Evolution, 4th Edition

(Amelia) #1
532 CHAPTER 20

same temperature regime throughout the last 3 My, despite the profound climate
changes during the Pleistocene [95]. Many groups of herbivorous insects have phy-
logenetically conservative diets; the larvae of all species of the butterfly tribe Heli-
coniini feed on plants in the passionflower family (Passifloraceae), and apparently
have done so since the tribe originated in the Oligocene. By occupying one niche
(e.g., host plant group, climate zone) rather than another, a species subjects itself to
some selection pressures and screens off others; it may even be said to “construct”
or determine its own niche, and therefore many aspects of its potential evolution-
ary future [59, 83]. Niches also may remain conservative because other species,
often by acting as competitors, may prevent a species from shifting or expanding
its niche.
More generally, if there is gene exchange among individuals that inhabit the
ancestral niche (e.g., microhabitat) and those that inhabit a novel niche, and if
there is a fitness trade-off between character states that improves fitness in the
two environments, then selection will generally favor the ancestral character
state (i.e., stabilizing selection will prevail) simply because most of the popula-
tion occupies the ancestral environment (see Chapter 11) [43]. As the degree
of adaptation to any one environment increases, adaptation to an alternative
environment may become steadily less likely. In some cases, a species may lose
the ability to vary in features that would be necessary for a substantial ecological
shift, even back to its ancestral niche. For example, in a bird-pollinated clade of
morning glories (Ipomoea) with red flowers, the genes required to synthesize the
blue and purple pigments that typify most other morning glories have become
nonfunctional pseudogenes [117]. In general, unused genes acquire disabling
mutations and become pseudogenes, as shown by the reduced number of func-
tional olfactory receptors in primates and the degeneration of many genes and
phenotypic functions in parasites.
Loss of functional genes is one of several reasons why there may exist genetic
constraints on the evolution of some characteristics or on adaptation to novel
environments (see Chapters 6, 8, and 15). Some features, such as tolerance to
higher temperatures, may have very limited genetic variation [42]. Because of
genetic correlations stemming from pleiotropy, genetic variation may exist for
certain combinations of characters, but not others, even if each character indi-
vidually is variable [7, 52]. If genes have consistent patterns of pleiotropy, such
character correlations may be very long lasting, and they could be an important
determinant, in the long run, of the pattern of genetic variation that is available
for evolutionary change [50]. Perhaps for this reason, researchers are increas-
ingly finding that the directions of evolutionary differences among species are
correlated with the “genetic lines of least resistance” that have been estimated
from genetic or phenotypic correlations in living populations (see Figures 6.21
and 15.20). For example, the evolution of shape characteristics in multiple lin-
eages of the early Paleogene ostracod crustacean Poseidonamicus proceeded
largely in the direction of those character combinations that showed the greatest
variation within living populations (FIGURE 20.15) [45]. Similarly, the pattern of
divergence in wing shape among various clades of Drosophila is broadly similar
to the genetic variance-covariance matrix in D. melanogaster, suggesting that the
pattern of genetic variation has been fairly consistent for more than 50 My [40].
Moreover, laboratory populations selected for a different wing shape rapidly
reverted to the normal shape when artificial selection was alleviated, because
of deleterious pleiotropic effects [8]. However, experiments have shown that in
some cases, pleiotropic correlations can be reduced by selection at other loci that
compensate for pleiotropic effects (see Chapter 15) [87]. Just how long genetic
constraints may persist is not yet known.

20_EVOL4E_CH20.indd 532 3/22/17 1:44 PM

Free download pdf