A survey of vitrinid land snails (Gastropoda: Pulmonata: Limacoidea)

There is general agreement that the vitrinids, a mainly Palaearctic group of semislugs, are higher stylommatophorans, but uncertainty still exists about their relationships. We therefore conducted a phylogenetic study based on morphological characters to determine relationships between genus-group taxa and to test monophyly and revise current taxonomy. Eighteen genus-group taxa those listed as valid in the recent literature plus the new genus Azorivitrina were considered using an exemplar approach, with individual species as terminals to avoid a priori assumption of monophyly. At least two species, one of which was the type species, were examined when possible. Phylogenetic analyses were performed using different outgroups consisting of a selection of western Palaearctic limacoidean genera (data partition I, DPI: outgroup consisting of Euconulus, Deroceras, Oxychilus, Tandonia, Vitrea and Troglaegopis; DPII: outgroup Euconulus, Oxychilus, Vitrea and Troglaegopis; DPIII: outgroup Euconulus and Oxychilus). Different approaches were used: parsimony analysis (MP: using P-AUP* 4.0; AP: TNT software) and bayesian inference (BI: using MRBAYES). Phylogeny analysis generally had low resolution. The parsimony analysis using implied weights (TNT software) provided the best results (AP found 11 supported clades, MP 11 and BI 8). It did not support monophyly of most genus-group taxa or the monophyletic groups found by Hausdorf (1995, 2002) and Alonso et al. (2000), or the subfamilies and tribes established by Schileyko (1986, 2003), or the family Vitrinidae when the outgroup consisted of a large selection of limacoidean taxa. Support was only found for Canarivitrina, Guerrina, Oligolimax and Vitrinobrachium and some support for Arabivitrina and Azorivitrina. On the contrary, no support was found for Eucobresia, Insulivitrina and Phenacolimax, or for Plutonia sensu Alonso et al. (2000). The most resolved group (but with low support values) was a clade including species of Oligolimax, Sardovitrina, Semilimacella and Vitrina supported by exclusive disposition of the penial retractor above the right ommatophore retractor. These results indicate that a different approach to vitrinid phylogeny and systematics is needed because morphological characters alone do not generate a realistic picture. In the meantime, we ranked the 18 genus-group taxa as distinct genera, although we are perfectly aware that some may be paraphyletic or polyphyletic. Finally we concisely surveyed all vitrinid genera, listing them in three distinct groups according to stimulator structure: vitrinids with vaginal stimulator (glandula amatoria); vitrinids with diverticular stimulator (atrial/atrial-vaginal/vaginal or penial diverticular stimulator); vitrinids without vaginal or diverticular stimulator. This approach enables easy comparison of taxa with similar distal genital structure and is not intended to have any systematic or phylogenetic value. A short diagnosis of each taxon is given together with re-description of the type species and remarks