Evolution, 4th Edition

182 CHAPTER 7

in a shark. But while the protein’s similarity does not correlate with the habitat that

these animals live in, it does correlate strongly with their phylogenetic relations.

Fossils show when the most recent common ancestors of these animals lived. The

number of differences between the hemoglobins of two species correlates very well

with the age of their most recent ancestors. These data show that changes in hemo-

globin molecules accumulate in time at a nearly constant rate. FIGURE 7.18 shows

plots in which differences in DNA and protein sequences between humans and nine

other mammals are compared with the time since our most recent common ances-

tors. Again, there is a very strong correlation between divergence time and sequence

differences. The figure also shows that different types of changes accumulate at dif-

ferent rates: synonymous differences build up about five times more rapidly than

nonsynonymous differences. But while these two types of changes evolve at differ-

ent rates, within each type the rate is quite constant.

Genes and proteins that evolve at roughly constant rates provide us with molec-

ular clocks. We use these molecules to learn about the evolutionary histories of

species that left no fossils. The idea is simply to count up the number of amino acid

differences between the hemoglobins of two species, then use a plot like that in

Figure 7.18 to estimate when their most recent common ancestor lived (see Figure

2.17). This strategy can be used with many proteins other than hemoglobin, and

with DNA sequences as well as amino acid sequences. Extensions of this approach

that use more sophisticated statistical analyses are the main tools used to construct

the phylogenetic tree of life that links all species on Earth (see Chapters 2 and 17).

The neutral theory of molecular evolution

What could explain the relatively constant evolutionary rates of molecules that

evolve in a clocklike way? Motoo Kimura, one of the giants in the history of evo-

lutionary biology, proposed the neutral theory of molecular evolution. One of the

many predictions that flows from that theory is that random genetic drift will cause

genes to evolve at relatively constant rates [17]. Kimura thought that beneficial

mutations are so rare that they make only a trivial contribution to the molecular

differences among species. Instead, Kimura believed that the vast majority of those

differences result from the fixation of neutral mutations (those with s << 1/Ne).

What does this theory imply about molecular clocks? Consider a locus that is

evolving by drift in two species that split apart t generations ago. We sample one

copy of the gene from each of the species. In the branch of the gene tree that leads

from the first species back to the most recent common ancestor of the two species,

on average μn t mutations will have occurred, where once again μn is the rate at

which neutral mutations occur at the locus per generation. Likewise, on average

μn t mutations will have occurred on the branch leading back from the second spe-

cies. If the amount of time, t, is not too long, there is a negligible chance that a

mutation will happen twice at any single DNA site. We therefore expect that the

total number of differences between the two copies of the gene, d, will equal the

total number of mutations that have occurred along the two branches. That is simply

d = 2 μn t (7.2)

Thus for genes that are accumulating differences by drift, the number of differ-

ences between two species is proportional to the time since their most recent com-

mon ancestor. This gives us a molecular clock: a linear relation between time and

divergence, consistent with what we see in Figure 7.18.

Recall from earlier in this chapter that purifying selection reduces the neu-

tral mutation rate. Equation 7.2 predicts that if most differences among species

are selectively neutral, sites in the genome that are free of purifying selection will

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

Evolution4e_07.18.ai Date 11-14-2016 01-18-17 01-24-17

Substitutions (per site)

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0

1.6

100 200

1

2

3

4

5

6

7

8

9

300

Divergence time (My) based

on fossil record

400 500

dS

dA

dN

FIGURE 7.18 Molecular clocks based on

three kinds of data run at different rates.

The x-axis shows time since the most

recent common ancestor of humans and

nine other vertebrates, as determined

from fossils. The y-axis plots the number of

amino acid substitutions (dA), number of

synonymous DNA substitutions (dS), and

number of nonsynonymous DNA substitu-

tions (dN) per site between humans and

those species, based on 4198 loci. Hu-

mans are compared with: 1 = chimpanzee,

2 = orangutan, 3 = macaque, 4 = mouse, 5

= cow, 6 = opossum, 7 = chicken, 8 = west-

ern clawed frog, 9 = zebrafish. (After [29].)

07_EVOL4E_CH07.indd 182 3/23/17 9:09 AM