A targeted gene phylogenetic framework to investigate diversification in the highly diverse yet geographically restricted red devil spiders (Araneae, Dysderidae).

The family Dysderidae is a highly diverse group of nocturnal ground-dwelling and active-hunter spiders. Dysderids are mostly restricted to the Western Palearctic, and particularly rich and abundant around the Mediterranean region. Interestingly, the distribution of species richness among its 24 genera and three subfamilies is highly biased-80% of its 644 documented species belong to just two genera, Dysdera (326) and Harpactea (211). Dysderidae provides an excellent study case for evolutionary and ecological research. It includes cases of trophic specialization, which are uncommon among spiders, and exhibit other remarkable biological (e.g. holocentric chromosomes), behavioural (e.g. cryptic female choice), evolutionary (e.g. adaptive radiation) and ecological features (e.g. recurrent colonization of the subterranean environment). The lack of a quantitative hypothesis on its phylogenetic structure has hampered its potential as a testing ground for evolutionary, biogeographical and ecological hypotheses. Here, we present the results of a target, multi-locus phylogenetic analysis, using mitochondrial (cox1, 16s and 12s) and nuclear genes (h3, 28s and 18s), of the most exhaustive taxonomic sample within Dysderidae (104 spp.) to date and across related families (Synspermiata) (83 spp.). We estimate divergence times using a combination of fossil and biogeographic node calibrations and use this timeline to identify shifts in diversification rates. Our results support the monophyly of the Dysderidae subfamilies Rhodinae and Dysderinae but reject Harpacteinae as currently defined. Moreover, the clades recovered within Harpacteinae do not support its current taxonomy. The origin of the family most likely post-dated the break-up of Pangea, and cave colonization may be older than previously considered. After correcting for the taxonomic artefacts, we identified a significant shift in diversification rates at the base of the genus Dysdera. Although the unique coexistence of specialist and generalist diets within the lineage could be suggested as the potential driver for the rate acceleration, further quantitative analyses would be necessary to test this hypothesis.