A classic monoglacialistic view was that continental species arrived postglacially to Europe by a ‘Siberian invasion’. Two assumptions strongly shaped these ideas: (i) The climatic effects were generally milder in Siberia because of the huge unglaciated areas, and (ii) No arboreal refuges could exist north of the Mediterranean peninsulas. This picture was radically changed by findings of full-glacial forests in Europe north of the transversal mountain ranges, the phylogeography of cold-tolerant animal species and the biogeography and faunal history of European mountain butterflies (reviewed by Varga 2003; Schmitt 2007; Varga & Schmitt 2008).
Full-glacial forests in Eastern Europe were often questioned, but also repeatedly confirmed by several macrofossil and palynological studies from the Carpathians which demonstrated that the most cold-resistant trees survived in refugial pockets in the hilly areas of the Carpathians during the last full glaciation (Fig. 1). Similarly, forest refugia were also proved for southern Moravia, the Eastern Alps, the Dinaric Mountains and in the eastern Balkans (e.g. Willis & Andel 2004; Magri et al. 2006). It was also predicted that the Mediterranean ‘sanctuaria’ in general were not the core areas of postglacial expansion into deglaciated areas (Widmer & Lexer 2001). Thus, there was a need for new paradigms for the phylogeography of cold-resistant arboreal species, and their associated flora and fauna (Steward & Lister 2001).
In this issue, Schmitt & Haubrich (2008) reveal the major patterns of recent glacial and postglacial history for the European mountain coniferous forests by studying the large ringlet, Erebia euryale, a well-known butterfly species. Most other species occurring in these forests have a large Eurasian or Euro-Siberian range and are thus hardly suitable for this sort of study. However, the large ringlet is a European endemic butterfly, is fairly widely dispersed in central and southern European mountains and is closely connected with the subalpine coniferous belt. It does not occur in northern Europe, Scandinavia or Asia. Therefore, survival of this species in European coniferous forest refugia during the last glaciation is the most plausible working hypothesis. These species can thus be used to answer questions about the allocation and extension of these refugia. Which core areas preserved a significant amount of genetic variation, and which refugial populations were involved in the postglacial re-population of suitable habitats?
The authors have thoroughly reviewed and evaluated both the earlier and more recent references and suggested a plausible explanation for the glacial survival and postglacial re-population for an important component of the European mountain fauna. They used the established methods of allozyme electrophoresis, which reflect only a minor fraction of the actual genetic changes. However, these data clearly show patterns of genetic differentiation along a geographical gradient, allowing conclusions on the importance of the glacial coniferous mountain forest refugia in Europe. Neighbour-joining phenograms based on genetic distances distinguished four major genetic lineages: (i) the Pyrenees, (ii) the western Alps, (iii) the eastern Alps, and (iv) southeastern Europe. In particular, the three southeastern European samples were interesting, as they clustered very closely together and showed the highest diversity values (particularly in the Carpathians and Rila Mountains.). The genetic distances among the three western Alps samples were considerably higher than between the eastern Alps or southeastern European populations. Also of interest, the sample from the Pyrenees was highly differentiated from all other samples. Differences of this magnitude have been observed also in other species with strong intraspecific genetic structures evolved during long-lasting periods of allopatry in different refugial areas (Habel et al. 2005; Schmitt et al. 2006, 2007). Similar patterns of glacial survival and genetic differentiation were already shown in several recent papers published in Molecular Ecology (e.g. Schönswetter et al. 2003a, b; Deffontaine et al. 2005; Pinceel et al. 2005; Muster & Berendonk 2006; Hofman et al. 2007).
Schmitt and Haubrich have concluded that the mountains of southeastern Europe were the most important reservoirs of the coniferous forests in Europe. The other important aspect of glacial survival and postglacial expansion was shown by the wide distribution of the ‘northern’ racial group of E. euryale (E. euryale isarica) from the Carpathians to the Pyrenees, as opposed to the more fragmented pattern of southern races (Figs 2, 3). These results essentially coincide with the conclusions drawn from the molecular biogeography of other extra-Mediterranean faunal elements of Europe. Extra-Mediterranean core areas have been found for the widely dispersed, cold-tolerant frogs and reptiles Rana arvalis, Zootoca vivipara, Vipera berus (Surget-Groba et al. 2001; Babik et al. 2004; Ursenbacher et al. 2006). Carpathian populations proved to be the sources of colonization into more northern areas for these animals, as well as others such as bank voles and brown bears (Kotlík et al. 2006; Saarma et al. 2007). The identification of Carpathian centres of genetic diversity for so many species shows the importance of habitat conservation in this region.
Additionally, these results show the power of this type of study, and draw the attention to some unresolved problems of European phylogeography. It would be very interesting to further explore, for example, the pattern of extended but genetically more uniform distributions north of the Alps (as observed in E. euryale isarica and other species), and more southern refugial populations which regularly show more scattered pattern of distribution but higher levels of genetic diversity.