Evolution, 4th Edition

gENETIC dRIfT: EvolUTIoN AT RANdoM 179

The relative contributions that adaptation and drift make to the differences

among species vary dramatically among groups of organisms [8, 16]. In flies and

bacteria that have very large Ne, about half of the amino acid differences in the

proteins of closely related species evolved by positive selection, that is, by the fixa-

tion of beneficial mutations. The other half of the differences were fixed by drift.

The picture is very different for our own species: only 15 percent (and maybe less)

of the differences between the proteins of humans and macaque monkeys result

from adaptive evolution, and the rest accumulated by drift.

Drift can cause deleterious mutations to spread to fixation. We saw earlier that

human effective population size was roughly 10,000 in our recent evolutionary

past. As a result, many mutations that reduce fitness by s = 10–5 became fixed in

our genomes [1]. The fixation of deleterious mutations is a well-known problem in

the small populations of animals and plants in zoos [10]. The same problem can

also contribute to the extinction of natural populations. The decline of fitness by

the fixation of deleterious mutations in a small population is called the inbreeding

load. (Note that the inbreeding load is distinct from inbreeding depression, which

is the loss of fitness of offspring from parents that are closely related compared

with offspring from unrelated parents; see Chapter 10.) An isolated population of

the adder Vipera berus had fewer than 40 individuals and was highly homozygous

[24]. Females had unusually small litter sizes, and many of their offspring were

deformed or stillborn. Twenty adult males were introduced from other populations,

left to interbreed with the residents for 4 years, and then removed. Soon after, the

population size rebounded dramatically (FIGURE 7.15). The snakes from the other

populations reintroduced the nondeleterious alleles that had been lost by drift,

which led to increased survival of offspring.

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

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

–2

0

1

2

3

4

–1 0 1

Deleterious

(A)

Benecial

Neutral

2

Ne s

Ne = 10,000

Ne = 1000

Relative xation probability

• 10 –4 0 10 –4

(B)

(C)

Selection coefcient, s

Frequency

Frequency

FIGURE 7.14 (A) The probability that a single copy of a mutation becomes fixed depends

on its selection coefficient (s) and the effective population size (Ne). Here we assume that

the relative fitnesses of A 1 A 1 , A 1 A 2 , and A 2 A 2 genotypes are 1, 1 + s, and 1 + 2s. The x-axis

shows the product of Ne and s. The y-axis shows the fixation probability relative to that of

a neutral mutation (whose fixation probability is 1/2Ne). For mutations with |Ne s| << 1, the

fixation probability is very close to that of a neutral mutation. (B) Hypothetical distribution

of fitness effects for new mutations. When Ne = 10,000, mutations with fitness effects in

the range –10–5 < s < 10–5 (shown in gray) evolve almost neutrally. (C) With the same distri-

bution of mutation effects as in (B) but with Ne = 1000, the range of mutations that evolve

almost neutrally expands to –10–4 < s < 10–4.

07_EVOL4E_CH07.indd 179 3/23/17 9:09 AM