Exploring the diversity of a neglected group of soil invertebrates (Chilopoda) across the South-Eastern Prealps

Despite their documented importance in maintaining ecosystem functioning, many groups of soildwelling invertebrates have been inadequately studied with regards to many facets of their biology. In this thesis, I focussed my attention on two poorly understood aspects of the diversity of soildwelling invertebrates: (i) local values of species richness and (ii) species boundaries. As a study system, I used one of the most neglected groups of soil invertebrates, i.e. the centipedes (Myriapoda: Chilopoda), and I carried out my research in the area of the Southeastern Prealps, which are known to harbour a large amount of biodiversity. The present work is in the form of a paper collection and contains a general introduction, two chapters and a conclusive paragraph. An original research article, dealing with the aforementioned topics, is presented at the end of each chapter. In chapter I, I addressed the dearth of knowledge on the local species richness of centipede communities. I applied statistical models to estimate the actual species richness of 10 centipede communities in the Southeastern Prealps and to compare richness values between these communities. I demonstrated that up to 27-28 species of centipedes can coexist in syntopy in temperate forests of the Southeastern Prealps, and that richness values can vary significantly among communities. In chapter II, I addressed the issue of species delimitation in poorly-vagile endogeic invertebrates, for which traditional morphology-based taxonomy has proven ineffective in identifying species boundaries. In particular, I applied a rigorous integrative approach, in order to test for the existence of species boundaries in a set of populations traditionally referred to a single species of endogeic centipede, i.e., Clinopodes carinthiacus (Chilopoda: Geophilomorpha). Taking advantage of the integration between different lines of evidence (sequences from three DNA loci and morphological characters) and different methods for species discovery (ABGD, GMYC, PTP and Expectation Maximisation cluster analysis with v-fold cross-validation), I found evidence for the existence of at least two candidate species within the investigated population system.