GRASS SNAKE POPULATION DIFFERENTIATION OVER DIFFERENT GEOGRAPHIC SCALES BARBARA MEISTER 1 ,SYLVAIN URSENBACHER, AND BRUNO BAUR Section of Conservation Biology, Department of Environmental Sciences, University of Basel, St. Johanns-Vorstadt 10, 4056 Basel, Switzerland ABSTRACT: The loss and fragmentation of pristine habitat restrict specialized species to remnants of original habitat patches in a less suitable landscape. This may lead to a genetic differentiation of the subpopulations and to a decline in biodiversity. We used seven microsatellite markers to examine the genetic population structure of the Grass Snake, Natrix natrix, sampled in remnants of pristine habitat in a former wetland in the Swiss lowlands and in a rural valley in the Alps. On a regional level, the population structures of N. natrix in these two areas were compared with that of Grass Snakes living in an intensively used agricultural area. The three study areas were 30–100 km apart, but were interconnected by the river Aare. At the local scale, no genetic differentiation was found in either of the N. natrix populations inhabiting the rural alpine valley or the intensively used agricultural area. However, two subpopulations in the former wetland area were genetically differentiated with a low but significant measure of genetic differentiation between subpopulations, F ST . This slight genetic differentiation can be explained by isolation by distance. At the regional scale, we found significant genetic differentiation between N. natrix populations inhabiting areas separated by 30–100 km. The genetic structure was highly related to isolation by distance with 85% of the among-populations genetic variance explained by the geographical distance between subpopulations. Euclidean distance explained genetic differentiation of Grass Snake populations better than the distance following watercourses. Our findings indicate regular gene flow between N. natrix subpopulations and show that this species also moves across intensively used terrestrial habitat. The genetic structure of Grass Snakes is mainly affected by geographic distance, while human activity and habitat alteration do not seem to reduce the snakes’ movements. Our results suggest that conservation actions in landscapes altered by humans should focus on the maintenance of a habitat mosaic with anuran breeding ponds and adequate oviposition sites. Key words: Grass Snake; Isolation by distance; Microsatellite DNA; Natrix natrix; Population structure HUMAN activities result in the loss and fragmentation of pristine habitat (Foley et al., 2005). Habitat fragmentation reduces the suitable area for organisms, leads to the isolation and size reduction of remnant populations, and increases the risk of local extinction (Saccheri et al., 1998). There is growing evidence that relictual populations are subject to significant genetic and demo- graphic changes, which may directly (e.g., via inbreeding depression or allele erosion) or indirectly (e.g., via reduced evolutionary potential) affect individual fitness and popu- lation viability (Young and Clarke, 2000; Rusterholz and Baur, 2010). Maintaining gene flow is therefore crucial for the long-term viability of populations. In animals, dispersing individuals may introduce novel alleles into a population. The dispersal rate depends on the population size, extent of resource competi- tion, habitat quality, and size and isolation of suitable habitat patches, as well as the species’ behavior (Bennett, 2003). A central topic of landscape genetics is examination of how different landscape fea- tures influence gene flow (Manel et al., 2003). Empirical evidence from a variety of animal taxa indicates that human infrastructure, including highways, roads, and railway lines, reduces landscape permeability (Trombulak and Frissell, 2000; Clark et al., 2010; Holder- egger and Di Giulio, 2010; but see Brown et al., 2006). However, even in the absence of barriers, cumulative effects of different land- scape features with low permeability may lead to genetic differentiation of populations. Landscape features such as ridges, rivers, and open shrub habitat influence dispersal in amphibians (Funk et al., 2005; Spear et al., 2005; Giordano et al., 2007). Dispersal in reptiles is affected by habitat discontinuities (Stow et al., 2001). As a consequence, the genetic population structure of reptile species is influenced by the size and configuration of suitable habitat patches and the permeability 1 CORRESPONDENCE: e-mail, barbara.meister@unibas.ch Herpetologica, 68(1), 2012, 134–145 E 2012 by The Herpetologists’ League, Inc. 134