Evolution, 4th Edition

THE GENETiCAl THEoRy of NATuRAl SElECTioN 107

A key point is that natural selection and evolution are not the same thing. If

selection on a trait occurs but the trait is not inherited, then evolution will not hap-

pen. You will see in later chapters that natural selection is not the only factor that

can cause evolution, and so evolution can also occur even if selection does not. But

if the twin conditions listed above are met, then the population will change across

generations. To make that basic idea more quantitative, let’s now consider how to

measure selection, account for inheritance, and predict the outcome of evolution.

fitness: The Currency of Selection

To understand how evolution by selection works, we need a way to measure selec-

tion. An individual’s absolute fitness is the number of zygotes (offspring) produced

over its lifetime. If an immature butterfly is eaten by a bird, its fitness is 0. If an oak

tree survives many years, its fitness could easily be in the millions. (Just think of

the number of acorns produced by a big tree.) It is also useful to consider the fit-

ness of an allele, a genotype, or a phenotype. In those cases, fitness is the average

of the fitnesses of the individuals with that allele, genotype, or phenotype. We use

the symbol W to represent absolute fitness.

Over the life cycle of any organism, many events affect fitness. An embryo may

develop to sexual maturity or die before that happens. If it survives to maturity, the

individual may or may not be able to mate. If he or she does successfully mate, the

individual’s gametes may or may not be successful at fertilization. These various

events are called fitness components (FIGURE 5.5). They can be divided even more

finely, or grouped together, depending on the application. Often it is useful to sim-

plify our thinking by viewing fitness as the product of just two fitness components:

W = (probability that the individual survives to maturity)

× (expected number of offspring if the individual does survive)

(5.1)

The strength of selection is determined by fitness differences. It is the relative (or

proportional) differences that matter. To see why this is so, consider an individual

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

Evolution4e_05.04.ai Date 12-12-2016 01-06-17

(A) Without selection (B) With selection

Mean size of parents

Mean size of offspring

Die Survive

O

P P

O

FIGURE 5.4 Evolution results when there is a correlation between the phenotypes of

parents and offspring and a correlation between the phenotypes of parents and their

fitness. Each dot represents a family. The horizontal axis shows the mean trait (in this case,

size) of the two parents, and the vertical axis the mean trait of their offspring. In both

panels, there is a correlation between the traits of parents and their offspring. (A) With no

selection, there is no evolution. The mean value of all offspring in the next generation,

O—, is the same as the mean value of all parents in the current generation, P—. No change

has happened. (B) When selection acts, evolution results. In this example, only the largest

individuals survive and become parents (solid circles to the right of the vertical dashed

line) that leave offspring. The mean trait in the offspring of the next generation is larger

than the mean of the parents before selection acted in the current generation.

FIGURE 5.5 Selection can affect fitness at

different points during the life cycle. This

sketch shows the life cycle of a sexually

reproducing species with four fitness com-

ponents that affect the number of descen-

dants that an individual leaves to the next

generation. (After [10].)

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

Evolution4e_05.05_v1.ai Date 12-29-2016

Zygotes

Parents

Adults

Females Males

Gametes

Fertilization

success

Viability

Mating

success

Fecundity

05_EVOL4E_CH05.indd 107 3/23/17 9:01 AM