.

tics I

An integrated Appro ch

Plant Systematics An Integrated Approach Third edition Gurcharan Singh University of Delhi Delhi, INDIA Science Publishers Enfield(NH) Jersey Plymouth CIP data will be provided on request Science Publishers 234 May Street www.scipub.net Post Office Box 699 Enfield, New Hampshire 03748 United States of America General enquiries [email protected] Editorial enquiries [email protected] Sales enquiries [email protected] Published by Science Publishers, Enfield, NH, USA An imprint of Edenbridge Ltd., British Channel Islands Printed in India 2010, copyright reserved ISBN 978-1-57808-668-9 The author and the publisher make no warranty of any kind, expressed or implied, with regard to programs contained in this companion CD. The authors and publish er shall not be liable in any event for incidental or consequential damages in c onnection with, or arising out of, the furnishing, performance, or use of these programs. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechan ical, photocopying or otherwise, without the prior permission of the publishers, in writing. The exception to this is when a reasonable part of the text is quot ed for purpose of book review, abstracting etc. This book is sold subject to the condition that it shall not, by way of trade or otherwise be lent, re-sold, hired out, or otherwise circulated without the publ isher s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being impo sed on the subsequent purchaser. Preface This third edition of integrated information on Plant Systematics has largely be en influenced by the developments of the first few years of twenty first century . Past two decades have seen development of new tools of biotechnology, vigorous utilization of molecular data in understanding phylogeny, and redefining affini

ties and arrangements of plant groups. Recent years have also seen disappearance of gaps between numerical and cladistic methodologies, and integration of forme r into the latter for complete understanding of phylogenetic relationships. Thes e trends have largely influenced the combination of numerical and cladistic meth ods under one chapter, and enlarged discussion on Molecular Systematics, discuss ing new concepts, tools and recent achievements. New chapters on Pteridophytes a nd Gymnosperms have been added for complete understanding of systematics of vasc ular plants. It is being increasingly realized that actual photographs of plants and plant pa rts enable better understanding of taxonomic information, the trend usefully exp loited by recent publications by Simpson (2006) and Judd et al. (3rd ed., 2008). The present edition incorporates more than 500 colour photographs of plants fro m diverse families of plants. High-resolution images of these as also the additi onal plants have been provided in the CD-ROM being supplied along with the book, latter including 772 photographs. This has largely been possible through the ki nd courtesy of my son Manpreet Singh and daughter-in-law Komal, who sponsored my recent visit to California, and provided me the opportunity to visit and photog raph temperate plants in and around California. The book as such contains images of both tropical plants (largely from Delhi), temperate American plants and pla nts from other parts of the World growing in the Botanical Gardens of University of California and San Francisco Botanical Garden. I wish to record the help ren dered by the members of TAXACOM in the identification of some of the American plants. The focus of the present edition has been to further consolidate the information on the principles of plant systematics, include detailed discussion on all majo r systems of classification, and significantly, also include discussion on the s elected families of vascular plants, without sacrificing the discussion on basic principles. The families included for discussion are largely those which have w ide representation, as also those that are less iv Plant Systematics known but significant in evaluating the phylogeny of angiosperms. The discussion of the families also has a considerable focus on their phylogenetic relationshi ps, as evidenced by recent cladistic studies, with liberal citation of molecular data. Several additional families have been included for detailed discussion in the present volume. Recent internet revolution has greatly helped in propagating taxonomic informati on, with numerous searchable databases, online programs for identification and d ata analysis available for ready reference. The information concerning these has been included at appropriate places in various chapters for easy utilization. I n light of this, the separate chapter on web has been omitted. The outputs of co mputer programs, especially used in molecular studies and construction of phylog enetic trees has been included based on actual or hypothetical data. This will a cquaint readers with the handling of raw data and working of computer programs. The author has attempted to strike a balance between classical fundamental infor mation and the recent developments in plant systematics. Special attention has b een devoted to the information on botanical nomenclature, identification and phy logeny of angiosperms with numerous relevant examples and detailed explanation o f the important nomenclatural problems. An attempt has been made to present a co ntinuity between orthodox and contemporary identification methods by working on a common example. The information on methods of identification using computers h as been further enhanced to help better online identification. For providing me inspiration for this book, I am indebted to my undergraduate st udents, who helped me to improve the material through frequent interactions. I a m also indebted to my wife Mrs. K.G. Singh for constant support and bearing with my overindulgence with this book. I also wish to acknowledge the help rendered by my son Kanwarpreet Singh at

various stages. I wish to record thanks to all the colleagues whose inputs have helped me to imp rove the information presented here. I also wish to place on record sincere than ks to Dr. Jef Veldkamp for valuable information on nomenclature, Dr. Gertrud Dah lgren for photographs and literature, Dr. P.F. Stevens for literature on APG II and trees from his APweb, Dr. Robert Thorne for making available his 2007 classi fication, Dr. James Reveal for his help on nomenclatural problems, Dr. D.L. Dilc her for his photograph, Dr. Julie Barcelona and Harry Wiriadinata for photograph s of Rafflesia, the authorities of New York Botanical Garden, Missouri Botanical Garden, USA, Royal Botanic Gardens Kew and University of California, Santa Cruz , for photographs used in the book. New Delhi Gurcharan Singh November 2009 Contents Preface iii 1. 1 14 PLANTS, TAXONOMY AND SYSTEMATICS Plants and Kingdoms of Life 1 Two Kingdom System 1 Two Empires Three Kingdoms 2 Five Kingdom System Six or Seven Kingdoms? The Plant Kingdom 2 2 6 Taxonomy and Systematics 7 Basic Components (Principles) of Systematics 8 Aims of Systematics 11 Advancement Levels in Systematics 12 2. BOTANICAL NOMENCLATURE 15 45 Need for Scientific names 15 Why Latin? 16 Development of Botanical Code 16 Contents of Botanical Code 17 Preamble 17 Principles of ICBN 18 Names of Taxa 18 The Type Method 23 Author Citation 25

Publication of Names 26 Rejection of Names 28 Principle of Priority 30 Names of Hybrids 34 Names of Cultivated Plants 35 vi Plant Systematics Unified Biological Nomenclature Draft BioCode PhyloCode 38 35 36 3. HIERARCHICAL CLASSIFICATION 46 55 Taxonomic groups, categories and ranks 46 Utilization of categories 48 Species concept 49 Infraspecific ranks 53 Genus 54 Family 54 4. DESCRIPTIVE TERMINOLOGY 56 90 Habit and life span 56 Habitat 57 Roots 57 Stems 58 Leaves 61 Leaf arrangement 62 Leaf duration 63 Leaf incision 63 Stipules 65 Leaf shape 65 Leaf margin 66 Leaf base 66 Leaf apex 67

Leaf surface 68 Venation 69 Inflorescence 69 Racemose types Cymose types 69 70 Specialized types 71 Flower 71 Calyx 73 Corolla 74 Perianth 74 Androecium 74 Gynoecium 77 Fruit 79 Simple fruits 80 Aggregate fruits 82 Multiple fruits 83 Floral formula 83 Floral diagram 83 5. PROCESS OF IDENTIFICATION 91 127 Specimen preparation 91 Fieldwork 91 Equipment 92 Collection 93 Pressing 93 Contents vii Handling special groups 94 Drying 94 Herbarium methods 95 Botanical gardens 95 Herbaria 101 Pest Control 105

Virtual herbarium 106 Identification methods Taxonomic literature Taxonomic keys 108 108 113 Cmputers in identification 120 Interactive keys Id. 121 6.VARIATION, BIOSYSTEMATICS, POPULATION GENETICS AND EVOLUTION 128 148 Types of variation 128 Developmental variation Environmental variation Genetic variation Variance analy sis 129 129 129 129 Reproductive systems 131 Outbreeding 131 Inbreeding 135 Apomixis 135 Population genetics 135 Allele frequencies 136 Mating systems 136 Hardy-Weinberg law 136 Evolution 139 Mutation 140 Migration 140 Random genetic drift 140 Natural selection 141 Molecular evolution 143 Neutral theory of evolution 143 Speciation 144 7. TAXONOMIC EVIDENCE 149 209 Morphology 149 Habit 149 Underground parts

150 Leaves 150 Flowers 150 Fruits 150 Anatomy 150 Wood anatomy 150 Trichomes 151 Epidermal features 153 Leaf anatomy 153 Floral anatomy 153 viii Plant Systematics Embryology 154 Families marked out by distinct embryological features 154 Specific examples of the role of embryological data 155 Palynology 156 Pollen aggregation 156 Pollen wall 157 Pollen aperture 157 Micromorphology and Ultrastructure 159 Micromorphology 159 Ultrastructure 161 Chromosomes 164 Chromosomal number Chromosomal structure 164 167 Chromosomal behaviour 168 Chemotaxonomy 168 Primary metabolites 169 Secondary metabolites 169 Non-semantide Macromolecules 178 Proteins 178 Molecular systematics 184

Molecular evolution Location of molecular data Molecular techniques DNA polymorp hism 184 186 187 199 Examples of molecular studies 204 Gene trees 209 8. DEVELOPING CLASSIFICATIONS 210 264 Phenetic methods 210 Principles of taxometrics 211 Cladistic methods 212 Phylogenetic terms 213 Phylogenetic diagrams 221 Phylogeny and classification 225 Phylogenetic data analysis 229 Taxa-Operational Units 229 Characters Measure of similarity Construction of trees The Consensus tree Automated trees 2 58 Gene trees and species trees Developing classification 263 229 234 237 251 262 9. PHYLOGENY OF ANGIOSPERMS 265 296 Origin of Angiosperms 265 What are Angiosperms? 265 What is the age of Angiosperms? 266 What is the place of their origin? 268 Are angiosperms monophyletic or polyphyletic? 270 Contents ix What are the possible ancestors? 270 Origin of monocotyledons 280 Basal living angiosperms 280 Casuarinaceae 281 Magnoliids 281 Paleoherbs 282 Evolutionary trends 285

Coevolution with animals Basic evolutionary trends Xylem evolution 287 Stamen ev olution 289 Pollen grain evolution 285 286 292 Carpel evolution 292 Evolution of inferior ovary 296 10. SYSTEMS OF CLASSIFICATION 297 358 Classifications based on gross morphology 297 Preliterate mankind 297 Early literate civilisations 297 Medieval Botany Renaissance 299 300 Sexual System 302 Carolus Linnaeus 303 Natural Systems 306 Michel Adanson 306 Jean B. P. Lamarck 306 de Jussieu family 306 de Candolle family 307 Robert Brown 308 George Bentham and Sir J. D. Hooker 308 Phylogenetic Systems Transitional Systems 312 312 Intentional phylogenetic systems Modern phylogenetic systems 316 324 11. FAMILIES OF PTERIDOPHYTES 359 383 Lycopodiophtes Lycopodiaceae 362 Selaginellaceae 363 Isoetaceae 365 Psilopsida Ophioglossaceae 366 Psilotaceae 368 Equisetopsida Equisetaceae 370 Pteropsida Osmundaceae

371 Marsileaceae 373 Salviniaceae 374 Cyatheaceae 376 Pteridaceae 377 x Plant Systematics Aspleniaceae 379 Dryopteridaceae 380 Polypodiaceae 382 12. FAMILIES OF GYMNOSPERMS 384 406 Cycadales Cycadaceae 386 Zamiaceae 387 Ginkgoales Ginkgoaceae 389 Coniferales Pinaceae 391 Cupressaceae 393 Podocarpaceae 395 Araucariaceae 396 Taxaceae 398 Gnetales Ephedraceae 399 Gnetaceae 401 13. MAJOR FAMILIES OF ANGIOSPERMS 407 678 Angiosperms roll of honour 408 Chloranthidae 409 Amborellaceae 409 Chloranthaceae 411 Austrobaileyaceae 413 Winteraceae Illiciaceae 415 Cabombaceae 417

Nymphaeaceae 419 Ceratophyllaceae 421 Magnolii dae 423 Magnoliaceae 423 Degeneriaceae 425 Annonaceae 427 Calycanthaceae 429 Lauraceae 431 Winteraceae 433 Saururaceae 435 Piperaceae 437 Alismatidae 439 Acoraceae 439 Araceae 441 Butomaceae 443 Alismataceae 445 Hydrocharitaceae 447 Potamogetonaceae 449 Liliidae 451 Pandanaceae 451 Dioscoreaceae 453 Smilacaceae 455 Contents xi Liliaceae 473 Orchidaceae 475 Iridaceae 478 Asphodelaceae 480 Alliaceae 482 Subfamily 484 Agavaceae 485 Commelinidae 487

Arecaceae 488 Commelinaceae 490 Musaceae 492 Zingiberaceae 494 Cannaceae 496 Juncaceae 498 Cyperaceae 500 Poaceae 502 Ranunculidae 505 Paeoniaceae 505 Berberidaceae 507 Ranunculaceae 509 Papaveraceae 512 Hamamelididae 514 Saxifragaceae 514 Fagaceae 517 Betulaceae 519 Casuarinaceae 521 Caryophyllidae 523 Portulacaceae 524 Cactaceae 526 Nyctaginaceae 528 Aizoaceae 530 Chenopodiaceae 532 Amaranthaceae 534 Caryophyllaceae 536 Polygonaceae 538 Droseraceae 540 Rosidae 542 Celastraceae 543 Violaceae

545 Salicaceae 547 Cucurbitaceae 550 Clusiaceae 552 Euphorbiaceae 554 Oxalidaceae 557 Zygophyllaceae 559 Geraniaceae 561 Rosaceae 563 Fabaceae 566 Myrtaceae 572 xii Plant Systematics Lythraceae 574 Onagraceae 577 Malvidae 579 Malvaceae 580 Grewiaceae 583 Dipterocarpaceae 584 Rhamnaceae 586 Ulmaceae 588 Moraceae 590 Urticaceae 592 Rafflesiaceae 595 Capparaceae 597 Cleomaceae 599 Brassicaceae 600 Rutaceae 603 Meliaceae 605 Anacardiaceae 607 Sapindaceae 610 Asteridae

612 Hydrangeaceae 613 Cornaceae 627 Balsaminaceae 629 Polemoniaceae 631 Ebenaceae 633 Sapotaceae 635 Primulaceae 637 Ericaceae 639 Adoxaceae 642 Apiaceae 644 Araliaceae 646 Asteraceae 649 Lamiidae 652 Solanaceae 652 Convolvulaceae 655 Boraginaceae 657 Rubiaceae 659 Apocynaceae 661 Plantaginaceae 664 Lamiaceae 666 Verbenaceae 669 Bignoniaceae 671 Acanthaceae 673 Scrophulariaceae 675 REFERENCES INDEX 703 742 669 702 Contents xiii Color Plate Section The page numbers referred below are those of the text pages where the B/W images of the figures appear. Stems 85 Leaves 86 Inflorescences 87

Fruits 88 Pteridophytes 403 Selaginellaceae, Osmundaceae, Blechnaceae 403 Gymnosperms Cycadaceae, Zamiaceae 404 Ginkgoaceae, Pinaceae, Cupressaceae 405 Angiosperms 457 Chloranthidae 457 Magnoliidae 458 Araceae, Alismataceae, Hydrocharitaceae, Liliaceae 459 Iridaceae, Asphodelaceae, Alliaceae 460 Hyacinthaceae, Agavaceae, Asparagaceae, Nolinaceae 461 Arecaceae, Musaceae, Commelinaceae, Cyperceae, Poaceae 462 Paeoniaceae, Berberidaceae, Papaveraceae 463 Ranunculaceae Grossulariaceae, Fagaceae, Nothofagaceae, Betulaceae Portulacaceae, Cactaceae, N yctaginaceae, Aizoaceae 464 465 466 Chenpodiaceae, Amaranthaceae, Caryophyllaceae, Polygonaceae 467 Celastraceae, Violaceae, Cucurbitaceae, Begoniaceae 468 Clusiaceae, Euphorbiaceae, Oxalidaceae 469 Geraniaceae, Rosaceae 470 Fabaceae 471 Myrtaceae, Lythraceae, Onagraceae 472 Malvaceae, Rhamnaceae, Moraceae 615 Rafflesiaceae, Brassicaceae 616 Rutaceae, Anacardiaceae, Meliaceae 617 Sapindaceae 618 Hydrangeaceae, Polemoniaceae, Cornaceae, Primulaceae Ericaceae, Adoxaceae Apiaceae, Araliaceae 619 620 621 Asteraceae 622

Solanaceae, Convolvulaceae, Boraginaceae 623 Rubiaceae, Apocynaceae Plantaginaceae, Lamiaceae 624 625 Verbenaceae, Bignoniaceae, Acanthaceae, Scrophulariaceae 626 Chapter 1 Plants, Taxonomy and Systematics Taxonomy (or systematics) is basically con-cerned with the classification of org anisms. Living organisms are placed in groups on the basis of similarities and d ifferences at the organismic, cellular, and molecular levels. The United Nations Environment Programme s Global Biodiversity Assessment estimates the number of de scribed species of living organisms as approximately 1.75 million. The list grow s longer every year. Clas-sifying these organisms has been a major challenge, an d the last few decades have seen a lot of realignments as additional ultrastructural and molecular information piles up. These realignments have primarily been the result of realization that the branches of the phylogenetic tree must be ba sed on the con-cept of monophyly, and each taxonomic group, kingdoms included, s hould be monoa single largest, most inclusive group. Classifying organisms and diverse forms of life is challenging task before the biologists. PLANTS AND KINGDOMS OF LIFE phyletic. Plants are man s prime companions in this universe, being the source of food and e n-ergy, shelter and clothing, drugs and beverages, oxygen and aesthetic environm ent, and as such they have been the dominant component of his taxonomic activity through the ages. Before attempting to explore the diversity of plant life it i s essential to understand as to what is our understanding of the term Plant, and the position of plants in the web of life. Traditionally the plants are delimit ed as organisms possessing cell wall, capable of photosynthesis, producing spore s and having sedentary life. A lot of rethinking has resulted in several differe nt interpretations of the term plant. Before attempting to classify the various organisms, it is necessary to identify and name them. A particular group of individuals, unique in several respects, i s given a unique binomial, and is recognized as a species. These species are gro uped into taxo-nomic groups, which are successively as-signed the ranks of gener a, families, orders, and the process continues till all the spe-cies have been a rranged (classified) under Two Kingdom System The living organisms were originally grouped into two kingdoms. Aristotle divide d all living things between plants, which generally do not move or have sensory organs, and animals. Linnaeus in his Systema naturae 2 Plant Systematics published in 1735 placed them under Animalia (Animals) and Vegetabilia (Plants) as two distinct kingdoms (Linnaeus placed minerals in the third kingdom Minerali a). Linnaeus divided each kingdom into classes, later grouped into phyla for ani mals and di-visions for plants. When single-celled organ-isms were first discove red, they were split between the two kingdoms: mobile forms in the animal phylum Protozoa, and colored algae and bacteria in the plant division Thal-lophyta or Protophyta. As a result, Ernst Haeckel (1866) suggested creating a third kingdom Protista for them, although this was not very popular until relatively recently (sometimes also known as Protoctista). Haeckel recognized three kingdoms: Pro-t ista, Plantae and Animalia. proposal, however, was not taken up immediately, because another classification was proposed by Herbert Copeland (1938), who gave the prokaryotes a separate kin

gdom, originally called Mycota but later referred to as Monera or Bacteria. Cope land later on (1956) proposed a four-kingdom system placing all eukaryotes other than animals and plants in the kingdom Protoctista, thus recognizing four kingd oms Monera, Protoctista, Plantae and Animalia. The importance of grouping these kingdoms in two empires, as suggested earlier by Chatton was popularized by Stan ier and van Niel (1962), and soon became widely accepted. Five Kingdom System Two Empires Three Kingdoms American biologist Robert H. Whittaker (1969) proposed the removal of fungi into a separate kingdom thus establishing a five kingdom system recognizing Monera, Pro-tista, Fungi, Plantae and Animalia as distinct kingdoms. The fungi like plan ts have a distinct cell wall but like animals lack autotrophic mode of nutrition . They, how-ever, unlike animals draw nutrition from decomposition of organic ma tter, have cell wall reinforced with chitin, cell membranes containing ergostero l instead of cholesterol and have a unique biosynthetic pathway for lysine. The classification was followed widely in textbooks. The subsequent discovery that bacteria are radically different from other organi sms in lacking a nucleus, led Chatton (1937) to pro-pose a division of life into two empires: or-ganisms with a nucleus in Eukaryota and organisms without in Pr okaryota. Prokary-otes do not have a nucleus, mitochondria or any other membrane bound organelles. In other words neither their DNA nor any other of their metab olic functions are collected to-gether in a discrete membrane enclosed area. Ins tead everything is openly accessible within the cell, though some bacteria have internal membranes as sites of metabolic activity these membranes do not enclose a separate area of the cytoplasm. Eukaryotes have a separate membrane bound nuc leus, numer-ous mitochondria and other organelles such as the Golgi Body within each of their cells. These areas are separated off from the main mass of the cel l s cytoplasm by their own membrane in order to allow them to be more specialized. The nucleus contains all the Eukaryote cell DNA, which gets organized into dist inct chromosomes during the pro-cess of mitosis and meiosis. The energy is gener ated in mitochondria. The exception to this rule are red blood cells which have no nucleus and do not live very long. Chatton s Six or Seven Kingdoms? Subsequent research concerning the organ-isms previously known as archebacteria has led to the recognition that these creatures form an entirely distinct kingdo m Archaea. These include anaerobic bacteria found in harsh oxygen-free condition s and are genetically and metabolically completely different from other, oxygenbreathing organisms. These bacteria, called Archaebacteria, or simply Archaea, a re said to be living fossils that have survived since the planet s very early ages, before the Earth s atmosphere even had free oxygen. This together with the emphasi s on phylogeny requiring groups to Plants, Taxonomy and Systematics 3 be monophyletic resulted in a six kingdom system proposed by Carl Woese et al. ( 1977). They grouped Archaebacteria and Eubacteria under Prokaryotes and rest of the four king-doms Protista, Fungi, Plantae and Animalia under Eukaryotes. They subsequently (1990) grouped these kingdoms into three domains Bacteria (containi ng Eubacteria), Archaea (containing Archaebacteria) and Eukarya (containing Prot ista, Fungi, Plantae and isms, but they differ in their treatment. Ross (2002, 2005) recognized Archaebac teria and Eubacteria as separate kingdoms, named as Protomonera and Monera, resp ectively again under separate superkingdoms (domains of earlier authors) Archaeb acteriae and Eubacteria. He added seventh kingdom Myxomycophyta of slime moulds under superkingdom Eukaryotes. Two additional superkingdoms of extinct organisms Progenotes (first cells) and Urkaryotes (prokaryotic cells that became eukaryot es) were added: Animalia). Margulis and Schwartz (1998) proposed term superkingdom for domains and recogniz ed two superkingdoms: Prokarya (Prokaryotae) and Eukarya (Eukaryotae). Former in cluded single kingdom Bacteria (Monera) divided into two subkingdoms Archaea and Eubacteria. Eukarya was divided into four kingdoms: Protoctista (Pro-

Superkingdom Progenotes first cells now extinct Superkingdom Archaebacteriae Kingdom Protomonera...archaic bacteria Superkingdom Eubacteria tista), Animalia, Plantae and Fungi. Kingdom Monera bacteria Several recent authors have attempted to recognize seventh kingdom of living org anSuperkingdom Urkaryotes ...prokaryoti cells that became eukaryotes GAiircnmosta _ .i~plallt rr. reolates aodophyta) :lliirialia VIyMmycota Plant Chloroplasts`" Eluli.-s Eatamch& } Cyanobacte xoplasma' Heterolobosca EEiib <i (,tell,, Kinetoplastids Agrobacteriurr rcrosponi armf Euglemids Plant Mitochond q Tnchomonads {;, EraerobaGLI. Diplomonads M1

Sulfolobus PROKARYOTA

Thernwplasma,f: Methambacteris Halob# x :~1 .elia eb a Gtel7,, >

Figure 1.1 Seven kingdoms of life and their possible phylogeny (after Patterson & Sogin 1992). 4 Plant Systematics Superkingdom Eukaryotes ninth kingdom, the Protozoa, are mainly phagotrophic, and have tubular or vesicu lar cristae (or lack mitochondria altogether), and lack tubular mastigonemes on their (primitively anterior) cilia; plastids if present have three-envelop membr anes, chlorophyll c2, and no internal starch, and a plastid endoplasmic reticulu m is absent. Kingdoms 4-9 are primitively anteriorly biciliate. A simpler system of five kingdoms suitable for very elementary teaching is possible by grouping the photosynthetic and fungal kingdoms in pairs. It was suggested that Various c ompromises are possible between the nine and five kingdoms systems; it is sugges ted that the best one for general scientific use is a system of seven kingdoms i n which the Eufungi and Ciliofungi become subkingdoms of the Kingdom Fungi, and the Cryptophyta and Chromophyta subkingdoms of the Kingdom Chromista; the Fungi, Viridiplantae, Biliphyta, and Chromista can be subject to the Botanical Code of Nomenclature, while the Zoological Code can govern the Kingdoms ...cells with nuclei Kingdom Protista. protozoans Kingdom Myxomycophyta...slime molds Kingdom Plantae plants Kingdom Fungi. fungi Kingdom Animalia animals Patterson & Sogin (1992; Figure 1.1) rec-ognized seven kingdoms, but included sl ime moulds under Protozoa (Protista) and instead established Chromista (diatoms) as seventh kingdom. Interestingly the traditional algae now find themselves dis tributed in three dif-ferent kingdoms: eubacterial prokaryotes (the blue-green c yanobacteria), chromistans (diatoms, kelps), and protozoans (green algae, red algae, dinoflagellates, euglenids). Cavalier-Smith (1981) suggested that Eukaryotes can be classified into nine kingdoms each defined in terms of a unique constella-tion of cell structures. Five kingdoms have plate-like mitochondrial c ristae: (1) Eufungi (the non-ciliated fungi, which unlike the other eight kingdo ms have unstacked Golgi cisternae), (2) Ciliofungi (the posteriorly cili-ated fu ngi), (3) Animalia (Animals, sponges, mesozoa, and choanociliates; phagotrophs w ith basically posterior ciliation), (4) Biliphyta (Non-phagotrophic, phycobiliso me-containing, algae; i.e. the Glaucophyceae and Rhodophyceae), (5) Viridiplanta e (Non-phagotrophic green plants, with starch-con-taining plastids). Kingdom (6) , the Euglenozoa, has disc-shaped cristae and an intraciliary dense rod and may be phagotrophic and/or phototrophic with plas-tids with three-membraned envelope s. King-dom (7), the Cryptophyta, has flattened tu-bular cristae, tubular mastig onemes on both cilia, and starch in the compartment between the plastid endoplas mic reticulum and the plastid envelope; their plastids, if present, have phycobi lins inside the paired thylakoids and chlorophyll c2. Kingdom (8), the Chromophy ta, has tubular cristae, together with tubular mastigonemes on one anterior cili um and/or a plastid endoplasmic reticu-lum and chlorophyll c1 + c2. Members of t

he Animalia, Protozoa and Euglenozoa. These 9 kingdoms together with two or one kingdom of prokaryotes total eleven or ten kingdoms of life. Subsequently, however, Cavalier-Smith (1998, 2000, 2004) reverted back to six kingdom classification recognizing Bacteria, Protozoa, Anim alia, Fungi, Plantae and Chromista under two empires Prokaryota and Eukaryota. P rokaryotes constitute a single kingdom, Bacteria, here divided into two new subk ingdoms: Negibacteria, with a cell envelope of two distinct genetic membranes, a nd Unibacteria, comprising the phyla Archaebacteria and Posibacteria. Outline of the classification is as under: Empire Prokaryota Kingdom Bacteria Subkingdom Negibacteria (phyla Eobacteria, Sphingobacteria, Spirochaetae, Proteo bacteria, Planctobacteria, Cyanobacteria) Subkingdom Unibacteria (phyla Posibacteria, Archaebacteria) Plants, Taxonomy and Systematics 5 Empire Eukaryota subunit chaperonin, prefoldin, core histones, small nucleolar ribonucleoproteins (snoRNPs), exosomes and similar replication, repair, transcription and translat ion machinery) that gave rise to archaebacteria and eukaryotes. It is, as such m ore appropriate to call Kingdom Protozoa Subkingdom Sarcomastigota (phyla Amoebozoa, Choanozoa) Subkingdom Biciliata Kingdom Animalia (Myxozoa and 21 other phyla) archaebacteria as metabacteria. Kingdom Fungi (phyla Archemycota, Microsporidia, Basidiomycota) Ascomycota, The eukaryotic host cell evolved from something intermediate between posibacteri a and metabacteria ( archaebacteria ), which had evolved many metabacterial features but not yet switched to ether-linked lipid membranes in a major way. They would no doubt cladistically fall out as primitive metabacteria, but whether such for ms are still extant is un- certain. There are lots of metabacteria out there whi ch are uncultured (only known from environmental sequences) or just undiscovKingdom Plantae Subkingdom Biliphyta (phyla Glaucophyta, Rhodophyta) Subkingdom Viridaeplantae (phyla Bryophyta, Chlorophyta, Tracheophyta) Kingdom Chromista Subkingdom Cryptista (phylum Cryptista: cryptophytes, goniomonads, ered, so who knows. katablepharids) The further shift from archaebacteria to Eukaryotes involved the transformation of circular DNA into a linear DNA bound with histones, formation of membrane bou nd nucleus enclosing chromosomes, develop-ment of mitosis, occurrence of meiosis in sexually reproducing organisms, appearance of membrane bound organelles such as endoplasmic reticulum, golgi bodies and lysosomes, appearance of cytoskeleta l elements like actin, myosin and tubulin, and the formation of mitochondria thr ough enSubkingdom Chromobiota The name archaebacteria seems to be con-fusing. They were so named because they were thought to be the most ancient (Greek archaio meaning ancient) and sometimes labelled as living fossils, since they can sur-vive in anaerobic conditions (met

hanogens-which use hydrogen gas to reduce carbon di-oxide to methane gas), high temperatures (thermophiles, which can survive in tem-peratures of up to 80 degre e C), or salty places (halophiles). They differ from bacteria in hav-ing methion ine as aminoacid that initiates protein synthesis as against formyl-methion-ine in bacteria, presence of introns in some genes, having several different RNA pol y-merases as against one in bacteria, absence of peptidoglycan in cell wall, and growth not inhibited by antibiotics like streptomycin and chloramphenicol. In s everal of these respects archaebacteria are more similar to eukary-otes. Bacteri a are thought to have diverged early from the evolutionary line (the clade neomu ra, with many common characters, notably obligately co-translational secretion o f N-linked glycoproteins, signal recognition particle with 7S RNA and translatio n-arrest domain, protein-spliced tRNA introns, eightdosymbiosis. A major shift in this eukaryotic line which excluded animal and fungi, involved the development of chloroplast by an eukaryotic cell engulfing a photosynthetic bacterial cell (probably a cyanobacterium). The bacterial cell continued to live and multiply inside the eukaryotic cell, provided high energy products, and in turn received a suitable environment to live in. The two thus shared endosymbios is. Over a period of time the bacterial cell lost ability to live independently, some of the bacterial genes getting transferred to eukaryotic host cell, making the two biochemically interdependent. Chloroplast evolution in Euglenoids and D i-noflagellates occurred through secondary endosymbiosis, wherein eukaryotic cel l 6 Plant Systematics aebacteacriArh

aeDeD ythe acteariB oos epp aaimAlin

guinF . Carpel, stamen -,l.-. Seeds C. Secondary growth G Vascular tissue Sporophyte independent Chloroplast TChloroplast g piosnermsA Embryo Gametangia

(secondary (secondary u Endosymbiosis) Endosymbiosis) .................... Greeenn cchhlloorrooppllaasstt Chloroplast (primary endosymbiosis) Mitochondria Cytoskeletal elements: actin, myosin, tubulin ER, Golgi, lysosomes Mitosis , Meiosis Membrane bound nucleus Linear DNA, with histones Figure 1.2 Cladogram showing the evolution of major groups of organisms and the associated apomorphies. Chloroplast evolution has occurred twice, once (primary endosymbios is) eukaryote cell engulfing a photosynthetic bacterial cell, and elewhere (seco ndary endosymbiosis) eukaryotic cell engulfing an eukaryotic cell containing chloropla st. engulfed an eukaryotic cell containing a chloroplast. This common evolutionary s e-quence is shared by green plants (including green algae; green chloroplast), r ed al-gae (red chloroplast) and brown algae and their relatives (commonly known as stramenopiles; brown chloroplast), in which diversification of chloroplast pi gments occurred, along with the thylakoid structure and a variety of storage products The Plant Kingdom It is now universally agreed that members of the plant kingdom include, without doubt the green algae, liverworts and mosses, pteriPlants, Taxonomy and Systematics 7 dophytes, gymnosperms and finally the an-giosperms, the largest group of plants. All these plants share a green chloroplast. Red algae, Brown algae and Glaucoph ytes, latter two together known as stramenophiles, also belong to this kingdom. All these groups share the presence of a chloroplast. All green plants share a g reen chloroplast with chlo-rophyll b, chlorophyll a, thylakoids and grana, and s tarch as storage food. Evolution of cuticle combined with gametangia and embryo characterizes embryophytes, includ-ing bryophytes, pteridophytes and seed plants . The development of vascular tissue of phloem and xylem, and independent sporophyte characterize tracheophytes including pteridophytes and seed plants. Second ary growth resulting in the formation of wood and seed habit differentiates seed plants. The final evolution of a distinct flower, carpels and stamens, together with vessels and sieve tubes set apart the angiosperms, the most TAXONOMY AND SYSTEMATICS There are slightly more than one third of a million species of plants known to m an to-day, the information having been accumulated through efforts of several mi llenniums. Although man has been classifying plants since the advent of civiliza tion, taxonomy was recognized as a formal subject only in 1813 by A. P. de Cando lle as a combination of Greek words taxis (arrangement) and no-mos (rules or law s) in his famous work Theorie elementaire de la botanique. For a long time plant taxonomy was considered as the science of identifying, naming, and classifying p lants (Lawrence, 1951). Since identification and nomenclature are important prere quisites for any classification, taxonomy is often defined as the science dealing with the study of classification, including its bases, principles, rules and pr ocedures (Davis and Heywood, 1963).

highly evolved group of plants. Although Systematics was recognized as a formal major field of study only during the latter half of twentieth century, the term had been in use for a considerab le period. Derived from the Latin word systema (organized whole), forming the ti tle of the famous work of Linnaeus Systema naturae (1735), the term Systematics first appeared in his Genera Plantarum (1737), though Huxley (1888) is often cre dited to have made the first use of the term in his article in Nature on the sys tematics of birds. Simpson (1961) defined systematics as a scientific study of th e kinds and diversity of organisms, and of any and all relationships between the m . It was recognized as a more inclusive field of study concerned with the study of diver-sity of plants and their naming, classification and evolution. The scop e of taxonomy has, however, been enlarged in recent years to make taxonomy and s ystematics synonymous. A broader definition (Stace, 1980) of taxonomy, to coinci de with systematics recognized it as the study and description of variation in or ganisms, the investigation of causes and consequences of this variation, and the manipulation of the data The species of living organisms on this planet include Monera-10,000; Protista-2 50,000; Fungi-100,000; Plantae-279,000; Animalia-1,130,000. Nearly three fourth of animals are insects (800,0000) and of these more than one third beetles (300, 000). Amongst plants nearly 15,000 species be-long to usually overlooked mosses and liv-erworts, 10,000 ferns and their allies, 820 to gymnosperms and 253,000 t o an-giosperms (belonging to about 485 families and 13,372 genera), considered t o be the most recent and vigorous group of plants that have occurred on earth. A ngiosperms occupy the majority of the terrestrial space on earth, and are the ma jor components of the world s vegetation. Brazil and Colombia, both located in the tropics, are considered to be countries with the most diverse angiosperms floras and which rank first and second. China , even though the main part of her land is not lo-cated in the tropics, the numb er of her an-giosperms still occupies the third place in the world, and has appr oximately 300 fami-lies, 3, 100 genera and 30,000 species. 8 Plant Systematics obtained to produce a system of classifi- with never-ending duties. The continuo us cation . flow of data necessitates rendering descriptive information, revising schemes of identification, revaluating and improving systems of classification and perceiv ing new relationships for a better understanding of the plants. The discipline a s such includes Realization of the fact that a good number of authors still consider taxonomy to be a more restricted term and systematics a more inclusive one has led recent authors to pre-fer the term systematics to i nclude discus-sion about all recent developments in their all activities that are a part of the effort to works. Modern approach to systematics aims organize and record the diversity of plants at reconstructing the entire chronicle of and appreciate the fascinating differences evolutionary events, including the formation among the species of pl ants. Systematic ac-of separate lineages and evolutionary modi- tivities are bas ic to all other biological scifications in characteristics of the organisms. enc es, but also depend, in turn, on other disIt ultimately aims at discovering all the ciplines for data and information useful in branches of the evolutionary tre e of life; and constructing classification. Certain disci-to document all the ch anges and to describe plines of biology such as cytology, genetics, all the spec ies which form the tips of these ecology, palynology, paleobotany and phytobranches. This won t be possible unless in-formation is consolidated in the form o f an unambiguous system of classification. This, however, is again impossible without a clear understanding of the basic identification and nomenclatural methods. Equally important

geography are so closely tied up with systematics that they can not be practiced with-out basic systematic information. Experiments cannot be carried out unless the or-ganisms are correctly identified and some information regarding their re lationship is is the understanding of the recent tools of available. The understanding of rela tion-data handling, newer concepts of ships is particularly useful in the applie d phylogenetics, expertise in the judicious uti- fields of plant breeding, horti culture, forestry lization of fast accumulating molecular data and pharmacology for exploring the usefulin understanding of affinities between taxa. ness of related species. Knowledge of sysPrior to the evolutionary theory of Darwin, relationships were expressed as natu ral af-finities on the basis of an overall similarity in morphological features. Darwin ushered in an era of assessing phylogenetic rela-tionships based on the course of evolutionary descent. With the introduction of com-puters and refined statistical procedures, overall similarity is represented as phenetic relationsh ip, which takes into account ev-ery available feature, derived from such di-vers e fields as anatomy, embryology, mor-phology, palynology, cytology, phytochemistry, physiology, ecology, phytogeography and tematics often guides the search for plants of potential commercial importance. Basic Components (Principles) of Systematics ultrastructure. Various systematic activities are directed towards the singular goal of construc ting an ideal system of classification that necessitates the procedures of ident ification, description, nomenclature and constructing affinities. This enables a better management of information to be utilized by different workers, investiga ting different aspects, structure and functioning of different species of plant. With the advancement of biological fields, new information flows continuously an d the taxonomists are faced with the challenge of integrating and providing a sy nthesis of all the available data. Systematics now is, thus, an unending synthes is, a dynamic science Identification Identification or determination is recognizing an unknown specimen with an alrea dy Plants, Taxonomy and Systematics 9 known taxon, and assigning a correct rank and position in an extant classificati on. In practice, it involves finding a name for an unknown specimen. This may be achieved by visiting a herbarium and comparing unknown specimen with duly ident ified specimens stored in the herbarium. Alternately, the specimen may also be s ent to an expert in Whereas the fresh specimens can be de-scribed conveniently, the dry specimens ne ed to be softened in boiling water or in a wet-ting agent before these could be described. Softening is often essential for dissection of flowers in order to st udy their details. Nomenclature the field who can help in the identification. Nomenclature deals with the determination of a correct name for a taxon. There a re Identification can also be achieved using various types of literature such as Fl oras, different sets of rules for different groups of Monographs or Manuals and making use of living organisms. Nomenclature of plants identification keys provided in these sources of literature. After the unknown specimen has been provisionally identified with the help of a key, the identific ation can be fur-ther confirmed by comparison with the de-tailed description of the taxon provided in the (including fungi) is governed by the Inter- national Code of Botanical Nomenclat ure (ICBN) through its rules and recommendaliterature source. tions. Updated every six years or so, the Botanical Code helps in picking up a s

ingle correct name out of numerous scientific names available for a taxon, with a particular circumscription, position and rank. To avoid inconvenient name chan ges for certain taxa, a list of conserved names is provided in the Code. Cultiva ted plants are governed by the International Code of No-menclature for Cultivate d Plants (ICNCP), slightly modified from and largely based on A method that is becoming popular over the recent years involves taking a photograph of the plant and its parts, uploading this picture on the website and info rming the members of appropriate electronic Lists or Newsgroups, who can see the photograph at the website and send their comments to the enquirer. Members of t he fraternity could thus help each other in identification in a much efficient m anner. the Botanical Code. Description Names of animals are governed by the In-ternational Code of Zoological Nomenclat ure (ICZN); those of bacteria by International Code for the Nomenclature of Bact eria (ICNB), now called Bacteriological Code (BC). A separate Code exists for vi ruses, named the International Code of Virus ClassificaThe description of a taxon involves listing its features by recording the approp riate character states. A shortened description consisting of only those taxonom ic charac-ters which help in separating a taxon from other closely related taxa, forms the diag-nosis, and the characters are termed as di-agnostic characters. The diagnostic char-acters for a taxon determine its circumscrip-tion. The descr iption is recorded in a set pat-tern (habit, stem, leaves, flower, sepals, pet-a ls, stamens, carpels, fruit, etc.). For each character, an appropriate character -state is listed. Flower colour (character) may thus be red, yellow, white, etc. (states). The descrip-tion is recorded in semi-technical language using specifi c terms for each character state to enable a proper documentation of data. tion and Nomenclature (ICVCN). With the onset of electronic revolution and the need to have a common database f or living organisms for global communication a common uniform code is being atte mpted. The Draft BioCode is the first public expression of these objectives. The first draft was prepared in 1995. After successive reviews the fourth draft, na med Draft BioCode (1997) prepared by the International Committee for Bionomencla ture was published by Greuter et al., (1998) and is now available on the web. Th e last decade of twentieth century also saw the development of rankless PhyloCod e based on the concepts of phylogenetic 10 Plant Systematics systematics. It omits all ranks except spe-cies and clades based on the concept of rec-ognition of monophyletic groups. The latest version of PhyloCode (PhyloCode 4b, 2007) is also available on the web. (a taxonomic group assigned to any rank; pl. taxa), dividing a taxon into smalle r units, uniting two or more taxa into one, transferPhylogeny ring its position from one group to another and altering its rank. Once establis hed, a classification provides an important mechanism of information storage, re trieval and usage. This ranked system of classification is popularly known as th e IIinnaean system. Taxonomic entities are classified in different fashions: Phylogeny is the study of the genealogy and evolutionary history of a taxonomic group. Genealogy is the study of ancestral relation-ships and lineages. Relation ships are depicted through a diagram better known as a phylogram (Stace, 1989), since the com-monly used term cladogram is more appro-priately used for a diagra m constructed through cladistic methodology. A phylogram is a branching diagram based on the degree of advancement (apomorphy) in the descen-dants, the longest branch representing the most advanced group. This is distinct from a phylogeneti c tree in which the vertical scale represents a geological time-scale and all li v-ing groups reach the top, with primitive ones near the centre and advanced one s near the periphery. Monophyletic groups, including all the descendants of a co mmon ancestor, are recognized and form entities in a classifica-tion system. Par aphyletic groups, wherein some descendants of a common ancestor are left out, ar

e reunited. Polyphyletic groups, with more than one common ancestor, are split t o form monophyletic groups. Phenetic infor-mation may often help in determining a phy-logenetic relationship. 1. Artificial classification is utilitarian, based on arbitrary, easily observable characters such as habit, colour, number, form or similar features. The sexual system of Linnaeus, which fits in this cate gory, utilized the number of stamens for primary classification of the flowering plants. 2. Natural classification uses overall similarity in grouping taxa, a concept initiated by M. Adanson and culminating i n the extensively used classification of Bentham and Hooker. Natural systems of the eighteenth and nineteenth centuries used morphology in delimiting the overal l similarity. The concept of overall similarity has undergone considerable refin e-ment in recent years. As against the sole morphological features as indicators of similarity in natural systems, overall similarity is now judged on the basis of features derived from all the available fields of taxonomic informaClassification tion (phenetic relationship). Classification is an arrangement of organ-isms into groups on the basis of simil ari-ties. The groups are, in turn, assembled into more inclusive groups, until a ll the organ-isms have been assembled into a single most inclusive group. In seq uence of in-creasing inclusiveness, the groups are as-signed to a fixed hierarch y of categories such as species, genus, family, order, class and division, the f inal arrangement consti-tuting a system of classification. The pro-cess of class ification includes assigning ap-propriate position and rank to a new taxon 3. Phenetic Classification makes the use of overall similarity in terms of a phenetic relationship based on data from all available sources such as morphology, anatomy, embryology, phytochemistry, ultrastructure and, in fact, all other fields of study. Phenetic classifications were strongly advocated by Sneath and Sokal (1973) but did not find much favour with major systems of classification of higher plants. Phenetic relationship has, however, been very prominently used Plants, Taxonomy and Systematics 11 in modern phylogenetic systems to decide the realignments within the Cronquist, Thorne and Dahlgren, are largely based on decisions in which phenetic information is liberally used in deciding the phylogenetic relationship between groups, differing largely on the weightage given to the system of classification. 4. Phylogenetic classification is based on the evolutionary descent of a group of organisms, the relationship de-picted either through a phylogram, phylogenetic tree or a cladogram. Classification is constructed with this premise in mind, that all the descen-dants of a common anc estor should be placed in the same group (i.e., group should be monophyletic). I f some de-scendents have been left out, render-ing the group paraphyletic, these are brought back into the group to make it monophyletic (merger of Asclepiadace ae with Apocynaceae, and the merger of Capparaceae with Brassicaceae in recent c lassifica-tions). Similarly, if the group is poly-phyletic (with members from mo re than one phyletic lines, it is split to create monophyletic taxa (Genus Arena ria split into Arenaria and Minuarlia). This approach, known as cladistics, is p racticed by cladists. cladistic or phenetic relationship. There have been suggestions to abandon the hierarchical contemporary classificat ions based on the Linnaean system, which employs various fixed ranks in an estab lished conventional sequence with a phylogenetic taxonomy in which monophyletic gr oups would be unranked names, defined in terms of a common ancestry, and diagnos ed by reference to synapomorphies (de Queiroz and Gauthier, 1990; Hibbett and Donoghue, 1998). 5. Evolutionary taxonomic classificaClassification not only helps in the place-ment of an entity in a logically orga

nized scheme of relationships, it also has a great predictive value. The presenc e of a valuable chemical component in one species of a par-ticular genus may pro mpt its search in other related species. The more a classification reflects phyl ogenetic relationships, the more predictive it is supposed to be. The meaning of a natural classification is gradually losing its traditional sense. A natural cl assification today is one visualized as truly phylogenetic, establishing monophyl etic groups making fair use of the phenetic information so that such groups also reflect a phenetic relationship (overall similarity) and the classification rep resents a reconstruction of the evolutionary descent. tion differs from a phylogenetic clas-sification in that the gaps in the varia-t ion pattern of phylogenetically adja-cent groups are regarded as more im-portant in recognizing groups. It ac-cepts leaving out certain descendants of a common ancestor (i.e. recogniz-ing paraphyletic groups) if the gaps are not significant , thus failing to pro-vide a true picture of the genealogical history. The chara cters considered to be of significance in the evolution (and the classification based on these) are dependent on expertise, authority and intuition of systemati sts. Such clas-sifications have been advocated by Simpson (1961), Ashlock (1979) , Mayr and Ashlock (1991) and Stuessy (1990). The approach, known as eclecticism , Aims of Systematics The activities of plant systematics are basic to all other biological sciences a nd, in turn, depend on the same for any additional information that might prove useful in con-structing a classification. These activities are directed towards achieving the undermentioned aims: 1. To provide a convenient method of is practiced by eclecticists. identification and communication. A workable classification having the taxa arra nged in hierarchy, detailed and diagnostic descriptions are essential The contemporary phylogenetic systems of classification, including those of Takh tajan, 12 Plant Systematics for identification. Properly identified and arranged herbarium specimens, dichot omous keys, polyclaves and com-puter-aided identification are impor-tant aids fo r identification. The Code (ICBN), written and documented through the efforts of IAPT (Internaprimitive groups towards the centre and the advanced ones towards the periphery. 5. To provide an integration of all availational Association of Plant Taxonomy), helps in deciding the single correct name acceptable to the whole botanible information. To gather information from all the fields of study, analysing t his information using statistical procedures with the help of computers, providi ng a synthesis of this information and developing a classification based on over all similarity. This synthesis is unending, however, since scientific progress w ill continue and new information will continue to pour and pose new challenges f or cal community. 2. To provide an inventory of the world s flora. Although a single world Flora is difficult to come by, floristic records of continents (Continental Floras; cf. Flora Europaea by Tutin et al.), regions or countries (Regional Floras; cf. Flora of British India by J. D. Hooker) and s tates or even counties (Local Floras; cf. Flora of Delhi by J. K. Maheshwari) ar e well documented. In addition, World Monographs for selected genera (e.g., The genus Crepis by Babcock) and fami-lies (e.g., Das pflanzenreich ed. by taxonomists. 6. To provide an information reference, supplying the methodology for information storage, retrieval, exchange and utili zation. To provide significantly valuable information concerning en-dangered spe

cies, unique elements, genetic and ecological diversity. 7. To provide new concepts, reinterpret A. Engler) are also available. the old, and develop new procedures for correct determination of taxonomic affin ities, in terms of phylogeny and 3. To detect evolution at work; to reconstruct the evolutionary history of the plant kingdom, determining the se-quence of evolutionary change and phenetics. 8. To provide integrated databases includcharacter modification. ing all species of plants (and possibly all organisms) across the globe. Several big organizations have come to-gether to establish online searchable databases of taxon names, images, descriptions, synonyms and molecular information. 4. To provide a system of classification which depicts the evolution within the group. The phylogenetic relationship betw een the groups is commonly de-picted with the help of a phylogram, wherein the l ongest branches repre-sent more advanced groups and the shorter, nearer the base , primitive ones. In addition, the groups are repAdvancement Levels in Systematics resented by balloons of different sizes that are proportional to the number of Plant systematics has made considerable s pecies in the respective groups. Such strides from herbarium records to dataa phylogram is popularly known as a banks, recording information on every bubble diagram. The phylogenetic re-lationship could also be presented in the fo rm of a phylogenetic tree (with vertical axis representing the geologi-cal time scale), where existing species reach the top and the bubble diagram may be a cro ss-section of the top with possible attribute of a plant. Because of extreme climatic diversity, floristic variability, inaccessibility of certain regions and economic disparity of differ ent regions, the present-day systematics finds itself in different stages of adv ancement in different parts of the world. Tropical Asia and tropical Plants, Taxonomy and Systematics 13 Africa are amongst the richest areas of the world in terms of floristic diversit y but amongst the poorest as far as the economic resources to pursue complete do cumenta-tion of systematic information. The whole of Europe, with more than 30 m square kilometres of landscape and numerous rich nations with their vast econom ic resources, have to account for slightly more than 6 thou-sand species of vasc ular plants. India, on the other hand, with meager resources, less This development is helpful in the preparation of Floras and Monographs. It also aids in better understanding of the degree of variation within a species. Two o r more herbarium specimens may appear to be sufficiently different and regarded as belonging to different species on the basis of a few available herbarium reco rds, but only a field study of populations involving thousands of specimens can help in reaching at a better understanding of their status. If there are than one tenth of landscape, has to account enough field specimens to fill in th e gaps for the study of at least four times more of in variation pattern, there is no justificathe vascular plants. A small country like tion in regarding them as separate species. Colombia, similarly, has estimated 4,5000 On the other hand , if there are distinct gaps different species, with only a few botanists in the variation pattern, it strengthens to study the flora. Great Britain, on the oth er their separate identity. In fact, many plants, hand, has approximately 1370 t axa (Wood- described as species on the basis of limited land, 1991), with thousa nds of professional material in the pioneer phase, are found to and amateur bota nists available to document be variants of other species in the consolithe infor mation. It is not strange, as such, dation phase. Most parts of central Europe, that there is lot of disparity in the level of North America and Japan are exper iencadvancement concerning knowledge about ing this phase.

respective floras. Taxonomic advancement today can be conveniently divided into four distinct phases encountered in different parts of the world: Experimental or BtOSystemattC Phase Exploratory or Pioneer Phase This phase marks the beginning of plant tax-onomy, collecting specimens and buil ding herbarium records. The few specimens of a During this phase, the herbarium records and variation studies are complete. In addition, information on biosystematics (studies on transplant experiments, bree ding behaviour and chromosomes) is also avail-able. Transplant experiments invol ve colspecies in the herbarium are the only record of its variation. These specimens are, how- lecting seeds, saplings or other pro pagules ever, useful in a preliminary inventory of from morphologically distinct populations flora through discovery, description, naming from different habitat s and growing them and identification of plants. Here, morphol- under common env ironmental conditions. If ogy and distribution provide the data on the differenc es between the original popuwhich the systematists must rely. Taxo- lations were purely ecological, the differnomic experience and judgement are par- ences woul d disappear under a common enticularly important in this phase. Most ar- vironme nt, and there is no justification in eas of tropical Africa and tropical Asia are regarding them as distinct taxonomi c entipassing through this phase. ties. On the other hand, if the differences still persist, these are evidently g enetically fixed. If these populations are allowed to grow together for several years, their breeding behaviours would further establish their status. If there are complete reproductive bar-riers between the populations, they will fail to i nterbreed, and maintain their separate Consolidation or Systematic Phase During this phase, herbarium records are ample and enough information is availab le concerning variation from field studies. 14 Plant Systematics identity. These evidently belong to different species. On the other hand, if the re is no reproductive isolation between them, over the years, they would interbr eed, form in-termediate hybrids, which will soon fill the gaps in their variatio n. Such populations evidently belong to the same species and better distinguishe d as ecotypes, subspecies or varieties. Further chromosomal studies can throw mo re light on their affinities and status. Central Europe has reached this phase o f plant systematics. is assembled, analyzed, and a meaningful synthesis of analysis is provided for u nder-standing phylogeny. Collection of data, analysis and synthesis are the jobs of an independent discipline of systematics, referred to as numerical taxonomy. Encyclopaedic or Holotaxonomic Phase The first two phases of systematics are often considered under alpha-taxonomy an d the last phase under omega-taxonomy. At present, only a few persons are involv ed in encyclopaedic work and that too, in a few isolated taxa. It may thus be sa fe to conclude that though in a few groups omega-taxonomy is within reach, for t he great majority of plants, mainly in the tropics, even the al- pha stage has not been crossed. The total integration of available information for the plant king dom is, thus, only a distant dream at present. Here, not only the previous three phases are attained, but information on all th e botani-cal fields is also available. This information Chapter 2 Botanical Nomenclature

Nomenclature deals with the application of a correct name to a plant or a taxono mic group. In practice, nomenclature is often combined with identification, sinc e while identifying an unknown plant specimen, the author chooses and applies th e correct name. The favourite temperate plant is correctly identified whether yo u call it Seb (vernacular Hindi name), Apple, Pyrus malus or Malus malus, but only by using the correct scientific name Malus domestica does one combine identification with nomenclature. The current ac-tivity of botanical nomenclature is gov erned by the International Code of Botanical Nomen-clature (ICBN) published by t he International are allowed across the codes. The generic name Cecropia applies to showy moths a s also to tropical trees. Genus Pieris, similarly, refers to some butterflies and shrubs. During the last decade, there have been attempts at developing unified code for all living organisms, for convenient handling of combined database for all organ isms. Draft BioCode and PhploCode, have been con-certed efforts in this directio n, but it will take a long time before acceptability of these endeavours can be determined. NEED FOR SCIENTIFIC NAMES Scientific names formulated in Latin are preAssociation of Plant Taxonomy (IAPT). The Code is revised after changes at each Inter- ferred over vernacular or common na mes national Botanical Congress. The naming of since the latter pose a number of pro blems: the animals is governed by the International 1. Vernacular names are not available for Code of Zoological Nomenclature (I that of bacteria by the International Code for the Nomenclature of Bacteria (ICN B; now known as Bacteriological Code-BC). Virus nomenclature is governed by Inte rnational Code of Virus Classification and Nomencla-ture (ICVCN). Naming of cult ivated plants is governed by the International Code of Nomen-clature for Cultiva ted Plants (ICNCP), which is largely based on ICBN with a few additional provisi ons. Whereas within the provisions of a particular code no two taxa can bear the same correct scientific name, same names and all the species known to man. 2. Vernacular names are restricted in their usage and are applicable in a single or a few languages only. They are not universal in their application. 3. Common names usually do not provide information indicating family or generic relationship. Roses belong to the genus Rosa; woodrose is a member of the genus Ipomoea and primrose be-longs to the ge nus Primula. The three genera, in turn, belong to three differ-ent families Rosace ae, Convolvu16 Plant Systematics laceae and Primulaceae, respectively. Oak is similarly common name for the speci es of genus Quercus, but Tanbark oak is Lithocarpus, poison oak a Rhus, silver o ak a Grevillea and Jerusalem DEVELOPMENT OF BOTANICAL CODE oak a Chenopodium. For several centuries, the names of plants appeared as polynomials long descriptiv e phrases, often difficult to remember. A species of willow, for example, was na med Salix 4. Frequently, especially in widely distributed plants, many common names may exist for the same species in the same langua ge in the same or differ-ent localities. Cornflower, bluebottle, bachelor s button and ragged robin all refer to the same species Centaurea pumila angustifolia altera by Clusius in his herbal (1583). Casper Bauhin (1623)

introcyanus. duced the concept of Binomial nomenclature under which the name of a species con sists of two parts, the first the name of the genus to which it belongs and the second the spe- cific epithet. Onion is thus appropriately named Allium cepa, Al lium being the generic name and cepa the specific epithet. Bauhin, however, did not use binomial nomenclature for all the species and it was left to Carolus Lin naeus to firmly establish this system of naming in his Species plantarum (1753). The early rules of nomenclature were set forth by Linnaeus in his Critica botan ica (1737) and further amplified in Philosophica botanica (1751). A. P. de Cando lle, in his Theorie elementaire de la botanique (1813), gave ex- plicit instruct ions on nomenclatural procedures, many taken from Linnaeus. Steudel, in Nomencla tor botanicus (1821), provided Latin names for all flowering plants known 5. Often, two or more unrelated species are known by the same common name. Bachelor s button, may thus be Tanacetum vulgar e, Knautia arvensis or Centaurea cyanus. Cockscomb, is simi-larly, a common name for Celosia cristata but is also applied to a seaweed Ploca-mium coccinium or t o Rhinanthus minor. Why Latin? Scientific names are treated as Latin regard-less of their origin. It is also ma ndatory to have a Latin diagnosis for any new taxon published 1 January 1935 onw ards. The cus-tom of Latinized names and texts originates from medieval scholars hip and custom con-tinued in most botanical publications until the middle of nin eteenth century. Descripto the author together with their synonyms. The first organized effort towards the detions of plants are not written in classical Latin of Cicero or of Horace, but i n the linvelopment of uniform botanical nomenclature was made by Alphonse de Candolle, wh o circulated a copy of his manuscript Lois de la nomenclature botanique. After d eliberations of the First International Botanical Congress at Paris (1867), the Paris Code, also known as de Candolle rules was adopted. Linnaeus (1753) was made the starting point for plant nomenclature and the rule of priority was made fund amental. Not satisfied with the Paris Code, the American botanists adopted a sep arate Rochester Code (1892), which introduced the concept of types, strict appli cation of rules of priority even if the name was a tautonym (specific epithet re peating the gua franca spoken and written by scholars during middle ages, based on popular La tin spoken by ordinary people in the classical times. The selection has several adva ntages over modern languages: i) Latin is a dead lan-guage and as such meanings and interpre-tation are not subject to changes unlike, En-glish and other langua ges; ii) Latin is spe-cific and exact in meaning; iii) grammatical sense of the word is commonly obvious (white translated as album-neuter, alba-feminine or alb us- masculine); and iv) Latin language employs the Roman alphabet, which fits we ll in the text of most languages. generic name, e.g. Malus malus). The Paris Code was replaced by the Vienna Code (1905), which established Species plantarum (1753) of Linnaeus as the starting Botanical Nomenclature 17 point; tautonym was not accepted, and Latin diagnosis was made essential for new spe-cies. In addition, a list of conserved generic names (Nomina generic conser vanda) was approved. Not satisfied with the Vienna Code also, adherents of the R ochester Code adopted the American Code (1907), which did not ac-cept the list o f conserved names and the reIIB. Nomina familiarum bryophytorum et spermatophytorum conservanda IIIA. Nomina generica conservanda et rejicienda IIIB. Nomina specifica conservanda et

rejicienda IV. Nomina utique rejicienda (A. Algae, B. Fungi, C.Bryophyta, D. Pteridoquirement for Latin diagnosis. phyta, E. Spermatophyta) V. Opera utique oppressa It was not until the 5th International Bo-tanical Congress (IBC) at Cambridge (1 930) that the differences were finally resolved and a truly International Code e volved, accept-ing the concept of type method, rejecting the tautonyms, making L atin diagnosis manda-tory for new groups and approving conserved generic names. The Code has since been constantly amended at each International Botanical Congr ess. The 15th IBC was held at Tokyo in 1993, 16th at St Louis in 1999 (published by Greuter et al., 2000). The Code discussed in the following pages is based on the 17th International Botanical Congress held at Vienna in 2005 (Published by McNeill et al., 2006- Code is generally published one year after the Congress). The 18th Interna-tional Botanical Congress would be held in Melbourne, Australia in 2011. The last three useful appendices were in-cluded for the first time in the Tokyo Code. The first (IIIB) includes the names of conserved and rejected specific nam es; the second (IV) lists the names and all combinations based on these names, w hich are ruled as rejected under Art. 56, and none is to be used; and the last ( V) the list of publications (and the category of taxa therein) which are not validly published according to the Code. Principles form the basis of the system of botanical nomenclature. There are 62 main rules (set out as articles) and associated recommendations. The object of t he rules is to put the nomenclature of the past into order and provide for that of the future; names contrary to the rules cannot be maintained. Recommendations deal with subsidiary points, and are meant for uniformity and clarity. Names co ntrary to the recommendations cannot, on that account, be rejected, but they are not examples to be followed. Conserved names include those that do not satisfy the principle of priority but are sanctioned for use. The various rules and reco mmendations are discussed here under relevant headings. CONTENTS OF BOTANICAL CODE Publication of the Code is based on the real-ization that botany requires a prec ise and simple system of nomenclature used by bota-nists in all countries. The C ode aims at pro-vision of a stable method of naming taxonomic groups, avoiding a nd rejecting the use of names which may cause error or ambiguity or throw scienc e into confusion. Preamble highlights the philosophy of the botanical Preamble 1. Botany requires a precise and simple Code. The Code is divided into 3 divisions: system of nomenclature used by botanists in all countries, dealing on the one ha nd with the terms which denote the ranks of taxonomic groups or units, and on th e other hand with the scientific names which are applied to the individual taxon omic groups of plants. The purpose of giving a name to a taxonomic group is not to indicate its characters or history, but to supply a I. Principles II. Rules and recommendations III. Provisions for the governance of the Code In addition, the Code includes th e following appendices: I. Names of hybrids IIA. Nomina familiarum algarum, fungorum, pteridophytorum et fossilium conservanda et rejicienda 18 Plant Systematics means of referring to it and to indicate its taxonomic rank. This Code aims at t he provision of a stable method of nam-ing taxonomic groups, avoiding and reject ing the use of names which may cause error or ambiguity or throw science into co nfusion. Next in impor-tance is the avoidance of the useless creation of names. Other consider-ations, such as absolute grammatical correctness, regularity or e

uphony of names, more or less prevailing custom, regard for persons, etc., notwi thstand-ing their undeniable importance, are and formation of names for special plant categories in agricultural, forestry, and horticultural nomenclature. 9. The only proper reasons for changing a name are either a more profound knowledge of the facts resulting from adequate taxonomic study or the necessity of giving up a nomenclature that is contrary to the rules. 10. In the absence of a relevant rule or where the consequences of rules are doubtful, established custom is followed. relatively accessory. 11. This edition of the Code supersedes all previous editions. 2. The Principles form the basis of the system of botanical nomenclature. Principles of ICBN 3. The detailed Provisions are divided into The International Code of Botanical Nomenclature is based on the following set o f six principles, which are the philosophical basis of the Code and provide guid elines for the tax-onomists who propose amendments or deliberate on the suggesti ons for modification of Rules, set out in the Articles, and Rec-ommendations. Examples (Ex.) are added t o the rules and recommendations to illustrate them. 4. The object of the Rules is to put the nomenclature of the past into order and to provide for that of the future; names contrary to a rule cannot be mainthe Code: 1. Botanical Nomenclature is independtained. ent of Zoological Nomenclature. The Code applies equally to the names of taxonom ic groups treated as plants whether or not these groups were 5. The Recommendations deal with subsidiary points, their object being to bring about greater uniformity and clar-it y, especially in future nomenclature; names contrary to a recommendation cannot, on that account, be rejected, originally so treated. 2. The application of names of taxonomic groups is determined by means of but they are not examples to be followed. nomenclatural types. 6. The provisions regulating the gover3. Nomenclature of a taxonomic group is based upon priority of publication. nance of this Code form its last division. 4. Each taxonomic group with a particu7. The rules and recommendations apply lar circumscription, position and rank can bear only one correct name, the earli est that is in accordance with the to all organisms traditionally treated as plants, whether fossil or non-fossil, e.g., blue-green algae, Cyanobacteria, fungi, including chytrids, oomycetes, and slime moulds, photosynthetic pro-tists and taxonomically related nonrules. 5. Scientific names of taxonomic groups are treated as Latin, regardless of derivation. photosynthetic groups. 6. The rules of nomenclature are retroactive, unless expressly limited. 8. The International code of nomenclature for cultivated plants is prepared under the authority of the International Com-m

ission for the Nomenclature of Cultivated Plants and deals with the use Names of Taxa Taxon (pl. taxa) refers to a taxonomic group Botanical Nomenclature 19 belonging to any rank. The system of nomen-clature provides a hierarchical arran gement of ranks. Every plant is treated as belonging to a number of taxa, each a ssigned a particu-lar rank. Onion thus belongs to Allium cepa (species rank), Al lium (genus rank), Alliaceae (family rank) and so on. The seven principal obliga tory ranks of taxa in descending se-quence are: kingdom (regnum), division or ph ylum (divisio, phylum), class (classis), or-der (ordo), family (familia), genus (genus), and species (species). The ending of the name indicates its rank: endin g -bionta denotes a kingdom, -phyta a division, -phytina a subdivision, -opsida a class, -opsidae or -idae a subclass, -ales an order, -ineae a suborder and -aceae a family. The detailed hierarchy of ranks and endings with examples is given in Table 2.1. Stevens (2005) describes this system of naming where endings determine ranks of taxa and suggest relative positions of groups i n local hierarchy as flagged hierarchy. The names of the groups belonging to ranks above the level of genus are uninomia ls in the plural case. Thus, it is appropriate to say Winteraceae are primitive d inappropriate when we say Winteraceae is primitive . Table 2.1 Ranks and endings provided by the ICBN Rank Ending Example Kingdom -bionta Chlorobionta Division -phyta Magnoliophyta -mycota (Fungi) Eumycota Subdivision -phytina Pterophytina -mycotina (Fungi) Eumycotina Class -opsida Magnoliopsida -phyceae (Algae) Chlorophyceae -mycetes (Fungi) Basidiomycetes Subclass -opsidae Pteropsidae -idae (Seed plants) Rosidae -physidae (Algae) Cyanophysidae -mycetidae (Fungi) Basidiomycetidae Order -ales

an

Rosales Suborder -ineae Rosineae Family -aceae Rosaceae Subfamily -oideae Rosoideae Tribe -eae Roseae Subtribe -inae Rosinae Genus -us, -um, -is, -a, -on Pyrus, Allium, Arabis, Rosa, Polypogon Subgenus Cuscuta subgenus Eucuscuta Scrophularia section Anastomosanthes Scrophularia sub section Vernales Scrophularia series Lateriflorae Section Subsection Series Species Rosa canina Crepis sancta subsp. bifida Lantana camara var. varia Subspecies Varietas Forma Tectona grandis f. punctata 20 Plant Systematics

The focus changes when we are mentioning the rank with it. Thus, r-aceae is primitive is a logically correct Traditional Alternate Type name name mm

the family Winte

genus statement. Cruciferae Brassicaceae Brassica The name of a taxon above the rank of fam-ily may be formed by replacing the ter mina-tion -aceae in the name of an included fam-ily by the termination denoting their rank (order Rosales from family Rosaceae, class Guttiferae Clusiaceae Clusia Leguminosae

Fabaceae Faba Umbelliferae Apiaceae Apium Compositae Asteraceae Aster Labiatae Lamiaceae Lamium Palmae Arecaceae Areca Magnoliopsida from family Magnoliaceae). Gramineae Poaceae Poa The name of a family is a plural adjective used as a noun. It is formed from the name of the type genus by replacing the genitive singular (gender) ending with the termina-tion -aceae in the genera of classical Latin or Greek origin (Family Rosaceae from genus Rosa, Potamogetonaceae from Potamogeton). For generic names of non-classical origin, when analogy with classical names is insufficient to d etermine the genitive singular, -aceae is added to the full word (Ginkgoaceae fr om Ginkgo). For generic names with alternative genitives the one implicitly used by the original author must be maintained (Nelumbonaceae from Nelumbo Nelumbonis declined by analogy The alternate names of these families which are in accordance with the ICBN rule s need to be encouraged. Under a unique exception to article 18 of the Code, the name Leguminosae is sanc tioned as alternate name for Fabaceae only as long as it includes all the three subfamilies: Faboideae (Papilionoideae), Caesalpinioideae and Mimosoideae. In ca se these are upgraded as families, then the name Papilionaceae is conserved agai nst Leguminosae for the first of these getting the name Fabaceae. The two altern ate names allowed then are Papilionaceae and Fabaceae. with umbo and umbonis). Fossil taxa may be treated as morphotaxa. A morphotaxon is defined as a fossil t axon, which for nomenclatural purposes, comprises only the parts, life-history s tages, or preservational states represented by the corresponding nomenclatural t ype. The endings for ranks, subclass and above are recommendations, whereas for order and below these are mandatory rules. It is, thus, nothing strange that group na mes such as Gymnosperms, Angiosperms, Bryophytes, Pteridophytes, Lignosae, Herba ceae, Dicotyledoneae, Monocotyledoneae, etc. have been used as valid group names for supraordinal taxa. Recently developed ver-sions of the APG classification r ecognize only informal group names such as Paleoherbs, Tricolpates (Eudicots), A sterids, Rosids, Euasterids, Eurosids above the order level as monophyletic clad es. No formal taxonomic names are used above the level of the order. The name of a family ends in -aceae. The following eight families of angiosperms, how-ever, whose original names are not in accor-dance with the rules but the use of these names has been sanctioned because of old traditional usage. The type genus of e ach family is listed: Genus The generic name is a uninomial singular word treated as a noun. The examples of the shortest generic name Aa as well as the long-est name Brassosophrolaeliocat tleya (26 characters), both belong to the family Orchidaceae. The genus may have

a masculine, neuter or feminine form as indicated by the ending: -us , -pogon c ommonly stand for masculine genera, -um for neuter and -a, -is for feminine gene ra. The first letter of the generic name is always capitalised. The name may be based on any source, but the common sources for generic names are as under: 1. Commemoration of a person commonly an author such as Bauhinia for Botanical Nomenclature 21 Bauhin, Benthamia and Benthamida for Bentham, Darwinia for Darwin, Hutchinsonia for Hutchinson, Lamarckia for Lamarck and Linnaea for Linnaeus. It may also be u sed for head of a state such as Victoria for Queen Victoria of England, Washingt onia for King George Washington, and Zinobia for Queen Zinobia of Palmyra. The n ames com-memorating a person, man or woman always take the feminine form. The na me of a genus is constructed by add-ing -ia if name of a person ends in a conson ant (Fuchsia after Fuchs), -a if it ends in a vowel (Ottoa after Otto), but -ea is added if it ends in -a (Collaea af-ter Colla). If the name ends in -er both a re permitted (Kernera for Kerner; Sesleria for Seslar). For Latinized per-sonal names ending with -us, this ter-mination is dropped before adding ap-propriate e nding (Linnaea after Linnaeus, Dillenia after Dillenius). The name may also be f ormed directly as in case of Victoria and Zinobia, as indiThe generic name of a tree, whatever be the ending, takes a feminine form, since trees are generally feminine in classical Latin. Pinus, Quercus and Prunus are, thus, all feminine genera. If two words are used to form a generic name, these have to be joined by a hyphen (generic name Uva-ursi). In case, however, the two words were combined into one word by the original author, the use of hyphen is not needed (generic name Quisqualis). The name of a genus may not coincide with a technical term currently used in morphology unless it was published before 1 J anuary 1912 and was accompanied by a specific name published in accordance with the binary system of Linnaeus. The generic name Tuber (published in 1780 was acc ompanied by a binary specific name Tuber gulosorum F. H. Wigg.) and is, therefor e, validly published. On the other hand the intended generic names Lanceolatus (Pl umstead, 1952) is, therefore, not validly published. Words such as radix , caulis , f ol ium , spina , etc., cannot now be validly published as generic names. cated above. 2. Based on a place such as Araucaria Species The name of a species is a binomial: consisting of two words, a generic name fol lowed by a specific epithet. The Code recommends that all specific epithets shou ld begin with a lower case initial letter. An upper case initial letter is somet imes used, however, for specific epithets derived from a person s name, former gen eric name or a common name. The Code discourages such usage for specific epithet s. A specific epithet may be derived from any source or composed arbitrarily. Th e following sources are commonly after Arauco a province of Chile, Caucasia for Caucasus in Russia, Salvadora for EL Salvadore, Arabis for Arabia and Sibiraea for Siberia. The name could also b e based on names of two places such as Austroamericium (Australia and America) o r place and author such as Austrobaileya (Australia and Bailey) 3. Based on an important character such as yellow wood in Zanthoxy-lum, liver-like leaves in Hepatica, marshy h abit of Hygrophila, trifoliate leaves of Trifolium, and spiny fruit of used: 1. Name of a person. The specific epithet Acanthospermum. named after a person may take genitive (possessive) or an adjectival form: 4. Aboriginal names taken directly from a language other than Latin with-out alteration of ending. Narcissus is the Greek name for daffodils named after the famous Greek god Narcissus, Ginkgo a C hinese, Vanda a Sanskrit and Sasa a Japanese aboriginal name. (i) When used in the genitive form the epithet takes its form depending on the ending of the person s name. For names end

ing in a vowel or -er the letter -i is added for a male person (roylei after Roy le, hookeri after 22 Plant Systematics Hooker), -ae for female person (laceae after Lace), and -orum for more than one persons with the same surname (hookerorum after Hooker & Hooker). If the name, h owever, ends in -a then -e is added (paulae after Paula). If the name ends in a consonant -ii is added male person (wallichii after Wallich), -iae for a female person (wilsoniae after Wilson), and -iorum for more than one persons with same surname and at least one male (verlotiorum after Verlot brothers), and -iarum if both are female (brauniarum for Braun sisters). For names of the persons alread y in Latin (e.g. Linnaeus), the Latin ending (-us in this case) has to be droppe d before adding the appropriate genitive end-ing. The specific epithets in genitive form are not related to the gender of the genus. Illustrative examples are listed in Table a. (ii) When used in adjectival form, the epithet takes its ending from the gender of the genus after adding ian if name of the person ends in a consonant, adding -an if the name ends in a vowel except when it ends in -a, wherein -n is added. Illustrative examples are given in Table b. 2. Place. The specific epithet may, similarly, be formed by using the place name as an adjective, when again the genus d etermines the ending after the addition of -ian or -ic and then the relTable a

Person Sex Specific epithet Binomial Royle roylei Impatiens roylei Hooker hookeri Iris hookeri Sengupta senguptae Euphorbia senguptae Wallich wallichii Euphorbia wallichii Todd toddiae Rosa toddiae Gepp & Gepp geppiorum Codiaeum geppiorum Linnaeus linnaei Indigofera linnaei

Table b

Author Genus Gender Specific epithet Binomial Webb Rosa Feminine webbiana Rosa webbiana Webb Delphinium Neuter webbianum Rheum webbianum Webb Astragalus Masculine webbianus Astragalus webbianus Kotschy Hieracium Neuter kotschyanum Hieracium kotschyanum Lagasca Centaurea Feminine lagascana Centaurea lagascana Table c Place Genus Gender Specific epithet Binomial Kashmir Iris Feminine kashmiriana Iris kashmiriana Delphinium Neuter

kashmirianum Delphinium kashmirianum Tragopogon Masculine kashmirianus Tragopogon kashmirianus India Rosa Feminine indica Rosa indica Solanum Neuter indicum Solanum indicum Euonymus Masculine indicus Euonymus indicus Botanical Nomenclature 23 evant gender ending as determined by the genus. The specific epithet is also for med by adding -ensis (for masculine and feminine genera, e.g. Hedera nepalensis, Rubus canadensis) or -ense (for neuter genera, e.g. Ligustrum nepalense) to the place name. Different situations are illustrated in Table c. adjectival form in species of the same genus is to be avoided, e.g. Iris hookeri and I. Hookeriana; Lysimachia hemsleyana Oliv. and L. hemsleyi Franch. 3. Character. Specific epithets based on Infraspecific taxa The names of subspecies are trinomials and are formed by adding a subspecific ep ithet to the name of a species, e.g. Angelica a character of the species are always archangelica ssp. himalaica. A variety in adjectival form and derive their (varieta) within a subspecies may accordgender from the genus. A name based ingly be quadrinomial as in Bupleurum on a w hite plant part may take the falcatum ssp. eufalcatum var. ho teri, or form alba (Rosa alba), album (Chenopo- it may just be a trinomial when no subspe dium album) or albus (Mallotus albus). cies is recognized within a species as in A common epithet used for cultivated Brassica oleracea var. capitata. A forma m ay plants may similarly take the form sativa (Oryza sativa), sativum (Allium sat ivum) or sativus (Lathyrus sativus) depending on the gender of the genus to whic h the epithet is assigned. Some epithets, however, such as bicolor (two-coloured ) and repens (creeping) remain unchanged, e.g. Ranunculus repens, also be assigned a name in a similar manner, e.g. Prunus cornuta forma villosa. The formation of the infraspecific epithet follows the same rules as the specifi c epithet. Infraspecific name may sometimes be a poly-nomial as Saxifraga aizoon var. aizoon subvar. brevifolia f. multicaulis subf. surculosa Engl. & Irmsch. Ludwigia repens and Trifolium repens. 4. Noun in apposition. A specific epithet may sometimes be a noun in apposi-tion carrying its own gender, and usu-ally in the nominative case. Binomial Pyrus malus is based on the Greek name malus for c ommon apple. In Al-lium cepa, similarly, cepa is the Latin The Type Method The names of different taxonomic groups are based on the type method, by which a certain representative of the group is the source of the name for the group. Th is representative is called the nomenclatural type or simply the type, and metho dology as typification. The type need not be the most typical member of the grou p, it only fixes the name of a particular taxon and the two are permanently asso ciated. Type may be correct name or even a synonym. Thus the tea family name (Th

eaceae) is derived from synonym Thea although the correct name for the genus is Camellia. Mimosa is the type for family Mimosaceae, but unlike most representati ves of the family that have pentamerous flowers, the genus Mimosa has tetramerou s flowers. The family Urticaceae, similarly, has Urtica as its type. When the or iginally large family was split into a number of smaller natural families, the n ame Urticaceae was retained for the group containing the genus Urtica, since the two cannot be separated. name for onion. Both the generic name and the spe-cific epithet are underlined when writ-ten or typed. When printed, they are in Italics or boldface. After the generic name in a species has been spelled out at least once, if used for other spe-cies, it may be abbreviated using the initial capital, e.g. Quercus dilatata, Q. suber, Q. Ilex, etc. A specific epithet is usually one word but when consisting of two words, these must be hy-phenated as in Capsella bursa-pastoris and Rh amnus vitis-idaea, or else the two words may be combined into one as in Narcissu s pseudonarcissus Although not leading to rejection, the use of same name in genitive form as well as 24 Plant Systematics The other splitter groups with family rank got the names Moraceae, Ulmaceae and Cannabaceae with type genera Morus, Llmus and Cannabis, respectively. The family Malvaceae has seen a lot of realignments, with Tiliaceae sometimes merged with Malvaceae. Thorne (2003) shifts Tilia to Malvaceae, but retains rest of the gene ra. This necessitates name change for former Tiliaceae (excluding genus Tilia) t o the same person. Often the collection number is also the same, differentiated as a, b, c, etc. 3. Syntype: Any one of the two or more specimens cited by the author when no holotype was designated, or any one of the two or more specimens simultaneously designated as types. Duplicate of a syntype is an isosyntype. 4. Paratype: A paratype is a specimen Grewiaceae, with Grewia as the type genus. cited in the protologue that is neither the holotype nor an isotype, nor one of the syntypes if two or more specimens were simultaneously designated as The type of a family and the higher groups is ultimately a genus, as indicated a bove. A type of a particular genus is a species, e.g. Poa pratensis for Poa. The type of name of a species or infraspecific taxon, where it ex-ists, is a single type specimen, preserved in a known herbarium and identified by the place of co llection, name of the collector and his collection number. It may also be an illustration of the plant. The Code recognizes several kinds of type, depending up on the way in which a type specimen is selected. types. 5. Lectotype: A specimen or any other elThese include: ement selected from the original ma-terial cited by the author when no holotype was originally selected or when it no longer exists. A lectotype is selected fro m isotypes or syntypes. In lectotype designation, an isotype must be chosen if s uch exists, or otherwise a syntype if such exists. If no isotype, syntype or iso syntype (duplicate of syntype) is extant, the lectotype must be chosen from amon g the paratypes if such exist. If no cited specimens ex-ist, the lectotype must be chosen from among the uncited specimens and cited and uncited illustrations w hich comprise the remaining original ma1. Holotype: A particular specimen or illustration designated by the author of the species to represent type of a spe-ci es. For the purpose of typification, a specimen is a gathering, or part of a gat hering, of a single species or infraspecific taxon made at one time, disregardin g admixtures. It may con-sist of a single plant, parts of one or several plants, or of multiple small plants. A specimen is usually mounted either on a single h

erbarium sheet or in an equivalent preparation, such as a box, packet, jar or mi croscope slide. Type specimens of names of taxa must be preserved permanently an d may not be living plants or cultures. However, cultures of fungi and algae, if preserved in a metabolically inactive state (e.g. by lyophilization or deep-fre ezing), are acceptable as types. It is now essen-tial to designate a holotype wh en pubterial, if such exist. 6. Neotype: A specimen or illustration selected to serve as nomenclatural type as long as all of the material on which th e name of the taxon was based is missing; a specimen or an illustration selected when no holotype, isotype, paratype or syntype exists. 7. Epitype: A specimen or illustration selishing a new species. lected to serve as an interpretative type when the holotype, lectotype or previo usly designated neotype, or all original material associated with a validly publ ished name, is demonstrably ambiguous and cannot be critically identified for pu rposes of the precise application of the name of a taxon. When an epitype is des ignated, the holotype, 2. Isotype: A specimen which is a duplicate of the holotype, collected from the same place, at the same time and by Botanical Nomenclature 25 lectotype or neotype that the epitype supports must be explicitly cited. G. Bentham, Hook. for William Hooker, Hook.f. for Sir J. D. Hooker (f. stands fo r filius, the son; J. D. Hooker was son of Will-iam Hooker), R.Br. for Robert Br own, Lam. for J. P. Lamarck, DC. for A. P. de Candolle, Wall. for Wallich, A. DC . for Alphonse de Candolle, Scop. for G. A. Scopoli and Pers. for C. H. Persoon. In most cases in which no holotype was designated there will also be no paratypes, since all the cited specimens will b e syntypes. However, when an author has des-ignated two or more specimens as typ es, any remaining cited specimens are paratypes and not syntypes. Single author Topotype is often the name given to a specimen collected from the same locality from which the holotype was originally colThe name of a single author follows the name of a species (or any other taxon) w hen a single author proposed a new name, e.g. Solanum nigrum L. lected. In cases where the type of a name is a culture permanently preserved in a metabo lically inactive state, any living isolates obtained from that should be referre d to as Multiple authors ex-type (ex typo), ex-holotype (ex holotypo), ex-isotype (ex isotypo), etc., in order to make The names of two or more authors may be associated with a name for a variety of reasons. These different situations are exhibited by citing the name of the auth ors differently: it clear they are derived from the type but are not themselves the nomenclatural type. When an infraspecific variant is recog-nized within a species for the first time , it automatically establishes two infraspecific taxa. The one, which includes t he type specimen of the species, must have the same epithet as that of the speci es, e.g. Aca-cia nilotica ssp. nilotica. Such a name is called an autonym, and t he specimen an autotype. The variant taxon would have its own holotype and is di fferentiated by an epi-thet different from the specific epithet, e.g. 1. Use of et: When two or more authors publish a new species or propose a new name, their names are linked by et, e.g. Delphinium viscosum Hook.f. et Thomson. 2. Use of parentheses: The rules of bo-

Acacia nilotica ssp. indica. tanical nomenclature specify that whenever the name of a taxon is changed by the transfer from one genus to another, or by upgrading or downgrading the level of the taxon, the original epithet should be retained. The name of the taxon provi ding the epithet is termed a basionym. The name of the original author or author s whose epithet is being used in the changed name is placed within parentheses, and the author or authors who made the name change outside the parentheses, e.g. Cynodon dactylon (Linn.) Pers., based on the basionym Panicum dactylon Linn., t he original name for the species. It must be borne in mind that the applica-tion of the type method or typificatio n is a methodology different from typology, which is a concept based on the idea that does not recognize variation within the taxa, and be-lieves that an ideali zed specimen or pattern can represent a natural taxon. This concept of typology was very much in vogue before Dar-win put forward his ideas about variations. Author Citation For a name to be complete, accurate and readily verifiable, it should be accompa nied by the name of the author or authors who first published the name validly. The names of the authors are commonly abbreviated, e.g. L. for Carolus Linnaeus, Benth. for 3. Use of ex: The names of two authors are linked by ex when the first author had proposed a name but was validly publi shed only by the second author, the first author failing to satisfy all or some of the requirements of the Code, e.g. Cerasus cornuta Wall. ex Royle. 26 Plant Systematics 4. Use of in: The names of authors are 1. sp. nov. for species nova, a species new linked using in when the first author published a new species or a name in a pub lication of another author, e.g. Carex kashmirensis Clarke in Hook.f. Clarke pub lished this new species in the Flora of British India whose author to science; 1ragopogon kashmirianus G. Singh, sp. nov. (published in 1976). 2. comb. nov. for combinatio nova, a was Sir J. D. Hooker. name change involving the epithet of the basionym, name of the original au-thor being kept within parentheses; Vallisneria natans (Lour.) Hara comb. nov. (publi shed in 1974 based on 5. Use of emend: The names of two authors are linked using emend. (emendavit: person making the correc-tion) when th e second author makes some change in the diagnosis or in cir-cumscription of a t axon without alter-ing the type, e.g. Phyllanthus Linn. Physkium natans Lour., 1790). 3. comb. et stat. nov. for combinatio et emend. Mull. status nova, when a new combination also involves the change of status. Epithet of the basionym will accordingly be used in the combination intended; Caragana o pulens Kom. var. licentiana (Hand.-Mazz.) Yakovl. comb. et stat. nov. (published in 1988 based on C. licentiana Hand.-Mazz., 1933; new combination also involved change of status from a species C. licentiana to a variety of Caragana opulens Kom.). 6. Use of square brackets: Square brackets are used to indicate prestarting point author. The generic name Lupinus was effectively published by Tournefort in 1719, but as it hap-pens to be earli er than 1753, the start-ing date for botanical nomenclature based on Species pla ntarum of Linnaeus, the appropriate citation for the genus is Lupinus [Tourne.] L. 4. nom. nov. for nomen novum, when the When naming an infraspecific taxon, the original name is replaced and its epithet cannot be used in the new name; Myrcia lucida McVaugh nom. nov. (pub-lished in 1969 to replace M. laevis O. Berg, 1862

, an illegitimate homonym authority is cited both for the specific epithet and the infraspecific epithet, e.g. Acacia nilotica (L.) Del. ssp. indica (Benth.) Brenan. In the case of an au tonym, how-ever, the infraspecific epithet does not bear the author s name since i t is based on the same type as the species, e.g. Acacia nilotica (L.) Del. ssp. nilotica. of M. laevis G. Don, 1832). Publication of Names These abbreviations are, however, used only when first published. In future refe r-ences, these are replaced by the name of the publication, page number and the year of publication for full citation, or at least the year of publication. Thus when first published in 1976 as a new species, 1ragopogon kashmirianus G. Singh sp. nov. appeared in a book titled Forest Flora of Srinagar on page 123, figure 4, any successive reference to this species would appear as: 1ragopogon kashmir ianus G. Singh, Forest Flora of Srinagar, p 123, f. 4, 1976 or 1ragopogon kashmi rianus G. Singh, 1976. The other names would be cited as Vallisneria natans (Lou r.) Hara, 1974, Caragana opulens Kom. var. licentiana (Hand.-Mazz.) Yakovl., 198 8 and Myrcia lucida McVaugh, 1969, specify-ing the year of publication. A new co mbination, or an avowed substitute (replacement The name of a taxon, when first published, should meet certain requirements so a s to become a legitimate name for consideration when deciding on a correct name. A valid publication should satisfy the following requirements: Formulation A name should be properly formulated and its nature indicated by a proper abbrev iation after the name of the author: Botanical Nomenclature 27 name, nomen novum), published on or after 1 January 1953 based on a previously a nd validly published name is not validly pub-lished unless its basionym (name-br inging or epithet-bringing synonym) or the replaced synonym (when a new name is proposed) is clearly indicated and a full and direct refer-ence given to its aut hor and place of valid publication, with page or plate reference and date. Autho rs should cite themselves by name after each new name they publish rather than r efer to themselves by expres-sions such as nobis (nob.) or mihi (m.). tion, the herbarium or institution in which the type is conserved must be specif ied. Names published on or after 1 January 2007 would require a specimen (and no t a mere illustration) as type, except only for micro-scopic algae or microfungi for which preservation of a type was technically difficult, and where illustrat ion is accepted as type. On or after 1 January 2001, lectotypification or neotyp ification of a name of a species or in-fraspecific taxon is not affected unless indicated by use of the term lectotypus or neotypus , its abbreviation, or its equiva lent in a modern language. The specimen selected as type should belong to a sing le gathering. Echinocereus sanpedroensis (Raudonat & Rischer, 1995) was based on a holotype con-sisting of a complete plant with roots, a de-tached branch, an entir e flower, a flower cut in halves, and two fruits, which according to the label w ere taken from the same cultivated individual at different times and preserved, in alcohol, in a single jar. This material be-longs to more than one gathering a nd can-not be accepted as a type. Raudonat & Rischer s name is thus not validly pu blished. Latin diagnosis Names of all new species (or other taxa new to science) published 1 January 1935 on-wards should have a Latin diagnosis (Latin translation of diagnostic feature s). Full de- scription of the species in any language can accompany the Latin di agnosis. A descrip-tion in any language, not accompanied by a Latin diagnosis is allowed for publications before 1 January 1935. For publications be-fore 1 Janu ary 1908, an illustration with analysis without any accompanying description is valid. Thus description in any lan-guage is essential from 1 January 1908 onward s and this accompanied by a Latin di-agnosis from 1 January 1935. For name chang es or new names of already known spe-cies, a full reference to the original publ ica-tion should be made. Typification

A holotype should be designated. Publica-tion on or after 1 January 1958 of the name of a new taxon of the rank of genus or below is valid only when the type of the name is indicated. For the name of a new taxon of the rank of genus or belo w published on or after 1 January 1990, an indication of the type must include o ne of the words typus or holotypus , or its abbreviation, or even its equivalent in a modern language. For the name of a new species or infraspecific taxon published on or after 1 January 1990 whose type is a specimen or unpublished illustraEffective publication The publication becomes effective by distribution in printed form, through sale, exchange or gift to the general public or at least the botanical institutions w ith libraries accessible to botanists generally. It is not affected by communica tion of new names at a public meeting, by the placing of names in collections or gardens open to the public; by the issue of microfilm made from manuscripts, ty pescripts or other unpublished ma-terial, by publication on-line, or by dissemin ation of distributable electronic media. The publication in newspapers and catal ogues (1 January 1953 onwards) and seed exchange lists (1 January 1977 onwards) is not an effective publication. A theses submitted for a higher degree on or af ter 1 January, 1953 is considered effectively published, only if it car-ries a s tatement of its publication or an in-ternal evidence (e.g. an ISBN, or a commerc ial publisher). Publication of handwritten 28 Plant Systematics material, reproduced by some mechanical or graphic process (indelible autograph) such as lithography, offset, or metallic etch-ing before 1 January 1953 is effe ctive. Salvia oxyodon Webb & Heldr. was effectively pub-lished in an indelible a utograph catalogue algae, but Vienna Code extended this provision also to organisms subsequently re cognized as fungi. The provision has benefitted the recognition of Microsporidia , long considered protozoa and now recognized as fungi. Similarly the species of Pneumocystis, not validly published because of lack of Latin placed on sale (Webb & Heldreich, Catalogus plantarum hispanicarum ... ab A. Blanco diagnosis or description, are now treate d as lectarum, Paris, Jul 1850, folio). The Journal validly published, since Lat in requirement of the International Conifer Preservation Soci- is not mandatory under Zoologica l Code, ety, Vol. 5[1]. 1997 ( 1998 ), consists of du- which originally covered thes e mammalian plicated sheets of typewritten text with hand- pathogens, now treate d as fungi. written additions and corrections in several places. The handwritten portions, b eing in-delible autograph published after 1 January 1953, are not effectively pu blished. Intended new combinations ( Abies koreana var. yuanbaoshanensis , p. 53), f or which the basionym reference is handwritten are not validly published. The en tirely handwritten account of a new taxon (p. 61: name, Latin description, state ment of type) is treated as The Tokyo Code included a rule (Art. 32. 1-2), subject to ratification by the XV I International Botanical Congress (St Louis, 1999) according to which new names of plants and fungi would have to be registered in order to be validly publishe d after the 1st of January 2000. A trial registration had already begun, on a no n-mandatory basis, for a two-year period starting 1 January 1998. The proposal w as, however, voted out at St. Louis and all references to the registration delet ed from unpublished. The date of a name is that of its valid pub-lication. When the various condition s for valid publication are not simultaneously ful-filled, the date is that on w hich the last con-dition was fulfilled. However, the name must always be explici tly accepted in the place of its validation. A name published on or after 1 Janu ary 1973 for which the various con-ditions for valid publication are not simulta -neously fulfilled is not validly published un-less a full and direct reference is given to the places where these requirements were the Code.

A correction of the original spelling of a name does not affect its date of vali d publication. Rejection of Names previously fulfilled. The process of selection of correct name for a taxon involves the identification of illegitimate names, those which do not satisfy the rules of botanical nomenc lature. A legitimate name must not be rejected merely because it, or its epithet , is inappropriate or disagree-able, or because another is preferable or better known or because it has lost its original meaning. The name Scilla peruviana L. (1753) is not to be rejected merely because the species does not grow in Peru. A ny one or more of the following situations leads to the rejecIn order to be accepted, a name of a new taxon of fossil plants published on or after 1 January 1996 must be accompanied by a Latin or English description or di agnosis or by a reference to a previously and effectively published Latin or English descri ption or diagnosis. tion of a name: For groups originally not covered by ICBN, the Code accepts them as validly publ ished if they meet the requirements of the perti-nent non-botanical Code, but ar e now recognized as organisms covered under bo-tanical Code. This provision earl ier covered organisms subsequently recognized as 1. Nomen nudum (abbreviated nom. nud.): A name with no accompanying description. Many names published by Wallich in his Catalogue (abbreviated Wall. Cat.) published in 1812 were nomen nudum. Th ese were either validated by another author at a later date Botanical Nomenclature 29 by providing a description (e.g. Cerasus cornuta Wall. ex Royle) or if by that t ime the name has already been used for another species by some other author, the nomen nudum even if validated is rejected and a new name has to be found (e.g. Quercus dilatata Wall., a nom. nud. rejected and replaced by Q. Archangeli 1882 for almond became a later homonym of Prunus communis Huds., 1762 which is a species of plums. P. communis (L.) Archangeli was as such replaced b y P. dulcis (Mill.) Webb, 1967 as the name for almonds. When two or more generic or specific names based on different types are so similar that they are likely to be confused (because they are applied to related taxa or for any other reason ) they are to be treated as homonyms. Names treated as homonyms include: Asteros temma Decne. (1838) and Astrostemma himalayana Bahadur, 1972). 2. Name not effectively published, not properly formulated, lacking typifica-tion or without a Latin diagnosis. 3. Tautonym: Whereas the Zoological Code allows binomials with identical generic name and specific epithet (e.g. Bis on bison), such names in Botanical nomenclature constitute tautonyms (e.g. Malus malus) and are rejected. The words in the tautonym are exactly iden-tical, and evidently names such as Cajanus cajan or Sesbania sesban are not tautonyms and t hus legitimate. Repetition of a specific epithet in an infraspecific epithet doe s not consti-tute a tautonym but a legitimate autonym (e.g. Acacia nilotica ssp. Benth. (1880); nilotica). Pleuropetalum Hook. f. (1846) and Pleuripetalum T. Durand (1888); Eschweilera DC . (1828) and Eschweileria Boerl. (1887); Skytanthus Meyen (1834) and Scytanthus Hook. (1844). The three generic names Bradlea Adans. (1763), Bradleja Banks ex G aertn. (1790), and Braddleya Vell. (1827), all commemorating Richard Bradley, ar e treated as homonyms because only one can be used without serious risk of confu sion. The follow-ing specific epithets under the same genus would also form homo nyms chinensis and sinensis; ceylanica and zeylanica; napaulensis, nepalensis, a nd 4. Later homonym: Just as a taxon

should have one correct name, the Code similarly does not allow the same name to be used for two different spe-cies (or taxa). Such, if existing, consti-tute ho monyms. The one published at an earlier date is termed the earlier homonym and t hat at a later date as the later homonym. The Code rejects later homonyms even i f the earlier homonym is illegitimate. Ziziphus jujuba Lam., 1789 had long been used as the correct name for the cultivated fruit jujube. This, however, was asc er-tained to be a later homonym of a re-lated species Z. jujuba Mill., 1768. The binomial Z. jujuba Lam., 1789 is thus rejected and jujube correctly named as Z. mauritiana Lam., 1789. Similarly, although the earliest name for al-monds is Am ygdalus communis L., 1753 when transferred to the genus Prunus the name Prunus c ommunis (L.) nipalensis. 5. Later isonym: When the same name, based on the same type, has been pub-lished independently at different times by different authors, then only the ear-liest of these so-called isonyms has nomencla tural status. The name is always to be cited from its original place of valid pu blication, and later isonyms may be disregarded. Baker (1892) and Christensen (190 5) independently published the name Alsophila kalbreyeri as a substitute for A. podophylla Baker (1891) non Hook. (1857). As published by Christensen, Alsophila kalbreyeri is a later isonym of A. kalbreyeri Baker, without nomenclatural status. 6. Nomen superfluum (abbreviated as nom. superfl.): A name is illegitimate 30 Plant Systematics and must be rejected when it was nomenclaturally superfluous when published, i.e ., if the taxon to which it was applied as circumscribed by its author included the type of a name or epithet which ought to have been adopted under the rules. Phys kium natans Lour., 1790 thus when trans-ferred to the genus Vallisneria, the epi -thet natans should have been retained but de Jussieu used the name Vallisneria physkium Juss., 1826 a name which becomes superfluous. The species has according ly been named correctly as Vallisneria natans (Lour.) Hara, 1974. A combination based on a superfluous name is also illegitimate. Picea excelsa (Lam.) Link is i llegitimate since it is based on a superfluous name Pinus excelsa Lam., 1778 for Pinus abies Linn., 1753. The legitimate combina-tion under Picea is thus Picea abies Rhinanthus crista-galli to a group of several varieties, which he later describe d under separate names, rejecting the name R. crista-galli L. Several later auth ors, however, continued to use this name for diverse occasions until Schwarz (19 39) finally listed this as Nomen dubium, and the name was finally rejected. 10. Name based on monstrosity: A name must be rejected if it is based on a monstrosity. The generic name Uropedium Lin dl., 1846 was based on a monstrosity of the species now referred to as Phragmidi um caudatum (Lindl.) Royle, 1896. The generic name Uropedium Lindl. must, theref ore, be rejected. The name Ornithogallum fragiferum Vill., 1787, is likewise, ba sed on a monstrosity and thus should be rejected. (Linn.) Karst., 1880. Principle of Priority 7. Nomen ambiguum (abbreviated as The principle of priority is concerned with the selection of a single correct na me for a taxonomic group. After identifying legitimate and illegitimate names, a nd rejecting the latter, a correct name has to be selected from among the legiti mate ones. If more than one legitimate names are available for a taxon, the corr ect name is the earliest legitimate name in the same rank. For taxa at the speci es level and below the correct name is eitherthe earliest legitimate name or a c ombination based on the earliest legitimate basionym, unless the combination bec omes a tautonym or later homonym, rendering it illegitimate. The following examp les illustrate the principle of priority: nom. ): A name is rejected if it is usedinn a different sense differ-

ent authors and has become aa source of persistent error. The name Rosa villosa L. is rejected because it has been applied to several different species and has become a source of error. 8. Nomen confusum (abbreviated as nom. confus.): A name is rejected if it is based on a type consisting of two or more entirely discordant elements, so that it is difficult to select a satisfac-tory lectotype. The characters of the genus Actinotinus, for example, were derived fr om two genera Viburnum and Aesculus, owing to the insertion of the inflorescence of Viburnum in the ter-minal bud of an Aesculus by a collector. The name Actino tinus must, therefore, 1. The three commonly known binomials be abandoned. for the same species of Nymphaea are N. nouchali Burm.f., 1768, N. acutiloba DC. , 1824, N. stellata Willd., 1799 and N. malabarica Poir., 1798. Using the priori ty criterion, N. nouchali Burm.f. is selected as the correct name as it bears th e earliest date of publication. The other three names are regarded as synonyms. The citation is written as: 9. Nomen dubium (abbreviated as nom. dub.): A name is rejected if it is dubious, i.e. it is of uncertain application because it is impossible to establish the taxon to which it should be referred. Linnaeus (1753) attributed the name Botanical Nomenclature 31 Nymphaea nouchali Burm.f., 1768 N. malabarica Poir., 1798 N. stellata Willd., 17 99 Vallisneria physkium are based on the same type as the correct name V. natans an d are thus known as nomenclatural synonyms or homotypic synonyms. These three wo uld remain together in all citations. The other two names V. gigantea and V. asi atica are based on separate types and may or may not be regarded as synonyms of V. natans depending on taxonomic judgement. Such a synonym, which is based on a type different from the correct name, is known as a taxonomic synonym or heterot ypic synonym. V. spiralis auct. (auctorumauthors) is misplaced identification of Asian specimens with V. Spiralis L. N. acutiloba DC., 1824 The following binomials for common maize plant exist: Zea mays Linn., 1753, Z. c uragua Molina, 1782, Z. indurata Sturtev., 1885 and Z. japonica Von Houtte, 1867 . Zea mays being the ear-liest validly published binomial is cho-sen as correct name, and others cited as its synonyms as under: Zea mays L., 1753 Z. curagua Molina, 1782 Z. japonica Von Houtte, 1867 Z. indurata Sturtev., 1885 2. Loureiro described a species under the name Physkium natans in 1790. It was subsequently transferred to the genus Valli sneria by A. L. de Jussieu in 1826, but unfortunately, he ignored the epithet na tans and instead used a binomial Vallisneria physkium, a super-fluous name. Two Asiatic species with independent typification were de-scribed subsequently under the names V. gigantea Graebner, 1912 and V. asiatica Miki, 1934. Hara on mak-in g a detailed study of Asiatic speci-mens concluded that all these names are syno nymous, and also that V. spiralis Linn. with which most of the Asiatic specimens were identified does not grow in Asia. As no legitimate com-bination based on P hyskium natans Lour. existed, he made one V. natans (Lour.) Hara in 1974. The synony my 3. The common apple was first described by Linnaeus under the name Pyrus malus in 1753. The species was sub-sequently tr ansferred to the genus Malus but the combination Malus malus (Linn.) Britt., 188 8 cannot be taken as the correct name since it becomes a tautonym. The other bin omial under Malus available for apple is M. domestica Borkh, 1803 which is accep ted as correct name and citation written as: Malus domestica Borkh Pyrus malus Linn., 1753 M. malus (Linn.) Britt., 1888

Tautonym M. pumila auct. (non Mill.) M. communis Desf., 1798 Nom. superfl. M. communis Desf. is based on same would be cited as under: type as Pyrus malus, and is as such a nomen superfluum. Apple has been assigned by some authors to M. pumila Mill., 1768, which however is small Vallisneria natans (Lour.) Hara, 1974 Physkium natans Lour.,1790 Basionym fruited Paradise apple. V. physkium Juss., 1826 nom. 4. Almond was first described by Linnaeus superfl. under the name Amygdalus communis in 1753. Miller described another species unde r the name A. dulcis in 1768. The two are now regarded as synonymous. The genus Amygdalus was sub-sequently merged with the genus Prunus and the combination Pru nus communis (L.) Archangeli made in 1882 V. gigantea Graebner, 1912 V. asiatica Miki, 1934 V. spiralis auct. (non L., 1753) The correct name of the species in this case, is the most recent name, but it is based on the earliest basionym. It must be noted that Physkium natans and 32 Plant Systematics based on the earlier name Amygdalus 1-5-1753. The starting dates for different communis Linn. It was discovered by Webb that the binomial Prunus commugroups include: Seed plants, Pteridophytes, Sphagnaceae nis had already been used by Hudson in 1762 for a different species render-ing P . communis (L.) Archangeli a later homonym which had to be conse-quently rejecte d. Webb accordingly used the next available basionym Amygdalus dulcis Mill., 176 8 and made a combination Prunus dulcis (Mill.) Webb, 1967 as the correct name fo r almond. Another binomial, Prunus amygdalus Batsch, 1801, cannot be taken up as it ignores the earlier epi-thets. The citation for almond would 1-5-1753 and lichens Hepaticae, most Algae, slime moulds Mosses (excluding Sphagnaceae) 1-1-1801 Fungi 31-12-1801 Fossils 31-12-1820 Algae (Nostocaceae) 1-1-1886 Algae (Oedogoniaceae) 1-1-1900 The publications before these dates for re-spective groups are ignored while dec iding the priority.

thus be: Starting date for suprageneric names was set at Vienna Congress as 4 August, 178 9, the date of publication of de Jussieu s Genera Plantarum .Double author citatio n is not justified or permitted at suprageneric ranks. Prunus dulcis (Mill.) Webb, 1967 Amygdalus dulcis Mill., 1768 basionym A. communis L., 1753 Not above family rank The principle of priority is applicable only up to the family rank, and not abov e. P. communis (L.) Arch., 1882 (non Huds., 1762) P. amygdalus Batsch, 1801 When two or more names simultaneously published are united, the first author to make such a merger has the right of choos-ing the correct name from these. Brown , 1818 was the first to unite Waltheria americana L., 1753 and W. indica L., 175 3. He adopted the name W. indica for the com-bined species, and this name is acc ordingly treated as having priority over W. americana. The generic names Thea L. and Camellia L. are treated as having been published simul-taneously on 1 May 1 753. The combined ge-nus bears the name Camellia, since Sweet, 1818, who was the first to unite the two genera, chose that name, and cited Thea as a synonym. Not outside the rank In choosing a correct name for a taxon, names or epithets available at that rank need to be considered. Only when a correct name at that rank is not available, can a combination be made using the epithet from another rank. Thus at the level of section the correct name is Campanula sect. Campanopsis R. Br., 1810 but whe n upgraded as a genus the correct name is Wahlenbergia Roth, 1821 and not Campan opsis (R. Br.) Kuntze, 1891. The following names are synonyms: Lespedza eriocarpa DC. var. falconeri Prain, 1897 L. meeboldii Schindler, 1911 Limitations to the principle of priority Nair, 1977 (Prain) Campylotropis eriocarpa var. falconeri C. meeboldii (Schindler) . Schindler, 1912 Application of the principle of priority has the following limitations: The correct name at the species level under the genus Campylotropis would be C. meeboldii, ignoring the earlier epithet at the varietal level. If treated as a v ariety ,the correct name would be C. eriocarpa var. falconeri, based on the earl iest epithet at that rank. Under the genus Lespedza, at the species Starting dates The principle of priority starts with the Spe-cies plantarum of Linnaeus publish ed on Botanical Nomenclature 33 level the correct name would be L. meeboldii, whereas at the varietal level, it would be same date of publication are united, the au-thor who unites them first has the c hoice of selecting the correct binomial. For a long time it was known that the f irst persons to unite the two species were Fiori and Paoletti in 1896 who select ed T. aestivum L. as the correct name. It was pointed out by Kergulen (1980), how ever, that the first person to unite these two species was actually Mrat (1821) a nd he had selected T. hybernum L. and not T. aestivum. This discovery led to the danger of T. aestivum L. being dropped in favour of T. hybernum L. A proposal f

or conserving the name T. aestivum L. was thus made by Hanelt and Schultze-Motel (1983), and being the numL. eriocarpa var. falconeri. Magnolia virginiana var. foetida L., 1753 when raised to specific rank is called M. gran-diflora L., 1759, not M. foetida (L.) Sarg., 1889. Nomina Conservanda Nomina conservanda (abbreviated as nom. cons.): Strict application of the princi ple of priority has resulted in numerous name changes. To avoid name changes of well-known families or genera especially those containing many species a list of con served generic and family names has been prepared and published in the Code with relevant changes. Such nomina conservanda are to be used as correct names repla cing the ear-lier legitimate names, which are rejected and constitute nomina rej icienda (abbrevi-ated nom. rejic.). The family name Theaceae D. Don, 1825 is thu s conserved against Ternstroemiaceae Mirbe, 1813. The genus Sesbania Scop., 1777 is conserved against Sesban Adans., 1763 and Agati Adans., 1763. ber one economic plant this was accepted at the Berlin Congress, removing any fu rther danger to the name Triticum aestivum L. Conservation of names of species In 1768 P. Miller proposed a new name, Lycopersicon esculentum for tomato, a spe cies described earlier by Linnaeus (1753) as Solanum lycopersicum. Karsten (1882 ) made the name change Lycopersicum lycopersicum (L.) Karst., retaining the epit het used by Linnaeus, but since the name became a tautonym it was not considered the correct name for tomato. Nicolson (1974) suggested an orthographic correcti on Lycopersicon lycopersicum (L.) Karst., suggesting that Lycopersicon and lycop ersicum are orthographic variants. Since the name Lycopersicon lycopersicum was no longer a tautonym, it was accepted as the correct name. But since Lycopersico n esculentum Mill., 1768 was a more widely known name, a proposal for its conser vation was made by Terrel (1983) and accepted at the Berlin Con-gress along with that of Triticum aestivum L. A list from a mere 5 in Tokyo Code has grown to ne arly 60 for Spermatophyta alone. Names listed in this Appendix fall under the sp ecial provisions of Art. 14.4. Neither a rejected name, nor any combination base d on a rejected name may be used for a taxon that includes the type of the corre sponding conserved name (Art. 14.7; see also Art. 14 Note 2). Combinations based on a conserved name are, therefore, in effect, similarly conserved. Given below are the major examples of In spite of several protests from agricultural botanists and horticulturists, wh o were dis-gusted with frequent name changes due to the strict application of th e principle of pri-ority, taxonomists for a long period did not agree upon conse rving names at the species level. The mounting pressure and the dis-covery that Triticum aestivum was not the cor-rect name of common wheat, compelled tax-onomi sts to agree at the Sydney Congress in 1981 upon the provision to conserve names of species of major economic impor-tance. As a result, Triticum aestivum Linn. was the first species name conserved at Ber-lin Congress in 1987 and published i n sub-sequent Code in 1988. Another species name also conserved along with was L ycopersicon esculentum Mill. Linnaeus described two species, Triticum aestivum and T. hybernum in his Species plantarum, both bearing the same date of publication in 1753. According to the rules of nomenclature when two species with the 34 Plant Systematics species names which have been declared nomina conservanda (each name followed by the (=) sign, indicating taxonomic synonym or a (= =) sign, indicating nomencla tural syn-onym and then the binomial against which it has been conserved). Some names listed as conserved have no corresponding nomina rejicienda because they w ere conserved solely to maintain a particular type: cated by including the sexual symbols ( : female; : male) in the formula, or by placing the female parent first. If a non-alphabetical sequence is used, its bas is should

be clearly indicated. Allium ampeloprasum L., 1753 A hybrid may either be interspecific (be-tween two species belonging to the same genus) or intergeneric (between two species belonging to two different genera). A binary name may be given to the interspecific hybrid or nothospecies (if it i s self-perpetuating and/or reproductively isolated) by placing the cross sign (i f mathematical sign is available it should be placed immediately before the spec ific epithet, otherwise x in lower case may be used with a gap) before the specifi c epithet as in the following cases (hybrid formula may be added within the (=) Allium porrum L., 1753 Amaryllis belladonna L. Bombax ceiba L. Carex filicina Nees, 1834 (=) Cyperus caricinus D. Don, 1825 Hedysarum cornutum L., 1763 (= =) Hedysarum spinosum L., 1759 parentheses if the parents are established): Lycopersicon esculentum Mill., 1768 1. Salix x capreola (S. aurita x S. caprea) (= =) Lycopersicon ly (L.) or Salix xcapreola (S. aurita x S. caprea) H.Kaarrsseenn,, 1882 2 Rosa x odorata (R. chinensis x R. Magnolia kobus DC., 1817 gigantea) or Rosa xodorata (R. chinensis x R. gigantea) Silene gallica L., 1753 The variants of interspecific hybrids are and nothovarieties, (=) Silene anglica L., 1753 (=) Silene lusitanica L., 1753 (=) Silene quinquevulnera L., 1753 e.g. Salix rubens nothovar. basfordiana. For an intergeneric hybrid, if given a disTriticum aestivum L., 1753 (=) Triticum hybernum L.,1753. tinct generic name, the name is formed as a condensed formula by using the first part (or whole) of one parental genus and last part (or whole) of another genus (but not the whole of both genera). A cross sign is placed before the generic n ame of the hybrid, e.g. xTriticosecale (or x Triticosecale) from Triticum and Se cale, xPyronia (or x Pyronia) from Pyrus and Cydonia, and Agropogon from Agrosti s and Polypogon. The names may be written as under: Names of Hybrids Hybridity is indicated by the use of the mul-tiplication sign, or by the additio n of the pre-fix notho- to the term denoting the rank of the taxon, the principal ranks being nothogenus and nothospecies. A hybrid between named taxa may be indi cated by placing the multiplication sign between the names of the taxa; the whol e expression is then called a hybrid formula: 1. xTriticosecale (Triticum x Secale) 2. xPyronia (Pyrus x Cydonia) 1. Agrostis Polypogon The nothogeneric name of an intergeneric hybrid derived from four or more genera is formed from the name of a person to which is added the termination -ara; no such name may exceed eight syllables. Such a name is regarded as a condensed for mula: 2. Agrostis stolonifera Polypogon monspeliensis 3. Salix aurita S. caprea It is usually preferable to place the names or epithets in a formula in alphabet ical or-der. The direction of a cross may be indixPotinara (Brassavola x Cattleya x Laelia x Sophronitis) Botanical Nomenclature 35

The nothogeneric name of a trigeneric hy-brid is either: (a) a condensed formula in which the three names adopted for the pa-rental genera are combined into a s ingle word not exceeding eight syllables, using the whole or first part of one, followed by the whole or any part of another, followed by the whole or last part of the third (but not the whole of all three) and, optionally, one or two conne cting vowels; or (b) a name formed like that of a nothogenus derived from four o r more genera, i.e., from a personal name to which is added the termination -ara : tal letter, and is not a Latin but rather a common name. It is either preceded b y cv. as in Rosaf loribunda cv. Blessings or simply within single quotation mark s, e.g. Rosa floribunda Blessings . Cultivars may also be named directly under a ge nus (e.g. Hosta Decorata ), under a hybrid (e.g. Rosa x paulii Rosea ) or directly und er a common name (e.g. Hybrid Tea Rose Red Lion ). The correct nothogeneric name fo r plants derived from the Triticum x Secale crosses is x Triticosecale Wittmack ex A. Camus. As no correct name at the species level is available for the common crop triticales, it is recommended that crop triticales be named by appending t he cultivar name to the nothogeneric name, e.g. x Triticosecale Newton . Since 1 Ja nuary 1959 new cultivar names should have a description published in any languag e and these names must not be the same as the botanical or common name of a genu s or a species. Thus, cultivar names Rose , Onion , etc., are not permitted as the nam e of a cultivar. It is recommended that cultivar names be registered with proper registering authorities to prevent duplication or misuse of cultivar names. Reg istering authorities exist separately for roses, orchids and several other group s or genera. xSophrolaeliocattleya (Sophronitis x Laelia x Cattleya) When a nothogeneric name is formed from the name of a person by adding the termination -ara, that person should pre ferably be a collector, grower, or student of the group. A binomial for the intergeneric hybrid may similarly be written as under: xAgropogon lutosus (Agrostis stolonifera x Polypogon monspeliensis) It is important to note that a binomial for an interspecific hybrid has a cross before the specific epithet, whereas in an intergeneric hybrid, it is before the generic name. Since the names of nothogenera and nothotaxa with the rank of a s ubdivision of a genus are condensed formulae or treated as such, they do not have types. UNIFIED BIOLOGICAL NOMENCLATURE Since the name of a nothotaxon at the rank of species or below has a type, state -ments of parentage play a secondary part in Biology as a science is unusual in the sense that the objects of its study can b e named according to five different Codes of nomenclature: International Code of Zoological Nomenclature (ICZN) for animals, International Code of Botanical Nom enclature (ICBN) for plants, International Code for the Nomenclature of Bacteria (ICNB) now called Bacteriological Code (BC) for bacteria, International Code of Nomenclature for Cultivated Plants (ICNCP) for plants under cultivation, and In ternational Code of Virus Classification and Nomenclature (ICVCN) for viruses. F or the general user of scientific names of organisms, there is thus inherent con fusion in many aspects of this situation: different sets of rules have different condetermining the application of the name. The grafts between two species are indi-cated by a plus sign between two grafted spe-cies as, for example, Rosa webbiana + R. flo-ribunda. Names of Cultivated Plants The names of cultivated plants are governed by the International Code of Nomencl ature for Cultivated Plants (ICNCP), last published in 1995 (Trehane et al.). Mo st of the rules are taken from ICBN with additional recog-nition of a rank culti var (abbreviated cv.) for cultivated varieties. The name of a cultivar is not wr

itten in Italics, it starts with a capi36 Plant Systematics ventions for citing names, provide for differ-ent forms for names at the same ra nk, and, although primarily each is based on priority of publication, differ som ewhat in how they Salient Features Largely on the pattern of the Botanical Code the salient features of this Draft BioCode in-clude: determine the choice of the correct name. This diversity of Codes can also create more serious problems as, for example, i n the determination of which Code to follow for those organisms that are not cle arly plants, animals or bacteria, the so-called ambiregnal organisms, or those w hose current genetic affinity may be well established but whose traditional trea tment has been in a different group (e.g. the cyanobacteria). Moreover, the deve lopment of electronic information re-trieval, by often using scientific names wi th-out clear taxonomic context, accentuates the problem of divergent methods of citation and makes homonymy between, for example, plants and animals a source of trouble and frequently confusion. BioCode and PhyloCode are two efforts towards a unified code, the former retaining the ranked hier-archy of Linnaean system, whereas the lat-ter developing a rankless system based on the concepts of phylog enetic systematics. 1. General points: No examples are listed, Notes omitted at the present stage, al-though some will no doubt be needed. A co nsiderable number of articles and paragraphs have been dropped; the Draft BioCod e has only 41 Articles, whereas the St. Louis Code has 62. 2. Taxa and Ranks: The present ranks of the Botanical Code are maintained in the Draft BioCode, and a few tentatively added: domain (above kingdom), in use for the pro-/eukaryotes, superfamily (in widespread use in zoology), and the option of adding the pre-fix super- to rank designations that are not already prefixed. The phrase family group refers to the ranks of superfamily, family and subfamily; subdivision of a family only to taxa o f a rank between family group and genus group; genus group refers to the ranks of genus and subgenus; subdivision of a genus only to taxa of a rank between genus gr oup and species group; species group to the ranks of species and subspecies; and t he term infra-subspecfic refers to ranks beDraft BioCode The desirability of seeking some harmoni-zation of all biological Codes has been ap-preciated for some time (see Hawksworth, 1995) and an exploratory meeting on the subject was held at Egham, UK in March 1994. Recognizing the crucial import ance of scientific names of organisms in global com-munication, these decisions included not only agreement to take steps to harmonize the existing terminology and procedures, but also the desirability of working towards a unified system of biological nomenclature. The Draft BioCode is the first public expres-sion of t hese objectives. The first draft was prepared in 1995. After successive reviews the fourth draft, named Draft BioCode (1997) prepared by the International Commi ttee for Bionomenclature (ICB) and published by Greuter et al., (1998) is now av ailable on the web: (http://www.rom.on.ca/biodiversity/ low the species group. 3. Status: For the purposes of this Code biocode/biocode97.html) from the Royal Established names are those that are published in accordance with relevant artic les of this Code or that, prior to 1 January 200n, were validly published or bec ame available under the relevant Special Code. Acceptable names are those that a re in accordance with the rules and are not unacceptable under homonymy rule, an d, for names pub-lished before 1 January 200n, are neither illegitimate nor juni or homonyms under the relevant Special Code. In the family group, genus group, o r species group, the accepted name of a taxon with a particular circumscription, position, and rank is the acceptable

Ontario Museum. Botanical Nomenclature 37 name which must be adopted for it under the rules. In ranks not belong-ing to th e family group, genus group, or species group, any established name of a taxon a dopted by a particular au-thor is an accepted name. In this Code, unless otherwi se indicated, the word name means an established name, whether it be acceptable or unaccept-able. The name of a taxon consisting of the name of a genus combined w ith one epithet is termed a binomen; the name of a species combined with an infr aspecific epithet is termed a trinomen; binomina or trinomina are after fulfillment of the present requirements for valid publication. 5. Typification: The type of a nominal also termed combinations. taxon in the rank of genus or subdivision of a genus is a nominal species. The t ype of a nominal taxon of the family group, or of a nominal taxon of a higher ra nk whose name is ultimately based on a generic name, is the nominal genus. For t he names of superspecies, species or infraspecific taxon is a specimen in a muse um jar, herbarium sheet, slide preparation, or mounted set of freeze-dried ampou les. It should be in metabolically inactive state. Type designations must be pub -lished and registered. The typeless ( descriptive ) names do not have a representat ive type and are formed based on some character/s, apply to taxa defined by circ umscription, and may be used unchanged at different ranks above the rank of a fa mily. 4. Establishment of names: In order to be established on or after 1 January 200n, a name of a taxon must be pub-lished as provided for by the rules for publication, which are essentially similar to t he Effective publication in botany. The rules for establishment (valid publicati on of Botanical Code) are generally similar to the Botanical Code with certain c hanges. The new taxon may have a Latin or English descrip-tion or diagnosis (thu s Latin diagnosis is not mandatory). Change of rank within the family group or g enus group, or elevation of rank within the species group do not require the for mal estab-lishment of a new name or combina-tion. In order to be established, a name of a new fossil botanical taxon of spe-cific or lower rank must be accompanied by an illustration or figure show-ing diagnostic characters, in addition to the description or diagnosis, or by a bibliographic reference to a previously p ublished illustration or figure. This re-quirement also applies to the names of new non-fossil algal taxa at these ranks. Only if the corresponding genus or spe cies name is established can the name of a subordinate taxon. Establish-ment (va lid publication) under the BioCode includes registration of names in the family group, genus group and species group as a last step 6. Registration: Registration is affected by submitting the published matter that includes the protologue(s) or nomenclatu ral acts to a registering office designated by the relevant inter-national body. It is pertinent to men-tion that this requirement was based on the Botanical Co de (Tokyo Code, 1994) where it has already been abandoned (St. Louis Code, 2000) , removing all references to registration in the Botanical Code. The date of a n ame is that of its registration, which is the date of receipt of the relevant ma tter at the registering office. When alter-native (homotypic) names are proposed simultaneously for registration for one and the same taxon (same rank and same position) neither is considered to be submitted. When one or more of the other c onditions for establishment have not been met prior to registration, the name mu st be resubmitted for registration after these conditions have been fulfilled. 7. Precedence (priority): For purposes of precedence, the date of a name is 38 Plant Systematics either the date attributed to it in an adopted List of Protected Names or, for u nlisted names, the date on which it was validly published under the botani-cal o r bacteriological Code, or became available under the zoological Code, or was es tablished under the present Code. Limitations of priority that un-der previous C

odes affected names in certain groups or of certain catego-ries even if not provid ed for in the present Code still apply to such names if they were published before 1 January 200n The limitations to prec-edence are largely similar to botany. Co nservation and rejection procedures would remain largely the same as at present. The botanical process of sanc-tioning concerns old names only and need be provi ded for in a future contexts where such citations are neither informative nor really appropriate. Th is may be a timely change, since the current attitude is showing signs of cracki ng (Garnock-Jones and Webb, 1996). Art. 40.1 is so worded as to reflect this new attitude. 10. Hybrids: The Appendix for Hybrids in the Botanical Code is replaced by a single Article in the Draft BioCode. This ex treme simplification should in no way disrupt the present and future usage of hy brid designations, but has some philosophical changes as its basis. Most importa ntly, taxonomy and nomenclature are disentangled, in conformity with Principle I . Cultivated plants are not covered under the BioCode. BioCode. PhyloCode 8. Homonymy: The major change with respect to the homonymy rule would be that in future, it would operate across th e kingdoms. In order that this provision be applicable, it is necessary that lis ts of established generic names of all organisms be publicly available, ideally in electronic format; most such, The PhyloCode is being developed by Inter-national Committee on Phylogenetic Nom enclature on the philosophy of Phylogenetic taxonomy replacing the multirank Lin naean system with a rankless system recognizing only species and clades. It is in tended to cover all biological entities, living as well as apparently, already exist, but are not fossil. Underlying principle of the Phylo Code yet generally accessible. A list of is that the primary purpose of a taxon name across-kingdom generic homonyms in is to provide a means of referring unamb igucurrent use is being prepared, and, as ously to a taxon, not to indicate its relationa next step, a list of binomina in the ships. The PhyloCode grew out of recognicorresponding genera is planned, so that future workers may avoid the crea-tion of new (illegal) homonymous binomina. Existing across-kingdom homonyms would not lose their status of acceptable names, but would be flagged for the benefit of biological in-dexers and users of indexes. Existing names are not affected by th e proposed rules. The practice of Secondary Homonymy in ICZN is not followed in tion that the current Linnaean system of no-menclature as embodied in the pre-exis ting botanical, zoological, and bacteriological Codes is not well suited to govern the naming of clades and species, the entities that compose the tree of life an d are the most significant entities above the organism level. Rank assignment is subjective and biologically meaningless. The PhyloCode will pro-vide rules for the express purpose of naming the parts of the tree of life both species and clade s by explicit reference to phylogeny. In doing so, the PhyloCode extends tree-think ing to nomenclature. The PhyloCode is designed so that it can be used concur-rent ly with the pre-existing Codes or (after BioCode. 9. Author citation: The Draft BioCode signals a departure from the botanical tradition of laying great emphasis on the use of author citations, even in Botanical Nomenclature 39 rules governing species names are added) as the sole code governing the names of taxa, if the scientific community ultimately dewell, by removing the linkage to a genus name. Under the PhyloCode, phylogenetic position can easily be indicated by associating the species name with the names of one or more clades to which it belongs. Another benefit of phylogenetic nome nclature is that abandonment of ranks eliminates the most subjective aspect of t axonomy. The arbitrary nature of ranking is not widely appreciated

cides that it should. The starting date of the PhyloCode has not yet been determined and is cited as 1 January 200n in the draft Code. Rules are pro-vided for naming clades and will event ually be provided also for naming species. In this system, the categories species and clade are not ranks but different kinds of biologi-cal entities. A species is a segment of a popu-lation lineage, while a clade is a monophyl-etic group of sp ecies. Fundamental differ-ences between the phylogenetic and tradi-tional system s in how supraspecific names are defined lead to operational differences in the determination of synonymy and hom-onymy. For example, under the PhyloCode, synon yms are names whose phylogenetic definitions specify the same clade, regard-less of prior associations with particular ranks; in contrast, under the pre-existin g Codes, synonyms are names of the same rank based on types within the group of con-cern, regardless of prior associations with particular clades. The requireme nt that all established names be registered will reduce by non-taxonomists. The PhyloCode is designed so that it can be used concurrently with the rank-base d codes or (after rules governing species names are added) as the sole code gove rning the names of taxa, if the scientific community ultimately decides that it should. The intent is not to replace existing names but to provide an alternativ e system for governing the application of both existing and newly proposed names . In developing the PhyloCode, much thought has been given to minimizing the dis ruption of the existing nomenclature. Thus, rules and recommendations have been included to ensure that most names will be applied in ways that approximate thei r current and/or historical use. However, names that apply to clades will be red efined in terms of phylogenetic the frequency of accidental homonyms. relationships rather than taxonomic rank and therefore will not be subject to th e subsequent changes that occur under the rank-based systems due to changes in r ank. Be-cause the taxon membership associated with particular names will sometim es differ between rank-based and phylogenetic systems, suggestions are provided for indicating which code governs a name when Phylogenetic nomenclature was pre-sumed to have several advantages over the trad itional system. In the case of clade names, it eliminates a major source of in-s tability under the pre-existing Codes name changes due solely to shifts in rank. It also facilitates the naming of new clades as they are discovered and not wait ing till a full classification is developed as in the case of existing Codes. Th is is a particularly sig-nificant when new advances in molecular biology and com puter technology have led to a burst of new information about phylogeny, much of which is not being translated into taxonomy at present. The availability of the PhyloCode will permit researchers to name newly discovered clades much more eas ily than they can under the pre-existing Codes. At present PhyloCode has rules o nly for clades but when extended to species, it will improve nomenclatural stabi lity here as there is a possibility of confusion. The concept of PhyloCode was first introduced by de Queiroz and Gauthier (1992). The theoretical development of PhyloCode re-sulted from a series of papers from 1990 on-wards and three symposia first in 1995, the second in 1996 at the Ranch o Santa Ana Botanic Garden in Claremont, California, U.S.A., organized by J. Mar k Porter and en-titled The Linnean Hierarch: Past Present and Future, and the thir d at the XVI International Botanical Congress in St. Louis, Missouri, U.S.A. (19 99), entitled Overview 40 Plant Systematics and Practical Implications of Phylogenetic Preamble Nomenclature . Practical shape to the PhyloCode was given at the first workshop held in 1998 at the Harvard University Herbaria, Cam-bridge, Massachusetts, U.S.A. The initial philosophy of unification of biological world was based on draft BioCode. The fi

rst public draft of the PhyloCode was posted on the internet in April 2000. A se cond workshop was held at Yale University in July 2002 wherein it was decided to publish separate documents governing clade names and species names. Modified ve rsions of PhyloCode were posted in October 2003 (PhyloCode2), December 2003 (Phy locode2a) and 2004 (PhyloCode2b), June 2006 (PhyloCode3), July 2007 (PhyloCode 4 a) and September 2007 (PhyloCode4b). The efforts crystallized into the establish ment of the International So-ciety for Phylogenetic Nomenclature (ISPN) at the F irst International Phyloge-netic Nomenclature Meeting, which took place in July 2004 in Paris, attended by about 70 systematic and evolutionary biologists from 11 nations. The Second International Phylogenetic Nomenclature Meeting was held btween June 28 - July 2, 2006 at Yale University (New Haven, Connecticut, U.S.A. ), and the Third July 21 23, 2008 at 1. Biology requires a precise, coherent, international system for naming clades and species of organisms. Species names h ave long been governed by the traditional codes (listed in Preamble item 4), but those codes do not provide a means to give stable, unambiguous names to clades. This code satisfies that need by providing rules for naming clades and describi ng the nomenclatural principles that form the basis for those rules. 2. This code is applicable to the names of all clades of organisms, whether extant or extinct. 3. This code may be used concurrently with the rank-based codes. 4. Although this code relies on the rank-based codes (i.e., International Code of Botanical Nomenclature (ICBN), Int ernational Code of Zoological Nomenclature (ICZN), International Code ofN omencl ature ofB acteria: Bacteriological Code (BC), International Code of Virus Classification and No-mencla ture (ICVCN)) to determine the acceptability of preexisting names, it governs th e application of those names independently from the rankDalhousie University, Halifax. The latest version of the PhyloCode (PhyloCode4b) was posted in September 2007 a nd includes many substantive modificabased codes. 5. This code includes rules, recommentions. The version is available at http:// dations, notes and examples. Rules are mandatory in that names contrary to them have no official standing under this code. Recommendations are not mandatory in that names contrary to them cannot be rejected on that basis. Systematists are e ncouraged to follow them in the interest of promoting nomenclatural uniformity a nd clarity, but editors and reviewers should not require that they be followed. Notes and examples are intended solely for www.ohiou.edu/phylocode/. The latest changes concern the name of species (Ar-ticle 21-Regulation of species names is left to rank-based Codes; The genus portion of the binomen, called the prenomen is treated as simply the first part of the spe-cies name and need not be establish ed un-der this code), Crown and total clade names (Art. 10-To have integrated sy stem of clade names and providing more nomenclatural freedom) and emendation of definitions(Art. 15- Unrestricted emendations can be pubclarification. 6. This code will take effect on the lished without CPN (Committee on Phylo-genetic Nomenclature) whereas a re-strict ed emendation needs CPN approval). publication of Phylonyms: a Companion to the PhyloCode, and it is not retroactiv e. Botanical Nomenclature 41 Principles Every individual organism (on Earth) belongs to at least one clade (i.e., the cl ade comprising all extant and extinct organisms, assuming that they share a sing

le origin). Each organism also belongs to a number of nested clades (though the ancestor of the clade comprising all life again assuming a single origin does not be long to any other clade). It is not necessary that all clades be named. Clades a re often either nested or mutually exclusive; how-ever, phenomena such as specia tion via hybridization, species fusion, and symbiogenesis can result in clades t hat are partially overlapping. This code does not prohibit, discourage, encourag e, or require the use of taxonomic ranks. In this code, the terms species and clade refer to different kinds of bio- logical entities, not ranks. The con-cepts of s ynonymy, homonymy, and precedence adopted in this code are, in contrast to the p re-existing codes, independent of categorical rank. 1. Reference. The primary purpose of taxon names is to provide a means of referring to taxa, as opposed to indicat-in g their characters, relationships, or membership. 2. Clarity. Taxon names should be unambiguous in their designation of par-ticular taxa. Nomenclatural clarity is ach ieved through explicit definitions, which describe the concept of the taxon desi gnated by the defined name. 3. Uniqueness. To promote clarity, each taxon should have only one accepted name, and each accepted name should refer to only one taxon. 4. Stability. The names of taxa should not change over time. As a corollary, it must be possible to name newly discovered t axa without changing the names of previously discovered taxa. 5. Phylogenetic context. This code is concerned with the naming of taxa and the application of taxon names in the context of phylogenetic concepts of taxa. 2. Publication: The provisions of the Code 6. Taxonomic freedom. This code perapply not only to the publication of names, but also to the publication of any n omenclatural act (e.g. a proposal to conserve a name). Publication, under this c ode, is defined as distribution of text (but not sound), with or without images, in a peer-reviewed book or periodical. To qualify as published, works must cons ist of at least 50, simultaneously obtainable, identical, durable, and unalterab le copies, some of which are distributed to major insti- tutional libraries (in at least five countries on three continents) so that the work is generally acces sible as a per- manent public record to the scientific community, be it through sale or exchange or gift, and subject to the re-strictions and qualifications in the mits freedom of taxonomic opinion with regard to hypotheses about relation-ships ; it only concerns how names are to be applied within the context of a given phylogenetic hypothesis. 7. There is no case law under this code. Nomenclatural problems are re-solved by the Committee on Phylogenetic Nome nclature (CPN) by direct application of the code; previ-ous decisions will be co nsidered, but the CPN is not obligated by precedents set in those decisions. Salient Features At present the Phylocode has rules only for clades. Rules for species will be ad ded later on. 1. Taxa: Taxa may be clades or species, present article. but only clade names are governed by the PhyloCode. In this code, a clade is an ancestor (an organism, population, or species) and all of its descendants. 3. Names-status and establishment: Established names are those that are published in accordance with rules of PhyloCode. In order to indicate which 42 Plant Systematics

names are established under this Code and therefore have explicit phylogenetic d efinitions (and whose endings are not reflective of rank), it may be desirable t o distinguish these names from the supraspecific names governed by pre-existing codes, particu-larly when both are used in the same publication. The letter P (bra cketed or in superscript) might be used to desig-nate names governed by the Phyl oCode, and the letter L to desig-nate names governed by the pre-exist-ing Linnaean codes. Using this con-vention, the name Ajugoideae[L] would apply to a plant subf amily which may or may not be a clade, whereas Teucrioideae[P] would apply to a cl ade which may or may not be a subfamily. Establishment of a name can only oc-cur on or after 1 January 200n, the starting date for this code. In order to be est ablished, a name of a taxon must be properly published, be adopted by the author (s), be registered, and the regis-tration number must be cited in the protologue . The accepted name of a taxon is the name that must be adopted for it under thi s code. It must; (1) be established; (2) have precedence over alternative uses o f the same name (homonyms) and alternative names for the same taxon (synonyms); and (3) not be rendered inapplicable by a qualify-ing clause in the context of a particuof preexisting names or introduction of new names. In order to be established, t he name of a clade must consist of a single word and begin with a capital letter . In order to be established, con-verted clade names must be clearly identified as such in the protologue by the designation converted clade name or nomen cladi co nversum . New clade names must be identified as such by the designation new clade n ame or nomen cladi novum . In order to be established, a clade name must be provided with a phylogenetic definit ion, written in English or Latin, linking it explicitly with a particular clade. The name applies to whatever clade fits the definition. Examples of phylogeneti c definitions are node-based, stem-based, and apomorphybased definitions. A node -based definition may take the form the clade stem-ming from the most recent comm on ancestor of A and B (and C, D, etc., as needed) or the least inclusive clade containing A and B (and C, D, etc.), where A-D are specifiers. A node-based lar phylogenetic hypothesis. definition may be abbreviated as Clade (A+B). A stem-based definition may take t he form the clade consisting of Y and all organisms that share a more recent comm on ancestor with Y than with W (or V or U, etc., as needed) or the most inclusive clade containing Y but not W (or V or U, etc.). A stem-based definition may be ab breviated as Clade (Y< W). An apomorphy-based defini4. Registration: In order for a name to be established under the PhyloCode, the name and other required informa-tion mus t be submitted to the PhyloCode registration database. A name may be submitted t o the data-base prior to acceptance for publication, but it is not registered (i .e. given a reg-istration number) until the author no-tifies the database that t he paper or book in which the name will appear has been accepted for publication . tion may take the form the clade stem-ming from the first species to possess char acter M synapomorphic with that in H . An apomorphy-based definition may be abbrev iated as Clade (M in H). When giving a new name for total clade, prefix Pan- mus t be used to the name of crown clade (separated by hyphen) and designated as pan clade. 6. Specifiers and Qualifying Clauses: 5. Clade Names: The names of clades Specifiers are species, specimens, or synapomorphies cited in a phylogenetic def inition of a name as reference points may be established through conversion Botanical Nomenclature 43 Table 2.2 Equivalence table of nomenclatural terms used in the Draft PhyloCode, the Draft

BioCode and the current biological codes (excluding Code for Vriuses). PhyloCode BioCode Bacteriological Code Botanical Code Zoological Code Publication and pre][][ oircedence of names published published effectively published effectively published published precedence precedence priority priority precedence earlier earlier senior earlier senior later later junior later junior Nomenclatural status established established available validly published validly published converted acceptable acceptable legitimate legitimate

potentially valid registration registration validation Taxonomic status accepted accepted

correct correct valid Synonymy and homonymy homodefinitional homotypic heterodefinitonal heterotypic objective nomenclatural objective subjective taxonomic subjective replacement name replacement name avowed substitute new replacement name Conservation and suppression conserved conserved conserved conserved conserved suppressed suppressed/ rejected rejected suppressed rejected that serve to specify the clade to which the name applies. All specifiers used i n node-based and stem-based defini-tions of clade names, and one of the specifie rs used in apomorphy-based definitions of clade names, are species or specimens. The other specifier used in an apomorphy-based definition of a clade name is a synapomorphy. If sub-ordinate clades are cited in a phylogenetic definition of a more inclu-sive clade, their specifiers must also be explicitly cited within th e definition of the more inclusive clade. An inter-nal specifier is one that is explicitly included in the clade whose name is being defined; an external specif ier is one that is explicitly excluded from it. All specifiers in node-based and apomor phy-based definitions are internal, but stem-based definitions must always have at least one specifier of each type. When a species is used as a specifier, the author and publication year of the species name must be cited. When a type specimen is used as a specifier, the species name that it typifies and the author and publication year of that speci es name must be cited. 7. Precedence: Although the entity to which precedence applies in this code is referred to as a name, it is really the combination of a name and its definition. In different cases, one or the 44 Plant Systematics other of these components is more important. Specifically, in the case of synony ms, precedence refers primarily to the name, whereas in the case of homonyms, pr ecedence refers prima-rily to the definition. Precedence is based on the date of establishment, with earlier-established names having precedence over later ones , except that later-established names may be constances and requires approval of the CPN. Once a name has been con-served, the e ntry for the affected name in the registration database is to be annotated to in dicate its conserved status relative to other names that are simultaneously supp ressed. An emendation is a formal change in a phylogenetic definition. A restric ted emendation (changes in definitional type, clade category, specifiers, and/or qualifying clauses) requires approval by the CPN, while an unrestricted emendat

ion (changes in specifiers or qualiserved over earlier ones. In the case of homonymy involving names gov-erned by t wo or more preexisting codes (e.g. the application of the same name to a group o f animals and a group of plants), precedence is based on the date of establishme nt under the PhyloCode. However, the International Committee on Phylogenetic Nom enclature has the power to conserve a later-established homonym over an earlierestablished homonym. This might be done if the later homonym is much more widely known than the earlier one. For the determination of precedence, the date of es tablishment is considered to be the date of publication, and not the date of fying clauses) may be published without CPN approval. 10. Provisions for hybrids: Hybrid origin of a clade may be indicated by placing the multiplication sign () in front of the name. The names of clades of hybrid origin otherwise follow the same rules as f or other clades. An organism that is a hybrid between named clades may be indica ted by placing the multiplication sign between the names of the clades; the whol e expression is then called a hybrid formula. registration. 11. Authorship of Names: A taxon name 8. Synonymy: Synonyms are names that is to be attributed to the author(s) of the protologue, even though authorship o f the publication as a whole may be different. In some cases, it may be desirabl e to cite the author(s) who established a name. If the author of a con-verted na me is cited, the author of the pre-existing name on which it is based must also be cited, but in square brackets[]. If the author of a replacement name is cited , the author of the definition of the replaced name must also be cited, but in b races{}. If the author of a homonym that has been conserved for the purpose of e mending a definition is cited, the author of the original are spelled differently but refer to the same taxon. In this code, synonyms must be established and may be homodefinitional (based on the same definition) or he terodefinitional (based on different definitions). Homodefini-tional synonyms ar e synonyms regard-less of the phylogenetic context in which the names are applie d. However, in the case of names with different definitions, the phylogenetic co ntext determines whether the names are heterodefinitional synonyms or not synony mous. When two or more syno-nyms have the same publication date, the one that wa s registered first (and therefore has the lowest registration using definition must also be cited, but by < and > symbols (e.g., number) takes precedence. Hypotheticus <St Maki). Phylocode 9. Conservation, supression and emenfollows the use of in but not ex. dation: Conservation of names is pos-sible only under extraordinary circum12. Species names: This code does not govern the establishment or precedence Botanical Nomenclature 45 of species names. To be considered available (ICZN) or validly published (ICBN, BC), a species name must sat-isfy the provisions of the appropriate rank-based c ode. Because this code is independent of categorical ranks, the first part of a species binomen is not interpreted as a genus name but in-stead as simply a pren omen, first part of the species name, and the second part of a species binomen i s associated with the species as a kind of biological entity, not as a rank. A p renomen has no necessary tie to any categorical rank under this code. This code also does not govern the establishment of names associated with ranks below that of species under the rank-based codes ( infraspecific names ); however, such names may be used in conjuncstated goals can t be met by proposals in current draft, which also fails to uphol

d its stated Principles. The internal contradictions include a cumbersome reinve ntion of the very aspect of the current Linnaean System that the advocates of Ph yloCode most often decry. The incompleteness is due to the fact that the drafter s cannot agree on what form the species names should take. Keller et al. (2003) pointed out inherent instabilities, fundamental flaws in its very foundation by exposing unsubstantiated philosophical assumptions preceding and subtending it. tion with phylogenetic nomenclature. A strong opposition to the PhyloCode was offered by Nixon et al. (2003) who conc luded that The PhyloCode is fatally flawed, and the Linnaean System can be easily fixed. They argued that the proponents of the PhyloCode have offered nothing re al to back up claims of greater stability for their new system. A rank free syst em of naming would be confusing at the best and would cripple our ability to tea ch, learn and use taxonomic names in the field or publications. They assured tha t the separate issue of stability in reference to rules of priority and rank can be easily addressed within the current Codes, by implementation of some simple changes. Thus there is no need to scrap the current Linnaean Codes for a poorly re asoned, logically in-consistent and fatally flawed new Code that will only bring chaos. 13. Governance: The PhyloCode will be managed by The Society for Phylogenetic Nomenclature (SPN) through its two commi ttees: International Committee on Phylogenetic No-menclature (ICPN) and the Regi stration Committee. The desirability of PhyloCode has been reviewed in several papers published over last few years. Nixon and Carpenter (2000) showed that Phylogenetic nomenclatur e would be less stable than existing systems. A critique of draft PhyloCode is p resented by Carpenter (2003), pointing out that its Chapter 3 H ication ierarchical Classif It would be total chaos to study and document information about more than a quar ter mil-lion species of vascular plants if there were no proper mechanism for gr ouping the same. Whatever may be the criterion for classifi-cation artificial char acters, overall mor-phology, phylogeny or phenetic relationship the basic steps a re the same. The organ-isms are first recognized and assembled into groups on th e basis of certain resemblance. These groups are in turn assembled into larger a nd more inclusive groups. The pro-cess is repeated until finally all the organ-i sms have been assembled into a single, largest most inclusive group. These group s (Taxonomic groups or Taxa) are arranged in order of their successive inclusive ness, the least inclusive at the bottom, and the the groups have been assigned categories and named, the process of classificatio n is complete, or the taxonomic structure of the whole largest most inclusive gr oup has been achieved. Because of the hierarchical arrangement of categories to which the groups are assigned, the classification achieved is known as hierarchi cal classification. This concept of categories, groups and taxonomic structure c an be illustrated in the form of a box-in-box figure (Figure 3.1) or a dendrogra m (resembling a pedigree chart, Figure 3.2). TAXONOMIC GROUPS, CATEGORIES AND RANKS most inclusive at the top. Taxonomic groups, categories and ranks are inseparable once a hierarchical class ification has been achieved. Rosa alba is thus nothing else but a species and Ro sa is nothThe groups thus formed and arranged are next assigned to various categories, having a fixed sequence of arrangement (taxo- ing other than a genus. However, the difn

omic hierarchy), the most inclusive group ferences do exist in concept and appli cation. assigned to the highest category (generally The categories are like shel ves of an a division) and the least inclusive to the low- almirah, having no sig nificance when est category (usually a species). The names empty, and importance and meaning only are assigned to the taxonomic groups in such after something h as been placed in them. a way that the name gives an indication of Thereafter, t he shelves will be known by the category to which it is assigned. Rosales, their contents: books, toys, clothes, shoes Myrtales, and Malvales all belong to the or- etc. Categories in that sense are artificial der category and Rosaceae, Myrt aceae and and subjective and have no basis in reality. Malvaceae to the family c ategory. Once all They correspond to nothing in nature. HowHierarchical Classification 47 Class Order Genus Division Subclass Fan Species

CD }

Figure 3.1 Processes of assembling taxonomic groups according to the hierarchica l system, depicted by box-in-box method. In the above example, there are 18 species grouped into 10 genera, 6 families, 4 orders, 3 subclasses, 2 classes and 1 division. ever, they have a fixed position in the hier-archy in relation to other categori es. But once a group has been assigned to a particu-lar category the two are ins eparable and the category gets a definite meaning because it now includes someth ing actually occurring in nature. The word genus does not carry a specific meani ng but the genus Rosa says a lot. We are now talking about roses. There is pract ically no difference between category and rank, except in the grammatical sense. Rosa thus belongs to the category genus, and has generic rank. If categories ar e like shelves, ranks are like partitions, each sepa-rating the given category f rom the category above. Taxonomic groups, on the other hand, are objective and n on-arbitrary to the extent that they represent discrete sets of organ-isms in na ture. Groups are biological enti-ties or a collection of such entities. By as-si gning them to a category and providing an appropriate ending to the name (Rosace ae with ending aceae signifies a family which among others also includes roses, belo ng-ing to the genus Rosa) we establish the position of taxonomic groups in the h ierarchical system of classification. Some important characteristics, which enab le a better understanding of the hierarchical system of classification, are enumerated below.

1. Different categories of the hierarchy are higher or lower according to whether they are occupied by more inclusive or less inclusive groups. Higher categories are occupied by more inclusive groups t han those occupying lower categories. 2. Plants are not classified into categories but into groups. It is important to note that a plant may be a member of se veral taxonomic groups, each of which is assigned to a taxonomic cat-egory, but is not itself a member of any taxonomic category. A plant collected 48 Plant Systematics Division Class Subclass Order Fanuliy Genus Species 11 0 11 11 11

11 11

11 11 11 11

0 0 0 0 Figure 3.2 Dendrogram method for depicting the hierarchical system based on same hypothetical example as in Figure 3.1. from the field may be identified as Poa annua (assigned to species category). It is a member of Poa (assigned to genus category), Poaceae (assigned to family ca tegory) and so on, but the plant can t be said to be belonging to the speguish the taxa. Dicots are thus conveniently separated from monocots by possessi on of two cotyledons, pentamerous flowers, reticulate venation and vascular bund les in a ring as against one cotyledon, trimerous flowers, parallel venation and scattered vascular bundles in monocots. But when taken individually, Smilax is a monocot with reticulate venation and Plantago is a dicot with parallel venatio n. Similarly Nymphaea, is a dicot with scattered bundles, and the flow-ers are t rimerous in Phyllanthus, which is a dicot. cies category. 3. A taxon may belong to other taxa, but it can be a member of only one category. Urtica dioica, thus, is a member of Urt ica, Urticaceae, Urticales, and so on, but it belongs only to species category. 4. Categories are not made up of lower categories. The category family is not made up of the genus category, since ther e is only one genus category. 5. The characters shared by all members UTILIZATION OF CATEGORIES of a taxon placed in a lower category provide the characters for the taxon immed iately above. Thus, the charac-ters shared by all the species of Brassica make u p the characters of the genus Brassica. The characters shared by Brassica and se veral other genera form distinguishing characters of the fam-ily Brassicaceae. I t is important to note that the higher a group is placed in the hierarchy, the f ewer will be the char-acters shared by the subordinate units. Many higher taxa, as such (e.g. Dicots: Magnoliopsida) can only be separated by a combination of c haracters; no sin-gle diagnostic character may distinTaxonomic categories possess only relative value and an empty category has no fo undation in reality and obviously can t be defined. An important step in the proce ss of classification is to assign taxa to an appropriate category. It thus becom es imperative to decide what should be the properties of taxa to be included in a particular category? Only with a proper utilization of the concept of categori es can their application in hierarchical systems be meaningful. The problem is f ar from resolved. An attempt will be made here to discuss the relevant aspects o f the inclusion of type of entities or groups of entities under different catego ries.

Hierarchical Classification 49 Species concept ing to the taxa at other ranks. Thus we can sort herbarium sheets into different species without difficulty, without knowing or bothering to know how many gener a are covered by these sheets. We cannot recognize genera or describe them witho ut reference to the included species. Species is thus the only category dealing directly with Darwin aptly said: Every biologist knows approximately what is meant when we talk about species, yet no other taxon has been subjected to such violent controvers ies as to its definition . A century and a half has passed, so much advancement in the taxo nomic knowledge has been achieved, yet the statement of Darwin is as t rue today as it was then. Numerous definitions of species have been proposed, ma king it futile to recount all of them. Some significant as- pects of the problem will be discussed here. Probably the best explanation of diversity of the plants. 3. Whether defined in terms of morphoopinions can be explained as under. logical discontinuity or restriction of gene exchange, species is unique in bein g non-arbitrary to both inclusion and exclusion. A group is non-arbitrary to inc lusion if all its members are con-tinuous by an appropriate criterion. It would be arbitrary to inclusion if it shows internal discontinuity. A group is non-arb itrary to exclusion if it is discontinuous from any other group by the same crit erion. A group not show-ing discontinuity with other groups is arbitrary. All hi gher taxa although non-arbitrary to exclusion are arbitrary to inclusion, i.e. t hey exhibit internal discontinuity as now species with external discontinuity fo rm part of these taxa. The species is a concept. Concepts are constructed by the human mind, and as humans think differently uYe have so many definitions of a species. Obviously a con cept can t have a single acceptable definition. The word species has different meaning for different botanists. According to ICB N, which has attempted to clarify the meaning of the word species, species are co nvenient classificatory units defined by trained bi ologists using all informati on available . The word species has a dual connotation in biological science. Firs t, the species is a naturally-occurring group of individual organisms that compr ises a basic unit of evo-lution. Second, the species is a category within a taxo nomic hierarchy governed by various rules of nomenclature. Species as Basic Unit of Ideal Species A perfect situation! Species that can be easily distinguished and have no proble m of identity. Such species, however, are very few; common examples include Apia ceae, Asteraceae and the genera Allium and Se- dum. The following characteristic s are exTaxonomy The following information serves to substan-tiate the view that species constitu tes the basic unit of classification or, for that matpected in an ideal species: 1. The species poses no taxonomic ter, taxonomy (systematics): problems and is easily recognized as a distinct entity on the basis of morpho1. Species is considered the basic unit of logical characters. taxonomy, since in the greater major-ity of cases, we do not have infraspecific names. This is especially common in families such as Apiaceae (Umbelli2. It exhibits no discontinuity of variation within, i.e. it contains no subspecies, varieties or formas. 3. It is genetically isolated from other ferae) and Liliaceae. species.

2. 4. 5. 50

Species, unlike other taxa, can be de-scribed and recognized without relatIt is sexually reproducing. It is at least partially outbreeding. Plant Systematics

Unfortunately, ideal species are rare among the plant kingdom and the greater ma jority of species pose situations contrary to one or more of the above criteria. pose of nomenclature, all organisms must be referable to species. Species, by th is concept, can be defined by the language of formal relations and not by proper ty of their organisms. The concept considers species to be a category in taxonom ic hierarchy and may correspond to a specific name Idea of Transmutation This is an ancient Greek idea which perin the binomial system of nomenclature. sisted as late as the seventeenth century. Greeks believed in the transmutation of The concept is logically sound but scien tifiwheat into barley, Crocus into Gladiolus, bar- cally irrelevant since the ultima te aim is to place a particular group of individuals in ley into oats, and many other plants, under certain conditions. The supporters of this notion often included professional bo tanists like Bobart (who swore that Crocus and Gladiolus, as likewise the Leucoj um, and Hyacinths by a long standing without re-planting have in his garden chan ged from one kind to the other) as reported by Robert Sharrock (1660) in his boo k History of the propagation and improvement of vegetables by the concurrence of art and nature. Sharrock fortunately, however, on investigation did not find an y proof of this in the field. So called transmutation can be explained as nothin g other than the result of unintentional mix-ing of seeds or other propagules of another a species. Typological Species Concept plant with a particular crop before plantation. This concept was first proposed by John Ray (1686) and further elaborated by C. Linnaeus in Critica botanica (1737). Linnaeus refuted the idea of transmutation of species. Linnaeus believed that although there is some variation within a spe cies, the species by themselves are fixed (fixity of species) as created by the Almighty Creator. The species, according to the concept, is a group of plants wh ich breed true within their limits of variation. Towards the later part of his l ife, however, Linnaeus moved away from idea of fixity of species and was con-vin ced that species can arise by hybridizaThe present author had a glimpse of this fallacy while studying the weeds in saffron tion. In his later publication (Fund amenta (Crocus sativus) fields of Kashmir valley. With fructificationis, 1762), Linnaeus imagined a few vegetative specimens of Iris reticulata that at the time of creation, there arose as (whose corms and leaves are closely similar to saff ron; the flowers are quite distinct) in his hand, the author tried in vain to co n-vince the saffron grower (who always thinks that he knows more about his crop) that the plant he was carrying was not saffron. The author managed to escape th e assault but was more convinced that this Iris (which does not grow elsewhere i n Kashmir valley) would have come unintentionally from Persia where it grows com monly, and from where the Kashmir saffron is supposed to have been introduced. T he concept of transmutation is now firmly rejected. many genera as were the individuals. These, in the course of time, were fertiliz ed by others and thus arose species until so many were produced as now exist. Th ese species were sometimes fertilized by other species of the same genus, giving rise to varieties. The typological concept, however, should not be confused wit h typification, which is a distinct methodology of nomenclature, providing names to taxonomic groups. Taxonomic Species Concept

The doctrine of fixity was challenged by Lamarck (1809) and finally Darwin (1859), who recognized continuous and disconti nuNominalistic Species Concept ous variation and developed his taxonomic species concept based on morphology, m ore appropriately known as the Morphological This nominalistic species concept is also only of academic interest now. For the purHierarchical Classification 51 species concept. According to this concept, the species is regarded as an assemb lage of individuals with morphological fea-tures in common, and separable from o ther such assemblages by correlated morphological discontinuity in a number off eatures. The supporters of this view be-lieve in the concept of continuous and d is-continuous variations. The individuals of a species show continuous variation , share certain characters and show a distinct dis-continuity with individuals b elonging to an-other species, with respect to all or some of 1. It is useful for general taxonomic purposes especially the field and herbarium identification of plants. 2. The concept is very widely applied and most species have been recognized using this concept. 3. The morphological and geographical features used in the application of this concept can be easily observed in populations. 4. Even experimental taxonomists who do not recognize this concept, apply this concept in cryptic form. these characters. 5. The greater majority of species recogDu Rietz (1930) modified the taxonomic species concept by also incorporating the role of geographic distribution of populations and developed the morpho-geograp hical species concept. The species was defined as the smallest population that i s permanently separated from other populations by disnized through this concept correspond to those established after experimental confirmation. The concept, however, also has some inherent drawbacks: 1. It is highly subjective and different sets of characters are used in differtinct discontinuity in a series of biotypes. ent groups of plants. The populations recognized as distinct spe-cies and occurring in separate geogra phical areas are generally quite stable and remain so even when grown together. There are, however, examples of a few species pairs which are morphologically qu ite distinct, well adapted to respective climates, but when grown together, they readily interbreed and form intermediate fertile hybrids, bridg2. It requires much experience to practice this concept because only after considerable observation and experience can a taxonomist decide the characters which are reliable in a particular taxonomic group. 3. The concept does not take into account the genetic relationships between plants. ing the discontinuity gap between the species. Examples are Platanus orientalis of the Biological Species Concept Mediter ranean region and P. occidentalis of The biological species concept was first de E. United States. Another well-known pair veloped by Mayr (1942) who defined spe cies is Catalpa ovata of Japan and China and C. bignonioides of America. Such pairs o

f spe-cies are known as vicarious species or vicariants and the phenomenon as as groups of actually or potentially interbreeding natural populations, which ar e reproductively isolated from other such groups. The words actually or potential ly , being meaningless, were subsequently dropped by Mayr (1969). Based on the sam e criteria, Grant (1957) defined species as a community of cross-fertilizing ind ividuals linked together by bonds of mating and reproductively isolatedf rom oth er species vicariance or vicariism. Morphological and morpho-geographical types of taxonomic species have been widel y accepted by taxonomists who even take into account the data from genetics, cyt ology, ecology, etc., but firmly believe that species recognized must be delimit ed by morphological characters. by barriers to mating. The recognition of biological species thus involve: (a) i nterThe taxonomic species concept has sev-eral advantages: breeding among populations of the same species; and (b) reproductive isolation 52 Plant Systematics between populations of different species. Valentine and Love (1958) pointed out that species could be defined in terms of gene exchange. If two populations are capable of exchanging genes freely either under natural or artificial conditions , the two are said to be conspecific (belonging to the same species). On the oth er hand, if the two populations are not capable of ex-changing genes freely and are reproduc-tively isolated, they should be considered specifically distinct. T he concept has sevspecies according to the biological species concept) but morphologically similar (single species according to the taxonomic species concept). Such species are known as sibling species. 4. Fertility-sterility is only of theoretical value in allopatric populations. 5. It is difficult and time consuming to carry out fertility-sterility tests. 6. Occurrence of reproductive barriers has no meaning in apomicts. 7. Necessary genetic and experimental eral advantages: data are available for only very few 1. It is objective and the same criterion is used for all the groups of plants. species. 2. It has a scientific basis as the popuStebbins (1950), it would appear, combined two concepts when he stated that spec ies must consist of systems of populations that are separated from each other by complete or at least sharp discontinuities in the variation pattern, and that t hese discontinuities must have a genetic basis These populations with isolating mechanisms (different species) may occur either in the same region (sympatric sp ecies) or lations showing reproductive isolation do not intermix and the morphological dif ferences are maintained even if the species grow in the same area. 3. The concept is based on the analysis of features and does not need experience to put it into practice. The concept, first developed for animals, holds true because animals as a rule a re sexually differentiated and polyploidy is very rare. When applying this conce pt to plants, however, a number of problems are encounin different regions (allopatric species). tered: Fortunately, although the taxonomic and biological concepts are based upon diffe rent principles, the species recognized by one concept, in the majority of cases

, stand the test of the other. Morphology provides the evidence for putting the genetic definition into practice. 1. A good majority of plants show only vegetative reproduction, and hence the concept of reproductive isolation as such cannot be applied. 2. Reproductive isolation is commonly Evolutionary Species Concept verified under experimental conditions, usually under cultivation. It may This concept was developed by Meglitsch (1954), Simpson (1961) and Wiley (1978). Al-though maintaining that interbreeding among sexually reproducing individuals is an important component in species cohesion, this concept is compatible with a broad range of reproductive modes. Wiley (1978) defines: an evolutionary speci es is a single lineage of ancestor-descendant populations which maintains its id entityf rom other such lineages, and which has its own evolutionary tendencies a nd historical fate. This concept avoids many of the problems of the biological c oncept. Lineage is a single series of demes (populations) that have no relevance for wild populations. 3. Genetic changes causing morphological differentiation and those causing reproductive barriers do not always go han d in hand. Salvia mellifera and S. apiana are morphologically distinct (two sepa rate species according to the taxo-nomic species concept) but not reproductively isolated (single species according to the biological species con-cept). Such sp ecies are known as compilospecies. Contrary to this, Gilia inconspicua and G. tr ansmontana are reproductively isolated (two separate Hierarchical Classification 53 share a common history of descent, not shared by other demes. The identity of sp e-cies is based on recognition systems that operate at various levels. In sexual ly repro-ducing species, such systems include rec-ognition because of phenotypic , behavioural and biochemical differences. In asexual spe-cies phenotypic, genot ypic differences main-tain the identity of species. Identity in both sexual and asexual species may also be due to distinct ecological roles. Viewed from the st andpoint of evolutionary species concept, however, the important question is not whether two species hybridize, but whether two species do or do not lose their distinct ecological and evolutionary roles. If, despite some hybridization, they do not merge, then they remain separate species in the evolulished by Linnaeus. Microspecies are distinct from cryptic species, which are mo rphologically similar but cytologically or physiologically different. Stace (198 9) uses the term semi-cryptic species for the latter. Biosystematic Species Concept tionary perspective. The term biosystematic species has been used by Grant (198 1) to refer to the ca tegories based on fertility relationships as de-termined by artificial hybridiza tion experiments. Ecotype refers to all members of a species that represent a pro duct of genetic response of a species towards a particular habitat . The ecotypes, which are able to exchange genes freely without loss of fertility or vigour in the offsprings, form an ecospecies. An ecospecies corresponds to a taxonomic spe cies. A group of ecospecies capable of limited genetic exchange constitutes a co enospecies. A coenospecies is considered equivalent to a subgenus. A group of re lated coenospecies between which hybrid-ization is possible directly or through in -termediates constitutes a comparium, which is considered equal to a genus. Compl ete sterility barriers exist between genera. Several other terms have been proposed to distinguish species based on specific criteria. Grant (1981) recognizes microspecies as populations of predominantly un iparen-tal plant groups which are themselves uni-form and are slightly different iated morpho-logically from one another ; they are often restricted to a limited g eographical area. Microspecies develop in inbreeding species, but are usually no t stable over longer peri-ods. They may undergo cross-fertilization sooner or la ter forming recombinant types which themselves become new microspecies. Several microspecies have been found in Erophila verna mostly repre-senting single bioty

pes or groups of similar biotypes some of which are marked by only one or two ch aracters. These may be dis-tinguished as clonal microspecies (repro-ducing by ve getative propagation, e.g. Phragmites), agamospermous microspecies (reproducing by agamospermy, e.g. Rubus), heterogamic microspecies (reproducing by genetic sy stems, e.g. Oenothera biennis or Rosa canina), and autogamous microspecies (pred ominantly autogamous and chromo-somally homozygous, e.g. Erophila). The term mic rospecies was first suggested by Jordan (1873) and as such they are often termed as Jardanons to distinguish them from Linnaeons, the normal species, first esta bInfraspecific ranks The species is regarded as the basic unit of classification and many works, incl uding the Flora of USSR, do not recognize infraspecific taxa. Many European, Ame rican and Asian Floras, however, do recognize taxa below the rank of species. Th e international Code of Botanical Nomenclature recognizes five infraspecific ran ks: subspecies, variety (Latin, varietas), subvariety, form (Latin, forma) and s ubform. Of these, three (subspecies, variety and form) have been widely used in the literature. Du Rietz (1930) defined subspecies as a population of several biotypes forming m ore or less a distinct regional facies of a species. Facies stands for race. Mor phologically distinct but interfertile populations of a species growing in diffe rent geographical regions are maintained as distinct subspe54 Plant Systematics cies due to the geographical isolation of the by cytogenetic and geographic information in relation to morphology. species. Du Rietz defined variety as a population of several biotypes, forming more or le ss a local facies of a species. The term variety is commonly used for morphologi cally dis-tinct populations occupying a restricted geo-graphical area. Emphasis is on a more lo-calized range of the variety, compared with the large-scale regi onal basis of a subspe-cies. Several varieties are often recognized within a sub species. The term variety is also used for variations whose precise nature is no t understood, a treatment often necessary 2. The genera should not be distinguished on a single character but a sum total of several characters. In a number of case s, genera are easily recognized on the basis of adaptive characters (adaptations in response to ecological niches), as in the case of establishing aquatic speci es of Ranunculus under a separate genus Batrachium. 3. There is no size requirement for a in the pioneer phase of taxonomy. genus. It may include a single species (monotypic genus) as Leitneria, Ginkgo, M ilula or many (Polytypic genus): Euphorbia (2100 species), Astragalus (2000) Car ex (1800), Senecio (1500) and Acacia (1300) being the examples of large genera. The genus Senecio was earlier included more than 2500 species, but it has now be en split into several genera. The only important criterion is that there should be a decided gap between the species of two genera. If the two genera are not re adily separable, then they can be merged into one and distinguished as subgenera or sections. Such an exercise should take into considera- tion the concept in o ther genera of the family, size of the genus (it is more convenient to have subg enera and sections in a larger genus) and Forma is often regarded as sporadic variant distinguished by a single or a few l inked characters. Little taxonomic signifi-cance is, however, attached to minor and random variations upon which the forms are normally based. Genus The concept of genus is as old as folk science itself as represented by names ro se, oak, daffodils, pine and so on. A genus represents a group of closely-relate d species. According to Rollins (1953), the function of the genus concept is to bring together species in a phylogenetic manner by placing the closest related s pecies within the general classification. When attempt-ing to place a species wi

thin a genus, the primary question would be, is it related to the undoubted spec ies of that genus? Mayr (1957) defined genus as a taxonomic cat-egory which cont ains either one species or a monophyletic group of species, and is separable fro m other genera by a de-cided discontinuity gap. It was earlier believed that a g enus should always be readily definable on the basis of a few tech-nical floral characters. A more rational recognition should take the following traditional usage. 4. When generic limits are being drawn, it is absolutely necessary that the group of species should be studied throughou t the range distribution of the group, because characters stable in one region m ay break down elsewhere. Family criteria into consideration: A family, similarly, represents a group of closely-related genera. Like genus, i t is also a very ancient concept because the natural groups now known as familie s, such as legumes, crucifers, umbels, grasses have been recognized by laymen an d taxonomists 1. The group, as far as possible, should be a natural one. The monophyletic nature of the group should be deduced Hierarchical Classification 55 alike for centuries. Ideally, families should be monophyletic groups. Like the g enus, the family may represent a single genus (Podophyllaceae, Hypecoaceae, etc. ) or several genera (Asteraceae: nearly 1100). Most taxonomists favour broadly-c onceived family concepts that lend stability to classification. Although there is no marked discontinuity between Lamiaceae (Labiatae) and Verbenaceae, the two are maintained as distinct families. The same traditio n prevents taxonomists from splitting Rosaceae, which exhibits considerable inte rnal differences. 56 Plant Systematics Chapter 4 Descriptive Terminology Any botanical analysis of a plant necessi-tates the availability of information about its characteristics. The descriptive information about the morphology of a plant (phytography) is suitably expressed in semitechnical language through a s et of terms, which provide an unambiguous representa-tion of the plant. The desc riptive terminol-ogy thus precedes any taxonomic or phylo-genetic analysis of a taxon. Whereas the veg-etative morphology of vascular plants (Tracheophyes) unif ormly includes informa-tion about the organs such as root, stem as leaves, the r eproductive morphology may dif-fer in different groups. The Pteridophytes are re presented by strobili, cones, sporophylls, microsporophylls, megasporophylls and spores, Gymnosperms by cones, megasporo-phylls, microsporophylls and seeds. The flow-ering plants have distinct inflorescences, flowers, seeds and fruits. All these organs show considerable variability, amply de-picted through a large voca bulary of descripHABIT AND LIFE SPAN Annual: A plant living and completing its life cycle in one growing season. Ephe merals are annuals surviving for one or two weeks (Boerhavia repens). Biennial: A plant living for two seasons, growing vegetatively during the first and flowering during the second. tive terms. Perennial: A plant living for more than two years and flowering several times du ring the life span (except in monocarpic plants which live for several years but perish after flow-ering, as in several species of Agave and bamboos). In herbac eous perennials, the aerial shoot dies back each winter, and the annual shoots a re produced from subaerial stock every year, those with a rhizome, tuber, corm o r bulb better known as geophytes. A woody perennial, on the other hand has woody aerial shoots which live for a number of years. A woody perennial may be a tree

(with a distinct trunk or bole from the top of which the branches arise deliques cent tree as in banyan, a totally unbranched caudex with a crown of leaves at to p as in palms, or the main stem continues to grow gradually narrowing and produc ing branches in acropetal order excurrent tree as in Polyalthia) or a shrub (with several distinct branches arising from the ground level). A Morphological terminology has been in use for description of species for several cen-turies and continues to be the principal source of taxonomic evidence. The descrip-tive terminology is very exhaustive, and as such only the most commonly used terms are illustrated here. Descriptive Terminology 57 suffrutescent plant is intermediate be-tween woody and herbaceous plants, with t he basal woody portion persisting year after year whereas the upper portion dies back every year. A weak climbing plant may be woody Assimilatory: Green chlorophyll-containing roots capable of carbon assimilation as in Tinospora cordifolia, and many species of Podostemaceae. (liana) or herbaceous (vine). Fibrous: Threadlike tough roots common in monocots, especially grasses, usually advenIt should be noted that the terms herb, shrub, suffrutescent plant and tree repr e-sent different forms of habit. Annual, bien-nial and perennial denote the life span or duration of the plant. titious in nature. Buttressed: enlarged, horizontally spread and vertically thickened roots at the base of certain trees of marshy areas. Fleshy: Thick and soft with a lot of storage tissue. Storage roots may be the mo dificaHABITAT grow in a variety of habitats. Terres-trial ppl ants ow on lnad, aQqu atic pplan ts in water and those on other plants as epi-phytes. Terrestrial plant may be a meso-phyte (growing in normal soil), xerophyte (growing on dry habitats: psammop hyte on sand, lithophyte on rock). An aquatic plant may be free-floating (occurr ing on water surtion taproot: (i)) Fusifom: Swollen in the middle and tapering on sides, as in radish (Raphanus sativus). (ii) Conical: Broadest on top and gradually narrowed below, as in carrot (Daucuscarota). (iii) Napiform: Highly swollen and almost globose and abruptly narrowed below, face), submerged or emersed (wholly under water), emergent (Anchored at bottom b ut with shoots exposed above water), floating-leaved (anchored at bottom but wit h float-ing leaves), or a helophyte (emergent marsh plant in very shallow waters ). A plant grow- ing in saline habitats (terrestrial or aquatic) is known as hal ophyte, whereas one in acidic soils as oxylophyte or oxyphyte. Saprophyte grow o n decaying organic mat-ter, parasite lives and depends on another organism. as in turnip (Brassica rapa). Modifications of the storage adventitious roots include: (i) Tuberous: Clusters of tubers growing out from stem nodes, as in sweet po- tato (Ipomoea batatas) and tapioca (Manihot esculenta). (ii) Fasciculated: Swollen roots occurring in clusters, as in Asparagus and some species of Dahlia. (iii) Nodulose: Only the apices of adventitious roots becoming swollen like beads, as in Curcuma amada and ROOTS Roots unlike stems lack nodes and intern-odes, have irregular branching and prod

uce endogenous lateral roots. Upon seed germi-nation, usually the radicle elonga tes into a primary root, forming a taproot, but several other variations may be encountered: Adventitious: Developing from any part Costus speciosus. (iv) Moniliform: Portions of a root are alternately swollen and constricted giving beaded appearance, as in Dioscorea alata. other than radicle or another root. Haustorial (sucking): Small roots penetrating the host xylem tissue for absorbin g water and nutrients as in partial parasites (Viscum) or also the photosyntheti c materials by penetrating the phloem tissue as well, Aerial: Grows in air. In epiphytes, the aerial roots termed epiphytic roots are found hanging from the orchids and are covered with a spongy velamen tissue. Orc hids also carry some clinging roots which penetrate crevices and help in anchora ge. as in total parasites (Cuscuta). Mycorrhizal: Roots infested with fungal mycelium which helps in root absorption. The fungal mycelium may penetrate cortical cells (endotrophic mycorrhizae found in 58 Plant Systematics it

H D N K Figure 4.1 Roots. A: Fusiform fleshy root of Raphanus sativus; B: Conical fleshy root of Daucus carota; C: Napiform fleshy root of Brassica rapa; D: Root-tuber of Ipomoea batat as; E: Fasciculated tuberous roots of Dahlia; F: Nodulose roots of Curcuma amada ; G: Moniliform roots; H: Pneumatophores of Avicennia; I: Stilt roots of Zea may s; J: Stilt roots of Pandanus; K: Prop roots of Ficus benghalensis; L: Aerial ro ots of Dendrobium; M: Haustorial roots of Viscum, sending haustoria only into th e host xylem; N: Mycorrhizal roots of Pinus orchids) or may largely form a mantle over the root with a few hyphae penetratin g be-tween the outer cells (ectotrophic mycor-rhizae found in conifers). In spec ialized VAM (vesicular arbuscular mycorrhizae) found in grasses, the fungal hyph ae penetrate cor-tical cells, forming a hyphal mass called large hanging prop roots of Ficus species are often used in bungee jumping sport . Stilt: Adventitious roots arising from the lower nodes of the plant and penetrat ing the soil in order to give increased anchorage as in maize (Zea mays), screwpines (Pandanus) and Rhizophora. arbusculum. Respiratory: Negatively geotropic roots of some mangroves (e.g. Avicennia) which grow vertically up and carry specialized lenticels (pneumathodes) with pores fo r gaseous ex-change. Such roots are also known as pneuSTEMS matophores. Stems represent the main axes of plants, being distinguished into nodes and inte rn-odes, and bearing leaves and axillary buds at the nodes. The buds grow out in to lateral shoots, inflorescences or flowers. A plant may lack stem (acaulescent ) or have a distinct stem (caulescent). The latter may be aerial (erect or weak) or even underground. Acaulescent: Apparently a stemless plant having very inconspicuous reduced stem.

The reduced stem may often elongate at the Prop: Elongated aerial roots arising from horizontal branches of a tree, strikin g the ground and providing increased anchorage and often replacing the main trun k as in several species of Ficus (e.g. the great banyan tree F. benghalensis in the Indian Botanical Garden at Sibpur, Kolkata). The Descriptive Terminology 59 time of flowering into a leafless flowering axis, known as scape as found in oni on. (iii) Axillary (lateral) bud: Bud located in the axil of a leaf. Arborescent: Becoming treelike and woody, usually with a single main trunk. (iv) Bulbil: Modified and commonly enAscending: Stem growing upward at about 45-60o angle from the horizontal. larged bud meant for propagation. In Agave and top onion (Allium x prolfierum) f lower buds get modified into bulbils. Bark: Outside covering of stem, mainly the trunk. Bark may be smooth, exfoliatin g (splitting in large sheets), fissured (split or cracked), or ringed (with circ ular fissures). Bud: Short embryonic stem covered with bud scales and developing leaves and ofte n found in leaf axils. Buds are frequently helpful in identification and may pre sent considerable diversity: (v) Dormant (winter) bud: Inactive well protected bud usually to survive winter in cold climates. (vi) Flower bud: Bud developing into flower. (vii) Mixed bud: A bud bearing both embryonic leaves and flowers. (viii) Naked bud: Not covered by bud scales. (ix) Pseudoterminal bud: Lateral bud near the apex appearing terminal due to death or non-development of terminal (i) Accessary bud: An extra bud on either bud. side (collateral bud) or above (super-posed bud or serial bud) the axillary (x) Scaly (covered) bud: Covered by bud scales. bud. (xi) Terminal bud: Located at stem tip. (ii) Adventitious bud: Bud developing from any place other than the node. (xii) Vegetative bud: Bearing embryonic leaves. A B F r C I Figure 4.2 Buds. A: Axillary bud with 2 collateral buds in Acer; B: Axillary bud and a superposed bud in Juglans regia; C: Scaly bud of Ficus covered with bud-scale; D: Winter bu ds in Salix; E: Terminal bud with two collateral buds; G: Intrapetiolar bud hidd en by petiole base; H: Same with petiole removed; I: Bulbil developing from one flower of Agave; J: Pseudoterminal bud, taking terminal position due to death or non-development of terminal bud; K: Vegetative bud of Brassica oleracea var. ca pitata (cabbage). 60 Plant Systematics

F E

H D T K Figure 4.3 Stem, subaerial and underground modifications. A: Tunicated bulb of A llium cepa; B: Same in vertical section, showing concentric layers of leaf sheaths; C: Scaly bu lb of Lilium with separate fleshy leaf sheaths; D: Stem tuber of Solanum tuberos um with eye buds; E: Rhizome of Zingiber officinale with fleshy branched horizon tal stem; F: Corm of Crocus sativus covered with scale leaves; G: Same in longit udinal section showing the solid inside as opposed to the bulb; H: Runner of Oxa lis, rooting at nodes; I: Stolon of Fragaria vesca, arching down to strike roots at nodes; J: Sucker in Chrysanthemu m, underground and rising up to produce shoot; K: Offset in Eichhornia crassipes , like runner but shorter and thicker. Caulescent: With a distinct stem. Caudiciform: Low swollen storage stem at groun d level, from which annual shoots arise as in Calibanus and some species of Subaerial: generally perennial partially hidden stems: (i) Runner: Elongated internodes trailing Dioscorea. along the ground and generally producing a daughter plant at its end as in Culm: Flowering and fruiting stem of grasses Cynodon and Oxalis. and sedges. (ii) Sobol: Like runner but partially unErect: Growing erect as an herb, shrub derground as in Saccharum spontaneum, and unlike rhizome, not or a tree. Lignotuber: Swollen woody stem at or below ground level, from which persistent w oody aerial branches arise, as in Manzanita. Pachycaul: Woody trunk-like stem swollen at base functioning for storage as in b ottle a storage organ. (iii) Stolon: Like runner but initially growing up and then arching down and striking roots in soil as in strawberry. (iv) Sucker: Like runner but underground tree Brachychiton. and growing up and striking roots to form new plant as in Chrysanthemum Phylloclade (cladophyll): Stem flattened and green like leaves bearing scale lea ves as in Opuntia. A phylloclade of one internode length found in Asparagus in k nown as and Mentha arvensis. (v) Offset: Shorter than runner and found in aquatic plants like Eichhornia crassipes. Cladode. Pseudobulb: Short erect aerial storage or propagating stem of certain epiphytic orchids. Subterranean (underground): Growing below the soil surface and often specially m odified: Descriptive Terminology 61 (i) Bulb: A reduced stem surrounded by Spine is like a thorn but generally weaker and developing from the leaf or stipu

le. Thorns may bear leaves (Duranta), flowers (Prunus), or may be branched (Cari ssa). thick fleshy scale leaves. The leaves may be arranged in a concentric man-ner su rrounded by a thin membranous scale leaf (tunicated bulb of onion Allium cepa) or leaves only overlapping along margins (scaly or imbricate bulb Weak: Plant not strong enough to grow erect: (i) Creeper: Growing closer to ground and often rooting at the nodes, as in Oxaof garlic Allium sativum) lis. (ii) Corm: A vertical fleshy underground (ii) Trailer: Trailing along the surface and stem covered with some scale leaves and with a terminal bud, as in Gladioften quite long. They are usually pros-trate or procumbent, lying flat on groun d as in Basella, but sometimes decumbent when the tips start growolus. (iii) Rhizome: A horizontal dorsiventral fleshy underground stem with nodes and internodes and covered with scale ing erect or ascending, as in Portulaca. (iii) Climber: Weak plant which uses a supleaves, as in Ginger. port to grow up and display leaves to-wards sunlight. This may be achieved (iv) Stem tuber: Underground portions of stem modifies into tubers as in potato. in a number of ways: Thorn: Branch or axillary bud modified into a hard sharp structure, being deep-s eated and having vascular connections as opposed to prickles which are mere supe rficial out-growths without vascular connections. (a) Twiner (stem climber): Stem coiling round the support due to special type of growth habit, as in Ipomoea and Convolvulus. (b) Root climber: Climbing with the help of adventitious roots which cling to the support, as in species of Piper. (c) Tendril climber: Climbing with the help of tendrils which may be modified stem (Passiflora, Vitis), modified inflor escence axis (Antigonon), modified leaf (Lathyrus aphaca), modified leaflets (Pi sum sativum), modified petiole (Clematis), modified leaf tip (Gloriosa), modifie d stipules (Smilax) or even modified root (Parthenocissus). (d) Scrambler: Spreading by leaning or resting on support, as in Rose. (e) Thorn climber: Climbing or reclining on the support with the help of thorns, as in Bougainvillea. (f) Hook climber: Climbing with the help of hooked structures (Galium). Figure 4.4 Stem, aerial modifications. A: Phylloclade of Opuntia; B: Cladodes in Asparagus; C: Portion enlarged to show whorl of cladodes in axil of scale-leaf; D: Phylloclades of Ruscus, leaf-like and bear ing flow-ers; E: Thorn of Prunus; F: Tendril of Luffa. LEAVES Leaves are green photosynthetic organs of a plant arising from the nodes. Leaves are usually flattened, either bifacial (dorsiventral) with adaxial side (upper surface facing stem 62 Plant Systematics E n D G Figure 4.5 Phyllotaxy of leaves. A: Rosulate; B: Alternate; C: Diagramatic repre sentation of

distchous (2-ranked) arrangement; D: Diagramatic representation of tristichous ( 3 ranked) arrangemt; E: Whorled leaves in Galium; F: Opposite and decussate leav es of Lamium; G: Opposite and superposed leaves of Quisqualis; H: Imbricated lea ves. axis) different from abaxial side (lower sur-face facing away from stem axis) or may be unifacial (isobilateral) with similar adaxial and abaxial surfaces. A le af is generally dif-ferentiated into a leaf blade (lamina) and a petiole. A leaf with a distinct petiole is termed petiolate, whereas one lacking a petiole is s essile. A petiole may be winged (Citrus), swollen (Eichhornia), modified into te ndril (Clematis), spine (Quisqualis) or become modi-fied into a flattened photos ynthetic phyllode (Australian Acacia). Two small stipules may be borne at the ba se of the petiole. The leaf terminology affords a wide diversity. The leaf base may sometimes be sheathing or pulvinate (swollen). all leaves are found to lie in a fixed number of vertical rows or orthostichies. The arrangement commonly agrees with the Fibonacci series (Schimper-Brown serie s), wherein numerator and denominator in each case are obtained by adding up the preceding two (1/2, 1/3, 1+1/2+3=2/5, 1+2/ 3+5=3/8, and so on) In grasses the l eaves are in two rows (2-ranked, distichous or 1/2 phyllotaxy), so that the thir d leaf is above the first leaf. Sedges have three rows of leaves (3-ranked, tris tichous, or 1/3 phyllotaxy), the fourth leaf above the first leaf. China rose an d banyan show pentastichous arrangement, where the sixth leaf lies above the fir st one, but in doing so leaves complete two spirals and the phyllotaxy is known as 2/5 phyllotaxy. Carica papaya depicts octastichous arrangement, wherein the n inth leaf lies above the first one and three spirals are completed in doing so, thus a 3/8 phyllotaxy. Leaf bases of date palm and sporophylls of pinecone are c losely packed Leaf arrangement (Phyllotaxy) Alternate: Bearing one leaf at each node. The successive leaves usually form a s piral pattern, in mathematical regularity so that Descriptive Terminology 63 and internodes are extremely short making it difficult to count the number of ro ws (orthostichies). Such an arrangement is A Simple leaf may be undivided or in-cised variously depending upon whether the incision progresses down to the midrib (pinknown as parastichous. nate) or towards the base (palmate): Imbricated: The leaves closely overlapping (i) Pinnatifid: The incision is less than one another, as in Cassiope. halfway towards the midrib. Opposite: Bearing pairs of leaves at each node. The pairs of successive leaves m ay be parallel (superposed) as in Quisqualis or at right angles (decussate) as i n Calotropis and (ii) Pinnatipartite: The incision is more than halfway towards the midrib. (iii) Pinnatisect: The incision reaches almost the midrib. Stellaria. (iv) Palmatifid: The incision is less than Whorled (verticillate): More than three leaves at each node as in Galium, Rubia and halfway towards the base. (v) Palmatipartite: The incision is more Nerium. than halfway towards the base of leaf Radical: Leaves borne at the stem base of-ten forming a rosette (rosulate) in re duced stems, as in Primula and Bellis. Cauline: Leaves borne on the stem. Ramal: Leaves borne on the branches. blade. (vi) Palmatisect: The incision reaches almost the base of leaf blade.

(vii) Pedate: Deeply palmately lobed leaves with lobes arranged like the claw of a Leaf duration bird. Leaves may stay and function for few days to many years, largely determined by t he adA compound leaf has incision reaching the midrib (or leaf base) so that there ar e more than one distinct blades called as leaflets or pinnae. It may similarly b e pinnate when the leaflets are borne separated along the rachis (cf. midrib of simple leaf) or pal-mate when the leaflets arise from a single point at the base . Pinnate compound leaves may be further differentiated: aptation to climatic conditions: Caducuous (fugacious): Falling off soon after formation, as in Opuntia. Deciduous: Falling at the end of growing sea-son so that the plant (tree or shru b) is leaf-less in winter/dormant season. In tropical climate, the tree may be l eafless for only a few days. Salix and Populus are common ex(i) Unipinnate (simple pinnate): The amples. leaflets are borne directly along the rachis. In paripinnate leaf (Cassia), the leaflets occur in pairs and as such the terminal leaflet is missing and there ar e even numbers of leaflets. In an imparipinnate (Rosa) leaf, on the other hand, there is a terminal leafEvergreen (persistent): Leaves persisting throughout the year, falling regularly so that tree is never leafless, as in mango, pines and palms. It must be noted that whereas the term persistent is used for the leaves, the term evergreen is c ommonly associated let, resulting in odd number of leaflets. with trees with such leaves. (ii) Bipinnate (twice pinnate): The pinMarcescent: Leaves not falling but wither-ing on the plant, as in several member s of Fagaceae. nae (primary leaflets) are again divided into pinnules, so that the leaflets (pi nnules) are borne on the primary branches of the rachis as in Mimosa Leaf incision/type of leaves A leaf with a single blade (divided or not) is termed simple, whereas one with t wo or more distinct blades (leaflets) is said to be compound. pudica. (iii) Tripinnate (thrice pinnate): The dissection goes to the third order so that the leaflets are borne on secondary bran ches of the rachis as in Moringa. 64 Plant Systematics Figure 4.6 Leaf incision. A: Undivided with pinnate venation; B: Pinnatifid; C: Pinnatipartite; D: Pinnatisect; E: Pinnate compound- imparipinnate leaf of Rosa; F: Pinnate comp ound-paripinnate leaf of Cassia; G: Bipinnate leaf of Acacia nilotica; H: Pinnat e-trifoliate leaf of Medicago, note middle leaflet with longer petiolule; I: Tri pinnate leaf of Moringa; J: Triternate leaf of Thalictrum; K: Undivided with pal mate venation; L: Palmatifid; M: Palmatipartite; N: Palmatisect; O: Palmate comp ound-digitate; P: Unifoliate leaf of Citrus; Q: Bifoliate; R: Trifoliate leaf of Trifolium, note all leaflets with equal petiolules as opposed to pinnate trifol iate leaf; S: Trifoliate leaf of Oxalis; T: Quadrifoliate leaf of Marsilea; U: p edate leaf of Vitis pedata. (iv) Decompound: Here the dissections go lower two leaflets are reduced and the terminal leaflet looks like a simple leaf but has a distinct joint at base, as beyond the third order, as in Fennel. The term is sometimes used for leaves more than once compound. seen in Citrus plants.

(v) Ternate: The leaflets are present in (ii) Bifoliate (binnate): A leaf with two leaflets, as found in Harduyickia. groups of three. Leaf may be ternate (pinnate with three leaflets, i.e. trifolia te), biternate (twice pinnate with three pinnae and three pinnules) (iii) Trifoliate (ternate): A leaf with three triternate or decompound ternate. leaflets, as in Trifolium. The trifoliate leaf of Medicago and Melilotus has ter minal leaflet with a longer petiolule (stalk of leaflet) than basal leaflets and Palmate compound leaf does not have a ra-chis and the leaflets arise from the to p of is accordingly a pinnate trifoliate leaf. the petiole: (iv) Quadrifoliate: A leaf with four leaflets, (i) Unifoliate: A modified situation in commonly a trifoliate leaf when the as in Paris and aquatic pteridophyle Marsilea. Descriptive Terminology 65 (v) Multifoliate (Digitate): A leaf with more than four leaflets, as in Bombax. Hastate: Shaped like an arrow head with two basal lobes directed outwards, as in Typhonium; also referring to hastate leaf Stipules The leaves of several species bear two small stipules as outgrowths from the lea f base. Leaves with stipules are termed stipulate and those without stipules as exstipulate. They show a lot of structural diversity: Free-lateral: Free and lying on either side of the petiole base, as in china-ros e (Hibisbase. Lanceolate: Shaped like a lance, much longer than broad and tapering from a broa d base towards the apex, as in bottle-brush plant (Callistemon lanceolatus). Linear: Long and narrow with nearly parallel sides as in grasses and onion. Lunate: Shaped like half-moon, as in cus rosa-sinensis). Passiflora lunata. Adnate: Attached to the base of petiole for Lyrate: Lyre-shaped; pinnatifid with large terminal lobe and smaller lower lobes , as in some distant, as in Rose. Intrapetiolar: The two stipules are coher-ent to form one, which lies in the axi l of a Brassica campestris. leaf as in Gardenia. Oblanceolate: Like lanceolate but with broadest part near apex. Interpetiolar: A stipule lying between the petioles of two adjacent leaves, comm only due to fusion and enlargement of two adja-cent stipules of different leaves as found in several members of Rubiaceae like Ixora. Ochreate: The two stipules united and form-ing a tubular structure ochrea, found in famObcordate: Like cordate but with broadest part and notch at apex, as in Bauhinia . Oblong: Uniformly broad along the whole length as in banana. Obovate: Ovate, but with broadest part near the apex, as in Terminalia catappa. ily Polygonaceae. Ovate: Egg-shaped, with broadest part near the base, as in Sida ovata. Foliaceous: Modified and enlarged to func-tion like leaves as in Lathyrus aphaca , where the whole leaf blade is modified into tendril Orbicular (rotund): Circular in outline. The peltate leaf of Nelumbo is orbicula r in out-

line. and stipules are foliaceous. Pandurate: Fiddle shaped; obovate with sinus or indentation on each side near th e base and with two small basal lobes, as in Tendrillar: Stipules modified into tendrils as in Smilax. Spiny: Stipules modified into spines as in Jatropha panduraefolia. Acacia. Peltate: Shield shaped with petiole attached to the lower surface of leaf (and n ot the marLeaf shape (outline of lamina) The shape of leaf/leaflet blade shows con-siderable variability and is of major taxogin), as in Nelumbo. Reniform: Kidney-shaped, as Centella asiatica. nomic value. Runcinate: Oblanceolate with lacerate or Acicular: Needle shaped, as in pine. Cordate: Heart shaped, with a deep notch at base, as in Piper betle. parted margin, as in Taraxacum. Sagittate: Shaped like an arrowhead with two basal lobes pointed downwards, as i n Sagittaria and Arum; also referring to sagitCuneate: Wedge-shaped, tapering towards the base, as in Pistia. tate leaf base. Deltoid: Triangular in shape. Elliptical: Shaped like an ellipse, a flattened circle usually more than twice a s long as broad, as in Catharanthus roseus. Spathulate (spatulate): Shaped like a spatula, broadest and rounded near the ape x, gradually narrowed towards the base, as in Euphorbia neriifolia. 66 Plant Systematics A

0 A , I~ F D E o j I

0 ` ~

1:5/1X-1 J " 0 N 1 Q Figure 4.7 Leaf outline. A: Acicular; B: Subulate; C: Linear, common in grasses; D: Lanceolate; E: Oblong; F: Spathulate; G: Cordate; H: Ovate; I: Obovate; J: Oblanceolate; K: Peltate; L: Reniform; M: Hastate; N: Runcinate; O: Lunate; P: Sagittate; Q: Pand urate; R: Deltoid; S: Lyrate; T: Elliptic. Subulate: Awl-shaped, tapering from a broad base to a sharp point. Double serrate (bi-serrate): The serrations are again serrate similarly as in Ul mus. Entire: Smooth, without any indentation, as Leaf margin The edge of a leaf blade is known as margin and may show any of the following co nditions: Crenate: With low rounded or blunt teeth, in Mango. Retroserrate: Teeth pointed downwards. Revolute: Margin rolled down. Serrate: With sharp teeth pointing upward as in Kalanchoe. like saw, as seen in rose. Crisped: Margin strongly winding in vertical plane giving ruffled appearance to leaf. Dentate: With sharp teeth pointing outSerrulate: Minutely or finely serrate. Sinuate: Margin winding strongly inward a s well as outward. wards. Undulate (repand, wavy): Margin winding gradually up and down and wavy, as in Po lyalthia. Denticulate: Minutely or finely dentate. Double crenate (bi-crenate): Rounded or blunt teeth are again crenate Leaf base In addition to the terms cordate, cuneate, hastate, sagittate already described above when referring to the leaf base, the following additional terms are freque ntly used: Double dentate: Sharp outward teeth are again dentate. The term bi-dentate, thou gh sometimes used here, is inappropriate, as it more correctly refers to a struc ture bear-ing two teeth. Descriptive Terminology 67 4

F /s

D F H I C K Figure 4.8 Leaf margin. A: Entire; B: Crenate; C: Crenulate; D: Dentate; E: Dent iculate; F: Serrate; G: Serrulate; H: Bi-serrate; I: Undulate; J: Sinuate; K: Cr ispate. Amplexicaul: The auriculate leaf base comLeaf apex pletely clasps the stem. Leaf apex may similarly present a number Attenuate: Showing a long gradual taper toof diverse terms: wards the base. Acute: Pointed tip with sides forming acute Auriculate: With ear like appendages at the angle, as in mango. base, as in Calotropis. Acuminate: Tapering gradually into a protracted point, as in Ficus religiosa. Ar istate: With a long bristle at the tip. Attenuate: Tip drawn out into a long tap erCuneate: Wedge shaped, with narrow end at the point of attachment. Decurrent: Extending down the stem and adnate to the petiole. ing point. Oblique: Asymmetrical with one side of the Caudate: Apex elongated and tail-like. Cirrhose: With slender coiled apex, as in blade lower on petiole than other. Perfoliate: The basal lobes of leaf fusing so that the stem appears to pass thro ugh the leaf, as in Swertia. When the bases of two opposite leaves fuse and the stem passes through them, it is termed connate perfolibanana. Cuspidate: Abruptly narrowed into sharp spiny tip, as in pineapple. Emarginate: With a shallow broad notch at tip, as in Bauhinia. ate as seen in Canscora. Mucronate: Broad apex with a small point, Rounded: With a broad arch at the base. Truncate: Appearing as if cut straight a cross. as in Catharanthus. Obtuse: Broad apex with two sides forming an obtuse angle, as in banyan. 68 Plant Systematics

R ( E F G n

R S Figure 4.9 Leaf apex and leaf base. Leaf apex. A: Acute; B: Acuminate; C: Arista te; D: Caudate; E: Emarginate; F: Retuse; G: Rounded; H: Mucronate; I: Truncate; J: Obtuse; K: C irrhose. Leaf base. L: Attenuate; M: Amplexicaul; N: Connate-perfoliate; O: Perfoliate; P: Cuneate; Q: Auriculate; R: Cordate; S: Truncate; T: Decurrent. Retuse: With a slight notch generally from an obtuse apex, as in Crotakaria retu sa. Truncate: Appearing as if cut straight across as in Paris Lanate: Wooly, with long intertwined hairs. Pilose: Covered with long distinct a nd scattered hairs. Puberulent: Minutely pubescent. Pubescent: Covered with soft short hairs. Leaf surface The surface of leaves, stems and other or-gans may present a variety of surface indumentation, whose characteristics are highly diagnostic in several taxa. The sur-face may be covered by trichomes (hairs, glands, scales, etc.) arranged vari ously: Arachnoid: Covered with entangled hairs giving a cobwebby appearance. Canescent: Covered with grey hairs. Ciliate: With marginal fringe of hairs. Floccose: Covered with irregular tufts o f Rugose: With wrinkled surface. Scabrous: Surface rough due to short rough points. Scurfy: Covered with scales. Sericeous: Covered with soft silky hairs, all directed towards one side. Stellate: Covered with branched star-shaped hairs. Strigose: Covered with stiff appressed hairs pointing in one direction. Tomentose: Covered with densely matted soft hairs, wooly in appearance. loosely tangled hairs. Velutinous: Covered with short velvety Glabrate: Nearly glabrous or becoming glahairs. brous with age Villous: Covered with long, fine soft hairs, Glabrous: Not covered with any hairs. Some-times but not always synonymous with shaggy in appearance.

smooth surface. The hairs covering the surface may be unicellular or multicellular, glandular or nonglandular. The hairs may be unbranched or branched variously. They may bear one row of cells (uniseriate), two rows (biseriate) or several rows (multiseriat e). Some species of plants, especially some acacias bear specialized glands doma tia at the leaf base, which house ants which protect plants from herbivores. Glaucous: Surface covered with a waxy coating, which easily rubs off. Glandular: Covered with glands or small secretory structures. Glandular-punctate (gland-dotted): Surface dotted with immersed glands, as in Ci trus. Hirsute: Covered with long stiff hairs. Hispid: Covered with stiff and rou gh hairs. Descriptive Terminology 69

E B D H'

H J N I M O K L Figure 4.10 Surface coverings. A: Arachnoid; B: Ciliate; C: Floccose; D: Glandul ar; E: Hirsute; F: Hispid; G: Pilose; H: Puberulent; I: Rugose; J: Scabrous; K: Sericeous; L: St ellate; M: Strigose; N: Tomentose; O: Villous. Venation monotelic) inflorescence, on the other hand, the main axis has limited growth, b eing ter-minated by the formation of a flower, and as each level of branching be ars one flower, there are generally a limited number of flow-ers, and the oldest flower is either in the centre, or flowers of different ages are mixed up. An i nflorescence is sometimes carried on a leafless axis. Such a leaf less axis aris ing from aerial stems is termed a peduncle (inflorescence pedunculate) and the o ne arising from basal rosette of leaves as scape (inflorescence scapigerous). The distribution of vascular bundles that are visible on the leaf surface as vei ns consti-tutes venation. Dicots exhibit a network of veins (reticulate venation ); whereas mono-cots usually have non-intersecting parallel veins (parallel vena tion). Each type of ve-nation may encounter a single midrib from which the secon dary veins arise (Unicostate or pinnate), or more than one equally strong veins entering the leaf blade (multicostate or palmate). In ferns and Ginkgo, the vena -tion is dichotomous with forked veins. INFLORESCENCE Racemose types The following variations of the racemose type are commonly encountered: Inflorescence is a modified shoot system bearing flowers (modified shoots). The

term inflorescence appropriately refers to the ar-rangement of flowers on the pl ant. The flow-ers may either occur singly (in leaf axils Raceme: A typical racemose inflorescence with single (unbranched) axis bearing f lowers on distinct pedicels, as in Delphinium. Panicle: Branched raceme, the flowers be-ing borne on the branches of the main a xis, solitary axillary or terminal on the stem solitary terminal) or may be organized into distinct inflorescences. Two principal types of inflorescences are differen tiated. In race-mose (indeterminate or polytelic), inflores-cence the axis is of unlimited growth, api-cal bud continuing to grow, thus bearing old-est flower t owards the base and youngest to-wards the top. In cymose (determinate or as in Yucca. Spike: Similar to raceme but with sessile flowers, as in Adhatoda. Spadix: Variation of a spike where the axis is fleshy and the flowers are covere d by a large bract known as spathe, as found in Alocasia and Arum. 70 Plant Systematics Corymb: Flat-topped racemose inflorescence with longer lower pedicels and shorte r up-per pedicels so that all flowers reach the more determinate branches arising below the terminal flower at each level: same level, as in Iberis amara. Monochasial (Uniparous) cyme: One branch arising at each node so that when the s ympodial (false) axis differentiates, a limited number of bract-opposed flowers (in-stead of many and axillary in raceme) are Corymbose-raceme: Intermediate between a typical raceme and a typical corymb, al l flowers not managing to reach the same height, as in Brassica campestris. formed. Two types of monochasia are found: Catkin (ament): A spike-like inflorescence of reduced unisexual flowers, as in M orus. Umbel: Flowers arising from one point due to condensation of axis, with ol dest flowers towards the periphery and youngest towards the center as in the fam ily Apiaceae (Umbelliferae). Compound umbel has branches bearing the umbels also borne in (i) Helicoid cyme: Successive branches (each forming one flower) are borne on same side so that the inflorescence is of ten coiled, as in the family Boraginaceae (e.g. Myosotis). (ii) Scorpioid cyme: Successive branches umbellate manner. (each forming one flower) are borne on alternate sides. In rhipidium found in So lanum nigrum, all the flowers lie in same plane as the main axis. Dichasial (Biparous) cyme: Two branches arising below the apical flower at each level so that the flower is between the fork of two branches, as in Stellaria an d Dianthus. Polychasial (multiparous) cyme: More than two branches arising at each node belo w the terminal flower so that a broad inflorescence of several flowers is formed , as in Viburnum. Cymose cluster: Cymose group of flowers arising from a point d ue to reduction of axis. Cymose umbel: Looking like an umbel but formed by group ing together of numerous Head: Flat-topped axis bearing crowded sessile flowers as in Acacia and Mimosa. Capitulum: Flat-topped inflorescence like head (and often known as head) but wit h distinct ray florets and disc florets (one or both types), surrounded by invol ucre bracts (phyllaries), as found in the family Asteraceae (Compositae). Cymose types A cymose inflorescence may be primarily dif-ferentiated on account of bearing on e or I t

B C Figure 4.11 Inflorescence: racemose types. A: Raceme of Linaria; B: Corymbose ra ceme of Brassica; C: Corymb of Cassia; D: Panicle of Yucca; E: Umbel of Prunus; F: Compound umbel of Foeniculum; G: Catkins of Betula; H: Spike of Achyranthes; I: Spadix of Descriptive Terminology 71 E t' Figure 4.12 Inflorescence: cymose and specialized types. Cymose types. A: Helico id cyme of Heliotropium; B: Scorpioid cyme of Ranunculus bulbosus; C: Biparous cyme of Dian thus; D: Multiparous cyme of Viburnum. Specialized types. E: Verticillaster of S alvia; F: Cyathium of Euphorbia; G: Hypanthodium of Ficus cunia. cymes so that the flowers of different ages are mixed up, as found in Allium. FLOWER A flower is a highly modified shoot bearing specialized floral leaves. The axis of the flower is condensed to form thalamus (torus or receptacle) commonly beari ng four whorls of floral parts: calyx (individual parts sepals), corolla (indivi dual parts petals), Androecium (individual parts stamens) and Gynoecium (individ ual parts carpels). In some plants, the calyx and corolla may not be differentia ted and represented by a single or two similar whorls of perianth (individual me mbers tepals: a term formerly restricted to petal like perianth of monocots). Th e flower is usually carried on a pedicel and may or may not be subtended by a re duced leaf known as bract. The pedicel may sometimes carry small bracteoles (if present usually two in dicots, one in monocots). As a general rule, members of d ifferent whorls alternate each other. The terms associated with the general desc ription of flower in usual sequence includes: Specialized types In addition to the typical determinate and indeterminate types, some mixed and s pe-cialized types are also encountered: Cyathium: Complex type of inflorescence met in genus Euphorbia, having a cup-sha ped involucre (formed by fused bracts) usually carrying five nectaries along the rim and enclosing numerous male flowers (in scorpioid cymes, without perianth a nd bear-ing a single stamen) in axils of bracts and single female flower in the centre. Verticillaster: Characteristic inflorescence of family Lamiaceae. Each node of t he inflorescence bears two opposite clusters of dichasial cymes, subsequently be coming monochasial as the number of flowers in each cluster exceeds three. Due t o the con-densation of the axis, flowers of different ages appear to form a fals e whorl or verticel. Hypanthodium: Typical inflorescence of figs having vessel l ike receptacle with a small opening at the top and bearing flowers along Bract Bracteate: Flower in the axil of a bract. Ebracteate: Bract absent. the inner wall. Bracteolate: Bracteoles present on pedicel. Thyrse: A mixed inflorescence with race-mose main axis but with cymose lateral c lus-ters as seen in grape vine. Pedicel Pedicellate: Pedicel distinct, often longer than flower. 72 Plant Systematics

} F i

Figure 4.13 Insertion of floral parts. A: Hypogynous with superior ovary; B: Per igynous with cupshaped hypanthium and superior ovary; C: Perigynous with flask-shaped hypanthium , ovary superior; D: Perigynous with partially immersed semi-inferior ovary; E: Epigynous with inferior ovary, without free hypanthium above the ovary; F: Epigy nous with inferior ovary and with free hypanthium above the ovary. Subsessile: Pedicel much shorter, often shorter than flower. all parts of perianth) more or less of the same shape and size. Sessile: Pedicel absent. Zygomorphic: Asymmetrical flower, which may be divided into equal halves by one or more but not all vertical planes. In practice such flower has parts of calyx and/or corolla (or perianth) of different Complete: All the four floral whorls present. Incomplete: One or more floral whorl lacking. Symmetry: Symmetry of a flower is largely based on relative shapes and sizes of sepals (or calyx lobes) in calyx whorl and/or rela-tive shapes and sizes of peta ls (or corolla lobes) in the corolla whorl. shapes and sizes. Sexuality Bisexual (perfect): Bearing both stamens and carpels. Unisexual (imperfect): Bearing either Actinomorphic: Symmetrical flower which can be divided into equal halves when cu t along any vertical plane. In prac-tice an actinomorphic flower has all parts o f the calyx and all parts of the corolla (or stamens or carpels. Staminate (male): Bearing stamens only. Pistillate (female): Bearing carpels onl y. Dioecious: With male and female flowers on the different plants. Descriptive Terminology 73 Monoecious: With male and female flowSpirocyclic: Calyx and corolla cyclic but stamens and carpels spirally arranged, as in Ranunculaceae. ers on the same plant. Polygamous: With male, female and bisexual flowers on the same plant. Insertion: Insertion of floral parts on the thalamus not only determines the sha pe of the thalamus, it also reflects on the rela-tive position of floral whorls, as also whether the ovary is superior (and, consequently, other whorls inferior ) or inferior (and, conCalyx Description of the calyx starts with the number of sepals in same whorl (5 typical on dicots, 3 typical of monocots), in two whorls (2+2, as in crucifers) or formin g two lips (1/ sequently, other whorls superior): 4 in Ocimum, 3/2 in Salvia):

Hypogynous: The thalamus is convex so that the other floral parts are inserted b e-low the ovary. The ovary in this case is superior and other floral whorls infe rior. Polysepalous (aposepalous, chorisepalous): Sepals free, and consequently more than one units (poly many). There is no hypanthium. Gamosepalous: Sepals fused. Once the ca-lyx is gamosepalous, it commonly gets di fferentiated two parts: calyx tube, the fused part and calyx lobes (no longer se pals), the free part. The shape of the calyx tube should be described. It may be campanulate (bell-shaped as in Hibiscus), urceolate (urn-shaped as in fruiting calyx of Withania ), tubular (tube-like as in Datura), or bilabiate Perigynous: The thalamus is depressed to the extent that the level of ovary is l ower than the other whorls and the thalamus forms either a saucer-shaped, cup-sh aped or flask-shaped hypanthium. It must be noted that although hypanthium sur-r ounds the ovary, it is free from the ovary, the other floral whorls are borne al ong the rim of the hypanthium, yet the ovary is morphologically still superior a nd other flo-ral whorls inferior. The ovary may some-times be partially immersed and thus semi-inferior. (two-lipped as in Ocimum) Caducous: Falling just after opening of flowers. Deciduous: Falling along with petals in mature flower. Epigynous: the hypanthium is fused with the ovary, so that the other floral whor ls appear to arise from the top of the ovary. The ovary is obviously inferior an d other floral whorls superior. There may or may not be a free hypanthium above the ovary; in the former case, other floral parts apPersistent: Persisting in fruit. Accrescent: Persisting and enlarging in fruit. Aestivation: Arrangement of sepals (or pet-als) in the flower bud. Term vernatio n is used exclusively for arrangement of young leaves in a bud. The following ma in types of aestivation are met: pear to arise from the top of ovary. (i) Valvate: Margins of sepals or calyx Pentamerous: Five members in each floral whorl (excluding stamens and carpels), typilobes not overlapping. (ii) Twisted: Overlapping in regular patcal of dicots. tern, with one margin of each sepal overlapping and other being overTetramerous: Four members in each floral whorl, as in crucifers. lapped. Trimerous: Three members in each floral (iii) Imbricate: With irregular overlapping. whorl, as in monocots. In Quincuncial imbricate, two sepals are with both margins outer, two with both margins inner, and fifth with one outer and one inner margin. Cyclic (tetracyclic): Calyx, corolla, androecium and gynoecium in four separate whorls. 74 Plant Systematics A B C E Figure 4.14 Aestivation of calyx and corolla parts. A: Valvate; B: Twisted; C: I mbricate; D: Quincuncial imbricate; E: Vexillary.

Description of aestivation may be followed by colour of sepals (green or petaloi d), and whether they are inferior or superior. variously coloured. In some cases, sepals or petals may bear a small pouch a con dition known as saccate (lateral sepals of Brassica or corolla of Cypripedium is more like slip-per and called calceolate). Sometimes the base may be produced in to a tube like structure known as spur (corolla as calcarate) as in Delphinium a nd Aquilegia. In some flow- ers (Aconitum), the corolla may be shaped like Corolla Description of the corolla follows the same pattern as calyx except that bilabia te corolla may be 4/1 or 2/3, corolla may be polypeta-lous (apopetalous, choripe talous), or gamo-petalous (sympetalous), corolla tube may be additionally infund ibuliform (funnel- shaped) as in Datura, rotate (tube very short with large lobe s spreading out at right angle to the tube like spokes of a wheel), as in Solanu m, or salverform (salver-shaped, hypocrateriform), as in Catharanthus. The junct ion of corolla tube and lobes (constituting limb) is known as throat. Petals may sometimes be narrowed into a stalk termed as claw, the broader part then consti tuting the limb. Specialized types of corolla are encountered in Brassicaceae (c ruciform four free petals arranged in the form of a cross), Caryophyllaceae (caryo phyllaceous five free clawed petals with limb at right angles to the claw), Rosace ae (rosaceous five sessile petals with limbs spreading out-wards) and Fabaceae (Pa pilionaceous re-sembling a butterfly with single large poste-rior petal vexillum o r standard, two lateral petals alae or wings, and two anterior pet-als slightly united to form keel or carina; the aestivation is vexillary or descending imbric ate, with the standard being the out-ermost, overlapping two wings, which in tur n overlap keel). The petals may similarly be a helmet, when it is termed as galeate. Present inner to corolla in some cases is an additional whorl generally attached to the throat of the corolla (or inner whorl of the perianth). Such a whorl is known as corona and may be consisting of appendages from perianth (Narcissus), c orolla (corolline corona as in Nerium) or from stamens (staminal co-rona as in H ymenocallis). The flower is known as coronate. Perianth The description of perianth in the flowers lacking distinct calyx and corolla fo llows the same pattern specifying the number, number of whorls, perianth being p olyphyllous (apotepalous) or gamophyllous (syntepalous), aestivation, and the co lour of the perianth. The parts when free are called tepals in place of sepals o r petals. Androecium Stamens representing the androecium present a more complicated architecture as c ompared to sepals and petals. Each stamen has an anther typically tetrasporangiate with two anther sacs (microsporangia) in Descriptive Terminology 75 A D E I K L Figure 4.15 Corolla types. A: Cruciform; B: Papilionaceous; C: Caryophyllaceous; D: Rosaceous; E: Campanulate; F: Tubular; G: Infundibuliform (Funnel-shaped); H: Hypocraterifo rm; I: Urceolate; J: Bilabiate; K: Spurred (Calcarate); L: Coronate; M: Personat e; N: Calceolate; O: Galeate. each of the two anther lobes , carried on a filament. The two anther lobes are oft en joined with the help of a connective, which in some primitive families, is a continua-tion of the filament. The description of androecium, likewise, starts w ith the num-ber of stamens in a single or more whorls. Epipetalous: Filaments attached to the petals, a characteristic feature of sympetalous families. Epiphyllous (epitepalous): Filaments at-

tached to the perianth. Major descriptive terms include: Relative size: Stamens in a flower are generally of the same size, but the follo wing variations may be encountered in some Fusion: Stamens may generally be free, but flowers: if fused it can take a variety of forms: Didynamous: Four stamens, two shorter Polyandrous: Stamens free throughout. Monadelphous: Filaments of all stamens united in a single group, as in family and two longer, as in Ocimum. Tetradynamous: Six stamens, two shorter in outer whorl and four longer in inner Malvaceae. whorl, as in crucifers. Diadelphous: Filaments of stamens united in two groups, as in Lathyrus. Polyadel phous: Filaments united in more Heterostemonous: Same flower with stamens of different sizes, as in Cassia. than two groups, as in Citrus Diplostemonous: Stamens in two whorls, the outer whorl alternating with petals a s Syngenesious (synantherous): Filaments free but anthers connate into a tube, as in Murraya. in family Asteraceae. Obdiplostemonous: Stamens in two whorls but outer whorl opposite the petals, as in the Synandrous: Stamens fused completely through filaments as well as anthers, as in Cucurbita. family Caryophyllaceae. Antipetalous: Stamens opposite the petals, as in the family Primulaceae. 76 Plant Systematics

I A C P D l 0 H M L 0 J Figure 4.16 Androecium types. A: Epipetalous staments. Length. B: Didynamous; C: Tetradynamous; D: Heterostemonous. Fusion. E: Diadelphous; F: Monadelphous; G: Syngenes ious. Attachment. H: Adnate; I: Basifixed; J: Dorsifixed; K: Versatile. Dehiscen ce. L: Longitudinal; M: Transverse; N: Poricidal; O: Valvular. P: Monothecous re niform anther. Bithecous: Stamen with two anther lobes (each anther lobe at maturity becomes un ilocular due to coalescence of two adja-cent microsporangia) so that anther is t wo(iv) Versatile: Filament attached nearly at the middle of connective so that anther can swing freely as, in Lilium and grass es.

celled at maturity. Dehiscence: Anther dehiscence commonly occurs by the formation of sutures along the point of contact of two anther sacs, but con-siderable variation in their lo cation may be Monothecous: Stamen with single anther lobe so that mature anther is single-cell ed, as in family Malvaceae. Attachment: Common modes of attachment found: of filament to the anther include: Longitudinal: The two sutures extend longitudinally, one on each anther lobe as in Datura. (i) Adnate: Filament continues into connective which is almost as broad, as found in Ranunculus. Transverse: Suture placed transversely, as in monothecous anthers of family (ii) Basifixed: The filament ends at the base of anther (when connective ex-tends up to base of anther) or at least base of connective (when anther lobes extend freely below the connective). The result ant anther is erect, as in Malvaceae. Poricidal (apical pores): Anther opening by pores at the tip of anther, as in So lanum nigrum. Brassica. Valvular: Portions of anther wall opening through flaps or valves, as in Laurus. (iii) Dorsifixed: Filament attached on the connective above the base. The result-ant anther is somewhat inclined, as in Ses bania. Centripetal: Developing from the outside to the inside so that the oldest stamen s are towards the periphery. Descriptive Terminology 77 Centrifugal: Developing from centre to-wards the periphery, so that the oldest f lownumber of rows of ovules (placental lines) would equal the number of united carp els. A solitary carpel would obviously have a single chamber with a single ovule or a single row of ovules. On the other hand, if ovary is more than one chamber ed, it obviously has more than one carpels, and the number of chambers would ind icate the number of carpels. There are, however, atypical cases. Single chambere d ovary may have a central column bearing ovules (since septa disappeared), or i n a single chambered ovary there may be single large ovule because all others (f rom one or more placental lines) have disappeared. In both these cases, the numb er of carpels can be known by counting the number of free styles, or if style is one the number of stigmas or stigmatic lobes. In extreme cases, even this may n ot help, as in Anagallis aruensis, when the number of suture lines on the fruit would help. The number of carpels are represented as monocarpellary (carpel one) , bicarpellary (carpels two), tricarpellary (carpels three), tetracarpellary (ca rpels four), pentacarpellary (carpels five), and multicarpellary (carpels more t han five). The number of chambers similarly are represented as unilocular, biloc ular, trilocular, tetralocular, pentalocular and multilocular. Gynoecium with fr ee carpels is apocarpous, whereas one with fused carers are towards the centre. Included: Stamens are shorter than the corolla. Exserted: Stamens protruding far beyond the petals as in Umbellifers. Introrse: Slits of the anther facing towards the centre. Extrorse: Slits of the anther facing towards outside.

Androphore: Extension of thalamus bearing stamens. Gynostegium: Structure formed by the fu-sion of stamens with the stigmatic disc, as in family Asclepiadaceae. Gynostemium: Structure formed by fusion of stamens with gynoecium, as in family Orchidaceae. Gynoecium Gynoecium represents a collection of car-pels in a flower. The distinction betwe en carpel and pistil is often ambiguous. In reality the carpels are components o f a gynoecium whereas the pistils represent visible units. Thus, if carpels are free, there would be as many pistils (simple pistils). On the other hand, if the carpels are united (and obviously more than one), the flower would have only on e pistil (compound pistil). Each carpel is differentiated into a broad basal ova ry containing ovules, an elongated style, and pollen-receptive apical part stigm a. Any attempt to describe gynoecium requires a transverse section through the o vary. An additional longitudinal section is always helpful. B I c E Carpel number and fusion D A flower having more than one separate pistils would have as many carpels, which are free. On the other hand, if the pistil is one, there could either be one ca rpel, or more than one fused carpels. A section through the ovary helps to resol ve the matter in most cases. If the ovary is single chambered, the Figure 4.17 Carpel fusion. A: Apocarpous; B: Apocarpous with fused styles and stigmas (which, in turn, also fused with anther s to form gynostegium); C: Syncarpous with free styles and stigmas (synovarious) ; D: Syncarpous with free stigmas (synstylovarious); E: Syncarpous. 78 Plant Systematics A B C D t E G H I Figure 4.18 Placentation. A: Marginal; B: Parietal with 3 carpels; C: Parietal w ith false septum in crucifers (parietal-axile); D: Parietal with false septa in cucurbits; E: Basal; F: Apical; G: Axile; H: Axile with false septa in Datura; I: Free central with usual c entral column attached at the base and top of the ovary; J: Free central in Prim ulaceae in Longitudinal section showing placental column projecting from the bas e; K: Superficial in Nymphaea. pels (at least ovaries fused) as syncarpous. Syncarpous gynoecium may have free styles and stigma (synovarious) or free stigmas (synstylovarious) or all fused. parietal-septate. In family Cucurbitaceae, the three parietal placentae intrude into ovary cavity and often meet in the centre making

false-axile placentation. Placentation Placentation refers to the distribution of pla- centae on the ovary wall and, co nsequently, the arrangement of ovules. The following major types are found: (iii) Axile: Ovary more than one chambered and placentae along the axis as in Hibiscus. (iv) Free-central: Ovary single chambered, ovules borne along the central column, as in family Caryophyllaceae. (i) Marginal: Single chambered ovary (v) Basal: ovary single chambered, with with single placental line commonly with single row of ovules, as in single ovule at the base, as found in family Asteraceae (Compositae). Lathyrus. (vi) Superficial: Multilocular ovary with (ii) Parietal: Single chambered ovary with whole inner wall of ovary lined with placentae as in Nymphaea. In laminar placen tation, the ovules arise from surface of septa. more than one discrete placental lines as, in family Capparaceae. In family Bras sicaceae, the ovary later becomes bilocular due to the formation of a false sept um, the ovules present at the junction of septum and ovary wall, a condition oft en known as parietal-ax-ile. In some members of Aizoaceae, the ovules arise from inner ovary walls of septate ovary, a condition known as Style and Stigma Simple: Single style or stigma resulting from single carpel or fused styles or s tigmas. Bifid: Style or stigma divided into two as in family Asteraceae. Descriptive Terminology 79 Plumose stigma: Feathery stigma with trichome-like branches as in Poaceae and Cy peraceae. Ovule Ovule represents megasporangium, attached to the placenta by funiculus, which jo ins the ovule at the hilum. Base of the ovule is known as chalaza, and the tip a s micropyle. Ovule has a female gametophyte (embryo sac) surrounded by nucellus, in turn, enveloped by two integuments. The following terms are commonly associa ted with ovules: Figure 4.19 Style and stigma. A: Lateral style; B: Gynobasic style; C: Bifid fea thOrthotropous (atropous): Straight erect ovule with funiculus, chalaza and micropyle in one line, as in family Polygonaceae. Anatropous: Inverted ovule with mic ropyle facing and closer to funiculus, as in ery stigma in Poaceae; D: Sessile and radiate stigma of Papaver; E: Tripartite f unnel-shaped stigma of Crocus; F: Capitate stigma of Alchemilla; G: Discoid stig mas of Hibiscus; H: Bifid stigma in Asteraceae. Ricinus. Amphitropous: Ovule placed at right angles to the funiculus, as in Ranunculus. C ampylotropous: Curved ovule so that micropyle is closer to chalaza, as in Brassicaceae. Terminal style: Arising from the tip of ovary, Circinotropous: Funiculus very long and surrounding the ovule, as in Opuntia. He mianatropous (hemitropous): Body half-inverted so that funiculus is attached nea r middle with micropyle terminal and the most common type. Gynobasic style: Arising from central base of the ovary, as in family Lamiaceae. Capitate: Stigma appearing like a head. Lateral style: Style arising from the side of the ovary, as in Mangifera and Alc hemilla. at right angles. Bitegmic: Ovule with two integuments, Stylar beak: Persistent style, extended into

common in polypetalous dicots. a long beak Unitegmic: Ovule with single integument, common in sympetalous dicots. Crassinuc ellate: Ovule with massive nucellus, found in primitive polypetalous Pistillode: Sterile pistil, devoid of any fer-tile ovules, as in ray floret of r adiate head of Helianthus. dicots. Radiate stigma: Sessile disc like with radiTenuinucellate: Ovule with thin layer of nucellus, as in sympetalous dicots. ating branches, as in Papaver. Stylopodium: Swollen basal part of style surrounded by nectary persisting in fru it of um-bellifers. FRUITS Sessile stigma: Seated directly on ovary, style being reduced as in Sambucus. Di scoid stigma: Disc-shaped stigma. Globose stigma: Stigma spherical in shape. A fruit is a matured and ripened ovary, wherein the ovary wall gets converted in to the fruit wall pericarp (differentiated into outer epicarp, middle mesocarp a nd inner endocarp), and the ovules into seeds. Three 80 Plant Systematics many-seeded segments, as in Mimosa. Sometimes considered as a type of schizocarpic fruit. Siliqua: Fruit developing from bicarpellary syncarpous superior ovary, which is initially one chambered but sub -sequently becomes two chambered due to the formation of a false septum, visible on the outside in the form of a rim known as replum. The fruit dehisces along b oth sutures from the base upwards, valves separating from septum and seeds remai ning attached to the rim (replum), characteristic of the family Brassicaceae. Th e fruit is narrower and longer, at least three times longer than broad, as in Br asA E F Figure 4.20 Ovules. A: Orthotropous; B: Anatsica and Sisymbrium. ropous; C: Campylotropous; D: Hemianatropous; E: Amphi-tropous; F: Circinotropou s. Silicula: Fruit similar to siliqua but shorter and broader, less than three time s longer than broad as seen in Capsella, Lepidium and Alyssum. Silicula is commo nly flattened at right angles to the false septum (Capsella, Lepidium) or parall el to main categories of fruits are recognized: simple fruits developing from a single ovary of the flower, aggregate fruits developing from several free carpels with in the flower, and composite fruits involving several flow-ers or the whole infl orescence. the false septum (Alyssum). Capsule: Fruit developing from syncarpous ovary and dehiscing in a variety of ways: Simple fruits Circumscissile (pyxis): Dehiscence transverse so that top comes off as a lid or operculum, as in Anagallis arvensis. Poricidal: Dehiscence through terminal A single fruit develops from a flower having a single carpel or several united c arpels so that the flower has a single ovary. Such a fruit may be dehiscent open ing by a suture exposing seeds or remain indehiscent. pores as in poppy (Papaver). Denticidal: Capsule opening at top ex- posing a number of teeth as in Primula an

d Cerastium. Dehiscent fruits Such fruits are generally dry and burst along the suture to release their seeds. Common Septicidal: Capsule splitting along septa and valves remaining attached to septa as in Linum. types are enumerated below: Loculicidal: Capsule splitting along locules and valves remaining attached to Follicle: Fruit developing from superior monocarpellary ovary and dehiscing alon g septa, as in family Malvaceae. one suture, as in Consolida. Septifragal: Capsule splitting so that valves fall off leaving seeds attached to central axis as in Datura. Legume or pod: Fruit developing like fol-licle from monocarpellary superior ovar y but dehiscing along two sutures, as in legumes. Schizocarpic fruits Lomentum: Modified legume, which splits transversely at constrictions into oneor This fruit type is intermediate between dehiscent and indehiscent fruits. The fr uit, Descriptive Terminology 81 P C s E G H L M A. J s Q P S U W Figure 4.21 Fruits. A: Achene of Ranunculus; B: Cypsela of Ageratum with scaly p appus; C: Nut of Castanea; D: Pod of Pisum; E: Single follicle of Calotropis; F: Siliqu a of Brassica; G: Silicula of Capsella bursa-pastoris; H: Capsule of Datura; I: Cremocarp in umbellifers; J: A pair of lomentum fruits in Mimosa; K: Double sama ra of Acer; L: Capsule of Primula dehiscing by apical teeth (denticidal); M: Ope rculate capsule of Papaver with poricidal dehiscence; N: Pyxis of Celosia with c ircumscissile dehiscence; O: Capsule of Abelmoschus esculentus with loculicidal dehiscence; P: Pome of Malus pumila; Q: Hip of Rosa with etaerio of achenes insi de; R: Drupe of Prunus; S: Berry of Lycopersicon esculentum; T: Pseudocarp of Fr agaria, an accessary fruit with etaerio of achenes; U: Etaerio of drupes in Rubu s; V: Syconium of Ficus developing from hypothodium inflorescence; W: Sorosis of Morus. 82 Plant Systematics

instead of dehiscing, rather splits into num-ber of segments, each containing on e or more seeds. Common examples of schizocarpic Utricle: Similar to nut but with papery often inflated pericarp as in Chenopodium. fruits are: Fleshy indehiscent fruits: Such fruits have fleshy and juicy pericarp even at ma turity. Common examples are: Drupe: Fruit with usually skinny epicarp, fibrous or juicy mesocarp and hard sto ny endocarp, enclosing single seed, as seen in mango, plums and coconut. Berry: Fruit with uniformly fleshy pericarp with numerous seeds inside, as seen in Sola num, tomato and brinjal. Pepo: Fruit formed from inferior ovary of cucurbits with epicarp forming tough Cremocarp: Fruit developing from bicarpellary syncarpous inferior ovary and spli tting into two one seeded segments known as mericarps, as in umbellifers. Carcer ulus: Fruit developing from bicarpellary syncarpous superior ovary and splitting into four one seeded seg-ments known as nutlets, as in family Lamiaceae. Double samara: Fruit developing from syn-carpous ovary, two or four chambered, p ericarp of each chamber forming a wing, fruit splitting into one-seeded winged s egrind. ments as in maple (Acer). It must be noted that single samara of Fraxinus, is a single-seeded dry winged indehiscent fruit and Hesperidium: Fruit developing from superior ovary with axile placentation, epica rp and mesocarp forming common rind and endocarp produced inside into juice vesi cles, as seen in citrus fruits. Pome: Fruit developing from inferior ovary, an example of accessory (false) frui t, wherein fleshy part is formed by thalamus and cartilaginous pericarp is not a schizocarpic fruit. Regma: Fruit developing from multicarpellary syncarpous ovary and splitting into one-seeded cocci, as in Rici-nus and Geranium. Indehiscent fruits inside, as seen in apple. Such fruits do not split open at maturity. Balausta: Fruit developing from inferior ovary, pericarp tough and leathery, see ds attached irregularly, succulent testa be-ing edible, as seen in pomegranate ( Punica granatum). They may be dry or fleshy: Dry indehiscent fruits: Such fruits have dry pericarp at maturity, and are repre sented by: Achene: Single seeded dry fruit developing from a single carpel with superior ov ary. Fruit wall is free from seed coat. Achenes are often aggregated, as in famAggregate fruits ily Ranunculaceae. Aggregate fruits develop from multi-carpellary apocarpous ovary. Each ovary form s a fruitlet, and the collection of fruitlets is known as etaerio. Common exampl es are etaerio of achenes in Ranunculaceae, etaerio of follicles in Calotropis, etaerio of drupes in raspberry (Rubus) and etaerio of berries in Polyalthia. In Rose the etaerio of achenes is surrounded by a cup like hypanthium forming a spe cialized accessory fruit known as hip. The fruit of strawberry Cypsela: Single seeded dry fruit, simi-lar to (and often named achene) but de-ve loping from bicarpellary syncarpous in-ferior ovary, as in family Asteraceae. Caryopsis: Fruit similar to above two but fruit wall fused with seed coat as see n in grasses. Nut: One-seeded, generally large fruit de-veloping from multicarpellary ovary an d with hard woody or bony pericarp, as seen in Quercus and Litchi. (Fragaria), though also an etaerio of achenes, is an accessory fruit, the edible

part being the fleshy thalamus. Descriptive Terminology 83

Multiple (composite) fruits Symbol Explanation Alternate A multiple fruit involves ovaries of more than Symbol one flower, commonly the whole inflorescence. Common examples are: Flowers actinomorphic * Sorosis: Composite fruit develops from the whole inflorescence and floral parts become edible, as seen in Morus (having fleshy peri-anth but dry seeds) and Arto carpus (with fleshy rachis, perianth and edible seeds). Syconium (syconus): Frui t developing from hypanthodium inflorescence of figs. There is a collection of a chenes borne on the inside of fleshy hollow receptacle. tor~jo Flowers zygomorphic X Flowers pistillate Flowers staminate K5 Sepals five and free CA5 K(5) Sepals five and united CA5 K2+2 Sepals 4 in two whorls CA2+2 FLORAL FORMULA The floral formula enables convenient graphical representation of essential flor al characteristics of a species, mainly incor-porating its sexuality, symmetry, number and fusion of floral parts and ovary position. It is more convenient to r epresent Calyx by K (or CA), Corolla by C (or CO), Perianth by P, Androecium by A and Gynoecium by G. The number of parts in a floral whorl are indi-cated by a numeral (as such when free, but when united within parentheses or a circle. Adna tion between whorls is indicated by a curve (above or below). Inferior ovary has a line above G, while the superior ovary has one below. Complete sequential rep resen-tation of components of floral formula with K(3/2)Calyx bilabiate, upper lip with 3 lobes, lower with 2 K4 5 Seeppaal s 4-5 in number CA4-5 C5 Petals 5 and free C(5) Petals 5 and united CO5 CO 5 C(2/3)Corolla bilabiate, upper lip with 2 lobes, lower with 3 P5 Perianth with 5 free tepals Perianth with 5 united tepals P(5) P3+3 Perianth with 6 free tepals in two whorls 5 Stamens five and free

A5 A(5) Stamens five and united major variations is given in Figure 4.22. A2+2 Representative floral formulae of some species of angiosperms are presented in F igure 4.23. Along side each floral formula is given a list of features of the sp ecies on which the floral formula is constructed. A2+2 Stamens didynamous 2+4 Stamens tetradynamous A2+4 A1+(9)Stamens diadelphous A1+9 ~1 FLORAL DIAGRAM C(5) A5Stamens epipetalous CO1A5 11 The floral diagram is a representation of the cross-section of the flower, flora l whorls ar-ranged as viewed from above. The floral diaG2 Carpels 2 and free, ovary superior G2 G(2) Carpels 2 and united, ovary inferior 2 gram not only shows the position of floral parts relative to the mother axis and each other, but also their number, fusion or not, overlapping, the presence and position of Figure 4.22 Representation of differen t feabracts, insertion of stamens, the number of anther sacs, whether the anthers are extures of floral whorls in a floral formula. 84 Plant Systematics

Family Species Floral Formula Solanaceae Solanum nigrum K (5)C (5)A 5 (~ (Z) petals 5 and united, stamens 5 free and epipetalous, Flowers actinomorphic, bisexual, sepals 5 and united, carpels 2 and united, ovary superior. Lamiaceae Ocimum hasilicum 7 K Flowers zygonamphic, bisexual, calyx bilabiate, upper lip with one lobe lower 4 lobed, corolla btlabiate, upper lip 4lobed, lower one lobed, stamens 4 and didyn amous, epipetalous, carpels 2 and united, ovary superior. (ll4) (4l1) 2+2 ~Z)

Wca83lc8csa0 $raSSICa Ca//7peSti'IS K C G Flowers actinomorphic, bisexual, sepals 4 free and in whorls, petals 4 cruciform , stamens 6, tetradynamous, carpels 2 and united, ovary superior. 2i+2 4X 2+4 l?') Fabaceae Lathyrus odoraius % 5 1+2+(2) AC T1 KC Flowers zygomorphic, bisexual, sepals 5 and united, petals 5 free and papilionac cous, stamens 10, diadelphous, 9 united 1 free, carpel 1, ovary superior. Malvaceae Hihiscus rosa-sinensis Epl K Flowers actinomorphic, bisexual, epicalyx 5-7 and free, 5 K(~C 5 A(0--)G (5)sepa ls 5 and united, petals 5 and free, stamens many. _ monadelphous, epipetalous, carpels 5 and united, ovary superior. Ray lforet / OtK C (5)A 0G ~2~pistillate, calyx represented by pappus, petals 5 a nd Flowers of 2 types. Ray florets zygomoprphic, Asteraceae Helianthus anmms united, stamens absent, carpels 2 and united, ovary inferior. Disc floret actino morphic, bisexual, calyx Disc floret KpappusC(5)A~5)G(_2) represented by pappus, corolla 5 and united , stamens 5 epipetalous and united (syngenesious), carpels 2 and united, ovary inferior. Chenopodiaceae Chenopodium album `+m p` )A G(2) Flowers actinomorphic, bisexual, tepals 5 and united, stamens 5 and free, carpel s 2 and united, ovary superior. Caryophyllaceae Stellaria media ~ K C A S55t 5 (3) Flowers actinonamphic, bisexual, sepals 5 and free, petals 5 and free, stamens 1 0 in two whorls, carpels 3 and united, ovary superior. Figure 4.23 Floral formulae of some representative species of few families of an giosperms depicting diversity of features depicted. The important features on which each for mula is based are shown in the right column. trorse or introrse, and more importantly, a section through the ovary, depicting the type of placentation, the number of ovules vis-ible in a section, and the p resence or ab-sence of a nectary. It also if some stamens are nonfunctional (rep resented by staminodes) and whether the ovary is func-tional or represented by a pistillode. The branch (or the inflorescence axis) bear-ing the flower is known as mother ax is, and the side of flower facing it as posterior side. The bract, if present is opposite the mother axis, and the side of flower facing it is the anterior side. The remaining components of the flower depending upon whether they are closer to the mother axis or the bract occupy postero-lateral and antero-late ral positions, respectively. The members of differ-ent floral whorls are shown a

rranged in concentric rings, calyx being the outermost and the gynoecium the inn ermost. A large ma-jority of dicot flowers are pentamerous, and as such the five members of each whorl (ex-cluding gynoecium in the centre) are arranged such a way that four of them occur in pairs (members of each pair occupying Descriptive Terminology 85 A

MWEI. I w

Figure 4.24 Stems. A: Arboreus stem (trunk) of Cyclobalanopsis glauca; B: Tendri l climbing stem of Luffa cylindrica; C: Scandent stem of Allamanda violacea; D: Creeping stem of Zebrina pendula; E: Offset of Eichhornia crassipes; F: Runner of Oxalis corniculata; G: Twining stem of Jacquemontia pentantha; H: Succulent stem of Echinopsis teresche ckii; I: Rhizome of Zingiber officinale; J: Phylloclade of Ruscus aculeatus; K: Bulb of Allium cepa; L: Tuber of Solanum tuberosum; M: Corm of Alocasia; N: Phyl loclade of Opuntia elatior. 86 Plant Systematics

C' DU fTIA

~1

y ra

Figure 4.25 Leaves. A: Alternate phyllotaxy in Citrus; B: Opposite decussate phy llotaxy in Calotropis procera; C: Whorled phyllotaxy in Alstonia scholaris; D: Ovate long a

cuminate leaf of Ficus religiosa; E: Sagittate leaf of Sagittaria sagitifolia; F : Palmately lobed leaf of Rubus trifidus; G: Palmately lobed leaf- lobes further pinnately lobed in Carica papaya; H: Palmate leaf of Acer palmatum; I: Unifolia te compound leaf of Citrus medica; J: Palmately trifoliate compound leaf of Oxal is corniculata; K: Palmate compound leaf of Cannabis sativa; L: Pinnate compound leaf of Rosa; M: Peltate orbicular leaf of Tropaeolum majus; N: Bipinnate compo und leaf of Leucaena leucocephala; O: Panduraeform leaf of Jatropha panduraefoli a; P: Grass leaf with leaf sheath and free lamina of Zea mays; Q: Pitcher leaf o f Sarracenia flava. Descriptive Terminology 87

:r r' rsa r

Figure 4.26 Inflorescences. A: Solitary flower of Malvaviscus arboreus; B: Corym bose-raceme of Brassica campestris; C: Corymb of Iberis amara; D: Rhipidium of Solanum nigrum; E: Cyathium of Euphorbia milii; F: Spike of Adhatoda vasica; G: Panicle of spike lets of Zea mays; H: Cob (spike of spikelets) of Zea mays; I: Spike of spikelets of Triticum aestivum; J: Raceme of verticillasters in Salvia splendens; K: Spik e of Acanthus spinosus; L: Raceme of Delphinium ajacis; M: Hypanthodium of Ficus religiosa; N: Cymose cluster with spathaceous bracts of Rhoeo discolor; O: Umbe l of Astrantia major; P: Radiate capitulum of Viguieria helianthoides; Q: Discoi d capitulum of Ageratum houstonianum; R: Spadix of Amorphophalus titanum; S: Cym ose umbel of Agapanthus umbellatus. 88 Plant Systematics r 110

'

N ~J

b 1 Figure 4.27 Fruits. A: Dehisced capsule of Gossypium hirsutum with exposed hairy seeds; B: Capsule of Papaver orientale; C: Dehisced capsule of Chiranthodendron pentada ctylon; D: Etaerio of achenes of Anemone occidentalis; E: Double samara of Acer griseum; F: Pod of Dalbergia sissoo; G: Cypsela of Haplopappus macrocephalus; H: Cypsela of Sonchus oleraceous; I: Schizocarp of Abutilon indicum; J: Carcerulus of Salvia splensens; K: Drupe of Juglans nigra; L: Drupe of Prunus persica; M: Pome of Malus pumila; N: Same in Longitudinal section; O: Pod of Clitoria ternat ea; P: Hesperidium of Citrus sinensis; Q: Same in Transverse section; R: Berry o f Lycopersicon esculentum; S: Same in Transverse section; T: Berry of Ribes menz iesii; U: Etaerio of drupes of Rubus nepalensis; V: Pepo of Cucumis sativus in T ransverse section; W: Whole pepo; X: Accessory fruit of Fragaria vesca; Y: Siliq ua of Brassica campestris; Z: Dehisced capsule of Stellaria media; a: Pod of Leu caena leucocephala; b: Multiple fruit of Liquidambar styracifolia; c: Multiple f ruit of Arbutus unedo; d: Balausta of Punica granatum. Descriptive Terminology 89

E B D A l J I K

mo/ I, M P N O Q Figure 4.28 Floral diagrams of some representative members of major families. A: Brassica campestris (Brassicaceae); B: Stellaria media (Caryophyllaceae); C: Hibiscus ros a-sinensis (Malvaceae); D: Lathyrus odoratus (Fabaceae-Faboideae); E: Acacia nil otica (FabaceaeMimosoideae); F: Foeniculum vulgare (Apiaceae); G: Ray floret of Helianthus annuus (Asteraceae); H: Disc floret of H. annuus; I: Calotropis proce ra (ApocynaceaeAsclepiadoideae); J: Withania somnifera (Solanaceae); K: Ocimum b asilicum (Lamiaceae); L: Male flower of Morus alba (Moraceae); M: Female flower of M. alba; N: Narcissus pseudo-narcissus (Amaryllidaceae); O: Avena sativa (Poa ceae), floral diagram of spike-let; P: Zea mays (Poaceae), floral diagram of fem ale spikelet; Q: Z. mays, floral diagram of male spikelet. complementary position) the fifth one is the odd member. It is also to be rememb ered that in large majority of dicots (except Fabaceae and few others), the odd sepal oc-cupies posterior position (of the remaining four, two form antero-later al pair, and the remaining two the postero-lateral pair). The different whorls u sually alternate each other, and accordingly the odd petal occupies anterior pos ition, the petals alternate with sepals. The stamens accordingly alternate with

petals and are opposite the sepals. In flowers with two whorls of stamens, the o uter whorl alternates with petals, whereas the inner is opposite the petals (bec ause it al-ternates with the outer whorl of stamens). The stamens are represente d in the floral diagram by anthers, each with two anther lobes (shown by a deep fissure) and lat ter, in turn, with two anther sacs (with a less deeper cleft). The lobes face to wards the outside in extrorse anthers and towards the ovary in introrse anthers. Epipetalous stamens are shown by a line joining the an-thers with the petals. A few representative types of floral diagram are shown in Figure 4.28. The floral diagram summarizes the information about the presence or absence of b racts and bracteoles, number, fusion and aestivation of sepals and petals (or te pals if there are no separate sepals and petals, as shown in Moraceae). The caly x and corolla forming bilabiate arrangement are appropri90 Plant Systematics ately shown with the number of lobes in up-per and lower lip (as seen in Lamiace ae). The stamens with united filaments are de-picted by joining anthers via line s (diadelp-hous condition in Fabaceae-Faboideae), whereas the united anthers are shown by physically touching anther margins. In fami-lies with complex floral a rrangement such as the cyathium in Euphorbia, floral diagram for the entire cyathium may be drawn, supplemented by floral diagrams of male an d female flowers. In family Poaceae also, it is helpful to make a floral diagram for the whole spikelet (shown in Avena sativa), or separate diagrams for male a nd female spikelets if the male and female flowers occur in separate inflorescen ces or at least separate spikelets (shown in Zea mays). Chapter 5 Process of Identif ication Recognizing an unknown plant is an impor-tant constituent taxonomic activity. A plant specimen is identified by comparison with already known herbarium specimen s in a herbarium, and by utilizing the available lit-erature and comparing the d escription of the unknown plant with the published descrip-tion/s. Since the bul k of our plant wealth grows in areas far removed from the centres of botanical r esearch and training, it becomes imperative to collect a large number of specime ns on each outing. For proper description and documentation, these speci-mens ha ve to be suitably prepared for incororation and permanent storage in a her-barium. This goes a long way in compiling floristic accounts of the different regions of the world. The availability of t he specimens properly prepared, can retain their essential features for a very long period, p roving to be immensely useful for future scientific studies, including compilati on of floras, taxonomic monographs and, in some cases, even experimental studies , since the seeds of several species can remain viable for many years even in dr y herbarium specimens. Fieldwork The fieldwork of specimen preparation in-volves plant collection, pressing and p artial drying of the specimens. The plants are collected for various purposes: b uilding new herbaria or enriching older ones, compilation of floras, material fo r museums and class work, ethnobotanical studies, and in- troduction of plants i n gardens. In addition, in the herbaria often provides reasonable information about the abundance or rarity of bulk collections are done for trade a nd drug manufacture. Depending on the purpose, rea species, and helps in preparing lists of rare or endangered species, and also provides sufficient inputs for efforts towards t heir con-servation.

sources, proximity of the area and duration of studies, fieldwork may be underta ken in different ways: SPECIMEN PREPARATION A specimen meant for incorporation in a her-barium needs to be carefully collect ed, pressed, dried, mounted and finally properly labelled, so that it can meet t he demands of rigorous taxonomic activity. Specimens, Collection trip: Such a trip is of short duration, usually one or two days, to a nearby place, for brief training in fieldwork, vegetation study and plant colle ction by groups of students. Exploration: This includes repeated visits to an area in different seasons, for a period 92 Plant Systematics of a few years, for intensive collection and study, aimed at compilation of flor istic accounts.

Expedition: Such a visit is undertaken to remote and difficult area, to study th e flora and fauna, and usually takes several months. Most of our early informati on on Himalayan flora and fauna has been the result of European and Japanese exp editions.

Equipment

The equipment for fieldwork may involve a long list, but the items essential for collection include plant press, field notebook, bags, vasculum, pencil, cutter, pruning shears, knife and a digging tool (Figure 5.1). Figure 5.2 Plant press containing pressed Plant Press A plant press consists of two wooden, plywood or wire mesh planks, each 12 inche s X 18 inches (30 cm X 45 cm), between which are specimens. Vasculum placed alongside (Photograph courtesy Mr. S. L. Kochhar). camp or the organization, is called the drying press. It is much heavier and has an increased number of corrugated sheets, one alternating each folded blotter c ontaining one folded newspaper. In countries such as India which use thick coars e paper for newsprint, blotters can be dispensed with, in at least subsequent ch anges, as the paper soaks sufficient moisture and serves the purpose of blotters as well. A B C D Figure 5.1 Common implements helpful in collection. A: Trowel; B: Prunning shears; C: Knife; D: Pickaxe. placed corrugated sheets, blotters and newsField Notebook A field notebook or field diary is an imporpaper sheets (Figure 5.2). Two straps, chains or belts are used to tighten the press. Cor- tant item for a collector. A well-d

esigned field rugated sheets or ventilators are made of notebook (Figure 5.3) ha s numbered sheets cardboard, and help ventilation and the con- with printed prof orma for entering field notes sequent drying of specimens. The ducts of such as scientific name, family, vernacular the corrugated sheet run across and not name , locality, altitude, date of collection and lengthwise to in order to afford sh orter dis- for recording any additional data collected in tance and larger number of ducts. the field. The multiple detachable slips at the lower end of the sheet, separate d by per-forated lines and bearing the serial number of the sheet, can be used a s tags for multiple specimens of a species collected from a site, and serve as r eady reference to the information recorded in the field notebook. The number als o serves as the collection number for the collector. The plant press carried in the field, and called a field press is light weight a nd generally has one corrugated sheet alternat-ing with one folded blotter conta ining ten newspaper sheets, one meant for each specimen. The plant press used for subsequent press-ing and drying of specimens, kept at t he base Process of Identification 93 WESCUlUM A vasculum is a metallic box with a tightly-fitted lid and a shoulder sling. It is used to store specimens temporarily before press-ing, and also to store bulky parts and fruits. It is generally painted white to deflect heat Collection The specimen collected should be as complete as possible. Herbs, very small shru bs, as far as possible, should be collected complete, in flowering condition, al ong with leaves and roots. Trees and shrubs should be collected with both vegeta tive and flowering shoots, to enable the representation of both leaves and flowe rs. All information con-cerning the plant should be recorded in the field notebo ok and a tag from the sheet at- tached to the concerned specimen. It is advisabl e to collect a few specimens of each species from the site, to ensure that reser ve specimens are available if one or more get destroyed, and also to ensure that duplicates can be deposited in different herbaria, when finally mounted on shee ts. Department of Botany University of Kashmir Srinagar- 190006 Date: 14- 3- 1970 No. 1068 Name: Iris ensata Local Name: KRISHM Pressing Family: Iridaceae Locality: Harwan, Kashmir The specimens should be placed in the field press at the first opportunity, eith er directly after collection, or sometimes after a temporary storage in a vascul um or a polythene bag. A specimen shorter than 15 inches (38 Altitude: 1900 m Habitat: Open grassland meadow

Collector: Gurcharan Singh Determinavit: Self Notes: Perennial herb, forming isolated patches, flowers light blue. r ~ a b c d No. 1068 No. 1068 Figure 5.4 a-c: Different methods of folding 0 0 0 0 O longer herbaceous plants; d, use of flexostat slips for holding plants in folded condition. Note that the tip of the plant (arrow) would al-ways be erect for co nvenient study of this important portion with leaves and flowers. Figure 5.3: A sheet from field notebook with relevant entries. and affords easy detection when left in the field. Being bulky, the vasculum is com-monly substituted by a polythene bag, which is almost weightless. A number o f polythene bags can be carried for easy stor-age, as these can be readily made airtight using a rubber band and, as such, the plants retain their freshness for many hours. cm) should be kept directly in the folded newspaper after loosely spreading the leaves and branches. Herbs, which are generally collected along with the roots, if longer than 15 inches, can be folded in the form of a V, N or W (Figure 5.4, a-c), always ensuring that the terminal part of the plant with leaves, 94 Plant Systematics flowers and fruits, is erect, and when finally mounted, the specimens can be eas ily stud-ied, without having to invert the herbarium sheet. Specimens of grasses and some other groups, which show considerable elasticity, are difficult to hol d in a folded condition. These specimens can be managed by using flexostat (a st rip of stiff paper or card with 2.5 cm long slit). One flexostat inserted at eac h corner (Figure 5.4 d) holds the speciSuch specimens are collected in bags and made to float in a tray filled with wat er, at the bottom of which a white sheet of paper is placed. The paper is lifted gently, carry-ing the specimen along and placed in a blot-ter and pressed. As t he slender water plant sticks to the paper, the sheet along with the specimen is shifted from one blotter to an-other during the process of drying, and finally pasted on the herbarium sheet as such. men in place. Succulents and cacti have a large amount of proliferated parenchyma storing wate r and, unless special care is taken, these plants readily rot and fungal infecti on sets in. Such plants are handled by giving slits on thick organs and scooping out the succulent tissue or, alternately, salt is sprinkled on the slits to dri ve out the moisture. The plants may also be killed by pretreatment To press bulky fruits, these may be thinly sliced. Large leaves can be trimmed t o re-tain any lateral half. It is useful to invert some leaves so that the under surface of the leaves can also be studied from a pressed leaf. Handling special groups

A few groups of plants such as conifers, wa-ter plants, succulents and mucilagin ous plants pose problems during collection and with ethyl alcohol or formaldehyde. need special methods. Mucilaginous plants such as members of the family Malvaceae stick to the blotter s and are difficult to process. These plants should be placed between waxed or t issue paper or else folds of muslin cloth. Only the blotter should be changed ev ery time the press is opened and the specimen separated from the tissue paper or muslin only when Conifers, although easy to collect and press, pose problems during drying. The t is-sues of conifers remain living for a long time and progressive desiccation du ring pressing and drying initiates an abscission layer at the base of leaves and sporophylls. As such a dry twig readily disintegrates, losing its leaves with a slight touch, a problem occasionally encountered in Abies, Picea, Cednus, and s ev-eral other genera. Before pressing, such twigs should be immersed in boiling water for one minute, a pretreatment that kills the tissues and prevents the abs cission formation dur-ing drying. Page (1979) has suggested pretreatment method involving immersion in 70% ethyl alcohol for 10 minutes, followed by immersion i n 50% aqueous glycerine solu-tion for four days. Since the pretreatment removes the bloom and waxes, and results in a slight colour change, an untreated portion of the plant should also be preserved, kept in a small pouch and attached to th e herbarium sheet along with the pretreated specimen, fully dry. Aroids and bulbous plants continue to grow even in a press even after they have pre-sumably been properly pressed and dried. These should be killed with ethyl a lcohol and formaldehyde prior to pressing. Drying Drying of pressed plant specimens is a slow process if no artificial heat is use d. Natural Drying for reference. Natural drying of specimens is a slow pro-cess, which may take up to one month f or complete drying. The plants, freshly col-lected, are placed in a press withou t corrugated sheets and the press is locked for 24 hours. During this sweating p eriod, plants lose some moisture, become flaccid and can be easily rearranged. T he folded sheet con-taining the specimen is lifted and placed in Water plants, especially with submerged leaves, readily collapse due to the abse nce of cuticle and are difficult to press normally. Process of Identification 95 a fresh dry folded blotter. In countries using thick coarse newsprint, changing the news-paper is also necessary, and the plant should be carefully transferred from one newspa-per to another. The use of a blotter in such a case can be dispe nsed with, especially af-ter one or two changes. The change of blot-ters or news paper sheets is repeated every few days, increasing the interval between the cha nges successively until the speci-mens are fully dry. The whole process of dry-i ng may take about 10 days to one month, depending on the specimens and the clima te of the area. drying. This solar drier, with practically no operational cost, should provide a right step towards energy conservation. The rapid drying of specimens using artificial heat has, however, inherent limit ations of rendering plants brittle, loss of bloom and some colour change in leaves. In arid regions, plants can be dried par-tially during travel, by placing the pr ess horizontally on the luggage rack of the vehicle, with the corrugate ducts fa cing front, forcing the dry wind through the corrugates as the vehicle moves forward. Drying With Artificial Heat Specimens pressed and dried are next mounted on herbarium sheets, and properly l abelled before these can be incorporated in a herbarium.

Drying with the help of artificial heat takes 12 hours to two days. The specimen s, after the initial sweating period in the field press, are transferred to a dr ying press, with an ample number of corrugated sheets, usually one alternating e very folded blotter contain-ing one specimen. The press is kept in a drier, a ca binet in which a kerosene lamp or electric bulb warms the air, drying the specim ens by movement through the cor-rugates. Use of a hot air blower in the cabi-net speeds up circulation of the hot air and, consequently, faster drying is achiev ed. Sinnott (1983) developed a solar powered drier capable of drying 100 specime ns on a sunny day, and attaining a temperature of up to 600 Celsius in the centr e of the press. The unit consists of a flat plate collector and a drying box to hold the press. The collector is composed of a wooden frame, a blackened alumini um absorber plate, insulation and a glass or Plexiglas glazing to retain and cha n-nel heat into the drying box. One-inch space is provided between the glazing a nd the ab-sorber plate. The air enters the collector at the open bottom of the c ollector panel, is heated by conduction from the absorber, rises by convection i nto the drying box, moves through the corrugates and finally exits from the unco vered top of the drying box, taking with it moisture from the plant specimens. D rying is accomplished in a single day, occasionally two days for complete HERBARIUM METHODS A herbarium is a collection of pressed and dried plant specimens, mounted on she ets bearing a label, arranged according to a sequence and available for referenc e or study. In practice, it is a name given to a place owned by an institution, which main-tains this orderly collection of plant specimens. Most of the well-kn own herbaria of the world made their beginning from botanical gardens. Botanicalga rdens Although gardens existed in ancient China, India, Egypt and Mesopotamia, these g ar-dens were not botanical gardens in the true sense. They existed for growing f ood plants, herbs, and ornamentals for aesthetic, religious and status reasons. The famous hanging gardens of Babylon in Mesopotamia is a typical example. The fir st garden for the purpose of science and education was main-tained by Theophrast us in his Lyceum at Athens, probably bequeathed to him by his teacher, Aristotle . Credit for establishment of the first modern botanical garden belongs to Luca Ghini (ca 1490-1556), a professor of botany who developed it at Pisa, Italy in 1 544. These were followed by botanical gardens at Padua and Florence in 1545. 96 Plant Systematics Roles of a Botanical Garden sity, as also in conserving rare and endangered species. The Proceedings of the Symposium on Threatened and Endangered species, sponsored by New York Botanical Garden in 1976, pub-lished as Extinction is Forever, and the conference on pract ical role of botanical gardens in conservation of rare and threatened species sp onsored by the Royal Botanical Garden, Kew and published as Survival and Extinct ion, are among the major examples of the role Botanical gardens have been instrumental in motivating several well-known author s to develop their own systems of classification while trying to fit the plants grown in the garden, into some previous system of clas-sification, e.g. Linnaeus , while working at Uppsala and Bernard de Jussieu at Versailles. Although the ma jority of the bo-tanical gardens house plant species which the climate of the ar ea can support, several well-known botanical gardens have conof botanical gardens in conservation. trolled enclosures to support specific plants. Tropical gardens often need indoo r growing space, screen houses for most plants and glasshouses for the majority of cacti and succulents in wet tropical and temperate gardens. Glasshouses in te mperate gardens often require winter heating. Botanical gar-dens play the follow ing important roles: 6. Seed exchange: More than 500 botanical gardens across the world operate an informal seed exchange scheme, offering annual lists of available species and a free exchange of seeds. 7. Herbarium and library: Several

1. Aesthetic appeal: Botanical gardens major botanical gardens of the world have herbaria and libraries as an in-tegral part of their facilities, and offer taxonomic material for research at a have an aesthetic appeal and attract a large number of visitors for observa-tion of general plant diversity as also the curious plants, as for example, the Grea t Banyan Tree (Ficus benghalensis) in the Indian Botanical single venue. 8. Public services: Botanical gardens Garden at Kolkotta. provide information to the general public on identification of native and exotic species, methods of propagation and also supply plant material through sale or exchange. 2. Material for botanical research: Botanical gardens generally have a wide range of species growing together and offe r ready material for botanical research, which can go a long way in understandin g taxonomic affinities. Major Botanical Gardens 3. On-site teaching: Collection of plants Thousands of botanical gardens located worldwide are maintained by various insti tutes. Of these, nearly 800 important gar-dens are documented in the Internation al Di-rectory of Botanical Gardens published by Henderson (1983). A botanical ga rden today is an area set aside and maintained by an organization for growing va rious groups of plants for study, aesthetic, conservation, economic, educational , recreational and scientific purposes. Some of the major botanical gardens are discussed below: is often displayed according to fami-lies, genera or habitats, and can be used f or self-instruction or demonstration purposes. 4. Integrated research projects: Botanical gardens with rich living material can support broad-based research projects which can integrate informa-tion from such diverse fields as anatomy, embryology , phytochemistry, cytology, physiology and ecology. New York Botanical Garden, USA: This garden was christened the New York Botanical Garden in 1891, when the Torre y Botanical Club adopted its foundation as a 5. Conservation: Botanical gardens are now gaining increased importance for their role in conserving genetic diverProcess of Identification 97 corporation chartered by the State. David Hosak founded the garden in 1801 as Al gin tanical garden and botanical research and resource centre in the world. The gard en was developed in the 1600s by Kew House owned by Richard Bennet. The widow of the Prince of Wales commissioned the garden in 1759 and William Aiton took over as its superintendent. Sir Joseph Banks introduced large collections from diffe rent parts of the world. In 1841, the management of the garden was transferred f rom the crown to the parliament and Sir William Hooker became its first offiBotanic Garden. Professor N. L. Britton, the most produc-tive taxonomist of his time, directed t he idea ^ g

the advancement of the garden, enlarging

and building a palm house. Sir J. D. Hooker, who succeeded his father as its Dir ector, added rhododendrons, and also authored several important publications. Jo hn Hutchinson worked and developed his faFigure 5.5 Haupt conservatory complex of New York Botanical Garden. mous system of classification here. of advancement of botanical knowledge through research at this botanical garden. The garden (Figure 5.5) today covers 100 ha. in the heart of New York City alon g the Bronx River. In addition 778 ha. Mary Flager Cary Arboretum at Millbrook h as been added to the jurisdiction of the garden. There are 15,000 species distri buted in the demonstration gardens, Montgomery conifer collection, Stout day lil y garden, Havemeyer lilac collection, Rhododendron and Azalea collection, Everet t rock garden, herb garden, rose garden, arboretum and conservatory complex. The garden has a systematic arrangement of trees and shrubs that make it a place of interest for the general public as well as botanists and horticulturists. The g arden plays a major role in conservation of rare and endangered species. The gar den has a well-maintained herbarium of over 5 million specimens from all over th e world, but mainly from the New World. The library houses over 200,000 volumes and over 500,000 items (including pamphlets, photographs, letters, etc.). It als o maintains a huge botanical database. The garden (Figure 5.6) has since grown into a premier Research and Educational _ Figure 5.6 Princess of Wales House at Royal Botanic Gardens Kew. Royal Botanic Gardens, Kew: More popularly known as Kew Gardens , this historical garden is undoubtedly the finest boInstitute with excellent herbarium and library. Originally the garden covered an area of 120 ha. The outstation of the Royal Botanic Gardens, Kew at Wakehurst P lace near Ardingly in West Sussex is a rural estate of 202 ha. with an Elizabeth an mansion, and was acquired in 1965. The Royal Botanic 98 Plant Systematics Gardens Kew has directed and financed its development so that Wakehurst Place no w makes a vital contribution in maintaining the international reputation of the Living Collections Department (LCD). In particular the practical in-situ conserv ation policies pursued, and the rich and diverse plant col-lections, which are m aintained, add greatly to the LCD s activities. The environmental conditions of th e High Weald of Sussex con-trast with those of Kew by offering varied to-pograph y, higher rainfall and more diverse and moisture retentive soils. These com-bine together to provide a range of microcli-mates, which make possible the successf ul cultivation of a great diversity of plants, mature temperate trees. Tropical plants are maintained indoors, including Aroid House, Palm House, Filmy Fern House etc. Several interesting plants such as Vict oria amazonica from South America and Welwitschia mirabilis from Angola are also growing here. Kew Herbarium, undoubtedly the most famous herbarium of the world, maintains ove r 6 million specimens of vascular plants and fungi from every country in the wor ld. There are over 275,000 type specimens as well. The library at Kew is very extensive with over 750,000 books and journals a reso urce for all Kew s research work. Kew Bulletin and Index Kewensis are its two cont inuing premier publications. many of which do not thrive at Kew. Kew maintains databases on plant names, taxonomic literature, economic botany, p lants for arid lands and on plant groups of special economic and conservation va

lue. Kew also makes about 10, 000 identifications a year through its Herbarium s ervice and provides specialist advice on taxonomy and nomenclature in difficult cases. Kew is in-volved in major biodiversity research programmes in many parts of the world, in-cluding tropical and West Asia, SE Asia, Africa, Madagascar, So uth America, and the Pacific and Indian Oceanic islands. The Herbarium runs an i nternational Diploma Course in Herbarium Techniques. The General Catalogue now c ontains over 122,000 records and is available throughout RBG Kew on the network. There are substantive differences in the layout and content of the collections a t Wakehurst Place which act to complement those at Kew. In particular the botani cal collections are laid out in a floristic manner reflecting the way that tempe rate plant communities have evolved. The botanical collections are supported by extensive ornamental displays exploiting the wide range of available biotopes an d acting as primary visitor attractants. A final element of the woodland cover is forestry plot s compris-ing high forest and Christmas tree planta-tions. Jodrell Laboratory at Kew has estab-lished itself as the world centre in the study of plant anatomy, cytogenetics and plant biochemistry. The Royal Botanic Gardens Living collections at Kew and Wakehurst Place are a mul -tilevel encyclopaedic reference collection re-flecting global plant diversity a nd providing a reference source which serves all the as-pects of botanical and h orticultural science within Kew, Great Britain and throughout the world. It is p robably the largest and most diverse living collection in the world. The two sit es provide quite different environ-ments, allowing the development of two dif-fe ring but complementary collections. The living collections at Kew are most diver se with 351 families, 5465 genera and over 28,000 species growing successfully. The ar-boretum covers the greatest area with large Missouri Botanical Garden, USA: Considered one of the top three botanical gardens in the world, the Missouri Bot anical Garden is a National Historical Landmark and a centre for botanical resea rch, education and horticultural display. The garden was founded by an Englishma n Henry Shaw and opened to public in 1859 with active help from Asa Gray and Sir William Hooker and Enelmann. Today, the garden covers 79 acres and operates the world s most active tropical botany research programme. Under the leadership of Dr Peter Raven, its Director, the Garden plays a leading role in strategies of conservation and sustainProcess of Identification 99 able living. The garden is known for its Climatron conservatory, a geodesic green -house dome with climatic control, support-ing a vibrant tropical rainforest, un der a 0.5 acre roof (Figure 5.7). It also has a Japanese Garden (Seiwa-en) cover ing 14 acres, the largest Japanese strolling Garden in North The Missouri Botanic Garden is one of the world s leading research centres for bot anical exploration and research, with nearly 25 major flora projects. The informati on is shared via website TROPICOS, the world s largest database, containing more t han 920,000 scientific plant names and over 1,800,000 specimen records. The gard en s highly regarded education programme seeks to improve science instruction in the St. Louis region, reaching more than 137,000 students each year. .._`~_-~ With more than 5.3 million specimens (mosses, ferns, gymnosperms, and angiosperm s), the herbarium ranks second in the USA and 6th in the world. It has collectio ns dating back to mid 1700s. The herbarium specializes in having collections of G. Boehmer, Joseph Banks, D. Solander (who accompanied Captain Jamers Cook in hi s first voyage around the world), and Charles Darwin. During the last five years , the her-barium has added an average of 120,000 mounted specimens per year to i ts collection. In addition to the many gift specimens sent to the specialists, t his herbarium loans an average of 34,000 specimens annually, and borrows about 2

7,000 specimens. The herbarium staff also provides identifications from their ar ea of expertise. The pace of development of the herbarium can be judged from the fact from being number 13th in the world in 1990 (Woodland, 1991), the herbariu m today has risen to number six. The reference library of the garden has over 22 0,000 volumes, including many rare %++~ -- - -- ~,..~

NC Figure 5.7 Climatron at the Missouri Botanibooks. cal Garden, a Geodesic dome with Climatic control and supporting tropical rainfo rest. (Photograph by Jack Jennings/Courtesy the Mis-souri Botanical Garden). Among its major research activities in-clude Flora of North America project, fiv e volumes having already been published, covering the plants of USA, Canada and Greenland. The garden also coordinates Flora of China project, 25 volumes being planned, America, with a proud collection of to be completed in 15 years starting from Hamerocallis, Iris, roses, Hosta, and several 1994. economic plants (Figure 5.8). There are also Chinese, English German and Victorian Gardens. Over 4,000 trees thrive on the grounds, including some rare and unusual varieties. Pisa Botanical Garden, Italy: The Pisa Garden, developed by Luca Ghini in 1544, is credited as the first modern botanical garden. The garden was known for the 100 Plant Systematics The botanical garden today comprises an area of 126 acres. About 20,000 differen t species of plants are cultivated here. The section on plant geography covering 39 acres, one of the biggest of its kind in the world, depicts the whole of the Northern Hemisphere. The arboretum and taxonomy section covers 42 acres and inc ludes around 1800 species of trees and shrubs and nearly 1000 species of herbace ous plants, the latter arranged according to the classification system of Adolph Engler. The botanical mu-seum specializes in the display of botanical exhibits, being the only museum of its kind in Central Europe with models of various

Figure 5.8 Japanese Garden at the Missouri

Botanical Garden. (Photograph by Jack Jennings/Courtesy the Missouri Botanical G arden). life-forms. The main tropical greenhouse (Figure finest specimens of Aesculus hippocastanum, Magnolia grandiflora, and several ot her species. Though the garden does not exist today, the records of its design d emonstrate geometric outlay of plantings that are characteristic of several cont inental gardens 5.9), with its length of 60 m and height of Padua Botanical Garden, Italy: The Garden, established in 1545. The specialty of this garden is the elegance and Ha lian taste, which has been wedded to the service of science. The elegance and be auty of Padua Botanical Garden are equalled by Kew Gardens only.

Berlin Botanic garden and Museum, Berlin-Dahlem: The Berlin Botanic Garden was set up in 1679 when the Grand Duke of Berlin gave instructions to open a n agricultural model garden in Schoneberg, a village near Berlin. Due to lack of space, it was later relocated to Dahlem. The garden developed largely due to th e efforts of C. L. Wildenow, who built it up from an old run-down royal garden. Adolph Engler and L. Diels, who were its subsequent directors, improved its qual ity and content. Much of the garden was destroyed during World War II. It was re built largely through the efforts of Robert Pilger, its then director. Figure 5.9 Tropical greenhouse of the Berlin Botanic Garden at Dahlem. 23 m, is one of the largest in the world, featuring tall trees with epiphytes, r ich ground vegetation and lianas, which give an idea of the vast variety of trop ical vegetation. Cambridge University Botanical Garden: The Cambridge University Botanical Garden was founded in 1762 as a small garden on 5 acres of land in the Process of Identification 101 centre of Cambridge. It was moved to the present location in 1831 when Prof. J.S . Henslow established it on newly-acquired land of the University covering 40 ac res. The garden is artistically landscaped with sys-tematic plantings, winter-ha rdy trails, an al-pine garden and a chronological bed. The latter is in the form of a narrow bed (300 x 7 feet) divided into 24 sections, each contain-ing plant s introduced during a 20-year pe-riod. Tropical houses are one of the major attr actions of the garden and contain palms and other tropical plants. Although the herbarium technique was a well-known botanical practice at the time of Linnaeus, he departed from the convention of mounting and binding the specim ens into volumes. He mounted specimens on single sheets, storing them horizontal ly, a practice followed even today. From isolated personal collections, her-baria have grown into large institutions of national and international stature with millions of specimens from different parts of the world. Index Herbariorum, edited by Patricia Holmgren (Figure 5.10 ) (Holmgren, Holmgren, & Barnett, 1990) lists the world s important herbaria. Each herbarium is identified by an abbreviation that is valuable in locating the typ e specimens of various species. The major herbaria of the world with approximate number of specimens in the orHerbaria It was again Luca Ghini who initiated the art of herbarium making by pressing an d sewing specimens on sheets of paper. This art was disseminated throughout Euro pe by his students who mounted sheets and bound them into book volumes. der of importance are listed in the table 5.1. In India Central National Herbarium (CAL) of the Indian Botanic Garden, Botanica l Survey of India, Kolkotta has over 1.3 million specimens. The herbarium of For

est Research Institute, Dehradun (DD) and National Botanical Research Institute, Lucknow (LUCK) are other major herbaria in India, with collections from all ove r the world. Roles of a Herbarium From a safe place for storing pressed specimens, especially type material, herba ria have gone a long way in becoming major centres of taxonomic research. Additi onally, herbaria also form an important link for research in other fields of stu dy. The classification of the world flora is primarily based on herbarium materi al and associated literature. More recently, the herbaria have gained importance for sources of informa-tion on endangered species and are of primary interest t o conservation groups. The

major roles played by a herbarium include: Figure 5.10 Patricia Holmgren Director 1. Repository of plant specimens:' PriEmerita ofthe Herbarium, New York Botanical Garden, the editor of Index Herbario rum and 2 volumes of Intermountain Flora. (Courtesy New York Botanical Garden, B ronx). mary role of a herbarium is to store dried plant specimens, safeguard these agai nst loss and destruction by insects, and make them available for study. 102 Plant Systematics Table 5.1 Major herbaria of the world, listed in the order of number of specimen s. Herbarium Abbr. No. of specimens 2. New York Botanical Garden, New York, USA 1. Museum National d Histoire Naturelle (Museum of Natural History), Paris, France * (Formerly Leningrad), Russia. (Conservatory and Botanical Garden), Geneva, Switzerland Sweden P, PC LE K G MO BM A, FH, GH, ECON, AMES S 9,377,300 7,000,000 7,000,000 7,000,000 6,000,000 6,000,000 5,200,000 5,000,500 4,400,000 3. Komarov Botanic Institute, Saint Petersburg 4. Royal Botanic Gardens, Kew, Surrey, UK 5. Conservatoire et Jardin Botaniques 6. Missouri Botanical Garden, Saint Louis, Missouri, USA

7. British Museum of Natural History, London, UK 8. Combined Herbaria, Harvard University, Cambridge, Massachusetts, USA** 9. Naturhistorika Riksmuseet, Stockholm, 10. US National Herbarium (Smithsonian), Washington, USA US 4,340,000 Information updated on 17 February, 2009 from Institutional websites. * Specimens are located at Laboratoire de Phanrogamie (P) and Laboratoire de Cryp togamie (PC). **Consists of Arnold Arboretum (A), Farlow Herbarium (FH), Gray s He rbarium (GH), The Economic Herbarium of Oakes Ames (ECON) and Oaks Ames Orchid H erbarium (AMES). 2. Safe custody of type specimens: Type 5. Identification of specimens: The specimens are the principal proof of the existence of a species or an infraspeci fic taxon. These are kept in safe custody, often in rooms with restricted access , in several major majority of herbaria have a wide-ranging collection of specimens and offer facil ities for on-site identification or having the specimens sent to the herbarium i dentified by experts. Re-searchers can personally identify their collection by c omparison with the duly identified herbarium specimens. herbaria. 3. Compilation of Floras, Manuals and Monographs: Herbarium specimens are the original documents upon which the knowledg e of taxonomy, evolution and plant distribution rests. Floras, manuals and monog raphs are 6. Information on geographical largely based on herbarium resources. distribution: Major herbaria have collections from different parts of the world and, thus, scrutiny of the specimens can provide information on the geographical distribution of a taxon. 4. Training in herbarium methods: Many herbaria carry facilities for training graduates and undergradu-ates in her barium practices, organiz-ing field trips and even expeditions to remote areas. 7. Preservation of voucher specimens: Voucher specimens preserved in various herbaria provide an index of specimens on which a chromosomal, phytochemical, ultrastructural, Process of Identification 103 micromorphological or any specialized study has been undertaken. In the case of a contradictory or doubtful report, the voucher specimens can be critically exam ined in order to arrive at a more satisfactory conclusion. paper envelops called fragment packets are often attached to the herbarium sheet to hold seeds, extra flowers or loose plant parts. Labelling Mounting of Specimens An herbarium label is an essential part of a permanent plant specimen. It primar ily con-tains the information recorded in the field diary (Field notebook) at th e time of collection, as also the results of any subsequent identification proce ss. The label is located on the lower right corner of the herbarium sheet (Figur e 5.11), with the necessary information recorded on the pre-printed proforma, pr inted directly on the sheet or on the paper slips which are pasted on the sheets . It is ideal to type the information. If handwritten, it should be in permanent ink Ball pens should never be used, as the ink Pressed and dried specimens are finally mounted on herbarium sheets. A standard herbarium sheet is 29 by 41.5 cm (111/2 by 161/2 inches), made of thick handmade pa-per or a card sheet. The sheet should be rela-tively stiff to prevent damage during handling of specimens. It should have a high rag content (preferable 100 per cent) with fibres running lengthwise. The specimens are attached to the sheet in a number of ways. Many older specimen

s in the herbaria are frequently found to have been sewn on the sheets. Use of a dhesive linen, paper or cellophane strips is an easier and faster method of fixi ng specimens. Archer method involves the use of small strips of liq-uid plastic extruded from a container with a narrow nozzle. Most of the contemporary specime ns are fixed using liquid paste or glue often spreads after some years. There is no agreement as to the size of a herbarium label, the recommendations b e-ing as diverse as 23/4 by 41/4 inches (Jones and Luchsinger, 1986) and 4 by 6 inches (Woodland, 1991). The information commonly recorded on the herbarium label includes: Name of the institution in one of the two ways, however: Scientific name (i) Paste or glue is applied to the Common or vernacular name backside (if distinguishable) of the specimen, which is later pressed onto the m ounting sheet and allowed to dry in the pressed condition for a few hours. This method is slower but more Family Locality Date of collection Collection number Name of the collector economical. Habit and habitat including field notes (ii) Paste or glue is smeared on a glass or An expert visiting a herbarium may want to correct an identification or record a name change. Such correction is never done on the original label but on a small annotation label or determination label, usually 2 by 11 cm and appended left o f the original la-bel. This label, in addition to the correction, records the na me of the person and the date on which the change was recorded. Such information is useful, especially when more than one annotation label is appended to a herb arium sheet. The last label is likely to be the correct one. plastic sheet, the specimen placed on the sheet and the glued specimen transferr ed to a mounting sheet. This method is more efficient but expensive. The use of methylcellulose as adhesive mixed in a solution of 40% alcohol, inste ad of pure water was suggested by Tillet (1989) for fixing herbarium specimens. It decreases the drying time and also prevents growth of micro-organisms. The st em and bulky parts may often require adhesive strips or even sewing for secure f ixing of specimens. Small 104 Plant Systematics Voucher herbarium specimens of a research study often have authentic infor-matio n about the specimens recorded in the form of a voucher label. most commonwealth countries) or Engler and Prantl (Europe and North America). Ma ny herbaria of the latter category follow the number code of families and genera given by Dalla Torre and Harms (1900-1907). Filing of Specimens The specimens belonging to a species are placed in a folder made of thin strong paper, termed species cover. The species covers belonging to a particular genus are often arranged alphabetically and placed inside a genus cover, a heavy manil la folder made of a thicker paper. More than one genus cover may be used if the number of species are more, or if the specimens are to be arranged geographicall y, and often differently Mounted, labelled and treated (to kill insect pests) specimens are finally incor porated in a herbarium, where they are properly stored coloured for different geographical regions. The genus covers of a family are arranged according to the system of classificat ion being followed. The demarcation between the two families (last genus of a fa mily and first genus of the next family) is done using a sheet of paper with a f

ront-hanging label, indicating the name of the next family. The folders are stac ked in pigeonholes of the herbarium cases and the arrangement is suitable for sh ifting of folders as the number of specimens increase with time. Unknown specimens are kept in separate folders marked dubia, placed towards the end of a genus (when the genus is identified) or a family (when the family is id entified but not the genus), as the case may be, so that an expert can examine t hem conveniently. Standard herbarium cases are insect- and dust-proof with two o r more tiers of pigeonholes, each 19 in. deep, 13 in. wide and 8 in. high (Figure 5.12). Type specimens are usually kept separately in distinct folders or often in separ ate herbarium cases, sometimes even separate Figure 5.11 A sample herbarium sheet with mounted specimen and a label. rooms, for better care and safety. and looked after. Small herbaria arrange A herbarium commonly maintains an index register in which all the genera in the herbarium are listed alphabetically and specimens alphabetically according to family, genus and species. Larger herbaria, against each genus is indicated fami ly however, follow a particular system of number and the genus number, the two h elp classification. Most herbaria usually follow in convenient incorporation and retrieval of Bentham and Hooker (British herbaria and specimens in a herbarium. Process of Identification 105 Pest control Herbarium specimens are generally sufficiently dry, and as such not attacked by bacteria or fungi. They are, however, easily attacked by pests such as silverfis h, (c) The embryo in the seed gets killed, thus destroying a valuable source of experimental research, as seeds fro m herbarium specimens are often used for growing new plants for research projects. 2. Use of repellents: Chemicals with an offensive odour or taste are kept in herbarium cases to keep pests away from spe cimens. Naphthalene and Paradi-chlorobenzene (PDB) are commonly used repellents, usually powdered and put in small muslin bags kept in pigeonholes. PDB is more toxic and as such prolonged exposure of workers should be avoided. For people wo rking 8 hours a day in a 5 day per week schedule, the upper exposure level for n aphthalene is 75 PPM and Figure 5.12 Herbarium cabinet with filed specimens used in New York Botanical Garden Herbarium (Photograph courtesy New York B otanical Garden). dermestid beetles (cigarette beetle, drugstore beetle and black carpet beetle). for PDB 10 PPM. 3. Fumigation: In spite of pre-treatment Control measures include: of specimens and the use of repel-lents, fumigation is necessary for proper herb arium management. Fu-migation involves exposing specimens to the vapours of cert ain volatile toxic substances. A mixture of ethylene dichloride (3 parts) and ca rbon tetrachloride (1 part) was once commonly used for fumigation. Ethylene dich loride is explosive without carbon tetrachloride, but the latter is extremely to xic to humans, causing liver damage, and as such the use of this fumigant has be en banned. Some herbaria also use Ethylene bromide, Ethylene oxide, Lauryl penta chlorophenate (Mystox), Methyl bromide (often used synergistic with carbon dioxi de) or Malathion for fumigation, 1. Treating incoming specimens: Specimens have to be pest free before they can be incorporated into a her-barium . This is achieved in three

ways: (i) Heating at temperatures up to 60oC for 4-8 hours in a heating cabinet. The method is effective but the specimens become brittle. (ii) Deep-freezers have now replaced heating cabinets in most herbaria of the world. A temperature of 20 to 60oC is maintained in most herbaria. (iii) Microwave ovens have been used by some herbaria, but as indicated by Hill (1983), the use of microwave ovens has some serious but these are also toxic for humans. shortcomings including: Integrated Pest Management pro-gram (IPMP) for safer herbarium use emphasises pr evention of insect entering the herbarium, rather than relying on toxic chemical s to kill them once inside the herbarium. It also encourages the use of: (a) Stems containing moisture burst due to sudden vapori-zation of the water inside. (b) Metal clips, staples on the sheets get overheated and may char the sheet. 106 Plant Systematics a) Deep freezing not only to treat inmoth, cigarette beetles and warecoming specimens but also to kill insects in the herbarium. The bundles of speci mens are placed in plastic bag, the bag sealed, and the bundle placed in freezer for several days. The bag (sealed) is next left at room temperature for 7-10 da ys to allow any eggs to hatch and then refrozen for 3-4 house beetles. e) Insect electrocuters are useful for detecting and controlling flying insects. These emit ultraviolet light that attracts flying insects particularly flies and moths. days. Where fumigation is essential use of safer fumigants like Pyrethrin, sulphuryl f luoride (Vikane), dichlorovos (no pest, Vapona resin strips or Raid strips are s uitable for herbarium cases; one-third of a strip is placed in each herbarium ca se for seven to ten days twice a year.), carbon dioxide and cyanogen (often used synergistic with carbon dioxide). Dowfume-75 has been cleared by the Environmen tal Protection Agency for use in herbaria. b) Anoxic treatment, involves the usage of bag impermeable to oxy-gen and containing specimens and a oxygen scavan ger. The pest is killed by depriving it of oxygen.

Virtual Herbarium Virtual herbarium is a database of consisting of images of Herbarium specimens a nd the supporting text, available over the internet. It is a huge advancement in herbarium use and design, coupling physical specimens directly with internet an d integrating complete specimen data, with resources or information generation a nd retrieval. Although a virtual herbarium cannot exist without a physical herba rium, it enjoys several advantages over a physical Figure 5.13 The researchers comparing specimens inside the herbarium at the Missouri Botanical Garden. (Photograph by Jack

Jennings/ Courtesy the Missouri Botanical Garden.) herbarium: 1. Images being available electronically, c) Small sticky trap placed in hidden user may not have to handle physical specimens, thus reducing the damage areas of the herbarium and herbarium cases to trap insects. Such traps should be checked regularly substantially. for insects trapped. 2. Whereas it may take months to sort out specimens of a collector or a cound) Pheromone traps involve the use try in a physical herbarium, the same can be done done in few seconds of natural scents which insects use to communicate with each other. Certain inse cts are at-tracted to these traps from the surrounding area and are very ef-fect ive. Specific traps are available for drug store beetles, Indian meal through a virtual herbarium. 3. Virtual herbarium greatly increases the user interaction. Only few hundred visitors may visit a physical herbarium i n a month, but during the same period thousands of users can Process of Identification 107 access the virtual herbarium sitting Fairchild Tropical Garden Virtual Herbarium (eFTG) has record of more than 100,0 00 specimens, more than 200,000 photographs (including data labels). There are m ore than 20,000 high resolution photographs of specimens, that can be zoomed in or out of the browser. Nearly 60, 000 records are are searchable online by famil y, genus, collector and other fields. eFTG is the first truly Virtual Herbarium as Web portal of the herbarium allows simultaneous search throughVirtual Herbari a of FTG (Fairchild Tropical Garden), FLAS (Florida Museum of Natural History), MO (TROPICOS-Missouri Botanical Garden), NY Cassia- New York Botanical Garden), S (LinneanHerbarium, Swedish Museum of Natural History, Leiden), BM (Btitish Mus eum of Natural History-including Clifford Herbarium) CAYM (National Trust for Ca yman Islands), INB (Instituto Nacional de Biodiversidad, Costa Rica), TAMU (Texa s A & M University). The virtual herbarium of FTG thus affectively includes not only specimens from Fairchild Herbarium, but also from other herbaria. It also p rovides species lists, interactive keys and photographs of living specimens in v arious datain the comfort of their offices. 4. Physical herbarium usually stores only specimens, and the user has to spend considerable time in the library to collect relevant information. A virtual herbarium on the other hand provides informatio n on descriptive details, geographical distribution, photographs, illustrations, manuscripts, published work, microscopic preparations, gene sequences and nomen clature through hyperlinks. 5. Physical herbarium can offer only own specimens for study, whereas portals of major virtual herbaria offer facility of searching several virtual herbaria simultaneously. 6. Physical specimens are prone to damage through handling or during hazardous situation. Thousands of specimens and v aluable holotypes preserved in Berlin herbarium were destroyed during second the World War. Digitized images, on the other hand can be saved on several computers, at different locations. 7. Majority of research projects don t bases and indices. need physical specimens, and as such electronic images can be utilized, saving t he time and cost for transAustralia s Virtual Herbarium (AVH) is a collaborative project of the State, Commo n-wealth and Territory herbaria, being developed under the auspices of the Counc

il of Heads of Australian Herbaria (CHAH), rep-resenting the major Australian co llections. It is an on-line botanical information re-source accessible via the w eb, providing immediate access to the wealth of data associated with scientific plant specimens in each portation of actual specimens. Virtual herbaria with searchable database have been been developed by many major organisations like New York Botanical Garden (KE EMu), Royal Botanic Gardens Me lbourne (AVH), Fairchild Tropical Garden (E-FTG), Australian Virtual Herbarium ( AVH) and Royal Botanic Gardens, Kew (ePIC). Australian herbarium. Though initiated in 1990, the NYpc project of New York Botanical Garden became o perational in 1995. The data was transferred to new platform KE EMu in 2004 Australian herbaria house over six mil-lion specimens that are a primary source of information on the classification and distribution of plants, algae and fungi . These specimens are the working tools of scienwith additional search and display capabilities. The Virtual herbarium tists who contribute to our knowledge and pr esently consists of digital collection of understanding of biodiversity and cons erva850,000 herbarium specimens and 120,000 tion through the discovery, classifi cation high resolution images, updated daily. and description of new species. Th ese will Garden pursues the goal of digitizing all of be enhanced by images, descriptive text and its 7 million plant and fungi specimens. identification tools. 108 Plant Systematics The AVH is accessed via the website of any participating herbarium. A gateway at each of these herbaria links to the databases of all the other herbaria, consol idating the combined data into a nation-wide view of the botanical information. Most data related to specimens will be stored by the custodial institution, and there will be some re-sources, such as the scientific names data- base (Australi an Plant Names Index, APNI) which will be common to all. More than 70% of the sp ecimens housed in Australian herbaria have been databased, providing a comprehen sive resource for accurate depiction of geographic distribution and occurrence, historical mapping, information valuable for understanding the threatening proce sses of vegetation clearance and weed invasion. Flexible on-line search options allow you to customise the data you generate to suit your ceeding and as of June 2006 the database consisted of c193,600 specimens. Furthe r digitization is likely to focus on type specimens. A potential c7,000,000 spec imens (c275,000 types) may eventually be digitized. IDENTIFICATION METHODS requirements. Identification of an unknown specimen is a common taxonomic activity, and often com- bined with determination of a correct name. The combined activity is approp riately re- ferred as specimen determination. Before the specimen can be identif ied, it is desirable to describe it and prepare a list of char-acter-states, mai nly pertaining to floral structure. Whereas fresh specimens may be described mor e conveniently, the dried specimens may be softened by immersing in water or a w etting agent such as aerosol OT (dicotyl sodium sulfosuccinate 1 per cent, disti lled water 74 per cent and Methanol 25 per cent). Pohl s softening agent is an exc ellent detergent solution for soften-ing flowers and fruits for dissection. The identification of an unknown plant may be achieved by comparison with identified her-barium specimens or through the help of taxonomic literature. Both methods may be Australia s Virtual Herbarium provides the opportunity to deliver descriptions of the flora dynamically linked to data and infor-mation from across the continent, and dis-tributed on-line as an electronic Australian Flora - a one-stop source of current infor-mation on the plants, algae and fungi of the entire Australian

continent. New observa-tions can be released with minimal delay as they are conf irmed and recorded in the combined for a more reliable identification. database. The unknown specimen meant for identification is sent to a herbarium, where an e xpert on the plant group examines and identifies it by comparison with duly iden tified specimens (Figure 5.13). The user can also visit a herbarium and personal ly compare The Strong ePIC database software of Royal Botanic Gardens, Kew also provides a win-dow for digitized herbarium specimens. The Herbarium s core digital collection and identify his specimens. programme was initiated in 2002 and since then digital resources have grown at a n in-creasing rate, as well as central Herbarium Catalogue, have an image server and many project databases with information about specimens that were built bef ore the Cata-logue became available. These are being moved into the Catalogue as resources per-mit. Label data from dry and spirit specimens of flowering plants , ferns and gymnosperms held in Kew's herbarium are being uploaded. Information recorded includes the plant name, collection and determination data, locality an d type status. Digitization is proComputers have entered in a big way into solving identification problems. Electr onic revolution in recent years has opened up a new, faster and more reliable me thod of identification. The photograph, description or illustration of parts can be put up on a website, with information to a relevant e-mail list, whose membe rs can help in achieving identification within hours. Taxonomic literature Various forms of literature incorporating description, illustrations and identif ication Process of Identification 109 keys are useful for proper identification of unknown plants. The library is, the refore, as important in taxonomic work as a her-barium, and knowledge of taxonom ic litera-ture is vital to the practicing taxonomist. The literature of taxonomy is one of the oldest and most complicated literatures of science. Several bibli ographic references, indexes and guides are available to help taxonomists to loc ate relevant literature concerning a taxonomic group or a geographical region. T he major forms of literature helpful in iden-tification are described below. Australiensis by G. Bentham (1863 1878). 4. Comprehensive treatments have a much broader scope. Although no world Flora has ever been written, several impor tant works have attempted a worldwide view. Examples: Genera plantarum of G. Ben tham and J.D. Hooker (1862-83), Die Naturlichen pflanzenfamilien of A. Engler an d K.A. Prantl (1887-1915) and Das Pflanzenreich of A. Engler (1900-1954). Floras Electronic Floras (eFloras) Last few years have seen the online avail- ability of digitized form of many pop ular floras. These Online Floras known as Electronic Floras (eFloras) provide op purtunity for users to work dynamically on floristic treat-ments, and to browse and search these treat-ments. One such effort by Missouri Botanical Gardens has resulted in the publication A Flora is an inventory of the plants of a defined geographical region. A Flora may be fairly exhaustive or simply synoptic. Lists of the Floras may be found in the Geographical Guide to the Floras of the World by S.F. Blake (Part I, 1941; Part II, 1961) and Guide to the Standard Floras of the World by Frodin (1984). D epending on the scope and the area covof www.eFloras.org/, combining together the ered, the Floras are categorized as: information from several Floras including Flora of Chile, Flora of China, Flora of Missouri, Flora of North America, Flora of Pakistan, Moss Flora of China, Tre es and Shrubs 1. Local Flora covers a limited geo-

graphical area, usually a state, county, city, a valley or a small mountain of Andes and Ecuador, as also the Annotated range. Examples: Flora of Delhi by Checklist of Flowering Plants of Nepal. J.K. Maheshwari (1963), Flora These Floras can be searched through comSimlensis by H. Collet (1921), Flora of mon search engine to obtain relevant inforTamil Nadu by K.M. Mathew (1983), mation. The hyperlinks to families, genera Flora of Missour i by J.A. Steyermark and species are very handy in identification (1963) and Flo ra of Central Texas by R.G. and retrieving information. The website also Reeves (1972). hosts the interactive Actkey Provided by the Harvard University Herbarium, allow ing visitors to locate and use a key for identify-ing an unknown specimen. The k eys for Families of Angiosperms by Bertel Hansen & Knud Rahn, Families of Dicoty ledons of the Western Hemisphere South of the United States, Generic Tree Flora of Madagascar, Key to Taxa of China in ActKey, Trees and Shrubs of Borneo, and W eeds Of Rain Fed Lowland Rice Fields Of Laos And Cambodia are already incorporat ed in a user friendly 2. Regional Flora includes a larger geo-graphical area, usually a large country or a botanical region. Examples: Flora ofB ritish India by Sir J.D. Hooker ( 1872-1897), Flora Malesiana by C.G. Steenis (1948), Flora Iranica by K.H. Rechin ger (1963), Flora of Turkey and East Aegean Islands by P.H. Davis and Flora SSSR by V. L. Komarov and B.K. Shishkin (1934-1964). A Flora cover-ing a country is more appropriately known as a National Flora. interface. 3. Continental Flora covers the entire Royal Botanical Gardens Kew has hosted eFlora Flora Zambesiaca providing not onl y an easy way of searching the information continent. Examples: Flora Europaea by T.G. Tutin et al., (1964-80) and Flora 110 Plant Systematics but also an identification tool. This website allows you to search for a plant n ame across the whole Flora, which would otherwise entail looking up separate ind exes. It also allows the creation of lists of: endemics, spe-cies from a particu lar division or country, species that match a particular habit and of A revision is less comprehensive than a monograph, incorporating less introducto ry material and including a synoptic literature review. A revision includes a co mplete synonymy but the descriptions are shorter and often confined to diagnosti c characters. The geographical scope is usually worldwide. species that occur at a specific altitude. As A conspectus is an effective outline of a far as posible no changes have been made revision, listing all the taxa, with al l or ma-to the information existing in the original jor synonyms, with or withou t short diagnotext and the information is presented in the sis and with a brief mention of the geographisame way as in the original. cal range. Species plantarum of C. Linnaeus (1753) is an ideal example. Manuals A synopsis is a list of taxa with much abbreviated diagnostic distinguishing sta te-ments, often in the form of keys. A manual is a more exhaustive treatment than a Flora, always having keys for ide nti-fication, description and glossary but gener-ally covering specialized group s of plants. Examples: Manual of Cultivated Plants by L.H. Bailey (1949), Manual of Cultivated Trees and Shrubs Hardy in North America by A. Rehder (1940) and M anual ofA quatic Plants Icones (Illustrations) by N.C. Fassett (1957). Illustrations, often with detailed analysis of the parts are usually published a long with the text in Floras and Monographs, but may sometimes be compiled exclu sively and of-ten serve as useful tools for identification. In fact, many specie

s of plants based on pub-lished illustrations only, without any accompanying des cription or diagnosis before 1 January 1908 have been accepted as validly publis hed. Two principal compilations of Icones are Hooker s Icones and Wight s Icones. Ot hers of interest include Illustrations of plants from Europe (Hegi, 1906-1931), North America (Gleason, 1963), Pacific states (Abrams, 1923-1960), Pacific coast trees (McMinn and Maino, 1946), Germany (Garcke, 1972), Korea (Lee, 1979). A manual differs from a monograph in the sense that the latter is a detailed tax onomic treatment of a taxonomic group. Monographs A monograph is a comprehensive taxonomic treatment of a taxonomic group, general ly a genus or a family, providing all taxonomic data relating to that group. Usu ally the geographical scope is worldwide since it is impossible to discuss a tax on without includ-ing all its members, and often all its species, subspecies, va rieties and forms are discussed. The monograph also includes an exhaustive revie w of literature, as also a report on author s research work. A mono-graph includes all information related to nomenclature, designated types, keys, exhaustive des cription, full synonymy and citation of specimens examined. Examples: The Genus Pinus by N.T. Mirov (1967), The Genus Crepis by E. B. Babcock (1947), A Mono-gra ph of the Genus Avena by B.R. Baum (1977), The Genus Datura by A.F. Blakeslee et al., (1959) and The Genus Iris by W.R. Dykes (1913). Journals Whereas Floras, manuals and monographs are published after a lot of taxonomic in put and it may take several decades before they are revised, if at all, taxonomi c journals pro-vide information on the results of ongoing research. A continuous update on additional taxa described or reported from a region, nomenclatural ch anges and other taxonomic information is essential for continuance of taxonomic activity. Reference to a publication in a journal includes volume number (all is sues within a year bear the same Process of Identification 111 volume number; trend not followed by a few journals), issue number (numbered wit hin a volume, a monthly journal would have 12 issues, quarterly 4 issues and so on) and page numbers on which a particular article appears. Common journals devo ted largely to taxonomic research include: ral History), Paris; a peer-reviewed journal of plant biology, devoted to the in ventory, analysis and interpretation of vascular plants biodiversity; publishes original results, in French or English, of botanical research, particularly in s ystematics and related fields; two issues appear each year. Adansonia continues as from 1997 the Bul- letin du Musum national d'Histoire naturelle, Taxon- The journal of the International Association for Plant Taxonomy devoted t o systematic and evolutionary biology with emphasis on botany; published quarter ly by the International Bureau for Plant Taxonomy and Nomenclature, Botanisches Institut der Universitaet Wien, Austria. section B, Adansonia, Botanique, phytochimie. Other important taxonomic journals in-clude Journal of the Arnold Arboretum (Har vard), Bulletin Botanical Survey of India (Calcutta), Botanical Magazine (Tokyo) and Systematic Botany (New York). Kew Bulletin- International peer-re-viewed Journal of Plant Taxonomy; published in four parts in one year by Royal Botanic Gardens, Kew; containing original art icles of interest mainly to vascular plant and mycological systematists; each pa rt illus-trated with line drawings and photographs and also features a Book Revi ew and Notices section. Supporting literature With a large amount of research material being published throughout the world, t here is always need for supporting literature to give consolidated information a bout the works published the world over. They also help in tracking down materia l concerning a particular taxon covering a certain period. Taxonomic Literature, an exhaustive series of Regnum vegetabile, covers full bibliographical details of literature extremely helpful in searching type material, priority of names, d ates of publication and biographic data on authors. Originally published in 1967 , it is under constant revision with 3 supplements of the 2nd edition published

between 1992 97 (Stafleu and Mennega). Plant Systematics and EvolutionPublished by Springer, Wien from 1974 on-wards; originally started in 1851 under the name sterreichisches Botanisches Wochenblatt was published between 1958 to 1 973 as sterreichische Botanische Zeitschrift; devoted to publishing original pape rs and reviews on plant systematics in the broad-est sense, encompassing evoluti onary, phylogenetic and biogeographical studies at the populational, specific, h igher taxonomic levels; taxonomic emphasis is on green plants; volumes each with four numbers published randomly, usually 6-7 volumes in one year. Abstracts or Abstracting journals: These provide a summary of different articles published in various journals thro ughout the world. Biological Abstracts and Current Advances in Plant Science are more general in approach. The Kew Record of Taxonomic Literature covers all art icles relevant to taxonomy. Botanical Journal of Linnaean Society- Published on behalf of Linnean Society by Blackwell Synergy, London; three volumes with four monthly issues each pub-lish ed in one year; publishes original re-search papers in the plant sciences. Index: An Index provides an alphabetic listing of taxa with reference to their p ublication. Index Kewensis is by far the most important reference tool, first pu blished in 2 volumes from Royal Botanic Gardens, Kew (1893-1895), covering names of species and genera of seed plants published between Adansonia- Published by Musum na-tional d'Histoire naturelle (Museum of Natu112 Plant Systematics AMEBIA, Regel, Pl. Nov. Fedsch. 58 (1882) err. typ Arnebia, Forsk. (Boragin.). AMERCARPUS, Benth in Lindl. Veg. Kingd. 554 (1847) Indigofera, Linn. (Legumin.). AMECHANIA, DC. Prodr. vii. 578 (1839)= Agarista, D. Don (Ericaceae). hispidula, DC. l. c. 579 (=Leucothoe hispidula). subcanescens, DC. l. c. (=Leucothoe subcanescens). AMELANCHIER, Medic. Phil. Bot. i. 135 (1789). ROSACEAE, Benth. & Hook.f. i. 628. ARONIA, Pers. Syn. ii. 39 (1807. PERAPHYLLUM, Nutt. in Torr. & Gray, Fl. N. Am. i. 474 (1840). XEROMALON, Rafin. New Fl. Am. iii. 11 (1836). alnifolia, Nutt. in Journ. Acad. Phil. vii. (1834)22. Amer. bor. asiatica, Endl. in Walp. Rep. ii. 55= canadensis Bartramiana, M. Roem. Syn. Rosifl. 145= canadensis. Botryapium, DC. Prodr. ii. 632= canadensis. canadensis, Medic. Gesch. 79; Torr. & Gray, Fl. N. Am. i. 473. __Am. bor.; As. or. chinensis, Hort. ex Koch, Dendrol. i. 186=Sorbus arbutifolia. Figure 5.14 Portion of a page from Index Kewensis. Generic name Amebia (Normal c aps) Regel is synonym of genus Arnebia (Bold small case) Forsk. of family Boraginaceae. Generi c name Amelanchier (Bold caps) Medic. is correct name with generic names Aronia Pers., Peraphyllum Nutt. and Xeromalon Rafin. as synonyms. Species names Amelanc hier alnifolia (Normal small case) Nutt. and A. canadensis Medic. are correct, w hereas the names A. asiatica (italics small case) Endl., A. Batramiana M. Roem.a nd A. Botryapium DC. are synonyms of A. canadensis. A. chinensis Hort. is simila rly synonym of Sorbus arbutifolia. 1753 and 1885. Regular Supplements used to be published every 5 years and 18 Sup ple-ments appeared up to 1985. Supplement 19 was published in 1991 covering the

years 1986 to 1990. Since then the listing has been published annually under the title Kew with bibliographic references to the place of first publication. At the beginnin g of the nineteen eighties the data was transferred to a computer database which continues to expand at the rate of approximately 6000 records per year. To make this data generally available, it was decided to publish the whole Index Kewens is as a CD-ROM in 1993. This contains almost 968,000 records. Index. Index Kewensis (Figure 5.14) is a list of new and changed names of seed-bearing plants Process of Identification 113 Illustrations of vascular plants can be lo-cated through Index Londinensis, whic h con-tains information up to 1935. More recent information can be found in the 2-volume work Flowering Plant Index ofI llustrations and was produced by Dr. Hu Shiu-ying (Arnold Arboretum of Harvard University) and hi s staff. The Hu Card Index was prepared in the early 1950s when the Arnold Arbor etum undertook a project to prepare a flora of China. Information compiled by R. T. Isaacson (1979). Royal Botanic Gardens Kew, The Harward University Herbaria, and the Australian N ational Herbarium, under the collaborative project, have developed International Plant Names Index (IPNI), a single web database which combines citation data fo r seed plants from Index Kewensis, the Gray Herbarium Card Index, and the Austra lian Plant Names Index (APNI). It provides information on names and associated b asic bibliographical details of all seed plants. Its goal is to eliminate the ne ed for repeated reference to primary sources for basic bibliographic information about plant names. The data are freely available and are gradually being standa rdized and checked. IPNI is intended to be a dynamic resource, depending on dire ct contributions by all members of the botaniA listing of all generic names can be found in Index Nominum Genericorum (ING) a 3-vol-ume work published in 1979 under the se-ries Regnum Vegetabile. The first supplement appeared in 1986. It has now been put on the database and can be dir ectly accessed through the Internet. Index Holmiensis (earlier Index Holmensis) is an alphabetic listing of distribut ion maps found in taxonomic literature of vascular plants. It commenced publication in 1969. Gray Herbarium Card Index is information on cards, which has now been set up on a database. Usually on the same pattern as Index Kewensis, the Index has been pu b-lished in 10 volumes between 1893 and 1967. A 2-volume supplement was publishe d by G. K. Hall in 1978. The Gray Herbarium Index Database currently includes 35 0,000 records of New World vascular plant taxa at the level of species and below . The Index in-cludes from its 1886 starting point, the names of plant genera, s pecies and all taxa of infraspecific rank. The Gray Index has in common with Ind ex Kewensis its involve-ment with taxon names, although they dif-fer in biologic al and geographical coverage. The Gray Index covers vascular plants of the Ameri cas; Index Kewensis includes seed cal community. Numerous valuable Dictionaries have been published but by far the most useful is Dictionary ofF lowering Plants and Ferns published by J. C. Willis. The 8th edi tion revised by Airy Shaw appeared in 1973. The book contains valuable informati on concerning genera and families providing name of the author, distribution, fa mily and the number of species in the genus. plants worldwide. Only the Gray Index has Taxonomic Keys nomenclatural synonyms cross-referenced Taxonomic keys are aids for rapid identi to basionyms. The information is now ac- fication of unknown plants. They consti cessible over the Internet via keyword tute important component of Floras, manus earches from the E-mail Data Server and als, monographs and other forms of liter athrough the Biodiversity and Biological Col- ture meant for the identifying pla

nts. In adlections Gopher. Indices covering other dition, identification methods in recent groups of plants have also been published: years have incorporated th e usage of keys Index Filicum for Pteridophytes, and Index based on cards, tables and computer p roMuscorum for Bryophytes. grams. The latter are primarily designed for identification by non-professionals . These keys are fundamentally based on characters, which are stable and reliabl e. The keys are helpful in a faster preliminary identificaThe Hu Card Index is a file of 158,844 cards for Chinese plant names, now housed in the Harvard University Herbaria building where it is available for use in person. The Index 114 Plant Systematics tion, which can be backed up by confirma-tion through comparison with the detail ed description of the taxon provisionally iden-tified with. Before identificatio n is at-tempted, however, it is necessary that the unknown plant is carefully st udied, de-scribed and a list of its character states pre-pared. Based on the arr angement of charac-ters and their utilization, two types of idenated, spur absent, petals without nectary. 3. Anemone: Plants herbaceous, fruit achene, calyx not differentiated, perianth petaloid, spur absent. 4. Clematis: Plants woody, fruit achene, calyx not differentiated, perianth petaloid, spur absent. 5. Caltha: Plants herbaceous, fruit follitification keys are differentiated: cle, calyx not differentiated, perianth 1. Single-access or sequential keys; and petaloid, spur absent. 2. Multi-access or multientry keys 6. Delphinium: Plants herbaceous, fruit (polyclaves). follicle, calyx not differentiated, perianth petaloid, spur one in number. Single Access or Sequential 7. Aquilegia: Plants herbaceous, fruit folKeys Single-access keys are usual components of Floras, manuals, monographs and other books meant for identification. The keys are based on diagnostic (important and conspicuous) characters (key characters) and as such the keys are known as diag nostic keys. Most of the keys in use are based on pairs of con-trasting choices and as such are dichotomous keys. They were first introduced by J. P. Lamarck in his Flore Francaise in 1778. The construction of a dichotomous key starts with the preparation of a list of reliable char-acters for the taxon for which the ke y is to be constructed. For each character the two con- trasting choices are det ermined (e.g., habit woody or herbaceous). Each choice constitutes a lead and th e two contrasting choices form a couplet. For characters having more than two av ailable choices the character can be split to make it dichotomous. Thus if flowe rs in a taxon could be red, yellow or white the first couplet would constitute f lowers red vs. non-red and the second couplet flowers yellow vs. white. We shall illustrate the construction of keys taking an example from family Ranunculaceae . The diagnostic characters of some representative genera are licle, calyx petaloid, not differentiated from corolla, spurs five in number. Based upon the above information the following couplets and leads can be identified: 1. Plants woody Plants herbaceous 2. Fruit achene

Fruit follicle 3. Calyx and corolla differentiated Calyx and corolla not differentiated 4. Spur present Spur absent 5. Number of spurs 1 Number of spurs 5 6. Petal with nectary at base Petal without nectary at base It must be noted that three choices are available for spur (absent, one, five). It has been broken into two couplets to maintain the dichotomy. Based on the arr angement of couplets and their leads, three main types of dichotomous keys are i n use: Yoked or Indented key, Bracketed or parallel key, and Serial or numbered key. listed below: 1. Yoked or Indented key: This is one of the most commonly used keys in Floras a nd manuals especially when the keys are smaller in size. In this type of key, th e state-ments (leads) and the taxa identified from them are arranged in visual g roups or yokes and additionally the subordinate couplets are indented below the primary one at a fixed distance from the margin, the distance increasing with ea ch subordinate couplet. We 1. Ranunculus: Plants herbaceous, fruit achene, distinct calyx and corolla, spur absent, petal with nectary at base. 2. Adonis: Plants herbaceous, fruit achene, calyx and corolla differentiProcess of Identification 115 1 Stem woody at base; achenes 3.5-5 mm S. pustulatus 1 Stem not woody; achenes 2-3.75 mm 2 Annual or biennial 3 Achenes smooth at least between the ribs; strongly compressed and + winged 1. asper 3 Achenes rugose or tuberculate between the ribs, neither strongly compressed nor winged 4 Leaf-lobes strongly constricted at base, or narrowly linear; terminal lobe usu ally about as large as lateral lobes; ligules longer than corolla-tube; achenes abruptly contracted at base 2. tenerrimus 4 Leaf-lobes (if present) not constricted at base; terminal lobe usually much la rger than lateral lobes; ligules about as long as corolla-tube; achenes graduall y nar rowed at base 3. oleraceous 2 Perennial Figure 5.15 Portion of a polythetic key of the yoked type used in Flora Europaea for genus Sonchus (vol. 4, p. 327). shall select the fruit type as the first cou-plet, as it divides the group into two almost equal halves and the taxa excluded would be almost equal whether the fruit in the un-known plant is an achene or a follicle. The yoked or indented ke y for the taxa under con-sideration is shown below: those appearing on a single page, but if the key is very long running into sever al pages, an Indented key exhibits important draw-backs. Firstly, it becomes dif ficult to locate the alternate leads of initial couplets, as they may appear on any page. Secondly, with the number of subordinate couplets increasing substanti ally, the key becomes more and more sloping, thus reducing the space avail-able for writing leads. This may result in wastage of a substantial page space. The p

roblem is clearly visible in Flora Europaea where attempts to reduce the indenta tion distance in longer keys has further complicated the usage of keys. These tw o disadvantages are taken care of in the Parallel or 1. Fruit achene. 2. Calyx differentiated from corolla. 3. Petal with basal nectary 1. Ranunculus 3. Petal without basal nectary..2. Adonis 2. Calyx not differentiated from corolla. 4. Clematis 4. Plants woody 4. Plants herbaceous 3. Anemone 1. Fruit follicle. 5. Spur present. Bracketed key. 6. Number of spurs 1 6. Delphinium 2. Bracketed or Parallel key: This type of key has been used in larger floras su ch as Flora of USSR, Plants of Central Asia, and Flora of British Isles. The two leads of a couplet are always together and the distance from the margin is alwa ys the same. Several variations of this are used wherein the second lead of the couplet is not numbered, as in Flora of British Isles or else the second lead is prefixed with a + sign as in Plants of Central Asia. The arrangement of couplet s in this type of key is useful for longer keys 5. Spur absent 6. Number of spurs 5 7. Aquilegia 5. Caltha It is important to note that all genera with achene fruit appear together and fo rm vi-sual groups; leads of subordinate couplets are at increasing distance from the margin and the leads of initial couplets are far separated, whereas those o f subsequent subordinate couplets are closer. Such an arrangement is very useful in shorter keys, especially 116 Plant Systematics as the location of alternate keys is no prob-lem (two are always together) and t here is no wastage of page space. There is, however, one associated drawback; th e statements are no longer in visual groups. The reference to primary lead is of ten difficult, but this problem is usually solved by indicating the number of pr imary lead within parenthesis as done in several Russian Floras such as Flora Si beria and Plants of Central Asia. A typical bracketed key is illustrated below: 8. Number of spurs 5 7. Aquilegia 9. (7) Spur absent 5. Caltha Such a key retains the visual groups of statements and taxa, alternate leads, ev en though separated, are easily located and the there is no wastage of page space. 1. Fruit achene 2 An inherent drawback of dichotomous keys is that the user has a single fixed cho ice of the sequence of characters decided by the person who constructs the key. In the said example if information about the fruit is not available, it is not p ossible to go be-yond the first couplet. 1. Fruit follicle 5 2. Calyx differentiated from corolla 3 Guidelines for dichotomous keys Certain basic considerations are important for the construction of dichotomous k eys. 2. Calyx not differentiated from corolla .... 4

3. Petal with basal nectary ... 1. Ranunculus 3. Petal without basal nectary 2. Adonis 4. Plants woody 4. Clematis These include: 4. Plants herbaceous .3. Anemone 1. The keys should be strictly dichoto5. Spur present 6 mous, consisting of couplets with only 5. Spur absent 5. Caltha two possible choices. 6. Number of spurs 1 6. Delphinium 2. The two leads of a couplet should be 6. Number of spurs 5 7. Aquilegia mutually exclusive, so that the accept-ance of one should automatically lead Retention of positive features of the Par-allel key and visual groups of the Yok ed key is achieved in the Serial key. to the rejection of another. 3. The statements of the leads should not 3. Serial or numbered key: Such a key has been used for the identification of an imals and also adopted in some botanical works. This key retains the arrangement of Yoked key, but with no indentation so that distance from the margin remains the same. The location of alternate leads is made possible by serial numbering o f couplets (or leads when separated) and indicating the serial number of the alt ernate lead within parentheses. A serial key for the taxa in question would appe ar as under: be overlapping. Thus, the two leads leaves 5-25 cm long and leaves 20-40 cm long wou ld find it difficult to place taxa with leaves that are between 20 and 25 cm in length. 4. The two leads of a couplet should start with the same initial word. In our example, both leads of the first couplet start with Fruit . 5. The leads of two successive couplets 1. (6) Fruit achene. 2. (4) Calyx differentiated from corolla. should not start with the same initial word. In our example the word spur appears in two successive couplets and as such in the second one the language has been c hanged to start with Number . If such a change were not possible it would be conven ient to pre-fix the second couplet with The . Thus, the other alternative for the s econd couplet would have the two leads worded 3. Petal with basal nectary... 1. Ranunculus 3. Petal without basal nectary 2. Adonis 4. (2) Calyx not differentiated from corolla. 5. Plants woody 4. Clematis 5. Plants herbaceous 3. Anemone 6. (1) Fruit follicle. 7. (9) Spur present. as The spur 1 and The spurs 5 . 8. Number of spurs 1 6. Delphinium 6. For identification of trees, two keys should be constructed based on Process of Identification 117

vegetative and reproductive charac-ters separately. As trees commonly have leave s throughout the major part of the year, and flowers appear briefly when in many trees leaves are not yet developed, such separate keys are essential for identi fication round the 3. The single character used in the couplet may be exceptional. Such likelihood is not possible when more than one char acter is used. Multi-Access Keys (Polyclaves) year. Such multientry order-free keys are user-oriented. Many choices of the sequence of characters are available. Eventually, it is the user who decides the sequence in which to use the characters, and even if the information about a few charact ers is not avail-able, the user can go ahead with identification. Interestingly, identification may often be achieved without having to use all the characters a vailable to the user. Such identification methods often make use of cards. Two b asic types of cards are in use: 7. Avoid usage of vague statements. Statements such as Flowers large vs. Flowers small may often be confusing during actual identification. 8. An initial couplet should be selected in such a way that it divides the group into more or less equal halves, and the character is easily available for study. Such a selection would make the process of exclusion faster, whichever lead is selected. 9. For dioecious plants, it is important Body-punched cards These cards are also named window cards or peek-a-boo cards, and make use of car ds with appropriate holes in the body of the card to have two keys based on male and female flowers separately. 10. The leads should be prefixed by numbers or letters. This makes location of leads easier. If left blank, the loca-ti on of leads is very difficult, especially (Figure 5.16). The process involves using one in longer keys. card for one attribute (character-state). In our example we shall need 11 cards (we have chosen only diagnostic characters above, whereas our list in polyclaves could include more characters, and thus more cards to The keys described above have a single character included in a couplet, with two contrasting statements about the character in the two leads. Such keys are know n as monothetic sequential keys. The common-est forms of keys used in floras, ho wever, have at least some couplets (Fig 5.15) with several statements about the different char-acters in each lead. These keys are known as polythetic sequentia l keys. Such polythetic keys, also known as synoptic keys are especially useful for constructing keys for higher categories. Such keys have three make it more flexible). It should be noted that we selected 12 leads and 6 couplets, with 4 leads for sp ur. Now we shall need only three actual attributes: spur absent , spur 1 and spurs 5 . N umbers are printed on the cards corresponding to the taxa for which the identifi cation key is meant. In our example, we use only 7 of these numbers correspondin g to our 7 genera. On each card, holes are punched corresponding to the taxa in which that attribute is present. basic advantages over the monothetic keys: In our example card Habit woody will have only one hole at number 4 (genus Clemati s), and the card Habit herbaceous will have holes at 1,2,3,5,6,7 (all seven except number 4). Once the holes are punched at appropriate positions in all the cards , we are ready for identification. The user studies the unknown plant and makes a list of characters, according to the sequence he wishes and the characters tha t are available to him 1. One or more characters may be

unobservable due to damage or nonoccurrence of requisite stage in the specimen. In such cases, a monothetic key becomes useless. 2. User can make a mistake in deciding about a single character. This error gets minimized if more than one character i s used. 118 Plant Systematics I Plants Herbaceous 4 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Figure 5.16 A body-punched card for herbaceous habit for the seven representativ e genera of Ranunculaceae: 1- Ranunculus, 2- Adonis, 3- Anemone, 4- Clematis, 5- Caltha, 6Delphinium, 7- Aquilegia. Note the diagonal trim on upper left corner of card fo r proper alignment of cards.

The user starts the identification process by picking up the first card concerni ng the first attribute in his list of attributes of the unknown plant. He next p icks up the sec-ond card concerning the second attribute from his list and place s it over the first card. This will close some holes of the first card and some of the second card. Only those holes will remain open which correspond to the ta xa, which contain both the at-tributes. The third card is subsequently placed ov er the first two and the process is repeated with additional cards until finally only one hole is visible through the pack of selected cards. The taxon to which this hole corresponds is the identification of the un-known plant. hole is clipped out to form an open notch instead of a circular hole along the edge. For actual identification, all the cards are held together as a pack. A needle i s inserted in the hole corresponding to the first at-tribute of the unknown plan t. As this needle is lifted up the taxa containing this attribute would fall dow n, and those lacking that at-tribute would remain in the pack lifted by the need le. The latter are rejected. The cards falling down are again arranged in a pack , the needle inserted in the hole corresponding to the next attribute of the unk nown plant. The process is repeated until finally a single card falls down. The taxon, which this card represents, is the identification of the unknown plant. Edge-punched cards Note that we may not have to explore all attributes of the unknown plant; identi fication may be achieved much before we have reached the end of the list of attr ibutes of the unknown plant. An edge-punched card differs from the body punched card in that there is one car d for each taxon and holes are punched all along the edge of the card, one for e ach attribute. In our example here, we shall need seven cards, one for each genu s. These holes are normally closed along the edge (Figure 5.17). For each attrib ute, present in the taxon the Tabular keys Tabular keys are essentially similar to the polyclaves in the sense that they ca n take care of exhaustive lists of attributes and are Process of Identification 119 Figure 5.17 Edge-punched card for genus Ranunculus. Only the attributes represen ted in the example above are pictured. Many more attributes could be added along the vacant holes to make the identification process more versatile. easier to use. The data are incorporated, however, not on cards but in tables wi th taxa along the rows and attributes along the col-umns. The attributes represe nted in each taxon are pictured with the help of appropri-ate symbols or drawing s (Figure 5.18). The attributes not represented in a taxon show a blank space in the column. Thus the table will have as many rows as taxa and as many columns a s the number of attributes for formulae are arranged in alphabetic order in the same manner as words in a dicti onary. Based on the attributes of the unknown plant, its taxonomic formula is co nstructed. The next step is as simple as locating a word in the dictionary. The formula is located in the alphabetic list and its identification read against the formula. which information is available. The above example of Ranunculaceae could be extended here by assigning alpha-bet s to the attributes: A: Woody; B: Herbaceous; C: Achene; D: Follicle; E: Spur ab sent; F: Spur 1; G: Spurs 5; H: Calyx differentiated from corolla; I: Calyx not differentiated from corolla, only perianth present; J: Nectary The identification process begins with a strip of paper whose width is equal to each row and vertical lines separated by the width of the columns. The attribute s present in the unknown plant are pictured on this strip of paper. The strip of paper is next placed towards the top of the table and slowly low-ered and compa red with each row. The row with which the entries match represents the identific ation of the unknown plant. present; K: Nectary absent.

The seven representative genera would thus have the formulae as given below: ACEIK Clematis BCEHJ Ranunculus BCEHK Adonis BCEIK Anemone Taxonomic formulae BDEIK Caltha A taxonomic formula is really an alphabetic formula based on a specific combinat ion of alphabets. The various attributes in this method are coded with alphabets . Each taxon thus gets a unique alphabetic formula. These BDFIJ Delphinium BDGIK Aquilegia Such formulae are really useful in the identification process and have been inco rporated in the written version of the multi120 Plant Systematics

F-

i i i i -

i i i

i i i _

Figure 5.18 Tabular key for the identification of representative genera of famil y Ranunculaceae. Only selected attributes as in the above example are pictured. More attributes c ould be added in additional columns to make the identification process more vers atile. access key to the Genera of Apiaceae in the Flora of Turkey (Hedge and Lamond, 1 972). tween the user and the computer. The computer program starts with the first coup

let of the key, enquires about the attribute in the unknown plant and on the inf ormation provided, and handles the key asking relevant questions until finally t he actual identification is achieved. Computers in Identification Over the years, computers have been in-creasingly used in data collection, proce ss-ing and integration. They have also found use in a big way in scanning and id entify-ing human ailments, which has greatly helped health management programmes . Computers have also found use in plant iden-tification, whereby we no longer n eed trained botanists for this task. The follow-ing main approaches are used in computer identification: Computer-Constructed Keys Appropriate programs may be developed which can construct a taxonomic key based on the taxonomic information about the taxa, in the same way and based on the sa me logic which is used by man to construct keys manually. Such keys permanently stored in a computer can be handled as above for the step-wise process of identi fication. Computer-Stored Keys Dichotomous keys are constructed in the usual manner, fed into a computer and ru n using an appropriate program, which may be appropriately designed for step-wis e pro-cessing of the key through a dialogue beSimultaneous Character-set Identification Taxonomic keys are an aid to rapid identification and always provide only a prov isional R an un c u-2 us Ad on]. s Anemone C..emat-2 Ca.2 tha De.2 phi n-IUm Aqui.2eg3a Process of Identification 121 identification, confirmation being achieved only after comparison with a detaile d descrip-tion of the specific taxon. This comparison with the detailed descript ion is not done in the first place, as comparing the description of the unknown plant with the description of all taxa of the group or the area would be la-bori ous, time consuming and often impos-sible. Such a comparison can be achieved thr ough a computer in a matter of seconds. With such an approach, the whole set of char-acters of the unknown plant may be fed into the computer simultaneously, an d a com-puter program used to compare the descrip-tion with the specific group a nd to suggest the taxon with which the description matches. In case complete inf ormation is not available, the computer program may be able to suggest possible alternate identifications. (DEscription Language for TAxonomy), which is a flexible and powerful method of recording taxonomic descriptions for pro-cessing by a computer. DELTA, a sharewa re program, has been adopted as a standard for data exchange by the Internationa l Taxonomic Databases Working Group. It enables the generation and typesetting o f descriptions and conventional keys, conversion of DELTA data for use by classi fication pro-grams, and the construction of Intkey packages for interactive iden tification and information retrieval. The System developed in the Natural Resour ces and Biodiversity Program of the CSIRO (Commonwealth Scientific and Industria l Research Organisation, Australia) Division of Entomology over a period of 20 y ears M.J. Dallwitz, T.A. Paine and E.J. Zurcher, is in use world-wide for divers

e kinds of organisms, including fungi, plants, and wood. The programs are contin ually refined and enhanced in response to feedback Automated Pattern Recogni-tion Methods from users. Computer technology has now developed to the extent that fully-automated identif ica-tion can be achieved. The computer fitted with optical scanners can observe and record features, compare the same with those al-ready known and make importa nt conclu-sions. Programs and techniques are already available for human diagnos is, including chemical spectra and photomicrographs of chromosomes, abnormality in human tis-sues and even in vegetation and agricultural surveys. The DELTA program Key generates con-ventional identification keys. Characters ar e selected by the program for inclusion in the key based on how well the charact ers divide the remaining taxa. This information is then balanced against subject ively determined weights, which specify the ease of use and reliability of the characters. Interactive Keys DELTA data can be readily converted to the forms required by programs for phylog enetic analysis, e.g. Paup, Hennig86 and MacClade. The characters and taxa for t hese analyses can be selected from the full dataset. Numeric characters are conv erted into multistate characters, as numeric char-acters cannot be handled by th ese programs. Printed descriptions can be generated to faLast two decades have seen the development of sophisticated computer based progr ams designed to collect, integrate and use it for organising descriptions and as sociated taxon data and also help in the identification of taxa through user fri endly interfaces. Some of the Major ones are briefly described here. cilitate checking of the data. Setting up a simple Delta identifiDELTA System cation: Although the DELTA system has capabilities of setting up of strong and s ophisticated identification procedures, a simple one can be built with basic kno wledge in computers. The first step in the pro-cess is to create a new data set (the existThe DELTA System is an integrated set of programs based on the DELTA format 122 Plant Systematics INTKEY : S. .x Fruit type Anemone Nectary Adonis Perianth Caltha Flower symmetry Delphinium Aquilegia Used Characters (1) Eliminated Taxa (1) Habit Herbaceous (1) Clematis ~ ~ ; 8:27 PM " ~{ ~) t '' ;Start 2112... Figure 5.19 Identification window of the Intkey (version 5) for the seven genera of Ranunculaceae. Selection of character state herbaceous (shown in the panel of used characters) leads to the rejection of one taxon (Clematis, which has woody habit) shown in t he panel of eliminated taxa. Selection further character states of the known pla

nt would eliminate further taxa till only one identified taxon remains. ing one can also be used, some even down-loaded from the internet). Create a new folder under Delta directory and give it an appro-priate name. Open Delta Edito r and click New Dataset from menu. This will open Attribute editor with 4 panels . A click in the upper left panel will open Item editor for the first taxon. Ent er its name (images, com-ments and change of settings can be added later on) and click Done to come back to the Attribute editor (else add image, sound, change settings and then click done). Now click in the upper right panel will open }cha racter editor. Give appropriate name or description to the character. Select the char-acter type from the list of Unorde red multistate (Say for flower colour with character states yellow, red, white, etc.; multistate includes binary characters also such as woody and herbaceous ha bit), Or-dered multistate (height range such as 1-10 cm, 11-20 cm, 21-30 cm, etc .; similarly two states with plants up to 20 cm tall and more than 20 cm tall), Integer numeric (say leaves per node), Real numeric (seed size say 2.4 cm) or te xt information (say about habitats). If Multistate character has been selected ( ordered or unordered), click states Process of Identification 123 tab (if not already done), enter first Charac-ter state in the lower right panel , it will automatically be defined in the left panel. Now click below this entry in the left panel and enter second Character state in the right panel. Repeat t his till all states (pos-sible in the other taxa included in the iden-tification process, but yet to be entered) have been entered. Next select character Num-be r 2, give it a name, and select type. If the numeric (real or integer) has been selected, states tab won t appear. You will select the unit (cm, mm, leaves per no de). For text char-acter, add appropriate notes in the notes tab. After all the characters have been selected, Action sets again, select Translate into Key Format and click Run and proceed simi larly. Now open Action sets again, change Tab from Confor to Key, select Confirma tory character RTF and click Run. In the Action sets again now change the Tab bac k to Confor, select Translate into Intkey format and click Run. Go to the Action s ets for the last time, change to Intkey Tab, select Intkey initializing File and c lick Run. The process will complete and Intkey program window will open (don t for get to add dataset to the Intkey Index when prompted when you close Intkey progr am; or else add dataset when you open Intkey program window next time) with four panes with the list of characters in the upper left pane and the list of taxa i n the upper right pane, both lower panes being click Done to go back to the Attribute editor. Now click in the left upper panel to add the second taxon and repeat this till a ll the taxa have been added. The Attribute editor will now show a list of taxa i n the upper left panel and the characters (identified as high-lighted U-unordere d multistate, O-ordered multistate, I-integer numeric, R-real numeric and T-text ) in the upper right panel. Any character missing from the list can be added and appropriately defined. Now select taxa one by one. For each taxon, enter (verif y) the state in the right panel after expanding the character icon (+ not expand ed, ex-panded) till information for all the taxa has been entered. Save the data set under the folder already created in the beginning. You can open the dataset now to add any images, comments or change settings if desired. The identificatio n program needs a large number of files in the folder created for a particular d ataset. The following procedure will create these files automatically. empty. Using Intkey program window, one can identify an unknown plant from this group o f taxa by reading the first character in the unknown plant, clicking the appropr iate character in the left upper pane and clicking or entering the right choice of the character state when prompted. This will eliminate and show certain taxa in the lower right pane and the used character in the lower left pane (Figure 5. 19). As you use more and more characters, some more taxa will be rejected and th e process will end when a single identified taxon is remains in the upper right pane. You can click i (information) icon to view image (if added) or read full description of taxon.

Open Dataset in the Delta editor and click File-->Export directive. Delta Files to export dialogue box appears. Click OK, subse-quently Done and then Close (if necessary from X on top). Open Delta editor (if closed). Click view-->Action set s. Print character list appears. See that the Confor Tab is active. Select Print character list-RTF . Click Run. In the next dialogue Box click Yes. Go to Ac-tion sets again, select Translate into Natu-ral Language-RTF-Single file for all taxa nd click Run, subsequently click Yes. Go to Intkey can also be used to access Delta data and images over the internet. For t his data files (such as iitems, ichars), intkey.ini, contents.ind (together with rtf files), and image files (optional) are put in a zip file (or self extractin g zip file) and uploaded to the website along with startup file (*.ink; which co ntains the information and the path of uploaded files of the project), intkey.in i, imagePath (optional) and InfoPath(optional). A data-set index file or link in WWW page must point to the special startup file (*.ink; not intkey.ini or intke y.ink). The startup file tells Intkey where the data set and its associated imag es are found on the website. 124 Plant Systematics When a person using an internet browser clicks on a link to an intkey startup fi le, the browser activates Intkey and passes it a copy of startup file. Intkey it self then retrieves the actual data set from the web, extracts its contents, and begins identification. For this web applicability, however, Intkey has to be in stalled on both Web server and each each client PC, and an association of files has to be developed by the manager of Web up the identification window with four panels, like Intkey. Character panel is u pper left window, but right upper panel shows character states, lower right pane l the the matching items panel showing match-ing or remaining taxa (click any ta xon to get its full description) and the lower left panel the query criteria pan el: display of previous (used) character state selections. NaviKey also allows c heckbox matching options to: a) Restrict view on used characters and character s tates of remaining items. b) Re-tain items unrecorded for the selected char-acte rs. c) Retain items matching at least one selected state of resp. characters. d) Use extreme interval validation, and e) Use overlapping interval validation.Nav iKey does not display the list of excluded taxa but the total number of taxa and number remainserver, where the project files are located. Intkey based web applications are avail-able for several families and Genera fro m Flora of China, Families of the World (Watson and Dallwitz), Grass Genera of t he World (Watson and Dallwitz), Grass Species of the World (RBG, Kew), Tree and Shrub Genera of Borneo (J.K. Jarvie & Ermayanti), indentifica-tion facility for the vascular flora of Western Australia, and is available in FloraBase. Ad-ditio nally, interactive keys (using Intkey) to the families and genera of flowering p lants in Western Australia are soon to be added to FloraBase, with specialist ke ys for cer-tain significant genera also well advanced. ing are displayed. The use of software as web application is very convenient. Just fill in the titl e and subtitle of the project being developed in NaviKeyAppletWebpageTemplate.ht ml using html editor (say Frontpage), upload the whole folder to your website, a nd provide a link to NaviKey.html page. As this page opens, the NaviKey Navikey is a simple Java based interactive java application gets loaded and the program identification key, a free program, which is ready for interactive identificatio n. works on Delta flat files (chars, items and specs- present in your folder if you have de-veloped a database ready for identification through Intkey, as detailed in preceding paragraphs). NaviKey v. 4 is developed in the frame of BIOTa Afric a project (An International Research Network on biodiversity, sustainable use an d conservation) by Dieter Neubacher and Gerhard Rambold (University of Bayreuth, Germany), based on an earlier version (NaviKey v. 2.3 by Michael Bartley and No

el Cross, Harvard University Her-barium, Boston, USA). The program can be downlo aded from www.NaviKey.net and can be used both as stand alone application or as web application. After downloading the and unzipping the the file, the folder wi ll have a number of files on your computer. Simply add the three flat files of y our project to this folder. For using it as stand alone applica-tion simply clic k NaviKey.jar, and it will open NaviKey identifications are available for several families and genera of Flora o f China and genus Arisaema (Guy Gusman & Eric Gouda) and Flowering Plant Familie s of Jamaica (Gerald Guala & Jimi Sadle) Lucid Systems Lucid software (Lucid3) is a commercial powerful and widely acclaimed Lucid Prof essional identification and diagnostic software developed by Centre for Biologic al Information Technology, The University of Queensland, Brisbane Australia. The Lucid3 system comprises a Builder and Player for creating and deploying effecti ve and power-ful identification and diagnostic keys. It allows creation of inter active, random-access keys that can be deployed over the World Wide Web or CD. The key when used for the identification of an unknown specimen progressively Process of Identification 125 eliminates entities that do not match the chosen features until only one or a fe w pos-sible entities remain. Further information and images can be accessed to c onfirm the XID System allows the user to randomly select characteristics that are consisten t with their specimen and skill level. If the user cannot decide upon a characte ristic, they may query the program, which will provide a list of suggestions in order of ease of use, effectiveness, and items remaining.In general, much more d ata is included on each item/ species than is necessary to identify it. With thi s abundance of data, the user can identify any of the items/species using the ch aracteristics most obvious and easy to describe. With each characteristic entere d by the user, the program eliminates all species that do identification. The basic elements of a Lucid3 key are: a list of entities; a list of features a nd states that may be used to describe those entities; a matrix of score data fo r the features asso-ciated with each of the entities for the fea-tures; and vari ous attachments ( images, web pages etc) for the entities and features, to provide extra information to users. The Lucid3 Builder provides all the tools necessary to create the entity and fea ture lists, encode the score data, and attach in-formation files to items. The p ackage in ad-dition includes Lucid Phoenix, a computer based dichotomous or path way key Builder and Player that enables traditional paper based identification k eys to be published on the Internet or CD. Phoenix keys are inter-active, can be enhanced with multimedia, and delivered across the Internet seamlessly. Additio nal Fact Sheet Fusion software is a tool to facilitate the rapid generation of s tandardised fact sheets in HTML (Hyper Text Markup Language) or XML (eXtensible Markup Language). not have the combination of features entered. XID also offers 1000 Weeds of North America CD ROM. This is the most comprehensi ve weed identification reference ever published in North America. Contains 140 g rass-like and 860 broadleaf weeds, features include Interactive key, color photo s of all species, illustrated glossary of terms, page number references to over 40 weed refer-ence books, searchable geographic data, and State level distributi on maps. RID (Expert Identification ActKey ActKey is a web based interactive identification program developed by Hong Song of the Missouri Botanical Garden. This Java- based program uses MySQL as the dat abase server, and can handle data sets in DELTA, Systems) XID Services Inc. produces commercial soft-ware with emphasis on biological scie nces, and is one of the leading providers of expert identification systems for m

ajor universi-ties and botanical gardens in United States. XID offers two identi fication packages: Pankey, a DOS bassed identification pro-gram, and XID Authori ng Systems, Windows based databases and Program for IdentificaMS Excel, MS Access and Lucid formats. tion. The XID Authoring System allows au-thors to create their own smart key or ra n-dom access expert system for the identifi-cation of plants, animals, or any ot her ob-ject. The elegant simplicity of the XID Sys-tem makes it extremely user f riendly, and is as useful for school teacher as for the pro-fessional scientist. ActKey identification is available for the floras of China, North America, Madag ascar, Borneo, at the Harvard University Herbaria Editorial Center, and hosted a t eFlora website. Examples include several keys to the large and medium-sized ge nera of China (also in Chinese); the genera of Brassi-caceae of the world by Ihs an Al-Shehbaz; Salix (Salicaceae) of North America by George W. Argus (also in C hinese); angiosperm families by B. Hansen and K. Rahn (also in Chinese and Spani sh); Trilliaceae (Trillium and Paris) of the world by Susan B. Farmer, the gener ic tree flora of Madagascar by George Schatz, and the trees & shrubs of Borneo b y James K. Jarvie & Ermayanti, respectively. 126 Plant Systematics Meka biodiversity documentation and species identification. Linnaeus II supports the creation of taxonomic databases, optimizes the construction of easy-to-use ident ification keys, expedites the display and comparison of distribution patterns, a nd promotes the use of taxonomic data for biodiversity studies. There are three 'modules' of Linnaeus II: the 'Builder' to manage your data and to create an inf ormation system, the 'Runtime' engine to publish completed information systems o n CD-ROM/DVD-ROM, and the 'Web Publisher' to publish your comMEKA (pronounced mecca ) is an interactive Multiple-Entry Key Algorithm to enable r apid identification of biological specimens, now designed to run under Windows. The program, distrubuted free, is developed by Christopher Meacham, Jepson Herba rium, Berkeley, CA. The user picks character states that are present in the spec imen from a list of possibilities. As the character states are scored by picking them, MEKA eliminates taxa that no longer match the list of scored character st ates. Different windows display different aspects of the underlying data base. A s the identification progresses the windows are updated automatically. An index screen makes it easy to find and score particular classes of character states. M EKA does not lead the user in a fixed stepwise progression through a series of q uestions. Instead, the user can perform identifications by scoring character sta tes in any order. This makes it possible to identify specimens that are much mor e fragmentary than is possible with dichotomous keys. New Windows version in-clu des a conversion function that can con-vert any MEKA key to the SLIKS (Stinger s L ight Weight Interactive Key Software) for-mat developed by Gerald Guala for Webbased identification. Thomas J. Rosatti has devel-oped many Meka keys to Califor nia plants, and Prof. Knud Ib Christensen of the Botanic Garden of the Universit y of Copenhagen key pleted project as a Web site. The package offers three identification modules: Text KeyTM- an electronic versi on of written dichotomous keys, The Pic-ture KeyTM - similar to the Text Key but picture- based, and IdentifyItTM, the most powerful identification tool. It is a multi-ple-entry key based on a matrix of taxa, characters, and character state s. Unlike the Species and Higher Taxa, which hold text descriptions of the taxa, in IdentifyIt taxa are described in a more structured format: as a series of ch aracter states. This allows you to easily obtain answers to specific questions l ike, "Which species are red and/or white" to Old World Crataegus. In addition to these interactive Keys Illustrations of plants from various parts of the World as also the illustrations of economic plants are put up at various websites hosted by different institutions, particularly one supporting Virtual her-

SLIKS software is a small free Javascript program developed to facilitate the use of baria and eFloras. These illustration s are interactive keys. SLIKS is written in simavailable for help in identification. ple Javascript and runs over the web or lo-cally on your machine. Users can down load their own copy or use it from your web site. It runs through the web browse r so it is es-sentially platform independant. A number of electronic lists are maintained by listservers. Taxacom is one such list very active on taxonomic matters, subscribed to by numerous active taxonomi sts all over the world. There is a regular exchange on matters of taxonomic inte rest. Any member with a problem can seek opinions from all members simultaneousl y. An unknown plant can be identified by sending its description to the list. St ill better, a photograph or illustration of the unknown plant can be put up on a website with information to the members. The memIndentifyIt IdentifyIt is identification software of com-prehensive commercial Linnaeus II m ultifunctional research tool developed by ETI BioInformatics, for systematists a nd biodiversity researchers. It facilitates Process of Identification 127 bers may go to the website, observe the photograph or illustration and send thei r comments to the member concerned or the list itself. Many users are being bene fitted DNA Barcoding through this web based interaction. DNA Barcoding is the most recent approach to fix the identity of different speci es, to ultimately facilitate a common database for living organisms. Consortium for the Barcode of Life (CBOL) is an international collaboration of natural hist ory museums, herbaria, biological repositories, and biodiversity inventory sites , together with academic and commercial experts in genomics, taxonomy, electroni cs, and computer science. The mission of CBOL is to rapidly accelerate compiling of DNA barcodes of known and newly discovered plant and animal species, establi sh a public library of sequences linked to named specimens, and promote developm ent of Last few years have seen the spurt of internet based exchange of information. In diantreepix (http://groups.google.co.in/ group/indiantreepix) is Google e-group devoted to creating awareness, helping in iden-tification, discussion and document ation of Indian Flora. It is one of busiest group consisting of experts from various fields. The group also aims at compiling a datab ase of photographs, nomenclature, relevant infor-mation and local names in diffe rent Lan-guages. The database is building up at rapid pace, and new taxa are add ed to the database after confirmation from several experts. FlowersOfIndia, anot her website (http:// portable devices for DNA barcoding. www.flowersofindia.net/index.html) devoted DNA barcoding is a technique for characterizing species of organisms using a sho rt DNA sequence from a standard and agreed-upon position in the genome. DNA barc ode sequences are very short relative to the entire genome and they can be obtai ned reasonably quickly and cheaply. The cytochrome c oxidase subunit 1 mitochond rial region (COI) is emerging as the standard barcode region for higher animals. Because of its slow rate of evolution in higher plants, however, is not suitabl e for barcoding, and after experimenting with chloroplast plastid trnH-psbA inte rgeneric spacer gene, botanists at the Proceedings of National Academy of Scienc es, Cameroon and Plant Working Group of CBOL in 2009, have decided to use two ge nes rbcL and matK for DNA barcoding of plants. Once the to Indian flowering plants has separate data-bases with links to photographs arr anged according to botanical names (alphabetic or sorted family wise) and common names. Plants of different categories such as Flowering trees, Orchids, Medicinal plants, Garden Flowers, Bulbous plants, Himalayan Flowers can be accessed through separate links. New images are being continuousl

y added after confirmation by experts. Vascular plant im-age library (http//bota ny.csdl. tamu.edu/ FLORA/gallery.htm) was developed originally with support from Texas Higher Education co-ordinating Board as a part of Digita l Flora of Texas. Links are provided family wise to the images of plant species in databases including Flowers of India, CalPhotos, Flora of Chile, Missouri plants, Floral images, Plants of Habarcodes of all species of plants are waii, Oregon Flora image project, and several established, identification of pla nts may be individual image collections. CalPhotos is a possible through a handh eld scanner. It may huge databas developed under a project of be useful for dete cting illegal plants at check BSCIT of University of California, Berkeley, point s, and also make the process of and contains more than 215,750 images of identif ication much simpler. However, at plants, animals, fossils, peoples and land- th is point, detection of closely related species scapes around the world. Nearly 1 18,000 im- may not be possible, and traditional methods ages of plants can be browsed alphabetically may be used before more refined met hods of and also searched through easy criteria. DNA barcoding are developed. Chapter 6 Varia tion, Biosystematics, Population Genetics and Evolution It is now a universally agreed upon fact that different species are not fixed en tities but systems of populations which exhibit varia-tion and wherein no two in dividuals are iden-tical. This concept of variations was first pro-posed by Lama rck and further developed by Darwin, culminating in his famous book Ori-gin of S pecies (1859). Systematics is a unique natural science concerned with the study of individual, population and taxon relation-ships for purposes of classificatio n. The study of plant systematics is based on the premise that in the tremendous variation in the plant world, there exist conceptual discrete units (usually na med as species) that can be recognized, classified, described, and named, on the further premise that logical relationships developed through evolution dynamic structure. Clausen et al. (1945) regard genetics, cytology, comparative morphology and ecology as furnishing the critical data which together, when appl ied to the study of organic evolution, make up biosystematics. These two differe nt approaches aim at the same problem, the study of variations. exist among these units. The study of biosystematics, mainly the experimental systematics and population genetics approach the common aim, al-though the methodology is different. The ex -perimental systematist usually begins with classical interpretation of species and works backwards so as to understand the genetic mechanisms involved. The pop ulation geneticist, on the other hand, begins with raw population, discarding an y classical concept in mind. He works into a series of group con-cepts which may or may not be comparable to the taxonomists concept of species. The studies on variations, experimental studies and hybridization studies in lig ht of genetic information are commonly covered under the term biosystematics. Th e term was first proposed by Camp & Gilly (1943 as Biosystematy) to delimit natu ral biotic units and to apply to these units a system of nomenclature adequate t o the task of con-veying precise information regarding their defined limits, rel ationships, variability and TYPES OF VARIATION The recognition of taxonomic units is based on the identification of the occurre nce and the degree of discontinuity in variation in the populations. The variati on may be con-tinuous when the individuals of a popula-

Variation, Biosystematics, Population Genetics and Evolution 129 tion are separable by infinitely small differ-ences in any of the attributes. In a discon-tinuous variation, however, there is a dis-tinct gap between two popul ations, each showing its own continuous variation for a particular attribute. Th e discontinuity be-tween the populations primarily results from isolation in nat ure. Isolation plays a major role in establishing and widening the gap between t he populations, allowing evolution to take its destined course with no distur-ba nce. Variation in plants includes three fundamental types: developmental, enviro nronmental conditions. Such populations are named ecophenes. In Epilobium; the su n-plants have small, thick leaves, many hairs and a short stature, whereas the s hade-plants have larger thinner leaves with fewer hairs and a taller stature. Genetic variation mental and genetic. Genetic variation may result from mutation or recombination. Mutation is the occ urrence of heritable change in the genotype of an organism that was not inherite d from its ancestors. It is the ultimate source of variation in a species and re plenishes the supply of genetic variability. A mutation may be as minute as the substitution of a single nucleotide pair in the DNA (point mutation), change in a sequence nucleotides controlling gene action (Gene mutation) or as great Developmental variation A distinct change in attributes is often found during different stages of develo pment. Juvenile leaves of Eucalyptus, Salix and Populus are often different from the mature leaves, as a major change in the chromosome and may often cause much confusion, but structure (chromosomal mutation). may pr ove equally useful when both types of Chromosomal mutation may be due to leaves are available from a plant. The first deletion, inversion, aneuploidy or polyplo idy. leaves of Phaseolus are opposite and simple, Recombination is a reassortmen t of chromothe later ones alternate and pinnately com- somes, bringing together via meiosis and pound. As the seedling stage is most critical fertilization the genetic material from in a plant s life, the characters present dur- different parents and producing a n ew ing this period surely have survival value. Takhtajan proposed a neotenous origi n for angiosperms on the assumption of juvenile simple leaves of seed ferns havi ng persisted in the adult forms, which were the direct progenitors of angiosperm s. genotype. VARIANCE ANALYSIS Environmental variation Since no two individuals in a population are similar, there is need for some obj ective analysis for useful comparison. It is, however, often impossible to colle ct information about all the individuals of a population, and as such is reasona ble to analyse a representative sample. It is essential that this sample should represent random subset of Environmental factors often play major role in shaping the appearance of a plant . Heterophylly is the common manifestation the population. The simplest tool is to of environmental variation. The submerged leaves of Ranunculus aquatilis are finely calculate the mean or average by addin g the dissected, whereas the emergent leaves of series of values and dividing the tota l by the number of values. The formula for calculatthe same plant are broadly lobed. The first submerged leaves of Sium suave are pin-nately dissected and flaccid; the older e merged leaves are pinnately compound and stiff. The individuals of a species oft en exhibit phenotypic plasticity, expressing different phenotypes under differen t envi-

ing the mean is: X = `, n where X represents the mean, E summation of all values of X, Xi represents the i ndividual 130 Plant Systematics Table 6.1 Mean, variance and standard deviaremoved from mean, the variance would be higher. The variance may either be calc ulated for a population, or a sample from the population. The variance for a pop ulation maybe calculated as: tion of two samples based to plant height. The two samples have the same mean bu t different variance and standard deviation, highlighting the significance of th ese calculations. ~n(Xi X)2 Height (cm) (Xi - X) (Xi - X)2 = i=1 M n ~ Sample Sample sample Sample sample sample To obtain the variance, the difference be~ ~ ~ ~ ~ tween each value of the attribute (X) and the mean is squared and a sum of these squares is divided by the number of observation (n). For calculating sample var iance (s2) the sum of squares is divided by n 1 instead of n. The formula for sa mple variance may be written as: ~~ ~~ ~~ ~ ~~ = 3 22- 15 = 7 49 1a 10 1a-1s = -1 10 -1s = -s 1 2s 16 06 16- 15 = 1 06- 15 = -9 1 81 ~~ ~~ ~~ ~ ~~ ~ ~ ~~ ~ ~~ ~ ~ ~ ~ 17 12 17-15=2 12-15=-3 4 9

19 19 19-15=4 19-15=4 16 16 In(Xi X)2 12 11 12-15=-3 11-15=-4 9 16 i =1 s = 14 27 14-15=-1 27-15=12 1 144 n 1 For the calculation of sample variance, 13 1a 13 -1s = -2 1a - 1s = -1 a 1 the reason for dividing by n 1 and not by n, is related to the degrees of freedo m. If we have a single value we can t compare it, if we have two we have one compa rison (2 1), if we have three values we have two comparisons (3 1), and if there are n values, n 1 comparisons are possible. While calculating population varian ce, with large number of values , the difference of one would be irrelevant, and as such the sum is divided directly by n. Two samples may have the same mean, b ut different variance (Table 16 21 16- 15 = 1 21 - 15 = 6 1 36 ~~ ~~ ~~ ~ ~~ ~ ~~ ~~ ~ ~~ ~ ~~ ~ ~~ 12 13 12- 15 = -3 13 - 15 = -2 9 4 ^x,. = 180 180 Mean X = ~~ ~~ Variance Sz = ^ (Xi X)2 5.09 3a n - 1 Standard deviation s = ^ sZ 2.256 6.17 6.1). The square root of variance is reprevalues of an attribute under study and n represents the number of values. Thus, five plants of a species with height 15 cm, 12 cm, 10 cm, 22 cm and 16 cm would have a mean of 15 cm ((15+12+10+22+16)/5). The extent of variation within a popu lation of a species is best represented by determining the variance. It is a mea sure of the spread of individual observations around the mean, i.e. how variable the individuals and their measurements are. It is defined as the av-erage squar ed deviationf rom the mean. If various individuals were not far from this mean t he variance would be minimum. On the other hand, if many individuals were far sented by standard deviation. Latter is often preferred over the variance becaus e it shares the same units as the original measurements, whereas the variance is

in the units squared. The reason for first squaring the values and then determi ning the square root, is to obtain the real picture of variation. If simple diff erence of each value and mean is summed, the negative values (measurements lower than the mean) may get cancelled by positive values (measurements higher than t he mean), and the result would be zero, and thus meaningless. The squaring conve rts all values to plus and thus a real diversion from the mean on Variation, Biosystematics, Population Genetics and Evolution 131 either side is taken into account. We may thus determine the standard deviation of a population as: This is significant in economical plants where it is important to determine whet her the attributes are related to environmental variations or genotype. If forme r is true, it is advisable to improve cultural practices, if it is related to ge notype, selective breeding would be the answer. V 62 and that for a sample as: s = s2 For our sample data, the sample variance would be [(15-15)2 + (12-15)2 + (10-15) 2 + (22-15)2 + (16-15)2]/4 = 21 and the sample standard deviation REPRODUCTIVE SYSTEMS 21 = 4.5825 The diverse mechanisms of reproduction in seed plants can be classified under fo ur major categories. Allogamy involves cross-fertilization between closely relat ed individuals growing at a suitable distance from each other, and results in th e formation of hybrids. Cross-fertilization promotes het-erozygosity, resulting in considerable variation and diversity in individuals. Autogamy involves self-f ertilization, resulting in inbred offsprings. It promotes homozygosity, yield-in g uniform populations. Agamospermy in-volves production of seeds resulting from the development of embryos from maternal tis- sue without fertilization. Finally , the repro-duction may result from vegetative propagation of somatic regions su ch as shoot segments, bulbs, rhizomes, corms and other vegetative structures. Bo th allogamy and auto-gamy are examples of sexual reproduction, involving meiosis and fertilization. The last two, circumvent sexual reproduction and multiplicat ion occurs through asexual reproduction, and are often termed apomixis. The prod ucts of asexual reproduction are known as ramets, where as products of sexual re production, which show genetic variation as genets. The determination of sample size from a population is crucial for the calculatio n of variance and standard deviation. Sample size n can be computed from magnitu de of standard deviation (this can be estimated from the smallest and the larges t value of an attribute (say smallest 5, largest 45, mean 25, deviation 20), lev el of confidence desired (z, say 0.95%) and maximum width of units from true val ue (d, say 5): z 2 0.2 n = d2 For the above parameters the adequate sample size is (0.95 x 0.95) = 0.902 x (20 x 20) = 360.8/(5 x 5) = 14.4. Thus 15 would be ideal sample size with these parameters. The analysis of variance data is often com-plicated, especially where more than one factors are responsible for variation. The technique of Analysis of variance (ANOVA) developed by Sir Ronald Fisher (1930) is com-monly used for dividing th e variance into components. It is a powerful statistical procedure for determini ng whether the differ-ences from the mean are significant, i.e. larger than expe cted by chance. Thus for example, if probability value of less than 0.002 (There is less than one per cent chance than variation obtained is due to chance) is o

btained through variance analy-sis, the results are due to factors others than c hance. The analysis involves partitioning the variance and comparing the role of vari-ous factors (environmental, genetical, etc.). Outbreeding Outbreeding, as mentioned earlier is largely responsible for genetic and pheneti c diver-sity in populations. It is also known as outcrossing, allogamy or xenogamy. It enables plants to adapt to wide range of envi ronmental conditions, and increases likelihood of survival and evolutionary change. A variety of mechanisms promote outbreeding. These are briefly described below: 132 Plant Systematics 1. Dioecy: The phenomenon involves the lization between the gametes derived from the same flower. Gametophytic self-inc ompatibility results from genetic composition of male gametophyte, and sporophyt ic self-incompatibility by genetic composition of sporophytic tissue such as sty le and stigma. occurrence of unisexual flowers, with male and female flowers in different indiv iduals. Some variations of this are also encountered as for example, some indivi duals having male flowers others bisexual flowers (androdioecy), some individual s having female flow-ers others bisexual flowers (gynodioecy), or some individua ls having male flowers others female and Hybridization still others bisexual flowers (trioecy). Although occurrence of breeding barriers is dominant criterion for distinction b etween the species, several cases of interspecific hybridization have been repor ted. Based on the studies of the Flora British Isles, Stace (1989) concluded tha t there are approximately 70, 000 different naturally occurring interspecific hy brids, accounting for more than one fourth of the total number of species of see d plants on this planet. Natural hybridization is common in Salix, Helianthus, Q uercus, Senecio and Tragopogon. It is more common in perennials as 2. Dichogamy: The situation reflects the maturation of male and female floral parts at different times. In some members o f Apiaceae and Asteraceae, pollen grains mature and are released before gynoeciu m is mature and re-ceptive, the phenomenon known as protandry. In others, like m embers of Chenopodiaceae, the gynoecium is mature and receptive before the polle n maturation and release, a feature known as protogyny. 3. Herkogamy: It results from physical compared to annuals. separation and stamens and carpels. This could be achieved by heterostyly, diffe rences in the length of stamens and carpels. In the phenomenon known as distyly some flowers have short style and longer stamens (thrum flowers), whereas others have long style and short stamens (Pin flowers). In a rarer situation known as trisyly, three types of stamen and carpel lengths occur. In other cases the styl e is curved away from stamens, either towards right (right-handed flowers) or le ft (left-handed flowers), the situation known as enanciostyly. In some genera li ke Mimulus, the stigmas close after being touched by a pollinator, thus preventi ng pollination from same flower (movement herkogamy). In others like Kalmia, the pollinator triggers the movement of one or more sta-mens, dusting insect with p ollen (trigHybridization between different species usually results in sterile offsprings, d ue to failure of pairing at meiosis, but in several genera like Senecio and Trag opogon, inter-specific hybridization is often followed by chromosomal duplicatio n, the resulting poly-ploid (Allopolyploid; Amphiploid) genera-tion is sexually stable due to normal meiosis of paired genomes. Many such polyploid species with distinct characters have been reported in these genera. Occurrence of intergeneric hybrids is much rarer, and there may be less than 300 naturally intergeneric hybrids world-wide. Such hybrids are reported mostly in

Poaceae and Orchidaceae, although in the latter family there are many artificial ly synthesized intergeneric hybrids, often involving five different genera. ger herkogamy). Introgressive hybridization The process of introgressive hybridization, also known as introgression involves the gradual infiltration of one species into that of another, and commonly invo lves species 4. Self-incompatibility: The phenom-enon refers to the prevention of fertiVariation, Biosystematics, Population Genetics and Evolution 133 with some degree of reproductive isolation. The phenomenon involves three steps: the formation of F1 hybrids, their backcrossing with one or another parental sp ecies, and natural selection of certain favourable re-combinant types. The hybri ds generally pro-duce a lot of variability through backcross-ing and F2 segregat ion, and may produce hy-brid swarms., occupying a variety of habi-tats. Backcros sing with parental species fre-quently results in reversion of hybrid off-spring s towards parental types. Backcross-ing may also result in movement of genes fro m one species to another via the hybrids and backcrosses. Introgression may lead to three diverse consequences. In some cases, as Gilia capitata, it may lead to merging of species. The species has eight geographical races, of which three ar e believed to have been distinct in Pliocene. Subsequent gene flow led to interg rading races, and as such they are included under single species. In-trogression may also transfer genetic ma-terial from one species to another without merging them. Introgressants, which get stabilized, may lead to the formation of new sp ecies. share intermediate area largely occupied by products of hybridization. Juniperus viryiniana, a mesophytic tree of eastern North America has shown introgression of bushy xerophyte J. ashei from dolomitic outcrops in Texas and Okhlahoma. Thro ughout the intermediate area are seen plants with partial recombination of chara cters between the two well differentiated species. It must be mentioned that the process of introgression may lead to the developme nt of variants with no taxonomic status, their recognition as ecotypes, subspeci es, or if the intermediate species is sufficiently distinct, recognition as dist inct species. Recognition of Hybrids The identification of hybrid nature of an off-spring is possible through the use of some important criteria. Two types of introgression are commonly recognized. Sympatric introgression commonly occurs between species occurring in the same general geographical area, bu t occupying different habitats. In California the introduced Helianthus annuus h as introgressed with native serpentine species H. bolanderi, and the vigorous we edy vari-ant of latter has spread into irrigated areas. Such introgression usual ly results in the wider spread of one species as compared to another. In England , for example Silene alba is spreading in weedy areas, whereas S. dioica, a wood land species is contracting. In Scotland, on the other hand, the more hu-mid cli mate allows S. dioica to flourish out-side woodlands, on hedgebanks and cliff Phenetic intermediacy: Hybrids tend to have phenetic intermediacy between the pu tative parents. It is easier to recognize morphological characters, which can be plot-ted on a scatter diagram (Figure 6.1). Hybrids can also detected by calcul ating hybrid in-dex. A list of characters by which the two species differ is pre pared. Each characterstate of one species is assigned zero score, whereas each c ontrasting character-state of another species given a score of 2. The hybrid ind ex of each species is calculated by summing up the score. Thus one species will have hybrid index of 0, and another species 2n (n refers to the number of charac ters by which two species differ). ledges. Reduced fertility: Hybrids between different species commonly tend to have reduc ed fertility, some being totally sterile. The degree of sterility is reflected u pon the degree of heterozygosity between genomes of parental species. A hybrid w hich perishes at zygote stage would represent maximum heterozygosity, whereas a

hybrid which manages to produce viable seeds, although less vigorous than either parents, would depict least heterozygosity between parents. Allopatric introgression occurs between species which are now fully allopatric, but had contact in the past. Such species are centered in different areas but 134 Plant Systematics ()mLeafmreadthb 11 9 810 5 4 Stabilization of hybrids Hybrids generally tend to obscure distinction between parental species, due to i nterme- diacy, segregation and consequent charac- ter combinations. The hybrids, however, often establish themselves as distinct taxa through a number of method s. Commonest of these is bypass sexual processes and perpetuate by asexually mea ns such as veg-etative propagation and agamospermy. The hybrid may similarly bec ome estab-lished sexually by hybridization followed by duplication of chromosome s (Amphiploidy), a phenomenon common in several genera, such as Senecio and Trag opogon. The hybrids may also establish through translocation heterozygosity, whe rein multivalent rings of chromosomes are formed at meiosis, as seen in Oenother a. In other cases unbalanced polyploidy, wherein female parents contributes grea ter number of chromosome sets, as compared to male parent. In genus Rosa, for ex ample hybrids are often established because female parent contributes four sets of chromosomes, and male parent A

~ 2 3 000 B 1 1 2 3 4 5 6 7 8 9 10 11 12 Leaf length (cm) Figure 6.1 Scatter diagram of populations of presumed parental species (A and B) and hybrid population. Latter shows structur al intermediacy. More characters can be added and depicted by appropriate symbol s. F2 segregation: Although F1 hybrids may tend to be normal, the next (F2) generat ion might show a lot of variability, exhibiting the segregation of parental char acters. There may thus be reappearance of parental forms, as also many new recom binations of parenonly one set. tal characters. In some genera such as Quercus, there is frequent hybridization between closely related species, resulting in the production hybrid swarms. Such sets of hybridi zing species constitute a syngameon, or semispecies. Distributional area: The hybrids between two species can also be verified by stu dying their distribution. In case of paren-tal species occupying the same area,

the hybrid populations would commonly be lo-cated in the same area. In case of s pecies occupying different but adjacent areas, hybrids would commonly be located in the con-tact area, or transitional area between the parental species. Outbreeders with internal barriers Artificial synthesis: Hybrids can often be created artificially through breeding tech-niques. The comparison of these artificial hybrids with suspected natural hybrids can help in confirming their identity. Several genera are reported to include species or species complexes, which inclu de races which are not morphologically very distinct, but are unable to interbre ed, owing to differences in chromosome number or structure (structural hybridity ), in others the chromosomal differences are not clear (cryptic structural hybri dity). Such inter-sterile races are often known as semi-cryptic species. Interst erile populations with no apparent morphological distinction are known as crypti c or sibling species. Variation, Biosystematics, Population Genetics and Evolution 135 Inbreeding seed formation occurs without sexual union. Vegetative apomixis is common in pla nts where sexual reproduction is not pos- sible. It is encountered in dioecious species of Elodea with flowers of only one sex, some species of Poa where flower s are replaced by bulbils, and sexually sterile species of Po-tentilla, Mentha a nd Circaea, where genetic reasons don t permit normal sexual repro-duction. Agamos permy may be manifested in a variety of ways. Embryo may be formed directly from the sporophytic tissue such as nucellus (adventive embryony), or from diploid g ametophyte where meiosis is bypassed (gametophytic apomixis), either archesporia l cells (diplospory) or somatic cells (apospory) developing directly into dip-lo id gametophyte. Embryo may develop from unreduced egg (parthenogenesis) or from a somatic cell (apogamy). Gametophytic apomixis occurs in several families such as Rosaceae and Asteraceae. Although male par-ent does not contribute towards em bryonic tissue, nevertheless pollination is necessary because one of the male nu cleus has to fuse with female nucleus to produce en-dosperm, the phenomenon know n as pseudogamy. Apogamy occurs is prevalent Also known as selfing, involves union of gametes from the same plant. It may eit her occur within the same flower (autogamy), or between different flowers of the same plant (geitonogamy). Although it ensures reproduction, even when there are fewer individuals, or pollinators are not available. It, however, reduced varia tion in populations and may even result in accumulation of del-eterious alleles, a phenomenon known as inbreeding depression. Inbreeding species tend to exist a s relatively uniform popula-tions, often differing considerably from one another , because of absence of gene flow between them. This commonly results in the pro duction of pure lines. In several genera of flowering plants, distinct inbreedin g popu-lations have been recognized as distinct spe-cies. Although they are most ly interfertile, but very low level of outbreeding, and very high level of inbre eding ensures that the taxa remain distinct. Those taxa with very minor differen ces, but reproductively iso-lated, are known as microspecies, also called as Jor danons, as they were first rec-ognized by Jordon (1873). Being the result of inb reeding, such microspecies are uniparen-tal in origin. Many microspecies are rec ognized with Andropogon virginicus species com-plex, where many inbreed due to c leisto-gamy. Several authors, however, avoid rec- ognizing them as distinct spe cies, because in ferns. of their great numbers. The populations produced by agamospermy may often show smaller differences, beca use any genetic mutation is preserved in population, and as these are stable thr ough generations, they are usually rec- ognized as distinct taxa, often as apomi ctic microspecies (agamospecies). Such microspecies are more stable than those p roduced through inbreeding (Jardanons). Agamospecies with better dispersal mecha nisms, as in Taraxacum and Hieracium, are widely spread, whereas those with poor ly de-veloped dispersal mechanisms as in RanunOutbreeding and inbreeding are, however, not isolated mechanisms. Some plants sh ow both, a phenomenon known as allautogamy. Species of Viola and Clarkia, for ex

ample, have two types of flowers: normally open (chasmogamous) flowers which exp erience cross-pollination. Others remain closed (cleistogamous), and are self-po llinated. culus are narrowly distributed. Apomixis The phenomenon of apomixis in a broader sense includes non-sexual reproduction, either through vegetative propagation (veg-etative apomixis) or agamospermy, whe re POPULATION GENETICS Population genetics deals with the application of genetic principles to populati ons of a particular species. A population constitutes 136 Plant Systematics a group of individuals growing in a specific area and freely interbreeding. A gr oup of interbreeding individuals who share a common set of genes constitute a Me nde-lian population. The widely distributed species often have separate populati ons in different geographical regions, known as subpopulations. Each subpopulati on grow-ing in a particular geographical area constitutes a local population. Th e entire set of genetic information covering all alleles in a population, forms its gene pool. The evolutionary process is best understood by studying the gene pool of Mendelian population and not the individual members. Mating systems Allele frequenc ies While analysing genetic data, it is more logi-cal to talk in terms of genotype f requency and allele frequency, instead of absolute numbers. Thus in a population with alleles A and a, and 100 aa individuals, 300 heterozygous Aa individuals a nd 600 homozygous AA individuals, the genotypic frequencies are calculated as: Three types of mating patterns are recognized, which determine the genotype freq uencies of populations. In random mating, the two genotypes mate in proportion t o their relative frequencies in the population. A population may undergo random mating with respect to some traits, but nonrandom with respect to others at the same time. Assortative mating is a type of nonrandom mating in which the mates a re formed on the basis of their degree of similarity in phenotype. In human popu lation positive assortative mating is more common, as mating partners are more s imilar in phenotype, for example skin colour. In several species of plants negat ive assortative mating is prevalent, the mating partners differing in phenotype. In Primula officinalis, for example pin type flowers (long style and short stam ens) produce pollen lower down in flower but receive pollen higher up, where as the thrum type flowers (long stamens short style) receive pollen lower down, and produce higher up. Consequently the insect pollinators that work deeper into fl ower collect pollen from pin types flowers and deposit on stigma of thrum type f lowers. Pollinators AA = 600/1000 = 0.6 Aa = 300/1000 = 0.3 aa 100/1000 = 0.1 working higher up (due to shorter mouth It must be noted AA genotype has 600+600 = parts) do the reverse, collecting pol len from 1200 A alleles, Aa genotype 300 A and 300 a thrum type flowers and depo siting on pin type alleles, and aa genotype 100+100 = 200 a flowers. The third type of mating involves inbreedalleles. This totals 1500 A alleles and 500 a alleles. Allele frequency as such would be calculated as: ing, mating between relatives, and in bisexual flowers, generally between gamete s of the same flower. A 1500/2000 0.75 a

500/2000 0.25 Hardy-Weinberg Law Please note, lower the allele frequency, rarer it is. Once the value touches 0, the allele is lost, and the other allele with value of 1 Most species of animals and plants, except inbreeding plants, random mating is p revalent form of reproduction. Thus each type of mating pair is formed as often as would be expected by chance encounter between the genotypes. In a randomly ma ting population with genotype consisting of alleles A and a, with allele frequen cy of p and q, respectively (note p + q = 1), genotypes formed from would get fixed. It is more appropriate to analyze popula-tions in terms of alleles, and not geno types, because genotypes are disrupted during the process of segregation in subs equent generations. Variation, Biosystematics, Population Genetics and Evolution 137 Table 6.2 Chi-square table showing the relationship between Chi-square vales, de grees of freegeseoefrD deeorm f dom and the probability. For a particular degree of freedom the nearest Chi-squa re value is located from the row, the the appropriate probability value read fro m the top row. Probability values lower than 0.05 are highly significant and do not support the hypothesis being tested. The values higher than 0.05 support the hypothesis.

Probability 0.95 0.90 0.70 0.50 0.30 0.20 0.10 Chi-square values 0.004 0.016 0.15 0.46 1.07 1.64 2.71 0.10 0.21 0.71 1.39 2.41 3.22 4.61 0.35 0.58 1.42 2.37 3.67 4.64 6.25 0.71 1.06

2.20 3.36 4.88 5.99 7.78 1.15 1.61 3.00 4.35 6.06 7.29 9.24 1.64 2.20 3.83 5.35 7.23 8.56 10.65 2.17 2.83 4.67 6.35 8.38 9.80 12.02 2.73 3.49 5.53 7.34 9.52 11.03 13.36 3.33 4.17 6.39 8.34 10.66 12.24 14.68 3.94 4.87 7.27 9.34 11.78 13.44 15.99 7.26 8.55. 11.72 14.34 17.32 19.31 22.31 10.85 12.44 16.27 19.34 22.78 25.04

28.41 18.49 20.60 25.51 29.34 33.53 36.25 40.26 34.76 37.69 44.31 49.34 54.72 58.16 63.17 0.05 3.84 5.99 7.82 9.49 11.07 12.59 14.07 15.51 16.92 18.31 25.00 31.41 43.77 67.51 0.01 6.64 9.21 11.35 13.28 15.09 16.81 18.48 20.09 21.67 23.21 30.58 37.57 50.89 76.15 0.001 10.83 13.82 16.27 18.47 20.52 22.46 24.32 26.13 27.88 29.59 37.70 45.32 59.70 86.66

1 2 3 4 5 6 7 8 9 10 15 20 30 50 fusion of A and a gametes of either parent would be AA, Aa, Aa and aa. In terms of allele frequencies the genotypes could be written as: (five assumptions) the frequencies of the alleles don s change over time. The law also concluded that as long as mating is random, the genotype frequencies remain in the proportion of p2, 2pq and q2. The sum of genotype frequencies equals 1, i.e. p2 + 2pq + q2 =1. The allelicf requencies remain con-stant from generation to generation, in AA: p2 AA: 2pq aa: q2 The foundations for these calculations-a landmark contribution in population genet-ics- were laid by Godfrey Hardy and Wi lhelm Weinberg, independently in 1908. The law is based on assumptions that in a infinitely large, randomly mating population, free from mutation, migration and natural selection such randomly mating populations. Although it is difficult for a population to be infinitely large in size, but a fairy large population satisfies the requirement. If the size of the population is limited, chance 138 Plant Systematics deviations from the expected rates can re-sult in changes in allelic frequency, a phe-nomenon known as genetic drift. It must also be remembered, however, that random mating does not always mean that popu-lations must be interbreeding rando mly for all the traits for the law to hold true. In human populations, for examp le, where mar-riages are chosen on the basis of religion, cast and colour, the m ating partners do not select blood groups or other such traits, which may thus s atisfy the Hardy Weinberg law. The law thus applies to any locus for which rando m mating occurs, even if matthe expected values. It helps us to conclude whether the departure is within the probability limits, or due to some other phenomenon operating on the population . Goodness of fit is conventionally measured in terms of chi-square . It is calc ulated as follows: c2 _ S (Observed value Expected value )2 Expected value ing is nonrandom for other loci (traits). Supposing phenotype X represents a population heterozygous for one pair of allel es, and Y homozygous recessive. Mating between the two is expected to produce pr ogeny with the two phenotypes in the ratio of 1:1 ( X = A + a, Y = a + a; phenot ypes Aa, Aa, aa, aa; X = 2, Y = 2). Supposing there is a progeny of 30 individua ls with 18 of X phenotype and 12 of Y phenotype, whereas the expected number for each phenotype is 15. For Hardy-Weinberg law to apply, the popu-lations must be free from mutation, mi gra-tion and natural selection. It is important to note that the condition of no

evolutionary change applies only to the trait (locus) in question. A population may be subject to evolutionary processes acting on some genes, while still meet ing the Hardy-Weinberg as-sumptions for other loci. The populations which satisf y the requirements of the law are said to be in genetic equilibrium or Hardy-Wei nberg equilibrium. If the ob-served genotype proportions are different from the expected, one or more assumptions of the law have been violated. (18 15)2 (12 15)2 9 9 c2 = + _ 12 15 15 15 15 Null hypothesis and ChiCloser the observed values are to the expected, lower the value of c2. If the tw o match perfectly the values would be zero, although it never happens in nature. Once the c2 value has been calculated, the good-ness of fit of this value to th e expected numbers is determined. Two parameters are essential for this interpre tation. First the number of degrees of freedom for a particular c2 test is calcu lated, as number of classes of data minus one. In our case there are two classes of data (phenotypes X and Y), hence one degree of freedom. The second parameter is the probability p, which can be deter-mined from graph of the Chi-square tes t. The graph shows range of c2 values along the X axis, probability along the Y axis and curves of different degrees of freedom running from base towards the ri ght top. The vertical line starting from the specific c2 value is located at the point it touches the relevant curve, and a horizontal line from this point towa rds the left touching the Y axis determines the probability values. The probabil ity can also be read from the statistical table for Square Test A population with random mating should rep-resent the progeny with numbers of va rious phenotypes closer to the expected numbers. For any population analysis it is important to determine, whether the observed values match (fit) the expected values or not. Null hypothesis states that there is no real dif-ference between the observed and the pre-dicted data. If the statistical analysis shows that the difference between predicted and observed values is due to chance, the null hyp othesis is proved, if not, it is rejected. Although we could never expect a perf ect match, the calculation of goodness of fit, gives us an indication of the dep arture from Variation, Biosystematics, Population Genetics and Evolution 139 Chi-square (Table 6.2). A value of p between 0.01 (1 per cent; worse fit; one in 100 stud-ies, this value would appear by chance) and 0.05 (5 percent; bad fit; one in 20 studies, this value would appear by chance) is considered significant. A value lower than 0.01 is con-sidered highly significant and the hypothesis is rejected outright. It is, however, safer to reject any value lower than 0.05. A value of p higher than 0.05 shows that the departure of observed values from ex pected values is not Female = XX = p2(XAXA); 2pq(XAXa); q2(XaXa) Male = XY = p(XAY); q(XaY) This is the obvious reason for the prevalence of X-linked recessive traits such as colour blindness and haemophilia among human males. Inbreeding and the Hardy-Weinberg Law

significant, and the hypothesis is supported. It is important to note that Chi-quare value is calculated on the basis of actua l numbers and not on the basis of percentages or ratios of various phenotypes. Inbreeding constitutes another important departure from random mating. It is oft en measured in terms of coefficient of inbreeding (F), which can be calculated as: Extension of Hardy-Weinberg F (Observed heterozygosity Expected heterozygosity) Law Expected heterozygosity Hardy-Weinberg Law may also be extended to situations such as multiple alleles a nd sex-linked alleles. Consider a population with three alleles A, B and C with allele frequencies of p, q and r respectively. The frequencies of various genoty pes would be represented as: Greater the value of F, the greater the reduction in heterozygosity relative to that expected from the Hardy-Weinberg expectation. If genotypes are in Hardy-Wei nberg pro-portions, F = 0, because observed heterozygosity equals expected one. In self-fertilization, common in plants, however, the decreases with every gener ation and homozygosity increases consequently. Supposing we start with a complet ely heterozygous popu(p + q + r)2 = p2 (AA) + 2pq(AB) + q2 (BB) + 2pr (AC) + 2qr (BC) + r2 (CC) lation Aa reproducing by self-fertilization. In a population with allele frequencies of After one generation the progeny will conp = 0.52, q = 0.31 and r = 0.17, the following sist of 1/4 AA, 1/2 Aa and 1/4 aa . In next genotypes are frequencies are expected if the population is in Hardy-Weinberg equilibAA progeny, aa will produce only aa progeny, generation homozygotes AA will produce only rium: whereas only heterozygotes will again segregate into half heterozygotes and half ho- mozygotes (1/4 AA, 1/4 aa). This will reduce heterozygotes to 1/4 of the to tal population. After large number of generations, there will be no heterozygote s, and the population will AA = p2 = (0.52)2 AB = 2pq = 2(0.52 x 0.31) BB = q2 = (0.31)2 AC = 2pr = 2(0.52 x 0.17) BC = 2qr = 2(0.31 x 0.17) CC = r2 = (0.17)2 0.27 0.32 = 0.10 = 0.18 = 0.11 be divided into half AA and half aa.

0.03 It should, however, be noted that although genotype frequencies change from one generation to another, the allele frequencies remain constant. The law can similarly be applied to sex-linked alleles. In human populations, fo r example males are XY and females XX. For X-linked alleles the female genotypes show normal Hardy-Weinberg distribution, whereas the male genotypes are distrib uted in the same frequencies as respective alleles EVOLUTION Evolution consists of progressive changes in the gene pool, associated with prog ressive

140 Plant Systematics adaptation of a population to its environment. The evolution is the end result o f four distinct processes, which result in changes in allelic frequencies, ultim ately resulting in cumulative changes in the genetic char-acteristics of populat ions. It is believed that these processes over a geological time lead to the evo lution of species. amino acid, when it is known as silent mutation. Mutations may be irreversible o r reversible. Change from A to a constitutes forward mutation, whereas from a to A as reverse mutation. In reversible mutations, forward mutations are offset by reverse mutations and as such an equilibrium in allele frequencies is reached. Mutation Migration It is now agreed that variation in heritable traits results from mutations. Muta tion is an important process in evolution. It was earlier believed that variatio ns result mostly from adaptive inheritable change induced by environment. This a daptation theory, based on Lamarckism, believed in the inheritance of acquired c haracteristics. The middle of twentieth Century saw the emer-gence of mutation t heory, to explain changes in several bacterial populations. Although some mutati ons occur only in the somatic cells, and not passed to the next generation (soma tic mutation), others occurring in germ cells are transmitted from one generatio n to another (genetic muta-tion; germ-line mutation). A mutation may involve cha nges within the same base type (purine to purine; pyrimidine to pyrimidine), whe n its is known as transition mutation; in others it may involve changes from a p urine to a pyrimidine or vice versa, when it is termed transversion mutation. A mutation in which base pair change in DNA causes a change in mRNA codon so that a different amino acid is inserted into the polypeptide chain constitutes missen se mutation. On the other hand, a change that results in mRNA codon for an amino acid to Migration is similar to mutation in that new alleles are introduced into the loc al population, although the new alleles are derived from another population, and not from mutation within the same population. In populations with no migrations , the genetic changes cause considerable differentiation in subpopulation of a p opulation. This genetic differentiation among subpopulations gets minimized when exchange of individuals through migration occurs. Only a small amount of migrat ion is sufficient to prevent the accumulation of high level of genetic different iation. However, in spite of migration, the genetic differentiation may continue if other evolutionary forces, such as natural selection for adaptation to local environment, are operating. Random Genetic Drift stop is known as nonsense mutation. Although population is supposed to be infinite in size as per Hardy-Weinberg Law , in practice they are finite or limited in size, although large enough so that chance factors have little effects on allelic frequencies. Some populations, how ever, may be small and the chance factors may produce large changes in allelic f requencies. Ran-dom change in allelic frequency due to chance is called random g enetic drift or simply genetic drift. Ronald Fisher and Sewall Wright, who laid the foundations of population genetics, were the first to describe how genetic d rift affects the evolution of Most of the newly arising mutations are harmful to the organism, and are elimina ted from the population in successive genera-tion. Some mutations may result in amino acid changes that cause detectable change in structure or function of the organism. Such neutral mutations also do not partici-pate in evolution. Similarl y a base pair may change mRNA codon that inserts the same populations. Imagine a small population with equal number of individuals with either of the t wo biallelic traits. In a population of twenty Variation, Biosystematics, Population Genetics and Evolution 141 individuals, say 15 carry homozygous domi-nant genotype, 5 heterozygous genotype and 10 homozygous recessive genotype. the frequency of a allele would be 5 + 20

= 25, i.e. 0.62. Supposing all the individuals carrying the dominant allele (AA or Aa genotypes) perish, the population will be left with only a allele, and he nce allelic frequency of a equals 1. The genetic drift, and consequently evoluti on has occurred in les may bounce a bit, but do not stray too much from steady state. Natural Selection the population due to chance. Natural selection is the deriving force of adaptive evolution. The theory of nat ural selection was first developed independently by Charles Darwin and Alfred Ru ssel Wallace and presented at the Linnean Society of London in 1858. Darwin furt her pursued this theory and published in his famous book, The Origin of Species (1859). The theory has undergone considerable refinement through the incorporati on of genetic concepts in the succeeding years. The theory is based on two premi ses. First- In all organisms, more offspring are produced than survive and repro duce (originally proposed by Thomas Malthus); Second- Organisms differ in their ability to survive and reproduce, and some of these differences are due to genot ype. It is deduced that in every generation, the genotypes that promote survival in the prevailing environment (favoured genotypes) must be present in excess am ong individuals of reproductive age , and hence the favoured genotypes will cont ribute disproportionally to the offsprings of the next generation. As a result o f this process, the alleles that enhance survival and reproduction increase in f requency from generation to generation, and the population becomes progressively better equipped to survive and reproduce in the prevailing environment (Surviva l of the fittest). This progressive genetic improvement constitutes the process of evolutionary adaptation. In addition to random factors such as floods, fires or cyclones which cause gene tic drift due to mortality, the genetic drift may also result from other causes. In small populations, although large number of gametes are produced, only a few will participate in fertilization and represented in the zygotes of next genera tion. This process of random sampling (sampling error) of gametes from generatio n to generation, may result in changes in allele frequencies by chance, and consequent genetic drift Genetic drift may cause a number of evolutionary changes in addition to changes in allelic frequencies. As different populations may experience genetic drift in different directions, and as a result, the populations diverge in allelic frequ encies. Also, we expect more genetic drift in smaller populations as compared to larger populations. Mutation versus Random Genetic Drift Mutation continues to introduce new genetic changes in populations, whereas rand om genetic drift removes certain traits from the population. How does the balanc e between two opposite forces is achieved? Infinite alleles model attempts to ex plain this balance. According to this model, each mutation in a gene is assumed to generate a novel allele that has never been seen before. The model also assum es that random genetic drift occurs by the repeated random sampling, as describe d earlier. In this situation the mutation and genetic drift balance each other, and an equilibrium is reached. In this state of balance, some alleDarwinian Fitness and Fitness Coefficient For natural selection to operate, survival of the fittest is not the only criter ia, the population should be able to reproduce and pass on the fit genes to the next generation. The relative reproductive ability of a genotypes is represented as Darwinian fitness. It is represented by W, and when comparing populations, t he one able to produce most 142 Plant Systematics offspring is given value of 1. Other popula-tions are assigned value relative to this. Supposing out of the four populations A produces 15 offspring, B 10, C 8 and D 5 offspring. The population A will be assigned W value of 1, B 10/15 = 0.6 6, C 8/15 = 0.53, being recessive, s equals 1, and the number of a alleles eliminated by selection will be proportional to q2s according to Hardy-Weinberg proportions, where as t

he number of new a alleles introduced by mutation will be proportional to m. At equilibrium the two should balance and as such q2s = m and D 5/15 = 0.33. It is, however, important to consider the number of offspring surviving, rather than number produced. Darwinian fitness can also be used to calculate Selection coefficient, which is a measure of relative intensity of selection against a gen otype, or selective disadvantage of a disfavoured genotype. It is symbolized as s and calculat-ed as: t q__ s + t where q stands for equilibrium value. In s = 1 - W those cases where the harmful allele shows partial dominance in having small det rimental effect on fitness of the heterozygous carriers then the fitness of geno types AA, Aa and aa can be written as 1 : 1 - hs : 1 - s, where hs stands for se lection coefficient against heterozygous carriers, h refers to Thus for above populations the selection coefficient would be calculated as: Pop ulathe degree of dominance. When h = 1, the tion A s = 1- 1 = 0, B 1 - 0.66 = 0.44, C 1 - 0.53 harmful allele is completely dominant, but = 0.47, D 1 - 0.33 = 0.67. Thus Population A when h = 0 it is comp letely recessive, and has zero selective disadvantage, B 44%, C when h = 1/2 the fitness of heterozygous 47%, and D 67% selective disadvantage. genotype is average of homozygous genotypes. For most harmful alleles that show partial dominance, the value of h is smaller than 1/2. Whereas the estimation of fitness is relatively easier with microorganisms with shorter life span, the same in higher organisms may take several years of study. In such organisms it is convenient to divide fitness into its component parts, and estimate these parts separately. The commonly used components include: viabi lity, the probability that the zygote of a genotypes survives up to reproductive stage, and fertility, the average number of offspring produced by a genotype du ring the reproduc-tive period. Heterozygote superiority Selection-Mutation Balance Whereas in most cases of selection, the fitness of heterozygote is intermediate between dominant and recessive genotypes, or equalling one of them, some cases o f se-lection lead to superior heterozygote. The phenomenon is known as overdomin ance, heterozygote superiority or heterosis. An equilibrium of allelic frequenci es is maintained, because both alleles are favoured in heterozygous state. Selec tion will lead to changes in allelic frequencies, but once the equilibrium is re ached, the frequencies will stabilize. With overdominance the fitness of genotyp es AA, Aa and aa may be written as 1 - s : 1 : 1 - t, where s and t are the sele ction coefficients against AA and aa, respectively. The proportion of A alleles eliminated by selection in event of random mating is p2s/ p = ps, and the propor tion of a alleles eliminated by selection is q2t/q = qt. When the The process of natural selection reduces the frequency of harmful alleles in a p opulation. These harmful alleles, however, are never eliminated, since mutations from wild types continuously produce harmful alleles. These two opposite forces maintain an equilibri-um within a population. In selection-mutation balance new mutations exactly off-set selective eliminations. In populations (with A a alle les) with harmful (lethal) allele Variation, Biosystematics, Population Genetics and Evolution 143 equilibrium is reached and the selective elimination of the two alleles is balan ced. views concerning the evolution of species. Kimura (1968) proposed neutral theory of evolution, according to which most genetic variation observed in natural pop ulations due to accumulation of neutral mutations. Whereas the nonneutral mutati on affects the phenotype of the organism and can be acted upon by natural select

ion, the neutral mutation does not affect the phenotype of the organism and not acted on by natural selection. Since neutral mutations do not affect phenotype, they spread throughout a population according to their frequency of appearance a nd to genetic drift. This non-Darwinian evolution has been called as survival of the luckiest as opposed to survival of the fittest as advocated by Darwin-ian t heory. Although Kimura agreed with Darwin that natural selection is responsible for adaptive changes in species during evolution, but he stressed that modern va riation in gene sequences is explained by neutral variation rather than adaptive qs t = s + t With overdominance of one allele, the allele frequencies ultimately reach equilib-rium, but the rate of approach depen ds on magnitudes of s and t. The equilibrium is reached much faster, when there is a strong selection against homozygotes. Molecular evolution Molecular evolution involves changes in populations at the level of DNA and prot ein sequences. It attempts to correlate these changes with evolution of new gene s and organisms. Whereas the population genet-ics is concerned with changes in g ene fre-quencies from generation to generation, the molecular evolution consider s much longer time frames, associated with speciation. variation. The field of molecular evolution is multidisciplinary, involving data from genet -ics , ecology, evolutionary biology, statistics and even computer science. The analysis of molecular data can help in unravelling the historical records preser ved in genomes, and identifying the dynamics behind evolution-ary processes to u nderstand and reconstruct In further elaboration of their theory, Kimura et al., (1974) developed five pri nciples that govern the evolution of genes at molecular level: 1. For each protein, the rate of evolution, in terms of amino acid substitutions, is approximately constant with regard to n eutral substitutions that do not affect protein structure or function. the chronology of change. Molecular evolution mostly operates by substitutions that lead to the change of codon for one amino acid to another. Leucine codon CUU, for example can be chang ed to isoleu-cine codon AUU by change in single base pair of DNA, where the chan ge to codon for chemically dissimilar asparagine (AAU), two base pair changes mu st occur. The meth-ods of analysing DNA sequence data, and its utilization in ph ylogenetic analysis is discussed in detail in the chapter on Molecular Systemati cs. 2. Proteins that are functionally less important for the survival of an organ-ism, or parts of a protein that are less im portant for its function, tend to evolve faster than more important pro-teins or regions of a protein. In other words, during evolution, less important proteins will accumulate amino acid substitutions more rapidly than important proteins. 3. Those amino acid substitutions that Neutral Theory of Evolution do not disrupt the existing structure and function of a protein (conserva- tive substitutions) occur more fre- quently in evolution than those which disrupt (di sruptive substitutions) ex-isting structure and function. Molecular studies over the last few decades have also seen emergence of alternat e 144 Plant Systematics 4. Gene duplication must always precede

America) are well established species but readily interbreed when brought the emergence of a gene having a new function. into the same area (vicarious species). 5. Selective elimination of definitively 2. Ecological isolation: Two species deleterious mutations and random fix-ing of selectively neutral or very slightly deleterious alleles occur far more frequently in evolution than Darwinian selec tion of definitely adoccupy the same general area but occupy different habitats. Silene alba grows on light soils in open places while S. dioica on heavy soils in shade. Their habit ats rarely overlap, but when they do, hybrids are encounvantageous mutants. Although, the DNA sequencing of hundreds of thousands of different genes from th ousands of species has provided compelling support for the five principles, ther e are, however, some geneticists called selectionists, who oppose neutralist the ory. It is, however, agreed by all that genetic drift and natural selection both play key roles in evolution. tered. 3. Seasonal isolation: Two species occur in the same region but flower at different seasons. Sambucus racemosa and S. nigra flower nearly 7 weeks apart. 4. Temporal isolation: Two species flower during the same period but during different times of the day. Agrostis te nuis flowers in the afternoon, whereas A. stolonifera flowers in the Speciation Speciation is a general term for a number of different processes which involve t he pro-duction of new species. The speciation commonly results from the developm ent of bar-riers to gene flow. Different types are isolat-ing mechanisms are res ponsible for the development of barriers. morning. 5. Ethological isolation: Two species are interfertile but have different pollinators. Humming-birds for example, are attracted to red flowers and hawk-moths to white ones. 6. Mechanical isolation: Pollination beIsolating Mechanisms Isolation is the key factor preventing inter-mixing of distinct species through preven-tion of hybridization. Based on whether isolating mechanisms operate befo re or after sexual fusion, two main types of mechanisms are distinguished: prezy gotic mechanisms and postzygotic mechanisms. A detailed classification of isolat ing mechanisms is presented below. tween two related species is prevented by structural differences between flow- e rs, as for example between Ophrys insectifera and O. Apifera. II. Post-pollination Mechanisms 1. Gametophytic isolation: This is the commonest isolating mechanism wherein cross-pollination occurs but the pollen tu be fails to germinate or if germinated, it can t reach and penetrate the embryo sac. 2. Gametic isolation: In such cases, reA. Prezygotic mechanisms (operating before sexual fusion) ported in several crop plants, the pollen tube releases the male gametes into th e embryo sac, but gametic and/ or endospermic fusion does not occur. I. Pre-pollination Mechanisms 1. Geographical isolation: Two species are separated geographically by a gap larger than their pollen and seed dis-pers al. Platanus orientalis (Mediterra-nean region) and P. occidentalis (North B. Postzygotic mechanisms (operating after sexual fusion) 1. Seed incompatibility: The zygote or even immature embryo is formed but

Variation, Biosystematics, Population Genetics and Evolution 145 fails to develop and as such a mature seed is not formed. The phenomenon is comm only encountered in cross between Primula elatior and P. veris. allopolyploids. Such allopolyploids depict normal pairing at meiosis and thus re present well-isolated, phenotypically as well as genotypically distinct species. Classical example hybridization followed by polyploidy is provided by evolution of bread wheat Triticum aestivum, although it has taken a period of several tho usand years starting from Neolithic times. The two diploid species Triticum mono coccum (AA) and Aegilops speltoides (BB) were involved in the evolution of tetra ploid T. dicoccum (AABB). Subsequent hybridization between the latter and Aegilo ps tauschii (Syn: A. squarrosa) (DD) and subsequent duplication resulted in the evolution of hexaploid T. aestivum (AABBDD). Abrupt speciation also results from the phenomenon of apomixes, and an increasing number of species are being forme d in genera such as Taraxacum, Euphrasia and Alchemilla, where the normal reprod uctive process is bypassed and any expressive mutations are retained within the population. 2. Hybrid inviability: Mature seed is formed and manages to germinate but the F1 hybrid dies before the flower-ing sta ge is reached. The phenom-enon is commonly encountered in crosses between Papave r dubium and P. rhoeas. 3. F1 hybrid sterility: F1 hybrids are fully viable and reach flowering stage but flowers may abort or abortion may occ ur as late as F2 embryo formation, with the result that the F1 hybrid fails to produce viable seeds. 4. F2 hybrid inviability or sterility: F2 hybrid dies much before reaching the flowering stage or fails to produce seeds. Once prezygotic isolation is partially achieved, there is a snowball effect in w hich the rate of accelerates. Individuals who engage in interspecific mating suf fer an in-creasing disadvantage until at last the bar-rier of gene flow is compl ete. New species may develop through the mechanism of abrupt speciation or gradu al speciation. Gradual speciation This is a more common phenomenon in nature. It may involve phyletic evolution wh en one species might evolve into something different from its ancestor over a pe riod of time (phyletic speciation). Alternatively, a population belonging to a s ingle species might differentiate into two evolutionary lines through divergent evolution (additive speciation). Abrupt speciation The phenomenon of abrupt speciation is commonly involves sympatric populations o f two different species and as such is also known as sympatric speciation. It is commonly the result of a sudden change in chro-mosome number or structure, prod ucing instantly an almost irreversible barrier between populations and thus effe ctively isolating them. Phenomenon is met in gen-era such as Tragopogon, Bromus and Senecio (see examples under Chromosomes in chap-ter on Taxonomic evidence) . The species are often well isolated and any chance hybridization fails to culmi nate into suc-cessful hybrids because of genomic differ-ences. In some cases, ho wever, hybridizaPhyletic Speciation The concept of phyletic speciation has been the subject of considerable debate. It is the sequential production of species within a single evolutionary lineage. Species A might, over a period of time, change through species B and C into spe cies D without ever splitting. The new species produced in this manner are vario usly called successional species, palaeospecies, and allochronic species. The sp ecies which have become extinct in the process are termed taxonomic tion may be accompanied by chromosome extinctions. Wiley (1981), while agreeing duplication resulting in the formation of with the concept of phyletic character trans-

146 Plant Systematics Ancestral species Ancestral species Environmental differentiation

Race A Race B C Subspecies B Adaptive radiations Subspecies A Reproductive Isolation Species B Reproductive isolation Species A

CC SpeciesB Species A Specis B Species A A B Figure 6.2 Allopatric speciation. A: Allopatric speciation through environmental differentiation, successive adaptive radiations and development of reproductive isolation. B: All opatric speciation resulting from geographical separation of populations of an a ncestral species. Variation, Biosystematics, Population Genetics and Evolution 147 formation, rejects the concept of phyletic mental differentiation. They then undergo adaptive radiations to develop physica l and physiological differences, which sooner or later get genetically fixed (ec otypes). With further morphological and physiological differentia-tion, they for m distinct varieties (or subspecies). Development of reproductive isolation esta blishes these as distinct species that will retain their identity even if a futu re chance should draw them together (Figure 6.2 A). speciation on the grounds that: 1. Recognition of phyletic species is an arbitrary practice. Mayr (1942) argues that delimitation of species, which do no t belong to the same time-scale, is difficult. 2. Arbitrary species result in arbitrary speciation mechanisms. 3. Phyletic speciation has never been satisfactorily demonstrated. Additive Speciation Additive speciation is the commonest mode of speciation, which adds to the diver

sity of living organisms. Mayr (1963) suggested the occurrence of reductive spec iation, whereby two previously independent species fuse into a third, new specie s, themselves becoming extinct. Hybridization likewise produces new species but this always leads to an addition 2. Allopatric introgressive speciation: Although origin of species through hybridization is commonly results from sympat ric species, examples of speciation involving two allopatric species, which had contacts in the past, are also reported. Quercus brandegei, now confined to Cape Region of Baja California, extended to west in Tertiin the number of species. ary times and had a narrow zone of contact in Edwards Plateau escarp-ment area, with Q. virginiana of S. E. Coastal plain of U.S.A. Allopatric introgression bet ween the two species occurred at the contact zone, but the genes spread slowly b ecause of limited contact of parental species and predominantly rhizomatous prop agation. The introgressed population, now in contact with only Q. virginiana, is sufficiently stabilized to be classified as It is impossible to imagine that two evo-lutionary species can actually fuse to pro-duce a third species and themselves become extinct. This may happen in a par ticular re-gion, but not over the entire range of these species. The various mod es of additive speciation are described below: 1. Allopatric speciation: Lineage independence and consequent speciation result from geographical separation of lineag es, i.e. the actual physical sepa-ration of two relatively large popu-lations of a single species. Over a period of time, such separation would enable these geo graphical races to de-velop and maintain gene combinations controlling their mor phological and physiological characters. The develop-ment of reproductive isolat ion would sooner or later result in the establisha distinct species Q. fusiformis. 3. Allo-parapatric speciation: Such ment of distinct species (Figure 6.2 B). speciation occurs when two populations of an ancestral species are separated, di fferentiate to a degree that is not sufficient for lineage independence, and the n develop lineage independence during a period of parapatry (limited sympatry). It differs from allopatric speciation in the sense that speciation is completed after a period of sympatry and the process of attaining lineage independence is potentially reversible because it is possible that two partly differentiated pop ulations could form a single evolutionary lineage showing clinal variaAllopatric speciation may also result from the development of new species along the boundaries of a large central population. These marginal populations (races) get separated from the main population during environ148 Plant Systematics tion after they meet rather than the period of sympatry reinforcing differsomal modifications. The resultant chromosome arrangement must be fully viable i n the homozygous state but of reduced viability in the heteroences between them. 4. Parapatric speciation: This occurs when two populations of an ancestral species differentiate despite the fact that no complete disjunction has oc-curred. The daughter species may share a small f raction of their respec-tive ranges and interbreed within this narrow contact zo ne and yet still difzygous state. 6. Sympatric speciation: The examples ferentiate. of sympatric speciation due to hybridization and apomixis have been discussed un der abrupt speciation. The process of ecological sympatric speciation is a slow one of gradual speciation. The ecological differences in the habitats result in adaptive radiations in populations which gradually evolve into new species. 5. Stasipatric speciation: This is simi-

lar to parapatric speciation except that it results from spontaneous chromoChapter 7 Taxonomic Evidence Over the last few decades, the affinities be-tween plant groups have been redefi ned as more and more information is accumulated from various sources. Newer appr oaches in recent years include (a) increasing reliance last two centuries, more and more microscopic characters of morphology were inco rporated. Although floral morphology has been the major material for classificat ions, other morphological characters have also contributed in specific groups of plants. The diveron phytochemical information (Chemotaxonomy); (b) studies on ultrastructure and sity of morphological features has alr eady micromorphology; (c) statistical analysis of been discussed in detail under Descriptive the available data without much a priori terminology in Chapter Fou r. weighting and providing a synthesis of all the available information (Taxometric s); and (d) analysis of phylogenetic data to con-struct phylogenetic relationshi p diagrams (Cladistics). The aforesaid disciplines con-stitute the major modern trends in tax-onomy. Data continues to flow from differ-ent disciplines, so that the process of analy-sis and synthesis is an ongoing activity. Tax-onomy (Syste matics) is as such a field of unending synthesis. The following disci-plines hav e contributed to a greater or lesser extent to a better understanding of taxo-no mic affinities between plants. Habit Life-forms though of little significance to taxonomy allow a means of estimating ada ptiveness and ecological adjustment to the habitat. In Pinus, bark characters ar e used for identification of species. Woody and herbaceous characters have been the primary basis of recognition of Lignosae and Herbaceae series within dicots by Hutchinson (1926, 1973). MORPHOLOGY For several decades it was believed that trees or shrubs with simple leaves repr esented the most primitive condition within angiosperms. Increased evidence over the last decade, however, is pointing towards the assumption that the perennial herbaceous condition in paleoherbs such as Ceratophyllaceae, Nymphaeaceae and P iperaceae represents the archetype of the most primitive angiosperms. Morphology has been the major criterion for classification over the last many ce nturies. The initial classifications were based on gross morphological character s. During the 150 Plant Systematics Underground parts used fruit morphology in identification of Indian genera of Compositae (Ligulifl orae). In Asteraceae the shape cypsela (usually called achene), presence or absenc e of pap-pus and whether the pappus is represented by hairs, scales or bristles, the presence or absence of beak, and its length, the number of ribs on the cyps ela constitute valuable identifying features. The number of capsule valves is used in segregating genera in fam- ily Caryophyllaceae (Melandrium, Silene, Cerastiu m). Seed characters are valuable identification features in the genus Veronica. Rhizome characteristics are important for identification of various species of t he genus Iris. Similarly, bulb structure (whether bulbs are clustered on rootsto ck or not) is an important taxonomic criterion in the genus Allium. Davis (1960) has divided Turk-ish species of the subgenus Ranunculus of genus Ranunculus bas ed on rootstock and habit. Leaves Leaves are important for identification in palms, Salix and Populus. The genus Azadirachta has been separated from Melia ANATOMY

among other features by the presence of Anatomical features have played an incre asunipinnate leaves as against bipinnate in the latter. Similarly, the genus Sorbus has been separated from Pyrus, and genus Sorbaria separated from Spiraea on the baingly important role in elucidation of phylogenetic relationships. Anatomical ch aracteristics are investigated with the help of a light microscope; whereas ultr astructure (finer details of contents) and micromorphology (finer details of surface featur es) are brought out using an electron microscope. Anatomical work of taxonomic s ignificance was largely undertaken by Bailey and his students. Carlquist (1996) has discussed the trends of xylem evolution, especially in the context of primit ive an-giosperms. sis of pinnate leaves. Stipules are an im-portant source for identification in V iola and Salix. Leaf venation is important for the iden-tification of the specie s in Ulmus and Tilia. Interpetiolar stipules are useful for identifi-cation with in family Rubiaceae. Flowers Floral characters are extensively used in de-limitation of taxa. These may inclu de the calyx (Lamiaceae), corolla (Fabaceae, Cory-dalis), stamens (Lamiaceae, Fa baceae-Mimosoideae), or carpels (Caryophyllaceae). A gynobasic style is characte ristic of Lamiaceae. Similarly, the gynostegium characterizes Asclepiadaceae (no w recog-nized as subfamily Asclepiadoideae of fam-ily Apocynaceae). Different sp ecies of Euphorbia have a distinctive cyathium inflorescence with clusters of ma le flowers each rep-resented by a single stamen. Wood anatomy Wood represents secondary xylem constituting the bulk of trees and shrubs, forme d through the activity of vascular cambium. It primarily consists of tracheids a nd vessels. Tracheids are long narrow elements with tapering ends, imperforate a t ends, and transfer of water and minerals occurring through pit-pairs (two adja cent pits of two traFruits cheids, separated by primary cell walls). The vessels, on the other hand, are co mposed of vessel elements, much broader than trac- heids and with perforation pl ates at ends (with opening not having primary walls unlike pit-pair). Vessel ele ments are joined end to end to form long tubes, the vessels. Perforation plate m ay be simple with a Fruit characteristics are very widely used in identification. Coode (1967) used only fruit characteristics in delimitation of species of the genus Valerianella. Singh et al. (1972) Taxonomic Evidence 151 single opening, or compound with several openings. Latter with elongated opening s in a row like a ladder is known as scalariform, a common type in primitive ang iosperms. associated gaps (lacunae) left in the vascular cylinder of stem at each node are distinctive for several groups. The node may have single gap (unilacunar) from single leaf trace or three leaf traces (two additional commonly entering stipule s) or three gaps (trila-cunar) associated with three leaf traces (Figure 7.1) The genus Illicium has been separated from Winteraceae because of unilacunar nodes, continuous pseudosiphonostel e and the absence of granular material in stomatal depressions. Trichomes Figure 7.1 Nodal anatomy. A: Unilacunar node with one leaf trace; B: Trilacunar Trichomes constitute appendages of epider-

mis which may be non-glandular or glandunode with three leaf traces; C: lar. Non-glandular trichomes may be in the Unilacunar node with three leaf form of simple unicellular or multicellular trac es. hairs (common in Brassicaceae, Lauraceae and Moraceae), in the form of vesicles, peltate hairs (Olea) or flattened scales. Vessels are absent in Gymnosperms, but present in Angiosperms. It is commonly be lieved that there has been a progressive evolution in angiosperms from tracheids to long, narrow vessel elements with slanted, sca-lariform perforation plates, to short, broad vessel elements with simple perforation plates. Studies on wood anatomy have contributed largely in arriving at the conclusion that Amentiferae constitute a relatively advanced group, and that Gnetales are not ancestral to a ngiosperms. Bailey (1944) concluded that vessels in angiosperms arose from trach eids with scalariform pitting, whereas in Gnetales they arose from tracheids wit h circular pitting, thus suggesting an independent origin of vessels in these tw o groups. Demonstration of vessel-less angiosperms (Winteraceae, Trochodendracea e), also having other primitive features, has led to the conclusion that angiosp erm ancestors were vessel-less. The separation of Paeonia into a distinct family Paeoniaceae and Austrobaileya into a sepa-rate family Austrobaileyaceae has bee n supf '

Figure 7.2 Trichomes. A: Simple unicellular pP orted bY studies of wood anatomYy . hair; BB:: Mulr hair; C: Scale; D: Canddellaabbraa h of Verbascum; E: Vesicullaar haaiirr of Atriplex; F: Peltate hair; G: Stellate hair Styrax; H: Secretary gland of Thymus; I: Stinging hair of Urtica. Nodal anatomy has considerable signifi-cance in angiosperm systematics. The number of vascular traces entering leaf base and 152 Plant Systematics 3

G Figure 7.3 Stomatal apparatus in Angiosperms. A: Anomocytic type with epidermal cells around stomata not differentiated; B: Paracytic type with two or more cells parallel to the guard cells differentiated as subsidiary cells; C: Diacytic type with two s ubsidiary cells at right angles to the guards cells; D: Anisocytic type with thr ee subsidiary cells of unequal size; E: Actinocytic type with stomata surrounded by a circle of radiating cells; F: Tetracytic type with four subsidiary cells; G: Cyclocytic type with concentric rings of subsidiary cells; H: Graminaceous ty pe with dumb-bell shaped guard cells with two small subsidiary cells parallel to the guard cells.

Branched hairs may be dendroid, stellate (Styrax) or candelabrum-like (Verbascum ). Glandular trichomes may be sessile or stinging hairs of Urtica are highly specialized with silica tip which readily br eaks when hair is touched. The broken tip is sharp like a syringe and easily pen etrates stalked and present a variety of forms. Unicellular glandular hairs of Atriplex are bladder-like (Figure 7.2) with few-c elled stalk and basal cell and they secrete salt. Others may secrete nectar (cal yx of Abutilon), mu-cilage (leaf base of Rheum and Rumex). The the skin injecting irritating cell contents. Trichomes hold considerable promise in systematics of angiosperms. Trichomes hav e been of considerable help in Cruciferae (Schulz, 1936), especially in the gene ra Taxonomic Evidence 153 Arabis and Arabidopsis. Trichome characters are very useful in the large genus A straga-lus (with more than 2000 species). The Hi-malayan species Hedera nepalens is is dis-tinguished from its European relative H. helix in having scaly trichom es as against stellate in the latter. In family Combretaceae the trichomes are o f immense significance in classification of genera, species or even varieties (S tace, 1973). Trichomes are also diagnostic characters for many species of Ver-no nia (Faust and Jones, 1973). stomata in the former as against anomocytic in the latter. The stomatal features , however, are not always reliable. In Streptocarpus (Sahasrabudhe and Stace, 19 79) cotyledons have anomocytic while mature organs have anisocytic stomata. In Phyla nodiflora (syn Lippia nodiflora) the same leaf may show anomocytic, anisocytic, diacytic and pa racytic stomata (Pant and Kidwai, 1964). Leaf anatomy Epidermal features The florets of Poaceae are reduced and do not offer much structural variability. Leaf anatomy has been of special taxonomic help in this family. The occurrence of the C-4 pathway and its association with Kranz anatomy (dense thick-walled ch lorenchymatous bundle sheath, mesophyll simple), has resulted in revised classif ication of several genera of grasses. Melville (1962, 1983) developed his gonoph yll theory largely on the basis of the study of venation pattern of leaves and f loral parts. The rejection of San-miguelia and Furcula as angiosperm fossils fro m the Triassic has largely been on the basis of detailed study of the venation p attern of leaves (Hickey and Doyle, 1977). The more recent rediscovery of Sanmig uelia from the Upper Triassic of Texas (Cornet 1986, 1989) points to presumed an giosperm incorporating features of both monocots and dicots. Discovery of the La te Triassic Pan-naulika (Cornet) from the Virginia-North Carolina border has reo pened the possibilities of Triassic origin of angiosperms. Epidermal features are also of considerable taxonomic interest (SEM epidermal fe atures are discussed under ultrastructure and micromorphology). Prat (1960) demo nstrated that one can distinguish a Festucoid type (simple silica cells, no bice llular hairs) and Panicoid type (complicated silica cells, bicellular hairs) of epidermis in grasses. Stomatal types (Figure 7.3) are distinctive of certain families such as Ranuncul aceae (anomocytic), Brassicaceae (anisocytic), Caryophyllaceae (diacytic), Rubiaceae ( paracytic), and Poaceae (graminaceous). Ano-mocytic type has ordinary epidermal surrounding the stomata. In others the epider-mal cells surrounding the stomata are dif-ferentiated as subsidiary cells. There may be two subsidiary cells at right angles to the guard cells (diacytic), two are more parallel to the guard cells (paracytic), or three sub-sidiary cells of unequal size (anisocytic). Other type s include actinocytic type with stomata surrounded by a ring of radiating cells, cyclocytic with more than one concentric rings of subsidiary cells and tetra-cy tic with four subsidiary cells. The stomatal complex of Poaceae is distinctive i n having two dumb-bell shaped guard cells with two small subsidiary cells parall

el to the guard Floral anatomy cells. Floral anatomy has been one of the thoroughly explored areas, with significant c on-tributions to the understanding of the phylogeny of angiosperms. Vascular tra ces in the carpels of various genera of the family Ranunculaceae have confirmed the origin of achene (Ranunculus, Thalictrum, etc.) from follicle (Delphinium, A quilegia, etc.) through successive reduction in the number of ovules ultimately to one. The additional Stace (1989) lists 35 types of stomata in vascular plants. Closely related famil ies Acanthaceae and Scrophulariaceae are distinguished by the presence of diacyt ic 154 Plant Systematics traces which would have gone to other ovules, now aborted, can be observed in ma ny genera. There, thus, is no justifica-tion for Hutchinson s separation of achene -bearing genera and follicle-bearing genera into separate families Ranunculaceae and species. Floral anatomy also supports the separation of Menyanthes from Gentiana ceae into a distinct family Menyanthaceae. The genus Centella is separated from Hydrocotyle on the basis of inflorescence being a cyme, and ovules receiving vas cular supply from alternate bundles. In Hydrocotyle, the inflorescence is an umb el and the ovules receive vascular supply from fusion of two adjacent bundles. P aeonia is a classical example of a genus, which was removed from family Ranuncul aceae into a distinct family Paeoniaceae. The separation has been supported by e vidence from morphology, embryology and chromosomes. Floral anatomy also support s this separation, as both sepals and petals have many traces, carpels have five traces and the stamens are centrifugal. Developmental studies have indicated th at some flowers, such as Apiaceae and Ericaceae, that appear to have free petals , are gamopetalous early in development. They are, therefore, considered to have evolved from gamopetalous ancestors. Helleboraceae, respectively. Melville (1962, 1983) developed his gonophyll theory after studying the vascu-la ture of carpel and other floral parts through the clearing technique. He believe d the an-giosperm carpel to be a modified dichotomous fertile branch adnate to t he petiole of a leaf. Sporne (1971) cautioned against such a dras-tic conclusion citing the example of bathroom loofah. The genus Melandrium was segregated from Silene on the basis of the ovary being unilocular as against partly septate in Silene. Detailed floral anatomy revealed that in all the species of both genera, the ovary is mul-tilocular, at least in the early stages of de-velopment. The septa break down to various degrees in di fferent species as the ovary de-velops. Thus structurally, the ovaries are simil ar. The two genera were consequently merged into the single genus Silene. EMBRYOLOGY The inferior ovary in angiosperms has been formed in two ways: appendicular oriEmbryology has made a relatively lesser contribution in understanding taxonomic affinities. This is primarily because of long preparatory work needed for embryo logical studies. More often, the study of hundreds of preparations may reveal ju st a single embryological characteristic of any significance. It may take many y ears of laborious and painstaking research to study even a few representatives o f a family. The embryo-logical features of major significance include microsporo genesis, development and structure of ovule, embryo sac development, endosperm a nd embryo development. gin (formed by fusion of calyx, corolla and their traces to the ovary wall; in t his case, all vascular traces have normal orientation, i.e. phloem towards the o utside) or by axial invagination (formed by depression of the thalamus; the inne r vascular traces have reverse orientation, i.e. phloem towards the inside). Stu dies on floral anatomy have confirmed that in a large majority of families, the

inferior ovary is of appendicular origin. Only in a few cases (Rosa, Cactaceae, etc.) is the origin by axial invagination of the thalamus. Floral anatomy has also supported the inclusion of Acer negundo under Acer, and does not support its separation into a distinct genus Negundo. Although this spe cies is specialized in having a dioecious habit and anemophily, the anatomy of t he flower shows unspecialized features of other Families marked out by distinct embryological features A number of families of angiosperms are characterized by unique embryological fe atures found in all members. These include: Taxonomic Evidence 155 Podostemaceae swollen petiole, semiepigynous disc and spiny fruit. The following embryological features support this separation: (i) pyramidal pollen grains with 3 folded cre sts (bluntly triangular and basin shaped in Onagraceae); (ii) ovary semi-inferio r, bilocular with single ovule in each loculus (not inferior, trilocular, with m any ovules); (iii) Polygonum type of embryo sac (not Oenothera type); (iv) en-do sperm absent (not present and nuclear); (v) embryo Solanad type (not Onagrad type); (vi) one cotyledon extremely reduced (both not equal); and (vii) fruit large one -seeded drupe (not loculicidal capsule). Family Podostemaceae includes perennial aquatic herbs, which have a unique embry o-logical feature in the formation of a pseudoembryo sac due to the disintegrati on of the nucellar tissue. The family is also characterized by the occurrence of pollen grains in pairs, bitegmic tenuinucellate ovules, bisporic embryo sac, so lanad type of embryogeny, prominent suspensor hausto-ria, and absence of triple fusion and, conse-quently, endosperm. Cyperaceae Family Cyperaceae is characterized by the formation of only one microspore per m i-crospore mother cell. Following meiosis, of the four microspore nuclei formed, only one gives rise to pollen grain. Besides Cyperaceae, only Epacridaceae in a few members shows the degeneration of three microspore nuclei. Cyperaceae is di stinct from these taxa in pollen shedding at the 3- celled stage, as against the 2-celled stage shedding in Epacridaceae. Paeonia Onagraceae The genus Paeonia was earlier included under the family Ranunculaceae (Bentham a nd Hooker; Engler and Prantl). Worsdell (1908) suggested its removal to a distin ct family, Paeoniaceae. This was supported on the basis of centrifugal stamens ( Corner, 1946), floral anatomy (Eames, 1961) and chromosomal information (Gregory , 1941). The genus as such has been placed in a distinct monogeneric family, Pae oniaceae, in all modern systems of classification. The separation is supported b y the following embryo-logical features: (i) centrifugal stamens (not centripeta l); (ii) pollen with reticulately-pitted exine with a large generative cell (not granular, papillate and smooth, small generative cell); (iii) unique embryogeny in which early divisions are free nuclear forming a coenocytic stage, later onl y the peripheral part becomes cellular (not onagrad or solanad type); and (iv) s eed arillate. Family Onagraceae is characterized by Oenothera type of embryo sac, not found in any other family except as an abnormality. This type of embryo sac is 4-nucleat e and is derived from the micropylar megaspore of the tetrad formed. Specific examples of the role of embryological data There are a few examples of the embryologi-cal data having been very useful in t he in-terpretation of taxonomic affinities: Exocarpos The genus Exocarpos (sometimes misspelled Exocarpus) is traditionally placed und er the family Santalaceae. Gagnepain and Boureau (1947) suggested its removal to a distinct family Exocarpaceae near Taxaceae under Gymnosperms on the basis of articulate pedicel, naked ovule and presence of a pollen chamber. Ram (1959) studi

ed the embryology of this genus and concluded that the flower shows the usual Trapa The genus Trapa was earlier (Bentham and Hooker, 1883) included under the family Onagraceae. It was subsequently removed to the family Trapaceae (Engler and Die ls, 1936; Hutchinson, 1959, 1973) on the basis of distinct aquatic habit, two ty pes of leaves, 156 Plant Systematics angiospermous character, the anther has a distinct endothecium and glandular tap e-tum, pollen grains shed at the 2-celled stage, embryo sac of the Polygonum typ e, endosperm cellular, and the division of zygote trans-verse. This confirms tha t the genus Exocarpos is undoubtedly an angiosperm, and a member of the family S antalaceae, with no justification for its removal to a dis-tinct family. The gen us is as such placed in Santalaceae in all the major systems of clasest. The number of nuclei present at the time of shedding is also significant. M ost primitive angiosperms are shed at 2-nucleate stage, whereas in more advanced groups pollen is shed at 3-nucleate stage. sification. Angiosperms mostly have pollen grains of radial symmetry, bilateral symmetry bei ng found in several gymnosperms. Most pollen grains are globose in shape, althou gh boatshaped, ellipsoidal and fusiforms are also met in different angiosperms. Since most pollen grains at least in early stages form tetrads, the outer end of grain is termed distal pole, whereas the inner end where grains meet as proxima l pole, and the line joining the two poles as polar axis. The line running aroun d the pollen at right angles to the polar axis is termed as equator. Loranthaceae The family Loranthaceae is traditionally divided into two subfamilies Loranthoidea e and Viscoideae largely on the basis of presence of a calyculus below the periant h in the former and its absence in the latter. Maheshwari (1964) noted that the Loranthoideae has triradiate pollen grains, Polygonum type of embryo sac, early embryog-eny is biseriate, embryo suspensor present, and viscid layer outside the vascular supply in fruit. As against this, Viscoideae have spherical pollen gra ins, Allium type of embryo sac, early embryogeny many tiered, embryo suspensor a bsent, and viscid layer inside the vascular supply of fruit. He thus advocated s eparation of the two as distinct families Loranthaceae and Viscaceae. The separa tion was accepted by Takhtajan (1980, 1987, 1997), Dahlgren (1980), Cronquist (1 981, 1988) and Thorne (1981, 1992). Pollen aggregation Microsporogenesis yields four microspores which mature into pollen grains. In la rge majority of angiosperms the pollen grains separate prior to release. Such si ngle pollen grains are known as monads. In rare cases pollen grains are released fused in pairs, when they are known as dyads. In many angiosperms the four micr ospores do not separate and the pollen grains form a tetrad. Five different type s of tetrads are differentiated: 1. Tetrahedral tetrad- four pollen grains form a tetrahedron: four grains compacted in a sphere. Such pollen grains are found in family Ericaceae. 2. Linear tetrad- four pollen grains PALYNOLOGY The pollen wall has been a subject of con-siderable attention, especially in an attempt to establish the evolutionary history of angiosperms. Some families, suc h as Asteraceae, show different types of pollen grains (eurypalynous), whereas s everal others have a single morphological pollen type (stenopalynous). Such sten opalynous groups are of considerable significance in systematic palynology. Poll en grains present a number of features of taxonomist interarranged in a straight line as in genus Typha. 3. Rhomboidal tetrad- four pollen grains in one plane, with two separated from one another by close contact of the

other two. 4. Tetragonal tetrad- four grains are in one plane and equally spaced as in Philydrum. 5. Decussate tetrad- four grains in two pairs, arranged at right angles to one another, as in genus Lachnanthes. Taxonomic Evidence 157 is lacking (atectate pollen grain), and the exine appears granular. Above layers of exine are clearly visible under an electron microscope, but when observed un der a light microscope, the inner layer known as nexine, includes endexine plus foot layer of ektexine. The upper layers consisting of columella, tectum and the supratectal sculpturing constitute sexine. i B Pollen wall sculpturing Figure 7.4 Fine structure of Pollen wall. Present on the outer surface of tectum are often certain supratectal projections , which provide a variety of sculpturing to exine wall. In some cases lacking te ctum, the sculpturing is formed by columellae. The common types of sculpturing i nclude: baculate (rod-shaped elements, each known as baculum), clavate (club-sha ped elements), echinate (spine-like elements longer than 1 micron), spinulose (s pine-like elements shorter than 1 micron; scabrate), foveolate (pitted surface w ith pores), reticulate (forming network, each element known as murus and space i n between as lumen), fossulate (longitudinal grooves), verrucate (short wart-lik e elements), gemmate (globose or ellipsoid el-ements), psilate (smooth surface), and striate (having thin striations on surface). A, Intine, B, Endexine, C, Foot layer, D, Baculum and E, Tectum. Note the apertu re formed due to break in the tectum and baculum layers. In some genera, such as Calliandra of Mimosoideae, the pollen grains are connate in a group of more than four. Such pollen grains constitute a polyad. A polyad generally consists of eight pollen grains, and rarely of more than ten. In some members of fam-ily Orchidaceae, as for example genus Piperia, large number of po llen grains form irregular groups, of which there are more than one groups in a theca. These are known as massulae. In subfamily Asclepiadoideae of family Apocy naceae, and several mem-bers of orchidaceae, all pollen grains of a theca are fu sed into a single mass known as pollinium. Pollen aperture Pollen wall The pollen grain wall is made of two princi-pal layers, outer exine and inner in tine. The exine is hard and impregnated with sporopollenin, a substance that mak es it resistant to decay, and enables preservation in fossil record. Exine is fu rther differenti-ated into two layers: outer ektexine and in-ner endexine. The e ktexine is further dis-tinguished into basal foot layer, radially elongate colum ella and roof like tectum Pollen aperture is a specialized region of pollen wall through which the pollen tube comes out. The exine may be inaperturate (with-out an aperture) or apertura te. An aperturate pollen may have a single pore (monoporate), a single slit runn ing at right angles to the equator (monocolpate), three slits (tricolpate), thre e pores (triporate) three slits each with a geminate pore in middle (tricolporat e), with many pores (multiporate) accompanied by a variety of surface ornamentat ions (Figure 7-5). Pollen with one or more slits located at the polar end is acc ordingly termed, monosulcate, disulcate and trisulcate, depending on the number of slits. Pollen grain with slits joined (Figure 7.4). In some taxa the columella may be replaced by granular middle layer. Simi-larly in some primitive angiosperms t ectum 158 Plant Systematics

41

l 11 11 Figure 7.5 SEM of pollen grains. A: Nonaperturate pollen grain of Persea america na; B: Monosulcate pollen grain of Magnolia grandiflora; C: Monoporate pollen grain of Siphonogloss a; D: Tricolporate pollen grain of Scaevola glabra; E: Polyporate spinose pollen grain of Ipomoaea wolcottiana; F: Tricolpate pollen grain of Disanthus cercidif olius. (A, after Fahn, 1982; C, after Mauseth, 1998 courtesy R. A. Hilsenbebeck, Sul Ross State University; F, after Endress, 1977; rest, after Gifford and Fost er, 1988). at poles is termed syncolpate. Aperture having three branches is termed trichoseveral genera with wind pollination. The vestigial scattered patches of adhesiv e layer on wind pollinated pollen have been considered as evidence of the deriva tion of tomosulcate. Monocolpate condition is widely spread in primitive dicots and a majority of mon ocots. The pollen of anemophilous plants is usually small, rounded, smooth, rath er thin-walled and dry with shallow furrows. Anemo-philous pollen is found in Po pulus, Poaceae, Cyperaceae, Betulaceae and several other families. Insect- and b ird-pollinated pollen, on the other hand, is large, sculptured and often coated with adhesive waxy or oily substance. The pollen of Asteraceae is generally high ly elaborate but simplification towards loss of sculpturing has occurred in anemophily from entomophily. Fossil studies over the last three decades have confirmed monosulcate pollen of Clavitopollenites described (Couper, 1958) from Barremian and Aptian strata of t he Early Cretaceous of southern England (132 to 112 Mya) to be the oldest record ed angiosperm fossil with distinct sculptured exine, resembling the pollen of ex tant genus Ascarina. Brenner and Bickoff (1992) recorded similar but inaperturate pollen grains from Taxonomic Evidence 159 the Valanginian (ca 135 Mya) from the Helez formation of Israel, now considered being the oldest record of angiosperm fossils (Taylor and Hickey, 1996). This la st discovery has led to the belief that the earliest angiosperm pollen were with out an opening, the through Transmission Electron Microscopy (TEM). On an average basis, the resolut ion power of SEM is 250A (20 times as good as optical microscope, but 20 times l esser than TEM). Behnke and Barthlott (1983) have made extensive studies of SEM and TEM characters. In most of the examples studied, Electron Microscopy (EM) ch aracters proved to be stable and unaffected by environmental conditions. monosulcate types developing later. Many claims of angiosperm records from the strata, earlier than the Cretaceous w ere made, but largely rejected. Erdtman (1948) described Eucommiidites as a tric olpate dicotyledonous pollen grain from the Jurassic. This, however, had bilater al symmetry instead of the radial symmetry of angiosperms (Hughes, 1961) and a g ranular exine with gymnospermous laminated endexine (Doyle et al., 1975). Among examples of the role of pollen grains in systematics is Nelumbo whose separation from Nymphaeaceae into a distinct family Nelumbonaceae is largely supported by the tricolpate pollen of Nelumbo as against the monosulcate condition in Micromorphology Nymphaeaceae. SEM studies have been made primarily on pollen grains, small seeds, trichomes an d surface features of various organs. In most of these organs (except pollen gra ins), the studies involved the epidermis. The value of epidermal studies lies in the fact that an epidermis covers almost all the organs and is always present,

even in herbarium specimens. The epidermis is thick and stable in SEM preparatio ns and is little affected by environment. However, it is important to note that only comparable epidermis should be studied (e.g. petals of all plants, leaves o f all plants, not petals of some and leaves of others). Most of SEM studies have been con-centrated on seed-coats which are usually thick-walled and stable in v acuum, thus faBrenner (1996) proposed a new model for the evolutionary sequence of angiosperm pol-len types. The earliest angiosperm pollen (from the Valanginian or earlier) was small, circular, tectate-columellate and without an aperture. In the Hauteri vian, there was pos-sible occurrence of thickening of the intine coupled with th ickened endexine and evo-lution of the sulcus. A considerable diversi-fication o f these monosulcate pollens oc-curred in the Barremian. Tricolpate pollen evolve d in northern Gondwana in the lower Aptian. Multicolpate and multiporate pollen arose at a later stager. cilitating quick preparation for SEM examination without the need for complicate d dehydration techniques. The micromorphology of the epidermis includes the foll owing as-pects: Primary Sculpture MICROMORPHOLOGY AND This refers to the arrangement and shape of cells. The arrangement of cells is s pe- cific for several taxa. In Papaveraceae, seedcoat cells by a particular arra ngement form a reticulate supercellular pattern (Figure ULTRASTRUCTURE Although widely used in lower plants, elec-tron microscopy has been a comparativ ely new approach for flowering plants. The finer details of external features (m icromorphol-ogy) have been explored in the recent years by Scanning Electron Mic roscopy (SEM), whereas the minute details of cell contents (ultrastructure) have been discerned 7.6-D), which is a family character. The members of Caryophyllaceae, Portulacaceae and Aizoaceae exhibit a specific arran gement and orientation of smaller and larger cells known as There is specific distribution of long and short cells over the veins in the fam ily centrospermoid pattern. 160 Plant Systematics

s~ i IF

Figure 7.6 SEM seed characteristics of angiosperms. A: Seed of Sceletium campact um (Aizoaceae) showing centrospermoid cell arrangement; B: Seed-coat of Aeginatia indica (Oroba nchaceae) with isodiametric deeply concave cells and reticulate secondary struct ure; C: Single isodiametric tetragonal cell of seed-coat of Matucana weberbaueri (Cactaceae) with heavy secondary sculpturing; D: Seed of Eschscholzia californi ca (Papaveraceae) with cells arranged to form a supercellular net-like pattern; E: Seed-coat of Jacaranda macarantha (Bignoniaceae) with stellate epicuticular s

culpture; F: Seed of Dichaea sp. (Orchidaceae) almost one cell long with heavy m arginal thickenings and irregular secondary sculpture. (From Barthlott, 1984). Poaceae. The shape of cells is mainly de-termined by the outline of the cells, b oundaries of the walls, relief, and cell wall cur-vature (flat, convex or concav e). Outline of cells may be isodiametric (usually tetrago-nal or hexagonal: Figu re 7.6-B and C), elon-gated in one direction (Figure 7.6-F). Cell boundaries of superficially visible anticlinal walls may be straight (Figure 7.6-B and C), irr egularly curved or undulated (S-, U-, omega-, V- types) and are of high taxonomi c significance in family Cactaceae and Orchidaceae. Relief of the anticlinal bou ndary may be channelled or raised. In primitive members of Cactaceae, cell junctions are depressed, whereas in derived Cactinae they are raised. The curvature of outer periclinal walls may be flat, concave Secondary Sculpture The secondary sculpture (Microrelief) is formed by the deposition of cuticle ove r the outer wall or due to secondary wall thickenings, often shrinking and colla psing in desiccated cells. It may be smooth, striate (Figure 7.6-C), reticulate (Figure 7.6-B) or (Figure 7.6-B) or convex. micro-papillate (verrucose). All members of Urticales have curved trichomes with silicified cuticular striations at the base, and micro-papillations on the tric home body. This single character of trichomes allows for precise circumscription of the order Urticales (Barthlott, 1981). Loasiflorae is circumscribed by unice llular irregularly hooked trichomes. Secondary wall thick-enings are always of a high taxonomic significance. In Orchidaceae, for example Taxonomic Evidence 161 longitudinal striations caused by underlying secondary thickenings are restricte d to all members of Catasetinae. Tertiary sculpture is generally lacking from seeds. In orchidaceae, however, cer tain tribes possess epicuticular waxes on their seed-coats. Seed-coats of Jacara nda Tertiary Sculpture (Figure 7.6-E) have stellate epicuticular Tertiary sculpture is formed by epicuticular secretions such as waxes and other mucilaginous adhesive lipophilic substances and shows a variety of patterns. Sec ondary and tertiary sculpturing are mutually exclusive as the presence of waxes would invariably mask the cuticle; the cuticle would be vis-ible only if there a re no wax deposits. Winteraceae have a particular type and distribution of wax-l ike secretions (alveolar material not soluble in lipid solvents) on their stomat a, similar to gymnosperms, and sculpture known as star scales , a feature characteristic of this genus. Many membe rs of Aizoaceae are characterized by seed-coat with epicuticular secretions form ing long upright rodlets and small rodlets lying on the cell surface. absent in all other angiosperms. Cactaceae is a huge family commonly divided into three subfamilies, of which Cac toideae includes 90 per cent species but its classification is difficult because of uniform floral characters, pollen morphology and plasticity. Barthlott and V oit (1979) analyzed 1050 species and 230 genera by SEM for seed coat structure i n the family Cactaceae. The simple unspecialized testa of Pereskoideae supports its ancestral position. Opuntioideae has a unique seed with a hard aril, thus co nfirming its isolated position, also indicated by pollen morphology. Cactoideae shows complex diversity, con-firming its advanced position and subtribes have be en recognized based on seed-coat structure, each subtribe possessing distinctive features. Thus the genus Astrophytum has been transferred from Notocactinae to In monocots orientation and pattern of epicuticular waxes seem to provide a new taxonomic character of high systematic signifi-cance. Four types of wax patterns and crys-talloids have been distinguished (Barthlott and Froelich, 1983): 1. Smooth wax layers in the form of thin films, common in angiosperms.

2. Non-oriented wax crystalloids in the form of rodlets or platelets with no regular pattern. These are common in dicots and Lilianae groups of monocots. Cactinae. 3. Strelitzia wax-type with massive Orchidaceae is another large family with complicated phylogenetic affinities. Mi nute dust seeds show microstructural diversity of the seed-coat. Studies of over 1 000 species (Barthlott, 1981) have helped in better subdivision into subfamilies and tribes. Barthlott also supports the merger of Cypripediaceae with Orchidace ae, a suggestion incorporated in several recent classification (Judd et al., 200 2; APG II, 2003; Thorne, 2003; Stevens, 2003). compound wax projections composed of rodlet-like subunits that form massive comp ound plates around the stomata. This wax type is found in Zingiberanae, Commelin anae, and Arecanae. It is also found in Velloziales, Bromeliales, and Typhales, which further differ from other Lilianae in a starchy endosperm. 4. Convallaria wax-type with small wax platelets arranged in parallel rows, which cross the stomata at right angle and form a close circle around each polar end of the stoma, similar to the lines of an electromagnetic field. This type is restricted to Lilianae only. Ultrastructure Ultrastructure studies of angiosperms have provided valuable taxonomic informati on from phloem tissue, mainly sieve tube elements. Besides this, information has also come from studies of seeds. 162 Plant Systematics Protein P III P I t PI P cs C C -, P PV ~ c %Tf* PV c'f PI cf P VI c v C: c

cf cfs x p =stral t mplex Steps of evolution Starch within subtypes OOQ S ' Some interactions "? between P- and S-subtypes Po15 o Figure 7.7 Various forms of sieve-tube plastids and their possible evolution (Af ter Behnke and Barthlott, 1983). Sieve-tube plastids (i) PI-subtype. The plastids contain Studies on sieve-tube plastids were first initiated by Behnke (1965) in the fami ly Dioscoreaceae. Since then, nearly all angiosperm families have been investiga ted for the taxonomic significance of these plas-tids. All sieve-element plastid s contain starch grains differing in number, size and shape. The protein accumul ates in specific plastids single crystalloids of different sizes and shapes and/or irregularly arranged fi laments. This subtype is thought to be the most primitive in flowering plants, m ainly Magnoliales, Laurales and Aristolochiales. (ii) PII-subtype. This subtype contains in the form of crystalloids and filaments. several cuneate crystalloids oriented towards the centre of the plastid. All inv estigated monocots contain this subtype. It is significant to note that only mem bers of dicots with this subtype, Asarum, and Saruma of Aristolochiaceae are wid ely regarded among the most primitive members of dicots, a possible link between monocots and dicots. Thus two types of plastids are distinguished: P-type which accumulate proteins a nd S-type which do not accumulate proteins. Starch accumulation is of no primary importance in classification, since it may be present or absent in both types o f plastids. P-type plas-tids are further divided into six subtypes (Behnke and B arthlott, 1983) (Figure 7.7): Taxonomic Evidence 163 ders Caryophyllineae, Chenopodineae and Phytolaccineae, earlier established by F riedrich (1956). Whereas Takhtajan had recognized these three suborders in his 1 983 revision, in his final revision of his classification, he merged Phytolaccin eae with Caryophyllineae, thus recognizing only two suborders Caryophyllineae an d Chenopodineae. Of the three orders recognized in Caryophyllidae of Takhtajan, only Caryophyllales contains PIII-subtype plastids while the other two orders, P olygonales and Plumbaginales, contain S-type plastids. Behnke (1977) as such, ad vocated retention of only Caryophyllales under Caryophyllidae and removal of the other two orders to subclass Rosidae whose members also contain S-type plastids . The suggestion was not accepted by Takhtajan (1987) and Cronquist (1988), who

retained all the three orders under Caryophyllidae. Takhtajan, however, placed t he three orders under separate superorders (Cronquist does not recognize superor ders). On further intensive studies of plastids within the group, Behnke (1997) advocated the removal of genus Sarcobatus from the family Chenopodiaceae on the basis of the presence of PIIIcf plastids and absence of PIIIf, which are charact eristic of family Chenopodiaceae. He places the genus in an independFigure 7.8 Different forms of P III subtype sieve-element plastids all with ring like bundle of filaments (F). 19 and 20: P II Ia with globular crys-talloid (C); 21: P IIIb with polygo-nal crystalloid; 22: P IIIc without crystalloid. (From Behnke, 1977). (iii) PIII-subtype. This subtype contains a ring-shaped bundle of filaments. PIII-subtype is confined to Centrospermae (Cary ophyllales) and the removal of Bataceae and Gyrostemonaceae has been supported b y the absence of this subtype in these families. Further, forms are recognized b ased on the presence or absence of crystalloids ent family, Sarcobataceae. (iv) PIV-subtype. The plastid contains a few polygonal crystalloids of variable size. This subtype is restricted to the order Fabales. (v) PV-subtype. The plastid contains (Figure 7.8) into PIIIa (globular crysmany crystalloids of different sizes and shapes. This subtype is found in the or der Ericales and family Rhizophoratalloid), PIIIb (hexagonal crystalloid) and PIIIc (without crystalloid). Based o n the distribution of these forms, Behnke (1976) proposed division of the order into three family-groups which exactly correspond to the three suborceae. (vi) PVI-subtype. The plastid contains a single circular crystalloid. This sub-type is found in family Buxaceae. 164 Plant Systematics Dilated Cisternae sieve-tube elements have also been reported in Boraginaceae, and may prove usefu l. Dilated Cisternae (DC) were first described by Bonnett and Newcomb (1965) as dil ated sections of endoplasmic reticulum in the root cells of Raphanus sativus. Or iginally found in Brassicaceae and Capparaceae, DC have now been found in severa l other families of angiosperms but are concentrated in the or-der Capparales (B rassicaceae and Capparaceae) and form a part of the charac-ter syndrome of this order. The DC may be utricular, irregular or vacuole-like in form with filamento us, tubular or granular contents. They have been proposed to be func-tionally as sociated with glucosinolates and myrosin cells found in this order. Non-phloematic TEM Characters Protein bodies in seeds through TEM, SEM and dispersive X-ray techniques have de monstrated their significance if qualitative and quantitative aspects are both t aken into account. Similarly, SEM studies of starch grains are also potential so urces of information of taxonomic significance. Phloem (p-) proteins CHROMOSOMES Chromosomes are the carriers of genetic information and as such have a considera ble significance in evolutionary studies. Increased knowledge about chromosomes and their behaviour has largely been responsible for extensive biosystematic stu dies and development of the biological species concept. During the first quarter of the twentieth century, chromosomal data were relatively sparse. Such informa tion has markedly increased over the last few decades, however, with ample usefu l information coming from studies of the banding pattern. Three types of chromos omal information have been of significance in Systematics. P-proteins are found only in sieve elements of angiosperms and occur in the form of fila-ments or tubules. These assemble into large discrete bodies, and are no t dissolved dur-ing maturation of the sieve-element, unlike single membrane orga

nelles. The composi-tion and three-dimensional arrangement of these proteins exh ibit taxonomic specific-ity. They are dispersed over the entire cell as the cell matures but in some dicots, a single non-dispersive (crystalline) body of vario us shapes may be found in addition to dispersive ones. Crystalline bodies are ab sent in monocots. Their shape is often spe-cific and thus of taxonomic importanc e. Globular crystalline bodies are found in Malvales and Urticales. Fabanae, whi ch is characterized by PIV-subtype plastids, has spindle-shaped crystalline bodi es in the family Fabaceae. The feature has supported the transfer of Swartzia to Fabaceae. Chromosomal number Extensive records of chromosome numbers are available in the works of Darlington and Janaki-Amal (1945), Darlington and Wylie Nuclear inclusions (1955), Federov (1969) and Lve et al. (1977). The International Association of Pl ant Taxonomy (IAPT) has also been publishing an Index to Plant Chromosome Number s in its series Regnum Degetabile. Between 1967 and 1977, the series published 9 volumes mostly forming annual lists of chromosome numbers. An updated server of the Missouri Botanical Garden maintains the records of chromosome numbers and c an be queried for online information about plant species. The chromosome counts are usually reported as diploid number (2n) from mitosis of sporophytic tissue b ut when based on mitosis in Nuclear inclusions in the form of protein crys-tals occur in phloem- and Ray par enchyma, primarily in the families of Asteridae. Five types of crystals have bee n differentiated. Structural differences are significant for classification with in Scrophulariaceae and Lamiaceae (Speta, 1979). Protein crystals in Taxonomic Evidence 165 gametophytic tissue or on meiosis studies, counts are reported as haploid (n). T he gametophytic chromosome number of diploid species is designated as base-numbe r (x). In diploid species as such n = x, whereas in polyploid species n is in mu ltiples of x. A hexaploid species with 2n = 42 will

thus have n = 21, n = 3x and 2n = 6x. The chromosome number in angiosperms exhibits considerable variation. The lowest number (n = 2) is recorded in Haplopappus gracilis (Asteraceae) and the highest (n = 132) in Poa littoroa (Poaceae). The alga Spirogyra cylindrica also contain s n = 2, whereas the record of the highest chromosome number (n = 630) is found in Ophioglossum reticulatum (Pteridophytes). Such a range of variation, (n = 2 t o n = 132), however, within nearly a quarter a million species of angiosperms, m ay not be very significant in taxonomic delimitation, but there have been instan ces of the isolated role of studies on chromo~/ , somes. Raven (1975) provided a review of Figure 7.9 Mitotic chromosomes of Trade s-chromosome numbers at the family level in angiosperms. He concluded that the o rigi-nal base-number for angiosperms is x = 7 and that comparisons at the family level are valid only when the base-number (and not n or 2n) is used. The family Ranunculaceae is dominated by genera with large chromo-somes (and x = 8). The t wo genera Thalictrum and Aquilegia originally placed in two separate subfamilies o r tribes (and even two separate families Ranunculaceae and Helleboraceae by Hutc hinson, 1959, 1973 along with other achene bearing and follicle bearing genera, respectively) are distinct in having small chromosomes (and x = 7) and as such hav e been segregated into a distinct tribe. The genus Paeonia with very large chrom osomes (and x = 5) has been separated into a distantly related family Paeoniacea

e, a placement which has been supported by morphological, anatomical and embryol ogi-cal data. Significant records in other families include Rosaceae with x = 17 in subfamily Pomoideae, whereas other sub-families have x = 7, 8 or 9. In Poace ae similarly subfamily Bambusoideae has x = 12, whereas Pooideae has x = 7. cantia spathacea (2n=12) with sister chromatids and centromere. Spartina was for long placed in the tribe Chlorideae (x = 10) although its chrom osomes (x= 7) were at variance. Marchant (1968) showed the genus to have, in fac t x = 10, thus securing placement within Chlorideae. The classical study of the genus Crepis (Babcock, 1947) based separation from th e closely related genera on chromosomal number and morphology. This led to the s eparation of the genus Youngia and merger of Pterotheca with Crepis. Similarly, in the genus Mentha which has small, structurally uniform chromosomes, the chrom osome numbers provide strong support for subdivision into sections Audibertia (x = 9), Pulegium (x = 10), Preslia (x = 18) and M entha (x = 12). The duplication of chromosome numbers leading to polyploidy may prove to be of t axonomic significance. The grass genus Vulpia contains diploid (2n = 14), tetrap loid (2n =28) and hexaploid (2n = 42) species. The genus is divided into five se ctions, of which three 166 Plant Systematics

Figure 7.10 Ideogram of the somatic complement of Allium ampeloprasm. Of the 32 somatic chromosomes, 8 show secondary constriction (courtesy Prof. R. N. Gohil). contain only diploids, one diploids and tetraploids and one all three levels of ploidy. It is presumed that tetraploid and hexaploid species of Vulpia arose fro m diploid progeni-tors. The duplication of chromosome num-ber of a diploid speci es may form a tetraploid (autopolyploid). Such a polyploid, however, does not sh ow any or at most may show minor differences from the diploid species, and is ra rely recognized as an independent taxonomic entity. The hybrid between two diplo id species contains one genome from either parent and thus, generally doesn t surv ive because of failure of chromosomal pairing during meiosis. Hybridization foll owed by duplication of chromosomes establishes a tetraploid (allopolyploid; amph iploid) with normal pairing as both genomes are in pairs. Such a tetraploid hybr id with distinct characteristics may be would exhibit the problem of pairing at meiosis but the hybridization followed b y du-plication leading to hexaploidy can form a perfectly normal independent spe cies. Such facts have led to the detection of hybrids or confirmation of suspect ed hybrids. Senecio (Asteraceae) includes the diploid S. squalidus (2n = 20), th e tetraploid S. vulgaris (2n = 40) and the hexaploid S. cambrensis (2n = 60). Th e last is intermediate in morphology between the first two and is found in the a rea where these two grow. Additionally, sterile triploid hybrids between two spe cies have been reported. It seems clear that S. cambrensis is an allohexaploid b etween the other two species (Stace, 1989). Similarly, based on chromosome numbe r and karyotype, Owenby (1950) concluded that Tragopogon mirus (2n = 24), a tetr aploid species arose as an amphiploid between two diploid species, T. dubius and T. porrfiolius recognized as an independent species. A triploid hybrid between a diploid species and a tetraploid species may, simila rly, not survive as genome from the diploid parent (2n = 12). Whereas a species generally shows a single chromosome number, certain Taxonomic Evidence 167 populations or infraspecific taxa (subspecies, variety, forma) may sometimes sho

w a different chromosome number (or even differ-ent chromosomal morphology). Suc h popula-tions or infraspecific taxa constitute cytotypes. similarity. This was supported by the distinctive bimodal karyotype of Agavaceae con-sisting of 5 large chromosomes and 25 small ones. Rudall et al. (1997) advo cated the transfer of Hosta (placed in Hostaceae; Hesperocallidaceae by Thorne, 1999), Camassia and Chlorogalum (both placed under Liliaceae by Hutchinson, 1973 ; Hyacinthaceae by Thorne, 1999) to family Agavaceae on the basis of possession of bimodal karyotype, a suggestion incorporated by Judd et al. (2002) and Thorne (2003). Rousi (1973), from his studies on the genus Leontodon, showed that data on the basic number, chromosome length, centromeric position and the occurrence of satellites provide evidence for the relegation of the former genus Thrincia (x = 4) as a section of subgenus Apargia along with section Asterothrix (x = 4, 7). The subgenus Leontodon is distinct with x = 6 or 7, and a different Chromosomal structure Chromosomes show considerable variation in size, position of centromere (Figure 7.9) and presence of secondary constriction. The chromosomes are commonly differ entiated as metacentric (with centromere in middle), submetacentric (away from m iddle), acro-centric (near the end) or telocentric (at the end). The chromosomes are also character-ized by their size. In addition the occurrence and position of secondary constriction, which demarcates a satellite is important in chromoso mal identification and charac-terization. The identification of satellites is of ten difficult, and especially when the sec-ondary constriction is very long, a s atellite may be counted as a distinct chromosome. This situation has often led t o erroneous chromosome counts. The structure of the chromosome set (genome) in a species is termed karyotype and is commonly dia-grammatically represented in th e form of an ideogram (Figure 7.10) or karyogram. An analysis of a large number of studies has led to the conclusion that a symmetrical karyotype (chromosomes e ssentially simi-lar and mainly metacentric) is primitive and an asymmetric karyo type (different types of chromosomes in a genome) advanced, the latter commonly found in plants with spe-cialized morphological features, such as Delchromosome morphology. Cyperaceae and Juncaceae were earlier placed far apart due to distinct floral st ructure. Both families have small chromosomes without distinctive centromeres, t he latter may be diffuse or non-localized. These families as such are now consid ered to be closely related. Such chromosomes (holocentric chromosomes) do not de pend on a discrete centromere for meiosis and mitosis and may undergo fragmentat ion with no deleterious effect. This may result in variable chromophinium and Aconitum. somal counts. In the Luzula spicata group, chromosomal counts are reported to be 2n = 12, 14, and 24. Interestingly, the total chromosomal volume is the same an d the higher chromosome number is the result of fragmentation (agmatoploidy) of these holocentric chromosomes. Different chromosome numbers may often occur in d ifferent cells of the same root-tip (mixoploidy). The occurrence of accessory ch romosomes (known as B-chromosomes) in higher plants generally does not have a si gnificant effect on morphology and, thus, is of little taxonomic importance. B-c hromosomes in bryophytes, contrarily, are very small (termed mchromosomes) and o ften highly diagnostic. An interesting example of utilization of chromosomal information is family Agava ceae. The family contains about 16 genera such as Agave (and others formerly pla ced in Amaryllidaceae due to inferior ovary) and Yucca (and others formerly plac ed in Liliaceae due to superior ovary). These genera were shifted and brought in to Agavaceae on the basis of great overall 168 Plant Systematics In recent years, considerable break-through has been achieved in the study of ba nding patterns of chromosomes using Giemsa and fluorochrome stains. Already diff erent techniques such as C-banding, G-banding, Q-banding and Hy-banding are in u se, and help in clearly distinguishing the heterochromatic and euchromatic regio ns. C-banding is very useful in indicat-ing the position of centromeres in cases where they cannot be identified by conven-

chromosomes without chiasmata so that chromosomes fall apart before metaphase (d esynapsis). In extreme cases, the entire genome may fail to pair. The genome ana lysis of suspected hybrids has helped in establishing the parentage of several p olyploid species. tional staining. A diploid hybrid between two species generally exhibits failure of meiotic pairi ng due to non-homology of genomes resulting in hybrid sterility, but when hybrid ization is followed by duplication of chromosomes to form a tetraploid hybrid, t he latter shows normal pairing between the two genomes derived from the same par ent and is generally fertile. A triploid hybrid may, similarly, be sterile but a hexaploid one fertile. Genome analysis has confirmed that the hexaploid Senecio cambrensis is allohexaploid between tetraploid S. vulgaris and diploid S. squal idus. Similarly the tetraploid Tragopogon mirus is the result of hybridization b etween the two diploid species T. dubius and T. porrifolius. The most significan t case, however, is the common bread wheat Triticum aestivum, a hexaploid with A ABBDD The technique of silver-staining has been developed to highlight NOR (nucleolar orga-nizing region). An interesting study of the chromosomes of top onion (vario usly recog-nized as Allium cepa var. viviparum or A. fistulosum var. proliferum) as also those of A. Cepa and A. fistulosum was done by Schubert, Ohle and Hanel t (1983). By Giemsa banding pattern and silver-staining studies, they concluded that some chromosomes of top onion resemble A. cepa and others resemble A. fistu losum. Of the two satellites, one resembles either species. Top onion is as such a pseudodiploid with no homologous pair. The study confirmed that top onion is genome. Genome analyses have confirmed a hybrid between the two aforesaid parents, that genome A is derived from the di ploid and thus would be better known as A. x T. monococcum, B from Aegilops spel toides, proliferum (Moench) Schrad. (based on Cepa both genomes being represente d in the proliferum Moench), and not as a variety of tetraploid T. dicoccum. Gen ome D is derived either species. Interestingly, the top onion from the diploid A egilops tauschii. owes its existence to the bulbils, which are produced in place of an inflorescence and ensure the multiplication of the hybrid, which is otherwise sterile. CHEMOTAXONOMY Chromosomal behaviour Chemotaxonomy of plants is an expanding field of study and seeks to utilize chem ical information to improve upon the classifica- tion of plants. Chemical eviden ce has, in fact, been used ever since man first began to classify plants as edib le and inedible, obviously based on their chemical differ-ences. Chemical inform ation about medicinal plants in herbals published nearly five centuries back was concerned with localization and application of physiologicallyactive secondary metabolites such as saponins and alkaloids. Knowledge about chemistry of plants greatly increased during the The fertility of a plant is highly dependent on the ability of meiotic chromosom es to pair (synapsis) and their subsequent sepa-ration. The meiotic behaviour of chromo-somes enables comparison between genomes to detect the degree of homolog y, especially when they are a result of hybrid-ization. A greater degree of geno mic non-homology results in either failure of pairing (asynapsis) or a loose pai ring of Taxonomic Evidence 169 eighteenth and nineteenth centuries. The greatest interest has been generated ov er the last 40 years, however, with the devel-opment of improved techniques for studying biological molecules, especially proteins and nucleic acids. In recent years, interest has focused on the study of allelochemy and re-alization of the concept that the animal king-dom and the plant kingdom have experienced a chemic al coevolution. Plants con-tinuously evolve new defensive chemical mechanisms to save themselves from preda-tors, and animals evolve methods to over-come these

defenses. In the process, some plant species have developed animal hor-mones, th us disturbing the hormonal levels The utilization of studies on DNA and RNA for understanding of phylogenetic rela tions has received a great boost over the last decade, meriting the establishmen t of a new field referred to as Molecular Systematics, and would be dealt separa tely after chemotaxonomy. Only proteins would be described in this section. Primary metabolites of animals if ingested. Primary metabolites include compounds, which are involved in vital metabolic pat h-ways. Most of them are universal in plants and of little taxonomic importance. Aconitic acid and citric acid, first discovered from Ac-onitum and Citrus respe ctively, participate in Krebs cycle of respiration and are found in all aerobic organisms. The same is true of the 22 or so amino acids forming proteins, and th e sugar molecules, which are involved in the Kalvin cycle of photosynthesis. The quantitative variations of these primary metabolites may, however, be of taxono mic significance sometimes. In Gilgiochloa indurata (Poaceae), alanine is the ma in amino acid in leaf extracts, proline in seed extracts and asparagine in flowe r extracts. Rosaceae is similarly rich in arginine. A large variety of chemical compounds are found in plants and quite often the bi osyn-thetic pathways producing these compounds differ in various plant groups. I n many in-stances the biosynthetic pathways corre-spond well with existing schem es of classi-fication based on morphology. In other cases, the results are at va riance, thus calling for revision of such schemes. The natural chemical constitu ents are conveniently di-vided as under: Micromolecules: Compounds with low molecular weight (less than 1000). Primary metabolites: Compounds involved in vital metabolic pathways citric acid, aconitic acid, protein amino Secondary metabolites Secondary metabolites perform non-vital functions and are less widespread in pla nts as compared to primary metabolites. These are generally the by-products of m etabolism. They were earlier considered to be waste products, having no importan t role. Recently, however, it was realized that they are important in chemical d efense against predators, pathogens, allelopathic agents and also help in pollin ation and dispersal (Swain, 1977). Gershenzon and Mabry (1983) have provided a c omprehensive review of the significance of secondary metabolites in higher class ification of angiosperms. The following major categories of secondary metabolite s are of taxonomic significance: acids, etc. Secondary metabolites: Compounds which are the by-products of metabo-lism and of ten perform non-vital func-tions non-protein amino acids, phe-nolic compounds, al kaloids, glucosino-lates, terpenes, etc. Macromolecules: Compounds with high molecular weight (1000 or more). Non-semantide macromolecules: Compounds not involved in information transfer starc hes, celluloses, etc. Semantides: Information carrying mol-ecules DNA, RNA and pro teins. 170 Plant Systematics

OH

`} OH `O` OH OH

OH Phenol Catechol Hydroquinone OH OH OH ~ OH p gel OH OH Phloroglucinol *HO Vl'\ -HO's N 'COO~~ O O conmw in HOOC '\/' COOH H Betanidin Fhwonoid nucleus Figure 7.11 Structure of important phenolic molecules and a betalain (Betanidin) . (*indicates the position of sugar.) Non-protein amino acids localized in vacuoles in cell, and are found in combination with sugars as glyco side. (Simple phenolics can be tested by extraction with ethanol. Take 5-6 gm of chopp ed leaf tissue in a beaker and add 30 ml of 70% ethanol; heat over water bath at 60-70 degree centigrade for 20 minutes; filter and concentrate filtrate over wa ter bath till about 0.5 ml is left; load the sample on Whatman paper (No. 1) usi ng BAW: Butanol, Acetic acid and water in ratio of 4:1:5; run chromatogram, dry and observe under UV light; spray with mixture (1:1) of 1% Ferric chloride and 1 % Potassium ferricyanide and calculate Rf value). These are widely distributed in the plant kingdom; common examples being catechol, hydroquinone, phlo- roglucinol and pyragallol. Coumarins, a group of natural phenolics, have a characteristic smell. The crushed leaves of Anthoxanthum odoratum can thus be i dentified by this char-acteristic odour. More than 300 coumarins have been repor ted from nearly 80 families of plants. They are a group of lactones formed by ri ng closure of hydroxycinnamic acid. A large number of amino acids not associated with proteins are known (more than 300 or so). Their distribution is not universal but specific to certain groups a nd, as such, holds promise for taxonomic significance. Lathyrine is, thus, known only from Lathyrus. Canavanine occurs only in Fabaceae and is shown (Bell, 1971 ) to be a protection against insect larvae. These amino acids are usually concen trated in storage roots and, as such, root extracts are generally used for their study. Phenolics Phenolic compounds form a loose class of compounds, based upon a phenol (C6 H5 O H). Simple phenolics are made of a single ring and differ in position and number of OH groups. These are water soluble and are Taxonomic Evidence 171 OH OCH3

00113 *HO //'V\ *HO ~,/~ ' *HO ~:~ Y\

II ~/ \% OH* OH N/ \%~~ OH* OH V OH* OH OH* OH* OH* f~.,alli[llll Pa-e-011 U1L1 P:elryruoilifhii O CHD OH O CH3 *HO ~f ~VA *HO ~~ ~Y /~OH *HO fi/ Y />- OH /~ OH ~r/~I OH* OCH3 \ i` /~ OH* OH S OH* OH OH OH* OH* iti-lalvi ni D'elpidliiil n P:ehuuilui Figure 7.12 Structure of anthocyanin forming molecules, differing in right three positions, middle position absent in Cyanidin, Paeonidin and Pelargonidin, but having OH in Malvid in, Delphinidin and Petunidin. Cyanidin has OH at both other positions, Paeonidi

n has one replaced by OCH3, Pelargonin upper is missing, Malvidin has both repla ced by OCH3, Delphinidin has OH at both and Petunidin one having OH and other OC H3. Lignin is a highly branched polymer of beakers to dryness on water bath uncovere d; to each three simple phenolic alcohols. Whereas add 1% methanolic HCl; load each sample on two circular whatman (No. 1) filter paper, load spot in gymnosperm lignin is composed of coniferyl centre using Forestall solvent: HCl, Acetic acid and alcohol subunits, the angiosperm lignin is water (3:10:30); mark spots visually; observe one a mixture of coniferyl and sinapyl alcohol set of chromatograms of A and B under UV; expose subunits. The alcohols are oxidized to free radicals by peroxidase enzyme, and t he freed radicals react to form lignin second set to ammonia vapours by rotating discs over open mouth of NH3 bottle an d observe under UV. Phenyl propanoids can be detected similarly except that extraction is done using diethyl ether instead of ethyl estate, only organic layer is retained Flavonoids, the more extensively studied compounds, are based on a flavonoid nuc leus consistin g of two benzene ringg s 'Dined b ya C3 open or closed structure (Figure 7.11). (Presence of flavonoids can be detected as fol-lows: Finely chop 5 gm of flower petals or tepals in beaker; add 20 ml of 2N HCl, cover with alumi nium foil, and heat at 80-90 degree centigrade for 30-40 minutes in water bath; filter and extract filtrate with 15-20 ml ethyl acetate in separating funnel, sh ake and allow solvent to evaporate; two layers are formed, upper organic and low er inorganic aqueous layer (mostly anthocyanins); collect them in two separate b eakers; label beaker with upper organic layer as B; if aqueous layer is coloured heat for 5-10 minutes to expel ethyl acetate; put back in separating funnel, ad d 2-4 ml of amyl alcohol, shake, transfer upper organic layer to beaker and mark it as A; heat both and the solvent use for chromatography is BAW (Butanol : Acetic acid: Distilled water-4:1:5). Common examples are flavonols (mainly colourless and commonly occurring as co-pi gments, yield bright yellow spot in chromatogram after acid hydrolysis), flavones (similar, yield dull brown spots), gly coflavones, biflavonyls (similar, yield dull absorbing spots on BAW), isoflavone s (colourless, often found in roots of legumes), flavanones (colourless, occur i n leaves, citrus fruits, yield red colour with HCl), chalcones or aurones (usual ly occur in yellow flowers, yield red colour with NH3), Anthocyanins (red, blue coloured water soluble) 172 Plant Systematics

Table 7.1 Classification of Betalain containing families and potential relatives . Structural Chemical Compromise Compromise classification classification classification-I classification-II (Engler & Prantl)

(Mabry, 1963) (Alston & Turner, 1971) (Mabry, 1976) Centrospermae Chenopodiales Caryophyllidae Caryphyllales Chenopodiaceae Chenopodiaceae Chenopodiales Chenopodinea Amaranthaceae Amaranthaceae Chenopodiaceae Chenopodiaceae Nyctaginaceae Nyctaginaceae Amaranthaceae Amaranthaceae Phytolaccaceae Phytolaccaceae Nyctaginaceae Nyctaginaceae Gyrostemonaceae Aizoaceae Phytolaccaceae Phytolaccaceae Aizoaceae Portulacaceae Aizoaceae Aizoaceae Portulacaceae Basellaceae Portulaceae Portulaceae Basellaceae Cactaceae Basellaceae Basellaceae Caryophyllaceae Didiereaceae Cactaceae Cactaceae Molluginaceae Didiereaceae Didiereaceae Caryophyllales Cactales Caryophyllaceae Caryophyllales Caryophyllineae Cactaceae Molluginaceae Caryophyllaceae Caryophyllaceae Molluginaceae Molluginaceae Sapindales

Didiereaceae and leucoanthocyanins (mainly colourless, mainly in heartwood and leaves of tree s, yield and having quite distinct metabolic pathways of synthesis. However, they carry t he same functions as anthocyanins. Betalains are mutually exclusive with anthocy anins, and concentrated in the traditional group Centrospermae of Engler and Pra ntl, now recognized as order Caryophyllales. Of the nine families which contain betalains, seven were included in Centrospermae, Cactaceae placed in Cactales or Opuntiales and the ninth was placed in Sapindales. Traditional Centrospermae al so included Gyrostemonaceae, Caryophyllaceae and Molluginaceae which lack betala ins and contain anthocyanins instead. Mabry et al., (1963) on the basis of separ ate structure and metabolic pathways, suggested the placement of only betalain-c ontaining families in Centrosperanthocyanins). Anthocyanins and Anthoxanthins are important pigments in the cell sap of petals providing red, blue (anthocyanins), and yel-low (anthoxanthins) colours in a lar ge num-ber of families of angiosperms. They are formed by anthocyanadins combini ng with different sugars at different places. Six main categories (Figure 7.12) of anthocyanin form-ing molecules are recognized providing dif-ferent colours: C yanidin-magenta; Pelargonidin-orange-red; Delphinidin-purple, blue, mauve; Petun idin-purplish; Paeonidin-magenta and Malvidin-purple. These pigments are absent in some fami-lies and replaced by highly different com -pounds, betacyanins and betaxanthins (to-gether known as betalains), which cons ist of heterocyclic nitrogen-containing rings mae, thus advocating the inclusion of Cactaceae and Didiereaceae and exclusion o f Gyrostemonaceae, Caryophyllaceae and Taxonomic Evidence 173 '

HDAC

in all members of Centrospermae and thus suggested a compromise by including all families within subclass Caryophyllidae with betalain-containing families place d under the order Chenopodiales and the other two (Caryophyllaceae and Mollugina ceae) placed under Caryophyllales. Interestingly, Mabry (1976), on the basis of DNA/RNA hybridization studies, found closer affinities between these fam-ilies a nd suggested the placement of all these families under Caryophyllales with the b etalain-containing families under the sub=o HoAC Hymenophyron flabellarum FBA order Chenopodiineae and the two non-betalain families under Caryophyllineae. Th is final compromise has met with mixed response in recent years with the morphol ogical, anatomical and DNA/RNA hybridization evidence overriding the betalain ev idence. Takhtajan (1997) places only Chenopodiaceae and Amaranthaceae in Chenopo diineae, Dahlgren (1989) and Cronquist (1988) and APG II (2003) did not recogniz

e suborders, and Thorne (2007) recognising four suborders, Caryophyllaceae and M olluginaceae separated under different suborders. Figure 7.13 Two-dimensional paper chromatoIt is interesting to note that the betalains have also been reported in Basidiom ycetes (Fungi), in some cases the same substance found in both fungi and angiosp erms. The above studies on the significance of distribution of betalains in Cent rospermae bring home the fact that chemical data are useful in taxonomic realign ments when such accord with data from other fields. The significance is reduced when larger evidence from elsewhere contradicts the chemical evidence. Thus, whe reas no questions were ever asked about the removal of Gyrostemonaceae and the i nclusion of Cactaceae and Didiereaceae, there has been no agreement about the re moval of Caryophyllaceae and Molluginaceae as it goes against the evidence from morphology, anatomy, ultrastructure and DNA/RNA hybridization. This also highlig hts the danger grams of the flavonoids in two species of Hymenophyton (after Markham et. al., 1 976). Molluginaceae (Table 7.1). Whereas the in-clusion of Cactaceae and Didiereaceae was readily accepted (thus bringing all betalain-containing families in the same order Cen-trospermae), the exclusion of Caryophyl-laceae and Molluginaceae was strongly op-posed on the basis of structural data. This clash between orthodox a nd chemical tax-onomy initiated renewed interest in the of relying too much on one type of evidence. group. Studies on phenolic compounds have helped in solving some specific problems. Bat e-Smith (1958) studied five phenolic characters of different sections in the Behnke and Turner (1971), on the basis of ultrastructure studies, reported P-III plastids 174 Plant Systematics and H. flabellatum, and that there is no o CH3 justification for their merger. 7H N~ H3C` Similar studies in the genus Baptisia (Fabaceae) by Alston and Turner (1963) hav e been very useful in the detection of hybrid-ization. Each species of the genus has a distinctive spectrum of flavonoids, and the hybrid can be easily identifi ed by the combination of flavonoid pattern of both parental o~`NI N CK3 Lupinine Caffeine H3co l a species in the suspected hybrid. /~ It is interesting to note that the ten taxa recognized (four parental and six hy brid), could not be differentiated on the basis of morphological or biochemical characters alone, but a combination of both enabled a complete separation. The f lavonoids in these studies were extracted from either flower or leaf. i N CH3 H3 Ca// N OCH3 Nicotine OCH3

Papaverine CH= CH 2 Ha Alkaloids O H3ca N Alkaloids are organic nitrogen-containing bases, usually with a heterocyclic rin g of some kind. They form one of the largest class of secondary metabolites, wit h nearly 10,000 different types reported. They are insoluble in water but solubl e in organic solvents, but their salts are soluble in water and insoluble in org anic solvents. Their distribution is restricted to some 20% of angiosperms. They are mostly present in storage tissues, seeds, fruits and roots. They act as che mical defence of plants against herbivory, and NCH 0-c N H2COH Quinine Atropine Figure 7.14 Main Examples of alkaloids found in the plant kingdom. The distri-bution of some is highly specific. genus Iris. The chemical evidence supported the division into various sections, but I. flavissima, originally placed in the section Pogoniris resembled species of the section Regelia on the basis of phenolic charac-teristics. Chromosomal ev idence also allelopathic reactions between plants. supported this transfer. Alkaloids are generally classified on the basis of predominant ring system prese nt in the molecule. They are synthesized from a few common amino acids like tyro sine, tryptophan, ornithine, argenine and lysine. Tobacco alkaloid Nicotine (Nic otiana) is synthesized from nicotinic acid and caffeine (coffee beans and tea le aves) from purine. Isoquinolene alkaloids morphine, codeine and papaverine are f ound in opium poppy (Papaver somniferum). Their distribution is often specific a nd thus taxonomically significant (Figure 7.14). Conalium is the simplest known alkaloid found in Conium maculatum (Apiaceae). Alkaloids are present in specialized parts of plant. Higher nicotine content The technique of two-directional paper chromatography, which brings about a more pronounced separation of flavonoids, has proved very useful in taxonomic studie s. Hymenophyton (Bryophytes) was considered by some researchers to be a monotypi c genus, but by others to include two species. Markham et al., (1976) on the bas is of rapid flavonoid extraction, two-dimensional chromatographic analysis and i dentification (Figure 7.13) concluded that the genus contains two distinct species, H. leptodotum Taxonomic Evidence 175 concentrate on water bath at 100 degree centigrade to one fourth volume; add NH4 OH or NH3 dropwise till pH rises to 9-10 and precipitate alkaloids; centrifuge a t 2000 rpm for 5-10 minutes; wash with 1% NH4OH; dissolve in few drops of ethano l; repeat above steps with tea, cigarette and coffee as control; take circular W hatman paper No. 1, immerse l S G I u c o syl R

G M e' ~ H SO2 A mustard oil g I u c o id in 5% Sodium dihydrogen citrate for 5 minutes, dry in oven; make pore in centre of disc, mark circle of 1 cm, load extract and run chromatogram using solvent (b utanol 217 ml, water 32 m, acetic acid 1.2 gm); and dry; on one chromatogram spr ay Iodoplatinate reagent (10 ml 5% Platinum chloride, add 240 ml 2% KI, make fin al volume to 500 ml with water) and to other Dragendroff s reagent (make two solut ions, one 0.6 gm Bismith subnitrate + 2 ml HCl+ 1m water; second 6 gm KI dissolv ed in 10 ml water; mix two solutions and filter, add 7 ml conc. HCl and 15 ml wa ter, dilute to 400 ml with water); dry chromatograms and observe colour under UV ; calculate Rf value. For Solvent extraction method take 2 gm of chopped plant m aterial in a beaker containing 20 ml methanol; heat on water bath for 30 minutes at 70 degree centigrade, cover-ing beaker with aluminium foil to prevent evapor ation of methanol; filter, suspend residue in 2 ml methanol, add 12 ml HCl to br eak cell wall and release cell sap, shake and filter; wash residue with 8 ml of 1% HCl, and filter; collect all three filtrate, add NH4OH or NH3 and adjust pH t o 10-11; add precipitate formed in separating funnel, and add chloroform, shake and collect lower organic layer; repeat procedure with chloroform, and label all collected solution as A; to acqeous layer in separating funnel add a pinch of s odium sulphate and 20 ml of mixture of ethanol and chloroform (2: 3 ratio); coll ect and label organic layer as B; put solutions A and B in test tubes heat in wa ter bath till volume is reduced to about 2 ml; add 2 ml of 1% HCl and 2 ml of ch loroform in both test tubes, two layers are formed; collect upper layer using dr opper; if turbid mustard oil R N C S _ ~ S H3C G K, N O S02 O B glucocapparine g I u c o s i n o I ate S. OH2OH Glucosyl

O H O O Spa C Figure 7.15 Glucosinolates. A: Mustard oil glucosid; B: Glucocapparine; C: General structure of Gluco-sinolates. is found in only older leaves. In Datura alka-loids occur only in seeds. Alkaloi ds are more widely distributed in dicots as compared to monocots. Some of them a re of medicinal importance at low concentration, but toxic at high concentration . Some such as lycotonine (Delphinium), scopolamine (Datura), and atropine (Atro pa) cause poison-ing of livestock. Alkaloids are generally ex-tracted from plant s in weak acid alcoholic solution and precipitated by NH3. Their pres-ence is te sted through chromatic method, CH 20 H O H 2 H 0

H OH H/ 0 O 0 H / OH HON H N am Y. ,MW O H OH and quantified by solvent extraction method. For chromatographic method, take 10 gm of chopped plant tissue in beaker contain ing 30 ml of 10% acetic acid in ethanol; heat in water bath at 60 degree centigr ade for 40 minutes and filter; Figure 7.16 Structure of Amygdalin, cyanogenic glycosides found in seeds of almond. 176 Plant Systematics filter through a piece of cotton film; put a drop of this extracted sample of al kaloids in a groove tile, add 2-3 drops of different reagents (1, Mayer s reagent: dissolve 1.63 gm HgCl2 in 60 ml of dis-tilled water; separately dissolve 5 gm K I in 10 ml of distilled water; mix two and distilled water to make 100 ml. 2, Wa gner s reagent: dissolve 1.27 gm iodine and 2 gm KI in 10 ml of distilled water, m ake volume to 100 ml. 3, Dragendroff s reagent, de-scribed above. 4, Scheilden s rea gent: dissolve 25 gm sodium tungstate and 17 gm of disodium hydrogen phosphate ( Na2HPO4) in 125 ml of distilled water; adjust pH to 4 or 5 by using conc. HNO3 a nd 0 OH Menthol Camphor 0 note the change of colour and precipitate formation.

Mears and Mabry (1971), in studies conducted on the family Fabaceae, observed th at the alkaloid hystrine occurs only in three genera Genista, Adenocarpus (both belonging to Genistae) and Ammodendron (originally placed in Sophorae). The latt er, however, lacks matrine, characteristic of Sophorae. This indicates that the transfer of the last genus also to Genistae is warranted. Families Papaveraceae and Fumariaceae are closely related. This affinity is supported by the occurrenc e of the alkaloid protopine DH ,CH3 ~ 0 '"H-:1 CHI 011, CH2 Gii10 `0o CH 3 Sweroside Amarolide gure 7. 17 oids: Menthol and Camphor, tthhee mmoononoids; Swe , a seco-iridoiidd frfroom Swertia,, GGeentiain both. naceae; Amarolide, a quassinoid triterpenoid derivative from Ailanthus, Simaroub aceae. Gershenzon and Mabry (1983) reported that tropane alkaloids of Solanaceae and Co nvolvulaceae are similar, suggesting a close relationship. The families are plac ed in the same order in recent systems. Papa-veraceae, earlier grouped with Cruc iferae and Capparaceae, is now removed to nearer Ranunculales on the basis of th e absence of glucosinolates and presence of benzyliso-quinolene. Nymphaeaceae an d Nelumbona-ceae differ in the sense that the former lacks benzylisoquinolene al kaloids. Benzyliso-quinolene, and the alkaloids that can be derived from it, are characteristic of Magnoliidae, as also family Rutaceae, some Rhamnaceae and gen us Croton. ae and Fumariaceae were placed in the same order, Rhoeadales. Chemical and other evidence, however, supported the placement of Cruciferae and Capparaceae in the order Capparales (on the basis of the presence of glucosinolates) and Papaverac eae and Fumariaceae in the order Papaverales or suborder Papaverineae of Ranuncula les (Thorne, 2003) (on the basis of the absence of glucosinolates and the presence of the alkaloid benzylisoquinolene). Bataceae and Gyrostemonaceae were once pla ced in Centrospermae (Caryophyllales) but subsequently removed due to the absenc e of betalains. This removal was supported by the presence of glucosinolates, wh ich are absent in Caryophyllales. Glucosinolates Glucosinolates are sulphur containing compounds found in 15 families of angiospe rms, mainly concentrated in the order Capparales (Figure 7.15). Mustard oils or isothiocyanates Cyanogenic Glycosides are hydrolytic products of glucosinolates. Orig-inally Cruciferae, Capparaceae, PapaveraceCyanogenic glycosides are phytotoxins which occur in at least 2000 plant species , of Taxonomic Evidence 177 which a number of species are used as food in some areas of the world. They are hydrolysed by various enzymes to release hydrogen cyanide, the process known as cyanogenesis, and the plants as cyanogenic plants. Cassava and sorghum are espec ially important staple foods containing cyanogenic glycosides. There are approxi mately 25 cyanogenic glycosides known. The major cyanogenic glycosides found in the edible parts of plants used for human or animal consumption include Amygdali

n (Figure 7.16; almonds, Prunus dulcis), Dhurrin (Sorghum album, S. bicolor). Li namarin (cassava, Manihot esculenta; lima beans, Phaseolus lunatus), Lotaustrali n (cassava, Manihot cartha-ginensis; lima beans, Phaseolus lunatus). Prunasin (s tone fruits, Prunus avium, P. padus, P. persica, P. macrophylla), and Taxiphylli n (bamboo shoots, Bambusa vulgaris). The po-tential toxicity of a cyanogenic pla nt depends primarily on the potential that its consump-tion will produce a conce ntration of HCN that is toxic to exposed animals or humans. Hy-drogen cyanide is released from the cyano-genic glycosides when fresh plant material is macerated as in chewing, which allows enzymes and cyanogenic glycosides to come together, releasing hydrogen cyanide. Cya-nides inhibit the oxidative processes of cells causing them to die very quickly. Because the body rapidly detoxifies cyanide, a n adult human can withstand 50-60 ppm for an hour without serious consequences. However, exposure to concentrations of 200-500 ppm for 30 minutes is usually fat al. Whereas most of the cyanogenic glycosides are widely can be fixed in the cap of screw vial; place strips in picric acid (5.7 gm in 50 0 ml of distilled water, saturated with sodium bicarbonate and filtered) for 5 m inutes and dry with drier or in oven, paste strips in the cap of the vial; put c hopped leaves or plant material in so as cover the entire base of vial, add 2-3 drops of distilled water and 1-2 drops of toluene, crush material with glass rod ; cap the vial; picrate paper in cap should not touch plant tissue or walls of v ial; incubate at 600 C for 2 hours ; observe the change of paper colour to red. Ferrous hydroxide paper test: Dip rectangular strips of Whatman paper in 10% FeS O4 solution for 5 min-utes, remove and dry; next immerse in 20% NaOH solution fo r 20-30 seconds and dry, and fxi ion cap of vial; put crushed leaves in vial, a dd 2-3 drops of distilled water and 2-3 drops of toluene, crush us- ing glass ro d; cap the vial and incubate at 600for 2 hours; immerse strips in 30% H2SO4; obs erve colour change to prussian blue due to formation of sodium ferric ferrocyanide. Terpenes Terpenes include a large group of compounds derived from the mevalonic acid prec ursor and are mostly polymerized isoprene derivatives. Common examples are camph or (Cinnamomum), menthol (Mentha), and carotenoids (Figure 7.17). They seem to h ave a definite role in the allelopathic effects of plants. They are lipid solubl e found in single membrane bound liposomes, in glandular cells as essential oils , and can be easily extracted with petroleum ether and chloroform, and can be se parated by GLC (Gas Liquid Chromatography), enabling qualitative as well as quan titative measure of chemical differences. spread, others such as cyclopentenoid cya-nogenic glycosides are restricted in d istribuTerpenes are isomeric unsaturated hydro-carbons of the basic 5-carbon isoprene ( CH2=C(CH3)-CH=CH2) present in Hamamelis japonica. Terpenoids, the common group o f terpenes are distinguished as 10-Carbon Monoterpenoids (Fennel, Menthol-Mentha , Gymnosperms), 15-C Sesquiterpenoids (Asteraceae as sesquiterpenol, ABA, some e ssential oils), 20-C Diterpenoids (TaxusTaxol, gibbrellins), 30-C Triterpenoids (sterols, steroids, saponins, betulin in Betula papyrifera), 40-C Tetraterpenoid s (Caro-tenoids) and poly-C Polyterpenoids (Rubber). They have been largely used in distinguishing tion mainly to Flacourtiaceae, Passifloraceae, Turneraceae, and Malesherbiaceae. Leucine derived cyanogenic glycosides are found in Rosaceae, Fabaceae, and Sapi ndaceae. Several families belonging to Magnoliales and Laurales contain Cyanogen ic glycosides derived from tyrosine. Presence of cyanogenic glycosides can be tested by two methods. Picrate paper te st: Cut Whatman no. 1 filter paper into rectangular strips so that they 178 Plant Systematics specific and subspecific entities, geographic races and detection of hybrids. St udies in Citrus have focused on determination of the origin of certain cultivars . Studies on Juniperus viginiana and J. ashei have refuted previous hypotheses a

bout extensive hybrid-ization and introgression between the two species. Their d istribution in Pinus has been used (Mirov, 1961) to understand relation-ships. P . jeffreyi has been considered a vari-ety of P. ponderosa, but turpentine distri bu-tion showed that it strongly resembles the group Macrocarpae and not Australe s to which P. ponderosa belongs. A major contribution of terpenoid chemistry has been the use of ses-quiterpene lactones in the family Compositae. Many tribes w ithin the family are characterized by distinct types of sesquit-erpene lactones they produce. This helped to establish that genus Vernonia has two cen-tres of d istribution one in the Neotropics and the other in Africa. Similarly, studies on X anthium strumarium (McMillan et al., 1976) have thrown some light on the origin of Old World and New World populations. Old World populations produce xanthinin/ or xanthinosin, whereas the New World popula-tions contain xanthinin or its ste reoisomer xanthumin. Plants of the chinense complex (from Louisiana) contain xan thumin and are believed to be the source of introduced lies. The occurrence of iridoids in several unrelated families, e.g. Hamamelidac eae and Meliaceae, however, suggests that iridoids could have arisen independent ly several times in the evolution of angiosperms. The occurrence of a distinctiv e iridoid aucubin in Budleja has been taken to sup-port its transfer from Logani aceae to Budlejaceae. Aucubin and geniposide, have shown antitumoral activities. Iridioid presence in plant tissues can be tested using Trim-Hill Reagent. Take 5 gm of chopped leaves in a beaker containing 5 ml of 1 % HCl; heat on water bath at 40-500C for 30 minutes; filter using coarse filter paper; take 0.1 ml of fil trate in a test tube; add 1 ml Trim-Hill Reagent (THR: 10 ml acetic acid, 1 ml 0 .2 % CuSO4 and 0.5 ml conc. HCl); warm over spirit lamp for few seconds; change of colour commonly to yellow, orange or red confirms presence of iridoids. Quant itative estimation can be done by reading absorption at 609 nm, concentration ca lculated on the basis of standard curve of aucubin. Cronquist (1977) proposed that chemical repellents had an important role in the evolution of major groups of dicots. The alkaloid Isoquinolene of Magnoliidae gave way to tannins of Hamamelidae, Rosida e and Dilleniidae, which in turn gave way to the most effective iridoids in Aste ridae, the family Compositae developing the most effective sesquiterpene lactone s. chinense populations in India and Australia. Non-Semantide Macromolecules Triterpenoids occur in several families. Betulin occurs only in bark of white bi rch (Betula papyrfiera) and its relatives, is wa-terproof highly flammable, and is taxonomi-cally useful at species level. Triterpene saponins occur in Apiaceae and Pittosporaceae In addition to DNA and RNA, which will be dealt under Molecular systematics, the macromolecules include proteins , and complex polysaccharides such as starches and celluloses. Starches are commonly found in the form of grains which may be c oncentric (Triticum, Zea) or eccentric (Solanum tuberosum) and present anatomica l charac- teristics which can be seen under a micro- scope. Detailed studies of starch grains under SEM also hold promise for taxonomic significance. and support their close relationship. Iridoids constitute another important group of terpenes (mostly monoterpene lactones). They are present in over 50 families and their presence is correlated w ith sympetaly, unitegmic tenuinucellate ovules, cellular endosperm and endosperm hausto ria. Assuming that independent origin of sev-eral groups with this combination of independent attributes is unlikely , Dahlgren brought together all iridoid-produc ing famiProteins Proteins, together with nucleic acids, are often called Semantides, which are pr imary Taxonomic Evidence 179 constituents of living organisms and are involved in information transfer. Based

on their position in the information transfer DNA is a primary semantide, RNA s econdspecies A, but when mixed with the protein extract of species B, the degree of p recipitin reaction would depend on the similarity between the proteins of the two species. ary semantide and proteins the tertiary semantides. Semantides are popular sourc es of taxonomic information, and most of this information has come from proteins . The information about DNA and RNA will be discussed under Molecular Systematic s in the next section; only proteins are being The antiserum obtained from the mammal normally contains several immunoglobulins that can bind to the same antigen, is said to be polyclonal. This is because an antigen activates several different lymphocytes within the animal, each produci ng a different antibody for the same antigen. Techniques have now been developed which can generate monoclonal antibodies. In a method developed by Milstein and Khler (1975), antibody-producing lymphocyte of mammal (which can not grow and di vide in cultures) was fused with malignant myoloma cell (cancer cell which can g row rapidly in cultures) to produce hybrid cells called hybridoma. These hybrido mas can grow, proliferate and produce large amount of monodiscussed here. Proteins are complex macromolecules made up of amino acids linked into a chain b y peptide bond, thus forming a polypeptide chain, organized into a three dimensi onal structure. Because of their complex structure, special techniques are necessary for the isolation, study and comparison o f proteins. These methods include serology, electrophoresis and amino acid sequencing. clonal antibody. Serology The field of systematic serology or serotaxo-nomy had its origin towards the tur n of the twentieth century with the discovery of serological reactions and devel opment of the discipline of immunology. Precipitin reactions were first reported by Kraus (1897). The technique was originally applied by J. Bordet (1899) in hi s work on birds, when he reported that immune reactions are relatively specific and the degree of cross reactivity was essentially proportional to the degree of relationship among organisms. The present technique of serology is based on imm unological reactions shown by mammals when invaded by foreign proteins. In the s tudy of estimat-ing relationships between plants, the plant extract of species A containing proteins (antigens) is injected into a mammal (usually a rabbit, mou se or goat). The latter will develop antibodies, each specific to an antigen wit h which it forms a precipitin reaction, coagu- lating and thus making it non-fun ctional. These antibodies are extracted from the body of the animal as antiserum . This antiserum is capable of coagulating all proteins in Antigens are mostly extracted from seeds and pollen. In early works, crude total comparison of precipitin reactions was done but now more refined methods have b een developed which can bring about individual antigen-antibody reactions. Major methods include: Antipru AMtgeris

' 1 t` uk+

B Figure 7. 18 Double-diffusion serology. A: Anetigens and antibodies moving towards each other on the gel. 1-10 refer to antigen mixtures from ten different taxa; B: Resultant precipitant lines. 180 Plant Systematics O Radio-immunoassay (RIA) In this technique the antibodies or antigens are labelled with radioactive molec ules enabling their detection even when present in minute quantities. A

Q Enzyme-linked immunosorbent U ~ U ' assay (ELISA) A $ C In this technique either the antibodies or antigens are labelled linked with enz ymes, thus enabling detection even in very small Figure 7.19 Immuno-electrophoresis. A: Antigen separation by electrophoresis; B: Antibodies and separated anti-gens diffusing towards each other; C: Resultant precipitant lines. quantities. Double-diffusion serology It must be noted that there are specific sites on proteins (determinants), which are capable of initiating production of immunoglobulins in specific cells of ma mmals. Determinants are regions consisting of 10-20 amino acids and one protein may comprise several different determinants and In this technique the antigen mixture and antiserum are allowed to diffuse towar ds one another in a gel (Figure 7.18). The different proteins travel at differen t rates and thus the reactions occur at different places on the gel. This method allows comparison of precipitin reactions of several antigen mix-tures from dif ferent taxa simultaneously on the same gel. In a modification of this method, th e antiserum is placed in a circu-lar well surrounded by a ring of several wells containing the samples of antigens. thus several antigens. Immuno-electrophoresis

Extensive studies of the immunoelectrophoretic patterns of the genus Bromus were done by Smith (1972, 1983). Results showed that North American diploids of the genus are reasonably diverse. The study also high-lighted that antisera raised f rom different species could provide different results. On the basis of serologic al studies, Smith established the distinct identity of B. Pseudosecalinus, previ ously recognized as a variety of B. secalinus. This separation was supported by cytological evidence also. Serological studies have also supported the removal o f Nelumbo from Nymphaeaceace into a separate family Nelumbonaceae, placement of Hydrastis in Ranunculaceae (and not Berberidaceae), and merger of In this technique the antigens are first separated unidirectionally in a gel by elec-trophoresis and then allowed to travel towards the antiserum (Figure 7.19). This method enables a better separation of con-stituent reactions but has the l imitation that only one antigen mixture can be handled on a single gel. Mahonia with Berberis (Fairbrothers, 1983). Absorption Protein mixtures from different species often contain a large number of common p ro-teins, especially those involved in common metabolic processes. The antibodie s for these common proteins (antigens) are first removed from the antiserum so t hat there is a more logical comparison of precipitin reactions. Serology may be done through comparison of protein mixtures or the comparison of single isolated and purified proteins. Schneider and Liedgens (1981) developed a complex but excellent procedure of monoclonal culture of antibodies, but unfor tunately used this for construction of a phylogenetic tree not parallel with accep ted evolutionary schemes. Fairbrothers (1983) cautioned that an evolutionary tre e should not be constructed on the reactions of a single Taxonomic Evidence 181 enzyme or a single species. Lee (1981) using purified protein for antigen and us ing different techniques concluded that Franse-ria (Asteraceae) should be merged with Ambrosia. remove water; pour 5% Stacking gel (0.82 ml Acrylamide-bisacrylamide, 3.27 ml di stilled water, 0.625 ml 0.5 M Tris-HCl, 0.05 ml 10% SDS, 0.25 ml 10% APS, 5l TEME D) up to one third volume of glass mould, insert comb carefully so that bubbles are not formed, allow to polymerize for 30 min; install gel assembly into electr ophoresis apparatus; add 5X SDS Running buffer (6.026 g Tris, 28.8 g Glycerine, 2 g SDS; make to 200 ml with distilled water) in upper and lower chamber, remove the comb from under the buffer; load protein samples of different concentration s into wells by micropipette (also load molecular weight marker proteins in one lane); connect electrodes and run current of 20 mA for 10-15 min, increase curre nt to 40 mA, track the mobility of sample; disconnect power, remove gel carefull y; stain gel with Coommassie Blue stain (200 ml methanol, 50 ml glacial acetic a cid, 250 ml water, 0.25% Coommasie Blue) for 15-30 min; Electropphoresis The technique of serology serves to compare the degree of similarity between the protein mixtures of different species and does not in-volve the identification of proteins. The sepa-ration and identification of proteins can be done by elect rophoresis. Separation is based on the amphoteric properties of proteins whereby they are positively or negatively charged to various extents according to the p H of the medium, and will travel through gel at various speeds across a voltage gradient, usually carried out in a polyacrylamide gel (polyacrylamide gel electr ophoresis PAGE). The procedure involves homogenizing the tissues (containing prote ins) in a buffer solu-tion. Sample is loaded into wells in the cen-tre of the ge l. The current is run for a specific time, and the proteins run up to differ-ent points on the gel. The gel, usually 1 cm thick, is cut into three thin slices, each about 3 mm thick. These slices are subjected to different staining techniqu es and proteins are identified using various criteria. In com-monly used Western blot technique the protein bands are transferred from the gel to nitrocellulose membrane for further process-ing. In disc-electrophoresis, a gel of larger pore s is placed over a gel of smaller pores. The former is used for crude separation and the latter for a complete separation. destain gel till bands are visible.

Method for Western blotting: Use gel from elec- trophoresis without staining; ma ke cut at bottom of gel for orientation; cut nitrocellulose sheet to the size of gel and dip in transfer buffer (14 g glycerine, 3 g Tris base, 0.75 g SDS, 100 ml methanol, make volume to 1 litre with distilled water); soak sponge in transf er buffer and place wet sponge on gel holder; place Whatman paper on sponge; pla ce gel over Whatman paper avoiding air bubbles; keep membrane with shining surfa ce towards gel and roll with glass pipette; place Whatman 3 mm paper over the me mbrane and a second sponge over the paper; place assembly in transfer tank conta ining sufficient transfer buffer to completely cover the blot; place assembly in case with gel facing the cathode and membrane the anode; run current for 4 hour s at 36 V; lift membrane and stain with Poinceau S Staining Solution. In the technique of isoelectric focusing, a gel of a single pore size, is set up with a pH gradient (usually 3-10), so that proteins come to lie on the gradient corresponding to their iso-electric point. These can be sub-sequently separated more completely by disc-electrophoresis. Isoelectric focusing of Rubisco (Ribul ose 1, 5 diphosphate carboxylase) has been very useful in determining relationsh ip between species of Avena, Brassica, Triticum, and several other genera. It is an excellent protein for helping to evaluMethod for SDS-PAGE Electrophoresis: Prepare working concentrations of 2, 5, 10 and 25 g of protein sample by diluting the stock solution; add loading buffer (5 ml 0.5 M Tris, pH 6.8, 8 ml 50% Glycerol, 8 ml 10% SDS, bromophenol blue, 2 ml ^ mercaptoethanol added immediately before loading); boil for 5 min; store in ice; clean, dry and assemble glass plates of casting assembly; prepare 10% resolving gel (3.24 ml Acrylamide-bisacrylamide, 3.5 ml distilled water, 2.5 ml 1.5 M Tri s-HCl, 0.1 ml 10% SDS, 0.5 ml 10% APS, 10l TEMED) and pour into assembly up to it s two third position; overlay with water ; allow gel to polymerise for 30 min, ate hybridization. Electrophoretic studies have supported the origin of hexaploid wheat (Triticum a estivum) from Aegilops tauschii and T. dicoccum. Johnson (1972), working on stor age proteins showed that T. aestivum (AABBDD) and T. dicoccum 182 Plant Systematics Fagopyrum esculentum Spinacea oleracea Pastinaca sativa Tropaeolum majus Guizotia abyssinica Helianthus animus Arum maculatum Allium porrum Zea mays Hordeum vulgare Triticum sp. Lycopersicon esculentum Ricinus communis Sesamum indicum Acer negundo Abutilon theophrasti Gossypium barbadense Sambucus nigra Nigella damaseena Cannabis sativa Phaseolus aureus Cucurbita maxima Brassica oleracea Ginkgo biloba

Figure 7.20 Cladogram of 25 species of seed plants based on the ancestral sequenc e method used by Boulter (1974) (after Boulter). (AABB) possess all proteins of the A genome of the diploid T. monococcum (AA). T hey also share proteins of the B genome of uncertain origin. The D genome is bel ieved to have come from Aegilops tauschii as evidenced by morphological and cytological data. By mixing proteins of A. tauschii and T. dic occum it was seen that the electrophoretic properties of the mixture closely res emble those of T. aestivum, thus proving the origin of the Taxonomic Evidence 183 latter from the two previous species. Electro-phoretic studies have also helped to assess species relationships in Chenopodium (Crawford and Julian, 1976), by c ombining data from flavonoids with proteins. A flavonoid survey of seven species showed that in some taxa, the flavonoid data were fully compatible with intersp ecific protein differences, but in some cases, did not agree. Thus, Chenopo-dium atrovirens and C. leptophyllum had identical flavonoid patterns but could be di stinguished by their different seed protein spectra. C. desiccatum and C. atrovi rens, on the other hand, were closely similar in seed proteins but differed in f lavonoids. Both flavonoid and protein evidence, however, dis-tinguished C. hians from C. leptophyllum, thus providing support to their recognition as sepa-rate species. Vaughan et al., (1966) through the study of serology and electrophoresi s have shown that Brassica campestris and B. oleracea differences between the proteins result from different sequences of amino acids in the polypeptide chain. It is now possible to break off the amino acids from t he polypeptide chain one by one, identify each chromatographically and build up the sequence of amino acids step by step. Cytochrome c is the most commonly used molecule and out of 113 amino acids, 79 vary from species to species, but alter ation of even one of the other 34 destroys the functioning of the molecule. Bein g present in all aerobic or-ganisms, it is ideal for comparative studies. Boulte r (1974) constructed a cladogram (Figure 7.20) of 25 species of spermatophytes are closer to each other than to B. nigra. using the ancestral sequence method . Ginkgo biloba, the only gymnosperm used occup ied isolated position in the cladogram. Ginkgo with an isolated phylogenetic pos ition is no new discovery, but rather a long established fact. But the fact that amino acid sequencing also produces a similar cladogram establishes the signifi cance of Electrophoresis has also made possible the separation of allozymes (different fo rms of the same enzyme with different alleles at one locus) and isozymes (or iso enzymes with different alleles at more than one locus). Barber (1970) showed tha t certain polyploids possess isozymes of all their progenitors plus some new one s. Backman (1964) crossed two strains of maize, each with three different isozym es. F1 possessed all six isozymes. The hybrids thus show such studies in understanding phylogeny. molecular complementation. Recent data from various fields have pointed to the merger of Aegilops with Trit icum. Autran et al., (1979) on the basis of N-terminal amino acid sequencing sup ported this merger. In general, the number of amino acid differences is roughly parallel to the distance between the organisms in traditional classifications, s uggesting that the method is broadly reliable. There are, how-ever, certain cont radictions. The number of differences between the cytochrome c of Zea mays and T riticum aestivum (both members of the same family Poaceae) is greater than Studies of the genus Tragopogon have con-firmed that the tetraploid T. mirus is a hybrid between two diploid species, T. dubius and T. porrfiolius. Whereas the parental diploids were found to be divergent at close to 40 per cent of the 20 e nzyme loci examined, the tetraploid hybrid possessed completely addi-tive enzyme patterns. The evidence thus sup-ported the recognition of a hybrid on the basis of morphological and chromosomal evidence. between Zea mays and certain dicotyledons. Amino acid sequencing It has been found that cytochrome c and plastocyanin (another protein commonly u

sed in amino acid sequencing studies) can exhibit a large number of parallel sub stitutions (identical changes from one amino acid to another at the same positio n in the protein in different organisms), thus rendering them unsuitable for con structing phylog- enies. The practical solution is to use evidence from a wide r ange of proteins, prefer-ably using different techniques. Since only 22 amino acids are known to be the constituents of proteins, the prim ary 184 Plant Systematics MOLECULAR SYSTEMATICS years ago the primary atmosphere consisted of only hydrogen and helium, but bein g too small a planet to hold these light gases, they floated away into space. Th e earth accumulated its secondary atmosphere because of volcanic activity in ear ly hot earth and the gases consisted of largely steam, variable amounts of CO2, N2, SO2, H2S, HCl, Sulphur and smaller quantities of H2, CH4, SO3, NH3. There wa s no free oxygen. Our present at-mosphere is of biological origin, in which meth ane and ammonia have largely been consumed, inert components like nitrogen remai ned unchanged, and oxygen produced by photosynthesis. This happened nearly 2500 m years ago when Cyanobacteria, the first photosynthetic bacteria made their app earance. The ultraviolet radiations from the sun, together with lightening disch arges caused the gases to react in the primeval atmosphere forming simple organi c compounds such as amino acids, sugars and nucleic acid bases. This mostly happ ened because of gases dissolving in primeval oceans and continuing to react form ing primitive soup, the precursor of life. Further reactions formed polymers, gl obules and The closing years of the past century saw the concentration of macromolecular st udies towards DNA and RNA, resulting in the establishment of an emerging field o f mo-lecular systematics. Although flavonoids and isozymes also constitute molec ular data, molecular systematics commonly deals ex-clusively with the utilizatio n of nucleic acid data. As molecular data reflects gene-level changes, it was be lieved to reflect true phy-logeny better than morphological data. It has, howeve r, been realized that molecular data may also pose similar problems, although th ere are more molecular characters avail-able and comparison is generally easier. Molecular evolution Traditionally, different taxa, especially the species have been characterized pr imarily on morphological differences (phenotypes). Additionally differences in p hysiology, bio-chemistry, anatomy, palynology, embryology, gross chromosome stru cture and behaviour, have been used in refining evolutionary trees. Although, it had been long recognized that evolution is based on genetic changes, only durin g the last two decades, there have been forceful drive to use genetic material f or a better understanding of evolutionary re-lationships. Those species that are closely related, are expected to have greater simi-larities in their genetic ma terial than the distantly related species. During the past decade, molecular gen etics has taken a dominant role in enabling us to understand speciation and evol ution clearly. Differences in the nucleotide sequences are quantita-tive and can be analyzed using mathemati-cal principles, utilizing the help of computer prog rams. Evolutionary changes at the DNA level can be objectively compared among di fferent species to establish evolutionary relationships. eventually the first primitive cell. Evolution of Nucleic acids The possibility of such reactions in the primitive atmosphere was demonstrated b y a Russian biochemist Alexander Oparin in 1920s who proposed that life evolved before there was any free oxygen in the atmosphere. The oxygen if present at tha t stage would have reacted with precursor organic molecules formed in the atmosp here, oxidizing them back into carbon dioxide and water. These reactions were mi micked by biochemist Stanley Miller in 1950s, who subjected a mixture of methane , ammonia and water vapour to high voltage discharge or to ultra-violet light, a nd the products allowed to dissolve and react in water. As long as oxygen was ex cluded, the results were similar pro-ducing several organic molecules such as am ino acids, formic acid, glycolic acid, lactic acid, acetic acid, propionic acid,

succinic acid, urea, purines, pyrimidines and sugars. These energy sources can also destroy these organic molecules present in the atmosphere. The and Proteins When Earth originated nearly 5000 million Taxonomic Evidence 185 occurrence of primeval oceanic atmosphere helped shielding and preserving these or-ganic molecules and prevented their destruction. Organic acids, particularly amino acids are soluble in water and non-volatile, and little nucleotide changes give rise to synonymous codons, that code for the same amino acid. Evolutionary Rates within chance of their returning to atmosphere. a Gene The polymerization of amino acids and other monomers to form macromolecules requ ires energy for formation of bonds and removal of water. Such polymers known as proteinoids can be generated by simple heatIt is now well established that different parts of genes evolve at widely differ ent rates, reflecting the extent of natural selection on each part. Some nucleot ides code for amino acid sequence of a protein (Coding sequences), whereas other s do not code for amino acids in a protein (noncoding sequences). Latter include introns, leader regions, trailer regions (all these are transcribed but not tra nslated), and 5 and 3 flanking sequences that are not transcribed. Pseudogenes, wh ich are nucleotide sequences that no longer produce functional gene products as they have accumulates inactivating mutations, also constitute noncoding sequence s. Even within coding regions of functional gene, not all nucleotide substitutio ns produce a corresponding change in the amino acid sequence of a protein. Many substitutions occurring at the third position of triplet codons have no effect o n the amino acid sequence of the pro-tein because such changes often produce ing of amino acids at around 1500 C for a few hours. Such heating could have occ urred near volcanoes or when pools left behind by changing coastline evaporated. Inorganic polyphosphates present in the primeval times would have helped in condensation. It is generally thought that RNA probably evolved first through polymerisation o f nucleotides present in primeval environ-ment. When RNA template is incubated w ith mixture of nucleotides and zinc as a cata-lyst, a complementary piece of RNA is synthesized. The complementary strand in turn will act as template to genera te more of original RNA molecule. It is assumed that RNA originated even before proteins. It is also believed that earliest organisms had both genes and en-zymes made of RNA and formed RNA world . The examples of enzymatically active RNA are found in Ribozymes and self-repli cating introns. Later proteins infiltrated and took over the role of enzymes. Th is was followed by the evolution of DNA as genetic material, and RNA relegated t o the role of intermedisynonymous codons. ate between genes and the enzymes. Although synonymous and nonsynonymous nucleotide changes are likely to arise in equal frequency (because enzymes re-sponsible for DNA replication and repair can not differentiate between the two), yet the rate of synonymous nucleotide chang es (con-servative substitutions of Kimura) is about five times greater than obse rved rate of non-synonymous changes (disruptive substitutions of Kimura). This i s because synonymous changes do not alter protein structure and function and are tolerated by natural seChanges in DNA sequences (mutations) lead to the changes in the codons, that in turn determine the sequence of various amino acids, deciding the final structure and function of a protein. These changes commonly result from changes of one or more lection, but the nonsynonymous changes are base pairs in a DNA sequence. Two types of usually detrimental and are exclu ded by nucleotide changes occur in the genome. natural selection. Synonymous sub

stitution Some changes give rise nonsynonymous rates and not nonsynonymous nucle otide codons, coding for different amino acid, and changes are, as such, the fai r reflection of thus resulting in a corresponding change in actual mutation rate within a genome. the amino acid sequence of a protein. Other Pseudogenes and 3 f lanking regions also 186 Plant Systematics show high evolutionary rates, comparable to synonymous changes. 5 flanking region s show a little slower rates, whereas leader and trailer regions show very low e volutionLarge Single copy region atpB ary rates, slightly higher than nonsynonymous changes. It is as such obvious tha t nucleotide changes in noncoding regions or codings that do not alter amino aci d se-quences, have high rate of evolution, whereas changes in coding regions, es pecially those affecting amino acid sequences show very low rate of evolution, a s most of matK I~ In them get filtered out by natural selection. R It is important note that whereas muta-tions are changes in nucleotide sequences that occur because of mistakes in DNA repli-cation or repair processes, the sub stitutions are mutations that have passed through the filter of selection at lea st at some level. nverted Repeat A ndhF Small Single copy region Figure 7.21 Chloroplast DNA with location of genes commonly used in molecuLocation of molecular data lar systematics. Systematists use molecular data from three different locations within a plant ce ll: chlo-roplast, mitochondrion and the nucleus, yielding three different types of genome (DNA). Chloroplast genome is the smallest ranging from 120 to 160 kbp (kilo base pairs) in higher plants (up to 2000 kbp in alga Ac-etabularia), mitoc hondrion genome 200 to 2500 kbp, whereas the nuclear genome is much larger often ranging between million to more than billion kbp. Although the former two are i nherited from the maternal parent, the latter is biparental. Mitochon-drion geno me undergoes a lot rearrange-ments, so that many different forms may be found wi thin the same cell, and hence is of little significance in interpreting phylogenetic relationships, whereas the other two are highly stable not only within the same cell, but also within a species, and present useful taxonomic tools. as each cell contains several mitochondria, the number of mtDNA molecules per ce ll could be very large. Most mtRNA molecules are circular, but linear in Chlamyd omonas reinhardtii. In vascular plants, mtDNA is considerably larger, circular, containing many noncoding sequences, including some that are duplicated. The physical mapping of genes in vascular plants has shown that these are located in different positions on mtDNA circles of different spec ies, even in fairly closely-related species. This renders mtDNA less useful in p hylogenetic studies. Chloroplast DNA Mitochondial DNA Studies of DNA in plants have largely been undertaken from chloroplast compared to the other two cellular genomes. This is because chloroplast DNA (cpDNA) can b e easily isolated and analyzed. It is also not altered by evolutionary processes such as gene dupli- cation and concerted evolution (in rRNA, having thousands o

f copies of repeated segments so that mutation in one sequence gets corrected to match other copies, this homogMitochondrial DNA has been studied from several species of plants. Each mitochon -drion contains several copies of mtDNA, and Taxonomic Evidence 187 enization process is termed as concerted evolution). It also has an added advant age in that it is highly conserved in organisation, size and primary sequence. C hloroplast DNA is closed circular molecule (Figure 7.21) with two regions that e ncode the same genes but in the opposite direction and known as IGS 18S (SSU) 5.8S 26S (LSU) 00000 00000 r ITS1 ITS2 inverted repeats. Between the inverted repeats are single copy regions. All cpDNA Figure 7.22 molecules carry basically the same set of genes, arranged differently in different spe-cies of plants. Th ese include genes for ribo-somal RNA, transfer RNA, ribosomal proteins and about 100 different polypeptides and subA portion an array of ribosomal genes. Each unit has three sub-units separated b y two ITS (internal transcribed spacer) regions. Adjacent units are separated by IGS (intergenic spacers). units of enzyme capturing CO2. Most studies of chloroplast DNA have focused on chloroplast gene rbcL, which enc odes large subunit of photosynthetic enzyme RuBisCO (ribulose-1,5-biphosphate ca rboxylase/ oxygenase, carbon acceptor in all photosynthetic eukaryotes and cyano bacteria. The gene occurs in all plants (except parasites), is fairly long (1428 bp), presents no problems of alignment, and has many copies available in the ce ll. Ready availability of PCR primers has made it pos-sible to generate over 200 0 sequences, primarily of seed plants. Other commonly used chloroplast genes inc lude atpB, (beta subunit of ATP synthetase involved in the synthesis ATP), matK (maturase involved in splicing type II introns from RNA tran-scripts), and ndhF (subunit of chloroplast NADH dehydrogenase, which functions in ited biparentally, the chloroplast genome is inherited maternally. Thus the hybr id plant will possess the nuclear complement of both parents but only the cpDNA of the maternal plant. The study of nuclear genes has traditionally involved ribosomal RNA. Ribosomal g enes are arranged in tandem arrays of up to several thousand copies. Each set of genes has a small subunit (18S) and a large subunit (26S) separated by a smalle r (5.8S) gene (Figure 7.22). It must be noted that 5S RNA although also a part of the unit, but of the unknown function is synthesized separ ately outside nucleolus. The three subunits are separated by internal transcribe d spacers (ITS: ITS1 and ITS2). Each set of genes is separated from adjacent one by a larger spacer (variously known as IGS- intergenic spacer, EGS- extragenic spacer or NTSconverting NADH to NAD+H, involved in reactions of respiration. Of these four nontranscribed spacer). Sequences of 18S and commonly used genes rbcL, atpB, and matK 26S genes have been used in phylog enetic belong to large single copy region, where studies, because they have some highly conas ndhF is located on small single copy servative regions which help in alignmen t, region. and other variable regions, which help to distinguish phylogenetic groups. Recen tly ITS region has been used to determine relation-ships among species. In gener al, the ITS region has supported relationships inferred from chloroplast studies

and morphology. Nuclear DNA The nuclear DNA, although more difficult to analyze, and hence used less frequen tly has two great advantages. Certain nuclear sequences evolve more rapidly than cpDNA sequences, and thus allow finer level of discrimination at population lev el than cpDNA. Also, whereas the nuclear genome is inherMolecular techniques The techniques of handling molecular data saw great advancements over past few d e188 Plant Systematics cades, starting with comparison of whole DNA molecules. It is now possible to br eak hydroxyapatite. Any radioactive strands (A) that have separated from the DNA duplexes DNA at specific sites, generate maps of in- pass through the column, and the amo unt is dividual genes, determine sequence of measured from their radioactivity. A graph genes, and make multicopies of a DNA showing the percentage of ssDNA at each through Polymerase chain reaction (PCR) temperature is drawn. The temperatu re at technique. These help in generating enough which 50% of the DNA duplexes (dsDNA) have molecular data for comparison. been denatured (T50H) is determined. Bolton (1966) found that only half nucleTotal DNA/DNA hybridization otide sequences in the DNA of Vicia villosa are similar (homologous) with those of Pisum, while only 1/5th are homologous between Phaseolus and Pisum. In the te chnique of DNA/RNA hybridization, the RNA is hybridized with the complementary D NA of related plants. Mabry (1976) used this technique in Centrospermae (Caryoph yllales) and con-cluded that the family Caryophyllaceae (al-though lacking betal ains) is quite close to betalain-containing families, but not as close as the la tter are to each other. The early studies on utilization of nucleic acids in systematics involved DNA/DN A hy-bridization using the whole DNA for study. In a method developed by Bolton and Mecarthy (1962), the extracted DNA is treated to make it single stranded. Th e DNA of another organism is, similarly, made single stranded. The two are subse quently allowed to hybridize in vitro. The degree of reassociation (annealing) e xpresses the degree of similarity in sequences of nucle-otides of the two organi sms. Procedure in-volves heating DNA so that it becomes de-natured into single s trands (ssDNA). The ttemperature is lowered just enough to allow the multiple sh ort sequences of repetitive DNA to rehybridize back into double-stranded DNA (ds DNA). The mixture of ssDNA (repre-senting single genes) and dsDNA (represent-ing repetitive DNA) is passed over a column packed with hydroxyapatite. The dsDNA s ticks to the hydroxyapatite; ssDNA does not and flows right through. The purpose of this step is to be able to compare the information-encoding portions of the genome mostly genes present in a single copy without having to worry about varyi ng amounts of noninformative repetitive DNA. The ssDNA of species A is made radi oactive. The radioactive ssDNA is then allowed to rehybridize with nonradioactiv e ssDNA of the same species (A) as well as in a sepa-rate tube the ssDNA of spec ies B. After hybridization is complete, the mixtures (A/A) and (A/B) are individ ually heated in small (2 3C) increments. At each higher temperature, an aliquot is p assed over Chromosome painting The technique of chromosome painting provides another way to compare entire geno mes. A fluorescent label is attached to the DNA of individual chromosomes of one species. These chromosomes are exposed to the chromosomes of another species. T he re-gions of gene homology will hybridize taking up the fluorescent label and the painted chromosomes can be examined under a microscope. The method is a modifi cation of fluorescence in situ hybridization (FISH). Chromosome painting studies

in humans have shown that human chromosome 6 has counterparts in chromosome 5 o f chimpan- zee, chromosome 7 of pig and chromosome 23 of cow as few examples. Unravelling DNA Structure Understanding DNA structure involves complex procedure to unravel the arrange-me nt of genes in DNA, and sequence of arrangement of nucleotides which differentia tes different genes and the DNA of differ-ent organisms. The procedure involves some distinct steps. Taxonomic Evidence 189 DNA Cleaving DWNTA This technique is a landmark development of 1970s that can be used to generate p hysical maps of individual genes or the entire genome. The DNA extracted from a species is cut (cleaved) at specific points (recogni-tion-site; restriction site ), yielding restriction fragments using restriction endonu-cleases (REs). The sp ecific enzymes are named using the first letter of the genus and the first two l etters of the species of the bacterium from which the enzyme is isolated. Thus, enzyme EcoRI which cleaves DNA at every site where it finds sequence GAATTC CiYA`IAtT:~TdC 3; 5f i0 C 1i' VAfAlU (Figure 7.23) is obtained from Escherichia coli. HindIII obtained from Haemophilus influenzae strain Rd cleaves DNA at AAGC TT, and BamHI from Bacillus amyloliquefaciens cleaves GGATCC. More than 400 rest riction enzymes have already been isolated. Their natural function is to inactiv ate invading viruses by cleaving the viral DNA. Majority of restriction enzymes recognize a 6-nucleotide sequence, but others recognize 4 nucleotide sequence. T hus AluI (from Arthrobacter luteus) recognizes AGCT, TaqI (from Thermus aquaticu s) TCGA, and HaeIII s. 3; (from Haemophilus aegypticus) GGCC. Each restriction enzyme can recognize a sequence four to six nucleotides long, h aving twofold rotational symmetry, because it can be rotated 1800 without change in the base sequence. Thus, sequence recognized by EcoliRI if read from 5 to 3 in b oth strands, of DNA segment would read GAATTC, but if read from 3 to 5 in both stran ds it would read CTTAAG. This symmetry is known as palidrome (as, for example, i n nonsense phrase: AND MADAM DNA that is read simi-larly from both ends). This f eature combined with the fact that most restriction enzymes give staggered cuts (and not straight cuts) wherein they cut two strands of DNA at dif-ferent points , produces complementary single-stranded termini that can be rejoined later usin g enzyme DNA ligase. Such enzymes produce sticky ends or cohesive ends. Others l ike AluI and HaeIII, however, make simple double stranded cut in the Figure 7.23 Cleavage of DNA using EcoRI restriction enzyme. The enzyme gives staggered cuts to ensure complimentary single stranded termini. middle of the recognition sequence, resulting in blunt end or flush end. The use of restriction enzymes allows the DNA to be dissected into a precisely-d efined set of specific segments. Using different enzymes, sites cleaved by diffe rent enzymes can be identified and ordered into a restriction map or physical ma

p. 190 Plant Systematics Method of DNA Cleaving: Label three sterile microfuge tubes as E (for EcoRI), H (for HindIII) and l (for control). In tube E add 12l distilled water, 2l of 10 X B uffer (prepared by dissolving 108 g Tris, 55 g Borate and 7.4 g EDTA in 700 ml o f distilled wa-ter, adjust pH to 8.3 and sterilize by autoclaving), 5ml of lamda DNA (1mg) and 1l of EcorRI restriction enzyme ( 2 U). In tube H add same chemica ls except 1l of HindIII restriction enzyme ( 2 U) instead of EcorRI. In l tube ad d same chemicals but replace distilled water for restriction enzyme. Flick all t ubes to mix well and spin for 5 minutes in microfuge. Incubate all tubes at 37oC for 60 minutes in water bath. Stop reaction by incubating tubes at 65oC for 5 m inutes. Subject contents of all three tubes to agar-ose gel electrophoresis for 1 hour at 100 V. Stain the gel with ethidium bromide. Fragments with be lined in each lane according to size. These can be com-pared and suitably analysed. and complementary ends of two different DNA molecules join to form double strand ed recombinant DNA, in the presence of DNA ligase. For successful cloning, one o f the parental DNAs incorporated into recombinant DNA molecule is capable of sel f-replication, and is known as cloning vector. In practice, the gene or DNA frag ment of interest is inserted into a specially-chosen cloning vector, which is us ed as a vehicle for carrying foreign DNA into a suitable host cell, such as a ba cterium. Method of Agarose Gel Electrophoresis: Prepare 1x6 agarose in 1X TBE buffer; heat the mixture on a hot plate or microwave until the solution becomes clear; add E thidium bromide (EtBr) (0.5 g/ml in 0.5X TBE buffer) to agarose solution when it cools to 45-500C; clean the casting tray, place on table and adjust equilibrium bubble; position the comb 1 mm above the plate; pour solution into gel tray en-s uring that there is no bubble between or under the teeth of the comb; allow the gel to set; pour some 1X TBE buffer over the gel; remove the comb; mix 1.5 l each of DNA sample and bromophenol blue track-ing dye (dissolve 70 g sucrose in 50 m l distilled water by heating, add 0.25 g bromophenol blue and 20 ml 0.5M EDTA, r aise volume to 100 ml) and slowly load mixture into wells; connect assembly with power supply and run the gel at voltage of 80 V until the dye has travelled 75x6 of the distance; turn off the equipment, remove the gel and view under UV; alte rnately stain gel with 0.025x6 meth-ylene blue for 20-30 minutes, destain with lu kewarm water for 30 minutes and observe under white light. Plasmid vector Plasmids are extra-chromosomal double-stranded circular DNA molecules present in microorganisms, especially bacteria. The plasmid chosen as vector contains a ge ne for antibiotic resistance. In the most commonly-employed technique (Figure 7. 24-I), the re-combinant plasmids (with foreign DNA in-serted into plasmid) are a dded to an E. coli bacterial culture pretreated with calcium ions. When subjecte d to brief heat shock, such bacteria are stimulated to take up DNA from their su rrounding medium. Once within the bacterial cell, the plasmid replicates au-tono mously and is passed on to the progeny during cell division. The bacteria contai ning recombinant plasmid can be separated by treatment with an antibiotic which removes bacterial cells without plasmid. Because a large number of different rec ombinant plasmids are formed, incorporating different segments, the one of inter est can be separated by combined procedure of replica plating and in situ hybrid ization (Figure 7.24-III). Through replica plating, numerous dishes with represe ntatives of the same bacterial colony are prepared. In one of the replica plates , cells are lysed and DNA fixed on to surface of nylon or nitrocellulose membran e. DNA is next denatured; membrane is incubated with labelled single stranded DN A probe, containing complementary sequence being sought. The unhybridized probe is washed away, and the location of labelled hybrids determined by autoradiograp hy. In DNA Cloning A detailed analysis of DNA requires avail-ability in sufficient quantity of DNA or its restriction fragments. DNA cloning is a technique to produce large quanti ties of a

specific DNA segment. The technique of cloning has largely been made possible through recombinant DNA technology. The DNA molecules from two dif-ferent sources are treated with restr iction enzyme that makes staggered cuts in DNA, leaving single-stranded tails in either of the cleaved DNA. These tails act as sticky ends Taxonomic Evidence 191 F 7 c~ li E3 A ]3 1 1 li

I7 E I] 17 F:

El (1nneJ hidC~'iuin`itLp _lim~d I ~ Ili Figure 7.24 DNA Cloning. I. Cloning using plasmid vector. A: DNA fragment of an organism; B: Cleaved plasmid DNA; C: Recombinant DNA molecule (5-10 kb); D: Bacterium; E: Bacterium with recombinant DNA. II. Cloning eukaryotic DNA using lambda phage. A : Mutant strain of lambda phage with DNA having two EcoRI cleavage sites; B: Ext racted DNA of phage treated with EcoRI; C: Two fragments of phage DNA, middle se gment discarded; D: DNA segment from eukaryotic cell (about 25 kp); E: Recombina nt DNA; F: Recombinant DNA packed into phage head; G: Culture dish with bacteria l culture with clear plaques of phage infection. III: Combined procedure of repl ica plating and in situ hybridization. A: Dish with bacterial colonies; B: Trans ferring bacterial cells from dish (a) to a filter paper (b); C: Filter paper wit h bacterial colonies; D: inoculating empty culture dish by pressing filter paper ; E: Dish with bacterial colonies; F: Culture dish with bacterial colonies for r eplica plating; G: Nitrocellulose membrane with replica of bacterial colonies; H : DNA separated by lysis of cells and denatured to become single stranded adheri ng to membrane; I: Radiograph of labelled hybrid. 192 Plant Systematics refined technique of fluorescence in situ hybridization (FISH), probe labelled w ith fluorescent dyes is used, and labelled hybrids A yeast gene can be cloned in shuttle vector, subjected to site-specific mutagen esis in E. coli, and then moved back to the yeast to examine the effects of indu ced modificalocalized with fluorescent microscope. The live representatives of the identified clones can be found on corresponding

sites on the original plates, these cells are grown into large colonies, which s erve to amplify recombinant DNA plasmid. After sufficient amplification, the DNA is extracted and recombinant plasmid DNA is separated from bacterial DNA. The r ecombinant plasmid DNA is again treated with the same restriction enzyme that re leases plasmid DNA from the cloned DNA segments. Latter can be sepa-rated from p lasmid DNA by centrifugation. tions in native host cells. Artificial chromosome vectors Attempts have been made over the recent years to develop vectors which can accom modate DNA sequences larger than 45 kb. One of the most important of these vecto rs is YAC (yeast artificial chromosome), which can accept DNA fragments as large as 1000 kb. More recently the use of BAC (bacterial artificial chromosome) has become more common. BACs are specialized bacterial plasmids (F factors) that con tain bacterial origin of replication, and can accommodate up to 300 kb of DNA se gments. Bacteriophage vector Bacteriophage ^ (lambda) is commonly used as a vector. The DNA of the phage is linear 50 kb in length. During treatment with reAmplification through PCR striction enzyme middle 15 kb segment of The earlier procedures for obtaining a large phage DNA which contains genes for lysis and quantity of DNA were very cum bersome, incan be dispensed with is replaced with foreign volving the cloning of genes into bacteria, DNA. The resultant recombinant DNA is which replicate genes along with their own genome. The development of PCR (polym erase chain reaction) technique has now made it possible to obtain large number of copies of a gene using enzyme in place of bacteria. Small pieces of single-st randed DNA with known sequence are used as prim-ers (Figure 7.25). These primers are built from templates of short regions of DNA that occur at either end (flan king) of DNA segment of interest, do not occur any where else in genome (unique) , and are invariable (conserved) in all taxa to be investigated. The extracted D NA from a species is mixed with the primer, DNA polymerase (usually taq polymera se, which can tolerate heat), buff-ers, salts and free nucleotides in a tube. Th e mixture is alternately heated and cooled. Heating denatures DNA making it sing le-stranded. The subsequent cooling allows primers to bind to the complementary DNA sequences. Polymers are designed so that they can not bind with each other. The temperature is then raised to make polymerase active, bind to the already fo rmed complex packed into phage heads in vitro (Figure 7.24-II). Phage particles can inject the recombi-nant DNA molecules into E. coli cells, wh ere they will replicate and produce clones of re-combinant DNA molecules. As lam bda heads can accommodate molecules of only 45 to 50 kb size, it can accommodate inserts (foreign DNA fragments) of only 10-15 kb. Cosmid vector For inserting larger DNA insertion, cosmid vectors are used. A cosmid is a hybri d be- tween plasmid and lambda phage. Cosmids combine plasmid s ability to replica te autonomously with in vitro packaging capacity of lamda phage. A cosmid vector can carry out inserts of 35 to 45 kb. Eukaryotic shuttle vectors Some of the most useful cloning vectors are shuttle vectors that can replicate i n both E. coli and another species. Such shuttle vec-tors are very useful for ge netic dissections. Taxonomic Evidence 193 (DNA + polymerase), and begin synthesis of complementary strand (at DNA region n ot bound by primers) using free nucleotides. The temperature is raised further t o dena-ture DNA and the cycle repeated, thus making enough copies of DNA. I Extracted DICTA

Method: Prepare PCR mix E by mixing (all previously placed in ice, taken out and thawed) 35l autoclaved water (pH 7.0), 5l 10X PCR buffer, 2l mixture of deoxyribon ucleotides (dNTPs- 10mM of II each nucleotide: dATP, dCTP, dGTP and dTTP), 1l Forward primer (10m), 1l Reverse pr imer Pruner (10m), 5l Genomic DNA template(25 ng/l) and 1l Taq DNA polymerase (5 units/l); prepar e one more PCR mix without DNA template and label as C; put 20-50l of each mixtur e in PCR tubes, briefly centrifuge, insert tubes into thermal cycler; program PC R machine for temperature 94C (step 1, 5 min: primary denaturation), 94C (step 2, 1 min: secondDenatured sire DICTA strands ary denaturation), Annealing temperature (about 5 10C less than Tm of the primers , step 3, 1 min),

72C (step 4 and 5, 1.5 min), hold, 4 min; take out tubes when temperature comes d own to 4C and place in ice; analyze further by gel electrophoresis as described e arlier. Primers binding complementary sequences DNA libraries DNA libraries are collections of cloned DNA fragments. Two basic types of DNA li braries can be created. Genomic libraries are produced from the total DNA extrac ted from the nuclei and contain all of the DNA sequences of the species. cDNA li braries (cDNA complementary DNA) on the other hand, are derived from DNA copies of usually the mes-senger RNA, and thus represent DNA sequences which are expres sed in the species. This is significant because a large number of DNA sequences do not express themselves, and are of little significance. Sometimes, individual chromosomes of an organism are isolated by a procedure that sorts chromosomes b ased on size and the DNA content. The DNAs from the isolated chromosomes are the n used to construct chro-mosome-specific DNA libraries, which facilitates the se arch for a gene that is known to reside on a particular chromosome. This is part icularly useful for organisms with I Synthesis of new strands progressing

Copy of double stranded DICTA generated Repeated cycles of steps I to large genomes, such as humans. To construct a DNA library, the DNA from a species is randomly cleaved using enz ymes Figure 7.25 Polymerase chain reaction technique.

194 Plant Systematics which recognize short nucleotide sequences, the fragments are incorporated into lambda phage and multiple copies of each recombi-nant DNA obtained. These are st ored and con-stitute a permanent collection of all DNA se-quences present in the genome of a species. To construct a cDNA library using mRNA, a complementary st and of DNA is constructed by reverse transcriptase. RNA-DNA duplexes are convert ed into double-stranded DNA mol-ecules by combined activity of ribonuclease H, D NA polymerase I, and DNA ligase. The double-stranded DNA is incorporated into la mbda phage and further processed as de-tailed above. probe. The segments of different sizes can be ordered to generate physical maps. Gene Mapping Method: Perform gel electrophoresis of DNA sample; look for fluorescent bands; t reat gel with 200 ml of 0.25 M HCl for 15 minutes; rinse with distilled water an d treat twice with 200 ml denaturing solution (DS: 1 M NaCl, 0.5 M NaOH) for 15 minutes each; neutalize the gel by soaking in 200 ml of Neatralising solution (2 .5 M NaCl, 0.5 M Tris-HCl (60.5 g/l), adjust pH to 7.4 with conc. HCl) for 15 mi nutes; Lay transfer buffer TB (20 X SSC, 3 M NaCl (175 g/l), 0.3 M Tri Sodium Ci trate (88 g/l)), prewetted double layer of Whatman paper onto transfer tray, so as to reach both ends of reservoir; put gel upside down on Whatman paper, roll g ently with pipette to remove bubbles; place TB-prewetted nylon membrane on the g el; put 3 layers of TBprewetted Whatman paper on top of it; place 1 dry Whatman paper on top; put 3 layers of blotting papers on top of Whatman papers; pour 200 -400 ml 20X SSC buffer into tray; add 10-20 cm layers or 2/3 of a pack of paper towels; place 500g weight on top and allow the transfer for 12-16 hours; take ou t membrane from assembly; crosslink DNA on nylon Above techniques contribute in developing the physical maps of gene. Whereas res tric-tion enzymes enable cleavage at specific membrane by exposing it to UV light for 3-5 minsites, the cloning and amplification tech- utes; for detection either expose mem brane to niques help in obtaining a large number of X-Ray film or else stain the membrane with 0.025% copies of fragments. methylene blue for 20 minutes and then destain in Identification of the location of genes and water. DNA sequences on restriction fragments separated by gel electrophoresis constitu tes an important step of genome mapping. The process of gene mapping has been si mpli-fied with the availability of cloned organelle genomes which are used as pr obes. In the commonly used Southern blot hybridiza-tion method (named after E. M . Southern, who published it in 1975), a cloned piece of chloroplast DNA (to be used as probe) is la-belled with radioactive phosphorus and de-natured to produc e single-stranded DNA. The cleaved DNA from the specific species, after electrop horetic separation of frag-ments, is placed on a nylon or nitrocellu-lose membra ne, and denatured by using alkaline solution and finally immobilized by drying o r UV irradiation. It is renatured and allowed to bind to the radioactive probe o n a nylon membrane. Only matching se-quences will bind, and carry the radioac-ti ve tag. When transferred to an X-ray film the bound bands will appear as dark ba nds, which will show the positions of DNA se-quences that have hybridized with t he The technique of Northern blot hybrid-ization, (so named as it is opposite of So uth-ern blot technique), is used to hybridize RNA molecules separated by electro phoresis. Denaturing is affected by formaldehyde, and after transfer to the memb rane, the RNA blot is hybridized either with RNA probe or DNA probe. The procedure of gene mapping is sufficiently complex. It involves crossing two plants, selfing F1 and producing a large number of F2 plants. Genotypes of paren ts and offsprings are determined using various markers. Although physical maps c an be constructed by identifying and aligning overlapping DNA fragments, more el

aborate genetic maps are constructed using genetic markers. Genetic map can be unified with physical map using molecular markers. The physic al map thus obtained will afford single framework for organizing and integrating diverse types of genetic information, including the position of chromosome band s, chromosome breakpoints, mutant genes, transcribed regions, and DNA sequences. Taxonomic Evidence 195 NH2 O CHq NH2 N ! H HN N 1 \N\ HN NH HN NH NH NH\2 N NH Adenine O O Guanine Cytosine Thymine Purines Pyrimidines OH OH PPhosphate OH 5' OH P \ P \ CH O Base \ CH Base H

OH H H H Nucleotide H HO CH P HO CH \ Base \ H H CH OH H H H Base Sugar OH H H H Nucleoside OH H 3' A short Segment of polynucleotide strand Figure 7.26 Structure of DNA. One of the four bases is joined to a deoxyribose s ugar to form nucleoside, which links with a phosphate to yield a nucleotide. A long chain of nucleotides forms the DNA strand having OH (3 terminus) group at one end and phos phate (5 terminus) at the other end. Purines bases have double ring structure, wh ereas pyrimidines have single ring structure. Gene Sequencing Sequencing determines the exact order of the bases (adenine, cytosine, guanine a nd thymine) constituting nucleotides in a por-tion of a DNA and thus building an ultimate fine structure map of a gene or chromosome (Figure 7.26). Today, sequencing is a routine laboratory procedure. A complete sequence of human genome has been develope d, as also the small annual weed 196 Plant Systematics Arabidopsis thaliana, developing into a strong genetic tool. Two main procedures of DNA cence as they pass through the well (tube or gel). The output is directly analyz ed by a computer, which analyses, records and prints sequencing are commonly used. In the first procedure developed by Allan Maxam and Walter Gilbert, the DNA chai n is cleaved using four different chemical reactions, each targeting A, G, C or C+T. In the second procedure developed by Fred Sanger (chain termination method) and col-leagues, there is in vitro synthesis of DNA in presence of radioactive nucleotides and spe-cific chain terminators to generate four populations of radi oactively-labelled frag-ments that end with As, Gs, Cs and Ts, reout the results. The PCR product can be sequenced di-rectly using restriction enzymes. Since re-s triction sites are spread at several places on the DNA, the results are less sen sitive to local vagaries of selection or differences in mutation rate. Sequencin g of both the strands often minimizes errors. Analysis of Sequence data

For the analysis of changes at the level of spectively. nucleotides and the amino acids, the align-ment of DNA sequences derived from di fferThe procedure begins with obtaining iden-tical DNA fragments up to about 500 bp us-ing a restriction enzyme. The preparation is divided into four samples. Each sample is denatured into single strands, incubated with a short radioactively-la belled oligo-nucleotide complementary to 3 end of single strands. To each sample is also added DNA polymerase and all the four deoxyribonu-clease triphosphate pr ecursors (dNTPs). To one sample is now added chain terminator ddATP (2 , 3 -dideoxy adenosine triphos-phate), to the second ddGTP (2 , 3 dideoxyguanosine triphosphate) , to the third ddCTP (2 , 3 -dideoxycytidine triphosphate), and to the fourth ddTTP (2 , 3 dideoxythymidine triphosphate). The first sample after reaction will have a ll the seg-ments terminated at As, the second at Gs, the third at Cs and the fou rth at Ts. The frag-ments are separated on gel electrophoresis, and their positi ons determined by autorad-iography. Different bands, representing dif-ferent seg ments will be arranged like a lad-der. By reading the ladder, a complete nucle-o tide sequence of DNA chain can be deter-mined. In conventional slab-gel procedur e, four different samples are loaded in four dif-ferent wells on a gel. Nowadays , automated DNA sequencing machines are used which make use fluorescent dyes ins tead of radio-active nucleotide. The products of all four samples are run throug h single well, and photocells are used to detect the fluoresent taxa constitutes an important step. Alignment helps in detection of insertio n, deletion or substitutions of one or more base pairs at different sites within a DNA. When comparing two sequences with L positions (nucleotides), of which D positio ns are different, the evolutionary distance counted A number of different models have been proposed to explain evolutionary distance be-tween two sequences on a ccount of nucleotide changes. Jukes-Cantor Model T. Jukes and C. Cantor (1969) realized, even before the DNA sequences were avail able for analysis, that alignments between sequences with many differences might cause a significant underestimation of the actual number of substitutions that occurred since sequences last shared a common ancestor. They assumed that each n ucleotide was as likely to change into any of the other three nucleotides. A can thus equally well change into T, C or G. Based on this assumption they created a mathematical model in which rate of change to any one of the three alternative nucleotides was assumed to be a, and the overall rate of substitution for any g iven nucleotide was 3a. According to this model, if a site within a gene was occ upied by a C (t = 0), then the probability (P) that this site would still be sam e nucleotide at time Taxonomic Evidence 197 1 (t = 1) would be PC(1) = 1 - 3a. On the other hand if C changed to some other nucleotide, the probability that after time t, the site would contain C can be c alculated as: If evolutionary rates between species are similar, substitution rates can help i n cal-culating the dates of evolutionary events. PC(t) = (3/4)e-4at Kimura two-parameter (K2P) Model The probability rate matrix for the changes in four nucleotides can be repre-sen tedas under: The model was proposed by M. Kimura (1980) and accounts for different rates of n ucleotide changes involving transitions and transversions. Supposing we assign v alue a for transitions and b for transversions, the probability rate matrix woul d be represented as: A G C

T A 1-3a a a a G a 1-3a a a A G C T C a a 1-3a a A 1-a-2b a b b T a a a 1-3a G a 1-a-2b b b C b b 1-a-2b a It was , however, subsequently realized that transitions (change from purine to purine; pyrimidine to pyrimidine) proceed at much faster rate than transversions (purine to pyrimidine or vice versa), but the Jukes-Cantor model can still be t aken into account for calculating the number of substitutions (the distance betw een two sequences) per site (K) when multiple substitutions were possible: T b b a 1-a-2b The number of substitutions per site (distance between two sequences could be ca l-culated as: Q ) 1/4 1n (1 2Q) K or dK2P = 1/2 1n (1-2P K or djc= 3/4 1n( 1 - (4/3)p) Where p is the fraction of the nucleotide where P and Q are observed fractions of aligned sites whose two bases are relate d by that a simple count reveals to be different between two sequences. It follows fr om the equation that if two sequences have fewer mismatches, p is small and the chance of multiple substitutions is also small. On the other hand if number of m

ismatches are large, the actual number of multiple substi-tutions per site will be considerably larger a transition or a transversion, respectively. than what is counted. Once the sequences are generated, they must be aligned. First the sequences of a given length are aligned by arranging ho-mologous nucleotides in corresponding columns. Alignment is simpler for conserved genes, where all taxa will have same number of nucleotides per gene. Some other genes which have some deletions, add itions, inversions or translocations in some taxa, are difficult to align. Simil arly DNA with multiple copies of a gene makes it difficult to assess homology. S everal computer programs are available to produce alignment, but the as-sumption s used in each program should be carefully examined before the program can Once number of substitutions per site (K) is calculated, knowing the time taken for divergence (T), the rate of substitution (r) can be calculated as: r = K/(2T) For calculating substitution rates, data from at least two species should be available. 198 Plant Systematics DNA Alignment Coding of characters 2 30 2 5 30 Species 1 1 4 1 2 3 4 5 A GTCCAAGACTCTCAGTGGTTCAATCGTCTGTT 2 0 2 4 3 B CTCCAAGTCTCTCACTG TCGTCAGTT 1 3 1 5 0 C CTCCAAGACTCTCAGTGGTTCAATCGTCTGTT 1 0 2 4 3 D GTCCAAGTCTCTCACTGGTTCAATCGTCTGTT 2 3 1 4 3 Figure 7.27 Alignment of a DNA sequence for 32 nucleotide positions in four spec ies. nucleotides at a particular position showing variation can be coded as characters, an d are shown in bold. The codes assigned to nucleotide states are A=0, C=1, G=2 a nd T=3. Character number four here involves deletion of a particular sequence in species B. Presence of this deletion is coded as 5 and absence of deletion as 4 . Computer programs (such as DNADIST and DNAPARS of PHYLIP) are available which can read DNA sequence data directly. be used for a particular set of taxa. In phylo-genetic analysis each nucleotide position is considered as one character, and each of the four nucleotides as one character state. A large number of nucleotide positions, how-ever don t show vari ation among taxa, and of others that are variable are often uninfor-mative becau se of being autapomorphic for a given taxon. This leaves only a small propor-tio n of nucleotide positions that can be used for phylogenetic analysis. Chromosoma l mu-tations such as additions, deletions and translocations are identified as e volutionary novelty, and are generally given more weightage than individual nucl

eotides (Figfrom A to G or vice versa; or from C to T and vice versa) are more common than t ransversions (A to T, A to C, G to C, G to T; C to A, C to G, T to A, T to G). T he latter are often given more weight depending upon the frequency of distributi on in the taxa, more frequently the transitions are distributed, greater weight is consequently given to transversions. Thus if transitions occur 4 times more t han transversions, a transition may be given weight of 1 and transversion a weig ht of 4. Computer programs such as DNAPARS, DNADIST, etc. of PHYLIP are avail-ab le, which can read and analyse the DNA sequence data directly. Details are descr ibed ure 7.27). Such chromosomal changes repunder chapter on Developing Classifications. resenting apomorphy are important and of-ten used in establishing a lineage. Thu s all members of subfamily Faboideae lack one of the inverted repeats found in t he chloroplast Whereas alignment of simple chloroplast genes such as rbCL is easier, others suc h as genes encoding RNAs, secondary structure (folding) of the molecule is also ac-counted for. The nucleotide differences that result in major changes in the s tructure of a product, such as ribosomal RNA or a pro-tein, and may have greater effect in the plant function, often receive greater weight than DNA of most angiosperms. In our example illustrated in Figure 7.27, the four nucleotides are given coding from 0 to 3 for different nucleotides. Other strate-gies could also be used. Tr ansitions (change Taxonomic Evidence 199 I II SNP SNP G-A-A-T-T-C G-A-A-T-T-C G-A-A-T-T-C G-A-A-T-T-C G-A-A-C-T-C G-A-A-T-T-C C-T-T-A-A-G C-T-T-A-A-G C-T-T-A-A-G C-T-T-A-A-G C-T-T-G-A-G C-T-T-A-A-G

Cleavage with EcoRI Cleavage with EcoRI SNP SNP G A-A-T-T-C G A-A-T-T-C G A-A-T-T-C G A-A-T-T-C G-A-A-C-T-C G A-A-T-T-C C-T-T-A-A G C-T-T-A-A G C-T-T-A-A G C-T-T-A-A G C-T-T-G-A-G C-T-T-A-A G r'

Figure 7.28 The effect of the location of SNPs in the restriction sites in two D NA molecules. DNA with TA nucleotide pair shows normal cleaving with EcoRI, resulting in two r estriction fragments. In another DNA (II) substitution of CG pair prevents cleav ing resulting in single large cleavage fragment. This results in unequal fragmen ts in different DNA molecules. those changes that do not affect the func-tion. Several computer algorithms are avail-able to evaluate and handle such analysis. ously developed. Some of the commonly used procedures are discussed below. DNA Polymorphism Single-Nucleotide Polymorphisms (SNPs) Utilization of sequence data in phylogenetic analysis involves the identificatio n of unique sequences which show certain differences in different organisms or p opulations. These sequences, which could be used as genetic markers in identific ation of character-state differences between the target taxa, and ultimate const ruction of phylogenetic trees. The phenomenon is also known as DNA Fingerprintin g or DNA polymorphism. The technique is now widely used in forensic investigatio ns. A variety of methods have been developed to detect this polymorphism. Each m ethod has its own advantages and limitations, and suitable for a particular situ ation. New methods are being continuDNA differences in a population may often be the result of differences in single nucleotide pair at a particular locus, say from C-G to T-A. This may result in three genotypes in a population: homozygous with C-G at corresponding sites on b oth homologous chromosomes, homozygous with T-A at correspond-ing sites on both homologous chromosomes, and heterozygous with C-G in one chromosome and T-A in h omologous chromosome. However, all SNPs are not located on coding sequences or g enes. In human genome, for example, any two randomly chosen DNA molecules differ at one SNP site about every 1000 3000 bp in protein coding DNA, but 500-1000 bp in noncoding DNA segments. SNPs are 200 Plant Systematics most common types of genetic differences among human populations, and are unifor mly Step1 Isolate DNA from species A and cleave using EcoRI distributed over the chromosomes. The SNPs can be easily detected if they are located in a cleavage site (Figure 7 .28). Thus a sequence GAATTC can be cleaved by EcoRI, but a corresponding GAACTC sequence can t be cleaved as T has been re-placed with C (and on the complementar y segment A replaced by G). This will result in larger DNA fragment in the latte r case. Step 2 Cleave these fragments using Km~III

Restriction Fragment Length Polymorphisms (RFLPs) Step 3 RFLP results from the fact that a mutation that causes changes in base sequence may result in loss or gain of a cleavage site, thus alleles differing in the pre sence or absence of a cleavage site. This may also result from SNPs located at c leavage sites as indicated earlier. As a result fragments of different lengths a re yielded. The method is widely used for identification of individuals, species or populations. The DNA from a species is cleaved using a restriction enzyme (s ay EcoRI) yielding a certain number of fragments (Figure 7.29). These fragments can be separated using Southern blotting proce-dure, and a map of these construc ted. These fragments are further fragmented using an-other enzyme (say HindIII), and the data incorporated into original map. Restriction site fragments obtaine d are coded as characters and character-states for phylogenetic Construct Restriction Site map EcoRI EcoRI 5000 6000 y 3500 5300 KnmI HVZW Step 4 Construct Restriction Site map for species B and compare EcoRI EcoRI 5000 6000 Species A 3500 5300 III EcoRI EcoRI EcoRI 3000 5000 6000 Species B 3400 5300 7800 analysis. r r~~~III The absence or presence of a restriction site in closely related species and the pre-sumed hybrids can also be detected by Southern blotting procedure. Species A, for example lacks restriction site at 3000 bp position (allele a, genotype aa ), where this site is present in species B (allele A, geno- type AA). Southern b lotting technique will yield longer first restriction fragment of 5000 bp length for aa genotype, whereas it will yield fragment of 3000 bp length in AA geno-ty pes. The heterozygous Aa genotype, pre-suming the alleles are codominant will yi eld Step 4 Code characters for species A & B EcoRI III III EcoRI III EcoRI III 3000 3400 3500 5000 5300 6000 7800 Species A

+ + + + Species B + + Figure 7.29 Major steps involved in Restriction Fragment Length Polymorphism (RFLP) procedure. Taxonomic Evidence 201 A B C D Random Amplified Polymorphisms (RAPDs) RAPD method is commonly used for population studies and involves short (10bp) ra ndom PCR primers that will bind to the matching sequences on genome. The approac h is useful for species where cloned DNA probes are not available (essential for Southern blotting method), or where DNA se-quences are not known (necessary for PCR amplification where oligonucleotide prim-ers have to be constructed). The m ethod uses PCR primers of 8-10 nucleotides with random sequence. These primers a re tried singly or in pairs in PCR reactions to am-plify segments of DNA from a species. These short primers anneal at multiple sites on DNA, and those that ann eal at suitable distance are able to amplify unknown region between them. The pr esence or absence of such amplified regions in different individuals can be suit ably coded for analysis. The procedure helps in identifying different genotypes in the population. The morphologic characters of interest are mapped accord-ing to their linkage to markers. The results of one of several primers used are show n in Figure 7.30. Gel electrophoresis yields 13 bands, of which four show polymo rphism, the 3 6 7 98 10 11 12 13 Figure 7.30 Results of one of the several primers used for RAPD procedure on rest nine are monomorphic. Each of the poly-DNA f rom four species. A total of morphic allele can be represented similarly 13 band s appear on electrophore- as + for the presence of band, - for its absis gel, 4 show polymorphism in species compared , whereas 9 are monomorphic. sence, and if + is dominant, both genotypes +/+ and +/ will show this band, where as it will be lacking in / genotype. Thus for the last band in the gel species B a nd C have / genotype, where as A and D are either +/+ or +/ . two fragments from homologous chromo-somes, one of 3000 bp length and another of 5000 bp length. RFLP analysis, however, contains much Amplified Fragment Length lesser data than complete DNA sequencing, accounting only for presence or absence of Polymorphisms (AFLPs) sites 6-8 base pairs long, but the method AFLP (amplified fragment length polymo raffords advantage of surveying larger seg- phism) technology is used for nuclei c acid ments of DNA. The use of this method has, fingerprinting, exploiting mole cular genetic however, declined with the development of variations existing betw een closely related improved and less expensive sequencing genomes in the form o f restriction fragment techniques in the recent years. length polymorphisms. 202 Plant Systematics

Step 1 appropriate restriction endonucleases (REs). For most plant DNAs , two REs are u sed: one a rare cutter having 6-bp recognition site, and the other a frequent cu tter with 4-bp recognition site. In the second step, specific double-stranded ol igodeoxynucleotide adapters (primer adapters) are ligated to the ends of the dig ested DNAs to generate chimeric molecules. These primers are so designed that th ey bind at both cut ends of fragments. In the third step the chimeric fragments are subjected to PCR amplification to provide sufficient template DNA for finger printing PCRs. During the fourth step, PCR products are resolved on through elec trophoresis using polyacrylamide sequencing gel, which separates the amplified D NA fragments that exhibit length polymorphisms, enabling the recognition of Isolate DNA from sources of interest and digest with 2 REs G-A-A-T-T-C T-C-G-A C-T-T-A-A-G A-G-C-T EcoRI Taq I G A-A-T-T-C T C-G-A \ / \ /

C-T-T-A-A G A-G-C T Step 2 Ligate adapters to digested DNAs numerous genetic markers. N-N-G-A-A-T-T-C T-C-G-A-N-N One of the earliest significant results of this method were obtained by Jansen a nd Palmer (1987), who found a unique order of genes in the large single-copy reg ion of the chloroplast genome in Asteraceae. This unique order could be explaine d by single

A-G-C-T N-N N-N-C-T-T-A-A-G Step 3 PCR amplification / all other angiosperms, strongly confirming that the Asteraceae family is monophy letic. The family Poaceae, similarly, has three inversions in the chloroplast ge nome. Out of these three inversions, one is unique to the family and confirms it s monophyletic status. Of the other two, one is shared with Joinvilleaceae and o ne with both families Joinvilleaceae and Restionaceae, suggesting that these two are the sister groups of Poaceae.

5 N-N-C-T-T-A-A-G N-N-G-A-A-T-T-C . ... j A-G-C-T-N-N T-C-G-A-N-N Step 4 Product separation and analysis morphisms (STRPs) Figure 7.31 Basic steps in Amplified Fragment Length Polymorphism (AFLP) pro-tocol. N-N represents a particular nucleotide seq uence. STRP results from the fact DNA molecules may differ in the number of copies of a se-quence of few nucleotides repeated in tandem at a particular locus. In TGTGT G sequence, for example, two base pairs are repeated. Such repeated nucleotides are known as tandem repeats. STRPs present at different loci may differ in seque nce and AFLP procedure involves four basic steps (Figure 7.31). In first step DNAs from different sources are isolated and digested with Taxonomic Evidence 203 SPECIES A 3 A-T-C-G-G-T-T-G-T-G-T-G-T G-T-G-T-G-T-G-A-G-G-T-T-A 5 SPECIES B 5 T-A-G-C-C-A-A-C-A-C-A-C-A C-A-C-A-C-A-C-T-C-C-A-A-T 3 3 A-T-C-G-G-T-T-G-T-G-T-G-T-G-T G-T-G-T-G-T-G-T-G-A-G-G-T-T-A 5 5 T-A-G-C-C-A-A-C-A-C-A-C-A-C-A Primer 5 T-A-G-C-C-A-A 3 A-T-C-G-G-T-T-G-T-G-T-G-T 5 T-A-G-C-C-A-A-C-A-C-A-C-A 3 G-A-G-G-T-T-A 5 Primer C-A-C-A-C-A-C-A-C-T-C-C-A-A-T G-T-G-T-G-T-G-A-G-G-T-T-A C-A-C-A-C-A-C-T-C-C-A-A-T 3 5

Primer 5 T-A-G-C-C-A-A 3 A-T-C-G-G-T-T-G-T-G-T-G-T-G-T G-T-G-T-G-T-G-T-G-A-G-G-T-T-A 5 5 T-A-G-C-C-A-A-C-A-C-A-C-A-C-A C-A-C-A-C-A-C-A-C-T-C-C-A-A-T 3 G-A-G-G-T-T-A 5 Primer 5 T-A-G-C-C-A-A-C-A-C-A-C-A C-A-C-A-C-A-C-T-C-C-A-A-T 3 3 A-T-C-G-G-T-T-G-T-G-T-G-T G-T-G-T-G-T-G-A-G-G-T-T-A 5 5 T-A-G-C-C-A-A-C-A-C-A-C-A . ... C-A-C-A-C-A-C-T-C-C-A-A-T 3 3 A-T-C-G-G-T-T-G-T-G-T-G-T

G-T-G-T-G-T-G-A-G-G-T-T-A 5 5 T-A-G-C-C-A-A-C-A-C-A-C-A-C-A C-A-C-A-C-A-C-A-C-T-C-C-A-A-T 3 3 A-T-C-G-G-T-T-G-T-G-T-G-T-G-T G-T-G-T-G-T-G-T-G-A-G-G-T-T-A 5 5 T-A-G-C-C-A-A-C-A-C-A-C-A-C-A C-A-C-A-C-A-C-A-C-T-C-C-A-A-T 3 3 A-T-C-G-G-T-T-G-T-G-T-G-T-G-T G-T-G-T-G-T-G-T-G-A-G-G-T-T-A 5 Figure 7.32 VNTR procedure. Specific primers to flank regions of tandem repeats are constructed and used for PCR amplification of DNA segments with tandem repeats, for comparis on of different species. length of repeating unit, and in minimum and maximum number of tandem copies occ urring in the DNA of a population. A re-peating sequence of 2-9 bp is often know n as microsatellite or SSLP (Simple se-quence length polymorphism), whereas one of 10-60 bp as minisatellite. If these repeated sequences show variation within a population or a species, they are known as variable number tandem repeats (VNT Rs). At a given locus in different indi-viduals, the length of tandem repeats ma y ers that flank tandem repeats, and then using PCR technology to generate multipl e copies of tandem repeat DNA, whose length can be determined by gel electrophor esis (Figure 7.32). VNTR technology generates data quickly and efficiently and is often used for population studies, for exami ning relationships within a species, or between closely related species. STRP is very useful in mapping, as a large number of alleles present in the popu lation often have high proportion of genotypes that are heterozygous for differe nt alleles. STRPs vary, because of irregularities of crossing over and replication, and as such can be are widely used in DNA typing (DNA fing er-used as genetic marker. Identification of printing) involving identification of human microsatellites involves constructing prim- individuals in criminal inv estigation. 204 Plant Systematics A B C D A B C D as character, and unique allele combina- tions as character states (Figure 7.33) . A comparison electrophoresis bands of four species for enzyme I with two alloz ymes can be coded as 0 for allozyme separated at position 18 and 1 for allozyme separated at position 21. Similarly Enzyme II with three allozymes can be coded as 0 for 27, 1 for 31 and 2 for 35. Enzyme I, as such would be coded as 1 for sp ecies A-C with band at 21 and 0 for D for band at 18. For enzyme II, similarly s pecies A has band at 31 coded as 1, B at 35 coded as 2 and C as 3 having bands a t 31 and 35. Allozyme data can also be coded as loss of each allele as one state and gain as another state. Allozyme data can also be coded on the basis of alle le frequency. A species with two alleles in the frequency of 90/ 10% would be co ded differently from another species with same alleles but with frequency of 40/ 60%. 35 M M M Enzyme M M M Enzyme 31 27

21 18 I II Enzyme Enzyme Species I II A 1 1 B 1 2 C 1 3 D 0 0 Figure 7.33 Allozyme electrophoresis data for two enzymes, enzyme I with two allozymes and enzyme II with three allozymes. The coding of data for four species, of which species D represents an outgroup is p resented below.Each locus is treated as character and combi-nation of character states as charExamples of Molecular studies acter states. Whereas considerable progress has been made in the mapping of chloroplast genome , similar success in nuclear genome is at its infancy. Questions of speciation a re being addressed through genome mapping in Helianthus. Some progress has also been made in grasses and the family Solanaceae. In Helianthus, Riesberg and his co-workers (1996) reported that H. annuus and H. petiolaris differed by at least seven translocations and three inversions, which affected recombination and pos sibilities of introgression. The genome of hybrid derivative H. anomalus, was re arranged relative to both parents, and the species was partially reproductively isolated from both parents. They also created new hybrids between the two parent al species and found that chromosomal rearrangements were similar to the natural ly occurring hybrid species, H. anomalus. Belford and Thomson (1979), us-ing sid e-copy sequence hybridization in Allozyymes Different forms of an enzyme differing in dif-ferent alleles at the same locus c onstitute allozymes, as distinct from isozymes show-ing differences at different loci. Allozymes are separated and detected using starch electrophoresis as agai nst gel electrophore-sis for DNA sequencing. Allozymes differing slightly in ami no acid composition will take different charges and migrate differently, and can be identified using specific stains. Allozymes have traditionally been used to assess genetic variation within a popu lation or a species, but they can also be used for phylogenetic analysis of clos ely related species. Allozyme data can be coded in a variety of Atriplex concluded that division into two subways. Each allele may be coded as a charac- genera in this genus is not correct. ter and its presence or absence as charac-ter states. Alternately a locus may be treated Bayer et al., (1999) on the basis of sequence analyses of the plastid atpB and r bcL Taxonomic Evidence 205 (A) Rice (B) Wheat (C) Maize

(D) Fox tail (E) Sugar cane (F) Sorghum (G) Ancestral millet cereal R1a W1 W2 R4a R7 R4b W3 W4 W5 W6 W7 R6a R8 F R1a R5a R5b H B C G A D I E F I H A C R10 R3a R3b G R12a R12b R11b D B S1 S2 S3 RRRRRRR8633361cb0aab S4 S5 S8 S6

S7 S9 S10 R4a R R1b R2 R3a R3b R3c R4a R4b R5a R5b R6a R6b R7 R9 R10 R11a R11b R12a R12b I 8 b 12 3 c 4 Rice

II Figure 7.34 Grass genome evolution. I: Conserved linkages (synteny groups) betwe en the rice genome and other grass species. A: Rice genome with chromosomes divided into blo cks of linked genes; B: Wheat genome with chromosomes showing correspondance wit h rice segments; C: Maize genome with duplicated blocks indicating ancient tetraploidy; D: Foxtail millet genome; E: Sugar cane genome; F: Sorghum genome; G: Inferred or reconstructed order of segments in a hypothetical ancestral creal geno me cosisting of a single chromosome pair. II. Circular arrangement of synteny gr oups in above grasses. Thin dashed lines indicate connections between blocks of genes. (After Moore et al., 1995). 206 Plant Systematics DNA, found a support for an expanded order Malvales, including most of the gener a pre-viously included in Sterculiaceae, Tiliaceae, Bombacaceae and Malvaceae. T hey propose to merge Sterculiaceae, Tiliaceae and Bombacaceae with Malvaceae and subdivide this enlarged family Malvaceae into nine subfamilies based on molecul ar, morphologi-cal and biogeographical data. Because of the synteny groups in the genomes, homologous genes can often be iden tified by location alone. It must, however, remembered that the circular diagram is only for convenient representation; there is not indication that the ancestr al grass chromosome was actually circular. It was a normal linear chromosome. Grass Genome A large number of workers have targeted the family Poaceae using different crite ria and techniques. All molecular phylogenies point to the Stipeae to be an earl y-diverging lineage. The morphological characters of the Stipeae are thus a mixt ure of synapomorphies linking them with pooids and symplesiomorphies, which they

share with many other grasses. The studies based on chloroplast gene: cpRFLP (D avis and Soreng, 1993), ndhF sequences (Catalan et al., 1997), and nuclear genes : through ITS (Hsiao et al., 1994), phytochrome b (Mathews and Sharrock, 1996), and granule bound starch synthase I (Mason-Gamer et al., 1998) all supported the same placement of Stipeae. Similar studies of comparison of results from chloro plast DNA and nuclear DNA in Triticeae, however, produced different results, alt hough two chloroplast phylogenies constructed from RFLP (Mason-Gamer and Kellog, 1996) and rpoA sequences (Petersen and Seberg, 1997) produced similar results. Genome analysis of cereal grasses has provided useful information. Of the common cereal grasses, rice has the smallest ge-nome (400 mb). Maize genome is 2500 mb , whereas the largest genome is found in wheat (17,000 mb). In spite of large va ria-tions in chromosome number and genome size, there are a number of genetic an d physical linkages between single-copy genes that are remarkably conserved amid a back-ground of very rapidly evolving repetitive DNA sequences. By comparison of rice chro-mosomes numbered R1 to R12 (Figure 7.34-I), with conserved regions marked in lower case (R1a, R1b, etc.), it is found that con-served regions homol ogous to rice are found in other cereals. The wheat monoploid chro-mosome set is designated W1 through W7. One region of W1 contains single-copy se-quences that are homologous to those in rice segment R5a, another contains single-copy seque nces that are homologous to those in rice segment R10, and still another contain s single-copy sequences homologous to those in rice segment R5b. Each of such co nserved physical and genetic linkages is called a synteny group. It is notable t hat maize genome has repetition of segments, confirming that maize is a complete , very ancient tetraploid with two duplicated genome blocks rearranged relative to each New World Tetraploid Cottons other. Genomic studies in genus Gossypium (Wendel et al., 1995) using isozymes, nuclear ITS sequences, and chloroplast re-striction site analysis, indicated that New o rld diploids are monophyletic, as are the Old World diploids. The New World tetr aploid cottons, including G. hirsutum were formed by allopolyploidy of genomes A (from the Old World) and D (from the New World). It was found that H. hirsutum has a chloroplast de-rived from one of the African species, and it must have acq uired it only about 1-2 mil-lion years ago, well after the formation of the Atla ntic Ocean. Simultaneous comparison of above cereal grass genomes is better represented with the help of a circular diagram (Figure 7.34-II). The segments are arranged into a circle in the same order in which they were aligned in the hypothetical ances tral chromosome. Taxonomic Evidence 207 3. A rapid life cycle (about 6 weeks from germination to mature seed). 4. Prolific seed production and easy cultivation in restricted space. 5. Efficient transformation methods utilizing Agrobacterium tumefaciens. 6. A large number of mutant lines and [YO genomic resources 7. Multinational research community of academic, government and industry laboratories. 8. Easy and inexpensive to grow. 9. Compared to other plants, it lacks the repeated, less-informative DNA sequences that complicate genome analysis. Figure 7.35 Arabidopsis thaliana, small annual The Ara bidopsis Genome Initiative (AGI) is an international collaboration to se quence the genome of the model plant Arabidopsis thaliana. Begun in 1996 with th e goal of completing the genome sequence by 2004, the herb from family Brassicaceae,

whose genome is most completely known among the angiosperms, is genome sequencing was completed at the aptly kno wn as the guinea-pig of end of 2000. Comprehensive information on plant kingdom. Arabidopsis genome is available on the Arabidopsis Genome internet via The Arabidopsis Information Resource (TAIR), which provides a compr e- hensive resource for the scientific community working with Arabidopsis thalia na. TAIR is a collaboration between the Carnegie Institution of Washington Depar tment of Plant Biology, Stanford, California, and the National Center for Genome Resources (NCGR), Santa Fe, New Mexico. Funding is provided Insignificant small crucifer, Arabidopsis thaliana (Figure 7.35), often ignored in the field, holds great promise for opening new frontiers of phylogenetic anal ysis. With its small genome size of 114.5 mbp (as com-pared to 165 mbp in Drosop hila melanogaster and 3000 mbp in humans), the species is the most completely kn own genetically among all flowering plants. During the last 8 to 10 years, Arabi dopsis thaliana has be-come universally recognized as a model plant for such stu dies. Although it is a non-commercial member of the mustard family, it is favore d among basic scientists because it develops, reproduces, and responds to by the National Science Foundation. Important studies on Arabidopsis thaliana have been devoted to genetic control o f de-velopment. Transgenic plants of this species have been created that either overexpress or underexpress cyclin B. Overexpression of cyclin B results in acce lerated rate of cell division; underexpression results in decelerated rate. Plants with faster rate of cell stress and disease in much the same way division contain more cells and are some -as many crop plants. The choice of what larger than their wild type counterArab idopsis as a genetic tool has been forced parts, but otherwise they look complet ely by the following attributes: 1. Small genome (114.5 Mb/125 Mb total). normal. Likewise, plants with the decreased rate of cell division have less than half the normal number of cells, but they grow at al-most the same rate and rea ch almost the 2. Extensive genetic and physical maps of all 5 chromosomes. 208 Plant Systematics Whorl The studies on genetic control of flower development in Arabidopsis have reveale d interesting results. During floral development (as in other tetracyclic plants ), each whorl of the floral parts (sepals, petals, stamens and carpels) arises f rom a separate whorl of initials. Three types of mutations result in three diffe rent phenotypes, one lacking sepals and petals, the second lacking petals and st amens and the third lacking stamens and carpels. Crosses between homozygous orga nisms have resulted in identification of four genetic groups (Table Gene 1 2 4 3 2 1 Activity 3 ag ap3 & pi ap2 Sepal Petal Stamen Carpel Stamen Sepal 7.2). Mutations in the gene ap2 (apetala-2) Figure 7.36 Graphic representation of control result in phenotype without sepals and pet-als. The phenotype lacking petals and

stamens is caused by mutation in either of two genes, ap3 (apetala-3) or pi (pi stillata). The genotype lacking stamens and carpels is caused by mutations in th e gene ag (agamous). Each of these genes has been cloned and sequenced. They are all transcription factors, members of MAD box family of transcription factors, each containing a of floral development in Arabidopsis thaliana by the overlap-ping action of four genes. Gene ap2 is expressed in the outer two whorls (sepals and petals), ap3 a nd pi are expressed in the middle two whorls (petals and stamens) and ag in the inner two (stamens and carpels). Each whorl has a unique combination o f active genes. sequence of 58 amino acids. same size as wild-type plants, because as the number of cells decrease, the indi vidual cells get larger. The plants thus have ability to adjust to abnormal grow th conditions, as opposed to animals which frequently develop proliferative canc er cells. An interesting finding from this study is that mutation in any of the genes elim inates two floral organs belonging to adjacent whorls. The pattern suggests that ap2 is necessary for sepals and petals, ap3 and pi are both necessary for stame ns and ag necessary for stamens and carpels. As mutant phenotypes are caused by loss-of-function in alleles, it may be inferred that ap2 is expressed in whorls 1 and 2, ap3 and pi expressed in whorls 2 and 3, and ag is expressed in whorls 3 and 4. The floral development in this plant is thus controlled by combinational effect of these four genes. Sepals develop from tissue in which ap2 is active; petals by combination of ap2, ap3 and pi, stamens by combination of ap3, pi and ag; and carpels where only gene ag is expressed. This is graphically represented Table 7.2 Floral development in mutants

of Arabidopsis thaliana. Genotype Whorl 1 2 3 4 wildtype sepals petals stamens carpels ap2/ap2 carpels stamens stamens carpels ap3/ap3 sepals sepals carpels carpels sepals sepals carpels carpels sepals petals petals sepals in figure 7.36. pi/pi It is pertinent to remember that ap2 expression and ag expression are mutually e xclusive. In presence of ap2 transcription factor ag is repressed, and in the pr esence of ag transcription factor, ap2 is repressed. ag/ag Taxonomic Evidence 209 Accordingly, in ap2 mutants, ag expression spreads to whorls 1 and 2, and, in ag mu-tants, ap2 expression spreads to whorls 3 and 4. This assumption enables us to explain the phenotypes of single and even double mu-tants. This pattern of ge ne expression has been assayed by in situ hybridization of RNA in floral cells w ith labelled probes for each of the genes. The results confirm the above as-sump tion of repressive action of concerned genes. It is significant that triple muta tion involves all the genes. The phenotype of ap2 pi ag triple mutant does not h ave any nor-mal floral organs. There are concentric whorls of leaves instead. single copy region), for b subunit of ATP synthase (AtpB), ITS region of ribosome,

phytochrome B, and granule bound starch synthaseI. An encouraging congruence of re-sults of these diverse studies was met in tribe Stipeae of grasses. In other cases, re-sults from chloroplast phylogeny and nuclear phylogeny did not agree, suggesting caution in relying on any attribute singly for constructing molecula r phylogenies. The gene trees constructed from rbcL have great utility in angios perms. Chase et al., (1993) at-tempted to yield the phylogeny of all seed plants using 499 rbcL sequences. The analysis proved a few sequences to be pseudogenes , and entire families were represented by single sequences. The data set have be en reanalyzed by other authors to yield parsimonious trees (Rice et al., 1997). RbcL data has supported that Caryophyllidae is monophyletic. It has also support ed the Gene trees Molecular systematics presents powerful tools for constructing phylogenetic tree s. Commonly used methods over the recent years include studies on chloroplast DNA using restriction site polymorphism (cpRFLP), union of family pairs Asclepiadace aeanalysis of chloroplast gene for subunit F of Apocynaceae, Araliaceae-Apiaceae , and NADP dehydrogenase (ndhF, in the small copy Brassicaceae-Capparaceae. The data a lso and b subunits of RNA supported the polyphyletic nature of Saxifraregion), for a polymerase II (rpoA and rpoC2, in a large gaceae and Caprifoliaceae. 210 Plant Systematics Chapter 8 Developing Classifications Systematics aims at developing classifica-tions based on different criteria and, often a distinct methodology is employed for the analysis of data. Data handlin g to establish relationships between the organisms often makes use of one of the two methods: phe-netic methods and phylogenetic methods, often providing differ ent types of classifica-tion. Distinction is sometimes also made be-tween phylog enetic and evolutionary clas-sification schemes. Phylogenetic methods aim at dev eloping a classification based on an analysis of phylogenetic data, and devel-op ing a diagram termed cladogram or phy-logenetic tree, or more recently, simply t ree, which depicts the genealogical descent of taxa. Biologists practicing this methodology are known as cladists, and the field of study as cladistics. The ter m, however, is slowly being replaced by phylogenetic sys-tematics. The phylogene tic concepts present a huge diversity of variation, unfor-tunately often contrad ictory, leading to dif-ferent interpretations of similar results. A brief unders tanding of these is, therefore, necessary before attempting to explore this comp lex field. Before the development of modern methods of cladistics, the numeri-ca l methods were largely used for drawing phylogenetic inferences from the data an aly-sis. The modern Phylogenetic methods, how-ever, integrate the concepts and p ractices of numerical taxonomy with cladistic methods. It is, however, essential to under stand the concepts of each, and the final integration in phylogeny reconstructio n. PHENETIC METHODS Numerical taxonomy received a great impetus with the development and advancement of computers. This field of study is also known as mathematical taxonomy (Jardi ne and Sibson, 1971), taxometrics (Mayr, 1966), taximetrics (Rogers, 1963), mult ivariate morphometrics (Blackith and Reyment, 1971) and phenetics. The modern me thods of numerical taxonomy had their beginning from the contributions of Sneath (1957), Michener and Sokal (1957), and Sokal and Michener (1958) which culminat ed in the publication of Principles of Numerical Tax-onomy (Sokal and Sneath, 19 63), with an expanded and updated version Numerical Tax-onomy (Sneath and Sokal, 1973). The latter authors define Numerical taxonomy as grouping by numerical me

thods of taxonomic units into taxa on the basis of their character states. Befor e the development of modern methods of cladistics, the numerical methods were al so used for drawing phylogenetic inferences from the data analysis. The last few decades have witnessed a forceful debate on the suitability of the Developing Classifications 211 empirical approach or operational approach in systematic studies. Empirical taxo nomy forms the classification on the basis of taxo-nomic judgment based on obser vation of data and not assumptions. Operational taxonomy, on the other hand, is based on operational methods, experimentation to evaluate the observed data, bef ore a final classification. Numerical taxonomy finds a balance be-tween the two as it is both empirical and Empirical

Operational I NuTr1~r1C~1 operational (Figure 8.1). It must be remembered that numerical taxonomy does not produce new data or a new system of classification, but is rather a new method of organizing data that co uld help in better understanding of relationships. Spe-cial classifications are based either on one or a few characters or on one set of data. Numerical taxonom y seeks to base classifications on a greater number of characters from many sets of data in an effort to produce an entirely phenetic classification of maximum predictivity. Figure 8.1 Relationship between empirical, operational and numerical taxonomy (after Sneath and Sokal, 1973). the better a given classification will be. 2. A priori, every character is of equal weight in creating natural taxa. 3. Overall similarity between any two enPrinciples of Taxometrics tities is a function of their individual similarities in each of the many charac ters in which they are being comThe philosophy of modern methods of nu-merical taxonomy is based on ideas that w ere first proposed by the French naturalist Michel Adanson (1763). He rejected t he idea of giving more importance to certain char- acters, and believed that nat ural taxa are based on the concept of similarity, which is measured by taking al l the characters into consideration. The principles of modern numerical taxonomy developed by Sneath and Sokal (1973) are based on the modern inter-pretation of the Adansonian principles and as such are termed neo-Adansonian prin-ciples. It would, however, be wrong to visu-alize Adanson as the founder of numerical taxo nomy, because he worked in a different academic environment from that of today, when tools of investigation were much dif-ferent. These principles of numerical tax-onomy are enumerated below. pared. 4. Distinct taxa can be recognized because correlations of characters differ in the groups of organisms under study. 5. Phylogenetic inferences can be made from the taxonomic structures of a group and also from character correlations, g iven certain assumptions about evolutionary pathways and mechanisms. 6. Taxonomy is viewed and practiced as an empirical science. 7. Classifications are based on phenetic similarity. 1. The greater the content of informa-

The methodology of numerical taxonomy involves the selection of operational unit s (populations, species, genera, etc., from which the information is collected) and char-acters. The information from these is recorded, and similarity (and/or distance) between units is determined using various tion in the taxa of a classification and the more characters it is based upon, 212 Plant Systematics statistical formulae. The ultimate analysis involves comparison of similarity da ta and constructing diagrams or models, which pro-vide a summary of the data ana lysis. These diagrams or models are used for final syn-thesis and better underst anding of the rela-tionships. The major advantages of numeri-cal taxonomy over c onventional taxonomy include: fined phenetic relationship as an arrange-ment by overall similarity, based on a ll available characters without any weight-ing. Sneath and Sokal (1973) define p henetic relationship as similarity (resemblance) based on a set of phenotypic ch aracteristics and not phylogeny of organ-isms under study. It is distinct from a cladistic relationship, which is an expression of the recency of common ancestr y and is represented by a branching network of an-cestor-descendant relationship s. Whereas the phenetic relationship is represented by a phenogram, the cladisti c relationship is depicted through a cladogram. 1. Numerical taxonomy has the power to integrate data from a variety of sources such as morphology, physiol-ogy, phytoc hemistry, embryology, anatomy, palynology, chromosomes, ultrastructure and micro morphology. This is very difficult to do by conventional taxonomy. 2. Considerable automation of the data CLADISTIC METHODS Although phylogenetic diagrams (now appropriately known as phylograms) have been used by Bessey (1915), Hutchinson (1959, 1973), and contemporary authors of cla ssification systems to depict the relationships between taxa, the cladograms are distinct in the sense that they are developed using a distinct methodology. Thi s method was first proposed by W. Hennig (1950, 1957), a German zoologist who fo unded the subject of phylogenetic systematics. The term cladistics for this meth odology was coined by Mayr (1969). An American Botanist, W. H. Wagner, working i ndependently, developed a method of constructing phylogenetic trees, called the groundplan-divergence method, in 1948. Over the years, cladistics has developed into a forceful methodology of developing phyloprocessing promotes efficiency and the work can be handled by even less skilled workers. 3. Data coded in numerical form can be integrated with existing data-process-ing systems in various institutions and us ed for the creation of descrip-tions, keys, catalogues, maps and other documents. 4. The methods, being quantitative, provide greater discrimination along the spectrum of taxonomic differences, and can provide better classifications and keys. 5. The creation of explicit data tables for numerical taxonomy necessitates the use of more and better described char-acters , which will necessarily improve conventional taxonomy as well. genetic classifications. 6. The application of numerical taxonomy Cladistics is a methodology that attempts to analyse phylogenetic data objective ly, in a manner parallel to taxometrics, which analyses phenetic data. Cladistic methods are largely based on the principle of parsimony according to which, the most likely evolutionary route is the shortest hypothetical pathway of changes that explains the pattern under observation. Taxa in a truly phylogenetic system should be monophyletic. It has been found that symplesiomorphy (possession of p rimitive or plesiomorphic

has posed some fresh questions concerning classification and initiated efforts for re-examination of classification systems. 7. A number of biological and evolutionary concepts have been reinterpreted, thus introducing renewed interest in biological research. Numerical taxonomy aims at determin-ing phenetic relationships between organ-ism s or taxa. Cain and Harrison (1960) deDeveloping Classifications 213 character-state in common by two or more taxa) does not necessarily indicate mon ophy-ly. Synapomorphy (possession of derived or apomorphic character-state in co mmon by two or more taxa), on the other hand, is a more reliable indicative of m onophyly. It is thus common to use homologous shared and derived character-state s for cladistic stud-ies. Before analysing the methodology of han-dling data for phylogenetic analysis, it is im-portant to understand the major terms and conce pts used in Phylogenetic Systematics. These families are primitive because they possess primitive characters (or charac terstates) and primitive characters (or character-states) are those which are po ssessed by these primitive families . Over the recent years, a better understandin g of these con-cepts has become possible. It is generally accepted that evolutio n has proceeded at different rates in different groups of plants so that among t he present-day organisms, some are more advanced than others. The first step in the determination of relative advancement of characters, is to ascertain which c haracters are plesiomorphic and which are apomorphic. Stebbins (1950) ar- gued t hat it is wrong to consider the charac- ters as separate entities, since it is t hrough the summation of characters peculiar to an individual, that natural selec tion operates. Sporne (1974) while agreeing with this, believed that it is scarc ely possible initially to avoid thinking in terms of separate characters, which can be treated better statistically. Given insufficient fossil records of the ea rliest angiosperms, comparative morphology has been largely used to decide the r elative advancement of characters. Many doctrines have been proposed but unfortu nately most rely on circular reasoning. Some of the Phylogenetic Terms Many important terms have been repeatedly used in discussions on the phylogeny o f an-giosperms, with diverse interpretation, which has often resulted in differe nt sets of conclusions. A prominent case in point is Melville (1983), who regard s the angiosperms as a monophyletic group. His justification several ancestral fo rms of the single fossil group Glossopteridae gave rise to an-giosperms renders hi s view as polyphyletic in the eyes of the greater majority of authors who believ e in the strict application of the concept of monophyly. The involvement of more than one ancestor makes angiosperms a polyphyletic group, a view that has been firmly rejected. A uniform thorough evalua-tion of these concepts is necessary f or proper understanding of angiosperm phylogeny. important doctrines are described below: Plesiomorphic and Apomorphic Characters The Doctrine of conservative regions holds that certain regions of plants have b een less susceptible to environmental in-fluence than others and, therefore, exh ibit primitive features. Unfortunately, however, over the years, every part of t he plant has been claimed as conservative region. Also, the assumption that a fl ower is more con-servative than the vegetative parts is derived from classificat ions which are based on this A central point to the determination of the phylogenetic position of a particula r group is the number of primitive (plesiomorphic) or advanced (apomorphic) char acters (al-though the term character is often used broadly in literature, more a ppropriately primitive or advanced and similarly plesiomorphic and apomorphic re fer to dif-ferent character-states of a character, and not different characters) that the group con-tains. In the past, most conclusions on primi-tiveness were based on circular reasoning: assumption.

The doctrine of recapitulation holds that early phases in development are suppos ed to exhibit primitive features, i.e. ontogeny repeats phylogeny . Gunderson (1939 ) used this theory to establish the following evolutionary trends: polypetaly to gamopetaly (since the petal primordia are initially separate, the tubular porti on of the corolla arises later); polysepaly to gamosepaly; 214 Plant Systematics actinomorphy to zygomorphy and apocarpy to syncarpy. The concept originally appl ied to animals does not always hold well in plants where ontogeny does not end w ith embryog-eny but continues throughout the adult life. Neoteny (persistence of juvenile features in mature organism) is an example wherein a persistent embryo nic form represents an adthat certain morphological characters are statistically correlated and the fact can be used in the study of evolution. Sinnot and Bailey (1914) demonstrated a p ositive correlation between trilacunar node and stipules. Frost (1930) believed that correlation between characters arises because rates of their evolution have been correlated. Sporne (1974) has, however, argued that correlation can be sho wn to occur even though the rates of evolution of characters are not the same. W ithin any taxonomic group, primitive characters may be expected to show positive correlation merely because their distribution is not random. By definition, pri mitive members of that group have retained a relatively high proportion of an-ce stral (plesiomorphic) characters, while ad-vanced members have dispensed with a relatively high proportion of these same char-acters either by loss or replacement with different (apomorphic) characters. It follows, therefore, that the distrib ution of plesiomorphic characters is displaced to-wards primitive members, which have a higher proportion of plesiomorphic characters, than the average for the group as a whole. Departure from the random can be statistically calculated in o rder to establish correlation among characters. Based on these calculations, Spo rne (1974) prepared a list of 24 characters in Dicotyledons and 14 in Monocotyle dons, which exhibit positive correlation. These characters, because of their dis tribution, have been categorized as magnoloid and amarylloid, respectively. Base d on the distribution of these characters, Sporne calculated an advancement inde x for each family and projected the placement of different families of an-giospe rms in the form of a circular diagram, with the most primitive families near the centre, and the most advanced along the periphery. That the earliest members of an-giosperms are extinct is clear from the fact that none of the present-day fa milies has the advancement index of zero. All living vanced condition. The doctrine of teratology was advocated by Sahni (1925), who argued that when a normal equilibrium is upset, an adjustment is often effected by falling back up on the surer basis of past experience. Thus, tera-tology (abnormality) is seen a s reminiscent of some remote ancestor. According to Heslop-Harrison (1952), some teratological phenomena are just likely to be progressive or retrogressive, and each case must be judged on its own merit. The doctrine of sequences advocates that if organisms are arranged in a series i n such a way as to show the gradation of a par-ticular organ or structure, then the two ends of the series represent apomorphy and plesiomorphy. The most crucia l decision, however, is from which end should the series be read. The doctrine of association advocates that if one structure has evolved from another, then the primitive condition of the derived one will be similar to the ge neral condition of the ancestral structure. Thus, if vessels have evolved from t racheids, then the vessels similar to tracheids (vessels with longer elements, s maller diameter, greater angularity, thinner walls and oblique end walls) repres ent a more primitive condition than vessels with broader, shorter, more circular elements with horizontal end walls. The doctrine of common ground plan ad-vocates that characters common to all members of a group must have been possessed by the original ancestor and must, ther efore, be primitive. The doctrine, however, cannot be applied to angiosperms in

which there is an exception for almost every character. The doctrine of character correlation was acknowledged during the second decade of the previous century when it was realized families have advanced in some respects. The concept of apomorphic and plesiomorphic characters in understanding Developing Classifications 215 the phylogeny of angiosperms has been con-siderably advanced with the recent dev elop-ment of cladistic methods. These employ a distinct methodology, somewhat si milar to taxometric methods in certain steps in-volved, leading to the construct ion of cla-dograms depicting evolutionary relation-ships within a group. Certain groups of an-giosperms are reported to have a combina-tion of both plesiomorphi c and apomorphic characters, a situation known as heterobathmy. Tetracentron has primitive vesselless wood but the pollen grains are ad-vanced, being tricolpate . homologous with leaves because their development is identical. During the latter half of the present century, phylogenetic interpretation has b een applied to these terms. Simpson (1961) defined homology as the resemblance d ue to inheritancef rom a common ancestry. Anal-ogy, similarly, represents functi onal similarity and not due to inheritance from a common fined homology as the occurrence of similar features in two or more organisms, w hich can be traced to the same feature in the common ancestor of these organisms . It is, as such, imperative that homology between two organisms can result only from their having evolved from a common ancestor, and the ancestor must also co n-tain the same feature or features for which try. Mayr (1969) similarly deHomology and Analogy Different organisms resemble one another in certain characters. Taxonomic groups or taxa are constructed based on overall resem-blances. The resemblances due to homology are real, whereas those due to analogy are generally superficial. A re al understanding of these terms is, thus, necessary in order to keep organisms w ith superficial resemblance in separate groups. The two terms as such play a ver y important role in underthe two organisms are homologous. Wiley (1981) has provided a detailed in-terpretation of these terms. Homology ma y either be between two characters, two char-acter states, or between two organi sms for a A 0 B tirti standing evolutionary biology. These terms were first used and defined by Owen (1848). He defined Homology as t he occurrence of the same organ in difA

C B tirti ferent animals under every variety of Figure 8.2 Homology between characters (or forms and functions. He defined Analogy as the occurrence of a part or an organ in one animal which has the same func-tion as another part or organ in a differ -ent animal. If applied to plants, the rhizome of ginger, the corm of colocasia, tuber of potato, and runner of lawn grass are all homologous, as they all repre

sent a stem. The tuber of potato and the tuber of sweet potato, on the other han d, are analogous as character states). In the first ex-ample, character A is plesiomorphic and B is apomorphic. In the second example, B is apomorphic in relation to A but plesiomo rphic in relation to C as all three belong to an evolutionary transformation ser ies. the latter represents a root. particular character or character state. Two characters (or character-states) ar e homologous if one is directly derivedf rom the other. Such a series of charact ers is called an evolutionary transformation series (also called morphoclines or phenoclines). The original, pre-existing character (or character- state) is ter med plesiomorphic and the derived one as apomorphic or evolutionary novelty. Darwin (1959) was the first to apply these terms to both animals and plants. He defined homology as that relationship between parts which results from their dev elop-mentf rom corresponding embryonic parts. The parts of a flower in different plants are thus homologous and these, in turn, are 216 Plant Systematics Three or more character-states may be homologous if they belong to the same evolutionary transformation series (ovary su-perior >half-inferior > inferior). The ter ms plesiomorphic and apomorphic are, however, relative. In an evolutionary trans formation series representing characters A, B and C 4. When the same relatively simple character is found in a large number of species, it is probably homologous in all th e species. Sets of characters may similarly be homologous. 5. If two organisms share the characters (Figure 8.2), B is apomorphic in relation to A of sufficient complexity and judged homologous, other characters shared by the o rganisms are also likely to be homologous. but it is plesiomorphic in relation to C. Two or more organisms may be homologous for a particular character (or charac-ter-state) if their immediate common ancestor also had this character. Such a character is called shared homologue. I f the character-state is present in the immediate common ancestor, but not in th e earlier ancestor (Figure 8.3), i.e. the character-state is a derived one, the situation is known as synapomorphy. If the character-state is present in the imm ediate common ancestor, as well as in the earlier ancestor, i.e. it is an origin al character-state, the situation A B C A B C is known as symplesiomorphy (note sym-). The homology between different organ-isms is termed special homology, as repre-s ented by different types of leaves in differ-ent species of plants. Different le aves in the same plant such as foliage leaves, bracts, floral leaves would also be homologous, representing serial homology. The following criteria may be helpf ul in identifying homol-ogy in practice: Figure 8.3 Homology between two organisms B and C. In diagram I, similarity is due to symplesiomorphy as the character was unchanged in the previous ancestor. In II, it is due to synapomorphy as the pre vious ancestor had a plesiomorphic char-acter and the two now share a derived ch aracter. 1. Morphological similarity with respect Parallelism and convergence to topographic position, geometric po-sition, or position in relation to other p arts. A branch, for example, occurs in the axil of a leaf, although it may

Unlike homology, if the character shared by two organisms is not traced to a com mon an-cestor, the similarity may be the result of homoplasy (sometimes consider ed synonym of analogy). It can result in three different ways. One, the organism s have a common ancestor but the character-state was not present in their common ancestor (parallelism). It could also result from two different characters in d ifferent ancestors evolving into identical character-states (convergence). Simil arity could also arise from loss of a particular character (reversal), thus reve rting to ancestral condition (loss of perianth in some families). All the three situ-ations represent false synapomorphy be modified in different ways. 2. Similar ontogeny. 3. Continuation through intermediates, as for example, the evolution of mam-malian year from gills of fishes, evo-lutio n of achene fruit from follicle in Ranunculaceae. Similarly, vessels having evol ved from tracheids, the primitive forms of vessels are more like tracheids, with elongated nar-rower elements with oblique end walls. Developing Classifications 217 because the similar character-state is dewind pollination in such unrelated families as Poaceae, Salicaceae and Urticaceae, pollinia in Asclepiadaceae rived and not traced to a common ancestor. Simpson (1961) defined parallelism as the independent occurrence of similar chan ges in groups with a common ances-try, and because they had a common an-cestry. The two species Ranunculus tripartitus and R. hederacea have a similar aquatic h abit and dissected leaves and have acquired these characters by parallel evolu-t ion. The development of vessels in Gnetales and dicotyledons also represents a c ase of and Orchidaceae. 3. Convergence may also be due to similar modes of dispersal, as seen in hairy seeds of Asteraceae, Asclepiadaceae and some Malvaceae. 4. Convergence commonly occurs beparallelism. tween relatively advanced members of respective groups. Arenaria and Minuartia f orm natural groups of species which were earlier placed within the same genus Ar enaria. The two species Arenaria leptocladus and Minuartia hybrida show more sim ilarity than between any two species of these two genera. If the similarity is p atristic (result of common ancestry), then the two species would represent the m ost primitive members of respective groups (Figure 8.5-I) and it would have been advisable to place all of the species in the same genus Arenaria. The Convergence implies increasing similar-ity between two distinct phyletic lines, ei-ther with regard to individual organ or to the whole organism. The similar fe a-tures in convergence arise separately in two or more genetically diverse and n ot closely related taxa or lineages. The similarities have arisen in spite of la ck of affinity and have probably been derived from different sys-tems of genes. Examples may be found in the occurrence of pollinia in Asclepiadaceae and Orchid aceae, and the switch habit (circular sheath at nodes) in Equisetum, Ephe-dra and Polygonum. The concepts of parallelism and convergence are illustrated in A B C A B C Figure 8.4. Convergence is generally brought about by similar climates and habitats, similar methods of pollination or dispersal. Once the convergence has been identified b etween two taxa, which have been grouped together, they are separated to make th

e groups natu-ral and monophyletic. The following criteria may help in the ident ification of converI 1Jl gence: 1. Convergence commonly results from adaptation to similar habitats. Wa-ter plants thus usually lack root hairs and r oot cap but contain air lacunae. Annuals are predominant in deserts, which also have a good number of suc-culent plants. The gross similarity between certain su cculent species of Euphorbiaceae and Cactaceae is a Figure 8.4 Examples of convergence (I) and parvery striking example of convergence. allelism (II) between organisms A and B. In convergence, similarity is between o rganisms derived from different lineages. In parallelism, the ancestor is common but both A and B have evolved an apomorphic character independently. In both ca ses, similarity represents false synapomorphy. Dissimilarity be-tween B and C in both diagrams is due to divergence. 2. Convergence may also result from similar modes of pollination such as 218 Plant Systematics F G H A B C D E studies have shown, however, that these two species are the most specialized in each group (Figure 8.5-II) and thus show convergence. Separation of the two gene ra is justified, because placing all the species within the same genus Arenaria would render the group polyphyletic, a situation that

evolutionary biologists avoid. m It is pertinent to mention that although the concepts parallelism and convergenc e seem to be distinct and theoretically sound, and often easy to apply when disc ussing homoplasious (non-homologous) similarity in the case of closely related o rganisms (parallelism), or distantly related organisms (convergence), the distin ction is not always clear. In Figure 8.5-I, for example if we did not know the e volutionary history of the group before level m, there was no way of telling whe ther all the eight species had a common ancestor or not. For practical reasons, it is always safer to refer homoplasious situations together. Some recent author s like Judd et al., (2002) treat parallelism and convergence F G H A B C D E as same. II Reversal is a common evolutionary process, wherein loss of a particular characte r may lead to apparent similarity with ancestral condition. The occurrence of re duced unisexual flowers without perianth or with reduced perianth in Amentiferae

was once considered to be primitive situation, but the evidence from wood anato my, floral anatomy and palynology have shown that apparent simplicity of these f lowers is due to evolu- tionary reduction (reversal), and as such the assumed si milarity to angiosperm ancestral condition is representation of homoplasy, a fal se similarity between an evolutionary advancement (secondary reduction) and ance stral simple condition. Figure 8.5 Two possible reasons for similaribetween species A and B. In (I), A (cf. Arenaria leptocladus) and B (cf. Minuar tia hybrida) are the most primitive members of respective lin-eages FGHA (cf. Ar enaria) and BCDE (cf. Minuartia). The two lineages have common ancestry and thus consti-tute a single monophyletic group (cf. Arenaria s. l.). In (II), A and B happen to be the most advanced members of the respective groups, the two lineage s are distinct and as such similarity between A and B is superficial due to conv ergence, justifying the independent recogni-tion of two lineages (cf. distinct g en-era Arenaria and Minuartia). Monophyly, Paraphyly and Polyphyly These terms have been commonly used in taxonomy and evolutionary literature with such varied interpretation that much con-fusion has arisen in their application . Developing Classifications 219 A B C D A B C D more lineages from one immediately ancestral taxon of the same or lower rank. Su ch a definition would be true if, say, genus B evolved from genus A through one species of the latter, since in that case, the genus B would monophyletic at the same rank (genus) as well as at the lower (species) rank. On the other hand, if genus B evolved from two species of genus A, it would be monophyletic at the ge nus level but polyphyletic at the lower rank. A B C D Most authors, however, including HeslopHarrison (1958) and Hennig (1966), adhere to a stricter interpretation of monophyly, namely the group should have evolved from a single immediately ancestral species which, may be considered as belongi ng to the group in question. There are thus two different levels of monophyly: a minimum monophyly wherein one supraspecific taxon is derived from another of eq ual rank (Simpson s definition), and a strict mono-phyly wherein one higher taxon is derived Figure 8.6 Concepts of monophyly, paraphyly and polyphyly. In (I) groups AB and CD are monophyletic as each has a common anc estor at level m. Similarly. group ABCD is monophyl-etic as it has a common ance stor at level n. In (II) group ABC is paraphyletic as we are leaving out descend ant D of the common an-cestor at level n. In (III) group BC is polyphyletic as t heir respective ancestors at level m do not belong to this group. from a single evolutionary species. Mayr (1969) and Melville (1983) follow the concept of minimum monophyly. Most au -thors, including Heslop-Harrison (1958), Hennig (1966), Ashlock (1971) and Wile y (1981), reject the idea of minimum monophyly. All supraspecific taxa are compo sed of individual lineages that evolve independent of each other and cannot be a ncestral to one another. Only a species can be an ancestor of a taxon. The supra specific ancestors and, for that matter, supraspecific taxa are not biologically

meaningful entities Defined broadly, the terms monophyly (deri-vation from a single ancestor) and po lyphyly (derivation from more than one ancestor) would have different meanings d epending upon how far back we are prepared to go in evolutionary history. If lif e arose only once on Earth, all organisms (even if you place an animal species a nd a plant species in the same group) are ultimately monophyletic in origin. The re is thus a need for a precise definition of these terms, to make them and are only evolutionary artifacts. meaningful in taxonomy. Hennig (1966) defined a monophyletic group as a group ofs pecies descendedf rom a single ( stem ) species, and which in-cludes all the descendants from this species . Briefly, a monophyletic group comprises all the descendants that at one time b elonged to a single species. A useful analysis of Hennig s concept of monophyly wa s made by Ashlock (1971). He distinguished between two types of monophyletic gro ups:: those that are holophyletic when all descendants of the most recent common ancestor are con-tained in the group (monophyletic sensu Simpson (1961) defined monophyly as the derivation of a taxon through one or 220 Plant Systematics

iim~, IluuuuW Monophyly Paraphyly Polyphyly Figure 8.7 The application of cutting rules to distinguish between monophyly, pa raphyly and polyphyly. The group is represented by lighter portion of the tree. Monophyletic group can be separated by a single cut below the group, a paraphyletic group by one cut below the group and one or more higher up. A polyphyletic is separated by more than one cut below the group. A monophyletic group represents one comple te branch, a paraphyletic group one larger portion of the branch; whereas the po lyphyletic group represents more than one pieces of a branch (based on Dahlgren et al., 1985). Hennig) and those that are paraphyletic and do not contain all descendants of th e most recent common ancestor of the group. A polyphyletic group, according to h im, is one whose most recent ancestor is not cla-distically a member of that gro up. The terms holophyletic and monophyletic are now considered synonymous. Diagr ammatic representations of Ashlock s concept of polyphyly, monophyly and paraphyly is pregroup, i.e., it represents more than one piece of a branch. sented in Figure 8.6. Gerhard Haszprunar (1987) introduced the term orthophyletic while discussing the phylogeny of Gastropods. An orthophyletic group is a stem group, i.e. a group t hat is paraphyletic because a single clade (the crown group), has been excluded. The term has not been followed in other groups, especially in botanical systema tics. Sosef (1997) compares the existent hierarchical models of classification. He argues that a phylogenetic tree can be subdivided according to a monophyletic hierarchical model, in which only monophyletic units figure or, according to a L innaean hierarchical model, in which both mono- and paraphyletic units occur. Mos t present-day phylogeneticists try to fit the monophyletic model within the set of nomenclatural conventions that fit the Lin-naean model. However, the two mode ls are intrinsically incongruent. The monophyletic model requires a system of cl assification of its own, at variance with currently accepted conventions. Since, however, the monoAn excellent representation of mono-phyly, paraphyly and polyphyly is presented by cutting rules , devised by Dahlgren and Rasmusen (1983). The distinction is base

d on how the group is separated from a repre-sentative evolutionary tree (Figure 8.7). A monophyletic group is separated by a single cut below the group, i.e. i t repre-sents one complete branch. A paraphyletic group is separated by one cut below the group and one or more cuts higher up, i.e. it represents one piece of a branch. A poly-phyletic group, on the other hand, is sepa-rated by more than o ne cut below the Developing Classifications 221 phyletic model is unable to cope with reticu-late evolutionary relationships; it is unsuited for the classification of nature. The Linnaean model is to be prefe rred. This renders the acceptance of paraphyletic a polyphyletic group, because they are derived from two separate ancestors at le vel m. If, however, A, B, and C are under one group, B and C would still now be components of a paraphyletic group, because one descendant of the common ancesto r at level n is kept out of the group. A natural group would be one, which inclu des all descendents of the common ancestor, or the group is monophyletic. supraspecific taxa inevitable. As is true for the distinction between par-allelism and convergence, similarly, the con-cepts of paraphyly and polyphyly (both of which are rejected by modern p hylogenetic systematics while constructing classification), hold good, when the former is applied to a group of closely related organisms and latter to disPhylogenetic Diagrams The affinities between the various groups tantly related organisms. The concepts of plants are commonly depicted with the become ambiguous when a small group of help of diagrams, with several innovation s. organisms is considered. In Figure 8.6-III, These diagrams also help in under standing taxa B and C if brought together would form the classification of included taxa. An Era Years ago Period (million) Tertiary 65 Cretaceous 146 Jurassic 208 235 Triassic Permian 280 Carboniferou 345 Devonian 395 430 Silurian 1 Ordovician 500 Silurian 570 Figure 8.8 A phylogenetic tree representing the evolutionary history of plants i ncluding angiosperms. The vertical axis represents the geological time scale. Only extant (liv ing) plants are shown reaching the top. 222 Plant Systematics understanding of these terms is necessary for a correct interpretation of putati ve relationships. These branching diagrams are broadly known as dendrograms. Any dia-

gram showing the evolutionary history of a group in the form of branches arising from one or more points has often been referred as a phylogenetic tree, but the use of terms is now becoming more precise, and more innovative diagrams are bei ng developed often providing useful information about different taxa mapped in the diagram. The most common form of diagram is one where the length of branch indicates the degree of apomorphy. Such diagrams were sometimes classified as cladograms (Stac e, 1980), but the term has now been restricted to diagrams constructed through t he dis-tinct methodology of cladistics (Stace, 1989). Diagrams with vertical axi s representing the degree of apomorphy are now more appro-priately known as phyl ograms. The earli-est well-known example of such a phylogram is Bessey s cactus (see Fig 10.11). In such diagrams the most primitive groups end near the base and th e most advanced reach Figure 8.9 Phylogenetic tree of angiosperms presented by Dahlgren (1975) with a section of the top (subsequently named phylo genetic shrub by Dahlgren, 1977). the farthest distance. history. The vertical axis in such a diagram represents the geological time scal e. In such a diagram, the origin of a group is depicted by the branch diverging from the main stock and its disappearance by the branch termination. Branches re presenting the fossil groups end in the geological time when the group became ex tinct, whereas the extant plant groups extend up to the top of the tree. Hutchinson (1959, 1973) presented his phylogram in the form of a line diagram (f ig-ure 10.13). The recent classifications of Takhtajan (1966, 1980, 1987) and Cr onquist (1981, 1988) have more innovative phylograms in which the groups are depicted in the form of balloons or bubbles As already mentioned, the relative adv ancewhose size corresponds to the number of ment of the living groups is indicated b y their species in the group (an approach also found distance from the centre, p rimitive groups in Besseyan cactus). Such phylograms thus being near the centre, and advanced gr oups not only depict phylogenetic relationships between the groups, they also show th e de-gree of advancement as also the relative number of species in different gro ups. Such towards the periphery. A phylogenetic tree representing possible relationships a nd the evolutionary history of seed plants is presented in Figure 8.8. diagrams have been popularly known as Dahlgren (1975) presented the phylogebubble diagrams. The bubble diagram of netic tree (preferred to call it phylogen etic Takhtajan (Figure 10.16) is more detailed shrub in 1977) of flowering plants with all and shows the relationship of the orders extant groups reaching the top , and the within the bubble ; as mentioned earlier, cross-section of the top of the phylogenetic Woodland (1991) aptly described it as tree was shown as top plane of this diagra m Takhtajan s flower garden . (Figure 8.9). In subsequent schemes of The phylogenetic tree is a commonly used diagram in relating the phylogenetic Dahlgren (1977, 1983, 1989), the branching portion of the diagram was dropped an d only Developing Classifications 223 Figure 8.10 Mapping of pollen grain dispersal stage in different dicotyledons on a two-dimensional diagram (Dahlgrenogram) of Dahlgren, representing transverse section through the top of a phylogenetic shrub. Pollen grain dispersal in 2-celled stage (unsha ded), 3-celled stage (dotted), or mixed (hatched). (Courtesy Gertrud Dahlgren). the top plane (cross-section of the top) pre-sented as a two-dimensional diagram

(Figthetical pathway of changes within a group that explains the present phenetic pa ttern, using the principle of parsimony. A cladogram is a representation of the inferred historical connections between the entities as evidenced by synapomorph ies. The vertical axis of the cladogram is always an implied, but usually non-ab solute time scale. Cladograms are ancestor-descendant sequences of populations. Each bifurcation of the cladogram represents a past speciaure 8.10), and this has been very useful in mapping the distribution of various charac-ters in different groups of angiosper ms, and the comparison of these provides a good measure of correspondence of var ious char-acters in phylogeny. This diagram has been popularly known as Dahlgrenogram . Thorne s diagram (2000) is similarly the top view of a phylogenetic shrub (Figure 10.23), in which the centre representing the extinct primitive angiosperms, now absent, tion that resulted in two separate lineages. is empty. It must be pointed out, however, that considerable confusion still exists betwee n application of the terms cladogram and phylogenetic tree. Wiley (1981) defines a cladogram as a branching diagram of entities A cladogram represents an evolutionary diagram utilizing cladistic methodology, which attempts to find the shortest hypo224 Plant Systematics Araceae Tofieli3iaceae V.Denotes-h7anch'es w~#h 50-7,0!cisupport HydrocFia7itaceae ^.Denofes hranrhes i w0i c5Qft:support; Oth'eir hranr es:With 6Uti]maCede su0OZrtkgreater;tha6 I ec u a I'',to Alismataceae LinTmOChaF.itacea CS C f1 CIC W2FeT12 1 C Sae 4p6nogetonacea_e ]uncaq ha'ceae * Posidoniaceae RUppia_ceae * i E,ymodoceaceae Z'6stera ce"a e Pota m ggeto n area e Figure 8.11 Tree (cladogram) for different families of the order Alismatales. Support indica ted for branches refers to bootstrap support, discussed in subsequent pages. (Re produced from APweb vesion 7 (June , 2008), with permission from Dr P. F. Steven s.) where the branching is based on inferred historical connections between the enti

changes, depending on which species is ancestral and being relegated lower down on the vertical axis. In the case of higher taxa, the number of cladograms and p hylogenetic trees could possibly be equal, because higher ties as evidenced by synapomorphies. It is, thus, a phylogenetic or historical d endrogram. He defines a phylogenetic tree as a branching diagram portraying hypoth- taxa cannot be ancestral to other higher taxa esized genealogical ties and sequences of since they are not units of evolu tion but historical events linking individual or- historical units composed of s eparately ganisms, populations, or taxa. At the spe- evolving species. cies and population level, the number of pos-sible phylogenetic trees could be more than cladograms for particular character Over the recent years, it has been thus becoming increasingly common to construc t evolutionary diagrams using cladistic Developing Classifications 225 methodology, assuming that these charac-ter-state changes (represented as evolut ionPhylogeny and Classification ary scale or tree length) correspond to the geological time scale, and call thes e evoluThe construction of phylogenetic classification involves two distinct steps: det ermining the phylogeny or evolutionary history of a group, and construction clas sification on the basis of this history. Imagine a lineage (or clade- a group of individuals producing suctionary diagrams as evolutionary tree (Judd et al., 2008), phylogenetic tree, or simply tree (Stevens, 2008), synonymous with a cladogram. A Phenogram is a diagram constructed on the basis of numerical analysis of phe-n etic data. Such a diagram is the result of utilization of a large number of char acters, usually from all available fields, and involves calculating the similari ty between taxa and constructing a diagram through cluster analysis. Such a diag ram (Figure 8.20) is very useful, firstly because it is based on a large number of characters, and secondly because a hierarchical classification can be achieve d by deciding upon the threshold levels of similarity between taxa assigned cessively, similar and genetically related in-dividuals, generally represented b y lines in a cladogram) with woody habit, alternate leaves, cymose inflorescence , 5 red petals, 5 stamens, 2 free carpels, and dry fruit with many seeds. Over a period of time, some population ac-quires herbaceous habit and the original lin eage splits into two, one with woody habit and the second with herbaceous habit (Figure 8.12). In the lineage with woody habit, one linto various ranks. eage emerges with fused carpels, while the other loses one of the two carpels. T he one with fused carpels loses 3 of the five stamens in one or more populations , and that with a single carpel doubles the number of stamens to ten in one or m ore populations. The herbaceous lineage, similarly, splits into one with yellow petals and one with white petals, the former developing fleshy fruits in one or more populations, and the latter having the number of seeds reduced to one in so me populations. The present descendents of the original ancestor are thus repres ented by eight lineages, which have developed a few apomorphic character-states, but also share plesiomorphic character-states such as alter-nate leaves, 5 peta ls, and cymose inflorescence. There must be hundreds of more plesiomorphic state s, but of little significance in classification, as the above three. Note that t he ancestral species at level I, II (woody habit, 2 free carpels), IA (herbaceou s habit, red pet-als), III (herbaceous habit, yellow petals and free carpels and dry fruit), and IV (herbaceous habit, white petals, 5 stamens and many seeds) a nd have disappeared, whereas those at level V, and VI are still represented (although with minor changes) in the form of E and G, respectively. Also note that

united car-pels have arisen twice independently. The same is also true for the l oss of three stamens. It must be pointed out that the modern phylogenetic methods, which aim at con-st ructing phylogenetic trees, also some-times use large number of characters for c omparison, especially when dealing with morphological data, and there seem to be a lot of similarities in data handling and com-putation, but are unique in the utilization of evolutionary markers and, consequently, produce slightly differen t results. With the incorporation of distance methods in the con-struction of tr ees, the classical difference between the terms is largely disappearing. Modern cladistic programs develop trees in which branch lengths are indicated, and plot ting programs offer the choice to indicate branch lengths (and often called phyl ograms) or not. In latter case branches may be square (line running vertically a nd horizon-tally- and often called phenogram; Figure 8:21) or V-shaped (cladogram; Figure 8.11). These may be presented as upright or as horizontal trees (prostrate trees) . Mod ern trees contain information about evolutionary markers such as bootstrap support, branch length, and Bremer support, as dis cussed in subsequent pages. 226 Plant Systematics A B C D E F G H O Fleshy fruit Single seed 2 stamens 10 stam 2 stamens Carpels uni IV VI 1 carpel Carpels united Yellow petals White petals II IA O, A Herbaceous habit I Figure 8.12 Evolutionary history of a hypothetical group of organisms which star ted with the ancestral species with woody habit, alternate leaves, cymose inflorescence, 5 re d petals, 5 stamens, 2 free carpels and dry fruit with many seeds. Eleven charac ter state transformations at different stages have resulted in 8 present day spe cies. Note that two of the changes (carpel union and loss of three stamens) have occurred twice, and as such only nine genetic switches are involved. The ancest ral species at levels I , IA, II, III an IV have disappeared. Having known the evolutionary history of the group, we could use synapomorphy an

d the concept of monophyly. Assuming that all eight lineages (groups of populati ons) are suf-ficiently distinct to be recognized as distinct species, we would h ave eight species. The simplest way would be to group these eight species into f our genera, each having a com-mon ancestor. Two of these common ances-tors have disappeared, but two are still liv-ing and would also be included in the respective genera (it will be more appropriate to regard E as ancestral to F and G anc estral to H). These could be further assembled into two families of four species each (two gen-era each) having a common ancestor at level IA and II, and these two families into one order with a common ancestor at level I. Please note that common ancestor at level genus, and include all 8 species (2 genera) in one family (and, of course, depen ding upon the degree of diversity from related families, this could still be a c onstituent of a monotypic order). The third option would be to have a single genus of eight species. Note the importance of synapomorphy in determining monophyly. Character-states a lternate leaves, cymose inflorescence and 5 petals (character-states of differen t characters not same) have been passed unchanged in all the eight descendents ( species), which as such are symplesiomorphic for each of these, and this symples iomorphy will be valid between any two (or more) I, IA, II, III and IV are also no longer living. species that you choose to combine into a genus, say, D and E, or C and F, or sa y ABCDE. On the other hand, if we consider only synapomorphy, the monophyly is e asily deciphered. A and B are accordingly synapomorphic for yellow petals, C and D for white petals, E and F for one carpel and The second option would be to include ABCD in one genus, and EFGH in another Developing Classifications 227 CO Stamens 2 >2 B MpraCOf bit lants herbaceousP -

tamens 2 C

O H ceo Free Plants Plants herbaceous

oody Fre e Plants ~ Stamens 2 ood e oody Un ed Plants

O oody Un ed 2 Plants O A E Figure 8.13 Diagrams based on evolutionary pattern depicted in Figure 10.11. A: Venn diagram based on the assumption that there are 21 woody species of which, 13 are with un ited carpels, and 8 with free carpels. Of the 13 with united carpels, 7 have 2 s tamens whereas 6 have more stamens. B: An unrooted tree based on Venn diagram. C: A possible rooted tree, if evolutionary history of the group was not kn own, 15 possible rooted trees could be drawn. D. Rooted tree based on knowledge that herbaceous habit arose from woody habit, and we know further evolution of w oody species. Other possibilities are discussed subsequently. E: Data matrix of above, where the number of species are not indicated. 228 Plant Systematics G and H for united carpels. Similarly, A, B, C and D are synapomorphic for herba ceous habit. The development of fleshy fruit in A, single seed in C, 10 stamens in F represent the occurrence of a derived character state in a single taxon, an d termed as autapomorphy. Such character states are not helpful in cladogram con struction, al-though indicative of divergence. Develop-ment of 2 stamens in D an d H independently represents homoplasy, and may lead to ar-tificial grouping of these two, if history of the group was not properly known. It must, how-ever, be noted that symplesiomorphy may sometimes be helpful in detecting monoleaves, cymose inflorescence and five pet-als. The species at level IA and above additionally share woody habit, VI and above ad-ditionally united carpels, V an d above one carpel (and not fused carpels). Similarly, species at level II and a bove share herbaceous habit (instead of woody habit) in addition to three common , at level III and above additionally yellow petals, and at level IV and above a dditionally (in place of yellow petals) white petals. The situation depicted above can be more easily represented through the concept of nested groups, more conveniently represented as a set of ovals, a Venn diagra m (Figphyly, especially where in some other taxa, it has evolved into another characte r-state. As such, out of the four plesiomorphic char-acter-states listed here, o nly the woody habit has changed to herbaceous habit, and as such in the remainin

g taxa (E, F, G and H) symplesiomorphy of woody character-state identifies monop hyly of the group ABCD. It must be remembered that synapomorphy and symplesiomor phy are a reflection of ho-mology for a particular character-state (or more than one character-states, each beure 8.13A). The diagram drawn here is based on the assumption that we have information from a large group in which there are 21 woody species of which 13 are with united carpels and 8 with free carpels. O f the 13 with united carpels 7 have 2 stamens where as 6 have more stamens. The information is presented in the form of an unrooted network (unrooted tree) (Figure 8.13B). The herbaceous species are longing to a different character). shown towards the left of left double arrow, and the woody species towards the r ight. Similarly, the species towards the left of the middle double arrow are wit h free carpels while those towards the right with fused car-pels. The species to wards the left of the right double arrow have more than 2 stamens and All above options would render (genus, fam-ily or order) monophyletic groups, th e ulti-mate goal of phylogenetic systematics. Any other options won t work. Keepin g D and E in one genus (or CDE or DEF) would make it polyphyletic, promptly reje cted once de-tected, because the group is derived from more than one ancestor. K eeping ABC un-der one genus, or FGH under one genus would make paraphyletic grou ps because we are not including all the descendents of the common ancestor (we a re leaving out D in first genus and E in second). In the same way, putting more than four species (but less than eight) under the same genus would make it parap hyletic. Paraphyletic taxa are strongly opposed by phylogenetic system-atists; t he classical case is the demise of traditional division of angiosperms into those towards the right just 2 stamens. monocots and dicots, over the last decade. It must be noted that in constructing the above Venn diagram and the unrooted ne t-work, only three character-state transformations are accounted for. We have co mpletely left out grouping of herbaceous plants and the woody plants with a sing le carpel. Inclusion of these would make the diagrams much more complicated, and present several alternatives. Also, the more meaning-ful trees have to be roote d (the most primitive end at the base), to reflect the phylogeny. Even with the phylogenetic history of the group known, there could be several variations of th e rooted tree, two simple ones being shown in the Figure 8.13C and 8.13D. If we did not know the evolutionary history of the group, a number of variations would be It must be noted that all eight species whatever way you classify share alternate Developing Classifications 229 possible, depending upon which character-state is plesiomorphic, and which chara cter (habit, carpel fusion or stamen number) forms the root, and what would be t he seTaxa-Operational Units quence of the character changes on the tree. The first step in data analysis involves the selection of Taxa for data collecti on, often called Operational Taxonomic Units (OTUs) in Taxometrics, Operational Evolutionary Units (OEUs) in cladistics, referring to the sample from which the data is collected. Al-though it would be ideal to select different individuals o f a population, practical considerations make it necessary to select the members of the next lower rank. Thus, for the analysis of a species would need selectio n of various populations, for the study of a genus they would be different speci es, and for a family they would be different genera. It is not advisable, howeve r, to use genera and higher ranks, as the majority of char-acters would show var iation from one species to another and thus would not be suit-able for compariso n. The practical solution would be to use one representative of each taxon. Thus , if a family is to be analysed and its genera to be compared, the data from one representative species of each genus can be used for analysis. Once the taxa ar e selected, a list of such taxa is prepared. A unique feature of cladistic studi

es, however, is that the list of taxa generally includes a hypothetical ancestor , the comparison with which re-veals crucial phylogenetic information, and is us ed for rooting of the tree. It is, increasingly being realized that only a speci es is the valid evolutionary entity, and all taxa at higher ranks are artifacts, constructed for the sake of convenience. A meaningful analysis would always be one derived from data from various species (taken from populations) and not any higher rank directly. The character-states chosen for analysis should necessarily be homologous (one d erived from another) and non-overlapping. The analysis becomes more meaningful w hen it is established that the evolution of a par-ticular character-state has be en the result of a corresponding genetic change, and not a mere plastic environm ental influence. This fact underlies the importance of the emerging field of mol ecular analysis in phylogenetic systematics. It is believed that the recognition of molecular character-states (nucleotide sequences) is often easier and more p recise, although there are always accompanying problems. The problem with vascular plants, espe-cially the angiosperms, is that we know v ery little about their evolutionary history. The fossil records, which generally give fair in-formation about evolution, are very scarce-ly represented. What we have available with us is a mixture of primitive, moderately ad-vanced and adva nced groups. Almost each group has some plesiomorphic and some apomorphic charac ter-states, and relative proportion of one or the another delimit the relative a dvancement of various groups. At-tempt to reconstruct the evolutionary history of the group involves comparative study its living members, sorting out plesi omor-phic and apomorphic character states, and distribution of these in various members. Once the evolutionary history of the group has been constructed, monoph yletic groups at various levels of inclusiveness are identified, assigned ranks, and given appropriate names, to arrive at a working system of classification. Characters PHYLOGENETIC DATA ANALYSIS A conventional definition of a taxonomic character is a characteristic that dist inguishes one taxon from another. Thus, white flowers may distinguish one specie s from another with red flowers. Hence, the white flower is one character and re d flower another. A more practical definition espoused The methodology of cladistics with incorpo-ration of numerical methods involves a num-ber of steps. 230 Plant Systematics by numerical taxonomists defines charac-ter (Michener and Sokal, 1957) as af eat ure, which varies from one organism to an-other. By this second definition, flow er colour (and not white flower or red flower) is a character, and the white flo wer and red flower are its two character-states. Some authors (Colless, 1967) us e the term at-tribute for character-state but the two are not always synonymous. When selecting a character for numerical analysis, it is iminformation is not available (a large number of plants in a population are not i n fruit) or the information is irrelevant (trichome type if a large number of pl ants are without trichomes), or the characters which show a much greater variati on within the same taxon. Such characters are omitted from the list. This consti tutes residual weighting of characters. The characters (leaves, bracts, carpels) or character states (simple leaf, palmate compound leaf, pinnate compound leaf) portant to select a unit character, which chosen should also be homologous, in t erms may be defined as a taxonomic character of sharing common ancestry or belon ging of two or more states, which within the to same evolutionary transformation series. study at hand cannot be subdivided logi- The petals of Anemone are modifi ed sepals cally, except for the subdivision brought and thus not homologous with the petals of about by the method of coding. Thus, tri- Ranunculus and hence no t comparable. Simichome type may be glandular or eglandular. larly, the tuber of sweet potato (a modified A glandular trichome may be sessile or root) cannot be compared with the tuber of stalked. An eglandular trichome may, simi-

larly, be unbranched or branched. In such a case, a glandular trichome may be recognized as a unit character and an eglandular trichome as another unit character. On the other hand, if all glandular trichomes in OTUs are of the same type and all eglandular trichomes are of the same type, the trichome potato (a modified stem). Binary and multistate characters type may be selected as a unit character. The characters most suitable for computer handling are two-state (binary or pres enceabsence) characters (habit woody or herbaceous). However, all characters may not be two-state. They may be qualitative multistate (flowers white, red, blue) or quantitative multistate (leaves two, three, four, five at each node). Such m ultistate characters can be converted into two-state (flowers white or coloured; leaves four or more vs leaves less than four). Or else the characters may be sp lit (flowers white vs not white, red vs not red, blue vs not blue; leaves two vs not two, three vs not three and so on). Such a splitting may, however, give mor e weightage to one original character (flower colour or number of leaves). It is essential that different character states identified are discrete or discontinu ous from one another. Discreteness of character states can be evaluated by compa ring the means, ranges, and standard deviations of each character for all taxa i n analysis. Additionally t-tests and multivariate analysis may also be used for evaluating character state disreteness. The first step in the handling of characters is to make a list of unit character s. A preliminary step involves character compat-ibility study in which each char acter is ex-amined to determine the proper sequence of character-state changes t hat take place as the evolution progresses (morphoclines or transformation serie s). The list should include all such characters concerning which information is available. A priori, all characters should be weighted equally (no weighting to be given to characters). Although some authors advocate that some characters sho uld subsequently be assigned more weightage than others (a posteriori weighting) , such considerations generally get nullified when a large number of char-acters is used. It is generally opined that numerical studies should involve not less than 60 characters, but more than 80 are desirable. For practical consideration, there may be some characters concerning which Developing Classifications 231 B A B A B A I A CC C D II III B A C A B C

A C B A C B IV B A C B C A A B A B C B A B A V C VI Figure 8.14 Ordering and polarity of character states. I: Binary character with single possible switch. II: Unordered three-state character with single possible switch. III: Un ordered Four-state character with single possible switch. IV: Ordered three-stat e character with two possible switches and three possible morphoclines. V: Polar ized binary character with two possible morphoclines. VI: Ordered and Polarized three-state character with 6 possible morphoclines. Ordering of Character-states and cytosine) at a particular locus, and as such present four character-states. As re-versals are common in these, these are al-ways left unordered. A binary character will have single step or switch (Figure 8.14-I) necessary for change. The minimum number of switches possible (Wagner parsimony) in a multist ate char-acter will depend whether the character states are ordered or left unor dered. In an unordered transformation series each char-acter state can evolve in to every other char-acter state, and represents a single switch Assigning Polarity (Figure 8.14-II, III). A three-state character It is, however, necessary to determine the relative ancestry of the character-st ates, or the assignment of polarity. The designation of polarity is often one of the more difficult and uncertain aspects of phylogenetic analysis. For this, th e comparison may be made within the concerned group (in-group comparison) or rel atives outside the group (outgroup comparison). The latter may often pro-vide us eful information, especially when the out-group used is the sister-group of the conwill have two switches or steps, and three possible morphoclines (Figure 8.14-IV ), four-state character three switches and several morphoclines. Whereas orderin g of two-state characters is relatively easy, multi-state characters are often d ifficult to order, and changes may often be reversible, and it is advisable to l eave them unordered, and idencerned group. If two character-states of a character are found in a single monop hyltify only one switch (Fitch parsimony). The etic group, the state that is also f

ound in a molecular characters are different DNA se- sister-group is likely to b e plesiomorphic and quences, that may differ in having one of that found only wi thin the concerned mono-the four bases (adenine, thymine, guanine phyletic group is likely to be apomorphic. 232 Plant Systematics A B C A B C D A B C D A 0 1 2 A 0 1 2 3 A 0 1 1 1 B 1 0 1 B 1 0 1 2 B 1 0 1 1 C 2 1 0 C 2 1 0 1 C 1 1 0 1 D

3 2 1 0 D 1 1 1 0 I II III A B C D A 0 1 5 5 B 1 0 B 1 0 5 5 IV C 5 5 0 1 D 5 5 1 0 V Figure 8.15 Data matrix of coded character states. I: Ordered three-state charac ter. II: ordered four-state character. III: Unordered character. IV: Binary character. V: Differe ntial weighting to character state changes; imagine A and B represent Purines (A denine and Guanine), C and D Pyrimidines (Cytosine and Thymine), purine to purin e or pyrimidine to pyrimidine change (transition) is given 1 step weight, but pu rine to pyrimidine change or reverse (transversion) given 5 steps weight. Ingroup comparison (also known as com-mon ground plan or commonality principle) is based on the presumption that in a given group (presumably monophyletic), the primi-tive structure would tend to be more common. Thus all 8 species of cladog ram in Fig-ure 8.12 share plesiomorphic character states: alternate leaves, cymo se inflorescence and five petals. Five species have plesiomorphic 5 stamens, two derived 2 sta-mens and one with 10 derived stamens. Similarly four species have red petals and two each with white and yellow petals. It is assumed that the evo lution of a derived con-dition will occur in only one of potentially numerous li neages of the group; thus the ancestral condition will tend to be in the ma-jori ty. As is evident from Figure 8.14, the number of possible morphoclines increase s after the polarity criterion is included and the selection of single appropriate morphocline representing the true sequence even more challenging. Character Weighting and Coding

The coding of character states is done by assigning non-negative integer values. Binary characters are conveniently assigned 0 and 1 for two states. If possible to distinguish, plesiomorphic state is assigned 0 and apomorphic state 1 code ( Figure 8.15-IV). It is often assumed that whereas the same character-state may a rise more than once within a group between closely related species (parallelism) , or between remotely related species (convergence; the distinction Developing Classifications 233 Table 8.1 A portion of the data matrix with hypothetical t OTUs and n characters . Binary coding involves for state a and 1 for state b. The NC code stands for characters not co mparable for that OTU. In this analysis a total of 100 characters were used but only nine are pictured here. Characters OTUs (t) dhb01eceoosaroyuw-iFtur lhfill10e ceacone-aryv O0 fiii1reorprounser -Laseev dli01ocnepmosump-lii01tttrraeersaaquc- li01tttaoe prrmonosuace-iiibi0ttc mgemngecu-lCpersa idf01teerenu-ldiPtass ibHttaa llPeno 1. 1 1 2. 0 0 3. luev 1 1 4. O1 5. 1 6. 0 7. 0 8. 1 9. 1 10. 1 11. 0 12. 0

1 1 1 1 0 10PIPIItt epepyy---1

0 0 NC

1 1 0

1 1 1

0 1 0

1 1 0 1

1 0 1 1 0 1 0 0 1

1 1 0 1 0 1 0 1 0

0 1 1 0 1 0 1 1 0

1 0 1 1 0 0 1 0 1

0 0 1 NC 1 0 1 0 1 1

0 0 1 0 1 0 1 0 0

0 1 1 1 0 0 1 0 1 1 0 0 1 1 1 1 1

13. 0 0 1 0 1 0 1 0 1 14. 1 0 1 0 1 0 1 0 1 15. 0 1 1 0 0 1 1 0 0 between parallelism and convergence is sometimes omitted) for a simple character , it is highly unlikely for more complex char-acters. It is also assumed that wh ereas many genes must change in order to create a mor-phological structure, one gene change is enough for its loss (reversal). This Dollo s law is taken into acco unt when choosing trees, gains of structures counted more than losses, a process known as Dollo parsimony. Such weighting of characters is often com-mon in phyl ogenetic analysis. In transforbe done as 0 for most primitive character-state (simple leaf), 1 for intermediat e char-acter-state (pinnately lobed leaf) and 2 more most advanced state (pinnat ely compound leaf) (Figure 8.15-I). In molecular data, transversions (Purine to pyrimidine or pyrimidine to purine changes) are given more weightage (Figure 8.1 5-V) over transitions (purine to purine or pyrimidine to pyrimidine), because th e latter occur more frequently and are easy to reverse, whereas the former is a less likely biochemical change. mation series leaf simple > pinnately lobed Restriction site gains may similarly be > pinnately compound, the development of weighted over site losses. A complex char-pinnate compound leaf from simple leaf oc- acter, presumably controlled by many genes, curred in two steps, and needs to be given may change less easily th an a simple char-more weightage. The coding may accordingly acter controlled by fewer genes. The former 234 Plant Systematics is often given more weighting over a simple Similarly sympetalous members tend to have epipetaly and tenuinucellate ovules. character. It may be assumed that leaf anatomy may not change easily but hairi- Such correlated characters receive less er ness may change readily. The number of weighting. If two characters are corre lated, steps between two character states is con- each gets 1/2 weighting, if th ree 1/3 weightveniently represented through character ing and so on. step matrix (Figure 8.15). One may, how-ever, be tempted to count leaf anatomy c har-acter as equivalent to two changes in hairi-ness. This may often be the resu lt of bias to obtain desired results. It is reasonable, how-ever, to adopt the a pproach of numerical tax-onomy to give equal weighting to all the characters in the preliminary analysis, identify those characters which show the least homopla sy and give them more weightage in the subsequent analysis, a pro-cess known as successive weighting. This avoids a bias towards a particular charac-ter, and as such enables rational treatment It is always advisable to identify and in-clude the most ancestral taxon (outgro up) as last taxon (or first taxon, as certain programs choose first taxon for ro oting) in the list of taxa. If it is possible to identify plesiomorphic and apom orphic character-states, 0 represents plesiomorphic character-state and 1 the ap omorphic character-state of a particular character. Outgroup taxon in the matrix gets 0 code for all character states (Table 8.4). For multistep changes, or unl ikely events appropriate codes as indicated in Figure 8.15 are transferred to th e matrix. Outgroup taxon in the matrix is essential in final rooting of the most parsimonious cladogram (tree). of available data. Residual weighting involves excluding a character from the list when information for a large number of taxa is not available, or is irrelevant. But in certain c ases, information may be available for a particular character for large number o f OTUs but not for a few. Alternately, the information may be irrel-evant for a few taxa (say, the number of spurs in a taxon, which lacks spurs). Such char-act ers are used in analysis but for the taxa for which information is not available or is irrelevant, an NC code (Not Comparable) is entered in the matrix. Wheneve

r the NC code is encountered, the program bypasses that particular character for comparing the concerned taxon. For data handling by com-puters, the NC code is assigned a particular (not 0 or 1) numeric value. Such residual weighting should , however, be avoided when appreciable number of taxa are not comparable for a p articular character. The coded data may be entered in the form of a matrix with t number of rows (OTUs) and n number of columns (character-states) with the dime nsion of the matrix (and the number of Measure of similarity Once the data have been codified and entered in the form of a matrix, the next s tep is to calculate the degree of resemblance between every pair of OTUs. A numb er of formulae have been proposed by various authors to calculate similarity or dissimilarity (taxonomic distance) between the OTUs. If we are calculating the s imilarity (or dissimilarity) based on binary data coded as 1 and 0, the followin g combinations are possible. 0,Uk 10 01 c d

TJ U J Number of matches m = a + d attributes) being t x n (Table 8.1). Number of positive matches a Number of mismatches u = b + c Sample size n = a + b + c + d = m + u j and k are two OTUs under comparison Certain characters in plants evolve together. Occurrence of stipules and trilacu nar nodes is usually correlated. Developing Classifications 235 Some of the common formulae are dis-cussed below: Yule coefficient Simple matching coefficient This coefficient has been less commonly used in numerical taxonomy. It is calcul ated as: This measure of similarity is convenient and highly suitable for data wherein 0 and 1 represent two states of a character, and 0 does not merely represent the a bsence of a character-state. The coefficient was introduced by Sokal and Michene r (1958). The coefficient is represented as: bc ac SY = ad bc Taxonomic distance Matches Taxonomic distance between the OTUs can be easily calculated as a value 1 minus similarity or 100 minus percentage similar-ity. It can also be directly calculat ed as Euclidean distance using formula proposed by Sokal (1961): SSM = Matches + Mismatches m X )2J 1/2 or m +u Ajk= MY9 i1 It is more convenient to record similarity in percentage (Table 8.2). In that ca se, the formula would read: The average distance would be represented as: m SSM = m + u

x 100 djk = jk When comparing a pair of OTUs, a match is scored when both OTUs show 1 or 0 for a par-ticular character. On the other hand, if one OTU shows 0 and another 1 for a particular character, a mismatch is scored. Other commonly used distance measures include Mean character difference (M.C.D.) proposed by Cain and Harrison (1958), Manhattan metric distance coefficient (La nce and Williams, 1967) and Coefficient Jaccard Coefficient of association of divergence (Clark, 1952). The coefficient was first developed by Jaccard (1908) and gives weightage to sco res of 1 only. This formula is thus suitable for data where absence-presence is coded and 1 rep-resents the presence of a particular char-acter-state, and 0 its absence. The formula is presented as: Once the similarity or distance between every pair of taxa has been calculated, the data are presented in a second matrix with t x t dimensions where both rows and col-umns represent taxa (Table 8.2; Table 8.3). It must be noted that diagon al t value in the matrix represents self-comparison of taxa and thus 100% simila rity. These values are redundant as such. The values in the triangle above this diagonal line would be similar to the triangle below. The effective number of si milarity values as such would be t x (t-1)/2. Thus if 15 OTUs are compared the n umber of values calculated would SJ = a + u a where a stands for number of characters that are present (scored 1) in both OTUs This can similarly be represented as a percentage similarity. be 15 x (15-1)/2 = 105. A data matrix with coded character-states for each taxon can be used for calcula ting 236 Plant Systematics Table 8.2 Similarity matrix of the representative hypothetical taxa presented as percentage simple matching coefficient. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 100 47.0 100 54.0 47.0 100 49.0 54.0 52.0 100 50.0 51.0 44.0 48.5 100 46.0 59.0 46.0 47.0 48.0 100 47.0 48.0 48.0 46.0 65.0 47.0 100 56.0 51.0 56.0 51.5 46.0 58.0 25.0 100

50.0 50.0 53.0 48.0 47.0 55.0 56.0 2 3 4 7 8 9 10 11 12 13 14 15 OTUs 56

45.0 45.0 54.0 47.0 44.0 46.0 45.0

49.0 54.0 49.0 49.0 49.0 55.0 57.0

50.0 50.5 45.5 50.0 49.5 51.5 53.0

60.0 58.0 65.0 58.0 59.0 57.0 54.0

40.0 41.0 51.0 42.0 44.0 44.0 44.0

79.0 77.0 92.0 81.0 68.0 72.0 67.0

30.0 36.0 31.0 30.0 41.0 39.0 40.0

100 92.0 75.0 96.0 81.0 81.0 78.0

100 73.0 94.0 83.0 81.0 72.0

100 75.0 62.0 72.0 67.0

100 81.0 100 81.0 74.0 100 74.0 67.0 87.0 100

Table 8.3 Dissimilarity matrix of the representative hypothetical taxa based on the similarity matrix in Table 8.2. 987 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 OTUs 1 0.0 2 53.0 0.0 3 46.0 53.0 0.0 4 51.0 46.0 48.0 0.0 5 50.0 49.0 56.0 51.5 0.0 6 54.0 41.0 54.0 53.0 52.0 0.0 53.0 52.0 52.0 54.0 35.0 53.0 0.0 44.0 49.0 44.0 48.5 54.0 42.0 75.0 0.0 50.0 55.0 51.0 50.0 40.0 60.0 21.0 70.0 0.0 10 50.0 55.0 46.0 49.5 42.0 59.0 23.0 64.0 8.0 0.0 47.0 46.0 51.0 54.5 35.0 49.0 8.0 69.0 25.0 27.0

11 0.0 12 52.0 53.0 51.0 50.0 42.0 58.0 19.0 70.0 4.0 6.0 25.0 0.0 53.0 56.0 51.0 50.5 41.0 56.0 32.0 59.0 19.0 17.0 38.0 19.0 0.0 45.0 54.0 45.0 48.5 43.0 56.0 28.0 61.0 19.0 19.0 28.0 19.0 26.0 0.0 44.0 55.0 43.0 47.0 46.0 56.0 33.0 60.0 22.0 28.0 33.0 26.0 33.0 13.0 0.0 13 14 15 Developing Classifications 237 the distance (and, consequently, the simi-larity) between every pair of taxa, in cluding the hypothetical ancestor. The distance is calculated as the total numbe r of character-state differences between two concerned taxa, the data presented as t x t matrix replace intuition with analysis. The method was based on determining the apomorp hic character-states present within a taxon and then linking the subtaxa based o n relative degree of apomorphy. Interestingly, whereas the method found little f avour with zoologists, it has been used in many botanical studies. Kluge and Far ris (1969) and Farris (1970) developed a comprehensive methodology for the devel opment of Wagner trees, based on the principle of parsimony. The method is the b asis of many phylogeny computer algorithms currently in use. A given dataset may , however, yield many possible equally parsimonious trees due to homoplasy, as m ore than one character-state change may occur during the evolutionary process of a Table 8.5). This method is closer to taxometric meth-ods, because both plesiomorphic and apo morphic character-states are given equal weightage, but the inclusion of hypo-th etical ancestor is always crucial for the study. Another method of calculating distance involves calculation of the number of apo morphic character-states common be-tween the pairs of concerned taxa, ignoring t he possession of plesiomorphic character-states in common (Table 8.6). Since onl y synapomorphy is likely to define monophyl-etic groups, this method is closer t o the original cladistic concept. particular group of organisms. The following steps are involved in the analysis: 1. Determine which of the various characters (or character-states) in a series of character transformations are Construction of Trees apomorphic. Different methods are available for the final analysis of cladistic information. Three of these commonly used in phylogenetic analy-sis include Parsimony-based methods, Dis-tance methods and Maximum likelihood method. 2. Assign the score of 0 to the Parsimony-based methods plesiomorphic character and 1 to the apomorphic character in each transformation series. If the transformation series contains more than two homologues, then th ese intermediate apomorphies may be scaled between 0 and 1. Thus, a transformation se- ries of three characters may be scored as 0, 0.5 and 1 (or 0, 1 and 2 depen dThe methods are largely based on the bio-logical principle that mutations are ra re events. The methods attempt to minimise the number of mutations that a phylog enetic tree must invoke for all taxa under consid-eration. A tree that invokes m inimum num-ber of mutations (changes) is considered to be the tree of maximum pa rsimony. The evolutionary polarity of taxa is decided for construction of such t rees. The Wagner groundplan divergence method, an example of this, was first dev eloped by H. W. Wagner in 1948 as a technique for deter-mining the phylogenetic relationships among organisms that he hoped would ing on the weighting assigned).

3. Construct a table of taxa (EUs) and coded characters (or character-states: see Table 8.3). 4. Determine the divergence index for each taxon by totalling up the values. Since apomorphic character-states are cod ed 1, the divergence index in effect represents the number of apomorphies (chara cter-states) in a taxon, except in cases of weighted cod-ing. For the data matri x in Table 8.4, the divergence index for 15 taxa would be calculated as: 238 Plant Systematics Table 8.4 Data matrix of t taxa and n characters scored as 0 (plesiomorphic) and 1 (apomorphic) character-states. Multistate character is assigned 0 for ancestr al state, 1 for intermediate and 2 for most advanced state.The matrix is similar to Table 9.1 but only 9 characters pictured are used for calculations. Also the last taxon included is the hypothetical ancestor in which all character-states are scored as 0 (plesio morphic), as it is presumed that the ancestor would possess all characters in a plesiomorphic state. (n)4 Taxa (t) y Characters 10hdbraeeousycwoo-llilh1 onaceceeF0 urf1fiiireorprounser -va O0 lbdidl120ceompounpemso- -Leaves lii01tttrraeersaaquc- ibHttaa llPeno 1l0itttaoe prrmonosaceu-1iib0itgc ecngmemu-id1tunef0reeldiPtass 1IPIIP0ttyp eeyp---ibHta 1. 1 1 1 2. 0 0 1 3. 1 1 1 4. eit vu 0 0 5. O 6. 0 7. 0 1 1 1 0 0 1 1 0 1 1 1 0 1 1 1 1

1 1 0 1 1

0 0 1 1 0

1 1 0 1 1

1 1 1 0 2

0 1 0 1 0

1 1 0 0 0 1

0 0 1 8. 0 0 0 1 0 0 1 1 1 9. 1 1 1 0 0 1 1 0 0 10. 0 0 0 1 1 0 1 1 1 11. 1 0 1 2 0 1 0 0 1 12. 1 1 0 0 1 1 0 0 1 13. 0 0 1 0 1 0 1 0 1 14. 1 0 1 0 1 0 1 0 1 15. 0 0 0 0 0 0 0 0 0 Taxon Divergence index equals its divergence index. The lines connecting the taxa are determined by sha red synapomorphies (see Table 444666055755575 8.6). The cladogram (Wagner tree) is presented in Figure 8.16. Not all cladistic methods apply the prin10 ciple of parsimony. The methods of compatibility analysis or clique analysis uti lize the concept of character compatibility. Such methods can detect and thus om it ho-moplasy. They can be carried out manually or using a computer program, and can generate both rooted as well as unrooted trees. Groups of mutually compatib le characters are termed cliques. Let us consider two characters, A and B, with two character-states each. Four character-state combina11 12 13 14 15 Note that the hypothetical ancestral taxon 15 has an index of 0. tions are possible: 5. Plot the taxa on a graph, placing each taxon on a concentric semicircle that Assuming the evolution has proceeded from A1 to A2 and from B1 to B2. If all the Developing Classifications 239 Table 8.5 t x t matrix presenting distance between taxa expressed as the number of characterstate differences between pairs of taxa. E us 1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 1 2 3 0 3 4 7 0 4 7 4 8 0 5 2 3 6 7 0 6 5 5 6 4 7 0 7 6 3 7 3 6 6 0 8 3 6 1 8 5 6 7 0 9 4 5 7 3 4 6 4 7 0 10 4 5 2 7 6 5 6 1 8 0 11 3 2 0

7 6 1 6 5 6 5 7 0 12 3 7 3 6 4 2 7 4 6 5 0 13 4 5 4 5 8 3 4 5 3 6 5 0 14 3 6 3 4 7 4 5 4 4 5 4 1 0 15 6 5 4 7 6 5 4 5 5 6

5 4 5 0 Table 8.6 t x t matrix presenting distance between taxa expressed as number of d erived (apom orphic) character-states common between pairs of taxa. 1 EUs 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 X 5 X 4 2 X 2 3 0 X 5 4 2 2 X 3 3 2 2 2 X 3 4 1 3 3 2 X 4 2 4 0 2 2

1 X 4 3 1 3 4 2 3 1 X 0 4 3 4 1 2 3 2 4 1 X 1 4 4 1 2 5 2 3 1 3 1 X 2 3 4 1 3 3 3 4 1 3 2 3 X 13 3 3 2 2 3 1 3 2 2 3 2 2

X 14 4 4 2 3 4 2 3 2 3 3 3 3 4 X 15 0 0 0 0 0 0 0 0 0 0 0 0 0 X

240 Plant Systematics four combinations are met in nature then obviously there must have been at least one reversal (A2 to A1) or parallelism (A1 to A2 occurring twice), and as such A and B are incompatible. On the other hand, if only two or three of the combina tions occur, then A and B are compatible. Cliques are formed by comparing all pa irs of characters and finding mutually compatible sets. The largest clique is se lected from the data to produce a cladogram. Finally, a rooted tree or network i s obtained according to whether or not a hy-pothetical ancestor was included in the analysis. 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 Divergence Index Fig. S. General representation of a Wagner tree. Multiple Trees praels unitedC p eaees Cfrrl small portion of the herbaceous lineage with yellow petals, again assuming that there are a total of 15 herbaceous species of which six are with red petals and 9 with yellow petals. Of these nine 4 are with united carpels and 5 with free carpels. The additional Venn diapraeC The unrooted tree constructed in Figure 8.13-B represented only a small portion of the evolutionary sequence. Extension of this tree would make it more complica ted and present a lot of possibilities. Let us add a yellowPetals dsaetelPr atssuoeebacPlnrh P praCelsunite Stamens 2> tamens 2S

db Plants d Plants herb

B A Figure 8.17 A: The Venn diagram for woody species, the same as Figure 10.12A. B: The Venn diagram for a small portion of the herbaceous lineage of assumed 15 species of w hich 6 are with red petals and 9 with yellow petals, latter with 4 species havin g united carpels and 5 free carpels. C: Extension of the unrooted tree of Figure 10.12B to include the species depicted in the Venn diagram B here. There are 5 actual character state changes but with 4 switches as united carpels have arisen twice. Developing Classifications 241 ytsoweealPll 00 0 e ydoJreefedti o s

H H d Plants wood puaCesrnl Plants herbaceous Petals red tamens2S> Petals yellow tamens2S O 0 Figure 8.18 Possible variation of the unrooted tree presented n figure 10.14C, i f we did not have any idea about the evolutionary history of the group. Note that tree length has increased to six, and habit has changed twice from woody to herbaceous and from red to yellow petals. Such homoplasious situations are uncommon. gram for the herbaceous species and the extended unrooted tree is presented in t he Figure 8.17-C. It must be noted that here we know the evolutionary history of

the group which normally is never known and the aim of phylogenetic analysis is to recon-struct and depict this evolutionary history through trees. The unrooted tr ee here has five character-state changes (actual changes, tree length) involved. The change from free to united carpels has occurred twice, and as such there ar e only four ge-netic switches involved. If we did not have the knowledge about t he evolutionary his-tory of the group, we would try a number of variations. One possible variation of the unrooted tree would be to link 4 herbaceous species wi th united carpels to the woody spe-cies with united carpels, thus presenting a s ingle change of free to united carpels. But this brings in further changes. Now, change from woody habit has occurred twice, change from red to yellow petals ha s occurred twice, and more significantly the number of actual changes (tree leng th) has increased to six being included in the tree, the number of options would increase. Also we have t o con-vert each unrooted tree into a rooted tree so that the most primitive basa l end of the tree is known, and different lineages pre-sented as the more advanc ed branches. This brings in many more options, as indicated earlier. In our exam ple, where we know the history of the tree, the tree can be rooted at R, as indi cated by an arrow (Figure 8.17-C), but in a large majority of cases, it is a com plicated process, and a lot of hypotheses, strategies and algorithms come into play. A number of sophisticated computer algorithms are available which compare trees and calculate their lengths. The widely used ones include NONA, PAUP, and PHYLIP . These programs determine the number of possible trees, and then sort out the s hort-est of all these. If we are dealing with three species, three rooted trees are possible [A (B, C)], [B (A, C)] and [C (A, B)] (Figure 8.19), if 4 taxa are mapped the 15 trees are possible, 5 then 105, for 10 taxa 34,459,425 trees and so on. (Figure 8.18), with same four genetic The number of possible rooted trees for n number of taxa can be calculated as: switches involved. With more descendents 242 Plant Systematics A B A B C B A C C B A B A I II III V A B C D A C B D A B C D A C B D

C Figure 8.19 Possible number of rooted and unrooted trees. I: Single rooted tree for two taxa. II-IV: Three possible rooted trees for three taxa. V: One possible unrooted tree for three taxa. VI-IX: Some of the possible 15 rooted trees for four taxa. Nr = (2n-3)! / [(2 n 2 ) X (n-2)! ] was not known we should expect 135135 rooted trees and 10395 unrooted trees. The figures also highlight the enormous challenges in reconstructing the evolutiona ry It can also be calculated as: Nr = P (2 i - 1) where P represents the product of all factors history of any group. (2 i -1) from i = 1 to i = n - 1. A large number of trees generated are sorted and, ones presenting the shortest e volutionary path, in agreement with the principle of parsimony, are shortlisted. A simpler way to calculate the possible number of rooted trees is as follows: Nr = (2(n + 1) 5) X number of trees for (n 1) taxa Distance methods As noted above, the number of possible rooted trees is much more than number of possible unrooted trees. Latter can be cal-culated as: Distance methods were originally developed for handling phenetic information and con-struction of phenograms, some of these have now been incorporated in cladis tic method-ology. Cluster analysis is the most commonly used method of construct ing trees. Nu = (2n - 5)! / [(2 n 3 ) X (n-3)! ] or more simply as: Nu = number of rooted trees for (n 1) taxa Cluster analysis Data presented in OTUs x OTUs (t x t) matrix are too exhaustive to provide any m eaningful picture and need to be further condensed to enable a comparison of uni ts. Cluster analysis is one such method in which OTUs are arranged in the in the Thus, for 3 taxa, 3 rooted trees and 1 unrooted trees are possible (Figure 8.19) , for 4 taxa 15 rooted trees and 3 unrooted trees and for 5 taxa, 105 rooted tre es are possible but only 15 unrooted trees. For our 8 spe-cies in Figure 8.12, i f evolutionary history Developing Classifications 243 order of decreasing similarity. The earlier methods of cluster analysis were cum ber-some and involved shifting of cells with simi-lar values in the matrix so th at OTUs with closely similar similarity values were brought together as clusters . Today, with the advancement of computer technology, pro-grams are available wh ich can perform an efficient cluster analysis and help in the con-struction of c luster diagrams or pheno-grams. The various clustering procedures are classified under two categories. technique or minimum method), the can-didate OTU for admission to a cluster has similarity to that cluster equal to the similarity to the closest member within the cluster. The connections between OTUs and clusters and between two clusters are established by single links between pairs of OTUs. This procedure frequently leads to long straggly clusters in comparison with other SAHN cluster methods. The phenogram for our data using this strategy is shown in Figure 8.20. Agglomerative methods The highest similarity value in our matrix (see Table 8.2) is 96.0 between OTUs 9 and 12, and as such they are linked at that level. The next similarity value o f 94.0 is between OTUs 10 and 12, but since 12 has already been clustered with 9 , 10 will join this cluster linked at 94.0. The process is repeated till all OTU s have been agglomerated into single cluster at similarity value Agglomerative methods start with t clusters equal to the number of OTUs. These a re suc-cessively merged until a single cluster has finally been formed. The most commonly used clustering method in biology is the Se-quential Agglomerative Hie

rarchic Non-overlapping clustering method (SAHN). The method is useful for achie ving hierarchical classifications. The procedure starts with the assumption that only those OTUs would be merged which show 100% similarity. As no two OTUs woul d show 100% similarity, we start with t number of clusters. Let us now lower the criterion for merger as 99% similarity; still no OTUs would be merged as in our example the highest similarity re-corded is 96.0%. The best logical solution wo uld be to pick up the highest similarity value (here 96.0) and merge the two con -cerned OTUs (here 9 and 12). By inference, if our criterion for merger is 96.0 we will have t-1 clusters. Subsequently the next lower similarity value is picke d up and the number of clusters reduced to t-2. The pro-cedure is continued unti l we are left with a single cluster at the lowest significant simi-larity value. Since at various steps of clus-tering a candidate OTU for merger would cluster with a group of OTUs, it is important to decide the value that would link the cl us-ters horizontally in a cluster diagram. A number of strategies are used for t he of 53.0. In the complete linkage clustering method (farthest neighbour or maximum method) the candidate OTU for admission to a cluster has similarity to that cluster equ al to its similarity to the farthest member within the cluster. This method will generally lead to tight discrete clusters that join others only with difficulty and at relatively low overall similarity values. purpose. In the average linkage clustering method, an average of similarity is calculated between a candidate OTU and a cluster or between two clusters. Several variatio ns of this average method are used. The unweighted pair-group method using arith metic averages (LPGMA) computes the average similarity or dissimilarity of a can didate OTU to a cluster, weighting each OTU in the cluster equally, regardless o f its structural subdivision. The method originally developed for the procedures of numerical taxonomy has been applied in phylogenetic analysis with relevant m odifications, and used for the construction of trees. LPGMA method procedure beg ins with as many clusters as the number of taxa. The two taxa with minimum dista nce merged to reduce the number of clusters by one. In the next In the commonly used single linkage clustering method (nearest neighbour 244 Plant Systematics 40 50 60 70 80 90

100 5 7 11 9 12 10 13 14 15 3 1 2 8 6 4 Figure 8.20 Cluster diagram of 15 OTUs based on similarity matrix in Table 8.2 u sing single linkage strategy. step average distance between new cluster and remaining taxa are determined by t ak-ing the average distance between these two members and all other remaining ta xa, weighting each taxon in the cluster equally regardless of its structural sub division, and merging the taxon with smallest distance to the first cluster. The process is repeated with this new cluster of three taxa, and the procedure cont inues till all the taxa are merged, the most distant taxon joining last of all. From measure of similarity or dissimi-larity of taxa (OEUs) as presented in Tabl e 8.5 and 8.6, a network presenting minimum dissimilarity is constructed. Analys is of data from first six taxa of table 8.4 is presented in Figure 8.21. The pro

cedure begins by unit-ing nearest taxa A and E (with minimum distance of 2). Nex t matrix in now con-structed in which distance between (AE) and rest of the taxa is recalculated. The lowest value in this matrix (step 1 matrix) is be-tween (A E) and E, which are next united at distance level 3 into (AE)B. The distance between this cluster and rest of the taxa is now recalculated as presented in step 2 matrix. The lowest distance in this matr ix is 4 between D and F which are united into one cluster. With this merger the distance between taxa/clusters is recalculated and presented in step 3 matrix. T he lowest distance 5.5 is now between clusters (AE)B and DF, which are next unit ed. Finally the distance between this enlarged cluster and C is recalculated as presented in step 4 matrix. Finally the two clusters (((AE)B)(DF)) and C are uni ted at distance of 6.5 to form final cluster ((((AE)B)(DF))C). The resulting phe nogram and reconstructed phylogenetic tree constructed from the analysis are presented in Figure 8.21. The distance matrix can similarly be generated from single nucleotide difference s between homologous DNA sequences derived from different species. Six such hypo thetical sequences from different species are presented in Figure 8.22. The dist ance matrix is based on number nucleotide differences between different sequence s. A and F with lowest distance are merged, followed Developing Classifications 245 Taxa A B C D E F Taxa AE B C D F Taxa (AE)B C D F A 0 AE 0 (AE)B 0 B 3 0 B 3 0 C 6 0 C 4 7 0 C 5 7 0 D 5.5 8 0 D 7 4 8 0 D 7 4 8 0 F 5.5 6 4 0 E 2 3 6 7 0 F 6 5 6 4 0 Step 2 matrix

F 5 5 6 4 7 0 Step 1 matrix Distance Matrix A E B D F C Taxa (AE)B FD C (AE)B 0 0 FD 5.5 0 c e C 6 7 0 4 Step 3 matrix 3

2 Taxa ((AE)B)(FD) C

((AE)B)(FD) 0 A E B D F C C 6.5 0 Taxa Phylogenetic tree Phenogram Figure 8.21 Construction of phenogram and phylogenetic tree (cladogram) based on distance matrix concerning first six taxa in Table 8.4 using UPGMA clustering method. The taxa with minimum distance are united and treated as single cluster in next mat rix, and distance values recalculated as average of distance from either of unit ed taxa. The procedure repeated till all taxa are united. The phylogenetic tree is constructed based on sequence of clustering of taxa. by recalculation of new matrix in which A and F form one cluster and value of ea ch taxon is calculated as average distance from A and F. Now lowest value is sha red by B and D which form second cluster. The values are recalculated similarly , and successively C joins AF cluster, and then E joins (AF)C cluster. The two c lusters are finally merged to enable construction of phylogenetic tree either as phenogram or as cladogram. Some types of genetic polymorphism data such as RAPD are best handled when sharing of 0 code by two taxa in the matrix is ignored wh en both taxa lack a given polymorphic band in gel electrophoresis. Jaccard coeff i-cient is best suited for handing such data. Figure 8.23 presents results of RAPD analysis of 8 taxa, where only polymorphic

bands are shown, monomorphic bands being omit-ted. The distance matrix based on similar-ity matrix was processed using PHYLIP, as shown in Figure 8.24. Distance methods are suitable for handling both morphological and molecular data, or a c ombination of both. These methods use all data with usually equal importance, wh ereas the parsimony methods use only informative molecular data. In general for a site to be informative, when handling sequence data, irrespective of how many sequences are aligned, it has to have at least two different nucleotides, and ea ch of these nucleotides has to be 246 Plant Systematics 10 20 30 40 50 A GCCAACGTCG ATGCCACGTT B GCCAACAATG ATACCACGCC C GGTAACGTCA ATGGGACGTT D GCCAACATTG ATACCACGCC E GCTAACGACA ATACCAGGCT F GCCAACATCG ATGGGACGTT Taxa A B C D E F Taxa AF B C D E Taxa AF BD C E A 0 AF 0 AF 0 B 13 0 B 14.5 0 BD 13.5 0 C 11 17 0 C 11.5 17 0 C 11.5 20.5 0 D 13 E 13 12 10 24 12 0 18 0 D 12.5 10 24

GTTTAGCACC GTCCAGCACC GTCCAGCACC GTTTAGCTGC GTCCAGCTCC GTTTAGCTCC

GGTTCTTGTC GATTCTCGTC GGTTCATGTC GACACTCGTC GGTTTACGTC GATTCATGTC

CGATCACAGA CGAGTACCGA CAAGCAGAGA CGATCACCAA CGAGCACAGA CAATCACCAA

TGT TGT TGT TGT TGT TGT

0 E 15.5 15 12 0 F 9 16 12 12 18 0 E 15.5 12 12 18 0 Matrix step 2 Distance matrix Matrix step 1 Taxa (AF)C BD E A F C E B D A F C E B (AF)C 0 0 BD 16.75 0 E 13.5 15 0 5 Matrix step 3 10 i Taxa ((AF)C)E BD 15 ((AF)C)E 0 20 Phenogram BD 15.875 0 Cladogram Final Matrix ((((A,F:9),C:11.5),E:13.5),:15.875 (B,D:10)); Figure 8.22 Construction of phylogenetic tree based on single nucleotide differe nces in hypothetical DNA sequences of six species. Distance matrix is constructed based on t he number of nucleotide differences between each pair of DNA sequences and prese nted in distance matrix. Further analysis proceeds as detailed in Figure 8.21, a nd also in the text on these pages. present at least twice. Thus in the sequence data presented in Figure 8.22, out of 28 sites showing nucleotide differences in six se-quences, there are only 12 informative sites which can be used in parsimony analysis. Procedures based on U PGMA method, how-ever, don t account for different rates of evolution occurring in different lineages. Some distance methods such as transforme d distance method and neighbour-joining method, although more complex are capabl e of incorporating different rates of evolution within the lineages. Developing Classifications 247 R A B C D E F G H r

A B C D E F G H 21 1 1 1 0 0 1 1 1 0 1 0 1 1 1 1 0 0 1 0 0 1 1 1 0 1 1 1 0 1 1 1 1 0 1 1 1 0 0 0 1 2 3 3 45 4 5 6 6 7 7 98 8 9 10 10 11 11 12 12 13 13 II I 1 2 3 4 5 6 7 8 9 10 11 12 13 A B C D E F G H A 1 0 1 1 1 0 1 1 0 0 1 0 0 1 1 1 0 1 0 1 0

0 1 0 0 1 1 0 1 1 0 0 1 1

1 0 1 1 0 0 1 1 0 1 0 0 1

0 1 1 0 1 1 1 0 0 1 0 1 1

1 1 0 1 0 1 1 0 1 0 0 0 1

1 0 1 0 1 1 0 1 0 1 0 1 1

0 1 1 0

1 1 0 1

0 1 0 1

0 1 1 0

1 0 1 1

1 1 0 1

1 1 1 0

0 1 1 1 0 1 1 0 0 1 1 0 1 0 0 1 0 1 1 0 1 1 1 0 0 1 0 1 1 1 1 0 1 0 1 1 0 1 0 0 0 1 1 0 1 0 1 1 0 1 0 1 0 1 1 A 1.0 0.46 1.0 0.33 0.30 1.0 0.36 0.45 0.25 1.0 0.36 0.33 0.66 0.16 1.0 0.33 0.54 0.45 0.50 0.36 1.0 0.15 0.60 0.36 0.40 0.40 0.36 1.0 0.60 0.42 0.33 0.42 0.50 0.60 0.25 1.0 B B C C D D E E F F G G H H III IV Figure 8.23 Phylogenetic analysis of data concerning polymorphic bands from gel electrophoresis from DNA of 8 taxa. I: Polymorphic bands of DNA from 8 taxa (A-H ), R representing reference bands; II: Binary coded matrix of the polymorphic bands; III: The same matrix presented in conventional format; IV: Lower triangular matrix of sim ilarity matrix using Jaccard coefficient, wherein sharing of 0 state (absence of bands) is ignored. Further handling of data using UPGMA program of PHYLIP is presented in Figure 8.24. Divisive methods Divisive methods as opposed to agglo-merative methods, start with all t OTUs as a single set, subdividing this into one or more subsets; this is continued until further subdivision is not necessary. The commonly used divisive method is asso ciation analysis (William, Lambert and Lance, 1966). The method has been mostly use d in ecological data employing two state characters. It builds a dendrogram from the top downwards as opposed to cluster analysis, which builds a diagram from t he bottom up. The first step in the analysis involves 248 Plant Systematics 8 Taxona 0.00 0.54 0.67 0.64 0.64 0.67 0.85 0.40 Taxonb 0.54 0.00 0.70 0.55 0.67 0 .46 0.40 0.58 Taxonc 0.67 0.70 0.00 0.75 0.34 0.55 0.64 0.67 Taxond 0.64 0.55 0. 75 0.00 0.84 0.50 0.60 0.58 Taxone 0.64 0.67 0.34 0.84 0.00 0.64 0.60 0.50 Taxon f 0.67 0.46 0.55 0.50 0.64 0.00 0.64 0.40 Taxong 0.85 0.40 0.64 0.60 0.60 0.64 0 .00 0.75

I Taxonh 0.40 0.58 0.67 0.58 0.50 0.40 0.75 0.00 (((Taxona:0.20000,Taxonh:0.20000):0.11000,(Taxonc:0.17000, Taxone:0.17000):0.140 00):0.01500,((Taxonb:0.20000,Taxong:0.20000):0.08125, II (Taxond:0.25000,Taxonf:0.25000):0.03125):0.04375);

III

IV Figure 8.24 Construction of phylogenetic tree based on polymorphic bands from ge l electrophoresis from DNA of 8 taxa using UPGMA program of PHYLIP; I: Square distance generated f rom Figure 8.23-IV, each value calculated 1-similarity value. II: Outtree file g enerated by UPGMA option of NEIGHBOUR program; III: Upright square tree (Phenogr am) plotted through DRAWGRAM program; IV: Cladogram, but with branch lengths omi tted. calculating chi square value between every pair of characters using the formula: For each character the sum of chi-square is computed and the character showing m aximum chi square value is chosen as the first differentiating character. The wh ole set of OTUs is divided into two clusters, one containing the OTUs which show the character-state a and another containing OTUs which show the character-stat e b. Within n(ad bc)2 Xh2i _ [(a + b)(a + c)(b + d)(c + d)] where i stand for the character being com-pared and h for any character other th an i. Developing Classifications 249 each cluster, again, the character with the next value of the sum of chi square is se-lected and the cluster subdivided into two clusters as before. The process is repeated till further subdivision is not significant. culated similarity values as 100 minus similarity (if similarity values are in p ercentage) or 1 minus similarity (if similarity values range between 0 to 1). A dissimilarity matrix based on Table 8.2 is presented in Table 8.3. Hierarchical classifications

The first step in the ordination starts with construction of the x-axis (horizon tal axis). In the commonly used method of polar ordination, the two most distant OTUs are se-lected as the end points (A and B) on x-axis. In our example, these are OTU 8 and 7 with a distance (dissimilarity value) of 75. The position of al l other OTUs on this axis can be plotted one by one. OTU 10 has a distance of 64 from A (OTU 8) and a distance of 23 from B (OTU 7). A compass with a radius of 64 units is swung from A and a compass with a radius of 23 units is swung from B , form-ing two arcs. A line joining the intersection of two arcs forms a perpend icular on the x-axis, and the point at which the line crosses the x-axis is the position of the OTU. The distance between the x-axis and the point of intersecti on of arcs is the poorness of fit of the concerned OTU. The location of OTU on t he axis from the left (point A) can also be calculated directly instead of plott ing: The phenogram constructed using any tech-nique or strategy can be used for attem pt-ing hierarchical classification, by deciding about certain threshold levels f or different ranks. One may tentatively decide 85 per cent similarity as the thr eshold for the spe-cies, 65 for genera and 45 for families and recognize these r anks on the basis of num-ber of clusters established at that thresh-old. Whereas such an assumption can help in hierarchical classification, the point of confli ct would always be the threshold level for a particular rank. Some may argue and a re justified in doing so to suggest 80 per cent (or any other value) as the thresh old for species. It is more common, therefore, to use terms 85 per cent phenon l ine, 65 per cent phenon line, and 45 per cent phenon lines. These terms may conv e-niently be used till such time that sufficient data are available to assign th em formal taxonomic ranks to the various phenon lines. L2 + dAC2 dBC 2 The results of cluster analysis are com-monly presented as dendrograms known as phenograms. They can also be presented as contour diagrams (Figure 8.25), origin ally developed under the name Wroclaw dia-gram by Polish phytosociologists. The con-tour diagram may also incorporate the levels at which clustering has taken p lace. x = 2L where x is the distance from the left end, L is the dissimilarity value between A and B (length of x-axis), dAC is dissimilarity be-tween A and the OTU under co nsideration and dBC as the dissimilarity between B and the OTU under considerati on. The poorness of fit (e) of this OTU can be calculated as: Ordination Ordination is a technique which determines the placement of OTUs in two-dimensio nal or three-dimensional space. The results of two-dimensional ordination are co nvee = dAC 2 x 2 niently represented with the help of a scat-ter diagram and those of three-dimen sional ordination with the help of a three-dimen-sional model. The procedure wor ks on dis-tance values calculated directly from the coded data or indirectly fro m the already calAfter the position of all OTUs has been de-termined and the poorness of fit calc ulated, a second axis (vertical axis or y-axis) has to be calculated. For this, the OTU with the highest poorness of fit (most poorly fitted to x-axis) is selec ted and this forms the first reference OTU of y-axis. The second refer-ence OTU is selected as that one with the 250 Plant Systematics

2.0000 1.6000 1.2000 0.8000 0.4000 0.0000 Figure 8.25 Contour diagram based on the phenogram shown alongside. highest dissimilarity to the first reference OTU of y-axis, but within 10 per ce nt (of the length of x-axis) distance on x-axis. The po-sition of all other OTUs on the y-axis and their poorness of fit is determined as ear-lier. By using the values of poorness of fit to y-axis, a z-axis can be similarly generated and th e position of all OTUs on z-axis deter-mined similarly. The values can be used f or constructing a scatter diagram or a threedimensional model. A commonly used ordination technique known as principal component analysis also calculates values for a two-dimensional scatter diagram. In this method, however , the values on the horizontal as well as the vertical axis are non-zero, rangin g from -1 to 1 (calculated as eigenvalues) and as such the Developing Classifications 251 scatter diagram is presented along four axes: positive horizontal, negative hori zontal, positive vertical and negative vertical (Fig 8.26). The technique is bas ed on the assumption that if a straight line represented a single character, all the OTUs could be placed along the line according to their value for that chara cter. If two characters were used, a two-di-mensional graph would suffice to loc ate all OTUs. With n characters, n-dimensional space is required to locate all O TUs as points 3 11 - +0.6 168 - +0.4 - +0.2 2e 0.8 - 0.6 0.4 +0.2 +0.4 +0 6 +0 8 - 0.2 6

12% 7 9 - -0.2 1 T in space. s. --04 13 14 1 Principal component analysis determines the line through the cloud of points tha t ac-counts for the greatest amount of variation. This is the first principal co

mponent axis. A second axis, produced perpendicular to the first, accounts for t he next greatest amount of variation. The procedure ultimately pro-duces axes on e less than the number of OTUs. The first two axes are generally plot-ted to pro duce a scatter diagram. The proce-dure also calculates eigenvectors, which indic ate the importance of a character to a particular axis. The larger the eigenvect or in absolute value, the more important is - -0.6 17` 15 5 - -.0.8 Figure 8.26 Plot of the results of the principal component analysis of 18 hypothetical taxa. that particular character. likelihood is selected. The method is especially suited to molecular data, where the probability of genetic changes can be mod-eled more easily. With this appro ach, the probabilities are considered for every individual nucleotide substituti on in a set of sequence alignments. It is commonly un- derstood that transitions occur three times more frequently as compared to transversions. Thus if C, T, a nd A occur in one column (representing one site), the se-quences with C and T (p yrimidines) are more likely to be closely related than sequence with A (Purine). Using objective criteria probability for each site and every possible tree that describes the relationship of sequences. The tree with highest aggregate probab ility is selected as representation of A related method of ordination is princi-pal co-ordinate analysis developed by G ower (1966). This technique enables computation of principal components of any E uclidean distance matrix without being in possession of original data matrix. Th e method is also applicable to non-Euclidean distance and as-sociation coefficie nts as long as the matrix has no large negative eigenvalues. Princi-pal co-ordin ate analysis also seems to be less disturbed by NC entries than principal com-po nents. a true phylogenetic tree. Maximum Likelihood method Using any one of the methods, a large dataset commonly used, and which includes many homoplasies, large number of short-est trees may be generated by these auto -mated algorithms. These short listed trees have to be further compared. The method is similar to distant method in that all data is taken into considera tion. In this method, similarly character-state trans-formations are compared, a nd the probabil-ity of changes determined. These probabilities are used to calcu late the likelihood that a given tree would lead to the particular data set obse rved, and the tree with maximum The Consensus Tree The use of automated methods based on par-simony, even after applying relevant s trate252 Plant Systematics gies yield several trees, all presenting short-est pathways, based on parsimony but with different linkages among the taxa (OEUs), and often presenting differen t evolutionary history. Molecular studies of Clusiaceae by Gustafsson et al., (2 002) for example, includ-ing 1409 nucleotides of chloroplast gene rbcL positions using PAUP*4.0b8a parsimony analysis method, yielded 8473 most parsimonious tre es for the 26 species compared. Interestingly, the number of trees generated was so large that search for trees 3 steps longer than most parsimonious trees was aborted. More significantly different data sets (molecular, morphology) may yiel d dif-ferent trees. While selecting the consensus tree, the commonest approach i s to identify the groups, which are found in all the short listed trees, and bui ld a consensus tree. This could be achieved in different ways. Majority-rule consensus tree Majority-rule consensus tree shows all the groups which appear in a majority of trees, say, more than 50 per cent of the trees. It is useful to indicate for eac

h group on the consensus tree the percentage number of the most parsimonious tre es in which the group appeared. Such a consensus tree, however, provides a parti al summary of the phylogenetic analyses, and may be inconsistent with the trees from which it is derived. Semi-strict consensus tree Strict consensus tree A semi-strict consensus tree is useful when comparing trees from different data sets, or with different terminal taxa. The consensus tree developed indicates al l the relationships supported by both type of trees or any one of these, but not contradicted by any. Thus, in Figure 8.27, tree II does not give us any in-form ation about the time of origin of E, F and G, the tree I indicates that they ori ginated successively. Similarly, tree I does not indicate any close relationship between C and D, whereas the tree II does. The semi-strict consensus tree IV as such presents such in-formation, not contradicted by either tree. A more conservative approach in building a consensus tree involves including onl y monophyletic groups that are common to all the trees. The tree developed this way is known as strict consensus tree. Consider the two most parsimonious trees (although there could often be numerous trees of same shortest length available for comparison) as Evaluating consensus tree shown in Figure 8.27-I and 8.27-II. Developing a consensus tree involves the use of intuition, making guesses and de veloping hypothesis. A number of evaluation strategies are used to test the soun dness of the tree and measuring support for either the tree as a whole or for it s individual branches. These values are generally pub-lished along with the tree , to allow the fair assessment of the final results for comparison of trees base d on different datasets. Imagine that all groups A to J are mono-phyletic. Tree I shows that A and B are very closely related, and so are H and I. C, D, E, and F are shown arising succe ssively and are related in that sequence. Tree II shows a similar relationship b etween H and I, and between A and B (but group J is shown re-lated to these two) . The tree also shows that C and D are closely related. As relationships between E, F, and G are ambiguous, they are shown arising from the same point in evo-lu tionary history. The consensus tree III would thus omit taxon J (which is absent from tree I), show A and B, as also H and I as in the two trees I and II. The o ther taxa C, D, E, F, and G are shown arising from the same point. Consistency Index The principle of Parsimony is based on a basic rule of science known as Ockham s r azor, which says do not generate a hypothesis any more complex than is demanded b y the data . Some information in the data may be representing homoplasy (reversals , Developing Classifications 253 A B C D E F G H I A B J C D E F G H I A B

C D E F G H I A B C D E F G H I Figure 8.27 Two most parsimonious trees for a particular group of organisms, wit h monophyletic taxa A to J. I: showing C, D, E and F arising successively. II: E, F, G are show n arising at the same time from a common point, C and D being closely related. I II: Strict consensus tree of trees I and II. IV: Semi-strict consensus tree of t rees I and II. parallelisms). Dollo parsimony (as indicated above) minimizes the use of homopla sious characters. The commonest measure of ho-moplasy is the Consistency Index ( CI), which is calculated by dividing the number of genetic switches by actual ge netic changes on the tree. changes. In the tree shown in Figure 8.13B, there are three character-state chan ges, each involving one switch, and, as such, the consistency index would measur e 3/3 = 1. The tree shown in Figure 8.17C has five actual character-state change s (tree length is 5), but it involves only four genetic switches. As carpel fusi on has occurred twice, the consistency index would accordingly be 4/5 = 0.8. In the tree shown in Figure 8.18, the number of genetic switches remains the same a s four but the tree length has increased to six due to two Consistency Index CI = Min /L Min stands for the minimum possible tree length or genetic switches, and L for t he ac-tual tree length or actual number of genetic 254 Plant Systematics parallel (or convergent) evolutions; the CI changes in the character with actual number of changes in the character. RI is c omputed by first calculating the maximum possible tree length, if the apomorphic character-state originated independently in every would be calculated as 4/6 = 0.66. Consistency Index may also be calculated for individual characters. In Figure 8. 13B as such CI for all characters is one, while in 8.17-C, it is 0.5 for carpel fusion (minimum taxon that it appears in, or say, t he taxa are number of changes possible one for binary unrelated for the said chara cter-state. The character divided by actual number of value of RI is calculated as: changes here 2 since the character has changed twice) and 1 for rest. In Figure 8. 18, CI is 0.5 for habit and petal colour, and 1 for stamen number and carpel fus ion. The char-acters that lower the CI of a tree (or which have lower CI) are co nsidered to be homoplasious. The inclusion of a larger number of homoplasious ch aracters in the analysis lowers CI for the tree and contra-dicts phylogeny. Ther e may also be a char-acter, which changes only in one (or a very few) species, a nd may be of no relevance in others. Suppose one species develops spiny fruits. The length or number of spines would not be of any relevance in rest of the species without spines. Such a situation (a single species having a particular char acter) is known as autapomorphy. Since such a character has changed only once, i t gives CI of 1, and as such the inclusion of many such characters would increas e the consistency index of the tree, and provide false support. Such uninformati ve characters are

Retention Index RI = (Max L)/(Max Min) Max stands for the maximum tree length possible, L the actual tree length and Mi n the minimum tree length possible. The tree in Figure 8.17-C thus has a maximum possible tree length of 9 (minimum length of 4 and actual length of 5 as we alr eady know) and the RI would be (9-5)/(9-4) = 0.8. Higher the RI, sounder is the tree. as such omitted before calculating CI. Bremer Support (Decay Index) The principle of parsimony, followed in phylogenetic analyses, aims at selecting the shortest tree. Some parts of the tree may beore reliable than others. This is commonly evaluated by comparing the shortest tree with those one or more step s longer. Decay index or Bremer Support is the measure of how many extra steps a re needed before the original clade (group) is not retained. Thus if an internod e has decay index of 3, then the clade (monophyletic group) arising from it is m aintained even in the cladogram 3 steps longer than the shortest tree (see FigThe Consistency index values are often dependent on the number of taxa analyzed. Any increase in number of taxa lowers CI values, and this is true for data from differ-ent sources, morphological or molecular. ure 8.29). Certain branches of the tree which Retention Index Although theoretically the value of CI could range between 0 and 1, it rarely go es below 0.5. For a character that, has changed five times on a tree (this is a remote possibil-ity), CI will be 0.2. More so, the value of CI for a tree, very rarely may go below 0.5, and the values thus range between 0.5 and 1. The Retent ion Index (RI) corrects this nar-row range of CI by comparing maximum (and not m inimum as in CI) possible number of appear in the shortest tree, but disappear, or collapse in the tree one step longe r, are not drawn in the strict consensus tree. Greater the decay index value, mo re robust is that internode of the cladogram. Branches of the tree may also be tested by comparing the number of genetic chang es leading up to a particular group, and the CI of individual characters involve d. Doyle et al., (1994) on the basis of morphological data, developed a tree hav ing 18 character changes leading to angiosperms. Of these 18 characters 11 had C I of 1, thus Developing Classifications 255 supporting the view that angiosperms form a unique group of plants. volves sampling fewer than the full number of characters. The user is asked for the fraction of characters to be sampled. Block-bootstrapping is useful for hand ling correlated characters. When this is thought to have occurred, we can correc t for it by sampling, not individual characters, but blocks of adjacent characte rs. Block bootstrap and was introduced by Knsch (1989). If the correlations are b elieved to extend over some number of characters, you choose a block size, B, th at is larger than this, and choose N/B blocks of size B. In its implementation h ere the block bootstrap wraps around at the end of the characters (so that if a bl ock starts in the last B-1 characters, it continues by wrapping around to the fi rst character after it reaches the last character). Note also that if you have a DNA sequence data set of an exon of a coding region, you can ensure that equal numbers of first, second, and third coding positions are sampled by using the bl ock bootstrap with B = 3. Partial block-bootstrapping is similar to partial boot strapping except sampling blocks rather Bootstrap Analysis Any realistic analysis requires that the data used is randomized. Many technique s are available for randomizing the data. Bootstrap analysis is the commonly use d method de-veloped by Bradley Efron (1979). Its use in phylogeny estimation was introduced by Felsenstein (1985). Matrix in the Figure 8.28-A contains informat ion on the basis of which the unrooted tree in Figure 8.17-C is constructed. Wit hout touching the rows, any column is chosen at random to become the first colum n; similarly any other as second and the process is repeated till the number of columns in the new matrix is the same as in the original matrix. As the columns

are picked up from the original matrix, the new matrix may contain some characte rs repre-sented several times (the same column may have been picked up at random more than once), while others may have been omitted (the columns were not picke d up at all). The method is known as random sampling with replacement. The resul tant matrix B shows that character carpel fusion was picked up twice, whereas th e random selection process than single characters. missed the stamen number. Jackknife analysis (Jackknifing) is similar to bootstrap analysis but differs in that each randomly selected character may be resampled only once, and not multi ple times, and the resultant resampled data matrix is smaller than the original. Delete-half-jackknifing involves sampling a ran-dom half of the characters, and including them in the data but dropping the others. The resulting data sets are half the size of the original, and no characters are duplicated. The random var iation from doing this should be very similar to that obtained from the bootstra p. The method is advocated by Wu (1986). Delete-fraction jackknifing was advocat ed by Farris et. al. (1996) and involves deleting a fraction 1/e (1/2.71828). Th is retains too many characters and will lead to overconfidence in the resulting groups when there are conflicting characters. This and the preceding options for m a part of the SEQBOOT program of Phylip software, and the user is asked to sup ply the fraction of Repeating the method of random selection, multiple such matrices (usually more t han 100) are constructed, and for each matrix the most parsimonious tree/trees f ound. The consensus tree is developed from these most parsimonious trees. In thi s consensus tree, the percentage number of trees (gen-erated by bootstrap analys is) that contain that clade is indicated as bootstrap support value of that clad e. Bootstrap analysis based on the assumption that differential weight-ing by re sampling of the original data will tend to produce same clades if the data are g ood, and reflect actual phylogeny and very little of homoplasy. A bootstrap valu e of 70 per cent or more is generally considered as good support to the clade. Several variations of bootstrap analysis are available. The partial bootstrappin g in256 Plant Systematics A H Stamens B

C it sl

Plants Plants Plants Plants Plants Plants Plants Plants Plants Plants Plants Figure 8.28 A: Matrix based on the tree 9:16A. B: One possible matrix after proc edure of random sampling with replacement. characters that are to be retained. The pro-gram also offers permuting method, w ith following alternatives. Permuting species within characters involves permuti ng the columns of the data matrix separately. This produces data matrices that h ave the same number and kinds of characters but no taxo-nomic structure. It is u sed for different pur-poses than the bootstrap, as it tests not the variation ar ound an estimated tree but the hypothesis that there is no taxonomic struc-ture in the data: if a statistic such as num-ber of steps is significantly smaller in the actual data than it is in replicates that are permuted, then we can argue t hat there is some taxonomic structure in the data (though perhaps it might be ju st the pres-ence of a pair of sibling species). Permuting characters simply perm utes the order of the characters, the same reordering being ap-plied to all spec ies. It is included as a pos-sible step in carrying out a permutation test of ho mogeneity of characters (such as the Incongruence Length Difference test). Per-m uting characters separately for each spe-cies permute data so as to destroy all phylo-genetic structure, while keeping the base composition of each species the same as before. It shuffles the character order separately for each species. It is a common practice, and consequently more informative, to indicate the bran ch length (number of steps needed to reach that clade), bootstrap or jackknife s up-port and Bremer support (decay index) for each clade in the consensus tree (F igure 8.29). Effect of Different Outgroups An important component of procedures generating rooted trees is the incorporatio n of an outgroup in the analysis. In morphological data, the outgroup choice can influence phylogenetic inference. In molecular data, one specific concern is th e levels of sequence divergence between outgroups and ingroups and the subsequen t possibility of spurious long-branch attraction (Albert et al., 1994). The robu stness of tree can be tested by using randomly-generated outgroup se-quences, ex cluding all outgroups, and using outgroups selectively. Sytsma and Baum (1996), investigating the molecular phylogenies of angiosperms, found that removal of al l outgroups generated 27 shortest unrooted trees. Using Ginkgo only as outgroup yielded Herbaceous

C sl United > 2 Yellow Herbaceous Free > 2 Yellow Herbaceous Free > 2 estalP Red Woody Free > 2 Red Woody United > 2 Red Woody United 2 Red Yellow United Herbaceous United Yellow Free

Herbaceous Free Red Free Herbaceous Free talseP Red Free Woody Free Red paesClr United Woody United Red United bitaH Woody United Developing Classifications 257 Ranunculus repens Aquilegia formosa Piper amalago 80 * -G +1 Piper marginata 60 Peperomia metallica +4 90 Saururus cernuus +2 Gymnotheca chinensis 81 +1 29** -A 22

+3 Houttuynia cordata 94 Anemopsis californica Acorus calamus >5 -G 96 Lilium tigrinum 28** +1 76 Scilla violacea Cabomba caroliniana 28 -G Lactoris fernadeziana 16 *** Saruma henryii Chloranthus spicatus +3 53 Magnolia denudata +1 47 Illicium floridanum Figure 8.29 Tree developed from the study of 16 species of paleoherbs and 2 outg roup taxa, using 58 morphological and ontogenetic characters. The cladogram requires 214 steps an d has CI = 0.51 and III = 0.65. Bootstrap values are underlined and indicated be low a branch. Decay Index i indicated above the branch. Ranunculus repens and Aq uilegia formosa were chosen as outgroup taxa. (Drawn from Tucker and Douglas, 19 96). s 258 Plant Systematics lineages identical with baseline study (which included all outgroups); when only conifers were used as outgroup, the consensus tree was less resolved and many n odes collapsed. Use of Gnetales us outgroup increased the number of steps needed to yield baseline to-pologies, and interestingly, Ceratophyllum is shown as sis ter to all angiosperms except eudicots. These programs are basically similar to those designed for development of phenog rams, but differing essentially in the requirement to select one taxon for rooti ng in most programs. PHYLIP (Phylogeny In-ference package), is a commonly used s et of programs for inferring phylogenies (evolutionary trees) by parsimony, comp atibility, distance matrix methods, and likelihood. It can also compute consensu s trees, compute Effect of Lineage Removal distances between trees, draw trees, Lineage removal strategy highlights the resample data sets by bootstrapping or p roblems of lineage extinction, which often jacknifing, edit trees, and compute d istance leads to a particular group (especially criti- matrices. It can handle d ata that are nuclecal in angiosperms where fossil record is otide sequences, pro tein sequences, gene meager) not being sampled in analysis, thus frequencies, re striction sites, restriction giving distorted phylogenies. The same may fragment

s, distances, discrete characters, also be true for extant taxa, for which very and continuous characters. Distance matrix little data is available. The removal of all ma- can be generated using programs such as jor lineages, one at a time (Sytsma and DNADIST (which handles nucleotide se-Baum, 1994), provided useful in formation. quence data; it gives you choice to set The removal of Ceratophyllum, paleoherbs IIb weightage for transversions/transitions), (Chloranthaceae, and M agnoliales) had no PRODIST (which works with protein se-effect on the remaining angiosperm topol- quences) and RESTDIST (which works with ogy, whereas the remov al of paleoherbs I restriction site data). The most commonly (Aristolochiales and Illiciales), Laurales and used programs of PHYLIP for handl ing diseudicots showed substantial changes. tance matrix data include FITCH, KITSCH, and NEIGHBOR These deal with data which comes in the form of a matrix of pairwise distances between all pairs of taxa N EIGH-BOR offers ITPGMA option in which no taxon needs to be selected for rooting , whereas neighbor-joining option of this program, as well as FITCH and KITSCH n eed one taxon to be selected for rooting, otherwise by de-fault first taxon is u sed for rooting. The outtree generated by these programs can be plotted using DR AWGRAM or DRAWTREE, latter plotting only unrooted trees. DRAWGRAM provides a var iety of options to choose from. The trees can be drawn horiEffect of Exemplars The large computational load in handling a large data is often reduced by using place-holders or exemplars. These are often used to represent large lineages. Th e use of ex-emplars can warn about the possible artifacts when sparsely-sampled lineages ap-pear in basal positions. In such cases, more taxa can be added to th e data set for further analyses. But in the case of basal clade where a large nu mber of taxa are extinct, the results could be ambiguous. The results from angiosperms have shown that clades shift around with ease when the number of zontal or vertical, branches square taxa sampled for each lineage is reduced, (phenogram), v-shaped (cladogram), cur ved or circular. The branch lengths may be deand the use of exemplars at times could give misleading results. picted (phylogram) on the tree. The DNA sequence data presented in Figure 8.22 w as analysed using PHYLIP programs. Outputs are presented in Figure 8.30. DNA seq uence data can also be handled by DNAPARS pro-gram which performs Parsimony anal ysis Automated Trees A number sophisticated computer programs are available to construct phylogenetic trees. Developing Classifications 259 6 53 Taxona GCCAACGTCG ATGCCACGTT GTTTAGCACC GGTTCTTGTC CGATCACAGA TGT Taxonb GCCAACA ATG ATACCACGCC GTCCAGCACC GATTCTCGTC CGAGTACCGA TGT Taxonc GGTAACGTCA ATGGGACGTT GTCCAGCACC GGTTCATGTC CAAGCAGAGA TGT Taxond GCCAACATTG ATACCACGCC GTTTAGCTGC GA CACTCGTC CGATCACCAA TGT Taxone GCTAACGACA ATACCAGGCT GTCCAGCTCC GGTTTACGTC CGAGC ACAGA TGT Taxonf GCCAACATCG ATGGGACGTT GTTTAGCTCC GATTCATGTC CAATCACCAA TGT I 6 Taxona 0.000000 0.297888 0.253844 0.301576 0.306645 0.199728 Taxonb 0.297888 0.000000 0.470560 0.229212 0.276609 0.401604 ((((Taxona:0.099 86,Taxonf:0.09986):0.03265,Taxonc:0.13251):0.04302, Taxonc 0.253844 0.470560 0.000000 0.736034 0.286112 0.276199 Taxone:0.17553): 0.02684,(Taxonb:0.11461,Taxond:0.11461):0.08776); Taxond 0.301576 0.229212 0.736034 0.000000 0.475079 0.278527 Taxone 0.306645 0.2 76609 0.286112 0.475079 0.000000 0.460436 Taxonf 0.199728 0.401604 0.276199 0.27 8527 0.460436 0.000000 III II IV

Taxono GGCAACGACG ATACCACGTT GTTTAGCTCC GGTTCTCGTC CCAGCAGCCA TGT

`;I _11

III

Iti

YII

"-ITT

I L

Figure 8.30 Analysis of the DNA sequence data presented in Figure 8.22 using PHY LIP. I: Infile, first line indicating number of taxa and number of nucleotides in each sequence; II: Square distance matrix (outfile) generated by DNADIST program; III: Outtree file generated by NEIGHBOR program using UPGMA option; IV: DNA sequence of the 7th hypothetical taxon (taxono) used for rooting; V-VI: Square tree (Phenogram) and V-shaped tree (cladogram); VII: Unrooted tree of same; VIII-XVI: Diagrams ba sed on 7-taxa sequences; VIII: Phenogram, UPGMA option; IX-XI: Phylogram, Phenog ram and Cladogram based on neighbour-joining option of NEIGHBOR; XII: Phenogram based on DNAML program; XIII: Phenogram based on FITCH program; XIV: Phenogram b ased on KITSCH program; XV: Tree (Phenogram) generated based on DNAPARS program; XVI: Majority-rule consensus tree based on CONSENSE program, using outtree file s of above six programs. (All trees except VII (plotted using DRAWTREE) plotted using different options of DRAWGRAM program). 260 Plant Systematics

15 9

LCW I ELI

C Taxon1 101001111 Taxon2 101011001 Taxon3 010000111 Taxon4 111010000 Taxon5 101101101 Taxon6 110011010 Taxon7 011011001 Taxon8 000000111 Taxon9 111001100 Taxon10 000010111 Taxon11 101101001 Taxon12 110011001

D Taxon13 001010101 Taxon14 101010101 Taxon15 000000000 A Figure 8.31 Construction of trees using MIX program of PHYLIP based on matrix in the Table 8.4. A: Input file with fourth character converted into binary (simple and compo und leaves) character. Out of the 34 parsimonious trees generated by Mix, Consen sus tree generated by CONSENSE program presented as Phenogram (B), Cladogram (C) and Phylogram (D). and selects the best tree. It gives you choice to select the number of trees to be saved, 10000 being the default. The program di-rectly yields the outtree file . PROTPARS, similarly performs Parsimony analysis of Protein sequences. The prot ein sequences are given by the one-letter code used by the late Margaret Dayhoff s group in the Atlas of Protein Sequences, and consistent with the IUB standard a bbreviations. DNAMOVE which handles data similar to DNAPARS, allows the user to choose an initial tree, and displays this tree on the screen. The user can look at different sites and the way the nucleotide states are dist ributed on that tree, given the most parsimonious reconstruction of state change s for that particular tree. The user then can specify how the tree is to be rear ranged, rerooted or written out to a file. By looking at different rearrangement s of the tree the user can manually search for the most parsimonious tree, and Developing Classifications 261 can get a feel for how different sites are af-fected by changes in the tree topo logy. DNAML program carries out analysis of DNA sequences using Maximum Likeliho od Method. The program uses both informative and non-informative sites and yield s the outtree file directly. RESTML similarly handles restriction site data usin g maxi-mum likelihood method. Binary data coded as 0 (ancestral state) and 1 (ad vanced state) is handled by MIX, which performs parsimony analysis and generates outtree which can be plotted using DRAWGRAM. Input data from Table 8.4 and most parsimonious tree generated using MIX program is presented in Figure 8.31. For this analysis fourth multistate character was converted into bi-nary character ( simple and compound leaves). Using Wagner parsimony the pro-gram was able to gen erate 34 trees. Taxon 15 was used for rooting. CONSENSE program was used to sele ct the majority rule consen-sus tree. MOVE handles binary data and is an interac tive program which allows the user to choose an initial tree, and displays this tree on the screen. The user can look at different characters and the way their states are distributed on that tree, given the most parsimonious reconstruction of state changes for that particular tree. The user then can specify how the tre e is to be rearraranged, rerooted or written out to a file. By looking at differ ent rearrangements of the tree the user can manually search for the most parsimo nious tree, and can get a feel for how different characters are afquencies contains number of species (or populations) and number of loci , where as the second line contains number of alleles for each locus. the default number of data for each species (A-all) contains one allele less for each locus. thus for three loci with 2, 3 and 2 alleles respectively there would be four values. Without A option, there should be 7 values. The values in dataset are preceded a nd followed by blanks. The data from continuous characters does not contain the second line, the data would include number of species and the number of characte rs in the first line (only line above species

data). PHYLIP also offers programs to yield con-sensus tree (CONSENSE), Bootstrapping ( SEQBOOT) and a host of related programs. The following information may be useful in handling DNA sequence data. fected by changes in the tree topology. Prepare infile of DNA sequences in which taxon name takes 10 characters followed by sequences in groups of 10 (separated by a space), last three nucleotides bei ng terminating codon. First taxon should be one in-tended as one used for rootin g. Number of taxa (sequences used) and number of nucleotides in each sequence fo rms first line of file. Longer sequences can be interleaved (giving first part o f sequences of all taxa and then next part of all taxa) or aligned (finishing on e sequence and then going to second). Save this file in text format in notepad ( ANSI code should be used; is default in notepad). Distance matrix can be prepare d using dnadist.exe program. When program asks for infile, type above file name along with .txt ending. Choose the ratio of transitions and transversions, so th at program can handle it accordingly. You can also choose distance model such as F84, Kimura, Jukes-Cantor and Logdet. Give name of your output file (preferably in txt format so that you can open and see it in notepad). The file can be save d as square matrix or only lower triangle. The above file can be used for genera ting clusters through Fitch, Kitsch, and Neighbor programs. Each program searche s for the short-est tree. When program asks for infile, type the name of above o utput file. Give name of Multistate data can similarly be handled by PARS, and can be converted into bina ry data by FACTOR program. Data from Gene frequencies and continuous characters is handled by CONTML (constructs maximum likelihood estimates of the phylogeny; handles both types of data), GENDIST (com-putes genetic distances for use in the dis-tance matrix programs; handles data from gene frequencies) and CONTRAST (ex am-ines correlation of traits as they evolve along a given phylogeny; handles co ntinuous char-acters data). The data matrix for gene fre262 Plant Systematics A B C D E F G H I A B C D E F G H I Figure 8.32 Attempts towards construction of monophyletic groups. I: Strict cons ensus tree as presented in the Figure 8.27-III. With poorly resolved phylogenies, the separati on of H and I in a group distinct and of the same rank as group CDEFG would crea te a paraphyly, as HI are left out of the descendents of common ancestor o. II: A consensus tree (hypothetical) with better resolved phylogenies. Both groups CD EFG and HI are monophyletic and, in turn could be assembled into more inclusive group with common ancestor at level o, now containing all descendents of the com mon ancestor. This group (CDEFG, HI) and AB (also monophyletic) could be assembl ed into one most inclusive monophyletic group, containing all descendents of the common ancestor at p. output file (of this program) or simply ask for replacement if program reports t hat file is already present. Neighbour provides a choice between Neighbour-joini ng (in which one taxon is to be chosen for rooting) and UPGMA (in which no taxon for rooting has be se-lected). After selection of choice press Y. Pro-gram will

generate outtree file, if already present replace it. You can read this file if saved in txt format. The above outtree file can be used for plotting trees usin g DRAWGRAM or DRAWTREE programs. Draw program asks for intree file. Type in the name of above outtree file. It will next ask for name of font file. Type font1 o r any other within the Phylip folder. The program pro-vides you number of choice s including phenogram and cladogram. It also provides choice between indicating branch lengths (construction of conventional phylogram) or not (conventional phe nograms and cla-dograms where taxa end at same height). On typing Y tree preview will appear. Press Print screen on keyboard and paste on new paint file to save as image file in Paint. You can change options by clicking Fi le->change parameters in tree preview and go back to drawgram to generate other types of trees. Binary data can similarly be input in infile with just replacing nucleotide alph abets with binary 0 and 1 data as presented in Table 8.31 and handled by various programs men-tioned earlier. Gene Trees and Species Trees Traditional the phylogenetically trees are constructed using data from multiple char-acters, and if genetic data is used, from analysis of multiple genes. Such trees, appropriately known as species trees reflect the evolutionary history of related groups of species, and consequently a single species. A phylogenetic tre e based on the divergence observed within a single homologous gene is most appro priately called a gene tree. The genes commonly used for the construction of gen e trees have been described in Developing Classifications 263 chapter 7. Although they have been broadly Consider the strict consensus tree repused in recent years in the construction of resented in the Figure 8.27-III. Thi s tree is phylogenies, a single gene may not always reproduced in the Figure 8.3 2-I. As noted reflect relationships between species, be- earlier, the phylogenet ic relationships because divergence within genes, especially tween taxa (these c ould be different species, the sequence polymorphism occurs before genera, etc.) C, D, E, F, and G are poorly rethe splitting of populations that give rise to solved, and as such they are show n arising new species. from the common point, and consequently common ancestor as level o. Although H a nd I form a distinct group with a common an-cestor as m, but leaving these two o ut of the group including CDEFG would render latter as paraphyletic (cf. traditi onal separation of dicots and monocots). The safest situation would be to includ e all the seven taxa into one group, which may be regarded as belong-into the sa me rank as the group including A and B. All the nine taxa may next be in- cluded into the single most inclusive group with common ancestry at level p. We are th us Developing Classification Once the phylogeny of a group has been de- veloped, the evolutionary process wit hin the group can be reconstructed, the morphological, physiological and genetic changes can be described, and the resultant information used in the classificat ion of the group. Phy-logenetic classifications are based on the recognition of monophyletic groups and avoid including paraphyletic and often completely reject paraphyletic groups. Such classifica tions are superior over classifications ba sed able to construct groups at two ranks only. on overall similarity in several respects: Now supposing the phylogenies of the taxa were better resolved and we had obtain ed a consensus tree as shown in Figure 8.32-II. Now taxa C,D,E,F and G belong to a lineage which diverged from the main lineage, successive to the divergence of the lineage formed by A and B. Placement of H and I into one group HI would not create any problem as both this group as well as the group CDEFG are monophylet ic with separate common ancestors at level m and q, respectively. The groups CED FG and HI could next be as-sembled into group CDEFGHI with common ancestor at o.

Note that the group AB can next be merged with CDEFGHI to form single most incl usive group ABCDEFGHI. Now we have been able to construct taxa at three 1. Such a classification reflects the genealogical history of the group much more accurately. 2. The classification based on monophyletic groups is more predictive and of greater value than classification based on some characteristics. 3. Phylogenetic classification is of major help in understanding distribution patterns, plant interactions, pollen biology, dispersal of seeds and fruits. 4. The classification can direct the search for genes, biocontrol agents and potential crop species. 5. The classification can be of considerranks instead of two from tree I. able help in conservation strategies. Supposing the taxa A to I included in the tree, are different species. From tree II, thus we are able to recognize three genera AB, CDEFG and HI. The last two a re next as-sembled into family CDEFGHI and the former a monotypic family AB. The other al-ternative was to place A and B in two separate monotypic genera (depen ding on the de-gree of morphological and genetic divergence obtained) which are then assembled into The evolutionary history of the of the group of 8 living species shown in Figure 8.12 was known with precise point of character trans-formations, and the constr uction of mono-phyletic groups, assembled into successively more inclusive group s, did not pose much problem. But it is often not the case. Even, most resolved consensus trees are often ambiguous in several respects. 264 Plant Systematics family AB. The two families may next be as-sembled into order ABEDEFGHI. There c ould be other possibilities also. The second rank could be a subfamily and the t hird a family. Similarly, a third rank could be a suborder instead of an order. These final decisions are often made, based on the size of the group, degree of divergence, and the reliabil-ity of characters. All the groups recognized above would be monophyletic at the respecaffinities with very poor support, with highly unstable position. tive ranks. The final decisions on the recognition of groups are, however, often based on pe rsonal interpretation of phylogenies. Chloranthus, in this tree as well in sever al others, is closer to Magnolia (Magnoliales) and Laurus (Laurales), but often finds different treatment. APG II places Chloranthaceae after Amborellaceae at t he start of Angiosperms. Judd et al., had earlier (1999) placed Chloranthaceae u nder order Laurales of Magnoliid complex, but have now (2008) shifted the family among basal ANITA Grade with uncertain position. APweb of Stevens (2008), which places the family under order Chloranthales. Thorne had earlier (1999, 2000) pl > M a g n o l i a n a e >Magnoliale aced Chloranthaceae in M a g n o l i i d a e s > Chloranthineae (other sub-orders within the order being Magnoliineae, and Laur ineae), but subsequently (2003) in-cluded the family after Amborellaceae under o rder Chloranthales, the first order of Magnoliidae, finally (2006, 2007) separat ed under subclass Chloranthidae, a placement somewhat similar to APG II. Further discus-sion on angiosperm affinities will be resumed in the next chapter. Next, let us look at the tree shown in the Figure 8.29, a study on paleoherbs. R anunculus and Aquilegia were used as outgroup representing family Ranunculaceae; the ir isolated position from paleoherbs is clearly depicted in the tree. Paleoherbs constitute a group of taxa of uncertain affinities, which have been placed diff erently in various clas-sification schemes, but a few points seem to have been r esolved. Piper, and Peperomia (both belong to Piperaceae) form a distinct group,

and so do Saururus, Gymnotheca, Houttuynia and Anemopsis (all four belonging to Saururaceae), and the two families a well supported (bootstrap support of 90 pe r cent). This was confirmed by comparison of seven published trees of paleoherbs . Cabomba, Lactoris and Saruma have least resolved Chapter 9 Phylogeny of Angiosperms Angiosperms form the most dominant group of plants with at least 253,300 species (Thorne, 2007), a number much greater than all other groups of plants combined to-gether. Not only in numbers, angiosperms are also found in a far greater rang e of habi-tats than any other group of land plants. The phylogeny of angiosperms has, however, been a much-debated subject, largely be-cause of very poor record s of the earliest an-giosperms. These earliest angiosperms prob-ably lived in ha bitats that were not best suited for fossilization. Before trying to evaluate th e phylogeny, it would be useful to have an understanding of the major terms and concepts concerning phylogeny in gen-eral, and with respect to angiosperms in pa rticular. giosperms dominate all major terrestrial vegetation zones, account for the major ity of primary production on land, and exhibit bewildering morphological diversi ty. Unfortunately, much less is known about the origin and early evolution of angiosperms, result-ing in a number of different views regarding their ancestors, the earlies t forms and course of evolution. The origin of an-giosperms may be conveniently discussed under the following considerations. What are Angiosperms? ORIGIN OF ANGIOSPERMS The origin and early evolution of an-giosperms are enigmas that have intrigued b otanists for well over a century. They con-stituted an abominable mystery to Darwi n. The mystery is slowly being sleuthed and at the present pace of Sherlock Holms r e-search, it may be no more mysterious within the next two decades than for any other major group. With the exception of co-nifer forest and moss-lichen tundra, anAngiosperms form a distinct group of seed plants sharing a unique combination of characters. These important characters include carpels enclosing the ovules, po llen grains germinating on the stigma, sieve tubes with companion cells, double fertilization result- ing in triploid endosperm, and highly reduced male and fem ale gametophytes. The an-giosperms also have vessels. The pollen grains of angio sperms are also unique in having non-laminate endexine and ectexine differentiat ed into a foot-layer, columellar layer and tectum (tectum absent in Amborellacea e). The angiosperm flower typically is a hermaphrodite structure with car-pels s urrounded by stamens and the latter by petals and sepals, since insect pollinati on prevails. Arbuscular mycorrhizae are 266 Plant Systematics Table 9.1 Geological time scale.

Time Era Period Epoch Stage m years (mya) 0.01

Quaternary Holocene 2.5 Pleistocene 7 Cenozoic Pliocene 26 Miocene 38 Tertiary Oligocene 54 Eocene 65 Palaeocene 74 Maestrichtian 83 Campanian 87 Santonian 89 Upper Coniacian 90 Turonian 97 Cenomanian 112 Cretaceous Albian 125

Aptian 132 Mesozoic Barremian 135 Lower Hauterivian 141 Valanginian 146 Berriasian Upper Jurassic Middle 208 Lower Upper Triassic Middle 235 Lower 280 Permian 345 Carboniferous 395 Devonian 430 Palaeozoic Silurian 500 Ordovician

570 Cambrian 2400 Precambrian Algonkian 4500 _ Archaean also unique to angiosperms (except anemophily. In spite of these and other exceptions, this combination of characte rs is unique to angiosperms and not found in any other group of seed plants. Amborellaceae, Nymphaeales and Austrobaileyales). The vessel elements of an-giosperms typically possess scalari form perforations. There may be individual exceptions to most of these characters. Vessels are abse nt in some angiosperms (Winteraceae) while some gymnosperms have vessels (Gnetal es). The flowers are unisexual without perianth in several Amentiferae, which al so exhibit What is the age of Angiosperms? The time of origin of angiosperms is a mat-ter of considerable debate. For many years, the earliest well-documented angiosperm Phylogeny of Angiosperms 267 fossil was considered to be the form-genus Clavitopollenites described (Couper, 1958) from Barremian and Aptian strata of Early Cretaceous (Table 9.1) of southe rn England (132 to 112 mya-million years), a monosulcate pollen with distinctly sculptured exine, resembling the pollen of the extant other fossil pollens from the Jurassic age attributed to Nymphaeaceae ultimately turned out to be gymnosperms. In the last few years Sun et al., (1998, 2002) have described fossils of Archaef ructus from Upper Cretaceous (nearly 124 mya) of China, with clearly defined spi rally arranged conduplicate carpels enclosing ovules, a feagenus Ascarina. Brenner and Bickoff (1992) recorded similar but inaperturate pollen ture not reported in earlier angiosperm s. grains from the Valanginian (ca 135 mya) of The fruit is a follicle. This is con sidered to the Helez formation of Israel, now consid- be the oldest record of angiosperm fl ower. ered to be the oldest record of angiosperm fossils (Taylor and Hickey, 1996). Also found in Late Hauterivian (Brenner, 1996) of Israel (ca 132 mya) were Pre-Afropollis (mostly inaperturate, few weakly monosulcate), Clavitopollenites (weakly monosulcate to inaperturate), and Liliacidites (monosulcate, sexine similar to monocots). From Late Barremian have been recorded Afropollis and Brenneripollis (both lacking columellae) and Tricolpites (the first appearance of tricolpate Several vegetative structures from the Triassic were also attributed to angiospe rms. Brown (1956) described Sanmiguilea leaves from the Late Triassic of Colorad o and suggested affinity with Palmae. A better understanding of the plant was ma

de by Cornet (1986, 1989), who regarded it as a presumed primitive angiosperm wi th features of monocots and dicots. Although its angiosperm venation was refuted by Hickey and Doyle (1977), Cornet (1989) established its an-giosperm venation and associated reproducpollen grains) The number and diversity of angiosperm tive structures. Our knowledge of this co nfossils increased suddenly and by the end of troversial taxon, however, is far f rom clear. the Early Cretaceous (ca 100 mya) period major groups of angiosperms, including her-baceous Magnoliidae, Magnoliales, Laurales, Winteroids and Liliopsida were w ell repre-sented. In Late Cretaceous, at least 50 per cent of the species in the fossil flora were angiosperms. By the end of the Cretaceous, many extant angios perm families had ap-peared. They subsequently increased expo-nentially and cons tituted the most dominant Marcouia leaves (earlier described as Ctenis neuropteroides by Daugherty, 1941) are recorded from the Upper Triassic of Arizona and New Mexico. Its angiosperm a ffinities are not clear. land flora, continuing up to the present. Harris (1932) described Furcula from the Upper Triassic of Greenland as bifurcat e leaf with dichotomous venation. Although it seems to approach dicots in venati on and cuticular structure, it has several non-angiospermous characters includin g bifurcating midrib and blade, higher vein orders with relatively acute angles of origin (Hickey and The trail in the reverse direction is in-complete and confusing. Many claims of an-giosperm records before the Cretaceous were made but largely rejected. Erdtma n (1948) described Eucommiidites as a tricolpate di-cotyledonous pollen grain fr om the Jurassic. This, however, had bilateral symmetry in-stead of the radial sy mmetry of angiosperms (Hughes, 1961) and granular exine with gym-nospermous lami nated endexine (Doyle et al., 1975). This pollen grain was also discov-ered in t he micropyle of seeds of the female cone of uncertain but clearly gymnosper-mous affinities (Brenner, 1963). Several Doyle, 1977). Cornet (1993) has described Pannaulika, a dicot-like leaf form from Late Triassi c from the Virginia-North Carolina border. It was considered to be a three-lobed palmately veined leaf. The associated reproductive structures were attributed t o angiosperms but it is not certain that any of the repro-ductive structures wer e produced by the plant that bore Pannaulika. Taylor and Hickey (1996), however, do not accept its angiosperm affinities, largely on the basis 268 Plant Systematics of the venation pattern, which resembles more that of ferns. Much more informati on is needed before the Triassic record of anwas made by Martin et al., (1989) using nine angiosperm sequences from gapC, the nuclear gene encoding GADPH (cytosolic glyceraldehydes-3-phosphate dehydrogenas e). The observed number of nonsynonymous substitutions between each pair of spec ies (Ka) was compared to estimated rates of Ka (substitutions per site per year) inferred from known divergence times (e.g. plants-animals, plants-yeast, mammal -chicken, human-rat). The results implied separation of monocots and dicots at 3 19 + 35 mya, a dicot radiation at 276 + 33 mya, and cereal grass divergence at 1 03 + 22 mya. The results were questioned by several au-thors, since the study us ed a single gene. Wolfe et al., (1989) attempted to date the monocot-dicot split using a large number of genes in chloroplast genome and using a three-tiered ap proach. They suggested Late Triassic (200 mya) as the likely estimate of monocot -dicot split. Martin et al., (1993) pro-vided new data to support Carboniferous origiosperms can be established. Cornet (1996) described Welwitschia like fossil as Archaestrobilus cupulanthus f rom the Late Triassic of Texas. The plant had simi-larly constructed male and fe

male spikes, each possessing hundreds of spirally ar-ranged macrocupules. The fo ssil has revived renewed interest in gnetopsids. Given the inconclusive pre-Cretaceous record of angiosperms, it is largely belie ved that angiosperms arose in the Late Juras-sic or very Early Cretaceous (Taylo r, 1981) nearly 130 to 135 mya ago (Jones and Luchsinger, 1986). Melville (1983), who strongly advocated his gonophyll theory, believed that angi osperms arose nearly 240 mya ago in the Permian and took nearly 140 mya before t hey spread widely in Cretaceous. The Glossopteridae which gave rise to angiosper ms met with a disaster in the Triassic and disappeared, this disaster slowing do wn the progress of an-giosperms slow until the Cretaceous when their curve enter ed an exponential phase. This idea has, however, found little favour. gin (~300 Mya) of angiosperms. They used both rbcL and gapC sequences for this s tudy. Sytsma and Baum (1996) conclude that the results strongly caution using th e molecular clock for dating unless extensive sampling of taxa and genes with qu ite different molecular evolution is completed. Thus, the resolution of angiospe rm phylogeny may have to wait for a more complete molecular data and its proper appraisal. There has been increasing realization in recent years (Troitsky et al., 1991; Do yle and Donoghue, 1993; and Crane et al., 1995) to distinguish two dates one in th e Triassic when the stem angiosperms ( angiophytes sensu Doyle and Donoghue, 1993 o r proangiosperms sensu Troitsky et al., 1991) separated from sister groups (Gnetal es, Bennettitales and Pentoxylales) and the second in the Late Jurassic when the crown group of angiosperms (crown angiophytes) split into extant subgroups (Fig ure 9.1). What is the place of their origin? Molecular Dating It was earlier believed that angiosperms arose in the Arctic region (Seward, 193 1), with subsequent southwards migration. Axelrod (1970) suggested that flowerin g plants evolved in mild uplands (upland theory) at low latitudes. Smith (1970) located the general area of South-East Asia, adjacent to Malaysia as the site wh ere angiosperms evolved when Gondwana and Laurasia were undergoing initial fragm entation. Stebbins (1974) suggested that their origin occurred in exposed habita ts in areas of seasonal drought. Takhtajan (1966, There have been a number of attempts to estimate the time of divergence of an-gi osperms (node B in Figure 9.1) by applying a molecular clock to nucleotide seque nce data. The results mostly pointing to much earlier origin of angiosperms have , however, been contradictory. The first detailed attempt Phylogeny of Angiosperms 269 1980), who believed in the neotenous origin of angiosperms, suggested that angio sperms arose under environmental stress, probably as a result of adaptation to m oderate sea-sonal draught on rocky mountain slopes in became widespread following changing sea levels during the Early Cretaceous. areas with monsoon climate. Although agreeing with the role of environmental stress, many authors in recent years (Hickey and Doyle, 1977; Upchurch and Wolfe, 1987; Hickey and Taylor, 1992 ) have suggested that early angiosperms lived along stream- and lake-margins (lo wland theory). Later, they appeared in more stable backswamp and channel sites, and lastly, on river terraces. Taylor and Hickey (1996) suggested that ancestral angiosperms were Retallack and Dilcher (1981) believed that the earliest angiosperms were probabl y woody, small-leaved plants occurring in the Rift valley system adjoining Afric a and South America. Some of these angiosperms adapted to the coastal environmen ts and yaCtoniaceae setaeGln

Angiosperms Bennettitale

Cenozoic b

Cretaceous I1I1I1I1 5 IIIIII II Jurassic Triassic 111111 ~ 111111 I Figure 9.1 Phylogenetic tree of anthophytes (angiosperm lineage and sister group s). Point (I) marks when angiosperm lineage separated from sister groups in the Late Triassic, and (II) marks the splitting of crown angiosperms into extant subgroups in the Late Jurassic. Dotted line represents conclusions for which fossil record is not available (diagram based on Doyle and Donoghue, 1993). 270 Plant Systematics perennial rhizomatous herbs and evolved along rivers and streams on sites of rel a-tively high disturbance with moderate amounts of alluviation. These sites woul d have been characterized by high nutrient levels and frequent loss of plant cov er due to periodic disturbances. monophyletic, as is a genus from a genus, a family from a family. The principle, according to him, is that to be monophyletic, a taxon of any rank must be deriv ed solely from another taxon of the same rank. Gloss-opteridae and Angiospermida e belong to the same rank subclass. Both taxa consist of minor lineages that may be likened to a rope with many strands, a situation called pachyphyletic. This explanation, however, conforms to the concept of minimum monophyly and does not satisfy the rule of strict monophyly, which is now, the accepted criterion for m onophyly. Are Angiosperms monophyl-etic or polyphyletic? Engler (1892) considered angiosperms to be polyphyletic, monocotyledons and dico tyle-dons having evolved separately. Consider-able diversity of angiosperms in t he Early Cretaceous and the extant angiosperms led several authors, including Me euse (1963) and Krassilov (1977) to develop models for polyphyletic origin of an giosperms. This view is largely supported by considerable diverWhat are the possible ancestors? sity in the early angiosperm fossils. Ancestry of angiosperms is perhaps one of the most controversial and vigorously

debated topics. In the absence of direct fossil evidence, almost all groups of f ossil and living gymnosperms have been considered as possible ancestors by one a uthority or the other. Some authors even suggested the Isoetes origin of monocot yledons because the plant has a superficial resemblance with onion, albeit with no trace of seed habit. The various theories have revolved around two basic theo ries, viz., the Euanthial theory and the Pseudanthial theory of angiosperm origi n. Some other theories projecting her-baceous ancestry for the angiosperms have also recently received attention, making the question of ancestry of angiosperms rather more ambiguous: Most recent authors, including Hutchinson (1959,1973), Cronquist (1981, 1988), T horne (1983, 1992, 2000, 2007), Dahlgren (1980, 1989), Takhtajan (1987, 1997), J udd et al. (2002, 2008), Bremer et al. (APG II, 2003), and Stevens (APweb, 2008) believe in the monophyletic origin of an-giosperms, monocotyledons having evolv ed from primitive dicotyledons. This view is supported by a unique combination o f char-acters such as closed carpels, sieve tubes, companion cells, four microsp orangia, trip-loid endosperm, 8-nucleate embryo sac and reduced gametophytes. Sp orne (1974), on the basis of statistical studies, also concluded that it is high ly improbable that such a unique combination of characters could have arisen mor e than once, independently from Euanthial Theory gymnosperm ancestors. Also known as Anthostrobilus theory, Euanthial theory was first proposed by Arbe r and Parkins (1907). According to this theory, the angiosperm flower is interpr eted as being derived from an unbranched bisexual strobilus bearing spirally arr anged ovulate and pollen organs, similar to the hermaphrodite reproductive struc tures of some extinct bennettitalean gymnosperms. The carpel is thus regarded as a modified me-gasporophyll (phyllosporous origin of carpel). It is interesting to note that Melville (1983) considered angiosperms to be mono -phyletic but the explanation that he offers clubs him with the proponents of po lyphyl-etic origin. He believes that angiosperms arose from several different ge nera of Glossopteridae. According to him, the spe-cies is not always to be consi dered as the ancestor for determining a monophyletic nature. A species from anot her species is Phylogeny of Angiosperms 271 The bisexual flower of Magnoliales has been considered to have evolved from such a struc-ture. Also agreeing with this general prin-ciple, various authors have tried to identify different gymnosperm groups as possible angiosperm ancestors: Cycadeoidea, which had an elongated receptacle with perianth-like bracts, a whor l of pollen-bearing microsporophylls surrounding the ovuliferous region having n umerous ovules and interseminal scales packed together. There were, however, sig ns of abscission at the base of the male structure, which would have shed, expos ing the ovular Cycadeoidales (Bennettitales) region. The group, now better known as Cycadeoidales, appeared in the Triassic and disap peared in the Cretaceous. Their poten-tial as angiosperm ancestors was largely b uilt upon the studies of Wieland (1906, 1916). Lemesle (1946) considered the gro up The plant was believed to look like cycads with a short compact trunk and a crow n of pinnate compound leaves (Figure 9.2-A). It was earlier suggested that the m icrosporophylls opened at maturity but the subsequent studies of Crepet (1974) showed that to be ancestral to angiosperms, primarily be- microsporophylls were pinnate, and distal cause of the hermaphrodite nature of tips of pinnae were fused, the open ing of the

D f

a'

:: A t. B Figure 9.2 Cycadeoidea. A: Suggested reconstruction of plant with a compact trun k and numerous pinnate leaves. B: Suggested reconstruction of the cone cut open to show the arrangement of microsporangia. Ovulate receptacle is in the centre (A, after De levoryas, 1971; B, after Crepet, 1974). 272 Plant Systematics had rachis with branching pinnae, each with a synangium of four microsporangia. The seed-bearing structure (Caytonia) had rachis with two rows of stalked cupule s (Figure 9.3 B). Each cupule contained several ovules borne in such a way that the cupule is re-curved, with a lip like projection (of ten called stigmatic surface) near the point of attachment (Figure 9.3-C). The discovery of pollen grains within the ovules was thought to suggest their tr ue gymnosperm position, however, rather than being angiosperm ancestors. Krassil ov (1977) and Doyle (1978) regarded the cupule as homologous to the carpel, wher eas Gaussen (1946) and Stebbins (1974) considered it the outer integument of the ovule. Cladistic studies of Doyle and Donoghue (1987) support the caytoniales-a ngiosperm lineage. Thorne (1996) agreed that angiosperms probably evolved during the Late Jurassic from some group of seed ferns.

B Figure 9.3 Caytoniaceae. A: Palmately compound leaf of Sagenopteris phillipsi. B: Caytonia nathorstii with two rows of cu pules. C: Reconstruction of cupule of Caytonia sewardii. (B and C from Dilcher, 1979; C, from Stewart and Rothwell, 1993).

Cycadales Sporne (1971) suggested possible links between Cycadales and angiosperms in the palm-like habit of Cycadales, the ovules being borne on leaf-like microsporophyl ls, trends in the reduction of sporophyll blade as seen in various species of Cy cas. Although it may be difficult to assume Cycadales as ancestral to angiosperm s, the fact that they have been derived from pteridosperms, and yet resemble ang iosperms further supports the origin of angiosperms from pteridosperms. region was not structurally possible, and they later disintegrated internally (F igure 9.2-B). The ovules were terminal in contrast to their position in carpels of angiosperms. Caytoniaceae Opinion has strongly inclined in the recent years towards the probability that a ngiosperms arose from Pteridosperms or seed ferns, often placed in the order Lyg inopteridales but more commonly under Caytoniales. Caytoniaceae was described fr om the Jurassic of Cayton Bay in Yorkshire by Thomas, and subsequently from Gree nland, England and Canada. The group appeared in the Late Triassic and disapPseudanthium Theory Commonly associated with the Englerian School, the theory was first proposed by Wettstein (1907), who postulated that an-giosperms were derived from the Gnetops ida, represented by Ephedra, Gnetum and Welwitschia (formerly all placed in the peared towards the end of the Cretaceous. same order Gnetales). The leaves (Sagenopteris ) were borne on twigs and not the trunk. These had two pairs of leaflets (rarely 3 to six leaflets) and were net veined. Male structure s (Caytonanthus) The group shows more angiosperm char-acteristics than any other group of living or fossil gymnosperms. These include the presence of vessels, reticulate dicot-l ike leaves Phylogeny of Angiosperms 273 vesselless living angiosperms (cf. Winteraceae). Amentiferae are now regarded as advanced due to floral reduction. Tricolpate pollen grains also represent an ad vanced condition. More importantly, Gnetopsida is a very young group. But this theory has been strongly supported by Young (1981), who challenged the view that first angiosperms were vesselless 6 E) F~ D

Figure 9.4 Ephedra. A: A small portion of plant with opposite scale-like leaves. B: Male strobili on a branch. C: A male strobil us with series of opposite : `. : - M . bracts, apical bract subtending male stalk with several microspo-rangia. D: Fema le strobilus with series of whorled bracts, uppermost closely clasping ovules. TL ' t F g.z va~

;,, 1R * ~ .,, .! LL (Gnetum), male flower with perianth and bracts, strong gametophyte reduction, an d fusion of the second male gametophyte with the ventral canal nucleus. Ephedra resembles Casuarina in habit. Wettstein ho-mologized the compound strobili of Gn etales with the inflorescences of wind-pollinated Amentiferae, and regarded the showy insect pollinated bisexual flowers of Magnolia as pseudanthia derived by a ggregation of uni-sexual units, the carpel thus representing a modified branch ( Stachyosporous origin i I

II I yI I

A ;, ll Figure 9.5 Resconstruction of Archaestrobilus of carpel). cupulanthus and isolated organs. A: Associated leaf of Pelourdea poleoensis. B: Associated sterile A number of features, however, refute this theory: different origin of vessels ( Bailey, 1944) in angiosperms (from tracheids with scalariform pitting) and Gneto sids (from tra-cheids with circular pitting), several lower part of strobilus. C: Female strobilus with numerous spirally arranged macrocupules. D: Dispersed seed. (After Cornet, 1996). 274 Plant Systematics B Figure 9.6 Pentoxylales. A: Suggested reconstruction of Pentoxylon sahnii with s trap-shaped leaves. B: Suggested reconstruction of seed cones (From Sahni, 1948) . and assumed that vessels were lost in sev-eral early lines. Muhammad and Sattler (1982) found scalariform perforations in ves-sel-elements of Gnetum, suggesting that an-giosperms may be derived from Gnetales af-ter all. Carlquist (1996), ho wever, concludes that this claim from Gnetum does not hold macrocupule had an axially curled (tubular) bract-like organ with a narrow shaft and expanded funnel shaped apex. The macrocupules contained an ovule (or seed) surrounded by sterile scales. Three to four very small bracts were present attac hed near the base and surrounding the when large samples are examined. macrocupule. The basal group of angiosperms accord-ing to this theory included amentiferous-h amamelid orders Casuarinales, Fagales, Myricales and Juglandales. It is signific ant to note that Wettstein (1907) also included in this basal group, Chloranthac eae and Piperaceae, which have been inviting conEach male macrocupule contained filament like appendages instead of sterile scal

es within. Outside, the macrocupule was crowded with numerous bivalved microspor ophylls, each with four pollen sacs attached to an inflated stalk. On the outsid e of the female macrocupule were present gland-like structures resembling the st alks bearing pollen sacs on the male macrocupule. This suggests an origin from a bisexual macrocupule. The pollen grains are radially symmetrical and monosulcat e. The plant is regarded as a gnetophyte more siderable attention in recent years. The importance of Gnetopsids in an-giosperm phylogeny has been further strengthe ned by the discovery of Welwitschia like fossil described by Cornet (1996) as Ar chaestrobilus cupulanthus from the Late Triassic of Texas (Figure 9.5). The plan t had similarly constructed male and female spikes, each possessing hundreds of spirally arranged macrocupules. Male spikes were borne in clusters of three, whe reas female spikes occurred singly. Each female primitive than extant Gnetales. Ephedra is generally considered to be the most primitive of the three living gen era of gnetopsids. Cornet believes that Archaestrobilus possessed characters tha t may be plesiomorphic even for Ephedra, such Phylogeny of Angiosperms 275 L B Figure 9.7 Glossopteridae. A: Dictypteridiumf eistmantelii (Glossopteris tenuine rvis) vegetative branch. B: Fertile branch (Gonophyll) of Lidgettonia mucronata C: Fertile branch of Denkania indica with cupules. (A: from Chandra and Surange, 1976; B and C from Surange an d Chandra, 1975) as radial symmetry of floral parts which are spirally arranged and not opposite. He be-lieves that Bennettitales, Gnetales, Pentoxylales and angiosperms had a c om-mon ancestry sometimes before Late Trias-sic. Gnetales are relatives of angio sperms and Bennettitales that underwent drastic floral reduction and aggregation in response gins (i.e. angiosperms are polyphyletic). In most Magnoliidae and their dicotyle donous derivatives, they are modified pluriaxial systems (holanthocorms) that or iginated from the gnetopsids via the Piperales, whereas the modification of an o riginally uniaxial system (gonoclad or anthoid) gave rise to flow-ers of Chloran thaceae. Meeuse (1963) postulated a separate origin of monocotyledons from the f ossil order Pentoxylales through to wind pollination. Taylor and Hickey (1996) have presented a hypothesis for the derivation of the f lower of Chloranthaceae from the inflorescence unit (anthion) of gnetopsids, wit h considerthe monocot order Pandanales. able reduction in the reproductive parts. Pentoxylales (Figure 9.6) were described from the Jurassic of India and New Zeal and. The stem (Pentoxylon) had five conducting strands. The pollen-bearing organ (Sahnia) was pinnate: free above and fused into a cup below. The seed-bearing s tructure was similar to a mulberry with about 20 sessile seeds, each having an o uter fleshy sarcotesta and the inner hard sclerotesta. The sarcotesta was consid ered homologous to the cupule of seed ferns. The carpel of angiosperms was regar ded as a composite structure being an Anthocorm Theory This theory is a modified version of the pseudanthial theory and was proposed by Neumayer (1924) and strongly advocated by Meeuse (1963, 1972). According to thi s theory, the angiosperm flower ( functional reproductive unit ) has several separat e ori276 Plant Systematics Anthophytes

Angiophytes (possible) Crown Angiophytes Triassic Jurassic reticulate reticulate Bennettitaleans Gnetopsids pollen Sanmiguillia pollen Angiosperms Figure 9.8 A consensus phylogeny of Anthophytes proposed by Taylor and Hickey (1 996). Note that Pentoxylon has been excluded from sister groups (now only Bennettitaleans a nd Gnetopsids) of angiophytes. ovule-bearing branch fused with a support-ing bract. It is interesting, however, to note that Taylor and Hickey (1996) no longer in-clude Pentoxylon as a member of Anthophytes, which include angiosperm lineage and its sister groups Bennetti tales and gnetopsids (see Figure 9.12). According to them, Pentoxylon lacks key anthophyte characteristics such as distal, medial and proximal positioning of fe male, male and sterile organs on the reproductive axis, as well as the enclosure of ovules by bract-de-rived organs. minal ovules on dichotomous groups of branches. Folding of the scutella along th e cluster of its ovules forms the angiosperm condition, an indication of this cl osure being found in the Permian fossil Breytenia. In Lidgettonia, the fertile b ranch consists of four to eight disc-like bearing several seeds. In Denkania, de scribed from Raniganj , India, about six seed-bearing cupules are attached to lo ng stalks borne from the midrib of fertile scale. Gonophyll Theory The leaves of Glossopteris (Figure 9.7) are lanceolate, with distinct reticulate venation. In Glossopteris, the fertile region is cone-like with a transition fr om leaves to fertile scales, spirally arranged and conforming to the anthostrobi lus. In Mudgea, there is a suggestion of anthofasciculi, i.e. leafy structures w ith two fertile branches, one male and the other female, forming the an-giosperm flowers as found in Ranunculus and The Gonophyll theory was developed by Melville, (1962, 1963, 1983) largely on th e basis of a study of the venation pattern. He derived angiosperms from Glossopt eridae, which formed important elements in the flora of Gondwanaland. He further derived angiosperm flower from gonophyll, a fertile branching axis adnate to a leaf. In simple Glossopterids Scutum and Ottokaria, the fer-tile branch consiste d of a bivalved scale (hav-ing two wings) called the scutella with terAcacia. Melville believed that angiosperms arose 240 mya ago in the Permian and took aro und 140 mya before they spread widely in the Cretaceous. It is pertinent to poin t out, as explained earlier, that although he considered Phylogeny of Angiosperms 277 Figure 9.9 Dr D. L. Dilcher palaeobotanist with the Florida Museum of Natural Hi story at the University of Florida, who has pioneered research on Angiosperm fossils with spe cimen (above left) and reconstruction (above right) of recently described (Sun, Dilcher et al., 2002) Archaefructus sinensis, believed to be the oldest angiospe rm fossil nearly 124 mya old. 278 Plant Systematics angiosperms to be monophyletic, his justifi-cation puts him among the proponents of the polyphyletic origin of angiosperms. lineage continuing as Angiophytes up to the Late Jurassic when it further split into stem Angiophytes (the early extinct angiosperms) and crown Angiophytes cons

tituting the exHerbaceous Origin Hypothesis tant groups of angiosperms (Figure 9.8). The herbaceous origin hypothesis resembles the Pseudanthial theory but the ances tral plant is considered to be a perennial rhizomatous herb instead of a tree. T he term paleoherb was first used by Donoghue and Doyle (1989) for a group of der ivative (not ancestral) forms of Magnoliidae having anomocytic stomata, two whor ls of perianth and trimerous flowers, including Krassilov, who believed in the polyphyletic origin of angiosperms, identified th ree Jurassic groups as proangiosperms: Caytoniales, Zcekanowskiales and Dirhopal ostachyaceae. Pollen germinating on the lip, according to him, would be rather d isappointing because these plants would then be classified as angiosperms and ex -cluded from discussion of their ancestors. He evolved the Laurales-Rosales seri es from Caytoniales. Zcekanowskiales had bivalved Lactoridaceae Aristolochiaceae, Cabombaceae, Piperales, Nymphaeaceae and monocots.

According to this hypothesis, ancestral angiosperms were small herbaceous plants with a rhizomatous to scrambling perennial habit. They had simple leaves that w ere reticulately veined and had a primary venation pattern that would have been indifferently pinnate to palmate, whereas the sec-ondary veins branched dichotom ously. The vegetative anatomy included sieve-tube el-ements and elongate trachea ry elements with both circular-bordered and scalariform pitting and oblique end walls. The flowers occurred in cymose to racemose inflorescences. The small mono sulcate pollen had perforate to reticulate sculpturing. Carpels were free, ascid iate (ovules attached proximally to the closure) with one or two orthotropous, b itegmic, crassinucellate ovule and dicotyledonous embryo. The aforesaid authors cite extreme rarity of fossil an-giosperm wood and abundance of leaf impres-

~\ Ip

Figure 9.10 Reconstruction of leafy branch sions in early fossils. with flower of Archaeanthus linnenbergeri from middle CretaConsensus is emerging from recent phy-logenetic studies that gnetopsids represen t the closest living relatives of angiosperms, whereas the closest fossil group is the Bennettitales. Angiosperm lineage, together with these two groups, consti tutes Anthophytes. The group is believed to have split in the Late Triassic, the

angiosperm ceous. (After Dilcher and Crane, 1984.) capsules provided with stigmatic bands and showed links with monocots. Dirhopalo stachyaceae had paired ovules exposed on shield-like lateral appendages and prob ably evolved in Hamamelidales. Phylogeny of Angiosperms 279 Using the oldest, most complete fossil anTransitional-Combinational giosperm on record, David Dilcher (Figure 9.9), a palaeobotanist with the Florida MuTheory seum of Natural History at the University of J. Stuessy (2004) published a trans itional-Florida, recently announced the discovery of combinational theory for th e origin of ana new basal angiosperm family of aquatic giosperms, initiating renewed interest in plant, Archaefructaceae. It was published in the journal Science with coauthors Ge Sun of the Research Center of Palaeontology at Jilin University, Qiang Ji of the Geological Institute of the Chinese Academy of Geo-sciences at Beijing and t hree others (Sun et al., 2002). The family is based on a single genus Archaefruc tus with two species, Archaefructus sinensis and Archaefructus lianogensis. Thes e were probably aquatics herbs and living at least 124 mya. Archaefructus has pe rfect flowers rather un-like those of extant angiosperms there is no perianth, th e receptacle is very elon-gated, and the stamens are paired. The fruits are smal l follicles formed from conduplicate carpels helically arranged. Adaxial elongat e stigmatic crests are conspicuous on each carpel. Earlier to this, Dilcher and Crane (1984) had described Archaeanthus linnenbergeri (Figure 9.10) from uppermo st Albian/mid-cenomanian (approx 110 mya) of middle Cretaceous as a primitive fl owering plant with simple bilobed leaves, terminal flower with numerous free car pels producangiosperms and also serving to explain several recent divergent viewpoints and find-ings. The theory suggests that the angiosperms evolved slowly from the seed ferns in the Jurassic. Carpel was the first to develop, followed by the double fertilization and then the development of flower. These three fundamental transi tions may have taken more than 100 million years to complete. The theory is prop osed in view of the difficulty in finding ancestors for angiosperms, yet also co nsidering their sudden appearance and explosive evolutionary success. The extant angiosperms did not appear until Early Cretaceous when the final combination of all three important angiosperm features occurred, as presented by fossil record . This combination provided the opportunity for explosive evolutionary diversifi cation, especially in response to selection from insect pollinators, as also the accompanying modifications in compatibility and breeding systems. The theory at -tempts to explain discrepancy between fossil and molecular phylogenetic data, l atter suggesting pre-Cretaceous origin of an- giosperms when DNA (and protein) s e-quences showed first changes accompany-ing carpel evolution, much earlier than final combination of all the three angiosperm features. The theory also suggest s that bar-ring extinct seed ferns, from which the car-pel arose, other gymnospe rms had no direct phylogenetic connections to modern ing follicle fruit. Archaefructus was about 50 cm high, rooted in the lake bottom and was partially sup-ported by the water. Thin stems reached to the water s surface. Pollen and see d organs extended above the water. The leaves were possibly submerged. Seeds pro bably dispersed on the water and floated towards the shore where they germinated in shallower areas, he added. This is considered to be the oldest record of an angiosperm flower. It is placed in a distinct family Archaefructaceae, prob-ably sister to all extant angiosperms. Accord-ing to Stevens (2005), It is unclear a s to how it relates to extant angiosperms and its flow-ers are perhaps better in terpreted as inflo-rescences (Zhou et al. 2003; Ji et al. 2004; Crepet et al. 20 04). angiosperms.

Stuessy suggests that meaningful morphological cladistic analyses should focus o n ties between pteridosperms and angiosperms directly, and not include rest of t he gymnosperms. He believes that new biology of pollination and breeding systems that favoured outcrossing and developed angiosperm pollen, took place only afte r the 280 Plant Systematics advanced vessels in an aquatic environment yet gave rise to terrestrial monocots with more primitive vessel elements in the met-axylem of roots. They thus favou red the origin of Alismataceae from terrestrial forms. According to Hutchinson (1973), monocots arose from Ranales along two lines, one (Ranunculoideae) giving rise to Alismata les and other (Helleboroideae) giving rise to Butomales. Takhtajan (1980, 1987) proposed a common origin for Nymphaeales and Alismatales from a hypothetical ter restrial herbaceous group of Magnoliidae. Dahlgren et al., (1985) believed that monocots ap-peared in the Early Cretaceous some 110 mya ago when the ancestors o f Magnoliiflorae must have already acquired some of the present attributes of th at group but were less differentiated; some other di-cotyledonous groups had alr eady branched off from the ancestral stock. Thorne (1996) believes that monocoty ledons appear to be very early offshoot of the most primitive di-cotyledons. In their rbcL sequence studies, Chase et al., (1993) and Qiu et al. (1993) found th e monocots to be monophyletic and derived from within monosulcate Magnoliidae. A corus, Melanthiaceae, and Butomus are re-garded to be the least specialized Mono cotyledons. F Figure 9.11 Casuarinaceae. Casuarina suberba. A: Branch with male inflores-cences; B: Portion of male inflo-rescence; C: Male flower with single stamen; D: Part of female inflorescence showing 3 flowers; E: Fruits; F: Seed with broad wing. flower had developed, explaining the absence of angiosperm pollen record prior t o 130 mya. Origin of monocotyledons It was originally believed (Engler, 1892) that monocotyledons arose before dicot yledons and are polyphyletic (Meeuse, 1963). It is now largely believed that mon ocotyledons evolved from dicots monophyletically. Ac-cording to Bailey (1944) an d Cheadle (1953), vessels had independent origin and special-ization in monocots and dicots, and thus monocots arose from vesselless dicots. Cronquist did not a gree with the indepen-dent origin of vessels in two groups. He con-sidered monoc ots to have an aquatic origin from ancestors resembling modern Nymphaeales. This was strongly refuted, however, by studies of vessels done by Kosakai, Mosely an d Cheadle (1970). They considered it difficult to believe that puta-tively primi tive Alismataceae evolved BASAL LIVING ANGIOSPERMS Angiosperms are now increasingly believed to have evolved in very Late Jurassic or very Early Cretaceous. The course of evolution within the group is being thor oughly examined with newer tools of research. There was general agreement for nearly a century that the early angiosperms were woody shrubs or small trees (herbaceous habit being derived), with simple everg reen entire and pinnately veined leaves with stipules. Concerning the most primi tive living angiosperms, there have been two op-posing points of view: Englerian school (Considering Amentiferae, particularly Casuarinaceae to be the most prim itive dicots) and the Besseyan school (Bisexual 282 Plant Systematics

have been assigned. This view is supported by the occurrence of similar stamens and carpels, absence of vessels, morphology similar to pteridosperms, high chrom osome number suggesting a long evolutionary history, and less specialized beetle pollination of Drimys compared to Magnolia. Takhtajan (1980, 1987) later acknowledged that moderate sized flowers of Degener ia and Winteraceae are primitive, and the large flowers of Magnolia and Nymphaea ceae are of secondary origin. How-ever, he considered Degeneriaceae to be the mo st primitive family of living angiosperms. Cronquist (1981, 1988) also discarded Magnolia while considering Winteraceae to be the most primitive.

ij 11 _,Rd 1 Degeneriaceae Takhtajan, who was earlier a strong sup-porter of Magnolia as the most primitive liv-ing angiosperm, has abandoned this view in favour of Degeneriaceae and Wint eraceae to be the basal angiosperm families, but has maintained since 1980 to re gard Figure 9.14 Degeneriaceae. Degeneria vitiensis. A: Branch with flowers; B: Sta- men; C: Transverse section of car- pel; D: Fruit . Degeneriaceae as the most primitive. ous spirally arranged tepals, and few ovulate carpels. Food bodies terminating t he stamen Degeneriaceae (Figure 9.14) may be rec-ognized by their spiral, entire, exstipul ate leaves and large, axillary flowers with many tepals and a single carpel. Ves sel elements have scalariform perforations. Leaves are spirally arranged and pol len boat-shaped. The most significant plesiomorphic features in-clude stigma run ning the entire length of the carpel, laminar stamens with three veins, the frui t a follicle and embryo with 3 to 4 cotyledons. connectives indicate beetle pollination. The family (Figure 9.15) is regarded as the most basal family of Laurales. It is interest-ing to note that genus Idiospermum (which was recognized as new genus based on Calycanthus australiensis by S. T. Blake in 1972) was considered as the most primitive flowering plant by these authors. Endress (1983) had described In all respects, Ideospermum gives the impression of a strange living fossil . Calycanthaceae Suggestions have also come projecting Calycanthaceae (Loconte and Stevenson, 199 1) as basic angiosperms with a series of vegetative and reproductive angiosperm plesiomorphies such as shrub habit, unilacunar two-trace nodes, vessels with sca lariform perforations, sieve-tube ele-ments with starch inclusions, opposite lea ves, strobilar flowers, leaf-like bracteopetals, poorly differentiated numerPaleoherbs The last decade of the twentieth century has seen the strong development of an a lternative herbaceous origin hypothesis for angiosperms (Taylor and Hickey, 1996 ) originally developed as paleoherb hypothesis. The most primitive angiosperms a re considered to be rhizomatous or scrambling perennial herbs with simple net-ve ined leaves, flowers in racemose or cymose inflorescences, Phylogeny of Angiosperms 283 atophyllaceae (towards the end of Magnoliid complex) as having uncertain positio n. APG II, like Thorne places Amborellaceae and Chloranthaceae at the beginning of angiosperms (but as unplaced families), whereas family Ceratophyllaceae is pl

aced before Magnoliids. Chloranthaceae Figure 9.15 Calycanthaceae. Calycanthus Taylor and Hickey (1996) consider Chloranthaceae (Figure 9.16) the basic angiosp erm family. The family shows several plesiomorphic characters such as flowers in an inflorescence, plants dioecious, carpels solitary, placentation apical, and fruit drupaceous with small seeds. The family is the oldest in the fossil record , the fossil genus Clavitopollenites being assigned to Chloranthaceae and closer to the genus Ascarina. The stems of Sarcandra are primitively vesselless, but C arlquist (1996) has reported vessels in this genus. The family is considered to be earliest to record wind occidentalis. A: Flowering twig with solitary terminal flower. B: L. S. of flowe r showing free car-pels. C: Flower with some tepals and stamens removed. pollination in angiosperms. with free carpels containing one or two ovules. A number of families are include d in the group. Thorne (2000) had placed all of them under Magnoliales, along wi th Magnoliaceae and Winteraceae. In his later revision (2003), however, placed A mborellaceae and Chloranthaceae (together with Trimeniaceae and Austrobaileyacea e) under Chloranthales, the first order of Magnoliidae (and accordingly angiospe rms), the families arranged in that order. Subsequently (2006, 2007) he separate d them under distinct subclass Chloranthidae, at the begining of angiosperms. Th e family Ceratophyllaceae is placed after the monocot families, towards the begi nning of Ranunculidae. The placement of Amborellaceae at the beginning of angios perms is found in the classification schemes of Judd et al. (2003), APG II (2003 ) and APweb (Stevens, 2003). The position of the other two families is, however, not set-tled. Judd. et al. and APweb consider both Chloranthaceae (towards the end of basal families before Magnoliid complex) and CerThe plants are mostly herbaceous, some species being shrubs. The flowers are hig hly reduced, subtended by a bract and without any perianth, and arranged in decu ssate

Figure 9.16 Chloranthaceae. A: Ascarina lanceolata, flowering branch. B: A male flower C: fr uit D: Bisexual flower of Chloranthus henryi with bract, three stamens and pisti l with tufted stigma. E: Bisexual flower of Sarcandra glabra. 284 Plant Systematics pairs. The flowers are unisexual in Ascarina, Hedyosmum and Ascarinopsis but bis exual in Chloranthus and Sarcandra. Stamens vary in number from 1 to 5. The carp el lacks style, and single orthotropous ovule is bitegmic. Taylor and Hickey believe in the origin of Chloranthaceae from gnetopsids, hypot hesizing that the ovule and the bract subtending the floral unit in Chloranthace ae are homologous with one of the terminal ovules and proanthophyll subtending t he anthion (inflorescence

unit) of gnetopsids. Chloranthaceae has undergone considerable reduction in its number of parts as we

ll as general level of elaborateness. They also believed that the outer integument of the angiosperm bitegmic ovule ha s ring-like origin and is homologous with the ovular bracts that form the second integument in the gnetopsids. Figure 9.17. Ceratophyllaceae. Ceratophyllum Ceratophyllaceae submersum. A: A portion of plant; B: Whorl of leaves at node; C: Male flower; D: Young stamen; E: Dehiscing stamen; F: Fruit; G: Longitudinal section of fruit w ith pendulous seed. Chase et al., (1993) on the basis of rbcL had expressed the view that Ceratophyl laceae represents the basal angiosperm family. The family has fossil record exte nding back to the Early Cretaceous. Cladistic studies by Sytsma and Baum (1996) based on molecular data support the placement of Ceratophyllum (Figure 9.17) at the base of angiosperms, but the authors cautioned that resolution of basal angi osperm relationships may have to await both the collection of ad-ditional molecu lar and morphological data as well as further theoretical advances in phylogenet ic systematics. Hickey and Taylor (1996) felt that aquatic plant with highly red uced vegetative body and pollen wall, tenuinucellate, unitegmic ovules is a poor candidate for the basal-most position. Thorne (1996) believed the family is hig hly special-ized and its relationships are highly obscured. Other specialized fe atures include lack of roots, dissected leaves, reduced vasapomorphic character-states. Parsimony analysis using PAUP resulted in 10 trees at 590 steps. Calycanthaceae appeared as first branch. Magnoliaceae, Winteraceae and Chloranthaceae hypotheses appeared two steps longer, whereas Ceratophyllace ae and Casuarinaceae hypotheses appeared six steps longer. Other families that are considered belong-ing to paleoherbs and share plesiomorp hic features include Saururaceae, Piperaceae, Aristolochiaceae, Barclyaceae, Cab ombaceae and Nymphaeaceae. They share characters of herbaceous habit, tectate-co lumellate monosulcate pollen, apocarpous gynoecia, and simple floral units. Amborellaceae culature and the lack of stomata. The family Amborellaceae has attracted considerable interest in the recent years , being unique in angiosperms in lacking pollen tectum and being inaperturate to lacerate. Amborellaceae (Figure 9.18) are Loconte (1996) carried out cladistic analy-sis of the above taxa proposed by dif ferent authors as most basal angiosperms. He in-cluded 69 taxa in the study scor ing 151 Phylogeny of Angiosperms 285 primitive living angiosperms being woody or herbaceous, the general features of primitive angiosperms are largely settled. They have simple alternate exstipulat e leaves, which are entire and petiolate with poorly organized reticulate venati on and with unilacunar, two-trace nodes. The vessels are absent or tracheid-like . Flowers are bi-sexual, radially symmetrical with spirally arranged floral part s. Stamens are broad, undifferentiated with marginal microsporangia. Carpels are broad with large number of ovules, stigma along the margin and not completely s ealed, ovules bitegmic, crassinucellate. Fruits are follicular. B Figure 9.18 Amborellaceae. Amborella Coevolution with Animals Studies on comparative morphology, pollination biology and biochemistry have cle arly elucidated the role of animals in the evolution of angiosperms. It is sugge sted that Animal kingdom and Plant kingdom, particularly the Angiosperms have un dergone a pro-cess of co-evolution, wherein the evolution of one has influenced

the other. This has proceeded in various ways. trichopoda. A: Fully opened fe-male flower; B: Close up of a branch; C: Male flo wer. (photo B, courtesy University of Cali-fornia, Santa Cruz; A, photo courtesy Missouri Botanical Garden). shrubs without vessels, with unilacunar nodes, 2-ranked, exstipulate leaves; the margins are both serrate and rather undu-late. The plant is dioecious and the f lowers are small in cymes, with an undifferenti-ated perianth of spirally arrang ed 5-8 tepals. The staminate flowers have 10-25 stamens, sessile anthers and pol len with granulate ektexine. The carpellate flowers have 1-2 staminodes and 5-6 whorled incompletely closed carpels that develop into drupelets with pock-marked stones and pockets of alPollination most resinous substances. Early seed plants, the gymnosperms were wind pollinated with sticky sap exuding from micropyles trapping the pollen. Early insects, the beetles were probably at tracted to this sap and pollen by chance. The better pollination and increased s eed set encouraged the selection towards showy flowers more at-tractive to insec ts, edible flower parts, pro-tein rich pollen, nectaries and bisexual flow-ers s o that same insect visit can both deposit the pollen and pick up for visit to an other flower. Increased visits by insects posed danger to the exposed seeds, res ulting in selection towards protection of seeds in closed carpel, a major step t owards the evolution of angiosperms. Increased protection of seeds encouraged sm aller seeds in increased numbers and shorter life cycle to overcome drought cond itions. Complete closure of carpel was accompanied by the Relationships at the base of the an-giosperm lineage are being clarified. Ambore llaceae are most likely to be sister to other angiosperms, Nymphaeaceae sis-ter to the rest, then Austrobaileyales. EVOLUTIONARY TRENDS Although there has been some recent con-troversy regarding the habit of the most 286 Plant Systematics differentiation of stigmatic region for receiv-ing pollen, and the distinct styl e to keep the stigma within the reach of insects. To suite to the floral mechani sms the early beetles were slowly replaced by higher insects such as moths, butt erflies, bees, wasps and flies, coinciding with the floral diversification of ingested by the insect larvae elevate the level of hormone, resulting in their d evelop-ment into abnormal asexual adults. The larvae as such, learn to avoid such plants. angiosperms. Some plant products help insects against predators. Monarch butterfly, for examp le, ingests cardiac glucoside from milkweed Asclepias. Such butterflies if inges ted by blue jays make latter violently sick. Blue jays learn to recognize the to xic brightly coloured monarch butterflies. The milkweed, thus helps to protect m onarch butterfly from blue jay. Beetle pollinated flowers are typically dull or white with fruity odours, edible petals and heavily protected seeds. Bee pollinated flow-ers are brightly colour ed (blue or yellow but not red) with honey guides and with lot of pol-len and ne ctar. Butterfly pollinated flowers are red, blue or yellow. Moth pollinated flow ers mostly open at night and have heavy fragrance to attract moths. Moth and But terfly pollinated flowers generally have long corolla tubes with nectaries at th e base. Bird polli-nated flowers are bright red or yellow, pro-duce large amount of nectar, with little or no fragrance. Bat pollinated flowers are dull coloure d, open at night and have fruity odour. Basic evolutionary trends Evolution within Angiosperms has proceeded along different lines in different gr oups. Nu- merous trends in the evolution of an-giosperms have been recognized fr om comparative studies of extant and fossil plants. The general processes involv ed in attaining diversity of angiosperms are underlined below. Biochemical coevolution Plants and their insect predators are believed to have undergone adaptive radiat

ion in stepwise manner, with the plant groups evolving new and highly effective chemical defenses against herbivores and the latter continually evolving means o f overcoming these defenses. Mustard oils of Brassicaceae are toxic for many ani mals, yet they attract other herbivores such as cabbage worm which uses the must ard oils to locate the cabbage plant for laying its eggs. The chemi-cal hyperici n in genus Hypericum repels al-most all herbivores but the beetle genus Chrysoli na can detoxify hypericin and use it Fusion During the course of evolution in an-giosperms, fusion of different parts has le d to floral complexity. Fusion of like parts has led to the development of gamos epaly, gamopetaly, synandry and syncarpy in various families of angiosperms. Sta mens have shown fusion to different degrees: fusion of filaments only (monadelph ous condition in Malvaceae), fusion of anthers only (syngenesious condition foun d in Asteraceae) or complete fusion (synandry as in Cucurbita). Carpels may simi larly be fused only by ovaries (Synovarious: to locate the plant. Caryophyllaceae), only by styles (synstylous: The evolution of new chemical defense of Apocynaceae) or complete fusion of both ova-plant has resulted in plants often acquiring ries and styles (Synstylovario us: Solanaceae, the growth hormones found in insect larvae. Primulaceae). Fusion of unlike parts has re-Proper levels of juvenile hormone in insect sulted in an epipetalous condition (fusion of larvae are essential for the hatching of in- p etals and stamens), formation of sect larvae into normal sexual adults. Sev- gyn ostegium (the fusion of stamens and gyeral species of plants such as Ageratum co n- noecium: Asclepiadaceae) and formation of tain hormone juvabione, similar to the ju- an inferior ovary (fusion of calyx with ovary: venile hormone of insects . Such plants if Apiaceae, Myrtaceae, etc.). Phylogeny of Angiosperms 287 Reduction Remoration The term was suggested by Melville (1983) to refer to evolutionary retrogression found in angiosperms and their fossil relatives. The fertile shoots of angiospe rms, accord-ing to him, show venation pattern changes progressively from vegetat ive leaves through successive older evolutionary stages in bracts and sepals, an d the most ancient in petals. The innermost parts in a bud as such represent the most primitive evolutionary condition, and the outermost the most reRelatively simple flowers of many families have primarily been the result of red uction. The loss of either stamens or carpels has resulted in unisexual flowers. The loss of one perianth whorl has resulted in monochlamydeous forms, and their total ab-sence in achlamydeous forms. There has also been individual reduction in the num-ber of perianth parts, number of stamens and carpels. Within the ovar y different genera have shown reduction in the number of ovules to ultimately on e, as seen in the transformation of follicle into achene within the family Ranun culaceae. There has also been reduction in the size of flowers, mani-fested in d iverse families such as Asteraceae and Poaceae. Reduction in the size of seeds h as been extreme in Orchidaceae. Male flower of Euphorbia pre-sents a single stam en, there being no peri-anth or any trace of a pistillode, only a joint indicate s the position of thalamus and the demarcation between the pedicel and the filam ent. cent condition. There has been some shift in the under-standing of angiosperm phylogeny to suppo rt the stachyosporous origin of angiosperm carpel (Taylor and Kirchner, 1996). W ith the acceptance of such a viewpoint, the repro-ductive axis with many flowers , few carpels per flower and few ovules per carpel are ancestral. Evolution proc eeded along two directions from this: one with few flowers, each of which had ma ny carpels and few ovules and the other with few flowers, each containing few ca rpels and many ovules. The evolutionary trends in angiosperms are thus often com

plicated and frequent reversal of trends may be encountered, as for example the secondary loss of vessels in some members. Change in Symmetry From simple radially symmetrical actinomorphic flowers in primitive flowers deve loped zygomorphic flowers in various fami-lies to suit insect pollination. The s ize of corolla tube and orientation of corolla lobes changed according to the mo uthparts of the pollinating insects, with striking specializa-tion achieved in t he turn-pipe mechanism of Salvia flowers, and female wasp like flow-ers of orchi d Ophrys. Xylem evolution Elaboration Xylem tissue of angiosperms largely consists of dead tracheids and vessels, supp orting fibers and living ray cells. Tracheids are elongate, imperforate water co nducting cells found in almost all lower vascular plants, gymnosperms and angios perms. Vessels are perforate elements largely restricted to an-giosperms, althou gh also found in extant gnetopsids, some species of Equisetum, Selaginella, Mars ilea and Pteridium. The presence of vessels in gnetopsids, the closest relatives of angiosperms, had given rise to the speculation that the latter arose from fo rmer. The studies of Bailey and associates, however, showed that the vessels in the two groups arose independently. In gnetopsids, This compensating mechanism has been found in several families. In Asteraceae an d Poaceae, the reduction in the size of flowers has been compensated by an incre ase in the number of flowers in the inflorescence. Similarly, reduction in the n umber of ovules has been accompanied by an increase in the size of ovule and ult imately seed, as seen in Juglans and Aesculus. 288 Plant Systematics

>sa

I { D

F E J A B H G Figure 9.19 Presumed evolutionary transformation of gymnosperm tracheid with cir cular pitting (A) to angiosperm tracheid with scalariform pitting (B), further to vessel-eleme nt with oblique perforation plate with numerous scaraliform bars (C). Further sh ortening and broadening of vessel-element, perforation plate becoming more and m ore horizontal and reduction in the number of bars in the perforation plate ulti mately led to shortest, broad vessel element with transverse simple perforation plate. Vessel-element E shows one scalariform and one simple perforation plate. Note conspicuous tails, a reminder of tracheids in vessel-elements D, E and F wi th still oblique perforation plates. J represents the vessel-element of Quercus alba, being more broader than long and with a simple large perforation.

they developed from tracheids with circular pitting and in angiosperms from trac heids with scalariform pitting. It is also pointed out by Carlquist (1996) that circular pitting in the vessels of gnetopsids, as also the gym-nosperm tracheids in general, is different from angiosperms in having pits with torus and pit mar go with pores much larger than those of angiosperms. Although some an-giosperms do have pit membrane with torus, the pit margo is always absent. Because all fossil and almost all extant gymnosperms possess tracheids with circ ular-bordered pitting, it has led to the conclusion that the tracheid is the most primitive type of tracheary element in the an giosperms. As tracheids have given rise to vessel elements, the most primitive t ype of vessels have long narrow vessel elements with tapering ends. The tracheid s in angiosperms have scalariform pitting, and as such it is assumed that these tracheids Phylogeny of Angiosperms 289 arose from tracheids of gymnsperms with circular pitting. In the transformation of tra-cheids with scalariform pitting to vessel el-ements with scalariform pitt ing, the earli-est elements had perforation plates with nu-merous scalariform ba rs. During further evo-lution of vessels, elements became smaller and broader, a nd perforation plates more horizontal. There has been an accompanied reduction i n the number of scalariform bars, resulting in shortest broadest vessel ele-ment s with simple perforation plate and transverse end wall (Figure 9.19) in most ad vanced forms. Hamamelididae were once regarded to be primitive due to their simp le floral structure, but have advanced vessel elements and thus considered advan ced over Magnoliidae with primitive elongated narrow vessel elements. This is su pported stem metaxylem by Takahashi, 1988). This led Carlquist to conclude that vessels have originated numerous times in dicotyledons. It had often been held that vessels arose first in the secondary xylem and later in the metaxylem, and that specialization has gradually advanced from the secon dary to primary xylem. Carlquist pointed out that scalariform pitting is widespr ead in the met-axylem of vascular plants and if primitive angiosperms were herbs , in accord with paleoherb hypothesis, metaxylem would be expected to have scala riform pitting of tracheary elements. The development of woody habit if featured p aedomorphosis would extend scalariform patterns in secondary xylem. by studies on floral anatomy and palynology. Stamen Evolution Carlquist (1996), based on a survey of wood anatomy, has identified a number of distinct evolutionary trends in angiosperms. The cambial initials have shortened , the ratio of length accompanying fibers to vessel ele-ments (F/V ratio) has sh own an increase from 1.00 in primitive dicotyledons to about 4.00 in the most sp ecialized woods, and an-gular outline of vessels changed to circular outline tog ether with widening of their di-ameter. There was also a progressive reduc-tion in the number of scalariform bars, ulti-mately resulting in simple perforation, fa-cilitating an easier flow of water. The lat-eral walls showed a shift from sc alariform to the opposite circular pits and finally to al-ternate circular pits, so as to provide better mechanical strength. Imperforate tracheids have shown a shift to fibre-tracheids to fi-nally libriform fibres. Shortening of fusiform i nitials is correlated with storeying of woods. It is interesting to note that cl adistic stud-ies have shown that present-day vessel-less angiosperms do not form a single clade and are distributed in diverse groups such as Hamamelidales (Tro chodendron and The most primitive type of Stamen in an-giosperms represented in genera like Deg eneria, Austrobaileya, Himantandra and Magnolia (Figure 9. 20) and other primiti ve genera is laminar, 3-veined leaf-like organ without any clear cut distinction of fertile and sterile parts. The pollen sacs (sporangia) are borne near the ce ntre either on the abaxial (dorsal) side (Degeneria, Annonaceae and Himatandrace ae) or on adaxial (ventral) side (Austrobaileya and Magnolia). Semilaminar stame ns occur in some other primitive families like Nymphaeaceae, Ceratophyllaceae an

d Eupomatiaceae. In further specialization of stamen, there has been reduction o f sterile tissues and retraction of marginal areas. The proximal part became fil ament and distal part the anther. The midvein re-gion formed the connective, and distal part the appendage, as seen in several genera. Tetracentron), Magnoliales (Amborellaceae, Winteraceae), and Laurales (Sarcandra : VesIn primitive families, connective forms a major part of anther. In more advanced families, the connective is highly reduced (Acanthaceae, Plantaginaceae) or may be almost absent. In some families such as Betulaceae, the connective as well a s the upper part of filament may become divided and two anther lobes get separat ed. In more primitive families the connective is produced above into an appendag e which sels have been, however, reported in root secondary xylem by Carlquist, 1987 and in 290 Plant Systematics -A,

lI

E / ~ M

T (i H Figure 9.20 Evolution of stamen in angiosperms. A-D: Primitive laminar stamens w ithout clear distinction of anthers and filaments; A: Austrobaileya scandens with adaxial pol len sacs (microsporangia); B: Himantandra baccata with abaxial pollen sacs; C: D egeneria vitiensis with abaxial pollen sacs; D: Magnolia maingayi with adaxial p ollen sacs; E: Laminar stamen of Magnolia nitida with marginal pollen sacs and p rolonged sterile appendage; F: Semilaminar stamen of Michelia fuscata with margi nal pollen sacs and narrowed filament; G: Outer semilaminar stamen of Nymphaea o

dorata with petaloid filament and narrow anthers; H: inner stamen with narrower filament and differentiated anther region; I: Stamen of Illicium parviflorum wit h reduced anther region and broad filaments; J: Stamen of Opuntia pusilla with w ell differentiated anthers and filament; K: Stamen of Poa pratensis with reduced connective and thread-like filament; L: Stamen of Penstemon canescens with well -defined filament and large anthers with distinct anther lobes but reduced conne ctive; M: Stamen of Betula nigra with divided connective and filament. disappears progressively in more advanced It is generally agreed that primitive stamen was laminar, with two pairs of spor angia, borne on adaxial or abaxial side, be-cause both situations are met in pri mitive families. During the course of evolution, the stamen became more slender, its laminar form slowly disappearing, and sporangia occupying marginal position . The transformation of broad laminar to narrow stamens is clearly depicted in N ymphaea from outer to inner stamens and in different species of the genus. families. Broad laminar filament, merging with the rest of stamen represents the most prim i-tive state. It becomes narrower and finally terete in advanced families. The s tamens with well marked narrow filament may have basal, dorsal or versatile atta chment with anthers. Basifixed condition is the most primitive, versatile the mo st advanced of-ten commonly seen in grasses and Amaryllidaceae. Phylogeny of Angiosperms 291 A typical anther of angiosperms is bithecous with two anther lobes with the two anther sacs in each, finally merging into one. The anther with a single anther l obe, as in Malvaceae, has a single final sac (theca) and as such monothecous. Th e monothecous anthers of Malvaceae and some other families result from splitting o f stamens, thus separating the two anther lobes. In others, like Salix, there ma y be a partial connation of two stamens resulting

C 191

IV G V VI w H VII I 4 VIII xi Walker and Walker, 1981 Brenner, 1996

Figure 9.21 Two different models for evolution of exine and pollen grains in Ang iosperms. Walker and Walker, 1984 (on left). A: Exine of ancestral gymnosperm with homogenous sex ine and laminated nexine; B: Same but sexine with granular infratectal layer; C: Pollen grain of same, boat shaped and monosulcate; D: Exine of most primitive a ngiosperm with smooth sexine and disappearance of laminated nexine; E: Same but with homogenous sexine; F: Monosulcate pollen of same; G: Exine with development of tectum, infratectal layer and homogenous nexine; H: Same but with loss of te ctum.: I: Monosulcate pollen grain with intectate exine. Brenner, 1996 (on right ). I: Exine of early angiosperm, tectate-columellate and without aperture; II: E xine with initiation of endexine (shaded solid black); III. Nonaperturate early pollen with circular outline; IV: Exine with complete endexine layer and initiat ion of sulcus; V: Exine with developed sulcus; VI: Monosulcate pollen of basal a ngiosperms; VII: Monosulcate boat-shaped pollen of Magnoliids and monocots; VIII : Circular monosulcate pollen of dicots; IX: Tricolpate pollen of Eudicots which might have developed from monosulcate or inaperturate forms; X: Uniporate polle n; XI: Uniporate pollen of Winteraceae in tetrads. (Modified from Brenner, 1996) . 292 Plant Systematics shows two independent vascular supplies derived from opposite sides of the recep tacle, as opposed to families where splitting of sta-mens occurs. lar in outline, tectate columellate, and with-out aperture. A possible intine th ickening was accompanied by developments of endexine layer above intine in Haute rivian. The next step involved evolution of sulcus and divergence of monocot and dicot pollen types from basic dicot stock. In Barremian diver-sification of mon osulcate pollen grains occurred with migration to different geographical regions . In Lower Aptian tricolpate pollen evolved from either monosulcate or inapertur ate forms in northern Gondwana, Pollen grain evolution A large number of families in monocots and several primitive dicots of the magno liid complex bear monosulcate pollen grains, a condition generally considered to be the primitive one in angiosperms. Walker and Doyle (1975) and Walker and Wal ker (1984) suggested that the primitive angiosperm pollen grain is large- to med ium-sized, boat-shaped, smooth-walled, with homogenous or granular infratectal l ayer, the tectum being absent (pollen atectate) and endexine (a layer unique to angiosperms, being absent in gymnosperms) either missing or poorly developed und er the apertural area. This type of pollen is found among extant an-giosperms in Annonaceae, Degeneriaceae and Magnoliaceae. The prototype of this was gymnosper m pollen which was monosulcate, large, boat-shaped with laminated nexine, homoge nous sexine or with granular infratectal layer (Figure 9.21), a pollen type comm on in Bennettitales, Gnetopsids (ex-cluding Gnetum) and Pentoxylon. In evolution of angiosperm pollen from this, the lami-nated nexine disappeared, nexine becam e granular to tectate with columella and re-ticulate surface to those with intec tate resulting in evolution of eudicots. It is suggested by Brenner that the for-mation of sulcus during Early Cretaceous may have been an adaptation that was a more effective way of releasing recognit ion proteins involved in pollen-tube development while the later development of tricolpate con-dition in the Aptian would be a further extension of this process. The formation of endexine before aperture development may reflect the developmen t of intine thickening, which is related to sulcus development. In extant angios perms, the intine beneath the aperture stores recognition proteins. Carpel evolution collumellae. Carpel is a structure unique to angiosperms enclosing and protecting ovules. The evolution of carpel probably played a major role in diversity and success of an giosperms as it not only protected seeds from predators, but also carried associ ated benefits. These included seed dissemination via the evolution of numerous d

ispersal mechanisms, effective fertilization by transport of pollen grains to th e stigma and growth of a pollen tube, promotion of outbreeding by insect pollina tors through the evolution of special structural mechanisms and through the deve lopments of intraspecific and interspecific Brenner and Bickoff (1992) recorded glo-bose inaperturate pollen grains from the Valanginian (ca 135 mya) of the Helez for-mation of Israel, now considered to b e the oldest record of angiosperm fossils. These pollen grains resemble those of Gnetum in general shape and lack of aperture, and also found in Chloranthaceae, Piperaceae, and Saururaceae, which are gaining increased attention as basal ang iosperm families. This led Brenner (1996) to postulate a new model for the evolu tion of angiosperm pollen. The earliest pollen in angiosperms, developed in Vala nginian or earlier from stock that also gave rise to Gnetum, had small pollen, c ircuincompatibilty. It is more common in recent years to differentiate three types of carpels, a ter minology developed by Taylor (1991). Ascidiate carpels have ovules attached prox imally to the closure, plicate ones have ovules Phylogeny of Angiosperms 293 attached along margins of the closure and ascoplicate carpels are intermediate b ePhyllosporous origin tween the two. Among the believers of phyllosporous origin, suggest that carpel is a folded lea f with adaxial surfaces (conduplicate), or involute abaxial surfaces in contact (involute) with many ovules along the margins (or submargins) of closure (Bailey and Swamy, 1951; Eames, 1961). Others suggest that the leaf is fundamentally pe ltate (Baum, 1949; Baum and Leinfellner, 1953), as many carThe nature of carpel in angiosperms has been the subject of considerable discuss ion. The dominant view supported by Bailey and Swamy (1951), Cronquist (1988), T akhtajan (1997) and several others considers carpel as homologous to megasporoph yll, appropri-ately named as phyllosporous origin of car-pel (Lam, 1961). Others believe that carpel consists of a subtending bract with placenta representing a shoot with distally placed ovules, concept named as stachyosporous origin (Pank ow, 1962; Sattler and Lacroix, 1988). pels have cup-shaped primordia. The conduplicate view of the origin of car-pel was advocated by Bailey and Swamy (1951). In primitive type of carpel, the stigma is represented by a crest exten ding from 1

P A C R G J K L M

Figure 9.22 Phyllosporous concept of carpel evolution. A-I, conduplicate closure ; J-N, involute closure; O-U, closer along margins. A: Carpel of Drimys piperita with long stigm atic crest; B: Transverse section of same showing partially closed margins; C: T ransverse section of carpel of Degeneria vitiensis with flared up margins and co nspicuous papillose growth; D-F: Stages in conduplicate closing of carpel with d isappearance of stigmatic region (broken lines) from the body of carpel and its localization towards the tip, resulting in marginal placentation; G: Fusion of a djacent carpels by lateral cohesion of open conduplicate carpels forming parieta l placentation; H: Fusion by adnation of free margins of conduplicate carpels to the thalamus forming axile placentation; I: Fusion by cohesion of ventral surfaces of the carpels formong axil e placentation. J-K: Involute closing of carpels by meeting of dorsal surfaces o f carpels. M-N: Examples of fused carpels with involute closure. L: Erythraea ce ntaurium; M: Isanthus brachiatus; N: Limnophila heterophylla. O-Q: Closure by fu sing margins of carpel with ultimate merging of ventral bundles; R: Fusion of ma rgins of adjacent opens carpels with merging of ventral bundles and formation of parietal placentation. S-U: Fusion of sides of closed carpels with merging of a djacent lateral bundles and ventral bundles resulting in axile placentation. (AH based on Bailey and Swamy, 1951; J-N based on Eames, 1961; O-U after Eames and MacDaniels, 1947). 294 Plant Systematics apex to base of carpel (decurrent) as in some species of Drimys, Himantandra and Degeneria. In Degeneria and Butomus, the double nature of the crest is evident in mar-gins flaring back from the line of contact. In Degeneria and Drimys, the margins of car-pels are incompletely closed by interlocking papillose cells of t he stigmatic crest. The carpels have three traces, one dorsal and two ventral, t he latter providing vascular supply to ovules. From this type of carpel, the clo sed tively primitive families. The ovules in these carpels derive their vascular sup ply chiefly from a smaller meshwork of the bundles, and rarely from dorsal bundl e or ventral bundles. There was consequent reduction in the number of ovules and their restriction to submarginal position with vas-cular supply coming from ven tral bundles. The evidence of this transition is seen in Winteraceae and Degeneriaceae. carpel of other angiosperms developed by cloClosing of carpels may also result from fusing of incurved margins of carpels. P rosure of adjacent adaxial surfaces (Figure gressive fusion results in final fusio n of ad9.22 D-F) and concentration of stigmatic jacent ventral bundles (Figure 9.22 O-P ) in margins to the upper part of the carpel, re- follicular carpel. Fusion of margin s of adjaduction in the number of ovules and their cent open carpels results in parietal placenrestriction to lower part differentiating as tation with only ventral bundles ul timately ovary, the middle sterile portion forming the merging, whereas the fusi on of sides of style. The fusion of adjacent carpels in the closed carpels resul ts in axile placentation formation of syncarpous gynoecium may with both adjacen t lateral bundles as well have proceeded along different directions. as ventral bundles merging. Lateral cohesion of open conduplicate car-pels resulted in unilocular ovary with pari-etal placentation (Figure 9.22 G). Axile pla-centation (with number of loc ules equalling the number of fusing carpels) may have re-sulted from adnation of free margins to the thalamus (Figure 9.22 H) or cohesion of ven-tral sutures of carpels (Figure 9.22 I). Differ-ent families of angiosperms exhibit differ-ent degrees of fusion, some like Caryophyllaceae with free styles and stig-mas, othe rs like Solanaceae with complete fusion of ovaries, styles and stigmas. Free cen tral placentation may result from disso-lution of septa from ovary with axile pl

acen-tation (Caryophyllaceae) the placental col-umn being attached to the base a nd top of the ovary. It may also result from protruding thalamus carrying the pl acenta from the base of the ovary (Primulaceae). The basal placentation with the number of ovules re-duced to basal one may be derived from one (Alismataceae) o r more than one carpels Involute closing of carpels was advocated by Joshi (1947), Puri (1960) and sever al other workers. In such carpels, the margins of the carpel are involuted and a baxial (and not adaxial surfaces) or margins are in contact. The example of such carpels involved in the formation of syncarpous gynoecium is seen in several ge nera (Figure 9.22 L-N). Although there have been suggestions that involute types may have evolved from conduplicate types, Eames (1961) considered it highly unl ikely, as such a derivation would involve change from contact by adaxial sides t o con-tact by abaxial sides, a major change, far more complicated and circuitous than usually found in evolutionary derivation. He suggested that several indepe ndent closure of carpels occurred in different phylogenetic lines. (Asteraceae). The theory that the carpel is a peltate leaf was strongly advocated by Baum and Leinfellner (1953). The peltate form of carpel is assumed to have arisen by turn ing upward (ventrally of the basal lobes of the lamina and their fusion, margin to margin, as in the formation of peltate leaves. A trans-verse meristem, known as cross zone, develops where the two marginal meristems meet. As the carpel pri mordium elongates, Laminar placentation with ovules scat-tered over the entire inner surface of the ovary wall is considered to be the most primi-tive type, present in Nymphaeacea e, Cabombaceae, Butomaceae and other relaPhylogeny of Angiosperms 295 Sister Groups Angiosperms Kirchner (1996). The theory is gaining increased interest with the renewed inter est in gnetopsids as close relatives of angiosperms. The theory holds that the c arpel envelope represents a bract and placenta homologous with a shoot bearing d istally placed ovules. This bract-terminal ovule system is directly homologous t o the one found in outgroups including Gnetales, the closest living sister-group . According to this theory, ascidiate carpel with few ovules represents an ances tral stage. It is believed that the origin of plicate (conduplicate) and ascopli cate carpel types would be due to in-tegration of the gynoecial primordia and ov uancestral derived few flowers many (> 7) carpels/flower Many flowers , many flowers 1 or 2 ovules/ 1 or 2 carpels/flower flower 1 or 2 ovules/carpel ew flowers 1 or 2 carpels/flower, many (> 7) ovules/carpel Figure 9.23 Evolution of carpel and floral types in angiosperms based on stachyosporous model proposed by Taylor and Kirchner (19 96). An-cestral type is based on suggested homologies between female struc-tures of sister groups. lar (placental) growth areas. Taylor and Kirchner found further evidence for stachyosporous origin from: 1. Ingroup phylogeny based on structural

and DNA sequences, rbcL cpDNA datasets which place either woody or herbaceous ma gnoliids as basal clades suggest that ascidiate carpels with 1 the cross zone, continuous with the marginal meristems, is said to build up a ve ntral strip of carpel wall, which, united with the lateral walls, forms a tubula r organ. Under this theory, the ovules are borne on the wall formed by the cross zone. Peltate carpels may be manifest peltate carpels with well-de-fined stalks and tubular lamina with well-marked cross zone (Thalictrum) or latent peltate w ith short tubular base and cross zone present only in early ontogeny (Calycanthu s). According to this theory, the achene of Ranunculus having a single ovule on latent cross zone is considered to be the first stage in the building of tubular follicle. This theory is reverse of the most commonly held view that the follic le represents a more primitive state, and that the achene of Ra-nunculus is deri ved. or 2 ovules represent ancestral state. 2. Outgroup analysis involving Bennettitales, Gnetales and Cordaitales which suggest that female reproductive structures are compound organs. 3. Morphogenic analysis involving unStachyosporous origin derstanding of floral development in Antirrhinum and Arabidopsis through mutagen ic analysis, study of the development of carpel in Datura by examining chromosom al chimeras (car-pel wall is similar to petals and leaves in development; carpel with two types of primordia, one forming the wall and one with distinct central ridge which develops into septum, placenta and false septum and functions like floral apex), study of carpel development in Nicotiana using Ac-GUS reporter sys tem using GUS bacterial gene as marker (showing that whereas carpel wall is comp osed only L1 and L2 lay-ers, the placental region has an addiThe idea was first developed by Hagerup (1934, 1936, 1938) who suggested that co n-duplicate carpels have two growth areas. The theory was further developed by L am (1961), Melville (1962, 1983 who proposed a varia-tion of this as gonophyll the ory) and more recently Taylor (1991) and Taylor and tional L3 layer). Based on new evidence, Taylor and Kirchner concluded that ancestral carpel is 296 Plant Systematics ascidiate with marginal stigma and basal to slightly lateral placentation of one or two orthotropous ovules. The evolution of curved ovules and placement of the ovules in other positions was to direct the micropyle away from the stigma or p ollen-tube transmission-tissue. They suggested that reproductive axes with many flowers, few carpels per flower, and few ovules per carpel were antheory (axial theory) developed by German school of botanists and supported by S chleide, Eichler, Sachs and others, according to which inferior ovary resulted f rom invagination of the floral receptacle which surrounds the ovary. cestral (Figure 9.23). The accumulating anatomical evidence has shown that inferior ovary has evolved a number of times in different groups of angiosperms, in some due to adnation of floral parts and in others due to axial invagination. In certain plants like Hed era, separate traces related to different floral organs are found, in others lik e Juglans, different stages of bundle fusion can be found in the inferior ovary. Such plants also have normal orientation of vascular bundles (phloem outside, x ylem inside) and evidently, the inferior From such an ancestral type developed two types of inflorescences: one with few flowers, each of which had many carpels and few ovules and the other with few fl owers each containing few carpels and many ovules. Gonophyll theory Melville (1962, 1983) developed his gonophyll theory largely on the basis of stu dies of vas- culature in leaves and floral whorls. This theory is a variation of

stachyosporous oriovary is appendicular in origin. An axial invagination of floral receptacle will eventually result in inverted va scular gin. According to him, the ovary consists of sterile leaves and ovule-bearing br anches attached to the petiole of the leaf. Each leaf, together with the fertile branch, is considered a unit and termed as gonophyll instead of carpel. This th eory has already been discussed under probable ancestors of an-giosperms. bundles (xylem on outside, phloem inside) in the inner part of inferior ovary, w ith nor- mal orientation in the outer part. This has been observed in the inferi or ovaries of Cactaceae and Santalaceae. In others like Rosa, the lower part of the fleshy receptacle has invaginated receptacular tissue whereas the upper port ion consists of fused floral parts. The adnation of floral parts above or surrou nding the ovary in a large number of plants forms hypanthium, a structure distin ct from but often confused with the calyx tube, the latter involving the cohesio n of sepals only. The development of inferior ovary has occurred within several families, as genera with superior ovary and those with inferior ovary may be enc ountered in the same family as seen in Rosaceae, Gesneriaceae, Nymphaeaceae and several others. In Nymphaeaceae, Nuphar has superior ovary, Nymphaea semi-inferi or and Euryale superior ovary. Evolution of Inferior ovary It has been universally agreed that the in-ferior ovary in angiosperms is a deri ved state. The nature of origin of this type had two opposing views. Linnaeus, d e Candolle and many early botanists believed in the origin of an inferior ovary through adnation of bases of outer floral whorls to the gyno-ecium, a view known as appendicular theory (Candollean theory, concrescence theory). Others believe d in the receptacular Chapter 10 Systems of Classif ication The urge to classify plants has been with man since he first set his foot on thi s planet, borne of a need to know what he should eat, avoid, use as cures for ai lments and utilize for his shelter. Initially, this information was accumulated and stored in the human brain and passed on to generations through word of mouth in dialects restricted to small communities. Slowly, man learned to put his Preliterate Mankind Although no written records of the activities of our preliterate ancestors are a vailable, it is safe to assume that they were practical taxonomists having acqui red knowledge as to which plants were edible and which cured their ailments. Pri mitive tribes in remote areas of the world still carry the tradition of preservi ng knowledge of the names and uses knowledge in black and white for others to share and improve upon. We have now of plants by word of mouth from one generareached a stage whereby a vast amount of tion to another. Such classifications o f information can be conveniently stored and utilized for far-reaching conclusions aimed plants developed by isolated communities through the need of the society and wit hout the influence of science are termed Folk taxat developing ideal systems of classification, which depict the putative relationships be- onomies; often parallel modern taxon omy. tween organisms. Historical development of The common English names grass a nd sedge classification has passed through four disare equivalent to the modern families tinct approaches, beginning with simple classifications based on gross morpholog y to the latest phylogenetic systems incorporating all types of phenetic informa tion. Poaceae and Cyperaceae and illustrate this parallel development between folk tax

onomy and modern taxonomy. Early Literate Civilizations CLASSIFICATIONS BASED ON GROSS MORPHOLOGY Early civilizations flourished in Babylonia, Egypt, China and India. Though the written records of Indian botany appeared several centuries before those of the Greeks, they re-mained in obscurity, not reaching the outside world. Moreover, t hey were written in San-skrit, a language not easily understood in the West. Cro ps such as wheat, barley, dates, Classifications based on features studied without microscopic aids continued unt il the seventeenth century, when the naked eye was the sole tool of observation. The trail backwards leads us to preliterate man. 298 Plant Systematics melons and cotton were grown during the Vedic Period (2000 BC to 800 BC). Indian s ob-viously knew about descriptive botany and cultural practices. The first wor ld symposium on medicinal plants was held in the Hima-layan region in the sevent h century BC. The Atharva Veda, written around 2000 BC de-scribes the medicinal uses of various plants. Leucippus or Alcippus, Theophrastus then proceeded to Athens, and became a membe r of the Platonic circle. After Plato s death, he attached himself to Aristotle , and after the latter s death, he inherited his library and the garden. He rose to become the head of the Lyceum at Athens. Theophrastus Father of botany Theophrastus (Figure 10.1) is credited with having authored more than 200 works most of which survive as fragments or as quotations in the works of other author s. Two of his botanical works have survived intact, how-ever, and are available in English translations: Enquiry into plants (1916) and The Causes of plants (19 27). Theophrastus described about 500 kinds of plants, classified into four majo r groups: the trees, shrubs, subshrubs and herbs. He also recognized the differe nces be-tween flowering plants and non-flowering plants, superior ovary and infe rior ovary, free and fused petals and also fruit types. He was aware of the fact that many cultivated plants do not breed true. Several names used by Theophrast us in his De Historia plantarum, e.g. Daucus, Crataegus and Narcissus, to name Theophrastus (372 BC to 287 BC), the suc-cessor of Aristotle in the Peripatelic School (those following the philosophy propagated by Aristotle), was a native of Eresus in Lesbos. W .

, t r v a few, are in use even today. ~L 8 Theophrastus was fortunate to have the patronage of Alexander the Great. During his conquests, Alexander made arrangements to send back materials to Athens, ena bling Theophrastus to write about exotic plants such as cotton, cinnamon and ban anas. Botanical knowledge at the Lyceum in Athens thus flourished during this tr uly golden age of learning, whose botanical advance Theophrastus was privileged

to steer. RM

C r AI< ! :ry -:.

Parasara Indian scholar Figure 10.1 Theophrastus (372 BC to 287 BC) the Greek philosopher, credited to Parasara (250 BC to 120 BC) was an Indian scholar who compiled Vrikshayurveda (S cience of plant life), one of the earliest works dealing with plant life from a scientific standpoint, a manuscript discovered a few decades ago. The book has s eparate chapters on morphology, properties of soil, forest types of India and de tails of internal structure, which suggest that the author possessed a magnifyin g apparatus of some kind. He also described the existence of cells (rasakosa) be the father of botany, wrote more than 200 manuscripts. His original name was Tyrtamus, but he later became known by the nickname Theophr astus given to him it is said by Aristotle, to indicate the grace of his con-versatio n. After receiving his first introduc-tion to philosophy in Lesbos from one Systems of Classification 299 in the leaf, transportation of the soil solu-tion from the root to the leaves wh ere it is digested by means of chlorophyll [ranjakena pacyamanat] into nutritive substance and Cicilia. Being a physician in the Roman army, he travelled extensively and gaine d firsthand knowledge about plants used for treating various ailments. He wrote a truly the by-products. Plants were classified into outstanding work, Materia medica, p resent-numerous families [ganas] on the basis of ing an account of nearly 600 me dicinal morphological features not known to the Eu- plants, nearly 100 more than Theophrastus. ropean classification until the eighteenth Excellent illustration s were added later. Writ-century. Samiganyan [Leguminosae] were ten in a straigh tforward style, the book was distinguished by hypogynous flowers, five an asset for any literate man for the next 15 petals of different sizes, gamosepalous cal yx centuries. No drug was recognized as genuand fruit being a legume. Svastikaga nyan ine unless mentioned in Materia medica. It [Cruciferae] similarly, were dif ferentiated as is no less a tribute to Dioscorides that a having calyx resemblin g a swastika, ovary beautiful illustrated copy of the book was pre-superior, 4 f ree sepals and petals each, six pared for Emperor Flavius Olybrius Onycius stame ns of which 2 are shorter and 2 car- around 500 AD, who presented it as a gift t o pels forming a bilocular fruit. Unfortunately, his beautiful daughter Princess Juliana this great scientific advance did not reach Anicia. The manuscript, bet ter known as Europe at that time, where scientific knowl- Codex Juliana, is a prize manuscrip t preedge was just making its debut. served in Vienna. Materia medica was not a deliberate attempt at classification but legumes, mints and umbels were described as separate groups. Among the other Indian scholars, Caraka (Charaka Ist century AD) wrote Caraka samhita (Charaka samhita) in which he rec-og nized trees without flowers, trees with flowers, herbs which wither after fructi fica-tion and other herbs with spreading stems as separate groups. This huge tre atise on Indian medicine, containing eight divisions, is largely based on a much

earlier treatise published by Agnivesh. A. C. Kaviratna translated it into Engl ish in 1897. Medieval Botany Caius Plinius Secundus During the Middle Ages (fifth to fifteenth century AD), little or no progress wa s made in botanical investigation. During this dark period in history, Europe an d Asia witnessed wars, famine and epidemics, and the only worthwhile contributio n was copying and recopying of earlier manuscripts, unfortu- nately often with e rrors added. The straw- berry plant was thus shown to have five leaf-lets instea d of three in several manuscripts. The manuscripts were lost at a faster rate th an they could be copied. Pliny the Elder The decline of the Greek Empire witnessed the emergence of the Romans. Pliny (23 AD to 79 AD), a naturalist who served under the Roman army, attempted a compila tion of everything known about the world in an exten-sive 37-volume work Histori a naturalis, 9 vol-umes of which were devoted to medicinal plants. In spite of a few errors and fanciful tales from travellers, the Europeans held this work in reverential awe for many centuries. Pliny died during the eruption of Vesuvius. Islamic Botany The ascent of the Muslim Empire between 610-1100 AD saw the revival of literacy. Greek manuscripts were translated and pre-served. Being practical people, they concentrated on agriculture and medicine and pro-duced lists of drug plants. Ibn -Sina, better known as Avicenna authored Canon of medi- cine, a scientific class ic along the lines of Materia medica. Another Muslim scholar, Ibual-Awwan, in th e twelfth century described Pedanios Dioscorides Dioscorides (first Century AD), of Greek par-entage, was a native of the Roman p rovince 300 Plant Systematics nearly 600 plants, and interpreted their sexuality as well as the role of insect s in fig pollination. Although Muslim scholars pro-duced several practical lists of drug plants, but did not develop any significant scheme of classification. German Fathers of Botany. Otto Brunfels (1464-1534) wrote Herbarium vivae eicone s in three volumes (1530-1536), a herbal that marked the beginning of modern tax onomy, and contained excellent illustrations pre-pared from living plants. The t ext, however, was of little value, comprising extracts from earlier writers. Jer ome Bock (Hieronymus Tragus), who lived between 1498 and 1554, wrote New kreuter buch in 1539, which contained no illustrations but did include accurate descript ions based on firsthand knowledge, also mentioning the localities and habitats. He described 567 species classified as herbs, shrubs and trees. The herbal, writ ten in German, was widely understood as compared to the manuscripts of earlier s cholars, which were in Greek and Latin, languages which had by then become obsol ete. Leonard Fuchs (1501-1566), regarded as more meritorious than his contempora ries, wrote De Historia stirpium in 1542, containing descriptions as well as ill ustrations of 487 species of medicinal Albertus Magnus Doctor Universalis Albertus Magnus (1193-1280 AD), called Doc-tor Universalis by his contemporaries and Aristotle of the Middle Ages by historians, is the best remembered naturali st of that period. He wrote on many subjects. The bo-tanical work De vegetabilis dealt with me-dicinal plants and provided descriptions of plants based on first hand information. Magnus is believed to be the first to recog-nize monocots and dicots based on stem structure. He also separated vascular and non-vascular plan ts. Renaissance The fifteenth century saw the onset of the Renaissance in Europe, with technical in-novations, mainly the printing machine and the science of navigation. Invent ion of the printing machine with movable type around 1440 ensured wide circulati on of manu-scripts. Navigation led to the successful ex-ploitation of botanical

wealth from distant places. plants (Figure 10.2). Valerius Cordus (1515-1544), whose tragic early death prevented him from becomin g the greatest of all herbalists, undertook to study plants from living material . He travelled in the forests of Germany and Italy, where unfortunately he fell ill and died at the young age of 29. His work Historia plantarum, pub-lished in 1561, many years after his death, contained accurate descriptions of 502 species , 66 apparently new. He was perhaps the first to show how to describe plants fro m nature accurately. Unfortunately, Konrad Gesner, the editor of his work chose to add illustrations, which were not only of poor quality, but also wrongly iden tified, and the work Herbalists Printing made books cheap. The first to be-come popular were medically-oriented books on plants. Specialists started producing their own botanical-medical books , which were easily understood as compared to ancient manuscripts. These came to be known as herbals and the authors who wrote these were known as herbalists. T he first herb-als were published under the name Gart der Gesundheit or Hortus sa nitatus. These were cheaply done and of poor quality. The out-standing herbals c ame from German herb-alists Otto Brunfels, Jerome Bock, Valerius Cordus and Leon ard Fuchs, constituting the suffered for no fault of Valerius Cordus At the times when herbals flourished in Germany, Pierandrea Mathiola was active in Italy, producing Commentarii in sex libros Pedacii Dioscorides in 1544, addin g many illustrations, though it was a commentary on Dioscorides. The Dutch Big T hree Rembert Dodoens, Carolus Clusius and Mathias de L obel spread the botanical kno wledge to Holland and France through Systems of Classification 301 Early Taxonomists With renewed interest in plants and extensive explorations of Europe, Asia, Afri ca and the New World, the list of plant names increased enormously, signifying t he need for a formalized scheme of classification, naming and description of pla nts. Botany, hitherto dependent on medicine, started to spread its wings as a sc ience per se.

4 ' ' Andrea Cesalpino (1519-1603) The first plant taxonomist Andrea Cesalpino was an Italian botanist who studied botany under Luca Ghini and became the Director of the Botanical Garden and later professor of botany and m edicine at Bologna. He went to Rome in 1592 as the personal physician to Pope Cl ement VIII. He prepared a herbarium of 768 well-mounted plants in 1563, which is still preserved in the Museum of Natural History at Florence. His work De Plant is libri in 16 volumes appeared in 1583 and contained descriptions of 1520 speci es of plants grouped as herbs and trees and further differentiated on fruit and seed characters. Cesalpino subscribed to the Aristotelian logic, taking decision s based on reasoning and not the study of features. It was not surprising, there fore, that he considered pith akin to the spinal cord of animals and leaves havi ng the sole role of protecting the apical bud. However, he highlighted the signi ficance of reproductive characters, an attitude not liked by his contemporaries, but having much bearing on the subsequent classifications of Ray, Tournefort an d Linnaeus. Figure 10.2 Illustration of Plantago major from Fuch s De Historia Stirpium (1542) (reproduced from Arber: Herbals, their origin and Evolution, 1938: used with per mission from Cam-bridge University Press). their herbals. William Turner, in his Herball (1551-1568), swept out many old su

perstitions concerning plants. Herball won for Turner the title of Father of British Botany. Herbalism also saw the emergence of the doctrine of signatures, a result of the urge to search for clues from plants. Many me-dicinal plants, the doctrine held, are stamped with a clear indication of their medicinal use. This was based upon the belief that plants and plant parts cured that particular portion of the hum an body, which they resembled. Thus, herbs with yellow sap would cure jaundice, the walnut kernel Joachin Jung (1587-1657) The first terminologist A brilliant teacher in Germany, Jung sucwould comfort the brain and maidenhair fern would prevent baldness. Paracelsus and ceeded in defining several terms such as Robert Turner were the main proponents of nodes, internodes, simple and compound this doctrine, later ridiculed when more leaves, stamens, styles, capitulum com posed knowledge concerning medicinal plants was of ray and disc florets. Though he left no pubacquired from the seventeenth century lications of his own, two of his pupils pr eonwards. served records of his teaching. 302 Plant Systematics Gaspard (Caspar) Bauhin (1560- 1624) Legislateur en botanique A Swiss botanist, Bauhin travelled exten-sively and formed a herbarium of 4000 s peci-mens. He published Phytopinax (1596), Prodromus theatri botanici (1620) and , lastly, Pinax theatri botanici (1623), containing a list of 6000 species of pl ants giving synonyms (other names used for a species by earlier authors) and int roducing the binomial no-menclature for several species which he named. He sough t to clarify in a single pub-lication the confusion regarding multiplic-ity of n ames for all species known at that time. Although he did not describe genera, he did recognize the difference between spe-cies and genera and several species we re included under the same generic name. His elder brother Jean Bauhin (1541-161 3) had earlier compiled a description of 5000 plants with more than 3500 figures , a work pub-lished under the name Historia plantarum universalis in 1650-51, se veral years after his death. It is tragic that the two brothers never collaborat ed and rather worked on identical lines independently. John Ray was the first to group together plants that resembled one another and s eparated those that differed more. His classification was a great advancement in plant sciences. It was evidently ahead of his time, groping at what later devel oped as natural systems, which were perfected by de Jussieu, de Candolle and Ben tham and Hooker. J. P. de Tournefort (1656-1708) Father of genus concept John Ray (1627-1705) A French botanist, de Tournefort studied under Pierre Magnol in the University o f Montpellier and later became the professor of botany at Jardin du Roy in Paris and later, Director of Jardin des Plantes in Paris. He published Elements de bo tanique in 1694, including 698 genera and 10,146 species. A Latin translation of this work with additions was published as Institutions rei herbariae in 1700. T ournefort travelled extensively in Greece and Asia Minor and brought back 1356 p lants, which were fitted into his system by his admirers. He was perhaps the fir st to give names and description of genera, merely listing the species. Casper B auhin, who did recognize genera and spe- cies, provided no such description. Tou rnefort was, thus, the first to establish genera. His system of classification, though inferior to that of Ray, was useful for identification, recognizing petal iferous and apetalous flowers, free and fused petals, and regular and irregular flowers. No doubt the system became very popular in Europe during the eighteenth

century. Ray was an English botanist who travelled extensively in Europe and published nu mer-ous works, the most significant being Methodus plantarum nova (1682) and His toria plantarum (1686-1704), a three-volume work. The last edition of Methodus, published in 1703, included 18000 species. Ray divided the plant kingdom as show n in the outline of his classification presented in Table 10.1. SEXUAL SYSTEM Table 10.1 Outline of classification of plants published by John Ray in Historia A turning point in the classification applantarum (1686-1704). proach was establishing the fact of sexuality in flowering plants by Camerarius in 1694. He concluded that stamens were male sex organs and pollen was necessary for seed set. He showed that the style and ovary form female sex organs of a fl ower. The thought regarding sexuality in plants, ridiculed by the church hithert o, once established saw renewal in botanical interest, amply 1. Herbae (Herbs) A. Imperfectae (Cryptogams) B. Perfectae (Seed plants) i. Monocotyledons ii. Dicotyledons 2. Arborae (Trees) A. Monocotyledons B. Dicotyledons Systems of Classification 303 exploited by Linnaeus for classifying flowering plants. sor Rudbeck. Under the able guidance of Prof. Rudbeck, Linnaeus published his fi rst paper on the sexuality of plants in 1729. Following Carolus Linnaeus Father of favourable publicity of his paper, he was aptaxonomy pointed as Demonstrator and subsequently promoted as Docent. In 1730, he publish ed Hortus upplandicus, enumerating the plants in the Uppsala Botanical Garden ac cording to the Tournefort s system. Faced with problem of increasing numbers of pl ants which he found hard to fit in Tournefort s system, he published a revised edi tion of Hortus upplandicus with plants classified according Carolus Linnaeus (1707-1778), was also known as Carl Linnaeus, Carl Linne, or Ca rl Von Linne. Whereas Darwin dominated bo-tanical thinking during the nineteenth century, Linnaeus did so during the eighteenth. Carl Linne, Latinized as Carl Linna eus or Carolus Linnaeus (Figure 10.3), born in Rashult, Sweden on 23 May 1707, h ad botany attached to him at birth, since Linnaeus is the Latin for Linn or Lind en tree (Tilia spp.). His father, a country Parson, wanted his son to become a p riest, but Linnaeus chose to enter the University of Lund in 1727 to learn medic ine. Although he had no money to buy books, his dedication impressed Professor K ilian Stobaeus, who not only allowed him full use of his library but also gave h im free boarding at his house. Lund not being a suit-able place for Medicine, Li nnaeus shifted to the University of Uppsala in 1729. In recog-nition of his enth usiasm for plants, Dean Olaf Celsius introduced him to botanist Profesto his own sexual system. ~ Linnaeus was sent on an expedition to Lapland in 1732, a trip that widened his k nowledge. He brought back 537 specimens. The results of the expedition were late r pub-lished as Flora Lapponica (1737). Linnaeus went to the Netherlands in 1735 and obtained an M. D. degree from the University of Haderwijk. While in the Net herlands, he met several prominent naturalists including John Frederick Gronoviu s and Hermann Boerhaave, the former financing the publication of Systema naturae (1735), presenting an outline of the sexual system of Linnaeus. He became the p ersonal physician of a wealthy person named George Clifford who was the Director of the Dutch East India Company, and this gave Linnaeus an opportunity to study numerous tropical and temperate plants grown by Clifford in his garden. It was

at Clifford s expense that Linnaeus published several manuscripts, including Hortu s cliffortianus and Genera plantarum (1737). Linnaeus then went to England, wher e he met Professor John Jacob Dillen, who initially thought of Linnaeus as this i s he who is bringing all botany into confusion , but he soon became the advocate o f the Linnaean system in England. He also met the de Jussieu brothers in France. Following the death of Professor Rudbeck, Linnaeus was appointed Professor of me dicine and botany at the University of Uppsala, a position he held until his dea th in 1778. He published his best known Species plantarum in 1753. His growing f ame and publications attracted large number of Figure 10.3 Carolus Linnaeus (1707-1778), the Father of taxonomy (reproduced with permission from Royal Botanic Gardens Kew). 304 Plant Systematics DIADELPHIA HEXANDRIA. 699 Classis XVII DIADELPHIA. H E X A N D R I A, FUMARIA. *Corollis bicalcaratis, I. FUMARIA Scapo nudo. Hort. Cliff. 251: * Gron. cucullaria, virg. 171. Rov.lugdb,. 393 Fumaria tuberosa insipida. Corn. canad. 127. Fumaria siliquosa, radice grumosa, flore bicorporeo ad labia conJucto, virginiana. Plak, alam. I62. t. 90. f 3 Raj. suppl. 475 Cucullaria. Juss. act. paris . 1743 Habitat in Virginia, Canada 2: Radix tuberosa;Folium radicale tricompositum. Scapus nudus, Racemo simplici; bracteae vix ullae; Nectarium duplex corollam basi bicor nem efficiens. 2. FUMARIA floribus postice bilobis, caule folioso. spectabilis. Habitat in Sibiria. D. Demidoff: Planta eximia floribus speciofissimis, maximis. Habitus Fumariae bulbosae, sed majora omnia. Rami ex alis rarioris. Caulis erectus. Race mus absque bracteis. Corollae magnitudine extimi articuli pollicis, pone in duos lobos aequales, rotundatos divisae. * Corollis unicalcaratis. 3. FUMARIA caule simplici, bracteis longitudine florum bulbosa Figure 10.4 A portion of a page from Species plantarum of Linnaeus (1753). Speci fic epithet (trivial name) is indicated towards the margin students, their number increasing every year and the botanical garden at Uppsala enAmong his enthusiastic students were Pe-ter Kalm and Peter Thunberg. Kalm collec ted plants extensively in Finland, Russia and America and when he returned with bundles of collection from America, Linnaeus was bedridden, but forgot his ailme nt and trans-ferred his concern to plants. Thunberg colriched considerably. Linnaeus botanical excursions every summer also included an annotator to take notes, a Fiscal to maintain discipline and marksmen to shoot birds. At the end of each trip, they marched back to the town with Linnaeus at the head, aided with French lected extensively in Japan and South Africa. horns, kettledrums and banners. Linnaeus first outlined his system in Systema naturae, which classified all know

n plants, animals and minerals. In his Gen-era plantarum, he listed and describe d all the plant genera known to him. In Species plantarum, he listed and describ ed all the known species of plants. For each species there was (Figure 10.4): In recognition of his contributions, Linnaeus was made Knight of the Polar Star in 1753, the first Swedish scientist to get this honour. In 1761, he was granted the patent of nobility and from this date came to be known as Carl von Linne. Systems of Classification 305 Table 10.2 Outline of the 24 classes recognized by Linnaeus in his Species plant arum (1753) on the basis of stamens. Classes 1. Monandria- stamen one 2. Diandria- stamens two 3. Triandria- stamens three 4. Tetrandria- stamens four 5. Pentandria- stamens five 6. Hexandria- stamens six 7. Heptandria- stamens seven 8. Octandria- stamens eight 9. Ennandria- stamens nine 10. Decandria- stamens ten 11. Dodecandria- stamens 11-19 12. Icosandria- stamens 20 or more, on the calyx 13. Polyandria- stamens 20 or more, on the receptacle 14. Didynamia- stamens didynamous; 2 short, 2 long 15. Tetradynamia- stamens tetradynamous; 4 long, 2 short 16. Monadelphia- stamens monadelphous; united in 1 group 17. Diadephia- stamens diadelphous; united in 2 groups 18. Polyadelphia- stamens polyadelphous; united in 3 or more groups 19. Syngenesia- stamens syngenesious; united by anthers only 20. Gynandria- stamens united with the gynoecium 21. Monoecia- plants monoecious 22. Dioecia- plants dioecious 23. Polygamia- plants polygamous 24. Cryptogamia- flowerless plants (i) a generic name; The system of Linnaeus, very simple in its application, recognized 24 classes (ii) a polynomial descriptive phrase or phrase-name commencing with generic name and of up to twelve words, intended to serve as description of the Table 10.2), mostly on the basis of stamens. species; These classes were further subdivided on the basis of carpel characteristics int o orders such as Monogynia, Digynia, etc. Such a classification based on stamens and carpels resulted in the artificial grouping of unre(iii) a trivial name or specific epithet on the margin; (iv) synonyms with reference to important lated taxa and separation of relatives. earlier literature; and Linnaeus knew that his system was more convenient than natural, but it was the n eed of the day when there was a tremendous in-crease in the number of plants kno wn to man, which necessitated quick identification and placement. This is exactl y what the sexual system of Linnaeus achieved with merit. His Species plantarum (1753) marks the starting point of botanical nomenclature (v) habitats and countries. The generic name followed by the trivial name formed the name for each species.

Linnaeus thus established the binomial nomenclature, first started by Caspar Bau hin and the generic concept, started by Tournefort. 306 Plant Systematics today. Linnaeus did aim at natural classifi-cation and in the 6th edition of his Genera plantarum (1764), he appended a list of 58 natural orders. It was, howev er, left to others to carry forward. Linnaeus had done based on the idea conceived by Adanson and now developed into Neo-Adansonian pri nciples. Jean B. P. Lamarck (1744-1829) his job according to the demands of the day. Following Linnaeus death in 1778, his son Carl received the post of Professor as well as the collections at Uppsala. When the latter died in 1783, the collection s went to the widow of Linnaeus, whose sole aim was to sell it the highest bidde r. Fortunately, this highest bidder of 1000 guineas was J.E. Smith, an English b otanist. Smith founded the Linnaean Society of London in 1788 and handed over th e herbarium to this society. Herbarium specimens have since been photographed an d are available in A French naturalist, Jean B. P. Lamarck authored Flore Francaise (1778), which i n addition to a key for identification of plants, contained principles concernin g the natural grouping of species, orders and families. He is better known for h is evolutionary theory, Lamarckism. de Jussieu family microfiche. Four well-known botanists belonged to this prominent French family. Of the three broth-ers Antoine (1686-1758), Bernard (1699 1776) and Joseph (1704-1779), the yo ungest spent many years in South America, where afterlosing his collections of f ive years, he became insane. The elder two studied at the University of Montpell ier under Pierre Magnol. Antoine succeeded Tournefort as Director de Jardin des Plantes, Paris and later added Bernard to the staff. Bernard started arranging p lants in the garden at La Trianon, Versailles, according to the classification t hat was initially similar to Fragmenta methodi naturalis of Linnaeus with some s imilarities to Ray s Methodus plantarum, introducing changes, so that when finally set, it had no resemblance with the Linnaean system. Bernard based his classifi cation on the number of cotyledons, presence or absence of petals and their fu-s ion. He never published his system and it was left to his nephew Antoine Laurent de Jussieu (1748-1836; Figure 10.5) to publish this classification, along with his own The Linnaean classification remained dominant for a long time. The 5th edition o f Species plantarum appeared as late as in 1797-1805, greatly enlarged and edite d by C.L. Wildenow in four large volumes. NATURAL SYSTEMS Linnaeus had provided a readily referable cataloguing scheme for a large number of plants, but it soon became evident that unrelated plants came together in suc h groupings. A need was realized for a more objective classification. France, wh ich was undergoing an intellectual ferment and where the Linnaean system never b ecame popular, took the lead in developing natural systems of classification. Michel Adanson (1727-1806) A French botanist, unimpressed with artifi-cial choice of characters, Michel Ada nson devised a classification of both animals and plants, on the equal use of as many features as possible. In his two-volume work Familles des plantes (1763), he recognized 58 natu-ral orders according to their natural affini-ties. Present -day Numerical taxonomy is changes in Genera plantarum (1789). In this classification, the plants were di-vided into three groups, further divi ded on corolla characteristics and ovary position to form 15 classes and 100 ord ers (till the beginning of present century class and order were mostly used as names of categori es now understood as order and family, respectively).

Systems of Classification 307 the most important one being Theorie elementaire de la botanique (1813), wherein he proposed a new classification scheme, outlined the important principles and intro-duced the term taxonomy. Table 10.3 Outline of classification proposed by A. P. de Candolle in his Theori e elementaire de la botanique (1813). }~1 r I. Vasculares (vascular bundles present) Class 1. Exogenae (dicots) A. Diplochlamydeae Thalamiflorae Calyciflorae ~Corolliflorae B. Monochlamydeae (also Figure 10.5 Antoine Laurent de Jussieu (1748including gymnosperms) 1836) the author of Genera plantarum (1789) largely based on the work of his Unc le Bernard de Jussieu (reproduced with permis-sion from the Royal Botanic GarClass 2. Endogenae A. Phanerogamae (monocots) B. Cryptogamae II. Cellulares (no vascular bundles) Class 1. Foliaceae (Mosses, Liverworts) dens, Kew). Class 2. Aphyllae (Algae, Fungi, Lichens) An outline of the classification is presented below: From 1816, until his death Augustin de Candolle worked in Geneva and undertook a monumental work, intended to describe every known species of vascular plants un der the title Prodromus systematis naturalis regni vegetabilis, the first volume appearing in 1824. He published seven volumes him-self. His son Alphonse de Can dolle and grandson Casimir de Candolle continued the work. Alphonse published te n more volumes, the last one in 1873, resulting in revision of sev1. Acotyledones 2. Monocotyledones 3. Dicotyledones i. Apetalae ii. Monopetalae iii. Polypetalae iv. Diclines irregulares Acotyledones, in addition to cryptogams, contained some hydrophytes whose reproduction was not known then. Diclines irregulares contained Amentiferae, Nettles, Euphorbias as also the Gymnosperms. eral families by specialists. de Candolle familyy The classification by A. P. de Candolle delimited 161 natural orders (the number was increased to 213 in the last revision of Theorie elementaire , edited by Alphonse in

1844), grouped primarily on the basis of the presence or absence of vascular str uctures The de Candolles were a Swiss family of bota-nists. Augustin Pyramus de Candolle (1778- 1841) was born in Geneva, Switzerland but took his education in Paris, w here he be-came the Professor of Botany at Montpellier Table 10.3). (Figure 10.6). He published several books, Ferns were provided a place co-ordinate with monocots and in contrary to de Juss ieu, Gymnosperms were given a place, although among dicots. The importance of an atomical 308 Plant Systematics botanist (Figure 10.7). He was extremely accomplished and wrote many important m onographs on families such as Labiatae, Ericaceae, Scrophulariaceae and Polygonaceae. He also published Handbook of British Flora (1858) and Flora Austr aliensis in 7 volumes (1863-78). Sir J. D. Hooker (1817-1911), who succeeded his father William Hooker as Director, Royal Botanic Gardens in Kew, England was a very well known botanist, having explored many parts of the world (Figure 10.8). He published Flora of British India in 7 volumes (1872-97), Student s Flora of th e British Islands (1870) and also revised later editions of Handbook of British Flora, which remained a major British Flora until 1952. He also supervised the p ublication of Index Kewensis (2 volumes, 1893), list-ing the names of all known species and their synonyms. Figure 10.6 Augustin Pyramus de Candolle (1778-1841) who first introduced the term taxonomy in his Theorie elementaire de la botanique (1813) (reproduced wi th permission from the Royal Botanic Gardens, Kew). features was highlighted and successfully employed in the classification. Robert Brown (1773-1858) Robert Brown was an English botanist, who did not propose a classification of hi s own but demonstrated that Gymnosperms were a group discrete from dicotyledons and had naked ovules. He also clarified the floral morphology and pollination of Asclepiadaceae and Orchidaceae, morphology of grass flower structure of cyathiu m in Euphorbiaceae and established several families. George Bentham & Sir J.D. Hooker The system of classification of seed plants presented by Bentham and Hooker, two English botanists, represented the most well developed natural system. The clas sification was published in a three-volume work Genera plantarum (1862-83). Geor ge Bentham (1800-1884) was a self-trained Figure 10.7 George Bentham (1800-1884), coauthor of Genera plantarum (with J. D. Hooker, 1862-1883), and the author of the 7-volume Flora Australiensis and several mono-graphs on major families (repro-d uced with permission from Royal Botanic Gardens, Kew). Systems of Classification 309 Table 10.4 Outline of the system of classification presented by Bentham and Hook er in Genera pkantarum (1862-1883). Phanerogams or seed plants Class 1. Dicotyledons (Seed with 2 cotyledons, flowers pentamerous or tetramer ous, leaves netveined) 14 series, 25 orders and 165 families Subclass 1. Polypetalae Series 1. Thalamiflorae (sepals and petals distinct, petals free) (flowers hypogynous, stamens many, disc absent) 6 orders: Ranales, Parietales, Polygalineae, Caryophyllineae, Guttiferales and Malvales 2. Disciflorae (Flowers hypogynous, disc present below the ovary)

4 orders: Geraniales, Olacales, Celastrales and Sapindales 3. Calyciflorae (flowers perigynous or epigynous) 5 orders: Rosales, Myrtales, Passiflorales, Ficoidales and Umbellales Subclass 2. Gamopetalae Series 1. Inferae (sepals and petals distinct, petals united) (ovary inferior) 3 orders: Rubiales, Asterales and Campanales 2. Heteromerae (ovary superior, stamens in one or two whorls, carpels more than 2) 3 orders: Er icales, Primulales and Ebenales 3. Bicarpellatae (ovary superior, stamens in one whorl, carpels 2) 4 orders: Gentianales, Polemoniales, Personales and Lamiales Subclass 3. Monochlamydeae (flowers apetalous; perianth lacking or if present no t differentiated into sepals and petals) Series 1. Curvembryeae (embryo coiled, ovule usually 1) (aquatic plants, ovules many) (terrestrial plan ts, ovules many) (embryo minute) 2. Multiovulatae aquaticae 3. Multiovulatae terrestres 4. Microembryeae 5. Daphnales (carpel 1, ovule 1) 6. Achlamydosporae (ovary inferior, unilocular, ovules 1-3) (flowers unisexual) (relationship uncertain) 7. Unisexuales 8. Ordines anomali Class 2. Gymnospermae (ovules naked) 3 families Class 3. Monocotyledons (flowers trimerous, venation parallel) 7 series, 34 families Series 1. Microspermae (ovary inferior, seeds minute) 2. Epigynae (ovary inferior, seeds large) 3.Coronarieae (ovary superior, carpels united, perianth coloured) (ovary superior, carpels uni ted, perianth green) (ovary superior, perianth absent) (ovary superior, carpels more than 1, free) 4. Calycinae 5. Nudiflorae 6. Apocarpae 7. Glumaceae (ovary superior, perianth reduced, flowers enclosed in glumes) 310 Plant Systematics The system divided Phanerogams or seed plants into three classes: Dicotyledons, Gymnospermae and Monocotyledons. Di-cotyledons were further subdivided into thre e subclasses: Polypetalae, Gamopetalae and Monochlamydeae based on the presence or absence of petals and their fusion. These subclasses, in turn, were subdivide d into series, orders (called cohorts by the two authors) and families (called n atural orders). No orders (cohorts) were recognized within Monochlamydeae and Mo nocotyledons, the series being directly divided into families (natural orders).

A broad outline of the classification is presented in Table 10.4. x,71

I o

Merits The fact notwithstanding that the system does not incorporate phylogeny and is m ore than 100 years old, it still enjoys a reputation of being a very sound syste m of classifiFigure 10.8 Sir Joseph Dalton Hooker (18171911), the famous British botanist who co-authored Genera Plantarum with George Bentham, besides authoring the 7- volume Flora of British India and several othe r pub-lications. He was the Director of the Royal Botanic Gardens, Kew (reproduc ed with permission from Royal Botanic Gardens, Kew). cation, owing to the following merits: 1. The system has great practical value for identification of plants. It is very easy to follow for routine identification. 2. The system is widely followed for the arrangement of specimens in the her-baria of many countries, including Britain and India. 3. The system is based on a careful comThe Genera plantarum of Bentham and Hooker provided the classification of seed p lants describing 202 families and 7569 gen-era. They estimated the seed plants t o in-clude 97,205 species. The classification was a refinement of the systems pr oposed by A. P. de Candolle and Lindley, which in turn were based on that of de Jussieu. The de-limitation of families and genera was based on natural affinitie s and was pre-Darwinian in concept. The descriptions were based on personal stud ies from specimens and not a mere compilation of known facts, an asset which mad e the classification so popular and authentic. Many important herbaria of the wo rld have specimens arranged according to this system. parative examination of actual specimens of all living genera of seed plants and is not a mere compilation of known facts. 4. Unlike de Candolle, the Gymnosperms are not placed among dicots but rather in an independent group. 5. Although the system is not a phylogenetic one, Ranales are placed in the beginning of Dicotyledons. The group Ranales (in the broader sense including families now separated under order Magn oliales) is generally regarded as primitive by most of the leading authors. 6. Dicotyledons are placed before the Monocotyledons, a position approved by all present-day authors. Systems of Classification 311 7. The description of families and genwith Caryophyllaceae. In Bentham and Hooker s system, however, Caryoera are precise. Keys to the identifi-cation are very useful. Larger genera have been divided into subgenera in phyllaceae are placed in Polypetalae, and the other two in Mono-chlamydeae. Simi

larly Podostemaceae which are placed in a separate series Multiovulatae aquatica e, better belong to the order Rosales (Cronquist, 1988). Chloran-thaceae placed by Bentham and Hooker under Micro-embryeae and Laurineae placed under Daphanales are closely allied to the order Magnoliales (Ranales s. l.) and are thus placed in the same subclass order to facilitate identification. 8. The arrangement of taxa is based on overall natural affinities decided on the basis of morphological features, which can be easily studied with the naked eye or with a hand lens. 9. Although a few important characters have been chosen to name a few groups, the grouping itself is based on a combina tion of characters, rather than any single character in the ma-jority of cases. Thus, although Del-phinium has fused petals, it has been kept in Ranunculaceae a long with the related genera and placed in Poly-petalae. Similarly, some gamopet alous genera of Cucurbitaceae are retained along with the polypetalous ones and Magnoliidae by Cronquist (1988). 4. In Monocotyledons, Liliaceae and placed in Polypetalae. Amaryllidaceae are generally regarded as closely related and often included in t he same order, some authors, including Cronquist merging Amaryllidaceae with Lil iaceae. In this system they are placed under different series, Amaryllidaceae un der Epigynae and Liliaceae 10. Heteromerae is rightly placed before Bicarpellatae. under Coronarieae. 5. Unisexuales is a loose assemblage of Demerits diverse families, which share only one major character, i.e. unisexual flow-ers. Cronquist (1988) separates these families under two distinct subclasses Hamamel idae and Rosidae and Takhtajan (1987) under HamameThe system being pre-Darwinian in approach, suffers from the following drawbacks: 1. The system does not incorporate phylogeny, although it was published after Darwin published his evolutionlididae and Dilleniidae. ary theory. 6. Bentham and Hooker did not know the 2. Gymnosperms are placed between Diaffinities of the families placed under Ordines anomali, and the families were t entatively grouped together. Cronquist (1988) and Takhtajan (1987) place Ceratop hyllaceae under sub-class Magnoliidae and the other three cotyledons and Monocotyledons, whereas their proper position is before the former, as they form a group independent from angiosperms. 3. Monochlamydeae is an unnatural assemblage of taxa, which belong else-where. The creation of this group has result ed in the separation of many closely-related families. Caryophylla-ceae, Illeceb raceae and Chenopodia-ceae are closely related families to the extent that they are placed in the same order in all major contemporary clas-sifications. Several authors including Takhtajan (1987) merge Illecebraceae under Dilleniidae. 7. Many large families, e.g. Urticaceae, Euphorbiaceae, Liliaceae and Saxifragaceae, are unnatural assemblages and repres ent polyphyletic groups. These have rightly been split by subsequent authors int o smaller, natural and monophyletic families. 8. Orchidaceae is an advanced family with inferior ovary and zygomorphic 312 Plant Systematics

flowers, but the family is placed towards the beginning of Monocotyledons. Cryptogamae and Phanerogamae, the latter further subdivided into Gymnospermae an d Angiospermae. Angiospermae was divided into two classes: Monocotyledons and Di cotyledons. Only two groups Choripetalae and Sympetalae were recognized in Dicot yledons. Gymnosperms thus found their separate identity before angiosperms, Mono chlamydeae found itself abolished and dispersed among the two groups. Monocotyle dons, strangely, found a place before Dicotyledons. 9. In Gamopetalae, Inferae with an inferior ovary is placed before the other two series having a superior ovary. The inferior ovary is now considered to have been derived from a superior ovary. 10. The system divides angiosperms into dicotyledons and monocotyledons, whereas the modern phylogenetic sys-tems place paleoherb families and Magnoliids before monocotyledons and Eudicots. Adolph Engler and Karl A Prantl PHYLOGENETIC SYSTEMS This is a system of classification of the entire plant kingdom, proposed jointly by two German botanists: Adolph Engler (1844 1930) (Figure 10.9) and Karl A. E. Prantl (1849 1893). The classification was published in a monumental work Die N atrlichen pflanzenfamilien in 23 volumes (1887-1915). Engler was Professor of Bot any at the University of Berlin and later Director, Berlin Botanic Garden. The s ystem provided classification and description down to the genus level, incorpora ting information on morpholThe publication of The Origin of species by Charles Darwin, with every copy of t he first edition sold on the first day, 24 November 1859 revolutionized biologic al thinking. The species was no longer regarded as a fixed entity having remaine d unchanged since its creation. Species were now looked upon as systems of popul ations, which are dynamic and change with time to give rise to lineages of close ly-related organisms. Once the ex-istence of this evolutionary process was ac-kn owledged, the systems of de Candolle as also of Bentham and Hooker were found to be inadequate and classifications, which made an attempt to reconstruct evoluti onogy, anatomy and geography. ary sequence, found immediate takers. The system is commonly known under Engler s name, who first published classificati on up to the family level under the title Syllabus der pflanzenfamilien in 1892. This scheme was constantly revised by Engler and continued by his followers aft er his death, the latest 12th edition appearing in 2 volumes, 1954 (ed. H. Melch ior and E. Werdermann) and 1964 (ed. M. Melchior). In this last edition, however , dicots were placed before Transitional systems The early systems were not intended to be phylogenetic. Rather, they were attemp ts to rearrange earlier natural systems in the light of the prevalent phylogenet ic theories. monocots. A.W. Eichler Engler also initiated an ambitious plan of providing taxonomic monographs of var ious families up to species level under the title Das pflanzenreich. Between 190 0 and 1953, 107 volumes were published covering 78 families of seed plants and o ne family Eichler (1839-1887) was a German botanist who proposed the rudiments of a system in 1875. This was elaborated into a unified sys-tem covering the entire plant k ingdom and finally published in the third edition of Syl-labus der vorlesungen.. .(1883). The plant kingdom was divided into two subgroups: (Sphagnaceae) of mosses. This system, often considered the begin-ning in phylogenetic schemes, was not st rictly phylogenetic in the modern sense. It was an arrangement of linear sequenc e Systems of Classification 313 starting with the simplest groups and arranged in the order of progressing com-p

lexity. In doing so, unfortunately, Engler misread angiosperms, where in many gr oups, the simplicity is a result of evoluthe beginning of dicots, also did not find much subsequent support. The system Table 10.5) became very popular, like that tionary reduction. of Bentham of Hooker, due to its comprehensive treatment and is still being foll owed in many herbaria of the world. Some recent floras including Flora Europaea (1964-1980) folThe system, however, had significant im-provements over Bentham and Hooker: Gymnosperms were placed before angiosperms, low this system. group Monochlamydeae was abolished and In this scheme of classification, the plant kingdom was divided into 13 division s (in its members distributed along with their the 11th edition of Syllabus der pflanz enpolypetalous relatives, and many large un- familien published in 1936, 14 divi sions and natural families were split into smaller in the 12th edition edited by Melchior 17 divinatural families. The placement of mono- sions were recognized) , of which the first 11 cots before dicots, another change made by dealt with Th allophytes, the 12th this system did not, however, get subsequent Embryophyta Asiphonogama (embryo su pport. The placement of the so-called group formed, no pollen tube) included Bry ophytes Amentiferae comprising families and Pteridophytes. The 13th and last div ision Betulaceae, Fagaceae, Juglandaceae, etc. in Embryophyta Siphonogama (embryo form ed, pollen tube developing) included seed plants. Merits The classification of Engler and Prantl has the following improvements over that of Bentham and Hooker: 1. This was the first major system to incorporate the ideas of organic evolution, and the first major step towards phylogenetic systems of classification. 2. The classification covers the entire plant kingdom and provides description and identification keys down to the level of family (in Syllabus der pflanzenfamilien), genus (in Die Natrlichen pflanzenf amilien) and even species for large number of families (in Das pflanzenreich). V aluable illustrations and information on anatomy and geography are also provided. 3. Gymnosperms are separated and placed before angiosperms. 4. Many large unnatural families of Figure 10.9 Adolph Engler (1844-1930), the famous German botanist who produced the most comp rehen-sive classification of the plant kingdom along with K. Prantl in a 20-volu me work Die Natrlichen pflanzenfamilien (1887-1915). Bentham and Hooker have been split into smaller and natural families. The family Urticaceae is thus split into Urticaceae, Ulmaceae and Moraceae. 5. Abolition of Monochlamydeae has resulted in bringing together several closely related families. Family 314 Plant Systematics Table 10.5 An outline of the system of classification presented by Engler and Pr antl. Plant Kingdom Division 1. }

} Thallophytes Division 11. } Division 12. Embryophyta Asiphonogama Subdivision 1. Bryophyta Subdivision 2. Pteridophyta Division 13. Embryophyta Siphonogama Subdivision 1. Gymnospermae Subdivision 2. Angiospermae Class 1. Monocotyledoneae 11 orders, 45 families Order 1. Pandanales (first family Pandanaceae) Order 11. Microspermae (last family Orchidaceae) Class 2. Dicotyledoneae 44 orders, 258 families Subclass 1. Archichlamydeae (petals absent or free) 33 orders, 201 families Order 1. Verticillatae (family Casuarinaceae only)) Order 33. Umbelliflorae (Last family Cornaceae) Subclass 2. Metachlamydeae (petals united) 11 orders, 57 families Order 34. Diapensiales (family Diapensiaceae only) Order 44. Campanulatae (Last family Compositae) Illecebraceae is merged with Caryophyllaceae. Chenopodiaceae and Caryophyllaceae are placed in the 8. Consideration of gamopetalous condition as advanced over polypetalous con-dition is in line with current phyletic same order, Centrospermae. views. 6. Compositae in dicots and Orchidaceae 9. The classification, being very thorough in monocots are advanced families with inferior ovary, zygomorphic and complex f lowers. These are rightly placed towards the end of dicots and has been widely used in textbooks, Flo-ras and herbaria around the world. 10. The terms cohort and natural order have been replaced by the appropriate monocots, respectively. terms order and family, respectively. 7. Several recent systems of classification 11. Closely related families Liliaceae and place monocots before true dicots (eudicots) Amaryllidaceae have been brought under the same order Liliiflorae. Systems of Classification 315 Table 10.6 Comparison of classification system of Bentham and Hooker with that o f Engler and Prantl. Bentham and Hooker 1. Published in Genera plantarum in 3 volumes (1862-83). 2. Includes only seed plants. 3. Gymnosperms placed in between Dicotyledons and Monocotyledons. 4. Dicotyledons placed before Monocotyledons.

5. Dicotyledons divided into 3 subclasses: Polypetalae, Gamopetalae and Monochlamydeae. 6. Subclasses are further subdivided into series, cohorts (representing orders) and natural orders (representing families). 7. Monocotyledons include 7 series and 34 natural orders. 8. Pre-Darwinian in concept. 9. Dicotyledons start with Ranales having bisexual flowers. 10. Monocotyledons start with Microspermae, including Orchidaceae. 11. Closely related families Caryophyllaceae, llecebraceae and Chenopodiaceae are kept apart, the first under Polypetalae and the other two under Monochlamydeae. 12. Closely related families Amaryllidaceae and Liliaceae placed in separate series Epigynae and Coronarieae, respectively. 13. Many larger families, e.g. Urticaceae, Saxifragaceae and Euphorbiaceae are unnatural heterogeneous groups. Engler and Prantl 1.Published in Die Natrlichen pflanzenfamilien in 23 volumes (1887-1915). 2. Includes the entire plant kingdom. 3. Gymnosperms separated and placed before the angiosperms. 4. Dicotyledons placed after Monocotyledons. 5. Dicotyledons divided into 2 subclasses: Archichlamydeae and Metachlamydeae. 6. Subclasses are further subdivided into orders and families, series not recognized. 7. Monocotyledons include 11 orders and 45 families. 8. Post-Darwinian in concept. 9. Dicotyledons start with Verticillatae with unisexual flowers. 10. Monocotyledons start with Pandanales. Microspermae are placed towards the end of Monocotyledons. 11. Family Illecebraceae is merged with Caryophyllaceae. Chepodiaceae and Caryophyllaceae are placed in the same order Centrospermae. 12. Liliaceae and Amaryllidaceae placed in the same order Liliiflorae. 13. Several larger families of Bentham and Hooker split into smaller homogeneous families. Urticaceae split into Urticaceae , Ulmaceae and Moraceae. Demerits 2. Monocotyledons are placed before With better understanding of the phylogenetic concepts in recent years, many dra w-backs of the system of Engler and Prantl have come to light. These primarily r esult from the fact that they applied the concept of simplicity representing prim itiveness even to the angiosperms, where evolutionDicotyledons. In the recent systems, paleoherbs and sometimes Magnoliids are placed before monocots. 3. The so called Amentiferae including ary reduction is a major phenomenon, not commonly seen in the lower groups. The major drawbacks of the system include: such families as Betulaceae, Juglandaceae and Fagaceae with reduced unisexual fl owers, having few floral members and borne in catkins, were considered primitive

. It has been established from studies on wood anatomy, palynology and floral an atomy that Amentiferae is an advanced group. The simplicity of flowers is due to evolutionary reduction and not 1. The system is not a phylogenetic one in the modern sense. Many ideas of Engler are now outdated. 316 Plant Systematics primitiveness. Cladistic studies of Loconte (1996) have shown that tree based on this hypothesis is six steps The above two systems of classification have been widely followed in different h erbaria around the world, as also in various regional and local Floras. Although based on basically different criteria, the two are similar in be-ing exhaustive in treatment, allowing the placement and identification of various genera with the help of valuable keys and detailed descriptions. Such a treatment is very ne cessary for distribution of specimens in a herbarium. It is also valuable in pre liminary identification of a specimen up to the generic level. Most of the conte mporary systems of classification lack treatment beyond the family level. Such s ystems of classification may be very sound in the placement of higher groups, bu t have little practical value for the longer than the shortest tree. 4. Dichlamydeous forms (distinct calyx and corolla) were considered to have evolved from the monochlamydeous forms (sin gle whorl of perianth). This view is not tenable. 5. Angiosperms were considered a polyphyletic group. Most of the recent evidence points towards monophyletic origin. 6. Araceae in Monocotyledons are now believed to have evolved from Liliaceae. In this classification, Araceae are inc luded in the order Spathiflorae which is placed before purpose of actual identification. Liliiflorae, including family Liliaceae. A comparison of the classification of Bentham and Hooker with that of Engler and Prantl is presented in Table 10.6. 7. Helobiae (including families Alismaceae, Butomaceae and Potamogetona-ceae) is a primitive group, but in this classi fication it is placed after Panda-nales, which is a relatively advanced Intentional phylogenetic systems group. 8. Derivation of free central placentation The natural systems rearranged in the light of phylogenetic information soon gav e way to systems that reflect evolutionary develop-ment. A beginning in this dir ection was made by an American botanist, Charles Bessey. from the parietal placentation, and of the latter from axile placentation is con trary to the evidence from floral anatomy. Free central placentation is now beli eved to have evolved from axile placentation through the disapCharles Bessey pearance of septa. C. A. Bessey (1845-1915) was an American botanist, who laid the foundations of m odern phylogenetic classifications (Figure 10.10). He was a student of Asa Gray and later be-came Professor of botany at the University of Nebraska. He was the first American to make a major contribution to plant classification, and also th e first botanist to develop intentional phylogenetic classification. He based hi s classification on Bentham and Hooker, modified in the light of his 28 dicta an d pub-lished in Ann. Mo. Bot. Gard. under the title The phylogenetic taxonomy of flowering 9. Ranales (in the broader senses. l.) are now considered as a primitive group with bisexual flowers, spirally arranged flo ral parts and numerous floral mem-bers. In this classification, they are placed

much lower down, after Amentiferae. 10. Family Liliaceae of Engler and Prantl is a large unnatural assemblage, which has been split into several smaller monop hyletic families like Liliaceae, Alliaceae, Asparagaceae, Asphodelaceae in the r ecent classifi-cation of Judd et al. (2002), APG II (2003) and Thorne (2006). plants (1915). Bessey considered angiosperms to have evolved monophyletically from Cycadophyta Systems of Classification 317 in angiosperms, a theory followed by many subsequent authors. Bessey believed in the strobiloid theory of the origin of the flower, the latter having originated from a vegetative shoot with spiral phyllomes, of which some modified to form sterile perianth, fertile stamens and carpels. Two evolutionary lines from such a flower formed strobiloideae (Ranalian line) with connation of like parts and cotyloideae (Rosalian line) with connation of unlike parts (Table 10.7). .~ Ranales in dicots and Alismatales in mono-cots were considered to be the most pr imitive in each group, a fact recognized by most subsequent authors. Ranalian pl ants were considered to be primitive angiosperms having given rise to monocots, but unfortunately j'4. monocots were placed before dicots. Figure 10.10 Charles Bessey (1845-1915) who Bessey also initiated the representation of evolutionary relationships through a n evolutionary diagram, a phylogram with primitive groups at the base and the mo st advanced at the tips of branches (Figure 10.11). His diagram, resembling a ca ctus plant is better known as Besseyan cactus. initiated the modern phyloge-netic systems of classification. He proposed his id eas in Ann. Mo. Bot. Gdn. (1915). belonging to implied bennettitalean ancestry. He was the pioneer to consider tha t the large-sized bisexual flowers of Magnoliaceae with spirally arranged floral parts represent the most primitive condition Hans Hallier Hallier (1868-1932) was a German botanist who developed a classification resembl ing Bessey s and starting with Ranales. Dicots Table 10.7 Outline of the classification of angiosperms proposed by Charles Bess ey (1915). Class 1. Alternifoliae (Monocotyledoneae) Subclass 1. Strobiloideae (5 orders) Subclass 2. Cotyloideae (3 orders) Class 2. Oppositifoliae (Dicotyledoneae) Subclass 1. Strobiloideae Superorder 1. Apopetalae-polycarpellatae (7 orders) Superorder 2. Sympetalae-pol ycarpellatae (3 orders) Superorder 3. Sympetalae-dicarpellatae (4 orders) Subclass 2. Cotyloideae Superorder 1. Apopetalae (7 orders) Superorder 2. Sympetalae (3 orders)

Rubiales w

ae

I'ridaln '1t* a

J Garanieles Figure 10.11 Besseyan cactus or Opuntia Besseyi showing the relationship of orde rs recognized by Bessey were, however, placed before monocots. Magnoliaceae were separated from Ranales and placed in a separate order Annonales. of his conclusions on phylogeny have been adopted in subsequent classifications. Alfred Rendle Wettstein Rendle (1865-1938), an English botanist associated with the British Museum of Na tural History, published Classification of Flowering Plants (1904, 1925), resemb ling that of Engler in considering monocots more primitive than dicots and Ament iferae a primitive group under dicots. He recognized three grades in dicots: Mon ochlamydeae, Dialapetalae (petals free) and Sympetalae. Wettstein (1862-1931) was an Austrian sys-tematist who published his classificat ion in Handbuch der systematischen botanik (1930, 1935). The classification rese mbled that of Engler in considering unisexual flowers primitive but treated mono cots advanced over dicots; and considered Helobiae to be primitive and Pandanale s advanced. Many Systems of Classification 319 In monocots Palmae were separated as a dis-tinct order and Lemnaceae considered to be advanced over Araceae. In addition to presenting his system of classification for angiosperms, Hutchins on also published valuable works such as Flora of West Tropical Africa (1927-29) , Common Wild Flowers (1945), A Botanist in South AfJohn Hutchinson John Hutchinson (1884-1972) was a British botanist associated with the Royal Bot anic Gardens, Kew, England who also served as keeper of Kew herbarium for many y ears rica (1946), Evolution and Classification of Rhododendrons (1946), British Flowe ring Plants (1948), More Common Wild Flowers (1948), Uncommon Wild Flowers (1950), British W ild (Figure 10.12). Hutchinson first proposed his Flowers (1955), Evolution and Phylogeny of Flowering Plants (1969) and Key to th e Famiclassification of angiosperms in his book The Families of Flowering Plants, the first volume on Dicotyledons appearing in 1926 and the second on Monocotyledons in 1934. The clas-sification was revised periodically, second edition in 1959 an d the 3rd in 1973, one year after his demise. lies of Flowering Plants of the World (1968). Hutchinson also embarked upon an am-bitious plan of revising Genera plantarum of Bentham and Hooker under the title The Genera of Flowering Plants. Unfortunatel y he could complete only 2 volumes of this work, published in 1964 and 1967, the

project cut short by his demise. The classification system of Hutchinson dealt only with the flowering plants, in cluded under Phylum Angiospermae as distinct from Phylum Gymnospermae. The class ification was based on 24 principles including General principles, Relating to G eneral Habit, Relating to General Structure of Flowering plants and those Relati ng to Flowers and Fruits. These principles are

outlined below: Other things being equal, it may be stated that: 1. Evolution is both upwards (sympetaly, epigyny) and downwards (apetaly, unisexuality). 2. Evolution does not necessarily involve all the organs of a plant at the same time; and one organ or set of organs may b e advancing while the other may be stationary or retrograding. 3. Evolution has generally been consistFigure 10.12 John Hutchinson (1884-1972), ent and when a particular progression or retrogression has set in, it is per-sis ted into the end of the phylum. Relating to the General Habit of the British botanist who worked as keeper of Kew Herbarium and published classif ication of an-giosperms as Families of Flower-ing Plants (1973), as also the Gen era of Flowering Plants (repro-duced with permission from Royal Botanic Gardens Kew). Plants 4. In certain groups, trees and shrubs are probably more primitive than herbs. 5. Trees and shrubs are older than climbers, the latter habit having been 320 Plant Systematics Table 10.8 Outline of the system of classification of flowering plants presented by Hutchinson in 3rd edition of The Families of Flowering Plants (1973). Phylum I. Gymnospermae Phylum II. Angiospermae Subphylum I. Dicotyledones Division I. Lignosae_ 54 orders Order 1. Magnoliales (first family Magnoliaceae) Order 54. Verbenales (last family Verbenaceae) Division II. Herbaceae_ 28 orders Order 55. Ranales (first family Paeoniaceae) Order 82. Lamiales (last family Lamiaceae) Subphylum II. Monocotyledones Division I. Calyciferae- 12 orders Order 83. Butomales (first family Butomaceae) Order 94. Zingiberales (last family Marantaceae) Division II. Corolliferae 14 orders Order 95. Liliales (first family Liliaceae) Order 108. Orchidales (family Orchidaceae only) Division III. Glumiflorae 3 orders Order 109. Juncales (first family Juncaceae) Order 111. Graminales (family Poaceae only) acquired through a particular environ-

rate in the members of the same family or genus), and the same may be said of ep iphytes, saprophytes and parament. 6. Perennials are older than biennials, and from them annuals have been derived. sites. 7. Aquatic Phanerogams are as a rule more recent than terrestrial (at any Relating to the General Structure of Flowering Plants Systems of Classification 321 8. Plants with collateral vascular bundles 24. Aggregate fruits are more recent than arranged in a cylinder (Dicotyledons) are more primitive in origin than those wi th scattered bundles (Mono-cotyle-dons), though it does not necessarily follow t hat the latter have been directly single fruits, and as a rule, the cap-sule precedes the drupe or berry. Following Bessey, Hutchinson considered derived from the former. flowering plants to be monophyletic, having evolved from the hypothetical cycade oid ancestral group which he gave the name of Proangiosperms. He recognized a nu mber of smaller groups, bound together by a combination of characters. He establ ished Magnoliales as an order distinct from Ranales, as he considered them to ha ve evolved on parallel lines. Hutchinson re-garded Magnoliaceae as the most prim itive family of the living angiosperms. He considered Dicotyledones to be more p rimitive and placed them (Table 10.8) before Monocotyledones, giving them a rank of Sub9. Spiral arrangement of leaves on the stem and of floral leaves precedes that of opposite and whorled types. 10. As a rule, simple leaves precede compound leaves. Relating to the Flowers and Fruits of Plants 11. Bisexual flowers precede unisexual flowers, and the dioecious is probably more recent than the monoecious condition. 12. Solitary flower is more primitive than phylum. the inflorescence. The groups Polypetalae, Gamopetalae and Monochlamydeae were totally abolished; i n-stead Hutchinson recognized two evolution13. Spirally imbricate floral parts are more primitive than whorled and valvate. 14. Many-parted flowers (polymerous) ary lines: division Lignosae (fundamentally woody group) and division Herbaceae (fundamentally herbaceous group) within Dicotyledones, the former starting with Magnoliaceae and ending with Verbenaceae. The Herbaceae started with Paeoniaceae and ended with Lamiaceae. Within Monocotyledones he recognized three evolutiona ry lines: division Calyciferae (calyx bearers), division Corolliferae (corolla-b earers) and division Glumiflorae (glume-bearers). A total of 411 families are re cognized, 342 in Dicotyledones and 69 in Monocotyledones. Lignosae includes 54 o rders, Herbaceae 29, Calyciferae 12, Corolliferae 14 and Glumiflorae 3. A diagra m (appropriately phylogram) showing phylogeny and evolution within Dicotyledones is presented in precede, and the type with few parts (oligomerous) follow from it, being accompa nied by progressive sterilization of reproductive parts. 15. Petaliferous flowers precede apetalous ones, the latter being the result of reduction. 16. Free petals (polypetaly) are more primitive than connate petals (sympetaly).

17. Actinomorphy of flower is an earlier type than zygomorphy. 18. Hypogyny is the primitive structure, and from it perigyny and epigyny were derived later. 19. Free carpels (apocarpy) are more primitive and from them connate carpels resulted. 20. Many carpels (polycarpy) precede few carpels (oligocarpy). Figure 10.13 21. The endospermic seed with small Whereas Hutchinson considered the woody habit to be primitive in dicots, in mono cots he considered the herbaceous habit to be primitive, and the woody forms der ived from the her-baceous forms. He also considered Monocotyledones also to be a monophyletic group derived from Ranales, Butomales having a link with Hellebora ceae and Alismatales embryo is primitive and the non-endospermic seed more recent. 22. In primitive flowers, there are many stamens; in more advanced flowers few stamens. 23. Separate stamens precede connate stamens. 322 Plant Systematics SAPINDALES APOCYNALES CASUARINALES f MELIALES VERBENAIES VIOLLALES TAMARICALI BI iUTALES LOGANIALES GNONIALES EB LES ASTERALES SANTALALES RURTICALES MYRT DALES ILA}IALEII MALPIGHIALES+EUPHORBIALES LEITNEflIALES \ CELASTRALES~ LAMIALES MYRICAL S GUI MALVALES l;ALERANALEB ES ABAE1ALfS SpL1C ERICALES S CACTALES PLANTAGINALES BORAGI ALES STYRACARE B

fAMPANALES HAMA ~PITI POLEMONIALES POOOSTEMONALES LEGUMIN0.LES ROSALES THEALE: S CHEN0p0OlAc. GENTIANALES'~ ZMBELRLES GElfAFSNIALES 0r SARRACENIALES rJ' pgSSIFLOflT RESEOALES CORl0.RIACFS DILIENIALES ,'LAEALESf PROEALES POLYGONALES AlfY0`""IFRAGAL PIPERALES~ BRASSICALES LAURALES ANNONALES BEflBEmu.c" RHOEAOALES1 ~MAGNOLIALES ARISTOLOCHIALESr RANALES~ To MONOCOTYLEDONES HERBACEAE (see part III LIGNOSAE (fundamentallywoody) (fundamentally herbaceous) OICOTYLEDONES HYPOTHETICA PROMIGIOSPERMS Figure 10.13 Hutchinson s diagram (phylogram) showing phylogeny and relationships of orders of Dicotyledones as presented in his 1973 classification. with Ranunculaceae. The presence of endosperm in seeds of Ranunculaceae and its absence from Butomaceae and Alismataceae, otherwise considered closer, is explai ned by Hutchinson to be the result of aquatic habit in the last two. A diagram ( phylogram) showing the probable phylogeny of various orders in Monocotyledones i s presented in Figure 10.14. few characters as in earlier systems is more logical. 4. Many large unnatural families have been split into smaller natural ones. Euphorbiaceae of Bentham and Hooker is spl it into Euphorbiaceae, Ricinaceae and Buxaceae. The family Urticaceae is similar ly split into Urticaceae, Moraceae, Ulmaceae and Merits The system of Hutchinson, being based on a number of sound phylogenetic principl es, and studies of a large number of plants at his dis-posal at Kew, shows the f ollowing improveCannabinaceae. 5. Standards of description are very high. Useful keys are provided for the identification of families. 6. Phylograms for dicots and monocots ments over earlier systems: are more superior to the Besseyan 1. The system is more phylogenetic than cactus. that of Engler and Prantl, as it is based on phylogenetic principles, generally

7. The classification of Monocotyledones recognized by most authors. is sounder and generally appreciated, even keys to the identification of 2. The treatment of Magnoliales as the genera have been provided. starting point in the evolutionary se-ries of Dicotyledones is in agreement 8. The derivation of Monocotyledones from Dicotyledones is widely agreed. with prevalent views. 9. The placement of Alismatales towards 3. The abolition of Polypetalae, Gamothe beginning of Monocotyledones finds petalae, Monochlamydeae, Archichla-mydeae and Metachla-mydeae and rearrangement of taxa on the combi-nation of characters and not one or a general acceptance. 10. Detailed classification up to the generic level, together with identification keys and description has been Systems of Classification 323 CYCLANNTHALES ARALES PANOANALES TYPHALES ZINGIBERALES ORCHIDALES BURMANNIALES IRIDALES PALMALES 1 NAJAOALES '-AGAVALES HAEMODORALES XYRIDALES ERIDCAULALES APONOGETONALES POTAMOCETONALES DIOSCOREALES f GRAMINALES ~-JUNLALES \/\ CYPERALES BUTOMALES ALISMATALES t HELLEBORACEAE RA\UNCLLAUAE ItANAI.I\ DICO'EYLEUOSES Figure 10.14 Hutchinson s diagram (phylogram) showing probable phylogeny and relat ionship of orders within Monocotyledones. provided for some families in the two volumes of The Genera of Flowering Plants. even separate divisions Herbaceae and Lignosae, respectively. 3. Hutchinson did not provide a full explanation for the majority of his evo-

Demerits The classification of Hutchinson has largely been ignored, as it mostly did not proceed beyond family level, and gave much impor-tance to the habit. The major d rawbacks of lutionary concepts. 4. He derives angiosperms from proangiosperms, but does not provide information about the nature of this hypothetical ancestral group. the system are listed below: 5. Although he has split several large un1. The system is not useful for practical natural families into natural units, in some cases he has even split some famili es which were already natural mono-phyletic groups. The family Ranunculaceae has been split into Ranunculaceae and Helleboraceae on the basis of achene and foll icle fruit, respectively. Studies on the floral anatomy have shown that evolutio nary stages in the reduction of ovule number can be seen in the genera of Helleb oraceae, and many genera of Ranunculaceae show traces which would have gone to n ow aborted ovules. Thus, the Ranuncula-ceae of Bentham and Hooker represents a m onophyletic group and need not be split. identification, arrangement in Floras and herbaria, as it does not proceed beyon d the family level in the greater majority of taxa. 2. The division of Dicotyledones into Lignosae and Herbaceae is most arti-ficial and has resulted in separation of clo sely related families Araliaceae and Apiaceae, in Lignosae and Herbaceae respect ively. Similarly, Lamiaceae and Verbenaceae are very closely related and often p laced in the same order in contemporary systems of classification. Hutchinson, o n the basis of habit, sepa-rated them under distinct orders and 324 Plant Systematics 6. The family Calycanthaceae is related techniques of the numerical analysis of avail-able data have helped in developin g classifications that have several features in common, though differing in some basic concepts. It is now largely agreed upon that Angiosperms are a monophylet ic group with dicots being more primitive than monocots. Vesselless Winteraceae and the paleoherb families are now generally regarded as among the basal living angiosperms. Out of the authors of the four major contemporary systems of classi fication, two A. Cronquist and R. Dahlgren have unfortunately left us during the pas t decade and a half. There is, how-ever, a positive trend of frequent updating o f classification schemes in electronic versions. During the last decade, the Ang iosperm Phylogeny Group (APG) has been working towards realization of monophylet ic groups. to Laurales, but placed here in Rosales. 7. Hutchinson regards Magnoliaceae as the most primitive family of living Dicotyledones, but most contemporary authors consider vessel-less Winteraceae, or paleoherbs be the most primitive. 8. The monocotyledons are placed after dicotyledons, whereas the recent clas-sifications place them between primi-tive angiosperms and the eudicots. 9. Family Liliaceae of Hutchinson is a large unnatural assemblage, which has been split into several smaller mono-phyle tic families like Liliaceae, Allia-ceae, Asparagaceae, Asphodelaceae in the rece nt classification of Judd et al. (2002), APG II (2003) and Thorne (2003). Lyman Benson Lyman Benson developed a classification de-signed for teaching botany and publis hed in his book Plant classification (1957). Dicotyle-dons are divided into five groups on features derived from the classifications of Bentham and Hooker and E ngler and Prantl. Monocoty-ledons are divided directly into 13 orders, starting with Alismales and ending with Pandanales. Although several realignments have be

en made by Benson, de Candolle as also Bentham and Hooker have been followed for grouping in dicots and Bessey s outline Armen Takhtajan* Armen Takhtajan (1910 2009) was a lead-ing Russian plant taxonomist (Figure 10.15) and chief of the Department of higher plants in V.L. Komarov Botanic Institute, USSR Academy of Sciences, Leningrad (now named St. Petersburg). He was an inter national authority on phytogeography, origin and phylogeny of flowering plants. He was the President of the 12th International Botanical Congress held in Leningrad in 1975. for classification of monocots: His classification was first published in 1954 in Russian, but came to be known out-side the Soviet Union only after its English translation Origin of Angiosper mous Plants was published in 1958. The system was elaborated in Die Evolution de r Angiospermen (1959), and Systema et Phylogenia Magnoliophytorum (1966), both i n Russian. The classification became popular with the English translation of the latter as Flower1. Thalamiflorae (hypogynous, free or no petals) 2. Corolliflorae (hypogynous, petals fused) 3. Calyciflorae (perigynous or epigynous, petals free or none) 4. Ovariflorae (epigynous, petals fused) 5. Amentiferae (catkin bearing) Modern phylogenetic systems ing Plants Origin and Dispersal by C. Jeffrey in 1969. The classification was publ ished in a revised form in Botanical Review in 1980. A more elaborate revision o f this classification appeared in the Russian work Sistema Magnoliophytov (Latin facsimile Systema Magnoliophytorum) in A number of contemporary workers are in-volved in improving schemes of classific ation based on new information from various sources. Recent data from paleobotan y, phytochemistry, ultrastructure and improved *Armen Takhtajan passed away on June 13, 2009, Systems of Classification 325 juvenile characters in the adult plant-also called paedomorphosis). Thus, in Mag noliales, considered most primitive among the living angiosperms, the simple ent ire, pinnately veined leaves represent the juvenile stage of frond-like leaves of the seed ferns. Takhtajan was of the opinion that angiosperms arose under environmental stress, probably as a result of adaptation to moderate seasonal drought on rocky mountai n slopes, in an area with monsoon climate. For many years, Takhtajan considered Winteraceae along with Degeneriaceae to rep resent the most primitive angiosperms. Finally, however, he chose Degeneriaceae as the most primitive family, placed under the order Magnoliales. He shifted Win teraceae to a separate order Winterales after Magnoliales (in his earlier 1987, classification he had placed Winteraceae towards the beginning in the phyletic l ine after orders Eupomatiales and Annonales). This is interFigure 10.15 Armen Takhtajan (1910-2009), leading Soviet authority on phy- esting in light of the fact that many contemtogeography and classification of porary authors regard Winteraceae or flowering plants. Published last paleoherbs (mainly Amborellaceae) among version of his classification in the most primitive groups of the living an-1997 , incorporating several modigiosperms. fications in his system. While deciding the placement of various

1987. Between 1980 and 1987, he proposed smaller revisions in 1983 (revision of di- teria based on his understanding of t he groups, Takhtajan has used a number of cricots only in Metcalfe and Chalk: Anatomy of available information. His major con cluDicotyledons, vol. 2) and 1986 (Takhtajan: Floristic Regions of the World). His final com-prehensive system of classification was published in 1997 (Diversity a nd Classifica-tion of Flowering Plants). Earlier Takhtajan, along with Cronquist and Zimmerman, had also provided a broad classification of sions are summarized below: 1. The most primitive angiosperms are regarded to be the small evergreen trees or shrubs, taller trees and deciduous h abit being later developments. 2. Simple leaves with entire margin and Embryobionta (1966). pinnate venation are primitive. Pinnately and palmately lobed leaves arose subse quently followed finally by the pinnate and palmate compound Takhtajan, who has provided a classifica-tion of angiosperms up to the family le vel, belongs to the Besseyan School and was strongly influenced by Hutchinson, H allier and the other more progressive German workers. He believes in the monophy letic origin of angiosperms, the group having evolved from seed ferns Lyginopter idophy-ta. According to Takhtajan, the angiosperms are of neotenous origin (rete ntion of leaves. 3. Primitive flower is moderate in size, in few flowered cymes, as in Degeneria. The large flowers of Magnolia and Nymphaeaceae are of secondary origin. 4. Petals have dual origin, from the bracts in Magnoliales (bracteopetals), and from the stamens in Caryophyllales 326 Plant Systematics (andropetals). Early angiosperms have numerous spirally arranged perianth of mod ified bracts. Distinct sepals and petals are secondary developments. These are further subdivided into subclasses (ending in -idae, e.g. Rosidae), su perorders (ending in -anae, e.g. Rosanae), orders and families. Cronquist, howev er, does not recognize superorders. Also, as against 11 and 6 (8 and 4 in 1987 c lassification) subclasses of dicots and monocots respectively in Takhtajan s syste m, Cronquist recognizes 6 and 5, respectively. Both systems are developed based on phylogenetic, as well as phenetic information from every field of study. Howe ver, whereas Cronquist gives more importance to phenetic information, Takhtajan 5. Primitive stamens were broad, laminar, 3-veined, not differentiated into filament and connective. The com-mon ancestral type had marginal sporangia and later on gave rise to the abaxial type s (extrorse as in Degeneria) and the adaxial types (introrse as in Magnolia). 6. Monocolpate pollen grains are primitive, from which arose tricolpate and relies more heavily on phylogenetic data. then the polycolpate types. These two systems of classification show a general agreement with the other two major classifications of angiosperms, developed by Thorne (1981, 1983, 1992,) an d Dahlgren (1981, 1983, 1989), although the recent revisions by Thorne (2000,200 3) are in more agreement with APG classifications, in abandoning traditional div ision into 7. Primitive carpels are free, unsealed, conduplicate, containing numerous ovules and with laminar placentation (as in De generia). Fusion is a later development. Fusion of closed carpels laterally resu lted in syncarpous gynoecium with axile placentation, the dissolution of septa s ubsequently resulting in lysicarpous gynoecium with free-central placentation. L

ateral fusion of open conduplicate carpels formed paracarpous gynoecium with monocots and dicots. Both Takhtajan and Cronquist prefer the name Magnoliophyta for angiosperms and a ppropriate names Magnoliopsida and Liliopsida for dicots and monocots, respectiv ely. An outline of the classification (1997 parietal placentation. 8. Outer integument arose from the version) is presented in Table 10.9. cupule of ancient gymnospermous ancestor. Unitegmic ovules arose by fusion of two integuments or abortion As against the classification proposed in 1980, the revision proposed in 1987 ha d one subclass each added to Magnoliopsida (only former superorder Asteranae now s plit into of one. 9. Takhtajan (and Cronquist) earlier Asteranae and Campanulanae retained in regarded monocotyledons as being of Asteridae, all remaining superorders placed aquatic origin from Nymphaeales via in a new subclass Lamiidae) and Liliopsida A lismatales. Later he regarded the (superorder Triuridanae separated from latter only as a lateral side branch of Liliidae into a new subclass Triurididae). Also monocotyledons, and proposed that 17 superorders, 56 orders and 96 families Nym phaeales and Alismatales had a were added to Magnoliopsida and 8 super-common or igin from hypothetical ex- orders, 17 orders and 27 families to Liliopsida. tinc t terrestrial group of Magnoliidae, In his 1997 revision he added three new subthe main monocotyledonous stock be- classes Nymphaeidae, Nelumbonidae ing terrestrial in origin. (separated from Magnoliidae) and Cornidae (separated from Rosidae) to Magnoliops ida and two Commelinidae (separated from Liliidae) and Aridae (separated from Ar ecidae) to Liliopsida. He further added 18 superorders, 47 orders and 29 familie s in Magnoliopsida and 20 orders and 27 families in Liliopsida. Takhtajan s classification approaches more closely that of Cronquist (1981, 1988) in naming angiosperms as division Magnoliophyta. Dicots and monocots are given t he rank of a class and named Magnoliopsida and Liliopsida, respectively. Systems of Classification 327 Table 10.9 Outline of the system of classification of Angiosperms proposed by Ta khtajan in 1997. Subclasses marked* did not exist in 1987 classification. Division. Magnoliophyta_ 2 classes, 17 subclasses, 71 superorders, 232 orders, 5 89 families (2 classes, 12 subclasses, 53 superorders, 166 orders, 533 families in 1987 classification); estimated genera_13,000, species_ 2,50,000 Class 1. Magnoliopsida (Dicotyledons)_ 11 subclasses, 55 superorders, 175 orders , 458 families (8 subclasses, 37 superorders, 128 orders, 429 families in 1987 classification); estimated genera_ 10,000, species_ 1,90,000 Subclass 1. Magnoliidae 2. Nymphaeidae* 3. Nelumbonidae* 4. Ranunculidae 5. Caryophyllidae 6. Hamamelididae 7. Dilleniidae 8. Rosidae 9. Cornidae* 10. Asteridae 11. Lamiidae Class 2. Liliopsida (Monocotyledons)_6 subclasses, 16 superorders, 57 orders and 131 families (4 subclasses, 16 superorders, 38 orders, 104 families in 1987

classification); estimated genera_3,000, species_ 60,000 Subclass 1. Liliidae 2. Commelinidae* 3. Arecidae 4. Alismatidae 5. Triurididae 6. Aridae* An interesting aspect about the 1987 clas-sification of Takhtajan was uncertaint y about the placement of the family Cynomoriaceae. The single genus Cynomorium e arlier placed in family Balanophoraceae (Hutchinson, 1973; Cronquist, 1988), was removed to the family Cynomoriaceae and placed next to Balanophoraceae under th e order Balanophorales by Takhtajan (1980), Thorne (1983, 1992, 2003) and Dahlgr en (1983, 1989). In his 1987 classification, Takhtajan had placed this family un der the order Cynomoriales, but not being certain about its affinities, has inse rted this order tenta-tively towards the end of Rosidae. In his 1997 classificat ion he has brought Cynomoriales under Magnoliidae within superorder Balanophoranae. It is interesting to note that Judd et al. (2002), APG II (2003) and APweb (2003) are uncertain about the placement of these two families. A major departure of Takhtajan from earlier versions is the recognition of Comme linidae as a distinct subclass in agreement with the position taken by Cronquist . Takhtajan, however, unlike Cronquist placed Liliidae at the beginning of Lilio psida, while the Alismatidae are placed higher up after Arecidae. Like other phylogenetic systems of classification, the presumed relationship of various subclasses and superorders is indicated with the help of a bubble diagra m (Figure 10.16 for dicots; Figure 10.17 for monocots) more appropriately a phylogram , the size 328 Plant Systematics CARYOPHYLLIDAE DILLENIIDAE ROSIDAE ASTERIDAE )horbianae ~aginanac Polygonanae astranae Sarracenianae r 1 LAMIIDAE Carvophyllanae Ma Fabanae Proteanae RANUNCULIDAE Lamianae Ericanae Solananae ` Violanae Vitanae

Them Myrtanae Loasanae Cornanae Dillenianae Gentianan ae Hamamlid Rosanae Eucommianae Rafflesianae Nelumbonanae / J - LILIOPSIDA HAMAMELIDIDAE Magnolianae MAGNOLIIDAE Figure 10.16 Bubble diagram of Takhtajan showing the probable relationship betwe en different subclasses and superorders of dicotyledons (based on Takhtajan, 1987). 1997 clas sification does not include a bubble diagram. of each bubble or balloon representing the relative size of each group, the bran ching pattern the phylogenetic relationship, and the length of bubble its evolut ionary advance-ment (degree of apomorphy). genetic as well as phenetic information for the delimitation of orders and famil ies. The genus Nelumbo was earlier placed in the family Nymphaeaceae under Nymph aeales. Takhtajan separated it to Nelumbonaceae under the order Nelumbonales on the basis of the occurrence of tricolpate pollen grains, embryo structure, absen ce of laticifers and chromosome morphology. He finally separated it to a separat e superorder Nelumbonanae under the distinct subclass Nelumbonidae. Thorne (1983 , 1992, 2000, 2003) also follows the separation into Nelumbonales (closer to Ran unculales) but under superorder Ranunculanae. APG II (2003) also places Nelumbon aceae Merits The latest classification of Takhtajan (1997) shows several improvements in ligh t of re-cent information on phylogeny and phenet-ics. Many merits achieved in th e earlier ver-sions have also been retained in the latest revision. The major ac hievements of this system include: 1. A general agreement with the major contemporary systems of Cronquist, Dahlgren and Thorne (earlier versions up to 1992) and incorporation of phyloSystems of Classification 329 LILIIDAE Cornmelinanae Zingiberanae Junca~nae ~ ederianae Lilianae ALISMATIDAE ARECIDAE Pandananae

Figure 10.17 Bubble diagram of Takhtajan showing the probable relationship betwe en different subclasses and superorders of monocotyledons (based on Takhtajan, 1 987). closer to Ranunculales (under Eudicots), but in order Proteales. Simi-larly, the genus Eucommia was earlier placed in the family Hamamelidaceae. Takhtajan remov ed it to the family Eucommiaceae under the order Eucommiales based on the presen ce of stipules, unilacunar nodes, unitegmic ovule and cellular endosperm, a sepa ration followed by Cronquist (1988). Thorne (1983, 1992) gave it a rank of a sub order under the order Hamamelidales (Hamame-lididae), but has now shifted it to Lamiidae under order Garryales, some-what similar to APG II (2003; Garryales of Euasterids I). de Soo (1975) placed it under a separate subclass Euco-mmiidae. S imilarly, the genus Paeonia, placed under the family Ranuncula-ceae in earlier c lassifications, was separated by Takhtajan to the family Paeoniaceae under the o rder Paeoniales on the basis of evidence from chromosomes (5 large chromo-somes) , floral anatomy (centrifugal sta-mens, many traces in sepals and pet-als, 5 in carpels), and embryology (unique embryogeny with coenocytic proembryo stage, reticulately pitted exine, l arge generative cell, thick fleshy carpels, broad stigmas, prominent lobed flesh y nectariferous disc surrounding the gynoecium). Thorne (1983, 1992, 2000, 2003) follows the separation of Paeoniaceae under the order Paeoniales. 2. The system is more phylogenetic than that of Hutchinson and other earlier authors and is based on now widely accepted phylogenetic principles. 3. The derivation of Monocotyledons from the terrestrial hypothetical ex-tinct group of Magnoliidae (often called pr oangiosperms), is largely favoured, as also the view that Alismatales and Nympha eales represent ancient side branches and have a common origin. 4. Abolition of artificial group names Polypetalae, Gamopetalae, Lignosae, Herbaceae etc. has resulted in more natural grouping of taxa. Lamiaceae and Verbenaceae are thus brought together under the order Lamiales (as against their separation under Lamiales and Verbenales and 330 Plant Systematics placement under separate groups Herbaceae and Lignosae, respectively, by Hutchin son). Caryophyllaceae, Chenopodiaceae and Portulacaceae have similarly been plac ed under the between Malvanae and Euphorbianae under Dilleniidae is more appropriate. Dahlgre n (1983) had pointed out affinities of Urticales with Malvales and Euphorbiales. The arrangement agrees with that of Thorne (2003) also. Cronquist (1988), howev er, placed Urticales in Hamamelididae, Malvales in Dilleniidae and Euphorbiales in order Caryophyllales. 5. Nomenclature is in accordance with the International Code of Botanical Nomenclature, even up to the level of division. 6. Clifford (1977) from numerical studies Rosidae. has largely supported the division of Monocotyledons into subclasses. 11. The placement of Dioncophyllaceae in 7. The placement of Magnoliidae as the a separate order Dioncophyllales is in line with the opinion presented by Metcal fe and Chalk (1983), who on the basis of anatomical evidence, proposed that the family occupied an isolated taxonomic position. Earlier, the family had been inc luded in the order Theales most primitive group of angiosperms, Dicotyledons before Monocotyledons, Magnoli ales at the beginning of Mag-noliopsida, finds general agreement with other authors. 8. Depiction of the putative relationships

next to the family Ancistrocladaceae. of major subclasses and superorders with the help of a bubble diagram is very us eful. It gives some idea about the relative size of different groups, point of c ladistic divergence and de-gree of advancement (apomorphy) reached. Larger group s are repre-sented by larger bubbles, vertical length the degree of advancement, and the point of separation of a branch its 12. Nymphaeales, whose position within the dicots, has been a matter of debate have been placed in a distinct subclass Nymphaeidae under Magnoliopsida. 13. The ending -anae, earlier opposed in favour of -florae has now been accepted by G. Dahlgren (1989) and Thorne (1992 o nwards) since the ending-florae restricts the usage to angio-sperms and is not u niversal in application. cladistic divergence. 14. The separation of Brassicaceae and 9. By splitting Asteridae into two subCapparaceae has found support from chloroplast sequence data (Hall, Sytsma and Iltis, 2002), consistent classes: Lamiidae and Asteridae, a more rational distribution of sympetalous fam ilies has been achieved. Separation of Asteridae and Lamiidae has also been foll owed by Thorne (2000, 2003) and APG II (2003; although under informal groups Eua sterids I and Euasterids II). Also Cornanae, which show affinities with the symp etalous families, have been placed in an independent subclass Cornidae (earlier placed under Rosidae). APG II also places Cornales in distinct informal group Eu asterids, although Thorne places Cornanae with morphological data. 15. The merger of Asclepiadaceae with Apocynaceae has been supported by molecular analyses by Judd et al., (1994) and Sennblad and Bremer (1998). Recognition of distinct Asclepiadaceae would render Apocynaceae as paraphyletic (Judd et al., 2002) Demerits under Asteridae. With the latest revision of his classification in 1997, Takhtajan attempted to r emove de-ficiencies in the earlier versions of his system. The critical appraisa l of his latest version, in future, may bring out some further drawbacks. The fo llowing limitations of the system can be recorded: 10. Removal of the order Urticales from Hamamelididae and its placement in an independent superorder Urticanae, Systems of Classification 331 1. The system, although very sound and and Polygonaceae under the same orhighly phylogenetic, is not helpful for identification and for adoption in her-b aria, as it provides classification only up to the family level. Also, keys to t he der Polygonales. 7. Further splitting and increase in the identification of taxa are not provided. number of families to 589 (533 in 1987) has resulted in a very narrow circumscri ption by the creation of numerous monotypic families such as Pottingeriaceae, Hydrastidaceae, 2. Dahlgren (1980, 1983) and Thorne (1983,1992, 2003) consider that the angiosperms deserve a class rank equivalent to the main groups of gym-nosperms such as Pinopsida, Barclayaceae, Nandinaceae, Griseliniaceae, Hypecoaceae, etc., and numerous oligotypic ones such as Balanoph oraceae, Sarraceniaceae, Cycadopsida etc. 3. Clifford (1977) by numerical analysis

Peganaceae and Agrophyllaceae. has shown that Arales are closer to Liliales. Dahlgren (1983, 1989) placed Arale s next to Liliiflorae (Lilianae). The recent studies have, shown the affini-ties of Araceae with Alismatales. As such, the family is included under Alismatales in APG II and under Alismatidae >Aranae >Arales by 8. Most authors regard the vesselless family Winteraceae or paleoherb Amborellaceae the most primitive among living an giosperms, but Takhtajan regarded Degeneriaceae as most primitive, considering Thorne (2003). Winteraceae as an isolated group and placing it in a separate order Winterales. The family Amborellaceae 4. Although the system is based on data finds place under Lauranae. derived from all sources, in final judg-ment more weightage is given to cladisti c information compared to 9. Takhtajan has made substantial phenetic information. changes in his scheme of classification in 1997 over his earlier version of 1987 . Unfortunately, however, he has failed to provide a bubble diagram, which was a positive feature of his ear-lier versions and was very useful in relating affin ities between the groups as also to know the relative sizes of the various group s. This is especially significant, as he has added three new subclasses under di cots and two new 5. Ehrendorfer (1983) points out that Hamamelididae do not represent an ancient side branch of Magnoliidae, but are re mnants of a transition from Magnoliidae to Dilleniidae-RosidaeAsteridae. 6. Behnke (1977) and Behnke and Barthlott (1983) point out that Caryophyllales have PIII-type plastids whereas under monocots. Polygonales and 10. Takhtajan suggested that smaller Plumbaginales have S-type plastids, and thus advocate their removal from Caryoph yllidae to Rosidae, retaining only Caryophyllales in the subclass Caryophyllidae . Though not agreeing on their removal, Takhtajan (1987, 1997) partly incorporat ed Behnke s suggestion by placing all the three or-ders under separate superorders Caryophyllanae, Polygonanae and Plumbaginanae, but within the same subclass, Ca ryophyllidae. Thorne (2003, 2006) places Plumbaginaceae families are more natural . According to Stevens (2003), this is incorrect. Monophy letic groups that include fewer taxa Takhtajan s smaller families do not necessarily have more apomorphies, even if all members of such groups are certainly likely to have more features in general in common. 11. Family Triuridaceae is removed under a separate subclass and separate superorder, but the evidence from 18S rDNA sequ encing (Chase et al., 2000) justifies its placement under 332 Plant Systematics Pandanales. Thorne (2003) shifts Triuridaceae under Pandananae, but (monocots). Cronquist includes only six subclasses in dicots and recognizes five in monocots. In dicotyledons, the Ranunculidae of Takhtajan are merged with Mag noliidae and Lamiidae are not given a separate rank distinct order Triuridales. 12. The monocotyledons are placed after dicotyledons, whereas the recent clas-sifications place them between primi-tive angiosperms and the eudicots.

at subclass level, but retained in Asteridae. 13. The Families Winteraceae and In monocotyledons, Zingiberidae are treated separate from Liliidae and Triuridal es kept under Alismatidae. As a major departure from the systems of Takhtajan, D ahlgren and Thorne, no super-orders are recognized, the subclasses are di-vided into orders directly. Also, as against 233 orders and 592 families recognized by Takhtajan, Cronquist recognizes 83 orders and 386 families. Cronquist agrees wi th Thorne (earlier versions up to 1992) in keeping the family Winteraceae (and n ot Degeneriaceae as done by Takhtajan) at the beginning of dicotyledons, and inc luded along with Degeneriaceae, Magnoliaceae, Annonaceae etc. in the same order Magnoliales. Paeoniaceae, unlike other Canellaceae, are placed in two sepa-rate orders, whereas the multigene analyses (Soltis et al., 1999; Zanis et al., 2002, 2003) have provided 99-100 per cent bo otstrap support in their relationship. The two are accordingly placed in the sam e order in APG II and APweb, and under the same suborder in Thorne (2003). The a ffinities be-tween these two families is also sup-ported by morphological studie s of Doyle and Endress (2000). Arthur Cronquist Arthur Cronquist (1919-1992), a leading American taxonomist, associated with the contemporary authors, are not separated by New York Botanical Garden (Figure 10.18), Cronquist into a distinct order Paeoni ales, produced a broad classification of but instead shifted to the order Dillen iales Embryobionta along with Takhtajan and under Dilleniidae. Zimmerman (1966). He produced a detailed classification of angiosperms in 1968 i n his book The Evolution and Classification ofF low-ering Plants. The classifica tion was further elaborated in 1981 in his book An Integrated System of Classifi cation of Flowering Plants. The final revision was published in the sec-ond edit ion (1988) of The Evolution and Clas-sification of Flowering Plants. Some realig n-ments in Dicotyledons were published in Another significant departure from Takhtajan s system is the merger of Amaryllidac eae with Liliaceae, under the order Liliales. Takhtajan places these two familie s in separate orders Amaryllidales and Liliales, respectively. Unlike most re-ce nt authors, Cronquist believed in the aquatic origin of monocotyledons, from a p rimitive vessel-less ancestor resembling present-day Nymphaeales. Nordic Journal of Botany in 1983. In contrast to Takhtajan s system, Nelumbonaceae are placed in Nymphaeales The classification is conceptually similar to that of Takhtajan s system, but differs in (and not a separate order Nelumbonal es), details. The classification, like that of Typhales in Commelinidae (and not Takhtajan, is based on evidence derived Arecidae) and sympetalous families of d icotyfrom all sources, but in contrast to Takhtajan ledons placed in a large sub class Asteridae who gives more importance to cladistics, (and not three subclass es Asteridae, Cronquist gave more importance to morphol- Cornidae and Lamiidae). Urticales are ogy (Ehrendorfer, 1983). included along with wind-pollinated families under Hamamelididae (and not with i ts 3related orders Malvales and Euphorbiales), and Malvales and Euphorbiales are kept in Following Takhtajan, the angiosperms are given the name Magnoliophyta and divided into Magnoliopsida (dicots) and Liliops ida Systems of Classification 333 evidence from all sources in arrangement of various groups. Paeonia and Nelumbo are thus placed under Paeoniaceae and Nelumbonaceae, although the orders Paeonia les and Nelumbonales are not recognized. Eucommia is also kept in a separate fam ily Eucommiaceae under a distinct order Eucommiales.

2. The revision of the classification in 1981 and 1988 was presented a in comprehensive form, giving detailed information on phytochemistry, anatomy, ultrastructure and chromoFigure 10.18 Arthur Cronquist (1919-1992) leading American Plant taxono-mist who published 2nd edition of his Evolution an d Classification of Flowering Plants in 1988. His classification is similar to t hat of Takhtajan in general outline (photograph courtesy Allen Rokach, The New Y ork Botanisomes besides morphology. 3. The text, being in English, has been readily adopted in books and floristic projects originating in the USA. 4. The system is highly phylogenetic and is based on now largely accepted phylogenetic principles. cal Garden, Bronx, New York). 5. The placement of Winteraceae at the separate subclasses Dilleniidae and Rosidae respectively (and not the same subcl ass Dilleniidae). Cronquist has provided a syn-optic arrangement of taxa, facili tating the process of identification up to the family level. An outline of Cronq uist s system is presented in Table 10.10. The system is beginning of dicotyledons is generally favoured by most authors including Ehrend orfer (1968), Gottsberger (1974) and Thorne (up to 1992). The family has vesselless wood similar to gymnosperms, great similarity between micro- and megasporop hylls, unifacial stamens and carpels, morphology similar to pteridosperms, high chromosome number suggesting long evolutionary history and less specialized beet le pollination as compared to the widely used in the USA. The relationships of various subclasses and orders (Figure 10.19) are shown with the help of a phylogram which takes the form of a bubble diagram, like other co ntemporary systems of classification. Merits genus Magnolia. 6. Abolition of artificial group names The classification of Cronquist is largely based on principles of phylogeny that find ac-ceptance with major contemporary authors. The system is merited with th e following achievements over the previous systems of such as Polypetalae, Gamopetalae, Lignosae, Herbaceae etc. has resulted in more natural grouping of taxa. Verbenaceae and Lamiaceae are thus brought under the o rder Lamiales. Caryophyllaceae, Chenopodiaceae and Portulacaceae are similarly p laced in the same order Caryophyllales. classification: 7. Nomenclature is in accordance with 1. It shows general agreement with the International Code of Botanical major contemporary systems of Takhtajan, Dahlgren and Thorne (earlier versions), and incorporates Nomenclature. 8. Placement of Magnoliidae as the most primitive group of angiosperms, 334 Plant Systematics Table 10.10 Broad outline of the classification of angiosperms presented by Cron quist (1988). Division. Magnoliophyta_ 2 classes, 11 subclasses, 83 orders and 386 families; 2 19,300 species

Class 1. Magnoliopsida (Dicotyledons)_ 6 subclasses, 64 orders, 320 families; 16 9,400 species Subclass 1. Magnoliidae (12 orders: Magnoliales, Laurales, Piperales, Aristoloch iales, Illiciales, Nymphaeales, Ranunculales and Papaverales) 2. Hamamelidae (11 orders: Trochodendrales, Hamamelidales, Daphniphyllales, Didymelales, Eucommiales, Urticales, Leitneriales, Juglandales, Myricales, Fagales and Casuarinales) 3. Caryophyllidae (3 orders: Caryophyllales, Polygonales and Plumbaginales) 4. Dilleniidae (13 orders: Dilleniales, Theales, Malvales, Lecythidales, Nepenthales, Violales, Salicales, Capparales, Batales, Ericales, Diapensiales, Ebenales and Primulales) 5. Rosidae (18 orders: Rosales, Fabales, Proteales, Podostemales, Haloragales, Myrtales, Rh izophorales, Cornales, Santalales, Rafflesiales, Celastrales, Euphorbiales, Rham nales, Linales, Polygalales, Sapindales, Geraniales and Apiales) 6. Asteridae (11 orders: Gentianales, Solanales, Lamiales, Callitrichales, Plantaginales, Scr ophulariales, Campanulales, Rubiales, Dipsacales Calycerales and Asterales) Class 2. Liliopsida (Monocotyledons)_ 5 subclasses, 19 orders, 66 families; 49,9 00 species Subclass 1. Alismatidae (4 orders: Alismatales, Hydrocharitales, Najadales, and Triuridales) (4 orders: Arecales, Cyclanthales, Pandanales and Arales) 2. Arecidae 3. Commelinidae (7 orders: Commelinales, Eriocaulales, Restionales, Juncales, Cyperales, Hydatellales and Typhales) 4. Zingiberidae (2 orders: Bromeliales and Zingiberales) 5. Liliidae (2 orders: Liliales and Orchidales) dicotyledons before monocotyledons, Magnoliales at the beginning of Magnoliidae and Butomaceae at the beginning of Liliopsida, finds general chloroplast sequence data (Hall, Sytsma and Iltis, 2002), consistent with morpho logical data. agreement with other authors. Demerits The system is becoming increasingly popular, especially in the USA, where many b ooks are following this system. The following draw9. Compositae in dicotyledons and Orchidaceae in monocotyledons are generally regarded as advanced families, and are rightly placed towards the backs, however, may be pointed out: end of each group, respectively. 1. In spite of being a highly phylogenetic 10. The relationship of various groups has and popular in the USA, the system is not very useful for identification and ado ption in herbaria since identification keys for genera, their distribution and d escription are not provided. been depicted with diagrams, which provide valuable information on rela-tive adv ancement, cladistic relation-ship and size of various subclasses. 11. The separation of Brassicaceae and 2. Dahlgren (1983,1989) and Thorne Capparaceae has found support from (1981, 2003) considered angiosperms Systems of Classification 335

5. Rosido dae 4. Dillen 2. Hamameli I _ viagnolil Magnoliopsida 3. Caryophyllidae

5. Liliidae 4. Zingiberidae 3. Commelinidae 7 2. Arecidae 1. Alismatidae:) Liliopsida Figure 10.19 Phylogram showing the relationship between various subclasses and o rders as presented by Cronquist (based on Cronquist 1988). to deserve a class rank, and not that 4. Clifford (1977) on the basis of nuof a division. merical studies has shown that Typhales are better placed in Arecidae. Cronquist places Typhales in Commelinidae. 3. Asteridae represent a loose assemblage of several diverse sympetalous families. 336 Plant Systematics 5. Superorder, as a rank above the or10. Metcalfe and Chalk (1983), on the der, is not recognized, thus showing a significant departure from the con-tempor ary systems of Takhtajan, basis of a unique combination of anatomical features, suggested that family Dion cophyllaceae should occupy an isolated taxonomic position, but it was placed by Cronquist in order Violales before family Thorne and Dahlgren. 6. Ehrendorfer (1983) pointed out that Hamamelidae do not represent an ancient side-branch of Magnoliidae but are remna nts of a transition from Magnoliidae to Dilleniidae-RosidaeAncistrocladaceae. 11. Cronquist (1988) recognized Asteridae. Physenaceae as a family under Order Urticales, but was not sure about its 7. Behnke (1977) and Behnke and exact placement. Barthlott (1983) advocate that Polygonales and Plumbaginales, with S-type plasti ds, should be re-moved to Rosidae and only Caryophyllales with PIII-type plastid s 12. The monocotyledons are placed after dicotyledons, whereas the recent classifications place them between primitive an

giosperms and the eudicots. 13. The family Winteraceae is placed toretained in Caryophyllidae. wards the beginning of Magnoliales and Canellaceae towards the end. The multigen e analyses (Soltis et al., 1999; Zanis et al., 2002, 2003) have provided 99-100 per cent bootstrap support in their relationship. The two are accord-ingly place d in a separate order in APG II and APweb, and under the same suborder in Thorne (2003). The affinities between these two families is also supported by morpholo gical studies of Doyle and Endress (2000). 8. Urticales are placed in Hamamelidae together with wind-pollinated families, whereas they are close to Malvales and E uphorbiales (Dahlgren, 1983,1989). Cronquist further sepa-rates Malvales in Dill eniidae and Euphorbiales in Rosidae. 9. Most recent authors do not believe in the aquatic ancestry of mono-cotyledons. Kosakai et al., (1970) have provided am ple evidence to refute the aquatic ancestry of monocotyledons on the basis of st udy of primary xylem in the roots of Nelumbo (Nymphaeales). Cronquist believed t hat mono-cotyledons arose from vesselless ancestors resembling presentday Nympha eales. Dahlgren et al., (1985) point out that Nymphaeales and Alismatales demons trate a case of multiple convergence, and only a few characters (sulcate pollen grains and trimerous flowers) are due to shared ancestry. The presence of two co tyle-dons, S-type sieve tube plastids, occur-rence of ellagic acid and perisperm ous seeds in Nymphaeales argue strongly against their position as a starting poi nt of monocotyledons, and none of these attributes occur in Alismatales. Rolf Dahlgren Rolf Dahlgren (1932-87), a Danish botanist working in Botanical Museum of the Un iversity of Kopenhagen first proposed his system and a new method of illustratin g phylogenetic relationships in a text book in Danish in 1974. The revised syste m in English and subsequent revisions were published in 1975, 1980, 1981, 1983. A useful detailed treatment of Monocotyledons was presented in a book The famili es of Monocotyledons (Dahlgren et al.,) in 1985. His diagram, a cross-section th rough the top of an imaginary phylogenetic tree became very popular for mapping the distribution of character-states in various orders of angiosperms, and is po pularly known as Dahlgrenogram. Systems of Classification 337 After Dahlgren s tragic death in a car accident in 1987, his wife Gertrud Dahlgren (Figure 10.20) continued his work and finally published the last Dahlgrenogram for dicotyledons, followed by a classification of monocotyledons, both in 1989, incorporating the latest ideas of Dahlgren, and b ringing up an updated classification of angiosperms. She also changed the ending s for the superorders from florae to anae, since the use of former term restricted its application to only flowering plants, and the change to anae was in the inte rest of nomenclatural uni-formity. This practice of using anae was initially star ted by Takhtajan, and has now been incorporated by Thorne (since 1992), who earl ier like Dahlgren, preferred the end-ing florae. Gertrud followed this up (1991) with the mapping of various embryological Figure 10.20 Rolf F. Dahlgren and his wife Gertrud Dahlgren who has continued his work on the classification of Angiosperms since his death in 1987. Gertrud has concentrated on evolutionary botany and sp ecies differentia-tion after 1990. (Photographs courtesy Gertrud Dahlgren.) character-states. a family in Arales, merging Sparganiaceae in Typhaceae, Thismiaceae in Burmannia ceae, and Geosiridaceae in Iridaceae, plus shifting of a few families. Gertrud a lso included the position of families in the bubble diagram. A broad outline of the classification, as presented by Gertrud Dahlgren is presented in Table 10.11 . The classification is closely similar to the earlier versions of Thorne in using the name Magnoliopsida for angiosperms, Magnoliidae for dicots, and Liliidae fo r monocots. The realignments are based on a large number of phenetic characteris

tics, mainly phytochemistry, ultrastructure and embryology. The system includes 25 super-orders in dicots and 10 in monocots. Several hundred such maps have bee n developed by Dahlgren and his associates. Dahlgren pointed out that recognitio n of Dicotyledons and Monocotyledons would not be allowed if one followed rigid cladistic approaches, but he nevertheless, considered Monocotyledons Merits The system of classification presented by Dahlgren has several advancements over the previous systems of classification. The salient advantages of the system include: 1. The system is a highly phylogenetic as a unique group worthy of subclass rank. one incorporating evidence from morphology, phytochemistry and The Dahlgrenogram (Figure 10.21) is a bubble diagram in which different orders a re represented as bubbles, whose size is rela-tive to the number of species in t he order, and their related positions reflect phyloge-netic affinities. The orde rs are combined into superorders, thus forming bubble com-plexes. While presenti ng a revision of Dahlgren s system in 1989, Gertrud made significant changes in th e superorders Theanae, Malvanae, Rutanae and Cornanae. Similarly, in monocotyled ons, the minor changes included recognizing Acoraceae as embryology. 2. The angiosperms are given a more agreeable rank of a class like Thorne and other recent systems. 3. Unlike recent phylogenetic systems, no family of angiosperms is left unplaced. 4. The Dahlgrenogram in the form of a bubble diagram is very useful in giving an idea about the relationships of super orders, orders and even families. It also gives an idea about the relative numbe r of species in each group. O yStematscsPlanti Feba Drovaia$ les pP 1 Rosales Cary0phyI1sIes I I \ \~ IPotyg0nates ales Cunoniales Elae agnates -5aaitrega1ea Myricale Jiendeie_ \ -0r03e'ales SaPindslesCeauadne$ / tmcho,)fndrel

Myrtales ~ es a Rutalea

Melanthtalee /

l Lill-6. IYgalal Linales l FiY :asstrales

' Annenab6 CMo"'thaI Win ` ag.oI1s1 PasoniaIaa Yitab6 DI sstaieY r\ Diosconalas tkeies Thulea ~J lliicies Lecythidales _i OLeesatee Cornsba Good _ CtIenta _ S tnlalelaa~ iPlttosporeee ~ Ebsnebs Ericaltss Bru.islIs

phorales \O Saliealss Arelialaa Dlabe 1'rimulalse \ MiDPUriGles ` 1 \ Tamarleslsa 1 \ Cucurbitales des ales . e$ Irate$ Capparelee 1 Figure 10.21 Two-dimensional diagram of angiosperm classification (both dicots a nd monocots included) showing orders, combined into superorders (Courtesy G. Dah lgren) Systems of Classification 339 Table 10.11 Outline of the updated Dahlgren s classification of angiosperms as pre sented by his wife Gertrud Dahlgren (1989). Dicotyledons 25 superorders, 87 orders and 343 families Sperorder Superorder 1. Magnolianae (10 orders) 14. Rutanae (9 orders) 2. Nymphaeanae (2 orders) 15. Vitanae (1 order) 3. Ranunculanae (2 orders) 16. Santalanae (1 order) 4. Caryophyllanae (1 order) 17. Balanophoranae (1 order) 5. Polygonanae (1 order) 18. Aralianae (2 orders) 6. Plumbaginanae (1 order) 19. Asteranae (2 orders) 7. Malvanae

(5 orders) 20. Solananae (2 orders) 8. Violanae (7 orders) 21. Ericanae (5 orders) 9. Theanae (4 orders) 22. Cornanae (3 orders) 10. Primulanae (2 orders) 23. Loasanae (1 order) 11. Rosanae (15 orders) 24. Gentiananae (3 orders) 12. Proteanae (2 orders) 25. Lamianae (3 orders) 13. Myrtanae (2 orders) Monocotyledons 10 superorders, 24 orders and 104 families 1. Alismatanae (2 orders) 6. Zingiberanae (1 order) 2. Triuridanae ( 1 order) 7. Commelinanae (3 orders) 3. Aranae (1 order) 8. Arecanae (2 orders) 4. Lilianae (6 orders) 9. Cyclanthanae (1 order) 5. Bromelianae (6 orders) 10. Pandananae (1 order) 5. The Dahlgrenogram has been widely 8. The merger of Scrophulariales with used for plotting and comparing the dis-tribution of various character-states in Lamiales has been followed in the recent classifications of Cronquist and angiosperms. APG. 6. The use of a superorder rank similar 9. The placement of Cornales closer to to Thorne and Takhtajan has resulted in a more realistic arrangement of families and orders. The use of ending Ericales is justified by Judd et al., (2002) and also followed up in Thorne (2003) and APG II (2003). anae is in line with other two authors. 10. The separation of Brassicaceae and 7. The separation of genus Acorus from Capparaceae has found support from chloroplast sequence data (Hall, Sytsma and I lt is, 2002), consistent with morphological data.

Araceae into a distinct family Acoraceae has been followed by recent systems of Takhtajan (1997), Thorne (2000, 2003) and APG II, who have even separated the fa mily under a distinct Demerits order Acorales. The genus is distinct Although the system of classification of from Araceae in ensiform leaves, glan- Dahlgren shows several improvements over dular tapetum, and the type of the earlier systems, it suffers from the folendothecial cells. lowing drawbacks: 340 Plant Systematics 1. The system covers only flowering 8. Dahlgren had regarded Budlejaceae plants and does not proceed below the family level, as such is not usefully for arranging specimens in a herbarium and Myoporaceae as distinct from Scrophulariaceae but morphological studies of B remer et al., (2001) and molecular (three gene analysis) by Olmstead et al., (20 01) supported their merger, which was followed by APG II (2003) and Thorne (2003 ). or for following in the Floras. 2. Dahlgren places Asteranae, Cornanae and Aralianae before Lamianae, whereas the data from molecular stud-ies justifie s placement of the group (with circumscription somewhat simi-lar to Euasterids I I) after Lamiidae Robert F Thorne Robert F. Thorne (b. 1920), an American tax-onomist, associated with the Rancho Santa Ana Botanic Garden, Claremont, California, has developed and periodically revised a system of classification. Earlier versions of the classification close ly approached the system proposed by Dahlgren in giving angiosperms a rank of a class, and dicots and monocots as subclasses. These were further subdivided into superorders, orders, suborders and families. In general approach of arrange-men t of orders and families, there was a con-siderable parallel development with ot her three contemporaries Cronquist, Dahlgren (Comparable to Euasterids I of APG II). 3. Dahlgren divides angiosperms into dicotyledons and monocotyledons, where as the recent classification of APG II (200 3) and Thorne (2000, 2003), the primitive angiosperms are placed separately. 4. Monocotyledons are placed after dicotyledons, whereas the recent classifica-tions place them between primitive angiosperms and the eudicots. 5. The family Ceratophyllaceae is placed under order Nymphaeales, but the studies of Zanis et al. (2002) and Whitlock et al., (2002) have shown that the family is a sister group of monocots as indicate d by microsporogenesis and structure of leaf margin. It is accord-ingly placed j ust before monocots in and Takhtajan. APG II. Thorne (Figure 10.22) first put forward his classification in 1968 and proposed revisions in 1974, 1976, 1981, 1983, 1992, 1999, 2000, 2003 and 2007. He earlier preferred the ending -florae over -anae of Takhtajan for superorders, but has n ow (1992 onwards) ac-cepted the ending -anae. Dr. Thorne is the recipient of the prestigious Asa Gray Award (2001) from the American Society of Plant Taxonomist s and the Merit (1996) and Centennial Award (2006) from the Botanical So6. Family Acoraceae is placed under order Arales, but according to molecular studies of Chase et al., (2000) and Fuse and Tamura (2000), it deserves placement before the rest of the monocots. 7. Family Winteraceae is placed in a sepa-

ciety of America. rate order much after Canellaceae, whereas the multigene analyses (Soltis et al. , 1999; Zanis et al., 2002, 2003) have provided 99-100 per cent bootstrap suppor t in their relation-ship. The two are accordingly placed in the same order in AP G II and APweb, and under the same suborder in Thorne (2000, 2003). The affiniti es between these two families are also supported by morphological studies of Doy le and Endress (2000). Thorne incorporated the role of phytochemistry in realignment of taxa, recognizing subfamilies more frequently and applied the principle of priority up to the class rank thus preferring name Anno nopsida for angiosperms, Annonidae for dicots, replacing Magnoliflorae by Annoni florae and Magnoliales by Annonales. Since 1992 he has, however, abandoned this departure from contemporary systems and adopted the generally accepted names Mag noliopsida, Magnoliidae and Magnoliales. Systems of Classification 341 Thorne s diagram indicating the relationship between different groups, is a phylog enetic shrub viewed from above, with the centre of the diagram left empty to ind icate extinct early angiosperms; those nearer the centre being the primitive gro ups and those nearer the periphery the advanced ones. The relative number of spe cies in different groups is indicated by balloons of different sizes (Figure 10.23). Since 1992, Thorne has been putting up electronic revisions of his classificatio n. The major revision put up in 1999 at: http:/ www.inform.umd.edu/PBIO/fam thorneangiosp99.html made a big departure from 1992 version abandoning traditional division into dicots and monocots thus con-forming to the current cladistic approach. The classification was published in 2000 in Botanical Review . Angiosperms were clas-sified into 10 subclasses where in the traditional monocot taxa were distributed in Figure 10.22 Robert Thorne of Rancho Santa Botanic Garden. His latest revi- sion of the classification of An- giosperms was published in 2007. Garden as www.rsabg.org/angiosperms/ three subclasses (Alismatidae, Liliidae angiosperms.pdf . The latest revision puband Magnoliidae and before Ranunculidae. The ten subclasses somewhat approaching the informal groups of APG included Magno-liidae, Alismatidae, Liliidae, Commelinid ae, Ranunculidae, Caryophyllidae, Dilleniidae, Rosidae, Asteridae and Lamiidae. This ar-rangement brought about a more realistic phylogenetic arrangement of dif ferent taxa, bringing the system much closer to that of Angiosperm Phylogeny Gro up. The classification recognized 31 superorders and 74 orders covering 471 families in an-giosperms. There were also 7 uncertain gen-era, 4 of them assigned tentatively to mono-typic families. The classificat ion is also very elaborate in that orders, where necessary, have been divided in to suborders, and famiCommelinidae) placed after lished in 2007 (New York Botanical Garden Press) recognizes 12 subclasses, 36 su per-orders, 85 orders and 485 families. The significant changes include: 1. The number of subclasses has been increased to 12 (as against 10 in 2000, 2003, and 11 in 2006) by adding Chlorand idae (earlier placed under Magnoliidae] at the beginning of classification and M alvidae (containing five superorders segregated from Rosidae) after Rosidae. Cer taophyllaceae placed in 2006 under Ranunculidae has been shifted under order Nym phaeales of Chloranthidae.

2. Magnoliidae now has only one superlies similarly into subfamilies. order Magnolianae (Nymphaeanae removed to Chloranthidae and Rafflesianae to Malv idae) with four or-ders (as against 1 in 2000 and four in 2003). Winteraceae (li ke in 2003) loses its position as the first family of The classification has undergone a ma-jor revision after the publication of APG II, in 2003 (10 subclasses, 33 superorders, 90 orders and 489 families), 2006 (1 1 sub-classes, 35 superorders, 89 orders and 486 families) classification displa yed on the website of the Rancho Santa Ana Botanic angiosperms. 3. Superorder Triuridanae in Alismatidae (Takhtajan 1987, 1997 took it to distinct subclass Triurididae) has been 342 Plant Systematics Table 10.12 Outline of the system of classification of Angiosperms proposed by T horne in 2007. Class Magnoliopsida 12 subclasses, 36 superorders, 85 orders, 485 families ; estimated genera- 13,37 2, species-2,53,300 Subclass 1. Chloranthidae 1 superorder, 2 orders , 9 families, 19 genera, 250 species Superorder 1. Chloranthanae 2. Magnoliidae 1 superorder, 4 orders , 20 families, 276 genera, 8805 species Superorder 1. Magnolianae 3. Alismatidae 3 superorders, 6 orders , 18 families, 235 genera, 3660 species Superorder 1. Acoranae 2. Aranae 3. Alismatanae 4. Liliidae 3 superorders, 5 orders ,51 families, 1261 genera, 29085 species Superorder 1. Pandananae 2. Dioscoreanae 3. Lilianae 5. Commelinidae 2 superorders, 10 orders , 35 families, 1116 genera, 23270 species Superorder 1. Arecanae 2. Commelinanae 6. Ranunculidae 2 superorders, 8 orders , 17 families, 298 genera, 6350 species Superorder 1. Proteanae 2. Ranunculanae 7. Hamamelididae 1 superorder, 4 orders , 22 families, 145 genera, 3870 species Superorder 1. Hamamelidanae 8. Caryophyllidae 5 superorders, 9 orders , 46 families, 889 genera, 13875 speci es Superorder 1. Berberidopsidanae 2. Caryophyllanae 3. Dillenianae 4. Santalanae 5. Balanophoranae 9. Rosidae 7 superorders, 12 orders , 83 families, 2258 genera, 48127 species 1. Celastranae

2. Violanae 3. Podostemanae 4. Oxalidanae 5. Geranianae 6. Rosanae 7. Myrtanae 10. Malvidae 5 superorders, 8 orders , 61 families, 1430 genera, 20430 species Superorder 1. Malvanae 2. Rafflesianae 3. Capparanae 4. Huerteanae 5. Rutanae 11. Asteridae 4 superorders, 13 orders , 78 families, 2677 genera, 44970 species Superorder 1. Cornanae 2. Ericanae 3. Aralianae 4. Asteranae 12. Lamiidae 2 superorders, 4 orders , 45 families, 2752 genera, 50310 species Superorder 1. Solananae 2. Lamianae Four genera (Haptanthus, Heteranthia, Pottingeria and Pteleocarpa) of uncertain position Systems of Classification 343 abolished, as also the order Triuridales. As in 2003, Family Petrosaviaceae is p laced order Petrosaviales (under superorder Acoranae). Triuridaceae is shifted t o Liliidae >Pandananae > 9. Subclass Caryophyllidae has seen major revision in 2006, recognising five superorders as against only one in 2003. 10. Malvidae has been added as new Pandanales. superorder after Rosidae in 2007, including four super Rosidae is the largest su bclass with 11 superorders Malvanae, Rafflesianae, Capparanae, Huerteanae and Ru tanae all segre4. Liliidae contains the same number of three superorders as 2003 (there were two in 2000) but Taccanae has been replace d by Dioscoreanae. The number of orders in have, however been reduced from 9 in 2003, 2006 to 5 in 2007, all 5 in Pandananae merged into gated from Rosidae. 11. Rosidae with the removal of 5 superorders from 2006 revision includes the a single order Pandanales. remaining 6 superorders. 5. Subclass Commelinidae has 2 super12. Subclass Asteridae has same four orders (same as in 2003) instead of 3 superorders as 2003 (Ericanae was added in 2003 mostly containing the members of Dillenianae which has now been restricted to include only Dilleniaceae under Ro sidae). The new version has two orders lesser. Hydrangeales has been abolished w ith its families distributed between Cornales and newly created order (in 2000), Hydatellanae being abolished. 6. Ranunculidae finds one superorder added to 2000 classification. Proteanae is shifted from Dilleniidae and broad>Hamamelidales), Buxaceae and Didymelaceae (shifted from Rosidae >Rosanae >Balanopa les) ened to include Platanaceae (shifted Rosidae > from

Rosanae and Desfontainiales. Sabiaceae (shifted from Dilleniidae >Rutanae >Rutales). These are recognized under four distinct orders Proteales, Platanales, Buxales and Sabiales. The placement broadly 13. Subclass Lamiidae has one order Garryales added (to 2000 version) in-cluding families Garryaceae, Aucubaceae, Eu commiaceae, Oncothecaceae and Icacinaceae. The arrangement is remains the same as in 2003 more or less same as in 2003 and 2006. 7. The largest subclass Dilleniidae (with 14. Genus Guametela, which was earlier 10 superorders, 19 orders and 160 families in 1999, 2000 version), which was abo lished in 2003 and its contents distributed mainly under Rosidae (see under Rosi dae). Rest of the members are distributed under Caryophyllidae listed as genus of uncertain position has been placed in 2007 revision under a d istinct family Guametelaceae under superorder Crossosomatales, Myrtanae, Rosidae ; Pottingeria has been listed as genus of uncertain position. and Asteridae 8. The new subclass Hamamelididae, Many of these changes are in line with APG II. Significant parallels include the (a) placement of Amborellaceae, Chloranthaceae and Austrobaileyaceae towards th e beginning of angiosperms; (b) recognition of Canellales, Piperales, Laurales a nd independent orders within Magnoliidae; (c) Shifting of Triuridaceae and Stemo naceae closer to Pandanaceae; (d) Bringing closer families Proteaceae, Platanace ae, established in place of Dilleniidae in 2003, includes orders Hamamelidales, Saxi fragales, Juglandales and Betulales, all shifted from Rosidae >Rosanae (of 2000). The revisions of 2006 and 2007, however, shift Vitaceae of Gunnerales to Malvid ae, Gunnera-ceae to Asteridae and Berberidopsi-dales to Caryophyllidae. 344 Plant Systematics ROSIDAE ASTERIDAE L.SV1=AE LILIDAE 4M.NY

I GFNT ANA SOLANANAE y~ PANDANANAE CORNAN.aE

TRNRIDANAE DII,LENimAE EXTINCT PROTOANGIOSPERMAE ACORANAEr. / \1

, CAPPARANAE ALISMATIDAE T iuuue E ALISAIATANAE AREOANAE

_ ~IOLANAI ` SANT.4LE_ I CELASTRANAE , MAGNOLIAVAE ;RAN[1NCULANAE RAFFLESIANAE _ CARYOPHTLLATAE BAAN HORAN~ rvuw vtuvt

COMAIM=AE CARYOPHYLLIDAE cULi AE__O__ DICOTYLEDONS 0 MONOCOTYLEDONS Figure 10.23 Thorne s Phylogenetic shrub of Angiospermae (2000 version of classifi cation). Buxaceae and Didymelaceae; (e) Shifting of Hamamelidaceae, Saxifragaceae, Vitace ae away from Rosidae to Hamamelididae; (f) groups, and focussing those which need further investigation. Salient features o f 2007 version are given in Table 10.12. placement of Ericaceae and related families under Asteridae and away Rosidae (and abol- Merits ished Dilleniidae) and (g) recognition of The classification of Thorne has kept pace Garryales as distinct order including with recent developments, and is being reg uGarryaceae, Aucubaceae and larly updated. The system is merited with the Eucommiaceae. following achievements over the previous Thorne, in his 2003 version had also in-troduced the concept of assigning the de gree of confidence in hierarchical level, circum-scription and alignment of taxa , continued in 2006 and 2007 revisions. A represents limited confidence, B for p

robably correct as-signment and C implies considerable confi-dence in assignment . Such an indication is very useful for future placements of the and contemporary systems of classification: 1. It is a highly phylogenetic system, incorporating the recent evidence from molecular systematics and chemotaxonomy, and balancing it with evidence from other sources. 2. The angiosperms are given a more agreeable rank of a class like Dahlgren and other recent systems. Systems of Classification 345 3. The system is more exhaustive than 12. Family Winteraceae and Canellaceae the contemporary systems in that families, where necessary have been divided int o subfamilies. Similarly suborders are recognized under sevare brought together under the same order. Their affinities are strongly sup-por ted by morphological studies and multigene analyses. eral orders. 13. The separation of Brassicaceae and 4. The system, unlike the APG II has Capparaceae has found support from chloroplast sequence data (Hall, Sytsma and Iltis, 2002), consistent found a place for all unplaced families of APG. 5. The placement of Amborellaceae, with morphological data. Chloranthaceae, Austrobaileyaceae, Nymphaeaceae and Cabombaceae (a major shift f rom 2006 version) at the beginning of angiosperms is generally favoured in the r ecent cladistic schemes of APG ( and supported by Qui et al, 2000; Soltis et al. , 2000). These have been placed under an independ14. The merger of Budlejaceae in Scrophulariaceae is supported by morphological studies of Bremer et al., (2001) and molecular (three gene analysis) by Olmstead et al. (2001). 15. Shifting Triuridaceae and Stemonaent subclass Chloranthidae. ceae closer to Pandanaceae is in line with recent APG schemes. The evidence from 18S rDNA sequencing (Chase et. al., 2000) justifies place-ment under Pandanales . Triurididae as an independent subclass is not jus6. Abolition of traditional groups dicots and monocots, and dividing angiosperms directly into various sub-classes (with c ircumscription largely paralleling informal groups of APG) is in line with the r ecent phylogenetic tified as indicated by recent evidence. 16. The placement of Cornales and Ericales together under Asteridae is in line with thinking. recent thinking of APG. 7. The subclass Magnoliidae placed after 17. Family Liliaceae of Hutchinson and paleoherb families is in line with APG earlier authors has been split into a number of monophyletic families such as Li liaceae, Alliaceae, Asphodelaceae, Asparagaceae, etc. in line with the arrangeme nt in APG classifications. classification. 8. The system is superior over APG II classification in that formal group names are given for all supraordinal ranks. 9. The recognition of superorders with 18. The concept of assigning the degree of ending anae has resulted in more realistic arrangement of the orders

confidence (A, B or C) in hierarchical level, circumscription and alignment of t axa is very useful for better under-standing of phylogenetic affinities. within subclasses. 10. The monocots families are arranged in between primitive angiosperms and more advanced dicots, and not towards the end of angiosperms, as in previous systems of Takhtajan, Dahlgren and Cronquist. Thi s treatment is in agree-ment with Angiosperm Phylogeny 19. The merger of Asclepiadaceae with Apocynaceae has been supported by molecular analyses by Judd et al., (1994) and Sennblad and Bremer (1998). Recognition of distinct Asclepiadaceae would render ApocynaGroup. ceae as paraphyletic (Judd et al., 2002). 11. Creation of superorder Malvanae and 20. Placement of Ceratophyllaceae under shifting several families of Rosidae here has resulted in more realistic arrange ment. Chloranthidae before monocots is in line with recent data. Studies of Zanis et a l. (2002) and Whitlock et al. (2002) 346 Plant Systematics have shown that the family is a sister group of monocots as indicated by microsp orogenesis and structure of leaf margin. monocots; 54 orders in dicots and 14 in monocots are recognized. Zheng-Yi Wu Demerits During the last decades Zheng-Yi Wu (b. 1916) and his associates have developed a system of classification of angiosperms, much different from contemporary syst ems in logic and treatment. Professor Wu , Director Emeritus, Academician of Chi nese Academy of Sciences is a leading Chinese taxonomist. He was appointed as a deputy director of the Beijing Institute of Botany, Chinese Academy of Sciences in 1950, be-came a member of the Chinese Academy in 1955, and the director of th e Kunming Institute of Botany in from 1974 through 1983. In addition to his nume rous publications on In spite of several improvements, the following drawbacks, however, may be pointed out: 1. Although highly phylogenetic the system is not very useful for identifica-tion and adoption in herbaria since identi fication keys for genera, their distribution and description are not provided. 2. Thorne places Asteridae before Lamiidae, whereas the data from mo-lecular studies justifies placement of the gr oup (with circumscription some-what similar to Euasterids II) after Lamiidae (Co mparable to Euasterids I of APG II) 3. Thorne is not clear about the affinities of four genera of angiosperms. 4. Grewiaceae (former Tiliaceae with Tilia excluded) is recognized as an independent family, whereas recent APG class ifications (Judd. et al., APG II and APweb) place all members of Tiliaceae, Bomb acaceae and Sterculiaceae under Malvaceae. 5. Thorne separates Cabomba and Brassenia under Cabombaceae on the basis of trimerous flowers, with distinct sep als and petals, 2-3 free carpels, and fruit a follicle, whereas the cladistic an alyses support their placement under Nymphaeaceae as done by APG II, APweb and J udd et al. The separation of Cabombaceae renders Nymphaeaceae as paraphyletic. C. R. de Soo From Budapest, Hungary, C. R. de Soo proposed (1975) a classification essentiall y similar to Takhtajan s but preferring the name Angiospermophyta for angiosperms, Dicotyledonopsida for dicots and Monocotyledonopsida for monocots. Five subclas ses are included in dicots and 3 in

Figure 10.24 Zheng-Yi Wu, a leading Chinese Taxonomist who spearheaded the publication of Flora of China and published his n ew eightclass classification of angiosperms. Systems of Classification 347 many plant families and genera, Professor Wu is credited for 28 major works in t ax-onomy, vegetation, floristics, biogeography, Chinese herbals, and diversity. He led or joined several botanical expeditions, espe-cially to Xizang. He descri bed about 300 new species and proposed 11 genera. He has de-voted himself to the research of the flora of China and East Asia since 1930s. He is the Chairman of editorial board for the publica-tion of Flora of China , which describes all the d iverse species of plants in China, a large-scale scientific work projected to co ver 80 volumes and 125 issues in Chinese. An English version has been published under the co-editorship of Dr. Peter H. Raven, Di-rector of the Missouri Botanic al Garden. From a global perspective, Dr. Wu has shown deep involvement with the attempts to pro-tect natural flora, specially the human-in-duced extinction of plant species and their impact on the global environment. His ef-forts have cont ributed to the establishment of national parks and natural reserves in polytopic. By polyphyletic the authors meant that during Early Cretaceous explosio n of angiosperms, there were many monophyletic groups due to extinction of many ancient species. By polytopic, the authors did not mean that the same group coul d have occurred on different continents at the same time, they rather believed t hat the modern inter-continental disjunctive distribution patterns of angiosperm s can be explained only by using plate tectonics and vicariance biogeography. Af ter their origin from Pangaea during Late Triassic to the Early Jurassic, the pr o-angiosperms might have undergone a process of differentiation, extinction and re-differentiation of several dozen million years, and then undergone a great ex plosive radiation in the Early Cretaceous. By that time, the major groups of ang iosperms might have appeared, forming 8 major lineages at early stage of differe ntia-tion of angiosperms. These eight lineages are circumscribed as 8 classes, t hus proposing a new 8-class system of classification. The system is outlined in Table 10.13 China. The ideas for this polyphyletic-polychronic-polytypic classification dividing an -giosperms into eight classes, were pre-sented in two papers of Wu et al. (1998a , 1998b). The synopsis of classification was published in 2002, and detailed des cription of families and genera represented in China Angiosperm Phylogeny Group (APG) in 2003. First serious attempts towards developing a cladistic classification were made b y Bremer (Figure 10.25) and Wanntorp (1978, 1981), who suggested that angiosperm s should be treated as subclass Magnoliidae of class Pinatae (seed plants). They argued that Monocotyledons and Dicotyledons should not be recognized because it will make the group paraphyletic, suggesting that angiosperms should be directl y divided into a number of superorders. The proposal was not taken seriously bec ause monocots and di-cots as separate groups were recognized in all major system of classifications up to the The classification has been developed on the basis of assumption that although a n-giosperms are monophyletic in their earli-est origin, yet owing to some intrin sic fac-tors in plants themselves and different ex-trinsic factors appearing on the Earth after the Early Cretaceous explosion of an-giosperms, some groups migh t have become isolated and continued to flourish for many generations. These gro ups might have given rise to many lineages, just like the situa-tion that many b ranches and leaves may sprout from a single shoot. Thus viewed from certain cros s section of time, some lineages are monophyletic-monochronic-monotopic, whereas others are polyphyletic-polychroniclast decade of last century. There has been a considerable revival of the cladistic concepts with the utiliza tion of molecular data and development of powerful tools of data handling. Durin g the last decade, concept has developed into APG

348 Plant Systematics Table 10.13 Broad outline of 8-class polyphyletic-polychronic-polytopic classifi cation of angiosperms proposed by Zheng-Yi Wu et al. (2002). Phylum Magnoliophyta (Angiospermae) (8 classes, 40 subclasses, 202 orders, 572 f amilies) Class Subclass Class Subclass

1. Magnoliopsida (5 subclasses, 11 orders, 17families) 20. Commelinidae 1. Magnoliidae 21. Juncidae 2. Annonidae 22. Poaoidae 3. Tlliciidae 23. Arecidae 4. Ceratophyllidae 6. Ranunculopsida 5. Nymphaeidae (4 subclasses, 9orders, 17families) 2. Lauropsida 24. Nelumbonidae (3 subclasses, 4orders, 9families) 25. Ranunculidae 6. Lauridae 26. Paeoniidae 7. Calycanthidae 27. Papaveridae 8. Chloranthidae 7. Hamamelidopsida 3. Piperopsida (3 subclasses, 11 orders, 21 families) (2 subclasses, 4 orders, 8families) 28. Trochodendridae 9. Aristolochidae 29. Hamamelididae 10. Piperidae 30. Betulidae 4. Caryophyllopsida 8. Rosopsida (3 subclasses, 8 orders, 20families) (10 subclasses, 112 orders, 361 families) 11. Caryophyllidae 31. Dilleniidae 12. Polygonidae 32. Malvidae 13. Plumbaginidae 33. Ericidae 5. Liliopsida 34. Rosidae (10 subclasses, 43 orders, 119families) 35. Myrtidae 14. Alismatidae

36. 15. 37. 16. 38. 17. 39. 18. 40.

Rutidae Triurididae Geraniidae Aridae Cornidae T.iliidae Asteridae Bromelidae lamiidae

19. 2ingiberidae classification by collaborative efforts of a group of dedicated workers of Angios perm Phylogeny Group (K. Bremer, A. Backlund, B. Briggs, B. Bremer, M. W. Chase, M. H. G. Gustafsson, S. B. Judd, F. A. Kellogg, P. F. Stevens, M. Thulin and sev eral others), who published a classification of 462 fami-lies of Angiosperms in 1998. These families were grouped into 40 putative monophyletic orders under a small number of inform al monophyletic higher groups: monocots, commelinoids, eudicots, core eudicots, rosids, eurosids I, eurosids II, asterids, euasterids I and euasterids II. Under these informal groups there were also listed a number of families without assig nment to order. Systems of Classification 349 0 ing a total of 51 orders and shifting some families from informal groups where t hey were placed directly in 1998 classification, to these orders. The book, howe ver, lists only major families and as such nearly 200 families, have been left o ut. A revision presented in the 2nd edition (2002) has further improvements in l ine with thinking of APG, and is largely similar to the APG II classification, with minor differences. Figure 10.25 Kre Bremer who first proposed A recent revision of APG (APG II, 2003), and continuous upgradation on Angiosper m Phylogeny website (APweb) by P. F. Stevens a cladistic classification of an(Figure 10.25)- htt www.mobot.or giosperms, and has played a leading role in development of MOBOT/research/APweb/, have resulted in APG classification along with his considerable refinement in the APG wife Birgit ta Bremer. These two, scheme, with more and more families (and along with severa l colleagues of some orders) coming out of the list of the Angiosperm Phylogeny unplaced taxa. A broad outline of APG II clasGroup, have been working at sificat ion is presented in Table 10.14. The Phylogenetic classification. classification recognizes 45 orders of An-(Publishe d with permission from Kre Bremer) giosperms, of which 44 are placed in 11 informal groups, considered more or less monophyletic. One order is unassigned at the beginning of Angiosperms. A total of 457 famiEleven unclassified families were included in the beginning. Also in the beginni ng under Angiosperms directly were 4 orders with no supraordinal grouping into i nformal groups. At the end of the system there was an additional list of 25 fami lies of uncertain position for which no firm data existed regarding placement an ywhere within the lies are recognized. The short history of APG classification makes interesting reading. A few trends are also emerging fast. The monocots are bet-ter placed under two groups, the co mmelinids and the rest of the monocots. These two groups find their place after primitive angiosperms (and possibly the Magnoliids).

system. The 1998 edition of APG classification had 82 unplaced families, of which 12 whe re placed towards the beginning and 25 towards the end, and 45 unplaced in the i nformal ten groups. In addition, 18 families classified in four orders, placed i n the beginning did not have any taxon at supraordinal rank. The number of these unplaced families has been reduced to mere 48 in APG II (in addition to 7 unpla ced genera), and the four orders in the beginning adjusted under Magnoliids (a c ombination of non-monocot paleoherbs and Magnoloid complex of Judd et al.), reco gnizing two unplaced orders in the beginning. The list of unplaced families at b eginning has been reduced to 4 and uncertain Recent cladistic analyses are revealing the phylogeny of flowering plants in inc reas-ing detail, and there is support for the mono-phyly of many major groups ab ove the family level. With many elements of the major branch-ing sequence of phylogeny established, a revised suprafamilial classification of flow-ering plants becomes both feasible and desirable. Cladistic information strongly points to the realization that sim plistic division of angiosperms into monocots and dicots do not reflect phylogenetic history. Some modifications of APG classification were presented Judd et al. (1999) recog niz350 Plant Systematics Table 10.14 Broad outline of APG II (2003) classification of Angiosperm Phylogen y Group. Magnoliophyta Group Order Group Order Unplaced families at base: Amborellaceae [+Cabombaceae], Chloranthaceae, Nymphae aceae 1. Austrobaileyales 2. Ceratophyllales 6. Rosids 1. Crossosomatales 1. Magnoliids 2. Geraniales 1. Canellales 3. Myrtales 2. Laurales 7. Eurosids I 3. Magnoliales 1. Celastrales 4. Piperales 2. Cucurbitales 2. Monocots 3. Fabales 1. Acorales 4. Fagales 2. Alismatales 5. Malpighiales 3. Asparagales 6. Oxalidales 4. Dioscoreales

7. Rosales 5. Liliales 8. Eurosids II 6. Pandanales 1. Brassicales 3. Commelinids 2. Malvales 1. Ericales 3. Sapindales 2. Commelinales 9. Asterids 3. Poales 1. Cornales 4. Zingiberales 2. Ericales 4. Eudicots 10. Euasterids I 1. Proteales 1. Garryales 2. Ranunculales 2. Gentianales 5. Core Eudicots 3. Lamiales 1. Gunnerales 4. Solanales 2. Caryophyllales 11. Euasterids II 3. Santalales 1. Apiales 4. Saxifragales 2. Aquifoliales 3. Asterales 4. Dipsacales Taxa of uncertain position:Aneulophus, Apodanthaceae, Bdalophyton, Balanophorace ae, Centroplacus, Cynomoriaceae, Cytinaceae, Dipentodontaceae, Gumillia, Hoplest igmataceae, Leptaulus, Medusandraceae, Metteniusaceae, Mitrastemonaceae, Potting eriaceae, Rafflesiaceae, Soyauxia and Trichostephanus. In addition the unplaced families in groups include 4 in the beginning (listed above),1 in monocots (Petr osaviaceae) 1 in Commelinids (Dasypogonaceae), 5 in Eudicots (Buxaceae, Didymela ceae, Sabiaceae, Trochodendraceae and Tetracentraceae), 3 in core eudicots (Aext oxicaceae, Berberidopsidaceae, Dilleniaceae), 6 in Rosids (Aphloiaceae, Geissolo mataceae, Ixerbaceae, Picramniaceae, Strasburgeriaceae, Vitaceae ), 3 in Eurosid s I (Zygophyllaceae, Krameriaceae, Huaceae), 1 in eurosids II (Tapisciaceae), 4 in Euasterids I (Boraginaceae, Icacinaceae, Oncothecaceae, Vahliaceae) and 9 in Euasterids II (Bruniaceae, Columelliaceae, Desfontainiacea, Eremosynaceae, Escalloniaceae, Paracryphiaceae, Polyosmaceae, Sphenostemonaceae, Tribelaceae). Systems of Classification 351 AmM1~,xllnceae Np,d ChlmanlM1amae Canelllales

Piperales LAUrAIes

magnoliids MAgnoliAles Arnrales Alismatales

Asp gales Dioscoleales Liliales Nnd-Aes _ Deuyl,uyanacrac Areeales Pwdes commelinids Commelinales Zingiberales Ceratophyllales RanuneulAles s.md Proteales

r d. dend. Gunnerales Ae i13erb Op,id eeae Dill i Calyophyllales Santalales SA.ifragAles CreS5d6nnlatale6 Gernaniales

My,-tales CelaslrAles Malpighiales Ozalidales Fabales euroslds i Rosales Cuenrbitales ragales Brassicales

Makales euroslds It Sapindales Corrodes Frie.leC Garryales Germliales

euasterids L Wes Solanales Aquifoliales

ApiAles euasterids II Asterales Dips-les Figure 10.26 Interrelationships of orders and some families presented in APG II classification (2003), having bootstrap support of more than 50 per cent. 352 Plant Systematics families towards the end to just 11 (plus 7 unplaced genera). A Cladogram presen ted by APG II, depicting the relationship of orders, informal higher groups, and some families is presented in Figure 10.26. Stevens in the periodically updated APweb (version 7, June 2008; Table 10.15) ha s reduced the number of unplaced families to 7, none in the beginning and the en d. He has added four orders of gymnosperms in APweb. He has also recognized 16 a dditional orders of angiosperms (total 61 as against 45 in APG II), of which 3 a

ccommodate unplaced families of APG II in the beginning of angiosperms (Amborell ales, Nymphaeales, 1 Chloranthales), (Petrosaviales), 2 in Commelinids (Ceratophyllalesrathe-rther between commelinid s and Magnoliids), 2 in Eudicots (Buxales, Trochodendrales; plus 1 Gunnerales sh ifted from Core eudicots), 3 in Core Eudicots (Berberidopsidales, Dilleniales, V itales) , and 1 in Eurosids I (Zygophyllales). The main tree (Figure 10.27) show s relationships of orders (there are no unplaced families in the beginning). The re are also useful tree links which lead to the in monocots Figure 10.27 Peter F. Stevens of Missouri Botanic Garden, who has been upgrading his APweb classification at the Angiosper m Phylogeny website. concept of monophyly, so sacred with the trees for individual orders. Angiosperm Phylogeny Group. A comprehensive treatment of the fami-lies of angiosperms based on APG II with s ome modifications is presented by Soltis et al. (2005). The number of unplaced f amilies has been reduced to 26, and three orders (Pet-rosaviales, Dasypogonales, Berberidop-sidales) added to the APG scheme, and one dropped (Ceratophyllales) thus recognising a total of 47 orders. An outline of the classiThe true phylogenetic picture of angiosperms is still far from clear. There seem s to be considerable unanimity in removing some of the primitive herbaceous fami lies of Magnoloid complex such as Amborellaceae, Cabombaceae, Paeoniaceae, Austr obaileyaceae, Trimeniaceae, Illiciaceae, and Schizandraceae and place them towar ds the beginning of the angiosperms. This renders both monocots and eudicots as truly monophyletic groups. The position of Magnoliids (after the removal of herb aceous families) seems to be settled before monocots. Earlier versions of APweb ( up to Version 6) and first edition of Judd et al. (1999) of their book, had pl aced Magnoliids after monocot, but version 7 of APweb (2008) in the second (2002 ) and Third (2008) editions of Judd et al., like APG II, Magnoliid complex is pl aced before Monocots. The classification of Judd et al., differs from APG II fication is presented in Table 10.16. Table 10.17 presents a comparison of the treatment given to the unplaced familie s (of APG or any other system) in five recent sys-tems of classification. The nu mber of unplaced families has been drastically re-duced in recents treatments of Stevens (2008) and Judd et al. (2008). Thorne does not belong to the Angiosperm Phylogeny Group, but has kept pace with the recent mo-lecular developments, and is trying to bal-ance hierarchical classification with the Systems of Classification 353 Table 10.15 Broad outline of APweb (version 7, June 2008) classification of Flow ering plants presented on Angiosperm Phylogeny website of P. F. Stevens. Magnoliophyta Group Order Group Order 1. 6. 2. 6. 3. 1. 4. 2. Amborellales Vitales Nymphaeales Rosids Austrobaileyales Geraniales Chloranthales Myrtales

3. Magnoliids 7. Eurosids I 1. Magnoliales 1. Zygophyllales 2. Laurales 2. Celastrales 3. Canellales 3. Oxalidales 4. Piperales 4. Malpighiales 1. Monocots 5. Cucurbitales 1. Acorales 6. Fagales 2. Alismatales 7. Fabales 3. Petrosaviales 8. Rosales 4. Dioscoreales 8. Eurosids II 5. Pandanales 1. Crossosomatales 6. Liliales 2. Picramniales 7. Asparagales 3. Sapindales 2. Coninielinids 4. Huerteales 1. Ericales 5. Brassicales 2. Poales 6. Malvales 3. Commelinales 4. Zingiberales * Ceratophyllales 9. Asterids 1. Cornales 2. Ericales 4. Eudicots 10. Euasterids I 1. Ranunculales 1. Garryales 2. Sabiales 2. Gentianales 3. Proteales 3. Lamiales 4. Trochodendrales 4. Solanales 5. Buxales 11. Euasterids II 6. Gunnerales 1. Aquifoliales 5. Core Eudicots 2. Asterales 1. Berberidopsidales 3. Escalloniales 2. Dilleniales 4. Bruniales 3. Caryophyllales** 5. Apiales 4. Santalales

6. Paracryphiales 5. Saxigragales 7. Dipsacales Unplaced families: 1 in Commelinids (Dasypogonaceae), 1 in Eudicots (Medusandrac eae) and 5 in Euasterids I (Oncothecaceae, Metteniusaceae, Icacinaceae, Boragina ceae, Vahliaceae). *considered closer to monocots, **closer to Asterids. 354 Plant Systematics T hie re'are di_Cect links f ram 'all terminal taxa and all' internal noti'es to , they relevant;;page :of the characterizations'. Tre"e ic6hw 1io1ka6-tir WH1,3fnk.161tree-for each erder. FernsllS,L Gyc adales Gi'riligoal'es -50i-na1es V GneFaies- ~T -a.Ambdr,ellales Nymphaeales Aust~_obaileyalesAT ChlarantiaIun M~agnolial'_es :4r Laural'es~l Ga ii el Ia I'es Llnk'[o ;Mod.el ,Or~anisml'T'ree Pi"p'e~ales. Acor,ales ;41i'sitlaial'es ~~ Pe#rosavial'es Paird`anale_s_~ Asparagales i Llilpl'aced Porealeas'est ~ Ctimmeli nales i -AP Zingifaer;al'es 4P Itanuneulal'es Sabial'cs T i]fIOCOtS Prdteaken Trochadendrales Buxales ~.Gunner,al'es Berberidopsidales aillerii'ales Carya,phyll-al'es :p SayE'aallgsr. -iSaxfragales :vital e s Unpla_ced Cros's__osom~Ea'Ie_s_I Nl yrym l'es'' jr Unpla~ed Celastrales pxalida~BS~ rosids

~ G_ucpuirggbit'aleessi eudicars .oc a rr Fagb"a3l2C!sGi -5cQcrscud1e3.ts `rHH'uaclwrtacleaSlci 4~T Cornales~ Erjcalesl 1p aster,i Unplaced Ga ~.yales rpl'aced Taka Unpi iCEW G'enti''anales~ :T Lamiales S'olanales iWrrplaced 4 ,45t~raiB5 Unplaced AI' pia s ~jp Figure 10.28 Main tree showing relationships between ms and the informal higher clades presented by Stevens in APweb ns link to the trees of the respective orders on the ssion from P. F. Stevens) Systems of Classification 355 Table 10.16 Broad outline of APG classification used sed on rearrangements of APG II. Angiosperms Group Order Group Order 1. Basal angiosperms: Amborellaceae, Nymphaeaceae, Ceratophyllaceae, Chloranthac eae 1. Austrobaileyales 1. Magnoliids 7. Eurosids I 1. Magnoliales 1. Rosales 2. Laurales 2. Fabales 3. Piperales 3. Cucurbitales 4. Canellales 4. Fagales 2. Monocots 5. Malpighiales 1. Acorales 6. Oxalidales

various orders of angiosper (version 7, 2008). Tree ico web. (Reproduced with permi by Soltis et al. (2005), ba

2. Alismatales 7. Celastrales 3. Petrosaviales 8. Eurosids II 4. Pandanales 1. Brassicales 5. Dioscoreales 2. Sapindales 6. Liliales 3. Malvales 7. Asparagales 9. Other Rosids 3. Commelinid Monocots 1. Myrtales 1. Ericales 2. Geraniales 2. Dasypogonales 3. Crossosomatales 3. Zingiberales 10. Asterids 4. Commelinales 1. Cornales 5. Poales 2. Ericales 4. Eudicots 11. Euasterids I 1. Ranunculales 2. Proteales 1. Garryales 2. Lamiales 5. Core Eudicots 3. Gentianales 1. Gunnerales 4. Solanales 2. Berberidopsidales 12. Euasterids II 3. Saxifragales 1. Aquifoliales 4. Santalales 2. Asterales 6. Caryophyllales 3. Dipsacales 4. Apiales Unplaced families: In addition to four unplaced families of basal angiosperms li sted above there 1 in commelinid monocots (Dasypogonaceae), 3 in eudicots (Sabia ceae, Buxaceae, Trochodendraceae), 3 in rosids (Vitaceae, Picramniaceae, Huaceae ), 2 in eurosids I (Zygophyllaceae, Krameriaceae), 1 in eurosids II (Tapisciacea e), 5 in Euasterids I (Oncothecaceae, Metteniusaceae, Icacinaceae, Boraginaceae, Vahliaceae) and 7 in Euasterids II (Escalloniaceae, Eremosynaceae, Bruniaceae, Columelliaceae (incl Desfontainiacea), Polyosmaceae, Tribelaceae, Sphenostemonaceae). 356 Plant Systematics in recognizing 3 orders Amborellales, Nym-phaeales and Austrobaileyales for basal fam-ilies now known as ANITA Grade and in-cluding Chloranthaceae (although with un-certain placement), shifting Ceratophyllace-ae (although with uncertain posit ion) to the end of Magnoliid complex, recognizing order Vitales under Rosid clad e (Vitaceae unplaced in Rosids in APG II), recognizing Zygophylla-les as a disti nct order under Eurosids I (Zy-gophyllaceae unplaced in Eurosids I in APG II) an d shifting the sequence of some orders, recognising 45 orders of Angiosperms. So

me informals names have also been introduced: ANITA Grade for early angiosperms, Fabids for Eurosids I, Malvids for Eurosids II, Sym-petalae for Asterid Clade, Lamiids for Euas3. Formal group names have been given mostly only up to the level of the order, where monophyly of the group has been firmly established. 4. The traditional division of angiosperms has been abandoned and various monocot taxa placed in between primitive angiospe rms and eudicots, thus overcoming the problem of paraphyly in the earlier recogn ized two groups monocot and dicots. 5. Although no formal names have been given for groups above the rank of order, there is constant endeavour to constru ct supraordinal monophyletic clades. 6. A number of cladograms are being preterids I and Campanulids for Euasterids II. sented for general affinities between various groups of angiosperms based on mol ecular as also on information Perhaps we have broken the jinx of dicot-monocot grouping of angiosperms, positi on of Magnoliid complex, which includes some of the most primitive representativ es of an-giosperms, also seems to be settled. The po-sition of Piperales seems t o be more or less stabilized towards the end of Magnoliids, but family Ceratophy llaceae has still to find a stable position. Judd et al., place the family under order Ceratophyllales after Piperales within Magnoliid complex with uncertain p o-sition, APG II place it before monocots, to-wards the beginning of angiosperms , and APweb towards end of Commelinids. from other fields. 7. The families with several primitive features are placed towards the beginning of angiosperms. The family Amborellace ae, which is unique in angiosperms in having granular and not tectate ectexine i s placed at the start. 8. Although there are four unplaced families in APG towards the beginning of the angiosperms, these have been accommodat ed under orders in APweb. 9. The number of unplaced families in Merits This newly emerging system of classifica-tion, which has undergone dramatic modi -fication over the last five years and is fast evolving, due to concerted effort s of a group of dedicated workers has several merits in various informal groups and uncertain families towards the end have been suffici ently reduced in APG II and APweb, finding ordinal places for many unplaced families of APG (1998). 10. The merger of Budlejaceae and APG II (and APweb): Myoporaceae with Scrophulariaceae has the support of morphological studies of Br emer et al., (2001) and molecular (three gene analysis) by Olmstead 1. The system is based on the sound phylogenetic principle of constructing taxa on the basis of established monophyly. et al. (2001). 2. The system is based on a synthesis of 11. Winteraceae and Canellaceae are information from mainly morphology, anatomy, embryology, phytochemistry and more strongly on molecular stud-ies. brought together under the same or-der. Their affinities are strongly sup-ported by morphological studies and multigene analyses. Systems of Classification 357 Table 10.17 Comparison of most recent phylogenetic systems of classifications in

dicating major groups and the number of unplaced taxa.

APG (1998) Modified APG APG II (2003) APweb Thorne 2007 (Judd et al., 2008) (Stevens 2008) Unplaced at start ANITA Grade * (@ 4 families * ( 4 orders) Chloranthidae *(@ 12 families (@ 1 family + 2 orders) + 4 orders) + 3 orders) Magnoliid complex (@ 1) Magnoliids Magnoliids Magnoliidae Monocots (@ 5) Monocots Monocots (@ 1) Monocots (@ 1) Alismatidae Liliidae Commelinoids (@ 6) Commelinoid clade Commelinids (@ 1) Commelinids Commelinidae Eudicots (@ 4) Tricolpates (Eudicots) Eudicots (@ 5) Eudicots (@ 1) Ranunculidae Basal Tricolpates (@ 2) Core Eudicots (@ 6) Core Tricolpates (Core Core Eudicots (@ 3) Core Eudicots Hamamelididae Eudicots) Caryophyllidae Rosids (@ 7) Rosid clade Rosids (@ 6) Rosids Eurosids I (@ 4) Eurosids I Eurosids I (@ 3) Eurosids Rosidae Eurosids II (@ 1) Eurosids II Eurosids II (@ 1) Eurosids II Malvidae Asterids

Asterid clade Asterids Asterids Asteridae Cornanae Ericanae !Aralianae !Asteranae Euasterids I (@ 3) Euasterids I Euasterid s I (@ 3) Euasterids I (@ 5) Lamiidae Solananae Lamianae Euasterids II Euasterids II (@ 1) Euasterids II (@ 10) Euasterids II (!Aralianae, Asteranae) ** (25 families) ** not listed **(10 families, 5 genera) **(none) **(4 genera) * Unplaced at the start of Angiosperms, orders have no supraordinal grouping. ** families of uncertain position. @ Families unplaced in various groups. ! Thorne includes Aralianae and Asteranae under Asteridae, which cover orders in cluded under Euasterids II, and is placed before Lamiidae, which roughly covers the orders included under Euasterids I. 12. Liliaceae of Hutchinson and earlier karyotype. The placement has been adopted by Judd et al., (2002, 2008), authors has been split to form several monophyletic families such as Liliaceae, Alliaceae, Asparagaceae, Thorne and APweb. 14. Circumscription of Malvaceae has Asphodelaceae, etc. been broadened to also include Tiliaceae, Sterculiaceae and Bombacaceae, thus fo rming monophyletic Malvaceae, as supported by morphological and molecular eviden ce. 13. Circumscription of Agavaceae has been further strengthened to include other genera like Hosta, Camassia and Chlor ogalwm, which also have bimodal 358 Plant Systematics 15. The merger of Asclepiadaceae with 6. Angiosperms have been given the Apocynaceae has been strengthened by molecular evidence Judd et al., (1994) and Sennblad and Bremer (1998). Recognition of distinct Asclepiadaceae would render Apocyna-ceae as paraphyletic (Judd et al., 2002, 2008). rank of a division, but there are no formal taxa between the rank of an order an d division, a rather unusual phenomena for classification systems. 7. Family Capparaceae has been merged Demerits Although the system is still evolving and con-tinuously improving, and will take a consid-erable time before it stabilizes and is tested by various parameters, a few shortcomings with Brassicaceae, but the Chloroplast sequence data points to the separation of these two families as also Cleomacae. Thorne (2006) recognizes Brassicaceae, Ca

pparaceae and Cleomaceae as distinct families. The developments of the last few years are obvious: have seen clear emergence of a few facts. The angiosperms are no longer to be di vided into traditional dicots and monocots. Commelinids are distinct from other mono-cots, and these two, forming the traditional monocots are better placed bet ween primitive angiosperms and the Eudicots. Primitive angiosperms include paleo herbs (Nymphaeaceae, Cabombaceae, Piperaceae, Amborellaceae, Ceratophyllaceae, e tc.) and true magnoliids (Magnoliales, Laurales, etc.) are better placed before monocots. Thorne (1999, 2000, 2003, 2006) has come up with a major revision of h is classification, bring-ing it on lines of APG, but maintaining the hierarchica l structure, and finding a place almost all families, with only 4 genera re-main ing unplaced. It is also interesting to note that his eleven subclasses are more or less complementary to the eleven informal groups of APG-II, the relationship somewhat reversed in Asterids. 1. Classification having not proceeded below the family level, the system is not useful in practice and for adoption in herbaria and floras. 2. Although a large number of families have been assembled into more or less monophyletic orders, there still exists a large number of unplaced families, and a few unplaced genera in both APG II and APweb. 3. Although most of the orders have been assembled into informal groups, no proper names conforming to the Botanical Code have been given for these groups. 4. Although APG II places all presumed primitive families of angiosperms be-fore monocots, APweb transfers the Magnilii ds to a position after monocots and commelinids. Chapter 11 Families of Pteridophytes Pteridophytes, Gymnosperms and AngioBold, Alexopoulos and Delevoryas (1987) sperms constitute Tracheophytes, a domi- included the same four group but prefer red nant group of green plants, characterized by name Microphyllophyta for Lycop odiophyta the presence of a well developed branched, and Arthrophyta for Equisetophyta. Re cent independent and dominant sporophyte, with evidence indicates that Pteridoph ytes often a vascular system consisting of xylem separated under Ferns and Fern A llies , form (tracheids-hence the name Tracheophytes, vessels in angiosperms) and phloem (sieve elements, sieve tubes in angiosperms). The group evolved nearly 42 0 million years ago, and is regarded as monophyletic. Pterido-phytes, the seedle ss vascular plants differ from higher Tracheophytes in lacking seed habit and ab sence of pollen tube, spores de-veloping freely into gametophytes, although few members exhibit heterospory and the re-duction of megaspore number to one, forea paraphyletic assemblage of groups, which represent distinct evolutionary lines and are lumped together for convenience. Recent genetic data has shown that Lyc opodiophyta are only distantly related to other vascular plants , having radiate d evolutionarily at the base of vascular plant clade, whereas Psilophyta and Equ isetophyta are much closer to true ferns. Tracheophytes are considered to includ e two major lineages: lycophytes and euphyllophytes. Latter in turn comprise two major lineages of living plants: ferns (including Psilotaceae and Equisetaceae) and spermatophytes or seed plants. Spermatophytes consist of two major lineages : gymnosperms and angiosperms. A more recent classification of Smith et al. (200 6), based on morphology as well as molecular data as such excludes Lycopodiophyt es from Ferns, divides ferns into two group: eusporangiate ferns including Marat tiaceae and leptosporangiate ferns, including the rest. The four groups, ranked

as classes are listed below: runner of seed habit. Pteridophytes form a complex heterogenous group reflecting antiquity and diverge nt evolutionary clades, and have been classified variously. Engler and Prantl re cognised Bryo-phyta and Pteridophyta as two subdivisions of the division Embryop hyta Asiphonogama. Cronquist, Takhtajan and Zimmermann (1966) recognised four gr oups within Pterido-phytes, each given the rank of a division: 1. Psilophyta 2. Lycopodiophyta 3. Equisetophyta 1. Psilotopsida 4. Polypodiophyta 2. Equisetopsida 360 Plant Systematics I?%Ip~puiii~:u~ fir_ ~~~--- 41t~nnt ii2Ce2e E~Fioly~posi s tl + ;~ Lum_~~crpsiti~r:929 E~polyLpa is+ ~t naaeaceae Eu oii'5Ll1' ingch"n_'aalb Polypaaia~as IT It Pitiridt~dti 'You --A ASN lem iaoB2 PolyNu41'S ~ P litn slds3B2e dtinn5idddii2e~B#~ ~ul}?~uiiic~ 9 tx c`durt~di2 2 NSp ~"~ LindSabatx~ae Mi~ixdy~i7B2 diCksuni20B2e ytheaeeae -01L oBae i~Uitidlds Co_ra,lapia~po~2ny~aibs_ .-.

41F~i u ~yy%2C6Be u 1090 ~u,xma~a~:bae TLhyr9uNib Kioe2e~ IS21vii5i21~s Mdr5iIB20B2B Schfiraeacea~ IAntim~txiati ISchi~aeales ~Lyyuiii2oB2e L~tuSpo~anyi2lBS=;~ ~~ :fi_,_IBiel1eni2lBs 41NOUHU-n-Gegae I 4iymit_Tq hy112oB2 };hey. Chcp1 1y1121B5 45rr~nnd s3eael 1M smun4ne 5 YMditdiles Fe ris+ ~Equi5~i2 ~2~1 , Equ~i21~~ , Equ~i~p~sici ~~,PsiI tapsi ~-- Psiltri20B2e ~Psdoi2ies ~phiaglassaaeael:~M1~~hiayloesales Consensus phylogeny representing relationships among ferns (after Smith et al., 2006). Families of Pteridophytes 361 3. Marattiopsida etic orders, and 37 families, 32 of which are 4. Pteropsida strongly supported as monophyletic. As a departure from traditional classifi-cations, Ophioglossaceae are included u n-der Psilopsida. The classification recogniz-es four monophyletic classes, 11 m onophylPeridophytes include nearly 10280 species, 1280 belonging to lycophytes and near ly 9000 to ferns. Latter comprise horsetails, whisk ferns, and all eusporangiate and Families of Pteridophytes (classification 0f Ferns according to Smith et al., 2006) Division: Tracheophyta- Vascular plants Lycopodiophytes (Lycophytes) Order: Salviniales Lycopodiaceae Marsileaceae Selaginellaceae Salviniaceae Isoetaceae Cyatheales Euphyllophytes Thyrsopteridaceae Monilophytes- Ferns Loxomataceae Class: Psilotopsida Culcitaceae

Order: Ophioglossales Plagiogyriaceae Ophioglossaceae Cibotiaceae Psilotales Cyatheaceae Psilotaceae Dicksoniaceae Class: Equisetopsida Metaxyaceae Order: Equisetales Polypodiales Equisetaceae Lindsaeaceae Class: Marattiopsida Saccolomataceae Order: Marattiales Dennstaedtiaceae Marattiaceae Pteridaceae Class: Pteridopsida Aspleniaceae Order: Osmundales Thelypteridaceae Osmundaceae Hymenophyllales Woodsiaceae Hymenophyllaceae Blechnaceae Gleicheniales Onocleaceae Gleicheniaceae Dipteridaceae Dryopteridaceae Lomariopsidaceae Matoniaceae Tectariaceae Schizaeales Oleandraceae Lygodiaceae Davalliaceae Anemiaceae Polypodiaceae Schizaeaceae Bold typed families are described and illustrated 362 Plant Systematics leptosporangiate ferns and are character-ized by lateral root origin in the endo der-mis, usually mesarch protoxylem in shoots, a pseudoendospore, plasmodial tap etum, and sperm cells with 30 1000 flagellae. A brief description of major families of Pteridophytes is presented in the follo wing pages. *************** Lycopodiaceae Mirbel Club Moss Family Cosmopoliton, diverse in tropical montane and alpine regions, rare in arid clima te. 3 Genera, 380 species Salient features: Terrestrial or epiphytic herbs, stems dichotomously branched; leaves simple, 1-veined, non-ligulate; spo-rangia in axils of sporophylls, homos

porous, usually aggregated into strobilus; spores with 3-branched scar. dent and reaching up to 2 m, arising from creeping rhizome, rarely highly reduce d as in Huperzia drummondii (Phylloglossum drummondii) scarcely exceeding 10 cm and all aerial parts dying off at the end growing sea-son only underground tuber persisting. Roots dichotomously branched, adventitious, root hairs paired. Stem dichotomously branched, erect, creeping or pendent, slender, protostelic; unequ al dichotomy often resulting in production of condensed axes forming bulbils for v egetative propagation (Lycopodium Genera: Huperzia (300 species), Lycopodiel-la (40) and Lycopodium (40). Description: Non-woody terrestrial or epiphytic perennial herbs, sometimes pena

Figure 11.1 Lycopodiaceae. Lycopodium phlegmaria. A: Pendulous branch with termi nal strobili; B: Vegetative leaf; C: Longitudinal section of strobilus; D: Sporophyll in adaxi al view; E: Same in abaxial view; F: Vertical section of fetile branch of L. luc idulum showing sporangia in axils of unmodified sporophylls. Families of Pteridophytes 363 selago). Leaves simple, small and 1-veined (microphylls), nonligulate, up to 2 c m long, covering the stem densely, spirally arranged or opposite, linear or scal e-like, appressed or spreading, usually entire, rarely serrate (Lycopodium serra tum). Sporangia large, kid-ney-shaped, sessile or short stalked, singly in axils (adaxial) of sporophylls which are similar to foliage leaves (and restricted to distal end of stem or in alternate sterile and fertile zones) or well different iated (smaller than foliage leaves and with dentate mar-gin) forming a strobilus , homosporous, sub-globose to reniform, shortly stalked, open-ing by transverse slit, sometimes folding back to expose spores; spores subglobose or tetrahedral, with a 3-branched scar. Game-tophyte green when on soil surface, non-green when subterranean, irregularly lobed, often living up to 25 years; antheridia sunk-e n, spermatozoids produced in large num-bers, biflagellate. flash of light. They were used in early Chinese fireworks, by magicians and sorc erers in Middle ages, and as flash in early photography, and first photocopying ma-chines. Spores have also been used as industrial lubricants and in surgical g loves and condomes. Plants of several species were ear-lier gathered for making Christmas wreaths. Phylogeny: Although the family is well circumscribed, the generic limits have un dergone considerable readjustment. Often treated under a single genus Lycopodium , or divided to five genera Lycopodium, Huperzia, Diphasiastrum, Lycopodiella an d Phylloglossum. Last genus includes a single peculiar species of reduced plants found in Australia New Zealand and Tasmania, but gametophyte morphology and rbc L studies support the inclusion under Huperzia. Diphasiastrum, a genus of nearly 16 species of low gymnosperm like plants, and often called as Ground-pine or Gr ound-cedar, is better placed under Lycopodium. The family is very old in fossil record, dating back to 380 MYA, mostly dominated by lycopod trees. Economic importance: Family is of little im-portance. The spores of Lycopodium c ontain a highly volatile oil and ignite rapidly into a *************** Selaginellaceae Wilk. Spike Moss Family Worldwide, predominant in tropics, a few species extending to temperate and arct ic region. 1 Genus, 750 species Salient features: Terrestrial or epiphytic in moist habitats, creeping (S. kraussiana) or erect (S. rupestris), few in semi

arid regions, rarely epiphytic. Roots dichotomous-ly branched, arising from dist al end of rhizophore, adventitious; rhizophores arising from stem. Stem dichotom ously branched, with both creeping and short erect branches or with only erect b ranches, protostelic, with vessels. Leaves simple, small and 1 veined (microphyl ls), ligulate (ligule axillary or near leaf base; tongue- or fan-shaped with hya line sheath at base), up to 1 cm long, covering the stem densely, spirally arran ged, herbs, stems dichotomously branched, producing rhizophores; leaves simple , 1- veined, ligulate, dimorphic, often 4-r anked; sporangia in axils of sporophylls, het-erosporous, aggregated into usuall y 4-angled strobilus. Genera: Single genus Selaginella (750 species). Description: Mainly terrestrial perennial plants, usually low-growing, mostly gr owing 364 Plant Systematics Figure 11.2 Selaginellaceae. Selaginella kraussiana. A: Portion of plant; B: A p art of same from upper view showing arrangement of leaves; C: Vertical section of ligule; D: Port ion of branch of S. pallescens; E: Small portion of same showing one megasporang ium (left) and one microsporangium (right); F: Strobilus of S. watsoni prolifera ting at apex into vegetative shoot. S. oregana. G: Vertical section of portion o f strobilus; H: Vertical section of microsporangium; I: Vertical section of mega sporangium. isomorphic (all similar) or dimorphic and 4- ranked, with two upper or dorsal ro ws of small-er leaves and two lower or lateral rows of larg-er leaves. Sporangia borne in axils of well-differentiated sporophylls, usually on 4-angled terminal strobili, heterosporous, microspospores tetrahedral with prominent triradiate mark and with characteristic orname ntation. Gametophyte unisexual, developed within respective spore walls; spermat ozoids biflagellate, smallest among vascular plants. Several species especially those growing in dry rangia and megasporangia occurring on climate can survive long periods of drough t same (with megasporangia in upper part of due to small leaves covered with thi ck strobilus, microsporangia in lower part- S. cuticle, and branches curling up into a ball. helvetica or in two opposite rows- S. oregana) Such plants revive f ast with availability of or different strobili; microsporangium with water and a re known as resurrection plants more than 100 microspores about 20-60 mm S. bryo pteris of India- regarded by some as in diameter; megasporangium with 4 large Sanjeevani booti of Ramayana legend, an d megaspores about 200-600 mm in diameter; S. lepidophylla of Mexico and Texas). Families of Pteridophytes 365 Economic importance: Family is of little im-portance. Only a few species are gro wn as ornamentals. Isoetaceae, it is differentiated by smaller leaves and superficial sporangia, le aves being 2-100 cm long, onion-like and sporangia initially embedded in Isoetac eae. Leaf dimorphism is considered as an adaptation to poor light, as the specie s commonly inhabits forest floor. Phylogeny: The family with single genus is well differentiated from Lycopodiacea e in ligulate leaves and heterosporous habit. From *************** Isoetaceae Reichenbach Quillwort Family Widely distributed in tropical and temperate Americas, Europe, Asia, Africa, Aus tralia and New Zealand. 1 Genus, 150 species

Salient features: Terrestrial perennial herbs of marshy areas, stem short with s econdary growth, , leaves simple , long, quill-like, 1-veined, ligulate; sporang ia borne singly, sunken at base of leaves, het-erosporous. bundles, ligulate with ligule inserted above sporangium, hardened scales and phy llopodia surrounding the leaves; all leaves potential sporophylls. Sporangia sol itary, borne adaxially embedded in cavity of swollen base of sporophylls, micros porophylls and megasporophylls usually borne in alternate cycles, sporangium cov ered partially or completely on adaxial side by velum; megasporangium with 50-30 0 megaspores; microsporangium with 0.15-1 million microspores; microspores elong ate, about 45 ^ long; megaspores 250-900 ^ in diameter; spores set free by disin tegration of sporanGenera: Single genus Isoetes (150 species). Description: Terrestrial tufted perennial plants usually found in marshy areas, often in periodically inundated pools. Roots firm, arising from grooves of lobed stem in radi-ating rows, unbranched or dichotomously branched, containing eccen tric vascular strand and surrounding lacuna. Stem short, erect, cormose (corm-li ke), rarely rhizoma-tous, having secondary growth, protostelic stele anchor-shap ed with upturned lobes (near base), lobed by a broad basal groove into 2-4 lobes , rough on sides due to slough-ing off of cortical tissue. Leaves simple, lin-ea r, long, quill-like, resembling narrow-leaved species of Allium , up to 100 cm l ong, swollen at base, 1-veined (microphylls), containing 4 transversely septate longitudinal lacunae, a central collateral vascular strand and frequently severa l peripheral fibrous gial wall, dehiscence zone not developed. Gametophyte unisexual, developed withi n re-spective spore walls (endosporic); microgametophyte 9-celled, antheridium s ingle, spore wall cracking to release 4 multiflagellate spermatozoids each with terminal vesicle; megagametophyte with 1-several archegonia, often with rhizoids , exposed by cracking of spore wall. Economic importance: None. Phylogeny: The family with single genus is well differentiated from Lycopodiacea e in ligulate leaves and heterosporous habit. 366 Plant Systematics C B

F D Figure 11.3 Isoetaceae. Isoetes engebmanni. A: Plant; B: Corm with attached leaf bases; C: Transverse section of leaf; D: Leaf base with ligule and sporangium; E: Same in verti cal section; F: Megaspore. From Selaginellaceae, it is differentiated by much longer linear leaves, short e rect cormose stem and sunken sporangia; leaves being less than 1 cm long, sporan gia superficial and stem branched in Selaginellaceae. Fossil tree Lepidodendron is m ore closely related to Isoetes among the extant lycophytes. *************** Ophioglossaceae C. Agardh Adder s-tongue Family Widely distributed in tropical and temperate regions, more abundant in old field s and disturbed pastures. 4 Genera, 73 species Salient features: Terrestrial perennial herbs with tubers or rhizomes, leaves wi th branched veins, fertile portion a simple or branched spike, arising from surface of leaf at the junction of petiole and blad e, stipules present, sporangia homosporous,

Families of Pteridophytes 367 C. i/ .A 3

B Figure 11.4 Ophioglossaceae. Ophioglossum reticulatum. A: Plant; B: Portion of f ertile spike; C: Small portion of same enlarged. Botrychium daucifolium. D: Leaf with fertile bra nch; E: Portion of fertile branch; F: Small portion of same showing sporangia. aggregated into sporophore, sporangium wall more than one cell thick, annulus ab sent, gametophytes subterranean. unfolding lengthwise (conduplicate) and not circinate, with branched veins (euph yll or macrophyll), venation reticulate (Ophioglossum) or open dichotomous (Botr ychium); sin-gle fertile branched (Botrychium) or simple (Ophioglossum) sporopho re arising from surface of leaf (trophophore), more than one sporophore arising in Cheiroglossa; petiole fleshy with expanded sheathing base; stipules present, sheathing, persisting even after decay of leaves. Sporangia aggregated in fertil e portion (sporophore), thick walled (eusporangiate-more than one celled thick), homosporous, not clustered in sori, separated (Botrychium) or forming synangia (Ophioglossum), exposed or embedded in spikelike sporophore, annulus absent; spo res thousands per sporangium, chlorophyllous. Major Genera: Botrychium (45 species), Ophioglossum (25), Cheiroglossa (1), Mank yua (1) and Helminthostachys (1). Description: Terrestrial perennial plants, very rarely epiphytic. Roots unbranch ed, adventitious, lacking root hairs. Stem short, tuberous or rhizomatous, aeria l portion per-ishing after growing season, rarely ever-green (Botrychium dissect um, B. multifidum). Leaves simple (Ophioglossum) or more or less palmately compo und (and looking like hand-Cheiroglossa) to many times pinnately compound (Botrychium), up to 50 cm long, 368 Plant Systematics Gametophyte subterranean, nongreen, mycorrhizal; antheridia and archegonia dis-t ributed over the surface of gametophyte; antheridia sunken producing multiflagel late spermatozoids in large numbers. alternate ternately compound leaf, open dichotomous venation and baculate spores like Helminthostachs; open dichotomous venation and horizontally dehiscent spor angia like Botrychium; fleshy spike, sunken horizontally dehiscent sporangia and vegetative leaf prop-agation like Ophioglossum; and nearly sessile sporophore l ike Cheiroglossa. Mankyua possesses a fertile spike (sporophore) originating dor siventrally from the adaxial side of the vegetative trophophore, eusporangiate s porangia without annuli, and noncircinate leaf vernation. These three characters , especially the first, place it in the Ophioglossaceae. Also, recent DNA analys es confirm the placement of Mankyua in Ophioglossaceae (Sun, 2002). The family O phioglossaceae is sometimes divided into two families: Ophioglossaceae and Botry chiaceae. The discovery of Mankyua unites the family. The mixture of character s tates pre-sents in Mankyua combines the distinguish-ing states for each family, and thus abolish-es the taxonomic boundaries between the two segregate families. Economic importance: None Phylogeny: The family shares apomorphies of unbranched roots and absence of root hairs. The two may represent a transitional stage for total absence of these in Psilotace-ae. The distinctness from other ferns and affinity with Psilotaceae i s supported by DNA sequence data, although there is little mor-phological suppor t. The sporophore repre-sents a unique apomorphy of the family. The family repre sents highest chromosome number in plants (n = 621 + 10 fragments). Mankyua chej uense is a recently discovered genus and species found in a lowland swampy area

on Cheju Island, off the southern coast of the Korean Peninsula (Sun et al., 200 1). Its affinities with the Ophioglossaceae are fairly clear. It has creeping rh izome, *************** Psilotaceae Kanitz Whisk Fern Family Pantropical and warm tropical regions, mainly in Southeast Asia and South Pacific. 2 Genera, 12 species Salient features: Perennial rhizomatous herbs without roots, stems dichotomously branched, leaves scale-like, 1-veined or veinless, 2-3 sporangia fused into thi ck walled synangium, homosporous, gametophyte free living, subterranean. producing gemmae for vegetative reproduction in Psilotum. Stem erect, pendent or creeping ; dichotomously branched, appearing like bundle of green forking stick s; vascular cylinder protostelic; branch tip flattened laterally and appearing l eaf-like in Tmesipteris. Leaves (euphylls) scale-like, spirally arranged, 2-rank ed, 1-veined (Tmesipteris) or veinless (Psilotum-termed enations) or nearly so, awl-shaped to lanceolate, simple or once forked. Sporangia homosporous with 2-3 celled thick wall, two or three together, fused into 2-3 lobed synangium (2-lobe d in Tmesipteris, 3-lobed Genera: Tmesipteris (10 species) and Psilotum (2). Description: Terrestrial or more commonly epiphytic perennial herbs, often pendu lous. Roots absent, plant anchored by rhizome with rhizoids and mycorrhizal fung i; rhizome Families of Pteridophytes 369 E Figure 11.5 Psilotaceae. Psilotum nudum. A: Plant; B: Portion of sterile branch with leaves; C: Portion of fertile branch with 3-lobed synangia; D: Synangium; E: Synangium i n cross section. Tmesipteris tannensis. F: Plant; G: Sporophyll; with 2-lobed sy nangium; H: Longitudinal section of synangium. in Psilotum), yellowish at maturity, subtend-ed by a forked appendage (Sporophyl l); spores numerous, bean-shaped, pale, in tetrads. Gametophyte free living, sub terranean or superficial, often with mycorrhizal fungi, ir-regularly branched; a ntheridia and archego-nia in large numbers all over the gametophyte; spermatozoi ds spirally coiled, multiflagellate. Phylogeny: The phylogeny of the family is a matter of considerable speculation. Wet-tstein (1901) agreed with Engler and Prantl in regarding the group as advanc ed over Selaginellales and Isoetales. Eames (1936), however, considered the fami ly as the most primitive of the extant vascular plants, lack-ing roots and true leaves. The recent morphological, chemical and molecular studies, however, suppo rt relationship with eusporangiate group Ophioglossales, and that the group lost roots secondarily. Economic importance: Family is of little im-portance, Psilotum nudum often grown as greenhouse plant. *************** 370 Plant Systematics Equisetaceae Michx. Almost worldwide except Australia, New Zealand and Antarctica, in temperate and tropical climate. 1 Genus, 15 species Horsetail Family Salient features: Annual or perennial rhi-zomatous herbs with jointed ribbed aer ial stems, leaves reduced with single vein, spo-rangia on peltate sporangiophore s, homo-sporous, spore wall with elaters, gametophyte green, thallus shaped, spe rmatozoids mul-tiflagellate. unbranched or with whorled branches, with swollen jointed nodes, internodes long itudinally ribbed, with ridges and grooved outside, hollow with central canal, w ith additional smaller canals under the ridges; growth intercalary; outer surfac

e covered with silica cells giving plant the texture of sand paper, hence the na me scouring rush . Leaves small, usually less than 1 cm, 1-veined, whorled and fuse d into a sheath, latter more or less swollen, each leaf corresponds to the ridge below, thus the number of leaves at each node as the number of ridges, the leav es of the successive nodes as well as ridges alternating. Sporangia large, lacki ng annulus, homosporous, hanging Genus: Single genus Equisetum (15 species). Description: Terrestrial or aquatic annual or perennial herbs, sometimes evergre en, with subterranean much-branched rhizome. Roots slender, arising from horizon tal subterranean rhizome. Stem subterranean as well as aerial, latter erect, gre en, E C r

Figure 11.6 Equisetaceae. A: Plant of Equisetum arvense showing rhizome, vegetat ive branch and fertile branches. E. telmateia. B: Portion of plant with sterile and fertile branches; C: Sporangiophore viewed from below; D-E: Spore with coiled elaters; F: Antherozoid. E. hyemale. G: Node with sheath of leaves; H: Strobilus; I: Spor angium. Families of Pteridophytes 371 from lower surface of peltate sporangiophore, whorls of sporangiophores aggregat ed into strobilus terminating green branches or unspecies, especially E. hyemale, were used for scouring pots and pans, hence the common name scouring rush. branched nongreen stems arising separately from rhizome; base of strobilus with collar of Phylogeny: The family is quite di stinct in fused sterile appendages; sporangium with its jointed stems with ridge s and grooves and 2-layered wall; sporangiophore with slender hollow within, who rled leaves and peltate stalk and hexagonal disc at distal end, 5-10 sporangioph ores. Earlier the family was con-sporangia hanging from each disc; spores sidere d distinct from ferns, but the molecuspherical, many thousand per sporangium, la r data and morphological characters such green, wrapped by 4-6 straplike elaters aris- as spermatozoids and root structure, support ing from outer wall, assisting in spore dis- placement with ferns. persal, wall with four apertures. Gameto-phyte pinhead sized, thallus shaped, gr een, developing near soil surface; antheridia and archegonia developing simultan eously; sper-matozoids multiflagellate. The genus is divided into two subgenera: Equisetum (8 species) with branched ste m and superficial stomata and Hippochaete (7 species) with unbranched stems and sunken stomata, sometimes recognized as distinct genera. Economic importance: Family is of lesser importance. Silica covered stems of sev eral The family is represented in the fossil record as early as Devonian 408-360 MYA. *************** Osmundaceae Bercht. & J. Presl Worldwide except very cold climates and Pacific Islands. 3 Genera, 18 species Royal Fern Family Salient features: Terrestrial plants with wiry roots, persistent stipe bases, di morphic fronds, pinnate compound sterile fronds, sporangia with many spores, ann ulus shield-like, gametophyte green, developing on soil surface. xylem strands, latter often conduplicate or twice conduplicate in cross section.

Leaves up to 2 m long, spirally arranged, usually dimorphic with distinct steri le and fertile fronds; sterile fronds green, once- thrice-pinnate compound, with expanded petiole base, circinate before unfolding, usually leathery, rarely fil my (Leptopteris hymenophylloides), usually covered with hairs especially when yo ung; fertile fronds usually brown, much narrower; sometimes sterile and fertile segments present on same leaf. Sporangia large, shortly stalked, intermediate be tween eusporangiate and leptosporangiate ferns arising from single initial (lept osporangiate) or many initials (eusporangiate) and archesporial cell tetrahedra Genera: Osmunda (10 species), Leptopteris (6) and Todea (2). Description: Terrestrial plants common in wetlands and lowland forests, sometime s tree-like (Leptopteris). Roots wiry, adventitious, generally two below each leaf base. Stem erect to decumbent, branched, ma s-sive, often covered by persistent leaf bases, ectophloic siphonostele with rin g of discrete 372 Plant Systematics i ~'Y/ E

c y B A Figure 11.7 Osmundaceae. Osmunda javanica A: Portion of plant with sterile and f ertile fronds; B: Portion of fertile frond; C: Fertile frond in cross section; D: Portion of fe rtile pinna of Todea africana; F: Portion of fertile pinna of Leptopteris hymeno phylloides. (leptosporangiate) or cubical (eusporan-giate), separate or in loose clusters, b orne on distinct fertile fronds (Osmunda) or on undersurface of foliage leaves a long veins (crowded in Todea, sparse in Leptopteris), not forming sori, annulus poorly developed shield-like plate or broad horizontal band around the sporangium, homosporous, producing 128-512 spores, sporangia opening by apical slit; spores green, subglobose, with triradiate mark. Gametophyte larg e green, cordate, developing on soil surface; antheridia emergent producing up t o 100 spermatozoids. Families of Pteridophytes 373 Economic importance: Family is of lesser importance. A few species like Osmunda cinto leptosporangiate ferns evidenced by nu-merous spores, rudimentary annulus, mo re than one celled thick wall, lack of sorus; position supported by rbcL sequenc e data. This conclusion is also supported by long fossil record, dating back to Permian 286-245 MYA. namomea (Cinnamon fern) and O. regalis (Royal fern) are grown as ornamentals. Phylogeny: The family is considered sister *************** Marsileaceae Mirbel -

Nearly worldwide in warm temperate and tropical areas. 3 Genera, 75 species Water-Clover Family Salient features: Aquatic plants with floating long-petioled leaves, leaflets 2 to 4, sori enclosed in hard sporocarp, heterosporous, sporangia without annulus, megasporangium with one megaspore, microsporangium with 16-64 microspores. Genera: Marsilea (69 species), Pilularia (5) and Regnellidium (1). Description: Plants perennial, aquatic or rooted in mud with creeping rhizome, r arely xerophytic and developing underground Figure 11.8 Marsileaceae. Marsilea quadrifolia. A: Plant with sporocarps; B: Ver tical section of sporocarp; C: Sporocarp in longitudinal section; D: Spore; E: Spermatozoid; F: P ortion of plant of M. polycarpa; G: Portion of plant of Regnellidium diphyllum. 374 Plant Systematics tubers from rhizome (M. hirsuta); land forms with short internodes, branched roo ts, long petiole and stomata on both leaf surfaces; aquatic forms with long inte rnodes, un-branched roots, flexible petiole and stoma-ta mainly on the upper sur face. Roots aris-ing from creeping rhizome, one or two at each node. Stem a slen der rhizome, creep-ing, growing on soil surface or subterra-nean, often with hai rs, dichotomously branched. Leaves floating or emergent, long petioled, blade fi liform (Pilularia) or di-vided into 2 (Regnellidium) or 4 clover-like (Marsilea) leaflets, circinate before unfold-ing, leaflets folded together upwards until n early mature, veins dichotomously branched but often fusing towards tips. Sporan gia heterosporous, arranged in sori, lat-ter without indusium, enclosed in hard pea-shaped, bean-shaped or subglobose sporo-carps borne singly on short stalks n ear or at base of petioles, sometimes stalk branched bearing 2-3 sporocarps (M. quadri-folia), rarely several sporocarps on one pet-iole (M. polycarpa); each sp orocarp with rows of sori along either side, with 2 (M. aegyptiaca) to 20 (M. quadrfiolia); sorus with large megasporangia along the crest and microsporangia along the sides; sporangia without annulus; megasporangium w ith one megaspore; microsporangium with 16 64 microspores; sporangia attached to a gelatinous ring-like structure called sorophore, that swells with water.; mic rospore small globular, producing 16 spermatozoids, latter multiflagellate and c orkscrew shaped with prominent vesicle; megagametophyte producing single archego nium. Economic importance: Family is of little importance, with Marsilea species often grown as curiosity. Phylogeny: The family is considered sister to leptosporangiate ferns due to pres ence of numerous spores, rudimentary annulus, more than one celled thick wall an d lack of sorus. This conclusion is also supported by evidence rbcL sequence dat a and long fossil record, dating back to Permian 286-245 MYA. Regnellidium is ex ceptional non-flowering plant with latex tubes. *************** Salviniaceae Martynov Worldwide in tropical and temperate climates. 2 Genera, 16 species Mosquito-fern Family Salient features: Aquatic free-floating plants, small simple leaf blades, veins branched, sporocarps flattened and soft, spo-rangia large, heterosporous, gameto phyte endosporous. zag stem (Azolla) or absent and represented by lower third row of leaves which a re mod-ified into root-like structures (Salvinia). Stem a rhizome, zigzag, horiz ontal, dichotomously branched, protostelic; stem fragile and readily breaks resu lting in proliferous vegetative propagation, and in often covering entire water surface. Leaves simple, sessile, less than 15 mm long, rounded to oblong, entire ; in Azolla imbricated in two rows harbouring nitrogen-fixing cyanobacterium Ana baena azollae in leaf cavities; in Genera: Salvinia (10 species) and Azolla (6). Description: Free-floating aquatic plants growing in lakes and ponds, often form ing dense floating mats, with slender branch-ing rhizome. Roots hanging down fro m zig-

Families of Pteridophytes 375

"~.. C

D F Figure 11.9 Salviniaceae. Salvinia natans. A: Plant; B: Two floating leaves and a submerged leaf. C: Sporocarps of S. rotundifolia with a megasporangium and a microsporangium; D: Vertical section megaspore. Azolla microphylla. E: Plant; F: Microsporocarp s plit open to show microsporangia; G: Megaspore. whorls of three appearing in three rows in Salvinia with upper two rows floating and cov-ered with hairs whereas the third lower row submerged and finely dissec ted into root-like structure (hairs septate unlike true root hairs); veins of le af-blades branched, free (Azolla) or anastaomosing (Salvinia). Spo-rangia hetero sporous, leptosporangiate, ar-ranged in sori; In Salvinia megasporangia and micr osporangia in different sporocarps (megasporocarp with up to 25 megasporan-gia; microsporocarp with numerous microsporangia); sporocarps soft; in Azolla spo-ran gia are enclosed in indusium with ei-ther one megasporangium or several mi-crosp orangia; megasporangium with single megaspore, microsporangium with several microspores (all microspores surrounded by hardened tapetal cytoplasm are shed as a single mass known as massula); annul us absent; spores globose, trilete. Gametophyte endosporous, megagametophytes an d microgametophytes protruding through sporangium wall; megagametophytes floatin g on water surface with archegonia directed downward; microgametophytes remainin g fixed to sporangium wall. Economic importance: Both genera are invasive weeds in warm climates, often chok ing lakes, ponds and drains. Salvinia is frequently sold as aquarium plant. Azol la is 376 Plant Systematics often used as green manure, owing to the presence of nitrogen-fixing cyanobacter ium. into two distinct families on the basis of absence of roots and sori in distinct sporocarps in Salvinia, as against the presence of roots and absence of sporoca rps in Azolla. Phylogeny: The family is sometimes split *************** Cyatheaceae Kaulf. Tropical and subtropical montane forests and cloud forests. 5 Genera, 600 specie

s Scaly Tree-fern Family Salient features: Tall arborescent palm-like ferns with thick trunk, leaves cove red with scales, large, pinnate to bipinnate, circinate before unfolding, homosp orous, sporangia in sori, annulus continuous. Major Genera: Alsophila (230 species), Sphaeropteris (120) and Cyathea (110). Description: Palm-like ferns with single erect arborescent trunk. Stem erect, a T

~ / ffi ~r . ~. B 71

tea ,

Figure 11.10 Cyatheaceae. Cyathea spinulosa. A: Portion of frond with fertile (b elow) and sterile (above) pinnules; B: Sterile pinnules enlarged; C: Portion of fertile frond of C yathea elegans; D: Portion of fertile frond of C. medullaris. Families of Pteridophytes 377 usually unbranched, reaching 25 m in height, rarely decumbent or creeping, dic-t yostelic, stem and leaves covered with large and small scales. Leaves large ofte n reach-ing 5 m in length, once to twice pinnate com-pound, with deeply pinnatel y lobed leaflets, rarely simple, circinate before unfolding; petioles with obvio us, usually discontinuous pneumathodes in two lines; veins free or anastomosing. Sporangia homosporous, ar-ranged in sori on abaxial surface of fronds, annulus continuous, not interrupted by sporangium stalk, oblique, allowing the sporangiu m to open horizontally; indusium completely covering the sorus or absent; spores usually 64 per sporangium, tetrahedral, trilete, variously ornamented. Game-tophyte green, cordate. Economic importance: Various members of the family are often grown as ornamental s, fibrous rhizomes often used as base for epiphytes in greenhouses. Phylogeny: The family is well characterised by uniform chromosome number n = 69. Hymenophyllopsis with small distinct stems is quite distinct, but shares the pr esence of scales with other members of the family. There is some molecular evide nce supporting the inclusion of this genus in Cyathea. *************** Pteridaceae Ching

Maidenhair Fern Family Tropical and subtropical montane forests and cloud forests. 50 Genera, 950 speci es Salient features: Terrestrial or epiphytic, rhizomes covered with scales, leaves all sim-ilar, petiole base with persistent scales, sori near margin, indusium a bsent. variously anastomosing and forming a reticulate pattern without included veinlet s. Sporangia homosporous, leptosporangiate, arranged in sori on abaxial surface near margin, forming a continuous band and pro-tected by reflexed leaf margin fo rming false indusium (Pteris) or two-lipped indusium with thick upper lip and th in papery lower lip (Pteridium), true indusium being absent, or arranged along a ll leaf veins; sporangium stalk with 2-3 rows of cells; sporangia each with a ve rtical, interrupted annulus cover-ing three-fourth of sporangium, a strip of 4 c ells below free end of annulus forms stomium, receptacles not or only obscurely raised; spores 64 per sporangium, globose or tetrahedral, trilete, variously orn amented. Gametophyte heart-shaped. Major Genera: Pteris (200 species), Adian-tum (150) and Cheilanthes (150). Description: Terrestrial or epiphytic, rarely aquatic, growing in wide variety o f habitats such as deserts, ponds, cultivated areas, forest canopies to mangrove swamps. Root adventitious, arising from lower surface of rhizome. Stem represen ted by rhizome, short, creeping to erect, sometimes deeply penetrating, dichotom ously branched, usually covered with scales, less often with hairs. Leaves all s imilar (monomorphic), dimorphic in few genera; petiole with persistent scales ne ar base; blade simple, 1-6 pinnate or pedate; veins free or forking, Economic importance: Some species of Adiantum and Pteris are grown as ornamental s. 378 Plant Systematics

s E C;

D Figure 11.11 Pteridaceae. Pteris pinnule. Pteris patens. D: Portion of fertile frond; E: Portion of pinnule. hii. A: Portion of frond; B: Petiole of same; C: Portion of a Phplogenp: The family shows a wide diversity due to its adaptation to diverse ha bitats. The family contains five clades that could be recognized as distinct fam ilies Parkeriaceae (Acrostichum and Ceratopteris), Adiantaceae (Adiantum and the ten vittarioid genera), Cryptogrammaceae (comprising Coniogramme, Cryptogramma, and Llavea;), Sinopteridaceae and Pteridaceae (Pteris and its immediate allies) , or else five subfamilies base number (x = 38 as against 29 or 30 in the family) and in having dimorphic s pores ( incipient heterospory ) but nests with other genera of Pteridaceae, subfam. Pteridoideae. Ceratopteris is distinct from other members of the family in coars ely ridged spores with parallel striations; spores 32 or fewer per sporangium; s porangia with ill-defined annuli; aquatic habitat; x = 38, often placed in disti nct family, but nests within Pteridaceae in all molecular analyses, and it appea

rs to be sister to Acrostichum. within Pteridaceae. Platyzoma, sometimes recognized as an isolated family, is aberrant in chromosome *************** Families of Pteridophytes 379 Aspleniaceae Newman Nearly worldwide, most diverse in tropics. 1 Genus, 700 species Spleenwort Family Salient features: Terrestrial or epiphytic, rhizomes covered with scales, leaves all sim-ilar, sori elongate, indusia linear, laterally attached, sporangium sta lk with one row of cells, spores bilateral. sented by rhizome, erect or creeping, usually covered with scales especially tow ards apex, dictyostelic. Leaves all similar (mono-morphic); petiole with scales near base, with C-shaped vascular strands fused distally back-to-back to form xshape; blade simple to 5-pinnate, often with glandular hairs and a few linear sc ales; veins pinnate or fork-ing, usually free, sometimes reticulate without incl uded veinlets. Sporangia homosporous, arranged in sori on abaxial surface Genus: Asplenium (700 species). Description: Terrestrial or epiphytic peren-nials, some members growing on rocks (epi-petric) in moist or wet forests. Stem repreFigure 11.12 Aspleniaceae. A: Plant of Asplenium ensiforme with simple sterile a nd fertile leaves; B: Leaf of A. alternans; C: Portion of same enlarged; D: Portion of fertile pinn a of A. bulbiferum; E: Sorus. 380 Plant Systematics along veins, linear or curved; indusia lin-ear, laterally attached; sporangium s talk with one row of cells; spores bilateral, reni-form, monolete, with winged p erine. 10 genera in the family, although large number of species fit well under Aspleni um. The segregate genera Camptosorus and Loxoscaphe , Diellia (endemic to Hawaii ), Pleurosorus, Phyllitis, Ceterach, and Thamnopteris clearly nest within Asplen ium s.l. Hymenasplenium, however, with a different chromosome base number than n early all of the other segregates, as well as distinct root characters (Schneide r et al., 2004), appears to represent the sister clade to the rest of the specie s in the family, and this name could be adopted as a well-supported segregate ge nus. Economic importance: Some species of Asplenium such as A. scolpendrium (heart s-to ngue fern) are grown as ornamentals. Phylogeny: The family is closely related to Blechnaceae, Onocleaceae and Thelypt eri-daceae. Some species have variously been removed to distinct genera to estab lish up to *************** Dryopteridaceae Ching Wood Fern Family Nearly worldwide, most diverse in tropics with several representatives in temper ate region. 45 Genus, 1700 species Salient features: Mostly terrestrial, rarely epiphytic, rhizomes covered with sc ales to-wards apex, leaves all similar, petioles with persistent scales towards base, sori round, covering the leaf surface, indusia round-reniform shaped, spor angium stalk with three rows of cells, spores bilateral. vascular strands arranged in a ring; blade simple to 5-pinnate, sometimes scaly or glandular, rarely hairy; veins pinnate or forking, free to variously anastomo sing, with or without included veinlets. Sporangia homosporous, arranged in sori on abaxial surface, sori rounded, closely spaced and covering the leaf surface; indusia round-reniform or peltate, rarely absent; sporangium stalk with 3 rows of cells; spore reniform monolete, perine winged. Major Genera: Elaphoglossum (500 species), Polystichum (260), Dryopteris (225) a nd Cteni-

tis (150). Description: Terrestrial, epiphytic or epipet-ric ferns. Stem represented by rhi zome, erect, ascending or creeping, sometimes scandent or climbing, usually cove red with non-clathrate scales especially towards apex, dictyostelic. Leaves all similar (monomor-phic), rarely dimorphic; petiole with persis-tent scales near b ase, with numerous round Economic importance: None. Phylogeny: Dryopteridaceae is well defined clade, except for three genera Didymo chlaena, Hypodematium and Leucostegia, whose inclusion renders this family parap hyletic. Hypodematium, often removed to a distinct family Hypodematiaceae and Le ucostegia is Families of Pteridophytes 381 Figure 11.13 Dryopteridaceae. Polystichum auriculatum. A: Plant with fronds; B: Portion of fertile frond with sori; C: Fertile pinna enlarged; D: Sterile pinna enlarged; E: Portio n of rachis of P. setiferum with fertile pinna; F: Sorus. nearly always placed in Davalliaceae be-cause of its similar indusia and sori te rmi-nal on the veins, but it differs from mem-bers of Davalliaceae in the terres trial hab-it, the more strongly verrucate spores with rugulate perispore. Tsutsumi & Kato (2006) found support for a sister relationsh ip between Hypodematium and Leucostegia, and also support for these as sister to the re-maining Eupolypods. *************** 382 Plant Systematics Polypodiaceae Bercht. & J. Presl. Widely distributed in tropics with few representatives in temperate region. 56 G enera, 1200 species Polypod Family Salient features: Mostly epiphytic, rhizomes covered with scales, leaves all sim ilar, usu-ally simple, petioles without scales, sori round, indusia absent, spor angium stalk with 1- three rows of cells. usually thick and coriaceous, rarely dimorphic; petioles cleanly abscising near their bases, leaving short phyllopodia; blade mostly simple to pinnatifid or 1-p innate; indument lacking or of hairs and scales; veins often anastomosing or ret iculate, sometimes with included veinlets, or veins free. Sporangia arranged in sori; sori abaxial, rarely marginal, round to oblong or elliptic, occasionally e longate, sometimes deeply embedded; indusium absent, sori sometimes covered by c aducous scales when young (Lepisorus, Pleopeltis); sporangia with 1 3-rowed, usual ly long stalks, frequently with paraphyses on sporangia or on receptacle; spores hyaline to yellowish, Major Genera: Grammitis (400 species), Poly-podium (150), Pleopeltis (50) and Ca mpyloneu-rum (50). Description: Mostly epiphytic and epipetric, a few terrestrial. Roots thick, wir y. Stem rep-resented by rhizome, long- to short-creeping, dictyostelic, bearing scales and hairs, scales often peltate (Pleopeltis). Leaves all similar,

E D A ~ m

r' Figure 11.14 Polypodiaceae. Polypodium wallii. A: Plant with fronds; B: Portion of fertile frond with sorus; C: Sorus; D-E: Sporangia in different views. Pleopeltis lanceolata.; F: Prtion of plant with fertile frond; G: Basal portion of fertile frond; H: Po rtion of leaf showing venation; I: Sorus; J: Peltate scale. Families of Pteridophytes 383 reniform, and monolete, or greenish and globose-tetrahedral, trilete; perine var ious, usually thin, not strongly winged or cristate. Davalliaceae, Oleandraceae and Tectariaceae. Grammitid ferns often removed as Gr am-mitidaceae (20 genera and 600 species) nest well within Polypodiaceae as evid enced by DNA sequence data (Schneider et al., 2004b), and share a large number o f morphological synapomorphies: veins free (mostly); scales lacking on blades; s etiform, often dark redbrown hairs on leaves; sporangial stalks 1 rowed; spores green, trilete; gametophytes ribbon-shaped. Economic importance: Species of tropical staghorn fern Platycerium, Phlebidium a nd Aglaomorpha are commonly cultivated in greenhouses and gardens for their inte rest-ing leaves. Phylogeny: The family is closely related to *************** Chapter 12 Families of Gymnosperms Gymnosperms comprise a small group of seed plants characterized by naked seeds ( gymno- naked, sperms- seeds) and absence of vessels (except Gnetopsids), endospe rm rangium produces numerous microspores arranged in tetrads, since each microspore mother cell undergoes meiosis to produce four haploid microspores. Microspore n u-cleus undergoes repeated divisions to form formation independent of fertilization and commonly resulting in halploid endosperm male gametophyte, which develops wall t o (absence of double fertilization), absence of become a pollen grain. The megas posieve tubes and companion cells. Group is rangium, known as ovule, on the othe r represented by nearly 15 families, 80 gen- hand, develops a single megaspore m other era and nearly 820 species, mostly consist- cell, surrounded by nucellus a nd integuing of evergreen trees and shrubs, distrib- ment, with an opening known as micropyuted worlwide and forming extensive forests le, at the end of integum ent. Of the four hap-in North America, Europe and Asia. They loid megaspores res ulting after meiosis, represent some of the largest (Sequoiaden-drod giganteum of California), tallest (Sequoia sempervirens of California and Oregon) and longest living (Pinus aristata) organ isms in the world. three degenerate, and only one megaspore is functional. Latter, after repeated n uclear divisions and wall formations produces a female gametophyte with several archegonia, consisting of an enlarged egg cell and two or four neck cells. Polle n grains of gym-nosperms are carried by wind, land on micropyle and adhere to st icky fluid released by the female gametophyte. The pollen germinates to produce a pollen tube, that grows through nucellus and releases two sperms. One fuses wi th the egg to form zygote after fertilization. Latter develops into an embryo Gymnosperms are woody trees or shrubs, herbaceous plants being absent from the g roup. The plants have well-developed tap root system, sometimes with symbiotic n i-trogen fixing cyanobacterium (coralloid roots of Cycas) or mycorrhizae (Pinus) . Vascular cylinder has xylem with tracheids with bor-dered pits and phloem with sieve cells. Leaves lack lateral veins, but are compen-sated by transfusion tis sue. Sporangia are heterosporous, microsporangia and me-gasporangia borne on mic rosporophylls and megasporophylls, respectively; latter often arranged in distin

ct cones. Each microspowithin matured ovule known as seed. Gymnosperms have been recognized as group distinct from angiosperms since Robert Brown (1827) established their identity. Four distinct groups of extant gymnosp erms viz: Cycads, Conifers, Gnetopsids and Families of Gymnosperms 385 monotypic Ginkgo have, however been treat-ed differently by various authors. Cha mber-lain (1935) divided gymnosperms into two classes: Cycadophytes (Cycadofilic ales, Ben-nettitales and Cycadales) and Conifero-phytes (Cordaitales, Ginkgoales , Coniferales and Gnetales). Arnold (1948) separated Gn-etalean members under a separate third class, raised to the level of divisions by Pant (1957) recognizin g Cyacadophyta, Chlamy-dospermophyta and Coniferophyta. Cron-quist, Takhtajan an d Zimmerman (1966) included gymnosperms under Pinophyta, di-vided into three sub divisions: Cycadicae (classes Lyginopteridatae, Cycadatae and Bennettitatae), Pi nicae (classes Ginkgoatfemale gametophyte. Early studies considered Gnetales to be more closely related (even sister) to angiosperms, making gym-nosperms paraphyletic. Recent studies based on molecular evidence, however, point out affinities between Gnetales and conifers. ae and Pinatae) and Gneticae. The monotypic genus Ginkgo (placed in monotypic Ginkgoales, Ginkgoaceae) is uniq ue and apparently unrelated to other gymnosperms, and often termed as living fos sil, retaining several primitive features like dichotomously veined fan-shaped l eaves, motile sperms and lack of pollen tubes. The genus is also unique in havin g sex chromosomes: two X chromosomes in female plants and XY in male plants. Rec ent molecular studies (Qui et al., 2006; Wu et al., 2007) have pointed to close relationship with Cycadales, with which the genus shares features like dioecious habit, branched pollen tube growing away from the ovule, motile male gametes wi th several flaGnetales (sometimes treated under three separate orders: Gnetales, Ephedrales an d Welwitschiales, and then collectively termed as Gnetopsids) are unique among g ymno-sperms in presence of vessels and occur-rence of double fertilization in Ep hedra, wherein one male nucleus fuses with ven-tral canal nucleus producing supe rnumer-ary embryos (and not endosperm of an-giosperms); endosperm, however, rema ining haploid. Ephedra also depicts flower-like re-productive structures and hig hly reduced gella and cell wall. Recent treatments of Gymnosperms following the APG tradition, prefer to treat th e four distinct groups of extant gymnosperms under distinct orders: Cycadales, G inkgoales, Coniferales and Gnetales. The same treatment is followed in the prese nt book. Families of Gymnosperms Division: Pinophyta Order: Cycadales 5. Taxaceae Family: 1. Cycadaceae 2. Zamiaceae Gnetales Ginkgoales 1. Ephedraceae 1. Ginkgoaceae 2. Gnetaceae Coniferales 3. Welwitschiaceae 1. Pinaceae 2. Cupressaceae 3. Podocarpaceae 4. Araucariaceae 386 Plant Systematics

Cycadaceae Pers. Tropics and subtropics of Africa, Southeast Asia, Malaysia, Phillipines and Cycad Family Polynesia. 1 Genus, 20 species Salient features: Palm-like, unbranched stem, leaves fern-like, pinnately compou nd, thick, young pinnae circinate, with mid-vein but no side-veins, plants dioec ious, me-gasporophylls leaf-like with ovules along mar-gin, sperms motile. Description: Palm-like plants with unbranched stem, rarely fern-like with underg round rhizome (C. siamensis); bulbils often developing from axils of leaf bases; plants dioecious. Roots adventitious, large, fleshy; some roots near soil surfa ce become inhabited with nitrogen fixing cyanobacteria, giving coralloid appeara nce (coralloid roots). Stem unbranched, covered with perMajor genera: Single genus Cycas (20 species).

H ''il ~Z. iImmM d $41

A F G E Figure 12.1 Cycadaceae. Cycas revoluta. A: Leaf; E: Megasporophyll. C. circinali s. B: Male cone; C: Microsporophyll; D: Portion of microsporophyll with sori; F: Megasporophyll; G: Longitudinal section of mature seed; H: Pollen grain Families of Gymnosperms 387 sistent leaf bases and scale leaves (cata-phylls) in alternate spiral bands; out line irregular in section, large pith, numerous small vascular bundles in a ring , a wide cortex; parenchyma cells of cortex and pith containing a lot of starch; secondary growth initiated very early in life of plant, but pro-duced in small amount (manoxylic); muci-lage canals abundant in stem; leaf traces forming girdl e in stem. Leaves large, com-pound, forming a crown at the apex of stem, circina te when young, pinnate compound, spirally arranged, alternately with scale leave s; each leaflet with single mid-vein, side veins absent and compensated by trans fusion tissue; leaf petiole with ome-ga-like pattern of vascular bundles. Mi-cro sporophylls aggregated into compact large strobilus (cone), arranged spirally, h ard, wedge-shaped with small sterile pro-jection coiled at tip; microsporangia b

orne on abaxial surface, in groups of 3-5, sur-rounded by hairs; male strobilus terminal in position, becoming lateral due to lateral bud continuing the stem gr owth, axis be-coming pseudopodial; pollen nonsaccate, with a single furrow; sper ms motile. Me-gasporophylls spirally arranged , somewhat leaflike, not forming a strobilus, with 2-8 large ovules (up to 7 cm-largest in plant kingdom) along ma rgin; upper sterile por-tion pinnate (C. revoluta) or reduced and ser-rated (C. circinalis); seeds elliptic or egg shaped, slightly flattened, fleshy, brightly coloured: orange or red. Economic importance: Several species no-tably C. revoluta and C. circinalis are common-ly grown as ornamentals. Sago starch is obtained from stem of C. circinal is and other species. Seeds may also be utilized for yielding starch, but after the removal of toxins. Chamarrow people of Guam who suffer from fatal neurologic al disease apparently consume frugivorous bats known as flying fox-es, who eat C ycas seeds. BMAA (b-methylamino-I-alanine), a neurotoxic non-protein amino acid, present in the seeds of Cycas, is produced by nitrogen fixing cyanobacteria in coralloid roots. Phylogeny: The family was formerly circumscribed to include all genera now remov ed under Zamiaceae. The genus Cycas is distinct from all these genera in circina te young leaves, leaflets with midvein and no side veins, leaf-like megasporophy lls not aggregated into strobilus and marginal ovules. Morphological and molecul ar studies have strong-ly supported Cycas to be sister genus to the rest of cyca ds. Cycas lineage may already have diverged from Zamiaceae by the Permian at lea st 250 million years before present (Hermsen et al. 2006) or as recently as 92 m illion years ago with diversification within the clade occurring ca 36 million y ears ago (Wink 2006). *************** Zamiaceae Horianow Coontie Family Tropical to warm temperate regions of New World, Africa, and Australia. 9 Genera , 111 species Salient features: Fern-like with rhizome or palm-like with unbranched stem, leav es fern-like, pinnately compound, thick, flat or conduplicate but not circinate, veins numerous, plants dioecious, megasporophylls re-duced, forming strobilus, sperms motile. Major Genera: Encephalartos (35 species), Zamia (35), Macrozamia (12), Ceratozam ia (10), Dioon (10) and Bouxenia (3). Description: Fern-like plants with rhizome or Palm-like plants with unbranched s tem; 388 Plant Systematics C Figure 12.2 Zamiaceae. Zamia floridana. A: Plant with female cone; B: Female con e; C: Megasporophyll; D: Male cone; E: Microsporophyll. Macrozamia. F: Microsporophyll; G: Mega sporophyll. plants dioecious. Roots adventitious, large, fleshy; some roots near soil surfac e become inhabited with nitrogen fixing cyanobacte-ria, giving coralloid appeara nce (coralloid roots). Stem unbranched, covered with persistent leaf bases and s cale leaves (cata-phylls) in alternate spiral bands; outline irregular in sectio n, large pith, numerous small vascular bundles in a ring, a wide cortex; parench yma cells of cortex and pith containing a lot of starch; secondary growth initia ted very early in life of plant, but produced in small amount (manoxylic); mu-ci lage canals abundant in stem; leaf traces forming girdle in stem. Leaves large, thick, leathery, compound, forming a crown at the apex of stem, conduplicate or flat, not circi-nate when young (circinate in Bowenia), pinnate compound or bipinnate compound (Bowenia), spirally arranged, alternately with scale leaves; each leaflet with numerous parallel veins (there being no mi

dvein); mid-vein present with dichotomous secondary veins in Stangeria; margin e ntire, dentate or with sharp spines; leaf petiole with omega-like pattern of vas cular bundles; petiole and rachis with or without stout spines. Microsporophylls aggregated into compact strobilus (cone), arranged spirally, hard, wedge-shaped with minute sterile projection; microsporangia numerous, borne on abaxial surfa ce, often clustered; pollen non-saccate, with a single furrow; sperms motile. Me gasporophylls aggregated into strobilus, densely crowded, symmetrical to asymmet rically peltate, valvate or Families of Gymnosperms 389 imbricate, each with two ovules; strobili 1- several per plant, globose, ovoid o r cylindri-cal, disintegrating at maturity; seeds large, 1-2 cm long, rounded in cross section, often brightly coloured with fleshy outer layer and hard inner l ayer; cotyledons 2. Cycadaceae in presence of lateral veins, megasporophylls reduced and aggregated into strobili, absence of circinate vernation, and two reflexed ovules per megas porophyll. Stan-geria is unique in the family with buds arising from roots, abse nce of scales and with presence of midvein and dichotomously branched lateral ve ins. It was removed into distinct family Stangeriaceae by Johnson (1959). Boweni a, similarly has bipinnate leaves, shows circinate vernation in leaflets and rem oved to Boweniaceae by Stevenson (1981), on subsequent morphological evidence in cluded it under Stangeriaceae. Molecular studies, however, include both genera u nder Zamiaceae (Zgurski et al., 2008). The relationships of these genera, and ot her members of the family, however, are not clear. Economic importance: Several species are commonly grown as ornamentals. Starchy underground rhizome of Zamia pumila of trop-ical America were used by early sett lers as flour. Seeds of Dioon edule are also ground into meal and eaten. Several species are also used for the production of sago starch. Removal of toxic glyco sides is however im-perative before consumption. This is easily achieved through boiling. Phylogeny: The family is distinct from *************** Ginkgoaceae Engler Maidenhair Tree Family Limited to remote mountain valleys of China. Possibly extinct in wild, but plant ed in temples and gardens in many parts of the world, especially in temperate cl imate. 1 Genus, 1 species Salient features: Deciduous tree with sim-ple fan-like leaves, veins dichotomous ly branched, dioecious, ovules paired on long stalks, seed juicy with unpleasant smelling outer layer. ranged, turning bright yellow in autumn, widely placed on long shoots, crowded a nd appearing whorled on stout short shoots, latter arising from old long shoots; veins dichotomously branched, without midvein. Plants dioecious. Microsporophyl ls spirally arranged, aggregated in small loose strobili resembling angiosperm c atkins, borne on short shoots; microsporophyll with short stalk and knob-like hu mp, bearing a pair of microsporangia; pollen tube not developed; sperms motile; pollination by wind. Ovules paired, on long stalks, on short shoots, hang-ing li ke cherries, naked; seeds one per stalk (other ovule not maturing), about 2.5 cm in diameter; outer coat fleshy, white-pink, unpleasant smelling (like rancid bu tter or human vomit); cotyledons 2-3. Major genera: Single genus Ginkgo (1 species). Description: Tall deciduous tree reaching 30 m. Roots diarch when young, older t et-rarch or hexarch, containing VAM. Stem branched, crown asymmetric with curved branches attached to stout trunk; bark fur-rowed, grey; secondary growth profus e in long shoots forming broad wood zone (pycnox-ylic), whereas short shoots sho w little of sec-ondary growth (manoxylic). Leaves simple, fan-shaped, bilobed or entire, spirally ar390 Plant Systematics

Figure 12.3 Ginkgoaceae. Ginkgo biloba. A: Shoot with seeds; B: Shoot with male strobilus; C: Shoot with paired ovules; D: Microsporophyll with two microsporangia; E: Pollen grain; F: Pair of ovules; G: Developing seed with aborted ovule near base; H: Lo ngitudinal section of mature seed. Economic importance: The tree has long been grown as ornamental near temples and religious institutions in Eastern Asia. Male plants are commonly grown as they do not pro-duce the unpleasant smell unlike female plants. More recently it has been planted in Canada, USA and Europe. It is relatively dis-ease resistant and tolerates high air pollution. The seeds boiled or fried, are delicacy in some Chinese dishes. Phylogeny: Ginkgo biloba is known in fossil record in Triassic and Jurassic peri ods, having appeared 200 million years ago, and re-productive structures seem to have ch anged little at least for last 120 m years, justifying Families of Gymnosperms 391 its being called as living fossil . It is not closely related to any extant group, but shares motile sperms with cycads. Absence of pollen tube is another primitiv e feature. The short shoots are manoxylic (like Cycadales), whereas the long sho ots are pycnoxylic (like conifers). Ginkgo is unique in having sex chromosomes: XY male and XX female. The leaves re semble Adiantum, the Maidenhair fern, hence the name Maidenhair tree for Ginkgo. *************** Pinaceae Adanson Pine Family Distributed extensively in Northern Hemisphere, from warm temperate to arctic re gions, extending to tree limit in mountains. 10 Genera, 220 species Salient features: Evergreen resinous trees with linear to needle-like leaves, ma le and female cones distinct, seeds several in woody cones hidden by scales, ovu liferous scales imbricate, flat, distinct from bract scales, seeds winged, 2 per scale. woody with persistent spirally arranged ovuliferous scales, each with two ovules , each scale in axil of but free from bract scale; bract longer or shorter than scales; ovules invert-ed, on adaxial surface of scale; archegonia few per ovule, ; seed with long terminal wing derived from scale; embryo straight, cotyledons 2-18, seeds shedding after elongation of cone axis, allowing scales to open; fem ale cones taking 1-2 years to mature. Major Genera: Pinus (100 species), Abies (40), Picea (40), Larix (10), Tsuga (10 ) and Pseudotsuga (5). Description: Trees, rarely shrubs, ever-green, rarely deciduous (Larix, Pseudola rix), with strong smell from bark and leaves. Roots containing ectomycorrhiza. S tem branched, trunk elongate with whorled or op-posite branches, rarely alternat e; resin ca-nals present in wood and leaves, crown py-ramidal or spreading; wood pycnoxylic; p-plas-tids in sieve cells. Leaves simple, linear to needle-like, s piral but often appearing 2- ranked by twisting of leaf base to bring leaves int o one plane, clustered in sheathed fasci-cles on short shoots of Pinus; sessile or short petioled; buds enclosed in bud scales. Plants monoecious. Male cones sm all, microsporo-phylls papery, spirally arranged; microspo-rangia abaxial, two p er microsporophyll; pol-len grains saccate with two saccae (saccae absent in Lar ix, Pseudotsuga). Female cone Economic importance: The family is the leading source of timber in the world. Th e wood of Pinus (Pines), Pseudotsuga (Douglas fir), Picea (Spruce), Abies (Fir), Tsuga (Hem-lock) and Cedrus (Cedars) is extensively used for timber, fence post s, furniture and paper pulp. Many of these are also grown as ornamentals. Seeds of several species of Pinus (pinon pines or pine nuts) particularly P. gerardian a of W. Himalayas (Chilgoza; Neoza) are eaten as nuts. Rosin and turpentine are

extracted from several species of pines. Phylogeny: Pinaceae is a well defined family. Numerous features such as inverted ovule, winged seed, woody cones, p-plastids in sieve cells, simple linear or ac icular leaves and absence of biflavonoid compounds strongly support the monophyl y of 392 Plant Systematics

2T, kka I

D a yr`` ~iJ \~

L M Figure 12.4 Pinaceae. Pinus gerardiana. A: Shoot with leaves in spurs; B: Spur w ith three needles; C: Mature female cone. Pinus wallichiana. D: Male cone E: Microsporophyll with t wo microsporangia; F: Female cone; G: Megasporophyll with two ovules; H: Microsp orophyll in lateral view. Pinus roxburghii. I: Microsporophyll; J: Pollen grain; K: Microsporophyll with two ovules; L: Microsporophyll with two seeds; M: Seed with wing. Cedrus deodara. N: Shoot with leaf clusters and mature female cones; O: Male cone; P: Microsporophyll in lateral view; Q: Microsporophyll in dorsal v iew. Families of Gymnosperms 393 family. The family is sister to rest of the conifers as evidenced by morphologic al (Hart, 1987) and molecular evidence Pinus and Pseudotsuga) and Abietoideae (Abies, Cedrus, Keteleria, Pseudolarix an d Tsuga). The separation is supported by data from chloroplast matK, mitochondri al nad5 and nuclear 4CL genes, although Cedrus is sister to rest of the family. (Quinn et al., 2002). The family is commonly divided into two subfamilies: Pinoideae (Cathaya, Larix, Picea, *************** Cupressaceae Gray Cypress Family Cosmopoliton distribution, mainly in warm and cold temperate climates, more abun

dant in Northern Hemisphere. 30 Genera, 133 species Salient features: Trees or shrubs, leaves scale-like or needle-like, persistent on branches after dying, cone scales usually valvate, pollen grains nonsaccate, ovulifer-ous scale fused with bract scale, seeds with 2-3 lateral wings. attached and flat or peltate, persistent, rarely deciduous (Taxodium), each with 1-20 ovules, inverted, on adaxial surface of scale, erect or inverted; archegon ia clustered; seed with 2-3 short lateral wings derived from seed coat; embryo s traight, cotyledons 2-15; cone maturing in 1-3 years, woody, rarely fleshy and b erry-like (Juniperus). Major Genera: Juniperus (68 species), Cal-litropis (18), Callitris (14), Cupress us (12), Chamaecyparis (7), Thuja (5), Taxodium (3), Sequoia (1) and Sequoiadend ron (1). Economic importance: The family is known for its ornamental shrubs and trees, pa rticularly Species of Cryptomeria (C. japonica), Cupressus, Thuja (T. orientalis -arbor vitae) and Juniperus, the leading source of timber in the world. The wood of Juniperus virginiana (Eastern red cedar) is used for cedar chests, to line c losets, for pencils and for shingles. Oil from the cones of J. communis is used to flavour gin. Juniper pollen is known to cause pulmonary allergies in humans. Description: Trees or shrubs with aromatic wood and foliage. Stem branched, trun k usu-ally short with diffuse branches, bark fibrous, shedding in long strings, buds without bud scales; branches shedding with age. Leaves simple, small, scale -like to needle-like, per-sistent, often closely appressed to branches or spread ing and shedding with them, spiral but often appearing 2-ranked, opposite or who rled, with resin canals; often linear leaves on leading branches and scale-like on later-al branches; scale-like leaves often dimor-phic: those towards base and top of branch flat, middle ones keeled and folded around branch. Plants monoeci ous, rarely dioecious (Juniperus). Male cones small, inconspicu-ous, microsporop hylls spiral or opposite; mi-crosporangia abaxial, 2-10 per microsporo-phyll; po llen grains nonsaccate. Female cones terminal or lateral on short branches; ovul if-erous scale and bract scale fused, basally Phylogeny: The family has often been con-sidered distinct from Taxodiaceae (incl uding Taxodium, Sequoia, Sequoiadendron, etc.) in either opposite and scale-like or whorled and linear leaves as against spiral and linear in latter (although M etasequoia has opposite, and Athrotaxis scale-like leaves). Page (1990) also sug gested fundamental differences in reproductive structures. The two, however, sha re features like fused ovuliferous and bract scale, wings derived from seed coat , more than 2 microsporangia per microsporophyll, 394 Plant Systematics Figure 12.5 Cupressaceae. A: Shoot of Cupressus torulosa with female cones; B: B rach of Juniperus indica with berry-like female cones. Thuja orientalis. C: Curved branch with con e; D: Megasporophyll in abaxial view with two ovules; E: Megasporophyll in later al view; F: Mature male cone; G: Side view of microsporophyll. Sequoiadendron gi ganteum. H: Portion of shoot with female cone; I: Microsporophyll; J: Female con e of Taxodium distichum; K: Megasporophyll of Sequoia sempervirens with several ovules. more than 2 seeds per scale, shedding of small branches, wingless pollen grains and clustered archegonia. Eckenwalder (1976) suggested the merger of two on the basis of phenetic evidence. The merger of two fami-lies is also supported by mol ecular evidence (Quinn et al., 2002; Farjon, 2005). Unified monophyletic Cupress aceae perhaps arose from the paraphyletic assemblage Taxodi which form a basal clade. The genera aceae from Northern Hemisphere and Southern Hemisphere, however, form two distinct cla des, cuppresoid clade and callitroid clade, respectively. The presence of severa l teeth on ovuliferous scales of Cryptomeria is perhaps reversal to plesiomorphi c condition. *************** Families of Gymnosperms 395

Podocarpaceae Endlicher Tropical and subtropical regions of Southern Hemisphere, extending northwards to Japan, Central America and West Indies. 17 Genera, 170 species Podocarp Family Salient features: Shrubs or trees, leaves linear or broader, persistent, microsp orophyll with two microsporangia, mature ovulifer-ous scale with one ovule, seed s surrounded by specialized scale called epimatium, bracts juicy, pollen saccate . Major Genera: Podocarpus (100 species), Dacrydium (20), Dacrycarpus (9), Afrocar pus (6) and Phyllocladus (5). Description: Shrubs or trees, rarely parasitic (Parasitaxus ustus parasitic on r oots of E G Figure 12.6 Podocarpaceae. Podocarpus spicatus.A: Vegetative shoot; B: Brach wit h male cones; C: Single cone showing arrangement of microsporophylls; D: Microsporophyll; E: P ollen grain; F: Vertical section of ovule; G: Mature seed. Phyllocladus alpinus. H: Branch with flattened phylloclads; I: Portion of branch with female cones. 396 Plant Systematics Falcatifolium taxoides, another podocarp), slightly resinous. Stem branched, tru nk short or long. Leaves simple, entire, thick, persistent, alternate, rarely op posite (Micro-cachrys), variable, scale-like to broadly lin-ear (sometimes upto 30 cm long and 5 cm broad). Plants monoecious, rarely dioecious. Male cones smal l, cylindical; microsporo-phylls numerous, spirally arranged, each with 2 micros porangia, pollen grains with 2 saccae. Female cones with 1-several ovu-liferous scales; each scale with single ovule, modified into juicy structure called epime -tium, thus cone appearing like a drupe; seed with 2 cotyledons. Phylogeny: The family has been considered closer to Taxaceae with which it share s features of resinous plants, ovuliferous scale with solitary ovule and cone mo re or less fleshy. In two families, however, the nature of fleshy structure is d ifferent. It represents ovuliferous scale in Podocarpaceae, whereas it represent s an aril, an outgrowth from the base of ovule, in Taxaceae (Quinn et al., 2002) . Genus Phyllocladus with branches flattened into phylloclads looking like leave s (leaves re-duced to scales) and with aril is often separated into distinct fam ily Phyllocladaceae. Recent rbcL studies, however, point to their being sister g roups (Quinn et al., 2002), or Phyllocladus embedded in Podocarpaceae (Wagstaff, 2004), latter conclusion also reached based on nuclear gene XDH (Peery et al., 2008). Economic importance: Species of Podocar-pus and Dacrydium are valuable sources o f timber.. Podocarpus macrophyllus is widely planted as ornamental. *************** Araucariaceae Henckel & W. Hochst Family Monkey-puzzle tree Nearly restricted to Southern Hemisphere mainly in S E Asia to Australia, New Ze aland, and S. America. 3 Genera, 32 species Salient features: Large trees with naked buds, highly resinous, leaves needle-li ke to lanceolate, persistent, shedding along with small branches, pollen grains nonsaccate, exine pitted, female cones woody, ovulifer-ous scale with single ovu le. to elliptic, persistent, shedding on small branches, spiral (Araucaria) or oppos ite (Agathis). Plants monoecious (Agathis) or dioecious (Araucaria). Male cones small, cylindrical; microsporophylls numerous, spirally arranged, each with 4-20 microsporangia, pollen grains nonsaccate, exine pitted; sperms nonmotile. Femal e cones solitary, more or less erect, maturing in 2-3 years, disintegrating on t ree, ovuliferous scales numerous, spirally arranged, flattened, linear to peltat e, longer than and fused with bract scale, each scale with single ovule; ovule f

ree from scale (Araucaria) or fused with it (Agathis); seed large, with (Agathis ) or without (Araucaria) marginal wings, with 2 cotyledons sometimes deeply divi ded and appearing 4. Major Genera: Araucaria (18 species), Agathis (13) and Wollemia (1). Description: Tall long-lived trees reaching up to 65 m in height and up to 6 m i n diam-eter, highly resinous, usually symmetrical with conical crown, buds naked . Stem branched, trunk stout and thick, small branches shedding along with leave s. Leaves simple, entire, varying in shape from awl-shaped, scale-like, linear, oblong Families of Gymnosperms 397 D G Il E 1

1 i

L Figure 12.7 Araucariaceae. Agathis australis. A: Shoot with female cone; B: Mega sporophyll; C: Shoot of Agathis alba with male cones. Araucaria angustifolia. D-E: Microspor ophylls in different views; F: Pollen grain Economic importance: Species of Araucar-ia mainly A. araucana (Monkey-puzzle tre e) with spectacular whorled branches and A. heterophylla (Norfolk Island pine) o f Chile are prized ornamentals grown as avenue trees and house plants. Many spec ies mainly Ag-athis australis (Kauri) with massive trees are utilized for timber . broader leaves, monoecious habit, ovule free from scale, winged seeds and latter with spiral linear leaves, dioecious habit, ovule fused with scale, and wingles s seeds. Mono-phyly of two genera is supported by rbcL sequence data. Phylogeny: The family is distinct, restrictWollemia nobilis, discovered only in 1994 from Wollemi National Park in Australi a was earlier known only from fossil record extending to 150 MYA. The species is represented by less than 50 trees and has unique dark ed to the Southern Hemisphere, fossil record of Araucaria extending to Jurassic. brown and knobbly bark described as bu bBoth genera are, Agathis and Araucaria bling chocolate . It is also multi-trunked are well separated, former with opposite appearing as clumps of trunks. *************** 398 Plant Systematics Taxaceae Bercht. & J. Presl. Largely distributed in Northern Hemisphere, extending south to Guatemala, Java a nd Caledonia. 6 Genera, 30 species Yew Family Salient features: Shrubs to small trees, wood without resin canals, leaves linea r, flat-tened, persistent, pollen grains nonsaccate, ovules solitary, not in con es, with fleshy aril at base surrounding the ovule fully at matu-rity and becomi ng bright red. Description: Shrubs or moderately sized trees, not resinous or slightly resinous , fragrant or not. Stem much branched, wood without resin canals. Leaves simple, entire, linear, flattened, with abruptly tapering apex, persistent for several years, shedding singly, spiral, often twisted at base to appear 2-ranked, leaf b ase with decurrent petiole. Plants dioecious or monoecious. Male cones with 6-14 microsporophylls, each with Genera: Taxus (10 species), Cephalotaxus (8), Torreya (5), Amenotaxus (5), Pseud

otaxus (1) and Austrotaxus (1). Figure 12.8 Taxaceae. Taxus baccata. A: Shoot with seeds; B: Seed surrounded by aril; C: Portion of branch with male cone; D: Male cone; E: Peltate microsporophyll with microspo rangia; F: Pollen grain Families of Gymnosperms 399 2-9 microsporangia radially arranged or on abaxial surface; pollen grains nonsac cate, sperms nonmotile. Ovules solitary, not formremoved to a distict order Taxales. The family, however, shares similar embryolo gy, wood anatomy, leaf and pollen morphology with rest of the conifers. The plac ement within Coniferales is supported by evidence from DNA studies (Chase et al. , 1993; Price, 2003) and micromorphology (Anderson and Owens, 2003). Two distinc t clades are established within the family: one including Tax-us, Austrotaxus an d Pseudotaxus (aril partly enclosing ovule, maturing in 6-8 months and mature se ed 5-8 mm long) and the other including Torreya, Cephalotaxus and Amenotaxus (ar il completely enclosing ovule, maturing in 18-20 months and mature seed 12 40 mm long). The two are sometimes placed in distinct families Taxaceae s. s. and Cep halotaxaceae, respectively. There is, however, strong molecular evidence to supp ort their merger into broadly circumscribed Taxaceae (Price, 2003; Rai et al., 2 008), as recognition of Cephalotaxaceae renders it para- or polyphyletic. ing cones, surrounded by fleshy aril, arising from base, aril brightly coloured at maturity; seed with hard seed coat, cotyledons 2. Economic importance: Various species of Taxus, especially T. baccata (English ye w) and T. cuspidata (Oriental yew) are commonly planted as ornamentals. Wood of yew family has been popular since Middle Ages for mak-ing bows, owing to the pre sence of extra spi-ral thickenings on the xylem cells. Wood of Taxus is used in high grade furniture. The presence of taxol, a highly toxic alkaloid hav-ing ant imitotic activity makes it potential agent for anticancer chemotherapeutic treat ment. Phylogeny: The family is unique in coni-fers in the absence of female cone, and in having solitary ovules, and has often been *************** Ephedraceae Dumort Joint Fir Family Worldwide in temperate regions except Australia, adapted to extremely arid condi tions, extending up to 4000 m in Andes and Himalayas. 1 Genus, 60 species Salient features: shrubs with jointed stems with clustered or whorled branches, vessels present, leaves scale-like, fused into sheath, microsporophylls stalked, pollen furrowed, nonsaccate, each ovule surrounded by pair of fused bracts, see d covered. time, but gradually replaced by adventitious roots. Stem much branched, aerial s tems arising from spreading rhizome; branches numerous, whorled or clustered, lo ngitudinally grooved, distinctly jointed with long internodes, usually green and photosynthetic, horsetail like, wood with vessels. Leaves scale-like, opposite or in whorls of 3-4 leaves, fused at base into a sheath at each node, often shed ding early, each leaf with two par-allel veins; with axillary buds; resin canals absent. Plants usually dioecious, rarely monoecious. Male strobili (inflorescen ce) in whorls of 1-10 at nodes in axil of scale leaf, Genus: Ephedra (60 species). Description: Small shrubs, often trailing, rarely climbing (E. foliata), very ra rely al-most tree-like reaching 30 cm in diameter and height of several metres ( E. triandra). Root a tap root in seedling, persisting for long 400 Plant Systematics r

.. ~..., ., ' `?f1 B

Figure 12.9 Ephedraceae. Ephedra gerardiana. A: Shoot with strobili; B: Male str obilus; C: Pollen grain; D: Surface of pollen grain with parallel ridges on exine; E: Female strob ilus with bracts and two ovules; F: Same with seeds. each with 2-8 successive pairs of cupped bracts, lower one or two pairs of bract s ster-ile rest bearing solitary microsporangiate shoots (microsporophylls, flow ers), each on short secondary axis (microsporangiophore) arising between each pa ir of fertile bracts and bearing two opposite scales (perianth) and into 2-10 mi crosporangia (stamens with filamented or sessile anthers), dehiscence by termina l pores; pollen furrowed, nonsac-cate, inaperturate, exine shed on germinaaround the ovule borne on a stalk (stalk and ovule constitute female flower); ov ule single or in pairs, with 2 integuments; seeds 1 2 per strobilus, surrounded by leathery yellow to dark brown cup; cotyledons 2. tion, pollen becoming naked. Female stroEconomic importance: Several species of Ephedra were used as beverage by early M ormon settlers, hence the name Mormon tea for the genus. The alkaloid drug ephed rine used as decongestant, treatment for cough and circulatory problems is obtai ned bili opposite or in whorls of 3-4 at branch from several species especially E. s inica (ma nodes, each with 2-10 successive pairs of huang), which has been used in China ear-bracts, uppermost fused to form a fleshy cup lier than 2500 B.C. Families of Gymnosperms 401 Phylogeny: The genus Ephedra shows superficial resemblance to Equisetum and Casu arina, all three exhibiting switch habit , owing to sheathed nodes and scale leaves . The presence of vessels, perianth-like bracts, extremely reduced gametophytes and the fusion of second male gamete with venter canal nucleus have been taken a s angiosperm affinities, although the oriare sister to a clade including all other seed plants based on studies of rbcL ( Seider et al., 2002) and nuclear genes (Rydin et al., 2002), rendering gymnosper ms as paraphyletic. The bulk of evidence in recent years, however, points to gym nosperms being monophyletic and Gnetales sister to Pinales. The binucleate sperm cells, basic proembryo structure, development of poly-embryony, etc., of Ephedr a agree with Pinales in general and perhaps Pinaceae in particular. gin of vessels in two has been separate and several primitive angiosperms are ve sselless. It has been suggested that Gnetales *************** Gnetaceae Lindley Gnetum Family Distributed in tropical regions of W. Africa, Brazil, India and Southeast Asia. 1 Genus, 30 species Salient features: Evergreen lianas, trees or shrubs, with vessels, leaves opposi te, angiosperm-like, net-veined, microsporo-phylls stalked, with bracts, pollen spinose, seeds large, fleshy.

each pair fused to form a cup known as cupule or collar; each bearing several mi crosporangiophores (microsporophylls, flowers) in 3-6 rings, with one ring of ab ortive female flowers or ovules above rings of male flowers; each microsporangio phore with two bracts (perianth) at base and two microsporangia (anthers) at top (one microsporangium in G. gnemonoides); perianth containing sclereids and late x tubes; pollen not striate, surface spinose. Female strobili with a pair of opp osite sheathing bracts at base, followed by 5-6 whorls of ovules; each whorl wit h 4-10 ovules in a single ring above the collar, each ovule with two bracts form -ing perianth which persists in seed, a terminal ovule sometimes present on the strobilus; only few ovules maturing into seeds, others aborting and falling down ; uniseriate hairs interspersed among the ovules; ovule with 2 integuments, subt ended by perianth forming third outer envelope; perianth and outer integument wi th sclereids and laticiferous ducts; seed large, fleshy. Genus: Gnetum (30 species). Description: Usually liana, sometimes scadent shrubs (G. contractum), rarely tre es (G. gnemon). Stem climbing often reaching top of tall trees, rarely erect, so metimes with two types of shoots: long shoots and short shoots, rarely all shoot s similar (G. gnemon), vessels present, phloem with companion cells, arising fro m cambium cells and not from mother-cell as in angiosperms. Leaves large, entire , oblong, elliptic or lanceolate, subsessile or short petiolate, opposite and de cussate, net-veined, appear-ing like a dicot leaf, appearing only on short shoot s in lianas, often reduced to scales on long shoots, stipules absent. Plants dio e-cious, rarely monoecious. Male strobili arising in axils of paired and decussa te scale leaves, slender, elongate, with several (10-25) pairs of decussate brac ts, bracts of Economic importance: The tree species G. gnemon from Malaya, is widely cultivate d 402 Plant Systematics

E F Figure 12.10 Gnetaceae. A: Shoot of Gnetum latifolium with opposite leaves and m ale strobili; B: Portion of branch with female strobili; C: Portion of female strobilus showin g ovules; D: Portion of female strobilus with two mature seeds; E: Longitudinal section of seed; F: Portion of male strobilus of G. ulva; G: Single microsporoph yll of same; H: Microsporophyll of G. gnemon; I: Pollen grain. as ornamental, and as food, with leaves and strobili cooked in coconut milk, and fibre for making ropes and fishing nets. The kernels of G. ulva yield an oil us ed for illumination and also as a massage in rheumatism. The plant of G. montanu m is used as fish poison. reduced gametophytes and presence of perianth like bracts. Gnetum appears more l ike angiosperms with net-veined leaves and presence of companion cells, although their origin is quite different. The reaction wood in Gnetum consists of gelati nous extraxylary (reaction) fibers in the adaxial position i.e., it is unique amon g seed plants (Tomlinson, 2003) and is unlike tension wood of angiosperms. Phplogenp: The genus Gnetum along with other members of Gnetales share an-giospe rm features of presence of vessels, *************** Families of Gymnosperms 403

A A . F9

A ~

~ti1 !:Yid ll

Figure 12.11 Selaginellaceae. A: Selaginella martensii. Equisetaceae. B: Equiset um telemateia var. braunii, vegetative shoots; C: Portion of stem enlarged. Osmundaceae. D: Osmunda regalis, portion of plant be enlarged; F: Todea barbara, plant with sterile and fertile fronds. Blechnaceae. sterile and fertile fronds; E: Portion of fertile frond Woodwardia semicordata. G: portion of lea all portion enlarged. 404 Plant Systematics

MR

'M

+Ie]

- +yy. T '~L.

` RSA

rr ~

s.

Figure 12.12 Cycadaceae. Cycas circinalis; A: Plant; B: Young circinate leaves; C: Portion of mature leaf of C. revoluta; Zamiaceae. D: : Macrozamia communis, apex with femal e cones; E: Plant with leaves and female cones; F: Plant of Zamia integrifolia; G: Encephalartos trispinosus, plant; H: Dioon edule, plant. Families of Gymnosperms 405

11

VE y~JS RI (as f r Lam. na

Figure 12.13 Ginkgoaceae. Ginkgo biloba. A: Branch with leaves; B: Portion enlar ged. Pinaceae. C: Cedrus deodara, branch with leaves and female cones; D: Portion of bark; E: Tsuga canadensis, portion of pla ecyparis lawsoniana, portion of twig with female cones; G: Sequoia sempervirens, portion of branch enlarged; H: Cryptomeriajaponica, portion of branch with fema le cones. emale cones. Cupressaceae. F: Chama406 Plant Systematics

r. n ]l

Figure 12.14 Podocarpaceae. A: Podocarpus totara; B: Podocarpus gracilior, branc h enlarged; C: Dacrydium cupressinum; D: Twig of same showing leaves; E: Twig of Phyllocladu s trichmanoides. Taxaceae. F: Taxus baccata, twig.G: Torreya californica, twig w ith seed enclosed in aril. Ephedraceae. H: Ephedra americana., plant with strobi li. Chapter 13 Major Families of Angiosperms The treatment of major families of he had preferred to retain primitive woody An giosperms presented in the following pages families Winteraceae, Chloranthaceae, is an attempted integration of the principal Magnoliaceae, Degeneriaceae, Annon aceae, systems of classification. The view that the Calycanthaceae, and Lauracea e before the division of Angiosperms into monocots and paleoherb taxa, followed by monocots. The dicots, renders the latter as paraphyletic, has latest version, however, has Amborellaceae, firmly been established. Views are further Chlorant haceae, Trimeniaceae and consolidating to interpolate monocot taxa Austrobaileya ceae placed before Winteraceae between primitive dicots and the more and other M agnoliids. It is only for the sake advanced ones, because it had long been of co nvenience of reference, circumscription suggested that monocots arose from primi tive of families that would have a direct bearing dicotyledons. The strong workf orce of on the number of genera and species within Angiosperm Phylogeny Group is attempting the family, that the arrangement of the to establish monophyletic gr oups, more families according to his system of consistently up to the family lev el. The classification has been followed. The number success above the family le vel is much short of genera and species is based on his 2007 of any level of con sistency for any meaningful revision. Nevertheless, the placement of the applica tion. As many as 30 families still have family visa-vis all recent major systems of to find their place. The artificial clades above classification is compared

and latest the order level are too arbitrary, and it may phylogenetic position l argely based on recent take a long time before the system may have developments in serotaxonomy and a practical approach for the purpose for which molecular sys tematics discussed. It should the classifications are meant. From the lot be bor ne in mind that the same group names of contemporary authors Thorne, with his ar e not always comparable. Thus whereas latest version (2007) his classification cla ss names Magnoliopsida and Liliopsida having undergone periodic electronic refer to dicots and monocots in the revisions has attempted to upgrade his classificati on systems of Cronquist and classification in light of recent developments Takht ajan, the term Magnoliopsida was used in the field of molecular systematics, but for Angiosperms (for which Takhtajan and retaining the hierarchical grouping, s o essential for any working system of classification. In his earlier version of 1999, Cronquist had used the division name Magnoliophyta) by both Dahlgren and Thorne. Both of them had used the name Magnoliidae 408 Plant Systematics

lit

~k L

/ r

m.`

J v 3

b J

I ~ rrrl S

Angiosperms roll of honour Top left: Victoria amazonica (Nymphaeaceae), having large floating leaves often reaching 2 m in diameter and can often support the weight of a child. Top right: Woo arrhiza (Lemnaceae), the smallest known angiosperm, barely about 1 mm in si ze, visible as scum on the surface of water. Above left: Eucalyptus regwns (Myrt aceae), the tallest tree, with recorded height of 97 m and girth of 7.5 m. Above right: Rafflesia arnoldii (Rafflesiaceae), a bizarre plant with plant body no more than a fungus mycelium, yet producing largest sized flower sometimes reaching 1 min diameter. Left: Arabit4sis thaliana (Brassicaceae), the guinea pig of plant kingdom, with most completely known genome. Major Families of Angiosperms 409 for dicots and Liliidae for monocots. However, Thorne has lately abandoned this distinction into traditional dicots and monocots, and instead used subclass name s more or less comparable to Takhtajan and Cronquist (but without the super grou ps dicots and monocots). He has also taken the bold decision of separating the p rimitive dicots from the more advanced ones by inserting monocots in between, th us bringing the classification system much nearer to the Angiosperm Phylogeny Gr oup, but retaining the essence of hierarchical arrangement through consistent us e of superorders and subclasses above the level of order. Thorne from 2003 onwar ds has also assigned the degree of confidence in hierarchical level, circumscrip tion and alignment of taxa. A represents limited confidence, B for probably correct assignment and C implies c onsiderable confidence in assignment. Only those with A or B level are indicated here. Rest belong to level C. It is not under the scope of the present book to include all the families of ang iosperms, but in addition to the major families, those which have been the subje ct of considerable phylogenetic interest over the recent years have been especia lly chosen for treatment. Angiosperms placed under Class Magnoliopsida are divid ed into 12 subclasses, 36 superorders, 85 orders and 485 families. Angiosperms a re estimated to include13372 genera and 253300 species (10760 genera and 196990 species of dicots; 2612 genera and 56310 species of monocots). Arrangement after Thorne (2007) Subclass 1.Chloranthidae (B) Superorder 1. Chloranthanae Suborder 2. Illiciineae Order 1.Chloranthales 1. Illiciaceae 2. Schisandraceae Suborder 1. Chloranthineae (B) Order 1. Nymphaeales Family 1. Amborellaceae# 1. Cabombaceae

2. Nymphaeaceae 3. Ceratophyllaceae 2. Chloranthaceae 3. Trimeniaceae 4. Austrobaileyaceae # Families in boldface are described in detail. Amborellaceaee Pichon 1 genus, with a single species Amborella trichopoda Endemic to the Island of New Caledonia in South Pacific Ocean. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Magnoliidae Magnoliidae Magnoliidae Chloranthidae Series+/Superorder Microembyeae+ Lauranae Magnolianae Chloranthanae Order Laurales Laurales Laurales Chloranthales Unplaced/ (Amborellales) B & H under family Monimiaceae. B & H- Bentham and Hooker 1862-1883; Cronquist, 1988; Takhtajan, 1997; Dahlgren- G. Dahlgren, 1989; Thorne, 2007; APG II, 2003; APweb, 2008 (Stevens) 410 Plant Systematics A B e

~A BARBIA ~ F~P IT 4

JJ {{i H

0 n Figure 13.1 Amborellaceae. Amborella trichopoda. A: A specimen growing at Arbore tum of University of California, Santa Cruz (Manager of Arboretum Brett Hall is squatting bes ides the tree) (Photo Tim Stephens); B: Fully opened female flower. C: Close up of a branch; D: Male flower. (photos A and C: courtesy University of California, Santa Cruz; B: photo courtesy Missouri Botanical Garden). Salient features: Shrubs lacking vessels, with simple alternate leaves, stipules lack-ing, nodes unilacunar, flowers unisexual, with multiseriate perianth, stam ens nu-merous, pollen with granulate ektexine, carpel incompletely closed, fruit aggregate of drupes. primary medullary rays narrow, sieve-tube plastids S-type. Leaves evergreen, alt ernate, spiral to two-ranked, simple, entire to pinnately lobed, venation pinnat e, stomata anomocytic, stipules absent, mesophyll with-out ethereal oils. Inflor escence cymose, plants dioecious. Flowers small, unisexual, hypanthium present. Perianth with 5-13 tepals, number more in staminate flowers than pistillate flow ers), slightly united at base, spirally arranged, not differentiated into sepals and petals. Androecium with 12 22(-100) stamens, free, in 3-5 whorls, outer Major genera: Only genus Amborella with 1 species. Description: Sprawling shrub; wood with tracheids but no vessels, nodes unilacun ar, Major Families of Angiosperms 411 whorl adnate to tepals at base, anthers adnate, dehiscence longitudinal, intrors e, microsporogenesis successive, pollen inaperturate to ulcerate, ektexine granu -late, pistillate flower with 1-2 staminodes. Gynoecium with 5-8 carpels in a si ngle whorl, free, ovary stalked, superior, carpel margins incompletely closed (u nsealed at tip), stigma sessile with two expanded flanges, ovule 1, placentation marginal, ovule pendulous, hemianatropous, sessile. Fruit an aggregate of drupe s with pock-marked stones, and pockets of resinous sub-stances, seeds endospermi c, embryo minute, cotyledons 2. an unknown common ancestor of all angiosperms. The family has traditionally been placed under Laurales (Cronquist, Dahlgren, Takhtajan). The multigene analyses (Qui et al., 1999; Soltis et al., 1999; Zanis et al., 2002), support this family as sister to all extant angiosperms, with Nymphaeaceae as subsequent sister to the rest of angiosperms, which may ultimately result in placing these families i n separate orders or in a common order. APG II, for the present regards the fami ly unplaced at the beginning of angiosperms. APweb (2008) places this family und er monotypic order Amborellales at the beginning of Angiosperms. Thorne had earl ier (1999, 2000) placed Winteraceae at the beginning of Angiosperms (and the Mag noliidae), and Amborellaceae in the third suborder, shifted (2003) Amborellaceae , like APG schemes, to the beginning of Magnoliidae under order Chloranthales. H e has further (2006, 2007) removed the order together with Nymphaeales under a d istinct subclass Chloranthidae, superorder Chloranthanae. Economic importance: No economic value known. Phylogeny: The family is unique in angiosperms in having granular ektexine, lack ing tectum. This, combined with the absence of vessels, partially closed carpels , places this family on the lowest branch of angiosperm family tree, having evol ved from * * * * * * * * * * * Chloranthaceae R. Br. Ex Lindl. Chloranthus family 4 genera, 75 species Tropical, subtropical and South Temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Magnoliidae Magnoliidae Magnoliidae Chloranthidae Series+/Superorder Microembyeae+ Magnolianae Magnolianae Chloranthanae Chloranthales Chloranthales Chloranthales Unplaced/(Chloranthales) Order Piperales Salient features: Leaves aromatic, opposite, simple with connate petiole bases, stipules small, flowers small, lacking perianth, stamens 1-3, connate in a mass, carpel 1, ovary inferior with a single ovule, fruit a small drupe. Major genera: Hedyosmum (25 species), Chloranthus (12), Ascarina (3) and Sarcand ra (3). Description: Herbs or evergreen shrubs or trees containing essential oils. Vesse ls 412 Plant Systematics

~D

y 1~ F Figure 13.2 Chloranthaceae. A: Flowering branch of Chloranthus inconspicuous; B: Bisexual flower of C. brachystachys; C: Bisexual flower of C. henryi with single bract, trimerou s stamen having bithecous (tetrasporangiate) middle anther and monothecous (bisp orangiate) lateral anthers, and a single pistil with tufted stigma. Sarcandra. D : Transverse section of fruit of S. chloranthoides; E: Seed; F: Bisexual flower of S. glabra. Ascarina lanceolata. G: Flowering branch; H: Male flower and I: Fr uit. absent in Sarcandra (vessels reported in roots, but not in stem), present in oth ers but with long tapering elements, with scalariform per-foration plates, nodes unilacunar or trilacunar, sieve-tube plastids S-type. Leaves aromatic, opposite , simple, usually serrate along margin, petioles often connate at base, stipules

small, interpetiolar, leaf mesophyll with spherical etherial oil cells. Inflore scence a spike, panicle or capitate, ultimate inflo-rescence units cymose. Flowe rs usually uni-sexual or pseudobisexual due to coherence of male and female flow ers, unisexual in Ascarina and Hedyosmum but bisexual in Chloranthus and Sarcand ra, flowers minute, bracteate, actinomorphic. Perianth absent in male flowers, r udimentary and calyx-like and adnate to ovary in female flowers, latter some-tim es completely naked (Ascarina) or enclosed by a cupular bract (Hedyosmum). Andro ecium with a single (Sarcandra) or three stamens connate into single mass with o ften bithecous middle anther and monothecous lateral anthers (Chloranthus), dehiscence longitudinal. Gynoecium with 1 carpel, ovary inferior, unilocular, ovule 1, orthotropous, pendulous, bitegmic, crassin ucellate, placentation apical, style very short or absent. Fruit an ovoid or glo bose drupe, seed with abundant oily endosperm and minute embryo, perisperm prese nt. Economic importance: Chloranthus glaberi is grown as an ornamental shrub. The le aves of C. officinalis are used to make a drink in parts of Malaya and Indonesia . The infusion of flowers and leaves of C. inconspicuous are used to treat cough s and the flowers used to flavour tea in various parts of East Asia. Ex-tract fr om the leaves of Hedyosmum brasiliense is used locally in tropical South America as tonic, to induce sweating and also to treat stomach complaints. Phylogeny: The family is traditionally placed in the magnoloid complex under ord er Major Families of Angiosperms 413 Piperales (Cronquist), Chloranthales (Takhtajan, Dahlgren, Thorne). Donoghue and Doyle (1989) placed Chloranthaceae under Laurales, but this position is not sup -ported by DNA-based cladistic analyses. Taylor and Hickey (1996) consider Chlor anthaceae as the basic angiosperm family. The family shows several plesiomorphic characters such as flowers in an inflorescence, plants dioecious, carpels solit ary, placentation apical, and fruit drupaceous with small seeds. The family is t he oldest in the fossil record, the fossil ge-nus Clavitopollenites being assign ed to Chloranthaceae and closer to the genus Ascarina. The stems of Sarcandra ar e primi-tively vessel-less. In other genera vessels are primitive with long vess el-elements, ta-pered and with many barred perforation plates. The family is con sidered to be earli-est to record wind pollination in angiosperms. Taylor and Hi ckey believe in the origin of Chloranthaceae from gnetopsids, hypoth-esizing tha t the ovule and the bract subtend-ing the floral unit in Chloranthaceae are homo logous with one of the terminal ovules and proanthophyll subtending the anthion (inflorescences unit) of gnetopsids. Chloranthaceae has undergone considerable r eduction in its number of parts as well as general level of elaborateness. Thorn e (1996) considered Trimeniaceae to be the closest relative of Chloranthaceae. The position of this family in APG system is uncertain. The family is sister to magnoliids + eudicots in the six-gene compartmentalized analysis with 84 per cen t bootstrap support (Zanis et al., 2003), but APG II prefers to keep the family unplaced at the beginning of angiosperms, without assigning it to any order. APw eb places it under Chloranthales, before Magnoliids, but after Commelinids. Thor ne had earlier (1999) placed family under Magnoliales under sub-order Chloranthi neae after Winterineae and Illicineae, but in subsequent revision (2003) placed it (along with Amborellaceae, Trimeniaceae and Austrobaileyaceae) under distinct order Chloranthales at the beginning of Magnoliidae (first subclass of angiospe rms), finally removed in 2006, 2007 together with Nymphaeales under a distinct s ubclass Chloranthidae, superorder Chloranthanae. * * * * * * * * * * * Austrobaileyaceae (Croizat) Croizat 1 genus, single species Austrobaileya scandens Native to Queensland, Australia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliidae Class Magnoliopsida Magnolipsida Magnoliidae Chloranthidae Magnolianae Chloranthanae Subclass Magnoliidae Series+/Superorder Magnolianae Order Not described then Magnoliales Austrobaileyales Annonales Chloranthales Austrobaileyales Salient features: Climbing shrubs with stamens numerous, laminar, inner modiMopp osite leaves, flowers solitary in leaf fied into staminodes, carpels several, fr ee, axils, bisexual with numerous tepals gradu- partially unsealed with bilobed style. ally intergrading from sepals to petals, 414 Plant Systematics uo 'bo 0

B Figure 13.3 Austrobaileyaceae. Austrobaileya scandens. A: Flowering branch; B: S tamen, broad and petal-like; C: Carpel with bifid style. Major genera: Single genus with one spe-cies. Originally, two species Austrobail eya maculata and A. scandens were described, but they have now been combined int o a single species under the latter name. gitudinal, introrse, pollen grains monosulcate. Gynoecium with (6 )9( 12) carpels, f ree, spirally arranged, ovary superior, 8 14 ovuled, placentation marginal (biseri ate), ovules collateral, anatropous, bitegmic, crassinucellate, style partially unsealed, bilobed. Fruit baccate, seeds with ruminate endosperm. Pollination by insects. Description: Large climbing shrubs bearing essential oils, nodes unilacunar, wit h two traces, vessel end-walls scalariform, sieve-tube plastids S-type. Leaves e vergreen, op-posite to sub-opposite, leathery, petiolate, simple, entire, pinnat ely veined, stipules intrapetiolar, caducous, small, mesophyll with spherical et herial oil cells. Inflores-cence with solitary axillary flowers. Flowers bisexua l, bracteate, pedicellate, bracteolate. Perianth with tepals nearly sequentially intergrading from sepals to petals, (9 ) 12( 14), free, imbricate. Androecium with 12 25 stamens, maturing centripetally, free, outer laminar, petaloid, fertile, inn er gradu-ally smaller, innermost reduced to staminodes, anthers adnate, dehiscen ce lon-

Economic importance: No economic value known. Phylogeny: When first described by C. T. White (1933), it was considered to be b elonging to Magnoliales, a Placement also followed by Cronquist, but due to uniq ue combination of characters, and for want of a better place, Hutchinson (1973) placed the family in Laurales, near Monimiaceae. According to Thorne (1996), the family Austrobaileyaceae is so intermediate between Magnoliales and Laurales, t hat Laurales should not be recognized as a separate order. He, Major Families of Angiosperms 415 accordingly, placed Magnoliaceae (and re-lated families), and Lauraceae (and rel ated families) under separate suborders Magnoliineae, and Laurineae, respectivel y. He later (1999, 2000) placed this family closer to Monimiaceae and Chlorantha ceae under suborder Chloranthineae, before suborder Magnoliineae under Magnolial es. In his sub-sequent revision (2003), however, he recogChloranthanae, order Chloranthales, suborder Chloranthineae. Dahlgren (1989) pla ced it under the first order Annonales of Magnolianae. Takhtajan removed it to o r-der Austrobaileyales within Magnolianae. APG II and APweb (2008) place this fa mily together with Trimeniaceae, Schisandraceae and Illiciaceae under the order Austrobaileyales, but with no informal higher rank, placed towards the beginning nized the suborders as independent orders, Chloranthales (upgraded Chloranthineae) of angiosperms. The combination of these being placed towards the beginning of families under single order received 99 p er Magnoliidae. In latest revision (2006, 2007), cent bootstrap support in sever al multigene however he has shifted this family under a analyses (Soltis, Soltis , and Chase, 1999; distinct Subclass Chloranthidae, superorder Soltis et al., 20 00). * * * * * * * * * * * Illiciaceae (DC) A. C. Smith Star Anis family 1 genus, 42 species Southeast United States, West Indies, Mexico, China, Japan and Southeast Asia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Magnoliidae Magnoliidae Chloranthidae Series+/Superorder Thalamiflorae+ Magnolianae Magnolianae Chloranthanae Order Ranales Illiciales

Illiciales Illiciales Chloranthales Austrobaileyales B & H under family Magnoliaceae Salient features: Trees and shrubs with simple alternate leaves, stipules lackin g, nodes unilacunar, flowers solitary with multiseriate perianth, stamens numero us, carpels free, in a single whorl, fruit a star-like aggregate of follicles. sclereids. Leaves evergreen, alternate, often clustered at tips of branches, som e-times subverticillate, entire, gland-dotted, containing terpenoids, simple wit h reticulate venation, stipules absent. Inflorescence with solitary flowers, axi llary or supra-axillary, rarely 2-3 together. Flowers usually bisexual, actinomo rphic, hypogynous. Perianth with numerous tepals, in several whorls, sepals and petals not differentiated, outermost somewhat sepal-like, inner gradually becomi ng smaller and petal-like. Major genera: Only genus Illicium (42 species). Description: Shrubs or small trees, contain-ing aromatic terpenoids and branched 416 Plant Systematics r x

E F Figure 13.4 Illiciaceae. Illicium floridanum. A: Fruiting branch; B: Flower; C: Stamens with broad filaments, left with almost petaloid filament; D: Tricolpate pollen grain; E: Longitudinal section through carpels, all other floral parts removed; F: Two seeds in dif ferent view. G: A dehiscing follicetum of I. anisatum. Androecium with many stamens, free, spirally arranged, filaments short and thick , anthers basifixed, dehiscence longitudinal, connective extending beyond anther lobes, pollen tricolpate. Gynoecium with 5-20 free ascidiate carpels, in a sing le whorl, ovary superior with a single ovule, placentation is used in flavourings. I. verum (star anis) and I. anisatum (Japanese star anis ) are sources of anethole, used in dentistry, flavourings and perfumes. basal, stigma extending down on style. Fruit Phylogeny: The family is closely related to Winteraceae, and although the vessel s are present, the elements are long, slender, ana star-like aggregation of follicles (follice- gular, thin and greatly overlappi ng end walls tum), embryo minute, endosperm conspicu- with many scalariform perf oration plates. ous, seeds glossy. Pollination primarily by The fruit is a primi tive whorl of single-flies. Dispersal by elastic opening of folli- seeded follic les. Although the pollen grains cles, shooting out seeds. Economic importance: The family is impor-tant for producing aromatic oils. Oil f rom the bark of Illicium parviflorum (yellow star anis) are tricolpate, but their corpus morphology is different from eudicots. Loconte (1996) considers Illiciales among the most basal lineages of angiosperms. Major Families of Angiosperms 417 The family has been traditionally placed in the Magnoloid complex under order Il liciales, but has been removed in APG II and APweb along with Austrobaileyaceae, Schisandraceae and Monimiaceae into a separate order Austrobaileyales placed to

-wards the beginning of angiosperms without any informal superclade. Eames (1961 ) con-sidered Schisandraceae to be the closest fam-ily. The family Illiciaceae h as been found to be very closely related to Austrobaileyaceae and Schisandraceae through multigene analyses (Soltis, Soltis, and Chase, 1999; Soltis et al., 2000) having received 99 per cent bootstrap support. APG II sugge sts optionally including Illiciaceae under Schisandraceae (because the latter is a priority name). Schisandraceae includes climbing or trail-ing shrubs. Thorne includes the two families under suborder Illicineae, which was earlier (1999) pl aced under order Magnoliales, but now shifted to Canellales. * * * * * * * * * * * Cabombaceae Richard ex A. Richard 2 genera, 6 species America, India, Australia and tropical Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magniolopsida Subclass Polypetalae Magnoliidae Nymphaeidae Magnoliidae Chloranthidae Series+/Superorder Thalamiflorae+ Chloranthanae Nymphaeanae Nymphaeales Hydropeltidales Nymphaeales Nymphaeales Unplaced / (Nymphaeales) Nymph Order Ranales B & H under family Nymphaeaceae APG II optionally under Nymphaeaceae Salient features: Aquatic herbs, leaves floating, long-petioled, peltate, flowers large acyclic. Calyx with 3 sepals, pe taloid, in one ers bisexual, 3-merous, cyclic, or partially on long pedicels, stamens numerous, fruit whorl, free. Corolla with 3 petals, in one spongy with several immersed seeds. whorl, free, yellow, or purple, or white, clawed, or sessile. Androecium with 3 6 (Cabomba) stamens, or 12 18 (Brasenia), maturing centripetally, free, filaments sl ightly flattened, anthers bithecous, dehisMajor genera: Cabomba (5 species) and Brasenia (1). Description: Perennial aquatic herbs, rhizomatous, secretary cavities present, v essels absent, sieve-tube plastids S-type. Leaves submerged, or submerged and fl oat-ing, similar (Brasenia), or heterophyllous with dissected submerged leaves a nd entire floating leaves (Cabomba), alternate or op-posite (submerged leaves of Cabomba), sim-ple or compound, peltate, stipules absent, without sclerenchymato us idioblasts. Inflo-rescence with solitary axillary flowers. Flowcence by longitudinal slits, extrorse, pollen grains monosulcate, sometimes trichotomosulcate. Gynoecium with (2 )3 18 carpels, free, ovary superior, with a lo ngitudinal stigmatic surface (Brasenia), or apically stigmatic (Cabomba), (1 )2(-3 ) ovuled, placentation parietal, ovules pendulous,

anatropous, bitegmic, crassinucellate, outer integument not con-tributing to the micropyle. Fruit aggr egate of follicles, many seeded spongy berry, some418 Plant Systematics

K B

C ... G J I E Figure 13.5 Cabombaceae. Cabomba carolinaria. A: Flowering branch with submerged dissected leaves and broad peltate floating leaves; B: Flower: C: Gynoecium with 3 free ca rpels; D: Longitudinal section of carpel; E: Fruit; F: Seed. Brasenia schreberi. G: Portion of plant with peltate leaves and small flowers; H: Submerged part of plant covered with thick jelly; I: Flower with three sepals and three petals, e ssentially similar; J: Two-seeded panduraeform nut-like fruit; K: Globose seed. times indehiscent and nut-like (Brasenia schreberi); seeds endospermic, perisper m present, cotyledons 2. primitive lineage among angiosperms. The paleoherb complex is characterized by s cat-tered vascular bundles, absence of vascular cambium, leaves alternate, usual ly palmately veined, adventitious root system and lack of etherial cells. Judd e t al., (2002) include genus Cabomba and Brasenia under Cabomboideae in Nymphaeac eae (APG II, optionally), because their separation would render Nymphaeaceae par aphyletic. Thorne (1999, 2003, 2006, 2007) and Stevens (APweb, 2008), separate t hem under Cabombaceae on the basis of trimerous flowers, with distinct sepals an d petals, 2-3 free carpels, and fruit a follicle. Economic importance: None Phylogeny: The family is considered to be more strongly linked to monocots rathe r than Nymphaeaceae. The family has been located under superorder Nymphaeanae af ter Magnolianae but Takhtajan has finally taken it under distinct subclass Nymph aeidae. During the last decade the family has been identified as a constituent o f Paleoherb complex, forming the most * * * * * * * * * * * Major Families of Angiosperms 419 Nymphaeaceae Salisbury Water Lily family (excluding Cabombaceae) Throughout the world forming floating masses in freshwater habitats.

6 genera, 62 species Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Magnoliidae Nymphaeidae Magnoliidae Chloranthidae Series+/Superorder Thalamiflorae+ Nymphaeanae Chloranthanae Nymphaeanae Nymphaeales Nymphaeales Nymphaeales Nymphaeales Unplaced / (Nymphaeales) Order Ranales l

v.r L < 4 C B c A E F G H Figure 13.6 Nymphaeaceae. A: Leaf and flower of Nymphaea coerulea. B: Longitudin al of seed of N. alba. C: Transverse section of seed of the same showing plumule lying in the cavity of one cotyledon. D: Longitudinal section of flower of N. odorata ; E-H: Successively outer to inner stamens. Nuphar sp. I: Flower; J: Longitudina l section of flower; K: Gynoecium with stigmatic disc; L: Seed. Salient features: Aquatic herb, leaves float-ing, long-petioled, peltate, flower s large on long pedicels, stamens numerous, fruit spongy with several immersed s eeds. Description: Perennial aquatic herbs with stout creeping rhizome. Stem with scat tered vascular bundles, numerous air canals and laticifers. Hairs simple, usuall y producing mucilage. Leaves floating (Nymphaea, Victoria, etc.) or immersed, of

ten very large (up to 2 m dia. in Victoria amazonica) usually Major genera: Nymphaea (40 species), Nuphar (15), and Victoria (3). 420 Plant Systematics alternate, rarely opposite or whorled, sim-ple, cordate or orbicular, often pelt ate with long petiole emerging from rhizome, stipules absent or present. Inflore scence of soli-tary axillary flowers. Flowers floating or raised above water, bi sexual, actinomorphic with spirally arranged stamens, hypogynous. Calyx with 4-1 2 sepals, free or connate, often petaloid. Corolla repre-sented by staminodes, a bsent or many, free or connate at base, often passing into sta-mens. Androecium with many stamens, free, spirally arranged, filaments flattened sometimes poorly differentiated from an-thers, sometimes adnate to petaloid staminodes, pollen g rains usually monosulcate or inaperturate. Gynoecium with 3-many free or connate carpels with several locules and parietal placentation, unilocular with single or many ovules, stig-mas often elongated and radiating from the disc, often surr ounding a central bump, ovary superior (Nuphar), semi-inferior (Nymphaea), or in ferior (Euryale). Fruit a spongy berry, rarely an aggregate of nuts or pods; see ds usually operculate, arillate, with small embryo, endosperm absent but with a conspicuous perisperm. Pollination by beetles, flies and bees. Flowers of Victor ia and some species Nymphaea have starch-filled apical appendages of carpels as insect attraction, providing food, heat and charac-teristic smell. The fruit, on maturity, splits to separate individual segments (Nuphar) or ruptures under wat er so as to release seeds. of a child. The seeds of Victoria, Nymphaea and Euryale (Makhana) are often cons umed. Phylogeny: The family has been a subject of considerable discussion, often stron gly linked with monocots, although traditionally classified with dicots. The fam ily has been located under superorder Nymphaeanae after Magnolianae but Takhtaja n has finally taken it under a distinct subclass Nymphaeidae. During the last de cade, the family has been identified as a constituent of Paleoherb complex, form ing the most primitive lineage among angiosperms. The paleoherb complex is chara cterized by scattered vascular bundles, absence of vascular cambium, leaves alte rnate, usually palmately veined, adventitious root system and lack of etherial c ells. The family formerly also included genus Nelumbo, which has now been separated under family Nelumbonaceae because of distinct tricolpate pollen grains and absence of latici fers. Takhtajan places it under separate subclass Nelumbonidae, whereas in APG I I classification it is removed under Tricolpates (Eudicots) clade. APG II (optio n-ally) and Judd. et al., (2002) include genus Cabomba and Brasenia under Cabomb oideae in Nymphaeaceae, because their separation would render Nymphaeaceae parap hyletic. Thorne (2000, 2003, 2006, 2007) and APweb (2008), separate them under C abombaceae. Earlier placed after the Magnoloid complex, Thorne has finally shifted Cabombaceae and Nymphaeaceae in Economic importance: Species of Nymphaea basal angiosperm clades as suggested by (water lily), Nuphar (yellow water lily, molecular studies (Qui et al., 2000; S oltis et spatterdock), and Victoria (Amazon lily) are al., 2000). Wikstrm et al. (2001) suggest an ornamentals grown in ponds and lakes. The age for the Nymphaea les clade some 171 leaves of Victoria amazonica (Royal water lily) 153 MYA, with divergence occuring 144-111 are so large that they can support the weight MYA * * * * * * * * * * * Major Families of Angiosperms 421 Ceratophyllaceae S. F. Gray Hornwort family 1 genus, 6 species Widespread, forming floating masses in fresh water bodies. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Liliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Magnoliidae Nymphaeidae Magnoliidae Series+/Superorder Ordines anomali+ Ceratophyllanae Nymphaeaneae Nymphaeanae Order

Chloranthidae

Nymphaeales Ceratophyllales Nymphaeales Nymphaeales Ceratophyllales Salient features: Submerged aquatic herbs, roots absent, leaves whorled, often d ichoto-mously dissected, flowers minute, uni-sexual, perianth with 7-numerous br act-like segments, stamens 10 to numerous, anthers with prolonged connective, ca rpel 1, ovary superior, placentation apical, fruit an achene with 2 or more proj ections. extrorse, connective prolonged beyond an-thers into two prominent often coloured teeth, staminodes absent in female flower; pollen grains inaperturate, exine re duced, pollen tubes branched. Gynoecium with 1 carpel, ovary superior, unilocula r with 1 ovule, placentation apical, ovule pendulous, style continuous with the ovary, stigma extending along one side of style. Fruit a nut tipped by persisten t spine-like style and of-ten with two or more projections; seed with straight e mbryo, endosperm absent. Major genera: Single genus Ceratophyllum (6 species). Description: Submerged aquatic herbs often forming floating masses, rootless but some-times with colourless root-like branches anchoring the plant; stems branch ed but with never more than one branch at one node, with single vascular strand with cen-tral air canal surrounded by starch-contain-ing cells, with tannins. Le aves whorled, 3- 10 at each node, once to four times dichotomously dissected, ul timate leaf-segments with two rows of minute teeth and tipped by two bristles, s tomata and cuticle absent, stip-ules absent. Inflorescence with solitary ax-illa ry flowers, usually one flower in a whorl of leaves. Flowers unisexual (plants m onoecious), male and female flowers usu-ally on alternate nodes, actinomorphic, very small. Perianth with 7 to numerous tepals, linear, bract-like, slightly con nate at base. Androecium with numerous stamens, fila-ments indistinct, anther ob long-linear, bithecous, dehiscence longitudinal, Economic importance:. Floating masses provide protection to fish fry. The masses also support bilharzia-carrying snails and malaria- or filaria-carrying mosquit o larvae. Fruits and foliage form food for migratory waterfowl. It is sometimes troublesome, choking waterways. Phylogeny: The phylogeny of the family has been a matter of great speculation. B entham and Hooker placed this family along with others of uncertain affinities u nder Ordines anomali. It is usually considered to be related to Nymphaeaceae (La wrence, 1951; Heywood, 1978 both include Nelumbo under Nymphaeaceae) particularly genus Nelumbo, which has now been removed to a distin ct family Nelumbonaceae. Cronquist (1988) places all the three families under the same order Nymphaeales. G. Dahlgren (1989) includes Nelumbonaceae under order Nelumb onales of 422 Plant Systematics

"ll". o E U A Nf B Figure 13.7 Ceratophyllaceae. Ceratophyllum submersum. A: Portion of plant; B: W horl of leaves at node spread out to show dichotomous segments; C: Male flower with perianth an d several sessile stamens; D: Young stamen with connective produced into two pro minent teeth; E: Dehiscing stamen; F: Fruit with persistent style and spiniscent projections; G: Longitudinal section of fruit with pendulous seed. superorder Magnolianae, whereas Nymphaea-ceae and Ceratophyllaceae are placed un der order Nymphaeales of superorder Nymphae-anae. Takhtajan (1997) removed Nelum bona-ceae under distinct order Nelumbonales, dis-tinct superorder Nelumbonanae a nd even a distinct subclass Nelumbonidae. Ceratophylla-ceae and Nymphaeaceae are placed under subclass Nymphaeidae but separate super-orders Ceratophyllanae and Nymphaeanae, under respective orders Ceratophyllales and Nymphaeales. Thorne (2 003) placed Nymphaeaceae and Cabombaceae under Nymphaeales, superorder Nymphaean ae, under subclass Magnoliidae, subsequently (2006) shifting to Chloranthidae. C eratophylla-ceae is taken closer to Nelumbonaceae under Ranunculanae of subclass Ranun-culidae but under separate orders Cerato-phyllales and Nelumbonales, resp ectively. Latest 2007 revisions shows a major shift of Ceratophyllaceae to Chloranthidae under order Nymphaeales. The family Ceratophyllaceae has attracted a lot of interest with morphological a nd fossil evidence (Les et al., 1991) and molecular evidence (Chase et al., 1993 ) suggesting basal placement in angiosperms. Hickey and Taylor (1996), however, suggested that the aquatic plant with highly reduced vegetative body and pollen wall, tenuinucellate unitegmic ovules and problematic fossil record is a poor ca ndidate for basal-most position. Ceratophyllaceae is possibly sister to monocots (e.g. Graham & Olmstead, 2000; Zanis et al., 2002; Whitlock et al., 2002), and accordingly in APG II and Apweb, placed in distinct order Ceratophyllales before monocots and Chloranthales, respectively without any supraordinal informal grou p. Nymphaeaceae is similarly placed under Nymphaeales but Nelumbonaceae removed to Eudicots. * * * * * * * * * * * Major Families of Angiosperms 423 Subclass 2. Magnoliidae Superorder 1. Magnolianae 4. Hernandiaceae 5. Atherospermataceae 6. Gomortegaceae Order 1. Magnoliales Family 1. Myristicaceae 7. Siparunaceae

2. Magnoliaceae 3. Degeneriaceae 4. Himantandraceae 5. Eupomatiaceae Order 3. Canellales 1. Winteraceae 2. Canellaceae 6. Annonaceae Order 4. Piperales Order 2. Laurales 1. Aristolochiaceae 1. Calycanthaceae 2. Lactoridaceae 2. Monimiaceae 3. Hydnoraceae 3. Lauraceae 4. Saururaceae 5 Piperaceae Magnoliaceae A. L. de Jussieu Magnolia family 7 genera, 182 species Warm temperate to tropical regions of Southeast, North and Central America, West Indies, Brazil, and East Asia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Nymphaeidae Magnoliidae Magnoliidae Series+/Superorder Thalamiflorae+ Magnolianae Magnolianae Magnolianae Magnoliids* Order Ranales Magnoliales Magnoliales Magnoliales Magnoliales Magnoliales Salient features: Trees or shrubs with alternate simple leaves, stipules caducou s, leaving a circular scar at the node, nodes multilacunar, flowers usually soli tary, bisexual, large, floral parts numerous, spi-rally arranged on elongated th alamus, tepals gradually passing from outer sepals to inner petals, stamens lami nar, carpels free, seed often suspended by thread like funiculus. vestured pits, wood parenchyma apotracheal (terminal), sieve-tube plastids S-typ e, or P-type and S-type; when P-type, subtype I (b). Leaves evergreen or deciduo

us, alternate, spiral, petiolate, simple, dissected (Lirio-dendron), pinnatifid or entire, pinnately veined, or palmately veined, stipules lar ge, sheathing, enclosing the terminal buds, caducous, leaving a ring-shaped scar at the no de, stomata paracytic, or anomocytic, minor leaf Major genera: Magnolia (80 species), Michelia veins without phloem transfer cell s (Magno(40), Talauma (40) and Liriodendron (2). lia). Inflorescence with usually solitary terminal, or axillary flowers. Flowers bracteate (the bracts spathaceous); large, regular, bisexual. Perianth with 6-1 8 tepals, free, sequentially intergrading from sepals Description: Trees or shrubs, nodes 5-lacunar or multilacunar, vessels-elements with scalariform ends, vessels without 424 Plant Systematics W

c Figure 13.8 Magnoliaceae. Magnolia virginiana. A: Flowering branch with single t erminal flower; B: Stamen, laminar and with apical sterile appendage; C: Longitu dinal section of gynoecium, two anatropous ovules in each carpel; D: Seed with fleshy seed coat r emoved; E: Longitudinal section of seed showing fleshy seed coat, copious endosp erm and small embryo. M. grandiflora. F: Flower bud; G: Vertical section of flow er; H: Floral receptacle with half of the stamens removed; I: Anther; J: Dehisce d fruit with arillate seeds hanging through thread-like funiculus. to petals, or petal-like (usually), usually spi-rally arranged, rarely 3 4 whorled , white, or cream, or pink, deciduous. Androecium with numerous (50 200) stamens, maturing cen-tripetally, free, spirally arranged, all fertile, usually laminar ( the four paired microsporangia embedded, the stamens of-ten more or less strap-s haped), anthers adnate, dehiscence longitudinal, through slits or valves, extror se (Liriodendron), or latrorse to introrse, bithecous, appendaged often by prolo ngation of the connective or unappendaged, pollen grains monosulcate. Gynoecium with (2 ) 20 200 free carpels, ovary superior, carpel fully or incompletely closed, 2 ( 20) ovuled, placentation marginal; ovules funicled, pendulous, biseriate (on t he ventral suture), anatropous, bitegmic, crassinucellate; stigma extending down the style, but sometimes terminal. Fruit an aggregation of follicles or indehiscent samaras (Liriodendron), or united into fleshy syncarp (Aromadendron); seeds endo spermic, endosperm oily, seeds usually large, often with long thread-like funicu lus. Pollination primarily by beetles. The fruits are primarily dispersed by ani mals, but the samaras of Liriodendron are wind dispersed. Economic importance: Various species of Magnolia (M. grandiflora, M. kobus, M. s tellata) and Michelia (M. fuscata, M. champaca sapu, also source of timber) are g rown as ornamentals. Liriodendron tulipifera (tulip tree or yellow poplar) is a valuable timber source in USA. Species of Magnolia (M. hypoleuca), and Michelia also constitute sources of timber. Major Families of Angiosperms 425 Phylogeny: The family was regarded as the most primitive of the extant angiosper ms for several decades in the classification systems of Hallier (1905), Hutchins on (1926, 1973), and earlier versions of Cronquist and Takhtajan. The view was f

irst challenged by Smith (1945), who considered that Magnoliaceae are rela-tivel y highly specialized both vegetatively and florally, casting some doubt on the a ssump-tion of the primitive nature of the family, and implying that groups such as Winteraceae, etc., may be at least as primitive. The status of Magnoliaceae a s the most primitive fam-ily was strongly challenged by Carlquist The family is considered to be mono-phyletic based on the support from rbcL and ndhF sequences (Qui et al., 1993, Kim et al., 2001). These studies, however, que stion the recognition of Talauma, Michelia and Manglietia as distinct genera, as it renders Magnolia as paraphyletic. Although Liriodendron is quite distinct, a ll other genera have been merged with Magnolia in the recent works. Figlar and N ootebroom (2004) divide the enlarged genus Magnolia into three subgenera: Magnol ia, Yulania and Gynopodium. Two clades are distinguished within the family one r epresented by Liriodendron, and the other by rest of the genera. Judd et al., (2 008), Stevens (2008) as such recognize only 2 genera Magnolia and Liriodendron w ithin the family. Thorne (2003, 2006,2007), places Magnolia and other 5 genera i n subfamily Magnolioideae, whereas Liriodendron is placed in mono-generic Liriod endroideae. (1969), Gottsberger (1974) and Thorne (1976), claiming Winteraceae to be the mos t primi-tive family. The primitive features of Magnoliaceae include spirally arr anged floral parts, laminar stamens, fruit a follicle, longer and narrower vesse l elements, monosulcate pollen grains and beetle pollination. * * * * * * * * * * * Degeneriaceae I.W. Bailey & A.C. Smith Degeneria family 1 genus, 2 species Endemic to Fiji. Placement: B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Class Magnoliopsida Magnoliopsida Subclass Magnoliidae Nymphaeidae Magnoliidae Magnoliidae Series+/Superorder Magnolianae Magnolianae Magnolianae Magnoliids* Order Not described then Magnoliales Magnoliales Magnoliales Magnoliales Magnoliales Salient features: Trees or shrubs with al-ternate simple leaves, stipules absent , nodes 5-lacunar, flowers usually solitary, bisexual, large, sepals and petals distinct, sepals 3, petals 12-18, stamens many, laminar, 3-veined, inner modifie d into staminodes, carpel single, incompletely sealed, fruit leathery with many seeds. Genus: Single genus Degeneria with 2 species, D. vitiensis and D. roseiflora. Description: Large trees; bearing essential oils, nodes 5-lacunar, vessel-elemen ts with oblique end walls, sieve-tube plastids P-type, pith with diaphragms. Lea ves alternate, petiolate, non-sheathing, gland-dotted, 426 Plant Systematics

rv

B 11 D Figure 13.9 Degeneriaceae. Degeneria vitiensis. A: Tree growing in natural habit at in Fiji.; B: A branch with flowers; C: Stamen, laminar with undifferentiated filament and anthe r part; D: Transverse section of carpel; E: Fruit. aromatic, simple, entire, pinnately veined, exstipulate, stomata paracytic, meso phyll with spherical etherial oil cells. Inflores-cence with solitary pendulous flowers, (su-pra) axillary. Flowers medium-sized to large, regular, polycyclic, thalamus shortly raised, sepals and petals distinct. Calyx with 3 se-pals, 1 who rled, free, persistent. Corolla with 12-18 petals, larger than the sepals, 3 5 wh orled, free, fleshy, deciduous, sessile. Androecium with about 30 50 stamens, matu ring centripetally, free, 3 6 whorled, innermost 3 10 modified into staminodes; fert ile stamens laminar, flattened, oblong, 3-veined; anthers bithecous, adnate, dehiscence longitudinal, with slits or valves, extrorse, tapetum glandular, pollen grains monosulcate. Gynoecium with single car-pel, ovary superior, single chamb ered, car-pel incompletely closed (largely unsealed at anthesis), style absent, stigma running nearly the entire length of carpel, placentation marginal, ovules 20-30, in two rows, long funicled, with a conspicu ous funicular obturator; anatropous, bitegmic, crassinucellate, outer integument not contributing to the micropyle. Fruit leathery, with a hard exocarp, dehisce nt, or indehiscent, 20 30 seeded; seeds flattened, more or less sculptured, with a n orange-red sarcotesta, embryo well differentiated but small, cotyledons 3 ( 4), copiously endospermic, endosperm ruminate, oily. Pollination by beetles. Economic importance: No economic value known. Phylogeny: The family was earlier included under Winteraceae, and was considered closer to Zygophyllum by Hutchinson (1973). It is now treated to be an independ ent family, more closely allied to Magnoliaceae and Himantandraceae. Takhtajan ( 1987, 1997), Major Families of Angiosperms 427 considers Winteraceae and Degeneriaceae to be more primitive families, but is pe rhaps is the only one among the recent authors to consider Degeneriaceae as the most primi-tive family of extant angiosperms. The primi-tive features of the fam ily include alternate simple leaves, numerous laminar stamens, partially closed carpel with stigma run ning the entire length of carpel, 3-4 cotyledons, and monosulcate pollen grains. Thorne (2003, 2006, 2007) places Degeneriaceae between Magnoliaceae and Himanta ndracea.

* * * * * * * * * * * Annonaceae A. L. de Jussieu Annona or Pawpaw family 128 genera, 2300 species Temperate and tropical regions of Eastern North America and Eastern Asia, and tr opical South America. Mainly distributed in Old World tropics, in moist forests. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Thalamiflorae+ Magnolianae Magnolianae Magnolianae Magnoliids* Order Ranales Magnoliales Annonales Annonales Magnoliales Magnoliales Salient features: Trees or shrubs with al-ternate distichous leaves, stipules ab sent, leaves glaucous or with metallic sheen, flow-ers fragrant, flowers trimero us with numer-ous spirally arranged stamens, many car-pels free, fruit an aggreg ate of berries, seed with ruminate endosperm. tire, pinnately veined, stipules absent, domatia recorded in 3 genera, stomata p aracytic, secretory cavities containing oil, mucilage, or resin. Inflorescence w ith soliMajor genera: Guateria (250 species), Xylopia (150), Uvaria (100), Artabotrys (1 00), Annona (100), and Polyalthia (100). tary flowers, or racemose. Flowers regular, cyclic, usually bisexual, rarely uni sexual, thalamus sometimes elongated (Mischogyne) Calyx usually with 3 sepals, o r 6 and 2 whorled, free, valvate. Corolla with 3-6 pet-als, 1 2 whorled, free, imb ricate or valvate. Androecium with 25 100 stamens, maturing centripetally; free, a ll equal, spirally arranged, rarely 3-6 whorled, rarely outer forming staminodes (e.g. in Uvaria), anthers bithecous, adnate, dehiscing by longitudinal slits or valves; extrorse, connective prolonged into appendage, tapetum glandular, polle n shed in aggregates (5 genera), or as single grains; when aggregated, in tetrad s (usually), or in polyads (octads in Trigynaea). Pollen Description: Trees, or shrubs, or lianas, bear-ing essential oils, nodes unilacu nar, or bilacunar, vessel end-walls horizontal, sim-ple, vessels without vesture d pits, wood dif-fuse porous; partially storied, sieve-tube plastids P-type, sub type I (a), pith commonly with diaphragms Leaves evergreen, alter-nate, disticho us, non-sheathing, simple, en428 Plant Systematics

D f C B E H Figure 13.10 Annonaceae. Asimina triloba. A: Flowering branch bearing solitary f lowers; B: Vertical section of flower ; C: Longitudinal section of carpel showing ovules ; D: Pollen grain; E: Fruit. Annona furfuracea. F: Flowering branch; G: Vertical section of receptacle showing male flowers towards the centre and female flowers towards the periphery; H: Longitudinal section of carpel showing basal ovule. grains monosulcate or nonaperturate, or with two parallel furrows at the equator , or ulcer-ate. Gynoecium with 10 100 carpels, usu-ally free, rarely united, place ntation of free carpels basal, when syncarpous 1 locular, or 2 15 locular, parieta l, or basal. Ovules 1-50, apotropous, with ventral raphe, bitegmic, crassinucell ate, outer integument not contri-buting to the micropyle. Fruit fleshy, com-monl y an aggregate of berries; seeds endoused in perfumes. The spicy fruits of West African Xylopia aethiopica are the so -called Negro pepper used as a condiment, and those of Monodora myristica used as substitute for nutmeg. Phylogeny: It is generally agreed upon that the family is derived from Magnoliac eous stock. Hutchinson placed the family under Annonales after Magnoliales, from which, spermic, endosperm ruminate, oily. Pollination according to him, they were clear ly derived. mostly by beetles. Dispersal especially of The primitive features in clude spirally arfleshy fruits by birds, mammals and turtles. ranged numerous stamens and carpels , conEconomic importance: Many species of nective prolonged into an appendage. The sepals and petals are more advanced tha n Magnoliaceae. Most of the recent authors Annona are cultivated for their edible fruits: A. squamosa (sweet sop), A. muricata (sour (except Dahlgren and Takhtajan, who p lace sop), A. reticulata (custard apple), and it under Annonales), however, incl ude this A. cherimola (cherimoya). Flowers of Cananga family under Magnoliales ( Stevens, 2008; odorata (ylang-ylang) and Mkiluaf ragrans are Thorne, 2003, 2006, 2007). The genera with Major Families of Angiosperms 429 connate carpels and with fleshy berries are considered more advanced than those with free carpels. It is proposed that the diversification of Annonaceae may have occured (84)82-57 mybp (Doyle et al. 2004; Schar aschkin & Doyle 2005). * * * * * * * * * * * Calycanthaceae 4 genera, 8 species Lindley Strawberry shrub family (including Idiospermaceae) Family with discontinuous distribution, found in North America, East Asia and Qu eensland. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Thalamiflorae+ Magnolianae Magnolianae Magnolianae Magnoliids* Order Ranales Laurales Calycanthales Laurales Laurales Laurales Salient features: Shrubs with opposite sim-ple leaves, stipules absent, flowers with nu-merous spirally arranged tepals, numerous stamens along the rim of cup-l ike recepta-cle, fruit single-seeded achene. veins, sequentially intergrading from sepals to petals, free, inserted along the rim of receptacle. Androecium with 15 55 stamens, maturing centripetally, free, s pirally arranged at the top of the hypanthium, laminar or linear, bithecous, inn er 10-25 modified into usually nectariferous staminodes, anthers adnate, dehisce nce by longitudinal slits, extrorse, connective extended into an appendage, poll en grains 2( 3) aperturate, sulcate. Gynoecium with 5 45 free carpels, spirally arra nged within the hypanthium, ovary superior, style distinct, stigma terminal, ova ry 2 ovuled (upper often abortive), placentation marginal, ovules ascending, apo tropous with ventral raphe, anatropous, bitegmic, crassinucellate, or pseudocras sinucellate, outer integument not contributing to the micropyle. Fruit an aggreg ate of achenes enclosed in the fleshy hypanthium; seeds nonendospermic, embryo w ell differentiated (large), cotyledons 2, spirally twisted. Pollination by insec ts, mainly beetles. Major genera: Calycanthus (3 species), Chimonanthus (3), Sinocalycanthus (1) and Idiospermum (1). Description: Small trees, or shrubs with aro-matic bark, bearing essential oils, nodes unilacunar, vessel-elements with oblique end-walls, sieve-tube plastids P -type, subtype I (a). Leaves opposite, leathery, petiolate, gland-dotted, simple , entire, pinnately veined, stipules absent, stomata paracytic, hairs unicellula r or absent, mesophyll with spherical etherial oil cells. Inflorescence with sol itary terminal flowers on specialized leafy short-shoots. Flowers medium-sized t o large, regular, bisexual with spirally ar-ranged floral parts, markedly perigy nous. Perianth with 15-30 tepals, each with 3-4 430 Plant Systematics r

k'1

E D

H Figure 13.11 Calycanthaceae. A: Flowering branch of Calycanthus laevigatus. Chim onanthus praecox. B: Flower bud. C: Vertical section of flower; D: Longitudinal section of ca rpel. Calycanthus floridus. E: Flower; F: Vertical section of flower; G: Longitu dinal section of carpel; H: Nut. Economic importance: Calycanthus floridus Laurales. Loconte and Stevenson (1991) , pro-(Carolina allspice) and C. occidentalis jected Calycanthaceae as basic ang iosperms are grown as ornamental shrubs. Bark of with a series of vegetative and reproductive C. fertilis and C. floridus yield medicinal ex- angiosperm plesiomorphies such a s shrub tracts. Chimonanthus praecox (winter sweet) is widely cultivated, and is one of the few spe-cies flowering in cold winter with snow around, the flowers used in Japan to make perfumes. habit, unilacunar two-trace nodes, vessels with scalariform perforations, sievetube elements with starch inclusions, opposite leaves, strobilar flowers, leaf-l ike bracteopetals, poorly differentiated numerous spirally arranged tepals, and few ovulate carpels. Food bodies terminating the stamen connectives indicate bee tle pollination. It is interesting to note that genus Idiospermum (which was rec ognized as new genus based on Calycanthus australiensis by S. T. Blake in 1972) was considered as the most primitive flowering plant by these authors. Blake had separated Idiospermum under distinct family Idiospermaceae, also recognized by Hutchinson, as distinct from CalycanthaPhylogeny: The family is closely related to Magnoliaceae and Annonaceae in its n u-merous spirally arranged floral parts and free carpels. Hutchinson included th e family under Rosales primarily because of perigynous flowers and free carpels, a posi-tion contested by several authors. Thorne (1996, 2000) regarded it to be closely related to Monimiaceae and placed under suborder Laurineae of Magnolial es, but subsequently (2003, 2006, 2007) under independent order Major Families of Angiosperms 431 ceae. Endress (1983) had described In all respects, Idiospermum gives the impress ion of a strange living fossil . Molecular studies suggest Calycanthaceae to be basal within Laurels, probably sister to all other Laurales (Doyle and Endress, 2000). * * * * * * * * * * * Lauraceae Durande Laurel family 50 genera, 3000 species Throughout tropical and subtropical regions of the world, primarily in rain fore sts of Southeast Asia and North America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Daphnales+ Lauranae Magnolianae Magnolianae Magnoliids* Order Laurales Laurales Laurales Laurales Laurales B & H as Laurineae Salient features: Aromatic trees or shrubs, leaves alternate, perianth small and undif-ferentiated, stamens in several whorls, fruit single seeded drupe or berr y. (14 genera) represented by pits, pockets, or hair tufts, stomata, paracytic, hai rs mostly unicellular, mesophyll usually with spherical etherial oil cells. Infl orescence cymose, or racemose, often umbelliform with involucral bracts, rarely solitary. Flowers small, often fragrant, regular, bisexual, rarely unisexual, us ually 3-merous, cyclic, with well-developed hypanthium. Tepals usuMajor genera: Litsea (400 species) Ocotea (350), Cinnamomum (250), Cryptocarya ( 250), Persea (200), Beilschmeidia (150) and Lindera (100). Description: Aromatic trees and shrubs, sometimes parasitic climbers (Cassytha) nodes unilacunar with two traces, vessel el-ements with scalariform or simple en d-walls, without vestured pits, wood partially storied, sieve-tube plastids P-ty pe, or S-type; when P-type, subtype I(b). Leaves nearly always ev-ergreen, usual ly alternate and spiral, rarely opposite or whorled, leathery, petiolate, non-sh eathing, gland-dotted, aromatic, simple, entire , sometimes lobed (Sassafras), p innately veined, stipules absent, domatia ally 6, sometimes 4, free, (1 )2( 3) whorled, similar, sepaloid to petaloid, green, or white, or cream, or yellow. Androecium with (3 )9( 26) stamens, free, equal, or markedly unequal, (1 )3( 4) whorled, inner sometimes modified into staminodes, somew hat laminar to petaloid by expansion of the filament and connective, filaments a ppendaged or not, anthers bithecous, basifixed, dehiscence longitudinal by valve s opening from base to apex, or dehiscing by pores (in Hexapora), usually intror se, sometimes extrorse, tapetum amoeboid (mostly), or glandular (in several gene ra), 432 Plant Systematics

Figure 13.12 Lauraceae. Litsea doshia. A: Flowering branch flowers in globose axillary clusters; B: Flower; C: Fruit. Cinnamomum tamala. D: Flowering branch terminal

paniculate inflorescence; E: Flower; F: Fruit; G: Anther dehiscing by valves. (A fter Polunin and Stainton, Flowers of the Himalaya, 1984). pollen grains nonaperturate, exine spiny. Gynoecium with 1 carpel, ovary usually su-perior, sometimes inferior (Hypodaphnis), style distinct with terminal stigm a, placentation apical; ovules pendulous, apotropous, with dorsal raphe, non-ari llate, anatropous, bitegmic, crassinucellate, outer integument not contributing to the micropyle. Fruit fleshy, drupaceous, or baccate, enclosed in the fleshy r eceptacle, 1 seeded; seeds nonendospermic, embryo well differentiated, cotyledon s massive, occasionally ruminate. Phylogeny: Lauraceae is generally considered to be a more specialized family pla ced closer to Monimiaceae and Calycanthaceae. The order Laurales is generally co nsidered to belong to the magnoloid complex, and represents an early divergent l ineage . The derived apomorphies of Lauraceae and Monimiaceae include ovary with single carpel, and spinose pollen grains. The families also share pollen lackin g aperture, stamens with paired appendages, and anthers opening by valves. Economic importance: The family contrib-utes some important spices from plants s uch as Laurus nobilis (bay leaves), Cinnamomum verum (cinnamon), C. camphora (ca mphor), and Sassafras albidum (sassafras). Persea americana (avocado) is an impo rtant tropical fruit. Aromatic oils are obtained from Lindera (benzoin) and Sass afras. Species of Litsea, and Ocotea yield fragrant woods used in cabi-net-makin g. The family is traditionally divided into two subfamilies Cassythoideae (Cassytha ) and Lauroideae (rest of genera). The latter is variously divided into 3 (Werff and Richter, 1996) or 5 tribes (Heywood, 1978). It has been suggested that the perianth in some Lauraceae may represent modified stamens (Chanderbali et al. 20 04) as both the tepals and the stamens of Persea have three traces. Hypodaphnis, with an inferior ovary is con-sidered to be sister to the rest of the family (S tevens, 2008). * * * * * * * * * * * Major Families of Angiosperms 433 Winteraceae R. Br. ex Lindley Winter s bark family 8 genera, 90 species Tropical, subtropical and temperate regions of Madagascar, South America, Mexico , Australia, New Caledonia and New Guinea. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Thalamiflorae+ Magnolianae Magnolianae Magnolianae

Magnoliids* Order Ranales Magnoliales Winterales Winterales Canellales Canellales B & H under family Magnoliaceae * informal clade names, no formal group names gi ven above order in APG II and Apweb

ill, D

J i urJII

A. ~u "WAR U r70 Figure 13-13 Winteraceae. Drimys winteri. A: Flowering branch of var. punctata; B: Vertical section of flower; C: fruits. Tasmannia sp. D: Flower; E: Vertical s ection of carpel. 434 Plant Systematics Salient features: Trees and shrubs with simple alternate leaves with glaucous un der surface, stipules lacking, nodes trilacunar, vessels absent, flower medium s ized in cymes, stamens numerous with flattened filament, pollen grains in tetrad s, stigma extending down on style and fruit a follicle. tance, used locally in South America as a tonic. It was also used once by marine rs for scurvy prevention. Some other species also have medicinal uses. The fruit s and seeds of D. lanceolata (mountain pepper) are used as pepper and allspice s ubstitute. Major genera: Tasmannia (40 species), Bubbia (30), and Drimys (4). Phylogeny: The family has gained consider-able phylogenetic significance during the last three decades and has been regarded as the most primitive living family of angiosperms, and Drimys (according to Eames, 1961 the combination of charact ers suggests Belliolum as most primitive genus of the family) as the most primit ive genus in the recent classifications of Thorne (pre-2003 versions) and Cronqu ist. Takhtajan also regarded this as a very primitive family but considered Dege neria (formerly under Winteraceae, but now removed to Degeneriaceae) to be the m ost primitive genus. The primitive position of Drimys is supported by the absenc e of vessels, narrow elongated tracheids, laminar stamens and more primitive bee

tle pollination. The fossil records of the family also go back to 100-140 years. Only Chloranthaceae is perhaps as old in fossil history of angiosperms. Thorne (1996) lists other primitive features of Drimys as alternate, entire, exstipulat e leaves, pollen grains in tetrads, long cambial initials and tracheids, heterog enous rays, and poorly-organized pinnate venation, small medium sized flowers in cymes, style-less carpel, partly sealed stigmatic marDescription: Trees or shrubs lacking ves-sels and with narrow elongated tracheid s, nodes trilacunar, sieve-tube plastids S-type. Leaves leathery, alternate, aro matic, gland-dotted, containing terpenoids, entire, sim-ple with reticulate vena tion, under surface glaucous due to waxy coating, stipules ab-sent. Inflorescenc e cymose or fasciculate, with medium-sized few flowers, solitary ter-minal in Zy gogynum. Flowers usually bi-sexual, rarely polygamous, actinomorphic with spiral ly arranged stamens, hypogynous. Calyx with 2-6 sepals, free or connate at base (Drimys), valvate, sometimes falling off as cap. Corolla with 5 or many petals, 2- or more-seriate, mostly conspicuous in bud, imbricate. Androecium with many s tamens, centrifugal, free, filaments flattened or al-most laminar, poorly differ entiated from an-thers, anthers bithecous, dehiscing longi-tudinally, introrse, connective frequently ex-tending beyond anthers, tapetum amoeboid or glandular, pollen uniporate, released in tetrads. Gynoecium with 1-many carpels, in a singl e whorl, usually free, sometimes slightly connate (Kxospermum) or syncarpous (Zy gogynum), ovary superior with parietal placentation, ovules 1-many, anatropous, bitegmic, crassinucellate, stigma extending down on style or capitate, carpels s ometimes partially unsealed (Drimys). Fruit an etaerio of berries or follicles, embryo minute, endosperm conspicuous. Pollination by small beetles (Drimys), fli es and moths, some spe-cies wind pollinated (Tasmannia). Dispersal especially of berries is by vertebrates. gins, and follicle fruit. Economic importance: The bark of Drimys winteri (winter s bark) is of medicinal im porThe position of Winteraceae at the base of angiosperms, however, has been refute d during the last decade, largely due to emergence of the herbaceous origin hypo thesis, and the results of cladistic studies largely based on molecular data. Yo ung (1981) in-terpreted neoteny in the family with a se-ries of reversals. It is also suggested that the family shares common ancestry with Illiciaceae (Doyle a nd Donoghue, 1993) and Amborellaceae (Loconte and Stevenson, 1991). Loconte (199 6), on comparison of various hypotheses concluded that the tree based on Wintera ceae hypothesis is two steps longer than one based on Calycanthaceae. Major Families of Angiosperms 435 The family has recently been placed along with Canellaceae in a separate order C anellales (APG II, APweb), not at the begin-ning of angiosperms but after Ambore llaceae, Chloranthaceae and Austrobaileyaceae. The sister-group rela-tionship of Winteraceae and Canellaceae has received bootstrap support of 99 or 100 per cen t in all recent multigene analyses (Qui et al., 1999; Soltis et al., 1999; Zanis et al., 2002, 2003). The two are characterised by apomorphine alkaloids, trilacunar nodes and sieve-tube plastids with starch and protein crystalloids. Thorne earl ier (2003) placed the two families under suborder Canellineae of order Canellale s, the other suborder being Illiciineae. However, With the shifting of latter un der Chloranthales, the two families are placed directly under Canellales (Thorne , 2007). * * * * * * * * * * * Saururaceae Martynov Lizard-tail family 4 genera, 6 species Temperate or subtropical Coasts of North America and East Asia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Microembryeae+ Piperanae Nymphaeanae Magnolianae Magnoliids* Order Piperales Piperales Piperales Piperales Piperales B & H under family Piperaceae Salient features: Perennial herbs, leaves alternate, stipules adnate to petiole, flowers reduced, in dense spikes, coloured bracts often surrounding the base of spike, looking like petals, the whole inflorescence looking like a flower, stam ens 6, somewhat attached to carpels, carpels free or united, fruit a capsule. end-walls, scalariform, sieve-tube plastids S-type. Leaves alternate; spiral to distichous, petiolate, aromatic, simple, pinnately or palmately veined, stipules intrapetiolar and adnate to the petiole, stomata cyclocytic, mesophyll with sph erical etherial oil cells. Inflorescence a raceme, or spike, often with involucr al bracts resembling petals (Houttuynia, Anemopsis) and inflorescence appearing like a flower, or without involucral bracts (Saururus, Gymnotheca). Flowers smal l, regular; bisexual, cyclic Perianth absent. Androecium with 3, or 6, or 8 stam ens, united with the gynoecium or not, free, 1 whorled (when 3), or 2 whorled (w hen six or eight), with slender filaments, anthers basifixed, dehiscence by long itudinal slits, Major genera: Saururus (2 species), Gymnotheca (2), Houttuynia (1), Anemopsis (1 ) and Circaeocarpus (1). Description: Perennial aromatic herbs bear-ing essential oils, rhizomatous, node s 5-lacunar, or multilacunar, vascular bundles in one ring, vessel elements with oblique 436 Plant Systematics xf. s 4

c B

D Figure 13.14 Saururaceae. Saururus cernuus. A: Flowering branch with elongated s pike. B: Flower with subtending bract; C: Vertical section of flower; D: Transverse section of f ruit. E: Flowering portion of Anemopsis californica with showy involucre bracts below the spike, and the plant with basal leaves. extrorse to latrorse to introrse, tapetum glan-dular, pollen grains aperturate, or nonaperturate, usually monosulcate. Gynoecium with 3 or 4( 5) carpels, free or united, semicarpous in Saururus (the conduplicate carpels distinct above the con nate base), superior (mostly), or inferior (Anemopsis); carpel in Saururus incom pletely closed, style with decurrent stigma, (1 )2 4 ovuled, placentation dispersed (laminar-lat-eral), ovary 1 locular, Styles 3 4( 5); in the genera other than Saurur us apical, stigmas 3 4( 5), placentation parietal, ovules in the single cavity 20 40( 50 ) (6 10 on each pla-centa), orthotropous to hemianatropous, bitegmic, tenuinucella te, or crassinucellate, outer integument contributing to the micro-pyle. Fruit a n aggregate (Saururus), or not, indehiscent (Saururus) or dehiscent, fleshy, a c apsule, or capsular-indehiscent; seeds scantily endospermic, perisperm present, embryo rudimentary. Economic importance: Houttuynia cordata forms a good ground cover and is commonl y cultivated. The leaves of this species are used as salad and for treating eye diseases in Vietnam. Saururus chinensis is also occasionally cultivated. The aro matic stoloniferous stock of Anemopsis californica was once fashioned into cylin drical necklace beads by American Indians, and hence the name Apache beads. The stock infused in water is also a reputed treatment for ma-laria and dysentery. Phylogeny: The family is considered to be less specialized than Piperaceae in it s free to united carpels and parietal placentation, and is believed to be belong ing to paleoherb complex, early basal branch of angiosperms. Hickey and Taylor ( 1996), who proposed the herbaceous origin hypothesis, believe that the flowers o f Piperaceae, as well as of Anemopsis and Houttuynia, arose through Major Families of Angiosperms 437 suppression of the system of inflorescence axes of gnetopsids and bracts to brin g either a single distal and one more proximal pair of anthions together above t he subtending bract of the second order inflorescence axis. The four carpelled flo wers of Saururus and Gymnotheca are the result of the reduction of an infloresce nce axis to a penultimate and ultimate pair of anthions. Zeng et al. (2002) on the basis of matR gene studies concluded that 4 genera and six species of Saururaceae form a monophyletic group . Circaeocarpus saurroides C. Y. Wu earlier placed in Saururaceae is conspecific with Zippelia begoniaefolia Blume and belongs to Piperaceae. * * * * * * * * * * * Piperaceae Batsch Pepper family 5 genera, 2,015 species Tropical and subtropical regions, mainly in rain forests. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne

APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Magnoliidae Magnoliidae Magnoliidae Magnoliidae Series+/Superorder Microembryeae+ Piperanae Nymph Magnolianae Magnoliids* Order Piperales Piperales Piperales Piperales Piperales Salient features: Herbs, shrubs or climbers with jointed nodes, vascular bundles scat-tered, leaves alternate, petioles sheathing the nodes, flowers small in de nse spikes, peri-anth absent, stamens 2-6, ovary with single ovule, embryo very small. mesophyll with spherical etherial oil cells. Inflorescence spadix or spike. Flow ers bracteate; minute to small, usually bisexual, sometimes unisexual. Perianth absent. Androecium with 1 10 stamens, adnate to the base of ovary or not, free, of ten more or less monadelphous, staminodes often present, anthers bithecous (mono thecous in Peperomia), dehiscence by longitudinal slits, extrorse, tapetum gland ular, pollen grains monosulcate or nonaperturate. Gynoecium with 2-4 united carp els, or single carpel (Peperomia), ovary superior, unilocular, stigmas 1 5, placen tation basal; ovule, ascending, orthotropous, bitegmic or unitegmic (Peperomia), crassinucellate. Fruit fleshy, usually a drupe; seeds scantily endospermic, per isperm copious, embryo minute. Major genera: Piper (970 species), Peperomia (961), Ottonia (70) and Pothomorphe (10). Description: Herbs, shrubs, or woody climb-ers, or small trees bearing essential oils, stems conspicuously jointed, nodes 3-lacunar to multilacunar, vascular bu ndles scattered, vessel-elements with scalariform or simple end-walls, sieve-tub e plastids S-type. Leaves alternate, spiral, herbaceous or fleshy, sim-ple, enti re, pinnately or palmately veined, petiolate sheathing, stipules intrapetiolar, adnate to petiole, hydathodes commonly present, stomata cyclocytic or anisocytic , Economic importance: Piper nigrum is the source of black and white pepper (ripe and 438 Plant Systematics

~i

B D Figure 13.15 Piperaceae. Piper guineense. A: Fruiting branch with pendulous spik e; B: Paired flowers and their bracts; C: Longitudinal section of female flower. Piper nigrum . D: Fruiting branch; E: Flower; F: Stamen. G: Peperomia griseo-argenteum, a clu ster of flowering shoots. unripe, respectively). The roots of P. methy-stichum are used for making the fam ous national beverage Kava in Polynesia. The leaves ( Paan leaves) of P. betle (bet el-vine) are used as masticatory in East Africa, India and Indonesia. Some speci es of Peperomia (P. caperata, P. hederaefolia, P. magnoliaefolia) are grown as o rnamental foliage plants. Piperales together with Aristolochiaceae, Hydnoraceae and Lactoridaceae. Piperac eae, like Saururaceae, are also monophyletic (Tucker et al., 1993). Peperomia is considered to be the most derived member of the family with numerous apomorphie s such as single carpel, monothecous anthers, unitegmic ovule, inaperturate poll en grains and succulent leaves, and is often removed to a distinct family Pepero miaceae. Thorne (2003, 2006, 2007) and Stevens (2008) place Peperomia under sepa rate subfamily Peperomioideae, whereas the other 4 genera are placed under Piper oideae. Phylogeny: Piperaceae, together with Saururaceae, constitute a monophyletic grou p often considered as order Piperales. Thorne had earlier (1999, 2000) placed th ese families under suborder Piperineae of the order Magnoliales, but has subsequ ently (2003, 2006, 2007) shifted them under * * * * * * * * * * * Major Families of Angiosperms 439 Subclass 3. Alismatidae (B) Superorder 1. Acoranae (A) Order 1. Alismatales Order 1. Petrosaviales (B) 1. Butomaceae 2. Limnocharitaceae 3. Alismataceae Family 1.Petrosaviaceae 2. Nartheciales (B) 4. Hydrocharitaceae 1.Tofieldiaceae (B) 2. Potamogetonales 2. Nartheciaceae (B) 1. Aponogetonaceae 2. Scheuchzeriaceae 3. Posidoniaceae 4. Cymodoceaceae 5. Ruppiaceae 6. Juncaginaceae 7. Potamogetonaceae 8. Zosteraceae 9. Zannichelliaceae 3. Acorales 1. Acoraceae Superorder 2. Aranae Order 1. Arales 1. Araceae Superorder 3. Alismatanae Acoraceae Martynov Acorus family 1 genus, 2 species North temperate region, Paleotropical. Frigid zone, temperate, and subtropical.

Celebes and New Guinea, Eastern Asia to Norway approaching the Arctic circle, Ce ntral and Western North America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Liliidae Alismatidae Aranae Acoranae Subclass Arecidae Aridae Series+/Superorder Nudiflorae+ Aranae Monocots* Order Arales Acorales Arales Acorales Acorales B & H under family Aroideae Salient features: Marshy herbs with rhi-zomes, lacking oxalate crystals, inflore s-cence a spadix without spathe, flowers small, bisexual, tepals 6 in two whorls , sta-mens in two whorls, carpels 3, united, fruit a berry. Description: Aromatic perennial marshy herbs with rhizomes, bearing essential oi ls, root xylem with vessels with scalariform end-walls. Leaves alternate, distic hous, flat, sessile, sheathing, entire, parallel-veined, mesophyll with spherica l etherial oil cells, lacking calcium oxalate crystals. Inflorescence scapigerou s, spadix without spathe. Flowers ebracteate, small, bisexual, Major genera: Single genus Acores with 2 species. 440 Plant Systematics

Ci 9 D C r J A Figure 13.16 Acoraceae. Acorus calamus. A: Rhizome with basal leaves; B: Spadix with subtending leaf; C: Flower; D: Gynoecium; E: Transverse section of ovary; F: Longitudin al section of ovary; G: Stamen. regular, 3-merous, cyclic. Perianth with 6 tepals, free, in two whorls, concave or hooded, similar, membranous. Androecium with 6 stamens, free, in two whorls, anthers basifixed, dehiscence by longitudinal slits, introrse, tapetum glandular , pollen grains monosulcate to sub-ulcerate. Gynoecium with 3, rarely 2 or 4 car pels, united, ovary superior, 3-locular (rarely with 2 or 4 locules), placentati on axile, ovules 2 4( 5) per locule, pendulous, orthotropous, bitegmic. Fruit fleshy , berry; seeds endospermic, perisperm present, cotyledon 1, no double fertilizat ion. Gymnostachys. Grayum (1987) justified removal of Acorus from Araceae, and this h as been followed in all recent classifications. Whereas Dahlgren and Cronquist i ncluded the family under the order Arales, Takhtajan and Thorne took it under a separate order Acorales. Thorne (1999) removed the order from Aranae and placed it under Acoranae, along with or-der Nartheciales. He has now (2003, 2006, 2007) added Petrosaviales (family Petrosaviaceae; regarded unplaced in monocots in AP G II ) to Acoranae. Subsequent molecular analyses portray it, alone or with Gymn ostachys, as the sister group of all other Monocotyledons. The rooting of Acorac eae as sister group of monocots is also supported by the studies of Chase et al. , (2000), Soltis et al., (2000) and Fuse & Tamura (2000), based on multigene ana lyses. APG II and APweb, accordingly place Acoraceae under a separate order Acor ales at the beginning of monocots. Stevens (APweb, 2008), has also recognized Pe trosaviales. Economic importance: Oleum calami is distilled from the rhizomes of A. calamus, fo r use in perfumery and medicine. Phylogeny: The genus was earlier included under family Araceae. Hutchinson (1973 ) placed it under tribe Acorae along with * * * * * * * * * * * Major Families of Angiosperms 441 Araceae A. L. de Jussieu Arum family 109 genera, 2,755 species Throughout world but mainly in tropical and subtropical regions, very common in tropical forests and wetlands, a few species in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb)

Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Aridae Liliidae Alismatidae Series+/Superorder Nudiflorae+ Aranae Aranae Aranae Monocots* Order Arales Arales Arales Arales Alismatales B & H as family Aroideae Salient features: Terrestrial erect or climb-ing or aquatic herbs with rhizomes or corms, leaves often large, mucilaginous, inflores-cence scapigerous, a spadix subtended by a large spathe, flowers very small, reduced, usually unisexual, fr uit a berry or utricle. spathe (spathe absent in Gymnostachys, Orontium). Flowers very small, sessile, v ery rarely subsessile (Pedicellarum), unisexual, or bisexual, ebracteate, often fragrant, or malodorous; regular to very irregular. Perianth lacking or with 4-6 tepals (rarely 12), free or connate, usually in two whorls, green. Androecium w ith 1 (Cryptocoryne) 6( 12) stamens, free or with connate filaments, usually in two whorls, anthers basifixed, dehiscence by pores, longitudinal slits, or transver sely, extrorse, tapetum amoeboid, pollen grains aperturate, or nonaperturate. Gy noecium with 2-3 carpels, rarely up to 8 carpels, united, ovary superior, usuall y unilocular, 1 5 ovuled, placentation apical, or marginal, rarely mutilocular wit h axile placentation, ovule orthotropous or anatropous, bitegmic, tenuinucellate (Pistia), or crassinucellate. Fruit usually a berry or drupe, rarely a utricle or capsule; seeds endospermic, or nonendospermic, cotyledon 1. Pollination mainl y by insects, especially beetles, flies and bees. Dispersal of berries by birds and animals.. Major genera: Anthurium (900 species), Philodendron (500), Arisaema (150), Amorp hophalus (100), Pothos (55), Dieffenbachia (40) and Syngonium (30). Description: Terrestrial or aquatic herbs, sometimes epiphytic or climbing (Poth os, Syngonium), usually with rhizomes, or corms, sometimes free floating (Pistia ), usu-ally mucilaginous, raphide crystals of cal-cium oxalate, containing chemi cals caus-ing irritation of mouth (or temporary dumb-ness: Dieffenbachia, the du mbcane), vessels absent in stem and leaves, sieve-tube plastids P-type, subtype II; roots with vessels having scalariform end-walls, rarely with velamen Leaves small to very large with sheathing base, alternate, spiral, or distichous, petio late or sessile (Pistia), with parallel, pinnate or palmate venation, often cord ate, or hastate, or sagittate; stipules absent; stomata paracytic, tetracytic, c yclocytic, or anomocytic. Inflorescence scapigerous, a spadix consisting of a de nse spike subtended and enclosed by a large Economic importance: The family furnishes numerous horticultural ornamentals suc

h as Pothos, Alocasia, Arum, Dieffenbachia, Monstera, Philodendron, Zantedeschia and Syngonium. Epipremnum aureum (money plant) is commonly grown as house plant . 442 Plant Systematics i AL

c G L O K Figure 13.17 Araceae. Arum maculatum. A: Plant with flowering spadix; B: Vertica l section of spadix and spathe; C: Spadix; D: Gynoecium; E: Fruit cut to show seeds; F: Seed. Pistia stratiotes. G: Inflorescence; H: Vertical section of inflorescence; I: L ongitudinal section of ovary; J: Longitudinal section of orthotropous ovule with placent al hairs; K: Portion of androecium; L: Seed. Cala palustris. M: Inflorescence; N : Mature fruits; O: Transverse section of carpel. The corms of Colocasia esculenta (taro or dasheen), Amorphophallus campanulatus (El-ephant-foot yam), Cryptosperma and Xanthosoma (tanier, yautia) and fruits of Monstera (Mexican breadfruit) are used as food. subclass Aridae; Dahlgren and Cronquist also includes Acoraceae under Arales). A PG II and APweb place Acoraceae under separate order, but include Araceae along with several others under order Alismatales, Lemnaceae being merged with Araceae . The family is considered to be a fairly early divergent lineage within monocot s and sister to remaining families (as circumscribed by APG II) of Alismatales. Thorne (2003) recognized 7 subfamilies under Araceae: Gymnostachyoideae (1 genus ), Orontioideae (3 genera), Pothoideae (4 genera), Monsteroideae (12 genera), La sioideae (10 genera), Calloideae (1 genus), and Aroideae (73 genera), treating L emnaceae as a distinct family. APweb (2003, 2008) also recognises 7 subfamilies but Monsteroideae is merged with Pothoideae and Lemnoideae (based on Phplogenp: The family is considered to be monophyletic. Hutchinson believed it h ad been derived from the tribe Aspidistreae of Liliaceae. The bisexual flowers a re considered to be more primitive, while those with unisexual flowers to be mor e highly evolved. Most recent classifications place this family under the order Arales along with Lemnaceae, Acoraceae having been removed to a separate order A corales. Arales has also been placed under independent superorder Aranae in thes e classifications (Takhtajan shifts Aranae under separate Major Families of Angiosperms 443

Lemnaceae) included instead. Thorne (2006, 2007) has followed this change, and i ncluded two additional subfamilies Philodendroideae (27 genera) and Schismatoglo ttidoideae (7 genera), both segragates from Aroideae (Keating, 2003a,2003b, 2004 a, 2004b), thus recognizing nine in all. Mayo et al., (2003) on the analysis of five plastid genes did not find a clear resolution of affinit ies within the family. However, the basal clade (Gymostachydoideae + Orontioidea e) remains the same, and Lemnoideae are strongly supported as sister to the rest of the family. * * * * * * * * * * * Butomaceae Richard Flowering rush family 1 genus, single species Butomus umbellatus North temperate region, widespread in Asia and Europe, naturalized in tropical A merica. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Alismatidae Alismatidae Liliidae Alismatidae Series+/Superorder Apocarpae+ Alismatanae Alismatanae Alismatanae Monocots* Order Alismatales Butomales Alismatales Alismatales Alismatales B & H under family Alismaceae Salient features: Aquatic or marsh plants with linear triquetrous leaves, inflor escence scapigerous, umbellate cymes, perianth in two whorls, outer tinged green , stamens 9, carpels free, fruit etaerio of follicles. tepals, free, 2 whorled, whorls similar or outer sepaloid and inner petaloid, ou ter often tinged with green. Androecium with 9 stamens, 2 whorled (6+3), anthers basifixed, dehiscence by longitudinal slits, latrorse, pollen grains monosulcat e. Gynoecium with 6 carpels, free or connate at base, ovary superior, carpel inc ompletely closed, style short, with ventral decurrent stigmatic region, ovules 2 0 100, placentation dispersed, stigma papillate. Fruit an aggregate of follicles, seeds nonendospermic, embryo straight (not curved), cotyledon 1. Major genera: Butomus (1 species).

Description: Aquatic or marshy plant with basal leaves, rhizomatous, secretory c avities present, with latex, root xylem with vessels with simple and scalariform end-walls, stem xylem without vessels, sieve-tube plastids P-type, subtype II. Leaves emergent, alter-nate, distichous, petiolate or sessile, sheathing, simple , entire, linear, tri-quetrous, parallel-veined, stomata paracytic. Inflorescenc e scapigerous, umbellate cymes, involucral bracts three. Flowers medium-sized, o n long pedicels, trimerous, regular, bisexual, cyclic. Perianth with 6 Economic importance: Cultivated as an ornamental. The rhizomes are edible when b aked. Phylogeny: The genus was earlier placed under the family Alismaceae by Bentham 444 Plant Systematics

F E

r G u

A D Figure 13.18 Butomaceae. Butomus umbellatus. A: Rhizome with basal leaves; B: Um bellate cymose inflorescence with involucral bracts; C: Vertical section of flower; D: C arpels; E: Longitudinal section of carpel showing scattered ovules; F: Stigmas; G: Seed. and Hooker. Buchenau in Engler s Das Helleboraceae, and the peculiar placentation Pjlanzanreich (1903), recognized the family of the ovules dispersed all over the surface Butomaceae to include all members of of the carpel more ancient characte r than Alismaceae (this original name of the fam- found in any herbaceous dicoty ledon except ily has now been replaced by Alismataceae) in Cabombaceae, which is similar in this with numerous ovules and parietal respect. Cabombaceae also hav e trimerous placentation. Pichon (1946) redefined the flowers and aquatic habit, and Butomaceae circumscription of these two families to shift is separated only on the basis of a single all genera with petioled leaves with ex- cotyledon and absence of endosperm. The panded blades, campylotropous ovules, and recent clad istic studies reveal this family seeds with curved embryos, to Alismaceae to be closer to Hydrocharitaceae. Accord-and retaining only genus Butomus under ing to Judd et al., (2002) Butomaceae, Butomaceae, a treatment followed in most Hydroc haritaceae and Alismataceae form of the recent publications. According to one aq uatic clade of Alismatales, supported Hutchinson (1973), who retained a broader by the apomorphies of perianth differenticircumscription of the family, the ated into sepals and petals, stamens more gynoecium of Butomaceae represents prob- t han six and carpels more than three, and ably the most ancient type of the the o vules scattered over the inner surmonocotyledons, the free carpels recalling fac e of locules. * * * * * * * * * * * Major Families of Angiosperms 445 Alismataceae Ventenat

Water Plantain family 13 genera, 80 species (excluding Limnocharitaceae) Throughout world, plants of fresh water marshes, swamps, lakes, rivers and strea ms. Majority of the species found in the New World. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Alismatidae Alismatidae Liliidae Alismatidae Series+/Superorder Apocarpae+ Alismatanae Alismatanae Alismatanae Monocots* Order Alismatales Alismatales Alismatales Alismatales Alismatales B & H as family Alismaceae Salient features: Aquatic or marsh plants with laticifers, leaves petiolate with well-developed blade, inflorescence scapigerous, perianth in two whorls, differ entiated into sepals and petals, stamens 6 to many, car-pels 6 to many, free, ov ule usually one per carpel, fruit an etaerio of achenes, embryo curved. with 6 stamens, rarely more with branched outer stamens, free, anthers bithecous , dehiscence by longitudinal slits, extrorse, pollen grains usually 2 3 aperturate . Gynoecium with 3 carpels, rarely more, free, ovary superior, 1 ovuled, placent ation basal, ovule anatropous, or amphitropous. Fruit etaerio of achenes, seeds nonendospermic, cotyledon 1, embryo strongly curved. Major genera: Echinodorus (35 species), Sagittaria (25), Alisma (9) and Burnatia (3). Description: Aquatic or marshy plant with basal leaves, rhizomatous, laticifers present with white latex, root xylem with vessels having scalariform to simple e nd-walls, ves-sels absent in stem and leaves. Leaves sub-merged and emergent, of ten heterophyllous, alternate, petiolate or sessile, sheathing, simple, pinnatel y, palmately, or parallel-veined, stomata paracytic or tetracytic, ax-illary sca les present. Inflorescence scapigerous, paniculate, ultimate branches cymose or racemose, sometimes umbellate or even solitary, with or without involucral bract s. Flowers bracteate, bisexual or uni-sexual and monoecious, or dioecious (Burna tia), regular, trimerous, cyclic. Peri-anth differentiated into calyx and coroll a. Sepals 3, free, imbricate. Petals 3, free, white, red, or pink. Androecium us ually Economic importance: Sagittaria sagittfiolia (arrowhead) is cultivated in China and Japan for its edible corms. Several species of Sagittaria, Alisma (water pla

ntain), and Echinodorus (bur-heads) are cultivated as poolside plants and used a s aquarium plants. Phylogeny: The family has been redefined (Pichon, 1946) to shift all genera with laticifers, petioled leaves with expanded blades, campylotropous ovules, and se eds with curved embryos, including several genera formerly included under Butoma ceae. According to Judd et al., (2002) Butomaceae, Hydrocharitaceae and Alismata ceae form one aquatic clade of Alismatales, supported by the apomorphies of peri anth differentiated into sepals and petals, stamens more than six and carpels mo re than three, and the ovules scattered over the inner surface 446 Plant Systematics

A B Figure 13.19 Alismataceae. Alisma plantago-aquatica. A: Basal part of plant with leaves; B: Inflorescence; C: Flower; D: Outer tepal; E: Inner tepal; F: Achene. Sagittaria sagit tifolia. G: Plant with sagittate leaves and basal part of scape; H: Inflorescenc e; I: Male flower with petals removed; J: Petal; K: Stamens in different views; L: Achene; M: Carpel.(After Sharma and Kachroo, Fl. Jammu, vol. 2, 1983) of locules. The genera with achenes (Sagittaria, Alisma, Echinodorus, etc.) may form a monophyletic group (Chase et al., 1993). The family is often regarded as primi-tive due to numerous stamens and carpels. The developmental and anatomical studies have, however, indicated that these numer-ous stamens are due to second ary increase, from ancestral condition of six stamens in two whorls. According to Hutchinson ( 1973) the family reminds of Ranunculaceae and but for solitary cotyledon and lac k of endosperm, the genus Ranalisma might be well be placed in Ranunculaceae. Ac cording to Soros & Les (2002) Echinodorus is polyphyletic and evidently needs sp litting. * * * * * * * * * * * Major Families of Angiosperms 447 Hydrocharitaceae A. L. de Jussieu Tape Grass family 18 genera, 110 species Throughout the world, mostly tropical and subtropical regions in fresh water and marine habitats. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Alismatidae Alismatidae Liliidae Alismatidae

Alismatanae Alismatanae Alismatanae Series+/Superorder Microspermae+ Monocots* Order Hydrocharitales Hydrocharitales Alismatales Alismatales Alismatales B & H as family Hydrocharideae Salient features: Aquatic fresh water or marine herbs, leaves submerged, usually ribbon-like, flowers subtended by paired bracts, male flowers often detaching a nd floating on water, carpels united, ovary in-ferior with many scattered ovules , fruit a capsule or berry. with connate filaments, anthers bithecous, dehiscence by longitudinal slits, inn ermost times modified into staminodes which of-ten act as sails (Lagarosiphon), pollen grains monosulcate or inaperturate, sometimes united into thread-like cha ins (Thalassia, Halophila). Gynoecium with 3-6 (rarely 15), connate carpels, sin gle in Najas, ovary in-ferior, unilocular, ovules many (1 in Najas), scattered o ver the surface, placentae often deeply intruded, styles often divided and twice the number of carpels, stigmas elongate and papillose. Fruit fleshy berry or dr y capsule rupturing irregularly, nut in Najas; seeds without endosperm, embryo s traight, cotyledon 1. Pollination in some (Vallisneria, Enhalus) by water, in ot hers (Egeria, Limnobium) by insects. Dispersal by water or animals. Major genera: Ottelia (32 species), Najas (32), Elodea (12), Vallisneria (8), Hy drocharis (6), Halophila (4) and Hydrilla (1). Description: Aquatic herbs, submerged or partly emergent, rooted in mud or unattached, in freshwater and marine habitats, annual or perennial. Leaves alternate (Nechamandra), opposite (Elodea some spe-cies) or whorled (Hydrilla, Lagarosiph on), in basal rosettes or cauline, simple, entire or serrate, with parallel or p almate venation, sheathing at base, small scales, at nodes inner to leaf base, s tipules absent. Inflo-rescence with solitary flower (female) or short cymes (usu ally male), subtended by two often connate bracts. Flowers bisexual or unisexual , male flowers often discon-nected and floating on water surface (Vallisneria, E nhalus, Lagarosiphon). Peri-anth often with distinct sepals and petals. Sepals 3 , free, valvate, green. Petals 3, free, usually white, imbricate, sometimes lack -ing (Thalassia, Halophila). Androecium with 2 to 3 stamens, rarely more (Egeria ), free Economic importance: Species of many genera (Hydrilla, Vallisneria, Elodea, Eger ia, etc.) are used as aquarium plants. Some species like Hydrilla verticillata, Elodea canadensis have become troublesome weeds in many parts of the world. Phylogeny: The family, along with Butomaceae and Alismataceae, forms a well-defi ned clade, as indicated by cladistic analysis. Although monophyletic (Dahlgren a nd Rasmussen, 1983), the family is morphologically heterogenous and divided into 448 Plant Systematics

D r ~ 1

a N,

Figure 13.20 Hydrocharitaceae. Vallisneria spiralis. A: Plant with creeping stem , strap-shaped leaves and female flowers on coiled long pedicels; B: Male flower which detaches and floats on water; C: Female flower; D: Vertical section of female flower; E: Male inflorescence subtended by two connate bracts; F: Transverse section of ov ary with parietal placentation. Ottelia cordata. G: Plant with leaves and emergi ng flower bud; H: Male inflorescence with spathe opened out; I: Female flower; J : Male flower with perianth removed, showing stamens and pistillodes; K: Transve rse section of ovary. 3 (Hutchinson, 1973; Thorne, 2003: Hydrocharitoideae, Thalassioideae and Halophi loideae) to 5 subfamilies (Dahlgren et al., 1985). Les et al., (1997) concluded that the family forms a well-defined lineage. Tanaka et al., (1997) suggest a se ries of quite well-supported nodes based on analysis of variation in two genes, the ultimate group-ings recognized are similar to those of Les et al. (1997). Th e family Najadaceae, though distinct with single carpel, single ovule and superi or ovary, has been included in Hydrocharitaceae by Thorne, APG II and APweb. It is possible Najas may be sister to the rest of Hydrocharitaceae, in which case it may probably be recognized as a separate family. APweb recognises 7 well defined groups (lineages) within Hyd rocharitaceae as established by the studies of Les et al., (1997). Thorne (2006, 2007) has revised the classification of the family recognising 4 sub-families: Hydrocharitoideae (2 genera), Stratiotoideae (1 genus), Anacharidoideae (7 gener a) and Hydrilloideae (8 genera). * * * * * * * * * * * Major Families of Angiosperms 449 Potamogetonaceae Dumortier Pondweed family 3 genera, 90 species

Throughout world, in ponds, ditches and lakes. (Excluding Ruppiaceae) Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Magnoliopsida Liliidae Alismatidae Alismatanae Alismatanae Subclass Alismatidae Alismatidae Series+/Superorder Apocarpae+ Alismatanae Monocots* Order Najadales

Liliopsida

Potamogetonales Najadales Potamogetonales Alismatales B & H under family Naiadaceae FAI B E

H a 0 D 1 Figure 13.21 Potamogetonaceae. Potamogeton perfoliatus. H: Plant with flowering and fruiting inflorescences; B: Leaves; C: Flower; D: Tepal with attached stamen; E: Carpel; F: Anther in two different views; G: Fruit; H: Seed enclosed in hard endocarp; I : Seed. (After Sharma and Kachroo, Fl. Jammu, vol. 2, 1983). Salient features: Aquatic herbs, leaves sub-merged as well as floating, flowers usually in spikes, bisexual, perianth with 4 free tepals, stamens 1-4, carpels 4 , free, fruit etaerio of achenes. Major genera: Potamogeton (83 species), Coleogeton (6) and Groelandia (1). Description: Perennial or rarely annual fresh water herbs with rhizomes, stems 450 Plant Systematics

mostly submerged, with reduced vascular bundles in a ring, with air cavities, ta nnins often present, root xylem with vessels hav-ing scalariform end-walls, stem without ves-sels, sieve-tube plastids P-type, subtype PII. Leaves submerged as well as floating, sheathing at the base, alternate or opposite (Groenlandia), si mple, entire, venation par-allel, submerged leaves thin, without cuti-cle and st omata, floating leaves thick, small scales present at nodes inner to the leaf sh eath. Inflorescence terminal or axillary spike, often carried on a long peduncle , raised above water surface, peduncle sur-rounded by sheath at base. Flowers eb racteate, regular, bisexual, hypogynous, cyclic. Perianth with 4 tepal (often in ter-preted as appendages from connective of the anthers and thus perianth absent ), free, fleshy, usually clawed. Androecium with 4 stamens, free, adnate to and opposite each tepal, anther sessile, dehiscence by longi-tudinal slits, pollen g rains globose, inaperturate. Gynoecium with 4 free carpels, ovary superior with basal to apical placentation, ovule 1, campylotropous, bitegmic, crassinucellate , style short or lacking, truncate to capitate. Fruit an etaerio of achenes or d rupes, seeds nonendospermic, with starch, cotyledon 1, embryo slightly curved. P ollination by wind, dispersal by water or animals. Phylogeny: The affinities of the family are not clear. The family is sometimes a lso in-terpreted to include Ruppia and/or Zannichelia. The studies of Les et al. , (1997) however, have shown that Zannichelia is rather weakly embedded in the f amily, and the inclusion of Ruppia makes family biphyletic. Potamogeton itself i s considered para- or polyphyletic (Les and Haynes 1995). Uhl (1947) had earlier supported the segregation of Zannichelia into separate family by Hutchinson (19 34). The tepals are often interpreted variously. Uhl suggested that so-called pe rianth parts are in fact in-dividual bracts, subtending and adnate to stamens an d the flower is fundamentally an inflorescence with staminate flowers (each repr esented by a monbracteate perianth) and naked female flowers, the view first pro posed by Kunth (1841) and sup-ported by Miki (1937). Most of the authors (Rendle , 1925; Watson and Dallwitz, 2000; Judd et al., 2008) consider tepals to represe nt appendages from the connective of the anther. As per Hutchinson (1973), these are normal perianth-segments on claws of which the extrorse anthers are sessile , a view also supported by Heywood (1978) and Woodland (1991). Hutchinson stress ed that in the petaloid monocotyledons the stamens are always opposite the peria nthsegments, and it is not a great step for those species of Aponogeton with mor e than one perianth-segment to Potamogeton. If the anther were introrse in Potam ogeton, then the petal-like organ might be regarded as an outgrowth from the bas e of the connective, a very unusual feature indeed in any flowering plant. Economic importance: The family is of lit-tle economic importance but biological ly an important source of food for aquatic life. Many species of Potamogeton are trouble-some weeds. Fleshy starchy rootstocks are sometimes used as food. * * * * * * * * * * * Major Families of Angiosperms 451 Subclass 4. Lili idae 2. Asteliaceae 3. Hypoxidaceae (B) Superorder 1. Pandananae 4. Lanariaceae Order 1. Pandanales 5. Blandfordiaceae 6. Orchidaceae Family 1. Velloziaceae 3. Iridales 2. Acanthochlamydaceae (B) 3. Pandanaceae 4. Cyclanthaceae 5. Pentastemonaceae 6. Stemonaceae Suborder 1. Iridineae

1. Doryanthaceae 2. Tecophilaeaceae (B) 3. Ixioliriaceae 7. Triuridaceae 4. Iridaceae Superorder 2. Dioscoreanae (B) 2. Asphodelineae (B) Order 1. Dioscoreales (B) 1. Xanthorrhoeaceae 2. Xeronemaceae (B) 3. Asphodelaceae (B) 4. Hemerocallidaceae (B) 1. Dioscoreaceae 2. Burmanniaceae 3. Thismiaceae (B) 5. Johnsoniaceae Superorder 3. Lilianae (B) 3. Hyacinthineae Order 1. Liliales (B) 1. Anthericaceae (B) 2. Alliaceae 3. Hyacinthaceae 4. Themidaceae 5. Behniaceae (B) 6. Anemarrhenaceae (B) 7. Herreriaceae (B) 1. Corsiaceae (B 2. Campynemataceae 3. Melanthiaceae 4. Trilliaceae 5. Petermanniaceae (B) 6. Luzuriagaceae (B) 7. Alstroemeriaceae (B) 8. Colchicaceae (B) 9. Riponogaceae 10. Philesiaceae (B) 11. Smilacaceae 8. Agavaceae 4. Asparagineae 1. Aphyllanthaceae 2. Laxmanniaceae 3. Asparagaceae 12. Liliaceae 4. Dracaenaceae 2. Orchidales (B) 1. Boryaceae (B) 5. Nolinaceae 6. Eriospermaceae Pandanaceae R. Brown Screw pine family 4 genera, 910 species Throughout tropics and subtropics of the Old World, mostly in coastal and marshy areas. Placement:

B & H Cronquist

Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Aridae Liliidae Liliidae Series+/Superorder Nudiflorae+ Pandananae Pandananae Pandananae Monocots* Order Pandanales Pandanales Pandanales Pandanales Pandanales 452 Plant Systematics E B Figure 13.22 Pandanaceae. Pandanus ceylonicus. A: Plant with leaves and female i nflorescence; B: Portion of leaf showing marginal prickles; C: Male inflorescence; D: Male flo wer with apiculate anthers; E: Female spadix split to show arrangement of flower s; F: Female flower with curved style; G: Longitudinal section of female flower; H: Drupe tipped by persistent style; I: Seed.(After Dassanayake, Fl. Ceylone, v ol. 3, 1981). Salient features: Large shrubs, trees or climbers with annual scars of leaf base s, bearing aerial roots, leaves 3-ranked, stiff, inflorescence a spadix with uni sexual flow-ers, flowers naked, male flowers with many stamens, carpels many, fr ee or united, fruit scalariform end-walls. Leaves forming terminal crown, 3-ranked or 4-ranked, some times appearing spirally arranged due to twisting of stem, long, narrow, usually stiff or sword-like, sheathing at base, keeled, often spinose along margin and keel, sometimes even grass-like. Inflorescence a spadix subtended by a brightly coloured spathe, and usually containing one type of flowers, male and female flo wers being borne on separate plants (plants dioecious), spadix lacking in Sarara nga and inflorescence paniculate. Flowers sessile, pedicellate in Sararanga, uni sexual, without perianth, hypogynous. Perianth absent or vestigial, sometimes fo rming a short cupule (Sararanga). Androecium in male flower with numerous a berry or multilocular drupe, often aggre-gated into a cone appearing like pine apple. Major genera: Pandanus (750 species), Freycinetia (123) and Sararanga (2). Description: Trees, shrubs or climbing perennials (Freycinetia), supported by ae rial roots, roots often penetrating supporting host (Freycinetia) or even absent (Sararanga), trunk bearing annual scars of leaf bases, stem and leaves also wit h xylem with Major Families of Angiosperms 453

stamens, filaments free or connate, anthers erect, bithecous, basifixed, dehisce nce by longitudinal slits, staminodes often present (Freycinetia) in female flow er. Gynoecium in female flower with many carpels, free or connate, ovary superio r, unilocular (if car-pels free) or multilocular (if united), ovules 1 (Pandanus ) to many (Freycinetia), anatropous, style short or absent, stigma nearly sessil e, rudimentary ovary often present in male flower. Fruit a berry or multilocular drupe, often aggregated to form oblong or globose syncarps resembling a pineapp le; seeds small with fleshy endosperm and minute embryo. poutpourris. Freycinetia banksii and Pandanus veitchii are used as ornamentals. Economic importance: Several species of Pandanus (screw pine) are useful sources of food. P. leram (Nicobar breadfruit) produces a large globose fruit which is boiled in water. Other species also yielding edible fruits in-clude P. utilis an d P. andamanensium. The leaves of P. odoratissimus are used for thatch-ing and w eaving. Fibres made from aerial roots are used for cordages and brushes. The flo wers of this species are used for popular Indian essence Kewra. The fragrant lea ves of P. odorus in Malaya are used in Phylogeny: Pandanaceae forms a well de-fined clade included under Pandanales eit her singly (Thorne, Takhtajan, Dahlgren and Cronquist, Hutchinson), or together with Cyclan-thaceae, Stemonaceae Velloziaceae, and Triuridaceae (APG II, APweb). The position of Cyclanthaceae does not show much departure as it has often been placed in the adjacent order. The placement of Triuridaceae within Pandanales i s, however, inter-esting. Takhtajan places it under a separate subclass, Cronqui st under Alismatidae, and Dahlgren under separate superorder Triuridanae. The pl acement of this family in Pandanales is supported by the studies of 18S rDNA (Ch ase et al. 2000). Thorne who had earlier (1999) placed Triuridaceae under Alisma tidae has finally (2007) shifted it under Liliidae > Pandananae >Pananales, not rec ognizing superorder Triuridanae. He has also made a major change in shifting Pan dananae to the beginning of Liliidae. He had earlier placed it towards the end o f the sub-class. Thorne (2003, 2007) recognizes two sub-families Pandanoideae an d Freycinetioideae. * * * * * * * * * * * Dioscoreaceae R. Brown Yam family 4 genera, 400 species Mainly tropical and subtropical, few in the temperate region. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Epigynae+ Dioscoreanae Lilianae Dioscoreanae Monocots*

Order Liliales Dioscoreales Dioscoreales Dioscoreales Dioscoreales Salient features: Woody or herbaceous climbers, leaves alternate, cordate, petio le with pulvinus at both ends, venation reticulate, inflorescence axillary racemes, spikes or umbels, flowers unisexual, fruit a capsule, seeds winged. 454 Plant Systematics Figure 13.23 Dioscoreaceae. Dioscorea esculenta. A: Male plant with alternate le aves; B: Male flower; C: Male flower opened. D. oppositifolia. D: Male plant with opposite lea ves and flowers; E: Female plant with fruits. Major genera: Dioscorea (375 species), Rajania (20) and Tamana (5). palmate, reticulate, stomata anomocytic, leaf axils often with bulbils. Inflores cence axillary panicles, racemes or spikes, flowers arranged singly or in 2-3 fl owered clusters. Flow-ers usually unisexual (plants dioecious), small, sessile o r rarely pedicellate, actinomorphic. Perianth with 6 tepals, in two whorls, free or connate at base into tube. Androecium with 6 stamens, in two whorls, attache d to the base of perianth, 3 sometimes reduced to staminodes, filaments free or slightly connate, anthers bithecous, connective sometimes broad; pollen grains m onosulcate or variously porate. Gynoecium with 3 carpels, united, ovary inferior , Description: Perennial herbaceous or woody climbers with tubers or rhizomes, a f ew dwarf shrubs, usually twining over the support, stem with vascular bundles in one or two rings. Leaves usually alternate, sometimes opposite (Dioscorea alata ) simple, cordate, sometimes palmately lobed or compound (D. pentaphylla), petio late, petiole with pulvinus both at base and above, sometimes with stip-ule like flanges on both sides, sometimes with superficial or internal glands contain-in g nitrogen fixing bacteria, venation Major Families of Angiosperms 455 trilocular with axile placentation, ovules 2 (rarely many) in each locule, style s 3, free or connate. Fruit a 3-valved capsule or berry, rarely samara; seeds us ually flattened and winged, with endosperm and small em-bryo, often with second scaly cotyledon, seed coat with yellow-brown to red pigments, and crystals. related families under Dioscoreales. Cronquist had earlier included it under bro adly circumscribed Liliales. The circumscription of Dioscoreales has been narrow ed in APG II and APweb to include only Nartheciaceae, Dioscoreaceae and Burmanni aceae, shifting Smilacaceae and Trilliaceae (latter under Melanthiaceae) to Liliales and Stemonaceae to Pandanales, but Economic importance: Many species of have merged Taccaceae, Stenomeridaceae Dios corea are cultivated for starchy tubers and Trichopodaceae with Dioscoreaceae (b oth (Yam). Some species are source of diosgenin, regarded as distinct families b y Thorne). The a steroidal sapogenin developed in recent narrowly circumscribed Dioscoreales is years for its use in oral contraceptives. monophyletic as supported by morphological Phylogeny: The family is often placed under order Dioscoreales. Dahlgren (1985) also in-cluded Smilacaceae and Trilliaceae and con-sidered Dioscoreales to repre sent primitive monocots. The primitive position of the or-der (also advocated by Stevenson and Loconte, 1995), is not supported by cladistic analysis based on r bcL sequence and morphological data (Chase et al., 1995). Thorne has included Di oscoreaceae along with Taccaceae under order Taccales, whereas Takhtajan took Ta ccaceae under monotypic order Taccales, keeping Dioscoreaceae along with closely and rbcL sequence evidence (Chase et al., 1995), although the Placement of Narth eciaceae received poor support in the recent studies of Chase et al. (2000) and Caddick et al., (2002a, 2002b). Thorne who had earlier (1999, 2000) placed Tacca

les after Liliales under Lilianae, removed (2003) under distinct superorder Tacc anae placed before Lilianae, but after Pandananae, which has been shifted to the beginning of Liliidae. In 2006, 2007 he has preferred the name Dioscoreanae for Taccanae and Dioscoreales for Taccales, as he has merged Taccaceae and Stenomer idaceae with Dioscoreaceae. * * * * * * * * * * * Smilacaceae Ventenat Catbrier family 3 genera, 320 species Mainly tropical and subtropical, extending into the temperate region. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Coronariae+ Dioscoreanae Lilianae Lilianae Monocots* Order Liliales Smilacales Dioscoreales Liliales Liliales B & H under family Liliaceae 456 Plant Systematics F

c Figure 13.24 Smilacaceae. Smilax aspera. A: Portion of plant with leaves, stipul e tendrils and inflorescence; B: Male flower; C: Tepal with stamen; D: Female fl ower; E: Transverse section of ovary; F: Infrutescence with berries. Salient features: Woody or herbaceous climbers, climbing by stipular tendrils, s tems sometimes prickly, leaves alternate, vena-tion reticulate, inflorescence ax illary ra-cemes, spikes or umbels, flowers unisexual, fruit 1-3 seeded berry. Perianth with 6 tepals, equal or subequal, in two whorls, free or united into a tube (Heterosmilax). Androecium with 6 stamens (rarely 3 or 9), free or united, anthers monothecous due to confluence of two locules, introrse, pollen grains in aperturate or monosulcate, staminodes present in female flower. Gynoecium with 3 united carpels, ovary superior, 3-locular, 1-2 ovules in each locule, placentat ion axile, ovules pendulous, orthotropous or semianatropous, stigmas 3. Fruit a berry with 1-3 seeds, embryo small, endosperm hard. Pollination by insects, disp ersal of fruits by birds. Major genera: Smilax (300 species), Heterosmilax (13) and Rhipogonum (7) Description: Herbaceous or woody climbers with paired stipular tendrils, stem pr ickly, with underground rhizomes or tubers. Leaves alternate or opposite, mostly leath-ery, petiolate, 3-veined, stipules (or leaf sheath) developing into tendr ils (Smilax), ve-nation reticulate. Inflorescence axillary raceme, spike or cyme . Flowers small, uni-sexual (plants dioecious) or bisexual (Rhipo-gonum), regula r, hypogynous, trimerous. Economic importance: Several species of Smilax are the source of sarsaparilla, u sed for treating rheumatism, and other ailments. The dried rhizomes of Smilax ch ina (China root) yield an extract used as a Major Families of Angiosperms 457

Figure 13.25 mb trichopoda. A: Plant; B: Flower. Cloranthaceae. C: Chloranthus glaber plant in flower. Calycanthaceae. D: Calycanthus occidental is, plant in flower. Nymphaeaceae. E: Nuphar polysepala, plant; F: Flower of N. lutea; G: Nymphaea odorata, plant in flower; H: Flower enlarged. 458 Plant Systematics

f TN

r~

~3 f Figure 13.26 Magnoliaceae. Magnolia grandiflora A: Tree in flower; B: Flower; C: Fruit; D: Leaf of Liriodendron chinense. Degeneriaceae. Degeneria vitiensis E: Plant; F: Fruit; G: Flower. Lauraceae. H: Laurus nobilis plant in flower; I: Plant of Neolitsea ser icea. Major Families of Angiosperms 459

Figure 13.27 Araceae. A: Amorphophallus titanum, unopened inflorescence; B: Part of fruiting inflorescence; C: Epipremnum aureum, plant; D: Anthurium andreanum, inflorescenc e. Alismataceae. E: Sagittaria sagittifolia, plant. Hydrocharitaceae. F: Hydrill a verticillata, plant. Liliaceae. G: Tulipa cultivar, flower; H: Stamens and pis til enlarged; I: Lilium longiflorum, plant in flower; J: Flower enlarged. 460 Plant Systematics

LI ~1 a

r,45MAN020

LM

Figure 13:28 Iridaceae. A: Dietes grandiflora plant; B: Flower enlarged; C: Iris germanica, flower. Asphodelaceae. D: Kniphofia thomsoni, plant; E: Part of inflorescence.; F: Aspho delus fistulosus, plant; G: Flower and fruit; Alliaceae. H: Clivia chrysanthifol ia, plant; I: Part of inflorescence; J: Allium cepa, plant; K: Inflorescence enl arged; L: Agapanthus praecox, inflorescence; M: Basal part with leaves and scape s. Major Families of Angiosperms 461 0 xa

G IO ~

Figure 13:29 Hyacinthaceae. Eucomis autumnalis plant with inflorescence; B: E. b icolor, inflorescence. Agavaceae. C: Yucca rupicola, plant with inflorescence; D: Part of inflor escence; E: Agave parrii, plant; F: A. wightii, inflorescence. Asparagaceae. G: Ophiopogon planiscaposus, plant; H: Ruscus aculeatus, plant; I: Flower; J: Aspar agus racemosus, fruit. Nolinaceae. K: Nolina recurvata, swollen base; L: N. nels oni, plant. 462 Plant Systematics

~a k

C 0 MIA I ,xAnMR IV51 X31

ld

t Figure 13:30 Arecaceae. A: Roystonea regia, plant; B: Trunk with leaf scars; C: Parajubaea coccoides, portion of plant with fruits; D: Caryota urens, plant; E: Inflorescence; F: Part in Fruit. Musaceae. G: Musa paradisiaca subsp. sapientum, plant with infloresce nc; H: Inflorescence with fruits. Commelinaceae. I: Tradescantia spathacea, plan t; J: Tradescantia pallida, plant; K: Flower. Cyperaceae. L: Cyperus alternifoli us, plant; M: Part of spike enlaged; N: Scirpoides holoschoenus, plant. Poaceae. O: Triticum aestivum, plant; P: Spike; Q: Portion of spike of Pennisetum glaucu m; R: Avena sativa, plant with inflorescence; S: Zea mays, male inflorescence. Major Families of Angiosperms 463

8 R IRE

~ J ~

L4

Wt0

01

Figure 13:31 Paeoniaceae. A: Paeonia suffruticosa, plant. Berberidaceae. B: Nand ina domestica, flowering branch; C: Branch with fruits. Papaveraceae. D: Bocconia glaucifolia, plant; E: Flower; F: Papaver orientale, plant; G: Flower; H: Romneya coulteri, p ortion of plant; I: Stamens and stigma. 464 Plant Systematics

r~ rr t

Figure 13:32 Ranunculaceae. A: Anemone occidentalis, plant; B: Fruit; C: Cimicif uga heracleifolia, plant; D: Clematis viticella, plant; E: Central part of flower. F: Caltha leptos epala, plant; G: Thalictrum polyganum, plant; H: Consolida ajacis; I: Flower; J: Ranunculus muricatus, plant; K: Flower; L: R. sceleratus, flower; M: Plant; N: Helleborus argutifolius, plant; O: H. orientalis, flower. P: Nigella damascena, plant. Major Families of Angiosperms 465 d It7w Figure 13.33 Grossulariaceae. A: Ribes menziesii var. leptosmum, plant with frui ts; B: Fruit. C: Ribes sanguineum var glutinosum, fruit. Fagaceae. D: Cyclobalanopsis glauca, portion of trunk; E: Vegetative branches. F: Lithocarpus densiflorus, plant in f lower. Nothofagaceae. G: Nothofagus obliqua, plant with fruits. Betulaceae. Betu la utilis. H: Bark; I: Branch with Fruit; J: Fruiting inflorescence; Branch of C orylus colurna. 466 Plant Systematics r

i~_ - tr3 .1 ~ rJ Figure 13.34 Portulacaceae. A: Portulacaria afra, plant. Cactaceae. B: Mammillar ia densispina, plant; C: Echinopsis terscheckii, plant; D: Portion of stem. Nyctaginaceae. E: B ougainvillea glabra, plant; F: flower enclosed in showy bracts; G: Mirabilis jal apa, plant; H: Flower. Aizoaceae. I: Mesembryanthemum criniflorum, plant; J: Flo wer. Major Families of Angiosperms 467 GAN

F 0 LJ OA

Figure 13.35 Chenopodiaceae. A: Chenopodium album, plant; B: Flower cluster. Ama ran e. C: Amaranthus caudatus, plant with inflorescence; D: Flower cluster; E: Celosia cristata, inflorescence. Caryophyllaceae. F: Stellaria media, plant; G: Flower; H: Dianthus barbatus, plant; I: Part of inflorescence. Polygonaceae. J: Rumex hy menosepalus, plant. K: Eriogonum latifolium, plant; L: Flower cluster; M: Polygo num davisae, portion of plant; 468 Plant Systematics

1!

A G

,W-11

J \ 11 Figure 13.36 Celastraceae. A: Euonymus grandiflorus, plant; B: Flowers. Violacea e. C: Viola tricolor, plant; D: Flower. Cucurbitaceae. E: Luffa cylindrica, plant; F: Female flower; G: Young fruit. Begoniaceae. H: Begonia foliosa, plant; I: B. sempervir ens, flower; J: Begonia Gene Daniels , flowers. Major Families of Angiosperms 469

13

A e

Clusiaceae. A: cum sp. , flower; Euphorbiaceae. D: Jatropha panduraefolia, plant; E: Female flo wer with fruit F: Male flowers G: Ricinus communis, plant; H: Male Flower I: Fem ale flower; J: Euphorbia milii, plant; K: Cyathia with showy bracts; L: E. pulch errima, plant; M: Phyllanthus emblica, branch in flower; Oxalidaceae. N: Oxalis spiralis, plant; O: Flower. ricum androsaemum, plant; B: H. monogynum, plant; C: Hyperi470 Plant Systematics

Figure 13.38 Geraniaceae. A: Pelargonium zonale, plant; B: Flower. Rosaceae. C: Prunus campanulata,

plant; D: Chaenomeles vilmoriana, plant; E: Flower; F: Cotoneaster microphyllus, fruit; G: Fragaria vesca, plant; H: Flower; I: Potentilla fruticosa, plant; J: Rubus trifidus, flower Major Families of Angiosperms 471

4.. it W. F

Figure 13.39 Fabaceae. A: Lupinus arboreus, plant; B: Clitoria ternatea, flower; C: Parkinsonia aculeata, plant; D: Poinciana pulcherrima, plant; E: Flower; F: Saraca asoka, plant with inflorescence; G: Senna candolleana, plant; H: Flower; I: Calliandra haematocephala, pla nt; J: Fruit; K: Leucaena leucocephata, plant; L: Fruits. 472 Plant Systematics a

za

K Cl

dYf~ 1 I Figure 13.40 Myrtaceae. A: Lophostemon confertus, plant; B: Flower; C: Myrceugen ella apiculata, branch; D: Metrosideros excelsa, inflorescence; Lythraceae E: Cuphea sp, floweri ng branch; F: Flower; G: Lagerstroemia speciosa, flowering twig. Onagraceae. H: Fuchsia hatschbachii, flowering branch; I: Flower; J: Fuchsia microphylla, flowe r. Major Families of Angiosperms 473 stimulant. Young stems and berries are sometimes used as food. ment under Liliales (Cronquist, 1988; Thorn; APG II; APweb). Cronquist also incl uded genera Luzuriaga, Petermannia and Philesia under the family Smilacaceae, bu t according to Chase et al., their inclusion makes Smilacaceae polyphyletic. The monophyly of the family (including only genera Smilax, Heterosmilax and Rhipogo num) is supported by morphological as well as molecular evidence. APG II and APw eb include Rhipogonum under a distinct family Rhipogonaceae (leaves opposite, po llen reticulate), whereas Judd et al. ,(2002) include it (Ripogonum) under Smila caceae. Thorne (2003) divided the family into two subfamilies Smilacoideae and R hipogonoideae, latter has been elevated to a distinct family in 2006, 2007 revis ions. Phylogeny: Earlier included under Family Liliaceae (Bentham and Hooker; Engler a nd Prantl), it was separated as a distinct family by Hutchinson (1934, 1973), ac cording to whom, the members are distinct from Liliaceae in habit, dioecious flo wers and con-fluent anther loculi. He considered Smilacaceae to be considerably advanced from the general stock of the Liliaceae. Dahlgren et al., (1985) consid ered the family to be related to Dioscoreaceae and included it under Dioscoreale s. The morphological studies (Conran, 1989) and rbcL sequences (Chase et al., 19 93), however, support place* * * * * * * * * * * Liliaceae M. Adanson Lily family 15 genera, 640 species Widely distributed in the Northern Hemisphere, mainly in the temperate regions. Placement:

B & H Cronquist

Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Coronariae+ Dioscoreanae Lilianae Lilianae Monocots* Order Liliales Liliales Liliales Liliales Cronquist also included Amaryllidaceae, Asparagaceae, Alliaceae and several othe rs under Liliaceae Salient features: Herbs with alternate or whorled leaves, base sheathing, flower s not subtended by spathaceous bracts, flowers bisexual, trimerous, perianth wit h 6 petaloid tepals, stamens 6, filaments free, carpels 3, united, ovary superio r, placentation ax-ile, fruit a capsule. Major genera: Mritillaria (90 species), Gagea (80), Tulipa (80) and Lilium (75). Description: Perennial herbs with underground bulb, generally with contractile r oots. Leaves mostly basal, alternate or whorled, usually linear or strap shaped, simple, 474 Plant Systematics L ~1 Figure 13.41 Liliaceae. A: Flowering portion of Lilium polyphyllum. B: Longitudi nal section of flower of L. canadense. C: Gynoecium of L. lancifolium. D: Transverse section of ovary of L. lancifolium. Gagea pseudoreticulata. E: Flower; F: Gynoecium; G: Tr ansverse section of ovary. (A, after Polunin and Stainton, Fl. Himal., 1984) entire venation parallel, stipules absent. In-florescence usually racemose (Lili um), sometimes solitary (Tulipa) or subumbellate (Gagea). Flowers showy, bisexua l, actinomorphic, rarely zygomorphic, trimerous, hypogynous. Perianth with 6 tep als, in two whorls (outer representing se-pals, inner petals), both whorls petal oid, of-ten spotted or with lines, often united into tube, nectary at the base o f tepal. Androecium with 6 stamens, in 2 whorls, epiphyllous, filaments free. Gy noecium with 3 carpels, united, ovary superior, trilocular with many ovules, pla centation axile, styles simple with 3-lobed stigma. Fruit a loculicidal capsule, rarely a berry; seeds usu-ally flat, with well-developed epidermis, seed coat not black, small embryo, endosperm copious. Pollination by insects, especia Liliales

lly bees, wasps, butterflies. Seeds are dispersed by water or wind. Economic importance: The family is important for its valuable ornamentals such a s lily (Lilium), tulip (Tulipa) and Fritillaria (Fritillary). Phylogeny: The circumscription of the family has undergone considerable reductio n over the recent years. The genera formerly included under the family have now been removed to diverse families: Colchicum (Colchicaceae with corm), Trillium (Tr illiaceae rhizome, leaves whorled, perianth Major Families of Angiosperms 475 with sepals and petals), Allium (Alliaceae bulb, inflorescence umbellate, with sp athaceous bracts, smell of onion, seeds black), Asphodelus and Aloe (Asphodela-c eae inflorescence racemose, seeds black, leaves succulent, often with coloured sap ), Asparagus (Asparagaceae fruit a berry, leaves rudimentary, seeds black), Ruscus (Ruscaceae leaves scarious, filaments connate), the last four were taken under a separate order Asparagales along with several other families, in the recent APG II and APweb classifications. Thorne had earlier (1999, 2000) included these fou r families (along with other splitters from broadly circumscribed Liliaceae) und er order Orchidales, but has shifted (2003) these together with others to order Iridales, also bringing about certain changes in circumscription (Ruscaceae and Convallariaceae included under Asparagaceae). In 2006, 2007 revisions he has slightly enlarged the circumscription of family Liliaceae by merging Tricyrtidaceae and Calochorta ceae, but has divided the family into four sub-families Medeoloideae (1 genus), Lilioideae (9 genera), Tricyrtidoideae (4 genera) and Calochortoideae (1 genus). The narrow circumscription of the family was first suggested by Dahlgren (1985) , and forms a monophyletic group as confirmed by the cladistic studies of Chase et al., (1995a, 1995b). Cronquist, on the other hand, broad-ened the circumscrip tion of the family, also including, in addition to the above families, large fam ily Amaryllidaceae within Liliaceae. * * * * * * * * * * * Orchidaceae A. L. de Jussieu Orchid family 788 genera, 18,500 species (Second largest family after Asteraceae) Widely distr ibuted, most common in moist tropical forests (where frequently epiphytic), also well distributed in subtropics and temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Microspermae+ Lilianae Lilianae Lilianae Monocots*

Order Orchidales Orchidales Orchidales Orchidales Asparagales Salient features: Herbaceous perennials, roots with velamen, leaves distichous, flow-ers trimerous, zygomorphic, corolla with 2 lateral petals and labellum, pol len in pollinia, ovary inferior, seeds minute. Epidendrum (800), Habeneria (580), Liparis (320), Malaxis (280), Oberonia (280), Calanthe (1000), Vanilla (100) and Vanda (60). Major genera: Pleorothallus (1100 species), Bulbophyllum (970), Dendrobium (900) , Description: Perennial herbs, terrestrial (Malaxis, Orchis), epiphytic (Oberonia , Dendrobium) or saprophytic (Gastrodia, Epigonium), rarely climbers (Vanilla), with 476 Plant Systematics 1 1 ,

rIA1 r / W0 C :

77", li 1. E

rpm.

B ~ F

Figure 13.42 Orchidaceae. A: Cymbidium hookeranum with leaves and flowers. B: Er ia muscicola, plant with inflorescences. Oberonia recurva. C: Epiphytic plant with ensiform le aves and pendulous inflorescence; D: Flower ; E: Floral parts separated showing from above downwards bract, three sepals, two lateral petals and anterior labell um. Vanda tessellata. F: Epiphytic plant with inflorescence; G: Floral parts sep arated; H: Pollinia from behind with gland and strap; I: Pollinia from front; J: Operculum from inside.(After Dassanayake and Fosberg, Fl. Ceylone, vol 2, 1981) rhizomes, tuberous roots, corms or root-stock, roots mycorrhizal, with multiseri ate epidermis of dead cells known as velamen. Stems foliate or scapose, base oft en thick-ened to form pseudobulb, aerial roots present. Leaves usually alternate , distichous, rarely opposite, sometimes reduced to scales, often fleshy, simple, entire, sheathing at base, sheath closed and en-circling stem, venation parallel, stipules absent, stomata tetracytic. In florescence racemose, spicate or paniculate, some-times with solitary flowers, r arely cleistogamous. Flowers usually bisexual, Major Families of Angiosperms 477 very rarely unisexual, zygomorphic, usually showy, often twisted 180o during dev elop-ment (resupinate), Perianth differentiated into sepals and petals. Sepals 3 , free or connate, usually petaloid, imbricate, similar or dorsal smaller, later al more or less adnate to the ovary. Petals 3, free; middle petal forming labell um or lip, often spotted and variously coloured, sometimes saccate or even spurr ed at base; lateral petals simi-lar to sepals. Androecium with usually 1 sta-men , sometimes 2 (Apostasia) or 3 (Neuwiedia), adnate to style and stigma forming a

column (gynostemium) opposite the lip, anther sessile on column, bithecous, deh iscence by longitudinal slit, introrse; pollen grains powdery or waxy, agglutina ted into pollinia, each pollinium with a sterile portion called caudicle, 2 to 8 pollinia formed in a flower. Gynoecium with 3 united carpels, ovary inferior, u nilocular with parietal placentation, rarely 3-locular with axile placentation ( Apostasia), stigmas 3, one often transformed into a sterile ros-tellum, latter o ften having a sticky pad called viscidium attached to the pollinia; ovules numer ous, minute, anatropous, tenuinucellate. Fruit a loculicidal capsule or a sausag e-shaped berry; seeds numer-ous, minute, embryo very minute, en-dosperm absent. Pollination mostly by in-sects such as bees, wasps, moths and but-terflies. Flow ers of Ophrys resemble the fe-male wasp and the pollination results from pseudoc opulation, male wasp attracted by the shape and smell of the flowers, mistak-ing it for a female wasp. Tiny dust-like seeds are dispersed by wind. Dendrobium, Cymbidium, Epidendrum, Vanda, Coelogyne and Brassia. The only food p roduct from the family is vanilla flavouring obtained from the fruits Vanilla pl anifolia. Phylogeny: The family is generally considered as a natural group, the monophyly of the family supported by morphology and rbcL sequences (Dressler, 1993; Dahlgr en et al., 1985). The family is commonly divided into three subfamilies: Apostas ioideae, Cypripedioideae, and Orchidoideae. The former two include one tribe eac h, but the last one, which includes nearly 99 per cent of the orchid species is divided into 4 tribes. Apostasioideae are considered to be sister to the remaini ng orchids (Dressler, 1993; Cameron et al., 1999), and monophyletic as supported by vessel-elements with simple perforations and distinctive seed type. Cypriped ioideae are usually considered clearly monophyletic (Judd et al., 1999) as suppo rted by their saccate labellum, two functional stamens and absence of pollinia. Members of Orchidoideae share acute anther apex, soft stems and lack silica bodi es. More recently however 5 subfamilies are recognized the other two being Vanil loideae and Epidendroideae (APweb, Thorne, 2003, 2006). The recent studies have, however, put some uncertainty over the position of Cypripedioideae. For instanc e, they may group (albeit weakly) with Vanilloideae (Freudenstein & Chase 2001) or be sister to Orchidaceae minus Apostasioideae (Cameron 2002; Stevens, 2003). Relation-ships within Orchidoideae are becoming fairly well resolved (Cameron 20 04). Thorne (2006, 2007) recognizes 6 subfamilies: Apostasioideae, Neuwiedioidea e, Cypripedioideae, Orchidoideae and Epidendroideae. Economic importance: The family is known for large number of ornamentals reputed for their showy flowers mainly Cattleya, * * * * * * * * * * * 478 Plant Systematics Iridaceae A. L. de Jussieu Iris family 60 genera, 1,845 species Widely distributed, in tropical and temperate climates, mainly distributed in So uth Africa, the Mediterranean region, Central and South America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida

Liliopsida

Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Epigynae+ Lilianae Lilianae Lilianae Monocots* Order Liliales Orchidales Liliales Iridales Asparagales Salient features: Herbaceous perennials, leaves equitant, leaf base sheathing, f low-ers bisexual, sepals petaloid, petals often spotted, ovary inferior, style p etaloid, fruit a capsule. rarely 2, filaments free or connate, some-times adnate to perianth, anthers bith ecous, dehiscence by longitudinal slits, extrorse, sometimes sticking to style b ranches. Gynoecium with 3 carpels, united, ovary inferior, rarely superior (Isop hysis), 3-locular with axile placentation, rarely unilocular with parietal place ntation (Hermodactylus), ovules few to numerous, anatropous or campylotropous, s tyle three-lobed in upper part, sometimes petaloid. Fruit a loculicidal capsule, dehiscing by valves, usually with marked circular scar at tip; seeds often with aril, copious endosperm and small embryo, seed coat usually fleshy. Flowers are prima-rily insect pollinated, especially by bees and flies, some species by bir ds (Rigidella), a few being wind pollinated (Dierama). Seeds are dispersed by wi nd or water Major genera: Iris (240 species), Gladiolus (230), Moraea (125), Sisyrinchium (1 00), Cro-cus (75), Ixia (45), Freesia (20) and Tigridia (12). Description: Perennial herbs with rhizomes (Iris), corms (Gladiolus) or bulbs, b undle sheaths with styloid crystals of calcium oxalate, tannins and terpenoids p resent. Leaves alternate, distichous, usually sessile, equitant (oriented edgewi se to the stem), simple, entire, sheathing at base, venation parallel, stipules absent. Inflorescence a cyme, raceme, spike or panicle, sometimes solitary, comm only subtended and enclosed by one or more spathaceous bracts. Flowers bisexual, showy, actinomorphic Economic importance: The family includes some of the most popular garden ornamen tals such as Gladiolus, Iris, Freesia, Sparaxis, Tigridia and Sisyrinchium. The stigmas of Crocus sativus yield saffron, widely used as a colouring agent and fo r flavouring food stuffs. Orris root, from Iris florentina is used in making per fumes and cosmetics. (Sisyrinchium) or zygomorphic (Gladiolus), trimerous, epigynous. Perianth 6 in t wo whorls often differentiated into sepals and petals. Sepals 3, free or united, imbricate, sometimes deflexed and with a patch of hairs (bearded Irises). Petal s 3, free (Moraea) or united (Crocus), adnate to the sepals form-ing a perianth tube, petals sometimes spot-ted, erect in bearded Irises (forming stand-ard). An droecium with usually 3 stamens, Phylogeny: The family is related to Liliaceae. The genus Isophysis with superior ovary is sometimes removed to a distinct family, but according to Hutchinson (1

973), it is Major Families of Angiosperms 479 1

E B

G A Figure 13.43 Iridaceae. Iris germanica. A: Rhizome and basal leaves; B: Flower; C: Vertical section of flower; D: Capsule dehiscing through valves; E: Longitudinal section of seed. Crocus vernus. F: Flower and leaves; G: Longitudinal section through entir e plant. Iridaceous in all the characters except based on rbcL sequence (Chase et al., 1995a). superior ovary, and hence included here only. The morphological studies, however (Chase The family is often divided into a number of et. Al., 1995b; Stevenson a nd Loconte) place tribes of which Sisyrincheae with free the family within Lilia les. The combined perianth, rhizome and undivided style studies of the two, plac es the family under branches is considered to be the most Asparagales. Rudall (2 001) included an primitive (Hutchinson). Gladioleae and inferior ovary as a syna pomorphy of the order, Antholyzeae are more advanced with noting that in higher As paragales (now zygomorphic perianth with curved tube and reduced in APG II to on ly two families Alliaceae hood-like dorsal lobe. The position of family is and A sparagaceae), there may well be a major uncertain. Whereas Hutchinson removed it reversal to superior ovaries. Thorne earlier to a distinct order Iridales, Takh tajan placed (1999, 2000) included Iridaceae under order it under Orchidales, an d Dahlgren and Orchidales but subsequently (2003, 2006, Cronquist under Liliales . In the recent 2007) shifted it along with several other classifications of APG , it is placed under a families to Iridales suborder Iridineae, broadly circumsc ribed Asparagales primarily restricting Orchidales to just 6 families. * * * * * * * * * * * 480 Plant Systematics Asphodelaceae A. L. de Jussieu Aloe family 13 genera, 1070 species Temperate and tropical regions of the Old World, especially South Africa, usuall y in arid habitats. Placement:

B & H Cronquist Takhtajan

Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Coronarieae+ Lilianae Lilianae Lilianae Monocots* Order Liliales Asparagales Asparagales Iridales Asparagales B & H, Cronquist under family Liliaceae Salient features: Rhizomatous herbs or shrubs, leaves in rosettes, often succule nt, vascular bundles in a ring surrounding mu-cilaginous central zone, infloresc ence race-mose, flowers bisexual, perianth not spotted, stamens 6, free, not adn ate to tepals, ovary 3 carpels, united, ovary superior, placentation axile, ovules many, orthotropous or anatropous, nectaries in septa of ovary, stigma discoid or 3-lobed. Fruit a loculicidal capsule, rarely a berry (Kniphofia); seeds flattened, black, usually with dry aril Pollination by insects and birds. Seeds mainly dispersed by wind. superior, nectaries in septa of the ovary, fruit a capsule, seeds black. Major genera: Aloe (340 species), Haworthia (55), Kniphofia (50), Bulbine (50) a nd Asphodelus (12). Economic importance: Several genera in-cluding Aloe, Haworthia, Kniphofia and Ga steria are grown as ornamentals. Several species of Aloe are used in cosmetics a nd as sources of medicine. Description: Rhizomatous herbs (rarely bulbous), shrubs or trees often with anomalous secondary growth, anthraquinone present. Phylogeny: The members of Asphodelaceae Leaves in rosettes at base or tips of were earlier included under Liliaceae, bu t branches, simple, usually succulent, not fi- now placed separately. Dahlgren e t al., (1985) brous, vascular bundles in a ring around mu- being the authors of first major classificacilaginous parenchyma, phloem with a cap tion to recognize this and several other of aloine cells containing coloured secre- smaller famil ies as indicated above. They tions, leaves sheathing at base, venation recognize d two subfamilies Asphodeloideae parallel, stipules absent. Inflorescence ra- an d Alooideae. The latter is clearly monoceme, spike or panicle. Flowers usually p hyletic as evidenced by apomorphies of bracteate, bisexual, hypogynous, often le aves with central gelatinous zone with showy, trimerous. Perianth with 6 tepals, aloine layer and dimorphic karyotype. free or slightly connate, petaloid, not s pot- Asphodeloideae includes Kniphofia, which ted. Androecium with 6 stamens, fr ee, not lacks aloine layer and has berry fruit, adnate to tepals, bithecous, bas ifixed or Bulbine which is closer to Alooideae, and dorsifixed, dehiscence longi

tudinal, introrse, Asphodelus. The recognition of Alooideae pollen grains monosu lcate. Gynoecium with renders Asphodeloideae as such Major Families of Angiosperms 481 B I A elk fEE E

K ~r

J I G F Figure 13.44 Asphodelaceae. Asphodelus fistulosus. A: Plant with scapigerous inf lorescence; B: Flower; C: Vertical section of flower; D: Flower with perianth removed showin g stamens and pistil; E: Outer tepal; F: Inner tepal; G: Pistil showing ovary, s imple style and 3-lobed stigma; H: Capsule; I: Seed. J: Eremurus olgae with basa l leaves and scapigerous inflorescence. K: Gasteria verrucosa with basal leaves. (A-G, After Sharma and Kachroo, Fl. Jammu, 1983). paraphyletic (Stevens, 2003). The family is included commonly under Asparagales (Dahlgren, Takhtajan, APG II, Judd et al., and APweb), but Thorne (2003, 2006, 2 007) places it under order Iridales under subor-der Asphodelineae. He like Dahlg ren, recognizes the two subfamilies under Asphodelaceae. Treutlein et al., (2003) on the basis of chloroplast DNA sequences (rbcL, matK) and from genomic finger-printin g (ISSR) concluded that generic limits around Aloe are decidedly unsatisfactory. * * * * * * * * * * * 482 Plant Systematics Alliaceae Batsch ex Borkh. Onion family 80 genera, 1,586 species Widely distributed in tropical and temperate climates, frequently in semiarid ha bitats. Includes five subfamilies: Allioideae (2 genera Allium and Milula; 750 species), Tulbaghioideae (1 genus, 22 species), Gilliesioideae (10 genera, 75 species; in cl. Nothoscordum), Agapanthoideae (1 genus, 9 species) and Narcissoideae (Syn: A maryllidoideae- 66 genera, 730 species). The discussion below includes first fou r, Narcissoideae, the largest of all is described separately. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne

APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Coronarieae+ Lilianae Lilianae Lilianae Monocots* Order Liliales Amaryllidales Asparagales B & H, Cronquist under family Liliaceae; B & H placed Amaryllidaceae under Epigy nae Salient features: Herbaceous perennials with bulbs, latex present, smell of onio n or garlic, leaf base sheathing, inflorescence scapigerous, umbellate cyme, sub tended by spathaceous bracts, flowers bisexual, peri-anth not spotted, stamens 6 , often adnate to perianth, ovary superior, fruit a capsule. ebracteate, bisexual, usually actinomorphic, rarely zygomorphic (Gilliesia) trim erous, hypogynous, pedicels often unequal in size. Perianth with 6 tepals, free or connate at base, in two whorls, petaloid, outer often with green midvein, som etimes with scale-like appendages forming corona (Tulbaghia). Androecium with 6 stamens, filaments free or connate, sometimes epitepalous, sometimes with append ages, bithecous, dehiscence longitudinal, rarely 3 or 4 stamens without anthers, pollen grains monosulcate. Gynoecium with 3 carpels, united, ovary superior, pl acentation axile, ovules 2 or more, anatropous or campylotropous, nectaries in s epta of ovary, style simple, stigma capitate to 3-lobed. Fruit a loculicidal cap sule; seeds globular or angular, seed coat black, embryo curved. Pollination by insects. Seeds wind or water dispersed. Major genera: Allium (499 species), Nothoscrodum (35), Tulbaghia (22), Agapanthu s (9), Miersia (5) and Gilliesia (3). Description: Perennial herbs with bulb and contractile roots, stem reduced, rare ly with corm (Milula) or rhizome (Agapanthus, Tulbaghia), vessel-elements with s imple per-forations, laticifers present, with onion or garlic scented sulphur co mpounds such as allyl sulphides, vinyl disulphide, etc. Leaves mostly basal, alt ernate, simple, cylindrical or flat, often fistular, entire with parallel venation, base sheathing and forming tunic of Economic importance: The family is u se-bulb, stipules absent. Inflorescence ful contributing garlic (Allium sativum) , onscapigerous, umbellate cyme, subtended by ion (A. cepa), leek (A. porrum), a nd chives spathaceous bracts, which enclose the flower (A. schoenoprasum), used as important vegbud, some species producing bulbils instead etables or flavourin gs. Onion seeds are of-of flowers, rarely spicate (Milula). Flowers ten used as substitute for nigella ( kalonji ). A Major Families of Angiosperms 483 Asparagales Iridales

h D G A F Figure 13.45 Alliaceae, Subfamily Allioideae. Allium victorialis. A: Plant with bulb covered with reticulate fibres, broadly laceolate leaves and scape; B: Uppe r part of scape with inflorescence; C: Tepals with stamens; D; Capsule with long style. A. humile. E: Plant with scape and inflorescence; F: Tepals with stamens; G: Capsule. H: Tepa l and stamens of A. roylei showing 2-toothed inner filaments. few species of Allium, Tulbaghia, and Gilliesia umbellate inflorescence (Agapant haceae, are cultivated as ornamentals. Amaryllidaceae and Alliaceae). APG II and APweb have optionally included the oth er two families under Alliaceae. The clade is characterized by bulbs, flavonols, saponins , laticifers, inflorescence scapigerous, umbellate, with scarious spat he, inflorescence bracts 2 (or more - external), pedicels not articulated, free or basally connate perianth, style long, and endosperm nuclear or helobial (Fay et al. 2000). Thorne had earlier (1999) considered Amaryllidaceae and Agapanthac eae as distinct families (like Apweb, 2005) but has finally (2003) merged them i n Alliaceae, recognizing 5 subfamilies Allioideae, Tulbaghioideae, Gilliesioidea e, Agapanthoideae and Amaryllidoideae (in 2006, 2007 revisions he changed it to Narcissoideae) under Alliaceae. The monophyly of Alliaceae is supported by morph ology, chemistry and rbcL sequences. Phplogenp: Originally included under Liliaceae, the first major shift was made b y Hutchinson (1934) who abandoned the traditional separation of Liliaceae and Am aryllidaceae on the basis of superior ovary in former and inferior in latter, an d included Allium and its relatives with superior ovary under Amaryllidaceae, la rgely on the basis of spathaceous bracts. Cronquist had subsequently (1981, 1988 ) merged Amaryllidaceae with Liliaceae. In the recent years a number of distinct families have been segregated, as indicated above. The family is closely relate d to Amaryllidaceae in umbellate inflorescence subtended by spathaceous bracts, bulbs, and the presence of scape. The recent cladistic analysis have resulted in merging all families with * * * * * * * * * * * 484 Plant Systematics ~ N

U D G F E F3 Figure 13.46 Alliaceae, subfamily Narcissoideae. Crinum asiaticum. A: Plant with

inflorescence; B: Flower with elongated perianth tube without corona. Hymenocallis narcissifoli a. C: Inflorescence with a part of scape; D: Longitudinal section of flower showing staminal corona. bxiolirion tataricum (now Ixiliriaceae). E: Inflorescence; F: Dehiscing capsule; G: Seed. H: Vertical section of flower of Narcissus poeticus. Subfamily Narcissoideae Haw. (Syn = Amaryllidoideae Burn.) (Syn = Amaryllidaceae J. St.-Hilaire Amaryllis or Daffodil family) 66 genera, 730 species Widely distributed, in tropical and temperate climates, especially in South Afri ca, South America and the Mediterranean region. Salient features: Herbaceous perennials having bulbs with contractile roots, lea f base sheathing, inflorescence scapigerous, umbellate cyme, subtended by spatha ceous bracts, flowers bisexual, perianth, not spotted, some-times with staminal corona, stamens 6, often adnate to perianth, ovary inferior, 3-chambered, fruit a capsule or berry. Major genera: Crinum (130 species), Hippeastrum (65), Zephyranthes (55), Hymenoc allis (48), and Narcissus (30). Major Families of Angiosperms 485 Description: Perennial herbs having bulb with contractile roots, stem reduced, v essel-elements with scalariform perforations, Leaves mostly basal, alternate, mo stly lin-ear or strap shaped, sometimes petiolate, base sheathing, venation para llel, stipules absent. Inflorescence usually scapigerous, cymose, often umbellat e clusters or solitary, flowers often subtended by spathaceous bracts. Flowers b racteate, showy, bisexual, actinomorphic or zygomorphic, epigynous. Perianth 6, in two whorls (outer representspider lily (Hymenocallis) and amaryllis (Hippeastrum). ing sepals, inner petals), both whorls petaloid, often united into tube, sometim es with a co-rona on throat of perianth tube. Androecium with 6 stamens, in 2 wh orls, epiphyllous, fila-ments free, sometimes expanded and connate forming stami nal corona (Hymenocallis, Pancratium). Gynoecium with 3 united carpels, ovary in ferior, trilocular with many ovules, placentation axile, styles simple with 3-lo bed stigma, nectaries present in septa of ovary. Fruit a loculicidal capsule, ra rely a berry; seeds usually black, with small curved embryo, endosperm fleshy. P ollination by insects and birds. Seeds are dispersed by wind or water. Phylogeny: The group had been tradition-ally circumscribed to include scapigerou s plants with spathaceous bracts in inflorescence and inferior ovary and regarde d as independent family Amaryllidaceae. Hutch-inson had also included genera wit h superior ovary (present Allioideae) under Amaryllidaceae. Cronquist had subseq uently (1981, 1988) merged Amaryllidaceae with Liliaceae. In the recent years a number of distinct families have been segregated, as indicated above. The clade is closely related to Alliaceae in umbellate inflorescence subtended by spathace ous bracts, bulbs, and the presence of scape. Monophyly of the clade is supporte d by inferior ovary, amaryllid alkaloids and rbcL se-quences (Chase et al., 1995 a). The recent cladistic analysis has resulted in optionally merging all familie s with umbellate inflorescence (Agapanthaceae, Amaryllidaceae and Alliaceae) and choosing Alliaceae as the priority name in APG II and APweb, as indicated above . Judd et al. (2002) recognize group as independent family. Thorne (2003) includ es it under Alliaceae in subfamily Amaryllidoideae, changed to Narcissoideae in 2006, 2007 revisions. Economic importance: The subfamily is important for its valuable ornamentals suc h as daffodils (Narcissus), swamp lily (Crinum), * * * * * * * * * * * Agavaceae S. L. Endlicher Agave family 16 genera, 355 species Throughout tropics and subtropics, mainly in arid climate. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Liliidae Liliidae Liliidae Liliidae Series+/Superorder Coronarieae/Epigynae+ Lilianae Lilianae Lilianae Monocots* Order Liliales Amaryllidales Asparagales Iridales Asparagales B & H Yucca (and others with superior ovary) under Liliaceae (series Coronarieae ), Agave (and others with inferior ovary) under Amaryllidaceae (series Epigynae) 486 Plant Systematics

hF

, F E Figure 13.47 Agavaceae. A: Yucca aloifolia with a number of developing infloresc ences. B: Part of inflorescence of Y. filamentosa. Agave americana. C: Plant with inflorescence; D : Stamen with versatile anther; E: Vertical section of flower; F: Transverse sec tion of ovary. Salient features: Large herbs, shrubs or trees with usually rosettes of leaves, leaves succulent with sharp spine at tip, fibrous, inflorescence paniculate, flo

wers bisexual, perianth 6, free or connate, stamens 6, ofnate, simple in rosettes, succulent, margin entire or spinose-serrate, tip with sharp spine, venation parallel, with thick tough fibres, base sheathing. Inflore scence usually terminal raceme or panicle. Flowers ten adnate to perianth, ovary inferior or su- usually bisexual, actinomorphic, t rimerous. perior, 3-chambered, nectaries in septa of Perianth with 6 tepals, fre e (Yucca) or ovary, fruit a capsule or berry, seed coat connate into tube (Agave ), petaloid, not spotblack, bimodal karyotype 5 large and 25 ted, usually white or yellow. Androecium small chromosomes. with 6 stamens, longer than perianth (Agave) or shorter (Furcraea), free, anthers Major genera: Agave (240 species), Yucca bithecous, basifixed (Doryanthes) or (40), Furcraea (20) and Polianthes (13). dorsifixed (Beschorneria), dehiscence by longitudinal slits, introrse. Gynoecium with 3 united carpels, ovary superior (Yucca) or in-ferior (Agave), 3-locular w ith axile placentation; ovules many, anatropous, nectaries in septa of ovary, st yle short or long, stigma minute. Fruit a loculicidal capsule; Description: Large rhizomatous herbs, shrubs or trees with basal or terminal rosettes of leaves, stems with anomalous sec-ondary growth, calcium oxalate crysta ls and steroidal saponins present. Leaves alterMajor Families of Angiosperms 487 seeds flat, black. Bimodal karyotype with 5 large and 25 small chromosomes. Poll ina-tion by moths (Yucca by moth Tegeticula), oth-ers by bats (Agave, some speci es) or birds (Beschornea). Seed dispersal by wind or ani-mals. 1988) circumscribed the family Agavaceae more broadly also to include genera whi ch have now been removed to Dracaenaceae, Nolinaceae, and Laxmanniaceae (incl. L omandraceae). Such a broadly defined family is heterogenous, united by woody hab it and clearly polyphyletic (Dahlgren et al., 1985; Rudall et al., 1997). The fa mily is also characterized by bimodal karyotype, also shared by genera Hosta (pl aced in Hostaceae; Hesperocallidaceae by Thorne, 1999), Camassia and Chlorogalum (both placed under Hyacinthaceae by Thorne, 1999). Rudall et al., (1997) advoca ted their transfer to family Agavaceae, a suggestion incorporated by Judd et al. (2002) and Thorne (2003, 2006, 2007). Judd et al. have recognized only 2 subfam ilies Yuccoideae and Agavoideae. Thorne recognized 4 subfamilies, adding Chlorogal oideae (Camassia, Chlorogalum, Hastingia and Schoenolirion) and Hesperocallidoid eae (Hesperocallis and Hosta). Monophyly of the family is supported by phenotypi c and DNA characters (Bogler and Simpson, 1996). The family is closely related t o Hyacinthaceae. APweb (2008) recognizes 5 groups (generic groups with no formal names). This grouping received 100 per cent support in three- and four-gene tre es (Chase et al., 2000a; Fay et al., 2000). Economic importance: Several species such as Agave sisalana (sisal hemp), A. het eracantha (Istle fibre or Mexican fibre), A. morrisii (Keratto fibre), are impor tant sources of fibre. A few species of Agave are fermented to produce tequila a nd mescal. The species of both Agave and Yucca are used in the manufacture of or al contraceptives. Several species of Agave, Yucca and Polianthes (P. tuberosatube rose) are also used as ornamentals. Phylogeny: The members of the family were earlier placed in Liliaceae and Amaryl lidaceae and were later removed to a separate family to include members with superior ovary (removed from Liliaceae) and inferior ovary (removed from Amaryllid aceae) representing advanced tribes in the respec-tive families (Hutchinson, 197 3), and lack-ing bulb, having arborescent habit, and in-florescence racemose (no t an umbel). Hutchinson (1973) and Cronquist (1981, * * * * * * * * * * * Subclass 5. Commelinidae Suborder 1. Musineae Superorder 1. Arecanae 1. Musaceae Order 1. Arecales

2. Lowiineae 1. Lowiaceae Family 1. Arecaceae 3. Strelitziineae Superorder 2. Commelinanae 1. Strelitziaceae Order 1. Commelinales (B) 4. Heliconiineae 1. Commelinaceae (B) 2. Hanguanaceae 3. Haemodoraceae 4. Pontederiaceae 1. Heliconiaceae 5. Cannineae 1. Cannaceae 2. Marantaceae 5. Philydraceae 6. Zingiberineae 1. Zingiberaceae 2. Cannales 488 Plant Systematics 2. Costaceae 7. Juncales 3. Dasypogonales (B) 1. Thurniaceae 1. Dasypogonaceae 2. Juncaceae 4. Bromeliales (B) 3. Cyperaceae 1. Bromeliaceae 8. Restionales 2. Rapateaceae 2. Anarthriaceae 3. Restionaceae 4. Hopkinsiaceae (B) 5. Lyginiaceae (B) 5. Typhales 1. Typhaceae 6. Xyridales Suborder 1. Xyridineae 6. Centrolepidaceae 1. Xyridaceae 9.. Poales 2. Hydatellaceae (B) 1. Flagellariaceae 2. Eriocaulineae 2. Joinvilleaceae 1. Eriocaulaceae 3. Ecdeiocoleaceae 2. Mayacaceae 4. Poaceae Arecaceae C. H. Schultz (=Palmae A. L. de Jussieu) Palm family 189 genera, 2,350 species Widespread in tropics of both hemispheres, a few in warm temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Commelinidae Liliidae Commelinanae Arecanae Commelinidae Series+/Superorder Calycinae+ Arecanae Commelinids* Order Arecales Commelinales Arecales Arecales Arecales B & H as family Palmae Salient features: Woody shrubs or trees, trunk with scars of fallen leaves, leav es large, fan-shaped or pinnately compound, with sheathing bases, inflorescence paniculate, spathes often present, flowers small. (Hyphaene, Nypa), with prominent scars of fallen leaves, sometimes spiny due to modified leaves roots or exposed fibres, some-times rhizomatous, tannins and pol yphenols often present, vascular bundles with hard fibrous sheath, apical bud we ll protected by leaf sheaths. Leaves alternate, usually form-ing a terminal crow n, petiolate (petiole of-ten with a flap called hastula at base), with pinnate ( feather palms) or palmate (fan palms) segments, sometimes pinnately or twice pin nately compound, plicate (folded like a fan), blades rarely entire (Licuala); le af segments folded V-shaped (induplicate) or Major genera: Calamus (350 species), Bactris (180), Pinanga (120), Licuala (105) , Daemonorops (100), Areca (60) and Phoenix (17). Description: Trees or shrubs with unbranched trunk, rarely branched Major Families of Angiosperms 489 D 3 Figure 13.48 Arecaceae. Cocos nucifera A: Habit; B: Inflorescence; C: Branch of inflorescence with female flowers towards base, male flowers towards the top; D: Male flower; E: Vertical section of male flower; F: Female flower; G: Vertical section of fem ale flower. Calamus pseudotenuis. H: Vegetative branch; I: Portion of stem showi ng thorns; J: Male inflorescence; K: Female inflorescence.

inverted-V-shaped (reduplicate) in cross sec-tion; leaves sometimes very large s ome-times over 20 m (Raphia fainifera with larg-est known leaf). Inflorescence a xillary or ter-minal, often covered with spathes, a repeat-edly branched panicle (Calamus) or almost spicate. Flowers bisexual (Licuala, Livistona) or unisexual with monoecious (Reinhardtia) or dioecious (Borassus, Rhapis) plants, flow-ers small, actinomorphic, usually sessile, trimerous, often with bracteoles connate below flowers. Perianth differentiated into sepals and petals, sometimes vestigi al (Nypa). Sepals 3, free (Arenga) or connate (Didymosperma), usually imbricate. Petals 3, free or connate, usually valvate in male flower and imbricate in fema le flower (valvate in female flowers of Arenga). Androecium usually with six sta mens in two whorls, sometimes numerous (Reinhardtia, Howea), rarely only 3 (Nypia), free, ra rely with connate filaments (Nypa), anthers bithecous, basifixed or dorsifixed, rarely versatile, dehiscence by longitudinal slits; pollen grains usually monosu lcate, smooth or echinulate. Gynoecium with usually 3 car-pels, free or united, only 1 fertile in Phoenix, carpels sometimes many, ovary superior, placentation usually axile, rarely parietal (Gronophyllum), stigma usually terminal, sometime s lateral (Heterospatha) or basal (Phloga), ovules usually 1, rarely upto 3, , o rthotropous or anatropous. Fruit single seeded berry or drupe, exocarp often fib rous or covered with reflexed scales; seeds free or adhering to endocarp, endosp erm present, embryo small. Largest seed in angiosperms formed in double coconut (Lodoicea maldivica). 490 Plant Systematics Economic importance: The family is of great economic importance. Most useful mem ber is Coconut palm (Cocos nucfiera), with almost every part put to use. Mesocar p of the fruit is the source of coir fibre, the seed endosperm (copra) yielding coconut oil, and the leaves used in thatching, basket making and a va-riety of t oys and decoration articles. Palm oil is extracted from Elaeis guineensis). Sago , a major source of carbohydrate food is obtained from Metroxylon sagu (sago pal m) and some species of Arenga and Caryota. Palm wine (toddy) is obtained from sp ecies of Borassus and Caryota. Fibre is also extracted from many species of palm s particularly belonging to Raphia (raffia), Caryota (kitul fibre) and Leopoldin ia (Piassava fibre). Dates are ob-tained from Date palm (Phoenix dactylfiera). V egetable ivory is obtained from the seeds of ivory nut palm (Phytelephas macroca rpa) and was once used for buttons and as a substi-tute for real ivory. Waxes ar e obtained from Copernicia (carnauba wax) and Ceroxylon. Be-tel nut are obtained from Areca catechu of Af-rica and Southeast Asia. The family also con-tributes a large number of ornamentals such as Royal palm (Roystonea regia), fishtail pal m (Caryota), Chinese fan palm (Livistona), and cabbage palm (Sabal). Various species of Ca-lamus are source of commercial cane used in furniture and polo sticks. Phylogeny: In spite of being very large and diverse, and often divided into nume rous subgroups, the family is distinct, easily recognized and monophyletic. APwe b (2005) recognizes 5 subfamilies: Calamoideae, Nypoideae, Coryphoideae, Ceroxyl oideae and Arecoideae. Thorne ( 2007) adds sixth Phytelephoideae segragated from Ceroxyloideae. Uhl et al., (1995) carried out cladistic analysis of the family u sing morphological data as well as cpDNA restriction site analysis and found sup port for placement of Nypa (Nypoideae) a sister of rest of the palms. More recen t studies of Asmussen et al., (2000) indicated that Nypoideae + Calamoideae (str ong support) + the rest of the family (moderate support) form a basal trichotomy ; other characters support these general relationships. However, other work sugg ests that details of the relationships of Nypoideae and Calamoideae to the rest of the family are unclear, and some morphological groupings are not supported by molecular data (Hahn 2002). * * * * * * * * * * * Commelinaceae R. Brown Spiderwort family

40 genera, 650 species Widespread in tropical, subtropical and warm temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Commelinidae Commelinidae Liliidae Commelinanae Commelinanae Commelinanae Commelinids* Commelinidae Series+/Superorder Coronarieae+ Order Commelinales Commelinales Commelinales Commelinales Commelinales Salient features: Herbs with succulent stems, nodes swollen, leaves with closed basal sheath, outer perianth whorl (sepals) green, inner (petals) coloured, flowers bisexual, trimerous, in axils of spathaceous bracts, filaments usually hairy, ovary superior, 3-chambered. Major Families of Angiosperms 491 N

r YI B N

6 [~yA, J

L C D E H t F Fifure 13.49 Commelinaceae. Tradescantia virginiana. A: Plant with flowers; B: V ertical section of flower; C: Flower with sepals and gynoecium; D: Stamen with hairy filament; E : Moniliform staminal hair; F: Transverse section of ovary showing one ovule in each chamber; G: Seed with aril. Commelina paludosa. H: Plant with flowers; I: F lower; J: Transverse section of ovary. C. kurzii. K: Flower; L: Stamen with larg e anther in different views; M: One of the lateral stamens; N: Staminode. (A-E, after Hutchinson, Fam. Fl. Pl, ed. 3, 1973; H-J, after Polunin and Stainton, Fl. Himal, 1984) Major genera: Commelina (170 species), Tradescantia (70), Aneilema (65), Murdann ia (50), Cyanotis (50), Dichorisandra (30) and Zebrina (4). (blue, violet or white) free (Tradescantia,) or connate into a tube (Cyanotis, Z ebrina), per-ishing soon after anthesis, imbricate, crumpled in bud. Androecium with 6 stamens (a few often reduced to staminodes), in 2 whorls, filaments free, often hairy with simple or moniliform hairs, sometimes adnate to petals, connec tive often flattened, anthers bithecous, dehiscence by longitudinal slits, rarel y by apical pores (Dichorisanda), pollen grains monosulcate. Gynoecium with 3 ca rpels, united, ovary superior, trilocular with 1-few orthotropous or anatropous ovules, placentation axile, styles simple with 3-lobed or capitate stigma. Fruit a loculicidal capsule, rarely a berry; seeds with aril, endosperm present, meal y. Description: Annual or perennial herbs, rarely climbers (Streptolirion) with com monly succulent stems and swollen nodes, often with mucilage cells or canals con taining raphides. Leaves alternate, simple, entire, flat or folded V-shaped in c ross section, leaf sheath closed at base, venation parallel, sto-mata tetracytic , stipules absent. Inflores-cence a helicoid cyme at the end of stem in leaf axi l, sometimes solitary, subtended by spathaceous bracts. Flowers bisexual (rarely unisexual) actinomorphic (zygomorphic in Commelina), hypogynous. Perianth 6, in two whorls, outer representing sepals, inner pet-als. Sepals green and free. Pe tals coloured Economic importance: The family is important for its valuable ornamentals such a s 492 Plant Systematics dayflower (Commelina), spiderwort (Tradescantia), Moses-in-the-bulrushes former is characterized by nonspiny pollen grains, medium to large chromosomes, actinomorphic flowers and moniliform hairs. (Rhoeo discolor, now Tradescantia spathacea) and wandering Jew (Zebrina). In Africa, Commelinoideae is characterized by spiny Aneilema beninense is used as a laxative. Leaf pollen, zygomorphic flowers, and filament sap of Floscopa scandens is used in tropical hairs not moniliform. Mon ophyly of the famAsia to treat inflammation of the eyes. The ily is supported by both morphological and young shoots and leaves of Tradescantia molecular data ( Evans et al., 2000). Recent virginiana and Commelina clavata are edible. treatments (APweb, 2008; Thorne, 20 03, Phylogeny: The family is commonly divided into two subfamilies Tradescantoideae and Commelinoideae well formed clades. The 2006, 2007) include the basal Cartonema and widely separated Triceratella in sub

family Cartonemoideae, merging the other two under Commelinoideae. * * * * * * * * * * * Musaceae A. L. de Jussieu Banana family 2 genera, 40 species Mainly wet tropical lowlands from West Africa to Pacific (Southern Japan to Quee nsland). Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Zingiberidae Commelinidae Liliidae Commelinidae Series+/Superorder Epigynae+ Zingiberanae Zingiberanae Commelinanae Commelinids* Order Zingiberales Musales Zingiberales Cannales Zingiberales B & H under family Scitamineae Salient features: Large herbs with pseudostems formed by leaf sheaths, leaves la rge with thick midrib, parallel venation, flowers unisexual, inflorescence subte nded by large spathaceous bracts, corolla 2-lipped, stamens 5 (sixth rudimentary ), carpels 3, ovary inferior, 3-locular, ovules numerous, fruit fleshy berry wit h numerous small black Description: Large usually tree-like perennial herbs with pseudostems formed fro m overlapping leaf sheaths, with laticifers, rhizomatous. Leaves large, spirally arranged, simple, entire, margin often torn and blade appearing pinnate, venati on parallel with stout midrib, sheathing at base. Inflorescence a panicle-like c yme with one or more spathes, axis arising from basal rhizome and growing up thr ough pseudostem. Flow-ers unisexual (plant monoecious), male within upper bracts , female in clusters seeds. Major genera: Musa (34 species) and Ensete (6). Major Families of Angiosperms 493

1 V F i b: ,4 ~Y1 D ~ E I Figure 13.50 Musaceae. Musa Paradisiaca. subsp. sapientum (A-C, F, G) A: Plant w ith inflorescence and split old leaves; B: Young plant; C: Apical portion of inflorescence; D: Male flower of M. rubra; E: Female flower of M. rubra; F: Vertical section th rough bisexual flower; G: Fruit partially opened to show edible berry sliced at top. Ensete edule. H: Bisexual flower; I: Fruit; J: Seed; K: Transverse section of seed showing pit of hilum. within lower bracts. Perianth 6 in two whorls, petaloid. Sepals 3, adnate to 2 p etals, narrowly tubular, soon splitting on one side, variously toothed at apex. Petals 3, somewhat 2-lipped, 2 adnate with sepals, 1 free. Androecium with 5 fer tile stamens and 1 forming staminode, adnate to petals, filaments free, anthers linear, bithecous, dehiscence by longitudinal slits, pollen sticky. Gynoecium wi th 3 united carpels, ovary inferior, 3-locular, ovules many, placentation axile, style filiform, stigma 3-lobed. Fruit elongated berry containing numerous seeds , fruits forming compact bunches; seed with copious and small embryo. in many tropical countries. Manila hemp or Abaca obtained from fibres of M. text ilis is used in making ropes and cordage. Inset or Abyssinian banana (Ensete ven tricosa) is cultivated for its fibre and for food; the stem pulp and young shoot s are eaten cooked. Some dwarf cultivars of Musa (M. acuminata Dwarf Cavendish ) ar e often grown as greenhouse plants in temperate climates. Phylogeny: The family is usually placed in Zingiberales (Cronquist, Dahlgren, AP G II, APweb) along with Cannaceae, Zingiberaceae, Marantaceae and other closely related families. The genus Heliconia, earlier placed in this family has been removed to a sepaEconomic importance: Banana (Musa rate family Heliconiaceae (Thorne, APG II, par adisiaca subsp. sapientum) is a staple food APweb) or placed under Strelitziacea e 494 Plant Systematics (Heywood, 1978). Takhtajan included Musaceae under distinct order Musales. Thorn e (2003, 2006, 2007) prefers name Cannales over Zingiberales for the broadly circumscribed order, divided into 6 suborders, Musaceae included under Musineae. Fossil record has been found in Eocene of W. North America. * * * * * * * * * * * Zingiberaceae Lindley Ginger family 46 genera, 1,275 species

Widespread in tropics mainly under forest shade and wetlands, chiefly distribute d in Indomalaysia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Zingiberidae Commelinidae Liliidae Commelinidae Series+/Superorder Epigynae+ Zingiberanae Zingiberanae Commelinanae Commelinids* Order Zingiberales Zingiberales Zingiberales Cannales Zingiberales B & H under family Scitamineae LJ W MI fy l: A' Figure 13.51 Zingiberaceae. Zingiber officinale. A: Plant with inflorescence; B: Flower; C: Rhizome. D: Flower of Roscoea alpina. Aframomum laurentii. E: Inflorescence; F: Lea f. Alpinia nutans. G: Plant with inflorescence; H: Flower; I: Vertical section o f flower; J: Transverse section of fruit. K: Longitudinal section of seed of Amo mum. Major Families of Angiosperms 495 Salient features: Perennial rhizomatous aromatic herbs, leaves alternate, distic hous, sheathing at base, flowers bisexual, zygomorphic, perianth 6 in two whorls , one petal often larger than others, fertile sta-men 1, staminodes 3 or 4, peta loid, two staminodes forming a lip or labellum, car-pels 3, united, ovary inferi or, placentation axile, fruit a capsule or berry. monosulcate or inaperturate. Gynoecium with 3 united carpels, ovary inferior, 3locular, rarely 2-locular, ovules usually many, placentation axile, or unilocula r with parietal (Globba) or basal (Haplochorema) placentation, style terminal, u ndivided, free or grasped by anther, sometimes 2-lipped or dentate, stigma funne

l-shaped, nectaries 2, on top of ovary. Fruit a fleshy capsule, indehiscent or l oculicidal, rarely berry; seeds globose or angular, with large aril, endosperm c opious, white, hard or mealy, perisperm present. Pollination by insects or birds . Fruits are dispersed by birds. Major genera: Alpinia (165 species), Amomum (130), Zingiber (95), Globba (65), C urcuma (55), Kaempheria (65), Hedychium (66) and Elettaria (7). Description: Perennial rhizomatous herbs, often with tuberous roots, aromatic, c ontain-ing etherial oils, terpenes and phenyl-propanoid compounds, aerial stems short, usually leafless, rarely foliate, vessels present in roots as well as ste m. Leaves al-ternate, distichous, arising from rhizome, with open or closed shea th at base, sessile or petiolate; petiole with air canals separated by diaphragm s with stellate cells; blade large, venation closely parallel, pinnate, divergin g obliquely from midrib, stipules absent, a ligule present at the junction of sh eath and petiole. Inflorescence usually surrounded by involucre (Geanthus) or wi thout involucre (Gastrochilus, Amomum), a dense spicate head (Amomum) or cyme, s ometimes racemose or paniculate (Elettaria), even solitary (Monocostus). Flowers bisexual, often sub-tended by a sheathing bract, usually zygomorphic, early wit hering, often complicated, epigynous, trimerous. Perianth dif-ferentiated into s epals and petals. Sepals 3, green, connate into a tube. Petals 3, showy, somewha t united, posterior petal often en-larged. Androecium with 1 fertile stamen and usually 4 staminodes, in two whorls; outer whorl with 2 staminodes often fused t o form 2-3 lobed lip or labellum (sometimes reduced to a tooth-Rhynchanthus), th ird be-ing absent; inner whorl with one fertile sta-men and two smaller staminod es which are free or connate with staminodes of the lip; fertile stamen with bit hecous anther, grooved and grasping style, pollen grains Economic importance: Many members of the family mainly Hedychium (ginger lily), Kaempheria, Costus, Nicolaia (torch ginger) and Alpinia are widely grown as orna mentals. The family also contributes important spices from the rhizome Zingiber officinale (ginger, adrak ), Curcuma longa (turmeric), or fruits of Amomum subulatu m ( Bengal cardamon, moti elaichi ) and Elettaria cardamomum (Malabar cardamon, chho ti elaichi ). East Indian arrowroot is obtained from tubers of Curcuma angustifoli a. A perfumed powder abir is obtained from the rhizomes of Hedychium spicatum. S pice Melegueta pepper is obtained from Aframomum melegueta. Rhizomes of Alpinia zedoaria are also sources of spice, tonic and perfume (zeodary), whereas those o f A. galanga are used in medicine and flavour-ing (galangal). Phylogeny: The family forms a monophyletic group along with other families in-cl uded under Zingiberales as indicated above. Genus Costus sometimes included unde r a separate subfamily within Zingiberaceae has been removed to a distinct famil y Costaceae. Monophyly of the family is supported by morphology (Kress, 1990) an d DNA information (Kress, 1995). The family was divided by Loesener (1930) into 2 sub-families Zingiberoideae and Costoideae (latter with four tribes Hedychieae , Globbeae and Zingibereae). Hutchinson (1934, 1973) treated the four groups as four tribes (fourth 496 Plant Systematics being Zingibereae). Kress et al., (2001, 2002) has redefined the classification of the fam-ily, recognizing four subfamilies: distribut-ing most genera of the t hree tribes (Costoideae has already been removed as distinct family) among two s ubfamilies Alpinoideae (rhizome fleshy, lateral stami-nodes of outer whorl very small or 0, labellum formed by the two staminodes of the inner whorl alone, frui t usually indehiscent, en-dosperm without starch) and Zingiberoideae (rhizomes f ibrous, lateral staminodes of outer whorl also free from labellum, labellum adnate to filament and forming tube), and recognizing two additional subfamilies Siphonochiloideae (single genus Siphonochilus rhizome fleshy, vertical; inflorescence a raceme, bracteoles 0) and Tamijioideae (single monotypic genus

Tamjiia flagellaris the only species; Rhizome fibrous; placentation parietal). T wo genes analyses by these authors indicated strong support for Siphonochiloidea e being sister to other three, and Tamijioideae to the other two Alpinioideae + Zingiberoideae. Thorne (2006, 2007) and Stevens (2008) follow this * * * * * * * * * * * Cannaceae A. L. de Jussieu Canna family 1 genus, 25 species Mainly tropical and subtropical America, several species having naturalized in A sia and Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Zingiberidae Commelinidae Liliidae Commelinidae Series+/Superorder Epigynae+ Zingiberanae Zingiberanae Commelinanae Commelinids* Order Zingiberales Cannales Zingiberales Cannales Zingiberales B & H under family Scitamineae Salient features: Herbaceous perennials, leaves broad, sheathing at base, flower s showy, bisexual, trimerous, perianth petaloid, stamens petaloid, only one fert ile, ovary inferior, style flat with marginal stigma, fruit covered with warts. distinct midrib containing air canals, alter-nate, spirally arranged, simple, ve nation par-allel, petiole sheathing the stem, stipules and ligule absent. Inflor escence a terminal raceme, panicle, or spike of commonly 2-flowered cincinni; ax is 3-angled in section with 3-ranked bracts, each bract associated with each 2-f lowered (rarely 1-flowered) cincinnus. Flowers showy, bisexual, zygomorphic, epi gynous, with a bract and a bracteole. Perianth 6, in two whorls (outer represent ing sepals, inner petals). Sepals 3 free, green or purple, persistent in fruit. Petals 3, connate Major genera: Single genus Canna (10 species) Description: Perennial herbs with under-ground rhizome. Stem with mucilage canal s. Leaves large, spirally arranged, broad, with Major Families of Angiosperms 497 a

Figure 13.52 Cannaceae. Canna indica. A: Plant with leaves and inflorescence; B: Open flower showing petaloid staminodes, half anther and tip of style; C: Fertile stamen wit h half petaloid staminode and half anther; D: Transverse section of ovary showin g axile placentation and ovary wall covered with warts; E: Dehiscing capsule cov ered with warts; F: Seed. at base and adnate to staminal column. Androecium with 6 stamens, in 2 whorls, starch from rhizome of C. edulis (Queens land connate and adnate to petals, three outer Arrowroot) is used as diet for in fants as it is modified into petaloid imbricate staminodes, easily digestible. of 3 inner 2 modified into petaloid staminodes and the third with one anther lob e fertile and other modified into petaloid staminode, pollen grains inaperturate . Gyn-oecium with 3 united carpels, ovary inferior, trilocular with many ovules, placentation axile, styles simple, petaloid with marginal stigma. Fruit a capsu le covered with warts, dehiscing by collapse of the pericarp; seed spherical, bl ack, with tuft of hairs (modified aril) with straight embryo, endosperm hard, pe risperm present. Most species self polli-nated. Seeds often dispersed by water. brids are grown as garden ornamentals. The Phylogeny: The family is closely related with other families such as Zingiberaceae, Musaceae, Marantaceae and Strelitziaceae, usually placed under order Zingiberales, differing from Zingiber aceae in lacking ligule. Thorne (2003, 2006, 2007) prefers name Cannales for the order, placing the family Cannaceae under suborder Cannineae. Takhtajan (1997) has narrowly circumscribed Cannales, to include Cannaceae and Marantaceae (subor der Cannineae of Thorne), Zingiberales restricted to include only Zingiberaceae and Economic importance: Various species of Costaceae. Lowiaceae is removed to Canna especially C. indica and various hy- Lowiales and Musaceae together 498 Plant Systematics Heliconiaceae and Strelitziaceae to Musales. All 4 orders are placed under super order Zingiberanae. Grootjen and Bouman (1988) described a pachychalaza in Canna ceae, with mitosis occurring during ovule development in the chalaza and basal part of the nucellus. This is unlike the other Zingiberalean families. The family is monophyletic, as supported by DN A and morphology (Kress, 1990, 1995). * * * * * * * * * * * Juncaceae A. L. de Jussieu Rush family 6 genera, 345 species Worldwide, mostly in cold temperate and montane regions, usually in damp habitat s. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division

Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Commelinidae Liliidae Commelinidae Series+/Superorder Calycinae+ Juncanae Commelinanae Commelinanae Commelinids* Order Juncales Juncales Cyperales Poales Salient features: Tufted herbs, leaves grass-like, sometimes reduced to basal sh eath, perianth 6, in two whorls, stamens 6, pollen in tetrads, ovary superior, p lacentation axile, stigmas 3, fruit a capsule. small, trimerous. Perianth with 6 tepals, in two whorls, free, green brown or bl ack, rarely scarious, inner sometimes smaller (Marsippospermum). Androecium with usually 6 stamens, rarely 3 (Voladeria), opposite the tepals, free, anthers bil ocular, basifixed, dehiscence by longitudinal slits, introrse, pollen in tetrads , monoporate. Gynoecium with 3 united carpels, ovary superior, placentation axil e, sometimes parietal (Marsippospermum), ovules many, styles 3 or 1, stigmas 3. Fruit a loculicidal capsule; seeds spherical or flat, sometimes pointed and spin dle-shaped (Marsippospermum), with small straight embryo, endosperm present. Win d pollinated. Major genera: Juncus (260 species), Luzula (65), Oxychloe (7) and Distichia (3). Description: Perennial or annual tufted herbs, often with rhizomes, stems cylind ri-cal and solid, usually foliate only at base. Leaves alternate, mostly basal, 3-ranked, rarely distichous (Distichia), cylindrical or flat, sheathing at base or reduced only to sheath, sheath open or closed, blade grass-like, entire, vena tion parallel, stipules and ligule absent. Inflorescence with cymes clustered in heads or forming panicles, corymbs or even solitary (Andesia). Flowers usually bisexual, sometimes unisexual (Distichia) and plants dioecious, rarely monoeciou s (Rostkovia), actinomorphic, very Economic importance: Family is not of much commercial use. Split rushes used in basket making are taken from stems of Juncus effusus (soft rush) and J. squarosu s (heath rush). Juncio, used in binding, is derived from Juncus maritimus (sea r ush). A few Major Families of Angiosperms 499 ]i ow 4 H Juncales

Figure 13.53 Juncaceae. Juncus articulatus. A: A plant with inflorescences; B: F lower; C: Perianth and stamens; D: Gynoecium; E: Capsule; F: Seeds. Luzula albida. G: Plant wi th inflorescence; H: Flower; I: Vertical section of flower; J: Gynoecium; K: Deh iscing fruit; L: Seed; M: Longitudinal section of seed. N: Flower of Juncus spha celatus. species of Juncus and Luzula are grown as ornamentals. tives of Juncaceae. The family is connected to Cyperaceae through genus Oreobolu s, the most primitive genus of that family. Muasya et al., (1998) suggest that O xychloe (Juncaceae) is sister to Cyperaceae, with moderate support, other Juncac eae are also basal and paraphyletic, but with poor sup-port, while Prionium is s ister to the whole clade, with good support. A study by Plunkett et al., (1995) placed Oxychloe within Cyperaceae. The relationships of the latter genus in part icular are still unclear. Accord-ing to the studies of Bremer (2002) Thurniaceae (including Prionium) are sister to Juncaceae plus Cyperaceae, (Oxychloe not inc luded) with strong support. Phplogenp: Juncaceae are closely related to Liliaceae (Hutchinson, 1973; Heywood , 1978) representing reduced forms derived from that stock. Prionium which was e arlier considered to be the most primitive genus of the family Juncaceae, linkin g it to Liliaceae, has now been removed to a dis-tinct family Prioniaceae (Thorn e, 1999, 2003), or under Thurniaceae (APG II; Apweb, 2008; Judd et al., 2008; Th orne, 2006, 2007). Takhtajan (1997) places Prionium under Juncaceae. Restionacea e, which have be-come totally dioecious, form the closest rela* * * * * * * * * * * 500 Plant Systematics Cyperaceae A. L. de Jussieu Sedge family 104 genera, 5,010 species Worldwide, mostly in cold temperate regions, usually in damp habitats. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Commelinidae Liliidae Commelinidae Series+/Superorder Glumaceae+ Juncanae Commelinanae Commelinanae Commelinids* Order

Cyperales Cyperales Cyperales Poales Juncales

Y 'A F 1 Figure 13.54 Cyperaceae. Eleocharis lankana. A: Plant in flower; B: Inflorescenc e; C: Fruiting gynoecium with hypogynous bristles. Carex ligulata. D: Lower part of plant; E: P art of inflorescence; F: Male spike; G: Female spike; H: Utricle; I: Nut. Major Families of Angiosperms 501 Salient features: Herbs, stems often 3-an-gled, solid, leaves 3-ranked, containi ng silica bodies, sheaths closed, ligule absent, glumes present, flower subtende d by a sin-gle bract, lodicules absent, perianth repre-sented by bristles, scale or absent, ovary su-perior with single ovule, fruit a nut. pous, placentation basal, style with 2-3 branches, stigmas 2 or 3. Fruit a nut ( often called achene but the latter is strictly derived from a single carpel), so metimes enclosed in a utricle (Carex), with often persistent style and associate d with persistent perianth bristles, bifacial or trigonous; seed erect, embryo s mall, endosperm conspicuous, mealy or fleshy. Major genera: Carex (1800 species), Cyperus (580), Fimbristylis (290), Scirpus ( 280), Rhynchospora (240), Scleria (200) and Eleocharis (190). Economic importance: Various species of Cyperaceae are useful in different ways. Stems of Cyperus papyrus (papyrus or paper weed) were much used in ancient time s for making paper, and is now commonly grown as an ornamental. The stems of Cla dium effusum (saw grass) are also source of cheap paper. Stems and leaves of Car ex brizoides and Lepironia mucronata are used for pack-ing and basket work. Unde rground organs of Cyperus esculentus (tigernut, Zulu nut or rush nut), Scirpus t uberosus and Eleocharis tuberosa (matai, Chinese water chest nut) are used as fo od. The stems of Scirpus totara are used for making canoes and rafts and those o f S. lacustris for basketwork, mats and chair seats. Description: Annual or perennial herbs, usually rhizomatous, stems mostly 3-angl ed, solid. Leaves alternate, 3-ranked, often crowded at the base of stem, simple , grass-like, with silica bodies, entire or serrulate, venation parallel, stipul es and ligule absent, sheath closed, stomata with dumbbell-shaped guard cells. I nflorescence consist-ing of small spikes (sometimes called spikelet but differen t from spikelet of grasses which has two basal glumes, and each floret enclosed in a lemma and a palea) each often subtended by a bract (prophyll) and bearing ( on the axis called rachilla) spirally arranged (Cladium) or distichous (Cyperus) bracts (glumes) , each subtending one flower; small spikes (spikelets) aggregat ed in spikes, panicles or even umbels, the whole inflorescence subtended by one or more usu-ally leaf-like involucral bracts. Flowers very small, bisexual (Cype rus, Scirpus) or uni-sexual (Scleria), subtended by bract (glume), female flower often with second bract sur-rounding the pistil and forming sac-like perigonium . Perianth represented by bris-tles, sometimes scales (Oreobolus, Lipocarpha), o r even absent (Bulbostylis, Scirpus). Androecium with 3 stamens, some-times more (6 in Arthrostylis; 12-22 in Evandra), free, anthers bithecous, basifixed, oblo ng or linear, dehiscence by longitudinal slits, pollen grains uniporate, in pseu -domonad (out of the four microspores, three degenerate and form the part of fou rth fer-tile forming pollen grain). Gynoecium with 2 (Kyllinga) or 3 (Cyperus) u nited carpels, ovary superior, unilocular, ovule 1, anatroPhylogeny: The family is considered to be monophyletic, connected to Juncaceae t hrough genus Oreobolus, the most primitive genus (Hutchinson, 1973). Muasya et a l., (1998) suggest that Oxychloe (Juncaceae) is sister to Cyperaceae, with moder ate sup-port. Plunkett et al. (1995) placed Oxychloe within Cyperaceae. The rela tionships of the latter genus in particular are still unclear. According to the

studies of Bremer (2002) Thurniaceae (including Prionium) are sister to Juncacea e plus Cyperaceae, with strong support. He did not include Oxychloe under Cypera ceae. According to him Cyperaceae, Juncaceae and Thurniaceae form a well de-fine d cyperid clade. The family is tradition-ally divided into 3 subfamilies (Engler ): Scirpoideae, Rhynchosporoideae and Caricoideae. Thorne (2006, 2007), however recognizes only two: Mapanioideae and Caricoideae. Hutchinson divided the family into 8 tribes. Simpson et al., (2003) based on 502 Plant Systematics the pollen and plastid DNA sequence data concluded that Mapanioideae (tribe Hypo lytreae of Hutchinson) are sister to the rest of the family, while Carex, sister to Eriophorum is embedded in the other clade. Within Caricoideae, especially tr ibe Cariceae, phylogenetic studies are begin-ning to resolve relationships (Star r et al. 2004, 2006). According to Ford et al. (2006), Carex itself includes a few other genera, and conventional wisdom in which a highly compound inflorescence is the plesiomorphic condition for the genus, taxa with simple branches being derived, perhaps several times, seems the exact opposite of what actually happened. * * * * * * * * * * * Poaceae Barnhart Grass family (= Gramineae A. L. de Jussieu) 678 genera, 10,230 species (Fourth largest family after Asteraceae, Orchidaceae and Fabaceae) Worldwide, distributed from poles to equator and from mountain peaks to sea leve l, in all types of climates and habitats. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Monocotyledons Liliopsida Liliopsida Liliopsida Magnoliopsida Subclass Arecidae Commelinidae Liliidae Commelinidae Series+/Superorder Glumaceae+ Poanae Commelinanae Commelinanae Commelinids* Order Cyperales Poales Poales

Poales Poales B & H as Gramineae Salient features: Herbs or shrubs with hollow internodes and jointed stems, leav es distichous with distinct sheath enclosing the stem and linear blade with ofte n a ligule at their junction, spikelet with two glumes, flowers reduced, enclose d in lemma and palea, perianth represented by lodicules, ovary superior, stigma feathery, fruit caryopsis. Description: Herbs or rarely woody shrubs or trees (bamboos), often with rhizome s, stolons or runners, frequently tillering (branching from ground level) to for m tufts of stems, stem (culm) with hollow internodes and jointed swollen nodes, with silica bodies. Leaves distichous, alternate, simple, with basal sheath surr ounding internode and free linear blade, a ligule often present at the junction of blade and sheath, margins of sheath overlapping but not fused, sometimes unit ed into a tube, venation parallel, leaf margins often rolled especially on dryin g, stipules absent. Inflorescence of spikelets arranged in racemes, panicles (Po a, Avena) or spikes Major genera: Poa (500 species), Panicum (450), Festuca (430), Paspalum (350), S tipa (300), Bromus (160), Elymus (150), Sporobolus (140), Bambusa (125), Setaria (100), Arundinaria (50) and Chloris (50). Major Families of Angiosperms 503

1 ~ Y

1 Figure 13.55 Poaceae. Zea mays. A: Plant with terminal male inflorescence and ax illary female inflorescence (Cob); B: Vertical section of female spikelet; C: Paired male spik elets; D: Male spikelet opened to show two fertile florets. Poa annua. E: Plant in flower; F: Spikelet. Avena sativa. G: Inflorescence; H: Spikelet opened; I: F ertile floret with awned lemma. (Triticum, Hordeum). Each spikelet with 2 (rarely 1 as in Monera) glumes enclosi ng 1 (Hordeum, Nardus) or more (Poa, Triticum) florets borne on an axis called r achilla, usually in 2 rows. Flowers small, reduced (floret), zygomorphic (due to only 2 lodicules displaced on one side), rarely actinomorphic, usually bisexual rarely unisexual (Zea), hypogynous, enclosed in lemma and palea (prophyll), lem ma often bearing dorsal (Avena), subterminal (Triticum) or terminal (Hordeum) aw n, or awn absent (Poa). Perianth absent or represented by 2 (rarely 3, as in Bam busa and Streptochaeta) lodicules. Androecium with usually 3, sometimes 6 (Oryza ) or more (Arundinaria), rarely 1-2 (Leptureae) stamens, filaments free, anthers bithecous, basifixed, usually sagittate, dehiscence longitudinal, pollen grains monoporate. Gynoecium variously interpreted as bicarpellary, tricarpellary (wit h one reduced style), syncarpous or monocarpellary, unilocular with 1 ovule, pla

centation basal, styles 2, sometimes 3 (Bamboos and Streptochaeta), very rarely 1 (Anomochloa), stigmas often feathery. Fruit a caryopsis, rarely nu t berry or utricle; seed fused with pericarp, embryo straight, endosperm starchy . Economic importance: The family is of great economic importance, being a source of important cereals such as rice (Oryza sativa), wheat (Triticum aestivum) and corn or maize (Zea mays). The family also includes other food crops such as barl ey (Hordeum vulgare), pearl millet (Pennisetum glaucum), oats (Avena sativa), ry e (Secale cereale) and sorghum (Sorghum vulgare). Grasses such as Cynodon, Axono pus and Agrostis are extensively used in lawns and turfs. Andropogon, Agropyron, and Phleum are major forage grasses. Sugarcane (Saccharum officinarum) is the m ajor source of commercial sugar. Bamboos are employed in big way in construction work, wickerwork and thatching in different parts of the world. Young bamboo sh oots are used as food and often pickled. Lemon grass (Cymbopogon) 504 Plant Systematics leaves are distilled to yield essential oil for imparting citronella scent. Grai ns of Coix lacryma-jobi (Job s tears) are use as necklace beads. Roots of Uetiveri a zizanioides (vetivar grass) are used for making fragrant cooling pads and extr action of vetiver oil. Centostecoideae occupies an isolated position although related to both Bambusoid eae and Panicoideae, and includes broad-leaved herbs with single- to several-flo wered spikelets. Studies of Clark et al., (1995) and Soreng and Davis (1998) sug gest that Arundinoideae, Chloridoideae and Panicoideae form a well supported cla de (often called PACC clade) based on embryological and DNA data. Arundinoideae as generally defined are not monophyletic, and many of their members such as Ari stida, Phragmites, etc. are spread over in other two subfamilies. Chloridoideae and Panicoideae are generally found to be monophyletic. Stevens (APWeb, Phplogenp: Although a very large assemblage Poaceae are easily recognized and fo rm a monophyletic group, as supported by morphology (lodicules, spikelets with g lumes, lemma and palea, fruit caryopsis) and DNA characters( rbcL and ndhF seque nces). Cronquist (1988) places Poaceae and Cyperaceae under the same order Cyper ales, but similar morphology of two is believed to be due to convergent evolutio n, Cyperaceae being more closely related to Juncaceae (Judd et al., 1999). The s tudies of Bremer (2002), using rbcL and taq analyses found strong support for Cy peraceae, Juncaceae, and Thurniaceae forming cyperid clade and Poaceae along wit h other families forming a 2003) and Thorne (2003) listed 12 subfamilies under Poaceae: Pharoideae, Puelioideae, Panicoideae, Arundinoideae, Centothecoideae, Chloridoid eae, Aristidoideae, Danthonioideae (six forming PACCAD clade), Bambusoideae, Ehr hartoideae, Pooideae (BEP clade). Subsequently (APWeb 2008, Thorne 2006, 2007), however, they have added 13th Micrairoideae, probably sister to the whole clade (Thorne prefers Chondrosoideae to Chloridoideae). There is great diversity in th e morphology and biochemistry of C4 photosynthesis in the family (Kellogg, 2000) . Studies based on gene expression (Ambrose et al., 2000) indicate that the pale a and perhaps even lemma are calycine in nature and the lodicules are corolline. Clark and Triplett (2006) discuss relationships within Bambusoideae, previously divided into the woody Bambuseae and the herbaceous Olyreae. However, the woody temperate bamboo group may be sister to the rest of the family. The duplication of AP1/FUL gene, apparently in stem-group Poaceae, may be involved in the evolu tion of the spikelet (Preston & Kellogg 2006). Malcomber and Kellogg (2005) sugg est that there has been duplication of LOFSEP genes within Poaceae, while there has been a duplication of the whole genome in a clade that includes at least Zea , Oryza, Hordeum and Sorghum (Schlueter et al. 2004). Developmental gene duplica tion and subsequent functional divergence seem to have played a very Anomochlooideae, graminoid clade. The nature of gynoecium in this family has been a matter of controversy. Most ea rly authors including Haeckel (1883), Rendle (1930) and Diels (1936) considered

it to consist of a single carpel terminated by 2-3 branched stigma. Lotsy (1911) , Weatherwax (1929) and Arber (1934) considered that it represents tricarpellary ovary having evolved from an ovary with parietal placentation, a view supported by studies on floral anatomy (Belk, 1939). Others believe that gynoecium consis ts of 2-3 carpels (depending on the number of stigmas visible; Cronquist 1988, Woodland, 1991). The family is variously classified by different authors. Hutchinson (1973) recog nized two subfamilies Pooideae (with 24 tribes) and Panicoideae (with 3 tribes). Heywood (1978) recognized 6 subfamilies (Bambusoideae, Centostecoideae (should be Centothecoideae as the genus Centosteca on which the name is based is listed neither in Willis, 1973 nor Hutchinson, 1973), Arundinoideae, Chloridoideae, Pan icoideae and Pooideae), further subdivided to include 50 tribes. Of these subfam ilies, Major Families of Angiosperms 505 important role in allowing the development of the baroque diversity of infloresc ences in the family (Malcomber et al. 2006). Indeed, there has been very extensive duplication of genes - API, AG and SEP families but not in the AP3 lineage (Zahn et al. 2005a). * * * * * * * * * * * Subclass 6. Ranunculidae (A) 2. Eupteleales (B) Superorder 1. Proteanae (A) l. Eupteleaceae Order 1. Proteales 3. Paeoniales (B) Family 1. Proteaceae 1. Paeoniaceae 2. Platanales (B) 2. Glaucidiaceae 1. Platanaceae 4. Ranunculales 3. Buxales (B) Suborder 1. Ranunculineae 1. Buxaceae 1. Lardizabalaceae 2. Circaeasteraceae 3. Menispermaceae 4. Berberidaceae 5. Hydrastidaceae 2. Didymelaceae 4. Sabiales (B) 1. Sabiaceae Superorder 2. Ranunculanae 6. Ranunculaceae Order 1. Nelumbonales 2. Papaverineae 1. Nelumbonaceae 1. Pteridophyllaceae (B) 2. Papaveraceae Paeoniaceae Rafinesque Peony family 1 genus, 33 species Mainly temperate regions of Asia and Europe, also in Northwest America Placement:

B & H

Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Liliopsida Polypetalae Dilleniidae Magnoliopsida Magnoliopsida Subclass Ranunculidae Magnoliidae Ranunculidae Ranunculanae Core Eudicots* Series+/Superorder Thalamiflorae+ Ranunculanae Theanae Order Ranales Dilleniales Paeoniales Paeoniales Paeoniales Saxifragales B & H under family Ranunculaceae Salient features: Perennial rhizomatous herbs or shrubs, leaves alternate, compo und or lobed, without stipules, flowers large, bisexual, sepals 5, green and lea thery, petals 5-10, coloured, stamens many, centrifugally arranged, carpels 5, free, ovules many, fruit etaerio of follicles. Major genera: Single genus Paeonia (33 species). 506 Plant Systematics

F H

D I A 1 Figure 13.56 Paeoniaceae. Paeonia emodi. A: Branch with flower; B: Flower bud wi th leafy bracts; C: Flower magnified to show numerous stamens; D: Petal; E: Stamen; F: Ovary cove red with hairs; G: Follicle splitting; H: Follicle dehisced to expose seeds; I: Seed with aril Description: Perennial herbs or soft shrubs, with tubers or rhizomes, stem base covered with scale-like sheaths. Leaves alternate, petiolate, pinnately to terna tely compound to highly dissected or lobed, stipules absent. Inflorescence usual ly with solitary flowers with leafy bracts at base. Flowers Large, showy, bisexu al, hypogynous, almost globu-lar in appearance. Calyx with 5 sepals, free, green , unequal, imbricate, subfoliaceous, persistent. Corolla with 5 petals, sometime s 6-10, large, free, orbicular, subequal, imbri-cate. Androecium with numerous s tamens, centrifugal, attached to fleshy disc present around the carpels, free, s pirally arranged, bithecous, basifixed, dehiscence by longitu-dinal slits, extro rse. Gynoecium with 5 carpels, sometimes upto 2, borne on fleshy disc, free, fle shy, ovary superior, unilocular, ovule 2-many, placentation marginal, stigma ses sile, thick, falcate, 2-lipped. Fruit an etaerio of leathery follicles, dehiscing by adaxial suture, seeds globose, with aril, red turn-ing black at maturity, with prominent umbilicus, embryo small, en dosperm copious. Economic importance: The family contributes many ornamentals cultivated for attr active flowers. The flowers of Paeonia officinalis may reach 15 cm in diameter. Phylogeny: The genus Paeonia was once in-cluded under family Ranunculaceae, from which, however, it is distinct in having 5 large chromosomes, centrifugal (and not centripetal) stamens, persistent sepals, disc and seeds with aril. The separ ation of the genus into a distinct family was first advocated by Worsdell (1908) on the basis of anatomical evidence. Corner (1946) considered the centrifugal d evelopment of stamens of considerable importance in phylogeny and advocated Major Families of Angiosperms 507 placing Paeoniaceae near Dilleniaceae, a placement followed by Cronquist (1981, 1988), order Saxifragales (Core Eudicots) by APG II and APweb. Paeoniaceae and another monogeneric family Glaucidiaceae are often but not supported by Hutchinson, who in 1969 placed Paeoniaceae before Helleboraceae considered related. They are together pl aced under order Ranales. The placement of in the same order Paeoniales by Dahlg ren Paeoniaceae near Dilleniaceae is also con- (1989) and Thorne (2003, 2006, 20 07), tradicted by difference in gynoecial develop- whereas as Takhtajan (1997, p laces them ment, nectary morphology (Stevens in Apweb, under two adjacent orders Paeoniales and 2008). Hutchinson considered Paeonia to be Glaucidiales under su perorder Ranuna link between the Magnoliaceae and culanae of subclass Ranunculidae. Mabberley Helleboraceae, but much more closely related to latter. Paeonia was linked with moderate support to the Crassulaceae clade, or, more weakly, with the Crassulace ae + Saxifraga-ceae clades in some analyses in Fishbein et al., (2001) and accor dingly placed under (1997) includes Glaucidium in Paeoniaceae. Hoot et al., (1998) included Glaucidi um and Hydrastis under Ranunculaceae, being sister to rest of the family, a trea tment followed in APG II and APweb, the latter under two sub-families Glaucidioi deae and Hydrastidoideae. * * * * * * * * * * * Berberidaceae Durande Barberry family 13 genera, 660 species

Widespread chiefly in North Temperate regions and the Andes of South America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Liliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Magnoliidae Ranunculidae Magnoliidae Ranunculidae Ranunculanae Ranunculanae Ranunculanae Eudicots* Series+/Superorder Thalamiflorae+ Order Ranales Ranunculales Berberidales Ranunculales Ranunculales Ranunculales Salient features: Herbs or shrubs, stipules absent, flowers bisexual, sepals and petals similar, stamens 6, outer whorl opposite the petals, anthers dehiscing b y valves, carpel 1, ovary superior, fruit a berry. Description: Perennial herbs or shrubs, rarely small trees, stem sometimes with scattered vascular bundles, wood usually col-oured yellow by berberine (an isoqu inolene alkaloid). Leaves usually alternate, rarely opposite (Podophyllum), simp le (Berberis), or palmately lobed (Podophyllum), or pinnate compound (Mahonia), rarely 2-3 times pinnately compound (Nandina), leaves of Major genera: Berberis (540 species), Mahonia (60), Podophyllum (12), Jeffersoni a (2) and Nandina (1). 508 Plant Systematics

d B III E

D w, E F G

H Figure 13.57 Berberidaceae. Berberis vulgaris. A: Twig with leaves, flowers and spines; B: Flower; C: Longitudinal section of flower; D: Fruit; E: Seed. B. stenophylla. F: Stamen with anther dehiscing by two valves; G: Ovary; H: Longitudinal section of ovary (F-H after Hutchinson, 1973). longer shoots sometimes modified into spines (Berberis), leaves entire or spinos e-serrate, venation pinnate or palmate, reticu-late, stipules absent. Infloresce nce a rastyle very short, stigma almost sessile, sometimes 3-lobed. Fruit usually a berr y, rarely a dehiscent capsule (Jeffersonia), or an achene (Achlys); seeds with s mall embryo, endosperm copious, sometimes with aril Pollination by insects. Disp ersal by birds or animals. Bladder-like capsule of Leontice dispersed by wind. I n Caulophyllum the fleshy blue seeds burst through the ovary wall and develop in completely exposed state. ceme, panicle (Nandina) or even solitary (Jeffersonia). actinomorphic, hypogynous. Calyx with 3 to 6 sepals, free, imbricate, green (Pod ophyl-lum) or petaloid (Berberis), rarely absent (Achlys). Corolla with 3-6 peta ls, sometimes more, free, inner whorl often in the form of petaliferous nectarie s, rarely absent (Achlys). Androecium with usually 6 sta-mens, opposite the peta ls, sometimes upto 18 (Podophyllum) or reduced to 4 (Epimedium), anthers bitheco us, dehiscence by longitu-dinal valves opening from base upwards, sometimes by l ongitudinal slits (Nandina, Podophyllum), pollen grains usually tricolpate. Gyno ecium with 1 carpel, ovary superior, unilocular with many ovules, sometimes with 1 ovule (Nandina), anatropous, placentation parietal or basal, Flowers bisexual, Economic importance: Many species of Berberis (B. buxfiolia, B. darwinii), Mahon ia (M. aquifolium) and Nandina (N. domestica) are commonly grown as ornamentals. The rhizomes of Podophyllum hexandrum (May apple) yield a resin which is used a s a purgative, and incorporated in many laxative pills. Phylogeny: The family includes genera which are quite distinct from one another. Chapman (1936) on the basis of carpellary anatomy proposed that Berberidaceae a nd Major Families of Angiosperms 509 Ranunculaceae arose by parallel evolution from a proranalian complex, and also d oubted whether any existing families may be related as the immediate predecessor s of Berberidaceae. She also demonstrated that single carpel of this family aros e from ancestors having 3 carpels with axile placentation, and that two carpels were suppressed and their placentae moved towards one side of the ovary, and the locules lost by compression, resulting in a unilocular condition. According to Kim & Jansen (1998) the gynoecia of the n = 6 clade alone (Epimedium, Podophyllu m, Jeffersonia) being derived from two carpels. Hutchinson (1973) separated the genera included here under three families: Berberidaceae (including woody genera Berberis and Mahonia in which anthers open by flaps), Nandinaceae (single woody genus with 2-3 times pinnate compound leaves and anthers opening by slits) and Podophyllaceae (including herbaceous genera). Interestingly, whereas former two were included under order Berberidales, the last family was included under Ranal es along with Ranunculaceae, Nymphaeaceae, Ceratophyllaceae, etc. Takhtajan also segregated Ranzaniaceae and included all the 4 families under order Berberidales. The recent classificat ions treat them under the same family Berberidaceae which is considered to be mo nophyletic as supported by morphology and DNA data. The family is placed under o rder Ranunculales along with Ranunculaceae and other related families in most of the recent systems. Nandina is considered to be sister to rest of the family an d often included under separate subfamily Nandinoideae, and rest of the genera u nder Berberidoideae. A number of distinct clades are recognized within Berberido ideae (Loconte, 1993). Leontice, Gymnospermum and Caulophyllum are characterized by petaliferous nectaries (staminodes), pollen with reticulate sculpturing and

basal placentation. Similarly Epimedium, Vancouveria and Jeffersonia are distinc t in the sense that large fleshy blue seed develops in an exposed condition. Tho rne (2003, 2006, 2007) recogniszes 4 subfamilies under Berberidaceae: Nandinoide ae, Berberidoideae, Leonticoideae and Podophylloideae. APweb (2008) recognizes o nly two, monogeneric Nandinoideae and Berberidoideae including rest of the gener a. * * * * * * * * * * * Ranunculaceae M. Adanson Buttercup or Crowfoot family 58 genera, 2,505 species Primarily in temperate and boreal regions of the Northern Hemisphere. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Ranunculidae Magnoliidae Ranunculidae Series+/Superorder Thalamiflorae+ Ranunculanae Ranunculanae Ranunculanae Eudicots* Order Ranales Ranunculales Ranunculales Ranunculales Ranunculales Ranunculales 510 Plant Systematics H Figure 13.58 Ranunculaceae. Ranunculus muricatus. A: A portion of plant with flo wers and fruits; B: Vertical section of flower; C: Petal with nectary; D: Stamen; E: Achene. Cons olida ajacis. F: A branch with young inflorescence and an expanded inflorescence ; G: Vertical section of flower; H: Stamen; I: Dehiscing follicle.(A-E, after Sh arma and Kachroo, Fl. Jammu, 1983). Salient features: Herbs, leaves with sheath-ing base, blade often divided, flowe rs bisexual, petals with nectary, stamens and carpels nu-merous, free and spiral ly arranged, ovary su-perior, fruit a follicle or achene. (Naravelia). Inflorescence of solitary flowers (Anemone) or cymose, sometimes ra cemes (Delphinium) or panicles (Clematis natans). Flowers bracteate (Clematis) o r ebracteate (Anemone) bisexual (unisexual in 7halictrum), actinomorphic (zygomo rphic in Delphinium) with spirally arranged stamens and carpels, hypogynous. Cal yx with 5 (4 in Clematis) or many sepals, free, one (Delphinium) or all five (Aq uilegia) sepals often produced into spur at base. Corolla with 5 or many (Helleb orus) petals, free, often with nectaries or represented only by nectaries (Delph inium), sometimes produced into spur which enters the spur formed by sepal, some times perianth is not differentiated (Anemone, Helleborus) into sepals and petal

s. Androecium with many stamens, free, spirally arranged, anthers often extrorse , dehiscence longitudinal. Gynoecium with Major genera: Ranunculus (400 species), Clematis (200), Delphinium (250), Aconit um (245), Anemone (150) and 7halictrum (100). Description: Mostly herbs, sometimes woody climbers (Clematis), or shrubs (Xanth orhiza). Stem with scattered or several rings of vascular bundles. Hairs simple. Leaves usually alternate (opposite in Clematis), undivided (Caltha) palmately l obed (Ranunculus) or compound (Clematis), stipules absent (present in 7halictrum ). Tendrils for support may sometimes be formed from petiole (Clematis) or termi nal leaflet Major Families of Angiosperms 511 single (Consolida) or many free (Delphinium) carpels (syncarpous in Nigella), un ilocular (multilocular in Nigella) with single (Ranunculus) or many (Delphinium) ovules, placentation marginal or basal, rarely axile (Nigella), ovary superior, style 1, sometimes feathery (Clematis), stigma 1. Fruit an achene (Ranunculus), follicle (Delphinium), berry (Actaea) or rarely a capsule (Nigella); seed with small embryo, endosperm present. Pollination usually by insects. Clematis and An emone, which lack nectaries are pollinated by pollen-gathering insects. Ranuncul us, Delphinium, etc., with nectaries by usually bees. Some species of Thalictrum are wind pollinated. Achenes may be provided with hairs for wind dispersal (Cle matis), with tubercles or hooked spines for dispersal by animals (Ranunculus). B erries of Actaea are mainly dispersed by birds. tis, with 3-merous perianth, vessels with scalariform perforations, ovule with t wo integuments, and fleshy follicles occupies a unique basal position along with Glaucidium, as evidenced by molecular data. Both these genera were removed by T akhtajan (1997) into distinct families Hydrastidaceae and Glaucidiaceae, under H ydrastidales and Glaucidiales, respectively. Thorne (2003) includes Glaucidiacea e under Paeoniales, but Hydrastidaceae near Ranunculaceae under Ranunculales. St udies based on cpDNA restriction sites and sequence data (Hoot, 1995) suggest th at these two genera along with other genera placed in Thalictroideae form basal paraphyletic group, thus justifying retain-ing all these genera within Ranuncula ceae. These basal genera retain plesiomorphies such as presence of berberine, ye llow creep-ing rhizomes, small hairs and small chromosomes, linking them to Berb eridaceae. Economic importance: Delphinium (LarkThe separation of follicle bearing genera spur), Anemone (windflower), Aquilegia under Helleboraceae by Hutchinson is re-( columbine), Ranunculus (buttercup), and jected by the evidence from floral anato my. Helleborus (hellebore) are grown as The reduction in the number of ovules pe r ornamentals. Aconitum napellus yields aco- carpel and the evolution of achenes has ocnite, whereas A. ferox is source of bikh poi- curred several times within the family. The son. Roots of Hydrastis (removed by separation is also negated by nucleotide seTakhtajan to Hydrastidaceae) are used for quences (Hoot, 1995). The petals with stomach ailments. Seeds of Nigella sativa nectary are often cons idered to represent (Nigella, black seed, Kalonji ) are used as fla- petaliferous n ectaries, the petals being abvouring, medicinally to treat asthma, bron- sent. A ccording to Erbar et al., (1999) they chitis and rheumatism. Thymoquinone ex- ar e interpreted as being derived from statracted from the seeds of this species ha ve mens, and that stamens are secondarily recently been found to be useful in treatment spiral. Thorne (2003, 2006) divide s family of cancer. Ranunculaceae into 3 subfamilies: Coptidoideae, Isopyroideae (Thalictroideae in 2007 revision) and Ranunculoideae. Stevens (APweb, 2006) recognizes 5, adding Hy drastidoideae and Glaucidioideae. Phylogeny: The family is largely considered to be a monophyletic group as suppor ted by morphology and molecular evidence. Hydras* * * * * * * * * * * 512 Plant Systematics

Papaveraceae A. L. de Jussieu Poppy family 50 genera 830 species Widely distributed, primarily in temperate regions of the Northern Hemisphere, a lso in Southern Africa and Eastern Australia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Magnoliidae Ranunculidae Magnoliidae Ranunculidae Series+/Superorder Thalamiflorae+ Ranunculanae Ranunculanae Ranunculanae Eudicots* Order Parietales Papaverales Papaverales Papaverales Ranunculales Ranunculales Salient features: Herb, sap usually milky or coloured, flowers bisexual, sepals caducous, petals crumpled in bud, stamens numerous in several whorls, ovary supe rior, unilocular, fruit a capsule. merous stamens (Papaver), sometimes 4 and opposite the petals (Corydalis) or 6 i n two bundles of 3 each (Dicentra), anthers bithecous, (in Fumaria of the 6 stam ens 2 are with bithecous anthers, 4 with monothecous anthers) dehiscence longitu dinal, pollen grains tricolpate to polyporate. Gynoecium with usually 2 united c arpels, sometimes loosely united and becoming free in fruit (Platystemon), ovary superior, unilocular with parietal placentation, sometimes be-coming multilocul ar due to intrusion of placentae, ovules numerous, sometimes 1 (Baccopia), anatr opous, stigma discoid or lobed, sometimes capitate. Fruit a capsule dehiscing by valves or splitting into 1-seeded segments, sometimes nut (Fumaria); seeds smal l, sometimes with aril, embryo minute, endosperm copious, fleshy or oily. Pollin ation usually by insects, rarely wind (Bocconia). Seeds are dispersed by explosi ve opening of capsules, those with aril often by ants. Major Genera: Corydalis (380 species), Pa-paver (100), Fumaria (50), Argemone (3 0) and Eschscholzia (10). Description: Annual or perennial herbs, rarely soft-wooded shrubs (Dendromecon), or small trees (Bocconia), vascular bundles of-ten in several rings, white or c oloured la-tex. Hairs simple, sometimes barbellate (Cathcartia). Leaves usually alternate, floral leaves sometimes subopposite (Platystemon), simple, often much dissected, sometimes entire (Dendromecon) or spinose (Argemone), venation retic ulate, stipules absent. Inflo-rescence usually with solitary flowers, scapigerou s in Sanguinaria, racemose in Eomecon, paniculate in Bocconia. Flowers bisexual, actinomorphic, sometimes zygomorphic (Fumaria, Corydalis). Calyx with 2 sepals, sometimes 3, caducous or calyptrate, free, usually enclosing bud. Co-rolla with usually 4 petals, sometimes 6 or even 8-12 (Sanguinaria), free, usually in two

whorls, two outer sometimes saccate or spurred containing nectary (Fumaria, Cory -dalis), inner sometimes connivent at tip (Fumaria), imbricate, often crumpled i n bud, absent in Bocconia. Androecium with nuEconomic importance: Many species of Pa-paver (poppy), Eschscholzia (Californian poppy), Argemone (Prickly poppy), Corydalis (harlequin), Sanguinaria (blood-roo t) and Dicentra (Dutchman s breeches, bleeding heart) are grown as ornamentals. Op ium poppy (Papaver somniferum) is the most valuable member yielding opium (obtai ned from the latex of capsules) and its derivatives heroin, morphine and codeine . Seeds of this Major Families of Angiosperms 513 i Figure 13.59 Papaveraceae. Papaver nudicaule. A: Plant with flowers; B: Fruit wi th bristly hairs. P. rhoeas. C: Vertical section of flower; D: Gynoecium with on e stamen still attached, others having shed; E: Transverse section of ovary with intruded placent ae; F: Fruit with glabrous surface and broad stigmatic disc. G: Argemone ochrole uca with flowers and a fruit towards the base with conspicuous style; H: Fruit o f A. mexicana with sessile style; I: Seed of A. mexicana.. (A-B after Polunin an d Stainton, Fl. Himal., 1984). species do not contain opium and as such are used in baking, and also yield a dr ying oil. Seeds of Glaucium flavum and Argemone mexicana also yield oils used in the manu-facture of soaps. sometimes treated under distinct family Fumariaceae (Hutchinson, 1926, 1973; Law rence, 1951; Cronquist, 1988; Dahlgren, 1989 and Takhtajan) but morphological an d nucleotide sequence data supported the monophyly of the family including these genera, which are better placed under subfamily Fumarioideae (Thorne, 2003; Jud d et al., 2002). APG II also optionally include Fumariaceae under Papaveraceae. There is, however, difference of opinion regarding basal genera. Loconte et al., (1995) proposed Platy-stemonoideae (Platystemon and relatives) with numerous sl ightly fused carpels and free Phylogeny: The family is considered to be related backwards to Helleboraceae (Hu tchinson, 1973) but with syncarpous gynoecium and parietal placentation, and ver y clearly forwards to Brassicaceae, which also has parietal placentation but wit h false septum. Genera with zygomorphic flowers, and with saccate or spurred pet als are 514 Plant Systematics stigmas as the basal clade. Hoot et al., (1997) on the other hand, on the basis of morphology and nucleotide sequence, regarded Pteridophyllam as sister to the remaining genera. This monotypic genus has been removed to a distinct family Pte ridophyllaceae by Takhtajan (1997), Thorne ( 2003, 2006, 2007) APG II and APWeb. Thorne (2007) divides Papaveraceae into 5 subfamilies: Papaveroideae, Eschscholzioideae, Chelidonioideae, Hypecoideae and Fumarioideae. The family was earlier placed closer to Brassicaceae and Capparaceae, due to the parietal placentation, but has now been shifted closer to (or under) Ranunculal es, the shift supported by chemical evidence absence of glucosinolates and the pr esence of alkaloid benzylisoquinolene. * * * * * * * * * * * Subclass 7. Hamamelididae (B) 6. Iteaceae Superorder 1. Hamamelidanae 7. Pterostemonaceae 8. Grossulariaceae Order 1. Hamamelidales 9. Haloragaceae (B) Suborder 1. Trochodendrineae 3. Juglandales Family 1. Trochodendraceae 1. Juglandineae

2. Cercidiphyllaceae 1. Rhoipteleaceae 2. Hamamelidineae 2. Juglandaceae 1. Altingiaceae 2. Myricineae 2. Hamamelidaceae 1. Myricaceae 3. Daphniphyllaceae 4. Betulales 2. Saxifragales 1. Nothofagaceae 1. Tetracarpaeaceae 2. Crassulaceae 3. Penthoraceae 4. Saxifragaceae 5. Aphanopetalaceae (B) 2. Fagaceae 3. Ticodendraceae 4. Betulaceae 5. Casuarinaceae Saxifragaceae A. L. de Jussieu Saxifrage family 30 genera 525 species Widespread but best represented in the Northern Hemisphere, mainly in the temper ate and arctic climate. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Hamamelididae Hamamelidanae Core Eudicots* Series+/Superorder Calyciflorae+ Saxifraganae Rosanae Order Rosales Rosales Saxifragales Saxifragales

Saxifragales Saxifragales B & H as Saxifrageae

Major Families of Angiosperms 515 F I

a=a, r5 J H Figure 13.60 Saxifragaceae. Bergenia ciliata. A: Plant with basal leaves and cor ymbose panicle carried on a long scape; B: Flower with petals distinctly larger than nearly cupshaped calyx. Saxifraga flagellaris. C: Plant with thick stolons, small leaves a nd few flowers; D: Flower with almost free sepals, both pedicel and sepals gland ular. E: Flower with calyx and gynoecium, petals and stamens removed; F: Stamen; G: Transverse section of ovary with axile placentation. Astilbe rivularis. H: P ortion of bipinnate leaf and a paniculate inflorescence alongside; I: Flower lac king petals, with 5 stamens and 2 carpels; J: Seed, tailed at both ends. Salient features: Perennial herbs, leaves alternate, gland-toothed, stipules abs ent, flowers actinomorphic, usually perigynous, sepals and petals 5 each, stamen s 5 to 10, carpel 2, united, ovary superior, placentation axile, fruit a capsule . Description: Perennial herbs, vessel elements with simple perforations, often wi th tannins, sometimes cyanogenic. Leaves alternate, usually in basal rosette, si mple or pinnately or palmately compound, venation pinnate or palmate, reticulate , stipules absent or represented by expanded margins of petiole base. Infloresce nce racemose or cymose, rarely with solitary flowers. Flowers bisexual, rarely u nisexual, (plants Major genera: Saxifraga (310 species), Heuchera (50), Chrysoplenium (45), Mitell a (18), Astilbe (18) and Bergenia (6). 516 Plant Systematics monoecious or dioecious), actinomorphic, rarely zygomorphic, usually perigynous with distinct hypanthium, rarely epigynous. Calyx usually with 5 sepals, rarely 4, free or connate, often persistent. Corolla usu-ally with 5 petals, free, ofte n clawed, imbri-cate or convolute, sometimes reduced or absent. Androecium with 5 to 10 stamens, free, anthers bithecous, dehiscence longi-tudinal, pollen grain s tricolporate. Gynoecium with usually 2 carpels, rarely up to 5, united, free o r adnate to hypanthium, ovary superior or inferior, placentation axile or pariet al, ovules nu-merous, styles free, stigmas free, capitate. Fruit a septicidal ca psule or follicle; seed with small straight embryo surrounded by endosperm. Poll ination mainly by insects. Seeds dispersed by wind or passing animals.

Asterids. Parnassiaceae is in agreement with data from floral anatomy (Bensel & Palser 197 5b, c). The family has long been considered as closely related to Rosaceae. Asti lbe of Saxifragaceae can be confused with Aruncus of Rosaceae but the former qui te often have opposite leaves, their carpels are usually two and connate at the base, and their stamens are fewer. These resemblances are mainly superficial. Th ere are two major clades in Saxifragaceae, Saxifraga s. str. and the Heuchera cl ade, members of the latter conSimilarly separation of Economic importance: The family has lit-tle economic importance with a few gener a Saxifraga, and Astilbe grown in rock gardens or perennial borders. taining the bulk of the floral variation in the family (Soltis et al., 2001). Ge neric limits are unclear; hybridization is extensive and there are various combi nations of chloroplast and nuclear genomes. For example, the chloroplast genome of Tellima is also found in Mitella (Soltis et al., 1993). However, the unitegmi c Darmera with scapigerous inflorescence is properly to be retained in Saxifraga ceae (Gornall 1989). Thorne had earlier placed Saxifragales under Rosidae but ha s subsequently (2003, 2006, 2007) shifted it to newly created subclass Hamamelid idae, under superorder Hamamelidanae, order Saxifragales. The family is monophyl etic as evidenced by data from cpDNA restriction sites, rbcL, matK, and 18S sequ ences and morphology. The members, in addition share an rpl2 intron deletion. Re cent studies have shown that genera like Saxfiraga and Mitella are not monophyle tic. In addition hybridization often causes taxonomic problems. Thorne (2006, 20 07) recognizes two subfamilies Astilboideae and Saxifragoideae. Phylogeny: The family was earlier broadly circumscribed to include genera, which have now been separated to different families such as Grossulariaceae (Ribes), Hydrangeaceae (Hydrangea separated by Thorne (2003) to Asteridae >Cornanae >Hydrangeal es (Cornales in 2007); Asterids >Cornales in APG II), Parnassiaceae (Parnassia under Rosidae >Celastranae >Celastrales by Thorne; Eurosids I >Celastrales by APG II and AP web), etc. Hydrangeaceae are woody, tenuinucellate and unitegmic and related to * * * * * * * * * * * Major Families of Angiosperms 517 Fagaceae Dumortier Oak family 9 genera 990 species Widespread in tropical and temperate regions of the Northern Hemisphere. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II /(APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Hamamelidae Hamamelididae Magnoliidae Hamamelididae Series+/Superorder Unisexuales+ Faganae Rosanae Hamamelidanae Eurosids I* Order Fagales

Fagales Fagales Betulales Fagales B & H under Cupuliferae Salient features: Trees or shrubs with tan-nin, leaves alternate, simple, entire or ser-rate, stipules present, inflorescence cymose, female flowers usually in groups of 1-3, as-sociated with scaly cupule, carpels usually 3, ovary inferior, placentation axile, fruit a nut, closely associated with cupule. in female flower. Gynoecium with usually 3 carpels, rarely upto 12, united, ovar y inferior, locules as many as carpels, placentation axile, ovules 2 in each cha mber but only one in whole ovary developing, pendulous, bitegmic, outer integume nt vascularized, styles free, stigmas porose or expanded along upper side of sty le, fertilization porogamous. Fruit a nut (acorn) closely associated with and su rrounded at base (Quercus) or completely (Castanea) with cupule, cupule often ha rdened and woody, sometimes spiny (Castanea), indehiscent (Quercus) or dehiscent by splitting of cupule like pericarp into valves (Castanea); seed single, witho ut endosperm. Flowers of Fagus and Quercus are wind pollinated, those of Castane a and Castanopsis produce strong odour and are pollinated by flies, beetles and bees. Fruits are dispersed by birds and rodents. Major genera: Quercus (430 species), Lithocarpus (280), Castanopsis (100), Casta -nea (12) and Fagus (8). Description: Trees or shrubs, deciduous or evergreen, tannins present, hairs sim ple or stellate, sometimes glandular. Leaves sim-ple, alternate, sometimes lobed , entire or serrate, venation pinnate, reticulate, stip-ules present, early deci duous, often narrowly triangular. Inflorescence cymose, male flow-ers in slender catkins or spikes, females flowers solitary or in groups of upto three, associa ted with a scaly cupule formed of sev-eral imbricate scales, male and female flo w-ers sometimes in the same inflorescence (Castanea, Lithocarpus). Flowers small , uni-sexual (plants monoecious), actinomorphic. Perianth with 4-6 tepals, reduc ed, free or slightly connate, imbricate. Androecium with 4-numerous stamens, fil aments free, filiform, anthers erect, bithecous, loculi of-ten contiguous, dehis cence longitudinal, pollen grains usually tricolporate or tricolpate, staminodes sometimes present Economic importance: Species of Castanea (chestnut) yield nuts which are eaten a fter roasting, but have a very short shelf life, turning rancid within a few day s. Fruits of some species of Quercus (oak) and Fagus (beech) are also occasional ly eaten. Cork is made from the bark of Quercus suber. Wood of several species i s a source of timber used for construction, furniture, barrels and cabinetry. Se veral species of Quercus, Fagus, Castanea and Castanopsis are grown as ornamenta ls. 518 Plant Systematics

F G Figure 13.61 Fagaceae. Quercus robur. A: Branch with lobed leaves and long pedun cled female flower; B: Young shoot with male catkin; C: Male flower; D: Fruit enclosed upto nearly half by cupule. E: Branch of Castanea sativa with long spikes each bearin g single female flower at base and numerous male flowers above, cupule spiny. Ca stanopsis indica. F: Branch with several spikes; G: Fruiting spike. Lithocarpus pachyphylla. H: Portion of branch with leaves; I: Portion of fruiting branch wit h nuts in groups of three. Phplogenp: The family is closely related to Betulaceae and the two are usually p laced under the same order, although Takhtajan places only Fagaceae and Nothofag aceae un-der Fagales and separates Betulaceae and others under Corylales. The fa mily is mono-phyletic as supported by morphology, cpDNA restriction sites (Manos et al., 1993) and matK sequences (Manos and Steele, 1997). Casta-nea, Lithocarp us and Chrysolepis have re-tained numerous plesiomorphic morphologi-cal characte rs such as monoecious inflores-cences, perianth better developed, exserted stame ns and minute stigmas. The nature of cupule has been a subject of considerable discussion. It is generally regarded t o represent a cymose inflorescence in which outer axes of the cyme are modified into cupule valves which bear scales or spines (Manos et al., 2001). The cupules of Nothofagus (which was earlier placed (Hutchinson, 1973, Cronquist, 1981) und er Fagaceae but now separated under Nothofagaceae) are composed of clustered bra cts and stipules and not homologous with the cupule of Fagaceae. Heywood (1977) recognized three subfamilies under Fagaceae: Fagoideae, Quercoideae and Castanoi deae. Fagus is sister to rest of the Major Families of Angiosperms 519 Fagaceae and placed singly under subfamily Fagoideae. Trigonobalanus is consider ed sis-ter to the rest of family (excluding Fagus) and as such removed together with Colombobalanus and Formanodendron under fourth subfamily Trigonobalanoi-dea e by Thorne. Thorne had earlier (1999) placed Betulales under Rosidae >Rosanae but has subsequently (2003, 2006, 2007) shifted it to Hamamelididae >Hamamelidanae. APG II and AP web prefer order name Fagales placed under the clade Eurosids I. * * * * * * * * * * * Betulaceae S. F. Gray Birch family 6 genera 140 species Widespread in temperate and boreal regions, Alnus being distributed in South Ame rica in Andes and Argentina. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II /(APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons

Subclass Monochlamydeae Hamamelidae Hamamelididae Magnoliidae Series+/Superorder Unisexuales+ Faganae Rosanae Hamamelidanae Eurosids I* Order Fagales Corylales Fagales Betulales

Hamamelididae

Fagales B & H under Cupuliferae Salient features: Trees or shrubs with tan-nin, bark sometimes exfoliating in th in lay-ers, leaves alternate, simple, doubly serrate, stipules present, inflores cence a catkin, male and female inflorescences distinct, tepals 2-4, stamens 2-4 , perianth absent in female flower, carpels usually 2, placentation axile, fruit a nut, surrounded by fused bract and bracteoles. secondary veins running into serrations, stipules present. Inflorescence a catki n, with separate male and female inflorescences but plant monoecious, male inflo rescence usually pendulous, female short and erect, flowers borne singly at each node of catkin or in cymose cluster of 2-3, adhering to involucre of bracts and bracteoles. In Alnus, female inflorescence has each cymose cluster with 2 flowe rs associated with 1 bract, two secondary bracteoles and 2 tertiary bracteoles a ll connate into woody persistent involucre. In Betula, there are 3 female flower s in the cluster with 1 bract, 2 bracteoles, all fused into a 3-lobed bract or inv olucre. Flowers small, unisexual (plants monoecious), actinomorphic. Perianth wi th usually 2-4 tepals, rarely 1 or upto 6, reduced, free, imbricate, absent in m ale (Coryloideae) or female flower (Betuloideae). Androecium Major genera: Betula (55 species), Alnus (30), Carpinus (28), Corylus (15), Ostr ya (10) and Ostryopsis (2). Description: Trees or shrubs, deciduous, tannins present, bark smooth or scaly, with prominent horizontal lenticels, sometimes exfoliating in thin layers, hairs simple, glan-dular or peltate. Leaves simple, alternate, doubly serrate, venati on pinnate, reticulate, 520 Plant Systematics Figure 13.62 Betulaceae. Betula utilis. A: Portion of branch with male catkins, appearing before or along with leaves; B: Branch with female spikes; C: Single stamen (second one removed) with bracteole and forked filament separating anthers; D: Bract and la teral bracteoles of male flower; E: Fused bract and bracteoles of female flower; F: Young winged carpel; G: Nut with 2 wings and persistent styles. Alnus nitida . H: Branch with slender male catkins towards top and ovoid female spikes lower down; I: Male flower with 4 tepals and 4 stamens; J: Nut with 2 wings. (A-B, aft er Polunin and Stainton, Fl. Himal, 1984). with 2 (Betula) or 4 (Alnus) stamens, rarely 1 or upto 12 (Coryloideae), sometim es appear-ing many due to close association of three flowers, filaments free or connate at base, anthers bithecous, loculi distinct or contigu-ous, dehiscence l ongitudinal, pollen grains usually 2-multiporate, staminodes absent in female fl ower. Gynoecium with 2 united carpels, ovary inferior, bilocular, placentation a xile, ovules 2 in each chamtimes bladder-like (Ostrya); seed solitary pen-dulous, embryo straight, cotyledo ns large, endosperm absent. Flowers are wind pollinated, and emerge before leave

s. Winged fruits of Betula and Alnus are dispersed by wind. Large nuts of Corylu s are dispersed by rodents. Economic importance: Papery bark of Betula utilis (birch) was used as a writing surface (bhojpatra) in place of paper in ancient Vedic ber but only one in whole ovary developing, pendulous, unitegmic, styles free, cylindri- manuscripts; also used for roofing and um-cal, stigma running along adaxial side of brella covers. B. lutea and B. lenta are imstyle, pistillode absent in male flower. Fruit portant hardwoods in North America provida single-seeded nut or 2-winged samara of- ing wood used for plywood, boxes and turn-ten with persistent styles, involucre of bract and bracteoles deciduous or persistent, scaly and woody or enlarged and foliaceous, someery. Alnus rubra (alder) provides a valuable timber which is a good imitation of ma-hogany. Nuts such as hazelnuts, filberts Major Families of Angiosperms 521 from species of Corylus are edible. Many spe-cies of Betula, Alnus, Corylus, Ost rya are grown as ornamentals. winged samara) and Coryloideae (male flower without perianth, female with perian th, stamens usually more than 3, involucre foliaceous, nut not flattened). Nucle ar ribosomal ITS and rbcL sequences also support these two subfamilies (Chen et al., 1999). Hutchinson treated them as distinct families, also proposing that as the female flow-ers of Betulaceae lack perianth, the ovary is superior, and tha t of Corylaceae with perianth and ovary is inferior, a contention not supported by other authors. Both groups are monophyletic, although monophyly of Ostrya and Carpinus is doubtful (Yoo & Wen, 2002). Phylogeny: The family is closely related to Fagaceae and the two are usually pla ced under the same order, although Takhtajan places only Fagaceae and Nothofagac eae under Fagales and has separated Betulaceae and others under Corylales. The f amily is usually divided into two sub-families: Betuloideae (male flowers with p erianth, female lacking perianth, stamens 2 or 4, involucre scaly or woody, frui t 2* * * * * * * * * * * Casuarinaceae R. Brown She-Oak family 4 genera 96 species Widespread in Southeast Asia and Australia, naturalized in the coastal regions o f tropical and subtropical Africa and America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Hamamelidae Hamamelididae Magnoliidae Hamamelididae Series+/Superorder Unisexuales+ Casuarinanae Rosanae Hamamelidanae Eurosids I*

Order Casuarinales Casuarinales Casuarinales Betulales Fagales Salient features: Usually trees with jointed stems, appearing like conifers, lea ves scale-like, whorled at nodes, inflorescence catkin, flowers unisexual, subte nded by bracts, peri-anth absent, stamen 1, carpels 2, united, ovary superior, f ruit a samara, fruits aggre-gated like cones. jointed with circular sheath at nodes (switch habit), branches grooved, photosyn thetic, sometimes aromatic (Allocasuarina), sieve-tube plastids S-type, nodes un ilacunar, roots with nodules containing nitrogen-fixing bacteria, tannins presen t . Leaves whorled (4 20 in a whorl), scale-like, connate forming a toothed shea th at each node, stipules absent. Inflorescence forming catkins at tips of later al branches. Flowers small, unisexual (plants monoecious or dioecious), actinomo rphic, solitary in axil of each bract of inflorescence, associated with 2 bracte oles. Perianth absent in female flower, sometimes represented by Major genera: Allocasuarina (55 species), Casuarina (25), Gymnostoma (14) and Ce uthostoma (2), often combined into single genus Casuarina. Description: Trees or shrubs with a weep-ing habit due to long slender branches, stems 522 Plant Systematics 1~ F

H Figure 13.63 Casuarinaceae. Casuarina suberba. A: Branch with male inflorescence s; B: Portion of male inflorescence; C: Male flower with single stamen; D: Portion of branch w ith female inflorescences; E: Part of female inflorescence showing 3 flowers; F: Female flower with bract, 2 small bracteoles and gynoecium with 2 long stylar b ranches; G: Fruits; H: Seed with broad wing; I: Longitudinal section of seed. 1-2 vestigial scales in male flower (often in-terpreted as inner bracteoles). An droecium with single stamen, anthers bithecous, incurved in bud, dehiscence long itudinal, pollen grains usually triporate. Gynoecium with 2 united carpels, ovar y superior, bilocular with axile placentation, one often reduced and ovary appea ring unilocular, ovules 2 but only one developing, orthotropous, bitegmic, crass inucellate, style short with 2 linear branches. Fruits crowded into cones with p ersistent bracts, fruit an indehiscent samara associated with 2 woody bracteoles which open like a capsule; seed with straight embryo, without endosperm. Wind p olli-nated. Fruits are also dispersed by wind. Economic importance: The wood of several species is extremely hard and valued fo r fur-niture making. Casuarina equisetifolia (red beefwood) is most widely culti vated as ornamental tree. Phylogeny: The family is monophyletic, as are the four genera recognized indepen dently or combined into Casuarina. The family is considered to be a part of the Hamamelid complex, now included along with the broadly circumscribed Rosalean co mplex Rosanae (Dahlgren) or Rosidae (Thorne, 1999 under order Casuarinales; subs equently shifted to Hamamelididae >Hamamelidanae >Betulales in 2003, Major Families of Angiosperms 523 2006 and 2007). APG II and APweb include the family under Fagales (under Eurosid s), shifting Hamamelidaceae and some other families of the complex to Saxifragal es. Fagales are the core of the old Englerian, Amentiferae which have since been demol-ished, several members shifted to otherwise entirely unrelated groups with in the Eudicots (Qiu et al., 1998). The family Casuarinaceae was once considered to be the most primi-tive among dicots (Engler and Prantl) derived from Ephedraceae. The studies of wood anatomy and floral anatomy have sh own that it is sufficiently advanced, having undergone considerable reduction in

floral features and vegetative morphology. The genus was split into four genera indicated above (Johnson and Wilson, 1993). Gymnostoma is sister to the rest of the family and has many plesiomorphous features (both carpels fertile, with 2 o vules in each carpel). * * * * * * * * * * * Subclass 8. Caryophyllidae (B) Superorder 1. Berberidopsidanae 1. Achatocarpaceae 2. Chenopodiaceae 3. Amaranthaceae Order 1. Berberidopsidales 4. Caryophyllineae 1. Aextoxicaceae 1. Caryophyllaceae 2. Berberidopsidaceae 3. Polygonales Superorder 2. Caryophyllanae 1. Polygonaceae Order 1. Physenales 2. Plumbaginaceae 1. Physenaceae (B) 4. Tamaricales 2. Asteropeiaceae (B) 1. Tamaricaceae 2. Caryophyllales 2. Frankeniaceae 1. Cactineae 5. Nepenthales (B) 1. Portulacaceae 1. Ancistrocladaceae 2. Dioncophyllaceae 3. Drosophyllaceae (B) 4. Droseraceae 2. Halophytaceae 3. Cactaceae 5. Basellaceae 6. Didiereaceae 5. Nepenthaceae 2. Phytolaccineae Superorder 3. Dillenianae 1. Rhabdodendraceae (A) 2. Simmondsiaceae (B) 3. Stegnospermataceae 4. Limeaceae (B) 5. Nyctaginaceae 6. Sarcobataceae (B) 7. Petiveriaceae 8. Agdestidaceae 9. Phytolaccaceae 10. Gisekiaceae 11. Lophiocarpaceae (B) 12. Aizoaceae 13. Barbeuiaceae Order 1. Dilleniales 1. Dilleniaceae Superorder 4. Santalanae Order 1. Santalales 1. Olacaceae 2. Medusandraceae 3. Misodendraceae 4. Loranthaceae

5. Opiliaceae 6. Santalaceae 7. Viscaceae Superorder 5. Balanophoranae 14. Molluginaceae (B) Order 1. Balanophorales 3. Chenopodiineae 1. Balanophoraceae 524 Plant Systematics Portulacaceae A. L. de Jussieu Purslane family 28 genera, 440 species Widely distributed in tropical and temperate regions, mainly North and South Ame rica. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Thalamiflorae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Caryophyllineae Caryophyllales Caryophyllales Caryophyllales Caryophyllales Cary ophyllales G D Figure 13.64 Portulacaeae. Porbulaca oleracea. A: Portion of plant with flowers; B: Flowers seen from above; C: Flower enlarged in vertical section; D: Stamen; E: Transverse section of ovary with free central placentation;. F: Seed; G: Embryo. Major Families of Angiosperms 525 Salient features: Usually succulent herbs, mucilaginous, roots somewhat tuberous , leaves simple, flowers in cymes or solitary, bisexual, sepals 2, carpels unite d with unilo-cular ovary, fruit a capsule, embryo curved. (Portulaca oleracea) commonly growing wild is frequently cultivated as pot herb. Many species of Montia (Miner s lettuce) were used earlier as green salad in Amer ica. Root stalks of Lewisia rediviva are also eaten in America. Rose moss (Portu laca grandiflora), Flame flower (Talinum spp.) and Rock purslane (Calandrinia sp

p.) are grown as ornamentals. Major Genera: Calandrinia (120 species), Portulaca (100), Claytonia (35), and Ta linum (30). Description: Annual or perennial herbs, somewhat succulents, Mucilage cells very common, containing betalains, often exhibiting CAM metabolism, hairs usually si mple. Stems erect or prostrate, herbaceous. Leaves alternate or opposite, simple , usually fleshy, entire, often clustered at ends of branches, stipules scarious or setose, rarely absent. Inflorescence cymose with few or of solitary flowers, sometimes racemose. Flowers showy, bisexual, usually actinomorphic, with short or elongated hypanthium. Calyx of 2 sepals, antero-posterior, green, free or uni ted at base. Corolla with usually 5 petals, rarely 4 or 6, free, rerely united a t base, imbricate, falling early. Androecium with usually as many stamens as pet als, opposite petals, filaments free from petals or epipetalous, dehiscence long itudinal, pollen tricolpate, polycolpate or polyporate. Gynoecium with 2 to 3 (r arely more) united carpels, ovary superior or half-inferior, single chambered, w ith single basal ovule or several ovules on free-central placenta attached at th e base of ovary, style simple or split above, stigma minute. Fruit a loculicidal or circumscissile capsule; seeds lens-shaped, smooth, shining, embryo curved, e ndosperm absent, perisperm present, aril sometimes developed. Pollination by ins ects like bees, flies and beetles, flowers opening briefly in full sunlight. See ds with aril dispersed by ants, smaaler ones by wind or water. Phylogeny: The family Portulaceae has traditionally been considered closely rela ted to Caryophyllaceae and Basellaceae, although the presence of betalains has o ften taken this family away from Caryophyllaceae. Phylogeny of Portulacaceae has been a mat- ter of considerable speculation. The separa- tion of Basellaceae an d Didiereaceae is sup- ported by morphological data. Cactaceae is considered as more closely related to Cactaceae (although separation is supported by ITS seque nce data) and Thorne (2006) has accordingly shifted family under suborder Cactin eae. Applequist and Wallace (2001) on the basis analysis of chloroplast gene ndh F in Portulacaceae, Basellaceae, Cactaceae, and Didiereaceae concluded that the group forms a monophyletic group with two major clades. The first included Portu laca, Anacampseros and its relatives, much of Talinum, Talinella, and Cactaceae; the sec-ond, weakly supported, included the remain-ing genera of Portulacaceae, Basellaceae, and Didiereaceae. The separation of these families from Portulacea e renders it paraphyletic. Subsequent studies of these authors (2006) resulted i n Stevens (2007) placing only genus Portulaca in the family Portulaceae, separat ing 10 genera (incl. Montia, Lewisia and Phemeranthus) to Montiaceae and two (Ta linum, Talinella) to Talinaceae. Thorne who earlier (2003, 2006) recognized Hect orellaceae as a distinct family has in the latest revision (2007) merged it with Portulacaceae. Economic importance: The family is of little economic importance. Purslane * * * * * * * * * * * 526 Plant Systematics Cactaceae A. L. de Jussieu Cactus family 122 genera, 1,810 species Mainly in arid climate, in deserts regions of North and South America, several s pecies introduced in Africa, India and Australia. Placement:

B & H Cronquist Takhtajan Dahlgren

Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Calyciflorae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Ficoidales ophyllales Caryophyllales Caryophyllales Caryophyllales Caryophyllales Cary

B & H as Cacteae Salient features: Succulents, fleshy habit, usually spiny herbs or shrubs, spine s ar-ranged in areoles, flowers solitary, petals many, stamens numerous, ovary i nferior, fruit a berry. Androecium with numerous stamens, free or adnate to the base of petals, bithecou s, dehiscence longitudinal, introrse, pollen grains tricolpate to polycolpate or polyporate. Gynoecium with 2 to numerous carpels, united, ovary inferior, rarel y semi-inferior (Pereskia) or even superior (some species of Pereskia), unilocul ar with numerous ovules, placentation parietal, sometimes divided by false septa or nearly basal (Pereskia), stigmas 2 to numerous, spreading, ovules campylotro pous, bitegmic, crassinucellate. Fruit a berry, often covered with spines and/ o r glochids; seeds numerous, immersed in pulp, testa often black, endosperm usual ly absent, embryo usually curved. Pollination by insects, birds or bats. Berries are dispersed by animals or birds. Major Genera: Opuntia (250 species), Mammillaria (190), Echinopsis (75), Cereus (55), Rhipsalis (50) and Cleistocactus (50). Description: Spiny stem succulents, herbs, sometimes tree-like, rarely non-succu lent (but with fleshy leaves-Pereskia) or epiphytic (Rhipsalis), stem cylindrica l or angled, some-times flattened, or even jointed, usually pho-tosynthetic, usu ally with vessels, some-times without vessels, usually without laticifers, rarel y with laticifers (Coryphantha), plastids PIII-A type. Leaves usually borne on l ong shoots and readily falling, alternate sim-ple, entire with pinnate or obscur e venation, leaves sometimes represented by spines, or absent, short shoots (are oles) with clusters of spines and tufts of hairs (glochids); stip-ules absent. I nflorescence with solitary flow-ers, usually sunk at the apex of branch and thus appearing axillary, rarely in clusters (Pereskia). Flowers bisexual, usually ac tinomorphic, with short or elongated hypanthium. Perianth sequentially intergrad ing from sepals to petals or all petaloid, spirally arranged, numerous, in-nermo st slightly coherent at base. Economic importance: The family is known for large number of ornamentals (cacti) such as Opuntia (prickly pear), Mammillaria (pin-cushion cactus), Cereus (hedge cactus), Echinopsis (sea-urchin cactus), Epiphyllum (orchid cactus), Schlumberg era (Christmas cactus) and Rhipsalis (mistletoe cactus). Fruits of several speci es of Opuntia are eaten raw or made into jams or syrups. Spines of cacti are oft en used as gramophone needles. Lophophora contains mescaline alkaloids and is ha llucinogenic. Cochineal dye is derived from small insects living on members of t his family. ~~3mJ101ri.y RIM-f~4' ~ ,

Figure 13.65 Cactaceae. Opuntia rafinesqui. A: Portion of plant with flowers and spines; B: Vertical section of flower showing many stamens, inferior ovary and parietal placenta tion; C: Ovules with long funiculus; D: Longitudinal section of fruit; E: Longit udinal section of seed. F: Carnegiea gigantea with characteristic branched habit and ribbed stem. Phplogenp: The family is unique in com-bining unspecialized floral characters wi th highly advanced vegetative organs. The fam-ily is commonly divided into three subother two subfamilies form a well-defined clade with solitary flowers sunken int o stem apices, inferior ovary and parietal placentation. Opuntioideae are mono-p hyletic based on synapomorphies of presfamilies: Pereskioideae, Opuntioideae and Cactoideae (Heywood, 1978). Thorne (1999, ence of glochids on the areoles, seeds coat 2003, 2006, 2007) places the two genera of with bony aril and cpDNA charac ters. Pereskoideae under two separate sub- Cactoideae similarly has monophyly su p-families Pereskioideae (Pereskia) and ported by extreme reduction of leaves an d a Maihuenioideae (Maihuenia), thus recogniz- deletion of the rpoCl intron in c hloroplast ing 4 subfamilies in all. Pereskia retains genome (Wallace and Gibson , 2002). The afseveral plesiomorphic features such as non- finities of Cactaceae have largely remained succulent stems, well-developed persistent uncertain. It is now considered to be closely leaves, cymose inflorescence, superior ovary rel ated to Portulacaceae, Phytolaccaceae, (in some species) with basal placentation , Basellaceae, Halophytaceae, Didiereaceae and is sister to rest of the Cactacea e. The and Aizoaceae. Phylogenetic relationships 528 Plant Systematics within Cactaceae are still rather unclear, with chloroplast and nuclear genes so me-times suggesting different major clades. A recent study by Nyffeler (2002) fo und rather weak support for the subfamilies and per-haps rather distressingly no clear monophyly for the basal Pereskia. The distinctive Blossfeldia was sister to the other Cact oideae. Edwards et al. (2005) confirm that Pereskia s.l. is probably paraphyleti c, which allows them to shed new light on the evolution of the cactus habit * * * * * * * * * * *

Nyctaginaceae A. L. de Jussieu Four O Clock family 31 genera, 400 species Widely distributed in tropical and subtropical regions, mainly in the New World. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Curvembryeae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Caryophyllales Caryophyllales Caryophyllales Caryophyllales Caryophyllales Salient features: Swollen nodes, wood quickly turning reddish-brown when cut, le aves usually opposite, sometimes unequal, without stipules, vascular bundles in con-centric rings, flowers bisexual with con-spicuous bracts, bracts often enlar ged, peri-anth 5 with long tube, stamens usually 5, carpel single, ovule single, basal, fruit an achene or nut. to reddish-brown when cut. Roots branched taprrot, sometimes thick and tuberous (Mirabilis) or fusiform (Boerhavia). Leaves opposite, leaves of a pair equal or unequal, simple, usually entire, net-veined, without stipules. Inflorescence wit h usually cymose clusters, usually dichasial, subsequently monochasial. Flowers bisexual, complete, actinomorphic, subtended by an involucre of 3 (Bougainvillea ) to 5 (Mirabilis) bracts which are often enlarged and coloured (Bougainvillea), sometimes reduced to small teeth (Boerhavia), hypogynous, flowers rarely unisex ual (Pisonia). Perianth with 5 united tepals, campanulate (Boerhavia) or more co mmonly tubular, lower part of tube persist-ent and surrounding fruit (and known as anthocarp), distal part usually petal-like and falling off. Androecium with 3 -5 (Mirabilis), 5-10 (Bougainvillea) , or more, filaments free or connate, equal or unequal, anthers Major genera: Neea (80 species), Guapira (70), Mirabilis (45), Pisonia (40), Abr onia (30), Boerhavia (20) and Bougainvillea. Description: Herbs (Boerhavia, Mirabilis) of-ten with swollen nodes, shrubs (Pis onia) or woody climbers (Boungainvillea), rarely small trees (Pisonia alba), usu ally with concentric rings of vascular bundles, containing betalains and raphide crystals of calcium oxalate, woody oxidising, i.e., turning orange Major Families of Angiosperms 529 Figure 13.66 Nyctaginaceae. Bougainvillea glabra A: Portion of plant with flower s; C: Carpel; B: Longitudinal section of flower of B. spectabilis showing perianth and stamens . Mirabilis jalapa D: A portion of branch with flowers; E: Longitudinal section of of ovary with subtending bracts; F: Anthocarp. Boerhavia repens G: Portion of branch with flowers and fruits; H: Longitudinal section of flower to show peria nth, stamens and ovary; I: Anthocarp.

bithecous, basifixed or dorsifixed, dehiscence longitudinal or by lateral slits, pollen grains tricolpate or polyporate. Gynoecium with single carpel, ovary sup erior, but often ap-pearing inferior due to closely associated persistent part o f perianth tube, placentation basal, ovule single, anatropous or campylotropous, style long and filiform, stigma capitate, nectar disc present. Fruit an achene or nut, usually enclosed in leath-ery or fleshy persistent perianth tube, lat-te r winged or or ribbed, often covered with glandular hairs; seed small with unequ al cotyledons, curved embryo, endosperm ab-sent, replaced by perisperm. Flowers are pollinated by bees, butterflies, moths and birds. Dispersal in species with fleshy perianth occurs by birds, those with glands and hooked hairs by exozoochory. Economic importance: Species Bougainvillea are commonly grown as hedges and for covering walls and fences. Species of Mirabilis (Four O Clock) are grown as garden ornamentals. Boerhavia repens is used as medicinal plant as a diuretic. Fisonia aculeata is used as hedge plant. Phylogeny: Hiemerl (1934) divided the family into 5 tribes: first four with glab rous 530 Plant Systematics ovary and stamens connate at base, and the fifth Leacastereae with hairy ovary a nd distinct stamens. Of the first four, Miralileae has straight embryo and large cotyledons, Pisoneae with shrubby habit, Boldoeae with herbaceous habit and alt ernate leaves, and Colignoneae with herbaceous habit and opposite leas; all thre e have curved embryo. The family is closely related to Phytolacca-ceae in anatom ical features, and ovary with single carpel. The close affinities of the two fam ilies together with genus Delosperma (Aizoaceae) were confirmed by the studies of D. Soltis et al. (2000). Monophyly of the family found moderate support in the studies of Douglas and Manos (2007). Stevens (2008) recognises three more or less monophyletic groups within the fam ily: Leucasterae (4 genera), Boldoeae (3 genera) and the Rest (24-25 genera). Th e South American Leucastereae and Mexican-Central American Boldoeae are successi vely sister taxa to the remainder of the family, positions that have moderate to strong support. * * * * * * * * * * * Aizoaceae Martynov Stone Plant family 122 genera, 1,790 species Widely distributed in tropical and subtropical regions, mainly in arid and coast al regions of South Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Calyciflorae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order

Ficoidales ophyllales

Caryophyllales Caryophyllales Caryophyllales Caryophyllales Cary

B & H as Ficoideae Salient features: Succulent herbs or small shrubs, usually with many stems, leav es with large bladder-like cells in epidermis, petals several of staminodal orig in, stamens usually many, carpels 2 or more, united, fruit a capsule, embryo cur ved, surrounded by mealy endosperm. from base, sometimes mat-forming, vascular bundles in concentric rings, with bet alains, alkaloids and raphide crystals of calcium oxalate, usually with CAM meta bolism. Leaves alternate or opposite, leaves of a pair equal (Sesuvium) or stron gly unequal (Trianthema) with branches arising from axil of smaller leaf, simple , usually entire and succulent, veins somewhat obscure, stipules scarious or fri nged, rarely absent, epidermis with large bladder-like cells. Inflorescence of c ymose axillary cymes, in pairs or solitary, terminal or axillary. Flowers bisexu al, subtended by a pair of laciniate bracts often fused with calyx tube, regular , with hypanthium. Calyx with 5 connate sepals, imbricate, Major genera: Conophytum (250 species), Delosperma (150), Lapranthus (150), Dros anthemum (100), Antimima (60), Lithops (40) Mesembranthemum (30), Trianthema (20 ), Sesuvium (8), Description: Succulent herbs or small shrubs, commonly with numerous stems Major Families of Angiosperms 531 Figure 13.67 Aizoaceae. Trianthema portulacastrum. A: Portion of plant with flow ers; B: Flower with two bracts and horned calyx lobes; C: Capsule; D: Seed with papilae. Sesuvi um verrucosum E: Portion of plant with flowers F: Flower with horny calyx. G: Se ed of S. maritimum.(A,B after Maheshwari, Fl. Delhi, 1963; D-G after Godfrey & W ooten, Aq.Wetland Pl. SE US, Vol. 2, 1981). often with horny protuberance on back below tip, persistent in fruit, margin sca rious. Co-rolla absent, represented by petaloid staminodes, numerous. Androecium with 5 to numerous stamens, many outer modified into petaloid staminodes, ferti le stamens 5 or more, free or with connate filaments, arising from hypanthium, p ollen tricolpate. Gynoecium with 2-5 united carpels, ovary superior or inferior, 2-5 locular, with axile, parietal or basal placentation, ovules 1 to many, anat ropous to campylotropous, disc usually present. Fruit a capsule, loculicidal, se pticidal, or circumscissile, sometimes berry, rarely a nut; seed with large curv ed embryo, sometimes with aril, endosperm ab-sent, replaced by perisperm. Flower s are pol-linated by bees, wasps, butterflies and bee-tles. Seeds are dispersed by wind or water. Dorotheanthus and Carpobrotus are grown as garden ornamentals. Some like Lithops (stone plant) and Titanopsis are grown as curiosities. Tetragonia is used as ve getable. Some species help stabilize sand dunes and road banks. Economic importance: Species of Mesembranthemum (ice plant), Lapranthus, Phylogeny: The family is monophyletic, represented by distinct clades, often rec ognised as four subfamilies Aizooideae, Mesembranthemoideae, Sesuvioideae and Ru schioideae (Klak et al., 2004; Stevens, 2008; Thorne, 2007). Members with numero us petaloid staminodes, placed in Mesembryanthemoideae and Ruschioideae form a m onophyletic group (Hartmann, 1993). Sesuvioideae (Sesuvium and related genera) p robably form a clade based on circumscissile capsules and arillate seeds. Genus Mesembryanthemum was previously circumscribed to include more than 1000 species, but has subsequently been split into 532 Plant Systematics numerous genera, a bulk of genera shifted to Ruschioideae. Brown (1920) had orig inally split the genus into more than 100 genera, a suggestion not incorporated by Pax and Hoffmann (1934), but subsequently followed by authors as more materia l became avalable. There have been further attempts to divide the genus further into much smaller genera, an attempt that would resu lt in many poorly characterized genera. A detailed phylogenetic analysis of the

family by Klak et al. (2007) has resulted in better resolution affinities within the the subfamily Mesembryanthemoideae. * * * * * * * * * * * Chenopodiaceae Ventenat Goosefoot family 97 genera, 1,305 species Widely distributed in temperate and tropical climates but common in arid and sem iarid saline habitats. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Curvembryeae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Caryophyllales Caryophyllales Caryophyllales Caryophyllales Caryophyllales APG II and APweb under family Amaranthaceae Salient features: Herbs or small shrubs usu-ally in saline habitats, often cover ed with white bloom, stipules absent, cuticle waxes with platelets, flowers smal l often greenish, bracts and perianth herbaceous, stamens opposite perianth lobe s, all fertile and simi-lar, carpels 2, ovary superior, fruit a nut enclosed in persistent perianth, embryo curved. unilacunar, vascular bundles in concentric rings, included phloem usually presen t, sieve-tube plastids PIII-C type, containing betalains instead of anthocyanins , cuticle waxes with platelets. Leaves minute to large, alternate, rarely opposi te (Salicornia, Nitrophila), petiolate to sessile, simple, entire or variously l obed, sometimes fleshy or re-duced to scales, stipules absent. Inflorescence cym ose, spikes or panicles, sometimes catkins. Flowers small, greenish, bisexual, r arely unisexual and plants dioecious (Grayia) or monoecious, actinomorphic, hypo gynous. Perianth (represented by sepals petals absent) with 2-5 united tepals, r arely free (Salsola), herbaceous, usually persistent and accrescent in fruit, an d appendaged with tubercles, spines or wings, sometimes absent. Major genera: Atriplex (300 species), Salsola (120), Chemnopodium (105), Suaeda (100) and Salicornia (35). Descriptiomn: Herbs or small shrubs, rarely small trees (Haloxylon), usually in saline habi-tats, sometimes succulent (Salicornia), often covered with whitish b loom, nodes Major Families of Angiosperms 533 Figure 13.68 Chenopodiaceae. Chenopodium album. A: Portion of plant in flower; B : Flower partially opened with stamens still included; C: Fruit from above; D: Seed. Beta vu lgaris. E: Flower; F: cluster of Fruits. Suaeda maritima. G: Portion of plant in

flower; H: Flower. Androecium with 5 stamens, rarely 3, opposite the perianth lobes, filaments free , anthers inflexed in bud, bithecous, dehis-cence longitudinal, pollen grains mu ltiporate, spinulose. Gynoecium with 2 carpels, united, rarely carpels upto 5, o vary superior, unilo-cular, ovule 1, placentation basal, styles 2 (rarely upto 5 ). Fruit a nut or utricle (when enclosed in membranous perianth); seed lens shap ed with curved or spiral embryo, endosperm absent, perisperm present. a source of sugar), spinach (Spinacea oleracea) and lambs quarters (Chenopodium album; bathoo in Hindi). Chenopodium ambrosioides is source of wormseed used as a vermifuge. Seeds and leaves of C. quinoa are eaten by Peruvians and Andes. Phylogeny: The family has been considered distinct from Amaranthaceae in its her baceous perianth, all fertile stamens equal in length, and free filaments, but h as been merged with Amaranthaceae in recent APG classifications (Judd et al., 20 02; APG II; APweb) because their separation leads to paraphyletic Chenopodiaceae (Downie et al., 1997; Rodman, 1994; Pratt et al., 2001). Economic importance: The family includes a few food plants such as beet (Beta vu lgaris: used as leafy vegetable {often confused with spinach}; root vegetable ma inly for salad and 534 Plant Systematics Cunoud et al. (2002) found Chenopodiaceae were perhaps monophyletic, but the bran ch collapsed in a strict consensus tree; the sampling was moderately good, but o nly one gene - matK - was sequenced and analysed. Sarcobatus has long been ackno wledged an anomalous member of this family, e.g. by Bentham and Hooker (1880), w ho presented it as a monogeneric tribe. Behnke (1997) proposes raising it to fam ily rank, because sieve-element plastid form supports recent chloroplast DNA seq uencing studies in portraying it nearer Phytolaccaceae than Chenopodiaceae. The studies of Cunoud et al. (2002), however, consider it sister to Nyctaginaceae, although with weak sup port. Thorne (2003) placed Sarcobataceae near Nyctaginaceae under suborder Phyto laccineae, whereas Chenopodiaceae along with Amaranthaceae placed under Chenopod iineae. Subsequently (2006) he enlarged the latter order by including the third family Achatocarpaceae, earlier included under monotypic suborder Achatocarpinea e. He earlier (2006) divided Chenopodiaceae into four subfamilies: Chenopodioide ae, Micro-teoideae, Salicornioideae and Salsoloideae, adding fifth .Suaedoideae in 2007 revision. * * * * * * * * * * * Amaranthaceae M. Adanson Amaranth family 72 genera, 1,020 species Cosmopolitan, mainly tropical, centred in Africa and America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II/(APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Curvembryeae+

Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Caryophyllales Caryophyllales Caryophyllales Caryophyllales Caryophyllales Salient features: Herbs or small shrubs, stipules absent, flowers small often gr een-ish, subtended by scarious or papery bracts, perianth papery, stamens opposi te perianth lobes, slightly connate at base, staminodes present, carpels 2-3, ov ary superior, fruit a capsule or utricle or nutlet, enclosed in per-sistent peri anth, embryo curved. nodes unilacunar, vascular bundles in con-centric rings, included phloem usually present, sieve-tube plastids PIII-A type, con-taining betalains instead of anth ocyanins. Leaves alternate or opposite, herbaceous, sometimes aggregated at base (Ptilotus), petiolate to sessile, simple, entire, stipules absent. Inflorescenc e cymose, spikes or panicles, with conspicuous persistent bracts and bracteoles. Flowers small, greenish, bisexual (rarely unisexual), actinomorphic, hypogynous , cyclic. Perianth (represented by sepals petals absent) with 3-5 free or united tepals, usually persistent, sometimes accrescent (Ptilotus) in fruit, usually d ry and Major genera: Gomphrena (120 species), Alternanthera (100), Iresine (80), Amaran thus (60) and Celosia (55). Description: Herbs or small shrubs, very rarely climbing, often with swollen nod es, Major Families of Angiosperms 535 r Figure 13.69 Amaranthaceae. Amaranthus spinosus. A: Part of plant in flower; B: Cymose cluster with one male and several female flowers; C: Female flower with 3 carpels; D : Mature fruit of same with enlarged persistent perianth; E: Mature utricle deve loped from flower with 2 carpels, perianth removed; F: Seed. Achyranthes aspera. G: Part of plant in flower; H: Flower with bract and perianth removed; I: Andro ecium showing stamens and staminodes; J: Bract; K: Bracteoles; L: Utricle with p ersistent style. (A, G-L, after Sharma and Kachroo, Fl. Jammu, 1983). scarious. Androecium with 5 stamens, rarely 3 or even 6-10, opposite the tepals, filaments slightly connate at base, often adnate to tepals, anthers inflexed in bud, bithecous (Amaranthus) or monothecous (Gomphrena), dehiscence longitudinal , pollen grains multiporate, spinulose, staminodes often present, usually 1-3. G ynoecium with 2-3 united carpels, ovary superior, unilocu-lar, ovule usually 1, placentation basal, rarely many (Celosia), styles 1-3. Fruit a circumscissile ca psule, or nut or utricle (when enclosed in membranous perianth); seed lens shape d with curved or spiral em-bryo, endosperm absent, perisperm present. Economic importance: The family includes several ornamentals such as Celosia (Co ckscomb), Amaranthus (amaranth), Gomphrena (globe amaranth) and Iresine (bloodle af). Species of Alternanthera and Tilanthera are grown as edge plants and have o rnamental leaves. Seeds and leaves of several species of Amaranthus are edible, as are also the leaves of Alternanthera sessilis. Phylogeny: The family is closely related to Chenopodiaceae (which is placed with in Amaranthaceae in APG classifications) but differentiated in scarious bracts a nd perianth, connate stamens and presence of 536 Plant Systematics staminodes. Hutchinson (1926, 1973) believes the family to have evolved from car yophyllaceous ancestors. Cunoud et al., (2002) found Amaranthaceae s. str. to be monophyletic, with very strong (97 per cent) support. The family, in the broader sense (including Chenopodiaceae), is monophyletic as supported by morphology, c hloroplast DNA restriction sites and rbcL sequences. Pollen structure is also si milar with thickened tecta, apertures with reduced pointed flecks of

seems to be arbitrary. Others like Pratt et al. (2001) consider Amaranthaceae to be polyphyletic. Thorne (2003, 2006, 2007) treats Amaranthaceae as distinct fro m Chenopodiaceae and including three subfamilies: Polycnemoideae, Amaranthoideae and Gomphrenoideae. In analysis of Mller and Borsch (2005; 2005b) from analysis of matK/trnK sequences concluded that Polycnemum and Nitrophila (100% support) were sister to the rest; they have exine underlain by lamellar plates, and a ordinary secondary thickening, imperfe ct thickened endexine. The separation of flowers, basally connate filaments, and Chenopodiaceae and Amaranthaceae unithecate anthers. * * * * * * * * * * * Caryophyllaceae A. L. de Jussieu Pink family 93 genera, 2,395 species Distributed mainly in all temperate parts of the world. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Thalamiflorae+ Caryophyllanae Caryophyllanae Caryophyllanae Core Eudicots* Order Caryophyllineae Caryophyllales Caryophyllales Caryophyllales Caryophyllales Cary ophyllales Salient features: Herbs with swollen nodes, leaves opposite, inflorescence usual ly a dichasial cyme, corolla caryophyllaceous, stamens ten or lesser, obdiploste monous, ovary unilocular with free central placentation, superior, fruit a capsu le open-ing by valves or teeth. (Paronychia), secondary veins often obscure. Inflorescence typically a dichasial cyme, rarely solitary flowers. Flowers bisexual, rarely unisexual (Lychnis alba ), actinomorphic, hypogynous. Calyx with 5 sepals, free (Stellaria) or connate ( Dianthus, Silene). Corolla with 5 petals, usually differentiated into a distinct claw and a limb, with an appendaged joint between the two, often Major genera: Silene (700 species), Dianthus (300), Arenaria (200), Gypsophila (150), notched or deeply bilobed at tip. Andro ecium Minuartia (150), Stellaria (150) and Cerastium with 10 or lesser number of stamens, (100). Description: Annual or perennial herbs with swollen nodes, anthocyanins present. Leaves opposite, simple, bases of opposite leaves of-ten connected, stipules ab sent or scarious

obdiplostemonous, free, anthers bithecous, dehiscence longitudinal, pollen grain s tricolpate to polyporate. Gynoecium with 2-5 (2 Dianthus, 3 Silene, 4 Sagina, 3-5 Stellaria) united carpels (syncarpous), unilocular with many ovules, placent ation free central, ovary Major Families of Angiosperms 537

A I C i V H Figure 13.70 Caryophyllaceae. Stellaria media. A: A part of plant in flower; B: Flower showing hairy sepals and deeply bilobed petals; C: Flower with sepals and petals removed to show stamens and pistil; D: Bilobed petal; E: Mature capsule with persistent calyx; F: Capsule dehiscing through valves; G: Seed. Silene conoidea. H: A port ion of plant in flower; I: Capsule with half of calyx removed to show dehiscence through teeth and remnants of petals and stamens; J: Seed. (After Sharma and Ka chroo, Fl. Jammu, 1983). superior, styles 2-5. Fruit a loculicidal cap-sule opening by valves or teeth, r arely a utri-cle (Paronychia). Seeds many, ornamented on surface, embryo curved, endosperm absent, often replaced by perisperm. have been the subject of considerable debate, several authors (Mabry, 1963) advo cating separation of betalain containing families into a separate order, and tho se lacking betalains (but containing anthocyanins as in Caryophyllaceae and Plum baginaceae) into another. Ultrastructure studies of sieve tube plastids (Behnke, 1975, 1977, 1983) showed that all members ( those with and without betalains) c ontained unique PIII plastids, affinity reinforced by studies of DNA/RNA hybridi zation (Mabry, 1975), throwing up a compromise of including all families within the same order, but separate suborders, a trend being followed by Takhtajan upto 1987 but finally discarded in 1997. Thorne (1999, 2003)had established Economic importance: The family is rep-resented by several ornamentals such as c arnation, pinks, sweet william (different species of Dianthus), baby s breath (Gyp sophila) and corn cockle (Agrostemma). Species of Arenaria, Cerastium and Stella ria are troublesome weeds. Phylogeny: The family Caryophyllaceae along with other members of the order Cary ophyllales (Centrospermae of Engler) 538 Plant Systematics 5 suborders (Achatocarpineae, Cactineae, Phy-tolaccineae, Chenopodineae and Cary o-phyllineae) within Caryophyllales, Caryo-phyllaceae being placed under monotyp ic suborder Caryophyllineae. In subsequent revision (2006, 2007) he merged Achat o-carpineae with Chenopodiineae, thus rec-ognising only 4 suborders. He recognis es two subfamilies Illecebroideae and Caryo-phylloideae within the family. The family forms a well defined monophyletic clade, as evidenced by morphology, and rbcL sequence. True petals are lacking in the family, and in most cases oute r 4 to 5

stamens are transformed into petals. Centrospermae is a classical case of proving a point that there should not be too much reliance on any single character, and con-clusions should be drawn only after the findings are reinforced by studi es from other fields. * * * * * * * * * * * Polygonaceae Durande Buckwheat family 52 genera, 1,105 species Distributed mainly in all temperate parts of the Northern Hemisphere, a few spec ies in tropics, arctic region and the Southern Hemisphere. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II /(APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Caryophyllidae Caryophyllidae Magnoliidae Caryophyllidae Series+/Superorder Curvembryeae+ Polygonanae Polygonanae Caryophyllanae Core Eudicots* Order Polygonales Polygonales Polygonales Polygonales Caryophyllales Salient features: Mostly herbs with swollen nodes, stipules forming ochrea at no des, flow-ers in spikes, heads or panicles, perianth usually petaloid, stamens 3 -8, carpels 3, united, ovule solitary, fruit a nut. simple, usually entire, venation pinnate, re-ticulate, stipules connate to form ochrea at node (ochrea absent in Eriogonum). Inflorescence in axillary cymose cl usters or form-ing spikes, heads or panicles. Flowers bisexual, rarely unisexual (Rumex), actinomorphic, hypogynous, showy (Antigonon) or inconspicuous (Rumex). Perianth with 6 tepals, in two whorls, usually petaloid, sometimes 5 due to fus ion of 2 tepals (Polygonum), free or slightly connate, imbricate, persistent, in ner whorl often enlarged in fruit and tubercled (Rumex) or not (Oxyria), periant h often 4 in two whorls (Oxyria). Androecium with usually 6 stamens, 8 in Fagopy rum, 9 in Rheum, filaments free or slightly connate, anthers bithecous, dehiscen ce longitudinal, pollen grains tricolpate to polyporate. Gynoecium Major genera: Eriogonum (250 species), Rumex (200), Persicaria (150), Coccoloba (120), Polygonum (60), Rheum (50) and Fagopyrum (15). Description: Annual or perennial herbs, shrubs, small trees (Triplaris) or climb ers with tendrils (Antigonon), with swollen nodes, usually with tannins, without laticifers, nodes pentalacunar or multilacunar, sieve-tube plastids S-type. Lea ves usually alter-nate, rarely opposite (Pterostegia) or whorled (Eriogonum), so metimes reduced (Coccoloba) Major Families of Angiosperms 539 Figure 13.71 Polygonaceae. Rumex nepalensis. A: Portion of plant in flower; B: F lower with par-

tially emergent anthers; C: Fruit with broad wings and hooked teeth. Polygonum t ortuosum. D: Portion of plant in flowers; E: Flower with included stamens; F: Fl ower from top; G: Seed. (After Polunin and Stainton, Fl. Himal., 1984). with 2-3 united carpels, ovary superior, unilocular with 1 ovule, orthotropous, place-ntation basal, sometimes partially divided by false septa, styles 2-3, a n ectary surround-ing the base of ovary, or paired glands asso-ciated with filamen ts. Fruit a trigonous or bifacial nut; seed with straight or curved embryo, endo sperm copious, mealy. Pollina-tion mostly by bees and flies. Fruits are usually dispersed by wind or water. salad. Species of Coccoloba, Antigonon, Muehlenbeckia, and Polygonum are often c ultivated as ornamentals. Fruits of Coccoloba are often used to make jellies. Le aves of Rumex acetosa and R. crispus are eaten as vegetables. Phylogeny: The basic floral pattern of the family according to Laubengayer (1937 ) is trimerous and whorled and the apparent spiral condition in some members is actually whorled as shown by anatomical study. In members with 5 tepals, one out er and one Economic importance: A few species of the family are of economic importance. Buck- inner tepal are fused. Laubengayer beli eved wheat (Fagopyrum) is an important source of the family to be allied and mor e advanced food (millets) in some areas. The petioles of than Caryophyllaceae an d the seemingly rhubarb (Rheum rhaponticum) are used as basal placentation of th e family is derived 540 Plant Systematics from free central placentation, the funicu-lus representing the greatly reduced free central placenta. Studies of Lamb Frye & Kron (2003) suggest that five peta ls is basic sheathing ochrea and Eriogonoideae with opposite leaves and more or less cymose and involucrate inflorescence. Thorne (1999, 2003) recognized third subfamily Co ccoloboideae characterized by reduced leaves, stems flattened and photos-yntheti c, for which, however, he subsequently (2006, 2007) preferred the name Brunnichi oideae. The members of Eriogonoideae have generally 6 tepals and may form a basa l paraphyletic complex (Cunoud et al., 2002; Lamb Frye & Kron 2003), thus justify ing merger with polygonoideae. condition in the family. The family is easily recognized and clearly monophyletic. The family is consider ed closer to Plumbaginaceae and according to Williams et al., (1994) although no plumbagin had been reported from the family, other quinones were found. Two sub families are commonly recognized in Polygonaceae: Polygonoideae with spiral leav es with * * * * * * * * * * * Droseraceae Salisbury Sundew family 3 genera, 110 species Widely distributed, mainly in marshy places low in nutrients, more commonly repr esented in Australia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta

Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Polypetalae Dilleniidae Dilleniidae Magnoliidae Caryophyllidae Caryophyllanae Core Eudicots* Series+/Superorder Thalamiflorae+ Nepenthanae Th Order Parietales Nepenthales Droserales Theales Caryophyllales Caryophyllales B & H under Sarraceniaceae; Dahlgren under Nepenthaceae Salient features: Insectivorous plants, leaves in rosettes, covered with sticky in-sect-catching hairs, circinate in bud, flow(Aldrovanda), annual or perennial. Leaves usually in basal rosette, upper altern ate, circinate in bud, simple, blade modified into hinged trap (Dionaea) or cove red with sticky glandular hairs (often called tentacles) for trapping insects (D rosera), venation obscure, stipules absent or present, petiole winged in Dionaea . Inflorescence scapigerous, determinate, cymose, often appearing like umbel, ra ceme or panicle, rarely solitary. Flowers bisexual, actinomorphic, hypo-gynous, pentamerous. Calyx with 5 sepals, ers bisexual, pentamerous, inflorescence determinate, stamens 5-many, carpels 35, fruit a capsule. Major genera: Drosera (108 species), Dionaea (1) and Aldrovanda (1). Description: Insectivorous herbs of bogs and wetlands, or submerged aquatic plan t Major Families of Angiosperms 541 09 B

J i o

`rt

D H Figure 13.72 Droseraceae. Dionaea muscipula. A: Plant with flowers; B: Leaf with winged petiole

and hinged leaf blade with marginal bristles; C: Flower; D: Ovary cut open to sh ow ovules; E: Pollen tetrad; F: Seed. Drosera intermedia G: Plant with flowers; H: Leaf blade with glandular hairs; I: Flower, top view; J: Seed covered with pa pillae. (After Godfrey and Wooten, Aq. Wetland Pl SE US, 1981). rarely 4, slightly connate at base, imbricate, persistent. Corolla with 5 petals , rarely 4, free, convolute, mostly white or pinkish. Androecium with usually 5 stamens talline-granular, seed variously reticulated or ornamented. Pollination usually by in-sects, protoandy resulting in outcrossing, but selfing may result when flo wers close by the end of the day. Small seeds are dispersed by wind or water. De tached leaves and inflores(Drosera), 10-20 in Dionaea, free or slightly connate at base, anthers bithecous , dehiscence longitudinal, pollen grains triporate to multiporate, released in tetrads. Gynoecium with 3 united carpels, rarely upto 5, ovary superior, unilocular with 3- many ovules, placentation parietal or basal, style single (Dionaea) or 3-5, each divided upto the base thus appearing 6-10 in number. Fruit a loculicidal ca psule; seed small, with straight embryo, endosperm present, cryscences may also produce new plants. The hinged leaves of Dionaea have two halves, each with marginal sensitive brist les. When an insect touches the bristles, the two halves close, entrapping the i nsect. The small glands on the leaf surface secrete enzymes, which digest the in sect. In Drosera, the gland-tipped sticky hairs when stimulated, entangle the in sect by bending 542 Plant Systematics inwards, pressing the insect against the leaf in order parietales near Violaceae and Ochnaceae on the basis of parietal placen tation. Insectivorous habit and aquatic habit exhibit homoplasy with family Lent ibulariaceae. Earlier the family also included 4th subwoody genus Drosophyllum, which has now been separated into a distinct family Drosophyllaceae (APG-II, 200 3; Thorne, 2006, 2007; APWeb, 2007), being sister to Dioncophyllaceae + Ancistro cladaceae. Within Droseraceae, Dionaea and Aldrovanda with snap-trap leaves and n=6 may be sister to rest of taxa. blade. Economic importance: The family is of lit-tle ecomic importance. Venus flytrap ( Dionaea muscipula) and various species of Drosera (Sundew) are grown as noveltie s. Leaves of Drosera yield a violet dye, but is no longer of commercial importan ce. Phylogeny: The family was earlier in-cluded under Sarraceniaceae (Bentham & Hook er), but is now recognised independ-ently. Wettstein (1907) placed Droseraceae Subclass 9. Rosidae (B) 2. Podostemales Superorder 1. Celastranae (B) 1. Bonnetiaceae 2. Clusiaceae Order 1. Celastrales 3. Elatinaceae 1. Celastraceae 2. Parnassiaceae 3. Lepidobotryaceae (B) 4. Podostemaceae (B) 3. Euphorbiales (B) Superorder 2. Violanae 1. Humiriaceae 2. Ctenolophonaceae 3. Hugoniaceae 4. Irvingiaceae (B) 5. Ixonanthaceae (B) 6. Corynocarpaceae (B)

7. Coriariaceae (B) 8. Linaceae 9. Rhizophoraceae 10. Erythroxylaceae 11. Euphorbiaceae 12. Puttrrananjjiivaceaaee (B) 13. Goeae (B) 14. Malpighiaceae 15. Peridiscaceae (B) 16. Chrysobalanaceae (A) 17. Balanopaceae 18. Trigoniaceae 19. Dichapetalaceae (B) Order 1. Violales Suborder 1. Violineae 1. Violaceae 2. Flacourtiaceae 3. Achariaceae (B) 4. Sa 5. Pssclaoeraacceae 6. urneaceae 7. MMaalleesshherbiaceae 8. Lophopyxidaceae (B) 2. Begoniineae 1. Cucurbitaceae 2. Begoniaceae 3. Datiscaceae 4. Tetramelaceae Superorder 3. Podostemanae (B) Order 1. Ochnales 20. Euphroniaceae (B) 1. Ochnaceae Superorder 4. Oxalidanae (B) 2. Quiinaceae Order 1. Oxalidales 3. Medusagynaceae 1. Huaceae 4. Caryocaraceae 2. Oxalidaceae Major Families of Angiosperms 543 3. Connaraceae 4. Anisophylleaceae (A) 5. Cunoniaceae 6. Brunelliaceae 7. Cephalotaceae (B) 8. Elaeocarpaceae 4. Surianaceae (B) Superorder 7. Myrtanae Order 1. Myrtales 1. Melastomatineae 1. Penaeaceae 9. Tremandraceae 2. Oliniaceae Superorder 5. Geranianae (B) 3. Rhynchocalycaceae 1. Zygophyllales (B) 4. Alzateaceae 1. Zygophyllaceae 5. Crypteroniaceae 2. Krameriaceae

6. Melastomataceae 2. Geraniales (B) 2. Myrtineae 1. Myrtaceae 1. Geraniaceae 2. Vochysiaceae 2. Hypseocharitaceae 3. Lythrineae (B) 3. Ledocarpaceae (B) 1. Lythraceae 4. Francoaceae 2. Onagraceae 5. Greyiaceae 3. Combretaceae 6. Melianthaceae 2. Crossosomatales (B) Superorder 6. Rosanae Order 1. Rosales 1. Staphyleaceae 2. Guamatelaceae (B) 3. Stachyuraceae 4. Crossosomataceae 5. Geissolomataceae 6. Aphloiaceae (B) 7. Ixerbaceae (B) 8. Strasburgeriaceae (B) 1. Rosaceae 2. Fabales 1. Quillajaceae (B) 2. Fabaceae 3. Polygalaceae Celastraceae R. Brown Spindle-Tree family 98 genera, 1210 species Widely spread, mainly in tropical and subtropical regions, a few species in temp erate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II /(APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Disciflorae+ Celastranae Rutanae Celastranae Eurosids I*

Order Celastrales Celastrales

Celastrales Celastrales Celastrales Celastrales Salient features: Trees, shrubs or climbers, leaves simple often leathery, stipu les usu-ally small, falling early or absent, flowers small, greenish, pentamerou s, in clusters, ovary superior with large fleshy disc at base fused with the base of stamens, ca rpels 2-5, seeds with brightly coloured coat or aril 544 Plant Systematics

a 1 J t

A -~ C ~' Figure 13.73 Celastraceae. Euonymus hamiltonianus. A: Portion of plant in flower ; B: Flower from above; C: Flower from base; D: Fruit . Celastrus paniculatus. E: Portion of plan t in flower; F: Flower; G: Fruit. (A-D, after Polunin and Stainton, Fl. Himal, 1 984). Major genera: Maytenus (200 species), Salacia (150), Euonymus (130), Hippocratea (120), Cassine (60) and Crossopetalum (50). solitaryor racemose. Flowers small, greenish or greenish-white, regular, bisexua l or unisexual, hypogynous pergynous or epigynous. Calyx with usually 4-5 small se-pals, distinct or connate at base, rarely 3, imbricate, rarely valvate. Corol la with 4-5 free petals, rarely 3, somewhat similar to sepals, rarely absent. An droecium with usually 3-5 stamens rarely many (Plagiopteron), free, often attach ed at base to enlarged disc, rarely connate at base, anthers bithecous, dehiscen ce longitudinal, pollen grains aperturate or colporate. Gynoecium with 2 5 unite d carpels, ovary superior, sometimes Description: Trees, shrubs or climbers (Hippocratea), stem smooth or with spines , with or withour laticifers, juice not coloured, nodes unilacunar, vessels with simple or scalariform end-walls, vestured pits absent. Leaves alternate (Mayten us) or opposite (Euonymus), simple, usually leathery, ser-rate, pinnately veined , stipules small and caducous or absent. Inflorescence of flat-topped axillary o r terminal clusters, rarely Major Families of Angiosperms 545 inferior (Empleuridium) due to elarged disc, placentation axile with 2-6 ovules

in each chamber, style short, terminal, stigma stigma capitate or lobed, dry typ e non-papil-late. Fruit a berry, drupe, capsule or samara; seed usually surround ed by brightly coloured aril, embryo large and straight, endosperm present. (2003) placed Celastales under superorder Celastanae of Rosidae. Thorne had earl ier placed this superorder as fifth after Rosanae, subsequently (2006, 2007) bri nging it to the begining of Rosidae. Celastales of Thorne (2006) included 4 fami lies Celastaceae, ParnassiaEconomic importance: Economically the fam-ily is of lesser importance, a few use d as ornamentals. Climber Celastrus scandens is grown for its attractive coloure d fruits and seeds. Various swpecies of Euonymus are grown for their attractive foliage, E. japonicus with shiny leathery leaves very popular as hedge plant alo ng pathways in temperate cli-mate. The toxic alkaloid maytansine (from Maytenus) , when delivered by antibodies, may have application in treating colon cancers. ceae, Lepidobotryaceae and Huaceae. Zhang & Simmons (2006) found that Huaceae we re sister to Oxalidales, with quite strong support (jacknife values over 80%); t hey suggest that Huaceae should be included in Oxalidales. Thorne (2007) and Ste vens( APWeb, 2008), as such exclude Huaceae from the order Celastrales. Whereas Stevens keeps it unplaced within Eurosids I, Thorne has shifted it under order O xalidales of superorder Oxalidanae. Thorne (2007) recognizes 4 sub-families with Celastraceae: Celastroideae, Hippocrateoideae, Macgregorioideae and Stackhousio ideae (Siphonodontoideae recognized as 5th subfamily in 2006 version having been merged with Celastroideae). The family appears to be more uniform with the remo val of Bhesa to Malpighiales (Zhang & Simmons 2006) and Perrottetia to Huerteale s (Crossosomatales of Thorne) near Tapiscia (M. Simmons in Matthews & Endress 20 05b). Interestingly, the inclusion of both these genera. Phylogeny: The family was earlier placed closer to Rhamnaceae, latter removed al ong with a few other families to Rhamnales by Hutchinson (1973). Dahlgren (1989) placed Stakhousiaceae, Lophopyxidaceae Card-iopteridaceae, Corynocarpaceae and Celas-traceae under Celastales under superorder Rutaneae. Both Takhtajan (1997) and Thorne * * * * * * * * * * * Violaceae Batsch Violet family 23 genera, 900 species Distributed widely, mainly in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II/ APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Rosidae Series+/Superorder Thalamiflorae+ Violanae

Violanae Violanae Eurosids I* Order Parietales Violales Violales Violales Violales Malpighiales 546 Plant Systematics

E H

. A

B Figure 13.74 Violaceae. Viola canescens. A: Plant with flowers; B: Vertical sect ion of flower; C: Gynoecium; D: Fruit with persistent calyx; E: Seed. V. tricolor. F: Portion o f plant with flowers; G: Vertical section of flower showing spurred lower petal; H: Stamen with spurred anther; I: Ovary with style and enlarged stigma; J: Tran sverse section of ovary with parietal placentation. Salient features: Herbs, leaves serrate, stipules present, flowers zygomorphic, bisexual, petals 5, anterior spurred, anthers with spur-like nectaries, carpels 3, united, placentation parietal, fruit a loculicidal capsule, seeds large dispe rsed explosively as the fruit wall closes round them and squeezes them out. entire to serrate, venation pinnate or palmate, veins often conspicuous, stipule s present, sometimes foliaceous (Viola). Inflorescence usually with solitary axi llary flower, sometimes in racemes or spikes. Flowers bisexual, rarely unisexual , actinomorphic (Rinorea) or zygomorphic (Viola), hypogynous, pentamerous, somet

imes cleistogamous. Calyx with 5 sepals, usually free, sometimes slightly connat e to form a ring around ovary, imbricate, persistent. Corolla with 5 petals, fre e, imbricate or convolute, unequal, anterior usually largest and spurred or sacc ate. Androecium with 5 stamens, filaments short, free or slightly connate at bas e, anthers erect, somewhat connivent forming a ring around ovary, 2 anterior ant hers often with spur-like nectaries, connective often Major genera: Viola (450 species), Rinorea (280), Hybanthus (110), Anchietia (8) and Leonia (6). Description: Herbs (Viola) shrubs (Rinorea) or trees (Rinorea maingayi), rarely climbers (Anchietia) with often saponins or alkaloids. Leaves alternate rarely o pposite (Hybanthus), mostly basal, simple, sometimes lobed, Major Families of Angiosperms 547 with triangular appendage, dehiscence lon-gitudinal, introrse, pollen grains usu ally tricolpate. Gynoecium with 3 united carpels, carpels rarely 2-5 (Melicystus ), ovary superior, unilocular with parietal placentation, ovules many, anatropou s, style 1, stigma often ex-panded but with small receptive region, some-times l obed. Fruit a loculicidal capsule; seeds with straight embryo, endosperm and ari l present. Pollination by insects, attracted by nectar in the spur. Seeds are of ten dispersed explosively as the fruit wall closes round them and squeezes them out. Ants also disperse seeds, attracted by oily aril Phylogeny: The family is clearly defined and uniformly placed in most classifica tions under Violanae of Dilleniid complex. Hutch-inson who had largely separated dicotyledons into woody (Lignosae) and herbaceous (Herbaceae) lineages, had esp ecially chosen to justify the position of largely herbaceous Viola in the predom inantly woody clade, and considered the herbaceous habit in this genus to be der ived from woody ancestors. APG II and APweb, however place this family in a broa dly circumscribed order Malpighiales. Two tribes are commonly recognized: Rinore ae with mainly actinomorphic flowers and Violeae with zygomorphic flowers. Altho ugh the family is clearly monophyletic, rbcL sequences suggest that neither trib e is monophyletic. Violaceae are weakly associated with Acharaciaceae (and Goupi aceae, Lacistemataceae and Ctenolophonaceae) in Chase et al., (2002). Thorne, wh o had earlier (1999, 2000) placed Violanae under subclass Dilleniidae has now (2 003) shifted it to Rosidae, Dilleniidae having been dismantled. He (2006, 2007) recognizes three subfamilies Violoideae, Leonioideae and Fusispermoideae under V iolaceae. Economic importance: The family is mainly known for ornamental pansy flowers (Vi ola) and green-violet (Hybanthus). Flowers of Viola odorata is largely grown in France for essen-tial oil used in the manufacture of perfumes, flavourings and t oiletries. The flowers are also preserved in sugar ( banafsha ). Hybanthus ipecacuan ha has been used as substitute for true ipecac (Psychotria ipecacuanha) as emeti c. Roots of Anchietia salutaris are used as an emetic and to treat sore throats and lymphatic tuberculosis. The roots of Corynostylis hybanthus are used as an e metic. * * * * * * * * * * * Salicaceae Mirbel Willow family 2 genera, 485 species Distributed widely, mainly in north temperate to arctic regions, in moist open h abitats. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne

APG II/ (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Dilleniidae Dilleniidae Magnoliidae Rosidae Series+/Superorder Ordines anomali+ Violanae Violanae Violanae Eurosids I* Order Salicales Salicales Salicales Violales Malpighiales 548 Plant Systematics B v ry 9

Figure 13.75 Salicaceae. Populus ciliata. A: Portion of a vegetative branch; B: Branch with male catkins; C: Fruiting female catkin. Salix alba. D: Portion of a shoot with femal e catkins; E: Portion of shoot with male catkins; F: Male flower with hairy brac t and 2 stamens; G: Female flower with bract and stipitate ovary; H: Longitudina l section of ovary with basal placentas. S. caroliniana. I: Male flower with bract, 2 nect ar glands and many stamens; J: Transverse section of ovary with parietal placent ation. Salient features: Deciduous trees and shrubs, leaves with salicoid teeth, stipul es conspicuous, flowers unisexual, infloressalicoid (vein entering tooth and associated with glandular seta), venation pinn ate to palmate, reticulate, stipules present, sometimes foliaceous and persisten t. Inflorescence erect or pendulous catkins, on short branches. Flowers unisexua l (plants dioecious), actinomorphic, reduced, usually subtended by hairy bracts. Calyx reduced to a glandular disc (Populus) or 1 to 2 fringed nectar gland (Sal ix). Corolla absent. Androecium with 2 to numerous stamens, filaments free or sl

ightly connate at base, anthers bithecous, dehiscence by cence a catkin, flowers naked, carpels 2, ovules many, seeds with hairs. Major genera: Salix (445 species; incl. Chosenia) and Populus (40). Description: Deciduous trees and shrubs containing phenolic heterosides salicin and populin, containing tannins. Leaves alter-nate, simple, serrate to dentate, teeth Major Families of Angiosperms 549 longitudinal slits, pollen grains usually tricolpate or triporate, rarely inaper turate. Gynoecium with 2-4 carpels, united, ovary superior, unilocular with pari etal placentation or with 2-4 basal placentas, ovules many, unitegmic, styles 24, stigmas 2-4, capitate, often expanded and lobed. Fruit a loculicidal capsule; seeds with a basal tuft of hairs, endosperm scanty or absent. Polli-nation by w ind, flowers of salix are pollinated by insects attracted by nectar. Seeds are o f-ten dispersed by wind aided by hairs. appropriate to designate them as perianth. He also separated the family under or der Salicales. This treatment has been followed by Hutchinson (1973; placing Sal icales after Hamamelidales), Cronquist (1988; placing Salicales under Dilleniida e after Violales and not in Hamamelidae), Dahlgren (1989; under Violanae after V iolales and Cucurbitales), Takhtajan (1997; DilleniidaeViolanae after Violales, Passiflorales and Caricales). Thorne (1999) placed it under Dilleniidae >Violanae >V iolales after Economic importance: The family is impor-tant for several species grown as ornam entals, usually avenue trees. Cricket bats and polo balls are usually made from willow (Salix spp.) wood. Twigs of willow are commonly used in basket making. Th e bark of Salix contains salicylic acid, which re-duces swellings and fever, and is constitu-ent of aspirin. Violaceae and Flacourtiaceae, subsequently (2003, 2006, 2007) placing it under R osidae owing to the abolition of Dilleniidae. APG II and APweb include this fami ly under Malpighiales (Eurosids I) towards the begin-ning after Achariaceae and Violaceae (APweb) or nearly towards the end (APG II). Phylogeny: The affinities of this family were not known to Bentham and Hooker wh o placed it along with other uncertain fami-lies under Ordines anomali of Monoch lamydeae. The reduced flowers of Salicaceae (and other members of Amentiferae) w ere considered to represent primitive dicots by Engler (1892) and Rendle (1904, 1930). Fisher (1928) on the basis of extensive studies concluded that the sim-pl icity of flowers is largely due to extreme reduction and not a representation of ar-chaic features, and that flowers in the an-cestral form possessed a perianth of 1 or 2 series, which is now represented by a cupule-like gland. Hutchinson ( 1926), placed the family under Hamamelidales, treating it as the most primitive within the group. Hjelmquist (1948) was of the opinion that the cup or finger li ke gland in the flower was formed by the reduction of an undifferenti-ated bract eal envelope and that it is not quite The family is clearly monophyletic, having affinities with Flacourtiaceae, which also exhibit salicoid teeth, presence of salicin and apetalous flowers in some genera. Molecular data also support close affinities. The family Salicaceae as s uch has been broadly circumscribed in APG II and APweb (55 genera, 1,010 species ) to include larger part of Flacourtiaceae and smaller diverse families such as Bembiciaceae, Homaliaceae, Poliothyrsidaceae, Prockiaceae, Samydaceae and Scypho steg-iaceae. Broadly circumscribed Salicaceae is defined by leaves with salicoid teeth, cocarcinogens and flowers in which sepals and petals if present are equa l in number, those with sepals and petals not equal shifted to Achariaceae. The genus Casearia (formerly in Flacourtiaceae), which may lack salicoid leaf teeth and has apetalous flowers with the disc on the basal-adaxial surface of the caly x, is sister to the rest of Salicaceae, although this position is weakly support ed by rbcL (Chase et al., 2002) but strongly supported by data based on three ge nes (Soltis et al., 2000a). * * * * * * * * * * * 550 Plant Systematics

Cucurbitaceae A. L. de Jussieu Cucurbit or Gourd family 122 genera, 780 species Mainly distributed in tropics and subtropics, in temperate regions often found i n cultivation. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Violanae Violanae Violanae Eurosids I* Order Passiflorales Violales Cucurbitales Cucurbitales Violales Cucurbitales

' G

Figure 13.76 Cucurbitaceae. Luffa cylindrica. A: Branch with male flowers on ped uncle towards the base and solitary axillary female flowers towards the top; B: Vertical section of

female flower; C: Vertical section of male flower. Coccinia cordifolia. D: Branc h with male flower; E: Branch with female flowers. Stamen types. F: Lagenaria wi th 3 stamens, 2 with bithecous anthers and 1 with monothecous anther; G: Cucurbi ta with anthers united into a column; H: Cyclanthera with anthers united into 2 rings running around the top; I: Sicyos with filaments as well as anthers united . Major Families of Angiosperms 551 Salient features: Tendril climbing plants, leaves palmately veined, flowers unis exual, stamens 5, variously united, carpels usually 3, united, ovary inferior, f ruit a berry or pepo. Economic importance: The family is economically important for its food plants su ch as cucumber (Cucumis sativus), water melon (Citrulus vulgaris), loofah (Luffa acutangula, L. cylindrica), bottle gourd (Lageneria siceraria), melon (Cucumis melo) and red pumpkin (Cucurbita maxima). The dried fruit of Luffa yields bathro om sponge loofah. Species of Bryonia, Cucumis, Momordica are of medicinal import ance. Major Genera: Cayaponia (60 species), Momordica (45), Gurania (40), Sicyos (40), Cucumis (30) and Cucurbita (27). Description. Climbing annuals with coiled tendrils, sometimes trailing (Ecballiu m), rarely xerophytic shrubs (Acanthosicyos horrida) or even trees (Dendrosicyos ), vascuPhylogeny: The family was earlier considered closely related to Passifloraceae a nd included under the same order. Hutchinson (1973) placed them under separate o rders, Cucurbitales derived from Passiflorales through formation of unisexual fl owers, parietal placentation, inferior ovary, and modification of stamens. The s eparation is followed by Takhtajan, Dahlgren, and APG group. Cronquist and Thorn e preferred to retain these and other families under the same order Violales. Th orne (1999) placed Cucurbitaceae along with Begoniaceae and Datiscaceae under a separate suborder Begoniineae. Subsequently (2003, 2006, 2007) he has added Tetr amelaceae (earlier included under Datiscaceae). It is interesting to note that o rder Cucurbitales of APG II and APweb include the same four families along with a few more. Cucurbitaceae and Begoniaceae share the apomorphies of inferior ovar y, strongly intruded placentae and imperfect flowers. Monophyly of Cucurbitales is supported by serological data and rbcL sequences. Cucurbitaceae is easily rec ognized and monophyletic, but of the two sub-families commonly recognized only C ucurbitoideae is monophyletic, Nhandiroboideae (Zanonioideae) being paraphyletic . Renner et al., (2002) from the molecular lar bundles usually bicollateral, often in two rings. Leaves alternate, simple, palmately veined, lobed or compound, rarely absent (Acanthosicyos horrida), stip ules absent. In-florescence cymose (Bryonia) or flowers soli-tary axillary (Luff a female flower), rarely in short racemes (Luffa male flowers), plants monoeciou s or dioecious. Flowers bracteate or ebracteate, unisexual, rarely bisexual (Sch izopepon), actinomorphic, epigynous, with long hypanthium. Calyx with 5 sepals, more or less united, fused to ovary wall. Co-rolla with 5 petals, free (Luffa, L agenaria, Benincasa) or united (Cucurbita, Cucumis), imbricate, commonly yellow or white. Androecium with 5 stamens, anthers monothecous, filaments free (Luffa) or connate, sometimes 4 of these fused in two pairs thus two stamens bithecous and third monothecous giving appearance of 3 sta-mens (Coccinia), rarely all fiv e fused (Cucurbita), pollen grains with 3 to many fur-rows. Gynoecium with 3 uni ted carpels (syncarpous), unilocular with many ovules, placentation parietal, pl acentae enlarged intruding and often meeting in centre form-ing pseudo-axile pla centation, ovary inferior, styles simple or 3-partite. Fruit a berry, pepo or ca psule; seeds many, embryo straight, studies multiple chloroplast loci. p. 169, endosperm absent. Pollination mostly by in- concluded that Nhandiroboideae form an sects. Dispersal by animals, capsules of unresolved basal group. Echinocystis open explosively. * * * * * * * * * * * 552 Plant Systematics

Clusiaceae LindleySt. John s Wort family 45 genera, 1010 species Distributed widely, mainly in moist tropics, some in temperate regions of New as well Old World.Hypericum and Triadenum distributed in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II/ APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Rosidae Podostemanae Eurosids I* Series+/Superorder Thalamiflorae+ Th Th Order Guttiferales Theales Hypericales Theales Podostemales Malpighiales Salient features: Leaves opposite or times 4, rarely 3 or 6, free, imbricate, pe r-whorled, dotted with resin or secretary cavi- sistent below fruit or shedding from mature ties or canals, margin entire, stipules ab- fruit. Corolla with 5 pe tals, sometimes 4, sent, sepals and petals free, stamens many rarely 3 or 6, ora nge-yellow in Hypericum, in bundles, ovary 3-5 chambered, superior, free, imbricate, often persisting as withered styles free, stigma papillate, fruit a capsule or berry. remains, nectar glands alternating with pet-als or absent. Androecium with many stamens, filaments free or united in 3-5 bunMajor Genera: Hypericum (350 species), dles (rarely 6-8 bundles), opposite the p etGarcinia (210), Calophyllum (180), Clusia als, filaments often persisting in f ruit, an-(150), Kayea (7), Mammea (65), Vismia (55), thers bithecous, dehiscence longitudinal, Chrysochlamys (55), Kielmeyera (50), staminodes often present in female flower, Harungana (50) and Triadenum (10). pollen grains tricolporate. Gynoecium with 3-5 united carpels, rarely many, ovary suDescription. Herbs (Hypericum), shrubs or perior, placentation axile, rarely uni locular trees (Garcinia), rarely as woody lianas or with parietal placentation w

ith deeply in-epiphytes (some species of Clusia) behav- truded placentas, ovules 2-many in each ing like strangling figs, with clear, black or locule, anatropou s, styles 3-5, free or coloured resin or secretion in cavities or connate at bas e, stigmas lobed or capitate. canals; hairs simple, multicellular, some- Fruit u sually a capsule, sometimes berry or times stellate. Leaves opposite or whorled, drupaceous; seeds many, embryo straight, simple, often with pellucid or black d ots endosperm absent, aril often present. Polli(punctate) or canals, margin enti re, nation mostly by bees and wasps. Dispersal unicostate reticulate venation, s tipules ab- by animals for fleshy fruits those with sent but sometimes with pair ed glands at arillate seeds, capsular fruits disperse seeds nodes. Inflorescence of terminal cymes or by wind or water. with solitary flowers, sometimes thyrsoid. Flowers with usually two bracteoles below Economic importance: The family is kno wn calyx, bisexual (Hypericum) or unisexual for edible fruits mangosteen (Garcin ia (Clusia, Garcinia) and polygamous or mangostana) and mammey apple (Mammea dio ecious, actinomorphic, hypogynous, americana). Fats and oils are obtained from p entamerous. Calyx with 5 sepals, some- the seeds of species of Calophyllum, Major Families of Angiosperms 553 Figure 13.77 Clusiaceae. Hypericum calycinum. A: Branch with terminal flower. H. lobbii B: Portion of branch showing one flower; C: Androecium D: Flower with petals and stame ns removed; E: Transverse section of ovary. Clusia purpurea. F: Branch with infl orescence; G: Flower. Garcinia mangostana. H: Branch with fruit; I: flower; J: V ertical section of flower; K: Fruit with rind from top removed. Hypericum myrtif olium. L: Portion of branch with flowers; M: Flower top view, enlarged. (L-M, af ter Goodfrey and Wooten, Aq. Wetland Pl. SE US, vol. 2, 1981; H-K, after Bailey, Man. Cult. Pl., 1949) Pentadesma, etc. The species of Harungana, Calophyllum yield hard and durable wo od. Drugs and cosmetics are obtained from the leaves of Hypericum spp. and Harun gana madagascariensis, and flowers of Mesua ferrea. Gums and pigments are extrac ted from the stems of Garcinia (source of gamboge) and Clusia (source of healing gum s). Species of Vismia, Psorospermum and Harungana yield drugs and dyes from bark . Species of Clusia and Hypericum with showy flowers are often grown as ornament als. 554 Plant Systematics Phylogeny: The family Hypericaceae was treated as distinct from Guttiferae (Clus iaceae) by Bentham and Hooker (1862). Engler and Prantl (1887) combined the two families under Clusiaceae, a treatment followed by Heywood (1978) and Cronquist (1988). Hutchinson (1973) justified separation of Hypericaceae on the basis of c onstantly bisexual flowers and gland-dotted leaves, as against unisexual flowers , close veins and secretary canals. He argued that Hyperica-ceae is fairly well c ircumscribed, and there seems little to be gained, in these days of smaller fami ly concepts , by including them in Clusiaceae (Guttiferae), as in Engler and Pra ntl system . The two are combined in treatments of Judd et al. (2002, 2008; under Clusiaceae) and Thorne (1999, 2000 and 2007 as Clusiaceae; 2003, 2006, prefered priority name Hypericaceae). Thorne (2007) divides Clusiaceae into five subfamil ies Kielmeye-roideae, Calophylloideae, Clusioideae, Chrysopioideae and Hypericoideae. APG II and APweb treat the two as distinct familie s, latter recognising two subfamilies Kielmeyeroideae and Clusioideae (including Calophylloideae) under Clusiaceae. Broadly circumscribed family (under name Clu siaceae or Hypericaceae) is assumed to be monophyletic on the basis of anatomica l and chemical evidence. The affinities within the group are not clearly resolve d as suggested by the studies of Chase et al., (2002) and Gustafsson et al., (20 02). Thorne (2003, 2006), placed Bonnetiaceae, Hypericaceae, Elatinaceae and Pod ostemaceae under order Hypericales, prefering name Podostemales in 2007 revision . The monophyly of this clade was not supported by the studies of Savolainen et al., (2000). Bonnetiaceae + Clusiaceae + Hypericaceae seem to be a distinct grou p with several potential synapomorphies, of which some are lost or highly modifi

ed in Podostemaceae. * * * * * * * * * * * Euphorbiaceae A. L. de Jussieu (including Phyllanthaceae) Spurge family 321 genera, 7,770 species Distributed widely in tropical and subtropical regions, with few species in temp erate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II/ APweb Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Rosidae Dilleniidae Magnoliidae Rosidae Podostemanae Eurosids I* Series+/Superorder Unisexuales+ Euphorbianae Malvanae Order Euphorbiales Euphorbiales Euphorbiales Euphorbiales Malpighiales Salient features: Plants usually with milky latex, leaves alternate, flowers uni sexual, carpels 3, ovary superior, 3-chambered, ovule with a caruncle. Major genera: Euphorbia (2100 species), Croton (720), Phyllanthus (500), Acalyph a (350), Glochidion (300), Antidesma (140), Manihot (160) and Jatropha (140). Major Families of Angiosperms 555 Description: Herbs (some species of Euphorbia, Phyllanthus) shrubs (Acalypha) or trees (Hevea) with often milky or coloured cal regions. The leaves of Cnidoscolus chayamansa are used as vegetable. The fruits of Antidesma bunias are also edible. latex, sometimes succulent and cactus-like, Aleurites moluccana (candlenut tree) and A. usually poisonous. Leaves alternate rarely fordii (Tung tree) are source s of oils used in opposite (some species of Euphorbia; the manufacture of paints and varnishes. Excoecaria) or whorled (Mischodon), some- Oil similar to tung is also obtained f rom the times modified into spines, simple or pal- species of Vernicia. Castor o il obtained from mate compound, venation pinnate or pal- Ricinus communis is used as purgative. T he mate, reticulate, stipules present, some- common ornamentals include Euphorbia t imes modified into spines (Euphorbia milii) pulcherrima, E. milii, Acalypha hisp

ida, Jator glandular, rarely absent. Inflorescence of ropha panduraefolia and Codiaeum v ariegatum. various types, commonly a cup shaped The fruit of Phyllanthus emblica ( amla ) is cy athium (Euphorbia) having a cup-shaped very rich source of vitamin C. The greasy involucre with usually 5 nectaries along the tallow surrounding the seeds of Sap ium rim and enclosing numerous male flowers sebiferum (Chinese tallow tree) is u sed for (arranged in scorpioid cymes, without peri-anth and represented by a sin gle stamen) and single female flower in the centre; sometimes a raceme (Croton) or panicle (Rici-nus). Flowers unisexual (monoecious or dioecious), actinomorphi c, hypogynous. Peri-anth usually with 5 tepals (representing sepals, petals abse nt), rarely 6 in two whorls (Phyllanthus) or absent (Euphorbia), petals usually absent but present in Jatropha and Aleurites, free or connate. Androecium with 1 stamen (Euphorbia), 3 with fused filaments (Phyllanthus), 5 (Bridelia) or many (Trewia), sometimes polyadelphous (or with repeatedly branched filaments) as in Ricinus, anthers bithecous (sometimes monothecous in Rici-nus due to splitting o f filament), dehiscence longitudinal. Gynoecium with 3 united car-pels, carpels rarely 4-many, ovary superior, trilocular with 1-2 ovules in each chamber, place ntation axile, styles usually 3. Fruit a schizocarpic capsule, a regma (Ricinus) , rarely a berry or drupe (Bridelia); seed often with conspicuous fleshy outgrow th called caruncle, embryo curved or straight, en-dosperm abundant or absent. making soaps and candles. Economic importance: The family includes a number of valuable plants. Hevea bras iliensis (Para rubber tree) is the source of natural rubber. Rubber is also obta ined from Manihot glaziovii (ceara rubber). Thick roots of Manihot esculentus (c assava or tapi-oca) are important source of starch in tropiPhylogeny: The family was earlier broadly circumscribed (Bentham and Hooker) to include genera which have now been separated under Buxaceae. Earlier considered related to Euphorbiaceae the family Buxaceae has been far removed to Sapindales (Engler and Prantl), Hamamelidales (Hutchinson), Buxales (Takhtajan: under Caryo phyllidae >Buxanae), or Balanopales (Thorne: under Rosidae >Rosanae near Hamamelid ales), Proteales (Judd et al., : under core tricolpates), Buxales (APweb) or unp laced at the beginning of Eudicots (APG II). Cronquist is the only recent author to include Buxaceae next to Euphorbiaceae under Euphorbiales (Rosidae). The gen us Ricinus is sometimes included under a separate family Ricinaceae but is more appropriately included under Euphorbiaceae. Webster (1967, 1994), who studied th is family extensively recognized five subfamilies: Phyllanthoideae, Oldfieldioid eae, Acalyphoideae, Crotonoideae and Euphorbioideae. These five are also recogni zed by Thorne (1999, 2003; 2007 prefers Hyaenanchoideae over Oldfieldioideae). T he former two on the basis of evidence from rbcL sequences have been separated i nto a distinct family Phyllanthaceae by APG II and APweb, as they do not seem to form a clade with other members of Euphorbiaceae. Putranjiva, Lingelsheimia and Irypetes have been removed to Putranjivaceae, and Paradrypetes 556 Plant Systematics

A D Figure 13.78 Euphorbiaceae. Euphorbia milii. A: Branch with umbellate cyathia an d spines; B: Vertical section of cyathium to depict showy scarlet bracts, single female fl ower and numerous male flowers, and nectaries along the rim of cyathium. E. hirt a C: Portion of plant showing opposite leaves and cyathia in heads; D: Cyathium wi th female flower protruding out and only 4 nectaries, showy bracts absent; E: Ve rtical section of cyathium. Phyllanthus fraternus. F: Portion of plant with flow ers; G: Male flower with monadelphous stamens; H: Female flower; I: Vertical sec tion of female flower. Croton bonplandianum. J: Portion of plant with flowers an

d fruits; K: Male flower with many stamens; L: Female flower; M: Vertical sectio n of female flower. shifted to Rhizophoraceae. Rest of the Euphor-biaceae including last three subfa milies form a well-defined clade with single ovule in each chamber. Thorne (2003 , 2006, 2007) has also recognized Putranjivaceae as distinct family. Sutter and Endress (1995) advocate a broadly delimited Euphorbiaceae (inc. both Phyllanthaceae and Putranjivaceae) but Major Families of Angiosperms 557 Huber (1991) for a narrower circumscription, with the biovulate taxa being consi dered to be closer to Linales. Within Euphorbiaceae s. str., molecular analyses by Wurdack and Chase (2002), suggest that substantial changes may be needed in the groupings recognized. Thorne had earlier (2003, 200 6) in-cluded Euphorbiaceae under superorder Geranianae, but in his latest revisi on (2007) shifted it under Podostemanae * * * * * * * * * * * Oxalidaceae R. Brown Wood sorrel family 6 genera, 700 species Distributed mainly in tropical and subtropical regions, a few species in tempera te regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II /(APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Disciflorae+ Geranianae Rutanae Oxalidanae Eurosids I* Order Geraniales Oxalidales Linales Oxalidales Oxalidales B & H under family Geraniaceae Salient features: Herbs or shrubs, leaves usually compound, with a sour taste, leaflets in cold or at night, entire, often Geraniales

emarginate, pulvinate at base, entire, stipules usually with pinnate or palmate reticulate venation, absent, flowers pentamerous, heterostylus, leaflets often with prominent pulvinus, stippetals clawed, stamens united, outer sta- ules smal l or absent. Inflorescence cymose oles forming phyllodes), leaflets often folding mens shorter, styles 5, seeds with conspicu- umbel, rarely solitary. Flowers bis exual, ous endosperm, arillate. actinomorphic, usually heterostylus, sometimes cleistogamous and apetalous (Oxal is acetosella). Calyx with 5 sepals, free, green, persistent. Corolla with 5 pet als, free or connate at base, often clawed, usually convolute, absent in cleisto gamous flowers. Androecium with 10 stamens, usually in two whorls, usually conna te at base, outer filaments usually shorter than inner, anthers bithecous, dehis cence by longitudinal slits, pollen grains tricolpate or triporate, nectar gland s at the base of filaments or Major genera: Oxalis (600 species), Biophytum (70) and Eichleria (2). Description: Herbs with bulbous tubers or fleshy rhizome, or shrubs, rarely tree s, of-ten with soluble and crystalline oxalates. Leaves alternate or all basal, pinnately (Biophytum) or palmately compound or trifoliate (some species of Oxali s), rarely re-placed by phyllode (Oxalis bupleurifolia, peti558 Plant Systematics \, F B I

G A H Figure 13.79 Oxalidaceae. Oxalis martiana. A: Plant with trifoliate leaves and u mbellate inflorescence; B: Flower with calyx and corolla removed; C: Calyx; D: Petal. O. cornicul ata. E: Portion of plant rooting at nodes and umbellate inflorescence; F: Flower ; G: Flower with calyx and corolla removed; H: Transverse section of ovary; I: F ruit with persistent calyx. (A-D, after Sharma and Kachroo, 1983) alternating with petals. Gynoecium with 5 united carpels, rarely free (Biophytum ), ovary superior, placentation axile, 1 or more ovules in each loculus, styles 5, free, per-sistent, stigmas capitate or shortly divided. Fruit a loculicidal c apsule or berry, often angled; seeds usually with an aril, embryo straight, endo sperm copious, testa often elastic turning inside out and ejecting seed. Pollina tion by insects, heterostyly result-ing in outcrossing. Mostly self dispersed by explosive inversion of testa and aril that of O. crenata boiled and eaten in Peru. The leaves of O. acetosella are som e-times used as salad. The bulbous stem of O. pescaprae (Bermuda buttercup) are sometimes used as vegetable in France and North Africa. Averrhoa carambola (cara mbola or star fruit) is cultivated widely for its edible fruit. Economic importance: The family is of little importance. The tubers of Oxalis tu berosa (oca) are eaten in Andean South America, and Phylogeny: The family was earlier included under Geraniaceae (Bentham and Hooker ) but now separated into a distinct family and separable by 5 distinct styles, p

ossession of arillate seeds and absence of stipules. Phylogenetic studies based on rbcL sequences (Chase et al., 1993) indicate that Oxalidaceae are more closel y related to Cunoniaceae and Major Families of Angiosperms 559 Cephalotaceae (and related families) and in-cluded under Oxalidales (Judd et al. , APG II, APweb), distinct from Geraniales. The family is also related to Linace ae with it was placed by Dahlgren under Linales. Woody genera in-cluding Averrho a are sometimes placed in a distinct family but are better placed here. The genu s Hypseocharis with united style, in-cluded here by Hutchinson and Cronquist has been optionally shifted to Geraniaceae (APG II, APweb), and to a distinct famil y Hypseocharitaceae by Takhtajan. Thorne who had earlier (1999) included Hypseocha ris under Geraniaceae and placed both Oxalidaceae and Geraniaceae closer together under Dilleniidae >Geranianae >Geraniales has subsequently (2003, 2006, 2007) placed Oxalidaceae under Rosidae >Oxalidanae >Oxalidales whereas Hypseocharitaceae and Geraniaceae are placed under Rosidae >Geranianae >Geraniales , far removed from Oxalidaceae. * * * * * * * * * * * Zygophyllaceae R. Brown Creosote Bush family 20 genera, 250 species Widespread in tropics and subtropics, especially in arid mainly in arid regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Disciflorae+ Geranianae Rutanae Geranianae Eurosids I * Order Geraniales Sapindales

Zygophyllales Geraniales Zygophyllales /Zygophyllales

B & H as Zygophylleae; APG II unplaced in Eurosids I Salient features: Leaves usually opposite, pinnate compound, stipules paired per sist-ent, flowers with disc, stamen with gland or appendage at base, ovary 4-5 l ocular, syle 1. simple or 2-foliate, strongly resinous, leaflets entire, venation reticulate, pi nnate or palmate, stipules paired, commonly spiny. In-florescence cymose, someti mes reduced to single flower. Flowers bisexual, rarely unisexual (Neoleuderitzia ), actinomorphic, rarely zygomorphic, hypogynous, usually pentamerous. Calyx wit h 5 sepals, rarely 4, free or slightly connate at base. Corolla with 5 petals, r arely 4 or absent, free, often clawed, imbricate. Androecium with 10 stamens, ra rely 15, usually in whorls of 5, outer whorl opposite the petals, free, each fil ament with a gland or appendage at base, anthers bithecous, basifixed, dehiscenc e by longituMajor genera: Zygophyllum (80 species), Fagonia (40), Tribulus (20), Balanites ( 20), Guaiacum (6), and Larrea (5). Description: Herbs (Tribulus), shrubs or trees (Guaiacum) with often jointed nod es, xylem elements arranged in horizontally aligned tier, with steroidal or trit erpenoid saponins and alkaloids. Leaves opposite, 2-ranked, rarely alternate, us ually paripinnate, rarely 560 Plant Systematics Figure 13.80 Zygophyllaceae. Tribulus terrestris. A: Part of plant with flowers and fruits; B: Flower; C: Vertical section of flower; D: Flower with sepals and petals remov ed; E: Ovary covered with hairs, short style and stigma; F: Transverse section o f ovary; G: Fruit; H: One of the cocci enlarged showing sharp spines. (A-B, D-E, after Sharma and Kachroo, Fl. Jammu, 1983) dinal slits, pollen grains usually tricolporate. Gynoecium with 5 united carpels , rarely 2- 6, ovary superior, usually furrowed or winged, placentation axile, l ocules as many, ovules one to many in each locule, pendulous, anatropous or orth otropous, style 1, short, stigma capitate or lobed, nectar disc present at the base of ovary. Fruit a usua lly a cap-sule, septicidal or loculicidal, rarely schizocarpic, berry or drupe, somtimes winged; seeds usually with aril, embryo curved or straight, endosperm u sually absent Major Families of Angiosperms 561 or scanty. Pollination by insects. Arillate seeds (Guaiacum) are dispersed by wi nd, schizocarpic winged fruits by wind and spiny fruits (Tribulus) by exozoochor y. Phylogeny: The family is usually placed under order Geraniales, although more re cently shifted to order Zygophyllales (Takhtajan 1997, Thorne 2007, Stevens 2008 ). The family is considered to be monophyletic after the exclusion of a few gene ra to Peganaceae and Nitrariaceae (Thorne to order Rutales) or Nitrariaceae (Sap indales of Eurosids II by APG II, APWeb of Stevens). Monophyly of Zygophyllaceae supported by morphology and DNA characters. The family is sister to Krameriacea e as supported by rbcL sequences (Soltis et al., 1998; Savolainen et al., 2000). The family is divided into 5 sub-families: Morkillioideae, Tribuloideae, Seetze nioideae, Larreoideae and Zygophylloideae. Balanites is very different from othe r Zygophyllaceae in floral, vegetative and seed anatomy, although tentatively in cluded under Tribuloideae. Hilu et al. (2003) reported Larrea to be weakly assoc iEconomic importance: The family is of mi-nor economic importance. Wood of Guaiac um officinale (lignum vitae), being the strong-est and heaviest wood, is highly prized tim-ber in tropical Central America and West Indies.The tree also yields medicinal resin guaiacum, once used to treat syphilis. Spe-cies of Bulsenia (B. arborea: Maracaibo lig-num vitae; B. sarmienti: Paraguay lignum vitae) yield val uable timber and perfume oil. Seeds of Peganum harmala are the source of dye tur key red. Tribulus terrestris is a trou-ble some weed whose spines on fruit are s imilar to sharp iron caltrops once used in battlefields to stab the feet of men and horses. They also often puncture cycle tyres; hence the names caltrops, punc

ture vine and goat head for the weed. Species of ated with Fabaceae in their rbcL analysis; they note that the possession of anth roZygophyllum are used as spices: buds of quinones is a possible synapomorphy Z. f abago used in sauces and fruits of between Zygophyllaceae and the N-fixing Z. coccinium as substitute for black pepper. clade Fabaceae. * * * * * * * * * * * Geraniaceae A. L. de Jussieu Geranium family 5 genera, 760 species Widespread mainly in temperate and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Disciflorae+ Geranianae Rutanae Geranianae Rosids * Order Geraniales Geraniales Geraniales Geraniales Geraniales Geraniales 562 Plant Systematics

P "

J / M N B

t L B f Ci 7n., V'A~'~' . D L J L t U F A K Figure 13.81 Geraniaceae. Geranium rotundifolium. A: Plant with palmately lobed leaves and umbellate inflorescence; B: Flower; C: Petal; D: Flower with sepals and petals rem oved to show androecium and gynoecium; E: Gynoecium with 5 carpels; F: One segme nt of capsule; G: Seed. Erodium cicutarium. H: Plant with pinnate leaves and umb ellate inflorescence; I: Sepal; J: Petal; K: Stamen; L: Staminode; M: Gynoecium; N: One mericarp with long coiled beak. Monsonia senegalensis. O: Flower having 15 stamens; P: Portion of flower to show stamens with filaments united in groups of 3, the androecium being pentadelphous. Salient features: Usually herbs, stems swol-len at nodes, leaves usually deeply lobed, stipules conspicuous, flowers pentamerous, petals clawed, stamens united, styles 1, fruit with elastic dehiscent schizocarps that curl on the beak, aril absent. pound, venation palmate, reticulate, stipules conspicuous. Inflorescence cymose umbel, rarely solitary. Flowers bisexual, actinomorphic, rarely zygomorphic (Pel argonium), hypogynous, pentamerous. Calyx with 5 sepals, free, green, persistent , sometimes spurred (Pelargonium). Corolla with 5 petals, rarely 4 or absent, fr ee, often clawed, imbricate, nectar glands alternating with petals or absent. An droecium with 10 (Geranium) or 15 (Monsonia) stamens, rarely 5 (other 5 sterileErodium), usually connate at base, someMajor genera: Geranium (300 species), Pel-argonium (250), Erodium (80) and Monso nia (25).

Description: Usually herbs, rarely undershrubs, sometimes aromatic (Pelargo-nium ), stems swollen at nodes, usually with stalked glandular hairs. Leaves alternat e or opposite, simple or palmately lobed, or comtimes pentadelphous (Monsonia), rarely free, anthers bithecous, dehiscence by lo ngitudinal slits, pollen grains tricolpate or triporate. Gynoecium with 5 united carpels, ovary Major Families of Angiosperms 563 superior, usually lobed, placentation axile, ovules usually 2 in each loculus, a natropous or campylotropous, style 1, slender and beak-like. Fruit a capsular de hiscent schizocarp with 5 1-seeded segments that separate elas-tically from cent ral column, and often open-ing to release seeds (Geranium), or indehiscent schiz ocarp (Biebersteinia); seeds usually without aril, pendulous, embryo curved, end osperm usually absent or scanty. Pollination by insects. Mostly self dispersed b y explosive opening of schizocarps throwing seeds several metres away. Phylogeny: The family is consistently placed under order Geraniales, sometimes a long with Oxalidaceae. The recent DNA based studies (Chase et al., 1993), howeve r, suggest that it is related to Crossosomataceae, Staphyleaceae, in a narrowed circumscribed order. Geraniaceae are well-defined monophyletic group based on rb cL sequences and loss of intron in the plastid gene rpl16 (Price and Palmer, 199 3). Hypseocharis, with capsular fruits and formerly placed under Oxalidaceae is sister to rest of the family. Takhtajan places it under a distinct family Hypseo charitaceae. APG II optionally include Hypseocharitaceae under Geraniaceae. APwe b treats Hypseocharis as distinct group within Geraniaceae. Thorne who had earli er (1999) included both Geraniaceae and Oxalidaceae adjacent to each other under Geraniales, has subsequently (2003, 2006, 2007) shifted Oxalidaceae under disti nct superorder Oxalidanae, order Oxalidales. He has also removed Hypseocharis to a distinct family Hypseocharitaceae, placed next to Geraniaceae. Economic importance: The family is known for Pelargonium (often marketed as Gera -nium), grown as ornamental in pots and also for geranium oil extracted from the leaves and shoots of mainly P. odoratissimum. Spe-cies of Geranium (crane-bill) and Erodium (storckbill) are also grown as ornamentals. The persistent dry styl e of Erodium, which is hygroscopic, is often used to indicate changes in humidit y. * * * * * * * * * * * Rosaceae A. L. de Jussieu Rose family 110 genera, 3,100 species Widespread but best represented in the Northern Hemisphere, mainly in the temper ate and arctic climate. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Polypetalae Rosidae Subclass Rosidae

Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Rosanae Rosanae Rosanae Eurosids I* Order Rosales Rosales Rosales Rosales Rosales Rosales Salient features: Herbs shrubs or trees, leaves usually serrate, stipules conspi cuous, flowers actinomorphic, usually perigynous and with hypanthium, sepals and petals 5 each, petals usually clawed, well-developed nectary on hypanthium or base of stamens, stamens 564 Plant Systematics numerous, carpel single or numerous and free, rarely united, fruit usually flesh y. aggregate (etaerio of achenes in Potentilla, etaerio of drupes in Rubus); seed w ith straight embryo, without endosperm. Pollination mainly by insects. Dispersal by birds, animals or wind. Major genera: Rubus (750 species), Potentilla (500), Prunus (430), Crataegus (24 0), Cotoneaster (230), Sorbus (230), Rosa (225), Alchemilla (220), Spiraea (100) , Pyrus (60), Malus (55) Geum (40) and Fragaria (15). Description: Herbs (Alchemilla, Fragaria), shrubs (Rosa, Rubus) or trees (Prunus , Malus, Pyrus), rarely climbing (some species of Rosa), sometimes with runners (Fragaria), often with prickles and thorns, without latex, nodes trilacunar, rar ely unilacunar. Leaves alter-nate, rarely opposite (Rhodotypos), simple (Malus, Prunus), palmately compound (Fragaria) or pinnate compound (Sorbaria), leaf blad e often with gland-tipped teeth, usually serrate, venation pinnate or palmate, r eticu-late, stipules present, often adnate to peti-ole. Inflorescence with solit ary flowers (some species of Rosa), racemes (Padus), panicles or cymose umbels ( Spiraea), sometimes corymbs (Crataegus), rarely catkin-like (Poterium). Flowers bisexual, rarely unisexual (Poterium; plants monoecious or dioecious), actinomor phic, rarely zygomorphic (Parinarium), usually perigynous with distinct hypanthi um (flat, cup-shaped or cylindrical); hypanthium free from or adnate to carpels, often enlarging in fruit, with nectar ring on inside, rarely epigynous (Malus). Calyx usu-ally with 5 sepals, united at base, sometimes with 3-5 epicalyx (Frag aria) on outside, often persistent. Corolla usually with 5 petals, free, often c lawed, imbricate. Androecium with numerous stamens, free , 4 in Sanguisorba, 2 i n Parastemon urophylla, anthers bithecous, rarely monothecous (Alchemilla), dehi scence longitudinal, pollen grains tricolporate. Gyn-oecium with 1 (Prunus), 2-3 (Crataegus) to many carpels (Rosa), usually free, rarely connate (Crataegus, Py rus), sometimes adnate to hypanthium, ovary superior or in-ferior, usually unilo cular, ovules 1,2 or more, unitegmic or bitegmic, crassinucellate, placentation basal, lateral or apical, rarely axile (Pyrus). Fruit a follicle (Spiraea), ache ne (Rosa), drupe (Prunus), pome (Malus), or Economic importance: The family is largely known for its temperate fruits: apple (Malus domestica), pear (Pyrus), plums (Prunusseveral species), cherries (Prunu s avium, P. cerasus) peaches (Prunus persica), almonds (Prunus dulcis), apricots

(Prunus armeniaca), strawberry (Fragaria vesca), loquots (Eriobotrya), raspberr ies (Rubus), quince (Cydonia), etc. Popular ornamentals include species of Rosa, (rose) Rubus (raspberry), Chaenomeles (flowering quince), Potentilla (cinquefoi l), Geum (avens), Cotoneaster, Crataegus (hawthorn), Pyracantha (firethorn), and Sorbus (mountain ash). Flowers of Rosa damascena are used for extracting attar of roses. The bark of Quillaja (soap-bark tree) contains saponin used as substit ute for soap in cleaning textiles, and also yield tannin. Bark of Moquilla utili s (pottery tree) of Amazon is used in making heat-resistant pots. The wood of Pr unus serotina is used for making furniture and cabinets. Several species are als o valuable sources of timber. Phylogeny: In spite of great morphological diversity the family Rosaceae is a we ll recognized group whose monophyly has been supported by rbcL sequences (Morgan et al., 1994). More than 27 family names have been proposed for groups of diffe rent genera taken out from Rosaceae, but according to Hutchinson (1973) if one o r two tribes of the family are taken out, at least 18 or 19 should follow suit, and the Rosaceae would be reduced to the genus Rosa only. He like most recent au thors follows a broader circumscription of the family, but does not recognize se paration of Chrysobalanaceae and Neuradaceae (established as distinct in 12th ed ition of the Engler s Syllabus published in 1964). These two last families have be en recognized as distinct in all major classifications. Cronquist places them to gether with Rosaceae under Rosales. Dahlgren places Major Families of Angiosperms 565

Figure 13.82 Rosaceae. Prunus domestica. A: Portion of a flowering twig; B: Flow er from above; C: Vertical section of flower, petals removed. Rubus ellipticus. D: Branch with terminal inflorescence; E: Vertical section of flower with petals removed; F: Pe tal; G: Fruit covered with persistent calyx. Duchesnia indica. H: Portion of a b ranch with trifoliate leaves and flower; I: Vertical section of flower with peta ls removed, 3 lobed bracteoles (epicalyx) present outside calyx; J: Calyx and 5 3-lobed bracteoles. Rosa pimpinellifolia. K: Branch with fruits; L: Flower and b ud; M: Vertical section of flower showing cup shaped hypanthium and numerous fre e carpels; N: Fruit (hip) enclosing achenes and with persistent calyx. Chrysobalanaceae under Theanae >Theales, but Neuradaceae along with Rosaceae in R osales. Takhtajan places Neuradaceae in Rosales along with Rosaceae, but Chrysob alanaceae in distinct order Chrysobalanales. Thorne (1999) shifted both families from Rosidae to Dilleniidae, Chrysobalanaceae under Dillenianae >Dilleniales and Neuradaceae under Malvanae >Malvales. In later revisions (2003, 2006, 2007), howe ver, he has abolished Dilleniidae. in his latest revision (2007) he placed Rosac eae under Rosidae >Rosanae >Rosales, Neuradaceae under Malvidae >Malvanae >Malvales >Cis tineaeand Chrysobalanaceae under Rosidae >Podostemanae >Euphorbiales. APG II and APweb have shifted Chrysobalanaceae to Eurosids I >Malpighiales and Neuradaceae to Eurosids I I >Malvales, retaining Rosaceae in Eurosids I >Rosales. The family has often been co nsidered closely related to Saxifragaceae and Crassulaceae but the rbcL data ide ntify Ulmaceae, Celtidaceae, Moraceae, Urticaceae and Rhamnaceae as sister group s (Savolainen et al., 2000a). Usually 4 subfamilies are recognized within Rosace ae: Maloideae (fruit a pome), Amygdaloideae (syn: Prunoideae; fruit a drupe, car pel 1, nectaries on petiole and 566 Plant Systematics lamina), Rosoideae (fruit achenes or drupelets) and Spiraeoideae (follicle or ca psule). Although Rosoideae and Maloideae are reasonable clades, little can yet b e said of larger patterns of relationship in the rest of the family (Potter et a l., 2002). Porteranthus is sister to Maloideae; Gillenia is sister to that whole clade (Potter et al., 2002; Evans

et al., 2002a, b). The position of Dryadeae (inc. Cercocarpus, Dryas, Cowania an d Chamaebatia) included in Rosoideae is uncertain, they lack phragmidiaceous rus ts; their roots are associated with N-fixing Frankia and their fruits are achene s with hairy styles. They are rather basal (Potter et al., 2002; Evans et al., 2002). * * * * * * * * * * * Fabaceae Lindley Bean or Pea family (=Leguminosae A. L. de Jussieu) 630 genera, 18,000 species (Third largest family after Asteraceae and Orchidaceae) Cosmopolitan in distribution, primarily in warm temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Fabanae Rutanae Rosanae Eurosids I* Order Rosales Fabales Fabales Fabales Fabales Fabales B & H as Leguminosae Takhtajan, Thorne, APG II and APweb as Fabaceae. Cronquist and Dahlgren recognize three independent families Fabaceae, Caesalpini aceae and Mimosaceae, thus restricting the name Fabaceae to include only papilionaceous (m embers of Papilionoideae) members, for which the alternate name is Papilionaceae and not Leguminosae. This large family has traditionally been di-vided into three subfamilies Papilio noideae (Faboideae), Caesalpinioideae and Mimosoi-deae. These have been recogniz

ed as inde-pendent families Fabaceae (Papilionaceae), Caesalpiniaceae and Mimosa ceae in several recent systems of classification, a trend that tends to be rever sing in last decade or so. It must be noted that name Fabaceae is valid for family sensu lato as well as for Papilionoideae upgraded as family. Legumino sae is the alternate name only for former whereas Papilionaceae is the alternate name for latter. Common features of the family include leaves usually compound with pulvinate base, odd sepal anterior, flowers perigynous, carpel 1 with margi nal placentation and fruit commonly a pod or lomentum. Major Families of Angiosperms 567 Subfamily Faboideae DC. B & H as Papilionoideae Takhtajan, Thorne, APG II and APweb as Faboideae Cronqui st and Dahlgren as family Fabaceae (Papilionaceae). (=Papilionoideae L. ex A. DC.) 440, genera 12,800 species Cosmopolitan in distribution, primarily in warm temperate regions.

K J N Q Figure 13.83 Fabaceae, subfamily Faboideae. Medicago polymorpha. A: Portion of p lant with trifoliate leaves, laciniate stipules and few flowered axillary clusters on long pedu ncles; B: Flower; C: Standard; D: Wing; E: Keel; F: Androecium with diadelphous (1 free, 9 with united filaments) stamens; G: Fruit covered with tubercles; H: S eed. Dalbergia sissoo. I: Flowering shoot with a fruiting twig; J: Flower; K: An droecium with 9 monadelphous stamens. Sophora mollis. L: Branch with flowers; M: Moniliform pod. Lathyrus odoratus. N: Portion of a flowering branch, upper leaf lets modified into tendrils; O: Vertical section of flower; P: Diadelphous andoe cium; Q: Pod. 568 Plant Systematics Salient features: Trees, shrubs or herbs, leaves usually pinnate compound with p ulvinate base, flowers zygomorphic with papilionaceous corolla, sepals united, o dd sepal anterior, stamens 10, usually diadelphous (1+(9)), carpel 1, ovary supe rior, fruit a pod. stamens, diadelphous (1 posterior free and filaments of nine fused into a tube w hich is open posteriorly), sometimes 5+5 as in Smithia, rarely monadelphous (Ono nis), or free (Sophora, Thermopsis) anthers bithecous, dehiscence longitudinal. Gynoecium with a single carpel, unilocular with many ovules, placentation margin al, ovary superior, style single, curved. Fruit a legume or pod, rarely a loment um (Desmodium), sometimes indehiscent (Melilotus), rarely spirally coiled (Medic ago); seeds 1-many, seed coat hard, endosperm minute or absent, food reserves in cotyledons. Pollination primarily by insects, mostly bees. Dispersal is commonl y by wind, but often exozoochorus (Medicago), or by mammals (Tamarindus). Major genera: Astragalus (2000 species), Indigofera (700), Crotalaria (600), Des modium (400), Tephrosia (400), Trifolium (300), Dalbergia (200), Lathyrus (150), Lotus (100), and Milletia (100).

Description: Trees (Dalbergia, Erythrina), shrubs (Tephrosia, Alhagi, Indigofera ) or herbs (Medicago, Melilotus), sometimes woody climbers (Wisteria), commonly with root nodules. Leaves alternate, pinnately (Pisum, Vicia) or palmately compo und (Trifolium), sometimes simple (Alysicarpus, Alhagi), whole leaf (Lathyrus ap haca) or upper leaf-lets (Vicia, Pisum) sometimes modified into tendrils, leaf b ase (sometimes also the base of leaflets) pulvinate, stipules present. Inflo-res cence racemose, in racemes, heads (Tri-folium) or spikes (Ononis), sometimes in clusters (Lotus, Caragana). Flowers bracteate (bracts often caducous), bisexual, zygomorphic, perigynous. Calyx with 5 se-pals, more or less united, usually cam panulate, odd sepal anterior. Corolla with 5 petals, free, papilionaceous consis t-ing of a posterior standard or vexillum, two lateral wings or alae and two ant erior petals fused along margin to form keel or carina which encloses stamens an d pistil, poste-rior petal outermost. Androecium with 10 Economic importance: The subfamily is of major economic importance, ranking second to Poaceae. It is the source of several pulse crops such as kidney bean (Pha seolus vulgaris), green gram (P. aureus), black gram (P. mungo), lentil (Lens es culenta), chick pea (Cicer arietinum), pea (Pisum sativum) and pigeon pea (Cajan us cajan). Soybean (Glycine max) and peanut (Arachis hypogaea) yield oil and hig h-protein food. Indigo dye is obtained from Indigofera tinctoria. The seeds of A brus precatorius are used in necklaces and rosaries, but are extremely poisonous and can be fatal if ingested. The important fodder plants include alfalfa (Medi cago sativa) and clover (Trifolium). Common ornamentals in-clude lupin (Lupinus) , sweet pea (Lathyrus odoratus), Wisteria (Wisteria), Laburnum, coral tree (Eryt hrina), false acacia (Robinia) and broom (Cytisus). Subfamily Caesalpinioideae DC. B & H, Takhtajan, Thorne, APG II and APweb as Caesalpinioideae Cronquist and Dah lgren as family Caesalpiniaceae. 150 genera, 2,700 species Distributed mainly in tropics and subtropics, a few species in the temperate regions. Major Families of Angiosperms 569 Salient features: Trees, shrubs or herbs, leaves usually pinnate compound with p ulvinate base, flowers zygomorphic corolla not papilionaceous, posterior petal i nner-most, sepals free, odd sepal anterior, stamens 10, usually free, in two who rls , ovary supe-rior, carpel 1, fruit a pod. woody climbers (Pterolobium, Bauhinia). Leaves alternate, pinnately or palmately compound, sometimes simple (Bauhinia), leaf base (sometimes also the base of le aflets) pulvinate, stipules present. Inflorescence racemose, in racemes or spike s (Dimorphandra). Flowers bracteate (bracts usually caducous) bisexual, zygomorp hic, perigynous. Calyx with 5 sepals, rarely 4 (Amherstia), free or rarely conna te (Bauhinia), odd sepal ante-rior. Corolla with 5 petals, rarely 3 (Amherstia), 1 (Pahuda) or even absent (Tamarindus), free, not papilionaceous, posterior pet al innermost. Androecium with Major genera: Chamaecrisia (260 species), Bauhinia (250), Senna (250), Caesalpin ia (120) and Cassia (30). Description: Trees (Delonix), shrubs (Cassia occidentalis) or herbs (Cassia obtu sa), rarely

c D Figure 13.84 Fabaceae, subfamily Caesalpinioideae. Cassia occidentalis. A: Porti on of plant with flowers and paripinnate leaves; B: Flower with sepals and petals removed, showin g gynoecium and stamens of three different sizes; C: A pair of pods. Caesalpinia decapetala. D: Portion of plant with bipinnate leaves and racemose inflorescenc e; E: Flower; F: One of the four large petals; G: Gynoecium; H: Pod; I: Seed. 570 Plant Systematics 10 stamens, sometimes lesser (3 in Tamarindus), rarely more, free, sometimes une qual in size (Cassia), anthers bithecous, dehiscence longitudinal or by apical p ores. Gynoecium with a single carpel, unilocular with many ovules, placentation marginal, ovary superior, style single, curved. Fruit a legume or pod, rarely a lomentum; seeds 1-many, seed coat hard, endosperm minute or absent, food reserve s in cotyledons. Economic importance: The Subfamily includes several ornamentals such as pride of Barbados (Caesalpinia pulcherrima), paulo verde (Parkinsonia), red bud (cercis canadensis), Gul-mohar (Delonix regia), and several species of Cassia and Senna. Many species of Senna are cultivated for leaves that yield drug senna. The hear twood of Haematoxylon campechianum (logwood) yields the dye hematoxylin. Subfamily Mimosoideae DC. B & H, Takhtajan, Thorne, APG II and APweb as Mimosoideae Cronquist and Dahlgren as family Mimosaceae. 40 genera, 2,500 species Distributed mainly in tropical and subtropical regions. Salient features: Trees, shrubs or herbs, leaves usually pinnate compound with p ulvinate base, flowers actinomorphic, co-rolla not papilionaceous, petals valvat e, se-pals united, odd sepal anterior, stamens 4- many, free or connate, filamen ts often long exserted and showy, ovary superior, carpel 1, fruit a pod or lomen tum. bracteate (bracts usually caducous), sessile, or short-pedicelled, bisexual, act inomorphic, perigynous. Calyx with 5 sepals (4 in Mimosa), connate, odd sepal an terior, usually valvate, teeth small. Corolla with 5 petals (4 in Mimosa), free or united (Acacia, Albizia), valvate. Androecium with 4-many (4 in Mimosa, 10 in Prosopis, numerous in Acacia and Albizia) stamens, free (Acacia, Prosopis) or f ilaments connate (Albizia), anthers bithecous, dehiscence longitudinal, filament s long and anthers usually exserted. Gynoecium with a single carpel, unilocular with many ovules, placentation marginal, ovary superior, style single, curved. F ruit a legume or lomentum (Mimosa, Acacia); seeds 1-many, seed coat hard, endosp erm minute or absent. Major genera: Acacia (1300 species), Mimosa (500), Inga (250), Pithecellobium (1 70), Calliandra (150) and Albizia (150). Description: Trees (Acacia, Albizia), shrubs (Calliandra) or herbs (Mimosa pudic a), rarely climbers (Entada), or aquatic plants (Neptunia). Leaves alternate, pi nnately or palmately compound, sometimes simple, leaf base (sometimes also the b ase of leaflets) pulvinate, petiole sometimes modified into phyllode (Acacia aur iculiformis), stipules present, sometimes spiny and hollow inside sheltering ant s (Acacia sphaerocephala), leaves of Mimosa pudica sensitive to touch and showin g sleeping movements. Inflores-cence racemose, in racemes (Adenanthera) or spike s (Prosopis), sometimes in cymose heads (Mimosa, Acacia). Flowers small, Economic importance: The subfamily is of lesser economic importance. Sensitive p lant touch-me-not (Mimosa pudica) is grown as a curiosity. Various species of Ac acia (A. senegal, A. stenocarpa) yield gum arabic. The pods and seeds of mesquit e (Prosopis juliflora) are used as animal feed, wood in cooking meats. Wood of X ylia is hard and used in ship building. Calliandra, Dichrostachys are grown Major Families of Angiosperms 571

E vn !/ H

K Figure 13.85 Fabaceae, subfamily Mimosoideae. Mimosa pudica. A: Branch with infl orescence heads; B: Lomentum fruits constricted between and splitting into 1-seeded segmen ts. Acacia nilotica. C: Branch with long spines and inflorescence heads; D: Flow er bud; E: Moniliform pod. A. farnesiana. F: Portion of a branch with spines, le af and inflorescence heads; G: Flower with numerous stamens; H: Pod. Albizia jul ibrissin. I: Part of a bipinnate leaf; J: Flower with monadelphous stamens; K: P art of a stamen showing anther. as ornamentals, Pithecellobium as a useful hedge plant. Dahlgren (1989). Takhtajan who also began with the same treatment, has in his la st two versions (1987, 1997) included all the three under broadly circumscribed Fabaceae, giving these three the rank of subfamily. Thorne has consistently incl uded all the three sub-families under broadly circumscribed Fabaceae, a position also justified by APG II. Thorne had earlier (1999) included Fabaceae along wit h 21 other families in a broadly circumscribed order Rutanae >Rutales under subord er Fabineae (containing Fabaceae, Surianaceae and Connaraceae). In his latest re vision (2003) he has placed Fabaceae, Surianaceae, Polygalaceae (earlier placed in Phplogenp of Fabaceae: The family is com-monly circumscribed to include all the three subfamilies. Hutchinson as early as 1926 had recognized these as independe nt families Fabaceae, Caesalpiniaceae and Mimosaceae, a position that he maintai ned even in his last revision in 1973, regarding Caesalpiniaceae as the most pri mitive of the three, Mimosaceae relatively advanced and Fabaceae to be the clima x group. The trend was followed and maintained in their latest classifications b y Cronquist (1988) and 572 Plant Systematics Dilleniidae >Geranianae >Polygalales) and Quillajaceae (earlier with uncertain posit ion) in separate order Fabales, a treat-ment similar to APG II and APweb. Affini ties with Rutales have been supported on the basis of wood anatomy and embryolog y (Thorne, 1992). Thorne (2006, 2007) shifted times bilobed; vestured pits, which they lack, are also absent in Cassieae. The flowers of Cercis are only superficially similar to those of Faboideae (Tucker 2002). the family under Rosanae >Fabales. Mimosoideae are largely monophyletic, Faboideae are monophyletic, Caesalpinioide ae are paraphyletic and basal. Wojciechowski et al., (2003) on basis of stud-ies on sequences of the plastid matK gene note than non-protein amino acids seem to have originated once in this clade. Fabaceae s. l. are often referred to their own order, as in both Cronquist (1981) and Takhtajan (1997), former placing it c loser to Rosales and latter closer to Sapindales. They can be confused with Conn araceae (Oxalidales), although the latter lack stipules, their flowers are radia lly symmetrical and have stamens of two distinctly different lengths, and their gynoecium is frequently multi-carpellate. However, in both the RP122 chloroplast gene has moved to the nucleus! Also, the ovaries of both have adaxial furrows ( cf. the ventral slit: Matthews & Endress, 2002). Fabaceae have also been linked with Sapindaceae, in the Eurosid II group in APG II and APweb. Recognition of broadly circumscribed Fabaceae, is supported by its monophyly as evidenced by common morphological fea-tures, and the results of rbcL sequence da

ta (Chappill 1994; Doyle 1994). Studies also in-dicated that Caesalpinioideae ar e paraphyletic with some genera more closely related to Mimosoideae, and others to Faboideae than they are to one another. It is now established that Swartzia a nd Sophora (and relatives) represent basal clades of Faboideae lack a 50kb inver sion in the trnL intron that is found in other members of the subfamily. Studies of Doyle et al., (2001) and Bruneau et al., (2001) suggest that Cercis and Bauh inia are basal in Fabaceae and as such discussed under distinct group Cercideae in APweb, characterized by simple leaves, some* * * * * * * * * * * Myrtaceae A. L. de Jussieu Myrtle family 137 genera, 2,050 species Mainly distributed in tropics and subtropics, abundant in Australia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Myrtanae Myrtanae Myrtanae Rosids * Order Myrtales Myrtales Myrtales Myrtales Myrtales Myrtales Major Families of Angiosperms 573

A H 1

J B D L Figure 13.86 Myrtaceae. Eucalyptus tereticornis. A: Portion of branch with axill ary umbellate inflorescences; B: Vertical section of flower bud; C: Open flower with cap shed; D: Transverse section of ovary with 4 axile placentas. Callistemon viminalis. E : Branch with spike proliferating into vegetative shoot; F: Flower with long exs erted numerous stamens; G: Vertical section of flower; H: Stamen with dorsal fix ation; I: Transverse section of ovary with 3 axile placentas. Syzygium cuminii. J: Branch with inflorescences borne on peduncles; K: Vertical section of flower; L: Fruits. Salient features: Shrubs or trees, bark flaky, leaves gland-dotted, entire, infr amar-ginal venation, stamens numerous, ovary in-ferior often united with hypanth ium. erating into vegetative shoots, giving appear-ance of a bottle-brush). Flowers b racteate (Eugenia) or ebracteate (Eucalyptus), bisexual, actinomorphic, epigynou s (sometimes perigynous). Calyx with 4-5 sepals, more or less connate into a tub e, imbricate, sometimes united into a cap (calyptra or operculum) which drops of f as flower opens. Corolla with 4 (Eugenia) to 5 (Psidium) petals, (rarely absen t), usually fugacious, free, rarely united with calyx to form cap like operculum (Eucalyptus) that falls off as the flower opens. Androecium with many stamens, filaments free or slightly connate at base (Callistemon), attached higher up on hypanthium, anthers bithecous, dehiscence longitudinal or by apical pores, polle n grains usually tricolpate with fused furrows Gynoecium with 2-5 united carpels (syncarpous), multilocular (locules as many as carpels) Major genera: Eugenia (600 species). Eucalyptus (500), Myrcia (300), Syzygium (300), Psidium (100), Melaleuca (100) and C allistemon (25). Description: Evergreen Shrubs (Myrtus) or large trees (Eucalyptus) often with fl aky bark, terpenes present. Leaves alternate (Barringtonia, Callistemon), opposi te (Eugenia) or whorled, simple, entire, gland-dotted, usu-ally coriaceous, vena tion often inframar-ginal, stipules absent. Inflorescence cymose (umbellate cyme in Eucalyptus) or racemose (Barringtonia), flowers sometimes solitary (Psidium) , or in spikes (Callistemon prolif574 Plant Systematics with 2-many ovules, placentation axile, rarely parietal (Rhodamnia) with intrude d placentae, ovary inferior, or semi-inferior (Melaleuca) style long with capita

te stigma. Fruit a fleshy berry (Eugenia), drupe (Barringtonia) or capsule (Euca lyptus), rarely one-seeded nut (Calycothrix); seeds 1-many, embryo curved or twi sted, endosperm absent. thors). Monophyly of the family, together with the morphological data, is eviden ced by molecular analysis through rbcL (Conti, 1994), matK (Wilson et. al., 1996 ), and ndhF (Sytsma. et al., 1998) sequences. Heteropyxis and Psiloxylon are bas al taxa with perigynous flowers and stamens in two whorls. The family is conside red to be closely related to Rosaceae, and probably the order Myrtales is derive d from Rosales. The family is traditionally divided into two subfamilies: Leptos permoideae (leaves spiral to opposite; fruit dry, dehiscent) and Myrtoideae (pol yhydroxyalkaloids common; leaves opposite; terpenoid-containing glands in the ap ex of the connective, stigma dry; fruit fleshy, indehiscent). The latter are lar gely derived. Leptospermoideae are basal and paraphyletic (Wilson et al., 2001; Salywon et al., 2002) as evidenced by molecular and morphological data. Genus Sy zygium, sometimes included under Eugenia, represents an independent acquisition of the fleshy fruit from that in Eugenia and the bulk of Myrtoideae. Thorne (200 3, 2006, 2007), places the family along with other two in suborder Myrtineae, Ly thraceae and Onagraceae being placed under Lythrineae. Economic importance: The family is the source of important oils such as eucalypt us oil (Eucalyptus) used as flavouring and inha-lant, clove oil (Syzygium aromat icum) used as clearing agent and in tooth aches, and oil of bay rum (Pimenta rac emosa). Callistemon is commonly grown as ornamental with its beautiful bottlebru sh like inflorescence (hence the name). Guava fruit is obtained from Psidium gua java. Clove and allspice (Pimenta dioica) include important spices. Fruits of Sy zygium cuminii (jambolan; jamun ) are edible and grown in India and China. Phylogeny: The family presents least taxo-nomic conflicts, almost universally pl aced under Myrtales under Rosids (whether rosid clade, Rasanae, or Rosidae depen ding upon the nomenclature followed by different au* * * * * * * * * * * Lythraceae J. St.-Hilaire Loosestrife family 31 genera, 460 species (including Trapaceae) Widely distributed mainly in tropics, more widespread in America, a few herbaceo us species in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Myrtanae Myrtanae Myrtanae

Rosids * Order Myrtales Myrtales Myrtales Myrtales Myrtales Myrtales Major Families of Angiosperms 575 z r i

E C D Figure 13.87 Lythraceae. Trapa bispinosa. A: Portion of plant with leaves (showi ng swollen petioles) and flowers; B: Flower with front sepals and petals removed; C: Pistil with disc; D: Fruit. E: Fruit of T. natans. Lythrum salicaria F: Branch with fl owers; G: Flower; H: Vertical section of flower; I: Transverse section of ovary. Ammania coccinia. J: Branch with flowers and fruits; K: Flower; L: Cluster of f ruits at node. (F-I, after Bailey, Man. Cult. Pl., 1949. J-L, after Godfrey and Wooten, Aq. wetland Pl. SE US, 1981) 576 Plant Systematics Salient features: Leaves opposite, simple, entire, flowers in racemes or panicle s, bisexual, hypanthium present, petals crum-pled, stamens unequal in length, in two whorls, ovary superior, fruit dry, indehiscent or dehiscent capsule.7 (Cuphea), sometimes by bats (Sonneratia). Cleistogamy prevalent in Peplis and Am mania. Dispersal of seeds occurs through wind or water. Major genera: Cuphea (280 species), Diplu-sodon (70), Lagerstroemia (55), Nesaea (50), Rotala (45), Lythrum (35), Ammania (20), and Trapa (3, sometimes split in to up to 30). Economic importance: The family is known for various ornamentals such as Crepe m yrtle (Lagerstroemia), Mexican heather (Cuphea) and loosestrife (Lythrum). Migon ette tree (Lawsonia inermis) is the source of henna, obtained from leaves. Leave s of Woodfordiaf ruticosa yield red colour, and the bark of Lafoensia pacari a r ed dye. Fleshy seeds of Pomegranate (Punica granatum) are edible, and also used as condiment after dry-ing and powdering. Description: Herbs, shrubs or trees (with flaky bark), rarely spinescent (Lawson ia), rarely aquatic herbs (Trapa). Leaves opposite or whorled , rarely alternate , simple, entire, toothed and with swollen petiole in Trapa, venation pinnate, s tipules absent or repre-sented by minute hairs . Inflorescence of solitary flowers or raceme, panicle or cymose. Flowers bisexual, actinomorphic, rare ly zygomorphic (Cuphea), perigynous with well developed ribbed hypanthium, epica lyx of connate pair of bracts sometimes present be-low hypanthium (Lythrum). Cal yx with 4-8 sepals, free or connate, valvate, often thick. Corolla with usually

4-8 petals, free, imbri-cate, usually attached along the inner rim of hypanthium , crumpled in bud and wrinkled at maturity, sometimes lacking (Peplis, Rotala). Androecium with usually twice as many stamens as petals, in two whorls, outer wh orl alternating with petals, sometimes only in one whorl, usually unequal in len gth, filaments free, anthers bithecous, dorsifixed, deiscence longitudinal, intr orse. Gynoecium with usually 2-6 united carpels, ovary supe-rior, locules as man y as carpels, rarely uni-locular, placentation axile, septa sometimes incomplete and disappearing in upper part of ovary, ovules 2-several, anatropous, ascend-i ng, nectaries at base of hypanthium, style simple, stigma discoid or capitate. F ruit a capsule, indehiscent or dehiscent, rarely berry; seeds usually flattened or winged, seed coat often with hairs becoming mucilaginous on wetting, embryo s traight, endosperm lack-ing. pollination by bees, beetles, flies or birds Phylogeny: The family is commonly placed under Myrtales although Hutchinson incl uded it under order Lythrales. The genus Trapa formerly included in Onagraceae w as separated to Trapaceae, but has now been shifted to Lythraceae (APG-II, 2003; APWeb, 2007; Thorne, 2006, 2007). Onagraceae and Lythraceae share features of t annins scarce, soluble oxalate present, wood with vessels in groups, petiole bun dle arcuate, inflorescence racemose and clawed petals. The family Lythraceae bro adly circumscribed to in-clude Trapaceae, Sonneratiaceae and Punicaceae is monop hyletic as supported by rbcL sequences. Two well defined clades are recognised w ithin the family: one containing Sonneratia, Duabanga, Punica, Lager-stroemia and Lawsonia is characterised b y determinate inflorescence, and wet stigmas; second containing the remaining ge nera racemose inflorescence, reduced number of carpels and dry stigmas. Thorne ( 2006) recognises five subfamilies: first Lythroideae containing bulk of genera ( 27), and the rest monogeneric Duabangoideae, Sonneratioideae, Punicoideae and Tr apoideae. Subsequently (2007) he has merged Punicoideae with Lythroideae and com bined the other three under Lagerstroemioideae, thus recognizing only two subfam ilies. * * * * * * * * * * * Major Families of Angiosperms 577 Onagraceae A. L. de Jussieu Evening Primrose family 17 genera, 650 species Widely distributed mainly in temperate and subtropical regions, very diverse in western North America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Rosidae Series+/Superorder Calyciflorae+ Myrtanae

Myrtanae Myrtanae Rosids * Order Myrtales Myrtales Myrtales Myrtales Myrtales Myrtales Salient features: Herbs and shrubs, leaves simple, flowers usually 4-merous, sep als, petals and stamens inserted on rim of hypanthium, carpels 4, united, placen tation axile, ovary inferior. petaloid, caducous, rarely persistent (Ludwigia). Corolla with usually 4 petals, rarely 2-7, free, sometimes clawed, imbricate, convolute or rarely valvate, rar ely absent. Androecium with usually 4 stamens, mostly as many as petals, sometim es twice as many, rarely only one fertile and one staminode (Lopezia), filaments free, inserted on inner rim of hypanthium, anthers bithecous, sometimes with cr oss partitions, dehiscence logitudinal; pollen grains usually triporate, sometim es tricolporate or biporate, with paracrystalline beaded outer exine, as-sociate d with viscin threads, which help pollen to adher together. Gynoecium with usual ly 4 united carpels, rarely 2 or 5, ovary in-ferior, usually 4 chambered with ax ile placentation, septa sometimes incomplete, or with parietal placentation, ovu les 1-many in each locule, anatropous, with monosporic 4-nucleate megagametophyt e (OenotheraMajor genera: Epilobium (180 species), Oenothera (120), Fuchsia (110), Ludwigia (80), Camisonia (60), Clarkia (45), Gaura (18, some-times merged with Oenothera) , Lopezia (17) and Circaea (12). Although Oenothera is the type genus of the fam ily, the name Oenotheraceae Warming, 1879 is antedated by Onagraceae, 1829 and a dopted by A. L. de Jussieu. Description: Usually herbs, sometimes shrubs (Fuchsia), rarely aquatic herbs (Ju ssiaea) or trees (Hauya), raphides present, stems with internal phloem, often wi th epi-dermal oil cells. Leaves alternate, opposite or whorled , simple, rarely pinnate, entire or toothed, sometimes lobed, venation pin-nate, stipules usually absent, if present caducous (Fuchsia, Circaea). Inflorescence of solitary flowe rs in leaf axils, sometimes spike or raceme, rarely panicle (Fuchsia). Flowers b isexual, actinomorphic, rarely zygomorphic (Lopezia), epigynous with well develo ped hypanthium often prolonged above ovary, . Calyx with usually 4 sepals, rarel y 2-7, free, rarely connate, valvate, sometimes type), nectary near or at base of hypanthium, style slender, stigma capitate or 4-lobed. Fruit a loculicidal capsule, rarely berry (Fuchsia), bristly 1-2 seeded nutlet (Circaea), or 1 seeded nut (Gaura); seeds commonly with hairy tufts (Epi lobium) or wings (Hauya), rarely smooth, embryo straight, endosperm lacking. Pol lination by bees, moths, flies and birds. Dispersal of winged and hairy-tuft see ds by wind, fleshy fruits of Fuchsia by birds, and hooked fruits of Circaea by e xozoochory. 578 Plant Systematics c G

H Figure 13.88 Onagraceae. Oenothera rosea. A: Portion of plant with flowers and f ruits; B: Flower; C: Stamen; ; D: Transverse section of ovary; E: Fruit. Ludwigia alternifolia F: Branch with flowers and fruits; G: Flower from above. L. linearis H: Flower, sid e view; I: Fruit. (A-E, after Sharma and Kachroo, Fl. Jammu, 1983. F-I, after Go dfrey and Wooten, Aq. wetland Pl. SE US, 1981). Epilobium commonly colonises burned areas, hence the name fireweed. Oenothera fl ow-ers often open in late afternoon and thus known as Evening primrose. Phylogeny: The family is commonly placed under Myrtales although Hutchinson (197 3) included it under order Lythrales. The genus Trapa formerly included in this family was separated to Trapaceae Economic importance: The family is known (Cronquist, 1988; Dahlgren, 1989; for s howing flowers. Species of Oenothera Takhtajan, 1997), but has now been shifted (Evening pirmrose) and Clarkia are grown as to Lythraceae (APG-II, 2003; APWeb, 2008; ornamentals in flower beds. Fuchsia shrubs Thorne, 2006, 2007). Onagraceae and are grown in greenhouses or in open warm Lythraceae share features of tannins sc arce, regions. soluble oxalate present, wood with vessels Major Families of Angiosperms 579 in groups, petiole bundle arcuate, inflores-cence racemose and clawed petals. Pe ntamerous members Decodon and Ludwigia are sister to the respective fami-lies Ly thraceae and Onagraceae. Two well defined subfamilies, Ludwigioideae (4-5 merous flowers, hypanthium absent, pollen in tetrads, stigma capitate) and Onagroideae (flowers 4-merous, hypanthium long, stigma divided) are commonly recognised in APWeb (2008), although it is conventional to divide the family into number of tr ibes. Raimann (1893) recognised 8 tribes within the family: Jussieae, Epilobieae , Hauyeae, Onagreae, Gaureae, Fuchsieae and Circeae. Gaureae is often included u nder Onagreae. Despite intensive morphological and molecular stud-ies of Onagrac eae, relationships within the family are not fully understood. Levin et al. Lythraceae found strong support for monophyly of Onagraceae, with Ludwigia as th e basal lineage and a sister-taxon relationship between Megacorax and Lopezia. M ost relationships within Onagreae are weakly resolved, suggesting a rapid divers ification of this group in western North America. Neither Camissonia nor Oenothe ra appears to be monophyletic. The study also showed that the small genus Gongyl ocarpus previously included in tribe Onagreae is strongly sup-ported as sister t o the rest of Onagreae + Epilobieae, and should be placed in its own tribe, Gong ylocarpeae. Subsequent studies of Levin et al. (2004) on two biggest tribes base d on DNA sequence data from one nuclear region (ITS) and two chloroplast regions trnL-trnF and rps16 strongly suggest that tribe gongylocarpeae is sister to tribes (2003) on the basis of parsimony and maxi- Epilobieae + Onagreae, both of which are mum likelihood analyses with rbcL and ndhF monophyletic. Within Onagreae, Ca missonia sequence data for 24 taxa representing all seems to be broadly paraphyl etic, and 17 Onagraceae genera and two outgroup Oenothera is also paraphyletic. * * * * * * * * * * * Subclass 10. Malvidae (B) 3. Muntingiaceae (B) 4. Dipterocarpaceae Superorder 1. Malvanae Order 1. Vitales 5. Sarcolaenaceae Family 1. Vitaceae 4. Thymelaeineae 2. Malvales 1. Thymelaeaceae Suborder 1.Malvineae 2. Tepuianthaceae

1. Malvaceae (B) 3. Rhamnales 2. Grewiaceae (B) 1. Rhamnaceae 3. Byttneriaceae 2. Elaeagnaceae 2. Cochlospermineae 3. Dirachmaceae (B) 1. Bixaceae 2. Diegodendraceae (B) 3. Cochlospermaceae 4. Barbeyaceae (B) 4. Urticaalleess 1. Uceae 4. Sphaerosepalaceae (B) 2. Moraceae 3. Cistineae 3. Urticaceae 1. Neuradaceae (A) 2. Cistaceae 4. Celtidaceae 5. Cannabaceae 580 Plant Systematics Superorder 2. Rafflesianae 17. Tropaeolaceae 18. Akaniaceae 19. Bretschneideraceae Order 1. Rafflesiales 1. Apodanthaceae 2. Rafflesiaceae Superorder 4. Huerteanae 3. Cytinaceae Order 1. Huerteales Superorder 3. Capparanae (B) 4. Mitrastemonaceae 1. Gerrardinaceae 2. Tapisciaceae 3. Dipentodontaceae Order 1. Capparales 1. Capparaceae 2. Cleomaceae 3. Brassicaceae 4. Setchellanthaceae 5. Stixaceae 6. Resedaceae 7. Gyrostemonaceae 8. Pentadiplandraceae (B) 9. Tovariaceae (B) 10. Emblingiaceae (B) 11. Koeberliniaceae (B) 12. Bataceae 13. Salvadoraceae 14. Limnanthaceae 15. Caricaceae 16. Moringaceae Superorder 5. Rutanae Order 1. Rutales 1. Rutineae 1. Rutaceae

2. Simaroubaceae 3. Kirkiaceae 4. Picramniaceae 5. Biebersteiniaceae (B) 6. Tetradiclidaceae (B) 7. Nitrariaceae (B) 8. Meliaceae 2. Anacardiineae 1. Burseraceae 2. Anacardiaceae 3. Sapindineae 1. Sapindaceae Malvaceae A. L. de Jussieu Mallow family 197 genera, 2,865 species (excluding Grewiaceae) Distributed in tropical and temperate climates, mainly in the South American tro pics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Dilleniidae Dilleniidae Magnoliidae Malvidae Thalamiflorae+ Series+/Superorder Malvanae Malvanae Malvanae Eurosids II* Order Malvales Malvales Malvales Malvales Malvales Malvales APG II Malvaceae includes Tiliaceae, Sterculiaceae and Bombacaceae Thorne under suborder Malvineae; Malvaceae, includes Sterculiaceae and Bombacace ae and truncated Tiliaceae (2 genera) but excludes Grewiaceae (majority genera o f former Tiliaceae) Major Families of Angiosperms 581

Salient features: Herbs and shrubs with stellate pubescence, often mucilaginous, leaves palmately veined, stipules prominent, flowers usually with epicalyx, sta mens nu-merous with united filaments, anthers monothecous, carpels five or more, ovary superior, placentation axile. Economic importance: The family is represented by several ornamentals such as Ch ina rose (Hibiscus rosa-sinensis), hollyhock (Althaea rosea) and rose of Sharon (Hibiscus syriacus), Young fruits of okra (Hibiscus esculentus; bhindi ) are used a s vegetable. Cotton is obtained from different species of Gossypium. Cocoa (choc olate source) is obtained from seeds of Theobroma cacao, Cola nitida (both forme rly under Sterculiaceae) yields cola. Seed hairs from Ceiba and Bombax (kapok) a re used as stuffing. Tilia is a tree valuable as timber (Basswood). The wood of T. cordata is particularly good for making furniture and musical instruments, al so grown as ornamental tree. Major genera: Hibiscus (300 species), Sterculia (300), Dombeya (300), Sida (200), Pavonia (200), Abutilon (100), Tilia (50 ), Adansonia (10), Gossypium (20) and Bombax (8). Description. Herbs or shrubs, rarely small (Thespesia) or large (Tilia) trees. P lants of-ten mucilaginous. Leaves alternate, simple, sometimes palmately lobed ( Gossypium), palmately veined, pubescence stellate or of peltate scales, stipules present. Inflores-cence cymose (Pavonia) or flowers solitary axillary. Flowers bracteate (Abutilon) or Phylogeny: The family has been considered quite distinct on the basis of monadel phous stamens with monothecous anthers, though it had been considered quite clos er to Tiliaceae, Bombacaceae and Sterculiaceae by Cronquist (1988) and Takhtajan (1997). These families share the features of presence of stellate hairs, mu-cil aginous cells, pericycle strands above phloem, similar size and pitting of vesse ls, and the distribution of xylem parenchyma. According to Judd et. al., (1999, 2002) the traditional distinctions between these families are arbitrary and inco nsistent, and the merger of four would form a monophyletic Malvaceae. They howev er, concede that genera such as Grewia, Corchorus, Triumfetta , etc., form a cla de which has lost calyx fusion, also suggesting that Grewioideae and Byttnerioid eae form distinct clades within Malvaceae. Traditional Tiliaceae was circumscrib ed by free stamens and bithecous anthers. Thorne (1999, 2000), obviously had kep t Tiliaceae distinct, merging the other two families with Malvaceae. Recent mole cular evidence (Alverson et al., 1998) suggests that half-anthers of the traditi onal Malvaceae are transversely septate bithecous anthers that are strongly conn ate. Earlier Hutchinson (1973) had proebracteate actinomorphic, hypogynous. Calyx with 5 sepals, more or less united, often subte nded by epicalyx (bracteoles), epicalyx 3 (Malva), 5-8 (Althaea) or absent (Sida ). Corolla with 5 petals, free, imbricate, often adnate at base to staminal tube . Androecium with many stamens, filaments united into a tube (monadelphous), epi petalous, anthers monothecous, dehiscence transverse, pollen grains large with s pinous exine, triporate or multiporate, tricolpate in Abutilon. Gyn-oecium with 2-many (usually 5) united car-pels (syncarpous), multilocular (locules as many a s carpels) with many ovules, placentation axile, ovary superior, styles branched above, stigmas as many as carpels or twice as many (Malvaviscus). Fruit a locul icidal capsule or schizocarp (Malva), follicles (Sterculia), rarely a berry (Mal va-viscus); seeds 1-many, embryo curved, endosperm absent. Flowers are insect po lli-nated, nectar usually produced by inner surface of calyx. Dispersal may occu r by wind, water, or animals. Large indehiscent pods of Adansonia are dispersed by large mammals. (Hibiscus) bisexual, 582 Plant Systematics

B H G 7~ I

C A Figure 13.89 Malvaceae. Malva parviflora. A: Plant in flower; B: Portion of flow er with 2 petals and longitudinally split androecium; C: Gynoecium; D: Fruit with persistent caly x. Abutilon indicum. E: Plant with flowers and fruits on long peduncles; F: Caly x; G: Gynoecium with several carpels; H: One fruiting carpel split to show seeds . posed that the monothecous anthers arose from splitting (chorisis) of the filame nts. Restriction site analysis of cpDNA has established that genera with loculic idal capsules and numerous seeds (Hibiscus, Gossypium), form a basal paraphyleti c complex. Genera with schizocarpic fruits, more than five carpels, and ovules o ne or two per carpel depict synapomorphies. APweb recognises following 9 subfami lies under the broadly circumscribes Malvaceae: Malvoideae, Bombacoideae, Stercu lioideae, Tilioideae, Dombeyoideae, Brownlowioideae, Helicteroideae, Grewioideae and Byttnerioi-deae. Thorne who had earlier recognized Tiliaceae as distinct family has finally (2003, 2006) shifte d Tilia and Craigia to Malvaceae under Tilioideae, the remaining genera of famil y Tiliaceae be-ing put under new family Grewiaceae. He recognizes 7 subfamilies under Malvaceae, recognizing Grewioideae and Byttnerioideae as independent famil ies Grewiaceae and Byttneriaceae, respectively. A major shift in his 2007 revisi on puts Malvanae under new subclass Malvidae. * * * * * * * * * * * Major Families of Angiosperms 583 Grewiaceae (Dippel) Thorne Grewia family 31 genera, 390 species Widely distributed in tropics and subtropics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II/(APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae

Magnoliidae Malvidae Series+/Superorder Thalamiflorae+ Malvanae Malvanae Malvanae Eurosids II* Order Malvales Malvales Malvales Malvales Malvales Malvales B & H, Cronquist, Takhtajan, and Dahlgren under family Tiliaceae. APG II and APw eb do not recognize Tiliaceae or Grewiaceae as separate family, merge with Malva ceae. Thorne under suborder Malvineae

L I

A G Figure 13.90 Grewiaceae. Grewia tenax. A: Portion of plant with flowers and frui ts; B: Flower with sepals and petals removed; C: Stamen; D: Transverse section of ovary. Corch orus capsularis. E: Portion of plant with flowers and fruits; F: Flower from abo ve; G: Fruit. H: Corchorus aestuans, portion of plant with fruit. 584 Plant Systematics

Salient features: Shrubs or trees, leaves with asymmetrical base, pubescence of branched hairs, stamens numerous with free or united filaments, anthers bithecou s, carpels five or more, ovary superior, placentation axile. placentation axile, ovary superior, style sin-gle, stigma lobed or capitate. Fru it a capsule or fleshy. Seeds 1-many, embryo straight, endosperm present. Major Genera: Grewia (150 species), Triumfetta (70) and Corchorus (50). Economic importance: Jute is obtained from stem fibres of Corchorus capsularis a nd C. olitorius. Leaves of C. olitorius are used for food in many eastern Medite rranean countries. Description: Shrubs or trees, rarely herbs (Corchorus, Triumfetta). Leaves alter nate, sim-ple, deciduous, bases asymmetrical, pubes-cence of branched hairs, sti pules present. Inflorescence cymose, usually in small clus-ters in leaf axils. F lowers bisexual, rarely uni-sexual, actinomorphic, hypogynous. Calyx with 3-5 se pals, free or connate, valvate. Co-rolla with 3-5 petals, free, imbricate or val vate, sometimes with glandular hairs at their bases, rarely absent. Androecium w ith many stamens, sometimes only 5 (Triumfetta pentandra), filaments free or uni ted into groups of 5 or 10 (polyadelphous), adnate to base of petals, anthers bi thecous, dehiscence longitudinal, or by apical pores. Gynoecium with 2-many unit ed carpels, multilocular (locules as many as carpels) with many ovules, Phylogeny: The family has been considered to be distinct for a long time, with r emoval of genera of the family Tiliaceae to Flacourtiaceae by Engler and Prantl (1887 1915). The family is distinct from now en-larged Malvaceae in free stamens and bithecous anthers. APG classifications (APG II, APweb) have merged Tiliacea e with Malvaceae. Thorne (1999), treated Tiliaceae as distinct family (including subfamilies of Malvaceae: Tilioideae and Grewioideae in APweb). Subsequently (2 003) he has merged Tilioideae (Tilia and Craigia) with Malvaceae, recognizing Gr ewioideae as independent new family Grewiaceae (new name necessitated due to shi fting of Tilia, the type of family Tiliaceae). * * * * * * * * * * * Dipterocarpaceae Blume Meranti family 17 genera, 550 species Distributed mainly in tropical Asia and Indomalaysia, also represented in Africa and South America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Malvidae Series+/Superorder Thalamiflorae+ Malvanae Malvanae Malvanae

Eurosids II* Order Guttiferales Malvales Malvales Malvales Malvales B & H as Dipterocarpeae. Thorne under suborder Cistineae Major Families of Angiosperms 585 Theales

II A B E Figure 13.91 Dipterocarpaceae. Dipterocarpus trinervis. A: Branch with flower; B : Calyx and corolla; C: Vertical section of flower; D: Longitudinal section of ovary; E: Stame n with sterile tip above anther; F: Transverse section of ovary. G: Fruit of D. pilosus with two long wings. H: Fruit of Parashorea stellata with five wings. Salient features: Small or large trees with buttressed bases, leaves evergreen, alter-nate, often with domatia, flowers perigynous or epigynous, in racemes or p anicles, sepals becoming winged in fruit, petals 5, often leathery, anthers with sterile tips, carpels 3, fruit a winged nut. bisexual, actinomorphic, often showy, fragrant, hypogynous. Calyx with 5 sepals, free or slightly connate, sometimes enlarged and winged in fruit. Corolla with 5 petals, free or connate at base, spirally twisted, often leathery. Androecium with 5-numerous stamens, filaments free or connate at base, anthers bithecous, dorsifixed (MonotoiMajor genera: Shorea (150 species), Hopea deae), or basifixed (Dipterocarpoideae ), (110), Dipterocarpus (80), Vatica (60) and dehiscence longitudinal, anthers w ith Monotes (26). sterile tip formed by extension of connective, Description: Small or large trees, often but-tressed at the base, trunk very lon g and smooth, branched at top with cauliflower-shaped crown, usually with specia l resin canals exuding aromatic dammar from wounds, nodes trilacunar or pentalac unar, roots with ectomycorrhiza. Leaves alternate, distichous, coriaceous, simpl e, evergreen, covered with fasciculate or stellate hairs, stipules present and f requently containing domatia housing insects, usually early shed-ding. Infloresc ence racemose, axillary or terminal racemes or panicles. Flowers pollen grains tricolpate or triporate. Gynoecium with 3 united carpels, ovary su perior or partly inferior (Anisoptera), 3-locular with 2 ovules in each chamber, placentation axile, ovules pendulus, anatropous, bitegmic, crassinucellate, onl y one ovule develops further. Fruit a single seeded nut with winged and membrano us calyx; seeds without endosperm, cotyledons often twisted, enclosing radicle. Economic importance: Many species of Dipterocarpus, Shorea, Hopea and Vatica usu ally grow together in tropical rain forests and 586 Plant Systematics are principal sources of hardwood timber. The wood is pale in colour and in grea

t demand for plywood and block wood. Dammar resin obtained from the tree is used for special varnishes. clade having plant with secretory canals, calyx imbricate, two outer members oft en different from the rest, filaments not articulated, ovules both anatro-pous a nd atropous; exotegmen curved inwards in chalazal region, and there is a strong case for merging former two in Dipterocarpaceae. Phylogenetic studies on family Dipterocarpaceae based on morphological and rbcL sequence data (Dayanandan et al ., 1999) have shown that Monotoideae and Pakaraimaeoideae are cladistically basa l, representing primitive members of the family. Thorne (2003) had earlier inclu ded the family (and the order Malvanae) under superorder Rosanae, but subsequent ly (2006) shifted to Malvanae. Phylogeny: The family is related to Ochnaceae, Elaeocarpaceae, Grewiaceae and ot her members of Malvales. Cronquist con-siders it closer to Guttiferae and Theace ae in addition to Ochnaceae. The family is usu-ally divided into 3 subfamilies: Monotoideae, Pakaraimaeoideae and Dipterocarpoideae. Molecular studies of Kubitz ki & Chase, (2002) have shown that Sarcolaenaceae, Cistaceae and Dipterocarpacea e form a well defined * * * * * * * * * * * Rhamnaceae A. L. de Jussieu Buckthorn family 53 genera, 875 species Distributed worldwide but more common tropical and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Malvidae Series+/Superorder Disciflorae+ Rutanae Rutanae Malvanae Eurosids I* Order Rhamnales Rhamnales Sapindales Sapindales Rhamnales Rosales Salient features: Trees and shrubs, leaves toothed with strong secodary veins, s tipu-late, flowers perigynous, petals strongly con-cave, stamens opposite the pe

tals, hypanthium with nectary inside, ovules on basal placentas. Description: Erect or climbing shrubs, trees, rarely herbs, sometimes thorny, sometimes associated Actinomycetes bacteria, stems often modified into thorns, tendrils or hooks. Lea ves usually opposite, sometimes alternate, simple, toothed, venation reticulate with strong secondary veins, stipules usually present and often modifies into sp ines, leaves sometimes rudimentary. Inflorescence axillary corymb or cymose clus ters, rarely solitary. with nitrogen-fixing Major genera: Rhamnus (150 species), Phylica (150), Ziziphus (100), Ceanothus (4 0), Gouania (35), Colubrina (15), Berchemia (10) and Sageretica (10). Major Families of Angiosperms 587

F D Figure 13.92 Rhamnaceae. Rhamnus purpurea. A: Branch with flowers; B: Flower; C: Vertical section of flower; D: Transverse section of ovary; E: Fruit. F: Fruiti ng shoot of Zizyphus mauritiana. G: Flowering shoot of Z. nummularia. (A, B and E, after Pol unin and Stainton, Fl. Himal., 1984; F and G, after Maheshwari Illus. Fl. Del., 1966). Flowers small, usually inconspicuous, actinomorphic, bisexual, rarely unisexual, perigynous, hypanthium present. Calyx with 5 sepals, rarely 4, free or united, lobes valvate. Corolla with 5 petals, rarely 4, free, sometimes absent, often co ncave and hooded (cucullate) over anthers, usually clawed. Androecium with as ma ny as petals and opposite them, arising from outside disc that lines the rim of hypanthium, anthers bithecous, dehiscence longitudinal, anthers with sterile tip formed by extension of con-nective, pollen grains tricolpate or triporate. Gyno ecium with 2-4 united carpels, rarely 5, locules as many, ovary superior or part ly inferior (Gouania) due to adnation with disc, ovule 1 in each locule on basal pl acenta, pendulus, anatropous, style one, often lobed or cleft. Fruit a drupe wit h 1-many endocarps, capsule or samaroid nut; seed large, usually straight, somet imes curved, without or with scanty endosperm. Economic importance: The family yields important fruits from species of Ziziphus (Z. jujuba, the jujube or Chinese date; Z. mauritiana, the Indian jujube; Z. lo tus, lotus fruit), also used to make jelly-like candy. Plants of the family yiel ds dyes include Rhamnus cathartica (green dye from 588 Plant Systematics sap), R. tinctoria (yellow dye from fruits), R. chlorophora (Chinese green indig o from bark) obtained from the members of this family include. Several species a re also used me-dicinally: fruits of R. cathartica and R. purshiana are strongly laxative; extract of Gouania bark are used as woud dressing in Africa; Ventilag o oblongifolia is used to treat cholera in Malaya. Many species of Ceanothus (te a bush) with beautiful panicles of blue, pink or white flowers are grown as orna mentals. Other members used as orna-mental include Hovenia (raisin tree), Berche mia (supplejack), Poliurus (Jerusalem thorn) and Reynosia. Elaeagnaceae, Dirachmaceae and Barbeyaceae under Rhamnales, but shifts Vitaceae to independent order Vitales under Malvanae. APG -II includes Vitaceae unplaced in Rosids, whereas APWeb included it under Vitales but towards end of Core Eudic ots. Takhtajan (1997) also separated two families under distinct orders. The fam

ily Rhamnaceae shows affinities with Rosales, under which it is placed in both A PG-II and APWeb. There are three main clades in the family are recognised by APW eb, the rhamnoids, which include Maesops and Ventilago (three tribes), the zizip hoids (five tribes), which include most of the rest of the family, and the ampel oziziphoids (three tribes). The four families included under the order by Thorne (2006) form a well defined clade (Sytsma et al. 2002), with dense curly hairs o n abaxial surface of leaf, a possible synapomorphy. Rhamnaceae is sister to this clade. Phylogeny: Family is often placed closer to Vitaceae, in the same order Rhamnale s (Rendle, Cronquist, Hutchinson but he also included Elaeagnaceae and Heteropyx ida-ceae). Thorne (2006) retains Rhamnaceae, * * * * * * * * * * * Ulmaceae Mirbel Elm family 7 genera, 40 species Mainly distributed in temperate region, also tropics and subtropics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Hamamelidae Dilleniidae Magnoliidae Malvidae Series+/Superorder Unisexuales+ Urticanae Malvanae Malvanae Eurosids I* Order Urticales Urticales Urticales Urticales Rosale B & H under family Urticaceae Salient features: Trees or shrubs with wa-tery sap, sieve-tube plastids P-type, leaves simple, serrate or biserrate, vascular bundles entering teeth, leaf base oblique, vena-tion pinnate, flowers often bisexual, fruit a winged samara or dru pe. Major genera: Ulmus (20 species), Zelkova (6), Phyllostylon (3) and Planera (1). Description: Shrubs or trees without laticifers, often with tannins, cystoliths

present, branching profusely and often Major Families of Angiosperms 589 1 i 4 ~

C L

n 4 Cl . '

a o A F I G K Figure 13.93 Ulmaceae. Ulmus chumlia. A: Short shoot, lower surface & corky bark ; B: Leaf of coppice shoot, upper surface; C: Normal adult short shoot, upper surface; D: Lea f margin; E: Indumentum of midrib portion, lower surface; F: Flowering shoot; G: Fruiting shoot; H: Flower; I: Slightly older flower; J: Gynoecium; K: Bract; L: Inner bud scale; M: Mature winged fruit. (After Melville and Heybroek, 1971) spreading. Leaves alternate, rarely opposite, simple, serrate or biserrate, base often ob-lique, venation pinnate, reticulate, vascu-lar bundles entering teeth, stipules present but falling early. Inflorescence consisting of axillary cymose clusters. Flowers small, actinomorphic, bisexual or unisexual and monoecious, h ypogynous or perigynous. Perianth with 4-9 tepals, free or connate, representing sepals (petals absent), imbricate. Androecium with 4-9 stamens, as many as tepals and opposite them, som e-times adnate to tepals, pollen grains 4-6 porate. Gynoecium with 2 united carp els, ovary superior, unilocular, ovule 1, placentation apical, stigmas 2, decurr ent on 590 Plant Systematics style. Fruit a nut or samara; seed flat with straight embryo, endosperm forming a sin-gle layer and appearing absent. Pollination by wind. Winged fruits are als o dispersed by wind, nut-like fruits of Planera dispersed by water. fruit a samara or nut. The family was earlier divided into two subfamilies: Celt idoideae (drupe-like fruit, three palmate veins, sieve tube plastids S-type; sty le with single vascular bundle, embryo curved) and Ulmoideae (fruit samara, vein

s pinnate; sieve tube plastids P-type, style with 3 vascular bundles, embryo str aight). The former has now been separated as a distinct family Celtidaceae. The family is often placed in order Urticales. Cronquist places it under Hamamelid c omplex but others including Dahlgren and Thorne place them along with other Malv anean groups. Takhtajan also places them closer to Malvales but under superorder Urticanae. APG classifications place them closer to Rosaceae, and Rhamnaceae un der order Rosales, Rosaceae being considered sister to rest of the families. Economic importance: Various species of Ulmus (elm) and Zelkova provide timber u sed for furniture, posts and under water pillings. Ulmus americana and other spe cies are grown as ornamentals and important shade trees. Mucilaginous inner bark of U. rubra has me-dicinal importance. Phylogeny: The family was earlier included under Urticaceae (Bentham and Hooker) but now separated due to veins of leaves running directly into teeth, flowers o ften bisexual and * * * * * * * * * * * Moraceae Link Mulberry family 37 genera, 1,100 species Distributed mainly in tropics and subtropics with some species in temperate regi ons. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Hamamelidae Dilleniidae Magnoliidae Malvidae Series+/Superorder Unisexuales+ Urticanae Malvanae Malvanae Eurosids I* Order Urticales Urticales Urticales Urticales Rosales B & H under family Urticaceae Salient features: Trees and shrubs with milky latex, leaves alternate, flowers u

ni-sexual, small, carpels usually 2, ovary supe-rior, single chambered, ovule 1. Description: Trees or shrubs, sometimes lianas, rarely epiphytic in early stage (strangling species of Ficus), often with milky latex distributed in all parench ymatous tissues, cystoliths present, usually globose, tannins often present. Lea ves alternate (rarely opposite), usually distichous, simple with entire or lobed margin, with pinnate or Major genera: Ficus (600 species), Dorstenia (110), Artocarpus (50), Morus (15) Maclura (12) and Broussonetia (8). Major Families of Angiosperms 591 D N I H" A

A C H B Figure 13.94 Moraceae. Ficus cunia. A: Branch with leaves; B: Branch bearing fig s; C: Longitudinal section of hypanthodium (fig, receptacle); D: Female flower; E: Gynoecium; F : Male flower with single stamen; G: Stamen. H: Twig of Ficus religiosa. Morus a lba. I: Male branch; J: Female branch; K: Female flower with closely appressed p erianth; L: Longitudinal section of female flower; M: Male flower with four tepa ls and 4 stamens; N: Multiple fruit (Sorosis).

palmate reticulate venation, stipules present and leaving a circular scar when s hed. Inflorescence of various types, erect or pendulous (catkin) spike (Morus), hypanthodium (Ficus), or raceme. Flowers small, unisexual (monoecious or dioecio us), actinomorphic, hypogynous. Perianth usu-ally with 4-6 tepals, (representing sepals, petals absent), free or united, often persist-ing and becoming fleshy i n fruit, sometimes absent. Androecium with 4-6 (as many as tepals) stamens, oppo site the tepals, fila-ments free, incurved in bud or straight, anthers bithecous or monothecous, dehiscence longitudinal, pollen grains multipora te or with 2-4 pores. Gynoecium with 2 united car-pels, ovary superior, unilocul ar, ovule 1, anatropous to campylotropous, placentation apical, styles usually 2 . Fruit usually a multiple fruit sorosis (Morus), syconium (syconus; Ficus), som etimes etaerio of drupes or a berry; seed with curved or straight embryo, endosp erm present or absent. Economic importance: The family is important for its fruits such as mulberry (Mo rus 592 Plant Systematics alba, M. nigra), fig (Ficus carica) and bread-fruit (Artocarpus altilis). Fruits of Artocarpus heterophyllus ( kathal ) are cooked as vegeta-ble, whereas those of A . lakoocha ( dheon ) are pickled. Leaves of Morus are also used for raring silkworms . Various species of Ficus including F. elastica (Indian rubber tree or rubber p lant) are grown as ornamentals. being considered sister to rest of the families. Cecropia and related genera ear lier included under Moraceae (Hutchinson and earlier authors), and separated und er Cecropiaceae by APG (1998), Thorne (1999, 2000) and Judd et al. (1999, 2002) are inter-mediate between Moraceae and Urticaceae, but closer to Urticaceae in r estriction of laticifers to bark, basal ovule, straight embryo and with one carp el aborted (pseudo-monomerous). The family Cecropiaceae has appropriately been m erged with Urticaceae by APG II (2003), APweb (2003) and Thorne (2003). The fami ly Moraceae as narrowly circumscribed here is monophyletic as sup-ported by rbcL sequences (Sytsma et al., 1996). The reduction of one carpel is also in-dicated in slightly or strongly unequal styles in Artocarpus, Dorstenia and Ficus. A co mplete loss of one of the two styles probably occurred in common ancestor of Cec ropiaceae +Urticaceae clade. Phylogeny: The family was earlier placed in Urticaceae (Bentham and Hooker) but now considered distinct in woody habit with milky latex, 2 carpels, ovary with s ingle apical ovule and usually curved embryo. Cronquist places Urticales (includ ing Urticaceae and related families) under Hamamelid complex but oth-ers includi ng Dahlgren and Thorne place them along with other Malvanean groups. Takhtajan a lso places them closer to Malvales but under superorder Urticanae. APG classi-fi cations place them closer to Rosaceae, and Rhamnaceae under order Rosales, Rosac eae * * * * * * * * * * * Urticaceae A. L. de Jussieu Nettle family 44 genera, 1080 species Widespread in tropics and temperate climates, poorly represented in Australia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne

APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Class Dicotyledons Subclass Monochlamydeae Hamamelidae Dilleniidae Magnoliidae Malvidae Series+/Superorder Unisexuales+ Urticanae Malvanae Malvanae Eurosids I* Order Urticales Urticales Urticales Urticales Rosales Salient features: Usually herbs with sting-ing hairs, leaves with stipules, flow ers small, unisexual, tepals and stamens usu-ally 4 each, carpel 1, style 1, fru it achene or fleshy drupe. Major genera: Pilea (370 species), Elatostema (170), Boehmeria (80), Urtica (50) Parieteria (30) and Laportea (20). Description: Usually herbs, rarely trees or Major Families of Angiosperms 593 le F C G I t

E A Figure 13.95 Urticaceae. Urtica hyperborea. A: Plant with axillary clusters of f lowers; B: Male flower with four tepals and four stamens; C: Female flower with unequal tepals; D: Achene surrounded by persistent perianth. Boehneria platyphylla. E: Plant wit h interrupted spikes; F: Female flower with bristly perianth and hairy style; G: Vertical section of female flower; H: Gynoecium; I: Transverse section of achen e. shrubs, sometimes climbers, with often milky latex restricted to bark or reduced with clear sap, cystoliths present, usually elongated, tannins often present, h airs simple, usually stinging. Leaves alternate or opposite, usually distichous, simple with entire or lobed margin, with pinnat e or palmate venation, reticulate, stipules present, leaf base cordate or assyme trical. Inflorescence cymose or heads, sometimes 594 Plant Systematics with solitary flowers. Flowers small, uni-sexual (monoecious or dioecious), acti nomorphic, hypogynous. Perianth usu-ally with 4 tepals (representing sepals, pet -als absent), rarely only 3 or upto 6, free or united, imbricate or valvate. And roecium with 4-5 (as many as tepals) stamens, op-posite the tepals, filaments fr ee, incurved in bud, reflexed at anthesis, anthers bithecous, dehiscence longitu dinal, pollen grains multiporate or with 2-3 pores. Gynoecium with single carpel (actually 2 but with one reduced: pseudo-monomerous), ovary superior, unilocula r, ovule 1, orthotropous, placentation basal, style 1, stigmas 1 or 2, extending on style or capi-tate. Fruit usually an achene, embryo straight, endosperm some times lacking. elongate cystoliths, laticifers restricted to the bark, with clear sap, incurved stamens, pseudomonomerous gynoecium and basal ovule. Thorne (1999) and the APG classifications (APG, 1998; Judd et al., 1999, 2002) also included in Urticaceae the genus PoikiloEconomic importance: In addition to being a noxious weed, Urtica dioica (common sting-ing nettle) yields silky bast fibre. Fibre is also extracted on the commer cial scale from Boehmeria nivea (ramie or china grass). Species of Pilea and Sol eirolia (baby s tears) provide important ornamentals. spermum, formerly placed in Cecropiaceae. Cronquist places Urticales (including Urticaceae and related families) under Hamamelid complex but others including Da hlgren and Thorne place them along with other Malvanean groups. Takhtajan also p laces them closer to Malvales but under superorder Urticanae. APG classification s place them closer to Rosaceae, and Rhamnaceae under order Rosales, Rosaceae be ing considered sister to rest of the families. Single carpel in the family has b een derived through abortion of the second carpel, as borne out by the aborted v ascular bundles in the ovary of Urtica and Laportea. The basal placentation has similarly been derived from apical placentation of Moraceae. This is inferred fr om the fact that in Boehmeria cylindrica the vascular bundle supplying the ovule ascends the carpel wall for a short distance and then reverses direction to ent er the ovule at the base of the ovary. The family Cecropiaceae, which was former ly recognized (APG, 1998; Judd et al., 1999,2002; Thorne, 1999) as distinct fami ly has finally been merged with Urticaceae (Thorne, 2003; APG II, 2003; APweb, 2 003). Phylogeny: The family was earlier broadly circumscribed (Bentham and Hooker) to

include families which have now been sepa-rated as Moraceae, Ulmaceae, Celtidace ae, etc. The family is now circumscribed to include mainly herbaceous species wi th * * * * * * * * * * * Major Families of Angiosperms 595 Rafflesiaceae Dumortier Ra esia family 3 genera, 20 species Southeast Asia, from India to Indonesia. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Monochlamydeae Rosidae Magnoliidae Magnoliidae Malvidae Rafflesianae Magnolianae Rafflesianae Series+/Superorder Multiovulatae Terrestres+ Order Rafflesiales Rafflesiales Rafflesiales Rafflesiales Uncertain B & H as family Cytinaceae Salient features: Parasitic on stem and roots, plant body filamentous like a fun gal myc-elium, flowers usually unisexual, with fleshy petaloid calyx, stamens in a column, ovary inferior, carpels fused, placentation parietal, fruit fleshy. with 4 8 carpels, ovary inferior, unilocular with 4-14 parietal placentas, or 3 10( 20 ) locular by deep intrusion of the placentas (Rafflesia), ovules 50 100 per locule , non-arillate, hemianatropous to anatropous, bitegmic, tenuinucellate, united, style expanded into an often large, complex disk, with stigmatal projections. Fr uit usually fleshy berry, seeds endospermic, minute, embryo rudimentary. Major genera: Rafflesia (16 species), Sapria (2) and Rhizanthes (2). Description: Total parasites on stems and roots of angiosperms, vegetative part filamentous, like fungal mycelium, rootless, permeating the host tissues, with o nly the flowers or the flowering stems emerging from the host tissue, xylem with out vessels. Economic importance: None. Leaves much reduced, present at the bases Phylogeny: The family is closely related to Hydnoraceae, often placed in Magnolo id complex under a distinct order Rafflesiales. Cronquist, however, placed this order under of flowering stems, or beneath the flower, or Rosidae. The family is often consi dered closer absent, alternate, opposite, or whorled, of to Aristolochiaceae bec ause of similar perimembranous scales, stomata absent. Inflo- anth. The recent c ladistic studies, however, rescence with solitary flowers. Flowers small place H

ydnoraceae and Aristolochiaceae unto very large (Rafflesia arnoldii, with the la rg- der Piperales based on multigene analyses. est known flowers in angiosperms, up to 1 m The position of Rafflesiaceae still remains in diameter), regular, usually unisexual, cy- uncertain in APG II and APweb. Nic krent clic. Perianth with tepal green or petaloid, 4, or 5( 10), free, or united into tu be, usually fleshy, imbricate, rarely valvate. Androecium with 5 100 stamens, unit ed with the gynoecium, free or filaments united into a tube round the stylar col umn, 1 whorled, fila-ments slender, or reduced, anthers monothecous or bithecous , dehiscing by lon-gitudinal slits, pores, or transversely, pollen grains usuall y nonaperturate. Gynoecium (2002) considers this family to be closer to Malvales. Thorne had earlier (1999) broadly circumscribed the family Rafflesiaceae, divided into 4 subfamilies: Mit rastemonoideae (Mitrastemon-flowers bisexual and solitary, ovary superior), Cyti noideae (Cytinus and Bdallophyton-flowers unisexual, in racemes, stamens in one ring, ovary inferior with 8 14 placentas), Apodanthoideae (Apodanthes, Pilostyle s and Berlinianche- flowers small, 596 Plant Systematics

- isC ,Y +r

k B

I ~1

~i 1 U Iii T Figure 13.96 Rafflesiales. Rafflesiaceae (A-C). A: Fully opened flower of Raffle sia speciosa (photo courtesy Julie Barcelona, Manilla, Philippines). B: Seed of R. arnoldii. C: Part ial section of seed showing undivided embryo. Cytinaceae (D-G). Cytinus hypocist is. D: Plant in flower; E: Vertical section of male flower; F: Vertical section

of female flower; G: Portion of transverse section of ovary. Apodanthaceae (H-J) Pilostyles berterii. H: Host twig with flowers of Pilostyles emerging out; I: V ertical section of male flower; J: A head of stamens. unisexual, solitary, stamens in 2 or 4 rings, ovary inferior with 4 placentas or 1 continu-ous placenta) and Rafflesioideae (Flowers soli-tary and unisexual, la rge, stamens in 1 ring, ovary inferior with many irregular chambers). These have now (2003) been recognized as independent families Mitrastemonaceae, Cytinaceae , Apodanthaceae and Rafflesia-ceae, respectively. APG II and APweb also recogniz e them as independent families but unplaced towards the end of angiosperms, the family placement of Bdallophyton be ing uncertain. According to Judd et al. (2002), Rafflesiaceae (also Balanophorac eae and Hydnoraceae) look so different from other flowering plants that no one h as been sure where to place them. Hydnoraceae appear to belong in Piperales, the other two, according to them are apparently dicots, but are no more precisely p laced than that. * * * * * * * * * * * Major Families of Angiosperms 597 Capparaceae A. L. de Jussieu Caper family 13 genera, 450 species Widespread in tropical and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Malvidae Series+/Superorder Thalamiflorae+ Violanae Violanae Capparanae Eurosids II* Order Parietales Capparales Capparales Capparales Capparales Brassicales B & H as Capparidaceae Cronquist, Thorne and Takhtajan as Capparaceae APG II and APweb do not recognize as separate family, rather merge with Brassica

ceae C E

A Figure 13.97 Capparaceae. A: Capparis decidua with flowers, leaves absent on bra nches. C. sepiaria. B: Portion of flowering branch; C: Small portion of fruiting branch ; D: Flower with numerous stamens. E: Transverse section of ovary with intruded placentae. 598 Plant Systematics Salient features: Shrubs or trees, sepals and petals 4 each, free, stamens many, ovary unilocular with parietal placentation, superior, sometimes with gynophore , fruit a capsule or berry. ( dela ) is pickled and also given to heart pa-tients. The dried floral buds of C. s pinosa are called capers and are used in seasoning. Major Genera: Capparis (350 species), Maerua (100), Boscia (37), Cadaba (30) and Crataeva (20 species) Phylogeny: Heterogeneity of broadly circumscribed Capparaceae was long recognize d. Hutchinson (1973), on the basis of extensive studies at Kew concluded that th e family consisted of two distinct groups which are not really phylogenetically related. True Capparids, according to him are woody plants with indehiscent frui ts, without a replum, and fairly closely related to Flacourtiaceae, whereas Cleo me and its relatives are herbs with dehiscent fruits with replum, as in family B rassicaceae. This view was confirmed by morphological studies (Judd et. al., 199 4) and rbcL sequences (Rodman et al., 1993) as mentioned under Brassicaceae, ult imately leading to the merger of Capparaceae with Brassicaceae in APG classifica tions. Al-though position was kept up in the APG II, but was pointed out that res urrection of Capparaceae and Cleomaceae may be appropriate in the future . This ch ange in position has largely been on account of the re-sults of the studies by H all, Sytsma and Iltis (2002), who on the basis of chloroplast DNA sequence data, concluded that the three form distinct strongly supported monophyletic groups, as is also supported by morphological data. According to Puri (1950), bicarpella ry syncarpous unilocular ovary with parietal placentation is derived from tetrac arpellary condition with axile placentation. Thorne who had earlier (1999) recog nized Capparaceae (also including Cleome and its relatives) and Brassicaceae as distinct families, has subsequently (2003) Description. Shrubs (Capparis), rarely trees (Crataeva) or climbers (Maerua). Le aves al-ternate, rarely opposite, simple, stipules present, sometimes reduced to glands or spines (Capparis). Inflorescence typically racemose, corymbose (Crata eva), or in um-bels (Capparis). Flowers bracteate (bracts often leafy), actinomo rphic or zygomorphic (Capparis), bisexual, rarely unisexual or po-lygamous (Crat aeva), hypogynous, thalamus often prolonged into androgynophore. Calyx with 4 se pals, rarely 2-8, free or connate (Maerua), in two whorls, sometimes in one whor l. Corolla with 4 petals, cruciform (ar-ranged in a cross), rarely 8 or even lac king (Maerua), clawed. Androecium with 4 or more stamens, free, often arising fr om androphore (lower portion of androgynophore), dehiscence longitudinal, nectaries often present near base of sta-me ns. Gynoecium with 2-12 united carpels (syncarpous), unilocular with one-many ov

ules, replum absent, placentation parietal, ovary superior, often on a gynophore (upper part of androgynophore), style 1, stigma capi-tate or bilobed. Fruit a b erry, capsule, drupe, or nut, often stalked; seeds 1-many, embryo curved, endosp erm usually absent. Economic importance: The family contribseparated Cleome and relatives under disutes a few ornamentals such as Capparis tinct family Cleomaceae, as suggested in and Crataeva. The fruit of Capparis decidua APG II. * * * * * * * * * * * Major Families of Angiosperms 599 Cleomaceae Horaninow Spider plant family 11 genera 300 species Widespread in tropical, subtropical and warm temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Malvidae Series+/Superorder Thalamiflorae+ Violanae Violanae Capparanae Eurosids II* Order Parietales Capparales Capparales Capparales Capparales Brassicales B & H under Capparidaceae Cronquist, Dahlgren and Takhtajan under Capparaceae AP G II and APweb under Brassicaceae Figure 13.98 Cleomaceae. Cleome gynandra. A: Lower part of the plant with palmat ely compound leaves; B: Inflorescence with flowers having conspicuous androgynop hore; C: Sepal; D: Petal; E: Stamen; F: Gynoecium with distinct gynophore; G: Transverse se ction of ovary with parietal placentation; H: Seed. 600 Plant Systematics Salient features: Herbs, sepals and petals 4 each, free, stamens many, ovary uni

locu-lar with parietal placentation, superior, sometimes with gynophore, replum present, fruit a capsule or follicle. capsule or siliqua, often stalked; seeds 1-many, embryo curved, endosperm usuall y absent. Major Genera: Cleome (200 species), Podandrogyne (10) and Polanisia (7). Economic importance: The family contributes a few ornamentals such as Cleome and Polanisia. The decoction of Cleome chelidonii is used to cure scabies. Description. Annual or perennial Herbs. Leaves alternate, rarely opposite, simpl e or palmately compound, stipules present. Inflorescence typically racemose, cor ymbose (Cleome). Flowers bracteate, bracts often leafy, actinomorphic, bisexual, hypogynous, thalamus often prolonged into androgyno-phore. Calyx with 4 sepals, rarely 2-8, free, in two whorls, sometimes in one whorl. Corolla with 4 petals, cruciform (arranged in a cross), clawed. Androecium with 4 or more stamens, fre e, often arising from androphore (lower portion of androgynophore), dehiscence l ongitudinal, nectaries often present near base of stamens. Gynoecium with 2-12 u nited carpels (syncarpous), unilocular (usually bilocular due to false septum an d with distinct replum) with one -many ovules, placentation parietal, ovary supe rior, often on a gynophore (upper part of androgynophore), style 1, stigma capit ate or bilobed. Fruit a Phylogeny: Members of the family are generally included under family Capparaceae . Hutchinson (1973) on the basis of extensive studies at Kew concluded that Cleo me and its relatives are distinct from capparids. APG II classification included Capparaceae (in-cluding Cleomaceae) under Brassicaceae, but was pointed out tha t resurrection of Capparaceae and Cleomaceae may be appropriate in the future . Thi s change in position has largely been on account of the re-sults of the studies Hall, Sytsma and Iltis (2002), who on the basis of chloroplast DNA sequence data , concluded that the three form distinct strongly supported monophyletic groups, as is also supported by morphological data. Thorne who had earlier (1999) inclu ded this family under Capparaceae, has subsequently (2003) separated Cleome and relatives under distinct family Cleomaceae, as suggested in APG II. * * * * * * * * * * * Brassicaceae Burnett Mustard family (=Cruciferae A. L. de Jussieu) 340 genera, 3,350 species A cosmopolitan family mainly distributed in North Temperate Zone, particularly t he Mediterranean region. Major Families of Angiosperms 601 Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Polypetalae Dilleniidae Magnoliidae Malvidae Series+/Superorder Thalamiflorae+ Violanae

Violanae Capparanae Eurosids II* Order Parietales Capparales Capparales Capparales Capparales Brassicales B & H as Cruciferae others as Brassicaceae APG II and APweb Brassicaceae includes Capparaceae Figure 13.99 Brassicaceae. Brassica campestris. A: Upper part of plant with infl orescence; B: Lower leaf; C: Vertical section of flower; D: Siliqua with persistent style f orming a long beak. Capsella bursa-pastoris. E: Plant with inflorescence; F: Flo wer; G: Flower with sepals and petals removed; H: Silicula with apical notch hav ing persistent style, fruit flattened at right angles to the septum and as such replum appearing as vertical rim. Coronopus didymus. I: Plant with highly dissec ted leaves and axillary racemes; J: Flower from above showing minute petals and 2 stamens; K: Stamen; L: Silicula, deeply bilobed and prominent replum. M: Silic ula of Lobularia maritima flattened parallel to the false septum and as such rep lum forming a ring around the fruit. N: Siliqua of Brassica nigra dehisced with valves separating and seeds attached to false septum. 602 Plant Systematics Salient features: Herbs, sap watery, sepals and petals 4 each, free, stamens tet radynamous, ovary with false septum and a thickened placental rim called replum, ovary superior, placentation parietal, fruit a siliqua or silicula. width), at dehiscence valves break away from below upward leaving seeds appresse d to false septum, fruit moniliform lomentum on Raphanus; seed with large embryo , endosperm scant or absent. Pollination by insects, failure of cross pollinatio n may result in self pollination. Seeds are usually dispersed by wind. Major Genera: Draba (350 species), Erysimum (180), Lepidium (170), Cardamine (160), Arabis (160), Alyssum (150), Sisymbr ium (90) and Brassica (50). Economic importance: The family contributes a number of food plants such as radi sh (Raphanus sativus), cabbage (Brassica oleracea var. capitata), cauliflower (B . oleracea var. botrytis), Brussels sprouts (B. oleracea var. gemmifera), kohlra bi (B. oleracea var. caulorapa) and turnip (B. rapa). Seeds of B. campestris yie ld cooking oil those of black mustard (B. nigra) are used as condiment. Woad was formerly used a blue dye obtained from leaves of Isatis tinctoria. Common ornam entals include stock (Mathiola), candy tuft (Iberis amara), alyssum (Alyssum), w all flower (Erysimum) and sweet alyssum (Lobularia). Description. Annual, biennial or perennial herbs (rarely undershrubs: Farsetia) with watery sap, containing glucosinolates (mus-tard oils) and with myrosin cell s. Hairs sim-ple, branched, stellate or peltate. Leaves al-ternate or in basal r osettes, simple, often dissected, rarely pinnate compound (Nastur-tium officinal e) sometimes bearing bulbils in axil (Dentaria bulbifera) or leaf surface (Carda mine pratensis), stipules absent. Inflo-rescence typically racemose, corymbose r a-ceme, or flat topped corymb (Iberis), Cardamine also produces subterranean cle istogamous flowers. Flowers ebracteate, rarely bracteate (Nasturtium montanum), bi-sexual, actinomorphic or rarely zygomorphic (Iberis), hypogynous (perigynous in Lepidium). Calyx with 4 sepals, free, in two whorls, se-pals of lateral pair sometimes saccate at base, green or slightly petaloid. Corolla with 4 petals, cr uciform (arranged in a cross), clawed, sometimes absent in Coronopus and Lepidiu m. Androecium with 6 stamens (2 in Coronopus, 4 in Cardamine hirsuta, 16 in Mega

carpaea), free, tetradynamous (2 short 4 long), dehiscence longitudinal, nectari es often present near base of stamens, pollen grains tricolporate or tricolpate. Gynoecium with two united (thus pistil single) carpels (syncarpous), rarely car pels 3 (Lepidium) or 4 (Tetrapoma), unilocular but becoming bilocular due to fal se septum that is sur-rounded by a thick placental rim called replum, ovules man y, rarely single ovules, placentation parietal, ovary superior, gynophore distin ct, style 1, stigmas 2. Fruit a siliqua (long: length thrice width or more) or s ilicula (short: length less than thrice Phylogeny: The family is regarded as mono-phyletic, supported by evidence from m orphology ( gynophore, exserted stamens), glucosinolates, dilated cisternae in e ndoplasmic reticulum and rbcL sequences. The or-der Brassicales (others prefer C apparales) had long been treated as a well defined group, with Brassicaceae and Capparaceae considered to be fairly close as suggested by evidence from morphology, dilated cister nae, but have been treated as distinct largely because of several stamens and ve ry long gynophore in Capparaceae. Judd et al., (1994) on the basis of morphological studies, and Rodman et al., (1 993) on the basis rbcL sequences, concluded that out of the traditional Capparac eae, Capparoideae and Cleomoideae do not form a monophyletic group, as also conc luded ear-lier by Hutchinson (1973). Capparoideae according to these authors for m basal paraphyletic group within Brassicaceae. Cleomoideae and Brassicoideae (t raditional Brassicaceae) form monophyletic group based on synapomorphies of herb aceous Major Families of Angiosperms 603 character, replum in fruit and rbcL se-quences. The merger of Capparaceae with B rassicaceae avoids arbitrarily delimited paraphyletic taxa, and thus forms monophyletic group with broadened circumscrip-tion. The two have been merged in APG II and APweb classifications. APweb recog-nizes 3 subfamilies under broadly circ um-scribed Brassicaceae: Capparoideae, Cleomoideae and Brassicoideae. It is pert iSoltis et al., (2000) and Hall et al., (2002), Brassicaceae (Brassicoideae) and Cleomaceae (Cleomoideae) are more closely related and form a monophyletic group based on synapomorphies of herbaceous habit, rbcL sequences and presence of replum. It is interesting to record that although Hutchinson had indicated heterogeneity within Capparaceae, and reasoned that nent to note that Thorne (1999), who has Cleome and its relatives were much clos er been updating his classification in light of to Brassicaceae, he had placed C apparaceae recent advances, has preferred to retain and Brassicaceae in two dist inct orders CapBrassicaceae and Capparaceae as distinct parales (in his diagram he used name Cap-families, also separating Cleomaceae in paridales) and Brassica les, even further recent revisions (2003, 2007), thus recog- separating them und er Lignosae and Herbanizing three subfamilies as independent ceae respectively, as he was obsessed with the families. According to the recent studies of distinc tion of woody and herbaceous habits. * * * * * * * * * * * Rutaceae A. L. de Jussieu Citrus or Rue family 162 genera, 1,650 species Distributed in warm temperate and tropical regions with the greatest diversity i n Australia and South Africa. Placement:

B & H

Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Malvidae Series+/Superorder Disciflorae+ Rutanae Rutanae Rutanae Eurosids II* Order Geraniales Sapindales Rutales Rutales Rutales Sapindales Salient features: Trees or shrubs, leaves usually compound and gland dotted, sta mens free or polyadelphous, sometimes obdiplostemonous, ovary superior, seated o n a nectary disc, fruit a berry. Description. Shrubs or trees, sometimes armed with thorns or spines, rarely herb s (Boenninghausenia), often aromatic, containing alkaloids and phenolic compound s. Leaves alternate, rarely opposite (Evodia), usually pinnate compound, sometim es unifoliate due to reduction of lower two leaflets (Citrus), less frequently s imple (Evodia), gland dotted, stipules absent. Inflorescence Major Genera: Zanthoxylum (200 species), Agathosma (180), Citrus (65), Ruta (60) and Murraya (12). 604 Plant Systematics

Ia ! MAN /

V E A

Figure 13.100 Rutaceae. Murraya paniculata. A: Branch with pinnate compound leav es and flowers; B: Vertical section of flower; C: Flower with petals removed showing 10 sta mens in 2 whorls; D: Gynoecium with nectary at base. Citrus paradisi. E: Portion of branch with unifoliate leaf having broadly winged petiole and cluster of flo wers with polyadelphous stamens; F: Fruit. Zanthoxylum armatum. G: Branch with p innate compound leaves, spines and inflorescences; H: Male flower with 6 free st amens and abortive ovary, the sepals are small and petals absent; I: Schizocarpi c fruit splitting into 2 segments. Haplophyllum acutifolium. J: Portion of plant with flowers; K: Flower with large petals, stamens with flattened filaments; L: Capsule covered with glands and deeply 5-lobed. cymose or flowers solitary (Triphasia), rarely racemose (Atlantia). Flowers ebra cteate, bi-sexual or rarely unisexual (Zanthoxylum) actinomorphic or rarely zygo morphic (Dictamus), hypogynous. Calyx with 4-5 se-pals, rarely 3 (Lunasia) free or more or less united, gland dotted. Corolla with 4-5 pet-als, rarely 3 (Tripha sia) free, rarely united (Correa) valvate or imbricate, sometimes absent. Androe cium with 8-10 rarely 5 (Skimmia), or many (Citrus) stamens, free (Murraya), or polyadelphous (Citrus), rarely monadelphous (Atlantia) sometimes obdiplostemonou s, anthers bithecous, dehiscence longitudinal, pollen grains 3-6-colpate. Gynoec ium with 2-5 united carpels (syncarpous), rare monocarpellary (Teclea), sometimes ovaries free (Zanthoxylum) and only styles united, multilocular (locul es as many as carpels) with 1-many ovules, placentation axile, rarely parietal ( Feronia), ovary superior and lobed, style 1, stigma small. Fruit a berry (Murray a), drupe (Spathelia), hesperidium (Citrus), samara (Ptelea), capsule (Ruta) or follicle (Zanthoxylum); seeds 1-many, embryo curved or straight, endosperm absen t or present. Pollination mainly by insects, chiefly bees and flies. Dispersal u sually by animals, rarely birds or even wind (Ptelea). Economic importance: The family is important for its citrus fruits such as lemon (Citrus limon), lime (C. aurantfiolia), sweet orange Major Families of Angiosperms 605 or mousmi (C. sinensis), orange, tangerine or santra (C. reticulata) and grapefruit (C. paradisi). Aegle marmelos (Bel tree), Fortunella (kumquat), and Casimiroa (w hite zapote) are grown for fruits. Murraya paniculata is cultiported by data from rbcL and atpB sequences. Subfamilies are characterized by ca rpel number, extent of fusion and fruit type. Whereas Dahlgren (1989) and Takhta jan (1997) preferred to separate order Sapindales from order Rutales, others lik e Cronquist vated as an ornamental shrub, whereas M. koenigii is cultivated for its curry leaves. (1988), Thorne (1999, 2003, 2007 ), and APG Leaves of Skimmia laureola are burnt in II prefer to merge the two. C ronquist and order to purify air. Ruta (rue), Zanthoxylum APG II use name Sapind ales whereas (toothache tree), and Casimiroa are medici- Thorne prefers the prio rity name Rutales. nal. Boenninghausenia ( Pisu-mar-buti ) is Thorne, however, place s Sapindaceae and used as an insecticidal. Ravenia spectabilis Rutaceae under se parate suborders. Earlier (Syn: Limonia spectabilis) is grown as orna- Hutchinso n (1973) had also separated the mental shrub. Phylogeny: Although the family presents a variety of fruit types, it is a well c ircum-scribed monophyletic taxon characterized by oil cavities appearing as pell ucid dots, suptwo orders, also separating Meliaceae under Meliales, and placing between the tw o orders. The affinities of Meliaceae, Rutaceae and Sapindaceae have been long r ecognized. Thorne (2007) recognizes 3 subfamilies: Rutoideae, Aurantioideae and Cneoroideae. Meliaceae A. L. de Jussieu Mahogany family 52 genera, 600 species

Distributed mainly in tropical and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Malvidae Series+/Superorder Disciflorae+ Rutanae Rutanae Rutanae Eurosids II* Order Geraniales Sapindales Rutales Rutales Rutales Sapindales Salient features: Trees or shrubs contain-ing bitter triterpenoid compounds, lea ves alternate, pinnate compound, flowers uni-sexual, sepals 4-5, petals 4-5, sta mens with connate filaments, ovary with axile placentation, stigma capitate, see ds dry and winged. Major genera: Aglaia (95 species), Trichilia (60), Turraea (60), Dysoxylum (58), Guarea (32), Toona (15), Melia (15), Cedrela (6) and Azadirachta (2). Description: Shrubs or trees, rarely herbs (Naregamia), commonly producing bitte r 606 Plant Systematics

I r. ~jit

rod Figure 13.101 Meliaceae. Melia azedarach. A: Bipinnate leaf; B: Inflorescence; C : Flower with long staminal tube; D: Drupe. Aglaia apiocarpa. E: Fruiting branch with pinnate compound leaf; F: Portion of male inflorescence; G: Female inflorescence; H: Ver tical section of male flower; I: Vertical section of female flower. Dysoxylum ch ampionii. J: Flowering branch; K: Vertical section of male flower; L: Tetrad of pollen grains; M: Transverse section of ovary. triterpenoid compounds, usually with scat-tered secretary cells, wood sometimes yel-low (Chloroxylon) or red (Cedrela), nodes pentalacunar, vessels with simple end-walls. Leaves alternate, once (Azadirachta) or twice pinnate (Melia), someti mes trifoliate (Sandoricum) or unifoliate (Turraea), venation pinnate, reticulat e, stipules absent. Inflo-rescence axillary or terminal panicles, usu-ally cymos e. Flowers ebracteate, bisexual or rarely unisexual (Amoora), actinomorphic, tri merous to pentamerous, cyclic. Calyx with 3 (Amoora), 4 (Dysoxylum) or 5 (Melia, Cedrela) sepals, free or united (Amoora, Melia), valvate or imbricate, green. C orolla with 3-5 petals, usually as many as sepals, free, rarely united (Munronia ), imbricate or valvate. Androecium with 3 (some species of Amoora), 4-6 (Cedrela), 5 (Aglaia) o r upto 12 (Melia), free (Walsura) or monadelphous (Melia), usually inserted on a nectariferous disc, anthers bithecous, dorsifixed or versa-tile, introrse, dehi scence longitudinal, pollen grains 2- to 5-colporate, sometimes in tetrads (Dyso xylum championii). Gynoecium with 2-6 united carpels, ovary superior, 2-5 chambe red with 1 or 2 (or more in Swietenia) ovules in each loculus, ovule orthotropou s or anatropous (Dysoxylum), placentation ax-ile, style 1, stigma capitate. Frui t a drupe, (Melia azedarach), berry (Walsura), or a cap-sule (Amoora); seeds win ged or with an aril (Melioideae), embryo curved or straight, en-dosperm present or absent. Major Families of Angiosperms 607 Economic importance: The family is highly prized for its true mahogany woods: Sw ietenia mahogani of the West Indies; Entandrophragm, Khaya and Lovoa of Africa; Cedrela odorata and Toona of Australia. These or Sapindales (Cronquist, APG II, APweb) or narrowly circumscribed Rutales (Takh tajan, Dahlgren). Thorne (1999,2003) combines Rutales with Sapindales but prefer s name Rutales for the broadly circumscribed order. Hutchinson (1926, 1973) segr egated the famare renowned for their excellent colour, working properties and finish. Oils for soap- ily to a distinct order Meliales p rimarily on making are extracted from the seeds of the basis of leaves usually n ot gland-dotted Trichilia emetica in Uganda. The oil from and stamens connate. T he family is distinct Malayan Chisocheton macrophyllus has been and monophyletic as supported by morpholused as an illuminant. The flowers of Aglaia ogy and rbc L sequences (Gadek et al., 1996). odorata are used in the East for flavouring Tw o subfamilies are commonly recognized: tea. Species of Melia, Aglaia, Chisocheto n and Melioideae (seeds not winged, naked buds) Turraea are grown as ornamentals . and Swietenioideae (flattened or winged Azadirachta indica (neem) from India h as seeds, and scaly buds. Thorne had earlier gained considerable importance in t he re- (1999) merged these two under Melioideae cent years as a bioinsecticide a nd a compo- and recognized two more Quivisianthoideae nent of several medicines and also tooth (single genus Quivisiantha) and Capuronipastes. The tree, for a l ong time, has been anthoideae (single genus Capuronianthus). grown as a shade tree and twigs used to In subsequent revision (2003) he resurbrush teeth (datun). rected Swietenioideae, thus recognizing four subfamilies under the family. The g enus Cedrela is distinct in its free stamens and erect petals, included by thorn

e under Swietenioideae. Phylogeny: The family is generally included closer to Rutaceae in broadly circum scribed Geraniales (Bentham and Hooker, Bessey) * * * * * * * * * * * Anacardiaceae R. Brown Cashew family 70 genera, 845 species Distributed mainly in tropics but several species extending to north temperate r egions of Asia, Europe and America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Malvidae Series+/Superorder Disciflorae+ Rutanae Rutanae Rutanae Eurosids II* Order Sapindales Sapindales Burserales Sapindales Rutales Sapindales 608 Plant Systematics

G Figure 13.102 Anacardiaceae. Rhus wallichii. A: Branch with inflorescence; B: Po rtion of inflorescence in fruit. Mangifera indica C: Branch with inflorescence; D: Flower; E: Vertical section of flower showing conspicuous disc; F: Single stamen. G: Longit udinal section of fruit, reduced view. H: Fruit of Anacardium occidentale. (A-B, af ter Polunin and Stainton, Fl. Himal., 1984). Salient features: Trees or shrubs with resin canals, alternate exstipulate leave s, flowers pentamerous, stamens inserted at the base of a disc, ovary unilocular , ovule one, fruit commonly a drupe. tissue; resin clear when fresh but drying black and often causing dermatitis (To

xicodendron, some species of Rhus). Leaves alternate, rarely opposite (Dobinea), usually pinnate compound (Rhus, Schinus), rarely simple (Mangifera), entire or serrate, venation pinnate, stipules absent or vestigial. Inflorescence paniculat e, axillary or terminal, usually thyrse with cymose branches. Flowers bracteate, bisexual or unisexual due to reduction of stamens or carpels, receptacle often swollen and fleshy, actinomorphic, small, hypogynous. Calyx with 5 sepals Major genera: Rhus (100 species), semecarpus (50), Lannea (40), Toxicodendron (3 0), Schinus (30) and Mangifera. Description: Trees, shrubs or lianas, rarely perennial herbs, resin canals in ba rk, big-ger veins of leaves and in parenchymatous Major Families of Angiosperms 609 (rarely 3-7), free or connate at base, imbri-cate, rarely much enlarged (Parishi a). Co-rolla with 5 petals (rarely 3-7), free, imbri-cate, rarely much enlarged (Swintonia). Androecium with 5-10 stamens, sometimes more or reduced to single f ertile stamen (Mangifera, Anacardium), filaments usually glabrous, distinct, rar ely connate at base, arising along outer (most genera) or inner margin (Mangifer a) of rim of the disc, dehis-cence longitudinal, pollen grains tricolporate or t riporate. Gynoecium with usually 3 car-pels, sometimes 5 (Buchanania) or only 1 (Mangifera), united, rarely free (Buchanania), ovary superior, locule usually 1 with basal or apical placentation, rarely multilocular with 1 ovule in each locu le, styles 1-3, free or connate, stigma capitate, disc present between stamens a nd petals. Fruit a drupe with resinous mesocarp, rarely berry or nut (Anacardium ), or samara (Loxopterygium); seed with straight or curved embryo, absent or sca nty. Pollination by insects, dioecious habit promoting outcrossing. Fruits are d is-persed by birds and mammals. of Cotinus (smoke tree), Rhus (Sumac) and Schinus (Brazilian pepper). Commercial supply of tannins is obtained from quebracho (Schinopsis lorentzii), Sicilian s umac (Rhus coriaria) and species of Cotinus and Pistacia. The first turpentine u sed by artists came from terebinth tree (Pistacia terebinthus). Phylogeny: Anacardiaceae is closely related to Burseraceae (Thorne, 2006 places two families separately under suborder Anarcardiineae) as supported by rbcL sequ ences (Gadek et al., 1996). The two share resin canals and biflavones. The famil y is usually placed under Sapindales, but shifted by Thorne under Rutales. Two c lades (subfamilies) are recognised within the family (Thorne; ApWeb): Anacardioi deae and Spondoideae, latter having retained many plesiomorphic features such as five carpels, multilocular ovary and fruit with thick endocarp having lignified and irregularly oriented sclereids, and sister to rest of the family i.e., Anac ardioideae (Aguilar-Ortigosa & Sosa 2004). Anacardioideae clade has 3 carpels, u nilocular ovary with apical placenta and endocarp with regularly arranged cells. wind-pollinated taxa of this subfamily, however, do not form a single group (Pe ll & Mitchell 2007). Buchanania in some analyses is quite well supported as sist er to Anacardioideae (Aguilar-Ortigosa & Sosa 2004; Wannan, 2007), consistent wi th its chemistry, endocarp anatomy and carpel number. Phylogeny of the family ha s been studied by Pell (2004) who covers the morphology of the whole family and Mitchell et al. (2006), who focus more on Spondoideae. Rhus and Toxicodendron, a re distinct in former having red glandular-pubescent fruits and latter glabrous greenish or white ones. These are often combined, but the resultant genus won t be monophyletic. Economic importance: The family contrib-utes important fruits such as mango (Man gifera indica), yellow mombin or Hog plum (Spondias mombin), Indian Hog plum (S. indica), Red mombin or Jamaica apple (S. purpurea), Otaheite apple (S. cytherea ) and Kaffir plum (Harpephyllum caffrum). Nuts of Cashew (Anacardium occidentale ) and Pista-chio (Pistacia vera) are eaten after roasting. Species of Rhus and T oxicodendron cause der-matitis due to the phenolic compound 3-n-pentadecycatecho l and should be touched with care. Lacquer is obtained from Varnish tree (Toxico dendron vernicifera) and Mastic tree (Pistacia lentiscus). Important ornamentals are contributed by the species * * * * * * * * * * * 610 Plant Systematics

Sapindaceae A. L. de Jussieu Soapberry family 145 genera 1490 species Distributed mainly in tropical and subtropical regions, a few genera in temperat e regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Malvidae Series+/Superorder Disciflorae+ Rutanae Rutanae Rutanae Eurosids II* Order Sapindales Sapindales Sapindales Sapindales Rutales Sapindales Salient features: Usually trees or shrubs, leaves alternate, pinnately-compound or palmately-compound, strongly swollen peti-ole base, inflorescences paniculate , flowers often borne in congested groups along the axis, flowers small and cons picuously hairy inside, nectar disc between petals and sta-mens, fruit with only one or two seeds in each chamber, often deeply lobed, seeds with aril zygomorphic (Cardiospermum). Calyx with usually 4 or 5 sepals, free or united. C orolla with 4 or 5 petals, usually as many as sepals, sometimes absent (Dodonaea ), free, usually clawed, with basal appendages inside, imbricate. Androecium wit h usually 4 (Glenniea) 10 stamens, rarely more (Deinbollia), usually borne on a nectar disc present between pet-als and stamens, filaments free and usually hair y, anthers bithecous, dehiscence by longitudinal slits, pollen grains tricolpate , fur-rows often fused. Gynoecium with 2-3 united carpels, rarely upto 6, ovary superior, placentation axile, 1 or 2 ovules in each loculus, ovule orthotropous or anatropous, funiculus lacking and ovule broadly attached to obturator (projec tion from placenta), style 1, stigma usually lobed. Fruit a drupe, berry, capsul e (3-winged-Bridgesia), samara (Acer) or schizocarp; seeds often with an aril, e mbryo curved, endosperm absent. Pollination by birds and insects, Dodonaea and s ome species of Acer being wind pollinated. Dispersal often by birds attracted by

aril, but inflated fruits and winged fruits are often dispersed by wind. Major genera: Serjania (200 species), Paullinia (140), Acer (100), Allophyllus ( 95), Dodonaea (60), Sapindus (18), Aesculus (13), Cardiospermum (12), Koelreuter ia (10) and Litchi (2). Description: Shrubs or trees, herbaceous or woody lianas with tendrils (Serjania ), rarely herbs (Cardiospermum), often with tannins, usually with triterpenoid s aponins in secretary cells. Leaves alternate, rarely opposite (Velenzuelia, Acer), once or twice pinnate, sometimes palmately compound (Aesculus) trifoliate (Billia) or simple (Litchi), leaflets entire or serrate, venation pinnate, reticu-late, base of pet iole strongly swollen, stipules absent, rarely present (Urvillea, Serjania). Inf lorescence usually cymose, aggregated in panicles, often quite congested along t he axis. Flowers unisexual (plants monoecious, dioecious or polygamous), actinom orphic or Economic importance: The family is important primarily for its fruits: Litchi si nensis (Litchi, lychy), Nephelium lappaceum (rambutan) and Euphoria (logan). The aril of the fruits of Blightia sapida (akee), a native Major Families of Angiosperms 611 rn ~ n

f K E G I B F Figure 13.103 Sapindaceae. Dodonaea viscosa. A: Flowering twig of female plant; B: Flowering twig of male plant; C: Male flower with 5 sepals and 8 stamens, petals absent; D : Female flower with sepals and gynoecium; E: Gynoecium. F: Fruit with with 2 wi ngs. Acer caesium. G: Portion of a fruiting branch; H: Portion of flowering bran ch; I: Flower. Aesculus indica. J: Portion of flowering branch with palmately co mpound leaf; K: Flower with long-clawed petals and exserted stamens; L: Capsule opening by 3 valves.(G-I after Fl. Himal., 1984) of West Africa are also eaten cooked, tast-ing like scrambled egg, but poisonous if eaten when unripe. The fruits of different species of Sapindus (soapberry) a re often used as a Acer saccharum (sugar maple) and some other species yield maple sugar. natural soap due to the presence of Phylogeny: Sapindaceae is sometimes narrowly circumscribed to exclude Hipposaponins. Paullinia cupana is the source of castanaceae and Acerceae (Hutchinson , drink guarana, popular in Brazil. Schleichera Takhtajan, Cronquist and Dahlgre n), but trjiuga is the source of macassar oil, used in their separation leads to paraphyletic ointments and for illumination. The species Sapindaceae (Judd et a l., 1994). The family of Aesculus have various medicinal uses, and is as such br oadly circumscribed to include extracts from some have been used by North both A ceraceae and Hippocastanaceae American Indians to stupefy fish. The fam- (Thorne

, Judd et al., APG II and APweb). ily also contributes a number of ornamentals M onophyly of the family is supported by such as Koelreuteria (goldenrain tree), m orphology and rbcL sequences. XanthoCardiospermum (balloon vine), Xanthoceras, c eras, with simply 5-merous, polysymmetric Acer (maple) and Aesculus (horse chest nut). flowers and complex, golden nectaries borne The trees of maple are prized for their beau- outside the eight stamens, is sister to rest tiful foliage and s pectacular autumn colours. of Sapindaceae, and the genera included in 612 Plant Systematics Aceraceae and Hippocastanaceae are monophyletic sister taxa (Savolainen et al., 2000). Sapindaceae are chemically similar to Leguminosae, and both have compound leaves, but they are unlikely to be immediately related. Thorne (1999, 2003) recognizes 5 subfamilies: Dodonaeoideae (Dodonaea), Koelreuterioideae, Sapindoideae, Hippocastanoideae (Aesculus, Billi a) and Aceroideae (Acer, Dipteronia). * * * * * * * * * * * Subclass 11. Asteridae (B) Superorder 1. Cornanae (B) 4. Samolaceae Order 1. Bruniales (B) 5. Primulaceae Family 1. Bruniaceae 6. Ericales 2. Geissolomataceae 3. Grubbiacerae 4. Curtisiaceae 5. Myrothamnaceae (B) 6. Gunneraceae 1. Theaceae 2. Pentaphylacaceae 3. Ternstroemiaceae 4. Sladeniaceae 5. Symplocaceae 7. Hydrostachyaceae (B) 6. Styracaceae (B) 2. Cornales 7. Diapensiaceae 1. Hydrangeaceae 8. Sarraceniaceae 2. Loasaceae 9. Roridulaceae (A) 3. Alangiaceae 10. Actinidiaceae 4. Cornaceae 11. Clethraceae 5. Nyssaceae 12. Cyrillaceae 2. Desfontainiales 13. Ericaceae 1. Escalloniaceae Superorder 3. Aralianae 2. Eremosynaceae Order 1. Aquifoliales (A) 3. Vahliaceae 1. Aquifoliaceae 2. Helwingiaceae 3. Phyllonomaceae (B) 4. Cardiopteridaceae (B) 5. Stemonuraceae (B) 6. Sphenostemonaceae (B)

4. Columelliaceae 5. Desfontainiaceae Superorder 2. Ericanae (B) Order 1. Balsaminales 1. Marcgraviaceae 2. Balsaminaceae 3. Tetrameristaceae (B) 7. Paracryphiaceae (B) 2. Dipsacales 2. Polemoniales 1. Adoxaceae 2. Caprifoliaceae 1. Fouquieriaceae 3. Diervillaceae (B) 2. Polemoniaceae 4. Linnaeaceae (B) 3. Lecythidales 5. Morinaceae 1. Lecythidaceae 6. Dipsacaceae 4. Sapotales 7. Triplostegiaceae (B) 1. Ebenaceae 8. Valerianaceae 2. Sapotaceae 3. Araliales 5. Primulales Suborder 1. Aralidiineae 1. Maesaceae 1. Torricelliaceae 2. Theophrastaceae 2. Aralidiaceae 3. Myrsinaceae 3. Melanophyllaceae (B) Major Families of Angiosperms 613 4. Griseliniaceae (B) 2. Campanulaceae 5. Pennantiaceae (B) 3. Stylidiaceae 2. Apiineae 2. Asterineae 1.Pittosporaceae 1. Rousseaceae (B) 2. Alseuosmiaceae 3. Argophyllaceae (B) 4. Phellinaceae (B) 5. Menyanthaceae 6. Goodeniaceae 7. Calyceraceae 8. Asteraceae 2. Apiaceae 3. Araliaceae 4. Myodocarpaceae (B) Superorder 4. Asteranae Order 1. Asterales 1. Campanulineae 1. Pentaphragmataceae Hydrangeaceae Dumortier Hydrangea family

17 genera, 250 species Mainly distributed in Northern Hemisphere from Himalayas to Japan to North Ameri ca and tropical Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Cornidae Magnoliidae Asteridae Series+/Superorder Calyciflorae+ Cornanae Cornanae Cornanae Asterids* Order Rosales Rosales Hydrangeales Cornales Cornales Cornales B & H under Saxiragaceae Salient features: Mostly shrubs, leaves sim-ple, usually opposite, stipules abse nt, flow-ers bisexual, sepals often enlarged and petaloid, ovary inferior or sem i-inferior, placentation axile or deeply intruded pari-etal, nectary present at top of ovary, fruit a capsule. Iridoids and raphide crystals. Leaves usually opposite, rarely alternate (Cordia ndra), simple, sometimes lobed, usually deciduous, rarely evergreen (Pileostegia ), venation pin-nate or palmate, reticulate, stipules absent. Inflorescence term inal racemes, cymes or corymbose, rarely solitary flowers. Flowers usually bisex ual, sometimes unisexual (Broussaisia, polygamo-dioecious) outer usually sterile with enlarged petaloid sepals, actinomorphic, perigynous or epigynous. Calyx wi th 4-5 sepals , united, calyx tube often adnate to ovary, sepals of outer steril e flowers often large and petaloid. Corolla with 4-5 petals, free, imbricate, co nvolute, rarely Major genera: Philadelphus (65 species), Deutzia (40), Hydrangea (30), Dichroa ( 13) and Fendlera (4). Description: Herbs (Cordiandra), soft-wooded shrubs (Hydrangea) or rarely small trees or climbers (Decumeria), often with tannins and

614 Plant Systematics

/I

C P E F A D Figure 13.104 Hydrangeaceae. Hydrangea heteromalla. A: Branch with corymbose inf lorescence with outer sterile flowers having 4 enlarged petaloid sepals. Kirengeshoma palma ta. B: Terminal part of flowering branch; C: Petal with adnate stamens; D: Flowe r with stamens and petals removed; E: Stamens with dorsifixed anthers; F: Transv erse section of ovary. (B-F after Hutchinson, 1973) valvate (Platycrater), usually white. Androecium with many stamens, some-times 8 -10 (Hydrangea), free or slightly connate at base, anthers bithecous, basifixed or dorsifixed, filaments often lobed or toothed, connective sometimes app-endage d at tip (Fendlera), pollen grains tricolpate or triporate. Gynoecium with 2-7 u nited carpels, ovary semi-inferior (Dichroa, Broussaisia), inferior (Philadelphu s, Deutzia, Hydrangea) or superior (Jamesia), 1-7 locu-lar, ovules numerous, pla centation axile or parietal with deeply intruded placentas, styles free, rarely united (Carpenteria), stigmas free, nectar disc usually pres ent at top of ovary. Fruit usually a loculicidal (Hydrangeae) or septicidal (Phi ladelpheae) capsule, rarely a berry (Dichroa); seeds nu-merous, small, sometimes winged, with fleshy endosperm and straight embryo. Pollination by insects aided by epigynous disc. Small seeds are dispersed by wind. Economic importance: The family is known for ornamental shrubs with showy flower s including Hydrangea, Decumeria (climbing Major Families of Angiosperms 615

rr.

~ n L~

r L`ZI ~

i' ~~. ~ art ' ~ a~~

Figure 13.105 Malvaceae. A: Hibiscus rosasinensis, flowering branch; B: stamens and stigmas; C: Malvaviscus arboreus, flowering branch; D: Flower; E: Lavatera assurgentiflor a, flowering branch; F: Flower; G: Gossypium hirsutum, flowering branch; H: Flow er; I: Fruit. Rhamnaceae. J: Colletia paradoxa, plant. Moraceae. K: Morus alba, plant with male catkins; L: Male catkin enlarged; M: Plant with female inflresce nces. N: Ficus religiosa, twig with young hypanthodia. 616 Plant Systematics RWk,

FM

:1

i~ Figure 13.106 Rafflesiaceae. A: Rafflesia speciosa, flower.Brassicaceae. B: Coro nopus didymus, plant; C: Cheiranthus cheiri, plant; D: Mathiola incana, plant; E: Brassica oler acea, flowering branch; F: Fruits; G: B. campestris, flowering branch; H: Fruits ; I: Flowers; J: Iberis amara, flowering branch; K: Cakile maritima, flowering b ranch. Major Families of Angiosperms 617 t~7l34 P.

LAA

r.. VP ~ k [;rli

DIVAN IVI

Figure 13.107 Rutaceae. A: Citrus limon, flowering branch; B: Fruit; C: Murraya paniculata, flowering branch; D: Flower. E: Citrus medica, flower; F: Ravenia spectabilis, plant i n flower; G: Flower. H: Murraya koenigii, flowering branch; I: Flowers. Anacardi aceae. J: Mangfiera indica, flowering branch; K: Portion of inlforescence enlarg ed; L: Female flower. Meliaceae. M: Melia azedarach, flowering branch. N: Flower ; O: Fruits. 618 Plant Systematics

Figure 13.108 Sapindaceae. A: Acer japonicum, branch; B: A. griseum, bark; C: Fr uit; D: Aesculus californica, flowering branch; E: Flowers; F: Ungnadia speciosa , fruiting branch; G: Fruit. Major Families of Angiosperms 619 yu 1 D

T Y4

Figure 13.109 Hydrangeaceae. A: Hydrangea macrophylla, flowering branch; B: Flow ers. Polemoniaceae. C: Phlox diffusa, flowering branch; D: Flowers. Cornaceae. E: Cor nus capitata, flowering branch; F: Flower. Primulaceae. G: Primula florindae, in florescence; H: Dodecantheon meadia, inflorescence; I: Flower. 620 Plant Systematics

U Figure 13.110 Ericaceae. A: Colluna vulgaris, flowering branch; B: Erica blanda, flowering branch; C: Arbutus unedo, flowering branch; D: Fruit; E: Rhododendron giersonianum, flow er cluster; F: R. occidentale, flowering branch. G: Phyllodoce breweri, flowerin g branch; H: Adoxaceae. E: Sambucus nigra, flowering branch; I: Viburnum cotinif olium, branch with young fruits; J: Branch with mature fruits. Major Families of Angiosperms 621

0,"Al1. -

10-11

w LWJ

v VXA-'~ih 0 R ~'ffAb W -1-M-71

Figure 13.111 Apiaceae. A: Astrantia major, plant; B: Umbel; C: Eryngium panicul atum, plant with inflorescences; D: Angelica pachycarpa, inflorescence; E: Foeniculum vulgare, pl ant; F: Part of inflorescence; G: Coriandrum sativum, plant; H: Part of inflores cence. Araliaceae. I: Pseudopanax crassifolium, flowering branch; J: Inflorescen ce; K: Hedera helix, plant. 622 Plant Systematics

i , e Figure 13.112 Asteraceae. A: Viguieria helianthoides, plant; B: Capitulum; C: Ac hillea millefolium, plant; D: Taraxacum officinale, plant with capitula; E: Pachystegia insignis, pl ant with capitula; F: Haplopappus macrocephalus, plant with capitula and fruitin g heads; G: Centaurea solstitialis, plant with inflorescence; H: Artemisia pycno cephala, plant; I: Calendula officinalis, capitulum; J: Sonchus oleraceous, plan t. Major Families of Angiosperms 623

li

~--NrAet

Figure 13.113 Solanaceae. A: Cestrum elegans, flowering branch; B: flowers; C: S olanum hispidum, flowering branch; D: Atropa belladonna, flowering branch; E: Datura suaveolens, flowering plant; F: Solanum melanogena, flowering branch; G: Flower; H: Fruit; I : Solanum nigrum, plant in flower; J: Young fruits. Convolvulaceae. K: Ipomoea c airica, plant in flower; L: Flower; M: Jacquemontia pentantha, plant in flower; N: Flowers. Boraginaceae. O: Ehretia laevis, plant; P: Flowers. 624 Plant Systematics

a C

Figure 13.114 Rubiaceae. A: Ixora fulgens, flowering branch; B: Inflorescence C: Hamelia patens, flowering branch; D: Flowers and young fruits; Apocynaceae E: Plumeria alba, flo wering branch; F: Allamanda catharatica, branch with flowers; G: Catharanthus ro seus, flowering branch; H: Flower; I: Nerium oleander, flowering branch; J: Flow er; K: Asclepias syriaca, flowering plant; L: A. fascicularis, flowering branch; M: Flowers; N: Calotropis procera, flowering branch.. Major Families of Angiosperms 625 ,{o ~ Ih Poe

i d4E, Figure 13.115 Plantaginaceae. A: Plantago lanceolata, plant; imulus cardinalis, plant; C: M. guttatus, flower cluster; D: M. puniceus, flowering branch; E: Digitalis p urpurea, flowering branch; F: Flowers enlarged. Lamiaceae. G: Salvia muelleri, f lower; H: S. mexicana, flowering branch; I: Flower; J: Origanum calcaratum, flow

ers; K: Salvia splendens, plant; L: Flower; M: S. scorodonifolia, flower; N: Lav andula angustifolia, flowering twig. 626 Plant Systematics

Ly ~I ~ 8 s Figure 13.116 Verbenaceae. A: Lantana camara, plant; B: Flowers. Bignoniaceae. C : Incarvillea arguta, flowering branch; D: Tecoma stans, flowering branch. Acanthaceae. E: Jus ticia brandegeana, plant; F: Acanthus spinosus, flowering branch; G: Flowers; H: Adhatoda vasica, flowering branch; I: Flowers; J: Thunbergia grandiflora, flowe r. Scrophulariaceae. K: Bowkeria gerardiana, flowering branch; L: Part of inflor escence; M: Vebascum thapsus, plant. Major Families of Angiosperms 627 hydrangea), Schizophragma (climbing hydran-gea), Philadelphus (mock orange) and Deutzia. Some species of Hydrangea are used as source of hydrangin, a compound u sed in medicine. rate subclass Cornidae (superorder Cornanae, order Hydrangeales), while Saxifrag aceae was retained in Rosidae. The recent classifications of APG II, APweb and T horne (2003, 2007) place family under Asterid complex (Thorne in Asteridae >Cornan ae >Cornales; APG classifications in Asterids >Cornales). The separation of Hydrange aceae away from Saxifragaceae has been supported by cladistic analysis, indicati ng that the two are distantly related, Phylogeny: The family was earlier included under Saxifragaceae (Bentham and Hook

er; Engler and Prantl). It was separated as dis-tinct family by Hutchinson (1927 ), who also treated Philadelphaceae as distinct family in his The Genera of Flow ering Plants (1964) followed up in his last revision of classifica-tion (1973). Philadelphaceae was merged with Hydrangeaceae in classifications of Cronquist (1 988) retaining the family in Rosales (placed under Rosidae) along with Saxifraga ceae. The merger of Philadelph-aceae with Hydrangea-ceae is supported by morphol ogy and DNA characteristics (Albach et al, 2001; Soltis et al., 1995; Hufford, 1 997) and the treatment has been followed in recent classifications. Dahlgren (19 83, 1989) shifted the family away from Rosanae to superorder Cornanae. Takhtajan (1997) placed family under sepaand Hydrangeaceae is closer to Cornaceae. The family Hydrangeaceae is divided into two subfamilies: Jamesioideae (2 genera Jamesia and Fendlera) and Hydrangeoideae (rest of genera). Hufford et al., (200 1) on the basis of analysis of sequences of matK and their combination with rbcL and evidence from morphology concluded that Jamesioideae may be sister to the r est of the family. They also divide Hydrangeoi-deae into two tribes Hydrangeae ( conspicuous sterile marginal flowers, valvate petals, loculicidal capsule) and P hiladelpheae (sterile flowers absent, petals imbricate, capsules septicidal). * * * * * * * * * * * Cornaceae Dumortier Dogwood family 2 genera, 83 species (including Alangiaceae DC.) Widely spread, mainly in north temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Cornidae Magnoliidae Asteridae Series+/Superorder Calyciflorae+ Cornanae Cornanae Cornanae Asterids* Order Umbellales Cornales Cornales Cornales

Cornales Cornales Salient features: Mostly shrubs, leaves sim-ple, usually opposite, stipules abse nt, flow-ers bisexual, sepals often enlarged and petaloid, ovary inferior or sem i-inferior, placentation axile or deeply intruded pari-etal, nectary present at top of ovary, fruit a capsule. Major genera: Cornus (65 species) and Alangium (20). Description: Trees and shrubs, rarely stoloniferous subshrubs, glabrous or hairy , usually with iridoids. Leaves usually opposite, sometimes alternate or distich ous, 628 Plant Systematics

rr E '

Figure 13.117 Cornaceae. Cornus macrophylla. A: Portion of plant with flowers; B : Flowertop, view; C: Transverse section of fruit.D: Flower head of C. capitata subtended by four bracts. C. stolonifera E: Flower; F: Vertical section of flower. (A-D, afte r Nasir and Ali, Fl. W. Pak.No. 88, 1975; E-F, Bailey, Man. Cult. Pl., 1949). simple, entire, venation pinnate, secondary veins usually smooth arching towards margin or forming a series of loops, stipules absent. Inflorescence terminal br anched cymes or heads, usually subtended by large showy bract, often forming inv olucre in heads. Flowers usually bisexual, rarely unisexual, actinomorphic, epig ynous. Calyx with 4-5 sepals , united, valvate, lobes rounded, sometimes represe nted by small teeth, rarely absent. Corolla with 4-5 petals, free, imbricate, im bricate or valvate. Androecium with 4-5 stamens, alternating with petals, rarely up to 10, filaments free, arising from the edge of disc, anthers bithecous, bas ifixed or dorsifixed, dehiscing laterally. Gynoecium with 2 united carpels, rarely 1-4, ovary inferior, usually 2-locular, single ovule in each locule, placentation axile, axis lacking vascula r bundles and the ovules attached to vascular bundles that arch over the top of each septum (apical axile placentation), style simple, stigma capitate or lobed, disc present on top of ovary; Fruit usually a drupe, 1-2 seeded, ridged or winged; seeds small, endosperm pres ent. Pollination by bees, flies and beetles. Drupes are dispersed by birdsand ma mmals. Economic importance: Species of Cornus (Dogwood) and Alangium are grown as ornam ental trees and shrubs. Major Families of Angiosperms 629 Phylogeny: The family has undergone a lot realignment in recent years. Bentham a nd Hooker placed it under Araliaceae, but sub-sequently recognised as independen t family under Umbelliflorae (Engler and Prantl). Hutchinson (1948) shifted Corn

aceae along with Araliaceae to order Cunoniales on the basis of woody habit and stem anatomy. The family was earlier broadly circumscribed to include 10-15 gene ra (Hutchinson, 1973; Cronquist, 1988), but was split into a number of families (Takhtajan, 1997) such as Davi-diaceae, Nyssaceae, Mastixiaceae, Curtisia-ceae, Cornaceae and Alangiaceae. APG- II recognises only Curtisiaceae and Cornaceae, N yssaceae optionally merged with latter. Thorne (2006, 2007) recognizes monogener ic Cornaceae, Alangiaceae, whereas rest of the genera are united under Nyssaceae, d ivided into subfamilies Davidioideae, Nyssoideae and Mastixioideae. APWeb (2007) united Alangiaceae and Cornaceae, thus including two genera under Cornaceae. An alysis of a combined matK and rbcL sequence data set by Xiang et al. (1998) esta blished CornusAlangium as distinct clade within Cornales, results confirmed by 2 6S rRNA and combined 26S rDNA-matK-rbcL sequence data of Fan and Xiang (2003) es tablishing Grubbiaceae (Grubbia and Curtisia), Cornaceae (Cornus and Alangium) a nd Nyssaceae (Nyssa, Davidia, Camptotweca, Mastixia and Diplopanax) as distinct clades; Hydrangiaceae and Loasaceae forming indepedent clade. * * * * * * * * * * * Balsaminaceae Bercht. & J. Presl. Balsam family 2 genera, 1000 species Widely distributed but more common in subtropics and tropics of Africa and Asia, and temperate regions New and Old world. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Rosidae Magnoliidae Asteridae Series+/Superorder Disciflorae+ Geranianae Rutanae Ericanae Asterids* Order Geraniales Geraniales Balsaminales Balsaminales Balsaminales Ericales B & H under Geraniaceae Salient features: Somewhat succulent herbs, flowers bisexual, zygomorphic, spurr ed, anthers connate, fliaments closely surrounding ovary, succulent capsule with elastic dehiscence.

Description: Herbs, rarely evergreen subshrubs, often with fleshy translucent st ems, sometimes aquatic, rarely epiphytes. Leaves alternate, opposite or in whorl s of three, simple, usually toothed, stipules as paired glands or absent. Inflor escence of soliMajor genera: Impatiens (997 species) and Hydrocera (3). tary flowers, cymes, racemes or panicles, on axillary peduncles. Flowers bisexua l, 630 Plant Systematics

G 1 H F D

E Figure 13.118 Balsaminaceae. Impatiens glandulifera. A: Portion of plant with fl owers and fruits; B: Flower, top view; C: Longitudinal section of flower with sepals and petals re moved; D: Lateral sepal; E: anterior petal; F: Lateral fused petals; G: Transver se section of ovary; H: Seed. spurred, nodding, often resupinate, hypo-gynous, pentamerous, some flowers cleis to-gamous and self pollinating. Calyx with usu-ally 3 sepals , sometimes upto 5, free, petaloid, imbricate, posterior (lower in ma-ture flower) largest and with nectar spur or pouch on back. Corolla with 5 petals, free, or lateral pairs con nate and appear as 3 petals, lower pair larger than upper. Andro-ecium with 5 st amens, filaments flattened, short, closely covering the ovary, free below, conna te towards top with connate anthers (syngenesious) forming lid (calyptra) over t he ovary, anthers bithecous, dehiscence longitudinal. Gynoecium with 5 united ca r-pels, ovary superior, 5-locular, ovules 3many in each locule, anatrpous, pendulous, placentation axile, style simple, oft en very short, stigmas 1-5. Fruit a fleshy 5-valved capsule, dehiscing explosive ly, valves coiling elastically and on splitting the tension forcibly distributin g the seeds, rarely a berry-like drupe; seeds with straight embryo, endosperm ab sent. Pollination by insects. Capsules dehiscing explosively by autochory or whe n touched slightly, hence the name touch me not . Economic importance: The family is known for its ornamental herbs with showy flo wers. Garden balsam (Impatiens balsamina), Himalayan balsam (I. glandulifera), B usy Major Families of Angiosperms 631 lizzie (I. holstii) and several hybrid cultivars are widely grown in temperate a s well as tropical climates. et al. (2004). Latter authors concluding that Balsaminaceae and Marcgraviaceae a re sister taxa, although there are no obvious synapomorphies. Balsaminaceae show the combination of leucoanthocyanins and raphides, rarely seen in herbs (Fische r, 2004). The family is vegetatively very uniform although florally diverse and duplication and probable subfunctionalisation of the class B DEF gene has occure d in this clade (Janssens et al. 2006; Geuten et al. 2006). Hydrocera and Impati ens are clearly sister taxa (Yuan et al. 2004; Janssens et al. 2006). Species of Impatiens with five sepals are scat-tered through the genus, so that condition is apparently at least sometimes derived. Phylogeny: The family was formerly placed under Geraniaceae, but differes in zyg omorphic flowers, spurred sepal and connate anthers. It is often allied to Tropa eolaceae, but the spur in latter family is derived from recepacular tissue and n

ot calyx as in Balsaminaceae. Balsaminaceae, Marcgraviaceae and Tetrameristaceae form a well defined clade and probably sister to rest of Ericales. Monophyly of these three fami-lies is well supported by the studies of Nandi et al. (1998), Soltis et al. (2000) and Geuten * * * * * * * * * * * Polemoniaceae Bromehead Phlox family 20 genera, 360 species Widely distributed, more common in temperate climate, especially western North A merica. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Asteridae Class Dicotyledons Subclass Gamopetalae Lamiidae Magnoliidae Asteridae Series+/Superorder Bicarpellatae+ Solananae Solananae Eri Asterids* Order Polemoniales Solanales Polemoniales Solanales Polemoniales Ericales Salient features: Mostly herbs, leaves alter-nate, stipules absent, sepals 5, un ited, pet-als 5, united, stamens 5, epipetalous, carpels 3, united, placentation axile, ovules many, style 1, stigmas 3, fruit a capsule. Description: Annual or perennial herbs, rarely shrubs or trees (Cantua) or climb ers (Cobaea), often with glandular hairs, nodes unilacunar. Leaves usually alter nate, rarely opposite or whorled (Gymnosteris), usually simple, sometimes dissec ted or pinnate compound (Polemonium), venation reticulate, stipules absent. Infl orescence terminal or axillary, often crowded into corymbs or heads, Major genera: Gilia (110 species), Phlox (75), Polemonium (40), Limnanthus (40), Ipomopsis (25), Collomia (15) and Cantua (12). 632 Plant Systematics Figure 13.119 Polemoniaceae. Polemonium caeruleum. A: Basal part of plant; B: Up per leaves and inflorescence; C: Flower from above; D: Vertical section of flower; E: Fruit enc

losed in persistent calyx. Phlox nivalis. F: Plant in flower; G: Vertical sectio n of flower ; H: Seed. rarely solitary. Flowers bisexual, actino-morphic, rarely zygomorphic, hypogynou s, usually showy. Calyx with 5 sepals, united, green. Corolla with 5 petals, uni ted, often with a narrow tube, lobes plicate or convolute . Androecium with 5 st amens, adnate to corolla tube (epipetalous; inserted), alter-nating with lobes, filaments free, anthers bithecous, dehiscence longitudinal, pollen grains 4-many aperturate, colpate or porate. Gynoecium with 3 united carpels, rarely 2, ovary superior, seated on a nectary disc, 3- locular, rarely 2-locular, ovules 1 or m ore, unitegmic, placentation axile, style single, elongated, branched above, sti gmas 3, rarely 2. Fruit a loculicidal capsule; seed with straight or curved embr yo, endosperm copious, seed-coat mucilaginous when moistened. Pollination mostly by bees and flies . Seeds are dispersed aided by mucilaginous coat, sometimes by wind or water. Economic importance: The family is known for its ornamentals Phlox (P. drummondi i being most popular), Gilia and Polemonium with showy flowers. Phylogeny: The family is considered to be closely related to Hydrophyllaceae, th e two having closer affinities with group consisting of Solanaceae, Nolanaceae a nd Convolvulaceae. Hutchinson, who places Cuscutaceae under the order Polemonial es, considers it to be most highly evolved within Major Families of Angiosperms 633 the order. Thorne (1999) shifted this family to Dilleniidae (Dillenianae) under separate order Polemoniales (in accordance with the findings of DNA sequence stu dies), retain-ing other families under Lamiidae, order Solanales. Takhtajan plac es all families in-cluding Polemoniaceae under Lamiidae, su-perorder Solananae, but under separate or-ders. Judd et al., (1999) had placed Polemoniaceae under S olanales because of actinomorphic flowers, and united plicate corolla. The studi es of Porter and Johnson (1998), based on morphology and DNA se-quences, however , indicate that the family belongs to Ericales. The family has accord-ingly been shifted to Ericales under Aterids in APG II, Judd et al., (2002) and APweb. Tho rne (2003), who has brought about ma-jor realignments, and abolished Dilleniidae , distributing its members among various subclasses, has also placed Polemoniaceae under Ericales (superorder Ericanae) unde r Asteridae. Traditionally two subfamilies Cobaeoideae and Polemonioideae are re cognized. Acanthogilia with dimorphic leaves and short shoots has been placed in its own subfamily (Porter et al., 2000). Thorne (2006, 2007) accordingly recogn iszes 3 subfamilies Cobaeoideae, Acanthogilioideae and Polemonioideae. Studies b ased on chloroplast gene ndhF (Prather et al., 2000), however, indicate that the genus Acanthogilia may be basal in the Cobaea lineage. Studies of Porter and Jo hnson (1998) have also indicated that woody tropical genera of Cobaeoideae form a paraphyletic basal complex, and the herbaceous genera of Polemonioideae mainly Ipomopsis, Linanthus, Polemonium, Phlox and Gila con-stitute a monophyletic gro up. * * * * * * * * * * * Ebenaceae Grcke Ebony family 5 genera, 600 species Widely distributed in tropics and subtropics, with a few species in temperate re gions of North America and Australia. Placement:

B & H Cronquist Takhtajan

Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Gamopetalae Dilleniidae Magnoliidae Asteridae Primulanae Primulanae Ericanae Series+/Superorder Heteromerae+ Asterids* Order Ebenales Ebenales Styracales Ebenales Sapotales Ericales Salient features: Trees and shrubs without milky latex, bark blackish, leaves di stichous, simple, unisexual flowers, mul-tilocular superior ovary, bitegmic ovul es in pairs. Description: Shrubs and trees with black charcoal-like bark, heartwood black, re d or green, lacking milky latex, buds covered with adpressed hairs. Leaves alter nate, distichous, simple, entire, coriaceous, venation reticulate, lower surface with dark coloured glands, turning blackish on drying due to presence of naptho quinones, stipules Major genera: Diospyros (472 species), Euclea (20) and Lissocarpa (8). 634 Plant Systematics 00 00 0 0 B F C M i Figure 13.120 Ebenaceae. Diospyros paniculata. A: Flowering branch. D. virginian a B: Staminate flower; C: Same in vertical section; D: Pistillate flower; E: Same enlarged in v ertical section; F: Transverse section of ovary. (A, after Brandis, Ind. Trees, 1918; B-F, after Bailey, Man. Cult. Pl., 1949). absent. Inflorescence axillary cymes or with solitary flowers; plants usually mo noecious, male flowers in larger numbers. Flowers unisexual, actinomorphic, male flowers with pistillode, female with staminodes, hypogynous, two bracteoles bel ow flower in Lissocarpa. Calyx with 3-7 sepals, united, persistent and often enl arged in fruit (accrescent). Corolla with 3-7 petals, united, urceolate, lobes i mbricate and contorted, coriaceous. Androecium with same number of stamens as pe

tals or twice as many, and as many whorls, free or united in pairs, epipetalous or free from petals, anthers bithecous, dehiscence longitudinal, introrse, rarel y by apical pores. Gynoecium with 3-8 united carpels, ovary superior, rarely inferior (Lissocarpa), locules many, plac entation axile, each locule with usually 2 ovules at-tached from the top but eac h ovule separated by partition, thus ovary with twice as many chambers and apica l-axile placentation, ovule pendulous, anatropous, integuments two, styles anatr opous to campylotropous, styles and stigmas 3-8, styles free or connate at base. Fruit a berry with often enlarged calyx; seed with straight embryo, endosperm c opious, hard and irregularly grooved or ridged. Economic importance: The family is important source of durable timber Macassar Major Families of Angiosperms 635 ebony (Diospyros ebenum) and Black ebony (D. reticulata). Other species of the g enus are sources of common fruits such as Japanese persimon (D. kaki), American persimon (D. virginiana), and Date plum (D. lotus). ovary) removed to a distinct unassigned family Lissocarpaceae in APG (1998), has been merged with Ebenaceae in APG-II (2003), Thorne (2006, 2007) and APWeb (200 7). Latter places it under subfamily Lissocarpoideae, other two genera under Ebe noideae. Savolainen et al. (2000) suggested removal of Lissocarpa to Rutaceae, b ut rbcL studies supported it as sister to Ebenaceae s. str. (Berry et al. 2001). Euclea and Royena (recognised as 4th genus within Ebenaceae by APWeb) were sist er to Diospyros, and within latter there were a number of well-supported clades, although relationships between them are unclear (Duangjai et al., 2006a, b). Phylogeny: Family is closely related to Sapotaceae but distinct in the absence o f milky latex, unisexual flowers, multilocular superior ovary, ovules in pairs a nd with 2 integuments. From Styracaceae it is dis-tinct by unisexual flowers and septate ovary. Lissocarpa (with large bracteoles, corolla with 8-lobed corona, connate filaments, an-ther with prolonged connective and inferior * * * * * * * * * * * Sapotaceae A. L. de Jussieu Sapodilla family 52 genera, 1250 species Widely distributed in tropics of New World as well as Old World, especially in w et lowland forests, with a few species in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Gamopetalae Dilleniidae Magnoliidae Asteridae Primulanae Primulanae Ericanae Series+/Superorder Heteromerae+ Asterids* Order

Ebenales Ebenales Sapotales Ebenales Sapotales Ericales Salient features: Mostly trees with milky latex, small terminal naked buds with appressed T-shaped hairs, leaves simple, coriaceous, bisexual flowers, multilocu lar superior ovary, single ovule in each cham-ber, unitegmic ovules, seeds with shiny testa. (110), Sideroxylon (75), Chrysophyllum (75), Manilkara (70) and Mimusops (60). Description: Mostly trees, rarely shrubs (Reptonia), with milky latex, with symp odial branches or thorns, silica bodies, triterpenoids and cyanogenic compounds often present, small naked buds brownish hairs, Major genera: Pouteria (325 species), latter T-shaped but one branch often reduc ed. Palaquium (110), Planchonella (100), Madhuca Leaves alternate, sometimes clu stered at 636 Plant Systematics

1r Xa, IR F A B E Figure 13.121 Sapotaceae. A: Flowering branch of Mimusops hexandra. B: Fruiting branch of Pouteria sapota. Chrysophyllum olivaeforme. C: Flowering branch; D: Flower; E: V ertical section of ovary; F: Transverse section of ovary. (A, after Brandis, Ind . Trees, 1918; B-F, after Bailey, Man. Cult. Pl., 1949). shoot apices, simple, entire, coriaceous, ve-nation reticulate, stipules usually absent, rarely present (Madhuca), fresh petioles bot-tle-shaped. Inflorescence axillary cymes, rarely terminal cymes (Madhuca) or with soli-tary flowers. Flowe rs bisexual, ebracteate (Manilkara zapota) or bracteate (Madhuca), actinomorphic , hypogynous. Calyx with 4- 12 sepals, free or connate at base, sometimes in two whorls or spirally arranged, imbricate, persistent. Corolla with 4-12, as many as sepals, united, sometimes with paired petaloid appendages and as such corolla lobes appear 18-24 in number, usually imbricate, rarely valvate. Androecium wit h usually 8-16 stamens, in two or 3 whorls but only inlongitudinal, pollen grains tricolpate or tetracolpate. Gynoecium with 2-many un ited carpels, ovary superior, locules many, placentation axile, each locule with 1 ovule, ovule anatropous with single integument, style single, protruding, sti gma capitate or lobed. Fruit a berry with often leathery bony layer; seed with h ard shiny testa and large hilum, endosperm usually fleshy, rarely absent. Pollin ation by insects. Dispersal of berries by birds and animals. Economic importance: The family provides several delicious tropical fruits like sapodilla (Manilkara zapota), mamey sapote (Pouteria sapota), eggfruit or yellow sapote (P. campechiana) and star apple (Chrysophyllum cainito). Latex of Manilk ara zapata provides chicle, the elastic substance in chewing

ner whorl fertile opoosite petals, others re-duced to staminodes, epipetalous, a nthers bithecous, basifixed or dorsifixed, dehiscence Major Families of Angiosperms 637 gum. Species of Palaquium provide gutta-percha, a latex substance used in golf b alls and submarine telephone cables as insula-tion and in dental stoppings. Seve ral species are used for timber in Malaya. genera having same number of stamens as corolla lobes constitute a well defined clade, those with twice as many stamens form a heterogenous complex, which is pr obably paraphyletic and basal. Thorne (2006, 2007) and APWeb (2008) recognise th ree subfamilies in Sapotaceae: Sarcospermatoideae, Sapotoideae and Chrysophylloi deae. Above three clades are supported by combined morphological molecular analy sis of Swenson and Anderberg (2005). They also concluded that the staminodes com mon in Chrysophylloideae, but derived within the clade, are perhaps not immediat ely comparable with the staminodes of other members of the family; the former ar e outside the staminal whorl, the latter in the same whorl as the stamens. Sarco sperma is regarded as sister to rest of family. Phylogeny: Family is closely related to Ebenaceae but distinct in the presence o f milky latex, bisexual flowers, ovules singly in chambers and with single integ ument. The family is closer to Hoplestigmataceae in single ovule in chamber and single integu-ment (Lawrence, 1951), but the latter family has been shifted to L amiidae by Thorne (2006, 2007), whereas APG-II places Hoplestigma among unplaced genera, and APWeb (2007) the family as unplaced eudicot. The studies of Penning ton (1991) from the study of corolla lobes and stamens concluded that whereas * * * * * * * * * * * Primulaceae Batsch ex Borkh. Primrose family 18 genera, 955 species Largely distributed in north temperate regions, mainly in the Mediterranean regi on, Alps, and Asia Minor. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Gamopetalae Dilleniidae Magnoliidae Asteridae Primulanae Primulanae Ericanae Series+/Superorder Heteromerae+ Asterids* Order Primulales Primulales Primulales Primulales

Primulales

Ericales Salient features: Herbs, leaves opposite or whorled or basal, petals united, ova ry supe-rior, stamens opposite the petals, carpels more than 2, seeds numerous. Description: Perennial herbs, usually with sympodial rhizomes (Primula) or tuber s (Cyclamen) rarely annuals (Anagallis) or subshrubs, sometimes aquatic (Hottoni a), nodes unilacunar, sieve-tube plastids S-type. Leaves opposite, whorled or al ternate, sometimes all basal, simple, sometimes dissected (Hottonia), venation r eticulate, stipules absent, rarely present (Coris). Inflorescence Major genera: Primula (500 species), Lysimachia (200), Androsace (90), Dodecanth eon (50), Anagallis (28) and Cyclamen (15). 638 Plant Systematics

1~ +J

= Y 1 A Figure 13.122 Primulaceae. Primula longiscapa. A: Plant with basal rosette of le aves and scapigerous inflorescence; B: Vertical section of flower to show long corolla tu be and epipetalous stamens; C: Transverse section of ovary with free-central pla centation; D: Fruit dehiscing by recurved apical teeth. Anagallis arvensis. E: P art of plant with opposite and whorled leaves and axillary flowers; F: Flower fr om above; G: Vertical section of flower; H: Fruit with persistent calyx and styl e; I: Pyxidium fruit dehiscing by terminal cap; J: Seed. with solitary axillary flowers (Anagallis), to paniculate (Lysimachia) or umbell ate (Primula), often scapigerous (Primula). Flow-ers bisexual, actinomorphic, ra rely zygo-morphic (Coris), hypogynous, rarely partly epigynous (Samolus), usuall y pentamerous. Calyx with 5 sepals, rarely 6 (Lysimachia) or even 9 (7rientalis) , united, inflated or tubu-lar, imbricate or twisted. Corolla with 5 pet-als, ra rely 4 (Centunculus), 6 (Lysimachia) or 9 (7rientalis), or absent (Glaux), unite d, the tube often short, rotate (Anagallis) or tubular (Primula), lobes imbricat e or twisted. Androecium with 5 stamens (rarely 4 or 6, depending on the number of petals), free, opposite the petals, epipetalous, anthers bi thecous, dehiscence longitudinal, sometimes with apical pores, sometimes with st aminodes alternating the petals. Gynoecium with 5 united carpels, ovary superior or half-inferior, unilocular, ovules many, anatropous to campylotropous, placen tation free central, style simple, stigma capitate or minute, heterostyly is pre valent in the genus Primula. Fruit a capsule, variously dehiscent, pyxidium in A nagallis, opening by a cap like Major Families of Angiosperms 639 cover; seeds with straight embryo, endosperm present, sometimes with aril. Polli nation by various insects. Small seeds are often dis-persed by wind or water, so me by ants at-tracted by the oily aril

Portulacaceae being a connecting link. Over the recent years, there have been at tempts to shift Anagallis, Lysimachia and other genera to Myrsinaceae, separatio n of Maesa from Myrsinaceae into a distinct family Maesaceae and shifting of Sam olus from Primulaceae to Theophrastaceae (Anderberg et al., 2000, 2001). On the basis of phylogenetic analysis based on DNA sequences of rbcL and ndhF Kallesjo et al., (2000) concluded that genera of tribe Lysimachieae (Anagallis, Cyclamen, Gaux, Lysimachia and Trientalis) as also the genera Coris and Ardisiandra shoul d be placed within expanded Myrsinaceae, re-stricting Primulaceae to herbaceous members with campanulate corolla and capsule fruits, a treatment followed in APw eb. Judd Economic importance: The family is impor-tant for several ornamental species of Primula and Cyclamen. Anagallis arvensis is of medicinal importance. Phylogeny: The family is well defined and usually placed under Primulales along with other groups of Dillenianae or Dilleniidae (whichever is recognized). In so me genera such as Samolus, there are five staminodes alternating with petals, in addition to 5 nor-mal stamens opposite the petals, suggestet al. (2002) and Thorne (2003), however, ing that antipetalous condition has resulted have followed broader circumscripti on of famduring course of evolution from the loss of ily Primulaceae retaining t hese genera. Re-the outer whorl (represented in some gen- cent studies have indi cated resemblances era by staminodes). The family is mostly of the family Primul aceae with Ericalean placed closer to Myrsinaceae. Hutchinson, complex under Ast erids. As such Thorne in however, advocated that the two are not re- his recent revision (2003, 2006, 2007) has lated and their free central placentation and sh ifted Primulales to Asteridae (under stamens opposite the petals are due to parEricanae), and Judd et al., (2002), APG II and allel evolution. He considers Pr imulaceae to APweb have placed the family under order have evolved from Caryophy llaceae with Ericales of Asterids. * * * * * * * * * * * Ericaceae A. L. de Jussieu Heath family 140 genera, 2,990 species Widely distributed throughout temperate and subtropical regions, and to some ext ent in subarctic and alpine regions, mainly on acidic soils. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Dilleniidae Class Dicotyledons Subclass Gamopetalae Dilleniidae Magnoliidae Asteridae Series+/Superorder Heteromerae+ Ericanae Ericanae Ericanae Asterids*

Order Ericales Ericales Ericales Ericales Ericales Ericales 640 Plant Systematics Figure 13.123 Ericaceae. Rhododendron glaucophyllum. A: Portion of plant with fl owers; B: Anther; C: Transverse section of ovary; Vaccinium vacillans D: Branch with flowers; E: F lower; F: Vertical section of flower; G: Anther. H: Flowering branch of Lyonia v illosa. (A, H, after Polunin and Stainton, Fl. East. Himal., 1984; rest, after B ailey, Man. Cult. Pl., 1949). Salient features: Mainly shrubs, leaves al-ternate, flowers campanulate to urceo late, stamens twice the number of corolla lobes, arising from nectariferous disc , anthers opening by terminal pores, ovary 4 or more chambered. herbs lacking chlorophyll (Monotropa). trichomes unicellular or multicellular, g landular or scaly, never stellate. Leaves alter-nate, sometimes opposite or whor led, simple, entire or serrate, often coriaceous, per-sistent, sometimes reduced to needles or scales, margin sometimes revolute, venation reticulate, usually p innate, sometimes palmate, blade sometimes reduced to mycoparasite, stipules abs ent. Inflorescence with solitary axillary flowers, cymes, racemes or panicles. F lowers bisexual, rarely unisexual (Empetrum), actinomorphic, sometimes slightly zygomorphic (Rhododendron), hypogynous, . Calyx with 5 sepals, rarely 4-7, disti nct or slightly connate at Major genera: Rhododendron (850 species), Erica (600), Vaccinium (450 incl. Agap etes: 90), Gaultheria (160), Leucopogon (150), Cavendishia (110), Arctostaphylos (70), Dracophyllum (50), Epacris (45), Pyrola (20), Cassiope (12) and Monotropa (5). Description: Shrubs, sometimes small, rarely lianas, epiphytes or mycoparasitic Major Families of Angiosperms 641 base, persistent. Corolla with 5 petals, rarely 4-7, united, corolla tubular, ca mpanulate or urceolate, lobes short, im-bricate or convolute. Androecium with 5 sta-mens sometimes twice the number of corolla lobes, arising from nectar disc, free, filaments flattened at base, sometimes connate at base (Vaccinium), straig ht or S-curved, free from corolla or epipetalous, anthers bithecous, sometimes w ith paired appendages (spurs) near base, dehiscence by apical pores, rarely smal l slits near tip, pollen in tetrads. Gynoecium with 5 united carpels, rarely 2-1 0, ovary usually superior, rarely half-inferior (Gaultheria) or inferior (Vaccin ium), disc present, placentation axile or parietal with deeply intruded placenta e, unilocular, ovules many, anatropous, style simple, conical or filiform, rarel y split above, stigma simple. Fruit a capsule, berry (Vaccinium, Cavendishia) or drupe (Arctosta-phylos, Styphelia); seeds small, embryo straight, endosperm pre sent, seed coat thin Pollination mostly by bees and wasps, aided by nectar. Visc in threads present in Rhodo-dendron and related genera help insects to pull out pollen tetrads. Capsule fruits are dispersed by wind, species with berries or dr upes are usually dispersed by birds. widely grown as ornamentals. Oil of winter-green (methyle salicylate) is obtaine d from Gaultheria procumbens. Foliage of Gaultheria shallon is sold as lemon leaf . Economic importance: The family provides edible fruits blueberry (Vaccinium spp. ), cran-berry (V. macrocarpum) and huckleberry (Gaylussacia; sometimes included

in Vaccinium). Several species of Rhododendron (rhododendrons, azaleas), Calluna (heather), Kalmia (K. latifolia, mountain laurel), Erica (heath), Oxydendrum (s ourwood), Arbutus (madrone) and Leucothoe (fetterbush) are Phylogeny: The broadly circumscribed Ericaceae includes Empetraceae, Epacridacea e, Monotropaceae, Pyrolaceae and Vacciniaceae, sometimes recognized as in-depend ent families, or included under two families Ericaceae (ovary superior, fruit ca p-sule) and Vacciniaceae (ovary semi-inferior or inferior, fruit a berry). Segre gation these families would render Ericaceae as paraphyletic. Monophyly of broad ly circumscribed Ericaceae is supported by evidence from morphology, rbcL and 18 S rDNA sequences (Kron, 1996; Soltis et al, 1997). Engler and Diels (1936) inclu ded four subfamilies: Rhododendroideae (septicidal capsule, seed winged or ribbe d, anthers without appendages), Arbutoideae (fruit berry or loculicidal capsule, seed not winged, anthers app-endaged, ovary superior), Vaccinioideae (ovary inf erior, rest similar to Arbutoideae) and Ericoideae (fruit loculicidal capsule or nut, ovary superior, seeds not winged, calyx persistented, anthers with apicall y spread-ing lobes). Thorne (2006) recognized eight subfamilies, merging Rhodode ndoideae with Ericoideae, and adding Enkianthoideae, Monotropoideae, Cassiopoide ae, Styphelioideae and Empetroideae, subsequently (2007) adding ninth Harrimanel loideae segregated from Cassiopoideae. APWeb (2008) also recognises 8, but inclu des Empetroideae under Ericoideae, recognising Harrimanelloideae as additional s ubfamily. Monotropa uniflora has smallest embryo, just two-celled (Olson 1991) * * * * * * * * * * * 642 Plant Systematics Adoxaceae E. Meyer Elderberry family 5 genera, 245 species (Inc. Sambucaceae Borkh.) Distributed mainly in northern temperate region, a few in montane tropical and s ubsubtropics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Cornidae Magnoliidae Asteridae Series+/Superorder Inferae+ Dipsacanae Cornanae Aralianae Euasterids II* Order Rubiales

Dipsacales Adoxales Cornales Dipsacales Dipsacales B & H under Caprifoliaceae, H

D Fb

` r f _ E I 1 G Figure 13.124 Adoxaceae. Viburnum mullaha. A: Branch with inflorescence; B: Flow er, top view; C: Vertical section of flower; D: Transverse section of ovary. Sambucus nigra. E : Branch with inflorescence; F: Flower, side view; G: Flower, top view; H: Fruit ; I: Seed. (A-B, after Polunin and Stainton, Fl. Himal, 1984) Major Families of Angiosperms 643 Salient features: Mainly shrubs or lianas, leaves opposite, usually simple. crassinucellate. Fruit a small drupe, or drupaceous berry with 3-5 stones, the e mbryo minute, straight, endosperm oily. Pollination by insects, musky flowers of Adoxa attract flies. Dispersal mostly by birds and animals. Major genera: Viburnum (200 species), Sam-bucus (40), Adoxa (1), Sinadoxa (1), t etradoxa (1). Description: Usually shrubs or small trees (Viburnum), tall herbs (some species of Sam-bucus), or small geophytic herbs with creep-ing rhizome (Adoxa), sometime s with storied cambium and crystal sand (Sambucus). Leaves opposite or rarely ve rticillate, in ba-sal rosette and alternate in Adoxa, simple, entire to variousl y toothed or lobed, with or without small stipules (Viburnum), or pinnately or b ipinnately compound, the leaf-lets mostly with serrate margins, and large foliac eous stipules (Sambucus), stipules sometimes reduced to glandular appendages or absent. Inflorescence usually flat-topped, arranged in terminal cymes, corymbs o

r panicles, or small head (Adoxa). Flowers small, bisexual or rarely unisexual, occa-sionally with some marginal flowers neutral and a greatly enlarged corolla, slightly zygomorphic. Calyx with 5 sepals, rarely 3 Economic importance: The family is known for its well known ornamentals belongin g to Sambucus and Viburnum. Black elder (S. nigra) is widely cultivated in temperate as well tropical regions, whereas snow ba ll (Viburnum opulus) is mainly grown in temperate climates. Phylogeny: The genera included here were originally placed in broadly circumscri bed Caprifoliaceae, which has now been reduced to mere 5 genera, including well known genus Lonicera. The genus Adoxa (and recently recognised Sinadoxa and Tetr adoxa) has traditionally been considered distinct from caprifoliaceae in absence of calyx, divided filaments with monothecous anthers, split styles and fruit wi th more than one stones, but now two major genera Viburnum and Sambucus have bee n included in expanded Adoxaceae. Donoghue et al. (2001) studies (Adoxa; usually interpreted to represent bract + 2 bracteoles, calyx treated as absent) the taxonomy of Dipsacales especi ally or 4, imbricate. Corolla with 5 petals, rarely Adoxaceae using rbcL and nuc lear ribosomal 3-4, united, rotate to campanulate, occasion- internal transcribed spacer (ITS) sequences. ally tubular, imbricate, lobes sometimes with nectary (Adoxa). Androecium with a s many stamens as petals, epipetalous, alternating with petals, anthers bithecou s, in Adoxa, Sinadoxa and Tetradoxa the stamens appear to be divided into two ha lf stamens, each with a separate filament and monothecal anther, half anther bei ng peltate in Tetradoxa, dehiscence by longitudinal slits. Gynoecium with 3-5 un ited carpels, 1 in Sinadoxa, ovary semi-inferior, rarely supe-rior (Tetradoxa), 3-5 locular, the style soli-tary and terminal, divided in Adoxa, Sinadoxa and Te tradoxa, the stigma mostly capitate, the ovule 1 per locule, pendulous, anatropo us, unitegmic and tenuinucellar or They concluded that the 5 genera formed a clearly defined clade with Viburnum be ing sister to rest of the four genera, followed by Sambucus, which is sister to remaining three. Data established Adoxaceae and Caprifoliaceae as two major clad es within the order. Within Adoxaceae Adoxa, Sinadoxa and Tetradoxa form Adoxina clade marked by herbaceous habit, reduction in the number of perianth parts, ne ctaries of multicellular hairs on the perianth, and bifid stamens. Thorne (2006, 2007) recognizes two subfamilies Adoxoideae (4 genera) and Opuloideae (Viburnum ). APWeb also recognizes these two clades as Adoxa clade and Viburnum clade, res pectively. * * * * * * * * * * * 644 Plant Systematics Apiaceae Lindley Carrot family (=Umbelliferae A. L. de Jussieu) 440 genera, 3,590 species Mainly distributed in north temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren

Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Cornidae Magnoliidae Asteridae Series+/Superorder Calyciflorae+ Aralianae Aralianae Aralianae Euasterids II* Order Umbellales Aralialales Araliales Araliales Apiales B & H as Umbelliferae, others as Apiaceae E 1~ K Figure 13.125 Apiaceae. Coriandrum sativum. A: Upper portion of plant with compo und umbels in flower and fruit; B: Part of lower leaf with broader segments; C: Inner actinomorphic flower; D: Outer zygomorphic flower; E: Vertical section of flower; F: Crem ocarp with persistent stylopodium at tip. Foeniculum vulgare. G: Portion of bran ch with compound umbels without bracts; H: Flower; I: Vertical section of flower ; J: Cremocarp with forked carpophore separating 2 mericarps. Bupleurum candollii. K: upper portion of plant with simple entire leaves (rare situation in this fami ly) and umbels; L: Cremocarp. M: Upper portion of plant of Eryngium biebersteini anum with spiny leaves and sessile head-like umbels. Major Families of Angiosperms 645 Salient features: Aromatic herbs with hollow internodes, leaves compound with sh eathing base, inflorescence umbel, petals incurved in bud, yellow or white, stam ens 5, inflexed in bud, ovary inferior, fruit a cremocarp with stylopodium at ap ex. called vittae inside. Seeds with small embryo, endosperm oily. Major genera: Eryngium (230 species), Ferula (150), Pimpinella (150), Bupleurum (100), Heracleum (60), Sanicula (40), and Chaerophyllum (40). Economic importance: The family is the source of food plants, spices and condime nts. Carrot (Daucus carota) and parsnip (Pastinaca sativa) are important root cr ops. Important flavouring plants include fennel (Foeniculum vulgare), coriander (Coriandrum sativum), caraway (Carum carvi), anise (Pimpinella anisum) and celer y (Apium graveolens). Cicuta, Conium (hemlock, which Socrates is said to have us ed for suicide) and Oenanthe include poisonous plants. Description. Herbs with hollow internodes, commonly aromatic, rarely shrubs (Ery ngium giganteum), or even climbers (Pseudocarpum), sometimes forming huge cushio ns (Azorella). Stems often fistular, with secretary canals containing ethereal o ils and resins, coumarins, and terpenes, plants character-istically containing u Apiales

mbelliferose, a trisac-charide storage product. Leaves alternate, rarely opposit e (Apiastrum), lobed or com-pound, rarely simple (Bupleurum), petioles with shea thing base, stipules absent. Inflo-rescence of simple or compound umbels, of-ten subtended by involucre of bracts (involu-cre bracts of umbel branches and involuc el bracts of flowers; absent in Foeniculum), sometimes like a head (Eryngium). Fl owers small, bracteate or ebracteate (Foeniculum), usually pedicelled, rarely se ssile (Eryngium) bisexual, rarely uni-sexual (Echinophora), actinomorphic (rarel y zygomorphic), epigynous. Calyx with 5 sepals, adnate to ovary, 5-lobed, lobes often very small. Corolla with 5 petals, free, valvate or slightly imbricate, in curved in bud, notched at tip. Androecium with 5 stamens, free, inflexed in bud, exserted in open flower, rarely included, anthers bithecous, dehiscence lon-git udinal, pollen grains usually tricolpate. Gy-noecium with 2 united carpels (sync arpous), with inferior ovary, bilocular with 1 ovule in each chamber, placentati on axile, style sur-rounded at base by bilobed nectary, the basal portion of sty le along with nectary persisting in fruit as stylopodium. Fruit schizocarpic kno wn as cremocarp splitting at maturity into two mericarps attached by a common st alk carpophore, mericarp containing oil canals Phylogeny: Apiaceae and Araliaceae have been considered as closely related famil ies for a long time, often included in the same order (Bentham and Hooker, Engle r and Prantl), a trend continued by almost all recent authors, though Hutchinson (1926, 1973) had separated the two under distinct orders, and even under differ ent groups Lignosae and Herbaceae. This separation was arbitrary and as such in most recent classifications they are placed closer together under Araliales (Dah lgren; Takhtajan, Thorne) or Apiales (Cronquist, APG II, APweb), Monophyly of th e family is supported by morphology, secondary metabolites, rbcL and matK sequen ces (Judd et al., 1994; Plunkett et al., 1997). Earlier studies (Judd et al., 19 99) had indicated that Apiaceae are most closely related to Pittosporaceae, but recent data (APweb; Plunkett, 2001) points to Pittosporaceae be-ing sister taxon of the whole group or Pittosporaceae may be embedded in Apiaceae + Araliaceae + other taxa. The family Apiaceae is usually divided into two sub-families: Sanic uloideae (Leaves often broad, with hairy or thorny leaf teeth, stylopodium separ ated from style by groove, fruit scaly or spiny, vittae often poorly developed) and Apioideae ( umbels compound, stylopodium lacking groove, carpophore free, bi fid, mericarps attached at apex). Recent molecular studies (Downie et al., 2000a , 2000b) have indicated that traditional division into tribes and genera may und ergo substantial rearrangement. The genera formerly included in 646 Plant Systematics Hydrocotyloideae (including genera Hydrocotyle, Centella, etc. ) form a polyphyl etic group and as such have been segregated to Araliaceae (Hydrocotyle) and Mack inlaya-ceae (Centella, Trachymene, etc.) by Downie et al., (2000) and Chandler & Plunkett (2003; 2004, quoted in APweb). Stevens (APweb, 2003) points out that s ampling must improve to resolve affinities especially with regard to Hydrocotyle and Trachymene. Thorne (2003), has shifted Centella and 5 other genera to Macki nlayaceae but placed Hydrocotyle and Trachymene in Araliaceae, recognizing only two subfamilies Apioideae and Saniculoi-deae under Apiaceae. Judd et al., (1999, 20 02) argued that if Apiaceae and Araliaceae, in close to their traditional circum scriptions, were recognized, they would be poorly characterized morphologically, and certain genera would have no well-supported familial placement. They accordingly merge Araliaceae and Mackinlayaceae with Apiaceae, recognizing three subfamilies Arali oideae, Apioideae and Saniculoideae. Thorne (2003) and APG II (2003) treat all t he three families as independent. APweb (2003 onwards) recognizes Araliaceae as an independent family, but relegates Mackinlayaceae to subfamily Mackinlayoideae , recognizing additional subfamily Azorelloideae (some former members of Hydroco tyloideae), thus recognizing a total of 4 subfamilies (other two being Apioideae and Saniculoideae), treatment followed by Thorne (2006, 2007). * * * * * * * * * * * Araliaceae A. L. de Jussieu

Aralia family 54 genera, 1325 species Widely distributed with most species in tropics and subtropics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Polypetalae Rosidae Cornidae Magnoliidae Asteridae Series+/Superorder Calyciflorae+ Aralianae Aralianae Aralianae Euasterids II* Order Umbellales Apiales Aralialales Araliales Araliales Apiales Salient features: Leaves often large, alter-nate, compound, leaving large scars on fall-ing, flowers small, actinomorphic, pentamerous, usually in umbels, ovary inferior, each locule with single ovule, fruit a berry. (50), Trachymene (45), Eleutherococcus (38), and Hedera (15). Major genera: Schefflera (650 species), Polyscias (150), Hydrocotyle (130), Oreo panax (85), Dendropanax (70), Aralia (65), Osmoxylon Description. Herbs, shrubs, trees or lianas with prickly or stellate hairs, some times palm-like, a few root-climbers (Hedera). Leaves alternate, rarely opposite or whorled, large, lobed or more commonly pinnately or palmately compound, rare ly simple, petioles Major Families of Angiosperms 647 B C y~>

F Figure 13.126 Araliaceae. Eleutherococcus cissifolius. A: Portion of plant with inflorescence B: Flower; C: Fruit. Hedera nepalensis D: Branch with inflorescence; E: Flower; F: Transverse section of ovary; G: Fruit. (A-C, after Polunin and Stainton, Fl. Himal., 1984) often with sheathing base usually formed by membranous stipules, leaving large s car on stem after falling. Inflorescence usually umbellate, rarely corymbose, ra cemose or panicled, spikes or heads. Flowers small, usually pedicelled, greenish or whitish, bi-sexual, rarely unisexual and dioecious, epigynous, rarely hypogy nous, bracts very small. Calyx with 5 sepals, adnate to ovary, lobes reduced to small teeth or seam-like rim. Corolla with 5 petals, rarely 3-10, free, broader at base, arising from dis c, valvate, caducous, falling separately or as calyptralike cap. Androecium with 5 stamens, rarely 3-10, as many as petals and alternating them, free, anthers b ithecous, dorsifixed, dehiscence longitudinal. Gynoecium with usually 5 united c arpels, sometimes 2-15, rarely 1, ovary inferior, locules as many as carpels, ov ule one in each locule, on apical648 Plant Systematics axile placentas, anatropous, raphe ventral, styles as many as carpels, free and recurved or connate into a column or cone (stylopo-dium), rarely absent and stig mas sessile. Fruit a berry or drupe, rarely schizocarpic and splitting into pyre nes or mericarps; seed with small embryo at one end, endosperm copious, sometime s ruminate. The genera formerly included in Hydrocotyloideae of Apiaceae (including genera H ydro-cotyle, Centella, etc. ) form a polyphyletic group and as such have been se gregated to Araliaceae (Hydrocotyle) and Mackinlayaceae (Centella, 1rachymene, e tc.) by Downie et al., (2000) and Chandler and Plunkett (2003; 2004). Stevens (A Pweb, 2003) points out that sampling must improve to resolve affinities especial ly with regard to Hydrocotyle and 1rachymene. Thorne (2003, 2006), has shifted C entella and 5 other genera to Mackinlayaceae but placed Hydrocotyle and 1rachyme ne in Araliaceae. Judd et al., (1999, 2002) argued that if Apiaceae and Araliace ae, in close to their traditional circumscriptions, were recognized, they would be poorly characterized morphologically, and certain genera would have no well-s upported familial placement. They accordingly merge Araliaceae and Mackinlayacea e with Apiaceae, recognizing three subfamilies Aralioideae, Apioideae and Sanicu loideae. Thorne (2003) and APG II (2003) treated all the three families as indep endent, but Thorne subsequently (2006, 2007) merged Saniculoideae and Mackinlaya ceae with Apiaceae and divided Araliaceae into two subfamilies Aralioideae and H ydrocotyloi-deae. APweb (2003, 2008) follows the same treatment. Basal Araliacea e may well be bicarpellate and have simple leaves. Both these are features of th e herbaceous Hydrocotyloideae, sister to the rest of the family (Chandler & Plun kett 2004; Plunkett et al. 2004a). Economic importance: The family is known for its ornamental foliage plants such as Schefflera, Fatsia japonica, Eleutherococcus, Aralia and Hedera. Many cultivars of Hedera helix (Ivy) are used as house plants. Roots of

Ginseng plant (Panax ginseng) from China and Korea yield drug Ginseng used medic i-nally as stimulant and aphrodisiac. Ameri-can Ginseng (P. quinquefolia) is als o being used as substitute for true ginseng. Aralia cordata and A. racemosa are also used me-dicinally. Phylogeny: Apiaceae and Araliaceae have been considered as closely related famil ies for a long time, often included in the same order (Bentham and Hooker, Engle r and Prantl), a trend continued by almost all re-cent authors, though Hutchinso n (1926, 1973) had separated the two under distinct orders, and even under diffe rent groups Lignosae and Herbaceae. This separation was arbitrary and as such in most recent clas-sifications they are placed closer together under Araliales (D ahlgren; Takhtajan, Thorne) or Apiales (Cronquist, APG II, APweb). * * * * * * * * * * * Major Families of Angiosperms 649 Asteraceae Martinov Sunflower or Aster family (= Compositae Giseke) 1,532 genera, 23,790 species (largest family of flowering plants) Worldwide in d istribution mainly in temperate and subtropical climates, mainly in mountain reg ions, also common in tropics. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Gamopetalae Asteridae Subclass Asteridae Magnoliidae Asteridae Series+/Superorder Inferae+ Asteranae Asteranae Asteranae Euasterids II* Order Asterales Asterales Asterales Asterales Asterales Asterales B & H as Compositae; others as Asteraceae Salient features: Usually herbs, lacking Iridoids, leaves usually alternate, sti pules absent, inflorescence a capitulum with ray florets and disc florets (one t

ype or both in a head), surrounded by involucre bracts (phyllaries), calyx repre sented by pappus, anthers united forming a cylinder around style, style with two branches, fruit a cypsela (commonly called achene, although typical achene is f ormed from single carpel and superior ovary), ovary inferior. ple, sometimes compound (Dahlia, Artemisia), rarely opposite (Dahlia) or whorled , stipules absent. Inflorescence a capitulum with broad receptacle containing di sc florets (discoid head- Ageratum, Vernonia), ray florets (ligulate head- Sonch us, Launaea) or both type of florets with latter towards the periphMajor genera: Senecio (1470 species), Verery (radiate head- Helianthus, Aster), all types of heads having florets surroun ded by involucre bracts (phyllaries), rarely capitulum with single floret (Echin ops) with capitula arranged into globose heads. Flowers bisexual (usually disc f lorets and ray florets of a ligulate head) or unisexual (commonly ray nonia (1050), Cousinia (600), Eupatorium (590), Centaurea (590), Hieracium (470), Helichrysum florets in a ligulate head, which may even (460), Saussurea (300), Cirsium (270), Aster be sterile), actinomorphic (usually disc flo(240), Bidens (210), Chrysanthemum (200), rets) or zygomorphic (usually ray florets), Crepis (200), Inula (200), Gnaphalium (140), epigynous. Calyx absent or represented by Solidago (110), Helianthus (100), Carduus (90), p appus in the form of scales (Helianthus), Lactuca (90), Taraxacum (80), Tragopogon (70), bristles (Bidens), simple hairs ( Sonchus) or Sonchus (50) and Calendula (30). plumose (Carduus). Corolla with 5 petals, united, tubular and 5-lobed (disc flor et) or ligulate with 3-5 teeth (ray floret: sometimes also bilabiate). Androeciu m with 5 stamens with free filaments and united anthers (syngenesious) forming a tube around the style, epipetalous, anthers bithecous, dehiscence longitudinal. Gynoecium with 2 united carpels, unilocular a single ovule, placentation basal, ovary inferior, style with Description: Usually herbs or shrubs, rarely trees (Vernonia arborea; Leucomeris ) or lianas (Vernonia scandens), sometimes producing tubers (Dahlia, Helianthus tuberosus), usually storing inulin, laticifers usually present, rarely lacking, terpenoids usually present, usually sesqueterpene lactones, Iridoids absent. Lea ves usually alternate and sim650 Plant Systematics Figure 13.127 Asteraceae. Helianthus annuus. A: Portion of plant with infloresce nces, the capitulum with both ray florets and disc florets (radiate head); B: Vertical secti on of ray floret lacking androecium; C: Vertical section of disc floret. Ageratu m houstonianum. D: Portion of plant with capitula in clusters, each with only di sc florets (discoid head); E: Vertical section of disc floret; F: Achene with pa ppus consisting of 5 scales. G: Plant of Sonchus oleraceous with auricled leaves and capitula with only ray florets (ligulate head). H: Plant of Launaea nudicau lis with ligulate heads. Carthamus lanatus. I: Portion of plant with spinose lea ves and discoid heads; J: Capitulum with spiny involucre bracts. Bidens chinensi s. K: Lower part of plant with pinnate leaves; L: Upper part with one flowering and one fruiting capitulum; M: Ray floret with three-toothed corolla; N: Disc fl oret; O: Disc floret with corolla partly removed to show androecium; P: Achene. Major Families of Angiosperms 651 two branches. Fruit a cypsela (often called achene which typically, however, is formed from single carpel with superior ovary) usu-ally with pappus at tip. Seed s 1, embryo straight, endosperm usually absent. Calyceraceae, Goodeniaceae and their sister group Menyanthaceae form a mono-phyl etic group. All the four families are placed under Asterales by Thorne (1999, 20 03). The relationships between the first three families are not very clear. The rbcL and ndhF (Krehed et al., 1999) and ndhF data (Olmstead et al., 2000) support Asteraceae and Calyceraceae as sister families whereas rbcL together with atpB and 18S rDNA (Soltis et al., 2000) support Goodeniaceae and Calyceraceae as sist er taxa. A combination of morphological data, and rbcL, ndhF and atpB sequences

provided a strong support for Calyceraceae and Asteraceae as sister groups (Lund berg and Bremer, 2002). Similar conclusion was also reached by the analyEconomic importance: Compared to the number of species included, the family is o f lesser economic importance. Common valuable ornamentals include species of Ast er, Dahlia, Chrysanthemum, Gerbera, Helichrysum, Tagetus and Zinnia. A few food plants include Lactuca (lettuce), Cynara (artichoke), Helianthus (sunflower oil) , and Cichorium (chicory, added to coffee). Safflower a red dye is obtained from Carthamus tinctorius. Latter is now more commonly cultivated for its seeds yiel ding safflower oil, used in cooking. Chrysanthemum cinerarie-folium is the sourc e of natural insecticide pyrethrum. sis of six DNA regions (Bremer et al., 2002). Phylogeny: Interestingly Asteraceae in spite of huge size form a well-defined cl ade, easily recognizable and evidently mono-phyletic. The family is often consid ered re-lated to Rubiaceae, Caprifoliaceae, Dipsacaceae, Valerianaceae, Campanul -aceae and a few others. The first four are basically cymose and also differ in biochemi-cal features. Stylidiaceae, Goode-niaceae and Brunoniaceae resemble the Asteraceae in being mostly racemose and in possess-ing inulin, but differ in bi ochemical features of taxonomic significance. Recent molecu-lar studies (Bremer et al., 2002; Lundberg and Bremer, 2002) indicate that Asteraceae, The family Asteraceae is usually divided into three subfamilies: Barnadesioideae (style papillate, stigma lobed; cypsela with spines; lacks chloroplast DNA inve rsion found in other two subfamilies), Cichorioideae (Latex present, style branc hes long with inner surface stigmatic, acute; those with ray florets often separ ated into a distinct subfamily Lactucoideae) and Asteroideae (latex absent, both disc and ray florets). Thorne (2006, 2007) recognizes Carduoideae including Cic horioideae and Lactucoideae. He sub-divides these three subfamilies further to i nclude a total of 25 tribes in Asteraceae. Heywood had earlier (1978) recognized 17 tribes under two subfamilies Lactucoideae and Asteroideae. APweb (2003) reco gnizes 11 subfamilies including one undefined The Stifftia group . * * * * * * * * * * * 652 Plant Systematics Subclass 11. Lamiidae 3. Tetrachondraceae (B) 4. Plocospermataceae (B) 5. Gratiolaceae (B) 6. Calceolariaceae (B) 7. Gesneriaceae 8. Plantaginaceae (B) 9. Pedaliaceae 10. Linderniaceae 11. Byblidaceae 12. Lamiaceae (B) 13. Paulowniaceae (B) 14. Orobanchaceae (B) 15. Phrymaceae 16. Nesogenaceae (B) 17. Schlegeliaceae (B) 18. Verbenaceae (B) 19. Martyniaceae 20. Petraeaceae (B) 21. Trapellaceae (B) 22. Bignoniaceae 23. Lentibulariaceae 24. Acanthaceae 25. Stilbaceae 26. Scrophulariaceae Superorder 1. Solananae Order 1. Garryales (B) Family 1. Garryaceae

2. Eucommiaceae 3. Aucubaceae 4. Oncothecaceae (B) 5. Icacinaceae (B) 2. Solanales Suborder 1. Solanineae 1. Solanaceae 2. Convolvulaceae 3. Hydroleaceae (B) 4. Sphenocleaceae 5. Montiniaceae (B) 2. Boraginineae 1. Boraginaceae 2. Hydrophyllaceae 3. Lennoaceae 4. Hoplestigmataceae (B) Superorder 2. Lamianae Order 1. Rubiales l. Gentianaceae 2. Loganiaceae Taxa Incertae Sedis Haptanthus Goldberg & Nelson (1). Hondu3. Gelsemiaceae 4. Rubiaceae ras. 5. Apocynaceae Heteranthia Nesse & C.Mart. (1) Brazil. 2. Lamiales *Pottingeria Prain (1) 1. Oleaceae Pteleocarpa Oliver (1) W. Malesia. 2. Carlemanniaceae (B) * Under Hydrangeales-> Pottingeriaceae by Takhtajan Solanaceae A. L. de Jussieu Nightshade or Potato family 98 genera, 2,715 species Cosmopolitan in distribution, found both in temperate and tropical climates with largest concentration in Central and South America. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnolipsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Solananae Solananae

Solananae Euasterids I* Order Polemoniales Solanales Solanales Solanales Solanales Solanales Major Families of Angiosperms 653 i A

E F I B

~ A

M G Figure 13.128 Solanaceae. Solanum nigrum. A: Branch with extra-axillary rhipidiu m infloresences and fruits; B: Vertical section of flower. Withania somnifera. C: Branch with ax illary cymose clusters of flowers; D: Flower; E: Vertical section of flower with bell-shaped corolla; F: Fruit enclosed in enlarged urceolate calyx. Datura inox ia. G: Branch with axillary funnel-shaped flowers; H: Transverse section of ovar y, tetralocular due to false septum; I: Capsule covered with tubercles and basal persistent portion of calyx. Physalis minima. J: Portion of plant with flowers; K: Flower; L: Vertical section of flower; M: Transverse section of ovary with s wollen placentae; N: Fruit with inflated calyx removed from one side; O: Seed; P : Flowering branch of Atropa belladona. Salient features: Leaves alternate, stipules absent, flowers actinomorphic, stam ens 5, carpels 2, ovary superior, 2-chambered, placenta swollen, septum oblique, ovules nu-merous, fruit a berry or capsule.

(110), Physalis (95), Lycium (90), Capsicum (50), Hyoscyamus (25) and Datura (10 ). Major genera: Solanum (1350 species), Lycianthus (190), Cestrum (160), Nicotiana Description: Herbs, shrubs (Brunfelsia, Cestrum) or small trees (Solanum verbasc ifolia; Dunalia), rarely lianas, often poison-ous, sometimes with prickles, unde rground tubers in Solanum tuberosum (potato), 654 Plant Systematics vascular bundles with both outer and inner phloem. Leaves alternate, simple, rar ely pinnately compound (potato), stipules absent, paired leaves adjacent on the stem are com-mon. Inflorescence cymose (Solanum) or of solitary flowers (Datura) . Flowers bisexual, actinomorphic, hypogynous. Calyx with 5 sepals, united, pers istent, sometimes en-larged and swollen in fruit (Withania, Physa-lis). Corolla with 5 petals, united, rotate (Solanum) or tubular (Cestrum), rarely funnel shap ed (Datura) or bilabiate (Schizanthus). Androecium with 5 stamens epipetalous, i nserted in corolla tube, filaments free, bithecous, anthers introrse, dehiscence lon-gitudinal or by apical pores. Gynoecium with 2 carpels, rarely 3-5 (Nicandr a), united, ovary superior, bilocular, axile placentation, pla-centa swollen, se ptum oblique, ovary often further divided by false septa, style 1, rarely gynoba sic (Nolana), stigma bilobed, ovary seated on a nectary. Fruit berry or capsule (Datura); seeds many, embryo straight, endosperm present. Pollination mostly by insects. Dispersal mostly by birds. ated in vascular bundles having outer and inner phloem, actinomorphic flowers an d oblique septum of the ovary. Schizanthus with zygomorphic flowers is borderlin e genus. The family also has close affinities with Convolvulaceae, Boraginaceae and Gesneriaceae. Nolanaceae with gynobasic style and lobed ovary has been merge d with Solanaceae. The following 7 subfamilies are recognized (Olmstead et al., 1999; APweb, 2003): Schwenckioideae (pericycle fibres present, stamens 4, didyna mous, or 3 staminodes; embryo straight, short), Schizanthoideae (pericycle fibre s absent, flowers zygomorphic, anterior petals connate, forming a keel, stamens 2, staminodes 3, embryo curved), Goetzeoideae (Fruit often a drupe, embryo curve d: Takhtajan as family Goetzeaceae), Cestroideae (pericyclic fibres present, sta mens 4 or 5, often didynamous), Petunioideae (flowers bisymmetric, embryo slight ly curved), Solanoideae (seeds flattened, embryo curved, often coiled), and Nico tianoideae (Cork superficial pericyclic fibres present or absent, stamens 4 or 5 , of two lengths, embryo straight or curved). The grouping (Petunioideae (Solano ideae + Nicotianoideae)) is well supported, although the relationships between t he more basal branches have only weak support, but Schwenkia is probably sister to the rest of the family (Olmstead et al., 1999). Family Sclerophylacaceae has been variously included under Solanaceae (Hutchinson, Cronquist, APG II), treate d as distinct family (Takhtajan, Dahlgren), or con-sidered unplaced (APweb) has been treated as subfamily Sclerophylacoideae of Solanaceae by Thorne (2003), who recognizes Browallioideae, Solanoideae and Goetzeoideae as other three subfamil ies In subsequent revisions in 2006 and 2007, he recognises 7 subfamilies, addin g Schizanthoideae, Petunoideae and Nicotianoideae. APweb (2008) likewise recogni ses 7 subfamilies, taking out Schwenckioideae doubtfully placed in Browallioidea e by Thorne, Sclerophylaceae still unplaced. Olmstead et al. (1999) considered S chwenckia sister to rest of family, but Martins and Barkman (2005) using the Economic importance: The family includes a number of food plants such as tomato (Lycopersicon esculentum), potato (Solanum tuberosum), egg plant or brinjal (S. melongena), ground cherry (Physalis peruvi-ana). Peppers (Capsicum annuum) are u sed both as a food source (young) and spices (ripe). Many poisonous species are impor-tant drug plants such as Atropa belladona (atropine), Hyoscyamus niger (he nbane-hyp-notic drug), Datura stramonium (stramonium) and Mandragora officinarum (mandrake). To-bacco (Nicotiana tabacum and N. rustica) con-tains toxic alkaloi d nicotine and is grown for chewing, smoking and snuff. Some orna-mental genera include Brunfelsia (lady-of-the-night; yesterday-today-and tomorrow), Cestrum ( Ra t ki Rani ; night blooming jessa-mine), Petunia, Physalis (ground cherry) and Sola num (nightshade). Phylogeny: The family is closely related to Scrophulariaceae from which it is di

fferentiMajor Families of Angiosperms 655 nuclear gene SAMT (salicylic acid methyl transferase), found Schizanthus to be s ister to the rest of the family, and with rather strong support, with Schwenkia weakly linked with Cestroideae. Furthermore, Wu et al. (2006) found a strongly supported group ing of [Solanoideae [Petunioideae + Nicotianoideae]. * * * * * * * * * * * Convolvulaceae A. L. de Jussieu Morning glory family 59 genera, 1,830 species Widely distributed, mostly in tropical and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Solananae Solananae Solananae Euasterids I* Order Polemoniales Solanales Convolvulales Solanales Solanales Solanales Salient features: Usually twining or climb-ing herbs, commonly with latex, leave s alternate, venation palmate, stipules absent, flowers actinomorphic, corolla f unnel-shaped, stamens 5, carpels 2, ovary superior, 2-cham-bered, , ovules 1 or 2, fruit a capsule. base, persistent. Corolla with 5 petals, united, funnel-shaped, usually plicate. Androecium with 5 stamens, epipetalous, inserted in corolla tube, often unequal , filaments free, bithecous, anthers introrse, dehiscence longitudinal or by api cal pores, pollen grains tricolpate or multiporate. Gynoecium with 2 united carp els, ovary superior, entire or deeply bilobed, bilocular, axile placentation, st yle 1, terminal or gynobasic, stigma bilobed, capitate or linear, ovary seated o n a nectary. Fruit a capsule; seeds 1 or 2 in each chamber, embryo straight or c urved, cotyledons folded. Pollination mostly by insects. Major genera: Ipomoea (550 species), Con-volvulus (240), Cuscuta (140), Jacquemo

ntia (110), Evolvulus (95) and Calystegia (25). Description: Twining or climbing herbs, of-ten rhizomatous, latex usually presen t, sometimes parasitic (Cuscuta), rarely tree (Humbertia), vascular bundles with both outer and inner phloem, sometimes with alkaloids, branching usually sympod ial. Leaves alternate, simple, rarely lobed or compound, sometimes absent (Cuscu ta), venation pal-mate, reticulate, stipules absent. Inflores-cence cymose or wi th solitary flowers. Flow-ers bisexual, actinomorphic, rarely zygomorphic (Humbe rtia), hypogynous. Calyx with 5 sepals, free or slightly connate at Economic importance: Ipomoea batatas (sweet potato) is important for its edible roots. Important ornamentals include Ipomoea (morning glory), Porana (Christmas vine) and Dichondra (ponyfoot). Roots of Convolvulus scammonia (scammony) and of Ipomoea purga (jalap) yield a drug used medicinally as cathartic. 656 Plant Systematics Figure 13.129 Convolvulaceae. Convolvulus arvensis. A: Branch with flowers and f ruits; B: Flower from above. Ipomoea arachnosperma. C: Branch with flowers; D: Flower. Rivea hypo crateriformis. E: Portion of plant with leaves; F: Flower; G: Fruit with persist ent calyx. H: Flower of Seddera latifolia. Phplogenp: The family is closely related to Solanaceae, Boraginaceae and Polemon ia-ceae. Cuscutaceae and Dichondraceae, some-times recognized as distinct famili es, are better placed in Convolvulaceae, their sepa-ration leading to paraphyly of Convolvulaceae. Monophyly of the family is supported by mor-phological charac ters. Thorne (1999, 2000) recognized 4 subfamilies: Humbertioideae (Humbertia; t ree), Dichondroideae (2 genera Dichondra, Falkia), Convolvuloideae and Cuscutoid eae (Cuscuta; without leaves), sub-sequently (2003) merging Dichondroideae in Convolvuloideae, but restored in 2006 and 2007 revisions. APweb recognizes only two Humbertioideae and Convolvuloideae (including both Cuscutoideae and Dichondr oideae). On the basis of DNA sequences of multiple chloroplast loci, Stefanovic et al., (2002) concluded that Poranae (including Porana, itself polyphyletic) an d Erycibeae successively form basal clades in Convolvuloideae. The basal Poranea e have foliaceous bracts and fruits that are utriculate. Erycibe (Erycibeae) has sessile stigmas. * * * * * * * * * * * Major Families of Angiosperms 657 Boraginaceae A. L. de Jussieu Borage family 117 genera, 2,435 species Widely distributed in temperate, tropical and subtropical regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Solananae Solananae Series+/Superorder Bicarpellatae+ Solananae Euasterids I*

Order Polemoniales Lamiales Boraginales Boraginales Solanales Unplaced in Euasterids I Salient features: Bristly herbs, stems cy-lindrical, leaves alternate, infloresc ence helicoid cymes, flowers pentamerous, actinomorphic, carpels 2, ovary 4-lobe d, style gynobasic, fruit with 4 nutlets. truncate, ovary seated on a nectary. Fruit a drupe with one 4-seeded, two 2-seed ed or four 1-seeded pits or schizocarp with four 1 seeded nutlets; seeds with em bryo straight or curved. Pollination mostly by insects. Drupaceous fruits are di spersed by birds, whereas corky ones (Argusia, Cordia) are carried away by water . Major genera: Cordia (300 species), Heliotropium (250) Tournefortia (240), Onosm a (140), Myosotis (90), Cynoglossum (75) and Ehretia (75). Economic importance: Several species of Heliotropium (heliotrope), Mertensia (vi rgin bluebells), Myosotis (forget-me-not), Cordia (Geiger tree), Cynoglossum (ho und s tongue) and Pulmonaria (lungwort) are grown as ornamentals. Several species such as Borago officinalis (borage), Symphytum officinalis (comphrey) and Lithos permum spp. (pucoon) have been used as medicinal herbs. Alkanna tinctoria (alkan et) is a source of red dye used to stain wood and marble and to colour medicines , wines and cosmetics. Description: Herbs, shrubs or trees (Cordia), sometimes lianas, inner phloem lac king, hairs with basal cystolith and often calcified or silicified and as such p lants bristly, rough to touch. Leaves alternate, simple, entire, venation pinnat e, reticulate, stipules ab-sent. Inflorescence usually of helicoid cymes, rarely scorpioid. Flowers bisexual, actinomorphic, rarely zygomorphic (Echium), hypogy nous, pentamerous. Calyx with 5 se-pals, free or slightly connate at base, persi stent. Corolla with 5 petals, united, rotate, tubular or funnel shaped, usually plicate. Androecium with 5 stamens, epipetalous, inserted in corolla tube, filam ents free, bithecous, anthers introrse, dehiscence lon-gitudinal, pollen grains tricolporate or multiporate, filaments often with nectar discs at base. Gynoeciu m with 2 united car-pels, ovary superior, deeply 4-lobed, bilocu-lar, axile plac entation, becoming 4-locular due to false septa, style 1, terminal or gynobasic, stigma 1 or bilobed, capitate or Phylogeny: The family is closely related to Solanaceae, Convolvulaceae and Polem oniaceae with which it shares alternate leaves and actinomorphic flowers and mos tly included under Boraginales next to Solanales (Dahlgren) or Solanales, Convol vulales and Polemoniales (Takhtajan), or under Solanales (Thorne). The family al so shows close affinities with Lamiaceae and Verbenaceae in having gynobasic sty le, 4 lobed ovary becoming 4-locular by false septum and usually schizocarpic fr uit. 658 Plant Systematics D Figure 13.130 Boraginaceae. Cynoglossum glochidiatum. A: Plant with terminal inf lorescences; B: Flower. Heliotropium eichwaldii. C: Portion of plant with terminal helicoid c ymes; D: Flower with bristly calyx; E: Vertical section of flower; F: Fruit with persistent calyx; G: Fruit. Trichodesma indicum. H: Portion of plant in flower; I: Flower with corolla cut away to show androecium; J: Transverse section of ov ary, 4-locular due to false septum; K: Fruit with two of the sepals removed; L: Seed. Cronquist accordingly places Boraginaceae under Lamiales closer to Lamiaceae and Verbenaceae. Although a number of mutigene analyses have been carried out the r esults are not conclusive and as such the family Boraginaceae is still unplaced

under Asterids I in APG II and APweb. The position of both Vahliaceae and Boragi naceae in Euasterids I is uncertain. Vahlia is placed sister to Lamiales, but wi th only 63 per cent bootstrap support (Albach et al., 2001), or is associated mo re specifically with Boraginaceae (Lundberg, 2001). Relationships be-tween Gentianales, Lamiales and Solanale s are also unclear (Albach et al., 2001). APweb recognizes 6 groups within Borag inaceae: four subfamilies Boraginoideae, Heliotropioideae, Cordioideae and Ehret ioideae and two family groups Hydrophyllaceae and Lennoaceae. Thorne (2003, 2006 , 2007) treats last two as independent families and recognizes 5 subfamilies und er Boraginaceae, adding Wellstedioideae as fifth subfamily. * * * * * * * * * * * Major Families of Angiosperms 659 Rubiaceae A. L. de Jussieu Madder family 650 genera, 13,000 species Worldwide in distribution, but mainly distributed in the tropics and subtropics, especially the woody members. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Gentiananae Loasanae Series+/Superorder Inferae+ Lamianae Euasterids I* Order Rubiales Rubiales Rubiales Gentianales Gentianales Salient features: Mainly shrubs and trees, leaves opposite or whorled, often tur ning blackish when dry, stipules interpetiolar, colleters present in leaf axils, inflorescence cymose, flowers pentamerous, stamens 5, ovary inferior. brightly coloured (Mussaenda). Corolla with 4-5 petals, (rarely 8-10), united, t ubular, rotate or funnel-shaped, valvate, imbricate or twisted. Androecium with 4-5 stamens, free, epipetalous, anthers bithecous, dehiscence longitudinal, intr Rubiales

orse, pollen grains usually tricolporate. Gynoecium with 2 (rarely 1-many) unite d carpels, ovary inferior, rarely superior (Pugama) or semi-inferior (Synaptanth a), bilocular (rarely 1-many locules) with 1-many ovules in each chamber, placen tation axile (rarely apical or basal), nectar disc usually present above the ova ry, style slender, stigma capitate or lobed. Fruit a berry, capsule, drupe or sc hizocarp; seeds 1-many, with small embryo, curved or straight, endosperm present or absent. Major genera: Psychotria (1450 species), Galium (410), Ixora (370), Pavetta (360 ), Hedyotis (360), Tarenia (350), Randia (240), Gardenia (240) and Mussaenda (19 0). Description: Trees (Adina, Neolamarckia) or shrubs (Ixora, Gardenia), rarely her bs (Galium), sometimes climbing (Rubia) with hooked hairs, rarely epiphytic (Myr mecodia) with large swellings on roots inhabiting ants, usually with Iridoids, r aphide crystals common. Leaves opposite, with interpetiolar stipules which often become as large a leaves and thus forming whorled arrangeEconomic importance: The family is economically important for being the source o f coffee, quinine and a large number of ment of leaves, simple, entire, often turn- ornamentals. Coffee is obtained from ing blackish when dry, with colleters in leaf roasted seeds of Coffea arabica a nd C. axils. Inflorescence cymose, sometimes canephora. Quinine, a remedy for ma laria is capitate (Adina), or solitary (Gardenia). Flow- derived from several sp ecies of Cinchona. ers bisexual, actinomorphic, rarely Madder (Rubia tinctoria) was formerly cultizygomorphic (Posoqueria) epigynous, some- vated for its red dy e alizarin. Important times dimorphic (Randia). Calyx with 4-5 ornamentals inclu de Gardenia, Ixora, sepals, adnate to ovary, 5-lobed, lobes often Hamelia, Neola marckia (cadamb tree) and very small, one sometimes enlarged and Mussaenda. 660 Plant Systematics

~ E D G

Yi

C L I Figure 13.131 Rubiaceae. Rubia manYith. A: Portion of plant with axillary inflor escences; B: Flower. Randia spinosa. C: Twig showing spines and flowers; D: Flower with corolla remov ed and calyx opened to show gynoecium; E: Corolla opened to show epipetalous sta mens; F: Transverse section of ovary; G: Fruit. Coffea arabica. H: Portion of tw ig with fruits; I: Flower. Mitragyna parvifolia. J: Twig with globose infloresce nces. K: Flower with mitraeform (head-gear) stigma; L: Corolla opened to show an droecium; M: Flower with corolla removed to show style and stigma. Phplogenp: Rubiaceae form a well-defined group which is clearly monophyletic as sup-ported by morphology (Bremer and Struwe, 1992) and rbcL sequences (Bremer et al., 1995). Affinities of the family lie with Gentianales (Dahlgren, APG II, AP web; Thorne places under Rubiales also contain-ing Gentianaceae and related fami lies; Takhtajan under Rubiales next to Gentia-nales under Lamiidae >Gentiananae) o r Dipsacales (Cronquist next to Dipsacales towards the end of Asteridae, Gentianal es towards the beginning), both having opposite leaves and 2 carpels. The separation exclusively on the basis of ovary being inf erior or superior, is slowly being abandoned as has also been done in the case o f certain monocots such as Alliaceae, Agavaceae and Amaryllidaceae. The family i s commonly di-vided into three subfamilies (Thorne, APweb): Cinchonoideae (mainl y woody, raphides absent, seeds with endosperm, heterostyly absent), Ixoroideae (woody, raphides absent, pollination plunger-mechanism as in Asteraceae), Rubioi deae (mainly herbaceous, raphides present in leaves, Major Families of Angiosperms 661 seeds with endosperm, heterostyly common). Molecular data (Fay et al., 2000a) pr ovide support for including Dialy-petalanthus (for-merly placed under Dialy-peta lanthaceae-Thorne, 1999) in Rubiaceae, under sub-family Ixoroideae (Thorne, 2003 , 2006, 2007). Recent molecular studies based on trnL-F and cpDNA data (Rova et al., 200 2) and broadly based molecular data involving several taxa (Bremer et. al., 1999 ) suggest that Cinchonoideae and Ixoroideae are sister taxa. * * * * * * * * * * * Apocynaceae A. L. de Jussieu Dogbane family 480 genera, 4,800 species Mostly tropical and subtropical with a few species in temperate regions. (including Asclepiadaceae Borkh.) Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass

Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Gentiananae Loasanae Series+/Superorder Bicarpellatae+ Lamianae Euasterids I* Order Gentianales Gentianales Apocynales Gentianales Rubiales Gentianales Salient features: Herbs shrubs or climbers, latex milky, leaves opposite or whor led, throat of corolla tube with scales, pollinia absent, ovary superior, fruit a follicle, seed with a tuft of hairs (Asclepiadoideae, formerly Asclepiadaceae separated by pollinia, anthers adnate to stigmatic disc, stigmas united into gyn ostegium). latex usually milky, Iridoids often present. Leaves simple, reduced or absent in some succulent species, opposite (Calotropis, Catharanthus) or whorled (Nerium) , simple, entire, venation pinnate, reticulate, stipules absent, colleters often present at the base of petiole. Inflorescence dichasial or monochasial cyme, ra cemose or umbellate (Calotropis), sometimes solitary (Vinca), or ax-illary cymos e pairs (Catharanthus). Flowers bisexual, actinomorphic, hypogynous, pentamerous , often coronate. Calyx with 5 sepals, distinct or basally connate, imbricate or valvate, often with glands (colleters) at base. Corolla with 5 united petals, t he tube often short campanulate (Calotropis), salver-shaped (Catharanthus) or fu nnel-shaped (Thevetia) and with contorted or valvate lobes. Corona of usually 5 scales or appendages Major genera: Asclepias (220 species), Tabernaemontana (220), Cynanchum (200), C eropegia (140), Hoya (140), Rauvolfia (105), Ervatamia (80), Allamanda (15) and Catharanthus (5). Description: Perennial herbs (Catharanthus), vines (Cryptostegia, Daemia), shrub s (Calotropis, Nerium), rarely trees (Alstonia), often fleshy (Hoya) or cactus-l ike (Stapelia), 662 Plant Systematics

a B A

D ~. G F N Figure 13.132 Apocynaceae. Catharanthus roseus. A: Portion of plant with flowers and fruits; B: Flower bud showing twisted corolla with a long tube; C: Vertical section of f lower from corolla throat showing free epipetalous stamens and calyptrate stigma ; D: Anther with dorsal fixation; E: Transverse section of flower passing throug h ovary showing sepals, corolla tube 2 lateral nectaries and 2 free ovaries; F: Pair of follicles; G: Seed. Nerium indicum. H: Branch with whorled leaves and te rminal inflorescence; I: Corolla opened to show corona of scales and anthers wit h tailed appendages forming single twisted hairy appendage; J: Pair of follicles . Thevetia nerifolia. K: Branch with subopposite and alternate leaves and large funnel-shaped flowers; L: Drupe fruit. Rauvolfia serpentina. M: Portion of plant with inflorescences in flower and fruit; N: Flower buds with twisted corolla; O : Seed. arising from corolla throat (corolline corona: Nerium, Cryptostegia), or from st amens (staminal corona: Calotropis, Asclepias), coro-nal appendages nectariferou s. Androecium with 5 stamens, filaments free (Apocynaceae sensu str.) or connate (Asclepiadoideae except Cry-ptostegia) anthers free with separate pol-len grain s (Apocynaceae sensu str.). In Asclepiadoideae anthers adherent to the stigmatic area forming a 5 angled disc gynostegium (gynandrium), pollen agglutinated within the anther sacs to form waxy pollinia (corpusculum or gland joining together two pollinia one each from adjacent anthers with the help of caudicles to form translator, an adaptation for insect pollination); pollen grain tricolp orate or biporate or triporate. Gynoecium bicarpellary of two apically united ca rpels, unilocular ovaries, marginal placentation, ovules 2 or more, unitegmic, s tyles 2, stigma 1, calyptrate (Catharanthus), dumb-bell shaped (Nerium), or Major Families of Angiosperms 663 Figure 13.133 Apocynaceae, Subfamily Asclepiadoideae. Cryptostegia grandiflora. A: Branch with terminal inflorescence; B: Vertical section of flower with corolloin corona and gynostegium. Calotropis procera. C: Portion of plant with umbellate inflore scences on axillary peduncles; D: Flower with purple-tipped corolla; E: Vertical section of flower with staminal corona and broad gynostegium and free ovaries; F: Translator with 2 pollinia joined by caudicles to common corpusculum; G: Pair of follicles. Pergularia daemia. H: Portion of plant with inflorescence; I: Flo wer; J: Flower with calyx removed and corolla lobes cut to show corona and stami nal tube; K: Stamen and corona in side view; L: Pair of follicles covered with b ristles. 5 lobed and fused with anthers to form gynostegium (Calotropis), carpels sometim es united by ovaries with axile placentation (Thevetia, Allamanda, Carissa). Fru it etaerio of 2 follicles (Nerium, Calotropis), sometimes drupe (Thevetia), caps ule or berry; seeds usu-ally numerous, flattened and comose with long silky hair s. Pollination by insects, helped by special translators in Asclepiadoideae. Dis persal mostly by wind, aided by hairs. (carrion flower), Plumeria (frangipani) and Tabernaemontana are grown as ornamen tals. Nerium and Thevetia are poisonous (can be fatal). Roots of Rauvolfia serpe ntina yield re-serpine used as tranquillizer for patients suffering from schizop hrenia and hypertension. Catharanthus provides antileukaemic drugs. Latex from P lumeria used for healing tooth-ache. Low quality down is obtained from seeds of

several species. Pitcher like leaves and roots of Dischidia chewed with betel. R ubbervine (Cryptostegia) is also as caoutchouc or rubber source. Tubers of Cerop egia are edible. Stem fibre of Calotropis and Economic importance: Nerium (oleander), Catharanthus (Madagascar periwinkle), As cle-pias (milkweed), Hoya (wax plant), Stapelia 664 Plant Systematics Leptadaenia is used for cordage. Asclepias is a livestock poison. 5 subfamilies (Thorne, 2000, 2006, 2007): Rauvolfioideae (Plumerioideae), Apocyn oideae, Periplocoideae, Secamonoideae and Asclepiadoideae. The generic limits ar e not clearly resolved. According to Sennblad and Bremer (2002), both Rauvolfioi deae and Apocynoideae may be quite wildly paraphyletic. The position of the Peri plocoideae as sister to Secamonoideae + Asclepiadoideae is also uncertain (Potge iter & Albert, 2001; Phylogeny: For a long time Apocynaceae was considered to be closely related but distinct from Asclepiadaceae, latter with pollinia, gynostegium and usually stam inal corona (Bentham and Hooker, Engler and Prantl, Hutchinson, Cronquist, Dahlg ren). The family Asclepiadaceae was merged with Apocynaceae by Thorne (1983) and practice was followed by Takhtajan (1987, 1997), Judd et al., (2002), APG II an d APweb. The separa-tion of Asclepiadaceae as distinct family would lead to para phyletic Apocynaceae (Judd et al., 1994; Endress et al., 1996). The family Apocy naceae is appropriately divided into Sennblad & Bremer, 2002). The family is usually placed in Gentianales, but Thorne has merged this order wi th broadly circumscribed Rubiales. Dahlgren and APG II classification prefer the name Gentianales for the broadly circumscribed order. * * * * * * * * * * * Plantaginaceae A. L. de Jussieu Snapdragon family 110 genera 2,000 species Widely distributed from temperate to tropical regions, more diverse in temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae Lamianae Euasterids I* Order

Personales Lamiales Lamiales

Scrophulariales Scrophulariales Lamiales

B & H, Cronquist, Takhtajan and Dahlgren under Scrophulariaceae Salient features: Leaves alternate or oppo-site, stipules absent, flowers zygomo rphic, stamens 4 or 2, anther opening by 2 distinct slits, anthers more or less sagittate at base, carpels 2, ovary superior, 2-chambered, many, fruit a capsule . (130), Linaria (110), Antirrhinum (40), Limnophila (32), Globularia (25), Gratio la (20), Scoparia (20) and Digitalis (20). Major genera: Veronica (350 species), Penstemon (250), Plantago (210), Mimulus Description: Herbs or small shrubs, rarely climbers (Antirrhinum cirrhosum), oft en with phenolic glycosides and triterpenoid saponins, and sometimes with cardia c glycosides, hairs usually simple, when Major Families of Angiosperms 665 J Figure 13.134 Plantaginaceae. Veronica persica. A: Portion of plant with axillar y flowers; B: Flower with 4 sepals and petals each and 2 stamens; C: Gynoecium; D: Fruit wi th persistent calyx and style; E: Seed. Digitalis purpurea. F: Branch with inflo rescence; G: Vertical section of flower; H: Transverse section of ovary with axi le placentation; I: Young anther; J: Anther dehisced through 2 slits. glandular with short discoid head lacking vertical partitions. Leaves alternate or opposite, rarely whorled (Russelia), simple, entire or dentate, venation pinn ate, reticu-late, stipules absent. Inflorescence race-mose: racemes or spikes. F lowers bisexual, zygomorphic, hypogynous. Calyx with 5 sepals, rarely 4 (Veronic a), connate, per-sistent. Corolla with 5 petals, rarely 4 (due to fusion of 2 pe tals as in Veronica), united, usually bilabiate, sometimes with nectar sac or sp ur, lower lip sometimes with a bulge obscuring the throat (personate), lobes imb ricate or valvate. Androecium with usually 4 stamens, didynamous, fifth stamen s ometimes present as a staminode (Penstemon), rarely 2 (Veronica), epipetalous, i nserted in corolla tube, filaments free, anther bilocular, locules distinct, opening by two longitudinal slits, pollen sa cs diver-gent (anther sagittate), pollen grains tricolporate. Gynoecium with 2 u nited carpels, rarely only 1 carpel developed (Globularia), ovary superior, bilo cular, ovules several in each chamber, rarely 1 or 2 (Globularia) unitegmic, axi le placentation, style 1, stigma bilobed, ovary seated on a nectary. Fruit a sep ticidal capsule; seed angular or winged, with curved or straight embryo, endospe rm present. Pollination by insects. Seeds or nutlets dispersed by wind. Economic importance: The family contributes a number of ornamentals such as Digi talis (Foxglove), Mimulus (monkey flower), Antirrhinum (snapdragon), Penstemon 666 Plant Systematics (beardtongue), Veronica (speedwell) and Russelia (firecracker plant). Species of Digi-talis, mainly D. purpurea and D. lanata, are used for the extraction of dr ugs digitalin and digoxin used as cardiac stimulants and ton-ics. The juice of L imnophila indica is used in fevers, tonic and as stomachic. Various spe-cies of Veronica yield glucoside rhinanthis and used for ulcers and burns. dehiscing by 2 slits and absence of vertical partitions in heads of glandular ha irs. Thorne had earlier (1999, 2000) removed all three Plantago, Callitriche, an d Hippuris to separate families Plantaginaceae, Callitrichaceae and Hippuridacea e, respectively, and used the name Antirrhinaceae for the family. Judd et al., ( 1999, 2002), APG II and APweb had, however, combined all the four families under Plantaginaceae, a placement also followed subsequently (2003) by Thorne. Monoph yly of the family is supported by cpDNA characters. Olmstead (2001) suggested th e removal of Calceolaria (with highly saccate corolla) and related genera to sep

arate family Calceolariaceae, a change that has been incorporated in APG II, APw eb and Thorne (2003, 2007). Phplogenp: The family is closely related to Scrophulariaceae and Acanthaceae sha ring the features of zygomorphic flowers, pentamerous flowers, stamens less than 5, bicarpellate superior ovary and fruit a cap-sule. The genera were originally included under Scrophulariaceae from which they differ in having distinct bithe cous anthers * * * * * * * * * * * Lamiaceae Martinov Mint family (=Labiatae A. L. de Jussieu) 264 genera, 6,990 species Worldwide in distribution, largely concentrated in the Mediterranean Region. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae Lamianae Euasterids I* Order Lamiales Lamiales Lamiales Lamiales Lamiales Lamiales B & H as Labiatae, others as Lamiaceae Salient features: Plants aromatic, stem 4- angled, nonglandular hairs multicellu lar, leaves opposite, stipules absent, inflores-cence with cymose lateral cluste rs, com-monly verticillaster, flowers zygomorphic, stamens 2-4, pollen exine not thickened near apertures, carpels 2, ovary superior, 2chambered, finally 4 chambered due to false septum, ovules 4, attached to the si

des of false septa, ovary deeply 4-lobed, style usually gynobasic, forked apical ly with inconspicuous stigmatic region at the tip of each style branch, fruit sc hizocarpic breaking into 4 nutlets. Major Families of Angiosperms 667 @ y A

Figure 13.135 Lamiaceae. Salvia splendens. A: Branch with axillary and terminal inflorescences; B: Vertical section of flower to show bilabiate corolla and stamen with turn-pip e mechanism, gynobasic style and nectary below ovary. Ocimum basilicum. C: Porti on of plant with inflorescences; D: Flower with bilabiate corolla and 4 didynamo us stamens; E: Corolla spread out to show epipetalous stamens of 2 sizes; F: Gyn oecium with gynobasic style and bifid stigma, nectary below the ovary. G: Lamium rhomboideum, plant with inflorescences. Clerodendrum viscosum. H: Branch with t erminal spreading inflorescence; I: Fruit with persistent calyx. Major genera: Salvia (700 species), Clerodendrum (400), Thymus (340), Plectranth us (300), Scutellaria (300), Stachys (300), Nepeta (260), Teucrium (200), Callicarpa (150), Ocimum (150), Lamium (50), Marrubium (40), Mentha (30), Lavandula ( 30) and Tectona (3). suckers (Mentha) or stolons (Ajuga), sometimes green and assimilatory (Hedeoma), usually with glandular hairs, nonglandular hairs when present multicellular. Le aves opposite (rarely alternate), simple or pinnate compound, usually aromatic, sometimes reduced (Hedeoma), stipules absent. Inflorescence verticillaster [two opposite whorls (verticels) of cymose clusters initially biparous and subsequent ly uniparous], arranged in raceme, spike or panicle. Flowers bracteate (Coleus) or ebracteate (Salvia), bisexual, zygomorphic, hypogynous, often Description: Aromatic herbs or shrubs (Rosmarinus, Teucrium), sometimes small (H yptis) or large (Tectona) trees, rarely climb-ers (Scutellaria), stem 4-angled, often with iridoids, phenolic glycosides, sometimes with 668 Plant Systematics bilabiate. Calyx with 5 sepals, united, often bilabiate 1/4 (Ocimum) or 3/2 (Sal via), per-sistent. Corolla with 5 petals, united, usu-ally bilabiate 4/1 (Ocimum ) or 2/3 (Salvia), upper lip sometimes absent (Ajuga), rarely corolla 4-lobed (P ogostemon). Androecium with 2 (Salvia) to 4 (Ocimum) stamens, epipetalous, usual ly didynamous, inserted in corolla tube, sometimes with turn-pipe mechanism (lev er mechanism) as in Salvia (anther lobes separated by a long connective and swin ging like a lever, one anther-lobe sterile, another fertile), filaments free, de -hiscence longitudinal, pollen grains tricolpate or 6-colpate. Gynoecium with 2 united carpels, ovary superior, bilocular, ovules 2 in each chamber, finally 4-l ocular due to false septa with 1 ovule in each cham-ber, anatropous, axile place ntation, ovules attached to the sides of the false septa, ovary 4-lobed, style 1 , gynobasic, rarely terminal (Ajuga), forked apically with inconspicuous stigmat ic region at the tip of each style branch, ovary seated on a nectary disc. Fruit a schizocarp (carcerulus) splitting into 4 nutlets or a drupe or indehiscent 4seeded pod; seed with straight embryo, endosperm minute or absent. Pollination b y insects, lower lip providing landing platform. Disper-sal by birds, wind or wa ter. species of Stachys are edible. Teak (Tectona grandis) is valuable timber, known for its hard and durable wood and extensively cultivated in India and Burma. Phylogeny: The family Lamiaceae is generally considered to be one of the most hi ghly evolved of all dicotyledonous families, and closely related to Vebenaceae.

The circumscription of the family has undergone con-siderable revision with seve ral genera (nearly two-thirds) from older Verbenaceae such as Clerodendrum, Call icarpa, Vitex and Tectona transferred to Lamiaceae (Judd et al., 2002; Thorne, 2 000, 2003; APG II, APweb). The family Lamiaceae is distinguished from Verbenacea e in cymose lateral whorls, ovules attached on sides of false septa, bilobed sty le with small stigmatic region, pollen exine not thickened near apertures, hairs multicellular, strongly bilabiate corolla and usually gynobasic style. Accordin g to Wagstaff et al., 1998 the following 5 clades (subfamilies) are distinct: Ne petoideae (pollen trinucleate, hexacolpate, style gynobasic; myxocarpy; endosper m absent, embryo investing), Lamioideae (laballenic acid in seed oils, em-bryo s ac with micropylar lobe longer and broader than chalazal lobe, style gynobasic), Pogostemonoideae (stamens 4, about the same length), Scutellarioideae (style bi labiate, with rounded lips; seeds tubercuEconomic importance: The family includes several plants used in cooking and flav our-ing such as spearmint (Mentha spicata), pep-permint (M. piperita) thyme (Thy mus vulgaris), sweet basil niazbo (Ocimum basilicum), pot marjoram (Origanum vulga re) and sage (Salvia officinalis). The family is also source of popu-lar perfume s such as lavender (Lavandula angustifolia) and rosemary (Rosmarinus officinalis ). Basil (Ocimum sanctum) is sacred in India. Common ornamentals include sage late), and Teucrioideae [(inc. Ajugoideae) exine with branched to granular colum ellae]. Bootstrap support for the family as circumscribed is 100 per cent (Wagst aff et al., 1998); Congea may be sister to the rest, but some relationships are still in a state of flux. Thorne (2000, 2003, 2007) adds two more subfamilies Sy mphorematoideae (Congea, Sphenodesme, Symphorema) and Prostantheroideae (17 gene ra incl. Chloanthoideae, Tectona), establish-ing Ajugoideae (instead of Teucrioi deae) thus recognizing a total of seven subfamilies. (Salvia), horsemint (Monarda), Molucella, and Coleus. The tubers of few * * * * * * * * * * * Major Families of Angiosperms 669 Verbenaceae Jaume St.-Hilaire Verbena family 36 genera, 1,035 species (including only Verbenoideae) Widely distributed, mainly in tropical regions, also in temperate regions, promi nent in new world. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae

Lamianae Euasterids I* Order Lamiales Lamiales Lamiales Lamiales Lamiales Lamiales Salient features: Plants aromatic, leaves opposite, serrate, stem often angular, non-glandular hairs if present unicellular, flow-ers zygomorphic, in racemes, s pikes or heads, pollen exine thickened near apertures, style simple with bilobed stigma, stigmatic area conspicuously swollen and glandular, ovary with four ovu les, ovules attached to the margin of false septa. gitudinal, pollen grains tricolpate, exine thickened near apertures. Gynoecium w ith 2 united carpels, ovary superior, bilocular, ovules 2 in each chamber, final ly 4-locular due to false septum with 1 ovule in each chamber, unitegmic, axile placentation, ovary not or slightly 4-lobed, style 1, terminal, style simple wit h bilobed stigma, stigmatic area conspicuously swollen and glandular, ovary seat ed on a nectary disc. Fruit a drupe with 2 or 4 pits, or schizocarp splitting in to 2 or 4 nutlets; seed with straight embryo , endosperm absent. Pollination by insects. Dispersal by birds, wind or water. Major genera: Verbena (200), Lippia (180) Lantana (140), Citharexylum (65), Glan du-laria (55), Duranta (28) and Phyla (10). Description: Aromatic herbs (Lippia), shrubs (Lantana), sometimes trees, rarely lianas, sometimes with prickles or thorns, stem usually 4-angled, often with iri doids and phe-nolic glycosides, usually with glandular hairs, nonglandular hairs if present unicel-lular. Leaves opposite, sometimes whorled, simple or sometime s lobed, usually aro-matic, entire to serrate, stipules absent. In-florescence r acemose: racemes, spikes or heads. Flowers bisexual, zygomorphic, hypogynous. Ca lyx with 5 sepals, united, tubular to campanulate, persistent, some-times enlarg ed in fruit. Corolla with 5 pet-als, sometimes appearing 4 due to fusion of two posterior petals, united, weakly bilabiate, lobes imbricate. Androecium with 4 s ta-mens, epipetalous, didynamous, inserted in corolla tube, filaments free, dehi scence lonEconomic importance: The family contributes some ornamentals such as Verbena, La ntana, Duranta, and Glandularia. Lippia (lemon verbena) and Privea are used as h erbal teas or yield essential oils. Verbena officinalis (vervain) is used for a number of herbal remedies including treatment of skin diseases. Phylogeny: The family is closely related to Lamiaceae. The circumscription of th e family has undergone considerable revision with several genera (nearly two-thi rds) from older Verbenaceae such as Clerodendrum, Callicarpa, Vitex and Tectona transferred to Lamiaceae (Judd et al., 2002; Thorne, 2000, 2003; APG II, APweb). The family is now 670 Plant Systematics Figure 13.136 Verbenaceae. Lantana camara. A: Branch with ovoid compact inflores cences on long peduncles; B: Flower with long corolla tube and zygomorphic limb; C: Coroll a spread out to show epipetalous stamens; D: Cluster of fruits; E: Fruit. Verben a officinalis. F: Plant with terminal spikes; G: Corolla with shorter broader tu be and zygomorphic limb; H: Corolla spread out to show epipetalous stamens; I: G ynoecium; J: Transverse section of ovary with 4 one-seeded chambers. Phyla nodif

lora. K: Portion of plant with pedunculate globose inflorescences; L: Flower wit h short broad tube and zygomorphic limb; M: Corolla spread out to show epipetalo us stamens; N: Fruit with persistent calyx; O: Longitudinal section of fruit. circumscribed to include only subfamily Verbenoideae. The traditionally delimite d Verbenaceae are paraphyletic and Lamia-ceae polyphyletic. With narrowly define d Verbenaceae and broadly defined Lamiaceae, both become monophyletic. The famil y is dis-tinguished from Lamiaceae in racemose inflorescence, ovules attached on margins of false septa, style simple with conspicuous bilobed stigma, pollen ex ine thickened near apertures, hairs unicellular, weakly bilabiate corolla and usually terminal styl e. Phryma (Phrymaceae) with one carpel aborted and ovary with single basal ovule may be closely related to Verbenaceae (Chadwell et al., 1992). Avicennia often included in distinct family or broadly circumscribed Verbenaceae is more appropr iately included in Acanthaceae (APweb). Thorne (2003) treats Phrymaceae and Avic enniaceae as distinct families. * * * * * * * * * * * Major Families of Angiosperms 671 Bignoniaceae A. L. de Jussieu Trumpet creeper family 113 genera, 800 species (excluding Paulownia) Widely distributed in tropical and subtropical regions, a few species in tempera te regions, most diverse in northern South America from temperate to tropical re gions, especially diverse in Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae Lamianae Euasterids I* Order Personales Lamiales Lamiales Salient features: Usually woody lianas or trees, leaves usually opposite, often com-pound, sometimes with tendrils, stipules absent, nectaries on leaves, flower Scrophulariales Scrophulariales Lamiales

s zygomorphic, showy, stamens 4, carpels 2, ovary superior, 2-chambered, ovules many, fruit a woody capsule, seeds often winged. epipetalous, inserted in corolla tube, filaments free, anther bithecous, sagitta te, de-hiscence longitudinal, pollen grains some-times in tetrads or polyads. Gy noecium with 2 united carpels, ovary superior, bilocular with axile placentation , rarely unilocular with free-central placentation, ovules many, anatropous, style short, stigma with unequal Major genera: Tabebuia (100 species), lobes. Fruit a woody capsule, occasionally a Arrabidaea (70), Adenocalyma (45), Jacaranda berry or pod; seed winged or fri nged with (40), Spathodia (20), Catalpa (11), Campsis (2) hairs, endosperm absent, cotyled ons deeply and Kigelia (1). bilobed. Pollination by insects. Seeds dispersed by wind. Description: Shrubs, trees or lianas (Bignonia, Campsis), lianas often with char -acteristic secondary growth resulting in lobed or furrowed xylem cylinder, usua lly with iridoids and phenolic glycosides. Leaves usu-ally opposite or whorled, pinnately or palmately compound, sometimes simple (Catalpa), venation pinnate to palmate, re-ticulate, some leaflets often modified into tendrils, stipules abse nt but glands often present at the base of petiole. Inflorescence cymose, raceme or panicle, rarely solitary. Flowers bisexual, zygomorphic, hypogynous, usually showy. Calyx with 5 sepals, connate. Corolla with 5 petals, united, showy, usually bilabiate, sometimes with sac or spur on the lower lip, imbricate. Androeci um with usually 4 stamens, fifth represented by a staminode, rarely 5 (Oroxylum) or 2 (Catalpa), Economic importance: The family contributes several ornamentals such as Spathodi a (African tulip tree), Kigelia (sausage tree), Tabebuia (poui, gold tree), Cres centia (calabash tree) and Tecoma. Common climbers include Bignonia (cross vine) , Campsis, Tecomaria (Cape honeysuckle) and Pyrostegia (flame vine). Tabebuia an d Catalpa are exploited as timbers, mostly for fence posts. Phylogeny: The family is closely related to Scrophulariaceae sharing the feature s of zygomorphic flowers, pentamerous flowers, stamens less than 5, bicarpellate superior ovary and fruit a capsule. The family is monophyletic as evidenced by morphology. Pinnate compound leaves are considered to be ancestral. The genus 672 Plant Systematics Figure 13.137 Bignoniaceae. Incarvillea emodi. A: Plant with terminal raceme and long linear capsules; B: Calyx with minute lobes; C: Stamen with arched filaments and spread ing villous anther lobes; D: Portion of corolla spread to show stamens; E: Seed, linear and fibrillate at both ends. Campsis radicans. F: Branch with flowers; G : Fruit ; H: Winged seed. Paulownia and Schlegelia often included in Bignoniaceae are intermediate between this family and the Scrophulariaceae and as such treated under distinct familie s Paulowniaceae and Schlegeliaceae respectively by Thorne, APG II and APweb. Paulownia is superficially like Catalp a but it has endosperm and lacks the ovary and seed anatomy of Bignoniaceae (Arm strong, 1985; Manning, 2000). * * * * * * * * * * * Major Families of Angiosperms 673 Acanthaceae A. L. de Jussieu Acanthus family 222 genera, 3,565 species Cosmopolitan in distribution, mainly in tropics and warm temperate regions. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae Lamianae Euasterids I* Order Personales Scrophulariales Scrophulariales Lamiales Lamiales Lamiales Salient features: Leaves opposite, stipules absent, flowers zygomorphic, with pr ominent bracts and bracteoles, stamens 2-4, anther lobes unequal in size, carpel s 2, ovary supe-rior, 2-chambered, ovules 4 or more, fruit a capsule, seeds with jaculators. equal in size, sometimes one lobe aborted, dehiscence longitudinal, tapetum glan dular, pollen grains 2-8 aperturate or colpate. Gynoecium with 2 united carpels, ovary superior, bilocular, ovules 2 in each chamber, axile placentation, style 1, stigma bilobed, ovary seated on a nectary. Fruit a loculicidal cap-sule; seed with jaculator or retinaculum a hook-shaped projection of funiculus, embryo lar ge, curved or straight, endosperm absent. Major genera: Justicia (400), Beloprone (300), Barleria (240), Strobilanthus (23 0), Ruellia (190), Dicliptera (140), Thunbergia (140) and Adhatoda (20). Description: Herbs or shrubs (Adhatoda), sometimes small trees (Strobilanthus) o r lianas (Thunbergia), a rarely aquatic herbs (Cardentha), sometimes spiny (Barl eria), usu-ally with anomalous secondary growth, often with Iridoids, alkaloids and diterpenoids, cystoliths often present, nodes unilacunar, vessels with simpl e end-walls. Leaves oppo-site, rarely alternate (Nelsonia, Elytraria), simple, e ntire or dentate, stipules absent. Inflorescence cymose, racemose (usually spike ) or of solitary flowers (Bontia). Flowers bisexual, zygomorphic, hypogynous, wi th prominent bracts and bracteoles. Calyx with 4 (Acanthus) to 5 (Adhatoda) sepa ls, free or united. Corolla with 5 petals, united, usu-ally bilabiate, sometimes nearly regular (Acanthus). Androecium with 2 (Adhatoda) or 4 (Acanthus, Ruellia ) didynamous stamens, rarely 5 (Pentstemon), epipetalous, inserted in corolla tu be, filaments free, anther lobes unEconomic importance: The family includes a large number of ornamentals such as B arleria, Thunbergia, Pachystachys, Eranthemum and Acanthus. Extract from Adhatod a vasica is component of some cough syrups. An extract from the boiled leaves of Acanthus ebracteatus (sea holy) is used as a cough medicine in parts of Malaya, whereas the roots of A. mollis (bear s breech) are used to treat diarrhoea in som

e parts of Europe. Phylogeny: The family is closely related to Scrophulariaceae sharing the feature s of zygomorphic flowers, pentamerous flowers, stamens less than 5, bicarpellate superior ovary and fruit a capsule. The Acanthaceae are distinct in absence of endosperm, anthers opening by two slits and the presence of retinaculum. Thorne had earlier (1999, 2000) recognized 5 subfamilies: Nelsonioideae, Thunbergioidea e, Mendoncioideae, Acanthoideae and Ruellioideae. The first two 674 Plant Systematics 9

G A

AWIF B D G Figure 13.138 Acanthaceae. Adhatoda vasica. A: Branch with flowers in spikes; B: Flower with bilabiate corolla; C: Longitudinal section of flower showing 2 epipetalous stame ns; D: Capsule with persistent calyx. Peristrophe bicalyculata. E: Branch with f lowers; F: Flower with bilabiate corolla; G: Vertical section of flower. Barleri a prionitis. H: portion of branch with spines at nodes and axillary clusters of flowers; I: Corolla tube opened to show stamens, corolla limb cut away; J: Spiny calyx and bracteoles; K: Transverse section of ovary. Blepharis maderaspatensis . L: Portion of branch with flowers; M: Flower; N: Corolla tube opened to show e pipetalous stamens, corolla limb partly cut away; O: Longitudinal section of gyn oecium. include aberrant genera. Nelsonoideae with sometimes alternate leaves, presence of endosperm and absence of retinacula may represent a paraphyletic basal group within the family. Nelsonioideae have often been placed in Scrophulariaceae s. l . or considered intermediate between Scrophulariaceae and Acanthaceae, but they ar e placed sister to rest of Acanthaceae s. l. in Hedren et al., (1995). According to Scotland and Vollesen (2000), the absence of retinacula or cystoliths, descending cochlear aestivation (i.e. the adaxial petals overlapping the abaxial petals in bud) are likely to b e plesiomorphies. Acanthoideae are clearly monophyletic (Scot-land, 1990) and ch aracterized by the absence of cystoliths, nodes not swollen, colpate pollen and monothecous anthers. In Mendoncioideae, one of the carpels is often aborted, fru it is a drupe and style bifid. Mendoncioideae Major Families of Angiosperms 675 and Ruellioideae have subsequently been merged under Thunbergioideae and Acantho ideae, respectively (APweb and Thorne, 2003). APweb includes Avicennioi-deae as fourth subfamily, stressing that the position of Avicenniaceae within Acantha-ce ae s.l. is fairly well established; it shows a rather weakly supported sister gr oup relation-ship with Thunbergioideae (Schwarzbach & McDade, 2002). This placement based on molecular evidence is also supported by a

rticulated nodes, inflorescence structure, flowers with bract and 2 bracteoles, a reduction in number of ovules and absence of en-dosperm (Judd et al., 2002). T horne who had earlier (2000, 2003) treated Avicenniaceae as distinct family has also finally (2007) relegated it to the subfamily level like APweb. * * * * * * * * * * * Scrophulariaceae A. L. de Jussieu Figwort family 42 genera, 1,460 species Widely distributed from temperate to tropical regions, especially diverse in Africa. Placement:

B & H Cronquist Takhtajan Dahlgren Thorne APG II / (APweb) Division Magnoliophyta Magnoliophyta Class Dicotyledons Magnoliopsida Magnoliopsida Magnoliopsida Magnoliopsida Subclass Gamopetalae Asteridae Lamiidae Magnoliidae Lamiidae Series+/Superorder Bicarpellatae+ Lamianae Lamianae Lamianae Euasterids I* Order Personales Scrophulariales Scrophulariales Lamiales Lamiales Lamiales Salient features: Leaves alternate or oppo-site, stipules absent, flowers zygomo rphic, anther commonly opening by single slit, car-pels 2, ovary superior, 2-cha mbered, ovules many, fruit a capsule. whorled, simple, entire or dentate, venation pinnate, reticulate, stipules absen t. Inflorescence racemose: racemes or spikes. Flowers bisexual, zygomorphic, or almost actinomorphic, hypogynous. Calyx with 3-5 sepals, connate, persistent. Co rolla with 4 5 petals, united, usually bilabiate, or with narrow tube broadening upwards, sometimes with nectar sac or spur, imbricate. Androecium with usually 5 stamens, rarely 4 or 2, epipetalous, inserted in corolla tube, filaments free, sometimes hairy (Verbascum), anther bilocular, anther sac confluent and opening by single slit right angles to filament, anther base not sagittate, pollen Major genera: Verbascum (360 species), Scrophularia (230), Selago (150), Sutera (140), Budleja (100), Manulea (55) and Nuxia (30). Description: Herbs or small shrubs (Budleja), often with Iridoids, hairs usually

simple, when glandular with short discoid head com-posed of many cells and with vertical parti-tions. Leaves alternate or opposite, rarely 676 Plant Systematics 3

Figure 13.139 Scrophulariaceae. Scrophularia elatior. A: Portion of plant with t erminal inflorescence; B: Flower with long-exserted stamens and style. Verbascum chinense. C: Lo wer part of plant with basal and lower cauline leaves; D: Upper part of inflores cence; E: Flower; F: Corolla spread to show epipetalous stamens; G: Stamen with glandular hairy filament; H: Flower after removal of corolla and one lobe of cal yx to show gynoecium; I: Capsule with persistent calyx. grains tricolporate. Gynoecium with 2 united carpels, ovary superior, bilocular, ovules several to 1 (Selago) in each chamber, axile placentation, style 1, stig ma bilobed, ovary seated on a nectary. Fruit a septicidal capsule, or schizocarp with two nutlets (Selago); seed with curved or straight embryo, endosperm prese nt. Pollination by insects. Seeds or nutlets dispersed by wind. and Nuxia are commonly grown as ornamentals. Phylogeny: The family is closely related to Acanthaceae sharing the features of zygomorphic flowers, pentamerous flowers, stamens less than 5, bicarpellate supe rior ovary and fruit a capsule. Scrophulariaceae are distinct in presence of end osperm, anthers opening by single slit and the absence of retinaculum. The gener a including Veronica, Linaria, Antirrhinum, Digitalis, etc. formerly included un der Scrophulariaceae Economic importance: The family has lit-tle economic importance. Verbascum is so metimes grown as ornamental. Budleja Major Families of Angiosperms 677 have been variously separated under Antirrhinaceae (Thorne, 1999, 2000) or Plant aginaceae (Judd et al., 2002, APG II and APweb, Thorne, 2006, 2007). The Budleja ceae and Selaginaceae have been merged with Scrophulariaceae in these sys-tems. Monophyly of Scrophulariaceae is clearly supported by morphology, rbcL and ndhF sequences (Olmstead and Reeves, 1995). Verbascum and Scrophularia form a clade characterised by hairy filaments, endosperm development and distinctive se eds and are sister to rest of genera. Selago and relatives (former Selaginaceae) form a clade based on uniovulate locules and achene-like fruits. Budleja is ver y much paraphyletic, but several lines of evidence place it here (Maldonado de M agnano, 1986b); Teedia and Oftia have strong support as the sister group to Budl eja s. l. (Wallick et al. 2001, 2002). * * * * * * * * * * * References Abrams, L. (1923-1960). An Illustrated Flora of Anderberg, A. A., C.-I. Peng, I. Trift and the Pacific States. 4 vols Vol. 4 by Roxana M. Kllersj. (2001). The Stimpsonia problem; evidence from DNA sequences of plastid Ferris). Stanford Univ. Press, Stanford, CA. Adanson, M. (1763). Familles des plantes, Paris, genes atpB, ndhF and rbcL. Bot. Jahrb. Syst. 2 vols. 123: 369-376. Aguilar-Ortigoza, C. J. and V. Sosa. 2004. The Anderson, E. (1940). The concept of the genus. evolution of toxic phenolic compounds in a group of Anacardiaceae genera. Taxon II. A survey of modern opinion. Bull. Torrey Bot. Club 67: 363-369. 53: 357-364. APG [= Angiosperm Phylogeny Group] II. 2003.

Albach, D. C., P. S. Soltis, D. E. Soltis and An update of the Angiosperm Phylogeny Group classification for the orders and fa milies of flowering plants: APG II. Bot. J. R. G. Olmstead. (2001). Phylogenetic analy-sis of Asterids based on sequences of four genes. Ann. Missouri Bot. Gard. 88: 163-212. Linn. Soc. 141: 399-436. 2001. Applequist, W. L. and R. S. Wallace. (2001). Albert, V. A., A. Backlund, K. Bremer, M. W. Phylogeny of the Portulacaceous Cohort Based on ndhF Sequence Data. Syst. Bot. Chase, J. R. Manhart, B. D. Mishler and K. C. Nixon. (1994). Functional constrai nts and rbcL evidence for land plant phylogeny. Ann. 26: 406-419. Applequist, W. L., W. L. Wagner, E. A. Zimmer, Missouri Bot. Gard. 81: 534-567. and M.Nepokroeff. (2006). Molecular evidence resolving the systematic position o f Hectorella (Portulacaceae). Syst. Bot. 31: 310-319. Alston, R. E. and B. L. Turner. (1963). Natural hybridisation among four species of Baptisia ( Leguminosae). Amer. J. Bot. 50:15 9-173. APG [= Angiosperm Phylogeny Group]. 1998. An Alverson, W. S., B. A. Whitlock, R. Nyffeler, and ordinal classification for the families of flowering plants. Ann. Missouri Bot. Gard. D. A. Baum. (1998). Phylogeny of Core Malvales: Evidence from ndhF sequence data . 85: 531-553. Amer. J. Bot. 86(6) Suppl.: 112. Arber, E. (1934). The Gramineae: a study of cereal, bamboo, and grass. Cambridge, England. Arber, E. (1938). Herbals: Their Origin and EvoluAmbrose, B. A., D. R. Lerner, P. Ciceri, C. M. Padilla, M. F. Yanofsky and R. J. Schmidt. (2000). Molecular and genetic analyse s of the Silky1 gene reveal conservation in floral or-gan specification between eudicots and tion (2nd ed.). Cambridge Univ. Press. Arber, E. and J. Parkins. (1907). On the Origin of Angiosperms. Bot. J. Linn. Soc. 38: 29-80. monocots. Molecular Cell 5: 569-579. Armstrong, J. E. (1985). The delimitation of Anderberg, A. A., X. Zhang and M. Kllersj. Bignoniaceae and Scrophulariaceae based on floral anatomy, and the placement of problem genera. Amer. J. Bot. 72: 755-766. (2000). Maesaceae, a new primuloid family in the order Ericales s. l. Taxon 49: 183-187. 680 Plant Systematics Arnold, C. A. (1948). Classification of gymnoBate-Smith, E.C. (1962). The phenolic constitusperms from the point of view of ents of plants and their taxonomic signifiPalaeobotany. Bot. Gaz. 110: 2-12. cance. J. Linn. Soc. Bot. 58: 95-173. Ashlock, P. H. (1971). Monophyly and related Bate-Smith, E.C. (1968). The phenolic constituterms. Syst. Zool. 20: 63-69. ents of plants and their taxonomic signifiAshlock, P. H. (1979). An evolutionary cance. J. Linn. Soc. Bot. 60: 325-383. Systematits s view of classification. Syst.

Bauhin, C. (1596). Phytopinax seu enumeratic Zool. 28: 441-450. plantarum Basel. Asmussen, C. B., W. J. Baker, and Dransfield. Bauhin, C. (1623). Pinax theatri botanici. Basel. (2000). Phylogeny of the palm family (Arecaceae) based on rps16 intron and trnLtrnF plastid DNA sequences. Pp. 525-535, in K. L. Wilson and D. A. Morrison (Eds .), Monocots: Systematics and Evolution. CSIRO, Baum, B. R. (1977). Oats: Wild and Cultivated. A Monograph of the Genus Avena L. (Poaceae). Minister of Supply and Services, Otta wa. Baum, H. (1949). Der einheitliche Bauplan der Collingwood. Angiospermen gynzeen und die Homologie ihrer fertilen Abschnitte. Bot. Jahrb. Sys t. Axelrod, D. I. (1970). Mesozoic paleogeography Pflanzen. 96: 64-82. and early angiosperm history. Bot. Rev. 36: Baum, H. and W. Leinfellner. (1953). Die 277-319. Peltationsnomeklatur der Karpelle. Bot. Babcock, E. B. (1947). The genus Crepis pt. 1. Jahrb. Syst. Pflanzen. 100: 424-426. The taxonomy, phylogeny distribution and evolution of Crepis. Univ. Calif. Publs . Bot. Bayer, C., M. F. Fay, A. Y. De Bruijn, V. 21: 1-197. Savolainen, C. M. Mortan, K. K. Kubitzki, W. S. Alverson and M. W. Chang. (1999) . Sup-port for an expanded family concept of Malvaceae within a recircumscribed order Malvales: a combined analysis of plastid atpB and rbcL DNA sequences. Bot. J. Linn. Soc. Bailey, I. W. (1944). The development of vessels in angiosperms and its significance in morphological research. Am. J. Bot. 31: 421-428. Bailey, L.H. (1949). Manual of Cultivated Plants (rev. Ed.). Macmillan, New York. 129: 267-303. Bailey, I. W. and B. G. L. Swamy. (1951). The Behnke, H. D. (1965). ber das phloem der conduplicate carpel of dicotyledons and its initial trend of specialization. Ame r. J. Bot. Dioscoreaceen unter besonderer 38: 373-379. Berficksichtigung ihrer phloembecken. II. Mitteilung: Elektronenoptische untersu chungen zur feinstruktur des phloembeckens. Z. Pflanzenphysiol. 53: 214 Barber, H. N. (1970). Hybridization and evolution of plants. Taxon 19: 154-160. Barthlott, W. (1981). Epidermal and seed sur244. face characters of plants: Systematic appli-cability and some evolutionary Aspec ts. NorBehnke, H. D. (1976). Ultrastructure of sievedic J. Bot. 1: 345-355. element plastids in Caryophyllales (Centrospermae); evidence for the delimitaBarthlott, W. (1984). Microstructural features of tion and classification of the order. Plant Syst. seed surfaces. In V. H. Heywood and D. M. Moore (Eds.). Current concepts in Plan t Tax-onomy. Systematic Association special

Evol. 126: 31-54. Behnke, H. D. (1977). Transmission electron Volume No. 25: 95-105. microscopy and systematics of flowering plants. In: K. Kubitzki. Flowering Plant s- Evolution and Classification of Higher Categories. Barthlott, W. and D. Froelich. (1983). Mikromorphologie und Orientierungsmuster epicuticularer Wachs-Kristalloide: Ein neues systematisches Merkmal bei Monocotylen. Plant Syst. Evol. Suppl. 1, 155-178. Behnke, H. D. (1997). Sarcobataceae a new family of Caryophyllales. Taxon 46: 49 5-507. Pl. Syst. Evol. 142: 171-185. Behnke, H. D. and W. Barthlott (1983). New Barthlott, W. and G. Voit. (1979). MikromorphoEvidence from the ultrastructural and micromorphological fields in Angiosperm cl aslogie der Samenschalen und Taxonomie der Cactaceae Ein raster-elektronem-microsc osification. Nordic. J. Bot. 1: 341-460. pischer uberblick. Pl. Syst. Evol. 132: 205-229. Bate-Smith, E.C. (1958). Plant phenolics as Belfod, H. S. and W. F. Thomson (1979). Single taxonomic guides. Proc. Linn. Soc. Lond. 169: copy DNA homologies and phylogeny of Atriplex. Carnegie Inst. Wash Year Book 78: 217-223. 198-211. References 681 Belk, E. (1939). Studies in the anatomy and morBordet, J. (1899). Sur l agglutination et la dissophology of the spikelet and flower of the Gramineae. Thesis (Ph.D.). Cornell Uni v. lution des globules rouges par le serum d animaux injectes de sang defibrin. Ann. Bell, G. A. (1971). Comparative Biochemistry of Inst. Pasteur 13: 225-250. non-protein amino acids. In: J. B. Harborne, D. Boulter and B. L. Turner (Eds.) Chemotax-onomy of Leguminosae. Academic Press, LonBoulter, D. (1974). The use of amino acid don, pp. 179-206. sequence data in the classification of higher plants. In: G. Bendz and J. Santes son (Eds.). Chemistry and Botanical Classification, Nobel Symposium, 25, pp. 211 -216. Acad. Press, Bensel, C.R. and B. F. Palser. (1975a). Floral anatomy in the Saxifragaceae sensu lato. II. Saxifragoideae and Iteoideae. Amer. J. Bot. London/New York. Bramwell, D. (1972). Endemism in the Flora of 62: 661-675. Canary Islands. In D. H. Valentine (ed.) Taxonomy, Phytogeography and Evolution 141-159. Bensel, C.R. and B. F. Palser. (1975b). Floral anatomy in the Saxifragaceae sensu lato. III. Kirengeshomoideae, Hydrangeoideae and Academic Press, London. Bremer, B. and L. Struwe. (1992). Phylogeny of Escallonioideae. Amer. J. Bot. 62: 676-687. the Rubiaceae and Loganiaceae: Congruence or conflict between morphologican and molecular data? Amer. J. Bot. 79: 1171-1184. Benson, L. (1957). Plant Classification. Oxford and IBH Co., New Delhi. Bentham, G. (1858). Handbook of British Flora.

Bremer, B., K. Andreasen and D. Olsson. (1995). (7th ed., revised by A. B. Rendle in 1930). Subfamilial and tribal relationships in the Rubiaceae based on rbcL sequence dat a. Ann. Ashford, Kent. Bentham, G. (1863-1878). Flora Australiensis. Missouri Bot. Gard. 82: 383-397. London, 7 volumes. Bremer, B., R. K. Jansen, B. Oxelman, Bentham, G. and J. D. Hooker (1862-83). GenM. Backlund, H. Lantz, and K. -J Kim. (1999). More characters or more taxa for a robust phylogeny-case study from the coffee family era Plantarum. London, 3 vols. Berry, P. E., V. Savolainen, K. J. Sytsma, J. C. Hall, and M. W. Chase. (2001). Lissocarpa is sister to Diospyros (Ebenaceae). Ke w Bull. (Rubiaceae). Syst. Bio. 48: 413-435. Bremer, B., K. Bremer, N. Heidari, P. Erixon, 56: 725-729. A. A. Anderberg, R. G. Olmstead, M. Kllersj, and E. Barkhordarian (2002). Phylogen etics of asterids based on three coding and three non-coding chloroplast DNA mar kers and the utility of non-coding DNA at higher taxonomic levels. Molecular Bessey, C. E. (1915). Phylogenetic taxonomy of flowering plants. Ann. Mo. Bot. Gard. 2: 109-164. Bhandari, M. M. (1978). Flora of the Indian Desert. Sc. Publ., Jodhpur. Bhattacharya, B. and B. M. Johri. 1998. FlowerPhylogenetics and and Evolution 24: 274-301. ing Plants: Taxonomy and Phylogeny. Narosa Bremer, K. (2002). Gondwanan evolution of the Publishing House, New Delhi. grass alliance of families (Poales). Evolution Blackith, R. E. and R. A. Reyment. (1971). Multi56: 1374-1387 variate Morphometrics. Acad. Press, London. Bremer, K. and H. -E. Wanntorp. (1978). PhyloBlake, S. F. (1961). Geographical Guide to the Flogenetic systematics in botany. Taxon 27: ras of the World. Part II. Misc. Publ. 797. U. 317-329. S. Dept. Agr., Washington, D. C. Bremer, K. and H. -E. Wanntorp. (1981). The claBlake, S. F. and A. C. Atwood. (1941). Geographidistic approach to plant classification. In Advances in Cladistics, 1, V. A. Fun k and D. R. Brroks (eds), 87-94, New York Botanical cal Guide to the Floras of the World. Part I. Misc. Publ. 401. U. S. Dept. Agr., Washington, D. C. Garden, New York. Blakeslee, A. F. , A. G. Avery, S. Satina and J. Bremer, K., A. Backlund, B. Sennblad, Rietsama. (1959). The genus Datura. Ronald U. Swenson, K. Andreasen, M. Hjertson, J. Lundberg, M. Backlund, and B. Bremer. (2001). A phylogenetic analaysis of 100+ genera and 50+ families of euasterids b ased on morphological and molecular data with notes on possible higher level mor phological synapomorphies. Plant Syst. Evol. 229: 137-169. Press, New York. Bogler, D. J. and B. B. Simpson. (1996). Phylogeny of Agavaceae based on ITS rDNA se-

quence variation. Amer. J. Bot. 83: 1225-1235. Bonnett, H. T. and E. H. Newcomb. (1965). Polyribosomes and cisternal accumulations in root cells of radish. J. Cell Biol. 27: 4 23-432. 682 Plant Systematics Brenner, G. H. and I. Bickoff. (1992). Palynology Caesalpino, A. (1583). De plantis libri. Florentiae. and age of the Lower Cretaceous basal Kurnub Group from the coastal plain to the northern Negev of Israel. Palynology 16: 137Cain, A. J. and G. A. Harrison. (1958). An analysis of the taxonomists judgement of affinity. Proc. Zool. Soc. Lond. 131: 85-98. 185. Cain, A. J. and G. A. Harrison. (1960). Phyletic Brenner, G. J. (1963). Spores and Pollen of weighting. Proc. Zool. Soc. Lond. 135: 1-31. Potomac Group of Maryland. Maryland Depart-ment of Geology, Mines, Water Resouce s BulleCameron, K. M. (2002). Intertribal relationships tin 27:1-215. within Orchidaceae as inferred from analyses of five plastid genes. P. 116, in B otany 2002: Botany in the Curriculum, Abstracts. Brenner, G. L. (1996). Evidence for the earliest stage of angiosperm pollen evolution. A paleoequatorial section from Israel. In: D. W. Taylor and L. J. Hickey (Eds.) Flowering Plant Origin, Evolution and Phyl ogeny. [Madison, Wisconsin.]. Cameron, K. M. . (2004). Utility of plastid psaB gene sequences for investigating intrafmilial relationships within Orchidaceae. Mol. Phyl. Chapman & Hall Inc., New York, pp. 91-115. Evol. 31: 1157-1180. Brown, R. (1827). Character and description of Cantino, P. D. (2000). Phylogenetic nomenclaKingia, a new genus of plants found on the southwest coast of New Holland with o bservations on the structure and its unimpregnated ovulum on female flower of Cycadeae and Coniferae in P. P. King s na r-rative of a survey of Intertropical and westerns coasts of Australia, performe d between 1818 ture: addressing some concerns. Taxon 49: 85-93. Carlquist, S. (1987). Presence of vessels in Sarcanda (Chloranthaceae); comments on vessel origin in angiosperms. Amer. J. Bo t. 64: 1765-1771. Carlquist, S. (1996). Wood anatomy of primitive and 1822, p 534-565. John Murray, London. Angiosperms: New perspective and syntheses. In: D. W. Taylor and L. J. Hickey (E ds.) Flowering Plant Origin, Evolution and Phylogeny. Chapman & Hall Inc., New Y ork, pp. Brown, N. E. (1920). New and old species of Mesembranthemum, with critical notes. Journ. Linn. Soc. Bot. 45: 53-140. Brown, R. W. (1956). Palmlike plants from the 68-90. Delores Formation (Triassic) in southwest-ern Colorado. U. S. Geological Survey ProfesCarpenter, J. M. (2003). Crique of pure folly. Bot.

Rev. 69(1): 79-92. sional Paper 274: 205-209. Cataln, P, E. A. Kellogg and R. G. Olmstead. Brunfels, O. (1530). Herbarium vivae eicones. (1997). Phylogeny of Poaceae subfamily Pooideae based on chloroplast ndhF gene s equences. Mol. Phylog. Evol. 8: 150-166. Cavalier-Smith, (1981). Eukaryotic king doms, Argentorati, 3 tomes. Bruneau, A., F. Forest, P. S. Herendeen, B. B. Klitgaard and G. P. Lewis. (2001). Phyloge-netic relationships in the Caesalpini oideae (Leguminosae) as inferred from chloroplast trnL intron sequences. Syst. B ot. 26: 487seven or nine? BioSystems 14: 461-481. Cavalier-Smith, (1998). A revised six-kingdom system of life. Biol. Rev. 73: 203-266. 514. Cavalier-Smith, (2000).A revised six kingdom Caddick, L. R. C. A. Furness, P. Wilkons and system of life. Biological Reviews. M. W. Chase. (2000). Yams and their allies: Systematics of Dioscoreales. Pp. 475 -487, in Wilson, K. L., & Morrison, D. A. (eds), Mono-cots: Systematics and Evol ution. CSIRO, Cavalier-Smith, . (2004). Only six kingdoms of life. Proceedings of the Royal Society B: Biological Sciences 271: 251-1262. Chamberlain, C. J. (1935). Gymnosperm structure Collingwood. and evolution. University of Chicago Press, Caddick, L. R. C. A. Furness, P. Wilkons, T. A. Chicago. J. Hedderson and M. W. Chase (2002a). Phylogenetics of Dioscoreales based on com -bined analyses of morphological and molecuChandra, S. and K. R. Surange. (1976). Cuticular lar data. Bot. J. Linnean Soc. 138: 123-144. studies of the reproductive organs of Glossopteris, Part I: Dictyopteridium feis manteli sp. Nov. attached on Glossopteris Caddick, L. R. , P. Wilkin,P. J. Rudall, T. A. J. tenuinervis. Palaeontographia B156: 87-102. Hedderson and M. W. Chase. (2002b). Yams reclassified: A recircumscription of Di oscoreaceae and Dioscoreales. Taxon 51: 103-114. Chandler, G. T, and G. M. Plunkett. (2003). The phylogenetic placement and evolutionary significance of the polyphyletic subfami ly Hydrocotyloideae (Apiaceae). P. 75 in Botany References 683 2003: Aquatic and Wetland Plants: Wet and Clark, P. J. (1952). An extension of the coeffiWild. [Mobile, Alabama.] cient of divergence for use with multiple Chandler, G. T, and G. M. Plunkett. (2004). Evocharacters. Copeia 2: 61-64. lution in Apiales: Nuclear and chloroplast markers together in (almost) perfect har-mony. Botanical J. Linnean Soc.(cited in Clark, L. G., W. Zhang and J. F. Wendel. (1995). A phylogeny of grass family (Poaceae) based on ndhF sequence data. Syst. Bot. 20 : 436 APweb) 460. Chapman, M. (1936). Carpel anatomy of Berberidaceae. Amer. J. Bot. 23: 340-348. Clark, L. G. & J. K. Triplett. (2006). Phylogeny of

Chappill, J. A. (1994). Cladistic analysis of the Bambusoideae (Poaceae): An update. P. 212, in Botany 2006 - Looking to the F uture, Conserving the Past. [Abstracts: Botanical SoLeguminosae. The development of an explicit hypothesis. In M. D. Crisp and J. J. Doyle (Eds.) Advances in legume systematics, part ciety of America, etc.] Clifford, H. T. (1977). Quantitative studies of in7., 1-9. Royal Botanic Gardens, Kew. ter-relationships amongst the Liliatae. In: K. Kubitzki. Flowering Plants-Evolut ion and Classification of Higher categories. Pl. Syst. Chase, M. W., D. E. Soltis, R. G. Olmstead et al. (1993). Phylogenetics of seed plants: an analysis of nucleotide sequences fr om the plastid gene rbcL. Ann. Missourie. Bot. Gdn. Evol. Suppl. 1, 77-95. Colless, D. H. (1967). An examination of certain 80: 528-580. concepts in phenetic taxonomy. Syst. Zool. Chase, M. W., M. R. Duvall, H. G. Hills, et al. 16: 6-27. (1995a). Molecular systematics of Lilianae: In M. J. Rudall et al. (Eds.) Monoco tyledons: Systematics and Evolution, 109-137. Royal Collett, H. (1921). Flora Simlensis, 2nd ed. Thacker, Spink, Calcutta. Botanic Gardens, Kew. Constance, L. (1964). Systematic Botany- an unending synthesis. Taxon. 13: 257-2 73. Chase, M. W. , D. W. Stevenson, P. Wilkin and Coode, M. J. E. (1967). Revision of Genus P. Rudall. (1995b). Monocot systematics: A combined analysis. In M. J. Rudall et al. (Eds.) Monocotyledons: Systematics and EvoValerianella in Turkey. Notes Roy. Bot. Gard. Edinb. 27: 219-256. Copeland, H. F. (1938). The kingdoms of organlution, 685-730. Royal Botanic Gardens, Kew. isms . Quart. Rev. Biol. 13: 383 420. Chase, M. W., D. E. Soltis, P. G. Rudall, M. F. Copeland, (1956). The Classification of Lower OrFay, W. J. Hahn, S. Sullivan, J. Joseph, M. Molvray, P. J. Kores, T. J. Givnish, K. J. Sytsma, and J. C. Pires. (2000). Higher level systematics of monocotyledo ns: An assess-ment of current knowledge and a new clas-sification. In K. L. Wils on and D. A. Morrison (Eds.) Systematics and evolution of monocots. Proceedings of the 2nd International monocot ganisms, Palo Alto: Pacific Books. Core, E. L. (1955). Plant Taxonomy. Prentice-Hall, Englewood Cliffs. Corner, E. J. H. (1946). Centrifugal Stamens. Journ. Arn. Arbor. 27: 423. Cornet, B. (1986). Reproductive structures and leaf venation of Late Triassic angiosperm, Sanmiguelia lewisii. Evol. Theory 7: 231-309. symposium. Melbourne: CSIRO, 3-16. Cornet, B. (1989). Reproductive morphology and Chase, M. W., S. Zmarzty, M. D. Lled, K. J. biology of Sanmiguelia lewisii and its bearing on angiosperm evolution in Late T riassic. Wurdack, S. M. Swensen, and M. F. Fay. (2002). When in doubt, put it in Flacourt iaceae: A molecular phylogenetic analysis based on plastid rbcL DNA seEvol. Trends Plants 3: 25-51. Cornet, B. (1993). Dicot-like leaf and flowers from quences. Kew Bull. 57: 141-181.

the Late Triassic tropical Newark Supergroup rift zone, U. S. A. Modern Geol. 19 : 81-99. Chatton, E. 1937. Titres et travaux scientifiques, Cornet, B. (1996). A New Gnetophyte from the Sette, Sottano, Italy. Late Carnian (Late Triassic) of Texas and its bearing on the origin of the angio sperm carpel and stamen. In: D. W. Taylor and L. J. Hickey (Eds.) Flowering Plan t Origin, EvoluCheadle, V.I. (1953). Independent origin of vessels in the monocotyledons and dicotyledons. Phytomorphology 3: 23-44. Chen, Z.-D., S. R. Manchester, & H. -Y. Sun. tion and Phylogeny. Chapman & Hall Inc., New York, . pp. 32-67. (1999). Phylogeny and evolution of Betulaceae as inferred from DNA sequences, mo rphology, and paleobotany. Amer. J. Bot. 86: 1168-1181. Couper, R. A. (1958). British Mesozoic microspores and pollen grains. Palaeontographica, Abt. B, 103: 75-179. 684 Plant Systematics Crane, P. R., E. M. Friis and K. R. Pedersen. Dahlgren, G. (1991). Steps towards a natural (1995). The origin and early diversification system of the dicotyledons: embryological of angiosperms. Nature 374: 27-33. characters. Aliso 13 (1): 107-165. Crawford, D. J. and E. A. Julian. (1976). Seed Dahlgren, R. (1975). A system of classification protein profiles in the narrow-leaved species of Chenopodium of the Western Unit ed States: Taxonomic value and comparison with distribution of flavonoid compoun ds. of angiosperms to be used to demonstrate the distribution of characters. Bot. No tiser 128: 119-147. Dahlgren, R. (1977). Commentary on a DiagramAm. J. Bot. 63: 302-308. matic Presentation of the Angiosperms. In: K. Kubitzki (ed.). Flowering Plants: Evolution and Classification of Higher Categories. Plant Systematics and Evoluti on Suppl. 1. Crepet, W. L. (1974). Investigations of North American Cycadeoides. The Reproductive bi-ology of Cycadeoidea Palaeontographia 148: 144-169. Springer-Verlag Wien/New York, pp. 253-283. Crepet, W. L., K. C. Nixon & M. A. Gandolfo. Dahlgren, R. (1980). A revised system of classi(2004). Fossil evidence and phylogeny: The age of major angiosperm clades based on mesofossil and macrofossil evidence from fication of angiosperms. Bot. J. Linn. Soc. 80: 91-124. Dahlgren, R. (1983). General aspects of anCretaceous deposits. American J. Bot. 91: giosperm evolution and macrosystematics. 1666-1682 Nordic. J. Bot. 3: 119-149 Cronquist, A. (1968). Evolution and Classification Dahlgren, R. and F. N. Rasmussen. (1983). Monoof Flowering Plants. Houghton Mifflin, New cotyledon evolution: characters and phylogenetic estimation. Evol. Biol. 16: 255 -395. York.

Cronquist, A. (1977). On the taxonomic signifiDahlgren, R., H. T. Clifford and P. F. Yeo (1985). cance of secondary metabolites in An-giosperms. Plant Syst. Evol., Suppl. 1: 179 The Families of Monocotyledons. SpringerVerlag, Berlin. 189. Dahlgren, R., S. Rosendal-Jensen and B. J. Cronquist, A. (1981). An Integrated System of ClasNielsen (1981). A revised classification of the angiosperms with comments on the correlation between chemical and other characters, In: D. A. Young and D. S. Se igler (Eds.). Phytochemistry and Angiosperm sification of Angiosperms.Columbia Univ. Press, New York. Cronquist, A. (1988). Evolution and Classification of Flowering Plants. (2nd ed.). New York Botanical Garden, Bronx, New York. Phylogeny. Praeger, New York, pp. 149-199. Cronquist, A., A. L. Takhtajan and W. Dahlgren, R. in cooperation with B. Hansen, Zimmerman (1966). On the higher taxa of K. Jakobsen and K. Larsen. (1974). Angiospermernes taxonomi, 1. (2 & 3 in 1975, 4 in 1976). In Danish. Kobenhsvn: Akademisk Forlag. Embryobionta. Taaoon 15: 129-134. Cunoud, P. (2002). Introduction to expanded Caryophyllales. pp. 1-4, in Kubitzki, K. (ed.), The Families and Genera of Vascu lar Plants. IV. Flowering Plants. Dicotyledons. Malvales, Capparales and Non-bet alain Caryophyllales. Darlington, C. D. and A. P. Wylie (1955). Chromosome Altas of Flowering Plants. Allen and Unwin, London. Darlington, C. D. and E. K. Janaki-Ammal. Springer, Berlin. (1945). Chromosome Atlas of Cultivated Plants. Cunoud, P, V. Savolainen, L. W. Chatrou, M. Allen and Unwin, London. Powell, R. J. Grayer, & M. W. Chase. (2002). Molecular phylogenetics of Caryophy llales based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences. American J. Darwin, C. (1859). The Origin of Species. London. Daugherty, L. H. (1941). The Upper Triassic Flora Bot. 89: 132-144. of Arizona with a Discussion on its Geological Occurrence. Contributions to Pale ontology 526, Carnegie Institution of Washington. Dahlgren, G. (1989). An updated angiosperm Davis, J. I., and R. Soreng. (1993). Phylogenetic classification. Bot. J. Linn. Soc. 100: 197-203. structure in grass family (Poaceae) as inferred from chloroplast DNA restriction site Dahlgren, G. (1989). The last Dahlgrenogram. System of classification of dicotyledons, pp. 249-260. In: K. Tan [ed.]. The Dav is and Hedge variation. Am. J. Bot. 80: 1444-1454. Davis, P. H. (1960). Materials for the Flora of Festschrift. Edinburgh Univ. Press, Edinburgh, pp. 249-260. Turkey. IV. Ranunculaceae, II. Notes Roy. Bot. Gard. Edinburgh, 23: 103-161. References 685 Davis, P. H. and V. H. Heywood. (1963). Principles lution within the apioid superclade. Canad. of Angiosperm Taxonomy. Oliver and Boyd,

J. Bot. 78: 506-528. London. Doyle, J. A. (1969). Cretaceous angiosperm polde Candolle, A. P. (1813). Theorie elementaire de len of Atlantic Coastal Plain and its evolutionary significance. J. Arnold Arbor etum. 50: la botanique. Paris. de Candolle, A. P. (1824-73). Prodromus systematis naturalis regni vegetabilis. Paris, 17 vols. de Jussieu, A. L. (1789). Genera plantarum. Paris. 1-35. Doyle, J. A. (1978). Origin of angiosperms. Ann. Rev. Ecol. Systematics. 9: 365-392. de Queiroz, K. and J. Gauthier. (1990). PhylogeDoyle, J. A. (2001). Significance of molecular netic taxonomy. Ann. Rev. Ecol. Syst. 23: 449phylogenetic analyses for paleobotanical in-vestigations on the origin of angios perms. 480. de Queiroz, K. and J. Gauthier. (1992). PhylogPalaeobotanist 50: 167-188. eny as central principle in taxonomy: Phylo-genetic definitions of taxon names. Syst. Doyle, J. A. and Donoghue. (1987). The origin of Zool. 39: 307-322. angiosperms: a cladistic approach. In: E. M. Friis, W. G. Chaloner and P. R. Cra ne (Eds.). The Origins of Angiosperms and Their Biological Consequences. Cambrid ge University de Soo, C. R. (1975). A review of new classification system of flowering plants (Angiospermatophyta, Magnoliophytina). Press, U. K, pp. 17-49. Taxon 24(5/6): 585-592. Doyle, J. A. and Donoghue. (1993). Phylogenies Delavoryas. (1971). Biotic provinces and the Juand angiosperm diversification. Paleobiology rassic-Cretaceous floral transition. Proc. N. 19: 141-167. Aer. Paleontol. L: 1660-1674. Doyle, J. A., M. Van Campo and B. Lugardon. Dilcher, D. L. (1979). Early angiosperm repro(1975). Observations on exine structure of Eucommiidites and Lower Cretaceous an giosperm pollen. Pollen and Spores 17: 429 ductions: An introductory report. Rev. Palaeobot. Palynol. 27: 291-328. Dilcher, D. L. and P. L. Crane. (1984). An early 486. Angiosperm from the Western Interior of Doyle, J. A. and P. K. Endress. (2000). MorphoNorth America. Ann. Missouri Bot. Gard. 71: logical phylogenetic analysis of basal an-giosperms: comparison and combination with molecular data. International Journal of Plant Sciences. 161 (6 suppl.): S1 21-S153. 380-388 Donoghue, M. J. and J. A. Doyle. (1989). Phylogenetic analysis of angiosperms and the re-lationships Hamamelidae. In: Pr. R. C rane and S. Blackmore (Eds.). Evolution, Systemat-ics and Fossil History of the Hamamelidae. Doyle, J. A., H. Sauquet, T. Scharaschkin & A. Claredon Press, Oxford, pp. 17-45. Le Thomas. (2004). Phylogeny, molecular and fossil dating, and biogeographic his

tory of Annonaceae and Myristicaceae (Magnoliales). Int. J. Plant Sci. 165(4 Sup pl.): S55-S67. Donoghue, M. J., T. Eriksson, P. A.Reeves and R. G. Olmstead. (2001). Phylogeny and phy-logenetic taxonomy of Dipsacales, with spe-cial reference to Sinadoxa and Tetradoxa Doyle, J. J. (1994). Phylogeny of legume family: An approach to understanding the origins of nodulation. Ann. Rev. Ecol. Syst. 25: 325-349. (Adoxaceae). Harvard Papers Bot. 6: 459-479. Doyle, J. J., J. A. Chappill, C. D. Bailey and T. Doweld, A. B. (2001). Tentamen Systematis Kajita. (2000). Towards a comprehensive phylogeny of legumes: Evidence from rbcL sequences and non-molecular data. Pp. 1 20, in Herendeen, P. S., & Bruneau, A. (eds), Advances in Legume Systematics, Part 9. Royal Plantarum Vascularium (Tracheophytorum). Moscow: GEOS. Downie, S, R., D. S. Katz-Downie, and M. F. Watson, (2000a). A phylogeny of the flower-ing plant family Apiaceae based on ch loro-plast rpl16 and rpoC1 sequences: Towards a suprageneric classification of s ubfamily Botanic Gardens, Kew. Downie, S. R. , D. S. Katz-Downie and M. F. Apioideae. Amer. J. Bot. 87: 273-292. Watson. (2000). A phylogeny of the flowering plant family Apiaceae based on chlo roplast rpl16 and rpoC1 sequences: Towards a suprageneric classification of subf amily Apioideae. American J. Bot. 87: 273-292. Downie, S. R., M. F. Watson, K. Spalik and D. S. Katz-Downie. (2000b). Molecular systemat-ics of Old World Apioideae (Apiaceae): Rela-tionships among some members of tribe Peucedaneae sensu lato, the placement of several island-endemic species, and resoDu Rietz. (1930). Fundamental units of biological taxonomy. Svensk bot. Tidskr. 24: 333 428. 686 Plant Systematics Duangjai, S., B. Wallnoeffer, R. Samuel, F. Engler, A. and Diels, L. (1936). Syllabus der Munzinger and M. W. Chase, M. W. (2006a). Phylogenetic relationships and infrafa milial classification of Ebenaceae s.l. based on six plastid markers. Pp. 218-21 9, in Botany 2006 - Looking to the Future - Conserving the Past. pflanzenfamilien. 11th ed. Berlin. Engler, A. and Prantl, K. (1887-1915). Die naturlichen pflanzenfamilien. Leipzig, 23 vols. Erbar, C., S. Kusma and P. Leins. (1999). Development and interpretation of nectary or[Abstracts: Botanical Society of America, etc.] gans in Ranunculaceae. Flora 194: 317-332. Duangjai, S., B. Wallnoeffer, R. Samuel, F. Erdtman, G. (1948). Did dicotyledonous plants Munzinger and M. W. Chase, M. W. (2006b). Generic delimitation and relationships in Ebenaceae sensu lato: Evidence from six plastid DNA regions. American J. Bot . 93: 1808-1827. exist in Early Jurrassic time? Geol. Fren. Stockholm Frh. 70: 265-271. Erdtman, G. (1966). Pollen Morphology and Plant Taxonomy. Angiosperms. (An Introduction to Palynology. I.). Hafner Publ. Co., Lo ndon. Dykes, W. R. (1913). The Genus Iris. Evans, R. C. and T. A. Dickinson. (2002). How do Eames, A. J. (1961). Morphology of Angiosperms.

studies of comparative ontogeny and morMcGraw-Hill Book Co., New York. phology aid in elucidation of relationships within the Rosaceae? P. 108, in Bota ny 2002: Botany in the Curriculum, Abstracts. [MadiEckenwalder, J. E. (1976). Re-evaluation of Cupressaceae and Taxodiaceae: A proposed merger. Madroo 23: 237-256. son, Wisconsin.] Ehrendorfer, F. (1968). Geographical and ecologiEvans, R. C., C. Campbell, D. Potter, D. Morcal aspects of infraspecific diiferentiation In: V. H. Heywood (ed.). Modern Met hods in Plant Taxonomy Acad. Press, New York, pp. 261-296. gan, T. Eriksson, L. Alice, S. -H Oh, E. Bortiri, F. Gao, J. Smedmark and M. Ars enault. (2002a). A Rosaceae phylogeny. P. 108, in Botany 2002: Botany in the Cur riculum, AbEhrendorfer, F. (1983). Summary Statement. stracts. [Madison, Wisconsin.] Nord. J. Bot. 3: 151-155. Evans, R. C., T. A. Dickinson, T. A., and C. Eichler, A. W. (1883). Syllabus der Vorlesungen Campbell. (2002b). The origin of the apple subfamily (Maloideae; Rosaceae) is cl arified ber Specielle und Medicinisch-Pharmaceutische Botanik. Leipzig. by DNA sequence data from duplicated Endress, M. E., B. Sennblad, S. Nilsson, L. GBSSI genes. American J. Bot. 89: 1478-1484. Civeyrel, M. W. Chase, S. Huysmans, E. Grafrm and B. Bremer. (1996). A phylogenet ic analysis of of Apocynaceae s. str. And some related taxa in Gentianales: A mu ltidisFairbrothers, D. E. (1983). Evidence from nucleic acid and protein chemistry, in particular serology, in angiosperm classification. Nordic. J. Bot. 3: 35-41. ciplinary approach Opera Bot. Belg. 7: 59-102. Farjon, A. (2005). A Monograph of Cupressaceae Endress, P. K. (1977). Evolutionary Tends in and Sciadopitys. Royal Botanic Gardens, Kew. Hamamelidales-Fagales-Group. In H. Kubitzki, Flowering Plants: Evolution and classification of higher categories. Plant. Syst. Evol. Fan, C. and Q.-Y. Xiang. (2003). Phylogenetic Suppl. 1, 321-347. analyses of Cornales based on 26S rRNA and combined 26S rDNA-MATK-RBCL sequence data. American J. Botany 90: 1357-1372. Engler, A. (1892). Syllabus der Pflanzenfamilien. Farr, E. R., J. A. Leussink and F. A. Stafleu (Eds.). Berlin. (1979). Index Nominum Genericorum (Plantarum). Engler, A. (ed.). (1900-1953). Das Pflanzenreich. Regnum Veg. 100-102: 1-1896. Regni vegetabilis conspectus Im Auftrage der Preus. Akademie der Wissenschaften, Leransgegeben von A. Engler, Berlin. (after Engler s death subsequent volumes, co ntinuFarr, E. R., J. A. Leussink and G. Zijlstra (Eds.). (1986). Index Nominum Genericorum (Plantarum) Supplementum I. Regnum Veg. 113: 1 -126. Fassett, N. C. (1957). A Manual of Aquatic Plants. ing upto 1953 were edited by other authors) Univ. Wisconin Press, Madison.

Engler, A. (H. Melchior and E. Werdermann, Faust, W. Z. and S. B. Jones. (1973). The sysEds.). (1954). Syllabus der pflanzen-familien.12th ed., vol. 1. Gebruder Borntra eger, tematic value of trichome complements in a North American Group of Vernonia Berlin. (Compositae). Rhodora 75: 517-528. Engler, A. (H. Melchior, ed.). (1964). Syllabus der Fay, M. F., P. J. Rudall, S. Sullivan, K. L. Stobart, pflanzenfamilien.12th ed., vol. 2. Gebruder Borntraeger, Berlin. A. Y. de Bruijn, G. Reeves, M. QamaruzZaman, W.-P. Hong, J. Joseph, W. J. Hahn, References 687 J. G. Conran, and M. W. Chase. (2000). Phy-logenetic studies of Asparagales base d on four plastid DNA regions. Pp. 360-371, in K. L. Wilson and D. A. Morrison (Eds.), Monocots: Systematics and Evolution. CSI RO, Sarcopus abberans Gagnepain. Bull. Soc. Bot. Fr. 94: 182-185. Garcke, A. (1972). Illustrierte Flore von Deutschland und Angrenzende Gebiete, 23rd ed. (revised by K. von Weihe ed.). Par ey, Collingwood. Berlin. Fay, M. F., B. Bremer, G. T. Prance, M. van der Garnock-Jones, P. J. and C. J. Webb. (1996). The Bank, D. Bridson and M. W. Chase. (2000a). Plastid rbcL sequence data show Dialy peta-lanthus to be member of Rubiaceae. Kew requirement to cite authors of plant names in botanical journals. Taxon, 45: 285 -286. Gaussen, H. (1946). Les Gymnosperms actuelles Bull. 55: 853-864. et fossiles. Pt. 3. Travaux du Laboratoire Federov, A. A. (ed.). (1969). Chromosome Numbers Forestier, Toulouse. of Flowering Plants. Akad. Nauk SSSR, Gershenzon, J. and T. J. Mabry. (1983). SecondLeningrad. ary metabolites and the higher classification of angiosperms. Nordic J. Bot. 3: 5-34. Fiori, A. and G. Paoletti. (1896). Flora analitica d Italia 1: 1-256. Padova. Geuten, K. E. Smets, P. Schols, Y. -M. Yuan, S. Fishbein, M., C. Hibsch-Jetter, D. E. Soltis and Janssens, P. Kpfer, and N. Pyck. (2004). Conflicting phylogenies of balsaminoid f amilies and the polytomy in Ericales: CombinL. Hufford. (2001). Phylogeny of Saxifragales (Angiosperms, Eudicots): Analysis of a rapid, ancient radiation. Syst. Biol. 50: 817-847. ing data in a Bayesian framework. Mol. Phyl. Fisher, M. J. (1928). Morphology and anatomy of Evol. 31: 711-729. flowers of Salicaceae. I. Amer. J. Bot. 15: 307Geuten, K., A. Becker, K. Kaufmann, P. Caris, 326. S. Janssens, T. Viaene, G. Theien and E. Smets. (2006). Petaloidy and petal ident ity MADS-box genes in the balsaminoid genera Impatiens and Marcgravia. Plant J. 47: 501-518. Fisher, R. A. (1930). Genetical Theory of Natural Selection. Oxford: Claredon Press. Fischer, E. (2004). Balsaminaceae. Pp. 20-25, in Kubitzki, K. (ed.), The Families and Genera of Gifford, E. M. and A. S. Foster.

(1988). MorpholVascular Plants. VI. Flowering Plants. Dicotyledons. Ceslastrales, Oxalidales, Rosales, ogy and evolution of vascular plants, 3rd ed. W. H. Freeman, New York. Cornales, Ericales. Springer, Berlin. Gleason, H. A. (1963). The New Britton and Brown Ford, B. A., M. Iranpour,R. F. C. Naczi, J. R. Starr Illustrated Flora. 3 Vols. Hafner, New York. & C. A. Jerome. (2006). Phylogeny of Carex subg. Vignea (Cyperaceae) based on no n-coding nrDNA sequence data. Syst. Bot. 31: Gornall, R. J. 1989. Anatomical evidence and the taxonomic position of Darmera (Saxifragaceae). Bot. J. Linnean Soc. 100: 173-182 . 70-82. Gottsberger, G. (1974). Structure and function Freudenstein, J. V, and M. W. Chase. (2001). of primitive angiosperm flower-A Discussion. Analysis of mitochondrial nad1b-c intron se-quences in Orchidaceae: Utility and coding Acta Bot. Neerl. 23: 461-471. Gower, J. C. (1966). Some distance properties of of length-change characters. Syst. Bot. 26: latent root and vector methods used in multivariate analysis. Biometrika 53: 325 -338. 643-657. Friedrich, H. C. (1956). Studien ber die Graham, S. W. and R. G. Olmstead. (2000a). natrliche verwandtschaft der Plumbaginales und Centrospermae. Phyton (Austria) 6: 220Evolutionary significance of an unusual chloroplast DNA inversion found in two b asal angiosperm lineages. Curr. Genet. 37: 183-188. 263. Frodin, D. G. (1984). Guide to the Standard Floras of the World. Cambridge Univ. Press. Graham, S. W. and R. G. Olmstead. (2000b). Frost, F. H. (1930). Specialization in secondary Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosp erms. xylem in dicotyledons. I. Origin of vessels. Botanical Gazette 89: 67-94. American J. Bot. 87: 1712-1730. Fuse, S. and M. N. Tamura. (2000). A phylogeGrant, V. (1957). The plant species in theory netic analysis of the plastid matK gene with emphasis on Melanthiaceae sensu lat o. Plant and practice. In: E. Mayr (Ed.). The Species Problem. Amer. Assoc. Adv. Sci. Was hington, Biology 2: 415-427. D. C., pp. 39-80. Gagnepain, F. And Boureau. (1947). Nouvelles considerations systmatische propos d u Grant, V. (1981). Plant Speciation (2nd ed.). Columbia Univ. Press, New York. 688 Plant Systematics Grayum, M. (1987). A summary of evidence and Harris, T. M. (1932). The fossil flora of Scorseby arguments supporting the removal of Acorus Sound East Greenland. Part 3: Caytoniales and Bennettitales. Meddelelser om Grnla nd from Araceae. Taxon 36: 723-729.

Gregory, W. C. (1941). Phylogenetic and cytologi85(5): 1-133. cal studies in the Ranunculaceae. Trans. Am. Phil. Soc. 3: 443-520. Hart, J. A. (1987). A cladistic analysis of conifers: Preliminary results. J. Arnold Greuter, W., D. L. Hawksworth, J. Mcneill, Arbor. 68: 269-307. M. A. Mayo, A. Minelli, P. H. A. Sneath, B. J. Tindall, P. Trehane, and P. Tubbs . (1998). Draft BioCode (1997): the prospective inter-national rules for the sci entific names of Hartl, D. L. and E. W. Jones. (1998). Genetics: Principles and Analysis. 4th ed. Jone and Bartlett Publishers, London. Hartmann, H. E. K. (1993). Aizoaceae. In: K. organisms. Taxon 47: 127-150. Kubitzki, J. G. Rohwer and V. Bittrich (Ed.) The Families and Genera of vascular plants, Vol. 2, Magnoliid, Hamamelid and CaryophyGreuter, W., J. Mcneill, F. R. Barrie, H. M. Burdet, V. Demoulin, T. S. Filgueiras, D. H. Nicolson, P. C. Silva, J. E. Skog, P. Trehane, N. J. Turland and D. L. Hawksworth (editors & compilers). (2000). In ternational code of botanical nomenclature (St Louis Code) adopted by the Sixtee nth International Bo-tanical Congress St. Louis, Missouri, July llid families, p 37-69. Springer-Verlag, Berlin. Haszprunar, G. (1987). Vetigastropoda and systematics of Streptpneuros Gastropoda (Mollusca). Journal of Zoology 211: 747-770 . Hawksworth, D.L. (1995). Steps along the road to a harmonized bionomenclature. Taxon, 44: August 1999. Regnum Veg. Vol 138. 447-456. Greuter, W., R.K. Brummitt, E. Farr, N. Kilian, Hedge, I. C. and J. M. Lamond. (1972). P. M.Kirk and P.C. Silva. (1993). Names in current use for extant plant genera. Koeltz, Knigstein, Germany. xxvii + 1464 pp. Regnum Umbelliferae. Multi-access key to the Turk-ish genera. In: P. H. Davis (ed.). Fl ora of Turkey. Edinburgh Univ. Press, Edinburgh, vol. veg. Vol. 129. 4, pp. 171-177. Gunderson, A. (1939). Flower buds and phylogHedrn, M., M. W. Chase, and R. G. Olmstead. eny of dicotyledons. Bull. Torrey Bot. Club (1995). Relationships in the Acanthaceae and related families as suggested by cl adistic analysis of rbcL nucleotide sequences. Plant 66: 287-295. Gustafsson, M. H. G. (2002). Phylogeny of Clusiaceae based on rbcL sequences. Int. Syst. Evol. 194: 93-109. J. Plant Sci. 163(6): 1045-1054. Hegi, G. (1906-1931). Illustrierte Flora Von Haeckel, E. 1866. Generelle Morphologie der Mitteleuropa. Ed. I, Munchen. Organismen. Reimer, Berlin. Heimerl, A. (1934). Nyctaginaceae in Engler and Haeckel, E. (1887). Echte Grser. Engler and Prantl Die natrlichen Pflanzenfamilien , II, 2. Prantl, Die naturlichen pflanzenfamilien, ed. 2, Bd. 16c, 86-134. Hahn, W. J. 2002. A molecular phylogenetic study Henderson, D. M. (1983). International Directory

of the Palmae (Arecaceae) based on atpB, rbcL and 18s nrDNA sequences. Syst. Bio l. 51: 92-112. of Botanical Gardens IV. Koeltz, Koenigstein. Hennig, W. (1950). Grundzge einer Theorie der phylogenetischen Systematik. Deutscher Hall, D. W. (1981). Microwave: a method to conZentralverlag, Berlin. trol herbarium insects. Taxon, 30: 818-819. Hennig, W. (1957). Systematik und Phylogenese. Hanelt, P. and J. Schultze-Motel. (1983). ProBer. Hundertjahrfeier Deutsch. Entomol. Ges., posal (715) to conserve Triticum aestivum L. (1753) against Triticum hybernum L. (1753) pp. 50-70. Hennig, W. (1966). Phylogenetic Systematics. (Gramineae). Taxon 32: 492-498. Translated by D. D. Davies and R. Zangerl. Hansen, A. (1920). Die Pflanzendecke der Erde. Univ. Illinois Press, Urbana. Leipzig. Hermsen, E. J., T. N. Taylor, E. L. Taylor and Harborne, J. B. and B. L. Turner. (1984). Plant Chemosystematics. Acad. Press, L ondon. D. W. Stevenson. (2006). Cataphylls of the Middle Triassic cycad Antarcticycas s chopfii and new insights into cycad evolution. AmeriHarris, J. G. and M. W. Harris. (1994). Plant identification terminology: An illustrated glos-sary. Spring Lake, Publishing, S pring Lake, UT. can J. Bot. 93: 724-738. Heslop-Harrison, J. (1952). A reconsideration of plant teratology. Phyton 4: 1934. References 689 Heslop-Harrison, J. (1958). The unisexual flower Pooideae (Poaceae) based on nuclear rDNA (ITS) sequences. Theor. Appl. Genet. 90 : 389 a reply to criticism. Phytomorphology 8: 177184. 398. Heywood, V. H. (ed.). 1978. Flowering Plants of Huber, H. (1991). Angiospermen. Leitfaden durch the World. Oxford University Press, London. die Ordnungen und Familien der Bedektsamer. Hibbett, D. and M. J. Donoghue. (1998). InteGustav Fischer, Stuttgart. grating phylogenetic analysis and classificaHughes, N. F. (1961). Further interpretation of tion in fungi. Mycologia 90: 347-356. Eucommiidites Erdtman, 1948. Palaeontology Hickey, L. J and D. W. Taylor. (1996). Origin of 4: 292-299. angiosperm flower. In: D. W. Taylor and L. J. Hickey (Eds.) Flowering Plant Orig in, EvoluHufford, L. (1997). A phylogenetic analysis of Hydrangeaceae based on morphological data. tion and Phylogeny. Chapman & Hall Inc., Int. J. Plant Sci. 158: 652-672. New York, pp. 176-231. Hufford, L. (2001). Ontogeny and morphology of Hickey, L. J. and D. W. Taylor. (1992). the fertile flowers of Hydrangea and allied

Paleobiology of early angiosperms: evidence from sedimentological associations i n Early Cretaceous Potomac Group of eastern U. S. A. Paleontological Soc. Spec. Publ. 6: 128. genera of tribe Hydrangeeae (Hydrangeaceae). Bot. J. Linn. Soc. 137: 139 187. Hutchinson, J and J. M. Dalzeil. (1927-1929). Hickey, L. J. and J. A. Doyle. (1977). Early CreFlora of West Tropical Africa. London, 2 vols. taceous fossil evidence for angiosperm evoHutchinson, J. (1946). A Botanist in South Africa. lution. Bot. Rev. 43: 1-104 London. Hill, S. R. (1983). Microwave and the herbarium Hutchinson, J. (1948). British Flowering Plants. specimens: potential dangers. Taxon 32: 614London. 615. Hutchinson, J. (1964-67). The Genera of FlowerHilu, K. , T. Borsch, K. Muller, D. E. Soltis, P. S. ing Plants. Claredon, Oxford, 2 vols. Soltis, V. Savolainen, M. W. Chase, M. P. Powell, L. A. Alice, R. Evans, H. Sauq uet, C. Neinhuis, T. A. B. Slotta, J. G. Rohwer, C. S. Campbell, and L. W. Chatr ou. (2003). An-giosperm phylogeny based on matK sequence Hutchinson, J. (1968). Key to the Families ofF lowering Plants of the World. Clarendon, Oxford, 117 pp. Hutchinson, J. (1969). Evolution and Phylogeny of Flowering Plants. Acad. Press, London. information. American J. Bot. 90: 1758-1766. Hutchinson, J. (1973). The Families of Flowering Hjelmquist, H. (1948). Studies on the floral morPlants. (3rd ed.). Oxford Univ. Press. (2nd ed. phology and phylogeny of the Amentiferae. 1959; Ist ed. 1926, 1934) Bot. Notiser, Suppl. 2: 1-171. Index Kewensis plantarum p (1893 Holmgren, P. K., N. H. Holmgren and L. C. 95), 2 vols. Oxford. 16 supplements up to Barnett. (1990). Index herbariorum. Part I: The Herbaria of the World (8th ed.). Regnum 1971. Jaccard, P. (1908). Nouvelles recherches sur la Veg.120. distribution florale. Bull. Soc. Vaud. Sci. Nat. Hooker, J. D. (1870). Student s Flora of British 44: 223-270. Isles. Macmillan and Co., London. (3rd ed. Janesen, R. K. and J. D. Palmer. (1987). A chloin 1884). roplast DNA inversionmarks an ancient evolutionary split in sunflower family Hooker, J. D. (1872-97). Flora of British India.. L. Reeve and Co., London, 7 vols. (Asteraceae). Proc. Nat. Acad. Sc., USA 84: 5818-5822. Hoot, S. B. (1995). Phylogeny of Ranunculaceae based on epidermal preliminary atpB, rbcL Janssens, S., K. Geuten, Y. M. Yuan, Y

. Song, and 18S nuclear ribosomal DNAsequence data. Plant Syst. Evol. Suppl. 9: 241-251. P. Kpfer and E. Smets. (2006). Phylogenetics of Impatiens and Hydrocera (Balsamin aceae) using chloroplast atpB-rbcL spacer sequences. Syst. Bot. 31: 171-180. Hoot, S. B., J. W. Kadereit, F. R. Blattner, K.B. Jork, A. E. Schwarzbach and P. R. Crane. (1998). Data congruence and phylogeny of Jardine, N. and R. Sibson. (1971). Math ematical the Papaveraceae s.l. based on four data sets: atpB and rbcL sequences, trnK res tric-tion sites, and morphological characters. Taxonomy. Wiley, London. Ji, Q., L. M. Bowe, Y. Liu, & D. W. Taylor. (2004). Syst. Bot. 22: 575-590. Early Cretaceous Archaefructus eoflora sp. nov., with bisexual flowers from Beip iao, Western Liaoning, China. Acta Geol. Sinica 78: 883-896. Hsiao, C. , N. J. Chatterton, K. H. Asay and K. B. Jensen. (1994). Molecular phy logeny of 690 Plant Systematics Johnson, B. L. (1972). Seed protein profiles and Keating, R. C. (2004b). Systematic occurrence the origin of the hexaploid wheats. Amer. J. of raphide crystals in Araceae. Ann. Missouri Bot. 59: 952-960. Bot. Gard. 91: 495-504. Johnson, L. A. S. (1959). The Families of Cycads Keller, R. (1996). Identification of tropical woody and Zamiaceae of Australia. Proc. Linn. Soc. plants in the absence offl owers and fruits: A 84: 64-117. field guide. Birkauser, Basel. Johnson, L. A. S. and K. L. Wilson. (1993). Keller, R. A., R. N. Boyd & Q. D. Wheeler. (2003). Casuarinaceae. Pp. 237-242, in Kubitzki, K., Rohwer, J. G. and V. Bittrich. (eds ), The Fami-lies and Genera of Vascular Plants. II. Flower-ing Plants: Dicotyled ons, Magnoliid, Hamamelid Illogical basis of Phylogenetic nomenclature. Bot. Rev. 69(1): 93-110. Kellogg, E. A. 2000. The grasses: A case study in macroevolution. Ann. Rev. Ecol. Syst. 31: 217 and Caryophyllid Families. Springer, Berlin. 38. Jones, S. B. Jr. and A. E. Luchsinger. (1986). Kergulen, M. (1980). Proposal (68) on article 57.2 Plant Systematics, 2nd ed. McGraw-Hill Book to correct the Triticum example. Taxon 29: Co., New York. 516-517. Jordan, A. (1873). Remarques sur le fait de l existence en socit l tat sauvage des Kimura, M. (1968). Evolutionary rate at the molecular level. Nature 217: 624 626. espces vgtales affines. Bull. Ass. Fr. Kimura, M. (1983). The Neutral Theory of MolecuAvanc. Sci. 2, session Lyon. lar Evolution. Cambridge University Press, Judd, W. S., C. S. Cambell, E. A. Kellogg, P. F. Cambridge. Stevens and M. J. Donoghue. (2002). Plant Systematics: A Phylogenetic Approach. 2nd ed. Klak, C., A. Khunou, G. Reeves, and T. A. J.

Sinauer Associates, Inc., USA (Ist ed., 1999). Hedderson. (2003). A phylogenetic hypothesis for the Aizoaceae (Caryophyllales) based on four plastid DNA regions. American J. Bot. Judd, W. S., R. W. Sanders, and M. J. Donoghue. (1994). Angiosperm family pairs: Preliminary cladistic analyses. Harvard Pap. Bo t. No. 5: 90: 1433-1445. Klak, C., P. V. Bruyns and T. A. J. Hedderson. 1-51. (2007). A phylogeny and new classification for Mesembryanthemoideae (Aizoaceae). Kallersjo, M., G. Bergqvist & A. A. Anderberg (2000). Generic realignment in primuloid families of the Ericales s.l. (Angiospe rms): A phylogenetic analysis based on DNA sequences of rbcL and ndhF. American J. Bot. Taxon 56: 737-756. Kluge, A. G. and J. S. Farris. (1969). Quantitative phyletics and the evolution of anurans. Syst. Zool. 18: 1-32. 87: 1325-1341 Komarov, V. L. And B. K. Shishkin. (1934-1964). Krehed, J., J. Lundberg, B. Bremer and K. Flora SSSR. AN SSSR Press, Moscow/ Bremer. (1999). Evolution of the Australian Australasian families Alseuosmiaceae , Argophyllaceae and Phellinaceae. Systematic Leningrad, 30 vols. Kosakai, H., M. F. Moseley and V. I. Cheadle. Botany 24: 660-682. (1970). Morphological studies in the Nymphaeaceae. V. Does Nelumbo have vesKarp, G. (2002). Cell and Molecular Biology: consels?. Amer. J. Bot. 57: 487-494. cepts and experiments. 3rd ed. John Wiley and Kramer, K. U. and P. S. Green (Eds.). (1990). Vol. Sons, New York. I, Pteridophytes and Gymnosperms; in K. Kubitzki (ed.) Families and Genera of Va scular Keating, R. C. (2003a). Anatomy of the Monocotyledons. IX. Acoraceae and Araceae (ed. Gre-gory, M., & Cutler, D. F.). Oxford Un iversity Plants. pp. 284-391, Springer-Verlag, Berlin. Krassilov, V. A. (1977). Contributions to the Press, Oxford. knowledge of the Caytoniales. Rev. Paleobot. Keating, R. C. (2003b). Vegetative anatomical data Palynology 24: 155-178. and its relationship to a revised classifica-tion of the genera of Araceae. Miss ouri Bot. Gard. Monogr. Syst. Bot. [in press; cited in Kraus, R. (1897). ber Specifishe Reactionin in APweb]. Keimfreien Filtraten aus Cholera, Typhus und Pestbouillon Culturen erzeugt durch homologes Serum. Weiner Klin. Wechenschr. Keating, R. C. (2004a). Vegetative anatomical 10: 136-138. data and its relationship to a revised classi-fication of the genera of Araceae. Ann. Mis-souri Bot. Gard. 91: 485-494. Kron, K. A. (1996). Phylogenetic relationships of Empetraceae, Epacridaceae, Ericaceae, Monotropaceae and Pyrolaceae. Evidence References 691 from nucleotide ribosomal 18S sequence Les, D. H. and R. R. Haynes. (1995). Systematdata. Ann. Bot. 77: 293-303.

ics of subclass Alismatidae: A synthesis of approaches. Pp. 353-377, in Rudall, P. J., Cribb, P. J., Cutler, D. F., & Humphries, C. J. (Eds.), Monocotyledons: S ystematics and Kubitzki, K. (ed.) (1993). The Families and Genera of Vascular Plants , Vol. II. Flowering Plants, Dicotyledons: Magnoliid, Hamam elid and Caryophyllid Families. Springer-Verlag, New Evolution, vol. 2. Royal Botanic Gardens, Kew. York. Les, D. H., M. A. Cleland, and M. Waycott. (1997). Kubitzki, K. and M. W. Chase. (2002). IntroducPhylogenetic studies in Alismatidae, II: Evolution of marine angiosperms (seagra sses) and hydrophily. Syst. Bot. 22: 443-463. tion to Malvales. Pp. 12-16, in Kubitzki, K. (ed.), The Families and Genera of V ascular Plants. IV. Flowering Plants. Dicotyledons. Malvales, Capparales and Non -betalain Levin, R. A., W. L. Wagner, P. C. Hoch, M. Caryophallales. Springer, Berlin. Nepokroeff, J. C. Pires, E. A. Zimmer and K. J. Sytsma. (2003). Family-level rel ationships of Onagraceae based on rbcL and ndhF data. Lam, H. J. (1961). Reflections on angiosperm phylogeny. I and II. Facts and theories. Koninkljike Akademie van Wetenschappen te Amsterdam, Afdruunken Natuurkunde, Amer. J. Bot. 90(1): 107-115. Levin, R. A., W. L. Wagner, P. C. Hoch, W. J. Procesverbaal 64: 251-276. Hahn, A. Rodriguez, D. A. Baum, L. Katinas, E. A. Zimmer and K. J. Sytsma. (2004 ). Paraphyly in Tribe Onagreae: Insights into Phylogenetic Relationships of Onag raceae Based on Nuclear and Chloroplast Sequence Data. Systematic Botany. 29(1): 147-164. Lamarck, J. B. P. (1778). Flore Francaise. Imprimerie Royale, Paris, 3 vols. Lamarck, J. B. P. (1809). Philosophie Zoologique. Paris. Lamb Frye, A. S. and K. A. Kron. (2003). RbcL Linnaeus, C. (1730). Hortus uplandicus. phylogeny and character evolution in Stockholm. Polygonaceae. Syst. Bot. 28: 326-332. Linnaeus, C. (1735). Systema naturae. (2nd ed.). Lance, G. N. and W. T. Williams. (1967). A gen-eral theory of classificatory sor ting strateLugduni Batavorum. Stockholm. Linnaeus, C. (1737). Critica botanica. Leyden. Linnaeus, C. (1737). Flora Lapponica. Amsterdam. gies. 1. Hierarchical systems. Computer J. 9: 373-380. Linnaeus, C. (1737). Genera plantarum. Lugduni Lapage, S.P., P.H.A. Sneath, E.F. Lessel, V.B.D. Batavorum. Skerman, H.P.R. Seeliger, and W.A. Clark (Eds.). (1992). International Code ofN omencla-ture of Bacteria (Bacteriological Code 1990 Revision). Amer. Soc. Microb iol., WashingLinnaeus, C. (1737). Hortus Cliffortianus. Amsterdam. Linnaeus, C. (1751). Philosophica botanica. Stockholm, 362 pp. ton, D.C. xlii + 189 pp. Linnaeus, C. (1753). Species plantarum. Laubengayer, R. A. (1937). Studies in the Stockholm, 2 vols. anatomy and morphology of the polygonaceous flower. Amer. J. Bot. 24: 329-

Linnaeus, C. (1762). Fundamenta fructificationis. Stockholm. 343. Loconte, H. (1993). Berberidaceae. In K. Kubitzki, Lawrence, G. H. M. (1951). Taxonomy of Vascular J. G. Rohwer and V. Bittrich (Eds.) The families and Genera of vascular plants, vol. 2, Magnoliid, hamamelid and caryophyllidf amilies. Plants. Macmillan, New York. Lee, T. B. (1979). Illustrated Flora of Korea. Hyangmunsa, Seoul. 147-152. Springer-Verlag, Berlin. Lee, Y. S. (1981). Serological investigations in Loconte, H. (1996). Comparison of Alternative Ambrosia (Compositae: Ambrosieae) and relaHypotheses for the Origin of Angiosperms. In D. W. Taylor and L. J. Hickey (Eds. ) Flowering Plant Origin, Evolution and Phylogeny, tives. Syst. Bot. 6: 113-125. Lemesle, R. (1946). Les divers types de fibres a ponctuations areolees chez les dicotyledones apocarpiques les plus archaiques et leur role dans la phylogenie. 267-285. Chapman and Hall, Inc, New York. Loconte, H. and D. W. Stevenson. (1991). Cladistics of Magnoliidae. Cladistics 7: 267-296. Ann. Sci. Nat. Bot. et Biol. Vegetal 7: 19-40. Lve, A., D. Lve and R. E. G. Pichi-Sermolli. Les, D. H., D. K. Garvin and C. F. Wimpee. (1977). Cytotaxonomic Atlas of Pteridophytes. (1991). Molecular evolutionary history of ancient aquatic angiosperms. Proc. Nat . Acad. Sc. USA 88: 10119-10123. Cramer, Koenigstein. Lundberg, J. (2001). A well resolved and supported phylogeny of Euasterids II ba sed on 692 Plant Systematics a Bayesian inference, with special empha-sis on Escalloniaceae and other incerta e sedis. Chapter V, in Lundberg, J. Phyloge-netic studies in the Euasterids II w ith particubers and their taxonomic significance. Bot. J. Linn. Soc. (London) 60: 411-417. Markham, K. R., L. J. Porter, E. O. Cambell, J. lar reference to Asterales and Escalloniaceae. Acta Universitatis Upsaliensis, U ppsala. Chopin and M. L. Bouillant. (1976). Phytochemical support for the existence of t wo species in the genus Hymenophyton. PhyLundberg, J. and K. Bremer. (2002). A phylogetochemistry 15: 1517-1521. netic study of the order Asterales using one large morphological and three molec ular data Martin, W., D. Lydiate, H. Brinkmann, G. sets. International Journal of Plant Sciences. Forkmann, H. Saedler and R. Cerff. (1993). Molecular phylogenies in angiosperm e voluMabberley, D. J. (1997). The Plant Book. Ed. 2. Cambridge University Press, Camb ridge. tion. Mol. Biol. Evol. 10: 140-162. Mason, H. L. (1936). The principles of geographic Mabry, T. J. (1976). Pigment dichotomy and DNAdistribution as applied to floral analysis. RNA hybridization data for Centrospermous

Madrono 12: 161-169. families. Pl. Syst. Evol. 126: 79-94. Mason-Gamer, R. J., and E. A. Kellogg. (1996). Maheshwari, J. K. (1963). Flora of Delhi. CSIR, Chloroplast DNA analysis of the monogenomic Triticeae: Phylogenetic implications and genome-specific markers. In P. Jauhaur (ed.) Methods of genome analysis in plants: Their merits and pitfalls. 301-325. CRC New Delhi. Maheshwari, P. (1964). Embryology in relation to taxonomy. In: W. B. Turril (ed.). Vistas in Botany. Pergamon Press, London, v ol. 4, pp. 55-97. Press, Boca Raton, FL. Malcomber, S. T., & E. A. Kellogg. (2005). Mason-Gamer, R. J., C. F. Weil and E. A. Kellogg. SEPALLATA gene diversification: Brave new (1998). Granule bound starch synthase: Structure, function and phylogenetic util ity. whorls. Trends Plant Sci. 10: 427-435. Malcomber, S. T., J. C. Preston, R. Reinheimer, Mol. Biol. Evol. 15: 1658-1673. J. Kossuth E. A. & Kellogg. (2006). Develop-mental gene evolution and the origin of grass inflorescence diversity. Adv. Bot. Res. 44: Mathew, K. M. (1983). The Flora of the Tamil Nadu Carnatic. The Rapinat Herbarium, St. Joseph s College, Tirucherapalli, India, 3 vols. 425-481. Mathews, S. and R. A. Sharrock. (1996). The phyMaldonado de Magnano, S. (1986). Estudios tochrome gene family in grasses (Poaceae): A phylogeny and evidence that grasses have a subset of loci found in dicot angiosperms. embriologicos en Buddleja (Buddlejaceae) I: Endosperma y episperma. Darwiniana 2 7: 225236. Mol. Biol. Evol. 13: 1141-1150. Meng, Shao-Wu, Zhi-Duan Chen, De-Zhu Li and Han-Xing Liang (2002). Phylogeny of Saururaceae Matthews, M. L. and P. K. Endress. (2002). Combased on mitochondrial matR gene sequence parative floral morphology and systematics in Oxalidales (Oxalidaceae, Connarace ae, Brunelliaceae, Cephalotaceae, Cunoniaceae, Elaeocarpaceae, Tremandraceae). B ot. J. Data. J. Plant Res. 115:71-76. Manning, S. D. (2000). The genera of Bignoniaceae in the Southeastern United Linn. Soc. 140: 321-381. States. Harvard Papers Bot. 5: 1-77. Mauseth, J. D. (1998). Botany: An Introduction to Manos, P. S., K. C. Nixon and J. J. Doyle. (1993). Plant Biology. 2/e multimedia enhanced edition. Jones and Bartlett Publishers, M assaCladistic analysis of restriction site varia-tion within the chloroplast DNA inv erted repeat region of selected Hamamelididae. chusetts. Mayo, S. J., L. Cabrera, G. Salazar and M. Chase. Syst. Bot. 18: 551-562. (2003). Aroids and their watery beginnings. Manos, P. S. and K. P. Steele. (1997). PhylogeMs. (cited in APweb) netic analysis of higher hamamelididae based on plastid sequence data. Syst. Bot. Mayr, E. (1942). Systematics and the Origin of Species. Columbia Univ. Press, Ne

w York. 84: 1407-1419. Mayr, E. (1957). The Species Problem. Amer. Manos, P. S. , Z.-K. Zhou, and C. H. Cannon. Assoc. Adv. Sci. Pub. No. 50. (2001). Systematics of Fagaceae: Phyloge-netic tests of reproductive trait evolu tion. Mayr, E. (1963). Animal Species and Evolution. Belknap Press, Harvard Univ. Press, CamInt. J. Plant Sci. 162: 1361-1379. bridge. Marchant, C. J. (1968). Evolution in Spartina (Gramineae) III. Species chromosom e numMayr, E. (1966). The proper spelling of taxonomy. Systematic Zool. 15: 88. References 693 Mayr, E. (1969). Principles of Systematic Zoology. McGraw-Hill, New York. Mirov, N. T. (1961). Composition of Gum Terpentines of Pines. U. S. Dept. Agric. Tech. Bull. 1239. Mayr, E. and P. D. Ashlock (1991). Principles of Mirov, N. T. (1967). The Genus Pinus. Ronald Systematic Zoology. 2nd. ed. McGraw-Hill, Press, New York. New York. Mitchell, J. D., D. C. Daly, S. K. Pell, and A. Mayr, E. (1998). Two empires or three? . ProRandrianasolo. (2006). Poupartiopsis gen. nov. and its context in Anacardiaceae classificaceedings of the National Academy of Science 95: 9720 9723. tion. Syst. Bot. 31: 337-348. McMillan, C., T. J. Mabry and P. I. Chavez. (1976). Moore, G., K. M. Devos, Z. Wang and M. D. Gale. Experimental hybridization of Xanthium strumarium (Compositae) from Asia and Ame rica, II.Sesquiterpene lactones of F1 (1995). Current Opinion Genet. Devel. 5: 737. Muasya, A. M., D. A. Simpson, M. W. Chase, hybrids. Amer. J. Bot. 63: 317-323. and A. Culham. (1998). An assessment of suprageneric phylogeny in Cyperaceae usi ng rbcL DNA sequences. Plant Syst. Evol. McMinn, H. E. and E. Maino. (1946). Illustrated Manual of Pacific Coast Trees, 2nd ed. Univ. 211: 257-271. of Calif. Press, Berkeley, CA. Muhammad, A. F. and R. Sattler. (1982). Vessel McNeill, J., Barrie, F. R. et al. (Eds.). (2006). Instructure of Gnetum and the origin of angiosperms. Amer. J. Bot. 69: 1004-1021. ternational Code of Botanical Nomenclature (Vienna Code) 2006. Adopted by the Se venMller, K., & T. Borsch. (2005a). Phylogenetics teenth International Botanical Congress, Vienna, Austria, July 2005. Series: Reg num Vegetabile Vol. 146. 568 pages. Gantner of Amaranthaceae based on matK/trnK se-quence data - evidence from parsimony, li kelihood, and Bayesian analysis. Ann. Missouri Verlag Bot. Gard. 92: 66-102. Meeuse, A. J. D. (1963). The multiple origins of Mller, K., & T. Borsch. (2005b). Multiple the angiosperms. Advancing Frontiers of Plant origins of a unique pollen feature: Stellate pore ornamentation in Amaranthaceae . Grana

Sciences, 1: 105-127. Meeuse, A. J. D. (1972). Facts and fiction in flo44: 266-281. ral morphology with special reference to the Polycarpicae. Acta Bot. Neerl. 21: 113-127, Naik, V. N. (1984). Taxonomy ofA ngiosperms. Tata McGraw Hill, New Delhi. 235-252, 351-365. Nandi, O. I. M. W. Chase and P. K. Endress. Meeuse, A. J. D. (1990). All about Angiosperms. (1998). A combined cladistic analysis of angiosperms using rbcL and non-molecula r data Eburon, Delft. Meglitsch, P. A. (1954). On the nature of spesets. Ann. Missouri Bot. Gard. 85: 137-212. cies. Zyst. Zool. 3: 49-65. Neumayer, H. (1924). Die Geschichte der Blte. Melville, R. (1962). A new theory of the anAbhandlung Zoologischen Botanische giosperm flower, I. The gynoecium. Kew Bull. Gesellschaft 14: 1-110. 16: 1-50. Nickrent, D. L. (2002). Orgenes filogenticos de Melville, R. (1963). A new theory of the an-giosperm flower, II. Kew Bull. 17: 1 -63. las plantas parsitas. Pp. 29-56 in LpezSez, J. A., Cataln & Sez, L. (eds), Plantas Pa rsitas de la Pennsula Ibrica e Islas Melville, R. (1983). Glossopteridae, Angiospermidae and the evidence of angiosperm oriBaleares. Mundi-Prensa, Madrid. gin. Bot. J. Linn. Soc. 86: 279-323. Melville, R. and H. M. Heybroek. (1971). The Elms of the Himalayas. Kew Bull. 26 (1): 5-28. Nicolson, D. H. (1974). Paratautonym, a comment on proposal 146. Taxon 24: 389-3 90. Nixon, K. C. and J. M. Carpenter. (2000). On Mrat, F. V. (1821). Nouvelle flore des environs de the other Phylogenetic Systematics . ClaParis ed. 2, 2 Paris, 107 pp. distics 16: 298-318. Metcalfe, C. R. and L. Chalk. (Eds.) (1983). Nixon, K. C., J. M. Carpenter and D. W. Anatomy of Dicotyledons.(2nd ed.). Claredon Press, Oxford. (Takhtajan s classifica tion, Stevenson. (2003).The PhyloCode is fatally flawed, and the Linnaean System can be vol., 2 pp. 258-300). easily fixed. Bot. Rev. 69 (1): 111-120. Michener, C. D. and R. R. Sokal. (1957). A quanNyananyo, B. L and S. I. Mensah. (2004). Distrititative approach to a problem in classificabution and origins of members of the family Portulaceae (Centrospermae). Journal of Applied Sciences & Environmental Management, 8 (2): 59-62 tion. Evolution 11: 130-162. Miki, S. (1937). The origin of Najas and Potamogeton. Bot. Mag. Tokyo 51: 290-48 0. 694 Plant Systematics Nyffeler, R. 2002. Phylogenetic relationships in Pell, S. K. 2004. Molecular Systematics of the the cactus family (Cactaceae) based on evidence from trnK/matK and trnL-trnF seq

uences. American J. Bot. 89: 312-326. Cashew Family (Anacardiaceae). Ph. D. Thesis, Louisiana State University. Pell, S. K. and J. D. Mitchell. (2007). EvolutionOlmstead, R. G. and P. A. Reeves. (1995). ary trends in Anacardiaceae inferred from nuclear and plastid molecular data and morphological evidence. P. 178, in Plant BiolEvidence for polyphyly of Scrophulariaceae based on chloroplast rbcL and ndhF se quences. Ann. Missouri Bot. Gard. 82: 176-193. ogy and Botany 2007. Program and Abstract Book. Chicago. Olmstead, R. G., J. A. Sweere, R. E. Spangler, Pennington, T. D. (1991). The genera of Sapotaceae. L. Bohs, and J. D. Palmer. (1999). Phylogeny and provisional classification of t he Solan-aceae based on chloroplast DNA. Pp. 111-137, in Nee, M., Symon, D., Les ter, R. N., & Jessop, J. P. (Eds.), Solanaceae IV: Advances in Biology and Utili zation. Royal Botanic GarRoyal Botanic Gardens, Kew and New York Botanic Gardens, Bronx. Petersen, G. and O. Seberg. (1997). Phylogenetic analysis of Triticeae (Poaceae) based on rpoA sequence data. Mol. Phylog. Evol. 7: 217-230. Pichon, H. (1946). Sur les Alismataces et les dens, Kew. Butomaces [includes Albidella, gen. nov., key to genera of redefined Alismaceae]. Not. Olmstead, R. G., K. -J Kim, R. K. Jansen and S. J. Wagstaff. (2000). The phylogeny of the Asteridae sensu lato based on chlor oplast ndhf gene sequences. Molecular Phylog. Evol. Syst. [Paris] 12: 170-183. Plunkett, G. M., D. E. Soltis, and P. S. Soltis. 16: 96-112. (1995). Phylogenetic relationships between Juncaceae and Cyperaceae: Insights fr om rbcL sequence data. American J. Bot. Olmstead, R. G., C. W. dePamphilis, A. D. Wolfe, N. D. Young, W. J. Elisens and P. A. Reeves. (2001). Disintegration of the Scrop hularia82: 520-525. ceae. American J. Bot. 88: 348-361. Plunkett, G. M., D. E. Soltis, and P. S. Soltis. (1997). Clarification of the re lationship beOlson, A. R. (1991). Postfertilization changes in tween Apiaceae and Araliaceae based on matK and rbcL sequence data. Amer. J. Bot . ovules of Monotropa uniflora L. (Monotropaceae). American J. Bot. 78: 99-107. 84: 567-580. Owen, R. (1848). Report on the archetype and Plunkett, G. M. (2001). Relationship of the orhomologies of vertebrate skeleton. Rep. 16th der Apiales to subclass Asteridae: A re-evaluation of morphological characters b ased on insights from molecular data. Edinburgh J. Meeting Brit. Assoc. Adv. Sci. 169-340. Owenby, M. (1950). Natural hybridisation and amphiploidy in the genus Tragopogon. Amer. Bot. 8: 183-200. J. Bot. 37(10): 487-499. Plunkett, G. M., J. Wen, and P. P. Lowry ( 2004). Page, C. N. (1979). The herbarium preservation of Conifer specimens. Taxon 28: 3 75-379.

Infrafamilial classifications and characters in Araliaceae: Insights from the ph ylogenetic analysis of nuclear (ITS) and plastid (trnLtrnF) sequence data. Plant Syst. Evol. Page, C. N. (1990). Pinata. Pp. 290-361, in Kubitzki, K. (Ed.), The Families and Genera of Vascular Plants. Volume 1. Pterid ophytes and 245: 1-39. Gymnosperms. Springer, Berlin. Porter, C. L. (1959). Taxononomy of Flowering Plants. W. H. Freeman, San Francis co. Pant, D. D. (1957). Classification of gymnospermous plants. Palaeobotanist 6: 65-70. Porter, E. A. and L. A. Johnson. (1998). PhylogePant, D. D. and P. F. Kidwai. (1964). On the dinetic relationships of Polemoniaceae: Inferences from mitochondrial nad1b intron seversity in the development and organisation of stomata in Phyla nodiflora Michx. Curr. Sci. quences. Aliso 17: 157-188. 33: 653-654. Porter, E. A., E. Nic Lughadha and M. S. J. Pax, F. and K. Hoffmann. (1934). Aizoaceae in Simmonds. (2000). Taxonomic significance of polyhydroxyalkaloids in the Myrtacea e. Kew Engler and Prantl, Die naturlichen pflanzenfamilien, 2ed, Bd. 16c, 179-233. Bull. 55: 615-632. Peery, R., K. R. Wilcox, C. Morton and L. A. Potgeiter, J. and V. A. Albert. (2001). PhylogeRaubeson. (2008). Exploring the utility of the nuclear XDH gene for gymonsperm p hylo-genetics. Pp. 196-197, in Botany 2008. Botany without Borders. Botanical So ciety of America. netic relationships within Apocynaceae s.l. based on trnL intron and trnL-F spac er se-quences and propagule characters. Ann. Missouri Bot. Gard. 88: 523-549. References 695 Potter, D., F. Gao, P. E. Bortiri, S. H. Oh and S. Stech, W. Frey, D. Quandt, & C. C. Davis. (2006). The deepest divergences in lan d plants inferred from phylogenomic evidence. Proc. National Acad. Sci. U.S.A 10 3: 15511 Baggett. (2002). Phylogenetic relationships in Rosaceae inferred from chloroplas t matK and trnL-trnF nucleotide sequence data. Plant Syst. evol. 231: 77-89. 15516. . Prat, W. (1960). Vers une classification Quinn, C. J., R. A. Price, and P. A. Gadek. (2002). naturelles des Gramines. Bull. Soc. Bot. Fr. Familial concepts and relationships in the conifers based on rbcL and matK seque nce 107: 32-79. Prather, C. A., C. J. Ferguson and R. K. Jansen. comparisons. Kew Bull. 57: 513-531. (2000). Polemoniaceae phylogeny and clas-sification: Implications of sequence da ta from the chloroplast gene ndhF. Amer. J. Bot. Radford, A. E. (1986). Fundamentals of Plant Systematics. Harper and Row, New York. Radford, A. E., W. C. Dickison., J. R. Massey 87: 1300-1308. and C. R. Bell. (1974). Vascular Plant Systematics. Harper and Row, New York. Pratt, D. B., L. G. Clark and R. S. Wallace. (2001).

A tale of two families: Phylogeny of the Chenopodiaceae-Amaranthaceae. P. 135, i n Botany 2001: Plants and People, Abstracts. Rai, H. S., P. A. Reeves, R. Peakall, R. G. [Albuquerque.]. Olmstead and S. W. Graham. (2008). Infer-ence of higher-order conifer relationsh ips from a multi-locus plastid data set. Botany Preston, J. C., & E. A. Kellogg. (2006). Recon86: 658-669. structing the evolutionary hiistory of paralogous APETALA1?FRUITFULL-like genes Raimann, R. (1893). Onagraceae. in Engler and Prantl, Die naturlichen Pflanzenfamilien in Grasses (Poaceae). III(7): 199-223. Price, R. A. (2003). Generic and familial relationRam, Manasi. (1959). Morphological and embryoships of the Taxaceae from rbcL and matK sequence comparisons. Pp. 235-238, in M ill, R. R. (ed.), IV International Conifer Conference. logical studies in the family Santalaceae II. Exocarpus, with a discussion on it s system-atic position. Phytomorphology 8: 4-19. [ISIS Acta Horticulturae 615.] Raven, P. H. (1975). The bases of angiosperm Qui, Y-L, M. W. Chase, D. H. Les and C. R. phylogeny: cytology. Ann. Miss. Bot. Gard. Parks. (1993). Molecular phylogenetics of the Magnoliidae: Cladistic analyses of nucle-otide sequences of plastid gene rbcL. Ann. 62: 725-764. Ray, J. (1682). Methodus plantarum nova. London, 3 vols. Missouri Bot. Gard. 80: 587-606. Ray, J. (1686-1704). Historia plantarum. London, Qiu, Y.-L., M. W. Chase, S. B. Hoot, E. Conti, P. 3 vols. R. Crane, K. J. Sytsma and C. R. Parks. (1998). Phylogenetics of the Hamamelidae and their allies: Parsimony analyses of nucleotide sequences of the plastid gen e Rechinger, K. H. (1963). Flora Iranica. Reeves, R. G. (1972). Flora of Central Texas. Prestige Press, Ft. Worth, TX, 320 pp. Rehder, A. (1940). Manual of Cultivated Trees and rbcL. Int. J. Plant Sci. 159: 891-905. Shrubs Hardy in North America, (2nd ed). Qui, Y-L, J. Lee, F. Bernasconi-Quadroni, D. E. Macmillan, New York. Soltis, P. S. Soltis, M. Zanis, E. A. Zimmer, Z. Chen, V. Savolainen and M. W. C hase. (1999). The earliest angiosperms: evidence from mitochondrial, plastid and nuclear geRendle, A. B. (1904). Classification of flowering plants. Cambridge, England. Vol. 2 1925; 2nd ed. Vol. 1 1930. Renner, S. S., A. Weerasooriya and M. E. Olson, nomes. Nature 402: 404-407. (2002). Phylogeny of Cucurbitaceae inferred from multiple chloroplast loci. P. 1 69, in Botany 2002: Botany in the Curriculum, Qui, Y-L, J. Lee, F. Bernasconi-Quadroni, D. E. Soltis, P. S. Soltis, M. Zanis, E. A. Zimmer, Z. Chen, V. Savolainen and M. W. C hase. (2000). Phylogeny of basal angiosperms: Analyses of five genes from three genomes. Int. J. Plant Sci. 161(6: suppl.): S3-S27. Abstracts. [Madison, Wisconsin.] Retallack, G. and D. L. Dilcher. (1981). A coastal Qiu, Y.-L., L. Li, B. Wang, Z. Chen, V. Knoop, M.

hypothesis for the dispersal and rise of dominance of flowering plants. In: K. J . Niklas (Ed.). Paleobotany, Paleoecology and Evolution. Groth-Malonek, O. Dombrovska, J. Lee, L. Kent, L., J. S. Rest, G. F. Estabrook, T. A. Hendry, D. W. Taylor, C. M. Testa, M. Ambros, B. Crandall-Stotler, R. J. D uff, M. Praeger, New York, Vol 2. Pp. 27-77. Riesberg, L. H., B. Sinervo, C. R. Linder, M. Ungerer and D. M. Arias. (1996). Role of gene interactionsin hybrid speciation:E vidence 696 Plant Systematics from ancient and experimental hybrids. (2000). Phylogenetics of flowering plants based on combined analysis of plastid atpB Science 272: 741-745. Rise, K. A., M. J. Donoghue and R. G. Olmstead. and rbcL sequences. Syst. Biol. 49: 306-362. (1997). Analyzing large data sets: rbcL 500 Salywon, A., N. Snow & L. R. Landrum. (2002). revisited. Syst. Biol. 46: 554-563. Phylogenetic relationships in the berry-fruited Myrtaceae as inferred from ITS s equences. P. 149, in Botany 2002: Botany in the Rodman, J. E., R. A. Price, K. Karol, E. Conti, K. J. Sytsma and J. D. Palmer. (1993). Nucle-otide sequences of rbcL gene indicate mono-phyly of mustard oil plants. Ann. Missouri Bot. Curriculum, Abstracts. [Madison, Wisconsin.] Scamardella , J. M. (1999). Not plants or aniGard. 80: 686-699. mals: a brief history of the origin of King-doms Protozoa, Protista and Protocti sta . InRogers, D. J. (1963). Taximetrics, new name, old concept. Brittonia 15: 285-290. ternational Microbiology 2: 207 221. Rollins, R. C. (1953). Cytogenetical approaches Scharaschkin, T., & J. A. Doyle. (2005). Phylogto the study of genera. Chronica Botanica eny and historical biogeography of Anaxagorea (Annonaceae) using morphology and non-coding chloroplast sequence data. Syst. Bot. 14(3): 133-139. Ross, K. L. (2004). All Living Things: In Seven Kingdoms.The Proceedings of the Friesian School, Fourth Series. http://www.fries ian. 30: 712-735. Schlueter, J. A., P. Dixon, C. Granger, D. Grant, com/life-1.htm#one. L. Clark, J. J.Doyle & R. C. Schoemaker. (2004). Mining EST databases ro resolve evolutionary events in major crop species. GeRousi, A. (1973). Studies on the cytotaxonomy and mode of reproduction of Leontodon (Compositae). Ann. Bot. Fenn. 10: 201-215. nome 47: 868-876. Rova, J. H. E., P. G. Delprete, L. Andersson, and Schneider, H. A. W. and W. Liedgens. (1981). An V. A. Albert. (2002). A trnL-F cpDNA sequence study of the Condamineeae-Rondelet ieaeevolutionary tree based on monoclonal an-tibody-recognised surface features of p lastid enzume (5-aminolevulinate dehydratase). Sipaneeae complex with implications on the phylogeny of the Rubiaceae. Amer. J. Bot. Z. Naturforsch. 36(c): 44-50.

89: 145-159. Schneider, H., S. J. Russell, C. J. Cox, F. Bakker, S. Rudall, P., C. A. Furness, M. W. Chase and M. Henderson, M. Gibby & J. C. Vogel. (2004). Chloroplast phylogeny of asplenioid f erns based on rbcL and trnL-F spacer sequences (Polypodiidae, Aspleniaceae) and its implications F. Fay. (1997). Microsporogenesis and pol-len sulcus type in Asparagales (Lilian ae). Canad. J. Bot. 75: 408-430. Rudall, P. (2001). Floral morphology of for the biogeography. Syst. Bot. 29: 260 274. Asparagales: unique structures and iterative evolutionary themes. P. 16, in Bota ny 2001: Schneider, H., A. R. Smith, R. Cranfill, T. E. Plants and People, Abstracts. [Albuquerque.] Hildebrand, C. H. Haufler & T. A. Ranker. (2004b). Unraveling the phylogeny of p olygrammoid ferns (Polypodiaceae and Grammitidaceae): exploring aspects of the d iversification of epiphytic plants. Molec. Rydin, C., M. Kllersj, and E. M. Friis. (2002). Seed plant relationships and the systematic position of Gnetales based on nuclea r and chloroplast data: Conflicting data, rooting problems, and the monophyly of conifers. Phylog. Evol. 31: 1041 1063. Schubert, I., H. Ohle and P. Hanelt. (1983). PhyInternat. J. Plant Sci. 163: 197-214. logenetic conclusions from Geisma banding and NOR staining in Top Onions (Liliac eae). Sahasrabudhe, S. and C. A. Stace. (1974). Developmental and structural variation in the trichomes and stomata of some Pl. Syst. Evol. 143: 245-256. Schulz, O. E. (1936). Cruciferae. In: E. Engler, Gesneriaceae. New Botanist 1: 46-62. Die Naturlichen Pflanzenfamilien, ed. 2, Sahni, B. (1925). Ontogeny of vascular plants and 17B: 227-658. theory of recapitulation. J. Indian Bot. Soc. Schwarzbach, A. E. and L. A. McDade. (2002). 4: 202-216. Phylogenetic relationships of the mangrove family Avicenniaceae based on chlorop last and nuclear ribosomal DNA sequences. Syst. Sahni, B. (1948). Pentoxyleae: A new group of Jurassic gymnosperms from the Rajmahal Hills of India. Bot. Gaz. 110: 47-80. Bot. 27: 84-98. Savolainen, V., M. W. Chase, S. B. Hoot, C. M. Scotland, R. W. (!990). Palynology and systematMorton, D. E. Soltis, C. Bayer, M. F. Fay, A. Y. de Bruijn, S. Sulllivan, and Y. -L. Qiu. ics of Acanthaceae. Ph. D. Thesis, University of Reading, England. References 697 Scotland, R. W. and K. Volleson. (2000). ClassifiSneath, P. H. A. and R. R. Sokal. (1973). Numerication of Acanthaceae. Kew Bull. 55: 513-589. cal Taxonomy. W. H. Freeman and Company, Sennblad, B and B. Bremer. (2002). ClassificaSan Francisco. tion of Apocynaceae s.l. according to a new approach combining Linnaean and phyl oge-netic taxonomy. Syst. Biol. 51: 389-409. Snustad, D. P. and M. J. Simmons. (2000). Principles of Genetics. 2nd ed. John Wiley and

Sons, New York. Seward, A. C. (1925). Arctic vegetation past and Sokal, R. R. (1961). Distance as a measure of present. J. Hort. Soc. 50, i. taxonomic similarity. Systematic Zool. 10: Shukla, P. and S. P. Misra. (1979). An introduc70-79. tion to Taxonomy of Angiosperms. Vikas PubSokal, R. R. and C. D. Michener. (1958). A stalishing House, New Delhi. tistical method for evaluating systematic relationships. Univ. Kansas Sci. Bull. 44: 467-507. Simpson, G. G. (1961). Principles of Animal Taxonomy. New York/London. Simpson, D. A., Furness, C. A., Hodkinson, T. Sokal, R. R. and P. H. A. Sneath. (1963). PrinR., Muasya, A. M., and M. W. Chase. (2003). Phylogenetic relationships in Cypera ceae subfamily Mapanioideae inferred from pol-len and plastid DNA sequence data. Americiples of Numerical Taxonomy. W. H. Freeman and Company, San Francisco. Solsbrig, O. T. (1970). Principles and Methods of Plant Biosystematics. Macmilla n, London. can J. Bot. 90: 1071-1086. Soltis, D. E., D. R. Morgan, A. Grable, P. S. Soltis Simpson, M. G.. (2006). Plant Systematics. 590 and R. Kuzoff. (1993). Molecular systematics of Saxifragaceae sensu stricto. Ame r. J. pp. Elsevier Academic Press, San Diego, CA. Singh, G. (1999). Plant Systematics. Science PubBot. 80: 1056-1081. lishers, New York. Soltis, D. E., P. S. Soltis, D. L. Nickrent, L. A. Singh, G. (1999). Plant Systematics- Theory and Johnson, W. J. Hahn, S. B. Hoot,J. A. Sweere, R. K. Kuzoff, K. A. Kron,M. W. Cha se, S. M. Swensen, E. A. Zimmer, S.-M. Chaw, L. J. Gillespie, W. J. Kresss and K . J. Sytsma. (1997). Angiosperm phylogeny inferred from 18S ribosomal DNA sequen ces. practice. Oxford & IBH Publishing Co. Pvt. Ltd., New Delhi. Singh, G., Bimal Misri and P. Kachroo. (1972). Achene morphology: An aid to the taxonomy of Indian Plants.1. Compositae, Liguli florae. J. Indian Bot. Soc. 51(3-4): 235-242. Ann. Missouri Bot. Garden 84: 1-49. Sinnot, E. W. and I. W. Bailey. (1914). InvestigaSoltis, D. E., P. S. Soltis, M. E. Mort, M. W. tions on the phylogeny of angiosperms. Part Chase, V. Savolainen, S. B. Hoot, and C. M. Morton. (1998). Inferring complex ph ylogenies using parsimony: An empirical approach using three large DNA data sets for angiosperms. Syst. Biol. 47: 32-42. 3. Amer. J. Bot. 1: 441-453. Sinnott, Q. P. (1983). A Solar Thermoconvective plant drier. Taxon 32: 611-613. Sivarajan, V. V. (1984). Introduction to Principles of Plant Taxonomy. Oxford & IBH, New Delhi. Soltis, D. E., P. S. Soltis, M. W. Chase, M. E. Smith, A. C. (1970). The Pacific as a key to flowerMort, D. C. Albach, M. Zanis, V. Savolainen, W. H. Hahn, S. B. Hoot, M. F. Fay, M. Axtell, S. M. Swensen, L. M. Prince, W. J. Kress, K. C. Nixon and J. A. Farri

s. (2000). Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences . Bot. Journ. Linn. Soc. 133: ing plant history. Harold L. Lyon Arboretum Lecture Number 1. Smith, A. R., M. K. Pryer, E. Schuettpelz, P. Korall, Harald Schneider & P. G. Wolf. (2006). Classification of extant Ferns. T axon 55(3): 705-731. 381-461. Smith, P. M. (1972). Serology and species relaSoltis, D. E., R. K. Kuzoff, M. E. Mort, M. Zanis, tionship in annual bromes (Bromus L sect. M. Fishbein, L. Hufford, J. Koontz and M. Arroyo. (2001). Elucidating deep-level phylogenetic relationships in Saxifragaceae using sequences for six chloroplast ic and nuclear DNA regions. Ann. Missouri Bot. Gard. Bromus). Ann. Bot. 36: 1-30. Smith, P. M. (1983). Protein, mimicry and microevolution in grasses. In: U. Jensen and D. E. Fairbrothers (Eds.). Proteins and Nucleic Acids in Plant Systematics. Springer-Verlag, 88: 669-693. Berlin, pp. 311-323. Soltis, P. S., D. E. Soltis and M. W. Chase. Sneath, P. H. A. (1957). The application of com(1999). Angiosperm phylogeny inferred from multiple genes as a tool for comparat ive biology. Nature 90: 461-470. puters to taxonomy. J. Gen. Microbiol. 17: 201-226. 698 Plant Systematics Soltis, P. S., D. E. Soltis , M. J. Zanis and S. Veg. 132 publ. 1995; suppl. 1 (Aa-Ba) Regnum Kim. (2000). Basal lineages of angiosperms: Relationships and implications for f loral evo-lution. International Journal of Plant Sciences Veg. 125 publ. 1992). Stanier, R. Y. & C. B. van Niel (1962). The concept of a bacterium . Arch. Microbiol. 161 (6, suppl.): S97-S107. 42: 17 35. Soltis, P. S., D. E. Soltis , M. W. Chase, M. E. Starr, J. R., S. A.Harris & D. A.Simpson. (2004). Mort, D. C. Albach, M. Zanis, V. Savolainen, W. H. Hahn, W. H. Hoot, M. F. Fay, M. Axtell, S. M. Swensen, L. M. Prince, W. J. Cress, K. C. Nixon and J. A. Farri s. (2000a). An-giosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequenc es. Bot. J. Linn. Soc. Phylogeny of the unispicate taxa in Cyperaceae tribe Cariceae I: Generic relatio nships and evolutionary scenarios. Syst. Bot. 29: 528-544. Starr, J. R., V. Teoh, E. Roalson, A. M. Muasya 133: 381-461. & D. A. Simpson. (2006). Towards a phylogenetic classification of sedges (Cypera ceae): chloroplast (rbcL, matK, NdhF) and nuclear (ADC) data. Pp. 258-259, in Bo tany 2006 Looking to the Future, Conserving the Past. [AbSoltis, D. E., P. S. Soltis, P. K. Endress, and M. W. Chase.(2005). Phylogeny and Evolution of Angiosperms. 370 pp. Sinauer Associa tes, Sunderland, MA. stracts: Botanical Society of America, etc.]. Soreng, R. J. and J. I. Davis. (1998). Phylogenetics Stebbins, G. L. (1950). Variation and Evolution in and character evolution in the grass family (Poaceae): Simultaneous analysis of

morpho-logical and chloroplast DNA restriction site Plants. Columbia Univ. Press, NY. Stebbins, G. L. (1974). Flowering Plants; Evol ucharacter sets. Bot. Rev. 64: 1-85. tion above the Species Level. The Belknap Press, Harvard Univ. Press, Cambridge. Soros, C. L. and D. H. Les. (2002). Phylogenetic Steenis, C. G. G. J. van (Ed.). (1948). Flora relationships in the Alismataceae. P. 152, in Botany 2002: Botany in the Curricu lum, AbMalesiana. Series I: Spermatophyta. Groninger, Jakarta. stracts. [Madison, Wisconsin.] Stefanovic, S., L. Krueger and R. G. Olmstead. Sosef, M. S. M. (1997). Hierarchical models, (2002). Monophyly of the Convolvulaceae and circumscription of their major linea ges based on DNA sequences of multiple chloroplast loci. American J. Botany 89: 1510-1522. reticulate evolution and the inevitability of paraphyletic supraspecific taxa. T axon 46: 75-85. Speta, F. (1979). Weitere untersuchungen fiber Stevens, P. F. (1994). The development of biologiproteinkrper in Zellkernen und ichre taxonomische Bedentung. Plant Syst. Evol. cal systematics. Columbia University Press, New York. 132: 1-126. Stevens, P. F. (1998). What kind of classificaSporne, K. R. (1971). The mysterious origin off lowtion should the practicing taxonomist use to be saved? in Plant diversity in Mal esia III, J. Dransfield, M. J. E. Coode and D. A. Simpson (Eds.) , 295-319. Roya l Botanical ering plants. Oxford Biology Readers 3, F. F. Head and O. E. Lowenstein (Eds.). Oxford Univ. Press, Oxford. Sporne, K. R. (1974). Morphology of Angiosperms. Gardens, Kew. Hutchinson Univ. Library, London. Stevens, P.F. (2003). Angiosperm Phylogeny Sporne, K. R. (1976). Character correlation among Website. Version 6, May 2005 [and more or less continuously updated since]. http :// angiosperms and the importance of fossil evidence in assessing their significanc e. In: C. B. Beck (Ed.). Origin and Early Evolution of www.mobot.org/MOBOT/research/APweb. Angiosperms. Columbia Univ. Press, New York. Website developed and maintained by Hilary Davis (last updation incorcorated 03/ 04/ Stace, C. A. (1973). Chromosome numbers in 2006). British Species of Calystegia and ConvolvuStevenson, D. W. (1981). Observations on ptyxis, lus. Watsonia 9:363-367. phenology and trichomes in the Cycadales and their systematic implications. Am. J. Bot. Stace, C. A. (1980). Plant Taxonomy and Biosystematics. Edward Arnold, London. 68: 1104-1114. Stace, C. A. (1989). Plant Taxonomy and BiosysStevenson, D. W. and H. Loconte. (1995). Clatematics. (2nd ed.) Edward Arnold, London. distic analysis on monocot families. In M. J. Rudall et al. (Eds.) Monocotyledon

s: Systematics and Evolution, 543-578. Royal Botanic Gardens, Kew. Stafleu, F. A. and E. A. Mennega. (1997). Taxonomic literature, 2nd ed., suppl. 4 (Ce-Cz). Regnum Veg. 134. (suppl. 3 (Br-Ca) Regnum References 699 Stewart, W. N. and G. W. Rothwell. (1993). Pa(Thunb.) Nakai (Chloranthaceae): additional evidence for the occurrence of vesse ls. Bot. leobotany and the evolution of plants. 2nd ed. Cambridge University press, Cambr idge. Mag. (Tokyo) 101: 387-395. Steyermark, J. A. (1963). Flora of Missouri. Iowa Takhtajan, A. L. (1958). Origin of Angiospermous State Univ. Press, Ames, IA, 1725 pp. Plants. Amer. Inst. Biol. Sci. (Translation Strahler, A. N. and A. H. Strahler. (1977). GeogRussian edition of 1954). raphy and Man s Environment. John Wiley and Takhtajan, A. L. (1959). Die Evolution Der Sons, New York. Angiospermen. Gustav Fischer Verlag, Jena. Stuessy, T. F. (1990). Plant Taxonomy. Columbia Takhtajan, A. L. (1966). Systema et Phylogenia Magnoliophytorum. Soviet Publishi ng InstituUniv. Press, New York. Sun, B.-Y. (2002). Characteristics of fern flora of tion, Nauka. Korea with emphasis on the endemic genus Mankyua (Ophioglossaceae) from Cheju Is land, Korea. First Korean Academy of Science and Technology/Hungary Academy of S cience Bilateral Symposium Proceedings 1: 62-68. Takhtajan, A. L. (1969). Flowering Plants-Origin and Dispersal.(English translation by C. Jeffrey). Smithsonian Institution Press, Washington. Takhtajan, A. L. (1980). Outline of classification of flowering plants (Magnoliophyta). Bot. Rev. Sun, B.-Y., M.H. Kim, C.H. Kim, and C.-W. Park. 46: 255-369. (2001). Mankyua (Ophioglossaceae): a new fern genus from Cheju Island, Korea. Ta xon Takhtajan, A. L. (1986). Floristic Regions of the World. Berkeley. 50: 1019-1024. Takhtajan, A. L. (1987). Systema MagnolioSun G., G. L. Dilcher, S. Zheng and Z Zhou. phytorum. Nauka, Leningrad. (1998) In search of the first flower: A jurassic angiosperm, archaefructus, from northeast Takhtajan, A. L. (1991). Evolutionary Trends in Flowering Plants. Columbia Univ. Press, New china Science. 282: 1692-1695. York. Sun, G., Q. Ji, D. L. Dilcher, S. Zheng, K. C. Takhtajan, A. L. (1997). Diversity and ClassificaNixon and X. Wang. (2002). Archaefru-ctaceae, a new basal angiosperm family. Sci tion of Flowering Plants. Columbia Univ. Press, New York, 642 pp. ence 296: 899-904. Tanaka, N[orio], H. Setoguchi and J.Murata. Surange, K. R. and S. Chandra. (1975). Morphol-

(1997). Phylogeny of the family Hydrocharitaceae inferred from rbcL and matK gen e ogy of gymnospermous fructifications of the Glossopteris flora and their relatio nships. Palaeontographia 149: 153-180. sequence data. J. Plant Res. 110: 329-337. Taylor, D. W. (1981). Paleobotany: An Introduction Sutter, D. and P. K. Endress. (1995). Aspects of to Fossil Plant Biology. McGraw Hill, New York. gynoecial structure and macrosystematics in Euphorbiaceae. Bot. Jahrb. Syst. 116 : 517Taylor, D. W. and G. Kirchner (1996). Origin and evolution of angiosperm carpel. In: D. W. Tay536. lor and L. J. Hickey (Eds.). Flowering Plant Origin, Evolution and Phylogeny. Ch apman & Swain, T. (1977). Secondary compounds as protective agents. Ann. Rev. Pl. Physiol. 28: 479-501. Hall Inc., New York, pp. 116-140. Swenson, U., & Anderberg, A. A. 2005. Phylogeny, Taylor, D. W. and L. J. Hickey. (1992). Phylogecharacter evolution, and classification of netic evidence for herbaceous origin of angiosperms. Plant Systematics and Evolu tion Sapotaceae (Ericales). Cladistics 21: 101-130. Sytsma, K. J. and D. A. Baum. (1996). Molecular 180: 137-156. phylogenies and the diversification of the angiosperms. In: D. W. Taylor and L. J. Hickey (Eds.) Flowering Plant Origin, EvoluTaylor, D. W. and L. J. Hickey. (1996). Evidence tion and Phylogeny. Chapman & Hall Inc., for and implications of an Herbaceous Origin of Angiosperms. In: D. W. Taylor an d L. J. Hickey (Eds.). Flowering Plant Origin, Evolution and Phylogeny. Chapman & Hall Inc., New York, pp. 314-340. Sytsma, K. J., J. Morawetz, J. C. Pires, M. New York, pp. 232-266. Nepokroeff, E. Conti, M. Zjhra, J. C. Hall, and M. W. Chase. (2002). Urticalean rosids: Circumscription, rosid ancestry, and phylogenetics based on rbcL, trnL-F , and ndhF Terrell, E. (1983). Proposal (695) to conserve the name of the tomato as Lycopersicon esculentum P. Miller and reject the combination Lycopersicon lycopersicum (L.) Karsten sequences. American J. Bot. 89: 1531-1546. (Solanaceae). Taxon 32: 310-313. Takahashi, A. (1988). Morphology and ontogeny of stem xylem elements in Sarcandr a glabra Theophrastus. De causis plantarum Translated by R. E. Dengler. Philadelphia (192 7). 700 Plant Systematics Theophrastus. Enquiry into plants. Translated by Treutlein, J., G. F. Smith, B. -E. van Wyk, and A. Hort. W. Heinemann, London (1916), M. Wink. (2003). Phylogenetic relationships in Asphodelaceae (subfamily Alooidea e) inferred from chloroplast DNA sequences (rbcL, matK) and from genomic fingerprint2 vols. Thorne, R. F. (1968). Synopsis of a putative phylogenetic classification of flowering plants.

Aliso 6(4): 57-66. ing (ISSR). Taxon 52: 193-207. Thorne, R. F. (1974). A phylogenetic classification of Annoniflorae. Aliso 8: 147-209. Thorne, R. F. (1976). A phylogenetic cla ssification of angiosperms. Evol. Biol. 9: 35-106. Troitsky, A. V., Y. F. Melekhovets, G. M. Thorne, R. F. (1981). Phytochemistry and anRakhimova, V. K. Bobrova, K. M. Valiegoroman and A. S. Antonov. (1991). Angiospe rm origin and early stages of seed plant evolution deduced from rRNA sequence co mparison. giosperm phylogeny: A summary statement, In D. A. Young and D. S. Seigler (Eds.) . Phy-tochemistry and Angiosperm Phylogeny. J. Molecular Evolution 32: 253-261. Tsutsumi, C. & M. Kato. (2006) Evolution of epiphytes in Davalliaceae and related ferns. Praeger, New York, pp. 233-295. Bot. J. Linn. Soc. Thorne, R. F. (1983). Proposed new realignments in angiosperms. Nordic J. Bot. 3:85-117. Thorne, R. F. (1992). An updated classi fication of the flowering plants. Aliso 13: 365-389. Tucker, S. C. and W. Douglas. (1996). Floral Thorne, R. F. (1992b). Classification and geograstructure, Development and Relationships of Paleoherbs: Saruma, Cabomba, Lactori s, and selected Piperales. In: D. W. Taylor and L. J. Hickey (Eds.) Flowering Pl ant Origin, Evolution and Phylogeny 141-175. Chapman and phy of the flowering plants. Bot. Rev. 58: 225348. Hall, Inc., New York. Thorne, R. F. (1996). The least specialized anTucker, S. C. (2002). Floral ontogeny of Cercis giosperms. In: D. W. Taylor and L. J. Hickey (Eds.). Flowering Plant Origin, Evo lution and Phylogeny. Chapman & Hall Inc., New York, (Leguminosae: Caesalpinioideae: Cercideae): Does it show convergence with Papili onoids? Int. J. Plant Sci. 163: 75-87.) pp. 286-313. Tutin, T. G. et al. (ed.) (1964-1980). Flora Thorne, R. F. (1999). An Updated Classification of Europaea.. Cambridge University Press, Camthe Class Angiospermae.http:// bridge, 5 vols. w w w . inform. u m d . e d u / P B I O / f a m Uhl, N. W. (1947) Studies in the floral morphology and anatomy of certain member s of Helothorneangiosp99.html. Website maintained (with nomenclatural additions) by Dr. J. L. biae. Ph.D. Thesis (Cited by Lawrence, 1951). Reveal of University of Maryland. Uhl, N. W., J. Dransfield, J. I. Davis, M. A. Thorne, R. F. (2000). The classification and Luckow, K, H. Nansen and J. J. Doyle. (1995). Phylogenetic relationships among p alms: Cladistic analysis of morphological and chloroplast DNA restriction site v ariation. In: P. J. Rudall, P. J. Cribb., D. F. Cut-ter and C. J. Humphries (Eds .). Monocotyledons: Systematics and Evolution. 623-661. geography of the monocotyledon subclasses Alismatidae, Liliidae, and Comelinidae . Pp. 75-124, in Nordenstam, B., El Ghazaly, G.,

& Kassas, M. (Eds.), Plant Systematics for the 21st Century. Portland, Oregon. Thorne, R. F. (2000) [2001]. The classification and geography of flowering plants: Dicotyle-dons of the class Angiospermae (subc lasses Magnoliidae, Ranunculidae, Caryophyllidae, Dilleniidae, Rosidae, Asterida e, and Upchurch, G. R. Jr. and J. A. Wolfe. (1987). MidLamiidae). Bot. Rev. 66: 441-647. Cretaceous to Early Tertiary vegetation and climate: evidence from fossil leaves and woods. In: E. M. Friis, W. G. Chaloner, and P. R. Crane (Eds.). The Origin of Angiosperms and their Biological Consequences. Cambridge Thorne, R. F. (2003). An Updated Classification of the Class Angiospermae. www.r sabg.org/pubUniv. Press, Cambridge, pp. 75-105. lications/angiosp.htm. Valentine, D. H. and A. Love. (1958). Taxonomic Tournefort, J. P. de. (1696). Elements de botanique. and biosystematic categories. Brittonia 10: Paris, 3 vols. 153-166. Tournefort, J. P. de. (1700). Institutiones rei Vegter, I. H. (1988). Index herb ariorum: a guide herbariae. Imprimerie Royale, Paris, 3 vols. to the location and contents of the world s public herbaria. Part 2(7). Collectors T-Z. Regnum Veg. 117. (Part 2(6). Collectors S. Regnum Veg. 114 publ. 1986). Trehane, P., et al. (1995). International code of nomenclature for cultivated plants. Regnum Veg. 133. References 701 Wallace, R. S. and A. C. Gibson. (2002). EvoluWiley, E. O. (1978). The evolutionary species concept reconsidered. Syst. Zool. 27: 17-26. tion and systematics. Pp. 1-21 in Nobel, P. S. (ed.), Cacti: Biology and Uses. U niversity Wiley, E. O. (1981). Phylogenetics-The Theory and of California Press, Berkeley Practice of Phylogenetic Systematics. John Wallick, K., W. Elisens, P. Kores and M. Molvray. Wiley and Sons, New York. (2001). Phylogenetic analysis of trnL-F se-quence variation indicates a monophyl etic Buddlejaceae and a paraphyletic Buddleia. Pp. 148-149, in Botany 2001: Plan ts and William, W. T., J. M. Lambert and G. N. Lance. (1966). Multivariate methods in plant ecology. V. Similarity analyses and inform ationanalysis. J. Ecol. 54: 427-445. People, Abstracts. [Albuquerque.] Willis, J. C. (1922). Age and Area. Cambridge. Wallick, K., W. Elisens, P. Kores and M. Molvray. Willis, J. C. (1973). A. Dictionary ofF lowering Plants (2002). Phylogenetic analysis of trnL-F se-quence variation indicates a monophyl etic Buddlejaceae and a paraphyletic Buddleia. Pp. 156-157 in Botany 2002: Botan y in the Cur-riculum, Abstracts. [Madison, Wisconsin.] and Ferns, 8th ed. (revised by H. K. Airy-Shaw). Cambridge Univ. Press, Cambridge, 1245 pp. Wilson, P. G., M. M. O Brien, P. A. Gadek and C. J. Quinn (2001). Myrtaceae revisited: A reassessment of infrafamilial groups. AmeriWalters, D. R. and D. J. Keil. (1995). Vascular can J. Bot. 88: 2013-2025. Plant Taxonomy, 4th ed. Kendall?Hunt,

Woese, C. R., W. E. Balch, L. J. Magrum, G. E. Dubuque, IA. Fox and R. S. Wolfe (1977). An ancient divergence among the bacteria . Journal of Wannan, B. S. (2006). Analysis of generic relationships in Anacardiaceae. Blumea 51: 165Molecular Evolution 9: 305 311. 195. Woese, C. R. O. Kandler and M. L. Wheelis Watson, L. and M. J. Dallwitz. (2000). The fami(1990). Towards a Natural System of Organ-isms: Proposal for the Domains Archaea, Bacteria, and Eucarya . Proceedings of the National Academy of Sciences 87: 4576 45 79. lies of flowering plants: Descriptions and i l l u s t r a t i o n s .Web site: h t t p : / / biodiversity.uno.edu/delta/ . Webster, G. L. (1967). The genera of Wojciechowski, M. F., M. Lavin and M. J. Euphorbiaceae in the southeastern United Sanderson (2003). A phylogeny of legumes based on sequences of the plastid matK gene, P. 99 in Botany 2003: Aquatic and Wetland Plants: Wet and Wild. Abstracts. [ Mobile, States. J. Arnold Arbor. 48: 303-430. Webster, G. L. (1994). Classification of the Euphorbiaceae. Ann. Missouri Bot. Gard. 81: 3-32. 1994. Alabama.] Wendel, J. F., A. Schnabel and T. Seelanan. Wolfe, K. H., M. Gouy, Y. -W. yang, P. M. Sharp (1995). An unusual ribosomal DNA dequence from Gossypium gossypioides reveals an cient, cryptic, intergenomic, introgression. Mol. Phylog. Evol. 4: 298-313. and W. -H. Li. (1989). Date of monocot-dicot divergence estimated from Choloropl ast DNA sequence data. Proc. Nat. Acad. Sci. USA 86: 6201-6205. Wettstein, R. R. von. (1907). Handbuch der Woodland, D. W. (1991). Contemporary Plant Syssystematischen Botanik (2nd ed.) Franz tematics. Prentice Hall. New Jersey. Deuticke, Leipzig. Worsdell, W. C. (1908). The affinities of Paeonia. Williams, S. E., V. A. Albert and M. W. Chase. J. Bot. (London) 46: 114 (1994). Relationships of Droseraceae: A cla-distic analysis of rbcL sequence and morphoWu, Z-Y, A. M. Lu, $ Y. C. Tang. (1998a). A Comlogical data. American J. Bot. 81: 1027-1037. prehensive study of Magnoliidae sensu lato - with special consideration on possibi lity and necessity for proposing a new polyphyleticpolychronic-polytopic system of angiosperms. In: Zhang, A. I & S. G. Wu (Eds.) Floristic Characteristics and Di versity of East Asian Plants. Beijing & Berlin: China Higher Whitlock, B. A., J. Lee, O. Dombrovska and Y. L. Qiu. (2002). Effects of rate heterogeneity on estimates of the age of angiosperm s. P. 158, in Botany 2002: Botany in the Curriculum, Abstracts. [Madison, Wisconsin.] Whittaker, R. H. (1969). New concepts of kingEducation Press & Springer Verlag, p. 269-334. doms of organisms . Science 163: 150 160. Wu, Z-Y, Y. C. Tang, A. M. Lu et al. (1998b).On Wieland, G. R. (1906). American fossil cycads. Carnegie Institute, Washington, D. C. Wieland, G. R. (1916). American fossil cyc ads vol. II. Carnegie Institute, Washington, D. C.

primary subdivisions of the Magnoliophytatowards a new scheme for an eight-class system of classification of the angiosperms. Acta. Phytotax Sinica. 36: 385-402 . 702 Plant Systematics Wu, Z-Y, A. M. Lu, Y-C Tang, Z-D Chen and D-Z Young, D. A. (1981). Are the angiosperms primitively vesselless? Syst. Bot. 6: 3 13-330. Li. (2002). Synopsis of a new polyphyletic-polychronic-polytopic system of the anYoung, D. J. and L. Watson. (1970). The classifigiosperms. Acta Phytotax. Sinica 40: 289-322. cation of the dicotyledons: A study of the upper levels of hierarchy. Aust. J. B ot. 8: 387 Wu, Z-Y, A. M. Lu, Y-C Tang, Z-D Chen and D-Z Li. (2003). The Families and Genera of An-giosperms in China: A Comparative Anal ysis. 433. Zahn, L. M., J. H. Leebens-Mack, C. W. Science Press, Beijing. dePamphilis, H. Ma & G. Theissen. (2005). To B or not to B a flower: The role of DEFICIENS and GLOBOSA orthologs in the evolution of the angiosperms. J. Hered. 96: Wu, H.-C., H.-J. Su, & J.-M. Hu. (2007). The identification of A-, B-, C-, and E-class MADS-box genes and implications for pe ri-anth evolution in the basal eudicot Trochodendron aralioides (Trochodendracea e). 225-240. Zanis, M. J., D. E. Soltis, P. S. Soltis, Y-L Qiu Internat. J. Plant Sci. 168: 775-799. and E. A. Zimmer. (2003). Phylogenetic analyses and perianth evolution in basal anWurdack, K. J. and M. W. Chase. (2002). Phylogenetics of Euphorbiaceae s. str. using plastid (rbcL and trnL-F) sequences . P. 160, in Botany 2002: Botany in the Curricugiosperms. Ann. Missouri Bot. Gard. Zanis, M. J., D. E. Soltis, P. S. Soltis,Y-L Qiu, lum, Abstracts. [Madison, Wisconsin.] S. Mathews, and M J. Donoghue. (2002). The root of the angiosperms revisited. Pr oc. Nat. Xiang, Q.-Y., D. E. Soltis, and P. S. Soltis. (1998). Acad. Sci. 99: 6848-6853. Phylogenetic relationships of Cornaceae and close relatives inferred from matK a nd rbcL sequences. Amer. J. Bot. 85: 285-297. Zgurski, J. M., H. S. Rai, Q. M. Fai, D. J. Bogler Yuan, Y.-M., Y. Song, K. Geuten, E. and J. Francisco-Ortega. (2008). How well do we understand the overall backbone of cycad phylogeny? New insights from a large, multigene plastid data set. Mol. Phyl. Evol. Rahelivololona, S. Wohlhauser, E. Fischer, E. Smets, and P. Kpfer. (2004). Phylog eny and biogeography of Balsaminaceae inferred from ITS sequences. Taxon 53: 391 -403. 47: 1232-1237. Zhou, Z.K., P. M. Barrett, & J. Hilton . (2003). Yoo, K.-O. and J. Wen. (2002). Phylogeny and An exceptionally preserved Lower Cretaceous ecosystem. Nature 421: 807-814. biogeography of Carpinus and subfamily Coryloideae (Betulaceae). Int. J. Plant S ci. 163: 641-650. Zomlefer, W. B. (1994). Guide to flowering plant families. University of Carolina Press, Chapel Hill.

2 HISTORY OF PLANT CLASSIFICATION It is customary to trace the history of all sciences with the Greeks inspite of the fact that many disciplines either originated in India or China or Egypt, Ass yria or even in ancient S. America (Maya, Inca and Aztec civilizations). As far as the plant sciences are concerned there are definite evidences that they origi nated much earlier to the early Greek writings. Hence we shall attempt to give a brief outline of this science in India first. 2.1. History of Plant Classification in India : According to Majumdar history of botany and plant taxonomy in India can be trace d back to the period of Rig Veda 3000 W.C. (and earlier ?) and later Hindu liter ature including Manu, Agnipurana, Vrikshayarveda etc. A broad classification of plants into trees, shrubs, herbs and creepers is to be found in the mantras (hym ns) of Rig Veda. Plants were further classified into those that produced fruits and flowers and those without fruits and flowers. Thus as early as 3000 B.C. we come across a classification approaching Phanerogamia and Cryptogamia of Eichler . Manu classified the plants more distinctly. They are : 1. Oshadhis annual plants. 2. Vanaspatis trees bearing fruits without evident flowers. 3. Vrikshas trees bearing flowers and fruits. 4. Guccha bushy or shrubby plants. 5. Gulmas succulent shrubs. 6. Trinas grasses. 7. Pratnas procumbent and decumbent herbs. 8. Vallis twiners. Charaka in his Charak Samhita also gives a classification similar to the above. Su sruta also follows a similar pattern. Apart from the above, the ancient Indians also classified plants according to their medicinal properties and dietic proper ties e.g. the ayurvedic systems of Charaka, Susruta and others. Susruta classifi ed medicinal plants under thirty seven ganas or sections. Plant classification, based upon their dietic properties was also attempted by Charaka and Susruta. Regarding the system of nomenclature the ancient Indian system adopted double na mes one based upon external features and the other based upon some special medic inal, dietic or other properties. This system of double naming is of course not synonymous with the binomial system devised by Linnaeus. 2.2. The Greek Period : rlL 1.;nraf,..3 #0Theophrastus (370-287 B.C.) was a student and contemporary of Aristotle and was particularly interested in plants. His work is entitled De Historia Plantarum in which he described about 480 kinds of plants and divided them primarily on the basis of habit into trees, undershrubs, (13) 14 Plant Taxonomy herbs, cultivated and wild plants. Theophrastus followed the Platonic method of logical division i.e. any given object is either A or not A. His main point of e nquiry was : what is the essential nature ? What is its difference (Thrill 1942) . After Theophrastus we do not get any notable name in the field of botany or as a matter of fact in any of the branches of science. It seems the Greek civiliza tiion degenerated and died out towards the beginning of the Christian era. 2.3. The Roman Period : With the advent of the Roman civilization we come across certain important works towards plant classification. Pliny the Elder (23-79 A.D.) He was a Roman naturalist and mentioned nearly a th ousand plants in his "Historical Naturalis". This is a series of 37 books dealin g with plants on the basis of medicinal properties, timber and practices of hort iculture. Dioscorides (first century A.D.) was a physician of Rome and worked in the milit ary of Emperor Nero. He complied De Materia medica in which he described about 600 species and gave t heir accounts and practical uses. The book included perfumes, oils, spices, cere

als, condiments, wines etc. He even got the idea of groups such as the Labiates and Umbellifers. The book was documented with original illustrations increasing its value and usefulness greatly. 2.4. The Dark Age, Herbals and the Transition period : During the middle age a dark period descended over Europe and there was no signi ficant botanical work. Most workers copied the work of Dioscorides without much addition. Wood cut of plants were prepared for illustrating them. Such were the herbals and included Albertus Magnus, Brunfels, Bock, Fuchs, Bauhin etc. Albertus Magnus (1193-1280) He was bishop of Ratisbon and is believed to have fi rst differentiated monocots from dicots on the basis of stem structure. Otto Brunfels (1464-1434) He was first to describe and to some extent illustrate plants and distinguish perfecti and imperfecti groups of plants on the presence and absence of flowers. Jerome Bock (1498-1554) He was a herbalist and classified plants as herbs, shrub s and trees providing suitable notes on natural distribution of many plants. His famous work "New Krenterbuch" gave some fine description of plants. Caesalpinio (1519-1637) Wrote the book "De Plantis" consisting of sixteen books with description of 1500 plant species. His classification was also based upon herbs, shrubs and trees but within these he recognised the significance of fruit and s eed characters. Caesalpinio is remembered by the genus Caesalpinia in the family Fabaceae. Jean Bauhin (1541-1631), a French and Swiss physician, His important contributio n was "Historia Plantarum Universalis", in which he dealt about 5000 plants. Gaspard Bauhin (1560-1624), brother of Jean Bauhin. Gaspard published "Pinax" (1 623) containing names and synonyms of about 6000 species. He was the first to di stinguish nomenclaturally between species and genus. Joachin Jung (1587-1657), a German mathematician and the first terminologist. He was the first to define the following terms nodes and internodes; blade and pet iole; simple' and compound leaf; stamens and styles etc. John Ray (1627-1705) was an English philosopher and naturalist. In his Methodus Plantarum' 18000 species were classified. He established for the first time the presence of one or two cotyledons in the angiosperms classification. His classif ication of the major groups was as follows : A, History of Plant Classification I. Herbae A. Imperfectae i.e. flowerless (the cryptogams). B. Perfectae i.e. flowering plants. Dicotyledones embryo with 2 cotyledons. Monoco tyledones embryo with 1 cotyledon. II. Arbores i.e. trees and shrubs. Dicotyledons. Monocotyledons. Pierre Magnol (1638-1715) was a French botanist. He divided the plants into grou ps, what he called families. He was the first to give the concept of modern fami lies. His name was commemorated by generic name Magnolia. Joseph Pitton de Tournefort (1656-1708) He recognised petal bearing and nonpetal bearing flowers, corollas with separate and with united petals and regular and irregular corollas. Rudolf Camerarius (1665-1721) He was professor and director of botanical garden at Germany. He established the fact of sexuality in flowering plants. 2.5. Modern Period : The modern period of classification starts with Linnaeus and continues even toda y. During this period numerous systems have been proposed. Broadly speaking they belong to the pre-Darwinian period and post-Darwinian period. Before Darwin the concept of evolution was not an accepted principle and the general belief was i n the fixity of species. In the post-Darwinian period evolutionary theory brough t about a fundamental change in the concept of classification. Generally speakin g in the pre-Darwinian period the systems proposed were either artificial or nat

ural. An artificial system is one in which plants classifies according to a few convenient characters for the purpose of identification without paying any atten tion to affinities. Linnaeus' sexual system is an example of artificial system. In fact all classifications prior to Linnaeus were also artificial. A natural sy stem is one which is based upon overall resemblances between plants and groups t aking external morphology as the main criterion. Systems of classifications afte r Linnaeus but prior to Darwin are supposed to be natural systems. A phylogeneti c system on the other hand tries to determine evolutionary relationships of plan ts and groups and arrange them accordingly. Thus those plants which were evoluti onarily more related would stand closer together than those which were distantly related. Such systems as those of Endlicher, Eichler, Engler, Rendle and Hutchi nson would come under this category. 2.5.1. Sexual (or artificial) System of Linnaeus Carolus Linnaeus (1707-1778), a Swedish naturalist published a sexual system of classification in Hortus Uplandicus (1730) and elaborated it in his Genera Plant arum (1737). The latter work is of importance in modern taxonomy as a source of description of 935 genera. The "Spegies Plantarum" published by Linnaeus in 1753 has been chosen by modern taxonomists as a starting point of present day nomenc lature. Linnaeus classified all plants into 24 classes as follows : 1. Monandria stamen one e.g. Lemna. 2. Diandria stamens two e.g. Veronica. 3. Triandria stamens three e.g. Iris. 4. Tetrandria stamens four e.g. Mentha. 5. Pentandria stamens five e.g. Primula. 6. Hexandria stamens six e.g. Rumex. 15 tr. 16 Plant Taxonomy 7. Heptandria-stamens seven. 8. Octandria-stamens eight. 9. Enneandria-stamens nine. 10. Decandria-stamens ten. 11. Dodecandria-stamens 11. 12. Icosandria-stamens more than twelve, atached to calyx, e.g. Rosa. 13. Polyandria-stamens more than 12 attached to the receptale e.g. Papaver. 14. Didynamia-stamens didynamous (2 long, 2 short), e.g. Linaria. 15. Tetradynamia-stamens tetradynamous (4 long, 2 short), e.g. members of Brassi caceae (Cruciferae). 16. Monadelphia-stamens monadelphous, (in one bundle) e.g. Malvaceae. 17. Diadelphia-stamens diadelphous (in 2 bundles) e.g. Lathyrus. 18. Polyadelphia-stamens polyadelphous (in several bundles). 19. Syngenesia-stamens syngenesious (with united antherlobes), e.g. Asteraceae ( Compositae). 20. Gynandria-stamens gynandrous (adnate to the pistil) e.g. Aristolochia. 21. Monoecia-plant monoecious (unisexual flowers on the same plant), e.g. Typha. 22. Dioecia-plant dioecious (unisexual flowers on the separate plants), e.g. Sal ix. 23. Polygamia-plants polygamous. 24. Cryptogamia-flowers concealed (the cryptogams-algae, fungi, mosses and ferns ). Michel Adanson (1727-1806) in his classical work : "Families des plantes" (1763) published for the first time a natural system of classification and described t axa more or less equivalent to modern orders and families. Jean. B. de Lamark (1744-1829) was a French biologist. In 1778 he published his classical work "Flore Francoise". In this work he laid down the principles of a natural classification : (1) the d etermination of which a plant precedes another in a natural series; (2) the rule s for natural grouping of species and (3) the treatment of order and sub-familie s. Lamark is well known for his theory - the Lamarckism.

De Jussieu (1699-1777), a French contemporary of Linnaeus. He did not publish hi s results, but his nephew, Antoine Laurent de Jussieu (1748-1836) published his uncle's plan along with his own in "Genera Plantarum Secundum Ordines Naturals D isposita" (1789). His natural classification included a hundred orders (now most ly recognised as families) classified in fifteen classes as follows : 1. Acotyledones-cotyledones absent, cryptogams and some aquatic flowering plants . 2. Monocotyledones, stamens hypogynous. 3. Monocotyledones, stamens perigynous. 4. Monocotyledones, stamens epigynous. 5. Dicotyledones, Apetalae - stamens epigynous. 6. Dicotyledones, Apetalae stamens perigynous. 7. Dicotyledones, Monopetalae - stamens hypogynous. 8. Dicotyledones, Monopetalae - corolla hypogynous. 9. Dicotyledones, Monopetalae - corolla perigynous. 10. Dicotyledones, Monopetalae-corolla epigynous, anthers united. 11. Dicotyledones, Monopetalae-corolla epigynous, anthers distinct. 12. Dicotyledones, Polypetalae-stamens epigynous. History of Plant Classification 17 13. Dicotyledones, Polypetalae stamens hypogynous. 14. Dicotyledones, Polypetalae stamens perigynous. 15. Diclines, irregulares Gymnosperms, Urticaceae, Euphorbias. Jussieu's conception of perigyny is different from that of the present day. Augustin Pyrame de Candolle (1778-1841). His views were expressed in his Theorie elementaire de la botanique (1813). He laid much stress on morphological charac ters and also recognised the significance of vestigial organs. The three generat ions of de Candolle collaborated for the preparations of the monumental work "Pr odromus systematis naturalis ragni vegetabilis" (1824-1823) in which every speci es then known to science was described and classified. 2.5.2. Natural System of A.P. de Candolle Synopsis of A.P. de Candolle's system of classification. I. Vascular Plants with Cotyledon. (I) Exogens or Dicotyledons; vessel arranged in a ring and embryo with two cotyl edons. (A) flowers with perianths differentiated into calyx and corolla, i.e. perianth leaves in two series. (a) Thalamiflorae Petals free on the receptacle. Cohort 1 Carpels many and stamens opposite to the petals. Natural orders 1-8, Ranunculaceae, etc. Cohort 2 parietal placentation, carpels solitary or many, connate. Natural orders Cruciferae, Violaceae etc. Cohort 3 Ovary single. Natural orders Caryophyllaceae, Rutaceae etc. Cohort 4 Fruit gynobasic. Natural order Simarubeae tc. (b) Calyciflorae Flowers perigynous, petals free or fused, ovary inferior or sup erior. Natural orders epigynous flowers with gamopetalous corolla. e.g. Rubiacea e, Compositae. (c) Corolliflorae Petals united, ovary superior. Natural order Acanthaceae, Ascl epiadaceae etc. (B) Monochlamydeae Perianth in one whorl. (II) Plants with scattered vascular bundles, cotyledon one. It includes Monocoty ledons and Cycadaceae with 27 orders. II. Plants without cotyledons (a) Foliaceous Mosses and Liverworts. (b) Non-foliaceous of unknown sexuality. Fungi, Lichens, etc. Robert Brown (1773-1858), a Scottish botanist. He published no system of his own but proved that the Gymnosperms were a discrete group with naked ovules and see ds. Adolphe Theodore Brongniart (1801-1876), a French botanist treated Apetalae as a

reduced members of a polypetalae. 2.5.3. Natural System Classification of Bentham and Hooker George Bentham (1800-1884) and Sir Joseph Hooker (1817-1911) British botanists a ssociated with Royal Botanic Gardens, Kew, contributed an important system of cl assification, as a 3 volume work in their Genera Plantarum (1862-1883). In this some 202 orders (now called families) were described. ( ot7 X 05 9fx,ie9) 18 Plant Taxonomy Summary of Bentham and Hooker's System of Classification* (1862-93) I. Dicotyledones Embryo with two cotyledons; stem with open bundles; usually reticulate venation; flowers usually tetra-or pentamerous : Group I. Polypetalae. Flowers usually with two whorls of perianth, inner whorl o r corolla free. Series I. Thalamiflorae. Sepals usually distinct and separate, free from ovary; petals 1, 2 or indefinite; stamens hypogynous,; ovary superior. \ It includes six orders from Ranales to Malvales. Series 2. Disciflorae. Sepals distinct or united, free from ovary; disk usually conspicuous as a ring cushion, or spread over the base of the calyx tube, or con fluent with the base of the ovary; stamens definite; ovary superior. It includes 4 orders from Geraniales to Sapindales. Series 32 Calyciflorae. Sepals united, rarely free, often adnate to ovary; petal s uniseriate, peri-or epi-gynous; disk adnate to base of calyx, rarely raised in to a torus or gynopliore; stamens perigynous, ovary often inferior. It includes 5 orders from Rosales to Umbellales. Group II. Gamopetalae. Petals are united. Series 1. Inferae. Ovary inferior; stamens usually many or as many.as corolla lo bes. It includes three orders i.e. Rubiales, Asterales, Campanulales. Series 2. Heteromerae. Ovary usually superior; stamens epipetalous or free from corolla, opposite or alternate to its segments or twice as many; carpels more th an two. It includes three orders from Ericales to Ebenales. Series 3. Bicarpellatae. Ovary usually superior; stamens as many as or less than corolla lobes, alternate to them, carpets two rarely one or three. It includes four order from Gentianales to Lamiales. Group III. Monochlamydeae or Incompletae. Flowers usually with one whorl of peri anth, commonly sepaloid or none. Series 1. Curvembryct e. Terrestrial plants with usually hermaphrodite flowers; stamens generally equal to perianth segments; ovule usually solitary; embryo cur ved. It includes seven families. Series 2. Multiovulatae Aquaticae. Aquatic plants with syncarpous ovary and inde finite ovules. Series 3. Multiovulatae Terrestres. Terrestrial plants with synca rpous ovary and indefinite ovules. Series 4. Microembryae. Ovary syncarpous or apocarpous; ovules solitary or few; embryo very small surrounded by endosperm. Series 5. Daphnales. Ovary usually of one carpel, ovules solitary or few; perian th perfect, sepaloid; stamens perigynous. Series 6. Achlamydosporeae. Ovary unilocular, one to three ovules; seed without testa. Series 7. Unisexuales. Flowers unisexual; ovary syncarpous or of one carp el; perianth much reduced or absent. Series 8. Anomalous Families. Unisexual families of doubtful or unknown affiniti es. II. Monocotyledones Embryo with one cotyledon; stem with closed vascular bundles; leaves with parall el venation; flowers usually trimerous. *Detail classification in appendix IV. I History of Plant Classification Series 1. Microspermae. Inner perianth petaloid; ovary inferior with parietal pl acentation; seed minute, exalbuminous. Series 2. Epigynae. Perianth partly petaloid; ovary usually inferior; endosperm abundant. Series 3. Coronarieae. Inner perianth petaloid; ovary usually free, su

perior; endosperm abundant Series 4. Calycinae. Perianth sepaloid, herbaceous, condition of ovary as in ser ies 3. Series 5. Nudiflorae. Perianth none or represented by hairs or scales; carpels o ne or several, syncarpous; ovary superior; ovules indefinite. Series 6. Apocarpae. Perianth in one or two whorls, or none, ovary superior, apo carpous; endosperm absent. Series 7. Glumaceae. Flowers solitary, sessile in the wdls of bracts and arrange d in heads or spikelety with bracts; perianth of scales or none; ovary unilocula r; one ovule; endosperm present. e.g. Gramineae, Cyperaceae. Critical assessment of Bentham and Hooker's System : This system though proposed after Darwin's work was not phylogenetic in approach . It was a natural system based upon Jussieu. and de Candollean scheme. The fami lies and groups are more natural and depended upon fusion and reduction of flora l parts. The Polypetalae were subdivided into Thalamiflorae, Disciflorae and Cal yciflorae. The gamopetalous families were grouped together under Gamopetalae and the apetalous families under Monochlamydeae. Merits and demerits of Bentham. and Hooker's System Merits. 1. The description of families and genera is very accurate. 2. The system is very handy for identification purposes. 3. The system is of great practical convenience. The British and Corimonw.ealth herbaria therefore still adopt this system in arr angement of families. 4. Each family had a synopsis at the beginning which is very useful in identific ation. 5. The system starts from Ranales, which are now universally considered to be mo st primitive living angiosperms. 6. Larger genera subdivided into subgenera and sections. 7. They believed in evolution through reduction and hence placed monocots after dicots; even in dicots, the dichlamydeous polypetalae and gamopetalae were place d before the uniseriate monochlamydeae. 8. The gamopetalae placed after polypetalae is justified since union of petals i s considered to be an advanced feature. 9. The polypetalae includes Thalamiflorae and Calyciflorae of de Candolle. But B entham and Hooker distinguished a new series Disciflorae which includes orders w hich cannot be assigned to Thalamiflorae or Calyciflorae. Thalamiflorae, Disciflorae and Calyciflorae show gradual evolut 10. The 3 series ionary advance from marked hypogyny to epigyny. 11. Treating Cucurbitaceae and Umballiferae (Apiaceae) at the end of Polypetalae as connecting links between poly and gamopetalous families. 12. Creation of Monochlamydeae at the end of Dicots. 13. Disputed families included in Ordines anomali. 14. Placing of unisexual monocot families after bisexual families e.g. Palmae an d Araceae after Liliaceae. 19 20 Plant Taxonomy 15. The series Glumaceae with extremely reduced flowers and inflorescences, plac ed at the end of the flowering plants. 16. The system was never conceived by its authors on the basis of phylogeny. The theory of organic evolution (theory of descent) was announced independently by Darwin and Wallace in 1859. So, any criticism of the system on the basis of phyl ogeny is not too justified. Demerits : 1. The system does not give any idea as to the evolutionary history of any genus , family or order. 2. In this system grouping of plants is mainly based on single and artificial ch aracters; with the result, that closely allied families are placed widely apart. 3. The group "Monochlaroydeae" is entirely artificial. 4. Gymnospermae is placed between the Dicotyledones and Monocotyledones, which i

s extremely anomalous. 5. The system does not show any phylogenetic relationship. The main demerit is t hat this system does' not give us any idea as to the evolutionary history of any genus, family or order nor does it give any idea of phylogenetic relationship b etween them. 6. Compositae (Asteraceae) is a highly advanced family and placed in Inferae at the beginning of Gamopetalae. 7. Advanced families like Orchidaceae and Scitamineae are treated in the beginni ng of monocots. 8. Liliaceae and Amaryllidaceae were kept, apart though they are very closely re lated. 9. The Amaryllidaceae is more allied to Liliaceae but is clubbed with Scitaminea e in series Epigynae, on account of inferior gynoecium. 10. The position of series Apocarpae is unsatisfactory due to its free and super ior 'carpels. 2.5.4. Phylogenetic Systems of Eichler August Wilhelem Eichler (1839-1887), a professor of botany in Germany. He modifi ed the system of Bentham and Hooker by placing gymnosperms in their proper seque nce. The outline of his system is as follows : 1. Cryptogamae Division : Thallophyta Bryophyta Pteridophyta 2. Phanaerogamae Division : Gymnospermae Angiospermae 2.5.5. Phylogenetic system of Engler and Prantl Adolf Engler (1844-1930), has adopted the main features of the Eichler's classif ication. Engler in association with Karl Prantl (1849-1893) published "Die Natur lichen Pflanzenfamilien" (1887-1915). In this monumental work were given details on the basis of morphology, anatt-thy and economic aspects of various families, profuse illustrations and keys to the genera dealt within a phylogenetic manner . Engler's system of classifications is based upon the trends of evolution in exis ting families. In this system plant kingdom is divided into 14 divisions viz., ( 1) Schizophyta (2) Myxothallophyta (3) Flagellate T4) Dinoflagellatae (5) Bacill ariophyta (6) Conjugate (7) Heterocontae (8) Chlorophyceae (9) Charophyta (10) P haeophyceae (11) Rhodophyta (12) Eumycetes (13) Archegoniatae (14) Embryophyta o r seed plants The division Embryophyta is sub-divided into two subdivisions - the Gymnospermae and Angiospermae. The subdivision Angiospermae is further divided as follows : History of Plant Classification 21 Sub-division 2. Angiospermae. Ovules enclosed in the ovary and tracheae or true vessels are present. Class 1.1VIonocotyledoneae.nyledon one; leaves with paralle l venation; flowers tiimerous; stem with closed vascular bundles. It is divided into II orders. Class 2. Dicotiledoneae. Cotyledons two; leaves with reticulate venation; flower tetra- or pentamerous; stem with open vascular bundles. It is divided into subclasses : Sub-class I. Archichlamydeae. Petals always free. It includes 33 orders. Sub-class II. Metachlamydeae. (Sympetalae). Petals always fused. It includes 11 orders. Critical assessment of Engler's System Engler's system is based upon his theoretical concept of a primitive flower. It supposed that the angiosperm flower was derived from a gymnosperm strobilus. The ancestral strobilus was unisexual either megasporangiate or microsporangiate al ong with their sporophylls. It produced the catkin type of inflorescence, in ang iosperms which are either staminate or pistillate. The bisexual flower was deriv ed from this unisexual type. Hence Engler placed the Amentiferae in the beginnin g of Archichlamydeae. Here the flowers were either naked or having a monochlamyd ous bract like perianth. The dichlamydous and sympetalous (gamopetalous) familie s were considered more evolved. Engler considered that the gymnosperms gave rise to the Amentiferae on the one hand and the monocots on the other. Angiosperms w ere therefore polyphyletic in origin. In fact Engler never believed in a linear

system of evolution and therefore never tried to draw an evolutionary tree. The primitive position of Amentiferae has been criticized by a number of botanis ts who have laid evidences that the anatomy and other features of this group can be best interpreted as a consequence of reduction rather than as a primitive co ndition. Amentiferae would thus be a highly specialized group derived from dichl amydous and bisexual flowers. But as recently as 1957 Helsop-Harrison has mentio ned that all unisexual angiosperms need not have arisen from bisexual types and it could even be a primitive condition. Thus this point remains unsolved. Richard von Wettstein (1862-1931), was an Austrian systematist. His classificati on was similar to that of Engler's in most respects. He considered the dicotyledons to be primitive and the monocotyledons to be deri ved from them through Ranales. In general his system of classification was a muc h better phylogenetic classification than that of the Engler's. 2.5.6. Phylogenetic System of Bessey Charles Edwin Bessey (1845-1915), was the first American to make a major contrib ution to the knowledge of plant relationship and classification. His system now known as Besseyan system, was evolved in "The Phylogenetic Taxonomy of Flowering Plants". The system was based upon a series of 'dicta' or statements of guiding principles used to determine the degree of primitiveness or advancement, of the plant groups. Bessey took Ranales as the basal group from which both monocotyle dons and other groups of dicotyledons have evolved, but also believed that all f lowering plants originated from strobiliferous, cycad ancestors. Hans Hallier (1868-1938), was a German botanist. His system closely resembled th at of Bessey. He made much use of ovule structure and position than other worker s. Alfred Barton Rendle (1865-1938), Director of British Museum London. In his two volumes "Classification of Flowering Plants" he classified plants chiefly accord ing to the Engler and Prantl'g system of classification. 2.5.7. Phylogenetic System of Hutchinson John Hutchinson of Royal Botanic Gardens, Kew has proposed a system which first appeared in Kew bulletins and later in his "Families of Flowering Plants (1926, 34) and ( 1959) Vol. I Dicotyledons and Vol. II Monocotyledons". His system is somewhat li ke that of Bessey. Hutchinson's revised classification has been based on 24 prin ciples which are similar to Bessey's dicta and are as follows : 22 Plant Taxonomy 1. Evolution is both upwards and downwards. 2. Evolution does not necessarily involve all organs of plant at a time. 3. Evolution has generally been consistent. 4. The trees and shrubs are more primitive than herbs in certain groups. 5. In one family or genus, the trees and shrubs are older than climbers. 6. Perennials are older than biennials and from biennials the annuals are derive d. 7. Aquatic flowering plants are derived from terrestrial ones. 8. Collateral vascular bundles arranged in a ring is a primitive condition e.g. dicots. 9. Spiral arrangement of leaves is a primitive character. 10. Simple leaves are primitive than compound leaves. , 11. Bisexual flowers are primitive than unisexual. 12. Solitary flowers are more primitive than being arranged in an inflorescence. 13. Spiral arrangement of floral parts is a primitive condition. 14. Polymerous condition is primitive than definite numbers. 15. Apetalous flowers are derived from petaliferous flowers. 16. Polypetaly is primitive than gamopetaly. 17. Actinomorphic condition is primitive than zygomorphic. 18. Hypogyny is primitive than epigyny and perigyny. 19. Apocarpous condition is primitive than syncarpous. 20. Many carpelled condition is primitive than few or one carpel. 21. Endospermic seeds are primitive. 22. Indefinite number of stamens are in primitive flowers.

23. Separate stamens precede connate stamens. 24. Simple fruits are more primitive than aggregate fruits. Critical assessment of Hutchinson's System Hutchinson imagined the angiosperms to have originated from a Bennettitaliean st ock like Cycadoidea along two lines. On one side the woody arborescent types aro se and side by side the non-woody herbaceous types also arose. They were termed Lignosae and Herbaceae respectively. The monochlamydeae was considered polyphyle tic from these groups and hence this group was altogether omitted. The Magnolial es represented the most primitive of Ligonsae and Ranales the most primitive of Herbaceae. This system separated families which were otherwise closely similar because they were either woody or herbaceous. Thus herbaceous Umbelliferae and woody Araliac eae or Cornaceae were separated even though their floral characters were very si milar. Similarly the Labiatae, from the woody Verbenaceae or the Bignoniaceae fr om the Scrophulariaceae and Genseriaceae. The main weakness of the system seems to be that all the emphasis has been place d only on the herbaceous and woody habit and neglecting other equally important floral characters. But it must be said to the credit of Hutchinson that his desc ription of families is very correct and that he provided a key which is extremel y useful in identification of a family. Hutchinson's system is not useful for he rbarium keeping and therefore it has not been adopted in any herbaria. Karl Christian Mez (1866-1944), a German professor of botany, evolved the theory that the relationships between the larger groups of genetically related plants could be determined by study and analysis of their protein reactions. This immun ological approach is known as serum diagonosis. 2.5.8. Oswald Tippo Oswald Tippo (born in 1911), formerly of the University of Illinois and now of t he New York University of U.S.A. published an outline of a projected classificat ion in 1942. He claims no History of Plant Classification originality for his system. Tippo's system is a compilation of systems proposed by G. Sirk,. for non-vascular plants and by A.J. Eames (1936) for vascular plant s. Linear Sequence of Order in Hutchinson's Revised System of Classification (1959/ 60) (with phylogenetic affinities and origin) HYPOTHETICAL PROANGIOSPERMS DICOTYLEDONAE 1 1 MONOCOTYLEDONAE HERBACEAE .l. Ranales Butomales Berberidales Alismatales Aristolochiales Triuridales 4 Piperales Juncaginales Rhoeadales Aponogitopales 4 Cruciales Potamogetonales 4 Resedales Najadales 4 Caryophyllales 4---- Commelinales Polygonales Xyridales Chenopodiales Eriocaulales Lythrales Bromeliales Gentianales Zinziberales Primulales -- Liliales Plantaginales Alstroemeriales Saxifragales4--Dioscoreales Sarraceniales Arales Podostemales . Typhales Umbellales 4 4. Amaryllidales Valerianales Iridales Campanales. --. Agavales

Goodeniales . Palmales Asterales Pandanales Solanales 4 . Cyclanthales Personales 4 Geraniales 4 Haemodorales. Burmanniales Polemoniales. Orchidales Boraginales. Juncales Lamiales 4 Cyperales Grarninales 4 DICOTYLEDONAE LIGNOSAE -- Magnoliales Annonales Laurales Dilleniales Coriariales Rosales Leguminales Cunoniales Styracales Araliales Hamamelidales4 Salicales Lietneriales Myricales Balanopsidales Fagales Juglandales Casuarinales Urticales Bixales Thymelae ales 1 Proteales Pittosporales Capparidales Tamaricales Violales Polygalales Lo as ales Passiflorales Cucurbitales Cactales Tiliales 4 Malvales Malpighiales Euphorbiales Theales Ochnales Ericales Guttiferales Myrtales Celastrales Olacales Santalales Rhamnales Myrsinales Ebenales Rutales Meliales Sapindales - Loganiales Apocynales Rubiales Bignoniales Verbenales Plant Taxonomy Synopsis of Tippo's System (1942) Kingdom Plantae Sub kingdom I. Thallophyta. Plant forming no embryos. Phylum 1. Cyanophyta. Phylum 2. Euglenophyta. vr, Phylum 3. Chlorophyta. Phylum 4. Chrysophyta. Phylum 5. Phaeophyta. Phylum 6. Rhodophyta. Phylum 7. Schizomycophyta. Phylum 8. Myxomycophyta. Phylum 9. Eumycophyta. Sub kingdom II. Embryophyta. Plants forming embryos. Phylum I. Bryophyta or Atracheata. Plants without vascular tissues. Phylum 2. Tr

acheophyta or Tracheata. Plants with vascular tissues. Sub phylum 1. Psilopsida. Root-less and leafless vascular plants. Sub phylum 2. Lycopsida. Scale like leaves and simple vascular tissues in plants. Sub phylum 3. Sphenopsida. Jointed, ribbed stem; scale like leaves and simple vasc ular tissues in plants. Sub phylum 4. Pteropsida. Large leaves and complex vascular tissues in plants. Class 1. Filicineae. Class 2. Gymnospermae. Class 3. Angiospermae. Sub class 1. Monocotyledonae. Sub class 2. Dicotyledonae. Alfred Gunderson in 1950 introduced a phylogenetic system of classification whic h is based upon chromosome number, anatomical and morphological characters. He h as divided dicots into 10 groups and 42 orders in his "Families of Dicotyledons" . He has not described the monocotyledons. Knowledge of chromosomes has contributed much to the understanding of varieties, species and genera but has contributed little to the understanding of classific ation of families or orders. 2.5.9. Lyman Benson Lyman Benson, professor of Botany, California, proposed a new system of classifi cation in 1957 in his book "Plant Classification". He has divided the vascular p lants into five divisions. Benson has mainly followed Bentham and Hooker's syste m. Synopsis of Benson's System (1957) Division Class Spermatophyta Angiospermae Conopsida (cone) Ephedropsida (Ephedra) Gnetopsida (Gnetum) Cyadopsida (Cycas) Pteridophyta Pteropsida Pteridophytes Sphenophyta Sphenopsida Lycophyta Lycopsida Psilophyta Psilospida 24 3 PLANT NOMENCLATURE Plant nomenclature or the system of naming plants is an important part of taxono my. It deals with the determination of correct name, to be applied to a known pl ant or taxon. It is well known that many plants have several common names in gen eral use in various parts of the country; and often the same common name is used for different plants and furthermore some plants have no common names. This is a confusing state. So it becomes essential to give to a particular plant a defin ite name which may be used all over the world. In general nomenclature may be defined as the system of naming plants, animals, or other objects or groups of animals, plants or objects. Before the middle of e ighteenth century, the names of plants were commonly polynomials i.e. composed o f several words in a series. Ultimately this system was superseded by the binomi al nomenclature or binomial system, which was used by Rivinus and later establis hed by Linnaeus in "Species Plantarum" (1753). In botanical nomenclature the nam es given to the plants are either Latin names or names are taken from some other language and Latinized. 3.1. Binomial System or Bin )mial nomenclature The botanical name of any plant is in two parts; the first part is the name of g enus or generic name and second is the specific epithet or attribute. For exampl e the botanical name of the pea is Pisum satiuum and the botanical name of wheat is Triticum satiuum. Botanical names should always be given in italics when pri nted or underlined when typed or written by hand. According to the International code, there can be only one group of plants with the name Pisum, reserved for p

ea group. In each genus there can be only one valid specific epithet satiuum, bu t the same specific epithet may be applied to plants that are not member of the same genus as in above example the specific epithet satiuum is applied to both t he plants. 3.1.1. Advantages of Binomial vomenclature : 1. These names are more definite and precise than ordinary names. 2. Being generally in Latin, they have a universal acceptance by people of all l anguages. 3. They are more comprehensible and easier to study. 4. They indicate the genetic relationship and descent of individual plant. 5. They are more independent. 6. They are usually descriptive. 3.2. The Generic Name It is always a noun, in the singular or nominative case and is always written wi th a capital initial letter. The generic name may be descriptive, aboriginal nam e of the plant or a name in honour of a person such as Caesalpinia for Caesalpin o, Linnaea for Linnaeus, Magnolia for Pierre Magnol, a French contemporary of Jo hn Ray. (33) 34 Plant Taxonomy These names have come from many sources viz., : 3.2.1. Honour of Person Many genera have been named in honour of some person, usually a botanist, or a p atron of botany or horticulture. A Dioscorea for the author of the great Materia Medica; Bauhinia, honours the two Bauhin brothers. Bignonia the Abbe Bignon. Ju ssieua, for Bernard de Jussieu; a Nicotiana for Jean Nicot, who introduced tobac co into Europe; Hookera for Hooker, Candollea for A. de Candolle. 3.2.2. Special Feature of Plant Many generic names are formed by combination of two or more Greek or Latin words , usually expressive of some feature of the plant. Examples are Leucadendron, "s ilver tree", from the silvery leaves of the silver tree; Xanthoxylum, "Yellow wood ", from the colour of the roots of the shrub yellow root; Oxydendrum, "sour tree", from the acidic leaves of the sourwood. 3.2.3. Land of its Discovery In many cases, when common names for plants existed in the lands in which they w ere discovered, these names from the native languages and were converted into La tin generic names. Among these may be mentioned Tsuga, from the Japanese; Nelumb o, from the Ceylonese; Revenala, from the Madagascarian; Pandanus, from the Mala yan. 3.2.4. Fancy or Mythology Another type of genetic names includes those of fanciful, mythological, or poeti c origin. For example Nymphaea, water lilies, refers to the beautiful water nymp hs; Circaea, the enchanter's nightshade, refers to Circe, the enchantrees; Theob roma (god's food) was applied to the chocolate plant. 3.3. The Specific Epithet The species may be regarded as morphological definable units, made up of groups of individuals (populations), which it is assumed are usually interbreeding, the containers and expression of one or more gene pools. These names indicate the k ind of plant and enables the members of a genus to distinguish from each other. They may be either an adjective indicating a distinguishing characteristic of th e species e.g. alba or referring to a locality e.g. indica from India, himalensi s from Himalaya or in honour of a person e.g. santapaui in honour of Rev. Fr.H. Santapau or indicating something about the habitat of the species. The specific epithet is always written with a small initial letter. It is second part of the scientific name, called by Linnaeus the "trivial name". It may be derived from many sources. 3.3.1. Descriptive Adjective It might be expected that this would continue to be the commonest form of the sp ecific epithet, as indeed it is. These may be indicative of (a) colour, as rubra (red), alba (white), (lava (Yellow), nigra (black), fusca (brown); (b) may indi

cate the size, shape or habit of the plant, as nana (dwarf), gigantea (giant), a lta (tall), crassa (thick), tenuis (thin), scandena (climbing), natans (swimming ), repens (creeping), pendulus (hanging); (c) may indicate the plant's relative abundance, as vulgaris (common), rara (rare); (d) may designate the plant's habi tat, as arvensis (in fields), palustris (in swamps), arenicola (in sand), saxati lis (among rocks), aquatica (in water), sylvatica (in woods), muralis (on walls) , (e) may describe the plant's uses, as sativus (sown for crops), esculentus or edulis (edible), (f) may indicate other characteristics, as foetida (ill scented ), Plant Nomenclature 35 fragrans (fragrant), spinosa (spiny), tomentosa (woolly), vernalis (of spring), aestivalis (of summer), autumnalis (of autumn), hiberna (or winter), acuminata ( long pointed), biennis (biennial), restrata (beaked), religiosa (religious), tox icaria (poisonous) or (g) may indicate the region where the plant was found, as canadensis (Canadian), virginica (Virginian), noveboracensis (of New York), chin ensis (Chinese), gallica (French), Japenica (Japanese), canarensis (of the Canar y Islands), capensis, nepalensis (of Nepal), australiana (of Australia), indica (of India). 3.3.2. Second Type Second type of specific epithet is a descriptive adjective formed by combining t wo or more Greek or Latin words and thereby making reference to some distinctive feature or characteristic of the plant, as latifolia (broad-leaved), angustifol ia (narrow-leaved), grandiflora (large-flowered), cordifloia (with heart-shaped leaves), quadrilocularig (four-celled). 3.3.3. Third Type A third type is formed from a noun, with a suffix indicating some resemblance or relationship, as bignonioides (resembling Bignonia), amaranthoides (resembling A maranthus), Zinniaeflora (with flowers like Zinnia). 3.3.4. Fourth Type A fourth type is in honour of some famous botanist or other person linked with t hat species. These may be written in two ways : as a noun in the possessive case , for example agharkari (Musa agharkarii) roxburghii (Vanda roxburghii), or as p ossessive adjective, as gesneriana (Tulipa gesneriana), Gesner tulip, Grayana (C rataegus grayana, Gray hawthorn), Muelleriana (Eucalyptus muelleriana, Muller st ringybark). 3.3.5. Fifth Type A fifth type is a noun itself. often an .old Latin or Greek word for some plant. Examples in Pyrus malus (Malus was the old Roman word for the apple), Prunus ce rasus (Cerasus was the Roman word for the sour cherry), Capsella bursa-pastoris (Bursa-pastoris was the classical word for the shepherd's purse. Other words of this type include names for plants taken from some vernacular or aboriginal lang uage, as Ipomoea batatas (Batatas was used for the sweet potato by the Tainos, a borigines of the West Indies), Theobroma cacao (Cacao, the chocolate, is from an Aztec name). Specific epithets, when adjectives, must agree grammatically with the generic na me, which is always in the singular number and in the nominative case. The gener ic names do, however, differ in gender, being either masculine, feminine, or neu ter. 3.4. The Family A family consists of a group of related genera and rarely of a single genus. The aceae to the stem of an legitimate na name of family is formed by adding suffix me of an included generic name, e.g. Ranunculaceae from Ranunculus, Papaveraceae from Papaver and Rosaceae from Rosa. A few families are still designated by old er names and such names are often merely the ancient common names and are not ba sed upon generic names. Some of these names are Cruciferae, Leguminosae, Umbelli ferae, Compositae and Gramineae, but according to International Code of Nomencla ture the new names of these are Brassicaceae, Fabaceae, Apiaceae, Asteraceae and Poaceae respectively. The sub family : This is a major sub-division of a family and is sometimes used wh en the size of the family justifies it. The name of the sub-family is formed by

adding oideae to the stem 36 Plant Taxonomy of an included generic name such as Prunoideae from Prunus and Rosoideae from Ro sa for the two sub families of Rosaceae. 3.5. The Tribe It is a sub-division of family and sub-ordinate to the sub-family. The name of t he tribe is formed by adding eae to the stem of an included generic name, such a s Roseae from Rosa of the family Rosaceae. 3.6. The Order This is a major taxon superior to the family. The name of the order or ordinal n ame is formed by adding ales to the stem of an included generic name e.g. Ranale s from Ranunculus, Rosales from Rosa, Malvales from Malva. 3.7. The Authority The name of person or persons written after the botanical name is called the aut hority of the name. Authority is given when some person or persons have original ly published the account of, and described each taxon whether family, genus or s pecies and gave it a name. For example Mimosa pudica first named and described b y Linnaeus, becomes the authority for that name and it is written as Mimosa pudi ca L. Similarly Eruca sativa Lamk shows that Lamarck, first named the species. 3.7.1. The Parenthetical Authority When the rank of a plant or plant group is changed or a species is transferred f rom one genus to another or similar changes are made then the name of the origin al author is placed in parenthesis and is followed by the name of the person mak ing the change. For example Medicago polymorpha L. variety orbicularis first nam ed by Linnaeus is raised to the rank of a species by Allioni and thus it becomes Medicago orbicularis (L) All. 3.7.1.1. Principle of Priority The first validly published name of a species or other taxon becomes its valid n ame and if other names are published subsequently for the same taxon then they b ecome synonyms. This means, in general, that names published earlier will take p recedence over names of same rank published later. To avoid this botanists have agreed that for the flowering plants the starting point shall be from Linnaeus " Species Plantarum" (1753). 3.8. Codes of Nomenclature or History of ICBN The rules of nomenclature had their beginning in Linnaeus' "Critica Botanica" (1 737) which was amplified later (1751) in "Philosophia Botanica" and established the real beginnings of a sound nomenclature for plants. The first important post Linnaean index was Steudel's Nomenclature Botanicus (1821). It was the forerunn er of Index Kewensis. 3.8.1. Paris code (1867) The first organised endeavour in the formulation of rules for nomenclature were at 1st International Botanical Congress at Paris. De Candolle prepared the laws of botanical nomenclature and after a revision, these laws were accepted as the best guide for nomenclature of plants. By these rules the starting point for all plant nomenclature was with Linnaeus. 3.8.2. Rochester code (1882) A set of rules of nomenclature prepared by a group of botanists in U.S.A. The ba sic difference between Paris and Rochester codes were (1) concept of types as a scientific and fundamental basis Plant Nomenclature 37 by which to determine the precise application of names, that is, each species to be based on a herbarium specimen to be known as type of the binomial and larger groups of classification to be based on a species, genus or family. (2) interpr etation of the rule of priority to apply not only to the date of publication but also to the precedence of a name within a publication, that is, if two names on a given page were later treated as belonging to the same plant the first had pr ecedence over the second name. 3.8.3. Vienna code, (1905) This third Botanical congress has accepted Linnaeus' Species Plantarum (1753) th e starting point of the nomenclature of vascular plants and descriptions of gene

ra by Linnaeus in Genera Plantarum (1754). 3.8.4. American Code, 1907 The significant action of this code was that a binomial may not be used again fo r a plant in any way, if it has been employed previously for another plant, even though the previous use may be illegitimate. 3.8.5. Brussel's Code, 1910 The Fourth Internal Congress convened in 1910 at Brussel. The important action o f this code was the establishment of different starting points for priority of n ames of non-vascular plants and recognition of the value of the 'type concept'. 3.8.6. International Rules of Botanical Nomenclature 1930 At Cambridge the Fifth International Congress was held where every effort was ma de to harmonise the basic difference between the Vienna and American codes. For the first time in botanical history, a code of nomenclature came into being that was international in function as well as in name. 3.8.7. Amsterdam Congress, 1935 At the 6th International Botanical Congress one major change provided for the ad vance from January 1, 1932 to January 1, 1935 as the date after which all diagno ses of plants new to science (excepting bacteria) must be in Latin. The rules of nomenclature are divided into chapters and sections and these are divided into articles and recommendations. The third edition of this rules contained an enume ration of the nomina generica conservanda as of 1935. In the supplement to that edition was given an enumeration of the nomina generica conservanda proposita an d additional names were proposed for conservation subsequently. 3.8.8. Stockholm Congress The Seventh International Botanical Congress met at Stockholm in 1950. It introd uced a certain number of definitions on types. 1. The term 'taxonomic group' will be changed to 'taxon' (plural taxa). 2. Terms 'legitimate, illegitimate, effective and valid' are defined under a new article. 3. Instead of article 18, dealing with the subject of selecting and designating types, five new articles and two recommendations are accepted. 4. Publication dates of works serving as starting points will be fixed either on January 1 or December 31. 5. The date of effective publication is the moment when the publication becomes available. Names published on labels of exsiccatae are treated as not validly pu blished. 6. Names of species or of taxa below the rank of species are regarded as not val idity published if the generic name which forms part of the binomial has not bee n validly publoished before or at the same time. 38 Plant Taxonomy 7. It is recommended to avoid the use of names previously published as `nomina n uda'. 8. In case of publication of new taxa by one author in the work of the another a uthor, the two authors should be connected by word 'in'. 9. If the rank of a taxon above that of a genus is changed, the root of the name must be preserved, the ending only has to be changed. 10. The congress rejected the principle of Nomina Specifica Conservanda by a bal lot of 320 votes to 40. 11. Alternative names shall be considered as illegitimate. 12. It is recommended that the publication of new taxon be accompanied by a figu re. 13. A name of which the type has been excluded, can only be maintained by 'conse rvation'. In this case the author of the name is the botanist who gives the new description. 14. Misidentification must not be cited in the synonymy, but at the end of the l atter. After the name `auct non' is added followed by the original name of the a uthor. 15. If the status of the taxon bearing a binnary name is changed from that of a species to that of an hybrid or inversely the name of the original author will b e maintained with the addition between brackets of 'pro sp' or 'pro hybr'.

16. In article 62 the sentence dealing with a list of nomina ambigua is deleted. 17. It is recommended to write the epithets of species of taxa of lower rank wit h a small initial. 18. Under the new rules of typification, the application of names of taxa is det ermined by means of nomenclatural types. A nomenclatural type is that constituen t element of a taxon to which the name of the taxon is permanently attached, whe ther as an accepted name or synonym. 3.8.9. Paris Congress, 1956 In 8th Congress principles and recommendations of the rules were treated separat ely. The chief decisions of this Congress were the following : 1. The adoption of a list of Nomina Specifica Rejicienda which has the same stab ilising effect as introducing a Nomina Specifica Conservanda. The-Code was divid ed into four main division : 1. General considerations. 2. Rules and Recommendations. 3. Provisions for the modification of the Code. 4. Appendices. 2. To the guiding principles as stated in Chapter 1 of the Stockholm code two mo re principles have been added, one stating that the application of names of taxo nomic group is determined by means of nomenclatural type' and the other stressin g the correct use of publications'. 3. Rules were laid for ,ffective publications, names of subdivisions of genera, transference of names between plant and animal kingdoms. 4. Proposals to delete article 35 were rejected. 3.8.10. 9th Congress or Montreal Congress (1961) The rules regarding nomenclature of fossils were prepared and list of Nomina Fam iliarum Conservanda was accepted. 3.8.11. In 10th Congress at Edinburgh (1964) Some changes in nomina familiera conservanda accepted in the previous congress w ere done. List of recently discovered gymnospermous families was prepared and a list of the meanings of difficult words relating to nomenclature was prepared. Plant Nomenclature 39 3.8.12. In the 11th Congress at Seaule (1972) Some changes in the name of genera and families accepted in the previous congres s were made. A committee was formed to deal with the problems relating to the po sition of taxa above the rank of family. This congress produced a Code published in 1972 which envisaged : 1. A list of nomina conservanda. 2. Priority of publication was given due importance. 3. Each group with particular circumscription and rank must bear only one name. 4. All scientific names are treated in Latin, irrespective of their derivation f rom any source. 5. The rank of taxa in ascending order are species, genus, family, order and cla sses. When greater number of taxa are to be incorporated additional taxa are con structed by adding the perifix sub to terms denoting the ranks of the individual . 6. Legitimate name of the epithet is the one which is suggested according to the rules of nomenclature. 7. Names suggested in contravention of the rules are termed as illegitimate. 8. Separate rules for giving names to the hybrids have been formulated. 3.8.13. In the 12th Congress at Leningrad (1975) Proposals regarding some changes in the type concept were made. The minor change s made in Code : 1. Individual paragraphs of all articles and Recommendations were numbered in a decimal-like system; some paragraphs were rearranged. 2. The concept of organ-genera was eliminated for fossil plants. 3. The principle of automatic typification was extended above the family level t hat were based on generic names, however, on priority basis. 4. It was made clear that Code does not apply to the names of the organism treat ed as Bacteria.

5. The Article 69 was modified as "a name be rejected if it has been widely and persistently used for a taxon not including its type". 6. The Articles 70 and 71 of the Seaule Code dealing with discordant elements an d monstrosities were deleted but the article numbers were retained as such. 7. A thorough rewritting of the section on Orthography was presented. 3.8.14. The 13th Congress at Sydney 1981 Articles 70, 71 and 74 were deleted. Article 59 and related provision were thoro ughly remodelled and a change in the starting point dates for fungi was adopted, but conferring a privileged nomenclature status on certain post Linnaean names. The Sydney Code is quite similar to the Leningrad Code except for a few minor ch anges which are enumerated below. 1. Numbering scheme introduced in this Code makes the reference to these more si mple. 2. More than one hundred proposals were accepted by Sydney Congress merely refin ing Leningrad Code. 3. Conservation of names of species was accepted but was not included in the Cod e. 4. The article dealing with spelling as used by Linnaeus (Article 74) was delete d at Sydney. 5. Articles 60 and 61 were remodelled. 6. Article 59 and related provisions were thoroughly remodelled and a change in starting point dates in fungi was adopted. 7. Some of the dates for the application of various rules were eliminated at Syd ney Congress. 40 Plant Taxonomy 8. The provisions for hybrids were thoroughly overhauled. 9. Autonyms were treated as validly published having priority over the names of the same date automatically established. 10. Appendix III Nomina Generica Conservanda and rejicienda was considerably rev ised. 3.8.15. The 14th Congress at Berlin 1988 The Berlin Code adopted by the 14th International Botanical Congress in Berlin 1 987, supersedes all the earlier Codes of the Botanical Nomenclature. The most important difference between Berlin Code and the previous editions of t he International Code of Botanical Nomenclature is that the text is in English o English, French and German which were o nly instead of three different languages perative since Paris code of 1956. In other respects, the Berlin Code is similar to the Sydney code. However, one new article and some new recommendations are a dded in this Code and some new numbers have been provided to certain paragraphs. Two articles dealing with illegitimate names (Article 66 and 67) were deleted by the Berlin Congress, 1987 and Recommendation 75A dealing with gender of generic names in Sydney Code appear as new article (Article 76). Articles 7-10 dealing with typification have been revised thoroughly and the por tion 'Guide for the determination of types' been given a numbered place as recom mendation 7B. Five additions regarding rules on types and typification have been incorporated. It is also mentioned that from January 1,1990, the place of prese rvation of the type of the name of taxon must be mentioned. The other changes relating to illegitimate, orthography and naming of hybrids ar e given in the present Code. Outline of Berlin Code, 1988 Preamble Division I. Principles (1-6) Division II. Rules and Recommendations (Articles 1-76) Chapter I. Ranks of taxa and the terms denoting them (Article 1-5) Chapter II. Names of taxa (Articles 6-15) Chapter III. Nomenclature of taxa acording to their rank (Articles 16-28) Chapter IV. Effective and valid publications (Articles 29-50). Chapter V. Retention, choice and rejection of names and epithets (Articles 51-72 ). Chapter VI. Orthography of names and epithets and gender of generic names (Artic

les 74-76). Division III. Provision of modification of the Code. Appendix I. Names of hybrids. Appendix HA. Nomina familiarum algarum, fungorum et pteridophytorum. Appendix IIB. Nomina familiarum bryophytorum et spermatophytorum conservanda. Appendix MA. Nomina generica conservanda et rejicienda. Appendix MB. Nomina specifica conservanda et rejicienda. Appendix IV. Nomina utique rejicienda. It is beyond the scope of this book to enter into full discussion of the Interna tional Code of Botanical Nomenclature. 3.9. International Code of Botanical Nomenclature (ICBN) The present code is divided into three parts or chapters viz., 1. Principles, 2. Rules and Recommendations and 3. Provision for the modification of code and app endices. Plant Nomenclature 41 3.9.1. Principles of ICBN 1. Botanical nomenclature is independent of zoological nomenclature. 2. The application of names of taxonomic group is determined by means of nomencl atural types. 3. The nomenclature of a taxonomic group is based upon priority of publication. 4. Each taxonomic group with a particular circumscription, position and rank can bear only one correct name the earliest that is in accordance with the Rules, e xcept in specified cases. 5. Scientific name of plants are treated as Latin regardless of their origin. 6. The rules of nomenclature are retroactive when expressly limited. Rules and Recommendations of ICBN : 3.9.2. Rules of Botanical Nomenclature : It is the product of intensive study by the best brains in the botanical science . The foundations of the code are to be found in Linnaeus "Philosophia Botanica" . In this he lays down the main points of nomenclature in the form of principles o r aphorisms. For example a few of these aphorisms or rules are as follows : "213" All those plants which belong to one genus must be designated by the same generic name. "214" All those plants which belong to different genera must be designated by di fferent generic names. "218" He who establishes a new genus should give it a name. "240" Those generic names are best which show the plants essential character or its appearance. "249" Generic names one and a half foot long or difficult to pronounce or unplea sant, are to be avoided. "257" The specific name must distinguish a plant from all its relatives. "260" Size does not distinguish species. "264" The original place of plant does not give specific difference. "284" A generic name must be applied to each plant species. "285" The specific name should always follow the generic name. The Linnaeus principles are quite simple. In course of time, it has become neces sary to specify the rules so as to make these precise. Ranks of taxa Taxon is a taxonomic group of plants or taxonomic unit of any rank. Taxa is plur al of taxon. For example the species Brassica campestris Linn. is a taxon. The f amily Brassicaceae is a taxon. These are coined as follows : sub family, subgenus, subs 1. By adding the prefix 'sub' to the terms, For example pecies, etc. _

2. By introducing supplementary terms, between two principal ranks of taxa; For example(i) between family and genus tribe; (ii) between genus and species section, series. (iii) Below species variety and form. 3.9.3. Names : In taxonomy all the names are treated as Latin names. The names are treated acco rding to the principle V of International Code of Botanical nomenclature (ICBN). The names are of following kind : 42 Plant Taxonomy 1. , Alternative names : Each taxon of the rank of family and below can bear onl y one correct name, but the following families have alternative names, which are perMifeed by ICBN. (Art. 18, Note 3). Names under usage Names according to rules .,Cruciferae Brassicaceae Compositae Asteraceae Gramineae Poaceae Labiatae Lamiaceae Leguminosae Fab ace ae Palmae Arecaceae Umbelliferae Apiaceae 2. Ambiguous name : The name is used in different senses, so become a long persi stent error. Ambiguous name should be rejected (Art. 69). 3. Autonym : It is an automatically established name. It is established below fa mily and genus level (Art. 26). 4. Confused name : The name based on a type consisting of discordant elements. 5. Conserved name : In view to avoid unadvantageous, changes, certain names of t he rank of family and below are conserved. The lists of these names are given in ICBN. (Art. 14) (see illegitimate name). 6. Correct name : The name which is adopted under the rules for a taxon with par ticular circumscription, position and rank (Principle IV). 7. Homonym : The names spelt identically bid based on different types (Art. 64). 8. Illegitimate name : It is validly published name contrary to one or more arti cles of ICBN (see conserved name). 9. Naked name : The name which is published without description or diagnosis or reference. These names are not validly published. 10. Synonym : When one of two or more names considered to apply to the same taxo n. 11. Tautonym : It is a name in which the specific epithet repeats exactly the ge neric name. These are not admissible in ICBN (Art. 23). 3.9.4.. Why name changes ? The name changes mainly due to the following reasons : 1. Taxonomic : When taxonomist gains more knowledge about the taxon, on which he is doing research he may either alter the rank or may change the position from one species to another or one genus to another genus or may merge two taxa into single taxon or may split one taxon into two or more taxa e.g. Acacia nilotica sp indica (Benth). Brenan, Kew Bull. 12 : 84. 1957 basionym* is Acacia arabica var. indica Benth. London J. Bot. 1 : 500. 1842. In this case there is double citation because the variety has been changed to th e rank of subspecies and also the position has been changed from the species A. arabica (Lam.) Willd. to Acacia nilotica (L) Willd. ex Delile. 2. Nomenclatural : The commonly used name may have to be changed because it is n ot in accordance with the rules of the International Code of Botanical Nomenclat ure. Nomenclatural example : Vinca, Catharanthus and Lochnera. In 1828 Reictrenbach split the genus Vinca L. into Vinca proper and Lochnera Rei ch. But Lochnera Reich was invalid. In 1837 G. Don renamed the genus thus separa ted from Vinca as Catharanthus G. Don gave its description. In 1838 Endlicher va lidated the name Lochnera by *Basionym A name or epithet required by a change in position or rank of a taxon and taken a previous name for the taxon is the basionym of the new name.

Plant Nomenclature 43 giving description, but it was one year later to Catharanthus. Therefore the cor rect name of the genus seggregated from Vinca is Cathananthus G. Don. The distinguishing characters given by H. M. Lawrence (1959) and which were repr oduced by Santapau (1961. Bull. Bot. Surv. India 3 : 15) are : Vinca Catharanthus 1. Stem woody, persisting for many years. 1. Stem annual or perennial herbaecous. 2. Leaves persistent. 2. Leaves deciduous in perennial plants. 3. Flowers solitary in leaf axils. 3. Flowers 2-3 in cyme axillary clu sters. 4. Corolla infundibuliform, the mouth closed by scales. 4. Corolla salverform, the mouth closed by bristle like hairs. 5. Filaments flattened much longer than anther sacs, connective spreading w ith clear coma like appendage. 5. Filaments round or half round, much shor ter than anther sacs, connective not spreading, without any appendage. 6. Nectaries much shorter than ovary. 6. Nectaries as long as or longer than ovary. 7. Stigma without hyaline "petticoat". 7. Stigma with basal hyalin e "petticoat". 8. Follicles mostly 6-8 seeded. 8. Follicles 15-30 seeded. The two species commonly found in India belong to Catharanthus. 3.9.5. Typification The application of names of taxa of the rank of family or below is determined by means of nomenclatural types A nomenclatural type (Typus) is than.constituent element of a taxon to which the name of the taxon is permanently attached, whether as a correct name or as a sy nonym (Art. 7). It is not necessarily the most typical specimen. The type is usu ally the gathering with which the name is first associated. Holotype : It is the one specimen or other element used by the author of the nam e, or designated by him, as the nomenclatural type i.e. the element to which the name of the taxon is permanently attached. 2. Lectotype : It is a specimen or other element selected from the original mate rial to serve as the nomenclatural type, when the holotype was not designated at the time of publication, or when the holotype is missing. 3., Neotype : It is a specimen selected to serve as the nomenclatural type of a toxon in a situation when all material on which the taxon was based is missing. 4. Paratylie : It is a specimen cited with the original description, other than the holotype. 5. Isotype : It is a specimen believed to be a duplicate of the holotype. 6. Syntype : It is one of the two or more specimens or elements used by an autho r when no holotype was designated, or in lieu of a holotype, or when one of two or more specimens were designated simultaneously as the type. 7. Schizotype : Any fragment of a type is known as Schizotype. 8. Merotype : It is a fragment of original holotype which has been divided into two or more pieces after having been used as a basis for description. 9. Clastotype : The portion of the same clone. 10. Paralectotype : The specimens whichwere left after the selection of lectotyp e out of syntypes. 11. Synonymotype : It is that constituent element of a taxon quoted by the autho r as being identical with the newly described taxon. Plant Taxonomy 3.9.6. Units of Classificaation The categories into which plants should be classified, constitute the units of c lassification. Their sequence and order of importance is fixed by the rules and changes are not allowed. The units of classification are arranged below in descending sequence from units of greatest magnitude

to units of least magnitude. Kingdom Division ( phyta) Sub division ( phytina) Class ( opsida) Sub class ( idae) Order (-ales) Sub order ( ineae) Family ( aceae) Sub family ( oideae) Tribe ( eae) Sub tribe ( inae) Genus Sub genus Series Species Sub species Varieties ( var) Subvarieties Forma ( f) Clona ( Cl). 44 In parenthesis is a suffix of unit e.g. aceae for family.

5 NUMERICAL TAXONOMY (50) 5.1. What is Numerical Taxonomy ? Numerical taxanomy is defined as a numerical evolution of similarity between gro ups of orOznisms and the ordering of these groups into higher-ranking taxa on th e basis of these similarities (Heywood, 1967). According to Sneath and Sokal (1973) the term numerical taxonomy means the group ing by numerical methods of taxonomic units into taxa* on the basis of their cha racter states. The numerical taxonomy includes the drawing of phylogenetic infer ences from the data by statistical or other mathematical methods to the extent t o which this is possible. These methods require the conversion of information ab out taxonomic entities into numerical quantities. Robert R. Sakai and Peter H. A. Sneath (1963) also described Numerical Taxonomy as the numerical evolution of affinity or similarity between taxonomic units and the ordering of these units into taxa on the basis their affinities. Jardine and Sibson (1971) suggest the name "mathematical taxonomy" and Blackith and Reyment (1971) employ "multivariate morphometrics". A synonym, "taxometrices ", has been suggested for numerical taxonomy (by Mayr, 1966, in lieu of "taximet rics", proposed by Rogers, 1963), and shall use the adjective "taxometric" occas ionally in place of the more cumbersome "numerical taxonomic". However, the term numerical taxonomy seems well established. 5.2. History of Numerical Taxonomy or Development of Numerical Methods in Taxono my The earliest attempts to apply numerical methods to taxonomy date from the rise of biometrics in last century as early as 1898. Early methods, specifically inte nded for taxonomy, are those of Forbes (1933), Anderson and Owenbey (1939), Stur tevant (1939, 1942), Boeke (1942), James (1953), Stallings and Turner (1957), Hu dson, Lanzillotti, and Edwards (1959), and Chillcot (1960), all based on variati ons of matching coefficients. Terentjev (1931, 1959) introduced a method of defi ning clusters of characters called "pleiades". These authors did not develop the ir methods sufficiently to meet the main needs of numerical taxonomy. The modern work in numerical taxonomy was initiated by the almost simultaneous p ublications of Sneath (1957a, b) at the National Institute for Medical Research, London, on bacteria, and of Michener and Sokal (1957) and Sokal and Michener (1

958) at the University of Kansas, on bees. There has been a very rapid increase in the development of methods for numerical taxonomy and in the application of these techniques after publication of book " Principles of Numerical Taxonomy". *The term taxon (plural taxa) as an abbreviation for taxonomic group of any natu re or rank, as suggested by H. J. Lam (in Lanjouw, 1950) and Rickett (1958). Numerical Taxonomy 51 5.2.1. Development of Numerical Taxonomy The subject is still in a period of active growth and change and lack the histor ical perspective to single out every significant landmark in its development, bu t the following chronological chart may give an idea of the development of the f ield. 1957-1961 Development of first methods and of theory of numerical taxonomy . 1962 First test of the nonspecificity hypothesis (Rohlf); first criterion of goodness of a classification (Sokal and Rohlf); first comprehensive publication of theory and methods (Sneath and Sokal). 1963 Publication of Principles of Numerical Taxonomy (Sokal and Sneath). 1964 First cladistic technique for continuous characters (Edwards and Cavalli S forza). 1965 First cladistic technique for discrete characters (Camin and Sokal); fir st critical analysis of the methodology of numerical taxonomy (Williams and Dale). 1966 First paper on taxonomy of scanned images (Sokal and Rohlf); electronic data processing linked to taxonomy (Sokal and Sneath); first generalization of hierai-chical clu stering methods (Lance and Williams). 1967 Numerical cladistics of proteins (Fitch and Margoliash); first attempts at definitions of operational homology (Jardine, Key, Sneath); congruence of phenetics with genome message (Heberlein, Del Ley, Tijtgat); first computer-oriented identification m ethods (Goodall, Morse, Dybowski, Franklin). 1968 Investigation of intra-OTU variation (Crovello). 1969 First generalized methods fgr numerical cladistics (Kluge and Farris); r e-examination of the logic of biological classifications (Jardine); attempts at statistical valid ation of clustering procedures (Mountford, Switzer). 1970 Re-examination of the biological species concept (Sokal and Crovello). 1971 Publication of Mathematical Taxonomy, stressing logically consistent cri teria for taxonomy and overlapping clusters (Jardine and Sibson). The application of high-spee electronic 'computers in taxonomy represents one of the major development in recent years and are being largely employed in the dev elopment of quantitative methods of classification in numerical taxonomy. Electr onic data processing (EDP) is also largely used in systematic botany concerned w ith the recording, storage and recovery of information on which a taxonomist dev otes most of his time. In this electronic data processing (EDP), the computer pe rforms all the arithmetic operations and calculations such as quantitative compa risons in respect to a large number of characters at a time at an extremely high speed. The time taken by the computer is of the order of a few seconds whereas a taxonomist himself would have taken a very long time for the same quantity of work. According to Heywood the numerical taxonomy is based on phenetic' evidence i.e., on the similarities shown by observed and recorded characters of taxa, and not on phylogenetic probabilities. Thus it is based on statements and hypotheses so formulated that they can be tested by observation and experiment. The outstandin g aims of numerical taxonomy are repeatability and objectivity. 5.3. Principles of Numerical Taxonomy or The Fundamental Position of Numerical T axonomy

Michel Adanson (1727-1806), a French botanist gave views regarding fundamental p osition of numerical taxonomy. The views are represented by certain principles. These principles are therefore frequently called neo Adansonian principles. Sneath a nd Sokal (1963) published a book "Principles of Numerical Taxonomy" which later was updated as Numerical Taxonomy (1973). In which they summerised the following principles : 52 Plant Taxonomy 1. The greater the content of information in the taxa of a classification and th e more characters on which it is based, the better a given classification will b e. 2. A priori, every character is of equal weight in creating natural taxa. 3. Overall similarity between any two entities is a function of their individual similarities in each of the many characters in which they are being compared. 4. Distinct taxa can be recognized because correlations of characters differ in the groups of organisms under study. 5. Phylogenetic inferences can be made from the taxonomic structures of a group and from character correlations, given certain assumptions about evolutionary pa thways and mechanisms. 6. Taxonomy is viewed and practiced as an empirical science. 7. Classifications are based on phenetic similarity. 5.4. The Advantages of Numerical Taxonomy The manifold advantages of numerical taxonomy are briefly following : 1. Numerical taxonomy has the power to integrate data from a variety of sources, such as morphology, physiology, chemistry, affinities between DNA strands, amin o acid sequences of proteins, and more. This is very difficult to do by conventi onal taxonomy. 2. Through the automation of large portions of the taxonomic process, greater ef ficiency is promoted (Sokal and Sneath, 1966). Thus, much taxonomic work can be done by less highly skilled workers or automata. 3. The data coded in numerical form can be integrated with existing electronic da ta processing systems in taxonomic institutions and used for the creation of des criptions, keys, catalogs, maps, and other documents. 4. Being quantitative, the methods provide greater discrimination along the spec trum of taxonomic differences and are more sensitive in delimiting taxa. Thus th ey should give better classifications and keys than can be obtained by the conve ntional methods. 5. The creation of explicit data tables for numerical taxonomy has already force d workers in this field to use more and better-described characters. 6. A fundamental advantage of numerical taxonomy has been the re-examination of the principles of taxonomy and of the purposes of classification. 7. Numerical taxonomy has led to the reinterpretation of a number of biological concepts and to the posing of new biological and evolutionary questions. Thus, t he method is coming into its own as a best tool in biological research. 5.5. Operational Steps in Numerical Taxonomy or Phenetics Numerical taxonomy is operational. It's operations are carried out in the follow ing sequence : organisms and characters are chosen and recorded; the resemblance s between organisms are calculated; taxa are based upon these resemblances; and last, generalizations are made about the taxa (such as inferences about their ph ylogeny, choice of discriminatory characters, etc.). It should be clear that generalizations about the taxa can not be made before on e has recognized the taxa; that taxa cannot be recognized before resemblances be tween organisms are known; and that these resemblances cannot be estimated befor e organisms and their characters have been examined. Therefore, although some of these steps may be in effect combined in certain computational methods, or the whole procedure may be repeated a second time for some special reason, the order of the steps within the procedure cannot be changed without destroying the rati onale of the classificatory process. Numerical Taxonomy 53 5.5.1. Phenetic Methodology In this study, referring to taxa by the species or generic names, is avoided. In

stead, these groupings are replaced with operational taxonomic units (OTU's). Fi rst, select the group of organisms to be studied. The character states be determ ined for each operational taxonomic unit (OTU) or entities to be classified. The successive steps in this study are : 5.5.2.1. Operational Taxonomic Unit (OTU) The lowest ranking taxa or taxonomic entities of lowest rank used or among item, individual convenient group or the basic units in any specific study are called operational taxonomic units (OTU's). 5.5.2.2. Unit Taxonomic Character A general definition of a character is : any attribute or descriptive phrase ref erring to form, structure or behaviour which the taxonomist separates from the w hole organism for a particular puspose such as comparison or interpretation. Cai n and Harrison (1958) describe a single character as "anything that can be consi dered as a variable independent of any other thing considered at the same time". Characters as such are strictly abstract entities : it is their expressions or s tates that taxonomists deal with. "Sepal length" for instance may be considered as a character; but "sepal length 12 mm" is an expression of that character. For purposes of comparison, a character must be divisible into two or more expressi dns or states (including here presence or absence in qualitative features). For practical purposes a character may be defined as any feature whose expressio n can be measured, counted or otherwise assessed (cf. Hedberg, 1957). Examples a re : seed weight, ratio of corolla to calyx length, number of teeth on a leaf, p resence or absence of tubercles in leaf axils, number of transverse septa in ant hers, to cite a few at random. For the purposes of numerical taxonomy Sneath and Sokal (1962) state that charac ters should be unit characters or if they are multiple they should be broken dow n into unit characters. As a working definition of a unit character they propose "a taxonomic character of two or more states, which within the study at hand ca n not be further subdivided logically". The character is a property or feature which varies from one kind of organism to another (Michener and Sokal, 1957). 5.6. Types of Basic Characters In numerical taxonomy the basic characters are of following major types : 5.6.1. Binnary Character A character having only two contrasting states or two state character e.g. 'Yes' or 'No' present ' or absent (+ and , or 1 and 0), "Possessed" or "Not Possessed" . 5.6.1.2. Multistate Character A character having three or more different states. It is of following two types : A. Qualitative multistate characters (unordered multistate characters). The mult istate characters possess three or more qualitatively contrasting forms, each ra nking equal e.g. flower colour red, yellow, blue. B. Quantitative multistate characters. These represent quantitative measures of the size on a continuous scale such as length, height or weight etc. 54 Plant Taxonomy 5.6.2. Choice of Characters In constructing a classification, all kinds of characters may be employed provid ed : 1. They are inherent in the organism or phenetic manifestations of the genome of an organism or taxon. 2. They are not susceptible to wide environmental modifications. The choice of character is a matter of the scientific judgement and experience o f the taxonomist. 5.6.3. The Requisite Number of Characters Maximum number of unit characters (OTU's) should be used to obtain a fairly and reliable classification. It is usually recemmonded that not less than 60 and pre ferably 100 or more characters should be used. According to Sneath and Sokal (19 77) these characters should be from all parts and all stages of life cycle. 5.6.4. Coding and Scaling of Characters

Coding and scaling are steps that convert the crude data into a form suitable fo r computation and also preserve the kind of information the taxonomist wishes to consider in making a classification. Coding is largely concerned with logical decisions and divisions, and scaling is most often some form of mathematical transformation. The characters selected then have to be coded. Coding of characters may be of fo llowing types : 5.6.4.1. Two state coding This is simplest form where characters are divided into + and which are used for simple coefficients of association. The symbol 'NC' meaning "no comparison" or 0, is used when no data are available or an organism is not possessing a particu lar character present in other organisms. 5.6.4.2. Multistate coding In this coding the multistate quantitative characters are divided into several s tates, such as 1, 2, 3, 4 etc., by breaking up a continuous variable into steps of equal weightage. The qualitative multistate characters can not be arranged in a reliable sequence of definite order. Therefore they are coded as some alphabetic symbols. 5.7. Data Matrix Data matrix is tabulation of data, such as taxonomic characters, to show differe nce between categories (taxa) often in a machine (computor) readable form. The d ata matrix is also called data chart, data table, comparison chart etc. In data matrix the data obtained by scoring the characters in the OTUs are now s et up in a table in matrix form, consisting of t OTUs scored for n characters. T he data matrix is called a t x n (or i x m) table the one column representing th e organisms, the other representing the characters (Table 1). Table 1. Coded data table (t x n table) Character states Taxa (OTUs) A B C D 1 + + NC 2 + + + + 3 + + + 4 + NC NC Numerical Taxonomy 5 + + + + 6 + + + 7 + + NC 8 NC + + 9 + + + + 10 + + + 5.8. Estimation of Resemblance The resemblance between a pair of operational taxonomic units (OTUs) can be expr essed in terms of similarity i.e. the percentage of characters in which they agr ee or disagree. Phenetic resemblance between the taxonomic entities or groups is generally estim ated by following three methods : 1. Coefficients of association 2. Correlation coefficients 3. Measurement of taxonomic distance. 5.8.1. Coefficients of Association It is simplest method and mainly refer to characters expressed as plus or minus (two-state). Coefficients of association are estimated by the following formulae : (a) Sheath (1957) Ns S Ns + Nd where S is the numerical index, Ns = number of positive features shared by any t wo OTUs : Nd = number of features positive in one OTU and negative in the other OTU.

(b) Grower (1971) has proposed a general coefficient of similarity applicable to all three types of characters, two-state, qualitative and quantitative characte rs. This is estimated by the formula : rz Grower coefficient (SG) E Sijk Where Sijk is a score on character i for comparison of OTU's j and k such that S ijk = 1 (1 x x 1/R). Here R is the range of character i over all t OTU's. If all characters are qualitative (1, 0) then SG becomes S. 5.8.2. Correlation Coefficients (r) Correlation coefficient is a measure of the linear relationship between the quan titative variables, indicating the degree to which they vary together; denoted b y r, values range from 1 (perfect negative correlation), through 0 (no linear rel ationship) to +1 (perfect positive correlation). These are estimated by the following formula : rz rjk [(xij - x i) ik . k) 4([ - x j)2] [I (xik x.k)2]) where j and k for two unites under comparison, xij = value of character i in uni t j; xik = value of character i in unit k; Yj and )7 k = mean for all characters in units j and k; n. = number of characters. 5.8.3. Measurement of Taxonomic Distance Taxonomic distance (d) is an expression of the relationship between individuals or taxa in terms of multidimensional space, where each dimension represents a ch aracter, based on quantitative estimates of dissimilarity. 55 56 Taxa A BC"E F CHI J A tt4:t. rf-tt+. B C D E F 70-phonon 90-phenon FO-phonon 90-phenon a E C F (1 J H Plant Taxonomy % similarity 100 90-99 80-89 70-79 ct0.60 1-1-1 50-59 ft ABtEl C F G J D H Taxa Fig. 5.1. Left: Schematic diagram showing a matrix of hypothetical similarity co efficients between pairs of taxa; the magnitude of the coefficients is shown by depth of shading. Right : The same coefficients as in left-hand diagram, arranged by placing simil ar taxa adjacent to each other. Note the formation of triangles of high similari ty values. (After Sneath, 1962) Taxonomic distance (d) is estimated by the following formula : djk =-\11E (xij x ik) 2 100 A B E --

90 Texa I C F G J D H 1 1 85% phonon line 110 75%.phenon line Co 70 65% phenon line 60 50 40 Fig. 5.2. Dendrogram representing the hypothetical hierarchy obtained from Fig. 5.1. Ordinate indicates magnitude of similarity coefficient at which stem join t o form higher ranking taxa. Horizontal lines delimit taxa of equal rank. Numerical Taxonomy 57 where xij is the character state of unit j for the character i; xik = character state of unit k; E = sum of n characters; d = taxonomic distance in a phenetic s pace divided by 5.8.4. Cluster Analysis Clusters are the groups of operational taxonomic units. Cluster analysis is a me thod or procedure for arranging OTU's into homogeneous clusters on the basis of their mutual similarities. Zn duster analysis, t`fte tom-pita scats out clusters ) the OTC's accor&mg to their overall similarity, that is, according to the numb er of characters or attributes in common. Cluster analysis can be done approximately by shading the elements of the matrix to indicate the degree of similarity so that very similar organisms are adjacen t and appear as dark triangles. The allows the main phenetic groups (phenons) to be identified visually in simple cases (Fig. 1). 5.8.5. Phenons and Rank Phenons are the groups of similar organisms recognised by cluster analysis. Phen ons may be prefaced by a number indicating the percentage of similarity at which they are formed, e.g. an 80 phenon indicates a group whose members are affiliated at no lower than 80 per cent. of the similarity scale used in the particular an alysis. In practice a phenon defines groups by drawing lines horizontally across the dendrograms* : these lines must be straight and parallel to the abscissa. I n this way similarity coefficients may be used in criteria of rank : we can ther eby delimit taxonomic groups objectively by choosing arbitrarily fixed levels of similarity. For example, in Fig. 2 by drawing a horizontal line across the dend rogram at a similarity level of 75 we have created four 75 phenons (ABEI, CFG, J, and DH). Phenons need not, however, be equated with classical taxonomic units su ch as species, genus, family etc. Phenetic classification is the grouping of taxa by overall similarity, regardles s of whether these similarities are symplesiomorphous or synapomorphous in phylo genetic sense. *Dendograra or tree diagram is a branching diagram in the form of a tree used to depict degrees of relationship or resemblance. Phenogram is a dendogram of phen etic relations. 6 CHEMOTAXONOMY 6.1. What is Chemotaxonomy ? The uses of Chemical characters of plants in classification or in solving taxono mic problems is called Chemotaxonomy, Chemical taxonomy, Plant Chemotaxonomy, Ch emosystematics, Chemical Plant Taxonomy. The basic chemical compounds used in Ch emotaxonomy are alkaloids, carotenoides, tannins, terpenoids flavonoids, polysac charide, betalins fatty acids, amino acids, aromatic compounds etc. The examples of presence or absence of chemical characters are : aromatic plants in Juglanda les, non-aromatic in Fagales; Plants produce betalains but not anthocyanins in t

he Caryophyllales, anthocyanins but not betalains in Polygonales; Plants are gen erally tanniferous in Sapindaceae. Plants generally have alkaloids in Solanaceae (e.g. Nicotiana, Datura), Plants have highly aromatic compounds in the Lamiacea e (Cronquist, 1981). Chemical contributions to the classification of plants are based on their chemic al constituents i.e., on their molecular characteristics. Such characters are ge netically controlled and have the advantage over morphological characters that t hey can be exactly described in terms of definite structural and chemical formul ae. The method of chemical taxonomy is thus exact and simple in principle consis ting of investigations of the distribution of chemical compounds or groups of bi osynthetically related compounds in series of related or supposed related plants . 6.2. Brief History Chemotaxonomy has a great antiquity and the chemical characteristics of plants h ave been noted and used by taxanomists. Its early history is largely a history o f uses of plants in medicine. The herbalists of past began to group plants posse ssing similar medicinal properties. Helen C. de S. Abbott (1886, 1887) a pioneer in the field wrote "The vegetable k ingdom does not usually claim our attention for its intellectual attainments, al though its members would certainly seem to possess greater chemical skill than a higher race of beings exhibit in laboratories". Greshoff (1909) published the presence of tannins, alkaloids, cyanogenetic compo unds and saponins in a wide variety of plants. He suggested the use of comparati ve chemistry in taxonomy. McNair, has tried to apply comparative chemistry to taxonomy. According to him a ny one alkaloid rarely occurs in more than one family, but may occur in several members of one family e.g., protopine in the Papaveraceae. This alkaloid protopi ne is not found in any plant of other families (Manske 1954). McNair (1935), has discussed Angiosperm phylogeny on a chemical basis and has tr ied to use alkaloids, glycerides (fats, oils etc.) and volatile oils as aids to taxonomy. According to him plants can be classified chemically in accordance wit h the substances made by them. Such chemical classification may be compard with or used as supplement to morphological classification and may (58) S. I Chemotaxonomy 59 be of some importance in the development of true natural system of angiosperm ph ylogeny. He further adds that the chemical products of more highly evolved plant s have larger molecules and that the iodine number of glycerides is higher in th e more highly evolved groups. According to Hutchinson Fumarioideae is quite distinctly separated as a group fr om Papaveraceae proper and closely allied to certain genera of the family Berber idaceae e.g., Epimedium, Aceranthus and Bongardia. That there is close affinity with certain genera of Papaveraceae especially Chelidonium is certain. He writes that this alliance is more apparent than actual, and that the Fumariaceae have not arisen directly from the ancestors of the present Papaveraceae. 6.3. Classification of chemical characters used in Taxonomy There is no suitable classification of chemical characters used in taxonomy. Mentzer (1966) recognised three principal kinds of chemical constituents which h elp in taxonomy; (1) primary or basic constituents like proteins, nucleic acid d erivative chlorophylls and polysaccharides; (2) Secondary constitute which lack nitrogen and are not involved in the basic metabolism of cells; (3) Miscellaneou s substances. Turner (1969) classified these according to their molecular size into following : (a) Micromolecules : Compounds of relatively low molecular weight alkaloids, amino acids, cyanogenic glucosides, glucosinolates (mustard oil glucosides), pigments (anthocyanins, betalains, and so on), phenolics (flavonoids and so on), and ter penoids.

(b) Macromolecules : Compounds of high molecular weight (over 1,000) such as pro teins, DNA, RNA, cytochrome c, ferredoxin complex polysaccharides. The term Semantides is also used for information carrying proteins such as DNA, RNA etc. Naik (1984) divided these chemical characters used in taxonomy into (i) directly visible, e.g. starch grains, raphides etc. (ii) plants products e.g. alkaloids, flavonoids and terpenoids (iii) proteins. Micromolecular data have practical utility in taxonomic problems at the generic level or below. They have been used to study variation caused by natural hybridi zation as well as other problems. In such studies, micromolecular data have been treated as characters used in classification. Generally macromolecular data are most helpful when applied to problems of phylogeny above the generic level. The taxonomically important chemical compounds and their systematic value, is as follows : 6.4. Tannins Tannins are amorphous, rarely crystalline substances which are widely distribute d in the vegetable kingdom. They are remarkable for their astringent taste and f or their ability to form coloured solutions and precipitates with iron and other metals. Families with all species tanniniferous : Anacardiaceae, Casuarinaceae, Lauracea e, Lythraceae, Magnoliaceae, Meliaceae, Moraceae, Myrtaceae, Oxalidaceae, Polygo naceae, Punicaceae, Rhizophoraceae, Sapindaceae, Sapotaceae, Sterculiaceae, Tili aceae, Ulmaceae, Urticaceae, Vitaceae, Winteraceae, etc. Families with all species without tannins : Acanthaceae, Amaranthaceae, Basellac eae, Campanulaceae, Capparidaceae, Caryophyllaceae, Chenopodiaceae, Convolvulace ae, Brassicaceae, Cucurbitaceae, Lamiacae, Papaveraceae, Portulacaceae, Solanace ae, Verbenaceae, Violaceae, Zygophyllaceae etc. 60 Plant Taxonomy 6.5. Steroids Steroids (high molecular weight solid alcohols) Cardenolides are cardiotonic gly cosides and these are present in Apocynaceae, Asclepiadaceae; Euonymus (Celastra ceae); Mallotus (Euphorbiaceae); Digitalis (Scrophulariaceae); Isoplexis (Ranunc ulaceae). estr Sex hormones (Steroids) : Three types of sex hormones are formed by animals ogens, gestogens, and androgens. Estrone occurs in date palm seed, pollen and in seeds of pomegranate, Malus sylvestris (Rosaceae), Khaya grandifoliola (Meliace ae). Estriol occurs in the catkins of Salix. Androgens occur in Haplopappus hete rophyllus. Steroids are sometimes referred to as sterols (Gr. stress, solid). Sterols occur in many plants viz., Type of sterol Plant (Family) Aegelin Aegle marmelos (Rutaceae) Calosterol Calotropis gigantea (Asclepiadaceae) Cholesterol Solanum tuberosum (Solanaceae) Phoenix (Palmaceae) Dioscorea (Dioscoreaceae) Ergosterol Hevea brasiliensis (Euphorbiaceae) Citrus (Rutaceae) Triticum sativum (Poaceae) 6.6. Terpenoids The odour of flowers, fruits, leaves (e.g. mint and Eucalyptus) and wood (e.g. S andal wood) is due to essential oils natural organic compounds called terpenes. The terpenoids have been classified on the basis of number of C5 (or isoprene) u nits present in them into : 1. Monoterpenoids with two `isopene units : C10 (e.g. Citral, Limonene). 2. Sesquiterpenoids with three Toprone' units : C15 (e.g. Farnesol). 3. Diterpenoids with four 'isoprene' units : C20 (e.g. Vitamin A). 4. Triterpenoids with six isoprene units : C30 (e.g. Squalene). 5. Polyterpenoid with several 'isoprene' units : (C5)n (e.g. Rubber). Terpenoid compounds are found in mints, umbellifers, citrus plants etc. Many tri bes of the Asteraceae (Compositae) are characterized by the sesquiterpene lacton

es they produce. Rind and leaf terpenoid patterns in Citrus were used to determine the origin of certain Citrus cultivars. Sesquiterpene lactones have been used to show that in the tribe Vernonieae, Vern onia has two major centres of distribution one in the Neotropics and the other i n Africa. On the basis of terpenoid distribution Ambrosia, Iva, Franseria and Xanthium are removed froth the tribe Heliantheae (family Asteraceae) and placed in a separat e tribe. 6.7. Flavonoids Flavonoids are the largest group of naturally occurring phenols. Phenols contain the hydroxyl group(s) attached directly to the aromatic nucleus (e.g. C6H5OH). They are mostly found in the vacuoles of the plant cell. Various types of flavon oids include flavones, flavonones, isoflavones and isoflavonoids, flavonols, ant hocynidins, chalcones, aurones and biflavonyls. Cronquist and Takhtajan placed the Juglandales in the Hamamelidae. But Thorne as signed Juglandaceae to the Rutales. But the presence of 5 methoxylated flavonols f avours relationships with the Hamamelidae. -4 Chemotaxonomy 61 Thorne considers that Fabaceae is allied to Rutales by the presence of phenylate d flavinoids in both Fabaceae and Rutaceae. Presence of flavinoids indicates that the Arecaceae is closely related to the Po aceae, due to the presence of tricin and lutealin (leaf flavonoids) in both the families. Sterculiaceae is closely related to Malvaceae by the presence of Cyanidin and go ssypetin in both the families. Hybrids resulting from accidental crossing of two species of Baptisia often look ed different from either parents. Alston and Turner (1960) found that the flavin oid profiles of the two parents were mutually exclusive in several compounds; ce rtain species specific compounds from each parent were found together in the hybri ds. 6.8. Sulphur Compounds These compounds are also acting as chemotaxonomic character. 1. Disulphides : These are known to occur in chopped onions, garlics, leeks etc. (Allium spp.). These substances are largely responsible for the odours of these plants. 2. Glucosilates : It is a group of nearly 85 thioglucosides which on hydrolysis from glucose and isothiocyanate. Glucosinolates (mustard oil glucosides) occur in the families Brassicaceae, Capp aridiaceae Tovariaceae, Moringaceae etc. On this basis Brassicaceae (e.g. odour of sulphur in juice) and Cappariaceae are placed in order capparales which were previously placed in one order Rhoeadales alongwith Fumariaceae and Papaveraceae . 6.9. Alkaloids Alkaloids are a heterogenous group of organic nitrogen containing bases, often w ith a heterocyclic ring. True alkaloids have a nitrogen-containing heterocyclic nucleus derived from a biogenetic amine and they can be related structurally to parent bases such as pyridine, piperidine, isoquinoline and tropane. The alkaloid protopine is present in all the species of Papaveraceae. Protopine is also present in all the species of Fumariaceae, showing their close relations hip with Papaveraceae. But Fumariaceae members have a combination of isoquinolin e derivatives which are not found in Papaveraceae. Berberidaceae, Fabaceae, Ranunculaceae and Solanaceae are rich in alkaloid conta ining species. The members of Papaveraceae synthesize isoquinoline alkaloids, th ose of the Fabaceae lupin alkaloids and those of the Solanaceae tropane derivati ves. Morphine is restricted to the opium poppy (Papaver somniferum), coniine to a few members of Apiaceae and strichnine to a few species of Strychnos. 6.10. Waxes These occur chiefly on the surfaces of leaves and stems Palmaceae produce large

amounts of wax on their leaves. The wax from Carnauba palm, Copernicia cerifera is collected and marketed in great quantity. Another palm Ceroxylon andicola pro duces wax on its trunk. Seeds of Simmondsia have a liquid wax. 6.11. Betalains These pigments are not flavinoids but are functionally equivalent to phenolics. Due to the presence of nitrogen they are excluded from the general definition of phenolics. The best-known example of chemical evidence resolving taxonomic problems at orde r level is of a order Centrospermae. 62 Plant Taxonomy All the 10 families Chenopodiaceae, Portulacaceae, Aizoaceae, Cactaceae, Nyctagi naceae, Phytolaccaceae, Stegnospermaceae, Basellaceae, Amaranthaceae and Didiera ceae, have the red and yellow pigments betacyanins and betaxanthins and are plac ed together in a single order. The families Molluginaceae and Caryophyllaceae ar e excluded from Centrospermae, as these families do not contain betalins. 6.12. Iridoids It is a separate class of monoterpenoid cyclopentanoid lactones. Iridoids provid e valuable taxonomic evidence. Several iridoid containing families have been sug gested as the putative progenitor of the Asteraceae, including Rubiaceae, Dipsac aceae, Calyceraceae, Caprifoliaceae, Scrophulariaceae and Cornaceae, all of whic h iridoids, which are not found in Asteraceae. Only the Campanulaceae, Araliacea e, and Apiaceae, which do not produce iridoids, remain as possible progenitor li nes. Asperuloside (iridoid) is common in Rubiaceae. Acubin (iridoid) is found in Corn aceae, Scrophulariaceae, Orobanchaceae etc., Buddleia, an acubin containing genu s is transferred from Loganiaceae to Buddleiaceae, with a position near Scrophul ariaceae. Iridoid glycosides (asperulosides and acubins) are distributed in the Rubiaceae, Scrophulariaceae, and related families, Fabaceae, Cornaceae etc. The interfamiliar relationship was evidenced between Cornales. Scrophulariales, Caprifoliales and Rubiales by the pressure of iridoids. 6.13. Non-Protein Aminoacids Presence of cyclopropyl amino acids in Sapindaceae and Aceraceae show their clos e relationship. Canavanine, a close analogue of arginine is found only in the Fabaceae. In Lathyrus, 7 intrageneric groups, and in Vicia, 14 intrageneric groups are rec ognised on the basis of distribution of amino acids. 6.14. Starch Grains The form of starch grains is useful in the taxonomy of Poaceae. In the Poaceae, the tribe Hordeae have compound grains; Nardus, Lolium, Parapholis have simple g rains. 6.15. Protein Behnke and Turner (1971) discovered characteristic protein inclusions in the sie ve-tube plastids of the plants in the Centrospermae. This evidence suggests that Cactaceae, Caryophyllaceae and Molluginaceae were correctly assigned to the Cen trospermae. 6.16. Nucleic Acids DNA of wheat is more similar to DNA of rye (Avena sativa) than that of barley (H ordeum vulgare). On the basis of DNA nucleotide the eight species of Atriplex (Chenopodiaceae) we re confined. 6.17. Ellagic Acid It is a phenolic compound and is absent from all members of subfamily Rosoideae of Rosaceae, except for the single tribe Kerrieae. 6.18. Melanins According to Thomson (1965) true melanins are probably produced by the legumes V icia faba, Cystisus nigricans and Sarahamnus scoparius; and the banana. Chemotaxonomy 63 6.19. Steroid Saponins The steroid saponins are present in Trigonella (Fabaceae); Solanum, Cestrum (Sol

anaceae); Digitalis (Scrophulariaceae). They are particularly prominent in 3 clo sely related families of monocotyledons : Liliaceae, Agavaceae and Dioscoreaceae . 6.20. Raphides Raphides are present in about 35 families of Angiosperms. But according to Gibbs (1958) raphides are absent in the monocot orders Alismatales, Hydrocharitales, Scheuzeriales, Potamogetonales, Triuridales, Najadales, etc. but present in the Liliales. 6.21. Cyanogenic Compounds The poisonous compounds release by plants are called cyanogenic compounds and pl ants are known as Cyanogenests or cyanogenetic plants. Some of the glycosides on hydrolysis yield HCN. This character is useful in taxonomy. Swain has recorded the production of hydrocyanic acid (HCN) from members of more than 80 families. All the genera of family Passifloraceae and of closely related Turneraceae yield HCN. No member of Cucurbitaceae and Violaceae is cyanogenetic. The 30 to 40 days old plants of Sorghum vulgare (Jowar) are cyanogetic but not t he mature plants. The use of chemical criteria adds a powerful weapon to the armoury of the taxono mist and when properly applied it is undoubtedly more useful than much subjective morphology. . "Strictly speaking one might demand that every accurate description of genus or of a new species should be accompanied by a short 'chemical description' of the plant (Swain, 1965)". 13 HERBARIUM A herbarium is a store house of plant specimens. In it dried, pressed, preserved and mounted plant specimens are arranged in a sequence of an accepted system of classification for future reference and study. The collected plant specimens fr om far and wide, mounted on appropriate sheets and kept in pigeon holes of steel or wooden cupboards for study at the same place and time. According to Fosberg and Sachet (1965) a modern herbarium is a great filling sys tem for information about plants both primary in the form of actual specimens an d secondary in the form of published information, pictures and recorded notes. 13.1. Functions of Herbarium The main and valuable functions of herbarium are : 13.1.1. Conservatory of material and data The herbarium is a storehouse of plant materials with valuable data mentioned on their labels. The material in the herbarium remains as permanent record of flor a of those regions even the natural topography and the vegetation have changed o r extinct. In these the herbarium provide evidence of what once existed. The herbarium specimens bear labels with adequate data on habitat, common name, native uses, ecological notes like frequency and abundance of species, associate d plants, habit etc. These types of data for the particular species, but from di fferent collections of different regions, when carefully studied and analysed, p rovides valuable material for almost complete description of various taxonomic c haracters such as morphological range of variation, distribution etc. A herbarium receives fresh material by collection of its staff, through gifts, e xchanges etc. So it is a conservatory of material and data. 13.1.2. Role in teaching and research Herbarium acts as an aid in teaching botany to degree and post-graduate students . While teaching a teacher can show herbarium specimens if fresh material is not available at the time of giving the course. It is of essential requirement for biosystematic research i.e. for correct ident ification and nomenclature besides a source of material. It provides research ma terial for anatomical, palynological and chemotaxonomical studies. According to Davis (1961) "in the herbarium we can compare all the related speci es of a genus in the same place, in the same state and at the same time".

13.2. Kinds of herbaria The kinds of herbaria depend upon the contents, purpose, region/place, plant gro ups present in these : (101) 102 Plant Taxonomy 13.2.1. The herbaria of medicinal plants The herbaria include specimens of plants of known medicinal value or supposed to 'have medicinal properties. 13.2.2. Herbaria of weeds The herbaria have weeds of cultivated fields and waste places. 13.2.3. Regiorial herbaria The herbaria of a particular region or place e.g. Herbarium of Arid Zone circle, Jodhpur; Herbarium of Central Circle, Allahabad; Herbarium of a University or I nstitution. 13.3. Important herbaria of the world The first herbarium of world has founded in 1545 in the University of Padua, Ita ly. In the world there are about one thousand five hundred recognised herbaria, besides many smaller herbaria with institutions, universities, pharmacies etc. S ome important herbaria of the world and India are listed below along with their country, year of founding and approximate number of sheets. Some Important Herbaria of the World Herbarium Place/Country Year of founding No. of sheets (approx.) Royal Botanic Garden, Kew V. L. Komarov London, U. K. 1853 6,500,000 Botanical Institute Leningard, USSR 1823 5,000,000 Museum National d'Historic Naturelle Laboratorie de Paris; France 1635 5,000,000 Phanerogamme., Conservatorie et Geneva, Switzerland 1817 4,000,000 Jardin Botaniques New York Botanic Garden New York, USA 1891 3,000,00 0 U.S. National Museum Washington USA 1868 3,000,000 Vienna Botanischer Gaertn. Vienna, U.S.A. 1748 2,500,000 National History Museum Chicago, Austria 1893 2,350,000 Royal Botanic Garden Edinburgh, UK 1761 2,500,000 Missouri Botanic Garden St. Louis, USA 1859 1,700,000 National Herbarium Melbourne, Australia 1857 1,500,000 Zurich Botanischer Gaertn Zurich, Germany 1834 1,500,000 Gray Herbarium Cambridge, USA 1807 1,485,000 Haravard University, Philadelphia Academy of Sciences. U.S.A. 1812 1,000,00 0 Philadelphia Arnold Arboretum Boston, U.S.A. 1872 700,000 Department of Agriculture Peradeniya, Ceylon 1817 85,000 Gardon college Rawalopindi, Pakistan 1893 60,000 Herbarium of the Rangoon University, Rangoon Burma 1947 15,000 Central National Herbarium Calcutta 1793 2,500,000 Forest Research Institute Herbarium Dehradun 1874 3,00,000 Botanical Survey of India, Southern Circle Coimbatore 1874 2,00,000 Blatter Herbarium St. Xavier's College Bombay 1906-7 100,000 Botanical Survey of India, Eastern circle Shillong 1956 100,000 Botanical Survey of India, Western Circle Poona 1956 135,000 Botanical Survey of India, Northern circle Dehradun 1956 60,000 Botanical Survey of India, Industrial section Calcutta 1887 50,000 Botanical Survey of India, Central circle Allahabad 1955 40,000 National Botanic Garden Herbarium Lucknow 1948 100,000 14 PLANT COLLECTION AND SPECIMEN PREPARATION Good collecting is the corner-stone of whole edifice. To built new herbarium or to enrich our herbaria, not to fill them with poor scraps, the botanical collect

ion must have a purpose. The purposes of botanical collections can be following : 1. To built new herbarium. 2. To enrich the herbaria. 3. To write a flora. 4. To carry out research on a particular genus/species. 14.1. What to collect ? What plants shall be collected depends on the purpose of study. The possible mai n approaches are : 1. To collect as much as one possibly can. This method is desirable in little-kn own ,areas but only when the collector is able to cope with the tiring work invo lved. 2. To collect certain groups, either those in which a taxonomist is particularly interested, or of which herbaria require better representation. In this. connec tion, expert advice should be sought. 3. It is more valuable to collect a limited amount of material really well, than to collect a great deal in a cursory manner. 4. In herbaceous plants the rootstock should be collected. Whenever possible spe cimens should be collected in flower and fruit. 5. Plant material collected should, as far as possible, be representative of the ir populations. 6. The size of specimens should be approximately 42 x 28 cm. because this is the standard size of card-sheet on which the specimens will finally be mounted. 7. Atleast six specimens of each plant are to be collected, to facilitate the ex change and detail studies. 14.2. Field Work Depending upon the purpose of plant collections, the field trip can be a i.e. co llection trip of only one day to some near by place, or exploration i.e. intensi ve four to six weeks collection done for preparing detailed floristic accounts o r expedition i.e. plant collection in remote and difficult areas for several mon ths may be made. 14.3. Procedure For Collecting The Plant Specimens 14.3.1. Planning Proper planning and making arrangements before undertaking a long trip is essent ial. The following points may be followed : (104) Plant Collection and Specimen Preparation 105 1. Try to get available maps and collect local information. 2. Try to have best guide and for carrying the equipment select preferably peopl e who are more or less acquainted with forest work. The local forest rangers and forest guards are of great help. 3. Calculate the food you must take with you especially the fast food. Keeping i n mind that a porter will carry about 15-18 kg. excluding his personal belonging s. 4. Try for camping in Forest Rest House or Travellers Bungalow. 14.3.2. Equipments The equipments for field work vary in different situations. The important ones a re : 1. A collecting pick for digging up roots and rhizomes of herbaceous plants so t hat the collected specimen is complete. 2. A strong knife or machete is sometimes required for cutting branches and othe r plant parts. 3. A pair of pruning shears or secateur for cutting woody and hard material. 4. A pair of forceps for flower bud study. 5. A vasculum for accommodating collected specimens. 6. A plant press with blotters or newsprint for pressing and keeping the collect ed specimens. 7. Several spare sheets of blotting paper (blotters). 8. A filled book (field diary) for noting down the details of collected plants. 9. Tags of thin tin or aluminium. The tag is numbered and tied to specimen in th

e field. This number should correspond with the number of the page of field book on which details of plant was noted. 10. Polythene bags or collecting bags. These are convenient containers for fresh plant specimens. After storing plants in these bags, the mouth of bags should b e tightly closed to avoid wilting of plant material. 11. Hand lens and pocket lens. 12. Camera and film. 13. Field press, folders of absorbent tissue papers or even newspapers. 14. General equipments First aid box, food stuff, water container, Haversack (Sh oulder bag), pen, pencil etc. 14.3.3. Field notes The data accompanying the specimens are as important as the specimen itself, bad ly labelled material being more of hinderance than a help. The main objective of field notes is to record data of taxonomic value that will not be apparent from the specimen. The following details should be recorded in field not book. 1. Date. 2. Number of plant a serial number or collector's name and number (e.g. Saxena 3 125) or collector's name, year and number (e.g. Saxena 1976/123). 3. Name of plant (as far as known even if only to family). 4. Province or whatever major divisions of country you wish (e.g. state, or dist rict). 5. Locality name of the place, distance and direction with reference to familiar or known place. 6. Altitude in metres. 7. Habitat. 8. Description Chief characters of specimens viz., colour of flower, aroma, late x, size, curious charcters. 106 Plant Taxonomy 14.3.4. Handling of fresh material There are three important methods of handling freshly collected plant material v iz., 1. To press each plant as it is collected in the field then and there. 2. To keep all the specimens collected in a vasculum and bring them to camp. The y are then pressed one by one. 3. This method is used more in the tropical rain forests than to temperate regio ns, is to carry collected specimens in a rucksack and brought to the camp where they are pressed. 14.4. Pressing of plant specimen Pressing is the process of placing specimens between the absorbent drying paper under heavy pressure in a plant press. Corrugated cardboard can be inserted betw een the drying paper to even out the pressure and to facilitate the aeration of the press. Pressing is most important step in the preparation of specimen and ne eds careful attention. The specimens should be carefully selected before pressin g and the following points may be taken note of : 1. Underground parts like root, rhizome, etc., must be included in herbs. 2. Roots and underground parts should be washed thoroughly before pressing to re move soil particles. 3. Select specimens in flowering or fruiting stage. 4. Ensure that specimens are free from insect feeding, fungal infection and any pathological symptom. 5. Fresh specimens should be arranged within the pressing paper. 6. Maximum surface of pressing paper is covered with the plant specimen to be pr essed. 7. Single specimen should be pressed in a folded pressing paper. 8. Herbaceous specimens, longer than 40 cm, may be folded in to V shaped, N shaped o r M shaped manner and then pressed. 9. Care should be taken to press the plant parts att4ched to the plant in their natural position. 10. A few leaves should be arranged with their lower surfaces facing upward. 11. A few flowers should be pressed separately of the specimens with gamopetalou

s corolla. Some of the flowers should be split open and then pressed to expose t heir essential organs and the nature of the thalamus. 12. Fleshy and bulky organs should be sliced open to facilitate quick drying. Pa per padding or cotton batting around such organs is helpful for drying. 13. To avoid decoloration and moulding, blotters should be changed frequently de pending upon the nature of specimen and humidity. 14. In case of pressing deliquescent corolla or perianth, a layer of absorbent c leaning tissue is put over and beneath the flower before pressing. This tissue i s peeled off after the specimen becomes dry. 14.5. Drying of specimens The essential thing is to dry the plant specimen as quickly as possible before t he tissues have time to decay. It is therefore necessary to change the drying pa per every day atleast for the first few days. For this press may be put in the s un or near the gentle artificial heat, when the specimens are quite dry they are ready for poisoning. Drying of specimens is carried out in two ways : 4 Plant Collection and Specimen Preparation 107 14.5.1. Drying of plant specimens without artificial heat It is followed universally and supposed to be most satisfactory. Plants are pres sed in pressing papers between blotters. Corrugates are not used in this method. Plant press is locked for 24 hours. This is called 'sweating period'. When swea ting period is over, press is unlocked, blotters are removed and pressing papers are turned back. Now the specimens are rearranged as the situation demands. Disadvantages in this method are : 1. Much longer time is needed for drying the specimens. 2. Blotters are to be changed at least 3 or 4 times during the entire process an d wet ones are to be dried in the sun before using them again. 3. A large number of blotters are needed. 4. Less number of specimens can be processed per collector. 5. Specimens dried without artificial heat are liable to be infected by fungi be cause much time is taken in complete drying. 6. Larvae present within the flowers eat away some of the important parts. 7. Labour cost of drying per specimen is higher. 14.5.2. Drying of plant specimens with artificial heat It is much preferred by the collectors. The specimens are locked tightly within the field press for 24 hours. It is called 'sweating period'. After the specimen s have sweated, press is unlocked, specimens rearranged and put again in the dry ing press, locked with less pressure and are allowed to dry with the aid of arti ficial heat. ' In this method corrugates are invariably used which provide air passage to the p lant press for circulation of dry and heated air. For providing heat, gasoline s toves, kerosene lamps, coal fire, electric bulbs and hot steam pipes are used. E ven large electric ovens may be taken into service and plant press is kept insid e it at a suitable temperature. Fernald (1945) has pointed out some disadvantages in drying of the specimens wit h the aid of artificial heat : 1. Plant specimens become brittle during drying. 2. Specimens loose their glaucescence or waxy bloom and colour. 3. Sometimes the specimens get permanently marked with the ridges of corrugates. 14.6. Poisoning Poisoning of the specimens should be done immediately after collection. Poisonin g kills the plants and thereby prevent the formation of abscission layer and dec ay. Generally used poisoning chemicals are mercuric chloride lauryl pentachlorophena te (LPCP), formalin and the fumigates (volatile poisonous liquids) like methylbr omide, carbon disulphide or carbon tetrachloride, paradichlorobenzene (PDB) etc. 14.7. Mounting When the specimen as well pressed, dried and poisoned, it is affixed with lebel on a mounting sheet (42 x 28 cm). While mounting following precautions should be

observed : 1. Mount only one specimen on one sheet. 2. Lower part of plant should be at the base of sheet. 3. Label should be be pasted/printed on the right hand bottom of the sheet. Moun ting is done either by glueing thespecimen on sheet or by stiching. 108 Plant Taxonomy 14.8. Labelling of specimen After mounting the specimen, herbarium sheet is given a final touch by pasting a label at its lower right hand corner. The label is of the size 11 x 7 cm but it s size may vary by 1.2 cm on either side. Some herbarium prefer to get the label printed on sheets. A label must provide the following informations : Flora of No. Family Botanical name Local name Habit Flower colour Habitat Locality Altitude Ecological notes Date of collection Collector Collection number Additional information 14.9. Storing of specimens When specimens are ready in all respect, these are stored in wooden or steel alm irahs for safe deposit and future record. These almirahs are essentially designe d with two to three tiers of pigeon holes. The specimens are arranged in almirah s according to any internationally recognised system of plant classification aft er identification (consult chapter on Plant Identification). Each species is kept in a folded species cover which is lighter in weight and sm aller in dimension than genus cover. Many species of a genus are kept together i n a genus cover. The genus cover usually has the name of genus written or printe d over it. After identification the specimen should be incorporated (filed) in the herbariu m. 14.10. Herbarium methods Herbarium methodology includes all activities viz., preservation, identification , labelling, documentation, incorporation, taxonomic revision, loan and exchange etc. connected with the specimens. 15 PLANT IDENTIFICATION Plant identification implies assiging a plant to a particular taxonomic group ul timately to the species. The identification of plant specimen is its determinati on of as being identical with or similar to another and already known plant. Ide ntification is the determination of a taxon as being identical with or similar t o another and already known elements; the determination may or may not be arrive d at by the aid of literature or by comparison with plant of known identity. No names need be involoved in the process of identifying a plant. The naming of pla nt or nomenclature is different than identification. When the unknown plant is collected from a known locality, the common practice i s to refer to a book accounting for the plants of that region. This contains usu ally the analytical keys and descriptions. 15.1. Methods of Identification 15.1.1. First Method The first step is the determination of the families to which the unknown plant b elongs. Knowing the name of the family one can turn the keys to genera for deter mining the generic name and then for the specific identity of the plant to the s pecies key. Since, for many reasons, the identity and name of the plant obtained

may be incorrect, it is always safe to check the description of the plant to en sure that there is a reasonable agreement between the characters observed in the unknown plant and those given in the description of the plant presumed to be. 15.1.2. Second Method Second method is the utilization of the latest floras and check list of the part icular region. These comprise usually an index to the plants known for the local ity and generally provide other pertinent habit, distributional and frequency da ta. By the process of elimination an unknown plant can be assigned to a genera h aving one or more species, and identification may be completed by comparison of characters with those given in any standard work accounting of the plants of tha t area. 15.1.3. Third Method Third method is the identification by means of monographs or revisionary works a ccounting for the particular family or genus. 15.2. Plant characters before its Identification Study the plant specimen to be identified in detail. Mention the following chara cters : 1. Nature of specimen - herbaceous, or woody; annual or perennial. 2. Phyllotaxy and venation. (109) 110 Plant Taxonomy 3. Inflorescence type Capitulum (e.g. Asteraceae), Cyathium (e.g. Euphorbiaceae) , Verticellaster (e.g. Lamiaceae) etc. 4. Flower and its parts actinomorphic or zygomorphic. 5. Presence of epicalyx (e.g. Malvaceae). 6. Number of sepals and petals or tepals, their aestivation. 7. Petals free (e.g. polypetalae) or fused (e.g. gamopetalae). 8. Number of stamens and their position antipetalous (e.g. Chenopodiaecae) alter nipetalous or obdiplostemonous, (e.g. Caryophyllaceae). Staminal tube (e.g. Malv aceae). 9. Count number of carpel/carpels, style gynobasic (e.g. Lamiaceae); stigmas. 10. Type of placentation. 15.3. Keys A key is a device for easily identifying an unknown plant by a sequence of choic es between two or more statements. A key is an artificial analytical device or a rrangement where by a choice is provided between two contradictory characters re sulting in the acceptance of one and the rejection of the other. Statements in t he keys are based on the characters of the plants (mentioned above). For example , a key might separate taxa using the following choices : (1) herbaceous versus woody if herbaceous, the woody plants are eliminated; (2) the next choice, zygom orphic flowers versus actinomorphic, if zygomorphic, the plants with actinomorph ic flowers are eliminated and so forth. Each time a choice is made, the number o f taxa that remain is reduced by the use of contrasting characters. If sufficien t numbers of ch6racters are contrasted, the number of possibilities is eventuall y reduced to one. 15.3.1. Types of Keys Keys used in floras are usually diagnostic, that is, identifying an unknown plan t by the conspicuous features. Diagnostic characters are sometimes referred to a s key characters. A key may be short and limited to a single pair of contradicto ry statements or propositions. A single pair of contradictory statements or each pair of choices is called a 'couplet'. Each statement of a couplet in a key is called 'Lead'. Keys do not offer descriptions of the plants concerned, but state only the essential diagnostic characters by means of which the taxa can be iden tified. Keys are of two types : punched card keys and dichotomous keys. 15.3.1.1. Punched Card Keys Punched card keys are used in the school, colleges etc. by the students. Punched card keys consist of cards of suitable size with names of all the taxa (all fam ilies, genera or species for which the key is meant) printed on each one of them . Each card has a number and any one character printed near one of the corners.

All the taxa showing this character are indicated by a perforation in front of t heir names, while those lacking this character are without any perforation. 15.3.1.2. Dichotomous Keys A dichotomous key presents two contrasting choices or couplet at each step. The key is designed so that one part of the couplet will be accepted and the other r ejected. The first contrasting characters in each couplet are referred to as the primary key characters. These are usually the best contrasting characters. Char acters following the lead are secondary key characters. The dichotomous keys are of two types, viz., Indented key (Yoked key) and Bracke ted key (Parallel key). Plant Identification 15.3.1.2.a. Indented key or Yoked key A dichotomous key in which the first part of a contrasting couplet is followed b y all subsequent couplets; each subordinate couplet being indented one step furt her to the right for clarity of presentation. The indented key is the one most w idely used in manuals for the identification of vascular plants. In the indented key, each of the couplets is indented a fixed distance from the left margin of the page. 15.3.1.2.b. Bracketed or Bracket Key or Parallel. Key A dichotomous key in which contrasting parts of a couplet are numbered and prese nted together, without intervening couplets, although the brackets joining each couplet are now omitted. Example : The plants used in the example are common genera of the family : Ranun culaceae, viz., Clematis, Anemone, Ranunculus, Aquilegia and Delphinium. A. Indented Key Fruit a group of achenes; flowers not spurred, petals absent Sepals usually 4; involucre none Clematis Sepals usually 5; involucre present Anemone Petals present Ranunculus Fruit a group of follicles; flowers spurred Flowers regular; spurs 5 Aquilegia Flowes irregular; spur 1 Delphinium The first choice, within the above genera is between "Fruit a group of achenes; flowers not spurred" and "Fruit a group of follicles; flowers spurred", these pa ired statements being given the same indention. If the latter choice is taken, t he next choice, shown of the indention, is between "Flowers regular, spurs 5" an d "Flowers irregular, spur '1'. Thus the plant in question has follicles and irr egular flowers with a single spur, it must be a Delphinium. B. Bracketed Key or Parallel Key In this the two couplets are always next to each other in censecutive lines on t he page. The same example of bracketed key is given below : f ,T) (i) Fruit a group of achenes; flowers unspurred (2) c7 (i) Fruit a group of follicles; flowers spurred (4) (2) Petals absent (3) (2) Petals present Ranunculus (3) Sepals usually 4; involucre absent Clematis (3) Sepals usually 5; involucre present Anemone (4) Flowers regular; spurs 5 Aquilegia (4) Flowers irregular; spur 1 Delphinium The number at the right end of a line in the bracket key indicates the next numb ered pair of choices to be considered. The keys use the most conspicuous and clearcut characters, without special regar d to those considered taxonomically the most important. For this reason the sequ ence of taxa is often quite artificial, and such keys are frequently termed arti ficial keys. Artificial key is an identification key based on convenient phenoty pic characters and not indicating phylogenetic relationships. Natural key is an identification key constructed from a natural classification and indicating the supposed evolutionary relationships of the group within the branching sequences

of the key. 111 Plant Taxonomy 15.3.2. Comparison of Indented Key and Bracketed Key Indented Key Bracketed Key 1. Each couplet has its 2 leads indented by the same amount from the left-h and margin of the page. 1. Each couplet has its 2 leads immediately adjacent under the same left-hand number. 2. The first couplet to be consulted is the one least indented and which ha s its first lead at the head of the key. 2. The first couplet to be consulted stands at the head of the key next to the number 1. 3. The next appropriate couplet to be consulted is the one with its first l ead immediately below the chosen lead of the previous couplet, its leads being t he next least indented pair below the latter. 3. The next appropriate cou plet to be consulted is indicated by the reference number to further down the ke y, placed on the right-hand side of the chosen lead. 15.3.3. Construction of Key In constructing a key following techniques may be followed : 1. Key should be dichotomous. 2. The first word of each lead of the couplet should be identical. For example, if the first lead of a couplet begins with the word fruit, the second lead of th e same couplet must begin with the word fruit as in example. 3. The two parts of the couplet should be made up of contradictory statements so that one part will apply and the other part will not i.e. rejected. 4. Do not use overlapping ranges or vague generalities in the couplets. 5. The couplets should be of positive statement e.g. "leaves narrow versus leave s not narrow". 6. Use distinct and readily observable features. 7. The leads of consecutive couplets should not begin with the same word, since this may cause confusion. 8. It may be necessary to provide two sets of keys in some groups; flowering ver sus fruiting material, vegetative versus flowering, or staminate versus pistilla te for dioecious plants. 9. Couplets of a key may be numbered or lettered, or may use some combination of lettering and numbering, or may be left blank in the case of indented keys. Keys are traditional method of identification in taxonomy. If keys are well writ ten with adequate specimens and carefully, then the specimen can be successfully identified. Keys, however, have several major disadvantages. The use of certain characters is required even if the character is not evident in the unknown spec imen. 15.4. Recent Methods of Identification Attempts to improve on the traditional identification process by using keys have resorted to either polyclaves or computer techniques for plant identification. These methods provide multientry (or multiaccess) opportunities that is, the pla ce of entry is not fixed by the format of the key. 15.4.1. Polyclave Identification A polyclave is a multientry, order-free key implemented in one of several differ ent formats. One form is a diagnostic key that uses cards placed on top of one a nother to eliminate taxa that disagree with the specimen to be identified. 112 Plant Identification 113 Polyclaves have a tremendous advantage over dichotomous system of plant identifi cation. In this system the user is free to select appropriate characters for eac h unknown specimen. The route taken to a particular taxon differ considerably fr om one plant specimen to another. The logic of identification with a polyclave i s the same as that in a key, but the user is free to choose any character, in an y sequence, thus avoiding the rigid format of traditional keys. The polyclaves in existence are (a) cards with holes commonly referred to as "pe ck-a-boo" or "window" cards (b) edge-punched or "key sort" cards and (c) semitra

nsparent overlays. (a) Peck-a-boo or Window Card Key The peck-a-boo or window card key to world angiosperm families uses a card for e ach character (Hansen and Rahn, 1969). Round holes are punched beside a family n umber for the families that have that character. The plant to be identified is e xamined and its characters noted. With a good specimen, it is possible to find 2 0 to 25 characters. One selects those cards corresponding to the characters foun d on the plant. The cards are then put on top of one another in a stack. Familie s having all the observed characters will be indicated by the holes, or "windows ", which are easily seen when the cards are held in front of a light. The logic is simple. Each time a card is added, families not perforated in this new card a re excluded. 15.4.2. Computerised Identification The plant identification by computer programme system has been introduced by Mor se (1974). The computerised efforts of plant identification are grouped into (a) computer-c onstructed keys (b) computer stored dichotomous' keys (c) automated pattern reco gnition systems and (d) simultaneous character-set methods. Recently polyclaves may be puched by computer or standard computer cards (Pankhu rst, 1974). Two general kinds of computer-based polyclaves have been developed. One kind dev eloped for qualitative taxon-character data, employs elimination. The other, dev eloped for use of taxon-character frequency tables, employes likelihood ratios o r other probabilistic techniques. Another form is a computer-stored multientry k ey. Still another polyclave is a printed table or matrix giving the status of va rious taxa and characters useful for separating the taxa. 15.5. Artificial Keys for the Identification of Major Groups 1. Plants with seeds; ovules enclosed within the ovary Angiosperms 2. Flowers mostly tetra- to pentamerous; two cotyledons Dicotyledons 3. Both sepals and petals present. 4. Petals are free; flowers with calyx and corolla Polypetalae 5. Polysepalous; petals hypogynous 6. Dome-shaped, unexpanded thalamus Thalamiflorae 6. Thalamus expanded into a disc Disciflorae 5. Gamosepalous; inferior ovary Calyciflorae 4. Petals are fused; flowers with calyx and corolla Gamopetalae 7. Ovary inferior Inferae 7. Ovary superior 8. More than two carpels Heteromerae 8. Only two carpels Bicarpellatae 3. Either sepals or petals are present; rarely both absent Monochlamydae 2. Flowers mostly trimerous; one cotyledon Monocotyledons 1. Plants with seeds; ovules exposed, i.e. not enclosed within the ovary Gymnosperms -m1111111.1111, 19 MODERN TRENDS IN PLANT TAXONOMY INTRODUCTION Modern taxonomists consider that the gross morphological characters are not alwa ys sufficient to provide means of differentiation in determining the genetical a nd evolutionary relationship between taxa. To achieve this the taxonomical evide nces from anatomy, embryology, palynology, cytology, palaeobotany, ecology, bioc hemistry etc. are discussed. Dr. V. Puri has said "One of the most significant modern trends in plant taxonom y is towards a synthesis between the older methods, outlook and more recent deve lopments in our knowledge of plants". Bailey (1949) has said, if a truly natural classification is to be attained, it must be based upon "the analysis and the harmonization of evidence from all orga ns, tissues and parts". 19.1. Morphology in relation to taxonomy

Gross morphology has no doubt provided the foundation and framework for taxonomy , but it has become increasingly clear that contributions to systematics may com e from almost any branch of biology. The modern taxonomist has, therefore, to ha ve a broader outlook than his predecessor a few years ago (see Constance, 1995, 1957; Keck, 1957). Morphology is the study of structure and form of plants, usually dealing with th e organism and its component organs : Morphology has been the most widely used tool in the classification of higher pl ants from a very long time. Morphological features have been so extensively stud ied by botanists and taxonomists in various classes of plants that it might be s aid that there is little left to learn. According to modern concepts, however it is not correct. Practically all the herbaria systems of classifications, botani cal manuals are based on comparative morphology and anatomy. Morphological characters are traditionally useful as evidence at all taxonomic l evels, but particularly at the specific and generic ranks. Morphological evidenc e provides the basic language for plant characterizations, identification, class ification, and relationships. Generally, morphological data are easily observabl e and obtainable, and thus most frequently used in taxonomic studies. Morphological characters are plants habit, root structural types, stem habit, st em structural types, bud structural types, leaf structural types, inflorescence types, flower types, perianth structural types, androecial types, stamen types, gynoecial types, carpel types, ovule types, fruit types and seed types. Modifica tion of flowers in stamen number, anther position, ovary position, style length, stigma shape, number of carpels, number and fusion of perianth parts etc, contr ibute to the reproductive success of the species. The growth habit (herbaceous o r woody) of plants may be of primary usefulness in classification. Brassicaceae and herbaceous; Asteraceae have both woody and herbaceous members. --e. (132) Modern Trend in Plant Taxonomy 133 19.1.1. Examples of morphological characters in taxonomy 19.1.2. Growth habit Herb, Shrub and Tree Hutchinson has divided dicotyledons into two major divisions Lignosae and Herbac eae on the basis of their woody and herbaceous habit. The Lignosae being a funda mentally woody group, with some herbs derived from woody plants and Herbaceae be ing a fundamentally herbaceous group with a few shrubby plants derived from herb aeous. Herb members of Ochidaceae; woody members of Fagaceae. 19.1.3. Root Adventitious Monocots, Tap Dicots. 19.1.4. Stem Bulb Allium, Rhizome Iris. 19.1.5. Leaf Pitcher Leaves in Sarracenia (pither plant); tenticular leaves in Drosera (sunde w). 19.1.6. Inflorescence Umbel in Apiaceae. 19.1.7. Corolla Cruciform in Brassicaceae, rosaceous in Rosaceae, pappillionaceous in Papilionac eae. The features of floral morphology are most important in the classification of fl owering plants. 19. . Anatomy in relation to taxonomy is believed that internal structure of plants can provide more information than external morphology. However, anatomical features cannot by themselves constitut e the basis of classification but can be used with advantage to supplement those morphological attributes on which classification has been built. The anatomical studies of organs of flowering plants can serve as an integral pa rt of taxonomy. Swamy and Bailey (1949) remarked that "before attempting to arra nge surviving Angiosperms in phylogenetic series, it is essential to obtain reli able evidence regarding salient trends of evolutionary specialisation in the var

ious organs and internal structures of these plants. Such evidence can be acquir ed only by comprehensive and time consuming investigations of the dicotyledons a nd monocotyledons as a whole". 19.2.1. Vegetative anatomy 19.2.1.1. Stomata Stomatal study is very useful in the fields of taxonomy. Stomata have been class ified according to the position of subsidiary cells, guard cells in relation to the aperture. They are of the following types : 1. Ranunculaceous or anomocytic (Anomocytic = irregular celled). This type is ch aracterised by having a limited number of subsidiary cells which are quite alike the remaining epidermal cells; the accessory or subsidiary cells may be four or five in number. In most of the cases, these subsidiary cells are just like the other epidermal cells. This type of stomata occur in Ranunculaceae, Capparidacea e, Malvaceae and some other families. 2. Cruciferous or anisocytic (Anisocytic = unequal celled). This type of stomata occur in Cruciferae (Brassicaceae) and many genera of Solanaceae. In this type, each stoma remains surrounded by three subsidiary cells of which one is distinc tly smaller than the other two. 134 Plant Taxonomy 3. Rubiaceous or paracytic (Paracytic = parallel celled). In this type, the stom a remains surrounded by two subsidiary or accessory cells which are parallel to the long axis of the pore and guard cells. This type of stomata occur in Rubiace ae and allied families. 4. Caryophyllaceous or diacytic (Diacytic = cross celled). In this type each sto ma remains surrounded by a pair of subsidiary or accessory cells and whose commo n wall is at right angles to the guard cells. This type of stomata occur in Cary ophyllaceae and allied families. 5. Gramineous type. The gramineous stoma possesses guard cells of which the midd le portions are much narrower than the ends so that the cells appear in surface view like dump-bells. They are commonly found in Gramineae (Poaceae) and Cyperac eae of monocotyledons. There are 31 known types of arrangement of subsidiary cells in the mature stomat al complex of vascular plants (Ditcher, 1974). At higher taxonomic levels, these distinct patterns play a big part in classification. Within the Combretaceae, the stomata are anomocytic in the Combretoideae and par acytic in the Strephonematoideae. While the stomata are anomocytic in the Scroph ulariaceae, they are diacytic in the closely related Acanthaceae. However, stoma tal traits are not always reliable, e.g. anisocytic, anomocytic, diacytic and pa racytic exist together in the same leaf of Lippia nodiflora (Verbenaceae). 19.2.2. Crystal Crystals of calcium oxalate are found in many families Apocynaceae, Begoniaceae, Cactaceae, Caricaceae and Euphorbiaceae. In these families sphaeraphides are co mmon. Raphides are very common in the monocotyledons like Araceae and Musaceae. According to Nair et al. (1977), the size and sliape of crystals in the abaxial epidermis of the leaves of Myristica fragrans can enable us to differentiate bet ween male and female trees even at the sapling phase. In general, simple crystal s dominate in male trees and compound ones (druses) in female trees. 19.2.3. Latex tissue Laticiferous tissue is found in a number of families. Butomaceae, Araceae, Musac eae, Urticaceae, Podostamanaceae, Nymphaeaecae, Papaveraceae, Fumariaceae, Resed aceae, Euphorbiaceae, Aceraceae Caricaceae, Cactaceae, Sapotaceae, Apocynaceae, Asclepiadaceae, Corivolvulaceae and Companulaceae possess milky or watery latex. 19.2.4. Cystoliths Crystals of calcium carbonate are found in the families Acanthaceae, Boraginacea e and Urticaceae. In the former two families the stalk of the cystolith is not c onspicuous where as in the latter it is characteristic. 19.2.5. Secretory cells Possession of lysigenous or scizogenous cavities with oils, resins or mucilages provide taxonomic character in families like Rutaceae, Myrtaceae and Apiaceae. 19.2.6. Vascular bundle

19.2.6.1. Types A. Scattered vascular bundles : These are found in monocot stems. B. Vascular bundles in ring : It is a characteristic of Dicot stem anatomy. Modern Trend in Plant Taxonomy 19.2.6.2. Bicollateral vascular bundle These are found in Cucurbitaceae and Gentianaceae. 19.2.6.3. Cortical bundles They are found in a number of families e.g., Araliaceae (Panax), Bombacaceae (Bo mbax), Cactaceae (Rhipsalis), Ficoideae (Mesembyanthemum), Oleaceae (Nyctanthus) and Asteraceae (Vernonia). 19.2.6.4. Medullary phloem Araliaceae (Aralia racemosa), Amaranthaceae, Chenopodiaceae, Crassulaceae (Echev aria), Euphorbiaceae (Ricinus, amphivasal bundles), Melastomaceae (Melastoma), N yctaginaceae, Polygonaceae, Piperaceae, Asteraceae (Dahlia, Lactuca and Sonchus) . 19.2.6.5. Internal phloem It occurs in Apocynaceae, Asclepiadaceae, Convolvulaceae, Cucurbitaceae, Lythrac eae, Melastomaceae, Myrtaceae, Onagraceae, Punicaceae, Solanaceae, Thymelaeaceae , Papilionaceae (Mucuna). 19.2.6.6. Interxylary phloem forminate type Salvadoraceae, Apocynaceae (Lysonia), Asclepiadaceae (Leptadenia lancifolia, Asc lepias obtusifolia and Ceropegia), Combretaceae (Calycopteris). 19.2.6.7. Interxylary phloem concentric type Chenopodiaceae (Majority of genera), Icacinaceae (Phytocrene), Loranthaceae (Nuy tsia floribunda) Menispermaceae (Anomospermum), Onagraceae (Epilobium and Oenoth era), Papilionaceae (Wistaria). Successive ring of cambium Amarantaceae, Chenopodiaceae, Menispermaceae and Nyctaginaceae. 19.2.6.8. Hard bast The pericycle opposite to the vascular bundles is made up of fibres of sclerench yma but it may be parenchymatous throughout as in many members of Ranunculaceae. It may form a continuous ring as in Cucurbitaceae, Piperaceae and Dioscoraceae. The pericycle may be made up of the phloem fibres with a thickening of cellulos e as in Nerium and Vinca. 19.3. Wood anatomy The evidentiary characters are wood cell type, wood cell size, wood cell shape, wood cell wall sculpture, wood cell patterns, stelar patterns, xylem maturation types, vascular bundle types; types of wood rays ground tissue parenchyma epider mis, mesophyll, stomata, sclereids, trichomes, crystals, nodes, petiole vasculat ion, venation, phloem cells etc. Based on wood or wood anatomy, Amborella, Tetracentrom and Trochodendron are tra nsferred to independent families. A new family Degeneriaceae is created. Rapatea ceae differs from Xyridiaceae in having silica bodies, tannin cells and type of chlorenchyma. Magnoliales are considered primitive on the basis of wood anatomy. 19.4. Floral anatomy Floral anatomy for delimiting taxa is used in Asclepiadaceae, Opuntia, Cuicuta a nd Evolvulus (Tyagi et al., 1966). According to him, Cactaceae is closely relate d to Calycanthaceae. V. Singh (1977) found primary pattern of floral development in all the taxa in Alismatales as trimerous. 135 136 Plant Taxonomy Eames (1953) suggested the removal of Paeonia from Ranunculaceae and elevation o f a new family Paeoniaceae on the basis of floral anatomy. The role of floral anatomy has been emphasised in solving taxonomic problems. Ac cording to Puri (1958), the gynoecium structure in the Capparidaceae, Cruciferae , Moringaceae and Papaveraceae is essentially alike. Though the placentation is parietal, the placental strands are inverted and occur on the inner side of the secondary marginal bundles which have normal orientation. This supports the view that the Moringaceae should be included in the Rhoeadales and that the parietal

placentation has been derived from that axile condition. As a similar phenomeno n, takes place in the Cucurbitaceae and Passifloraceae, it lends support to the suggested alliance between the Moringaceae and Passifloraceae. From a study of floral anatomy of four species of the Onagraceae, Roy (1949) obs erved significant variations in the vascular supply of the ovary of Trapa, justi fying its removal to a separate family (Trapaceae). The separation of Hydroc, ty le asiatica as Centella asiatica has been confirmed on anatomical grounds by Mit ra (1995). 19.5. Embryology in relation to taxonomy A recognition of the value of Embryology in taxonomy was delayed because of the time and trouble involved in collecting embryological data. According to Maheshw ari (1964) and Bhojwani and Bhatnagar (1978), the characters of taxonomic value in delimiting plant groups include the (a) anther; (b) quadripartition of the microspore mother cell; (6) development a nd organisation of the pollen-grain, (d) development and structure of the ovule; (e) origin and extent of the sporogenous tissue in the ovule; (f) megasporogene sis and development of the embryosac; (g) form and organisation of the mature em bryosac; (h) fertilisation; (i) endosperm; (j) embryo and (k) seed-coat. 19.5.1. Cyperaceae In the family Cyperaceae, while all four microspor.e nuclei are produced after m eiosis, three of them are cut off on one side of the pollen grain and only the f ourth develops to form the generative cell and then the male gametes. All genera and species of Cyperaceae studied in Europe and Japan (and at Delhi by Mr. C.K. Shah) show this character and it is possible to identify a member of this famil y just as definitely by a microscopic study of its anthers as by other floral ch aracters. Further, the simultaneous type of microspore formation and the functio ning of all the four microspores in the Juncaceae indicate that it is this famil y from which the Cyperaceae have probably been derived. 19.5.2. Centrospermales The Cactaceae agrees with the rest of the Centrospermales in having the followin g embryological characters : (a) glandular anther tapetum whose cells become two t o four nucleate; (b) microspore-mother cells is which two meiotic divisions are succeeded by a si multaneous quadripartition into the microspores; (c) trinucleate pollen-grains; (d) campylotropous ovules with strongly curved funiculi and massive nucellic; (e ) a hypodermal archesporial cell which cuts of a wall cell; (f) a micropyle form ed by the swollen tips of the inner integument which protrude out and approach t he functions; (g) formation of a nucellar cap originating from, periclinal divis ion of cells of the nucellar epidermis; (h) functioning of the chalazal megaspor e of the tetrad : (i) formation of a monospork eight-nucleate embryosac; (j) fun ctioning of the perisperm as the main storage region; (k) disappearance of most of the endosperm in the mature seed generally leaving merely a single-layered ca p over the radicle (Maheshwari, 1964). Modern Trend in Plant Taxonomy 19.5.3. Loranthaceae The studies of Johri and associates (1957) on the Loranthaceae show that the Lor anthoideae is embryologically different from the Viscoideae as regards mode of d evelopment of embryosac, endosperm, embryo and in the location of the viscid zon e of the fruit and that the subfamilies should be raised to the status of famili es. 19.5.4. Onagraceae and Trapaceae A monosporic tetranucleate embryo-sac is characteristic of all members of the On agraceae and is not found in any other family of angiosperms. The genus Trapa ha ving an eight-nucleate embryo-sac, which was once placed in the Onagraceae, has since been removed and assigned to a new family Trapaceae (Pulle, 1938). Manasi Ram's (1956) work on Trapa bispinosa fully confirms this view. Earlier, Eames (1 953) expressed the view that on anatomical evidence also Trapa does not belong t o the Onagraceae and is not even closely related to it. Table I presents the emb ryological differences between the families Onagraceae and Trapaceae. Table I

Trapaceae Onagraceae Ovary Semi-inferior and bilocular with a single Inferior, mostly tetralo cular. Usually many pendulous, anatropous, bitegmic ovule in anatropous bitegmic ovules in each each chamber. chamber. Megasporogenesis The chalazal 'megaspore invariably Usually the micropylar megaspore functions and the embryo-sac is 8-nucleate functions. The embryo-sa c is tetranucleate Endosperm Endoserm lacking Endosperm Nuclear Proembryo Solanad type Onagrad type Suspensor Extremely well developed and haustorial Short and inconspicuous haustorial Embryo Cotyledons extremely unequal Cotyledons equal Fruit Large, one-seeded drupe Usually a loculicidal capsule. 19.5.5. Santalaceae Gagnepain & Boureau (1946, 1947), raised doubts about the position of Exocarpus and stated that instead of being regarded as an angiosperm it should be assigned to the gymnosperms and given a place somewhere near the Taxaceae. Lam (1948) co mmented as follows : "At any rate, Exocarpus seems an interesting case and proba bly represents a transition between the protangiospermous gymnosperms and the Mo nochlamydeae". The embryological studies of Manasi Ram (1958) have clearly shown , however, that Exocarpus is a perfectly valid angiosperm with an archesporial c ell functioning as a megaspore mother cell, an embryo-sac of the Polygonum type, a cellular endosperm with a chalazal haustorium, and a pericarp derived from th e wall of the ovary. Its correct position, therefore, lies in the Santalaceae to which it was assigned by previous systematists. 19.5.6. Other taxonomal cases settled by embryological data 19.5.6.1. Paeonia was previously kept in Ranunculaceae but later on it was found that by a natomy and pollen characters, this genus differs from other members of Ranuncula ceae. Embryological studies found that embryogeny of Paeonia is very different f rom other members of Ranunculaceae and this genus was then shifted from Ranuncul aceae. 19.5.6.2. In Butomaceae Butomus is unique in having a monosporic 8-nucleate embryo sac; Bu tomopsis, Limnocharis and Hydrocleis has bisporic, 5-nucleate embryo sacs. Butom us is retained in Butomaceae while all the others are transferred to Alismatacea e. 137 Plant Taxonomy 19.5.6.3. Embryological evidences supported Hutchinson's (1959) treatment of Alismataceae and Butomaceae as separate families and the placement of the latter along with H ydrocharitaceae in the same order. 19.5.6.4. The separation of Hippocratea from Celastraceae into a separate family Hippocrat eaceae. 19.5.6.5. The genus Sphenoclea is placed under Campanulaceae based on embryological studie s. 19.5.6.6. The splitting of Gentianaceae into Menyanthaceae and Gentianaceae; Malvaceae int o Bombacaceae and Malvaceae. 19.5.6.7. Embryogeny of Lemna, Wolffia and Arisaema confirms older views of an intimate re lationship of the duck weeds with the aroids (Maheswari & Khanna, 1957). It may be emphasized that while the embryologist lays no claim to erect a phylog enetic system of its own, embryological data need to be considered a ng with inf

ormation from other sources in order to approach a natural system of classificat ion. 19.6. Palynology in relation to taxonomy* 19.6.1. Pollen characters Polynology is the science of pollen and spores and its applications. It is deriv ed from the Greek word palynein meaning to scatter. The significance of pollen a ttributes in taxonomy has been realised during the last three decades (Erdtman, 1952; Wagenitz, 1995; Stix, 1960; Raj, 1961; Chanda, 1972; Nair, 1974). The oute r wall of pollen-grains is endowed with unique structural traits which are broad ly categorised in order of their importance in phylogeny into the apertures, exin e ornamentation, exine strata, shape and size. According to Bailey and Nast (1943), "there are families of dicotyledons in whic h the pollen is of very considerable taxonomic significance not only in the diff erentiation of subfamilies and tribes but also of genera and species". A few exa mples will illustrate the statement. The Caryophyllales is recognised by centros permous type of pollen with a spinulose and punctate-perforate tectum. The Malva ceae and the Compositae contain typically spinulose exine, the Plumbaginaceae ve rrucate pollen and the Gramineae smooth, sulcate ones. 19.6.2. NPC system The pollen classification is based on number-position character analysis, called NPC system. As a rule, the number of apertures is only one in the cryptogams an d gymnosperms whenever the grains are aperturate but the position is distal in t he latter and proximal in the former and the apertures are non-trichotomous (e.g . monolete) or trichotomous (e.g. trilete) in character. The apertural conditions of pollen-grains have been looked upon as strong charac ters in solving taxonomic problems. It has been recommended that taxa with the s ame general NPC formula be grouped together and those showing aberrant NPC separ ately. For example, the Parietales of Engler Diels has been probed by means of `pa lynological compass needle'. The NPC formula for the order is 345, i.e. pollen-g rains are 3-termate (N3), zonotreme (P4) and colporate (C5). A significant varia nt is the Canellaceae where the pollen has one distal colpus; here NPC is 133, h ence Hutchinson (1959) proposed shifting of Canellaceae from Parietales. *Erdtman, G. 1954. Pollen morphology and Plant Taxonomy. Bot. Not. 107: 56-81. 138 Modern Trend in Plant Taxonomy 139 19.6.3. Apertual conditions A general analysis of the apertural conditions in the plant kingdom reveals that apertures are ill-developed (primorphous) in thallophytes. In the archegoniates (comprising bryophytes, pteridophytes and gymnosperms), pollen-grains are trimo rphous and in the angiosperms polymorphous. The above morphological situation pr ovided a logical base for classifying the plant kingdom into the Primorphosporat ae (Syn. Thallophyta), Trimorphosporatae (Syn. Archegoniatae) and Polymorphospor atae (Syn. Angiospermae) by Nair (1974). This analysis has thrown light on the p hylogeny and evolution of the primitive angiosperms. 19.6.4. Stenopalynous and Eurypalynous taxa In angiospermic taxa, termed as "Stenopalynous", the pollen type is constant; wh ile in others, called "Eurypalynous", there are different pollen types varying i n size, shape, aperture, exine stratification, etc. The stenopalynous taxa are, as a rule, very natural. The occurrence of a pollen type, representative of a stenopalynous taxon in a plant of doubtful affinity, m ay provide an important indication of its taxonomic position. Amongst the eurypalynous taxa some may be natural and others quite heterogeneous . Different pollen types in eurypalynous taxa furnish important indications in c lassifying them into subgroups and then arranging these subgroups according to v arying degrees of similarity between each other. In certain cases this may even result in the splitting up of a taxon or submerging of others. Numerous examples of such rearrangements in families and genera have been given by Erdtman. As pointed out by Bailey and Nast (1943, p. 343) "There are families of dicotyle dons in which the pollen is of very considerable taxonomic significance, not onl y in the differentiation of subfamilies and tribes, but also of genera and speci

es". For example, after the studies of Lindau and Bremekamp pollen structure has become an indispensable character in the taxonomy of Acanthaceae. 19.6.5. Number of nuclei in pollen The number of nuclei in the pollen at the time of dispersal has been used by tax onomists (Brewbaker, 1967). The angiosperm pollen is either binucleate or trinuc leate according to the precocity of division of the generative nucleus. The binu cleate condition is considered as more primitive than the trinucleate. In the Ce ntrospermae, the pollen is uniformly trinucleate. The monocot (Liliaceae) is binucleate, the apetalous d polypetalous dicots ar e binucleate and gamopetalous members trinucleate. 19.7. Cytology in relation to taxonomy The applkation of cytological data in elucidation of taxonomic problems, it is s een that various attributes of chromosomes like number, morphology, size, behavi our in crosses and aberrations in reproduction are all important. 19.7.1. Chromosome number The haploid number of chromosomes in angiosperms ranges from n = 12 in Halopappu s gracilis (Asteraceae) to around n = 132 in Poa littoroa (Poaceae). Most of the angiosperms have chromosome numbers ranging between n = 7 and n = 12. About 35 to 40% per cent of the flowering plants are polyploids (S. Jones 1987). It is usually seen that closely related plants, like the different species of a genus, show chromosome numbers which reveal an arithmetic relation with one anot often in multiples of base number of characteristics of the genus. For examp her le the different species of Piper show -.1111 11111MINM'IMINIMIRIMi. 111W III ^ ^11/1^I^Mili 140 Plant Taxonomy chromosome numbers in multiples of 26, like 2n = 52 in P. nigrum, 2n = 78 in P. betle, and 2n = 104 in wild species of Piper (Mathew 1958). In Morus nigra has the highest chromosome number (2n = 308), in M. cathyana ther e are forms with 2n = 56, 84; 112 (Janaki Ammal, 1960). Solanum nigrum is a good example of the existence of a species complex, comprising diploid (n = 12), tet raploid (n = 24) and hexaploid (n = 36) forms (Tandon and Rao, 1966). 19.7.2. Chromosome morphology A study of chromosome morphology is informative to the taxonomist in assessing a ffinities and modes of origin of separate species. The important work of Babcock in Crepis show how chromosome morphology coupled with chromosome number is of c onsiderable importance in the genetic and taxonomic phases of study. In genera l ike Crepis and Plantago the large size and small number of their chromosomes hav e been of great value in this type of study. 19.7.3. Chromosome size It has already been discovered that evolutionary development involves in additio n to alterations in chromosome number, their size, changes in structure etc., so that analysis of these cytological characters may also shed important light on species relationships. Recent observations on the Menispermaceae (Mathew, 1958b) have shown that this aspect of cytology is sometimes valuable in taxonomic disc ussions. In this family the genera Stephania, Cyclea and Cissampelos are grouped under the Cissampelideae by Hooker. While Cyclea and Cissampelos show reduction in the perianth parts of the female flower Stephania shows no such reduction. C ytologically Cyclea and Cissampelos are seen to be based on 12, while Stephania shows n = 13. It is seen that the number n = 13 is characteristic of the tribe C occuleae, which further shows chromosomes of small size. Stephania also shows ch romosomes of small size. Stephania also shows chromosomes of small size while Cy clea and Cissampelos have much larger chromosomes. On cytological grounds theref ore Stephania may be transferred to the tribe Cocculeae. This is an instance of how chromosome size coupled with chromosome number provide data of taxonomic imp ortance. Large chromosomes, low chromosome number and symmetrical karyotype represent a pr imitive status, while small chromosomes, high number and extreme asymmetry indic

ate advancement (Sharma, 1964). The application of these principles has yielded interesting results in the Pandanales, Alismataceae, Hydrocharitaceae, Liliaceae , Amaryllidaceae and Dioscoreaceae. Within the Geraniales, chromosome study does not justify the exclusion of the Batsaminaceae as proposed by Engler. 19.7.4. Chromosome behaviour in crosses The behaviour of chromosomes in crosses is a reliable factor in assessing relati onships. Pathak (1940) made a careful karyotypic analysis of various species of Aegilops, Secale and Triticum, suggesting that the hexaploid T spelta and T vulg are were probably derived through hybridisation between a tetraploid wheat and R oy (1959) carried out detailed genome analysis of Aegilops longissima and A. sha ronensis. On the basis of chromosome pairing and fertility of F1 hybrids and the derived amphidiploidy, he thought that the two species are closely related. Fro m a study of the karyotypes of species of Aegilops, Cheenaveeraiah (1962) postul ated that the section Sitopsis should be shifted from Aegilops to Triticum or gi ven the rank of a new genus. Sisymbrium irio, a polytypic species, where plants of various sizes are availabl e, differing in the size and shape of leaves, flowers and fruits as well as bran ching pattern. According to Khoshoo (1960), variation and evolution within this species is primarily due to hybridisation and polyploidy and secondarily to gene mutation and structural changes in chromosomes. Modern Trend in Plant Taxonomy 141 Often there exists a close parallelism between taxonomical and cytogenetical stu dies. In Prunus persica for example the systematist satisfactorily classified th e variations; the geneticist found that these variations were inherited as per s imple Mendelian laws and the cytologists found the plants to be ordinary diploid s. Again in Pyrus malus the systematist, in spite of all the care bestowed, foun d the classification difficult and involved, and the cytologist found them to be seccAtfary polyploids while the geneticist found the analysis of the inheritanc e of characters extremely complicated on account of involved polyploid ratios (C rane, 1940). In Boragineae there had been considerable disagreement regarding the position of Brunnera macrophylla. Johnston (1924) recognised Brunnera as a distinct genus a nd Smith's (1932) cytological studies not only supported Johnston but also affir med that segregation of Anchusa myostdiflora and A. sempervirens into Brunnera m acrophylla and Caryolopha sempervirens. Removal of Brunnera macrophylla from the genus Anchusa is substantialized on three grounds, viz., (1) size of chromosome s, (2) form of chromosomes and (3) number of chromosomes. "Despite a healthy scepticism of the value of the chromosomes in easing taxonomi c decision- making, the annual outpouring of cytological data continues to illum inate the taxonomic study of many groups. With increasing knowledge the chromoso mes will further reflect the manifold ways in which evolution has proceeded and will thus become an increasing refined tool, among the many available to the dis criminating taxonomist" (Moore, 1978). 19.8. Ecology in relation to taxonomy The ecological criteria are of comparatively little direct inportance in taxonom y, though ecological criteria at the interspecific level can not be neglected. In flowering plants, tolerance and plasticity are widespread. The tolerance of a plant population is determined by its ability to survive and reproduce upon exp osure to a range of environmental factors. The tolerance is greater. when the ra nge is wider. On the other hand, plasticity is ascertained by the degree to whic h the appearance of plants vary in moving from one set of factors to the other. 19.8.1. Ecotypes Ecological studies can help in determining the taxonomic status of a species. On the basis of developmental morphology, culture experiments and other analytical data, the two species of Lindenbergia, L. polyantha and L. urticaefolia, have b een demonstrated to be ecotypes of the same species, the former being a calcicol ous or a miniature form of the latter (Misra and Siva Rao, 1948). Euphorbia thymifolia exhibits ecotypic differentiation in response to the calciu m content of the soil (Ramakrishnan, 1961) : the red ecotype is a facultative ca lcicole and the green one an obligate calcifuge. Pursuing the matter further, th

e red plant shows three physiologically distinct ecotypes. Depending on the soil exchangeable calcium, Boerhaavia diffusa, Gomphrena celosioides and Mecardonia dianthera give rise to appropriate calcicolous and calcifugous ecotypes. Euphorbia hirta bears three forms : one upright form, growing in protected habit ats, is a distinct ecotype from the other prostrate one; the latter also appears as a compact form of the , footpath and a diffuse one of the grazed lands which are interconvertiable in reciprocal transplants ( (Misra, 1967). ,IIN ^^^... 24 RANUNCULACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Dicotyledons Dicotyiedoneae Dicotyledones Polypetalae Archichlamydeae Herbaceae Thalamiflorae Ranales Ranales Ranales Ranunculaceae Ranunculaceae Ranunculaceae -Diagonostic features Herbs, leaves 'exstipulate, incised blades, sheathing bases, flowers hypogynous, spiral or spirocyclic; sepals often decidous, usually petaloid; calyx and corol la free; stamens indefinite, free; carpels polycarpellary, apocarpous; fruit agg regate. Distribution It is commonly known as buttercup family. It includes 35 genera and 1500 species out of which 163 species are confined to India. They are chiefly found in tempe rate and arctic regions. Vegetative characters Habit The plants areannual or perennial herbs or a climbing shrubs (Clematis, Na ravelia), rarely trees. They perennate by means of tuberous roots (Aconitum) or rhizomes. Root Tap root, adventitious or tuberous (Ranunculus spp. and Aconitum). The tap root system is in the initial stage but sooner or later replaced by the adventit ious roots. Stem Herbaceous, in some climbing (Clematis) or underground rhizome or erect, br anched. Leaves Generally simple, alternate, or opposite (Clematis) exstipulate rarely st ipulate (Thalictrum), sheathing leaf base, petiolate rarely sessile (Delphinium) . In some aquatic species leaves may show dimorphy (Ranunculus aquatilis); unico state or multicostate reticulate venation. Floral characters Inflorescence Solitary terminal (Anemone), axillary (Clematis), raceme (Aconitum , Delphinium) and cymose (Ranunculus spp.). Flower Pedicellate, ebracteate rarely bracteate, hermaphrodite, (unisexual in Th alictrum). Mostly actinomorphic (Ranunculus) rarely zygomorphic (Delphinium and Aconitum) hypogynous, complete, pentamerous. Calyx There is no distinction of calyx and corolla in most of the flowers. Sepal s 5, caducous, polysepalous, petaloid, imbricate or valvate aestivation. Corolla Petals 5, polypetalous, variously coloured, caducous or wanting; nectari es present - at the base of petals. Petals are united to form spur (Delphinium). (181)

182

Plant Taxonomy

0 (C) (D) Fig. 24.1. (A) Anemone numerosa; the outer whorl is of involucre. (B) Clematis, with only 4 petaloid sepals. (C) Myosurug, only sepals and petals shown; sepals have a spur each (D) Adonis (after Eichler) Androecium Stamens indefinite, _polyandrous, spirally'arranged on the thalamus, inferior; anthers dithecous, extrorse and adnate. Gynoecium Polycarpellary (one carpel in Delphinium and 3 to 5 in Aconitum), onoc arpous rarely syncarpous (Nigella), ovary superior, marginal placentation (axile in Nigella). Fruit Aggregate, etario of achenes (Ranunculus), etario of follicle (Aconitum), follicle (Delphinium), septicidal capsule (Nigella) or berry (Actaea), etario of drupes (Adonis), etario of berries (Hydrastis) and simple pod (Xanthorhiza). Seed Small, oily and endospermic. Pollination Generally entomophilous (Delphinium, Aconitum, Aquilegia) and anemop hilous in Thalictrum. Floral formula ED V K4-5 C5 A Ga Range of floral structure Three sub-families have been recognised on the basis of floral structure viz. (a ) Helleboreae (b) Anemoneae (c) Paeonieae. Rendle is inclined to believe that th e Anemoneae is more primitive than the Helleboreae. In the Anemoneae the flowers are actinomorphic. In Anemone, for instance, flower s are solitary with five or six or more perianth leaves which are petaloid. True petals are altogether absent and as many as 13 rows of stamens along with numer ous free carpels are found. In Fig. 24.2 Thalictrum, the inflorescence is a corymb or panicle; the perianth consists of 4 -5 sepaloid structures. Stamens are numerous but carpels may be few. In Clematis there are 4 or more petaloid sepals with several stamens and carpels. In all th e above three genera there is a pendulous ovule while in Ranunculus the ovule is erect. In this genus there are 5 sepals followed by 5 petals. The stamens and c arpels are numerous and free. The arrangement of stamens and carpels is typicall y spiral. In Myosurus all the five sepals have a basal spur; the stamens are few but carpels are many. In Adonis five sepals are followed by 8-16 coloured petal s. As many as 21 rows of stamens may be counted; carpels also numerous. In the Helleboreae the perianth is spirally arranged in the primitive genera but in the advanced forms it is arranged in a cyclic manner. Two distinct tendencie s are noticeable in this subfamily viz., (i) with actinomorphic flowers-considered more primitive. (ii) with zygomorphic flowers considered more advanced. In Helleborus the perianth has five members followed by a number of honey leaves . These are nothing else but modified stamens belonging to the outermost row. Th e number of stamens may Ranunculaceae 183 0 4 te .4'; la (A) "fttzcozr." (B) (D) 184 Plant Taxonomy 0 ( (A) (B)

Fig. 24.3 0 be 100 arranged in 13 rows. In Isopyrum and Coptis also there are 13-rows of sta mens but their total number is decreased. In Aquilegia the number of stamens may be only 15 or sometimes 20-25; there are only five carpels. Xanthorhiza also has a typical pentamerous d9c9 c2z, condition with 5-10 stamens and 5 carpels. In Cimicifuga the number e9d9 N4 a of carpels has been reduced. In Actaea the reduction in carpel has Cps ib gone further so that only one carpel is found. %c:z. Zygomorphic tendencies are found in Trollius, Nigella, cP6s9 Delphinium, Aconitum etc. In Trollius there are 5-15 perianth leaves, 21 rows of stamen and 5-10 carpels. In Nigella we come across 5 perianth l eaves, 8 honey-leaves, 8 rows of stamens and 5-12 syncarpous carpels. In Delphin ium and Aconitum zygomorphy arises due to spur on the posterior side. Normally t here are only three carpels but in D. ajacis only one carpel is found. In the sub-family Paeonieae the flowers have become hemicyclic or acylic. In Pae onia the flower is pentamerous with 5 sepals and petals each, indefinite stamens but only two to three carpels. Definite nectariferous structures cannot be foun d but a disc or ring like swelling around the carpels may develop. Economic importance Condiment The seeds of Nigella sativa (H. Kalongi) are used as spice in pickles. Medicinal Aconitum hererophyllum and A. napellus yield a number of alkaloids spe cially aconitin., This is used in acute and inflammatory diseases, The roots of Thalictrum yields "mamira", which is used in opthalamia. Anemone pulsatilla is m ostly used in feminine diseases and in gastric derangements. Pulsatilla obtained from Anemone pulsatilla is a good medicine for menstrual disorder. Cimicifuga racemosa gives the black Snake root containing resins. This has been recommended for treatment of cholera and nervous pain. Helleborus niger and H. f oetida produce Dyp 1-11 auk-s')e)s caylazu2413) Fig. 24.4. Ranunculaceae 185 glycosides useful as purgatives in veterinary practices. Delphinium staphisagria is, used as antiparasitic ointment. Ornamental Some of the plants are cultivated in gardens for their beautiful flow ers viz., Ranunculus, Delphinium, Naravelia, Clematis, Nigella and Caltha. C Kr. ,6 Melo I (9 Affinities of the family The family Ranunculaceae is one of the most primitive of the dicotyledons. Hutch inson, Bentham and Hooker have placed the family in the class of very early dico tyledons. Engler, Rendle and others put the family under the archichlamydeae. The family is in its close relationship with the monocotyledons due to the forma tion of rhizomes, development of sheath-a-4166f base and copious endosperm. The family can be linked withRosaceae on account of free and numerous stamens an d carpels. The dimorphic leaves and hypogynous followers of Ranunculaceae trace the relationship with the family Nymphaeaceae. Primitive characters 1. Perennial habit (Aconitum, Clematis). 2. Presence of tree or shrub (Paeonia). 3. Presence of stipules (Thalictrum). 4. Leaves with reticulate venation. 5. Flowers hermaphrodite. 6. Actinomorphic symmetry. 7. Large number of petals. 8. Calyx and corolla free. 9. Gynoecium polycarpellary and apocarpous. 10. Ovule anatropous. Advanced characters 1. Plants are generally herbs.

2. Leaves compound (Clematis). 3. Stipules are absent except in Thalictrum. 4. Zygomorphic symmetry (Delphinium). 5. Unisexual flowers (Thalictrum). 6. Petals fused and forming spur (Delphinium). 7. Gynoecium syncarpous (Nigella). Common plants of the family 1. Aconitum Herb with medicinal properties due to the presence of several alkalo ids. 2. Clematis Climber. 3. Delphinium Ornamentanal, cultivated in gardens. 4. Ranunculus Annual or perennial herb. 5. Nigella The seeds of Nigella sativa are used as a condiment. 6. Thalictrum Perennial herb with medicinal properties. Division of the family and chief genera Engler and Prantl have divided the family into three tribes viz., Helleboreae, A nemoneae and Paeoniaeae. According to Hutchinson (1923) the family Ranunculaceae is divided into two sub-families. 186 Plant Taxonomy I. Helleboroideae Carpels with more than one ovule; fruit follicle or berry. Tribe (i) Helleboreae Flowers actinomorphic. Nigella, Helleborus, Aquilegia. Tribe (ii) Delphinieae Flowers zygomorphic. Delphinium, Aconitum. H. Ranunculoideae Carpels with one ovule; fruit a bunch of dry achenes very rare ly berry. Tribe (iii) Ranunculeae Leaves alternate, sepal imbricate, flowers are not subte nded by involucral leaves, sepal mostly caducous. Ranunculus, Thalictrum. Tribe (iv) Anemoneae Leaves alternate, sepals mostly coloured and persistent, fl owers subtended by an involucral leaves. Anemone, Hepatica. Tribe (v) Clematideae Leaves opposite, sepals valvate or imbricate; corolla abse nt or represented by staminodes. Clematis, Clematopsis, Naravelia. Important types 1. Ranunculus scleratus, Linn. (Fig. 24.5). ,SEPAL; ,PETALS STAMENS THALAMUS CARPEL L we OVARY - SmEarsEE CM.* D..T S. iheimmIE lq stranEemens el carpals NEC1 ARV' E Flom/ dialms Fig. 24.5. Ranunculus scleratus. 25 MAGNOLIACEAE Classification : Bentham & Hooker (1862-1893) Engler & Prantl (1931) Hutchinson (1959) Dicotyledons Dicotyledoneae Dicotyledones Polypetalae Archechlamydeae Lignosae Thalamiflorae Ranales Magnoliales Ranales Magnoliaceae Magnoliaceae Magnoliaceae Diagnostic characters Trees and shrubs; two ranked stipulate leaves, stipules enclose young buds; flow

ers heknaphrodite, actinomorphic, large; perianth usually trimerous, whorled or spiral; stamens and carpels numerous; apocarpous, spirally arranged on elongated axis, fruit an etario of follicles or berries, sometimes samara. Distribution Magnoliaceae or the Magnolia family embraces 10 genera and about 100 species. Th e members of this family belong to the temperate regions of northern hemisphere, with centres of distribution in eastern Asia, Malaysia, eastern North America, West Indies, Brazil and North-east and south east India. Vegetative characters Habit Trees or shrubs sometimes climbing. Oil sacks present in stem and leaves. Root Tap branched. Stem Erect, aerial, woody, branched. Leaves Alternate, simple, entire, commonly ever-green, coriaceous, stipules larg e (Magnolia) covering young leaves. Floral characters Inflorescence Solitary terminal or axillary. Flower Largest-and sometimes, 25 cm in diameter (Magnolia, Fraseri), complete, r egular, actinomorphic, unisexual (Drimys), usually bisexual, hypogynous, aromati c. Floral axis (torus) long to long convex. Perianth Nine to many, free, all alike and petaloid or the three outer ones gree n (Liriodendron), arranged in whorls of three, imbricate and cyclic (Magnolia an d Michelia) or acyclic (spiral) arranged on an elongated or semi-elongated conve x torus, free, inferior. Androecium Stamens many, free, often spirally arranged in a beautiful series, fi laments short or absent, anther lobes linear, with a prolonged connective. (189) 190 Plant Taxonomy Gynoecium Carpels numerous, free, superior, arranged spirally on a cone-shaped e longated thalamus (gynophore), rarely carpels are fused, e.g., Zygogynum, placentation ma rginal. Fruit An aggregate of berries or follicles, sometimes, a samara as in Liriodendr on. Seed Large, with abundant oily endosperm, and bright or orange testa which makes them highly decorative. Pollination Entomophilous due to large and scented flowers. Floral formula G P9 or a Au Ga Economic importance Medicinal The root bark and dried roots of Michelia champaca are used as purgati ve, while the flowers and fruits are used as carminative and in certain renal tr oubles and venerial diseases like gonorrhoea. The bark of Drimys winteri and Illicium uerum is useful in medicine. Timber The wood of Michelia excelsa is an excellent commercial timber known as " white wood". M. acuminata, Manglietia hookeri, Michelia baillonii, M. dottsopa a nd Pachyla rnax pleiocarpa produce valuable timber used for mill work, furniture , musical instruments, toys etc. Ornamentals The species of Magnolia and Michelia are of surprising beauty becaus e of their conspicuous white and yellow-tinted, fragrant flowers. Flowers of Mic helia champaca are used by womens to ornament their hair and also offered in tem ples. Michelia champaca yields `champaca oil' from the flowers, camphor from the wood and scented water from the leaves. , , Affinities of the Family Magnoliaceae is akin to Annonaceae, Engler, Rendle and other botanists put the M agnoliaceae under the Ranales. But Hutchinson placed the family within the Magno liales and considered it as the most primitive among the dicotyledons. The primi tiveness of this family is shown by the spiral arrangement of stamens and carpel

s and apocarpous pistil characters also shared with the Ranunculaceae. Smith, ho wever, cast some doubt on the assumption of the primitive nature of the Magnolia ceae and thought that the family is relatively specialised from vegetative and r eproductive aspects. From the stand-point of construction and organisations of flowers, the Magnoliac eae is allied tc the Annonaceae. The two families, however, can be set apart fro m each other by the following features : Magnoliaceae Annonaceae Stipules present. Stipules absent. Flowers large, showy. Flower small not showy. Corolla imbricate Corolla valvate. Anthers introtse. Anthers extrorse. Endosperm watery-fleshy, not ruminate. Endosperm ruminate or marbled. The Magnoliaceae has a great deal of affinity with the gymnosperms. The most str iking thing about the wood of Drimys is the presence of tracheids with bordered pits in the xylem. This coniferous nature of the wood is "discounted by the fact that the structure of the bast is characteristically dicotyledonous, having sie ve-tubes with companion cells". Furthermore, the receptacle of Magnolia flower b ears numerous spirally arranged free sporophylls. This can be compared with the strobilus of Bennettites, consisting of a central conical axis covered with mega sporophylls. This has led to the suggesticn that the order Bennettitales may be ancestral to the modern angiosperms initiated by the Magnoliaceae. Magnoliaceae 191 Common plants of the family. 1. Liriodendron tulipifera L., A large tree with handsome flowers, wood as a tim ber. 2. Michelia champaca L. "Champa", A garden plant cultivated for its flowers whic h are sweet scented; used in perfumery. 3. Michelia grandiflora L. 'Bari Champa' or 'Him Champa" A small tree with fragr ant flowers. 4. Magnolia acuminata Cucumber tree. Largest tree of Magnolias. Its green fruits resemble a cucumber; wood light, soft and durable. Division of the family and chief genera On the basis of the habit, presence or absence of stipules and degree of develop ment of thalamus, the Magnoliaceae is separated into three subfamilies : Subfamily I. Magnolioideae. Leaves with sheathing stipules. Flowers bisexual. Th alamus long. Example : Magnolia. Subfamily II. Illiciodeae. Leaves with no stipules. Flowers bisexual or unisexua l. Thalamus short. Example : Illicium. Subfamily III. Schisandroideae. Climbing shrub. Leaves with no stipules. Flowers unisexual. Example : Schisandra. Important type 1. Michelia champaca (Champa) (Fig. 25.1). Habit and Habitat A tall, graceful, ever-green tree with dark-grey bark; cultiva ted, flowering mainly during April June. Root Tap root; branched. Stem Aerial; erect; branched; woody; solid. Leaf Ramal; alternate; stipulate, stipules convolute; petiolate; simple; ovate o r oblong-lanceolate; entire with acute apex; coriaceous; glabrous; dark green; r eticulate unicostate. Inflorescence Clusters of axillary flowers. Flower Bracteate; pedicellate, pedicel short; ebracteolate; complete; regular; a ctinomorphic; hermaphrodite; hypogynous; spirocyclic; pale-yellow, very fragrant , large. Perianth Usually nine tepals, 3 whorls of 3 each; either all the 3 whorls petall oid or sometimes the outermost of 3 tepals becomes sepaloid than calyx and corol la and described as under. Calyx Three; free; greenish; valvate, inferior. Corolla Six, in two whorls of three each; choripetalous (free), those of the out er whorls valvate whereas those of the inner whorl narrow and twisted; pale yell ow; fragment; inferior. Androecium Indefinite; polyandrous; spirally arranged; distinguished into filame

nt, connective and anther lobes; dithecous, basifixed; dehiscence longitudinal, extrorse. Gynoecium Polycarpellary; apocarpous; superior; spirally arranged on a cone like and elongated thalamus; each carpel with a swollen ovary, curved style and beak ed and simple stigma; unilocular; placentation marginal; many ovules in each car pel. Fruit An etaerio of follicles, dark coloured with white specks. Seed Large; endospermic, endosperm copious; embryo small. Pollination Entomophilous. Floral Formula Br ED V P3+3+3 A,, Ga or Br V K3 C3+3 A G_ 34 ) CARYOPHYLLACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Polypetalae Archichlamydeae Herbaceae Thalamiflorae Centrospermae Caryophyllales Caryophyllinae Caryophyllaceae Caryophyllaceae Caryophyllaceae Diagnostic features Leaves opposite, decussate on swollen nodes, inflorescence dichasial cyme, flowe r actinomorphic, hypogynous, corolla caryophyllaceous, stamens twice the number of petals, obdiplostemonous, 2-5 carpels, syncarpous, ovary superior, unilocular with free central placentation. Distribution The family is commonly called pink or chickweed family. The plants of this family are mostly distributed in cold climates of temperate r egions specially north temperate regions. According to Rendle it consists of abo bt 80 genera and 2000 species out of which 107 species are found in India. The m ain centre of distribution is Mediterranean region. Vegetative characters Habit Mostly annual or perennial herbs and a few under shrub (Acanthophyllum). Root Branched tap root. Stem Erect or partially decumbent, swollen nodes, glaucous. Leaves Simple, sessile, opposite rarely alternate, exstipulate, if present scari ous (thin, dry and membranous) margin entire or serrate, shape linear to lanceolate, leaf base connate, perfoliate or sheathing. Floral characters Inflorescence Dichasial cyme, sometimes it ends in monochasial cyme rarely solit ary. (Githago, Arenaria, Drymaria). Flower Bracteate, pedicellate, actinomorphic, hermaphrodite rarely unisexual, co mplete, pentamerous rarely tetramerous; caryophyllaceous and hypogynous. Calyx Sepals 5, polysepalous or gamosepalous, persistent, imbricate aestivation. Corolla Petals 5, polypetalous, clawed, notched or bifid, caryophyllaceous, imbr icate aestivation. Androecium Stamens 8 to 10, polyandrous, obdiplostemonous; anthers dithecous, de hiscing longitudinally sometimes staminodes are present. (233) r 234 Plant Taxonomy Gynoecium Typically 5 carpels, may be 2 to 5, syncarpous; ovary superior, unilcc ular; free central placentation, ovules one to many on each placentum, style fre e and corresponds to the number of carpels; so also the stigmas. Fruit Capsule rarely berry (Cucubalus). Seed Small, many and albuminous. Floral formula m # K5 C5 A5+5 or 4+4 G (5 or 3) Floral variation Two types of flower patterns are known In the Alsinoideae a simple type of flowe

r structure is found while in the Silenoideae elaboration has taken place. In the Alsinoideae the most complete floral type can be represented by the formu la; K5, C5, A5 + 5, G (51. Commonly the number of carpels may be reduced to 3 so that the formula becomes K5, C5, A5+5, Gam. Commonly the stamens are obdiploste monous but by reduction it may be 5 or only 3. When it is fiv5the stamens are an tipetalous. Petals may also be reduced e.g. Polycarpaea where the petals are abs ent. In Lyallia the petals are absent and the stamens are also reduced to only 2 . The gynoecium is typically tricarpellary syncarpous, free central placentation ; number of styles equal to the number of carpels. The fruit is a capsule. In the Silenoideae the calyx is tubular and the sepals are gamosepalous. The pet als are distinguished into a stalk and a limb and at the junction of the two the re is often a ligular outgrowth (Dianthus and Lychnis). Between the sepals and p etals there is often an internode so that the rest of the floral whorls arise at a higher level. The floral formula is K (5), C5, A5+5, G (5) or (3). The stamen s are again obdiplostemonous. Carpels commonly 3 but may be 2 in Dianthus and Sa ponaria. Placentation is typically free central. Fruit capsular but baccate in C ucubalus. The pollination mechanism in Silenoideae is more elaborate. Because th e flowers are tubular only long tongued insects can reach the nectar. Cross poll ination is helped by protandry. Economic importance The family is not of great economic importance. Medicinal Dianthus caryophyllus is considered as cardiac tonic, nervine and anti septic. The dried plants of Dianthus chinensis are used as diuretic and abortifacient. Ornamental Some plants are cultivated for their flowers viz. Gypsophylla, Sapona ria, Cerastium, Dianthus etc. The roots and leaves of Saponaria officinalis contain saponin which yield lather with water. Affinities of the family There are diverse opinions among the systematists regarding the origin of the fa mily Caryophyllaceae. Eichler (1875) considered that the family Caryophyllaceae has originated from Phytolacaceae by the modification of the outer stamen whorls to petals and of outer carpel whorl to stamens. This view has also been support ed to some extent by Pax, Wettstein and Rendle. Wernham (1911) considered it to have evolved from Ranalian ancestors. This view has been accepted by Bessey and Hutchinson. Dickson (1930) suggested that the family Caryophyllaceae was derived from Gerani aceae. Primitive characters 1. Leaves simple and stipulate in some species (Dianthus). 2. Flowers hermaphrodite, hypogynous and actinomorphic. 3. Calyx and corolla free. 4. Stamens polyandrous. 5. Seeds endospermic 235 O Fig. 34.1. Floral diagram of Dianthus. Caryophyllaceae Advanced characters 1. Plant are generally annual herbs. 2. Inflorescence dichasial cyme. 3. Petals absent in some species (Sagina apetala). 4. Gynoecium syncarpous and reduction of carpel number from five (Viscaria, Gyps ophila) to two (Dianthus). 5. Fruit simple. Common plants of the family 1. Dianthus chinensis This plant is called Flower of Jupiter or Carnation for it s beauty and fragrance. 2. Saponaria Plant with abundant soapy juice.

3. Stellaria media A herbaceous weed, commonly called Chickweed. 4. Spergula arvensis Common weed in or near the wheat and barley fields. 5. Lychnis alba Flowers unisexual and found in hills. Division of the family and chief genera Porter (1959) has divided the family into three sub-families : 1. Paronychioideae Sepals distinct; or united petals lacking, fruit a utricle. P aronychia, Scleranthus. 2. Alsinoideae Sepals distinct; petals present, fruit a capsule. Spergula, Stell aria, Arenaria. 3. Silenoideae (Caryophylloideae) Sepals united or gamosepalous; petals present; fruit a capsule. Silene, Dianthus, Lychnis. Important types 1. Dianthus caryophyllatus, Linn. (Fig. 34.1 and 34.2). Habit An annual ornamental herb. Root Tap, branched and annual. Stem Herbaceous, erect, branched, cylindrical, fistular, glabrous and green. Leaves Opposite, simple, sessile, exstipulate, linear, entire margin, acute, uni costate reticulate venation. Inflorescence Axillary or terminal dichasial cyme. Flower Pedicellate, bracteate, complete, hermaphrodite, actinomorphic, pentamero us, hypogynous. Calyx Sepals 5, gamosepalous, tubular, imbricate aestivation, inferior. Corolla Petals 5, polypetalous, clawed, twisted aestivation, inferior. Androecium Stamens 10, in two whorls of 5 each, polyandrous, obdiplostemonous, f ilament long, dorsifixed, dithecous, introrse. Gynoecium Bicarpellary, syncarpous, ovary superior, unilocular, free central pla centation, style 2; stigma 2. Floral formula Br. V K(5) C5 A5 + 5 G f_21. 236 Plant Taxonomy El-Entire Rower A-Flowering twig D-Gynoecasm E-T. S. ovary Fig. 34.2. Dianthus. 2. Stellaria media, Cryill. (Fig. 34.3) Habit An annual herb. Root Tap, branched and annual. Stem Erect, herbaceous, cylindrical, branched, minute hairy, green, branching is dichotomous. Leaves Ramal, cauline, opposite, simple, exstipulate, sessile, lamina ovate usua lly cordate, margin entire, acute apex, unicostate reticulate venation. Inflores cence Dichasial cyme. \ Ilk THALAMUS PEDICEL L.S. FLOWER A FLOWERING TWIG Caryophvllaceae 237 STIGMA STAMEN 11 OVULE OVARY SEPAL FLORAL DIAGRAM Fig. 34.3. Stellaria media L. (The Chickweed)

Flower Bracteate, pedicellate, complete, hermaphrodite, actinomorphic, pentamero us, hypogynous. Calyx Sepals 5, polysepalous, hairy, glandular, quincuncial aestivation, inferio r. Corolla Petals 5, polypetalous each petals is deeply lobed white, imbricate aest ivation, bifid, white. unilocular, free central placentation, Plant Taxonomy polyandrous, basifixed, dithecous, Androecium Stamens 10; 5+5, obdiplostemonous, introrse. Gy-noecium Tricarpellary, syncarpous, ovary superior, style three, stigma 3. Floral formula Br. ED V K5 C5 A5+5 G(3). 3. Spergula arvensis, Linn. (Fig. 34.4) Habit An annual herb. Root Tap, branched and annual. 238 STIGMA FLOWER STAMEN PETAL SEPAL OVARY PEDICEL L.S. FLOWER ROOTS FLORAL DIAGRAM Fig. 34.4. Spergula arvensis Linn. AN ENTIRE PLANT , Caryophyllaceae 239 Stem Herbaceous, erect, cylindrical, solid, branched, pubescent, green. Leaves Opposite, sessile, stipulate, with minute and scarious stipules, simple, green, needle like, acicular, unicostate reticulate venation. Inflorescence Dichasial cyme. Flower Pedicellate, bracteate, complete, hermaphrodite, actinomorphic, pentamero us, hypogynous, greenish white. Calyx Sepals 5, polysepalous, green, smooth, quincuncial aestivation, inferior. Corolla Petals 5, polypetalous, caryophyllaceous, white, imbricate aestivation, inferior. Androecium Stamens 10, arranged in two whorls of 5 each, obdiplostamonous, polyandrous, dithecous, dorsifixed, introrse. Gynoecium Penta or tricarpellary, syncarpous, ovary superior, unilocular, free-c entral placentation, style 5 or 3, stigma 5 or 3. Floral formula Br. 0 V K5 C5 A5+5 G (5 or 3). r. Literature 1. Pal, N. 1952 : A contribution to the life history of Stellaria media L. and P olycarpon loefingiae B. & H. Proc. Nat. Ins. Sci. Ind. 18 : 363-378. 2. Pax, F. and Hoffman, K. 1934: Caryophyllaceae. Die. Nat. Pflanzenfamilien ed. 2, Band 16C, 275-364. 3. Rendle, A.B. 1959: The Classification of Flowering Plants Vol. II, 93-123. 1 4 40 MELIACAE Classification :

Bentham & Hooker Engler & Prantl Hutchinson Polypetalae Archichlamydeae Lignosae Disciflorae Geraniales Meliales Geraniales Meliaceae Meliaceae Meliaceae Diagnostic characters Plants woody trees, leaves pinnately compound exstipulate; leaflets asymmetrical , margin serrate, inflorescence cymose panicles, flowers actinomorphic, hermaphr odite, calyx and corolla sometimes united, stamens 8 to 10, monadelphous, obdipl ostemonous : annular necticiferous disc between petals and stamens, gynoecium pe ntacarpellary, syncarpous, superior, fewer or multilocular with 1-2 rarely more ovules in each locule; single style; fruits various capsular or drupaceous. Distribution This family is also called Mahogany family. It includes 50 genera and 1400 speci es according to Willis. In India it is represented by 20 genera and 70 species. It is widely distributed in tropics of both the hemispheres. Vegetative characters Habit Mostly woody trees rarely shrubs, often with a characteristic smell. Roots Much branched tap root. Stem Woody much branched, erect, solid. Leaves Alternate, exstipulate, pinnately compound rarely simple, without transpa rent dots or glandular dots, serrate margin. Floral characters Inflorescence Cymose panicles often axillary. Flower Pedicellate, bracteate, complete, hermaphrodite or polygamous, actinomorp hic, hypogynous, pentamerous, with a necticiferous disc. Calyx Sepals 4-5, small poly or gamoseplaous (connate at the base), imbricate rar ely valvate aestivation, inferior. Corolla Petals 4-5 rarely 3 to 8, polypetalous rarely connate at the base or adn ate to the staminal tube, imbricate or twisted aestivation, inferior. (267) 268 Plant Taxonomy Androecium Stamens 8-10, inserted outside the base of hypogynous disc, filaments united to form a columnar tube (monadelphous; Cedrela), anthers bithecous, erec t, introrse, longitudinal dehiscence, necticiferous disc present or absent. Gynoecium Carpels 2-5, syncarpous, superior, 2-5 Jocular, axile placentation, wi th 1-2 ovules in each loculus, single style, stigma capitate, discoid or lobed. Fruit Berry, capsule or drupe. Seed Winged, albuminous or exalbuminous. Pollination Entomophilous, due to the presence of nectaries. Floral formula $, i K4_5 or (4-5) C4_5 A(8-io) G Economic importance The family is not of great economic importance. Oil The seeds of Melia azadirachta (H. Neem) yield the `margossa' oil of commerc e. The oil is used in soap industry and medicinally in rheumatism and skin disea ses. The oil of Carapa guianensis is used as an illuminant. The oil of Cedrela i .e. cedar oil is used in microscopy. Medicines Almost every part of Melia azadirachta possesses some medicinal proper ties. The bark is a bitter tonic, astringent and antiperiodic. The bark, root ba rk and young fruits are used as a tonic antiperiodic and alterative. Leaves are used as poultice and applied to boils, the twigs as tooth brushes. Decoction of leaves is antiseptic and used to wash ulcers and eczema. The oil is used in rheu matism and skin diseases. Dry flowers are used as a tonic and stomachic. It is b lood purifier. The bark of Cedrela tonna is used as astringent, tonic and antiperiodic in chron ic dysentory. Timber The wood of Cedrela toona (H. Tun), is used for furniture, carving and also for cigar boxes. The Swietenia mahoganii, Khaya senegalensis pr oduce cabinet wood. Ornamentals Some of the plants viz., Melia, Amoora, Swietenia are grown in garde

ns. Primitive characters 1. Plants are mostly trees or shrubs. 2. Leaves alternate. 3. Presence of secretary cells. 4. Wood rays heterogenous. 5. Flowers hermaphrodite and hypogynous. 6. Flowers actinomorphic. 7. Calyx and corolla free. 8. Stamens free. 9. Pollination by insects. Advance characters 1. Leaves mostly compound. 2. Leaves exstipulate. 3. Unisexual flowers in Amoora, Lansium. 4. Flowers are small in size. 5. Inflorescence cymose or racemose. 6. Reduction in number of sepals and petals e.g. Amoora (3 sepals and 3 petals). 7. Androecium is monadelphous. 8. Reduction in number of stamens (Melia 10-12, in Heynea 8-10, Cedrela 4-6, Amo ora 3). 9. Gynoecium syncarpous. Meliaceae 269 10. Placentation axile. 11. Fruit simple. 12. Seed non-endospermic in many genera. The family is regarded as one of the advance among Geraniales. Affinities of the family Hallier placed the family in his order Terebinthales and considered that it is d erived from Rutaceae. Both the families resembled each other in many characters viz., plants are mostly trees and shrubs, leaves exstipulate and compound, small scented flowers. Bessey, Engler and Prantl and Bentham and Hooker retained fami ly Meliaceae in Geraniales. Rendle has placed it as the last family of Rutales. Cronquist (1968) included this family under order Sapindales of sub-class Roside ae and class Magnoliatae. Takhtajan (1969) included Rutaceae and Meliaceae in or der Rutales. Hutchinson included this family as the only family in his order Mel iales. Common plants of the family contains margosic acid. 1. Melia azadirachta (H. Neem, Margosa tree) CIT 5 6 7 8 Fig. 40.1. Melia azadirachta Linn. 1. Flowering twig and Leaf. 2. Flower open to show. 3 & 8. Flower. 4, 5, & 6. Gy noecium. 7. Fruits. 270 Plant Tormataser 2. Cedrela toona (H. Tun., Indian Mahogany) cultivated for wood; flowers yield d ire "Vas anti". 3. Chloroxylon (stain wood) valuable timber tree. 4. Carapa obovata Small tree of Sundarbans. 5. Swietenia mahogani beautiful tree with shining wood. Division of the family and chief genera Engler divided the family into three sub-families as below I. Cedreioideae stamens free, from capsule and seed winged. Cedrela. II. Swietenioideae Stamens monadelphous, fruit capsule, seed winged, Swietenia, Khaya. III. Melioideae Stamens monadelphous, fruit capsule, berry or drupe, seeds not w inged. Melia. Important type 1. Melia azadirachta (Fig. 40.1)

Habit A perennial tree. Root Tap, well branched, with deep feeders. Stem Woody, erect, aerial, cylindrical, solid, branched, loWer stem covered by b ark. Leaves Alternate, exstipulate, compound imparipinnate. Leaflet Petiolate, ovate, serrate margin, acute apex, unicostate reticulate. Inflorescence Axillary panicle. Flower Pedicellate, complete, hermaphrodite, hyp ogynous, actinomorphic, pentamerous, white, scented. Calyx Sepals 5, gamosepalou s, green, valvate, inferior. Corolla Petals 5, polypetalous, imbricate, white, i nferior. Androecium Stamens 10, monadelphous, forming a tube, with anthers at the top, on the inside, opposite the teeth of the staminal tube, basifixed, bithecous, inferior. Fig. 40.2. Floral digram Gynoecium Tricarpellary to multicarpellary, syncarpous, superior, number of locu les is equal to number of carpels, one or two ovules per locules, axile placenta tion; style long, stigma capitate, nectar disc below the ovary. Fruit Drupe. Seed Non-endospermic. Floral formula Ic5), C5 of Melia. A(10) G(3 - a) 54 CACTACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Polypetalae Archichlamydeae Lignosae Calyciflorae Opuntiales Cactales Fucoidales Cactaceae Cactaceae Cactaceae Diagnostic characters Plant fleshy with tufts of thorns; stems flattened, jointed, succulent, apparent ly leafless; flowers large, usually solitary, hermaphrodite, epigynous; tepals i ndefinite with gradual transition from sepals to petals, stamens indefinite, Car pels syncarpous, ovary inferior unilocular, parietal placentae; fruit berry. Distribution Cactaceae consists of about 87 genera and 2,000 species. Members are distributed mainly in semi-desert and hot-desert regions of tropic and sub-tropic of Americ a, Mexico. Vegetative characters Habit The plants are mostly perennial xerophytic shrubs rarely trees, sometimes epiphytes. Root Tap-root, raher small and shallow. Stem Fleshy, of various shapes, rarely bearing normal leaves, usually provided w ith spines or barbed bristles (glochidia) which are generally considered to be m odified leaves; spines and glochidia develop from small cushion-like structures called areoles; in several genera, the stem assumes leaf-like flattened from phy lloclades. Leaf Mostly leafless, leaves reduced to spines, in Opuntia, true leaves are very small, appear early and fall very soon leaving the phylloclades to do the assimi lation work, spines or tuft of bristles developed from areoles. Floral characters Inflorescence Solitary rarely paniculate. Flower Exceptionally beautiful, large, showy but only very rare (Mammillaria, Op untia, Cereus, Echinocereus, Schiumbergera); sessile, mostly bisexual and actino morphic or nearly zygomorphic, epigynous. Periantl Tepals numerous showing a gradual transition from sepals to petals; spi rally arranged; sepals often petalloid; petals epigynous and present in several series; often fuse to form a perianth tube or hypanthium. (335) 1

336 Plant Taxonomy Androecium Stamens numerous, in several series or groups, epipetalous or inserte d at the base of petals; anthers 2-celled, basifixed, introrse; longitudinally d ehiscent. Gynoecium Two or many carpels, syncarpous; ovary inferior, unilocular, numerous anatropous ovules; parietal placentation, very rarely basal placentation; style 1 and simple; stigmas often as many as the carpels. Fruit Many seeded berry. Seed Scanty or no endosperm. Pollination Entomophilous. Floral formula AD P - - a or (a) Aci G(3 - a) Economic importance Food Fruits ofLophophora williamsii, Nopalea cochenillifera and several species of Opuntia are edible. Fodder In emergency i.e. fodder famine prevails the spines of Opuntia are burnt, by scorching and then used as fodder. Dye A scarlet dye cochineal is prepared from Cochineal insects that feed upon Op untia tuna, Napalea cochenillifera. Medicinal Some species of Opuntia and Cercus have medicinal properties. Ornamentals In rock gardens and unusual environment the use of Cacti Mammillaria , Cereus, Echinocactus, Rebutia, Opuntia etc has become extensive. Affinities of the family Cactaceae is of special interest to botanists for its combination of a primitive , unspecialized flower with highly advanced vegetative organs. To the taxonomist it presents great problems, being apparently still in a state of active evoluti on. As regards the position and relationship of Cactaceae, many views have been put forward. Engler retained it in the Opuntiales near the Myrtiflorae and admitted an alliance with the Aizoaceae. Bessey regarded the Cactales to be related to th e Cucurbitales. Wettstein included the Cactaceae in the Centrospermae. Bentham a nd Hooker placed the family in the Ficoidales before the Umbellales and after th e Passiflorales. Rendle opined that "the affinity of the family is very doubtful "; he set it up within the Opuntiales, lying between the Myrtiflorae and Umbelli florae. In Hutchinson's arrangement, the Cactaceae appeared under the Cactales w hich was again put in between the Cucurbitales and Theales. On the basis of epid ermal structures, floral plan and serological data, Mez considered the family as linked to the Loasaceae and the Parietales complex. Buxbaum (1944) derived the Cactaceae from the Phytolaccaceae and took it to exem plify parallel evolution with the Aizoaceae. Evidence pooled from anatomy, embry ology and floral morphology led Maheshwari (1945) and others to suggest that thi s family should be assigned to a position within or near the Centrospermae. Gund ersen (1950) treated the family in the Cactales near the Ranales. Mitra (1964) s upported Gundersen's contention of a ranalian relationship as expressed by Anema rrhena type of seedling, spirocyclic flowers (simulating Euryale), transition fr om sepals to petals and fasciculate androecium. Common plants of the family 1. Opuntia elatior is an example of Cladode. 2. Opuntia dillenii Prickly pear H. Chittar thor. 3. Cactus indicus grown in hills as a hedge and sometimes found in plains. 4. Cereus peruvianus Phylloclades 5-angular. Cactaceae 5. Mammillaria exceptionally beautiful large and showy. Division of the family and chief genera The Cactaceae is divided into three subfamilies : Sub-family I. Cereoideae Succulents with leaves reduced to minute scales and are oles Mammillaria, etc. without glochidia. Flowers funnelform of salverform. Examples Cereus, Echinocact us,

B 337 Fig. 54.1. Selenicereus stem/with Britt. et Rose (Syn. Cereus flagelliformis). A. Part of the plant bearing flower on the floral diagram. B. L.S. of flowe r showing fasciculate stamens. C. Fruit in vertical section showing numerous seeds embedded in pulp. D. Seed in L.S. showing the embryo. E. Floral diagram. A E 338 Plant Taxonomy Sub-family II. Opuntioideae Succulents with usually flattened jointed stems, sma ll cylindrical to subulate caducous leaves and glochidiate areoles. Flowers rota te. Examples Nopalea, Opuntia, etc. Sub-family III. Pereskioideae Plants with flat leaves. Flowers in panicles. Exam ple Pereskia. Important type 1. Selenicercus flagelliformis (Fig. 54.1) 2. Opuntia dillenii (Prickly-pear; Vern. Chittar thor) (Fig. 54.2) Habit A spiny, gregarious, perennial shrub about 1.5 metres high, xerophytic. Root Tap-root. Stem Flattened into phylloclades, jointed, joints, 30-40 cm. by 15-20 cm. broadl y obovate, undulate, not very thick, bluish to bluish-green, covered with little groups of spines. Leaf Reduced, 2in. long, palegreen, caducous, represented by clusters of spines as areoles, each tuft of 4-6 prickles, the largest (glochidia) very stout, subulate, firm and sharp, 1 to 11 in. long usually somewhat curved, yellowish. Inflorescence Solitary. Flowers Bright yellow tinged with orange, 1-3 in. across, sessile regular, bisex ual, acyclic, epigynous. Perianth Indefinite, not sharply distinguishable from one another, the outer sep aloid showing gradual transition into petals in several series, spirally arrange d, spreading imbricate, the innermost largest, obovate rounded, mucronate; the o uter ovate, acute or rounded with membranous margins, superior. Androecium Stamens indefinite, free, of unequal length inserted to the base of t he inner perianth (epiphyllous) on the margin of the calyx-tube, superior, dorsi fixed bithecous, introrse. Fig. 54.2. Opuntia dillenii Haw. A. Part of the plant with flower, fruit and flo wer bud on stem (phylloclade) beset with spines. B. Flower in L.S. showing fasciculate stamens and epigyny. C. Fruit in L.S., the seed are embedded in pulp. D. Vertical section of seed showing embryo. E. Floral diagram. Cactaceae 339 Gynoecium Carpels many, syncarpous; ovary inferior, unilocular, with 5-8, many o vuled parietal placentae, style single; stigmas 5-8, erect. Fruit Berry, pyriform truncate, depressed at the apex, often angular, warty when not fully ripened, deep reddish-purple when ripe, edible. Seed Non-endospermic. Floral formula m V P,,, A,. G(,,c). SO 56 RUBIACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Gamopetalae Sympetalae Lignosae Inferae Rubiales Rubiales Rubiales Rubiaceae Rubiaceae Rubiaceae Diagnostic features Trees or herbs; leaves alternate or opposite; stipules interpetiolar or intrapet

iolar; inflorescence cymose; flowers tetra or pentamerous, hermaphrodite, actino morphic, epigynous, corolla, gamopetalous; stamens 4-5; epipetalous, introrse, d ithecous; ovary inferior, bilocular with one or many ovules in each loculus; fru it capsule or berry. Distribution It is commonly known as Madder or Coffee family. It includes 6000 species and 50 0 genera. In India it is represented by 551 species. The members of this family are distributed in tropics, sub-tropics and temperate regions. Vegetative characters Habit Mostly shrubs (Gardenia, Ixora, Mussaenda, Hamelia); trees (Morinda, Adina ) and a few herbs (Galium, Rubia). Root Much branched tap root system. Stem Erect, herbaceous or woody or twinning (Manettia), climbing by hooks (Uncar ia), branched, cylindrical or angular, hairy or smooth. Leaves Cauline, ramal, opposite or verticillate, simple, entire or toothed, stip ulate, stipules bristle like (Pentas) and leafy (Galium, Rubia), stipules mostly interpetiolar or sometimes intrapetiolar; unicostate reticulate venation. Floral characters Inflorescence Solitary (Gardenia) usually cymose or globose head (Adina), or pan icled cyme; may be axillary (Coffea arabica) or terminal cyme (Mussaenda glabra) . Flower Actinomorphic, rarely zygomorphic (some what bilabiate as in Henriquezia) , mostly hermaphrodite, rarely unisexual, epigynous, pedicellate or sessile (Gre enia, Randia), bracteate or ebracteate, complete, tetra or pentamerous, cyclic, variously joloured. Calyx Sepals 4 or 5, gamosepalous, superior, sometimes one sepal modified into c oloured bract like structure (Mussaenda), valyate. (346) Rubiaceae 347 Corolla Petals 4 or 5, gamopetalous, lobed, generally funnel shaped (Asperula), tubular (Ixora), valvate to twisted or imbricate, superior. Androecium Stamens 4 or 5, rarely many (Gardenia), epipetalous, alternipetalous, inserted near the mouth of corolla tube, stamens dithecous, introrse, dehiscing longitudinally, superior. Gynoecium Bicarpellary, rarely polycarpellary, syncarpous, inferior rarely half inferior (Synaptanthera) or superior (Paganea), sometimes unilocular (Gardenia) with one to many anatropous ovules in each loculus, axile placentation (parietal placentation in Gardenia), style one sometimes bifid or multifid, stigma simple or bilobed. Fruit Capsular (Anotis), berry (Mussaenda, Hamelia, Ixora). Seed Endospermic, sometimes winged. Pollination Entomophilous; ant pollination is well known. Floral formula $ or oho # K (4-5) C (4-5) A4-5 G (2-5). Economic importance I. Medicinal plants Bark of Cinchona officinalis yields an alkaloid called Quini ne which is the best remedy for malarial fever. TIT rotes of Rubia cordifolia ar e also used as nie-aii-cine. H. Beverage plants The seeds of Coffea arabica, C. liberica and C. robusta are r oasted and ground to give coffee powder. cvnji) &of te.P.-. HI. Ornamental plants Rubia, Hamelia, Gardenia, Ixora, Mussaenda are cultivated in gardens for their beautiful flowers. Affinities of the family The family is closely related to Adoxaceae and Caprifoliaceae in the presence of inferior ovary, actinomorphic and hermaphrodite flowers, type of inflorescence, reduction in the number of carpels and ovules. The Rubiaceae also resembles the Loganiaceae in well developed stipules, glandular trichomes, nuclear endosperm. Rubiaceae differ from Loganiaceae in possessing inferior ovary. Bessey traced the origin of Rubiales from the Umbellales which in turn originate

d from the Rosales. The Rubiales resemble the Umbellales in aggregation of flowe rs, reduction in the number of calyx and corolla lobes, stamens and carpels; in both epigynous flowers with nectar secretory disc are present. This suggests a c lose taxonomic relationship among the Rubiales and the Umbellales. Primitive characters 1. Plants mostly trees and shrubs. 2. Leaves simple and stipulate. 3. Flowers mostly hermaphrodite and actinomorphic. 4. Stamens polyandrous. 5. Ovules anatropous and many in some genera. 6. Seeds endospermic. Advanced characters 1. A few plants are herbs (Rubia, Galium). 2. Leaves opposite or whorled. 3. Flowers epigynous and rarely unisexual, zygomorphic. 4. Calyx and corolla fused. 5. Stamens epipetalous. 6. Carpel number reduced to two. 7. Fruit simple. 348 Plant Taxonomy Common plant of the family 1. Coffea arabica (Coffee) An evergreen shrub, cultivated for seeds. 2. Cinchona officinalis (Quinine) Tree cultivated for medicinal importance. 3. Hamelia A large evergreen shrub with reduced flowers. 4. Rubia A common climbing herb. 5. Gardenia A resinous shrub or tree. 6. Ixora Evergreen shrub with showy flowers. Many species are cultivated in gard ens as ornamentals. 7. Mussaenda A shrub, very conspicuous during flowering. Division of the family and chief genera Verdcourt (1958) divided the family into three sub-families : Sub-family 1. Cinchonoideae Seeds endospermic, trees, shrubs rarely herbs, ovule s indefinite in each loculus on axile placentation. It includes eleven tribes viz., Nucleaceae, Cinchoneae, Rendeletieae, Mussaendea e, Catesbaceae, Gardenieae, Ixoreae, Alberteae, Vanguerieae, Retiniphylleae, Chi ococceae. Sub-family 2. Rubioideae Trees, shrubs or herbs, ovary one to many locular, ovul e single or many, erect or pendulous. It includes 18 tribes. The chief genera ar e Rubia, Hamelia, Morinda. Sub-family 3. Guettardoideae Seeds non-endospermic, or endosperm very little tre es or shrubs; aestivation im,bricate or valvate; ovary two to many locular with one pendulous ovule per loculus. Important types 1. Hamelia patens Habit Evergreen perennial shrub. Root Branched tap root. Stem Erect, woody, cylindrical, solid, brownish-green with nodes and internodes, branching simple, hairy. Leaves Whorled-3 or more leaves at each node, simple, petiolate, lanceolate, ent ire, acute, smooth, unicostate reticulate. Inflorescence Multiparous cyme; each branch helicoid. Flower Ebracteate, pedicellate, complete, hermaphrodite, pentamerous, actinomorp hic, epigynous, reddish. Calyx Sepals 5, gamosepalous, adnate to ovary, greenish brown in colour, superio r, valvate aestivation, persistent. Corolla Petals 5, gamopetalous, tubular, dark red, twisted.

Androecium Stamens 5, epipetalous, alternipetalous, linear, dithecous, anthers v ery long, filament short, basifixed dehiscing by longitudinal slit, introrse. Gynoecium Pentacarpellary, syncarpous, pentalocular, axile placentation, ovary i nferior, many ovules per loculus, style distinct, stigma single, terminal. Fruit Capsular. Floral formula K(5) C(5) A5 G (5). 2. Ixora parvillora Habit Evergreen shrub. Root Branched tap root system. Stem Erect, woody, cylindrical, solid, branched, green. Rubiaceae 349 Leaves Opposite, decussate, sessile, simple, stipulate, simple interpetiolar, cu spidate, oblong, entire, acute, unicostate reticulate. Inflorescence Terminal cyme. Flower Bracteate, pedicellate, complete, hermaphrodite, actinomorphic, tetramero us, epigynous, white or pink. Calyx Sepals 4, gamosepalous, calyx tubular, persistent, valvate. BA_ S. Flower C -Gynoeceum M E -Floral diagram PETALOID SEPAL Co-T. S. ovary Fig. 56.1. Mussaenda. 350 Plant Taxonomy Corolla Petals 4, gamopetalous, tubular, twisted in bud, white or pink. Androecium Stamens 4, polyandrous, epipetalous, on the mouth of corolla tube, al ternipetalous, filament short, anthers dithecous, introrse. Gynoecium Bicarpellary, syncarpous, bilocular, ovary inferior, axile placentatio n, one ovule in each loculus, style simple, filiform, hairy; stigma slender, fus iform, bifid. Floral formula ED V Ic 4 ) C(4) A4 G (2). 3. Mussaenda (Fig. 56.1) Habit An ornamental, perennial, evergreen, shrub. Root Branched, tap root. Stem Herbaceous but woody in the lower portions, erect, cylindrical, branched, differentiating into nodes and internodes; internodes are swollen, hairy and gre en. Leaf Ramal and cauline, simple, opposite decussate, sub-sessile, interpetiolar s tipules are present, lanceolate, entire, acute, unicostate reticulate. Inflorescence Dichasial cyme. Flower Bracteate, pedicellate, complete, hermaphrodite, actinomorphic but mature and older flowers are zygomorphic, pentamerous, epigynous, whitish yellow. Calyx Sepals 5, polysepalous or sometimes gamosepalous; 4 sepals are smaller and 5th one is modified into a yellow leafy bract in zygomorphic flowers; all sepals are of same shape, size and colour in actinomorphic flowers; persistent, valvate,. green. Corolla Petals 5, gamopetalous, corolla tube is elongated and funnel shaped, val vate or rarely imbricate, yellow; coronary structures are present in the form of silky h airs. Androecium Stamens 5, alternipetalous, polyandrous, epipetalous, dithecous, basi fixed or dorsifixed introrse.

Gynoecium Bicarpellary, syncarpous, inferior, bilocular, many ovules, axile plac entation; style long with two stigmatic lobes. Fruit A berry. Seed Many, small, endospermic. Floral formula (a) Young Flower Br, 0 V, K5, C(5) A5, G. (b) Older flowers Br, m , V K5, CO) A59 G(2). Literature 1. Joshi, A. C. 1938 : A note on the morphology of ovule of the Rubiaceae with s pecial reference to Cinchona and Coffea. Curr. Sci. 70 : 236-237. 2. Majumdar, G. P. and Pal, P. 1958 : The stipules of the Rubiaceae a review. Tr ans. Bose. Res. Inst. 22 : 57-67. 3. Schumann, K. 1891 : Rubiaceae in Die. Nat. Pflanzenfamilien IV (4) : 1-55. SO 59 OLEACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Gamopetalae Sympetalae Lignosae Bicarpellatae Contortae Gentianales Gentianales Oleaceae Oleaceae Oleaceae Diagnostic characters Trees or shrubs, leaves opposite, simple or pinnately compound, stipulate; 2 merou s flowers, the 2 anthers with cells back to back; 2 loculed, superior ovary with g enerally 2 ovules each. Distribution Oleaceae or olive family includes 29 genera and 600 species approximately. Its m embers are cosmopolitan in distribution but occur in great diversity in temperat ate and tropical Asia. Vegetative characters Habit Trees or woody shrubs or undershrubs sometimes scandent or twining. Root Tap and branched. Stem Woody, erect or twining, branched. Leaf Simple or compound pinnately opposite exstipulate, alternate in Jasminium h umile; entire, unicostate reticulate venation. Floral characters Inflorescence Cymose or compound racemes, panicle. Flower Actinomorphic, hermaphrodite rarely unisexual (Fraxinus spp. and some spe cies of Olea), tetramerous, hypogynous. Calyx Sepals usually 4 or 5 in Fraxinus excelsior and sometimes more than 5 (Jas minum spp.) more or less united, bell-shaped or tubular, usually small, toothed or nearly entire, free in some species, persistent. Corolla Petals 4-5, sometimes more (Jasminum spp.), gamopetalous, salver or funn el shaped or rotate, valvate, imbricate (Jasminum), sometimes as in Fraxinus, 4 petals, rarely free (Fraxinus ornus) or even absent. Androecium Stamens 2, sometimes 4 (Hesperelaetz and Tessarandra), inserted on th e corolla-tube or at the base of the petals. Filaments usually short, anthers 2celled, extrose. Gynoecium Carpels 2, syncarpous; ovary superior, 2-celled, with usually 2 anatro pous, ascending or pendulous ovules in each loculus, rarely 4-10 ovules in each loculus; style usually short, simple; stigma terminal simple or bilobed. (362) Oleaceae 363 Fig. 59.1. Nyctanthus arbortristis. Fruit Variable capsule, samara, berry or drupe. Seed Endospermic and oily. Pollina tion Entomophilous. Floral formula Ki4_5) C(4_5) A2_4 Gn. Economic importance

Food Fraxinus ornus is cultivated in the Mediterranean countries for the sake of saccharine juice which flows out and coagulates into "Manna" a sugary substance . Medicinal Leaves of Jasminum sambac (Arabian Jasmine) are used in India as lacti fuge and are said to be as efficacious as belladona. A decoction of the leaves a nd roots are used for eye-sore. A decoction of roots of Jasminum pubescence has some repute as an antidote for cobra venom while those of J. humile and J. offic inale are said to be usful in curing ringworm. The leaves and flowers of Jasmini um grardiflorum contain a resin, salicylic acid, an alkaloid Vasminine' and an a stringent. The whole plant is considered to be anthelmintic, diuretic and emmena gogue. Inflorescence Stigma Floral diagram Flowering branch 364 Plant Taxonomy Oil Olive oil is obtained from the fruit-pulp and seeds of Olea europea. Perfumes The sweet-scented flowers of Jasminum grandiflorum and J. sambac contai n an essential oil which is used in the preparation of well-known perfume, Jasmi ne. Timber The wood of Fraxinus excelsior (Ash), Olea dioica and Olga robusta, yield excellent timber which is hard and durable. Dye The corolla of Nyctanthus yields an orange dye.Ornamentals Many species of Jasminum, e.g., J. sambac and J. grandiflorum are cu ltivated as ornamental shrubs on account oftheir elegant foliage and beautiful s weet-scented flowers. Nyctanthes-arbortristis (Harsinghir) is very popular for i ts sweet-scented flowers which open at night. Many species of Ligustrum, Syringa (Lilac) and Forsythia are also beautiful ornamentals. Affinities of the family Hutchinson has placed the family under the order Loganiales related to Loganiace ae. He was, however, of the opinion that the family is an unnatural assemblage c onsidering Fraxinus to be more akin to the Sapindaceae and Ligustrum to Loganiac eae. He establishes its relationship with Rubiaceae and Apocynaceae from which i t can be distinguished by the aestivation of corolla, number of stamens, positio n of ovules and the nature of endosperm. Common plants of the family 1. Jasminum name derived from Jasmine, the Persian name signifying fragrant; cul tivated in gardens. 2. Nyctanthus from Greek nyx, nycols, night and anthos, flower, the flowers open in the evening and fall off the following morning. Nyctanthus arbor tristic Harsi ngar ornamental tree. 3. Olea Latin name of olive, a tree. 4. Osmanthus a shrub or small ever-green tree. 5. Syringa a musical pipe, its long, straight branches from which the pith is ea sily removed to make flute or whistle. Division of the family and chief genera The Oleaceae is divided into two sub-families : Sub-family I. Oleoideae Seeds pendulous. Fruit not vertically constricted. Tribe I. Fraxineae Fruit samara, Genus : Fraxinus. Tribe II. Syringeae Fruit loculicidal capsule. Genus : Syringa. Sub-family II. Jasminoideae Seeds erect. Fruit vertically constricted and divide d into two parts by constriction. Genus : Jasminum. Important type Jasminum (H. Chambeli) (Fig. 59.2) Habit Shrubby or climbing cultivated in gardens. Root Tap and branched. Stem Branched, erect or climbing. Leaf Opposite rarely alternate, simple, usually compound, imparipinnate, leaflet s opposite, nearly sessile, entire, end one the longest.

Inflorescence Terminal cymose clusters. Flower Bracteate, hermaphrodite, actinomorphic, pedicellate, complete. Flower Floral diagram Fig. 59.2. Jasminum Calyx Bell shaped sometimes cytindrical with 4-10 lobes, teeth short or long, gamosepalous, inferior. Corolla Gamopetalous, tube slender, 4-10 lobed, to be spreading. Androecium Stamens 2, epipetalous, bithecous, anther oblong, filaments short, co nnective shortly produced. Gynoecium Bicarpellary, syncarpous, bilocular, superior, 2 ovules in each loculu s; style slender, stigma capitate or linear. Fruit A berry. Floral formula -eV K, -- 4-10) C(4-10) A2 G.

65 0'191SOLANACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Gamopetalae Sympetalae Herbaceae Bicarpellate Tubiflorae Solanales Polemoniales Solanaceae Solanaceae Solanaceae Diagnostic features Plants herbs, shurbs rarely trees; leaves alternate, flowers solitary or in cyme s; axillary or terminal; flowers pentamerous, actinomorphic, hypogynous, hermaph rodite, calyx persistent, gamosepalous, corolla gamopetalous, campanulate; stame ns epipetalous; gynoecium bicarpellary, syncarpous, ovary obliquelyplace51, axil e placentation; swollen placentae; ovules many in each locules; fruit capsule or berry. Distribution The family is commonly called 'Potato family'. It is a large family well distrib uted in tropics and sub-tropics, though a few members are found in temperate zon e. The family includes 2,000 species belonging to 90 genera. In India it is repr esented by 70 species of 21 genera. Several members are cultivated through out t he world for their great economic importance; among them are Solanum tuberosum ( Potato), Solanum melongena (Bringal), Lycopersicum esculentum (Tomato) etc. Vegetative characters Habit Mostly herbs (Petunia, Withania), shrubs and trees. Root A branched tap root system. Stem Aerial, erect, climbing (Solanum jasminoides), herbaceous, or woody, cylind rical, branched, solid or hollow, hairy, or glabrous, underground stem in Solarium tuhe msum. Leaves Cauline, ramal, exstipulate, petiolate or sessile, alternate sometimes op posite, simple, entire pinnatisect in Lycopersicum, unicostate reticulate venation. Inflorescence Solitary axillary, umbellate cyme, or helicoid cyme in Solanum.Flower Bracteate or ebracteate, pedicellate, complete, hermaphrodite, actinomorp hic, pentamerous, hypogynous. Calyx Sepals 5, gamosepalous, tubular or campanulate, valvate or imbricate, pers istent,

green or coloured, hairy, inferior. Corolla Petals 5, gamopetalous, tubular or infundibuliform, valvate or imbricate aestivation, scale or hair-like outgrowth may arise from the throat of the corol la tube, coloured, inferior. (390) Solanaceae 391 Androecium Stamens 5, epipetalous, polyandrous, alternipetalous, filaments inser ted deep in the corolla tube, anthers dithecous, usually basifixed or dorsifixed , introrse, inferior. Gynoecium Bicarpellary, syncarpous, ovary superior, bilocular, unilocular in Hen oonia, axile placentation placentae swollen, many ovules in each loculus, ovary obliquely placed; in some cases nectariferous disc is present; style simple; sti gma bifid or capitate. Fruit A capsule or beery. Seed Endospermic. Pollination Entomophilous. Floral formula Br e V K(5) C(5) A5 Gam. Economic importance The family is of great economic importance. Food Many members viz., Solanum tuberosum (Potato), Solanum melongena (Brinjal), Lycopersicum esculentum (Tomato), Capsicum (H. Mirch) etc. are used as vegetabl es. Physalis peruviana (H. Rasbhari) produces edible berries. Medicinal Atropa belladona contains alkaloid Atropine; this is used in Belladona plaster. Atropine is used in eye testing. Nicotiana tabacum (tobacco) yields Ni cotine. Hyoscyamus niger, Solanum nigrum, Datura (H. Dhatura), Withania somnifer a (Ashwagandha) are used medicinally. Narcotics Tobacco is obtained from leaves of Nicotiana tabacum and variously use d in cigars, bidi, chewing, jarda etc. Ornamentals Petunia, Cestrum, Lycium, Salpiglossis, Schizanthus are cultivated i n gardens for ornamentals. Primitive characters 1. Shrubs, trees and perennial climbers. 2. Leaves simple and alternate. 3. Inflorescence solitary axillary or terminal. 4. Flowers actinomorphic, hermaphrodite and hypogynous. 5. Pollination by insects. 6. Ovules numerous in each loculus. 7. Stamens dithecous. 8. Seeds endospermic. Advanced characters 1. Most of the plants are herbaceous and many are annuals. 2. Leaves exstipulate, in some finely divided. 3. Calyx and corolla are gamosepalous and gamopetalous. 4. Stamens epipetalous. 5. Reduction in the number of carpels to two. 6. Gynoecium is syncarpous. 7. Fruit is simple. Affinities of the family Hallier regarded Solanaceae as a primitive member of the tubiflorae together wit h the Scrophulariaceae and both have arisen very likely from the Linaceae. Wetts tein placed the family in the Tubiflorae along with Convolvulaceae. Rendle place d the family in the Tubiflorae, assigning a separate position for the Convolvula ceae under the order Convolvulales. 392 Plant Taxonomy Solanaceae bears a close relationship to the Boraginaceae in alternate leaves, r egular flowers and five stamens. It is related to Convolvulaceae in the presence of persistent calyx, twisted corolla and false septum. It is allied to Scrophul ariceae but the latter differs from it in actinomorphic flowers and obliquely pl

aced carpels. Common plants of the family 1. Browallia An ornamental genus characterised by the presence of 4 didynamous s tamens. 2. Cestrum nocturnum (H. Rat Ki Rani or Queen of the night) A garden plant with flowers emitting sweet smell at night. 3. Datura metal (H. Dhatura or Thorn apple) A herb with highly poisonous fruits and seeds. 4. Petunia A garden ornamental. 5. Lycoperscium esculentum A tall herb with red globose pulpy fruits. 6. Physalis peruviana Produces edible reddish berries that are enclosed in a bla ddery and persistent calyx. 7. Solanum nigrum (H. Mako) A roadside herb S. xanthocarpum a spinous xerophytic herb. 8. Withania somnifera (H. Asgandh, Sanskrit Ashwangandha) A perennial tall herb. Division of the family and chief genera Von Wettstein divided the family in to five tribes and two groups : A. Embryo curved like a semicircle, stamens 5,-filaments equal. Tribe 1. Nicandreae Gynoecium bicarpellary with 3-5 locules e.g. Nicandra. Tribe 2. Solaneae Ovary bilocular; e.g. Solanum, Capsicum. Tribe 3. Datureae Ovary 4-celled, septum dividing placenta equally; e.g. Datura. B. All stamens or 2-4 stamens fertile; embryo slightly or almost straight. Tribe 4. Cestreae All 5 fertile stamens; seeds not flat; e.g. Cestrum. Tribe 5. Salpiglossideae Stamens 2-4 fertile; flowers zygomorphic; e.g. Browalli a, Salpiglossis. Important types 1. Solanum nigrum (Fig. 65.1) Habit A wild herb. Root Branched tap root system. Stem Erect, aerial, woody below and herbaceous above, cylindrical with distinct ribs, solid, branched, green. Leaves Alternate, simple, exstipulate, petirAate, ovate, repand, acute, glabrous , unicostate reticulate venation. Inflorescence Extra-axillary helicoid cymes. Extra axillary position is due to f usion. Flower Ebracteate; pedicellate, complete, hermaphrodite, actinomorphic, pentamer ous, hypogynous, small and white. Calyx Sepals 5, gamosepalous, pentafid, valvate, persistent, green, hairy, infer ior. Corolla Petals 5, gamopetalous, rotate, valvate, five lobed, white, inferior. Androecium Stamens 5, polyandrous epipetalous, alternipetalous, filaments shorts , equal in length, anthers long and conniving, basifixed, dithecous, and dehiscence by a pical pores. C-L S.Frower B -Entire Flower Solanaceae Gynoecaum 4-Entire plant r- Moral dlognom Fig. 65.1. SoSwum nifpnin In A numbers 1, 2, 3, 4, denote successive youngefildwers-showing helicoid cymos e inflorescence. Gynoecium Bicarpellary, syncarpous, ovary superior, bilocular, axile placentatio n, placentae swollen, ovules many in each loculus, ovary obliquely placed; style simple, hairy; stigma bilobed. Fruit A berry.

Seed Endospermic. rTh Floral formula -EK( C G 5) _ (5) _ _(5) _ Q. 2. Withania somnifera (Fig. 65.2) Habit Wild perennial herb. Root Branched, tap root system. Stem Erect, aerial, cylindrical, solid, woody below and herbaceous above, branch ed, covered with hair. 393. 394 Plant Taxonomy C - Gynoeceenn 9 E Floral diagram Fig. 65.2. Withania somnifera. Leaves Cauline, ramal, opposite, simple, exstipulate, petiolate ovate, entire, a cute, covered with glandular branched hair, unicostate reticulate venation. Inflorescence Axillary cyme. Flower Ebracteate, sub-sessile, complete, hermaphrodite, actinomorphic, pentamer ous, hypogynous. Calyx Sepals 5, gamosepalous, campanulate, pentafid, valvate, hairy, green, infe rior. Corolla Petals 5, gamopetalous, pentafid, campanulate, valvate aestivation , corolla lobes inflexed, greenish yellow. Androecium Stamens 5, polyandrous, epipetalous, alternipetalous inserted near th e base of corolla tube, filaments equal in length; basifixed, dithecous, intrors e B S. Flower A - flowering awls SWOLLEN PLACENTUM OVULE SEPTUM Solanaceae 395 Gynoecium Bicarpellary, syncarpous, ovary superior, bilocular, ovules many, axil e placentation, placentae swollen, ovary obliquely placed; hairy, style simple; stigma bifid. Fruit A berry enveloped by persistent calyx. Seed Endospermic. it". Floral formula ED V lc(5) C(5) A5 G. 3. Datura metel Habit An annual herb. Root Tap, branched, annual. STYLE COROLLA TUBES THALAMUS FLOWERING TWIG V.S. FLOWER FLORAL DIAGRAM Fig. 65.3. Petunia alba. 396 Plant Taxonomy Stem Herbaceous, erect, cylindrical, branched, slightly woody and slightly fistu lar, hairy Leaf Alternate, opposite in floral region, simple, exstipulate, petio late, ovate-lanceolate, acuminate or acute apex, unequal at base, sinuate-toothed or repand margin, glab rous on both sides, unicostate reticulate venation. Inflorescence Solitary axillary. Flower S hortly pedicellate, hypogynous, pentamerous. Calyx Sepals 5, gamosepalous, tubular, angulate, 5-toothed, teeth triangular lan ceolate, acuminate, green, persistent, twisted aestivation, inferior. Corolla Petals 5, gamosepalous, campanulate, 5 or 6 cuspidate acute angles, viol et outsi twisted aestivation, inferior.

Androecium Stamens 5, epipetalous, filaments long, anthers basifixed, introrse, dithecous_ Gynoecium Bicarpellary, syncarpous, ovary superior, obliquely placed, four celle d due to the formation of false septum, axile placentation, many ovules in each loculus, style long, stigma bilobed. bracteate, hermaphrodite, actinomorphic, complete. Floral formula Br ED V K(5) C(5) 5) _ (5) _ A _5 Gam. (g). 4. Petunia alba (Fig. 65.3) Habit A cultivated annual herb. Root Tap, branched. Stem Herbaceous, erect, cylindrical, solid, branched, hairy, green. Leaf Cauline, simple opposite decussate, exstipulate, ovate, unicostate reticula te venation. Inflorescence Axillary or terminal condensed cymes. Flower Pedicellate, bracteate, hermaphrodite, complete, actinomorphic, pentame hypogynous. Calyx Sepals 5, gamosepalous, campanulate, sepals free above and fused below, gr een, hairy, imbricate aestivation, inferior. Corolla Petals 5, gamopetalous, campanulate, tube hairy, twisted aestivation, in ferior. Androecium Stamens 5, polyandrous, epipetalous,- alternating with the petals, fi laments long, anthers basifixed, dithecous, introrse. Gynoecium Bicarpellary, syncarpous, ovary superior, obliquely placed, bilocular; swollen axile placentation; ovules many; style long, slightly, twisted, stigma capitate, bilobed. Floral formula Br V Ic< n 5 ) _ C (5) A5 G. 00 67 BIGNONIACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Gamopetalae Sympetalae Lignosae Bicarpellatae Tubiflorae Bignoniales Personales Bignoniaceae Bignoniaceae Bignoniaceae Diagnostic characters Plant predominently lianous, compound leaves, zygomorphic flowers, anthers conni vent in pairs; numberous ovule, silique-like woody capsule, large winged seed an d non-endospermic. Distribution Bignoniaceae or Bignonia family is primarily tropical or subtropical family comp rising 120 genera and 800 species of trees or shrubs, often climbing or twining vines and rarely herbs. Vegetative characters Habit Mostly trees or shrubs; often climbing or twining vine, rarely herbs. Root Tap, deep, branched. Stem Hard, woody and branched; weak in climbers and twiners, rootlet or tendril climbers. Leaf Usually pinnately compound, opposite, decussate, rarely simple or alternate, exstipulate, terminal leaflet modified into tendril, adhesive disc o r hook. Floral characters Inflorescence Usually dichasial cyme with bracts and bracteoles. Flower Bracteate, bracteolate, hermaphrodite, hypogynous, zygomophic, complete. Calyx Sepals 5, gamopetalous, lobed or bilabiate, valvate.

Corolla Petals 5, obliquely campanulate or infundibuliform, imbricate, gamopetal ous, lobes or teeth 5, sometimes bilabiate. Androecium Stamens 4, slidynamous, epipetalous, posterior staminode, sometimes 2 (Catalpa); anther two-lobed, lobes divaricate, disc present. Gynoecium Carpels 2, syncarpous; ovary superior, bilocular with axile placentati on, each loculus many ovuled; sometime unilocular (Eccremocarpus) with two bifid parietal placentae; style terminal and single; stigma bifid. Fruit Capsule two valved septicidal or loculicidal or berry. Seed Non-endospermic, flattened, winged. Pollination Entomophilous. (401) 402 Plant Taxonomy Fig. 67.1. Tecoma stans (L). H.B.K., A. Flowering twig with pinnate leaves. Flor al formula Br olo V K(5) C(5) A4 + 1 std. Gn. Economic importance Timber Catalpa bignonioides, Spathodea companulata, Tabebuia pentaphylla, Oroxylum are prized for timber. Dye The leaves of Cybistax antisyphilitica are used as source of blue dye. Ornamental The garden ornamental plants are represented by Pyrostegia venusa (Sy n. Bignonia venusta), Spathodea campanulata, Tecoma stans, Affinities of the family Bignoniaceae is allied to Acanthaceae, Pedaliaceae and Scrophulariaceae. Wettste in considered the family a little advanced over Scrophulariaceae. Bessey placed the family in Scrophulariales, the origin of which was traced from Polemoniales (Boraginales). Hallier included L.S. Flower Fruit T.S. Ovary Floral diagram Bignoniaceae 403 it in the Tubiflorae and placed it near Pedaliaceae and Acanthaceae to which it is also very allied. Hutchinson (1959) put the family under his 59th order Bigno niales which also includes family Pedaliaceae; according to him the origin of Bi gnoniales was from Loganiales. Common plants of the family 1. Crescentia cujete Calabash fruit large, ground-like. 2. Jacaranda acutifolius Jack tree a road-side avenue tree with wonderful masses of purplish flowers. 3. Catalpa speciosa tree grown for valuable wood. 4. Bignonia vensusta Climber, ornamental plant. 5. Millingtonia hortensis Indian cork tree. 6. Tecoma stans garden shrub. 7. Tabebuia West Indian Box tree. Division of the family and chief genera According to Schumann the Bignoniaceae is separated into four tribes : Tribe I. Bignonieae Ovary completely bilocular, compressed, parallel to the sept um or ovary may be cylindrical. Capsule septifragal. Seeds winged. Mostly climbi ng shrubs by means of leaf tendrils or sometimes trees. Genera : Glaziovia, Bign onia, Oroxylum, Millingtonia etc. Tribe II. Tecomeae Ovary bilocular compressed, at right angles to the septum, or ovary may be cylindrical. Capsule loculicidal with winged seeds. Erect shrubs o r trees, rarely climbers by tendrils. Genera : Tecoma, Catalpha, Spathodea, Hete rophragma, Pajanelia etc. Tribe III. Eccremocarpeae Ovary unilocular. Capsule dehiscing from base upwards. Seeds winged. Climbing shrubs by tendrils. Only genus Eccremocarpus (Peru, S. A merica). Tribe IV. Crescentieae Ovary uni- or bilocular. Fruit berry or dry indehiscent.

Seeds not winged. Erect trees or shrubs. Genera : Kigelia, Crescentia etc. 00 75 84,rth- wet,/ ARISTOLOCHIA EAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Monochlamydeae Archichlamydeae Herbaceae Multiovulatae Terrestres Aristolochiales Aristolochiales Aristolochiaceae Aristolochiaceae Aristolochiaceae Diagnostic features Plants twinner, flower actinomorphic, trimerous, stamens 6-36 free or attached w ith the style forming gynostemium, generally 6-loculed inferior ovary. Distribution The family Aristolochiaceae or Birthwort family contains 7 genera with 625 speci es (Rendle) having fairly wide range of distribution, the main centre of distrib ution being north temperate and tropical regions of the globe. Aristolochia is m ainly tropical. According to Willis genera 5, species 300. Vegetative characters Habit Mostly climbing herbs or shrubs with woody stems or low herbs. Root Tap and branched. Stem Softly woody or herbaceous, branched twinner. Leaf Simple, alternate, with oil glands, entire; petioled, exstipulate. Floral characters Inflorescence Either solitary or clustered or in racemes of spikes. Flower Hermaphrodite, actinomorphic or zygomorphic, epigynous and with haplochla mydous trimerous perianth usually. In Aristolochia flowers are zygomorphic, the perianth members are united to form a pitcher like structure having constricted neck, the interior of which is line d with downwardly directed hairs. F'er petaloid, united, 3-lobed or unilateral often bizarrely coloured occasio nally an inner whorl of 3 minute teeth (vestigial corolla) present. Androecium Stamens 6-36. free or adnate to style and producing a column or gynos temium, filaments short and thick or anthers sessile, bithecous dehiscing longit udinally. *--ynoecium Carpels usually 6 rarely 4, connate in an inferior six cha mbered, rarely four chambered ovary. Ovules anatropous, many on mile palcentas. Fruit Septicidal capsule with basal dehiscence. Seed Endospermic and small. Pollination Entomophilous. (436) Aristolochiaceae 437 Fig. 75.1. Aristolochia clematits L. A. Flower in I.s. flowing many ovules, infe rior ovary and gynostemium anther adnate to the style forming gynostemium. Floral formula ED or olo # P(3) (3) A6-36 or (6-36) G. Economic importance Medicinal The roots of Aristolochia, serpentaria and Bragantia wallichii serve a s a cure for snake-bites. The roots of Aristolochia indica are used by snake-cha rmers for catching snakes. Ornamental Aistolochia grandiflora, A. clematis A. microphylla, A. gigas are orn amental plants. Affinities of the family Hutchinson placed the family as a terminal derivative from the Ranales via herba ceous members of Berberidaceae. It is presumed that the more or less woody climb ing members of the family may have derived from the herbaceous ones. According to Bentham and Hooker, it is most closely related to certain petalifer ous families such as Menispermaceae._ Ranales Berberidales (Menispermaceae) -4 Aristolochiales. Common plant of the family 1. Aristolochia bracteolata a decumbent, slender perennial.

2. A. tagala twining perennial shrub. Flowering branch Flower L.S. Flower Floral diagram Anther Ovary Flower without perianth 438 Plant Taxonomy Division of the family and chief genera The Aristolochiaceae is divided into three tribes viz., I. Aristolochieae Perennial herbs or shrubs; flowers zygomorphic, solitary or in cymes or racemes. Petals absent. Stamens in one row, united with style. Ovary i nferior. Fruit a capsule e.g., Aristolochia, Euglypha and Helostylis. II. Bragantieae Shrubs or semi shrubs, flowers actinomorphic in cymes or racemes . Petals absent. Stamens in two rows, free or united with the style. Ovary infer ior, fruit a capsule e.g. Apana, Thottea. III. Sarumeae Perennial herbs, flowers actinomorphic, solitary. Petals present o r absent. Stamens in two rows, free or united at base of filaments. Ovary inferi or or semi-inferior. Fruit a capsule or follicle e.g. Asarum and Saruma. 76 LORANTHACEAE (Mistletoe Family) Classification : Bentham & Hooker Engler & Prantl Hutchinson Monochlamydeae Santalales Lignosae Achlamydeae Loranthaceae Santalales Loranthaceae Loranthaceae Diagnostic features Aerial parasite usually, cup-shaped receptacle; inferior ovary and the absence o f distinct ovules. Distribution Mistletoe family consists of 36 genera and 1300 species, mostly distributed in t ropical and temperate regions. About 10 genera and 52 species have been reported from India. Vegetative characters Habit Herbs, perennial, aerial parasitic or semi-parasitic on trees. Root Modified adventitious rooshtaTefoinoihaTTisoiTI7uystia (W. Australia) roote d in earth. Stem Herbaceous, soft, branched, branching dichotomous, swollen nodes. Leaf Simple, opposite, entire, exstipulate, thick and leathery, ever-green, ofte n reduced to mere scales. Floral characters Inflorescence Flowers are solitary or in racemes and cymes. Flower Hermaphrodite or unisexual, e.g., Viscum (when so, plants dioecious), act inomorphic or slightly irregular, often brightly coloured, dimerous or trimerous , apetalous, epigynous with cup-shaped or disc-like receptacle. Perianth Biseriate, the two whorls similar and 2-3 merous, both green and sepallike (Viscum), or both large, brilliantly coloured and petaloid, e.g., Loranthus (no apparent differentiation into calyx and corolla), an irregular rim, called calyculus, is sometimes present below the perianth and is interpreted as calyx. Androecium Stamen 4-6, opposite to the perianth members, and borne on them (epip hyllous) or at their base; anthers 2-celled, dehisencing by longitudinal or by t ransverse slits or by pores (in staminate flowers a rudimentary pistil may be pr esent). Gynoecium Carpels 1-3 or 5, syncarpous, inferior, unilocoular, ovules not differ entiated from Placenta, basal placentation, style simple or absent, stigma 1 and often sessile.

(439) 440 Plant Taxonomy Fig. 76.1. Viscum album. A. Portion of the twig bearing staminate flowers. B. Ma le flower, C. Staminate flower in L.S.; p, perianth; a, anther. D & E. Pistillate flower, e ntire and in L.S. respectively; p, perianth member; F. Floral digram femae flower (9); G. Floral diagram male flower (1). Fruit Berry or drupe, often viscid. Seed Fleshy endospermic. Pollination Entomophilous. Floral formulae Br Brl f P4_6 4-6 A4-6 G1 r-Th Br Br! ED 0 P4_6 A4_6 Go Br Brl S P4_6 4-6 AO G1 Affinities of the family The family is closely related to Santalaceae but can be distinguished by its aer ial parasitic habit; cup-shaped receptacle; inferior ovary, lacking a clearly de fined locule and the absence of distinct ovules. D E B c MALE FLOWER (1) (:N) F FLORAL DIAGRAM FEMALE FLOWER ANTHER G di FLORAL DIAGRAM MALE FLOWER .4y A Loranthaceae 441 The Loranthoideae and Viscoideae show great variations in the floral structure. Moreover, the mode of development of the embryosac, endosperm and embryo and loc ation of the viscous part of the fruit are different in the two sub-families. The Loranthoideae contains bisexual or unisexual flowers generally associated wi th calyculus. The ovary is characterised by a collenchyma pad and contains a mam elon which may or may not be lobed. The viscid zone of the fruit is located outs ide the vascular bundles of the corolla. Other features peculiar to the Lorantho ideae include the Polygonum type of embryosac, protrusion of tip of the embryosa c into the style, formation of composite endosperm and vertical division of the zygote. The Viscoideae contains only unisexual flowers where the calyculus in invariably absent. The viscous layer of the fruit is situated within the vascular bundles of the perigone. The bisporic embryosac with a slow but steady curvature, endosp erm developing from a single embryosac and embryo without a suspensor are unique to the Viscoideae. In view of the above differences, Johri and his associates suggested that the Lo ranthoideae is distinct from the Viscoideae. These differences, according to the m, warrant the elevation of the status of the sub-families to that of families L oranthaceae and Viscaceae. Common plants of the family. 1. Arceuthobium minutissimum Hook. f., parasitic on the twigs of Pinus wallichia na, is the smallest known dicotyledonous plant, known in the world, is endemic i n India. 2. Dendropthoe falcata Parasite on mango and other trees. 3. Viscum album True Misletoe- was used as ceremonial plant by the Druids and ot her early Europeans. 4. Loranthus cordifolius Abundant on Oak (Quercus dilatata) in Giri valley, Cent ral India. Division of the family and chief genera The Loranthaceae is divided into two sub-families : Sub-family I. Loranthoideae Flowers bisexual or unisexual. Calyculus present. Ex

amples : Loranthus, Struthanthus, etc. Sub-family II. Viscoideae Flower unisexual. Calyculus absent. Examples : Arceuth obium, Viscum etc. 77 ,A7111( EUPHORBIACEAE Classification : Bentham & Hooker Engler & Prantl Hutchinson Monochlamydeae Archichlamydeae Lignosae Unisexuales Geraniales Euphorbiales Euphorbiaceae Euphorbiaceae Euphorbiaceae Diagnostic features Plants shrubs or trees, a few annual herbs; milky latex present; leaves simple, stipulate, alternate, inflorescence cyathium, flowers unisexual, hypogynous, per ianth one or two whorls or absent; stamens one or many, free or monadelphous or polyadelphous, gynoecium tricarrpellary, syncarpous, trilocular, ovary superior, axile placentation; stigma trifid; fruit schizocarpic r seeds endospermic. Distribution The family is commonly known as Spurge family. It is one of the largest families consisting of 300 genera and 7500 species. The members are cosmopolitan in dist ribution except arctic region. In India the family is represented by 450 species belonging to 70 genera. Vegetative characters Habit Herbaceous' shrubby or tree habit are known. Species of Croton, Phyllanthu s are herbs; those of Acalypha and Jatrapha are shrubs; and fully developed tree habit is met with in Putranjiva roxburghii, Bischofia sp. and Emblica officinal is (H. Amla); Ricinus communis produces small tree like plants; climbers found i n 7'ragia. The genus Euphorbia, with more than seven hundred species, shows vari ous gradations of habit from small herbs (E. hirta and E. thymifolia) are spiny perennial herbs (E. mili syn. E. splendens) to shrubs (E. pulcherrima) and final ly to cactus like shape as in Euphorbias of tropical Africa or Indian species of E. royleana. In the section xylophylla of Phyllanthus the ultimate branches bec ome phylloclades with the leaves reduced to scales. In Phyllanthus niruri the la teral branches become small and leaves arranged in two rows so as to resemble a pinnately compound leaf. Root Much branched tap root. Stem Aerial, herbaceous or woody, erect or prostrate, cylindrical, or angular, s ometimes spiny, branched, in someslaclodes (Euphorbia sp.), hollow or solid, mil ky latex present. Leaves Alternate, sometimes opposite or whorled, petiolate or sessile, stipulate (Euphorbia, Croton) or exstipulate, petiolate or sessile, compound (Hevea), ent ire or serrate, unicostate (Jatropha) multicostate (Ricinus, Manihot), reticulat e venation. In some leaves are variegated (Codiaeum variegatum). (442) Euphorbiaceae 443 Floral characters Inflorescence - Terminal spike like racemes or spikes (Croton); axillary interru pted spikes (Mercurialis); axillary (Phyllanthus); terminal panicled raceme (Ric inus); catkin with pistillate flowers at the base and staminate flowers at the a pex of the same inflorescence (Acalypha); and cyathium (Euphorbia and Synadenium ). A cyathium is a cymose inflorescence of which the terminal flower is female; below it there are four or five bracts which are connate foiming a calyx like in volucre. In the axil of each bract there is a scorpioid cyme of male flowers; ea ch male flower consists of a single stamen. In between involucral bracts coloure d glands are found. These glands are free in Euphorbia but always fused in Synad enium. Flowers - Bracteate and pedicellate but incomplete unisexual and apetalous, some times both the perianth whorls are absent. Where perianth is present it is regul ar and hypogynous. In Weilandia a pentamerous flower with both sepals and petals are found; carpels epipetalous. In Croton sepals and petals are five with numer

ous stamens in male flowers but only three carpels in female flowers, Apetalous flowers in Manihot, Acalypha, Phyllanthus. Both the perianth whorls totally abse nt in Euphorbia where both the male and the female flowers are naked. Male flower : Perianth- Absent or 3 or 5. Androecium - Stamens single in Euphorbia, 3 in Phyllanthus, 5 in Crotonopsis; hi ghly branched but in 5 groups so as to give a miniature tree like appearance in Ricinus; 8-16 in Acalypha and slightly monadelphous; 10 in Jatropha; 80-100 free stamens in Croton; anthers dithecous. In Phyllanthus cyclanthera synadrous cond ition is found with both filaments and anthers fused to form a closed ring like structure (compare the genus Cyclanthera of the Cucurbitaceae). Female Flower : Perianth - As in male flower. Staminodes may be present or absent. Gynoecium - Tricarpellary, syncarpous, superior, trilocular, axile placentation, one or two ovules per loculus; styles 3, may be forked; stigma 3 or 6, linear o r filiform, plumose and red in Ricinus. In Crotonopsis there are three carpels b ut the ovary is unilocular and one ovuled. Fruit - Schizocarpic capsule or regma (Ricinus) rarely drupe (Hippomane), or ber ry (Bischofia). Seed - Endospermic, cotyledons flat or folded; caruncle found in Ricinus, Mercur ialis etc. Pollination - Entomophilous and in some anemophilous (Ricinus). Floral formulae - Male flower - P(3-5) (3-5) or 3+3 or 0 Al a GO Female flower - P(3-5) or 3+3 or 0 AO or staminode GO1 Economic importance Food - The tuberous roots of Manihot utilissima yield tapioca and farina which a re starchy. The fruits of Emblica officinalis are used in manufacture of pickle and jam. Fruits of Bridelia are edible. Oil - Ricinus communis has a highly oil rich endosperm which is chiefly used as a strong purgative. Recently it has found its use as a lubricant in airplane eng ines. Aleurites moluccana produces candle-nut oil from its seeds which is used as a su bstitute for linseed oil in making paints, varnishes etc. The oil cake is poison ous and can be used only as a fertilizer. 444 Plant Taxonomy Croton tiglium produces an oil from its seeds which is perhaps one of the most p owerful purgatives known. Medicinal The milky juice of Manihot, Buxus, Euphorbia, Mercurialis is poisonous . The fruits of Emblica officinalis (syn. Phyllanthus emblica) are very rich in vitamin C and are used in medicines. In the Ayurvedic system of medicine the fru its are used in a wide variety of medicines. The juice of leaves of the Phyllanthus reticulatus is diuretic and also used to stop bleeding from gums. From Ricinus communis the alkaloids ricinine and toxalbum in ricin are obtained. The plants of Acalypha paniculata and A. indica cure pneu monia and asthma. Putranjiva roxburghii is considered mysteriously beneficial to children by Hindu women. Jatropha curcas gives Curcin which is a powerful purga tive. Rubber Rubber is obtained from Hevea brasiliensis, Manihot etc. by tapping the b ark and coagulating the latex. Euphorbia intisy produces rubber in Madagascar. Other uses Plants of Cleistanthus collinus yield saponin, tanin and the glycosid e aduvin. The seed oil of Jatropha curcas is used for manufacturing soap, candle s, lubricant and also for illumination. Ornamentals Euphorbia pulcherrima, Codiaeum variegatum, Jatropha, Acalypha are c ultivated in garden. Primitive characters 1. Many plants are shrubs and trees. 2. Leaves are simple and alternate. 3. Leaves are stipulate in many genera.

4. Pollination by wind. 5. Seed endospermic. Advanced characters 1. Mostly plants are annual herbs. 2. Leaves in some genera are compound and exstipulate. 3. Flowers are arranged in distinct and specialised inflorescence. 4. Flowers incospicuous and unisexual (Euphorbia, Phyllanthus). 5. In some male flowers have only one stamen (Euphorbia). 6. Gynoecium syncarpous and axile placentation. 7. Fruit simple capsule or drupe. 8. Pollination by insects. Affinities of the family Hallier included Euphorbiaceae in his order Passionales, while Wettstein and Ren dle have placed it in the Tricoccae (created order) between Malvales and Gerania les. The family Euphorbiaceae is very closely related to the members of the Malvales particularly Sterculiaceae in the floral structure such as the occurrence of mon adelphous androecium in 25% of the total genera, valvate sepals, structure of pe tals when present, presence of pistillode and staminode, presence of caruncle an d nature of embryo. It also shows resemblances with the Geraniales and the Sapindales in the occasio nal presence of disc and the number and structure of ovules. The family Euphorbiaceae might have evolved from ancestors of Malvales due to su ppression or reduction of the inner whorl of perianth and abortion of one of the members o f essential whorls viz., stamens or carpels during the course of evolution. Pax (1924) regarded the Euphorbiaceae as advanced without any primitive characte rs. According to him the unisexuality in flower is due to reduction from early b isexual types. Euphorbiaceae 445 Common plants of the family 1. Bischofia javanica A large handsome deciduous tree. 2. Croton aromaticus Cultivated as garden ornamental. 3. Emblica officinalis (H. Amla) Fruits are the richest source of vitamin C. 4. Euphorbia A large genus with great variation in habit. 5. Hevea brasiliensis (Indian rubber plant) Rubber is obtained from its latex. 6. Jatropha An evergreen shrub, good for hedge. 7. Phyllanthus A common glabrous annual weed. 8. Ricinus communis (Castor plant) An evergreen small tree. Division of the family and chief genera Pax classified the family into four sub-families and 10 tribes as follows : Group A. Platylobeae Cotyledons are broader than the radicle. Sub family I. Phyllanthoideae Latex absent; two ovules per loculus. It includes tw o tribes viz., Phyllantheae (e.g. Phyllanthus); Bridelieae (e.g. Bridelia). Sub family II. Euphorbioideae (Crotonoideae) Each loculus has one ovule, latex pre sent. It includes 8 tribes viz., Crotoneae (e.g. Croton); Acalypheae (e.g. Acalypha); Jat ropheae (e.g. Jatropha); Andrianeae (e.g. Manihot); Clutieae (e.g. Clutia); Hippomaneae (e.g. Sapium); Euphorbieae (Euphorbia). Group B. Stenolobeae Cotyledons narrow and as broad as radicle. Sub family III. Porantheroideae Ovules two in each loculus; e.g. Poranthera. Sub family IV. Ricinocarpoideae Ovule one per loculus; e.g. Ricinocarpus. Important types 1. Euphorbia pulcherrima (Fig. 77.1) Habit An annual herb. Root Tap, branched. Stem Herbaceous, green, solid, smooth, erect, unbranched, hairy, cylindrical, mi lky latex;

when branched, branches 4 angled. Leaf Simple, shortly petiolate, stipulate, alternate, entire margin, acute apex, lanceolate, stipules caducous, unicostate reticulate venation. Inflorescence Cyathium arranged in axillary clusters. Flower Pedicellate, bracteate, unisexual, central female flower and surrounded b y many male flowers and all are enclosed in a cup shaped involucre of bracts. Male Flower Naked with no perianth, bracteate, represented by a stamen. Androecium Stamen 1, stalked, filament long, jointed, anthers 2 celled. Female flower Naked, with no perianth, bracteate. Gynoecium Tricarpellary, syncarpous, ovary superior, trilocular, axile placentat ion, pistil represents the entire female flower, styles three, stigma 3, each bifid. Floral formulae Male Flower Br dPo Al Go Female flower Br 9 Po Ao Go2 2. Phyllanthus niruri (Fig. 77.2) Habit An annual herbaceous weed. Root Tap and branched. Stem Herbaceous, erect, aerial, cylindrical, green, branched, solid smooth. Leaf Alternate, small, simple, sub-sessile, elliptic oblong, glacous below, arrang e in two rows so that branches resemble compound leaves. 446 Plant Taxonomy Fig. 77.1. Euphorbia pulcherrima. F.D. of cyathium. Inflorescence Axillary. Male flower Usually solitary or in a clusters of 2-3, unisexual, incomplete, bra cteate, minute, shortly pedicellate, actinomorphic. Perianth 6 perianth leaves, polyphyllous, sepaloid arranged in two whorls of 3 e ach. 1 Euphorbiaceae 447 0-0 Flower F -0+ F lower Floral diagram Fig. 77.2. Phyllanthus niruri. Androecium Stamens 3, monadelphous, sessile; anthers dithecous, filaments fused to form a column. Gynoecium Absent. Female flower Unisexual, shortly pedicellate, bracteate, incomplete, actinomorph ic. Perianth 6 perianth leaves, arranged in two whorls of 3 each, polyphyllous, sepaloid. 448 Plant Taxonomy Gynoecium Tricarpellary, syncarpous, ovary superior, trilocular, axile placentat ion, 2 ovules in each loculus, style 3, each stigma bifi d. Floral formulae Male Flower : Br. (33, P3+3 A(3) G0. Female flower : Br. ED 9 P3 +3 AO Gill. 3. Ricinus communis (Fig. 77.3) Habit Shrub or a small tree. OVULE PLACENTUM E T. S Ovary 0 O G Flow., Fig. 77.3. Ricinus communis. Euphorbiaceae 449 Root Tap, branched. Stem Erect, aerial, woody, cylindrical, solid, branched, glabrous. Leaf Alternate, petiolate, exstipulate, simple, palmately lobed, serrate margin, acute apex, multicostate reticulate venation. Inflorescence Panicled terminal cyme.

Flower Bracteate, pedicellate, unisexual, actinomorphic, incomplete, male toward s the lower portion of inflorescence and female towards the upper portion; female flow ers hypogynous. Male flower Perianth 5 lobed, polyphyllous, valvate, green. Androecium Stamens 5, polyandrous, filaments profusely branched, anthers on smal l branches, basifixed, introrse. Gynoecium Absent. Female flower Perianth 3 lobed, polyphyllous, spathaceous, caducous. Androecium Absent. Gynoecium Tricarpellary, syncarpous, ovary superior, trilocular, axile placentat ion, one ovule in each loculus, style 3, stigma 3 and each bifid, feathery or plumose. Flower formulae Male flower : Br ED di P5 A5 Go Female flower : Br ED 9 P3 Ao 80 CASUARINACEAE (CASUARINA FAMILY) Classification Bentham & Hooker Engler & Prantl Hutchinson Monochlamydeae Archichlamydeae Lignosae Unisexuales Verticillatae Casuarinales Casuarinaceae Casuarinaceae Casuarinaceae Diagnostic features Tree with jointed branches; whorled minute, scaly leaves forming sheaths; woody cone-line, fruits. Distribution Casuarinaceae or Casuarina family is a monotypic or unigeneric family having onl y one genus Casuarina with about 65 species, mainly distributed in north east Au stralia but also in New Caledonia, Fiji, Malaysia and Mascarenes Islands. Only o ne species Casuarina equisetifolia occurs in India. Vegetative characters Habit Evergreen, xerophytic, woody, much branched trees or shrubs. Root Tap-root. Stem Hard, woody, jointed, striated; branches, green, when young. Leaves Whorled, scale-like, whorls of 4-16, usually linear to lanceolate, connat e basally, forming a sheath around the twing, and the leaf tips appearing only as teeth, in ternodes with striate grooves as many as leaves. Floral characters Inflorescence Spike or catkin-like, usually borne at the apices. Flower Unisexual, with or without perianth, arising in whorls, bracteate and bra cteolate, atinomorphic, dimerous, hypogynous. Male flower : Peianth 2 or 2+2, thin membranous, polyphyllous. Androecium Stamen 1, anther basifixed, 4-celled dehiscing by vertical slits. Female flower : Perianth As in male flower. Gynoecium Bicarpellary, syncarpous, overy superior, bilocular at first, but some times unilocular, ovules two in each loculus, placentation parietal; style 1, very sho rt; stigma bifid, linear. (458) Casuarinaceae 459

Fruit 1-seeded, small-winged samara, enclosed by two woody bracteoles and a brac t, many fruits and their bracteoles aggregated into a dry, woody, cone-like mult iple fruit. Seed Non-endospermic. A ; et ) 14 M 1 ! p e 1 e t i T # 1 ; ; I ill 4 1 ,v, s w , ' V A . 1 1 I i 1 I 1 / I r V ( 1 I I I \\VI 1 I I 1 fa P h \1iN 1\IV I Il \ \III\I 1111 1 I I I 1411 I 111 1 \ \ \t1 1111 / I \\ \\0: i II / 1 \ :\\ NI 1 II:. , . . . ' ' ' ' OPPIIIP 0 41 ' Vt , B Sheath of bracts stamen Perianth lobe Bracteoles Fig. 80.1. Casuarina equisetifolia Forst. A. Branch with staminate inflorescence s and fruits. B. A part of the staminate inflorescence, 2 whorls. C. Pistillate flower. D. Dia gram of whorl of staminate flowers. E. Diagram of a pistillate flower. Economic importance Timber Casuarina is important as timber tree. The timber is used in making oars, felloes of wheels, mine props, rafters etc. The tree is gro wn on the coasts of India in order to reclaim sandy dunes. 460 Plant Taxonomy Affinities of the family The Casuarinaceae is an isolated taxon under the Monochlamydeae. It was thought by Engler as the most primitive of dicots constituting the Verticillatae. It als o formed the 8th family of Rendle's Monochlamydeae and 1st family of Core's Arch ichlamydeae. Hjelmquist (1948) held the view that the family is not a component of the Amentiferae (Core, 1955), though it is a primitive dicot. The primitivene ss of the family is ascribed to the presence of apparent catkin-like male inflor escence, simple wind-pollinated flowers and large-rayed wood. A number of workers have regarded the Casuarinaceae as advanced: At first, Besse y (1897) took it as a member of the Sapindales; later (1915), he made it look li ke a "leafless representative of and derived from the Hamamelidaceae within the Rosales". Hallier treated it as a member of his Amentacees. Hutchinson derived t he group from the Hamamelidales, indicating it as a climax family. Rendle and Wettstein placed it in a Amentiforae and interpreted it as the most p rimitive angiosperm derived from Ephederacee. But Hallier, Bessey Hutchinson and Tippo recognise it to be highly advanced family originating from Hamamelidaceae . Eames (1961) favoured a relationship of the Casuarinaceae with the Betulaceae. T his is based on the similarity of pollen structure, two-trace stamen character a

nd other floral aspects. The family is related to gymnosperms. The habit of Casuarina recalls that of Eph edra in the branches beaming alternating whorls of scale-like leaves. In both th e genera, the male and female flowers are simple and reduced. Thus, the Casuarin aceae plays an important role in the system which seeks the origin of the angios perms from gymnosperms. In fact, Wettstein postulated a theory of the developmen t of the angiosperm flowers from gymnosperm ancestors like Gnetales. Lam (1961) went to the extent of removing Casuarina from angiosperms, to be included in the Protoangiosperms along with Ephedra, Welwitschia and Gentum. Hutchinson (1969) conceived the Casuarinaceae "as an extreme reduction adopted m ostly to dryer- nnatic condition,but with no vary relationsstill living.. Casuarina on chemical correlation basis provides a mysterious correlation in pro ducing biflavonyls compounds atypical of angiosperms but similar to many conifer s, ginkgo and cyeads. It has an extremely low relative proportion of Syringyl Su bunits, found only in certain conifers like Podocarpous and not in angiosperms. In spite of some external morphological gymnospermous characters the family Cusu arinaceae is a typical angiospermous moderately advanced family. Common plant of the family 1. Casuarina equisetifolia Australian pine or Beef-wood tree is a big tree, usua lly planted in avenues. 88 ARECACEAE (PALMACEAE OR PALMAE) Classification : Bentham & Hooker Engler & Prantl Hutchinson Monocotyledons Monocotyledoneae Monocotyledons Calycinae Principes Corolliferae Arecaceae (Palmae) Arecaceae (Palmae) Palmales Arecaceae (Palmae) Diagnostic features Mainly trees with stout unbranched stem ending in crown of leaves; leaves large, compound, alternate, young leaves are plicate, exstipulate with long petioles; inflorescence enclosed in a persistent spathe; flowers unisexual; perianth 6 in two whorls of 3 each; in male flower 6 stamens in two whorls, anthers versatile; in female flowers carpels three; ariocarpous or syncarpous, superior, trilocula r or rarely unilocular; fruit berry or drupe; seed endospermic. Distribution The family is commonly known as "Palm family". It includes 217 genera and 2500 s pecies. The members are confined to tropics in both the hemispheres and extendin g in the warmer regions of the world. In India it is represented by 225 species belonging to 25 genera. Vegetative characters Habit Large unbranched trees (Phoenix, Areca catechu), shrubs or garden palms, t railing (Calamus), herbs (Reinhardtia). Root Adventitious roots arising from the base of bulbous stem. Thick aerial root s are also found in some species of Manicaria. Stem Aerial, woody, erect, unbranched, very rarely branched, (Hyphaene), in some short rhizome (Nipa), cylindrical, hairy, old stem protected by woody leaf base s, climbing (Calamus). Leaves Alternate crowded at the apex of stem giving palmlike appearance to the p lant; petiolate, leaf-base sheathing, broad and persistent; exstipulate, compoun d pinnately (Phoenix, Areca), palmately (Borassus), acute, thick, leathery, para llel venation. In some palms (Copernica) the petiole is prolonged into a ligule like structure called histula. Floral characters Inflorescence It is simple or compound, spike or branched panicle, usually a spa dix with a woody spathe which opens by two valves; spadix may have sessile or pe dicellate flowers, simple racemose (Borassus), or compound racemose (Cocos) or e ven profusely branched panicle (Daemonorops). (492) Arecaceae (Palmaceae or Palmae) 493

Flower Sessile or shortly pedicellate, bracteate, mostly unisexual (Phoenix) or hermaphrodite (Livingstonia), actinomorphic, incomplete or complete, hypogynous trimerous, flowers are of small size and produced in large numbers. Plant may be monoecious or dioecious. In monoecious flower the position of male and female f lowers is variable i.e. male flowers at the base or at the apex and the female f lowers at the upper part (Ruffia, Raphis) or male and female flowers are inter-m ingled or female flowers in the centre, male on the either side as the Cocos, Ca ryota. A- Entire plant Fig. 88.1. Areca catechu. 494 Plant Taxonomy Perianth Tepals 6, in two whorls of 3 each, polyphyllous or slightly connate at the base; perianth lobes tough, persistent, coriaceous, leathery or fleshy, valv ate or imbricate aestivation, white or petaloid. Androecium In male or hermaphrodite flowers, stamens are 6 in number, two whorls of 3 each, polyandrous, staminodes may be present in the female flowers; anther s versatile, dithecous, basifixed or dorsifixed, introrse, filament short and di stinct. Gynoecium In female or hermaphrodite flower carpels 3 in number, apocarpous or syn carpous, ovary superior, trilocular, axile placentation, single ovule in each lo culus; style short, stigma small or broad or 3 lobed. Fruit Usually a berry, fleshy or fibrous waxy coating on the fruit; the mature f ruit contains a single seed (Phoenix); drupe (Cocos nucifera). Seed Endospermic. Pollination Anemophilous or entomophilous. Floral formulae Male flower : 63, d' P3+3 A3+3 Go Female flower : m 9 P3+3 A0 Gil or 3 Hermaphrodite flower : m V P3+3 A3+3 Go_l or 3. Economic importance Food Pith of Metroxylon rumphii and M. laeve (Sago palm) yield sago of commerce. The sap of Borassus yields a sugar, which on fermentation gives alcoholic drink "Toddy'. Fruits of Phoenix dactylifera are very delicious and eaten throughout the Arab world. The nuts of Areca catechu serve as a asteringent and used with b etel leaves. The milk of Cocos nucifera makes a refreshing drink, endosperm is e aten raw and stored when dry. Medicinal Tender leaves of Calamus travancoricus are given in bilousness, worms and dyspepsia. Fibres Mesocarps of the drupes of Coconut are extensively used for stuffing pill ows and sofa sets. The cane of commerce is obtained from Calamus tenuis and C. r otang and are used for making mats, baskets and other furniture. Borassus flabellifer yields palmyra fibres which are used to prepare brushes and brooms. The leaves are used in the manufacture of hand fans, umbrellas, baskets and mats. Wax and oil Wax is obtained from the leaves of Copernicia cerifera and Ceroxylon andicola. The wax is used in making gramophone records, candles and models. Coconut oil is obtained from the Cocos nucifera and is used as hair oil, in soap industry and also for cooking. Ornamentals Roystonea regia (Royal palm), Corypha elata (Talipot palm). Primitive characters 1. Mostly plants are trees. 2. Leaves are spirally arranged. 3. Flowers are actinomorphic, hypogynous and hermaphrodite. 4. Gynoecium is apocarpous (Phoenix, Rhapis). 5. Ovary superior. Advanced characters 1. Small herbaceous forms are also present. 2. Leaves are compound and exstipulate. 3. Inflorescence is a spadix. Arecaceae (Palmaceae or Palmae) 495 4. Perianth is present.

5. Flowers are usually unisexual (Phoenix, Cocos). 6. Flowers trimerous. 7. Stamens epiphyllous. 8. Gynoecium tricarpellary, syncarpous rarely unilocular. 9. Style very short or absent. 10. Axile placentation. According to Eames (1961) "The palms give evidence of great age; they are primit ive taxon that has become greatly diversified and advanced in many characters, e ach character giving evidence of long specialization." Affinities of the family Rendle placed the family together with the Araceae under Spadiciflorae due to un isexual flowers and occurrence of spadix. Hutchinson (1959) traces the origin of Palms from Liliflorean stock directly fro m Liliaceae through Dracaena Cordyline. Erdtman also reports similar pollen struct ure of Palms and Dracaena. Palmaceae is closely related to Liliaceae in palm-like habit of Yucca, Dracaena of Liliaceae, perianth segments, stamens in two whorls, tricarpellary, syncarpou s ovary, and structure of pollen grains (Dracaena). Common plants of the family 1. Areca catechu (H. Supari; Betelnut palm) : Graceful single stemmed palm. 2. Caryota urens (Fish-tail palm) : Toddy is tapped from its stem. 3. Corypha umbraculifera (Talipot palm) : Planted in gardens. 4. Cocos nucifera (H. Nariyal) : a tall palm, widespread along sea shore in trop ics and sub-tropics. 5. Calamus tenuis and C. rotang Bent) : climbing palm. 6. Metroxylon : Fruits take 3 years to mature and pith yields 'sago'. 7. Nipa fruitcans (Water coconut) : palm with delicate round leaves used as ciga rette paper; stemless palm of Sunderbans. 8. Phoenix dactylifera (Date palm) : tall palm with rough trunk due to persisten t leaf bases; fruits are delicious. Division of the family and chief genera The family Arecaceae is divided into seven tribes as follows : Tribe 1. Areceae Leaves pinnatisect; flowers hermaphrodite; fruit a berry or fib rous drupe e.g. Areca. Tribe 2. Borassieae Leaves pinnatisect; flowers unisexual, e.g. Borassus. Tribe 3. Cococeae Leaves pinnatisect; flowers hermaphrodite, fruit fibrous drupe not covered with scales e.g. Cocos. Tribe 4. Phoeniceae Leaves pinnatisect; flowers unisexual; fruit one seeded berr y e.g. Phoenix. Tribe 5. Phytelephanteae Leaves pinnatisect; male and female inflorescences sepa rate e.g. Nipa. Tribe 6. Lepidocaryeae Leaves palmatisect or pinnatisect; spadices terminal or i nterfoliar e.g. Calamus. Tribe 7. Sabaleae Leaves palmatisect; flowers unisexual or bisexual, e.g. Coryph a. 496 Plant Taxonomy Important type Phoenix sylvestris (Date palm)' (Fig. 88.2) Habit A tree. INFLORESCENC WITH SUBTENDING SPATE OPENED OUT 0 INFLORESCENCE

WITH SUBTENDING SPATHE OPENED OUT dINFLORESCENCE 9 INFLORESCENCE e K(3) C3 Ao G3 +" OR . P(3)+3 CCO K(3) C3A3+ 3 GO OR P,3i.3 Fig. 88.2. Phoenix sylvestris. Arecaceae (Palmaceae or Palmae) 497 Root Adventitious. Stem Aerial, woody, erect, cylindrical, rough, covered with persistent leaf base s, unbranched, solid, brown. Leaves Forming a dense terminal crown, exstipulate, compound, unipinnate, petiol ate, glabrous. Leaflets Sub-sessile, lanceolate, entire, acute, unicostate parallel venation. Inflorescence Spadix-branched, erect, long, enclosed by spathe. Flower Small, actinomorphic, hypogynous, unisexual, bracteate, incomplete. Male Flower Bracteate, sessile, incomplete, numerous, angular, actinomorphic, hypogynous, trimerous. Perianth Tepals 6, in two whorls of 3 each, white, angular, free and inferior. Androecium Stamens 6, in two whorls of 3 each, polyandrous, filament short; anth ers dithecous, dorsifixed, introrse. Gynoecium Absent. Female flower Bracteate, sessile, incomplete, actinomorphic, hypogynous, trimero us. Perianth As in male flower. Androecium Absent. Gynoecium Tricarpellary, syncarpous, ovary superior, one ovule in each carpel; s tyle absent; stigma hooked. Fruit One seeded berry, orange yellow. Seed Hard and endospermic. Floral formulae Male flower : ED 8' P3+3 A3+3 Go Female flower : O y P3+3 k = 89 TYPHACEAE (Water Grass or Cat-tail Family) Classification : Bentham & Hooker Engler & Prantl Hutchinson Monocotyledons Monocotyledoneae Monocotyledons Nudiflorae Pandanales Typhales Typhaceae Typhaceae Typhaceae Diagnostic characters Herb tall of marshy places, perennial, leaves erect, linear, distichous, thick a nd spongy; flowers monoecious, minute, densely crowded in cylindrical bracteate spikes; perianth of 5 hair or 0; male flower with stamens 2-5, filament capillar y, anther erect, bi-celled; female flower ovary 1, unilocular; fruits nutlets; s eeds endospermic. Distribution Small family of monotypic genus Typha and with _15-20 species, which are distrib uted both in temperate and tropical regions. Vegetative characters Habit Aquatic or marshy, perennial herbs occurring in ponds, ditches, lakes on r iver banks. Root Adventitious, fibrous. Stem Rhizome long, creeping, thick, clothed.

Leaf Scale leaves in two lateral rows on the rhizome, in their axils arise aeria l shoots bearing two-ranked foliage leaves with narrow, linear, thin and spongy, more or less twisted lamina and sheathing leaf-base. The lower leaves reduced t o long sheaths represent a transitory stage between the aerial and scale leaves on the rhizome. Floral characters Inflorescence Spike, bearing yellow male flowers above and brownish female flowe rs below, often intermixed with hair like bracteoles with dilated tips. Flower Small, unisexual, each subtended by one caducous bract like spathe and encl osed by hairs or scales which according to some botanists represent reduced perianth. Male flower Perianth Represented by 3 or 6 hair like outgrowth free, inferior or absent (Typha minima). Androecium Stamens two to five or reduced to one, e.g., Typha minima, variously monadelphous; anthers basifixed, linear, connective long, projecting beyond the anthers; pollen grains simple or in tetrads. Female flower Perianth As in male flower but represented by many one celled hair. (498) Typhaceae (Water Grass and Cat-tail Family) 499 Gynoecium Bracteate or ebracreate, e.g., Typha latifolia; carpel one, unilocular with a single, pendulous ovule; gynophore long beset with numrous hairs which h elp in the dispersal of fruit; style capillary with a narrow or ligulate stigma, clavate tipped, pistillodes frequent among the flowers in some species. CREEPING RHIZOME 0 ct Po A(3) Go FLORAL DIAGRAM MALE FLOWER A PLANT PROLONGED CONNECTIVE STAMINAT FLOWERS ANTHER GYNOPHORE OVULE BRACT HAIRS PISTILLATE FLOWER STAMINATE FLOWER 0 Po Ao Gi FLORAL DIAGRAM FEMALE FLOWER Fig. 89.1. Typha angusta. PISTILLATE FLOWERS 500 Plant Taxonomy Fruit An achene. Seed Endospermic, small. Pollination Anemophilous. Floral formulae Male flower : as d' PO or Prepresented by hair A1-5 GO Female flower : 9 1:), Ao G1. Economic importance Medicinal The root stock of T angustata is astringent and diuretic. The down of the ripe fruit of T elephantina is used as an application to wounds and ulcers, which acts as a medicated cotton. The root stocks are used in measels, dysentery and gonorrhoea. The stamens and pollen grains of T laxmanni are astringent and styptic. General The leaves of T elephantina are used for thatching. The leaves of T lati folia are used to manufacture baskets, mats, and seats of chair.' According to C ook (1958) the flowers of T angustata are used in the preparation of a yellow ca ke called Bur and is eaten by the natives all over Sind area. Affinities of the family

Bentham and Hooker placed the family in the series Nudiflorae while Engler and R endle placed it in series Pandanales. According to Engler and Rendle it is the f irst and most primitive in the Monocotyledons because of simple unisexual flower s consisting of one for few essential organs; where as Hutchinson placed it in s eparate order Typhales with Sp'arganiceae after Arales, he regards Typhales as v ery reduced order derived fromiiliaceae on parallel line with Araceae. Typhacea e has some affinity with Araecae. Common plant of the family 1. Typha elephantina growing in marshes. 2. T angustata gregarious, found throughout India. 92 POACEAE (GRAMINEAE) Classification : Bentham & Hooker Engler & Prantl Hutchinson Monocotyledons Monocotyledoneae Monocotyledons Glumaceae Glumiflorae Graminales Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Diagnostic features Mostly herbs, stem jointed, fistular, cylindrical; leaves simple, alternate, she athing, sheath open, ligulate; inflorescence compound spike; flowers zygomorphic , hypogynous, protected by palea; perianth represented by 2 or 3 minute scales ( lodicules); stamens 3, versatile; carpel one, style 2 or 3, stigmas feathery, ba sal placentation; fruit caryopsis; testa fused with pericarp. Distribution The family is commonly known as grass family. It is one of the largest among the angiospermic families. It consists of 620 genera and 6,000 species. The members are cosmopolitan in distribution. The plants represent all the 3 ecological typ es as hydrophytes, xerophytes and mesophytes. In India it is represented by 850 species. Vegetative characters Habit Herbs, annuals or perennials or shrubs, sometimes tree like (Bambusa, Dend rocalamus). Root Adventitious, fibrous, branched, fascicled or stilt (Zea mays). Stem Underground rhizome in all perennial grasses, cylindrical, culm with conspi cuous nodes and internodes, internodes hollow, herbaceous or woody, glabrous or glaucous, vegetative shoots are arising from the base of aerial stem or from und erground stems are called tillers. Leaves Alternate, simple, distichous, exstipulate, sessile, ligulate (absent in Echinochloa), leaf base forming tubular sheath, sheath open, surrounding interno de incompletely, ligule is present at the junction of the lamina and sheath, ent ire, hairy or rough, linear, parallel venation. Floral characters Inflorescence Compound spike which may be sessile or stalked. Each unit of inflo rescence is spikelet. The spikelets are arranged in various ways on the main axi s called rachilla. A compound inflorescence may be spike of spikelets (Triticum) , panicle of spikelets (Auena). The spikelet consists of a short axis called rachilla on which 1 to many sessile or short stalked flowers ,are borne. The florets may be arranged in alternate o r opposite manner on the central axis. (509) 510 Plant Taxonomy At the base of rachilla two sterile scales, called glumes, are present. The glum es are placed one above the other on opposite sides. The lower one is called fir st glume and the upper is called second glume. Both the glumes are boat shaped a nd sterile. Above the glumes a series of florets are present. Each floret has an inferior palea or lemma and above it a superior palea. The lemma frequently bea rs a long, stiff hair called awn. Flower Bracteate and bracteolate, sessile, incomplete, hermaphrodite, or unisexu al (Zea mays), irregular, zygomorphic, hypogynous, cyclic. Perianth Represented by membranous scales called the lodicules. The lodicules ar

e situated above and opposite the superior palea or may be absent, or many (Ochl andra), or 2 or 3. Androecium Usually stamens 3, rarely 6 (Bambusa, Oryza) and one in various speci es of Anrostis, Lepturus; polyandrous, filaments long, anthers dithecous, versat ile, linear, extrorse; pollen grains dry. Gynoecium Monocarpellary, according to some authors carpels 3, of which 2 are ab ortive, ovary superior, unilocular with single ovule, basal placentation, style short or absent; stigmas two feathery or papillate and branched. Fruit Caryopsis (achene with pericarp completely united or adherent with the see d coat) or rarely nut (Dendrocalamus) or berry (Bambusa). Seed Endospermic and containing a single cotyledon called scutellum, which is sh eild shaped and pressed against the endosperm. Floral formula 010 PO or 2 (Lodicules) A3 or 6 G1. Economic importance The family stands first and foremost in respect of economic importance in whole of Angiosperms. The staple food grains of the population of world is derived fro m Oryza sativa (Rice) and Triticurn aestivum (Wheat). They are cultivated from t ime immemorial. The family has been divided on economic basis as follows : Food Triticum aestivum, Oryza sativa, Zea mays (Maize), Hordeum vulgare (Jaw), S orghum vulgare (Jowar), Avena sativa (Oats), Pennisetum typhoides (Bajra) are cu ltivated for cereals and food grains. Fooder Many grasses as Cynodon dactylon, Panicurn, Cymbopogon, Agrostis, Poa are grown for fodder. Sugar Saccharum officinarum (Sugarcane; H. Ganna) is cultivated for gur and suga r. Building material Some species of Bambusa e.g. B. tulda, B. vulgaris are used for scaffolding, thatching huts etc. Furniture Species of Dendrocalamus (H. Bent), Arundinaria, Melocalamus are used in manufacture of furniture. Aromatic grasses Many grasses yield scented oils which are used in perfumery viz . Vetiveria zizanioides (H. Khus khus) yields vetiver oil from the roots. The ro ots are also woven into curtains. Andropogon odoratus (Ginger grass), Cymbopogon citratus (Lemon grass), Cymbopogon martini (Geranium grass), Cymbopogon jawaran cusa etc. also yield oil. Medicinal Phragmites karka, Cymbopogon schoenanthus etc. are medicinal. Secale cereale is cultivated for infection of its inflorescence by Claviceps pur purea for production of Ergot and for extraction of ergotine. Ergotine is an exc ellent remedy for uterine contraction. Paper It is manufactured from certain spe cies of grasses and bamboos. Ornamental Rhynchelytrum repens, Cortaderia selloana and some species of the tri be Bambusoideae are ornamentals. Besides these a number of grasses are grown to form fine lawns, play grounds etc . Poaceae (Gramineae) 511 Primitive characters 1. A few plants arboreal in habit. 2. All florets in a spikelet are fertile. 3. Glumes are persistent. 4. Lemmas are herbaceous and leafy. 5. Stigmas are three. 6. Leaves are simple and alternate. 7. Flowers are hypogynous and hermaphrodite. 8. Seeds are endospermic. Advanced characters 1. Plants are mostly herbaceous, annuals and perennials. 2. Leaves are exstipulate. 3. Flowers are arranged in distinct inflorescence. 4. Flowers are small, incospicuous and zygomorphic. 5. Perianth is represented by lodicules. 6. Stamens are reduced to 3. 7. Gynoecium is monocarpellary and unilocular.

8. Basal placentation. 9. Fruit is caryopsis. 10. Seeds are small sized. Affinities of the family The family Poaceae (Gramineae) closely resembles the family Cyperaceae and the t wo families have been placed in same order Glumiflorae by Engler and Prantl, and Glumaceae by Bentham and Hooker. Hutchinson (1964), butzin (1965) and Takhtajan (1969) placed the families into 2 separate orders, the Cyperales and Graminales on the basis of many differences viz., 1. leaf sheath, jointed and unjointed stem, 3. single bract and lemma and palea, 4. seed coat et c. Cronquist (1968) considered that the families are also related on account of som e chemical data. Hegnauer (1963) has also supported the same view. Hutchinson (1959) believes that the origin of Grasses i.e., the family Poaceae ( Gramineae) took place on parallel line with Cyperaceae through Juncaceae which w as derived directly from Liliforean stock. Division of the family and chief genera The family is divided into 2 sub-families : Sub family I. Poideae Spikelets one to many flowered; mature spikelets break up an d fall above the persistent glumes leaving them intact, if spikelets falling ent ire then not 2 flowered or if 2 flowered falling entire; the glumes and lemmas all m embranous with wrinkled seed. Rachilla is continued above the floret. This sub fam ily is divided into 8 tribes. Sub family II. Panicoideae Spikelets mostly two flowered, rarely one flowered, fal ling entire at maturity with their stalks and glumes. In a spikelet one floret i s perfect i.e. hermaphrodite, the other is sterile or male usually. Rachilla is not continued above the upper floret. This sub family is divided into 3 tribes. Important types 1. Triticum aestivum (Fig. 92.1) Habit A cultivated annual crop plant. Plant Taxonomy Root Adventitious. Stem Herbaceous, erect, cylindrical, fistular, with distinct nodes and internode s, unbranched, glabrous, a number of tillers. Leaf Simple, alternate, green, exstipulate, entire margin, acute apex, sheathing leaf base, at the junction of leaf-sheath and leaf-blade membranous ligule pres ent, parallel venation. Inflorescence Spike of spikelets. Flower Bracteate, sessile, hermaphrodite, zygomorphic, incomplete, hypogynous, f lower lies between superior and inferior palea. Perianth 2 membranous scales the lodicules. Androecium Stamens 3, polyandrous, filament long, anthers dorsifixed when young and versatile when mature. Gynoecium Monocarpellary, theoritically tricarpellary, ovary superior, unilocula r, single ovule, basal placentation, style short; stigma 2, feathery. STIGMA 0, II STEM. A Flowering twig LAMINA Floret with lemma and pal ea removed LEMMA Fig. 92.1. Triticum aestivum. AWN FLORET L EMMA GLUMES STIGMA STAMEN ---- OVARY PALEA RACHILLA C-A Floret Opened out e-sr%..1.4

SPIKE OF SPIKELET PALE A Poaceae (Gramineae) 513 Floral formula :oho # P2 2 (lodicules) A3 G1. 2. Zea mays Habit An annual cultivated crop plant. Root Adventitious stilt roots. Stem Erect, cylindrical, herbaceous, solid, distinct nodes and internodes, smoot h, green. Leaf Alternate, simple, long ligule present, linear, entire margin, green, multi costate parallel venation, sheathing leaf base. FLOWERING GLUME FLOWERING GLUME EMPTY GLUMES FLOWERING GLUME EMPTY GLUMES Fig. 92.2. Oryza sativa. 514 Plant Taxonomy Inflorescence Plant monoecious, diclinous; two kinds of flowers in different inf lorescences; male in staminate and female in pistillate inflorescence respectively. Male flower Each spikelet has two flowers one pedicellate other sessile; each has a pair of bracts the inferior and superior palea, zygomorphic. Perianth Represented by 2 scales the lodicules. Androecium Stamen 3, linear, versatile, extrorse. Gynoecium Rudimentary or absent. Female flower Each spikelet is two flowered and subtended by two glumes the ster ile and fertile. Each flower is enclosed by inferior and superior palea, zygomorphic . Perianth Represented by 2 lodicules. Androecium Absent. Gynoecium Monocarpellary, unilocular, (theoritically tricarpellary) superior, si ngle ovule, basal placentation; style filiform, long; silky, stigma hairy. Floral formulae Male flower : 010 6 P2 (lodicules) A3 GO Female flower : 9 P2 (lodicules) AO Gl. 3. Oryza sativa (Fig. 92.2) Habit A cultivated herb for food. Root Adventitious, fibrous roots. Stem Herbaceous, erect, cylindrical, fistular, nodes and internodes are present, green. Leaf Long, narrow, with a sheathing base, ligulate, membranous, linear to lanceolate, entire, parallel venation. Inflorescence Panicle of spikelets, Spikelet one loosely arranged on the branche s of the panicle. At the base of the each spikelet there are two small, membrano us, persistent bracts called the empty Glumes; above them lies the single flower which is enveloped in two larger bracts. The lower one (in the axil of which th e flower develop) is the flowering glume, which may or may not have at its apex a narrow elongated outgrowth called the awn; the upper one insterted at little a bove the flowering glume is called Palea. Flower Bracteate (Lemma or inferior palda) pedicellate, complete, hermaphrodite, zygomorphic, small hypogynous. Perianth Represented by 2 scales called lodicules, fleshy and small. Androecium Stamens six, polyandrous, anthers linear, versatile, introrse. Gynoecium Monocarpellary, ovary superior, unilocular with only one ovule. Style 2, short; stigma 2 feathery and lateral. Fruit A caryopsis completely enclosed by the flowering glume and palea. Floral formula Br. Brl 010 V P2 As G1.