Relyea, R. A. 2002. Local population differences in phenotypic plasticity: Predator-induced changes in wood frog tadpoles. Ecological Monographs 72:77-93

Taxa that are divided into separate populations with low levels of interpopulation dispersal have the potential to evolve genetically based differences in their phenotypes and the plasticity of those phenotypes. These differences can be due to random processes, including genetic drift and founder effects, or they can be the result of different selection pressures among populations. I investigated population-level differences in predator- induced phenotypic plasticity in eight populations of larval wood frogs (Rana sylvatica) over a small geographic scale (interpopulation distances of 0.3–8 km). Using a common-garden experiment containing predator and no-predator environments, I found population differences in behavior, morphology, and life history. These responses exhibited a habitat-related pattern: the four populations from closed-canopy ponds did not differ from each other in any of their phenotypes whereas the four populations from opencanopy ponds did differ from each other in these traits. This phenotypic pattern matches the pattern of competitors and predators found in these two types of ponds. Based on two years of pond surveys, the four closed-canopy ponds contained very similar competitor and predator assemblages while the assemblages of the four open-canopy ponds were more diverse and highly variable among open-canopy ponds. When combined with past studies, which demonstrate that predators and competitors select for alternative behavioral and morphological traits, these patterns suggest that the population differences may have arisen via natural selection and not via random mutation or drift. In a second experiment, I cross-transplanted two of the populations into each other’s ponds to determine if the populations were locally adapted to the conditions of their native pond (using low and high competition crossed with the presence or absence of a lethal predator). The populations continued to exhibit phenotypic differences, and one of the two populations tested exhibited superior growth in its native pond. This suggests that some wood frog populations are adapted to the local conditions of their natal pond and that localized selection by predation and competition may be the underlying mechanism. Collectively, these experiments indicate that taxa that are divided into discrete populations and face different predator and competitor environments can evolve different phenotypically plastic responses.