These

lady-bird beetles,
from the Chiricahua
Mountains in Arizona,
show considerable
phenotypic variation.

25 ■
CHAPTER CONCEPTS
Most populations and species harbor considerable genetic
variation.
Population ■ This variation is reflected as allele differences distributed
among populations of a species.
and Evolutionary ■ The relationship between allele frequencies and genotype
frequencies in an ideal population is described by the
Genetics ■
Hardy–Weinberg Law.
Selection, migration, and genetic drift can cause changes in
allele frequency.
■ Mutation creates new alleles in a population’s gene pool.
■ Nonrandom mating changes population genotype frequency
but not allele frequency.
■ A reduction in gene flow between populations, accompanied
by selection or genetic drift, can lead to reproductive isola-
tion and speciation.
■ Genetic differences between populations or species are used
to reconstruct evolutionary history.
I
698 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

n the mid-nineteenth century, Alfred Russel Wallace 25.1

and Charles Darwin identified natural selection as
the mechanism of evolution. In his 1859 book, On the
Genetic Variation Is Present
Origin of Species, Darwin presented comprehensive in Most Populations and Species
evidence that populations and species are not fixed,
but change, or evolve, over time as a result of natural se- A population is a group of individuals belonging to the
lection. However, Wallace and Darwin could not explain same species that live in a defined geographic area and actu-
either the origin of the variations that provide the raw ma- ally or potentially interbreed. In thinking about the human
terial for evolution or the mechanisms by which such varia- population, we can define it as everyone who lives in the
tions are passed from parents to offspring. Gregor Mendel United States, or in Sri Lanka, or we can specify a popula-
published his work on the inheritance of traits in 1866, but tion as all the residents of a particular small town or village.
it received little notice at the time. The rediscovery of Men- The genetic information carried by members of a
del’s work in 1900 began a 30-year effort to reconcile the population constitutes that population’s gene pool. At first
concept of genes and alleles with the theory of evolution by glance, it might seem that a population that is well adapted
natural selection. As twentieth-century biologists applied to its environment must be highly homozygous because you
the principles of Mendelian genetics to populations, both assume that the most favorable allele at each locus is present
the source of variation (mutation) and the mechanism of at a high frequency. In addition, a look at most populations
inheritance (segregation of alleles) were explained. We now of plants and animals reveals many phenotypic similarities
view evolution as a consequence of changes in alleles and among individuals. However, a large body of evidence in-
allele frequencies in populations over time. This union of dicates that, in reality, most populations contain a high de-
population genetics with the theory of natural selection gree of heterozygosity. This built-in genetic diversity is often
generated a new view of the evolutionary process, called concealed because it is not necessarily apparent phenotypi-
neo-Darwinism. cally; hence, detecting it is not a simple task. Nevertheless,
In addition to natural selection, other forces including the diversity within a population can be revealed by several
mutation, migration, and drift, individually and collectively, methods.
alter allele frequencies and bring about evolutionary diver-
gence that eventually may result in speciation, the formation Detecting Genetic Variation
of new species. Speciation is facilitated by environmental by Artificial Selection
diversity. If a population is spread over a geographic range
One way to determine whether genetic variation affects a
encompassing a number of ecologically distinct subenvi-
phenotypic character is to use artificial selection. A pheno-
ronments with different selection pressures, isolated pop-
type that is not associated with variation in the genes that
ulations occupying these areas may gradually adapt and
control the phenotype will not respond to selection; if ge-
become genetically distinct from one another. Genetically
netic variation does exist, the phenotype can change over
differentiated populations may remain in existence, become
a few generations. A dramatic example of this test is the
extinct, reunite with each other, or continue to diverge until
domestic dog. The broad array of sizes, shapes, colors, and
they become reproductively isolated. Populations that are
behaviors seen in different breeds of dogs all arose from the
reproductively isolated are regarded as separate species. Ge-
effects of selection on the genetic variation present in wild
netic changes within populations can modify a species over
wolves, from which all domestic dogs are descended. Genetic
time, transform it into another species, or cause it to split
and archaeological evidence indicates that the domestica-
into two or more species.
tion of dogs took place at least 15,000 years ago and possibly
Population geneticists investigate patterns of genetic
much earlier. On a shorter time scale, laboratory selection
variation within and among groups of interbreeding indi-
experiments on the fruit fly Drosophila melanogaster have
viduals. Because mutations change the genetic structure of
generated significant changes over a few generations in al-
populations and form the basis for evolutionary change,
most every phenotype imaginable, including size, shape,
population genetics has become an important subdisci-
developmental rate, fecundity, and behavior.
pline of evolutionary biology. In this chapter, we examine
the processes of microevolution—defined as evolution-
ary change within populations of a species—and then Variations in Nucleotide Sequence
consider how molecular aspects of these processes can be The most direct way to estimate genetic variation is to com-
extended to macroevolution—defined as evolutionary pare the nucleotide sequences of genes carried by individu-
events leading to the emergence of new species and other als in a population. To do this, Martin Kreitman analyzed
taxonomic groups. variation in the alcohol dehydrogenase gene (Adh) in Dro-
 2 5. 1 GE NE TIC VA RIA TION IS PRE S E NT IN MOS T POPU LATIONS AND S PE C I ES 699

100 bp Polyadenylation Exon 3 Intron 3 Exon 4

site Consensus
Exon 1 Exon 2 Exon 3 Exon 4 Adh sequence: C C C C GGA A T C T C C A* C T A G
5’ 3’
Strain
Protein-encoding Wa-S T T • A CA • T A AC • • • • • • •
region Fl1-S T T • A CA • T A AC • • • • • • •
3’ untranslated
region Ja-S • • • • • • • • • • • • T • T • CA
Fl-F • • • • • • • • • • • GT C T CC •
FI G U R E 2 5 –1 Organization of the Adh locus of Drosophila Ja-F • • A • • • G • • • • GT C T CC •
melanogaster.
© 1983 Macmillan Publishers Ltd.

FIGUR E 25–2 DNA sequence variation in parts of the

sophila melanogaster (Figure 25–1). At the protein level, this Drosophila Adh gene in a sample of the 11 laboratory strains
gene has two alleles, Adh-f and Adh-s. The encoded proteins derived from the five natural populations. The dots represent
differ by only a single amino acid (thr versus lys at codon nucleotides that are the same as the consensus sequence; letters
represent nucleotide polymorphisms. An A/C polymorphism
192). To determine whether the amount of genetic variation
(A*) in codon 192 creates the two Adh alleles (F and S). All
detectable at the protein level (one amino acid difference) other polymorphisms are silent or noncoding.
corresponds to the variation at the nucleotide level, Kreitman
cloned and sequenced Adh genes from five natural popula- and lungs, leading to susceptibility to bacterial infections.
tions of Drosophila (Figure 25–2). Almost 1800 different mutations in the CFTR gene have
The 11 cloned genes isolated from these five popula- been identified. Among these are missense mutations, ami-
tions contained a total of 43 nucleotide variations in the no acid deletions, nonsense mutations, frameshifts, and
Adh sequence of 2721 base pairs. These variations are dis- splice defects.
tributed throughout the gene: 14 in exon-coding regions, Figure 25–3 shows a map of the 27 exons in the CFTR
18 in introns, and 11 in the untranslated flanking regions. gene, with many exons identified by function. The histo-
Of the 14 variations in coding regions, only one leads to an gram above the map shows the locations of some of the
amino acid replacement—the one in codon 192, produc- disease-causing mutations and the number of copies of each
ing the two alleles. The other 13 coding-region nucleotide mutation that have been identified. One mutation, a 3-bp
substitutions do not change the encoded amino acid, and as deletion in exon 10 called F508 accounts for 67 percent of
such, are silent variations in this gene. all mutant cystic fibrosis alleles, but several other mutations
As another example, consider the CF gene, which encodes were present in at least 100 of the chromosomes surveyed.
the cystic fibrosis transmembrane conductance regulator In populations of European ancestry, between 1 in 44 and 1
(CFTR), one of the most intensively studied human genes. in 20 individuals are heterozygous carriers of mutant
Recessive loss-of-function mutations in the CFTR gene alleles. Note that Figure 25–3 includes only the sequence
cause cystic fibrosis, a disease that affects secretory glands variants that alter the function of the CFTR protein. There are

10,000 © 1992 Elsevier

F508
Number of copies of

1,000
mutant alleles

100 FIGUR E 25–3 The locations of

disease-causing mutations in the
cystic fibrosis gene. The histogram
10 shows the number of copies of each
mutation geneticists have found.
(The vertical axis is on a logarithmic
1
scale.) The genetic map below the
5’ 3’ histogram shows the locations and
12 3 4 5 6a 6b 7 8 9 10 1112 13 14a 14b 15 16 17a 17b 18 19 20 21 22 23 24
relative sizes of the 27 exons of the
CFTR locus. The boxes at the bottom
Membrane ATP R-domain Membrane ATP indicate the functions of different
spanning binding spanning binding domains of the CFTR protein.
700 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

undoubtedly many more CFTR alleles with silent sequence genetic variation is neutral and what fraction is subject to se-
variants that do not change the amino acid sequence of lection. The neutral theory nonetheless serves a crucial func-
the protein and do not affect its function. tion: It points out that some genetic variation is expected
Studies of other organisms, including the rat, the simply as a result of mutation and drift. In addition, the
mouse, and the mustard plant Arabidopsis thaliana, have neutral theory provides a working hypothesis for studies of
produced similar estimates of nucleotide diversity in vari- molecular evolution. In other words, biologists must find
ous genes. The results point to the presence of an enormous positive evidence that selection is acting on allele frequen-
reservoir of genetic variability within most populations and cies at a particular locus before they can reject the simpler
show that, at the DNA level most, and perhaps all, genes ex- assumption that only mutation and drift are at work.
hibit diversity from individual to individual. Alleles repre-
senting these variations are distributed among members of 25.2
a population.
The Hardy–Weinberg Law
Explaining the High Level of Genetic Describes Allele Frequencies
Variation in Populations and Genotype Frequencies
The finding that populations harbor considerable genetic in Populations
variation at the amino acid and nucleotide levels came as
a surprise to many evolutionary biologists. The early con- Populations are dynamic; they expand and contract through
sensus had been that selection would favor a single optimal changes in birth and death rates, migration, or contact with
(wild-type) allele at each locus and that, as a result, popula- other populations. Often some individuals within a popu-
tions would have high levels of homozygosity. This expecta- lation will produce more offspring than others, contribut-
tion was shown conclusively to be wrong, and considerable ing a disproportionate fraction of their alleles to the next
research and argument has ensued concerning the forces generation. Thus, differential reproduction in a population
that maintain such high levels of genetic variation. can, over time, lead to changes in the allele and genotype
The neutral theory of molecular evolution, proposed frequencies in subsequent generations. Changes in allele fre-
by Motoo Kimura in 1968, proposes that mutations lead- quencies in a population that do not result in reproductive
ing to amino acid substitutions are usually detrimental, with isolation are examples of microevolution. In the following
only a very small fraction being favorable. Some mutations sections, we will discuss microevolutionary changes in pop-
are neutral, that is, they are functionally equivalent to the al- ulation gene pools, and later in this chapter we will consider
lele they replace. Mutations that are favorable or detrimental macroevolution and the process of speciation.
are preserved or removed from the population, respectively, Often when we examine a single genetic locus in a pop-
by natural selection. However, the frequency of the neutral ulation, we find that the distribution of alleles at this locus
alleles in a population will be determined by mutation rates produces individuals with different genotypes. The calcula-
and random genetic drift, and not by selection. Some neu- tion of allele frequencies and genotype frequencies in the
tral mutations will drift to fixation in the population; other population, and the determination of how these frequencies
neutral mutations will be lost. At any given time, a popula- change from one generation to the next are key elements
tion may contain several neutral alleles at any particular of population genetics. Population geneticists use these cal-
locus. The diversity of alleles at most loci does not, there- culations to answer questions such as: How much genetic
fore, reflect the action of natural selection, but instead is a variation is present in a population? Are genotypes randomly
function of population size (larger populations have more distributed in time and space, or do discernible patterns ex-
variation) and the fraction of mutations that are neutral. ist? What external and internal factors affect the composi-
The alternative explanation for the surprisingly high tion of a population’s gene pool? Do these factors produce
genetic variation in populations is natural selection. There genetic divergence among populations that may lead to the
are several examples in which enzyme or protein variations formation of new species?
are maintained by adaptation to certain environmental con- The relationship between the relative proportions of al-
ditions. The well-known advantage of sickle-cell anemia leles in the gene pool and the frequencies of different geno-
heterozygotes when infected by malarial parasites is such an types in a population was elegantly described in the early
example. 1900s in a simple mathematical model developed indepen-
Fitness differences of a fraction of a percent would be dently by the British mathematician Godfrey H. Hardy
sufficient to maintain such a variation, but at that level their and the German physician Wilhelm Weinberg. This model,
presence would be difficult to measure. Current data are called the Hardy–Weinberg Law, describes what happens
therefore insufficient to determine what fraction of molecular to alleles and genotypes in an “ideal” population that is
25.2 T H E H ARD Y–W EIN B ERG LAW D ESCRIB ES A LLE LE FRE QU E NC IE S A ND G E NOTYPE FRE QU E NC IE S IN POPU LA TIO N S 701

infinitely large with random mating and is not subject to Sperm

any evolutionary forces such as mutation, migration, or se- fr(A) = 0.7 fr(a) = 0.3
lection. Under these conditions, the Hardy–Weinberg model
fr(AA) = fr(Aa) =
makes two predictions: fr(A) = 0.7 0.7  0.7 0.7  0.3
= 0.49 = 0.21
1. The frequencies of alleles in the gene pool do not change
Eggs
over time.
fr(aA) = fr(aa) =
2. If two alleles at a locus, A and a, are considered, then as fr(a) = 0.3 0.3  0.7 0.3  0.3
we will show later in this chapter, after one generation = 0.21 = 0.09
of random mating, the frequencies of the genotypes
AA:Aa:aa in the population can be calculated as
FIGUR E 25–4 Calculating genotype frequencies from allele
p2 + 2pq + q2 = 1 frequencies. Gametes represent samples drawn from the gene
pool to form the genotypes of the next generation. In this popu-
where p = frequency of allele A and q = frequency of lation, the frequency of the A allele is 0.7, and the frequency of
allele a. the a allele is 0.3. The frequencies of the genotypes in the next
generation are calculated as 0.49 for AA, 0.42 for Aa, and 0.09
A population that meets these criteria, and in which the fre- for aa. Under the Hardy–Weinberg Law, the frequencies of A and
quencies of p and q, two alleles at a given locus result in a remain constant from generation to generation.
the predicted genotypic frequencies, is said to be in Hardy–
Weinberg equilibrium. As we will see later in this chapter, it
is rare for a real population to conform totally to the Hardy– the frequency distribution of alleles in the new gene pool?
Weinberg model and for all allele and genotype frequencies Under the assumptions of the Hardy–Weinberg Law, we
to remain unchanged for generation after generation. assume that all genotypes have equal rates of survival and
The Hardy–Weinberg model uses the Mendelian prin- reproduction. This means that in the next generation, all
ciples of segregation along with simple probability to explain genotypes contribute equally to the new gene pool. The AA
the relationship between allele and genotype frequencies individuals constitute 49 percent of the population, and we
in a population. To demonstrate how this works, we will can predict that the gametes they produce will constitute 49
consider a single autosomal locus with two alleles, A and percent of the gene pool. These gametes all carry allele A.
a, in a population where the frequency of A is 0.7 and the Similarly, Aa individuals constitute 42 percent of the popu-
frequency of a is 0.3. Note that 0.7 + 0.3 = 1, indicating lation, so we predict that their gametes will constitute 42
that all the alleles for gene A present in the gene pool are percent of the new gene pool. Half (0.5) of these gametes
accounted for. We assume that individuals mate randomly, will carry allele A. Thus, the frequency of allele A in the gene
following Hardy–Weinberg requirements, so for any one zy- pool is 0.49  (0.5)0.42  0.7. The other half of the gametes
gote, the probability that the female gamete will contain A produced by Aa individuals will carry allele a. The aa indi-
is 0.7, and the probability that the male gamete will contain viduals constitute 9 percent of the population, so their gam-
A is also 0.7. The probability that both gametes will con- etes will constitute 9 percent of the new gene pool. All these
tain A is 0.7 * 0.7 = 0.49. Thus we predict that genotype gametes carry allele a. Thus, we can predict that the allele a
AA will occur 49 percent of the time. The probability that in the new gene pool is (0.5)0.42  0.09  0.3. As a check
a zygote will be formed from a female gamete carrying A on our calculation, note that 0.7 + 0.3 = 1.0, accounting
and a male gamete carrying a is 0.7 * 0.3 = 0.21, and the for all of the gametes in the gene pool of the new generation.
probability of a female gamete carrying a being fertilized We have arrived where we began: with a gene pool
by a male gamete carrying A is 0.3 * 0.7 = 0.21, so the where the frequency of allele A is 0.7 and the frequency of
frequency of genotype Aa is 0.21 + 0.21 = 0.42 = 42 allele a is 0.3. For the general case of the Hardy–Weinberg
percent. Finally, the probability that a zygote will be formed model, we use variables instead of numerical values for
from two gametes carrying a is 0.3 * 0.3 = 0.09, so the the allele frequencies. Imagine a gene pool in which the
frequency of genotype aa is 9 percent. As a check on our cal- frequency of allele A is p and the frequency of allele a is q,
culations, note that 0.49 + 0.42 + 0.09 = 1.0, confirming such that p + q = 1. If we randomly draw male and female
that we have accounted for all of the zygotes. These calcula- gametes from the gene pool and pair them to make a zygote,
tions are summarized in Figure 25–4. the probability that both will carry allele A is p  p. Thus,
We started with the frequency of a particular allele in the frequency of genotype AA among the zygotes is p2. The
a specific gene pool, and we calculated the probability that probability that the female gamete carries A and the male
certain genotypes would be produced from this pool. When gamete carries a is p  q, and the probability that the female
the zygotes develop into adults and reproduce, what will be gamete carries a and the male gamete carries A is q  p.
702 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

Sperm quantified. It also can be used to identify “neutral genes” in

fr(A) = p fr(a) = q a population gene pool—those not being operated on by the
forces of evolution.
The Hardy–Weinberg Law has three additional important
fr(A) = p fr(AA) = p2 fr(Aa) = pq
consequences:

Eggs 1. It shows that dominant traits do not necessarily in-

crease from one generation to the next.
fr(a) = q fr(aA) = qp fr(aa) = q 2 2. It demonstrates that genetic variability can be main-
tained in a population since, once established in an ideal
population, allele frequencies remain unchanged.
FIG U R E 2 5 –5 The general description of allele and genotype 3. Using Hardy–Weinberg assumptions, if we know the
frequencies under Hardy–Weinberg assumptions. The frequency frequency of just one genotype we can calculate the
of allele A is p, and the frequency of allele a is q. After mating,
the three genotypes AA, Aa, and aa have the frequencies p2, 2pq,
population frequencies of all other genotypes at that
and q2, respectively. locus. This is particularly useful in human genetics be-
cause we can calculate the frequency of heterozygous
carriers for recessive genetic disorders even when all we
Thus, the frequency of genotype Aa among the zygotes is know is the frequency of affected individuals.
2pq. Finally, the probability that both gametes carry a is q  q,
making the frequency of genotype aa in the zygotes q2.
Therefore, the distribution of genotypes among the zygotes is

p2 + 2pq + q2 = 1 25–1 The ability to taste the compound PTC is controlled

by a dominant allele T, while individuals homozygous for
This is summarized in Figure 25–5. the recessive allele t are unable to taste PTC. In a genetics
These calculations demonstrate the two main predic- class of 125 students, 88 can taste PTC and 37 cannot. Cal-
tions of the Hardy–Weinberg Law: (1) allele frequencies in culate the frequency of the T and t alleles in this population
our population do not change from one generation to the and the frequency of the genotypes.
next, and (2) genotype frequencies after one generation of
HINT: This problem involves an understanding of how to determine
random mating can be predicted from the allele frequencies.
allele and genotype frequencies using the Hardy–Weinberg Law.
In other words, this population does not change or evolve
The key to its solution lies in determining which allele frequency
with respect to the locus we have examined. Remember,
(p or q) you must estimate first when homozygous dominant and
however, the assumptions about the theoretical population heterozygous genotypes have the same phenotype.
described by the Hardy–Weinberg Law:
1. Individuals of all genotypes have equal rates of survival and
equal reproductive success—that is, there is no selection.
25.3
2. No new alleles are created or converted from one allele
into another by mutation. The Hardy–Weinberg Law Can Be
3. Individuals do not migrate into or out of the population.
Applied to Human Populations
4. The population is infinitely large, which in practical To show how allele frequencies are measured in a real popu-
terms means that the population is large enough that lation, let’s consider a gene that influences susceptibility to
sampling errors and other random effects are negligible. infection by HIV-1, the virus responsible for AIDS (acquired
immunodeficiency syndrome). A small number of individ-
5. Individuals in the population mate randomly.
uals who make high-risk choices (such as unprotected sex
These assumptions are what make the Hardy–Weinberg Law with HIV-positive partners) remain uninfected. Some of
so useful in population genetics research. By specifying the these individuals are homozygous for a mutant allele of a
conditions under which the population cannot evolve, the gene called CCR5.
Hardy–Weinberg Law can be used to identify the real-world
forces that cause allele frequencies to change. In other words, Calculating Allele Frequency
by holding certain conditions constant, use of Hardy–Wein- The CCR5 gene (Figure 25–6) encodes a protein called the
berg isolates the forces of evolution and allows them to be C-C chemokine receptor-5, often abbreviated CCR5.
 25. 3 THE HA RDY–WE INBE RG LA W C A N BE A PPLIE D TO HU MA N POPU LA TIONS 703

CCR5 Can we expect the CCR5-32 allele to

increase in human populations in which it is
currently rare? If these populations meet the
1 2 3 4 five assumptions of the Hardy–Weinberg Law,
then the frequency of the 32 allele will not
change. However, as we shall see in later sec-
32 tions of this chapter, when the assumptions
of the Hardy–Weinberg Law are not met—
F I G U R E 2 5 –6 Organization of the CCR5 gene in region 3p21.3. The gene because of natural selection, mutation, migra-
contains 4 exons and 2 introns. The arrow shows the location of the 32-bp deletion tion, or genetic drift—the allele frequencies in
in exon 4 that confers resistance to HIV-1 infection.
a population may change from one generation
to the next. The Hardy–Weinberg Law tells geneticists where
Chemokines are signaling molecules associated with the to look to find the causes of evolution in populations.
immune system. The CCR5 protein is also a receptor for
strains of HIV-1, allowing it to gain entry to cells. The mutant Testing for Hardy–Weinberg
allele of the CCR5 gene contains a 32-bp deletion in a coding Equilibrium
region, making the encoded protein shorter and nonfunc- One way to establish whether one (or more) of the Hardy–
tional. In individuals homozygous for this mutation, HIV-1 Weinberg assumptions does not hold in a given population
cannot enter their cells. The normal allele is called CCR51 is to determine whether the population’s genotypes are in
(also called 1), and the mutant allele is called CCR5-32 equilibrium. To do this, we first determine the frequen-
(also called 32). cies of the genotypes, either directly from the phenotypes
Homozygous 32/32 individuals are resistant to HIV-1 (if heterozygotes are recognizable) or by analyzing proteins
infection; heterozygotes (1/32) are susceptible to HIV-1 or DNA sequences. We then calculate the allele frequencies
infection but progress more slowly to AIDS. Table 25.1 from the genotype frequencies, as demonstrated earlier.
summarizes the genotypes possible at the CCR5 locus and Finally, we use the allele frequencies in the parental genera-
the phenotypes associated with each. tion to predict the offspring’s genotype frequencies. Accord-
The discovery of the CCR5-32 allele generates two ing to the Hardy–Weinberg Law, the genotype frequencies
important questions: Which human populations harbor the are predicted to fit the p2 + 2pq + q2 = 1 relationship. If
32 allele, and how common is it? To address these ques- they do not, then one or more of the assumptions are in-
tions, teams of researchers surveyed people from a variety valid for the population in question.
of populations. Genotypes were determined by direct DNA To demonstrate this, let’s start with a population
analysis (Figure 25–7). In one population, 79 individu- that includes 283 individuals, of which 223 have geno-
als had genotype 1/1, 20 were 1/32, and 1 was 32/32. type 1/1; 57 have genotype 1/32; and 3 have genotype
This population has 158 1 alleles carried by the 1/1 indi-
viduals plus 20 1 alleles carried by 1/32 individuals, for
a total of 178. The frequency of the CCR51 allele in the
32/ 32

sample population is thus 178/200 = 0.89 = 89 percent.

1/ 32
1/1

Copies of the CCR5-32 allele were carried by 20 1/32 in-

dividuals, plus 2 carried by the 32/32 individual, for a
403 bp
total of 22. The frequency of the CCR5-32 allele is thus
22/200 = 0.11 = 11%. Notice that p + q = 1, confirming 371 bp
that we have accounted for the entire gene pool. Table 25.2
shows two methods for computing the frequencies of the 1
and 32 alleles in the population surveyed. 332 bp

TA B L E 2 5 .1
FIGUR E 25–7 Allelic variation in the CCR5 gene. Michel
CCR5 Genotypes and Phenotypes Samson and colleagues used PCR to amplify a part of the CCR5
Genotype Phenotype gene containing the site of the 32-bp deletion, cut the resulting
DNA fragments with a restriction enzyme, and ran the fragments
1>1 Susceptible to sexually transmitted strains of HIV-1 on an electrophoresis gel. Each lane reveals the genotype of a
1> 32 Susceptible but may progress to AIDS slowly single individual. The 1 allele produces a 332-bp fragment and
32> 32 Resistant to most sexually transmitted strains of HIV-1 a 403-bp fragment; the  32 allele produces a 332-bp fragment
and a 371-bp fragment. Heterozygotes produce three bands.
704 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

TA BL E 2 5 . 2 Calculating Frequencies
Methods of Determining Allele Frequencies from Data on Genotypes for Multiple Alleles
(a) Counting Alleles Genotype in Hardy–Weinberg
1/1 1/32 32/32 Total Populations
Number of individuals 79 20 1 100 We commonly find several alleles of a
Number of 1 alleles 158 20 0 178 single gene in a population. The ABO
blood group in humans (discussed in
Number of 32 alleles 0 20 2 22
Chapter 4) is such an example. Re-
Total number of alleles 158 40 2 200 call that the locus I (isoagglutinin)
Frequency of CCR51 in sample: 178/200 = 0.89 = 89, has three alleles IA, IB, and i, yielding
Frequency of CCR5-32 in sample: 22/200 = 0.11 = 11, six possible genotypic combinations
(IAIA, IBi , ii , IAIB, IAi , IBi). Remem-
(b) From Genotype Frequencies Genotype
ber that in this case IA and IB are
1/1 1/32 32/32 Total
codominant alleles and that both of
Number of individuals 79 20 1 100 these are dominant to i. The result
Genotype frequency 79/100 = 0.79 20/100 = 0.20 1/100 = 0.01 1.00 is that homozygous IAIA and het-
Frequency of CCR51 in sample: 0.79 + (0.5)0.20 = 0.89 = 89, erozygous IAi individuals are pheno-
Frequency of CCR5-32 in sample: (0.5)0.20 + 0.01 = 0.11 = 11, typically identical, as are IBIB and IBi
individuals, so we can distinguish
only four phenotypic combinations.
32/32. These numbers represent genotype frequencies By adding another variable to the Hardy–Weinberg
of 223>283 = 0.788, 57>283 = 0.201, and 3>283 = 0.011, equation, we can calculate both the genotype and allele fre-
respectively. From the genotype frequencies, we compute quencies for the situation involving three alleles. Let p, q,
the CCR5-1 allele frequency as 0.89 and the frequency of the and r represent the frequencies of alleles IA, IB, and i, respec-
CCR5-32 allele as 0.11. From these allele frequencies, we tively. Note that because there are three alleles
can use the Hardy–Weinberg Law to determine whether this
population is in equilibrium. The allele frequencies predict p + q + r = 1
the genotype frequencies as follows:
Expected frequency of genotype Under Hardy–Weinberg assumptions, the frequencies
of the genotypes are given by
1>1 = p2 = (0.89)2 = 0.792
(p + q + r)2 = p2 + q2 + r2 + 2pq + 2pr + 2pq = 1
Expected frequency of genotype
If we know the frequencies of blood types for a population,
1>32 = 2pq = 2(0.89)(0.11) = 0.196 we can then estimate the frequencies for the three alleles
of the ABO system. For example, in one population sam-
Expected frequency of genotype
pled, the following blood-type frequencies are observed:
32>32 = q2 = (0.11)2 = 0.012 A = 0.53, B = 0.13, O = 0.26. Because the i allele is reces-
sive, the population’s frequency of type O blood equals the
These expected frequencies are nearly identical to the proportion of the recessive genotype r 2. Thus,
observed frequencies. Our test of this population has failed
to provide evidence that Hardy–Weinberg assumptions are r2 = 0.26
being violated. The conclusion is confirmed by a x2 analysis r = 20.26
(see Chapter 3). The x2 value in this case is tiny: 0.00023. r = 0.51
To reject the null hypothesis at even the most generous,
accepted level, p = 0.05, the x2 value would have to be 3.84. Using r, we can calculate the allele frequencies for the
(In a test for Hardy–Weinberg equilibrium, the degrees of I and IB alleles. The IA allele is present in two genotypes,
A
freedom are given by k - 1 - m, where k is the number of IAIA and IAi. The frequency of the IAIA genotype is repre-
genotypes and m is the number of independent allele sented by p2 and the IAi genotype by 2pr. Therefore, the
frequencies estimated from the data. Here, k = 3 and combined frequency of type A blood and type O blood is
m = 1 since calculating only one allele frequency allows given by
us to determine the other by subtraction. Thus, we have
3 - 1 - 1 = 1 degree of freedom.) p2 + 2pr + r2 = 0.53 + 0.26
 25. 3 THE HA RDY–WE INBE RG LA W C A N BE A PPLIE D TO HU MA N POPU LA TIONS 705

TA B L E 2 5 . 3 To illustrate this for a recessive X-

Calculating Genotype Frequencies for Multiple Alleles in a Hardy–Weinberg Population linked trait, let’s consider the example
Where the Frequency of Allele I ⴝ 0.38, Allele I ⴝ 0.11, and Allele i ⴝ 0.51
A B of red-green color blindness, which af-
Genotype Genotype Frequency Phenotype Phenotype Frequency fects 8 percent of human males. The
frequency of the color blindness al-
IA IA p2 = (0.38)2 = 0.14 A 0.53
2pr = 2(0.38)(0.51) = 0.39 lele is therefore 0.08; in other words,
IA i
8 percent of X chromosomes carry it.
IBIB q2 = (0.11)2 = 0.01 B 0.12
The other 92 percent of X chromo-
IBi 2qr = 2(0.11)(0.51) = 0.11
somes carry the dominant allele for
IAIB 2pr = 2(0.38)(0.11) = 0.084 AB 0.08 normal red-green color vision. If we
ii r2 = (0.51)2 = 0.26 O 0.26 define p as the frequency of the nor-
mal allele and q as the frequency of the
color-blindness allele, then p = 0.92
and q = 0.08. The frequency of
If we factor the left side of the equation and take the color-blind females (with two affected X chromosomes)
sum of the terms on the right is q2 = (0.08)2 = 0.0064, and the frequency of carrier fe-
males (having one normal and one affected X chromosome)
(p + r) = 0.79
2 is 2pq = 2(0.08)(0.92) = 0.147. In other words, 14.7 per-
p + r = 20.79 cent of females carry the allele for red-green color blindness
p = 0.89 - r and can pass it to their children, although they themselves
have normal color vision.
p = 0.89 - 0.51 = 0.38
An important consequence of the difference in allele
frequency for X-linked genes between male and female
Having calculated p and r, the frequencies of allele IA and gametes is that for a rare recessive allele, the trait will be
allele i, we can now calculate the frequency for the IB allele: expressed at a much higher frequency among XY individu-
als than among those who are XX. So, for example, diseases
p + q + r = 1 such as hemophilia and Duchenne muscular dystrophy
q = 1 - p - r (DMD) in humans, both of which are caused by recessive
= 1 - 0.38 - 0.51 mutations on the X chromosome, are much more common
= 0.11 in boys, who need only inherit a single copy of the mutated
allele to suffer from the disease. Girls who inherit two af-
fected X chromosomes will also have the disease; but with a
The phenotypic and genotypic frequencies for this popula- rare allele, the probability of this occurrence is small.
tion are summarized in Table 25.3.
Calculating Heterozygote Frequency
In another application, the Hardy–Weinberg Law allows us
Calculating Allele Frequencies to estimate the frequency of heterozygotes in a population.
for X-linked Traits The frequency of a recessive trait can usually be determined
The Hardy–Weinberg Law can be used to calculate allele by counting such individuals in a sample of the population.
and genotype frequencies for X-linked traits, as long as we With this information and the Hardy–Weinberg Law, we
remember that in an XY sex-determination system, the ho- can then calculate the allele and genotype frequencies.
mogametic (XX) sex will have two copies of an X-linked al- Cystic fibrosis, an autosomal recessive trait, has an in-
lele, whereas the heterogametic sex (XY) only has one copy. cidence of about 1>2500 = 0.0004 in people of northern
Thus, for mammals (including humans) where the female European ancestry. Individuals with cystic fibrosis are easily
is XX and the male is XY, the frequency of the X-linked al- distinguished from the population at large by such symp-
lele in the gene pool and the frequency of males express- toms as extra-salty sweat, excess amounts of thick mucus in
ing the X-linked trait will be the same. This is because each the lungs, and susceptibility to bacterial infections. Because
male only has one X chromosome, and the probability of this is a recessive trait, individuals with cystic fibrosis must be
any individual male receiving an X chromosome with the al- homozygous. Their frequency in a population is represented
lele in question must be equal to the frequency of the allele. by q2, provided that mating has been random in the previous
The probability of any individual female having the allele in generation. The frequency of the recessive allele is therefore
question on both X chromosomes will be q2, where q is the
frequency of the allele. q = 2q2 = 20.0004 = 0.02
706 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

Since p + q = 1, then the frequency of p is survival, competing with other members of the species
for scarce resources. Offspring also must avoid predators,
p = 1 - q = 1 - 0.02 = 0.98
and in sexually reproducing species, adults must com-
In the Hardy–Weinberg equation, the frequency of hetero- pete for mates.
zygotes is 2pq. Thus,
4. In the struggle for survival, individuals with particular
2pq = 2(0.98) (0.02) phenotypes will be more successful than others, allow-
= 0.04 or 4 percent, or 1>25 ing the former to survive and reproduce at higher rates.

Heterozygotes for cystic fibrosis are rather common in the As a consequence of natural selection, populations and
population (about 1>25, or 4 percent), even though the in- species change. The phenotypes that confer improved abil-
cidence of homozygous recessives is only 1>2500, or 0.04 ity to survive and reproduce become more common, and
percent. Calculations such as these are estimates because the the phenotypes that confer poor prospects for survival and
population may not meet all Hardy–Weinberg assumptions. reproduction may eventually disappear. Under certain con-
ditions, populations that at one time could interbreed may
lose that capability, thus segregating their adaptations into
particular niches. If selection continues, it may result in the
appearance of new species.
25–2 If the albino phenotype occurs in 1>10,000 individu-
als in a population at equilibrium and albinism is caused
by an autosomal recessive allele a, calculate the frequency Detecting Natural Selection
of (a) the recessive mutant allele; (b) the normal dominant in Populations
allele; (c) heterozygotes in the population; and (d) matings Recall that measuring allele frequencies and genotype fre-
between heterozygotes. quencies using the Hardy–Weinberg Law is based on certain
H IN T: This problem involves an understanding of how to use the assumptions about an ideal population: large population
Hardy–Weinberg Law to calculate allele frequencies and heterozy- size, lack of migration, presence of random mating, ab-
gote frequencies. The key to its solution lies in solving the problem sence of selection and mutation, and equal survival rates of
in the order presented. offspring.
However, if all genotypes do not have equal rates of sur-
vival or do not leave equal numbers of offspring, then allele
frequencies may change from one generation to the next.
To see why, let’s imagine a population of 100 individuals in
25.4
which the frequency of allele A is 0.5 and that of allele a
Natural Selection Is a Major Force is 0.5. Assuming the previous generation mated randomly,
we find that the genotype frequencies in the present genera-
Driving Allele Frequency
tion are (0.5)2  0.25 for AA, 2(0.5)(0.5) = 0.5 for Aa, and
To understand evolution, we must understand the forces (0.5)2  0.25 for aa. Because our population contains 100
that transform the gene pools of populations and can lead individuals, we have 25 AA individuals, 50 Aa individuals,
to the formation of new species. Chief among the mecha- and 25 aa individuals. Now suppose that individuals with
nisms transforming populations is natural selection, different genotypes have different rates of survival: All 25
discovered independently by Alfred Russel Wallace and AA individuals survive to reproduce, 90 percent or 45 of the
Charles Darwin. The Wallace–Darwin concept of natural Aa individuals survive to reproduce, and 80 percent or 20 of
selection can be summarized as follows: the aa individuals survive to reproduce. When the survivors
reproduce, each contributes two gametes to the new gene
1. Individuals of a species exhibit variations in phenotype— pool, giving us 2(25)  2(45)  2(20)  180 gametes. What
for example, differences in size, agility, coloration, defenses are the frequencies of the two alleles in the surviving popu-
against enemies, ability to obtain food, courtship be- lation? We have 50 A gametes from AA individuals, plus 45
haviors, and flowering times. A gametes from Aa individuals, so the frequency of allele
A is (50 + 45)>180  0.53. We have 45 a gametes from Aa
2. Many of these variations, even small and seemingly in-
individuals, plus 40 a gametes from aa individuals, so the
significant ones, are heritable and passed on to offspring.
frequency of allele a is (45 + 40)>180 = 0.47.
3. Organisms tend to reproduce in an exponential fash- These differ from the frequencies we started with. The
ion. More offspring are produced than can survive. This frequency of allele A has increased, and the frequency of allele
causes members of a species to engage in a struggle for a has decreased. A difference among individuals in survival
 25. 4 NA TU RA L S E LE C TION IS A MAJOR FORC E DRIVING A LLE LE FRE QU E NCY 707

or reproduction rate (or both) is an example of natural se- 0.50

lection. Natural selection is the principal force that shifts

Frequency of allele a
allele frequencies within large populations and is one of the 0.40
most important agents of evolutionary change.
0.30
0.25
Fitness and Selection 0.20
Selection occurs whenever individuals with a particular
genotype enjoy an advantage in survival and reproduction 0.10
over other genotypes. However, selection may vary from less
0
than 1 to 100 percent. In the previous hypothetical example, 0 2 10 20 100
selection was strong. Weak selection might involve just a Generation
fraction of a percent difference in the survival rates of dif-
ferent genotypes. Advantages in survival and reproduction FIGUR E 25–8 Change in the frequency of a lethal recessive
ultimately translate into increased genetic contribution to allele, a. The frequency of a is halved in two generations and
future generations. An individual’s genetic contribution to halved again by the sixth generation. Subsequent reductions
future generations is called its fitness. Genotypes associated occur slowly because the majority of a alleles are carried by
heterozygotes.
with high rates of reproductive success are said to have high
fitness, whereas genotypes associated with low reproductive
success are said to have low fitness.
Hardy–Weinberg analysis also allows us to examine fit- Figure 25–9 shows the outcome of different degrees of
ness. By convention, population geneticists use the letter w selection against a nonlethal recessive allele, a. In this case,
to represent fitness. Thus, wAA represents the relative fitness the intensity of selection varies from strong (red curve) to
of genotype AA, wAa the relative fitness of genotype Aa, and weak (blue curve), as well as intermediate values (yellow,
waa the relative fitness of genotype aa. Assigning the values purple, and green curves). In each example, the frequency of
wAA = 1, wAa = 0.9, and waa = 0.8 would mean, for ex- the deleterious allele, a, starts at 0.99 and declines over time.
ample, that all AA individuals survive, 90 percent of the Aa However, the rate of decline depends heavily on the strength
individuals survive, and 80 percent of the aa individuals of selection. When selection is strong and only 90 percent
survive, as in the previous hypothetical case. of the heterozygotes and 80 percent of the aa homozygotes
Let’s consider selection against deleterious alleles. Fitness survive (red curve), the frequency of allele a drops from 0.99
values wAA = 1, wAa = 1, and waa = 0 describe a situation to less than 0.01 in about 85 generations. However, when
in which a is a homozygous lethal allele. Because homozy- selection is weak, and 99.8 percent of the heterozygotes and
gous recessive individuals die without leaving offspring, the 99.6 percent of the aa homozygotes survive (blue curve),
frequency of allele a will decrease. The change in the fre- it takes 1000 generations for the frequency of allele a to
quency of allele a is described by the equation drop from 0.99 to 0.93. Two important conclusions can be
drawn from this example. First, over thousands of genera-
q0 tions, even weak selection can cause substantial changes in
qg =
1 + gq0

where qg is the frequency of allele a in generation g, q0 is the

1.0
frequency of a in generation zero, and g is the number of
Frequency of allele a

generations that have passed. 0.8

Figure 25–8 shows what happens to a lethal recessive
0.6
allele with an initial frequency of 0.5. At first, because of the
high percentage of aa genotypes, the frequency of allele a 0.4
declines rapidly. The frequency of a is halved in only two 0.2
generations. By the sixth generation, the frequency is halved
again. By now, however, the majority of a alleles are carried 0
0 200 400 600 800 1000
by heterozygotes. Because a is recessive, these heterozygotes Generation
are not selected against. Consequently, as more time passes,
the frequency of allele a declines ever more slowly. As long FIGUR E 25–9 The effect of selection on allele frequency.
as heterozygotes continue to mate, it is difficult for selection The rate at which a deleterious allele is removed from a popula-
to completely eliminate a recessive allele from a population. tion depends heavily on the strength of selection.
708 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

allele frequencies; because evolution generally occurs over population had beaks less than 7 mm deep, while others had
a large number of generations, selection is a powerful force beaks more than 12 mm deep. In 1977, a severe drought
in evolutionary change. Second, for selection to produce killed some 80 percent of the finches. Big-beaked birds sur-
rapid changes in allele frequencies, the differences in fitness vived at higher rates than small-beaked birds because when
among genotypes must be large. food became scarce, the big-beaked birds were able to eat a
The manner in which selection affects allele frequen- greater variety of seeds. When the drought ended in 1978,
cies allows us to make some inferences about the CCR5-32 the offspring of the survivors inherited their parents’ big
allele that we discussed earlier. Because individuals with beaks. Between 1976 and 1978, the beak depth of the aver-
genotype 3232 are resistant to most sexually transmit- age finch in the Daphne Major population increased by just
ted strains of HIV-1, while individuals with genotypes 11 over 0.5 mm, shifting the average beak size in the direction
and 132 are susceptible, we might expect AIDS to act as of one phenotypic extreme.
a selective force causing the frequency of the 32 allele to Stabilizing selection tends to favor intermediate phe-
increase over time. Indeed, it probably will, but the increase notypes, with those at both extremes being selected against.
in frequency is likely to be slow in human terms. In fact, it Over time, this will reduce the phenotypic variance in the
will take about 100 generations (about 2000 years) for the population but without a significant shift in the mean. One
frequency of the 32 allele to reach just 0.11. In other words, of the clearest demonstrations of stabilizing selection is
the frequency of the 32 allele will probably not change shown by a study of human birth weight and survival for
much over the next few generations in most populations 13,730 children born over an 11-year period. Figure 25–11
that currently harbor it. shows the distribution of birth weight and the percentage
of mortality at 4 weeks of age. Infant mortality increases on
either side of the optimal birth weight of 7.5 pounds. Stabi-
There Are Several Types of Selection lizing selection acts to keep a population well adapted to its
The phenotype is the result of the combined influence of present environment.
the individual’s genotype at many different loci and the Disruptive selection is a case where selection acts against
effects of the environment. Selection for these complex intermediate phenotypes and in favor of both phenotypic ex-
traits can be classified as (1) directional, (2) stabilizing, or tremes. It can be viewed as the opposite of stabilizing selec-
(3) disruptive. tion because the intermediate types are selected against. This
In directional selection (Figure 25–10) phenotypes will result in a population with an increasingly bimodal dis-
at one end of the spectrum become selected for or against, tribution for the trait, as we can see in Figure 25–12. In one
usually as a result of changes in the environment. A care- set of experiments using Drosophila, after several generations
fully documented example comes from research by Peter of disruptive artificial selection for bristle number, in which
and Rosemary Grant and their colleagues, who study the only flies with high- or low-bristle numbers were allowed to
medium ground finches (Geospiza fortis) of Daphne Major breed, most flies could be easily placed in a low- or high-bristle
Island in the Galapagos Islands. The beak size of these birds category. In natural populations, such a situation might exist
varies enormously. For example, in 1976, some birds in the for a population in a heterogeneous environment.

10.0
Mean beak depth (mm)

9.8 Wet year 80 20

Percent of births in population

70
Infant births
Percentage mortality

9.6
60 15
Infant
9.4 50 deaths
Dry Dry Dry
year year year 40 10
9.2
30
9.0 20 5

1977 1980 1982 1984 10

0 0
2 4 6 8 10
FIG U R E 2 5 –10 Beak size in finches during dry years in-
creases because of strong selection. Between droughts, selection Birth weight (pounds)
for large beak size is not as strong, and birds with smaller beak
sizes survive and reproduce, increasing the number of birds with FIGUR E 25–11 Relationship between birth weight and
smaller beaks. mortality in humans.
 25.5 MU TA TION C RE ATE S NE W ALLE LE S IN A G E NE POO L 709

To determine whether mutation is a significant force

in changing allele frequencies, we must measure the rate at
1
which mutations are produced. As most mutations are re-
cessive, it is difficult to observe mutation rates directly in
2 diploid organisms. Indirect methods use probability and
statistics or large-scale screening programs to estimate mu-
3 tation rates. For certain dominant mutations, however, a
direct method of measurement can be used. To ensure ac-
4 curacy, several conditions must be met:
1. The allele must produce a distinctive phenotype that
Generations of selection

5 can be distinguished from similar phenotypes pro-

duced by recessive alleles.
6 2. The trait must be fully expressed or completely pen-
etrant so that mutant individuals can be identified.
7
3. An identical phenotype must never be produced by
nongenetic agents such as drugs or chemicals.
8
Mutation rates can be defined as the number of new
9
mutant alleles per given number of gametes. Suppose that
for a certain recessively inherited gene that undergoes
mutation to a dominant allele, 2 out of 100,000 births ex-
10
hibit a mutant phenotype. In these cases, both sets of par-
ents are phenotypically normal. Because the zygotes that
11 produced these births each carry two copies of the gene,
we have actually surveyed 200,000 copies of the gene (or
12 200,000 gametes). If we assume that the affected births are
10 15 20 25 30 35 each heterozygous, we have uncovered 2 new mutant al-
Number of bristles leles out of 200,000 alleles surveyed. Thus, the mutation
rate is 2>200,000 or 1>100,000, which in scientific notation
F I G U R E 2 5 –12 The effect of disruptive selection on bristle is written as 1  10–5. In humans, a dominantly inherited
number in Drosophila. When individuals with the highest and form of dwarfism known as achondroplasia fulfills the re-
lowest bristle number were selected, the population showed a quirements for measuring mutation rates. Individuals with
nonoverlapping divergence in only 12 generations. this skeletal disorder have an enlarged skull, short arms and
legs, and can be diagnosed by X-ray examination at birth. In
a survey of almost 250,000 births, the mutation rate (m) for
achondroplasia has been calculated as
25.5
m = 1.4 * 10 - 5 { 0.5 * 10 - 5
Mutation Creates New Alleles
in a Gene Pool Knowing the rate of mutation, we can estimate the extent to
which mutation can cause allele frequencies to change from
Each generation, a population’s gene pool is reshuffled to one generation to the next. We represent the normal allele as
produce new genotypes in the offspring. The enormous ge- d and the allele for achondroplasia as D.
netic variation present in the gene pool allows assortment Imagine a population of 500,000 individuals in which
and recombination to produce new genotypic combinations everyone has genotype dd. The initial frequency of d is
continuously. But assortment and recombination do not 1.0, and the initial frequency of D is 0. If each individual
produce new alleles. Mutation alone acts to create new al- contributes two gametes to the gene pool, the gene pool
leles. It is important to keep in mind that mutational events will contain 1,000,000 gametes, all carrying allele d. In
occur at random—that is, without regard for any possible this collection of alleles, 1.4 of every 100,000 d alleles
benefit or disadvantage to the organism. In this section, we mutate into a D allele. The frequency of allele d is
consider whether mutation, by itself, is a significant factor in now (1,000,000 - 14)>1,000,000 = 0.999986, and the fre-
changing allele frequencies. quency of allele D is 14 >1,000,000 = 0.000014. From these
710 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

1.0 In this case, migration from the mainland has changed the
Mutation rate (A a) = 1.0  10–5
frequency of A on the island from 0.40 to 0.42 in a single
0.8 generation.
Frequency of A

These calculations reveal that the change in allele fre-

0.6
quency attributable to migration is proportional to the dif-
0.5
ferences in allele frequency between the donor and recipient
0.4
populations and to the rate of migration. If either m is large
0.2 or pm is very different from pi, then a rather large change in
70,000 the frequency of A can occur in a single generation. If mi-
0 gration is the only force acting to change the allele frequency
50,000 100,000 150,000 200,000
on the island, then equilibrium will be attained only when
Number of generations
pi = pm. These guidelines can often be used to estimate mi-
gration in cases where it is difficult to quantify. As m can
FI G U RE 2 5 – 13 Replacement rate of an allele by mutation
alone, assuming an average mutation rate of 1.03  10–5. have a wide range of values, the effect of migration can sub-
stantially alter allele frequencies in populations, as shown
for the IB allele of the ABO blood group in Figure 25–14.

numbers, it will clearly be a long time before mutation, by

25.7
itself, causes any appreciable change in the allele frequencies
in this population (Figure 25–13). In other words, muta- Genetic Drift Causes Random
tion generates new alleles but by itself does not alter allele
frequencies at an appreciable rate.
Changes in Allele Frequency
in Small Populations
25.6 In small populations, significant random fluctuations in al-
Migration and Gene Flow Can lele frequencies are possible by chance alone. The degree of
fluctuation increases as the population size decreases, a situ-
Alter Allele Frequencies ation known as genetic drift. In addition to small popula-
tion size, drift can arise through the founder effect, which
Occasionally, a species becomes divided into smaller, geo-
occurs when a population originates from a small number
graphically separated populations. In such populations, var-
of individuals, whose gene pool may not reflect that of the
ious evolutionary forces, including selection, migration, and
larger population from which the founders are drawn. Al-
drift, can alter allele frequencies. Migration occurs when in-
though the population may later increase to a large size, the
dividuals move between the populations. Imagine a species
genes carried by all members are derived only from those of
in which a given locus has two alleles, A and a. There are two
the founders (assuming no mutation, migration, or selec-
populations of this species, one on a mainland and one on an
tion, and the presence of random mating). Drift can also
island. The frequency of A on the mainland is represented
arise via a genetic bottleneck. Bottlenecks develop when a
by pm, and the frequency of A on the island is pi . If there is
large population undergoes a drastic but temporary reduc-
migration from the mainland to the island, the frequency of
tion in numbers. Even though the population recovers, its
A in the next generation on the island (pi) is given by
genetic diversity has been greatly reduced.
pi = (1 - m)pi + mpm
Founder Effects in Human Populations
where m represents migrants from the mainland to the island.
Allele frequencies in certain human populations demon-
As an example of how migration might affect the fre-
strate the role of genetic drift in natural populations. Native
quency of A in the next generation on the island (pi), as-
Americans living in the southwestern United States have a
sume that pi = 0.4 and pm = 0.6 and that 10 percent of the
high frequency of oculocutaneous albinism (OCA). In the
parents of the next generation are migrants from the main-
Navajo, who live primarily in northeast Arizona, albinism
land (m = 0.1). In the next generation, the frequency of al-
occurs with a frequency of 1 in 1500–2000, compared with
lele A on the island will therefore be
whites (1 in 36,000) and African-Americans (1 in 10,000).
There are four different forms of OCA (OCA1–4), all with
pi = 3 (1 - 0.1) * 0.4 4 + (0.1 * 0.6)
varying degrees of melanin deficiency in the skin, eyes, and
= 0.36 + 0.06 hair. OCA2 is caused by mutations in the P gene, which
= 0.42 encodes a plasma membrane protein. To investigate the
 2 5. 7 GEN ETIC D RIF T CAUSES RANDOM C HA NG E S IN ALLE LE FRE QU E NC Y IN S MALL POPU LATIONS 711

FIGUR E 25–14 Affected

migration as a force in evolution.
20–25%
The IB allele of the ABO locus is
present in a gradient from east to
west. This allele shows the highest
frequency in central Asia and the
lowest in northeast Spain. The
25–30% gradient parallels the waves of
Mongol migration into Europe
following the fall of the Roman
15–20% Empire and is a genetic relic of
human history.

5–10%

10–15%
0–5%

genetic basis of albinism in the Navajo, researchers screened pigmented Navajo and 42 members of the Apache, a tribe
for mutations in the P gene. In their study, all Navajo with closely related to the Navajo. Based on this sample, the
albinism were homozygous for a 122.5-kb deletion in the P heterozygote frequency in the Navajo is estimated to be
gene, spanning exons 10–20 (Figure 25–15). This deletion 4.5 percent. No carriers were found in the Apache popula-
allele was not present in 34 individuals belonging to other tion that was studied.
Native American populations. The 122.5-kb deletion allele causing OCA2 was found
Using a set of PCR primers, researchers were able to only in the Navajo population and not in members of other
identify homozygous affected individuals and heterozy- Native American tribes in the southwestern United States,
gous carriers (Figure 25–16) and surveyed 134 normally suggesting that the mutant allele is specific to the Navajo

(a) (b)
BamHI Xbal BglI BamHI Xbal BglI
N5 C N5 C N5 C N5 C N5 C N5 C
kb kb
9
11
FIGUR E 25–15 Genomic DNA digests
6 from a Navajo affected with albinism (N5)
6.5 and a normally pigmented individual (C).
4 (a) Hybridization with a probe covering exons
5
11–15 of the P gene; there are no hybridizing
fragments detected in N5. (b) Hybridization
with a probe covering exons 15–20 of the
P gene; there are no hybridizing fragments
detected in N5. This confirms the presence of
a deletion in affected individuals.
Courtesy of Murray Brilliant, “A 122.5 kilobase deletion of
P gene underlies the high prevalence of oculocutaneous albinism
2.2 1.3 type 2 in the Navajo population.” From: American Journal
Human Genetics 72:62–72, Figure 1, p. 65. Published by
University of Chicago Press.
712 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

M N5 C N2 N3 N4 Nonrandom mating can take one of

several forms. In positive assortive mating
similar genotypes are more likely to mate
than dissimilar ones. This often occurs in
humans: A number of studies have indicat-
606 bp ed that many people are more attracted to
individuals who physically resemble them
(and are therefore more likely to be geneti-
cally similar as well). Negative assortive
mating occurs when dissimilar genotypes
257 bp are more likely to mate; some plant species
have inbuilt pollen/stigma recognition sys-
tems that prevent fertilization between in-
dividuals with the same alleles at key loci.
FIG U R E 2 5 –16 PCR screens of Navajo affected with albinism (N4 and N5)
and the parents of N4 (N2 and N3). Affected individuals (N4 and N5), heterozy- However, the form of nonrandom mating
gous carriers (N2 and N3), and a homozygous normal individual (C) each give a most commonly found to affect genotype
distinctive band pattern, allowing detection of heterozygous carriers in the popula- frequencies in population genetics is in-
tion. Molecular size markers (M) are in the first lane. breeding.
Courtesy of Murray Brilliant, “A 122.5 kilobase deletion of P gene underlies the high prevalence of
oculocutaneous albinism type 2 in the Navajo population.” From: American Journal Human Genetics
72:62–72, Figure 3, p. 67. Published by University of Chicago Press.

Inbreeding
and may have arisen in a single individual who was one of a
Inbreeding occurs when mating individuals are more closely
small number of founders of the Navajo population. Using
related than any two individuals drawn from the population
other methods, workers estimated the age of the mutation
at random; loosely defined, inbreeding is mating among rela-
to be between 400 and 11,000 years. To narrow this range,
tives. For a given allele, inbreeding increases the proportion of
they relied on tribal history. Navajo oral tradition indicates
homozygotes in the population. A completely inbred popula-
that the Navajo and Apache became separate populations
tion will theoretically consist only of homozygous genotypes.
between 600 and 1000 years ago. Because the deletion is
To describe the amount of inbreeding in a population,
not found in the Apaches, it probably arose in the Navajo
geneticist Sewall Wright devised the coefficient of inbreed-
population after the tribes split. On this basis, the deletion is
ing (F). F quantifies the probability that the two alleles of a
estimated to be 400 to 1000 years old and probably arose as
given gene in an individual are identical because they are de-
a founder mutation.
scended from the same single copy of the allele in an ancestor.
If F = 1, all individuals in the population are homozygous,
25.8 and both alleles in every individual are derived from the
Nonrandom Mating Changes same ancestral copy. If F = 0, no individual has two alleles
derived from a common ancestral copy.
Genotype Frequency but Not One method of estimating F for an individual is shown
Allele Frequency in Figure 25–17. In this pedigree, the fourth-generation
female (shaded pink) is the daughter of first cousins (yel-
We have explored how violations of the first four assump- low). Suppose her great-grandmother (green) was a car-
tions of the Hardy–Weinberg Law, in the form of selection, rier of a recessive lethal allele, a. What is the probability that
mutation, migration, and genetic drift can cause allele fre- this fourth-generation female will inherit two copies of her
quencies to change. The fifth assumption is that the mem- great-grandmother’s lethal allele? For this to happen, (1) the
bers of a population mate at random; in other words, any great-grandmother had to pass a copy of the allele to her son,
one genotype has an equal probability of mating with any (2) her son had to pass it to his daughter, and (3) his daugh-
other genotype in the population. Nonrandom mating can ter had to pass it to her daughter (the pink female). Also,
change the frequencies of genotypes in a population. Sub- (4) the great-grandmother had to pass a copy of the allele to
sequent selection for or against certain genotypes has the her daughter, (5) her daughter had to pass it to her son, and
potential to affect the overall frequencies of the alleles they (6) her son had to pass it to his daughter (the pink female).
contain, but it is important to note that nonrandom mating Each of the six necessary events has an individual probability
does not itself directly change allele frequencies. of 1>2, and they all have to happen, so the probability that the
 2 5. 9 RED UCED GENE FLOW, S E LE C TION, AND G E NE TIC DRIFT C A N LE A D TO S PE C IA TIO N 713

Aa AA organisms, speciation transforms the gene

1 pool of the parental species or divides a
1 2 single gene pool into two or more sepa-
2
rate and distinct gene pools. Changes in
1 1 morphology or physiology and adaptation
2 2 to an ecological niche may also occur but
are not necessary components of the
The chance that this female will inherit two
1 1 copies of her great-grandmother’s a allele is speciation event. Speciation can take place
2 2 1 1 1
F=      = 1 1 1 1 gradually or within a few generations.
2 2 2 2 2 2 64 As we saw earlier, most populations
Because the female’s two alleles could be identical
by descent from any of four different alleles, contain considerable genetic variation,
1 = 1 and different populations within a spe-
F = 4  64 16 cies may carry different alleles or allele
frequencies at a variety of loci. The ge-
F I G U R E 2 5 –17 Calculating the coefficient of inbreeding (F) for the offspring of a netic divergence of these populations
first-cousin marriage. can be caused by natural selection, ge-
netic drift, or both. In an earlier section,
pink female will inherit two copies of her great-grandmother’s we saw that the migration of individuals between popula-
lethal allele is (1>2)6 = 1>64. To calculate an overall value of tions, together with the gene flow that accompanies that
F for the pink female as a child of a first-cousin marriage, re- migration, tends to homogenize allele frequencies among
member that she could also inherit two copies of any of the populations. In other words, migration counteracts the ten-
other three alleles present in her great-grandparents. Because dency of populations to diverge.
any of four possibilities would give the pink female two alleles When gene flow between populations is reduced or ab-
identical by descent from an ancestral copy, sent, the populations may diverge to the point that members
of one population are no longer able to interbreed success-
F = 4 * (1>64) = 1>16 fully with members of the other. When populations reach
the point where they are reproductively isolated from one
another, they have become different species, according to the
25–3 A prospective groom, who is healthy, has a sister with biological species concept. The genetic changes that result in
cystic fibrosis (CF), an autosomal recessive disease. Their reproductive isolation between or among populations and
parents are normal. The brother plans to marry a woman that lead to the formation of new species (or higher taxo-
who has no history of CF in her family. What is the prob- nomic groups) are an example of macroevolution.
ability that they will produce a CF child? They are both Cau- The biological barriers that prevent or reduce inter-
casian, and the overall frequency of CF in the Caucasian breeding between populations are called reproductive
population is 1>2500—that is, 1 affected child per 2500. isolating mechanisms, classified in Table 25.4. These mech-
(Assume the population meets the Hardy–Weinberg as- anisms may be ecological, behavioral, seasonal, mechanical,
sumptions.)
or physiological.
HINT: This problem involves an understanding of how recessive Prezygotic isolating mechanisms prevent individuals
traits are inherited and how to use family history to calculate the from mating in the first place. Individuals from different
probability that an individual is heterozygous. The key to its solu- populations may not find each other at the right time, may
tion lies in knowing the risk factors for the man and the woman. not recognize each other as suitable mates, or may try to
mate but find that they are unable to do so.
Postzygotic isolating mechanisms create reproduc-
tive isolation even when the members of two populations
25.9 are willing and able to mate with each other. For example,
Reduced Gene Flow, Selection, genetic divergence may have reached the stage where the vi-
ability or fertility of hybrids is reduced. Hybrid zygotes may
and Genetic Drift Can Lead be formed, but all or most may be inviable. Alternatively,
to Speciation the hybrids may be viable, but may be sterile or suffer from
reduced fertility. Yet again, the hybrids themselves may be
A species can be defined as a group of actually or potentially fertile, but their progeny may have lowered viability or fer-
interbreeding organisms that is reproductively isolated in tility. In all these situations, hybrids will not reproduce and
nature from all other such groups. In sexually reproducing are genetic dead-ends. These postzygotic mechanisms act at
714 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

TA BL E 2 5 . 4

Reproductive Isolating Mechanisms

Prezygotic Mechanisms
Prevent fertilization and zygote formation
1. Geographic or ecological: The populations live in the same regions but occupy different habitats.
2. Seasonal or temporal: The populations live in the same regions but are sexually mature at different times.
3. Behavioral (only in animals): The populations are isolated by different and incompatible behavior before mating.
4. Mechanical: Cross-fertilization is prevented or restricted by differences in reproductive structures (genitalia in animals, flowers in plants).
5. Physiological: Gametes fail to survive in alien reproductive tracts.
Postzygotic Mechanisms
Fertilization takes place and hybrid zygotes are formed, but these are nonviable or give rise to weak or sterile hybrids.
1. Hybrid nonviability or weakness.
2. Developmental hybrid sterility : Hybrids are sterile because gonads develop abnormally or meiosis breaks down before completion.
3. Segregational hybrid sterility : Hybrids are sterile because of abnormal segregation into gametes of whole chromosomes, chromosome
segments, or combinations of genes.
4. F2 breakdown: F1 hybrids are normal, vigorous, and fertile, but the F2 contains many weak or sterile individuals.
Source: From Processes of Organic Evolution, 3rd Edition by G. Ledyard Stebbins, © 1977, p. 143. Reprinted by permission of Pearson Education, Inc.

or beyond the level of the zygote and are generated by genetic a similar Pacific species to form a pair. Members of each pair
divergence. were closer to each other in structure and appearance than
Postzygotic isolating mechanisms waste gametes and either was to any other species in its own ocean. Analysis of
zygotes and lower the reproductive fitness of hybrid sur- allele frequencies and mitochondrial DNA sequences con-
vivors. Selection will therefore favor the spread of alleles firmed that the members of each pair were one another’s
that reduce the formation of hybrids, leading to the devel- closest genetic relatives.
opment of prezygotic isolating mechanisms, which in turn The interpretation of these data is that, prior to the for-
prevent interbreeding and the formation of hybrid zygotes mation of the isthmus, the ancestors of each pair were mem-
and offspring. In animal evolution, one of the most effective bers of a single species. When the isthmus closed, each of the
prezygotic mechanisms is behavioral isolation, involving seven ancestral species was divided into two separate popula-
courtship behavior. tions, one in the Caribbean and the other in the Pacific.
Meeting in a dish in a lab for the first time in 3 mil-
Changes Leading to Speciation lion years, would Caribbean and Pacific members of a spe-
The Isthmus of Panama, which created a land bridge con- cies pair recognize each other as suitable mates? Males and
necting North and South America and simultaneously sep- females were paired together, and the relative inclination
arated the Caribbean Sea from the Pacific Ocean, formed of Caribbean–Pacific couples to mate versus that of Carib-
roughly 3 million years ago. To study populations separated bean–Caribbean or Pacific–Pacific couples was calculated.
by the formation of the isthmus, researchers matched seven For three of the seven species pairs, transoceanic couples
Caribbean species of snapping shrimp (Figure 25–18) with refused to mate altogether. For the other four species pairs,
transoceanic couples were 33, 45, 67, and 86 percent as likely
to mate with each other as were same-ocean pairs. Of the
same-ocean couples that mated, 60 percent produced viable
clutches of eggs. Of the transoceanic couples that mated,
only 1 percent produced viable clutches. We can conclude
from these results that 3 million years of separation has re-
sulted in complete or nearly complete speciation, involving
strong pre- and postzygotic isolating mechanisms for all
seven species pairs.

The Rate of Macroevolution

and Speciation
How much time is required for speciation? In many cases,
divergence of populations and the appearance of new spe-
FIG U R E 2 5 –18 A snapping shrimp (genus Alpheus). cies occur over a long period of time. In other cases, however,
 2 5. 9 RED UCED GENE FLOW, S E LE C TION, AND G E NE TIC DRIFT C A N LE A D TO S PE C IA TIO N 715

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
(a)
Juba 480 bp
160 bp
Lake
Turkana (b)

Lake UGANDA
Albert
(c)
KENYA

Kampala

Lake
FIGUR E 25–20 Use of repetitive DNA elements to trace spe-
Victoria Nairobi
RWANDA cies relationships in cichlids from Lake Tanganyika. (a) Photo-
Lake
Kivu Kigali graph of an agarose gel showing PCR fragments generated from
primers that flank the AFC family of SINES. Large fragments
BURUNDI containing this family are present in all samples of genomic
Bujumbura DNA from members of the Lamprologini tribe of cichlids (lanes
9–20). DNA from the species in lane 2 has a similar but shorter
fragment, which may represent another repetitive sequence.
DNA from other species (lanes 3–8 and 21–24) produce short,
TANZANIA non-SINE-containing fragments. (b) A Southern blot of the gel
Kalemie Lake from (a) probed with DNA from the AFC family of SINES. AFC
Tanganyika is present in DNA from all species in the Lamprologini tribe
(lanes 9–20), but not in DNA from other species (lanes 2–8 and
21–24). (c) A second Southern blot of the gel from (a) probed
with the genomic sequence at which the AFC SINE inserts. All
species examined (lanes 2–24) contain the insertion sequence.
Mbeya The larger fragments in Lamprologini DNA (lanes 9–20) cor-
respond to those in the previous blot, showing that the SINE is
inserted at the same site in all tribal species. In sum, these data
F I G U R E 2 5 –19 The lakes in the Rift Valley of east Africa are show that the AFC SINE is present only in members of this tribe,
home to most species of cichlids, a fresh water fish. is present in all member species, and is inserted at the same site
in all cases. These results are interpreted as showing a common
origin for the species of the tribe in question.
genetic divergence, reproductive isolation, and speciation
can be surprisingly rapid. irreversible event. If a SINE is present at the same locus in
The Rift Valley lakes of east Africa (Figure 25–19) sup- the genome of all species examined, this is strong evidence
port hundreds of species of cichlid fish. Cichlids are highly that all of those species descend from a common ancestor.
specialized for different niches. Some eat algae floating on Using a SINE called AFC, Okada’s team screened 33 species
the water’s surface, whereas others are bottom feeders, insect of cichlids belonging to four tribes. In each tribe, the SINE
feeders, mollusk eaters, and predators on other fish species. was present at all the sites tested, indicating that the species
Lake Tanganyika is an old lake, formed about 20 million years in each tribe are descended from a single ancestral species
ago and more than 200 species of cichlids have been identi- (Figure 25–20). Cichlid speciation in Lake Tankanyika has
fied in this lake, with many more remaining to be discovered. taken place over several million years and produced the most
Genetic analysis indicates that a majority of the species in diverse collection of species of this fresh water fish.
this lake are descended from a single common ancestor and In contrast, Lake Victoria, a neighboring Rift Valley lake,
that they diverged within their home lake. is only about 400,000 years old and has dried out several
The cichlid species in Lake Tanganyika are organized into times in its history, most recently, about 15,000 years ago. The
16 groups called tribes; all species in a tribe are thought to be 500 or so cichlid species in this lake today are thought to have
descended from a common ancestral species. In a study of diversified from a very small number of founding species. If
cichlid origins in Lake Tanganyika, Norihiro Okada and col- true, this means that new species were formed at a very rapid
leagues examined the insertion of a novel family of repetitive rate, averaging 1 species about every 30 years. If further re-
DNA sequences called short interspersed elements (SINES) search using SINES and other molecular markers can con-
(discussed in Chapter 11) into the genomes of cichlid species firm the number of founding species and establish their tribal
in Lake Tanganyika. SINES are a type of retroposon, and the organization, it would provide evidence for the fastest evolu-
random integration of a SINE at a locus is most likely an tionary divergence of species ever documented in vertebrates.
716 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

25.10 TA BLE 25.5

Amino Acid Differences and Minimal Mutational Distances

Phylogeny Can Be Used to Analyze between Cytochrome c in Humans and Other Organisms
Evolutionary History (a) Amino Acid (b) Minimal Mutational
Organism Differences Distance
Speciation is associated with changes in the genetic structure Human 0 0
of populations and with genetic divergence of those popula- Chimpanzee 0 0
tions. Therefore, we should be able to use genetic differences Rhesus monkey 1 1
among present-day species to reconstruct their evolutionary
Rabbit 9 12
histories, or phylogenies. These relationships are most often
Pig 10 13
presented in the form of phylogenetic trees (Figure 25–21),
which show the ancestral relationships among a group of Dog 10 13
organisms. These groups can be species, or larger groups. In Horse 12 17
a phylogenetic tree, branches represent lineages over time. Penguin 11 18
The length of a branch can be arbitrary, or it can be derived

©AAAS
Moth 24 36
from a time scale, showing the length of time between spe- Yeast 38 56
ciation events. Points at which lines diverge, called nodes, Source: From W.M. Fitch and E. Margoliash, Construction of phylogenetic trees,
show when a species split into two or more species. Each Science 155: 279–284, January 20, 1967. Reprinted with permission from AAAS.
node represents a common ancestor of the species diverging
at that node. The tips of the branches represent species (or
a larger group) alive today (or those that ended in extinc-
tion). Groups that consist of an ancestral species and all its
descendants are called a monophyletic group. The root of a Column (a) of Table 25.5 shows the number of amino
phylogenetic tree represents the oldest common ancestor to acid differences between cytochrome c in humans and in
all the groups shown in the tree. various other species. The data in this table are broadly con-
sistent with our intuitions about how closely related we are
Constructing Phylogenetic Trees to these other species. For example, most people would agree
from Amino Acid Sequences that we are more closely related to other mammals than we
In an important early example of phylogenetic reconstruction, are to insects, and we are more closely related to insects than
W. M. Fitch and E. Margoliash assembled data on the amino we are to yeast. Accordingly, our cytochrome c differs in 10
acid sequence for cytochrome c in a variety of organisms. amino acids from that of dogs, in 24 amino acids from that
Cytochrome c is a component of the electron transport of moths, and in 38 amino acids from that of yeast.
chain in mitochondria, and its amino acid sequence has Amino acid changes are the product of nucleotide
evolved very slowly. For example, its amino acid sequence changes, and more than one nucleotide change may be re-
in humans and chimpanzees is identical, and humans and quired to change a given amino acid. When the necessary
rhesus monkeys show only one amino acid difference. This nucleotide changes that account for all amino acid differ-
similarity is remarkable considering that the fossil record ences in a protein are totaled, the minimal mutational dis-
indicates that the lines leading to humans and monkeys di- tance between the genes of any two species is established.
verged from a common ancestor approximately 20 million Column (b) in Table 25.5 shows such an analysis of the
years ago. genes encoding cytochrome c. As expected, these values are
larger than the corresponding number of amino acids sepa-
rating humans from the other nine organisms listed.
Branch
Species A Data on the minimal mutational distances between the
cytochrome c genes of 19 organisms was used to reconstruct
Species B
Root their evolutionary history. The result is a phylogenetic
Node
Species C tree showing the relationships among the species studied
Tip (Figure 25–22). The black dots on the tips of the branches
represent existing species, whose inferred common ances-
FIG U R E 2 5 –21 Elements of a phylogenetic tree showing the tors are linked to them by green lines that diverge at nodes
relationships among species. The root represents a common (red dots). The common ancestors are connected to still
ancestor to all species on the tree. Branches represent lineages
earlier common ancestors, culminating in a single common
through time. The points at which branches separate are called
nodes, and at the tips of branches are living species (or those ancestor for all the species on the tree, represented by the
that have gone extinct. red dot on the extreme left.
 25. 1 0 PHYLOG E NY C AN BE U S E D TO A NA LYZ E E VOLU TIONARY HIS TO R Y 717

©AAAS

8
Human

0.
Monkey

0.
6.9

2
Dog
3.0

9
Horse

0.
1.4 1.
7 2.9 Donkey

0.
1.3

1
1.
–0.6 2.7 Pig

4
1.1
4.6 Rabbit
Kangaroo
3.
3.3 3 Pigeon
1.6
1 . 1 Duck
–0.2 1.2 1 .0 Chicken

.5
1.0
16.5 1.1 Penguin
5.4 17.2
4.9
Turtle
Snake
5.7 Tuna
15.2
6.5 Screwworm fly
9.9
Moth
2.1 28.1 Bread mold
17.4 Yeast
9.6
30 25 20 15 10 5 0
Average minimal mutation distance

F I G U R E 2 5 –2 2 Phylogenetic tree constructed by comparing homologies in cytochrome c amino acid sequences. Reprinted with
permission from Fitch, W.M. and Margoliash, E. 1967. Construction of phylogenetic trees. Science 279: 279–284, Figure 2.

Molecular Clocks Measure Molecular clocks must be carefully calibrated and used
the Rate of Evolutionary Change with caution. For example, the data indicate that strains of
In many cases, we would like to estimate not only which influenza A that jump from birds to humans have evolved
members of a set of species are most closely related, but also much more rapidly than strains that have remained in birds.
when their common ancestors lived. Sometimes we can do Hence, a molecular clock calibrated from human strains
so, thanks to molecular clocks—amino acid sequences or of the virus would be highly misleading if applied to bird
nucleotide sequences in which evolutionary changes accu- strains.
mulate at a constant rate over time.
Research on the influenza A virus shows how molecular Analysis of Genetic Divergence
clocks are constructed and used. Parts of the hemagglutinin between Neanderthals and
gene from viral strains isolated at various times over a 20- Modern Humans
year period were sequenced. A phylogenetic tree was con- Fossil evidence indicates that the Neanderthals, Homo ne-
structed using the number of nucleotide differences among anderthalensis, lived in Europe and western Asia from some
these strains [Figure 25–23(b)]. Most of the strains are now 300,000 years ago until they disappeared about 30,000 years
extinct and have no descendants. Plotting the number of ago. For at least 30,000 years, Neanderthals coexisted with
nucleotide substitutions between strains against the year anatomically modern humans (Homo sapiens) in several re-
in which they were isolated shows that the points fall very gions. Genetic analysis using DNA sequencing has helped
close to a straight line [Figure 25–23(a)]. This means that answer several questions about Neanderthals and modern
the nucleotide substitutions in the hemagglutinin gene have humans: (1) Were Neanderthals direct ancestors of modern
accumulated at a steady rate, and thus serve as a molecular humans? (2) Did Neanderthals and our species, H. sapiens,
clock to track the evolution of the influenza A virus. interbreed, so that our genomes carry Neanderthal genes?
This molecular clock is also used to compare the se- Or did the Neanderthals die off and become extinct leaving
quence of the hemagglutinin gene of new flu viruses as they no genetic heritage? (3) What can we say about the simi-
appear each year and to estimate the time that has passed larities and differences between our genome and that of the
since each diverged from a common ancestor. Neanderthals?
718 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

(a) (b)
120
WYO87

HST85

23
18
100 EQU86

15
7
ALA81

80 11 12 BKK79
6
3
Nucleotide substitutions

BKK79 1
1 ENG77
TX77
15 6 7

60 VIC76
4
5 ALE76
2 VIC75

8 8
17

6
40
PTC73 10
4
5 MEM72
2
ENG72
5 7 MEM71
20 HKK71 QNS70
8
13 5
16 6 ENG69
6
AIC68 2
2 2 1NT68 1
0
65 70 75 80 85 90
Year of isolation

FIG U R E 2 5 –2 3 Phylogenetic molecular clock in the influenza A hemagglutinin gene. (a) Number of nucleotide differences between
the first isolate as a function of year of isolation. (b) Estimate of the phylogeny of the isolates.

To answer these and other questions, researchers used Neanderthals were a separate hominid line and did not con-
two approaches, both of which involved sequencing DNA tribute mitochondrial genes to H. sapiens.
recovered from Neanderthal bones. One approach ana-
lyzed mitochondrial DNA recovered from skeletons found The Neanderthal Genome Project
in Feldhofer cave in Germany and Mezmaiskaya cave in the The second approach to studying Neanderthal DNA was an
Caucasus Mountains east of the Black Sea. The Feldhofer ambitious project to isolate and sequence the Neanderthal
DNA and mitochondrial DNA (mtDNA) from more than genome. The international research team began by isolat-
2000 present-day humans were used to construct a phyloge- ing DNA from three female Neanderthals whose skeletons
netic tree [Figure 25–24(a)]. From the structure of the tree, were recovered from a cave in Croatia. Using newly devel-
researchers concluded that Neanderthals are a distant rela- oped DNA sequencing techniques, they assembled a draft
tive of modern humans. The mtDNA recovered from the sequence of the Neanderthal genome. As part of the project,
Mezmaiskaya skeleton was sequenced and compared to the they also sequenced the genomes of five individuals living in
sequence from the Feldhofer skeleton. Although the two Ne- different parts of the world.
anderthal mtDNA sequences are from locations more than The Neanderthal genome contains about 3.2 billion
1000 miles apart, they vary by only about 3.5 percent. This base pairs, the same as the genome of our species, and is
indicates that the Neanderthal samples derive from a single 99.7 percent identical to our genome. Both the Neanderthal
gene pool. Phylogenetic analysis comparing the two Nean- genome and our genome are about 98.8 percent identical to
derthal sequences with those of modern humans and chim- that of the chimpanzee.
panzees places the Neanderthals in a group that is clearly By comparing the Neanderthal genome sequence with
distinct from modern humans [Figure 25–24(b)]. The con- the genomes of five present-day humans and with the chim-
clusion from the study of mitochondrial DNA is that while panzee genome, researchers were able to identify amino
Neanderthals and humans have a common ancestor, the acid-coding differences in 78 genes that developed after the
 25. 1 0 PHYLOG E NY C AN BE U S E D TO A NA LYZ E E VOLU TIONARY HIS TO R Y 719

(a) (b) to African genomes were detected, but it is

Neanderthal
possible that a larger sampling of African
populations will determine whether some
5846 Modern
Africans humans Africans carry Neanderthal genes.
and Comparative genomics combined with
non-Africans the fossil record has allowed scientists to
Mezmaiskaya
construct a phylogenetic tree showing the
1 African Feldhofer pattern and times of divergence of Nean-
1 African derthals and our species (Figure 25–25). As-
1 African-American Chimpanzees suming that chimpanzees and humans last
4 Africans
shared a common ancestor about 6.5 million
© 1997 Elsevier © 2000 Macmillan Publishers Ltd.
years ago, the tree shows that Neanderthals
F I G U R E 2 5 –2 4 (a) A phylogenetic tree estimated from mitochondrial DNA se-
and humans last shared a common ancestor
quences of one Neanderthal and over 2000 modern humans. From Krings, M. et al., about 706,000 years ago and that the isolat-
1997. Cell 90: 19–30. Figure 7A, p. 26, copyright 1997 with permission from Elsevier; ing split between Neanderthals and human
(b) A phylogenetic tree estimated from analysis of over 5000 modern humans, the populations occurred about 370,000 years
Neanderthal samples from the Feldhofer cave and the Mezmaiskaya cave, and from ago. From these studies, several conclusions
chimpanzees.
can be drawn. First, Neanderthals are not
From Ovchinnikov, I.V. et al., 2000. Molecular analysis of Neanderthal DNA from the northern Caucasus. Nature
404: 490–493 copyright 2000 Macmillan Publishers Ltd. direct ancestors of our species. Second, Ne-
anderthals and members of our species may
split between the Neanderthal and human lineages. These have interbred, but from the results available, it appears
genes are involved in cognitive development, skin morphology, that Neanderthals did not make major contributions to our
and skeletal development. In addition, there is evidence that genome. As a species, Neanderthals are extinct, but some of
several regions of the modern human genome have under- their genes survive as part of our genome. Third, from what
gone positive selection in the interval since Neanderthals we know about the Neanderthal genome, we share most
and modern humans last shared a common ancestor. genes and other sequences with them.
Perhaps the most unexpected finding in this analysis More exciting answers to the question about the simi-
is that the genomes of present-day humans in Europe and larities and differences between our genome and that of
Asia, but not Africa, are composed of 1–4 percent Neander- Neanderthals will be derived from further analysis and will
thal sequences. Interbreeding with Neanderthals may have allow us to identify key differences that define our spe-
occurred somewhere in the Middle East, before humans mi- cies and, in the process, revolutionize the field of human
grated into Europe and Asia. No Neanderthal contributions evolution.

~370,000 y.a.
Split of ancestral ~41,000 y.a.
human and Neanderthal Earliest modern
populations humans in Europe

706,000 y.a. ~195,000 y.a.

Coalescence of Earliest known ~28,000 y.a.
human and Neanderthal anatomically Most recent known
reference sequences modern humans Neanderthal remains

Modern
human

FIGUR E 25–25 Estimated times of

divergence of human and Neanderthal
genomic sequences and, subsequently, of
Neanderthal their populations, relative to landmark
events in both human and Neanderthal
Genomic data
evolution. These estimates are based on
Fossil data ©AAAS
sequencing about 65,000 base pairs of
Neanderthal DNA (y.a. = years ago).
Evolutionary lineage of human and Neanderthal reference sequences
From Noonan, J.P. et al. 2006. Sequencing and analysis
Evolutionary lineage of ancestral human and Neanderthal populations of Neanderthal genomic DNA. Science 314: 1113–1118.
720 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

G E N E T I C S , T E C H N O L O G Y, A N D S O C I E T Y

Tracking Our Genetic Footprints out of Africa

O ver the last century, paleoan-

thropologists have applied a
variety of sophisticated scien-
tific tools to explore our origins and hu-
man kinships.
Asia around 60,000 years ago. By about
40,000 years ago, populations of H. sapiens
reached Australia and later migrated into
North America. In the out-of-Africa model,
H. sapiens replaced all the preexisting
debated. As methods to sequence DNA
from ancient fossils improve, it may be
possible to fill the gaps in the genetic path-
way leading out of Africa and to resolve
those age-old questions about our origins.
Based on the physical traits and distribu- H. erectus types, without interbreeding. In
tion of hominid fossils, most paleoanthro- this way, H. sapiens became the only species Your Turn
pologists agree that a large-brained, tool-using in the genus by about 30,000 years ago.
1. The sequencing of Neanderthal DNA
hominid they call Homo erectus appeared in Although the out-of-Africa hypothesis
shows that it is so different from ours
east Africa about 2 million years ago. This is still debated, most genetic evidence ap-
that Neanderthals may have been a
species used simple stone tools and hunted, pears to support it. Humans all over the
separate species and that Neanderthals
but did not fish, build houses or fireplaces, globe are remarkably similar genetically.
and H. sapiens diverged about 600,000
or follow ritual burial practices. About 1.7 DNA sequences from any two people cho-
years ago. However, the debate contin-
million years ago, H. erectus spread into Eur- sen at random are 99.9 percent identical.
ues about whether H. sapiens and Nean-
asia and south Asia. Most scientists also More genetic identity exists between two
derthals interbred, and if so, to what
agree that H. erectus likely developed into persons chosen at random from a human
extent. Discuss evidence that supports,
several hominid types, including Neander- population than between two chimpan-
and refutes, the idea that Neanderthals
thals (in Europe) and Peking man or Java zees chosen at random from a chimpan-
and H. sapiens interbred.
man (in Asia). These hominids were ana- zee population. Interestingly, about 90
tomically robust, with large, heavy skeletons percent of the genetic differences that Start your investigations by reading Green,
and skulls. Neanderthals and other H. erectus do exist occur between individuals rather R.E. et al. 2010. A draft sequence of the
groups disappeared 50,000 to 30,000 years than between populations. This unusu- Neanderthal genome. Science 328: 710–722.
ago—around the same time that anatomi- ally high degree of genetic relatedness in
2. If all people on Earth are very similar
cally modern humans (H. sapiens) appeared all humans around the world supports the
genetically, how did we come to have
all over the world. It is at this point in our idea that our species arose recently from a
such a range of physical differences,
history—when ancient hominids gave way to small founding group of humans.
which some describe as racial differ-
anatomically modern humans—that contro- Studies of mitochondrial DNA sequences
ences? How has modern genomics
versy arises. from current human populations reveal
contributed to the debate about the
At present, two main hypotheses ex- that the highest levels of genetic variation
validity and definition of the term race?
plain the origins of modern humans: the occur within African populations. Afri-
multiregional hypothesis and the out-of- cans show twice the mitochondrial DNA- An interesting discussion of race and how
Africa hypothesis. The multiregional hy- sequence diversity of non-Africans. This genomic studies are changing our concepts of
pothesis is based primarily on archaeolog- implies that the earliest branches of human ancestry can be found on the Wikipedia
ical and fossil evidence. It proposes that H. sapiens diverged in Africa and had a site: http://en.wikipedia.org/wiki/Race_
H. sapiens developed gradually and simul- longer time to accumulate mitochondrial (classification_of_humans). A study of hu-
taneously all over the world from existing DNA mutations, which are thought to man population structure, based on microsatellite
H. erectus groups, including Neanderthals. accumulate at a constant rate over time. analysis, can be found at Rosenberg, N.A. et
Interbreeding between these groups even- DNA sequences from mitochondrial, al. 2002. Genetic structure of human pop-
tually made H. sapiens a genetically homo- Y-chromosome, and chromosome-21 ulations. Science 298: 2381–2385.
geneous species. Natural selection over markers support the idea that human
3. Geneticists study mitochondrial and Y-
1.5 million years then created the regional roots are in east Africa and that the mi-
chromosome DNA to determine the an-
variants that we see today. In the multi- gration out of Africa occurred through
cestry of modern humans. Why are these
regional view, our genetic makeup should Ethiopia, along the coast of the Arabian
two types of DNA used in lineage studies?
include contributions from many H. erectus Peninsula, and outward to Eurasia and
What is meant by the terms mitochondrial
groups, including Neanderthals. In con- Southeast Asia. Recent data based on
Eve and Y-chromosome Adam?
trast, the out-of-Africa hypothesis, based nuclear microsatellite variants and whole-
primarily on genetic analyses of modern genome single nucleotide polymorphism To read the original paper hypothesizing a
human populations, contends that H. (SNP) analysis further support the notion mitochondrial Eve, see Cann, R.L. et al. 1987.
sapiens evolved from the descendants of that humans migrated out of Africa and Mitochondrial DNA and human evolu-
H. erectus in sub-Saharan Africa about dispersed throughout the world from a tion. Nature 325: 31–36. For a discussion of
200,000 years ago. A small band of H. small founding population. Y-chromosome Adam, see Gibbons, A. 1997.
sapiens (fewer than 1000) then left Africa, As with any explanation of human ori- Y Chromosome shows that Adam was an
expanded, and migrated into Europe and gins, the out-of-Africa hypothesis is actively African. Science 278: 804–805.
 INS IG HTS A ND S OLU TIONS 721

CASE STUDY An unexpected outcome

A
newborn screening program identified a baby with a rare 1. Since it appears that the unaffected parents are heterozy-
autosomal recessive disorder called arginosuccinate ac- gotes, would it be considered unusual that there would be no
iduria (AGA), which causes high levels of ammonia to family history of the disorder? How would they be counseled
accumulate in the blood. Symptoms usually appear in the first about risks to future children?
week after birth and can progress to include severe liver damage,
2. If the disorder is so rare, what is the frequency of heterozy-
developmental delay, and mental retardation. AGA occurs with
gous carriers in the population?
a frequency of about 1 in 70,000 births. There is no history of
this disorder in either the father’s or mother’s family. This case 3. What are the chances that two heterozygotes will meet and
raises several questions: have an affected child?

Summary Points For activities, animations, and review quizzes, go

to the study area at www.masteringgenetics.com

1. Genetic variation is a characteristic of most populations. In and genetic drift. Nonrandom mating changes genotype fre-
some cases, this can be observed at the phenotypic level, but quencies but does not change allele frequencies.
analysis at the amino acid and nucleotide levels provides a 6. The formation of new species depends on the formation of
more direct way to estimate genetic variation. subpopulations and the accumulation of enough genetic dif-
2. Using the assumptions of the Hardy–Weinberg Law, it is pos- ferences that, when reunited, members of the separated popu-
sible to estimate allele and gene frequencies in populations. lations cannot interbreed.
3. The Hardy-Weinberg Law can be applied to determining allele 7. Phlyogenetic analysis using morphology, amino acid sequences,
and genotype frequencies for multiple alleles and X-linked al- or nucleotide sequences can be used to construct phylogenetic
leles, as well as calculating the frequency of heterozygotes for a trees showing the evolutionary relationships among a group of
given gene. organisms. When calibrated with molecular clocks, the evolu-
4. Natural selection changes allele frequency in populations lead- tionary changes on a phylogenetic tree can be calibrated with a
ing to evolutionary change. Selection for quantitative traits can time scale.
involve directional selection, stabilizing selection, or disruptive 8. Phylogenetic analysis combined with genome sequence data
selection. from humans, Neanderthals, and other hominids has helped
5. In addition to natural selection, other forces act on allele fre- scientists reconstruct the relationship between Neanderthals
quencies in populations. These include mutation, migration, and our species.

INSIGHTS AND SOLUTIONS

1. Tay–Sachs disease is caused by loss-of-function mutations in 2. A single plant twice the size of others in the same population
a gene on chromosome 15 that encodes a lysosomal enzyme. suddenly appears. Normally, plants of that species reproduce
Tay–Sachs is inherited as an autosomal recessive condition. by self-fertilization and by cross-fertilization. Is this new gi-
Among Ashkenazi Jews of Central European ancestry, about 1 ant plant simply a variant, or could it be a new species? How
in 3600 children is born with the disease. What fraction of the would you determine which it is?
individuals in this population are carriers? Solution: One of the most widespread mechanisms of spe-
Solution: If we let p represent the frequency of the wild-type ciation in higher plants is polyploidy, the multiplication of
enzyme allele and q the total frequency of recessive loss-of- entire sets of chromosomes. The result of polyploidy is usu-
function alleles, and if we assume that the population is in ally a larger plant with larger flowers and seeds. There are
Hardy–Weinberg equilibrium, then the frequencies of the two ways of testing the new variant to determine whether
genotypes are given by p2 for homozygous normal, 2pq for it is a new species. First, the giant plant should be crossed
carriers, and q2 for individuals with Tay–Sachs. The frequency with a normal-sized plant to see whether the giant plant
of Tay–Sachs alleles is thus produces viable, fertile offspring. If it does not, then the two
1 different types of plants would appear to be reproductively
q = 2q2 = = 0.017 isolated. Second, the giant plant should be cytogenetically
A 3600
screened to examine its chromosome complement. If it has
Since p + q = 1, we have twice the number of its normal-sized neighbors, it is a tet-
p = 1 - q = 1 - 0.017 = 0.983 raploid that may have arisen spontaneously. If the chromo-
Therefore, we can estimate that the frequency of carriers is some number differs by a factor of two and the new plant is
reproductively isolated from its normal-sized neighbors, it
2pq = 2(0.983)(0.017) = 0.033 or about 1 in 30 is a new species.
722 25 P OP ULATION AN D EVOLUTION ARY GENE TIC S

Problems and Discussion Questions For instructor-assigned tutorials and

problems, go to www.masteringgenetics.com

9. Determine whether the following two sets of data represent

HOW DO WE KNOW ?
1. In this chapter we focused on how changes in gene frequency
populations that are in Hardy–Weinberg equilibrium (use x2
analysis if necessary):
are related to the process of species formation. At the same time, (a) CCR5 genotypes: 1/1, 60 percent; 1/32, 35.1 percent;
we found many opportunities to consider the methods and rea- 32/32, 4.9 percent
soning by which much of this information was acquired. From (b) Sickle-cell hemoglobin: AA, 75.6 percent; AS, 24.2 percent;
the explanations given in the chapter, what answers would you SS, 0.2 percent
propose to the following fundamental questions? 10. If 4 percent of a population in equilibrium expresses a recessive
(a) How do we determine how much genetic variation exists trait, what is the probability that the offspring of two individu-
in a population? als who do not express the trait will express it?
(b) What evidence has been obtained from laboratory stud- 11. Consider a population in which the frequency of allele A is
ies to indicate that natural selection is the cause of genetic
p = 0.7 and the frequency of allele a is q = 0.3, and in which
differences among natural populations of a species?
the alleles are codominant. What will be the allele frequencies
(c) How can the minimum genetic divergence between two
after one generation if the following occurs?
species be determined?
(a) wAA = 1, wAa = 0.9, and waa = 0.8
(d) How can we determine the last common ancestor shared
(b) wAA = 1, wAa = 0.95, and waa = 0.9
by two divergent species?
(c) wAA = 1, wAa = 0.99, waa = 0.98
2. What types of nucleotide substitutions will not be detected by
(d) wAA = 0.8, wAa = 1, waa = 0.8
amino acid sequencing of a gene’s protein product?
12. If the initial allele frequencies are p = 0.5 and q = 0.5 and al-
3. The genetic difference between two Drosophila species, D.
lele a is a lethal recessive, what will be the frequencies after 1, 5,
heteroneura and D. sylvestris, as measured by nucleotide diver-
10, 25, 100, and 1000 generations?
sity, is about 1.8 percent. The difference between chimpanzees
13. Under what circumstances might a lethal dominant allele persist
(P. troglodytes) and humans (H. sapiens) is about the same, yet
in a population?
the latter species are classified in different genera. In your opin-
14. Assume that a recessive autosomal disorder occurs in 1 of
ion, is this valid? Explain why.
10,000 individuals (0.0001) in the general population and that
4. The use of nucleotide sequence data to measure genetic vari-
in this population about 2 percent (0.02) of the individuals are
ability is complicated by the fact that the genes of higher
carriers for the disorder. Estimate the probability of this disor-
eukaryotes are complex in organization and contain 5 and
der occurring in the offspring of a marriage between first cous-
3 flanking regions as well as introns. Researchers have com-
ins. Compare this probability to the population at large.
pared the nucleotide sequence of two cloned alleles of the
15. One of the first Mendelian traits identified in humans was a
g-globin gene from a single individual and found a variation
dominant condition known as brachydactyly. This gene causes
of 1 percent. Those differences include 13 substitutions of one
an abnormal shortening of the fingers or toes (or both). At the
nucleotide for another and 3 short DNA segments that have
time, some researchers thought that the dominant trait would
been inserted in one allele or deleted in the other. None of the
spread until 75 percent of the population would be affected
changes takes place in the gene’s exons (coding regions). Why
(because the phenotypic ratio of dominant to recessive is 3:1).
do you think this is so, and should it change our concept of
Show that the reasoning was incorrect.
genetic variation?
16. Achondroplasia is a dominant trait that causes a characteristic
5. Calculate the frequencies of the AA, Aa, and aa genotypes after
form of dwarfism. In a survey of 50,000 births, five infants with
one generation if the initial population consists of 0.2 AA, 0.6
achondroplasia were identified. Three of the affected infants
Aa, and 0.2 aa genotypes and meets the requirements of the
had affected parents, while two had normal parents. Calculate
Hardy–Weinberg relationship. What genotype frequencies will
the mutation rate for achondroplasia and express the rate as the
occur after a second generation?
number of mutant genes per given number of gametes.
6. Consider a rare disorder in a population caused by an autoso-
17. A recent study examining the mutation rates of 5669 mamma-
mal recessive mutation. From the frequencies of the disorder in
lian genes (17,208 sequences) indicates that, contrary to popu-
the population given, calculate the percentage of heterozygous
lar belief, mutation rates among lineages with vastly different
carriers:
generation lengths and physiological attributes are remarkably
(a) 0.0064
constant (Kumar, S., and Subramanian, S. 2002. Proc. Natl.
(b) 0.000081
Acad. Sci. [USA] 99: 803–808). The average rate is estimated
(c) 0.09
at 12.2  10–9 per bp per year. What is the significance of this
(d) 0.01
finding in terms of mammalian evolution?
(e) 0.10
18. List the barriers that prevent interbreeding and give an example
7. What must be assumed in order to validate the answers in
of each.
Problem 6?
19. What are the two groups of reproductive isolating mechanisms?
8. In a population where only the total number of individuals
Which of these is regarded as more efficient, and why?
with the dominant phenotype is known, how can you calculate
20. What is the neutral theory of molecular evolution as proposed
the percentage of carriers and homozygous recessives?
by Kimura (1968)? What determines the frequency of neutral
alleles in a population?