Understanding the nature of adaptive evolution has been the recent focus of research detailing the genetic basis of adaptation and theoretical work describing the mechanics of adaptive evolution. Nevertheless, key questions regarding the process of adaptive evolution remain. Ultimately, a detailed description of the ecological context, evolutionary history, and genetic basis of adaptations is required to advance our understanding of adaptive evolution. To address some of the contemporary issues surrounding adaptive evolution, I examine phenotypic and genotypic changes in a snake feeding adaptation. Adaptations can arise through fixation of novel mutations or recruitment of existing variation. Some populations of the garter snakes Thamnophis sirtalis, T. couchii, and T. atratus possess elevated resistance to tetrodotoxin (TTX), the lethal toxin of their newt prey. I show that TTX resistance has evolved independently through amino acid changes at critical sites in a voltage-gated sodium channel protein (Nav1.4) targeted by TTX. Thus, adaptive evolution has occurred multiple times in garter snakes via de novo acquisition of beneficial mutations. Detailing the genetic basis of adaptive variation in natural populations is the first step towards understanding the tempo and mode of adaptive evolution. I evaluate the contribution of Nav1.4 alleles to TTX resistance in two garter snake species from central coastal California. Allelic variation in Nav1.4 explains 29% and 98% of the variation in TTX resistance in T. atratus and T. sirtalis, respectively, demonstrating that Nav1.4 is a major effect locus. The simple genetic architecture of TTX resistance in garter snakes may significantly impact the dynamics of trait change and coevolution. Patterns of convergent evolution are cited as some of the most compelling examples of the strength of natural selection in shaping organismal diversity. Yet repeated patterns may tell us as much about the constraints that restrict evolution as about the importance of natural selection. I present data on convergent molecular adaptations in parallel arms races between diverse snakes and amphibians from across the globe. Six snake species that prey on TTX bearing amphibians have independently acquired amino acid changes in Nav1.4. The derived mutations are clustered in two portions of the gene, often involving the same sites and substitutions. While a number of amino acid changes can make Nav1.4 insensitive to TTX, most of these negatively impact or abolish the ion-conducting function of the protein. Thus, intramolecular pleiotropy likely prevents most replacements from becoming fixed and imposes limits on protein evolution.