The relationship between diet, toxicity, and color
Aposematic polytypism, having multiple aposematic (warning coloration) signals among populations, is a perplexing phenomenon because in order to fascilitate predator learning, selection should favor little variation in signal. Poison Frogs (Dendrobatidae), however, show polytypism in many species, so the question remains as to what evolutionary forces would promote such variation.
Recent research suggests that colors are honest indicators of toxicity for birds (the hypothesized predators driving this evolution), which means that the brighter the color, the more toxic the frog. Poison Frogs derive their toxicity from dietary items, which would suggest that different colored frogs are selecting different prey items. To examine this, I have collected stomachs of thirteen different populations of the highly polytypic Strawberry Poison Frog (Oophaga pumilio) and am in the process of analysize toxins (using both GC-MS and LC-MS processes) and identifying prey contents using next generation Illumina DNA sequencing technology.
Interestingly, a number of other Poison Frogs are sympatric with the polytypic Strawberry Poison Frog but are not polytypic among populations. If honest signaling exists in Poison Frogs, this would mean that these monotypic (here meaning a single morph among all populations) are likely eating the same type of invertebrate prey and/or have similar toxin profiles. As such I am using the same methods in analyzing the Strawberry Poison Frog diets and toxin profiles to analyze diet and toxin profiles in five sympatric species: Allobates talamancae, Andinobates claudiae, Silverstoneia flotator, Phyllobates lugubris, and Dendrobates auratus. These species are known to vary in toxicity among species (i.e., A. talamancae is considered to be nontoxic and P. lugubris is considered to be highly toxic).
This research is ongoing...