Evolution of genital asymmetry, exaggerated eye stalks, and extreme palpal elongation in Panjange spiders (Araneae: Pholcidae)

. We describe several new species of the previously monotypic Panjange lanthana species group from the Philippines and document their extraordinary morphology. Some species show strong male genital asymmetry, a phenomenon that seems to be exceedingly rare in spiders. Males of most species have eye stalks, and in two species these eye stalks are among the longest ever recorded in spiders. Some species show a tendency for male genital (pedipalp) elongation, and one species has the longest and thinnest palps ever recorded in Pholcidae. A cladistic analysis is performed including all described and several undescribed species of Panjange (except for one “problem


Introduction
Panjange Deeleman-Reinhold & Deeleman, 1983 spiders are leaf-dwelling pholcids widely distributed in insular Southeast Asia, from Borneo and the Philippines to northern Australia (Huber 2011).They exhibit some of the most extraordinary morphology among Pholcidae.Males of most species have eye stalks, sometimes with long pointed processes; males of some species have unusually elongated pedipalps, which in spiders function as copulatory organs; and females of some species have external portions of their genitalia strongly folded and extensible (Huber 2011).The biological significances of these sexual modifications remain unclear.
The genus has long been divided into three species groups: the nigrifrons group on Borneo, the cavicola group ranging from Sulawesi to northern Australia, and the lanthana group in the Philippines (Deeleman-Reinhold & Platnick 1986;Huber 2011).However, the latter "group" so far consisted of only one species, the very unique type species Panjange lanthana Deeleman-Reinhold & Deeleman, 1983, known only from the type series from Luzon.No further Panjange species has been known from the Philippines.Here we describe several new species that appear closely related to Pa. lanthana, and that show some very unique morphological features, not only among Pholcidae but among spiders in general: genital asymmetry, exaggerated eye stalks, and extreme palpal elongation.
Genital asymmetry is very common in some arthropod groups (such as insects) but very rare in others (such as spiders) (reviewed in Huber et al. 2007).Ultimately, the reason for this may be the fact that insects have an unpaired median intromittent organ (the aedeagus) while spider males copulate with a pair of pedipalps (Huber 2010).However, while there are plausible evolutionary routes to asymmetry in insects (related to changes in mating position that in turn are driven by sexual selection; Huber 2010), the origin of spider genital asymmetry is still poorly understood.Two of the new species described below are among the most asymmetric spiders known so far.
Eye stalks are also widespread in arthropods such as crustaceans (e.g., Barnes 1968;Bowman 1984;Zeil et al. 1986) and insects (in particular flies: e.g., Burkhardt & de la Motte 1988;Wilkinson & Dodson 1997;Baker & Wilkinson 2001), but very rare in arachnids.Fairly well studied exceptions are certain linyphiid species where the male "head" or ocular area is modified, presumably as a result of sexual selection: males take advantage of the fact that the female mouthparts contact their ocular region during copulation and modify this region in an attempt to manipulate females (e.g., Schaible et al. 1986;Vanacker et al. 2003;Kunz et al. 2012).Usually, these modifications come in the form of low humps or turrets, but rarely also as (unpaired median) stalks, such as in Walckenaeria acuminata Blackwall, 1833.In Pholcidae, a similar mating position may explain the repeated origin of male "head" modifications, such as in the genera Modisimus Simon, 1893, Smeringopina Kraus, 1957, Pholcus Walckenaer, 1805, Leptopholcus Simon, 1893, and Panjange (see e.g., Huber 1997a, 2011, 2013).Two of the new species described below have the longest eye stalks known in Pholcidae, and possibly in spiders.
Elongated male copulatory organs (pedipalps) occur in various spider groups, but the palps are rarely longer than the body (exceptions occur for example in Filistatidae; Marusik & Zamani 2015).The significance of this elongation is poorly understood, and the plausible idea that long palps may save the male from cannibalism has to our knowledge never been tested.In Pholcidae, palpal strength seems to play an important role, often leading to very thick palpal segments (Huber & Eberhard 1997).As a result, elongated (and thus necessarily thin) palps are rare in pholcids, and apparently limited to Southeast Asian leaf dwelling species.The only cases known to us occur in the genera Calapnita Simon, 1892 (most notably C. semengoh Huber, 2011), in most species of the Panjange cavicola group (Huber 2011) and in certain species currently placed in Pholcus but possibly close to Panjange (such as Pholcus bohorok Huber, 2011).In all of them, females have extensible external genitalia (or scapes), suggesting a functional correlation between palp length and scape length.A new species described below has by far the longest and thinnest male palps known in Pholcidae, as well as the longest female scape.
Species descriptions are arranged as in the cladogram in Fig. 1.Methods and terminology are as in recent revisions (Huber 2011(Huber , 2013)).Measurements are in mm unless otherwise noted.Eye measurements are +/-5 µm.Epigyna were cleared in warm NaOH solution and stained with chlorazol black.For SEM photographs, 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.The distribution map was generated with ArcMap 10.0.

Note on species groups
Previous work on Panjange has identified three species groups, the nigrifrons group on Borneo, the widespread cavicola group (Sulawesi to northern Australia) and the previously monotypic lanthana group on the Philippines (Deeleman-Reinhold & Platnick 1986;Huber 2011).The present cladistic analysis largely recovers these three groups, but the cavicola group is considered paraphyletic (Fig. 1).
The present paper focuses on the lanthana group A separate paper on the nigrifrons group is in preparation, as is a molecular phylogeny of the entire genus and of its relationships to other Pholcinae.

Identification key
This key is primarily designed to identify species groups of Panjange and species of the lanthana group.It mostly works for males and females, but females remain difficult to identify.An examination of the cleared internal genitalia may be necessary (step 7); geography rather than morphology may help to separate females (step 2); in one species, the female is unknown (Pa. bukidnon Huber sp. nov.;step 4).

Cladistic analysis
Preliminary analyses including all available species of Panjange (all 18 described species plus five undescribed species) revealed that the highly aberrant Pa.bukidnon Huber sp.nov.was also the most difficult to place in the cladogram.It ended up at three different positions, resulting in many trees and a consensus with large polytomies.For this reason, Pa.bukidnon Huber sp.nov.was deleted from the final matrix.It may be correctly placed in the lanthana group but its sister taxon remains unknown (see Monophyly and relationships below).
One potentially informative character system was also largely deleted from the final matrix.The apophyses and processes on the male palpal trochanter come in a wide variety of numbers, shapes and positions, and it seemed impossible to code them in a convincing way.Only one detail was left in the matrix (Appendix 3, character 10).
Using NONA with hold/100, mult*200 (or hold/10; mult*10.000),and amb-for the matrix in Appendix 1 and equal character weights resulted in two most parsimonious cladograms with a length of 37 (Ci = 70; Ri = 90).Variation occurred only among outgroup taxa and in the support for the monophyly of Panjange (characters 15 and 20, or only character 20).The cladogram shown in Fig. 1 is the strict consensus of these two cladograms.
Using TNT with the implicit enumeration algorithm resulted in the same two trees, as did successive weighting in NONA (with the consistency index as weighting function).Implied weighting in Pee-Wee (which resolves character conflict in favor of the characters that have less homoplasy) was used with various settings of the constant of concavity K (1)(2)(3)(4)(5)(6) to explore the stability of different clades under different weighting regimes.Major differences to the tree shown in Fig. 1 occurred only with K=1 and K=2, which resulted in large polytomies and no support for the monophyly of Panjange.However, even in these trees, the monophyly of the lanthana group was consistently recovered as was its close relationship to the cavicola group.The only difference within the lanthana group concerned the position of Pa.hamiguitan Huber sp.nov., which was not sister to Pa. isarog Huber sp.nov.+ (Pa.dinagat Huber sp.nov.+ Pa.marilog Huber sp.nov.) but placed in a basal tetrachotomy.

Diagnosis
Males are easily distinguished from congeners in other species groups of Panjange by the unique processes of the genital bulb (embolus and appendix), pointing in opposite directions (Figs 18,64,69).
Females differ from members of the nigrifrons group by their long folded scapes (Figs 22,67,72) but are in general similar to females of the cavicola group.Other leaf-dwelling pholcids in the Philippines differ from Panjange as follows: Calapnita with longer abdomen that is monochromous and dorso-posteriorly not angular; male eyes on very low humps; embolus parallel to bulbal appendix.An undescribed group of leaf-dwelling Pholcus species differs from Philippine Panjange species as follows: dorsal process on procursus; embolus parallel to bulbal appendix; short male eye stalks without pointed processes; and epigynum without long folded scape.
Color.In life  mostly pale ochre-yellow to whitish, ocular area with brown mark, sternum whitish, legs with dark patellae and tibia-metatarsus joints; abdomen with dark brown to black marks dorsally and laterally, sometimes partly reddish.

Female
Similar to male but without eye stalks; eye triads on low humps (Fig. 46); clypeus and chelicerae unmodified; legs slightly shorter than in male.Epigynum weakly sclerotized, with more or less long, folded and extensible scape directed either toward posterior or toward anterior (e.g., Figs 22,36,72); sometimes widened at about half length (Figs 77,82).Internal genitalia with pair of pore plates of variable shape.

Monophyly and relationships
The monophyly of the Pa.lanthana group is consistently supported by two characters: the small ventral process distally on the proximal part of the procursus (character 17;Figs 25,70), and the unique direction of the embolus, pointing in an opposite direction than the appendix rather than lying parallel to it (character 22;Figs 18,64,69).
Entirely dubious is the position of Pa.bukidnon Huber sp.nov.(excluded from the analysis for reasons detailed above; see "Cladistic analysis").In preliminary analyses including Pa. bukidnon Huber sp.nov., it ended up at three different positions, indicated in Fig. 1 by asterisks: (1) as sister to the lanthana + cavicola groups; (2) as sister to the lanthana group; or (3) as sister to Pa. isarog Huber sp.nov.+ (Pa.dinagat Huber sp.nov.+ Pa.marilog Huber sp.nov.).It shares one of the synapomorphies of the lanthana group (embolus pointing in opposite direction than appendix) but not the other (ventral process on procursus).The parallel tips of the ocular spines (Fig. 90) remind of Pa.casaroro Huber sp.nov.+ Pa.camiguin Huber sp.nov.while the small rounded process between appendix and genital bulb (Fig. 91) reminds of Pa.dinagat Huber sp.nov.+ Pa.marilog Huber sp.nov.The procursus is highly autapomorphic.It seems that only the discovery of the unknown female and/or molecular data may solve this issue.

Natural history
Most species seem to have low abundances and very patchy distributions, with few specimens found within a very small area and none in surrounding areas of apparently similar vegetation.This may have been related to the low abundance and patchiness of suitable large leaves.Specimens were usually collected between about 50 cm above ground to about 2 m, and they are likely to occur also in higher strata of the forest, possibly in higher abundances.The web is a domed sheet, most of which is closely attached to the underside of a leaf.The extremely fine silk is poorly visible except when the sheet is viewed directly from the side.Egg sacs are only slightly elongated and covered by a barely visible layer of silk (Figs 3,8,12).In one case (Pa. camiguin Huber sp. nov.;Fig. 8), eight of ten eggs in an egg sac were parasitized by a parasitic wasp.

Composition
As construed here, the Panjange lanthana group now contains nine species (including the dubious Pa.bukidnon Huber sp.nov.).Judging from known distribution patterns and from the large number of poorly sampled islands, the group may well contain several times as many species.

Distribution
Known from the Philippines only (Fig. 16).Panjange lanthana -Huber 2011: 111-114, figs 436-58, 765-66, 786, 788 (♂♀).With the new species from the Philippines described below, the lanthana group can finally be characterized based on several species, and the emended diagnosis below takes account of these new species.

Natural history
All specimens were found in a shady ravine, with their domed webs connected to the undersides of green leaves among the vegetation, about 1-2 m above the ground.

Distribution
Known from two localities on southern Luzon Island (Fig. 16).

Etymology
The species name is derived from the type locality; noun in apposition.Color.Carapace pale ochre yellow with indistinct posterior median mark, ocular area and clypeus brown, black mark in AME area; sternum whitish; legs ochre-yellow with dark brown patellae and tibiametatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.Body.Habitus as in Fig. 5; ocular area raised, each triad on short stalk, medially with pointed process directed toward frontal and ventral (Fig. 17); carapace without median furrow; clypeus unmodified; sternum wider than long (0.65/0.55), unmodified.
CheliCerae.As in Fig. 17, with simple pair of weakly sclerotized processes in rather frontal position, without modified hairs; without stridulatory ridges.
PalPs.As in  proximal segments symmetric in shape but slightly larger on left side (e.g., length and diameter of right tibia about 95-97% of left tibia); coxa with strong ventro-distal rim but otherwise unmodified; trochanter with pointed retrolateral apophysis and rounded ventral hump; femur with dorsal hump and weakly sclerotized semitransparent process retrolatero-ventrally; tarsus with long whitish elongation with terminal tarsal organ; procursi of left and right palps strongly different, distal part of procursus clearly hinged in left palp, not hinged in right palp; bulb slightly smaller in right palp, with long processes extending in opposite directions (dorsal embolus; ventral appendix) almost symmetric (slightly longer in left palp).

Male (variation)
Not seen.Tibia 1 missing in other male.

Female
In general similar to male but eye triads on low humps and closer together (distance PME-PME 210 µm).Tibia 1 in 3 females: 6.9, 7.0, 7.9.Epigynum large elongate plate, anterior part weakly sclerotized, with symmetric or at most slightly asymmetric internal structures visible through cuticle; posterior part orange, with extensible scape with distinct transversal folds and small median process ("knob") at tip (Fig. 22); internal genitalia as in Fig. 23.

Natural history
The spiders were found in very low abundance on the undersides of leaves about 50-200 cm above the ground in a disturbed part of the forest.

Distribution
Known from type locality on Mindanao Island only (Fig. 16). ) also by very long eye stalks and contiguous tips of male ocular processes (Fig. 28); from Pa. camiguin Huber sp.nov.also by proximally unmodified male chelicerae (densely covered with small scales in Pa.camiguin Huber sp.nov.; cf.Fig. 47) and by small process between eye stalks (Fig. 28).

Etymology
The species name is derived from the type locality; noun in apposition.Color.Carapace and clypeus pale ochre to whitish, only base of eye stalks dark; sternum whitish; legs ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen pale gray, with black marks dorsally, monochromous ventrally.

PHILIPPINES
Body.Habitus as in Figs 6-7; ocular area raised, triads on long stalks with further distal processes whose contiguous tips seem to form a functional unit (Fig. 28); with small process between eye stalks; carapace without median furrow; clypeus unmodified; sternum wider than long (0.65/0.55), unmodified.
PalPs.As in ; proximal segments symmetric in shape but slightly larger on left side (e.g., length and diameter of right tibia about 90-95% of left tibia); coxa with strong ventro-distal rim but otherwise unmodified; trochanter with long weakly sclerotized retrolateral process directed toward dorsal; femur with weakly sclerotized finger-shaped process retrolatero-ventrally and more heavily sclerotized apophysis prolatero-dorsally; tibia with retrolateral trichobothrium in rather proximal position; tarsus with long whitish elongation, distally more strongly widened (club-shaped) in left palp; procursi of left and right palps strongly different, in each case with ventral process arising from proximal part (simple in right palp; bifid in left palp); distal part of procursus clearly hinged in left palp, not hinged in right palp; bulb much smaller in right palp, long processes extending in opposite directions (dorsal embolus; ventral appendix) almost symmetric (slightly longer in left palp).

Female
In general similar to male but eye triads on low humps and much closer together (distance PME-PME 265 µm); with dark mark in place of AME.Tibia 1 in 4 females: 6.7, 7.0, 7.1, 7.4.Epigynum mostly weakly sclerotized, with apparently slightly asymmetric internal structures visible through cuticle; distinctive scape with small median process at tip and two long extensible projections curved toward dorsal (Figs 29,33,34); internal genitalia as in Fig. 30.

Natural history
The spiders were found on the undersides of large leaves about 50-100 cm above the ground.

Etymology
The species name is derived from the type locality; noun in apposition.
Color.Carapace pale ochre with darker triangular mark posteriorly, pair of dark marks lateral of base of eye stalks, and median mark in AME area; clypeus without pattern; sternum whitish; legs ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen pale gray, with black marks dorsally, monochromous ventrally.
CheliCerae.As in Fig. 47, with pair of curved, weakly sclerotized lateral processes, proximally densely covered with small scales; without modified hairs; without stridulatory ridges.
PalPs.As in Figs 38-39, 50-51; proximal segments symmetric in shape and apparently also in size (Figs 53-54); coxa with strong ventro-distal rim but otherwise unmodified; trochanter with distinctive pair of apophyses (retrolateral and dorsal; Figs 53, 54); femur with weakly sclerotized retrolateral process directed toward dorsal and short ventral process directed toward proximal (Figs 53, 54); patella very small; tibia large; tarsus with long whitish elongation; procursi of left and right palps slightly but clearly different, distal part of procursus hinged in left palp, not hinged in right palp ; bulb symmetric in shape, with long processes extending in opposite directions (dorsal embolus; ventral appendix), apparently slightly longer in left palp.legs.Without spines and curved hairs; few vertical hairs; retrolateral trichobothrium on tibia 1 at 3%; prolateral trichobothrium absent on tibia 1, present on other tibiae; tarsus 1 with > 20 pseudosegments, only distally fairly distinct.

Male (variation)
Tibia 1 in 11 other males : 8.6-10.1 (mean 9.5).Males from the three eastern sites on Bohol Island (Datag, Marawis, Nan-od) differ slightly but consistently from other specimens in the tip of the right procursus that has a distinct retrolateral process.Specimens from these localities are therefore assigned tentatively.

Female
In general similar to male but eye triads on low humps and much closer together (distance PME-PME 260 µm), without AME remnants (Fig. 46); posterior mark on carapace smaller, often medially divided; dark abdomen spots in some specimens very indistinct or even absent.Tibia 1 in 32 females: 6.4-8.2 (mean 7.1).Epigynum apparently symmetric, mostly weakly sclerotized, with strong internal transversal sclerite visible through cuticle; extensible scape with strong transversal folds, laterally light brown to orange, with small median process at tip (Figs 36,61); internal genitalia as in Fig. 37, also apparently symmetric.

Natural history
The spiders were found on the undersides of large leaves to which they attach fairly dense but barely visible sheets of silk.One female was collected on 5 Mar.2014 with a parasitized egg sac from which all eggs hatched on 12 Mar.2014 (two spiders and eight wasps).

Diagnosis
Easily distinguished from known congeners by extremely elongated male palpal elements and female epigynal scape 67).

Etymology
The species name is derived from the type locality; noun in apposition.Color.Carapace pale ochre-yellow with pair of light brown marks, ocular area dark brown, with black marks in place of AME, clypeus pale ochre to whitish; sternum whitish; legs pale ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen pale gray, with many black marks dorsally and laterally; ventrally with pair of small dark marks near book-lung covers.

Female
In general similar to male but eye triads on low humps, ocular area with narrow transversal dark mark including black marks in place of AME.Tibia 1: 6.8 (missing in other females).Epigynum pale, weakly sclerotized, with very long strongly folded and extensible scape (Fig. 67); internal genitalia as in Fig. 68.

Etymology
The species name is derived from the type locality; noun in apposition.
Color.Carapace pale ochre yellow to whitish, without posterior mark, ocular area and clypeus dark brown, with black mark in AME area; sternum whitish; legs ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.Pa. marilog Huber sp. nov. and Pa. dinagat Huber sp. nov. (cf. Figs 11,14); ocular area raised, each triad on long stalk, curved process arising from near PME with distinctive widened tip directed toward anterior (Fig. 71); carapace without median furrow; clypeus with distinct field of ~25 short spines, with slightly longer but not stronger hairs below black mark; sternum wider than long (0.70/0.55), unmodified.

Body. Habitus similar to
CheliCerae.As in Fig. 71, with pair of simple, weakly sclerotized processes in rather frontal position directed toward median; without modified hairs; without stridulatory ridges.
PalPs.As in Figs 69-70; symmetric; coxa with strong ventro-distal rim but otherwise unmodified; trochanter relatively long, with two simple retrolateral processes; femur short, with two short dorsal processes; tibia short and wide (width ~65% of length); tarsus with short whitish elongation with subterminal tarsal organ; procursus proximal part with pointed ventral process, distal part hinged, very complex, with distinctive dorsal transparent flap; bulb with strong proximal sclerite, with processes extending in opposite directions (dorsal embolus; ventral appendix with many small scales), with small retrolateral toothed process arising from proximal bulbal sclerite, without small round sclerite between appendix and bulb.legs.Without spines and curved hairs; few vertical hairs; retrolateral trichobothrium on tibia 1 at 2%; prolateral trichobothrium absent on tibia 1, present on other tibiae; tarsus 1 with > 30 pseudosegments, only distally distinct.

Distribution
Known from type locality on Luzon Island only (Fig. 16).

Etymology
The species name is derived from the type locality; noun in apposition.
Color.Carapace pale ochre yellow to whitish, without posterior mark, ocular area and clypeus dark brown, with black mark in AME area; sternum whitish; legs ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.
Body.Habitus as in Fig. 11; ocular area raised, each triad on long stalk, with pointed curved process arising from near PME and directed toward anterior (Fig. 76); carapace without median furrow; clypeus with two patches of modified hairs: slightly stronger hairs below black mark and distinct field of ~20 short spines more distally; sternum wider than long (0.65/0.50), unmodified.
CheliCerae.As in Fig. 76, with pair of simple, weakly sclerotized processes in rather frontal position directed toward median; without modified hairs; without stridulatory ridges.
PalPs.As in Figs 74-75; symmetric; coxa with strong ventro-distal rim but otherwise unmodified; trochanter relatively long, with simple retrolatero-dorsal process and complex retrolatero-ventral apophysis, distally curved toward prolateral, with two distinctive subdistal branches directed toward distal and retrolateral respectively; femur short, with two short dorsal processes and very indistinct prolateral hump; tibia very thick (width ~70% of length); tarsus with short whitish elongation with terminal tarsal organ; procursus proximal part with rounded ventral process, distal part hinged, with distinctive retrolateral process and simple flat prolateral process; bulb with strong proximal sclerite, long processes extending in opposite directions (dorsal embolus; ventral appendix), with retrolateral toothed process arising from proximal bulbal sclerite, with small round sclerite between appendix and bulb.legs.Without spines and curved hairs; few vertical hairs; retrolateral trichobothrium on tibia 1 at 2%; prolateral trichobothrium absent on tibia 1, present on other tibiae; tarsus 1 pseudosegments barely visible in dissecting microscope.

Natural history
The spiders were found on the undersides of large leaves about 0.5-2 m above the ground.

Distribution
Known from type locality on Dinagat Island only (Fig. 16).

Etymology
The species name is derived from the type locality; noun in apposition.Color.Carapace pale ochre yellow to whitish, without posterior mark, ocular area and clypeus dark brown, with black mark in AME area; sternum whitish; legs ochre-yellow with dark brown patellae and tibia-metatarsus joints; abdomen ochre-gray, with black marks dorsally, monochromous ventrally.
Body.Habitus as in Fig. 14; ocular area raised, each triad on long stalk, with pointed curved process arising from near PME and directed toward anterior (Figs 81,84,87); carapace without median furrow; clypeus with two patches of modified hairs: slightly stronger hairs below black mark and distinct field of ~20 short spines more distally ; sternum wider than long (0.70/0.60), unmodified.
CheliCerae.As in Figs 81 and 84, with pair of simple, weakly sclerotized processes in rather frontal position directed toward median; without modified hairs; without stridulatory ridges.
PalPs.As in Figs 79-80; symmetric; coxa with strong ventro-distal rim but otherwise unmodified; trochanter relatively long, with simple retrolatero-dorsal process and complex retrolatero-ventral apophysis distally curved toward prolateral, with two distinctive subdistal branches directed toward distal and retrolateral respectively; femur short, with two short dorsal processes and indistinct prolateral hump; tibia very thick (tibia width ~60% of length); tarsus with whitish elongation with terminal capsulate tarsal organ (Fig. 89); procursus proximal part with ventral process, distal part hinged, with distinctive retrolateral process and simple flat prolateral process; bulb with strong proximal sclerite, long processes extending in opposite directions (dorsal embolus; ventral appendix), with retrolateral process arising from proximal bulbal sclerite, with small round sclerite between appendix and bulb.legs.Without spines and curved hairs; few vertical hairs; retrolateral trichobothrium on tibia 1 at 2%; prolateral trichobothrium absent on tibia 1, present on other tibiae; tarsus 1 with > 30 pseudosegments, only distally fairly distinct.

Natural history
The spiders were found on the undersides of leaves about 0.5-1 m above the ground.

Distribution
Known from two localities on Mindanao Island (Fig. 16).

Diagnosis
Easily distinguished from known congeners (and other similar relatives currently in Pholcus) by modifications of male ocular area (Fig. 90; short stalks and distinctive processes arising from near PME, with contiguous pointed tips), and by morphology of male palp (Figs 91-92; highly distinctive procursus; shapes of bulbal processes).Female unknown.

Etymology
The species name is derived from the type locality; noun in apposition.

Other material examined
None.
Color.Entire spider mostly pale ochre-gray to whitish, except large dark brown mark in ocular area including clypeus, pair of indistinct dark marks posteriorly on carapace, few small dark spots on abdomen dorsally, and brown patellae and tibia-metatarsus joints.
CheliCerae.As in Fig. 90, with pair of simple, weakly sclerotized proximal processes in rather frontal position; each lamina with additional small tooth; without modified hairs; without stridulatory ridges.short processes (embolus and appendix) extending in opposite directions, without retrolateral process arising from proximal bulbal process, with small round sclerite between appendix and bulb.legs.Without spines and curved hairs; few vertical hairs; retrolateral trichobothrium on tibia 1 at 4%; prolateral trichobothrium absent on tibia 1, present on other tibiae.

Female
Unknown.

Distribution
Known from type locality on Mindanao Island only (Fig. 16).

Discussion
The Panjange lanthana group, previously monotypic (Deeleman-Reinhold & Platnick 1986; Huber 2011), is a diverse group apparently restricted to the Philippines and supported by two synapomorphies.The configuration of the male bulbal sclerites (embolus pointing in opposite direction than appendix) is unique among Pholcidae.A close relationship of the lanthana group with the widespread cavicola group (Sulawesi to Australia) is strongly supported by the cladistic analysis.However, the relationship of these two groups to the third species group (nigrifrons group on Borneo) remains unclear.The characters supporting the monophyly of Panjange do not appear particularly convincing: (1) the reduction of the uncus has occurred repeatedly within Pholcus (Huber 2011); and (2) the ridges on the procursus are barely visible or absent in some species.In fact, preliminary molecular data (A.Valdez-Mondragón, D. Dimitrov, B.A. Huber unpublished data) suggest Panjange might not be monophyletic.
Spider genital asymmetry is apparently exceedingly rare, and to our knowledge has been documented in only nine of currently 3959 extant spider genera (genus count from World Spider Catalog 2015).Most cases are in Pholcidae (independent origins in Mesabolivar González-Sponga, 1998, Metagonia Simon, 1893, andPanjange;Huber 1997bHuber , 2004Huber , 2006;;herein), and Oonopidae (presumably independent origins in Lionneta Benoit, 1979, Escaphiella Platnick & Dupérré, 2009, Aschnaoonops Makhan & Ezzatpanah, 2011, and Paradysderina Platnick & Dupérré, 2011;Burger 2010;Platnick & Dupérré 2009, 2011;Platnick et al. 2013).Further cases have been reported in Theridiidae (Asygyna Agnarsson, 2006;Agnarsson 2006) andOxyopidae (Oxyopes Latreille, 1804;Brady 1964Brady , 1975).An attempt to explain the rarity of this phenomenon in spiders (as opposed to the ubiquity in insects) was made in Huber et al. (2007).This explanation centers on the fact that one-sided mating positions are an option for neopteran insects with their unpaired intromittent organ but not for spiders with their paired palps.While this may explain the rarity in spiders, it does not explain those few cases where asymmetry did evolve.A previous idea, that spider asymmetry may usually originate in the female as a result of space constraints (Huber 2004(Huber , 2006;;Huber et al. 2007) may apply to certain taxa (e.g., Asygyna, Mesabolivar, Metagonia) but apparently not to others.In Oonopidae, most cases of asymmetry seem to be restricted to males.However, routine preparations of female internal genitalia may simply fail to reveal subtle asymmetries.In fact, the only oonopid case of female asymmetry was discovered by serial sectioning and SEM of internal structures (Burger 2010).The same may apply to the three Panjange spiders with male asymmetric genitalia described herein.In two of them (Pa. malagos Huber sp. nov.; Pa. camiguin Huber sp. nov.), the female genitalia appear symmetric; in the third (Pa.casaroro Huber sp.nov.), it is not clear if the apparent asymmetry is an artifact of preparation or not.Further anatomical study with adequate sample sizes is needed to evaluate female symmetry vs. asymmetry in these species.
Sexually dimorphic eye-stalks have been studied in much detail in stalk-eyed flies (e.g., Burkhardt & de la Motte 1988;Wilkinson & Dodson 1997;Panhuis & Wilkinson 1999;Baker & Wilkinson 2001).For these insects, the general consensus is that sexual selection in the context of a resource-defense mating system has been the driving force in the repeated evolution of exaggerated male eye stalks, both by male-male competition and by female choice.In linyphiid spiders, modifications of the male "head" are apparently always accompanied by glands that discharge their products through pores easily visible in the SEM (Schaible et al. 1986).Sexual selection by female choice (via gustatorial courtship) is the most plausible explanation for this group of spiders (Vanacker et al. 2003;Kunz et al. 2012), but the few species with real eye stalks (such as Walckenaeria acuminata Blackwall, 1833) have never been studied in detail.The same is unfortunately true of certain theridiid spiders with similar long eye stalks [e.g., Phoroncidia longiceps (Keyserling, 1886); see Levi 1964].In Pholcidae, nothing is known about the use of the highly sexually dimorphic eye stalks.Glands and pores have been found in the male eye turrets of Modisimus culicinus (Simon, 1893) (Huber 1997a) but in none of the Panjange spiders studied herein and in none of the stalk-eyed Pholcus studied previously with the SEM (Huber 2011).Neither fights nor matings have ever been observed in any Panjange species or in any stalk-eyed Pholcus.The facts that these spiders were hard to find at all localities and that the low abundance seemed to result from the rarity of suitable large leaves suggest that a resource-defense mating system like in stalk-eyed flies is a plausible working hypothesis for future studies.Elongated genital structures in spiders are widespread, but it is usually the male intromittent structures and the corresponding female internal ducts that become long and exaggerated.In most cases, this concerns the sperm transfer organ (the embolus) and the corresponding insemination ducts (reviewed in Eberhard & Huber 2010).However, other intromittent structures may go through analogous developments, as for example male procursus and female uterus externus folds in Mesabolivar yuruani (Huber, 2000) (see Huber 2006;under Kaliana yuruani); or male tibia apophysis and a novel female tubular system in Hortipes gigapophysalis Jocqué, Bosselaers & Henrard, 2012(Jocqué et al. 2012).In Filistatidae, it is the basal palpal segments rather than the intromittent embolus that may become elongated (Marusik & Zamani 2015).In this group, cannibalism might appear a plausible explanation because the long palp allows the male body to remain out of reach for the female jaws (e.g., Gerhardt 1928, fig.2).However, observations have suggested female cooperation rather than aggression (Gerhard 1928(Gerhard , 1930)).In Panjange spiders, elongation likewise involves male structures that are very likely not inserted into the female.Instead, the palpal elongation in this group appears related to the extensible scape in the female.Both in Calapnita and in Panjange, the species with the longest male palps also have the longest female scapes (Calapnita semengoh Huber, 2011;Pa. hamiguitan Huber sp. nov.).This suggests that males and females may be engaged in an arms race (cf.Kuntner et al. 2009), with ever longer scapes requiring ever longer palps.The fact that pholcid chelicerae contact the female during copulation in all cases studied (reviewed in Huber 1999) suggests that palpal elongation is not related to cannibalism avoidance by keeping the body far from the female.In sum, a wide variety of factors may lead to genital elongation in spiders, but direct functional observations are almost entirely missing.01--0000000-000-0001000--0 Pholcus phalangioides 01--0000000-000-0000000--0 Pholcus debilis 01--0000000-000-0000000--0 Pholcus phui Arachnida Cuvier, 1812 Order Araneae Clerck, 1757 Family Pholcidae C.L. Koch, 1851 Genus Panjange Deeleman-Reinhold & Deeleman, 1983 Panjange Deeleman-Reinhold & Deeleman, 1983: 123-124.Type species by original designation: Panjange lanthana Deeleman-Reinhold & Deeleman, 1983.

Fig. 1 .
Fig. 1.Strict consensus of two most parsimonious cladograms of Panjange Deeleman-Reinhold & Deeleman, 1983 resulting from analyses of the matrix in Appendix 1 using equal character weights, successive weighting, and implicit enumeration.Only unambiguous character changes are shown.A = clade with asymmetric male palps; * = positions of Pa.bukidnon Huber sp.nov. in preliminary cladistic analyses.See Cladistic analysis section for further details.