Evolutionary consequences of hybridization in the Lycaeides species complex




Gompert, Zachariah

Journal Title

Journal ISSN

Volume Title



Hybridization-here defined as mating between differentiated populations-has several evolutionary consequences (Arnold 1997, Mallet 2005). If hybrids have reduced fitness compared to the parental populations, then selection may lead to the evolution of increased assortative mating, and thus, decrease the rate of hybridization (Mallet 2005). This process is referred to as reinforcement (Mayr 1942). Alternatively, if the populations are not sufficiently differentiated, hybridization may lead to their fusion (Arnold 1997). In both of these examples hybridization can be viewed as a transient phenomenon. This is not always the case, as hybrid zones may be long lasting. The maintenance of a hybrid zone is likely when 1) the hybrid zone occurs along an environmental cline, or 2) when the hybrid zone is maintained by a selection-dispersal balance (in this case the hybrid zone is referred to as a tension zone) (Arnold 1997). Hybrid zones may allow alleles to pass from one of the parental populations into the other via backcrosses between the hybrids and parentals. Finally, the hybrids may come to occupy a novel niche and become self-sustaining and reproductively isolated from their parental populations (e.g. Reiseberg 2003). Here I examine the effects of hybridization within the North American Lycaeides idas-L. melissa species complex (Lepidoptera: Lycaenidae). A previous phylogeographic study based on variation in the control region of the mitochondrial genome demonstrated I that populations of Lycaeides were relegated to at least three refugia during Pleistocene glacial maxima (Nice et al. 2005). Secondary contact has occurred among these refugial populations following post Pleistocene range expansion (Nice et al. 2005). Based on discordance between the mitochondrial and morphological (male genitalic measurements and wing patterns) characters, Nice et al. (2005) concluded that introgressive hybridization had occurred at two of these contact zones, one near the Great Lakes and one along the Sierra Nevada. My first study investigated the contact zone near the Great Lakes, which involved the North American endangered Kamer Blue Butterfly (Lycaeides melissa samuelis) and the closely related L. m. melissa. These butterflies can be distinguished based on differences in life history and morphology. Western populations of L. m. samuelis share mitochondrial haplotypes with L. m. melissa populations, while eastern populations of L. m. samuelis have divergent haplotypes (Nice et al. 2005). I tested two hypotheses concerning the presence of L. m. melissa mitochondrial haplotypes in western L. m. samuelis populations: (i) mitochondrial introgression has occurred from L. m. melissa populations into western L. m. samuelis populations, or (ii) western populations of the nominal L. m. samuelis are more closely related to L. m. melissa than to eastern L. m. samuelis populations yet are phenotypically similar to the latter. A Bayesian algorithm was used to cluster 190 L. melissa individuals based on 143 AFLP loci. This method clearly differentiated L. m. samuelis and L. m. melissa. Thus, genomic divergence was greater between western L. m. samuelis</i> populations and L. m. melissa populations than it was between western and eastern populations of L. m. samuelis. These findings support the hypothesis that the presence of L. m. melissa mitochondrial haplotypes in western L. m. samuelis</i> populations is the result of mitochondrial introgression. These data provide valuable information for conservation and management plans for the endangered L. m. samuelis, and illustrate the risks of using data from a single locus for diagnosing significant units of biodiversity for conservation. The second study I conducted involved the contact zone between L. idas and L. melissa along the Sierra Nevada. Here I investigated the possibility that alpine adapted Lycaeides populations in the Sierra Nevada arose via hybrid speciation, following hybridization between L. idas and L. melissa. Theory predicts that homoploid hybrid speciation is facilitated by adaptation to a novel or extreme habitat; heretofore this has not been documented in animals (Buerkle et al. 2000). Using molecular data and data from ecological experiments, I demonstrated that the alpine-adapted butterflies in the genus Lycaeides are the product of hybrid speciation. I showed that the alpine populations possess a mosaic genome derived from both L. melissa and L. idas. These alpine populations are differentiated from, and have a more recent origin than, their putative parental species. Adaptive traits allow persistence in the environmentally extreme alpine habitat and reproductively isolate these populations from their parental species, as predicted by theory. These studies demonstrate that hybridization has several important evolutionary consequences, some of which may be far from transient. Indeed hybridization may represent a mechanism driving biological diversification. The first chapter presented has been published in Molecular Ecology, and thus, is formatted for that journal. The second chapter has been accepted at Science. It is formatted for that journal with some minor modifications.



Lycaeides, Evolution, Hybridization, Ecology


Gompert, Z. (2006). <i>Evolutionary consequences of hybridization in the Lycaeides species complex</i> (Unpublished thesis). Texas State University-San Marcos, San Marcos, Texas.


Rights Holder

Rights License

Rights URI