Data from: Toxicity and population structure of the Rough-Skinned Newt (Taricha granulosa) outside the range of an arms race with resistant predators
Hague, Michael T. J. et al. (2017), Data from: Toxicity and population structure of the Rough-Skinned Newt (Taricha granulosa) outside the range of an arms race with resistant predators, Dryad, Dataset, https://doi.org/10.5061/dryad.3c212
Species interactions, and their fitness consequences, vary across the geographic range of a coevolutionary relationship. This spatial heterogeneity in reciprocal selection is predicted to generate a geographic mosaic of local adaptation, wherein coevolutionary traits are phenotypically variable from one location to the next. Under this framework, allopatric populations should lack variation in coevolutionary traits due to the absence of reciprocal selection. We examine phenotypic variation in tetrodotoxin (TTX) toxicity of the Rough-Skinned Newt (Taricha granulosa) in regions of allopatry with its TTX-resistant predator, the Common Garter Snake (Thamnophis sirtalis). In sympatry, geographic patterns of phenotypic exaggeration in toxicity and toxin-resistance are closely correlated in prey and predator, implying that reciprocal selection drives phenotypic variation in coevolutionary traits. Therefore, in allopatry with TTX-resistant predators, we expect to find uniformly low levels of newt toxicity. We characterized TTX toxicity in northwestern North America, including the Alaskan panhandle where Ta. granulosa occur in allopatry with Th. sirtalis. First, we used microsatellite markers to estimate population genetic structure and determine if any phenotypic variation in toxicity might be explained by historical divergence. We found northern populations of Ta. granulosa generally lacked population structure in a pattern consistent with northern range expansion after the Pleistocene. Next, we chose a cluster of sites in Alaska, which uniformly lacked genetic divergence, to test for phenotypic divergence in toxicity. As predicted, overall levels of newt toxicity were low; however, we also detected unexpected among- and within-population variation in toxicity. Most notably, a small number of individuals contained large doses of TTX that rival means of toxic populations in sympatry with Th. sirtalis. Phenotypic variation in toxicity, despite limited neutral genetic divergence, suggests that factors other than reciprocal selection with Th. sirtalis likely contribute to geographic patterns of toxicity in Ta. granulosa.