosities ranging from a low of easy to identify with the problem of measuring the distance to remote
400 solar luminosities (400 at great distances objects. Recall that the inverse square law (discussed at
L ⊙) to a high of about 1 million within the galaxy. length in Chapter 4) enables us to determine distance when
L ⊙. A typical globular cluster (1) we know the luminosity of a star and (2) we can measure
consists of 500,000 stars packed into a volume of space with its brightness. Rearranging that law for our present purpose,
a radius of only 15 light-years. Let’s put this density into we can calculate the star’s distance:
_____
d=
√ 4πb
L
_____
star
.
star
(a)
Translated into words, this equation tells us that the distance

of a star is proportional to the square root of the ratio of its
luminosity (L star) to its brightness (bstar) as seen from Earth.
Some types of stars are especially useful for determin-
ing distances. These are stars that are very luminous (so
that they can be seen at great
distances) and have known We use stars of
luminosities. Globular clus- known luminosity as
ters, located in the Milky Way standard candles to
and other galaxies, contain measure distances.
good standard candles (Cep-
heid variables) and are also good standard candles them-
selves. Figure 20.4 shows the H-R diagram for the stars in
globular cluster M92. Recall from Chapter 17 that star clus-
ters offer snapshots of stellar evolution. The main-sequence (b) 107
turnoff in this cluster’s H-R diagram occurs for stars with
106
E X C U R S I O N S 2 0 .1
105
Nightfall 104
The location of the main-

Isaac Asimov (1920–1992), the famous writer of science and 103
Luminosity (L£)
sequence turnoff shows

science fiction, imagined what might happen to a civilization that this globular cluster
on a planet orbiting within a system of six stars located in is 13 billion years old.
102
the heart of a giant globular cluster. His short story “Night-
fall” has become one of the more famous works of science
101
fiction. On Asimov’s fictional planet Lagash, at least one of
its six stars is almost always above the horizon. “Nightfall”
occurs on Lagash only once every 2,049 years. The story tells 1
of the great madness that affl icts the inhabitants on this one
night as they recoil in fear from a sky fi lled with hundreds 10–1
of thousands of bright stars.
Asimov’s story leads us to consider how our perspectives 10–2
as human beings are formed by the circumstances in which
we live. Yet we need not go to science fiction to ask these
10–3
kinds of questions. Our perspective on the universe changes 1.3 1 0.8 0.6 0.5 0.4 0.3
with time. At the moment, we are traversing an open, rather bvisual/binfrared Color
dust-free part of the Milky Way. On a moonless night we see FIGURE 20. 4 (a) The globular cluster M92 and (b) an H-R
a dark sky and gaze with our telescopes into a universe of diagram of the stars it contains. (The color plotted on the
galaxies, but that will not always be the case. Just “down the diagram is based on the ratio of visible to infrared light.) The
road,” astronomically speaking, are dark interstellar clouds main-sequence turnoff at 0.8 solar mass (0.8 M⊙) indicates that
and star-forming regions filled with glowing gas and obscur- the cluster is about 13 billion years old.
ing dust through which the Sun and its entourage of planets
will occasionally pass. How different would our view of
the universe be if, instead of a dark sky, we looked up each masses of about 0.8 M⊙, which corresponds to a main-
night and saw a sky filled with a soft green glow, punctu- sequence lifetime of close to 13 billion years. The age of this
ated by a few points of intense light? How much different globular cluster is similar to those of many others, making
would our history be if, at some point during the rise of our them the oldest objects known in our galaxy or in any nearby
civilization, we had suddenly emerged over the course of galaxy. Globular clusters must have formed when the uni-
just a few years from within a molecular cloud and gotten verse and our galaxy were very young. Compared to
our fi rst look at the larger universe? globular-cluster stars, our Sun, at about 5 billion years old,
is a relatively young member of our galaxy.
FIGURE 20. 5 Two stereoscopic views of the

distribution of globular clusters in the Milky Way
Galaxy. Colors represent the average colors of
stars in each cluster. The green cross shows the
location of the Sun. The light gray area represents
the plane of the galaxy. (See Figure 13.1 for
viewing instructions.)
a roughly spherical region of space with a

d ia meter of
about 300,000 Harlow Shapley used
l ig ht-yea r s! globular clusters
These globular to determine the
clusters trace size of our galaxy.
out the halo of
the Milky Way Galaxy, as shown in Figure
20.5, which reflects the modern view of the
globular-cluster distribution.
The globular clusters around the Milky
Way are moving about under the gravita-
tional influence of the galaxy just as stars
in an elliptical galaxy move about under the
gravitational influence of that kind of galaxy.
Symmetry therefore requires that the center
of the distribution of globular clusters coin-
cide with the gravitational center of the gal-
axy itself. Shapley realized that, because he
could determine the distance to the center
In Chapter 17 we also found that there is a region in the of this distribution, he had actually determined the Sun’s
H-R diagram, called the instability strip, in which stars pul- distance from the center of the Milky Way, as well as the
sate. As they do so, their lumi- size of the galaxy itself.
nosity changes. In an old Globular clusters Figure 20.6 identifies the disk, bulge, and halo of the
cluster like a globular cluster, contain RR Lyrae Milky Way. Modern determinations indicate that the Sun
the horizontal branch of the standard candles. is located about 27,000 light-years from the center of the
H-R diagram crosses the insta- galaxy, or roughly halfway out toward the edge of the disk.
bility strip. Recall that horizontal branch stars that lie in Armed with strong evidence, in 1915 Shapley confidently
the instability strip are variable stars called RR Lyrae vari- presented the then-new view of our greatly expanded gal-
ables. RR Lyrae stars in globular clusters are easy to spot axy. It took all of human history up to 1610 to go from an
because they are very luminous (horizontal branch stars Earth-centered universe to a Sun-centered Solar System. It
are giant stars) and because their periodic changes in bright- took 305 years to go from a small galaxy to a large galaxy.
ness signal their identity. As with Cepheid variables, the Yet it was barely a decade later that Edwin Hubble, using
time it takes for an RR Lyrae star to undergo one pulsation Cepheid and RR Lyrae stars and the period-luminosity rela-
is related to the star’s luminosity. Harlow Shapley used tionship, proved that Shapley’s enormous Milky Way is but
Henrietta Leavitt’s determination of this period-luminosity one of billions of galaxies in the universe.
relationship to calculate the luminosities of RR Lyrae stars
in globular clusters. He then used the inverse square law of
radiation to combine these luminosities with measured We Use 21-cm Radiation to
brightnesses to determine the distances to globular clusters.
Finally, Shapley cross-checked his results by noting that
Measure Rotation of the Galaxy
more distant clusters (as measured with his standard can- There is much we would like to know about the Milky Way
dle) also tended to appear smaller in the sky, as expected. apart from its size and the fact that it is a spiral galaxy. But as
Knowing both the distances to globular clusters and we’ve noted many times already, there is a problem: because
where they appear in the sky, Shapley was able to make a we live inside the dusty disk of the galaxy itself, our visible-
three-dimensional map of the distribution of globular clus- light view of the Milky Way is badly obscured. If you go out
ters in space. This map showed that globular clusters occupy on a dark night, away from any streetlights, and look in the
FIGURE 20.6
This diagram of the
disk, bulge, and halo
Galactic halo of the Milky Way
Galaxy also shows
the Magellanic
Clouds and the
300,000 light-years location of the Sun
within the Milky Way
disk.
100,000 light-years
Globular clusters
Galactic bulge
Sun
Galactic disk
27,000 light-years
Magellanic Clouds
Toward galactic center

Redshifted
(moving away)
200
emission from hydrogen
Doppler shift of 21-cm
3 Looking toward galactic center,

we see gas moving neither
toward nor away from us.
Emission is not Doppler-shifted.
0
–200
1 Looking in this direction, we Blueshifted
see gas clouds moving away
(moving toward us)
from us. Emission is redshifted.
–180 –90 0 90 180
Viewing direction from Sun
FIGURE 20.7 Doppler velocities

2 Looking in this direction, we measured from observations of 21-cm
see gas clouds moving toward emission from interstellar clouds of neutral
us. Emission is blueshifted.
£
hydrogen. These velocities vary between
redshift and blueshift as we look around in
the plane of our galaxy. Notice that from our
perspective within the Solar System (sight
Sun’s velocity lines are shown as dashed white lines), we
see the clear signature of a rotating disk
Galactic rotation when looking on either side of the galactic
center.
FIGURE 20.8 From the outside,

the Milky Way would look much like
Irr this barred spiral galaxy, NGC 6744.
Sd SBd
Sc SBc
Sb SBb
1 The Milky Way would
lie here on the Hubble
tuning fork…
Sa SBa
S0 SB0
2 …and look much like
this SBbc galaxy.
E5
are more difficult to interpret at a glance. Even so, observed

E3 velocities of neutral hydrogen enable us to measure our gal-
axy’s rotation curve and even determine the structure pres-
ent throughout the disk of our galaxy.
Figure 20.2b shows an artist’s reconstruction of the Milky
E0
Way based on Spitzer Space Telescope infrared observations
and 21-cm radio data. Our galaxy has a substantial bar with
a modest bulge at its center, as indicated by infrared studies
of the distribution and motions of stars toward the center of
direction of the center of our galaxy (located in the constel- our galaxy. Two major spiral arms—Scutum-Centaurus and
lation Sagittarius), instead of a bright spot you will see a dark Perseus—connect to the ends of the central bar and sweep
lane of dusty clouds, as shown in Figure 20.1a. To probe the through the galaxy’s disk, just like the arms we saw in exter-
structure of our galaxy, we must therefore use long-wave- nal spiral galaxies. Putting all the available information
length infrared and radio radiation that can penetrate the together, we conclude that the Milky Way is a middle-of-the-
disk without being affected much by dust. The most powerful road giant barred spiral. From outside, our galaxy probably
tool for this work is the same 21-centimeter (cm) line from looks much like the galaxy shown in Figure 20.8, and we
neutral interstellar hydrogen that we used in the previous would place the Milky Way about halfway along the right
chapter to measure the rotation of other galaxies. tine of the Hubble tuning fork diagram (see Figure 19.4). In
The velocities of interstellar hydrogen measured from fact, we classify the Milky Way as an SBbc galaxy.
21-cm radiation are plotted in Figure 20.7 as a function of
the direction in which we are looking. Looking in the region
around the center of the gal-
axy, we see that on one side Doppler velocities
The Milky Way Is Mostly Dark Matter
hydrogen clouds are moving measured from 21-cm As we saw in the previous chapter, observations of rotation
toward us while on the other radiation show that curves have led us to conclude that the masses of spiral
side clouds are moving away we live inside a galaxies consist mostly of
from us. This is a pattern that rotating galaxy. dark matter. We can use the Our galaxy has a flat
we have seen before—it is the rotation of our own galaxy to rotation curve.
pattern of the rotation velocity of gas in a disk (see Figures determine whether the Milky
19.15 and 19.16). The only difference is that instead of look- Way is also dominated by dark matter. Figure 20.9 shows
ing at it from outside, we see our own galaxy’s rotation curve the rotation curve of the Milky Way as inferred primarily
from a vantage point located within—and rotating with—the from 21-cm observations. The orbital motion of the nearby
galaxy. In other directions, the velocities we see are compli- dwarf galaxy called the Large Magellanic Cloud provides
cated by our moving vantage point within the disk and so data for the outermost point in the rotation curve, at a dis-
300 FIGURE 20.9 A plot

showing rotation velocity
250 Large
versus distance from the
Rotation velocity (km/s)
Magellanic
center of the Milky Way. The
Cloud
200 most distant point comes from
measurements of the orbit of
150 the Large Magellanic Cloud.
The nearly flat rotation curve
100 The rotation curve of …over 160,000 light-years indicates that dark matter
the Milky Way remains from the center of the galaxy. dominates the outer parts of
nearly flat…
50 our galaxy.
0
0 10 20 30 40 50155 160 165
Distance from center of galaxy (thousands of light-years)
(a) (b)
1 A telescope looks
at a distant star. Massive compact Red light
object
Brightness
Distant
star
Brightness
Time
2 A passing massive object
distorts spacetime… 6 A lensing event affects
all wavelengths equally.
Blue light
Brightness
3 …directing more 4 The star looks Brightness
starlight toward temporarily
the telescope. brighter…
Time
FIGURE 20.10 (a) The light

from a distant star is affected by
a compact object crossing our
line of sight. (b) The observed light
Brightness curves of a real star experiencing
5 … and then
dimmer again as
a lensing event. Gravity affects
the object passes. all wavelengths equally, so real
“lensing events” should look the
same in all colors.
tance of roughly 160,000 light-years from the center of the distribution of dark and normal matter within the Milky
galaxy. As we have come to expect for spiral galaxies, the Way is also much like what we have seen in other galax-
Milky Way has a fairly flat rotation curve. ies. Visible matter dominates the inner part of our galaxy.
By applying the same techniques laid out in the previ- Dark matter dominates its outer parts, at least to a distance
ous chapter to the rotation curve of the Milky Way, we infer of 150,000 light-years from the galaxy’s center.
that the galaxy’s mass must be about 6 × 1011 to 10 × 1011 What is dark matter? Is it exotic, or could it perhaps consist
M⊙. However, if we instead estimate the mass of the Milky of very dim, compact objects of the kind we have discussed
Way by measuring the light from its stars and knowing how so far? There is a very clever way to search for dark matter
much stellar mass is needed to produce that much light within the halo of our own galaxy, when the dark matter con-
(using infrared to see through the dust), we fi nd a much sists of compact objects such as low-mass stars (for example,
lower value. As in other spiral galaxies, we can therefore small main-sequence M stars), planets, white dwarfs, neu-
infer that the Milky Way is mostly dark matter. The spatial tron stars, or black holes. We refer to such dark matter candi-
dates as MACHOs, which stands for “massive compact halo through the disk. Using the ages, chemical abundances, and
objects.” If the dark matter in our halo consists of MACHOs, motions of nearby stars, we can differentiate between disk and
there must be a lot of these objects, and they must each exert halo stars to learn more about the galaxy’s structure.
gravitational force but not emit much light.
How might we detect MACHOs? Because of their grav-
ity, MACHOs can gravitationally deflect light according to We See Stars of Different Age and
Einstein’s general theory of relativity. If we were observing a
distant star and a MACHO passed between us and the star,
Different Chemical Composition
the star’s light would be deflected and perhaps focused by the Stellar ages and their chemical abundances provide the
intervening MACHO as it passed across our line of sight, as most fundamental categories into which populations of
illustrated in Figure 20.10a. Because gravity affects all wave- stars can be grouped. Conveniently, some stars come pre-
lengths equally, such “lensing events” should look the same in packaged into distinct groups that split up along just these
all colors—a fact that rules out other causes of variability. two lines. Recall from Chapter 17 that there are two dif-
We would be remarkably lucky if such an event occurred
just as we were observing a single distant star. However,
astronomers monitored the stars in the Large and Small FIGURE 20.11 (a) The open star cluster NGC 6530, located
Magellanic Clouds (two of the small companion galaxies 5,200 light-years away, in the disk of our galaxy. (b) The H-R
diagram of stars in this cluster shows that it has an age of a few
of the Milky Way Galaxy), observing tens of millions of
million years or less.
stars for several years. They found a number of examples
of events of the sort shown in Figure 20.10b, some believed (a)
to be small main-sequence M stars, but not nearly enough
to account for the amount of dark matter in the halo of our
galaxy. Thus, it was concluded that the dark matter in our
galaxy is probably not composed primarily of MACHOs.
(MACHOs are not to be confused with another dark matter
candidate called “weakly interacting massive particles,” or
WIMPs. Who says astronomers have no sense of humor?)
The cluster is near the

remains of the cloud
20.3 Studying the Milky from which it formed.
Way Galaxy Up Close (b)

The presence of massive,
luminous, hot stars shows that
107
and Personal this open cluster is very young.
106 1,0
Earlier in the chapter we concentrated on the disadvantages 00
R
of our perspective that make it hard to see the Milky Way as £
105
a galaxy. However, our perspective also has advantages. For
example, we can study the stellar content of the Milky Way 10
0R
at very close range, star by star, looking at subtle aspects of 104 £
the populations of stars that give us direct clues about how

spiral galaxies like ours form. For example, stellar orbits 103 10
R
Luminosity (L£)
determine the shapes of the different parts of our galaxy, £
and it is much easier to measure stellar orbits in our own 102

galaxy than it would be to measure them in other galaxies. 1R
£
The stars in the Milky Way’s disk rotate about the center of 101
M
AI
the galaxy, just like the gas and dust in the disk. The stars N
0.1 S
in the halo move in orbits similar to those of stars in ellip-
EQ
1 R
£
UE
tical galaxies. The bar in the bulge of our galaxy is shaped

CE
N
primarily by stars and gas moving both in highly elongated 10–1 0.0
1R
orbits up and down the long axis of the bar and in short £
orbits aligned perpendicular to the bar.

10–2
Our Sun is a middle-aged disk star located among other 0.0
01
middle-aged stars that orbit around the galaxy within the disk. R
£
10–3
Yet near the Sun are other stars, usually much older, that are 40,000 30,000 10,000 6,000 3,000
a part of the galactic halo and whose orbits are carrying them Surface temperature (K)
ferent varieties of star clusters in the Milky Way. Globular galaxy, star formation seems not to have gotten started until
clusters—the densely packed collections of millions of stars later, but it has been continuing ever since. The process that
studied by Shapley—are mostly halo objects. With ages of formed the stars in the massive, compact globular clusters
up to 13 billion years, they are among the oldest objects must have also been much different from the more sedate
known. In contrast, open clusters, like the one shown in process responsible for creating stars in the less massive,
Figure 20.11a, are much less tightly bound collections of a more scattered open clusters.
few dozen to a few thousand stars that are found orbiting in It is obvious why we would be so interested in grouping
the disk of our galaxy. As with globular clusters, the stars stars according to their ages, but why would we focus on
in an open cluster all formed in the same region at about the chemical composition of stars? We have discussed the
the same time. When we study the H-R diagrams of open chemical evolution of the universe on a number of occasions.
clusters (Figure 20.11b), we fi nd a wide range of ages. Some When the universe was very young, only the least massive
open clusters contain the very youngest stars known. Other of elements existed. All elements more massive than boron
open clusters contain stars that are somewhat older than must have formed by nucleosynthesis in stars. For this rea-
the Sun. There is no overlap in age between open clusters son, the abundance of massive elements in the interstellar
and globular clusters, however. Even the youngest globular medium provides a record of the cumulative amount of star
clusters are several billion years older than the oldest open formation that has taken place up to the present time. Gas
clusters. Open clusters do not survive long in the disk of our that shows large abundances of massive elements must have
galaxy, because they are loosely bound together and easily gone through a great deal of stellar processing (see Chapters
disrupted by the gravitational tug from nearby objects. 13 and 17), whereas gas with low abundances of massive
The differences in ages between globular and open clus- elements is more pristine.
ters immediately tell us something interesting about the In turn, the abundance of massive elements in the atmo-
history of star formation in our galaxy. Stars in the halo sphere of a star provides a snapshot of the chemical com-
formed fi rst, but this epoch of star formation did not last position of the interstellar medium at the time the star
long. No young globular clusters are seen. In the disk of the formed. (In main-sequence stars, material from the core
Interstellar medium
Matter moves from the interstellar

medium into stars and back again…
Supernovae Star formation

and stellar
mass loss
…in a progressive cycle that
has enriched today’s universe
with massive elements.
Stellar
nucleosynthesis
Red and green dots represent
massive elements formed in stars.
Big Bang
Emission lines from Absorption lines from Emission lines from Absorption lines from
hydrogen and helium. hydrogen and helium. massive elements. massive elements.
Intensity
Intensity
Intensity
Intensity
Wavelength Wavelength Wavelength Wavelength
1 Only the least massive elements, 2 Massive elements are 3 …and then ejected into space 4 …chemically enriching the
hydrogen and helium, were formed formed in stars… in supernovae, planetary interstellar medium and
when the universe was born. nebulae, and stellar winds… subsequent generations of stars.
does not mix with material in after generation of disk stars have further enriched the
the atmosphere, so the abun- Abundances of interstellar medium with the products of their nucleosyn-
dances of chemical elements massive elements thesis. Within the disk, younger stars typically have higher
inferred from the spectra of a record the cumulative abundances of massive elements than do older stars. Simi-
star are the same as the abun- history of star larly, older stars in the outer parts of our galaxy’s bulge
dances in the interstellar gas formation. have lower massive-element abundances than do young
from which the star formed.) stars in the disk. Such lower abundances of heavy elements
As illustrated in Figure 20.12, the chemical composition of characterize not only globular-cluster stars, but all of the
a star’s atmosphere reflects the cumulative amount of star stars in our galaxy’s halo, where globular-cluster stars com-
formation that has occurred up to that moment. prise only a minority among the total number of stars in
If our ideas about the chemical evolution of the universe the galactic halo.
are correct, we expect to see large differences in massive- Within the galaxy’s disk, we can even see differences
element abundances between in abundances of massive elements from place to place that
globular and open clusters. Younger stars typically are related to the rate of star formation in different regions.
Stars in globular clusters, have higher massive- Star formation is generally more active in the inner part of
being among the earliest stars element abundances the Milky Way than in the outer parts. This higher level of
to form, should contain only than older stars have. activity is a result of the denser concentrations of gas that
very small amounts of mas- are found in the inner galaxy. If such activity has contin-
sive elements. And that is exactly what we see. Some ued throughout the history of our galaxy, we might predict
globular-cluster stars contain only 0.5 percent as much of massive elements to be more abundant in the inner parts of
these massive elements as our Sun has. This relationship our galaxy than in the outer parts. Observations of chemi-
between age and abundances of massive elements is evident cal abundances in the interstellar medium, based both on
throughout much of the galaxy. The chemical evolution of interstellar absorption lines in the spectra of stars and on
the Milky Way has continued within the disk as generation emission lines in glowing H II regions, confi rm this predic-
FIGURE 20.12 As
subsequent generations of
stars form, live, and die, they
enrich the interstellar medium
with massive elements—
the products of stellar
nucleosynthesis. The chemical
evolution of our galaxy and
other galaxies can be traced
in many ways, including by
the strength of interstellar
emission lines and stellar
absorption lines.
Intensity
Intensity
Wavelength Wavelength
5 With each subsequent generation of 6 Evidence of progressive chemical 7 …and in the strength of massive-element
stars, the amount of massive elements enrichment is found in interstellar absorption lines in spectra of stars formed
in the universe has increased. emission from massive elements… in different places and at different times.
FIGURE 20.13 In the

“galactic fountain” model of the
disk of a spiral galaxy, gas is
pushed away from the plane of
the galaxy by energy released by
young stars and supernovae and
then falls back onto the disk.
Galactic
“fountain”
“Chimney” wall
Dense molecular Newly formed

clouds massive stars Disk of galaxy
Energy from young stars …through which hot gas

and supernovae may blow and UV radiation escape
“chimneys” out of the disk… into the galaxy’s halo.
tion. As expected, there is a smooth decline in abundances the very oldest globular-cluster stars contain some amount
of massive elements from the inner to the outer parts of of massive chemical elements. The implication is clear:
the disk. Astronomers have documented similar trends in globular-cluster stars and
other galaxies. These trends can be seen with stars as well. other halo stars were not the Generations of stars
Within a galactic disk, relatively old stars near the center fi rst stars in our galaxy to must have formed
of a galaxy often have greater massive-element abundances form. There must have been at even before globular
than young stars in the outer parts of the disk have. least one generation of mas- clusters did.
Our basic idea about higher massive-element abun- sive stars that lived and died,
dances following the more prodigious star formation in ejecting newly synthesized massive elements into space,
the inner galaxy seems correct; but as always, the full pic- before even the oldest globular clusters formed. Further,
ture is not this simple. The chemical composition of the every star less massive than about 0.8 M⊙ that ever formed
interstellar medium at any location depends on a wealth of is still around today. Even so, we find no disk stars with
factors. New material falling into the galaxy might affect exceptionally low massive-element abundances. We would
interstellar chemical abundances. Chemical elements pro- have found these stars by now if they existed. The gas that
duced in the inner disk might be blasted into the halo in wound up in the plane of the Milky Way must have seen a
great “fountains” powered by the energy of massive stars significant amount of star formation before it settled into
(as illustrated in Figure 20.13), only to fall back onto the the disk of the galaxy and made stars.
disk elsewhere. Past interactions with other galaxies might We have placed great emphasis on variations in chemi-
have stirred the Milky Way’s interstellar medium, mixing cal abundances from place to place. These variations tell
gas from those other galaxies with our own. The variation us a lot about the history of our galaxy and a lot about the
of chemical abundances within the Milky Way and other origin of the material that we are made from. It is impor-
galaxies—and what these variations tell us about the his- tant to remember, however, that even a chemically “rich”
tory of star formation and nucleosynthesis—remain active star like the Sun, which is made of gas processed through
topics of research. approximately 9 billion years of previous generations of
Although the details are complex, there are several clear stars, is still composed of less than 2 percent massive ele-
and important lessons to be learned from patterns in mas- ments. Luminous matter in the universe is still dominated
sive-element abundances in the galaxy. The first is that even by hydrogen and helium formed long before the fi rst stars.
toward the midplane of the disk of a spiral galaxy. Although

Looking at a Cross Section stars are free to pass back and forth from one side of the
disk to the other, cold, dense clouds of interstellar gas
through the Disk settle down into the central plane of the disk. These clouds
The youngest stars in our galaxy are most strongly con- are seen as the concentrated
centrated in the galactic plane, defi ning a disk about 1,000 dust lanes that slice the disks Gas tends to
light-years thick (but over 100,000 light-years across—thin of spiral galaxies (as shown concentrate into
indeed). The older population of disk stars, distinguish- in Figure 20.1), like a layer of thin sheets.
able by lower abundances of massive elements, has a much bologna in a very thin, fl at
“thicker” distribution, about 12,000 light-years thick. Figure sandwich. This thin lane slicing through the midplane of
20.14 illustrates how the population of stars changes with the disk is the place both where new stars form and where
distance from the galactic plane. Not too surprisingly, the new stars are found.
youngest stars are concentrated closest to the plane of the The interstellar medium is a dynamic place—energy
galaxy for the simple reason that this is where the molecu- from star-forming regions can shape it into impressively
lar clouds are. Older stars make up the thicker parts of the large structures. We mentioned earlier that energy from
disk. There are two hypotheses for the origin of this thicker regions of star formation can impose interesting structure
disk. One suggests that these stars formed in the midplane on the interstellar medium, clearing out large regions of gas
of the disk long ago but have since been kicked up out of the in the disk of a galaxy. Many massive stars forming in the
plane of the galaxy, primarily by gravitational interactions same region can blow “chimneys” out through the disk of
with massive molecular clouds (see Figure 20.14). The other the galaxy via a combination of supernova explosions and
hypothesis suggests that these stars were acquired from the strong stellar winds. If enough massive stars are formed
merging process that formed our Milky Way Galaxy. together, sufficient energy may be deposited to blast holes
When we discussed the formation of accretion disks in all the way through the plane of the galaxy. In the process,
Chapters 6 and 15, we found that a rotating cloud of gas dense interstellar gas can be thrown high above the plane
naturally collapses into a thin disk as a consequence of of the galaxy (see Figure 20.13). Maps of the 21-cm emis-
gas falling from one direction (above the disk) running sion from neutral hydrogen in our galaxy and visible-light
into gas falling from the other direction (below the disk.) images of hydrogen emission from some edge-on external
Clouds of gas cannot pass through each other, so the col- galaxies show a wealth of vertical structures in the inter-
liding gas clouds instead settle into a disk. The same pro- stellar medium of disk galaxies. These vertical structures
cess applies to clouds of gas that are pulled by gravity are often interpreted as the “walls” of chimneys.
FIGURE 20.14 A vertical

profile of the disk of the Milky
Way Galaxy. Gas and young stars
are concentrated in a thin layer
in the center of the disk. Older
4 populations of stars make up the
thicker portions of the disk.
Newly formed stars (blue)

and molecular clouds
Thousands of light-years
2 Older stars (red) have

concentrated in a thin disk. had time to move away
from the midplane and
form a thicker disk.
Mid-
plane
4
are not a form of electromagnetic radiation. (They were

The Halo Is More Than named before their true nature was known.) Rather, they
are charged particles moving at nearly the speed of light.
Globular Clusters Cosmic rays are continually hitting Earth.
Earlier, when discussing the halo of our galaxy, we con- Most cosmic-ray particles are protons, but some are
centrated our attention on its globular clusters. These are nuclei of helium, carbon, and other elements produced by
very important to astronomy, in part because they tell us nucleosynthesis. A few are high-energy electrons and other
a great deal about the history of star formation in the halo. subatomic particles. What makes cosmic rays especially
Yet globular clusters account for only about 1 percent of the fascinating is that they span an enormous range in particle
total mass of stars in the halo. As halo stars fall through the energy. We can observe the lowest-energy cosmic rays using
disk of the Milky Way, some pass close to the Sun, giving us interplanetary spacecraft. These cosmic rays have energies
a sample of the halo that we can study at close range. as low as about 10 –11 joules (J), which corresponds to the
Because most of the stars near our Sun are, like the Sun, energy of a proton moving at a velocity of a few tenths the
disk stars, you might be wondering how we can tell them speed of light. In contrast, the most energetic cosmic rays
apart from nearby halo stars. They are distinguishable in are 10 trillion (1013) times as energetic as the lowest-energy
two ways. First, most halo stars have much lower abun- cosmic rays. These high-energy cosmic rays are known from
dances of massive elements than disk stars have. Second, the showers of elementary particles that they cause when
halo stars appear to be whizzing by us at high velocities. crashing through Earth’s atmosphere.
Actually, it is often not the halo stars that are moving fast, Astronomers hypothesize that cosmic rays are acceler-
but we who are moving fast relative to them. The reason ated to these incredible energies by the shock waves pro-
is that halo stars do not rotate about the center of the gal- duced in supernova explosions. The highest-energy cosmic
axy in the same way that disk stars do, but rather are in rays are more difficult to explain. These are as much as a
the same kind of odd-shaped orbits as stars in elliptical billion times more energetic
galaxies. In contrast, the disk stars near the Sun are mov- than any particle ever pro- Some cosmic rays
ing, mostly together, in a 220-kilometer-per-second (km/s) duced in a particle accelerator have enormously
rotation about the center of our galaxy. Just as it is easy to on Earth. These cosmic rays high energies.
tell the difference between a person sitting beside you on have energies up to 100 J, cor-
a bus and the people who (in your frame of reference) are responding to the energy of a proton moving at a velocity
whizzing by outside the window, it is easy to tell the dif- that is 99.9999999999999999999999 percent of the speed of
ference between disk stars that share our motion and halo light. To get a better sense of just how much energy we’re
stars that, to us, are just passing through. talking about, consider this: if you were to drop your copy
Reflecting once again the bias of the observer, astrono- of 21st Century Astronomy (this irreplaceable textbook) from
mers call the halo stars high-velocity stars, even though it the ceiling, the energy it would have when it hit the floor is
is usually the disk stars that are moving faster relative to about the same as the energy concentrated into a single
the galaxy as a whole! Enough halo stars are near us that high-energy cosmic-ray proton!
we can measure their distances and map out their orbits, Most cosmic rays are trapped within our galaxy by mag-
presenting a very detailed look at the kinds of orbits that netic fields. The interstellar medium of the Milky Way is
stars in a halo (or in an elliptical galaxy) have. The picture laced with measurable mag-
we get is complex. About half the halo stars are orbiting in netic fields that are wound up Cosmic rays are
the same direction as the disk rotation, while the other half and strengthened by the rota- trapped by the Milky
are moving in the opposite direction. We can deduce the tion of the galaxy’s disk. The Way’s magnetic field.
complete orbits of the halo stars near us by measuring their interstellar magnetic field,
motions. Such orbits suggest that halo stars fill a volume however, is weak compared to Earth’s magnetic field—a
of space similar to that occupied by the globular clusters million times weaker. When we discussed Earth’s magne-
in the halo. Halo stars and globular clusters both paint the tosphere in Chapter 8, we saw that charged particles spiral
same picture of the halo of our galaxy. around magnetic fields, with a net motion along the field
rather than across it. The other side of this coin is that mag-
netic fields cannot freely escape from a cloud of gas contain-
Magnetic Fields and Cosmic ing charged particles. Earlier we mentioned that dense
clouds of interstellar gas are concentrated by gravity in the
Rays Fill the Galaxy midplane of the Milky Way. The mass of these clouds
Almost all that we know about the galaxy we learn from anchors the galaxy’s magnetic field and its charged-particle
analyzing the electromagnetic radiation it emits—almost, cosmic rays to the disk, as shown in Figure 20.15.
but not quite, all. An important component of the galaxy’s Are all cosmic rays confi ned to our galaxy? The disk of
disk, whose energy we can measure directly on Earth, is our galaxy glows from synchrotron radiation produced by
made up of cosmic rays. Despite their name, cosmic rays cosmic rays (mostly electrons) spiraling around the galaxy’s
20.4 The Milky Way Hosts a Supermassive Black Hole 577
Disk of galaxy
1 Gravity keeps dense 2 Magnetic fields are

molecular clouds near anchored in molecular Synchrotron radiation
the galaxy’s midplane. clouds.
Disk of galaxy
Dense molecular Magnetic field

cloud
Cosmic rays
3 Energetic cosmic rays 4 In this way the weight

are trapped by the of molecular clouds holds
galaxy’s magnetic field. cosmic rays to the galaxy.
FIGURE 20.15 The weight of interstellar clouds anchors the magnetic field of the Milky Way to
the disk of the galaxy. The magnetic field in turn traps the galaxy’s cosmic rays, much as a planetary
magnetosphere traps charged particles.
magnetic field. Such synchrotron emission is seen in the

disks of other spiral galaxies as well, telling us that they,
20.4 The Milky Way Hosts
too, have magnetic fields and populations of energetic cos- a Supermassive
mic rays. Even so, the very highest-energy cosmic rays are
moving much too fast to be confined to our galaxy. Any such Black Hole
cosmic rays formed in the Milky Way soon stream away
from the galaxy into intergalactic space. Thus, although we We began this chapter by pointing out that dense clouds of
do not currently understand their origin, it is likely that a dust and gas hide our visible-light view of the Milky Way’s
fraction of the energetic cosmic rays reaching Earth origi- center. Yet it is here that we might expect to find a super-
nated in energetic events outside our galaxy. massive black hole, as is often found in other galaxies with
The total energy of all of the cosmic rays in the galactic central bulges. Fortunately, we can overcome the obscura-
disk can be estimated from the energy of the cosmic rays tion problem by utilizing the penetrating ability of X-ray,
reaching Earth. The strength of the interstellar magnetic infrared, and radio radiation. Figure 20.16 shows images
field can be measured in a variety of ways, including the of the Milky Way’s center taken with the Chandra X-ray
effect that it has on the properties of radio waves passing Observatory and the Spitzer Space Telescope. The X-ray
through the interstellar medium. These measurements indi- view (Figure 20.16a) shows the location of a strong radio
cate that in our galaxy, the magnetic-field energy and the source called Sagittarius A* (abbreviated Sgr A*), which
cosmic-ray energy are about equal to each other. Both are astronomers believe lies at the exact center of the Milky
comparable to the energy present in other energetic compo- Way. The infrared image (Figure 20.16b) cuts through the
nents of the galaxy, including the motions of interstellar gas dust to reveal the crowded, dense core of the galaxy con-
and the total energy of electromagnetic radiation within the taining hundreds of thousands of stars.
galaxy. Magnetic fields and cosmic rays are not bit players Studies of the motions of stars closest to the Sgr A*
in the production we call the Milky Way. source suggest a central mass very much greater than
(a) (b)
FIGURE 20.16 (a) An X-ray view of the Milky Way’s central region
showing the active source, Sgr A*, as the brightest spot at the middle
of the image. Lobes of superheated gas (shown in red) are evidence of
recent, violent explosions happening near Sgr A*. (b) This wide infrared view
(measuring 640 × 890 light-years) of the central core of the Milky Way
shows hundreds of thousands of stars. The bright white spot at the far right
marks the galaxy’s center, home of a supermassive black hole.
recently, astronomers have studied the orbits of stars less

than 0.1 light-year from the galaxy’s center. These stars fol-
low Kepler’s laws as they orbit about the center, and they
are so close that it takes them only a matter of a dozen or
so years to complete an orbit! This means we can observe
the positions of these stars changing noticeably over time,
and we can see them speed up as they whip around what
can only be a supermassive black hole at the focus of their
elliptical orbits (Figure 20.17). Using Kepler’s third law,
we can then estimate that the black hole at the center of
our own galaxy is a relative lightweight, having a mass of
“only” 3.7 × 106 M⊙.
Although invisible to our eyes, the central region of our
Milky Way Galaxy is awash in radio, infrared, X-ray, and
gamma radiation. Radio obser-
vations (Figure 20.18) reveal The center of our
synchrotron emission from galaxy shows rapid
fascinating wisps and loops rotation and
of material dist ributed AGN-like emission.
throughout the region. This is
FIGURE 20.17 Orbits of several stars within 0.1 light-year reminiscent of the synchrotron emission seen from active
(about 6,000 astronomical units, or AU) of the Milky Way’s galactic nuclei (AGNs), except that this emission is seen at
center. The Keplerian motions of these stars reveal the presence far lower levels.
of a 3.7-million-M⊙ supermassive black hole at the galaxy’s center. Clouds of interstellar gas at the galaxy’s center are heated
Colored dots show the measured positions of the stars over a to millions of degrees by shock waves from supernova explo-
12-year interval. sions and colliding stellar winds blown outward by young
massive stars. Superheated gas produces X-rays, and the
Chandra X-ray Observatory has detected more than 2,000
that of the few hundred stars orbiting there. Furthermore, X-ray sources within the central region of the galaxy. These
observations of our galaxy’s rotation curve also show include frequent, short-lived X-ray flares near Sgr A* (Fig-
rapid rotation velocities very close to its center. This is ure 20.19), which provide direct evidence that matter fall-
the same sort of evidence that led us to surmise the pres- ing toward the supermassive black hole fuels the energetic
ence of supermassive black holes in other galaxies. More activity at the galaxy’s center. Huge lobes of 20-million-K
20.5 The Milky Way Offers Clues about How Galaxies Form 579
FIGURE 20.18 Radio

observations of the center
of the Milky Way reveal
wispy molecular clouds
(purple) glowing from strong
synchrotron emission. Cold
dust (20–30 kelvins, or K)
associated with molecular
clouds is shown in orange.
Diffuse infrared emission
appears in blue-green. The
galactic center (Sgr A*) lies
within the bright area to the
right of center.
cano, however, what we are seeing in the inner Milky Way

is a reminder that our galaxy was almost certainly “active”
in the past and, like an episodic volcano, the beast could
become active once again.
20.5 The Milky Way Offers

Clues about How
Galaxies Form
One of the fundamental goals of stellar astronomy is to
understand the life cycle of stars, including how stars form
from clouds of interstellar gas. In Chapter 15 we were able
to tell a fairly complete story of this process, at least as it
occurs today, and tie this story strongly to observations of
our galactic neighborhood. Galactic astronomy has the same
basic goal. That is, astronomers would like very much to have
a complete and well-tested theory of how our galaxy formed.
Unfortunately, such a complete theory is not yet in hand.
FIGURE 20.19 The central region of the Milky Way is also a
source of strong X-rays, with evidence of the “beast” at its very
Even so, what we have seen so far offers us many clues.
center. The bright, pointlike feature in the middle is a huge X-ray Important among these clues are the properties of globu-
flare that occurred within 1 AU of the supermassive black hole. lar clusters and high-velocity stars in the halo of the galaxy.
The flare developed within a matter of minutes and declined to For reasons discussed earlier, these objects must have been
previous levels after 3 hours. among the fi rst stars formed
that still exist today. The fact At least one
that they are not concentrated generation of stars
gas surrounding Sgr A* (see Figure 20.16a) bear witness to in the disk or bulge of the gal- had to exist before
several such explosions that have occurred within the past axy says that they formed today’s halo stars
10,000 years. Nevertheless, this activity is not as intense as from clouds of gas well before were formed.
that seen in other galaxies with central black holes. What those clouds had settled into
does this mean? It tells us that today our supermassive black the galaxy’s disk. The observations that globular clusters
hole is starving. Without a source of large amounts of mate- are very old and that the youngest cluster is older than the
rial to feed the beast, the Milky Way’s center is relatively oldest disk stars agree with this hypothesis. The presence
quiet. Like steam rising from the cauldron of a dormant vol- of small amounts of massive elements in the atmospheres
of halo stars also tells us that at least one generation of stars and speeds of the dwarf galaxies about the Milky Way can
must have lived and died before the formation of the halo lead to new estimates of the dark matter mass within the
stars we see today. We have yet to fi nd any stars from that Milky Way itself.
fi rst generation in our galaxy today. The remainder of these protogalaxies merged to form
From these and other clues, we have come to understand the barred spiral galaxy we call the Milky Way. In this pro-
that our galaxy must have formed when the gas within a cess, stars were formed in the
huge “clump” of dark matter collapsed into a large number halo, in the bulge, and in the Our galaxy formed
of small protogalaxies. Some of these smaller clumps are disk. The fi rst stars to form from the mergers of
still around today in the form of small, satellite dwarf gal- ended up in the halo—some many smaller
axies near our own. The larger among them are the Large in globular clusters, but many protogalaxies.
and Small Magellanic Clouds (Figure 20.20), which are eas- not. Gas that settled into the
ily seen by the naked eye and appear much like detached disk of the Milky Way quickly formed several generations
pieces of the Milky Way. Among the closest is the Sagit- of stars. This process resulted in the formation of a dense,
tarius Dwarf, which at this moment is plowing through concentrated mass that quickly grew into the supermassive
the disk of the Milky Way on the other side of the bulge. It black hole at the center of the galaxy. The details of this
seems likely that at some point the Sagittarius Dwarf will process are sketchy, but some calculations indicate that so
become incorporated into the Milky Way—an indication much mass was concentrated in this small region that
that our galaxy is still growing. There are more than 20 of almost any sequence of events would have led to the forma-
these satellite dwarf galaxies, although we can’t be sure tion of a massive black hole.
that all are gravitationally bound to the Milky Way. Many We need to be careful not to get ahead of ourselves. The
were discovered only very recently, because of their low Milky Way offers many clues about the way galaxies form,
luminosity and the possibility that they are dominated by but much of what we know of the process comes from look-
an even greater fraction of invisible dark matter than we ing beyond our local system. Images of distant galaxies
fi nd in other known galaxies. Observations of the motions (which we see as they existed billions of years ago), as well
as observations of the glow left behind by the formation of
the universe itself, provide equally important pieces of the
puzzle. We will have to leave the story of galaxy formation
as we currently understand it unfi nished, for now, as we
Milky Way Galaxy turn our attention to the immensely larger structure of the
universe as a whole.
LMC SMC FIGURE 20. 20 Our Milky Way is surrounded by more than
20 dwarf companion galaxies, the largest among them being
Magellanic
Clouds the Large and Small Magellanic Clouds. The Magellanic Clouds
were named for Ferdinand Magellan (c. 1480–1521), who headed
the first European expedition that ventured far enough into the
Southern Hemisphere to see them.
Large Magellanic Cloud Small Magellanic Cloud

about the stars in globular and open clusters. These col-

Seeing the Forest for the Trees lections of stars not only tell us about the Milky Way as
a galaxy, but also provide the data we use to test and
Many books of puzzles contain optical illusions in refi ne our ideas about the way stars everywhere evolve.
which familiar patterns lie hidden in jumbles of lines In the Milky Way we see stars in the bulge, stars in the
and shapes. When looking at such a picture, at fi rst we thin and thicker portions of the disk, and stars in the
see nothing but confusion. Then in a flash of insight we halo that are falling through the disk near us. We mea-
suddenly see the pattern! Once the pattern is recognized, sure their properties—luminosity, size, temperature,
our perception shifts: a moment earlier all we saw was a and abundance of the chemical elements of which they
mess; but now when we look at the picture, the “totally are made—and we learn what each variety of stars in a
obvious” pattern jumps off the page at us. Such is the case spiral galaxy is like. Spiral arms in the Milky Way are
with observing our galaxy. Go outside on a dark, moon- not just bright swirls against the background of a spi-
less summer night, away from the city lights, and stare at ral disk. Rather, they are collections of young stars, old
the sky. In addition to the dazzling jewels of individual stars, and clouds of gas and dust that can be studied at
stars that fill your view, you will see a faint, patchy river close range. We know that stars form in the arms of other
of light running from horizon to horizon. This river—this spiral galaxies because we see the brightest of those
“Milky Way”—has been known since earliest prehistory. stars, along with the generally blue color of the arms.
Even so, its nature was unknown until a remarkably We know that stars form in the spiral arms of the Milky
short time ago. As recently as the early 20th century, it Way because that is where we see dense molecular-cloud
was unclear how this river of light fit into models of our cores containing individual young stars surrounded by
stellar system based on counts of individual stars. Yet their accretion disks.
once the answer is known, there it is for all to see—how It is difficult to study the Milky Way as a galaxy, but
could we ever have been so blind as to miss the obvious? it is worth the investment. In our study of the grand pat-
Stretched across our sky lies an unmistakable spiral gal- terns within the Milky Way, we are guided by what we
axy, as viewed edge on from within the disk. know of other galaxies; conversely, in the Milky Way we
Our conclusion from this fl ash of insight is con- can learn the details of processes that are at work shap-
fi rmed by the results of countless observational stud- ing all of those other islands of activity spread out across
ies. Whether we look at the distribution of globular the expanse of the universe. These details are beginning
clusters, 21-cm observations of the motions of hydro- to help us clarify our understanding of some of the “big”
gen gas, or any number of other tracers of structure, the questions about the way galaxies form and evolve.
answer is always the same. The Milky Way is a giant Our location in the Milky Way may not give us the
“plain vanilla,” dark matter–dominated, barred spiral best perspective from which to study our galaxy, but it
galaxy with a supermassive black hole at its center, just is far from the worst. Fate has been kind. Since the Sun
like countless others. The Sun and Earth are located in and Earth formed, they have made numerous trips around
a relatively open and clear region of interstellar space the galaxy, passing back and forth through the disk and
partway out in the galaxy’s disk. encountering a wide range of environments. What if
Our vantage point within the Milky Way is hardly the our species had become aware at a moment when Earth
best from which to answer questions about global proper- was in the depths of a molecular cloud and the sky held
ties of our galaxy. If ever there was a problem in astron- nothing of interest other than the Sun and planets? That
omy of not being able to see the forest for the trees, this would have brought a whole new and very literal mean-
is it. For example, we can determine the morphological ing to the notion of the “dark ages.”
classification of the Andromeda Galaxy by glancing at a In some sense this portion of our journey has been
single picture. In contrast, obtaining a reliable and com- about context. We have learned to see stars within the
plete classification of our own Milky Way is a task that context of the history of the Milky Way. We have seen
has taken decades of detailed study. The classification of star-forming molecular clouds within the context of the
the Milky Way as an SBbc galaxy was difficult to arrive sweep of spiral arms. We have learned to see the Milky
at and remains at least somewhat uncertain today. Way within the context of its similarities with other gal-
On the other hand, the Milky Way gives us the ulti- axies. On the next leg of our journey we will pose the
mate insider’s view. For example, stars in the Milky Way ultimate question of context as we learn to see galaxies,
are not just part of a large blur, but are instead individu- including our Milky Way, within the context of the uni-
als that can be studied as such. Here is where we learn verse itself.
Summary b. Why do globulars have such high masses while open

clusters have low masses?
c. Are these issues related? Explain.
• We live in the disk of a barred spiral SBbc galaxy called
the Milky Way, which is 100,000 light-years across. *8. Old stars in the inner disk of our galaxy have higher
abundances of massive elements than do young stars in
• We use variable stars of known luminosity to find the
the outer disk. Shouldn’t old stars always have a lower
distances to globular clusters, which enable us to mea-
abundance of massive elements than young stars have?
sure the size of the Milky Way’s extended halo.
Explain your reasoning.
• The Sun is about 27,000 light-years from the Milky Way’s
9. Do you think it is likely that astronomers will fi nd a
center.
young star that has no heavy elements in it? Why or why
• The Doppler velocities of radio lines show that the rota- not? How about an old star?
tion curve of the Milky Way is flat, like those of other
10. How do 21-cm radio observations reveal the rotation of
galaxies, and that the mass of our galaxy is mostly in
our galaxy?
the form of dark matter.
11. What does the rotation curve of our galaxy say about the
• The chemical composition of the Milky Way has evolved
presence of dark matter in the galaxy?
with time, and there must have been a generation of stars
before the oldest halo and globular-cluster stars we see 12. Explain the observational evidence that shows we live
today formed. in a spiral galaxy, not an elliptical galaxy.
• Star formation is actively occurring in the disk of our 13. In the Hubble scheme for classifying galaxies, how is
galaxy, leading to complex structures within the disk. the Milky Way classified?
• At the center of the galaxy is a massive black hole, which 14. What does the abundance of a star’s massive elements
produces rapid orbital velocities nearby. tell us about the age of the star?
15. Where do we fi nd the youngest stars in our galaxy?
SmartWork, Norton’s online homework system, 16. Halo stars are found in the vicinity of the Sun. What
includes algorithmically generated versions of these observational evidence distinguishes them from disk
questions, plus additional conceptual exercises. If stars?
17. Are cosmic rays a form of electromagnetic radiation?
18. Can a cosmic ray travel at the speed of light? Why or why
Student Questions not?
*19. What is one source of synchrotron radiation in our gal-
axy, and where is it found?
1. What can you see for yourself in the night sky (unaided
20. Why must we use X-ray, infrared, and 21-cm radio obser-
by a telescope) that suggests we live in a spiral galaxy?
vations to probe the center of our galaxy?
2. Why is it so difficult for astronomers to get an overall
21. What is Sgr A* and how was it detected?
picture of the structure of our galaxy?
22. Explain the evidence for a supermassive black hole at
3. What do astronomers mean by standard candle?
the center of the Milky Way.
4. Why are globular clusters so useful as standard candles
23. How does the mass of the supermassive black hole at the
in determining the size of the galaxy?
center of our galaxy compare with that found in most
5. Describe the distribution of globular clusters within the other spiral galaxies?
galaxy and what that implies about the size of the galaxy
24. To observers in Earth’s Southern Hemisphere, the Large
and our distance from its center.
and Small Magellanic Clouds look like detached pieces
6. How do we know that the stars in globular clusters are of the Milky Way. What are these “clouds,” and why is
among the oldest stars in our galaxy? it not surprising that they look so much like pieces of
the Milky Way?
7. Compare globular and open clusters.
a. What are the main differences in the gas out of which *25. What is the origin of the Milky Way’s satellite
globular and open clusters formed? galaxies?
26. What has been the fate of most of the Milky Way’s satel- 34. Given what you have learned about the distribution of
lite galaxies? massive elements in the Milky Way, and what you know
about the terrestrial planets, where do you think such
27. Why are most of the Milky Way’s satellite galaxies so
planets are most likely and least likely to form?
difficult to detect?
35. A cosmic-ray proton is traveling at nearly the speed of
28. Use your imagination to describe how our skies might
light (3 × 108 meters per second [m/s]).
appear if our Sun and Solar System were located
a. Using Einstein’s familiar relationship between mass
a. Near the center of the galaxy.
and energy (E = mc2), show how much energy (in
b. Near the center of a large globular cluster.
joules) the cosmic-ray proton would have if m were
c. Near the center of a large, dense molecular cloud.
based only on the proton’s rest mass (1.7 × 10 –27 kilo-
grams [kg]).
b. The actual measured energy of the cosmic-ray pro-
A P P LY I N G T H E C O N C E P T S ton is 100 J. What, then, is the relativistic mass of the
cosmic-ray proton?
29. The Sun completes one trip around the center of the
c. How much greater is the relativistic mass of this
galaxy in approximately 230 million years. How many
cosmic-ray proton than the mass of a proton at rest?
times has our Solar System made the circuit since its
formation 4.6 billion years ago? **36. One of the fastest cosmic rays ever observed had a speed
of (1.0 minus 10 –24) × c. Assume that the cosmic ray and
30. The Sun is located about 27,000 light-years from the cen-
a photon left a source at the same instant. To a stationary
ter of the galaxy, and the galaxy’s disk probably extends
observer, how far behind the photon would the cosmic
another 30,000 light-years farther out from the center.
ray be after traveling for 100 million years?
Assume that the Sun’s orbit takes 230 million years to
complete. 37. Consider a black hole with a mass of 5 million solar
a. With a truly flat rotation curve, how long would it take masses (M⊙ = 2 × 1030 kg.) A star orbiting the black hole
a globular cluster located near the edge of the disk to has a semimajor axis of 0.02 light-year (1.9 × 1014 meters).
complete one trip around the center of the galaxy? Using the universal gravitational constant, G = 6.67 ×
b. How many times has that globular cluster made the 10 –11 newton times square meters per square kilogram
circuit since its formation about 13 billion years (N m2/kg2), calculate the star’s orbital period. (Hint: For
ago? this and the questions that follow, you may want to refer
back to Chapter 3.)
31. Parallax measurements of the variable star RR Lyrae
indicate that it is located 750 light-years from the Sun. 38. A star in a circular orbit about the black hole at the cen-
A similar star observed in a globular cluster located far ter of our galaxy (MBH = 7.36 × 1036 kg) has an orbital
above the galactic plane appears 160,000 times fainter radius of 0.0131 light-year (1.24 × 1014 meters). What is
than RR Lyrae. the average speed of this star in its orbit?
a. How far from the Sun is this globular cluster?
39. How large is the black hole at the center of our galaxy
b. What does your answer to part (a) tell you about the
(that is, where is its event horizon)?
size of the galaxy’s halo compared to the size of its
disk? 40. A star is observed in a circular orbit about a black hole
with an orbital radius of 1.5 × 1011 kilometers (km) and
32. Although the flat rotation curve indicates that the total
an average speed of 2,000 km/s. What is the mass of this
mass of our galaxy is approximately 8 × 1011 M⊙, elec-
black hole in solar masses?
tromagnetic radiation associated with normal matter
suggests a total mass of only 3 × 1010 M⊙. Given this
information, calculate the fraction of our galaxy’s mass
that is made up of dark matter.
*33. Compare the H-R diagram for the young cluster NGC vides a Study Plan for each chapter of 21st Century
6530 (see Figure 20.11) with H-R diagrams in previous Astronomy. Study Plans include animations, read-
chapters to determine how soon the most massive stars ing outlines, vocabulary flashcards, and multiple-choice
will explode, and how massive the protostars are that quizzes, plus links to premium content in SmartWork and
are still contracting onto the main sequence. the ebook. Visit www.wwnorton.com/studyspace.
There is a theory which states that if ever anyone discovers
exactly what the universe is for and why it is here, it will
instantly disappear and be replaced by something even more
bizarre and inexplicable. There is another which states that
this has already happened.
DOUGL AS ADAMS (1952–2001)
To understand the large-scale universe, we must understand everything from the smallest
subatomic particles to the largest structures we see.
CHAP TER 21
Modern Cosmology
K E Y CONCEP T S
21.1 The Universe Has a Cosmology is the study of our large-scale universe,
Destiny and a Shape including its nature, origin, evolution, and ultimate
destiny. As we study and try to understand our universe
Cosmology is the study of the universe on the very grandest on the largest possible scales, we will learn
of scales. During our journey we have already learned many
• That gravity acts to slow the expansion of the universe.
things about the universe and its evolution. We learned that
the universe is expanding, having originated in a Big Bang • How the universe can have a shape or geometry
nearly 14 billion years ago; that it was once very hot, filled different from our standard Euclidean geometry.
with thermal radiation that has now cooled to a tempera- • How the history, shape, and fate of the universe are
ture of 2.7 kelvins (K); that light elements in the universe determined by how much mass it contains and by a
were produced within the fi rst few minutes after the Big poorly understood cosmological constant.
Bang. But there is still more to learn. In this chapter we
take a closer look at the nature of the universe and how • That the expansion of our universe appears to be
it has evolved over time, and we contemplate its ultimate accelerating.
fate. We also develop an understanding of particle physics, • What observations of the cosmic background radiation
which is necessary for describing the very smallest pieces tell us about the properties of the young universe.
of the universe.
• The universe may have undergone a period of huge
At fi rst glance, particle physics and cosmology might
expansion that we call inflation.
seem to have almost nothing in common. Whereas particle
physics is the study of the quantum mechanical world that • That matter and the fundamental forces of nature
exists on the tiniest scales imaginable, cosmology is the “froze out” of the uniformity of the expanding and
study of the changing structure of a universe that extends cooling universe moments after the Big Bang.
for billions of light-years and probably much farther. Yet the
last quarter of the 20th century saw the boundary between
these two fields fade and eventually disappear as cosmolo-
gists and particle physicists came to realize that the struc-
ture of the universe and the fundamental nature of matter
are two sides of the same scientific coin.
We know we live in an expanding universe, but will
that expansion continue forever? This is clearly one of the
great questions of modern cosmology. What is the fate of
585
586 Chapter 21 Modern Cosmology
the universe? The answer depends in part on the amount by gravity alone is greater than 1, then gravity is strong
of distributed mass the universe contains on very large enough to turn the expansion around. Like a projectile fi red
scales. The gravitational effect of this distributed matter is from the Moon at less than the escape velocity, the expan-
only one of the factors—and the fi rst one we discuss—that sion of such a universe will
determines how the universe evolves. slow and eventually stop, and Ωmass determines the
To see how gravity affects the expansion of the universe, the universe will then fall fate of a universe
it will help to recall gravity’s effects on the motion of pro- back in on itself, culminating governed exclusively
jectiles. Think back to our discussion of escape velocity in the “Big Crunch.” Con- by gravity.
in Chapter 3 and take the case of the Moon as an example. versely, if Ωmass in a possible
The fate of a projectile fi red straight up from the surface universe is less than 1, that universe will expand forever,
of the Moon depends on its speed. As long as the speed is but slowed down to some extent by gravity. This is like a
less than the Moon’s escape velocity (2.4 kilometers per projectile fi red at more than the escape velocity from the
second [km/s]), gravity will eventually stop the rise of the Moon. The dividing line, where Ωmass equals 1, corresponds
projectile and pull it back to the Moon’s surface. But if the to a universe that expands more and more slowly—continu-
speed of the projectile is greater than the Moon’s escape ing forever, but never quite stopping. Such a universe is
velocity, then gravity will lose. Although the projectile expanding at exactly the escape velocity.
will slow down, it will never stop. It will escape from the Look again at the three plots in Figure 21.1 and you will
Moon entirely. see that they are not straight lines. Curvature in these plots
Just as the gravity of the Moon pulls on a projectile, represents the effect that gravity is having on expansion. As
slowing its climb, the gravity arising from the mass con- the value of Ωmass increases (corresponding to greater density
tained in the universe slows in the universe), the amount of curvature also increases. In
its expansion. If there is Gravity slows the the case where Ωmass is greater than 1, the scale factor begins
enough mass in the universe, expansion of to decrease and gravity wins.
then gravity will be strong the universe. Until the closing years of the 20th century, most astrono-
enough to stop the expansion. mers thought that this straightforward application of grav-
And in that case, the universe will slow, stop, and eventu- ity to the universe was all there was to the question of
ally collapse in on itself in a catastrophic “Big Crunch.” But expansion and collapse. Researchers focused great efforts
if there is not enough mass, then the expansion of the union carefully measuring the mass of galaxies and assem-
verse may slow, but it will never stop. The universe will blages of galaxies in the belief that this study would reveal
expand forever. the density and therefore the fate of the universe. As we
A planet’s mass and radius determine the escape veloc- have seen time and again, measuring masses of objects in
ity from its surface. The “escape velocity” of the universe the universe is tricky. When we calculate Ωmass using just
is also determined by its mass and size—specifically, its the luminous matter that we see in galaxies and groups of
average density. If the universe is denser on average than a galaxies, we get a value of about 0.02. In Chapter 19 we saw
particular value, called the critical density, then we expect that galaxies contain about 10 times as much dark matter
gravity will be strong enough to eventually stop and reverse as normal matter, so adding in the dark matter in galaxies
the expansion. If the universe is less dense than this value, pushes the value of Ωmass up to about 0.2. Finally, when we
we expect gravity will be too weak and the universe will include the mass of dark matter between galaxies (a subject
expand forever. we will return to in the next chapter), Ωmass could increase
The faster the universe is expanding, the more mass is to 0.3 or higher. In other words, by our current accounting
needed to turn that expansion around. For that reason, the there is only about a third as much mass in the universe as
critical density depends on the value of the Hubble constant, is needed to stop the universe’s expansion.
H0. Assuming that H0 = 22 kilometers per second per mega-
light-year (km/s/Mly) and that gravity is the only thing we
have to worry about, the universe’s critical density has a
value of 8 × 10 –27 kilograms per cubic meter (kg/m3). Rather
Reviving Einstein’s “Biggest Blunder”
than trying to keep track of such awkward numbers, we As they were closing in on a good value for Ωmass, it seemed
will instead talk about the ratio of the actual density of the to astronomers that our understanding of the expansion
universe to its critical density. We call this ratio Ωmass (proof the universe was almost complete. Then the other shoe
nounced “omega sub mass”). Note that Ωmass is a dimension- dropped.
less pure number (a number that has no units). If the expansion of the universe has been slowing with
We can follow the expansion of different possible uni- time, as these simple models using gravity predict, then
verses in Figure 21.1, which shows a plot of the scale factor when the universe was young it must have been expanding
RU (see Chapter 18) versus time for different values of Ωmass. more rapidly than it is today. That is a prediction that we
(In these plots we have assumed that gravity alone controls can go to our telescopes and check. Objects that are very far
the fate of the universe—an assumption that we will revisit away (so that we see them as they were long ago) should have
in the following subsection.) If Ωmass in a universe controlled larger velocities than our local Hubble’s law would lead us
Actual density of a universe

Ωmass =
Critical density of the universe
The fate of a universe controlled

by gravity is determined by the
density of that universe divided
by the critical density.
Big
Bang
Gravity is too weak
to stop expansion: the
universe expands forever.
Borderline: expansion
slows, but never stops.
Scale factor (RU)
Big
1
<
Bang
=1
s
as
m
ss
Ω
Ω ma Gravity stops expansion

and the universe collapses.
Ωmass > 1
Big
Bang
Time
FIGURE 2 1.1 Three possible scenarios for the fate of the universe based on the critical density of
the universe, but ignoring any cosmological constant.
to expect. This is the same as saying that the plots in Figure fact that similar findings have been obtained independently
21.1 are all curving away from a straight line. (If gravity were by different groups of astronomers lends credence to the
not present, the Hubble expansion would not be changing, results. In addition, results obtained early in the 21st cen-
and the plots would instead be straight lines.) tury by the WMAP experiment (see Chapter 18) provide
During the 1990s, some groups of astronomers began independent confi rmation of this increasing rate of expan-
using tools such as the Hubble Space Telescope and the sion of the universe.
giant Keck Observatory tele- How could it be that the rate of expansion of the universe
scopes in Hawaii to test this Evidence suggests has increased over time? For this to be the case, there would
prediction. They measured that the expansion have to be some force at work to push the universe out-
the brightnesses of standard of the universe ward in opposition to normal gravity. Physicists have some
candles (in particular, Type I is accelerating. ideas about how such a hypothetical repulsive force might
supernovae) in very distant originate. These ideas are related to theories from particle
galaxies and compared those brightnesses with expected physics concerning the nature of what we call the vacuum.
brightness based on the redshifts of those galaxies. The Normally we think of a vacuum as “empty space,” but it
fi ndings of these studies sent a wave of excitement through turns out that even empty space has some very interesting
the astronomical community. Rather than showing that the physical properties. Discussion of these ideas (which are
expansion of the universe has slowed down over time, the related to Hawking radiation from black holes, discussed in
data indicated that it is speeding up! Chapter 17) will have to wait until later in the chapter, when
The observational data plotted in Figure 21.2 indicate we take up fundamental questions related to the nature of
that the expansion of the universe is speeding up. The obser- the interactions of matter. For now we turn our attention to
vations on which this conclusion is based are extremely the effects that such a force would have.
difficult to carry out, and we must take into account many The idea of a repulsive force opposing the attractive
different factors when interpreting the data. Even so, the force of gravity is not new. When Einstein used his newly
Billions of years from today

–20 –10 0 10
The cosmological constant
ΩΛ opposes gravity.
A nonzero cosmological constant

ΩΛ can cause a universe to expand
faster and faster…
0
ΩΛ > 0
Scale factor (RU)

After inflation, the
expansion either ΩΛ = 0
0.5 first decelerated and
Redshift
then accelerated…
1
1.5 Compare with the ΩΛ = 0
2 …or always curves in Figure 21.1.
decelerated. ΩΛ > 0
3
Past Today Future
0.0001 0.001 0.01 0.1 1

Ωmass < 1 …and can even prevent
Relative brightness the collapse of a universe
FIGURE 2 1. 2 Observed brightness of Type I supernovae, with Ωmass > 1.
plotted as a function of their redshift. The observations indicate
that the redshifts are too small for their distances—evidence
ΩΛ = 0
that the universe is expanding faster today than in the past. (Each ΩΛ > 0
Ωmass > 1
point represents the brightness and redshift of an individual Type
1 supernova. Overall, the observations show that the universe is Time
expanding faster today than it did in the past.) FIGURE 2 1. 3 Plots of scale factor R U versus time for
cosmologies with and without a cosmological constant, ΩΛ. If
there is enough mass in a universe, gravity could still overcome
the cosmological constant and cause that universe to collapse.
Any universe without enough mass to eventually collapse will
formulated equations of general relativity to calculate the
instead end up expanding at an ever-increasing rate.
structure of spacetime in the universe, he was greatly trou-
bled. The equations clearly indicated that any universe
containing mass could not be static, any more than a ball
can hang motionless in the air. He found the same result ΩΛ relative to the critical density.) The cosmological constant
that Figure 21.1 shows—namely, that gravity always makes acts as a repulsive force in the equations, opposing gravity
the universe move toward slower expansion or even col- and allowing galaxies to remain stationary despite their
lapse. However, Einstein’s formulation of spacetime came mutual gravitational attraction. The cosmological constant
more than a decade before Hubble did his epic work on the was Einstein’s version of the magician’s trick that allows a
expansion of the universe, and the conventional wisdom subject to hang apparently unsupported in midair.
at the time was that the universe was indeed static—that it When Hubble announced his discovery that the universe
neither expands nor collapses. is expanding, Einstein realized his mistake. The unmodi-
In order to force his new fied equations of general relativity demand that the struc-
general theory of relativity Einstein invented a ture of the universe be dynamic. Instead of doctoring them
to allow for a static universe, cosmological constant with the inclusion of ΩΛ, Einstein realized he should have
Einstein inserted a “fudge to oppose gravity in a predicted that the universe must either be expanding or
factor” into his equations. He static universe. contracting with time. What a coup it would have been for
called this fudge factor the his new theory to successfully predict such an amazing and
cosmological constant, which we write as ΩΛ(pronounced previously unsuspected result. He called the introduction
“omega sub lambda”).1 (Just as we did with Ωmass, we defi ne of his fudge factor, the cosmological constant, the “biggest
blunder” of his career as a scientist. It is ironic that with
1
Einstein probably never wrote ΩΛ, referring instead to the related the new results on the brightness of Type I supernovae,
quantity as Λ. Einstein’s biggest blunder has returned to center stage. The
3 FIGURE 2 1. 4 Current
observations from different
No Big Bang
sources—Type I supernovae,
measurements of mass in
galaxies and clusters, and detailed
2 Supernovae observations of the structure of
the cosmic background radiation—
suggest that the best current
estimate for ΩΛ is about 0.7 and
about 0.3 for Ωmass, which means
1 that the expansion of the universe
is accelerating.
ΩΛ
CBR
er
Expands forev
0
Recollapses eventually
C
Galaxy lo
se
clusters Fl d
at
–1 O
pe
n
0 1 2 3
Ωmass
repulsive force represented by the infamous ΩΛ in Einstein’s

equations is just what is needed to describe a universe that
is expanding at an ever-accelerating rate.
How does the possibility of a nonzero value for ΩΛ affect
the possible fates that await our universe? If ΩΛ is not zero,
the fate of the universe is no longer controlled exclusively
by Ωmass. If something is effectively pushing outward from
within the universe, adding to its expansion, then grav- FIGURE 2 1. 5 Cartoon from
ity will have a harder time turning the expansion around. physicist Willem de Sitter showing
In that case, the mass needed to halt the expansion of the that the cosmological constant is the
universe will be greater than the critical mass we already force blowing up the balloon that is the
discussed. universe. Note that the cartoon figure
Figure 21.3 shows plots of the scale factor RU versus time looks like lambda.
that are similar to those shown in Figure 21.1; but now we
have included the effects of a nonzero cosmological con-
stant. The evolution of a universe that collapses back on on itself, a large enough cosmological constant could make
itself (below the horizontal line in Figure 21.4) is similar, that same universe expand forever. One of the fi rst scientists
regardless of whether ΩΛ is zero. In contrast, if a universe to work with the cosmological constant was Willem de Sit-
with a nonzero cosmological constant expands forever, its ter (1872–1934). He provided an early cartoon (Figure 21.5)
evolution will look drastically different from that of any showing that it is the cosmological constant that drives the
universe in which ΩΛ is zero. As a universe expands, grav- expansion of the universe.
ity gets weaker and weaker because the mass is more and When Einstein added the cosmological constant to his
more spread out. Unlike gravity, however, the effect of the equations of general relativity, he introduced what he con-
cosmological constant becomes increasingly greater. While sidered a new fundamental constant, in many ways similar
a universe is young and compact, gravity is strong enough to Newton’s universal gravitational constant G. Today we
to dominate the effect of the cosmological constant. Unless realize that the vacuum can have distinct physical proper-
gravity is able to turn the expansion around, however, the ties of its own. For example, the vacuum can have a non-
cosmological constant wins out in the end, causing the zero energy even in the total absence of matter. We call this
expansion to continue accelerating forever. Indeed, even energy dark energy. It turns out that dark energy produces
for a universe that otherwise would have collapsed back exactly the same kind of repulsive force that Einstein’s
cosmological constant does. To astronomers, therefore, the see would change to a deceleration as the mass once again
terms dark energy and cosmological constant mean the same dominated over dark energy. In fact, a universe much denser
thing. But knowing that the vacuum can have dark energy than ours could collapse to the “Big Crunch.” On the other
associated with it is not the same as understanding why it hand, if dark energy’s effect were to increase with time, the
must have exactly the amount of dark energy we observe in universe would accelerate its expansion at an ever-increas-
the universe. That is still an unanswered question. ing rate. Ultimately, expansion could be so rapid that the
What fate, then, actually awaits our universe? Figure scale factor would become infi nite within a fi nite period of
21.4 shows the range of values for Ωmass and ΩΛ that are time—a phenomenon we call the Big Rip. Here the repul-
allowed by current observations. Assuming these observa- sive force of dark energy becomes so dominating that the
tions are correct, the values for Ωmass and ΩΛ are about 0.3 entire universe comes apart. First, clusters of galaxies rip
and 0.7, respectively. Indeed, apart; then, gravity can no longer hold individual galaxies
the data from Type I superno- The universe appears together; and so on. Just before the end, the Solar System
vae, from WMAP, and from to be accelerating and comes apart, and even atoms are ripped into their constitu-
clusters of galaxies (which will expand forever. ent components. But don’t worry too much. As this book
will be discussed in Chapter goes to press, the best observational data seem consistent
22) are all consistent with these values. Thus, it appears with constant dark energy.
that the expansion of our universe is already accelerating
under the dominant effect of the cosmological constant.
Before leaving this subject, there is one more speculative
but fascinating idea we should consider: what if the cosmo-
The Age of the Universe
logical constant is not constant with time? At first, this might Our values for Ωmass and ΩΛ not only affect our predictions
seem a strange supposition, because Einstein introduced for the future of the universe; they also influence our inter-
the cosmological constant as a true constant. But suppose pretation of the past. Figure 21.7 shows plots of the scale fac-
dark energy was not really a constant of nature, but instead tor of the universe versus time. Measurement of the Hubble
was an effect of the vacuum that arises from fundamental constant (H0) tells us how fast the universe is expanding
physics. If this were the case, dark energy’s effect could be today. That is, it tells us the slope of the curves in Figure
either increasing or decreasing with time. In truth, we do 21.7 at the current time. As we saw earlier, if the expan-
not yet understand the origin of dark energy, so it is entirely sion of the universe has not changed in time, then the plot
possible that its effects could vary with time. of RU versus time is the straight red line in Figure 21.7. The
Such a possibility would significantly change the future age of the universe in this case is equal to the Hubble time:
of the universe, as illustrated in Figure 21.6. For example, 1/H0. If the expansion of the universe has been slowing
if dark energy were to decrease rapidly enough with time, down with time (the green line in Figure 21.7), then the
the accelerating expansion of the universe that we now universe is actually younger than the Hubble time. (The
If the effects of dark energy increase FIGURE 2 1.6 How the scale
quickly enough, the scale of the universe factor R U of the universe varies
becomes infinite at the time of the Big depending on how dark energy
Rip. Just before that happens, galaxies, changes with time. If dark energy
solar systems, and even atoms are torn
is constant, then the universe first
apart by the dark energy.
decelerates and then accelerates
its expansion as dark energy
If dark energy (or the Big Rip Constant dominates gravity. If dark energy
cosmological constant) dark energy decreases with time, then dark
Scale factor (RU)
is constant, the universe energy can cause acceleration for

first decelerates and then
Deceleration some period of time, but mass
accelerates forever. If dark energy decreases with time, the
density later dominates gravity
universe first decelerates, and then
accelerates as dark energy dominates and the universe decelerates
gravity. It then decelerates again as again, and could even collapse for
dark energy subsides. If dense enough, high enough density. When the
tion
Accelera the universe could collapse. effect of dark energy increases
n
tio with time, the acceleration of the
ra
universe gets faster and faster,

ele
Dec
and the scale factor can become

infinite at a time called the Big Rip.
Big Bang Present
Time
clearly cannot be older than the universe that contains them!)

H0 gives the rate of On the other hand, if the expansion of the universe has sped
expansion today.
up with time, as suggested by the observations of Type I
1 supernovae and by WMAP, then the universe is about 13.7 bil-
Scale factor (RU)
lion years old—comfortably older than globular clusters.
An accelerating universe
is older than 1/H0. Decelerating
expansion The Universe Has a Shape
Accelerating
A decelerating universe We have already discussed such properties of the universe
expansion
is younger than 1/H0.
as density, dark energy, and age. But the universe also has
The Big Bang another key property: its shape in spacetime. By this time
occurs at RU = 0. the concept of spacetime as described by general relativity
0 is getting to be a familiar friend. Space is a “rubber sheet”
Age of accelerating universe Time
that has stretched outward from the Big Bang. In Chapter 17
Hubble time (1/H0)
we saw that the rubber sheet of space is also curved by the
Age of
decelerating
presence of mass. We saw how the shape of space around a
universe massive object can be detected through changes in simple
Today geometric relationships, such as the ratio of the circumfer-
FIGURE 2 1.7 A plot of the scale factor RU versus time for ence of a circle to its radius, or the sum of the angles in a
three possible universes. If the universe has expanded at a triangle. If the mass of a star, planet, or black hole causes a
constant rate, then its age is equal to the Hubble time, 1/H0. If distortion in the shape of space, then should not the mass of
the expansion of the universe has slowed with time, it is younger everything in the universe—including galaxies, dark mat-
than the Hubble time. If the expansion has sped up with time, ter, and dark energy—also distort the shape of the universe
then the universe is older than 1/H0. as a whole? The answer is yes.
Three basic shapes are possible for our universe. Which
shape actually describes the universe is determined by the
curve crosses RU = 0 at a point more recent than 1/H0.) If, total amount of mass and energy—in other words, the sum
on the other hand, the expansion of the universe has been of Ωmass and ΩΛ. Continuing with the rubber sheet analogy,
speeding up with time (blue line), then the true age of the the fi rst possibility, corresponding to Ωmass + ΩΛ = 1, is that
universe is greater than the Hubble time. we live in a flat universe. A flat universe is described over-
The current measured value for H0 (22 km/s/Mly) cor- all by the rules of Euclidean geometry. As shown in Figure
responds to a Hubble time (1/H0) of about 13.6 billion years. 21.8a, circles in a flat universe have a circumference of 2π
If expansion of the universe has slowed over time, the uni- times their radius, and triangles contain angles whose sum
verse is actually younger than 13.6 billion years. Having a is 180°. A flat universe stretches on forever.
younger universe is a problem if the measured ages of globu- The second possibility is that the universe is shaped
lar clusters—13 billion years—is correct. (Globular clusters something like the surface of a saddle (Figure 21.8b). This
FIGURE 2 1.8 Two-dimensional representations of the possible geometries that space can have in
a universe. In a flat universe (a), Euclidean geometry holds: triangles have angles that sum to 180°,
and the circumference of a circle equals 2π times the radius. In an open universe (b) or a closed
universe (c), these relationships are no longer correct over very large distances.
(a) Flat geometry (b) Open (saddle) geometry (c) Closed (spherical) geometry
r
r
Sum of angles > 180°
Circumference
Sum of angles Circumference > 2π r r
Sum of angles
= 180° = 2π r
< 180°
Circumference < 2π r
A universe with Ωmass + ΩΛ = 1 A universe with Ωmass + ΩΛ < 1 A universe with Ωmass + ΩΛ > 1
has a flat geometry. has an open geometry. has a closed geometry.
type of universe, in which Ωmass + ΩΛ < 1, is also infinite and the universe has grown and our observations of the cosmic
is referred to as an open universe. In this type of universe background radiation have improved, a number of puzzles
the circumference of a circle is greater than 2π times its have arisen. The solution to these puzzles has forced us to
radius, and triangles contain less than 180°. consider some remarkable ideas about how our universe
The final possibility, in which Ωmass + ΩΛ > 1, is a uni- expanded when it was very young.
verse shaped like the surface of a sphere (Figure 21.8c). The
geometric relationships on a sphere are similar to those in
the vicinity of a massive object, as discussed in Chapter 17.
The circumference of a circle on a sphere is less than 2π
The Universe Is Much Too Flat
times its radius, and triangles contain more than 180°. This The fi rst puzzle that we run into when observing our uni-
possibility is called a closed universe because space is finite verse is that the universe is too flat. In fact, the universe is
and closes back on itself. much too close to being exactly flat for this to have happened
Again we face the ques- Recent evidence by chance. To see why this is a problem, imagine a model
tion “Which of these shapes suggests that universe in which ΩΛ = 0. (This is a reasonable approxima-
describes the universe in our universe is tion for the very early days of any possible universe. When
which we live?” The measure- remarkably flat. a universe is very young, it is also very dense, and Ωmass is
ments are difficult. Even so, as all that matters.) As our model universe expands out of its
seen in Figure 21.4, Ωmass + ΩΛ is close to 1 (0.3 + 0.7), mean- own version of the Big Bang, its density falls. At the same
ing that our universe is very nearly flat. time, because the universe is expanding more slowly, the
critical density needed to eventually stop the expansion
falls as well. If Ωmass is exactly 1, the decline in the actual
density and the decline in the critical density go hand in
hand: the ratio between the two, Ωmass, remains 1 for all time,
21.2 Inflation as shown by the middle curve in Figure 21.9. A universe
that starts out perfectly flat stays perfectly flat.
Even with our understanding of the expansion, accelera- On the other hand, a universe that does not start out
tion, and shape of the universe, it is remarkable that Big perfectly flat has a very different fate. If a universe started
Bang cosmology makes so many correct predictions about out with Ωmass even slightly greater than 1, its expansion
the properties of the universe in which we live. A century would slow more rapidly than that of the flat universe,
ago, astronomers were struggling just to get a handle on meaning that less and less density would be required to
the size of the universe. Today we have a comprehensive stop the expansion. At the same time, the actual density
theory that ties together many diverse facts about nature: would be falling less rapidly than in the flat universe.
the constancy of the speed of light, the properties of grav- This disparity between the actual density of the universe
ity, the motions of galaxies, and even the origins of the very and the critical density would increase, causing the ratio
atoms of which we are made. The case for the Big Bang is between the two, Ωmass, to skyrocket. This condition is
compelling. Even so, as our knowledge of the expansion of shown by the blue curves that climb toward the top of
1,000 A universe starting with Ωmass

FIGURE 2 1.9 Universes in
slightly greater than 1 quickly
Universe closed
Ωmass(1 s) = 1.01 which Ωmass has slightly different

becomes obviously closed.
100 1.001 values at an age of 1 second.
Notice that even tiny differences
from Ωmass = 1 are rapidly
10
1.0001 amplified as the model universe
1.0000… expands. To have a value of Ωmass
Ωmass 1.0
0.9999 so close to 1 today, our universe
must have started out with a
Universe open
0.1 0.999 value of Ωmass exquisitely close

Ωmass(1 s) = 0.99 to 1. (The value of ΩΛ does not
A universe starting with Ωmass
0.01 affect the calculations shown.)
slightly less than 1 quickly
becomes obviously open.
0.001
100 102 104 106
Time (seconds)
1 second 1 minute 1 hour 1 day 1 week

21.2 Inflation 593
Figure 21.9. A universe that starts out even slightly closed discovered the bizarre world of quantum mechanics that
rapidly becomes obviously closed, and would collapse long shapes the world of atoms, light, and elementary particles.
before stars could form. When the universe was extremely young, it was so small
Conversely, if a universe started with Ωmass even a tiny bit that quantum mechanical effects played a role in shaping
less than 1, the expansion would slow less rapidly than in the structure of the universe
a flat universe. As time passed, more and more mass would as a whole. In particular, the We expect
be required for gravity to stop the too-rapidly-expanding early universe was subject to nonuniformities in the
universe. At the same time, the actual density of the uni- the quantum mechanical early universe.
verse would be dropping faster than in a flat universe. In uncertainty principle. The
this case, Ωmass (the ratio between the actual density and the uncertainty principle says that as we look at a system at
critical density) would plummet, leading to the red curves smaller and smaller scales, the properties of that system
that dive toward the bottom of Figure 21.9. become less and less well determined. This principle applies
Adding ΩΛ to the picture makes the math a bit more com- whether we are talking about the properties of an electron
plex, but it does not change the basic results. Try balancing in an atomic orbital, or about our entire universe at the time
a razor blade on its edge. If the blade is tipped just a tiny when it would have fit within the size of an atom.
bit in one direction, it quickly falls that way. If the blade Let’s look at a simple analogy of how the uncertainty
is tipped just a tiny bit in the other direction, it quickly principle applies to the universe. Imagine sitting on the
falls in the other direction instead. By all rights, we would beach looking out across the ocean. Off in the distance,
expect our universe to be either obviously open or obviously the surface of the ocean appears smooth and fl at. The
closed—analogous to the tipped razor blade. We fi nd our- horizon looks almost like a geometric straight line. Yet
selves instead in a universe in which Ωmass + ΩΛ is so close to the apparent smoothness of the ocean as a whole hides
1 that we have difficulty telling which way the razor blade the tumultuous structure present at smaller scales, where
is tipped at all! Discovering that Ωmass + ΩΛ is extremely waves and ripples upon waves fluctuate dramatically from
close to 1 after more than 13 billion years is like balancing place to place. In similar fashion, quantum mechanics says
a razor blade on its edge and coming back 10 years later to that as we look at smaller and smaller scales in the uni-
fi nd that it still has not tipped over! verse, conditions must fluctuate in unpredictable ways. In
For the present-day value of Ωmass + ΩΛ to be as close particular, quantum mechanics says that the smaller the
to 1 as it is, Ωmass + ΩΛ could not have differed from 1 by universe we consider (that is, the earlier in the history of
more than one part in 100,000 when the universe was 2,000 our universe we go), the more dramatic those fluctuations
years old. When the universe was 1 second old, it had to be become. When the universe was young, it could not have
flat by at least one part in 10 billion. At even earlier times, been smooth. There must have been dramatic variations
it had to be much flatter still. This is simply too special a (“ripples”) in the density and temperature of the universe
situation to be the result of chance—a fact referred to in from place to place.
cosmology as the flatness problem. Something about the If the universe had expanded slowly, those ripples would
early universe must have forced Ωmass + ΩΛ to have a value have smoothed themselves out. But the universe expanded
incredibly close to 1. much too rapidly for this. Dif-
ferent parts of the universe The CBR is smoother
could not have “communi- than the early
The Cosmic Background cated” with each other (telling universe should
each other to smooth out the have been.
Radiation Is Much Too Smooth ripples) rapidly enough to
The second problem faced by our cosmological models smooth these ripples out. There just wasn’t enough time
is that the cosmic background radiation is surprisingly after the Big Bang for a signal to travel from one region to
smooth. Following the discovery of the CBR in the 1960s, the other. So when we look at the universe today, we should
many observers turned their attention to mapping this back- see the fingerprint of those early ripples imprinted on the
ground glow. At fi rst they were reassured as result after cosmic background radiation—but we do not! The fact that
result showed that the temperature of the CBR is remark- the CBR is so smooth is referred to as the horizon problem
ably constant, regardless of where one looks in the sky. Yet in cosmology: different parts of the universe are too much
over time this strong confi rmation of Big Bang cosmology like other parts of the universe that should have been “over
turned instead into a puzzle that challenged our view of their horizon” and beyond the reach of any signals that
the early universe. Once we remove our motion relative to might have smoothed out the early quantum fluctuations.
the CBR from the picture, the CBR is not just smooth—it Basically, the horizon problem is this: how can different
is too smooth. parts of the universe that underwent different fluctuations
Why should we expect the CBR to be less uniform than and were never able to communicate with one another still
it is? To understand the answer, we need to shift our atten- show the very same temperature in the cosmic background
tion from the very large to the very small. In Chapter 4 we radiation?
traveling faster than the speed of light?” Inflation does not

Inflation Solves the Problems violate the rule that no signal can travel through space at
In the early 1980s, American physicist Alan Guth (1947–) greater than the speed of light. During inflation, space itself
offered a solution to the flatness and horizon problems of expanded so rapidly that the distances between points in
cosmology. Guth suggested space increased faster than the speed of light.
that for a brief time the young Inflation was a period To understand how infl ation solves the fl atness and
universe had undergone a of intense expansion horizon problems of cosmology, imagine that you are
period of inf lation during of the universe. an ant living in the two-dimensional universe defi ned
which the universe itself had by the surface of a golf ball, as shown in Figure 21.10.
expanded at a rate far in excess of the speed of light. Like Your universe would have two very apparent character-
so many things we have seen on our journey, the numbers istics: First, it would be obviously curved. If you were to
used to describe inflation are far beyond our ability to grasp walk around the circumference of a circle in your two-
intuitively. Between the ages of about 10 –35 and 10 –33 second, dimensional universe and then measure the radius of
the scale factor RU of the universe increased by a factor of the circle, you would fi nd the circumference to be less
at least 1030 and perhaps very much more. In that incompre- than 2π times the radius. If you were to draw a triangle
hensibly brief instant, the size of the observable universe in your universe, the sum of its angles would be greater
grew from 10-trillionths the size of the nucleus of an atom than 180°. The second obvious characteristic would be
to a region about 3 meters across. That is like a grain of very the dimples, approximately a half millimeter deep, on
fi ne sand growing to the size of today’s entire universe—all the surface of the golf ball.
in a billionth the time it takes light to cross the nucleus of Now imagine how this situation would change if your
an atom! golf ball universe suddenly grew to the size of Earth.
At this point you may well be saying to yourself, “Wait (Granted, this change is nowhere near comparable to the
a minute! What happened to all that business about nothing infl ation experienced by the real universe, but you get the
1 It is obvious to a cosmologically
minded ant that a golf ball is
curved and bumpy.
Circumference
of circle < 2πr
r 3 …and as far as the ant

can tell its new “universe”
is perfectly flat…
Circumference = 2πr
4 cm 2 But inflate that
golf ball to the
size of Earth…
VISUAL ANALOGY FIGURE A “dimple”

2 1.10 If a round, lumpy golf
ball were suddenly inflated to 1,300 km wide
the size of Earth, it would seem
Less than an 4 …and perfectly smooth.
extraordinarily flat and smooth atom deep
to an ant on its surface.
Similarly, following inflation,
5 Regardless of how curved and
any universe would seem both bumpy it starts out, any universe
extremely flat and extremely that undergoes inflation will become
smooth, regardless of the exact “too flat” and “too smooth” to have
geometry and irregularities it just been born that way.
started with.
13,000 km
Size of Earth
idea.) First, the curvature of your universe would no lon- verse. The slight irregularities that we do see in the CBR
ger be apparent. An ant walking along the surface of Earth are the faint ghosts of quantum fluctuations that occurred
would be hard-pressed to tell that Earth is not flat. The as the universe inflated.
circumference of a circle would be 2π times its radius, An early era of inflation in the history of the universe
and there would be 180° in a offers a handy way of solving the horizon and flatness prob-
triangle. In fact, it took us Inflating the size of a lems, but why would the real universe have the audacity
most of our history as a spe- dimpled sphere makes to do such a thing? It seems quite remarkable that the uni-
cies to realize that Earth it seem flatter. verse should undergo a period during which it expanded
really is a sphere. In the case at such an “astronomical” rate. But there are many remark-
of infl ationary cosmology, the universe after infl ation able things about the universe. It is remarkable that space-
would be extraordinarily fl at (that is, with Ω mass + Ω Λ time has a shape. It is remarkable that light is both a wave
extraordinarily close to 1) regardless of what the geometry and a particle. It is remarkable that galaxies and stars and
of the universe was before infl ation. Because the universe planets exist at all. Still, it is fair to ask what caused the
was infl ated by a factor of at least 1030, Ω mass + Ω Λ imme- universe to undergo inflation in the fi rst place, and the
diately after infl ation must have been one to within one answer surely lies in the fundamental physics that governs
part in 1060, which is fl at enough for Ω mass + Ω Λ to remain the behavior of matter and energy at the earliest moments
close to 1 today. Today’s universe is not fl at by chance. It of the universe.
is fl at because any universe that underwent infl ation
would emerge with a value for Ω mass + Ω Λ that was within
a gnat’s eyelash of 1.
So much for the flatness problem. What about the hori-
zon problem? When our golf ball universe inflated to the 21.3 The Earliest Moments
size of Earth, the dimples that
covered the surface of the golf Huge expansion also To understand the universe requires that we understand
ball were stretched out as smooths out the forces that govern the behavior of all matter and energy
well. Instead of being a half inhomogeneities. in the universe. There are four
millimeter or so deep and a fundamental forces in nature, There are four
few millimeters across, these dimples now are only an atom and everything in the uni- fundamental forces
deep but are hundreds of kilometers across. Again, our ant verse is a result of their action in nature.
would be hard-pressed to detect any dimples at all. In the (Table 21.1). Chemistry and
case of the real universe, inflation took the large fluctua- light are products of the electromagnetic force acting
tions in conditions caused by quantum uncertainty in the between protons and electrons in atoms and molecules. The
preinflationary universe and stretched them out so much energy produced in fusion reactions in the heart of the Sun
that they are unmeasurable in today’s postinflationary uni- comes from the strong nuclear force that binds together the
TA B L E 2 1 .1
The Four Fundamental Forces of Nature
Particles That
Relative Range of Can Carry the Example of
Force Strength Force Force What the Force Does
Strong nuclear 1 10 –15 m Gluons Holds protons and neutrons

together in atomic nuclei.
Electromagnetic 10 –2 Infi nite Photons Binds the electrons in an atom

to the nucleus.
Weak nuclear 10 –4 10 –16 m W +, W –, and Z 0 Responsible for beta decay.
Gravitational 10 –38 Infi nite Gravitons Holds you to Earth; binds

both the Solar System and the
Milky Way together.
(a) (b)
protons and neutrons in the nuclei of atoms. Beta decay of
nuclei, in which a neutron decays into a proton, an electron,
and an antineutrino, is governed by the weak nuclear force.
Finally, there is our old friend gravity, which has played
such a major role at every point along our journey. We sum-
marize these four forces in Table 21.1. How these forces— e
these physical laws—came into being is part of the history
of the universe as well.
e
May the Forces Be with You
In Chapter 4 we spoke of electromagnetism using electric
and magnetic fields, but we also spoke of the quantum Proton (+) Proton (+)
mechanical description of light as a stream of particles
called photons. Because there is only one reality, both of Virtual
these descriptions of electromagnetism have to coexist. photons
The branch of physics that deals with this reconciliation
is called quantum electrodynamics, or QED.
QED treats charged particles almost as if they were base-
ball players engaged in an endless game of catch. As base-
ball players throw and catch baseballs, they experience
forces. Similarly, in QED, charged particles “throw” and Electric field
“catch” an endless stream of “virtual photons,” as illustrated
in Figure 21.11. Earlier we
grappled with the idea that In QED, photons carry
quantum mechanics is a sci- the electromagnetic In QED, exchange of virtual
ence of probabilities rather force. Classically, the force photons between a proton
than certainties. The QED of a proton’s electric field and an electron alters the
alters an electron’s path. electron’s path.
description of the electromagnetic interaction between two
charged particles is an average of all the possible ways that FIGURE 2 1.11 (a) The classical view of an electron being
the particles could throw photons back and forth. The deflected from its course by the electric field from a proton.
resulting force acts, over large scales, like the classical elec- (b) According to quantum electrodynamics, the interaction
tric and magnetic fields described by Maxwell’s equations. is properly viewed as an ongoing exchange of virtual photons
Physicists speak of the electromagnetic force being “medi- between the two particles.
ated by the exchange of photons.” As is always the case with
quantum mechanics, the world described by QED is hard
to picture. Even so, QED is one of the most accurate, well
tested, and precise branches of physics. As of this writing, The strong nuclear force is described by a third theory,
not even the tiniest measurable difference between the pre- called quantum chromodynamics, or QCD. In this theory,
dictions of the theory and the outcome of an actual experi- particles such as protons and neutrons are composed of
ment has been found. more fundamental building blocks, called quarks, that are
The central idea of QED—forces mediated by the bound toget her by t he
exchange of carrier particles—provides a template for under- exchange of another type of Electroweak theory +
standing two of the other carrier particle, dubbed glu- QCD = The standard
three fundamental forces in The weak nuclear and ons. Together, electroweak model of particle
nature. The electromagnetic electromagnetic theory and QCD are referred physics.
and weak nuclear forces have forces combine in to as the standard model of
been combined into a single electroweak theory. particle physics. A deeper investigation of the standard
theory called electroweak model must await another journey. Here we leave the dis-
theory. This theory predicts the existence of three parti- cussion by pointing out that, excluding gravity, the stan-
cles—labeled W +, W –, and Z 0 —that mediate the weak dard model is able to explain all the currently observed
nuclear force. In the 1980s, physicists identified these par- interactions of matter and has made many predictions that
ticles in laboratory experiments, thus confirming the essen- were subsequently confi rmed by laboratory experiments.
tial predictions of electroweak theory. However, the standard model leaves many questions unan-
(a) Particle-antiparticle pairs

(b) …or gamma rays can (E = 2 × mec2) collide, then the two photons may disappear
can annihilate, producing interact, producing and leave an electron-positron pair behind in their place. If
gamma rays… particle-antiparticle pairs. the photons have more than the necessary energy, then the
Electron () Electron () extra energy goes into the kinetic energy of the two newly
formed particles.
Now we apply this idea to a hot universe awash in a
-ray -ray
photons photons bath of Planck radiation. Using Wien’s law from Chapter
4, which is stated as λpeak = (2,900
m K)/T, and the expres-
sion for photon energy (E = hc/λ), we can show that when
Positron (+) Positron (+) the universe was less than about 100 seconds old and had
a temperature greater than a billion kelvins, it was fi lled
FIGURE 2 1.12 (a) An electron and a positron annihilate,
with photons that had enough energy to create electron-
creating two gamma-ray photons that carry away the energy of
positron pairs. Under these conditions, photons were con-
the particles. (b) In the reverse process, pair production, two
gamma-ray photons collide to create an electron-positron pair.
stantly colliding, creating electron-positron pairs; and
electron-positron pairs were constantly annihilating each
other, creating pairs of gamma-ray photons. The whole
process reached an equilibrium, determined strictly by
swered, such as whether neutrinos have mass, or why temperature, in which pair
strong interactions are so much stronger than weak creation and pair annihila- The early universe was
interactions. tion exactly balanced each filled with photons,
other. Rather than being filled electrons, and
only with a swarm of pho- positrons.
A Universe of Particles tons, at this time the universe
was fi lled with a swarm of photons, electrons, and posi-
and Antiparticles trons, as illustrated in Figure 21.13a. Similarly, when the
For modern theories of particle physics to make any sense, universe was even hotter, photons would have produced
every type of particle in nature must also have an alter a swarm of protons and antiprotons.
ego—an antiparticle—that is the opposite of that particle in
every way described by quantum mechanics. For the elec-
tron there is the antielectron, otherwise known as the posi-
tron. For the proton there is the antiproton, for the neutron
The Frontiers of Physics
the antineutron, and so on down the list. One fascinating Physics has given us the tools to understand all the struc-
property of these particle-antiparticle pairs is that if you tures we have seen so far on our journey. There are still
bring such a pair together, the two particles will annihi- many gaps in what we know, but we are confident that
late each other. everything from planets to stars to galaxies can be under-
When a particle-antiparticle pair annihilates, the mass stood if we apply our current knowledge of the four funda-
of the two particles is converted into energy in accord with mental forces. In the fi rst decade of the 21st century, most
Einstein’s special theory of relativity (E = mc2). For example, of astronomy has been engaged in a struggle with the com-
in Figure 21.12a an electron and a positron annihilate each plexity of the universe, rather than a shortcoming of our
other, and the energy is carried away by a pair of photons. understanding of the fundamental laws governing matter,
(This is the idea behind Star Trek’s “antimatter” engines.) energy, and spacetime. As we push further back toward the
When we run time backward (as we are allowed to do in Big Bang itself, however, the nature of the game changes.
particle physics), we see two high-energy photons collid- We need new physical theories.
ing with each other, as in Figure 21.12b, creating in their In Foundations 10.1 we explored the power of symme-
place an electron-positron pair. This is an example of how try—the idea that we can learn a lot about nature just by
an energetic event can create a particle and its correspond- thinking about the way one part of something matches up
ing antiparticle—a process called pair production. with another. There we were talking about the gravitational
In principle, any type of particle and its antiparticle forces that hold planets together, but other kinds of sym-
can be created in this way. The only limitation comes when metry exist as well. In the process of pair creation there is
there is not enough energy available to supply the rest mass a symmetry between matter and antimatter: for every par-
of the particles being created. A specific example helps to ticle created, its antiparticle is created as well.
show how this works. An electron (or positron) has a mass As the universe cooled, there was no longer enough
of me = 9.11 × 10 –31 kilograms (kg), which corresponds to an energy to support the creation of particle pairs, so the swarm
energy (E = mec2) of 8.20 × 10 –14 joules (J). If two gamma-ray of particles and antiparticles that filled the early universe
photons with a combined energy greater than 16.40 × 10 –14 J annihilated each other and were not replaced. When this
In the early universe,

(a) pair creation and annihilation (b)
formed particles and antiparticles.
-ray
photon
Electron
Positron
One lonely
proton Proton
e
For every 10 billion After annihilation of
positrons, there were 10 pairs, only the leftover
billion and one electrons. electron remained.
FIGURE 2 1.13 (a) A swarm of electrons, positrons, and photons in the very early universe.
For every 10 billion positrons, there were 10 billion and one electrons. (b) After these particles
annihilated, only the one electron was left.
happened, every electron should have been annihilated by forces into a single grand, unified force, is referred to as a
a positron; and every proton should have been annihilated grand unified theory, or GUT.
by an antiproton. This was Many possible grand unified theories exist, and they
almost the case, but not quite. For every 10 billion make many predictions about the universe. Unfortunately,
For every electron in the uni- positrons there were most of those predictions are
verse today, there were 10 bil- 10 billion and one impossible to test with even GUTs predict that
lion and one electrons in the electrons. the largest of today’s particle even the proton
early universe, but only 10 bil- colliders. The problem is that will decay.
lion positrons. This one-part-in-10-billion excess of elec- the particles that carry the
trons over positrons meant that when electron-positron force or mediate GUTs are so massive that it takes enormous
pairs finished annihilating each other, some electrons were amounts of energy to bring them into existence—roughly a
left over—enough to account for all the electrons in all the trillion times as much energy as can be achieved in today’s
atoms in the universe today (Figure 21.13b). Similarly, there particle accelerators! Even so, some predictions of GUTs are
was an excess of protons over antiprotons in the early uni- testable with current technology. For example, GUTs predict
verse, and the protons we see today are all that is left from that protons should be unstable particles that, given enough
the annihilation of proton-antiproton pairs. time, will decay into other types of elementary particles.
If the standard model of particle physics were a com- This is a very slow process. Over the course of your life,
plete description of nature, then the one-part-in-10-billion GUTs predict that there may be as much as a 1 percent
imbalance between matter and antimatter would not have chance that one of the 1028 or so protons in your body will
been there in the early universe. The symmetry between decay. As of this writing, proton decay has yet to be observed,
matter and antimatter would have been complete. No mat- and only the very simplest of GUTs have been ruled out. As
ter at all would have survived into today’s universe, and we speak, however, large arrays of detectors, such as those
we would not exist. The fact that you are reading this page at Super-Kamiokande in Japan (pictured in Figure 21.14)
demonstrates that something more needs to be added to are peering into huge tanks of water, waiting to see the sig-
the model. nature of such a proton decay event. Perhaps soon we will
According to current ideas, the symmetry between mat- see a news story heralding the confi rmation of this central
ter and antimatter may be broken in a theory that joins the prediction of GUTs.
electroweak and strong nuclear forces together in much the Just as grand unified theories predict that the proton
same way that electroweak theory unified our understand- will decay, they can also explain why the universe is com-
ing of electromagnetism and the weak nuclear force. Such posed of matter rather than antimatter. Recall that in the
a theory, which combines three of the four fundamental early universe, for every 10 billion positrons there were 10
(a) FIGURE 2 1.14 Water tank and

detectors used to search for proton
decay at the Super-Kamiokande
neutrino observatory in Japan. (a)
The empty tank. (b) The tank filled
with water.
(b)
billion plus one electrons. GUTs provide a possible explana- energy available for these particles to be freely created. Dur-
tion for this excess of particles over antiparticles. To predict ing this time, the distinction among the electromagnetic,
the exact amount of this excess, however, would require us weak nuclear, and strong nuclear forces had not yet come
to know the correct grand unified theory—which as yet we into being. There was only the one grand, unified force.
don’t. Still, this class of models can at least explain why Welcome to the era of GUTs.
there is an excess of matter over antimatter. During this era of GUTs, the apparent size of the entire
The particles that mediate GUTs may be beyond the observable universe was less than a trillionth the size of a
reach of today’s high-energy physics labs. But when the single proton. Such an infinitesimally small size may seem
universe was very young (younger than about 10 –35 second) virtually incomprehensible, yet the basic ideas needed to
and very hot (hotter than about 1027 K), there was enough understand this universe are already in place. As we move
backward in time, however, there is one threshold we have The conflict between the continuous and the discrete—
yet to cross. The story of advances in our understanding of between general relativity and quantum mechanics—brings
physical law has been a story of unification—of the elec- us to the current limits of human knowledge. The physics
tromagnetic and weak forces into the electroweak theory, that we know can take us back to a time when the universe
and then of these and the strong nuclear force into a GUT. was a millionth of a trillionth of a trillionth of a trillionth
But this program is incomplete. How does gravity fit into of a second old; but to push back any further, we need
this scheme? something new. We need a theory that combines general
General relativity provides a beautifully successful relativity and quantum mechanics into a single theoreti-
description of gravity that correctly predicts the orbits of cal framework unifying all four of the fundamental forces.
planets, describes the ulti- Here we have reached the holy grail of modern physics. To
mate collapse of stars, and Gravity does not fit understand the earliest moments of the universe, we need
even enables us to calculate into the GUT picture. a theory of everything (TOE).
the structure of the universe. A successful theory of everything would do more than
Yet general relativity’s description of gravity “looks” very unify general relativity with quantum mechanics. It
different from our theories of the other three forces. Rather would tell us which of the
than talking about the exchange of photons or gluons or possible GUTs is correct and Superstring theory is
other carrier particles, general relativity talks about the would provide an answer for a possible theory
smooth, continuous canvas of spacetime upon which events the nature of dark matter. A of everything.
are painted. successful theory of every-
We might be tempted to say, “Oh well. Gravity works thing would also necessarily answer several outstanding
one way and the other forces work another way, and that issues in cosmology, including the how, when, and why
is how the universe happens to be.” In practice, that is of inflation, and the underlying physics of the dark energy
exactly what we do when we call on quantum mechanics that is accelerating the expansion of the universe. Physi-
to tell us about the properties of atoms and then use rela- cists are currently grappling with what a TOE might look
tivity to describe the passage of time or the expansion of like. Currently, the most promising contender for the title
the universe. Even the era of GUTs is described perfectly if is superstring theory. Here, elementary particles are
we treat gravity as a separate force. As we push back even viewed not as points but as tiny loops called “strings.” A
closer to the moment of the Big Bang, however, this happy guitar string vibrates in one way to play an F, another
coexistence between relativity and quantum mechanics way to play a G, and yet another way to play an A. Accord-
turns instead into a brutal confrontation. ing to superstring theory, different types of elementary
particles are like different “notes” played by vibrating
loops of string.
In principle, superstring theory provides a way to rec-
Toward a Theory of Everything oncile general relativity and quantum mechanics, but there
When the universe was younger than about 10 –42 second old, is a price to pay for this success. To make superstring theory
its density was incomprehensibly high. The observable uni- work, we have to imagine that these tiny loops of string are
verse was so small that 1060 universes would have fit into vibrating in a universe with 10 spatial dimensions! (Adding
the volume of a single proton! Under these extreme condi- time to the list would make our universe 11-dimensional.)
tions, quantum mechanical fluctuations in the matter and How can that be, when we clearly experience only 3 spatial
radiation making up the universe involved immense amounts dimensions? Whereas the
of mass—so much mass that three spatial dimensions that Superstring theory
quantum fluctuations made In the Planck era, we know spread out across the predicts 10 spatial
mincemeat of spacetime. the whole universe vastness of our universe, the dimensions,
Rather than a smooth sheet, was a quantum other seven dimensions pre- but we experience
spacetime was a quantum mechanical froth. dicted by superstring theory only 3 of them.
mechanical froth. The failure wrap tightly around them-
of general relativity to describe this early universe is remi- selves (Figure 21.15), extending no further today than they
niscent of the failure of Newtonian mechanics to describe did a brief instant after the Big Bang.
the structure of atoms. An electron in an atom must be To better appreciate how this bizarre notion works,
thought of in terms of probabilities rather than certainties. imagine what it would be like to live in a three-dimensional
Similarly, there is no unique history for the earliest moments universe (like the one we experience) in which one of those
after the Big Bang. This era in the history of the universe is dimensions extended for only a tiny distance. Living in
referred to as the Planck era, signifying that we can under- such a universe would be like living within a thin sheet
stand the structure of the universe itself during this period of paper that extended billions of light-years in two direc-
only by using the ideas of quantum mechanics. tions but was far smaller than an atom in the third. In such
Order “Froze Out” of the

Cooling Universe
To understand the very earliest moments in the history of
the universe, we have been looking backward to higher and
higher energies and correspondingly to earlier and earlier
times. We have seen how the four forces become unified in
stages—from four separate forces to electroweak theory to
GUTs and then to a TOE. We find that our universe started out
with one theory of everything, and as the universe expanded
and cooled the various forces emerged to take on lives of their
own. Figure 21.16 illustrates how the four fundamental forces
emerged in the evolving universe. In the first 10 –43 second
FIGURE 2 1.15 It is virtually impossible to visualize what seven
after the Big Bang, as described by the TOE, the physics of
spatial dimensions wrapped up into structures far smaller than
elementary particles and the physics of spacetime were one
the nucleus of an atom would be like. Here such geometries are
projected onto the two-dimensional plane of the paper.
and the same. As the universe expanded and cooled, gravity
parted ways with the forces described by the GUT. Spacetime
took on the properties described by general relativity. Infla-
tion may also have been taking place at this time.
As the universe continued to expand and its temperature
a universe we would easily be aware of length and width— fell further, less and less energy was available for the cre-
we could move in those directions at will. In contrast, we ation of particle-antiparticle
would have no freedom to move in the third dimension at pairs. When the particles As the universe
all, and we might not even recognize that the third dimen- responsible for GUT interac- cooled, atomic nuclei
sion existed. Perhaps our only inkling of the true nature of tions could no longer form, formed first, followed
space would come from the fact that in order to explain the the strong force split off from by atoms.
results of particle physics experiments, we would have to the electroweak force. One
assume that particles extended into a third, unseen dimen- might speak of this transition as the strong force “freezing
sion. In like fashion, if superstring theory is correct, we see out” of the GUT because this and other similar transitions
three spatial dimensions extending on possibly forever, but are reminiscent of the phase change that occurs as water
we are unaware of the fact that each point in our three- changes to ice and molecules become more constrained in
dimensional space actually has a tiny but fi nite extent in their motions. Somewhere along the line, as the unity of the
seven other dimensions at the same time! original TOE was lost, the symmetry between matter and
Superstring theory is only a pale shadow of the sort antimatter was broken. As a result, the universe ended up
of well-tested theories that we have made use of through- with more matter than antimatter.
out this book. In some respects, superstring theory is no The next big change took place when the particles
more than a promising idea providing direction to theorists responsible for unifying the electromagnetic and weak
searching for a TOE. It is worth noting that we will prob- nuclear forces froze out, leaving these two forces indepen-
ably never be able to build particle accelerators that enable dent of each other. The four fundamental forces of nature
us to directly search for the most fundamental particles that govern today’s universe had now come into their own.
predicted by a TOE. The energies required are simply too The temperature of the universe had fallen to a chilly 1016
high. Fortunately, however, nature has provided us with K, and a 10-trillionth of a second had ticked off the cosmic
the ultimate particle accelerator: the Big Bang itself! The clock. It was a full minute or two before the universe cooled
structure of the universe that we see around us today is the to the billion-kelvin mark, below which not even pairs of
observable result of that grand experiment. electrons and positrons could form.
One of the most intriguing questions that a successful Although the universe was now too cool to form addi-
TOE may answer is whether the universe could have been tional particles and their antiparticles, it was still hot
different. Do we live in only enough for the thermal motions of protons to overcome the
one of many possible uni- Is ours the only electric barriers between them, allowing nuclear reactions
verses, each with different possible universe? to take place. These reactions formed the least massive ele-
physical laws? Or are the ments, including helium, lithium, beryllium, and boron,
physical laws that govern our universe the only consistent but could not create more massive elements.
set of physical laws that could exist? We do not know the Big Bang nucleosynthesis had come to an end by the
answer yet, but possibly within your lifetime we will. time the universe was 5 minutes old, and the tempera-
Today
t = 14 109 yr
T = 2.73 K
d = 10–26 kg/m3
Galaxies and stars form
t = 109 yr
T = 18 K
d = 10–24 kg/m3
Transparent
t = 105 yr
Strong Weak Electro- Fog Recombination T = 4,000 K
nuclear nuclear magnetic Gravity d = 10–17 kg/m3
force force force
Temperature of
Sun’s surface
Temperature of
Sun’s center
Big Bang nucleosynthesis t = 3 min
T = 109 K
d = 105 kg/m3
Electron-positron pair annihilation t = 15 s

T = 3 109 K
d = 107 kg/m3
Electro-
Weak magnetic Density of white dwarf
force force
Density of neutron star
Electroweak breaks t = 10–13 s
Strong Electro- T = 1016 K
force weak force q d = 1035 kg/m3
q W+
q W–
t = 10–35 s
GUT breaks T = 1028 K
d = 1083 kg/m3
Grand Gravity
unified
Inflation
theory
t = 10–43 s
TOE breaks T = 1032 K
d = 1099 kg/m3
Planck
Theory
Era
of
everything
FIGURE 2 1.16 Eras in the evolution of the universe. Big

As the universe expanded and cooled following the Big Bang
Bang, it went through a series of phases determined
by what types of particles could be created freely at
that temperature. Later the structure of the universe
was set by the gravitational collapse of material to
form galaxies and stars, and by the chemistry made
possible by elements formed in stars.
Summary 603
ture of the universe had dropped below about 800 mil- to form neutral atoms. We have encountered this event
lion K (see Section 18.5). The density of the universe at before. This was the era of recombination, which we see
this point had fallen to only about a tenth that of water. directly when we look at the cosmic background radia-
Normal matter in the universe now consisted of atomic tion. At this stage the radiation background could no
nuclei and electrons, awash in a bath of radiation. So the longer dominate over matter, and gravity began playing
universe remained for the next several hundred thou- its role in forming the vast structure of the universe that
sand years, until fi nally the temperature dropped so far we now see. Understanding this structure will be the
that electrons were able to combine with atomic nuclei subject of the next chapter.
borne out time and again, in everything from observa-

Seeing the Forest for the Trees tions of the cosmic background radiation, to the predic-
tions the theory makes about the abundances of chemical
The journey that we have taken in 21st Century Astron- elements, to the fact of the expansion of the universe
omy has involved many amazing discoveries. We have itself. Although the jury is still out, the best models of
peered into the heart of the Sun, watched as planetary the universe early in the 21st century tell of a future in
systems formed, and witnessed the violent death throes which the expansion of the universe continues forever
of massive stars. We have seen our ideas of cause and at an ever-accelerating pace.
effect shattered by the quantum mechanical world of Seldom have novelists, theologians, or philosophers
atoms, and we have come to accept that space and time dared dream of anything so extraordinary as the universe
are joined in a four-dimensional fabric of spacetime of modern cosmology. Yet we arrive at our conclusion
that is bent and even torn by the presence of mass. not by fantasy or conjecture, but by following a path laid
Yet even when witnessed within the context of such down by hard-won observation and well-tested physical
a remarkable journey of discovery, what we see when law. In this chapter we have discovered the shape of the
we look at the structure of the universe itself is truly canvas on which our existence is painted. Yet our world
mind-boggling. is not defined by spacetime alone. The trees and planets
We live in a universe in which the spacetime described and stars are still there for us to see. We are part of a uni-
by general relativity has been expanding outward for verse that is filled with structure. How did that structure
about 14 billion years—an expansion that can be traced arise? What are the acts in the grand play that continues
back to a single moment when spacetime and all that it on the cosmic stage? In the next chapter we will run the
contains came into existence. The Big Bang is a theory, clock forward, watching as the structure and complex-
but one of the lessons we have learned along the way is ity of today’s universe emerge.
that a well-tested and well-corroborated scientific the- We have described the canvas of the universe in this
ory is the closest that humans can ever come to certain chapter. It is now time to move on to the painting itself.
knowledge. The predictions of the Big Bang have been Hold on to your hats!
Summary • Understanding the very earliest moments in the universe

requires that we also understand how the four funda-
mental forces of nature were all unified into one basic
• Both gravity and the cosmological constant (or dark
phenomenon at early times.
energy) determine the fate of the universe. Observations
suggest that rather than slowing down, the expansion
of the universe is accelerating.
• The very early universe may have gone through a brief includes algorithmically generated versions of these
but dramatic period of exceptionally rapid expansion, questions, plus additional conceptual exercises. If
called inflation. If true, inflation would explain both your instructor assigns questions in SmartWork, log in at
the flatness and the homogeneity of the universe we smartwork.wwnorton.com.
see today.
Student Questions a. How would the fate of the universe be affected if

Newton’s gravitational constant changed with time?
b. What if, instead, the electric force between charged
particles changed with time?
1. As applied to the universe, what is the meaning of criti-
17. Of the four fundamental forces in nature, which one
cal density?
depends on electric charge?
2. What set of circumstances would cause an expanding
18. The standard model cannot explain why neutrinos have
universe to reverse its expansion and end up in a “Big
mass, or why electron-positron asymmetry existed in the
Crunch”?
early universe. Do these failings make it an incomplete
3. What is the principal difference between normal matter theory? Should all of its predictions be ignored until the
and dark matter? theory can solve these remaining issues?
*4. Describe the observational evidence suggesting that Ein- 19. Explain how quantum electrodynamics describes
stein’s cosmological constant (a repulsive force) may the electromagnetic interaction between two charged
be needed to explain the historical expansion of the particles.
universe.
20. What are quarks?
5. Describe what astronomers mean by dark energy.
*21. Explain how quantum chromodynamics describes the
6. If the universe is being forced apart by dark energy, structure and stability of protons and neutrons.
why isn’t our galaxy, Solar System, or planet being torn 22. What is an antiparticle, and how do its properties com-
apart? pare with a normal particle?
7. Describe the cause and consequences of a “Big Rip.” 23. Explain what happens when you bring a particle and an
8. In Chapter 18, we said we could estimate the age of the antiparticle together.
universe with Hubble time (1/H0). Why does that not give 24. Why are there so few antiparticles in the universe?
us the best answer?
25. Explain the process of pair creation.
9. If you could accurately measure a triangle and a circle
drawn on the surface of Earth, you would fi nd that the 26. Describe the Planck era.
sum of the triangle’s interior angles is more than 180° 27. Is it possible that some planets, stars, or even galaxies
and the circle’s circumference is less than 2πr. These are made entirely of antimatter? If so, how might we
results are contrary to what you probably learned about distinguish these from identical objects made of regular
triangles and circles in your introductory geometry matter?
class. Explain why we get these results and how they
relate to a spherical universe. 28. What are the basic differences between a grand unified
theory (GUT) and the theory of everything (TOE)?
10. Explain the geometry of an open universe.
29. In this chapter we have spoken about GUTs and the TOE
11. Do astronomers believe we live in a closed, flat, or open as if they exist and we have only to specify and test them.
universe? Explain your answer. Is this a safe assumption?
*12. What is the flatness problem and why has it been a prob- 30. As the sensitivity of our instrumentation increases, we
lem for cosmologists? are able to look ever farther into space and, therefore,
ever further back in time. When we reach the era of
13. During the period of inflation, the universe may have
recombination, however, we run into a wall and can see
briefly expanded at 1030 (a million trillion trillion) or
no further back in time. Explain why.
more times the speed of light. Why did this ultra-rapid
expansion not violate Einstein’s special theory of relativ-
ity, which says that neither matter nor communication
can travel faster than the speed of light?
31. Currently, the Hubble constant has an uncertainty of
14. Why is high-energy physics important to our under-
about 5 percent. What are the corresponding maximum
standing of the early universe?
and minimum ages allowed for the universe?
15. Name the four fundamental forces in nature.
32. How many hydrogen atoms need to be in 1 cubic
**16. The fundamental forces of the universe are generally meter (m3) of space to equal the critical density of the
assumed not to change. universe?
**33. Dark energy seems to have strong effects only in the a. Calculate how much energy (in joules) would be
regions of space between groups of galaxies. Estimate released as the ordinary matter and antimatter hydro-
the density of mass below which dark energy operates. gen atoms annihilated one another.
b. Compare this amount of energy with the energy
34. The universe today has an average density ρ 0 = 3 × 10 –28
released by a 1-megaton hydrogen bomb (1.6 ×
kg/m3. Assuming that the average density depends on
1014 J).
the scale factor, as ρ = ρ 0/RU3, what was the scale fac-
tor of the universe when its average density was about **38. One GUT theory predicts that a proton will decay in
the same as Earth’s atmosphere at sea level (ρ = 1.23 about 1031 years, which means if you have 1031 protons,
kg/m3)? you should see one decay per year. The Super-Kamio-
kande observatory in Japan holds about 20 million kg of
35. The proton and antiproton each have the same mass, mp
water in its main detector, and it did not see any decays
= 1.67 × 10 –27 kg. What is the energy (in joules) of each
in 5 years of continual operation. What limit does this
of the two gamma rays created in a proton-antiproton
observation place on proton decay and on the GUT the-
annihilation?
ory described here?
**36. There are about 500 million CBR photons in the uni-
verse for every hydrogen atom. What is the equivalent
mass of these photons? Is it large enough to factor into
the overall density of the universe?
37. Suppose you brought together a gram of ordinary-matter Astronomy. Study Plans include animations, read-
hydrogen atoms (each composed of a proton and an elec- ing outlines, vocabulary flashcards, and multiple-choice
tron) and a gram of antimatter hydrogen atoms (each quizzes, plus links to premium content in SmartWork and
composed of an antiproton and a positron). Keeping in the ebook. Visit www.wwnorton.com/studyspace.
mind that 2 grams is less than the mass of a dime,
listen; there’s a hell of a good universe next door: let’s go.
E. E. CUMMINGS (189 4 –1962)
An example of galactic cannibalism, the NGC 1532/1531 galaxy pair shows how a large galaxy
grows by accreting and subsuming a small galaxy that strays too close.
CHAP TER 22
The Origin of Structure
K E Y CONCEP T S
22.1 Whence Structure?
The universe that emerged from the Big Bang was
Throughout the ages there have been as many different incredibly uniform, wholly unlike today’s universe of
ideas about the origin of the universe as there have been galaxies, stars, and planets. As our journey nears its
cultural traditions. Thoughts about what the universe was end, we tackle face on the question that has been with
like once upon a time have been part of the mythologies and us all along: Where does structure in the universe
traditions of all great civilizations. We live at a remarkable come from? Here we find that complex structure is
moment in history, when the nature of the early universe a natural, unavoidable consequence of the action of
has moved from philosophical speculation to the realm of
physical law in an evolving universe.
scientific fact. All we have to do to see the early universe
is point our microwave telescopes at the sky. The glow of • Just as stars cluster together in galaxies, galaxies
the cosmic background radiation is there for anyone with gather together to form large and small clusters of
the appropriate technology to see. galaxies, which themselves cluster together into giant
The early universe was an extraordinary place—an superclusters.
expanding fi reball that was far more uniform than the blue • Galaxies formed as slight ripples in the dark matter
of the bluest sky on the clearest day. How different that uni-
emerging from the Big Bang, which then collapsed under
verse was from the universe we see about us today! Today’s
the force of gravity, pulling in normal matter as well.
universe is one of stars and galaxies, viewed from the sur-
face of a planet with oceans and mountains and uncount- • Larger galaxies form from the merging together of
able species of living things. The contrast between these small protogalaxies.
two realities cries out for explanation. How did we get from • Galaxies were drawn together by gravity to form the
there to here? Thus we come to one of the most philosophi- large groupings of galaxies we see today.
cally intriguing questions in the history of human thought:
What is the origin of structure? • There are two types of dark matter: cold and hot.
• Our own observable universe may be but one among
an infinite collection of universes.
• Our current understanding of physics suggests that
over the long course of time, our universe will grow
very cold and structure itself may slowly disappear.
607
608 Chapter 22 The Origin of Structure
FIGURE 22 .1 A graphical
Sextans B 1 million ly map of the galaxies in our
Sextans
SextansAA
own Local Group of galaxies.
Most are dwarf galaxies. Spiral
galaxies are shown in yellow.
Leo A
NGC Antlia
Antlia
3109 Dwarf
Dwarf 180°
Leo I Leo II Canes NGC NGC
Dwarf 185 147
M110
IC10
270° 90°
Milky
Andromeda
Way
Galaxy
NGC 6822 M32
Triangulum
Galaxy Andromeda
Phoenix 0° II, III, and I
Dwarf Pegasus
IC
1613 Dwarf
Cetus LGS 3
Tucana Dwarf Aquarius
Dwarf Dwarf
WLM
SagDIG
22.2 Galaxies Form galaxy clusters contain far more dwarf galaxies than giant
galaxies. Nevertheless, most of the mass in galaxy clusters
Groups, Clusters, and resides in the giant galaxies. In addition, although spiral
galaxies are common in most systems, elliptical galaxies
Larger Structures are prevalent in only about one-fourth of galaxy clusters.
The Virgo Cluster (Figure 22.3a), located 53 Mly from the
Just as stars and clouds of glowing gas show us the struc- Local Group (see Figure 22.2), is an example of a cluster
ture of our galaxy, the distribution of galaxies shows us containing mostly spiral galaxies. The more distant Coma
the structure of our universe. And just as it is gravity that Cluster (Figure 22.3b) is dominated by giant elliptical and
holds galaxies together, giving them their shape, it is grav- S0 galaxies.
ity that shapes the universe itself. No galaxy exists in utter Clusters and groups of galaxies themselves bunch
isolation. The vast majority of galaxies are parts of gravi- together to form enormous superclusters, which contain
tationally bound collections of galaxies. The smallest and tens of thousands or even hundreds of thousands of galax-
most common of these are called galaxy groups. A galaxy ies and span regions of space typically more than 100 Mly
group is a structure containing as many as several dozen in size. Our Local Group is part of the Virgo Supercluster,
galaxies, most of them dwarf galaxies. Our Milky Way is which also includes the Virgo Cluster.
a member of the Local Group, which consists of two giant Hubble’s law is a powerful tool for mapping the distri-
spirals (the Milky Way and the Andromeda Galaxy), along bution of galaxies, groups, clusters, and superclusters in
with at least 30 smaller dwarf galaxies in a volume of space space. Using Hubble’s law, all
about 4 million light-years (Mly) in diameter (Figure 22.1). we need to determine the dis- Galaxy redshift
Almost 98 percent of all the galaxy mass in the Local Group tance to a galaxy is a single surveys measure
resides in just its two giant galaxies. spectrum from which we can distances to large
Larger systems of galaxies, called galaxy clusters, may measure the galaxy’s red- numbers of galaxies.
consist of hundreds of galaxies, often with a more regular shift. Although it is easy in
structure than is found in galaxy groups. Galaxy clusters principle to measure the redshift of a galaxy, in practice
are larger than groups, typically occupying a volume of this can be a time-consuming process. Astronomers mea-
space 10–15 Mly across. In many ways, our own Local Group sured the fi rst redshifts from spectra recorded on photo-
can be considered a small cluster of galaxies, and we show graphic plates. These plates required exposures of several
its location relative to two well-known clusters (the Virgo hours to capture the feeble signal, and results rolled in at
Cluster and the Coma Cluster) in Figure 22.2. Like groups, the breakneck pace of one or two redshifts per night of
22.2 Galaxies Form Groups, Clusters, and Larger Structures 609
Coma Cluster (a)
Virgo Cluster
Local Group
M33/
Triangulum Andromeda
(b)
M32
Milky Way
SMC LMC
FIGURE 22 . 2 This figure shows a close-up view of the Local

Group (bottom) and its position relative to the Virgo Cluster,
which is the nearest cluster, and to the more distant Coma
Cluster.
FIGURE 22 . 3 (a) The Virgo Cluster contains a large fraction of
spiral galaxies. (b) The larger, richer Coma Cluster is dominated
painstaking observation. By 1975, astronomers had docu- by elliptical galaxies.
mented redshifts for only about a thousand of the hundred
billion or so galaxies we can see. Since that time, there
has been a happy marriage of larger telescopes, newer the clusters and superclusters, rather than being scattered
instruments (such as CCD detectors and spectrographs randomly through space, are linked together in an intricate
capable of observing many galaxies at once), and more network of “fi laments” and
powerful computers that process larger amounts of data “walls.” The concentrations of Large-scale structure
in rapid, automated fashion. By 1990, astronomers had galaxies, in turn, surround fills the universe.
measured the redshifts of over 10,000 galaxies, and that large voids, regions of space
number is now well over 1 million. that are largely empty of galaxies. These voids represent
The Harvard-Smithsonian Center for Astrophysics con- some of the largest “structures” seen in the universe. Though
ducted the first large redshift survey and, in 1986, presented the voids may seem like “a whole lot of nothing,” we do not
the astronomical community with a “slice of the universe,” know that they are empty of matter—only that they are
as displayed in Figure 22.4a. The observations show that largely empty of observable galaxies. Clusters and super-
Hubble’s law is used

to measure galaxy
(a) distances… (b) 0.2
0.05
0.2
0.04
Volume
Walls Redshift sampled
0.03
0.1 in (a)
Redshift 0.1
0.02
0.01
Angle Angle Angle

0
on sky on sky on sky
0.1
0.2
1 1
0.3
0.4 Voids
Billion
Billion 0.5 2
…and map large- light-years 2
light-years
0.6 scale structure
in the universe.
FIGURE 22 . 4 Redshift surveys use Hubble’s law to map the universe.

(c) (a) In 1986, the Harvard-Smithsonian Center for Astrophysics redshift
survey, called “A Slice of the Universe,” was the first to show that
clusters and superclusters of galaxies are part of even larger-scale
structures. (b) The 2dF Galaxy Redshift Survey, completed in 2003,
shows similar structures at even larger distances. (c) The 2008 Sloan
Digital Sky Survey map of the universe extends outward to a distance of
about 2 billion light-years. Shown here is a sample of 67,000 galaxies
colored according to the ages of their stars, with the redder, more
strongly clustered points showing galaxies that are made of older stars.
0.02
0.04
0.06
0.08
0.10
0.12
0.14
Redshift z
22.3 Gravity Forms Large-

Scale Structure
Large-scale structure cries out for explanation. Clearly it
is the consequence of well-defi ned physical processes, but
what processes? Cosmologists have proposed a number of
ideas. For example, early on it was suggested that voids were
the result of huge expanding blast waves from tremendous
explosions that might have occurred in the early universe.
clusters are located within the walls and filaments. This The correct answer has turned out to be less fanciful, but
structure, however, is not peculiar to the “nearby” universe. far more satisfying: large-scale structure is the fingerprint
Subsequent surveys have looked at much larger volumes of of our old friend gravity.
space. Figure 22.4b shows the results of one such survey In our discussion of star formation in Chapter 15, we
conducted with the Anglo-Australian Telescope at the Sid- learned about gravitational instabilities. As illustrated in
ing Spring Observatory in Australia. Results from a more Figure 15.14, this concept shows that star formation can
recent survey, the Sloan Digital Sky Survey (SDSS), are begin with a molecular cloud with clumps inside it. Grav-
shown in Figure 22.4c. For as far out as our observations ity then causes those clumps to collapse faster than their
can currently measure, the universe has a porous structure surroundings. If conditions are right, what began as rela-
reminiscent of a sponge. Together, galaxies and the larger tively minor variations in the density of a cloud are turned
groupings in which they are found are referred to as large- by gravity into stars. The same gravitational instability is
scale structure. responsible for galaxy formation. If you replace the words
molecular cloud with universe and star with galaxy in the ture that will emerge following this episode of rapid expan-
sentence above referring to Figure 15.14, you will have a sion. These are among the predictions of the inflation model
good starting point for understanding the way galaxies that will be tested in the years to come. These predictions
formed from slight variations in the distribution of matter are especially important to test because they tie together
following recombination. the large-scale structure of
Recall from the previous chapter that the slight ripples today’s universe with our Smaller structures
in the glow of the cosmic background radiation (CBR) prob- most basic ideas about what form first. Larger
ably result from quantum the universe was like in the structures form later.
mechanical variations that Ripples in the early briefest instant after the Big
imprinted structure on the universe were the Bang. While we await fi nal answers about the details of
early universe at the time of seeds of galaxies and structure, we know enough to say that the early universe
inflation. These variations large-scale structure. was more irregular on smaller (galaxy-sized) spatial scales
provided the seeds, or than it was on the scales of the clusters, superclusters, fila-
“clumps,” from which galaxies and collections of galaxies ments, and voids that are shown in Figure 22.4. An imme-
grew. In many ways, it is truly amazing that quantum diate consequence of this fact is that smaller structures
mechanics, the physics that governs atomic nuclei, atoms, (such as galaxies and even subgalactic clumps) formed first,
and molecules but that is almost undetectable in our own whereas larger structures took more time to form. The idea
life experience, is responsible for seeding the very largest of small structure forming fi rst and larger structure form-
structures that we can see in our universe. ing later is referred to as hierarchical clustering. Hierarchi-
It is one thing to say that galaxies and larger structures cal clustering has become one of the most important themes
formed from gravitational instabilities that began with slight in our growing understanding of how structure in the uni-
irregularities in the early universe formed.
verse. It is quite another to Models show how The third thing we need if we are to model the formation
turn this statement into a real gravity turns ripples of large-scale structure is a complete list of the types and
scientific theory with testable into large-scale amounts of ingredients that existed in the early universe.
predictions. To accomplish structure. Specifically, we need to know the balance between radia-
that, we return to the same tion, normal matter, and dark matter. We also need to make
basic technique that we have used to look into the centers some choices about the nature of the dark matter that we
of planets and stars and to answer many other questions use in our model. We’ll discuss this further in a bit.
about things we cannot observe directly. We combine the Once we have these three pieces of information—the
ideas we want to test with the laws of physics, construct a shape of the universe we are modeling, the way matter
model, and then compare the predictions of that model with inhomogeneities developed, and the nature and mixture of
observations of the universe. different forms of matter we are using—the rest is physics
To build a model of the formation of large-scale struc- and calculations; we simply need to “turn the crank,” as
ture, we have to begin with three key pieces of informa- the expression goes. As we discuss later, there are some real
tion. First we have to decide what universe we are going to difficulties in carrying out this strategy. For example, our
model. That is, what values of Ωmass and ΩA (see Chapter 21) uncertainties about how gas turns into stars make it diffi-
are we going to assume? These values are important in part cult to compare our models with the real universe. Despite
because they determine how rapidly the universe expands technical difficulties, we are aiming to answer this key
and therefore how difficult it is for gravity to overcome this question: What choices lead to a model universe that most
expansion in a particular region. The more rapidly a uni- resembles the real universe in which we live?
verse is expanding, or the less mass it contains, the more
difficult it will be for gravity to pull material together into
galaxies and larger-scale structures. Galaxies Formed Because
The second thing we need to know to construct our
model is what the early bumps in the density of the universe
of Dark Matter
looked like. How large and how concentrated were these When we actually observe the cosmic background radiation
early bumps? There are several ways to approach this ques- using such satellites as COBE and WMAP, we find variations
tion. One such method uses observations of variations in the in the background radiation of about one part in 100,000.
CBR. The data from both COBE and WMAP (see Figure 18.20 The theoretical models clearly show that such tiny varia-
and the opening image for Chapter 18) provide a good pic- tions at the time of recombination (when the universe was
ture of structure in the CBR, from which we can infer what about half a million years old) are far too slight to explain
the early inhomogeneities in the universe must have looked the structure we see in today’s universe. Gravity is simply
like. Future observations, such as those from the European not strong enough to grow galaxies and clusters of galax-
Space Agency’s Planck mission, will do even better. ies from such poor “seeds.” These models indicate that for
A different way to approach this question is to look at ripples in the density of the universe to have collapsed to
models of inflation, which make predictions about the struc- form today’s galaxies, the density of those ripples must have
been at least 0.2 percent greater than the average density of ments. Therefore, the much greater amount of dark matter
the universe at the time of recombination. cannot be made of normal
If normal luminous matter in the early universe had been matter consisting of neutrons, Dark matter provides
clumped at this level, the variations in the CBR today would protons, and electrons. If it the seeds for
be at least 30 times larger than what is observed. At fi rst were, it would have affected observed structure.
glance this discrepancy might seem to be an insurmountable the formation of chemical ele-
problem with our understanding of the origin of structure ments in the early universe. The abundances of several iso-
in the universe, but it has turned out instead to be a crucial topes of the least massive elements would be quite different
result that ties together a number of pieces of the puzzle. The from what we find in nature. Dark matter must be something
theme of this story is “dark matter,” which turns out to be else—something that has no electric charge (so it does not
an essential ingredient enabling us to understand how the interact with electromagnetic radiation) and that interacts
kind of structure we see in the universe originated. only feebly with normal matter. Clumps of such dark mat-
Dark matter fi rst appeared on our journey back in Chap- ter in the early universe would not have interacted with
ter 19 as an ad hoc construction—an annoyance, really— radiation or normal matter, so we would not see them
used by astronomers to explain the oddly flat rotation curves directly when looking at the CBR. This unseen dark matter
of spiral galaxies. When we turn our attention to clusters of solves the problems of modeling the formation of galaxies
galaxies, we fi nd that dark matter dominates normal mat- and clusters of galaxies.
ter on those scales as well. Now we fi nd that irregularities What is so different about the behavior of dark matter
in normal matter in the early universe were too slight to and normal matter in the early universe? First, pressure
form galaxies. So where will we turn for an answer to the waves and radiation did a very good job of smoothing out
question of how structure in the universe did form? You ripples in the distribution of normal matter in the early
guessed it: dark matter. universe. Feebly interacting dark matter would be immune
In Chapters 19 and 21 we learned that there is much to these processes, so clumps of dark matter survived long
more dark matter in the universe than there is normal mat- after bumps in the normal matter had been smoothed out, as
ter—consisting of neutrons, protons, and electrons—which illustrated in Figure 22.5. Second, the dark matter in these
we fi nd in stars and gas throughout the universe. Yet our clumps does not glow, so we do not see it directly when we
discussion of Big Bang nucleosynthesis (in Chapter 18) look at the cosmic background radiation. Although these
revealed that the amount of normal matter we see in the clumps of dark matter do cause slight gravitational redshifts
universe predicts just the right abundances of light ele- in the light coming from normal matter, the resulting varia-
1 Clumps of dark matter, normal

FIGURE 22 . 5 Radiation
matter, and radiation emerge
from the very early universe. pressure and other processes
2 Pressure of radiation in the early universe smoothed
Dark matter smooths out normal out variations in normal matter,
matter clumps as the but irregularities in the dark
universe expands…
matter survived to become the
seeds of galaxy formation.
Dark matter
Normal matter 3 …leaving behind dark matter

and radiation clumps in a far more uniform
background of normal matter
Time and radiation.
4 Very slight irregularities remain

in the cosmic background
radiation that we see.
Normal matter and radiation

tions fit well with current observations of the CBR. Thus, exist. We have seen that neutrinos interact with matter so
clumps of dark matter could have (and, according to our feebly that they are able to flow freely outward from the
theories of the very early universe, should have) existed in center of the Sun, passing
the early universe, even though we are unable to see evi- through the dense overlying Hot dark matter
dence of them directly. layers of matter as if they were consists of less
However, just because we cannot see clumps of dark mat- not there. There is no question massive, rapidly
ter in the early universe does not mean they had no effect. In that the universe is filled with moving particles.
the same way that dark matter dominates the gravitational neutrinos. Calculations indi-
fields of today’s galaxies and clusters, the mass of dark mat- cate that about 300 million of these cosmic relics of the Big
ter controlled the growth of gravitational instabilities in Bang fill each cubic meter of space. Measurements of the
the early universe. neutrino mass obtained at the Super-Kamiokande detector
in Japan, at the Sudbury Neutrino Observatory in Canada,
and by the Main Injector Neutrino Oscillation Search
(MINOS) experiment in the United States suggest that neu-
There Are Two Classes of Dark Matter trinos might account for as much as 5 percent of the mass
Here is the story of galaxy formation in a nutshell. Dark mat- of the universe. Although this percentage is not enough to
ter in the early universe was much more strongly clumped account for all of the dark matter in the universe, it may be
than normal matter. Within a few million years after recom- enough to have had a noticeable effect on the formation of
bination, these dark matter clumps pulled in the surround- structure.
ing normal matter. Later, gravitational instabilities caused The different effects of cold and hot dark matter on struc-
these clumps to collapse. The normal matter in the clumps ture formation have to do with differences in the way the
went on to form visible galaxies. This story seems plausible two types of dark matter clus-
enough, but the details of how this happened depend greatly ter in a gravitational field. The Cold dark matter
on the properties of dark matter itself. Even though we do faster an object is moving, the forms galaxies.
not yet know exactly what dark matter in the universe is harder it is for gravity to hold
made of, we can talk about two broad classes of dark matter on to it. This is the same basic idea that we used to explain
that are based on how it behaves. why hydrogen and helium are able to escape Earth’s atmo-
One possibility is that dark matter consists of feebly sphere but molecules such as oxygen and nitrogen are not.
interacting particles that are moving about relatively slowly, It is also the same idea we used to explain why the hot gas
like the slowly moving atoms in elliptical galaxies extends far beyond their visible bound-
and molecules in a cold gas. Cold dark matter aries. Slow-moving particles are more easily corralled by
For obvious reasons, this type consists of relatively gravity than are fast-moving particles, so particles of cold
of dark matter is called cold massive, slowly dark matter clump together more easily into galaxy-sized
dark matter. There are several moving particles. structures than do particles of hot dark matter. As a result,
candidates for cold dark mat- models show that on the largest scales of massive super-
ter. It is possible that cold dark matter consists of tiny black clusters, both hot dark matter and cold dark matter can form
holes that might have been produced in the early universe. the kinds of structures we see; but on much smaller scales,
Few physicists and cosmologists favor this idea, however. only in models of cold dark matter can the dark matter
Most think instead that cold dark matter consists of an clump enough to produce structures like the galaxies we
unknown elementary particle. One candidate is the axion, see filling the universe. The result of the calculations of
a hypothetical particle fi rst proposed to explain some galaxy formation is clear: to account for the formation of
observed properties of neutrons. Even though axions would today’s galaxies, we need cold dark matter.
have very low mass, they would have been produced in
great abundance in the Big Bang. Another candidate is the
photino, an elementary particle related to the photon. Some A Galaxy Forms within a Collapsing
theories of particle physics predict that the photino exists
and has a mass about 10,000 times that of the proton. Our
Clump of Dark Matter
state of knowledge about the particles that make up cold We can best see how models of galaxy formation work by
dark matter could soon change; photinos might be detected following the events predicted by the models step by step.
in current particle accelerators, and experiments are under Let’s consider the case of a universe made up of 90 percent
way to search for axions and photinos that are trapped in cold dark matter and 10 percent normal matter, clumped
the dark matter halo of our galaxy. together in a manner consistent with observations of the
Since the fi rst class of dark matter is called cold dark cosmic background radiation. On the scale of an individual
matter, the other class of dark matter is quite logically galaxy, the effect of the cosmological constant is so small
known as—are you ready?—hot dark matter. Hot dark mat- that it can be ignored.
ter consists of particles that are moving very rapidly. Neu- Figure 22.6a shows the state of affairs of one clump of
trinos are one example of hot dark matter that we know dark matter at the time of recombination. The dark matter
is less uniformly distributed than normal matter; but over- closer (Figure 22.6d). The clump of cold dark matter is now
all, the distribution of matter is still remarkably uniform. given its shape by the orbits of its particles, in the same way
By a few million years after recombination (Figure 22.6b), that an elliptical galaxy is given its shape by the orbits of
the universe of our model calculation has expanded sever- the stars it contains.
alfold. Spacetime is expanding, so the clump of dark matter Unlike dark matter (which cannot emit radiation), the
is also expanding. However, the clump of dark matter is not normal matter in the clump is able to radiate away energy
expanding as rapidly as its surroundings are, because its and cool. As the gas cools, it
self-gravity has slowed down its expansion. The clump now loses pressure; and as it loses Normal matter cools
stands out more with respect to its surroundings. Another pressure, it collapses. Models and falls toward the
change that has taken place is that the gravity of the dark show that small concentra- center of the dark
matter clump has begun to pull in normal matter as well. By tions of normal matter within matter halo.
the stage shown in Figure 22.6c, normal matter is clumped the dark matter clump collapse
in much the same way as dark matter. under their own gravity to form clumps of normal matter that
A ball thrown in the air will slow, stop, and then fall range from the size of globular clusters to the size of dwarf
back to Earth. In like fashion, the clump of dark matter will galaxies. These clumps of normal matter then fall inward
stop expanding when its own toward the center of the dark matter clump, as shown in Fig-
self-gravity slows and then Clumps of dark ure 22.6e. We discussed a similar chain of events in Chapter
stops its initial expansion. By matter can collapse 15: the collapse of a molecular-cloud core on its way to becom-
the time the universe is about only so far. ing a star. It is important to note that, according to models,
a billion years old, the clump gas in our universe can cool quickly enough to fall in toward
of dark matter has reached its maximum size and is begin- the center of the dark matter clump only if the clump has a
ning to collapse (see Figure 22.6c). The collapse of the dark mass of 108–1012 solar masses (M⊙). This is just the range of
matter clump stops when the clump is about half its maxi- masses of observed galaxies! This agreement between theory
mum size, however, because the particles making up the and observation is an important success of the theory that
cold dark matter are moving too rapidly to be pulled in any galaxies form from cold dark matter.
1 At recombination, dark matter

clumps exist in a relatively
uniform background of normal FIGURE 22 .6 Stages in the
matter and radiation. 2 A few million years later, gravity formation of a spiral galaxy
is slowing the expansion of a from the collapse of a clump of
dark matter clump. cold dark matter.
3 Within a few hundred million
years the clump reaches its
maximum size. Normal matter
has been drawn into the clump.
(a)
Dark Normal (b)

matter matter
(green) (red) (c)
4 Normal and dark matter continue

collapsing until the dark matter can
collapse no further.
5 Normal matter, which can cool

by radiation, continues to collapse,
6 …and finally into first into smaller clumps…
a spiral galaxy.
(d)
(f)
(e)
As discussed in Connections 22.1, there are many simi- And just as the accretion disk around the Sun provided the
larities between the collapse of a molecular cloud to form raw materials for the planets of our Solar System, the disk
stars and the collapse of a clump in the early universe to formed by the collapse of each protogalactic clump becomes
form galaxies. Another process that we saw at work in star the disk of a spiral galaxy.
formation also comes into play at this point. The clumps
of matter collapsing to form galaxies do not exist in isola-
tion. They have been tugged on by the gravity of neighbor-
ing clumps and have been pushed around by the pressure
Searching for Signs of Dark Matter
waves that ran through the young universe, smoothing out Dark matter has come to play a very important role in our
its structure. As a result, each protogalactic clump has a understanding of the universe. We might reasonably be
little bit of rotation when it begins its collapse. As normal uncomfortable having our models of galaxy formation rely
matter falls inward toward the center of the dark matter so heavily on something we cannot see directly. Fortunately,
clump, this rotation forces much of the gas to settle into a dark matter shows itself in a variety of ways. The flat rota-
rotating disk (Figure 22.6f), just as the collapsing cloud that tion curves of spiral galaxies, for example, are compelling
was to become our Sun settled fi rst into an accretion disk. evidence for the existence of extended halos made up of dark
C O N N E C T I O N S 2 2 .1
Parallels between Galaxy and Star Formation

As you read about galaxy formation, you might fi nd it enlight- ture also climbs, and it, too, reaches a point where there
ening to think back to our discussion of star formation in Chap- is a balance between gravity and the thermal motions of
ter 15. Both processes involve the gravitational collapse of vast the dark matter. However, dark matter is not able to radiate
clouds to form denser, more concentrated structures. To help away energy, so once this balance is reached, the collapse
you with the comparison, we list here a few of the similarities of the dark matter is over. Only the normal matter within
and differences between the two. the cloud of dark matter is able to radiate away thermal en-
ergy and continue collapsing. That is why normal matter
Gravitational instability. In both star and galaxy formation,
collapses to form galaxies, while the dark matter remains
the collapse begins with a gravitational instability. Regions
in much larger dark matter halos.
only slightly denser than their surroundings are pulled
As galaxies form, dark matter remains in extended
together by their own self-gravity. As the matter in these
halos. Dark matter is too hot to settle into galaxy disks.
regions becomes more compact, gravity becomes stronger,
It is far too hot to become concentrated into even smaller
and the collapse process snowballs. One key difference is
structures such as molecular clouds or to take part in the
that for a galaxy to form, the dark matter clump must col-
formation of stars. Dark matter may be the dominant form
lapse rapidly enough to counteract the overall expansion
of matter in the universe, and it may determine the struc-
of the universe itself.
ture of galaxies; but dark matter can never collapse enough
Fragmentation. In both cases the original cloud separates to play a role in the processes that shape stars, planets, or
into smaller pieces as a result of the gravitational instabil- the interstellar medium.
ity. However, the order of fragmentation differs between
star and galaxy formation. In molecular clouds, fi rst large Angular momentum and the formation of disks. Conserva-
regions begin to collapse, and then they fragment further to tion of angular momentum is responsible for the formation
form individual stars. In contrast to this “top-down” proof disk galaxies, just as it is responsible for the formation
cess, galaxy formation is “bottom-up”: smaller structures of the accretion disks around young stars. The Milky Way
collapse fi rst and then merge to form galaxies and eventu- and the Solar System are both flat for the same reasons.
ally assemblages of galaxies. The origin of the angular momentum is different, though.
Whereas turbulent motions within star-forming molecular
Compression, heating, and thermal support. As an interstel- clouds produce the net angular momentum for stellar disks,
lar molecular cloud collapses, its temperature climbs and gravitational interactions with nearby clumps are respon-
the pressure in the cloud increases. The higher pressure sible for the angular momentum of the Milky Way.
would eventually be enough to prevent further collapse,
except that the cloud core is able to radiate away thermal The end product. Once a stellar accretion disk forms, most
energy. That is the bright infrared radiation that enables us of the matter moves inward and is collected into a star. In
to see star-forming cores. Compare this process with galaxy contrast, much of the matter in a spiral galaxy remains in
formation: As a dark matter clump collapses, its tempera- the disk.
FIGURE 22 .7 (a) An X-ray

(a) For gravity to hold (b) …this galaxy cluster image of the massive cluster
onto this hot X-ray must be more massive Abell 1689, about 2.3 billion
emitting gas… than the galaxies we see.
light-years away, shows that the
cluster is full of hot, tenuous
gas. (b) An image of the same
field of view taken in visible
light. (c) This combined image
of Abell 1689 shows both the
galaxies detected by the Hubble
Space Telescope and the X-ray
image of the 100-million-K gas
taken with the Chandra X-ray
Observatory.
again ask, How strong must gravity be to hold this cluster

(c) Galaxy clusters are together? And again, the answer is that the total mass of
mostly dark matter. the clusters, including dark matter, must be about 10 times
greater than the normal matter they contain.
A third way to look for dark matter relies on the predic-
tions of Einstein’s general theory of relativity, which states
that mass distorts the geometry of spacetime, causing even
light to bend near a massive object. In particular, light from
a distant object is bent by the gravity of a galaxy or cluster
of galaxies, so that images of the distant object can be seen
magnified on either side of the intervening galaxy or cluster.
The result is a gravitational lens (Figure 22.8a).
We have already encountered gravitational lensing, in
Chapter 20’s discussion of MACHOs, where we found that
lenses can make background
objects appear brighter. When Gravitational lenses
considering the complex are one way to
three-dimensional geometry measure the masses
of gravitational lenses, we find of galaxy clusters.
that lenses can show multiple
matter. Similarly, the ability of elliptical galaxies to hold on images of background objects, and that these magnified
to hot gas convincingly demonstrates that they contain far images are often drawn out into an arclike appearance. The
more mass than can be accounted for by their stars alone. greater the gravitational lensing, the greater the mass that
We can use a similar approach to look for evidence of must be in the cluster. Figure 22.8b shows an image of a gal-
dark matter on larger scales. Galaxy clusters are filled with axy cluster that is acting as a gravitational lens for a number
extremely hot gas (10 million to 100 million kelvins [K]), of background galaxies. By analyzing such images, we can
which occupies the space between galaxies (Figure 22.7). determine the mass of the lensing cluster. Figure 22.8c shows
Even though this gas is extremely tenuous, the volume of six examples of more distant gravitational lenses from the
space that it occupies is enormous: the mass of this hot gas Cosmological Evolution Survey (COSMOS), which is using
can be up to five times the mass of all the stars in that clus- most of the major space-based telescopes and several ground-
ter. X-ray spectra show that this gas contains significant based telescopes to catalog these lenses.
amounts of massive elements that must have formed in stars. Regardless of how we measure the masses of galaxy
This chemically enriched gas has been either blown out of clusters—by looking at the
galaxies in winds driven by the energy of massive stars, or motions of their galaxies, by Dark matter
stripped from galaxies during encounters with neighbor- measuring their hot gas, or by dominates the mass
ing galaxies. Were it not for the gravity of the dark matter using them as gravitational of galaxy groups
filling the volume of the cluster, this hot gas would have lenses—the results are the and clusters.
dispersed long ago. same. By mass, galaxy clus-
We can also look at the motions of the cluster’s galaxies ters, like the galaxies they are made of, are dominated by
(which behave like very large “atoms” in this context) and the dark matter they contain.
Light from a distant galaxy We see a magnified,

passes through curved space distorted image through
around a foreground cluster. this gravitational “lens.”
(a) FIGURE 22 .8 (a) The
geometry of a gravitational
lens. A mass can gravitationally
focus the light from a distant
object, thereby magnifying
and distorting the image. (b) A
Hubble Space Telescope image
of the cluster Abell 2218,
showing many gravitationally
(b)
lensed galaxies, seen as arcs. (c) Gravitational lenses
from several distant individual galaxies.
Foreground cluster
galaxies
Gravitationally lensed
background galaxies 22.4 Observations
Help Fill Gaps
in the Models
The picture of galaxy formation given in the
previous sections is much cleaner and more ide-
alized than what happened in reality. In our
hierarchical clustering picture, smaller struc-
tures collapsed fi rst. When an original clump
of galaxy-sized dark matter contained a sub-
structure, that substructure would itself have
(c)
collapsed into subgalaxy-sized objects before
the galaxy as a whole fi nished forming. This
hierarchical model gives us a way to understand
the existence of such objects as the globular star
clusters and dwarf companion galaxies near the
Milky Way and other galaxies. These objects
formed inside the larger clump of cold dark mat-
ter when the luminous matter in the clump was
still settling toward what would become the disk
of the galaxy.
Mergers Play a Large Role

in Galaxy Formation
As we saw in star formation, galaxies do not
always form in isolation. It is likely that the same
larger clump often produced more than one gal-
axy, and that these galaxies later interacted or
even merged. The merging of two spiral (or pro-
tospiral) galaxies made all sorts of commotion likely. The tidal interactions between the galax-
ies and the collisions between gas clouds in the galaxies probably triggered many regions of star
formation throughout the combined system. Recall our discussion of galaxy mergers at the end of
Section 19.4. The collision and merging of two giant clusters is nicely illustrated in Figure 22.9,
which shows the Bullet Cluster of galaxies at a redshift of z = 0.3. The stars contained within the
galaxies of this system have less total mass than the X-ray–emitting gas, which is itself much less
FIGURE 22 .9 The Bullet

Cluster of galaxies, at a redshift
of z = 0.3, represents the
collision of two giant clusters
of galaxies in the process
of merging. The galaxies in
this system constitute less
mass than the X-ray–emitting
gas (shown in red), which
is itself much less massive
than the dark matter (shown
in blue) measured from the
gravitational lensing by this
cluster.
massive than the dark matter, whose mass we deduce from clear consequences for what we should expect to see when
the gravitational lensing produced by the cluster. The fact looking back to a time in the history of the universe when
that the cluster’s galaxies, hot gas, and dark matter are so galaxies were actively forming. Rather than the well-formed
displaced from one another is testament to the violence of spirals and ellipticals that dominate today’s universe, the
the collision that formed this cluster. early universe should have contained many clumpy, irregu-
The merging of galaxies also answers another puzzle. We lar objects. As expected, images of the young universe such
have seen how spiral galaxies could form in the early uni- as the Hubble Space Telescope image shown in Figure 18.1
verse, but what about the for- are filled with lumpy, distorted objects that are still in the
mation of elliptical galaxies? Ellipticals form process of settling in. We show some of these high-redshift
Ellipticals are now thought to from the mergers objects as insets in Figure 22.11.
result from the merger of two of spirals. When we look at the bright, active galaxies that populate
or more spiral galaxies. Figure the early universe, it is important to remember that what we
22.10 shows a computer simulation of such a merger. If the actually see is only the tip of the iceberg. The luminous gal-
merging galaxies were not originally spinning in the same axies are small in comparison with the dark matter clumps
direction, then the resulting merged galaxy loses its disklike that surround them. These clumps are still found in today’s
character. The dark matter halos of the galaxies merge, and universe as the invisible dark matter halos responsible for
the stars eventually settle down into the bloblike shape of the flat rotation curves of spiral galaxies.
an elliptical galaxy. As this picture suggests, elliptical gal-
axies are known to be more common in dense clusters where
mergers are likely to have been more frequent. Star Formation Leaves a
Such events also played a major role in feeding super-
massive black holes, which themselves must have formed
Gap in the Models
very early in the collapse of galaxies and protogalaxies. The We would like to be able to turn our models of collapsing
same mergers and interactions that formed the giant galax- clumps of cold dark matter into detailed models of what
ies we see today also provided the fuel to power the quasars galaxies in the young universe should look like, but we have
and other AGNs that were common in the early universe. not yet reached such a level of sophistication. One major
With galaxies crashing together, supermassive black problem is that we lack a good understanding of how stars
holes forming, quasars flooding the young universe with form in young galaxies. We know that most of the normal
intense radiation and powerful jets, and star formation run- matter in a galaxy winds up as stars; but when and where
ning amok, galaxy formation in the young universe must do those stars form, and how long does it take? Although
have been a violent, messy business. This conclusion has we can say a great deal about how individual stars form
FIGURE 22 .10 In this

computer simulation, two
spiral galaxies merge to
form an elliptical galaxy.
Numbers indicate time since
the interaction began, where
1.00 is the time it took for the
galaxies to fall together.
in our galaxy, as we did in Chapter 15, we do not yet have helium gas within the overall clump of matter of which
theories or models that predict such important pieces of the they were a part.
puzzle as what fractions of stars should have what masses. Although a detailed understanding of star formation in
Nor do we clearly understand the differences between star a forming galaxy is in the future, we can still make a num-
formation in the early universe and the star formation ber of factual statements about when and where stars form
going on around us today. In a universe devoid of massive in a collapsing galaxy. These
elements, there were no dense, dusty molecular clouds clues come from the study of Star formation
from which stars might form, and no spiral disks in which our own galaxy. The fact that began very early in
such clouds might congregate. Instead, the fi rst stars must the atmospheres of even the collapsing galaxies.
have formed from the collapse of clouds of hydrogen and oldest halo stars in the Milky
FIGURE 22 .11 Another

picture of the Hubble Ultra
Deep Field (HUDF)—the farthest
glimpse into the universe’s
distant past. The insets show
high-redshift objects that are
examples of the youngest
galaxies in the process of
forming and merging.
Way contain some amount of massive elements tells us that current models, in which galaxies formed from collapsing
a significant amount of star formation must have taken place clumps of cold dark matter.
very early during the collapse of the Milky Way. If we could Uncertainties in our understanding of star formation
closely inspect a collapsing protogalactic clump in the stages complicate quantitative comparisons between models of
depicted in Figures 22.6d and e, we would expect to see stars galaxy formation and observations of the early universe.
already forming and supernovae exploding from place to Fortunately, though, when we talk about structure on scales
place in the collapsing halo. The halo stars that we see today much larger than galaxies, star formation becomes less of
must have formed while the Milky Way was still collapsing, an issue. To understand the formation of clusters and larger
before the gas from which they formed coalesced into the structures, we need worry only about the way that galaxy-
disk. In contrast, disk stars (all of which have relatively high sized clumps of matter themselves fall together under the
abundances of massive elements) formed from gas that was force of gravity.
enriched by early generations of halo stars before it settled Figure 22.12 shows the results of one computer simula-
into the galaxy’s disk. tion that follows the motions of millions of clumps of dark
Thus, although we cannot say exactly how stars formed matter as they fall through
in the early universe, we can say that star formation must space under their mutual Clusters, filaments,
have been going on throughout the process of galaxy forma- gravitational attraction. The and voids form after
tion. The pattern of stellar ages and massive-element abun- scale of this simulation is so galaxies form.
dances evident in our galaxy today fits naturally with our large that we cannot follow
the outcome of individual galaxies but instead are looking

at the overall distribution of dark matter. The simulation Scale factor
shows that the smaller-scale structures develop fi rst. Dur- RU = 0.03
ing the fi rst few billion years, dark matter falls together into
structures comparable in size to today’s clusters of galaxies.
Only later do the spongelike filaments, walls, and voids
become well defi ned. The young universe
was very uniform.
The similarities between the results of the models and
observations of large-scale structure, such as those shown in
Figure 22.4, are quite remarkable. Not just any model will do,
however. Only model universes with certain combinations
of shape, mass, nature of ripples, type of dark matter, and
values for the cosmological constant will produce structure RU = 0.09
similar to what we actually see. This is a very important
result. Our models contain assumptions consistent with
our knowledge of the early universe, and they predict the
Gravity caused dark matter
formation of large-scale structure similar to what we actu-
clumps to expand more slowly
ally see in today’s universe. Figure 22.13 provides a succinct than the universe as a whole…
summary of the galaxy formation process, in which smaller
objects form fi rst and merge into ever-larger structure, lead-
ing ultimately to the Hubble Ultra Deep Field picture shown
in Figures 22.11 and 18.1.
RU = 0.16
Peculiar Velocities Trace the

Distribution of Mass
…and eventually to collapse,
In our comparison of models and observations, there is an drawing in normal matter.
important caveat: whereas the models show where the mass
is, our observations show
where the light is. We have The distribution
already noted that our under- of light may not be RU = 0.25
standing of star formation in the same as the
the early universe is incom- distribution of mass.
plete at best. What if a clump
of dark matter formed and collapsed, but for some reason
the normal matter associated with that clump did not pro- Gravity first pulled matter
duce stars? Such clumps could still contain a significant into smaller structures…
amount of mass and could even affect the geometry of the
universe, but they would themselves not be seen in images RU = 0.50
of the sky.1
There is a way to get at the question of the overall dis-
tribution of dark matter more directly. If the clumping of
galaxies is a result of the action of gravity, we should be
able to look at how galaxies are moving and see them fall-
ing together. In fact, measurements of the way galaxies fall
together should enable us to infer the gravitational field
RU = 1.0 (Today)
they are experiencing, and hence the overall distribution
of dark matter.
1
Clumps of dark matter that are devoid of stars might be revealed
indirectly by gravitational lensing.
…and later pulled galaxies and

clusters of galaxies into the large-
scale structure we see today.
FIGURE 22 .12 A computer simulation of the formation of very
large-scale structure in a universe filled with cold dark matter.
FIGURE 22 .13 A schematic view of how structure formed in the universe, from smaller systems to
larger ones, leading ultimately to the kind of Hubble Ultra Deep Field picture shown in Figure 22.11.
In Chapter 18 we learned that the cosmic background comes from a region dubbed the Great Attractor, so called
radiation is blueshifted in one direction in space and red- because of its strong gravitational tug.
shifted in the opposite direction, telling us of our peculiar The picture painted in this chapter of the formation of
velocity relative to the CBR. Peculiar velocities of galaxies large-scale structure is almost certainly correct in its broad
other than our own are difficult to measure. They require outlines. Even so, uncertainties in star formation, our
us to accurately determine the distances to galaxies using understanding of the exact nature of dark matter, and our
standard candles, and then to use those distances along with knowledge of the shape of the universe limit the amount
Hubble’s law to predict what the redshifts of those galaxies of detail in any comparisons we can make between models
should be. Comparison of observed redshifts with redshifts and observations. In the fi rst decade of the 21st century,
predicted on the basis of Hubble expansion then tells us how these areas were a focus of very active research. Sev-
fast these galaxies are moving with respect to the cosmic eral projects—such as the Sloan Digital Sky Survey (cur-
background radiation (at least along our line of sight). rently collecting large amounts of data on the distribution
Figure 22.14a shows one such map of the peculiar veloci- and redshifts of galaxies), the WMAP satellite, and ESA’s
ties inferred for galaxies in our part of space. If we use obser- Planck mission—can dramatically improve our knowledge
vations of peculiar velocities of the structure in the CBR. Larger and more sophisticated
to map concentrations of mass Peculiar velocities models are being built all the time, while studies of star-
in this part of the universe, we tell us how dark forming regions in our galaxy and in galaxies with lower
get a picture that looks like matter is distributed. massive-element abundances are continuing to improve
two mountains surrounded by our understanding of this important process. There is also
foothills and valleys, as shown in Figure 22.14b. Unfortu- hope that new theories and laboratory experiments will
nately, these two “mountains” of mass, which exert the great- identify the particles that make up dark matter. A more
est gravitational pulls on our own galaxy, are located in complete understanding of the formation of galaxies and
regions of the sky that are heavily obscured by the Milky large-scale structure in the universe will not likely come
Way’s dusty disk. The largest pull that we are experiencing in a single “eureka” moment, but will instead arrive in bits
22.5 Multiple Multiverses 623
A map of peculiar
(a) (b)
velocities of galaxies…
FIGURE 22 .14 (a) A map of
Great Attractor peculiar velocities of galaxies in
Great
Attractor our neighborhood. Arrows show
the velocities of material lying
in the plane of our own local
supercluster. Motions seem to
be converging onto the position
725 Mly
of the Great Attractor. The
black cross marks the location
of the Milky Way Galaxy. (b) A
map of the mass distribution
Average
inferred from observations of
density
peculiar velocities.
…reveals the presence of of universe
mass concentrations whose
725 Mly
gravity attracts those galaxies.
and pieces throughout the fi rst part of the 21st century as It is true, of course, that as our universe ages and expands
a result of advances in all of these different areas. gradually, the size of our observable universe expands as well,
and we can see more and more
material. However, when the Our own observable
universe becomes completely part of the universe
dominated by dark energy (see will become isolated
22.5 Multiple Multiverses Chapter 21), it will expand so from all others.
fast that the separate observ-
We have described in detail the origin of structure in our able universes will never be observable from each other. Our
universe from the scale of asteroids and planets all the way own observable part of the universe will become truly iso-
to the largest clusters and superclusters. We’ll now turn to lated from all the other parallel observable universes.
ideas that are more speculative. Have you ever wondered What are these other parallel universes like? We may never
if there is even more structure on an even larger scale— know for sure in detail, but we can say several things. First,
perhaps parallel universes to our own either separated in if the cosmological principle
space or even occupying the exact same space as ours? These holds, then on large scales, There are parallel
hypothetical universes seem like material for science fiction each of these observable uni- universes identical
rather than for a serious science textbook, but it might not verses should look pretty much to our own.
be as strange as you may think. In fact, many cosmologists like our own. But the details
think seriously about the idea of multiverses, or collections could be very different among these parallel universes. Still,
of parallel universes. in an infinite universe there must be an infinite number of
Let’s begin with the simplest example of such paral- observable universes exactly like ours, with an exact genetic
lel universes, illustrated in Figure 22.15. We know that and experiential copy of you reading an identical version of
the age of the universe—that is, the amount of time that 21st Century Astronomy. We know this because if our own
has passed since the Big Bang—is 13.7 billion years. That observable universe is cooler everywhere than about 108 K,
means that light reaching us today can have traveled a then there can be no more than 10118 particles in the observ-
distance of only about 13.7 billion light-years (Gly). (Actu- able universe, and only so many ways that those particles can
ally, the real distance is a bit larger than that because of be distributed. If you then ask how far you must go before you
the past expansion of the universe, but we’ll ignore that are sure to find an observable universe just like—and we
118
difference.) Therefore, everything that we can possibly mean identical to—our own, the answer is about 1010 Mly.
observe today—that is, our observable universe—must Yes, that’s 10 raised to the power 10118, and for numbers that
be within a sphere of radius 13.7 Gly. We cannot possi- large, it doesn’t matter if you measure the distance in mega-
bly see anything outside our observable universe. Yet if light-years or in nanometers. So, in an infinite universe—as
the geometry of space is fl at, as our cosmological param- enormous as it might be—we still know how far away an
eters already suggest, then our universe is truly infi nite identical copy of you must be!
in size and must therefore contain an infi nite number of The parallel observable universes we just described rep-
disjoint observable universes like our own. These disjoint resent the fi rst of four types of multiverses about which cos-
universes are simply too far away for us ever to be able mologists have theorized (Figure 22.16a). The inflationary
to observe them. universe model forms the basis of the second type of multi-
The process of forming such bubbles is much like the

condensation of water vapor to form drops in clouds or on
the outside of a cold glass. In an eternally inflating universe,
the bubbles might form because of quantum fluctuations—
occasionally, because of quantum variations, a region may
slow down just a little bit from the rest of the universe, and
14,000
Mly
as a result the whole region may form a bubble whose inflat-
ing phase will soon end. Exactly the same physics should
govern all of the bubbles in such a universe, and depending
Our universe Parallel universe
on the theory of everything (TOE), it is possible that each of
the bubbles could have different physical constants.
We need to better understand the underlying physics of
the inflationary universe to know whether that possibility
is real. In an eternally inflating universe, our portion of the
universe is but one of these bubbles, which must itself be
fi nite in size. Given that these bubble universes are finite,
could we be living near the
edge of such a bubble? This New big bangs
Parallel universe
scenario is possible, but the are forever beginning
multiverses may be so huge in an eternally
that it is very unlikely we are inflating universe.
FIGURE 22 .15 Today, in anywhere near the edge. One
our universe, we can see out nice feature of eternal inflation is that it neatly answers the
118
1010 to only the finite distance that question of what there was before the Big Bang. Since the
Mly light can have traveled since universe has inflated and will inflate forever, there was no
the Big Bang. If we call that our beginning or end. Our own bubble or parallel universe sepa-
observable universe, then there rated from the rest of the universe at a time we call the Big
are many (perhaps an infinite Bang, but other bubbles are constantly separating and
number) of parallel observable
becoming their own parallel-universe big bangs.
universes. An infinite number
The third type of multiverse has its origins purely in
of parallel observable universes
quantum mechanics (Figure 22.16c). As we described in
would imply that there are
others exactly like our own. Chapter 4, the quantum description of the world is a prob-
abilistic description. There is a certain probability that a
radioactive element will decay within a day or that an atom
in an excited state will decay in a second. One interpretation
of this probability is that the atomic state either decays or it
doesn’t decay in that second, with the relative likelihoods
determined by quantum physics. A second interpretation,
however, is that the atom splits into two parallel universes:
one in which it decays and one in which it does not decay. If
we are observing that atomic state, watching for it to decay,
then we, too, split into two parallel universes: one in which
we see the decay and one in which we do not. With this
interpretation of quantum mechanics, new universes are
forming all the time and occupying the same space, even
Identical though they cannot communicate with one another.
parallel universe The fourth type of multiverse is characterized by paral-
lel universes that have different mathematical structure to
describe the different physics within these universes (Fig-
verse. Imagine a universe that undergoes inflation forever, ure 22.16d). This type is the all-encompassing case in which
with no beginning or end to the inflation. This idea, called almost any behavior for the universe is possible.
eternal inflation, was hypothesized by Russian-American It is fair to ask whether the idea of parallel universes,
physicist Andrei Linde (1948–), who realized that if such a or multiverses, is really science. Throughout this book we
universe exists, then individual regions could occasionally have emphasized over and over again that any legitimate
slow their expansion relative to the rest of the universe and scientific theory must be testable and ultimately falsifiable.
form an expanding bubble within a surrounding inflation- Are there tests capable of proving these multiverse ideas
ary universe (Figure 22.16b). to be wrong? Possibly. For example, we do test the Type I
22.6 The Deep Future 625
(a) Type I: Regions beyond our cosmic horizon (b) Type II: Other postinflation bubbles
(c) Type III: The many worlds of quantum physics (d) Type IV: Other mathematical structures
FIGURE 22 .16 Cosmologists have proposed four types of multiverses: (a) parallel
observable universes; (b) the inflationary model; (c) the quantum mechanical model; (d) parallel
universes with different physics.
multiverse, involving distinct observable bubbles, when we tests, the concern is that the tests are not truly meaning-
measure the isotropy of the CBR or the large-scale distri- ful. For example, you might want to test Newton’s theory of
bution of galaxies. Since this type follows inevitably from gravity by releasing an apple and watching whether it falls
our standard cosmological models, it makes predictions upward or downward. But this test is not very discerning—
that can be tested readily. The eternal inflation model is it could not always distinguish two sensible theories from
harder to test, because we will never directly observe par- one another. Similarly, for the multiverse hypothesis it is
allel universes. But if we obtain a theory of everything that still debatable whether the tests that seem possible would
predicts eternal inflation, and if that theory of everything be meaningful tests of the theory.
is itself falsifiable, then there is a connection between eter-
nal inflation and observation. In addition, if the bubbles in
eternal inflation have very different properties from one
another, we can also test whether such bubbles can expand
into a parallel universe in which human beings can evolve. 22.6 The Deep Future
If such parallel universes evolve too fast or don’t produce
stars and planets, then our very existence would be unlikely Let’s turn back now to our own observable universe and
and such models would be highly suspect. Types III and IV ask what the future holds in store for our universe and the
might also be modeled with quantum computers and with structure it has formed. In Chapter 21 we discussed how the
an ultimate TOE. universe evolved to its current state, and we found that if
There is considerable debate within the scientific com- the mass of the universe is large enough and the cosmologi-
munity as to whether the multiverse hypotheses, except for cal constant is small enough, gravity will win in the end.
the Type I case, can be truly falsified. Even if there are some Hubble expansion will eventually reverse, and the universe
FIGURE 22 .17 How

No structure structure in the universe
Dark Era evolves—from the Primordial
10100 Era to the present, and then
Galaxy-size
Black Hole Era to the future Dark Era—as
black holes decay
postulated by Fred Adams and
Gregory Laughlin.
1075
Stellar-mass
black holes decay
1050
Degenerate Era Neutron

Stars Degenerate stars decay
collide dwarfs decay
Time (years)
1025
Last star
Stelliferous Era Stars Galaxies Life
forms (no
form form begins
more gas)
Primordial Era Universe becomes transparent
1
Recombination
Electron
Electrons and
positrons annihilate
UV photon
+
γ-ray Primordial
Proton
photons Nucleosynthesis
10–25
− H He4
Inflation
10–50
Big
Bang
will collapse back into a state resembling that of the young, Using well-established physics, we might speculate how
hot universe. Galaxies, stars, planets, molecules, atoms, and the existing structures in the universe will evolve over
subatomic particles—all might cease to exist as matter is a very long time. In 1997, American astrophysicists Fred
replaced by pure energy. Perhaps from such a state, a new Adams and Gregory Laughlin published their calculations
universe would emerge. Early in the 21st century, however, of the great eras, past and future, in the history of the uni-
few cosmologists see such a “Big Crunch” in the future of verse. We illustrate these eras in Figure 22.17. During the
the universe. It appears that our universe is expanding at fi rst era, the “Primordial Era”—the fi rst 500,000 years after
an ever-accelerating rate. As we discussed in Chapter 21, if the Big Bang and before recombination—the universe was
the dark energy’s effects are increasing with time, the uni- a swarm of radiation and elementary particles. Today we
verse may expand so fast that it becomes infinite in size in live during the second era, the “Stelliferous Era” (“Era of
a fi nite time—the “Big Rip” that would tear all structure Stars”); but this era, too, will end. Some 100 trillion (1014)
apart. However, the cosmological parameters we measure years from now the last molecular cloud will collapse to
suggest that the universe will expand forever, perhaps even form stars, and a mere 10 trillion years later the least mas-
at an ever-accelerating pace. Does this mean the universe sive of these stars will evolve to form white dwarfs.
will go on expanding without end? Yes, but its constituents Following the Stelliferous Era, most of the normal mat-
will appear very different from what we see today. ter in the universe will be locked up in degenerate stellar
objects: brown dwarfs, white dwarfs, and neutron stars. years, even the largest of the black holes will be gone. A
During this “Degenerate Era,” the occasional star will still universe vastly larger than ours will contain little but pho-
flare up as ancient substellar brown dwarfs collide, merging tons with colossal wavelengths, neutrinos, electrons, posi-
to form low-mass stars that burn out in a trillion years or trons, and other waste products of black hole evaporation.
so. However, the main source of energy during this era will The “Dark Era” will have arrived as the universe continues
come from the decay of protons and neutrons and the anni- to expand forever—into the long, cold, dark night of eter-
hilation of particles of dark matter. Even these processes will nity. This may be the fi nal victory of entropy—the heat
eventually run out of fuel. In 1039 years, white dwarfs will death of the universe.
have been destroyed by proton decay (see Chapter 21), and If you think about it, the evolution of structure in our
neutron stars will have been destroyed by the beta decay of universe, from the earliest moments to the distant future,
neutrons (see Chapter 14.) is a journey of highs and lows. Any chaotic structure pres-
As the Degenerate Era comes to an end, the only signifi- ent in the Big Bang is soon smoothed out by the enormous
cant concentrations of mass left will be black holes. These expansion of the universe during inflation, the very time
will range from black holes when the seeds of later structure arise from quantum fluc-
with the masses of single stars Ultimately, all tuations. Then, until recombination, the radiation in the
to greedy monsters that grew structure in the universe resists the pull of gravity to form structure, after
during the Degenerate Era to universe will decay. which the dark matter begins to form galaxies, pulling the
have masses as large as those normal matter along with it. Then stars form and recycle
of galaxy clusters. During the period that follows—the their gas into newer stars, until all the gas is used up. Stellar
“Black Hole Era”—these black holes will slowly evaporate evolution eventually ceases, leaving behind white dwarfs,
into elementary particles through the emission of Hawking neutron stars, and black holes, which in turn ultimately
radiation (see Chapter 17). A black hole with a mass of a few must decay through quantum effects. This leaves our uni-
solar masses will evaporate into elementary particles in 1065 verse a very dull and lifeless place with no structure at
years, and galaxy-sized black holes will evaporate in about all—unless there are parallel universes still undergoing
1098 years. By the time the universe reaches an age of 10100 structure formation.
mental forces of nature. As you watch the clouds build

Seeing the Forest for the Trees before a thunderstorm, think of galaxies coalescing within
halos of dark matter. Glance at a crystalline snowflake as
As our journey nears its end, we have reached high it lands on the sleeve of your coat, and in your mind’s eye
ground, a vantage point from which we can look back see a star that condensed out of clouds of interstellar gas
and survey the terrain we have covered. From this per- and dust. The origin of structure is everywhere around
spective it becomes clear that from the beginning, our us and is no less remarkable for its familiarity.
journey has been guided by a single underlying quest: One of the lessons of our journey is that structure in
to understand how the structures we see in the universe the universe does not revolve around us. Just as Coper-
came to be. Whether we are talking about the formation nicus showed that we do not occupy a special place at
of galaxies and clusters, the origin and evolution of stars, the center of the Solar System, we have now seen that
the activity of black holes, the history of our Solar System, the material of Earth and of ourselves—the protons,
the geology of our own world, the changes that shaped neutrons, and electrons that comprise normal matter—
the universe itself during its earliest moments, or the eras is not the major constituent of the universe. In fact the
of the far future, 21st century astronomy (the science, as dark matter in our universe is essential for galaxies
well as the book) is organized around the desire to better like our Milky Way to even form. Our own constitu-
understand the origin of structure in the universe. ents could never coalesce to form the galaxies we see.
During the past century we have made remarkable And of course, without galaxies, there could be no stel-
strides toward this goal, and along the way a pattern has lar nucleosynthesis of heavy elements and therefore no
emerged. Look back at the quotation opening Chapter Earth. We have even seen that the mysterious dark matter
16. What statement remains true in all circumstances? itself is not the major constituent of our universe, which
“And this, too, shall pass away.” Structure is ephemeral. is dominated by the even more mysterious dark energy.
The universe is not about a fi nal destination; it is a place Indeed, we have learned that our own universe may not
of process and change. The things that “matter” are the even be special—just one among an infi nite number of
things that happen along the way! parallel universes.
When you see your breath fog up on a cold winter day, Another lesson we have learned is that the physics of
think about elementary particles freezing out of the early the smallest particles is really crucial on the very largest
universe, giving rise not only to matter but to the funda- scales. Dark energy and dark matter, described by par-
ticle theory, dominate the universe on its largest scales. In virtually all cases—from tiny planets to stars, black
And the quantum physics that is virtually irrelevant in holes, galaxies, and superclusters of galaxies—it is our
our macroscopic world turns out to be crucial on large old friend gravity that forms structure in our universe
scales. Quantum fluctuations may be responsible for par- and holds it together. Without gravity holding structure
allel universes being formed. In addition, it is quantum together, our universe would be a dull place indeed. But
fluctuations that produce the ripples in the early universe as we complete our journey in the next chapter, we will
that in turn seed the formation of galaxies and clusters of study one fascinating type of structure that does not form
galaxies, the very largest structures we can see. through the action of gravity: the structure we call life.
Summary *2. How do we know that the early universe was “far more
uniform than the blue of the bluest sky on the clearest
• Galaxies reside in groups, clusters, and larger structures, day,” as stated in Section 22.1?
which generally formed after galaxies did. 3. As clumps containing cold dark matter and normal
• On very large scales, the structures in the universe reside matter collapse, they heat up. When a clump collapses
mainly on filaments and walls surrounding large regions to about half its maximum size, the increased thermal
devoid of galaxies. motion of particles tends to inhibit further collapse.
Whereas normal matter can overcome this effect and
• Structure formed in the universe through the gravita- continue to collapse, dark matter cannot. Explain the
tional collapse of inhomogeneities in cold dark matter reason for this difference.
that arise from the early universe.
4. What are the main differences between galaxy groups
• Observed galaxies come from complex mergers, with and galaxy clusters?
the visible gas in galaxies cooling and falling inward to
form the visible stars, which are surrounded by a dark 5. Why is it reasonable to call our Local Group of galaxies
matter halo. a “group” rather than a cluster? Why is it safe to call our
Local Group a “cluster”?
• In the standard Big Bang model, our observable universe
6. What is the difference between a galaxy cluster and a
is but one of an infinite number of simultaneously exist-
supercluster? Is our galaxy part of either? How do we
ing observable universes, so there must be universes
know this?
identical to our own.
7. How do astronomers use the following to measure
• If our universe is eternally inflating and forming new
the amount of dark matter contained in a cluster of
bubbles, as some scientists have speculated, then it had
galaxies?
no beginning and has been expanding forever.
a. Motions of individual members of the cluster.
• Structure will continue to form in our universe for at b. Extremely hot gas that fills the intergalactic space
least another 1015 years, but our current understanding within the cluster.
of quantum effects suggests that they ultimately lead to c. Gravitational lensing by the cluster.
the decay of all structure, from planets to black holes.
8. Can a galaxy be located inside a void? Explain.
9. Is it likely that voids are filled with dark matter? Why
or why not?
includes algorithmically generated versions of these 10. Imagine that there are galaxies in the universe composed
questions, plus additional conceptual exercises. If mostly of dark matter with relatively few stars or other
your instructor assigns questions in SmartWork, log in at luminous normal matter. If this were true, how might
smartwork.wwnorton.com. we learn of the existence of such galaxies?
11. How are the processes of star formation and galaxy for-
mation similar? How do they differ?
Student Questions 12. What is the origin of large-scale structure?
THINK ING A BOUT THE CONCEP T S 13. Why is dark matter so essential to the galaxy formation
process?
1. Suppose you could view the early universe at a time
when galaxies were fi rst forming. How would it be dif- 14. Which do we believe is the correct evolutionary sequence:
ferent from the universe we see today? (a) small star clusters formed fi rst, which were bound
together into galaxies, which were later bound together difficult to model the star formation history of a young
in clusters and superclusters; or (b) supercluster-sized galaxy? Is this difficulty a failure of our theories?
regions collapsed to form clusters, which then later col-
**31. If 300 million neutrinos fill each cubic meter of space,
lapsed to form galaxies, which formed small clusters of
and if neutrinos account for only 5 percent of the mass
stars? Justify your answer.
density (including dark energy) of the universe, estimate
15. Why does the current model of large-scale structure the mass of a neutrino.
require that we include the effects of dark matter?
32. What is the approximate mass of
16. What do we think dark matter is made of, or at least isn’t a. An average group of galaxies?
made of? b. An average cluster?
c. An average supercluster?
17. Why do we think that some hot dark matter exists?
33. The lifetime of a black hole varies in direct proportion
18. Why do we ignore the cosmological constant (dark energy)
to the cube of the black hole’s mass. How much lon-
when thinking about the formation of a galaxy?
ger does it take a supermassive black hole of 1030 solar
19. Describe the stages of galaxy formation, for both the dark masses to decay compared to a stellar black hole of 3
matter and gas components. solar masses?
20. Why can gas consisting of hydrogen and helium col- 34. Knowing what elliptical galaxies are made of, estimate
lapse to a much smaller size than can nonbaryonic dark how old they must be. Knowing that ellipticals form via
matter? mergers of spirals, and knowing when we believe galax-
ies fi rst formed, estimate how long it took to complete
21. How does a roughly spherical cloud of gas collapse to
the merging events that formed the elliptical galaxies
form a disklike, rotating spiral galaxy?
we see today.
22. What are some of the observational signs that dark mat-
35. Currently, the Milky Way and Andromeda galaxies (M31)
ter exists?
are separated by about 2.3 million light-years and mov-
23. How do astronomers believe elliptical galaxies formed? ing toward each other at about 120 kilometers per sec-
Does this formation process explain the observed dif- ond (km/s). Estimate how long it may take for the two
ferences between spirals and ellipticals? to collide. Why do you think this may or may not be a
good estimate of how long it will take these galaxies to
24. On the basis of our understanding of galaxy formation,
fully merge?
describe how galaxies should appear as you look further
back in time. Are the features you described observed? 36. Compare the timescales you found in the previous two
questions. Is our impending merger with the Andromeda
25. How do we know that the dark matter in the universe
Galaxy typical of those that happened in the past? Can
must be composed mostly of cold, rather than hot, dark
this timescale place any constraints on the distances or
matter?
ages of galaxies that merged to form the elliptical galax-
26. Using broad strokes, describe the process of structure ies seen today?
formation in the universe, starting at recombination (half
*37. The Great Attractor is about 200 Mly from the Milky
a billion years after the Big Bang) and ending today.
Way, and by some estimates the Milky Way is moving
*27. How can we be certain that gravity, and not the other funda- toward it at a speed of about 1,000 km/s. Will we collide
mental forces, is responsible for large-scale structure? with it? Why or why not?
28. We have never observed a star with zero heavy metals 38. It is likely that the initial fluctuations leading to large-
in its atmosphere. What does this fact imply about the scale structure arose from quantum fluctuations in the
history of star formation in the early universe? early universe. How would the universe look different
today if those fluctuations were ten times bigger? What
about ten times smaller?
*29. Our theory of cosmology assumes that on large scales,

the structure in the universe is uniform no matter where
you look. Maps of structure, like that in Figure 22.4c,
support this assumption. Does the presence of large
masses like the Great Attractor violate this principle?
30. In previous chapters we painted a fairly comprehen- quizzes, plus links to premium content in SmartWork and
sive picture of how and why stars form. Why, then, is it the ebook. Visit www.wwnorton.com/studyspace.
If one could conclude as to the nature of the Creator from a
study of his creation it would appear that God has a special
fondness for stars and beetles.
J. B. S. HALDANE (1892–1964)
The search for signs of intelligent life in other stellar systems has been an ongoing pursuit of
modern science.
FPO
CHAP TER 23
Life
K E Y CONCEP T S
23.1 Life Is a Form
Our journey has brought us to a new understanding of
of Structure the universe and all it contains. Still, one very important
and very personal component remains: you, the reader
As we near the end of our journey, let’s briefly appreciate of this book. What combination of events—some
the distance we have covered. We have followed the origin probable, others much less likely—has led to your
of structure in the universe from the earliest instants after
existence as a sentient being, living on a small rocky
the Big Bang, when the fundamental forces of nature came
planet, orbiting a typical middle-aged star? Are you
to be, to formation of the galaxies and other large-scale
structure that is visible today. Earlier on our journey, we and your fellow humans unique, or are there others like
watched as stars (including our Sun) formed from clouds you inhabiting planets scattered throughout the vast
of gas and dust within these galaxies, and planets (includ- universe? Here we will learn that
ing our Earth) formed around those stars. We learned of • Like planets, stars, and galaxies, life is a form of
the geological processes that shaped early Earth into the structure that evolved through the action of the
planet we see around us today. In short, we have traced physical processes that shape the universe—the
the origin of structure from the instant the universe came universe may teem with life.
into existence up until the modern day. To help put these
events into perspective, Excursions 23.1 shows the relative • Life is a form of complex chemistry.
timing of the major events in the history of the universe if • Terrestrial life likely originated in Earth’s oceans.
scaled down to fit into a single 24-hour day that began with • Evolution is a natural process.
the Big Bang and ended today. But there is one piece of the
story that we have left out. Although the focus of our jour- • The conditions needed for life to form and evolve are
ney is astronomy, no discussion of how structure evolved commonplace throughout the universe.
in the universe would be complete without some consid- • Although the search goes on, there is not yet any
eration of the origin of the particular type of structure we evidence for extraterrestrial life.
refer to as life.
• All life on Earth must eventually come to an end.
631
632 Chapter 23 Life
obvious, but it is not immediately self-evident. Many sci-

23.2 Life on Earth entists will tell us that there is no universal defi nition of
life. Although we may have a feeling for what we mean
Before delving into the questions of how life fi rst appeared when speaking about terrestrial life, a universal defi nition
on Earth and how it has since evolved, we should fi rst would necessarily have to encompass very different forms
ask, How do we defi ne life? The answer might at fi rst seem of life that may occur elsewhere in the universe. To arrive
E X C U R S I O N S 2 3 .1
Forever in a Day
We have seen that the events that ultimately led to our exis- 4:05 P.M. A Mars-sized planetesimal crashes into Earth,
tence here as inhabitants of planet Earth stretched out over forming the Moon.
many billions of years—intervals that even astronomers have 5:20 P.M. The fi rst primitive life appears on Earth. It evolves
difficulty visualizing. We can sometimes get a better grasp of into the simplest life-forms: unicellular organisms such as
such enormous spans of time by compressing them into much bacteria, cyanobacteria, and archaebacteria.
shorter intervals with which we have day-to-day experience.
8:40 P.M. More complex single-celled organisms appear,
Try, then, to imagine the age of the universe and those impor-
making it possible for multicellular life to develop.
tant events associated with our origins as if they were all tak-
ing place within a single day. Our cosmic day begins at the 9:30 P.M. The fi rst multicellular organisms (fungi) appear
stroke of midnight: on dry land.
12:00:00 A.M. The embryonic universe is a mixed broth— 11:00 P.M. Multicellular organisms become abundant. Their
minute specks of matter suspended in a vast soup of radia- evolution paves the way for still larger and more complex
tion. The entire universe exists only as an extraordinarily life-forms.
hot bath of photons and a zoo of elementary particles. 11:20 P.M. The fi rst animals (fish) make the transition from
ocean to dry land—a major step toward our existence.
12:00:02 A.M. It is now just 2 seconds after midnight. All of
the early eras in the history of the universe have passed. The 11:35 P.M. The fi rst dinosaurs make their appearance. Vari-
fundamental forces have frozen out, Big Bang nucleosynthe- ous small animals appear as well, but they remain domi-
sis has formed the universe’s original complement of atomic nated by larger and more prevalent life-forms.
nuclei, and things have cooled down enough for atomic nu- 11:53:10 P.M. A large comet or asteroid crashes into Mexico’s
clei to combine with electrons to produce neutral atoms. Yucatán Peninsula. Seventy percent of all species on Earth
Both normal and dark matter are now available to create (including the dinosaurs) suddenly vanish. In the minutes
galaxies and stars, but that process will take some time. that follow, the mammals, being more adaptable to the
changed environment, gain prominence.
12:20 A.M. The fi rst stars, quasars, and galaxies appear. At
some point—we are not sure exactly when—our own galaxy 11:59:25 P.M. Our earliest human ancestors fi nally appear
becomes visible as star formation begins. Throughout the on the plains of Africa just 35 seconds before the end of
cosmic day, stars will continue to form. The more massive our cosmic day.
stars each go through their brief life cycles in only 5–10 11:59:59.8 P.M. Modern humans arrive with only a fi fth
seconds of our imaginary 24-hour clock. Each massive of a second to spare—a fraction of a heartbeat before the
star shines briefly, creates its heavy elements, and then day’s end.
disperses this material throughout interstellar space as it
12:00:00 A.M. Just as our cosmic day draws to a close, will
dies in a violent supernova explosion. Stars similar to our
a worldwide catastrophe occur? Will 21st century humans
Sun go through less dramatic life cycles, each lasting about
permanently scar the face of the planet, driving many of its
16 cosmic hours. Stars with mass less than 0.8 solar mass
life-forms into oblivion by polluting the atmosphere and
(M⊙) will last several dozen cosmic hours and will survive
poisoning the land, the rivers, and the oceans? Will our
beyond the end of our cosmic day.
progeny fi nally break free of their gravitational bondage
4:00 P.M. Our Solar System forms out of a giant cloud of gas to their planetary home and begin to claim the rest of the
and dust. Collapse of the cloud’s protostellar core and then the Solar System as their own? In comparison to all that came
appearance of the Sun and the planets—including Earth—all before, what happens in the next century will occur in a
take place within the span of a single cosmic minute. blur—less than a blink of the eye.
at a complete definition, we would have to take into account simple molecules, incapable of carrying out life’s complex
not only life-forms of which we are completely unaware, chemistry. However, early Earth had abundant sources of
but also those that exceed the limits of our imagination. We energy, such as lightning and ultraviolet solar radiation, that
will speculate about alien life later in the chapter. could tear these relatively simple molecules apart, thereby
creating fragments that could reassemble into molecules of
greater mass and complexity. As rain carried them out of
the atmosphere, these heavier organic molecules ended up
A Definition of Life in Earth’s oceans, forming a “primordial soup.”
For now, we’ll turn to a defi nition of life that meets our ter- In 1952, American chemists Harold Urey (1893–1981)
restrial experience. Life is fi rst of all a chemical process. and Stanley Miller (1930–2007) attempted to create condi-
Complex biochemical pro- tions similar to what they believed existed on early Earth.
cesses enable living organ- Life is a chemical To a laboratory jar containing liquid water as an “ocean,”
isms to grow and sustain process. they added methane, ammonia, and hydrogen as a primitive
themselves by drawing energy atmosphere; electric sparks simulated lightning as a source
from their environment. With the assistance of special mol- of energy (Figure 23.1). Within a week, the Urey-Miller
ecules such as ribonucleic acid (RNA) and deoxyribonucleic experiment yielded 11 of the 20 basic amino acids that link
acid (DNA), organisms are able to reproduce and evolve. All together to form proteins, the structural molecules of life.
terrestrial life involves carbon-based chemistry and employs Other organic molecules that are components of nucleic
liquid water as its biochemical “solvent.” (A more detailed acids, the precursors of RNA and DNA, also appeared
look at the chemistry of life appears in Section 23.3.) in the mix. We now know that Earth’s early secondary
With at least a basic idea of what we mean by life, we atmosphere contained carbon dioxide and nitrogen, rather
turn to the greater question of how it got its start here on than hydrogen, but recent experiments with more realistic
Earth. As we learned in earlier chapters, Earth’s second- atmospheric compositions have produced results similar
ary atmosphere was formed in part by carbon dioxide and to those of Urey and Miller. From laboratory experiments
water vapor that poured forth as a product of volcanism; a such as these, scientists have developed various models in
heavy bombardment of comets likely added large quantities which life got its start in Earth’s oceans, rich in prebiotic
of methane and ammonia to the mix. Note that these are all organic molecules.
The details of where and how these prebiotic molecules
evolved into the molecules of life is not so clear. Some biolo-
gists believe life began in the
ocean depths, where volcanic Terrestrial life
vents provided the hydrother- probably began in
mal energy needed to create Earth’s oceans.
the highly organized mole-
cules responsible for biochemistry (Figure 23.2). Others
believe that life originated in tide pools, where lightning
and ultraviolet radiation supplied the energy (Figure 23.3).
In either case, short strands of self-replicating molecules
may have formed fi rst, later evolving into RNA, and fi nally
into DNA, the huge molecule that serves as the biological
“blueprint” for self-replicating organisms.
Finally, we should briefly mention the suggestion by a
few scientists that life on Earth may have been “seeded”
from space in the form of microbes brought here by mete-
oroids or comets. Although this hypothesis might tell us
how life began on Earth, it does not explain its origin else-
where in the Solar System or beyond. We note that there is
no scientific evidence at this time to support the “seeding”
hypothesis.
The First Life

If life did indeed get its start in Earth’s oceans, when did it
FIGURE 23.1 Chemist Stanley Miller, with his experiment happen? In Chapter 6 we learned that young Earth suffered
simulating an early-Earth atmosphere. severe bombardment by Solar System debris for several
634 Chapter 23 Life
(a) (b)
FIGURE 23. 2 (a) Life on Earth may have arisen near ocean geothermal vents like this one. Similar
environments might exist elsewhere in the Solar System. (b) Living organisms around geothermal vents,
such as the giant tube worms shown here, rely on geothermal rather than solar energy for their survival.
hundred million years following its formation roughly 5 billion to 3.8 billion years. Fossilized stromatolites have
billion years ago. These were hardly the conditions under been found in western Australia and southern Africa, and
which life could form and living examples still exist to this day (Figure 23.4). We may
gain a foothold. However, Life on Earth may never know the exact date when life fi rst appeared on Earth,
once the bombardment had have begun more than but if the material found in the Greenland rocks is indeed
abated and Earth’s oceans had 3.6 billion years ago. of biological origin, then terrestrial life quickly took advan-
appeared, the opportunities tage of the favorable environment that followed cessation
for living organisms to evolve were greatly improved. The of Earth’s pummeling by errant planetesimal fragments.
earliest indirect evidence for terrestrial life is carbonized Thus, the current evidence suggests that the earliest life-
material found in Greenland rocks dating back to 3.85 bil- forms appeared within a billion years after the formation
lion years ago. Stronger and more direct evidence for early of the Solar System, and within 500 million years of the
life appears in the form of fossilized masses of simple end of young Earth’s catastrophic bombardment by leftover
microbes called “stromatolites,” which go back about 3.6 planetesimals.
FIGURE 23. 3 Life may have

begun in tide pools.
more oxygen did atmospheric levels of the gas begin to

rise. Oxygenation of Earth’s atmosphere and oceans began
about 2 billion years ago, and the current level was reached
only about 250 million years ago (see Chapter 8). Without
cyanobacteria and other photosynthesizing organisms,
Earth’s atmosphere would be as oxygen-free as the atmo-
spheres of Venus and Mars.
Biologists comparing genetic DNA sequences fi nd that
terrestrial life is divided into three major domains: Bacte-
ria, Archaea, and Eukarya—known informally as bacteria,
archaeans, and eukaryotes. Bacteria and archaeans are
organisms known as “prokaryotes” and consist of free-
floating DNA inside a cell wall. (Archaeans are similar in
appearance to bacteria but have a metabolic chemistry closer
to that of eukaryotes.) Prokaryotes are simple organisms,
lacking both cell structure and a nucleus (Figure 23.5a).
Even today, the largest component of Earth’s total biomass
FIGURE 23. 4 Modern-day stromatolites growing in colonies is in the form of simple microbes. Eighty percent of the
along an Australian shore. history of terrestrial life has been microbial. “Eukaryotes”
have a more complex form of DNA contained within the
cell’s membrane-enclosed nucleus (Figure 23.5b). The fi rst
The earliest organisms are examples of what biologists eukaryote fossils go back about 2 billion years, coincident
call “extremophiles.” These life-forms not only survive, but with the rise of free oxygen in the oceans and atmosphere,
thrive, under extreme environmental conditions. Extremo- although the fi rst multicellular eukaryotes did not appear
philes include organisms living in subfreezing environ- until a billion years later.
ments or in water temperatures as high as 120°C, which
occur in the vicinity of deep-ocean hydrothermal vents.
Other extremophiles are found under the severe condi-
tions of extraordinary salinity, pressure, dryness, acid-
Life Becomes More Complex
ity, or alkalinity. Among these earlier life-forms was an All life, whether prokaryotic or eukaryotic, shares a similar
ancestral form of “cyanobacteria,” single-celled organisms genetic code, which may have originated from a common
otherwise known as “blue-green algae.” Many scientists ancestor. DNA sequencing enables biologists to trace back-
have found supporting evidence that cyanobacteria were ward to the time when different types of life fi rst appeared
responsible for creating oxygen in Earth’s atmosphere by on Earth, and to identify the species they evolved from. Sci-
photosynthesizing carbon dioxide and releasing oxygen as entists have used DNA sequencing to establish the so-called
a waste product. Oxygen, however, is a highly reactive gas, evolutionary tree of life, which describes the interconnec-
and the newly released oxygen was quickly removed from tivity of all species. The tree has revealed some interesting
Earth’s atmosphere by the oxidation of surface minerals. relationships; for example, it places animals (including us)
Only when the exposed minerals could no longer absorb as most similar to fungi, which branched off the evolution-
(a) Typical prokaryote cell (b) Typical eukaryote cell FIGURE 23. 5 (a) A simple
prokaryotic cell contains little
more than the cell’s genetic
material. (b) A eukaryotic cell
contains several membrane-
enclosed structures, including
a nucleus that houses the cell’s
genetic material.
Strand of DNA Cell wall
Membrane enclosed
nucleus (contains DNA)
636 Chapter 23 Life
ary tree after slime molds and plants! The earliest primates shields us from deadly solar ultraviolet radiation. With a
branched off from other mammals about 70 million years ago, protective ozone layer in place, life was free to leave the
and the great apes (gorillas, chimpanzees, bonobos, and orang- oceans and move to land (Figure 23.6).
utans) split off from the lesser apes about 20 million years ago. The fi rst plants appeared on land about 475 million
DNA tests show that humans and chimpanzees share about years ago, and large forests and insects go back 360 million
98 percent of their DNA; the two groups are believed to have years. The age of dinosaurs began 230 million years ago and
evolved from a common ancestor about 6 million years ago. ended abruptly 65 million years ago, when a small aster-
The pace of evolution proceeded very slowly over the oid or comet collided with Earth and exterminated more
eons that followed the fi rst appearance of terrestrial life. than 70 percent of all existing plant and animal species
Living creatures in Earth’s oceans remained much the (see Chapter 7). Animals were the big winners in the after-
same—a mixture of single-celled and relatively primitive math, creating an evolutionary pathway that has led to us
multicellular organisms, for more than 3 billion years. Then, as human beings. Our earliest human ancestors appeared a
between 540 million and 500 million years ago, something few million years ago, and the fi rst civilizations occurred a
extraordinary happened. The number and diversity of bio- mere 10,000 years ago. Our industrial society, barely more
logical species increased than two centuries old, is but a footnote in the history of
spectacularly. Biologists call The number of new life on our planet.
this event the “Cambrian species exploded At the beginning of this chapter we mentioned that you,
explosion.” The cause of this about a half-billion the student, are here today because of a series of events that
sudden surge in biodiversity years ago. have occurred throughout the
remains unknown, but pos- history of the universe. Some You exist because of a
sibilities include rising oxygen levels, an increase in genetic of these events seem inevita- series of events—
complexity, major climate change, or a combination thereof. ble, such as the creation of some inevitable and
The “Snowball Earth” hypothesis suggests that prior to the heavy elements by earlier gen- others more random.
Cambrian explosion, Earth was in a period of extreme cold erations of stars and the for-
between about 750 million and 550 million years ago and mation of life-supporting planets, including Earth. Other
was covered almost entirely by ice. Astrophysicists specu- events may have been less likely, such as the creation of
late that the Sun’s output at that time was 5–10 percent self-replicating molecules that led to Earth’s earliest life. A
weaker than it is at present. During this period, predatory few events stand out as random, such as the life-destroying
animals may have died out, thus making it easier for new impact by a piece of space debris 65 million years ago—an
species to adapt and thrive. Another possibility is that a incident that eliminated the dominant species of that era.
marked increase in atmospheric oxygen (see Figure 8.5) This event—fortuitous for us, but not for the dinosaurs—
would have been accompanied by a corresponding increase made possible the evolution of advanced forms of mamma-
in stratospheric ozone, which, as we learned in Chapter 8, lian life and, ultimately, our existence as human beings. We
FIGURE 23.6 A fossil

recently found in the Canadian
Arctic shows Tiktaalik, a fish
with limblike fins and ribs—an
animal in a midevolutionary step
of leaving the water for dry
land.
might then ask the question, Would you be here today read- eties of molecules that were more successful than others in
ing this text if Earth hadn’t experienced that fateful encoun- reproducing themselves became more numerous. Varieties
ter? The answer is, probably not! With that sobering thought that could break down other varieties of self-replicating
in mind, let’s now return to the question of how prebiotic molecules and use them as
molecules became the molecules of life. raw material were especially Success breeds
successful in this world of success.
limited resources. Competi-
Evolution Is a Means of tion, predation, and even cooperation had entered the pic-
ture. After a few generations, certain species of molecules
Change and Advancement came to dominate the mix, while less successful varieties
Imagine that just once during the fi rst few hundred million became less and less common. This process, in which better-
years after the formation of Earth, a single molecule formed adapted molecules thrive and less well-adapted molecules
by chance somewhere in Earth’s oceans. That molecule had die out, is referred to as natural selection.
a very special property: Chemical reactions between that Four billion years is a long time—enough time for the
molecule and other molecules in the surrounding water combined effects of heredity and natural selection to shape
resulted in that molecule’s making a copy of itself. Now the descendants of that early self-copying molecule into a
there were two such molecules. Chemical reactions would huge variety of complex, competitive, successful structures.
produce copies of each of these molecules as well, making Geological processes on Earth, such as sedimentation, have
four. Four became 8, 8 became 16, 16 became 32, and so on. preserved a fossil record of the history of these structures
By the time the original molecule had copied itself just 100 (Figure 23.7). Among these descendants are “structures”
times, over a million trillion trillion (1030) of these molecules capable of thinking about their own existence and unrav-
would have existed. That is eling the mysteries of the stars.
100 million times more of To create life, only one The molecules of DNA (Figure 23.8) that make up the
these molecules than there self-replicating chromosomes in the nuclei of the cells throughout your
are stars in the observable molecule needed to body are direct descendants
universe! Such unconstrained form by chance. of those early self-duplicating Evolution is inevitable.
replication would be highly molecules that flourished in
unlikely for a number of reasons, such as the limited avail- the oceans of a young Earth. Although the game played by
ability of raw materials needed for reproduction and com- the molecules of DNA in your body is far more elaborate
petition from similar molecules. The point we would like than the game played by those early molecules in Earth’s
to make here is this: the likelihood that a copying error will oceans, the fundamental rules remain the same. We now
occur while a molecule is replicating increases significantly realize that this process is inevitable: any system that com-
with the number of copies being made. bines the elements of heredity, mutation, and natural selec-
Chemical reactions are never perfect. Sometimes when tion must and will evolve.
a copy is attempted, it is not an exact duplicate of its pre- In The Selfish Gene (1976), Richard Dawkins points out
decessor. The imperfection in the attempted copy is called that, when viewed from a purely utilitarian perspective, a
a mutation. Most of the time such an error is devastating, human being is a machine whose “purpose” is to produce
leading to a molecule that can no longer reproduce at all. copies of its genetic material. The remarkable thing about
But occasionally a mutation is actually helpful, leading to a humanity is that our intelligence (itself a very powerful tool
molecule that is more successful in duplicating itself than for survival in a competitive world) has also enabled us to
the original was. Even if imperfections in the copying pro- develop more noble pursuits (for example, astronomy) than
cess crop up only once every 100,000 times that a molecule those dictated by our basic biochemical imperative.
reproduces itself, and even if only one out of 100,000 of these
errors turns out to be beneficial, after only 100 generations
there will still be a hundred million trillion (1020) errors
that, by blind luck, might improve on the original design.
Copies of each of these improved molecules will inherit the 23.3 Life beyond Earth
change. These molecules will have happened upon a form of
heredity—the ability of one generation of structure to pass The story of the formation and evolution of life cannot
on its characteristics to future generations. be separated from the narrative of astronomy in the 21st
As molecules of the early Earth continued to interact century. We know that we live in a universe full of stars
with their surroundings and make copies of themselves, and that systems of planets orbit many—probably most—
they split into many different varieties. Eventually, these of those stars. Many biologists also believe that there is
descendants of our original molecule became so numerous nothing mysterious or unique about the origin of life or
that the building blocks they needed in order to reproduce the processes that cause it to evolve. In fact, as discussed
became scarce. In the face of this scarcity of resources, vari- in Connections 23.1, the evolution of life on Earth is but
638 Chapter 23 Life
FIGURE 23.7 Fossils

record the history of the
evolution of life on Earth.
one of the many examples that we have encountered of the

emergence of structure in an evolving universe. This point
leads naturally to one of the more profound questions that
we ask about the universe: Has life arisen elsewhere? To
explore this question, we need to take a closer look at the
chemistry of life on Earth.
The Chemistry of Life

When we speak about the chemistry of life, what we really
mean is life itself. All known organisms are composed of a
more or less common suite of chemicals, and very complex
ones at that. We begin by looking at ourselves. Approxi-
mately two-thirds of the atoms in our bodies are hydrogen
(H), about one-fourth are oxygen (O), a tenth are carbon (C),
and a few hundredths are nitrogen (N). The remaining
atomic elements, and there are several dozen of them, make
up only 0.2 percent of our total inventory. It turns out that
all living creatures—at least
those that we know of—are an All known life is
assemblage of molecules com- composed primarily of
posed almost entirely of these only six elements.
four atomic elements (some-
times called CHON), along with small amounts of phospho-
rus and sulfur. Some of these molecules are enormous.
Consider DNA, which is responsible for our genetic code.
DNA is made entirely from only five atomic elements: CHON
and phosphorus. But the DNA in each cell of our bodies is
composed of combinations of tens of billions of atoms of
these same five elements. Then there are proteins, the huge FIGURE 23.8 DNA is the blueprint for life.
C O N N E C T I O N S 2 3 .1
Life, the Universe, and Everything

Popular discussions of the origin of life frequently grapple with something increases its entropy. The decrease in entropy due
the question of how complex, highly ordered structures such to the formation of the drops is more than made up for by the
as living things could have emerged from simpler, less ordered increase in the entropy of their surroundings. Ordered structure
antecedents. Physicists discuss the degree of order of a system spontaneously emerged, but overall, disorder increased.
using the concept of entropy, which is a measure of the number It is important to note that energy can be used to reduce
of different ways a system could be rearranged and still appear entropy at one location while increasing entropy somewhere
the same. For example, place a drop of ink in a glass of water and else. An air conditioner provides a good example. Electric energy
watch what happens. The ink spreads out, diffusing through produced at a power plant is used to pump thermal energy from
the water until the only sign that the ink is there is the fact inside a home and dump that thermal energy outside. When an
that the water is a different color. The order represented by the air conditioner is run in reverse to provide heating, it is referred
discrete drop of ink naturally fades away to as a “heat pump,” but “entropy pump”
as the ink spreads out through the water. might be a more accurate description. As
Yet no matter how long we watch, we will you sit in your armchair on a summer day,
never see that drop of ink spontaneously all you immediately notice is that when
reassemble itself. the air conditioner comes on, the tempera-
Initially, the drop of ink and the glass ture drops. Entropy decreases inside your
of clear water form a neatly ordered system house. If you look at the system as a whole,
with low entropy. By contrast, the glass of on the other hand—including the heating of
inky water, which can be stirred or turned air by the outside coils and the entropy pro-
and still look the same, is a disordered sys- duced by the burning of coal or natural gas
tem with higher entropy. The second law of during the production of electricity—you
thermodynamics says that, left on its own, see that turning on the air-conditioning
an isolated system will always move toward causes an overall increase in entropy.
higher entropy—that is, from order toward This idea is especially important
disorder. This is common sense. As time when we consider the origin of life. A liv-
goes by, we are more likely to fi nd a system ing thing, whether an amoeba or a human
in a state that is, well, more likely. being, represents a huge local increase in
In light of the inescapable march toward order. However, no violation of the sec-
disorder that is dictated by the second law, ond law of thermodynamics is involved.
how can ordered structure emerge sponta- In our everyday lives, the food we eat
neously? Creationist or intelligent-design FIGURE 23.9 On a humid day, gives us the energy we need to stave off
claims that the origin of life fl ies in the water condenses into droplets on the the relentless advance of entropy. Viewed
surface of a cold glass. The second
face of the second law of thermodynamics even more broadly, the evolution of life on
law of thermodynamics does not
are about as common as reports of Elvis Earth was powered primarily by energy
mean that ordered structure cannot
sightings in supermarket tabloids. Yet such spontaneously emerge. striking Earth in the form of sunlight. A
claims make a crucial mistake: they focus local increase in order on Earth (such as
attention on one player while ignoring the rest of the game. you) is “paid for” in the end by the much greater decrease in
We have all seen what happens when we set a glass of ice order that accompanies thermonuclear fusion in the heart of
water out on a hot, humid day (Figure 23.9). Water vapor from the the Sun. Order emerges in localized regions within a system,
surrounding air condenses into drops of liquid water on the sur- but the second law of thermodynamics is obeyed overall.
face of the cold glass. This is amazing: We have just watched as The unifying theme of these last two chapters and, in many
ordered structure (drops of water) spontaneously emerged from ways, this entire book has been understanding the origin of
disorder (water vapor in the air). It is almost as if we saw the drop structure. The answer we have found is clear. Whether we
of ink reassemble itself. This simple, everyday event appears to are discussing the freezing out of matter and the fundamental
violate the second law of thermodynamics—but it does not. forces in the young universe; the gravitational collapse of stars,
To understand why, we need to step back and look at more planets, and galaxies; the evolution of life; or water beading up
than just the drops of water on the glass. When the drops con- on the outside of a cold glass—ordered structure does emerge
densed, they released a small amount of thermal energy that spontaneously as an unavoidable consequence of the action of
slightly heated both the glass and the surrounding air. Heating physical law.
640 Chapter 23 Life
molecules responsible for the structure and function of liv- that are also carbon-based but have chemistries quite dif-
ing organisms. Proteins are long chains of smaller molecules ferent from our own. For example, consider amino acids,
called amino acids. Terrestrial life employs 20 specific the building blocks of protein. There are countless variet-
amino acids, which also contain no more than five atomic ies of amino acids, but only 20 among them are employed
elements—in this case CHON plus sulfur instead of by terrestrial life. Furthermore, molecules other than RNA
phosphorus. and DNA may be capable of self-replication.
Can this be all there is to our body chemistry? No; the We have pointed out that carbon is the lightest of the
chemistry of life is far too complex to get by with a mere half- tetravalent atoms, but silicon is also tetravalent. Does this
dozen atomic species. Many others are present in smaller mean that somewhere in the
amounts but are essential to the complicated chemical universe there might exist a Life-forms other than
processes that make living organisms tick. They include form of life that is silicon- carbon-based are
sodium, chlorine, potassium, calcium, magnesium, iron, based? We see no reason why possible.
manganese, and iodine. Finally, there are the so-called trace not. As a potential life-
elements, such as copper, zinc, selenium, and cobalt. Trace enabling atom, silicon has both advantages and disadvan-
elements also play a crucial role in life chemistry but are tages when compared to carbon. An important advantage
needed in only tiny amounts. is that silicon-based molecules remain stable at much higher
We know that the infant universe was composed basi- temperatures than carbon-based molecules, enabling a pos-
cally of hydrogen and helium and very little else. But fast- sible silicon-based life to thrive in high-temperature envi-
forward to a later time—some 9 billion years later. By this ronments, perhaps on exoplanets that orbit close to their
time, all the heavier chemical elements essential to life parent star. But silicon has a serious disadvantage. Silicon
were present and available in the molecular cloud that gave is a larger and more massive atom than carbon, and it lacks
birth to our Solar System. As we learned in Chapters 16 the ability to form molecules as complex as those based on
and 17, those heavy elements—up to and including iron— carbon. Any silicon-based life would necessarily be simpler
were forged in the nuclear furnaces of earlier generations of than what we have here on Earth. Even so, we cannot rule
low- and high-mass stars and were then dispersed into out the possibility of silicon-based life existing in high-
space. At times, this dispersal was passive. For example, temperature niches somewhere within this almost limitless
low-mass stars, such as dying red giants, lose their gravi- universe.
tational grip on their overly extended atmospheres. Along Finally, although carbon’s unique properties make it
with hydrogen and helium, the newly created heavy ele- readily adaptable to the chemistry of life on Earth, we
ments are blown off into space, eventually finding their way don’t know what other chemistries life might adopt. As we
into molecular clouds. Other dispersals were more violent. learned from examples of terrestrial extremophiles here
Most of the trace elements essential to biology are more and in Chapter 11, life is highly adaptable and tenacious;
massive than iron, so they are not produced in the interiors when it comes to the form that extraterrestrial life might
of main-sequence stars. They are instead created within a take, nothing should surprise us.
matter of minutes during the violent supernova explosions
that mark the death of high-mass stars. They, too, are thrown
into the chemical mix found in molecular clouds.
Heavy elements produced by generations of stars provide
Life within Our Solar System
the raw materials for life—but what kind of life? The only In our quest to fi nd evidence of extraterrestrial life, the
form that we have discussed so far is terrestrial, because, logical place to start would be right here in our own Solar
after all, it is the only one we know of. As we have already System. As humans we have long been curious about the
pointed out, terrestrial life is possibility of life beyond Earth. Some early conjectures
organic, or carbon-based. Terrestrial life is seem ridiculous, considering what we now know about the
What, then, is so special about based on carbon. Solar System. Two centuries ago, the eminent astronomer Sir
carbon? Carbon is the lightest William Herschel, discoverer of Uranus, proclaimed, “We
among the “tetravalent” atoms. (Tetra means “four,” and need not hesitate to admit that the Sun is richly stored with
valence refers to an atom’s ability to attach to or “bond” with inhabitants.” In 1877, Italian astronomer Giovanni Schia-
other atoms or molecules.) This chemical property allows parelli (1835–1910) observed what appeared to be linear fea-
as many as four other atoms or molecules to bind themselves tures on Mars and dubbed them canali, meaning “channels”
to each carbon atom. Now, consider what can happen if the in Italian. In one of astronomy’s great ironies, the famous
attached molecules also contain carbon. The result can be American observer of Mars, Percival Lowell (1855–1916),
an enormous variety of long-chain molecules. This great misinterpreted Schiaparelli’s canali as “canals,” suggest-
versatility enables carbon to form the complex molecules ing that they were constructed by intelligent beings. Initial
that provide the basis for terrestrial life’s chemistry. public fascination with Martians turned to hysteria in 1938
Does this mean that all carbon-based life should be like when Orson Welles aired H. G. Wells’s fictitious War of the
us? Not at all. There could be forms of extraterrestrial life Worlds, as a “live” radio news coverage of militant Martians
invading Earth. Panic ensued when many listeners believed find direct evidence of life, so the long-standing question of
that the “invasion” was actually happening. whether there is life on Mars remains unanswered.
The next several decades saw only modest progress in The 1980s brought NASA’s instrumented robots to the
the search for life in the Solar System. During the mid–20th outer Solar System, and what they found surprised many
century, groundbased telescopes discovered that Mars pos- astrobiologists. Although the outer planets themselves did
sesses an atmosphere and water,1 both considered essential not appear to be habitats for life, some of their moons became
for any terrestrial-type life to get its start and evolve. Dur- objects of special interest. Jupiter’s moon Europa is covered
ing the 1960s, the United States and the Soviet Union sent with a layer of water ice that appears to overlie a great ocean
reconnaissance spacecraft to the Moon, Venus, and Mars, of liquid water. Impacts by
but the instrumentation carried aboard these spacecraft was comet nuclei may have added Life might exist on icy
more suited to learning about the physical and geological a mix of organic material, moons in the outer
properties of these bodies than to searching for life. Seri- another essential ingredient Solar System.
ous efforts to look for signs of life—past or present—would for life. Once thought to be a
have to await advanced spacecraft with specialized bio- frozen, inhospitable world, Europa is now a candidate for
instrumentation. biological exploration. Saturn’s moon Titan appears to be
In the meantime, astronomers and biologists alike were rich in organic chemicals, many of which are thought to be
discussing where to look and what to look for; and thus was precursor molecules of a type that existed on prebiotic
born the science of astrobiology, the study of the origin, evo- Earth. That 1980s message from Titan was clear: the chem-
lution, distribution, and future of life in the universe. The istry necessary for life does indeed exist elsewhere in the
question of where to look was a relatively easy one. Astro- Solar System.
biologists knew that Mercury and the Moon lacked atmo- Two decades later, the Cassini mission found additional
spheres, ruling them out as possibilities. The giant planets evidence for a variety of prebiotic molecules in Titan’s atmo-
and their moons were thought to be too remote and too cold sphere, and it identified another potential site for life: Sat-
to sustain life. Venus, they urn’s tiny moon Enceladus. In one of the bigger surprises
knew, was far too hot. As Gol- Mars is a planet from Cassini, the spacecraft detected water-ice crystals
dilocks might have concluded, that may harbor spouting from cryovolcanoes near the moon’s south pole.
Mars was just right. In the extraterrestrial life in Liquid water must lie beneath its icy surface, and Enceladus
mid-1970s, two American our own Solar System. therefore joins Europa as a possible habitat of extremophile
Viking spacecraft were sent to life, perhaps similar to that found near geothermal vents
Mars with detachable landers containing a suite of instru- deep within Earth’s oceans.
ments designed to fi nd evidence of a terrestrial type of life. Our efforts to find evidence of life on other bodies within
(Some scientists were critical of the specific sites chosen, our own Solar System have so far been unsuccessful, but
claiming that higher-latitude locations, where water ice the quest continues. The discovery of life on even one Solar
might exist, would have been preferable.) When the Viking System body beyond Earth would be exciting: if life arose
landers failed to find any evidence of life on Mars, hopes independently twice in the same planetary system, then
faded for finding life on any other body orbiting our Sun. that might increase the likelihood that life exists through-
Since that time, however, optimism has been renewed. A out the universe.
better understanding of the history of Mars indicates that at
one time the planet was wetter and warmer, leading many
scientists to believe that fossil life or even living microbes
might yet be buried under the planet’s surface. The first
Habitable Zones around Parent Stars
decade of the 21st century saw a return to Mars and a con- Searching for life among the countless exoplanets within
tinuation in the search for evidence of current or preexist- our own Milky Way Galaxy can be daunting. Nevertheless,
ing life (see Chapter 7). In 2008, NASA’s Phoenix spacecraft astronomers are narrowing the search by concentrating on
landed at a far-northern latitude, inside the planet’s arctic planets that provide environments conducive to the forma-
circle, where specialized instruments dug into and analyzed tion and evolution of life (at least, life as we understand it),
the martian water-ice permafrost (see Figure 7.37). Phoenix while eliminating planets that are clearly unsuitable for
found that the martian arctic soil has a chemistry simi- supporting life. As we learned in Chapter 6, most of the
lar to the Antarctic dry valleys on Earth, where life exists exoplanets discovered so far have environments that are
deep below the surface at the ice-soil boundary. Aqueous much too harsh to provide a haven for any life-forms that
minerals, such as calcium carbonate, reveal the existence we know of.
of ancient oceans on the planet. However, Phoenix did not To begin their quest, astronomers can optimize their
search by focusing on planetary systems that are stable.
Planets in stable systems remain in well-behaved, nearly
1
The martian polar caps are made up of frozen water ice and carbon circular orbits that preserve relatively uniform climato-
dioxide ice. logical and oceanic environments. Planets in noncircular
642 Chapter 23 Life
orbits can experience wild temperature swings that could Yet this limit may be too narrow, because it does not account
be detrimental to the survival of life. A stable temperature for possible extremophiles that may be thriving in liquid
that maintains the existence of water in a liquid state might water beneath the surfaces of some icy moons of Jupiter
be important. We know that liquid water was essential for and Saturn.
the formation and evolution of life on Earth. Of course, we Astronomers must also think about the type of star they
don’t know if liquid water is an absolute requirement for are observing in their search for life-supporting planets
life elsewhere, but it’s a good starting point when we think (Figure 23.10). Stars that are less massive than the Sun and
about where to look. On planets that are too close to their thus cooler will have narrower habitable zones, minimiz-
parent star, water would exist only as a vapor—if at all. On ing the chance that a habitable planet will just happen to
planets that are too far from their star, water would be perform within that slender zone. Stars that are more massive
manently frozen as ice. Still another consideration is planet than the Sun are hotter and will have a larger habitable
size. Large planets such as Jupiter retain most of their light zone. As we learned in Chapter 16, however, a star’s main-
gases—hydrogen and helium—during formation and so sequence lifetime depends on its mass. For example, a star
become gas giants without a surface. Planets that are very of 2 M⊙ would enjoy relative stability on the main sequence
small may have insufficient surface gravity to retain their for only about a billion years before the helium flash incin-
atmospheric gases and so end up like our Moon. erated everything around it! On Earth, a billion years was
We know that the location of a planet relative to its par- long enough for bacterial life to form and cover the planet,
ent star is critical. A candidate planet should orbit at a dis- but insufficient for anything more advanced to evolve. Of
tance from its star that provides a range of temperature course, we don’t know if evolution might happen at a differ-
suitable for life. Astronomers refer to this as a habitable ent pace elsewhere; we have only our one terrestrial case as
zone. Let’s think about the habitable zone in our own Solar an example! Still, main-sequence lifetime is a sufficiently
System. We can see that strong consideration that most astronomers prefer to focus
Venus, which orbits at 0.7 Exoplanets harboring their efforts on stars with longer lifetimes—specifically,
times Earth’s distance from life would most likely spectral types F, G, and K.
the Sun, has become an orbit within a star’s Finally, some astronomers think about a “galactic hab-
inferno because of its run- habitable zone. itable zone,” referring to a star’s location within the Milky
away greenhouse effect (see Way Galaxy. Stars that are situated too far from the galac-
Chapter 8). Any liquid water that might once have existed tic center may have protoplanetary disks with insufficient
on Venus has long since evaporated and been lost to space. quantities of heavy elements—such as oxygen and silicon
Mars orbits about 1.5 times farther from the Sun than Earth, (silicates), iron, and nickel—that make up terrestrial-type
and all of the water that we see on Mars today is frozen. planets like Earth. Stars that are too close to the galactic
But the orbit of Mars is more elliptical and variable than center also face problems. These regions experience less star
Earth’s, giving the planet a greater variety of climate, includ- formation, and therefore less recycling of heavy elements.
ing long-term cycles that might occasionally permit liquid Perhaps more serious is the high-energy radiation environ-
water to exist. Most astrobiologists put the habitable zone ment near the galactic center (X-ray and gamma-ray), which
of our Solar System at about 0.9–1.4 astronomical units is damaging to RNA and DNA. Even so, for many stars the
(AU), which includes Earth but just misses Venus and Mars. galactic habitable zone may not remain as a permanent
FIGURE 23.10 The distance and

extent of a habitable life zone (green)
surrounding a star depends on the
star’s temperature—in other words,
its mass. Regions too close to the
Hotter stars star are too hot (red), and those too
far are too cold (blue.)
Sunlike stars
Cooler stars
home if they tend to migrate within the galaxy and change

their distance from the galactic center over time. In short,
astronomers try to narrow down the vast numbers of stars
as they conduct their search, but they acknowledge that
these types of arguments—based on what worked well for
planet Earth—might not be applicable when we are looking
at other planetary systems.
The search for Earth-like planets from space is already
under way. In 2006, the European Space Agency (ESA)
launched CoRoT, the fi rst spacecraft dedicated to extra-
solar planet detection. CoRoT monitors nearby stars for
changes in brightness that would indicate the presence of
a transiting exoplanet (see Chapter 6). The ESA spacecraft
has already made several discoveries, including one planet
that is only 1.7 times the size of Earth. In 2009, NASA
launched Kepler into a solar orbit that trails Earth in its
own orbit about the Sun. Staying well behind Earth keeps
our planet from getting in Kepler’s field of view, thereby
enabling uninterrupted monitoring of Kepler’s target stars.
Over the next several years, the spacecraft’s photometer FIGURE 23.11 The plaque included with the Pioneer 11 probe,
will simultaneously and continually monitor the bright- which was launched in 1973 and will eventually leave the Solar
System to travel through the millennia in interstellar space.
ness of more than 100,000 stars, looking for transiting exo-
planets. Kepler can detect planets as small as 0.8–2.2 times
the size of Earth.
During the 1970s, humans made preliminary efforts to
“reach out and touch someone.” The Pioneer 11 spacecraft,
which will probably spend eternity drifting through inter-
23.4 The Search for Signs stellar space, carries the plaque shown in Figure 23.11. It
describes ourselves and our location to any future interstel-
of Intelligent Life lar traveler who might happen to find it. Another message
to the cosmos accompanied the two Voyager spacecraft on
We have come to a point where it seems that, given the right identical phonograph records that contain greetings from
conditions—an Earth-type planet orbiting within a habit- planet Earth in 60 languages, samples of music, animal
able zone around a solar-mass star—life of some kind might sounds, and a message from then-President Jimmy Carter.
invariably arise. It took but a few hundred million years for Interestingly, these messages created concern among some
life to form in Earth’s oceans. We know that this earliest life politicians and nonscientists who felt that scientists were
was primitive, and that it took another 3.5 billion years for dangerously advertising our location in the galaxy, even
this early life to climb the evolutionary ladder and arrive at though radio signals had been broadcast into space for
a species with the intelligence and technological capability nearly 80 years at the time. The messages were also criti-
to contemplate the universe that gave it birth and begin a cized by some philosophers, who claimed that we were
search through the vastness of space for others like itself. making ridiculous anthropomorphic assumptions about the
As humans, we have reached that stage at a time when our aliens being sufficiently like us to decode these messages.
star is only halfway through its lifetime. Such would not (Ironically, most Earthlings could not play those phono-
be the case for life arising on planets orbiting stars much graph records today!)
more massive than the Sun. As we learned in Chapter 16, a Sending messages on spacecraft may not be the most
3-M⊙ star has a main-sequence lifetime of only a few hun- efficient way to make contact with the universe, but it was
dred million years. Our terrestrial experience suggests that a significant gesture nonetheless. A somewhat more prac-
just as life was getting its start on one of the star’s habitable tical effort was made in 1974, when astronomers used the
planets, it would quickly be doomed as the star reached the 300-meter-wide dish of the Arecibo radio telescope to beam
end of its brief period of stability. A star 11⁄2 times as massive a message in binary code2 (Figure 23.12) toward the star
as the Sun has a main-sequence lifetime of a couple of bil- cluster M13. (If someone from M13 answers, we will know
lion years, insufficient for any life to reach the equivalent of in 48,000 years!)
Earth’s “Cambrian explosion.” If we are to search for signs
of intelligent life, we should look to planets surrounding 2
Binary code consists only of zeros and ones, with zero equivalent
stars of solar mass or smaller. And now that we have some to “off” and one equivalent to “on.” It may be a universal means of
ideas about where to search for intelligent life, how do we communication that does not depend on any specific language or
establish communication? numerical system.
644 Chapter 23 Life
FIGURE 23.12 The message it is possible that moons orbiting giant planets near stars
we beamed toward the star could also harbor life. For purposes of calculation, we
cluster M13 in 1974. This binary- put the value of npm between 0.1 and 2.
encoded message contains the
numbers 1–10, hydrogen and 4. f l is the fraction of planets and moons capable of sup-
carbon atoms, some interesting porting life on which life actually arises. Remember that
molecules, DNA, a human with just a single self-replicating molecule may be enough to
description, basics of our Solar get the ball rolling. Many biochemists now believe that
System, and basics of the if the right chemical and environmental conditions are
Arecibo telescope. A reply may be
present, then life will develop. If they are correct, then
forthcoming in 48,000 years.
f l is close to 1. For f l we will use a range of 0.01 to 1.
5. fc is the fraction of those planets harboring life that
eventually develop technologically advanced civiliza-
tions. With only one example of a technological civili-
The Drake Equation zation to work with, fc is hard to estimate. Intelligence
is certainly the kind of survival trait that might often
The fi rst serious effort to search be strongly favored by natural selection. On the other
for intelligent extraterres- hand, on Earth it took 4 billion years—half the expected
trial life was made by astron- lifetime of our star—to evolve tool-building intelligence.
omer Frank Drake (1930–) in The correct value for fc might be close to 1, or it might be
1960. Drake used what was closer to one in 1,000. The truth is, we just don’t know.
then astronomy’s most power-
ful radio telescope to listen for 6. L is the likelihood that any such civilization exists
signals from two nearby stars. today. This factor is certainly the most difficult of all to
Although his search revealed estimate because it depends on the long-term stability
nothing unusual, it prompted of advanced civilizations. We have had a technological
him to develop an equation that civilization on Earth for about 100 years, and during
still bears his name. The Drake that time we have developed, deployed, and used weap-
equation estimates the likely number (N) of intelligent civi- ons with the potential to eradicate our civilization and
lizations that may exist within the Milky Way Galaxy: render Earth hostile to life for many years to come. We
have also so degraded our planet’s ecosystem that many
N = Ffp npm fl fc L respectable biologists and climatologists wonder if we
The Drake equation are nearing the brink. Do all technological civilizations
estimates the number destroy themselves within a thousand years? A thousand
The six factors on the right
of technologically years is only one 10-millionth (10 –7) of the lifetime of
side of the equation relate to
advanced civilizations our star. Conversely, if most technological civilizations
the conditions that Drake
in the galaxy. learn to use their technology for survival rather than
thought must be met for a civ-
ilization to exist: self-destruction, might they instead survive for a 100
million years, or about a hundredth (10 –2) of the Sun’s
1. F is the total number of stars in our galaxy. We know lifetime? For L in our calculation, we will use a range
this to be several hundred billion stars. of between 10 –7 (for civilizations that live 1,000 years)
and 10 –2 (for civilizations that live 100 million years).
2. fp is the fraction of stars that form planetary systems.
Questions remain, but we do know that planets form as
a natural by-product of star formation and that many— As illustrated in Figure 23.13, the conclusions we draw
perhaps most—other stars have planets. For this calcu- using the Drake equation depend a great deal on the assump-
lation we will assume that fp is between 0.5 and 1. tions we make. For the most pessimistic of our estimates,
the Drake equation sets the likelihood of fi nding a techno-
3. npm is the average number of planets and moons in each logical civilization in our galaxy at 1 percent, or one chance
planetary system capable of supporting life. If such plan- in 100. If this is correct, then in all likelihood we are the
ets are like Earth, they will have to be of the terrestrial only technological civilization in the Milky Way. In fact,
type and orbit within a “habitable zone,” where tempera- this result would suggest that only one in 100 galaxies would
tures are neither too hot nor too cold. If our own Solar contain a technological civi-
System is a good example, there would be one or two lization at all. Such a universe The nearest advanced
such planets in each planetary system. (In addition to would still be full of intelli- civilization may
Earth, Mars appears capable of supporting life, whether gent life. With a hundred bil- be as far away as
or not life presently exists there.) Although we now know lion galaxies in the observable 30 Mly . . .
that many planetary systems do not resemble our own, universe, even these pessi-
The Drake equation estimates White dots show stars mistic assumptions would mean that there are a billion
the likely number of technological with possible civilizations. technological civilizations out there somewhere. On the
civilizations in our galaxy. other hand, we would have to go a very long way (30 mil-
lion light-years [Mly] or so) to fi nd our nearest neighbors.
At the other extreme, what if we take the most optimis-
tic numbers, assuming that intelligent life arises and sur-
vives everywhere it gets the
chance? The Drake equation . . . or as near as 40
then says that there should be light-years.
MORE OPTIMISTIC MORE PESSIMISTIC
40 million technological civi-
ASSUMPTIONS ASSUMPTIONS
lizations in our galaxy alone! In this case, our nearest neigh-
Eliminate stars bors may be “only” 40 or 50 light-years away. If scientists
without planets. in that civilization are listening to the universe with their
own radio telescopes, hoping to answer the question of other
life in the universe for themselves, then as you read this
page they may be puzzling over an episode of the original
Star Trek television series.
If we did run across another technologically advanced
civilization, what would it be like? Looking back at the
Eliminate planets Drake equation, we can see it
hostile to life. is highly unlikely that we Any civilization we
have neighbors nearby, unless discover will almost
civilizations typically live for certainly be advanced.
many thousands or even mil-
lions of years. In this case, any civilization that we encoun-
tered would almost certainly have been around for much
longer than we have. Having survived that long, would its
Eliminate hospitable members have learned the value of peace, or would they
planets without life. have developed strategies for controlling their pesky neigh-
bors? Movie theaters and the science fiction shelves of librar-
ies and bookstores are filled with amusing and thoughtful
stories that explore what life in the universe might be like
(Figure 23.14). For the moment, though, let’s set aside such
speculation to look at the question as scientists. How might
we go about fi nding a real answer?
Eliminate life
without civilization.
Technologically Advanced Civilizations

During lunch with colleagues, the famous physicist Enrico
Fermi (1901–1954), a fi rm believer in extraterrestrial beings,
is reported to have asked, “If the universe is teeming with
aliens . . . where is everybody?” To which Hungarian-born
Eliminate civilizations physicist Leo Szilard (1898–1964) is said to have replied,
not alive today. “They are here among us, but they call themselves Hun-
garians.” We can put Szilard’s tongue-in-cheek reply aside,
but Fermi’s question—fi rst posed in 1950 and sometimes
called the “Fermi paradox”—remains unanswered. Consider
also the following closely related question: If intelligent
life-forms are common but interstellar travel is difficult or
Do civilizations abound… …or are they very rare?
impossible, why don’t we detect their radio transmissions?
The truth is that we have failed so far to detect any alien
FIGURE 23.13 Estimates of the existence of intelligent radio signals—but it is not for lack of trying.
civilizations in our galaxy based on the Drake equation. Notice Drake’s original project of listening for radio signals
how widely these estimates vary, given optimistic and pessimistic from intelligent life around two nearby stars has grown
assumptions about the six factors (see the text) affecting the over the years into a much more elaborate program that is
prevalence of intelligent extraterrestrial life. referred to as the Search for Extraterrestrial Intelligence,
646 Chapter 23 Life
resources. One project, known as SETI@home, involves

the use of hundreds of thousands of personal computers
around the world to analyze the institute’s data. SETI screen
saver programs installed on personal computers worldwide
download radio observations from the SETI Institute over
the Internet, analyze these data while the computers’ own-
ers are off living their lives, and then report the results of
their searches back to the institute. Perhaps the fi rst sign
of intelligent life in the universe might be found by a com-
puter sitting on your desk.
SETI’s Allen Telescope Array (ATA), named for Micro-
soft cofounder Paul Allen, who provided much of the
initial fi nancing for the project, uses a similar low-cost
approach. A joint venture between the SETI Institute and
the University of California, the ATA consists of a “farm”
of small, inexpensive radio dishes like those used to cap-
ture signals from orbiting communication satellites (Figure
23.15). The ATA searches the sky 24 hours a day, 7 days
a week, for signs of intelligent life. Eventually there will
be 350 dishes, each with a diameter of 6.1 meters and col-
lectively yielding a total signal-receiving area greater than
that of a 100-meter radio telescope. Just as your brain can
sort out sounds coming from different directions, this array
of radio telescopes will be able to determine the direc-
tion a signal is coming from, allowing it to listen to many
stars at the same time. Over several years’ time, astrono-
mers using the ATA are expected to survey as many as a
FIGURE 23.14 The classic 1951 film The Day the Earth Stood
million stars, hoping to fi nd a civilization that has sent a
Still portrayed intelligent extraterrestrials who threatened to signal in our direction. If reality is anything like the more
destroy Earth if we carried our violent ways into space. optimistic of the assumptions we used in evaluating the
Drake equation (see Figure 23.13), this project will stand
a good chance of success.
Finding even one nearby civilization in our galaxy—
or SETI. Scientists from around the world have thought that is, a second technological civilization in our own small
carefully about what strategies might be useful for fi nding corner of the universe—could mean that the universe as a
life in the universe. Most of whole is literally teeming with intelligent life. SETI may
these have focused on the SETI listens for radio not be in the mainstream of astronomy, and the likelihood
idea of using radio telescopes signals from other of its success is difficult to predict, but its potential payoff
to listen for signals from civilizations. is enormous. Few discoveries would do more to change our
space that bear an unambigu- understanding of ourselves than certain knowledge that we
ous signature of an intelligent source. Some have listened are not alone.
intently at certain key frequencies, such as the frequency Science fiction is filled with tales of humans who leave
of the interstellar 21-centimeter (cm) line from hydrogen Earth to “seek out new life and new civilizations.” Unfor-
gas. The assumption behind this approach is that if a civi- tunately, these scenarios are not realistic. The distances
lization wanted to be heard, its denizens would tune their to the stars and their planets are enormous; to explore a
broadcasts to a channel that astronomers throughout the significant sample of stars would require extending the
galaxy should be listening to. More recent searches have physical search over tens or hundreds of light-years. As we
made use of advances in technology to record as broad a learned in Chapter 17, special relativity limits how fast we
range of radio signals from space as possible. Analysts then can travel. The speed of light is the limit, and even at that
use computers to search these databases for types of regu- rate it would take over 4 years to get to the nearest star. Time
larity in the signals that might suggest they are intelligent dilation would favor the astronauts themselves, and they
in origin. would return to Earth younger than if they had stayed at
Unlike much astronomical research, SETI receives its home. But suppose they visited a star 15 light-years distant.
funding from private rather than government sources, and Even if they traveled at speeds close to that of light, by the
SETI researchers have found ingenious ways to continue time they returned to Earth, 30 years would have passed!
the search for extraterrestrial civilizations with limited Some science fiction enthusiasts get around this problem
23.5 The Future 647
FIGURE 23.15 When

complete, the Allen Telescope
Array will listen for evidence of
intelligent life from as many as
a million stellar systems.
by invoking “warp speed” or “hyperdrive” to travel faster become a red giant and later an asymptotic giant branch (AGB)
than the speed of light, or by using wormholes as shortcuts star, swelling to hundreds of times its present size. The giant
across the galaxy. In truth, there is absolutely no evidence planets, orbiting outside the extended red giant atmosphere,
that any of these shortcuts are possible. may survive the Sun’s cranky old age in some form. Even so,
Some people claim that the aliens have already visited they will suffer the blistering radiation from a Sun grown
us: tabloid newspapers, books, and websites are filled with thousands of times more luminous than it is today.
tales of UFO sightings, government conspiracies and cover- The terrestrial planets will not fare as well. Some—per-
ups, alleged alien abductions, and UFO religious cults. haps all—of the worlds of the inner Solar System will be
However, none of these reports meet the basic standards engulfed by the expanding Sun. Just as an artificial satellite
of science, and we must conclude that there is no scientific is slowed by drag in Earth’s tenuous outer atmosphere and
evidence for any alien visitations. eventually falls to the ground in a dazzling streak of white-
hot light, so, too, will a terrestrial planet caught in the Sun’s
atmosphere be consumed by the burgeoning star. If this is
Earth’s fate, our home world will leave no trace other than
a slight increase in the amount of massive elements in the
23.5 The Future Sun’s atmosphere. As the Sun loses more and more of its
atmosphere in an AGB wind, our atoms may be expelled
We have used our understanding of physics and cosmology back into the reaches of interstellar space from which they
to look back through time and watch as structure formed came, perhaps to become incorporated into new generations
throughout the universe. We have peered into the future and of stars, planets, and even life itself.
seen the ultimate fate of the universe, from its enormous Another planetary fate is possible, however. In this sce-
clusters of galaxies down to its tiniest components. Now, nario, as the red giant Sun loses mass in a powerful wind,
as our journey nears its end, we examine our own destiny its gravitational grasp on the planets will weaken, and the
and contemplate the fate that awaits Earth, humanity, and orbits of both the inner and
the star on which all life depends. outer planets will spiral out- Far-future Earth
ward. If Earth moves out far will be consumed by
enough, it may survive as a the Sun or left as an
seared cinder, orbiting the icy cinder.
Eventually, the Sun Must Die white dwarf that the Sun will
As we have learned, about 5 billion years from now the Sun become. Barely larger than Earth and with its nuclear fuel
will end its long period of relative stability. Shedding its exhausted, the white dwarf Sun will slowly cool, eventu-
identity as the passive, benevolent star that has nurtured life ally becoming a cold, inert sphere of degenerate carbon,
on Earth for nearly 4 billion years, the Sun will expand to orbited by what remains of its retinue of planets. Thus, the
648 Chapter 23 Life
ultimate outcome for our Earth—consumed in the heart of Perhaps an explosion like the 1908 Tunguska blast in Rus-
the Sun or left behind as a frigid, burned rock orbiting a sia (see Chapter 12), occurring over New York or Paris, would
long-dead white dwarf—is not yet certain. In either case, be enough to convince us that such precautions are
Earth’s status as a garden spot will be at an end. worthwhile.
We might protect ourselves from the fate of the dino-
saurs, but in the long run the descendants of humanity will
either leave this world or die out. Planetary systems sur-
The Fate of Life on Earth round other stars, and all that we know tells us that many
Life on our planet has even less of a future, for it will not other Earth-like planets should exist throughout our galaxy.
survive even long enough to witness the Sun’s departure Colonizing other planets is currently the stuff of science
from the main sequence. Well before that cataclysmic event fiction, but if our descendants are ultimately to survive
takes place, the luminosity of our heretofore benevolent star the death of our home planet, off-Earth colonization must
will begin to rise. As solar luminosity increases, so will tem- become science fact at some point in the future.
peratures on all the planets, including our own. Eventually Although humankind may soon be capable of protecting
Earth’s temperatures will climb so high that all animal and Earth from life-threatening comet and asteroid impacts, in
plant life will perish. Even the extremophiles that inhabit other ways we are our own
the oceanic depths will die, as the oceans themselves boil worst enemy. We are poison- Humanity is the worst
away. Models of the Sun’s evolution are still too imprecise ing the atmosphere, the water, threat to its own
to predict with certainty when that fatal event will occur, and the land that form the survival.
but the end of all terrestrial life may be only 1 billion or 2 habitat for all terrestrial life.
billion years away. That is, of course, a comfortably distant As our population grows unchecked, we are occupying more
time from now, but it is well short of the Sun’s departure and more of Earth’s land and consuming more and more of
from the main sequence. its resources, while sending thousands of species of plants
It is far from certain, however, that the descendants of and animals to their extinction each year. At the same time,
today’s humanity will even be around a billion years from human activities are dramatically affecting the balances of
now. Some of the threats that await us come from beyond atmospheric gases. The climate and ecosystem of Earth con-
Earth. For the remainder of the Sun’s life, the terrestrial stitute a fi nely balanced, complex system capable of exhib-
planets, including Earth, will continue to be bombarded by iting chaotic behavior. The fossil record shows that Earth
asteroids and comets. Perhaps a hundred or more of these has undergone sudden and dramatic climatic changes in
impacts will involve kilometer-sized objects, capable of response to even minor perturbations. Such drastic changes
causing the kind of devastation that eradicated the dino- in the overall balance of nature would certainly have con-
saurs and most other species 65 million years ago. Although sequences for our own survival. When politics is added to
these events may create new surface scars, they will have the mix, even more immediate dangers await. For the fi rst
little effect on the integrity of Earth itself. Earth’s geological time in human history, we possess the means to unleash
record is filled with such events, and each time they hap- nuclear or biological disasters that could threaten the very
pen, life manages to recover and reorganize. survival of our species. In the end, the fate of humanity will
It seems likely, then, that some form of life will survive depend more than anything on whether we accept steward-
to see the Sun begin its march toward instability. However, ship of ourselves and of our planet.
individual species do not necessarily fare so well when
faced with cosmic cataclysm. If the descendants of human-
kind survive, it will be because we chose to become players
in the game by changing the odds of such planetwide bio-
Life in a Future Universe
logical upheavals. We are rap- The universe of the far distant future seems likely to be an
idly developing technology To survive, humanity extremely dull and lifeless place—a universe filled only
that could enable us to detect must learn to manage with decaying black holes,
most threatening asteroids the threat of impacts. magnetic fields, plasmas, and We cannot predict
and modify their orbits well ever-cooling radiation. But what structure might
before they could strike Earth. Comets are more difficult to will that necessarily be so? arise in the far future.
guard against because long-period comets appear from the Imagine for a moment that
outer Solar System with little warning. To offer protection, intelligent life somehow evolved amid the swarm of exotic
various means of defending ourselves would have to be in particles, such as free quarks and gluons, that filled the uni-
place, ready to be used on very short notice. So far, we have verse shortly after the Big Bang. Such organisms would have
been slow to take such threats seriously. Although impacts been far smaller than today’s atoms and perhaps would have
from kilometer-sized objects are infrequent, objects only a lived out their lives in 10 –40 second or so. To such organ-
few dozen meters in size, carrying the punch of a several- isms the universe—all 3 meters of it—would have seemed
megaton bomb, strike Earth about once every 100 years. incomprehensibly vast. These creatures and their entire
civilization would have had to evolve, live, and die out in stars, galaxies, planets, and intelligent creatures for whom
a millionth of a trillionth of a trillionth of the time that it a single thought took longer than a billion trillion times the
takes for a single synapse in our brains to fi re. If such crea- entire history of the universe they knew.
tures ever calculated the conditions in today’s universe, Now turn the tables, and think forward to a time when
they would have recoiled in horror. They would have imag- the universe is 1050 times older than it is today. Who can
ined a frozen time when the temperature of the universe say that there will not be life then as well? Perhaps the life-
was only a thousandth of a trillionth of a trillionth of what forms of this distant future will be organisms of magnetic
they knew—a time when most matter had ceased to exist fields and tenuous electron plasmas, spread across count-
altogether and the tiny fraction that remained was spread less trillions of light-years of space, whose lives unfold over
out over a volume of space 1075 times greater than that of untold eons of time. For such organisms, if they ever exist,
their universe. In short, such creatures would have looked it will be we who are the impossible creatures, alive for the
forward and seen our universe as the frozen, desolate future. briefest of instants, still immersed in the momentary fi re-
It seems doubtful they could have foreseen the existence of ball of the Big Bang.
planets, and every other component of the universe that

Seeing the Forest for the Trees we observe. Everything we see says that the universe
formed 14 billion years ago, not 6,000. Cosmology is the
As it is for galaxies and stars and planets, so it is for our- ultimate evolutionary science—the science of the ori-
selves. A recurring theme on our journey of discovery gin and evolution of the universe itself. During the 20th
has been the systematic dismantling of conceptual walls century, geology became the science of the evolution of
that in our minds separated us from the larger universe. the surface of Earth, while planetary science applied
Humanity does not stand outside the processes that shape our understanding of terrestrial geology to understand-
the universe. We are instead one more variety of the struc- ing the evolution of other planets and their moons. The
ture to which the universe has given birth—a way station most fundamental questions in physics concern the ori-
on a long road of evolving structure stretching back 14 gin and evolution of physical laws. So, too, is the case
billion years. Few single words are capable of eliciting as in modern biology, which simply makes no sense until
much emotional reaction from some people as evolution. it is organized around the theme of evolution by natural
Yet the public controversy surrounding evolution can- selection.
not change the scientific standing of this theory as one Evolution is anything but a scientific appendix that
of the best-tested and most successful theories in all of can be harmlessly removed. It is an observed fact. It is
science. As stated by Stephen Jay Gould, “The theory of the backbone and central nervous system of modern
evolution is in about as much trouble as the theory that science. In Darwin’s Dangerous Idea (1995), Daniel Den-
Earth revolves around the Sun.” By any reasonable stan- nett speaks of the concept of evolution as “universal
dards of scientific knowledge, evolution is a fact. acid: it eats through just about every traditional concept,
Opponents of evolutionary theory often act as if evolu- and leaves in its wake a revolutionized worldview.” As
tion were a tiny piece of science that could simply be cast the 21st century enters its second decade, evolution of
aside without doing violence to the rest. Having traveled structure is the unifying theme that ties the breadth of
the journey of 21st Century Astronomy, you should see modern science together into a beautiful, powerful, com-
clearly how absurd such a claim is. Modern astronomy prehensive whole. If we tried to pull the thread of evolu-
would simply cease to be, were it stripped of our under- tion out of this tapestry, the whole cloth would unravel
standing of the origin and evolution of galaxies, stars, before our eyes.
650 Chapter 23 Life
Summary 6. Today, most known life enjoys moderate climates and

temperatures. Compare this environment to some of the
conditions in which early life developed.
• Life is basically a form of complex carbon-based chem-
istry, made possible by special molecules capable of 7. Discuss some extremophiles living on Earth today.
reproducing themselves.
8. How was Earth’s carbon dioxide atmosphere changed
• Life likely formed in Earth’s oceans, evolving chemically into today’s oxygen-rich atmosphere. How long did that
from an organic “soup” of prebiotic molecules into self- transformation take?
replicating organisms.
9. What are the similarities and differences between
• All terrestrial life is composed mostly of only six ele- prokaryotes and eukaryotes?
ments: carbon, hydrogen, oxygen, nitrogen, sulfur, and
10. Tracing back on the evolutionary tree, what kinds of
phosphorus.
life do we fi nd humans are similar to, and when did
• Life-forms that are very different from ours, including the evolutionary steps leading from those kinds of life
those based on silicon chemistry, cannot be ruled out. to humans happen?
• Space-based instruments have begun the search for 11. What was the Cambrian explosion and what might have
Earth-like planets. caused it?
• Astronomers prefer to focus their search for extrater- 12. Why do you suppose plants and forests appeared in large
restrial life on Earth-like planets orbiting in habitable numbers before large animals did?
zones surrounding solar-type stars.
13. Why was evolution inevitable on Earth?
• Although the Milky Way Galaxy may be teeming with
14. What were the general conditions needed on our planet
extraterrestrial life, none has yet been detected.
for life to arise?
• Long before the Sun ends its period of stability on the
15. Is biological evolution underway on Earth today? If so,
main sequence, all terrestrial life will have perished.
how might humans continue to evolve?
16. Where did all the atoms in your body come from?
17. The Viking spacecraft did not fi nd evidence of life on
SmartWork, Norton’s online homework system, Mars when it visited that planet in the late 1970s, nor
includes algorithmically generated versions of these did the Phoenix lander when it examined the martian
questions, plus additional conceptual exercises. If soil in 2009. Does this imply that life never existed on
your instructor assigns questions in SmartWork, log in at the planet? Why or why not?
18. A few scientists believe we may be the only advanced
life in the galaxy today. If this were indeed the case,
which factors in the Drake equation would have to be
extremely small?
Student Questions
19. The second law of thermodynamics says that the entropy
THINK ING A BOUT THE CONCEP T S (a measure of disorder) of the universe is always increas-
ing. Yet living organisms exist as a result of order being
1. Why do we generally talk about molecules such as DNA
created from disorder. Why does this not violate the
or RNA when discussing life on Earth?
second law of thermodynamics?
**2. Provide a defi nition for life that does not rely on RNA
20. What is a habitable zone? What defines its boundaries?
or DNA. Is this defi nition sufficient to account for new
forms that may be discovered in the future, either on 21. In searching for intelligent life elsewhere, why is listen-
Earth or elsewhere? ing with radio telescopes currently our best method?
3. How do we suspect that the building blocks of DNA fi rst 22. Why is it likely that life on Earth as we know it will end
formed on Earth? long before the Sun runs out of nuclear fuel?
4. If processes on Earth were energetic enough to form
DNA, shouldn’t they have been able to destroy it as well?
Why, then, does life exist today? A P P LY I N G T H E C O N C E P T S
5. When do we believe life fi rst appeared on Earth? What 23. Excursions 23.1 (“Forever in a Day”) takes events spread
evidence supports this belief? out over enormous intervals of time and compresses
them into the more comprehensible interval of a single (from question 27) after 5,000 generations if Beta has a
24-hour day. Make your own “Life in a Day” by com- mere 0.1 percent survivability advantage over Alpha.
pressing all the important events of your lifetime into
29. Suppose the early Earth had oceans of ammonia (NH3)
a single day, starting with your birth at the stroke of
rather than water (H2O). Could life have evolved in these
midnight and continuing to the present at the end of the
conditions? Why or why not? What if Earth had been
day.
totally dry—that is, with no liquid of any kind? Explain
24. As noted in Section 23.2, some scientists suspect the how this may have affected the formation of terrestrial
early Earth was “seeded” with primitive life stored in life.
comets and meteoroids. Knowing when and how our
30. Knowing the history and composition of Mars, specu-
Solar System, galaxy, and universe formed, what time
late on whether life evolved there before the planet
line is required for such seeding to be possible?
became geologically dead. If so, how complex did the
25. Most life on Earth today relies directly or indirectly on life become?
the Sun. For example, the food chain of life near the
31. Speculate on whether life could have evolved on Jupi-
ocean’s surface relies on algae, which use photosynthe-
ter’s moon Europa. If so, how complex do you predict it
sis. Organisms living on ocean floors generally feed from
to be?
surface material that sank, so these bottom dwellers rely
indirectly on the Sun. Are there any forms of life today 32. Speculate on whether life could have evolved on Sat-
that do not rely in any way on solar energy for life? If urn’s moon Titan. If so, how complex do you predict it
so, what does this imply about the prospects of life in to be?
remote places in our Solar System?
33. Describe the properties of a planet, and its star, that would
26. If the chance that a given molecule will make a copy- make it more probable for life to have evolved there.
ing error is one in 100,000, how many generations are
34. Using your textbook or the Internet, suggest a few nearby
needed before, on average, at least one mutation has
stars that could be potential candidates for life-bearing
occurred?
planets. Explain why you selected each one.
27. Consider an organism Beta that, because of a genetic
35. Why do you think we sent a coded radio signal to the
mutation, has a 5 percent greater probability of survival
globular cluster M13 in 1974—rather than, say, to a
than its nonmutated form, Alpha. Alpha has only a 95
nearby star?
percent probability (pr) of reproducing itself compared
to Beta. After n generations, Alpha’s population within
the species would be Sp = (pr)n compared to Beta’s. Cal-
culate Alpha’s relative population after 100 generations.
(Note: You may need a scientific calculator or help from
your instructor to evaluate the quantity 0.95100.)
28. To fully appreciate the power of heredity, mutation, and quizzes, plus links to premium content in SmartWork and
natural selection, consider Alpha’s relative population the ebook. Visit www.wwnorton.com/studyspace.
Sometimes the light’s all shining on me
Other times I can barely see
Lately it occurs to me
What a long strange trip it’s been.
ROBERT HUN T ER (1941–)
Epilogue: The Universe and You
The Long and Winding Road we saw a realm of gods and angels, of mysticism and magic,
of the perfect and unchanging fifth element.
Go out at night and look at the stars, and feel the same sense So it remained for thousands of years until, at the dawn
of wonder and awe that our kind has always felt at the sight. of the Renaissance, a Polish monk dared to challenge the
Take it in, be amazed at the majestic canopy overhead, and wisdom of the ages and to think the unthinkable. Reviving
let your imagination roam, just as our ancestors have done a notion that had been discarded long before by the Greeks,
for thousands of years. As you do, drift back down the road Nicolaus Copernicus allowed himself to imagine that per-
we have followed in this book. Reflect on all we have come haps it was the motion of Earth, rather than the motions of
to know about the universe, and on how much more mag- the Sun and stars, that shaped the passage of the days and
nificent the heavens are than our ancestors ever could have the years. In so doing, he not only conceptually dislodged
imagined. Our journey has been more than a description of Earth from its moorings, but he also broke the shackles that
the universe and what it contains. It has been a travelogue had for so long constrained the human mind. What began
of the struggle and triumph of the human mind and spirit. as a crack in the foundation of our preconceptions would,
Ever since humans fi rst recognized that the patterns shap- in the end, turn that old view of the world to rubble. In its
ing our lives are echoed in the sky, we have searched for place we would construct an edifice of knowledge that has
the threads connecting us to the cosmos and have sought to given us dominion over our world and carried our thoughts
understand our place in it. We have the privilege of being to the farthest reaches of the universe.
among the fi rst generations of humans to fi nd those threads Copernicus was one of a succession of people with great
and to learn real answers to those age-old questions. minds who could not rest without fi rst picking at the loose
We have no way to count how many different stories threads of the ideas they had been raised to believe in. Water
have been told about the heavens, but we do know that for runs into the cracks in a slab of granite and freezes, expand-
thousands of years most of those stories shared a common ing and pushing the cracks open, exposing the flaws in the
foundation. One cornerstone was the belief that Earth occu- rock. As the seasons come and go, the imposing boulder
pies a special place in the scheme of things. In our minds stands no chance in the face of this persistent onslaught.
we were at the center of the universe, fi xed and immov- In like fashion, the persistent questioning and probing by
able, and all that we saw was present only to give meaning great minds would eventually shatter the reign of enforced
to our existence. A second cornerstone of this traditional ignorance and entrenched authority. We have met a few of
worldview was the belief that the heavens are fundamen- these great minds on our journey; there were many oth-
tally different from Earth. To our ancestors, our world was ers. They were of different nationalities, different upbring-
the realm of the ordinary and mundane—a terrestrial exis- ings, and different dispositions and beliefs. But their work
tence built from Aristotle’s earth, wind, fi re, and water. In shared a common theme: the answers to questions about
contrast, the heavens had their own separate reality. There the world come not from the pronouncements of authority
E-1
E-2 Epilogue: The Universe and You
or the prejudices taught in childhood, but from observing knowledge, even today, are (1) “It is true because I believe it,”
nature itself and thinking carefully, honestly, and openly (2) “It is true because we believe it,” and (3) “It is true because
about what we see. I want to believe it.” Of course, none of these has anything
In Isaac Newton’s famous thought experiment, a ball to do with what really is true. To learn about the universe
fi red from an imaginary cannon moves rapidly enough and our world, we have had to set aside these notions, which
that it falls around the world in a circle. Such a “cannon” blur the line between reality and fantasy, and replace them
could not have been built with the technology available in with a tough, unforgiving, and very different standard: “It
Newton’s day. That achievement would have to await the is provisionally true because we have worked very hard to
launch of Sputnik hundreds of years later. Yet although show that it is false but so far have failed.” This standard
Newton could not make his cannon a reality, he did not alone places reality itself in front of our parochial ideas and
have to. He had only to look at the sky and watch as the beliefs. It puts what is true ahead of what we would like to
Moon traced out its monthly path. Newton realized that be true. Only by testing the falsifiable predictions of our
the force holding the Moon in its orbit about Earth is the theories about the world have we learned to push aside the
same force that gives us weight and guides the path of a comfortable notions that for so long prevented us from truly
ball thrown into the air. In fact, all of us ride Newton’s seeing our world and our universe.
cannonball as the force of the Sun’s gravity holds Earth
in its yearly orbit.
This thought experiment was an important step in
Newton’s work. Eventually it led him to invent calculus, We Are Stardust in
which he used to calculate the motions of the planets,
making predictions that were confi rmed by Johannes Human Form
Kepler’s empirical laws. The philosophical and scientific
significance of Newton’s thought experiment goes far As evidenced by the obelisks of Stonehenge or the ruins
deeper, however: it signifies the fi nal collapse of the bar- of a Mayan pyramid, humans have always built temples to
riers that humankind had placed between Earth and the the stars. We still build temples to the stars today. They are
heavens. With Newton’s insight we came to see the heav- seen as an array of radio telescopes spread across the high
ens as part of the world around us—made of the same sub- desert of New Mexico, or a city of domes atop the summit
stance and shaped by the same physical laws. It is ironic of a dormant Hawaiian volcano, or a satellite telescope car-
that for knowledge to progress, we had to turn the early ried into orbit and subsequently repaired by space shuttle
cornerstones of our thinking upside down. The modern astronauts, or tiny rovers crawling across the surface of
foundation of our understanding of the universe—the cos- Mars. These modern temples are the legacy of insights by
mological principle—is the literal negation of those early Copernicus, Kepler, Galileo, Newton, Einstein, Hubble, and
beliefs: Not only is Earth not the center of the universe, countless others.
but Earth occupies no special place in the universe at The discoveries that pour forth from these modern-day
all. The heavens are not “the other,” but are instead “the temples stretch the mind and stir the imagination. We have
same.” We can know the heavens by going into terrestrial walked on the Moon and have come to see the planets not as
laboratories and learning about the nature of matter and points of light in the sky, but as worlds as rich and complex
energy and radiation, and then applying this knowledge as our own. We have looked at the remnants of stars that
to careful observations of a universe of stars, planets, and exploded long ago and have peered into eerie columns of
galaxies that is governed by physical law. glowing interstellar gas within which new stars are being
In our journey we have followed the trail of discovery born. We have gazed back in time at galaxies forming when
that grew from this profound change in our understanding. the universe was young, and we have even learned to rec-
We have watched as stars and planets formed, as stars lived ognize the birth of the universe itself in the faint glow of
out their lives and died, and as galaxies coalesced out of the cosmic background radiation. As we contemplate those
the primordial fi reball of the Big Bang. We have followed wonders, the words from act I, scene V, of Shakespeare’s
our physics back to the very briefest of instants after the Hamlet seem almost frighteningly appropriate. As Hamlet
event that brought space and time into existence, and we faces the challenges and revelations brought by the ghost
have seen the hints of theories that may in our lifetimes of his father, Horatio cries out,
carry us the fi nal step. There is no doubting the wonder
of what we have seen. Yet for you, there is another aspect O day and night, but this is wondrous strange!
to our journey that, in a practical sense, should be even
more significant. While learning about the universe, you To this comes Hamlet’s immortal reply:
have also come to better appreciate how we know those
things—and in so doing you have found a powerful, work- And therefore as a stranger give it welcome.
able defi nition for what it means “to know.” There are more things in heaven and earth, Horatio,
The three most common standards that people apply to Than are dreamt of in your philosophy.
Epilogue: The Universe and You E-3
Here is a message to shout back through the ages. At envision the flat, rotating accretion disk from which our
the start of our journey we asked the most basic question Solar System formed. Once you realize what you are look-
about what we see in the sky—“How big is it?”—and the ing at, the cradle of our world and ourselves hangs there in
answers were enough to expose the comedy of humanity’s the night sky for all to see.
ancient conceits. Traveling at the speed of a modern jetliner, Viewed in this way, the sunset takes on a whole new
it would take us over 5 million years to cross the distance significance. It will never be the same. Yet even the maj-
to even the nearest star beyond our Sun. Even so, we live esty of the planets spread across the sky fails to capture the
in a galaxy containing hundreds of billions of such stars, intimacy of our connection with the universe. In the most
which are themselves outnumbered by other galaxies filling basic sense, the question “What are we?” is easily answered.
a universe that may stretch on forever. Using the speed of Humans and all other terrestrial life are an organized assem-
light as our yardstick, we have come to realize that Earth— blage of various organic molecules, most of which are very
the world of our birth and the stage on which all of human complex. Counting the numbers of atoms in our bodies, we
history has been played—is to the expanse of the observable are approximately 60 percent hydrogen, 26 percent oxygen,
universe what a single snap of our fi ngers is to the aeons 11 percent carbon, and 2 percent nitrogen, with a small
that have transpired since time itself came into existence, fraction of metals and other heavy elements mixed in. Over
roughly 14 billion years ago. the course of our journey we have witnessed the history of
As we stare at the images that have come to symbolize those atoms.
modern astronomy and consider what they show, it is easy A very long time ago—roughly 14 billion years—some-
to understand why some people recoil from these insights. thing wonderful happened. The universe came into being,
“Wouldn’t it be nice,” they say, “if we could just go back and time began. From an infinitesimally small volume of
to imagining that Earth is only 6,000 years old, and that concentrated energy, the universe expanded, growing ever
humanity occupies a special place at the pinnacle of cre- larger in size. Within the first few minutes, particles of solid
ation?” Indeed, if our story ended here—with the fact of matter, including protons and electrons, condensed out of
our seeming insignificance in the universe—we might all this dense, primordial ball of energy. Nuclear reactions
long for an excuse to retreat into ignorance. Fortunately, caused some of the protons to fuse into other light nuclei.
this is not where our story ends. Rather, this is where our Several hundred thousand years later, when the universe
true journey of discovery begins. Although modern science had cooled to a temperature of a few thousand kelvins, those
may have shattered our egotistical views about our exalted nuclei combined with electrons to form atoms. Of those
place in the scheme of things, it has also offered us a won- atoms, roughly 90 percent were hydrogen atoms and 10 per-
derful new appreciation and understanding of ourselves cent were helium atoms. There were traces of lithium, beryl-
to fill that void. lium, and boron as well, but that was all that existed in the
When we look at distant galaxies, the light we see is way of normal luminous matter as the universe expanded
produced by stars like our Sun. Each of those stars formed past the threshold of recombination.
when a cloud of interstellar gas and dust collapsed under The hydrogen atoms in our bodies date back to this early
the same force of gravity that guided Newton’s cannonball. time in the history of the universe, but what of the rest? Hav-
As each of those clouds collapsed, it spun faster and faster, ing taken our journey of discovery, you know the answers.
obeying the same laws of motion that accelerate the spin of As the universe emerged from the Big Bang, clumps of dark
an Olympic skater as she pulls her arms and leg ever more matter began to collapse under the force of gravity, pulling
tightly to her body. This spin prevented those clouds from normal matter along with it. Within these collapsing pro-
collapsing directly into stars, forcing them instead to settle togalaxies the fi rst generations of stars formed—nuclear
into flat, rotating disks. We see such disks today when we furnaces powered by the fusion of those original hydrogen
look at the newest generation of stars. Inside those disks, atoms into increasingly massive elements. Carbon, oxygen,
grains of dust stick together to make larger grains, which silicon, sulfur—elements all the way up to iron and nickel—
stick together to make still larger grains—this is the begin- were formed in those stellar infernos. As those fi rst genera-
ning of a bottom-up process that culminates with planetesi- tions of stars ended their lives, they blasted this nuclear ash
mals crashing together to make planetary worlds. Our Earth back into the reaches of interstellar space.
is one such world. We have come to view our Sun, Earth, Nucleosynthesis did not end with fusion, however. In the
and Solar System as products of natural processes still going extreme environments of supernovae, free neutrons were
on around us today. As we watch new generations of stars captured by the products of fusion, building more massive
form and we search for the planets that surround them, we elements still. Atoms of copper, zinc, tin, silver, and gold—
are witnessing a replay of the birth of our own world, 4.6 all the way up to the most massive naturally occurring ele-
billion years ago. We have found our roots in the stars. ment, uranium—were formed and expelled into space. Here
Go out at sunset on an evening when a crescent Moon were the chemical elements to fill the periodic table and to
hangs low above the western horizon and several planets build the compounds of life. As early protogalaxies merged
stretch out across the darkening sky. The plane of the eclip- to form early galaxies, more generations of stars continued
tic is there in front of you, and your mind’s eye might even to enrich the universe with the fruits of their alchemy. As
E-4 Epilogue: The Universe and You
early galaxies settled into the well-ordered ellipticals and or choosing to push them from our minds will not make
spirals of today’s universe, still more generations of stars them go away. Yet at the same time modern medicine, food
came and went, adding to the chemical richness of the production, transportation, and a thousand other technolo-
universe. When our Solar System appeared on the scene gies that push back the ancient scourges of humanity are
4.6 billion years ago, it formed from interstellar material equally real. Differences among people remain, but modern
that carried the chemical building blocks of planets and of communication offers the hope of spreading understanding.
life—atoms produced both in the Big Bang and in the hearts Meanwhile, a picture of Earthrise above the lunar horizon
of generations of stars that had lived and died during the 9 taken by the Apollo 8 astronauts more than four decades ago
billion years that had transpired since the universe began. (see Figure 7.1) remains forever a part of the human expe-
“What are we, and how did we get here?” We are stardust rience and perspective. It shows us, as nothing else could,
in human form. that Earth is a tiny fragile island to be cherished. Whether
Even as we contemplate our own place in the universe, we like it or not, humanity is a single, interrelated, interde-
however, we have come to realize that ordinary matter—the pendent global village that will face the future together—
atoms and molecules that constitute our bodies, our planet, or not at all.
our Sun, and the stars and galaxies that surround us—is not At the beginning of the 21st century, science has shown
the main constituent of our universe. Most of the universe us the wonders of the universe and at the same time has
consists of dark matter and dark energy, unseen components given us the knowledge and power to shape our world and
that we do not yet understand. In summary, we have now choose our future. The future of humanity may depend
learned a key lesson about the importance of humankind entirely on how well we treat Earth and ourselves over the
in the universe. First, Copernicus taught us that our home, few decades and centuries ahead. If 21st century astronomy
our Earth, is not the center of the universe. Then, Darwin does nothing else, it forces us to change our perspective. As
reminded us that humans are but a stage in the inevitable we study the laws that govern the workings of atoms, we
evolution of life on Earth. And now, modern cosmology tells learn something about the conditions of our own lives: the
us that we and all that we can see about us are but a tiny future is not yet written. As we use our telescopes to stare
component of all that is. It is true we are stardust—but we at galaxies 10 billion light-years away, collecting light from
aren’t as important in the overall scheme of the universe as stars that died billions of years before our Sun was even
we might like to think. born, manifest destiny seems a pretty silly concept. There
are no guarantees that things will work out in the end for
one tiny world or for the species to which it gave birth. If we
choose to destroy our world, through either direct action or
The Future Arrives Every Day simple neglect, so be it. The universe as a whole will carry
on in sublime indifference to our fate.
While traveling the highways and back roads of 21st century This is not, however, a message of despair. Instead, it is
astronomy, we have come to see our world and ourselves in a message of hope, responsibility, and maturity. We have
a very different light. Even so, nothing we have seen has the power to make our Earth a paradise or to leave our
changed the most basic circumstances of our day-to-day exis- children’s children to cope with a world choking from our
tence. Earth remains our world, our home. The hopes and shortsighted excess. The choices are ours, whether we want
dreams we humans feel are no less real today than they were them or not. To paraphrase from the book of Genesis, we
a thousand years ago. In 1968, Stanley Kubrick and Arthur have truly “tasted of the tree of knowledge.” As we stand at
C. Clarke collaborated to make the film 2001: A Space Odys- the second decade of the 21st century, we face a future filled
sey. This provocative piece of speculative fiction captured with choices; but one choice we are not allowed is refusing
the imagination of a generation, and the year 2001 came to to acknowledge responsibility for our own destiny.
signify the future. That future is now here: yet little of our With that thought, we come to the end of our journey,
modern-day life is recognizable in those cinematic prognos- with the hope that it has helped open your eyes to the won-
tications dating from more than four decades ago. It is ironic ders of the world and the universe around you. Even more,
that while we can forecast the future of the Sun with great we hope this journey has given you pause to reflect on who
accuracy and can even calculate the fate of the universe itself, we are as humans and on our place in the larger reality in
our vision of our own future is so much less certain. which we fi nd ourselves. Finally, we hope this journey has
There is no denying the fact that we live in an evolving changed the shape of the way you think—not only about
universe—a place of ongoing and unending change that the sights you see in the night sky, but also about the events
continues to shape humanity as certainly as it shapes the of your daily life. If any or all of these hopes are fulfilled,
cosmos itself. The grim prospects mentioned in the closing then the journey will have been worth taking—worthy of
sections of Chapter 23 are real. Saying that they do not exist your time and thought and of ours.
APPENDIX 1
Mathematical Tools
Luminosity of A Brightness of A
( Distance of B )
Distance of A 2
Working with Proportionalities _____________
Luminosity of B
= _____________ × ___________ .
Brightness of B
Most of the mathematics in 21st Century Astronomy
If we use the symbols L, b, and d to represent luminos-
involves proportionalities—statements about the way that
ity, brightness, and distance, respectively, this equation
one physical quantity changes when another quantity
becomes
changes. Here we offer a practical guide to working with
proportionalities.
To use a statement of proportionality to compare two
LA
__
LB
b
bB ( )
dA 2
= __A × __
dB
.
objects, begin by turning the proportionality into a ratio. As an example, suppose that star A appears twice as
For example, the price of a bag of apples is proportional to bright in the sky as star B, but star A is located 10 times
the weight of the bag: as far away as star B. Compare the luminosities of the two
stars. Because we know that
Price ∝ Weight.
Here the symbol “∝” is pronounced “is proportional Luminosity ∝ Brightness × Distance2,
to.” What this means is that the ratio of the prices of two
bags of apples is equal to the ratio of the weights of the we write
two bags: Luminosity of A _____________
Brightness of A
( Distance of B )
Distance of A 2
_____________ = × ___________
Price of A Weight of A Luminosity of A Brightness of B
Price ∝ Weight means ________ = __________ .
Price of B Weight of B
Let’s work a specific example. Suppose bag A weighs 2
2
( )
10 2
= __ × __ = 200.
1 1
pounds, and bag B weighs 1 pound. That means, of course,
that bag A will cost twice as much as bag B. We can turn In other words, star A is 200 times as luminous as star B.
our proportionality into this equation:: A fi nal note: In our original example we said that the
price of a bag of apples is proportional to the weight of the
Price of A Weight
________ of A 2 lb bag, which allowed us to say that a 2-pound bag of apples
= __________ = ____ = 2.
Price of B Weight of B 1 lb costs twice as much as a 1-pound bag of apples. This is a
In other words, the price of bag A is two times the price of statement about the way the world works. A 2-pound bag
bag B. of apples costs more than a 1-pound bag, not less than a
Now let’s work another, more complicated example. In 1-pound bag. To figure out how much a bag of apples actu-
Chapter 13 we discuss how the luminosity, brightness, and ally will cost, we need another piece of information: the
distance of stars are related. The luminosity of a star—the price per pound. The price per pound is an example of a
total energy that the star radiates each second—is propor- constant of proportionality. Wrapped up in the price per
tional to the star’s brightness multiplied by the square of pound is all sorts of information about the cost of growing
its distance: apples, the cost of transporting them, what the market is
like at the moment, the profit the grocer needs to make, and
Luminosity ∝ Brightness × Distance2. so forth. In other words, this constant of proportionality is
a statement about the way the world is.
What this proportionality means is that if we have two Proportionalities help us understand how the world works
stars—call them A and B—then and let us compare one object to another. Constants of pro-
A-1
A-2 Appendix
portionality enable us to calculate real values for things. In which also means “1 multiplied by six factors of 10.”
21st Century Astronomy, it is usually the “understanding”— Moving to the right of the decimal place, each step we
the proportionality itself—that we care about. take removes a power of 10 from the number. One-millionth
can be written
1 1 1 1 1 1
1 millionth = 1 × ___ × ___ × ___ × ___ × ___ × ___ .
Scientific Notation 10 10 10 10 10 10
Astronomy is a science of both the very large and the very We are removing powers of 10, so we express this number
small. The mass of an electron, for example, is using a negative exponent. We write
1
___
0.0000000000000000000000000000009109 kilograms (kg), = 10–1
10
and
whereas the distance to a galaxy far, far away might be
about 1 millionth = 1 × 10–1 × 10–1 × 10–1 × 10–1 × 10–1 × 10–1
100,000,000,000,000,000,000,000,000 meters. = 1 × 10–6
All it takes is a quick glance at these two numbers to see Returning to our earlier examples, the mass of an elec-
why astronomers, like most other scientists, make heavy tron is 9.109 × 10 –31 kg, and the distant galaxy is located 1
use of scientific notation to express numbers. × 1026 meters away. These are much more convenient ways
of writing this information. Notice that the exponent in
scientific notation gives you a feel for the size of a number
at a glance. The exponent of 10 in the electron mass is –31,
Powers of 10 which tells us instantly that it is a very small number. The
Our number system is based on powers of 10. Going to the exponent of 10 in the distance to a remote galaxy, +26, tells
left of the decimal place, us immediately that it is a very large number. This exponent
is often called the “order of magnitude” of a number. When
10 = 10 × 1, you see a number written in scientific notation while read-
ing 21st Century Astronomy (or elsewhere), just remember
100 = 10 × 10 × 1, to look at the exponent to better understand what the num-
1,000 = 10 × 10 × 10 × 1, ber is telling you.
Scientific notation is also convenient because it makes
and so on. Going to the right of the decimal place. multiplying and dividing numbers easier. Again, let’s look
1 at an example. Two billion multiplied by eight-thousandths
0.1 = ___ × 1, can be written
10
1 1 2,000,000,000 × 0.008,
0.01 = ___ × ___ × 1,
10 10
1 1 1 but it is more convenient to write these two numbers using
0.001 = ___ × ___ × ___ × 1,
10 10 10 scientific notation, as
and so on, for as long as we care to continue. In other ( 2 × 109 ) × ( 8 × 10–3 ).
words, each place to the right or left of the decimal place
in a number represents a power of 10. For example, 1 mil- We can regroup these expressions in the following form:
lion can be written
(2 × 8 ) × ( 109 × 10–3 )
1 million = 1,000,000
= 1 × 10 × 10 × 10 × 10 × 10 × 10. The fi rst part of the problem is just 2 × 8 = 16. The more
interesting part of the problem is the multiplication in the
right-hand parentheses. The fi rst number, 109, is just short-
That is, 1 million is “1 multiplied by six factors of 10.” hand for 10 × 10 × 10 . . . nine times. That is, it represents
Scientific notation combines these factors of 10 in conve- nine factors of 10. The second number stands for three fac-
nient shorthand. Rather than writing out all six factors of tors of 1 ⁄ 10—or removing three factors of 10 if you prefer to
10, instead we combine them in easy shorthand using an think of it that way. Altogether, that makes 9 – 3 = 6 factors
exponent: of 10. In other words,
1 million = 1 × 106 109 × 10–3 = 109–3 = 106.

Appendix A-3
Putting the problem together, we get number. For example, if we know that the distance to the
store (call it d) is between 0.8 and 1.2 km, we can write
(2 × 109 ) × ( 8 × 10–3 ) = ( 2 × 8 ) × ( 109 × 10–3 )
d = 1.0 ± 0.2 km,
= 16 × 106.
where the symbol ± is pronounced “plus or minus.” In this
By convention, when a number is written in scientific nota- example, d is between 1.0 – 0.2 = 0.8 km and 1.0 + 0.2 = 1.2
tion, only one digit is placed to the left of the decimal point. km. This is an unambiguous statement about the limitations
In this case, there are two. However, 16 is 1.6 × 10, so we on our knowledge of the value of d, but carrying along the
can add this additional factor of 10 to the exponent at right, formal errors with every number we write would be cum-
making the fi nal answer bersome at best. Instead, we keep track of the approximate
precision of a number by using “significant figures.”
1.6 × 107. The convention for significant figures is this: We assume
the number we write has been rounded from a number that
Dividing is just the inverse of multiplication. Dividing had one additional digit to the right of the decimal point.
by 103 means removing three factors of 10 from a number. If we say that a quantity d, which might represent the dis-
Using the previous number. tance to the store, is “1.”, what we mean is that d is close to
1. It is likely not as small as “0.”, and it is likely not as large
( 1.6 × 107 ) ÷ ( 2 × 103 ) = ( 1.6 ÷ 2 ) × ( 107 ÷ 103 ) as “2.”. If we say instead
= 0.8 × 107–3
d = 1.0,
= 0.8 × 104.
then we mean that d is likely not 0.9 and is likely not 1.1. It
This time we have only a zero to the left of the decimal point. is roughly 1.0 to the nearest tenth. The greater the number
To get the number into proper form, we can substitute 8 × of significant figures, the more precisely the number is being
10 –1 for 0.8, giving us specified. For example, 1.00000 is not the same number as
1.00. The first number, 1.00000, represents a value that is
0.8 × 104 = ( 8 × 10–1 ) × 104 = 8 × 103. probably not as small as 0.99999 and probably not as large as
1.00001. The second number, 1.00, represents a value that is
Adding and subtracting numbers in scientific notation probably not as small as 0.99 nor as large as 1.01. The number
is somewhat more difficult, because all numbers must be 1.00000 is much more precise than the number 1.00.
written as values multiplied by the same power of 10 before When we carry out mathematical operations, signifi-
they can be added or subtracted. However, almost all mod- cant figures are important. For example, 2.0 × 1.6 = 3.2. It
ern calculators (and even some cell phones) have scientific does not equal 3.20000000000. The product of two numbers
notation built in. They keep up with the powers of 10 for cannot be known to any greater accuracy than the numbers
you. If you do not have such a calculator, you may want to themselves! As a general rule, when we multiply and divide,
buy one and learn to use it before tackling the mathemati- the answer should have the same number of significant fig-
cal problems in this book. There are more examples on our ures as the less precise of the numbers being multiplied or
website that you can use to better learn how to work with divided. In other words, 2.0 × 1.602583475 = 3.2. Because
scientific notation. all we know is that the fi rst factor is probably closer to 2.0
than to 1.9 or 2.1, all we know about the product is that it
is between about 3.0 and 3.4. It is 3.2. It is not 3.205166950
(even if that is the answer your calculator gives!). The rest of
Significant Figures the digits to the right of 3.2 just do not mean anything.
In the previous example, we actually broke some rules in When we add and subtract, the rules are a bit different.
the interest of explaining how powers of 10 are treated in If one number has a significant figure with a particular
scientific notation. The rules we broke involve the preci- place value but another number does not, their sum or dif-
sion of the numbers we are expressing. In everyday speech ference cannot have a significant figure in that place value.
we might say, “The store is a kilometer away,” by which we For example,
probably mean that the store is roughly a kilometer away. If 1,045.
it turned out to be 0.8 or 1.2 kilometers (km), it is unlikely
+1.34567
that the recipient of our directions would quibble. But when
expressing quantities in science, it is extremely important 1,046.
to know not only the value of a number, but also how pre-
cise that value is. The answer is “1,046.”, not “1,046.34567”. Again, the extra
The most complete way to keep track of the precision of digits to the right of the decimal place have no meaning
numbers is to actually write down the uncertainty in the because “1,045.” is not known to that accuracy.
A-4 Appendix
There is always a fly in the ointment, and this is no that. Algebra provides rules for manipulating the symbols
exception. What is the precision of the number 1,000,000? used to represent quantities. We begin with a bit of nota-
As it is written, the answer is unclear. Are all those zeros tion for “powers” and “roots.” When we talk about raising
really significant, or are they placeholders? If we write the a quantity to a power, we mean multiplying the quantity by
number in scientific notation, on the other hand, there is itself some number of times. For example, if S is a symbol for
never a question. Instead of 1,000,000, we write 1.0 × 106 for something (anything), then S2 (pronounced “S squared” or
a number that is known to the nearest hundred thousand “S to the second power”) means S × S, and S 3 (pronounced
or so; or we write 1.00000 × 106 for a number that is known “S cubed” or “S to the third power”) means S × S × S. Sup-
to the nearest 10. pose S represents the length of the side of a square. The area
So our earlier example would have been more correct of the square is given by
had we said
Area = S × S = S 2.
( 2.0 × 10 9)
× ( 8.0 ×10 –3 )
= 1.6 × 10 .
7
If S = 3 meters (m), then the area of the square is
S2 = 3 m × 3 m = 9 m2
Algebra
(pronounced “9 square meters”). It should be obvious why
There are many branches of mathematics. The branch that raising a quantity to the second power is called “squaring”
tells us about the relationships between quantities is called the quantity. We could have done the same thing for the
algebra. If you are reading this book, you have almost cer- sides of a cube and found that the volume of the cube is
tainly taken an algebra class, but you are not alone if you
feel a little review is in order. Basically, algebra begins by
Volume = S × S × S = S3.
using symbols to represent quantities. For example, we
might write the distance you travel in a day as d. As it
If S = 3 meters, then the volume of the cube is
stands, d has no value. It might be 10,000 miles. It might
be 30 feet. It does, however, have units—in this case, the
S3 = 3 m × 3 m × 3 m = 27 m3
units of distance.
The average speed at which you travel is equal to the
distance you travel divided by the time you take. If we use (pronounced “27 cubic meters”). Again, it is clear why rais-
the symbol v to represent your average speed and the sym- ing a quantity to the third power is called “cubing” the
bol t to represent the time you take, then instead of writing quantity.
out, “Your average speed is equal to the distance you travel Roots are the reverse of this process. The square root of
divided by the time you take,” we can write a quantity is the value that, when squared, gives the origi-
nal quantity. The square root of 4 is 2, which means that 2
d × 2 = 4. The square root of 9 is 3, which means that 3 × 3
v = __.
t = 9. Similarly, the cube root of a quantity is the value that,
The meaning of this algebraic expression is exactly the same when cubed, gives the original quantity. The cube root of
as the sentence quoted before it, but it is much more concise. 8 is 2, which means _
that 2 × 2 × 2 = 8. Roots are written
As it stands, v, d, and t still have no specific values. There are with the symbol √ . For example, we write
no numbers assigned to them yet. However, this expression _
√9 = 3
tells us what the relationship between those numbers will
be when we do look at a specific example. For example, if
you go 500 km (d = 500 km) in 10 hours (t = 10 hours), this for the square root of 9, and
expression tells you that your average speed is _
3
√8 =2
d 500 km
v = __ = _______ = 50 km/h. for the cube root of 8. If the volume of a cube is V = S 3, we
t 10 hours
can also write
Notice that the units in this expression act exactly like the __ __
3 3
numerical values. They are just multiplicative factors. When S = √V = √S 3 .
we say “500 km,” what we really mean is “500 × kilometers.”
Likewise, 10 hours means “10 × hours.” When we divide Roots can also be written as powers. Powers and roots
the two, we fi nd that the units of v are kilometers divided behave exactly like the exponents of 10 in our discussion
by hours, or km/h (pronounced “kilometers per hour”). of scientific notation. (They had better—the exponents used
We introduced algebra as shorthand for expressing rela- in scientific notation are just powers of 10.) For example, if
tions between quantities, but it is far more powerful than a, n, and m are all algebraic quantities, then
Appendix A-5
an
an × am = an+m, and __ = an–m. The natural way to measure angles is to use a unit called
am the radian. As shown in Figure A1.1a, the size of an angle
(To see if you understand all this, explain why the square in radians is just the length of the arc subtending the angle,
1
root of a can also be written a ⎯2 and the cube root of a can divided by the radius of the circle. In the figure, the angle
1
be written a⎯3.) x = S/r radians.
Some of the rules of algebra are listed next. These are Because the circumference of a circle is 2π multiplied by
really no more than the rules of arithmetic applied to the the radius, C = 2πr, a complete circle has an angular mea-
symbolic quantities of algebra. The important thing is this: sure of (2πr)/r = 2π radians. In more conventional angular
as long as we apply the rules of algebra properly, then the measure, a complete circle is 360°, so we can say that
relationships among symbols we arrive at through our alge-
braic manipulations remain true for the physical quantities 360° = 2π radians
that those symbols represent.
Here we summarize a few algebraic rules and relation- or that
ships. In this summary, a, b, c, m, n, r, x, and y are all alge- 360°
braic quantities: 1 radian = ____ = 57.2958°.
2π
Associative rule: When talking about stars and galaxies, we often use
seconds of arc (arcseconds) to measure angles. A degree is
a × b × c = (a × b) × c = a × (b × c) broken into 60 minutes of arc (arcminutes), each of which
is broken into 60 seconds of arc—so there are 3,600 seconds
Commutative rule: of arc in a degree. Therefore,
a×b=b×a arcseconds degrees arcseconds

3,600 _________ × 57.2958 ______ = 206,265 _________ .
degree radian radian
Distributive rule:
a × (b + c) = (a × b) + (a × c) (a) The natural unit for measuring

angles, called the radian, is
equal to the arc subtended by
Cross multiplication: an angle, divided by the radius
of the arc.
a c
If __ = __ , then ad = bc.
b d
Working with exponents: Angle x S
r
1
__ x = S radians
n
= a–n an am = an+m r
a
n
If an angle is very small,
a n
an
__
am = a
n‒m
(an)m = an × m ()
a
__
b
= __n
b
there is no real difference
between the pie
slice–shaped wedge and
(b)
Equation of a line with slope m and y-intercept b: a long, skinny triangle.
r
S
y = mx + b
x = S radians
Equation of a circle with radius r centered at x = 0, y = 0: r
r
x2 + y2 = r2 S
x(radians)
r
Angles and Distances Size
Size = x(radians) distance As long as the angle is small,

The farther away something is, the smaller it appears. This is the size of an object is equal to
= x(arcseconds) distance the angle it subtends, measured
common sense and everyday experience. In astronomy, where 206,265
in radians, multiplied by the
we seldom get to walk up to the object we are studying and distance to the object.
measure it with a meterstick, our knowledge about the sizes of
things usually depends on knowing how the size of an object,
its distance, and the angle it covers in the sky are related. FIGURE A1.1 Measuring angles.
A-6 Appendix
If the angle is small enough (which it usually is in astron-

omy), there is precious little difference between the pie slice
Circles and Spheres
just described and a long skinny triangle with a short side
of length S, as Figure A1.1b illustrates. So, if we know the To round out our mathematical tools, here are a few useful
distance d to an object and we can measure the angular size formulas for circles and spheres. The circle or sphere in
x of the object, then the size of the object is just each case has a radius r.
x (in degrees)
S = x (in radians) × d = ___________________ × d Circumferencecircle = 2πr
57.2958 degrees/radian
x (in arcseconds) Areacircle = πr 2
= ______________________ × d,
206,265 arcseconds/radian
Surface areasphere = 4πr 2
which is all we need to turn our knowledge of the angular
4
size and the distance to an object into a measurement of the Volumesphere = __πr 3
3
object’s physical size.
APPENDIX 2
Physical Constants and Units
Fundamental Physical Constants
Constant Symbol Value
Speed of light in a vacuum c 2.99792 × 108 m/s
Universal gravitational constant G 6.673 × 10 –11 N m 2/kg2
Planck constant h 6.62607 × 10 –34 J-s
Electric charge of electron or proton e 1.60218 × 10 –19 C
Boltzmann constant k 1.38065 × 10 –23 J/K
Stefan-Boltzmann constant σ 5.67040 × 10 –8 W/(m 2 K4)
Mass of electron me 9.10938 × 10 –31 kg
Mass of proton mp 1.67262 × 10 –27 kg
Source: Data from the National Institute of Standards and Technology (http://physics.nist.gov).
Unit Prefixes
Prefi x* Name Factor†
n nano- 10 –9
μ micro- 10 –6
m milli- 10 –3
k kilo- 103
M mega- 106
G giga- 109
T tera- 1012
These prefi xes (*), when appended to a unit, change the size of the
unit by the factor (†) given. For example, 1 km (kilometer) is 103
meters (m).
A-7
A-8 Appendix
Units and Values
Quantity Fundamental Unit Values

Length meters (m) radius of Sun (R ⊙) = 6.96265 × 108 m
astronomical unit (AU) = 1.49598 × 1011 m
1 AU = 149,598,000 km
Light-year (ly) = 9.4605 × 1015 m
1 ly = 6.324 × 104 AU
1 parsec (pc) = 3.261 ly
= 3.0857 × 1016 m
1 m = 3.281 feet
Volume meters3 (m3) 1 m3 = 1,000 liters

= 264.2 gallons
Mass kilograms (kg) 1 kg = 1,000 grams

mass of Earth (M ⊕ ) = 5.9736 × 1024 kg
mass of Sun (M ⊙ ) = 1.9891 × 1030 kg
Time seconds (s) 1 hour (h) = 60 minutes (min) = 3,600 s

solar day (noon to noon) = 86,400 s
sidereal day (Earth rotation period) = 86,164.1 s
tropical year (equinox to equinox) = 365.24219 days
= 3.15569 × 107 s
sidereal year (Earth orbital period) = 365.25636 days
= 3.15581 × 107 s
Speed meters/second (m/s) 1 m/s = 2.236 miles/h

1 km/s = 1,000 m/s = 3,600 km/h
c = 3.00 × 108 m/s = 300,000 km/s
Acceleration meters/second 2 (m/s2) gravitational acceleration on Earth (g) = 9.81 m/s2
Energy joules (J) 1 J = 1 kg m 2/s2

1 megaton = 4.19 × 1015 J
Power watts (W) 1 W = 1 J/s

Solar luminosity (L ⊙ ) = 3.827 × 1026 W
Force newtons (N) 1 N = 1 kg m/s2

1 pound (lb) = 4.448 N
1 N = 0.22481 lb
Pressure newtons/meter2 (N/m 2) atmospheric pressure at sea level = 1.013 × 105 N/m 2
= 1.013 bar
Temperature kelvins (K) absolute zero = 0 K = –273.15˚C = –459.67˚F
Sources: Data from the National Space Science Data Center (2002); Observer’s Handbook 2002 (Royal Astronomical
Society of Canada, 2001); and National Institute of Standards and Technology (2002).
APPENDIX 3
Periodic Table of the Elements
1 PERIODIC TABLE OF THE ELEMENTS 18

1A 8A
1 1 Atomic number Metals 2
H H Symbol
Metalloids
He
1 Hydrogen 2 Hydrogen Name 13 14 15 16 17 Helium
1.00794 2A 1.00794 Average atomic mass Nonmetals 3A 4A 5A 6A 7A 4.002602
3 4 5 6 7 8 9 10
Li Be B C N O F Ne
2 Lithium Beryllium Boron Carbon Nitrogen Oxygen Fluorine Neon
6.941 9.012182 10.811 12.0107 14.0067 15.9994 18.9984032 20.1797

11 12 13 14 15 16 17 18
Na Mg Al Si P S Cl Ar
3 Sodium Magnesium 3 4 5 6 7 8 9 10 11 12 Aluminum Silicon Phosphorus Sulfur Chlorine Argon
22.98976928 24.3050 3B 4B 5B 6B 7B 8B 1B 2B 26.9815386 28.0855 30.973762 32.065 35.453 39.948

19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
4 Potassium Calcium Scandium Titanium Vanadium Chromium Manganese Iron Cobalt Nickel Copper Zinc Gallium Germanium Arsenic Selenium Bromine Krypton
39.0983 40.078 44.955912 47.867 50.9415 51.9961 54.938045 55.845 58.933195 58.6934 63.546 65.409 69.723 72.64 74.92160 78.96 79.904 83.798
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
5 Rubidium Strontium Yttrium Zirconium Niobium Molybdenum Technetium Ruthenium Rhodium Palladium Silver Cadmium Indium Tin Antimony Tellurium Iodine Xenon
85.4678 87.62 88.90585 91.224 92.90638 95.94 [98] 101.07 102.90550 106.42 107.8682 112.411 114.818 118.710 121.760 127.60 126.90447 131.293
55 56 57 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
Cs Ba La Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
6 Cesium Barium Lanthanum Hafnium Tantalum Tungsten Rhenium Osmium Iridium Platinum Gold Mercury Thallium Lead Bismuth Polonium Astatine Radon
132.9054519 137.327 138.90547 178.49 180.94788 183.84 186.207 190.23 192.217 195.084 196.966569 200.59 204.3833 207.2 208.98040 [209] [210] [222]
87 88 89 104 105 106 107 108 109 110 111
Fr Ra Ac Rf Db Sg Bh Hs Mt Ds Rg
7 Francium Radium Actinium Rutherfordium Dubnium Seaborgium Bohrium Hassium Meitnerium Darmstadtium Roentgenium
[223] [226] [227] [261] [262] [266] [264] [277] [268] [271] [272]
58 59 60 61 62 63 64 65 66 67 68 69 70 71
Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
6 Lanthanides Cerium Praseodymium Neodymium Promethium Samarium Europium Gadolinium Terbium Dysprosium Holmium Erbium Thulium Ytterbium Lutetium
140.116 140.90765 144.242 [145] 150.36 151.964 157.25 158.92535 162.500 164.93032 167.259 168.93421 173.04 174.967
90 91 92 93 94 95 96 97 98 99 100 101 102 103
Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
7 Actinides Thorium Protactinium Uranium Neptunium Plutonium Americium Curium Berkelium Californium Einsteinium Fermium Mendelevium Nobelium Lawrencium
232.03806 231.03588 238.02891 [237] [244] [243] [247] [247] [251] [252] [257] [258] [259] [262]
We have used the United States system as well as the system recommended by the International Union of Pure and Applied Chemistry (IUPAC) to label the groups in this periodic table. The system used
in the United States includes a letter and a number (1A, 2A, 3B, 4B, etc.), which is close to the system developed by Mendeleev. The IUPAC system uses numbers 1–18 and has been recommended by the
American Chemical Society (ACS). Elements with atomic numbers 112 and higher have been reported, but not yet fully authenticated.
A-9
APPENDIX 4
Properties of Planets,
Dwarf Planets, and Moons
Physical Data for Planets and Dwarf Planets

Tilt of
Rotation Equatorial
Average Axis Surface Average
Equatorial Density Rotation (degrees, Gravity Escape Surface
Radius Mass (relative Period relative (relative to Velocity Temperature
Planet (km) (R/R ⊕) (kg) (M/M ⊕) to water*) (days) to orbit) Earth†) (km/s) (K)
Mercury 2,440 0.383 3.30 × 1023 0.055 5.427 58.64 0.01 0.378 4.3 340 (100, 725)§
Venus 6,052 0.949 4.87 × 1024 0.815 5.243 243.02‡ 177.36 0.907 10.36 737
Earth 6,378 1.000 5.97 × 1024 1.000 5.515 1.000 23.45 1.000 11.19 288 (183, 331)§
Mars 3,397 0.533 6.42 × 1023 0.107 3.933 1.0260 25.19 0.377 5.03 210 (133, 293)§
Ceres 475 0.075 9.60 × 1020 0.0002 2.100 0.378 3.0 0.27 0.51 200
Jupiter 71,492 11.209 1.90 × 1027 317.83 1.326 0.4136 3.13 2.364 59.5 165
Saturn 60,268 9.449 5.68 × 1026 95.16 0.687 0.4440 26.73 0.916 35.5 134
Uranus 25,559 4.007 8.68 × 1025 14.537 1.270 0.7183‡ 97.77 0.889 21.3 76
Neptune 24,764 3.883 1.02 × 1026 17.147 1.638 0.6713 28.32 1.12 23.5 58
Pluto 1,195 0.187 1.25 × 1022 0.0021 1.750 6.387‡ 122.53 0.083 1.3 40
21
Haumea ~700 0.11 4.0 × 10 0.0007 ~3 0.163 ? 0.045 0.84 <50
Makemake 750 0.12 4.18 × 1021 0.0007 ~2 0.32 ? 0.048 0.8 ~30
Eris 1,200 0.188 1.5 × 1022 0.0025 ~2 >0.3? ? 0.082 ~1.3 30

(est.) (est.)
*The density of water is 1,000 kg/m3.

†
The surface gravity of Earth is 9.81 m/s2.
‡
Venus, Uranus, and Pluto rotate opposite to the directions of their orbits. Their north poles are south of their orbital planes.
§
Where given, values in parentheses give extremes of recorded temperatures.
A-10
Appendix A-11
Orbital Data for Planets and Dwarf Planets

Inclination
Mean Distance from Sun (A*)
Orbital Period (P) (degrees, relative Average Speed
Planet (106 km) (AU) (sidereal years) Eccentricity to ecliptic) (km/s)
Mercury 57.9 0.387 0.2408 0.2056 7.005 47.87
Venus 108.2 0.723 0.6152 0.0067 3.395 35.02
Earth 149.6 1.000 1.000 0.0167 0.000 29.78
Mars 227.9 1.524 1.8809 0.0935 1.850 24.13
Ceres 413.9 2.767 4.6027 0.097 9.73 17.88
Jupiter 778.6 5.204 11.8618 0.0489 1.304 13.07
Saturn 1,433.5 9.582 29.4566 0.0565 2.485 9.69
Uranus 2,872.5 19.201 84.0106 0.0457 0.772 6.81
Neptune 4,495.1 30.047 164.7856 0.0113 1.769 5.43
Pluto 5,906.38 39.48 247.6753 0.2488 17.16 4.72
Haumea 6,447.8 43.1 283.3 0.195 28.22 4.48
Makemake 6,836.7 45.7 309.9 0.159 28.96 4.42
Eris 10,123 67.668 557. 0.4418 44.187 3.44
*A is the semimajor axis of the planet’s elliptical orbit.
Properties of Selected Moons*

Orbital Properties Physical Properties
Relative
P A R M Density†
Planet Moon (days) (103 km) (km) (1020 kg) (water = 1.00)
Earth (1 moon) Moon 27.32 384.4 1,737.4 735 3.34
Mars (2 moons) Phobos 0.32 9.38 13.5 × 10.8 × 9.4 0.0001 2.0
Deimos 1.26 23.46 7.5 × 6.1 × 5.5 0.00002 1.5
Jupiter (63 known moons) Metis 0.29 127.97 20 0.00036 0.9
Amalthea 0.50 181.30 131 × 73 × 67 0.0208 0.9
Io 1.77 421.60 1,815 894 3.55
Europa 3.55 670.90 1,569 480 3.01
Ganymede 7.16 1,070 2,631 1,480 1.94
Callisto 16.69 1,883 2,403 1,080 1.86
Himalia 250.57 11,480 93 0.042 2.8
Pasiphae 735‡ 23,500 25 0.0019 2.9
Callirrhoe 759‡ 24,100 4.3 0.00001 2.6
(continued)
Properties of Selected Moons*
(continued) Orbital Properties Physical Properties
Relative
P A R M Density†
Planet Moon (days) (103 km) (km) (1020 kg) (water = 1.00)
Saturn (61 known moons) Pan 0.58 133.58 14 0.00005 —
Prometheus 0.61 139.35 72.5 × 42.5 × 32.5 0.0016 0.5
Pandora 0.63 141.70 57 × 42 × 31 0.0014 0.5
Mimas 0.94 185.52 196 0.38 1.17
Enceladus 1.37 238.02 250 1.08 1.6
Tethys 1.89 294.66 530 6.17 0.97
Dione 2.74 377.40 560 10.5 1.43
Rhea 4.52 527.04 765 24.9 1.23
Titan 15.95 1,222 2,575 1,350 1.88
Hyperion 21.28 1,481 205 × 130 × 110 0.0558 0.57
Iapetus 79.33 3,561 735 18.8 1.21
Phoebe 550.48‡ 12,952 107 0.08 1.6
Paaliaq 686.9 15.200 11 0.0001 2.3
Uranus (27 known moons) Cordelia 0.34 49.75 20 0.0004 1.3
Miranda 1.41 129.78 236 0.64 1.15
Ariel 2.52 191.24 579 13.6 1.66
Umbriel 4.14 265.97 585 11.7 1.39
Titania 8.71 435.84 789 34.9 1.70
Oberon 13.46 582.60 761 30.3 1.64
Setebos 2,225‡ 17,418 24 0.0008 1.5
Neptune (13 known moons) Naiad 0.29 48.0 48 × 30 × 26 0.002 1.3
Larissa 0.55 73.6 108 × 102 × 84 0.05 1.3
Proteus 1.12 117.6 210 × 208 × 202 0.5 1.3
Triton 5.88‡ 354.8 1,353 214 2.07
Nereid 360.14 5,513.40 170 0.3 1.5
Pluto (3 moons) Charon 6.39 19.60 593 16.2 1.85
Haumea (2 moons) Namaka 18 25.66 85 0.018 ?
Hi’iaka 49 49.88 155 0.179 ?
Eris Dysnomia 15.8 37.4 50–125? ? ?
*Innermost, outermost, largest, and/or a few other moons for each planet.
†
The density of water is 1,000 kg/m3.
‡
Irregular moon (has retrograde orbit).
APPENDIX 5
Nearest and Brightest Stars
Stars within 12 Light-Years of Earth
Distance Relative Visual Luminosity‡ Apparent Absolute

†
Name* (ly) Spectral Type (Sun = 1.000) Magnitude Magnitude
Sun 1.58 × 10 –5 G2V 1.000 –26.74 4.83
Alpha Centauri C 4.24 M5.5V 0.000052 11.09 15.53

(Proxima Centauri)
Alpha Centauri A 4.36 G2V 1.5 0.01 4.38
Alpha Centauri B 4.36 K0V 0.44 1.34 5.71
Barnard’s star 5.96 M4Ve 0.00044 9.53 13.22
CN Leonis 7.78 M5.5 0.000020 13.44 16.55
BD +36-2147 8.29 M2.0V 0.0056 7.47 10.44
Sirius A 8.58 A1V 22 –1.43 1.47
Sirius B 8.58 DA2 0.0025 8.44 11.34
BL Ceti 8.73 M5.5V 0.000059 12.54 15.40
UV Ceti 8.73 M6.0 0.000039 12.99 15.85
V1216 Sagittarii 9.68 M3.5V 0.00050 10.43 13.07
HH Andromedae 10.32 M5.5V 0.00010 12.29 14.79
Epsilon Eridani 10.52 K2V 0.28 3.73 6.19
Lacaille 9352 10.74 M1.5V 0.011 7.34 9.75
FI Virginis 10.92 M4.0V 0.00033 11.13 13.51
EZ Aquarii A 11.26 M5.0V 0.000047 13.33 15.64
EZ Aquarii B 11.26 M5e 0.000050 13.27 15.58
EZ Aquarii C 11.26 — 0.000025 14.03 16.34
Procyon A 11.40 F5IV-V 7.31 0.38 2.66
(continued)
A-13
A-14 Appendix
Stars within 12 Light-Years of Earth

(continued)
Distance Relative Visual Luminosity‡ Apparent Absolute
Name* (ly) Spectral Type† (Sun = 1.000) Magnitude Magnitude
Procyon B 11.40 DA 0.00054 10.70 12.98
61 Cygni A 11.40 K5.0V 0.086 5.21 7.49
61 Cygni B 11.40 K7.0V 0.040 6.03 8.31
Gliese 725 A 11.52 M3.0V 0.0029 8.90 11.16
Gliese 725 B 11.52 M3.5V 0.0014 9.69 11.95
Groombridge 34 A 11.62 M1.5V 0.0063 8.08 10.32
Groombridge 34 B 11.62 M3.5V 0.00041 11.06 13.30
Epsilon Indi A 11.82 K5Ve 0.15 4.69 6.89
Epsilon Indi B (brown 11.82 T1.0 — — —

dwarf)
Epsilon Indi C (brown 11.82 T6.0 — — —

dwarf)
DX Cancri 11.82 M6.5V 0.000014 14.78 16.98
Tau Ceti 11.88 G8V 0.45 3.49 5.68
Gliese 1061 11.99 M5.5V 0.000067 13.09 15.26
*Stars may carry many names, including common names (such as Sirius), names based on their prominence within a constellation (such as
Alpha Canis Majoris, another name for Sirius), or names based on their inclusion in a catalog (such as BD +36-2147). Addition of letters A, B, and
so on, or superscripts indicates membership in a multiple-star system.
†
Spectral types such as M3 are discussed in Chapter 13. Other letters or numbers provide additional information. For example, V after the spec-
tral type indicates a main-sequence star, and III indicates a giant star. Stars of spectral type T are brown dwarfs.
‡
Luminosity in this table refers only to radiation in “visual” light.
Appendix A-15
The 25 Brightest Stars in the Sky

Relative Visual Apparent Absolute
Common Distance Luminosity* Visual Visual
Name Name (ly) Spectral Type (Sun = 1.000) Magnitude Magnitude
Sun Sun 1.58 × 10 –5 G2V 1.000 –26.8 4.82
Alpha Canis Sirius 8.60 A1V 22.9 –1.47 1.42

Majoris
Alpha Carinae Canopus 313 F0II 13,800 –0.72 –5.53
Alpha Bootis Arcturus 36.7 K1.5IIIFe-0.5 111 –0.04 –0.29
Alpha1 Centauri Rigel Kentaurus 4.39 G2V 1.50 –0.01 4.38
Alpha Lyrae Vega 25.3 A0Va 49.7 0.03 0.58
Alpha Aurigae Capella 42.2 G5IIIe+G0III 132 0.08 –0.48
Beta Orionis Rigel 770 B8Ia: 40,200 0.12 –6.69
Alpha Canis Procyon 11.4 F5IV-V 7.38 0.34 2.65

Minoris
Alpha Eridani Achernar 144 B3Vpe 1,090 0.50 –2.77
Alpha Orionis Betelgeuse 427 M1-2Ia-Iab 9,600 0.58 –5.14
Beta Centauri Hadar 350 B1III 8,700 0.60 –5.03
Alpha Aquilae Altair 16.8 A7V 11.1 0.77 2.21
Alpha Tauri Aldebaran 65.1 K5+III 151 0.85 0.63
Alpha Virginis Spica 262 B1III-IV+B2V 2,230 1.04 –3.55
Alpha Scorpii Antares 604 M1.5Iab-Ib+B4Ve 11,000 1.09 –5.28
Beta Geminorum Pollux 33.7 K0IIIb 31.0 1.15 1.09
Alpha Piscis Fomalhaut 25.1 A3V 17.2 1.16 1.73
Alpha Cygni Deneb 3,200 A2Ia 260,000 1.25 –8.73
Beta Crucis Becrux 353 B0.5III 3,130 1.30 –3.92
Alpha2 Centauri Alpha Centauri B 4.39 K1V 0.44 1.33 5.71
Alpha Leonis Regulus 77.5 B7V 137 1.35 –0.52
Alpha Crucis Acrux 321 B0.5IV 4,000 1.40 –4.19
Epsilon Canis Adara 431 B2II 3,700 1.51 –4.11

Majoris
Gamma Crucis Gacrux 88.0 M3.5III 142 1.59 –0.56
Sources: Data from The Hipparcos and Tycho Catalogues, 1997, European Space Agency SP-1200; SIMBAD Astronomical Database (http://
simbad.u-strasbg.fr/simbad); and Research Consortium on Nearby Stars (www.chara.gsu.edu/RECONS).
*Luminosity in this table refers only to radiation in “visual” light
APPENDIX 6
Observing the Sky
The purpose of this appendix is to provide enough informa- The latitude-like coordinate on the celestial sphere is
tion so that you can make sense of a star chart or list of astro- called “declination,” often signified with the Greek letter δ
nomical objects, as well as fi nd a few objects in the sky. (delta). The celestial equator has δ = 0°. The north celestial
pole has δ = +90°. The south celestial pole has δ = –90°.
(See Chapter 2 if you need to refresh your memory about
the celestial equator or celestial poles.) Declination is usu-
Celestial Coordinates ally expressed in degrees, minutes of arc, and seconds of
arc. For example, Sirius, the brightest star in the sky, has δ
In Chapter 2 we discuss the celestial sphere—the imaginary = –16° 42′ 58″, meaning that it is located not quite 17° south
sphere with Earth at its center upon which celestial objects of the celestial equator.
appear to lie. A number of different coordinate systems On Earth, east–west position is specified by longitude.
are used to specify the positions of objects on the celestial Lines of constant longitude run north–south from one pole
sphere. The simplest of these is the “altitude-azimuth coor- to the other. Unlike latitude, for which the equator provides a
dinate system.” The altitude-azimuth coordinate system is natural place to call “zero,” there is no natural starting point
based on the “map” direction to an object (the object’s azi- for longitude, so we just have to invent one. By arbitrary con-
muth, with north = 0°, east = 90°, south = 180°, and west vention, the Royal Observatory in Greenwich, England, is
= 270°) combined with how high the object is above the defined to lie at a longitude of 0°. On the celestial sphere the
horizon (the object’s altitude, with the horizon at 0° and longitude-like coordinate is called “right ascension,” often
the zenith at 90°). For example, an object that is 10° above signified with the Greek letter α (alpha). Unlike the case with
the eastern horizon has an altitude of 10° and an azimuth longitude, there is a natural point on the celestial sphere to
of 90°. An object that is 45° above the horizon in the south- use as the starting point for right ascension: the vernal equi-
west is at altitude 45°, azimuth 225°. nox, or the point at which the ecliptic crosses the celestial
The altitude-azimuth coordinate system is the simplest equator with the Sun moving from the southern sky into the
way to tell someone where in the sky to look at the moment, northern sky. The vernal equinox defines the line of right
but it is not a good coordinate system for cataloging the ascension at which α = 0°. The autumnal equinox, located
positions of objects. The altitude and azimuth of an object on the opposite side of the sky, is at α = 180°.
are different for each observer, depending on the observ- Normally, right ascension is measured in units of time
er’s position on Earth, and they are constantly changing as rather than degrees. It takes Earth 24 hours (of sidereal
Earth rotates on its axis. If we need to specify the direc- time) to rotate on its axis, so the celestial sphere is divided
tion to an object in a way that is the same for everyone, we into 24 hours of right ascension, with each hour of right
need a coordinate system that is fi xed relative to the celes- ascension corresponding to 15°. Hours of right ascension
tial sphere. The most common such coordinates are called are then subdivided into minutes and seconds of time.
“celestial coordinates.” Right ascension increases going to the east. The right ascen-
Celestial coordinates are illustrated in Figure A6.1. sion of Sirius, for example, is α = 06h 45m 08.9s, meaning
Celestial coordinates are much like the traditional system that Sirius is about 101° (that is, 06h 45m) east of the vernal
of latitude and longitude used on the surface of Earth. On equinox. Time is a natural unit for measuring right ascen-
Earth, latitude specifies how far you are from Earth’s equa- sion because time naturally tracks the motion of objects
tor, as discussed in Chapter 2. If you are on Earth’s equa- due to Earth’s rotation on its axis. If stars on the meridian
tor, your latitude is 0°. If you are at Earth’s North Pole, your at a certain time have α = 06h, then an hour later the stars
latitude is 90° north. If you are at Earth’s South Pole, your on the meridian will have 07h, and an hour after that they
latitude is 90° south. will have 08h. The “local sidereal time,” or “star time,” at
A-16
Appendix A-17
North
celestial
pole
+ 80°
1 The latitude-like
celestial coordinate,
measured in degrees,
is called declination.
+ 60°
A star at this location
has these coordinates:
= 4h
= +40°
on ( )
+ 40°
Declinati
20 h + 20°
h 5h
21 4h
h h
22 3
23h 0h 2h
1h
q uator
0° Celestial e
Ecliptic on ( )
Right ascensi
Vernal
–20° equinox
2 The longitude-like celestial
coordinate, measured in hours,
minutes, and seconds of time,
is called right ascension.
–40°
3 Right ascension and
declination are both zero
FIGURE A6.1 Celestial coordinates. at the vernal equinox.
your location right now is equal to the right ascension of Sirius about 45m east of the meridian, not quite 17° south of
the stars that are on your meridian at the moment. Because the celestial equator.
of Earth’s motion around the Sun, a sidereal day is about There is just one final caveat. As we discussed in Chap-
4 minutes shorter than a solar day, so local sidereal time ter 2, the directions of the celestial equator, celestial poles,
constantly gains on solar time. At midnight on September and vernal equinox are constantly changing as Earth’s axis
23, the local sidereal time is 0h. By midnight on December wobbles like the axis of a spinning top. In Chapter 2 we
22, local sidereal time has advanced to 06h. On March 21, called this 26,000-year wobble the precession of the equi-
local sidereal time at midnight is 12h. And at midnight on noxes, meaning that the location of the equinoxes is slowly
June 21, local sidereal time is 18h. advancing along the ecliptic. So when we specify the celes-
Putting this all together, right ascension and declina- tial coordinates of an object, we need to specify the date
tion provide a convenient way to specify the location of at which the positions of the vernal equinox and celestial
any object on the celestial sphere. Sirius is located at α = poles were measured. By convention, coordinates are usu-
06h 45m 08.9s, or ‒16° 42′ 58″, which means that at midnight ally referred to with the position of the vernal equinox on
on December 22 (local sidereal time = 06h) you will find January 1, 2000. A complete, formal specification of the
A-18 Appendix
coordinates of Sirius would then be α(2000)06h 45m 08.9s, or if you use a star catalog. The magnitude system is “loga-
δ(2000) = ‒16° 42′ 58″, where the “2000” in parentheses refers rithmic.” Other common systems of logarithmic measure-
to the equinox of the coordinates. ments that you may have encountered include decibels for
measuring sound levels, and the Richter scale for measuring
the strength of earthquakes. For example, an earthquake of
magnitude 6 is not just a little stronger than an earthquake
Constellations and Names of magnitude 5; it is, in fact, 10 times stronger.
As discussed in Math Tools 13.1, a difference of five
Although it is certainly possible to exactly specify any magnitudes between the apparent brightness of two stars
location on the surface of Earth by giving its latitude and (say, a star with m = 6 and a star with m = 1), corresponds
longitude, it is usually convenient to use a more descrip- to a 100-fold difference in brightness, and the greater the
tive address. We might say, for example, that one of the magnitude, the fainter the object. If five steps in magnitude
coauthors of this book works near latitude 37° north, lon- correspond to a factor of 100 in brightness, then one step
gitude 122° west; but it would probably mean a lot more in magnitude must correspond to the fifth root of 100—that
to you if we said that George Blumenthal works in Santa is, a factor of 1001/5 = 102/5 = approximately 2.512 in bright-
Cruz, California. ness (1001/5 × 1001/5 × 1001/5 × 1001/5 × 1001/5 = 100). The
Just as the surface of Earth is divided into nations and easiest way to write the relationship between brightness
states, the celestial sphere is divided into 88 constellations, and magnitude is to use common (base-10) logarithms. If
the names of which are often used to refer to objects within star 1 has a brightness of b1 and star 2 has a brightness of
their boundaries (see the star charts in Figure A6.2). We b2, then the difference in magnitude (m2 – m1) between the
refer to the brightest stars within the boundaries of a con- two stars is
stellation using a Greek letter combined with the name of
the constellation. For example, the star Sirius is the bright- b
m2 ‒ m1 = ‒2.5 log10 __2 .
est star in the constellation Canis Major (the “Great Dog”), b1
so it is referred to as “α Canis Majoris.” The bright red star
To convert from magnitude differences to brightness ratios,
in the northeastern corner of the constellation of Orion is
divide by –2.5 (that is, multiply by –0.4) and raise 10 to the
referred to as “α Orionis,” also known as Betelgeuse. Rigel,
resulting power:
the bright blue star in the southwest corner of Orion, is also
called “β Orionis.”
b2
__
Astronomical objects can take on a bewildering range of = 10–0.4 × (m2 – m1).
b1
names. For example, the bright southern star Canopus, also
known as “α Carinae” (the brightest star in the constella- The following table shows some examples using the two
tion of Carina), has no fewer than 34 different names, most preceding equations.
of which are about as memorable as “SAO 234480” (num-
ber 234,480 in the Smithsonian Astrophysical Observatory
catalog of stars). Apparent Magnitude Ratio of Apparent
Difference (m2 – m1) Brightness (b1/b2)
You may have noticed a slight difference in the way
a constellation is spelled when it becomes part of a star’s 1 2.512
name. For example, we see that Sirius is called “α Canis
2 2.5122 = 6.3
Majoris,” not “α Canis Major”; Rigel is referred to as
“β Orionis,” not “β Orion”; and Canopus becomes “α Cari- 3 2.5123 = 15.8
nae,” not “α Carina.” This is because we use the Latin geni-
4 2.5124 = 39.8
tive or possessive case with star names; for example, Orionis
means “of Orion.” 5 2.5125 = 100
10 2.51210 = 1002 = 10,000
15 2.51215 = 1003 = 1,000,000
Astronomical Magnitudes 20 2.51220 = 1004 = 108
25 2.51225 = 1005 = 1010
Apparent Magnitudes
We first introduced magnitudes in Math Tools 13.1; here we
provide some additional information. Throughout the text
we refer to the brightness of objects; but when discussing
Absolute Magnitudes
the appearance of an object in the sky, astronomers normally Recall that stars differ in their brightness for two reasons:
speak instead of the object’s apparent magnitude. You may the amount of light they are actually emitting, and their
come across this system if you take a lab course in astronomy distance from Earth. The magnitude system is also used for
Appendix A-19
luminosity, as well as for brightness with the same scale: a The following table shows how the difference between
difference of five magnitudes corresponds to a 100-fold dif- an object’s apparent and absolute magnitudes solves for its
ference in luminosity. We call these absolute magnitudes distance in parsecs.
(Mabs), and the idea is to imagine how bright the star would
be if it were at a distance of 10 parsecs (pc). Absolute mag-
nitudes allow us to compare how luminous two stars really m – Mabs Distance (pc)
are, without the factor of distance. Our Sun is very bright
because we are so close (apparent visual magnitude = –27); –3 2.5
but if the Sun were at a distance of 10 pc, its magnitude –2 4.0
would be only about 5.1 Thus we say the absolute magni-
tude of the Sun is Mabs = 5. Recall that we usually express –1 6.3
the luminosity of stars by comparing it with the luminosity 0 10
of the Sun. Thus, a star that is 10,000 times more luminous
1 16
than the Sun will be 10 absolute magnitudes brighter, or
Mabs = -5, and a star that is 100 times fainter than the Sun 2 25
will be 5 absolute magnitudes fainter, or Mabs = 10.
3 40
The following is a table of luminosity (where L ⊙= 1) ver-
sus absolute magnitude. 4 63
5 100
L star/L Sun Mabs 10 1,000
1,000,000 –10 15 10,000
10,000 –5 20 100,000
100 0
1 5 Although the system of astronomical magnitudes is con-

1/100 10 venient in many ways—which is why astronomers continue
to use it—it can also be confusing to new students. Just
1/10,000 15
remember three things and you will probably get by:
1. The greater the magnitude, the fainter the object.

Distance Modulus 2. One magnitude smaller means about two and a half times
The difference between the apparent magnitude and the brighter.
absolute magnitude depends on the star’s distance. A star at 3. The brightest stars in the sky have magnitudes of less
a distance of exactly 10 pc will have an apparent magnitude than 1, and the faintest stars that can be seen with the
equal to its absolute magnitude. We can always measure the naked eye on a dark night have magnitudes of about 6.
brightness of a star and thus its apparent magnitude, and
we can estimate the luminosity of a star and thus its abso- A fi nal note: In Chapters 13 and onward, we used “col-
lute magnitude using the H-R diagram. This is the way the ors” based on the ratio of the brightness of a star as seen in
distances to most stars are found. two different parts of the spectrum. The “bB/bV color,” for
Using the preceding equations and the defi nition of example, was just the ratio of the brightness of a star seen
absolute magnitude, we can get to this relatively simple through a blue filter, divided by the brightness of a star seen
expression: through a yellow-green (visual) filter. Normally, astronomers
instead discuss the “B-V color” of a star, which is equal
m - Mabs = 5 log10 d - 5,
to the difference between a star’s blue magnitude and its
where distance d is in parsecs. visual magnitude. We can use the previous expression for
We can rewrite this equation as a magnitude difference to write
d = 10
( m - M5
___________
).
abs + 5
B-V color = mB - mV = -2.5 log10 (bB/bV).
Thus, a star with a bB/b V color of 1.0 has a B-V color of 0.0,
and a star with a bB/b V color of 1.4 has a B-V color of –0.37.
1
The apparent and absolute magnitudes of the Sun are –26.74 and Notice that, as with magnitudes, B-V colors are “backward.”
+4.83, respectively. We use +5 for the Sun’s absolute magnitude as The bluer a star, the greater its bB/bV color but the less its
an approximation. B-V color.
A-20 Appendix
Key to star maps

Constellation
boundaries (blue)
Star names (match
star color)
Constellation
figures, names Meridian at
(yellow) midnight, September 23
h h
60° 4 20 60°
h h
Lines of right 2 h 22
0
ascension and
declination
(green)
40° 40°
20° 20°
0° 0°
–20° –20°
–40° –40°
h
h 0 h
2 22
Magnitude –60° h h –60°
4 20
FIGURE A6. 2 A The sky from right ascension 20h to 04h and declination -60° to +60°.
Appendix A-21
Meridian at
h
midnight, December 22 h
60° 10 2 60°
h h
8 h 4
6
40° 40°
20° 20°
0° 0°
–20° –20°
–40° –40°
h
h 6 h
8 4
10 2
FIGURE A6. 2B The sky from right ascension 02h to 10h and declination -60° to +60°.
A-22 Appendix
Key to star maps

Constellation
boundaries (blue)
Star names (match
star color)
Constellation
figures, names Meridian at
(yellow) midnight, March 21
h h
60° 16 8 60°
h h
Lines of right 14 h 10
12
ascension and
declination
(green)
40° 40°
20° 20°
0° 0°
–20° –20°
–40° –40°
h
h 12 h
14 10
16 8
FIGURE A6. 2C The sky from right ascension 08h to 16h and declination -60° to +60°.
Appendix A-23
Meridian at
h
midnight, June 21 h
60° 22 14 60°
h h
20 h 16
18
40° 40°
20° 20°
0° 0°
–20° –20°
–40° –40°
h
h 18 h
20 16
22 14
FIGURE A6. 2D The sky from right ascension 14h to 22h and declination -60° to +60°.
A-24 Appendix
12 h
h h
10 14
h h
8 16
6h 18 h
4
h 20 h
2
h 22 h
h
0h
0
2h 22 h
4h 20 h
h h
6 18
Magnitude
8h 16h
10 h 14h
12 h
FIGURE A6. 2E The regions of the sky north of declination +40° and south of declination -40°.
NCP, north celestial pole; SCP, south celestial pole.
APPENDIX 7
Uniform Circular Motion

and Circular Orbits
Uniform Circular Motion Remember that the acceleration is always directed toward
the center of the circle, whereas the velocity of the ball is
In Chapter 3 (see Section 3.5 and Figure 3.24) we discuss the always perpendicular to the acceleration. The ball’s veloc-
motion of an object moving in a circle at a constant speed. ity and its acceleration are always at right angles to each
This motion, called uniform circular motion, is the result other. As the object moves around the circle, the direc-
of the fact that centripetal force always acts toward the tion of motion and the direction of the acceleration change
center of the circle. The key question when thinking about together in lockstep.
uniform circular motion is, How hard do I have to pull to Figure A7.1 contains two triangles. Triangle 1 shows the
keep the object moving in a circle? Part of the answer to this velocity (speed and direction) at each of the two times. The
question is pretty obvious: the more massive an object is, arrow labeled “4v” connecting the heads of the two veloc-
the harder it will be to keep it moving on its circular path. ity arrows shows how much the velocity changed between
According to Newton’s second law of motion, F = ma; or in time 1 (t1) and time 2 (t2). This change is the effect of the
this case, the centripetal force equals the mass multiplied by centripetal acceleration. If we imagine that points 1 and 2
the centripetal acceleration. The larger the mass, the greater are very close together—so close that the direction of the
the force required to keep it moving in its circle. centripetal acceleration does not change by much between
The centripetal force needed to keep an object moving in the two—then we can say that the centripetal accelera-
constant circular motion also depends on two other quan- tion equals the change in the velocity divided by the time
tities: the speed of the object and the size of the circle. The between the two, 4t = t2 – t1. So, we have 4v = a4t.
faster an object is moving, the more rapidly it has to change Triangle 2 shows something similar. Here the arrow
direction to stay on a circle of a given size. The second quan- labeled “4r” indicates the change in the position of the
tity that influences the needed acceleration is the radius of ball between time 1 and time 2. Again, if we imagine that
the circle. The smaller the circle, the greater the pull needed the time between the two points is very short, we can say
to keep it on track. You can understand this by looking at that 4r is equal to the velocity multiplied by the time, or
the motion. A small circle requires a continuous “hard” 4r = v4t.
turn, whereas a larger circle requires a more gentle change The line between the center of the circle and the ball is
in direction. It takes more force to keep an object moving always perpendicular to the velocity of the ball. So if the
faster in a smaller circle than it does to keep the same object direction of the ball’s velocity changes by an angle α, then
moving more slowly in a larger circle. (To get a better feel for the direction of the line between the ball and the center of
how this works, think about the difference between riding the circle must also change by the same angle α. In other
in a car that is taking a tight curve at high speed and a car words, triangles 1 and 2 are “similar triangles.” They have
that is moving slowly around a gentle curve.) the same shape. If the triangles are the same shape, the
To arrive at the circular velocity and other results dis- ratio of two sides of triangle 1 must equal the ratio of the
cussed in Chapter 3, we need to turn these intuitive ideas two corresponding sides of triangle 2. Using this fact, we
about uniform circular motion into a quantitative expression can write
of exactly how much centripetal acceleration is needed to
a4t ____
____ v4t
keep an object moving in a circle with radius r at speed v. = .
v r
Figure A7.1 shows a ball moving around a circle of radius
r at a constant speed v at two different times. The centrip- If we divide by 4t on both sides of the equation and then
etal acceleration that is keeping the ball on the circle is a. cross-multiply, we obtain
A-25
A-26 Appendix
FIGURE A7.1 Similar triangles are

v 1 An object moving on a circular path used to find the centripetal force that
with radius r at constant speed v is needed to keep an object moving at
experiences a centripetal acceleration a. a constant speed on a circular path.
a
v
a
α 2 In a time Δt, the direction

r of the object’s velocity…
r
v Triangle 1
Δr = v Δ t α
v
Δv = a Δ t
3 …and the object’s

α location move through
r
r the same angle α, forming
two similar triangles.
Triangle 2
Mm
ar = v2, Fgravity = G ____ .
r2
which, after we divide both sides of the equation by r, If gravity is responsible for holding the mass in its circular
becomes motion, then it should be true that Fgravity = Fcentripetal. That
is, if mass m is moving in a circle under the force of grav-
v2
acentripetal = __ ity, the force provided by gravity must equal the centrip-
r.
etal force needed to explain that circular motion. Setting
We have added the subscript “centripetal” to a to signify our two expressions for Fcentripetal and Fgravity equal to each
that this is the centripetal acceleration needed to keep the other gives
object moving in a circle of radius r at speed v. The centrip-
mv 2
____ Mm
etal force required to keep an object of mass m moving on = G ____
2 .
r r
such a circle is then
All that remains is a bit of algebra. Dividing by m on both
mv 2
Fcentripetal = macentripetal = ____ sides of the equation and multiplying both sides by r gives
r.
M
v 2 = G __.
r
Circular Orbits Taking the square root of both sides then brings us to the
desired result:
In the case of an object moving in a circular orbit, there is ____
no string to hold the ball on its circular path. Instead, this
force is provided by gravity.
vcircular = √ GMr .
___
Think about an object with mass m in orbit about a much This is the circular velocity we presented in Chapter 3. It
larger object with mass M. The orbit is circular, and the dis- is the velocity at which an object in a circular orbit must
tance between the two objects is given by r. The force needed be moving. If the object were not moving at this velocity,
to keep the smaller object moving at speed v in a circle with then gravity would not be providing the needed centripetal
radius r is given by the previous expression for Fcentripetal. The force, and the object would not move in a circle.
force actually provided by gravity (see Chapter 3) is
APPENDIX 8*
IAU 2006 Resolutions:

Definition of a Planet in the
Solar System, and Pluto
August 24, 2006, Prague (1) A “planet”1 is a celestial body that (a) is in orbit around
the Sun, (b) has sufficient mass for its self-gravity to over-
come rigid body forces so that it assumes a hydrostatic
equilibrium (nearly round) shape, and (c) has cleared
Resolutions the neighborhood around its orbit.
(2) A “dwarf planet” is a celestial body that (a) is in orbit
Resolution 5 is the principal defi nition for the IAU usage of around the Sun, (b) has sufficient mass for its self-gravity
“planet” and related terms. to overcome rigid body forces so that it assumes a hydro-
Resolution 6 creates for IAU usage a new class of objects, static equilibrium (nearly round) shape,2 (c) has not
for which Pluto is the prototype. The IAU will set up a pro- cleared the neighborhood around its orbit, and (d) is not
cess to name these objects. a satellite.
(3) All other objects,3 except satellites, orbiting the Sun
shall be referred to collectively as “Small Solar-System
Bodies.”
Resolution 5
Definition of a Planet in Resolution 6
the Solar System
Contemporary observations are changing our understanding Pluto
of planetary systems, and it is important that our nomen-
The IAU further resolves:
clature for objects reflect our current understanding. This
Pluto is a “dwarf planet” by the above defi nition and
applies, in particular, to the designation “planets.” The
is recognized as the prototype of a new category of Trans-
word “planet” originally described “wanderers” that were
Neptunian Objects.4
known only as moving lights in the sky. Recent discover-
ies lead us to create a new defi nition, which we can make
using currently available scientific information. 1
The eight planets are: Mercury, Venus, Earth, Mars, Jupiter, Saturn,
The IAU therefore resolves that planets and other bodies, Uranus, and Neptune.
2
except satellites, in our Solar System be defi ned into three An IAU process will be established to assign borderline objects into
distinct categories in the following way: either dwarf planet and other categories.
3
These currently include most of the Solar System asteroids, most
Trans-Neptunian Objects (TNOs), comets, and other small bodies.
4
An IAU process will be established to select a name for this
*Source: International Astronomical Union. category.
A-27
A-27
A-28
Appendix
M
er
cu
ry
Ve
nu
s
Ea
rth
Ce M
re
s ar
s
Ju
p ite
r
Sa
tu
rn
Ha
um
ea
M Ur
ak an
em u s
ak
e
Ne
pt
Pl un
ut e
o
Er
is
Dwarf
Planets
Planets
Glossary
AGN See active galactic nucleus. aperture The clear diameter of a telescope’s
A albedo The fraction of electromagnetic objective lens or primary mirror.
aberration of starlight The apparent dis- radiation incident on a surface that is aphelion (pl. aphelia) The point in a solar
placement in the position of a star that is reflected by the surface. orbit that is farthest from the Sun. Com-
due to the fi nite speed of light and Earth’s algebra A branch of mathematics in which pare perihelion.
orbital motion around the Sun. numeric variables are represented by Apollos A group of asteroids whose orbits
absolute magnitude A measure of the letters. cross the orbits of both Earth and Mars.
intrinsic brightness of a celestial object, alpha particle A 4He nucleus, consist- Compare Amors and Atens.
generally a star. Specifically, the appar- ing of two protons and two neutrons. apparent magnitude A measure of the
ent brightness of an object, such as a star, Alpha particles get their name from the apparent brightness of a celestial object,
if it were located at a standard distance fact that they are given off in the type of generally a star. Compare absolute
of 10 parsecs (pc). Compare apparent radioactive decay referred to as “alpha magnitude.
magnitude. decay.” arcminute (arcmin) A minute of arc ('), a
absolute zero The temperature at which Amors A group of asteroids whose orbits unit used for measuring angles. An arc-
thermal motions cease. The lowest pos- cross the orbit of Mars but not the orbit of minute is 1 ⁄ 60 of a degree of arc.
sible temperature. Zero on the Kelvin Earth. Compare Apollos and Atens. arcsecond (arcsec) A second of arc ("),
temperature scale. amplitude In a wave, the maximum devia- a unit used for measuring very small
absorption The capture of electromagnetic tion from its undisturbed or relaxed posi- angles. An arcsecond is 1 ⁄ 60 of an arc-
radiation by matter. Compare emission. tion. For example, in a water wave the minute, or 1/3,600 of a degree of arc.
absorption line An intensity minimum in amplitude is the vertical distance from Arctic Circle The circle on Earth with lati-
a spectrum that is due to the absorption crests to the undisturbed water level. tude 66.5 north, marking the southern limit
of electromagnetic radiation at a specific angular momentum A conserved property where at least 1 day per year is in 24-hour
wavelength determined by the energy of a rotating or revolving system whose daylight. Compare Antarctic Circle.
levels of an atom or molecule. Compare value depends on the velocity and distri- asteroid Also called minor planet. A
emission line. bution of the system’s mass. primitive rocky or metallic body (plan-
acceleration The rate at which the speed annular solar eclipse The type of solar etesimal) that has survived planetary
and/or direction of an object’s motion is eclipse that occurs when the apparent accretion. Asteroids are parent bodies of
changing. diameter of the Moon is less than that meteoroids.
accretion disk A flat, rotating disk of gas of the Sun, leaving a visible ring of light asteroid belt Also called main asteroid
and dust surrounding an object, such as (“annulus”) surrounding the dark disk of belt. The region between the orbits of
a young stellar object, a forming planet, the Moon. Compare partial solar eclipse Mars and Jupiter that contains most of the
a collapsed star in a binary system, or a and total solar eclipse. asteroids in our Solar System.
black hole. Antarctic Circle The circle on Earth with astrobiology An interdisciplinary science
achondrite A stony meteorite that does not latitude 66.5° south, marking the north- combining astronomy, biology, chemistry,
contain chondrules. Compare chondrite. ern limit where at least 1 day per year geology, and physics to study life in the
active comet A comet nucleus that is in 24-hour daylight. Compare Arctic cosmos.
approaches close enough to the Sun to Circle. astrology The belief that the positions and
show signs of activity, such as the pro- anticyclonic motion The rotation of a aspects of stars and planets influence
duction of a coma and tail. weather system resulting from the Corio- human affairs and characteristics, as well
active galactic nucleus (AGN) A highly lis effect as air moves outward from a as terrestrial events.
luminous, compact galactic nucleus region of high atmospheric pressure. astronomical seeing A measurement of
whose luminosity may exceed that of the Compare cyclonic motion. the degree to which Earth’s atmosphere
rest of the galaxy. antimatter Matter made from antiparticles. degrades the resolution of a telescope’s
adaptive optics Electro-optical systems antiparticle An elementary particle of anti- view of astronomical objects.
that largely compensate for image distor- matter identical in mass but opposite in astronomical unit (AU) The average dis-
tion caused by Earth’s atmosphere. charge and all other properties to its cor- tance from the Sun to Earth: approxi-
AGB See asymptotic giant branch. responding ordinary matter particle. mately 150 million kilometers (km).
G-1
G-2 Glossary
astronomy The scientific study of plan- bar A unit of pressure. One bar is boundary between the region dominated
ets, stars, galaxies, and the universe as a equivalent to 105 newtons per square by the solar wind and the region domi-
whole. meter—approximately equal to Earth’s nated by a planet’s magnetosphere.
astrophysics The application of physi- atmospheric pressure at sea level. 2. The interface between strong colli-
cal laws to the understanding of plan- barred spiral A spiral galaxy with a bulge mated gas and dust outflow from a star
ets, stars, galaxies, and the universe as a having an elongated, barlike shape. and the interstellar medium.
whole. basalt Gray to black volcanic rock, rich in brightness The apparent intensity of
asymptotic giant branch (AGB) The path on iron and magnesium. light from a luminous object. Bright-
the H-R diagram that goes from the hori- beta decay 1. The decay of a neutron into ness depends on both the luminosity of
zontal branch toward higher luminosities a proton by emission of an electron (beta a source and its distance. Units at the
and lower temperatures, asymptotically ray) and an antineutrino. 2. The decay of detector: watts per square meter (W/m2).
approaching and then rising above the a proton into a neutron by emission of a brown dwarf A “failed” star that is not
red giant branch. positron and a neutrino. massive enough to cause hydrogen fusion
Atens A group of asteroids whose orbits Big Bang The event that occurred 13.7 bil- in its core. An object whose mass is inter-
cross the orbit of Earth but not the orbit of lion years ago that marks the beginning of mediate between that of the least massive
Mars. Compare Amors and Apollos. time and the universe. stars and that of supermassive planets.
atmosphere The gravitationally bound, Big Bang nucleosynthesis The formation of bulge The central region of a spiral galaxy
outer gaseous envelope surrounding a low-mass nuclei (H, He, Li, Be, B) during that is similar in appearance to a small
planet, moon, or star. the fi rst few minutes following the Big elliptical galaxy.
atmospheric greenhouse effect A warm- Bang.
ing of planetary surfaces produced by Big Rip A hypothetical cosmic event in
atmospheric gases that transmit optical which all matter in the universe, from C
solar radiation but partially trap infrared stars to subatomic particles, is progres- C See Celsius.
radiation. Compare greenhouse effect. sively torn apart by expansion of the C-type asteroid An asteroid made of mate-
atmospheric probe An instrumented pack- universe. rial that has largely been unmodified
age designed to provide in situ measure- binary star A system in which two stars since the formation of the Solar Sys-
ments of the chemical and/or physical are in gravitationally bound orbits about tem; the most primitive type of asteroid.
properties of a planetary atmosphere. their common center of mass. Compare M-type asteroid and S-type
atmospheric window A region of the elec- binding energy The minimum energy asteroid.
tromagnetic spectrum in which radiation required to separate an atomic nucleus carbon-nitrogen-oxygen (CNO) cycle One
is able to penetrate a planet’s atmosphere. into its component protons and neutrons. of the ways in which hydrogen is con-
atom The smallest piece of a chemical biosphere The global sum of all liv- verted to helium (hydrogen burning) in
element that retains the properties of ing organisms on Earth (or any planet the interiors of main-sequence stars. See
that element. Each atom is composed of or moon.) Compare hydrosphere and also triple-alpha process and proton-
a nucleus (neutrons and protons) sur- lithosphere. proton chain.
rounded by a cloud of electrons. bipolar outflow Material streaming away in carbon star A cool red giant or asymptotic
AU See astronomical unit. opposite directions from either side of the giant branch star that has an excess of
aurora Emission in the upper atmosphere accretion disk of a young star. carbon in its atmosphere.
of a planet from atoms that have been black hole An object so dense that its carbonaceous chondrite A primitive stony
excited by collisions with energetic par- escape velocity exceeds the speed of meteorite that contains chondrules and is
ticles from the planet’s magnetosphere. light; a singularity in spacetime. rich in carbon and volatile materials.
autumnal equinox 1. One of two points blackbody An object that absorbs and Cassini Division The largest gap in Saturn’s
where the Sun crosses the celestial equa- can reemit all electromagnetic energy it rings, discovered by Jean-Dominique
tor. 2. The day on which the Sun appears receives. Cassini in 1675.
at this location, marking the fi rst day blackbody spectrum See Planck spectrum. catalyst An atomic and molecular struc-
of autumn (about September 23 in the blueshift The Doppler shift toward shorter ture that permits or encourages chemi-
Northern Hemisphere and March 21 in wavelengths of light from an approaching cal and nuclear reactions but does not
the Southern Hemisphere). Compare object. Compare redshift. change its own chemical or nuclear
vernal equinox. Bohr model A model of the atom, proposed properties.
axion A hypothetical elementary particle by Niels Bohr in 1913, in which a small CBR See cosmic background radiation.
fi rst proposed to explain certain proper- positively charged nucleus is surrounded CCD See charge-coupled device.
ties of the neutron and now considered a by orbiting electrons, similar to a minia- celestial equator The imaginary great circle
candidate for cold dark matter. ture solar system. that is the projection of Earth’s equator
bolide A very bright, exploding meteor. onto the celestial sphere.
bound orbit A closed orbit in which the celestial sphere An imaginary sphere with
B velocity is less than the escape velocity. celestial objects on its inner surface and
backlighting Illumination from behind a Compare unbound orbit. Earth at its center. The celestial sphere
subject as seen by an observer. Fine mate- bow shock 1. The boundary at which has no physical existence but is a conve-
rial such as human hair and dust in plan- the speed of the solar wind abruptly nient tool for picturing the directions in
etary rings stands out best when viewed drops from supersonic to subsonic in its which celestial objects are seen from the
under backlighting conditions. approach to a planet’s magnetosphere; the surface of Earth.
Glossary G-3
Celsius (C) Also called centigrade scale. ets of the Solar System: Mercury, Venus, isolated, closed system does not change
The arbitrary temperature scale—defi ned Earth, Mars, Jupiter, Saturn, Uranus, and over time.
by Anders Celsius (1701–1744)—that Neptune. constant of proportionality The multipli-
defi nes 0°C as the freezing point of water climate The state of an atmosphere aver- cative factor by which one quantity is
and 100°C as the boiling point of water at aged over an extended time. related to another.
sea level. Unit: °C. Compare Fahrenheit closed universe A fi nite universe with a constellation An imaginary image formed
and Kelvin scale. curved spatial structure such that the by patterns of stars; any of 88 defi ned
center of mass The location associated with sum of the angles of a triangle always areas on the celestial sphere used by
an object system at which we may regard exceeds 180°. Compare flat universe and astronomers to locate celestial objects.
the entire mass of the system as being open universe. constructive interference A state in which
concentrated. The point in any isolated CNO cycle See carbon-nitrogen-oxygen the amplitudes of two intersecting waves
system that moves according to Newton’s cycle. reinforce one another. Compare destruc-
fi rst law of motion. cold dark matter Particles of dark matter tive interference.
centigrade scale See Celsius. that move slowly enough to be gravita- continental drift The slow motion (centi-
centripetal force A force directed toward tionally bound even in the smallest gal- meters per year) of Earth’s continents rel-
the center of curvature of an object’s axies. Compare hot dark matter. ative to each other and to Earth’s mantle.
curved path. color index The color of a celestial object, See also plate tectonics.
Cepheid variable An evolved high-mass generally a star, based on the ratio of its continuous radiation Electromagnetic radi-
star with an atmosphere that is pulsating, brightness in blue light (b B) to its bright- ation with intensity that varies smoothly
leading to variability in the star’s lumi- ness in “visual” (or yellow-green) light over a wide range of wavelengths.
nosity and color. (b V). The difference between an object’s convection The transport of thermal energy
Chandrasekhar limit The upper limit on blue (B) magnitude and visual (V) magni- from the lower (hotter) to the higher
the mass of an object supported by elec- tude, B-V. (cooler) layers of a fluid by motions
tron degeneracy pressure; approximately coma (pl. comae) The nearly spherical within the fluid driven by variations in
1.4 solar mass (M⊙). cloud of gas and dust surrounding the buoyancy.
chaos Behavior in complex, interrelated nucleus of an active comet. convective zone A region within a star in
systems in which tiny differences in the comet A complex object consisting of a which energy is transported outward by
initial configuration of a system result in small, solid, icy nucleus; an atmospheric convection.
dramatic differences in the system’s later halo; and a tail of gas and dust. core 1. The innermost region of a planetary
evolution. comet nucleus A primitive planetesimal, interior. 2. The innermost part of a star.
charge-coupled device (CCD) A common composed of ices and refractory materi- core accretion A process for forming giant
type of solid-state detector of electro- als, that has survived planetary accre- planets, whereby large quantities of sur-
magnetic radiation that transforms the tion. The “heart” of a comet, containing rounding hydrogen and helium are gravi-
intensity of light directly into electric nearly the entire mass of the comet. A tationally captured onto a massive rocky
signals. “dirty snowball.” core.
chemistry The study of the composition, comparative planetology The study of plan- Coriolis effect The apparent displacement
structure, and properties of substances. ets through comparison of their chemical of objects in a direction perpendicular
chondrite A stony meteorite containing and physical properties. to their true motion as viewed from a
chondrules. Compare achondrite. complex system An interrelated system rotating frame of reference. On a rotating
chondrule A small, crystallized, spherical capable of exhibiting chaotic behavior. planet, different latitudes rotating at dif-
inclusion of rapidly cooled molten drop- See also chaos. ferent speeds cause this effect.
lets found inside some meteorites. composite volcano A large, cone-shaped corona The hot, outermost part of the Sun’s
chromatic aberration A detrimental prop- volcano formed by viscous, pasty lava atmosphere.
erty of a lens in which rays of different flows alternating with pyroclastic (explo- coronal hole A low-density region in the
wavelengths are brought to different focal sively generated) rock deposits. Compare solar corona containing “open” magnetic
distances from the lens. shield volcano. field lines along which coronal mate-
chromosphere The region in the Sun’s compound lens A lens made up of two or rial is free to stream into interplanetary
atmosphere located between the photo- more elements of differing refractive space.
sphere and the corona. index, the purpose of which is to mini- coronal mass ejection An eruption on the
circular velocity The orbital velocity mize chromatic aberration. Sun that ejects hot gas and energetic
needed to keep an object moving in a cir- conservation law A physical law stating particles at much higher speeds than are
cular orbit. that the amount of a particular physi- typical in the solar wind.
circumpolar Referring to the part of the cal quantity (such as energy or angular cosmic background radiation (CBR) Iso-
sky, near either celestial pole, that can momentum) of an isolated system does tropic microwave radiation from every
always be seen above the horizon from a not change over time. direction in the sky having a 2.73-kelvin
specific location on Earth. conservation of angular momentum The (K) Planck spectrum. The CBR is residual
circumstellar disk See protoplanetary disk. physical law stating that the amount of radiation from the Big Bang.
classical mechanics The science of apply- angular momentum of an isolated system cosmic ray A very fast-moving particle
ing Newton’s laws to the motion of does not change over time. (usually an atomic nucleus); cosmic rays
objects. conservation of energy The physical law fi ll the disk of our galaxy.
classical planets The eight major plan- stating that the amount of energy of an cosmological constant A constant, intro-
G-4 Glossary
duced into general relativity by Einstein, dark matter halo The centrally condensed, dwarf planet A body with characteris-
that characterizes an extra, repulsive greatly extended dark matter component tics similar to those of a classical planet
force in the universe due to the vacuum of a galaxy that contains up to 95 percent except that it has not cleared smaller
of space itself. of the galaxy’s mass. bodies from the neighboring regions
cosmological principle The (testable) daughter product An element resulting around its orbit. Compare planet (defi ni-
assumption that the same physical laws from radioactive decay of a more massive tion 2).
that apply here and now also apply every- parent element. dynamic equilibrium A state in which a
where and at all times, and that there are decay 1. The process of a radioactive system is constantly changing but its con-
no special locations or directions in the nucleus changing into its daughter prod- figuration remains the same because one
universe. uct. 2. The process of an atom or molecule source of change is exactly balanced by
cosmological redshift (z) The redshift dropping from a higher-energy state to a another source of change. Compare static
that results from the expansion of the lower-energy state. equilibrium.
universe rather than from the motions density The measure of an object’s mass dynamo A device that converts mechanical
of galaxies or gravity (see gravitational per unit of volume. Units: kilograms per energy into electric energy in the form of
redshift). cubic meter (kg/m3). electric currents and magnetic fields. The
cosmology The study of the large-scale destructive interference A state in which “dynamo effect” is thought to create mag-
structure and evolution of the universe as the amplitudes of two intersecting waves netic fields in planets and stars by electri-
a whole. cancel one another. Compare constructive cally charged currents of material flowing
Crab Nebula The remnant of the Type II interference. within their cores.
supernova explosion witnessed by Chi- differential rotation Rotation of different
nese astronomers in 1054 C.E. parts of a system at different rates.
crescent Any phase of the Moon, Mercury, differentiation The process by which E
or Venus in which the object appears less materials of higher density sink toward eccentricity (e) A measure of the departure
than half illuminated by the Sun. Com- the center of a molten or fluid planetary of an ellipse from circularity; the ratio of
pare gibbous. interior. the distance between the two foci of an
Cretaceous-Tertiary (K-T) boundary The diffraction The spreading of a wave after ellipse to its major axis.
boundary between the Cretaceous and it passes through an opening or past the eclipse 1. The total or partial obscuration of
Tertiary periods in Earth’s history. This edge of an object. one celestial body by another. 2. The total
boundary corresponds to the time of the diffraction limit The limit of a telescope’s or partial obscuration of light from one
impact of an asteroid or comet and the angular resolution caused by diffraction. celestial body as it passes through the
extinction of the dinosaurs. diffuse ring A sparsely populated plan- shadow of another celestial body.
critical density The value of the mass den- etary ring spread out both horizontally eclipse season Any time during the year
sity of the universe that, ignoring any cos- and vertically. when the Moon's line of nodes is suf-
mological constant, is just barely capable dispersion The separation of rays of light ficiently close to the Sun for eclipses to
of halting expansion of the universe. into their component wavelengths. occur.
crust The relatively thin, outermost, hard distance ladder A sequence of techniques eclipsing binary A binary system in
layer of a planet, which is chemically dis- for measuring cosmic distances; each which the orbital plane is oriented
tinct from the interior. method is calibrated using the results such that the two stars appear to pass
cryovolcanism Low-temperature volcanism from other methods that have been in front of one another as seen from
in which the magmas are composed of applied to closer objects. Earth. Compare visual binary and spec-
molten ices rather than rocky material. Doppler effect The change in wavelength of troscopic binary.
cyclone See hurricane. sound or light that is due to the relative ecliptic 1. The apparent annual path of the
cyclonic motion The rotation of a weather motion of the source toward or away from Sun against the background of stars. 2.
system resulting from the Coriolis effect the observer. The projection of Earth’s orbital plane
as air moves toward a region of low atmo- Doppler redshift See redshift. onto the celestial sphere.
spheric pressure. Compare anticyclonic Doppler shift The amount by which the ecliptic plane The plane of Earth’s orbit
motion. wavelength of light is shifted by the Dop- around the Sun. The ecliptic is the pro-
Cygnus X-1 A binary X-ray source and pler effect. jection of this plane on the celestial
probable black hole. Drake equation A prescription for estimat- sphere.
ing the number of intelligent civilizations effective temperature The temperature
existing elsewhere. at which a black body, such as a star,
D dust devil A small tornado-like column of appears to radiate.
dark energy A form of energy that perme- air containing dust or sand. ejecta 1. Material thrown outward by the
ates all of space (including the vacuum) dust tail A type of comet tail consisting of impact of an asteroid or comet on a plan-
producing a repulsive force that acceler- dust particles that are pushed away from etary surface, leaving a crater behind.
ates the expansion of the universe. the comet’s head by radiation pressure 2. Material thrown outward by a stellar
dark matter Matter in galaxies that does from the Sun. Compare ion tail. explosion.
not emit or absorb electromagnetic radia- dwarf galaxy A small galaxy with a lumi- electric field A field that is able to exert a
tion. Dark matter is thought to comprise nosity ranging from 1 million to 1 billion force on a charged object, whether at rest
most of the mass in the universe. Com- solar luminosities (L ⊙). Compare giant or moving. Compare magnetic field.
pare luminous matter. galaxy. electric force The force exerted on a
Glossary G-5
charged particle by an electric field. Com- trum that is due to sharply defi ned emis- black hole. Nothing inside this surface—
pare magnetic force. sion of electromagnetic radiation in a not even light—can escape from a black
electromagnetic force The force, including narrow range of wavelengths. Compare hole.
both electric and magnetic forces, that absorption line. evolutionary track The path that a star fol-
acts on electrically charged particles. empirical science Scientific investigation lows across the H-R diagram as it evolves
One of four fundamental forces of nature. that is based primarily on observations through its lifetime.
The force mediated by photons. and experimental data. It is descrip- excited state An energy level of a particular
electromagnetic radiation A traveling dis- tive rather than based on theoretical atom, molecule, or particle that is higher
turbance in the electric and magnetic inference. than its ground state. Compare ground
fields caused by accelerating electric energy The conserved quantity that gives state.
charges. In quantum mechanics, a stream objects and systems the ability to do exoplanet See extrasolar planet.
of photons. Light. work. Units: joules (J). extrasolar planet Also called exoplanet. A
electromagnetic spectrum The spectrum energy transport The transfer of energy planet orbiting a star other than the Sun.
made up of all possible frequencies or from one location to another. In stars,
wavelengths of electromagnetic radiation, energy transport is carried out mainly by
ranging from gamma rays through radio radiation or convection. F
waves and including the portion our eyes entropy A measure of the disorder of a F See Fahrenheit.
can use. system related to the number of ways Fahrenheit (F) The arbitrary temperature
electromagnetic wave A wave consisting of a system can be rearranged without its scale—defi ned by Daniel Gabriel Fahren-
oscillations in the electric-field strength appearance being affected. heit (1686–1736)—that defi nes 32°F as the
and the magnetic-field strength. equator The imaginary great circle on the melting point of water and 212°F as the
electron (e –) A subatomic particle having a surface of a body midway between its boiling point of water at sea level. Unit:
negative charge of 1.6 × 10 –19 coulomb (C), poles that divides the body into northern °F. Compare Celsius and Kelvin scale.
a rest mass of 9.1 × 10 –31 kilograms (kg), and southern hemispheres. The equato- fault A fracture in the crust of a planet or
and mass-equivalent energy of 8 × 10 –14 rial plane passes through the center of the moon along which blocks of material can
joules (J). The antiparticle of the positron. body and is perpendicular to its rotation slide.
Compare proton and neutron. axis. Compare meridian. filter An instrument element that transmits
electron-degenerate Describing the state equilibrium The state of an object in which a limited wavelength range of electro-
of material compressed to the point physical processes balance each other so magnetic radiation. For the optical range,
at which electron density reaches the that its properties or conditions remain such elements are typically made of dif-
limit imposed by the rules of quantum constant. ferent kinds of glass and take on the hue
mechanics. equinox Literally, “equal night.” 1. One of the light they transmit.
electroweak theory The quantum the- of two positions on the ecliptic where it fi rst quarter Moon The phase of the Moon
ory that combines descriptions of both intersects the celestial equator. 2. Either in which only the western half of the
the electromagnetic force and the weak of the two times of year (the autumnal Moon, as viewed from Earth, is illumi-
nuclear force. equinox and vernal equinox) when the nated by the Sun. It occurs about a week
element One of 92 naturally occurring sub- Sun is at one of these two positions. At after a new Moon. Compare third quarter
stances (such as hydrogen, oxygen, and this time, night and day are of the same Moon.
uranium) and more than 20 human-made length everywhere on Earth. Compare fissure A fracture in the planetary litho-
ones (such as plutonium). Each element is solstice. sphere from which magma emerges.
chemically defi ned by the specific num- equivalence principle The principle stat- flat rotation curve A rotation curve of a spi-
ber of protons in the nuclei of its atoms. ing that there is no difference between a ral galaxy in which rotation rates do not
elementary particle One of the basic build- frame of reference that is freely floating decline in the outer part of the galaxy, but
ing blocks of nature that is not known to through space and one that is freely fall- remain relatively constant to the outer-
have substructure, such as the electron ing within a gravitational field. most points.
and the quark. erosion The degradation of a planet’s sur- flat universe An infi nite universe whose
ellipse A conic section produced by the face topography by the mechanical action spatial structure obeys Euclidean geom-
intersection of a plane with a cone when of wind and/or water. etry, such that the sum of the angles of
the plane is passed through the cone at an escape velocity The minimum velocity a triangle always equals 180°. Compare
angle to the axis other than 0° or 90°. needed for an object to achieve a para- closed universe and open universe.
elliptical galaxy A galaxy of Hubble type bolic trajectory and thus permanently flatness problem The surprising result
“E” class, with a circular to elliptical out- leave the gravitational grasp of another that the sum of Ωmass plus ΩΛ is extremely
line on the sky containing almost no disk mass. close to unity in the present-day universe;
and a population of old stars. Compare eternal inflation The idea that a universe equivalent to saying that it is surprising
spiral galaxy, S0 galaxy, and irregular might inflate forever. In such a universe, the universe is so close to being exactly
galaxy. quantum effects could randomly cause flat.
emission The release of electromagnetic regions to slow their expansion, even- flux The total amount of energy passing
energy when an atom, molecule, or par- tually stop inflating, and experience an through each square meter of a surface
ticle drops from a higher-energy state to a explosion resembling our Big Bang. each second. Units: watts per square
lower-energy state. Compare absorption. event A particular location in spacetime. meter (W/m2).
emission line An intensity peak in a spec- event horizon The effective “surface” of a flux tube A strong magnetic field contained
G-6 Glossary
within a tubelike structure. Flux tubes slowly in a gravitational field than in the gravitationally focuses the light of a
are found in the solar atmosphere and absence of a gravitational field. Compare more distant object to produce multiple
connecting the space between Jupiter and time dilation. brighter, magnified, possibly distorted
its moon Io. general relativity See general theory of images.
flyby A spacecraft that fi rst approaches relativity. gravitational lensing The bending of light
and then continues flying past a planet or general theory of relativity Sometimes by gravity.
moon. Flybys can visit multiple objects, referred to as simply general relativ- gravitational potential energy The stored
but they remain in the vicinity of their ity. Einstein’s theory explaining gravity energy in an object that is due solely to its
targets only briefly. Compare orbiter. as the distortion of spacetime by mas- position within a gravitational field.
focal length The optical distance between a sive objects, such that particles travel gravitational redshift The shifting to lon-
telescope’s objective lens or primary mir- on the shortest path between two events ger wavelengths of radiation from an
ror and the plane (called the focal plane) in spacetime. This theory deals with all object deep within a gravitational well.
on which the light from a distant object types of motion. Compare special theory gravitational wave A wave in the fabric
is focused. of relativity. of spacetime emitted by accelerating
focal plane The plane, perpendicular to the geocentric A coordinate system having the masses.
optical axis of a lens or mirror, in which center of Earth as its center. Compare gravity 1. The mutually attractive force
an image is formed. heliocentric. between massive objects. 2. An effect
focus (pl. foci) 1. One of two points that geodesic The path an object will follow arising from the bending of spacetime by
defi ne an ellipse. 2. A point in the focal through spacetime in the absence of massive objects. 3. One of four fundamen-
plane of a telescope. external forces. tal forces of nature.
frame of reference A coordinate system geometry A branch of mathematics that great circle Any circle on a sphere that has
within which an observer measures posi- deals with points, lines, angles, and as its center the center of the sphere. The
tions and motions. shapes. celestial equator, the meridian, and the
free fall The motion of an object when the giant galaxy A galaxy with luminosity ecliptic are all great circles on the sphere
only force acting on it is gravity. greater than about 1 billion solar lumi- of the sky, as is any circle drawn through
frequency The number of times per second nosities (L ⊙). Compare dwarf galaxy. the zenith.
that a periodic process occurs. Unit: hertz giant molecular cloud An interstellar cloud Great Red Spot The giant, oval, brick-red
(Hz), 1/s. composed primarily of molecular gas and anticyclone seen in Jupiter’s southern
full Moon The phase of the Moon in which dust, having hundreds of thousands of hemisphere.
the near side of the Moon, as viewed from solar masses. greenhouse effect The solar heating of air
Earth, is fully illuminated by the Sun. It giant planet One of the largest planets in in an enclosed space, such as a closed
occurs about two weeks after a new Moon. the Solar System (Saturn, Jupiter, Uranus, building or car, resulting primar-
or Neptune), typically 10 times the size ily from the inability of the hot air to
and many times the mass of any terres- escape. Compare atmospheric green-
G trial planet and lacking a solid surface. house effect.
galaxy A gravitationally bound system that gibbous Any phase of the Moon, Mercury, greenhouse molecule One of a group of
consists of stars and star clusters, gas, or Venus in which the object appears atmospheric molecules such as carbon
dust, and dark matter; typically greater more than half illuminated by the Sun. dioxide that are transparent to visible
than 1,000 light-years across and recog- Compare crescent. radiation but absorb infrared radiation.
nizable as a discrete, single object. global circulation The overall, planet- Gregorian calendar The modern calendar.
galaxy cluster A large, gravitationally wide circulation pattern of a planet’s A modification of the Julian calendar
bound collection of galaxies containing atmosphere. decreed by Pope Gregory XIII in 1582. By
hundreds to thousands of members; typi- globular cluster A spherically symmetric, this time the less accurate Julian calendar
cally 10–15 mega-light-years (Mly) across. highly condensed group of stars, contain- had developed an error of 10 days over
Compare galaxy group and supercluster. ing tens of thousands to a million mem- the 13 centuries since its inception.
galaxy group A small, gravitationally bers. Compare open cluster. ground state The lowest possible energy
bound collection of galaxies containing gluon The particle that carries (or, equiva- state for a system or part of a system, such
from several to a hundred members; typi- lently, mediates) interactions due to the as an atom, molecule, or particle. Com-
cally 4–6 mega-light-years (Mly) across. strong nuclear force. pare excited state.
Compare galaxy cluster and supercluster. grand unified theory (GUT) A unified GUT See grand unified theory.
gamma ray Electromagnetic radiation with quantum theory that combines the strong
higher frequency, higher photon energy, nuclear, weak nuclear, and electromag-
and shorter wavelength than all other netic forces but does not include gravity. H
types of electromagnetic radiation. granite Rock that is cooled from magma H II region A region of interstellar gas that
gas giant A giant planet formed mostly of and is relatively rich in silicon and has been ionized by UV radiation from
hydrogen and helium. In our Solar Sys- oxygen. nearby hot massive stars.
tem, Jupiter and Saturn are the gas giants. grating An optical surface containing many H-R diagram The Hertzsprung-Russell dia-
Compare ice giant. narrow, closely and equally spaced par- gram, which is a plot of the luminosities
gauss A basic unit of magnetic flux density. allel grooves or slits that spectrally dis- versus the surface temperatures of stars.
general relativistic time dilation The veri- perse reflected or transmitted light. The evolving properties of stars are plot-
fied prediction that time passes more gravitational lens A massive object that ted as tracks across the H-R diagram.
Glossary G-7
habitable zone The distance from its star at Hertzsprung-Russell diagram See H-R in excess of 300 kilometers per hour
which a planet must be located in order to diagram. (km/h).
have a temperature suitable for life; often HH object See Herbig-Haro object. hydrogen burning The release of energy
assumed to be temperatures at which hierarchical clustering The “bottom- from the nuclear fusion of four hydrogen
water exists in a liquid state. up” process of forming large-scale atoms into a single helium atom.
Hadley circulation A simplified, and there- structure. Small-scale structure fi rst hydrogen shell burning The fusion of
fore uncommon, atmospheric global produces groups of galaxies, which in hydrogen in a shell surrounding a stellar
circulation that carries thermal energy turn form clusters, which then form core that may be either degenerate or fus-
directly from the equator to the polar superclusters. ing more massive elements.
regions of a planet. high-mass star A star with a main- hydrosphere The portion of Earth that is
half-life The time it takes half a sample of a sequence mass greater than about 9 solar largely liquid water. Compare biosphere
particular radioactive parent element to masses (M⊙). Compare intermediate-mass and lithosphere.
decay to a daughter product. star and low-mass star. hydrostatic equilibrium The condition in
halo The spherically symmetric, low- high-velocity star A star belonging to the which the weight bearing down at a par-
density distribution of stars and dark halo found near the Sun, distinguished ticular point within an object is balanced
matter that defi nes the outermost regions from disk stars by moving far faster and by the pressure within the object.
of a galaxy. often in the direction opposite to the rota- hypothesis A well-thought-out idea, based
harmonic law See Kepler’s third law. tion of the disk and its stars. on scientific principles and knowledge,
Hawking radiation Radiation from a black homogeneous In cosmology, describing a that leads to testable predictions. Com-
hole. universe in which observers in any loca- pare theory.
Hayashi track The path that a protostar fol- tion would observe the same properties. Hz See hertz.
lows on the H-R diagram as it contracts horizon The boundary that separates the
toward the main sequence. sky from the ground.
head The part of a comet that includes both horizon problem The puzzling observation I
the nucleus and the inner part of the coma. that the cosmic background radiation is ice The solid form of a volatile material;
heat death The possible eventual fate of so uniform in all directions, despite the sometimes the volatile material itself,
an open universe, in which entropy has fact that widely separated regions should regardless of its physical form.
triumphed and all energy- and structure- have been “over the horizon” from each ice giant A giant planet formed mostly of
producing processes have come to an end. other in the early universe. the liquid form of volatile substances
heavy element Also called massive ele- horizontal branch A region on the H-R dia- (ices.) In our Solar System, Uranus and
ment. 1. In astronomy, any element more gram defi ned by stars burning helium to Neptune are the ice giants. Compare gas
massive than helium. 2. In other sci- carbon in a stable core. giant.
ences (and sometimes also in astron- hot dark matter Particles of dark mat- ideal gas law The relationship of pressure
omy), any of the most massive elements ter that move so fast that gravity cannot (P) to number density of particles (n) and
in the periodic table, such as uranium confi ne them to the volume occupied by temperature (T) expressed as P = nkT,
and plutonium. a galaxy’s normal luminous matter. Com- where k is Boltzmann’s constant.
Heisenberg uncertainty principle The pare cold dark matter. igneous activity The formation and action
physical limitation that the product of the hot Jupiter A large, Jovian-type extraso- of molten rock (magma).
position and the momentum of a particle lar planet located very close to its parent impact crater The scar of the impact left
cannot be smaller than a well-defi ned star. on a solid planetary or moon surface by
value, Planck’s constant (h). hot spot A place where hot plumes of collision with another object. Compare
heliocentric A coordinate system having the mantle material rise near the surface of a secondary crater.
center of the Sun as its center. Compare planet. impact cratering A process involving colli-
geocentric. Hubble constant (H 0) The constant of pro- sions between solid planetary objects.
helioseismology The use of solar oscilla- portionality relating the recession veloci- index of refraction (n) The ratio of the
tions to study the interior of the Sun. ties of galaxies to their distances. See also speed of light in a vacuum (c) to the speed
helium flash The runaway explosive burn- Hubble time. of light in an optical medium (v).
ing of helium in the degenerate helium Hubble time An estimate of the age of the inert gas A gaseous element that combines
core of a red giant star. universe from the inverse of the Hubble with other elements only under condi-
Hayashi track The path that a protostar fol- constant, 1/H0. tions of extreme temperature and pres-
lows on the H-R diagram as it contracts Hubble’s law The law stating that the speed sure. Examples include helium, neon,
toward the main sequence. at which a galaxy is moving away from and argon.
Herbig-Haro (HH) object A glowing, rap- us is proportional to the distance of that inertia The tendency for objects to retain
idly moving knot of gas and dust that is galaxy. their state of motion.
excited by bipolar outflows in very young hurricane Also called cyclone or typhoon. inertial frame of reference 1. A frame of
stars. A large tropical cyclonic system circu- reference that is not accelerating. 2. In
heredity The process by which one genera- lating counterclockwise in the Northern general relativity, a frame of reference
tion passes on its characteristics to future Hemisphere and clockwise in the South- that is falling freely in a gravitational
generations. ern Hemisphere. Hurricanes can extend field.
hertz (Hz) A unit of frequency equivalent to outward from their center to more than inflation An extremely brief phase of ultra-
cycles per second. 600 kilometers (km) and generate winds rapid expansion of the very early uni-
G-8 Glossary
verse. Following inflation, the standard ionize The process by which electrons are Kepler’s laws The three rules of planetary
Big Bang models of expansion apply. stripped free from an atom or molecule, motion inferred by Johannes Kepler from
infrared (IR) radiation Electromagnetic resulting in free electrons and a posi- the data acquired by Tycho Brahe.
radiation with frequencies and pho- tively charged atom or molecule. Kepler’s second law Also called law of
ton energies occurring in the spectral ionosphere A layer high in Earth’s atmo- equal areas. A rule of planetary motion,
region between those of visible light sphere in which most of the atoms are inferred by Johannes Kepler, stating that
and microwaves. Compare ultraviolet ionized by solar radiation. a line drawn from the Sun to a planet
radiation. IR Infrared. sweeps out equal areas in equal times as
instability strip A region of the H-R dia- iron meteorite A metallic meteorite com- the planet orbits the Sun.
gram containing stars that pulsate with a posed mostly of iron-nickel alloys. Kepler’s third law Also called harmonic
periodic variation in luminosity. Compare stony-iron meteorite and stony law. A rule of planetary motion inferred
integration time The time interval during meteorite. by Johannes Kepler that describes the
which photons are collected and added irregular galaxy A galaxy without regular relationship between the period of a
up in a detecting device. or symmetric appearance. Compare ellip- planet’s orbit and its distance from the
intensity (of light) The amount of radiant tical galaxy, spiral galaxy, and S0 galaxy. Sun. The law states that the square of the
energy emitted per second per unit area. irregular moon A moon that has been cap- period of a planet’s orbit, measured in
Units for electromagnetic radiation: watts tured by a planet. Some irregular moons years, is equal to the cube of the semima-
per square meter (W/m2). revolve in the opposite direction from the jor axis of the planet’s orbit, measured in
intercloud gas A low-density region of the rotation of the planet, and many are in astronomical units: (Pyears)2 = (AAU)3.
interstellar medium that fi lls the space distant, unstable orbits. Compare regular kinetic energy (EK) The energy of an object
between interstellar clouds. moon. due to its motions. EK = 1 ⁄ 2 mv2. Units:
interference The interaction of two sets of isotopes Forms of the same element with joules (J).
waves producing high and low intensity, differing numbers of neutrons. Kirkwood gap A gap in the main asteroid
depending on whether their amplitudes isotropic In cosmology, describing a uni- belt related to orbital resonances with
reinforce (constructive interference) or verse whose properties observers fi nd to Jupiter.
cancel (destructive interference). be the same in all directions. Kuiper Belt A disk-shaped population of
interferometer Also called interferomet- comet nuclei extending from Neptune’s
ric array. A group or array of separate but orbit to perhaps several thousand astro-
linked optical or radio telescopes whose J nomical units (AU) from the Sun. The
overall separation determines the angular J See joule. highly populated innermost part of the
resolution of the system. jansky (Jy) The basic unit of flux density. Kuiper Belt has an outer edge approxi-
interferometric array See interferometer. Units: watts per square meter per hertz mately 50 AU from the Sun.
intermediate-mass star A star with a main- (W/m2/Hz). Kuiper Belt object (KBO) Also called trans-
sequence mass between about 3 and 9 jet 1. A stream of gas and dust ejected from Neptunian object. An icy planetesimal
solar masses (M⊙). Compare high-mass a comet nucleus by solar heating. 2. A (comet nucleus) that orbits within the
star and low-mass star. collimated linear feature of bright emis- Kuiper Belt beyond the orbit of Neptune.
interstellar cloud A discrete, high-density sion extending from a protostar or active
region of the interstellar medium made galactic nucleus.
up mostly of atomic or molecular hydro- joule (J) A unit of energy or work. 1 J = 1 L
gen and dust. newton meter. Lagrangian equilibrium point One of five
interstellar dust Small particles or grains Jy See jansky. points of equilibrium in a system con-
(0.01–10 micrometers [
m]) of matter, pri- sisting of two massive objects in nearly
marily carbon and silicates, distributed circular orbit around a common center
throughout interstellar space. K of mass. Only two Lagrangian points (L 4
interstellar extinction The dimming of vis- K See kelvin. and L 5) represent stable equilibrium. A
ible and ultraviolet light by interstellar K-T boundary See Cretaceous-Tertiary third smaller body located at one of the
dust. boundary. five points will move in lockstep with the
interstellar medium The gas and dust that KBO See Kuiper Belt object. center of mass of the larger bodies.
fi ll the space between the stars within a kelvin (K) The basic unit of the Kelvin scale lander An instrumented spacecraft
galaxy. of temperature. designed to land on a planet or moon.
inverse square law The rule that a quantity Kelvin scale The temperature scale— Compare rover.
or effect diminishes with the square of defi ned by William Thomson, better large-scale structure Observable aggre-
the distance from the source. known as Lord Kelvin (1824–1907)—that gates on the largest scales in the universe,
ion An atom or molecule that has lost or uses Celsius-sized degrees, but defi nes 0 including galaxy groups, clusters, and
gained one or more electrons. K as absolute zero instead of as the melt- superclusters.
ion tail A type of comet tail consisting of ing point of water. Compare Celsius and latitude The angular distance north (+) or
ionized gas. Particles in the ion tail are Fahrenheit. south (–) from the equatorial plane of a
pushed directly away from the comet’s Kepler’s fi rst law A rule of planetary nearly spherical body.
head in the antisolar direction at high motion, inferred by Johannes Kepler, stat- law of equal areas See Kepler’s second law.
speeds by the solar wind. Compare dust ing that planets move in orbits of ellipti- law of gravitation See universal law of
tail. cal shapes with the Sun at one focus. gravitation.
Glossary G-9
leap year A year that contains 366 days. an object. Unit: watts (W). See also lation (such as a star cluster) where stars
Leap years occur every 4 years when the brightness. have just evolved off the main sequence.
year is divisible by 4, correcting for the luminosity class A spectral classification The position of the main-sequence turn-
accumulated excess time in a normal based on stellar size, from the largest off is determined by the age of the stellar
year, which is approximately 3651 ⁄4 days supergiants to the smallest white dwarfs. population.
long. luminosity-temperature-radius relationship mantle The solid portion of a rocky planet
Leonids A November meteor shower asso- A relationship among these three proper- that lies between the crust and the core.
ciated with the dust debris left by comet ties of stars indicating that if any two are mare (pl. maria) A dark region on the
Tempel-Tuttle. Compare Perseids. known, the third can be calculated. Moon, composed of basaltic lava flows.
libration The apparent wobble of an orbit- luminous matter Also called normal mat- mass 1. Inertial mass: the property of mat-
ing body that is tidally locked to its com- ter. Matter in galaxies—including stars, ter that resists changes in motion. 2.
panion (such as Earth’s Moon) resulting gas, and dust—that emits electromagnetic Gravitational mass: the property of mat-
from the fact that its orbit is elliptical radiation. Compare dark matter. ter defi ned by its attractive force on other
rather than circular. lunar eclipse An eclipse that occurs when objects. According to general relativity,
life A biochemical process in which liv- the Moon is partially or entirely in the two are equivalent.
ing organisms can reproduce, evolve, and Earth’s shadow. Compare solar eclipse. mass-luminosity relationship An empiri-
sustain themselves by drawing energy lunar tide A tide on Earth that is due to cal relationship between the luminosity
from their environment. All terrestrial the differential gravitational pull of the (L) and mass (M) of main-sequence stars
life involves carbon-based chemistry, Moon. Compare solar tide. See also tide expressed as a power law—for example,
assisted by the self-replicating molecules (defi nition 2). L ∝ M 3.5.
ribonucleic acid (RNA) and deoxyribonu- ly See light-year. mass transfer The transfer of mass from
cleic acid (DNA). one member of a binary star system to its
light All electromagnetic radiation, which companion. Mass transfer occurs when
comprises the entire electromagnetic M one of the stars evolves to the point that
spectrum. M-type asteroid An asteroid that was once it overfi lls its Roche lobe, so that its
light-year (ly) The distance that light trav- part of the metallic core of a larger, dif- outer layers are pulled toward its binary
els in 1 year—about 9 trillion kilometers ferentiated body that has since been companion.
(km). broken into pieces; made mostly of iron massive element See heavy element.
limb darkening The darker appearance and nickel. Compare C-type asteroid and matter 1. Objects made of particles that
caused by increased atmospheric absorp- S-type asteroid. have mass, such as protons, neutrons,
tion near the limb of a planet or star. MACHO Literally, “massive compact halo and electrons. 2. Anything that occupies
line of nodes 1. A line defi ned by the inter- object.” MACHO’s include brown dwarfs, space and has mass.
section of two orbital planes. 2. The line white dwarfs, and black holes, which are Maunder Minimum The period from 1645
defi ned by the intersection of Earth’s candidates for being considered dark mat- to 1715, when very few sunspots were
equatorial plane and the plane of the ter. Compare WIMP. observed.
ecliptic. magma Molten rock, often containing dis- megabar A unit of pressure equal to 1 mil-
lithosphere The solid, brittle part of Earth solved gases and solid minerals. lion bars.
(or any planet or moon), including the magnetic field A field that is able to exert a mega-light-year (Mly) A unit of distance
crust and the upper part of the mantle. force on a moving electric charge. Com- equal to 1 million light-years.
Compare biosphere and hydrosphere. pare electric field. meridian The imaginary arc in the sky
lithospheric plate A separate piece of magnetic force A force associated with, or running from the horizon at due north
Earth’s lithosphere capable of moving caused by, the relative motion of charges. through the zenith to the horizon at due
independently. See also continental drift Compare electric force. See also electro- south. The meridian divides the observ-
and plate tectonics. magnetic force. er’s sky into eastern and western halves.
Local Group The small group of galaxies of magnetosphere The region surrounding a Compare equator.
which the Milky Way and the Androm- planet that is filled with relatively intense meteor The incandescent trail produced
eda galaxies are members. magnetic fields and plasmas. by a small piece of interplanetary debris
long-period comet A comet with an orbital magnitude A system used by astronomers as it travels through the atmosphere at
period of greater than 200 years. Compare to describe the brightness or luminosity very high speeds. Compare meteoroid and
short-period comet. of stars. The brighter the star, the smaller meteorite.
longitudinal wave A wave that oscillates its magnitude. meteor shower A larger-than-normal dis-
parallel to the direction of the wave’s main asteroid belt See asteroid belt. play of meteors, occurring when Earth
propagation. Compare transverse wave. main sequence The strip on the H-R dia- passes through the orbit of a disintegrat-
look-back time The time that it has taken gram where most stars are found. Main- ing comet, sweeping up its debris.
the light from an astronomical object to sequence stars are fusing hydrogen to meteorite A meteoroid that survives to
reach Earth. helium in their cores. reach a planet’s surface. Compare meteor
low-mass star A star with a main-sequence main-sequence lifetime The amount of time and meteoroid.
mass of less than about 3 solar masses a star spends on the main sequence, fus- meteoroid A small cometary or asteroi-
(M⊙). Compare high-mass star and inter- ing hydrogen into helium in its core. dal fragment ranging in size from 100
mediate-mass star. main-sequence turnoff The location on the micrometers (
m) to 100 meters. When
luminosity The total flux emitted by H-R diagram of a single-aged stellar popu- entering a planetary atmosphere, the
G-10 Glossary
meteoroid creates a meteor, which is adapted to their environment become Newton’s laws See Newton’s first law of
an atmospheric phenomenon. Compare more common than less well-adapted motion, Newton’s second law of motion,
meteor and meteorite; also planetesimal forms. and Newton’s third law of motion.
and zodiacal dust. NCP See north celestial pole. Newton’s second law of motion The law,
micrometer ( m) Also called micron. 10 –6 neap tide An especially weak tide that formulated by Isaac Newton, stating
meter; a unit of length used for the wave- occurs around the time of the fi rst or that if an unbalanced force acts on a
length of electromagnetic radiation. Com- third quarter Moon when the gravita- body, the body will have an accelera-
pare nanometer. tional forces of the Moon and the Sun on tion proportional to the unbalanced
micron See micrometer. Earth are at right angles to each other, force and inversely proportional to the
microwave radiation Electromagnetic thus producing the least pronounced object’s mass: a = F/m. The acceleration
radiation with frequencies and photon tides. Compare spring tide. See also tide will be in the direction of the unbal-
energies occurring in the spectral region (defi nition 2). anced force.
between those of infrared radiation and near-Earth asteroid An asteroid whose Newton’s third law of motion The law, for-
radio waves. orbit brings it close to the orbit of Earth. mulated by Isaac Newton, stating that for
Milky Way Galaxy The galaxy in which See also near-Earth object. every force there is an equal and opposite
our Sun and Solar System reside. near-Earth object (NEO) An asteroid, force.
minor planet See asteroid. comet, or large meteoroid whose orbit nm See nanometer.
minute of arc See arcminute. intersects Earth’s orbit. normal matter See luminous matter.
Mly See mega-light-year. nebula (pl. nebulae) A cloud of interstel- north celestial pole (NCP) The northward
m See micrometer. lar gas and dust, either illuminated by projection of Earth’s rotation axis onto
modern physics Usually, the physical prin- stars (bright nebula) or seen in silhou- the celestial sphere. Compare south celes-
ciples, including relativity and quantum ette against a brighter background (dark tial pole.
mechanics, that have been developed nebula). North Pole The location in the Northern
since James Maxwell’s equations were nebular hypothesis The fi rst plausible Hemisphere where Earth’s rotation axis
published. theory of the formation of the Solar Sys- intersects the surface of Earth. Compare
molecular cloud An interstellar cloud com- tem, proposed by Immanuel Kant in 1734. South Pole.
posed primarily of molecular hydrogen. Kant hypothesized that the Solar System nova (pl. novae) A stellar explosion that
molecular-cloud core A dense clump formed from the collapse of an interstel- results from runaway nuclear fusion in a
within a molecular cloud that forms as lar cloud of rotating gas. layer of material on the surface of a white
the cloud collapses and fragments. Proto- NEO See near-Earth object. dwarf in a binary system.
stars form from molecular-cloud cores. neutrino A very low-mass, electrically neu- nuclear burning Release of energy by
molecule Generally, the smallest particle tral particle emitted during beta decay. fusion of low-mass elements.
of a substance that retains its chemi- Neutrinos interact with matter only very nuclear fusion The combination of two less
cal properties and is composed of two or feebly and so can penetrate through great massive atomic nuclei into a single more
more atoms. A very few types of mol- quantities of matter. massive atomic nucleus.
ecules, such as helium, are composed of neutrino cooling The process in which nucleosynthesis The formation of more
single atoms. thermal energy is carried out of the center massive atomic nuclei from less massive
momentum The product of the mass and of a star by neutrinos rather than by elec- nuclei, either in the Big Bang (Big Bang
velocity of a particle. Units: kilograms tromagnetic radiation or convection. nucleosynthesis) or in the interiors of
times meters per second (kg m/s). neutron A subatomic particle having no net stars (stellar nucleosynthesis).
moon A less massive satellite orbiting a electric charge, and a rest mass and rest nucleus (pl. nuclei) 1. The dense, central
more massive object. Moons are found energy nearly equal to that of the proton. part of an atom. 2. The central core of a
around planets, dwarf planets, asteroids, Compare electron and proton. galaxy, comet, or other diffuse object.
and Kuiper Belt objects. neutron star The neutron-degenerate rem-
multiverse A collection of parallel uni- nant left behind by a Type II supernova.
verses that together comprise all that is. new Moon The phase of the Moon in which O
mutation In biology, an imperfect reproduc- the Moon is between Earth and the Sun, objective lens The primary optical element
tion of self-replicating material. and from Earth we see only the side of the in a telescope or camera that produces an
Moon not being illuminated by the Sun. image of an object.
Compare full Moon. oblateness The flattening of an otherwise
N newton (N) The force required to accelerate spherical planet or star caused by its
N See newton. a 1-kilogram (kg) mass at a rate of 1 meter rapid rotation.
nadir The point on the celestial sphere per second per second (m/s2). Units: kilo- obliquity The inclination of a celestial
located directly below an observer, oppo- grams times meters per second squared body’s equator to its orbital plane.
site the zenith. (kg m/s2). observational uncertainty The fact that
nanometer (nm) One billionth (10 –9) of a Newton’s fi rst law of motion The law, real measurements are never perfect;
meter; a unit of length used for the wave- formulated by Isaac Newton, stating that all observations are uncertain by some
length of light. Compare micrometer. an object will remain at rest or will con- amount.
natural selection The process by which tinue moving along a straight line at a Occam’s razor The principle that the sim-
forms of structure, ranging from mol- constant speed until an unbalanced force plest hypothesis is the most likely; named
ecules to whole organisms, that are best acts on it. after William of Occam (circa 1285–1349),
Glossary G-11
the medieval English cleric to whom the peculiar velocity The motion of a galaxy photosphere The apparent surface of the
idea is attributed. relative to the overall expansion of the Sun as seen in visible light.
Oort Cloud A spherical distribution of universe. physical law A broad statement that pre-
comet nuclei stretching from beyond the penumbra (pl. penumbrae) 1. The outer dicts a particular aspect of how the
Kuiper Belt to more than 50,000 astro- part of a shadow, where the source of physical universe behaves and that is
nomical units (AU) from the Sun. light is only partially blocked. 2. The supported by many empirical tests. See
opacity A measure of how effectively region surrounding the umbra of a sun- also theory.
a material blocks the radiation going spot. The penumbra is cooler and darker pixel The smallest picture element in a
through it. than the surrounding surface of the Sun digital image array.
open cluster A loosely bound group of a but not as cool or dark as the umbra of the Planck era The early time, just after the Big
few dozen to a few thousand stars that sunspot. Bang, when the universe as a whole must
formed together in the disk of a spiral gal- penumbral lunar eclipse A lunar eclipse in be described with quantum mechanics.
axy. Compare globular cluster. which the Moon passes through the pen- Planck spectrum Also called blackbody
open universe An infi nite universe with a umbra of Earth’s shadow. Compare total spectrum. The spectrum of electromag-
negatively curved spatial structure (much lunar eclipse. netic energy emitted by a blackbody per
like the surface of a saddle) such that the perihelion (pl. perihelia) The point in a unit area per second, which is deter-
sum of the angles of a triangle is always solar orbit that is closest to the Sun. Com- mined only by the temperature of the
less than 180°. Compare closed universe pare aphelion. object.
and flat universe. period The time it takes for a regularly Planck’s constant (h) The constant of pro-
orbit The path taken by one object moving repetitive process to complete one cycle. portionality between the energy of a
around another object under the influ- period-luminosity relationship The rela- photon and the frequency of the photon.
ence of their mutual gravitational or elec- tionship between the period of variability This constant defi nes how much energy
tric attraction. of a pulsating variable star, such as a Cep- a single photon of a given frequency or
orbital resonance A situation in which the heid or RR Lyrae variable, and the lumi- wavelength has. Value: h = 6.63 × 10 –34
orbital periods of two objects are related nosity of the star. Longer-period Cepheid joule-second.
by a ratio of small integers. or RR Lyrae variables are more luminous planet 1. A large body that orbits the Sun
orbiter A spacecraft that is placed in orbit than their shorter-period cousins. or other star that shines only by light
around a planet or moon. Compare flyby. Perseids A prominent August meteor reflected from the Sun or star. 2. In the
organic Describing a substance, not neces- shower associated with the dust debris Solar System, a body that orbits the
sarily of biological origin, that contains left by comet Swift-Tuttle. Compare Sun, has sufficient mass for self-gravity
the element carbon. Leonids. to overcome rigid body forces so that
phase One of the various appearances of it assumes a spherical shape, and has
the sunlit surface of the Moon or a planet cleared smaller bodies from the neigh-
P caused by the change in viewing location borhood around its orbit. Compare dwarf
P wave See primary wave. of Earth relative to both the Sun and the planet.
pair production The creation of a particle- object. Examples include crescent phase planet migration The theory that a planet
antiparticle pair from a source of electro- and gibbous phase. can move from its formation distance
magnetic energy. photino An elementary particle related to around its parent star to a different dis-
paleomagnetism The record of Earth’s mag- the photon. One of the leading candidates tance though gravitational interactions
netic field as preserved in rocks. for cold dark matter. with other bodies or loss of orbital energy
parallax 1. The apparent shift in the posi- photochemical reaction A chemical reac- from interaction with gas in the proto-
tion of one object relative to another tion driven by the absorption of electro- planetary disk.
object, caused by the changing perspec- magnetic radiation. planetary nebula The expanding shell of
tive of the observer. 2. In astronomy, the photodissociation The breaking apart of material ejected by a dying asymptotic
displacement in the apparent position molecules into smaller fragments or indi- giant branch star. A planetary neb-
of a nearby star caused by the changing vidual atoms by the action of photons. ula glows from fluorescence caused by
location of Earth in its orbit. photoelectric effect An effect whereby intense ultraviolet light coming from the
parent element A radioactive element that electrons are emitted from a substance hot, stellar remnant at its center.
decays to form more stable daughter illuminated by photons above a certain planetary system A system of planets and
products. critical frequency. other smaller objects in orbit around a
parsec (pc) The distance to a star with a photometry The process of measuring the star.
parallax of 1 arcsecond using a base of brightness of a source of light, generally planetesimal A primitive body of rock
1 astronomical unit (AU). One parsec is over a specific range of wavelength. and ice, 100 meters or more in diame-
approximately 3.26 light-years. photon Also called quantum of light. A ter, that combines with others to form a
partial solar eclipse The type of eclipse discrete unit or particle of electromag- planet. Compare meteoroid and zodia-
that occurs when Earth passes through netic radiation. The energy of a photon is cal dust.
the penumbra of the Moon’s shadow, so equal to Planck’s constant (h) multiplied plasma A gas that is composed largely of
that the Moon blocks only a portion of the by the frequency (f ) of its electromagnetic charged particles but also may include
Sun’s disk. Compare annular solar eclipse radiation: Ephoton = h × f. The photon is some neutral atoms.
and total solar eclipse. the particle that mediates the electromag- plate tectonics The geological theory
pc See parsec. netic force. concerning the motions of lithospheric
G-12 Glossary
plates, which in turn provides the theo- energy, rather than from nuclear reac- radio galaxy A type of elliptical galaxy
retical basis for continental drift. tions in its core. with an active galactic nucleus at its
positron A positively charged subatomic pulsar A rapidly rotating neutron star that center and having very strong emission
particle; the antiparticle of the electron. beams radiation into space in two search- (1035–1038 watts [W]) in the radio part of
power The rate at which work is done or light-like beams. To a distant observer, the electromagnetic spectrum. Compare
at which energy is delivered. Unit: watts the star appears to flash on and off, earn- Seyfert galaxy.
(W) or joules per second (J/s). ing its name. radio telescope An instrument for detect-
precession of the equinoxes The slow pulsating variable star A variable star that ing and measuring radio frequency emis-
change in orientation between the eclip- undergoes periodic radial pulsations. sions from celestial sources.
tic plane and the celestial equator caused radio wave Electromagnetic radiation
by the wobbling of Earth’s axis. in the extreme long-wavelength region
pressure Force per unit area. Units: new- Q of the spectrum, beyond the region of
tons per square meter (N/m2) or bars. QCD See quantum chromodynamics. microwaves.
primary atmosphere An atmosphere, com- QED See quantum electrodynamics. radiometric dating Use of radioactive decay
posed mostly of hydrogen and helium, quantized Describing a quantity that exists to measure the ages of materials such as
that forms at the same time as its host as discrete, irreducible units. minerals.
planet. Compare secondary atmosphere. quantum chromodynamics (QCD) The ratio The relationship in quantity or size
primary mirror The principal optical mir- quantum mechanical theory describ- between two or more things.
ror in a reflecting telescope. The primary ing the strong nuclear force and its ray 1. A beam of electromagnetic radia-
mirror determines the telescope’s light- mediation by gluons. Compare quantum tion. 2. A bright streak emanating from a
gathering power and resolution. Compare electrodynamics. young impact crater.
secondary mirror. quantum efficiency The fraction of photons recombination 1. The combining of ions
primary wave Also called P wave. A lon- falling on a detector that actually pro- and electrons to form neutral atoms. 2.
gitudinal seismic wave, in which the duces a response in the detector. An event early in the evolution of the
oscillations involve compression and quantum electrodynamics (QED) The quan- universe in which hydrogen and helium
decompression parallel to the direction of tum theory describing the electromag- nuclei combined with electrons to form
travel (that is, a pressure wave). Compare netic force and its mediation by photons. neutral atoms. The removal of electrons
secondary wave. Compare quantum chromodynamics. caused the universe to become transpar-
principle A general idea or sense about how quantum mechanics The branch of physics ent to electromagnetic radiation.
the universe is that guides us in con- that deals with the quantized and proba- red giant A low-mass star that has evolved
structing new scientific theories. Prin- bilistic behavior of atoms and subatomic beyond the main sequence and is now
ciples can be testable theories. particles. fusing hydrogen in a shell surrounding a
prograde motion 1. Rotational or orbital quantum of light See photon. degenerate helium core.
motion of a moon that is in the same quark The building block of protons and red giant branch A region on the H-R dia-
sense as the planet it orbits. 2. The coun- neutrons. gram defi ned by low-mass stars evolving
terclockwise orbital motion of Solar Sys- quasar Short for quasi-stellar radio source. from the main sequence toward the hori-
tem objects as seen from above Earth’s The most luminous of the active galactic zontal branch.
orbital plane. Compare retrograde motion. nuclei, seen only at great distances from reddening The effect by which stars and
prominence An archlike projection above our galaxy. other objects, when viewed through
the solar photosphere often associated interstellar dust, appear redder than they
with a sunspot. actually are. Reddening is caused by
proportional Describing two things whose R the fact that blue light is more strongly
ratio is a constant. radial velocity The component of velocity absorbed and scattered than red light.
proton (p or p+) A subatomic particle hav- that is directed toward or away from the redshift Also called Doppler redshift.
ing a positive electric charge of 1.6 × 10 –19 observer. The shift toward longer wavelengths of
coulomb (C), a mass of 1.67 × 10 –27 kilo- radian The angle at the center of a circle light by any of several effects, including
grams (kg), and a rest energy of 1.5 × 10 –10 subtended by an arc equal to the length of Doppler shifts, gravitational redshift,
joules (J). Compare electron and neutron. the circle’s radius. Therefore, 2π radians or cosmological redshift. Compare
proton-proton chain One of the ways in equals 360°, and 1 radian equals approxi- blueshift.
which hydrogen burning can take place. mately 57.3°. reflecting telescope A telescope that uses
This is the most important path for radiant The direction in the sky from mirrors for collecting and focusing
hydrogen burning in low-mass stars such which the meteors in a meteor shower incoming electromagnetic radiation to
as the Sun. See also carbon-nitrogen-oxy- seem to come. form an image in their focal planes. The
gen cycle and triple-alpha process. radiation belt A toroidal ring of high- size of a reflecting telescope is defi ned by
protoplanetary disk Also called circum- energy particles surrounding a planet. the diameter of the primary mirror. Com-
stellar disk. The remains of the accretion radiative transfer The transport of energy pare refracting telescope.
disk around a young star from which a from one location to another by electro- reflection The redirection of a beam of light
planetary system may form. magnetic radiation. that is incident on, but does not cross, the
protostar A young stellar object that radiative zone A region in the interior of a surface between two media having dif-
derives its luminosity from the conver- star through which energy is transported ferent refractive indices. If the surface is
sion of gravitational energy to thermal outward by radiation. flat and smooth, the angle of incidence
Glossary G-13
equals the angle of reflection. Compare Roche limit The distance at which a plan- isolated system always increases as the
refraction. et’s tidal forces exceed the self-gravity of system evolves.
refracting telescope A telescope that uses a smaller object, such as a moon, aster- second of arc See arcsecond.
objective lenses to collect and focus light. oid, or comet, causing the object to break secondary atmosphere An atmosphere that
Compare reflecting telescope. apart. formed—as a result of volcanism, comet
refraction The redirection or bending of a Roche lobe The hourglass- or figure impacts, or another process—sometime
beam of light when it crosses the bound- eight-shaped volume of space sur- after its host planet formed. Compare pri-
ary between two media having different rounding two stars, which constrains mary atmosphere.
refractive indices. Compare reflection. material that is gravitationally bound secondary crater A crater formed from
refractory material Material that remains by one or the other. ejecta thrown from an impact crater.
solid at high temperatures. Compare vola- rotation curve A plot showing how the secondary mirror A small mirror placed on
tile material. orbital velocity of stars and gas in a gal- the optical axis of a reflecting telescope
regular moon A moon that formed together axy changes with radial distance from that returns the beam back through a
with the planet it orbits. Compare irregu- the galaxy’s center. small hole in the primary mirror, thereby
lar moon. rover A remotely controlled instrumented shortening the mechanical length of the
relative humidity The amount of water vehicle designed to traverse and explore telescope.
vapor held by a volume of air at a given the surface of a terrestrial planet or moon. secondary wave Also called S wave. A
temperature compared (stated as a per- Compare lander. transverse seismic wave, which involves
centage) to the total amount of water that RR Lyrae variable A variable giant star the sideways motion of material. Compare
could be held by the same volume of air whose regularly timed pulsations are primary wave.
at the same temperature. good predictors of its luminosity. RR seismic wave A vibration due to an earth-
relative motion The difference in motion Lyrae stars are used for distance mea- quake, a large explosion, or an impact on
between two individual frames of surements to globular clusters. the surface that travels through a planet’s
reference. interior.
relativistic Describing physical processes seismometer An instrument that measures
that take place in systems traveling at S the amplitude and frequency of seismic
nearly the speed of light or located in S-type asteroid An asteroid made of mate- waves.
the vicinity of very strong gravitational rial that has been modified from its self-gravity The gravitational attraction
fields. original state, likely as the outer part of a among all the parts of the same object.
relativistic beaming The effect created larger, differentiated body that has since semimajor axis Half of the longer axis of an
when material moving at nearly the speed broken into pieces. Compare C-type aster- ellipse.
of light beams the radiation it emits in the oid and M-type asteroid. SETI The Search for Extraterrestrial Intelli-
direction of its motion. S wave See secondary wave. gence project, which uses advanced tech-
remote sensing The use of images, spectra, S0 galaxy A galaxy with a bulge and a disk- nology combined with radio telescopes
radar, or other techniques to measure the like spiral, but smooth in appearance like to search for evidence of intelligent life
properties of an object from a distance. ellipticals. Compare elliptical galaxy, spi- elsewhere in the universe.
resolution The ability of a telescope to ral galaxy, and irregular galaxy. Seyfert galaxy A type of spiral galaxy with
separate two point sources of light. Reso- satellite 1. An object in orbit about a more an active galactic nucleus at its center;
lution is determined by the telescope’s massive body. 2. A moon. fi rst discovered in 1943 by Carl Seyfert.
aperture and the wavelength of light it scale factor (R U) A dimensionless number Compare radio galaxy.
receives. proportional to the distance between two shepherd moon A moon that orbits close
rest wavelength The wavelength of light points in space. The scale factor increases to rings and gravitationally confi nes the
we see coming from an object at rest with as the universe expands. orbits of the ring particles.
respect to the observer. scattering The random change in the direc- shield volcano A volcano formed by very
retrograde motion 1. Rotation or orbital tion of travel of photons, caused by their fluid lava flowing from a single source
motion of a moon that is in the oppo- interactions with molecules or dust and spreading out from that source. Com-
site sense to the rotation of the planet it particles. pare composite volcano.
orbits. 2. The clockwise orbital motion of scientific method The formal procedure— short-period comet A comet with an orbital
Solar System objects as seen from above including hypothesis, prediction, and period of less than 200 years. Compare
Earth’s orbital plane. Compare prograde experiment or observation—used to test long-period comet.
motion. (attempt to falsify) the validity of scien- sidereal period An object’s orbital or rota-
ring An aggregation of small particles tific hypotheses and theories. tional period measured with respect to
orbiting a planet or star. The rings of the scientific notation The standard expression the stars. Compare synodic period.
four giant planets of the Solar System are of numbers with one digit (which can be silicate One of the family of minerals com-
composed variously of silicates, organic zero) to the left of the decimal point and posed of silicon and oxygen in combina-
materials, and ices. multiplied by 10 to the exponent required tion with other elements.
ring arc A discontinuous, higher-density to give the number its correct value. singularity The point where a mathemati-
region within an otherwise continuous, Example: 2.99 × 108 = 299,000,000. cal expression or equation becomes
narrow ring. SCP See south celestial pole. meaningless, such as the denominator
ringlet A narrowly confi ned concentration second law of thermodynamics The law of a fraction approaching zero. See also
of ring particles. stating that the entropy or disorder of an black hole.
G-14 Glossary
solar abundance The relative amount of an that the speed of light is a constant affects of tiny particles. Their origin is not well
element detected in the atmosphere of the nonaccelerating frames of reference. Com- understood.
Sun, expressed as the ratio of the number pare general theory of relativity. sporadic meteor A meteor that is not asso-
of atoms of that element to the number of spectral type A classification system for ciated with a specific meteor shower.
hydrogen atoms. stars based on the presence and relative spreading center A zone from which two
solar day The 24-hour period of Earth’s strength of absorption lines in their spec- tectonic plates diverge.
axial rotation that brings the Sun back to tra. Spectral type is related to the surface spring tide An especially strong tide that
the same local meridian where the rota- temperature of a star. occurs near the time of a new or full
tion started. spectrograph A device that spreads out the Moon, when lunar tides and solar tides
solar eclipse An eclipse that occurs when light from an object into its component reinforce each other. Compare neap tide.
the Sun is partially or entirely blocked by wavelengths. See also spectrometer. See also tide (defi nition 2).
the Moon. Compare lunar eclipse. spectrometer A spectrograph in which the stable equilibrium An equilibrium state
solar flare Explosive events on the Sun’s spectrum is generally recorded digitally in which the system returns to its former
surface associated with complex sunspot by electronic means. condition after a small disturbance. Com-
groups and strong magnetic fields. spectroscopic binary A binary star pair pare unstable equilibrium.
solar maximum (pl. maxima) The time, whose existence and properties are standard candle An object whose lumi-
occurring about every 11 years, when the revealed only by the Doppler shift of its nosity either is known or can be pre-
Sun is at its peak activity, meaning that spectral lines. Most spectroscopic bina- dicted in a distance-independent way,
sunspot activity and related phenomena ries are close pairs. Compare eclipsing so its brightness can be used to deter-
(such as prominences, flares, and coronal binary and visual binary. mine its distance via the inverse square
mass ejections) are at their peak. spectroscopic parallax Use of the spectro- law of radiation.
solar neutrino problem The historical scopically determined luminosity and the standard model The theory of particle
observation that only about a third as observed brightness of a star to determine physics that combines electroweak the-
many neutrinos as predicted by theory the star’s distance. ory with quantum chromodynamics to
seemed to be coming from the Sun. spectroscopy The study of electromagnetic describe the structure of known forms of
Solar System The gravitationally bound radiation from an object in terms of its matter.
system made up of the Sun, planets, component wavelengths. star A luminous ball of gas that is held
dwarf planets, moons, asteroids, comets, spectrum (pl. spectra) 1. The intensity of together by gravity. A normal star is pow-
and Kuiper Belt objects, along with their electromagnetic radiation as a func- ered by nuclear reactions in its interior.
associated gas and dust. tion of wavelength. 2. Waves sorted by star cluster A group of stars that all formed
solar tide A tide on Earth that is due to the wavelength. at the same time and in the same general
differential gravitational pull of the Sun. speed The rate of change of an object’s location.
Compare lunar tide. See also tide (defi ni- position with time, without regard to the static equilibrium A state in which the
tion 2). direction of movement. Units: meters forces within a system are all in balance
solar wind The stream of charged particles per second (m/s) or kilometers per hour so that the system does not change. Com-
emitted by the Sun that flows at high (km/h). Compare velocity. pare dynamic equilibrium.
speeds through interplanetary space. spherically symmetric Describing an object Stefan-Boltzmann constant (σ) The pro-
solstice Literally, “sun standing still.” whose properties depend only on dis- portionality constant that relates the flux
1. One of the two most northerly and tance from the object’s center, so that the emitted by an object to the fourth power
southerly points on the ecliptic. 2. Either object has the same form viewed from of its absolute temperature. Value: 5.67 ×
of the two times of year (the summer sol- any direction. 10 –8 W/(m2 K4) (W = watts, m = meters, K
stice and winter solstice) when the Sun spin-orbit resonance A relationship = kelvin).
is at one of these two positions. Compare between the orbital and rotation peri- Stefan-Boltzmann law The law stating that
equinox. ods of an object such that the ratio of the amount of electromagnetic energy
south celestial pole (SCP) The southward their periods can be expressed by simple emitted from the surface of a body,
projection of Earth’s rotation axis onto integers. summed over the energies of all photons
the celestial sphere. Compare north celes- spiral density wave A stable, spiral-shaped of all wavelengths emitted, is propor-
tial pole. change in the local gravity of a galactic tional to the fourth power of the tempera-
South Pole The location in the Southern disk that can be produced by periodic ture of the body.
Hemisphere where Earth’s rotation axis gravitational kicks from neighboring gal- stellar mass loss The loss of mass from the
intersects the surface of Earth. Compare axies or from nonspherical bulges and outermost parts of a star’s atmosphere
North Pole. bars in spiral galaxies. during the course of its evolution.
spacetime The four-dimensional contin- spiral galaxy A galaxy of Hubble type “S” stellar occultation An event in which a
uum in which we live, and which we class, with a discernible disk in which planet or other Solar System body moves
experience as three spatial dimensions large spiral patterns exist. Compare ellipti- between the observer and a star, eclipsing
plus time. cal galaxy, S0 galaxy, and irregular galaxy. the light emitted by that star.
special relativity See special theory of spoke One of several narrow radial fea- stellar population A group of stars with
relativity. tures seen occasionally in Saturn’s B similar ages, chemical compositions, and
special theory of relativity Sometimes ring. Spokes appear dark in backscattered dynamic properties.
referred to as simply special relativity. light and bright in forward, scattering stereoscopic vision The way an animal’s
Einstein’s theory explaining how the fact light, indicating that they are composed brain combines the different information
Glossary G-15
from its two eyes to perceive the dis- axy clusters and galaxy groups; typically, wind and by radiation pressure from the
tances to objects around it. more than 100 million light-years (Mly) Sun.
stony-iron meteorite A meteorite consisting in size and containing tens of thousands tectonism Deformation of the lithosphere of
of a mixture of silicate minerals and iron- to hundreds of thousands of galaxies. a planet.
nickel alloys. Compare iron meteorite and Compare galaxy cluster and galaxy group. telescope The basic tool of astronomers.
stony meteorite. superluminal motion The appearance Working over the entire range from
stony meteorite A meteorite composed (though not the reality) that a jet is mov- gamma rays to radio, astronomical tele-
primarily of silicate minerals, similar to ing faster than the speed of light. scopes collect and concentrate electro-
those found on Earth. Compare iron mete- supermassive black hole A black hole of magnetic radiation from celestial objects.
orite and stony-iron meteorite. 1,000 solar masses (M⊙) or more that temperature A measure of the average
stratosphere The atmospheric layer imme- resides in the center of a galaxy, and kinetic energy of the atoms or molecules
diately above the troposphere. On Earth it whose gravity powers active galactic in a gas, solid, or liquid.
extends upward to an altitude of 50 kilo- nuclei. terrestrial planet An Earth-like planet,
meters (km). supernova (pl. supernovae) A stellar explo- made of rock and metal and having a
strong nuclear force The attractive short- sion resulting in the release of tremen- solid surface. In our Solar System, the
range force between protons and neutrons dous amounts of energy, including the terrestrial planets are Mercury, Venus,
that holds atomic nuclei together; one of high-speed ejection of matter into the Earth, and Mars. Compare giant planet.
the four fundamental forces of nature, interstellar medium. See also Type I theoretical model A detailed description
mediated by the exchange of gluons. supernova and Type II supernova. of the properties of a particular object or
Compare weak nuclear force. supersonic Moving within a medium at a system in terms of known physical laws
subduction zone A region where two tec- speed faster than the speed of sound in or theories. Often, a computer calculation
tonic plates converge, with one plate that medium. Compare subsonic. of predicted properties based on such a
sliding under the other and being drawn superstring theory The theory that con- description.
downward into the interior. ceives of particles as strings in 11 dimen- theory A well-developed idea or group
subgiant A giant star smaller and lower in sions of space and time; the current of ideas that are tied solidly to known
luminosity than normal giant stars of the contender for a theory of everything. physical laws and make testable predic-
same spectral type. Subgiants evolve to surface brightness The amount of electro- tions about the world. A very well-tested
become giants. magnetic radiation emitted or reflected theory may be called a physical law, or
subgiant branch A region of the H-R dia- per unit area. simply a fact. Compare hypothesis.
gram defi ned by stars that have left the surface wave A seismic wave that travels theory of everything (TOE) A theory that
main sequence but have not yet reached on the surface of a planet or moon. unifies all four fundamental forces of
the red giant branch. symmetry 1. The property that an object nature: strong nuclear, weak nuclear,
sublimation The process in which a solid has if the object is unchanged by rota- electromagnetic, and gravitational forces.
becomes a gas without fi rst becoming a tion or reflection about a particular thermal conduction The transfer of energy
liquid. point, line, or plane. 2. In theoretical in which the thermal energy of particles
subsonic Moving within a medium at a physics, the correspondence of different is transferred to adjacent particles by col-
speed slower than the speed of sound in aspects of physical laws or systems, such lisions or other interactions. Conduction
that medium. Compare supersonic. as the symmetry between matter and is the most important way that thermal
summer solstice 1. One of two points where antimatter. energy is transported in solid matter.
the Sun is at its greatest distance from the synchronous rotation The case in which thermal energy The energy that resides in
celestial equator. 2. The day on which the the period of rotation of a body on its axis the random motion of atoms, molecules,
Sun appears at this location, marking the equals the period of revolution in its orbit and particles, by which we measure their
fi rst day of summer (about June 21 in the around another body. A special type of temperature.
Northern Hemisphere and December 22 spin-orbit resonance. thermal equilibrium The state in which
in the Southern Hemisphere). Compare synchrotron radiation Radiation from elec- the rate of thermal-energy emission by
winter solstice. trons moving at close to the speed of light an object is equal to the rate of thermal-
sungrazer A comet whose perihelion is as they spiral in a strong magnetic field; energy absorption.
within a few solar diameters of the sur- named because this kind of radiation was thermal motion The random motion of
face of the Sun. fi rst identified on Earth in particle accel- atoms, molecules, and particles that gives
sunspot A cooler, transitory region on the erators called “synchrotrons.” rise to thermal radiation.
solar surface produced when loops of synodic period An object’s orbital or rota- thermal radiation Electromagnetic radia-
magnetic flux break through the surface tional period measured with respect to tion resulting from the random motion of
of the Sun. the Sun. Compare sidereal period. the charged particles in every substance.
sunspot cycle The approximate 11-year third quarter Moon The phase of the Moon
cycle during which sunspot activity in which only the eastern half of the
increases and then decreases. This is T Moon, as viewed from Earth, is illumi-
one-half of a full 22-year cycle, in which T Tauri star A young stellar object that has nated by the Sun. It occurs about one
the magnetic polarity of the Sun fi rst dispersed enough of the material sur- week after the full Moon. Compare first
reverses and then returns to its original rounding it to be seen in visible light. quarter Moon.
configuration. tail A stream of gas and dust swept away tidal bulge A distortion of a body resulting
supercluster A large conglomeration of gal- from the coma of a comet by the solar from tidal stresses.
G-16 Glossary
tidal locking Synchronous rotation of an tudes 23.5° south and 23.5° north, and in universal gravitational constant (G) The
object caused by internal friction as the which the Sun appears directly overhead constant of proportionality in the univer-
object rotates through its tidal bulge. twice during the year. sal law of gravity. Value: G = 6.673 × 10 –11
tidal stress Stress due to differences in the tropopause The top of a planet’s Newtons times meters squared per kilo-
gravitational force of one mass on differ- troposphere. gram squared (N m2/kg2).
ent parts of another mass. troposphere The convection-dominated universal law of gravitation The law stat-
tide 1. The deformation of a mass due to layer of a planet’s atmosphere. On ing that the gravitational force between
differential gravitational effects of one Earth, the atmospheric region clos- any two objects is proportional to the
mass on another because of the extended est to the ground within which most product of their masses and inversely
size of the masses. 2. On Earth, the rise weather phenomena take place. Compare proportional to the square of the distance
and fall of the oceans as Earth rotates stratosphere. between them: F ∝ (m1m2/r 2).
through a tidal bulge caused by the Moon tuning fork diagram The two-pronged universe All of space and everything con-
and Sun. diagram showing Hubble’s classification tained therein.
time dilation The relativistic “stretching” of galaxies into ellipticals, S0s, spirals, unstable equilibrium An equilibrium state
of time. Compare general relativistic time barred S0s and spirals, and irregular in which a small disturbance will cause
dilation. galaxies. a system to move away from equilibrium.
TNO See trans-Neptunian object. turbulence The random motion of blobs of Compare stable equilibrium.
TOE See theory of everything. gas within a larger cloud of gas. UV Ultraviolet.
topographic relief The differences in eleva- Type I supernova A supernova explosion in
tion from point to point on a planetary which no trace of hydrogen is seen in the
surface. ejected material. Most Type I supernovae V
tornado A violent rotating column of air, are thought to be the result of runaway vacuum A region of space that contains
typically 75 meters across with 200- carbon burning in a white dwarf star onto very little matter. In quantum mechan-
kilometer-per-hour (km/h) winds. Some which material is being deposited by a ics and general relativity, however,
tornadoes can be more than 3 km across, binary companion. even a perfect vacuum has physical
and winds up to 500 km/h have been Type II supernova A supernova explo- properties.
observed. sion in which the degenerate core of an variable star A star with varying luminos-
total lunar eclipse A lunar eclipse in which evolved massive star suddenly collapses ity. Many periodic variables are found
the Moon passes through the umbra of and rebounds. within the instability strip on the H-R
Earth’s shadow. Compare penumbral typhoon See hurricane. diagram.
lunar eclipse. velocity The rate and direction of change
total solar eclipse The type of eclipse that of an object’s position with time. Units:
occurs when Earth passes through the U meters per second (m/s) or kilometers per
umbra of the Moon’s shadow, so that the ultraviolet (UV) radiation Electromagnetic hour (km/h). Compare speed.
Moon completely blocks the disk of the radiation with frequencies and photon vernal equinox 1. One of two points
Sun. Compare annular solar eclipse and energies greater than those of visible light where the Sun crosses the celestial
partial solar eclipse. but less than those of X-rays, and wave- equator. 2. The day on which
transform fault The actively slipping seg- lengths shorter than those of visible light the Sun appears at this location, mark-
ment of a fracture zone between litho- but longer than those of X-rays. Compare ing the fi rst day of spring (about March
spheric plates. infrared radiation. 21 in the Northern Hemisphere and
trans-Neptunian object (TNO) See Kuiper umbra (pl. umbrae) 1. The darkest part of a September 23 in the Southern Hemi-
Belt object. shadow, where the source of light is com- sphere). Compare autumnal equinox.
transverse wave A wave that oscillates per- pletely blocked. 2. The darkest, innermost virtual particle A particle that, accord-
pendicular to the direction of the wave’s part of a sunspot. Compare penumbra. ing to quantum mechanics, comes into
propagation. Compare longitudinal wave. unbalanced force The nonzero net force existence only momentarily. According to
triple-alpha process The nuclear fusion acting on a body. theory, fundamental forces are mediated
reaction that combines three helium unbound orbit An orbit in which the veloc- by the exchange of virtual particles.
nuclei (alpha particles) together into a ity is greater than the escape velocity. visual binary A binary system in which
single nucleus of carbon. See also carbon- Compare bound orbit. the two stars can be seen individually
nitrogen-oxygen cycle and proton-proton uncertainty principle See Heisenberg from Earth. Compare eclipsing binary and
chain. uncertainty principle. spectroscopic binary.
Trojan asteroid One of a group of aster- unified model of AGNs A model in which void A region in space containing little or
oids orbiting in the L 4 and L 5 Lagrangian many different types of activity in the no matter. Examples include regions in
points of Jupiter’s orbit. nuclei of galaxies are all explained by cosmological space that are largely empty
tropical year The time between one cross- accretion of matter around a supermas- of galaxies.
ing of the vernal equinox and the next. sive black hole. volatile material Sometimes called ice.
Because of the precession of the equi- uniform circular motion Motion in a circu- Material that remains gaseous at mod-
noxes, a tropical year is slightly shorter lar path at a constant speed. erate temperature. Compare refractory
than the time that it takes for Earth to unit A fundamental quantity of measure- material.
orbit once about the Sun. ment—for example, metric units or Eng- volcanism The occurrence of volcanic
Tropics The region on Earth between lati- lish units. activity on a planet or moon.
Glossary G-17
vortex (pl. vortices) Any circulating fluid weight 1. The force equal to the mass of but less than those of gamma rays, and
system. Specifically, 1. an atmospheric an object multiplied by the local accel- wavelengths shorter than those of UV
anticyclone or cyclone; 2. a whirlpool or eration due to gravity. 2. In general light but longer than those of gamma
eddy. relativity, the force equal to the mass of rays.
an object multiplied by the acceleration X-ray binary A binary system in which
of the frame of reference in which the mass from an evolving star spills over
W object is observed. onto a collapsed companion, such as a
W See watt. white dwarf The stellar remnant left at the neutron star or black hole. The material
waning The changing phases of the Moon end of the evolution of a low-mass star. falling in is heated to such high tempera-
as it becomes less fully illuminated A typical white dwarf has a mass of 0.6 tures that it glows brightly in X-rays.
between full Moon and new Moon as seen solar mass (M⊙) and a size about equal to
from Earth. Compare waxing. that of Earth; it is made of nonburning,
watt (W) A measure of power. Units: joules electron-degenerate carbon. Y
per second (J/s). Wien’s law A relationship describing how year The time it takes Earth to make one
wave A disturbance moving along a surface the peak wavelength, and therefore revolution around the Sun. A solar year
or passing through a space or a medium. the color, of electromagnetic radiation is measured from equinox to equinox. A
wavefront The imaginary surface of an from a glowing blackbody changes with sidereal year, Earth’s true orbital period,
electromagnetic wave, either plane or temperature. is measured relative to the stars.
spherical, oriented perpendicular to the WIMP Literally, “weakly interacting mas-
direction of travel. sive particle.” A hypothetical massive
wavelength The distance on a wave particle that interacts through gravity but Z
between two adjacent points having iden- not with electromagnetic radiation and zenith The point on the celestial sphere
tical characteristics. The distance a wave is a candidate for dark matter. Compare located directly overhead from an
travels in one period. Unit: meter. MACHO. observer. Compare nadir.
waxing The changing phases of the Moon winter solstice 1. One of two points where zodiac The 12 constellations lying along
as it becomes more fully illuminated the Sun is at its greatest distance from the the plane of the ecliptic.
between new Moon and full Moon as seen celestial equator. 2. The day on which zodiacal dust Particles of cometary and
from Earth. Compare waning. the Sun appears at this location, marking asteroidal debris less than 100 microm-
weak nuclear force The force underlying the fi rst day of winter (about December 22 eters (
m) in size that orbit the inner
some forms of radioactivity and certain in the Northern Hemisphere and June 21 Solar System close to the plane of
interactions between subatomic particles. in the Southern Hemisphere). Compare the ecliptic. Compare meteoroid and
It is responsible for radioactive beta decay summer solstice. planetesimal.
and for the initial proton-proton interac- zodiacal light A band of light in the night
tions that lead to nuclear fusion in the sky caused by sunlight reflected by zodia-
Sun and other stars. One of the four fun- X cal dust.
damental forces of nature, mediated by X-ray Electromagnetic radiation hav- zonal wind The planetwide circulation of
the exchange of W and Z particles. Com- ing frequencies and photon energies air that moves in directions parallel to
pare strong nuclear force. greater than those of ultraviolet light the planet’s equator.
Credits
These pages contain credits for all 23 chapters in the complete volume of 21st Century
Astronomy, Third Edition. The Solar System Edition does not include Chapters
15–22. The Stars and Galaxies Edition does not include Chapters 6–12.
2.14b Larry Landolfi/Photo Research. 2.16 ASU Physics Instructional Resource Team.
Chapter 1 top left AFP/Getty Images; 2.16 top right AP 5.13c Steve Percival/Photo Researchers, Inc.
Opener NASA, ESA, Hubble Heritage Images; 2.16 bottom left Lindsay Hebberd/ 5.14 ASU Physics Instructional Resource
(STScI/AURA); D. Carter (LJMU) et al. and Corbis; 2.16 bottom right Lynn Goldsmith/ Team. 5.17a, b top Keck Observatory; 5.17a,
the Coma HST ACS Treasury Team. 1.2a, Corbis. 2.21 Roger Ressmeyer/Corbis. 2.23 b bottom Gemini Observatory, US National
b Neil Ryder Hoos; 1.2c Owen Franken/ Nick Quinn. 2.24 Dennis L. Mammana; Science Foundation, and University of
Corbis; 1.2d Monkey Business Images/ 2.25a Reuters New Media Inc./Corbis; 2.25b Hawaii Institute for Astronomy. 5.18 left
Dreamstime.com; 1.2e PhotoDisc; 1.2f ©2001 by Fred Espenak, courtesy of www. to right NASA; NASA/CXC/SAO; Graphic
American Museum of Natural History; 1.2g MrEclipse.com. 2.27 Fred Espenak NASA/ courtesy Orbital Sciences Corp; NOAO/
American Museum of Natural History; GSFC. 2.28a Johannes Schedler; 2.28b AURA/NSF; NASA/JPL/Caltech; Joint
1.2h NASA/JPL/Caltech. 1.3 Bettmann/ Anthony Ayiomamitis (TWAN). Astronomy Center in Hilo, Hawaii; NRAO
Corbis. 1.4 left NASA/STScI/Arizona VLA Image Gallery; The National Radio
State University/Hester/Ressmeyer/Corbis. Astronomy Observatory, Green Bank; Dave
1.4 right Ron Watts/Corbis. 1.5a Michael Chapter 3 Finley, courtesy National Radio Astronomy
J. Tuttle (NASM/NASA); 1.5b NSSDC. Opener NASA; 3.1 Nicolaus Copernicus Observatory and Associated Universities,
1.6 NRAO/AUI, James J. Condon, John J. (1473–1543) (oil on canvas), Pomeranian Inc. 5.19a Science Photo Library; 5.19b
Broderick, and George A. Seielstad. 1.7 School, (16th century)/Nicolaus Copernicus Hulton-Deutsch Collection/Corbis. 5.20
Fermilab. 1.9 top left Special Relativity: Museum, Frombork, Poland/Lauros/ NASA/IPAC/NED/JPL/Caltech/SSDS. 5.21b
The M.I.T. Introductory Physics Series by Giraudon/The Bridgeman Art Library. 3.2 Jean-Charles Cuillandre (CFHT); 5.21c
A. P. French. ©1968, 1966 by Massachusetts SSPL/The Image Works. 3.3a Science Photo Junenoire Photography. 5.22a Courtesy of
Institute of Technology. Used by permission Library. 3.4 Bettmann/Corbis. 3.5 Tnuç Ed Grafton; 5.22b Steve Larson/University
of W. W. Norton & Company, Inc; 1.9 center Tezel. 3.6 The Granger Collection, NY. 3.12 of Arizona. 5.24a Roger Ressmeyer/Corbis;
Seth Joel/Corbis; 1.9 bottom left Musée Lebrecht Music and Arts Photo Library/ 5.24b David Parker/Photo Researchers,
d’Orsay, Paris, France. Photo: Réunion Alamy. 3.13 Portrait of Galileo Galilei Inc. 5.25 Photo by Dave Finley, courtesy
des Musées Nationaux/Art Resource, (1564–1642) (oil on canvas), Sustermans, National Radio Astronomy Observatory
NY. Photograph by Herve Lewandowski; Justus (1597–1681) (school of)/Galleria and Associated Universities, Inc. 5.26 ESO.
1.9 bottom right Robbie Jack/Corbis. 1.10 Palatina, Palazzo Pitti, Florence, Italy/The 5.27 NASA/NSSDC. 5.28 a, b NASA/JPL.
Bettmann/Corbis. 1.11a Howard Bond Bridgeman Art Library. 3.18 Courtesy of 5.29 U.S. Department of Energy. 5.30 CERN.
(Space Telescope Science Institute), Robin Alan Bean. 5.31 Frank Summers, Space Telescope
Ciardullo (Pennsylvania State University) Science Institute; Chris Mihos, Case
and NASA; 1.11b NASA and The Hubble Western Reserve University; Lars Hernquist,
Heritage Team (STScI/AURA). 1.12 Four Chapter 4 Harvard University.
seasons photographs, Jim Schwabel/ Opener David Malin. 4.7 Eyewire
Index Stock Imagery/PictureQuest. 1.13 Collection/Getty Images. 4.14b Tom
NON SEQUITUR by Wiley Miller. Dist. by Pantages. 4.24 NASA/NSSDC/GSFC. Chapter 6
Universal Press Syndicate. Reprinted with Opener Caltech/NASA/JPL. 6.2a STScI
permission. All rights reserved. 1.14 Craig photo: Karl Stapelfeldt (JPL); 6.2 b–d
Lovell/Corbis. Chapter 5 D. Padgett (IPAC/Caltech), W. Brandner
Opener NASA/JPL. 5.1 Junenoire (IPAC), K. Stapelfeldt (JPL) and NASA. 6.3
Photography. 5.2 Gianni Tortoli/Photo Photograph by Pelisson, SaharaMet. 6.5
Chapter 2 Researchers, Inc. 5.3 Jim Sugar/Corbis. 5.4a Reuters/Corbis. 6.12 NASA/Johns Hopkins
Opener Anthony Ayiomamitis (http://www. Yerkes Observatory photographs: Richard University Applied Physics Laboratory/
perseus.gr). 2.1 British Library. 2.8 left Dreiser; 5.4b W. M. Keck Observatory/ Carnegie Institution of Washington. 6.13
Pekka Parviainen/Photo Researchers, Inc.; CARA. 5.5 ESO. 5.8b ASU Physics NASA, ESA, J. R. Graham and P. Kalas
2.8 right David Nunuk/Photo Researchers, Instructional Resource Team. 5.9b ASU (University of California, Berkeley),
Inc. 2.14a Larry Landolfi/Photo Research; Physics Instructional Resource Team. 5.11 and B. Matthews (Hertzberg Institute of
C-1
C-2 Credits
Astrophysics). 6.16a C. Marois, National 8.18 Eric Nguyen/Jim Reed Photography/ New Hope ©1977 & 1997 Lucasfilm Ltd.
Research Council Canada; 6.16b ESO/A.-M. Corbis. 8.19 NASA/JPL. 8.21 NASA. 8.22a & TM, all rights reserved. 11.19 NASA/
Lagrange et al. 6.17 NASA, ESA, P. Kalas, J. Craig Aurness/Corbis; 8.22b Philip James JPL/Caltech. 11.20 NASA/JPL/DLR. 11.22
Graham, E. Chiang, E. Kite (University of Corwin/Corbis. 8.23 NASA/JPL/Caltech. NASA/JPL/Caltech. 11.23 NASA/JPL/
California, Berkeley), M. Clampin (NASA 8.24 NASA/JPL/Caltech. 8.25a NASA/JPL/ Caltech. 11.24 Edward Kinsman/Photo
Goddard Space Flight Center), M. Fitzgerald Caltech; 8.25b NASA/JPL-Caltech/MSSS. Researchers, Inc. 11.25 NASA/JPL/Caltech.
(Lawrence Livermore National Laboratory), 8.26 NASA/STScI. 11.26a NASA/JPL/Caltech; 11.26b NASA/
and K. Stapelfeldt and J. Krist (NASA Jet JPL/Caltech. 11.27 NASA/JPL/Caltech.
Propulsion Laboratory). 11.28a NASA/JPL/Caltech; 11.28b NASA/
Chapter 9 JPL/Space Science Institute. 11.29a NASA/
Opener NASA/JPL/Space Science Institute. JPL/Space Science Institute; 11.29b NASA/
Chapter 7 9.1a NASA/JPL/University of Arizona; 9.1b JPL/Caltech. 11.30a, b, c NASA/JPL/Space
Opener Graeme L. White & Glen Cozens NASA and E. Karkoschka (University of Science Institute. 11.31 NASA/JPL/Space
(James Cook University). 7.1 NASA. 7.2 Arizona). 9.2a NASA/JPL; 9.2b NASA/JPL. Science Institute.
Stockli, Nelson, Hasler, Goddard Space 9.3 NASA/JPL/Caltech. 9.6 NASA/STScI. 9.7
Flight Center/NASA. 7.3 USGS Hawaiian CICLOPS/NASA/JPL/University of Arizona.
Volcano Observatory. 7.4b NASA/JSC; 7.4c 9.8a NASA/JPL/Caltech. 9.9a NASA/JPL/ Chapter 12
NASA/Johns Hopkins University Applied Space Science Institute; 9.9a inset NASA/ Opener Jerry Lodriguss/Photo Researchers,
Physics Laboratory/Carnegie Institution JPL/Space Science Institute; 9.9b NASA/ Inc. 12.1a, b Dr. R. Albrecht, ESA/ESO
of Washington. 7.5 a–f Montes De Oca & JPL. 9.10a, b NASA, L. Sromovsky, and P. Space Telescope European Coordinating
Associates. 7.6 Photograph by D. J. Roddy Fry (University of Wisconsin-Madison). Facility/NASA. 12.3 M. Brown (Caltech),
and K. A. Zeller, USGS, Flagstaff, AZ. 9.11 NASA/JPL. 9.12a NASA/JPL/Caltech; C. Trujillo (Gemini), D. Rabinowitz (Yale),
7.7 Don Davis. 7.8 NASA/JSC. 7.9 NASA/ 9.12b NASA/JPL/Caltech. 9.15 NASA, NSF, NASA. 12.4 NASA, ESA, J. Parker
JPL/Caltech. 7.10 Dennis Flaherty/Photo ESA, L. Sromovsky and P. Fry (University (Southwest Research Institute), P. Thomas
Researchers, Inc. 7.11 NASA/JPL/Caltech. of Wisconsin), H. Hammel (Space Science (Cornell University), and L. McFadden
7.15 Joe Tucciarone/Science Photo Library/ Institute), and K. Rages (SETI Institute). (University of Maryland, College Park).
Photo Researchers, Inc. 7.16 Grant Heilman 9.21 N. M. Schneider, J. T. Trauger, Catalina 12.5 From Meteorites and Their Parent
Photography. 7.17 Photo courtesy of Ron Observatory. 9.22a J. Clarke (University of Planets, H. Y. McSween, Cambridge.
Greeley. 7.22 NASA/JSC. 7.23 NASA/ Michigan), NASA; 9.22b Courtesy of John 12.6a, b, c, d Courtesy of Ron Greeley.
Johns Hopkins University Applied Clark, Boston University and NASA/STScI. 12.7 NASA/JPL/Cornell. 12.10 NASA/
Physics Laboratory/Carnegie Institution JPL/Caltech. 12.11 NEAR Project, NLR,
of Washington. 7.24a ESA/DLR/FU Berlin JHUAPL, Goddard SVS, NASA. 12.12a
(G. Neukum); 7.24b NSSDC/NASA. 7.25 Chapter 10 NASA/JPL/University of Arizona; 12.12b
NASA/Magellan Image/JPL. 7.27 NASA/JSC. Opener NASA/JPL. 10.7 a, b Christopher ESA/DLR/FU Berlin (G. Neukum). 12.14b
7.28 K. C. Pau. 7.29 NASA/Johns Hopkins Mackay. 10.12 NASA/JPL/Caltech. 10.13a Courtesy of Terry Acomb. 12.15a Dr Robert
University Applied Physics Laboratory/ NASA, ESA, H. Weaver, and T. Smith McNaught; 12.15b Don Goldman. 12.16a
Carnegie Institution of Washington. 7.30 (STScI); 10.13b New Mexico State University NASA/NSSDC/GSFC; 12.16b HST/WFPC2
ESA/DLR/FU Berlin (G. Neukum). 7.31a Astronomy Department. 10.14 NASA/ H. Weaver (ARC), NASA. 12.17 NASA/JPL/
NASA/JPL/Malin Space Science Systems; AURA. 10.17 NASA, ESA and E. Karkoschka Caltech. 12.18 NASA/JPL/Caltech/UMD.
7.31b NASA/MOLA Science Team. 7.32 (University of Arizona). 12.19 NASA/JPL/Caltech/UMD. 12.21
NASA/JPL-Caltech/University of Arizona/ Roger Lynds/NOAO/AURA/NSF. 12.23
Texas A&M University. 7.33 HiRISE, MRO, NASA. 12.24a NASA/JPL/MPIA; 12.24b
LPL (U. Arizona), NASA. 7.34 a, b NASA/ Chapter 11 NASA/STScI; 12.24c R. Evans, J. T. Trauger,
JPL/Caltech. 7.35 NASA/JPL/Caltech, Opener NASA/JPL/Space Science Institute/ H. Hammel and the HST Comet Science
courtesy Alfred McEwen, USGS; 7.35 Photo Researchers, Inc. 11.1 NASA/JPL/ Team, and NASA. 12.25 Courtesy of the
inset Malin Space Science Systems, MGS/ Caltech. 11.3 NASA/JPL/Space Science Wolbach Library, Harvard-Smithsonian
JPL/NASA. 7.36a NASA/JPL/Malin Space Institute. 11.4 NASA/JPL/Caltech. 11.5 Center for Astrophysics, Cambridge, MA.
Science Systems; 7.36b NASA/JPL/Cornell. NASA/JPL/Caltech. 11.6 NASA/JPL/ 12.26b Barrie Rokeach/Getty Images. 12.27
7.37 ESA/DLR/FU Berlin (G. Neukum). 7.38 Caltech. 11.7 NASA/JPL/Space Science Armagh Observatory. 12.28 Tony Hallas/
NASA/JPL-Caltech/University of Arizona/ Institute. 11.8a NASA/JPL/Space Science Science Faction. 12.29 Courtesy of Joe
Texas A&M University. 7.39 NASA/JPL- Institute; 11.8b NASA/JPL/Space Science Orman. 12.30 NASA.
Caltech/UTA/UA/MSSS/ESA/DLR/JPL Institute.11.9 NASA/JPL/Caltech. 11.10
Solar System Visualization Project. NASA/JPL/Caltech. 11.11 NASA/JPL/
Caltech. 11.12 NASA/JPL/Space Science Chapter 13
Institute. 11.13 NASA/JPL. 11.14a NASA/ Opener Jon Lomberg/Photo Researchers,
Chapter 8 JPL/ESA/University of Arizona; 11.14b Inc.
Opener Adam Block/NOAO/AURA/NSF. NASA/JPL. 11.15 NASA/JPL/ESA/
8.1a NASA/NSSDC; 8.1b NASA/NSSDC; University of Arizona. 11.16a NASA. 11.17a
8.1c NASA/STScI. 8.9a Louis A. Frank, The NASA/JPL/Space Science Institute; 11.17b Chapter 14
University of Iowa; 8.9b Raymond Gehman/ NASA/JPL/Space Science Institute. 11.18a Opener Neil deGrasse Tyson, 2001.
Corbis. 8.14 Lyndon State College/Nolan NASA/JPL/Caltech; 11.18b Courtesy of 14.9a Brookhaven National Laboratory;
Atkins. 8.16 Peter Arnold. 8.17 NOAA. Lucasfi lms Ltd, Star Wars: Episode IV—A 14.9b Courtesy of Tomasz Barszczak,
Credits C-3
University of California, Irvine, for the Heritage Team (STScI/AURA); 16.12 center AURA)-ESA/Hubble Collaboration. 19.5
Super-Kamiokande Collaboration; 14.9c left A. Hajian (USNO) et al., Hubble Heritage NOAO/AURA/NSF. 19.6 David Malin.
Lawrence Berkely National Laboratory/ Team (STScI/AURA), NASA; 16.12 bottom 19.7 NASA and the Hubble Heritage
Photo Researchers, Inc.; 14.9d Evelyn left J. P. Harrington and K. J. Borkowski Team (STScI/AURA). 19.9a, b European
Malkus/ IceCube Research. 14.10 NOAO/ (University of Maryland), HST, NASA; Southern Observatory (ESO). 19.10a, b
AURA/NSF. 14.11 SOHO/ESA/NASA. 16.12 top right NASA, A. Fruchter and Johannes Schedler/Panther Observatory.
14.13 Nigel Sharp, NOAO/NSO/Kitt Peak the ERO team (STScI); 16.12 center right 19.11a NASA/JPL-Caltech; 19.11b John
FTS/AURA/NSF. 14.14a Stefan Seip, www. Anglo-Australian Telescope, photograph Gleason, Celestial Images. 19.12a Todd
astromeeting.de; 14.14b, c 2001 by Fred by David Malin; 16.12 bottom right Bruce Boroson/NOAO/AURA/NSF; 19.12b Richard
Espenak, courtesy of www.MrEclipse. Balick (University of Washington), Vincent Rand, University of New Mexico. 19.14
com. 14.15 NASA/Photo Researchers, Inc. Icke (Leiden University, The Netherlands), image100/SuperStock. 19.15 David Malin.
14.16 SOHO/ESA/NASA. 14.19 SOHO/ Garrelt Mellema (Stockholm University), 19.17 X-ray: NASA/CXC/Penn State/G.
ESA/NASA. 14.20 T. Rimmele (NSO), and NASA. Garmire; Optical: NASA/ESA/STScI/M.
M. Hanna (NOAO)/AURA/NSF, Göran West. 19.18 top left to right Courtesy of
Scharmer/Institute for Solar Physics of Dr. Michael Brotherton; 19.18 bottom left
the Royal Swedish Academy of Sciences, Chapter 17 and center left NASA, ESA, ESO, Frédéric
Sweden. 14.22d SOHO/ESA/NASA. 14.24a Opener NASA/HST/ASU/J. Hester. 17.7 Courbin (Ecole Polytechnique Federale de
SOHO/ESA/NASA; 14.24b SOHO-EIT N. Smith, J. A. Morse (U. Colorado) et Lausanne, Switzerland) & Pierre Magain
Consortium, ESA, NASA. 14.25a SOHO/ al., NASA. 17.10a, b Anglo-Australian (Universite de Liege, Belgium); 19.18 bottom
EIT; 14.25b SOHO/LASCO/ESA and Observatory, photographs by David Malin. right and center right HST. 19.19 NRAO/
NASA. 14.26 SOHO/ESA and NASA. 17.11a Joachim Trumper, Max-Planck- Alan Bridle et al. 19.22b Courtesy Bill
Institut für extraterrestrische Physik; 17.11b Keel, University of Alabama, and NASA;
Courtesy Nancy Levenson and colleagues, 19.22c Holland Ford, STScI/Johns Hopkins
Chapter 15 American Astronomical Society; 17.11c Jeff University, and NASA; 19.22d R. P. van
Opener Canada-France-Hawaii Telescope/ Hester, Arizona State University, and NASA. der Marel, STScI, F. C. van den Bosch,
J.-C. Cuillandre/Coelum. 15.1a Image 17.15a European Southern Observatory, University of Washington, and NASA. 19.23
courtesy of Dr. A. Mellinger; 15.1b NASA/ ESO PR Photo 40f199; 17.15b, c Jeff Hester, F. Owen, NRAO, with J. Biretta, STScI,
NSSDC/GSFC Diffuse Infrared Background Arizona State University, and NASA. 17.31 and J. Eilek, NMIMT. 19.24 Holland Ford,
Experiment (DIRBE) instrument on the ESA/Hubble. STScI/Johns Hopkins University; Richard
Cosmic Background Explorer (COBE). Harms, Applied Research Corp.; Zlatan
15.4a NSSDC/GSFC Infrared Astronomical Tsvetanov, Arthur Davidsen, and Gerard
Satellite (IRAS); 15.4b JAXA. 15.5 NSSDC/ Chapter 18 Kriss at Johns Hopkins University; Ralph
GSFC Roentgen Satellite (ROSAT). 15.6 Opener NASA/WMAP Science Team. Bohlin and George Hartig at STScI; Linda
Dr. Douglas Finkbeiner. 15.7a Courtesy of 18.1 NASA, ESA, S. Beckwith (STScI) Dressel and Ajay K. Kochhar at Applied
Stefan Seip; 15.7b Courtesy of Emmanuel and the HUDF Team. 18.2 Jason T. Ware/ Research Corp. in Landover, MD; and Bruce
Mallart www.astrophoto-mallart.com Photo Researchers, Inc. 18.4 TIMEPIX/ Margon from the University of Washington,
& www.axisinstruments.com. 15.8 ESO Getty Images. 18.5 Courtesy of Jeff Hester. Seattle/NASA. 19.25 John Biretta, STScI.
Telescope Systems Division. 15.9 C. Jones, 18.5 NASA William C. Keel (University of 19.26 NASA, ESA, the Hubble Heritage
Smithsonian Astrophysical Observatory. Alabama, Tuscaloosa). 18.8 WIYN/NOAO/ (STScI/AURA)-ESA/Hubble Collaboration,
15.10a Canada-France-Hawaii Telescope NSF, WIYN Consortium, Inc. 18.17 Lucent and A. Evans (University of Virginia,
& Coelum; 15.10b, c ESO. 15.15 Jeff Technologies’ Bell Labs. 18.18 Ann and Rob Charlottesville/NRAO/Stony Brook
Hester and Paul Scowen (Arizona State Simpson. 18.20 a, b, c NASA COBE Science University).
University), Bradford Smith (University Team.
of Hawaii), Roger Thompson (University
of Arizona), and NASA. 15.19a inset Karl Chapter 20
Stapelfeldt/JPL/NASA; 15.19b NASA/ Chapter 19 Opener Tony Rowell/Astrophotostore.
JPL-Caltech/UIUC. 15.20a, b, c, d Jeff Opener NASA/ESA/STScI/Photo com. 20.1a Axel Mellinger; 20.1b C. Howk
Hester (Arizona State University), WFPC2 Researchers, Inc. 19.2 top left NASA; 19.2 and B. Savage (University of Wisconsin);
Team, NASA. 15.21a Anglo-Australian center left NASA; 19.2 bottom left NASA. N. Sharp (NOAO)/WIYN, Inc. 20.2a
Observatory/Royal Observatory, 19.2 top right Canada-France-Hawaii US Patent Office Library; 20.2b NASA/
Edinburgh, photograph from UK Schmidt, Telescope/J.-C. Cuillandre/Coelum. 19.2 JPL-Caltech. 20.3 Hubble Heritage Team
plates by David Malin; 15.21b NASA/JPL- center right NASA; 19.2 bottom right NASA. (AURA/STScI/NASA). 20.4a NOAO/
Caltech/J. Stauffer (SSC/Caltech). 19.3 The Hubble Heritage Team (AURA/ AURA/NSF. 20.5 Courtesy of Jeff Hester.
STScI/NASA). 19.4 top left NASA, ESA, and 20.8 Anglo-Australian Observatory. 20.11
the Hubble Heritage Team (STScI/AURA); The Electronic Universe Project. 20.16a
Chapter 16 19.4 top right NASA, ESA, A. Aloisi (STScI/ NASA/CXC/MIT/F. K. Baganoff; 20.16b
Opener NASA, ESA and H. E. Bond (STScI). ESA), and the Hubble Heritage (STScI/ Hubble, NASA, ESA, & D. Q. Wang (U.
16.11a Minnesota Astronomical Society; AURA)-ESA/Hubble Collaboration; 19.4 Mass, Amherst); Spitzer: NASA, JPL, & S.
16.11b NASA and The Hubble Heritage bottom left NASA, ESA, and the Hubble Stolovy (SSC/Caltech). 20.17 This image
Team (STScI/AURA). 16.12 top left Dr. Heritage (STScI/AURA)-ESA/Hubble was created by Prof. Andrea Ghez and
Raghvendra Sahai (JPL) and Dr. Arsen R. Collaboration; 19.4 bottom right NASA, her research team at UCLA and is from
Hajian (USNO), NASA and The Hubble ESA, and the Hubble Heritage Team (STScI/ data sets obtained with the W. M. Keck
C-4 Credits
Telescopes. Image creators include Andrea Chile–2008. 22.3a Anglo-Australian

Ghez, Angelle Tanner, Seth Hornstein, and Observatory; 22.3b Omar Lopez-Cruz and Chapter 23
Jessica Lu. 20.18 NRAO/AUI. 20.19 NASA/ Ian Shelton/NOAO/AURA/NSF. 22.4a, b Opener SETI/NASA. 23.1 Jim Sugar/
MIT/F.Baganoff. 20.20 Anglo-Austrailian Max Tegmark/SDSS Collaboration; 22.4c Corbis. 23.2a Dr. Michael Perfit, University
Observatory, David Malin. Jordan Raddick/SDSS. 22.7a NASA/CXC/ of Florida, Robert Embley/NOAA; 23.2b
MIT/E.-H Peng et al.; Optical: NASA/ Woods Hole Oceanographic Institution,
STScI; 22.7b NASA; ESA; L. Bradley Deep Submergence Operations Group,
Chapter 21 (Johns Hopkins University); R. Bouwens Dan Fornari. 23.3 Gaertner/Alamy.
Opener NASA/JPL/Origins. 21.4 top NASA, (University of California, Santa Cruz); H. 23.4 Chris Boydell/Australian Picture
ESA, The Hubble Key Project Team, and The Ford (Johns Hopkins University); and G. Library/Corbis. 23.6 Courtesy of National
High-Z Supernova Search Team; 21.4 bottom Illingworth (University of California, Santa Science Foundation. 23.7 top left Mark
Smithsonian Astrophysical Observatory, Cruz); 22.7c NASA/CXC/MIT/E.-H Peng et A. Schneider/Visuals Unlimited; 23.7
1993; 21.4 center NASA/WMAP Science al.; NASA/STScI. 22.8b NASA, A. Fruchter center middle left K. Sandved/Visuals
Team. 21.5 Robert Kirshner, Harvard and the ERO team (STScI, ST-ECF), NASA, Unlimited; 23.7 bottom left Ken Lucas/
University. 21.14 Kamioka Observatory, ESA; 22.8c Faure (Zentrum Für Astronomie, Visuals Unlimited; 23.7 top right Science
ICRR (Institute for Cosmic Ray Research), University of Heidelberg) and J. P. Kneib VU/National Museum of Kenya/Visuals
The University of Tokyo. 21.15 From The (Laboratoire d’Astrophysique de Marseille). Unlimited; 23.7 middle right Ken Lucas/
Elegant Universe by Brian Greene. ©1999 by 22.9 NASA/CXC/CfA/M. Markevitch et al.; Visuals Unlimited; 23.7 bottom right Ken
Brian R. Greene. Used by permission of W. and NASA/STScI; Magellan/U. Arizona/D. Lucas/Visuals Unlimited. 23.8 Kenneth
W. Norton & Company, Inc Clowe et al.; Lensing Map: NASA/STScI; E. Ward/Biografx/Photo Researchers,
ESO WFI; Magellan/U. Arizona/D. Clowe Inc. 23.9 Jeff J. Daly/Visuals Unlimited.
et al. 22.10 Images produced by Donna 23.10 Kepler Mission NASA. 23.11 NASA.
Chapter 22 Cox Smithsonian Institution and Motorola 23.12 NASA. 23.13 F. Drake (UCSC) et al.,
Opener Robert Gendler, Jan-Erik Ovaldsen, Corporation. 22.11 NASA, ESA, and N. Arecibo Observatory (Cornell, NAIC). 23.14
Allan Hornstrup, IDA Image data: ESO/ Pirzkal (STScI/ESA). 22.13 NASA, ESA, and Twentieth Century Fox Film Corporation/
Danish 1.5m telescope at La Silla, A. Feild (STScI). Photofest. 23.15 Seti/ NASA.
Index
This index contains page references for all 23 perspective on, 553, 554–56, 554 of molecular clouds, 426–27, 428, 428
chapters in the complete volume of 21st Century size of, 551–52, 552 angular resolution, 142
Astronomy, Third Edition. The Solar System unified model of, 552–58, 553 annular solar eclipse, 46, 47, 47, 49
Edition does not include Chapters 15–22 (pp. see also quasars Antarctica, 175, 307, 332–33
416–629). The Stars and Galaxies Edition does Adams, Fred, 626, 626 Antarctic Circle, 40
not include Chapters 6–12 (pp. 154–359). Adams, John Couch, 249 anticyclonic motion, 254, 255, 262, 262
Adams, Ring, 317, 319, 321 antimatter, 351, 597
Page numbers in italics refer to illustrations. An adaptive optics, 135–36, 137 symmetry between matter and, 598, 601
italic n after a number refers to a footnote. Adrastea, 319 antiparticles, 394, 503, 503, 596, 597, 597
Aeschylus, 22 see also positrons
Abell 2218 (galaxy cluster), 617 AGB, see asymptotic giant branch (AGB) star aperture, 125, 132, 133, 137, 142
aberration of starlight, 34–36, 35, 81 AGNs, see active galactic nuclei aphelion, 347–48
absolute magnitude, 365, 379 (table) air pressure, Coriolis effect and, 232 Apollinaire, Guillaume, 562
absolute zero, 113 AKARI, 420, 421 Apollo missions, 175, 180, 182, 186, 186, 188,
absorption, 104–7, 104, 422 Alabama, meteorite crash in, 181 195, 200, 284
absorption lines, 105, 110, 113, 368–71, 374, Alaska, troposphere in, 225 Apollo 10, 200
384, 546, 549 albedo, 118 Apollo 15, 8, 204
defi ned, 105 algebra, defi ned, 15, 16 Apollos, 336, 337
interstellar, 422 Allen, Paul, 646 apparent daily motion:
as spectral fi ngerprints of atoms, 106–7 Allen Telescope Array (ATA), 646 latitude and, 23, 28–32, 29, 30, 31, 32
in spectrum of Sun, 113, 402–4, 404 Alpha Canis Majoris, see Sirius of stars, 23, 25–32, 27, 28, 30, 31, 32–33,
star classification and, 368–70, 369, 370 Alpha Centauri, 365, 518 32, 39, 51–52, 53
star composition and, 368–71, 422 alpha particles, 453 of sun, 26, 34
acceleration, 373 Alpher, Ralph, 525–26 view from the poles of, 26, 27–28, 28
defi ned, 68, 69 altitude: apparent magnitude, 365
gravitational, 73, 78, 80, 85, 279–80, 279, atmospheric pressure and, 215, 257, 259 Aquarius, 33, 41n
285, 449, 456, 492, 492 temperature and, 225, 226, 227, 229, 257, Arab culture, 12
miniature solar system model of atom 259 archaeans, 635
and, 100 Alvin, 264 archaeology, sky patterns and, 24
Newton’s second law of motion and, Amalthea, 319 arcminute, 134
68–70, 69, 72 American Astronomical Society, 528 arcsecond, 134n, 135
accretion, 217 amino acids, 307, 633, 640 Arctic Circle, 40
accretion disks: ammonia, 166, 344, 633 Arecibo radio telescope, 137, 142, 142, 643
in binary system, 462, 482, 482 giant planets and, 252, 259, 265 argon, 106, 400, 424
defi ned, 161 terrestrial planets and, 217, 219 Ariel, 300, 312
formation of, 157, 158, 160–67, 162, 175, ammonium hydrosulfide, 252, 259 Ariel 1 satellite, 144
176, 428, 428, 443, 462, 575, 614–16 Amors, 336 A Ring, 314, 316, 321–22
in galaxy formation, 614–16 amplitude, of wave, 96, 96 Aristotle, 25, 67, 339
around supermassive black holes, 552–56, anaerobes, 224n Arizona, Meteor Crater in, 183, 183
553 Andromeda Galaxy, 4, 511, 543, 581, 608 Artemis, 43, 283
around white dwarfs, 462, 552 Andromeda Nebula, 511 artillery targets, Earth’s rotation and, 231,
see also protoplanetary disks; protostellar angle of incidence, 130, 131 232
disks angle of reflection, 130, 131 arts, science compared with, 3, 10
achondrites, 333–35, 334 Anglo-Australian Telescope, 610 Asimov, Isaac, 566
active comets, 340, 341–42, 342 angular momentum, 100, 158–60, 167, 462, A stars, 369, 473, 475
active galactic nuclei (AGNs), 535, 549–59, 502 asteroid belt, 169
618 law of conservation of, 158–60, 164, 175, Kirkwood gaps in, 289, 290, 320, 336
normal galaxies and, 556–57 236, 285, 426–27, 428, 428, 482, 615 main, 299, 327, 336, 337
I-1
I-2 Index
asteroids, 4, 7, 145, 211, 251, 320, 328, 331, comets and, 215–18, 339, 351, 357 Bay of Fundy, tides in, 281–82, 282
332–39, 358, 443 distortion caused by, 135–37, 135, 144 bB/bv color, 368
Amors, 336 electromagnetic radiation blocked by, Bean, Alan, 74
Atens, 336 136, 137, 144, 421 Bell Laboratories, 139, 141
Apollos, 336 impacts and, 184 beryllium, 5, 459, 480, 529, 601
close-up view of, 337–39, 338 layers of, 225, 226, 227–28 beryllium-8 nucleus, 453
comet nuclei compared with, 341 meteorites and, 332–33 Bessel, F. W., 365
C-type, 334, 338 orbital decay and, 80n beta decay, 394, 596, 627
defi ned, 332 primary, 168, 216 Beta Pictoris, 173
density of, 336, 338, 353–54 role of life in, 215, 222–25, 224, 244 Bethe, Hans, 525–26
discovery of, 331 secondary, 168, 216–21, 223, 243 Big Bang theory, 501, 509, 518–31, 521, 572,
as fractured rock, 336 solar activity and, 411–12 600, 601, 602, 603, 607, 611, 613, 624
impacts and, 184–85, 335–36, 350–53 weather and, 229–32, 230, 234–38 abundance of the least massive elements
key concepts about, 327 wind and, 215, 229–32, 230, 234–37 predicted by, 529–30, 530
landslides on, 207 atmospheric distortion, 125, 135–37, 135, center of, 523
mass of, 336 144 confi rmation of predictions of, 509, 525–
meteorites related to, 332–36, 335 atmospheric greenhouse effect, 215, 219–21, 30, 526–630
M-type, 335 219, 220–21, 263, 306 expanding universe, 585, 589, 623
near-Earth, 336, 337, 338 on Venus, 219, 220–21, 238, 244, 263 nucleosynthesis and, 529–30, 530, 601,
orbital resonances and, 275, 289, 290 atmospheric probes, 146, 147 602, 603
orbit of, 287, 289, 290, 332, 336, 337, 352 atmospheric windows, 137, 137, 141 problems with, 592–95, 592, 594
S-type, 334–35, 337–38 atomic bomb, impacts compared with, 183, radiation left over from, 525–27, 526, 527,
Trojan, 293, 293, 336 184–85 529
astrobiology, 7, 641 atoms, 72n, 99–108, 121, 222–23, 601 spacetime and, 501, 509, 588, 591, 601
astrology, 24, 25, 33, 51–52 Bohr model of, 100 “Big Crunch,” 590, 626
astronauts, 229 decay of, 102–5, 103, 106, 107–8 Big Dipper (Ursa Major of the Great Bear), 24
Apollo, 74, 175, 180, 182, 186, 186, 188, in degenerate cores, 443 “Big Rip,” 590, 590, 626
195, 204 emission and absorption determined by binary pulsars, 149
in free fall, 80, 80 energy levels of, 102–5, 103, 104, 105 binary stars, 373–77
Newton’s laws of motion and, 72, 73 empty space in, 448 eclipsing, 376–77, 376, 377, 384
astronomical seeing, 134–37 energy state of, see energy states, atomic evolution of, 443, 459, 461–64, 462, 463,
astronomical unit (AU), 36, 58, 64, 92, 168, ground state of, 102, 102 481–82
330, 340–41, 363, 407 in human body, 3, 5, 7 formation of, 438
astronomy: ideal gas law and, 222–23 mass transfer and, 461–64, 462, 463
amateur, 125, 140, 140 interaction of light with, 99–108 neutron stars in, 481–82, 482, 487–88,
defi ned, 4 isolated vs. molecular, 424 501, 552
overview of, 3–7 miniature solar system model of, 100, 101 orbits of, 361, 373–74, 374, 375, 461, 501,
tools of, 124–53, 170 in neutrino detectors, 399–400 504
see also specific topics pressure of gas and, 222–23, 223 Roche lobes of, 288, 461–79, 482, 482
astrophysics, 85, 86 in primary atmosphere, 216 tidal locking of, 288
asymptotic giant branch (AGB) star, 460, in protoplanetary disk, 164 total mass of, 375–76
460, 464–65, 471, 472, 486 spectral fi ngerprints of, 91, 106–7, 106 visual, 376
mass loss in, 455–56, 457, 459 thermal motions of, 112–13, 113, 452–54, white dwarf, 459, 462–64, 463, 482, 482,
structure of, 455–56, 456 601 518
Atacama Desert, 307 AU Microscopii, 170 X-ray, 469, 482, 482, 504, 552
Atens, 336 auroras, 215, 229, 229, 247, 270–71, 270, 412 binding energy, 476, 476
atmosphere: Australia Telescope Compact Array, 178 biosphere, 180
escape of, 216–19, 244 authority, challenging of, 67 bipolar outflow, 435, 436
formation of, 155, 167, 215–18, 217, 244 autumn, 39 blackbody, 114, 120
of giant planets, 167, 216, 247, 252, 254– autumnal equinox, 39 blackbody spectrum, see Planck (blackbody)
63, 254, 255, 256, 257, 258, 261, 271 axions, 613 spectrum
key concepts about, 215 Aztecs, 43 Black Hole Era, 626, 627
of moons, 300, 303, 304–9, 308 black holes, 469, 482, 501–4, 546, 549, 558–
primary, 167, 168, 216, 218, 244 backlighting, 317, 317, 318, 322 59, 570, 627
of protostars, 433, 449 bacteria, 307, 635 defi ned, 488
secondary, 167, 168, 216–21, 223, 244 “ball in the car” experiment, 231, 231, 488, event horizon of, 502, 503, 524
see also specific planets 489 in Milky Way Galaxy, 563, 577–79, 579
atmosphere of Earth, 7–8, 144n, 180, 181, Baltic Sea, 281 “seeing,” 503
207, 215–21, 216, 221–38, 356 Bar (unit), 221 as singularity, 502, 502
atmospheric greenhouse effect and, 215, Barnard 68, 425 supermassive, 501, 504, 535, 552–59, 553,
219–21, 237–38, 238, 243, 263 basalt, 191, 203, 206 555, 577–79, 618
chemical composition of, 192, 215–17, baseball: tidal effects and, 288
218–21, 218 (table), 222–25, 224, 243, acceleration and, 70 trip into, 502–3, 502
257, 613 Coriolis effect and, 233 “blueberries,” 209
Index I-3
blue light, 367–68 carbon, 5, 107, 166, 252, 269, 319, 333, 371 Chandra X-Ray Observatory, 137, 144, 145
interstellar dust and, 419, 420 (table), 391, 481, 530, 576, 640 (table)
of quasars, 549 in comets, 344, 351 chaos, chaotic behavior, 293–95, 306, 307
scattering of, 240, 240 in interstellar medium, 418, 425 orbits and, 294–95, 295, 301
temperature and, 113–15 in planetary nebulae, 459 charge-coupled devices (CCD), 139–40, 140,
blueshifted light, 108, 109, 109, 121, 374, 528 in stellar core, 443, 453, 453, 455, 456, 457, 141, 149
body, human: 464 Charon, 285, 299, 300, 328–29, 329, 341
as stardust, 3, 7, 7 in white dwarfs, 459, 460, 464 chemical reactions, self-sustaining, 476
temperature of, 111–12 carbonaceous chondrites, 333, 338 chemosynthesis, 307
thermal equilibrium of, 111–12 carbonates, 219, 224, 331 Chile, 307
see also brain, human; eyes, human carbon burning, 393 China:
Bohr, Niels, atomic model of, 100 in high-mass stars, 472, 472 (table), 473, astronomy in, 482–83, 484
bolide, 354 475, 477 culture of, 12
Bolshevik Revolution, 42 in white dwarfs, 464 China National Space Administration
bookcase, energy state of atoms as, 102, 102, carbon dioxide, 209, 303–4, 342, 344–46, (CNSA), 145
107 633 chlorofluorocarbons (CFCs), 220, 243
boron, 5, 459, 480, 572 in atmosphere, 167, 218–21, 218 (table), CHON (carbon, hydrogen, oxygen, nitrogen),
in early universe, 529, 530, 601 219, 220, 221 (table), 224–25, 238, 238, 638, 640
Borrelly, Comet, 345–46 241, 412 chondrites, 333, 334
BO supergiant, 504 greenhouse effect and, 219–21, 219, 220, chondrules, 333
bound orbits, 84, 85 221, 238, 243 chromatic aberration, 127, 131, 132, 132
bow shock, 229 carbon monoxide, 166, 217, 344, 349, 425, chromosphere, 404, 410–11
Bradley, James, 36 544, 544 circle, defi ned, 63
Bragg, William Henry, 90 carbon-nitrogen-oxygen (CNO) cycle, 470, circular orbits, 36, 63, 79–84, 284–85, 286
Brahe, Tycho, 59, 59, 62, 339 470, 481, 505 Newton’s prediction of, 79–82, 79, 81
Braille, 338 carbon stars, 459 of stars, 539
brain, human: car speeds: circular symmetry, 278
color and, 367–68 acceleration and, 67–69, 68 circular velocity, 81–83
limitations of, 100, 120 direction of rainfall and, 34–35, 35 circumpolar portion of sky, 31, 31, 40
stereoscopic vision and, 362, 362 frame of reference and, 34, 231, 231, 488 circumstellar disk, 170, 417
brightness: relative motions and, 34, 35, 231, 488 Clarke, Arthur C., 175n
of CCD images, 149 Carter, Jimmy, 643 classical mechanics, see Newtonian
defi ned, 114, 116, 366 Cascade Range, 211 (classical) mechanics
distance and, 117, 366, 366 Cassini, Jean-Dominique, 313 classical planets, 4, 7, 65, 65, 119, 145, 155,
luminosity vs., 113–14, 116 Cassini Division, 291, 291, 313, 314, 316, 316 179, 250
of stars, 138, 171, 171, 361, 366–67, 366 Cassini mission, 146, 146, 246, 248, 250, 255, Clementine mission, 210
surface, 138 256, 298, 304, 304, 305, 308, 308, 311, climate, 181, 224, 225–26, 262–63, 271–72,
B Ring, 314, 316, 322, 322 316, 324 283
brown dwarfs, 170, 172, 433, 439, 627 Catalina Observatory, 140 defi ned, 237
Bruno, Giordano, 67 catalysts, 225 latitudinal changes in, 38, 52
B stars, 369, 444, 475, 543 Catholics, 42 modification of, 215, 220, 237–38, 238, 243
bubbles, atmospheric, 135, 135 Cat’s Eye Nebula, 458, 458 seasonal change in, 36–38, 39–40, 52
bulge, 539 cause-and-effect relationships vs. quantum solar activity and, 412
Bullet Cluster of galaxies, 617, 618 probabilities, 107 clocks, 92
butane, 344 CBR, see cosmic background radiation cratering, 179, 186, 188
CCD detectors, 608–9 in Einstein’s thought experiment, 490,
calcium, 334, 405 celestial equator, 27, 29, 30, 31–32 491
calculus, 15, 79n, 82 defi ned, 26 relativity theory and, 490, 491, 492–93,
calendars, 42 seasons and, 36–37, 39 499, 500
California, San Andreas Fault in, 198 wobbling of, 42 see also time
Callisto, 169, 300, 307, 312–13 celestial poles, 26–32, 27, 29, 30 closed universe, 591, 592
Caloris Basin, 204 celestial sphere, 25 clouds:
Cambrian explosion, 636, 643 as useful fiction, 26, 27, 32 cumulonimbus, 234, 236
Canada, 194, 281–82, 282 center of mass, 373, 374 cumulus, 234
canali, 640 centripetal force, 81, 81 of giant planets, 247, 252, 254–59, 254,
Cancer, 41 Cepheid variables, 473, 474, 505, 511, 565, 257, 258, 260, 272
Canis Major, 24 567 interstellar, see interstellar clouds;
Cannon, Annie Jump, 369, 384 Cepheus, 416 molecular clouds
cannonball: Ceres, 64 (table), 156, 169, 327, 330–31, 330, see also intercloud gas
Coriolis effect and, 232–33, 232 331, 340 Clown Nebula, 458
law of gravity and, 75 CERN (European Organization for Nuclear CNO cycle, see carbon-nitrogen-oxygen
in Newton’s thought experiment, 79–82, Research), 148, 148 (CNO) cycle
79, 85, 558 Chandrasekhar, Subrahmanyan, 464 cobalt, 640
Capricorn, 41 Chandrasekhar limit, 464, 487, 504, 518 COBE, see Cosmic Background Explorer
I-4 Index
coins, compared to galaxies, 535–36, 536 extreme environments and origin of life, cork, bobbing of, electromagnetic waves
cold, cooling, 111–13, 229–30 307 compared to, 95, 95
defi ned, 112 general relativity and structure of the corona, 389, 403, 404–6, 405, 410–11, 411,
in evolution of planetary interiors, universe, 501 413
192–94 gravitation, 498 coronal holes, 405, 406
subjective experience of, 112 inverse square laws, 76 coronal mass ejections, 389, 411, 411
of universe, 601, 602, 603 life, the universe, and everything, 639 CoRoT, 643
cold dark matter, 613, 613 lunar tides and origin of life, 283 Cosmic Background Explorer (COBE), 528–
colliders, high-energy, 147–48, 148 parallels, between galaxy and star 29, 528, 529, 611
collimating lenses, 141 formation, 615 cosmic background radiation (CBR), 508,
Columbus, Christopher, 25, 33 seeding the universe with new chemical 509, 525–29, 526, 527, 585, 592, 611
coma, 342, 342, 344, 344 elements, 459 COBE map of, 528, 529
Coma cluster, 2, 608, 609 tides, 288 COBE measurement of, 528, 528, 529
comets, 4, 7, 184, 211, 249, 332, 339–59, 443 conservation laws, 159 excessive smoothness of, 593
abode of, 340–41, 340 for angular momentum, 158–60, 164, 175, horizon problem and, 593–95
active, 340, 341–42, 342 236, 285, 426–27, 428, 428, 482, 615 structure of universe and, 611–13, 622
atmosphere and, 155, 167, 215–18, 244, for energy, 159, 164, 165, 431 cosmic rays, 144, 493, 493, 576–77
339, 350–51, 357 see also motion, Newton’s laws of, fi rst cosmological constant, 585, 587, 588, 588–
chaos and, 294, 295 constants of proportionality, 71 91, 621
coma of, 342, 342, 344, 344 constellations, 24, 33, 34, 361 identical to dark energy, 590
composition of, 167, 341–46, 349, 351, in human mind, 24, 24 Cosmological Evolution Survey (COSMOS),
352 see also specific constellations 616
defi ned, 169 constructive interference, 132 cosmological principle, 10, 57, 211, 370, 384,
gravity of, 343 contamination, planetary samples and, 147 509, 512–13
head of, 342, 342, 347 continental drift, 196–97 Copernicus and, 57, 85, 511–12
impacts and, 184, 217, 289, 349–53, 350– continuous radiation, 113 homogeneity and, 512–15, 512, 516
51, 353, 358 convection, 194, 201 isotropy and, 512, 513, 514
Jupiter family of, 345, 347 atmosphere and, 215, 219–20, 227, 227, Newton and, 79, 85, 512
key concepts about, 327 232–33, 234–36, 234, 247, 259, 261, 261, as testable theory, 10, 79
Kuiper Belt, 217, 340, 340, 341, 344, 347, 262, 272 cosmological redshift, 524
350, 352 in energy transport, 396, 396, 433, 436 cosmology, 512–31
mass of, 343–44 plate tectonics and, 197–98, 198, 199 Big Bang, see Big Bang theory
meteor showers and, 354–55 in stars, 396, 396, 459, 471, 471, 477 defi ned, 585
nucleus of, 217, 328, 340–49, 342, 344, thunderstorms and, 233, 234, 234, 236, Hubble as pioneer of, 513–18
349, 352, 354 261 particle physics vs., 600
Oort cloud, 217, 340, 340, 341, 344, 347, and weather, 234–37 see also expanding universe; structure of
349, 352 convective zone, 396, 396, 402, 403 the universe; universe
orbits of, 58, 83, 217, 340–45, 346–49, 348, Copernicus, Nicholas, 25, 58 61, 66, 67, 85, Crab Nebula, 468, 484, 484, 551
352, 354 86, 511, 558 Crane, Stephen, 508
planetary rings and, 299, 319 Sun-centered theory of, 57–58, 59, 85, 86, Crater, 528
shattered by tides, 275, 287, 323–24, 350, 175, 299 craters:
461 copper, 640 on asteroids, 338
short-period, 345, 347–48, 348 Cordelia, 321 impact, see impact cratering, impact
superstitions about, 331, 339–40, 357 core, 167 craters
tail of, 267, 339, 342, 342, 346, 346, 349, of Earth, 5, 190, 191, 193, 195, 202, 264 on Mars, 186, 186, 209, 209, 210
405 of Ganymede, 301, 301 secondary, 183, 183
see also specific comets of giant planets, 168, 252, 263–66, 264, creationism, 639
comparative planetology, 180 272 crescent moon, 44
competition, 637 of other terrestrial planets, 195–96, 264 Cretaceous period, 184
complex system, 294 core, stellar, 294–95, 443, 445, 446–49, 450– Cretaceous-Tertiary boundary (K-T
composite volcano, 202 55, 450, 451–54 boundary), 184, 349
compound lens, 131, 132 collapse of, 471, 477–79, 478 C Ring (Crepe Ring), 314, 320
Compton orbital telescope, 137 electron-degenerate, see electron- critical mass density, 586, 592
computers, computer models, 128, 149, 150 degenerate core crust, 264
of galaxy formation, 620–21, 621 of high-mass stars, 470–72, 475, 477–79, of Earth, 190, 191, 193, 198, 202
of galaxy-galaxy interactions, 557 481–82 cryovolcanism, 304–6, 305, 308
of stellar evolution, 478 neutron-degenerate, see neutron stars C-type asteroids, 334, 338
concave mirrors, 130, 131 of Sun, 389, 391, 392–98, 399, 413, 448 culture, 52
cones, 137 core accretion, 168 light metaphors and, 91
connections boxes: Coriolis effect, 231–33, 232 myth and legends, 24, 25, 52
comets, asteroids, and life, 352 on giant planets, 247, 259–62, 261, 262, science and, 13
Earth, the Moon, and conservation of 272 winter festivals, 39, 40
angular momentum, 285 wind patterns and, 230–32, 230, 231–33, Cummings, E. E., 606
explained, 18 247, 259–62, 261, 262, 272 cumulonimbus clouds, 234, 236
Index I-5
cumulus clouds, 234 in gas giants, 264 magnetic fields of planets as, 271, 410
Curtis, H. D., 510–11 of meteorites, 191, 333 in protostar, 436–37
cyanobacteria, 223, 635 diffraction, 132, 133, 134, 135, 142 in Sun, 410
cyclonic motion, 232, 233, 254 diffraction grating, 132, 133, 141
cyclotron, 148 diffraction limit, 134 Eagle Nebula, 429, 430
Cygnus, 504 diffuse nebulae, 509–10 Earth, 3, 4, 4, 67, 172, 174–75, 178–207, 178,
Cygnus Loop, 480 dinosaurs, extinction of, 184–85, 185, 349, 185, 211, 336–37, 349, 350–53
Cygnus X-1, 504, 504 352, 357, 636–37 atmosphere of, see atmosphere of Earth
Dione, 293, 293, 300, 310 as center of universe, 12, 25, 57, 58, 59, 60,
Dactyl, 338, 338 direct imaging, 171–72 567
daguerreotypes, 138 Discorsi e dimostrazioni mathematiche core of, 5, 190, 191, 193, 195, 202, 264
Dante Alighieri, 214, 215, 243 intorno a due nuove scienze attenenti cosmic background radiation and, 528–
dark energy, 522, 589, 590 alla meccanica (“Discourses and 29, 529
identical to cosmological constant, 590, Mathematical Demonstrations Relating cosmological principle and, 57
590 to Two New Sciences”) (Galileo), 67 craters on, 180, 183–84, 183, 184, 186
Dark Era, 626, 627 disk instability, 169 crust of, 190, 191, 193, 198
dark matter, 521, 530, 535, 546–49, 548, 559, dispersion, 131–41, 131, 141 density of, 188–91
586, 611–23, 627 distance: eclipses and, 46–50, 48, 51, 51
classes of, 613 brightness and, 116–17, 117, 366–67, 366 effects of solar activity on, 389, 405, 407,
cold, 607 from Earth to stars, 362–67, 363, 364, 378 411–12
collapse of, 613–21, 613 (table), 381 erosion on, 179, 181, 184, 186, 206–7, 211
defi ned, 547 expressed as units of time, 92–93 evolution of life on, 7, 52, 221, 224, 349,
in galaxy formation, 549, 607, 611–23, 612, distance ladder, 517–18, 517 352, 357, 417, 633–37
613, 618 distortion, atmospheric, 135–37 extinctions on, 184–85, 185, 349, 352, 358
hot, 607 diurnal motion, 60 formation of, 168, 169
in Milky Way, 563, 569–71, 570, 580, 613 DNA (deoxyribonucleic acid), 307 fossil record of, 184–85, 194, 196
searching for signs of, 615–16, 616, 617 Doppler, Christian, 108 giant planets as viewed from, 249, 250,
dark matter halo, 547–48, 548, 613, 614–18 Doppler effect, 108–10, 524, 546 250, 254
of Milky Way, 563, 570–71, 613 AGNs and, 551, 554, 555 giant planets compared with, 247, 254,
Darwin, Charles, 307 relativistic beaming, 555 257, 260, 261, 262–63, 264, 266, 271–72
daughter products, 187 Doppler shift, 108, 151, 170, 371, 424, 484, gravity of, 58, 74, 76, 79–80, 82–84, 85,
Dawkins, Richard, 637 556 157, 188, 192, 216, 275–82, 276–77, 278,
Dawn mission, 331 of binary stars, 374, 375, 376, 377 461, 492, 492, 494
day, 52, 229 defi ned, 109 impacts in shaping of, 179, 180–86, 634
length of, 37–38, 252–53, 283–84 gravitational redshift compared with, 499 interior of, 179, 188–95, 190, 195, 283,
Day the Earth Stood Still, The (1951), 646 helioseismology and, 400, 402 399–400
Dead Sea, 307 structure of Milky Way and, 563, 568 location of, 220, 221, 581
decay, atomic, 102–5, 103, 106, 304, 308 see also blueshifted light; redshifted light magnetic field of, 194–95, 195, 196–97,
spontaneous, 107 Draconids, 354 (table) 215, 228–29, 228, 266, 412
Deep Impact mission, 345, 345 Draper, John W., 138, 138 magnetosphere of, 215, 228–29, 228, 229,
Deep Space 1 mission, 338, 345 D Ring, 314 266, 407, 412, 576
Defense Department, U.S., 210 Dumbbell Nebula, 458 mantle of, 5, 190, 191, 197–98, 198, 202,
deformable surface, of mirrors, 136 dust: 264
Degenerate Era, 626, 627 in comets, 342–46, 346, 349 mass of, 74–75, 78, 85, 168, 188–89, 218,
Deimos, 300, 323, 338–39, 339 in diffuse nebulae, 510 220, 221, 250, 251, 276–79, 276, 284
Delta Cephei, 473 interplanetary, 300–301, 333n measurement of distance to stars from,
deoxyribonucleic acid (DNA), 633, 635–37, in spiral galaxies, 535, 539–40, 541, 544, 362–67, 363, 364, 378 (table), 381
638, 640, 642 544, 546, 579 Moon’s orbit around, 23, 42–46, 43, 44,
De revolutionibus orbium coelestrium zodiacal, 355–56, 356 45, 47, 50, 51, 53, 169, 279, 284–86, 284,
(“On the Revolution of the Celestial dust, interstellar, 5, 58, 154, 155–58, 158, 285, 319
Spheres”) (Copernicus), 57 418–20, 419, 420–21, 429, 438–39, 443, night sky as seen from, 563, 564
deserts, 220 563 oceans of, see oceans of Earth
Desiderata (Ehrmann), 154 light blocked by, 418–20, 419, 420 precession of axis of, 293
de Sitter, Willem, 589, 589 dust devils, 236–37, 237 radiation belts of, 229, 268
destructive interference, 132 dust storms, on Mars, 239, 241–42 as seen from space, 180, 181
deuterium, 394, 529, 530 dust tail, 342, 342, 346, 346 size of, 5, 33, 188, 194, 247
Dialogo sopra i due massimi sistomi del dwarf galaxies, 4, 542, 543, 574, 614, 617 solar evolution and, 456
mondo (“Dialogue on Two Great World see also Large Magellanic Cloud; Small as spherically symmetric, 276–77, 278
Systems”) (Galileo), 67 Magellanic Cloud tectonism on, 179, 180, 186, 196–202, 196,
Diana, 43, 283 dwarf planets, 4, 63, 65, 65, 156, 169, 176, 197, 198, 199, 200, 211, 302
Dicke, Robert, 526 250, 299, 300, 327, 328–31, 328 (table), temperature of, 111–12, 119, 166, 192–93,
differential rotation, 407 330 219, 220, 222, 237–38, 243, 412
differentiation, 191 264, 333, 334–35 dynamic equilibrium, 110, 111 thermal equilibrium of, 112
of asteroids, 327, 332, 334–35, 335 dynamo: tidal locking of Moon to, 284–85, 284
I-6 Index
Earth (continued) 136, 141, 144, 147, 151, 268–69, 551, elliptical orbits, 16, 36n, 64, 82–85, 83, 294,
tides and, 193, 275–87, 279–85 611–12 537
Venus compared with, 200–202, 220–21, see also light; spectrum of binary stars, 373–74, 374
238 electromagnetic waves, 101, 235 as bound orbits, 85
volcanism on, 179, 181, 181, 186, 187, 195, defi ned, 94 of comets, 347
202–6, 211, 218, 633 energy of photons and, 99, 103 eccentricity of, 63, 63, 83, 83, 374
winds on, 207, 229–32, 233, 234–37, 234 gravity waves compared with, 499 focus of, 16, 62, 63, 63
wobbling axis of, 42, 293 light as, 91, 93–99, 95, 99–100, 490, 491 Kepler’s fi rst law and, 62–63, 62, 63
Earth, orbit of, 4, 16, 25, 32–42, 35, 51, 53, electron-degenerate core, 448–49, 465, 471 Kepler’s third law and, 58
81–82, 171, 218, 279, 336, 354, 497 atomic nuclei in, 452–53, 452 Newton’s prediction of, 79, 82, 497–98
aberration of starlight in measurement of, carbon, 455, 456, 458 tides and, 285–87, 285, 286
34–36, 35 of post-AGB stars, 457, 458 El Niño, 238
chaos and, 295 of red giants, 443, 449, 450, 451, 452–55, emission, 102–7, 103, 104, 424
eccentricity of, 63 452, 456, 463, 477 defi ned, 103
as ellipse, 36n, 62–63 of white dwarfs, 443, 459, 487, 505, 518 stimulated, 107
Kepler’s laws and, 62–64 electron microscopes, 134 emission lines, 102–7, 104, 106, 110, 113,
night stars and, 23 electron-positron pair annihilation, 597, 370–71, 482, 499, 549
seasons and, 23, 36–41, 52 597, 598, 602 of AGNs, 551, 554
speed of light and, 92 electrons, 72n, 106, 114, 140, 148n, 227, 262, defi ned, 103
tilt of axis and, 23, 36–39, 37, 38 268, 269, 271, 394, 405, 423, 482, 554, interstellar, 422, 424, 439
Earth, rotation of, 23, 25, 26, 51–52, 53 627 molecular, 424
change of seasons and, 36–41, 52 antiparticle of, see positrons as spectral fi ngerprints of atoms, 106–7,
Coriolis effect and, 231–33, 232 in atomic structure, 100, 101, 101, 448, 106
latitude and, 23, 28–32, 29, 231 595–96 star classification and, 369, 369
magnetic field and, 195, 195 “clouds” or “waves” of, 100–101, 101, 102 Type I supernova and, 464n
tides and, 279–85, 281 deflected by electric field from proton, empirical rules:
view from the poles and, 25–32, 26, 28 596 Kepler’s discovery of, see Kepler’s laws
weather patterns and, 23, 231, 232–33 in early universe, 597, 598, 601 physical laws vs., 65
earthquakes, 180, 189–91, 399 magnetized, 423 empirical science, 57, 59, 85, 86
plate tectonics and, 198, 200 of neutron stars, 482, 484, 484 Enceladus, 302, 304, 304, 305, 314, 320, 323,
eccentricity, of orbit, 63, 63, 83, 83, 374 Pauli exclusion principle and, 448n 330, 641
ECHO satellites, 525 radiation blocked by, 527, 527 cryovolcanism on, 304, 305
eclipses, 23, 46–51 electroweak theory, 596, 598 Encke, Comet, 354 (table)
stellar, see stellar occultations elementary particles, 394, 613 Encke Division, 316, 321–22, 321
Stonehenge and, 46, 46 elements, chemical, 5 energy, 12, 19
varieties of, 46–50; see also lunar isotopes of, 187, 394, 463, 476, 477, 530, equivalence between mass and, 392, 492,
eclipses; solar eclipse 530, 612 597
eclipse seasons, 51 least massive, Big Bang theory and, 529– grand unified theory and, 599
eclipsing binary, 376–77, 376, 377, 384 30, 530, 585 kinetic, see kinetic energy
ecliptic, 27, 33, 34, 36, 39, 40–41, 355 massive, see massive (heavy) elements law of conservation of, 159, 164, 165, 431
Eddington, Sir Arthur, 498, 499, 501 origin of, 5, 7, 458–59 light in transport of, 91, 95, 99, 120
Edgeworth, Kenneth, 340n parent, 187 potential, see potential energy
Edgeworth-Kuiper Belt, 340n periodic table of, 253, 480 primordial, 263
see also Kuiper Belt relative abundance of, 480, 481 relativity theory and, 492
Égalité, 317n solar abundance and, 252 thermal, see thermal energy
Einstein, Albert, 2, 17, 392, 490, 494 see also specific elements energy states, atomic, 100–105, 102
“great blunder” of, 586–89 ellipses: emission and absorption determined by,
particle theory of light and, 99, 100 defi ned, 62 102–5
quantum mechanics and, 12, 101, 107–8 eccentricity of, 62–63 excited, 102–5, 107, 423
scientific revolutions and, 12, 12, 107–8 focus of, 62–63, 62, 63, 85 energy transport, 393–95
see also general relativity theory; special semimajor axis of, 62, 63, 63 England, calendar in, 42
relativity theory elliptical galaxies, 510–11, 537, 543, 559, 614 Ensisheim, meteorite fall at, 331, 331
ejecta, 183, 185, 186 bulge in, 539 (table), 557 entropy, 627, 639
electric field, 93–94, 94, 114, 120, 596, 596 classification of, 537–42, 538, 539 (table), epicycle, 60, 60
in electromagnetic waves, 94, 95, 96 540 Epimetheus, 291
neon signs and, 104 color of, 540 Epsilon Ring, 317, 320, 321, 322
electric force, 93, 490 dark matter in, 547–49 equator, 26, 28
electricity: disk of, 539 celestial, see celestial equator
lightning and, 234–35 formation of, 618, 619 Coriolis effect and, 232
potential, 234 globular clusters compared with, 565 of giant planets, 248 (table), 253, 260
resistance and, 235 mass of, 547–49 seasons near, 40
electromagnetic force, 93–94, 595–97, 598, radio, see radio galaxies wind patterns and, 229–32
601, 602 stellar orbits in, 537, 539, 540, 571 equilibrium:
electromagnetic radiation, 86, 91–123, 125, elliptical nebulae, 510 basic properties of, 110–11, 112
Index I-7
hydrostatic, 191, 221, 390, 398, 426 motions of the planets in the sky, 60–61 electric, 93, 490
Lagrangian points and, 292–93, 293, 336, neutrino astronomy, 400–401 electromagnetic, 93–94, 595–97, 598, 601,
401 nightfall, 566 602
temperature, 117–19, 118, 119, 262–63, scattering, 240 electroweak theory and, 596, 598
419, 420 strange behavior in the realm of ultrahigh grand unified theory and, 598–603, 602
thermal, 112 pressure, 265 gravitational, see gravity
equinoxes: uniqueness of stars, 472 magnetic, 93, 490
autumnal, 39 expanding universe, 501, 509–31, 601, 603, in Newton’s third law, 70, 72, 72, 73
precession of, 42 611 strong nuclear, 391, 446n, 453, 470, 478,
vernal, 39, 42, 362 acceleration of, 587–88, 588 595–96, 597, 598, 601, 602
equivalence principle, 494, 495, 496 fate of, 5, 509, 531, 585–92, 587–91, 626–27 unbalanced, see unbalanced force
Eratosthenes, 33, 78 gravity and, 586, 587, 589, 611 weak nuclear, 596, 598, 601, 602
Eridanus, 534 Hubble’s law and, 513–21, 516, 519, 520, Formalhaut, 173
E Ring, 304, 304, 314–15, 319–20, 323 521, 522 Formalhaut b, 173
Eris, 4, 64 (table), 169, 299, 327, 329–30, 330, inflation period of, 593–95, 594, 601, 602, 47 Tucanae (star cluster), 486, 487
340 611 fossil record, 184–85, 194, 196
Eros, 338, 339 key concepts about, 509 foundations boxes:
erosion, 203 recombination in, 527, 527, 602, 603, 611– coriolis effect, 231–33
defi ned, 181 13, 613 determining the ages of rocks, 187
in shaping terrestrial planets, 179, 181– scale factor of, 522–25, 586, 587, 588, 589, direction of light, 130–33
82, 184, 186, 206–10, 207, 208, 209 591 energy, 165
by water, 181, 206–10, 208, 221 shape of, 509, 591–95, 591, 592, 611 equilibrium, 110–11
by wind, 181, 206–7, 207, 208 see also Big Bang theory explained, 18
escape velocity, 83–85, 216, 530 extinction of species, 184–85, 185, 349, 352, gas, 222–23
black holes and, 488, 502 358 proportionality, 71
defi ned, 83 extrasolar planet, 170–71 proton-proton chain, 394
Eta Aquariids, 354 (table) extremophiles, 306, 307, 635 symmetry, 278
Eta Carinae, 7, 475, 475 eyepiece, 126 synchrotron emission, 269
eternal inflation, 624, 625 eyes, human, 129, 137 when redshift exceeds one, 524
ethane, 259, 308 color and, 367–68 frame of reference:
Euclidian geometry, 585 limitations of, 129, 134–35, 136–38 accelerated, 492–93, 495, 496
eukaryotes, 635, 635 as refracting telescopes, 129 change in, 492–93
Europa, 169, 300, 306, 307, 323, 641 stereoscopic vision of, 361, 362, 362, concept of, 23
European Extremely Large Telescope 367 direction of rainfall and, 34–35, 35
(E-ELT), 127, 128 equivalence principle and, 494, 495, 496
European GALLEX experiment, 400 Fahrenheit (F), 112 inertial, see inertial frame of reference
European Southern Observatory (ESO), 127, Fairbanks, Alaska, 32 motion of object and, 34–35, 39, 231, 231,
128, 143–44, 143, 423 faults, 196 488–90
European Space Agency (ESA), 145, 205, Feibelman, Walter A., 314 relativity theory and, 490–93, 494–95,
209, 401, 611, 622, 643 Fermi, Enrico, 645 494, 496
event, in relativity theory, 490 Fermi Gamma-Ray Space Telescope, 145 stationary vs. high velocity, 494–95, 494
event horizon, 502–3, 503, 524 (table) Fraternité, 317n
evolution: Fermi National Accelerator Laboratory, 148, free fall, 80, 80
of life, 52, 221, 224, 283, 307, 349, 352, 357, 148 frequency, of wave, 96, 99, 103, 108
417, 445 festivals, winter, 39, 40 energy of photon and, 99, 103
stellar, see high-mass stars, evolution of; fiducial marks, 139 for sound, 108
low-mass stars, evolution of; stellar fi lters, 368 F Ring, 320–21, 320–21
evolution fi re truck siren, Doppler effect and, 108 full Moon, 43, 45, 50–51, 51
of universe, 601, 602, 603; see also fi rst quarter Moon, 45, 275, 282, 283 tides and, 275, 282, 283
universe, chemical evolution of fissures, 202 Fuse orbital telescope, 137
evolutionary track, 434, 435 flatness problem, 593 future, of expanding universe, 509, 585–92,
excited states of atoms, 102–5, 107, 423 flat rotation curves, 547, 548, 569–70, 570, 587–91, 626–27
excursions boxes: 612, 615, 618
atmospheres and origin of life, 224 flat universe, 591–93, 591, 592 “galactic fountain” model of disk of spiral
backlighting phenomenon, 317 flux, 115 galaxy, 574, 575
chaos, 294–95 flux tube, 270, 271, 271, 405 Galápagos Islands, 307
Comet Shoemaker-Levy, 350–51 prominences as, 411 Galatea, 318
explained, 18 flyby, 145–46 galaxies, 487
extinction of dinosaurs, 184–85 focal length, 125, 129, 130, 131, 133, 135 appearance of, 535–41, 536
forever in a day, 632 focus, of ellipse, 16, 62, 62, 63, 85 centers of, 501, 504, 535, 549–59, 550, 569,
Jupiter, 263 focus, focal plane, 129, 129, 131, 131, 133 570
leap years, 42 fold, 196 classification of, 509–10, 535–41, 536–41,
meteors, meteoroids, meteorites, and forces, 595–97, 595 539 (table), 542–43; see also elliptical
comets, 333 centripetal, 81, 81 galaxies; spiral galaxies
I-8 Index
galaxies (continued) planetary rings and, 318, 319 George III, King of England, 249
clusters of, 85, 608–13, 609, 610, 616, 616, Galle, Johann, 249 geothermal vents, 634
617 gallium, 400 germanium, 400
coins compared to, 535–36, 536 Galveston, Tex., 236 geysers, on moons, 299, 300
defi ned, 510 gamma rays, 98, 98, 125, 137, 137, 144, 411, Giacobini-Zinner, Comet, 354 (table)
discovery of, 509–10 453, 477, 484, 551 giant galaxies, 542
dwarf, 4, 542, 574, 614, 617; see also Large photons of, 394, 597, 597 giant molecular clouds, 425–26
Magellanic Cloud; Small Magellanic Gamow, George, 525–26 giant planets, 155, 167, 168–69, 172, 195,
Cloud Ganymede, 169, 300–301, 300, 301, 310, 311 247–72
elliptical, see elliptical galaxies gas, interstellar, 154, 155–58, 158, 417–18, atmosphere of, 167, 216, 235, 247, 252,
expansion of space and, 509, 519–20, 522, 421–28, 438–39, 443, 459, 573 254–63, 254, 255, 256, 257, 258, 261
523 ionization of, 417, 422–23 cometary orbits disrupted by, 217
giant, 542, 543 jets from protostars and, 436, 436 defi ned, 247
gravity of, 149, 517, 537, 544, 558, 567, 575, in Milky Way, 571–75, 574, 575, 577 extrasolar, 171–72
616, 621 temperatures and densities of, 421–26, gas vs. ice, 247–48, 248, 249, 266
intergalactic space compared with, 421 422, 423 interiors of, 247, 263–66, 264, 272
irregular, 537, 538, 539 (table), 540 see also intercloud gas; interstellar clouds key concepts about, 247
key concepts about, 535 gases: as magnetic powerhouses, 266–71, 267,
Local Group of, 4, 608 in comet, 342–46, 349 268, 269, 270, 271
luminosity of, 541 defi nition and properties of, 222–23 rings around, see rings, planetary;
mass of, 85, 535, 542, 542, 545–49, 547, in elliptical galaxies, 540, 548 specific rings and planets
548, 559, 586, 621, 623 in galaxy clusters, 616, 616 terrestrial planets compared with, 247,
measurement of distance to, 622 inert, 252 250–54, 257, 260, 261, 262–63, 264, 266,
mergers and interactions of, 149, 150, 288, in mass of galaxies, 535, 546, 547, 548–49 271–72
288, 517, 535, 544–45, 550, 557–58, 557, photons absorbed by, 104–5, 105 see also Jupiter; Neptune; Saturn; Uranus
574, 606, 607, 617–18, 618, 619 photons emitted by, 103–4, 104 gibbous, 44, 45
motions of, 509 in spiral galaxies, 540, 544, 544, 545 Giotto spacecraft, 343–44
peculiar velocities of, 517, 622, 623 see also specific gases global circulation, 230, 239, 262
recession velocity of, 514, 518, 519 gas giants, 248, 248 Global Oscillation Network Group (GONG),
redshift of, 509, 513–14, 516, 518, 523–25, Gaspra, 337–38 401
524, 525, 587, 588, 608–9, 610, 622 gauss, 266 Global Positioning System (GPS), 499–500
size of, 541–42, 542–43 Gauss, Carl Friedrich, 266 global warming, 220, 237–38, 238, 243
spiral, see spiral galaxies Gemini, 39, 41 globular clusters, 510, 541, 563, 565–67, 565,
in structure of universe, 607, 608–10, 609, Geminids, 354 (table) 566, 567, 571–76
610 Gemini telescopes, 128 (table) age of, 566, 566, 571–74, 579–80, 591
superclusters of, 4, 608–10, 610, 611 general relativistic time dilation, 499 distribution of, 567, 567
supermassive black holes in, 501, 504, general relativity theory, 12, 77, 494–501, formation of, 617
535, 552–59, 553, 555, 577–79, 617–18 494, 591, 603 massive elements in, 572–73, 574
S0 (S-zero), 537, 538, 539 (table) confrontation between quantum open clusters compared with, 571–73
tidal interactions and, 288, 288, 557–58, mechanics and, 600 gluons, 596, 627
557, 617 cosmological constant and, 588–89 golf ball analogy, for interstellar medium,
see also Milky Way Galaxy equivalence principle and, 494, 495, 496, 417–18
Galaxy Evolution Explorer (GALEX), 145 504 Gossamer Rings, 319
(table) expansion of space and, 509, 521 Grand Canyon, 187, 187, 200
galaxy formation, 558, 571, 579–80, 602, 607, frame of reference and, 494–95, 494, 496 grand unified theory (GUT), 598–603, 602
610–21 as geometrical theory, 496, 497 granite, 191
computer simulation of, 620–21, 621 MACHOs and, 571 Gran Telescopio Canarias, 128 (table)
dark matter in, 549, 607, 611–23, 612, 613 Newton’s law of gravity and, 497–99 granules, 407
mergers in, 617–18 Newton’s laws of motion and, 497 grating, diffraction, 132, 133, 141
star formation compared with, 610–11, observable consequences of, 497–501, 499 gravitational constant, universal (G), 71, 76,
614–16 redshift of galaxies and, 523–25 78
galaxy groups, 608–9 structure of universe and, 499, 501, 600 gravitational lensing, 171, 498–99, 499
Galaxy Redshift Survey, 610 genetic code, 52 in search for dark matter, 570, 571, 616,
Galilean moons, see Jupiter, moons of geocentric model of the universe, 25, 60 617, 618
Galileo Galilei, 25, 66, 85, 86, 124, 126–27, see also Earth, as center of universe gravitational potential energy, 164–66, 263
129, 151, 249, 299, 323, 512 geodesic, 495, 497 gravitational redshift, 499, 500, 504, 524,
gravity and, 73, 75, 78, 86 geology, geologists, 180–210 529
law of inertia of, 67 history of Solar System as viewed by, 156 gravitational wave detectors, 147
motion of objects as viewed by, 57, 67, 80 model of Earth constructed by, 191 gravity, 25, 57, 72–86, 275–96, 596, 602, 603
speed of light and, 92 see also erosion; impact cratering, impact acceleration and, 73, 78, 80, 85, 279–80,
telescope of, 66, 126–27, 126 craters; tectonism; volcanism 279, 284, 285, 449, 456, 492, 492
Galileo probe, 250, 257, 259, 304 306, 337– geometry, 75 atmosphere and, 167, 216–17
38, 351 defi ned, 15, 16 in binary system, 461–63, 464
moons and, 303, 303, 307, 307, 310 of spacetime, 496, 497 of black hole, 488, 501, 502, 502, 504, 505
Index I-9
centripetal force and, 81, 81 warming trend and, 220, 238, 238, 243 in early stellar formation, 619
Copernicus and, 57–58 greenhouse molecules, 220 in giant planets, 167, 168, 247, 252, 263–
as distortion of spacetime, 469, 494–505 Gregorian calendar, 42 65, 272
electron-degenerate core and, 443, 448– G Ring, 204n, 314–15, 318 in interstellar medium, 418
49, 453 ground state of atom, 102–4 liquid, 263
equivalence between effect of inertia and, Gusev crater, 209 in nuclear fusion, 389, 391–94, 394, 413,
76–77, 495, 496, 504 GUT, see grand unified theory 431, 438, 444–48
expansion of the universe and, 585, 586, Guth, Alan, 594 stellar build-up of, 448, 450, 470–72
587, 589, 611 in stellar composition, 370–71, 371 (table),
Galileo’s insight and observation about, habitable zones, 641–43, 642 402, 405, 413, 443, 448, 448
73, 75, 78, 86 Hadley cells, 230 helium burning:
general relativity and, 496–97, 600 Hadley circulation, 230, 230, 239 in high-mass stars, 471–72, 472 (table),
grand unified theory and, 600 hail, 234 473, 475, 477, 480
of high-mass stars, 469, 475, 477 Haldane, J. B. S., 630 in horizontal branch star, 454–55, 464
of horizontal branch stars, 455 Hale, George Ellery, 410 in intermediate-mass stars, 472
as increasing outward from object’s Hale-Bopp, Comet, 339–40, 342, 349 in low-mass stars, 443, 452–59, 453, 454,
center, 277, 277, 278 half-life, 187 464
of interstellar cloud, 58, 156, 426–28, 426, Halley, Comet, 319n, 339, 343–44, 344, 346, triple-alpha process and, 452–53, 453,
428, 544 347–48, 354 (table) 456–57, 476, 481, 530
as inverse square law, 75–77, 77, 86, 117, Halley, Edmund, 347–48 helium flash, 453–55, 454, 460, 463, 464, 472
288, 293–94, 324, 397, 427, 431, 498 halo, 565 helium shell burning, 454, 456, 462, 472
key concepts about, 275 halogens, 225 hematite, 209
large-scale structure formed by, 607, 610– Halo Ring, 319 Herbig-Haro (HH) objects, 436, 436, 554
21, 613, 617, 621 Harmonice mundi (“The Harmony of the heredity, 637
of molecular cloud, 428, 429 Worlds”) (Kepler), 58 Herschel, Caroline, 332, 509
orbital resonance and, 275, 289, 290, 292 Harmony of the Worlds (Kepler’s third law), Herschel, Sir William, 58, 249, 312, 332, 509,
place-to-place differences in, 276–78 57, 64 (table), 65, 65, 82, 85, 251, 375–76 534, 640
planetary rings and, 299, 314, 319, 320–21 Harvard Observatory, 369, 510 model of Milky Way proposed by, 564,
of planetesimals, 163, 163, 168 Harvard-Smithsonian Center for 564
of planets, see specific planets Astrophysics, 609, 610 Herschel Crater, 312, 312
in protoplanetary disk, 168, 172 Haumea, 327 Herschel Space Observatory, 145 (table)
of protostars, 158, 169, 417, 429–33, 431, Hawaiian Islands, 201, 211 hertz, 96
439, 443, 445, 552 volcanism in, 181, 203, 203, 205, 205 Hertz, Heinrich, 96
protostellar disk and, 158 Hawking, Stephen, 503 Hertzsprung, Einar, 376–78, 384
of stellar core, 294–95 Hawking radiation, 503, 503, 587, 627 Hertzsprung-Russell (H-R) diagram, 360,
of Sun, 58, 75, 275, 282, 289, 294, 296, 341, Hayashi, Chushiro, 433 361, 376–83, 379, 487
389, 390, 390, 397–98, 405, 413, 444–45, Hayashi track, 434, 435, 436, 438, 444, 450, age of stars and, 469, 485–87, 485, 486,
448–49 451, 486 487
symmetries and, 275, 278, 598 HD226868 (star), 504 of globular clusters, 566, 566
tectonism and, 196 head (comet), defi ned, 342 instability strip of, 471, 473, 567
tides and, 275, 279–87, 279–85, 287, 288, heat, heating, 111–16 main sequence and, 361, 379–84, 380,
289, 294, 296 blue light and, 113–15 381, 382 (table), 433, 435, 439, 461, 485,
gravity, Newton’s law of, 57, 72–85, 77, 149, defi ned, 112 486–87, 566
188, 222, 251, 275, 293–95, 323–24, 347, in evolution of planetary interiors, non-main sequence and, 383, 449, 451
456, 497–98, 589, 625 192–94 of open clusters, 571, 572
Newton’s laws of motion applied to, 73, luminosity and, 113–14 post-main sequence stellar evolution and,
74–75, 78–79 subjective experience of, 112 443, 449–50, 451, 455–57, 456, 457, 459,
playing with, 78 tidal, 193 460, 461–62, 471, 471, 473, 486
reasoning toward, 73–76 Wien’s law and, 115, 116 of pre-main sequence stars, 434, 435
testing of, 79 heat death, entropy and, 627 of star clusters, 469, 485, 486–87, 486, 487
gravity waves, 149, 501 Heaven’s Gate, 340n star formation and, 417, 433–35, 434, 439
Great Attractor, 622, 623 heavy elements, see massive (heavy) Hickson Compact Group, 537
great circle, 32, 39 elements hierarchical clustering, 611, 617
Great Dark Spot (GDS), 255–57, 257, 261 Heisenberg, Werner, 101, 101 high-energy colliders, 147–48, 148
Great Debate, 510–11 Heisenberg uncertainty principle, 101 high-mass stars, “chimneys” of, 575
Great Red Spot (GRS), 254–55, 257, 261, 271 heliocentric model of the universe, 61 high-mass stars, evolution of, 444–45, 465,
anticyclonic motion of, 254, 255 see also sun, as center of universe 468–507
Greeks, ancient, 25, 33, 43, 57, 60, 66–67, helioseismology, 399–402, 402 CNO cycle in, 470, 470, 481
150, 283, 331 helium, 5, 106, 106, 107, 459, 484, 574, 576, core collapse in, 472, 477–79, 478, 501
Green Bank radio telescope, 137 613, 640 evolution of intermediate-mass stars
greenhouse effect, 219–21 atmosphere formation and, 167, 168, compared with, 472
atmospheric, see atmospheric greenhouse 215–16 evolution of low-mass stars compared
effect birth of universe and, 480, 529, 530, 601 with, 469–72, 473, 475, 477, 485–86
ozone hole vs., 225n CNO cycle and, 470, 470 fi nal days in life of, 476–77
I-10 Index
high-mass stars (continued) moons and, 329, 329 in terrestrial planets, 210, 210
H-R diagram and, 471, 471, 475, 485–86, Shoemaker-Levy 9 and, 351, 351 ice ages, 237, 412
485 stellar evolution and, 458, 475, 480 IceCube Neutrino Observatory, 401
instability strip and, 471, 473 Hubble time, 591 ice giants, 248, 249
key concepts about, 469 origin of, 518–19, 520 Iceland, 197, 202
main sequence, 470–71, 470, 471, 475, Hubble Ultra Deep Field (HUDF), 620, 621 ice skater, angular momentum of, 158–59,
485–86 humidity, relative, 227 159
mass loss in, 475, 475 hurricanes, 235–36, 236 Ida, 338, 338
neutrino cooling in, 477 Coriolis effect and, 232, 233, 235–36, 261 ideal gas law, 222–23
post-main sequence, 470–79, 473, 475, Huygens, Christiaan, 313, 314 igneous activity, 181, 202–6, 211
476, 478 Huygens probe, 309, 310 see also volcanism
stages of nuclear burning in, 469, 470–72, Hyakutake, Comet, 326, 340, 344, 344, 352 Ikeya-Seki, Comet, 346, 347, 349
472 (table), 473, 475–76, 640 hydroelectricity, 165 image-processing software, 139
see also black holes hydrogen, 5, 167, 344, 459, 574, 638, 640 image size, 125, 132
high-mass stars, in spiral galaxies, 545, 575 in atmosphere of terrestrial planets, 168, impact cratering, impact craters, 180–86
high-velocity stars, 576, 579 215–16, 217, 222, 229, 613 comets and, 184, 217, 345–46, 350–52,
Hi’iaka, 330 in early stellar formation, 619 353, 357
H II regions, 422–23, 423, 430, 457, 543–44, energy states, of, 106, 106, 107, 423 defi ned, 181, 183
554, 573 in giant planets, 167, 168, 247, 252, 263– on Earth, 180, 183–84, 183, 184, 186, 198,
Hipparchus, 365 66, 264, 272 338n, 350–53
Hipparcos mission, 365, 380, 380 in interstellar medium, 418, 422–23, 422, formation of features in, 183, 183
Hobby-Eberly telescope, 128 (table) 424, 438–39 on Moon, 180, 182, 184, 186, 186, 188, 204,
Holmes, Comet, 342, 343 molecular, 424–25 206, 210
Homestake neutrino detector, 400, 401 in protostellar and stellar atmospheres, on moons, 299, 302, 304, 306, 310, 311
homogeneity, of distribution of galaxies, 433–34, 449 in shaping terrestrial planets, 168, 168,
512–13, 512 quasars and, 549 169, 179, 180–86, 182, 183, 185, 186,
horizon, 26–32, 36 radiation and, 527 200, 206, 211, 336, 349–52
horizon problem, 595 spectral lines of, 106, 106, 464n, 546, 549 stages in formation of, 182
horizontal branch stars, 454, 455, 456, 460, spiral galaxies and, 544, 544, 568, 569, incidence, angle of, 130, 131
464, 471, 486 569, 574, 575 index of refraction, 92
instability strip and, 473, 566 star classification and, 369 Indian Space Research Organization (ISRO),
Horsehead Nebula, 423, 425 in stellar composition, 370–71, 371 (table), 145
hot dark matter, 613 413, 444–49, 448 inert gases, 252
see also neutrinos hydrogen burning, 454–55, 457, 463, 463 inertia, 67–70, 80, 86, 492
hot Jupiters, 172 CNO cycle and, 470, 470, 481 defi ned, 67
hot spots, 201–3, 205–6 in high-mass stars, 470–87, 470, 472, 473, equivalence between effect of gravity
H-R diagram, see Hertzsprung-Russell (H-R) 475–77, 481 and, 76–77, 495, 496, 504
diagram in Sun, 389, 391–94, 394, 413, 431, 438, Galileo’s law of, 67
HR 8799, 173 444–48 measure of, see mass
HST, see Hubble Space Telescope hydrogen cyanide, 307, 344, 346 Newton’s fi rst law of motion and, 67–68,
Hubble, Edwin P., 511, 531, 558, 567, 586–88 hydrogen shell burning, 448–50, 450, 451, 495
Cepheid variables discovered by, 565 452, 455, 456, 464, 472 inertial frame of reference, 67–68, 68, 488,
classification of galaxies by, 531, 537, 538, hydrogen sulfide, 252 490, 521
539 (table), 541, 542, 559 hydrologic cycle, 309 preferred, 528
on cover of Time, 513 hydrosphere, 180 special relativity and, 494, 496
Great Debate and, 511 hydrostatic equilibrium, 191, 221, 390, 398, inflation model of universe, 585, 593–95,
Hubble constant, 514, 515, 516–18, 590–91 426 594, 601, 602, 611, 623–25
Hubble flow, 517 hydrostatic jump, 546, 549 Infrared Astronomical Satellite (IRAS), 420
Hubble’s law, 509, 513–21, 522, 531 hydrothermal vents, 307 infrared (IR) radiation, 99, 136–37, 137, 141,
discovery of, 509, 513–15, 516 hyperbolic orbits, 85 144, 227, 356
structure of the universe and, 519, 520, Hyperion, 301 atmospheric greenhouse effect and, 219,
521, 608–9, 622 hypothesis, defi ned, 9 238
universe mapped in space and time by, interstellar dust and, 417, 419–20, 419,
509, 518 Iapetus, 310, 311 420
Hubble Space Telescope (HST), 134, 135, ice, 166, 167–68, 221, 234, 332 of protostar, 429
136, 144, 145 (table), 157, 162, 242, 248, in comets, 167, 341–42, 344, 349, 352 star formation and, 417, 429, 430
287, 313, 316, 341, 344, 365, 411–12, 430, defi ned, 248n Inquisition, 67
518, 551n, 587 dry, 342 instability strip, 471, 473, 566
AGNs and, 555, 555, 558 on dwarf planets, 330, 331 integration time, 137–38
deep image of “blank” sky made by, 509, in erosion, 206 intelligent design, 639
510, 618 in giant planets, 166–67, 248, 249, 259, intensity of light, 99
of galaxy cluster, 617 265, 272 intercloud gas, 421–23, 421, 424, 425–26
giant planets and, 250, 254, 257, 270 in moons, 299, 301, 301, 302, 304, 310, interference, 132, 133
HH objects and, 436, 436 312, 319, 329 interferometer, interferometric array, 142
Index I-11
intergalactic space, galaxies compared with, Irwin, James B., 8 Keck telescopes, 127, 127, 128 (table), 133,
421 “island universes,” 509, 510 135, 587
intermediate-mass stars, 472 isotopes, 187, 394, 463, 476, 477, 612 kelvin (K), 113, 164, 166
International Astronomical Union (IAU), 4, in Big Bang, 530, 530 Kelvin scale, 113
170, 172, 329, 331 isotropy: Kepler, Johannes, 25, 58–59, 58, 62–65, 75, 86
International Space Station (ISS), 56, 144 of cosmic background radiation, 528, 625 Keplerian shear, 322
Internet, 149, 349 of distribution of galaxies, 509, 512, 513, Kepler mission, 643
interstellar clouds, 154, 341, 352, 479, 479, 514 Kepler orbital spacecraft, 145 (table), 171
480 Kepler’s laws, 58, 62–65, 79–87, 151, 161,
collapse of, 156–62, 161, 164, 263, 357, James Webb Space Telescope, 144, 145 222, 276, 294, 323, 328, 547
426–28, 427, 428, 443, 482, 485, 544 (table), 412 fi rst, 62–63, 62, 63
defi ned, 421 jansky, 141 mass of binary stars and, 361, 373, 375–
gravity of, 58, 156, 426–28, 426, 428, 544 Jansky, Karl, 141 76, 378 (table), 4761
in origins of solar system, 5, 58, 156–62, Janus, 291 Newton’s explanation of, 79–82, 83, 85,
158 Japan Aerospace Exploration Agency 92, 275, 375, 559
size of, 160 (JAXA), 145 planetary rings and, 320, 322
in star formation, 156–62, 161, 164, 417, JCMT microwave telescope, 137 second (Law of Equal Areas), 63–64, 63,
426–28, 426–30, 544 Jefferson, Thomas, 56, 331 83, 290
see also molecular clouds jet, 343 third (Harmony of the Worlds), 57, 64
interstellar extinction, 418–19, 419 jet stream, 255, 255, 260 (table), 65, 65, 82, 85, 251, 375–76
interstellar medium, 406, 407, 479, 480, 572 joule, 115 Kilmer, Joyce, 388
dust in, see dust, interstellar Jupiter, 7, 60–61, 146, 247–69, 248, 271–72, kinetic energy, 165, 182–83, 597
gas in, see gas, interstellar; intercloud 323, 335–36 of atmospheric atoms and molecules, 216,
gas; interstellar clouds atmosphere of, 167, 235, 252, 254–59, 255, 222
key concepts about, 417 257, 287, 350–51 of meteoroids, 333
of Milky Way, 573, 573, 575, 576–77 auroras of, 270–71, 270 of supernovae, 479
star formation and, 156–62, 418–40, 573– chemical composition of, 247, 252, 257, Kirkwood, Daniel, 289
74, 573 259, 263–66, 272 Kirkwood gaps, 289, 290, 320, 336
inverse square law: collapse of, 439 Kitt Peak National Observatory, 214
defi ned, 75 colorful bands of, 254, 259 knowledge, 10, 17, 57, 66–67, 600–601
distance of star determined with, 565 comet crash on, 184, 287, 289, 349–50, complacency about, 58
gravity as, 75–77, 77, 86, 117, 288, 293–94, 350–51, 357 as power, 24
324, 397, 427, 431, 498 density of, 248 (table), 252, 263–65 scientific revolution and, 12, 490
radiation as, 76, 116–17, 117, 371–72, 378 differentiation in, 264 scientific theory and, 9–10
(table), 381 formation of, 168, 252, 264–66 sky patterns and, 24
stellar luminosity and, 371–72, 378 (table), gravity of, 167, 168, 265, 266, 287, 289–90, Kohoutek, Comet, 339
381 290 Krakatoa, 202
tidal stresses and, 288 Great Red Spot (GRS) on, 254–55, 255, 257, Kuhn, Thomas, 11
Io, 169, 270, 300 261, 271 Kuiper, Gerard, 340
flux tube and, 270, 271, 271, 302, 405 interior of, 263–66, 264, 272 Kuiper Airborne Observatory (KAO), 144
plasma torus associated with, 269–70, magnetic field of, 266–67, 267, 270–71, Kuiper Belt, 217, 340–41, 340, 341, 344–45,
270, 271, 271, 302 319 347, 350, 352
tidal effects and, 193, 286–87, 287 magnetosphere of, 266–71, 268, 551 Kuiper Belt objects (KBOs), 156, 166–67, 328,
volcanism on, 274, 286–87, 287, 302–4, mass of, 167, 168–69, 170, 248 (table), 251, 330, 340–41
303, 304, 319–20, 323 252, 263, 264–65, 266
ionization, 235 moons of, 92, 93, 126, 193, 292, 294, 295, laboratories, physics, 399
ionosphere, 228, 229 299, 300–301, 300, 302–10, 303–8, 309, Lagrange, Joseph, 292
ion tail, 342, 346, 346 312–13, 319; see also Europa; Io Lagrangian equilibrium points, 292–93, 293,
IR, see infrared (IR) radiation orbital resonance and, 289, 290 336, 401
iridium, 184 orbit of, 64, 170, 171, 248, 248 (table), 289, Lamba Ring, 318
iron, 166, 167, 264, 334–35, 344, 346, 418, 290, 332, 335–36, 344, 347, 349 lambda peak, 116
459, 464, 640, 642 radiation belts of, 268–69 landers, 146–47
chemical vs. nuclear burning of, 476n rings of, 292, 314, 315, 317, 318, 320 landslides, 207
on Earth, 5 Roche limit of, 287, 289 Laplace, Pierre-Simon, 156
in high-mass stars, 475–80, 505 rotation of, 248 (table), 252–53 Large Binocular Telescope, 128 (table)
in Jupiter, 252 satellites captured by, 294, 295 Large Hadron Collider, 148, 148
magnetic field and, 194, 195 size of, 247, 250, 250, 265 Large Magellanic Cloud (LMC), 400, 423,
in planetesimals, 334–35 as “star that failed,” 263 479, 479, 504, 517, 580, 580
in Sun, 371 (table), 405 temperature of, 166, 169, 249, 263–65, 272 orbit of, 569, 570, 571
in terrestrial planets, 191, 224, 239, 264, tidal effect of, 286–87, 287, 289, 332 large-scale structure, 601, 602, 610
336 winds on, 259–60, 261–62, 261 computer simulations of, 620–21, 621
iron meteorites, 334, 334, 335, 357 defi ned, 610
irregular galaxies, 537, 538, 539 (table), 540 Kant, Immanuel, 156, 509, 510, 535 gravity in formation of, 607, 610–21, 613,
irregular moons, 302 Katrina, Hurricane, 236, 236 617, 621
I-12 Index
latitude: wave/particle nature of, 91, 99, 100, 120, last evolutionary stages of, 455–60, 456,
apparent daily motion and, 23, 28–32, 29, 121, 134, 135 457, 458, 460
30, 31, 32 light, speed of, 94, 130, 131, 399, 411, 469, main sequence, 443–48, 447 (table), 448,
climate and, 38, 52 488 459–60, 460, 464, 485–86, 485
defi ned, 28 AGNs and, 551, 555–56, 556 mass loss in, 455–59, 457, 458
speed differences and Coriolis effect and, Doppler effect and, 110 post-main sequence, 443, 447–60, 460,
231–33 electromagnetic waves and, 94, 96 473; see also specific stages
Laughlin, Gregory, 626, 626 Galileo’s failure and, 92 see also red giants, red giant branch;
Law of Equal Areas (Kepler’s second law), measurement of, 92–93, 93, 490 white dwarfs
63–64, 63, 83 Newtonian physics and, 488–90, 489 Lucifer’s Hammer (Niven and Pournelle),
lead, 187 photons and, 99 352
leap years, 42 relativity theory and, 490–93, 491, 494 luminosity, 113–16
Leavitt, Henrietta, 473, 511, 567 Rømer’s work on, 92, 93 of active galactic nuclei, 549–51
legends, 24, 52 size of universe and, 5, 6 brightness vs., 113–14, 116, 365
lenses, 126, 129, 131–32, 132, 133 as ultimate speed limit, 121, 492, 492, defi ned, 113–14
collimating, 141 646 of galaxies, 541–42, 542–43
compound, 131, 132, 132 lightbulbs: of globular clusters, 565–67
convex, 131 brightness of, 116–17 of protostars, 166, 429, 433–34
objective, 134, 135 luminosity of, 116–17 Stefan’s law and, 115–16
Leo, 34 spectrum produced by, 104–5, 105, 113 of Type I supernovae, 465, 518, 518
Leonids, 354–55, 354 (table), 355, 356 thermal radiation emitted by, 114 luminosity, stellar:
Le Verrier, Urbain-Jean-Joseph, 249 lightning, 214, 234–35, 262, 271 of asymptotic giant branch stars, 455–56,
L5 Society, 293 light-years, 364, 381, 400, 425 457
Liberté, 317n defi ned, 93, 159, 364 distance determined by, 565
libration, 284n, 285–87, 286 limb darkening, 402, 403 electron-degenerate core and, 448–50, 452
life: limestone, 221, 243 after helium flash, 454
chemistry of, 638–40 Lincoln, Abraham, 442 of high-mass stars, 469, 471, 473, 475, 487
defi ned, 633 Linde, Andrei, 624 H-R diagram and, 379–84, 379
Earth’s atmosphere reshaped by, 215, 222– linearity, 139 of main sequence stars, 379–83, 381, 382,
25, 224, 243 line of nodes, 51, 51 382 (table), 445, 447 (table)
evolution of, 7, 52, 224, 283, 307, 349, 352, Lippershey, Hans, 126, 151 mass and, 376, 444–45, 487
357, 417, 631, 633–37, 634, 635, 639 liquid: measurement of, 361, 366–67, 366, 378
extraterrestrial, 172, 174, 224, 242, 299, in core of giant planets, 263–65, 264 (table)
306, 336n, 630, 631, 637–43 defi ned, 222 of red giants, 449–52, 451, 454
as form of structure, 631 lithium, 5, 459, 480, 529, 601 size of stars and, 371–72
key concepts about, 631 lithosphere, 180, 191, 197–203, 206 of Sun, 367, 389–90, 397, 445, 449, 455–56,
patterns and, 14, 24 lithospheric plates, 197–203 464
light, 75, 86, 91–123 LMC X-3, 504 of white dwarf, 459
astronomical importance of, 91–92 Local Group, 4, 608, 608 luminosity-temperature-radius relationship,
brightness of, 116–17, 117 local midnight, 26 373, 378 (table), 379–80
diffraction of, 132, 133, 134–35 Loihi Volcano, 203, 203 luminous (normal) matter, 546, 547–49, 570,
dispersion of, 131–41, 131, 141 longitudinal wave, defi ned, 189 574, 586
Doppler effect and, 108–10, 108 long-period comets, 347, 348–49, 348, 352 density of, 529–30, 530
as electromagnetic radiation, 136–37 look-back time, 509, 518, 524 in galaxy formation, 549, 607, 611–15, 613,
as electromagnetic wave, 91, 93–99, 95, Lorentz, Hendrik, 492n 617
99–100, 488, 491 Lorentz factor, 492n structure of the universe and, 611–14,
energy transported by, 91, 95, 99, 120 low earth orbit (LEO), 144 612, 613, 617, 621
evolution of picture of, 91–99 Lowell, Percival, 640 Luna 3 mission, 145, 146
heat and, 113–16 Lowell Observatory, 513 lunar eclipses, 48, 50–51
intensity of, 99 low-mass stars, 570, 640 penumbral, 50
interactions of atoms with, 99–108 defi ned, 444 total, 50
interference and, 132, 133 in spiral galaxies, 545 lunar month, 285
key concepts about, 91 low-mass stars, evolution of, 443–67 Lunar Prospector, 210
particle theory of, 91, 93, 99–110, 116–17 in binary systems, 443, 459, 461–64, 462, Lunar Reconnaissance Orbiter, 210
of pulsars, 482, 483, 484 463 Lyell, Sir Charles, 298
quantized, 99; see also photons evolution of high-mass stars compared Lyrids, 354 (table), 562
reflection of, 130–31, 131 with, 469–72, 473, 475, 477, 485–86
refraction of, 130, 130 evolution of intermediate-mass stars M 51 Galaxy, 139
spectrum of, see spectrum compared with, 472 MACHOs (massive compact halo objects),
as symbol, 91 H-R diagram and, 443, 449–50, 451, 455– 570, 571, 616
telescopes in capture and analysis of, 57, 456, 457, 459–60, 462, 473 McNaught, Comet, 342, 343, 349
128 hydrogen shell burning in, 448–50, 450, Magellan, Ferdinand, 580
visible, 97–98, 99, 106, 107, 136, 141 451, 452, 455, 456, 464 Magellan mission, 200–201, 201, 239
wavelengths of, see wavelengths of light key concepts about, 443 Magellan telescopes, 128 (table)
Index I-13
magma, 181, 202–3, 203, 206 moons of, 207, 299, 300, 323, 338–39 of Sun, 82, 85, 373, 389, 397, 398, 435, 437,
magnesium, 191, 344, 371 (table), 460 orbit of, 241, 289, 332, 336, 346 443, 444, 445, 447, 521
in high-mass stars, 472, 475 samples of, 336 of white dwarfs, 459, 462–63, 518
magnetic bottle, 228, 229 tectonics on, 194, 200, 201, 211 massive (heavy) elements, 572, 620, 640
magnetic field, 93, 120, 307, 596 temperature of, 119, 166, 194, 226, 239, defi ned, 252, 371
in electromagnetic waves, 94, 95 241, 243 origin of, 469, 480, 484, 505
in Milky Way, 576–77, 577 Tycho’s observations of, 62 in stars, 371, 371 (table), 475–76, 480, 571–
of molecular clouds, 427, 484 volcanism on, 179, 194, 200, 204–6, 205, 74, 573, 576, 579, 640, 642
motion of charged particles in, 228–29, 206, 211, 218, 225 mass transfer:
228 water on, 185, 186, 207, 208, 209–10, 209, in binary systems, 461–64, 462, 463
of neutron stars, 483, 484 221, 242, 243 as “tidal stripping,” 461
of planets, 194–96, 247, 266–71, 267, 410; wind on, 207, 207, 208, 237, 237, 242 mathematics, 15–16, 24, 52, 295
see also specific planets Mars Express mission, 205, 209, 210, 339 discomfort with, 15, 18
of protostar, 436 Mars Global Surveyor, 196, 205, 208, 237 tools of, 16, 65
synchrotron radiation and, 268, 269, 269, Mars Orbiter laser altimeter, 205 Mathilde, 338
551 Mars Pathfinder mission, 237 math tools boxes:
magnetic force, 93 Mars Reconnaissance Orbiter, 210, 210, 339 achieving circular velocity, 82
magnetism, 93, 490 Mars rovers, 124, 146–47, 207, 207, 209 angular momentum, 160
magnetosphere: mass, 69, 70 calculating escape velocities, 84
of Crab Nebula, 484 atmosphere formation and, 167 calculating the recession velocity and
of Earth, 215, 228–29, 228, 229, 266, 407, of atoms, 100 distance of galaxies, 515
412, 576 black holes and, 488, 501, 502, 503, 535, diffraction limit, 134
of giant planets, 247, 266–71, 268 552–58, 553, 555, 577–79, 617–21, 627 estimating main-sequence lifetimes, 446
of neutron stars, 482, 484 center of, 373, 374 expansion and the age of the universe,
magnitudes, 366, 382n defi ned, 70, 492 520
magnitude system of brightness, 365 equivalence between energy and, 392, explained, 18
main sequence lifetime, 444, 447 (table) 492, 597 how planets cool off, 194
main sequence of stars, 417, 438 escape velocity and, 84 how to estimate the size of Earth, 33
binary stars on, 461–62 of galaxies, 85, 535, 542, 542, 545–49, 547, how to read cosmic clocks, 187
defi ned, 380, 444 548, 586, 621, 623 how wind speeds on distant planets are
evolution of, 443–48, 447 (table), 448, of galaxy clusters, 616 measured, 260
459–60, 460, 464, 470–71, 470, 471, 473, geometry of spacetime distorted by, 496, the magnitude system, 365
485, 486–87 497, 616 making use of the Doppler effect, 109
H-R diagram and, 361, 379–84, 380, 381, gravity and, 72–86, 251, 275–79, 276, 278, mathematical tools, 16
382 (table), 433–34, 435, 439, 461, 485, 373, 426–27, 444, 469, 487–88, 547 playing with Newton’s Laws of Motion
486–87, 566 inertial vs. gravitational, 76–77, 494–95 and Gravitation, 78
properties of, 379–84, 381, 382, 382 (table) momentum and, 101 the source of the sun’s energy, 392
see also Sun of neutrinos, 399, 399 using a bathroom scale to measure the
main sequence turnoff, 485, 486–87, 487, Newton’s laws of motion and, 70, 74–75, masses of an eclipsing binary pair, 377
566, 566 76 working with electromagnetic radiation,
Makemake, 327 Newton’s theory as tool for measurement 97
mantle, 197–203, 206 of, 85 working with the Stefan-Boltzmann and
of Earth, 5, 190, 191, 197–98, 198, 202–3, particle-antiparticle pair annihilation Wien’s Laws, 115
264 and, 597 matter, 12, 19
marbles, random motions of, 164, 164 planetary, see specific planets atomic theory of, 100–102, 101, 222–23;
marching band, AGNs compared with, planetary, measurement of, 251–52 see also atoms
551–52, 552 of protostar, 429–30, 433 defi ned, 100
mare, 203 relativity theory and, 12, 77, 492, 495, 496, electron-degenerate, 448–49
Mare Ibrium, 204 497, 616 forms of, 222–23
Mariana Trench, 198 required for star formation, 169 interaction of electromagnetic waves
Mars, 7, 36, 60–61, 61, 146, 192, 192, 224, 323 of universe, 85, 501, 586, 587, 592, 611, with, 95
atmosphere of, 168, 180, 207, 215, 216, 626–27 interaction of light with, 99–108
217–20, 218 (table), 224, 226, 229, 239, weight vs., 74 luminous, see luminous (normal) matter
241–42, 241, 243, 247 mass, stellar, 7, 169 symmetry between antimatter and, 598,
craters on, 185–86, 186, 205, 209, 209, 210 of asymptotic giant branch, 455–59, 457 601
dust storms on, 239, 241–42, 242 Chandrasekhar limit and, 464 Mauna Kea Observatory, 127, 127, 144, 225,
erosion on, 207, 207, 208, 209–10, 209 H-R diagram and, 381, 382–83, 382, 382 227
formation of, 168, 169 (table), 566 Mauna Loa, 203, 205, 205
gravity of, 216, 219, 242, 243 lifetime of star and, 443–45, 447 (table) Maunder minimum, 410, 410
Kepler’s second law and, 64 measurement of, 361, 373–76, 374, 375, Maxwell, James Clerk, 314
life on, 242, 336n 377, 378 (table), 461, 504 electromagnetic wave theory of, 93–96,
magnetic field of, 196 of post-AGB stars, 456, 457 99, 100, 490, 491, 596
mass of, 218 of red giants, 449, 452 Mediterranean Sea, 281
meteorites compared with, 336 star formation and, 435–38, 444 Megabar, 264
I-14 Index
mega-light year, (Mly), 514 chemical evolution of, 572–73 dark side of, 43
M80 (globular cluster), 565 cross section through the disk of, 575, 575 eclipses and, 23, 46–51
M87 (radio galaxy), 554, 555–56, 555 dark matter in, 563, 569–71, 580, 613 erosion on, 207
Mercury, 58, 60–61, 178, 215, 217–18, 249, disk of, 563, 564, 565–66, 567, 568, 569, escape velocity from, 530, 586
251, 300, 310 571–80, 577, 615, 617, 620, 622 formation of, 169, 294–95, 300
craters of, 168, 168, 169, 186, 188, 200, 210 dust in, 443, 540, 565, 567, 569, 570, 622 gravity of, 74, 275, 279–80, 283, 461, 530
erosion on, 207 gas in, 571–75, 574, 575, 576–77 ice on, 210
formation of, 168, 169 globular clusters in, 563, 565–67, 565, 566, interior of, 191, 192, 193, 195
ice on, 210 567, 571–76, 579–80 lack of atmosphere on, 168, 207, 210, 215,
interior of, 192, 193, 195, 200 Great Attractor in, 622, 623 217–18, 242, 641
lack of atmosphere on, 168, 207, 210, 215, halo of, 563, 565, 567, 568, 570–72, 573–74, lack of magnetic field on, 195
217–18, 242, 641 576, 579–80 libration and, 286, 286
lack of moon for, 299 Herschel’s model of, 564, 564 orbit of, 23, 42–46, 43, 44, 45, 47, 50–51,
lack of rings for, 323 interstellar medium in, 418–26, 419, 423, 53, 169, 279, 284, 284, 285, 286, 319
magnetic field of, 195–96 564 phases of, 23, 43–46, 44, 45, 50–51, 51, 53,
orbit of, 64, 75, 287, 348, 497–98 key concepts about, 563 275, 282, 283
spin-orbit resonance of, 287 location of Sun in, 566, 567, 567, 568 rotation of, 284–86, 284
tectonics on, 200, 210 magnetic fields and cosmic rays in, 576– same face of, 43, 43
temperature of, 111, 119, 166 77, 577 samples of, 175, 184, 186, 188, 188, 336
volcanism on, 204, 204, 206 mass of, 85 sidereal period of, 45, 45
meridian, 26, 45 novae in, 463 size of, 195
mesosphere, 226, 227 rotation curve of, 569–70, 570, 578 synchronous rotation of, 43, 43, 284, 287,
Messier, Charles, 509 rotation of, 563, 568, 569–70 294
Messier’s catalog, 555n size of, 510–11, 563–69 synodic period of, 45, 45
Meteor Crater, 183, 183 star formation in, 563, 571–76, 571, 573, tectonics on, 200, 200
meteorites, 147, 157, 157, 175, 302, 332–36, 574, 619–20 tidal effects of, 193, 275, 280, 281, 283,
353, 358 structure of, 341n, 563, 564, 569–70, 569, 283, 284, 380
blasted from surface of Mars, 206 608, 615 as tidally locked to Earth, 284–85, 284
categories of, 333–34 Miller, Stanley, 633, 633 volcanism on, 204, 204, 206
chemical composition of, 184, 189, 191, Milton, John, 246 walk on, 7, 8, 74, 179
319, 357 Mimas, 169, 291, 300, 304, 312, 312, 320 waning of, 45
defi ned, 333 minute of arc (arcminute), 364 waxing of, 44–45
differentiation and, 191, 333 Miranda, 300, 312, 312 moons, 251, 299–302, 323–24, 338, 444
Earth hit by, 181, 184–85, 206, 331, 332, mirrors, in reflecting telescopes, 126, 130– atmosphere of, 300, 303, 306–10, 308
336 31, 131, 133, 134, 134, 135, 136 composition of, 167, 299–301, 319
Earth’s interior and, 189 M92 (globular cluster), 566, 566 co-orbital, 291
key concepts about, 327 modern physics, 12, 100, 490, 600 dead, 299, 312–13
meteoroids, 292, 322, 328, 332, 333, 353–54 molecular cloud cores, 417, 427, 427, 428, defi ned, 169
meteors, 181, 332, 333, 353–54 435–38, 439, 615 formation of, 167, 169, 299–302
sporadic, 354 molecular clouds, 417, 424–28, 425, 479, 579, formerly active, 310–12
meteor showers, 331, 354–55, 354 (table), 640 geological activity on, 286–87, 287, 299,
355, 356 collapse of, 426–28, 427, 428, 429, 435–37, 300, 302–7, 303–8, 310–13, 319–20, 323
methane, 166, 217, 220, 224, 252, 256, 257, 438, 482, 610–11, 614, 615 irregular, 302
259, 265, 271, 272, 306–9, 330, 344, 633 giant, 425–26 key concepts about, 299
Metis, 319 star formation and, 417, 426–28, 426–30, landslides on, 207
metric system, 16 435–38, 485, 575 oceans on, 299, 301, 306, 307
Mexico, 184–85, 185 molecules, 99, 112, 113, 595 as origin of ring material, 299, 304n, 319
M51 Galaxy, 138 in atmosphere, 217–20, 235 regular, 301
micrometeorites, 304n emission lines of, 424 as satellites, 80
micrometeoroids, 184, 207 forms of matter and, 222 shepherd, 292, 320–22, 321
micrometer (micron), 97 greenhouse, 220 as small worlds, 302
microscopes, electron, 134 and pressure of gas, 222–23, 223 tidal locking of, 301, 323, 461
Microsoft, 646 in protoplanetary disk, 164–65 see also specific moons and planets
microwave radiation, 99, 137, 141n, 144 self-replicating, 601 motion, 23–89
Mid-Atlantic Ridge, 202 Molyneux, Samuel, 36 anticyclonic, 254, 255, 262, 262
midnight, 45 momentum, 101 apparent, see apparent daily motion
local, 26 M104 (galaxy), 541 of charged particles in magnetic field,
Milky Way Galaxy, 4, 4, 5, 12–13, 13, 25, month, lunar, 285 228–29, 228
126–27, 138, 141, 149, 341, 443, 444, 535, Moon, 5, 22, 126, 138, 145, 146, 150, 178, 300, cyclonic, 232, 233, 254
556, 557, 558–59, 562–83, 562 300, 301, 365, 641 diurnal, 60
arms of, 569, 569, 581 ancient depictions of, 17 Doppler effect and, 108–10, 108, 109
black hole in, 563, 577–79, 579 apparent size of, 47 frame of reference and, 34–35, 39, 231,
bulge of, 565, 567, 568, 569, 574, 579, 580 craters of, 180, 182, 184, 186, 186, 188, 231, 488, 490
chemical composition of, 418, 563, 571–74 204, 206, 210 Galileo’s views on, 57, 66–67, 80
Index I-15
key concepts about, 23, 57 orbit of, 248–49, 248 (table), 328–29, 348 Niven, Larry, 352
length of object in, 493 rings of, 314, 315, 317, 318, 319, 319, 320, noon:
libration and, 285–87, 286 321 local, 26
of Moon, 23, 25, 42–46, 43, 44, 45, 47, rotation of, 248 (table), 252–53, 266 phases of the moon at, 45
50–51, 53 size of, 247–49, 250, 250 normal matter, see luminous (normal)
planetary, see elliptical orbits; specific winds on, 260, 261, 262 matter
planets neutrino cooling, 477 north celestial pole, 26–32, 27, 30, 41
prograde, 61 neutrinos, 148n, 613, 627 Northern Hemisphere, 31, 229, 236
relative, see relative motions defi ned, 394 Coriolis effect in, 232–33, 232
retrograde, 60–61 detection of, 147, 148–49 phases of the moon in, 46
uniform circular, 81 from high-mass stars, 477, 478 seasons in, 36–39, 37, 38, 39
motion, Newton’s laws of, 57, 65–72, 66, 76, solar, 389, 393, 398–400, 399, 401, 413 tilt of Earth’s axis and, 36–37, 37
79, 82–86, 151, 222, 249, 251, 293, 328, from supernovae, 398, 400, 479 North Pole, 26–32, 26, 44, 194
344, 373, 384 neutrino wave detectors, 147 latitude of, 28
as approximations to special relativity neutron, 148n seasons near, 40
and quantum mechanics, 498 neutron-degenerate core, see neutron stars and tilt of Earth’s axis, 36–37, 37, 39
fi rst, 67–68, 73, 81, 86, 159 neutrons, 530, 613 view of sky from, 26, 26, 28
playing with, 78 in atomic structure, 100, 101, 391, 476, north star (Polaris), 32, 41–42
second (a = F/m), 69, 70, 73, 74, 78, 86, 495 476n, 596 novae, 443, 459, 463–64, 463, 481–82, 552
speed of light and, 488–90, 489 decay of, 627 nuclear fusion, 5
third, 70, 72, 72, 73, 73, 86 free, 480 defi ned, 391–92
mountain climbing, journey of discovery in isotopes, 187 effects of photodisintegration on, 477
compared with, 3, 19 in quantum chromodynamics, 597 in star formation, 263, 417, 429, 432, 433,
Mt. Wilson Observatory, 511 neutron stars, 469, 481–82, 482, 487–88, 505, 439
M stars, 370, 379, 433, 437–38, 444 570, 627 stellar evolution and, 443, 444–45
M13 (star cluster), 643 in binary systems, 481–82, 482, 487–88, in Sun, 389–93, 393, 394, 398, 413, 443–
M-type asteroids, 335 501, 552 48, 595
multiverses, see universe, multiple general relativistic time dilation and, 499 see also carbon burning; helium burning;
muons, 493, 493 pulsars, 482, 483, 484, 484 hydrogen burning; neon burning
mutation, 637 New Horizons mission, 329, 341 nucleons, 476
myths, stars in, 24 new Moon, 44, 50, 51, 51 nucleosynthesis:
tides and, 275, 282, 283 Big Bang, 529–30, 530, 601, 602, 603
nadir, 26 New Orleans, La., 236 stellar, 480, 481, 484, 572, 573, 574; see
Namaka, 330 Newton, Sir Isaac, 12, 25, 66–87, 66, 348, also stellar mass loss
NASA, 145, 145 (table), 171, 172, 209–10, 210, 488, 512 nucleus, atomic, 100, 101, 106, 444–45, 448,
331, 336n, 345, 352, 356, 401, 420, 643 Kepler’s laws derived by, 79–82, 83, 85, 481, 603
Nasmyth, James, 274 92, 275, 375, 559 binding energy of, 476–77, 476
National Radio Astronomy Observatory, 8 law of gravity of, see gravity, Newton’s in electron-degenerate core, 452–53,
natural selection, 637 law of 452
nature, science in relation to, 10, 12 laws of motion of, see motion, Newton’s
navigation, stars used for, 32 laws of Oberon, 300, 312
neap tides, 282, 283 light as viewed by, 93 objective truth, 12
near earth object (NEO), 352 telescope of, 126, 127 oblateness, 253, 330
NEAR Shoemaker spacecraft, 338, 339 Newtonian (classical) mechanics, 12, 66, obliquity, 248 (table), 253, 283
nebulae, use of term, 12–13, 156, 509, 511 67–75, 86, 488 observational uncertainty, 365
nebular hypothesis, 156 clockwork universe and, 107, 120 observing time, 128
neon, 106, 252, 371 (table), 460 special relativity and, 490–91, 494 Occam’s razor, 9–10
neon burning, in high-mass stars, 472, 472 see also motion, Newton’s laws of occultations, stellar, 250–51, 251, 313
(table), 477 newtons (N), 74 oceans:
neon signs, emission and, 104 NGC 891 (spiral galaxy), 564 of giant planets, 247, 252, 264
Neptune, 67, 137, 247–52, 249, 323, 341 NGC 1132 (elliptical galaxy), 548 on moons, 299, 301, 306, 307
atmosphere of, 235, 254–59, 260, 256, 261 NGC 1532/1531 (galaxy pair), 606 oceans of Earth, 180, 184–85, 207, 221, 236,
chemical composition of, 247, 252, 259, NGC 2362 (star cluster), 487 238, 252, 264, 307
265–66 NGC 3198 (spiral galaxy), 548 development of life in, 221, 283
density of, 248 (table), 252, 265 NGC 6744 (barred spiral galaxy), 569 formation of, 217, 220
discovery of, 249–50, 317 NGC 7052 (galaxy), 555 magnetic structure of, 196–97, 197
formation of, 168, 252, 266 nickel, 5, 167, 191, 335, 459, 464, 642 rifts in floor of, 196–97
Great Dark Spot (GDS) on, 255–57, 257, night, 52, 229, 252–53 temperature of, 238
261 Nightfall (Asimov), 566 tidal forces and, 279–87, 281, 282
interior of, 264–66, 264, 272 nitrogen, 5, 344, 371 (table), 481, 633, 638 waves of, 419
magnetic field of, 266, 267–68, 267 in atmosphere, 217–19, 218 (table), 221 Olympus Mons, 205–6, 205, 211
magnetosphere of, 266–67 (table), 222, 229, 241, 306–9, 613 Oort, Jan, 340
mass of, 248 (table), 251, 265 in giant planets, 252, 269 Oort Cloud, 217, 340–41, 340, 344, 347, 349,
moon of, 249, 300, 302, 304–6, 318 in planetary nebulae, 459 352
I-16 Index
opacity, 395–96 oxygen burning, in high-mass stars, 472, 472 Phoebe, 302
open clusters, 571, 572–73 (table), 475, 477 Phoenix, Ariz., 32, 33
open universe, 591, 592 ozone, 225, 241, 243, 422 Phoenix lander mission, 209, 210
Ophelia, 321 ozone hole, 225 Phoenix Mars lander, 207
Opportunity Mars rover, 207, 209, 209 phosphorus, 259, 271, 638
optical detectors, 137–41 pacemakers, 295 photinos, 613
optics, limitations of, 129, 132 pair production, 597 photochemical, 257
orbital resonance (orbital interaction), 275, paleomagnetism, 194, 196, 197 photodisintegration, 477
289–93, 323, 328–29 palimpsests, 310, 311 photodissociation, 224, 308
asteroids and, 275, 289, 290 Pandora, 320–21, 320 photoelectric effect, 99
Cassini Division and, 291 parabolic mirrors, 127, 130–31, 131 photoelectric photometers, 139
Lagrangian points and, 292–93, 293 parabolic orbits, 85 photographic plates, 138–39, 139, 141
orbiter, 145–46 parallax, 362–67, 364, 378 (table), 381, 565 photography, in astronomy, 138–39, 138,
orbits, 79–87 distance ladder and, 517 139, 140
of binary stars, 361, 373–74, 374, 375, 461, spectroscopic, 381, 385 photometers, 139
501, 504 parent elements, 187 photons, 91, 99, 114, 116–17, 121, 137, 138,
bound, 84, 85 parsec, 364–66 140, 399, 596, 613, 627
chaotic, unpredictable, 294–95, 295 Parsons, William, 138 absorption of, 104–7, 104, 422
circular, see circular orbits partial solar eclipse, 46, 47 atomic decay and, 102–5
of comets, 58, 83, 217, 340–45, 346–49, particle accelerators, 147 defi ned, 99
348, 352, 354 particles, particle physics, 148–49, 411, 503, in early universe, 597, 598
decay of, 80n, 412 503, 585, 595–603, 597 emission of, 102–7, 104, 422, 424
defi ned, 80 cosmology vs., 600 gamma ray, 394, 597, 597
eccentric, 172 massless, 99 in interstellar medium, 422–23, 424
elliptical, see elliptical orbits quantum electrodynamics and, 595, 596 in quantum electrodynamics, 596, 596
escape velocity and, 82–83 standard model of, 597 as quantum of light, 99
gravity and, 58, 79–87, 79, 80, 81, 83, 84 temperature as measure of movement of, in radiative transfer, 395, 395
hyperbolic, 85 113 scattering of, 240, 240
of Jupiter, 170, 171 wavelike properties of, 100, 101 photosphere, 389, 402–4, 403, 404, 407, 413
Kepler’s second law and, 63–64, 63 see also axions; electrons; neutrinos; photosynthesis, 59
of Moon, 23, 42–46, 43, 44, 45, 47, 50–51, neutrons; photinos; photons; protons physical laws, 3, 10, 19, 120, 188, 191, 211,
53, 169, 279, 284–86, 284, 285, 319 Pasiphae, orbit, 294, 295 296
of moons, 301–2, 304 Pathfinder mission, 206 defi ned, 9
noncircular, tides and, 285–87, 286 patterns, 19 discovery of, 59, 65
parabolic, 85 life and science made possible by, 14, 14 empirical rules vs., 65
period of, 64–65, 64 (table), 82, 284–85, mathematics as language of, 15, 15 hypothetical, 65
284, 286–92, 290, 346–49, 348, 375–76 proportionality and, 57 of Newton, see gravity, Newton’s law of;
planetary, see specific planets understanding and, 3, 15 motion, Newton’s laws of
of planetary rings, 314, 322 see also sky patterns proportionality and, 71
prograde vs. retrograde, 347, 348 Pauli exclusion principle, 448n structure and, 595–601, 596, 597, 598, 601,
of protostellar disk, 158 peculiar velocities, 517, 622, 623 607
retrograde, 302 Pelée, Mount, 202 physics, modern, 12
stellar galaxies shaped by, 535, 537, 539, penumbra, 407, 408 physics laboratories, 399
539 (table), 540, 541, 571 defi ned, 47 Piazzi, Giuseppe, 330, 331
unbound, 85 lunar eclipse and, 50 Pioneer 11 probe, 250, 269, 314, 320, 643, 643
organic materials, 166, 167, 168, 259, 308, partial solar eclipse and, 46, 47, 48 Pioneer Venus mission, 235
319, 327, 425 Penzias, Arno, 525–26, 526, 528 Pisces, 39, 41n
in comets, 306, 342, 344, 346, 349, 352 perihelion, 287, 346, 348 pixels, 139, 140
prebiotic, 633 period: Planck, Max, 99, 100, 114–15
in terrestrial life, 640 of orbits, 64–65, 64 (table), 82, 92, 284–85, Planck era, 600, 602
Orionids, 354 (table) 284, 286–92, 290, 314, 346–49, 348, Planck mission, 611, 622
Orion Nebula, 423, 423, 425 375–76 Planck’s constant, 99, 101, 103
O stars, 444, 457, 464, 475, 543–44, 554 of wave, 96, 189 Planck (blackbody) spectrum, 114–15, 114,
formation of, 422 period-luminosity relationship, for Cepheid 115–17, 120, 269, 367–69, 368, 369, 372,
in H-R diagram, 379 variables, 473, 511, 567 395, 503, 597
life span of, 422 Perseids, 354, 354 (table), 355 Big Bang and, 525–28, 526, 527
spectra of, 370, 370 Perseus, 564 of Sun, 403
outcrops, 209 perspiration, thermal equilibrium plane, 16
Owl Nebula, 458 maintained by, 112 planetary motion, 25, 57–87, 58, 537
oxygen, 5, 107, 216, 218, 269, 334, 344, 371 petroglyphs, 17 chaos and, 294–95
(table), 460, 481, 638, 642 Phaethon, Asteroid, 354 (table) Kepler’s fi rst law of, 62–63, 62, 63
in atmosphere, 218–21, 218 (table), 221 philosophy, 19, 57 Kepler’s second law of (Law of Equal
(table), 222, 224, 229, 241, 243, 613 Greek, 25 Areas), 63–64, 63, 83
in planetary nebulae, 459 Phobos, 300, 323, 338–39, 339 Kepler’s third law of (Harmony of the
Index I-17
Worlds), 57, 64–65, 64 (table), 65, 82, 85, of giant planets, 260, 261 aggregation in, 155–56, 162–63, 163, 438
251, 375–76 wind patterns and, 230 composition of, 166, 166, 264, 266, 352,
Newton’s theory of, 78–86, 249, 251, 294 Pompeii, 202 417
origins of Solar System and, 157 Poseidon, Temple of, 22 defi ned, 158
Sun’s influence on, 76, 79 positrons, 148n, 394, 482, 484, 484, 597–98, temperature of, 155, 164–66, 164, 166, 169
thought experiment and, 79–82, 79 597, 627 protostars, 166, 169, 427–38, 443
see also elliptical orbits; specific planets in early universe, 597, 598 atmosphere of, 433, 449
planetary nebula, 13, 443, 456–58, 457, 458, post-AGB stars, 456–59, 457, 459 collapse of, 159–60, 417, 433, 435–37, 439,
460, 464, 472, 572 see also white dwarfs 443, 445, 449, 451, 453
Crab Nebula compared with, 484 potential energy, 164–66, 234 defi ned, 158
planetary systems, 170–72, 174–75 gravitational, 164–66, 263, 264 evolution of, 428–35, 431, 432, 434, 438–
birth and evolution of, 155–57 Pournelle, Jerry, 352 39, 443–44, 449, 552–53
planetesimals, 163, 163, 166, 166, 168, 168, power, defi ned, 113 formation of accretion disk around, see
169, 254, 266, 292, 300–301, 319, 327– prebiotic organic molecules, 633, 641 protoplanetary disks
28, 331, 332, 333n, 352, 634 precession of the equinoxes, 42 H-R diagram and, 417
fate of, 334–35, 335 predation, 637 protostellar disks, 155–56, 158–59, 161, 175,
Trojan asteroids, 293, 293 pressure, 221, 222–23, 223, 229 176, 427–28, 428, 429, 443
see also asteroids; comets atmospheric, 215, 218, 218 (table), 221, material flow patterns in, 435, 436–37
planet migration, 167 225, 235, 238 Proxima Centauri, 364
planets, 62–63, 125, 150, 159, 166, 444 hydrostatic equilibrium and, 191, 221 pseudoscience vs. science, 501
albedo of, 118 in protostar, 429–30, 431, 432, 444, 445 Ptolemy, 24, 41, 60, 60
classical, 4 size and, 265 pulsars, 469, 482, 483, 484, 501, 505
composition of, 172 structure of Sun and, 389, 390, 390, 391, in Crab Nebula, 484, 484
defi nition of, 172 397–98, 448–49 pulsating variable stars, 473, 567
determination of temperature of, 110–11, ultrahigh, giant planets and, 265, 272 Cepheid, 473, 474, 505, 511, 565, 567
112, 117–20, 118, 119 weight and, 191, 193, 202, 221, 390, 397– pupil, 129, 137
Earth-like, see terrestrial planets 98, 469, 471
escape velocity from, 530 primary atmosphere, 167, 168, 216, 218, Quadrantids, 354 (table)
exploration of, 146–47, 207, 209–10 244 quantum chromodynamics (QCD), 596
extrasolar, 7, 170–72, 171, 173, 174–75, primary mirror, 127, 133, 135 quantum efficiency, 137, 138, 139
174, 176, 224 primary waves, 189–90, 190 quantum electrodynamics (QED), 596, 596
formation of, 155–70, 216, 252, 263, 264– primordial energy, 263 quantum mechanics, 12, 106, 295, 423, 503
65, 357 Primordial Era, 626 confrontation between relativity theory
“naked eye,” 60 principles, scientific, defi ned, 9–10 and, 600
as satellites, 80 prisms, 131, 131, 141 defi ned, 99, 585
thermal equilibrium of, 112 probes, robotic, 124, 129 density of stellar core and, 448, 452, 455,
thermal radiation from, 114–15, 117–20, see also rovers 477, 481
118, 119 prograde motion, 61 early universe and, 593–95, 611, 624
see also giant planets; terrestrial planets; prokaryotes, 635, 635 Einstein’s problems with, 12, 101, 107–8
specific planets Prometheus, 320, 321 Newton’s laws of motion and, 498
plant life, atmospheric oxygen and, 223–24 prominences, 403, 410, 411 probabilities and, 101, 107, 120
plasma, 227–28, 267, 269–70, 411 proportionality, 57, 70 thinking outside the box required by, 100
associated with Io, 269–70, 270, 271, 271, constants of, 71 uncertainty principle in, 593–95
302 defi ned, 16, 71 Quaoar, 341
plate tectonics, 179, 197–203, 205–6, 211, inverse square laws and, 76 quarks, 596
307 in universal law of gravitation, 76 quartz, 183
convection and, 197–98, 198, 199 proteins, 307, 633, 638 quasars, 269, 501, 504, 549–53, 550, 551, 555,
defi ned, 197 Protestants, 42 558, 559, 618
paleomagnetism as record of, 196, 197 protogalaxies, 580 Quaternary period, 184
volcanism and, 198, 199, 202–3 proton-proton chain, 393–94, 394, 398, 445,
Pleiades (Seven Sisters), 437, 437 470, 470 radar, 36
Pluto, 4, 63, 119, 156, 169, 250, 254, 299, 300, protons, 72n, 106, 148, 148n, 228, 269, 391– radial velocity, defi ned, 109
327, 328–31, 329, 340–41 94, 422–23, 480, 554, 576 radial velocity, spectroscopic, 170, 171
chemical composition of, 329 in atomic structure, 100, 101, 391, 476, radian, 364
discovery of, 4, 250, 328 476n, 596 radiant, 354, 355
as dwarf planet, 4, 13 decay of, 598, 627 radiation, 8, 477
mass of, 328 in early universe, 529, 530 continuous, 113
moon of, 285, 299, 300, 328–30, 329 grand unified theory and, 598 cosmic background, see cosmic
orbit of, 63, 63, 254, 294, 329 in isotopes, 187 background radiation
size of, 328–29 magnetized, 423–24 electromagnetic, see electromagnetic
temperature of, 111, 119, 166 in motion vs. at rest, 492 radiation; light
Polaris (north star), 32, 41 in quantum chromodynamics, 597 Hawking, 503, 503, 587, 627
polar regions: protoplanetary disks, 162–69, 176, 216, 417, interstellar extinction and, 418–19, 419
on Earth, 26–32, 26, 44, 194–95 438 interstellar medium and, 418–26, 419–25
I-18 Index
radiation (continued) refracting telescopes, 125, 126–27, 126, 129, see also specific planets
as inverse square law, 76, 116–17, 117, 129, 132, 134 rotational symmetry, 278
371–72, 378 (table), 381 refraction, 130, 130, 133 rotation curve, 546, 547
from quasars, 501 index of, 92, 130, 130, 130n, 131 flat, 547, 548, 570, 570, 612, 615, 618
structure of universe and, 612, 612 refractory materials, 166–67, 166 of Milky Way, 569–70, 570, 578
thermal, see thermal radiation regular moon, 301 rovers, 124, 146, 147, 207, 207, 209
21–cm, rotation of Milky Way measured relative humidity, 227 RR Lyrae variables, 473, 567
with, 568, 569 relative motions, 34, 151, 231 Rømer, Ole, 92, 93
radiation belts, 229, 268–71 gravity and, 279–80 Russell, Henry Norris, 376–78, 384
radiative transfer, 395, 395, 396 inertia and, 67 Russia, 42
radiative zone, 395, 396 relativistic beaming, 555 Russian Federal Space Agency (Roscosmos),
radioactive (radiometric) dating, 186, 187, relativity theory, see general relativity 145
188, 195, 197, 333 theory; special relativity theory
radio frequency maps, 141 remote sensing, 145 Sagittarius, 39, 41, 418, 569
radio galaxies, 550, 551, 553, 555 Renaissance, 12, 19 Sagittarius Dwarf, 574, 580
M87, 554, 555–56 reproduction, 637 San Andreas Fault, 198
radiometric dating, 186, 187, 188, 195, 197, resistance, electrical, 235 sand dunes, 207, 207
333 resolution, 125, 133, 134–35 satellite observations, 7–8
radio telescopes, 8, 8, 137, 141–44, 142, 551 angular, 142–43 satellites, 80–85
cosmic background radiation discovered rest wavelength, 108–9 artificial, 7–8, 8, 80–81, 412
with, 525–26, 526 retina, 129, 129, 137, 138 disruption of, 288
radio waves, 8, 98, 99, 125, 137, 141–42, 228, retrograde motion, 60, 60, 61 Global Positioning System, 499–500
268, 269, 353, 482, 484, 549–51, 550 retrograde orbits, 302 planet capture of, 294, 295
of interstellar medium, 417, 424, 425 ribonucleic acid (RNA), 633, 640, 642 see also Moon; moons; planets; specific
photons of, 99 Rima Ariadaeus, 200 moons and planets
rain, 227, 233, 234, 234, 243, 262 ring arcs, 314 saturation, 227
rainbow, spectrum of, 97, 98 ringlets, 314, 316 Saturn, 60–61, 128, 146, 167, 246, 247–55,
rainfall, direction of, frame of reference Ring Nebula, 458 248, 259–69, 323
and, 34–35, 35 rings, planetary, 313–23, 315, 324, 461 atmosphere of, 235, 252, 254, 254, 255,
ratios, defi ned, 16 backlighting and, 317, 317, 318, 322 255, 256, 259–60, 314
Reber, Grote, 141 origin of material for, 299, 319 auroral rings of, 270
reddening effect, 419, 420 see also specific planets and rings Cassini Division and, 291, 291, 314, 316,
red giants, red giant branch, 418, 472, 477 RNA (ribonucleic acid), 307 316
asymptotic giant branch compared with, robotic probes, 124, 129 chemical composition of, 247, 252, 259,
455–56, 456 Roche, Edouard A., 288, 461 263–66, 272
degenerate cores of, 443, 449, 450, 451, Roche limit, 287, 288, 289, 292, 296, 319, density of, 248 (table), 252, 263–65
452–55, 452, 456, 463, 477 323, 461 differentiation in, 264
evolution of, 443, 450–54, 456, 460, 460, Roche lobes, of binary stars, 288, 461, 482, formation of, 168, 252, 264–66
463, 465, 472 482 interior of, 263–66, 264, 272
helium burning in, 443, 452–55, 453 rocket ship, speed of light and, 492 magnetic field of, 266, 267, 268, 322
red light, 367–68 rocks, 166–67, 323 magnetosphere of, 268, 270
scattering of, 240, 240 of Earth, 187, 189–91, 194–95, 198, 206 mass of, 248 (table), 251, 264, 265
spiral galaxy viewed in, 544 in giant planets, 264, 265–66 moons of, 169, 217, 270, 291–92, 293, 293,
redshifted light, 612 lunar, 184, 186, 188, 188, 204, 204, 206, 299, 300, 301, 302, 304, 304, 308, 310,
cosmological, 524 207 312, 314, 319–21, 320–21, 323
defi ned, 109, 121 on Mars, 207, 209 orbit of, 248 (table), 267, 268, 322
discovery of Hubble’s law and, 509, 513– in moons, 300, 302, 304–6, 310, 310, 312, rings of, 169, 254, 272, 291–92, 294, 298,
14, 516, 524 323 304, 315, 316, 320–23, 324
Doppler effect and, 108, 109, 109, 374, 523, in planetesimals, 168 rotation of, 248 (table), 252–53
524, 524, 528 radioactive dating of, 179, 186, 187, 188, size of, 247, 250, 250, 254
of galaxies, 509, 513–14, 516, 518, 523–25, 195, 197 storms on, 254, 254, 256
524, 587, 588, 608–9, 610, 622 rods, 137 temperature of, 166, 169, 265
gravitational, 499, 500, 524, 529 Roentgen, W. C., 98 winds on, 254, 255, 256, 260, 261
of Planck spectrum, 525, 526, 527, 528 Romans, ancient, 43, 66–67, 283 scale factor, of universe, 522–25, 523, 586,
of quasars, 549 “room at the center of the Earth” problem, 587, 588, 589, 591
scale factor and, 523–25, 525 277, 277, 278 scattering, 239, 240, 240, 252, 259
redshift surveys, 608–9, 610 Rooney, Andy, 178 Schiaparelli, Giovanni, 640
red supergiants, 373 rotation: Schmidt, Maarten, 549
reflecting telescopes, 125, 126, 127, 129, of asteroids, 336 science:
133–34, 134, 136 equal to orbit, 287 aesthetics of, 296
reflection: of moons, 301–2 arts compared with, 3, 10, 11
angle of, 130, 131 of object, tied to orbit by tides, 275, 284– collaborations in, 59
of light, 130–31, 131 87, 284 as creative activity, 3, 10
reflection symmetry, 278 synchronous, see synchronous rotation culture and, 13
Index I-19
empirical, 57, 59, 85, 86 Shoemaker, Gene, 338n Earth’s magnetic field and, 215, 228, 229
lasting value of, 10 Shoemaker-Levy 9, 349–50 magnetosphere and, 267, 268, 269, 270,
mathematics as language of, 15 destruction of, 184, 287, 289, 349–50, 350– 407
patterns and, 14–15 51, 357 measurements of, 401
rise of, 34 short-period comets, 345, 347–48, 348 solids:
technology compared with, 9 Siberia, 350–51, 353, 357 defi ned, 222
theoretical, 65 sidereal period, 45, 45 thermal conduction in, 393
as way of viewing the world, 9–13 Sikhote-Alin region, 352n solstice:
science fiction, 566 silica, 191 summer, 37, 39, 40–41, 362
scientific method, 9, 15, 85, 86, 209n silicates, 166, 265, 301, 301, 304, 306, 319, winter, 39, 40
scientific notation, 15, 16 334, 642 sound:
scientific revolutions, 12, 19, 85 silicon, 5, 252, 334, 371 (table), 418, 640, 642 Doppler effect and, 108
of Copernicus, 57, 175 in high-mass stars, 475, 477 speed of, 551
of Einstein, 12, 12, 490 singularity, 502, 502 south celestial pole, 26, 27, 28, 30, 31
Kuhn’s views on, 11 Sirius (Alpha Canis Majoris or the Dog Star), Southern African Large Telescope, 128
scientific theory: 24, 365, 459 (table)
cosmological principle as, 79 61 Cygni, 365 Southern Hemisphere, 26, 28, 32, 229, 236,
failure of, 119 sky patterns, 23–25 580
as falsifiable, 9, 501, 624 ancient views of, 14, 23–25 constellations visible from, 24
interplay between observation and, 191 Earth’s orbit and, 32, 35–41, 35, 51–52 Coriolis effect and, 233
as nonarbitrary, 13, 73 Earth’s rotation and, 26–32, 26–32 seasons in, 37, 37, 39
old vs. new, 497 life patterns and, 24, 25 tilt of Earth’s axis and, 37, 37
rise of, 65–68, 85, 86 of phases of the Moon, see Moon, phases South Pole, 194
sources of, 73–76 of latitude of, 28
testable predictions of, 9, 73, 75, 151, 501 “Slice of the universe, A” (redshift survey), seasons near, 40
use of term, 9 609–10, 610 tilt of the Earth’s axis and, 37
see also specific theories Slipher, Vesto, 513–14 view of sky from, 28
Scott, David, 74 Sloan Digital Sky Survey, 611, 622 Soviet-American Gallium Experiment
Scutum-Centaurus, 564 Small Magellanic Cloud, 517, 571, 580, 580 (SAGE), 400
Search for Extraterrestrial Intelligence snowball rolling downhill, red giant Soviet Union, 186, 239, 343
(SETI), 645–47 compared with, 450 Sputnik 1 launched by, 7, 8
seasons: sodium, 5, 106, 269, 302, 472 spacecraft, 144–47, 207, 209–10
change of, 23, 24, 36–43, 52, 253 software, image-processing, 139 flyby missions of, 145–46
Earth’s wobbling axis and, 41–42 solar abundance, 252 information about giant planets supplied
leap years and, 42 Solar and Heliospheric Observatory (SOHO), by, 250, 252
on Mars, 241–42 401, 407, 409, 411 in measurement of mass of planet, 250,
near Earth’s poles, 40 solar eclipse, 46–51, 46, 47, 404, 405 252
near Equator, 40 annular, 46, 47, 49 orbital missions of, 145–46, 171, 172
tilt of Earth’s axis and, 23, 36–38, 52 gravitational lensing and, 498–99, 499 orbit of, 58, 80, 80, 276, 412
secondary atmosphere, 167, 217–20, 223, 244 lunar eclipse compared with, 48, 50 reconnaissance, 145
secondary craters, 183, 183 partial, 46, 47, 408 samples returned by, 147, 204
secondary mirrors, 127, 134 total, 46–50, 47 stationary vs. high velocity, 494, 495
secondary waves, 189–90, 190 solar flares, 389, 411, 411, 412 see also specific spacecraft
second of arc (arcsecond), 364 solar maximum, 409, 410, 412 space exploration, 7–8, 52, 144–47, 207, 209,
sedimentation, 187, 207, 209 solar neutrino problem, 399–400, 399, 402 210
Sedna, 341 solar neutrinos, 389, 393, 398–400, 399, 401 see also specific missions
seismic waves, 189–91, 190 solar neutrino telescopes, 398–400, 401 space probes, 7–8, 8, 157, 179
seismometers, 190–91 Solar System, 4, 4, 16, 120, 125, 141, 154– “spaceship traveling at half the speed of
selection effect, 172 357, 156, 421, 614–16 light” experiment, 488–89, 489
selenium, 640 age of, 333 space shuttle, 491, 495
self-gravity, 276–79, 276, 277, 287, 346, 464, AGNs compared with, 551 spacetime, 491, 600
614, 615 debris in, 353–56 Big Bang and, 501, 509, 588, 591
of molecular clouds, 426, 426, 427 exploration of, 7–8, 8, 144–47, 179, 207, gravity as distortion of, 469, 494–505
Selfish Gene, The (Dawkins), 637 209, 210 mass in distortion of, 496–97, 497, 505
semimajor axis, 62, 63, 63 as gravitational symphony, 58 as rubber sheet, 496, 497, 591
SETI inner vs. outer, 299 see also time
SETI Institute, 646 key concepts about, 155 special relativity theory, 12, 110, 392, 489,
Seyfert, Carl, 550 measurement of distances in, 36 490–93, 495, 524, 556, 646
Seyfert galaxies, 550–51, 553, 558 nonsolar mass of, 251 frame of reference and, 490–93
Sgr A*, 572, 572, 579 origins of, 5, 58, 151, 155–70, 323, 332, implications of, 491–93
Shapley, Harlow, 510–11, 565 333, 341, 344, 357, 417, 428, 438–39 particle-antiparticle pair annihilation
size of Milky Way determined by, 567 see also specific planets and moons and, 596
shepherd moons, 292, 320–22, 321 solar wind, 147, 242, 244, 346, 389, 405, 406, species, extinction of, 184–85, 185, 349, 352,
shield volcanoes, 202–3, 206 407, 413 358
I-20 Index
spectral fi ngerprints, 91, 106–7, 106 Type I supernovae as, 517–18, 518, 587, main sequence of, see main sequence of
spectral range, 139 588 stars; Sun
spectral types, 369–70, 370, 380, 382 (table) see also globular clusters in mass of galaxies, 535
spectrographs, spectrometers, 132, 141, 141, standard model, 596–97 nuclear fusion reactions within, see
170, 224, 307, 609 star clusters, 288, 437, 437, 438–39 nuclear fusion
spectroscope parallax, 381 as snapshots of evolution, 469, 484–87, parallax of, see parallax
spectroscopic binary stars, 376 485, 486, 487 patterns of, 24–25, 361
spectroscopic radial velocity, 170, 171 see also globular clusters; open clusters properties of, 361, 366–67, 379–84, 379
spectroscopy, 141 Stardust mission, 345, 345 protostar vs., 433, 449
spectrum, 99, 103–10, 141 star formation, 156–62, 166–67, 169, 174, shape of galaxy due to, 535, 537, 539, 539
of asteroids, 335, 338 263, 294, 378, 416, 443–44, 484–87, 572, (table), 540, 541
composition of stars and, 361, 368–71, 602 size of, 361, 367, 371–72, 372, 378 (table),
369, 370, 378 (table) accretion disks and, 157, 158–62, 158, 379–84, 389, 396–97, 449–50, 640
defi ned, 97 614–16 as source of chemical elements, 5, 7, 7,
Doppler effect and, 108–9, 108 conservation of angular momentum and, 458–59, 496–479–481
electromagnetic, 97–99, 99, 125, 136–37, 158–60, 164, 175 spectral type of, 368–70, 370, 380, 382
141, 144, 151, 367–68, 484; see also in galaxies, 535, 540–45, 559 (table)
specific kinds of radiation galaxy formation compared with, 610–11, spectroscopic binary, 376
Planck, see Planck (blackbody) spectrum 614–16 tidal effects and, 289–90
of Sun, 402–4, 404, 405 gravitational energy converted to thermal uniqueness of, 472
temperature and, 367–70, 369, 370, 372, energy in, 429–33, 431, 432, 439, 553, see also Sun; specific types of stars
419, 420 615 Star Trek, 597
Wien’s law and, 115, 116, 367–68, 370, 372, gravitational instability and, 610–11, 615 Star Wars, “Death Star” from, 312, 312
419 interstellar medium and, 417–40, 573–74, static equilibrium, 110
see also absorption lines; emission lines 573 Stefan, Josef, 115
sphere, net gravitational force at surface of, key concepts about, 417 Stefan-Boltzmann constant, 115
276, 276 in Milky Way, 563, 571–76, 571, 573, 574, Stefan’s law, 115–16, 118, 151, 263, 428
spherical symmetry, 276, 278 619–20 size of stars and, 367, 371–72, 372, 378
spin-orbit resonance, 287, 289 planetary systems as by-product of, 156– (table), 379, 384, 434
spiral density waves, 545 57, 172, 174, 174 stellar evolution:
spiral galaxies, 13, 418, 510–11 in young galaxies, 619–20, 621 of intermediate-mass stars, 472
arms of, 535, 542–43, 544, 545, 559 see also protostars mass and, 443–45, 463–64; see also high-
barred, 534, 538, 539 (table), 541, 545, 563, stars, starlight, 277, 549 mass stars, evolution of; low-mass
569, 569, 580, 582 aberration of, 34–36, 35, 81 stars, evolution of
bulges in, 539, 539 (table), 541, 544–45, age of, 469, 484–87, 485, 486, 487, 566, star clusters as snapshots of, 469, 484–87,
557 571–74, 571 485, 486, 487
classification of, 537–42, 538, 539 (table) ancient depictions of, 17 stellar explosions, 417, 418
color of, 540 apparent movement of, 23, 25–32, 28, 30, see also supernovae
dark matter in, 547–49, 548, 563, 569–71, 31, 32–33, 32, 39 stellar luminosity, see luminosity, stellar
580, 618 binary, see binary stars stellar mass, see mass, stellar
disks of, 539–45, 563 brightness of, 138, 171, 171, 361, 366–67, stellar mass loss:
edge-on, 563, 564 366 in high-mass stars, 475, 475
formation of, 544–45, 613, 614–18 circumpolar, 31 in low-mass stars, 455–59, 457, 458
light of, 546, 547 classification of, 361, 368–71, 369, 370 stellar occultations, 250–51, 251, 313
mass of, 546, 547, 548 color of, 361, 367–68, 368, 380–81, 382 stellar population, 487
merger of, 618, 619 (table), 487 stellar temperature, see temperature,
rotation curve of, 547, 547, 548, 569–71, composition of, 149, 361, 370–71, 371 stellar
570, 578, 612, 615, 618 (table), 378 (table), 383, 396–98, 402, stellar winds, 544, 572, 575, 578
rotation of, 544–45, 545, 546 404–5, 413, 418, 444–49, 448, 454–55, Stelliferous Era, 626, 626
Seyfert, 550–51, 553, 558 459, 460, 571–74, 573 stereoscopic vision, 361, 362, 362, 367
star formation in, 540–45, 544, 559 core of, see core, stellar Stonehenge, 46, 46
unbarred, 538, 539 (table) death of, 5, 7, 265, 294, 378, 418, 477–80 stony-iron meteorites, 334, 334, 335, 357
see also Milky Way Galaxy density of, 367 stony meteorites, 333, 334, 334
spiral nebulae, 509–11, 535, 564 failed, see brown dwarfs storms:
Spirit Mars rover, 207, 209 halo vs. disk, 576, 579–80, 620 on giant planets, 247, 254, 254, 256,
Spitzer Space Telescope, 144, 437 high-velocity, 576, 579 261–62
spoke, 322 interiors of, 367n see also dust storms; hurricanes;
spreading centers, 196, 198, 202–3 key concepts about, 361, 417, 443 thunderstorms
spring, 39, 42 in legends and myths, 24 stovetop convection, 197–98
spring balance, 430, 431 life of, 5, 443–65; see also stellar stratosphere, 226, 227, 229, 234, 234, 257
spring tides, 282, 283 evolution Stratospheric Observatory for Infrared
Sputnik 1, launch of, 7, 8 luminosity of, see luminosity, stellar Astronomy (SOFIA), 144
stable equilibrium, 110–11 luminosity-temperature-radius stromatolites, 634, 635
standard candles, 566, 567, 622 relationship for, 373 strong force, 601, 602
Index I-21
strong nuclear force, 391–92, 446n, 453, 470, gravitational redshift on, 499 intercloud gas and, 421, 421
478, 595–96, 598, 601, 602 gravity of, 58, 75, 275, 282, 289, 294, 296, neutrinos from, 398, 400, 479
quantum chromodynamics and, 596 341, 389, 390, 390, 397–98, 405, 413, synchrotron radiation and, 269
Structure of Scientific Revolutions, The 444, 445, 448–49 Type I, see Type I supernovae
(Kuhn), 11 great circle of, 32, 39 Type II, see Type II supernovae
structure of the universe, 607–27 greenhouse effect and, 219–20 superstring theory, 600–601, 601
distribution of galaxies and, 607, 608–10, hypothetical star compared with, 396–98, surface brightness, 138
609, 610 397 surface waves, 189
evolution of, 607 importance of, 4, 389 Swift Gamma-Ray Burst Mission, 145 (table)
gaps in models of, 621, 621 interior of, 390–402, 390, 391, 395, 396, Swift-Tuttle, Comet, 354, 354 (table)
general relativity and, 498, 501, 600 402, 403, 410, 413 symmetry, 598
hierarchical clustering in, 611, 617 Jupiter compared with, 252, 265 gravity and, 275, 278, 598
Hubble’s law and, 519–21, 520, 608–9, 622 key concepts about, 389 types of, 278
key concepts about, 607 location of, 566, 567, 567, 568, 581 synchronous rotation, 43, 43, 284, 287, 294,
life and, 607 luminosity of, 367, 389–90, 397, 445, 449, 301
physical law and, 595–601, 596, 597, 598, 455–56, 464 of Pluto-Charon system, 328, 329
601, 607 magnetic activity on, 389, 404–7, 405–12, synchrotron radiation, 268, 269, 269, 271–72
S-type asteroids, 334, 337–38 409–11 from AGNs, 551, 554, 578
Subaru telescope, 128 (table) mass of, 82, 85, 373, 389, 397, 398, 435, from cosmic rays, 576–77
subduction zones, 198, 202 437, 443, 444, 445, 448, 521 of Crab Nebula, 484, 484, 551
subgiant stars, subgiant branch, 449, 460 Moon’s orbit around, 42–43 in Milky Way, 576–77, 579
sublimation, 342, 344 at noon, 26 synodic period, 45, 45, 50
Sudbury Neutrino Observatory, 613 other stars compared with, 361, 367, 371, Système International d’Unités (SI), 16
sulfur, 209, 252, 259, 269, 271, 302–4, 351, 373, 389 S0 (S-zero) galaxies, 537, 538, 539 (table), 559
371 (table), 475, 638 phases of Moon and, 43–46 Szilard, Leo, 645
sulfur dioxide, 302–3, 349 photosphere of, 389, 402–4, 403, 407
sulfuric acid, 238 planetary atmospheres and, 217–20 Taurids, 354 (table)
Sumatra, 202 planetary motion influenced by, 75, 79 Taurus, 39, 41, 437, 437, 482
summer, 36–41, 37, 38, 40, 229 pressure of, 389, 390, 390, 391, 413, 444, Taurus-Gemini star field, 61
summer solstice, 37, 39, 40–41, 362 445 Tecciztecatl, 43
Sun, 3, 5, 150, 195, 285, 365, 389–413 as protostar, 158, 168, 417 technology, 19
age of, 566, 571, 572 red giant branch star compared with, 450 science compared with, 9
ancient depictions of, 17 red giant phase of, 443 see also tools, astronomical
apparent motion of, 25, 26, 34, 39, 39 rotation of, 407 tectonism:
apparent size of, 47 seasonal changes and, 36–41, 37, 38, 39 defi ned, 180
atmosphere of, 383, 389, 402–5, 403 seismic vibrations on, 389, 399–402, 402 on moons, 304, 306, 310
backlighting and, 317, 317, 318 size of, 389, 396–97, 450 in shaping terrestrial planets, 179, 180–
balance in structure of, 389–98, 390, 391, spectrum of, 103, 113, 115 81, 186, 196–202, 196, 197, 198, 199, 200,
444, 445, 447, 473 temperature of, 115, 389, 390, 391, 393, 211, 302
as center of universe, 57, 58, 61, 67, 85, 175 395–96, 395, 402–4, 403, 413, 433, 444, telescopes, 8, 341, 551
chemical composition of, 371, 371 (table), 445, 448–49 airborne, 144
397, 402, 404–5, 412–13, 418, 445–49, and temperature of planets, 111–12, 117– aperture of, 125, 132, 133, 142
448, 574 20, 118, 119 atmospheric distortion and, 125
circumpolar, 40 testing models of, 389, 398–402, 401, 402 cost of, 128
comets and, 339–45, 346–49, 358 tidal effects of, 193, 275, 282, 283, 287, 289 diffraction limit of, 134
core of, 389, 391, 392–98, 399, 413, 447–49 22–year cycle of, 389, 410, 410, 413 focal length of, 125
density of, 390, 391, 392–93, 396–98, 400– volume of, 417 Galileo’s use of, 67, 249, 299
401, 402, 403, 413, 444, 448 sungrazers, 349 history of, 126–27
as disk star, 576 sunlight (solar radiation): image size of, 125, 132
Earth’s orbit around, see Earth, orbit of photons emitted by, 103 infrared, 141, 144
eclipse of, see solar eclipse received by Earth, 411–12, 412, 639 key concepts about, 125
ecliptic path of, 33, 36, 39, 41 sunspot cycle, 408, 409 light-gathering power of, 132
eleven-year cycle of, 389, 408, 409, 410, sunspots, 389, 403, 407–12, 407, 409–11 limitations of, 132, 133
410, 413 butterfly diagram for, 408, 409 neutrino, 398–400, 401, 479
energy conversion in, 389–95, 393, 394, historical record of, 408, 410, 410 optical, 125–41, 143–44
398, 413, 438, 443–48, 595 superclusters, 4, 608–9, 610, 611 optical detectors for, 137–41
energy transport in, 389, 393–96, 395, Super-Kamiokande detector, 400, 401, 613 orbital, 125, 135, 137, 140, 144, 145 (table),
396, 398, 413 superluminal motion, 556 151
as focus, 16, 62–63, 62, 63 supermassive black hole, 552 and origin of Solar System, 157, 157
formation of, 5, 58, 155–58, 158, 162, 168, supernova 1987A, 479, 479 planets discovered by, 249
175, 176, 327, 429, 438, 445, 448 supernovae, 426, 443, 486, 505, 544, 571, radio, 137, 141–44, 142
future evolution of, see low-mass stars, 578, 640 reflecting, 125, 126, 127, 129, 133–34, 136
evolution of binary stars and, 443, 463, 464 refracting, 125, 126–27, 126, 129, 129, 132,
giant planets and, 249, 252, 265 in “galactic fountain” model, 574, 575 134
I-22 Index
telescopes (continued) tectonism on, 179, 180–81, 186, 196–202, tidal locking, 275, 284–85, 284
resolution of, 125, 133, 134–35 196, 197, 198, 199, 200, 201, 211, 302 of binary stars, 288
X-ray, 421 volcanism on, 179, 181, 181, 186, 187, 193, of moons, 301, 323, 461
see also specific telescopes 195, 202–6, 204, 205, 211 tidal stresses, 279–89, 292, 303–8, 323–24
Tempel 1, Comet, 345, 345 see also Earth; Mars; Mercury; Venus tide pools, 634
Tempel-Tuttle, Comet, 354 (table) Tertiary period, 184 tides, 193
temperature, 91, 107, 111–20 Tethys, 293, 293, 300, 310 black holes and, 288
absolute zero, 113 tetravalent atoms, 640 destructiveness of, 275, 287, 288, 288,
atmosphere and, 135, 135, 216, 218, 225, Thatcher, Comet, 354 (table) 289–90, 289, 319, 323–24, 332
226, 227, 229–30, 234, 235–36, 237–38, Thebe, 319 and differences in gravity from external
238, 239, 241, 257, 259, 263 theoretical models, 9 objects, 275, 279–80, 279
atmospheric greenhouse effects and, theoretical science, defi ned, 65 heating and, 193, 286–87, 297
219–21 theory, see scientific theory high vs. low, 281–82, 282
of cosmic background radiation, 528, 528 Theory of Everything (TOE), 296, 600–601, interactions between galaxies and, 288,
defi ned, 112, 121 602, 624, 625 288, 557–58, 557, 617
determination of, 112–13, 117–20, 118, theory of evolution, 307 lunar, 279–84, 283
419, 420 thermal conduction, 393 noncircular orbits and, 285–87, 286
of gas, 222–23 thermal energy, 91, 112–13, 192, 216, 304–10 object’s rotation tied to its orbit by, 275,
of interstellar medium, 419–24, 420, 422, in accretion disk, 166 284–87, 284
423 convection and, 197–98, 201, 202, 219–20, oceans flow in response to, 279–84, 281,
Kelvin scale of, 113 227, 227, 261 282
luminosity and, 113–16 in Earth’s interior, 192, 201, 202–3, 283 solar, 282, 283, 287, 289
as measure of how energetically particles evolution of binary stars and, 462, 464 spring vs. neap, 282, 283
are in movement, 112–13, 121 gravitational energy converted to, 429–33, Tiktaalic, 636
of Planck spectrum, 525, 526, 528, 528 431, 553–54 time, 5, 19
of planets, see specific planets greenhouse effect and, 219–20 atomic decay and, 107
of protostars, 429, 430–35, 432, 434 hurricanes and, 236 gravity and, 58
Stefan’s law and, 114, 115–16 instability strip and, 473 Hubble, 518–19, 520, 591
thermal equilibrium and, 112 from interior of giant planets, 247, 261, look-back, 509, 518, 524
Wien’s law and, 115, 116, 367–68, 370, 372, 263, 272 Newtonian physics and, 490–91
378 (table), 419 magma and, 202–3 relativity theory and, 12, 77, 490–93
see also cold, cooling; heat, heating of protostars, 428–33, 431, 432, 439, 553 see also spacetime
temperature, stellar, 265, 367–73, 445 in self-sustaining reactions, 476 Time, 12, 513
and classification of stars, 368–70 of Sun, 389, 391–96, 391, 393, 394, 395, time dilation, 492, 493
color and, 368, 368, 378 (table) 396, 398, 413 Titan, 7, 217, 270, 300, 308, 308, 309, 310, 641
of high-mass stars, 471, 472, 472 (table) thunderstorms and, 233, 234, 235 Titania, 300, 312
of horizontal branch stars, 455 transport of, 393–96, 395, 396–97, 396, TOE (Theory of Everything), 296, 600–601,
H-R diagram and, 378–80, 379, 381, 382– 413, 453 602, 624, 625
83, 382 (table), 383 thermal equilibrium, 112 toilet bowls, Coriolis effect and, 233
measurement of, 361, 367–68, 372, 378 thermal motions, 112–13, 113, 452–54, 601 Tolkien, J. R. R., 468
(table), 383 thermal noise, 139 Tolstoy, Leo, 326
of post-AGB stars, 457 thermal radiation, 113–16, 367–68, 585 Tombaugh, Clyde, 328
of red giants, 449, 452–53, 454, 454 defi ned, 113 tools, astronomical, 124–53, 170
size of stars and, 371–72 equilibrium temperature of planets and, topographic relief, 181
of Sun, 115, 389, 390, 391, 393, 395–98, 117–20, 118, 119 tornadoes, 236, 237
395, 402–4, 403, 413, 433, 444, 445, Planck spectrum and, 114–15, 114, 367– total eclipses:
448–49 69, 368, 369, 372 lunar, 50
temperature gradient, 395–96 temperature and size of stars and, 361, solar, 46–50, 47
Temple-Tuttle Comet, 354–55 367–69, 372, 385, 479 toys, “glow in the dark”, atomic decay and,
Terrestrial Planet Finder (TPF), 172 thermodynamics, second law of, 639, 639 107
terrestrial planets, 168, 180 (table), 194, 264 thermonuclear reactions, 263, 453–54, 463, traffic circle without exits, 161–62, 162
atmosphere of, 168, 215–45, 216, 217, 218 464, 639 transform fault, 198
(table), 219, 221 (table), 224, 252 thermosphere, 226, 227 transit method, of planet detection, 170–71,
erosion on, 179, 181–82, 186, 206–10, 207, third quarter Moon, 45, 275, 282, 283 171
208, 209 30 Doradus, 423, 423 trans-Neptunian objects (TNOs), 328
extrasolar, 170, 171, 172, 174, 643 thought experiments: transverse waves, 189
giant planets compared with, 247, 250– “ball in the car,” 231, 231, 488 triple-alpha process, 453, 453, 456, 476, 481,
54, 257, 260, 261, 262–63, 264, 266, of Einstein, 490–91 530
271–72 equivalence principle and, 495, 496 Triton, 249, 300, 302, 304–6, 305, 306
impacts on, 168, 169, 179, 180–86, 182, of Newton, 79–82, 79, 85, 558 cantaloupe terrain of, 305, 305
183, 185, 186, 206, 211, 336, 349–52 3C273, 551 Trojan asteroids, 293, 293, 336
interiors of, 188–96, 192, 264 thrust faults, 200 tropical year, 42
key concepts about, 179, 215 thunderstorms, 233, 234, 234, 236 Tropic of Cancer, 41
magnetic fields of, see specific planets tidal bulge, 279–87, 281, 284, 285 Tropic of Capricorn, 41
Index I-23
tropopause, 226, 227, 229, 234, 236 escape velocity of, 530 momentum and, 101
troposphere, 225, 226, 227, 229, 259 expansion of, see Big Bang theory; peculiar, 517, 622, 623
T Tauri star, 437 expanding universe radial, 109
Tunguska River, 349–52, 353 homogeneity of, 512–15, 512, 516 thermal, 164
tuning fork diagram, 537, 538, 569 inflation model of, 585, 593–95, 594, 601, Venera 9 Soviet lander, 235
turbulence, 234 602, 611, 623–25 vents, hydrothermal, 307
Tuttle, Comet, 354 (table) isotropy of, 512, 513, 514, 528 Venus, 7, 36, 60–61, 126, 146, 184, 192, 192,
Twain, Mark, 348 local, 4 249, 323, 365, 404
Type I supernovae, 459, 463, 464, 465, 479, mass of, 85, 501, 585, 586, 587, 592, 611, atmosphere of, 168, 180, 184, 207, 215, 216,
482, 552 626–27 217–20, 218 (table), 220, 226, 229, 235,
nucleosynthesis and, 480 measuring distances in, 518, 565 238–39, 241, 244, 351
as standard candles, 517–18, 518, 587, multiple, 607, 623–25, 624, 625 atmospheric greenhouse effect on, 219,
588 perspective on, 566, 567 220–21, 238, 244, 263
Type II supernovae, 464n, 478, 479–84, 505 remoteness and difference of, 5, 6 craters of, 186, 351
energetic and chemical legacy of, 479–84, scientific revolutions and, 12 Earth compared with, 200–202, 220–21,
480, 481, 481, 482, 483, 484 size of, 5, 6, 155, 365, 510–11, 530 238
nucleosynthesis and, 480 structure of, see structure of universe erosion on, 207, 208
typhoons, 233 Sun-centered, 57, 67, 85, 175 formation of, 168
uniformity of, 601, 607 gravity of, 200, 219
UFOs, 340n, 351 University of California, 646 lack of magnetic field on, 195
UGC 12343 (spiral galaxy), 544 unstable equilibrium, 110 lack of moon for, 299
Uhuru X-ray satellite, 504 uranium, 480 lack of rings for, 323
ultraviolet (UV) radiation, 98, 136, 137, 141, half-life, 187 mass of, 200, 218, 219, 251
144, 217, 241, 257, 270, 308, 411, 411, isotopes of, 187 orbit of, 218, 220, 254, 348
553–54 Uranus, 247–55, 249, 265–68, 323, 341 rotation of, 239
interstellar medium and, 419, 420, 422, atmosphere of, 235, 252, 254, 255, 256, tectonics on, 200–202, 201, 211
423 259, 317 temperature of, 119, 166, 219, 220–21, 226,
ozone and, 225 chemical composition of, 252, 259, 266, 238, 239, 641
photons of, 99, 259 272 volcanism on, 179, 201, 206, 218–19, 220,
of post-AGB stars, 457 density of, 248 (table), 252, 265 239
spiral galaxy viewed in, 542–43, 544 discovery of, 249, 640 water on, 220–21
star formation and, 544 formation of, 168, 169, 252, 266 wind on, 207, 208, 239
umbra, 50, 407, 408 interior of, 264, 265–66, 272 vernal equinox, 39, 42, 362
defi ned, 46 magnetic field of, 266, 267–68, 267 Very Large Array (VLA) radio telescope, 137,
total solar eclipse and, 46, 46, 48 magnetosphere of, 266–67, 268 142, 143
Umbriel, 312, 313 mass of, 248 (table), 251, 265 Very Large Telescope (VLT), 128 (table),
unbalanced force, 67–70 moons of, 300, 312–13, 312, 320, 321 143–44, 143
motion is changed by (Newton’s second obliquity of, 248 (table), 253–54 Very Long Baseline Array (VLBA) radio
law) (a = F/m), 69, 70, 73, 74, 78, 86, 495 orbit of, 248 (table), 167, 249, 253–54, 313, telescope, 142–43
unbound orbits, 83, 85 328–29, 345 Vesta, 331, 335, 338
uncertainty principle, 101, 593–95 rings of, 256, 313, 315, 317, 318, 319, 320, Vesuvius, Mount, 202, 203
understanding, 17–18 321, 322 vidicons, 139
patterns and, 3, 14–15 rotation of, 248 (table), 252–53 Viking Mars missions, 201, 205, 208, 241,
scientific, 10, 12, 17–18 size of, 247–49, 250, 250 641
unified model of AGNs, 553–58, 553 temperature of, 166, 260 Viking orbiters, 237
uniform circular motion, 81 winds on, 261 Virgo, 39
universal gravitational constant (G), 71, 76, Urey, Harold, 633 cluster, 608, 609
78 Ursa Major, see Big Dipper supercluster, 4, 4, 608
universe, 3–7, 19, 148, 174, 175 Ursids, 354 (table) virtual particles, 503, 503
age of, 590–91, 591 visual binary, 376
birth of, 7, 155 vacuum, 587, 589 visualization, 17
chemical evolution of, 5, 7, 7, 371, 459, Valhalla, 312–13 volatile materials, 155, 166, 166, 167, 168,
529–30, 530, 572–73, 585, 602 Valles Marineris, 200, 201 169, 192, 304
clockwork, 107, 120 variable stars, pulsating, see pulsating atmosphere formation and, 215, 217, 244
collapse of, 590; see also “Big Crunch” variable stars in comets, 342, 349
cosmological principle and, 9–10, 85, vector notation, 76n in giant planets, 247, 257, 258, 259, 262
512–13 Vega, 368 volcanic domes, 202
defi ned, 510 VEGA spacecraft, 343, 344 volcanism, 218, 224
dynamic vs. static, 588–89 V838 Monocerotis, 442 atmosphere and, 155, 167, 215–19, 244,
early, 620 velocity: 633
early, AGN activity in, 556, 558 change in, see acceleration defi ned, 181
Earth-centered, 25, 57, 567 circular, 83 on moons, 274, 286–87, 287, 299, 300, 302–
effect of new tools on view of, 7–8, 8 defi ned, 69 7, 303, 304, 310–13, 319–20, 323
eras in evolution of, 601, 602, 603, 626 escape, see escape velocity plate tectonics and, 198, 199, 202–3, 307
I-24 Index
volcanism (continued) neutrino, 147 Wild 2, Comet, 345, 345

in shaping terrestrial planets, 179, 181, wavefronts, 132, 135, 136 Wilkinson Microwave Anisotropy Probe
186, 187, 193, 195, 202–6, 204, 205, 211, wavelength: (WMAP), 145 (table), 529, 529, 611,
633 defi ned, 96, 96 622
tide-powered, 275, 286–87, 287 of radio waves, 141–44 Wilson, Robert, 525–26, 526, 528
vortices, of giant planets, 254–55, 255, 260, relationship between period and, 96 WIMPs (weakly interacting massive
261–62 of sound, 108 particles), 571n
Voyager missions, 145, 146, 643 wavelengths of light, 131, 141, 141, 367–68 wind, 229–32, 234–37, 234, 475
to giant planets, 249, 250, 254–55, 255, 257, Doppler shift of, 108, 109 Coriolis effect and, 230–32, 230, 233, 247,
260, 262, 266, 268, 306, 310, 311, 312, energy levels of atoms in determination of 259–62, 261, 262
313, 314, 316, 318, 319, 319, 320, 324 emission and absorption of, 102–5 erosion by, 181, 206–7, 208
Voyager 1, 262, 268, 313, 314, 320–22, 320, energy of photon and, 99, 103 on giant planets, 247, 259–63, 261, 262,
406, 407 Planck spectrum and, 114, 115, 116, 271–72
Voyager 2, 255, 255, 256, 260, 266, 306, 367–68 measurement of speed of, on distant
313, 316, 317, 319, 320 refraction and, 133 planets, 260
rest, 108–9 solar, see solar wind
Waller, Edmund, 360 scattering and, 239, 240, 240, 259 zonal, 230, 259, 260, 262, 262
War of the Worlds (Wells), 640 spectral fi ngerprints and, 106–7, 106 see also specific planets
watches, “glow in the dark”, atomic decay spectrum and, 97, 99 wind streaks, 207, 208
and, 107 waves, radio, 125, 137, 141–42 winter, 36–41, 37, 38, 229
water, 166, 166, 167, 303, 304, 352, 357, 642 weak nuclear force, 596, 598, 601, 602 festivals in, 39, 40
in atmosphere, 217, 218, 218 (table), 220– weather, 225, 227 winter solstice, 39, 40
21, 227, 233, 234, 234, 243 convection and, 234–37
in comets, 346, 349, 357 Earth’s rotation and, 23, 233 X-ray binaries, 469, 482, 482
on dwarf planets, 330, 331 on giant planets, 259–63, 261, 271–72 black holes and, 504, 552
erosion by, 181, 207, 208, 209–10, 209, interstellar medium compared with, 426 X-rays, 99, 136, 137, 144, 269, 462, 553, 578,
221 wind and, 229–32, 234–37, 234 579
on giant planets, 247, 252, 257, 259, 264– weathering, 206 of clusters of galaxies, 616, 616
66, 271 Wegener, Alfred, 196 Crab pulsar observed in, 484, 484
on Mars, 185, 186, 207, 208, 209–10, 209, weight: defi ned, 98
210, 221, 239, 241–42, 242, 243 defi ned, 74–75 emitted by gas in elliptical galaxies, 540,
on moons, 306–7, 306, 310, 312–13 pressure and, 191, 193, 202, 221, 390, 397– 548
thunderstorms and, 233, 234, 234 98, 469, 471 as evidence of black holes, 504, 504
see also oceans; oceans of Earth weightlessness, 277, 277 of interstellar medium, 417, 421
waterfalls, 7 Welles, Orson, 640 of Sun, 405, 406
waterspouts, 236 Wells, H. G., 640 Sun as source of, 405, 411
water vapor, 144, 633 West, Comet, 287, 289, 346 supernovae remnants and, 480
wave, waves: “What if?” questions, 18
amplitude of, 96, 96 Whipple, Fred, 342 years, 32, 42
defi ned, 93 white dwarfs, 373, 443, 457, 459–60, 472, Yellowstone National Park, 203, 307
electromagnetic, see electromagnetic 549, 570, 627 Yerkes 1–m telescope, 127, 127
waves accretion disks around, 462, 552 Yucatan Peninsula, 184–85, 185, 357
frequency of, 96, 99, 103, 108 in binary system, 459, 462–64, 463, 482,
gravity, 501 482, 518 Zenith, 26, 40
longitudinal, 189 degenerate core of, 443, 459, 487, 505, 518 zinc, 640
period of, 96, 189 gravitational redshift of, 499 zodiac, 33, 34, 52
sound, 108 neutron star compared with, 487 zodiacal dust, 355–56, 356
transverse, 189 Whitehead, Alfred North, 416 zodiacal light, 355–56
wave detectors: Wien’s law, 116, 151, 367–68, 370, 372, 378 zonal winds, 230
gravitational, 147 (table), 419, 457, 525, 527, 597 on giant planets, 259, 260, 262, 262

21st Century Astronomy PDF

Uploaded by

Copyright:

Available Formats

21st Century Astronomy PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

21st Century Astronomy PDF

Uploaded by

Copyright:

Available Formats

This page intentionally left blank

21st Century Astronomy’s text, art, and

Both text and

AstroTour animations help

StudySpace is a free and

Brad Smith • U N I V E R S I T Y O F H AWA I I — M A N O A

Laura Kay • B A R N A R D COLLEGE

Howard Voss • A R I Z O N A S TAT E U N I V E R S I T Y

All rights reserved

Editor: Rob Bellinger

Library of Congress Cataloging-in-Publication Data

ISBN: 978-0-393-11664-9 (pdf.)

Laura Kay thanks her partner, M.P.M.

Brad Smith dedicates this Third Edition to his patient and

PART I Introduction to Astronomy

PART II The Solar System

PART III Stars and Stellar Evolution

PART IV Galaxies, the Universe, and Cosmology

EPILOGUE The Universe and You

About the Authors xxxv

PART I Introduction to Astronomy

CHAPTER 1 Why Learn Astronomy?

1.2 Science Involves Exploration and Discovery 7

1.3 Science Is a Way of Viewing the World 9

1.4 Patterns Make Our Lives and Science Possible 14

Math Tools 1.1 Mathematical Tools 16

1.5 Thinking like an Astronomer 17

Seeing the Forest for the Trees 19

CHAPTER 2 Patterns in the Sky—Motions of Earth

2.2 Earth Spins on Its Axis 25

Math Tools 2.1 How to Estimate the Size of Earth 33

Tiny Deviations in the Direction of Starlight Provide a Measure

2.4 The Motions and Phases of the Moon 42

2.5 Eclipses: Passing through a Shadow 46

Seeing the Forest for the Trees 52

CHAPTER 3 Gravity and Orbits—A Celestial Ballet

3.2 An Empirical Beginning: Kepler Describes the Observed Motions

Excursions 3.1 The Motions of the Planets in the Sky 60

Kepler’s First Law: Planets Move on Elliptical Orbits

3.3 The Rise of Scientiﬁc Theory: Newton’s Laws Govern

Foundations 3.1 Proportionality 71

3.4 Gravity Is a Force between Any Two Objects

Connections 3.1 Inverse Square Laws 76

Putting the Pieces Together: A Universal Law for Gravitation 76

3.5 Orbits Are One Body “Falling around” Another 79

Planets Are Just like Newton’s Cannonball 81

Math Tools 3.2 Achieving Circular Velocity 82

Most Orbits in the Real World Are Not Circles 82

Math Tools 3.3 Calculating Escape Velocities 84

Newton’s Theory Is a Powerful Tool for Measuring Mass 85

Seeing the Forest for the Trees 86

4.2 How Our Understanding of Light Evolved 92

Math Tools 4.1 Working with Electromagnetic Radiation 97

A Wide Range of Wavelengths Makes Up the Electromagnetic

4.3 The Quantum View of Matter 100

4.4 The Doppler Effect—Is It Moving toward Us or Away from

Math Tools 4.2 Making Use of the Doppler Effect 109

Foundations 4.1 Equilibrium Means Balance 110

4.5 Why Mercury Is Hot and Pluto Is Not 111