Into the Chamber of Horrors: A proposal for the resolution of nomenclatural chaos in the Scaphiophryne calcarata complex (Anura: Microhylidae), with a new species-level phylogenetic hypothesis for Scaphiophryninae

The genus Scaphiophryne (Anura: Microhylidae) contains at least nine species that, together with their probable sister genus Paradoxophyla, form the Madagascar-endemic subfamily Scaphiophryninae. Scaphiophryne are robust burrowing frogs with explosive breeding behavior and are characterized by a unique larval morphology, being intermediate between that of filter-feeding and generalized tadpole types. Based on tadpole morphology, the nominal subgenus Scaphiophryne can be distinguished from the subgenus Pseudohemisus, a clade occurring in arid biomes with a largely unsolved taxonomy and nomenclature. Here we combine a comprehensive molecular phylogeny of all scaphiophrynines based on five mitochondrial and two nuclear genes with a re-examination of the name-bearing type specimens of the seven specieslevel nomina assignable to the subgenus Pseudohemisus. Our molecular phylogeny supports monophyly of the two subgenera and assigns the enigmatic S. matsoko to the subgenus Scaphiophryne. In this subgenus, our analysis confirms all currently accepted species as independent evolutionary lineages based on concordant differentiation in mitochondrial and nuclear genes, and provides evidence for a combination of low mitochondrial divergence, distinct nuclear divergence, and high nuclear heterozygosity. Yet, contrary to previous hypotheses, we do not find evidence of ongoing hybridization. In the subgenus Pseudohemisus we find support for three clearly delimited species-level lineages, one of which contains two deep conspecific lineages that warrant further study. To stabilize the taxonomy in this subgenus, and considering the very poor state of preservation of several type series and the heterogeneity of some of them, we here designate lectotypes for the nomina: Hemisus obscurus Grandidier, 1872; Calophrynus calcaratus Mocquard, 1895; Pseudohemisus verrucosus Angel, 1930; Pseudohemisus longimanus Angel, 1930; and Pseudohemisus longimanus var. melanopleura Angel, 1934. Taxonomically, we accept Scaphiophryne brevis (Boulenger, 1896) as a well-defined species; restrict S. calcarata (Mocquard, 1995) to the southern lineage of the S. calcarata complex; resurrect Hemisus obscurus Grandidier, 1872 (previously considered to be a dubious name or nomen inquirendum) as Scaphiophryne obscura to refer to the western and north-western lineage of the S. calcarata complex; consider Pseudohemisus verrucosus Angel, 1930 as junior synonym of Scaphiophryne brevis; and consider Pseudohemisus longimanus Angel, 1930 and Pseudohemisus longimanus var. melanopleura Angel, 1934 as junior synonyms of Scaphiophryne obscura. Finally, to accelerate taxonomic progress, we (i) outline a set of criteria by which such old and ambiguous nomina can be applied to known lineages, (ii) suggest that the International Code of Zoological Nomenclature should slightly broaden the conditions under which neotypes are to be designated, and (iii) propose to redouble discussions on the introduction of the concept of the epitype to zoology.