Stag beetles of the genus Dorcus MacLeay in North America ( Coleoptera , Lucanidae )

General confusion has surrounded the status of Nearctic Dorcus species since the 19th Century. In this paper the two Nearctic species are clarifi ed and compared, and morphological characters discussed that will readily distinguish them. Examination of the type specimens of D. mexicanus reveals that they are actually mislabeled specimens and that the species is a new synonym of the Palearctic D. parallelipipedus (L.).


Introduction
Th e status of the North American species of Dorcus MacLeay has been the subject of debate since the late 1800s. On more than one occasion I have been asked how many species are present in our fauna, and so the purpose of this paper is to discuss the morphology of the North American species and provide the relevant characters for identifi cation. Factors that have complicated the taxonomy are the relatively small numbers of known specimens of D. brevis, the markedly diff erent morphology of large and small specimens due to allometric growth, and the general trend in Dorcus species for larger individuals to be less punctate.
Say described both Lucanus parallelus (1823) and L. brevis (1825). Th e former was subsequently transferred to Dorcus by Burmeister (1847), and the name L. oblongus de Charpentier, 1825 (based on a pair mislabeled as coming from the Pyrenees) was treated as a synonym. Th e description of new species based on mislabeled specimens is a frequent occurrence in Lucanidae, especially in Dorcus; D. caucasicus Ganglbauer, 1886 (mislabeled as coming from the Caucasus Mountains) was reduced to a synonym of D. brevis by Reitter (1892). Angell (1916) described the form D. carnochani, a synonym of D. parallelus.
A history of the ensuing debate over the distinctness of D. brevis and D. parallelus was provided in Benesh (1937). Th ese species were frequently declared to be identical or merely forms of the same species (Parry 1870; Fuchs 1882; Horn 1892; Wickham 1899). Skinner (1911) planted the fi rst seeds of doubt concerning the synonymy but stopped short of arguing for the 'specifi c value' of D. brevis. Benesh (1937) distinguished the two species by the form of their posterior angles of the pronotum and overall body shape. His concepts of both species appear to be more or less correct based on specimens illustrated. However, his illustration of the male genitalia of neither species matches the actual morphology, which could indicate that the genitalia studied were broken during dissection or improperly relaxed. If male specimens with short fl agella were identifi ed by Benesh as D. parallelus, the locality information given in that paper and in Benesh (1942) may be suspect. Also, his characterization of the female mandibles as unidentate in D. brevis and bidentate in D. parallelus does not appear to be correct because no obvious diff erence in the dentition of female mandibles occurs in these species. However, based on the specimens examined it appears that D. brevis is widely distributed in the eastern United States. Benesh (1944) described a third North American species, Dorcus mexicanus, from a male and female specimen in the Field Museum (FMNH) collection, the male with the locality as Jalapa, Mexico. Reyes-Castillo and Boucher (2003) noted that the distribution of the species in Mexico was unknown. Other than the type series, no other specimens of this species have been reported. Only one element of the Nearctic lucanid fauna, Lucanus mazama (LeConte), is known to extend into northern Mexico (Paulsen 2005). In a biogeographical sense, it seems unlikely that a species in the Holarctic genus Dorcus would be endemic to Neotropical Mexico. For this reason, I examined and compared the type specimens of D. mexicanus with the known species in the genus.

Dorcus brevis and D. parallelus
Th e overall broader form of the body of D. brevis (Fig. 1) is often adequate to distinguish it from the relatively narrower D. parallelus (Fig. 2). However, the most useful and obvious external character for separating the North American species is the shape of the posterior pronotal angle. In D. brevis, the posterior angles are in close proximity to the elytral bases ( Fig. 3). In D. parallelus, the angles are distinctly removed from the base of the elytra (Fig.  4). Th e humeral angles of D. brevis are generally more strongly dentate and the humeri produced forward of the scutellum, while in D. parallelus the humerus is less strongly dentate and more or less in line with the scutellum. In addition, males of D. parallelus have a dense fi eld of setae on the internal face of the metatibia (Fig. 5), but this patch is not present on males of D. brevis (Fig. 6). Th e clypeus is distinctly broader in males of D. parallelus.
In both species, the mandibles of major males have a single large dorsal tooth, but the dentition of the mandibles of minor males is clearly distinct. Even the smallest minor males of D. parallelus have mandibles that are of the same basic shape as major males, but they are simply reduced in size (Fig. 7). In contrast, minor males of D. brevis have two distinct internal teeth below the dorsal tooth and are abruptly curved internally (Fig. 8).
Th e form of the male genitalia is radically diff erent in the two species. Th e fl agellum of D. parallelus is longer than the entire length of the body and is weakly fl ared at the apex (Fig.  9). Th e fl agellum of D. brevis is less than half as long, with a large lobe-like sac at the apex (Fig. 10). Th e genitalia of the European species, D. parallelipipedus, diff ers from either species in being more strongly expanded medially and in possessing a trilobed apex (Fig. 11).

Dorcus mexicanus
Examination of the type specimens of D. mexicanus revealed that they are Dorcus parallelipipedus (L.), a Palearctic species. Th e identical male genitalia (Fig. 11), protruding clypeus of males, irregularly punctate elytra, and bituberculate frons of the female provide overwhelming evidence to support this conclusion. Th us, the name D. mexicanus Benesh is here reduced to a synonym of D. parallelipipedus (new synonymy). Coincidentally, I encountered a second pair of D. parallelipipedus recently accessioned at the FMNH that bore handwritten labels indicating "Mexico, Nuevo Leon, Monterey (sic), Aug 1976". Th e presence of a second, more recent pair of D. parallelipipedus from Mexico at fi rst suggested that the species might be introduced there. However, on further examination I found that other handwritten locality labels of the same style from the accessioned collection were erroneous (with incorrect countries). While it is not impossible that the species is adventive in Mexico, it is more likely given the nomenclatural history of Dorcus that the common D. parallelipipedus has been mislabeled in these two instances.