Contrasting response to Pleistocene climate change by ground-living and arboreal Mandarina snails from the oceanic Hahajima archipelago.

While the genetic impact of Pleistocene climate change on temperate species has been well characterized, especially in Europe and North America, an effect on the diversification of species on oceanic islands has been less well studied. This is perhaps a surprising observation given the traditional and continuing contribution of island species (e.g. Darwin's finches, Partula snails, Lord Howe Island palms) to understand speciation. Here, we combine mitochondrial and microsatellite data from the ground-living and arboreal Mandarina snails of the oceanic, subtropical Hahajima archipelago (Ogasawara, colloquially 'Galápagos of the Orient') to enable a comparative approach to understand the impact of the Pleistocene glaciations on their phylogeography. Prior work suggested that several narrowly divergent, ground-living species pairs of Mandarina populations on the outlying islands, as well as the low-lying southern and central parts of Hahajima, probably underwent bottlenecks and subsequent expansions during the recent Pleistocene. Here, the most striking finding is that largely arboreal species have deeply divergent, geographically restricted mitochondrial lineages, in contrast to a census size that is at least an order of magnitude lower than ground-living snails. As populations of both types are highly polymorphic at microsatellite loci, the systematic difference at the mitochondrial locus probably indicates a contrasting effect of the Pleistocene climate cycles on the two groups. We speculate that this may have partly come about owing to a reduced efficacy of natural selection on the more greatly structured populations of arboreal snails. If so, then a prediction is that the genome of other snails, or other species with limited mobility, will show a similar response to the Pleistocene climate cycles.