Prey availability and phenotypic differences between local terrestrial salamander populations

Population comparisons are a common means of studying the processes that drive natural phenotypic variation. Accurately interpreting patterns of trait variation requires a clear understanding of the “context” in which a trait varies; that is the differences in the environments across which a trait varies, and correlated variation in other traits. Using a series of comparative studies between local populations of red-backed salamanders, Plethodon cinereus, I correlated life-history, behavioral, and morphological differences between local populations with differences in diet. Salamander diets from one site consisted of larger prey, particularly earthworms, than that found in salamander diets from another local site. The frequency of earthworms in salamander diets declined between rains, and salamander diets at the two sites were similar between rains. Periodic access to larger prey resulted in a greater amount of prey for salamanders at that site. Salamander tail condition and female fecundity were greater among salamanders from the “high-food” site than the “low-food” site. Both the numbers of eggs and mean egg size were greater at the high-food site, which is consistent with a direct effect of diet on fecundity differences between sites. Salamanders at high-food sites showed reduced or no territoriality while salamander behavior at the low-food site was consistent with food-based territoriality. Though low-food site salamanders took smaller prey, they had disproportionately larger heads than conspecifics from the high-food/large-prey site. This pattern contradicts other studies of holophagic predators that show positive relationships between prey size and geographic variation in head size. This suggests that factors besides prey size could drive geographic variation in head morphology. The difference in head morphology between populations may be related to the difference in the social environments, with disproportionately larger heads occurring in the territorial population. This hypothesis is consistent with other studies of geographic variation in lizards, which show correlated increases in size and territoriality in low-food environments. Establishing a robust hypothesis to explain the ‘apparently’ contradictory morphological differences between these populations was only possible with a clear understanding of differences in the environments of those populations and correlated variation in other traits including social behavior.