The spine armament of the legs as an important means for the characterisation of the genera of Corydiinae and their relationships (Blattodea, Corydiidae)

The Palaearctic species of Corydiinae having no apical spine on the femora of the frontleg (48 species distributed to the genera Heterogamisca, Heterogamodes, Hemelytroblatta, Leiopteroblatta, Mononychoblatta, Nymphrytria, Psammoblatta) were studied in mainly three characters: tibia armament, features of the right phallomere, and distribution of bristles on the subgenital plate. The results suggest considerable changes in the classification with now 5 genera and several subgenera: (1) Heterogamodes with subgenera Heterogamodes, Heterogamisca, and Atuberculoblatta subgen. nov.; (2) Hemelytroblatta with subgenera Hemelytroblatta and Mollidentoblatta subgen. nov.; (3) Nymphrytria with subgenera Nymphrytria, Mononychoblatta, and Leiopteroblatta; (4) Psammoblatta; and (5) Fissidentoblatta gen. nov. The presumed relationships between the genera (1)–(3) are shown in a cladogram (Fig. 23). Four species of the subgenus Heterogamisca are described as new: H. longipilosa, H. sulcata, H. brevipennis, and H. pallens; a fifth species, H. jeffreyana, member of the Maltese fauna, had recently been described (Bohn the Arenivaga-group with the Nearctic genera Arenivaga and Eremoblatta and the Palaearctic genus Polyphagina; and the Polyphaga-group containing the genera Anisogamia, Eupolyphaga, Polyphaga and all genera already mentioned in the first paragraph. The Therea-group most likely represents the most basal group of the Corydiinae, but appears as already split into two branches, with the Nearctic Homoeogamia on the one branch finally terminating in the Arenivaga-group, and the three other Old World genera on the other branch terminating in the Polyphaga-group (Fig. 24). The proposed phylogenetic relationships can only be seen as tentative and need proof on both fields of methods: on the morphological side the complex structures of the phallomeres still offer a rich arsenal of different structures hitherto certainly only partly used; on the side of the molecular methods a large extension of the number of genera to be studied is necessary including all known genera of Corydiinae. The unition of Princis’ Polyphagidae and Homoeogamiidae to one subfamily, Corydiinae, by Grandcolas 1994a appears justified, but with the exclusion of Cryptocercus. The genera Austropolyphaga, and Polyphagoides, later added by Roth (2003), have to be removed from Corydiinae: shape and size of the postclypeal shield exclude their belonging to this subfamily and there are no relevant characters known which the two genera share with Corydiinae. The same applies to the presumably erroneous addition of Hypercompsa to Corydiinae by Roth. Tivia inconspicua (Tiviinae) differs in several important characters from the other Tivia species. It is placed in a newly errected genus: Tiviella gen. nov..