Atractus

MYERS, CHARLES W., 2003, Rare Snakes-Five New Species from Eastern Panama: Reviews of Northern Atractus and Southern Geophis (Colubridae: Dipsadinae), American Museum Novitates 3391, pp. 1-48 : 8-9

GENUS ATRACTUS

Atractus is a large South American genus whose range is now known to include the eastern half of Panama (map 1), where it seems to be exceedingly rare. The four whole specimens (figs. 1, 2) and the one head (fig. 10) known to me represent five species! Two of these specimens come from the low uplands east of the Canal Zone and three are from Darién; all localities are in well­drained (100 to> 500 m) monsoon rain forest (Myers, 1969) .

I have been unable to match any of the Panamanian specimens with previously named South American species. But identifying Atractus is often difficult owing to the many inadequate, poorly illustrated descriptions that usually emphasize characters present in a majority of species. A major review of the genus is needed. There have been good taxonomic and variational studies for Ecuador ( Savage, 1960), Venezuela ( Roze, 1961, 1966), Surinam ( Hoogmoed, 1980), and eastern and central Amazonia ( Cunha and Nascimento, 1983; Martins and Oliveira, 1993)—but the situation is less good for Colombia, which has a rich Atractus fauna of several dozen named species, whose variation and relationships are virtually unknown (lists of nominal species in Daniel, 1949; Peters and Orejas­Miranda, 1970; Sanchez­C. et al., 1987; Pérez­Santos and Moreno, 1988). Although a majority of Colombian Atractus are Andean, it would not be surprising if an older Colombian (or other South American) name were eventually found to be applicable to one of the purportedly new species herein described as Panamanian endemics. To ease the burden of workers who may have to decide this point without ready access to all holotypes, the five Panamanian specimens are described and illustrated in some detail. I additionally discuss and illustrate a Colombian specimen of Atractus clarki Dunn and Bailey , a species heretofore reported only from the Panamanian holotype.

This genus has been confused with Geo­ phis, but, in Panama, known Atractus are easily distinguished by the generic characteristic of a single pair of genials (two pairs in southern Geophis ) and 17 rows of smooth scales (15 rows of scales, smooth or posteriorly keeled in southern Geophis ); also, the southern Geophis have a usually diamondshaped or rhomboidal frontal plate that is strongly angular anteriorly (as in fig. 12), whereas Panamanian Atractus have a more normal colubrid frontal that is roughly triangular or pentagonal, with only a small anterior apex at the prefrontal suture (fig. 3).

Unfortunately, the four whole specimens of Panamanian Atractus are all females and furnish no data on male genitalia, although opportunity is taken to describe and illustrate a hemipenis from the Colombian specimen of A. clarki . Hemipenial differences mentioned by Savage (1960: 30) have since broken down with further study of Geophis (see Downs, 1967: 184, and discussion herein un­ der Geophis bellus ), and the distinction is further blurred by the hemipenis of Atractus clarki (q.v.). Cadle (1984) considered Atractus a South American member of a mainly Central American xenodontine clade—now recognized as the subfamily Dipsadinae —for which the most diagnostic of several derived hemipenial features is the distal division of the sulcus spermaticus ( Myers and Cadle, 1994: 27; Zaher, 1999: 33). The sulcus spermaticus in the Dipsadinae usually divides close to or within the capitulum; the hemipenis of A. clarki is noncapitate and the sulcus divides slightly below the midpoint of the organ (at least in the retracted condition)—somewhat lower than in most dipsadines but still higher than the basal bifurcation in the great majority of Xenodontinae (in which the sulcate bifurcation may rarely approach the midpoint of the organ, e.g., see Zaher, 1999: fig. 64, upper).

Atractus and Geophis have been often compared, but the closeness of their relationship is questionable and they are probably convergent to a similar life­style. On the basis of immunological comparisons, Cadle (1984) saw a closer relationship for Geophis with Ninia and other Middle American dipsadines than with Peruvian Atractus elaps and A. major . Zaher (1999: 33) suggested, on the basis of a muscle character, an espe­