Speciation in rock-dwelling land snails

Geographic speciation is generally accepted as the most common form of diversification in animals, and it is especially easy to invoke in systems where organisms have low dispersal capability together with fragmented range due to patchy distribution of their habitat. Isolation is often assumed to be the sole explanation for the origin of diversity in these insular settings. As a result, the potential influence of selection and the possible contribution of non-allopatric modes of speciation have often been ignored, and several aspects of the evolutionary process remain unrecognized in such systems. In the focus of our study is the extensive radiation of the Balkan-endemic clausiliid (door- snail) genus Montenegrina. As a rock-dwelling gastropod genus, Montenegrina lives on isolated and scattered rocky outcrops in topographically complex terrain and comprises large numbers of narrow- range, morphologically-distinguishable taxa, providing an excellent system for studying stages and modes of speciation. We plan to analyse a comprehensive sample (each known taxon and populations from almost every known occurrence). As prerequisites for delivering further objectives, we will reconstruct the phylogeny of Montenegrina and carry out a long-needed taxonomic revision using several molecular genetic markers and morphology-based methods. Based on this taxonomic scaffold, we will reconstruct the biogeographic history of the genus and study the relationships between phylogenetic, morphological and ecological divergences. Primary objectives are to shed light on the role of selection and to assess the relative weight of non-allopatric modes of speciation in this rock- dwelling gastropod radiation. The novelty and intellectual merit of the project arises from the integrative approach: It is the first attempt to combine methods of traditional taxonomy, morphometrics, molecular phylogenetics, population genetics phylogeography and niche modeling in a comprehensively sampled, diverse land snail group. Aside from answering fundamental questions about speciation, this project is expected to bring methodological advances in two fields: First, it will lead to new developments in the use of niche models for studying the role of ecology during speciation. And second, Montenegrina will be the first clausiliid taxon for which microsatellite primers will be established, and these newly developed primers enable the application of microsatellite analysis in population genetic studies of other taxa in this species-rich family.

From the point of view of biodiversity research, rock-dwelling gastropods are very remarkable because of their high proportion of narrow-range species and subspecies. They contribute significantly to the total species diversity in some limestone regions of the world. Many species have extremely narrow ranges, some are known from single locations only. Still, we know little about processes and forces that led to this amazing diversity in rock-dwelling snails. In such insular settings, isolation and genetic drift (without niche differentiation) had been proposed as main drivers of diversity. In the present project we tested this "non- adaptive radiation" hypothesis in the morphologically hyper-diverse door snail genus Montenegrina, which is distributed in the western part of the Balkan Peninsula. As the majority of its previously described species and subspecies were described exclusively by certain shell traits, some might question whether such a species richness reflects reality and is not solely a man-created artefact. Therefore, the establishment of a sound phylogenetic framework and a taxonomic assessment were prerequisites. We analysed mitochondrial and nuclear DNA sequences to establish a phylogenetic tree. Comparing the phylogeny with current taxonomy we found only about half of the described species and subspecies as monophyletic indicating that for such a morphologically variable genus shell traits alone can provide only a rough taxonomic framework. To obtain a broader set of traits, the inner genital morphology was comprehensively studied, which was done the first time for a door snail genus. Genital morphology was constant within populations, but there was an astonishing variability between species and subspecies. Moreover, there were remarkable differences between some populations of the same shell morphotypes. As a strikingly different genital setup might easily be a reproductive barrier, we interpreted different genital setups as taxon distinguishing features. In an "integrative taxonomy" approach combining all available data we re-evaluated the taxonomy of Montenegrina. It turned out that this genus is indeed a hyper-diverse group, even more diverse than believed at the start of the project: in the currently revised taxonomy, it comprises 70 species (114 subspecies) of which about three out of four are known only from one or two locations. This implies that current knowledge on species richness of other rock-dwelling gastropod taxa is likely an underestimation too and that integrative taxonomy might substantially re-arrange their systems as well. Finally, based on a newly developed method, the range constrained co- occurrence simulation (RaCoCOS), we demonstrated that early stages of speciation are driven primarily by non-adaptive processes and not by the adaptation to different environments. Our study of population size and distribution revealed extreme site fidelity: Individuals remained mostly in a very small area throughout their lives, suggesting a severely limited gene flow. Together with the findings on internal genital morphology, this suggests that sexual selection might enhance the effects of genetic drift and founder effects and that speciation of these small-scale populations may occur in very short evolutionary periods.