The ant genus Polyrhachis F. Smith in sub-Saharan Africa, with descriptions of ten new species. (Hymenoptera: Formicidae) Author Rigato, Fabrizio text Zootaxa 2016 4088 1 journal volume 10.11646/zootaxa.4088.1.1 4019c4cc-9ef3-402e-888b-f258264b3617 1175-5326 261854 7F80636F-C96A-40B8-9DC6-BD341EF0D5AE Polyrhachis omissa n. sp. ( Figures 11 a–c) Diagnosis. A stout species in the viscosa -group, with strongly reduced pilosity, opaque integument, propodeal dorsum and declivity separated by a medially protruding transverse ridge, and petiole armed with a lateral pair of spines and a dorsal pair of teeth. Very similar to P. v i s c o s a , but with ordinary shaped scape and first funicular joint. Holotype worker . HL 1.62, HW 1.49, CI 92, SL 1.80, SI 121, FW 0.41, FI 28 PW 1.33, WL 2.10, HTL 1.70. Clypeus ecarinate, its anterior margin evenly convex and medially crenulate. Head widely oval, distinctly wider around the level of the eyes and much narrower at the level of mandibular insertions. Frons relatively narrow, frontal carinae sinuous. Antennae moderately long. Eyes large and flat, placed close to the posterior corners of the head. Mesosoma stout, nearly flat in profile, mesonotum more than twice as wide as long. Promesonotal suture narrow, but well marked, metanotal suture faint, hardly visible. Pronotal teeth well developed and slightly diverging. Propodeal dorsum bearing an upturned small tooth at each posterior corner; propodeal dorsum and declivity separated by a thin ridge strongly medially raised as an antero-posteriorly flattened lobe. Petiolar scale wide, armed with 4 equidistant spines and teeth: a lateral pair of spines and a dorsal pair of sharp teeth; the space between dorsal teeth straight. First gastral tergite anteriorly concave. Integument matt; ground sculpture finely reticulate-punctate all over the body and more superficially so on appendages. A superimposed, irregular reticulate rugulosity covers most of head and mesosoma in a somewhat areolate pattern. Standing hairs almost lacking: occurring only at the anterior clypeal margin and on gastral tergites III–V and all sternites. Pubescence very short and sparse, hardly visible on the body. Colour black throughout. Paratype gynes (n=12). HL 1.76–1.99, HW 1.49–1.72, CI 82–88, SL 1.81–2.13, SI 116–128, FW 0.42–0.50, FI 28–30, ScW 1.73–2.05, MnL 2.13–2.63, WL 2.93–3.50, HTL 1.85–2.21, Anterior wing length 8.1-9.3. With the usual caste differences from the worker and with weakly convex eyes. Wings moderately infuscated. FIGURE 11. Polyrhachis omissa , holotype worker: A, body profile; B, dorsum of mesosoma; C, head in full-face view. (Photos by M. Zilioli) Paratype male . HL 1.24, HW 1.10, CI 89, SL 1.40, SI 127, ScW 1.53, MnL 2.16, WL 2.87, HTL 1.95. (I confidently assign to this taxon a single male collected together with several gynes in Yemen). Mandibles narrow, almost unarmed, with a short masticatory margin bearing a blunt apical cleft tooth and 1 or 2 minute blunt denticles. Anterior clypeal margin entire, evenly convex. Head round with relatively small eyes (maximum eye length: 0.47), Ocelli well developed: MOD (mid-ocellus diameter): 0.18; distance between mid ocellus and each lateral one <MOD. Distance between lateral ocelli: 0.46. Length of anterior wing: approx. 7.4. Petiolar scale thick, wide and low; in frontal view the petiolar dorsum is evenly weakly convex and is separated from sides by a weak blunt angle. Integument subopaque, finely reticulate-punctate; only the head dorsum bears a trace of rugulo-reticulation recalling that of female castes. Pilosity reduced, mostly as in female castes. Body and antennae blackish; legs dark brown, mandibles and most articulations slightly paler. Wings as in gyne. Paratype workers (n=10). HL 1.56–1.80, HW 1.39–1.64, CI 88–92, SL 1.68–2.00, SI 118–127, FW 0.35–0.44, FI 25–28, PW 1.20–1.56, WL 1.90–2.32, HTL 1.56–1.98. Mostly consistent with the holotype, but with some minute variations. Eyes more or less slightly convex, metanotal suture sometimes completely lacking dorsally, and median lobe of propodeum reduced to a low convexity. Holotype worker. SOMALIA : Balad, 28.ix.1986 ( L. Bartolozzi ) (MSNM). Paratypes. YEMEN : Sokna (Tihama), m 200, 20.viii.1965 ( G. Scortecci ) (17 g, 1 m, MSNM). ETHIOPIA : “da Dimé al Bass Narok”, viii–ix.1896 ( Bottego ) (1 g, MSNG) [misidentified by Emery (1899) as P. viscosa ]; Banno, Sagan-Omo, 10.v.1939 ( E. Zavattari ) (1 w, MSNG); Caschei, Sagan-Omo, 6.vii.1939 ( E. Zavattari ) (3 g, MSNG). SOMALIA : “M. Umberto I”, iii.1892–93 [?] ( E. Ruspoli ) (1 g, MSNG); Ganana, iii.1892–93 [?] ( E. Ruspoli ) (2 g, MSNG); “Boran Galla, Medio Ganale”, vi.1893 ( V. Bottego) (1 w, MSNG); “Basso Ganana”, vii– viii.1893 ( V. Bottego) (3 w, MSNG); Eil (Nogal), iii–iv.1938 ( S. Venzo ) (2 g, MSNM); Eil ( A. Falzoni ) (1 g, MSNM); Gardo, 810m, 21.x.1957 ( G. Scortecci ) (2 g, MSNM); same data, but 22.x.1957 (2 g, MSNM); Afgoi, v.1972 ( L. Masutti ) (2 g, MSNM); Afgoi, 2.x.1986 ( L. Bartolozzi ) (1 w, MSNM). KENYA : Mackinnon (30 km ca. N-NW of Mombasa), ix.1946 ( Meneghetti ) (1 w, 1 g, MSNM); Malindi, xii.1993 ( R. Regalin ) (3 w, MSNM); Archer’s Post, Uaso Nyiro river, 2300’, 6.xii.1969 ( M.E. Irwin & E.S. Ross ) (4 g, CAS). Comment. I often found specimens of P. o m i s s a labelled as P. v i s c o s a and mixed with it, but omissa always lacks the distinctive antennal features that separate viscosa from all other African Polyrhachis known so far. Polyrhachis viscosa has a strongly widened apex of the scape and a strongly depressed first funicular joint (see Fig. 22 in Bolton, 1973). This feature is unique to viscosa and seemingly constant. Although I found several gynes assignable to P. omissa collected in Yemen, Collingwood and Agosti (1996) in their survey of Arabian ants mentioned P. v i s c o s a and P. lacteipennis F. Smith only. Polyrhachis lacteipennis superficially recalls viscosa and omissa , but belongs to the non-African subgenus Myrmhopla and has a slender and immarginate mesosoma and other very distinctive features. I also examined the types of P. an t i n or i i Emery, which Bolton (l.c.) did not see and synonymised with P. viscosa . I can confirm that antinorii is conspecific with viscosa ; therefore no old available names are assignable to omissa . The main features separating omissa from viscosa workers can be summarized as follows:

Polyrhachis omissa	Polyrhachis viscosa
Scape and first funicular joint of usual shape	Scape strongly widened apically and first funicular joint strongly flattened proximally
Head apperaring almost round, CI 88 or more	Head appearing somewhat elongate, CI 86 or less
Anterior clypeal margin entire, medially crenulate and at most faintly notched	Anterior clypeal margin usually with a small, but distinct, median notch
Frons narrower, FI 28 or less	Frons wider, FI 30 or more
Mesosoma stouter: pronotum about twice as wide as propodeum; mesonotum in dorsal view about 2.5 times as wide as long	Mesosoma more slender: pronotum about 1.5 times as wide as propodeum; mesonotum in dorsal view about twice as wide as long

In addition, most omissa specimens have a few standing setae on the second gastral tergite, while all of the approximetely 20 viscosa specimens I examined have no erect hairs on the second gastral tergite; but this difference might be due to population differences, age or degree of abrasion of specimens. Among the viscosa -group species listed by Bolton only viscosa itself and nigrita share the same arrangement of petiolar spines as in omissa : a long lateral pair, and a much shorter, often tooth-like, dorsal pair. Moreover, nigrita differs from viscosa and omissa especially by its longer propodeal teeth and by the propodeal dorsum evenly rounding into the declivity.