Phylogeographical studies are typically based on haplotype data, occasionally on nuclear markers such as microsatellites, but rarely combine both. This is unfortunate because the use of markers with contrasting modes of inheritance and rates of evolution might provide a more accurate and comprehensive understanding of a species' history. Here we present a detailed study of the phylogeography of the greater horseshoe bat, Rhinolophus ferrumequinum, using 1098 bp of the mitochondrial ND2 gene from 45 localities from across its Palaearctic range to infer population history. In addition, we re-analysed a large microsatellite data set available for this species and compared the results of both markers to infer population relationships and the historical processes influencing them. We show that mtDNA, the most popular marker in phylogeography studies, yielded a misleading result, and would have led us to conclude erroneously that a single expansion had taken place in Europe. Only by combining the mitochondrial and microsatellite data sets are we able to reconstruct the species' history and show two colonization events in Europe, one before the Last Glacial Maximum (LGM) and one after it. Combining markers also revealed the importance of Asia Minor as an ancient refugium for this species and a source population for the expansion of the greater horseshoe bat into Europe before the LGM.

The distribution of genetic variability across a species' range can provide valuable insights into colonization history. To assess the relative importance of European and Asian refugia in shaping current levels of genetic variation in the greater horseshoe bats, we applied a microsatellite-based approach to data collected from 56 localities ranging from the UK to Japan. A decline in allelic richness from west Asia to the UK and analyses of F(ST) both imply a northwestward colonization across Europe. However, sharp discontinuities in gene frequencies within Europe and between the Balkans and west Asia (Syria/Russia) are consistent with suture zones following expansion from multiple refugia, and a lack of recent gene flow from Asia Minor. Together, these results suggest European populations originated from west Asia in the ancient past, and experienced a more recent range expansion since the Last Glacial Maximum. Current populations in central Europe appear to originate from the Balkans and those from west Europe from either Iberia and/or Italy. Comparisons of R(ST )and F(ST) suggest that stepwise mutation has contributed to differentiation between island and continental populations (France/UK and China/Japan) and also among distant samples. However, pairwise R(ST) values between distant populations appear to be unreliable, probably due to size homoplasy. Our findings also highlight two priorities for conservation. First, stronger genetic subdivision within the UK than across 4000 km of continental Eurasia is most likely the result of population fragmentation and highlights the need to maintain gene flow in this species. Second, deep splits within China and between Europe and China are indicative of cryptic taxonomic divisions which need further investigation.

No abstract

No abstract