Evolution, 4th Edition

536 CHAPTER 20

peak, stabilizing new adaptations that otherwise would slip back due to gene flow

and have only an ephemeral existence.

Trends, Predictability, and Progress

For decades after the publication of On the Origin of Species, many of those who

accepted the historical reality of evolution viewed it as a cosmic history of progress.

As humanity had been the highest earthly link in the pre-evolutionary Great Chain

of Being, just below the angels (see Chapter 1), so humans were seen as the supreme

achievement of the evolutionary process (and Western Europeans as

the pinnacle of human evolution). Darwin distinguished himself from

his contemporaries by denying the necessity of progress or improve-

ment in evolution [24], but almost everyone else viewed progress as an

intrinsic, even defining, property of evolution.

In this section, we examine the nature and possible causes of

trends in evolution and ask whether the concept of evolutionary

progress is meaningful. A trend may be described objectively as a

directional shift over time. “Progress” implies improvement or bet-

terment, which requires a criterion for judging improvement, or a

value judgment of what “better” might mean.

Trends: Kinds and causes

A trend is a persistent, directional change in the average value of a

feature, or perhaps its maximal (or minimal) value, in a clade over

the course of time. It may describe evolution in a particular clade, in

diverse clades, or in all of life. Trends can also be classified as passive

or driven [73]. In a passive trend, lineages in the clade evolve in both

directions with equal probability, but if there is a strong constraint

in one direction (e.g., a minimal possible body size), the variation

among lineages can expand only in the other direction. Because the

variance expands, so do the mean and the maximum. Although the

mean increases, some lineages may remain near the ancestral char-

acter state (FIGURE 20.21A). In a driven, or active, trend, changes in

one direction are more likely than changes in the other (i.e., there is

a “bias” in direction), so both the maximal and the minimal charac-

ter values change along with the mean (FIGURE 20.21B).

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

Evolution4e_20.20.ai Date 12-26-2016

Character divergence

(A) (B) (C)

Time

Character divergence Character divergence

Geographic barrier in place

Barrier breaks down

Prezygotic or postzygotic

reproductive isolation

FIGURE 20.20 A model of how specia-

tion might facilitate long-term evolutionary

change in morphological and other pheno-

typic characters. (A) An allopatric population

diverges in characteristics that adapt it to a

different habitat. It does not evolve prezy-

gotic or postzygotic reproductive isolation,

so when the barrier breaks down and the

populations meet, the divergence may be

lost due to interbreeding. (B) If the diver-

gent population evolves reproductive isola-

tion (red bar) while allopatric, it can become

sympatric with the other form and retain its

distinctive character. (C) The same process

might occur again and again, so that each

speciation event enables more divergence

to evolve and persist indefinitely.

Futuyma Kirkpatrick Evolution, 4e

Sinauer Associates

Troutt Visual Services

Evolution4e_20.21.ai Date 12-16-2016

(A) Passive (B) Active (driven)

0

Decreases Increases

Character state

Decreases Increases

Character state

Time Time

0

FIGURE 20.21 Computer simulations of the diversifica-

tion of a clade. The original character state is marked 0.

(A) A passive trend. A character shift in either direction is

equally likely, but the character state cannot go beyond

the boundary at left. The mean increases, but many

lineages retain the original character value. (B) A driven

trend. The entire distribution of character states is shifted

by a bias in the direction of change. (After [73].)

20_EVOL4E_CH20.indd 536 3/22/17 1:44 PM