Cochlostoma obscurum ( Draparnaud, 1805 )

Gofas, Serge, 2001, The systematics of Pyrenean and Cantabrian Cochlostoma (Gastropoda, Cyclophoroidea) revisited, Journal of Natural History 35 (9), pp. 1277-1369 : 1288-1291

Cochlostoma obscurum ( Draparnaud, 1805)

(®gures 3, 5±9, 12±13)

Cyclostoma obscurum Draparnaud, 1805: 39 ±40, pl. 1, ®gure 13. Type locality` France septentrionale’, restricted to Burgundy by neotype designation (see remarks).

Pomatias Rayianum Bourguignat, 1857: 28 ±29, pl. 4 ®gure 5±9. Type locality: Bar s/ Seine , Riceys , Clairvaux (Aube, France).

Pomatias subobscurus Fagot , in Gourdon, 1890: 249. Type locality: central and western Pyrenees .

Description (based on populations from Burgundy)

Shell up to 10 Ö 4.8 mm with 7 to 8.5 quite rounded whorls. Early whorls with a peripheral angle which is totally concealed by the suture; body whorl hardly constricted, rounded. Sculpture very weak, wrinkled, on the ®rst embryonic half whorl, then with very ®ne axial ribs on the following one and one half whorls. Spire whorls with quite strong and irregular ribs (8±9 ribs/mm) on fourth and ®fth whorls, then with ribs becoming gradually more crowded (up to 12±13 ribs/mm); no terminal thickening of ribs along the suture.

Aperture moderately thickened and expanded into a collar, without a distinct rim inside. Peristome not distinctly auriculated, gradually narrowing and slightly deēcted inwards at the columellar insertion.

Shell colour tawny, with a distinct series of brown subsutural blotches, an interrupted suprasutural brown band continued on the periphery of the body whorl, and a rather broad, continued peri-umbilical brown band; whitish intervals between brown blotches formed by pale segments of the ribs. Peristome white.

Female and male genital systems as in general description.

Geographic range and variation

The populations assigned to C. obscurum are found in Burgundy in the deÂpartements of Yonne and CoÃte-d’Or, and then again in the Pyrenean foothills with a distribution gap of over 500 km.

Some populations from the Pyrenean foothills could not be recognized morphologically from those of Burgundy (®gure 8a,b), and for this reason are tentatively considered conspeci®c. All lowland populations have a rather thin outer lip, ®ne ribbing on the last whorl, and a contrasted colour pattern with a rather broad brown band on the body whorl. The size of the largest individuals in populations checked with electrophoresis did not exceed 10 mm in Burgundy, and 11.5 mm in the Pyrenean foothills (against over 13 mm in the Pyrenean populations assigned to C. crassilabrum ). However, there are specimens reaching 12.5 mm in other Pyrenean foothill localities which were not checked for allozymes (®gure 8c), and for which speci®c assignment remains uncertain.

Genetic distances calculated between Burgundy and Pyrenean foothills range between 0.18 and 0.21. Distances between foothill ( C. obscurum ) and montane ( C. crassilabum ) Pyrenean populations are higher and range between 0.32 and 0.42, thus approaching the range of interspeci®c distances (e.g. 0.45 to 0.67 for sympatric C. partioti vs. C. crassilabrum ). These do not share any allele for Octopine dehydrogenase. In the foothills, the St Martory population has three fast-moving electromorphs at ODH locus (C, D and N), and the population from Pau is ®xed for D. All three Burgundy populations are ®xed for the same D allele as in Burgundy. Conversely, the montane populations have a variable proportion of the slowmigrating alleles G and H, without any occurrence of C, D or N over 122 individuals examined (see ®gure 13). At the PGI locus, populations from Burgundy, Pau and St. Martory share an allele (PGI A) which is diOEerent from that ®xed in the montane populations (PGI D). The allozyme characters were the main basis for the recognition of two separate species, although admittedly in the absence of sympatry, it could still be tenable to treat C. crassilabrum as diOEerentiated populations of C. obscurum , and so to give it only subspecies status.

Habitat

Among limestone, with a preference for loose mounds of blocks in shaded places or woodland.

Remarks on nomenclature, type locality and type specimens

The nomenclatural implications of the decision to treat the populations from Burgundy and the Pyrenees as distinct species are not straightforward. The name Cyclostoma obscurum is often misdated from Draparnaud (1801) in the literature, but actually ®rst appeared in Draparnaud (1805) with`le Nord de la France’ (northern France) as the type locality. This wording is suggestive of Burgundy, and so has been understood by all later authors including Raven (1990) who designated a neotype from Arcy sur Cure (deÂpartement of Yonne, Burgundy, France). However the introduction of Draparnaud (1801: 35) points out that` France Septentrionale’ means the part of France where the Olive tree does not grow (i.e. extra-Mediterranean) and thus does not preclude the Pyrenees.

The neotype designation of Raven (1990) is anyway invalid under the provisions of the fourth edition of the International Code of Zoological Nomenclature, which states that original type material has precedence over a neotype. The specimen in the Draparnaud collection, ®gured in Wagner (1897: 67; pl. 3) is still extant in Naturhistorisches Museum, Vienna, Austria (G. Falkner, personal commmunication). This specimen measures 13 mm and such size can only be found in populations from the French Pyrenees (deÂpartements of PyreÂneÂes Atlantiques and Hautes PyreÂneÂes). Morphologically, the specimen has a rather weakly calci®ed outer lip and somewhat resembles the lowland forms, although larger.

Thus, this specimen would not allow a ®nal decision on whether the name obscurum should be applied to lowland populations (with ODH C, D or N) or to the montane populations. The ®rst option may be suggested by the shape of the specimen but is highly questionable because the diagnostic lowland allozymes were not found, and are unlikely to be found in a specimen of that size. The second option is not desirable because, whenever authors intended to distinguish the lowland and montane populations at the species level (e.g. Dupuy, 1849; Locard, 1894; Germain, 1931; Kerney et al., 1983), the name C. obscurum in a restricted sense was used for the lowland and C. crassilabrum was used for the Pyrenees. For these reasons, I consider that Raven’ s (1990) neotype best serves the stability of nomenclature, and will apply to the International Commission for Zoological Nomenclature to request that it be validated.

An alternative to using the name C. obscurum is reviving an earlier name, C. conicum [Vallot, 1801] introduced in an anonymous and little known leaēt describing some of the molluscs of Burgundy, issued in 1801 and later attributed to J. N. Vallot by Moquin-Tandon (1855). This name was known to Raven (1990) who did not use it and intended (but did not proceed) to request its suppression. However this leaēt was printed as an examination subject for Ecole Centrale du deÂpartement de la CoÃte d’ Or, and contains (p. 8) a list of addressees. I consider that this fails to meet the criterion of being`issued publicly for the purpose of permanent scienti®c record’ and will treat the names contained therein as not available. The following speci®c name with a type locality in Burgundy, Pomatias rayianum , is based on abnormal specimens having a keeled body whorl, and for this reason should not be used.

The name Pomatias subobscurum was intended for Pyrenean representatives of C. obscurum of earlier authors (i.e. foothill populations) by Fagot (1891), but there are also nomenclatural problems with it. The name actually appeared earlier in Gourdon (1890) but neither Gourdon (1890) nor Fagot (1891) gave a description. The statement `Pomatias obscurum of earlier authors’ can be considered as an indication, but this is questionable. The ®rst formal description which unquestionabl y makes the name available was published by Locard (1894) but could apply to any of the Pyrenean morphs, and syntypes in the Locard collection in MNHN originate from montane localities and have a crassilabrum morphology.