INSECTA MUNDI, Vol. 12, Nos. 1 & 2, March-June, 1998 81

Generic definitions and species assignments in the Family ()

Edward L. Mockford Department of Biological Sciences Illinois State University Normal, IL 61790-4120

Abstract: The family Epipsocidae is defined, with the addition of one character not previously used. The genus Di­ midistriata Li and Mockford is removed from Epipsocidae and tentatively placed in the family Dolabellopsocidae. The ge- nus Parepipsocus Badonnel remains unplaceable to family. Eleven genera are recognized within Epipsocidae. Bertkauia Kolbe, which had been synonymized with Epipsocus, is recognized as a valid genus. Definitions of genera based on the type species and seemingly close relatives result in 38 new combinations. Epipsocus Hagen, which previously held most of the species, now serves a dual function, with 15 species in the strict sense and 15 species 'incertae sedis' retained in it, pending further investigation. Goja Navas, previously with 2 species, now contains 10, with 8 transferred from Epip­ soc us. Mesepipsocus Badonnel, previously with 5 species, receives 24 more from Epipsocus and 1 from Dicropsocus. Eight species are transferred from Epipsocus to Epipsocopsis Badonnel. Notable consistency in geographic distributions of gen- era results from these transfers. Some evolutionary trends within the family are discussed. Epipsocus delicatus (Hagen) and E. completus Banks, which had been assigned by recent authors to the genus Pseudocaecilius Enderlein (Family Pseudocaeciliidae), are returned to the Epipsocidae and assigned to Epipsocopsis. A lectotype is designated for E. delica­ tus.

Introduction midistriata Li and Mockford (1997) was assigned The Family Epipsocidae is a group of 133 de- by its authors to the Epipsocidae. It probably be- scribed species, mostly tropical and subtropical in longs in the family Dolabellopsocidae because of distribution. This is 1 of 5 families defined by its labral characters. Eertmoed (1973) for the family group . Published keys to species of Epipsocus, s. lat. In this study, the definition of the Family in local faunas, such as that of New (1972) and Epipsocidae is reviewed and augmented. The gen- Mockford (1996) remain useful, but the user must era of this family are defined, and the described keep in mind that an author may have meant this species of each genus are listed. Prior to the pres- genus to include the entire family Epipsocidae at ent study, numerous species belonging in other the time, or an even broader set of species. genera had been placed in the genus Epipsocus The earlier literature contains a few errors Hagen (1866). In the present study 36 species are which have been perpetuated by subsequent transferred out of Epipsocus to other genera. The authors and require correction. Enderlein (1919) need for these transfers has arisen in part from placed Psocus delicatus Hagen in the genus Pseu­ genera having been defined initially on too few docaecilius. In 1991 I looked at the type material characters and in part from authors not having of this species, which consisted of 2 specimens on adhered strictly to existing definitions. The ulti- points in the Museum of Comparative Zoology, mate goal of this study is the establishment of Cambridge, Massachusetts. The species is clearly monophyletic taxa, but attainment of that goal an epipsocid. Both specimens bear label data must await further studies. Unfortunately, 15 spe- "Ceylon, type 10114". One, a female, is here desig- cies, about 11 % of the named species, cannot be nated lectotype and bears the additional label placed to genus on existing information. They are "Lectotype, Epipsocus delicatus, (Hagen), E. L. listed as incertae sed is but are retained for the Mockford Nov. 1991 (unpub.)". The other speci- present in Epipsocus. Thus, Epipsocus now has a men, a male, is here designated paralectotype. dual function. In the strict sense, it is a group of Smithers (1967) placed Epipsocus completus closely-related species, and in the broad sense, it is Banks in the synonymy of "Pseudocaecilius" deli­ a holding taxon for species of Epipsocidae incertae catus (Hagen). I also examined the unique type of sedis. The genus Parepipsocus Badonnel (1986) E. completus, and confirm it is a true epipsocid. I must be placed 'incertae sedis' at the family level do not believe that the synonymy with P. delicatus (as its author did) because of its extreme neoteny is justified. The wing markings of both species in- and lack of diagnostic characters. The genus Di- dicate placement in the genus Epipsocopsis. 82 Vol. 12, Nos. 1 & 2, March-June, 1998, INSECTA MUNDI

Materials and Methods Results and Discussion

The characters for each genus were abstracted 1. Definition and included genera of the from the type species and seemingly close rela- family Epipsocidae. This family is here regarded tives. These characters were obtained from de- as including all genera of the group Epipsocetae in scriptions in the literature, verified where possible which tarsi are 2-segmented; the antennal scape is with specimens at hand. These same methods membranous over much of its anterior (ventral) were used for assigning species to genera. A. N. surface, being well sclerotized only at the base Garcia Aldrete has kindly sent particulars about (Fig. 1); the pair of sclerotic rods of the labrum several species which I have not seen. No new (labral sclerites) run the entire length of the genera are proposed here, although some may be labrum and curve outward at the base to reach the found necessary in future studies of the group. sides of the labrum (Fig. 2); anteriorly the labral The characters which appear to be important sclerites are joined by only a weak sclerotic con- in distinguishing genera in this family are the fol- nection, if any; macropterous individuals have lowing: only one anal vein in the forewing; the pterostigma 1) lacinial tip: whether or not denticles are lacks crossveins; the first ovipositor valvula (vI) is present in the outer cusp, and the nature of denti- present or absent, and the third (v3) is represented cles when present; as a swelling or field of setae on the side of the 2) presence or absence of a row of cuticular second (v2) (Fig. 3). cones arising on setal bases on the fore and hind Type genus: Epipsocus Hagen (1866). Other femur; included genera: Bertkauia Kolbe (1882), Cu­ 3) presence or absence of a pre apical denticle bitiglabra Li (1995a), Dichoepipsocus Li and Mock- on the pretarsal claw; ford (1997), Dicropsocus Smithers and Thornton (1977), Epipsocopsis Goja 4) extent of development of wings in females; Badonnel (1955), Navas states are apterous, micropterous, brachypterous (1927), Heteroepipsocus Li (1995a), Hinduipsocus Mesepipsocus (rare), and fully winged; Badonnel (1981), Badonnel (1969), Odontopsocus 5) extent of development of multiple veins in Badonnel (1987). Rs and M, primarily in the forewing (this charac- Although all of these genera are probably not ter should not be used alone for recognition of gen- equivalent cladistically, they all appear to be use- era); ful in designating sets of related species. 6) nature of the Rs-M junction in the hind- 2. Definitions of genera and lists of in- wing; states are long fusion, short fusion, at a cluded species: point, or by a crossvein; 7) nature of the female sub genital plate: Epipsocus Hagen, 1866 whether or not the hind margin is rounded or ex- tended as a process, and nature of the process if Type species: Psocus auus Roesler, 1943 present; (replacement name for Psocus ciliatus Pictet- 8) presence or absence of vI (= ventral valvula) Baraban and Hagen, 1856, preoccupied), (only of the ovipositor and, if present, nature of its basal original species). attachment; Definition: Outer cusp of lacinial tip den- 9) nature of the composite v2+3 (= dorsal + lat- ticulate (as in Fig. 4). No row of cones on fore or eral valvulae): whether or not v3 forms a lobe on hind femur. Preapical denticle present on pretar- the side of v2 and nature of the distal process; sal claw. Adults of both sexes fully winged. Vena- 10) extent of development of external para- tion as in Figs. 5 and 6. Multiple veins extremely meres of the phallosome; states are absent, rudi- rare. Rs and M in hindwing fused for a distance. mentary, and well developed; Female sub genital plate rounded posteriorly. In 11) presence or absence of endophallic scleroti- ovipositor (Fig. 3) vI present, usually joined by a zations and their nature when present; sclerotic strip to clunium; v3 developed as a field of 12) nature of the anterior margin of the phal- setae on side of v2. Phallosome (Fig. 7) with exter- losome; states are membranous (open phallosome) nal parameres absent or vestigial, aedeagal arch and well sclerotized (closed phallosome). generally terminating in a slender process; endo- phallus without sclerotizations or these forming a INSECTA MUNDI, Vol. 12, Nos. 1 & 2, March-June, 1998 83 central mass; anterior margin of phallosome mem- ticulate (Fig. 4). No row of cones on fore or hind branous. femur. Pre apical denticle present on pretarsal Included species: (note: a question mark claws. Males fully winged, females completely ap- precedes the name here and subsequently where terous. No development of multiple veins except some doubt remains about the placement. In each an occasional extra Rs or M branch in forewing such case the name is annotated with my reason sometimes expressed as an unilateral anomaly. In for the placement): hindwing Rs and M fused for a distance. Oviposi- ?acanthus New, 1980, Brazil (Amazonas). Re- tor with vI present, generally based in membrane; tained here because it appears to be close to E. v3 a large bulge on side of v2; v2 terminating in a foliatus Mockford, which is known to belong very long, acuminate process. Phallosome (Fig. 8) here. membranous anteriorly, with broad, basally ar- ?argutus New, 1980, Brazil (Amazonas). Retained ticulated external parameres, no endophallic scle- here for same reason as noted for E. acanthus rotization. New (above). Included species: avus (Roesler), 1943, Baltic amber. crosbyana Chapman, 1930, U.s.A. badonneli Mockford, 1991, Brazil (Roraima). lepicidinaria Chapman, 1930, U.S.A. foliatus Mockford, 1991, Brazil (Roraima). loebli Badonnel, 1981, India latistigma Roesler, 1940, Brazil (Santa Catharina). lucifuga (Rambur), 1842, Europe. meruleus New, 1980, Brazil (Amazonas). remyi (Badonnel), 1966, Reunion lsI., new combi- ?pereirai Badonnel, 1974, Brazil (Mato Grosso). nation from Epipsocus. Retained here on basis of relatively few denti- reticularis Li and Mockford, 1997, China. cles in lacinial tip. Discussion: The genus Bertkauia was syn- petenensis Mockford, 1957, Guatemala. onymized with Epipsocus by Pearman (1935) when quurcus Roesler, 1940, Brazil (Santa Catharina). he ascertained that the wing venation of the male ?roraimensis Mockford, 1991, Brazil (Roraima). of B. lucifuga is like that of the type of Epipsocus. Retained here for same reason as for E. acan­ Several authors have accepted this synonymy, thus New (above). while others have not. The above definitions differ serenus Roesler, 1940, Brazil (Santa Catharina). sufficiently that it seems reasonable to accept uniformis New, 1972, Brazil (Mato Grosso). Bertkauia as a valid genus. This small genus is ?verrucosus New, 1980, Brazil (Amazonas). Re- primarily Holarctic in distribution, with one spe- tained here for same reason as for E. acanthus cies extending south in the mountains of eastern New (above). Mexico and one on the island of La Reunion in the willineri New, 1972, Brazil (Mato Grosso). Indian Ocean. Males of most species are exceed- Discussion: Details about the type species of ingly rare. Epipsocus are provided by Enderlein's (1911) and Cubitiglabra Li, 1995a Hagen's (1882, 1884) descriptions and figures of this amber fossil. Character states not known for Type species: C. quadripunctata Li, 1995a. the type (presence or absence of a row of cones on Definition: Lacinial tip broad but not den- the front and hind femur, nature of the attach- ticulate. Pretarsal claws lacking preapical denticle ment of vI, extent of development of v3, and the or with a minute one. Males fully winged (females characters of the phallosome) are provided by the unknown). In forewing Rs 4-branched, M 6- extant South- and Central American species as- branched. In hindwing Rs and M joined by a cross- signed to this genus. These are species which vein or fused a short distance. Phallosome scle- agree with the type in its known generic charac- rotized ('closed') basally, with well developed ex- ters. The genus appears to be restricted now to the ternal parameres, lacking endophallic sclerotiza- American Tropics but seems probably to have had tions. a much wider distribution in the past. Included species: quadripunctata Li, 1995a, China. Bertkauia Kolbe, 1882 polyphebia Li, 1995b, China. (= Lapithes. Bertkau, 1883) Discussion: The genus Cubitiglabra appears to be very close to the genus Heteroepipsocus. It Type species: B. prisca Kolbe, 1882=Psocus differs primarily by having extra veins in Rs and lucifugus Rambur, 1842, (only original species). M of the forewing. In the family Epipsocidae such Definition: Outer cusp of lacinial tip den- a difference, alone, seems unreliable as a generic 84 Vol. 12, Nos. 1 & 2, March-June, 1998, INSECTA MUNDI character, and it may be necessary to combine this vein complex has arisen in Rs and M of the genus with Heteroepipsocus when more material is forewing. known. The 2 genera appear to be restricted to southeastern Asia and nearby island groups. Epipsocopsis Badonnel, 1955

Dichoepipsocus Li and Mockford, 1997 Type species: E. machadoi Badonnel, 1955 (original designation). Type species: D. micropterus Li and Mock- Definition: Outer cusp of lacinial tip with few ford, 1997 (original designation). or no denticles, frequently with acuminate tip (Fig. Definition: Lacinial tip with short outer cusp 9). Pretarsal claws with preapical denticle. Front bearing few, short denticles. Pretarsal claws lack- and hind femora with a row of cones at bases of ing pre apical denticle. Females micropterous setae (Fig. 10) (row sometimes much reduced). (males unknown). Sub genital plate extended pos- Both sexes fully winged or females occasionally teriorly in a short, median tongue with truncate or brachypterous. Generally no tendency for multiple shallowly bifid apex. Ovipositor with vI present, veins. In hindwing Rs and M usually fused for a joined at base to base of v2 or to clunium; v3 distance (rarely, a short Rs-M crossvein). Oviposi- forming a decided swelling on side of v2. tor lacking vI; v3 represented only by a field of Included species: setae on v2. Phallosome with well developed ex- micropterus Li and Mockford, 1997, China. ternal parameres; anterior margin membranous; thimphuensis (New), 1978, Bhutan, new combina- endophallus unsclerotized. tion from Epipsocus. Included species: Discussion: Although this genus resembles angolensis (Badonnel), 1955, Angola. Transferred Hinduipsocus in shape of the sub genital plate, dif- from Epipsocus by Badonnel, 1969. ferences in head shape, structure of the lacinial apicalis New and Thornton, 1975, Malaysia. tip, and structure of the pretarsal claw seem to basalis New and Thornton, 1975, Malaysia. rule out the possibility of a close relationship. cameronensis New and Lee, 1991, Malaysia. cincta Badonnel, 1969, Gabon. Dicropsocus Smithers and Thornton, 1977 completa (Banks), 1916, Philippines, new combina- tion from Epipsocus. Type species: D. montanus Smithers and costalis (Banks), 1914, India (Assam), Indonesia. Thornton, 1977 (original designation). Transferred from Epipsocus by Thornton, Definition: Outer cusp of lacinial tip un- 1984. toothed, with acuminate tip. Pretarsal claws with delicata (Hagen), 1859, (Sri Lanka), new combina- preapical denticle. No row of cones on front or hind tion from Psocus. fem ur. Both sexes fully winged. In forewing Rs dubia (Karny), 1925. Sarawak. Transferred from and M multi-branched with at least one Rs branch Epipsocus by Thornton, 1984. and at least one M branch re-branching. In hind- fasciata Smithers and Thornton, 1977, New wing Rs and M fused for a distance. Ovipositor Guinea. lacking vI, with v3 forming only a field of setae on formosa Li, 1992, China, new combination from v2. Phallosome membranous anteriorly, with Epipsocus. broad external parameres; aedeagal arch broad- fumipennis (Banks), 1920, Philippines, transferred tipped; endophallus with only slight median den- from Epipsocus by New and Thornton, 1975. ticulation. greeni New, 1977, Sri Lanka. Included species: hakgalensis (New), 1977, Sri Lanka, new combina- complexus Smithers and Thornton, 1977, New tion from Epipsocus. Guinea. hyalina (Banks), 1920, Singapore, transferred montanus Smithers and Thornton, 1977, New from Epipsocus by New and Thornton, 1975. Guinea. longiceps (Enderlein), 1926, Java, new combina- rugosus Smithers and Thornton, 1977, New Brit- tion from Epipsocus. ain. machadoi Badonnel, 1955, Angola. Discussion: Dicropsocus appears to be an off- maclurei New and Thornton, 1975, Malaysia. shoot of Epipsocopsis, in which the cones of the macrostigma Smithers and Thornton, 1977, New front and hind femur have been suppressed (or Guinea. were absent in the parental form), and a multi- maculata New and Thornton, 1975, Malaysia. INSECTA MUNDI, Vol. 12, Nos. 1 & 2, March-June, 1998 85 magna (New and Thornton), 1975, Malaysia, new Included species: combination from Epipsocus~ auiceps (Badonnel), 1986, Colombia, new combina- mouldsi Smithers, 1976, Australia. tions from Epipsocus. ?murcus (Enderlein), 1903, Malaysia, new combi- bogotana (Roesler), 1940, Colombia, new combina- nation from Epipsocus. The figure of the fe- tion from Epipsocus. male genitalia accompanying the original de- cubitalis (Mockford), 1996, Venezuela, new combi- scription seems to place the species here. nation from Epipsocus. murphyi New and Thornton, 1975, Malaysia. ditata Navas, 1927, Costa Rica. nebulifera Smithers and Thornton, 1977, Solomon molinai (Williner), 1949, Bolivia, new combination Islands. from Epipsocus. nubilipennis Karny, 1925, Sarawak. nebulosa (Roesler), 1940, Brazil (Santa Catharina), obuduensis New, 1973, Nigeria. new combination from Epipsocus. paraselena Vaughan et al., 1989, Indonesia pechi (Williner), 1949, Bolivia, new combination (Krakatau, Java). from Epipsocus. peradenayense New, 1977, Sri Lanka. ?picta (Banks), 1920, Brazil, new combination from prominens (Banks) 1937, Philippines, new combi- Epipsocus. Placed here because it appears to nation from Epipsocus. be close to G. plaumanni (Roesler), which is punctata Smithers and Thornton, 1977, New known to belong here. Guinea. plaumanni (Roesler), 1940, Brazil (Santa Ca- sclerota New and Thornton, 1975, Malaysia. tharina), new combination from Epipsocus. selena New and Thornton, 1975, Malaysia. semiaptera Mockford, 1996, Venezuela. singaporense New and Thornton, 1975, Singapore. Discussion: The genus Goja was based on a spatulata Smithers, 1964, Madagascar. species with multiple veins in Rs and M of both stuckenbergi Smithers, 1957, Madagascar. the forewing and the hindwing. The genus is held taprobanensis (New), 1977, Sri Lanka, new combi- together by an array of characters, and within it nation from Epipsocus. are species with normal venation, i.e., no extra thailandensis New, 1973, Thailand. veins in either fore- or hindwing. This genus ap- truncatula Badonnel, 1967, Madagascar. pears to be restricted to the American Tropics and uilhenai Badonnel, 1955, Angola. south to about 40° in Chile. Discussion: Epipsocopsis is restricted to the Old World Tropics but is widely distributed there. Heteroepipsocus Li, 1995a Species occur from West Africa east to the Philip- pines, with species on Madagascar and northern Type species: H. longicellus Li, 1995a Australia. As noted above, it apparently gave rise (original designation). to the genus Dicropsocus in the New Guinea re- Definition: Lacinial tip slender, non- gion. denticulate. Pre tarsal claws lacking preapical den- Goja Navas, 1927 ticle. Both sexes fully winged (? see note under H. inornatus [Banks], below). Venation normal, with Type species: G. ditata Navas, 1927 (only no extra veins in Rs or M. In hindwing Rs and M original species). joined by a short crossvein or at a point. Ovipositor Definition. Lacinial tip with a few large denti- with vI present, slender, attached at base to cles in outer cusp. Pre tarsal claws with pre apical clunium; v3 represented only by field of setae on denticle. Fore and hind femora lacking rows of v2; v2 terminating in extremely slender process. cones. Males fully winged, females micropterous. Phallosome with anterior margin sclerotized In male fore- and hindwing venation either normal (closed), external parameres well developed; endo- or with multiple veins in Rs and M. In hindwing phallus without sclerotizations. Rs and M joined either by a crossvein, at a point, Included species: or by a very short fusion. Ovipositor with vI pres- breuicellus Li, 1995a, China ent, either based in membrane or joined to ?inornatus (Banks), 1916, Philippines, Indonesia, clunium by a sclerotic strip; v3 forming a bulge on new combination from Epipsocus. If the figures side of v2. Phallosome (Fig. 11) with anterior mar- by Vaughan et al. (1989) truly represent this gin membranous, paired lateral endophallic scle- species, the wing-venational details suggest rotizations, no external parameres, or these very placement here. small and partially sunk into endophallus. longicellus Li, 1995a, China. 86 Vol. 12, Nos. 1 & 2, March-June, 1998, INSECTA MUNDI

Discussion: As noted in the discussion of Cu- terior margin, with rudimentary or no external bitiglabra, a close relationship seems to exist be- parameres; endophallus without sclerotizations. tween these 2 genera. The Rs-M crossvein in the Included species: hindwing, seen also in Goja, probably does not in- andrewsi Turner, 1975, Jamaica, new combination dicate a close relationship to that genus. Several from Epipsocus. other characters separate them. Regardless of the antillanus Banks, 1924, Jamaica, new combination accuracy of determination of the species deter- from Epipsocus. mined as Epipsocus inornatus Banks from arborescens New and Thornton, 1988, Peru, new Krakatau by Vaughan, et al. (1989), it seems to combination from Epipsocus. belong here and thus to provide female characters bordoni (Badonnel), 1987, Venezuela, new combi- for the genus. nation from Epipsocus. brazilianus (New), 1972, Brazil (Mato Grosso), Hinduipsocus Badonnel, 1981 transferred by Badonnel, 1974 braziliensis (New), 1980, Brazil (Amazonas), new Type species: H. annulipes Badonnel, 1981 combination from Dicropsocus (original designation). brevistigma (New), 1972, Brazil (Mato Grosso), Definition: Outer cusp of lacinial tip with transferred by Badonnel, 1974 numerous denticles. Pre tarsal claw with pre apical broadheadi Turner, 1975, Jamaica, new combina- denticle. No row of cones on fore or hind femur. tion from Epipsocus Females micropterous (males unknown). Subgeni- brunellus (New), 1972, Brazil (Mato Grosso), tal plate extended posteriorly as a median tongue transferred by Badonnel, 1974 bearing 2 slender processes at tip. Ovipositor with campanulatus (Thornton and Woo), 1973, Galapa- vI present, based in membrane; v3 represented by gos, new combination from Epipsocus. a field of setae on side of v2. capitulatus (New) 1980, Brazil (Amazonas), new Included species: combination from Epipsocus. annulipes Badonnel, 1981, India. clarus (Mockford), 1969, Mexico (amber), new atratus Badonnel, 1981, India. combination from Epipsocus. coleoptratus New, 1987, Nepal. fuscatus (New), 1972, Brazil (Mato Grosso), trans- hongkongensis Li and Mockford, 1997, Hong Kong. ferred by Badonnel, 1974 Discussion: As noted above, Hinduipsocus re- fuscivenatus (New and Thornton), 1988, Peru, new sembles Dichoepipsocus in sub genital plate, but combination from Epipsocus. differs in several other characters. It shows much icarus (Banks), 1941, Dominican Republic, new similarity to Bertkauia, differing in the known combination from Epipsocus. characters only in the presence of minute winglets latiphallus (New and Thornton), 1988, Peru, new and in the posterior extension of the subgenital combination from Epipsocus. plate. mobilis (Hagen), 1861, Cuba and Gabon Mesepipsocus Badonnel, 1969 newi Badonnel, 1974, Brazil (Sao Paulo) niger (New), 1972, Brazil (Mato Grosso), trans- Type species: M. grassei Badonnel, 1969 ferred by Badonnel, 1974 (=Psocus mobilis Hagen, 1861). obscurus New, 1972, Brazil (Mato Grosso), trans- Definition: Outer cusp of lacinial tip with ferred by Badonnel, 1974 numerous small denticles. Pre tarsal claws with ornatus (Mockford), 1974, Cuba, new combination preapical denticle. No row of cones on front or hind from Epipsocus. femur. Both sexes fully winged. Venation normal, peruanus New and Thornton, 1988, Peru, new rarely with extra veins, except 4 M veins dichoto- combination from Epipsocus. mously branched in forewing of several South proctus New and Thornton, 1988, Peru, new com- American and Antillean species, 2 South American bination from Epipsocus. species with 5-branched M, and 1 of these with roesleri New, 1972, Brazil (Mato Grosso), trans- extra veins in Rs and Cula. In hindwing Rs and M ferred by Badonnel, 1974 fused for a distance. Ovipositor lacking vl(a very roncadorensis New, 1972, Brazil (Mato Grosso), short, slender vI present in some undescribed spe- transferred by Badonnel, 1974 cies noted by A. N. Garcia Aldrete, in litt.); v3 usu- semiclarus (Mockford), 1991, Brazil (Roraima), ally a lobe on side of v2, but the lobe sometimes new combination from Epipsocus. not conspicuous. Phallosome membranous on an- INSECTA MUNDI, Vol. 12, Nos. 1 & 2, March-June, 1998 87

sinuatus New, 1972, Brazil (Mato Grosso), trans- Epipsocus conspersus Banks, 1914, India (Assam). ferred by Badonnel, 1974 Epipsocus fuscareolatus New, 1980, Brazil taitubai New, 1972, Brazil (Mato Grosso), trans- (Amazonas). ferred by Badonnel, 1974 Epipsocus hageni Banks, 1937a, Taiwan. tambopatensis New and Thornton, 1988, Peru, Epipsocus maculithorax New, 1980, Brazil new combinations from Epipsocus. (Amazonas). umbratus New and Thornton, 1988, Peru, new Epipsocus marginatus Enderlein, 1903, New combination from Epipsocus. Guinea. Discussion: This large assemblage of species Epipsocus nepos Enderlein, 1900, Peru. is primarily restricted to tropical America, except Epipsocus opticus New and Thornton, 1988, Peru. for one species found in Cuba and West Mrica. Epipsocus pennyi New, 1980, Brazil (Amazonas). Several species and species complexes within this Epipsocus phaeus New, 1980, Brazil (Amazonas). genus are very distinctive, and it may be useful to Epipsocus stigmaticus Mockford, 1991, Brazil break this genus into several genera at some fu- (Roraima). ture time. 3. General discussion Odontopsocus Badonnel, 1987 The Family Epipsocidae differs from the Fam- Type species: 0. orghidani Badonnel, 1987 ily Neurostigmatidae (Eertmoed, 1973) by only a (original designation). few known characters. These are the presence in Definition. Outer cusp of lacinial tip elongate, the latter Family of crossveins in the pterostigma, slender, with several denticles. Pretarsal claws Cula arising from the wing margin, and a series of lacking pre apical denticle. Front and hind femora tubercles on the preclunial abdominal terga. These lacking row of cones. Females apterous (males un- few characters confer such a striking difference in known). Sub genital plate not extended on poste- overall appearance between members of the 2 rior margin in middle. Ovipositor with vI present, Groups that it seems reasonable at present to re- based in membrane; v3 a conspicuous lobe on side gard them as different families. At present, how- ofv2. ever, we cannot assume a Sister-group relation- Included species: ship of these 2 Families. There is no information to badonneli Mockford, 1996, Venezuela. rule out the possibility that Neurostigmatidae has orghidani Badonnel, 1987, Venezuela. arisen from within Epipsocidae. Discussion: Odontopsocus shows a number of The reassignment of species from Epipsocus to similarities to Dichoepipsocus. In both genera the other Genera has resulted in greater consistency head is rounded, the outer cusp of the lacinial tip of geographic distribution shown by the Genera. is elongate and denticulate, the pretarsal claw Thus, all of the extant species of Epipsocus are lacks a preapical denticle, the winglets are devel- tropical American (although E. avus from the Bal- oped to the same extent, and vI is present. They tic amber suggests a wider distribution in Eocene- differ in the somewhat shorter, rounder head and Oligocene times). All of the species of Bertkauia shorter outer cusp of the lacinial tip in Dichoepip­ are Holarctic except for E. remyi on La Reunion socus, also in the posterior extension of the sub- Island; all of those of Epipsocopsis are Old World genital plate and in the nature of the basal at- tropical, etc. tachment of vI in that genus. Males of neither ge- There is not yet enough information to allow nus are known. A macropterous nymph of Odon­ one to determine polarity of character states. topsocus is at hand, but its sex cannot be deter- Therefore, we cannot yet propose a cladistic classi- mined. fication for the Family. Nevertheless, by compari- son with other Families of Group Epipsocetae and Species incertae sedis Suborder , it is possible to point out a few evolutionary trends within the family. These Epipsocus argentinus Badonnel, 1962, Argentina include the following. (described from a nymph). 1) A lacinial tip with broad outer cusp bearing Epipsocus atratus New, 1980, Brazil (Amazonas). numerous small denticles is seen in the Genera Epipsocus beguiristaini Williner, 1949, Bolivia. Epipsocus, Mesepipsocus, Goja, Hinduipsocus, and Epipsocus blandus New and Thornton, 1988, Peru. Bertkauia. It also is seen in the Families Cladiop- Epipsocus borgmeieri R. Karny, 1926, BraziL socidae, Ptiloneuridae, and Neurostigmatidae, as 88 VoL 12, Nos. 1 & 2, March-June, 1998, INSECTA MUNDI

well as in some members of the Families Asiopso- same is true for females of Dichoepipsocus and cidae and Caeciliusidae of the Group Caecilietae. Odontopsocus, 2 Genera which show enough other It is probably the plesiomorphous state in the similarities to suggest a common origin. In Hin­ Epipsocidae, and probably from it have been de- duipsocus, the winglets of females are mere sac- rived the narrower but still denticulate outer cusp like extensions of the thoracic terga with little or of Odontopsocus and Dichoepipsocus, the condition no basal articulation. Unfortunately, males remain in Epipsocopsis, wherein some species have a few unknown for these last 3 genera. In Bertkauia, large denticles and others none, and the condition with fully winged males, females are completely in the generic pair Cubitiglabra and Heteroepip­ wingless. socus, in which there is no trace of denticles and 5) Absence of vI is seen in the 2 genera Mese­ an outer cusp cannot be clearly separated from an pipsocus and Epipsocopsis. It has arisen, then, ei- inner cusp. ther once or twice (or more if Mesepipsocus repre- 2) A trend toward multiple Rs and M branches sents more than 1 genus). It would be tautological in the forewing is identifiable. Without much to suggest that these 2 genera are closely related doubt, the plesiomorphous condition for the Fam- based on a single shared character. Their relation- ily is a 2-branched Rs and a 3-branched M. This ship will have to be resolved on other still un- condition is called "normal venation" in the generic known characters. definitions and discussions above. A dichotomously 6) Reduction of the external parameres is seen 4-branched M has arisen at least once in the genus in several genera. Well developed external Mesepipsocus. More complex venation is present in parameres exist in the Genera Bertkauia, Cu­ M. taitubai, where Rs may be dichotomously 4- bitiglabra, Dicropsocus, Epipsocopsis (most spe- branched, a pterostigma-Rs crossvein is present, cies), and Heteroepipsocus. They are also found in and M may be up to 5-branched. A still more com- members of the other Families of Epipsocetae (see plex venation is seen in M. brasiliensis, in which figures of Eertmoed, 1973). They are either greatly there are, in addition to the complexities of M. tai­ reduced, sunk into the endophallus, or lost entirely tubai, a crossvein within the pterostigma, two in Epipsocus, Mesepipsocus, Goja, and at least one small closed cells below the pterostigma, a M-Cula species of Epipsocopsis (note Fig. 54 in New and crossvein, and a 2-branched Cula. In the genus Thornton, 1975). Such reduction has probably oc- Goja some species have normal venation, whereas curred in more than one evolutionary line in the others have up to a 4-branched Rs and 7-branched family. M in the forewing. The hindwing may also be af- 7) Changes in endophallic sclerotization have fected in this genus, with up to 4 branches of Rs occurred, but it is not possible at present to state and 5 of M. A complex system of multiple Rs and which direction they have gone. Well developed M veins is also seen in the forewing of Dicrop­ endophallic sclerotization is relatively rare in the socus, which has probably arisen out of normal Epipsocidae. It is restricted to the genus Goja and venation in Epipsocopsis. a few species of Epipsocus. Elsewhere in the Epip- 3) The Rs-M junction in the hindwing appears socetae it is seen in the Dolabellopsocidae, Pti- to be normally via a rather long fusion. This is loneuridae, and Neurostigmatidae, but not in the seen in all species of Epipsocus, Epipsocopsis (with Cladiopsocidae (Eertmoed, 1973, 1986). It is not at a few exceptions), Mesepipsocus, and males of all obvious that these structures are homologous Bertkauia. It is also seen in all of the other fami- from one family to another within the Epipsocetae lies of Group Epipsocetae. At least 3 times in the or between Epipsocus and Goja in the Epipsocidae. Epipsocidae an Rs-M crossvein has arisen in the In Goja it seems likely that some of them have hindwing, namely in the genus Goja, in the ge- arisen from the bases of the external parameres, neric pair Cubitiglabra and Heteroepipsocus, and which seems not to be the case in Epipsocus. in Epipsocopsis costalis and a few close relatives. 4) Wing reduction is seen in several lines. In Acknow ledgments Epipsocopsis a single species, E. angolensis, has brachypterous females in which venation persists I thank A. N. Garcia Aldrete and S. Sakaluk in the short wings. In Goja, males are fully winged for critical reading of the manuscript resulting, in while females are micropterous. The winglet is useful comments. articulated basally but shows no venation. The INSECTA MUNDI, VoL 12, Nos. 1 & 2, March-June, 1998 89

References Banks, N. 1941. New neuropteroid from the Antilles. Mem. Soc. Cub. Rist. Nat. 15:385- Badonnel, A. 1955. Psocopteres de l'Angola. PubL 402, plates 44, 45. Cult. Cia. Diamant. Angola, Lisbon 26:1-267. Bertkau, P. 1883. Ueber einen auffalenden Badonnel, A. 1962. Psocopteres. In Biologie de Geschlechtsdimorphismus bei Psociden nebst l'Amerique australe. VoL I. Etudes sur la Beschreibung einiger neuer Gattungen und Faune du SoL Eds. Centre Nat. Recherche Sci., Arten. Arch. fur. Naturg. 49:97-101. Paris (pp. 185-229). Chapman, P. J. 1930. Corrodentia of the United Badonnel, A. 1966. Sur quelques Psocopteres des States of America: I. Suborder Isotecnomera. iles Mascareignes. BulL Soc. Ent. France. J. New York EntomoL Soc. 38:219-290, 319- 71:234-238. 383, plates xii-xxi. Badonnel, A. 1967. Faune de Madagascar XXIII: Eertmoed, G. 1973. The phenetic relationships of Insectes Psocopteres. O.R.S.T.O.M. and the Epipsocetae (Psocoptera): the higher taxa C.N.R.S., Paris, pp. 1-237. and the species of two new families. Trans. Badonnel, A. 1969. Psocopteres de l'Angola et de Amer. Ent. Soc. 99:393-414. pays voisins, avec revision de types Africains Eertmoed, G. 1986. The redefinition of Cladiop­ d'Enderlein (1902) et de Ribaga (1911). Publ. socus (Psocoptera: Cladiopsocidae). Studies on Cult. Cia. Diamant. Angola, Lisbon 79:1-152. Neotropical Fauna and Environment 21:207- Badonnel, A. 1974. Sur quelques especes bresili- 229. ennes du groupe des Epipsocetae. Bull. Soc. Enderlein, G. 1900. Die Psocidenfauna Perus. Ent. France. 79:192-197. ZooL Jahrb., Abt. f. Syst., Geogr. u. BioL Badonnel, A. 1981. Psocopteres (Insecta: Psocop- Thiere. 14:133-139, plates 8,9. tera) de l'Inde. Mission Besuchet-Lobl (1978) Enderlein, G. 1903. Die Copeognathen des indo- et voyage entomologique LobI 1979. Rev. Su- australischen Faunengebietes. Ann. hist-nat. isse ZooL 88:381-411. Mus. Natl. Rung. 1:179-344, plates III-XIV. Badonnel, A. 1986. Psocopteres de Colombie Enderlein, G. 1919. Collections zoologiques du (Insecta, Psocoptera). Missions ecologiques du Baron Edm. de Selys Longchamps. Catalogue Professeur Sturm (1956 a 1978). Spixiana systematique et descriptif. Fasc. III (2e partie), 9:179-223. Copeognatha. Bruxelles, 55 pp. Badonnel, A. 1987. No. 17. Psocopteres du Vene- Enderlein, G. 1926. Die Copeognathenfauna zuela et de la Republique Argentine (pp. 173- J avas. Zool. Meded. 9:50-70 182). In V. Decu et al. eds., Fauna hipogea y Hagen, H. A. 1859. Synopsis der Neuropteren hemiedMica de Venezuela y de otros paises de Ceylons. Pars II. Verh. Zool.-Bot. Ges. Wien America del Sur. 1 (num. 1-22). Editura Aca- 9:199-212. demiei Republicii Socialiste Romania, Bucha- Hagen, H. A. 1861. Synopsis of the Neuroptera of rest. North America; with a list of South American Banks, N. 1914. Zoological results of the Abor Ex- species. Smithson. Misc. ColI. 4:i-xx, 1-347. pedition, 1911-1912. Neuropteroid insects. Re- Hagen, H. A. 1866. Psocinorum et Embidinorum cords ofthe Indian Museum 8:351-356, p1.xxv. synopsis synonymica. Verh. ZooL-Bot. Ges. Banks, N. 1916. Neuropteroid insects of the Phil- Wien 16:201-222. ippine Islands. Philippine Journ. Sci. (D) Karny, H. H. 1925. On the Copeognatha from Mt. 11:195-217, 2 pIs. Murud and Mt. Dulit, Sarawak. Sarawak Mu- Banks, N. 1920. New neuropteroid insects. Bull. seum Journal 3:63-74, plate 3. Mus. Compo Zool. 64:299-362, plates 1-7. Karny, R. 1926. Uma nova especie de Epipsocus Banks, N. 1924. Descriptions of new neuropteroid do Brasil. Publica<;6es do Museo nacional do insects. Bull. Mus. Compo Zool. 65:421-455, Rio de Janeiro 8:3-7. plates 1-4. Kolbe, H. J. 1882. Neue Psociden del' Paliiark- Banks, N. 1937a. Neuropteroid insects from For- tischen Region. Entomol. Nachrichten 8:207- mosa. Philippine Journ. Sci. 62:255-289, plates 212. 1-3. Li, Fasheng. 1992. A new genus and three new Banks, N. 1937b. Philippine neuropteroid . species of Elipsocidae and Epipsocidae Philippine Journ. Sci. 63:125-174, plates 1-6. (Psocoptera: Psocomorpha) from China. Ento- motaxonomia 14:197-202. 90 Vol. 12, Nos. 1 & 2, March-June, 1998, INSECTA MUNDI

Li, Fasheng. 1995a. Psocoptera. In Insects and New, T. R., and I. W. B. Thornton. 1975. Pso- macrofungi of Gutianshan, Zhejiang. Science comorpha (Psocoptera) from the Malayan and Technology Publishing House of Zhejiang Peninsula. Oriental Insects 9:375-418. Province, Hangzhou, 60-85. New, T. R., and I. W. B. Thornton. 1988. Epip- Li, Fasheng. 1995b. Psocoptera. In Insects of socetae (Psocoptera) from Peru. Studies on Baishanzu Mountain, Eastern China. Forestry Neotropical Fauna and Environment 23:225- Publishing House, Beijing (pp. 136-216). 250. Li, Fasheng, and E. L. Mockford. 1997. Two Pearman, J. V. 1935. Notes on some dimorphic new genera and four new species of Epipsoci- psocids. Ent. Mon. Mag. 71:82-85. dae (Psocoptera) from China. Oriental Insects Pictet-Baraban, F. and H. A. Hagen. 1856. Die 31:139-148. im Bernstein befindlichen Neuroptera del' Mockford, E. L. 1957. Some Psocoptera from Ti- Vorwelt (pp. 41-125). In G. Berendt. Die im kal, Guatemala. Ent. News 68:197-205. Bernstein befindlichen organischen Reste del' Mockford, E. L. 1969. Fossil insects of the order Vorwelt. Berlin. Psocoptera from Tertiary amber of Chiapas, Rambur, J. P. 1842. Histoire naturelle des insec- Mexico. Journal of Paleontology 43: 1267 -1273. tes. Neuropteres. Paris. xvii + 534 pp., 12 Mockford, E. L. 1974. Records and descriptions of plates. Cuban Psocoptera. Entomologica Americana Roesler, R. 1940. Neue und wenig bekannte 48: 103-215. Copeognathengattungen II. Zool. Anzeiger Mockford, E. L. 1991. New species and records of 130:1-25. Psocoptera (Insecta) from Roraima State, Bra- Roesler, R. 1943. Uber einige Copeognathen- zil. Acta Amazonica 21:211-318. genera. Stett. Ent. Zeit. 104:1-14. Mockford, E. L. 1996. New species and records of Smithers, C. N. 1957. Notes et descriptions sur Psocoptera from northern Venezuela. Tropical les Psocopteres de Madagascar. Naturaliste Zoology, Special Issue 2:1-98. malgache 9:273-280. Navas, L. 1927. Communicaciones entomol6gicas. Smithers, C. N. 1964. On the Psocoptera of 8. Soc6pteros del Museo de Hamburgo. Rev. Madagascar. Rev. Zool. Bot. Mr. 70:209-294. Acad. Cienc. Zaragoza 11:37-52. Smithers, C. N. 1967. A catalogue of the Psocop- New, T. R. 1972. Some Epipsocetae (Psocoptera) tera of the world. Austr. Zool. 14:1-145. from central Brazil. Trans. Roy. Entomol. Soc. Smithers, C. N. 1976. Epipsocopsis mouldsi sp. n. Lond. 123:455-497. representing a family (Psocoptera: Epipsoci- New, T. R. 1973. A collection of Psocoptera from dae) new to Australia. Austral. Ent. Mag. 3:32- Nigeria. Occ. Pub. Entomol. Soc. Nigeria 10:1- 34. 22. Smithers, C. N., and I. W. B. Thornton. 1977. A New, T. R. 1977. Epipsocidae and Pseudocaecilii- new genus and some new species of Epipsoci- dae (Psocoptera) from Sri Lanka. Oriental In- dae from the Melanesian arc. Proc. Linn. Soc. sects 11:409-420. New South Wales 102:60-75. New, T. R. 1978. Ergebnisse del' Bhutan- Thornton, I. W. B. 1984. Psocoptera from Wal- Expedition 1972 des naturhistorischen Muse- lace's Line: collections from the islands of Bali ums in Basel. Psocoptera. Entomol. Basiliensia and Lombok. Treubia 29:83-177. 3:67-86. Thornton, I. W. B., and A. Woo. 1973. Psocop- New, T. R. 1980. Epipsocetae from the Reserva tera of the Galapagos Islands. Pacific Insects Ducke, Amazonas. Acta Amazonica 10:179- 15:1-58. 206. Turner, B. D. 1975. The Psocoptera of Jamaica. New, T. R. 1987. Further Psocoptera from the Trans. R. Ent. Soc. Lond. 126:533-609. Nepal Himalayas, collected by the Martens Vaughan, P. J., I. W. B. Thornton, and T. R. Expeditions (Insecta). Cour. Forschungsinst. New. 1989. The Psocoptera of the Krakatau Senckenb. 93:353-358. Islands, Indonesia. Treubia 30:1-93. New, T. R., and S. S. Lee. 1991. Epipsocidae Williner, G. J. 1949. Corrodentios de Bolivia. Rev. (Psocoptera) from West Malaysia. Oriental In- Inst. Nac. Invest. Cienc. Nat., Buenos Aires sects 25:121-126. 1:95-126. INSECTA MUNDI, Vol. 12, Nos. 1 & 2, March-June, 1998 91

1

I~

1.0

9

8 I 7 /

Figs. 1-11. Fig. 1. Epipsocus sp., antennal scape; m = membranous area of anterior face. Fig. 2. Epip­ socus sp., labrum. Fig. 3. Epipsocus sp. ~,ovipositor valvulae. Fig. 4. Bertkauia crosbyana Chapman ~, lacinial tip. Fig. 5. Epipsocus sp. ~,forewing. Fig. 6. Epipsocus sp. ~,hindwing. Fig. 7. Epipsocus sp. a, phallosome. Fig. 8. Berthauia crosbyana Chapman a, phallosome. Fig. 9. Epipsocopsis sp. ~,lacinial tip. Fig. 10. Epipsocopsis sp. ~,base of front femur with cone-based setae. Fig. 11. Goja sp. a, phallosome.

Scale lines = 0.1 mm unless otherwise indicated.