The Panjange nigrifrons group in Borneo ( Araneae : Pholcidae ) : high diversity in Sarawak , apparent absence in Sabah

Abstract. We revise the Panjange nigrifrons group in Borneo and document an unexpected diversity in western Sarawak forests. Five species occur within 80 km from Kuching, each species being known from its type locality only. Further species occur east until Niah, but the genus seems to be absent from Sabah. We contrast this with another pholcid genus (Aetana Huber, 2005), which is diverse in Sabah and westward until Niah, but does not seem to occur in central and western Sarawak. Five species are newly described: Panjange kapit Huber, sp. nov., Panjange kubah Huber, sp. nov., Panjange niah Huber, sp. nov., Panjange pueh Huber, sp. nov., Panjange seowi Huber, sp. nov.; Panjange tahai (Huber, 2011) comb. nov. is transferred from Pholcus.

Unfortunately, little is known about the relationships among leaf-dwellers and between leaf-dwellers and their closest relatives in other microhabitats.In only a few cases, phylogenetic analyses of morphological and molecular data have provided a basis for reconstructing the direction of evolutionary shifts among microhabitats.For example, the leaf-dwelling African Smeringopus cylingrogaster (Simon, 1907) and a clade of four Hispaniolan leaf-dwelling species in the genus Modisimus Simon, 1893 are quite clearly derived from near-ground dwelling ancestors (Huber et al. 2010;Huber 2012;Dimitrov et al. 2013); on the other hand, litter-dwelling Metagonia Simon, 1893 species in Brazil and the African litter-dwelling Pholcus kribi Huber, 2011 are apparently derived from leaf-dwelling ancestors (Dimitrov et al. 2013).For Southeast Asian taxa, molecular data have barely been available, and previous efforts at resolving phylogenetic relationships using morphology have often failed to convincingly resolve the relevant nodes (Huber 2011;Huber & Nuñeza 2015).
The present paper focuses on taxonomy, but is part of a major effort to resolve relationships among Southeast Asian pholcids and to reconstruct their evolutionary histories.It deals with Panjange, a genus that consists of leaf-dwelling species only, but whose internal and external relationships are far from clear.Bornean representatives of Panjange (i.e., representatives of the Panjange nigrifrons group) are in fact more similar to species currently placed in Pholcus from Sumatra, the Malay Peninsula, and Sri Lanka than to other species groups of Panjange in the Philippines (including the type species) and east of the Wallace Line (Huber 2011;Huber & Nuñeza 2015).It is not yet clear if these similarities are just plesiomorphies or if they reflect the non-monophyly of Panjange.Preliminary molecular data suggest the latter (A. Valdez-Mondragón, D. Dimitrov, B.A. Huber, unpubl. data).In this contribution we will deal with the taxonomy of the Panjange nigrifrons group, we provide basic microhabitat data, and we document a biogeographic peculiarity that has been known in other groups such as birds (e.g., Gawin et al. 2014;Sheldon et al. 2015) but to our knowledge not in spiders: an apparent distributional limit between Sarawak and Sabah.

Material and methods
Most of the material studied herein was collected during a recent expedition to northern Borneo (July-August 2014).The material is currently deposited at Sarawak Museum, Kuching (SMK), Malaysia and Zoologisches Forschungsmuseum Alexander Koenig, Bonn (ZFMK), Germany.Additional material came from the American Museum of Natural History, New York (AMNH), U.S.A. and the Netherlands Centre for Biodiversity Naturalis, Leiden (RMNH), the Netherlands.Species descriptions are arranged as in the cladogram in Huber & Nuñeza (2015).Methods and terminology used are as in recent revisions (Huber 2011(Huber , 2013)).Measurements are in mm unless otherwise specified.Eye measurements are ± 5 µm.Epigyna were cleared in warm NaOH solution and stained with chlorazol black.For SEM photos, specimens were dried in HMDS (Brown 1993) and photographed with a Hitachi S-2460 scanning electron microscope.SEM data are presented within the descriptions but are not based on the holotype specimens described.Locality coordinates are in round brackets when copied from labels and original publications or when received directly from collectors, in square brackets when originating from some other source (such as online gazetteers, Google Earth, etc.).The distribution maps were generated with ArcMap 10.0.

Note on species groups
Previous work on Panjange has identified three species groups, the nigrifrons group on Borneo (Fig. 1), the lanthana group on the Philippines, and the widespread cavicola group (Sulawesi to northern Australia) (Deeleman-Reinhold & Platnick 1986;Huber 2011;Huber & Nuñeza 2015).From here on, we will deal only with the nigrifrons group.A separate paper on the lanthana group has recently been published (Huber & Nuñeza 2015), and a molecular phylogeny of the entire genus and its closest relatives is in preparation (A. Valdez-Mondragón, D. Dimitrov, B.A. Huber, unpubl. data).

Female
Similar to male but without eye stalks; eye triads on low humps and never with pointed processes (Fig. 38); chelicerae unmodified; legs slightly shorter than in male.Epigynum weakly sclerotized, with distinct 'knob'; either without scape (Fig. 19), with very short scape (Figs 31,72,84), or with medium to long scape (Figs 36,87).Internal genitalia with pair of pore plates of variable shape, sometimes with very complex system of internal folds of unknown function (Figs 59,90,93).

Monophyly and relationships
The cladistic analysis in Huber & Nuñeza (2015) included all available species of Panjange except for Pa.tahai comb.nov.It resolved the Pa.nigrifrons group as monophyletic, but with weak support (only one synapomorphy: the ventral apophysis on the male palpal femur).Internal relationships and relationships to other species groups in Panjange and to Pholcus also remained unconvincing.Preliminary molecular data (including six species of the Pa.nigrifrons group and six species of the Pa.lanthana group; A. Valdez-Mondragón, D. Dimitrov, B.A. Huber, unpubl. data) suggest that Panjange is not monophyletic and that the Pa.nigrifrons group is in fact more closely related to species currently in Pholcus than to the Pa.lanthana group.Morphological data partly point in the same direction (e.g., the bipartite cheliceral apophyses that occur in the Pa.nigrifrons group as well as in the Pholcus minang group) but problems arise from character conflict, dubious homologies of bulbal sclerites, and missing SEM data for many species (especially of the potentially closely related Pholcus minang group).
Pholcus tahai [now Panjange tahai (Huber, 2011) comb. nov.] from Kalimantan was not included in the cladistic analysis in Huber & Nuñeza (2015).Depending on how the main bulbal process beside the embolus is coded (as unknown, as uncus, or as appendix) the species is either resolved as sister to Panjange or as member of the Pa.nigrifrons group.Each solution has its problems: the procursus of Panjange tahai comb.nov.partly looks very different from that of representatives of the Pa.nigrifrons group: it lacks ventral ridges and it has a hinge dividing proximal and distal parts; on the other hand, the tip of the procursus (compare fig. 684  Pa.tahai comb.nov.occurs on Borneo like all other species of the Pa.nigrifrons group, while all species of the Pholcus minang group (to which the species was previously assigned) are from Sumatra and the Malay Peninsula (Huber 2011).

Natural history
All species observed in the field share a very similar microhabitat and web.They were consistently found in webs among low vegetation in well preserved forests, usually at about 0.5 m above the ground or even lower.The domed webs had a diameter of about 15-20 cm and were at their apex connected to the underside of a leaf where the spiders rested.At some localities, large numbers of cecidomyiid flies were found hanging in the webs together with the spiders.Egg-sacs are slightly elongated and contain approximately 20-30 eggs (e.g., Figs 13,62).

Composition
As construed here, the Panjange nigrifrons group now includes ten species: Panjange bako Huber, 2011;Pa. iban Huber, 2011;Pa. kapit (Huber, 2011) comb.nov.Judging from known distribution patterns in Sarawak and from the large poorly sampled forests of Kalimantan, the group is likely to contain several times as many species.However, the absence of Panjange in northeastern Borneo may be real rather than an artifact of poor sampling.Our own collecting at nine localities east of Niah did not produce a single specimen of Panjange (Fig. 95; see Discussion).

Distribution
Known from type locality in Sarawak only (Fig. 1).

Etymology
Named for the type locality; noun in apposition.Color.Carapace pale ochre yellow to whitish, without posterior mark, ocular area and clypeus dark brown, with single black mark in AME area; sternum whitish; legs ochre-orange with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.

Natural history
The webs were found among the vegetation at 0.5-1 m above the ground, with the apex of the domed sheet connected to the underside of a leaf.
Color.Carapace pale ochre yellow to whitish, posterior mark (Fig. 11) lost in ethanol, ocular area and clypeus dark brown (black in life), without black spots in AME area; sternum whitish; legs ochre-orange with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.

Natural history
Webs were found among vegetation close to the ground, with the apex of the domed sheet connected to the underside of a leaf.Large numbers of Cecidomyiidae (deposited in ZFMK) were seen hanging from the silk lines in most webs.Egg sacs were slightly elongated, and contained about 25-30 eggs each (n = 2).

Diagnosis
Easily distinguished from congeners by highly distinctive procursus (very long, with hinge between proximal and distal parts; fig.674

Natural history
The webs had a diameter of about 10-20 cm and were found among vegetation in primary forest close to the ground where the domed sheets were attached to the undersides of leaves.

Distribution
Known from type locality in Sarawak only (Fig. 1).

Description -amendments
Male distal cheliceral apophyses clearly bipartite as in other species described herein (this was correctly noted in the original description but not correctly drawn in fig.11).Appendix with very distinct and unique pocket at ~60% of its length.Tibia 1 L/d: 100; tibia 2/tibia 4 length: 1.09 (not 1.33 as in original description); distance PME-PME: 420 µm; retrolateral trichobothrium at 2%.

Diagnosis
Easily distinguished from most congeners by distinctive transversal sclerite on procursus (Fig. 70); from very similar Pa.seowi sp.nov.by other details of procursus (unique prolateral process, Figs 69, 77-78; distal element longer, Fig. 70; absence of prolateral ridges on procursus, Fig. 69).Females are easily distinguished from Pa. nigrifrons by short scape but difficult to distinguish externally from other congeners in Sarawak (distinctive shape of pore plates and pair of median sclerites; similar only in Pa.seowi sp.nov.).

Etymology
Named for the type locality; noun in apposition.
Color.Carapace pale ochre yellow to whitish, posterior mark (Fig. 63) lost in ethanol, ocular area and clypeus dark brown (black in life), without black marks in AME area; sternum whitish; legs ochreorange with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.
PalPs.As in Figs 69-70; coxa unmodified; trochanter with slightly curved pointed retrolatero-ventral apophysis; femur with curved finger-shaped ventral apophysis; procursus with row of about 17 ventral ridges, with distinctive transversal sclerite and long prolateral process close to large flat distal element (Figs 77-78), with two distinct spiny processes in distal pit (Fig. 80); bulb with strong proximal sclerite, slightly curved appendix, and long partly sclerotized embolus with distinct distal fringes .

Natural history
Most specimens were found in a very limited area close to a waterfall.The domed webs had a diameter of about 15-20 cm and in each case the apex of the dome was attached to the underside of a leaf where the spider rested.

Diagnosis
Easily distinguished from most congeners by distinctive transversal sclerite on procursus (Fig. 83); from the very similar Pa.kubah sp.nov.by other details of procursus (prolateral ridges on procursus, Fig. 82; small transparent process close to transversal sclerite, Fig. 82; absence of long membranous prolateral process; distal element shorter).Females are easily distinguished from Pa. nigrifrons by short scape but difficult to distinguish externally from other congeners in Sarawak (distinctive shape of pore plates and pair of median sclerites; similar only in Pa.kubah sp.nov.).

Etymology
Named for Francis Seow-Choen from Singapore, surgeon and expert on stick insects.

Natural history
The domed webs were found under green leaves among vegetation, usually about 0.5 m above the ground.

Distribution
Known from type locality in Sarawak only (Fig. 1).

Discussion
We high levels of diversity and endemism of Panjange in western Sarawak, but it remains unclear to which extent this can be extrapolated to other, mostly very poorly sampled regions of Borneo.
Panjange does occur in Kalimantan, but it might be significantly less diverse there than in western Sarawak.A large part of Kalimantan is (or was originally) covered by peat swamp forest that in general is thought to have distinctive but limited diversity (e.g., Mohamedsaid & Holloway 1999;Posa et al. 2011).However, at least northern West Kalimantan is expected to have similar high diversity and endemism as western Sarawak, but this region seems to be among the most poorly explored regions worldwide, as far as spiders are concerned.
An extrapolation from Sarawak to unexplored regions in Borneo is particularly problematic in light of our surprising finding that Panjange seems to be absent from Sabah.Absence is of course difficult to document but the available data strongly suggest a real pattern rather than an artifact of sampling.First, we sampled with the same methodology and diligence at six localities west of Niah (sea level to 930 m a.s.l.) as we did at nine localities east of Niah (sea level to 1650 m a.s.l.).Panjange was found at all six localities west of Niah (and in Niah), but in none of the nine localities east of Niah.Second, the same pattern is apparent in the large collections made by C. Deeleman-Reinhold since the 1980s in Sarawak (Bako, Semengoh) and Sabah (Mt.Kinabalu, Poring, Sepilok) (Pholcidae mostly published in Huber 2011).Third, extensive more recent quantitative sampling by A. Floren in Sabah (Crocker Range, Mt.Kinabalu, Tawau) also did not produce a single specimen of Panjange (Huber 2011 and B.A. Huber, unpubl. data).
Further evidence suggesting a real pattern comes from another pholcid genus, Aetana Huber, 2005.This genus is not particularly close to Panjange (same subfamily but not part of the Pholcus group of genera like Panjange; Huber 2011) but is relevant in this context because it shows the exactly opposite distribution in northern Borneo.We found various species of Aetana at eight of nine localities east of Niah (and in Niah), but in none of the six localities west of Niah (Huber et al. 2015).Figure 95 summarizes current knowledge about the distributions of Panjange and Aetana in northern Borneo, combining all known records.Obviously, large sampling gaps remain, and a meaningful interpretation will have to build on a much more intense sampling including Kalimantan.However, data on better studied organisms such as birds have also supported a biogeographical divide between Sabah and Sarawak despite continuous rainforest (Gawin et al. 2014;Sheldon et al. 2015).These authors mainly suggest historical reasons for this pattern, such as early Pleistocene rainforest refugia in Sabah versus dry habitats in Sarawak (at that time part of central Sundaland).Whether the same explanations apply to spiders as to birds remains to be established.