Cerapachys sexspinus

Borowiec, M. L., 2009, New ant species related to Cerapachys sexspinus and discussion of the status of Yunodorylus (Hymenoptera: Formicidae)., Zootaxa 2069, pp. 43-58 : 53-57

Cerapachys sexspinus (Xu, 2000)

Figs. 9-11

Yunodorylus sexspinus Xu , 2000: 298-299. Figs 1-6. Holotype and paratype workers, CHINA: Yunnan province, Mengla county, Bubang village, 730 m, No. A97-2064, 17 VIII 1997, nest in soil, seasonal rainforest (G Zeng); further paratype workers, Yunnan province, Mengla county, Longlin village, 1040 m, No. A98-775, 14 III 1998 (G Zeng) [ ISAS, MCZC] ( MCZC paratypes examined).

Cerapachys sexspinus : Bolton 2003: 269. First combination in Cerapachys .

Cerapachyssexspenus: Jaitrong & Nabhitabhata 2005: 18. Incorrect subsequent spelling. Record in Thailand.

Worker measurements: HW 0.53-0.68, HL 0.63-0.74, SL 0.30-0.36, MH 0.38-0.44, ML. 0.75-0.94, PrW 0.36-0.49, PW 0.29-0.36, PL 0.24-0.29, IIIAW 0.40-0.49, IIIAL 0.29-0.33, IVAW 0.53-0.64, IVAL 0.30-0.37, FFeL 0.38-0.46, HFel 0.36-0.45, FTiL 0.31-0.38, HTiL 0.38-0.48, FBaL 0.19-0.23, HBaL 0.28-0.35, CI 109-119, MI 197-214, PI 81-83 [2 measured] Head slightly longer than wide and widest at about midlength; sides parallel, convex. Vertexal margin concave. Parafrontal ridges completely absent. Mandibles triangular; when closed, basal margin not separated from anterior clypeal region by gap. Basal margin meeting masticatory at right angle; masicatory margin with fine crenulation. Laterolypeal teeth small, blunt, and projecting forwards. Antennae 12-segmented. Palp formula 2,2 (after Xu 2000).

Mesosoma moderately stout, rectangular in dorsal view; dorsal surface flattened, bordered at lateral sides by distinct angle but not marginate. Openings of propodeal spiracles irregularly circular, directed sideways. Declivous face of propodeum immarginate above propodeal lobes. Propodeal lobes well developed, broadly rounded. Front femur moderately short and broad, laterally compressed.

Petiole wider than long, with well developed dorsal and posterior faces. Subpetiolar process relatively narrow and short, simple with ventral margin straight, evenly sloping towards posterior end; semi-translucent narrowing present along posterior two thirds of ventral margin.

Abdominal tergite III wide relative to following segment, in side view the whole segment is smaller than following, but with developed anterior, perpendicular face (Fig. 10).

Pygidial field small, with five to nine modified, peg-like setae on each side, arranged in one or two rows. Number and arrangement of setae varying with worker size.

Hypopygidium unarmed.

Mandibles densely sculptured with large, deep punctures and interspaces smooth and shining. Head with large but shallow, regular punctures, spaced from about half to more than once their diameter. Similar sculpture on dorsal surface of mesosoma, with punctures more shallow. All interspaces smooth and shining. Lateral sides of promesonotum with small punctures in upper part and extremely finely microreticulate, appearing matt; remaining mesosoma and sides of petiole similarly microreticulate.

Body pilosity composed of (1) dense, decumbent or subdecumbent hairs present on head, mesosoma, and abdominal segments and (2) moderately abundant, twice to more than three times longer than preceding, mostly suberect hairs present on head, mesosomal dorsum, petiole and posterior margins of gastral segments. Outer surface of middle tibiae without modified setae.

Color: body unicolored, yellowish.

Gyne and male unknown.

Diagnosis and discussion. This species can be distinguished from C. eguchii by differences in color, sculpture and shape of subpetiolar process. See diagnosis and discussion under eguchii for more details.

Xu (2000) reports that this species constructs nests in soil with colonies ranging from 20 to 385 individuals, foraging probably in soil and under leaf litter. The habitats where this species has been found include seasonal rainforest, mountain rainforest, deciduous monsoon forest, and warm deciduous broad-leaved forest, ranging from 730 to 1280 m in elevation.

In addition to material examined by the author from Mengla county, Xu (2000) reports this species from two other localities in Yunnan: Menghai and Jinghong counties. It is unknown whether specimens mentioned from Thailand (Jaitrong & Nabhitabhata 2005) represent this species.

Material examined. Paratypes. 2 workers, CHINA: Yunnan province, Mengla county, Bubang village, 730 m, A97-2064, 17 VIII 1997 (G. Zeng) [ MCZC]

General discussion

The original description of Yunodorylus sexspinus by Xu (2000) placed the newly established genus in Dorylinae on account of general habitus, single segmented waist, and polymorphic worker caste. However, due to certain characters found also in Cerapachyinae , namely triangular mandibles, modified peg-like setae on pygidium, and absence of promesonotal suture, Xu concluded that the new genus may represent some evolutionary link between cerapachyines and dorylines. He also provided a key to the subfamily Dorylinae , incorporating and differentiating Yunodorylus and Dorylus . In 2003, Bolton synonymized Yunodorylus with Cerapachys , based on the fact that the genus diagnosis provided by Xu included defining characters of Cerapachyinae but did not include any characters that could account for placement in the Dorylinae : Yunodorylus lacks a defined promesonotal suture, large propodeal spiracle situated high and far forward, mesosomal endophragmal pit, and bidentate pygidium characteristic for Dorylinae and possesses an externally visible sting, which is reduced and non-functional in Dorylus . The only putative character shared by Yunodorylus and Dorylus and not found in cerapachyines then, would be the absence of propodeal lobes, which was stated in the original description. However, Bolton (2003) was able to examine two related species from Borneo (described here as paradoxus and doryloides ), concluding that " Yunodorylus " species, contrary to Xu's account, do possess propodeal lobes. Examination of sexspinus and other material in the course of this study confirms that all the species have propodeal lobes developed. Moreover, Bolton (2003) pointed out that he had already shown (Bolton 1990) morphoclinal reduction in constriction between abdominal segments III and IV among Cerapachys , leading to the condition observed in the sexspinus-group. Thus, he concluded (2003), there is nothing differentiating " Yunodorylus " from Cerapachys as this genus is currently understood. In addition to characters discussed above, Brady & Ward (2005) investigated the character of mesosomal endophragmal pit, universal in dorylines and absent in cerapachyines. These authors also recognized a thin comb of the metatibial spur as a synapomorphy of the clade of the true army ants ( Aenictinae , Aenictogitoninae , Dorylinae , Ecitoninae ) as opposed to the broadly pectinate comb observed in other dorylomorphs, incuding Cerapachyinae . Species of the sexspinus-group invariably have the metatibial spurs broadly pectinate, thus providing another character invalidating placement within Dorylinae .

Recently, two important phylogenies of the Formicidae derived from molecular data have been published (Moreau et al. 2006, Brady et al. 2006). Both included material of Cerapachys sexspinus in the analysis, concluding with different results. In Moreau et al. (2006) this species appears on the tree as Yunodorylus and constitutes the sister branch to Sphinctomyrmex species from Australia. In analysis by Brady et al. (2006) Cerapachys sexspinus appears sister to the clade of Aenictinae + Ecitoninae . However, in both studies these nodes were relatively weakly supported. On account of this fact and the strong morphological evidence discussed above, it seems highly improbable that this group is more closely related to the true army ants than to other cerapachyines. However, it is worth bearing in mind that the internal phylogeny of the Cerapachyinae is unresolved and recent studies suggest paraphyly of this subfamily (Brady & Ward 2005, Moreau et al. 2006, Brady et al. 2006).

Species in the Cerapachys sexspinus-group possess a unique character combination that could easily account for placement in a genus of their own, as compared with generic concepts in other ant groups. Although these species are very distinctive within Cerapachys , most notably by having a single segmented waist and unique spur formula, it seems premature to resurrect the genus name Yunodorylus . The genus Cerapachys , as presently understood, encompasses an enormous diversity of morphological traits (Brown 1975, Bolton 1990, Bolton 2003). For example, species related to C. crawleyi Wheeler , 1924 have quite exceptional morphology and at times have been placed in a genus of their own, Chrysapace Crawley, 1924. They also possess unique palp (5,3) and spur (2p,2p) formulas, and on this ground would deserve reviving their generic status. Therefore, until broader revision of genera and their status in the subfamily is attempted, rank change of the species-group discussed above would be an act of inconsequence. Full generic revision of the subfamily is beyond the scope of this study, and it is recommended that any changes at the generic level should be postponed until the internal phylogeny of the Cerapachyinae is better understood.