Vitalius nondescriptus comb . nov . ( Araneae : Theraphosidae : Theraphosinae ) : an example of theraphosid taxonomic chaos

The male holotype of Hapalopus nondescriptus Mello-Leitão, 1926 is redescribed, illustrated and compared with freshly collected specimens from the type locality. The only difference noted among the holotype and the new material concerns the development of the subapical keel. Its taxonomic position is reinterpreted and discussed, resulting in its transfer to the genus Vitalius Lucas, Silva Junior & Bertani, 1993, and thus making the new combination Vitalius nondescriptus (Mello-Leitão, 1926) comb. nov. The female is described for the first time and the morphological variations in two males, born from the female used in the description, is presented and illustrated. The male differs from those of other Vitalius species by the palpal bulb with short apical keel and bifid tibial spur with narrow prolateral branch and almost straight retrolateral branch. The female differs from those of other Vitalius species by urticating hair of ‘type I’ having the region ‘a’ shorter than region ‘b’. Hapalopus nondescriptus has a confusing taxonomic history, since the holotype specimen was also used to describe another theraphosid species (Cyclosternum melloleitaoi Bücherl, Thimoteo & Lucas, 1971) which was, consequently, considered its objective synonym. Thus, we consider it a clear example of theraphosid taxonomical chaos.

The mygalomorph Theraphosidae comprises most of the largest living spiders, commonly called "tarantulas".With almost one thousand species described (PLATNICK 2012), it is one of the richest spider families.The taxonomic difficulties of the family are also well known.RAVEN (1990) called attention to the lack of taxonomic revisions in Theraphosidae, "either in its entirety or in any region"; and said that "a group so long left without revision becomes a nomenclatural and taxonomic nightmare".Despite the progress in the taxonomy of Theraphosidae in the last twenty years, much is yet to be done.
In this work we elucidate the identity of the species Hapalopus nondescriptus Mello-Leitão, 1926, which has a confusing history.It was described based on a single male from Ouro Preto, state of Minas Gerais, Brazil.The author considered it close to Hapalopus flovohirtus Simon, 1889 (now Catanduba flavohirta).Many years later, BÜCHERL et al. (1971) reviewed various types established by Mello-Leitão housed at the Museu Nacional, Rio de Janeiro.They found a specimen labeled "Hapalopus sp.", from Ouro Preto, but did not recognize it as a Mello-Leitão type.They considered it an undescribed species of Cyclosternum Ausserer, 1871 and decided to describe this new species, naming it Cyclosternum melloleitaoi Bücherl, Timotheo & Lucas, 1971.
Recently, a catalog of spider types described by Mello-Leitão was published (SILVA-MOREIRA et al. 2010) and the authors stated that there is no doubt that both species were described using the same specimen as a holotype, making Cyclosternum melloleitaoi an objective synonym of Hapalopus nondescriptus.However, PLATNICK (2012) considered that SILVA-MOREIRA et al. (2010) "provided no evidence justifying a placement of the species in either genus" and that "their conclusion should be checked by a theraphosid systematist before any change is accepted".Both species are thus listed as valid in Platnick's Catalog (PLATNICK 2012).
The case involving H. nondescriptus and C. melloleitaoi is a clear example of the theraphosid "taxonomic chaos" to which RAVEN (1990) referred.All efforts to clarify the identity of one or more theraphosid species are important to reverse this taxonomic chaos.We redescribe the species, describe the female for the first time, and reinterpret its taxonomic position after studying the holotype and additional specimens.
All measurements are in millimeters and were obtained with a Mitutoyo ® digital caliper with an error of 0.005 mm, rounded up to two significant decimals.Leg and palp measure-ABSTRACT.The male holotype of Hapalopus nondescriptus Mello-Leitão, 1926 is redescribed, illustrated and compared with freshly collected specimens from the type locality.The only difference noted among the holotype and the new material concerns the development of the subapical keel.Its taxonomic position is reinterpreted and discussed, resulting in its transfer to the genus Vitalius Lucas, Silva Junior & Bertani, 1993, and thus making the new combination Vitalius nondescriptus (Mello-Leitão, 1926) comb.nov.The female is described for the first time and the morphological variations in two males, born from the female used in the description, is presented and illustrated.The male differs from those of other Vitalius species by the palpal bulb with short apical keel and bifid tibial spur with narrow prolateral branch and almost straight retrolateral branch.The female differs from those of other Vitalius species by urticating hair of 'type I' having the region 'a' shorter than region 'b'.Hapalopus nondescriptus has a confusing taxonomic history, since the holotype specimen was also used to describe another theraphosid species (Cyclosternum melloleitaoi Bücherl, Thimoteo & Lucas, 1971) which was, consequently, considered its objective synonym.Thus, we consider it a clear example of theraphosid taxonomical chaos.KEY WORDS.Brazil; Cyclosternum; Hapalopus; objeticve synonym; taxonomy.
Geographical coordinates: primary sources are between round brackets, secondary sources (Google Earth © ) between square brackets.

TAXONOMY
Vitalius nondescriptus (Mello-Leitão, 1926) comb.nov.Diagnosis.The male differs from those of other Vitalius species by the palpal bulb with short apical keel (Figs 1-2 and 5-8) and bifid tibial spur with narrow prolateral branch and almost straight retrolateral branch (Fig. 3).The female differs from those of other Vitalius species by urticating hair of type I having the region 'a' shorter than region 'b' (Fig. 9).
Description.Female (Fig. 11).MZSP 465351.Carapace 12.55 long, 11.87 wide, chelicerae 7.80.Leg measurements in Table III ZOOLOGIA 29 (5): 467-473, October, 2012 Natural History.The female MZSP 465351 was found with egg sac on 11 January 2005 under a fallen tree trunk.The mother and the egg sac were maintained in captivity and on 28 January 2005, 52 spiderlings emerged from the egg sac.Of these, two males reached maturity, in January 2011 and February 2012.

DISCUSSION
When comparing the holotype of H. nondescriptus with freshly collected specimens from the type locality, we found agreement in most morphological characteristics such as size, color pattern, legs and palpal spination, tibial spur shape, folding of metatarsus I contacting the retrolateral spur branch, the short embolus with short apical keel (Figs 1-2), and the morphology of type I urticating hair with region 'a' shorter than region 'b' (Fig. 9).On the other hand, the subapical keel in the new examined material is well-developed, similar to those seen in Vitalius Lucas, Silva Junior & Bertani, 1993, Nhandu Lucas, 1982and Lasiodora C.L. Koch, 1850 species (Figs 5-8).Thus, the only difference noted among the holotype and the new material concerns the development of the subapical keel, which can vary for some taxa, e.g.Vitalius vellutinus (Mello-Leitão, 1923) from small to vestigial (BERTANI 2001).It is more plausible to consider that the holotype presents developmental variation of the subapical keel rather than having two very similar, sympatric species share all the characters shown above, yet only differing by a minor variance in bulb morphology.
The species herein under study is a theraphosine with type I urticating hair on the abdomen dorsum, spermathecae fused at a small area (Fig. 4), absence of long hairs on female's carapace, a tibial spur with two branches originating from a common base (Fig. 3), presence of retrolateral scopulae on femur IV, absence of stridulatory setae on trochanter and coxae of all legs and palp, and presence of the following keels on male bulb: prolateral superior, prolateral inferior, retrolateral, apical and subapical keel (Figs 1-2 and 5-8).This set of characters indicates close affinities with Vitalius.Therefore, we transfered Hapalopus nondescriptus to Vitalius, making the new combination Vitalius nondescriptus (Mello-Leitao, 1926) comb.nov.

Table III .
Vitalius nondescriptus comb.nov.Female MZSP 465351 from Ouro Preto, state of Minas Gerais, Brazil.Length and midwidths of right legs and palpal segments.
PÉREZ-MILES et al. 1996gstoniSchmidt & Weinmann, 1996 (PÉREZ- MILES et al. 2008), but not in Vitalius(BERTANI 2001).The presence of this urticating hair type I modification is clearly homoplasy among the three genera (above) and V. nondescriptus comb.nov., because they are not closely related (see cladograms ofPÉREZ-MILES et al. 1996 for Citharacanthus, BERTANI 2001 for  Proshapalopus, and PÉREZ-MILES et al. 2008for Ami).On the other hand, the characters short apical keel and tibial spur shape could represent homoplasious or plesiomorphic characters in relation to the related Lasiodora (short apical keel and tibial spur shape) and Nhandu (tibial spur shape).