population increases has no effect on optimal age at maturity. In contrast, when
maximizing localr, rapid population growth in that patch selects for earlier
maturity as fitness is increased by producing offspring rapidly (as we argued in
the discussion of Eq. (3.1)).
Conditions favouring the evolution of adaptive plasticity in life histories
include environmental change, which is predictable and which is encountered
sufficiently frequently during evolution to have a strong selective effect.
However, it should change sufficiently infrequently, or at least over a long
enough period during ontogeny, to allow fitness benefits to accrue from life-
history shifts that may take time to implement and/or be costly to reverse
(Atkinson & Thorndyke, 2001 ; Doughty & Reznick, 2004 ).
Testing the adaptive significance of body-size plasticity is a major chal-
lenge, but approaches have been identified to ensure that tests are rigorous
(Gotthard & Nylin, 1995 ; Atkinson & Thorndyke, 2001 ; Doughty & Reznick,
2004 ). For instance, even though the plastic shifts in cladoceran adult size in
response to predator kairomones appear to be adaptive, it is still necessary to
check that the life-history responses were not caused by some other effect of the
kairomones (e.g. toxic or nutrient) rather than their information content about
predation risk.
Interpreting phenotypic variation
Variation in a trait along an environmental gradient will result from the com-
bined effects of genetic variation in the trait between individuals, phenotypic
plasticity, and the differences in plasticity between genotypes (genotype
environment interaction). A reaction norm can be in the same direction as the
selective effects of the environment on the trait: this is co-gradient variation. An
example is how fish predation on large cladocerans can select among genotypes,
favouring a reduced mean size at maturity, and can select for reaction norms
that reduce age and size at maturity within clones in response to fish kairo-
mones (Fig.3.1). But counter-gradient variation can sometimes occur, where
genetically based mechanisms evolve to resist environmental perturbation of a
trait resulting in phenotypic similarity along an environmental gradient
(Conover & Schultz, 1995 ).
An example is provided by pumpkinseed sunfish,Lepomis gibbosus, which
when small competes with bluegill sunfish,L. macrochirus(Arendt & Wilson,
1999 ). Raising pumpkinseed with bluegill, irrespective of whether the pump-
kinseed were derived from lakes with or without bluegill, slowed pumpkinseed
growth. However, pumpkinseed derived from lakes with bluegill grew faster
than those from lakes without bluegill under all experimental treatments,
following the predicted evolutionary acceleration of growth to allow passage
through the competing size classes as quickly as possible (Arendt & Wilson,
1999 ; Fig.3.2).
LIFE HISTORIES AND BODY SIZE 39