Body medium-sized (ca 20–30 mm long, ca 2.0–3.2 mm wide), with 20 segments. Paraterga moderately to strongly developed. Sterna unmodified or with a cone near each coxa. Sternal lobe or cone(s) between ♂ coxae 4 present. First pair of ♂ legs with or without femoral adenostyles. At least some male legs with ventral brushes on tarsi, sometimes also on distal halves of tibiae.

Gonopod with a quite short, subcylindrical, distoventrally densely setose coxite; prefemoral (= setose) part of telopodite short to very short, a quarter or less as long as acropodite, delimited from extremely short femorite by a mesal sulcus, ridge or spine; femorite supporting a long, spiniform, sometimes deeply bipartite femoral process (fp) and a similarly long to longer, flagelliform solenomere (sl), both more or less clearly sheathed by a mesal fold in basal half of acropodite. The latter suberect to strongly unciform, distal quarter to third with or without evident processes, clearly fimbriate and/or fringed, often phylloid as well.

Tetracentrosternus subspinosus Pocock, 1895; by monotypy.

Other species include Tetracentrosternus hoffmani Golovatch, 2013 and Tetracentrosternus theelorsuensis sp. n.

Remarks. Both Tetracentrosternus Pocock, 1895 and Tetracentrosternus subspinosus Pocock, 1895 were originally diagnosed and described too briefly to be readily recognizable. The species was based on a few specimens taken by L. Fea and E. W. Oates at Puepoli (900–1, 200 m a.s.l.) and Bia-po, now Carin Cheba (1, 000–1, 200 m), both in Myanmar (Pocock 1895, Thorell 1895). These two localities lie in the Karin Hills of southeastern Myanmar, in the Tenasserim Mountain Range adjacent to Thailand (Fig. 4).

The sole male of the species, from Puepoli and kept in the British Museum, was redescribed by Hoffman (1963) who designated it as lectotype. Hoffman was the first to provide clear illustrations of its gonopod structure, resulting in the establishment of a new tribe, Alogolykini, to encompass Tetracentrosternus, Alogolykus Attems, 1936 and Touranella Attems, 1937 (Attems 1936, 1937, 1938). Jeekel (1965) redescribed this species in more detail not only from the lectotype, but also from the paralectotypes (from Carin Cheba) still housed in the Genoa Museum. He made new, even more accurate illustrations and a highly detailed redescription, fully accepting the tribe and adding thereto also Yuennannina Attems, 1936. A little later, Jeekel (1968) elevated the tribe Alogolykini to the status of a subfamily, Alogolykinae, adding thereto a new tribe, Polydrepanini. Jeekel transferred Touranella from Alogolykini to Polydrepanini, because the genus lacks adenostyles on the femora of the male first legs, as do the other constituent genera of the latter tribe.

Tetracentrosternus hoffmani Golovatch, 2013 stems from Mount Gaoligong Shan, western Yunnan, China (Golovatch 2013). Like Tetracentrosternus subspinosus, it shows adenostyles on male femora 1, a suberect gonopod telopodite with a fringed/fimbriate/spiculate and phylloid apex, and a long, unipartite gonofemoral process.

The new species described below, despite its relative geographical proximity to Tetracentrosternus subspinosus, shows a number of characters so different that the diagnosis of the genus needs to be refined. Thus, because there are no adenostyles in femora 1 of the male of Tetracentrosternus theelorsuensis sp. n., this trait can be regarded as only species-specific not only in Tetracentrosternus, but in the entire subfamily Alogolykinae. So, following Hoffman (1963), it seems best to return Touranella (three species in Vietnam, one in Nepal) to Alogolykini, as its close similarities to Tetracentrosternus are apparent. The only meaningful differences lie in gonopod structure, the femoral portion in Touranella sometimes being a little longer, the solenomere a little thicker, suberect and rod-shaped rather than flagelliform, while the femoral process when present is considerably shorter (Golovatch 2009a, 2009b). In our opinion, a strong, rod-shaped solenomere versus a thin, flagelliform one remains the basic difference between the tribes Alogolykini and Polydrepanini, respectively. Thereby the subfamily Alogolykinae seems to be best characterized only by the absence of a clear-cut division of the solenophore or solenophore-like structure near/around the solenomere into a membranous lamina medialis and/or a similarly membranous lamina lateralis, in Polydrepanini often coupled with a twisted, helicoid course of the seminal groove. In addition, the gonofemorite in Alogolykinae is often strongly abbreviated while many species show adenostyles on the male first femora. The latter two traits are also characteristic of the subfamily Australiosomatinae, but the solenophore branch or branches remain free and never sheath a primitively long, strong and rod-like solenomere. A more detailed review of these subfamilies and their tribes (see Jeekel 1968) lies far beyond the scope of the present note.

The gonopod telopodite in Tetracentrosternus theelorsuensis sp. n. is strongly elongate and unciform, bearing three evident processes in the distal half, whereas the solenomere is particularly long and nearly as long as the solenophore, while the femoral process is also very long, but deeply bipartite. Besides this, as in Tetracentrosternus subspinosus, the strongly abbreviated gonofemoral part in the new species is delimited distally by a distinct ridge, as opposed to a strong spine in Tetracentrosternus hoffmani.

In terms of metatergal structure, Tetracentrosternus theelorsuensis sp. n. is somewhat intermediate between both congeners, the tegument being only moderately rugulose as opposed to nearly smooth in Tetracentrosternus subspinosus or rather clearly tuberculate in the rear halves of metaterga in Tetracentrosternus hoffmani.

There is little doubt that more species of Tetracentrosternus await discovery at least in and between eastern Myanmar and southern China, including Thailand (Fig. 4). The same certainly holds true for some other Alogolykini as well, e.g. Touranella.

♂ (CUMZ), Thailand, Tak Province, Umphang District, Thee Lor Sue Waterfall, 590 m a.s.l., 15°55'38"N, 98°45'13"E (converted from GPS data), 19.01.2011, leg. N. Likhitrakarn.

♂ (CUMZ), same District, Mokro Subdistrict, roadside, 1, 168 m a.s.l., 16°14'14"N, 98°59'23"E, 20.01.2011, leg. N. Likhitrakarn. 5 ♀, 1 juv. (CUMZ), same District, Pa Wai Waterfall, 804 m a.s.l., 16°34'30"N, 98°50'3"E, 20.01.2011, leg. S. Panha, C. Sutcharit & N. Likhitrakarn.

After Thee Lor Sue Waterfall, the type locality, which is the largest and highest waterfall in Thailand.

Differs from congeners mainly by the first pair of ♂ legs lacking femoral adenostyles, coupled with unmodified sterna and the distal half of the gonopod telopodite being strongly curved, elongate and distally carrying three evident processes. See also Remarks above and Key below.

Length 21–23.5 (♂) or 19.5–22.5 mm (♀), width of midbody pro- and metazona 1.92–1.98 and 2.54–2.75 mm (♂) or 2.07-2.35 and 2.77–3.12 mm (♀), respectively.

Coloration of live animals dark brown (Fig. 1A); paraterga, legs and epiproct light brown, head, collum and metazona 2–4 blackish, following terga with a blackish triangle covering both pro- and metazona; coloration of alcohol material after 2 years of preservation faded to whitish or yellow with a pattern of a contrasting dark brown triangle at posterior edge of pro- and metazona, castaneous brown below paraterga; head and antennomeres 6 and 7 brown to castaneous brown; venter and a few basal podomeres light brown to yellow-brown, legs growing infuscate (brown) distally; tip of antenna pallid (Fig. 1A–H).

Clypeolabral region and vertex sparsely setose, epicranial suture distinct. Antennae moderately long (Fig. 1A, B), reaching body segment 4 (♂) or 3 (♀) when stretched dorsally. In width, head < collum < segment 3 = 4 < 2 < 5–16 (♂, ♀), thereafter body gently and gradually tapering. Collum with three transverse rows of setae: 5+5 in anterior, 3+3 in intermediate and 4+4 in posterior row; a very faint incision laterally in posterior 1/3; caudal corner of paraterga rounded, slightly declined ventrad, not extending behind rear tergal margin (Fig. 1B, D). Tegument smooth and shining, prozona very finely shagreened, metaterga smooth and delicately rugulose, leathery; surface below paraterga finely microgranulate. Postcollum metaterga with two transverse rows of setae: 2+2 in anterior and 3+3 in posterior row, traceable at least as insertion points when setae broken off. Tergal setae long, strong and slender, about 1/3 of metatergal length. Axial line visible, barely traceable on proterga (♂). Paraterga strongly developed (Fig. 1A–H), especially so in ♂, mostly slightly upturned, lying high (at about upper third of midbody height); caudal corner nearly or fully pointed; either clearly (♂) or only very slightly extending beyond rear tergal margin (♀); paraterga very thin in lateral view, like blunt blades, a little thicker only on pore-bearing segments. Calluses on paraterga 2 delimited by a sulcus both dorsally and ventrally, on following paraterga only dorsally, rather broad on pore-bearing segments. Paraterga 2 broad, anterior edge angular, a little better so in ♀, lateral edge with two small, but evident incisions in anterior half; posterior edge clearly concave (Fig. 1B, D). Anterior edge of paraterga 3–9 clearly convex, of paraterga 10–18 nearly straight and slightly bordered. Lateral edge of paraterga with a slight, but evident incision in anterior 1/3. Posterior edge of paraterga clearly concave, especially strongly so in segments 16–19 (Fig. 1A–D). Ozopores evident, lateral, lying in an ovoid groove at about 1/4 in front of caudal corner. Transverse sulcus usually distinct (Fig. 1C, D, F), slightly incomplete on segments 4, 18 and 19, complete on metaterga 5–17, deep, not reaching bases of paraterga, at most faintly beaded at bottom, a little better developed in ♀. Stricture between pro- and metazona wide, evidently ribbed at bottom down to base of paraterga (Fig. 1B–D, F–H). Pleurosternal carinae complete crests with a sharp caudal tooth in segment 2, a small, caudal, mostly sharp tooth until segment 7 (♂) or 6 (♀) (Fig. 1B, G, H). Epiproct (Fig. 1F, H) conical, flattened dorsoventrally, with two small apical papillae; tip subtruncate; pre-apical papillae small, but visible, lying not too close to tip. Hypoproct roundly subtriangular, setiferous knobs at caudal edge well-separated and evident.

Sterna moderately setose, without modifications; a linguiform, sternal lobe between ♂ coxae 4 (Fig. 1I, J). Adenostyles absent on femur 1 (Fig. 3A). A paramedian pair of evident tubercles in front of gonopod aperture. Legs moderately long and slender, slightly incrassate in ♂, midbody ones ca 1.2–1.3 (♂) or 0.8–1.1 times (♀) as long as midbody height, prefemora without modifications, ♂ tarsal brushes present in all legs.

Gonopods (Figs 2, 3B, C) complex. Coxa a little curved, short, subcylindrical, rather densely setose distoventrally. Prefemur densely setose, length less than 1/4 of femorite + “postfemoral” part. Femorite as usual, very short, delimited distally by a ridge (R). Femoral process (fp) a slender, distally deeply bipartite flagellum, with branch fp2 somewhat longer than fp1. Acropodite (= solenophore) elongate, clearly unciform, in proximal 1/3 with a small, but distinct, mesal, apically denticulate process (Z), in distal 1/4 with a similar, but lateral and more strongly dentate process (Y), as well as a spine (X) at its base. Solenomere (sl) branching off level to fp, very long, flagelliform, mostly sheathed by a slightly longer solenophore.