Incomplete reproductive isolation promotes gene flow between diverging taxa. However, any gene encoding for traits involved in the reproductive barriers will be less prone to introgression than neutral markers. Comparing introgression rates among loci is thus informative of the number and functions of loci involved in the reproductive barriers. This study aimed at identifying possible mechanisms of restriction to gene flow across a zone of recent secondary contact between Larus argentatus and Larus cachinnans by comparing introgression patterns for nine microsatellite loci, a fragment of mitochondrial DNA and a set of phenotypic traits. The low linkage disequilibrium between neutral nuclear markers indicated introgression without any barrier to gene flow. However, asymmetric introgression of mitochondrial DNA suggested that interspecific crosses may be more successful in one direction. The introgression rate for phenotypic traits was variable and low compared to neutral molecular markers. This was particularly evident in colouration of bare parts: individuals with intermediate colouration were scarcer in sympatry than expected if the genomes recombined freely. We hypothesized that one of these variables, the orbital ring colour, may play a role in mate choice, acting as an incomplete premating barrier through assortative mating. This study emphasizes that multilocus approaches are useful to discriminate among possible mechanisms responsible for the maintenance of hybrid zones.

No abstract

Current evolutionary forces and historical processes interact to shape the distribution of neutral genetic variability within and among populations. Focusing on the genetics of recently introduced organisms offers a good opportunity to understand the relative importance of these factors. This study concerns variation at 8 polymorphic microsatellite loci in 30 populations of Biomphalaria pfeifferi. The sampling area spans most of the species' range in Madagascar where it was probably introduced recently. Extremely low variation was found within all populations studied, which may partly result from high selfing rates. However, this cannot account for the variance of variation across populations, which is better explained by habitat openness (that reflects environmental stochasticity), the prevalence of the parasitic trematode Schistosoma mansoni and historical demography (colonization and subsequent bottlenecks). Large global differentiation was also observed, suggesting that current gene flow among populations is limited to small distances, within watersheds and to few individuals. Our data set also allowed us to test several hypotheses regarding colonization, based on bottleneck and admixture tests. The observed pattern requires at least two independent introductions from slightly differentiated genetic sources in the western part of Madagascar. Another introduction, from a very different genetic origin, should also be postulated to explain the genetic composition of eastern populations. That this introduction occurred recently suggests that the colonization of Madagascar by B. pfeifferi is an ongoing process.