The hazel dormouse, Muscardinus avellanarius is a hibernating rodent, member of the Gliridae family. It is present in most of Europe and western Russia with the exception of the Iberian Peninsula and the South-West of France , the North of Belgium and the Netherlands as well as the greater part of northern Germany and Poland. In its North-Western distribution area, the populations are declining because of the interplay between the biological requirements of the species complex and the habitat loss and fragmentation. Concrete conservation measures must be implemented to maintain long-term viable populations of this species. At the European level, the dormouse is a strictly protected species listed in Annex IV of the " Fauna-Flora -Habitat Directive" ( Directive 92/43/EEC ) and Annex II of the Bern Convention.
To develop appropriate tools for conservation, it is essential to understand the origin and the evolution of this species whose ancestor appeared 17 million years ago. To this aim, it is fundamental to analyze the evolutionary history of the species in both a geographic dimension (spatial) and a temporal scale. The temporal scale allows to better understanding how the species answered to past climate changes. On the other hand, understanding the impact of past climatic events may facilitate predictions of how species will respond to future climate changes. A better understanding of the genetic diversity’s spatial distribution is also important to determine regions with higher genetic diversity than others. This information may be particularly important to estimate the sensitivity of these species to environmental changes (pollution, fragmentation, climate) .
Both mitochondrial and nuclear markers revealed a complex phylogeographic pattern for the hazel dormouse with the presence of two highly divergent and allopatric genetic lineages, respectively distributed in South-Western and Central-Eastern Europe. A finer structure within each group was also observed. Divergence time analyses gave an ancient separation between the two lineages which is estimated in the late Miocene (6 million years ) while the intra lineage differentiation within M. avellanarius populations happened during the first Quaternary glaciations (2.8 million years). These two periods are characterized by important climate oscillations, by the definitive decline of Tertiary forests in north-western Europe and by deep geological modifications. All these factors may have triggered the differentiation of the hazel dormouse into several genetic lineages. Similar studies on the edible and the garden dormouse retrieved deeply divergent lineages, which predated the Quaternary glaciations. However, these closely related species with common physiological traits such as hibernation, seem to have reacted differently to past environmental changes. In line with expectations, our preliminary study of ecological niche modeling shows that these three species would react differently to future climate changes.
In a more regional and local scale, our study tends to show that hazel dormice populations are highly fragmented. It seems that this fragmentation is the result of past environmental and anthropogenic changes that occurred during the Holocene. Furthermore the common dormice inhabiting forests fragments represent genetically isolated populations and few migrations seem to happen between patches. These results evidence an important isolation among the observed populations, which may lead to inbreeding depression. They might be therefore much more vulnerable to environmental changes. Our study strongly suggests that the dormice populations are particularly vulnerable and important conservation measures must be taken for its survivalFinally, the presence of two highly differentiated phylogenetic lineages might question the taxonomic status of the hazel dormouse. Our work has highlighted the difficulty of delineating species boundaries. A sequence-based species delimitation method completed by the analysis of the level of genetic divergence was used to define species boundaries within our samples and revealed that the number of possible/putative species is between 1 and 10. The genetic evidence on its own does not provide conclusive evidence on species limits. Our results stressed the importance of integrating complementary approaches to achieve further taxonomic work. We used a quantitative analysis of intra- specific phenotypic variation as a complementary tool of the molecular approach by studying the upper first molar (UM1) in dormouse populations. Preliminary geometric morphometric analyses failed to confirm the presence of two lineages. On the other hand, we evidenced an intraspecific variation within lineages that might be explained by local adaptation. These results should be confirmed by further analysis. In conclusion, the knowledge acquired through this thesis will be a significant piece of the puzzle for the understanding of the evolutionary history of the hazel dormouse and may have important implications for its conservation.