Psyllipsocus stupendus sp. nov. Figs 1-4 Holotype: MHNG; male (slide-mounted); Mexico, Morelos state, Biosphere Reserve Sierra de Huautla, 2.5 km N and 4 km W of the Centro de Educación Ambiental e Investigación Sierra de Huautla (CEAMISH), 940 m; 12.-17.iii.1996, Malaise trap 2; 18˚ 27’.671 N, 99˚ 02’.475 W [right wings and left maxillary palp lacking]. Paratypes: MHNG and CNIN (slide-mounted or in alcohol). – 6 males, 6 females (one of them allotype); same locality as for holotype; ii.-vi.1996, Malaise traps. – 1 female, 1 nymph; same locality, on rock outcrop; 9.vi.1996, leg. T. Martínez. – 1 female; Mexico, Morelos state, Biosphere Reserve Sierra de Huautla, Municipio de Tepalcingo, El Limón de Cuauchichinola, 1200 m, on rock outcrop; 11.v.1996, leg T. Martínez. Diagnosis: This new species can be distinguished from all known species of the genus Psyllipsocus by the very particular male and female genitalia, the general structure of which is already observable in nondissected specimens in alcohol. Etymology: The species epithet (Latin: stupendus, -a, -um) refers to the astonishing male and female genitalic structures. Description: General colouration yellowish to medium brown. Head without particular facial markings, compound eye brown, maxillary palp uniformly light to medium brown, antenna light brown. Legs: femur light brown, tibia medium brown with whitish sub-basal zone, tarsus medium brown. Forewing hyaline with more or less intense brown markings along several veins in apical half (Fig. 1 A), hindwing hyaline (Fig. 1 B). Abdomen whitish, basally and laterally with some brown hypodermal pigment, terminalia brown. Both sexes macropterous (Fig. 1 AB). Forewing: Rs and M fused for a length; distal closed cell slightly shorter than marginal length of pterostigma and much shorter than basal closed cell (bcc/dcc ≈ 3.6); first portion of pterostigmal R1 about equal in length to R1-Rs crossvein; CuA1 strongly curved, AP relativly short and high. Hindwing: Basal portion of Rs not differentiated and R1 originating from Rs-M fusion, thus closed cell triangular. Three ocelli present. Pilosity of frons and vertex almost uniform. Antenna with at least 23 segments (most antennae damaged); antennal flagellomeres with uneven surface (due to insertion points of long and relatively thick setae), in basal half of antenna maximal length of flagellar hairs about 5x greatest width of their flagellomeres; pedicellar microspades organ weakly developed (with only one unit). P2 with a weakly differentiated stout sensillum in basal half (somewhat thicker than other setae of similar length); P4 slender hatchet-shaped, with a prominent subapical sensillum (Fig. 1 F). Lacinial tip as in Fig. 1 E. Pretarsal claws simple, symmetrical, with a small preapical denticle; hind legs with well-developed coxal organ. Epiproct and paraproct simple in both sexes (Fig. 1 G), anal spine long, setal organ consisting of two fine setae of about equal length, but the ventral seta somewhat thicker than the dorsal one, paraproctal sensorium with 6-7 fine trichobothria on basal florets and one normal seta. Male genitalia: Hypandrium and phallosome forming a ventrally bulged structure of about half the length of the abdomen, laterally articulated to the sclerotized rod-like antero-lateral margin of the clunium (Fig. 1 C). Hypandrium with a distinct posterior lobe (Fig. 1 D); some long hairs on the lobe and near its base, remaining pilosity shorter (pilosity not shown in the figures); near base of the posterior lobe a pair of small dorsal hooked sclerites (Figs 1 D, 2B), probably linked by ligaments to the phallic cradle and/or to the endophallic tube. Phallic cradle differentiated as a well-sclerotized frame on inner wall of hypandrium (Fig. 2 A). Phallosome complex (Figs 1 D, 2C), with a pair of slender, movable, slightly curved and acutely pointed claspers; endophallic tube basally on each side with a half-moon shaped plate and distally tapering towards a pair of narrowly rounded and slightly curved tips; a pair of posteriorly directed basal struts inserted medially near anterior margin of hypandrium (forming a posteriorly opened V in ventral view; see Fig. 1 CD, interrupted lines), each strut linked to phallic cradle by a less sclerotized foliaceous structure (Fig. 2 C, dotted lines) attached on a small lateral denticle situated in about one third of its length. Female genitalia: Complex sclerotized structures of v1, v2, vagina and spermapore plate readily visible through the largely membranous subgenital plate in ventral view (Fig. 1 G). Subgenital plate membranous dorsally, at most weakly sclerotized ventrally, roughly triangular, bearing hairs of medium length on ventral surface but no particularly long setae on hind margin. Gonapophysis v3 typical for the genus (Fig. 3 A), almost rectangular, weakly sclerotized, bearing some stouter laterodistal setae (somewhat thicker than other setae of same length); v1 and v2 well-developed, elongated and strongly sclerotized (Fig. 3 AB), basally widened and connected to the sclerotized and longitudinally folded dorsal wall of the vagina (Fig. 3 B). Posterior part of this sclerotized vaginal tube laterally articulated to its anterior part by a joint-like connection (x in Fig. 3 B). Dorsal wall of the genital chamber transversally double-folded, resulting in a ventral main compartment, dorsally delimited by the sclerotized vaginal wall, and a dorsal secondary compartment, dorsally delimited by the spermapore plate (Fig. 4 B). Spermapore plate elongated and wellsclerotized, with a longitudinal ventral bulge on each side covering a region of glandular tissue (Fig. 3 B, interrupted lines); spermapore opened into an ovally semicircular sclerite at the anterior end of the spermapore plate (Figs 1 H, 3B), situated proximally in the dorsal compartment of the genital chamber (Fig. 4 B). Spermatheca thinwalled (Fig. 1 G), very fragile, its duct greatly coiled; spermatophore only observed in one female (Fig. 1 G: the structure figured may correspond to two simple spermatophores or to one more complex spermatophore). Dimensions: Male holotype: BL = 1.7 mm; FW = 1820 μm; FWw = 650 μm; FW/FWw = 2.8; HW = 1510 μm; F = 342 μm; T = 720 μm; t1 = 270 μm; t2 = 45 μm; t3 = 54 μm; IO/D = 1.41. – Female allotype: BL = 1.8 mm; FW = 1750 μm; FWw = 650 μm; FW/FWw = 2.7; HW = 1440 μm; F = 348 μm; T = 715 μm; t1 = 260 μm; t2 = 47 μm; t3 = 51 μm; IO/D = 1.44. Distribution: Mexico, Morelos state, mountains of the Sierra de Huautla Biosphere Reserve, on rock outcrops (but most known specimens collected in Malaise traps without indication of biotope or microhabitat). Type locality: Mexico, Morelos state, Sierra de Huautla Biosphere Reserve, 2.5 km N and 4 km W of the Centro de Educación Ambiental e Investigación Sierra de Huautla (CEAMISH), 940 m, 18˚ 27’.671 N, 99˚ 02’.475 W. Remarks: The new species seems to be related to a group of North and Middle American species characterized by distally elongated phallic structures with proximal insertion of a V-shaped pair of basal struts. The following species listed by Mockford (2011) show this type of male genitalia: P. albipalpus Mockford, 2011 (Mexico), P. apache Mockford, 2011 (Southwestern USA), P. flexuosus Mockford, 2011 (Guatemala), P. maculatus Garcia Aldrete, 1993 (Mexico, USA), P. oculatus Gurney, 1943 (Mexico, USA), P. poblanus Mockford, 2011 (Mexico), P. regiomontanus Mockford, 2011 (Mexico), P. squamatus Mockford, 2011 (Mexico), P. subterraneus Mockford, 2011 (USA: Texas). In these species the arms of the basal struts are anteriorly directed, as in all other species of the genus that have well-developed basal struts. In P. stupendus, the structures which are here considered as homologous to the basal struts (see description) are posteriorly directed (forming a posteriorly opened V), probably due to the presence of extremely elongated phallic sclerites. The brachypterous species P. kintpuashi Mockford, 2011, the male of which is not known, may also be related to P. stupendus. Its female has some stouter laterodistal setae on v3 and the spermathecal duct is greatly coiled, similar to P. stupendus. A tendency to sclerotization of the vaginal wall has never previously been observed in the genus Psyllipsocus, and in all known species v2 and v3 are relatively short and membranous or only weakly sclerotized. In spite of the striking autapomorphic genital characters of P. stupendus it is at present not justified to consider this species as a representative of a new genus. Concerning the non-genitalic structures, it fits well in the genus Psyllipsocus, and it is almost certain that this genus would become paraphyletic by a generic separation of P. stupendus. For further discussion see below. DISCUSSION In two of the three dissected females of P. stupendus, the genitalia were in normal resting position (see Figs 1 G, 4B), as in the other alcohol-preserved females of the present material. That of the female no. 8164 show an evaginated dorsal compartment of the genital chamber (Fig. 4 D), although this may be an artifact, possibly due to stress during fixation in alcohol. Without detailed observations of copulation and oviposition it is impossible to make a sound functional interpretation of the unusual genital morphology in P. stupendus. It may nontheless be of interest to present some preliminary hypotheses about the functional morphology of the complex male and female genitalic structures in this species. Fig. 4A shows the standard structure of female terminalia in Psocoptera. In some psocids the vagina (genital chamber) is subdivided by a membranous transversal fold of its dorsal wall in a ventral main compartment and a smaller dorsal compartment (e. g. Stenopsocus, see Badonnel, 1934). The spermathecal duct opens directly into the main compartment (Weber, 1931) or into the dorsal compartment near the distal part of the ventral compartment (Badonnel, 1934, 1951). For fertilization the sperm is released from the spermatheca and reaches the egg when it moves past the spermapore (Fig. 4A). In the female of P. stupendus the dorsal wall of the genital chamber is transversally double-folded; the ventral compartment is dorsally delimited by the sclerotized part of the vagina; the dorsal compartment is ventrally delimited by a thin membrane and dorsally by the elongate spermapore plate which bears proximally the opening of the spermathecal duct (Fig. 4 B). In the resting position, the distal end of the spermapore plate slightly surpasses the posterior end of the sclerotized vaginal wall (Figs 1 G, 4B). The situation after dissection of the terminalia (Fig. 4 C) and the observation of the evaginated dorsal compartment in the female no. 8164 (Fig. 4 DE) show that the membranous ventral wall of the dorsal compartment is attached to the distal end of the sclerotized part of the vaginal wall and entirely separated from the sclerotized dorsal wall of the ventral compartment. Thus, the position of the spermapore plate may vary during copulation and oviposition depending on the extent of evagination of the eversible dorsal compartment. Assuming that the egg passes through the ventral compartment while female genitalia are in resting position (Fig. 4 B), it cannot enter in direct contact with the spermapore for fertilization. In this case, the sperm released from the spermatheca would have to swim toward the distal end of the spermapore plate before reaching the egg, probably passing between the lateral bulges of the spermapore plate which contain glandular tissue (Fig. 3 B). Fertilization of the eggs seems therefore more complicated than in the standard situation, due to the presence of the sclerotized proximal part of the vaginal wall. The presence of a pair of strongly sclerotized, apically curved and pointed phallic claspers (Fig. 2 C), previously unknown in the genus Psyllipsocus, suggests that the sclerotized part of the vagina may have a protective function against traumatic effects of copulation, reminiscent of particular sclerotized structures associated with the vulvar area in Miridae (Heteroptera) which are interpreted as a defence system for the female against potential wounding by the male organ (Pluot-Sigwalt & Matocq, 2006). But the sclerotized part of the vagina may also provide new anchoring sites on which males can hold the mate by using their unique phallic claspers. Alternatively, it is also possible that the sclerotized vagina forms a tube-like functional unit together with the sclerotized first and second ovipositor valvulae. The presence of much reduced and weakly sclerotized v1 and v2 in all other species of Psyllipsocus suggests that these ovipositor valvulae may have a particular function in P. stupendus, possibly in combination with the adjacent sclerotized vaginal wall. In the genus Neotrogla Lienhard, 2010, belonging to the related trogiomorphan family Prionoglarididae, structures associated with the spermapore are inserted deep into the male genital chamber during copulation (Yoshizawa et al., 2014), which suggests the possibility of a somewhat similar mechanism in P. stupendus. The close contact between the female spermapore and the opening of the male seminal duct, indispensable for the transmission of the spermatophore in the suborder Trogiomorpha (Klier, 1956; Yoshizawa et al., 2014) may be obtained in P. stupendus by evagination of the dorsal compartment of the female genital chamber and the shallow insertion of the spermapore plate into the male genitalia. During copulation the phallic claspers may drag the extendable part of the female genitalia into the male body. An example of female genitalia shallowly inserted into the male abdomen in this way is known in the cricket Gryllus bimaculatus De Geer, 1773; in this species the female copulatory papilla, bearing the opening of the spermathecal duct, is protruded and enters the male genital cavity where genital coupling is achieved (Sakai & Kumashiro, 2004). Thus, the unique structures of male and female genitalia in P. stupendus may result from a reproductive biology which differs considerably from that of the other species of the genus. A special form of selective pressure based on sexual selection may be at the origin of the very particular genital structures in this species, while the natural (ecological) selection remains similar to most other species of Psyllipsocus, favouring the uniform general morphology (see the similar hypotheses for Zoraptera in Mashimo et al., 2014).