Plethodontid salamanders do not conform to "general rules" for ectotherm life histories: insights from allocation models about why simple models do not make accurate predictions

Recent theoretical models hold that temperature imposes unalterable physiological effects on ectotherm growth and development such that their life histories are dictated by local biophysical environments. Models relying on this premise have been offered to explain many life history phenotypes including threshold traits such as age/size at metamorphosis/maturity. Because threshold traits arc thought to influence adult fitness components by affecting performance of an individual in its new habitat, they are evolutionarily important components of complex life cycles. Consequently, the ecological and genetical basis of variation in such traits has been the focus of a large research program by evolutionary biologists, and amphibians have been model systems for these studies in the last three decades. Smith-Gill and Berven proposed a physiological model to explain commonly observed clinal patterns of variation in metamorphosis, and it appears to account successfully for patterns observed in a few species of pond-breeding frogs and salamanders (Rana and Ambystoma). However, six species of stream-breeding salamanders (family Plethodontidae: Desmognathus, Pseudotriton, Eurycea) contradict both the phenotypic patterns found in nature for pond-breeding species as well as the predictions of this model. Four of the six species of plethodontids have significantly larger metamorphs at lower, warmer elevations (or more southerly latitudes) rather than at higher, cooler sites; two species show no clinal pattern. In light of these results, we critically examine the assumptions of and support for SGB. We propose alternative hypotheses to explain patterns of variation in metamorphic traits along thermal gradients, focusing on a) differences between pond-breeding Amphibia and plethodontids in basic biology and larval habitats, b) gradients in other biotic and abiotic factors, and c) other effects of temperature on organismal function. Finally, we discuss our results in the context of current models of how ectotherm life histories are affected by temperature.