Biology Reference
In-Depth Information
FIGURE 12.4
Norms of reaction for shape of three genotypes reared in four
environments. Genotype A varies little over the four environments. Genotype B
varies over the four environments, occupying a larger region of the phenotype
space. Genotype C responds to one environmental factor but not to the others.
An exemplary recent analysis illustrates the methods used to study plasticity, including
the structure of the comparisons that can be made and the methods for testing hypotheses.
In this experiment, Hollander and colleagues examined two ecomorphs of one species
of snail,
Littorina saxatilis
and another species,
L
.
littorea
(
Hollander et al., 2006
).
L
.
saxatilis
is polymorphic in shell traits; its morphs differ in shell, thickness, and aperture area.
An ecologically important distinction between the two species is that
L
.
saxatilis
releases miniature snails on the shore and therefore the juveniles experience the same
habitat as the (sedentary) adults, whereas
L
.
littorea
releases drifting larvae which therefore
experience a different environment from the adults. Because the environment of
L
.
saxatilis
does not vary across life-history stages, they are predicted to be less plastic than
L
.
littorea
even though the environment of the adults of both species is highly heterogeneous; adults
of both species are found in wave-exposed rocky shores that are free of crabs (a predator)
as well as in sheltered areas that are rich in the predator. The drifting larvae of
L
.
littorea
encounter the rocky-shore habitats largely at random and their environments need not
resemble that of their parents, whereas the larvae of
L
.
saxatilis
remain in their parents'
environment. The prediction was that
L
.
littorea
would be more plastic.
To test this hypothesis, the juveniles of both morphs of
L
.
saxatilis
and juveniles of
L
.
littorea
, were collected from various locations on the west coast of Sweden. The juveniles
were pooled so that the samples encompassed the variation among local sites. The juve-
niles were exposed to simulated wave action, or a predator cue (water-borne effluent from
a predator or water-borne effluent from crushed conspecifics) or to no treatment. Each
group contained at least 50 individuals and each was replicated in three aquaria. Analysis
of the variation among ecomorphs and species showed that the two ecomorphs are distinct
as are the two species, thus the null model for the statistical analyses included these
expected differences. Analyses were done to determine whether ecomorphs of
L
.
saxatilis
and
L
.
littorea
exhibit plasticity, whether all three groups exhibit equal levels of plasticity
within and across treatments, and to identify the largest contrasts between treatments for
each group. Multivariate analysis of variance (MANOVA) showed the treatments had a
statistically significant impact on mean shape of each group.
To quantify the overall magnitude of plasticity, Hollander and colleagues measured dis-
parity within each group and compared it across groups. Although the measure of dispar-
ity was introduced in Chapter 10, we repeat it here. Disparity is the variance across
