Batesian mimicry is characterized by phenotypic convergence between an unpalatable model and a palatable mimic. However, because convergent evolution may arise via alternative evolutionary mechanisms, putative examples of Batesian mimicry must be rigorously tested. Here, we used artificial butterfly facsimiles (N = 4000) to test the prediction that (1) palatable Limenitis lorquini butterflies should experience reduced predation when in sympatry with their putative model, Adelpha californica, (2) protection from predation on L. lorquini should erode outside of the geographical range of the model, and (3) mimetic color pattern traits are more variable in allopatry, consistent with relaxed selection for mimicry. We find support for these predictions, implying that this convergence is the result of selection for Batesian mimicry. Additionally, we conducted mark-recapture studies to examine the effect of mimicry and found that mimics survive significantly longer at sites where the model is abundant. Finally, in contrast to theoretical predictions, we found evidence that the Batesian model (A. californica) is protected from predation outside of its geographic range. We discuss these results considering the ongoing hybridization between L. lorquini and its sister species, L. weidemeyerii, and growing evidence that selection for mimicry predictably leads to a reduction in gene flow between nascent species.

Understanding the origin and maintenance of adaptive phenotypic novelty is a central goal of evolutionary biology. However, both hybridization and incomplete lineage sorting can lead to genealogical discordance between the regions of the genome underlying adaptive traits and the remainder of the genome, decoupling inferences about character evolution from population history. Here, to disentangle these effects, we investigated the evolutionary origins and maintenance of Batesian mimicry between North American admiral butterflies (Limenitis arthemis) and their chemically defended model (Battus philenor) using a combination of de novo genome sequencing, whole-genome resequencing, and statistical introgression mapping. Our results suggest that balancing selection, arising from geographic variation in the presence or absence of the unpalatable model, has maintained two deeply divergent color patterning haplotypes that have been repeatedly sieved among distinct mimetic and nonmimetic lineages of Limenitis via introgressive hybridization.