Coremiocnemis valida Pocock 1895

West, Rick C. & Nunn, Steven C., 2010, 2443, Zootaxa 2443, pp. 1-64 : 56-63

Coremiocnemis valida Pocock 1895

( Figs 192–217)

Coremiocnemis validus Pocock 1895, p. 175 , pl. 10, fig. 5

Type: female holotype, BM 1894.60.15, Entoyat River (4°12' N, 114°26' E), Baram District, Sarawak, East Malaysia, NHM –images (R. Raven, pers. comm.)–examined .

N.B.: Although the type was recorded from “East Indies” by Pocock, locality data in type jar is Entoyat River (4° 12' N, 114° 26' E), Baram District , Sarawak, East Malaysia (see Remarks below) .

Other material: Coremiocnemis valida Pocock 1895 , female non– type, ZRC ARA.751, Baram , Sarawak, East Malaysia, col. Sept. 1970 (unknown col.), RMBR –images (D. Court, pers. comm.)–examined .

Etymology: A Latin adjective actually meaning 'strong, efficacious, able' but Dr H. D. Cameron (pers. comm.) believes Pocock intended it to have the conventional taxonomical meaning 'valid,' as in a valid species or a vaild name. Pocock’s description seems to reflect some uneasiness whether his new species is genuine. He says, " Apparently resembling the type C. cunicularius [sic] ... but differing in many ... characters. (p. 175)" The species name reflects his conclusion that his species is indeed valid.

Diagnosis: Differs from all other Coremiocnemis species by the black endocuticular studs, between the posterior coxae and sternal borders and spermathecae (apically swollen, heavily sclerotized with greatly elongated clear shafts).

Description: Female holotype BM 1894.60.15 ( Fig. 196) with body length: 46.

Color (in life, Fig. 192): anteriorly pale brown, posteriorly same with slight purple sheen, slightly darker abdomen. All setation bright rust red.

Carapace: length 20, width 17.5. Fovea procurved, deep, almost as wide as OT.

Eyes ( Fig. 197): Anterior row transverse, posterior row slightly recurved. Eyes: ALE, AME, PLE, PME.

Chelicerae: Ectal lyrate region a series of strikers (>90) in>8 horizontal rows ( Figs 213, 216). Longest strikers on lowest rows. Each striker needleform, with some filiform ends. Teeth,15,>90 basomesals. Intercheliceral pegs>5 larger (1 very large),>20 smaller in tight cluster on basodorsal surface ( Figs 212, 214). Retrolaterally, basomedially with small cluster of short strong spines (>7), point distad ( Fig. 215).

Maxillae ( Fig. 207): prolaterally plano-convex, anterior lobe well pronounced, many cuspules (> 160) on inner basoventral surface. Lyra ( Figs 207, 210): many bacilliform rods (>160) in dense, ovoid patch on prolateral face, lowest row with>16 bacillae, longest rods medially in lowest rows. Rods paddle-shaped, medium to long shafts, largest with distal black tips (not blades). At widest point, lyrate patch>10 rows deep, smallest rods dorsally. Posterior end of patch truncate, anterior end slightly pointed distad/rounded. Dorsal to suture>35 small elongated spines on anterior margin in unorganised rows.

Labium ( Fig. 198): length 3.15, width 4.27. Many small cuspules (>550) along anterior 1/3 surface.

Sternum ( Figs 199, 202): length 9.35, width 8.94. Slightly wider posteriorly than anteriorly, with thin even cover of short setae, interspersed with longer darker setae. Spines lacking along sternal border. Posteriorly between left–right cox. IV, sternal border slightly acuminate, lateral sternal points slightly acuminate. Between posterior sternal borders and cox. III and IV, many dark black studs ( Figs 199, 202–203) on endocuticle. Sternal sigilla: 3 pairs (not including labio-sternal sigilla), posterior medium size, ovoid morphology, 2.5 lengths apart, 1.0–1.5 of their length from sternal margin. Median pair 1/3 size of posterior, similar form, 1.3– 1.8 of their length from sternal margin. Anterior pair somewhat obscured, border sternal margin. Labio-sternal sigilla larger than PSS.

Leg setation: All segments distally with band of short blunt setae. Leg IV with distinctly longer setae on all segments, dense on tib and met. IV giving “bottlebrush” appearance ( Figs 192, 209). Retrolaterally on met. and tar. IV, longer setae show distinct recurvature.

Legs ( Figs 204–206, 209): formula; length IV, I, II, III; width III, IV, I, II. Leg RF ~80.88. Leg lengths: palp: 35. I: 55. II: 48. III: 46.5. IV: 68. Tar. IV transversely integral, shows no weakening medially, but with pallid transverse region.

Scopulae: Met. and tar. I–III undivided, met. and tar. IV divided. Met. I, II: entire, III: 4/5, IV: 1/2 prolateral side of division, entire retrolaterally ( Figs 205, 206).

Coxae: some small black thorns prolatero-dorsally, no thorns retrolaterally on I–IV. Coxae easily seen dorsally. I longest, about 1.2 times length of II. IV widest, just longer than III, basally rectangular with defined corners. Small ventral thorns prolaterally on III–IV (small setae only on I–II). I–III ventrally with many long thick blunt setae proximally, pallid. No short black setae. IV with mixture of long thick blunt setae entirely with mix of longer setae most concentrated along retrolateral ledge, pallid intermixed with shorter thin pallid setae. Ventral I–IV gently sloping anteriorly.

Trichobothria: on all tarsi basal filiform field slightly wider than clavate field, merges evenly. Clavates on tar. I in distal 2/3, long filiforms only in basal 1/2, shorter filiforms intermixed with clavates distally. Clavate extent on II–IV cf. I, in distal 2/3. Shorter filiforms for length. Short epitrichobothrial field on I shorter than clavates, increasing in length proximally.

Tarsal organ not evident on legs I–IV under stereoscope.

Spines: met. I with 1 DV, 1 DPV, met. II with 1 DV, 1 DPV, 1 DRV, met. III with 1 DV, 1 DPV, 1 DRV, 1 DD and met. IV with 2 DV, 1 DPV, 1 DRV, 1 DD, 1 DPD.

Claws: paired claws on all legs and palpal claw unarmed. Reduced third claw on leg IV.

Abdomen: 25 long. Ovular, elongated, yellow brown (in alcohol), covered with mat of fine short dark setae with sparse cover of longer pale yellowish setae. Longer hairs concentrate ventrally toward spinnerets, point distad. Strongly darker anterior to booklung–epigynum areas.

Genitalia [ Figs 200 (holotype), 201 (nontype)]: Spermathecae paired, not fused, bilobular basally, each lobe dividing at 0.2 of total length proximally, with strong apical rounded appendages (apically swollen), heavily sclerotized compared to clear spermathecal shafts (same morphology in both specimens examined).

Distribution and natural history ( Figs 194, 195, 217): Known from the Baram and Entoyat River, Baram District, Sarawak, East Malaysia, north-eastward to the Liang River, Belait District, Brunei Darussalam, of the northern coast on the island of Borneo. C. valida is an opportunistic fossorial theraphosid. Adults and subadults excavate their own burrow in sloped ground which consisted mainly of a silken collar that incorporates surrounding leaf litter and other organic debris ( Fig. 195). The soil is moist and a mix of sand and clay. Depending on the size of the specimen, burrows ranged between 25–35 cm deep. Burrows did not terminate in a flask-like chamber but, instead, into a maze of root fibers from the surrounding vegetation. Excavating live specimens proved difficult with specimens disappearing into the root maze at the end of their retreats. C. valida burrows were all found in lowland moist dipterocarp forest ( Fig. 194) at 80–120 meters of elevation (T. C. Chen, pers. obs.). Nothing else is known about the natural history of C. valida . A mature male was recently located and photographed by T. C. Chen ( Fig. 193). Unfortunately the specimen was not located in time to fully describe within this revision; however, it will be described in a future paper.

Remarks: In late 2008, J. Beccaloni (NHM) loaned Pocock’s female holotype C. validus specimen, as well as two C. validus (Abraham 1924) non–type specimens (mature male and female from the Ridley Collection, held largely at the RMBR), to the QM to be examined for this revision. Dr Raven examined the holotype and subsequently discovered two major factors surrounding the previously dubious position of this species.

First, Raven, with assistance from R. West with supporting evidence, was able to positively confirm the holotype original collection label. The label is handwritten in pencil and is faded, but clearly reads: “Entoyat River, Nov/94 Baram, Hose Cole”. Further research through the NHM revealed the location of the Entoyat River , Baram District , Sarawak, East Malaysia and the collector to be Mr Charles Hose. Raven also identified specific diagnostic traits (heavily sclerotized and studded endocuticular region between proximal surfaces of coxae III and IV, and the sternum) eliminating this species from all those previously known to occur in West Malaysia (see Fig. 203). Immediately we began checking material lists of the RMBR collection provided by D. Court and located a mature female collected at Baram ( ZRC ARA.751) that, after detailed examination was found to be conspecific to Pocock’s holotype and the only other known example to date. Firstly , this supports that new diagnostic traits are not teratological or pathenogenetic (i.e., a developmental or unusual growth). Secondly , the precise type locality as per the oldest label is correct (despite Pocock naming “ East Indies ” in the original description) .

The second key factor noted by Raven was a prior unsuccessful attempt to extract the spermathecae. Raven, however, found the spermathecae, extracted, and cleared them until the morphology of the organ was evident ( Fig. 200). This is the first time spermathecae of C. valida has ever been noted (NB.: the Ridley collection of Abraham’s is not this species, but is C. cunicularia from Penang Island, Penang, West Malaysia). Support for constant spermathecal morphology was found in the non– type conspecifc ZRC ARA.751 ( Fig. 201).

Discussion: Ontogenetic considerations: During the course of examination of selenocosmiine material, two constant complex issues arose, juvenile types as singular representatives of species and notable changes of significance between penultimate and postultimate material. Much of the type material examined consisted of singular specimens, fixed at varying life stages. Many are clearly either juvenile or young sexual matures (adult males excluded) and, as such, may not display many “ultimate” character traits at species level (all arguments about species definitions aside, but purely with ontogeny in mind). Recognition of mature adults by the authors were considered as those specimens with developed sexual morphology, combined with anterior/posterior gradualism ( Raven 1985), focusing on consistency of clavate tarsal trichobothria on legs I–IV in particular. The combination was utilized because in singular specimens, the mere presence of spermathecal receptors do not unequivocally define a specimen with “ultimate” specific morphology, the spermathecae could be in early stages of development and may or may not be fully functional at that time (in the C. cunicularia nontype spermathecae, see: Fig. 17, deposited sperm can be seen distally within the receptors). The character state combination determination is not a total solution, but does provide stronger evidence of maturity of any specimen in question. Species only known from single specimens may also show traits not present in conspecifics (e.g., pathenogenic), thus displaying false diagnostic morphology that require serious examination against non-type conspecific animals if possible (see: C. valida , Remarks). Secondly, in many Coremiocnemis species , morphological ontogenetic changes from older juvenile to post/ultimate instar females are important. In C. hoggi sp. nov., two specimens (QM S88077 , QM S88079 ) are small females. Both examples possess spermathecae with shorter length lobes than in all fully mature female types and further, the longer setae on tib. and met. IV are much shorter and show a distinct recurvature, not seen 1–2 instars further in sexual maturity (RCW, pers. obs.). The total leg width formula varies significantly too, in the same smaller females the leg width formula is III, IV, I, II, however in all larger female types it is IV, III, I, II. In C. brachyramosa sp. nov., penultimate females display coloration very similar to C. hoggi sp. nov. (blue femora with distally cinnamon leg segments), yet at ultimate instar become an intense iridescent blue entirely (J.-M. Verdez, pers. comm.). In C. jeremyhuffi sp. nov., postultimate female types display a dull brown coloration compared to extremely bright orange older juvenile material, so much so that adults and older juveniles look like distinctly differing species. It appears that given the high level of speciation within the group, following “Haeckels Law” ( Gould, 1977), multiple examples in varying life stages have assisted to better map late ontogenetic variation of significance in Coremiocnemis . This work should assist to determine and identify more stable traits within each species through multiple specimen typifying (where possible) and further, aid in clearer character, and more importantly, terminal unit (TU) selection (see: Mishler 2005) for any phylogenetic data analyses involving Coremiocnemis species as TU’s.

Species groups within Coremiocnemis: Within Coremiocnemis , two groups are clearly evident, the first group (Group 1) comprises the type species, C. cunicularia and C. hoggi , C. kotacana , C. obscura and C. valida . The second group (Group 2) comprises C. brachyramosa , C. gnathospina and C. jeremyhuffi . Species of group 1 share leg morphology with leg RF ~ 78.26–83.04, distinct “bottlebrush” setae on tibiae and metatarsi IV, 3/4–entire retrolateral metatarsi IV scopulation, and apically swollen spermathecal lobes, which are distinctly heavy in sclerotization, compared to their weakly sclerotized shafts. Species of group 2 share leg RF ~ 89.03–94.81, reduced bottlebrush setae on leg IV, retrolateral metatarsi IV scopula only extending 1/3 distally and spermathecal lobes not apically swollen with less distinction in sclerotization entirely. Lyrate morphology, however, is not congruent within the above groups.

While the type species, C. cunicularia (Group 1) has bacillae on the lowest row with quite long shafts, so do C. brachyramosa (Group 2) and C. gnathospina (Group 2). However, a reduction in length of the bacillate shafts is clearly seen in C. kotacana (Group 1), C. valida (Group 1) and C. jeremyhuffi (Group 2). Both groups have the same palpal bulb morphology: “popcorn kernel”-shaped tegulum with lanceolate, terete embolus emerging anteriorly from the tegulum, with distinct distal “kiss curl” (little to no flaring distally). Both groups also share a foveal groove of less width than of the OT (except C. gnathospina ), low cephalothoracic profile, reduced third claw on leg IV, divided leg IV scopulae and bilobular spermathecae with the lateral lobes with terete shafts, similar in length and morphology to the medials (all species). Lobes also increase in sclerotization apically (except C. brachyramosa , with uniform heavy sclerotization of highly developed spemathecal lobes).

Within all Selenocosmiinae examined, the distal embolic “kiss curl” of the lanceolate and terete male embolus is not known outside the Coremiocnemis . Long retrolateral recurving setae on leg IV are also only known within this genus (most species), as are bilobular spermathecae with lateral lobes sharing similar morphology and length to that of the medials (all species, differing to the unilocular spermathecae of the type Selenoytpus plumipes Pocock 1895 , which does seem to show some artificial similarities to Coremiocnemis , particularly leg IV traits). These three character states are considered apomorphic to the group, providing strong support that all Coremiocnemis described and revised, herein, form a strong and distinct clade. The division between Groups 1 and 2 needs to be further investigated as new material from the region comes to light, however, at this time, there is simply not enough clear evidence to warrant splitting the clade and erecting a new genus. Systematic work on selenocosmiine fauna from southern Johor State, West Malaysia, and Singapore (Nunn, West and Court, in prep.) aims to resolve this question, as will a better understanding of S. javanensis , type species for Selenocosmia .

During the course of this revision, J. Beccaloni (NHM) examined the holotype of Selenocosmia tahanensis Abraham 1924 on our behalf, in order to determine, firstly, whether a new species with male types were distinct from it and, secondly, if it was better placed within either Coremiocnemis or Lyrognathus . The distal embolus tips of the holotype are not broken, but do not show the apomorphic kiss curl of the Coremiocnemis and an embolus keel is absent (of the three genera, known only in Lyrognathus ). The authors were unable to place this species into either genus, but further research is planned to resolve generic placement of this species, as it shows several character traits not typical of Selenocosmia in the region (elongated oval lyrate patch, third claw on tarsi IV, transverse suture along tarsi IV, etc).