A survey of the Porrhoclubiona Lohmander, 1944 from Central Asia (Araneae, Clubiondae)

Abstract Clubiona Latreille, 1804, with more than 500 named species, is one of the largest genera of Araneae. The genus has 15 synonyms, most of which are not listed in the World Spider Catalog (2018) and unknown to many arachnologists. The most comprehensive survey of Clubiona sensu lato by Wunderlich (2011) also lacked a few synonyms. In this paper all genus group names described in Clubiona are listed with their type species. Most of these names correspond to the species groups recognised in Clubiona sensu lato. We agree that Porrhoclubiona Lohmander, 1944 (= Clubionagenevensis-group) deserves a status of a separate genus and provide the diagnosis of this taxon. Three species of Porrhoclubiona that occur in Central Asia are surveyed, and two of them are described as new to science: P.laudata (O. Pickard-Cambridge, 1885), comb. n. (♂♀, Xinjiang, Tibet, China), P.bosmansisp. n. (♂♀, Tajikistan), and P.moradmandisp. n. (♂♀, Fars, Iran). It seems that all records of P.genevensis L. Koch, 1866 from China refer to P.laudata. The records of Clubionavegeta Simon, 1918 from Tajikistan and Iran refer to P.bosmansi sp. n. and P.moradmandi sp. n., respectively. The following new combinations have been established: Porrhoclubionadecora (Blackwall, 1859), comb. n., P.diniensis (Simon, 1878), comb. n., P.leucaspis (Simon, 1932), comb. n., P.minor (Wunderlich, 1987), comb. n., P.pseudominor (Wunderlich, 1987), comb. n., P.pteronetoides (Deeleman-Reinhold, 2001), comb. n., P.vegeta (Simon, 1918), comb. n., P.viridula (Ono, 1989), comb. n., and P.wunderlichi (Mikhailov 1992), comb. n. (all ex. Clubiona). SEM study of the structure considered earlier as scopula in Clubiona and Porrhoclubiona reveals that it is represented by several lateral rows of movable macrosetae (spines) with a locking mechanism.


Introduction
Clubiona Latreille, 1804 with more than 500 species (WSC 2018) is one of the largest genera of the order Araneae. There have been several attempts to split this genus either to genera and subgenera (Lohmander 1944;Wunderlich 2011) or to species groups (Mikhailov 1990(Mikhailov , 1992(Mikhailov , 1995(Mikhailov , 2003Deeleman-Reinhold 2001). One of the most distinct groups of the genus is the Clubiona genevensis-group (Bosmans et al. 2017) or subgenus Porrhoclubiona Lohmander, 1944belonging to Microclubiona Lohmander, 1944 Both subgenus and genus are currently considered in the genus Clubiona, although both sexes have autapomorphies lacking in other clubionids. While studying spiders described by O. Pickard-Cambridge from the Himalayas, we recognised one species of Clubiona belonging to the C. genevensis-group. While tubes with types from the Himalayas are lacking any name or geographical labels it was easy to identify these specimens as C. laudata due to the figures in Pickard-Cambridge (1885). Because this species is very similar to C. genevensis, we decided to compare it to available specimens. Comparison of this species with specimens identified as C. genevensis from Tajikistan and southern Iran revealed differences between them as well as with syntypes of C. laudata. The goals of this paper are 1) to provide the first redescription of C. laudata, 2) the description of two new species, 3) revalidation, re-diagnosis, and re-delimitation of Porrhoclubiona.

Materials and methods
Specimens were photographed with a Canon EOS 7D camera attached to an Olympus SZX16 stereomicroscope and with a SEM JEOL JSM-5200 scanning microscope at the Zoological Museum, University of Turku, Finland, and digital images were montaged using "Zerene Stacker" image stacking software. Epigynes were cleared in a 10% KOH/ water solution until soft tissues were dissolved. Photographs were taken in dishes with cotton on the bottom to hold the specimens in an appropriate position. All measurements are given in millimetres. Abbreviations used are as follows: Comments. While trying to rediagnose Porrhoclubiona we noticed a peculiar modification of leg I and II in females: they have a kind of scopula (Figs 2d,3a). A similar modification was documented for Clubiona comta C.L. Koch, 1839 (= Hyloclubiona c.) by Marusik and Kunt (2010). We thought that it was a diagnostic character for two related genera, but checking Clubiona pallidula, the generotype (Fig. 2e) and some other species revealed that this character is present in many species of Clubiona s. l. Locket & Millidge (1951: 125) mentioned scopulae on legs I and II present in all British Clubiona, that it was well developed only in females and can be reduced to a single row in a smaller species. Deeleman-Reinhold (2001) also reported the presence of scopulae in the Clubiona pteronetoides-group without specifying in which sexes.
Light microscopy (Fig. 2d, e) indicated that the modified setae cannot be considered as scopula. They are absent ventrally on the tarsus-tibia but located ventrolaterally and additionally are adpressed and not erect. SEM microscopy reveals that that "setae" in "scopula" are movable spines and have locking mechanisms (Lm), at least on the metatarsi and tibiae (Fig. 3e). A locking spine mechanism is known in several unrelated groups of spiders like Oonopidae, Corinnidae, Phrurolithidae, etc. (cf. Marusik et al. 2013: figs 7-10). In that groups, the locking mechanism is present in both sexes and such as ventral paired spines only. These spines are long, and when erect act as a catching basket for prey capture. The function of such spines arranged in 3-4 rows on each side of the segment is unclear.

Porrhoclubiona Lohmander, 1944
Porrhoclubiona Lohmander, 1944: 20  Note. Above we listed only two of the most recent publications dealing with this species group.
Comments. Aside from those mentioned in the diagnosis, characters that separate Porrhoclubiona from all other genera previously considered in Clubiona, such as the presence of a patch/brush of modified setae on the cymbium, a cymbial extension that can be considered a tutaculum (Tu) and a tegular groove (Tg) serving as a conductor, a simple retrolateral tibial apophysis, and the presence of a prolateral apophysis, a few more characters should be mentioned. The two genera differ by spination of leg I: Porrhoclubiona is lacking metatarsal spines which are present in Clubiona and have fewer ventral tibial spines (cf. Fig. 2d and Fig. 2e). Porrhoclubiona has better developed "scopula", which stretch along the entirety of tibia I, while in Clubiona it occupies only the distal ½ of the tibia (cf. Fig. 2d and Fig. 2e).
Although the retrolateral tibial apophysis of the male palp looks simple, from SEM figures it is rather broad (Fig. 5e) and the tip has a kind of filamentous extension (Fl). This tip can be long, like in P. vegeta (Bosmans et al. 2017: fig. 65) or P. moradmandi sp. n. (Fig. 5e), or rather short like in P. bosmansi sp. n. or P. genevensis (Figs 4c,5f). Although the base of the embolus looks like one sclerite, in fact it is composed of two sclerites (Figs 4b, e, h, 9c, d, 10b'), heavily sclerotised part of the tegulum (Ts) and the base of the embolus (Eb).
Some species can be separated based on the proportions of the cymbial setae. Porrhoclubiona lecucaspis has distinctly longer basal part of the setae (Sb) than P. moradmandi sp. n. and P. bosmansi sp. n. (cf. Fig. 5c and Fig. 5a, b, respectively).
The haematodocha in Porrhoclubiona is rather large, but the subtegulum is strongly reduced and located posterior to the embolus base (Figs 4h,6a,d,7a,d); however, it is not large and or located prolaterally as in all other Clubiona s. l. It appears that species in this genus can be separated by the shape of the sperm duct course and relative width of the sperm duct (cf. Fig. 9c and Fig. 9d).
While studying morphology of the Porrhoclubiona with SEM, we found several notable characters: -The femur has few bald areas (Ba), not covered with a transversal furrow as other parts of the cuticle (Fig. 3d). Such bald areas are known in several unrelated families.
-The tarsal organ (To) in Porrhoclubiona is (if we recognized it correctly) slit like (Fig. 3c).
Aside from the species mentioned above, we consider three more species in this genus, P. laudata (O. Pickard-Cambridge, 1885), comb. n., ex. Clubiona and two new species, P. bosmansi sp. n. and P. moradmandi sp. n.  (2018) indicates that the species was described based on the male only, although Pickard-Cambridge (1885: 24) also described a female. The species distribution is indicated as China (Yarkand), although Pickard-Cambridge (1885) mentioned the species was described based on specimens collected on the road from Yarkand (lying in SW Xinjiang, China) to Bursi (lying in the Leh District of the Jammu and Kashmir, India).

Diagnosis.
Porrhoclubiona laudata differs from P. leucaspis by the conical tibial apophysis (vs. broad and rounded at the tip, Figs 4i, 6e) and thinner basal part of the embolus. Tibetan species differ from other species occurring in Central Asia by the large palp (cf . Figs 10b-d).  Abdomen without distinct pattern. Palp as in Figs 6a-c, 10c. Tibial apophysis gradually tapering, subconical; anterior margin of cymbium broad; base of embolus (Be) equal to ½ of tegular length, basal part of embolus (Eb) as long as approx. 2/3 of the base.
Female. Lacking among type series. Pickard-Cambridge (1885) described it as: "The female is rather larger, but resembles the male in colours and markings, except that the abdomen is less marked and streaked with rusty red; the form of the genital aperture, which is quite small, is characteristic". Description of C. genevensis from Western China seems to refer to this species. Epigyne as long as wide.
Distribution. Exact type locality is unknown and in either in southwestern Xinjiang (China) or in adjacent Northeastern Jammu and Kashmir (India). It seems that all records of C. genevensis from China (Xinjiang and Tibet) refer to this species.
Diagnosis. Porrhoclubiona bosmansi sp. n. differs from P. laudata by having a smaller carapace (1.7-1.77 vs. 2.11), smaller palp (cf. Fig. 10c and Fig. 10d) and thinner tibial apophysis. The new species differs from P. moradmandi sp. n. by the smaller palps and shorter modified cymbial hairs (cf. Fig. 5a and Fig. 5b), fewer pro-and retro- lateral spines on metatarsus III (2-2 vs. 3-3), and inclined anterior edge of the embolic base (vs. almost horizontal, cf. Fig. 10b' and Fig. 10d'). Female of P. bosmansi sp. n. differs from those of P. moradmandi sp. n. by the shape of the epigynal fovea, which is more transverse and lacking posterior notch (cf. Fig. 8a-c, d-f ). Females of the two species differ by the shape of the copulatory ducts and relative position of hyaline and sclerotised receptacles: sclerotised receptacles located anterior to the hyaline receptacles in P. bosmansi sp. n. and posteriorly in P. moradmandi sp. n. (cf. Fig. 9a and 9b).
Leg lengths (paratype).   Abdomen yellow-reddish at dorsal part with dark-reddish cordial mark. Lateral sides of abdomen reddish, ventral side yellowish.

Fe
Palp as in Figs 4a-c, 5a, g, 7d-f, 10d. Tibial apophysis triangular, wider than long; anterior edge of cymbium almost flat (horizontal, not rounded); setae in cymbial brush not dense, approx. 1/3 of cymbial length; anterior part of tegulum (At) longer than base of embolus (Be); posterior edge (Pb) of the basal part of embolus inclined as well as anterior part of embolic base.
Female. Carapace 2.1 long, 1.43 wide. Abdomen 3.38 long, 2.4 wide. Total length 5.5. Coloration as in males, but somewhat lighter. Chelicerae with 4 prolateral and 3 retrolateral teeth.   Epigyne as in Figs 8d, e, 9a. Fovea oval, more than twice as wide as long, posterior notch absent; translucent sclerotised receptacles (Sr) spaced by approx. one radius in intact epigyne; hyaline receptacles (Hr) located posterior to sclerotised receptacle; hyaline receptacles 1.3 times larger than sclerotised receptacles; loop of copulatory duct (Dl) directed posteriorly and spaced from each other by approximately one diameter.
Distribution. Hatlon Province of Tajikistan.  Figs 10b' and 10d')) and relatively longer tibia -length/width ratio approx. 2 (vs. short-  er, ratio ca. 1.5). Females of P. moradmandi sp. n. can be distinguished from P. bosmansi sp. n. by the shape of the epigyne: epigynal fovea pentagonal (vs. oval) with distinct posterior notch (vs. lacking), anterior position of hyaline receptacles (vs. sclerotised receptacle located anteriorly), anteriorly directed loop of copulatory duct (vs. posteriorly). Porrhoclubiona moradmandi sp. n. is very similar to P. leucaspis by the abdominal pattern, palp shape, and particularly by having a filamentous extension (Fl) of the tibial apophysis, although the female differs by the shape of the copulatory ducts and receptacle proportions (cf. Fig. 8a-c and Fig. 9b  Leg spination Abdomen yellow with greyish V-shaped stripes (indistinct due to poor condition of the specimen) dorsally.
Palp as in Figs 4d-f, 5b, d, e, 7a-c, 9d, 10b. Tibial apophysis subtriangular, wider than long; tip with filamentous extension (Fl); anterior edge of cymbium rounded, with one distinct macroseta; modified setae of cymbial brush dense and long almost ½ of cymbial length; basal part of embolus ca. ½ of embolus base height, anterior edge of embolic base and posterior edge of basal part of embolus horizontal; base of embolus shorter than anterior part of tegulum.
Leg lengths Leg spination Epigyne as in Figs 8a-c, 9b. Fovea pentagonal with deep posterior notch; translucent receptacles spaced by less than radius in intact epigyne; copulatory duct well distinct in ventral view; hyaline receptacles located anteriorly from the sclerotised re- ceptacles; loop of copulatory duct directed anteriorly; mesal part of copulatory ducts spaced by more than 3 times their diameters. Distribution. It is known from the type locality only.
vided us with specimens of Porrhoclubiona from Central Asia. We also thank Seppo Koponen (Turku, Finland) for providing museum facilities, and specimens of Porrhoclubiona from Europe. The English of the earlier draft was kindly checked by Sarah Crews (San Francisco, USA).