A revision of the trapdoor spider genus Liphistius (Mesothelae: Liphistiidae) in peninsular Malaysia; part 1 Author Peter J. Schwendinger text Revue suisse de Zoologie 2017 2017-09-30 124 2 391 445 journal article 31806 10.5281/zenodo.893555 4a2efe36-c48b-456b-9223-7b5793025a72 0035-418 893555 Liphistius malayanus Abraham, 1923 Figs 3B, D, F, H ; 4 - 5 Liphistius malayanus Abraham, 1923b : 770 -774, text-fig. 1a-b (description of female). – Abraham (1929: 674- 676, pl. 1, figs 3-4, pl. 2, figs 11-16; first description of male) . For synonymy and other taxonomically relevant publications see the World Spider Catalog, 2017 . Liphistius malayanus cameroni Haupt, 1983 : 282 , figs 3e, 4d, 5f, 6f (description of male and female). New synonym Type material: BMNH; female holotype of L. malayanus (not examined); Negri Sembilan , Gunung Angsi near Seremban ; XII.1922 ; leg. F. de la Mare Norris. – ZMH (A16/84); male holotype and female paratype of L. malayanus cameroni (examined) ; Pahang , Cameron Highlands , Berinchang ; 4.III.1981 ; leg. J. Haupt & T. Dach. Remark: A L. malayanus male (not examined) from Fraser’s Hill was described by Abraham (1929) and deposited in the BMNH. Haupt (1983 : 281, figs 3d, 4c) re-described this specimen and referred to it as the paratype , but it has no type status, having been collected in November 1928 , after the publication of the original description of the species. Material examined: ZMH (A16/84); male holotype [left palp detached, macerated and partly collapsed, cymbium and hematodocha cut open to expose tendons inside, all strong spines deformed; right palp without tegulum, contrategulum and embolus complex] (matured VII.1981 ) and female paratype of Liphistius malayanus cameroni ; Pahang , Cameron Highlands , Berinchang ; 4.III.1981 ; leg. J. Haupt & T. Dach. – Collection of Joseph Koh , n° JK.14.10.05.0001; 1 male ; Pahang , Cameron Highlands , Taman Tringkap NE of Brinchang , 4°28’26”N , 101°22’58”E ; 5.X.2014 ; leg. N. Bay. – MHNG ; 1 female , 1 juv. male; Perak , Cameron Highlands , ca 1 km SW of Ringlet , 1060 m ; 21.I.1995 ; leg. P.J. Schwendinger. – MHNG (sample TM-15); 1 juv. male; Pahang , Cameron Highlands , Tanah Rata , trail n° 9, 4°27.620’N , 101°23.400’E , 1210 m ; 29.IX.2001 ; leg. L. Monod. – SMF 7425 /2 (ex coll. C. F. Roewer ); 2 juv. females (labelled as “ 1 male and 1 female , det. Roewer , 1935”); Pahang , “ Ginting Kial Highlands ” [probably a peak in the Cameron Highlands ] . – SMF 40602; 1 male ; Pahang , Fraser’s Hill , roadside at Jalan Girdle ; 19.XII.2001 ; leg. S. Huber. – SMF 56206; 1 female ; Pahang , Fraser’s Hill , roadside at Jalan Girdle ; 19.XII.2001 ; leg. S. Huber. – MHNG (ex coll. S. Huber, sample O-4, C); 1 male (matured end March 2002 ), 1 female (moulted 24.VI.2002 ); Pahang , Fraser’s Hill , Jalan Guillemard ; 18.XII.2001 ; leg. S. Huber. – SMF 64093; 1 female ; Pahang , Fraser’s Hill , 1300 m , 3°43.105’N , 101°45.164’E ; 17.VI.2013 ; leg. P. Jäger. – MHNG (sample MAL-04/02); 1 male (matured 18.IX.2004 ), 2 females (moulted 30.VII.2004 , 24.I., 28.VIII. , 8.XII.2005 ; 19.X.2004 , 30.VII.2005 ), 4 juveniles ; Pahang , Genting Highlands , 3°25’42”N , 101°47’41”E , 1650 m ; 18.-19.V.2004 ; leg. P.J. Schwendinger. – SMF 21946/1; 1 female; without locality data; 10.XI.1933 ; ex coll. Wiehle, don. W. S. Bristowe. – SMF 60037; 1 female ; “ Selangor ” [should be Pahang ], Genting Highlands , 1800 m ; 3.II.1989 ; leg. U. Maschwitz , don. H. Steiner. – SMF 40016; 1 female ; Pahang , Genting Highlands , 1200 m ; 27.VII.2001 ; leg. A. Kovac. – MHNG (sample SIM-01/14); 1 male (matured end VIII.2003 ), 1 female (moulted 31.XII.2001 ), 1 juvenile ; Selangor , Templer Park , 3°17’55”N , 101°37’13”E , 230 m ; 13.VII.2001 ; leg. P.J. Schwendinger. Fig. 3. SEM-micrographs of embolus proper of left palpal organ of two males: Liphistius desultor from Penang Hill (A, C, E, G) and Liphistius malayanus from Fraser’s Hill (B, D, F, H). (A) Retroventral-distal view. (B) Dorsal-distal view. (C) Distal view (dorsal side to the right). (D) Distal view (prolateral side to the right). (E) Retrodorsal view. (F) Prodorsal view. (G) Proventral view. (H) Prolateral view. Abbreviations: f - major embolic fold; m - membranous part of embolus proper; s - sclerotised part of embolus proper. Fig. 4. Liphistius malayanus , details of palp of four males: Templer Park (A, C-D, H-I); holotype of L. malayanus cameroni (B, E, G); Cameron Highlands, ZRC (F, J), Fraser’s Hill, MHNG (K-L). (A) Paracymbium and tibial apophysis of left palp, retroventral view. (B) Tibial apophysis of right palp, ventral view. (C) Distal part of left palpal organ, proventral view (arrow indicating V-shaped row of denticles at proventral end of distal edge of contrategulum). (D, F) Same, retrodorsal and slightly proximal view. (E) Same, retrodorsal view. (G-H) Left palpal organ, distal view (dorsal side up). (I, K) Distal edge of contrategulum of right palp, distal view (dorsal side to the left). (J) Same, distal and slighly prolateral view. (L) Same of left palp, distal and slighly prolateral view (dorsal side to the right). Abbreviations: a - dorsal apex of contrategulum; de - distal edge of contrategulum. Scale lines: 1.0 mm (A; B; C-D; E, G; F; H-L). Diagnosis: Large, dark-coloured species in both sexes. Males distinguished from those of other species in the malayanus -group by tibial apophysis situated distinctly lower (more proximal) than distal margin of palpal tibia ( Fig. 4B ); distal edge of contrategulum carrying a series of small denticles ( Fig. 4 G-L), those at proventral end arranged in a V-shaped row ( Fig. 4C ), 1-2 at prodorsal end isolated and slightly enlarged; dorsal apex of contrategulum large, wide, tounge-shaped and asymmetrical, its prolateral margin more strongly arched than retrolateral margin ( Fig. 4 G-L); dorsal wall of embolus proper as wide as ventral wall ( Fig. 4 G-H); membranous part of embolus proper distally narrower than proximally ( Figs 3F, H , 4C ). Females distinguished by vulval plate ( Fig. 5 ) with widely trapezium-shaped (posteriorly widest) posterior stalk not connected to pigmented lateral patches in genital atrium; poreplate with distinct anterolateral lobes; CDO large, very variable in shape, mostly longer than wide; receptacular cluster very large, in most cases protruding beyond straight or procurved anterior margin of poreplate. Additions to description of male: Scopula : See paragraph “Variation”. Palp : Both apical lobes of cymbium equally short, dorsal one usually slightly more pointed than ventral one. Retrolateral apophysis of palpal tibia entire, situated at a clearly more proximal level than distal margin of article ( Fig. 4B ), carrying four long apical megaspines ( Fig. 4 A-B). Paracymbium relatively small and shallow, about as deep as cymbium in retroventral view ( Fig. 4A ; Platnick & Sedgwick, 1984 : fig. 64); cumulus low, carrying a compact group of moderately long thick bristles. Subtegulum without apophysis. Tegulum with quite few small teeth only in retrodorsal portion of proximal edge, distal margin not drawn into an edge ( Fig. 4 D-F). Contrategulum with very indistinct, widely rounded ventral process ( Fig. 4 G-H); distal edge with a series of small denticles, those at proventral end arranged in a V-shaped row ( Fig. 4C ), 1-2 enlarged denticles or a sharp keel in prolateral to prodorsal section of distal edge ( Fig. 4 G-L); dorsal apex of contrategulum developed as a large, linguiform, strongly projecting horizontal plate, its prolateral margin being more strongly arched than its retrolateral margin ( Fig. 4 G-L). Embolus complex with para-embolic plate developed as a more or less distinctly elevated, rounded edge ( Fig. 4 D-F); below it a second small edge present in some cases ( Fig. 4D ); sclerotised and membranous part of embolus proper in contact for almost their entire length ( Fig. 3D, F, H ); sclerotized part with numerous longitudinal ribs carrying tiny denticles distally ( Fig. 3B, D, F, H ), its dorsal and ventral walls equally wide and parallel to each other ( Figs 3D ; 4G-H), dorsal wall ending in indistinct rounded lobe ( Fig. 4 C-F); membranous part distally narrower than proximally ( Fig. 3D, F, H ), its proximal portion essentially unpigmented. Additions to description of female: Vulval plate ( Fig. 5 ) always with few to many hairs laterally in genital atrium, rarely also posteromedially ( Fig. 5I ). Posterior stalk more or less distinctly trapezium-shaped (posteriorly wider than anteriorly), densely pitted in posterior portion. Poreplate wider than long, with a pair of anterolateral lobes projecting anteriad and slightly bent ventrad; several of these lobes more or less distinctly constricted at base ( Fig. 5 A-C, E); anterior margin of poreplate straight to procurved; lateral and posterolateral margins on ventral side of pore plate bulging ( Fig. 5B, J ). CDO large and of variable shape, mostly longer than wide, rarely wider than long ( Fig. 5E ). Receptacular cluster large and strongly racemose, in most cases protruding beyond anterior margin of poreplate, rarely not ( Fig. 5G ). Taxonomic remarks: Liphistius malayanus cameroni has been the only subspecies in the genus Liphistius . As specimens later collected near the type locality show no relevant differences from specimens of the nominal subspecies, this case became doubtful. Thus a re-examination of the types of Liphistius malayanus cameroni was necessary, and it revealed that they differ considerably from the illustrations and explanations given in the original description of this subspecies. The vulva of the female paratype was cleared of tissue, but the ventral wall of the genital region (with bristles still attached) was not removed, thus not allowing a clear view of the ventral side of the poreplate. It therefore appears that Haupt’s (1983) illustrations were made with the help of a compound microscope without using a drawing tube, which would explain why the poreplate in the original illustrations is wider than in reality and why the outlines of the posterior stalk are too rounded. The same is probably also the case in the dorsal view of the same poreplate, it being too wide, with a triangular CDO leading to the receptacular cluster and with a peculiar double posterior margin of the posterior stalk (see Haupt, 1983 : figs 5f, 6f). In fact the vulva of the paratype of L. malayanus cameroni does not differ from the vulvae of the other L. malayanus females examined ( Fig. 5 A-B cf. Fig. 5 C-J) in any way that would warrant a distinct taxonomic status. Fig. 5. Liphistius malayanus , vulval plate of eight females: paratype of L. m. cameroni from Berinchang, Cameron Highlands (A- B); exuvia of female from Ringlet, Cameron Highlands (C); female (SMF 64093) from Fraser’s Hill (D); female (moulted 24.VI.2002) from Fraser’s Hill; poreplate slightly deformed by brief immersion in cold KOH (E); female (SMF 56206) from Fraser’s Hill (F); female (SMF 60037) from Genting Highlands (G); female (moulted 8.XII.2005) from Genting Highlands (H); exuvia of female from Templer Park (I-J). (A, C-I) Dorsal view. (B, J) Ventral view. Scale lines: 1.0 mm (A-B; C, F; D; E, G; H; I-J). The illustrated differences between the male holotype of L. malayanus cameroni and the male that Haupt incorrectly referred to as a paratype of L. malayanus (in BMNH, not examined) are mostly due to a slightly different view of the palp and due to the fact that the palpal organ of the BMNH male is expanded (turned clockwise by about 90°) ( Haupt, 1983: figs 3D, 4C cf. figs 3E, 4D ). None of the L. malayanus males examined has the tibial apophysis of the male palp situated as far distally as illustrated by Haupt (1983: figs 3D, 4C) , therefore this may also be due to a different view. The embolus proper of the left palp of the L. m. cameroni holotype (that of the right palp is missing, together with tegulum and contrategulum) is slightly narrower than that of the second male from the Cameron Highlands and that of other conspecific males examined. This difference is presumably connected to size: the L. m. cameroni holotype is the smallest among the L. malayanus males examined. The character in which the holotype of L. m. cameroni most visibly differs from other conspecific males is a slightly elevated para-embolic plate ( Fig. 4E ), which is also present in the second male from the Cameron Highlands ( Fig. 4F ). However, an indistinctly elevated para-embolic plate is present in one of the two males from the Fraser’s Hill, whereas all other conspecific males possess only a distinct edge at this place ( Fig. 4D ). Thus the holotype of L. m. cameroni differs from other conspecific males examined in possessing the most extreme states of three quite variable characters, but it is not representative for the population in the Cameron Highlands . I consequently place Liphistius malayanus cameroni in the synonymy of Liphistius malayanus which makes the species monotypic again. Variation: Carapace lengths in males (n=6; including the holotype of L. m. cameroni ) 7.90-11.23, carapace widths 7.04-9.69; in the largest female from each locality (n=4; not including the paratype of L. m. cameroni ) 10.86-16.67 and 10.00-14.07, respectively. The holotype of L. m. cameroni is the smallest male examined, with carapace length 7.90, width 7.04; in the second male from the Cameron Highlands it is 9.63 and 8.40, respectively. The largest female is from the Templer Park, measuring 16.67 and 14.07; the second largest female is from the Fraser’s Hill, measuring 16.54 and 14.07. The female paratype of L. m. cameroni measures 9.26 and 7.53, the second female from the Cameron Highlands 10.86 and 10.12. The apical megaspines on the tibial apophysis are quite variable in length, but not as thin as illustrated in Murphy & Platnick (1981: figs 8, 11, 14, 17, 20) and in Platnick & Sedgwick (1984: figs 63-67) ; the ventral-most of them is mostly bent distad and has a thin, long apex ( Fig. 4 A-B). The ventral megaspine is bent in the male from the Templer Park ( Fig. 4A ), whereas in the holotype of L. m. cameroni ( Fig. 4B ), in a male from the Genting Highlands , and also in a second male from the Cameron Highlands it is slightly sigmoid. The distal edge of the contrategulum carries one or two slightly enlarged dorsal denticles at some distance from the projecting dorsal apex ( Fig. 4G , I-L); in two males there is a sharp edge at the same place ( Fig. 4H ), but only on one palp. These larger denticles are followed by a variable number of smaller proventral denticles. The para-embolic plate is always low, most visibly elevated (but not with a sharp edge) in the holotype of L. m. cameroni ( Fig. 4E ), in a second male from the Cameron Highlands ( Fig. 4F ) and in a male from the Fraser’s Hill. In the second male from the Fraser’s Hill it is not elevated; the remaining males examined are intermediate ( Fig. 4D ). An indistinct second edge, situated below the indistinct para-embolic plate, is found in the only available male from the Genting Highlands and in the only available male from the Templer Park ( Fig. 4D ), but not in other males examined ( Fig. 4 E-F). The size of the embolus proper varies from relatively thick in the male from the Templer Park ( Fig. 4D ) to distinctly more slender in the holotype of L. m. cameroni ( Fig. 4E ); the other males are intermediate ( Fig. 4F ). The longitudinal ribs on the sclerotised part of the embolus proper vary from indistinct to distinct ( Fig. 3B, D, F, H ), so do the denticles on them. There is considerable variation in the extent of the tarsal scopulae in males. The scopula on tarsus I is always thin and undivided, covering two-thirds of the ventral side in most males, only one half in the male from the Genting Highlands . The scopula on tarsus II is slightly denser than on tarsus I, undivided and covering twothirds of the ventral side in all males. Tarsal scopula III is like tarsal scopula II in most males, in the non-type male from the Cameron Highlands it is light. Tarsal scopula IV is dense in all males except for the non-type male from the Cameron Highlands (light), undivided in all males except for the same male from the Cameron Highlands (distinctly divided) and for the male from the Genting Highlands (indistinctly divided), covering two-thirds of the ventral surface in most males except for a male from the Cameron Highlands (covering only half of the surface) and the male from the Templer Park (covering three-quarters). The male holotype of L. malayanus cameroni had already been returned to its depository when the tarsal scopulae were examined. Even if both males from the Cameron Highlands differed from those of other localities in shape and extent of their tarsal scopulae (especially on tarsus IV), this would not warrant a subspecific distinction. The tarsal scopula in this species is too variable to be of high taxonomic value. The vulval plates of large females have more lateral hairs in the genital atrium than those of small females ( Fig. 5I cf. Fig. 5H ). In the largest female (from the Templer Park) hairs are also present posteromedially in the genital atrium ( Fig. 5I ). The shape of the posterior stalk is quite variable but mostly trapezium-shaped ( Fig. 5 ). Even more variable is the size and shape of the CDO of the poreplate, ranging from (mostly) longer than wide to (rarely) wider than long, from quadrangular to pentangular and from near-triangular to near-circular ( Fig. 5A , C-I). The receptacular cluster is always large, undivided and has a complex structure, covering most of the ventral side of the poreplate ( Fig. 5B, J ); only in one female examined (from the Genting Highlands ) does it not protrude beyond the anterior margin of the poreplate ( Fig. 5G ). The anterior poreplate margin varies from more or less strongly invaginated (in most cases; Fig. 5 A-E, G-J) to slightly arched medially ( Fig. 5F ). Distribution: Liphistius malayanus is known from several lowland and upland localities in the western part of central Malaysia ( Perak , Pahang , Selangor and Negeri Sembilan ; Fig. 1 , localities 1-5). See also Platnick & Sedgwick (1984: 24) . Liphistius records from other localties near Kuala Lumpur (e.g. Klang Gates; Murphy & Murphy, 2000 : plate 2.5) can also be attributed to L. malayanus . This species has a relatively wide geographical range (the northernmost localitiy, in the Cameron Highlands , is about 280 km away from the southernmost locality, Gunung Angsi) and a large vertical distribution (from 230 to 1800 m altitude). Biology: Spiders were collected from quite different habitats: soil on exposed sides of roads and trails inside and outside forests (but never very far from a forest), sides of erosion gullies, sloping forest floor, and in decomposing wood of logs lying on the forest floor. Trapdoors of females were up to 3.8 cm long and 6.5 cm wide. Signal lines (not more than eight) were usually about 10 cm long, in one immature male even 24 cm long. The longest burrow measured was 20 cm long. Two penultimate males had trapdoors with a length of 1.9-2.2 cm and a width of 3.0- 3.4 cm , which is probably normal. A juvenile male (moulted again but died before reaching maturity) from near Ringlet in the Cameron Highlands had a surprising 2.9 cm long and 5.0 cm wide trapdoor. The female paratype of L. m. cameroni and two other females (from Fraser’s Hill and Templer Park) show bite marks on their carapaces and chelicerae. These were probably caused by specimens of Ljunghia bristowi , a species of ectoparasitic laelapid mites originally described from L. malayanus (see Halliday & Juvara- Bals, 2016: 837-845). Mature males were collected in the field in October and December; in captivity males matured between July and March. The mating period in L. malayanus appears to be much longer than in other congeneric species. I did not find any egg cases.