What is the historical demography of the mimic convergence between H. erato and H. melpomene?

Heliconius erato and H. melpomene represent two of the most divergent lineages within Heliconius, yet they have nearly identical wing pattern phenotypes and display parallel geographic patterns of divergence in their aposematic wing color patterns. The striking similarities in their wing color patterns and overlapping pattern diversification has lead some researchers to speculate that they have diverge in parrellel during Pliestocene glacial cycle.

We are using sequence variation in rapidly evolving introns of single-copy nuclear genes to complement previous mtDNA work and explore the history of racial diversification in H. erato and H. melpomene. This work (PNAS, in prep.) demonstrates that, despite the concordant geographic patterns of diversification, the co-mimics have had completely different demographic histories over the recent past. Heliconius erato is the older species, probably forming in the mid-Pliocene, and shows a demographic history marked by high gene flow and large effective population sizes throughout much of the Pleistocene. In contrast, H. melpomene evolved much more recently and displays a demographic history marked by substantial population genetic bottlenecks and rapid coalescence of nuclear (and mtDNA) variation into regional clades. These genetic data i) argue strongly against the hypothesis that the parallel patterns of racial evolution in the two species was the result of strict co-evolutionary change, ii) further erode a simple Pleistocene vicariant hypothesis as an omnipotent explanation for the biological diversity of the Amazon region, and iii) highlight the remarkable evolutionary plasticity of butterfly wing patterns. Current research priorities are centered refining our understanding aboutthe population history of these species. We are presently developing markers more tightly linked to the major color pattern switch genes that underlie adaptive change in wing patterns in the two species.

Research also focuses on the behavioral, genetic, and historical processes that initiate speciation in coral reef fishes. There has been relatively little critical work on speciation in the marine realm, where taxa often possess vast ranges, huge population sizes and extended dispersal capabilities due to long-lived planktonic larval stages. Most evolutionary models predict that organisms with these types of population dynamics should be buffered from speciation; yet recent paleontological and genetic research has exposed the speed at which evolutionary change can occur in these groups.

We are actively involved in collaborative research with UNC-W (Dr. Mike McCarthney) and STRI (Dr. Biff Bermingham) to study speciation and species boundaries in Caribbean hamlets. Hamlets offer unparalleled research possibilities for blending high-resolution microsatellite data with field observations of mating behavior to study the incipient stages of speciation in coral reef fishes. There are currently 12 recognized species which are morphologically similar, but show striking color pattern differences. Mating is strongly assortative between similarly colored fishes, but hybridization occurs at low frequency. Nonetheless, the color pattern distinctions that define species are maintained across the Caribbean, and in the face of syntopic distribution patterns that can yield as many as 6 species in a single area of reef. The sympatric coexistence of the closely related hamlets brings to mind Darwin¹s finches and haplochromine cichlids.

MtDNA and hypervariable microsatellite loci data reveal an interesting genetic mosaic that highlights a complex interaction between population history, mating behavior, and geography in the hamlet species flock (Molecular Ecology, in press). Speciation has clearly occurred very recently, within the last 500,000 years, and there are only very slight genetic differences within and among incipient forms. Nevertheless, there is evidence for genetic differentiation across population and color pattern boundaries. Populations of the same species collected in Puerto Rico and Panama showed slight, yet significant, genetic differences indicating some regional within-species population structure. In addition, we observed genetic differentiation among different species collected from the same reef; however, this result varied by location. In Puerto Rico, there was clear evidence for assortative mating by color pattern. In our Panama collection, however, the same complement of species was genetically indistinguishable suggesting that mating behavior may vary geographically. We are currently expanding this study to better understand contemporary and historical forces shaping genetic variation in this group. Current research priorities include i) developing likelihood models to estimate interbreeding probabilities among sympatric color pattern species and to determine if these probabilities vary geographically, ii) increasing the scope of our genetic work to include more individuals and more high resolution microsatellite loci, and iii) modeling the buildup of genetic differences under different evolutionary scenarios.