J ouirnsLi or Hymenoptera esearch August 1995 ISSN #1070-9428 CONTENTS BANKS, D. Male nest defense in the digger wasp Cerceris binodis (Hymenoptera: Spheci- dae) 77 CARVER, M. Euryischomyia Girault (Hymenoptera: Chalcidoidea: Aphelinidae: Eriaporinae: EuryischiLni) 64 FINNAMORE, A. T. Revision of the world genera of the subtribe Stigmina (Hymenoptera: Apoidea: Sphecidae: Pemphredoninae), part 1 204 GESS, F. W. Descriptions of the male of Riekia nocatunga Richards, the male and two strik- ingly distinct sympatric colour forms of Riekia confluens (Snelling) and the male of Rolandia angulata (Richards) (Hymenoptera: Vespidae: Masarinae) from Aus- tralia 33 GESS, F. W., S. K. GESS and R. W. GESS. An Australian masarine, Rolandia angulata (Rich- ards) (Hymenoptera: Vespidae): nesting and evaluation of association with Good- enia (Goodeniaceae) 25 GOKHMAN, V. E. and D. L. J. QUICKE. The last twenty years of parasitic Hymenoptera karyology: an update and phylogenetic implications 41 HANSSON, C. Revised key to the Nearctic species of Chrysocharis Forster (Hymenoptera: Eulophidae), including three new species 80 HEYDON, S. L. A review of the North American species of Thinodytes Graham and Mauleus Graham (Hymenoptera: Pteromalidae) 1 LATTKE, J. E. Revision of the ant genus Gnamptogenys in the New World (Hymenoptera: Formicidae) 137 PULAWSKI, W. J. The wasp genus Tachytella Brauns, 1906 (Hymenoptera: Sphecidae) . . . 121 (Continued on back cover) INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991 OFFICERS FOR 1995 Donald L. J. Quicke, President James M. Carpenter, President-Elect Michael E. Schauff, Secretary John T. Huber, Treasurer Paul M. Marsh, Editor Subject Editors Eric Grissell, Robert Matthews, Wojciech Pulawski, David Rosen, Mark Shaw All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, Department of Biology, Imperial College at Silwood Park, Ascot, Berks SL5 7PY, England; President-Elect, Department of Entomology, American Museum of Nat- ural History, Central Park West at 79th Street, New York, New York 10024; Secretary, Systematic Entomology Laboratory, USDA, 7v National Museum of Natural History, NHB 168, Washington, D.C. 20560; Treasurer, Biological Resources Division, CLBRR, Agriculture Canada, K.W. Neatby Building, Ottawa, Ontario, Canada KIA 0C6; Editor, P. O. Box 384, North Newton, Kansas 67117. Membership. Members shall be persons who have demonstrated interest in the science of ento- mology. Annual dues for members are $25.00 (U.S. currency) per year, payable to The Interna- tional Society of Hymenopterists. Requests for membership should be sent to the Treasurer (ad- dress above). Journal. The Journal is published once a year by the International Society of Hymenopterists, % Department of Entomology NHB 168, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Members in good standing receive the Journal of Hymenoptera Research. Nonmember subscriptions are $50.00 (U.S. currency) per year. The Society does not exchange its publications for those of other societies. Please see inside back cover of this issue for information regarding preparation of manuscripts. Statement of Ownership Title of Publication: Journal of Hymenoptera Research. Frequency of Issue: Once a year (currently). Location of Office of Publication, Business Office of Publisher and Owner: International Society of Hymenopterists, % Department of Entomology, NHB 168, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Editor: Paul M. Marsh, P. O. Box 384, North Newton, Kansas 67117. Managing Editor and Known Bondholders or other Security Holders: none This issue was mailed 25 September 1995 J. HYM. RES. Vol. 4, 1995, pp. 1-24 A Review of the North American Species of Thinodytes Graham and Mauleus Graham (Hymenoptera: Pteromalidae) Steven L. Heydon Bohart Museum, Department of Entomology, University of California, Davis, CA 95616-8584 USA Abstract. — The Halticoptera-group is defined as containing those pteromaline pteromalid genera with a reticulate body, acarinate pronotal collar, weakly developed notauli, weakly delimited frenum, propodeum with the median carina and plicae connected posteriorly by a W-shaped carina, petiole with a basal flange, and the hind margin of the first gastral tergite sinuous laterally and usually emarginate medially. Genera included in this group are Halticoptera Spinola, Halti- copterina Erdos, Andersena Boucek, Thinodytes Graham, Syntomopiis Walker, Mauleus Graham, and Ploskana Boucek. Thinodytes and Mauleus are revised for the Nearctic region and keys to the world's described species are given. New species include T. caroticus n. sp., T. cyzicopsis n. sp., T. petiolatus n. sp., M. cultratus n. sp., M. iligneus n. sp., and M. venetus n. sp. Polycystus nigritus Howard is transferred to Mauleus as M. nigritus n. comb, and Gastrancistrus cephalon Walker is transferred to Thinodytes as T. cephalon n. comb. Bubekia fallax Gahan n. syn. is synonymized with T. cephalon Walker. INTRODUCTION are Notoglyptus Masi, Sphegigaster Spino- la, and Schimitschekia Boucek. The Halhcoptera-group is herein de- ^ ^^^ Thinodytes and Mauleus, the fined as containing those pteromaline North American species of all the genera pteromalid genera with: body reticulately ^^ ^^e Halticoptera-group occurring in the sculptured, pronotal collar acarmate (Figs Nearctic region were recently reviewed 21, 22), notauli weakly developed (Figs ^^ ^^e currently being studied. Andersena 21, 22), frenum weakly delimited (Figs 21, includes only one species (Boucek 1993). 22), propodeum with a sharp median ca- Anderson (1990) reviewed Halticopterina rina and the plicae connected posteriorly f^^ North America and is currently revis- by a W-shaped carina (Figs 17-20), peti- i^g Halticoptera. The world species of No- ole braced basally by an anteriorly di- togh/ptus and the Nearctic species of Si/n- rected lateral and ventral flange (Figs 17- tomopus and Sphegigaster were reviewed 20), and hind margin of first gastral ter- by Heydon (1988),' Heydon (1993), and gite sinuous laterally and usually emar- Heydon and LaBerge (1988), respectively, ginate medially (Figs 21, 22). Genera of This paper reviews the Holarctic species this group include Halticopitera Spinola, of Thinodytes and Mauleus. [Ploskana was Halticopterina Erdos, Andersena Boucek, reviewed by Boucek (1976).] Research by Thinodytes Graham, Syntomopus Walker, the author has revealed the presence of Ploskana Boucek, and Mauleus Graham, numerous species of both these genera in Three other genera that show many of the Central and South America, characteristics of the Halticoptera-group Thinodytes previously contained two and which probably are more or less described species, the Palearctic species closely related to the Halticoptera-^roup T. cyzicus (Walker) 1839, and the Carib- Journal of Hymenoptera Research Figs. 1-10. 1, Thinodytes caroticus n. sp., female hind leg; 2, Thinodytes cephalon (Walker), 2, female hind leg; 3, 4, Thinodytes cyzicopsis n. sp., female head dorsal view, 4, male antenna; 5, 6, Thinodytes petiolatus n. sp., 5, female head dorsal view, 6, male antenna; 7, Thinodytes cyzicus n. sp., female head dorsal view; 8, Mauleus cultratus n. sp., female habitus; 9, Mauleus iligneus n. sp., female antenna; 10, Mauleus venetus n. sp., female antenna. bean species T. clypeatus (Girault) 1918. To these I add the New World species T. cephalon (Walker) 1843, n. comb., and de- scribe four new Nearctic species — T. car- oticus n. sp., T. cyzicopsis n. sp., T. petiol- atus n. sp., and T. santerna n. sp. Mauleus was created for the species M. maderensis Graham, 1981 from Madeira. However, this species is associated with plants na- tive to Mexico and is suspected to be na- tive to the New World (Boucek and Ras- plus 1991). The New World origin of this Volume 4, 1995 Figs. 11-16. 11, Thinodytes petiolatus n. sp., female clypeus; 12, Thinodyies cephahn (Walker), female clypeus; 13, Maiik'iis iligueus n. sp., male clypeus; 14, Synlomopiis amerkanus Ashmead, female head dorsal view; 15, Maidens iligneiis n. sp., male head posterior aspect; 16, Halticoptera sp., male head posterior aspect. species is given more credence by the new Nearctic species — M. cultratus n. sp., presence of a described Neotropical Man- M. iligneus n. sp., and M. venetus n. sp. /('HS species, M. ///^^r/fj/s (Howard), 1897 n. These three new species are described comb, and the author's discovery of three herein. Journal of Hymenoptera Research Figs. 17-22. 17, Tliinodijtes ceplmlon (Walker), female propodeum and petiole; 18, Tlunodi/tes ci/zicopsis n. sp., male propodeum and petiole; 19, Tlii)iodi/tes petiolatiis n. sp., female propodeum and petiole; 20, 21, Maidens digneus n. sp., 20, male propodeum and petiole, 21, female habitus; 22, Syntomopus arpedes Heydon, female habitus. METHODS ^\^ jg used instead of clava. In addition, Terminology generally follows that of the gastral tergites are numbered T1-T7 be- Graham (1969), except that genal concav- ginning with the first tergite after the pet- ity is used instead of genal hollow and iole. The following abbreviations are used: Volume 4, 1995 the median ocellar diameter is MOD, the ocellar-ocular distance is OOL, the poste- rior ocellar distance is POL, the lateral ocellar distance is LOL, the multiporous plate sensilla are MPP sensilla, the lower ocular line is LOcL, and the antennal ha- nicular segments are Fl through F6. The measurements given in the descriptions can be converted to millimeters by multi- plying by 0.02. The acronyms for the mu- seums from which material was borrowed are listed in the acknowledgments section. TJiinodytes Graham Dicydus Thomson, 1876:221, 253. Type species: Miscogaster cyzicus Walker, 1839:200; by monotypy (examined). New name needed because of Dicydus Walker, 1833:371, 455. Thinodytes Graham, 1956:261. Type species: Miscogaster cyzicus Walker, 1839:200; by orig- inal designation (examined). Peck, Boucek, and Hoffer, 1964:41. Graham, 1969:150, 167. Hedqvist, 1975:167. Dzhanokmen, 1978:82. Boucek and Rasplus, 1991:32. Description. — Body color varying from almost wholly black to metallic green; scape metallic or nonmetallic. Head and mesosoma mostly alveolate except sculp- turing sometimes smooth on frenum (T. cyzicopsis, T. petiolatiis, and T. santenm) and median panels of propodeum (T. san- terna); petiole alveolate in species with quadrate to elongate petiole (Figs 18-20), smooth in species with transverse petiole (Fig. 17); gastral tergites nearly smooth. Clypeus variable, with either three asym- metrically arranged denticles (Fig. 11) (T. cyzicopsis, T. cyzicus, T. petiolatiis), three symmetrically arranged denticles (T. san- terna and T. clypeatus) or with single broad asymmetrically placed denticle (Fig. 12) (T. cephalon and T. caroticus). Head with short genal concavity often present; anten- nal torulus above LOcL. Antenna with for- mula 1:1:2:6:3; scape cylindrical, slender; length of flagellum plus pedicel of female less than head width (subequal in length in T. cyzicopsis), equal to or slightly greater than head width in male; funicular seg- ments cylindrical; MPP sensilla usually in single row, their length about equal to length of funicular segment (Figs 4, 6); fe- male club simple apically and with small patch of micropilosity on ventral side of terminal segment, except T. cephalon with terminal spine and large patch of micro- pilosity. Male maxilla with stipites unen- larged; palps slender (Fig. 15). Mesosoma arched dorsally; pronotum with collar very short medially, anterior edge round- ed; mesoscutum with notaulus incomplete or extending to hind margin of mesoscu- tum as impressed line; scutellum as long as wide, lacking anterior median sulcus, with two or three pairs of lateral setae (ex- cept T. santenm with many pairs), frenum not set off by sulcus; dorsellum a short transverse ridge; propodeum with width of median panels about 1.6X their length (Figs 17-19), plicae and median carina dis- tinct and connected posteriorly by W- shaped carina (Figs 17-19), basal fovea sometimes bordered mesally by short straight carina (Fig. 17) (T. clypeatus, T. cy- zicopsis, and T. cephalon), or by long sinu- ous carina (Figs 18, 19) (T. cyzicopsis, T. cyzicus, and T. petiolatiis); spiracles ovate. Fore wing with postmarginal vein about as long as marginal vein but postmarginal vein sometimes distinctly shorter (T. car- oticus and T. cephalon); stigma small, but its height half the distance between stigma and anterior wing margin in T. clypeatus; costal cell with one complete and two or three partial distal setal rows; basal cell bare except sometimes a few setae distally (T. clypK^atus and T. cyzicus); basal vein se- tose except in T. cephalon; speculum de- veloped and open posteriorly. Petiole braced basally by lateral and ventral an- teriorly directed lamellate flange (Figs 17- 19); petiole variable, sometimes trans- verse, unsclerotized ventrally, and with- out lateral setae (Fig. 17) (T. caroticus, T. cephalon, and T. santcrna), or quadrate to elongate and sclerotized ventrally (7. cly- peatus, T. cyzicopsis, T. cyzicus, and T. pe- tiolatiis); lateral setae sometimes present Journal of Hymenoptera Research (Fig. 19) (T. cyzicopsis and T. petiolatus); weak median carina present in T. clypea- tiis. Gaster of female ovate, acuminate api- cally, 1.4-1.8X as long as wide; hypopy- gium extending V2 to % gastral length; hind margin of Tl sinuous laterally and emarginate or straight medially. Discussion. — Giving a few characters to separate Thinodytes from the other genera of the Halticoptera-group, particularly Hal- ticoptera, Maiileus, and Si/ntomopiis, is im- possible because Thinodytes is what is left when the more distinct genera of the Hal- ticoptera-group are characterized. Apomor- phic characters among related genera com- mon to all Thinodytes species such as the reticulate body, weakly developed notauli, poorly delimited frenum, propodeum with sharp median carina and plicae connected posteriorly by W-shaped carina, petiole with a basal bracing consisting of an ante- riorly directed lateral and ventral flange, and the hind margin of the first gastral ter- gite being sinuous laterally and usually emarginate medially are the same charac- ters defining the Hnlticoptera-group itself. So presently, Thinodytes can only be de- fined negatively. There are many South American species, some described and some not, that fit within the present defi- nition of Thinodytes. Once these are inves- tigated, it may be possible to divide Thi- nodytes into monophyletic generic units. Halticoptera is distinguished from Thinod- ytes by a number of apomorphic character states including a bidentate clypeus, the antennal torulus located at or below the LOcL, the scape usually nonmetallic, the male maxilla with lamellately expanded palps and usually with another lobe on the stipites, and a median longitudinal carina on the petiole. Thinodytes has the clypeus with variable numbers and arrangements of clypeal denticles, but no known Thinod- ytes species has a bidentate clypeus. The antennal torulus in Thinodytes is located distinctly above the LOcL (except in T. pe- tiolatus), and the scape usually has metallic coloration. The male maxilla of Thinodytes lacks any expansion of the palps or any lobes on the stipites. Thinodytes clypeatus is the only species of Thinodytes that has a median carina on the petiole. Halticoptera species are commonly bright metallic green; those of Thinodytes are usually dark. Mauleus is distinguished from Thinodytes by the bidentate clypeus (Fig. 13), the dor- sum of the mesosoma as high as the vertex (Figs 8, 21), the median panels of the pro- podeum short (2X as wide as long) (Fig. 20), and the lateral flanges of the petiole enlarged and thickened (Fig. 20). In Thi- nodytes, the clypeus has various numbers and arrangements of clypeal teeth, but is never bidentate, the dorsum of the meso- soma is distinctly lower than the vertex, the median panels of the propodeum are lon- ger (Figs. 17-19), and the basal flanges of the petiole are lamellate (Figs. 17-19). Syntomopus is distinguished from Thi- nodytes by having an elongate pronotum (length about V3 its width) (Fig. 22), three broad symmetrically arranged clypeal denticles (Fig. 14), and usually a flattened mesosoma (Fig. 22). The pronotum of Thi- nodytes is much shorter, those Thinodytes species having symmetrically arranged denticles have them fingerlike rather than broad (Fig. 11), and the mesosoma is nev- er so flattened as it usually is in Syntomo- pus species. Biology. — The known hosts of Thinodytes species are all small Diptera living in plants as leaf or stem miners. KEY TO HOLARCTIC SPECIES OF THINODYTES GRAHAM 1. Petiole transverse, smooth (Fig. 17) 2 - Petiole as long as wide or longer than wide, reticulate (Figs 18-20) 4 2. Scape and legs beyond coxae pale, nonmetallic. Frenum and median panels of propodeum smooth. Plicae rounded and smoothly convergent posteriorly santema n. sp. Volume 4, 1995 7 - Scape mostly or completely metallic. Legs beyond coxae with metallic coloration or dark bands on femora and tibiae. Frenum and median panels of propodeum weakly to strongly alveolate. Plicae with distinct angle between parallel basal part and convergent posterior part 3 3. Hind margin of Tl emarginate medially. Dark bands on tibiae with sharp borders (Fig. 1). Female club with apex simple caroticus n. sp. - Hind margin of Tl entire medially. Dark bands on tibiae with diffuse borders (Fig. 2). Female club with terminal spine at apex cephalon (Walker) 4. Clypeal denticles symmetrically arranged. Petiole with weak median carina. Ovipositor exserted for length equal to that of T7 clypeattis (Girault) - Clypeal denticles asymmetrically arranged, the median denticle displaced to the left (Fig. 11). Petiole without a median carina (Figs 18-19). Ovipositor sheaths hardly exserted .... 5 5. Costal cell with three rows of setae distally. Eye length <4X as long as the temple length (Fig. 5). Body more or less all dark cyzicus (Walker) - Costal cell with two rows of setae distally. Eye length >4X as long as the temple (Figs 3, 7). Body dark with diffuse metallic patches on head and mesosoma 6 6. Male with terminal segment of funicle appearing as wide or wider than long (Fig. 4). Petiole usually less than 1.7X as long as wide (Fig. 18). Female with petiole usually less than 1.5X as long as wide. Both sexes with hind margins of Tl and T2 as long medially as laterally, (eastern United States and Canada) cyzicopsis n. sp. - Males with terminal segment of funicle appearing longer than wide (Fig. 6). Petiole usually more than 1.7x as long as wide. Female with petiole usually more than 1.5 x as long as wide (Fig. 19). Both sexes with hind margins of Tl and T2 usually longer laterally than medially, (far western United States and Canada) petiolatus n. sp. Thinodytes caroticus Heydon, new species (Fig. 1) Holotype, female. — Color: Body dark blue-green except flagellum, fore tarsus, wing veins brown; pretarsi black; knees, apical K; of tibiae, middle and hind tarsi white, border between the light ends and dark median band of tibiae sharp (Fig. 1). Sculpture: Clypeus smooth; head and mesoscutum delicately and regularly al- veolate; scutellum, frenum finely alveo- late; median panels of propodeum ob- scurely alveolate. Structure: Body length 1.2 mm. Head width 1.4X height (23:17), 2.3X length (23: 10); anterior margin of clypeus with single broad tooth; malar distance 5x length of genal concavity; eye height 1.2X length (10:8); 2.0X malar distance (10:5), eye length 4.0 X temple length (8:2); ratio of MOD, OOL, POL, LOL as 2.0:3.0:6.5:3.0; vertex rounding regularly into occiput; to- rulus Vi own diameter above LOcL. An- tenna with length of pedicel plus flagel- lum 0.87X head width (20:23); relative lengths of scape, pedicel, annelli, Fl-6, club as 6.5:2.5:0.5:2.0:2.0:2.0:2.0:2.0:2.0:5.0; widths of Fl, F6, club as 2:3:3; apical club segment simple apically, with micropilos- ity ventrally. Mesosoma arched dorsally, length 1.3X width (25:19); dorsellum short smooth band; propodeum with basal fo- vea part of continuous groove across an- terior margin of median panels, with short weak longitudinal carina crossing groove halfway between plica and median carina, with spiracles on anterior margin of pro- podeum, with nucha lunate strip and car- inate anteriorly. Fore wing with ratio of lengths of submarginal, marginal, post- marginal, stigmal veins as 19.0:11.5:9.0:5.0; stigma small; basal cell bare; basal vein with row of 4 setae. Petiole conical, trans- verse, smooth, with median carina. Gaster fusiform, length 1.8X width (30.0:16.5); hind margin of Tl emarginate medially; ovipositor sheaths hardly exserted; hypo- pygium extending Vi gastral length. Journal of Hymenoptera Research Allotype, male. — Color pattern similar to holotype except basic body color dark blue; fore tarsi pale brown; pale portions of legs pale yellow-brown instead of white. Body length 1.2 mm. Antenna with length of pedicel plus flagellum 0.98 X head width (21.0:21.5); relative lengths of scape, pedicel, annelli, Fl-6, and club as 6.0:2.5:0.5:2.0:2.0:2.0:2.0:2.0:2.0:6.0; widths of Fl, F6, club as 2.5:3.0:3.0; funicular setae sparse, reclinate. Gaster ovate, length 1.5X width (23:15). Variation. — The body length of females examined varied between 1.2 and 1.8 mm and males between 1.0 and 1.4 mm. The color of the dorsum of the mesosoma var- ies from dark blue-green to dark blue. The dark bands on the legs are always distinct, but the intensity of their metallic colora- tion is variable. Discussion. — Thinodi/tes caroticus most closely resembles T. cephalon because both species are dark in color; have a single, broad, asymmetrically placed clypeal den- ticle; and have smooth, transverse peti- oles. Thinodytes caroticus differs from T. ce- phalon in the following: 1. The hind mar- gin of Tl is emarginate in T. caroticus, but straight in T. cephalon. 2. The basal vein of the fore wing is setose in T. caroticus, but bare in T. cephalon. 3. The median panels of the propodeum are weakly sculptured in T. caroticus, but distinctly alveolate in T. cephalon. 4. The hypopygium extends about Vi the length of the gaster in T. car- oticus, but about % the gastral length in T. cephalon. 5. The dark bands on the tibiae are distinct with sharp borders in T. caro- ticus, but are less distinct and have diffuse borders in T. cephalon. These distinctive tibial color bands are unique to T. caroticus and will identify the species at a glance (Fig. 1). Etymology. — The species name comes from the Greek karotikos, meaning stupe- fying or soporific, and refers to the general nondescript appearance of this species. Type Material. — The holotype, allotype (both UCDC) and one male paratype were all reared by the author from the leaf mines of Calycomyza promissa (Frick) (Dip- tera: Agromyzidae) collected 30 June 1985 on the South Farms of the University of Illinois, near Champaign, Illinois. Fifty- nine additional paratypes seen were col- lected as follows (CNCI, INHS, SEMC, UCDC, USNM): Bermuda. DEVONSHIRE PARISH: Devonshire Marsh, 27.VI.1988, 1 male. FAGOT PARISH: Berry Hill Road, 29.VI.1988, 2 females, 3 males; Botanical Garden, 27. VI. 1988, 1 female; Camdon Marsh, 29.VI.1988, 1 female. SMITHS PARISH: Spittal Pond, 27.VI.1988, 1 fe- male. Canada. ONTARIO: Chatham, 1952 (mass reared from Hessian fly straw), 1 male. United States. CALIFORNIA: Haw- thorne, IX. 1940 (ex Aster blotch), 3 fe- males, 1 male; Jepson Prairie Preserve (13 km s. Dixon), 20.V.1983, 1 female; Lake Hennessy (11 km ese. St. Helena), 28.x. 1990, 1 female; Los Angeles County (bred from dipterous leaf miner), 1 female; Sacramento, 16. IX. 1924 [ex Agromyza pus- ilia {prob.= Liriomyza pusilla)], 1 female; So- quel, 26.VIII.1948 (ex Agromyza sp.), 2 fe- males, 1 male; 11 km e. St. Helena (Lake Hennessey), 28. X. 1990 (on Baccharis), 1 fe- male, 7.IX.1991 (on Heraculeum), 1 male; 6 km e. Suisun City, 11.VIII.1990, 2 males; 1000 Palms, 29.III.1977, 1 female; West- wood Hills (Los Angeles County), 5.XI.1940 (ex serpentine leaf miner in Zin- nia), 4 females, 1 male. FLORIDA: Braden- ton, 27.XI.1946 [ex Liriomyza pissilla {proh.= Liriomyza pusilla)], 1 male; Jackson- ville, 2 females. GEORGIA: Savannah, 5. VI. 1943 (parasite of goldenrod leafmin- er), 3 females. ILLINOIS: South Farms of the University of Illinois, 19.V.1985, 1 fe- male; White Heath, 24.IX.1939, 1 female. INDIANA: 4 miles s. New Harmony (Har- mony State Park), 28. VI. 1983, 2 females; 2 miles s. New Lisbon, 14. VII. 1981, 1 male. IOWA: Sioux City (reared from leaf miner on sunflower leaf), 1 female, 5. VIII. 1921 (reared from mine of leaf on sunflower), 1 female. KANSAS: Lawrence, 14.V.1955, 1 female. MINNESOTA: Albert Lea, Volume 4, 1995 26.VII.1960, 1 female. NEW MEXICO: Ci- marron, 1909, 1 female; Springer, 1909, 1 female, 3 males. TEXAS: Clarendon, 19. IX. 1905 (on Grindelia squarrosa (Pursk) Dunal (Compositae), 1 female; Ennis, 27.IX.1905 [ex dipterous leaf miner of Ma- chaeranthera annua (Rydb.) Shinner (Com- positae)], 2 females; Roma, 26. III. 1948 (ex pupa of dipterous leaf miner), 3 females, 3 males. Biology. — This species has been reared from leaf-mining Agromyzidae, mostly on Compositae. Known hosts include Ca- lycomyza promissa (Prick) and Liriomyza pusilla (Meigen). Because L. pusilla is a Pa- learctic agromyzid species (Spencer 1976) and is not known from the Nearctic region (Spencer and Steyskal 1986), it is likely that this host record is in error. There is one record from Chatham, Ontario from Hessian fly straw. Thinodytes caroticus was also reared from leaf miners on Machaer- anthera annua, Zinnia, goldenrod, sunflow- er, and "Aster". It has been reared from both linear-mining and blotch-making leaf miners. Other plant associations of a more uncertain nature include the com- posites Baccharis sp., Heracleum sp., and Grindelia squarrosa. Thinodytes clypeatus (Girault) Polycystus clypeatus Girault, 1918:128. Holotype, female (USNM); Hym. Type No. 20682; (ex- amined). Thinodytes clypeatus (Girault): Heydon, 1989: 193. Redescription. — Holotype, female. Color: Mesosoma, petiole black, with blue tints on propodeum and pleural regions; gaster dark brown; legs with basal % of femora brown, remainder of legs white. Sculpture: Dorsum of mesosoma regu- larly and delicately reticulate; median panels of propodeum weakly alveolate; petiole alveolate. Structure: Head with anterior margin of clypeus with three symmetrically ar- ranged fingerlike denticles, torulus above LOcL. Antenna with lengths and widths of segments as follows (in iJim): scape 19.5X3.6: pedicel 6.0x4.5: annelli 2.6x3.6: Fl 5X6: F2 4.6X6.0: F3 5.2x6.2: F4 4.6x6.5: P5 4.6X6.8: F6 missing: club 13.4X6.8; club simple apically. Mesosoma with dorsum rather flat, length 1.4X width (33:24); pro- notum with collar width 11 X length (22: 2), sides converging posteriorly; notauli shallow posteriorly; scutellum with two pairs of lateral setae, frenum almost indis- tinguishable from remainder of scutellum; propodeum with length of median panels 0.45 X width, with row of foveae separated by carinae along anterior margin, spiracle on anterior margin of propodeum, nucha a raised smooth crescent. Fore wing with relative lengths of submarginal, marginal, postmarginal, stigmal veins as 21:12:11:6; stigma large, height Vi distance between stigma and anterior margin of wing; basal cell with one seta; basal vein with three setae. Petiole length 1.2X width (6:5); with weak median carina; lacking lateral setae. Gaster fusiform, length 1.7X width (31.0: 18.5); Tl emarginate medially; hypopy- gium extending % length of gaster; ovi- positor sheaths exserted for length equal to that of T7. Discussion. — The holotype has the head removed and crushed on a slide. Thinod- ytes clypeatus differs from all other de- scribed Thinodytes species because it has an enlarged stigma, exserted ovipositor sheaths, and a median carina on the peti- ole. Thinodytes clypeatus has three sym- metrically arranged denticles like Synto- mopus species, but T. clypeatus differs from Syntomopus species in several ways. The denticles of T. clypeatus are fingerlike lobes like those of the other Thinodytes species with three denticles, whereas the denticles of Syntomopus species are rather broadly triangular (Fig. 14). The pronotal collar of Thinodytes clypeatus is short, eleven times as wide as long; the pronotal collar in Syn- tomopus is only about three times as wide as long (Fig. 22). Most species of Halticop- tera have a median carina on the petiole, but there are no other characters to indi- 10 Journal of Hymenoptera Research cate a particularly close relationship be- tween T. clypeatus and Halticoptera. Distribution and Biology. — The type spec- imen was reared from a leaf miner on corn on 2 May 1916, on St. Vincent, British West Indies by F. Watts. De Santis (1979) reports this species from Barbados also, and gives the name of its host as Agromyza parvicornis Loew (Diptera: Agromyzidae). Thinodytes cephnlon (Walker), new combination Figs 1, 17 ?Pteromnlus Rhseo Walker, Walker 1839b:88-89. Lectotype, female (BMNH); Hym. Type No. 5.772 (examined). Gastmncistrus cephalon Walker 1843:30. Lecto- type, male (BMNH); Hym. Type No. 5.661 (examined). Bubekia fallax Gahan, 1933:114-116. Holotype, female (USNM); Type No. 44841 (examined). Allen and Painter, 1937:225. Nikol'skaya, 1937:25. Peck, 1951:538. Thompson, 1958:587. Peck, 1963:610-611. Morrill and Kieckhefer, 1971:1130. Allen and Pienkowski, 1973:616, 617. Burks, 1979:789. Hendrickson, 1979:300, 302. n. syn. Discussion. — Thinodytes cephalon is easily distinguished by its single broad clypeal denticle, apical spine on the female club, notauli traceable to the hind margin of the mesoscutum as impressed lines, bare dor- sal vein, smooth and short petiole (Fig. 17), and first gastral tergite having the hind margin entire. A possible senior synonym of T. cephal- on is Pteromalus rhaeo described from Chil- oe Island, Chile. The lectotype female (designated herein) of P. rhaeo is unfortu- nately missing the head. It appears similar to specimens of T. cephalon examined, of which many specimens in my collection and the CNC collection were from Chiloe Island, but it is significantly larger than any other specimen examined. The medi- an panels of the propodeum in P. rhaeo are entirely rugose and the plicae are poste- riorly divergent. In other T. cephalon ex- amined, the median panels have few ru- gae which are located mostly posteriorly, and the plicae are parallel. The hind tibia in P. rhaeo is almost uniform in color; in T. cephalon, the hind tibia are distinctly dark over the middle half or more. Pter- omalus rhaeo may be within the range of variation of what is recognized as T. ce- phalon, but I am reluctant to formally syn- onymize the name until more is known about the range of variation of T. cephalon in South America. Distribution. — Thinodytes cephalon is one of the most commonly collected New World pteromalids. Its distribution ex- tends throughout most of North and South America (from Chile to Canada and also Bermuda). Biology. — Thinodytes cephalon is a para- sitoid of dipterous larvae that mine leaves or grass stems. Thinodytes cephalon was originally described from puparia of the Hessian fly, Phytophaga destructor (Say) (Diptera: Cecidomyiidae)(Gahan 1933). It has also been reared from the wheat stem maggot, Meromyza americana Fitch (Dip- tera: Chloropidae), in a number of studies (Gahan 1933; Allen and Painter 1937; and Morrill and Kieckhefer 1971). Allan and Painter report that T. cephalon probably oviposits into the larval stage of this host. One additional host added herein is Lir- iomyza trifoliearum Spencer (Diptera: Agro- myzidae) on alfalfa. Thinodytes cyzicopsis Heydon, new species Figs. 3, 4, 18 Holotype, female. — Color: Head, mesoso- ma black with metallic green patches as follows: entire dorsellum, pairs of spots alongside median ocellus, lateral hind cor- ners of pronotum and upper epimeron; metallic blue patches as follows: on inner orbits, anterior part of lateral lobe of scu- tum and propodeum. Antenna with scape, pedicel dark blue; flagellum black. Petiole black. Caster brownish black with metallic blue reflections dorsally, green ventrally. Legs with coxae black with weak blue re- Volume 4, 1995 11 flections; trochanters, femora brownish black with weak green reflections, knees yellow, tibiae with diffuse-edged dark brown band extending % their length; fore tarsi brown; middle and hind tarsi with pretarsi black, remainder yellow. Sculpture: Head, mesosoma, regularly and delicately alveolate; frenum, propo- deum shallowly and indistinctly alveolate; petiole finely alveolate. Structure: Body length 1.6 mm. Head width 1.2X height (27:22), 2.2 X length (27: 12); anterior margin of clypeus with three minute asymmetrically arranged apical denticles; genal concavity absent; eye height 1.4X length (13:9), 1.9X malar dis- tance (13:7), length 4.5 X temple length (9: 2) (Fig. 3); ratio of MOD, OOL, POL, LOL as 2.5:4.0:7.0:3.0; vertex rounding smooth- ly into occiput; torulus V2 own diameter above LOcL. Antenna with length of ped- icel plus flagellum l.Ox head width (27: 27); relative lengths of scape, pedicel, an- nelli, Fl-6, club as 11.0:3.0:1.0:2.5:3.0:2.75: 2.75:2.5:2.5:7.0; widths of Fl, F6, club as 3.0:3.5:3.5, Fl^ appearing quadrate, F5-6 slightly transverse; club simple apically, with minute patch of micropilosity on ter- minal segment. Mesosoma arched dorsal- ly, length 1.6X width (36:22); pronotum with humeral angles squared; mesoscu- tum with notauli shallow; dorsellum bandlike, short, weakly alveolate; propo- deum with basal fovea bordered mesally by sinuous carina; nucha a raised band, weakly carinate anteriorly; spiracle < 0.5 X own diameter from anterior margin of propodeum. Fore wing with ratio of lengths of submarginal, marginal, post- marginal, stigmal veins as 25:13:14:8; stig- ma small, maximum width only slightly more than stigmal vein width; basal cell bare; basal vein setose. Petiole length 1.2X width (6:5); without median carina; with one pair of lateral setae. Gaster length 1.4X width (31:22); hypopygium extend- ing to about % length of gaster; ovipositor sheaths hardly exserted. Allotype, male. — Color similar to holo- type except frons blue, mesoscutum with a pair of large diffuse green spots; axilla green. Body length 1.3 mm. Antenna (Fig. 4) with length of pedicel plus flagellum l.Ox head width (24:23); relative lengths of scape, pedicel, annelli, Fl-6, club as 7.0: 3.0:1.0:2.0:2.5:2.5:2.5:2.5:2:6.5; widths of Fl, F6, club as 2.0:2.5:2.5; setae fine, reclinate. Petiole length 1.2X width (5:4)(Fig. 18). Gaster ovate, length 1.2X width (20.0: 16.5). Variation. — The body color varies from mostly dark, like the holotype, to the pat- tern shown by the allotype male. The in- tensity, size, number, and hue of the me- tallic spots on the body are highly vari- able. The color of the metallic patches var- ies from coppery green to green to blue- green. The body length ranges between 1.4 and 1.8 mm in females and 1.2 and 1.7 mm in males. The length of the marginal vein varies from 0.9 to 1.2 times the length of the postmarginal vein but averages about equal its length [x = 1.02± (S.E. = )0.02 (n = 12)]. The petiole averages 1.39±0.045 (n=7, range 1.2-1.5) times as long as wide in the females and 1.57 ±0.061 (n = 7, range 1.2-1.9) times in males (Fig. 18). The pair of lateral setae on the petiole are sometimes difficult to see because they tend to lie near the anterior flanges of the petiole. Discussion. -Thinodytes cyzicopsis resem- bles the Palearctic species T. cyzicus (Thomson), but differs from that species in the following: 1. Thinodytes cyzicopsis has a series of metallic patches on the body, which are lacking in T. cyzicus. 2. The ratio of the eye length to temple length aver- ages 4.25±0.12 (n=6, range 3.8^.5) in fe- male T. cyzicofisis (Fig. 3) but measured only 2.3 and 3.6 in two female T. cyzicus specimens examined (Fig. 5). 3. The costal cell of the fore wing of T. cyzicopsis has two rows of setae distally, whereas there are three rows distally in the costal cell of T. cyzicus. 4. The petiole has a pair of lat- eral setae in T. cyzicopsis, which are lack- ing from T. cyzicus. 12 Journal of Hymenoptera Research Thinodytes cyzicopsis and T. petiolatus are very similar species and specimens cannot always be confidently separated. These two species are distinct from other Thinod- ytes in their common possession of a sim- ilar pattern of metallic patches on the head and mesosoma. Individuals of T. cyzicopsis differ from those of T. petiolatus in the fol- lowing: 1. The ratio of eye height to malar distance in T. cyzicopsis averages 1.74±0.035 (n=8, range 1.5-1.8) in females and 2.04 ±0.032 (n=9, range 1.9 to 2.0) in males; in T. petiolatus the ratio is 1.48 ±0.23 (n=10, range 1.4 to 1.6) in females and 1.77±0.038 (n=10, range 1.6 to 2.0) in males. 2. All funicular segments of the male antenna are transverse to quadrate in T. cyzicopsis; all funicular segments are usually longer than wide in male T. pe- tiolatus. 3. The ratio of petiole length to width in T. cyzicopsis averages 1.39 ±0.045 (n=7, range 1.2 to 1.5) in females and 1.57±0.061 (n = 7, range 1.2 to 1.9) in males; in T. petiolatus it averages 1.93±0.062 (n=10, range 1.5 to 2.0) in fe- males (Fig. 19) and 1.95±0.071 (n=10, range 1.6 to 2.4) in males. 4. The first gas- tral tergite in T. cyzicopsis is about as long medially as laterally, whereas Tl is often shorter medially than laterally in T. petiol- atus. 5. The hind margin of T2 is straight to just noticeably concave in T. cyzicopsis, whereas it is usually distinctly concave in T. petiolatus. 6. The terminal segment of the male funicle appears quadrate to slightly transverse in T. cyzicopsis, whereas it appears elongate in T. petiolatus. In ad- dition to these structural characters, T. cy- zicopsis is found east of the Rocky Moun- tains, and T. petiolatus occurs from the Rockies west. Type Material. — The holotype (CNCI) is from Kouchibouquac National Park, New Brunswick, and was collected 9 August 1977 by S. J. Miller. The allotype (USNM) is from Ithaca, New York, and was col- lected on vernal alfalfa on 22 June 1968 by A. G. Wheeler. Sixteen paratypes were col- lected as follows (CNCI, UCDC, USNM): Canada. ALBERTA: Elkwater Lake, 21.VII.1956, 1 female; Lethbridge, 1 fe- male. NEW BRUNSWICK: Kouchibou- quac National Park, 20.IX.1977, 1 female. NOVA SCOTIA: Alton, IX. 1964 [ex Phy- tobia {Calycomyza) solidaginis on Solidago], 1 male; Crosby, 31 .VII. 1952 (on apple), 1 male. ONTARIO: Ottawa, 22.VI.1972 (swept from Salix hlanda), 1 male, 29.VI.1972 (swept from Salix blanda), 3 males, 24. VII. 1972 (swept from Salix hlan- da), 1 male. QUEBEC: Lac Brule, 21.VII.1947, 1 female, 25.VII.1947 (swept from Rosa rugosa), 1 female, 9.VIII.1945, 1 female. United States. ILLINOIS: 2 miles e. Shumway, 7.VII.1980, 1 female; South Farms of the University of Illinois, nr. Champaign, 23.VI.1981, 1 male. MASSA- CHUSETTS: Hopkinton, 9.VIII.1951 (ex Ilex leaf miner), 1 female. NEW MEXICO: Mesilla, 4. V. 1909, 1 female. WEST VIR- GINIA: Winchester, 16.VI.1964, 1 male, 17.VI.1964, 1 male. Etymology. — The species name is de- rived from the species name of Thinodytes cyzicus and the Greek suffix -opsis, mean- ing like or similar in appearance, and re- fers to the morphological similarity be- tween T. cyzicopsis and T. cyzicus. Biology. — Known hosts of T. cyzicopsis include Phytobia {Calycomyza) solidaginis (Agromyzidae)[on Solidago sp. (Composi- tae)] and an Ilex leafminer. The species has also been taken on Salix blanda (Salica- ceae), Rosa rugosa (Rosaceae), and alfalfa. Salix and Rosa are probably primarily nec- tar sources because neither has many leaf- mining agromyzids, but these plants are important sources of nectar and /or hon- eydew for many parasitic Hymenoptera. Thinodytes cyzicus (Walker) Miscogaster cyzicus Walker, 1839a:200. Lecto- type, female (BMNH); Hym. Type No. 5.2570 (examined). Syntomopus cyzicus (Walker): Walker, 1846:28. Schmiedeknecht, 1909:376. Dicyclus circulus Thomson, 1876:253. Lectotype, female (LUND), not seen. Volume 4, 1995 13 Thinodytes cyzicus (Walker): Graham, 1956:261. Graham, 1969:167. Askew, 1970:380. Hedqv- ist, 1975:180. Boucek, 1977:56. Kamijo, 1978: 457. Takada and Kamijo, 1979:21, 22, 23, 25. Hedqvist, 1983:167. Boucek and Rasplus, 1991:32. Discussion. — The synonymy of Dicyclus circiilus with Thinodytes cyzicus is accepted on the authority of Graham (1969). Thi- nodytes cyzicus resembles T. cyzicopsis and T. petiolatus, because all three species have three small, sharp, asymmetrically ar- ranged clypeal denticles (Fig. 11) and a long, reticulate petiole. Besides the geo- graphic separation of their ranges, T. cy- zicus can be distinguished from the two Nearctic species by its body color. Thino- dytes cyzicus is uniformly dark and lacks the distinctive diffuse metallic patches on the head and mesosoma present in T. cy- zicopsis and T. petiolatus. Thinodytes cyzicus also has three rows of setae distally in the costal cell, whereas T. cyzicopsis and T. pe- tiolatus have only two rows. The eye length in T. cyzicus is 4X or less the length of the temple (Fig. 5), whereas T. cyzicopsis and T. petiolatus have the eye length more than 4X the length of the temple (Figs 3, 7). Distribution. — Thinodytes cyzicus occurs throughout the Palearctic region, from Britain to Japan (Graham 1969; Boucek 1970; Kamijo 1978). Biology. — The hosts of T. cyzicus are all Agromyzidae. Askew (1970) recorded this species from an agromyzid, probably Phy- tomyza atricornis Meigen, on Senecio jaco- baea Linnaeus. Kamijo (1978) recorded T. cyzicus from Chromatomyia horticola (Gou- reau)(Diptera: Agromyzidae) on pea and from an agromyzid on Lathy rus maritimus. Takada and Kamijo (1979) recorded T. cy- zicus as emerging from the puparium of Chromatomyia horticola and speculated that T. cyzicus may parasitize the larval stage of its host. Thinodytes petiolatus Heydon, new species Figs 6, 7, 11, 19 Holotype, female. — Color: Body black but frenum, metanotum coppery; frons, ver- tex, gena, lateral region of pronotum, lat- eral lobe of mesoscutum, middle lobe of mesoscutum with pair of large diffuse me- tallic patches, anterior lateral corner of ax- illa dark metallic green; gaster with green- ish reflections. Antenna with scape dark green; remainder brown, pedicel with weak green reflections. Legs with coxae black with greenish reflections; femora brown with greenish reflections; remain- der of legs mostly yellow-brown, tibiae with broad diffuse brown band mesally, apical two tarsi brown. Wing veins pale reddish brown. Sculpture: Clypeus weakly alveolate; face alveolate, cells elongate in radiating fashion from clypeus; remainder of head alveolate; mesoscutum, scutellum deli- cately and regularly alveolate except fre- num smooth; median panels of propo- deum alveolate; petiole finely alveolate. Structure: Body length 1.8 mm. Head width 1.4X height (28:21), 2.6 X length (28.5:11.0); clypeus with three small asym- metrically arranged clypeal denticles (Fig. 11); weak genal concavity extending Vs malar distance; eye height 1.3X length (12.0:9.5), 1.5X malar distance (12:8), length 4.8X temple length (9.5:2.0)(Fig. 7) ratio of MOD, OOL, POL, LOL as 2:4:7:3 vertex rounding regularly into occiput antennal torulus just above LOcL. Anten- na with length of pedicel plus flagellum 0.91 X head width (26:28.5); ratio of lengths of scape, pedicel, annelli, Fl-6, club as 11.5:3.5:1.0:2.5:2.5:2.5:2.5:2.5:2.5:6.0; widths of Fl, F6, club as 2:3:3; club simple apically, with small patch of micropilosity ventrally on apical segment. Mesosoma arched dorsally, length 1.7X width (37:22); notauli shallow posteriorly; propodeum (Fig. 19) with basal fovea margined mesal- ly by long sinuous carina; nucha a weakly 14 Journal of Hymenoptera Research sculptured band, carinate anteriorly; spi- arate these two species are given in the racle <0.5X own diameter from anterior discussion section for T. cyzicopsis. The margin of propodeum. Fore wing with rel- hind margin of T2 in many specimens of ative lengths of submarginal, marginal, T. petiolatus is distinctly concave, but the postmarginal, stigmal veins as 27.0:14.5: visibility of this character depends on how 15.0:9.0; stigma small; basal vein with row the specimen has dried. This character is of setae along length. Petiole (Fig. 19) more distinct in the males than in the fe- length 2.0 X width (8:4); without median males. carina; with two pairs of lateral setae. Gas- Type Material: — The holotype (USNM) ter ovate-acuminate, length 1.4X width was collected at Roseworth, Idaho, on 19 (33:24); hind margin of Tl strongly sinu- June 1931, on Salsola pestifer by D. E. Fox ous laterally, emarginate mesally; ovipos- and the allotype (USNM) was collected at itor sheaths hardly exserted; hypopygium Twin Falls, Idaho, on 5 August 1920 by R. extending % gastral length. H. Smith. Sixty paratypes were collected Allotype, male.— Body color similar to as follows (CNCI, INHS, UCDC, USNM): holotype except metallic spots more dif- Canada. ALBERTA: Banff (Sunshine fuse and extensive; frenum, metanotum Lodge, 7500'), 24. VII. 1962, 1 female; Elk- blue-green; dark bands on tibiae very water, 9.VI.1956, 1 male; Elkwater Lake, weak. Body length 1.3 mm. Antenna (Fig. 21.VII.1956, 1 male; Lethbridge, 7.VI.1956 6) with length of pedicel plus flagellum (swept from barley), 1 female, 5.VIII.1956 1.1 X head width (28:25); ratio of lengths (swept from barley), 1 male; nr. Leth- of scape, pedicel, annelli, Fl-6, club as 9.0: bridge, 1924.1925, 1 female. BRITISH CO- 3.5:1.0:2.5:3.0:3.0:3.0:3.0:3.0:8.0; widths of LUMBIA: Bowser, 28.V.1955, 1 female; Fl, F6, club as 2:2:3; setae of flagellomeres McQueen Lake, 10 miles n. Kamloops, fine, reclinate. Petiole length 2.0 X width 18.VI.1973, 1 male. United States. ARI- (8:4). Gaster length 1.2X width (20:16); ZONA: Mesa, 3.VI.1958 (swept from alfal- truncate apically. fa), 1 male. CALIFORNIA: Albany, Variation. — The body color varies from 29. IV. 1958, 1 male; Alpine Lake, VI. 1971, like that of the holotype to very dark, with 1 female; Apple Valley, 8.V.1955, 2 males; the green areas on the head and the me- 15 miles w. Baker, 6.V.1977, 2 females, 1 sosoma reduced and obscure, the frenum male; Boca, 22. VII. 1970, 1 female; Bolinas, and metanotum green, and the tibiae dark 5.X.1975, 2 males; Camino, 21.VII.1948 (ex metallic green. The body length of females Phytomyza aqiiilegiana), 10 females; Cerro examined varied between 1.3 and 1.8 mm Noroeste (sw. corner of Kern Co.), and males varied between 1.0 and 1.6 mm. 15.VII.1965, 1 female, Cuyler Harbor (San The sinuous carina on the propodeum Miguel Island), 11. VII. 1970, 1 male; Dar- sometimes extends only Va the length of win Falls (nr. Panamint Springs), the propodeum. The row of setae on the 29. III. 1984 (on Encelia), 1 male; Emeryville, basal vein sometimes curls proximally, 28. V. 1958, 1 male; Eureka Dunes (Inyo running a short distance along the cubital Co.), 15.V.1979, 1 male; Lake Tenaya, vein. 23.VII.1949 (host Recurraria milleri), 1 fe- Disciission. — Thinodytes petiolatus and T. male; Lake Tahoe, 29. VI. 1927, 1 female; cyzicopsis are very similar species and are Lily Pond (alpine lake), VI. 1971, 1 female; distinct from other Thinodytes species in Los Angeles {Agromyza playptera Thom.), 1 the similar pattern of metallic patches on female; Los Angeles Co., 1 male; McClure an otherwise dark body. Individuals of T. Beach (Marin Co.), 18.VII.1970, 1 male; Mt. petiolatus and T. cyzicopsis, especially fe- Ingalls (Placer Co.), 11. VII. 1964, 1 male; males, cannot always be distinguished near Nicasio, 15.11.1991 (sweeping Salix), 2 with absolute certainty. Characters to sep- males; 22 miles w. Panamint Springs, Volume 4, 1995 15 7.V.1961, 2 males; Placer Co., Vlll, 1 male; (5), 1 male; Moscow, 6.VIII.1926, 1 male; Sagehen Creek (near Hobart Mills), 24. Murtaugh, 29.V.1930 (3 & 5), 1 male; Oak- VI.1970, 1 female, 21-25.VI.1982 (sweep- ley, 7.VIII.1929 (1, 3 & 5), 1 female; Rupert, ing Primus), 1 male, 12.VII.1972, 1 male, 29.V.1930, 1 female; Tuttle, 22.V.1931 (4), 11-15.VII.1982, 1 female, 23.VII.1968, 1 fe- 1 female; Twin Falls, 7.VI.1930, 1 male; male, 1.VIII.1970, 1 female; San Bernardi- Wendell, 22.V.1931 (5), 3 males. MON- no Co., San Gorgonio Wilderness, TANA: Big Butte, 8.VIII.195?, 1 male. OR- 19. VII. 1982, 1 male; Santa Cruz, EGON: Corvallis, 15.VI.1981, 1 male, 22.V11I.1948 (ex PJnjtonn/za sp. B), 1 female; 26.VI.1985, 1 female, 3 males; Near Cor- Santa Rosa, 25.V.1990, 2 males; Sheppard vallis (St. Mary's Peak), 15.VIII.1984 (road- Pass Trail (Inyo Co.), 16. VII. 1985 (on side vegetation), 1 female; Eugene, 6- Sphenosciadium capitellatum),! male; Shive- 12.VI.1984 (Malaise trap), 1 female, 11 ly, 19. VI. 1959, 1 female; Sonora Pass males; McKinzie Pass (Mt. Washington (Mono Co., 9624'), 20.V1II.1960, 1 female; Wilderness), 17.VIII.1984, 2 females, Soquel, 26.VIII.1960 (ex Agromijza sp.), 2 20.VIII.1984, 5 females, 2 males; 1 mile w. females; Titus Canyon (Death Valley Na- McKinzie Pass, 1 female, 1 male. UTAH: tional Monument), 24.III.1984 (on Cownia Logan Canyon, 21. VII. 1976, 1 female; mexicana), 1 male; Tomales Bay State Park Monte Cristo, 6.VII.1976, 1 male; My ton, 3 (Marin Co.), 14.IV.1961, 1 male; Trinidad, males; Wellsville, 13.V.1964, 1 male; 1.5 24.IX.1977, 1 female; Truckee, 20.VII.1970, miles w. Wild Horse Butte (Wild Horse 1 female; White Water, Snow Creek Creek), 31.VII.1982, 1 female. WASHING- (1500'), 29.III.1955, 1 male; 6 miles w. Bas- TON: San Juan Island (Barney's Place), setts (Yuba Pass), 9.VII.1970, 2 females, 2 23.VII.1944, 1 male. Vancouver, 15.VI. males. COLORADO: Chambers Lake (Lar- 1911, 1 female, 1 male, 16.VI.1911, 1 male, imer Co.), 16.VIII.1966, 1 male; 16 km n. 20.VI.1911, 1 female. WYOMING: Kem- Colorado Springs, 25. VII. 1991, 1 male; Ft. merer, 12. VII. 1985 {Eriogomim), 2 females, Collins, 20. VIII. 1895 (on boxelder foliage), 13.VII.1985 [Artemesia spinosa {?=spines- 1 female, 1 male; Glacier Basin, Rocky cens D.C. Eaton], 1 male; Snowy Range, Mountain National Park, 24.VII.1977, 1 23.VIII.1951, 2 females, 1 male, male; Echo Lake (Mt. Evans, 10,500'), Etymologxj. — The species name refers to 4. VIII. 1961, 1 female. IDAHO: Boise, the long petiole, characteristic of this spe- 28.V.1984 (Malaise trap), 2 males; Buhl, cies. 27.V.1929 (1 & 50, 2 males; Burley, B/o/ogi/.— The host(s) of T. petiolatus is 14.VI.1930 (3), 1 female, 9. VII. 1931, 1 fe- unknown but it has been taken in associ- male; Eden, 11. VIII. 1930 (2), 1 female; ation with a number of plants, such as Ar- Hobbs Butte, 22.V.1931, 1 male, 6. VI. 1931, temesia sp., Descurainia sophia, Eriogomim, 1 male; Hollister, 16.V.1931 (3), 1 male, Salsola pestifer, and Sisymbrium altissimum 20.V.1931, 2 males, 2.VI.1931, 1 female, 1 in Idaho and Encelia in California. It has male, 5. VI. 1931 (3), 1 female, 7. VI. 1931 (3 also been collected in association with & 5), 1 female, 13.VI.1931, 1 female; Hubbs crop plants, such as beets in Idaho and Butte, 22.V.1931, 1 male, 6.VI.1931, 1 male; barley in Alberta. Jerome, 11.VII1.1930 (2), 1 male; Kimberly, 27.VI.1931, 2 females; Milner, 29.V.1930 Thinodytes santerna Heydon, new species Holotype, female. — Color: Body black with blue-green reflections on vertex, side ' The host plants for the specimens from Idaho are j^^^^g ^f mesoscutum, SCUtellum, gaster; numbered as follows: 1= A.? rosae. 2= Beta viiharis n ■ <- i i i L. 3= Soplua sophia {=Descuraima sophia (L.) V.B. coppery reflections on frenum and dorsel- Webb). 4= Salsola pestifer A. Nelson. 5= N.(ortn ) al- lum; yellow-green reflechons on middle tissiiniim or S. altissiina { = Sisymbri\nu altissimum L.). lobe of SCUtellum and propodeum. Anten- 16 Journal of Hymenoptera Research na with scape yellow, flagellum brown dorsally, brownish white ventrally. Legs yellow beyond coxae, tarsi white, pretarsi brown. Fore wing veins pale brown. Sculpture: Clypeus alveolate; frenum, dorsellum, propodeum smooth; gaster with T5-7 coriaceous, remainder smooth. Structure: Body length 1.6 mm. Head width 1.4X height (28:20), 2.3 X length (28: 12); clypeus with three small symmetri- cally arranged clypeal denticles; weak genal concavity extending Va malar dis- tance; eye height 1.3 X length (11:10), 2.2 X malar distance (11:6), length 5.0 X temple length (10:2); ratio of MOD, OOL, POL, LOL as 2:3:6:3; vertex rounding regularly into occiput; antennal torulus just above LOcL. Antenna with length of pedicel plus flagellum 0.86 X head width (24:28); ratio of lengths of scape, pedicel, annelli, Fl-6, club as 10:3:1:2:2:2:2:2:2:6; widths of Fl, F6, club as 2:2:2; club simple apically, with small patch of micropilosity ventrally on apical segment. Mesosoma arched dorsal- ly, length 1.5 X width (32:22); notauli shal- low posteriorly; propodeum with basal fo- vea obscure, nucha lunate strip and cari- nate anteriorly, spiracles on anterior mar- gin of propodeum. Fore wing with relative lengths of submarginal, marginal, postmarginal, stigmal veins as 24:12:12:6; stigma small; basal vein with row of setae along length. Petiole transverse, smooth. Gaster ovate acuminate, length 1.6X width (36:23); hind margin of Tl nearly straight, slightly convex mesally; hypo- pygium extending Vi gastral length. Discussion. — This species is distinct from all other Thinodytes species by the characters listed in the key: scape and legs beyond coxae pale, nonmetallic; frenum and median panels of propodeum smooth; plicae rounded and smoothly convergent posteriorly. In these characters, T. santerna is phenetically similar to species of Noto- glyptiis. Type Material— The holotype (USNM) was collected 3 April 1984, at the Eagle Borax Works in Death Valley National Monument, Inyo County, California (USA) by E. E. Grissell on Distichlis in a brackish marsh. Etymology. — The specific epithet of this species is from the Latin noun santerna, meaning borax, and refers to the locality where the type specimen was collected. Biology. — Nothing is know of the host(s) of T. santerna. Mauleus Graham, 1981 Type Species: Mauleus maderensis Gra- ham, 1981 (examined); original designa- tion. Description. — Body very dark green or blue; scape brownish yellow, nonmetallic. Head, pronotum, mesoscutum, scutellum (including frenum), dorsellum, median panels of propodeum, petiole alveolate; gastral tergites nearly smooth. Head with clypeus bidentate (left tooth compound) (Fig. 13), lateral part of mouth margin with short shallow genal concavity; anten- nal torulus IX own diameter above LOcL. Antenna with scape cylindrical, >6X as long as wide; flagellum length less than head width in females, about equal to head width in males; funicular segments cylindrical; MPP sensilla in single row; fe- male club simple apically and with small patch of micropilosity on apical segment. Male maxilla with palps slender, stipites unenlarged. Mesosoma (Figs 8, 21) arched dorsally; pronotum with collar short (Figs 23-25), nearly level with vertex dorsally, anterior edge rounded; mesoscutum with notauli shallow, impressed lines at most; scutellum as long as wide, lacking anterior median groove, frenum indistinguishable from remainder of scutellum; dorsellum short, length about equal to length of ridge across anterior margin of propo- deum; propodeum (Fig. 20) with median panels short (width about 2X median length), plicae and median carina well de- veloped and connected posteriorly by W- shaped carina; spiracles strongly ovate al- most linear. Fore wing with relative lengths of veins as follows: submarginal > Volume 4, 1995 17 Figs. 23-25. 23, Maidens maderensis Graham, female pronotum and mesonotum; 24, Maidens iligneus n. sp., female pronotum and mesonotum; 25, Maidens venetiis n. sp., female pronotum and mesonotum. marginal > postmarginal > stigmal; stig- ma small, width about 2X width of stig- mal vein; costal cell with complete row of setae and sometimes a partial second row; basal cell bare; basal vein setose; specu- lum present, open posteriorly. Petiole (Fig. 20) longer than wide, with basal flange thickened laterally, without median cari- na, without lateral setae. Gaster of females lanceolate, length 1.6 or more times width; hypopygium extending V3 length of gaster or more; in both sexes, hind margin of Tl sinuous laterally, emarginate medially (Figs 8, 21). Discussion. — This genus is placed in the Halticoptera-group as defined in this paper by: the rounded pronotum, shallow no- tauli, undifferentiated frenum, propo- deum with a median carina and plicae connected by a W-shaped carina, reticu- late petiole with a complete basal flange, and Tl that is sinuous laterally and emar- ginate medially. Maidens and Halticopitera exhibit considerable phenetic similarity. The clypeus of Mauleus is secondarily bi- dentate (Fig. 13); the left hand clypeal den- ticle is divided by a sulcus formed as a result of the near fusion of the two ap- proximated left hand denticles. These den- ticles are shown in their plesiomorphic separated state in Thinodytes (Fig. 11). The same compound condition of the left den- ticle is found in Halticoptera Spinola (Hey- don, unpublished data). [In contrast, in the unrelated genus Sphegigaster Spinola, the left tooth in the bidentate clypeus is a single unit (Heydon, unpublished data)]. The compound clypeal denticular struc- ture and the nonmetallic scape are possi- ble evidence of a close phylogenetic rela- tionship between Halticoptera and Mauleus. Halticoptera is well-defined cladistically relative to Mauleus by the very low inser- tion of the antennae (at or below the LOcL) and by the male maxilla, which has the terminal two segments of the palps flattened, expanded, and nearly always yellow and often the stipites also expand- ed. Two apomorphic characters readily define Mauleus relative to Halticoptera and related genera: 1. The propodeum is short- ened in Mauleus (median panels over 2X as wide as long in Mauleus species exam- ined). 2. The basal flanges of the petiole are exceptionally large and thick, giving the petiole a connate appearance. The bas- al flanges in related genera are generally free-standing lamella and the petiole is more or less cylindrical. Biology. — Mauleus iligneus has been reared from pupae of the native holly leaf miner, Phytouiyza ilicicola Loew (Diptera: Agromyzidae). Like many other genera of the Miscogastrinae, it is likely that Mau- leus species are parasitoids of the pupal stages of leaf-mining or stem-mining Dip- tera. 18 Journal of Hymenoptera Research KEY TO DESCRIBED SPECIES OF MAULEUS GRAHAM 1. Pronotal collar with sides parallel in dorsal view (Fig. 23) 2 - Pronotal collar with sides widest near anterior margin and converging posteriorly in dorsal view (Figs 24, 25) 3 2. Vertex and mesoscutum with conspicuous pale setae. Propodeum with reticulations much less coarse than on scutellum, median panel each with broad shallow groove along anterior margin nigritus (Howard) - Vertex and mesoscutum with indistinct dark setae. Propodeum with reticulations as coarse as on scutellum, median panel each with a pair of elongate sublateral depressions along anterior margin maderensis Graham 3. Propodeum with anterior depression between basal foveae; median panels with extensive area of weak, almost smooth sculpture. Pronotum with humeral angles acute, coming to blunt points in dorsal view (Fig. 25). MPP sensilla in two or more rows on funicular segments venetus Heydon - Propodeum without distinct anterior depression between basal foveae; median panels al- most entirely alveolate, with at most a small central patch of weak sculpturing. Pronotum with humeral angles either squared or slightly and smoothly convergent posteriorly (Fig. 24). MPP sensilla in single row on funicular segments (male of M. cultratus unknown) ... 4 4. Eye height 2.5-2.7 X genal distance. Female with combined length of head and mesosoma longer than gaster; gaster less than twice as long as wide; hypopygium extending around % gastral length (United States) iligneus Heydon - Eye height 2. 1-2.4 X genal distance. Female with combined length of head and mesosoma less than or equal to length of gaster (Fig. 8); gaster more than twice as long as wide; hypopygium extending to near tip of gaster (southern Mexico to Argentina) cultratus Heydon Maidens cultratus Heydon, new species Structure: Body length (excluding ovi- Fig. 8 positor sheaths) 1.6 mm. Head width 1.2X height (26:21), 2.2 X length (26:12); genal Holotype. female. -Color: Head, pleural concavity extending Vs malar distance; eye regions, propodeum dark blue; pehole j^^-g^t 1.4X length (13.5:9.5), 2.2X malar bluish black; collar, dorsum of mesosoma distance (13.5:6.0), length 4.8X temple greenish black; gaster dark brown, Tl length; ratio of MOD, POL, OOL, LOL as with dark blue reflections. Antenna with 2:4:6:3; torulus IX own diameter above scape brownish yellow with weak metallic lqcL. Antenna with length of pedicel plus reflections; pedicel, flagellum brown. Legs flagellum 0.81 X head width (21:26); ratio with coxae, trochanters, femora dark blue; of lengths of scape, pedicel, annelli, Fl-6, tibiae brown except basal and apical tips dub as 9.0:3.5:1.0:1.5:2.0:2.0:2.0:2.0:2.0:5.0; brownish yellow; fore tarsi brown, middle widths of Fl, F6, club as 2:3:3; MPP sparse, and hind tarsi yellow-brown with pretarsi only one or two visible per segment from brown. Wing veins pale brown. single view. Mesosoma length 1.4X width Sculpture: Clypeus, median portion of (33:23); pronotum with sides converging face finely alveolate; remainder of head, posteriorly; notauli incomplete; propo- mesonotum, scutellum, frenum, median deum with width of median panels 2.2 X panels of propodeum alveolate; petiole length (11:5); basal fovea a shallow, tri- finely alveolate; gaster smooth except T5- angular depression; plicae fading out in 7 weakly coriaceous. anterior Vz; spiracles 0.5 X own diameter Volume 4, 1995 19 from anterior margin of propodeum. Wing with relative lengths of submargin- al, marginal, postmarginal, stigmal veins as 27:16:14:7; costal cell with single com- plete row of setae; basal vein setose with one seta posteriorly on cubital vein. Peti- ole length 1.5X width (9:6); sides narrow- ing posteriorly. Gaster lanceolate (Fig. 8), length 2.6 X width (42:16), length l.OX that of head and gaster (42:42); hypopygium reaching to apex of T7; ovipositor sheaths exserted for distance equal to half length of hind tibia. Variation. — The color of the head, pleu- ral regions, and petiole varies from dark blue, as in the holotype, to dark green and nearly concolorous with the dorsum of the mesosoma. The body length of specimens from the type locality is about 1.9 mm, whereas the length of the specimen from Ixtapan is 2.3 mm, the one from Morelia is 2.4 mm, and the female from Panama is 3.1 mm in length. The ovipositor sheaths are exserted for a distance equal to ¥i to Vi the length of the hind tibia. Discussion. — Maidens cultratus can be distinguished from M. iligneus by the char- acters given in the discussion section for that species. Eti/mologx/. — The species name comes from the Latin word cultratus, meaning knife-shaped, and refers to the shape to the female gaster. Type Material. — The holotype female (CNCI) and one paratype female were col- lected from San Cristobal de las Casas, Mexico, on 29 June 1969. Eight other para- type females were collected as follows (CASC, CDAE, CNCI, SEMC, USNM): Ar- gentina. SALTA: Rosario de Lerma, 4- 8.XI.1983 (Malaise trap). Mexico. CHIA- PAS: San Cristobal de las Casas, 1- 12.V.1969; MICHOACAN: Morelia, 6.IX.- 1938; MEXICO: Ixtapan, 9.VII.1954; TA- MAULIPAS: 6 miles n. Ciudad Victoria. Panama. Chiriqui, XII. 1946. Biology. — The host(s) of this species are unknown. Mauleus iligneus Heydon, new species Figs. 13, 20, 21, 24 Holotype, female. — Color: Like that of M. cultrata except pedicel not metallic, and middle and hind tarsi white. Sculpture. Clypeus and immediate vi- cinity finely alveolate (Fig. 13); remainder of head, mesoscutum, scutellum, frenum, median panels of propodeum alveolate; petiole finely alveolate; gaster smooth ex- cept T6 and T7 coriaceous. Structure. — Body length 1.8 mm. Head width 1.3X height (29:23), 2.2X length (29: 13); eye height 1.4X length (15.0:10.5), 2.5 X malar distance (15:6), length 5.2 X temple length (10.5:2.0); ratio of MOD, OOL, POL, LOL as 2.5:4.0:7.0:3.0; torulus located IX own diameter above LOcL. An- tenna with length of pedicel plus flagel- lum 0.88 X head width (25.5:29.0); ratio of lengths of scape, pedicel, annelli, Fl-6, club as 11.0:3.0:1.0:2.5:2.5:2.5:2.5:2.5:2.5:7.0; widths of Fl, F6, club as 2.0:3.0:3.5. Me- sosoma length 1.5X width (36.0:24.5); pro- notum with humeral angles squared but sides convergent posteriorly (Fig. 24); no- tauli extending to hind margin of mesos- cutum as impressed lines; propodeum (Fig. 20) with plicae fading out in anterior half, basal fovea extending halfway down median panels and bordered mesally by carina, nuchal area raised but acarinate anteriorly, spiracles almost on anterior margin of propodeum. Fore wing with rel- ative lengths of submarginal, marginal, postmarginal, stigmal veins as 25:15:11:6; costal cell with one complete setal row and one partial row distally; basal vein with row of three setae. Petiole (Fig. 20) length 1.3X width (8:6); narrowing poste- riorly; basal flanges large. Gaster fusiform (Fig. 21), length 0.87X length of head and mesosoma (39:45); length 1.8X width (39: 22); hypopygium extending % length of gaster; ovipc^sitor sheath exserted for a distance equal to/ length of hind tibia. Allotype. — Male. Color similar to holo- type except frons, callus green; flagellum 20 Journal of Hymenoptera Research paler ventrally; tibiae brownish yellow, allotype (USNM), and an additional nine Body length 1.8 mm. Head with ocelli rel- female and seventeen male paratypes atively larger, ratio of MOD, OOL, POL, (UCDC, USNM) were reared from Phijto- LOL as 3.0:3.0:7.5:3.5. Antenna with mi/za ilicicola on Ilex opaca Solander at Lex- lengths of pedicel plus flagellum l.Ox ington, Kentucky by D. A. Potter collected head width (31:31); relative lengths of 16 May 1984. An additional 43 paratypes scape, pedicel, annelli, Fl-6, club as 10.0: were collected as follows (CNCI, UCDC): 3.5:1.0:3.0:3.5:3.0:3.0:3.0:3.0:8.0; widths of ILLINOIS: Cave-in-Rock State Park, near Fl, F6, club as 2.5:3.0:3.0; setae reclinate. Cave-in-Rock, 4.VI.1981, 1 female. KAN- Gaster ovate, length 1.8X width (42: SAS: Oswego, 17.V.1976, 1 female. MARY- 23)(Specimen critical-point dried so gaster LAND: College Park, V.1954 (ex holly leaf in air-dried specimen will be shorter). miner) 1 male, VII-VIII.1937 (ex Phytomyza Variation. — Length of female specimens ilicis), 4 females, 8 males; Laurel, varies between 1.5 and 2.1 mm and males 14.V.1965, 1 female. NEW JERSEY: New between 1.1 and 1.8 mm. Other than the Brunswick, 26.V.1947 (ex Phytomyza ilici- rather large variation in size, this is a mor- cola), 1 female. TEXAS: Houston, 8.XII.- phologically uniform species. 1929 (ex leaf miner on Ilex vomitoria Solan- Discussion. — Mauleus iligneus differs der), 1 female, 1 male; Sanderson, 9.V.- from M. cultratus in the following: 1. The 1912, 1 female. VIRGINIA: Norfolk, V.- ratio of the eye height to the genal dis- 1929 (ex Phytomyza ilicis), 2 females, 6 tance is more in M. iligneus [x= males; Richmond, V.1938 (ex P. ilicis), 10 2.56±(S.E.)0.022 (n = 12); range 2.5-2.7] females, 4 males. than in M. cultratus [x = 2.24 ±0.038 (n = 7); Biology. — This species has been reared range 2.1-2.4]. 2. Mauleus iligneus has a numerous times from the complex of Phy- partial second row of setae in the costal tomyza mining the leaves of holly [Potter cell of the fore wing, which is lacking from and Gordon 1985 (as Sphegigaster sp.), Un- M. cultratus. 3. The ratio of the length of derhill 1943 (as Sphegigastrinae, new ge- the head and mesosoma to the length of nus, new species), and Langford and Cory the gaster is relatively greater in M. ilig- 1936 (as Sphegigaster sp.)]. It may also be neus [x = 1.23 ±0.03 (n = 10); range 1.2- the species called Halticoptera sp. by Kulp 1.4] (Fig. 21) than in M. cultratus (1968), but voucher specimens from Kulp [x=0.918±0.035 (n=6); range 0.8-1.0](Fig. have not been seen. Host records prior to 8). 4. The ratio of the gastral length divid- 1968 are questionable because the Phy to- ed by its width is less for M. iligneus myza complex on holly was not studied in [x=1.76±0.04 (n=10); range 1.7-1.9] than detail until then (Kulp 1968). for M. cultratus [x=2.56±0.11 (n=6); range Potter and Gordon (1985) reported that 2.2-3.1)]. 5. The hypopygium extends to Mauleus iligneus was a primary parasite about % the length of the gaster in M. //- when its agromyzid host, the native holly /^news (Fig. 21), but is nearly even with the leafminer, Phytomyza ilicicola, was unpar- tip of T7 in M. cultratus (Fig. 8). 6. The asitized, but was a facultative hyperpar- ovipositor sheaths are exserted for a asite on Opius striativentris Gahan (Hy- length equal to Va the length of the hind menoptera: Braconidae) when that species tibia in M. iligneus (Fig. 21), but for about had already parasitized the agromyzid V3 to Vi the length of the hind tibia in M. maggot. Potter and Gordon reared no cultratus (Fig. 8). Mauleus iligneus from puparia collected Etymology. — The specific epithet for just a few days prior to the emergence of Mauleus iligneus is an adjective based on Phytomyza ilicicola and speculate that Mau- the Latin word ilex, meaning holly. leus iligneus parasitizes the fly late in its Type Material. — The holotype (USNM), pupal stage. Volume 4, 1995 21 Mauleus maderensis Graham Fig. 23 Mauleus maderensis Graham, 1981:8. Holotype, female (BMNH) Hym. Type No. 5.3454 (examined). Boucek and Rasplus, 1991:41. Diagnosis. — Mauleus maderensis differs from the three newly described species in that it has the sides of the pronotum in dorsal view as for most other Pteromali- dae — more or less parallel, but weakly convex, with the broadest point being about halfway between its anterior and posterior edge (Fig. 23). The three new species of Mauleus described in this paper have the pronotum in dorsal view broad- est near its anterior margin and distinctly convergent posteriorly. Mauleus luaderensis is similar in propodeal structure to the Ca- ribbean species M. niritus. However, these two species are easily distinguished by the characters given in the key. Distribution. — Madeira (Pico das Arru- das, near Sao Martinho), and possibly Mexico (Boucek and Rasplus, 1991) be- cause the species is associated with plants of Mexican origin (Boucek, pers. comm.). Biology. — The insect host(s) of M. mad- erensis remain unknown. Mauleus nigritus (Howard), new combination Polycystus nigritus Howard, 1897:142. Holotype, female (BMNH); Hym. Type No. 5.876 (examined). Heydon, 1989:193. Diagnosis. — The type of Mauleus nigritus is fragmented with most of the antennae gone and the gaster and petiole mounted on the card separately from the remainder of the body. Enough is remaining how- ever, to confirm that this species belongs in Mauleus. Mauleus nigritus differs from the three newly described species in that it has the sides of the pronotum in dorsal view more or less parallel, with the broad- est point being about halfway between its anterior and posterior edge. The three new species of Mauleus described in this paper have the pronotum in dorsal view broad- est near its anterior margin and distinctly convergent posteriorly. Mauleus nigritus is similar in pronotal structure to M. mader- ensis; however, these two species are eas- ily distinguished by the characters given in the key. Distribution. — Mauleus nigitus is known to me only from the holotype collected at Baltazar, on the windward side of Grena- da (West Indies) by H. H. Smith. Biology. — The insect hosts of M. nigritus remain unknown. Mauleus venetus Heydon, new species Fig. 25 Holotype, female. — Color: Body black ex- cept anterior aspect of head, collar, me- soscutum steel blue and lateral portions of metanotum, propodeum, Tl blue. Anten- na with scape, ventral side of pedicel and Fl brownish yellow, remainder of pedicel and flagellum brown. Legs with coxae dark blue; femora brown with metallic blue reflections, except basal and apical tips brownish yellow; tibiae brownish yel- low with weak brown bands mesally; tarsi white with pretarsi brown. Fore wing clear with veins brownish yellow, paras- tigma darker. Sculpture. — Clypeus and immediate vi- cinity finely alveolate; remainder of head, mesoscutum, scutellum, frenum, median panels of propodeum alveolate; petiole finely alveolate; gaster smooth except T6 and T7 coriaceous. Structure. — Body length 2.2 mm. Head width 1.4X height (42:30), 2.5 X length (42.0:16.5); eye height 1.5X length (20:13), 2.0X malar distance (20:10), length 7.5X temple length (13:2); torulus 2X own di- ameter above LOcL; ratio of MOD, OOL, POL, LOL as 3.0:6.5:9.0:4.0. Antenna with length of pedicel plus flagellum 0.90 X head width (38:42); relative lengths of scape, pedicel, annelli, Fl-6, club as 15.0: 4.0:1.0:4.5:4.5:4.5:4.0:4.0:3.5:8.0; relative widths of Fl, F6, club as 3.5:4.0:4.0; MPP sensilla in two rows on each flagellar seg- 22 Journal of Hymenoptera Research merit; terminal club segment with small ventral patch of micropilosity. Mesosoma length 1.5 X width (26.5:18.0); pronotum with humeral angles acute (Fig. 25); dor- sellum short, anterior and posterior edges parallel; propodeum with region between basal foveae depressed (this depression shorter mesally than laterally and bound- ed posteriorly by weak carina), plicae fad- ing out before reaching anterior margin of propodeum, nuchal region raised, mar- gined anteriorly by carina, spiracles al- most on anterior margin of propodeum. Fore wing with relative lengths of sub- marginal, marginal, postmarginal, stigmal veins as 37:23:15:8; costal cell with 1 com- plete and 1 distal partial row of setae; bas- al cell bare; basal vein with row of setae; speculum open posteriorly. Petiole length 1.1 X width (9:8). Gaster length 1.4X width (43:31), 0.66 X combined length of head and mesosoma (43:65); hypopygium ex- tending about Vi gastral length; ovipositor sheaths hardly extending beyond hind margin of T7. Allotype, male. — Similar to female ex- cept: body length 1.8 mm. Head with ratio of MOD, OOL, POL, LOL as 3.5:4.0:7.0:3.0. Antenna with length of pedicel plus fla- gellum 1.1 X head width (37:34); relative lengths of scape, pedicel, annelli, Fl-6, club as 11.0:3.0:1.0:4.5:4.5:4.0:4.0:4.0:4.0:9.0; relative widths of Fl, F6, club as 3.0:3.5: 3.0; MPP sensilla numerous and distrib- uted over funicular segments in many rows; setae reclinate, nearly absent. Gaster length 1.1 X width (29:26). Diagnosis. — Mauleus venetus is distin- guished from the other two Nearctic Mau- leus species by the acute humeral angles of the pronotum, the broad depression along the anterior margin of the propo- deum, and the lack of sculpture over most of the median panels of the propodeum. Mauleus venetus most closely resembles M. iligneus — the coloration of the two species is almost identical except the propodeum is brighter blue in M. venetus. In addition, the MPP sensilla of M. venetus are ar- ranged in two or more rows on each fu- nicular segment in both sexes, whereas they are arranged in only one row in M. iligneus. Etymology. — The species name comes from the Latin word venetus, meaning blue or sea-blue, and refers to the distinc- tive blue propodeum in this species. Type Material. — The holotype (IRCW) is a female, collected in Grant Co. (T6N, R6W, SI 7), Wisconsin (USA), in a malaise trap exposed 7-14.VL1976. The allotype (UCDC) was collected by the author on the South Farms of the University of Illi- nois, near Champaign, Illinois (USA) on 26.V.1985. Six paratypes were collected as follows (CNCI, IRCW, UCDC, USNM): Canada. QUEBEC: La Trappe, 12.VII.1942, 1 female. United States. ILLINOIS: Uni- versity of Illinois South Farms, near Champaign, 26.V.1985, 1 male. MICHI- GAN: Midland Co., 2.VII.1943, 1 male. VIRGINIA: Winchester, 16.VI.1964, 1 male. WISCONSIN: Grant, T6N, R6W, S17, 3-8.VI.1976 (gypsy moth Malaise trap), 1 female, 14-21. VI.1976 (gypsy moth Malaise trap), 1 female. Biology. — The host(s) of M. venetus are unknown. ACKNOWLEDGMENTS I thank Melissa Bennett and two anonymous re- views for their careful reading of this manuscript. I also thank the following people for the loan of ma- terial used in this study: Dr. J. S. Noyes, The Natural History Museum, London, ENGLAND (BMNH); Dr. W. J. Pulawski, California Academy of Sciences, San Francisco, CA (CASC); Dr. F. G. Andrews, California State Collection of Arthropods, Sacramento, CA (CDAE); Dr. G. A. P. Gibson, Canadian National Col- lection, Ottawa, ON (CNCI); Dr. W. E. LaBerge, Illi- nois Natural History Survey, Champaign, IL (INHS); S. Krauth, University of Wisconsin, Madison, WI (IRCW); Dr. R. W. Brooks, Snow Entomological Museum, University of Kansas, Lawrence, KS (SEMC); Dr. E. E. Grissell, United States National Mu- seum, Washington, D.C. (USNM). The acronym for the collection of the Bohart Museum at the University of California, Davis, CA is UCDC. Volume 4, 1995 23 LITERATURE CITED Allen, M. W. and R. H. Painter. 1937. Observations on the biology of the wheat maggot in Kansas. Journal of Agricultural Research 55:215-238. Allen, W. A. and R. L. Pienkowski. 1973. Parasites reared from puparia of the frit fly, Oscinella frit, in Virginia. Environmental Entomology 2:615-617. Anderson, H. 1990. Three new species of Halticopter- ina Erdos (Hymenoptera: Pteromalidae). Pan-Pa- cific Entomologist 66:131-139. Askew, R. R. 1970. Observations on the hosts and host food plants of some Pteromalidae [Hym., Chalcidoidea]. Entomophaga 15:379-385. Boucek, Z. 1970. Contribution to the knowledge of Italian Chalcidoidea, based mainly on a study at the Institute of Entomology in Turin, with de- scriptions of some new European species. Me- morie della Societa Entomologica Italiana 49:35-102. Boucek, Z. 1976. African Pteromalidae (Hymenop- tera): new taxa, synonymies and combinations. Journal of the Entomological Societi/ of Southern Af- rica 39:9-31. Boucek, Z. 1977. A faunistic review of the Yugosla- vian Chalcidoidea (parasitic Hymenoptera). Acta Entomologica Jugoslavica 13(Suppl.): 3-145. Boucek, Z. 1993. New taxa of North American Pter- omalidae and Tetracampidae (Hymenoptera), with notes. Journal of Natural History 27:1239- 1313. Boucek, Z. and J.-Y. Rasplus. 1991. Illustrated Key to West-Pa lea rctic Genera of Pteromalidae (Hy- menoptera: Chalcidoidea). Institut National de la Recherche Agronomique. Paris. Burks, B. D. 1979. Family Pteromalidae. pp.768-835. In, K. V. Krombein, P. D. Hurd, D. R. Smith, and B. D. Burks (eds.). Catalog of Hi/menoptera in America North of Mexico. Smithsonian Institution Press. Washington, D. C. De Santis, L. 1979. Catdlogo de los Himenopteros Cal- cidoideos de America al sur de los Estados Unidos. Comision de Investigaciones Cienti'ficas de la Provincia de Buenos Aires. La Plata. Dzanokmen, K. A. 1978. Identification of the insects of the European part of the USSR. Vol. 3. Hy- menoptera. Second Part. Pteromalidae. Opredeli- teti Faune SSSR 120:57-228. Gahan, A. B. 1933. The serphoid and chalcidoid par- asites of the Hessian fly. U.S. Department of Ag- riculture, Miscellaneous Publications 174:1-147. Girault, A. A. 1918. New and old West Indian and North American chalcid-flies (Hym.). Entomolog- ical News 29:125-130. Graham, M. W. R. de V. 1956. A revision of the Walker types of Pteromalidae (Hym., Chalcidoi- dea). Part 2 (including descriptions of new gen- era and species). Entomologist's Monthly Magazine 92:246-263. Graham, M. W. R. de V. 1969. The Pteromalidae of northwestern Europe (Hymenoptera, Chalcidoi- dea). Bulletin of the British Museum (Natural His- tory) Entomology Supplement 16:1-908. Graham, M. V. R. de V. 1981. A survey of Madeiran Chalcidoidea (Insecta: Hymenoptera) with addi- tions and descriptions of new taxa. Bocagiana 58: 1-20. Hedqvist, K.-J. 1975. Notes on Chalcidoidea 7. A key to Swedish species of the genus Halticoptera Spin. and three related genera (Hymenoptera: Pter- omalidae). Entomologica Scandinavica 6:167-181. Hedqvist, K.-J. 1983. Malmsteklar pa Olands Stora alvar. Entomolgisk Tidskrift 104:165-168. Hendrickson, Jr., R. M. 1979. Field studies and par- asites of Liriomyza trifoliearum (Diptera: Agro- myzidae) in northern USA. Journal of the Neiv York Entomological Society 87:299-303. Heydon, S, L. 1988. A review of the world species of NotoglypHus Masi (Hymenoptera: Pteromali- dae). Proceedings of the Entomological Society of Washington 9l:U2-\23. Heydon, S. L. 1989. Relationships among Holarctic genera in the Cyrtogaster-group with a review of the species of North America north of Mexico. (Hymenoptera: Pteromalidae). Journal of the New York Entomological Society 97:192-217. Heydon, S. L. 1993. Syntomopus Walker: The nearctic species with a review of known host associations (Hymenoptera: Pteromalidae). Journal of Hyme- noptera Research 2:107-116. Heydon, S. L. and W. E. LaBerge 1988. A review of North American species of Sphegigaster north of Mexico and the biology of their hosts (Hyme- noptera: Pteromalidae). Journal of the Kansas En- tomological Society 61: 258-277. Howard, L.O. 1897. On the Chalcididae of the island of Grenada, B.W.I. Journal of the Linnean Society of London. Zoology 26:129-178. Kamijo, K. 1978. Chalcidoid parasites (Hymenop- tera) of Agromyzidae in Japan, with description of a new species. Kontyu 46:455—469. Kulp, L. A. 1968. The taxonomic status of dipterous holly leaf miners (Diptera: Agromyzidae). Uni- versity of Maryland Agricultural Experiment Station Bulletin A-155:l-42. Langford, G. S. and E. N. Cory. 1936. The holly leaf- miner and its control. Proceedings of the National Shade Tree Conference 13:109-112. Morrill, W. L. and R. W. Kieckhefer. 1971. Parasitism of the wheat stem maggot in South Dakota. Jour- nal of Economic Entomology 64:1129-1131. Nikol'skaya, M. N. 1937. The chalcidoid parasites (Hymenoptera) of some injurious flies of grain crops. Entomological Review 27:1-27. Peck, O. 1951. Superfamily Chalcidoidea. pp. 410- 594. In, C. F. W. Muesbeck, K. V. Krombein, and H. K. Townes. Hymenoptera of America North of 24 Journal of Hymenoptera Research Mexico, Synoptic catalog. U.S. Department of Ag- riculture, Agriculture Monograph 2:1-1420. Peck, O. 1963. A catalogue of the Nearctic Chalci- doidea (Insecta: Hymenoptera). Canadian Ento- mologist Supplement 30:1-1092. Peck, O., Z. Boucek, and A. Hoffer. 1964. Keys to the Chalcidoidea of Czechoslovakia. Memoirs of the Entomological Society of Canada 34:1-120. Potter, D. A. and F. C. Gordon. 1985. Parasites as- sociated with the native holly leafminer, Phyto- myza ilicicola Loew (Diptera: Agromyzidae), on American holly in Kentucky. Journal of the Kansas Entomological Society 58:727-730. Schmiedeknecht, O. 1909. Hymenoptera family Chalcididae. Genera Insectorum 97:1-550. Spencer, K. A. 1976. The Agromyzidae (Diptera) of Fennoscandia and Denmark. Fauna Entomologica Scandinavica 5:1-606. Spencer, K. A. and G. C. Steyskal. 1986. Manual of the Agromyzidae (Diptera) of the United States. U.S. Department of Agriculture, Agriculture Hand- book 638:1^78. Takada, H. and K. Kamijo. 1979. Parasite complex of the garden pea leaf-miner, Phytomyza horticola Goureau in Japan. Kontyu 47:18-37. Thomson, C. G. 1876. Hymenoptera Scandinaviae. 4. Pteromalus (Svederus). Lund. Thomson, W. R. 1958. Catalogue of parasites and pred- ators of Insect pests. Section 2. Host parasite cata- logue. Part 5. Hosts of the Hymenoptera (Mis- cogasteridae to Trigonalidae), Lepidoptera and Strepsiptera. p. 562-698. Commonwealth Insti- tute of Biological Control. Ottawa, Ontario. Underhill, G. W. 1943. Some insect pests of orna- mental plants. Virginia Agricultural Experiment Station Bulletin 349:27-38. Walker, F. 1833. Monographia Chalciditum. Ento- mologist's Magazine 1:367-384, 455^66. Walker, F. 1839a. Monographia Chalciditum, 1. Lon- don. Walker, F. 1839b. Monographia Chalciditum, 2. Lon- don. Walker, F. 1843. Descriptions of Chalcidites discov- ered near Conception, in South America, by C. Darwin, Esq. Annals and Magazine of Natural His- tory 12:30-32. Walker, F. 1846. List of the specimens of Hymenopterous insects in the collection of the British Museum. Part I-Chalcidites. London. J. HYM. RES. Vol. 4, 1995, pp. 25-32 An Australian Masarine, Rolandia angulata (Richards) (Hymenoptera: Vespidae): Nesting and Evaluation of Association with Goodenia (Goodeniaceae) Friedrich W. Gess, Sarah K. Gess and Robert W. Gess Albany Museum, Grahamstown, South Africa Abstract. — An account is given of some aspects of the nesting of Rolandia angulata (Richards) (Masarinae). This wasp nests in multicellular burrows in compacted sandy soil. The sloping en- trance to the burrow, not surmounted by a turret, is concealed beneath a pebble, a plantlet or a suitable item of debris. The main thrust of the shaft is vertically downwards, however, at intervals it curves outwards to end in a horizontal cell so that each cell, except the last excavated and therefore deepest one, appears to be accessed by a lateral shaft. The cells are unlined. The archi- tecture of the nest and the method of its construction are discussed. The association between Rolandia angulata and Goodenia (Goodeniaceae) flowers, the source of nectar and pollen, is evalu- ated. It is shown that the association with Goodenia pinnatifida Schldl., at least, is mutually bene- ficial. Indeed it is suggested that in some areas, at some times, R. angulata may be the most important potential pollinator of this plant. The genus Rolandia Richards, 1962 is (28.03S, 148.30E), and 85 km E of St listed by van der Vecht and Carpenter George (28.03S, 148.30E) (27-29.X.1993, (1990) as a junior synonym of Metaparagia F.W., S.K. and R.W. Gess). Meade-Waldo, 1911. As the analysis on There is only one published observation which this is based has not yet been pub- on the nesting of Rolandia — that of R. ma- lished the generic name Rolandia is used culata entering a turretless, simple, in the present paper. Rolandia is restricted oblique, blindly ending burrow in sandy to Australia. Four species are known, two, ground (Houston, 1984). However, Hous- R. maculata (Meade-Waldo) and R. hous- ^^^ is currently investigating further the toni Snelling, from Western Australia, a riesting of this species and has made some third, R. borreriae Snelling, from Northern ^^^^i^^ available for comparison. Territory, and the fourth, R. angulata ^^ investigations concerning R. angii- (Richards), the subject of this paper, from ^"^" published m the present paper were Queensland and New South Wales. undertaken by the authors during the r, .^ J- 1 I u u J J course of a fieldtrip to Australia in Octo- Rolandia angulata has been recorded , ^^^^ w , £ .u .. r-i 1 J J ber 1993. Voucher specimens have been from southwestern Queensland and , . , . , »„ ,, ^ - , o .1 T»r , r deposited in the Albany Museum (R. an- northwestern New South Wales, from [,.,<- ^ . i .i i n u no o/ic 1/ir /.niri /o- u j gulata, the forage plants, and the bee visi- Cunnamulla [28.04S, 145.40E (Richards ; «. i.u r i x. x .^u a .. i- .^^„ T^. , . , , ^ ,. . tors to the forage plants), the Australian 1968 as Riekia auQulata and SneWine, 1986) xt .• it .. V- n ..• /- u /o , , '^ t> / National Insect Collection, Canberra [R. m the north to 90 km W of Cobar, Barnato ,,,^,,,,^,,1 and the Australian National Tanks [31.38S, 144.59E] (about 400 km Herbarium, Canberra (the forage plants). south of Cunnamulla) (Richards 1968) in the south, and from three sites to the east NESTING of Cunnamulla, 80 km E of Cunnamulla Description of the Nesting Areas. — Two (28.04S, 145.40E), 27 km W of St George nesting areas of Rolandia angulata were lo- 26 Journal of Hymenoptera Research Figs. 1-4. 1, Eucalyptus woodland between St George and Cunnamulla, southwestern Queensland; 2, Acacia scrub 80 km east of Cunnamulla, southwestern Queensland; 3, a nesting site of Rolandia ai^gulata 27 km west of St George, southwestern Queensland; 4, a nest entrance of Rolandia angulata beneath a piece of plant debris (approximately X 3.6). cated between St George and Cunnamulla in southern Queensland, one 27 km west of St George on 28 October and the other 80 km east of Cunnamulla on 29 October, the former in dry open woodland domi- nated by Eucalyptus (Fig. 1) and the latter in dry open scrub dominated by Acacia (Fig. 2). Both areas had recently received rain in the form of localized thundershow- ers which had resulted in a growth of an- nuals which were in flower on the road verges and in the low lying areas. The soil in both areas was sandy, compact but fri- able. It was increasingly moist at least to the depth of the deepest nest cells, 38 cm. Flozvers Visited. — The only previous rec- ord of flower visiting by R. angulata is that of Richards (1968) for females collected on Goodenia cycloptera R.Br, in C.Sturt (Good- eniaceae) at Barnato Tanks. In the present study females and males were collected and observed visiting the yellow flowers of a prostrate herb also identified as G. cy- cloptera (Fig. 11) and an erect herb Good- enia pinnatifida Schldl. (Fig. 6) in the two areas where nests were discovered and Volume 4, 1995 27 also in the area 85 km east of St George. At the most easterly site, that is 85 km east of St George, both species of Goodenia were common. Travelling westwards to Cunnamulla it was observed that G. cy- cloptera rapidly became uncommon whereas G. puinntifida became increasingly common and at the site 80 km east of Cun- namulla was the most common roadside herb. The abundance of R. angiilata fol- lowed a similar pattern to that of G. pin- natifida. Visits to the flowers by female R. angii- lata were abundant by lOhOO and contin- ued through the heat of the day and the afternoon. Visits became fewer in the late afternoon and ceased after IThOO. Males were observed to patrol the flowers and only later in the day to visit them for nec- tar. All plants flowering together with the two Goodenia species, most notably several Asteraceae and a Wahlenbergia species (Campanulaceae), were sampled for flow- er visitors. None was being visited by R. angulata. Provision. — Provision in the form of a firm, white pollen loaf was obtained from each of three cells. Pollen from the loaves was examined microscopically and found to match that obtained from the Goodenia flowers. Mate location. — During the morning males patrolled the flowers where they sought the females. They did not alight on the flowers but rested on a neighbouring plant, for example on a grass stem, or on the ground. Numerous attempted copu- lations at flowers were observed and a sin- gle attempt to copulate with a nest-exca- vating female was noted. Several instances of a male and female grappling on the ground next to plants and of "hot pur- suit" were noted. Description of the Nest. — The nest (n = 8) consists of a subterranean burrow (Fig. 5) excavated in horizontal sandy soil in a clearing (Fig. 3). The entrance is concealed beneath a pebble, a plantlet or a suitable item of debris (Fig. 4). It is a simple hole 3-4 mm in diameter, not surmounted by a superstructure. For approximately the first 10 mm the shaft slopes gently down- wards. Thereafter its main thrust is verti- cally downwards, however, at intervals it curves outwards to end in a horizontal cell so that each cell, except the last excavated and therefore deepest one, appears to be accessed by a lateral shaft. These "lateral shafts" radiate out through 360° each deeper than that preceding it. The diameter of the shaft narrows at a depth of approximately 80 mm and then continues constant. The first cell in the nests investigated was at a depth of be- tween 180 and 370 mm, the "lateral shafts" were 30 mm in length, and the cells 13 mm long and 4 mm in diameter at the widest point. A lateral shaft is packed with sand after the cell which ter- minates it has been provisioned. Up to seven cells per nest were recorded. The cells are unlined. Method of Construction of the Nest and Provisioning. — Water is not required for the excavation of the nest as the sandy soil, though compact, is friable. The sand extracted from the burrow is carried out of the shaft held between the head and the prosternum, the genae being fringed with ammochaetae. Whilst excavation is in progress, the female, when leaving the nest, backs out. During the initial stages of burrow excavation the extracted sand is dropped in flight in a more or less con- stant area to one side of the entrance and about 120 mm from it. As the burrow be- comes deeper the extracted sand is dropped further from the nest in a con- stant arc about 250 mm from the entrance. The sand being scattered, there is no de- tectable accumulation and so no tumulus develops. The presence of a recently hatched larva in a sealed, fully provisioned cell indicat- ed that mass provisioning is practised. The sand used for packing a shaft leading to a cell is almost certainly obtained with- 28 Journal of Hymenoptera Research in the nest during the excavation of the shaft leading to the next cell. The fully fed larva spins a white cocoon which completely fills the cell. Like the sealed cell it is therefore rounded at the inner end and truncate at the outer end. Discussion of Nesting. — The nest of R. an- gulata is essentially similar to that of R. maculata, which is described by Houston (pers. comm.) as a vertical burrow in sandy soil, about 300 mm deep and un- lined, with unlined cells at the lower end. Thus the basic nest type for these two spe- cies of Rolnndia can be defined as a mul- ticellular sub-vertical burrow in horizontal ground excavated by the nester, without an entrance turret and with excavated cells not containing constructed cells. When compared with the seven basic nest types recognized for the Masarinae as a whole by Gess and Gess (1992) this ap- pears to fit Nest type 1 except for the lack of an entrance turret. When the method of construction is compared with that of the species listed for this nest type, a basic dif- ference is apparent. Water is used in the excavation and construction of Nests of type 1, like nests of types 2 and 3, whereas the nests of the two Rolandia species are excavated without the use of water. This is possible due to the friable nature of the soil in which they are sited. As noted in Gess and Gess (1992) nesting in friable soil in the Vespidae is probably derived rather than primitive as in the Pompilidae and Sphecidae. The nest type of the two Rolan- dia species is therefore seen as a sub-type which can be derived from Nest type 1. Both species of Rolandia carry sand, ex- tracted from the shaft, out of the burrow held between the head and the proster- num, the genae being fringed with am- Fig. 5. Plan of vertical section of a nest of Rola7tdia attgulata. Cells 2 — 5 followed cell I, radiating out in sequence through 360° and at successively greater depths. 50mm 29 Figs. 6-10. 6, Goodenia pinnatifida, an erect herb; 7, a flower of Goodenia piiiiiatifida, indusium concealed by the bases of the adaxial petals; 8, Rolandia angulata entering a flower of Goodenia pinnatifida, indusium of the flower exposed (X 2.8); 9, Rolandia angulata in nectar drinking position in a flower of Goodenia pinnatifida, indusium of the flower fitting snugly over the wasp's mesosoma (X 2.8); 10, Rolandia angulata withdrawing from a flower of Goodenia pinnatifida, showing dusting of pollen on the head, dorsum of the prothorax, and anterior part of the mesoscutum (X 2.8). mochaetae. The possesion of ammochae- tae fringing the genae is a generic char- acter (Snelling 1986) and it is therefore ex- pected that all members of the genus excavate their nests in a similar manner. The only other vespid genus recorded as having a psammophore for the removal of the spoils of excavation is Pterochihis (Eu- meninae) (Bohart 1940) for which nesting in vertical burrows in friable soil by two species has been described (Isely 1914 and Evans 1956). EVALUATION OF ASSOCIATION WITH FLOWERS VISITED Both R. angulata males and females ob- tain nectar and possibly pollen for their own nourishment and females collect nec- tar and pollen for provisioning their nest- cells apparently solely from Goodenia flow- ers. They are therefore probably depend- ant on Goodenia flowers. To determine whether or not the association between the wasp and the flowers is mutually ben- eficial or not, that is whether or not the wasp in addition pollinates the flowers, requires a consideration of the functional morphology of the flowers, the behaviour of the wasps in the flowers, and wasp/ flower fit. Goodenia flowers are distinctly two lipped. The lower wings of the two ad- axial petals are differentiated in their low- er parts into auricles which envelop the 30 Journal of Hymenoptera Research Figs. 11-13. 11, Goodenia cycloptera, a prostrate herb; 12, three flowers of Goodenia cycloptera, indusium con- cealed by the bases of the adaxial petals; 13, Rolandia angulata in nectar drinking position in a flower of Goodenia cycloptera, indusium of the flower pressed down on the wasp's folded wings (X 3.3). indusium in the flower (Figs 7 and 12) so that it is only exposed when an insect vis- itor pushes the corolla lobes apart as it seeks the nectar at the base of the flower (Figs 8 and 13). The indusium is a cup at the top of the style and the surrounding stigmatic initial that collects and retains pollen from the stamens, which dehisce in the bud, and presents it thus to the pollen vector (Carolin, Rajput and Morrison 1992). Later the stigmatic initials mature and grow out of the indusium and collect pollen from pollen vectors. R. angulata when visiting the flowers for nectar always alights on the lower lip and then pushes its way in beneath the indu- sium so that, if the flower is in the pollen presenting phase, the wasp receives pollen on its head. When it is in the nectar drink- ing position in a flower of G. pinnatifida, the mesosoma (Fig. 9) is snugly capped by the indusium but, when it is in a flower of G. cycloptera, which is deeper, it is the folded wings which are pressed beneath the indusium (Fig. 13). On emerging from a flower of G. pinnatifida it can be clearly Volume 4, 1995 31 seen to have been well dusted with pollen not only on the head but also on the dor- sum of the prothorax and the anterior part of the mesonotum (Fig. 10). When coming thus laden with pollen to a flower with receptive stigmas the wasp would be ide- ally suited to pollinate it. It is not clear, however, whether R. angulata would effec- tively pollinate the deeper flowers of G. cycloptera. It is possible that pollen may be transferred from its head to a receptive stigma but the pollen received on the wasp's wings is unlikely to be successfully transferred. It is concluded that, within its limited distribution, R. angulata is a potential pol- linator, at least, of the widespread species G. pinnatifida with which it therefore has a mutually beneficial association. However, at none of the sites was R. angulata the sole visitor to the Goodenia flowers. At the most easterly site a second ma- sarine, Riekia nocatunga Richards, which was also recorded from Goodenia fascicu- laris F.Muell. & Tate at Kondar to the southeast, was a relatively abundant vis- itor to G. pinnatifida. Furthermore the flowers of both species of Goodenia, par- ticularly those of G. cycloptera, were vis- ited abundantly by bees. The most com- mon species was Leioproctus {Chri/socolle- tes) moretonianus (Cockerell) (Colletidae) which was also recorded from Goodeni- aceae at several sites in Queensland by Michener (1965). Less common visitors were a second, but slightly smaller, spe- cies of Leioproctus {Chrysocolletes) and a species of Megachile (Megachilidae). Rel- atively uncommon visitors were six fur- ther species of Megachile, an additional coUetid, a halictid, a few anthophorids and the honey-bee. Travelling westwards the bees were uncommon at the site 27 km west of St George and at the site 80 km east of Cunnamulla all but the second species of Leioproctus (Chrysocolletes) were absent. At the latter site a third much smaller species of Leioproctus was also re- corded. However, R. angulata was the only abundant visitor suggesting that in some areas, at some times, it may be the most important potential pollinator of G. pinnatifida. ACKNOWLEDGMENTS The following are thanked with appreciation: Terry Houston of the Western Australian Museum for per- mission to quote his unpublished observations con- cerning the nesting of Rolandia maciilata; Jo Cardale of the Australian National Insect Collection, Canberra for facilitating the identification of forage plants; Kir- sten Cowley, Australian National Herbarium, Can- berra for the identifications of the forage plants; Alan Weaving of the Albany Museum for producing black and white negatives from the authors' colour trans- parencies for Figs 1-4 and 6-13; and the South Afri- can Foundation for Research Development for a roll- ing support grant to F.W.Gess, which made possible the fieldwork in Australia. The manuscript was re- viewed by Karl V. Krombein and an anonymous re- viewer, who are thanked for their constructive com- ments. LITERATURE CITED Bohart, R.M. 1940. A revision of the North American species of Pterocheilus and notes on related gen- era (Hymenoptera, Vespidae). Annals of the En- tomological Society of America 33(1): 162-208. Carolin, R.C., M.T.M. Rajput, and D. Morrison. 1992. Goodeniaceae. In: George, A.S. ed., Flora of Aus- tralia, volume 35: Brunoniaceae, Goodeniaceae. Can- berra: Australian Government Publishing Ser- vice, pp. 4-300. Evans, H.E. 1956. Notes on the biology of four spe- cies of ground-nesting Vespidae (Hymenoptera). Proceedings of the Entomological Society of Washing- ton 58(5): 265-270. Gess, F.W. and S.K. Gess. 1992. Ethology of three southern African ground nesting Masarinae, two Celonites species and a silk-spinning Quartinia species, with a discussion of nesting by the sub- family as a whole (Hymenoptera: Vespidae). Journal of Hymenoptera Research 1(1): 145-155. Houston, T.F. 1984. Bionomics of a pollen-collecting wasp, Paragia tricolor (Hymenoptera: Vespidae: Masarinae), in Western Australia. Records of the Western Australian Museum 11(2): 141-151. Isely, D. 1914. The biology of some Kansas Eumen- idae. Kansas University Scietice Bulletin (2)8(7): 233-309. Meade-Waldo. G. 1911. Notes on the family Masar- idae (Hymenoptera), with descriptions of a new genus and three new species. Annals and Maga- zine of Natural History (8)8: 747-750. Michener, CD. 1965. A classification of the bees of the Australian and South I'acific regions. Bulletin 32 Journal of Hymenoptera Research of the American Museum of Natural History 130: 1- 362. Richards, O.W. 1962. A revisional study of the Masarid wasps {Hymenoptera, Vespoidea). London: British Museum (natural History). Richards, O.W. 1968. New records and new species of Australian Masaridae (Hymenoptera: Vespo- idea). Journal of the Australian Entomological Soci- ety 7: 101-104.' Snelling, R.R. 1986. The taxonomy and nomenclature of some Australian paragiine wasps (Hymenop- tera: Masaridae). Contributions in Science 378: 1- 19. van der Vecht, J. and J.M. Carpenter. 1990. A cata- logue of the genera of the Vespidae (Hymenop- tera). Zoologische Verhandelingen 26: 1-62. J. HYM. RES. Vol. 4, 1995, pp. 33-10 Descriptions of the Male of Riekia nocatiinga Richards, the Male and Two Strikingly Distinct Sympatric Colour Forms of Riekia confluens (Snelling) and the Male of Rolandia angiilata (Richards) (Hymenoptera: Vespidae: Masarinae) from Australia Friedrich W. Gess Albany Museum, Grahamstown, 6140, South Africa Abstract. — The male of Riekia nocatunga Richards from southern Queensland, the male and two strikingly distinct sympatric colour forms of Riekia confluens (Snelling), comb, nov., from Western Australia, and the male of Rolandia angnlata (Richards) from southern Queensland are described. The characters distinguishing R. nocatunga and R. confluens are discussed. The Australian Masarinae and their nat- ural history are relatively poorly known. During a recent collecting trip to Australia by the author, S.K.Gess and R.W.Gess, material collected included males of three species, Riekia nocatunga Richards, Riekia confluens (Snelling), comb, nov., and Rolan- dia angnlata (Richards), hitherto described only from the female sex. Two strikingly distinct sympatric colour forms of Riekia confluens were found to occur. One of these was previously unknown and helps to elucidate the identity of a single male Riekia collected by T.F.Houston and dis- cussed by Snelling (1986) who was unable to place it in a species. Riekia Richards, 1962 and Rolandia Rich- ards, 1962 have been sunk as junior sub- jective synonyms of Metaparagia Meade- Waldo, 1911 by van der Vecht and Car- penter (1990) on the basis of an as yet un- published cladistic analysis by Carpenter. Until such time as this analysis is pub- lished and may be studied, the present au- thor considers it best to continue to accept Riekia and Rolandia as genera in their own right. Institutions in which the material stud- ied is deposited are: Albany Museum, Grahamstown, South Africa (AMG); Aus- tralian National Insect Collection, Canber- ra (ANIC); Western Australian Museum, Perth (WAM). Riekia nocatunga Richards Riekia nocatunga Richards, 1962: 55-57, female. This species was described from 7 miles N of Nocatunga [on maps as Nockatunga] [27.40S, 142.40E], Queensland (not New South Wales as in Richards 1962; correct in Richards 1968) (holotype female and 4 female paratypes) and Bourke [30.09S, 145.59E], New South Wales (1 female paratype). Subsequently an additional five females from 90 km W of Cobar, Barnato Tanks [31.38S, 144.59E], New South Wales, were attributed to this species by Richards (1968). He noted that they differed from his original description with regard to the pale markings of which he gave some de- tails. Two of these females, in the collec- tion of the Natural History Museum, Lon- don, have been examined by the present author. They are confirmed as being R. no- catunga. They are less melanistic than the specimens originally described. The dark areas are less extensive and some, in par- 34 Journal of Hymenoptera Research ticular those on the pronotum and tergites each side of mesoscutum where its fur- 1 and 2, are brownish rather than black, rows meet pronotum (usually but not al- The light areas are more extensive and ways present in females) absent in both their colour is a strong yellow rather than males examined, pale lemon-yellow. Antennal scape (with radicle) very Snelling (1986) figured and briefly dis- slightly longer than interantennal dis- cussed a male Riekia from 43 km ENE Lan- tance; first flagellomere less than half (0.4) dor Homestead (25.08S, 116.54E) in West- as long as scape (with radicle), one and a ern Australia. Whereas it agreed generally third times as long as wide at its distal with the original description of R. nocatun- end, and one and one fifth as long as the ga, Snelling stated that there was no cer- second flagellomere. Last three flagellom- tainty that it was the opposite sex of that eres flattened and slightly concave be- species and that he suspected that it was neath; ultimate flagellomere narrowing not. He listed some discrepancies, allow- apically and distinctly curved to form a ance being made for sexual differences, hook. and stated that more material would have Tergite 7 evenly rounded apically. to be available before the specific status of Genitalia (Figs 3 and 4); parameral spine the male could be determined. broad and thick, evenly curved over its Recently collected Riekia material, of rel- length and not hook-like apically; ventral evance to the clarification of the above tax- process of each paramere relatively broad onomic question, consists of associated fe- and short, subtriangular, without a nar- males and males from both Queensland row elongation; digitus short, and Western Australia. The specimens Length 7.5-7.8 mm; length of forewing from Queensland are certainly R. nocatun- 5.8 mm, hamuli 14. ga, as established by the comparison by New Material Examined. — Queensland: the author of several of the females with Lundavra State School, Kondar (28. OSS, the holotype in the Australian National In- 149.51E), 25.X.1993 (F.W.,S.K.& R.W.Gess) sect Collection, Canberra. The specimens 11 females, 1 male (in yellow flowers of from Western Australia on the other hand Goodenia fascicularis F.Muell. & Tate, represent two strikingly distinct colour Goodeniaceae); Southwood Road, western forms of Riekia confluens. One is superfi- boundary of Southwood National Park cially similar looking to Riekia nocatunga (27.56S, 149.30E), 26.x. 1993 (F.W.,S.K.& but morphologically distinct. The males of R.W.Gess) 2 females (at water); 85 km E this form show the characters noted by of St George (28.03S, 148.30E), 27.X.1993 Snelling for his male. (F.W.,S.K.& R.W.Gess) 8 females, 1 male Male. — (Figs 3 and 4). Colour pattern (in yellow flowers of Goodenia pinnatifida very similar to that of female but differing Schldl., Goodeniaceae). Two females and in that the following parts are pale lemon- 1 male in ANIC; 2 females in WAM; rest yellow: small streak on scapes distally, en- of material in AMG. tire clypeus other than for narrow lateral and anterior margins, single large frontal Rekia confluens (Snelling) comb. nov. spot between and above antennal sockets, .1 . . 1 , .. Paravia confluens Snelling, 1986: 14 and Fie. 28, narrow streak marginmg lower orbits, c \ & ' elongate spot in lower half of each ocular p. , • ^ n- mo^ ^ o j t- mm ^ , ^ . . ^ ^ Riekia sp. Snelling, 1986: 6, 8 and Figs 10, 19- sinus (not rising above upper limit of fron- 21 male tal spot and separated from it by about width of antennal socket), uninterrupted Snelling (1986) erroneously described but posteromedially narrowly emarginate confluens as a species of Paragia Shuck- pronotal band connecting humeri. Spot on ard, 1837 despite the fact that, in his own Volume 4, 1995 35 Figs. 1-4. Riekia nocatunga. 1, dorsofrontal view of vertex and dorsal view of anterior third of thorax of female (X17.6); 2, dorsal view of posterior third of thorax of female (X17.6); 3-4, ventral and ventrolateral views of genitalia of male (X57). Figs. 5-8. Riekia confluens. 5, dorsofrontal view of vertex and dorsal view of anterior third of thorax of female (X17.6); 6, dorsal view of posterior third of thorax of female (X17.6); 7-8, ventral and ventrolateral views of genitalia of male (X63); [all lemon-yellow colour form] generic key, the species with its uncon- stricted gastral tergum 2 runs down to Riekia Richards, 1962. Further, with the exception of the number of spine-Hke se- tae at the apex of the mid- and hindtibiae (a specific character) the characters given by him in his diagnosis of the species are all common to both coiiflnciii^ and nocn- tungn and may therefore be considered to be generic characters pertaining to Riekia. Similarly the confluence of the postocular and preoccipital carinae (leading to the name confluens), or alter- natively the absence of the postocular ca- 36 Journal of Hymenoptera Research rina, is a character shared with nocatiin- ga. Conversely, the same characters set confluens (and nocatiinga) generically apart from Paragia. Material collected at a single site near Carnarvon, Western Australia, at distanc- es of 165 km and 260 km in a westerly and west-south-westerly direction respectively from the type localities of confluens (16 km WSW Lyons River Homestead, 24.38S, 115.20E — paratype female, and 36 km ESE Minnie Creek Homestead, 24.00S, 115.42E — holotype female) consists of 53 females and 4 males. The material is di- visible into two very distinct groups, one black and reddish-brown and the other black and lemon-yellow. Whereas intra- group variability is negligible, inter-group differences with respect to colour gener- ally and to colour pattern on the metaso- ma in particular are striking. The distinctness of the two sympatric groups and the notable absence of any in- termediate forms initially led to the belief that two species were represented. How- ever, no morphological characters could be found that supported this view and it is concluded that the apparent difference in overall facies is a product of the differ- ences in colour and colour pattern. The reddish-brown colour form, consistent with the description of Snelling's two fe- males (particularly with that of the para- type), is represented in the present mate- rial by females only, the lemon-yellow col- our form by both sexes. To facilitate intra- specific comparison, descriptions limited to colour pattern are given of both female forms whereas for the purpose of inter- specific comparison with nocatiinga the de- scription of the male is given in greater detail. In passing it is noted that in Snelling's description it is incorrectly stated that the probasitarsus is slightly more than twice wider than long. This is clearly an inad- vertent error as it is in fact slightly less than half as wide as long. Reddish-brown (RB) Colour Form Female. — Black. The following reddish- brown: mandibles other than for teeth and extreme base, upper half of clypeal disc, a variously developed transverse band (seemingly formed of a pair of medially fused spots) between and above antennal sockets and narrowly separated ventrally (where sometimes angularly emarginate) from clypeal marking, variously sized elongate spot in upper half of each ocular sinus (rising to level of anterior ocellus and exceeding upper margin of frontal marking) [median frontal transverse band and lateral spots sometimes broadly fused above], a large oval spot behind eyes dor- sally, entire upper surface of prothorax other than for narrow streak bordering posterior margin, tegulae except for clear testaceous central spot and narrow mar- gin, scutellar disc other than for its ante- rior margin, a large spot on raised part of axillae and a minute spot between it and tegula, a diffuse spot on metanotum me- dially, a large spot on angles of propo- deum, large spot on mesopleuron below tegula, distal end of mid- and hindcoxae, part of foretrochanter and whole of mid- and hindtrochanters, femur, tibia and tar- someres of all legs, gastral tergite 1 other than for anterior declivity, tergite 2 other than for a variously developed anterior black band which may be triangularly produced in the middle and for a pair of widely separated diffuse dark markings on posterior margin, an anteriorly point- ing triangular spot on posterior margin of tergite 3, tergite 4 other than for a narrow anterior black band, tergite 5, diffuse transverse posterior band on gastral ster- nites 2-5 (those of sternites 2 and 3 and sometimes 4 strongly and widely anteri- orly produced in the middle). Lemon-yellow (LY) Colour Form Female. — (Figs 5 and 6). Black. The fol- lowing reddish-brown: tip of mandibles except actual teeth, flagellomeres 1-10 be- Volume 4, 1995 37 neath, trochanters of mid- and hindlegs, eral and anterior margins, single large proximal half to three quarters of outer as- frontal spot between and above antennal pect of femora (where moderately to sockets, lower aspects of mid- and hind- heavily black suffused) and inner and coxae, entire outer aspect of midfemora, lower aspect of same, inner and lower as- entire outer aspect of all tibiae (including pect and distal end of tibiae, all tarsomer- distal ends), narrow longitudinal streak on es, posterolateral portions of dark mark- all basitarsi, transverse posterior band on ings on sternites 2-4 and sternite 6 apical- gastral sternites 2-5 (all with median lobe ly (all black-suffused). The following lem- strongly and widely anteriorly produced, on-yellow: proximal half to two-thirds of together giving the effect of a wide me- mandibles, subapical spots on scapes, dian longitudinal band), clypeus other than for narrow lateral and Antennal scape (with radicle) very anterior margins and in some specimens slightly longer than interantennal dis- arcuate black lines on distal half, a pair of tance; first flagellomere slightly more than irregularly shaped frontal spots between half (0.53) as long as scape (with radicle), and above antennae, a narrow streak on one and three quarter times as long as lower inner orbits flanking clypeus, a wide at its distal end, and twice as long as broad mark in ocular sinus rising to level the second flagellomere. Last three flagel- of anterior ocellus, an oval spot behind lomeres flattened and slightly concave be- eyes dorsally, uninterrupted but postero- neath; ultimate flagellomere narrowing medially narrowly emarginate pronotal apically and distinctly curved to form a band connecting humeri, posterior corner hook. of pronotum, tegulae except for clear tes- Tergite 7 truncate and narrowly trans- taceous central spot and narrow margin, a verse apically. large transverse suboval shield-like spot Genitalia (Figs 7 and 8); similar to those on scutellum (not quite reaching anterior of R. nocatunga but differing in detail par- margin), a small spot on raised part of ax- ticularly with respect to the form of the illae, a large spot on angles of propodeum, inwardly directed lobe of the volsella be- large spot on mesopleuron below tegula, low the digitus. a small spot on mesosternum anterior to Length 7.8-8.3 mm; length of forewing coxal cavities, a small lateral spot on mid- 5.5-5.7 mm; hamuli 12-13. and hindcoxae, distal half of outer aspect Material Examined. — Western Australia: of forefemora, distal spot on outer aspect 8 km NE of Carnarvon (24.51S, 113.45E) of mid- and hindfemora, outer aspect of on road to Bibbawarra Bore, 3-5. x. 1993 all tibiae (except distal ends), wide ante- (F.W.,S.K.& R.W.Gess) 16 RB females, 29 riorly trilobed posterior band on gastral LY females and 4 LY males (in yellow tergites 1-5, tergite 6 (other than for vari- flowers of Lechenaiiltia sp., Goodeniaceae), ously developed posteriorly pointing V- 4 RB females, 1 LY female (at water), 2 RB shaped black mark leaving lateral and me- females, 1 LY female (without biological dian spots or almost eliminating these), data). Two RB females, 2 LY females and transverse posterior band on gastral ster- 1 LY male in both ANIC and WAM; rest nites 2-5 (those of sternites 2 and 3 tri- of material in AMG. lobed with median lobe strongly and Discussion. — R. confliieiis may be distin- widely anteriorly produced). guished from R. nocatunga in both sexes Male. — (Figs 7 and 8). Colour pattern by the colour pattern and the below indi- very similar to that of female but differing cated morphological characters, in that the following parts are pale lemon- In R. confluens the ratio of POL (distance yellow: entire anterior aspect of scapes, between posterior ocelli):OOL (distance entire clypeus other than for narrow lat- between a posterior ocellus and nearest 38 Journal of Hymenoptera Research eye margin) is 1:1.3-1.4, whereas in R. no- the species differ in the male sex in the catunga it is 1:1.0-1.1. relative proportions of the first two flagel- In R. confliiens the pilosity of the pro- lomeres, in the form of gastral tergite 7 notum is only moderately dense, the in- and in slight differences in the genitalia, dividual hairs being slender and only Present data would indicate that the slightly curved apically whereas in R. no- distributions of the species are widely sep- catunga the pilosity is much denser and arated, R. nocatungci occurring in New the individual hairs are more robust and South Wales and Queensland and R. con- markedly and evenly curved. fluens in Western Australia. In R. confluens the scutellum (Fig. 6) has its disc posteriorly widely and evenly to Rolandia angulata (Richards) bluntly rounded and abruptly declivitous ^.^^^.^ ^,^^^^^^^^ j^.^j^^^^^^ ^^^^. ^q^_^q2, female, to the metanotum which it slightly over- ^^,^„^^,.^ ^^^^^^^^^ (Richards), Snelling, 1986: 8. hangs; the posterolateral free edge of each lateral wing is almost straight. The meta- This species was described from Cun- notum in its median third is not depressed namulla [28.04S, 145. 40E], Queensland middorsally and is therefore of greatest (holotype female and 1 female paratype) depth in the middle; it is vertical, not vis- and from 90 km W of Cobar, Barnato ible from above. In R. nocatunga the scu- Tanks [31.38S, 144.59E] (about 400 km tellum (Fig. 2) has its disc posteriorly nar- south of Cunnamulla), New South Wales rowly rounded and gradually descendant (12 female paratypes). Snelling (1986) re- to metanotum which it does not overhang; corded an additional 8 females from Cun- the posterolateral free edge of each lateral namulla. Hitherto the male appears to wing is widely and shallowly concave, have been unknown. One of the present The metanotum in its median third is females was compared with the holotype somewhat depressed middorsally and is in the Australian National Collection of therefore of even depth; it is sloping, vis- Insects, Canberra, and found to be identi- ible from above. cal. In R. confluens gastral tergite 1 is two Male. — (Figs 9 and 10). Colour pattern and two-fifths to two and a half times very similar to that of female, differing wider than long whereas in R. nocatunga chiefly in that the yellow markings are it is two and a quarter times wider than lighter in hue and brighter and contrast long. more strongly with the light reddish In R. confluens the number of spine-like brown areas. The bright yellow mandi- setae on the apex of the mesotibia is 1 bles, clypeus and transverse lower frontal whereas in R. nocatunga it is 2-4, most marking, and the distinct median and lat- commonly 3; the number of such setae on eral spots on the gaster facilitate recogni- the apex of the metatibia is 1-2, most com- tion of the male even in the field, monly 1, and 2-3 respectively. Antennal scape (with radicle) shorter The species differ in the female sex in than interantennal distance; first flagello- the form of the pronotum as seen from mere three times as long as broad at its above. In R. confluens (Fig. 5) the anterior distal end, about two-thirds as long as and lateral margins describe a semicircle scape (with radicle) [three-quarters as long whereas in R. nocatunga (Fig. 1) the as scape without radicle], and a little more "shoulders" are anterolaterally produced than twice as long as second flagellomere; so that the anterior and lateral margins do flagellomeres 3-9 subequal in length and not describe a semicircle but the anterior with the exception of 3 wider than long; margin appears subtransverse. 3-7 increasing in width; 7 and 8 of same As may be seen from the descriptions, width; 9 as wide as 6; 10 slightly longer Volume 4, 1995 39 Figs. 9-10. Rolandia angulata. 9, posterior view of right fore leg of male (X44.5); 10, ventral view of genitalia of male (X44.5). than preceding flagellomeres, narrowing and apically rounded; distal flagellomeres not modifled beneath. Foreleg (Fig. 9); process of foretrochan- ter large, compressed, broadly digitiform in posterior view and slightly forwardly curved, its apex attaining level of angulate base of forefemur; forefemur in posterior view broad and ventrally strongly angu- late at base and widely but shallowly in- curved from there to about midlength, with ventral margin angulate along its length but becoming lamellate at basal an- gle, with ventral surface basally (i.e. ad- jacent to lamellate ventral angle) excavat- ed. Genitalia (Fig. 10); parameral spines apically sharply pointed and ventrally re- curved; ventral process of each paramere tapering towards apex in lateral view but in ventral view more or less of even width over most of its length and apically nar- rowly rounded; digitus and inwardly di- rected lobe of volsella as figured. Length 8.7-9.2 mm; length of forewing 6.3-6.5 mm, hamuli 16-18. Material Examined. — Queensland: 85 km E of St George (28.03S, 148.30E), 27.X.1993 (F.W.,S.K.& R.W.Gess) 6 females (in yel- low flowers of Goodenia cycloptera R.Br.in C.Sturt, Goodeniaceae), 7 females (in yel- low flowers of Goodenia pinnatifida Schldl., Goodeniaceae); 27 km W of St George (28.03S, 148.30E), 28.X.1993 (F.W.,S.K.& R.W.Gess) 31 females, 10 males (in yellow flowers of Goodenia cycloptera R.Br.in C.Sturt, Goodeniaceae), 1 female (in yel- low flowers of Goodenia pinnatifida Schldl., Goodeniaceae), 2 females (nesting in sandy soil), 2 females, 3 males (without bi- ological data); 80 km E of Cunnamulla (28.04S, 145.40E), 29.X.1993 (F.W.,S.K.& R.W.Gess) 49 females, 31 males (in yellow flowers of Goodenia pinnatifida Schldl., Goodeniaceae), 4 females (nesting in sandy soil), 9 females, 2 males (without bi- ological data). Three females and 3 males in ANIC; 3 females and 3 males in WAM; rest of material in AMG. Discussion. — Snelling (1986) was correct in his assumption that the males of R. an- gulata would differ from those of the very similar R. houstoni in the same characters as do the females. Of particular note is the scutellum which in both sexes of R. an- gulata slopes evenly to the metanotum and 40 Journal of Hymenoptera Research lacks a definite posterior face. The males run down satisfactorily in Snelling's key, allowance being made for the longer than broad second flagellomere. ACKNOWLEDGEMENTS The author wishes to thank Ian Naumann and Jo Cardale of the Australian National Insect Collection, Canberra for hospitality in their department and for facilitating the comparison of some of the present specimens with type material in their care, Kirsten Cowley of the Australian National Herbarium, Can- berra for identifying the forage plants, Suzanne Lewis of the Natural History Museum, London for lending material under her care, and Robin Cross of the Elec- tron Microscopy Unit, Rhodes University, Grahams- town for producing the scanning electron micro- graphs reproduced as Figs 1-10. An earlier version of the manuscript was reviewed by Karl V. Krombein of the Smithsonian Institution, Washington and James M. Carpenter of the American Museum of Nat- ural History, New York. The latter, in particular, is thanked for his constructive critique and for drawing attention to a potential synonymy arising from the fact that R. confluens had been described by Roy Snell- ing as a species of Paragia, a genus to which the wasp in question quite clearly does not belong. Gratitude to the South African Foundation for Research Devel- opment is expressed for a rolling support grant which made the field work in Australia possible. LITERATURE CITED Meade-Waldo, G. 1911. Description of new species of African Diploptera in the collection of the Brit- ish Museum. Annals and Magazine of Natural His- tory (8)8: 448-457. Richards, O.W. 1962. A revisional study of the Masarid ivasps (Hymenoptera, Vespoidea). London: British Museum (natural History). Richards, O.W. 1968. New records and new species of Australian Masaridae (Hymenoptera: Vespo- idea). Journal of the Australian Entomological Soci- ety 7: 101-104." Shuckard, W.E. 1837. Descriptions of new exotic acu- leate Hymenoptera. Transactions of the Entomolog- ical Society of London 2: 68-82. Snelling, R.R. 1986. The taxonomy and nomenclature of some Australian paragiine wasps (Hymenop- tera: Masaridae). Contributions in Science 378: 1- 19. van der Vecht, J. and Carpenter, J.M. 1990. A cata- logue of the genera of the Vespidae (Hymenop- tera). Zoologische Verhandelingen 26: 1-62. J. HYM. RES. Vol. 4, 1995, pp. 41-63 The Last Twenty Years of Parasitic Hymenoptera Karyology: An Update and Phylogenetic Implications Vladimir E. Gokhman and Donald L. J. Quicke (VEG) Botanical Garden, Moscow State University, Moscow 119899, Russia; (DLJQ) Department of Biology, Imperial College at Silwood Park, Ascot, Berks SL5 7PY, U.K. Abstract. — A survey of karyological data in the parasitic Hymenoptera shows a considerable uniformity in chromosome numbers within higher taxonomic groupings, with most chalcidoids having n = 5-6 and most cynipoids, ichneumonoids and those very few species of Diapriidae and Scelionidae examined to date having n = 10-12. The Encyrtidae and Eurytomidae differ from other chalcidoids examined to date in having higher n values (8-11). The braconid subfamilies Aphidiinae {ii - 4-7), Charmontiinae (n = 5) and Exothecinae (» = 6) are shown to have lower values than do other braconids. New data are presented for 32 species, including the first records of chromosome numbers for the Eupelmidae, Bethylidae, Diapriidae, Ormyridae, the ichneumonid subfamilies Banchinae, Pimplinae, and Tryphoninae, and the braconid subfamilies Aphidiinae, Charmontinae, Doryctinae, Exothecinae, Macrocentrinae, Meteorinae, and Miracinae. Chromo- some number data are discussed in relation to current views on Hymenoptera phylogeny. INTRODUCTION There has long been a tendency for tax- onomists to ignore many potential sources of systematic evidence in favour of more traditional morphological, and nowadays molecular, data. Whilst this lack has not been ubiquitous amongst taxonomists, where it has occurred it may have been due to a number of reasons including lack of understanding of novel characters, lack of facilities for investigating them and lack of availability of information about such characters. In this respect, it is not sur- prising that chromosome number and oth- er karyological features have played little role in considerations of hymenopteran phylogeny, and this is especially true for the parasitic families. Reasons for this are manifold. Hymenopteran chromosomes have traditionally been studied in imma- ture stages in which dividing cells are common, but larvae of parasitic wasps are perhaps especially difficult to identify and obtain except when species are in culture. Recent developments in the study of hy- menopteran chromosomes (Imai et al. 1988, Baldanza et al. 1991b), have opened the possibility of greatly expanding our knowledge of parasitic wasp karyology. Further, it is now possible to obtain good results working with adult tissues such as ovaries (Gokhman 1985, 1990a), and such protocols have allowed, for the first time, an examination of chromosome number and structure in natural populations. Karyological studies have, nevertheless, played an important role in the systemat- ics and species level taxonomy of various groups of the non-parasitic Hymenoptera including, sawflies, ants, bees and social wasps (principally Polistinae) and these have been reviewed in detail elsewhere (see for example, Kerr 1972, Kerr and da Silveira 1972, Naito 1982, Imai et al. 1977, Moritz 1986, Pompolo and Takahashi 1987, 1990, Hoshiba, Matsuura and Imai 1989, Costa et al. 1993). Wide variation in n is well known among the ants (Formic- idae) and ranges from 1 to 47 (Imai and Taylor 1989, Imai et al. 1990), but is modal at 11 (Imai et al. 1988). Indeed, the greater 42 Journal of Hymenoptera Research part of this range can even be found with- in the single ant genus Myrmecia. How- ever, within the parasitic Hymenoptera there appears to be be rather more consis- tency, and for most famiUes the observed range of variation is far narrower. Outside of the Hymenoptera, both chro- mosome numbers, structure and size have all been found useful in phylogenetic re- construction, usually but not exclusively in combination with other morphochar- acters, and this has been particularly true of plant systematics. For example, George & Geethamma (1992) have recently pro- posed a phylogeny of jasmines based on chromosome numbers and assumed poly- ploidy events. Among the insects, chro- mosome numbers have been used exten- sively in the systematic treatment and phylogeny of the Hemiptera (Blackman 1980, Emeljanov & Kirillova 1989, 1991). Kuznetsova (1985), for example, has con- cluded that the homopteran subfamily Or- geriinae is monophyletic based on an au- tosomal fusion giving 2m =27 (in males) compared with the plesiomorphic value of 2n = 29 (in males), and similarly that the tribe Almanini of the Orgeriinae is mono- phyletic on the basis of an autosome-sex chromosome fusion (giving 2n=26). In this paper we review the current state of knowledge of the karyology of parasitic wasps and, in addition, provide new data for approximately 32 taxa including the first records of chromosome numbers for the Eupelmidae, Bethylidae, Diapriidae, Ormyridae, the ichneumonid subfamilies Banchinae, Pimplinae, and Tryphoninae, and the braconid subfamilies Aphidiinae, Charmontinae, Doryctinae, Exothecinae, Macrocentrinae, Meteorinae, and Miraci- nae. MATERIALS AND METHODS New karyotypic data were obtained us- ing ovarian tissues of adult wasps which were either collected from the wild at Sil- wood Park, Berkshire, U.K., during early September 1995, or obtained from labora- tory cultures. Chromosome preparations were obtained according to the schedule described in Appendix I. Voucher speci- mens for Ichneumoninae are deposited in the collection of Moscow State University, those for other taxa are in the Natural His- tory Museum, London. RESULTS AND DISCUSSION The last review of chromosome number in the parasitic Hymenoptera was by Cro- zier (1975) at which time values were known for only about twenty species (with data published before 1930 not in- cluded for the reason of insufficient reli- ability); his data are summarized in Figure 1. Since then, chromosomes of nearly 140 additional species have been studied (Gokhman 1994), and for this review we have investigated 32 more in order to help fill in a number of gaps and to confirm some previous findings. These data are presented in Table 1 and summarized in Figure 2. The great majority of published chromosome numbers come from studies on the ichneumonid subfamily Ichneu- moninae by Gokhman (1985, 1987, 1989, 1990a, 1990b, 1991a, 1991b, 1993a), though they also include a substantial body of ev- idence especially relating to various fam- ilies of Chalcidoidea. Within the better studied parasitoid families (i.e. Ichneu- monidae, Cynipidae, Torymidae) chro- mosome numbers were found to be rela- tively stable, and were generally uniform at the generic level. This contrasts mark- edly with observations for many aculeate Hymenoptera which often show striking chromosomal variability even within gen- era (Imai and Taylor 1989, Imai et al. 1990). The apparent bimodality in haploid number of parasitic Hymenoptera appar- ent in Crozier's small sample (Fig. 1) now appears to be well founded (Fig. 2). How- ever, the exact modal numbers are slightly different (n = 5 and 11 compared with n = 5 and 10 of Crozier), probably due to the strong bias in favour of the ichneumo- Volume 4, 1995 14 -T- 12 - 10 43 6 - 2 -- + + + + 4- 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Haploid chromosome number Fig. 1. Histogram of haploid chromosome numbers for parasitic Hymenoptera, data modified after Crozier (1975) to take into account subsequent taxonomic changes. nid subfamily Ichneumoninae in the pres- ent enlarged data set. As will be apparent from the figures, the peak in haploid val- ues around 5 largely comprises members of the Chalcidoidea and there is only slight overlap between these and the other peak centred around 11. The same is true, though less pronounced, when the data are plotted for genera rather than species (Figure 3: each n value in a genus being represented only once in the histogram). Members of three of the chalcidoid fami- lies investigated to date, the Eurytomidae (Eiiri/toma), Encyrtidae (6 species in three genera) and Aphelinidae (one species, Pteroptrix (= Archenomiis) orientalis Silves- tri, out of 5 investigated; see below) differ from other chalcidoids in having haploid numbers ranging between 8 and 11, whilst members of the braconid subfamilies Aphidiinae {Aphidius, Diaretiella, Ephcdriis, Praon), Charmontinae [Charmon) and Ex- othecinae [Rhysipolis) are atypical in hav- ing haploid numbers of 4 to 7, 6 and 5 respectively, compared with other ichneu- monoids whose haploid n values range from 8 to 17. GENOMIC SIZE AND ITS IMPLICATIONS Rasch et nl. (1975, 1977) calculated the haploid genomic size of the braconid wasps, Habrobracon juglandis and H. seri- nopae both to be 0.154).16 x 10 ^^g dNA and that of the pteromalid, Nnsonin (as Mormoniclla) vitripennis to be 0.33-0.34 X 10 '-^g. These values translate (using an average molecular weight of a base pair as 660) to base pair numbers of approxi- mately 1.4 X 10** and 3.0 X 10^ base pairs for the ichneumonoid and the chalcidoid respectively. More recently. Bigot et al. (1991) using DNA reassociation kinetics calculated the haploid genomic sizes of the ichneumonid Diadroniiis piilchclliis and the chalcidoid, Eiipclmus viiillcti, as 1-2 X 10'^ base pairs and 10 X 10"^ base pairs re- spectively. Unfortunately, no data for the numbers of chromosomes are available for Eupelmus vuiUeti but we have investigated 44 Journal of Hymenoptera Research Table 1. Chromosome number in parasitic wasps Taxon 2/!* Reterence(h)t Diaprioidea Diapriidae Beli/ta depressa Thomson Cinetus lanceolatus Thomson Scelionoidea Scelionidae Telejtomus fariai Lima Chalcidoidea Aphelinidae Apheliniis mali Haldeman Aphytis mytilaspidus (Le Baron) Archenonnis orientalis Silvestri Encarsia berlesei Howard Encarsia pergandiella Howard Chalcididae Bradiyineria intermedia Nees B. lasus Burks B. ovata Say Dirhiniis himalayanus Westwood Encyrtidae Ageniaspis fiiscicoUis Dalman Copidosoma buyssoni Mayr C. gelechiae Howard C. gelechiae Howard C. truncatelliim (Dalman) (= Ifloridanum Ashmead) C. floridanum Ashmead C. floridanum Ashmead Apoanagyrus lopezi (DeSantis) Eulophidae Cirrospilus diallus Walker Colpoclypeiis florus Walker Melittobia chalybii Ashmead Tetrastichus gigas Burks T. megachilidis Burks Eupelmidae Anastatus catalonicus Bolivar Macroneura vesicularis (Retzius) Eurytomidae Eudecatoma bigidtata (Swederus) Eurytoma californica Ashmead Leucospidae Leucospis affinis Say Ormyridae Ormynis sp. Pteromalidae Anisoptcromalus calandrae (Howard) Coelopisthia extenta (Walker) Dibrachys sp. Lariophagus distinguendus Foerster Muscidifurax zaraptor Legner Nasonia vitripennis (Walker) 8 10 10 6 6 5 6 6 5 5 9 10 ?6 7 5 5 5 5 5 16 present papertt 20 present paper 20 Dreyfus & Breuer 1944 5 10 Viggiani 1967 5 10 Rossler & De Bach 1973 11 22 Baldanza et al. 1991a 5 10 Baldanza et al. 1991b 6 12 Hunter et al. 1993 3 6 Hung 1986 5 10 Hung 1986 5 10 Hung 1986 5 10 Amalin et al. 1988 c. 10 C.20 Silvestri 1908; Martin 1914 12 24 Silvestri 1914 11 22 Hegner 1915 10 20 Patterson 1921 10 20 Hunter & Bartlett 1975 8 16 Leiby 1922; Patterson 1917, 1921 Patterson & Porter 1917 11 22 Strand & Ode 1990 10 20 Dijken 1991 12 present paper 12 Dijkstra 1986 10 Schmieder 1938 12 Goodpasture 1974 12 Goodpasture 1974 10 present paper 10 present paper 18 present paper 20 Goodpashire 1974 12 Goodpasture 1974 ?12 present paper 14 present paper 10 present paper 10 Goodpashjre 1974 10 Gershenzon 1968 10 Goodpashare 1974 10 Gershenzon 1946, 1968; Pennypacker 1958; Whiting 1960, 1968; Wahr- man & Zhu 1993 Volume 4, 1995 45 Table 1. Continued Taxon Reference(s)t Nasonia vitripennis (Walker) 5 + 0-1 IB 10 Nasonia vitripiennis (Walker) 6 12 Ptewmalus puparum L. 5 10 P. venustus Walker 5 10 Torymidae Monodontoinerus dementi Grissell 6 12 M. moniivagus Ashmead 6 J2 M. obscurus (Westwood) 4 8 M. obscurus (Westwood) 6 12 M. saltuosus Grissel 5 10 Tory mils baccharidis Huber 6 12 T. californicus (Ashmead) 6 12 T. capillaceus Huber 6 12 T. koebelci Huber 5 10 T. occidentalis Huber 6 12 T. tubicola Osten-Sacken 6 12 T. umbilicatus Gahan 5 10 T. vesiculi Moser 6 12 T. warreni (Cockerell) 6 12 Trichogrammatidae Trichogmmma cliilonis Ischii 5 10 T. dcicvi Pinto & Oatman 5 10 T. dendrolimi Matsumura 5 10 T. evanescens Westwood 5 10 T. nubialale Ertle & Davis 5 10 T. pretiosum Riley 5 10 T. spp. (7 strains) Cynipoidea Cynipidae Andricus curvator Hartig A. feciindator Hartig A. kollari Hartig A. quercuscalicis Burgsdorf Aulacidea hiemcii Bouche Biorrhiza pallida Olivier Callirhi/tis palmiformis Ashmead Cynips divisa Hartig Diastrophus nebulosus Osten-Sacken Diplolepis elganteriae Hartig D. nervosum Curtis D. rosae L. D. rosae L. D. spinosissimae Girault Dryocosmus kuripliilus Yasumatsu Dryophanta erinacea Mayr Neurotcrus laeviusculus Schenck N. numismalis Fourcroy N. quercusbaccarum L. Trigo)iaspis megaptera Panzer Xestopluvics potcntillac Retzius 10 Nur et al. 1988; Werren 1991 Goodpasture 1974 Guhl & Dozortseva 1934, Dozortseva 1936 McDonald & Krunic 1971 Grissell 1973b; Goodpasture 1975a Goodpasture 1975a Goodpasture 1975a McDonald & Krunic 1971 Grissell 1973b; Goodpasture 1975a Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 Hung 1982 Stouthamer & Kazmer 1994 Liu & Xiong 1988 Hung 1982 Hung 1982 Hung 1982; Stouthamer & Kazmer 1994 Fukada & Takemura 1943 (cited by Hung 1982) 10 20 Dodds 1938; Sanderson 1988 10 20 Dodds 1938 10 20 Hogben 1920; Sanderson 1988 10 ?20 Sanderson 1988 10 20 Dodds 1938 10 20 Dodds 1938 w 20 Goodpasture 1979b 10 20 Sanderson 1988 10 20 Hegner 1915' 9 27(3n) Sanderson 1988 9 18 Sanderson 1988 9 18 Henking 1892; Stille & Davring 1980; Sanderson 1988 ?12 ?24 Schleip 1909 9 18 Sanderson 1988 10 20 Abe 1994 c. 12 24 Wieman 1915 10 20 Sanderson 1988 10 20 Dodds 1938 10 20 Doncaster 1910, 1911, 1916; Dodds 1938 10 20 Dodds 1938 10 20 Dodds 1938 46 Journal of Hymenoptera Research Table 1. Continued Taxon ;;' 2n* Reference(s)t Eucoilidae Pseudeucoila bochei Weld 10 20 Jungen cited in Crozier 1975 Ichneumonoidea Braconidae Aphidius rhopalosiphi De Stefani Perez 7 14 present paper Biosteres carbonarius (Nees) U 28 present paper Charmon cruentatiis Haliday 5 10 present paper Cotesia gloineratiis (L.) (as Apanteles) 12 24 Hegner 1915i Dacnusa sp. 17 34 present paper Diaeretiella rapae Mcintosh 6 12 present paper Ephedrus sp. 17 ?14 present paper Habwbracon hebetor Say 10 20 Torvik-Greb 1935 H. jiiglandis Ashmead^ 10 c. 20 Torvik-Greb 1935; Speicher 1937; Rasch et al. 1977 H. jiiglandis Ashmead^ 11 22 Whiting 1927 H. pectinophorae Watanabe 10 20 Inaba cited in Makino 1959 H. serinopae Ramakrishna 10 20 Rasch et al. 1977 Heterospihis prosopidis Viereck 17 34 present paper Macrocentrus thomcicus (Nees) 13 26 present paper Meteorus versicolor Wesmael 8 16 present paper Meteorus gyrator Thunberg 10 20 present paper Meteorus pallipes Wesmael 10 20 present paper Mirax sp. 10 20 present paper Phaenocarpa persimilis Papp 17 34 Prince & Stace (cited by Crozier 1977) Praon abjectum Haliday 4 8 present paper Rln/sipolis decorator (Haliday) 6 12 present paper Ichneumonidae Aethercerus discolor Wesmael 11 22 Gokhman 1985 Ae. dispar Wesmael 12 24 Gokhman 1991 Ae. nitidus Wesmael 11 22 Gokhman 1990a Ae. raniui Gokhman 11 22 Gokhman 1991 Agrothereutes extrematus (Cresson) 10 20 Koonz 1939 Aophis pulchricornis (Gravenhorst) 13 26 Gokhman 1990b Aptesis puucticollis (Thomson) 8 16 Gokhman 1990a Baeosemus dent if er Gokhman 11 22 Gokhman unpubl. obs. Baranisobas ridibundus (Gravenhorst) 11 22 Gokhman 1990b Chasmias niotatorius (Fabricius) 17 34 Gokhman 1985 Coelichneiimon cyaniventris (Gravenhorst) 13 26 Gokhman 1990a C. sugillatorius (L.) 13 26 Gokhman 1990a Colpognathus celerator (Gravenhorst) 11 22 Gokhman unpubl. obs. Cratichneumon viator (Scopoli) 14 28 Gokhman 1985 Diadroinus prosopius Holmgren 11 22 Gokhman 1990a D. pulchellus Wesmael 11 22 Hedderwick et al. 1985 D. subtilicornis (Gravenhorst) 11 22 Gokhman 1990b D. troglodytes (Gravenhorst) 11 22 Gokhman 1990a D. varicolor Wesmael 11 22 Gokhman pers. obs. Dicaelotus sp. nr. parvulus (Gravenhorst) 11 22 Gokhman 1990a D. piimilis (Gravenhorst) 11 22 present paper Diphyus latebricola (Wesmael) 12 24 Gokhman 1990a D. raptorius (L.) 12 24 Gokhman 1990a Dirophanes callopus (Wesmael) 9 18 Gokhman 1987 D. fidvitarsis (Wesmael) 10 20 Gokhman 1990a,b D. invisor (Thunberg) 10 20 Gokhman 1987 Dyspetes arrogator Heinrich 10 20 present paper Volume 4, 1995 47 Table 1. Continued Taxon Ret'erence(s)+ Ephialtes manifestator (L.) 15 30 Eurylahiis tonms Wesmael 10 20 Gelis sp. 13 26 Gli/pta lapponica Holmgren 9 18 Heterischnus nigricollis (Wesmael) 11 22 H. tnincator (Fabricius) 11 22 Homothenis locutor (Thunberg) 11 22 Ichneimion albiger Wesmael 12 24 /. auip}iibolus Kriechbaumer 12 24 /. bucciilcntus Wesmael 12 24 /. confiisor Gravenhorst 12 24 /. crassifemiir Thomson 12 24 /. croceipes Wesmael 12 24 /. extensorius L. 12,13 24, 26^ /. cf. extensorius L. 11 22 /. formosus Gravenhorst 11 22 /. gracilentus Wesmael 12 24 (25) /. gracilicomis Wesmael 11 22 /. ingratiis Hellen 12 24 /. inquinatus Wesmael 13 26 /. iiisidiosus Wesmael 12 24 /. liigens Gravenhorst 12 24 /. melanotis Holmgren 12 24 /. mmutorius Desvignes 12 24 /. molitorius L. 12 24 /. nereni Thomson 11 22 /. sarcitorius L. 11 22 /. stramentarius Gravenhorst 12 24 I. submarginatus Gravenhorst 10 20 /. subquadmtiis Thomson 12 24 /. suspiciosus Wesmael 12 24 (26) /. validicornis Holmgren 22 24 Lissonota sp. 11 22 Mastnis smithii (Packard) 13 26 Oroiwtus binotatus (Gravenhorst) 11 22 Orthocentrus sp. 14 28 Orthopelma mediator (Thunberg) 11 22 Patrodoides chah/beatiis (Gravenhorst) 8 16 Phaeogencs melaiiogonos (Gmelin) 11 22 Ph. nigridens Wesmael 11 22 Ph. semivulpimis (Gravenhorst) 9 18 Ph. spiuiger (Gravenhorst) 11 22 Polysphmcta tiiberosa Gravenhorst 9 18 Pseudoambhjteles homocerus (Wesmael) 9 18 Stenichneumon cidpator (Schrank) 14 28 Sycaonia sicaria (Gravenhorst) 11 22 Syspasis alboguttata (Gravenhorst) 11 22 S. scutellator (Gravenhorst) 11 22 Tycherus australogemimis Gokhman 11 22 T. bellicontis (Wesmael) 10, U 20(21) T. dilleri Ranin n 22 T. fuscicornis (Wesmael) n 22 T. ischiomelinus (Gravenhorst) 9 18 present paper Gokhman 1987 present paper present paper Gokhman 1990b Gokhman unpubl. obs. Gokhman 1990b Gokhman 1990a Gokhman 1990b Gokhman 1993a Gokhman 1985 Gokhman 1985 Gokhman 1990b Gokhman 1993a Gokhman 1990a Gokhman 1990a Gokhman 1993a Gokhman 1990a Gokhman 1990b Gokhman 1993a Gokhman 1990a Gokhman unpubl. obs. Gokhman 1990a Gokhman 1987 Gokhman 1990a Gokhman 1990a Gokhman 1990a Gokhman 1987 Gokhman 1990a Gokhman 1990b Gokhman 1993a Gokhman unpubl. obs. present paper Koonz 1936 Gokhman 1987 Gokhman 1990a Hogben 1920 Gokhman 1993a Gokhman 1990b Gokhman 1990b Gokhman unpubl. obs. Gokhman 1990a present paper Gokhman unpubl. obs. Gokham 1985 Gokham 19990b Gokhman 1985 Gokhman unpubl. obs. Gokhman 1991 Gokhman 1989 Gokhman 1989 Gokhman 1990b Gokhman 1991 48 Journal of Hymenoptera Research Table 1. Continued Taxon n* 2n* Reference(s)+ T. ophthalmicus (Wesmael) 22 Gokhman 1990a T. osculator (Thunberg) 22 Gokhman 1989 T. suspicax (Wesmael) 22 Gokhman 1987 Ventiiria canescens (Gravenhorst) 22 Speicher 1937 Virgichneumon digrammus (Gravenhorst) 34 Gokhman 1990a V. f annus (Gravenhorst) 22 Gokhman 1990a Vulgichneumon snturatorius (L.) 9 18 Gokhman 1987 Chrysidoidea Bethylidae Epyris niger Westwood 14 28 present paper Laelius utilis Cockerell 10 20 present paper * In papers which only quote n or 2)i, the other value has been surmised and is given in italics, t Data appearing in works before 1930 should be considered with great caution as most resulted from his- tological rather than cytological protocols, involving sectioned material rather than squash preparations and also often involving fixation techniques not well suited for the study of chromosomes, though some of these earlier findings are clearly correct including the oldest one (Henking 1892). tt For some new data we were not able to obtain an unambiguous chromosome number but our best ap- proximation is presented (data indicated in table with a "?") as in some cases these still provide potentially valuable information. ' Hegner (1915) did not provide a definitive statement on chromosome number and the data here come from his rather stylized figures; such data need therefore to be considered with extreme caution. ^ Some workers consider H. juglandis to be a junior synonym of H. Iwbetor, however, this is not yet absolutely confirmed and therefore we prefer to keep these records separate. ^ A single, probably aneuploid, female specimen with 25 chromosomes has also been found. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Haploid chromosome number Fig. 2. Histogram of haploid chromosome numbers of parasitic Hymenoptera, data from Table 1 based on n values of each species. Volume 4, 1995 30 T 25 - 49 20 - 15 - 10 7 8 9 10 11 12 13 Haploid chromosome number 14 15 16 17 Fig. 3. Histogram of haploid ciiromosome numbers of parasitic Hymenoptera, data from Table 1, each n value occurring in a genus being represented only once. two other eupelmid species in the genera Macroneura and Anastatiis, which were both found to have haploid numbers of 5 in common with Nasonia. Although the re- sults of Rasch et al. (1975, 1977) and of Big- ot et al. (1991) don't quite agree in the or- der of magnitude of base pairs they esti- mate, probably due to differences in pro- cedure, they did both show the chalcidoid to have a larger genome than the ichneu- monid. Thus, if these values are roughly representative of other members of their superfamilies, then the chromosomes of most chalcidoids would be expected to have between four and twenty times as much DNA on average than chromosomes of ichneumonids. CHROMOSOME NUMBER IN RELATION TO HYMENOPTERA PHYLOGENY In karyological studies it is common practice to interpret modal chromosome numbers as representing the initial (ances- tral) number, though to many cladists this would be interpreted as the application of the much decried commonality principle. The 'common equals primitive' associa- tion is of course probabilistic rather than deterministic (Watrous and Wheeler 1981, Frohlich 1987, Quicke 1993). Some further insight into whether modal chromosome number is likely to reflect the ancestral number can come from the comparison of modal and median numbers. If these numbers coincide, and the whole distri- bution may be approximated to a normal one, then the data provide no evidence of a directed change, though it must be re- membered that there is always the possi- bility that an evolutionary change in chro- mosome number early in the evolution of the group could lead to the same distri- bution. In the Ichneumonoidea and the Cynipoidea both the median and modal chromosome numbers are the same 11 and 10 respectively; in contrast, for example, ants (Formicidae) have a modal number of 11 but the median is 15. However, cladis- tic analyses based on independent char- 50 Journal of Hymenoptera Research acter systems, via outgroup comparisons, provide the most reliable means of deter- mining plesiomorphic chromosome num- bers and, where possible, this is the ra- tional that we have employed. According to currently accepted views of Hymenoptera phylogeny, the sawflies ('Symphyta') form a paraphyletic group with respect to the Apocrita with the latter being the sister group of the Orussidae, and the Apocrita + Orussidae in turn most probably being the sister group of the Xiphydriidae (Konigsmann 1977, Rasnit- syn 1980, 1988, Gibson 1985). Unfortunate- ly, chromosome numbers are not known either for the Orussidae or for the Xiphy- driidae, although they are known for members of two other sawfly families with claims for a close relationship with the Apocrita, viz the Siricidae and Cephi- dae (Konigsmann 1977, Basibuyuk & Quicke 1994, 1995). In the Cephidae n ranges from 9 to 22-26 (Mackay 1955, Cro- zier and Taschenberg 1972), whilst in the Siricidae, according to Sanderson (1932, 1970), the haploid number varies between 8 and 18. As regards other, less derived sawflies, haploid chromosome number ranges from 5 to 22 in the Tenthredino- idea, with three quarters of species having an n value ranging between 7 and 10 (Nai- to 1982). Taking the Siricidae and Cephi- dae as the two sawfly families closest to the ancestral lineage of the Apocrita (i.e. putative sister groups of the Apocrita + Orussidae) for which chromosome num- bers are available, it seems reasonable to conclude that the plesiomorphic haploid chromosome number in the latter was at least 8 and possibly rather higher. Within the Apocrita there is a picture emerging from independent investiga- tions of phylogenetic relationships (Ras- nitsyn 1988, Johnson 1988, Gibson 1985, Mason 1983, Quicke et al. 1993, 1994, Her- aty et al. 1994) that the group divided rel- atively early in its history into a lineage giving rise to the Ichneumono- idea+Aculeata and a second comprising the bulk of the taxa currently regarded as 'Microhymenoptera' including Chalcidoi- dea, Cynipoidea, Scelionoidea, Diaprioi- dea and Proctotrupoidea s.s. (Fig. 4). Our data show that the modal n value in the Ichneumonoidea, the probable sister group of the Aculeata, is 11. Further, our limited data for the less derived aculeates of the family Bethylidae {Epyris and Lae- liiis; Fig. 10), whilst demonstrating some degree of variation in haploid number be- tween 10 and 14, when considered togeth- er with data for other aculeates suggest the ancestral aculeate may have had a haploid number around 11, as was also concluded by Hoshiba, Matsuura and Imai (1989). Similarly, available values for three other parasitoid superfamilies, the Diaprioidea (Fig. 5), Scelionoidea and Cy- nipoidea, are similar. According to Ras- nitsyn's (1988) phylogenetic hypothesis (see Fig. 4), the Scelionoidea are putatively the sister group of the Chalcidoidea s.l., and thus taking the former as the out- group, the plesiomorphic haploid chro- mosome number for the Chalcidoidea may be postulated as being approximately 10. Therefore the data collectively support the hypothesis that the small values of n (from 3 to 7) found in the majority of Chal- cidoidea are likely to be apomorphous. Unfortunately, there are no well founded views of relationships within the Chalci- doidea (Trjapitzin 1978, LaSalle 1987, Bou- cek 1988a, Woolley 1988, Noyes 1990, Gib- son 1990), largely perhaps because of the considerable plasticity in adult morphol- ogy displayed by many of the families, which may result because of the undoubt- edly polyphyletic natures of some family level taxa. Trjapitzin (1978) made few pro- posals about higher level relationships, and only suggested two possible group- ings, his 'pteromaloid' group comprising Pteromalidae, Tanaostigmatidae, Eupel- midae and Encyrtidae, and a 'tetracam- poid' group comprising Tetracampidae, Eulophidae, Elasmidae and Aphelinidae. A relationship between the Aphelinidae Volume 4, 1995 51 Chalckjidae Apheiinidae Ormyridae Pteromalidae Torymidae Trichogrammatidae Encyrtidae Cynipoidea EvantOKJea CeraphronoKiaa Trigonatyoidea Megatyroidea Stephanoidea Diaprioidea Ichneumonidae Braconidae Chrysidoidea (lO-l'^ Vespoidea + Apoidea Cephoidea Tenthredinoide Megaiodontoidea Xyeloidea Fig. 4. Chromosome number and evolution of the parasitic Hymenoptera, with haploid values overlain on a cladogram in which superfamilial relationships are those proposed by Rasnitsyn (1988) but with relation- ships within the Chalcidoidea based on current karyological evidence for clarity. Ranges and (in parentheses) modal values. and the Encyrtidae has also been pro- posed by a number of workers, but Gib- son (1986) and WooUey (1988) considered that the supposed synapomorphies might be better regarded as resulting from con- vergence. LaSalle (1987) upheld Trjapit- zin's view that the Tanaostigmatidae, Eu- pelmidae and Encyrtidae form a mono- phyletic group, citing two putative syna- pomorphies. Boucek (1988b) suggested that four families, the Chalcididae, Eury- tomidae, Torymidae and [some] Pterom- alidae were relatively 'ancient', partly be- cause of their 5 segmented tarsi, compar- 52 Journal of Hymenoptera Research © © ® »<^\ ^'^Z *v> © ® I ¥ r Fig. 5-10. Photomicrographs of chromosomes of adult female Hymenoptera revealed by Giemsa staining of ovarian tissue. 5, Behjia depressa (Diapriidae), meiosis, diakinesis in mature egg {2n = 16); 6, 7, Anastatus catalonicus (Eupelmidae), 6, meiosis, diplotene figures in developing oocyte, 7, metaphase mitosis (2« = 10); 8, Diaretiella rapae (Braconidae, Aphidiinae), metaphase mitosis {2n = 12); 9, Meteorus versicolor (Braconidae, Meteorinae), mitosis, one cell at pro-metaphase and one at metaphase {2n = 16); 10, Laelius utilis (Bethylidae), metaphase mitosis {2« = 20). Volume 4, 1995 53 atively large size and thoracic structure, they do not agree well with the hypothesis and he agreed with Trjapitzin and LaSalle of a close relationship between Eupelmi- that at least the Eupelmidae and Encyrti- dae and Encyrtidae (Trjapitzin 1978, La- dae might form a closely related group. Salle 1987, Boucek 1988a, Woolley 1988, However, this view was not supported by Noyes 1990) which have haploid numbers Gibson (1989, 1990) who considered both of 5 (Figs 6, 7) and 8-11 respectively, be- that the Eupelmidae might not be mono- cause if a modal haploid chromosome phyletic and that the characters used to number of 5 represents a synapomorphy, unite them with the Encyrtidae and Tan- then this relationship would require either aostigmatidae are ". . . either primitive two separate reductions (from c. 10) or a features or apparently were derived in- reversal. However, a close relationship be- dependently several times". tween Eupelmidae and Encyrtidae is not The present data could suggest that universally accepted (Gibson 1989, 1990), there has been an approximate halving of and chromosome number should be in- chromosome numbers within the Chalci- corporated in future cladistic analyses of doidea, from around a modal number of the superfamilies as an independent char- 10 as shown by the Eurytomidae and En- acter. cyrtidae to 5 or 6 in Aphelinidae (but see Potentially significant variation occurs below), Chalcididae, Eulophidae, Leucos- within the Braconidae and the Ichneu- pidae, Ormyridae, Pteromalidae, Torymi- monidae. In the former family, several dae and Trichogrammatidae. It should be subfamilies have haploid numbers be- noted that within the Aphelinidae, whilst tween 8 and 11, for example, the Bracon- most taxa examined have haploid values inae (Habrobrncon), Meteorinae {Meteonis; of 5, Pteropterix orientalis has an n value of Fig. 9), and Miracinae {Mirnx). However, 11 (referred to as Archenomus orientalis by rather higher numbers from 12 to 17 are Baldanza et al. 1991a). However, P. orien- found in the Doryctinae (Heterospilus), talis is an highly derived species within Alysiinae {Phaenocarpa, Dacnusa) and Opi- the Coccophaginae, being either a sister inae (Biosteres), whereas substantially low- group of Coccophagoides or representing a er numbers {n = 4 to 7) are found in the derived branch within Encarsia itself (A. Aphidiinae {Aphidius, Diareticlla (Fig. 8), Polaszek personal communication). In ei- Ephedrus, Praon), in the exothecine genus ther case, the high haploid chromosome Rln/sipolis (n = 6), and in the unrelated number in this taxon is clearly derived Charmontinae {Charmon; n = 5). The pos- with respect to the values of 5 (and 6) dis- session of low values within the four gen- played by the other aphelinids studied era of Aphidiinae examined provides a (collectively representing both the Cocco- potential synapomorphy for the group, phaginae and Aphelininae), and these which otherwise seems an heterogeneous lower values may therefore be taken as assemblage in which members are united representing the ancestral range of values by few characters other than biology, and for the family as a whole. The n value of more taxa will need to be examined before 11 in Pteropterix orientalis, being nearly this can be confirmed. The low n values twice that found in the other aphelinids found in Rln/sipolis might be an autapo- investigated, is further suggestive that this morphy, and it would be interesting to taxon could have originated through a know something about chromosome num- polyploidy event. bers in the apparently closely related Cli- The karyological data summarized here nocentrini and other Rogadinae s.s.. The lend some support to Boucek's view that haploid number of 5 found in Charmon the Eurytomidae are a relatively ancient was especially surprising as the subfamily and underived family of Chalcidoidea but seems to be close to the Macrocentrinae VARIATION AND IMPLICATIONS FOR HYMENOPTERA SYSTEMATICS 54 Journal of Hymenoptera Research (Quicke and Achterberg 1990, Quicke et any great extent in studies of Hymenop- al. 1994) for which we obtained a value of tera systematics. ^^Within the Ichneumonidae, most data ,, . J^J^l ^^D INTEFSPECIFIC available up until now were for the sub- family Ichneumoninae but a few chromo- some numbers for the Campopleginae Only with the advent of techniques for (Venturia), Cryptinae {Agr other eutes, Mas- examining chromosome number and mor- triis), Orthocentrinae (Orthocentrus), and phology in adult Hymenoptera has it be- Orthopelmatinae (Orthopelma) had also come possible to study intraspecific vari- been published. The overwhelming major- ation in nature and thus even to reveal the ity of the species had haploid chromo- presence of hitherto unsuspected cryptic some numbers modal at 11. We extended species or species complexes. As will be this data set by making chromosomal apparent from Table 1, chromosome num- preparations also for the Banchinae {Glyp- ber is often relatively consistent within a ta, Lissonotn), Pimplinae {Ephialtes and Po- single genus. For example, in the Ichneu- lysphincta), and Tryphoninae (Dyspetes). monidae, the 5 species of Diadromus for These additional data generally support which chromosome numbers are available the earlier findings, although the ephial- all have a haploid value of 11. Even for tine genus Ephialtes had an n value of 15, the large genus Ichneumon, 18 of the 25 rather higher than appears typical for the species examined have n = 12. family, while the polysphinctine genus Po- Although chromosome numbers are lysphincta had the more typical, even low, usually considered as differentiating char- haploid value of 9. acters, serving to help distinguish between In addition to chromosome number, closely related forms, they may also be chromosome size and structure have been used in an integrative fashion, providing used extensively for cytotaxonomic pur- evidence for uniting related forms if the poses in other groups of organisms. Dis- chromosome number represents a syna- cussion of size and centromere position in pomorphy. For example, the discovery the parasitic Hymenoptera is currently se- that all members of the cynipid genus Di- verely hampered by the relative paucity of plolepis have n=9, whereas the haploid data; in fact, in many illustrations, and es- values for all members of the other cyni- pecially among the earlier ones, centro- pid genera investigated to date for which meres are hardly (if at all) discernable. reliable figures are available is 10, pro- However, if we look at the karyotypes of vides additional evidence for the mono- the best studied groups (i.e. Ichneumoni- phyly of Diplolepis. dae, Torymidae, Cynipidae), bi-armed Chromosome numbers are fixed in the (metacentric in the broad sense) chromo- great majority of species of parasitic somes predominate in most cases (Figs 7- wasps whose populations have been stud- 10). Some Hymenotera, for example sev- ied in detail in the field. Two possible ex- eral Diplolepis species (Sanderson 1988), ceptions, however, are the ichneumonines may also have numerous acrocentric chro- Icheumon extensorius and I. suspiciosus, mosomes. Even less can be said about both of which were revealed by Gokhman chromosome size, though in general in the (1993a) to comprise individuals with two parasitic Hymenoptera it is inversely re- different diploid numbers, 24 and 26. lated to chromosome number. Much more Since specimens with the intermediate by way of comparative and quantitative chromosome number were not found data will have to be assembled before it (with one possible exception in /. extenso- will be possible to use these features to rius) the possibility that these represent Volume 4, 1995 55 Table 2. Sibling species in the parasitic Hymenoptera detected by karyological features Family Species Chromosomal characteristics Reference Ichneumonidae Ichneumonidae Ichneumonidae Ichneumonidae Encyrtidae Pteromalidae Torymidae Torymidae Aethecerus dispar Wesmael Aethecerus ranuii Gokhman Tycherus australogemiiius Gokhman Tycherus ischionieliniis (Gravenhorst) Ichneumon extensorius L. Ichneumon suspiciosus Wes- mael Copidosoma "tnincatellum (Dalman)'" Copidosoma floridanum (Ash- mead) C. floridamim (Ashmead) Nasonia vitripennis (Walker) Nasonia vitripennis (Walker) Torymus californicus (Ash- mead) Totrymus warreni (Cocker- ell)^ Monodontomerus obscurus (Westwood)'* Monodontomerus obscurus (Westwood)^ 2n = 24 2» = 22 2n = 22 2« = 18 2n = 24, 26' 2n = 24, 26 n = 10, 2n = 20 n = 11, 2n = 22 n = 8,2n = 16 n = 5 + 0 - IB, 2n = 10 n = 6, 2« = 12 n = 6(6M), 2n = 12 n = 6(5M + lA), 2n = 12 n = 6, 2n = 12 Gokhman 1991a Gokhman 1991a Gokhman 1993a Gokhamn 1993a Hunter & Bartlett 1975 Strand & Ode 1970 Leiby 1922; Patterson 1917, 1921; Patterson & Porter 1917 Nur ct al. 1988 and others Goodpasture 1974 Goodpasture & Grissell 1975 Goodpasture & Grissell 1975 McDonald & Krunic 1971 Abbreviations: A = acrocentric; M = metacentric (in a broad sense). ' A single, probably aneuploid, female specimen with 25 chromosomes has also been found. ^Apparently belongs to C. floridamim. ■ Apart from T. californicus, the second chromosome pair in T. ivarreni has secondary constrictions. ■* Apparently belongs to M. laticornis Grissell & Zerova. two sibling species rather than chromo- somal races, seems much more likely, though the reverse cannot at present be discounted. Despite intensive effort, in neither case was it possible to detect any external morphological criteria to permit delineation of these putative taxa in the absense of karyological evidence. Further, in one instance, a chromosomal polymor- phism in terms of C-banding pattern has been detected in the ichneumonine spe- cies, Dirophanes invisor (Thunberg). The C- banding patterns of the two homologous chromosomes of the second pair of meta- centrics do not differ from one another in some individuals, but in others, this pair is obviously heteromorphic, one member of the pair being substantially longer than the other, and its segment of pericentric heterochromatin is also much more devel- oped (Gokhman 1993b). Several other recently discovered ex- amples of apparent interspecific variation within other parasitic Hymenoptera are summarized in Table 2 and discussed be- low in more detail. The torymid chalci- doids Torymus californicus and T. ivarreni were considered to form a single species by Grissell (1973a). However, karyological analysis has shown that despite the fact that these two wasps have the same chro- mosome number (2n = 12), T. ivarreni has 56 Journal of Hymenoptera Research a pair of acrocentric chromosomes and has secondary constrictions on the second largest pair of submetacentrics, whilst in T. californicus all the chromosomes are bi- armed and the second pair has no con- strictions (Goodpasture and Grissell 1975). Thus these two may well be best inter- preted as sibling species. Other cases of possible karyologically-detected sibling species in the Chalcidoidea are more prob- lematical. Goodpasture (1975a) and Mc- Donald and Krunic (1971) reported n = 4 and n = 6 respectively for apparently the same torymid species, Monodontomerus ob- scurns. The most likely explanation for this difference is that one of these works in- volved a misidentified species with the specimens examined by McDonald and Krunic actually belonging to the very sim- ilar species, M. laticornis Grissell & Zerova, described 14 years later (Zerova and Gris- sell 1985). Two new ichneumonid species of the subfamily Ichneumoninae, Tycherus australogeminus and Aetheceriis ranini, were originally detected on the basis of karyo- logical evidence, but in each case reliable morphological differences were also found (Gokhman 1991a; see also Table 2). Application of karyology may also be important in laboratory cultures as avail- able evidence suggests that, at least in some instances, strains that were believed to belong to a single species may in fact represent more than one, with different laboratories working on different entities. For example. Hunter and Bartlett (1975) working with what they referred to as Copidosoma truncatellum, reported it as having a haploid number of 10. C. trun- catellum was subsequently partly synony- mized with C. floridanum, but Strand & Ode (1990) reported n = 11 for apparently the same species. Several earlier workers had reported the haploid number for C. floridanum (as Paracopidosomopsis floridan- us) to be 8 (Leiby 1922, Patterson 1917, 1921, Patterson and Porter 1917), but their findings have to be treated with consid- erable caution as the techniques for fixa- tion and preparation used in pre-1930 studies are often unreliable. Also in the genus Copidosoma, Hegner (1915) reported n = 11 and Patterson (1921) reported n = 10 for C. gelechiae. Again these data may not be fully reliable for technical reasons. However, it is harder to interpret the ap- parent conflict in reported numbers for the widely studied pteromalid, Nasonia vitri- pennis. Many workers (e.g. Gershenzon 1946, 1968, Pennypacker 1958, Whiting 1960, 1968) have reported an n value of 5, but Goodpasture (1974) working on the University of California at Davis culture found n = 6. As with the case of the ich- neumonines discussed below, intensive morphological investigation of these and other strains by Goodpasture failed to re- veal any differences. Therefore, the possi- bility that the Davis culture had devel- oped as a unique chromosomal race must be considered. Such variants are not un- common in cultures of other organisms. However, it should be noted that Darling and Werren (1990) recently discovered two cryptic species of Nasonia in North America, and the karyological results could also reflect a sibling species com- plex. The discovery of sibling species that can only reliably be separated by karyotype may pose a considerable nomenclatural problem, since the current Code of the Zoological Nomenclature requires new taxa to be differentiated from existing ones. As it is not normally possible to ob- tain karyological data from the type spec- imens of the species that have already been described, it would not possible to give scientific names to both of them (Gokhman 1993a) unless it were possible (for instance through geographic distri- bution) to infer the karyotype of the de- scribed taxon. Intraspecific karyotypic variation may also be of interest for population cytoge- netics. This type of variation is favoured in Hymenoptera by some characteristics of their genetic system, which allows the sur- Volume 4, 1995 57 vival and comparatively high viability of aneuploids (Imai et al. 1984). The data ob- tained, for example, for the ichneumonid Tycherus heUicornis suggest the long-term persistence of a chromosomal population polymorphism, probably induced by a translocation and subsequent non-disjunc- tion of chromosomes (Gokhman 1989). Occasional aneuploid specimens were also found in Ichneumon extensor ius and /. gra- cilentiis. Perhaps the most interesting case of numerical chromosomal polymorphism is described by Nur et al. (1988) and Wer- ren (1991) and occurs in the pteromalid, Nasonia vitripennis. This type of polymor- phism implies the existence of a particular B chromosome. Being transmitted pater- nally into the diploid zygotes, this chro- mosome eliminates all other chromosomes of the paternal set from the zygote, thus converting the originally diploid zygote into a haploid one. Therefore this B chro- mosome may be considered as the most selfish genetic element ever known. FUNCTIONAL IMPLICATIONS Reduced chromosome numbers will generally be associated with reduced lev- els of recombination (Vorontsov 1966, White 1973). Under many circumstances, a reduction in recombination will be mal- adaptive. However, situations that favour high levels of inbreeding, for example, ei- ther parasitisation of aggregated hosts or gregarious parasitism combined with mat- ing near the emergence site, may lead to selection in favour of low intrinsic levels of genetic diversity. Such situations may, for example, favour parthenogenesis or in the case of the parasitic Hymenoptera, thelytoky. Similarly, under such circum- stances, high levels of recombination, or large numbers of separate linkage groups, will not be favoured, and consequently, mutations leading to reduced recombina- tion will not necessarily be so deleterious. Such circumstances will permit a reduc- tion in chromosome number more readily than will situations favouring high levels of genetic diversity. Consideration of the data currently available for parasitic Hy- menoptera may be relevant in this respect. For example, in the Aphidiinae the chro- mosome number is markedly lower than in the rest of the Braconidae. Aphidiines parasitise aphids which in turn often form clonal patches. In chalcidoids, many spe- cies are also gregarious or attack clustered hosts, perhaps in part as a result of their small size and dispersal capabilities. It would be difficult to quantify this, but the possibility exists that inbreeding may be more common in the Chalcidoidea as a whole than in the Ichneumonoidea or Cy- nipoidea (Askew 1968). Information on re- combination levels in parasitic Hymenop- tera is extremely limited to date (Crozier, 1975), being based on linkage data {Habro- bracon hebetor) and chiasmata per bivalent arm {Aphytis mytilaspidiis). Further studies taking into account cross-over and levels of heterozygosity in parasitic wasps with different biologies and different chromo- some numbers and morphologies might provide additional evidence in this re- spect. FURTHER PERSPECTIVES Though we have managed for the first time to obtain cytogenetic information for several major groups of parasitic Hyme- noptera, the overwhelming majority of these insects still remains untouched by karyological investigation, and this in- cludes some entire superfamilies such as the Stephanoidea, Megalyroidea, Trigon- alyoidea, Evanioidea and Ceraphronoidea (Fig. 4). The data available at present, however, suggest that the new chromo- somal evidence may be a substantial help in future phylogenetic and taxonomic studies. As regards the higher level phy- logeny of the parasitic Hymenoptera, new karyological evidence is especially needed for various groups of Proctotrupoidea sen- su lato and of Chalcidoidea, especially of such apparently underived taxa as the Mymaridae, Tetracampidae and Rotoiti- 58 Journal of Hymenoptera Research dae (Noyes 1990) and the pteromalid sub- family Cleonyminae (Boucek personal communication). Recent investigations (e.g. Gokhman 1991b, 1994, Costa et al. 1993, Odierna et al. 1993) also show that other karyological data, especially those obtained using dif- ferential chromosome staining (e.g. C- banding), can still provide useful infor- mation even in the absence of differences in chromosome number. More detailed morphological investigations are therefore particularly likely to be of use in future species level work. ACKNOWLEDGEMENTS We would like to thank Efi Kazantzidou (Horti- cultural Research International, Wellesbourne), Prof. Helmut van Emden (University of Reading) and Robert Belshaw, Peter Mayhew and Patricia Reader (Imperial College) for providing living material for original investigation, Lynn Kimsey for enabling us to see C. Goodpasture's Ph.D. thesis, Andy Polaszek for helpful discussions pertaining to the Aphelini- dae, and Zdenek Boucek, John LaSalle, and Mark Shaw for identifying various specimens. This work was partly supported by grants from the British Council and International Science Foundation to VEG; DLJQ is supported by the NERC Initiative in Taxonomy. LITERATURE CITED Abe, Y. 1994. The karyotype of the chestnut gall wasp Dn/ocosmus kiiriphihis (Hymenoptera, Cy- nipidae). Applied Entomology and Zoology 29:299- 300. Amalin, D. M., L. M. Rueda and A. A. Barrion. 1988. Cytology of a parasitic wasp, Dirhinus himalay- anus Westwood (Chalcididae: Hymenoptera). Philippine Entomologist 7:272-274. Askew, R. R. 1968. Considerations on speciation in Chalcidoidea (Hymenoptera). Evolution 22:642- 645. Baldanza, F., L. Gaudio and G. Viggiani. 1991a. Ri- cerche cariologiche suW Archenomus orientalis Sil- vestri (Hymenoptera: Aphelinidae), parassitoide di Pseudaulacapis pentagona (Targioni Tozzeti) (Homoptera: Diaspididae). Atti XVI Congresso na- zionale italiano di Entomologia Bari-Martina Franca (Ta) 23/28 settembre 1991:457^61. Baldanza, P., G. Odierna, and G. Viggiani. 1991b. A new method for studying chromosomes of par- asitic Hymenoptera, used on Encarsia berlesei (Howard) (Hymenoptera: Aphelinidae). Bolletti- no del Laboratorio di Entomologia Agraria "Filippo Silvestri" 48:29-34. Basibuyuk, H. H. and D. L. J. Quicke. 1994. Evolu- tion of antennal cleaner structure in the Hyme- noptera (Insecta). Norwegian Journal of Agricultur- al Science, Supplement 16:199-206. Basibuyuk, H.H. and D. L. J. Quicke. 1995. Mor- phology of the antenna cleaner in the Hymenop- tera with particular reference to non-aculeate families (Insecta). Zoologica Scripta 24:157-177. Bigot, Y., M. H. Hamelin and G. Periquet. 1991. Mo- lecular analysis of the genomic organization of the Hymenoptera Diadromus pulchellus and Eu- pelmus iniilleti. Journal of Evolutionary Biology 4: 541-556. Blackman, R. L. 1980. Chromosome numbers in the Aphididae and their taxonomic significance. Sys- tematic Entomology 5: 7-25. Boucek, Z. 1988a. An overview of the higher classi- fication of the Chalcidoidea (Parasitic Hymenop- tera). In, V. K. Gupta (Ed.). Advances in Parasitic Hymenoptera Research. E. J. Brill: Leiden, pp. 11- 23. Boucek, Z. 1988b. Australasian Chalcidoidea (Hymenop- tera). A biosystematic revision of genera of four- teen families, with a reclassification of species. CAB International: Wallingford. 832 pp. Costa, M. A., G. A. R. de Melo, S. das Gracas Pom- polo, and L. A. O. Campos. 1993. Karyotype and heterochromatin distribution (C-band patterns) in three species of Microstigmus wasps (Hyme- noptera, Sphecidae, Pemphredoninae). Revista Brasileira de Genetica 16: 923-926. Crozier, R. H. 1975. Animal Cytogenetics. Vol. 3, Part 7. Gebriider Borntraeger: Berlin, pp. 1-95. Crozier, R. H. 1977. Evolutionary genetics of the Hy- menoptera. Annual Review of Entomology 22: 263- 288. Crozier, R. H. and E. F. Taschenberg. 1972. Chro- mosome number polymorphism in the sawfly Ja- nus integer (Hymenoptera: Cephidae). Psyche 79: 111-119. Darling, D. C. and J. H. Werren. 1990. Biosystematics of Nasonia (Hymenoptera: Pteromalidae): Two new species reared from birds' nests in North America. Annals of the Entomological Society of America 83: 352-370. Dessart, M. P. 1987. Quelques cas de la determina- tion du sexe chez les hymenopteres. Bulletin et Annales de la Societe Royale Entomologique de Bel- gique 123:55-58. Dijken, M. J. van, 1991. A cytological method to de- termine primary sex ratio in the solitary parasit- oid Epidinocarsis lopezi. Entomologia Experimentalis et Applicata 60:301-304. Dijkstra, L. J. 1986. Optimal selection and exploita- tion of hosts in the parasitic wasp Colpoclypeus Volume 4, 1995 59 floras (Hym., Eulophidae). Netherlands Journal of Zoology 36:177-301. Dodds, K. S. 1938. Chromosome numbers and sper- matogenesis in some species of the hymenopter- ous family Cynipidae. Genetica 21:177-190. Doncaster, L. 1910. Gametogenesis in the gall-fly, Neurolenis lenticularis (Spntliegaster baccaruni). Part I. Proceedings of the Royal Society, Series B 82: 88-112. Doncaster, L. 1911. Gametogenesis in the gall-fly, Neuroteriis lenticularis. Part II. Proceedings of the Royal Society, Series B 83:476-489. Doncaster, L. 1916. Gametogenesis and sex-deter- mination in the gall-fly, Neuroterus lenticularis (Spathegaster baccaruni). Part III. Proceedings of the Royal Society, Series B 89:183-200. Dozortseva R. L., 1936. Chromosome morphology in the parasitic wasp Pteromalus puparum. Bulletin of the Academy of Sciemces of the USSR. 6:1220-1231. (In Russian with English summary) Dreyfus, A. and M. E. Breuer. 1944. Chromosomes and sex determination in the parasitic hymenop- teran Telenonnis fariai Lima. Genetics 29:75-82. Emeljanov, A. F. and V. I. Kirillova. 1989. Trends and modes of karyotype evolution in the Cicadina (Homoptera). I. Entomologicheskoye Obozreniye 68: 587-603. Emeljanov, A. F. and V. 1. Kirillova. 1991. Trends and modes of karyotype evolution in the Cicadina (Homoptera). II. Entomologicheskoye Obozreniye 70: 796-817. Frohlich, M. W. 1987. Common-is-primitive: A par- tial validation by tree counting. Systematic Botany 12:217-237. Fukada, H. and M. Takemura. 1943. Genetical stud- ies of Trichogramma. Japanese Journal of Genetics 19:275-281.' Gauld, I. D. and B. Bolton. 1988. The Hymenoptera. British Museum (Natural History) /Oxford Uni- versity Press:Oxford. 332 pp. George, K. and S. Geethamma. 1992. Cytology and evolution of jasmines. Cytologia, Tokyo 57: 27-32. Gershenzon, S. M. 1946. The genetic structure of the natural populations of Mormoniella vitripennis Wlk. (Chalcididae: Hymenoptera). Zhurnal Ob- shchei Biologii 7:165-173. (in Russian). Gershenzon, S. M. 1968. The chromosomes and sex determination in the parasitic wasp, Mormoniella vitripennis Wlk. Tsitologiya i Genetika 2:3-13 (in Russian). Gibson, G. A. P. 1985. Some pro- and mesothoracic structures important for phylogenetic analysis of Hymenoptera, with a review of terms used for the structures. Canadian Entomologist 117: 1395- 1443. Gibson, G. A. P. 1986. Evidence for monophyly and relationships of Chalcidoidea, Mymaridae, and Mymarommatidae (Hymenoptera: Terebrantes). Canadian Entomologist 118:205-240. Gibson, G. A. P. 1989. Phylogeny and classification of Eupelmidae, with a revision of the world gen- era of Calosotinae and Metapelmatinae (Hyme- noptera: Chalcidoidea). Memoirs of the Entomolog- ical Society of Canada 149: 1-121. Gibson, G. A. P. 1990. A word on chalcidoid classi- fication. Chalcid Forum 13: 7-9. Gokhman, V. E. 1985. Chromosome sets in some Ich- neumoninae (Hymenoptera: Ichneumonidae). Zoologichesky Zhurnal 64: 1409-1413. (In Russian) Gokhman, V. E. 1987. Chromosomes in the Ichneu- moninae (Hymenoptera: Ichneumonidae). Zoolo- gichesky Zhurnal 66:543-548. (In Russian) Gokhman, V. E. 1989. Karyotypes of ichneumon flies of the Tycherus osculator group (Hymenoptera: Ichneumonidae). Entomologicheskoye Obozreniye 68: 710-714. (In Russian) Gokhman, V. E. 1990a. Karyology and systematics of the subfamily Ichneumoninae (Hymenoptera: Ichneumonidae). Summary of the PhD thesis. Moscow State University: Moscow. 17 pp. (In Russian) Gokhman, V. E. 1990b. Main trends of the karyotype evolution in Ichneumoninae (Hymenoptera: Ich- neumonidae). Zoologichesky Zhurnal 69:70-80. (In Russian) Gokhman, V. E. 1991a. New species of Phaeogenini (Hymenoptera: Ichneumonidae) from the Euro- pean part of the USSR. Zoologichesky Zhurnal 70: 73-80. (In Russian) Gokhman, V. E. 1991b. Taxonomic aspects of kar- yology of the Ichneumoninae (Hymenoptera: Ichneumonidae). / AlTRiissian Conference on Insect Genetics. Abstracts. Moscow, p. 32. (In Russian) Gokhman, V. E. 1993a. New data on the karyology of Ichneumonina (Hymenoptera: Ichneumoni- dae). Zoologichesky Zhurnal 72:85-91. (In Russian) Gokhman, V. E. 1993b. Sibling species and chromo- somal polymorphism in natural populations of Ichneumoninae (Hymenoptera, Ichneumonidae). Karyosystematics of the invertebrate animals. II. S. Petersburg, pp. 25-27. (In Russian) Gokhman, V. E. 1994. Cytogenetics of the parasitic Hymenoptera: karyotype diversity and taxonom- ic implications. 5th European Congress of Ento- mology. Uni\'ersity of York, UK. 29 August-2 September 1994. Abstracts, p. 87. Goodpasture, C. 1974. Cytological data and classifi- cation of the Hymenoptera. Unpublished Ph.D. thesis. University of California: Davis. 178 pp. Goodpasture, C. 1975a. Comparative courtship be- haviour and karyology in Monodontomcrus (Hy- menoptera: Torymidae). Annals of the Entomolog- ical Society of America 68:391-397. Goodpasture, C. 1975b. The karyotype of the cynipid 60 Journal of Hymenoptera Research Callirhytis palmiformis (Ashmead). Annals of the Entomological Society of America 68:801-802. Goodpasture, C. and E. E. Grissell. 1975. A karyo- logical study of nine species of Torymus (Hyme- noptera: Torymidae). Canadian journal of Genetics and Cytology 17:413-432. Grissell, E. E. 1973a. Revision of western Nearctic species of Torx/mus Dalman. Unpublished Ph.D. thesis. University of California: Davis. Grissell, E. E. 1973b. New species of North American Torymidae (Hymenoptera). Pan-Pacific Entomolo- gist 49:232-239. Guhl, A. P. and R. L. Dozortseva, 1934. A contribu- tion to the knowledge of sex determination in Hymenoptera. Compiles Rendiis Academic Sciences Russie. 3:524-526. (In Russian with English sum- mary) Hedderwick, M. P., M. el Agoze, P. Garaud and G. Periquet. 1985. Mise en evidence de males het- erozygotes chez I'hymenoptere Diadromus pul- chellus (Ichneumonide). Genetique Selection et Evo- lution 17:303-310. Hegner, R. W. 1915. Studies of germ cells. IV. Pro- toplasmic differentiation in the oocytes of certain Hymenoptera. Journal of Morphology 26:495-561. Henking, H. 1892. Untersuchen iiber die ersten En- twicklungsvorgange in der Eiern der Insekten. III. Specielles und AUgemeines. Zeitschrift fiir Wissenschaftliche Zoologie 54:1-274. Heraty, J. M., J. B. WooUey, and C. Darling. 1994. Phylogenetic implications of the mesofurca and mesopostnotum in Hymenoptera. Journal of Hy- menoptera Research 3: 241-277. Hogben, L. T. 1920. Studies on Synapsis. I. Oogenesis in the Hymenoptera. Proceedings of the Royal So- ciety, Series B 91:268-293. Hoshiba, H., M. Matsuura and H. T. Imai. 1989. Karyotype evolution in the social wasps (Hy- menoptera, Vespidae). Japanese Journal of Genetics 64:209-222. Hung, A. C. F. 1982. Chromosome and isozyme studies in Trichogramma (Hymenoptera: Tricho- grammatidae). Proceedings of the Entomological So- ciety of Washington 84:791-796. Hung, A. C. F. 1986. Chromosomes of three Brachy- meria species (Hymenoptera: Chalcidoidea). Ex- perientia 42:579-580. Hung, A. C. F., Day, W. H. and Hedlung, R. C. 1988. Genetic variability in arrhenotokous and thelyt- okous forms of Mesochonis nigripes Ratzeburg. (Hym.: Ichneumonidae). Entomophaga 33:7-15. Hunter, M. S. 1993. Sex allocation in a field popu- lation of an autoparasitoid. Oecologia 93:421^28. Hunter, K. W., Jr., and A. C. Bartlett, 1975. Chro- mosome number of the parasitic encyrtid Copi- dosoma truncatellum (Dalman). Annals of the En- tomological Society of America 68:61-62. Hunter, M. S., U. Nur and J. H. Werren. 1993. Origin of males by genome loss in an autoparasitoid wasp. Heredity 70:162-171. Imai, H. T., R. H. Crozier and R. W. Taylor. 1977. Karyotype evolution in Australian ants. Chro- mosoma 59:341-393. Imai, H. T. and R. W. Taylor. 1989. Chromosomal polymorphism involving telomere fusion, cen- tromeric inactivation and centromere shift in the ant Myrmecia (pilosula) n = 1. Chromosoma 98:456- 460. Imai, H. T., R. W. Taylor, M. W. J. Crosland and R. H. Crozier. 1988. Modes of spontaneous chro- mosomal mutation and karyotype evolution in ants with reference to the minimum interaction hypothesis. Japanese Journal of Genetics 63:159- 185. Imai, H. T., R. W. Taylor, M. Kubota, K. Ogata, and M. Y. Wada. 1990. Notes on the remarkable kar- yology of the primitive ant Nothomyrmecia ma- crops, and of the related genus Myrmecia (Hy- menoptera: Formicidae). Psyche 97:133-140. Johnson, N. F. 1988. Midcoxal articulations and the phylogeny of the order Hymenoptera. Annals of the Entomological Society of America 81:870-881. Kerr, W. E. 1972. Numbers of chromosomes in some species of bees. Jouriial of the Kansas Entomological Society 45: 111-122. Kerr, W. E. and Z. V. da Silveira. 1972. Karyotypic evolution of bees and corresponding taxonomic impications. Evolution 26:197-202. Konigsmann, E. 1977. Das phylogenetische System der Hymenoptera. Teil 2: Symphyta. Deutsche Entomologische Zeitschrift 24:1-40. Koonz, K. H. 1936. Some unusual cytological phe- nomena in the spermatogenesis of haploid par- thenogenetic Hymenopteran, Aenoplex smithii (Packard). Biological Bulletin 71: 375-385. Koonz, K. H. 1939. Spermatogenesis of a haploid parthenogenetic hymenopteran, Spilocryptus ex- trematus (Cresson). Transactions of the American Microscopical Society 58: 292-303. Kuznetsova, V. G. 1985. Phylogenetic analysis of the chromosome variability and karyosystematics of the leaf-hoppers of the family Dictyopharidae (Homoptera, Auchenorryncha). Entonwlogiches- koye Obozreniye 64: 539-553. (in Russian) LaSalle, J. 1987. New World Tanaostigmatidae (Hy- menoptera, Chalcidoidea). Contributions of the American Entomological Institute 23: 1-181. Leiby, R. W. 1922. The polyembryonic development of Copidosoma gelechiae with notes on its biology. Journal of Morphology 37:195-285. Liu, W. and P. Xiong, 1988. Karyotype study in Trichogramma dendrolimi. Journal of the Wuhan University 2: 105-108. (In Chinese) McDonald, M. D. and M. D. Krunic. 1971. Chromo- some numbers of Monodontomerus obscurus and Volume 4, 1995 61 Pteromahis venustus, chalcid parasites of Megachi- le rotiindata. Arhiv Bioloskih Naiika 23: 9pp. Mackay, M. R. 1955. Cytology and parthenogenesis of the wheatstem sawfly, Cephus ductus Nort. (Hymenoptera; Cephidae). Cauadiau journal of Zoology 33: 161-174. Makino, S. 1951. An Atlas of Chromosome Numbers iu Animals. Iowa State College Press: Ames. 290pp. Martin, F. 1914. Zur Entwicklungsgeschichte der po- lyembryonalen chalcidiers Ageuiaspis (Eucyrtus) fuscicollis Dalm. Zeitschrift fur Wissenschaftliche Zoologie 110: 419^79. Mason, W. R. M. 1983. The phylogeny of the Apo- crita (Unpublished lecture notes cited in Gauld & Bolton, 1988). Moritz, R. F. A. 1986. Genetics of bees other than Apis mellifera. pp. 121-154. In T. E. Rinderer (ed.) Bee Genetics and Breeding. Academic Press: Orlan- do. Naito, T. 1982. Chromosome number differentiation in sawflies and its systematic implication (Hy- menoptera, Tenthredinidae). Kontyu, Tokyo 50: 569-587. Noyes, J. S. 1990. A word on chalcidoid classifica- tion. Chalcid Forum 13: 6, 7. Nur, U., J. H. Werren, D. G. Eickbush, W. D. Burke, and T. H. Eickbush. 1988. A 'selfish' B chromo- some that enhances its transmission by eliminat- ing the paternal genome. Science 240: 512-514. Odierna, G., F. Baldanza, G. Aprea, and E. Olmi. 1993. Occurrence of G-banding in metaphase chromosomes of Encarsia berlesei (Hymenoptera: Aphelinidae). Genome 36: 662-667. Patterson, J. T. 1917. Studies on the biology of Para- copidosomopsis. III. Maturation and fertilization. Biological Bulletin of the Marine Biological Labora- tory Woods Hole, Mass. 33: 38-50. Patterson, J. T. (1921) 1922. The development of Par- acopidosomopsis. Journal of Morphology 36:1-70. Patterson, J. T. and L. T. Porter. 1917. Studies on the biology of Paracopidosomof)sis. II. Spermatogene- sis of males reared from unfertilized eggs. Biolog- ical Bulletin of the Marine Biological Laboratory Woods Hole, Mass. 33:28. Peacock, A. D. and R. A. R. Gresson. 1932. Male hap- loidy and female diploidy in Sirex cyaneus F. (Hy- men.). Proceedings of the Royal Society of Edinburgh 57: 97-103. Pennypacker, M. I. 1958. The chromosomes of the parasitic wasp Mormoniella vitripennis. Part I. In spermatogenesis of haploid and diploid males. Archives de Biologic, Liege 69:483-495. Pompolo, S. G. and C. S. Takahashi. 1987. Cytoge- netics of Brazilian Polybiini wasps (Hymenop- tera, Vespidae, Polistinae). Revista Brasileira de Genetica 10: 483-497. Pompolo, S. G. and C. S. Takahashi. 1990. Karyotype of ten species of social wasps (Hymenoptera, Pol- istinae, Polybiini). Revista Brasileira de Genetica 13: 469-477. Quicke, D. L. J. 1993. Prijiciples ami Techniques of Con- temporary Taxonomy. Blackie Academic and Pro- fessional: Glasgow. 311 pp. Quicke, D. L. J. 1994. Phylogenetic and biological transitions in the Braconidae (Hymenoptera: Ich- neumonoidea). Nonvegian journal of Agricultural Sciences. Supplement 16: 155-162. Quicke, D. L. J. and C. van. Achterberg. 1990. Phy- logeny of the subfamilies of the family Braconi- dae (Hymenoptera: Ichneumonoidea). Zoologis- che Verhandelingen, Leiden, 258:1-95. Quicke, D. L. ]., M. G. Fitton and S. Ingram. 1992. Phylogenetic implications of the structure and distribution of ovipositor valvilli in the Hyme- noptera (Insecta). journal of Natural History 26: 587-608. Quicke, D. L. J., M. G. Fitton, J, R. Tunstead, S. In- gram and P. V. Gaitens. 1994. Ovipositor struc- ture and relationships within the Hymenoptera, with special reference to the Ichneumonoidea. Journal of Natural History 28:635-682. Rasch, E. M., J. D. Cassidy and R. C. King. 1975. Es- timates of genome size in haploid-diploid species of parasitoid wasps. Journal of Histochemistry and Cytochemistry. 23: 317. Rasch, E. M., J. D. Cassidy and R. C. King. 1977. Ev- idence for dosage compensation in parthenoge- netic Hymenoptera. Chromosoma 59:323-340. Rasnitsyn, A. P. 1980. Origin and evolution of Hy- menoptera. Trudy Palcontologicheskogo Instituta 174:1-190. (in Russian) Rasnitsyn, A. P. 1988. An outline of evolution of the hymenopterous insects (Order Vespida). Oriental Insects 22:115-145. Rossler, Y. and P. De Bach. 1973. Genetic variability in a thelytokous form of Aphytis mytilaspidus (Le Baron) (Hymenoptera: Aphelinidae). Hilgardia 42:149-175. Sanderson, A. R. 1932. The cytology of parthenogen- esis in Tenthredinidae. Genetica (The Hague) 14: 321^94. Sanderson, A. R. 1970. Further studies on the cytol- ogy of sawflies. Proceedings of the Royal Society of Edinburgh B61:29-40. Sanderson, A. R. 1988. Cytological investigations of parthenogenesis in gall wasps (Cynipidae, Hy- menoptera). Genetica (The Hague) 77:189-216. Schleip, W. (1909) 1910. Die Reifung des Eies von Rhodites rosae L. und einige allgemeine Bemer- kungen uber die Chromosomen bei parthenoge- netischer Fortpflanzung. Zoologischer Anzeigcr 35: 203-213. Schmieder, R.G., 1938. The sex ratio in Melittobia chah/bii Ashmead, gametogenesis and cleavage in females and in haploid males (Hymenoptera: Chalcidoidea). The Biological Bulletin. 74:256-266. 62 Journal of Hymenoptera Research Silvestri, F. 1908. Contribuzioni alia conoscenza biol- ogica degli Imenotteri Parassiti. II. (1) Sviluppo dell' Ageniaspis fuscicollis (Dalm.) e note biografi- che. Bolletino del Laboratorio di Zoologia Generale e Agraria della R. Scuola Superiore d'Agricoltum in Portici 3:29-53. Silvestri, F. 1914. Prime fast de sviluppo del Copido- soma bin/ssoni (Mayr), imenottere calcidide. Aua- tomische Anzeiger 47:45-56. Speicher, B. R. 1937. Oogenesis in a thelytokous wasp, Nemeritis canescens (Grav.). Journal of Mor- phologij 61:453-467. Speicher, K. G. and B. R. Speicher. 1938. Diploids from unfertilized eggs in Habrobracon. Biological Bulletin 74:247-252. Speicher, B. R. and K. G. Speicher. 1940. The occur- rence of diploid males in Habrobracon brevicornis. American Naturalist 74:379-382. Stille, B. and L. Davring. 1980. Meiosis and repro- ductive strategy in the parthenogenetic gall wasp Diplolepis rosae (L.) (Hymenoptera, Cynipidae). Hereditas 92:353-362. Stouthamer, R. and D. J. Kazmer. 1994. Cytogenetics of microbe-associated parthenogenesis and its consequences for gene flow in Trichogramma wasps. Heredity 73: 317-327. Strand, M. R. and P. J. Ode. 1990. Chromosome num- ber of the polyembryonic parasitoid Copidosoma floridanum (Hymenoptera: Encyrtidae). Annals of the Entomological Society of America 83:834-837. Torvik-Greb, M. 1935. The chromosomes of Habro- bracon. Biological Bulletin 68:25-34. Trjapitzin, V. A. 1978. Superfamily Chalcidoidea. In G. S. Medvedev (ed.) Keys to the Identification of Insects of the European Part of the U.S.S.R. Nauka Publishers: Leningrad, pp. 39-58. Viggiani, G. 1967. Richerche sugli Hymenoptera Chalcidoidea. XV. Osservazioni caryolgische pre- liminari sulli Aphelinus mali. Bollettino del Labor- atorio di Entomologia Agraria "Filippo Silvestri" 25: 326-330. Vorontsov, N. N. 1966. The karyotype evolution. In: A Manual on Cytology. Vol. 2. Nauka Publishers, Moscow /Leningrad, pp. 59-389. [in Russian] Wahrman, M. Z. and S. Zhu. 1993. Haploid and dip- loid cell cultures from a haplo-diploid insect. In- vertebrate Reproduction and Development 24: 79-86. Watrous, L. E. and Q. D. Wheeler. 1981. The out- group comparison method of character analysis. Systematic Zoology 30:1-11. Werren, J. H. 1991. The patemal-sex-ratio chromo- some of Nasonia. American Naturalist 137:392^02. White, M. J. D. 1973. Animal Cytology and Evolution. Cambridge University Press: Cambridge. 961 pp. Whiting, A. R. 1927. Genetic evidence for diploid males in Habrobracon. Biological Bulletin 53:438- 449. Whiting, P. W. 1960. Polyploidy in Morinoniella. Ge- netics 45:949-970. Whiting, P. W. 1968. The chromosomes of Mormon- iella. Journal of Heredity 59:19-22. Wieman, H. L. 1915. Observations on the spermato- genesis of the gall-fly, Dryophanta erinacei (Mayr). Biological Bulletin of the Marine Biological Labora- tory VJoods Hole, Mass. 28:34-46. Woolley, J. B. 1988. Phylogeny and classification of the Signiphoridae (Hymenoptera: Chalcidoidea). Systematic Entomologi/ 13: 465-501. Zerova, M.D. and E. E. Grissell. 1985. A new species of the genus Monodontomerus Westw. (Hymenop- tera, Torymidae) — Parasite of leafcutting bees. Entomologicheskoye Obozreniye 64:203-206. (in Russian) APPENDIX I. METHODOLOGY FOR STAINING CHROMOSOMES IN ADULT PARASITIC WASPS General comments. Chromosomes can be stained in many adult parasitic wasp females especially in recently eclosed individuals using ovarian tissue. As always in insects, care should be taken to count mul- tiple cells so that the occasional polyploid cell can be recognised and discounted. In common with Cro- zier (1975) we recommend counting at least ten in- dividual metaphase plates though occasionally this may not be possible. If fewer plates are available, one may gain extra confidence in the results if it is possible to identify through size and morphology, particular pairs of chromosomes (see for example. Fig. 9). Usually, mitotic divisions are most evident, but in some individuals and taxa, meiosis may also be observed, sometimes with very clear spreads. The number of plates may also be increased if the wasp is fed on honey water containing colchicine for a few hours before preparation, though beyond about 5 hours one stands the risk of increasing the propor- tion of polyploid cells. The general method de- scribed below is modified after that of Imai et al. (1988). Method. Metasomas of adult female wasps are dissected in hypotonic sodium citrate solution with colchicine (Solution A). Ovaries are incubated in this solution at room temperature (c. 25°C) for 20 min- utes (optimal times for different taxa may vary slightly). Following incubation, ovaries are trans- ferred to a thoroughly cleaned microscope slide, ex- cess citrate solution pipetted off and the slide is gently flushed with Fixative I taking care not to wash off the ovaries. Whilst still moist with Fixative I, the ovaries are disrupted (e.g. using fine mounted needles) and their cells spread gently over the mid- dle part of the slide. One or two drops of Fixative II are then applied to the centre of the area of spread cells and the more aqueous phase which is displaced to the edge of the slide is blotted off. The same pro- Volume 4, 1995 63 cedure is then performed with Fixative III. The slide is then air dried before staining for at least 20 min- utes. Excess stain should be washed off with dis- tilled water and the slide examined dry or under emersion oil (do not apply mounting media). Slides can be restained if the initial result was insufficient- ly intense, or they can be destained with alcohol in the reverse situation. Solutions (A) Hypotonic sodium citrate with colchicine Ig Na citrate.2H,0 5mg colchicine 100 ml distilled water (B) Fixative I 3 parts Ethanol 3 parts Glacial acetic acid 4 parts distilled water (C) Fixative II 1 part Ethanol 1 part Glacial acetic acid (C) Fixative III Glacial acetic acid (D) Stain 2 ml Giemsa solution 50 ml 0.089M Na.HPO, 50 ml 0.066M KH>0, J. HYM. RES. Vol. 4, 1995, pp. 64-76 Euryischomyia Girault (Hymenoptera: Chalcidoidea: Aphelinidae: Eriaporinae: Euryischiini) Mary Carver Division of Entomology, CSIRO, G.P.O. Box 1700, Canberra, A.C.T. 2601, Australia Abstract. — Types of some Australian Euryischiini described by A. A. Girault have been exam- ined, and compared with other material identified by Girault, together with other reared and collected material of Euryischomyia Girault. Euryischomyia saintpverrei Girault, 1915, syn. nov.; Myiocnema marmorativentris Girault, 1915, syn. nov.; E.fasciata Girault, 1916, syn. nov.; and £. setosa Girault, 1929, syn. nov.; are placed in synonymy with £. flaznthorax Girault & Dodd, 1915; and £. alami Shafee, 1975, syn. nov., (described fom India), is placed in synonymy with E. washingtoni Girault, 1914. A key to the 2 remaining species of Euryischomyia is provided. £. washingtoni and £. flavithorax are redescribed. £. zuasJiingtoni is known as a parasite of pseudococcids, and £. flavithorax as a hyperparasite of psyllids, aphids, coccids and pseudococcids. INTRODUCTION compressed hind coxae, large, bristle-like np, , ill. J i.- setae on the lees and elsewhere, postaxil- The placement, status and composition , ,, of the subfamily Eriaporinae (Chalcidoi- ^^^^^ ^^'^^ scutellum and a large propo- dea: Aphelinidae) have long been a matter of uncertainty. The subfamily was erected '^^^ nomenclatural history of the Eu- in the family Aphelinidae by Ghesquiere ^yischiini is complicated by the fact that (1955). Members have been variously ^^^^^^ ^^^^^^ ^^^ ^^'"l^^^ ^^^^^^^ ^^^ f^^" placed in the Aphelinidae (e.g. Ashmead ^^y Euryischidae {sic) to contam Euryischia 1900; Compere 1947; Ferriere 1965) and ^^^' subsequently, he included Myiocnema Elasmidae (e.g. Girault 1914; Smith and ^^^ Euryischomyia in the family also (Sha- Compere 1928; Nikol'skaya 1963). Cur- ^^^ 1^75) but, later, Shafee and Rizvi rently, the Eriaporinae are included in the (^^^O) retained only Euryischia in the fam- Aphelinidae (e.g. Hayat 1994), albeit with i^Y- transferring the other 2 genera to the some reservation (Goulet and Ruber Aphelinidae. The elevation of these genera 1993). Contained taxa of Eriaporinae pos- ^ family status, and their separation, are sess an unbroken, posteriorly expanded "^o* accepted herein. parastigma in the fore wing, which is usu- Although Euryischia is well character- ally continued posterobasally and bears 1- ised, the limits of the other genera of Eu- 3 coarse, bristle-like setae. ryischiini have not been understood and Hayat and Verma (1980) created 2 tribes their history is also confusing. Essentially, within the Eriaporinae, namely, the Eria- Girault (1914) described the new genus porini to contain the genera Promuscidea and species Euryischomyia washingtoni and, Girault and Eunotiscus Compere; and the in 1917, erected the genus Paramyiocnema Euryischini (sic) for 4 genera: EuryiscJiia for Myiocnema marmorativentris Girault but Riley, 1889; Myiocnema Ashmead, 1900; did not describe or diagnose the genus. Euryischomyia Girault, 1914; and Para- Girault (1930) placed Euryischomyia in syn- myiocnema Girault, 1917. Euryischiini are onymy with Myiocnema. Hayat and Verma characterised by possession of enlarged, (1980) transferred Euryischomyia flavithorax Vt^LUME 4, 1995 65 Girault & Dodd to Parnmyiocnema. They also opined that E. saintpierrei Girault, E. fasciata Girault and E. setosa Girault might also belong in Paramyiocnema. Shafee et al. (1985) placed Paramyiocnema in synonymy with Myiocnema and Hayat (1994) placed it in synonymy with Eiiryischomx/ia. The types of Eiiryischomyia washingtoni, E. flavithorax, E. saintpierrei, Myiocnema marmorativentris, E. fasciata, and £. setosa have been examined by the author, to- gether with other material identified by Girault, and more recently reared and col- lected material of EuryiscJioirn/ia. As a re- sult, Eiiryischomyia saintpierrei, Myiocnema marmorativentris, E. fasciata, and £. setosa are placed in synonymy with £. flavithorax. In addition, £. alami Shafee, 1975, de- scribed from India, is proposed as a junior synonym of £. washingtoni, thus reducing Eiiryischomyia to 2 species, namely, £. washingtoni and £. flavithorax. METHODS All museum material was examined as provided, as dry-mounts, or as uncleared or cleared slide-mounts. Girault's material was recognizable from information pro- vided by Dahms (1984, 1986), and by rec- ognition of Girault's or Mrs E.J. Girault's handwriting on the labels. Whenever feasible, reared material was obtained by caging live, field-collected material to allow maximal mummification of already-parasitized hosts. Aphid mum- mies (mummified, parchment-like, some- times silk-lined, aphid skins containing immature parasites) were removed and placed individually in gelatin capsules. Except when slide-mounted, each emer- gent parasite was preserved in association with its mummy. Reared material was dry-mounted, or preserved in 80% etha- nol, or cleared and slide-mounted using 10% potassium hydroxide, chloral-phenol, and gum-chloral mounting media. Micrometer eyepieces were used for measurements, all of which are of maxi- mal values; those of the scape do not in- clude the radicle; those of the metasoma include the ovipositor. Length of mesoso- ma + metasoma is given rather than body length because the head of most of Gi- rault's specimens is separated from the trunk and slide-mounted fronto-occipital- ly and, in those instances of two or more mountings per coverslip, head and trunk could not be matched with confidence. Girault customarily enumerated the number of specimens he used in a descrip- tion of a new species, and then implicitly designated only 1 or 2 types from that ma- terial. Dahms (1983) considered as types all specimens stated by Girault to have been used by him in a description. Dahms' practice is followed herein. Dahms (1984, 1986) and Hayat and Ver- ma (1980) catalogued Girault's material of Eiiryischomyia in detail, obviating the ne- cessity to do so herein. The information contained in the lists of specimens exam- ined has therefore been simplified. Where appropriate, label data are enclosed in quotation marks; added or interpreted in- formation is enclosed in square brackets. Latitudes and longitudes have in most cases been provided by the author. The terminology adopted is that of Nau- mann (1991), except for the following: An- tennal formula: representation of the num- ber of segments comprising, respectively, the scape, pedicel, anellus, funiculus, club. Postaxilla: a small triangular sclerite pos- terior to, and apparently a division of, the mesonotal axilla; a term proposed by Sha- fee (1975). Parastigmal spur: a posterobas- ally orientated extension of the parastig- mal vein, bearing 1-3 stout, bristles. Abbreviations used include the follow- ing: ANIC, Australian National Insect Col- lection, Canberra; BMNH, British Museum (Natural History), London; QM, Queens- land Museum, Brisbane; SAM, South Aus- tralian Museum, Adelaide; USNM, United States National Museum of Natural His- tory, Washington DC; WADA, Western Australian Department of Agriculture, 66 Journal of Hymenoptera Research South Perth; ZMA, Zoological Museum, Aligarh. Euryischomyia Girault Euryischomyia Girault, 1914: 285. — Girault 1915a: 178; Girault 1930: 1; Shafee 1975: 27; Hayat & Verma 1980: 35; Hayat 1983: 87; Dahms 1986: 689; Viggiani 1988: 101; Shafee & Rizvi 1990: 103. Euryisdiomyiia [sic] Girault, 1929: 331. Paramyiocnema Girault, 1917: 93. — Hayat 1994: 85. Type species Euryischomyia washingtoni Gi- rault, by monotypy and original designation. The following characters, in combina- tion, are diagnostic for Euryischomyia: Head with 7 pairs of long, stiff setae near ocelli and inner orbital margins; max- illary and labial palps 2-segmented; man- dibles tridentate (or, depending on inter- pretation, bidentate, the wide dorsal tooth being apically concave and truncate); se- tose. Antennae: scape unexpanded in both sexes; 1-2 anelli; funicle 3-segmented; club 3-segmented. fronotum with row of long, stiff setae, 3 on each side, posterosublaterally. Me- soscutum between notauli with 2-3 rows of shorter, finer setae anterior to row of 4 and row of 2 longer, stiffer setae; scutel- lum wider than long, with 2 pairs of long, stout setae, otherwise bare. Postaxillae present, bare. Phragma evenlv tapered, apically incised. Metanotum not produced posteromedially. Propodeum wide, not excavated posterolaterally. Fore wings hyaline or medially infus- cated; parastigmal spur little- or well- developed, with 1-2, rarely 3, long, bris- tle-like setae of similar size to those on submarginal vein; 0-2 bristle-like setae posterobasal to parastigma, or area pos- terior to submarginal vein and proximal of parastigmal spur with 6-13 shorter, stiff setae in 2-3 irregular rows; this area of disc otherwise bare; discal setation proximally regularly or irregularly mar- gined. Legs: hind coxa enlarged, elongate- ovate, bilaterally compressed ventrally; hind femur elongate, broad, laterally com- pressed, laminose and arcuate in section; fore and mid femora also broad and lat- erally compressed. Only one hind tibial spur. Long, bristle-like setae on legs most commonly disposed in differently sized pairs (as subapically on coxae) or in lon- gitudinal rows (as on mid and hind tibi- ae). Petiole wide, boomerang-shaped. Bases of cerci low, tuberculate. Notes. — Euryischomyia is most closely re- lated to (the monotypic) Myiocnema, which differs in possessing 2 hind tibial spurs, 3- segmented maxillary palps, 3 anelli, and a distinctive fore wing chaetotaxy; the costal cell bears numerous short, stiff setae, and the basal discal area bears both a very long bristle-like seta posterior to the par- astigmal spur and several irregular rows of shorter, stiff setae in the angle between the submarginal vein and the parastigmal spur. The chaetotactic differences may be of specific value onlv. Studies in progress by J.B. Woolley and M. Hayat show that Myiocnema differs significantly from Eu- ryischomyia in not possessing a looped me- sofurca (Hayat 1994). Euryischia is very distinctive in possess- ing almost circular, disc-like hind coxae, a posterolaterally excavated propodeum to accommodate the coxae, and a postero- medial metanotal process ('elasmid' char- acteristics). Euryischia is further distin- guishable from Euryischomyia in possess- ing 2 hind tibial spurs, 4-segmented max- illary palps, and cylindrical cereal bases. The scape of the male of some species of Euryischia is greatly enlarged, and the chaetotaxy of the mesonotum and fore wing of Euryischia is interspecifically vari- able. Hayat and Verma (1980) provided keys to the genera. Taxa of Euryischiini clearly demonstrate different degrees of development of eria- porine and euryischiine features, namely, development of the parastigmal spur; spi- Volume 4, 1995 67 nosity of the legs, fore wing and body; en- thorax, Myiocnema comperei Ashmead and largement and compression of the hind Euryischia spp. are successively more en- legs and concomitant modifications of the dowed in these respects, mesosoma. Euryischia washingtoni, E. flavi- KEY TO SPECIES OF EURYISCHOMYIA GIRAULT 1. Parastigmal spur well-developed, bearing 2, rarely 3, long, bristle-like setae. Area imme- diately posterior to submarginal vein and proximal to parastigmal spur with 2-3 irregular rows of 6-13 relatively small, stiff setae. Subapical margin of costal cell with 4-7 short, stiff setae anterior to cluster of fine, ventral setae; otherwise bare. Fore wings hyaline, not narrow. Mesoscutum with 11-21 setae in 2-3 irregular rows anterior to row of 4 and row of 2 bristle-like setae. Body coloration yellow, orange, brown and black. Distribution: Australia £. flavithorax Girault - Parastigmal spur scarcely developed, bearing 1-2 bristle-like setae. Area posterobasal to parastigma bearing 0-2 large, bristle-like setae. Costal cell ventro-subapically with some pale, fine setae; otherwise bare. Forewings narrow, 3.5 times longer than width; medially with broad band of infuscation. Mesoscutum with only 8 setae anterior to row of 4 and row of 2 setae. Body coloration predominantly black, legs predominantly lemon-yellow. Distribution: Australia and India E. washingtoni Girault Euryischomyia washingtoni Girault Euryischomyia washingtoni Girault, 1914: 285. — Girault 1915a: 178; Hayat & Verma 1980: 37; Dahms 1986: 649. Euryischomyia alami Shafee, 1975: 27. — Hayat & Verma 1980: 37. Syn. nov. Type Material Examined. — Euryischomyia washingtoni Girault: Lectotype 9, paralec- totype 6 (both on 1 slide, also parts of ho- lotype of Pleurotropomyia seditiosus Gi- rault; all separate); paralectotype 9, (on tag, minus head). (QM 4060; [Hy]2742). QUEENSLAND: 20°32'S 145°24'E Cape- ville (Pentland), by sweeping in forest along the banks of Cape River, 8. i. 1913 (Girault 1914). Other Specimen Examined; Identified by A. A. Girault. — "Euryiscliomyiia [sic] wash- ingtoni" 9 (on tag and on slide; slide also contains a paratype of Amicromelus gran- daevus Girault). QUEENSLAND: 17°23'S 145°19'E Watsonville (QM). Other Records. — Euryischomyia washing- toni: QUEENSLAND: 19°15'S 146°48'E Townsville, sweeping in forest, 19. i. 1913, 9 (Girault 1915a). Euryischomyia alami: ex Nipaecoccus viri- dis (Newstead) on Mangifera indica, MY- SORE: Bangalore, Channapatna, 2.xii.l968, holotype 9, paratype 9, S.A. Shafee (Sha- fee 1975) (ZMA). Euryischomyia alami: ex Nipaecoccus viri- dis, TAMIL NADU: Shencottah, 2 9 9, M. Hayat (Hayat and Verma 1980) (ZMA). Redescription. — Coloration of 9 : Body shiny, blackish brown to black; mesono- tum obscurely dark green metallic. Body setae mostly pale. Legs whitish except for following: fore coxae dusky basally; mid and hind coxae black except apically; mid and hind femora black except apically and basally; tarsi brown apically; sockets of hind tibial bristles black. Fore wings each with broad band of pale to dark infusca- tion extending from anterior to posterior margins in region of marginal and post- marginal veins, otherwise hyaline. Morphology: Length mesosoma + me- tasoma: 0.90 mm (paralectotype 9); 1.05 mm (lectotype 9); 1.40 mm (9 t'.v Watson- ville); 0.84 mm (paralectotype 6). All bris- tle-like setae smaller and finer than cor- responding ones of £. flavithorax. 68 Journal of Hymenoptera Research Head: submedian frontal setae (of 6): (48|JLm); interocellar setae: (68) ixm. Max- illary and labial palps longer and more slender than in E. flavithorax. Dorsalmost tooth of mandibles pointed. Antennae of 9 ex Watsonville (and S): antennal formula: 1:1:1-2:3:3; relatively longer and more slender than in E. flavi- thorax. Scape: length 5.56 (4.44) X maximal width; 1.82 (1.73) X length of pedicel; 1.22 (1.11) X length of club. Pedicel: 2.5 (2.28) X maximal width. Anellus small, wedge- shaped; a 2nd anellus may be partly de- limited from it. Funicular segments mostly as long as or longer than wide; segment 1 (fl): length, 1.18 (1.0) X width; f2: length, 1.41 (0.94) X width; f3: length 1.04 (1.14) X width. Club: length: 2.56 (3.56) X width. Lengths of segments: scape, 150 (107) |xm; pedicel, 83 (62) fxm; fl: 30 (20) fxm; f2: 47 (21) |xm; f3: 41 (30)|jLm; club: 123 (96) jim. Mesosoma of lectotype 9 (and 6): Pro- notum: 6 bristle-like setae postero-subla- terally, 68 (75) fxm long. Mesoscutum be- tween notauli with row of 2, and row of 6, fine, bristle-like setae, 33 (26-29) |xm long; row of 4 (38) iJim long; and row of 2 thicker, bristle-like setae (51) ixm long. 2 midsublateral and 2 posterolateral scutel- lar setae, (45 ixm) and 90 (93) fxm long, respectively. Axilla with 2 setae; posterior seta (34) ixm long. Legs of lectotype 9 (and 6): Hind coxa, 0.21 (0.22) mm long, 1.87 (1.74) X width, slightly shorter than hind femur, which is shorter than hind tibia; subapical setae of hind coxa: 30 (36) and 48 (54) |xm long; hind femur, 0.23 (0.26) mm long, 2.53 (2.74) X width, subapical seta, 26 (30) |xm; hind tibia, 0.31 mm long, gradually broad- ening apicad, preaxial row of setae maxi- mally 50 |xm long, subapical setae 44 and 62 fxm long, spur 48 jxm long. Fore wings of lectotype 9 (and 6 ) nar- row, 3.5 X width. Parastigmal spur hardly developed, bearing 1-2 bristle-like setae, 45 |JLm long; 1 or 2 bristle-like setae pos- terobasal of parastigma, 31 (35) ixm and (18) fxm long, respectively. Subapical area of costal cell devoid of marginal setae. Discal setation with regular margin. Metasoma in dry-mounted 9 9 up- turned at an angle of approximately 60°, strongly tapered apicad. Ovipositor sheaths 0.18 mm long, 0.59 X hind tibial length, slightly protrusive beyond apex of metasoma, densely and evenly furnished with short, bristle-like setae. Notes. — Based on a knowledge of the chaetotactic variability within the tribe, £. alami is hereby placed in synonymy with E. washingtoni, despite the fact that the types of E. alami have not been seen by the pres- ent author. £ alami reportedly differs from £. zvashingtoni in not bearing discal setae posterior to the parastigma, and in possess- ing 3 setae anteriorly on the axilla (Shafee 1975; Hayat and Verma 1980). These differ- ences are too small to warrant separation at the species level. Hayat and Verma (1980), who studied the above-listed specimens from Tamil Nadu, are also of the opinion that the differences between the 2 taxa ap- pear to be of doubtful specific value. The above description of £. zuashingtoni supplements those of Girault (1914, 1915a) and Hayat and Verma (1980) of £. wash- ingtoni and that of Shafee (1975) of £. ala- mi. Because of the paucity and poor con- dition of the material available for study, a composite and comparative method of description is adopted. £. flavithorax, the standard of comparison, is conventionally described below in greater detail. The lem- on yellow coloration described by Girault (1914, 1915a) can be presumed to have faded with time to white. The antenna of the paralectotype 9 (on slide) is com- pressed; measurements and ratios of the 9 ex Watsonville are given instead. £. washingtoni has less developed eu- ryischiine features than the other taxa of Euryischiini, having a scarcely developed parastigmal spur, and shorter, finer bris- tle-like setae on the body, limbs and fore wings. Volume 4, 1995 69 ,.J^. I I'tnAniri Fig. 1. Euryischomyia flavithorax, male (X67). Cleared, slide-mounted specimen. Biology. — Parasite of unknown status of Nipaecoccus viridis, Pseudococcidae. Distribution.— AXJSTR ALIA: Queens- land; possibly also Beenleigh, Victoria (Viggiani 1988); INDIA: Mysore, Tamil Nadu. Euryischomyia flavithorax Girault & Dodd (Figure 1) Euryischomyia flavithorax Girault & Dodd, 1915 in Girault 1915a: 178.— Dahms 1984: 614. Euryischomyia saintpierrei Girault, 1915a: 178. — Hayat & Verma 1980: 37; Dahms 1986: 495. Syn. nov. Myiociicma iimnuorativentris Girault, 1915b: 64.— Dahms 1984: 795. Syn. nov. Euryiscliomyia fasciata Girault, 1916: 212. — Hay- at & Verma 1980: 37; Dahms 1984: 579. Syn. nov. Paramyiocnema marmorativentris (Girault). — Gi- rault 1917: 93; Hayat & Verma 1980: 35. Euryischomi/i[i]a sctosa Girault, 1929: 331.— Hay- at & Verma 1980: 37; Dahms 1986: 522. Syn. nov. Paramyiocnema flavithorax (Girault & Dodd). — Hayat & Verma 1980: 33. Type Material Exmuiued. — Eiin/isdioun/ia flavithorax Girault & Dodd: Lectotype 9 (on slide); paralectotype 9 (on tag; head and hind tibia on same slide as lectotype, herein called paralectotype 1); paralecto- type 9 (also on same slide as lectotype, herein called paralectotype 2). QUEENS- LAND: 17°05'S 145°47'E "Gordonvale (Cairns)" (QM 3903; Hy2744). Euryischomyia saintpierrei Girault: Holo- type 9 (torso and 1 fore wing on tag, fore wing separate, not listed by Dahms (1986); head, funicular segments 2 and 3 and club, both hind legs on slide) (QM 3905; Hy2743). QUEENSLAND: 27°28'S 153° 02'E Brisbane (Girault 1915a). Myiocnenia marmorativentris Girault: Ho- lotype 9 (on slide) QUEENSLAND: 17°05'S 145°47'E "Gordonvale (Cairns)" (QM 3907; Hy2959). Euryischomyia fasciata Girault: Holotype 70 Journal of Hymenoptera Research 6 (not ?, as described by Girault; on tag minus antennae, right wings and leg parts; antenna minus scape, fore wing, hind femur and hind tibia on slide; anten- na not listed by Dahms (1984)). WESTERN AUSTRALIA: 31°57'S 115°51'E "Perth, W. Austr., G. Compere, Collector, 837" (USNM 19679). Euryischomyiia [sic] setosa Girault: Holo- type 9 (on tag, minus head, with 1 fore wing on slide) SOUTH AUSTRALIA: 32°49'S 138°11'E "Melrose, Oct., A.M. Lea" (SAM). Other Specimens Examined: identified, as follows, by A.A. Girault. — "Myiocnema flav- ithorax Gir. [sic], brown scale [Coccus hes- peridum] parasites, [WESTERN AUSTRA- LIA:] 9, [2]6S, coll. L.J. Newman, 70". (On 1 slide. Host identified as Coccus hes- peridum in F. Wilson (I960)). (WADA). Euryischomyiia [sic] flavithorax Gir. [sic], 9 . Host Lecanidae [Coccidae] sp. Ent. Div. Dep. Ag. & Stk Qld." (QM Hy 10). "Myiocnema saintpierrei Gir., S , 9 . Host Saissetia sp., Sydney [33°53'S 151°12'E], N.S.W., June 19-31, S.E. Flanders. Univ. Cal. Citrus Exp. Sta. Ace. No. 27)." (On slide; only 2 of 6 specimens easily observ- able). (QM). "Myiocnema setosa Gir., 9, Myiocnema saintpierrei Gir., 6 , 9, on lime scale [Coc- cidae?], [QUEENSLAND: 26°45'S 150°- 38'E] Chinchilla, Aug. 14, 1931, S.E. Flan- ders". (On slide; only 29 9, 26 S easily observable) (QM). Other Specimens Examined. — QUEENS- LAND: 15°39'S 144°31'E Split Rock, 28.v.- 26.vi.1993, 9, P. Zborowski, I.D. Nau- mann (ANIC). WESTERN AUSTRALIA: 33°45'S 122° 32 'E Condingup, 55km E of Esperance, 30.xii.l986, 6, J.S. Noyes (BMNH). ex Psyllaephagus sp. / Cardiaspina albitex- tura Taylor on Eucalyptus blakelyi, VICTO- RIA: nr 36°21'S 146°19'E Wangaratta, ll.iLl955, 39 9,76 6, E. Lewis (ANIC). ex Cardiaspina sp. on Eucalyptus blakelyi, AUSTRALIAN CAPITAL TERRITORY: 35°18'S 149°08'E Canberra (site 15a), 3.ii.l954, 2 9 9, A. Magassy. NEW SOUTH WALES: 35°50'S 147°15'E Woomargana, 3.xii.l954, 9, E. Lewis (all in ANIC). ex Creiis sp., WESTERN AUSTRALIA: 31°59'S 115°52'E Western Australian Insti- tute of Technology, South Perth, 1982, 9, S.J. Curry (ANIC). ex Aphidius colemani Viereck / Br achy - caudus helichrysi (Kaltenbach) on safflow- er, NEW SOUTH WALES: 29°14'S 149°51'E 25km N. of Moree, 12.ix.l993, 3 9 9, Id, J. Stanley (ANIC). ex Aphidius colemani / Myzus persicae (Sulzer) on nectarine, AUSTRALIAN CAPITAL TERRITORY: 35°15'S 149°05'E Aranda, Canberra, 15.xi.l991, 29 9, P.J. Hart (ANIC). ex Trioxys complanatus Quilis / Therioa- phis trifolii (Monell) f. maculata on lucerne, VICTORIA: 36°25'S 145°56'E Goorambat, 1979, 9, P. Ridland; 36°33'S 145°59'E Ben- alla, 9.ii.l978, 9, 26 6, L.T. Woolcock; SOUTH AUSTRALIA: 34°40'S 138°34'E Virginia, 21.ii.l980, 6, C. Wilson; 34°56'S 138°36'E Northfield, Adelaide, 20.iii.l978, 8.ii.l979, 49 9, 36 6, L.T. Woolcock; 35°41'S 139°20'E Meningie, 27.ii.1979, 6, C. Wilson (all in ANIC). ex Aphelinus mali (Haldeman) / Eriosoma lanigerum (Hausmann) on apple, NEW SOUTH WALES: 30°31'S 151°40'E Armi- dale, December 1992, 6 9 9, S. Asante (ANIC). ex Pseudococcus sp., on Acacia melanoxy- lon, VICTORIA: 37°42'S 145°04'E LaTrobe University, Bundoora, Melbourne, l.xii.- 1980, 29 9,26 6, G. Farrell (BMNH). Other records.— QUEENSLAND: 15°28'S 145°15'E Cooktown, in forest, 2.iii.l914, 9, A.P. Dodd (Girault 1915a; Dahms 1984). WESTERN AUSTRALIA: 27°41'S 114°- 33 'E Kalbarri National Park, December 1986, 9, J.S. Noyes; 31°36'S 116°13'E Avon Valley National Park, 50km NE of Perth, 25.xii.1986, 9, J.S. Noyes; 31°43'S 116°04'E Walyunga National Park, 40km NE of Perth, 24.xii.1986, 6, J.S. Noyes (BMNH) (Identified by L.D. Coote). ex Ceroplastes rubens Maskell on Schef- Volume 4, 1995 71 flera actinophi/lla, QUEENSLAND: 27°30'S Morphology: Length mesosoma + me- 153°01'E Brisbane, May 1980, 46 6, J.S. tasoma, 0.7-L3 mm long (mean, LOl mm; Noyes (BMNH) (Recorded as Paramyioc- n = 2Q; lectotype, L06 mm). Integument nema sp. by Viggiani (1988)). variously reticulate-imbricate, except that Redescription of Female. — Coloration: of most of metanotum and propodeum. Most body setae dark. Head yellow, or- which is smooth; imbrication on tibiae ange or tawny except for following: finely setose at interstices. Large setae Ion- brown, linear macula suborbitally on gitudinally ridged. gena; small, brown macula surrounding Head: 2 pairs of large, stiff, bristle-like posterior half of median ocellus and an- setae on frons near margins of eyes; 1 sub- terior half of each lateral ocellus; exten- median pair between these (lOOixm long in sive, broad, brown to black, inverted U- lectotype); 1 pair lateral to median ocellus; shaped macula in region adjacent to pro- 1 pair near margin of eyes anterolateral to thorax; mandibles apically testaceous. An- lateral ocelli; 1 pair between lateral ocelli tennae yellow, scape may be dusky. Eyes (113 ixm); 1 pair posteromedial to lateral red; margin and, internally, medial half of ocelli. Dorsalmost tooth of mandible usu- eye socket darkly sclerotized (conspicuous ally slightly blunt, only in cleared specimens). Ocelli red. Antennae inserted not far distant from Mesosoma, orange or tawny except for clypeal margin, the distance subequal to following dark sclerotization: extensive, distance between insertions; setae short, brown to black macula on prothorax ad- with stouter, stiffer hairs concentrated on jacent to occipital region of head; notauli one side; antennal formula: 1:1:1-2:3:3. anteriorly, boundaries of anterior angles Scape: length, 2.9-3.4 X width (mean, 3.2); of axillae, anterior and posterior bound- 1.7-2.3 (2) x length of pedicel; 1-1.2 (1.1) aries of pronotum, scutellum and meta- X length of club. Pedicel: length, 1.3-2.5 notum mav be brown; 1-2 small, brown (1.8) X width; 0.6-1 (0.75) X length of to black maculae near base of fore wing; anellus(i) + funicle. Anellus 1 large sometimes 1 macula at base of posterolat- enough to be considered funicular, finely eral setae of scutellum and submarginally setose but non-sensoriated; a 2nd, very on metanotum; propodeum light to dark small, anellus sometimes variouslv devei- brown anteriorly and on each lateral third oped on half of apex of anellus 1, mav be except for spiracles, lateral margins may setose. Funicular segments all wider than be darker, or whole segment may be long, becoming progressivelv wider and brown and laterally black. Ventrally and longer apicad; segment 1 (fl) length, 0.58- sometimes pleurally, mesosoma light 0.87 (0.7) X width; f2: length, 0.47-0.82 brown to dark brown. (0.74) X width; f3: length, 0.64-0.9 (0.79) Legs yellow except for following: fore X width; length of anellus (i) + funicle, femur with linear macula postaxially; mid 0.6-0.9 (0.7) x club. Club 3-segmented; coxa may be slightly dusky basally; hind segments separable by segmentally ar- coxa brown to dark brown except usually ranged setal and sensorial patterns, not bv apically, sometimes also dark brown mar- sutures; length, 1.5-2.7 (2) X width; width ginally; hind femur sometimes streakily 1-1.8 X width of f3. Lengths of segments: dusky; sockets of hind tibial bristles black, scape, 99-135 fim (mean, 117 |xm; n = 20); Metasoma light brown to black; petiole pedicel, 45-68 (58) |xm; fl, 14-24 (17.5) usually brown, often black on lateral mar- |xm; f2, 12-37 (24) |xm; f3, 24^1 (31) |jim; gins; base of gaster medially and gastral club, 93-123 (107) |xm. tergites 2-4 usually darker than rest of Mesosoma: Pronotum with row of 3+3 gaster, and fasciate, especially under high long setae (115 ixm long in lectotype) pos- magnification. terosublaterally. Mesoscutum anteriorly. 72 Journal of Hymenoptera Research between notauli, with 11-21 fine, stiff se- tae (41 |jLm) in 2-3 irregular rows ; a mid- dle transverse row of 4 long, stout setae (98 fxm), and posterior row of 2 long, stout setae (90 |xm). Scutellum with 2 long setae anterosublateral of midpoint (87 ixm) and 2 very long, stout setae posterolaterally (168 |xm). Axillae each with long, stout seta (84 |xm) anteriorly and 2-3, rarely 4, shorter, differently sized setae (26-36 ixm) near posterior angle. Mesosomal setae of smaller specimens shorter and less stout. Legs bearing long, stiff setae, mostly in rows or in differently sized pairs; other se- tae mostly disposed dorsoapically on fe- mur and longitudinally on tibia; Fore coxa (of 1 Northfield specimen): pair of long, stiff setae subapically (39 and 57 ixm long); fore femur: subapically, 1 seta ventrally (29|jLm), 1 dorsally (45 lam); fore tibia ap- pearing fusiform, calcar plumose (57 ixm), 3 small, stiff setae apically (Slfxm), and row of 7 setae (maximally 31 ixm) preaxi- ally. Mid coxa: pair of setae subapically (42 and 57 lam); mid femur: very long pair subapically (65 (xm ventrally, 78 |xm dor- sally); mid tibia: spur long (87 |xm), plu- mose, 1 smaller seta near spur (18 |xm), 1 subapically (53 lam), preaxial row (maxi- mally 32 (xm) hardly differentiated from other setal rows. Hind coxa (0.22 mm long), shorter than hind femur (0.28 mm), which is slightly shorter than hind tibia (0.32 mm) and subequal to hind tarsus (0.27 mm); hind coxa large, elongate- ovate, length 1.44-1.69 X width, bilater- ally compressed ventrally, pair of setae subapically (44 and 66 ixm); hind femur broad, approximately parallel-sided, length 2.83-3.14 X width, 1 seta subapi- cally (58 |JLm); hind tibia, basally narrow, gradually and slightly widening apicad, spur almost smooth (54 |xm long), pair of very long, stout setae subapically (68 and 89 |jLm) and 1 longitudinal row of 8 very long, stout setae (maximally 75 iJim). Mid basitarsus longer than fore or mid basitar- sus, but not conspicuously long. Fore wings: 6-13 (mean, 10) moderately long and fine, stiff setae (30^0 |xm long in 1 Northfield specimen) present in 2-3 rows below submarginal vein and proxi- mal of parastigmal spur; 2-3 parastigmal setae (70-76 |xm); costal cell in distal third with 4-7 short, stiff, marginal setae (25 fxm) anterior to cluster of fine, stiff, ventral setae; margin of discal setation irregular. Metasoma not upturned or strongly ta- pered: Petiole boomerang-shaped except for truncated ends. Variously sized, stiff setae metamerically disposed laterad and apicad on dorsum of gaster. Ovipositor sheaths 0.10-0.11 mm long, 0.23-0.33 X hind tibial length, each fur- nished with about 10 mostly subapical se- tae. Description of Male. — Similar to female. Colour pattern similar but yellower be- cause of less extensive dark sclerotization. Linear macula usually present, anterome- dially, on vertex; suborbital, linear macula of gena absent and U-shaped macula of occipital region absent. Mesosoma later- ally pale. Gaster: dark sclerotization con- fined to narrow area adjacent to petiole and to medially coalesced fascia on gastral tergites 2-4, extending medially on gastral tergite 5 and apicolaterad. Antennae (1 specimen): scape: length, 111 |xm, 3.7 X width, 2.18 X length of ped- icel, 1.19 X length of club; pedicel: length, 51|xm, 1.7 X width, 0.71 X length of anel- lus + funicle; funicular segment 1 (fl): length, 12|jLm, 0.47 X width; f2: length, 21|jLm, 0.66 X width; f3: length, 27 |am, 0.66 X width; anellus + funicle: length, 0.77 X length of club; club: length, 0.93 |jLm, 1.32 X width. Hind coxa 0.24 mm long, twice as long as wide; hind femur 0.24 mm long, 3.2 X width; hind tibia 0.3 mm long. Mesosoma + metasoma 0.9 mm long. Anterior mesoscutal setae, 33^5 juim long; mid and posterior setae, 70 and 81 ixm long respectively. Anterior and posterior scutel- lar setae 64 and 121 \xm, respectively; only 1 posterior axillar seta, 24 |jLm long. Notes. — The type material of £. flavitlior- Volume 4, 1995 73 ax comprises the slide-mounted lectotype as possessing 2 anelli, which may have designated by Hayat in 1979 (Hayat and been the reason why he considered it a Verma 1980), and paralectotype 1, with species distinct from the other species of parts on a tag and parts on the same slide Euryischomyia, and why he identified spec- as the lectotype but mounted separately, imens from 'lime scale' as 2 species, M. inbetween the lectotype and paralectotype saintpierrei and M. setosa (see above list of 2 (see list of type material examined). Par- specimens identified by Girault). These alectotype 2 is the specimen stated by specimens, on 1 slide, have antennae each Hayat and Verma (1980) to be a species with either 1 or 2 anelli but are otherwise different from the lectotype, resembling similar to one another. Most of the speci- marmorativentris, and probably not part of mens of £. flavithorax examined in this the original type material. Consequent study have both antennae with 1 anellus; upon the above statements, the specimen others have 2; or 1 antenna may bear 1 was referred to by Dahms (1984) as an un- anellus and the other bear 2; or 1 small identified chalcidoid. Paralectotype 2 is, anellus may be partly delimited from a without doubt, Euri/ischoniyia flavithorax. larger one. Smaller and yellower than the lectotype, it Distribution. — AUSTRALIA: Queens- does indeed resemble the holotype Myioc- land. New South Wales, Australian Capi- nema marmorativentris, which was collect- tal Territory, Victoria, South Australia, ed in the same locality a week or so later. Western Australia (Figure 2). The female collected by A. P. Dodd at Biology. — E. flavithorax is not highly Cooktown has still not been located (see host-specific, its recorded hosts being chal- list of other records above). It may be part cidoid and ichneumonoid Hymenoptera, of Dodd's collection. and psylloid, aphidoid and coccoid He- £. flavithorax was originally described miptera Sternorrhyncha (Table 1). E. flav- from 3 swept specimens, and each of the ithorax was reared occasionally from im- synonymized species was described from mature Aphidiinae and Aphclimis mali, en- single specimens. The relevant parts of the doparasitic in aphidine and drepanosiphi- descriptions are based mainly on colora- ne, and pemphigine aphids, respectively, tion. Other characters described by Girault as part of a long-term study by the author are supra-specific or are not interspecifi- of aphids and their insect natural enemies; cally comparable. However, all of them fit and, also occasionally, from Psyllacphagus into the above conception of £. flavithorax. sp(p). parasitic in lerp-forming psyllid £. flavithorax is variable in coloration. In nymphs, as part of an ecological study by general, smaller females are yellower and L.R. Clark in 1950-74 of psyllids associat- less darkly sclerotized than larger ones, ed with eucalypts (Riek 1962). and males are smaller and yellower than £. flavithorax, however, is not a true hy- females. The differences are a reflection of perparasite of aphids if, like certain other differences in extent of dark sclerotization. chalcidoids and proctotrupc^ids {e.g. Pachy- The pattern of coloration, however, is cc:)n- neuron spp. (Pteromalidae) and Dcndroce- stant, and is best expressed in the lighter- rus spp. (Megaspilidae) (Takada 1973a,b)), coloured specimens. The lectotype of £. it does not parasitize the hymenopterous flavithorax, for instance, belongs near the host until after the latter has consumed all large, dark end of the scale of variability, of its aphid host except for the dead, dry whereas paralectotype 2 of £. flavithorax aphid skin (mummy); i.e. if it parasitizes and the holotype M. marmorativentris be- the hymenopterous contents of a capsule long near the small, pale end. The small, it is strictly a primary parasite. Such a pale holotype £. fasciata is a male. probability is supported by its broad host Girault (1913a) described £. saintpierrei spectrum; parasitizing live hosts of such 74 Journal of Hymenoptera Research ■ y c rt -eadon, and Eiicerceris (Evans and O'Neill 1988; O'Neill 1990). In this mating system, males apply volatile secretions originating from mandibular glands to a substrate with clypeal brushes, and defend these non-resource based territories from con- specific males. Philanthine females typi- cally nest in dispersed conditions, mate once at the beginning of the nesting period away from the nest, presumably are at- tracted to mandibular gland secretions, and enter territories to copulate (Evans and O'Neill 1988). The mating system of C. binodis departs from these conditions. Females nest in aggregations, and mate multiply at the nest. Resident C. binodis males stand on or directly adjacent to nests, exclude conspecific non-resident males from nests, and do not scent mark territories. Of the eight philanthine gen- era, male nest defense has been described only in Philanthiis, and presumably is de- rived within that genus (Evans and O'Neill 1988). Male nest defense is de- scribed here for the first time in the genus Cerceris, and also presumably represents a derived condition within this taxon. METHODS An aggregation of nesting C. binodis fe- males was discovered on Barbour Lathrop trail on Barro Colorado Island, Republic of Panama on July 8, 1994. A group of five nests that were close enough together to be observed simultaneously were marked with numbered tags. Male and female wasps were individually marked with enamel paint on the thorax, and head widths were measured to the nearest 0.05 mm with Manostat calipers. The nests were observed on two days after individ- uals were marked. Observations were made from 1 355-1 525h on the first day, and from 1515-1650h on the second day. The behaviors recorded were the amount of time that males spent standing on or adjacent tc^ nests, defensive interactions between males, and copulations. 78 Journal of Hymenoptera Research RESULTS When a female returned to a nest with prey, and a resident male was present, she would land on the nest tumulus, and cop- ulate. However, a female would fly di- rectly into her nest if no male was present. This behavior is also described in the sphecid genus Oxybelus (Crabroninae) (Peckham et. al. 1973). Females were pro- visioning nests in all cases where males defended nests and mated. Not only did males defend nests while females were on provisioning trips, but on two occassions, a male was observed to defend a nest while the female was inside the nest. These males were observed to bite at the females, and attempted to pull them out of the nest with their mandibles. One of these males mated with the female when she exited the nest. Nest defense duration was determined by the length of time a resident male was either in contact with a nest, directly ad- jacent to a nest, or defending a nest from non-resident males. On day one, males YB, RB and BY, spent a total of 66.7, 21.1, and 12.2 percent of the observation period defending nests, respectively. On day two, males WRB, YB, RY, BB and RB spent a total of 67.3, 41.1, 16.8, 1.2 and 0.8 percent of the observation period defending nests, respectively. Males YB and WRB obtained 71% (5/7) of the observed copulations, and there is a positive correlation between the duration of time which males defend nests and the number of copulations that males obtain; r = 0.86. At times, multiple males were simulta- neously defending different nests in the aggregation. Male-male interactions were considered a territorial defense when a resident male flew from his nest to chase a non-resident male. These chases rarely ended in grappling on the ground; other- wise, there was no physical contact. Four- ty eight aggressive interactions were re- corded between resident and non-resident males. Resident males won significantly more (95%) interactions than non-resident males (Chi-square = 40.1, df = 2, P< 0.001). In 11 of the 48 interactions, identity and size of both males were known; resi- dent males won 7 of these 11 interactions. Mean head widths of winner and looser males were 3.83mm and 3.76mm, respec- tively. The difference in mean headwidths between winner and looser males is not statistically significant (t-test = .82, df = 20, P > 0.4). DISCUSSION Size has been demonstrated to deter- mine dominance in territorial sphecids such as Philanthus (O'Neill 1983a, b), and in many other insect mating systems (Thornhill and Alcock 1983). In C. binodis, however, territory ownership, not size, ap- parently determines dominance (one should note the small sample size (n=ll) where identity of both males was known). This phenomenon has also been demon- strated in other insect mating systems (Da- vies 1978, Eickwort and Ginsberg 1980, Wang and Greenfield 1991). Data also in- dicate that as the total time invested in nest defense per male increases, the num- ber of copulations obtained per male also increases. This is not surprising, since fe- males mate with the resident male which is defending her nest. In many Cerceris species, females nest in aggregations, but males do not defend nests from conspecific males. The require- ment which is likely responsible for the difference between the mating systems of these species and C. binodis is that C. bin- odis females retain their receptivity after the initial mating. This might maintain male territoriality, particularly if there is last male sperm precedence (Hook and Matthews 1980, Brockmann and Grafen 1989). However, males in several species of Philanthus defend territories which con- tain nesting females that do not display continuous receptivity (Gwynne 1980, Evans and O'Neill 1988). Data collected from this study do not indicate that fe- Volume 4, 1995 79 males are continuously receptive through- out the nesting period, only that they mate multiply. In addition, last male sperm pre- cedence is not demonstrated in this spe- cies, but often occurs in insects (Thornhill and Alcock 1983). CONCLUSION This is the first description of male nest defense in Cerceris. This derived mating system shows striking convergence with the mating systems of both Oxyhehis (Peckham et. al 1973, Hook and Matthews 1980), and Philanthus (Evans and O'Neill 1988) species. Moreover, C. binodis males do not scent mark territories, and their clypeal brushes are reduced. Since clypeal brushes are a synapomorphy of Philanthi- nae, and are present in most species of Cerceris (Alexander 1992), reduction of the clypeal brushes in C. binodis represents a derived condition. Evans and O'Neill (1988) state that scent marking occurs in all territorial philanthines where the mat- ing systems have been described. There- fore, this mating system represents an ex- ception to their statement (see also Alcock and Gamboa 1975). ACKNOWLEDGEMENTS Helpful comments and suggestions were provided by B.A. Alexander, W. T. Wcislo, D. Windsor, R. L. Minckley, and four anonymous reviewers. This re- search was funded by a Smithsonian Tropical Re- search Institute 10 week graduate fellowship, and by a Sigma Xi research grant. LITERATURE CITED Alcock, J. 1975a. Male mating strategies of some phi- lanthine wasps(Hymenoptera: Sphecidae). Jour- nal of the Kansas Eiitoniological Society 48: 532-545. Alcock, J. 1975b. Territorial behavior by males of Phi- lanthus multimaculatus (Hymenoptera: Sphecidae) with a review of territoriality in male Sphecidae. Animal Behavior 23: 889-895. Alcock, J. and G. Gamboa. 1975. Home ranges of male Cerceris simplex macrosticta (Hymenoptera: Sphecidae). Psyche 81: 528-533. Alcock, J., C.E. Jones and S.L. Buchmann. 1977. Male mating strategies in the bee Centris pallida Fox (Hymenoptera: Athophoridae). American Natu- ralist 111: 145-155. Alexander, B.A. 1992. A cladistic analysis of the sub- family Philanthinae (Hymenoptera: Sphecidae). Systematic Entomolog]/ 17: 91-108. Bohart, R.M. and A.S Menke. 1976. Sphecid Wasps of the World: A Generic Revision. Berkely: Uni- versity of California Press. 695 pp. Brockman, H. J. and A. Grafen. 1989. Mate conflict and male behavior in a soitary wasp. Trypoxylon (Trypargilum) polifum (Hymenoptera: Sphecidae). Animal Behavior 37: 232-255. Davies, N.B. Territorial defense in the speckled wood butterfly (Pararge aegeria): the resident always wins. Animal Behavior 26: 137-147. Eickwort, G.C. and H. S. Ginsberg. 1980. Foraging and mating behavior in Apoidea. Annual Review in Entomology 25: 421-426. Evans, H.E. 1971. Obser\'ations on the nesting be- havior of wasps of the tribe Cercerini. journal of the Kansas Entomological Society 44: 500-523. Evans, H. E. and K. M. O'Neill. 1985. Male territorial behavior in four species of the tribe Cercerini (Sphecidae: Philanthinae). journal of the New York Entomological Society 93(3): 1033-1040. Evans, H. E. and K. M. O'Neill. 1988. The Natural History and Behavior of North American Bee- wolves. Comstock Publishing Associates. Hook, A. W. and R. W. Matthews. 1980. Nesting bi- ology of Oxybelus sericeus with a discussion of nest guarding by male sphecid wasps (Hyme- noptera). Psyche 87: 21-37. Gwynne, D.T. 1980. Female defense polygyny in the bumblebee wolf Pliilanthus hicinctus (Hymenop- tera: Sphecidae). Behavioral Ecolology and Socio- btology 7: 213-225. O'Neill, K.M. 1983a. Territoriality, body size, and spacing in males of the beewolf Phihvithus basi- laris (Hymenoptera: Sphecidae). Behavior 86: 295: 321. O'Neill, K.M. 1983b. The significance of body size in territorial interactions of male beewolves (Hy- menoptera: Sphecidae, Philanthus). Animal Behav- ior 3\A0i-AU. O'Neill, K.M. 1990. Female nesting behavior and male territoriality in Aphilanthops subfrigidus Dunning (Hymenoptera: Sphecidae). Pan-Pacific Entomologist 66: 19-23. Peckham, D.J., F.E. Kurczewski and D.B. Peckham. 1973. Nesting behavior of Nearctic species of Ox- ybelus (Hymenoptera: Sphecidae). Annals of the Entomological Society of America bb: 647-661. Peckham, D. J. 1977. Reduction of miltogrammine cleptoparasitism by male Oxybelus subulatus (Hy- menoptera: Sphecidae). Annals of the Entomologi- cal Society of America 70: 823-828. Thornhill, R.'and J. Alcock. 1983. The Evolution of Insect Mating Systems. Harvard Press. Wang, G. and M.D. Greenfield. 1991. Effects of ter- ritory ownership on dominanace in the desert clicker (Orthoptera: Acrididae). Animal Behavior 42: 579-587. J. HYM. RES. Vol. 4, 1995, pp. 80-98 Revised Key to the Nearctic Species of Chrysocharis Forster (Hymenoptera: Eulophidae), Including Three New Species Christer Hansson Department of Systematic Zoology, Lund University, Helgonavagen 3, S-223 62 Lund, Sweden Abstract. — The Nearctic species of Chrysocharis are keyed, altogether 63 species. The paper in- cludes the description of three new species: waJili, cornigera, sentenaca and two new combinations from Chrysocharis to Omphale Haliday: varia Hansson, gracilicornis Hansson. The Nearctic and Neotropical species of The page number after each species in Chrysocharis were revised by Hansson the revised key refers to Hansson (1987). (1987). Sixty one (61) species were treated The descriptions, diagnoses, distribution and 18 of the species were described as and hosts of Nearctic species of Chrysochar- new. From the Nearctic region 57 species is are found in Hansson (1985, 1986, 1987). were recognized. However, the kev for the identificatton of the species is difficult and RECOGNIZING CHRYSOCHARIS laborious to use, mainly because too many To facilitate the recognition of Chryso- characters are used under each item, and charis, following should be helpful. Anten- characters used in the two alternatives un- na with apical two segments fused, or der each item are not always exclusively with all flagellar segments free (exception: the same. To improve the key I have re- imhrasiis with three apical segments fused, arranged it so that the most obvious char- recognized as a Chrysocharis through the acters are used first, and under each item long postmarginal vein — 1.5X stigmal only 1-2 characters are used — only occa- vein — and antennal scrobes join below sionally are more than two characters frontal suture in female). Clypeus not de- used. The characters selected are those limited (exception: flaviclypeus, recognized easy to see and that vary as little as pos- as a Chrysocharis through female antennal sible. In some cases a species varies in scrobes that join below frontal suture), characters used, and this species is found Postmarginal vein longer than stigmal under both alternatives. vein (exception zvahli Sacornigera, recog- In the revised key I have also included nized as members of Chrysocharis through changes that has taken place since the the possession of the three abovemen- publication of the previous key: Entedon tioned characters). Without a sulcus sur- imbrasus Walker has been transferred to rounding ocellar triangle (this separates Chrysocharis from Neochrysocharis Kurdju- Chrysocharis from Derostemis). mov (Hansson in press); the species treat- ed as Zaommomyia Girault by Hansson (1986) have been transferred to Chryso- charis (Schauff 1991). The three species de- The terms used in the key and in the scribed as new in this paper are also in- text are indicated on Figs 1-8. eluded. Furthermore, two species regard- Abbreviations of collections: CNC= Ca- ed as Chrysocharis by Hansson (1987) are nadian National Collection of Insects and hereby transferred to Omphale Haliday: Arachnids, Ottawa; LUZM= Lund Uni- gracilicornis Hansson, varia Hansson, both versify Zoological Museum; TAMU= Tex- n. comb. as A&M University, College Station. TERMINOLOGY AND ABBREVIATIONS Volume 4, 1995 81 KEY TO NEARCTIC SPECIES OF CHRYSOCHARIS 1. Flagellum with all five segments free (e.g. Figs 35, 39) 2 - Flagellum with apical two or three segments fused, i.e. the constrictions between the basal funicular segments are narrower than the constrictions between the apical seg- ments (e.g. Figs 37, 38) 19 2. Forewing with a complete row of setae on underside of costal cell (Figs 48, 49) 3 - Costal cell without row of setae 4 3. Forewing speculum closed (Fig. 49); 5th flagellar segment (including the narrow tip) 1.3X as long as 4th segment (Fig. 35) C. chilo (Walker) (female, male) p. 30 - Forewing speculum open laterally (Fig. 48); 5th flagellar segment 0.9-l.Ox as long as 4th segment (Fig. 36) C. pilosa Delucchi (female, male unknown) p. 30 4. Pronotal collar with a transverse carina — at least on median pronotum — parts of pro- notum behind carina smooth and shiny (Fig. 87) 5 - Pronotum without transverse carina, hind margin of pronotum at most with a very narrow smooth strip (Fig. 86) 7 5. Petiole longer than wide (Fig. 62) C. acoris (Walker) (female, male) p. 64 - Petiole at most as long as wide (Figs 56, 59) 6 6. Scape and pedicel bright orange-yellow; clypeus pale yellow C. iUnstris Graham (male) p. 59 - Scape whitish or pale brown, pedicel brown; clypeus metallic bluish-green C. occidentalis (Girault) (female, male) p. 54 7. Propodeal callus with 2 setae 8 - Propodeal callus with at least 3 setae 14 8. Petiole pale; male flagellar segments with a single whorl of setae at the base of each segment (Fig. 42) C. beckeri Yoshimoto (female, male) - Petiole dark; male flagellar segments also with setae in middle and at apex 9 9. Propodeum with 2 complete and parallel submedian grooves (Fig. 84) C. sulcata (Hansson) (female, male) - Propodeum without such grooves 10 10. Forewing speculum open below (Fig. 47) C. vonones (Walker) (female, male) - Forewing speculum closed below 11 11. Scape bright orange-yellow; reticulation on thoracic dorsum fine and engraved C. fulviscapus Hansson (male, female unknown) p. 45 - Scape whitish or brown; reticulation on thoracic dorsum raised and strong 12 12. Occipital margin with a complete (reaching from eye to eye), high and sharp carina (Fig. 91); frontal suture raised C. liriomyzae Delucchi (male) p. 26 - Occipital margin without a complete carina; frontal suture not raised 13 13. All femora predominantly dark C. phijtomyzivora Hansson (male) p. 20 - Femora pale C. cerodonthae Hansson (male) p. 21 14. Petiole longer than length of median propodeum (Fig. 85) . . C. viridis (Nees) (male) p. 29 - Petiole at most as long as length of median propodeum 15 15. Postmarginal vein 3.0-3.5 x as long as stigmal vein 16 - Postmarginal vein 2.0-2.5X as long as stigmal vein 17 16. Petiole about as long as median propodeum; dorsellum usually excavated and divided in two parts by a median carina and hind part with a median incision (Fig. 69) C. entedonoides (Walker) (female, male) p. 28 - Petiole shorter than median propodeum; dorsellum convex to flat, without median in- cision and carina (Fig. 63) C. amyite (Walker) (female, male) p. 28 17. Hind coxa conspicuously long and slender, about 2.5 X as long as wide (Fig. 52); petiolar foramen triangular C. longicoxa Hansson (male) p. 26 - Hind coxa stouter, about 1.5X as long as wide; petiolar foramen semicircular to qua- drangular 18 82 Journal of Hymenoptera Research Figs. 1-8. Terminology. 1, beckeri, female: cc= costal cell; dm= dorsellum; m= marginal vein; ml= midlobe; msc= mesoscutum; om= occipital margin; OOL= ocell-ocular line; pc= propodeal callus; pet= petiolus; pm = postmarginal vein; POL= postocellar line; POO= distance between hind ocelli and occipital margin; prm= Volume 4, 1995 g3 18. Petiole as long as wide (as in Fig. 77); femora usually pale — in a few specimens pre- dominantly dark C. avia Hansson (female, male) p. 28 - Petiole transverse (Fig. 80); femora always predominantly dark C. phytomyzivora Hansson (male) p. 20 19. Flagellum with apical three segments fused (Fig. 41) C. imbrasus (Walker) (female, male unknown) - Flagellum with apical two segments fused 20 20. Forewing with a complete row of setae on underside of costal cell (as in Figs 48, 49) . . C robusta Yoshimoto (female, male) p. 62 - Costal cell without row of setae 21 21. Pronotal collar with transverse carina— at least on median pronotum (Fig. 87) 22 - Pronotum without transverse carina (Fig. 86) 55 22. Flagellum yellow; frontal suture missing (Fig. 26) C. lualleyi Yoshimoto (male) p. 60 - Flagellum brown; frontal suture present 23 23. Petiole at least 1.5x as long as wide 24 - Petiole at most as long as wide 26 24. Malar space very narrow, 1/15 the width of mouth opening (Fig. 92); frontal suture smoothly curved; with a procession between antennal toruli C. prodice (Walker) (female, male) p. 65 - Malar space not as narrow, 1/7 the width of mouth opening (Fig. 93); frontal suture more straight; without procession between toruli 25 25. Occipital margin with a low sharp carina; propodeal callus with 2 (3) setae; female frons above fork usually with rather strong reticulation; male frons smooth or with weak reticulation C. acoris (Walker) (female, male) p. 64 - Occipital margin rounded and smooth (in a few cases with a very weak carina); pro- podeal callus with 4-5 (3) setae; female frons above fork smooth or with very weak reticulation (male unknown) C. amasis (Walker) (female) p. 65 26. Propodeum with a strong and complete median carina (Fig. 58) C. walleyi Yoshimoto (female) p. 60 - Propodeum without a complete median carina 27 27. Posteromedian part of propodeum with two submedian, slightly curved carinae (Fig. 70); male pedicel bright orange-yellow C. gemma (Walker) (female, male) p. 59 - Posteromedian propodeum without carinae, or with 2-6 short and straight carinae (e.g. Fig. 73); male pedicel brown or whitish 28 28. Anteromedian part of propodeum with a wide and strong median carina — reaching half the length of propodeum and then dividing into 2 weaker carinae which diverge towards the hind edge of propodeum (Fig. 73) C. ilhistris Graham (female) p. 59 - Median carina on anteromedian part of propodeum weaker or missing 29 29. Clypeus partly or completely pale yellow 30 - Clypeus completely dark and metallic 31 30. Reticulation on thoracic dorsum very dense and strong, almost like punctulation; pro- podeal callus with 5 setae C. flaviclypeiis Hansson (female, male unknown) p. 63 propodeum; prn= pronotum; s= stigmal vein; scu= scutellum; sl= sidelobe; spe= speculum; sul= spiracular sulcus. 2, coUaris, female: pcl= pronotal collar. 3, Antenna, coinpresiiconiis, female: an= anelli; cla= clava; fla = flagellum; fun= funiculus; MPS= multiporous plate sensilla; pcl= pedicel; sca= scape. 4, Head, side view, phytoniyzh'om, female. 5, Head, front view, clarknc male: cly= clypeus; fs= frontal suture; he= height of eye; mo= width of mouth opening; ms= width of malar space; scr= antennal scrobes; tor= antennal toruli. 6, Mesothorax, side view, longicoxa, female: b= borderline between upper and lower mesepimeron. 7, Meso+metathorax, side view, tristis, female: b= as in Fig. 6; C2= mid coxa; C3= hind coxa; pc= propodeal callus. 8, iiicdiivia, female: pf= petiolar foramen. 84 Journal of Hymenoptera Research 10 11 Figs. 9-15. Chrysocharis n.spp. 9-12. cornigera. 9, Head, side view, female. 10, Head, front view, female. 11, Head, dorsal view, female. 12, Antenna, male. 13-15. wahli. 13, Head, side view, female. 14, Head, front view, female. 15, Antenna, male. Volume 4, 1995 85 Figs. 16-26. Head, front view. 16-17, ainslici, female. 18, Ditto, male. 19, oaciiiidi^, female. 20, Ditto, male. 21, chromatomyiae, female. 22, Ditto, male. 23, perditor, female. 24, Ditto, male. 25, amasis, female. 26, wallcyi, male. Figs. 27-40. 27, Head, front view, tristis, female. 28, Ditto male. 29, Head, lateral view, coptodiscae, male. 30, Ditto, pallidigaster, female. 31, Ditto, submiitica, male. 32, Head, dorsal view, occidentalis, female. 33, Ditto, coptodiscae, female. 34-40. Antennae. 34, occidentalis, female. 35, chilo, female. 36, Apical 2 flagellar segments, pilosa, female. 37, coptodiscae, female. 38, liriomyzae, female. 39, Ditto, male. 40, Flagellum, compressicornis, male. Figs. 41-56. 41-45. Antennae. 41, /»/^ras/- supposed species were found in close rorator that the senior author has seen in proximity to each other, i.e., north central that the two spines at the apex of meta- China and Mongolia, further supporting somal tergum 3 are not as sharp as those their conspecific identity. In their recent of S. irrorator. The variation in this char- revision of the Chinese species. He et al. acter in S. irrorator does not appear to be did not consider S. mongoliciis. It is the se- great, but series of specimens from the nior author's opinion that S. mongoliciis it- east Palaearctic must be checked. These self is likely synonymous with S. irrorator two species are therefore provisionally ac- because they apparently differ only in col- cepted. ^Q2 Journal of Hymenoptera Research PHYLOGENETIC CONSIDERATIONS is another character that was not included in the Achterberg and Austin data set but No hypothesis of the phylogenetic re- ^^^^^^ .^ ^^^ ^^^^^ ^^ ^^^ cladogram a lationships of Sigalphus species has been posteriori. An assessment of the distribu- published. Achterberg and Austin (1992) ^.^^ ^^ ^j^.^ character state shows that the analyzed the genera of the Sigalphinae. ^^^^^i^^j.^ of the non-Ethiopian species Despite this valiant first attempt at ap- ^^ Sigalphus are relatively more costate proximating relationships among the gen- ^^^^ ^^^^^ ^f the Ethiopian species which era, the monophyly of Sigalphus is ques- ^^^ mostly rugose, as in species of Min- tionable due to the inclusion of the Ethi- ^^^^ ^^^ Acampsis. Again, this character opian species (S. fiilvus, S. testaceus, and S. ^^^^^ ^^es not support the monophyly of neavei). Achterberg and Austin (1992) in sigalphus. their cladogram of sigalphine genera cite 3 -p^j-gt tergite with complete and 4 synapomorphies for the genus; they are g^^ong dorsal carina." Although this char- all suspect. These will be treated here in ^^^^^ ^^^^^ -g included in the data matrix, the order they appear on the cladogram. ^^^ cladogram does not present the most 1. "Third tergite with lobes or teeth parsimonious distribution. The data ma- ventro-posteriorly." This character state is ^^^^ j^g^g ^j^g following taxa as possessing not possessed by any of the three Ethio- ^^^ derived character state: Sigalphus, Min- pian species, conversely it is found in all ^^^^^ Acampsis, and Afrocampsis. This dis- members of Aposigalphus and Notosigal- tribution is correct, but on the cladogram phus in a condition almost identical to that Achterberg and Austin indicated that the of S. romeroi (Fig. 2a). Although the char- character state is an autapomorphy for Si- acter state appears on the cladogram of galphus and ignore the rest of the character Achterberg and Austin it is not included g^^te distribution. There are several parsi- in their data set. Species of Minanga also monious reconstructions of the character have teeth ventro-posteriorly on the third \y^^^ ^n indicate that the presence of strong tergum but since these do not emanate dorsal carina on the first tergite is plesiom- from the margin of the tergum as they do orphic with reference to Sigalphus. in the aforementioned groups their ho- 4 "Face with shallow semicircular de- mology is doubtful. The character state pression." This depression apparently re- has three possible origins given the phy- f^j-g to the two longitudinal depressions logeny of Achterberg and Austin, viz. a) a t^gt define the raised medial area of the synapomorphy for Sigalphus + (Aposigal- f^^e. Some specimens of S. irrorator appear phus + Notosigalphus) with a loss in the ^ have these depressions somewhat ex- Ethiopian species of Sigalphus, b) a syna- panded laterally but in other species of Si- pomorphy for the non-Ethiopian species galphus there is no discernable difference of Sigalphus and a synapomorphy (conver- between these depressions and those of gence) for Aposigalphus + Notosigalphus, or most other genera of Sigalphinae except c) a synapomorphy for Minanga + (Sigal- Afrocampsis which does not have them. phus + (Aposigalphus + Notosigalphus)) in summary, there is no evidence for the with a loss in the Ethiopian species of Si- monophyly of Sigalphus as presently de- galphus. This last hypothesis assumes that fined. That all four of the putative syna- the condition in Minanga and the other pomorphies proposed by Achterberg and genera is homologous. None of these char- Austin are incorrect suggests that their en- acter state distributions provides support tire analysis is suspect. We do not wish to for the concept of Sigalphus including the address generic concepts in this paper, Ethiopian species. only to draw attention to the fact that care 2. "Strongly costate metapleuron." This must be taken in interpreting their re- Volume 4, 1995 103 b Fig. 2. a, posterior aspect of metasoma of Sigalphus nmieroi showing ridges on ventral apex (d = denticles); b, dorsal aspect of metasomal tergum 2 of Sigalpliiis masoni; c, posteroventral aspect of metasoma of Sigalpluis nmsoiii showing denticles at ventral apex (d = denticles, o = ovipositor, v.f. = ventral foramen); d, dorsal aspect of metasoma of Sigalphus romcwi. 104 Journal of Hymenoptera Research ^V««Kvv,5*^<^VV*S^'^''"''^'' ,,<^x>'/''- .^yyy-^y-^yyy^ ,^^- m-cu U CU-a ^^^^^^VVVXVVXVSVVV,V.VNV .v>S-^" .vV-V---^ ^v^*- .v=N^^ *»»-^\\V«^^CSVV»^XVVNXVV^^ JJ.'tN"" "^"^^^--V-..ssvv.vs^-^-^ Fig. 3. a, fore wing of Sigalphiis romeroi; h, hind wing of Sigalphus romeroi; c, fore wing of Sigalphus inasoni. suits. A phylogenetic analysis of the spe- paper. The non— Ethiopian members of Si- des of Sigalphus must be preceded by a galpJius may be monophyletic, based on reexamination of the relationships of the the presence of denticles at the ventropos- genera and this is beyond the scope of this terior margin of tergum 3. KEY TO WORLD SPECIES OF SIGALPHUS 1. New World species 2 1.' Old World species 3 2. Metasoma orange; posterior flange of ventral foramen of metasoma smooth apically, ridg- es reduced to small crenulae not reaching apex of flange (south eastern Canada and east- ern U. S. A. south to Florida) S. bicolor (Cresson) Volume 4, 1995 105 2'. Metasoma black; posterior flange of ventral foramen of metasoma with ridges extending to apex (Fig. 2a) (known only from the type locality in northwestern Costa Rica, but the two species of noctuid hosts occur throughout the Mesoamerican Pacific dry deciduous forest and this parasitoid may be found there as well) S. romeroi n. sp. 3. Head and mesosoma entirely black 4 3.' Head and mesosoma not enhrely black, rather with pale coloration 9 4. Apex of ventral foramen of metasoma lacking denticles and two rounded lobes (southern Africa) S. neavei (Turner) 4'. Apex of ventral foramen of metasoma with two denticles or two rounded lobes (Fig. 2c) (Oriental, Palaearctic) 5 5. Apex of ventral foramen of metasoma with two sharp denticles 6 5'. Apex of ventral foramen of metasoma with two rounded lobes 7 6. Metasoma entirely black; stigma black (Palaearctic) S. irrorator (Fabricius) 6'. Metasoma orange and black; stigma yellow (Mongolia and northern China) S. tuougolicus Tobias 7. Tergum 2 with two parallel median longitudinal carinae with a median longitudmal groove between them 8 7'. Tergum 2 without two parallel median longitudinal carinae and without a median lon- gitudinal groove S. masoni n. sp. 8. Fore wing entirely melanic-infuscate; tergum 3 as wide as tergum 2 S. gyrodontus He and Chen 8'. Fore wing melanic-infuscate in distal half only; tergum 3 wider than tergum 2 S. rtifiabdominalis He and Chen 9. Vertex of head and apex of hind tibia black (southern Africa) S. fiilviis Brues. 9'. Head and hind tibia entirely reddish brown (Madagascar) S. testaceus Granger. Sigalphus romeroi Sharkey n. sp. (Figs 2a, d, 3a, b) MALE. Length. — 8.4 mm. Color. — Entirely black except fore tarsus brown, wings evenly infuscate. Head. — Antenna with 42-44 flagello- meres. Mesosoma. — Pronotum smooth except for weak crenulae along posterior border; subpronope present; notaulus deeply im- pressed and with weak punctures; scutel- lar furrow (sulcus) smooth except for me- dian longitudinal carina; posterior scutel- lar depression composed of two large pits; median depression of metanotum semicir- cular (rounded posteriorly); propodeum entirely areolate, with pair of strong lon- gitudinal carinae defining medial area; sternaulus complete to epicnemial carina; fore wing (Fig. 3a) with veins M and m- cu not distinctly converging anteriorly; crossvein a of fore wing not present; sec- ondary crossveins r and m-cu of hind wing present though weak and not tubu- lar (Fig. 3b). Metasoma (Figs 2a, d). — All terga rugose; first tergum with two precurrent longitu- dinal carinae; second tergum with median longitudinal carina, and without deep an- terolateral depressions (as in S. bicolor); tergum 3 lacking long dense setae; ventral margins of carapace crenulate; apex of ventral metasomal foramen with 6 riciges forming weak, sharp, denticles (Fig. 2a); ventral foramen of carapace not extending to apex of metasoma, i.e., tergum 3 curved around apex of metasoma forming poste- rior surface. Material Examined. — Holotype male: COSTA RICA, Guanacaste Province, Guanacaste Conservation Area, Sector Santa Rosa; 300 m. Janzen rearing data base voucher number 90-SRNP-828. Lar- val host Epithisanotia saiictijolunuia^i (Noc- tuidae) collected 24 June 1990, wasp 106 Journal of Hymenoptera Research eclosed 26 May 1991. Institute Nacional de Biodiversidad (INBio), Santo Domingo de Heredia, Costa Rica. Paratype male: COSTA RICA, Guana- caste Province, Guanacaste Conservation Area, Sector Santa Rosa, 300m. Janzen rearing data base voucher number 92- SRNP-2777. Larval host Erochia miimmia (species complex) (Noctuidae) collected 7 July 1992, wasp enclosed 10 May 1993. Ca- nadian National Collection, Ottawa, Can- ada. Remarks. — The female of this species is unknown. S. romeroi is easily distin- guished from the only other New World species, S. bicolor by the characters given in the key. Eti/mologi/. — The species is named in honor of Sra. Luz Maria Romero in rec- ognition of her outstanding drive and in- spiration in guiding and developing the Biological Education Program of the Guanacaste Conservation Area in north- western Costa Rica. Biogeographical Notes. — This is the first record of members of the genus Signlphiis in the Neotropics and only the second spe- cies to be discovered in the New World. The genus is unknown in South America. As suggested in the section on phyloge- ny, the non- Ethiopian members of Sigal- phus may be monophyletic based on the presence of denticles at the ventroposterior margin of tergum 3. If this is so, the distri- bution of the New World species is inter- esting from a biogeographic perspective. It is generally agreed that the closest phylogenetic relationships of most tem- perate North American taxa are found in the northern hemisphere of the Old World (Lavin and Luckow 1993). In contrast the usual hypothesis put forward to explain North American tropical fauna is dispers- al from South America. If the non-Ethiopian members of Signl- phus are monophyletic, it suggests a north- ern hemisphere origin of S. romeroi, the only Neotropical member of the genus. This distribution is consistent with what has been referred to as the boreotropics hypothesis, which suggests a historically more widespread tropical northern hemi- sphere biota. (Wolfe 1975; Tiffney 1985a, b; Lavin and Luckow 1993). Biology. — The larval hosts, Epithisanotia snnctijohannasi and Erochia miimmia (both: Noctuidae, Agaristinae), are common to exceedingly rare (depending on the year) and univoltine in the primary to early suc- cessional dry forests in eastern Sector Santa Rosa of the Guanacaste Conserva- tion Area in northwestern Costa Rica (200-300m elevation; about 35 km north of the town of Liberia, to the west of the In- teramerican Highway) (see Janzen 1987, 1988a, 1988b, 1993 for a general site de- scription of this forest and its caterpillars). E. sanctijohannasi and E. miimmia are most easily located when they are feeding on the large woody vine Tetracera vohibilis (Dilleniaceae, Dilleniales) (hundreds of re- cords), and £. sanctijohannasi has also been found feeding on Curatela americana (4 rec- ords). These larvae eat mature leaves and are non-gregarious feeders in the penulti- mate and ultimate instars (earlier instars unknown) on both host plants. Both spe- cies are univoltine (last instar caterpillars occurring in late June and early July, the second and third months of the rainy sea- son) and the pupae remain dormant through the remainder of the six month rainy season and all of the dry season, with the moths eclosing in May at the be- ginning of the rains (see Janzen 1987, 1988a, 1993 for other examples of the same phenomenon). Of 36 £. sanctijohannasi 31 £. miimmia last instar larvae collected and reared from T. voliibilis in 1992 and 1993, 84 and 93 percent respectively produced last instar larvae of Sigalphus romeroi. This extremely high percent of parasitization is not rep- resentative of caterpillar-parasitoid inter- actions in this forest, where 1-30% para- sitization by a given species of parasitoid is usually recorded (D. H. Janzen, unpub- lished rearing records). The very large Volume 4, 1995 107 number of Sigalphiis romeroi larvae did not generate a large paratype series because, as explained below, the larvae proved to be very incompetent at spinning cocoons and surviving subsequently under artifi- cial rearing conditions. The parasitized last instar caterpillars of £. sanctijohannnsi and E. uiiimmia appear to be perfectly normal in feeding, fleeing and walking behaviour. The prepupal caterpil- lar searches in the litter for a piece of rot- ting wood, tunnels into the wood, and makes its pupal chamber there. Within 1- 3 days of the caterpillar entering the wood the wasp larva eats nearly all of the con- tents of the prepupal caterpillar and exits the cadaver. The wasp larva is so large that there does not appear to be sufficient food for more than one per caterpillar of either host. The wasp larva then spins an ovoid white cocoon in the caterpillar's pu- pal chamber (late June to early July), where it remains dormant as a prepupal larva until the following May, when it pu- pates and exits through a round hole cut in the cocoon, and then burrows out through the wood chip-filled entrance tunnel made initially by the caterpillar. The biology described in the previous paragraphs is based on observation of all of this behaviour with captive caterpillars individually confined in large plastic bags and bottles in an open air laboratory a few meters from the forest where the caterpil- lars and wasps are found naturally. Dor- mancy of moth pupae and of wasp larvae in their cocoons occurred in the individ- uals confined under three conditions of captivity, i.e., under moist conditions throughout the wet season with a change to dry conditions during the dry season, entirely under dry conditions throught^ut the wet and dry seasons, and for those captured in the dry season they remained in a dry environment in captivity. Moths in the pupal stage and wasp larvae both use some environmental cue that is per- ceivable through the walls of a plastic bag to initiate further development in late April or early May, and this cue is most likely the abrupt drop in temperature that accompanies the first rains (see Janzen 1993). By eclosing at the same time as the moths, the adult wasps are presumably present at exactly the same time as are the first instar larvae of the next generation of caterpillars which, extrapolating from the known biology of S. bicolor, is the stage most likely to be attacked. The larva of Sigalphus romeroi is excep- tional, among the species of parasitoid wasps that the junior author has reared in this forest, in its inability to spin a cocoon except in the pupal chamber of its host. It was usually not able to spin a cocoon be- tween leaves in the litter, among crum- pled layers of dry or moist toilet tissue, or in tubular chambers of toilet tissue. Of ten larvae that did succeed in spinning co- coons among toilet tissue layers, all died of unknown causes over the following seven months. Malaise traps run for the equivalent of tens of years in this dry forest between 1984 and 1988 (e.g., Gauld 1988, Gaston and Gauld 1993) have captured no speci- mens of S. romeroi (I. D. Gauld and P. A. Mitchell, personal communication). This result is particularly striking given that the Malaise traps were located within sev- eral meters of the host plants and in all ages of forest succession. In contrast to S. romeroi being large, sol- itary, and univoltine, Signlplnis bicolor, the only North American member of the ge- nus, is small, gregarious and multivoltine. S. bicolor oviposits in the first instar larvae of a noctuid caterpillar, Acronicta clarescens (= Apatela clnrescetis), and the larv^ae come out of the prepupal caterpillar and spin their cocoons within the host cocoon (Cushman 1913, who described the wasp's natural history under the name Sphaeropyx bicolor). The only western Palaearctic spe- cies of the genus, S. irrorator has also been reared from Acronicta in Europe, but be- cause of its large size (about the same as S. romeroi) it is inferred to be solitary (S. 108 Journal of Hymenoptera Research Ward, personal communication). You, Zhou and Tong (1991) report S. irrorator (as S. anomis) as a parasitoid of Anomis flava (Noctuidae). The biology of all other spe- cies of Sigalphus is unknown. Sigalphus masoni Sharkey n. sp. (Figs. 2b, c, 3c) HOLOTYPE FEMALE. Length. — 10.9 mm. Color. — Mostly black with yellow on bases and ventral sides of all tibiae and on metasomal sterna, and orange on meta- somal terga 1, 2, and anterior % of tergum 3; fore wing entirely infuscate though somewhat paler basally; hind wing clear in basal Vs, infuscate in distal %. Head. — Antenna broken. Mesosoma. — Pronotum weakly rugose; subpronope present; notaulus deeply im- pressed and crenulate; scutellar furrow (sulcus) smooth except for three longitu- dinal carinae; posterior scutellar depres- sion composed of many small pits; medi- an depression of metanotum sharply tri- angular (acute posteriorly); propodeum weakly rugose laterally with pair of lon- gitudinal carinae defining medial area; sternaulus complete to epicnemial carina; fore wing with veins M and m-cu distinct- ly converging anteriorly; crossvein a of fore wing present and distinct; hind wing with secondary crossvein r present and secondary crossvein m-cu absent. Metasomn (Figs 2b, c). — All terga rugose; first tergum with two precurrent longitu- dinal carinae; second tergum with median longitudinal carina and an equally strong pair of lateral longitudinal carinae that converge towards the mid-line posterior- ly; second tergum with moderately deep anterolateral depressions (as in S. bicolor); tergum 3 with long dense setae; ventral margins of carapace crenulate; apex of ventral metasomal foramen with two rounded denticles (Fig. 2c); ventral fora- men of carapace (Fig. 2c) not extending to apex of metasoma. Material Examined. — Holotype 9: NE- PAL: Katmandu, Godavari, 5,000 ft. (1524 m.), 6. VIII. 1967, Canadian Nepal Expedi- tion, [Canadian National Collection, Otta- wa, Canada] Biology. — Unknown Remarks. — Members of S. masoni are similar to those of S. irrorator and S. mon- golicus in some respects, particularly in their general size and the long setae on metasomal tergum 3. In S. masoni, these setae are longer and denser than in the New World species of Sigalphus, though they are not as exaggerated as those of S. irrorator, nor are they gold colored. Mem- bers of the two species differ in color, those of S. masoni having the metasoma predominantly orange versus entirely black in S. irrorator. The species also differ in the shape of the denticles at the apex of the ventral foramen, i.e., those of S. masoni are wide and blunt whereas those of S. ir- rorator and S. mongolicus are relatively nar- row and acute. The most distinctive fea- ture distinguishing S. masoni from other species of Sigalphus is the sculpture of the second metasomal tergum. All other Pa- laearctic and Oriental species of Sigalphus have a median longitudinal groove bor- dered laterally by longitudinal carinae. These are not present in S. masoni (Fig. 2d). Etymology. — The species is named after Dr. W. R. M. Mason for his many contri- butions to Hymenopterology, and also, though it is not stated on the label, be- cause he is thought to be the collector of the unique specimen. ACKNOWLEDGMENTS This study was supported by NSF grant BSR 90- 24770 and DEB 94-00829 to D. H. Janzen. Junhua He is especially acknowledged for the loan of the type specimens of S. hunanus and S. anoniis. Sondra Ward and Ian Gauld are gratefully acknowledged for hav- ing expressed great excitement over the first speci- men reared. We thank Ian Gauld for having stimu- lated this dry forest parasitoid study in the first place. We thank R. W. Poole for identifying the moths. Hen- ri Goulet drew and Susan Rigby inked the habitus of S. bicolor which was first published in the "The Hy- menoptera Of The World" Goulet and Huber (1993). I Volume 4, 1995 109 Jason Hardis drew the fore wing S. masoni. All other drawings were done by Barry Flahey. We thank Da- vid Althoff, Paul Marsh, Lubomir Masner, Eric Rick- ey, Ales Smetana and two anonymous reviewers for valuable comments on early drafts. LITERATURE CITED Achterberg, C. van and A. D. Austin. 1992. Revision of the genera of the subfamily Sigalphinae (Hy- menoptera: Braconidae), including a revision of the Australian species. Zoologische Veiiuvniclingen 280:1-44. Cushman, R. A. 1913. Biological notes on a few rare or little known parasitic Hymenoptera. Proceed- ings of the Entomological Society of Washington 15: 153-161. Gaston, K. J. and I. D. Gauld 1993. How many spe- cies of pimplines (Hymenoptera: Ichneumoni- dae) are there in Costa Rica? Journal of Tropical Ecology 9:491^99. Gauld, I. D. 1988. A survey of the Ophioninae (Hy- menoptera: Ichneumonidae) of tropical Mesoam- erica with special reference to the fauna of Costa Rica. Bulletin of the British Museum (Natural His- tory), Entomology Series 57 :l-309. Goulet, H. and N.B. Huber. 1993. Hymenoptera of the world: an identification guide to families. Research Branch Agriculture Canada Publication 1894 /E. 668pp. He, J. and X. Chen. 1993. Descriptions of two species of the genus Sigalphus Latreille from China (Hy- menoptera: Braconidae: Sigalphinae). Acta Ento- mologica Sinica 34:225-229. (In Chinese with En- glish summary). He, J., Chen, X. and Ma H. 1994. Revision of Sigal- phus species from China with descriptions of two new species. Journal Zhejiang Agricultural Univer- sity 20:441-448. Janzen, D. H. 1987. How moths pass the dry season in a Costa Rican dry forest. Insect Science and its Application 8:489-500. Janzen, D. H. 1988a. Ecological characterization of a Costa Rican dry forest caterpillar fauna. Biotro- pica 20:120-135. Janzen, D. H. 1988b. Guanacaste National Park: Tropical ecological and biocultural restoration. In: J. Cairns, Jr., ed.: Rehabilitation of damaged eco- systems, Vol. II, CRC Press, Boca Raton, Florida, pp. 143-192. Janzen, D. H. 1993. Caterpillar seasonality in a Costa Rican dry forest, hi: N. E. Stamp and T. M. Casey eds.: Caterpillars. Ecological and evolutionary con- straints on foraging, Chapman and Hall, New York, pp. 448-477. Lavin, M. and M. Luckow. 1993. Origins and rela- tionships of tropical North America in the con- text of the boreotropics hypothesis. American Journal of Botany 80:1-14. Shenefelt, R.D. 1973. Braconidae 6 Cheloninae. Hy- menopterorum Catalogus (n.e.) 10:813-936. Tiffney, B. H. 1985a. Perspectives on the origin of the floristic similarity between eastern Asia and east- ern North America. Journal of the Arnold Arbore- tum 66: 73-94. Tiffney, B. H. 1985b. The Eocene North Atlantic land bridge: its importance in Tertiary and modern phytogeography of the northern hemisphere. Journal of the Arnold Arboretum 66: 243-273. Tobias, V.I. 1974. Contributions to the fauna of Bra- conidae (Hymenoptera) of Mongolia. [In Rus- sian]. The Insects Of Mongolia 2:261-274. Lenin- grad Publishing House. You, L., Zhou, Z. and X. Tong. 1991. Two new spe- cies of Sigalphus Latreille, 1802 from Hunan Province (Hymenoptera: Braconidae: Cheloni- nae). Acta Entomologica Sinca 34:225-229. You, L. and Z. Zhou. 1994. Revision of the genus Sigalphus Latreille, 1802 of the subfamily Sigal- phinae (Hymenoptera: Braconidae) in China. Journal of Hunan Agricultural College 20:363-365. Wolfe, J. A. 1975. Some aspects of plant geography of the northern hemisphere during the late Cre- taceous and Tertiary. Annals of the Missouri Bo- tanical Garden 62: 264-279. J. HYM. RES. Vol. 4, 1995, pp. 110-120 Ovipositor Steering Mechanisms in Braconid Wasps Donald L. J. Quicke, Mike Fitton, Jane Harris (DLJQ) Department of Biology, Imperial College at Silwood Park, Ascot, Berks SL5 7PY, U.K.; (MF, DLJQ) Department of Entomology, The Natural History Museum, London SW7 5BD, U.K.; (JH) Department of Animal and Plant Sciences, PO Box 601, University of Sheffield, Sheffield SIO 2UQ, U.K. Abstract. — Ovipositor features which allow the ovipositor tip to be manipulated and which have restricted distributions among the subfamilies of Braconidae are described and illustrated for the first time. Members of the Amicrocentrinae, Charmontinae, Helconinae and Macrocentrinae have the rhachis (the tongue-like part of the mechanism that interlocks the upper valve with the lower ones) with an increased density of scale-like sculpture and often also swollen pre-apically. This modification leads to increased friction and restricted movement between the upper and lower valves and is also associated with an ovipositor bending mechanism in which retraction of the lower ovipositor valves relative to the upper one causes the ovipositor tip to bend ventrally. Many members of the subfamilies Agathidinae and Orgilinae possess a pair of pre-apical boss-like pro- jections on the upper ovipositor valve (gonapophyses 9) and a corresponding structure on each lower valve (gonapophyses 8) that together enable the wasps to bend and manipulate the ovi- positor apex by retraction of the lower ovipositor valves relative to the upper one. Most members of the Doryctinae have the aulax (the groove-like part of the mechanism that interlocks the lower valve with the upper one) constricted opposite modified pre-apical, dorso-lateral tooth-like struc- tures. These constrictions restrict the relative movement between the upper and lower valves and thus operate as an ovipositor bending mechanism. The potential phylogenetic significance of these mechanisms is discussed, and preliminary parsimony analyses are described which suggest that the pre-apical boss-like projections of the Agathidinae and Orgilinae may represent a synapo- morphy uniting these two subfamilies. INTRODUCTION Oviposition behaviour and, as a conse- quence, the ovipositor have been of im- mense importance in the evolution of the Hymenoptera, and are central to the par- asitic way of life (Gauld & Bolton 1988). The commonly held vievv^ that the ovipos- itors of parasitic w^asps are fairly simple analogues of hypodermic needles is there- fore likely to be a considerable over-sim- plification as has in fact been shown by several previous investigations of func- tional morphology (Oeser 1961; Austin & Browning 1981; Quicke et al. 1992a,b, 1994). Several braconid genera have recently been shown to possess highly modified ovipositors whose tips are capable of be- ing independently actively manipulated by the wasp even though the ovipositor valves themselves are devoid of intrinsic musculature and their relative movements are controlled by muscles within the me- tasoma. Mechanisms, as found for exam- ple in the braconine genus Zagli/ptogastra Ashmead, were apparent because of very conspicuous features of the intact ovipos- itor (Quicke 1991; Quicke & Marsh 1992). However, an ability to manipulate or steer the ovipositor tip has been observed in a number of other parasitic Hymenoptera whose ovipositors appear more or less un- modified externally (Delanoue & Aram- bourg 1965; Compton & Nefdt 1988). Fur- ther, our own observations have also Volume 4, 1995 111 shown that an ability to manipulate the ovipositor tip is present in members of several subfamilies of Braconidae that possess moderately long ovipositors. We have therefore examined ovipositor tip structure in detail across a range of brac- onid taxa. Here we describe three different types of morphological specialisation found in the Braconidae that appear to un- derlie the manipulatory abilities observed. These modifications all utilise the same mechanical principle but they achieve this through modification of different struc- tures. In these cases, the structures in- volved can often only be revealed by ex- amination of separated ovipositor valves using scanning electron microscopy; how- ever, their existence supports the view that the ovipositors of parasitic wasps are highly specialised and complex structures. Two different ovipositor steering mech- anisms, one involving a sinuous apical re- gion of the ovipositor the other involving the development of an aulaciform rod be- tween the largely separated halves of the upper ovipositor valve have been report- ed previously (Quicke 1991; Quicke et al. 1994). The first of these is possessed by several unrelated taxa of Braconidae and Ichneumonidae, the second so far is known only from members of the Ichneu- monidae. During investigations of ovipos- itors belonging to a large number of other ichneumonid wasps collectively repre- senting most subfamilies, no steering spe- cialisations similar to any of the three de- scribed here for Braconidae have been de- tected. Functionally similar but morpho- logically different mechanisms in members of two other families of parasitic Hyme- noptera, Gasteruptiidae and Aulacidae, are being described elsewhere (Quicke & Fitton, in press). GENERAL MORPHOLOGY The ovipositors of Hymenoptera are comprised of components that are be- lieved to be derived primitively from two pairs of metasomal appendages (Smith 1968, 1970). The two gonapophyses of ab- dominal segment 9 are usually fused along their length and form the single up- per valve, while those of segment 8 re- main separate and form the two lower valves. The upper valve is interlocked with each lower valve by a k:)ngitudinal tongue and groove joint referred to as the olistheter. The tongue or rhachis situated ventro-laterally on each side of the upper valve is T'-shaped in transverse section and in life runs within the T'-shaped groove or aulax on the dorso-lateral face of the lower valve (Fig. 1). MATERIALS AND METHODS Ovipositors of 70 species of Braconidae, collectively representing members of 19 of the 40-or-so currently recognised subfam- ilies (Quicke & van Achterberg 1990; Shaw & Huddleston 1991) have been examined using scanning electron microscopy. For most of these, the upper and lower valves were carefully separated after soaking in- tact wasp metasomas or excised oviposi- tors in either water or dilute aqueous mild detergent. In the case of a few specimens in which one or both lower ovipositor valves were protruding beyond the apex of the dorsal valve, the intact ovipositor was examined. Limited observ^ations of ovipositor ma- nipulation were made on some hand-held living female wasps. In addition, experi- ments were carried out using excised ovi- positors from freshly killed wasps. Ovi- positors were submerged under water and the effects observed of pulling and push- ing the cut ends of the basal portions of the upper and lower valves differentially with fine forceps. The taxa investigated are listed below according to subfamily. Where more than one individual of a species was examined the number is given in parentheses. Agathidinae: Braunsia sp., Cameroun (3); Cnmptothlipsis sp., Kenya; Mcgngnthis sp.. Sierra Leone; Microdiis rufipcs Nees, UK. Alysiinae: Phacnocarpa conspurcator 112 Journal of Hymenoptera Research 2a. Macrocentrus type o < < 1. Stylized TS ovipositor 2b. Agathidine type upper valve aul lower valve c ■o •D n> < < (D O S < < 2c. Doryctine type upper valve retracted 2d. Doryctine type upper valve protracted lower valve upper valve lower valve rh aul Figs. 1,2. 1. Diagramatic transverse section (TS) through mid-region of braconid ovipositor to show olistheter mechanism: (ol) olistheter, (rh) rhachis, (aul) aulax. 2. Diagramatic representations of how the three ovipositor bending mechanisms are believed to operate, a, Macrocentrinae, Amicrocentrinae, and Charmontinae, lateral views showing preapical swelling of rhachis; b, Agathidinae and Orgilinae, lateral views showing pre-apical bosses; c, d, Doryctinae, dorsal showing in the postulated increased grip against the rhachis when the upper valve is protracted. (Haliday), Netherlands. Amicrocentrinae: Amicrocentrum sp., Tanzania. Blacinae: Ar- tiocus spinarius Achterberg, Brazil (2); Bla- cus hastatus Haliday, Sweden; B. longipen- nis (Gravenhorst), UK; B. maculipes Wes- mael, Austria; B. nigricornis Haeselbarth, UK; B. paganus Haliday, UK; B. pallipes Haliday, UK; B. ruficornis Nees, Hungary, Russia (2); Dyscoletes lancifer (Haliday), UK, Germany (2). Braconinae: Digonogas- tra sp., Trinidad; Eiiurohracon yokahamae Dalla Torre, Japan; Glyptomorpha sp., Gam- bia; Nesaiilax sp., Papua New Guinea; Ode- sia sp., Kenya; Virgulihracon endoxylapha- gus Quicke & Ingram, Australia. Cardi- ochilinae: Cardiochiles sp., Cameroun. Gen- Volume 4, 1995 113 ocoeliinae: Capitonius spp., Brazil (2). Charmontinae: Charmon cruentatus Haliday, UK (2). Doryctinae: Acanthodoryctes tomen- tosa (Szepligeti), Australia; Acropjliasmiis fernigineus (Marsh), USA; Allorhogas pyr- alophagus Marsh, Trinidad; Binarea sp., Ar- gentina; Doryctes erythromeles (Brulle), USA; Liobracon sp., Brazil; Megaloproctiis platyantennus Marsh, Brazil; Monarea sp., Brazil; Rhaconotus sp., USA; Rutheia sp., Bolivia; Syngaster sp., Australia; Spathiiis sp., Kenya; Zombrus sp., Kenya. Helconi- nae: Austrohelcou sp., Australia; Aspicolpus carinator Nees, Hungary; Aspidocolpus sp., Nigeria; Baecis abietis (Ratzeburg), Poland; Briilleia latiannulatus (Cameron), New Guinea; Calohelcon obsciiripennis Turner, Australia; Diospilus capita (Nees), Hungary (2); Diospilus morosus Reinhard, Austria; Eubazus pallipes Nees, Romania; £. tibialis Haliday, Bulgaria; £. (Brachistes) spp., Ar- gentina, UK (2); Helcon uunciator Fabricius, Hungary; Helconidea tentator (Fabricius), Belgium; Polydegmon sinuatus Foerster, Hungary; Taphaeus hiator Thunberg, Hun- gary; Wroiightonia sp., Canada; unde- scribed genus A, Australia; undescribed genus B, Australia. Homolobinae: Homo- lobiis truncator (Say), Bulgaria (2); H. {Oiil- ophiis) sp., USA. Macrocentrinae: Austroz- ele uniformis Provancher, USA; Macrocen- trus linearis Nees, Hungary (2); Macrocen- triis sp., Poland. Microgastrinae: Sathon falcatus (Nees), Germany. Microtypinae: Microtypus sp., Hungary. Opiinae: Doryc- tobracon sp., Trinidad. Orgilinae: Orgilus leptocephnlus (Hartig), Germany; Orgilus spp., Cameroun, France (3); Orgilonia sp., Cameroun; Stantonia sp., Cameroun. Pam- bolinae: Monitoriella elongatn Hedqvist, USA. Pselaphaninae: Pselaphanus trogoides Szepligeti, Surinam. Rhysipolinae: Rhysi- polis sp., UK RESULTS Upper Valve Rhachis Scaling and Expansions In all Hymenoptera both the rhachis and the margins of the aulax are furnished with scale-like sculpture which, in the case of the parasitic Apocrita and the Orussi- dae, are generally rather widely spaced (Figs. 16, 17). In the Amicrocentrinae, Bla- cinae, Charmontinae and Helconinae the density of scaling was found to be consid- erably increased over a short region pre- apically (Figs. 3-11). Further, in the Char- montinae and Macrocentrinae this is ac- companied by a distinct swelling of the rhachis (Figs. 3-6; arrowed) and slight widening of the corresponding length of the aulax. Observation of living Macrocentrus fe- males held in the hand showed that they are capable of bending the apex of their ovipositors ventrally through nearly 90°. It seems likely that this is achieved in an analogous way to that proposed for the Agathidinae and Orgilinae (see below), with the swollen and more coarsely sculp- tured region of the rhachis providing in- creased resistance to relative movement of the ovipositor valves at the apical part of the ovipositor. Thus, if the dorsal valve is protracted relative to the ventral ones, they will not be able to slide past one an- other freely at the apex, and so the ovi- positor will be forced to curve ventrally in order to accommodate the relative exten- sion of the upper valve in a manner anal- ogous to the bending of a bimetal strip in a thermostat, as illustrated diagrammati- cally in Fig. 2a. The plausibility of this was conclusion was substantiated by manipu- lation of the cut ends of an excised Macro- cent rus ovipositor which showed that pro- traction of the upper valve relative to the lower one causes a ventral curving of the ovipositor tip. Upper and Lower Valve Bosses In all members of the Orgilinae and most Agathidinae examined, the pre-api- cal lateral margins of both the upper (Figs. 12, 13, 16, 17) and lower (Figs. 14, 15) ovi- positor valves possess an abrupt escarp- ment-like boss in lateral aspect. These structures are located lateral tt) the olis- 114 Journal of Hymenoptera Research Figs. 3-6. Scanning electron micrographs of apical parts of isolated upper ovipositor valves showing preap- ical expansion of rhachis (r) with denser scaling (arrows). 3, 4 (detail), Clmrniou cruoitntus. 5, Macrocentrus linearis. 6, Macrocentrus sp. Scale line: 3, 5: 50|jim; 4, 6: 25(xm. Volume 4, 1995 115 Figs. 7-11. Scanning electron micrographs of apical parts of isolated upper ovipositor valves showing weak preapical expansion of rhachis (r) with denser scaling (arrows). 7, 8, Amicrocenirum sp. 9, Hckonidea lentator. 10, Wwiightonia sp. 11, BniUeia latmwnilatiis. Scale line: 7: 100|xm; 8: ZOjim; 9, 50n.m; 10, 11: 40^,01. theter. In the case of the dorsal valve, the steep (scarp) face of the stop is at the distal end of the modified region whereas it is at the proximal end in the lower valve. The positioning of these modifications is such that when the ovipositor is in its rest- ing position with neither upper nor lower valves apically protruding beyonci the other, the scarp faces of both processes are closely apposed. The presence of these stops limits the relative longitudinal movement of the upper and lower valves such that the upper valve cannot be pro- tracted apically, relative to the lower valves, beyond the resting position though the upper valve can be retracted basally. The functional significance of this modifi- cation was revealed when living agathi- dines of the genus Agathis were held in the hand and the apical part of the ovipositor was observed to change its degree of ven- tral curvature as the wasp attempted to 116 Journal of Hymenoptera Research I Figs. 12-17. Scanning electron micrographs of apical parts of isolated upper (12, 13, 16, 17) and lower ovi- positor valves (14, 15) showing (arrows) corresponding preapical, bosses lateral to the rhachis (r) and aulax (a) of the upper and lower valves respectively. 12, 13, 15, Braunsia sp. 14, 16, Megagathis sp. 17, Orgilonia sp. Scale line: 12: 50|jLm; 13, 16: 20|xm; 14, 15: 30|xm; 17: 35|xm. free itself. Such bending appeared to be posing scarp faces of their bosses. Because achieved by the wasp protracting the up- the valves are interlocked along their per valve posteriorly relative to the lower whole length by the olistheter mechanism, valves, thereby forcing together the op- the ovipositor is forced to curve ventrally. Volume 4, 1995 117 as illustrated in Figure 2b. The plausibility of this was conclusion was substantiated by manipulation of the cut ends of an ex- cised Agnthis ovipositor which showed that protraction of the upper valve relative to the lower one causes a ventral curving of the ovipositor tip. No other braconids examined possessed this modification and since neither the Agathidinae nor the Orgilinae are likely to be the most basal of extant braconids (Quicke & van Achterberg 1990; Wharton et al. 1992; van Achterberg & Quicke 1992) (for example, because of their koinobiont endoparasitoid life history), it seems likely that the presence of a scarp-like stop mechanism is apomorphous. The potential implications of this are dealt with in the discussion. Pre-apical Constrictions of the Aulax Unlike other braconid subfamilies, most members of the subfamily Doryctinae have the apex of the ovipositor much darker (piceous or black) and much more strongly sclerotized than the remainder (Quicke et al. 1992a) such that it is usually extremely difficult to separate the valves for study even after softening in aqueous potassium hydroxide; though with persis- tent manipulation, the three parts could usually be separated. Scanning electron microscopy shows that in many genera (e.g. Binnrea, Liobracon, Monarea, Syngasier, Zotnbrus) the apical part of the lower valves, as well as having ventro-lateral serrations, also possess one or two appar- ently distinct dorso-lateral tooth-like structures (Figs. 18-21; arrows) (Quicke et al. 1992a). The significance of these in terms of ovipositor steering has been re- vealed by examination of the adjacent part of the aulax. The aulax is an apprc^ximate- ly parallel-sided groove along most of its length, in isolated ventral valves it is strongly constricted opposite these addi- tional teeth by an infolding of the lateral margin (Figs. 22-24). Together with the in- creased sclerotization of the ovipositor tip and our observation that it is very difficult to separate the dorsal and ventral valves at the tip region, we conclude that the nar- rowed region of the aulax acts to grip the rhachis tightly, thereby reducing the free- dom of movement of the valves near to the ovipositor tip. The asymmetric ar- rangement of the constrictions of the aulax are such that retraction of the upper valves is likely to afford less resistance than protraction (Figs. 2c cf d) which will cause the lateral wall of the aulax to roll inwards thus increasing the friction against the rhachis. This prediction was borne out by observations of living Hcca- holiis females which were found \.o be able to bend their ovipositors ventrally but not dorsally. DISCUSSION In all the cases in which we have found to possess probable ovipositor steering mechanisms, their hosts are concealed in- sect larvae. Orgilines mostly attack weakly concealed hosts such as leaf miners, and they probe into their hosts' tunnels often through frass holes rather than by boring directly through the plant material (Flan- ders & Oatman 1982; Shaw & Huddleston 1991). Agathidines, charmontines and ma- crocentrines similarly attack weakly con- cealed hosts such as leaf rollers, leaf tiers, shoot borers, and, particularly in the case of the Agathidinae, leaf miners /tunnellers. The steering mechanisms may therefore enable the wasps to locate their hosts better within their concealed feeding places. It may also enable the wasps to place their eggs accurately on or within their hosts. In the case of agathidines, for example, it is known that eggs are usually placed very precisely within or next to the host's cen- tral nervous system (references in Shaw & Huddleston 1991). Given that their hosts are concealed and often relatively small, such accurate egg placement would be fa- cilitated by the existence of some sort of directional control of the ovipositor tip. In contrast to the above mentioned sub- 118 Journal of Hymenoptera Research Figs. 18-24. Scanning electron micrographs of apical parts of intact ovipositors (18, 19) and isolated lower ovipositor valves (20-24) of doryctine braconids showing dorso-lateral tooth like structures (arrow) and in 24, the corresponding narrowing of the aulax (a). 18, Megalopwctus plaiyantenniis. 19, 24, Binarea sp. 20, Acroplias- mus ferrugineus. 21, Spathius sp. 22, Syngaster sp. 23, Zombrus sp. Scale line: 18: 130(xm; 19: 70|xm; 20, 21: 60|xm; 22, 23: 30|xm; 24: 10|xm. families, many members of the Doryctinae attack more strongly concealed insects such as wood-boring beetle larvae, and since they are idiobiont ectoparasitoids, the exact placement of the paralysing sting or of the egg is usually likely to be of little importance. The process of boring through bark or other hard substrates may not be very accurate and is also quite time consuming, which in turn probably places the wasp under an increased risk of pre- dation. Locating a host tunnel might therefore be a good first step towards reaching a host if the wasp can subse- quently manipulate its ovipositor tip. Steering mechanisms might be seen in these cases as an adaptation which over- comes some of the potential problems of reaching a concealed host directly through a thick layer of substrate. Of some surprise was that we failed to find any morphological modifications that Volume 4, 1995 119 could be involved in ovipositor steering in a number of taxa with long to very long ovipositors such as members of the bra- conine genera Archibracon, Euurobracou, Glyptomorpha, Nesaidax, Odesia and Virgii- libracon. Nor were any steering mecha- nisms identified in the Alysiinae, Cardi- ochilinae, or Cenocoeliinae examined. Ob- servations of living females of these taxa would be valuable as it is quite possible that they have some other, perhaps less conspicuous, mechanisms that enable them to manipulate their ovipositors or ovipositor tips. Certainly for many of the braconines with long ovipositors, ovipo- sition involves boring through hard wood (van Achterberg 1986) and this may pose limitations on the bending mechanisms they might possess. Alternatively, it may be that most braconines that attack deeply concealed, wood-boring hosts are simply not adapted to hit the host's tunnel first and then to use the ovipositor to follow the tunnel to the host, but rather reach their hosts by drilling directly through the substrate to them. If this is generally the case, then the strong, possibly defensive, smell of most braconines (Quicke 1988; Quicke & Wharton in preparation) could be an adaptation to protect them during protracted periods of drillng for hosts, during which time they may be particu- larly vulnerable to predation. Interestingly, the Agathidinae and Or- gilinae did not appear together in the phy- logenetic analyses of Quicke & van Ach- terberg (1990) and Wharton et al. (1992) though their relationships were rather la- bile. These two subfamilies have a similar general appearance and perhaps more im- portantly, have two other putative syna- pomorphies, a relatively narrow forewing marginal cell with an approximately straight vein 3-SR, and in those species with a complete 2nd submarginal cell, this is small and usually triangular or at least has vein 2-SR very short. The discovery of a new apomorphous character present in both, further suggests the possibility that they could be sister groups, though the possibility also exists that the ovipositor bending mechanisms possessed by mem- bers of the two subfamilies reflect conver- gence as a result of the similarities in their hosts' biologies. In a preliminary investi- gation, we have added the presence or ab- sence of an ovipositor boss as an addition- al character to a data matrix of braconid subfamilies based on that of Quicke & van Achterberg (1990) modified in accordance with the suggestions of Wharton et al. (1992) and of van Achterberg & Quicke (1992). Whereas the Agathidinae and Or- gilinae did not appear as sister groups in the most parsimonious trees (length 391), trees with them so arranged were only one step longer than the most parsimoni- ous ones obtained. The absence of this character in the Pselaphaninae might be taken to indicate that they are not the sis- ter group of the Agathidinae (van Achter- berg 1985), however, it should be noted that in Pselaphanus, the ovipositor is rather robust and is probably unsuited to "steer- ing" and that such a mechanism could have been secondarily lost. In this context, it should also be noted that members of some genera of Agathidinae that have shorter and more robust ovipositors were also found to lack the bosses, and there- fore this feature may be readily lost in spe- cies whose ovipositors evolve to become less flexible. We therefore conducted a sec- ond phylogenetic analysis on the data ma- trix described above but this time scoring the Pselaphaninae as unknown for the new ovipositor character, thus allowing the possibility of a reversal in that group without adding to tree length. In this case the most parsimonious trees included a subset containing the clade (Orgilinae -I- (Agathidinae + Pselaphaninae)). We therefore suggest that the newly discov- ered ovipositor features should be incor- porated in future phylogenetic analyses of the Braconidae. ACKNOWLEDGMENTS We would like to th.ink Rachel Kruft and Laraine Ficken for their assistance with scanning electron mi- 120 Journal of Hymenoptera Research croscopy, and Kees van Achterberg, Paul Marsh and Jeno Papp for providing some of the material exam- ined. Our thanks also to Mark Shaw and Robert Bel- shaw who provided many helpful comments on an earlier version of the manuscript. The work was part- ly supported by a NERC research grant (GR3/8060) to Donald Quicke and Mike Fitton and by the NERC Initiative in Taxonomy. LITERATURE CITED Achterberg, C. van. 1985. Notes on Braconidae. V. The systematic position of the genera Ecnomios Mason and Pselaphanus Szepligeti (Hymenop- tera: Braconidae). Zoologische Verhandeliiigen, Lei- den 59: 341-348. Achterberg, C. van. 1986. The oviposition behaviour of parasitic wasps with very long ovipositors (Ichneumonoidea: Braconidae). Eutomologische Berichten 46: 113-115. Achterberg, C. van and D. L. J. Quicke. 1992. Phy- logeny of the subfamilies of the family Braconi- dae: A reassessment assessed. Cladistics 8: 237- 264. Austin, A. D. and T. O. Browning 1981. A mecha- nism for movement of eggs along insect ovipos- itors. International Journal of Insect Morphologx/ and Emhryologi/ 10: 93-108. Compton, S. and R. Nefdt. 1988. Extra-long ovipos- itors in chalcid wasps; some examples and ob- servations. Antenna 12: 102-105. Delanoue, P. and Y. Arambourg. 1965. Contribution a I'etude en laboratoire d'Eupelmus urozonus Dalm. (Hym. Chalcidoidea Eupelmidae). Annales de la Societe Entomologique de France (N.S.) 1: 817- 842. Flanders, R. V. and E. R. Oatman. 1982. Laboratory studies on the biology of Orgiliis jenniae (Hyme- noptera: Braconidae), a parasitoid of the potato tuberworm, Phthorimaea operculella (Lepidoptera; gelechiidae). Hilgardia 50: 1-33. Gauld, I. D. & B. Bolton (eds). 1988. The Hymenoptera. British Museum (Natural History/Oxford Uni- versity Press, Oxford. 332pp. Oeser, R. 1961. Vergleichend-MorphologischeUnter- suchungen iiber den Ovipositor der Hymenop- teren. Mitteihmgen aiis dem Zoologische Museum in Berlin 37: 1-119. Quicke, D. L. J. 1988. Host relationships in the Bra- coninae (Hymenoptera: Braconidae) — How little we know! Entomological Society of Queensland News Bulletin 16: 85-92. Quicke, D. L. J. 1991. Ovipositor mechanics of the braconine wasp genus Zaglyptogastra and the ich- neumonid genus Pristomerus. Journal of Natural History 25: 971-977. Quicke, D. L. J. and C. van Achterberg. 1990: Phy- togeny of the subfamilies of the family Braconi- dae (Hymenoptera). Zoologische Verhandelingen 258: 1-95. Quicke, D. L. J., L. C. Ficken, and M. G. Fitton. 1992a. New diagnostic ovipositor characters for doryc- tine wasps (Hymenoptera, Braconidae). Journal of Natural History 26: 1035-1046. Quicke, D. L. J., L. C. Ficken, and M. G. Fitton. 1992a. New diagnostic ovipositor characters for doryc- tine wasps (Hymenoptera, Braconidae). Journal of Natural History 26: 1035-1046. Quicke, D. L. J. and M. G. Fitton. In press. Ovipositor steering mechanisms in parasitic wasps of the families Gasteruptiidae and Aulacidae (Hyme- noptera). Preceedings of the Royal Society. Quicke, D. L. J., M. G. Fitton, J. R. Tunstead, S. N. Ingram, and P. V. Gaitens. 1994. Ovipositor structure and relationships within the Hymenop- tera, with special reference to the Ichneumono- idea. Journal of Natural Histoty 28: 635-682. Quicke, D. L. J. and P. M. Marsh. 1992. Two new species of Neotropical parasitic wasps with high- ly modified ovipositors (Hymenoptera, Braconi- dae: Braconinae and Doryctinae). Proceedings of the Entomological Society of Washington 94: 559- 567. Quicke, D. L. J. and R. A. Wharton. In prep. Antero- lateral scent glands of braconine wasps (Hyme- noptera: Braconidae). Shaw, M. R. and T. Huddleston. 1991. Classification and biology of braconid wasps. Handbooks for the Identification of British Insects. Vol. 7, Part 11. Roy- al Entomological Society, London. Smith, E. L. 1968. Biosystematics and morphology of Symphyta. I. Stem-galling Euura of the California region, and a new female genitalic nomenclature (Hymenoptera: Tenthredinidae). Annals of the En- tomological Society of America 61: 1389-1407. Smith, E. L. 1970. Evolutionary morphology of the external insect genitalia. 2. Hymenoptera. Annals of the Entomological Society of America 63: 1-27. Wharton, R. A., S. R. Shaw, M. J. Sharkey, D. B. Wahl, J. B. Woolley, J. B. Whitfield, P. M. Marsh and J. W. Johnson 1992. Phylogeny of the subfamilies of the family Braconidae (Hymenoptera: Ichneu- monoidea): a reassessment. Cladistics 8: 199-235. J. HYM. RES. Vol. 4, 1995, pp. 121-136 The Wasp Genus Tachytella Brauns, 1906 (Hymenoptera: Sphecidae) WOJCIECH J. PULAWSKI California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA Abstract.— The southern African genus Tachytella Brauns includes aurcopilosa Brauns, 1906 (west- ern South Africa) as well as two new species: heliophila (western Namibia) and nama (southern Namibia, northwestern South Africa). A lectotype is designated for Tachytella aureopilosa var. nana Arnold, 1936, the latter name being synonymized with Tachytella aureopilosa Brauns, 1906. The genus is redescribed using previously known and newly discovered characters, a key to the spe- cies is provided, and phylogenetic relationships between the species are discussed. INTRODUCTION General. — Tachytella Brauns, a little known genus from southern Africa, was thought to be monotypic until now. Spec- imens are rarely collected (Bohart and Menke 1976, for example, knew of only six). A decade ago, I discovered an unde- scribed species in materials from the State Museum Windhoek that Ole Lomholdt, then at the Zoological Museum, Copen- hagen, kindly had transshipped to me. During our expedition to Namibia in 1990, Maximilian Schwarz and I collected spec- imens that represented another new spe- cies. These findings prompted me to re- vise the genus. I follow the format used for my revisions of Kohliella Brauns, Holo- tacfn/spliex de Beaumont, and Gastrosericus Spinola (Pulawski 1991, 1992, 1995). Technical Terms. — Most morphological terms follow Bohart and Menke (1976), but the mandibular terms are taken from Michener and Fraser (1978). The following terms are here redefined for convenience: Clypeal lobe: the projecting medioventral portion of the clypeus. Mandible: — adductor ridge: extends distad from the adductor swellings (on the inner side) and gradually becomes visible from the outside, constituting the distal part of the mandibular posterior margin; the two portions differ in size, the distal one being higher than the basal one; — condylar ridge: arises from the condyle, extends distad, and forms the basal por- tion of the posterior mandibular mar- gin; it is angulate distally in many Lar- rinae; — condyle: mandibular articulation on the occipital side of the head capsule; — notch: an emargination on the posterior margin, delimited basally by the con- dylar ridge and distally by the expand- ed portion of the adductor ridge; — posterior margin: extends between the condyle and mandibular apex; called externoventral margin by Bohart and Menke (1976) and lower margin by Michener and Fraser (1978); it actually consists of two components: the con- dylar ridge basally and the adductor ridge distally (the term posterior is pre- ferred because the head is hypognath- ous, and this edge is thus oriented pos- terad); Sternum, tergum: shortened terms gastral sternum, gastral tergum. Vertex: for — length: the distance between a hind- ocellus and an imaginary line connecting eye hindcorners (i.e., the point where the 122 Journal of Hymenoptera Research inner and the posterior portions of the or- bit meet); — width: the shortest interocular distance in the ocellar region. specimens that have Origin of Material. — The 29 examined apparently are all been collected so far. Institutional or per- sonal collections in which the material is deposited are abbreviated in the text as follows (names of contact persons are giv- en in parentheses): AMG: Albany Museum, Grahams- town, South Africa (Friedrich W. Cess). BMNH: British Museum (Natural His- tory), London, current nonstat- utory name: The Natural Histo- ry Museum (Loraine Ficken). CAS: California Academy of Sciences, San Francisco, California (Woj- ciech J. Pulawski). FSCA: Florida State Collection of Ar- thropods, Gainesville, Florida (Lionel A. Stange, James R. Wi- ley). MS: Maximilian Schwarz, Ansfelden bei Linz, Austria (personal col- lection). SAM: South African Museum, Cape Town, South Africa, including G. Arnold collection previously housed in Bulawayo, Zimbabwe (Margie A. Cochrane, Vincent B. Whitehead). FMP: Transvaal Museum, Pretoria, South Africa (Ilema Fourie). USNM: Smithsonian Institution, Na- tional Museum of Natural His- tory, Washington, D.C. (Arnold S. Menke). Genus Tachytella Tachytella Brauns, 1906:56. Type species: Tach- ytella aureopilosa Brauns, 1906, by monotypy. Diagnosis. — Tachytella is a member of Larrini (Bohart and Menke 1976) and thus has the hindocelli modified to flat, elon- gate scars (Figs. 8, 12, 20), the only rem- nant of a lens being a narrow, translucent band along the scar's outer margin (the band is broadly interrupted on the scar's posterolateral or lateral side, depending on its orientation). Within the tribe, Tach- ytella can be recognized by the following four characters: 1. The frons has a median, slightly raised, flat area that is delimited by a lat- eral sulcus (Figs. 6, 10, 18, 22, 28, 31), the sulci being subparallel and close to the or- bits. Similar sulci are found in some An- cistromma W. Fox, e.g., capax W. Fox, and in less specialized Parapiagetia Kohl such as genicularis (F. Morawitz) and odontosto- ma (Kohl), in which however, the frons is not raised. 2. The pronotal collar is rounded and reaches the scutum level, a condition also found in some Gastrosericiis, e.g. baobabiciis Pulawski and eurypus Pulawski. Tachytel- la, however, has three submarginal cells, whereas Gastrosericiis has two. 3. The submarginal cell III in the fore- wing is essentially rhomboidal (Fig. 1), a previously unused character, with the an- terior and posterior margins equal in length or nearly so; the proximal and dis- tal margins are weaklv sinuous, almost straight, nearlv parallel to each other. 4. The hindwing jugal lobe is short and ends well before crossvem cu-a; conse- quently, the jugal and anal excisions are widely separated (Fig. 1). A similar con- dition is found in some specialized Gastro- sericiis such as braunsi Arnold, herero Pu- lawski, and pulchellus R. Turner (Pulawski 1995). Description. — Bohart and Menke (1976) regarded Tachytella as monotypic and, consequently, considered as generic some structures that are found only in aureopi- losa but not in the other two species (e.g., such male structures as the edentate inner mandibular margin, dentate clypeal free margin, and sterna VI-VIII with dense, erect setae). They thought that the prono- tal collar not closely appressed to scutum Volume 4, 1995 123 0.5 mm 0.25 mm Figs. 1-3. Some of the generic characters of Tachytella. 1, fore and hindwing of aureopilosa (drawn from the holotype, the model Bohart and Menke 1976, used for their illustration); arrows indicate third submarginal cell, jugal excision, and anal excision. 2, anterior part of thorax of aureopilosa (arrow indicates pronotum). 3, foretarsus of female nama. 124 Journal of Hymenoptera Research also was a generic character, but in fact it is only individual. The following rede- scription takes into account the three spe- cies now recognized and also some struc- tures not considered previously. Included are those features that vary within other Larrini but are universal in Tachytella. Labrum flat, not emarginate apically, not or barely protruding beyond clypeal free margin; prementum convex but not compressed laterally; stipes nearly flat. Mouthparts not elongate. Paramandibular process broadly separated from back side of clypeus (mandibular socket open). Pos- terior mandibular margin notched; notch delimited proximally by angulate apex of condylar ridge, and distally by marked, roundly expanded distal portion of ad- ductor ridge. Clypeus produced into mes- al lobe, free margin of lateral section con- cave. Antennal socket narrowly separated from frontoclypeal suture. Inner orbits convergent above. Frons flat, with no gla- brous swelling above each antennal sock- et. Hindocellar scars elongate, broadly di- verging anterolaterad (their long axes forming an angle of about 120-140°), shorter than distance that separates each one from midocellus. Occipital carina ef- faced just short of hypostomal carina. Pro- notum anterodorsally with transverse groove; collar swollen, reaching level of scutum. Postspiracular carina evanescent, about as long as midocellar diameter. Ep- isternal sulcus originating near middle of subalar fossa, ending well before reaching anteroventral mesopleural margin; scrobal sulcus absent; subalar fossa not margined below; metapleural flange not expanded. Axilla the usual shape, i.e., not expanded or carinate. Propodeum short, distance be- tween spiracle and metanotum less than spiracle's length; dorsum setose through- out. No additional sclerites between me- tasternal apex and propodeum (no "pro- podeal sternum"). Forewing with three submarginal cells, none petiolate; recur- rent veins separate or (some aureopilosa) forming short petiole. Hindwing crossvein cu-a vertical (not inclined). Forecoxal apex not expanded into process. Hindcoxal dor- sum: inner margin not carinate. Hindtibia not margined. Forebasitarsus and apical tarsomeres without ventral spines. Claws in each pair equal in size. Hindtarsomere II long (0.6-0.7 X hindtarsomere I). Base of tergum I variable: with or without short, oblique carina that extends from each anterolateral corner. Lateral carina of tergum I complete, reaching tergal hind- margin. Tergum II not carinate laterally. Body without erect setae (except setae on apical sterna in male of aureopilosa); those on propodeal dorsum inclined obliquely toward head. Female. Inner mandibular margin with tooth. Forecoxa not carinate anteriorly, venter slightly convex; outer surface of foretibia without preapical spines. Foretar- sus with rake that consists of stiff, flat- tened, widely spaced spines (four on for- ebasitarsus, two on foretarsomere II); tar- someres IV: length more than apical width, apicoventral margin emarginate; tarsomeres V: apicoventral margin slightly arcuate. Claws without teeth. Tergum VI fairly convex: angle between lateral mar- gin of tergum and lateral margin of pygid- ial plate, in side view about 30-40°. Pygid- ial plate well defined (margined laterally), rounded apically, sparsely punctate and with thin, inconspicuous setae (Figs. 9, 21, 30). Sting sheaths somewhat flattened dor- soventrally. Male. Foretrochanter and forefemur not emarginate; forebasitarsus 1 with one rake spine (in basal half); outer, apical spine of foretarsomere II shorter than foretarso- mere III; sterna without velvety patches; tergum VII carinate apicolaterally, not de- pressed apically; sterna III- VI impunctate at least mesally; sternum VIII rounded apically or (some aureopilosa) minimally emarginate. Venter of penis valve finely denticulate subapically (Figs. 17, 27, 35). Character Polarization. — The tribe Larri- ni, of which Tachytella is a member, con- sists of two lineages, the subtribes Larri- Volume 4, 1995 125 heliophila aureopilosa nama aureopilosa heliophila nama Figs. 4a and b. Cladograms showing hypothetical cladistic relationships of the three species of Tachytella; 2: modified cleft of inner mandibular margin, 8: penis valve with medioventral tooth. na and Tachytina (Bohart and Menke 1976). The autapomorphy of Larrina is a unique frons: the midocellus is placed in a depression, and the transverse sweUing delimiting the depression anteriorly joins a linear swelling along the inner orbit. The hindocelli are oriented along a trans- verse line, and the transparent lens rem- nant is horseshoe-like, with the two branches parallel and close to each other, apparently the plesiomorphic condition within the tribe. In Tachytina, the frons is the usual plesiomorphic condition, with no depression or swelling, but the hin- docelli are oriented in a V-shaped pattern and the lens remnant is clearly derived: the anterior branch is longer than the pos- terior or (in Parapiagetia) the branches are roughly circular (open on the lateral side). Because of the shape of the lens remnant, Tacln/tella is a member of Tach- ytina, but its relationships within the sub- tribe are unclear. Bohart and Menke (1976) noted that Tacln/telln resembled Ancistrotnma and Larropsis Patton in the shape of the mandible, ocellar scar, fe- male foretarsal rake, and male sternum VIII. None of these similarities, however, appears to be a synapomorphy. I have found no single derived character linking Tacln/telln with any of the other nine gen- era of the subtribe. For the purpose of this analysis, I regard the Larrina and the re- maining Tachytina as the outgroup. The characters found in both Tacln/telln and the outgroup are considered plesiom- orphic, unless there is strong evidence that they have been acquired indepen- dently. The characters present only in some Tncln/tella but nc^t the outgroup are regarded as apomorphic. The character codings are: 0. ancestral, 1 and 2. derived. The following characters have been con- sidered: 1. Female mandible inner margin: sub- basal tooth: 0. well defined, 1. inconspic- uous, obtuse (as in Tnclnjtelln nureopilosa). The tooth is present in most members of the outgroup, but absent in Lnrrn Fabri- cius, the subgenus Motes Kohl of Liris Fa- bricius, in Parapiagetia, several specialized Gastrosericus, and in two Tachysphex Kohl {nefarius Pulawski and ramses Pulawski). Because of their many specializations, none of these are likely to share a unique common ancestor with Tacln/telln, and consequently I regard the presence of a well defined mandibular tooth in Tacliy- tella as ancestral. 2. Female mandible inner margin: cleft: 0. narrow, 1. shallow, obtusely angulate (as in Tacln/tella aureopilosa , 2. deep, broad (as in Tacln/tella nama). A cleft is present and narrow in most Larrinae (including Tacln/tella heliophila), but absent in Larra, the subgenus Motes of Liris, Dalara Ritse- 126 Journal of Hymenoptera Research 9 aureopilosa Q heliophiia nama ( Fig. 5. Collecting localities of Tachytella. ma, Paraliris Kohl, Holotachysphex, many specialized Gastrosericus, Parnpiagetia geni- cularis (F. Morawitz), Tachysphex nefarius and ramses. None of these is likely to share a unique common ancestor with Tachytel- la, and I therefore accept that a narrow cleft is ancestral in the latter genus. The broad, deep cleft is unique within the tribe. 3. Male clypeus: 0. without longitudinal swellings, 1. swellings present. The clyp- eal swellings of aureopilosa are unique within Larrinae and thus apomorphic. 4. Propodeum: 0. without V-shaped im- pression, 1. with V-shaped impression. The impression is unique to Tachytella au- reopilosa and thus apomorphic. 5. Male foretarsus: 0. tarsomere II not expanded, 1. tarsomere II expanded. The expanded tarsomere of Tachytella aureopi- losa is unique within the tribe and thus apomorphic. 6. Tergum I: 0. with a pair of short, bas- al carinae, 1. carinae absent. Within Lar- rini, the carinae are absent in Gastrosericus, Holotachysphex, Kohliella, Parapiagetia, Tach- Volume 4, 1995 127 0.25 mm 0.5 mm 8 f-mm 0.25 mm Figs. 6-9. Tachytella aiireopilosa, female. t>, head frontally. 7, clypeus. S, head dorsally. 9, pygidial plate. i/sphex, manv Tachvtes Panzer, and Larrop- sis chilopsidis (Cockerell and Fox). They are present m all other Larrini and other Sphecidae (those with a petiolate gaster are obvious exceptions). 1 therefore regard the absence of carinae in Tachytella aiireo- pilosa as derived. 7. Male gaster: 0. sternal setae all short, 1. setae of sterna VI-VIII long, erect. Ster- nal setae are short, appressed in most Lar- rinae, but long, erect on apical sterna in many Liris and in Tachytella aiireopilosa. These species are not likely to share a sin- gle common ancestor. In addition, the se- tal configuration c^f T. aiireopilosa is unique and I regard it to be derived. 8. Penis valve: 0. without medioventral tooth, 1. with medioventral tooth. This tooth, found m Tachuteila heliophila and nama, is unique within Larrinae and thus apomorpnic. The following 5 characters cannot be polarized because both states are com- monly found in the outgroup: 9. Gena: a. thin in dorsal view, b. thick in dorsal view. 10. Mesopleural vestiture: a. not con- cealing integument, b. concealing integu- ment. 11. Hindtarsomere IV: a. length about 1.0 X apical width, b. length about 1.2-1.3 X apical width. 12. Pygidial plate of female: a. apex with a row of setigerous punctures. 128 Journal of Hymenoptera Research 0.5 mm 0.5 mm 12 0.25 mm 11 0.25 mm 13 Figs. 10-13. Tachytella aureopilosa, male. 10, head frontally. 11, clypeus. 12, head dorsally with outline of vertex showing variation. 13, scape. b. apex without a row of setigerous punc- tures. 13. Male mandible: a. inner margin an- gulate, b. inner margin simple. Cladistic Analysis. — Distribution of the 8 polarized characters is given below: Number: ancestor: aureopilosa: heliophila: nama: 12345678 00000000 11111110 00000001 02000001 Only two derived characters are found in more than one species: character 2 (a modified cleft on inner mandibular mar- gin) is shared by aureopilosa and nama, and character 8 (penis valve with medioventral tooth) by heliophila and nama. The clado- gram resulting from character 2 (Fig. 4a) is heliophila + {aureopilosa + nama), the cladogram based on character 8 (Fig. 4b) is aureopilosa + {heliophila + nama). I see no reason to prefer one over the other. Life History. — Unknown. The presence, in the female, of a foretarsal rake and a pygidial plate suggests ground nesting. Geographic Distribution (Fig. 5). — Namib- ia and western half of South Africa. Volume 4, 1995 129 KEY TO THE SPECIES 1. Propodeal dorsum with V-shaped impression that begins behind each spiracle and ends at hindmargin (impression inconspicuous in smallest individuals); tergum I basally without pair of short, oblique carinae. Female: clypeus (except basally) with finely, closely punctate median impression that separates two unsculptured, longitudinal areas (Fig. 6); imier man- dibular margin with inconspicuous subbasal tooth (Fig. 6). Male: clypeus with pair of glabrous, longitudinal swellings, free margin with small, median tooth (Fig. 10); sterna VI- VIII with dense rows or tufts of erect setae (Fig. 15); penis valve without ventral tooth (Fig. 17) aurcopilosa Brauns - Propodeal dorsum without such impression; tergum I basally with pair of short, oblique carinae. Female: clypeal bevel without median row of small punctures; inner mandibular margin with well-defined subbasal tooth (Figs. 18, 28). Male: clypeus with no longitudinal swellings, free margin without small, median tooth; sterna without dense rows or tufts of erect setae; penis valve with ventral tooth (Figs. 27, 35) 2 2. Mesopleural vestiture conspicuous, fully concealing integument. Length of hindtarsomere IV 1.2-1.3 X apical width. Female: free margin of clypeal lobe weakly arcuate, corner well defined but not prominent (Fig. 18); inner mandibular margin with usual, narrow cleft (Fig. 18). Male: clypeal lobe broad, nonprominent (Fig. 22); inner mandibular margin sim- ple (Fig. 22) heliophila sp. n. - Mesopleural vestiture inconspicuous, integument easily visible. Length of hindtarsomere IV about 1.0 X apical width. Female: free margin of clypeal lobe sinuate, with prominent corner (Fig. 28); inner mandibular margin with broadly expanded cleft (Fig. 28). Male: clypeal lobe prominent, markedly narrowing anterad (Fig. 31); trimmal margin obtusely dentate (Fig. 31) nama sp. n. DESCRIPTIONS OF SPECIES (TMP), present designation, examined. New synonym. — As Tachytella aiireopilosa nana: Diagnostic characters for each species ^^^^^^ ^^^ ^^^^^ ^^7^^253 (illustration of are given m the key and are not repeated ^^^^ ^ead), 256 (new status, listed), individually in the descriptions below. Synonymy. — Arnold (1936) described a Tachytella aiireopilosa Brauns yar. nana for specimens that had red gas- Figures 6-17 tral segments I-III, shallower V-shaped „ , ^ ,, ,1, inn^ err i • imprcssion on thc propodcal dorsum, and Tnan/tella aiireo-vilosa Brauns, 1906:57, male, in- '^ ,,,,,, • . . : , , . .. u 1 . ^ truncate rather than shallowly emarginate correct original hyphenation. Holotype: -^ " male. South Africa: Cape Province: Willow- male sternum VIII. Because these charac- more (TMP), examined.— Arnold 1923:218 ters vary individually, I regard var. nana (revision), 1930:5 (listed); Bohart and Menke as a synonym of aiireopilosa. 1976:231 (illustration of ocelli), 253 (illustra- Description. — Gena thin in dorsal view tion of male head), 255 (illustration of (Figs. 8, 12). Vertex with triangular, gla- wings), 257 (listed), 258 (illustrations of pro- brous, impunctate, dull area that extends notum and scutum), 268 (illustration of fe- f^.^^ Q^bit toward midline. Head, thorax, male foretarsus), 279 (illustration of male ^^^^^^^ ^^^ femora minutely punctate, sternum VIII), 280 (illustration of male geni- , i.u i ^- I^ .*. ,^ ^^^..f talia); Gess 1981:19 (South Africa: nefting punctures no more than 1 diameter apart, probably in friable soils). hence integument mat except for shiny Tachytella aureo-pilosa race nana Arnold, 1936:29, clypeal swellings in male. Propodeal dor- female, male. Lectotype: male. South Africa: sum, in most specimens studied, with Cape Province: Resolution in Albany District pair of impressions that meet postero- 130 Journal of Hymenoptera Research 0.25 mm 0.25 mm 15 0.1 mm Figs. 14-17. Tachytella aiireopilosa, male. 14, foretarsomere. 15, gastral apex in profile. 16, volsella. 17, penis valve. mesally and delimit triangular area re- throughout. Length of hindtarsomere IV sembling propodeal enclosure of most about 1.0 X apical width. Gastral seg- Sphecidae (impressions barely visible in ment I without basolateral, oblique cari- smallest specimens, apparently a result of na. allometric growth). Propodeal hindface Mesopleural vestiture concealing integ- not ridged. Femora and tibiae setose ument. i Volume 4, 1995 131 0.5 mm 20 0.25 mm 19 0.1 mm Figs. 18-21. Tacln/tt'lla heliophila, female. 18, head frontally. 19, clypeus. 20, head dorsally. 21, pygidial plate. Head and thorax black but the follow- ing are reddish: mandible (apical third dark), scape, tegula, and humeral plate. Gastral segment I to I-III red, remainder black. Mid- and hindfemora black, forefe- mur (except dorsally), tibiae and tarsi red. Wings nearly hyaline. Terga I-V silvery fasciate apically. Female. — Mandible: inner margin di- vided by an obtuse angle into basal and distal portions, with shallow, obtusely an- gulate cleft. Clypeus with median, finely, densely punctate impression that sepa- rates two unsculptured, longitudinal areas (Figs. 6, 7); lobe free margin arcuate except markedly concave near corner, which is prominent (Figs. 6, 7). Length of flagello- mere I 2.0 X apical width. Vertex width 1.4 X length. Tergum V punctate and se- tose throughout. Pygidial plate with preapical row of setigerous punctures (Fig. 9). Length 7.3 mm Male. — Mandible: inner margin simple, not angulate (Fig. 10). Clypeus (Figs. 10, 11): lobe free margin with median tooth, corner well defined but not prominent; distance between corners about 0.8 X dis- tance between corner and orbit; middle section with a pair of longitudinal swell- ings that are either microsculptured, dull, or unsculptured, shiny. Scape elongate, swollen apicoventrally (Fig. 13). Length of flagellomere I 1.7-1.8 X apical width. Ver- tex width 1.6 X length. Foretarsomere II enlarged (Fig. 14). Tergum VII: punctures averaging about 1 diameter apart, some of them about 3 diameters apart. Apex of sternum VI, sternum VII, and VIII with 132 Journal of Hymenoptera Research 0.5 mm 24 0.25 mm 23 0.25 mm 25 Figs. 22-25. Tachytella heliophila, male. 22, head frontally. 23, clypeus. 24, head dorsally. 25, scape. erect setae; longest setae equal to basal mandibular width. Length: 6.1-8.5 mm. Volsella: Fig. 16. Penis valve without ven- tral tooth (Fig. 17). Geographic Distribution. — Western South Africa. Records.— SOUTH AFRICA: Cape Prov- ince: Colesberg (1 female, SAM), Gra- hamstown: Hilton, sand pit (1 male, AMG), NE Ouberg Pass 31 road km NE Montagu at 33°39'S 20°18'E (1 male, AMG), Resolution in Albany District (1 fe- male, 1 male, AMG; 1 male, SAM; 1 head- less female, 1 male lectotype of nana, TMP), 10 km W Steinkopf (1 female, FSCA), Willowmore (1 male, TMP, holo- type of aureopilosa), Worcester (3 males, BMNH; 1 male, USNM). Tachytella heliophila Pulawski, new species Figures 18-27 Name Derivation. — Heliophila, a Neolatin feminine adjective derived from two Greek words: helios, sun; and philos, a friend; with reference to the open, sunny habitat where the species was found. Description. — Gena thin in dorsal view (Figs. 20, 24). Vertex without impunctate, triangular area. Head, thorax, coxae, fem- ora and tibiae minutely punctate, punc- tures no more than 1 diameter apart, hence integument mat except for the un- sculptured, shiny clypeal bevel. Propodeal hindface not ridged. Femora and tibiae se- tose throughout. Length of hindtarsomere IV 1.2-1.3 X apical width. Gastral segment I basolaterally with short, oblique carina. Volume 4, 1995 133 mTTT^. 0.1 mm Figs. 26 and 27. Tachytella heliophila. 26, volsella. 27, penis valve. Mesopleural vestiture concealing integ- ument. Head and thorax black but the follow- ing are reddish: mandible (apical third black), clypeal bevel anteriorly, scapal venter apically, pronotal lobe, tegula, and humeral plate. Gaster red. Coxae and fem- ora black (femora red apically), tibiae and tarsi red. Wings hyaline. Terga I-V silvery fasciate apically. Female. — Mandible: inner margin with small subbasal tooth and narrow cleft (Fig. 18). Clypeus (Figs. 18, 19): free margin of lobe arcuate, corner angulate but not prominent; apicomedian portion of mid- dle section shiny, practically unsculptured (with only a few microscopic punctures). Length of flagellomere I 2.75 X apical width. Vertex width 1.7 X length. Tergum V punctate and setose throughout. Pygid- ial plate with preapical row of setigerous punctures (Fig. 21). Foretibia without preapical spines. Length 6.6-7.5 mm. Male. — Mandible: inner margin simple. not angulate (Fig. 22). Clypeus (Figs. 22, 23): lobe free margin arcuate, corner ob- tusely angulate; surface finely punctate and setose throughout except for narrow apical lip; distance between corners about 0.9 X distance between corner and orbit. Scape swollen ventrally (Fig. 25). Length of flagellomere I 2.1 X apical width. Ver- tex width 3.0 X length. Tergum VII: punc- tures averaging less than one diameter apart. Length 6.4 mm. Volsella: Fig. 26. Pe- nis valve with ventral tooth near mid- length (Fig. 27). Geographic Distribution. — Known from a single area in Namib desert, western Na- mibia. Records. — Holotype: female, Namibia: Karibib District: 65 km SW Usakos, 24 Feb. 1990, W.J. Pulawski (CAS). Paratypes: NAMIBIA: Karibib District: 55 km SW Usakos, 1 Mar. 1990, W.j. Pu- lawski (1 female, CAS); 65 km SW Usakos, M. Schwarz, 24 Feb. 1990 (1 female, MS), 1 Mar. 1990 (1 female, MS; 1 male, CAS). 134 Journal of Hymenoptera Research 0.5 mm 29 0.25 mm Figs. 28-30. Tachytella nama, female. 28, head frontally. 29, head dorsally. 30, pygidial plate. Tachytella nama Pulawski, new species Figures 28-35 Name Derivation. — Nama, a Hottentot tribe that immigrated from southern Af- rica into the central Namib; a noun in ap- position to the generic name. Description. — Gena thick in dorsal view (Figs. 29, 32). Vertex without impunctate, triangular area. Scutal punctures varying from mostly about 1 diameter apart (fe- male from Goegap) to many diameters apart (most specimens). Mesothoracic ven- ter with minute punctures that are several diameters apart. Propodeal hindface mi- croscopically, densely ridged and with ev- anescent punctures. Fore- and midfemo.ral venter with a few, scattered punctures, as- etose. Outer side of foretibia as well as mid- and hindtibial dorsum impunctate, asetose. Length of hindtarsomere IV about 1.0 X apical width. Gastral segment I ba- solaterally with short, oblique carina. Mesopleural vestiture inconspicuous, not concealing integument. Head and thorax black except mandible yellowish basally and reddish distally, middle clypeal lobe of male reddish, male flagellum yellowish brown ventrally. Gas- tral segments I-III red, remainder black. Wings almost hyaline. Terga I-III silvery fasciate apically, but fasciae evanescent in most specimens. Female. — Mandible: inner margin with subbasal tooth and broadly expanded cleft (Fig. 28). Clypeus (Fig. 28): lobe free margin sinuous, corner prominent; sur- face all or largely unsculptured (at most punctate basally). Length of flagellomere I 2.0 X apical width. Vertex width 2.4 X length. Tergum V with impunctate, gla- Volume 4, 1995 135 0.5 mm 32 0.5 mm 0.1 mm Figs. 31-35. Tachytella nama, male. 31, head frontally. 32, head dorsally. 33, scape. 34, volsella. 35, penis valve. brous apical depression. Pygidial plate with no preapical row of setigerous punc- tures (Fig. 30). Foretibia with two spines between inner surface and dorsum: one at midlength, the other at two thirds of length. Length 7.5 mm. Male. — Mandible: inner margin obtuse- ly angulate (Fig. 31). Clypeus (Fig. 31): middle section markedly convex; lobe prominent, markedly narrowing anterad, with unsculptured apical part and shal- lowly emarginate anterior margin. Scape minimally swollen ventrally (Fig. 33). Length of flagellomere I 1.5 X apical width. Vertex width 2.4 X length. Tergum VII: punctures averaging more than one 136 Journal of Hymenoptera Research diameter apart. Length 7.2 mm. Volsella: Fig. 34. Penis valve with ventral tooth (Fig. 35) near midlength. Geographic Distribution. — Southern Na- mibia and northwestern South Africa. Records. — Holotype: female. South Afri- ca, Cape Province: Hester Malan [now Goegap] Nature Reserve near Springbok, 15-21 Oct. 1987, F.W. and S.K. Gess (AMG). Paratypes: NAMIBIA: Liideritz District: Namuskluft 88, 2716 Dd [= between 27°45' and 28°00'S and 16°45' and 17°00'E], collector unknown, 12-15 Sept. 1973 (1 female, SMW), 7-15 Oct. 1970 (1 male, SMW). SOUTH AFRICA: Cape Province: Anenous, 29°14'30"S, 17°34'45"E, 11-13 Oct. 1988, D.W. Gess (2 females, AMG); Dikbome, Merweville Koup, [South Afri- can] Museum Expedition, Oct. 1952 (1 male, SAM); Hester Malan [now Goegap] Nature Reserve near Springbok, 15-21 Oct. 1987, F.W. and S.K. Gess (1 female, CAS); same data but 10-12 Oct. 1988 (1 male, AMG); between Kamieskroon and Springbok, [South African] Museum Staff, Oct. 1952 (2 females, CAS, SAM). ACKNOWLEDGMENTS Illustrations were prepared by Mary Ann Tenorio on a Macintosh computer using the Adobe Illustrator program. Fieldwork in Namibia in 1989, where I col- lected specimens, was supported by the National Sci- ence Foundation (grant number BSR-8722030) and museum work in South Africa the same year by the California Academy of Sciences through its In-House Research Fund. I sincerely thank persons who sent specimens for study. I am indebted to Arnold S. Men- ke and Friedrich W. Gess for their critical reviews of the manuscript. I sincerely thank my friend and trav- el companion to Namibia, Herr Maximilian Schwarz, who generously donated the unique male of Tachy- tella heliophila to the California Academy of Sciences. LITERATURE CITED Arnold, G. 1923. The Sphegidae of South Africa. Part III. Annals of the Transvaal Museum 9:191-253. Arnold, G. 1930. A check-list of the Sphegidae of the Ethiopian Region. University Press, Cambridge, England. 21 pp. Arnold, G. 1936. New African Hymenoptera No. 3. Occasional Papers of the Rhodesian Museum 5:1-38, pi. I. Bohart, R.M., and A.S. Menke. 1976. Sphecid zoasps of the world. A generic revision. University of Cali- fornia Press, Berkeley, Los Angeles, London. 1 color plate, IX + 695 pp. Brauns, H. 1906. Zur Kenntnis der siidafrikanischen Hymenopteren. Verhandlungen der Zoologisch-Bo- tanischen Gesellschaft in Wien 56:43-59. Gess, F. 1981. Some aspects of an ethological study of the aculeate wasps and the bees of a karroid area in the vicinity of Grahamstown, South Af- rica. Annals of the Cape Provincial Museums (Nat- ural History) 14:1-80. Michener, CD. and A. Fraser. 1978. A comparative anatomical study of mandibular structures in bees. The University of Kansas Science Bulletin 51: 463^82. Pulawski, W.J. 1991. A revision of the wasp genus Kohliella (Hymenoptera: Sphecidae). Proceedings of the California Academy of Sciences 47:289-302. Pulawski, W.J. 1992. World species of the wasp ge- nus Holotachysphex de Beaumont (Hymenoptera: Sphecidae). Proceedings of the Etitomological Society of Washington 94:223-242. Pulawski, W.J. 1995. The wasp genus Gastrosericus Spinola, 1839 (Hymenoptera: Sphecidae). Mem- oirs of the California Academy of Sciences (in press). J. HYM. RES. Vol. 4, 1995, pp. 137-193 Revision of the Ant Genus Gnantptogenys in the New World (Hymenoptera: Formicidae) John E. Lattke Institute de Zoologia Agricola, Universidad Central de Venezuela, Apartado 4579, Maracay, Venezuela 2101-A Abstract. — Ants of the genus Gnmnptogenys Roger in the New World are revised. The study is based mostly on worker morphology; 73 species are recognized, fifteen of which are new: G. andina, G. hoUviensis, G. brunnea, G ciineiforma, G. ejuncida, G. extra, G. gentri/i, G. latkephala, G. ilimaui, G. nigrivitrea, G. pilosa, G. siapensis, G. stellae, G. transversa and G. volcano. G. ericae is revalidated. In the present work sixteen names have been synonymized, and a morphological synopisis of the genus is given. The species range from southern U.S. to northern Argentina, but most are found in Central America and tropical South American forests. The evolutionary history of the genus is studied using phylogenetic systematics criteria. Six species groups are recognized: striatula group (22 species); niiniita group (9 species); rastrata group (14 species); sulcata group (8 species); nwrdax group (17 species), and the concinna group (3 species). A key for the identification of workers is presented. Illustrations of morphological features supplement the key and descrip- tions. INTRODUCTION The genus Gnamptogem/s Roger is a di- verse group of ponerines with a distribu- tion that covers the Neotropics, Indoma- laysia and parts of the Nearctic and Aus- tralia. The genus as treated here is mostly the result of extensive synonymy by Brown (1958) in his revision of the Tribe Ectatommini. In this treatment, however, the minuta group species (= Alfarin Forel group of previous authors), are excluded and treated elsewhere (Brandao and Latt- ke 1990; Lattke 1992). Since Brown (1959), research on Gnamptogem/s has consisted mostly of the description of new forms and scattered bits of ecological informa- tion. The only known fossils of the genus are twc^ species reported from Dominican Amber by Baroni Urbani (1980). Lattke (1990) reviews the Venezuelan species and their natural history. Due to the growing number of specimens in collections, and the additional information published c^r accumulated in notes, a revision of the New World species seemed desirable. MATERIALS AND METHODS The revision is based primarily on worker morphology and the species have been defined conservatively because of considerable interpopulation variation. The criteria for judging allopatric popula- tions as conspecific are as in Ward (1984: 131): if they are at least as dissimilar as closely related, sympatric species and are not connected by known intermediate forms. For a phylogenetic analysis, the ge- nus Ectatomma Fr. Smith is used as an out- group in order to help determine the char- acter state polarities. This genus is consid- ered very close to Gnamptogemjs and pre- sents a greater number of primitive character states than either Gnamptogemjs or Rhytidoponera Mayr (Lattke 1994). Spe- cies groups were determined and then compared using a character matrix (Table 1) and Farris's Hennig '86 v. 1.5 phyloge- netic inference program. The "ie-" option was used to generate the most parsimo- nious tree (Fig. 19) and a nelsen consensus 138 Journal of Hymenoptera Research Table 1. Character state matrix 123456789 0123456789 012345 Ectatomma 000000000 0000000000 000000 LEMQ G. haenschi 110001101 0010000000 012110 G. concinna 100011111 1110000001 000001 G. schmitti 100001021 1010000011 001010 strigata sbgp. 010010020 0000011201 000100 porcata sbgp. 000010020 0000011201 000100 LNKD striatuln sbgp. 000010020 0000011200 000100 moniax group 111111121 111-011102 000011 MCZC sulcata group 110111121 1111011122 001011 rastrata group 110011120 0010111202 000010 miuiita group 100011000 0010010001 110110 MHNG MIZA tree calculated from the 13 most parsi- moniuos trees calculated by the "ie" op- tion. Due to logistical difficulties, it was not possible to examine some types, this is especially true for European specimens. Thus this revision depends, in such cases, upon Brown's concept of the species in- volved, either through examining species determined by him during the course of his studies, reading the notes he took while visiting European collections in 1963 or from his 1958 revision. When type material was examined it is followed by a "T" in the species list. Specimens were examined from or de- posited in the following collections: BMNH The Natural History Museum, London, U.K. CMLB Coleccion Martha Lucia Baena, Cali, Colombia CASC California Academy of Sciences, San Francisco, California, U.S.A. CFFC Coleccion Fernando Fernandez, Bogota, Colombia CKWJ Coleccion Klaus Werner Jaffe, Universidad Simon Bolivar, Ca- racas, Venezuela CNIC Cornell University Insect Col- lection, Ithaca, New York, U.S.A. CUNB Coleccion Universidad Nacion- al, Bogota, Colombia JTLC John T. Longino Collection, Ev- ergreen College, Olympia, Wash- ington, U.S.A. LACM Los Angeles County Museum of Natural History, Los Ange- les, California, U.S.A. Lyman Entomological Muse- um, McGill University, Ste. Anne de Bellevue, Quebec, Canada. Landessammlungen fiAr Natur- kunde, Karlsruhe, Germany Museum of Comparative Zool- ogy, Harvard University, Cam- bridge, Massachusetts, U.S.A. Museum d'Histoire naturelle, Geneva, Switzerland. Museo del Instituto de Zoologia Agricola, Universidad Central de Venezuela, Maracay, Vene- zuela MUSP Museu Universidade de Sao Paulo, Sao Paulo, Brazil NHMW Naturhistorisches Museum, Vi- enna, Austria. PSWC Philip S. Ward Collection, Uni- versity of California, Davis, California, U.S.A. USNM United States National Muse- um, Washington, D.C., U.S.A. WPMC William P. MacKay Collection, University of Texas, El Paso, Texas, U.S.A. Measurements and Indices Morphological measurements were made at magnifications of up to 100 X with an ocular micrometer on a Wild M5 ster- eoscope. Since ratios derived from the measurements are conveniently scaled in- dices by themselves, the orthodox proce- dure of multiplying them by 100 was not followed. The measurements follow those of Lattke (1990). HL Head length: midline length of head proper, measured in full-face (dorsal) view, from the anterior clypeal margin to the midpoint of a line drawn across the vertexal margin. ML Mandibular length: length of Volume 4, 1995 139 closed mandibles, from the anteri- or clypeal margin to the apices of the mandibles, measured in the same plane as HL. SL Scape length: length of the first an- tennal segment, excluding the neck and basal condyle. ED Eye Diameter: diameter of eye, measured along its long axis in lat- eral view HW Head width: maximum width of head, measured in full-face (dor- sal = frontal) view, excluding the eyes. WL Weber's length of the mesosoma (alitrunk): diagonal length mea- sured in lateral view, from the an- terior margin of the pronotum (ex- cluding collar) to the posterior ex- tremity of the metapleural lobe. CI Cephalic index: HW/HL MI Mandibular index: ML/HW SI Scape index: SL/HW OI Ocular index: ED/HW SSC Scape setal count: the number of standing hairs (not pubescence), i.e. those forming an angle of 45° or more with the cuticular surface, visible in outline on the upper sur- face of the scape, with the line of view at right angles to the plane of funicular inflexion. New World Gnamptogenys Roger Generic Synopsis (Worker). — Sculpture consisting mostly of even parallel costae, costulae or strigae, occasionally rugose; distinct dorsomedian cephalic carinae usually absent, at most never extending more than half-way between clypeus and vertex; eyes at midlength of head or pos- terad, surrounded by a fine groove; lobes of frontal carinae broadly convex and par- tially raised, incompletely covering anten- nal condyles (except miniitn group); fun- niculus filiform or incrassate, never with a distinct club; anterior clypeal border with narrow lamella of variable length; mese- pisternum anteroventrally bordered by narrow lamella; pronotum unarmed, without protuberances; propodeal spira- cles round or slightly ovoid, never slit- like; transverse sutures and grooves on mesosomal dorsum present or absent; mesonotum never prominently convex and bulging; anterior prostemal process bidentate; anterior mesepisternal process produced as thin triangular lobe with pointed or bluntly pointed apex; metepis- ternum with deep posterior cleft for peti- olar insertion and continuous with open metacoxal fossae; metepisternal process located anterad of cleft and of variable de- velopment; metacoxal dorsum usually with denticle, lobe or tubercle, absent in some species; helcium protruding medi- anly on anterior postpetiolar face; tergite of helcium much larger than sternite; for- tibial apex lacking stout moveable setae; outer border of foretarsal comb (opposite calcar) usually with single prominent seta; meso- and metatibial spurs one or two (weakly developed), barbulate or simple; empoida lacking. In most species the pre- tarsal claws are bidentate on all legs, and the median tooth may vary in its position among species, and the claws may not necessarily be alike on all legs. In small species the claws may be hard to observe. Malpighian tubule number six (Brown 1988). The genus is so diverse that char- acterizing it can be difficult. The following characters can be considered synapomor- phies of the genus which distinguish it from its closest relatives: Ectatomnia and Rhi/tidoponera: the single stout moveable seta on the foretibial apex and a spine or tubercle on the metacoxal dorsal surface. This former trait is lacking in some spe- cies. The following combination of char- acters are synapomorhic for the genus if one excludes the problem species dis- cussed further on in the text: inconspicu- ous metanotum, disappearance of the me- dian cephalic carina, lack of row of stout setae on foretarsal base opposite the strig- il, leaving only one seta. Characters. — The choice of characters for 140 Journal of Hymenoptera Research 1 8 Figs. 1-8. Lateral view of mandibles and anterior section of head of Ectatomma and Gnmnpiogemjs. Transverse section of mandible depicted in black with chewing border to the left and frontal surface below. 1, E. niidum; 2, G. porcata; 3, G. moellerius; 4, G. liaenschi; 5, G. triangularis; 6, G. mordax; 7, G. tortuolosa; 8, G. concinna. use is complicated by convergence in sev- eral instances. Some character states con- sidered apomorphic have apparently been reached independently by one or two spe- cies in different lineages. Examples are the smooth and shining scapes of G. rastrata and of G. mordax, or the conspicuous me- tanotal groove of G. bisulca and G. brimnea. There are also some apparent reversals, as well as retention of plesiomorphies pres- ent in the outgroup. But these cases are few, isolated and do not represent the ma- jor trends of their respective species groups. Therefore, they were not taken into account for defining group character states. Mandibular shape is an apparent useful indicator of phylogenetic trends, with the general direction being from ple- siomorphic triangular mandibles to apo- morphic forms as subtriangular or subfal- cate. Subtriangular mandibles have been independently derived in the haenschi, sul- cata, and mordax groups, as indicated by other mandibular traits such as the config- uration of the chewing border, sculpture and cross section. In the last two groups (Figs. 6-8), the mandibular cross section reveals thick, robust mandibles as op- posed to the slim cuneiform section of G. haenschi, which is the plesiomorphic state Volume 4, 1995 141 11 Figs. 9-14. Lateral view of propodeum of Ectatomma and Gnamptogenys. 9, E. lugeus; 10, £. quadridens; 11, E. tuberculatum; 12, G. bispinosa; 13, G. perspicax; 14, G. triangularis. exhibited by the outgroup and the striatula and rastrata groups (Figs. 1-5). The sub- petiolar process in its primitive state is tri- angular, lobe-like and anteriorly protrud- ing. Modification has been towards a sub- quadrate shape, but some species, such as G. striatula, exhibit both conditions, and in the derived mordax subgroup the triangu- lar subpetiolar porocess is probably a re- versal. Some characters are shared by the out- group and certain groups or subgroups of Gnamptogenys. Two of these are particu- larly important: the propodeal denticles and the setae on the foretarsal base. The 16 Figs. 15-16. Outer lateral view of apex of tibia and base of tarsus of foreleg. 15, typical configuration in Ectatomma with single calcaria on tarsal apex and row of setae in foretarsal concavity; 16, same in most Gnamptogenys with only one stout seta on foretarsal concavity. propodeal denticles of Ectatomma may be absent in some species and when present, they are quite modest, usually lobe-like and directed dorsally. The declivitous pro- podeal face runs evenly and directly to the denticular apex (Figs. 9-11). The denticles in the rastrata group are stout and inclined posterad, and their base forms a more abrupt separation from the propodeal de- clivity (Figs. 12-14). With the exception of a few highly derived species, propodeal denticles are lacking in Rhytidoponcra and the striatula group of Gnamptogenys. This suggests that the presence of propodeal denticles is probably independently de- rived in Ectatomma and Gnamptogenys and consequently was lacking in the ancestral stock of this group of genera. For this rea- son this state has been given value 0 in the character matrix for Ectatomma. The base of the foretarsus, in the concavity oppos- ing the strigil, has an outer row of stout setae in Ectatomma, but only one promi- nent seta in most Gnamptogenys (Figs. 15- 16). The exceptions are all the minuta- group species besides three other species, and each of these exceptions are unique within the genus as will be discussed later on. The brief anterior peduncle of the pet- iole, typical of Ectatomma is considered plesiomorphic for the porcata subgroup. 142 Journal of Hymenoptera Research 0^ 17 18 Figs. 17-18. Ventral view of the anterior postpetiolar process. 17: a, G. triajjgularis; b, G. haenschi; c, G. striatiila. 18, G. sulcata. but autapomorphic for the mordax sub- group. The following characters were taken into account for all the species groups or individual species in the matrix, but have been restricted to the minuta sub- group of the minuta group (Lattke 1991). Characters 8, 17, 18 were considered non- additive. The following character states were taken into account: 1. Head shape in frontal view: 0= wider posterad than anterad, l=subquadrate/ other. 2. Funiculus shape: 0 = filiform to subfi- liform, l = incrassate. 3. Scape length: 0= surpassing vertexal border (Figs. 39, 41), l = does not surpass, at most reaching the vertexal border (Fig. 43). 4. Scape sculpture: 0= scapes with ap- preciable sculpture, 1 = smooth and shin- ing. The aforementioned sculpture usually consists of punctulae, punctae or striae. 5. Development of the clypeal lamella: 0= poorly developed, very short, l=well developed. This character refers to how far anterad the lamella extends. In poorly developed cases the longitudinal cephalic costulation is not clearly delimited from the lamella. 6. Shape of anterior clypeal margin: 0= convex clypeus (Figs. 33, 37, 39, 41), 1 = straight/concave (Figs. 34-36, 62, 64). 7. Shape of lateral extremities of ante- rior clypeal lamella: O=softly curves (Figs. 33, 37, 39, 41), 1= angular /pointed (Figs. 34-35). 8. Presence of median clypeal carinae: 0= present, 1= variable, 2 = absent or indis- tinct. 9. Mandibular shape: 0 = triangular (Figs. 39, 41), l=subtriangular/subfalcate (Figs. 43, 54, 69, 62). 10. Shape of cross section at apical one third of mandible: 0 = thin and cuneiform (Figs. 2-5), l=wide and robust (Figs. 6, 7). 11. Sculpture of mandibular dorsum: 0=striae/rugulae (Fig. 54), l=smooth and shining (Fig. 43). 12. Development of promesonotal su- ture: 0 = deeply impressed, cleaving sculp- ture (Figs. 46, 49, 52), 1= weakly im- pressed, effaced (Fig. 55). 13. Development of metanotal groove: 0= deeply impressed (Fig. 55), l=weak to effaced (Figs. 46, 49, 52). 14. Development of propodeal denti- cles: 0= absent (Figs. 46, 49, 52), 1= present (Figs. 12-14). 15. Position of spiracle on propodeal side, in lateral view: 0 = anterad, well sep- arated from declivitous propodeal face, 1= posterad, close to declivitous propo- deal face. 16. Presence of setae on foretarsal base opposite the strigil: O = row of stout setae (Fig. 15), l=one prominent seta only (Fig. 16). 17. Development of dorsal metacoxal Volume 4, 1995 143 process: 0 = absent, 1 = variable, 2= well de- veloped (Figs. 46, 49). 18. Shape of the metasternal process: 0 = stout denticles or tubercles, l=acicular, straight denticles, 2=acicular, arched den- ticles. 19. Shape of petiolar node: 0=erect and high node (Figs. 46, 49, 52), l=high and posterad, 2 = subcylindrical/other (Figs. 61, 63, 65). 20. Petiolar spiracles: 0 = directed obliquely; 1= directed ventrally. 21. Petiolar spiracles: 0 = -at the same level as the surrounding integument; l=in a distinct depression. 22. Shape of postpetiolar process in lat- eral view: O=broad (Fig. 17), l=bluntly pointed (Fig. 18). 23. A brief median carina on the dor- sum of the postpetiolar process: 0= absent, 1= present. This is a low carinae that starts at the anterior margin of the process and briefly extends posterad. 24. Abdominal segment IV: 0=with a ventral stridulitrum, 1= without a ventral stridulitrum. Usually recognizable as a longitudinal band that refracts light into rainbow colors. 25. Shape of the fourth abdominal seg- ment: 0 = ventrally reflexed (Fig. 55), l=straight (Fig. 44). Species Groups. — Within the New World Gnamptogenys six species groups are de- fined here: striatula, sulcata, concinna, min- uta, mordax and rastrata. The internal clas- sification of the genus is depicted in Table 2. The characters used in defining the fol- lowing informal groupings are diverse, thus they may or may not be synapomor- phies. A question mark following the spe- cies name means that its position within the group is not clear. \. striatula group: mandibles triangular with rugulae or striae on frontal surface; convex anterior clypeal margin; scapes usually surpassing vertexal margin, sculp- turing punctate or vermiculate, never smooth and shining; eyes slightly behind cephalic midlength; head wider posterad Table 2. Internal classification of Gnamptogenys (except minuta group) striatula group porcata subgroup: acuta, brunnea, ejuncida, extra, gentryi, gracilis, nigrivitrea, pilosa, pleurodon, porcata striatula subgroup: ammophila, moelleri, striatula strigata subgroup strigata complex: andina, bisulca, ilimani, pitti- eri, strigata haytiana complex: haytiana, inina, reichenspergi, rclicta rastrata group rastrata subgroup: hispinosa complex: bispinosa, perspicax rastrata complex: cuneiforma, ingeborgae, lanei, lineolata, lucaris, mecotyle, meiiozzi, rastrata, triangu- laris banksi subgroup: banksi complex: laticephala, mediatrix semiferox complex: semiferox sulcata group sulcata subgroup: fernandezi complex: fernandezi sulcata complex: acuminata, curvoclypeata, tor- tuolosa ericae subgroup: ericae complex: ericae lucaris complex: lucaris siapensis volcano complex: volcano mordax group regularis subgroup: hartmaui complex: bruchi, hartmani, transversa regularis complex: horni, nana, pristina, regularis, rimulosa ainndata complex: annulata, kempfi mordax subgroup: continua, boliviensis, interrup- ta, mordax, stellae alfaroi subgroup: alfaroi levinates subgroup: levitjates concinna group concinna, haenschi, schmitti than anterad; propodeal spiracle close to declivity; anterolateral propodeal declivity without spines or tubercles; anterior pros- ternal process broadly concave medianly; metacoxal dorsum always with denticle or lobe; high petiolar node; relatively wide anteroventral postpetiolar process; second 144 Journal of Hymenoptera Research gastric segment ventrally arched. Most view; anterior clypeal margin usually seem to be generalist predators, but noth- straight; mandibular front usually striate ing is yet known about the diet of the stri- or rugulose, sometimes smooth; scapes gata subgroup species. usually surpassing vertex, sometimes with Three subgroups are recognizable: 1) longitudinal rugulae, vermiculate or striatula subgroup: petiolar node erect and smooth and shining; promesonotal suture sessile; subpetiolar process variably feebly impressed to absent, never totally shaped, either subquadrate or triangular dividing sculpture; metanotal suture well and projecting anterad in lateral view. 2) impressed; propodeum usually armed porcata subgroup: petiolar node posterior- with denticles or spines; petiolar node ly inclined, with short anterior peduncle; low; subpetiolar process shape variable, subpetiolar process lobe-like, projecting usually projecting anterad but sometimes anterad in lateral view. 3) strigata sub- subquadrate; metacoxal teeth always pres- group: petiolar node variably shaped; sub- ent, usually acicular; second gastric seg- petiolar process subquadrate with a cu- ment ventrally arched. The ants of this neiform ventral edge (Fig. 30) as opposed group are specialized millipede predators to the uniformly parallel sides of the other and can be subdivided into two sub- striatula group species (Fig. 29). Many of groups. 1) rastrata subgroup: mandibles the species have their propodeal spiracles triangular. Within the subgroup two com- elevated on small prominences. Most spe- plexes are defined: a) bispinosa complex: cies are smaller in size than those of other large species, anterior clypeal margin con- subgroups and all are cryptobiotic leaflit- vex, propodeal teeth above level of spira- ter inhabitats. Two species complexes can cles; anterior lobe of subpetiolar process be pointed out in this subgroup: a) strigata prominent, b) rastrata complex: smaller complex: cephalic vertex sculptured; no species; anterior clypeal margin straight; distinct propodeal lobes; petiolar node not propodeal teeth at same level as spiracles; parallel-sided in lateral view, anterior subpetiolar process subquadrate. 2) banksi margin usually convex (Figs. 40, 57). All subgroup: mandibles subfalcate to falcate, species inhabitat cloud forests, mostly of Two species complexes are defined, a) the Andes, with the exception of one banksi complex: metacoxae and propo- found in the Cordillera de la Costa of Ven- deum with denticles, b) semiferox complex: ezuela. b) haytiana complex: cephalic ver- metacoxae and propodeum lacking denti- tex mostly smooth and shining; small pro- cles. In a study of the sting apparatus of podeal lobes present; petiolar node several species of Gnamptogenys Kugler strongly compressed with subparallel an- (1991) grouped triangularis and bispinosa, terior and posterior faces (Fig. 21); some the only members of the rastrata group as species have modified sculpturing from defined above, into one group, the usual costulate patterns of the genus. IIL mordax group: antennal scapes usu- Most species inhabitat the Amazon-Ori- ally do not reach vertex and are com- noco river basin, except for one species pressed, smooth and shining; mandibles found on Hispaniola Island. Kugler (1991) subtriangular to subfalcate with front obtained similar results studying the sting smooth and shining; head subquadrate to apparatus of brunnea, nr. strigata, porcata, elongate; mesosoma mostly parallel sided; moelleri and gracilis. They were the only small propodeal lobes frequently present, members of the striatula group included, denticles absent (except one species); me- as defined above, and were grouped to- tacoxal denticles or lobes frequently lack- gether. ing; second gastric segment straight, with- II. rastrata group: head subquadrate or out ventral arching. Considerable inter- wider anterad than posterad in frontal population variability in the average size Volume 4, 1995 145 is observed in species of this group. The group can be divided into 4 subgroups: 1) regulciris subgroup: mandibles subtrian- gular; anterior clypeal border straight and usually with laterally acutely pointed la- mella; no transverse dorsal mesosomal su- tures; petiolar node lacking anterior pe- duncle; subpetiolar process subquadrate. Within this subgroup three species com- plexes are recognizable, a) hartfiiaiii com- plex: mandibles triangular to subtriangu- lar, frontal surface partially sculptured; vertical sculpturing on declivitous propo- deal face; b) regiilaris complex: mandibles subtriangular, their frontal sides smooth and shining; vertical sculpturing on de- clivitous propodeal face; c) annulata com- plex: mandibles subtriangular, front smooth and shining; horizontal sculptur- ing on propodeal declivity. 2) mordax sub- group: mandibles subfalcate, frontal sur- face smooth and shining; anterior clypeal margin concave with rounded lamellar sides; well impressed metanotal groove; petiolar node slightly pedunculate; sub- petiolar process projects anterad. 3) alfaroi subgroup: mandibles triangular, head prominently elongate; denticles on pro- podeum. The elongate head and armed propodeum makes this subgroup quite distinct from the others. 4) levinates sub- group: subtriangular mandibles, frontal surface sculptured; scapes surpassing ver- tex; subpetiolar process subquadrate. Its clypeus projects anterad more than in any extant species of the group. IV. sulcata group: mandibles subtrian- gular, front smooth and shining; head su- bquadrate; anterior clypeal border usually straight; lamella usually laterally angular; scapes smooth and shining; metacoxal denticles wanting in some species; usually no transverse mesosomal sutures; petiolar node low, lacking peduncle; metasternal process acicular and arched; second gas- tric segment relatively straight; anteroven- tral postpetiolar process relatively narrow and bluntly pointed, in contrast with wid- er process of other species. Two sub- groups can be pointed out. 1) sulcata sub- group: posterior face of petiolar node with horizontal costulae. Two species complex- es are recognizable: a) fernandezi complex: mandibles triangular; clypeal lamella con- vex, b) tornata complex: mandibles sub- triangular; clypeal lamella straight. 2) sul- cata subgroup: posterior node face with longitudinal costulae. This subgroup is made up of three complexes, a) ericae com- plex: clypeal lamella laterally acutely an- gulate; mandibles subtriangular. b) lucaris complex: clypeal lamella laterally round- ed; mandibles subtriangular. c) volcano complex: clypeal lamella laterally obtusely angulate; mandibles triangular. V. concinna group: head subquadrate; scapes surpassing vertex in frontal view; median clypeal carina frequently present; body sculpture densely striate; small pit frequently situated medianly on weakly impressed promesonotal suture; metaster- nal process slender to acicular; row of stout setae on base of foretarsus opposite strigil present; petiolar node with no an- terior peduncle; metacoxae and propo- deum unarmed. All three members of the group are large species. This group is nev- ertheless heterogeneous and its three members were considered individually in the matrix. VI. minuta group: head subquadrate; frontal carinae broadly expanded laterad; row of stout setae on base of foretarsus opposite strigil present; petiolar spiracles facing directly ventrad and sunken within a pit. Phylogenetics. — The possible evolution- ary history of the six species groups is de- picted in Fig. 19. Fig. 19 was computed by the "ie-" option of Hennig '86. Figure 20 is a nelsen consensus tree calculated from 13 trees generated by the "ie" option of Henning '86. The striatula group subgroups are con- sidered a sister group to the rest of the Gnamptogem/s species considered. The sul- cata and mordax groups are the most de- rived and are considered sister groups. 146 Journal of Hymenoptera Research 19 Ectatomma strigata subgroup porcata subgroup striatula subgroup minuta group rastrata group G. schmitti G. haenschi G. concinna mordax group sulcata group 20 Ectatomma strigata subgroup porcata subgroup striatula subgroup G. haenschi G. concinna G. schmitti rastrata group minuta group mordax group sulcata group Figs. 19-20. Cladograms generated from the data in Table 1 by Hennig '86. 19, most parsimonious tree found by the "ie-" option, length 51, ci 58. 20, Nelsen consensus tree generated from the 13 most parsimonious trees found by the "ie" option, length 51, ci 58. These results closely parallel Kugler's (1991) findings based upon the sting ap- paratus of several Gnamptogeni/s species. Within each of these groups we find spe- cies that apparently represent the transi- tion from striatida-\ike ancestors and thus give additional support to the program re- sults: the hartmani complexes of the mordax group, and the volcano and fernandezi com- plexes of the sulcata group. The rastrata group is probably closer to the striatula group than to the mordax or sulcata groups on account of characters such as triangular mandibles, long and usually sculptured scapes, the convex clypeal lamella of the bispinosa complex and the well developed metacoxal tooth. The consensus tree rec- ognizes two major groupings within the genus: striatula, and the rest. The "rest" make up a six-branched polytomy. Clearly further studies will be needed in the fu- ture in order to gain a better picture of the phylogenetic history of Gnamptogenys. The concinna group is probably not monophyletic. It is diverse in mandibular shape, and in other details such as the de- velopment of the clypeal lamella or me- dian clypeal carina. These traits as well as vestigial promesonotal suture are unusual remnants of plesiomorphies when com- pared with most other members of the ge- nus. G. haenschi retains more plesiomor- Volume 4, 1995 147 phies than the other two species and could be considered closer to the rootstock of the group. All are apparently specialized, ei- ther in feeding or nesting habits. This group opens a number of questions which can not be answered to satisfaction at present. Are the unarmed metacoxae ple- siomorphic or apomorphic for this group? Could they represent relicts of an other- wise extinct lineage (or lineages) in Gnamptogenys history? Interesting paral- lels can be found in the shallow areolate ground sculpture and well-developed ver- texal lobes of mimita subgroup members and those of several Old World species groups. A comparative study with Old World species could shed new light on the phylogeny of the genus. The striate sculp- ture of the concinna group can also be found, in a lesser degree, in the suniilans complex. The origins of Ectatommini can be traced to the late Cretaceous to early Ter- tiary, at least after the separation of Africa from Gondwana (80-90 Ma), but before the separation of South America from Pa- leoantarctica (36-41 Ma) (Lattke 1994). Given the presence of Gnamptogenys in the Indomalaysian Region and their absence in Africa, the origin of the genus also ex- tends at least to this time period and geo- graphically could be related to southern South America -Palaeoantarctica. The spread of the ancestral fauna from South America to Central America probably took place during the mid Tertiary (30-38 Ma) as the Panama land bridge devel- oped. Chances for an earlier exchange through possible connections between parts of the Greater Antilles and South America could have existed during the late Cretaceous (Donelley 1988:26), but the geological history of the Caribbean region is still fraught with controversy (Rull 1989). The mordax group fossils from Do- minican Amber show that by the late Oli- gocene to early Miocene (15-25 Ma), di- versification of the American Gnamptogen- ys fauna was well underway. Potential ter- restrial connections between Central America and the Greater Antilles could have existed during the Cenozoic via the presently submerged Nicaraguan rise which lays between Honduras and Jamai- ca (Donnelly 1988:28). It offered possibili- ties for dispersal until its subsidence dur- ing the middle Cenozoic. The presence of the Hispaniolan endemic species G. hayti- ana, G. schmitti, and G. semiferox lend sup- port to such a connection and to the di- versification of Gnamptogenys lineages some 26-34 Ma. Ecology. — The ecology of Gnamptogenys found in Venezuela is treated in Lattke (1990) and should be consulted for more information. With the exception of a few additional tidbits of information little is added here. Generally most species are dwellers of mesic, forested habitats from Texas and Louisiana to northern Argenti- na. Nests are generally constructed in de- composing wood on the ground and tend to be small, rarely exceeding 500 adults. A few species exhibit polygyny. The striatula group species are mostly generalist predators, but other species groups have developed varying degrees of specialization. This is taken to an extreme by the rastrata group species which prey only on certain diplopods. Beetles are the favorite of many members of other groups, and some species such as G. hart- manni and G. horni have also become fond of other ants as prey. New World Gnamptogenys: A Synonymic List When type material of a determined species was personally studied a "T" fol- lows the distribution, plus the acronym of the collection where the specimen is de- posited. Type material, in this case, in- cludes specimens labelled either holotype, paratype, cotype, type series, or syntype. acuniiiiaia Emery 1869. Tropical S. America. acuta (Brown 1957a). Colombia-Bolivia. T:MCZC alfaroi Emery 1894. Costa Rica-Ecuador aitimopliila Lattke 1990. se Venezuela. T:MIZA 148 Journal of Hymenoptera Research andina Lattke new sp. sw Colombia-Ecuador. annulata Mayr 1887. Costa Rica-tropical S. America. hanksi (Wheeler 1930). Panama-Ecuador. T:MCZC bispinosa (Emery 1890). Costa Rica-Colombia. bisiilca Kempf y Brown 1968. Costa Rica-Ecuador. T: MCZC bufonis (Mann 1926). s Mexico-Nicaragua. boliviensis Lattke, new species, n Bolivia. briichi (Santschi 1922). Argentina. T:MCZC brunnea Lattke, new species, nw Colombia. caelata Kempf 1967. Colombia-se Brazil. = soror Kempf and Brown 1968 concinna (F. Smith 1858). Mexico-tropical S. America. T:BMNH = romani Wheeler 1923. = semicircularis Borgmeier 1929. = conica Borgmeier 1929. continiia Mayr 1887 Mexico-tropical S. America. = panamensis Santschi 1931 = exarata Emery, 1901. new synonymy. cuneiforma Lattke, new species. Panama. curvoclypeata Lattke 1990. Colombia-Venezuela. T: MCZC ejuncida Lattke, new species, s Colombia. ericae Forel 1912, revalidated, n S. America. = bufonum Weber 1938, new synonymy. extra Lattke, new species. Colombia-Ecuador. falcifera Kempf 1967. n S. America. fernandezi Lattke 1990. Colombia-Ecuador. T:MIZA fieldi Lattke 1990. n central Venezuela. T:MIZA gentryi Lattke, new species, sw Colombia. gracilis (Santschi 1929). Guiana Shield. T:MCZC laticephala Lattke, new species. Ecuador. haenschi Emery 1902. Costa Rica-tropical S. America. hartmanni (Wheeler 1915). s United States-n S. Amer- ica. T:MCZC = nigrifrons Borgmeier, 1949. new synonymy. T: MUSP = turmalis Kempf y Brown, 1968. new synonymy. T:MCZC haytiana (Wheeler and Mann 1914). Hispaniola. T: MCZC horni Santschi 1929. Panama-Bolivia. T:MCZC ilimani Lattke, new species. Bolivian Andes. ingeborgae Brown 1992. e Colombia. T:MCZC interrupta Mayr 1887. Mexico-tropical S. America, Ja- maica. kempfi Lenko 1964. Brazil-Peru. lanei Kempf 1960. Brazilian Amazonas. levinates Baroni Urbani 1980. Dominican Amber. lineolata Brown 1992. Hispaniola. T:MCZC lucaris Kempf 1968. s Brazil. mecotyle Brown 1958. Panama-tropical S. America. T: MCZC mediatrix Brown 1958. Amazon drainage. T:MCZC menozzi Borgmeier 1948. s Brazil. TiMUSP = schubarti (Borgmeier, 1948) new synonymy. T: MUSP mina (Brown 1957). w S. America. T:MCZC minuta (Emery 1896). Belize-Brazil. = bufonis (Mann 1922) = emeryi (Forel 1901) = scabrosus (Mann 1922) = mus (Santschi 1931) = panamensis (Weber 1940) = carinata (Weber 1940) = pneodonax Kempf 1968 moelleri (Forel 1912). Tropical S. America. = splendens (Santschi 1929) = concinna (Santschi 1929) = teffensis (Santschi 1929) new synonymy. T:MCZC niordax (F. Smith 1858). Mexico-tropical S. America. T:BMNH = nodosa (Latreille 1802) = purensis Forel 1912 = sebastiani Borgmeier 1937 nana Kempf 1960. Brazil. nigrivitrea Lattke, new species, sw Colombia. perspicax Kempf and Brown 1970. Colombia-Ecuador. T:MCZC petiscapa Lattke 1990. ne Venezuela. T:MIZA pilosa Lattke, new species, sw Colombia. pittieri Lattke 1990. n Central Venezuela. T:MIZA pleurodon (Emery 1896) Tropical S. America. = emeryi (Santschi 1929). = vidua (Santschi 1929). porcata (Emery 1896). Honduras-Bolivia. = magnifica (Santschi 1921). new synonymy. pristina Baroni Urbani 1980. Dominican Amber. rastrata (Mayr 1866). s Brazil. = trigona Emery 1905. new synonymy. regularis Mayr 1870. Mexico-Paraguay. = splendida Pergande 1895. T:CASC = fiebrigi Forel 1909. = arcuata (Santschi 1929). reichenspergi (Santschi 1929). Amazon-Orinoco drain- age. relicta (Mann 1916). Amazon-Orinoco drainage. T: USNM rimulosa (Roger 1861). s Brazil. schmitti (Forel 1901). Hispaniola. = minor (Wheeler 1936). T:MCZC semiferox Brown 1958. Hispaniola. T:MCZC siapensis Lattke, new species, s. Venezuela. simulans (Emery 1896). Costa Rica. T:USNM stellae Lattke, new species. Costa Rica. striatula Mayr 1883. Mexico-Argentina, Caribbean. T: BMNH = curtula (Emery 1896) new synonymy. = stolli (Forel 1899) new synonymy. = brasiliensis (Emery, 1902). = angustiloba (Forel 1908). = paulina (Forel 1908). = simplicoides (Forel 1908) new synonymy. = pernambucana (Santschi 1929). = calcarata (Santschi 1929). Volume 4, 1995 149 = antillana (Santschi 1929). T:MCZC = hybrida (Santschi 1929). = mayri (Santschi 1929). = isthmica (Santschi 1929). = recta (Santschi 1929). T:USNM = rustica (Santschi 1929) new synonymy. = wheeleri (Santschi 1929) new synonymy. T:MCZC = arcuata (Santschi 1929) new synonymy. = wasmanni (Santschi 1929) new synonymy. striolata (Borgmeier 1957). se Brazil. T:MUSP strigata (Norton 1871). Mexico-Colombia. = simplex (Emery 1896). sulcata (F. Smith 1858). Mexico-tropical S. America. T: BMNH = tornata (Roger 1861). new synonymy. = lineata Mayr 1870. = cearensis Forel 1912. = nitens Mann 1916. TiUSNM = ypirangensis Borgmeier 1928. tortuolosa (F. Smith 1858). Amazon-Orinoco drainage. T:BMNH = quitensis Forel 1920. transversa Lattke, new species. Panama. triangularis Mayr 1887. Panama-Argentina; se United States. = richteri (Forel 1920) = aculeaticoxae (Santschi 1921) new synonymy. volcano Lattke, new species. Costa Rica. vriesi Brandao & Lattke, 1990. Ecuador. KEY TO THE WORKERS OF THE NEW WORLD SPECIES OF GNAMPTOGENYS. Note: In some species the difference between the promesonotal suture and the metanotal groove may be hard to establish when only one of the two is present, which is the usual case. In a dorsal view the lateral margins of the pronotum converge posterad and at the end of the convergence, when the margins more or less straighten out, the promesonotal suture is usually slightly arched anteriorly, while the metanotal groove is usually straight. 1. Petiolar spiracle directed ventrally and within a depression .... mn;Hfrt group (see Lattke 1991) - Petiolar spiracle directed anteroventrally and not depressed; Pronotum separated from mesonotum by very distinct suture which completely cuts dorsal sculpture 2 - Petiolar spiracle directed ventrolaterally and not depressed; Promesonotal suture absent to well impressed, but never completely cuts sculpture 23 2. Petiolar in lateral view with subparallel anterior and posterior margins sharply set off at right angles to relatively brief dorsal margin (Fig. 21); vertex usually smooth and shining, occasionally with weak transverse triae 3 - Petiole not as above, in lateral view with anterior margin either broadly curving into dorsal margin or separated by broad angle (Figs. 22-4, 46, 49, 52); vertex strongly sculp- tured 6 3. Mesosomal dorsum with extensive smooth and shining areas 4 - Mesosomal dorsum totally sculptured, usually rugulose or costulate 5 4. Metanotal groove well impressed, breaking sculpture; metacoxae armed; anterolateral lobes on declivitous propodeal face lacking (Amazon-Orinoco drainage) relicta - Metanotal groove absent; metacoxae unarmed; small lobes present on anterolateral de- clivitous propodeal face (Amazon-Orinoco drainage) reichenspergi 5. Postpetiolar dorsum transversely rugulose; body color yellowish (Hispaniola) .... haytiana - Postpetiolar dorsum longitudinally costulate; black to dark brown (w South America) .... mina 6. Mesosomal dorsum with well developed metanotal groove 7 - Metanotal groove absent 9 7. Dorsum of gastric tergites I and II smooth and shining (sw Colombia) gentryi n. sp. - Dorsum of gaster with longitudinal costulae 8 8. Subpetiolar process subquadrate, with posterior angle (Fig. 21; Costa Rica to Ecuador, Andes) bisulca - Subpetiolar process shaped as an anteriorly projecting lobe with no posterior angle Fig. 42) (nw Colombia) brunnea n. sp. 9. Petiolar node in lateral view relatively erect, posterodorsal and anterodorsal angles not differing notably (Fig. 22) 10 - Petiole in lateral view posteriorly inclined, anterior margin joins dorsal margin through broad convexity that contrasts with sharp angle between posterior and dorsal margins (Figs. 23, 24, 46, 49, 52) 12 150 Journal of Hymenoptera Research 10. SL < 1.08 mm 11 - SL > 1.08 mm (tropical S. America) moelleri 11. Declivitous propodeal face with 5-11 longitudinal costulae between spiracles and 1-3 transverse costulae between each spiracle and longitudinal costulae (Fig. 26); decumbent pubescence on scapes sparse, < six decumbent and suberect hairs (Mexico to Argentina, Caribbean) striatula - Declivitous face with 13-16 longitudinal costulae between spiracles and no transverse costulae between each spiracle and longitudinal costulae (Fig. 25); scapes with dense and uniform decumbent pubescence, 1-3 subdecumbent hairs (se Venezuela) ammophila 12. Posterodorsal angle of petiolar node forms a sharp to bluntly pointed apex that overhangs posterior margin (Figs. 23, 24, 46, 49, 52); subpetiolar process in ventral view with uni- formly narrow ventral edge, posteriorly widening only at junction with node (Fig. 29) 13 - Petiolar node not, or slightly overhanging posterior margin (Figs. 40, 57); subpetiolar process in ventral view with cuneiform ventral edge: anteriorly thin, wider posterad with bifurcate apex that forms posterior angle of subquadrate process (Fig. 30) 20 13. Petiolar node with a blunt posterodorsal apex (Figs. 24, 46, 49, 52) 14 - Petiolar node with acute posterodorsal point (Fig. 23; Colombia to Bolivia) acuta 14. Postpetiolar sternum totally costulate or striate; vertex usually with one (occasionally up to 3) transverse costulae next to vertexal carinae 15 - Postpetiolar sternum with a basal median smooth and shining area; vertex with 4-5 trans- verse costulae next to vertexal carinae (Guiana Shield) gracilis 15. SSC < 10 (Fig. 27) 18 - SSC > 10 (Fig. 28) 16 16. Body and legs brown, sculpture costulate to striate; body with abundant decumbent pu- bescence; subpetiolar process lobiform and without acute projection 17 - Body piceous and legs ferruginous; body with abundant pubescence; subpetiolar process usually with an acute anteroventral projection, sometimes blunt, especially in specimens from Central America (Honduras to Bolivia) porcata 17. Femora with abundant erect to suberect hairs; ventral surfaces of meso- and metacoxae with erect hairs; gastric sculpture striate (sw Colombia) pilosa, n. sp. - Femora with decumbent to subdecumbent hairs and few or no erect to suberect hairs; ventral surfaces of meso- and metacoaxe wirthout erect hairs; gastric sculpture costulate (s Colombia) ejuncida, n. sp. 18. Five or more standing hairs on scapes and tibiae 19 - One or no suberect and subdecumbent hairs on scapes and tibiae (Colombia-Ecuador) extra, n. sp. 19. Declivitous propodeal face with transverse costulae; petiolar node with anterior peduncle (Fig. 24; tropical S. America) pleurodon - Declivitous propodeal face with longitudinal costulae; node without peduncle (Fig. 59; sw Colombia) nigrivitrea, n. sp. 20. Postpetiolar sternum with well defined costulae or striae; dorsal propodeal face not no- tably depressed below mesonotum, dorsal and declivitous faces confluent 21 - Postpetiolar sternum with weak irregular rugae; in lateral view dorsal propodeal face separated from mesonotum by notable depression, dorsal and declivitous faces sharply separated by a ridge (n Central Venezuela) pittieri 21. Scapes with moderate to weak pubescence, five or more standing hairs present 22 - Scapes with dense, white subdecumbent pubescence, 0-2 standing hairs present (Bolivian Andes) ilimani n. sp. 22. HW > 0.84, WL > 0.35; Ol < 0.16 (sw Colombia-Ecuador) andina, n. sp. - HW < 0.84, WL < 0.35, OI > 0.16 (Mexico-Colombia) strigata 23. With head in frontal view antennal scapes barely or do not reach the vertexal margin, never beyond, when laid back as straight as possible from insertions 24 - Scapes clearly reach beyond vertexal margin 38 Volume 4, 1995 151 24. Mandibular dorsum with continuous striae or costulae over the basal one-fourth or more of length 25 - Mandibular dorsum entirely smooth and shining, sometimes with scattered punctures 27 25. Distinct median carinae visible on clypeus; eyes small, OI < 0.1; mandibular dorsum entirely and densely striate (Costa Rica-tropical S. America) haenschi - No distinct median carinae on clypeus; eyes larger, OI> 0.2; mandibular dorsum without dense striae 26 26. Mandibular dorsum with costulae on basal one fourth or less, rest smooth and shining; propodeal declivity with longitudinal costulae (s United States-n South America) .... hartmani (in part) - Mandibular dorsum mostly rugulose; propodeal declivity with transverse costulae . . . transversa n. sp. 27. Propodeal dorsum mostly transversely striate 28 - Propodeal dorsum mostly longitudinally costulate or striate, continuous with rest of notal sculpturing 30 28. Head subquadrate, CI > 0.78; body brown to dark brown 29 - Head notably rectangular and elongate, CI< 0.78; body black (Costa Rica-Ecuador) . . . .alfaroi 29. Costulae on mesosomal and petiolar dorsum subopaque; propodeum with small teeth; metacoxal teeth (Brazil and Peru) kempfi - Very fine striae on mesosoma and node; no teeth on propodeum nor on metacoxae (Costa Rica to tropical S. America) attnulata 30. Clypeal lamella medianly straight and laterally acutely angulose; subpetioloar process subquadrate, usually with acute posteroventral tooth 36 - Clypeal lamella usually concave and laterally rounded, never acutely angular (rarely with slight median projection and laterally bluntly angular); subpetiolar process usually pro- jecting anteriorly as subtriangular lobe, sometimes subquadrate, but never with acute posteroventral tooth 31 32. Mandibles subtriangular to subfalcate; metanotal groove well-impressed 32 - Mandibles triangular; metanotal groove absent (Brazil) nana 32. Metacoxa with dorsobasal lobe or tooth 34 - Metacoxa lacking tooth or lobe, at most small swelling or low tubercle present 33 33. Body sculpture striate; gastric tergum 2 smooth and shining; clypeal lamella laterally bluntly angulose, slightly projecting anterad medianly and with small concavity in middle of projection (Fig. 60; Costa Rica) stellae n. sp. - Body sculpture costulate; gastric tergum 2 longitudinally costulate; clypeal lamella lat- erally rounded and medianly concave (Fig. 43; Mexico to tropical S. America, Jamaica) interrupta 34. Metacoxal dorsum with low triangular lobe; HW < 1.12, WL < 1.81 mm 35 - Metacoxal dorsum with high, approximately parallel-sided lobe or tooth; HW usually > 1.12, WL > 1.81 mm (Mexico to tropical S. America) mordax 35. Cephalic dorsum striate; anterolateral lobes of declivitous propodeal face weakly devel- oped (n Bolivia) holiviensis, n. sp. - Cephalic dorsum costulate; anterolateral propodeal lobes well developed (Mexico to trop- ical S. America) continua 36. Metanotal groove weakly impressed, visible only with limited angles of view; subpetiolar process with acute posteroventral tooth; anterolateral angles of declivitous propodeal face bordered by raised carinae that cross over costulate sculpture 37 - Metanotal groove well-impressed, visible with any angle of view; subpetiolar process without acute posterior tooth; declivitous propodeal face lacking carinae different from costulate sculpture (s. Brazil) rimiilosa 37. Declivitous propodeal face with longitudinal costulae that converge posterad; anterior nodal face with transverse costulae; meso-metapleural suture absent or very weakly im- pressed (Panama to Bolivia) homi 152 Journal of Hymenoptera Research - Declivitous propodeal face with longitudinal parallel costulae; anterior nodal face mostly longitudinally costulate, basal transverse costulae frequently present and rarely occupy all; meso-metapleural suture distinctly impressed (Mexico to Paraguay) regularis 38. Mandibles very elongate: falcate or subfalcate, only their apices touching or crossing when closed (Figs. 35, 54) 39 - Mandibles triangular or subtriangular, apical and basal margins separated by rounded or angular basal angle (Figs. 33-34, 36-37) 43 39. Declivitous propodeal face with spines or denticles; metacoxal tooth present 40 - Declivitous propodeal face rounded, no trace of spines or denticles; coxal tooth absent 42 40. Inner basal mandibular border with short triangular tooth that represents basal angle (Fig. 54) 41 - Inner basal mandibular border with broadly convex lobe or flange (Amazon drainage) mediatrix 41. Clypeus with two lobes betwen antennal fossae and anterior lamella, lobe protrude over lamella in front view; mesothoracic spiracles not set in deep depressions (Ecuador) . . laticephala, n. sp. - Clypeus without lobes between antennal fossae and anterior lamella; spiracles of meso- thorax set in deep depressions (Panama to Ecuador) banksi 42. Mandibles long and slender (Fig. 36); dorsal propodeal face evenly and transversely cos- tulate (Hispaniola) schmitti - Mandibles shorter and wider; dorsal propodeal face subvermiculately and longitudinally costulate (Hispaniola) semiferox 43. Mandibles mostly smooth and shining, with scattered punctures, subtriangular to trian- gular, sometimes striae or costulae present on basal one-third or less; usually without transverse sutures on mesosomal dorsum (except concinna) 44 - Mandibles with longitudinal rugulae or striae (sometimes weakly impressed), always triangular, never subtriangular; metanotal suture distinctly impressed 55 44. HW > 1.43; WL > 2.27 mm 45 - HW < 1.43; WL < 2.27 mm 46 45. Sculpture costulate; metacoxae with dorsal tooth; petiolar node ends in blunt angle; black (Amazon-Orinoco drainage) tortuolosa - Sculpture striate; metacoxae without dorsal tooth, with low tubercle at most; petiole ends in acute point or beak; ferruginous (Mexico to tropical S. America) concinna 46. Sculpture striate; anterolateral corners of declivitous propodeal face borderd by carinae that cross costulae; antennal scape when laid back does not surpass vertexal margin by more than one apical width 47 - Sculpture costate; declivitous propodeal face without carinae distinct from costae; scapes surpass vertexal margin by more than an apical width 48 47. Posterior face of petiolar node longitudinally costulate (s United States to n S. America) hartmani (in part) - Posterior nodal face transversely costulate (Argentina) bruchi 48. Posterior face of petiolar node transversely costulate 49 - Posterior face of node longitudinally costulate 51 49. Anterior clypeal lamella laterally acutely angular; mandibles subtriangular, basal and apical margins joined by very broad continuous convexity (Fig. 34) 50 - Clypeal lamella very evenly convex laterad; mandibles triangular, basal and apical mar- gins separated by angle (Fig. 33; Venezuela to Ecuador) femandezi 50. Petiolar node with posterior pointed projection; declivitous propodeal face with longi- tudinal costulae; body color uniformly brown (tropical S. America) acuminata - Node sometimes with slight posterior projection, but never pointed; declivitous propodeal face with transverse costulae, rarely longitudinal; color variable, usually with brown and ferruginous areas, rarely totally brown (Mexico to tropical S. America) sulcata 51. Anterior clypeal lamella laterally acutely angulate 52 Volume 4, 1995 153 - Clypeal lamella laterally rounded or obtusely angulate 53 52. HW < 1.11, WL < 1.72 mm; anterior clypeal lamella straight (Fig. 31; n S. America) . . . ericae reval. sp. - HW > 1.11, WL > 1.72 mm; anterior clypeal lamella medially convex and laterally con- cave (Fig. 32; Venezuela and Colombia) ciirvoclypeata 53. Declivitous propodeal face with longitudinal costae; clypeal lamella straight anterad and laterally rounded; mandibles subtriangular, basal and apical margins meeting through broad convexity (Fig. 62) 54 - Declivitous propodeal face with transverse costae; clypeal lamella excised mesad and with two lateral convexities; mandibles roughly triangular, apical and basal margins separated by notch (Fig. 64; Costa Rica) volcano, n. sp. 54. Petiolar node with dorsal margin more or less evenly convex in lateral view; metacoxal tooth denticle-like; subpetiolar process with acutely pointed angles (s Brazil) lucaris - Petiolar node with dorsal surface flat; metacoxal tooth broadly triangular; subpetiolar process bluntly angulate (s Venezuela) siapensis, n. sp. 55. Antennal scapes usually smooth and shining 56 - Scapes with longitudinal striae or rugulae (sometimes partially effaced) 60 56. Scapes surpass vertexal margin by two or more apical widths; HW > 1.07, WL > 1.64 mm 59 - Scapes when laid back do not surpass vertexal margin by more than one apical width; HW < 1.07, WL < 1.64 mm 57 57. Body sculpture striate; no propodeal denticles; ventral petiolar process in lateral view projecting anterad as a lobe, lacking a posterior angle; Hispaniola Island lineolata - Sculpturing costulate; propodeal denticles present; petiolar ventral process with a pointed posterior corner; South America 58 58. Costulae on dorsum of metanotum and propodeum well defined; petiole in dorsal view with concentric costulae (s Brazil) rastrata - Costulae on dorsum of metanotum and propodeum mostly effaced; petiole in dorsal view transversely costulate (e Colombia) ingeborgae 59. Propodeal teeth longer than diameter of propodeal spiracles; petiolar node longer than wide and subpetiolar process projects anterad as triangular lobe (Brazilian Amazon drain- age) land - Propodeal teeth shorter than diameter of propodeal spiracles; petiolar node wider than long; subpetiolar process subquadrate (s Brazil) menozzi 60. Anterior pronotal margins with at least five transverse costulae or rugulae; eyes promi- nent and bulging (Fig. 37) 61 - Anterior pronotal margin with longitudinal costulae that extend caudad across rest of mesosomal dorsum; eyes not bulging and prominent, relatively flat (Fig. 38; Florida, Pan- ama to Argentina) triangularis 61. Vertex with 4-5 transverse costulae; anterior one-half of postpetiolar tergum rugose ... 62 - Vertex with longitudinal costulae; anterior one-half of postpetiolar tergum longitudinally costulate 63 62. Dorsum of metanotum and all of propodeum with transverse costulae (Panama) .... cuneiforma n. sp. - Dorsum of metanotum and propodeum longitudinally costulate; declivitous propodeal face with weak longitudinal costulae, smooth and shining mesad (Panama to tropical S. America) mecotyle 63. Propodeal spines at least as long as distance between inner sides of their base (Fig. 12); ferruginous species (Costa Rica to Colombia) bispinosa - Propodeal spines shorter (Fig. 13); black species (Colombia to Ecuador) perspicax 154 Journal of Hymenoptera Research New World species of Gnamptogenys In this section the species are arranged in alphabetical order. In the synonymy listings, when more than one citation fol- lows a junior synonym, the first author, after the author of the species, is respon- sible for that synonymy. The diagnosis is meant to complement the results obtained in the key. In order to avoid repeating in- formation already published the reader should consult Lattke (1990) for additional data concerning ecology and localities of species previously known from Venezue- la. For the sake of brevity the dates and collector's names have been omitted. Gnamptogenys acuminata Emery Gnamptogenys acuminata Emery 1896:50; Brown 1958:228,^236, 299; Kempf 1961:491; Kempf 1972:111; Kempf 1976:51; Lattke 1990:9. Diagnosis. — Mandibles subtriangular; longitudinal costulae on mesosoma and declivitous propodeal face; transverse su- tures weak, visible only with certain an- gles of light; body dark brown to black; coxae and basal two-thirds of femora tes- taceous to brown-testaceous; tibiae, apex of femora, antenna and mandibles brown. Ecology. — Found in humid forests, nest- ing in rotting wood. Comments. — The acute posterior tooth on the petiolar node, combined with lon- gitudinal costulae on the declivitous pro- podeal face, seem to provide the most re- liable distinction from tornata. Color in ac- uminata is not so variable but light-pig- mented forms can occasionally be found. The mesometepisternal suture is usually absent. Specimens Examined. — BRAZIL, Amazo- nas: Ig. Marianil, Rio Branco Rd, 24 km NE Manuas. COLOMBIA, Meta: Cano EL Buque, S Villavicenicio, 480 m. PERU, Lo- reto: 15 km WSW Yurimagua, 5°59'S 76°13'W, 220 m; Panguana: 9°37'N 74°56'W, 220 m. VENEZUELA, Bolivar: Rio Cuyuni, 66 km SSE El Dorado, 6° 09' N 61° 30'W, 250 m; 10 km E Icabaru, 700 m. Amazonas: Rio Baria, 0°50'N 66°10'W, 140 m. Gnamptogenys acuta Brown Holcoponera acuta Brown 1957:491. Gnamptogenys acuta (Brown); Brown 1958:300; Kempf 1972:111. Diagnosis. — Similar to striatula; four to six transverse costulae on anterior prono- tal face, rest of mesosoma with longitudi- nal costulae; longitudinal costulae on an- terior face of postpetiole; body dark brown to black, legs brown. Figs. 21-24. Lateral view of petiole. 21, G. relicta; 22, G. inoelleri; 23, G. acuta; 24, G. pleurodon. Volume 4, 1995 155 26 Figs. 25-26. View of declivitous propodeal face. 25, G. ammophila; 26, G. striatiila. Figs. 27-28. Frontal view of scapes, showing hairs and pilosity. 27, G. pleurodon; 28, G. porcata. Ecology. — Found in low to premontane jungles. Comments. — Not frequently collected, knowledge of its range has expanded slowly. Specimens Examined. — ECUADOR, Napo: 2-8 miles N Puyo, Pastaza, 935 m; original locality unknown: Quarantine Station, San Pedro, California. PERU, Ju- nin: Colonia Perene, Rio Perene, 18 km NE La Merced; Loreto: Boqueron, 500 m; CO- LOMBIA, Meta, Mesetas, 780 m. Gnamptogenys alfaroi Emery Gnamptogenys alfaroi Emery 1894:145; Brown 1958:301;'Kempf 1972:111. Diagnosis. — Mandibular dorsum smooth and shining with some punctulae; meta- coxal spine present; subpetiolar process anteriorly projected; gastric tergum 2 with basal one-half longitudinally costulate, the rest smooth and shining; disk of pospetio- lar sternum smooth and shining; body black; legs and antennae brown. Ecology. — Known from montane rain forest. Comments. — The Ecuador record ex- tends its distribution from the only pre- viously known localities in Costa Rica. This species is a striking member of the mordax group due to its elongated, almost cylindrical body. Its relationship with oth- er members of the mordax group is not clear. It is rarely collected. Specimens Examined. — COSTA RICA, Turrialba. ECUADOR, Guayas: 3 km SW Bucay (MCZC). Gnamptogenys ammophila Lattke Gnamptogenys ammop>hila Lattke 1990:9. Diagnosis. — Black, finely costulate with brown scapes and mandibles, light brown legs. Subpetiolar process subquadrate, not as projecting anterad as in striatula. Me- sosomal dorsum with longitudinal costu- lae and erect hairs up to 0.30 mm in length. Ecology. — This species is unique within the genus due to its savanna nest sites as opposed to the forest preferences of most species. Its range is apparently restricted to high (>1200m), cool savannas of the Upper Rio Caroni watershed (La Gran Sa- bana). It has not been found in lower sa- vannas, despite the fact that another ant, Pogonomyrmex naegeli, is widespread in sa- vannas isolated in the midst of great ex- panses of forest. The extent of savannas in the Guiana Shield was much broader dur- ing the last glaciation (75,000-11,000 B.P.), as were generally cooler temperatures and a drier climate (Schubert 1988; Clapperton 1993). Such conditions could concievably have favored the origin of this species from its apparent sister species G. striatula during the course of the last glaciation. Comments. — See Lattke 1990. Specimens Examined. — VENEZUELA, Bo- livar: via El Doradc-)-Santa Elena, km 141, Rio Apanwao, 5°52'N 61°26'W, 1500 m. 156 Journal of Hymenoptera Research ^: iT 31 29 30 32 Figs. 29-30. Ventral view of petiole showing shape of subpetiolar process (outlined in black). 29, G. gracilis; 30, G. andina. Figs. 31-32. Frontal view of clypeal lamella. 31, G. sulcata; 32, G. curvoclypeata. Gnamptogenys andina Lattke, new species (Figs. 39-40) Type Specimens Examined. — Holotype worker. COLOMBIA, Valle: Parque Far- allones de Cali, El Topacio, 20 km NW Cali, 76 37'N 2 30'W, 1550 m, 31-XII-1981, J. Lattke No. 220. Deposited in MIZA. Paratypes (all from Colombia, Valle): (1). 22 workers from tye same nest series as the holotype. (2) 21 workers, same data as holotype except collection numbers 220, 214, 218. (3). 18 workers, Cali-Buenaven- tura road, km 21, 15 km NE Cali, 1300- 200 m, l-X-1975, J. Lattke, leg. (4). 6 work- ers, CVC station near Pance, 15 km W Cali, 1700 m, 12-XII-1975, J. Lattke, leg. Additional Specimens (not paratypes). — ECUADOR, Bolivar, 20 km de Palzabam- ba, 1500-1800 m, 18-X1-87, N. Zavala, leg. Six paratypes deposited in MCZC and three in each of the following : LACM, BMNH, CASC, MUSP. The remainder of paratypes and other specimens deposited in MIZA. Worker. — Holotype (Paratypes) mea- surements: HL 1.04 (1.02-1.08); ML 0.40 (0.26-0.36); HW 0.88 (0.82-0.94); SL 0.86 (0.86-0.92); ED 0.12 (0.10-0.14); WL 1.36 (1.36-1.60) mm; CI 0.85 (0.83-0.90); SI 0.98 (0.93-1.10); OI 0.14 (0.11-0.17) n = 7. Typical striatida group member: Rough- ly costulate sculpture. Vertexal margin concave in frontal view; clypeal costulae extend slightly onto anterior lamella; man- dibles triangular and mostly striate except for smooth and shining lamellate basal corner and margin of chewing border; compound eyes relatively small; in lateral view dorsal propodeal margin not notably depressed below rest of mesosomal dor- sum; propodeal spiracle slightly elevated above rest of sculpture and at edge of de- clivitous surface; petiolar node posteriorly inclined; subpetiolar process subquadrate, typical of strigata subgroup. Longitudinal costulae present on declivitous postpetio- lar face. Weak transverse costulae on post- petiolar ventrum. Fairly abundant standing hairs on body, but very sparse appressed pubescence. Legs smooth and shining, except for nu- merous piligerous punctures. Metacoxal dorsum with a well-developed denticle. Body dark brown to black with yellowish legs. Male. — Longitudinal costulae on cephal- ic dorsum; mandibles triangular and cos- tulate; pronotum smooth and shining with piligerous punctures; pronotum with pili- gerous punctures; mesonotum mostly smooth and shining except for narrow an- terior band of transverse striae; declivitous propodeal face with longitudinal rugulae that do not reach anterior dorsal face; gas- ter smooth and shining. Queen. — Unknown. Discussion. -This species is near strigata and though the differences are not striking, they seem consistent enough to define a Volume 4, 1995 157 35 Figs. 33-38. Frontal view of head. 33, G. fernandezi; 34, G. sulcata; 35, G. haenschi; 36, G. schmitti; 37, G. bispinosa; 38, G. triangularis. species. G. strigata is smaller (HW 0.74- 0.80); WL 1.20-1.34 mm), with relatively larger eyes (OI 0.17-0.20) and shorter scapes (0.70-0.78 mm) and dorsal propo- deal face in lateral view is usually de- pressed below the rest of the mesosomal dorsum. It is probably sympatric with nn- dina in part of its range, having been found only 51 km away from the andina type lo- cality. The male of strigata has a median area of rugosity on the mesonotum and to- tally rugulose propodeum. Individuals of the Ecuadorean sample are slightly larger than the Colombian specimens. Etymology. — The name alludes to the Andes, a portion of which is inhabited by this species. Gnamptogenys annulata Mayr Ectntomma {Gnamptogenys) rimulosa var. annulata May 1887:543. Ectatomma {Gnamptogenys) annulatutn Mayr; Mann 1916:408; Luederwalt 1926:23. Gnamptogenys annulatum Mayr; Mann 1922:3; Brown 1958:300; Kempf 1960:390; Kempf 1961:491; Kempf 1969:275; Kempf 1972:112; Kempf 1976:51; Lattke 1990:11. Diagnosis. — Finely striate species; longi- tudinal striae on head, promesonotum and gastric segments 2 and 3; petiolar dor- sum with transversely arched striae; man- dibles subtriangular; metacoxal spine ab- sent; body reddish brown; coxae ferrugi- nous yellow; tibiae and femora yellow to testaceous but apically ferruginous yellow to ferruginous. Ecology. — J. Longino (pers. comm.) re- ports from a nest midden chitinous frag- ments of Coleoptera, aradid bug nymphs, an unidentified male ant, and a Cauipotio- tus worker. An apparently recently killed 158 Journal of Hymenoptera Research 42 Figs. 39-42. Frontal view of head, and lateral view of mesosoma and petiole. 39-40, G. andina; 41-41, G. brunnea. Pheidole (cephalica?) queen was also found in a nest chamber. Comments. — The gauge of costulation can vary from striate to finely costulate, but coloration is fairly stable. SPECIMENS EXAMINED: BOLIVIA, Caranavi: vie. Radio, 800 m. BRAZIL, Amazonas: 80 km NNE Manaus, 2°25'S 59°46'W, 80 m. COLOMBIA, Valle: Bajo Calima; Cauca: Isla La Gorgona. COSTA RICA, vie. Guapiles, Rio Toro Amarillo; Prto. Limon; Limon: Tortuguero, < 100m, 10°35'N 83°31'W; Heredia: 10°20'N 84°84'W, 500 m; 3 km S Puerto Viejo, 10°26'N 84°00'N, 50 m; Pq. Nac. Braulio Carrillo, Carrillo, 500 m; Peninsula de Osa, Sirena, 8°28'N 83°35'W; Alajeula: Rio Peii- as Blancas, 10°19'N 84°43'W, 800 m. EC- UADOR, Pichincha: 6 km W Sto. Domingo de los Colorados, 953 m; Maquipucuna, 5 km ESE Nanegal, 0°70'N 78°38'W, 1250m; Sucua. GUIANA, Kartabo. PERU, Pangu- ana: 9°37'S 74°56'W, 220 m; Tingo Maria: Valle de Monson; Madre de Dios: Res. Rio Tambopata, 30 km SW Prto. Maldonado, 290 m; Loreto: Boqueron, 500m; San Mar- tin: Davidcillo, 30 km NNE Tarapoto, 6°15'S 76°15'W. VENEZUELA, Tachira: Uribante-Caparo, Las Cuevas, 7°47'N 71°46'W, 500 m; Barinas: Ticoporo, 8°04'N 70°48'W, 250 m; Amazonas: Alto Mavaca, 2°01'N 65°07'W, 200 m; Alto Siapa, 1°42'N 64°38'W, 500 m. Gnamptogenys banks! (Wheeler) Emeryella banksi Wheeler 1930:10. Gnamptogenys banksi (Wheeler); Brown 1958: 227,233; Kempf 1972:112. Volume 4, 1995 159 Diagnosis. — Finely costulate; head broad from the first sighting to the prey was 6 with longitudinally striated scapes; meso- m. I followed a returning worker as far as metanotal suture not deeply impressed, I could into a tree fall tangle, 10 m from but distinct. It lacks the long, fine mandib- prey." The second attack involved 46 ular sulcus present in laticepJmla. workers tackling a 6 cm long millipede Ecology. — Found in rotten logs in low- within a rc^tten log. Once the millipede land rain forest, this species is a specialist was subdued they carried it back to the millipede predator. nest. Comments. — It is rarely collected. A Two millipedes taken from the nest by specimen from Barro Colorado has two Longino were determined by R.L. Hoff- deep fossae on the posterolateral margin man as Trichomorpha sp. (Polydesmida: of the mesonotum. The type was exam- Chelodesmidae), and Epinannolene sp. ined. (Spirostreptida: Epinannolenidae). Specimens Examined. — ECUADOR, Comments. — This unique ant was previ- Guayas: 3 km S Bucay. PANAMA, Isla ously only known from a restricted area Barro Colorado (MCZC). in Central America. Along with perspicax, it represents a highly specialized form in Gnamptogenys bispinosa (Emery) ^^^ lineage of millipede hunters. The an- Ectatomma {Poneracantha) bispinosum Emery terior clypeal setae are more developed 1897:547. than in any other of the rastrata group spe- Ectatomma (Holcoponera) bispinosum Emery; Em- cies and an interesting autapomorphy is a ery 1890:40. brief anteromedian longitudinal sulcus on Holcoponera bispinosa (Emery); Wheeler 1952: \\^q clypeus. 132-133. Specimens Examined. — COLOMBIA, Val- Gnamptogenys bispinosa (Emery); Brown 1958: ^^. ^QSTA RICA- Heredia- 10°20'N ?g79 i^r^^ ''''^ ^'°'^'' ^^^^''^^^'' ^'""^^ 84°04'W, 500m; Peninsula de Osa, Pq. Nac. Corcovado, Sirena, < 100m, 8°28'N Diagnosis. — Eyes subglobulose, scapes 83°35'W; Puntarenas: 2 km S San Vito, Fin- long and surpassing vertexal margin; ca Las Cruces, 1219 m; Rio Reventazon, 4 mandibles edentate. Promesonotal suture km E Turrialba (MCZC). breaks sculpture dorsally but not laterally; mesometanotal suture deep and broad. Gnamptogenys bisulca Kempf and Head, mesosoma and postpetiole rugose; Brown „ ^ ■ . . . . Gnamptogenys bisulca Kempf and Brown 1968: £co/o^/.-This wet forest species is a 92-3; ^ Wheeler and Wheeler 1971:1202; millipede specialist. J. Longino (pers. Kempf 1972112 comm.) observed the following two raids from the same colony, both at 11:00 AM: Diagnosis. — Longitudinal costulae on " A column of 20 workers was moving propodeal declivity, metacoxal spines along a liana on the forest floor. They left hooked and triangular, not acicular; well the liana and moved very slowly across defined transverse costulae on postpetio- the leaf litter, frequently bunching up. lar sternum; propodeal spiracle as low as They all went under a leaf and then agi- surrounding sculpture, not raised, tated ants could be seen coming out from Ecology. — Occurs in mesic forested hab- under the leaf, 2 or 3 ants at a time, clean- itats. The localities suggest preference for ing their antennae and mandibles. I re- relatively cool sites as the lowest altitudes moved the leaf to find the ants attacking are found at higher latitudes (10°N) and a millipede. Some were stinging and some as latitude approaches the equator the lo- were pulling on the legs. The distance calities ascend in altitude. 160 Journal of Hymenoptera Research Comments. — Within the strigata complex this species is unique due to the develop- ment of its metanotal groove, and within the striatula group two other species also show this character state. Recent collec- tions show a wider range than previously known. Specimens Examined. — COLOMBIA, Choco: 10 km SW San Jose del Palmar, Rio Torito, Finca Los Guaduales, 800 m, 1/4- VI-78; Valle: vie. Lago Calima, 1600 m; vie. Saladito, 1900-2100 m. COSTA RICA, He- redia: 17 km S. Puerto Viejo, 10°18'N 84°02'W, 550 m; 17 km N Volcan Barba, 10°17'N 84°05'W, 880-1020 m. ECUA- DOR, Pichincha: 25 km ENE Alluriquin, via Chiribaga, 1400-1800 m; Tinalandia, 16 km SE Sto. Domingo de los Colorados, 9-VI-76; 3 km E Tandapi, 1300 m. Gnamptogenys boliviensis Lattke, new species (Figs. 43, 44) Type Material. — Holotype worker. BO- LIVIA, Tumupasa, W.M. Mann, leg. De- posited in the USNM. Paratypes: Seven workers deposited in the USNM; One worker and one male in each of the fol- lowing: BMNH, MIZA, LACM, MCZC. One worker in MUSP. All from same lo- cality and probably from the same nest. One specimen bears an additional label: "Mulford Biological Expedition, 1921- 1922." Consultation of maps locates Tu- mupasa at 14°09'S 67°55'W in lowland for- ests of the upper Rio Beni watershed. Worker. — Holotype (Paratypes) mea- surements: HL 0.88 (0.83-0.88); ML 0.36 (0.39-0.41); HW 0.70 (0.69-0.70); SL 0.49 (0.48-0.50); ED 0.10 (0.10-0.12); WL 1.23 (1.06-1.15) mm; CI 0.70 (0.80-0.84); SI 0.70 (0.69-0.71); OI 0.15 (0.15-0.18). n=4. Head in frontal view elongate, lateral margins fairly parallel, vertexal margin concave; anterior lamella of clypeus later- ally rounded and medianly convex; eyes situated at mid-length; antennae smooth and shining, dorsoventrally compressed widest apicad; mandibles smooth and shining dorsally, with sparse punctae, su- bfalcate, laterobasally with small area of longitudinal rugae, including small sulcus never longer tha one-fourth of the man- dible; gula with longitudinal costulae, front of head with longitudinal striae. Occasional transverse costulae on pron- otal collar; mesosomal dorsum with lon- gitudinal costulae, slightly thicker on metanotum and propdeum than on pro- mesonotum; pronotal suture softly im- pressed, visible only in certain angles of view; metanotal suture deeply impressed but doesn't break longitudinal sculpture; declivitous propodeal face with oblique to transverse costulae and two superolateral ridges; pronotum laterally with transverse costulae, smooth and shining along ven- tral margin; anepisternum elongate and obliquely costulate; katepisternum and metepisternum with transverse costulae; costulae on metepisternum have same di- rection as on katepisternum, but costulae on lateral propodeal face are more hori- zontal; propodeal spiracle round and not prominently higher than surrounding sculpture; petiole slighty pedunculate, dorsally with longitudinal costulae, ante- rior face with transverse costula or rugae, sometimes partially effaced, laterally with longitudinal costulae with slight oblique tendency; node in lateral view with softly convex dorsal margin, anteriorly concave; anterior and posterior faces fairly sharply separated from dorsal face; posterior face with 4-5 convex costulae; subpetiolar pro- cess varies from subquadrate anteriorly and posteriorly concave. Anterior postpetiolar face smooth and shining; gastric terga I and II with longi- tudinal striae, sternum I laterally costula- te, but discal area smooth and shining as is rest of gaster; procoxa anteriorly and anterolaterally smooth and shining, pos- terolaterally costulate; mesocoxae with transverse costulae that tend to be effaced; dorsum of metacoxa with transverse cos- tulae and basal, low triangular lobe; tibia and femora smooth and shining. Body fer- Volume 4, 1995 161 Figs. 43-44. Micrographs of G. boliviciiisis. Scale bars = 0.5 mm. 43, head; 44, thorax and gaster. 162 Journal of Hymenoptera Research ruginous brown; legs and antennae testa- ceous. Queen. — Unknown. Male. — Mandibles triangular, apical edge serrate, dorsally smooth and shining; cephalic dorsum with longitudinal striae that tend to be effaced before reaching mid-ocellus, faint transverse rugae above the posterior ocelli but rest of head smooth and shining; propodeum rugose; petiole with lateral rugulae, dorsum shin- ing, with slight roughened aspect and me- dian longitudinal costulae; gastric sterna and terga smooth and shining; vestigial arolea present. Discussion. — Several specimens were found with determination labels, probably written by Mann, indicating that he rec- ognized this species as new. G. boliviensis is near to G. continua and the possibility exists that the collection and study of more material may show them to be con- specific. In G. continua the sculpture is coarser, costate-costulate, and the mandi- bles are more robust, without a concave inner edge. Etymology. — The name is derived from the name of the type locality country, Bo- livia. Gnamptogenys bruchi (Santschi) Ectatomma (Parectatomma) bruchi Santschi 1922: 241. Gnamptogenys bruchi (Santschi); Brown 1958: 227; Kempf and Brown 1968:92; Kempf 1969: 275; Kempf 1972:112. Diagnosis. — Striate body sculpture with weak transverse impressions; metacoxal process is shaped as a low, triangular lobe; clypeal lamella broadly convex anterad. Ecology. — Apparently a predator of Tra- chymyrmex and other leaf cutting ants (Kempf and Brown, 1968:92). Comments. — Kempf (1969) notes bruchi as close to hartmani and that a synonymy could be possible. He also mentions the variability of the extent of impression of the metanotal groove and discards it as an indicator of species separation. This could very well be only a variant of hartmanni but due to differences of sculpture on the posterior petiole face, postpetiolar sternal process and clypeal lamella, I have chosen to conserve this name until more material can be studied. The sculpture on the dor- sum of the petiole can vary from longitu- dinal to whorled. Specimens Examined. — ARGENTINA, Cordoba: Alta Gracia (cotype:MCZC). BRAZIL, Sao Paulo: Fazenda B. Rico. Gnamptogenys brunnea Lattke, new species (Figs. 41,42) Type Material. — Holotype worker. CO- LOMBIA, Choco, 10 km SW San Jose del Palmar, Finca Los Guaduales, 760 m, 2-VI- 1978, C. Kugler, leg. Deposited in MIZA. Para types: One queen and 18 nidotype workers. Queen and 6 workers deposited in MIZA. Two workers deposited in each of the following: MCZC, LACM, BMNH, MUSP, USNM. Worker. — Holotype (Paratypes) mea- surements: HL 0.90 (0.88-0.92); ML 0.30 (0.30-0.32); HW 0.74 (0.68-0.80); SL 0.74 (0.80-0.84); WL 1.20 (1.10-1.28) mm; CI 0.82 (0.77-0.87); SI 1.00 (0.95-1.09); OI 0.18 (0.16-0.18) n=5. Cephalic dorsum longitudinally costu- late; in frontal view vertexal margin con- cave and eyes slightly behind mid-length; mandibles triangular, with dorsal costu- lae, basal and apical margins with narrow smooth and shining strip that is sharply separated and lower than the rest of man- dibular surface; anterior pronotal margin with 6-8 transverse costulae which arch around and become longitudinal on dor- sum; mesometanotal suture well-im- pressed; mesonotum narrow and trans- verse; in lateral view dorsal propodeal face slightly depressed below rest of me- sosomal dorsal margin; declivitous pro- podeal face distinctly separated from rest of mesosoma by low lateral ridges; mostly transverse costulae with longitudinal cos- tulae from dorsum extending only to an- Volume 4, 1995 163 terior one-fourth of declivitous face; peti- Diagnosis. — A large finely striate species olar node in lateral view transversely cos- with transverse striae on propodeum, de- tulate and posteriorly inclined: a broad clivity medianly smooth; promesonotal anterior convexity with sharp posterior suture weakly impressed or absent; me- drop and concave posterior margin; sub- sometanotal suture deep and wide; meta- petiolar process shaped as an anteriorly coxal tooth absent, low crest or tubercle at projecting lobe; anterior postpetiolar face most. and sternum with transverse costulae. Ion- Ecology. — Consistently collected and ob- gitudinal on rest; costulae on the gastric served on trees, this species is an arboreal tergum II finer than on preceding seg- nester. Longino found a nest inside a ment; metacoxal spine present. large, deep knothole and records prey Body with sparse decumbent pubes- items of the following: Heteroptera: Pen- cence and erect to suberect hairs; pubes- tatomidae, Aradidae; Coleoptera: Ceram- cence denser on legs; scapes with abun- bycidae, Histeridae, Platypodidae, Chrys- dant decumbent pubescence, no erect or omelidae, Tenebrionidae, and Passalidae. suberect hairs. Body reddish-brown and One observed foraging group was made legs brownish-yellow. up of 9 workers and a dealate queen. Ob- Queen. —HL 0.98; HW 0.86; ML 0.32; SL servations of group foraging of up to 40 0.84; ED 0.18; WL 1.44 mm; CI 0.88; SI single-file workers, plus the following 0.97; Ol 0.21. Like workers with exception Longino observation, indicate trail and /or of usual caste differences. recruiting pheromone communication in Male. — Unknown. this species. "The foragers walked with Discussion. — This species is close to ni- their gasters curled and touching the sub- grivitrea on account of size, posteriorly tilt- strate. When they were together near the ed petiolar node, sleek appearance, sub- prey I could often make out a tiny white petiolar process shape, sparse pilosity and area (gland?) exposed at the very tip of the lack of any trace of anterior petiolar pe- gaster, which other workers would inves- duncle. But nigrivitrea is different in its pi- tigate". ceous body coloration with dark brown Comments. — Color can vary from ferru- legs, lack of metanotal groove, slightly ginous to light-brown, and the striae on more elevated propodeal spiracles, longer the posterior nodal face may be effaced, pilosity, and a sleeker petiolar node with Specimens Examined. — BRAZIL, Amazo- a more prominent apical point. nas: Ilha de Curari, (varzea); Amapa: Villa Etymolog}/. — The name comes from the Amazonas. COLOMBIA, Meta: Reserva Latin term for brown, brunneus. La Macarena, Cafio La Curia, 580 m. COS- TA RICA, Peninsula de Osa: Pq. Nac. Cor- Gnamptogenys concinna (Smith) covado, Llorona; Heredia: Finca La Selva. PANAMA, Isla Barro Colorado. PERU, Ectatommn concinnna F. Smith 1858:103. Madre de Dios: 15 km E Prto. Maldonado, Ectakvnma{Gnamptogenys)concinmim{¥.Smm; 200 m; Amazonas, Panguana, 9°37'S Mann 1916:406; Wheeler 1922:2. ^a^^c^'wi nnn ^ x- \a ^ \7 n a r. ,^ 74 56 W, 220 m; Tmgo Maria: Valle de tctatomma (Gnamptogenys) concinnum var. ro- ^. . -roiMTMr-v a r^ a • \i w i r-^ ■ AAfu 1 in-^oo D ir^ron-^r, Monson. TRININDAD, Arima Valley, 152 mnn? Wheeler 1923:2; Brown 1958:227. a/cxtc^ttct a a ai/ r,- Lcia\omma{Gnampto^em,s)concmnumv,.r.comca "^- VENEZUELA, Amazonas: Alto Rio Borgme.er 1928:196; Brown 1958:227. Mavaca, 2 02'N 65°06'W, 200m. Ectatommn {GnarnptoQenys) concimmni var. senii- ^ ^ .• »,r , D . ir.or.ir^r r, irvr-r, Gnamptogenys contiHua Mavr circularis Borgmeier 1929:195; Brown 1958: r & j / 227. Ectatommn [Gnnmptogenys) conti)U(uiii Mayr Grwmptogcnys co}tci)uia (Smith); Brown 1958: 1887:544. 227; Kempf 1972:112; Lattke 1990:11. Ectatommn {Giinmptogcin/^) cxaratum Emery 164 Journal of Hymenoptera Research 1901:50; Brown 1958:303; Kempf 1972:112; Lattke 1990:13. NEW SYNONYMY. Ectaomma {Gnamptogenys) continuum var. pana- mensis Santschi 1931:265; Brown 1958:228, 304. Gnamptogenys continua (Mayr); Brown 1958:228, 304; Kempf 1972:112; Lattke 1990:12. Diagnosis. — Mandibles with linear ten- dency; mesometanotal suture impressed; declivituous propodeal face with small su- perolateral lobes; upper half with longi- tudinal costulae and inferior half with transverse costulae; sternal disc of post- petiole smooth; anterior nodal face rugose with some inferior transverse costulae. Ecology. — Collected in leaflitter samples from humid forsets. Comments. — The synonymy of exarata was inevitable as more material accumu- lated and it became impossible to separate the two forms. The gap between OD and apical scape width closed. There are changes in the gauge of the costulation, degree of anterior clypeal convexity, ef- facement of the sculpture, shape of the subpetiolar process, extent of the petiolar peduncle, and size (HW 0.56-0.96; HL 0.64-1.16; WL 0.88-1.62). The occiput has a median dent that is most noticeable when the occiput is smooth and shining as opposed to costulate forms. Deposited in the MCZC is a specimen from Vera Cruz that was compared with the type by Brown in 1954. Specimens Examined. — BELIZE, Caves Branch, 4 km S Belmopan. BRAZIL, Rio de Janeiro: Itatiara; Sao Paulo: Fazenda Con- quista, Rio Preto; Guquitiba, BRE km 76. HONDURAS, La Lima. ECUADORA, Pi- chicincha: Tinalandia, 16 km SE Sto. Do- mingo del los Colorados, 680 m. COLOM- BIA, Meta: Quebrada Susumuko, 23 km NW Villavicencio, 1000 m; Magdalena: Parque Tayrona, Pueblito, 360 m. MEXI- CO, Vera Cruz, Tuxtla, Las Hamacas, 17 km N Santiago. PERU, Panguana, 9°37'S 74°56'W, 220 m; Huanuco: 16 km SW Las Palmas, 1000 m; Pasco: Pozuzo, 1000, SU- RINAM, Kartabo, Bartica Dam. VENE- ZUELA, Barinas: 17 km SSW Ciudad Bo- livia, 8°02'N 70°46'W, 240 m; Guarico: 24 km N Altagracia de Orituco, 300 m. Gnamptogenys cuneiforma Lattke, new species Type Material. — Holotype worker: PAN- AMA, Bocas del Toro, 8°47'N 82°11'W, 500m; 16/18-VII-87, D.M. Olson, leg. De- posited in MCZC. Paratypes: One worker deposited in MCZC and two workers de- posited in MIZA, all with same locality data as the holotype. Worker. — Holotype (Paratype) measure- ments: HL 0.90 (0.84-0.88); ML 0.62 (0.60- 0.64); HW 0.84 (0.82-0.84); SL 0.82 (0.80- 0.82); ED 0.20 (0.21-0.22); WL 1.40 (1.38- 1.40) mm; CI 1.07 (1.02-1.05); SI 0.98 (0.98); OI 0.24 (0.26-0.28), n=4. Head subquadrate in frontal view, sides very broadly convex, slightly wider ante- riorly than posteriorly, posterior margin broadly concave, longitudinally costulate. Anterior clypeal margin with a small la- mella, medianly straight and laterally rounded. A brief longitudinal carinae, thicker than surrounding sculpture, pres- ent between frontal lobes. Mandibles tri- angular and elongate, masticatory border concave, dorsal surface of blade convex and with longitudinal rugulae. Scapes shining and with low rugosities. Scapes without pilosity, only decumbent hairs. Anterior pronotal face with 4-5 trans- verse costulae, sometimes quite rough and resembling rugae, longitudinal costulae on the dorsum posterad including the mesonotum. Metanotum and propodeum with transverse costae. Promesonotal suture well to poorly im- pressed, metanotal groove well im- pressed. In lateral view the mesosoma has a very broadly convex dorsal margin, pro- podeal denticles brief and triangular, de- clivity concave. Sides of pronotum with regular longitudinal, slightly oblique cos- tae. Anepisternum fulcrum shaped. Kat- episternum with a well-developed flange. Mesopleura, propodeum and dorsal two- Volume 4, 1995 165 thirds of metapleura with transverse/ oblique costae; metapleura ventrad of spi- racle with longitudinal costae. Petiole in lateral view with a low and broadly convex dorsal margin, ventral process projects anterad. Dorsally with transverse costae, posterior margin ru- gose. Gastric dorsum with longitudinal costulation; ventrum of postpetiole with transverse rugae along posterior half; ir- regular rugulae anterad to the process. Base of foretarsus with a single stout se- tae opposite the strigil. Forecoxae with transverse costulae on lateral face, as well as dorsal faces of meso- and metacoxae. Legs shining and with abundant shallow punctulae. Very sparse pilosity on femora and tibiae. Color ferruginous to dark brown; mandibles, antennae and legs fer- ruginous. Queen, Male. — Unknown. Discussion. — This species may be con- fused with two other small rastrata group species: mstrata and mecoti/le. It is separa- ble from the former on account of the lon- ger scapes (surpassing the vertexal mar- gin), transverse costulae on the propo- deum as opposed to longitudinal, and a well-defined metapleura, separated from the propodeum laterally by an impressed line, lacking in rnstratn. The costulae of ine- cotyle have a coarse aspect in contrast with the smooth regularity in cuneiforma, tneco- tyle also has the striae on the scapes and dorsal mandibuar surface better defined and the masticatory border lacks the con- cavity of cuneiforma. Etymology. — The name alludes to the shape of the anepisternum. Gnamptogenys curvoclypeata Lattke Gnamptogenys curvoclypeata Lattke 1990: 12. Diagnosis. — Piceous species with ferru- ginous antenna, legs and coxae; prome- sonotal suture very weakly impressed, visible only with certain angles of light; meso- and metacoxae smooth and shining on median and basal lateral faces; meta- coxae with weakly developed triangular dorsal lobe. Ecology.— See Lattke (1990). Comments. — See Lattke (1990). Specimens Examined. — COLOMBIA, Amazonas: 7 km N Leticia (MCZC). Gnamptogenys ejuncida, Lattke, new species (Figs. 48, 50) Type Material. — Holotype worker. CO- LOMBIA, Putumayo, Mocoa, 610 m, 4-1- 77, C. Kugler, leg. Deposited in MIZA. Worker. — Holotype measurements: HL 1.10; ML 0.26; HW 0.88; SL 1.02; ED 0.18; WL 1.54 mm, CI 0.80; SI 1.56; OI 0.21. With head in frontal view: vertexal mar- gins slightly convex, lateral margins broadly convex and anterior clypeal mar- gin convex; mandibles triangular and with rugose dorsum; anterior pronotal face with six transverse costulae, dorsum with longitudinal costulae; mesonotum with anteriorly convex costulae, transverse at metanotum and posteriorly convex on dorsal propodeal face; anepisternum with three transverse costulae that descend from mesonotum; rest of mesosomal side obliquely costulate, some curve around propodeal spiracles; declivitous propodeal face with longitudinal costulae; in lateral view dorsal mesosomal outline evenly convex, with no abrupt breaks or depres- sions; petiolar node posteriorly inclined: in lateral view transversely costulate with convex anterior margin, concave posterior margin and overhanging apex; subpetiolar process triangular and projecting anterad. Anterior face to dorsal two-thirds of post- petiolar dorsum with transverse costulae; rest of gastric sculpture longitudinal; post- petiolar sternum anteriorly transversely costulate, posteriorly divergent; transverse costulae on procoxae and anterior faces of meso- and metacoxae; metacoxal tooth well developed. Abundant suberect hairs and pubescence on body and extremities. Queen, Male. — Unknown. Ecology. — Unknown. 166 Journal of Hymenoptera Research Figs. 45-53. Lateral view of posterior leg, mesosoma and petiole, and frontal view of head. 45-47, G. extra; 48-50, G. ejuncida; 51-53, G. pilosa. Discussion. — On account of size, sculp- ture, pilosity and posteriorly inclined pet- iole node this species seems to be most closely related to pilosa, a very similar spe- cies found in an Andean canyon NW of Cali, Colombia. The sculpture of pilosa is much finer and its pilosity is a bit denser, including abundant erect hairs, practically lacking in ejuncida. The presence of erect to suberect hairs in pilosa vs. subdecum- bent to decumbent hairs in ejuncida is es- pecially notable on the femora. In lateral view the angle formed by the junction of the dorsal and declivitous propodeal faces is greater in ejuncida than in pilosa, giving it a more slender appearance. Etymology. — This sleek species inspired the use of the Latin word for slender, ejun- cida. Gnamptogenys ericae Forel, revalidated Gnamptogenys tornata var. ericae Forel 1912:33; Brown 1958:229. Grtamptogenys sulcata (F. Smith); Brown 1958: 329; Lattke 1989:24. Gnamptogenys sulcatum bufonum Weber 1938: 208. NEW SYNONYMY. Diagnosis. — Mandibles subtriangular; posterior mesosomal dorsum and propo- deum with longitudinal costulae, trans- verse sutures barely visible under certain angles of lighting; small metacoxal tooth. Body black; antennae, legs and mandibles brown to dark brown. Ecology. — Can be found nesting in dis- turbed forests and coffee plantations as well as in undisturbed forests. Comments. — Much confusion has accu- Volume 4, 1995 167 mulated regarding the identity of this spe- cies and the similar sulcata. Many previous references to sulcata are actually ericae, but the confusion makes it quite difficult to distinguish between each reference. Dr. Ivan Lobl of the MHNG kindly examined the ericae type and confirmed the longitu- dinal direction of the costulae on the pos- terior face of the node of the specimen. A specimen in the MCZC from Guiana with a determination label as bufonum in N. Weber's handwriting and another label stating "in Bufo marinus stomach" coin- cides with the information in Weber's de- scription. Even though it has no label des- ignating it as a type, it seems reasonable to conclude that Weber's bufonum is a syn- onym of ericae. For a broader discussion on the separation of the forms known as tornata and sulcata the reader should turn to the "Discussion" for G. sulcata. Color in this species is relatively constant but oc- casional ferruginous, brown or mottled specimens will turn up. Size and shape of the petiolar node does not vary as much as in sulcata. Workers seem to always have longitudinal costulae on the propodeal de- clivity, though queens may have trans- verse costulae. Specimens Examined. — BOLIVIA, Rosa- rio. COLOMBIA, Magdalena: Minca, 610 m; Guajira: Serrania de Macuira, 7 km S Nazareth, 70-200m; Meta: San Juan de Ar- ama, 914m; La Macarena, Rio Guayabero, 260m; 65 km E Puerto Lopez. GUIANA, Forest Settlement, R. Mazaruni. TRINI- DAD, Port of Spain, Federation Park. VENEZUELA, Merida: Santa Cruz de Mora, 600m; Guarico: Hato Masaguaral; Bolivar: Campamento Rio Grande, 8°07'N 6r48'W, 280 m; Anzoategui: 20 km S El Tigre. Gnamptogenys extra Lattke, new species (Figs. 45-47) T\/pe Material. — Holotype worker. EC- UADOR, Pichincha: Tinalandia, 16 km SE Santo Domingo de Los Colorados, VI- 1976, S. & J. Peck, leg. Deposited in MIZA. Paratypes: (1) One worker with the same data as the holotype. Deposited in MCZC. (2) Two workers from COLOMBIA, Cho- c6: Lloro, Vereda Pefialoza, 26-III-1988, V. Jaimes, leg. One worker in BMNH and LACM, respectively. Worker. — Holotype (Paratypes) mea- surements: HL 1.20 (1.10-1.12); HW 1.00 (0.94-1.00); ML 0.34 (0.30-0.34); ED 0.16 (0.14-0.18); SL 1.26 (1.00-1.20); WL 1.68 (1.44-1.50) mm; CI 0.83 (0.85-0.98); SI 1.26 (1.06-1.20); lO 0.16 (0.15-0.19) n=4. With head in frontal view: vertexal mar- gin straight, laterally convex with eyes slightly behind midlength, and anterior clypeal margin convex; mandibles trian- gular with dorsal striae; scapes relatively thick in appearance and surpassing ver- texal margin; vertex with 2-3 transverse costulae; anterior pronotal face with 5-7 transverse costulae, dorsally with anteri- orly arching costulae; costulae on meso- notal concentric or longitudinal; costulae on metanotum and dorsal propodeal face transverse to longitudinal, sometimes arched; declivitous propt^deal face with longitudinal costulae that diverge poste- riorly, basally with 2-3 transverse costu- lae. Petiolar node in lateral view subquad- rate, with broad anterodorsal convexity and sharp posterior drop with overhang- ing apex; subpetiolar apex is an anteriorlv projecting lobe; anterior postpetiolar face with transverse costulae that arch back and bect")me longitudinal posteriorlv; rest of dorsal gastric cc^stulae longitudinal, di- vergent on posterior postpetiolar sternum; ventral stridulitrum present on fourth ab- dominal segment. Scapes with abdundant decumbent pilosity and no suberect or subdecumbent hairs; decumbent pilosity also present on coxae and legs, with no standing hairs; sparse suberect hairs on rest of body. Body dark brown, legs and antennae slightly lighter. Male, Female. — Unknown. Discussion. — This species has a fairly pointed petiolar node apex, but it doesn't 168 Journal of Hymenoptera Research approach the extremely pointed condition of G. acuta. The node is more robust and subquadrate (lateral view) than in other porcata subgroup species and is reminis- cent of the condition in G. tortiiolosa. This species could be closely related to G. stria- tula on account of the node shape (erect) and the lacking anterior peduncle. Gnamptogenys fernandezi Lattke Gnamptogenys fernandezi Latike 1990: 14. Diagnosis. — Triangular mandibles with a smooth and shining dorsum; anterior la- mella of clypeus convex; promesonotal su- ture lightly impressed and mesometanotal suture absent; low, elongate petiole, pos- terodorsally bluntly pointed; metacoxal teeth absent. Ecologif. — A rarely collected, but fairly widespread species in n South America. In Venezuela known only from 2 forested lo- calities in the Cordillera de la Costa be- tween 650-1100 m. Comments. — See Lattke (1990). Specimens Examined. — COLOMBIA: Val- le: 50 m (no other data). ECUADOR. Mo- rora: Santiago, Los Tayos. PERU, Madre de Dios: 5 km SE Prto. Maldonado, Rio Tambopata, 260m; Cuzco Amazonico, 15 km NE Prto. Maldonado, 200m. VENE- ZUELA, Carabobo: Hacienda Palmichal, Canoabo 900 m. Gnamptogenys gentryi Lattke, new species Type Material. — Holotype worker. CO- LOMBIA, Valle, 2250 m, Insp. Pance, Re- serva Natural Hato Viejo, J.B. Hillaire, leg. Deposited in MIZA. Thank you to Fernan- do Fernandez for making the specimen available. Worker. — Holotype measurements: HL 1.22; ML 0.32; HW 0.92; ED 0.22; SL 1.04; WL 1.64 mm; CI 0.75; SI 1.13; Ol 0.24. Head in frontal view elongate, anterior clypeal margin convex, posterior margin concave, eyes relatively flat, finely costu- late throughout, though diverging at ver- tex, with a band of three transverse cos- tulae next to the vertexal carinae; mandi- bles triangular, frontal surface rugulose. Scapes mostly smooth and shining except for sparse punctulae. Pronotal dorsum with whorled costulae almost enclosing 3 brief longitudinal cos- tulae, rest of mesosomal dorsum with lon- gitudinal costulae, metanotal groove deeper than pronotal suture. In lateral view the mesosoma has a convex pronotal profile that becomes an inclined and more or less straight slope that bends sharply at the propodeal declivity, which has a broadly concave profile. Node in lateral view tilted backwards, with oblique to longitudinal rugulae laterally and trans- verse costulae on the anterior and poste- rior faces, ventral process subquadrate, with anterior corner projecting more than posterior corner. Gastric tergite I mostly smooth and shining except for anterior face which has a small area of transverse costulae and the sternite with transverse striae; the rest of the gaster is smooth and shining. Color black. Queen, Male. — Unknown. Discussion. — This striatula group species is quite easy to separate from all others just considering the presence of a metan- otal groove and the mostly smooth and shining gaster. Etymology. — Named in honor of the late botanist Alwyn Gentry, he died in the line of duty in the Andes of Ecuador. Gnamptogenys gracilis Santschi Holcoponera gracilis Santschi 1929:468. Gnamptogenys gracilis (Santschi); Brown 1958: 228, 232; kempt 1972:112; Lattke 1990:17. Diagnosis. — Finely costulae; pronotal costulae concentric to arched; mesonotum frequently with longitudinal costulae sur- rounded by concentric costulae; metano- tum and propodeal dorsum with trans- verse costulae, and longitudinal on decliv- itous propodeal face; subpetiolar process anteriorly triangular and posteriorly an- gular. Volume 4, 1995 169 Ecology. — This is an epigaeic forager known only from primary and secondary wet forests of the Guiana Shield and its remnants in the Venezuelan Amazonas, mostly between altitudes of 850-1600 m. Within the range of 1000-1200 it can be locally abundant. One record from 550 is next to Angel Falls. This suggests that proximity with the higher forested talus slopes of Auyan Tepuy could explain its ocurrence at such a low elevation. This species obviously prefers cool, mesic hab- itats. This distribution plus climatic evi- dence (Schubert 1988) suggests that G. gracilis may have been more widely dis- tributed in this area during the last glacial period. Comments. — On account of size, similar sculpture, posteriorly inclined node and fair amount of pilosity, its nearest relative appears to be pilosa, also an apparent en- demic, of the Colombian Cordillera N and NNW of Cali. An occasional specimen may have a second, smaller seta just be- low the prominent seta of the foretarsal base. Specimens Examined. — VENEZUELA, Bolivar: Auyan Tepuy, Sector Aonda, 6°02'N 62°37'W, 1600 m; Auyan Tepui, Cafion del Diablo, Isla Raton, 550 m; San Ignacio de Yuruani, 5°00'N 61°08'W, 800 m (MIZA). Gnamptogenys haenschi Emery Gnamptogcmis haenschi Emery 1902:27; Brown 1958:302;"Kempf 1972:112. Diagnosis. — Vertexal margin concave in frontal view; mandibles with longitudinal costulae; declivitous propodeal face with longitudinal striae superiorly and basally transverse rugae; petiolar node transverse; postpetiolar sternum mostly transversely strigulate; no metacoxal tooth. Ecology. — This species inhabitats forest and apparently nests in rotten wood as well as in soil. It has also been taken in oil palm plantations. Some specimens were collected from carrion traps baited with iguana meat. Longino (pers. comm.) re- ports finding a nest beneath a rotten log; the ants burrowed into the soil on expo- sure. Comments. — This species is possibly the only survivor of an otherwise extinct lin- eage in the genus. See discussion on phy- logeny of the different species groups. Specimens. — BOLIVIA, Lower Rio Mad- idi; El Montero, 70 km N Sta. Cruz. CO- LOMBIA, Cauca: Isla La Gorgona; Guaji- ra: Quebrada Guacoche, vie. Don Diego, 10 m; Magdalena: 4 km N San Pablo, 10°57'N 74°03'W, 550 m. COSTA RICA, Peninsula de Osa: Corcovado, 8°28'N 83°35'W; Heredia: La Selva; 3 km S Prto. Viejo, 10°26'N 84°00'W, 50m. ECUADOR, Napo: Prto. Misahualli, 350 m. PERU, Tin- go Maria: Valle de Monson; Puerto Mal- donado, 260m; Loreto: Ramon Castillo, 5 km NW Leticia. PANAMA, Darien: Rio Tacaruna, 580m. VENEZUELA, Alto Rio Siapa, 1°40'N 64°35'W, 530m. Gnamptogenys hartmani (Wheeler) Ectatoniuw {Parectatomma) hartmani Wheeler 1915:390. Gnamptogenys txartmani (Wheeler); Brown 1958: 228, 302; Echols, 1964:137; Kempf 1972:112; MacKay 1988:127. Gnamptogenys nigrifnvis Borgmeier 1948; Brown 1958: 228, 236; Kempf 1972:114. NEW SYN- ONYMY. Gnamptogenys turmalis Kempf and Brown 1968: 93; Kempf 1972:116. NEW SYNONYMY. Diagnosis. — Superolateral corners of de- clivitous propodeal face with small lobes or carinae; mandibles triangular to sub- triangular; metanotal groove vaguely im- pressed and posterior nodal face has lon- gitudinal costulae; metacoxal dorsum with lobes; subpetiolar process subquad- rate. Ecology. — One record from soil in a ba- nana farm and another from soil in a de- stroyed Trachymyrmex nest. These ants have on several occasions been reported as predators of Trachymrmex ants (Echols, 1964:137; Kempf and Brown, 1968:94). J. 170 Journal of Hymenoptera Research Longino (pers. comm.) reports from Costa Rican field notes of Dana Myer: "a nest was found found in leaf litter amidst the remains of a Trachymrmex nest and many cut up workers and a queen of the attines were also found along with many wound- ed Gnamptogenys workers." Longino has observed this species carrying its brood in a loose 3 m column, fleeing from raiding Eciton. One specimen was taken from the stomach contents of Dendrobates lecomelns. Comments. — Kempf and Brown de- scribed G. turmalis as being close to G. ni- grifrons but slightly larger and more ro- bust, with finer sculpturing and a uniform ferruginous color. These authors realized that G. hartmani, G. nigrifrons and G. tur- malis were so close as to constitute possi- ble synonyms, and they were right. The study of specimens at hand show colors vary enough to make it an unreliable char- acter for separating species and the same is true for irregularities in the sculpture. Specimens from more southern localities have finer costulation. Also found to vary continously was the length vs. width of the petiolar node, as well as other size in- dicators. Specimens Examined. — BRAZIL, Bahia, Ilheus-Itab., km 22, (CEPEC). COSTA RICA, Peninsula de Osa: Sirena, 8°28'N 83°35'W, 50m. GUYANA, Kartabo; HON- DURAS, La Lima, Zapote farm. MEXICO, Tamaulipas: 10 km W EL Encino, 23°N 08'W 99°10'W; PANAMA, Isla Barro Col- orado. PERU, Valle de Chanchamayo, 800 m. USA, Louisiana: Lucky; Texas: Bent- son-Rio Grande State Park, 30. VENE- ZUELA, Amazonas, vie. Cerro Ya-Pakana. Gnamptogenys haytiana (Wheeler and Mann) Spaniopone haytiana Wheeler and Mann 1911:11. Gnamptogenys hai/tiana (Wheeler and Mann); Brown 1958:228, 316; Kempf 1972:113. Diagnosis. — Petiole node disciform, with anterior and posterior faces more or less parallel to one another, dorsal and lateral nodal faces with transverse rugulae; me- tanotal groove barely impressed; propo- deal spiracles mounted on turrets at mid- height of the lateral edge of declivitous propodeal face. Ecology. — Taken from forest and coffee plantation leaf litter. The dramatic defor- estation of Hispaniola has considerably re- duced the range of this species. Comments. — Endemic to Hispaniola Is- land. On account of the yellowish color, more pronounced curvature of the second gastric segment and different sculpture, this species is outstanding amongst the other members of its species complex. It is possible that it represents an independent development from the strigata complex. The holotype was examined in the MCZC. Specimens Examined. — DOMINICAN REPUBLIC, Barahana: 2 km N Polo, 1000m. Gnamptogenys horni Santschi Gnamptogenys regularis horni Santschi 1929:475. Gnamptogenys horni Santschi; Brown 1958:228, 235, 316;' Kempf 1961:491; Kempf 1972:113; Lattke 1990:17. Diagnosis. — Subtriangular mandibles; dorsum of petiolar node with posterome- dian longitudinal costulae, transversely arched anteriorly; episternal costulae curve on to declivitous propodeal face. Body dark brown; legs and antennae fer- ruginous. Ecology. — A series was taken from the stomach of Colostethus nubicola. See Lattke 1990. Comments. — This common species has a wide range in the neotropics. There is fair size variation and smaller specimens usu- ally have more vermiculate costulation. The position of the propodeal spiracle can vary from mid-heigth to below mid-heigth and the metacoxal tooth from triangular to denticulate. Nests of small forms can sometimes be found in the same locality as larger forms. Intermediates were found in other localities. A specimen labelled as a "cotype" was examined in the MCZC. Volume 4, 1995 171 Specimens Examined. — BRAZIL, Amazo- nas: Rio Taruma; Benjamin Constant; Re- serva Ducke; via Manaus-Itacoatiara km 50; Manaus; Rondonia: Vilhema; Para: Ca- rajas; Utinga Tract vie. Belem. BOLIVIA, Caranavi, vie. radio, 800m. COLOMBIA, Choco, Rio Napipi. ECUADOR, Pichincha: Estacion Rio Palenque; Sucua; Morona: Santiago, Los Tayos; Esmeraldas: 48 km S Atacames. GUYANA, Kamakusa. PANA- MA, Isla Barro Colorado. PERU, Pangu- ana, 9°37'S 74°56'W, 220m; Tingo Maria: Valle del Monson. SURINAM, Tambah- redjo; Dirkshoop; La Poolle; Maripa- hueuei, Vank. VENEZUELA, Amazonas: Alto Rio Mavaca, 2°02'N 65°06'W, 200m; Alto Rio Siapa, 1°40'N 64°35'W, 600m. Gnamptogenys ilimani Lattke, new species Type Material. — Holotype worker. BO- LIVIA, 22 km N Caranavi, Vivero Ilimani, 1700 m, 22-VI-81, C. Kugler, leg. Primary forest clearing with Cinchona, nest in rot- ten wood. Deposited in the MIZA. Para- types: paranidotypic workers from the same nest as the holotype. Deposited in each of the following: BMNH, LACM, MUSP, MCZC, MIZA. Worker. — Holotypes (Paratypes) mea- surements: HL 0.90 (0.84-0.90); HW 0.78 (0.72-0.78); ML 0.30 (0.26-0.32); ED 0.14 (0.14-0.18); SL 0.70 (0.70-0.72); WL 1.20 (1.12-1.20) mm; CI 0.87 (0.84-0.88); SI 0.90 (0.92-0.97); OI 0.19 (0.18-0.27) n=5. Coarsely costulate strigata group spe- cies. With head in frontal view vertexal margin convex; anterior pronotal face with 3-4 costulae that sharply bend back laterally to become longitudinal; declivi- tous propodeal face with longitudinal cos- tulae; node anteroposteriorly compressed, its posterior face sharply marginate later- ally with a flat to slightly concave surface and effaced sculpture; anterior postpetio- lar face with 2-3 transverse costulae, rest longitudinal; postpetiolar sternum with longitudinal costulae; lateral mesocoxal face with oblique, rough costulae and that of metacoxae with rough rugulae; meta- coxal tooth low and triangular. Body black with yellow-brown legs and scapes; abun- dant appressed pilosity on legs and scapes, but no standing hairs. Qi/mj.— Measurements: HL 0.88 (0.88); HW 0.72 (0.74); ED 0.20 (0.18); SL 0.72 (0.72); WL 1.28 (1.20) n = 2. Differences from workers are the usual; caste differ- ences, though the node is more disciform. Male. — Unknown. Discussion. — This species is nearest to G. strigata and G. pittieri but they are sepa- rable on several on several accounts: the posterior petiolar node face is not sharply set off and has raised costulae; in dorsal view their node is anteriorly convex, not straight; standing hairs on the body and especially the scapes are abundant, scape pilosity is longer. In G. pittieri the anterior transverse costulae of the pronotum curve around gently at the sides. Etymology. — The species name alludes to the type locality. Gnamptogenys ingeborgae Brown Gnamptogenys ingeborgae Brown 1992:279. Diagnosis. — Mandibular masticatory margin denticulate, promesonotal suture present, at least partially, and the longi- tudinal costulae on the propodeal dorsum and postpetiolar tergite is frequently ill- defined to effaced. Ecology. — A specialized millipede feeder from Colombia. See Brown 1992. Comments. — See Brown 1992. Specimens Examined. — COLOMBIA, Cundinamarca, Bogota-Villavicencio road, km 79, 1100 m (type series: MCZC). Gnamptogenys interrupta Mayr Ectatomuia {Gnamptogenys) interruptum Mayr 1887:543. Gnamptogenys interruptum (Mayr); Mann 1922:3; Brown 1958:228, 303; Kempf 1968:377; Kempf 1972: 113; Lattke 1990:18. Diagnosis. — Sublinear mandibles. Ce- phalic dorsum, mesosoma, and gastric ter- 172 Journal of Hymenoptera Research gum I with longitudinal costulae; gastric tergum II smooth; pleura also with smooth patches; metacoxal tooth absent. Body reddish brown; legs and antennae ferruginous. Ecology. — Found in humid forests of lowlands and mountains (cloud forest). Taken from leaf litter samples and beneath bark of rotting logs on ground. Comments. — Lateral mesosomal costulae can be effaced to a variable degree and the second gastric segment can ocassionally have weak longitudinal costulae, medi- anly effaced. Declivitous propodeal face with longitudinal costulae and weakly de- veloped anterolateral lobes. Specimens Examined. — COLOMBIA, Magdalena: 3 km SE Minca, 11°08'N 74° 06'W, 1050m; Valle: Puerto Merizalde, 10 m. COSTA RICA, Puntarenas: Montever- de, 10°18'N 84°48'W, 1500m. HONDU- RAS, Lombardia; JAMAICA, Mandeville. Gnamptogenys kempfi Lenko Gnamptogenys kempfi Lenko 1964:257; Kempf 1972:114. Diagnosis. — Mandibles subtriangular, dorsally smooth and shining, transverse costulae on mesosomal dorsum and node, no apparent transverse sutures on meso- somal dorsum. Propodeal spiracle large and metacoxal tooth present. Ecology. — Apparently from lowland for- est, one nest was found in rotten wood on the ground. Comments. — A widepread but uncom- mon species. Specimens Examined. — COLOMBIA, Amazonas: 7 km N Leticia. PERU, Pan- guana, 9°37'S 74°56'W, 220m (MCZC). Gnamptogenys lanei Kempf Gnamptogenys lanei Kempf 1960:388-90; Kempf 1968:377; Kempf 1970: 325; Kempf 1972:114. Diagnosis. — Relatively small eyes; clyp- eal lamella medianly concave; promeson- otal suture vestigial and small denticles on the propodeum; node elongate, with transverse costulae. Ecology. — Its morphology puts it in the rastrata group of millipede hunters. Comments. — Kempf 1968:377 reports a series with transverse costulae on the an- terior pronotal face, differing from the lon- gitudinal sculpture of the type series. MflfenVi/.— BRAZIL, Para: Belem, Rio Guama (MUSP). Gnamptogenys laticephala Lattke, new species (Figs. 54, 55) Type Material. — Holotype worker. EC- UADOR, Guayas: 3 km S Bucay, 24-VII- 73, W.L. Brown, leg. Deposited in MCZC. Worker. — Holotype measurements: HL 1.71; ML 1.01; HW 1.94; SL 1.53; ED 0.34; WL 2.32 mm; CI 1.13; SI 0.79; OI 0.18. Head in frontal view broad, vertexal margin fairly straight, lateral margins slightly convex and anteriorly diverging; eyes large and situated at cephalic mid- length; anterior clypeal lamella broadly concave with lateral triangular teeth near mandibular insertions; clypeus with two small anteriorly projecting lobes between anterior border of antennal fossae and an- terior lamella; cephalic dorsum with lon- gitudinal rugae that diverge slightly pos- terad on to vertex; rugae between frontal lobes and eyes more irregular than rugae between and behind frontal lobes; abun- dant piligerous punctures present. Anten- nal scapes smooth and shining, surpassing vertexal margin; gula with median longi- tudinal carinae and transverse costae at each side; mandibles falcate, longitudinal- ly costulate with interior glabrous sulcus that ends at apical tooth, basal tooth marks the end of basal margin and is fol- lowed by two more pre-apical teeth; an- terior pronotal face opaque, not as shiny as rest of pronotum and with faint trans- verse rugulae; mesosomal dorsm with parallel longitudinal costulae, promeson- otal suture well impressed but does not break longitudinal sculpture, mesometa- Volume 4, 1995 173 Fig. 54-55. Micrographs of G. laticcplmla. Scale bars =1.0 mm. 54, head; 55, body. 174 Journal of Hymenoptera Research notal suture deep and interrupting sculp- ture except for some ridges that join a few costulae on either side of suture; lateral pronotal face with slightly oblique parallel costulae; pronotal-mesopleural suture very broad and deep; meso- and metepis- terna with parallel, longitudinal costulae; mesometepisternal suture well impressed and metepisternal-propodeal suture dis- tinct; metepisternal lobe well developed; declivitous propodeal face with longitu- dinal costulae; petiolar node in lateral view low, anterodorsal margin convex and posterodorsal margin sharply convex; transversely costulate; subpetiolar process anteriorly triangular, with slighty round- ed posterior lobe. Postpetiolar costulation longitudinal, weakly roughened and becoming rugulo- se laterally, ventrally with transverse cos- tulae that tend to fade medially; gastric tergum II strongly vaulted anteroventrally with dorsal and lateral sculpture as on preceding segment; apical gastric seg- ments with faint transverse strigulae, tending to smooth and shining; anterior face of procoxae smooth and shining lat- erally with oblique costulae; meso- and metacoxae laterally smooth and shining, dorsally transversely costulate; metacoxal teeth well developed; body with sparse suberect and decumbent hairs, longest on gastric apex and clypeus, no appressed pi- losity on scapes, just suberect and decum- bent hairs. Mesosoma dark reddish brown; head, node and gaster darker; legs and antennae brown; mandibles brown to yellow brown. Queen, Male. — Uknown. Discussion. — The two lobes between the antennal fossae and clypeal lamella distin- guish this species from all other New World members of the group, which lack such lobes, including its closest relative, banksi. Other characters in which banksi differs are the following: areolate sculp- ture on the cephalic dorsum, occiput and anterior pronotal face with transverse ru- gae, deep lateral and mesonotal pits for the spiracles (such pits lacking in frauda- trix), pronotal suture interrupts sculpture, and the inner mandibular mandibular sul- cus only reaches the basal angle. In gen- eral the sculpture of banksi is rougher, mostly of gaster I, which is reticuate ru- gose and gaster II has rough, widely spaced, dorsal longitudinal rugae, becom- ing more irregular laterally. The form of the pronotal-mesopleural suture and the lobes formed by the pronotum and anepi- sternum at the base of the suture are rem- iniscent of some minuta group species. The possibility exists that this could be an ab- errant specimen of banksi (it was found de- termined as such), but given the observed differences I have opted for the most con- servative course. It is hoped that the cap- ture of additional specimens may throw more light onto the status of laticepJiala. Etymology. — The name alludes to the relatively broad head when seen in frontal view. Gnamptogenys lineolata Brown Gnamptogenys lineolata Brown 1992:275. Diagnosis. — Head in frontal view su- bquadrate, scapes smooth and shining; mandibles denticulate; sculpturing of body striate, lacking propodeal denticle. Ecology. — A specialized millipede feeder endemic to Hispaniola. Comments. — See Brown 1992. Specimens Examined. — DOMINICAN REPUBLIC, La Vega, La Cienaga, 1100 m. Gnamptogenys lucaris Kempf Ectatotwna tornatum Luederwalt 1926:236 (nee Roger 1862). Gnamptogenys lucaris Kempf 1968:379; Kempf 1972:114. Diagnosis. — Mandibles subtrianglar; clypeal lamella laterally bluntly rectangu- lar, not acutely dentate; propodeum with longitudianal costulae; node low, its de- clivitous face narrow. Ecology. — Luederwalt (1926) reports a Volume 4, 1995 175 Specimen on a shrub devouring the re- mains of a beetle. Comments. — This species is very close to G. siapensis and the possiblity exists that they are conspecific. Specimens Examined. — None. Gnamptogenys mecotyle Brown Gnauipfogcm/s mecotyle Brown 1958:318; Kempf 1972:114.' Diagnosis. — Promesonotal suture and metanotal groove impressed but not breaking sculpture; propodeal spiracle opening large and round, right behind teeth; metacoxal teeth long and sharp; oc- cipital carinae wide and visible in frontal view; anterior clypeal lamella laterally rounded and anteriorly straight to very broadly convex. Ecology. — Millipede feeder captured in litter samples from wet forest. Some spec- imens found in stomach contents of Den- drobates histroniciis. Comments. — The cephalic sculpture can vary from evenly costulate to roughly ru- gose and the petiole length vs. width is also variable. Specimens Examined. — COLOMBIA, An- tioquia, Providencia, Estacion Biologica; Choco: Serrania de Baudo, Camino de Yupe, 500-700m; Amazonas: 7 km N Le- ticia. COSTA RICA, Puntarenas: 6 km SW Monteverde. 10°16'N 84°50'W, 900m. PANAMA, Isla Barro Colorado. Gnamptogenys mediatrix Brown Gnamptogenys mediatrix Brown 1958:362; Kempf 1972:n4.' Diagnosis. — Mandibles elongate and tri- angular, their basal % rugulose and apical Vi smooth and shining; clypeal lamella me- dianly concave; scapes with no erect hairs; petiolar costulation mostly transverse, node broader than long; metacoxal teeth very slender, propodeal teeth short. Ecology. — Apparently of lowland for- ests, one series taken from a varzea habi- tat. Its morphology puts it within the ras- trata group, and it is quite probably a mil- lipede hunter. Comments. — Rarely collected, it is a close relative of G. laticephala. Specimens Examined. — BRAZIL, Amazo- nas: Ilha de Curari; Para; Igarape Mana (MCZC). Gnamptogenys menozii (Borgmeier) Ectatomma {Parectatomma) menozzi Borgmeier 1928:32 Gnamptogenys menozzi (Borgmeier); Brown 1958: 228, 316, 321; Kempf 1972:113 Ectatomma {Parectatomma) schubarti Borgmeier 1948:198. NEW SYNONYMY. Gnamptogenys schubarti (Borgmeier); Brown 1958:316; kempf 1960a:390; Kempf 1972:114. Diagnosis. — Eyes convex; mandibles; promesonotal suture impressed but not deep, metanotal groove very deep; pro- podeal teeth short; mesosoma longitudi- nally costulate. Ecology. — Found in forested areas, in- cluding montane forests. Comments. — Borgmeier described G. schubarti upon characters always "a bit more" than for G. menozzi and it is curious to note how authors frequently refer to these two forms together when discussing their differences from other species. Brown (1958:316) discusses differences be- tween the two forms: size, number of cos- tulae, color and the shape of the subpetio- lar process. He also admits that the forms could belong to one species. Size differ- ences between the two forms overlap and the color and gauge of the costulation are also continuously variable. The differences in the shape of the subpetiolar process are well within infraspecific variation, at least as observed in other Gnamptogenys. Specimens Examined. — BRAZIL, Rio Grande do Sul {menozzi cotype: MUSP); Sao Paulo: Monte Alegre, 900m {schubarti cotype: MUSP); Rio Corocovado; Espiritu Santo: Res. Nova Lombardia, 4 km N Santa Tereza, 900 m. 176 Journal of Hymenoptera Research Gnamptogenys mina (Brown) Holcoponera mina Brown 1957:494. GnampHogeiJi/s mina Brown 1958:220, 228; Kempf 1972:114;'Lattke 1990:18. Diagnosis. — Mandibular dorsum smooth and shining and apex of scapes surpassing the vertexal margin; declivitous propodeal face flat and separated from the dorsal face by a sharp angle; propodeal spiracles elevated at the apex of cylindrical tuber- cles. Ecology. — Found in leaf litter and earth samples from humid forests. Comments. — The series from near Ka- vanayen show very sharply defined cos- tulae with fine vermiculation in contrast with the glazed and effaced aspect of oth- er series. Specimens Examined. — COLOMBIA, Pu- tumayo: Villa Garzon (MCZC). VENE- ZUELA, Bolivar: 10 km E Kavanayen, 1200 m (MIZA). Gnamptogenys moelleri (Forel) Holcoponera moelleri Forel 1912:34; Mann 1916: 404. Holcoponera teffensis Santschi 1929:449. NEW SYNONYMY. Holcoponera teffensis var. concinna Santschi 1929: 449 (nee. F. Smith 1858); Brown 1958:229. Holcoponera moelleri var. splendens Santschi 1929: 450; Brown 1958:228. Gnamptogenys moelleri (Forel); Brown 1958:329; Kempf 1972:113; Kempf 1976:51. Gnamptogenys teffensis (Santschi); Brown 1958: 329; Kempf 1972:115; Lattke 1990:24. Diagnosis. — Slightly roughened costu- lae; pronotum with concentric costulae; posterodorsal mesosomal face with longi- tudinal costulae; subpetiolar process sub- quadrate, sometimes anteriorly projecting into a lobe. Ecology. — Epigaeic forager of mostly lowland forested areas. Comments. — Comparison of specimens determined as G. moelleri with G. teffensis forms reveal nothing beyond normal in- fraspecific variation. Santschi described G. teffensis as a relative of G striatida differ- ing in gauge of costulation, pilosity, and size. The use of length vs. width of the funnicular segments in separating G moel- leri from G teffensis is useless as even in members of the same nest series the ratio will differ either way. Specimens Examined. — ARGENTINA, 50 km S. Oran. BRAZIL, Amazonas: Fazenda Esteio, 80 km NNE Manaus, 2°25'S 59°46'W, 80m. COLOMBIA, Amazonas: 7 km N Leticia; Putumayo: 5 km S Mocoa, 610m; Meta: Rio Guayabero, Isla Angos- tura, 279 m; Reserva La Macarena, 580m; Choco: Finca Los Guaduales, 800m; Nari- fio: 28 km SE Mocoa, 510m. ECUADOR, Napo: Limoncocha, 250m; 59 km E Quev- edo, Los Rios, 2000m; PARAGUAY, Neembuca, Pilar. PERU, Panguana, 9°37'S 74°56'W, 220m; Puerto Maldonado, 260m. VENEZUELA, Barinas: 17 km SW Ciudad Bolivia, 8°04'N 70°48'W, 240 m. Gnamptogenys mordax (Smith) Ponera mordax F. Smith 1858:98. Ponera nodosa Latreille 1802:217; Brown 1958: 228. Ectatomma (Gnamptogenys) mordax (Smith); Em- ery 1896:49. Gnamptogenys mordax var. purensis Forel 1912; Brown 1958:228. Gnamptogenys mordax (Smith); Mann 1922:3; Brown 1958:319; Kempf 1972:113; Kempf 1976:51; Lattke 1990:19. Gnamptogenys mordax var. sebastiani Borgmeier 1937:220;'Brown 1958:228. Diagnosis. — Mandibles subtriangular; declivitous propodeal face transversely costulate; gastric tergum II can vary from smooth to longitudinally costulate or cos- tulate-rugose. Body very dark, brown, al- most black; legs dark brown. Smaller in- dividuals can be confused with G. contin- iia. Ecology. — A common dweller of humid forests. Cojnments. —Size (HW 1.08-2.08; WL 1.66-2.92 mm) and sculpture patterns of Volume 4, 1995 177 57 59 Figs. 56-59. Frontal view of head, and lateral view of mesosoma and petiole. 56-57, G. iliniani; 58-59, G. nigrivitrea. this widespread and timid species is quite variable. Specimens Examined. — COSTA RICA, Heredia: 17 km N. Volcan Barba, 10°17'N 84°05'W, 880 m. GUYANA, Kaietur. MEX- ICO, Chiapas: Jetja. PERU, Panguana, 9°37'S 74°56'W, 220 m; Tingo Maria: Mon- son Valley; 69 km E Tingo Maria, 1200m. VENEZUELA, Amazonas: Alto Rio Siapa, 1°40'N 64°35'W, 600m. Gnamptogenys nana Kempf Gnamptogein/s nana Kempf 1960b:422; Kempf 1972:113.' Parectatomma dina Kusnezov, 1969:35; Kempf 1978:35. Diagnosis. — Head subquadrate, mandi- bles subtriangular with striae at base, clypeal lamella straight, no transverse me- sosomal impressions, propodeal declivity with vertical costulae. Ecology. — Unknown. Its nearest rela- tives, horni and regularis include beetles as a significant part of their diets, while horni also preys on ants. Comments. — A rarely collected species, the specimen mentioned below is topotyp- ic (MCZC). Specimens Examined. — BRAZIL, Sao Pau- lo: Agudos. Gnamptogenys nigrivitrea Lattke, new species (Figs. 58, 59) Type Material. — Holotype worker: CO- LOMBIA, Valle: Reserva Forestal de Yo- toco, 3°50'N 76°25'W, 42 km NNE Call, 1100-1500 m, 6-1-84, W.P. MacKay, leg. No. 7221. Deposited in MIZA. Paratype: One worker with same locality data as ho- lotype, deposited in LACM. VSlorker. — Holotype (paratype) measure- 178 Journal of Hymenoptera Research merits: HL 1.10 (1.10); HW 0.86 (0.82); ML 0.86 (0.82); ED 0.14 (0.12); SL 0.82 (0.84); WL 1.40 (1.36) mm; CI 0.79 (0.79); SI 0.95 (1.02); OI 0.16 (0.15). Head elongate; fairly parallel sided in frontal view; posterior margin slightly concave; clypeal lamella widest medianly, giving it a bluntly point- ed aspect; eyes moderately protuberant. Mandibles triangular and shallowly cos- tulate; no trace of mesometanotal suture; pronotum anterolaterally convex; declivi- tous propodeal face longitudinally costu- late; metacoxal tooth triangular; node pos- teriorly inclined, in lateral view with a broadly convex anterior margin and a bluntly pointed apex which slightly over- hangs the posterior margin, in dorsal view with longitudinal costulae; subpetiolar process is an anteriorly projecting lobe; anterior face of postpetiole with 3 trans- verse costulae, costulation longitudinal on dorsal face; postpetiolar sternum with dis- tinct costulation; scape with few standing hairs and moderate amount of decumbent pilosity. Body black; antennae, legs, me- tapleural gland area, and gastric apex brown. Coxae very dark brown. Female, Male. — Unknown. Discussion. — The closest species to G. ni- grivitrea is G. hrunnnea, but its well im- pressed mesometanotal groove, angulate anterolateral pronotal area and depressed apical and basal mandibular margins per- mit easy separation. Etymology. — The species name is de- rived from the Latin adjectives for black, nigra, and for glassy, vitra, and alludes to the black and shining sculpture. Gnamptogenys perspicax Kempf and Brown Gnamptogenys perspicax Kempf and Brown 1970: 316; Kempf 1972:114. Diagnosis. — Eye slightly behind cephalic mid-length; scape longitudinally striate with abundant standing hairs; petiole node more or less evenly convex; postpe- tiolar sternum with transverse costulae or rugae. Ecology. — A millipede feeder found in humid forests. Comments. — Apparently a sister species of bispinosa. Specimens Examined. — ECUADOR, Pi- chincha, Estacion Rio Palenque (MCZC). Gnamptogenys pilosa Lattke, new species (Figs. 51-53) Type Material. — Holotype worker. CO- LOMBIA, Valle: Calima, Canon El Pital, F. Castano, leg. Deposited in MIZA. Para- types: Three workers with same locality data as holotype. One deposited in each of the following: LACM, BMNH, MCZC. The late Mr. F. Castano kindly provided additional information about the type lo- cality. The site is by a stream called Que- brada El Pital (tributary of Rio Calima) in a gorge 900-1300 m above sea level. Ap- proximate coordinates: 3°08'N 76°40'W. The site will soon be flooded due to con- struction of a dam. The specimens were found on leaf litter in a cloud forest. Worker. — Holotype (paratypes) mea- surements: HL 1.44 (1.40-1.48); HW 1.20 (1.20-1.22); ML 0.54 (0.46-0.54); ED 0.28 (0.28-0.30); SL 1.46 (1.50-1.57); WL 2.00 (2.00-2.06) mm; CI 0.83 (0.81-0.86); SI 1.21 (1.25-1.28); OI 0.23 (0.23-0.25) n=4. With head in frontal view, vertexal mar- gin fairly straight, sides broadly convex; anterior clypeal margin convex, lamella well developed; eyes prominent on pos- terior one half of head; mandibles trian- gular, apical margin denticulate, dorsum striate; cephalic dorsum with longitudinal parallel striae, diverging posterad at ver- tex; 1-2 transverse striae parallel to occip- ital carina; anterior prontal face trans- versely striate; dorsal mesosomal surface, including all of propodeum, longitudinal- ly striate; striae curve around propodeal spiracles with no transverse striae be- tween spiracles; lateral pronotal surface obliquely to transversely striate, striation Volume 4, 1995 179 on pleura and lateral propodeal face par- allel with dorsal striae; inner procoxal fac- es punctate, anterolaterally transversely striae; lateral faces of meso- and metacox- ae weakly striate; metacoxa with well-de- veloped tooth. Petiolar node transversely striate, low; viewed laterally with convex anterior margin and concave posterior margin, apex overhanging; subpetiolar process with a projecting anterior lobe and su- bquadrate posteriorly; transverse striae anteriorly on gastric tergum, up to three- fourths length of tergum, rest of gaster longitudinally striate; anteroventral post- petiolar process prominent; sternite with longitudinal striation, diverging caudad; abundant erect to suberect hairs on body and extremities; abundant pilosity on cox- ae, propodeum, petiole, antennae and legs. Queen, Male. — Unknown. Discussion. — This is the hairiest of all Gnamptogem/s species. On account of size, slender habitus, posteriorly inclined peti- olar node, fine costulation, and abundant pilosity its nearest reatives appear to be G. gracilis and G. ejuncida. The sculpture of ejuncida is coarser, the pilosity is not as dense and the presence of erect hairs is considerably less. The break between the dorsal and declivitous propodeal faces is sharper in G. pilosa, giving it a more ro- bust aspect. G. gracilis is an endemic from the Guiana Shield area, and has the post- petiolar sternal disc smooth and shining. Etymology. — The species name is de- rived from the Greek word for hair: pilos. Gnamptogenys pittieri Lattke Gnamptogenys pittieri Lattke 1990:21. Diagnosis. — Sculpture with rough as- pect; frontal lobes each with a median emargination; eyes protuberant and subgl- obular; propodeal spiracles mounted at the apex of protuberances. Ecology. — Hypogeic leaf litter (including bamboo) dweller of cloud forests of the Venezuelan Cordillera de la Costa. Comments. — The series from near Tejer- ias differs from the type series in having the notch on the frontal lobes shallower, the propodeal spiracles on higher turrets, finer costulae on the anterior postpetiolar face and metacoxal spines more acute. Specimens Examined. — VENEZUELA, Aragua: Pq. Nac. Henri Pittier, vie. Ran- cho Grande, 1500m; 17 km S Las Tejerias, 1300 m (MIZA). Gnamptogenys pleurodon (Emery) Ectatomma (Holcoponera) pleurodon Emery 1896: 47. Holcoponera emery i Santschi 1929: 463 (nee Forel 1901). Holcoponera vidua Santschi 1929:467; Brown 1958:229. Gnamptogenys pleurodon (Emery); Brown 1958: 320; Kempf 1961:390; Kempf 1970:325; Kempf 1972:114; Lattke 1990:22. Diagnosis. — Pronotum with concentric costulae; mesonotum with longitudinal costulae, sometimes surrounded by con- centric costulae; anterior postpetiolar face with 3-4 transverse costulae and dorsal face with longitudinal costulae; metacoxal tooth acicular. Body dark brown, legs brown. Ecologxj. — Kempf (1970) notes a nest preference for pre-existing cavities in plants. It is also a frequent forager on trees and shrubs. Usually found in lowland to premontane humid forests. Comments. — An occasional specimens may have vertical costulation on the de- clivitous propodeal face, as in striatula, but the backwards tilt of the petiolar node and its brief anterior peduncle will identify pleurodon. Specimens Examined. — BOLIVIA, Tumu- pasa. BRAZIL, Amazonas: Benjamin Con- stant, Fazenda Esteio, 80 km NNE Ma- naus, 80 m; Iriboca, Pirelli Plantation. CO- LOMBIA, Amazonas: 7 km N Leticia; Pu- tumayo: 5 km S Mocoa, 610m. ECUADOR, Napo: Sushijindi; Tena, 400m. PERU, Ma- 180 Journal of Hymenoptera Research dre de Dios: Estacion Biologica Cocha Ca- shu, 400; Tingo Maria: Yurac, 108 km E Tingo Maria. SURINAM, Brownsberg Na- tuur Park. VENEZUELA, Amazonas, Alto Rio Mavaca, 2°01'N 65°07'W, 200m; Boli- var, Auyan Tepui, Cafion del Diablo, Isla Raton, 500m. Gnamptogenys porcata (Emery) Holcoponera porcata Emery 1896:48. Holcoponera uiaguifica Santschi 1921:81. NEW SYNONYMY. Gnamptogenys magnifica (Santschi); Brown 1959: 320; Kempf 1972:113. Gnamptogenys porcata (Emery); Brown 1958:320; Kempf 1972:114; Lattke 1990:23. Diagnosis. — Costate species, median lon- gitudinal costae on pronotum inscribed anteriorly by transverse costae; postero- dorsal mesosomal sculpture variable: con- centric ovaloids with longitudinal or transverse orientation, or with longitudi- nal costae inscribed within whorls; pice- ous body with ferruginous legs. Ecology. — A dweller of premontane to montane humid forests. J. Longino (pers. comm.) found an incipent nest in Cecropia insignis. One nest midden contained most- ly the remains of beetles: scotylids, bostri- chids, curculionids, besides isopods. Nests from one population (Venezuela, Tachira) consistently had their brood chambers ir- regularly wall-papered with the remains of pupal cocoons. Comments. — Santschi described magnifi- ca as close to porcata but differing in its larger size and the transverse costulae on the posterior dorsal mesosoma. Brown (pers. comm.) studied the magnifica type in 1963 and noted it was "very close" to por- cata with the former slightly larger and with a higher, thinner node as seen later- ally. Specimens in the USNM from Bolivia collected by Mann during 1956 bear anon- ymous Holcoponera magnifica determina- tions. The study of series from Venezuela representing several colonies from differ- ent populations show great variability for dorsal mesosomal sculpturing. Size differ- ences between the two forms are no more than infraspecific. One Honduran worker has a wide, lobe-like subpetiolar process. Specimens Examined. — BOLIVIA (no oth- er data); COLOMBIA, Antioquia: Rio For- ce, 1020m; Cundinamarca: above Tena, 1300-1600 m; Valle: 3,2 km above Rio Agua CLara, old Cali-Buenaventura rd; Rio San Juan (tributary of Rio Digua), vie. Queremal, 1300m; Municipio Buenaven- tura, 650 m. COSTA RICA, Rio Toro Am- arillo, vie. Guapiles, < 40 m; Heredia: 10°19'N 84°43'W, 800m. ECUADOR, Pi- chincha: Tinalandia, 16 km SE Sto. Domin- go de los Colorados, 680m. HONDURAS, 14 km S La Ceiba; Lombardia. VENEZUE- LA, Tachira: vie. Siberia, 39,7 km WNW San Cristobal, 1200m; Trujillo: 15 ESE Bo- cono, 1160m. Gnamptogenys rastrata (Mayr) Ectatomma (Parectatomma) rastrata Mayr 1866:89. Gnamptogenys trigona Emery 1905:114; Brown 1958:321; Kempf 1972:116. NEW SYNONY- MY. Gnamptogenys rastrata (Mayr); Brown 1958:322; Kempf 1972:114; Kempf 1976:52. Diagnosis. — Antennal scapes surpass vertexal margin by no more than their api- cal width; triangular and edentate man- dibles with a varying degree of costulae and rugae; small tubercle-like propodeal teeth next to conspicuous spiracles which are slightly elevated above rest of sur- rounding cuticle; postpetiolar sternum with median longitudinal smooth and shining areas. Ecology. — Captured in wet forests, in- cluding montane areas. A millipede feed- er. Comments. — G. trigona was separated from rastrata on account of differences in the gauge and form of the costulation, as well as size. Variability of mandibular sculpture on basal flange and dorsum is nothing beyond infraspecific. The sculp- ture on the node as seen dorsally can vary from concentric costulae, with transverse or longitudinal costulae in the middle, or Volume 4, 1995 181 completely transverse. Costulae transverse on anterior petiolar node face. Brown (1958:321) discusses the differences in size, costulae counts, length vs. width of node, subpetiolar process shape and variability of sculpture on node between the speci- mens of G. rastrata and G. trigona he ex- amined. As more specimens became avail- able the gaps were bridged. The petiolar node can be wider tha long or the oppo- site. Size range is HL 0.74-0.90; WL 1.10- 1.28 mm. Specimens Examined. — Brazil, Rio de Ja- neiro: Jussaral; Sao Paulo: Sao Bernardo do Campo; Nova Teutonia, 27°11'S 52°23'W, 300-500m (MCZC). Gnamptogenys regularis (Mayr) Gnamptogenys regularis Mayr 1870:965; Mann 1922:3; Brown 1959:229, 319; Kempf 1972:114; Lattke 1990:23. Gnamptogenys rimulosa var. splendida Pergande 1896:871; Brown 1959:229. Gnamptogenys fiebrigi Forel 1909:253; Brown 1958:229.^ Diagnosis. — Declivitous propodeal face abruptly separated from dorsal face, su- perolaterally with small angulate lobes; metacoxal tooth apically rounded and not triangular; dark brown body, testaceous legs. Ecology. — J. Longino (pers. comm.) re- ports observing a short linear column of workers walk to an apparent nest entrance on the ground. He excavated it before any activity had begun and found a small col- ony of Pseudomyrmex boopns. The related G. horjti has a dietary preference for ants and beetles (Lattke, 1990). One specimen was removed from the stomach of a Tamandua tetrndactylus. Comments. — The degree of transverse costulation on the anterior nodal face has traditionally been used to separate this species from horni, but as additional ma- terial accumulated it became impossible to rely on the character. Transverse costula- tion can be totally absent or occupy all of the anterior nodal face. Females tend to have the anterior nodal face totally trans- versely costulate. This was noted for one specimen by Mayr (1871:965) who con- cluded it was not normal and that a fe- male with longitudinal costulae would eventually be found. The longitudinal cos- tulae on the sides of the mesosoma do not curve onto the declivitous propodeal face. The type of splendida was examined in the CASC. Specimens Examined. — ARGENTINA, Tintina. BRAZIL, Amazonas: Reserva Ducke, vie. Manaus; Maturaca. COLOM- BIA, Magdalena: 4 km N San Pedro, 10°57'N 74°03'W, 550 m. COSTA RICA, Alajuela: vie. Volcan Arenal, 10°29'N 84°44'W, 550 m. Heredia: 10°20'N 84°04'W, 500 m; Peninsula de Osa: Sirena, 8°28'N 83°35'W, 50m; Puntarenas: Res. Biol. Carara, 9°47'N 84°36'W, 500m; vie. Guapiles, Rio Toro Amarillo. BRAZIL, Goias: Anapolis; Sao Paulo: Holambra; Agudos; Distrito Federal: Deodoro; Ama- zonas: Maturaca; Minas Gerais: Carmo da Cachoeira; Bahia: Salvador. ECUADOR, Pichincha: Rio Palenque, Centro Cientifi- co; Santo Domingo de los Colorados; Pi- chilingue: Los Rios. GUIANA, (NY Quar- antine Stn.). MEXICO, Tepic; Veracruz: 10 km NNW Sontecomapan, 200m. PERU, Tingo Maria: Valle del Monson; Junin: Co- lonia Perene, Rio Perene, 18 km NE La Merced; Puerto Maldonado, 260 m. VEN- EZUELA, Sucre: El Rincon, 10°36'N 63°12'W, 100 m; Guarico: Hato Masaguar- al. Gnamptogenys reichenspergeri (Santschi) Acanthoponera (Anacanthoponera) reichenspergeri Santschi 1929:274. Gnamptogenys reichenspergeri (Santschi); Brown 1958:324; Kempf 1972:114. Diagnosis. — Head in frontal view with concave vertexal margin, vertex smooth and shining; anterior face of postpetiole and median disc of gastric tergum II smooth and shining; metacoxal spine ab- 182 Journal of Hymenoptera Research sent and declivitous face of propodeum with small superolateral lobes, spiracles lateral and not elevated; metanotal suture absent. Ecology. — From leaf-litter samples of hu- mid forests. . Comments. — A rarely collected ant, its nearest relative is apparently relicta. Specimens Examined. — BRAZIL, Jussaral: Angra dos Rios (MCZC). VENEZUELA, Amazonas: Cerro Ya-pakana (LACM). Gnamptogenys relicta (Mann) Rhopalopone relicta Mann 1916:403. Holcoponera relicta (Mann); Brown 1957:491. Gnamptogenys relicta (Mann): Brown 1958:229; Kempf 1972:114; Wheeler & Wheeler 1975: 119. Diagnosis. — Mandibles, vertex and an- terior postpetiolar face smooth and shin- ing, occiput sometimes with very faint transverse costulae; last 3 antennal seg- ments form vague club; declivitous pro- podeal face with transverse costulae, pro- podeal spiracles elevated above rest of cu- ticle; pronotal suture present but fine, me- tanotal suture well impressed, breaking sculpture; no anterolateral lobes on declivitous propodeal face; metacoxal spine present. Ecology. — Sifted from leaf litter in rain forests. Comments. — The degree of effacement of sculpture, and size can vary considerably. S. Cover reports a specimen from the N. Range of Trinidad. Specimens Examined. — BRAZIL, Amazo- nas: Rio Taruma, High Falls; Benjamin Constant; Ponta Negra, N of Manaus; 66 km N Manaus on Carcari road. COLOM- BIA, Amazonas: 7 km N. Leticia. SURI- NAM, Raleigh Vallen-Voltzberg Res., Voltzberg Camp, 90m. VENEZUELA, Bo- livar: 49 km ESE Tumeremo, 7°28'N 61°06'W, 200 m. Gnamptogenys rimulosa (Roger) Ponera rimulosa Roger 1861:18. Gnamptogenys rimulosa (Roger); Brown 1958:324; Kempf 1972:114. Diagnosis. — Mandibles with small basal patch of costulae, rest smooth and shining; metacoxa with broad, basal lobe; petiolar node longitudinally costulate throughout; postpetiolar sternum smooth and shining. Ecology. — Unknown. Comments. — This species seems to be particularly close to G. regularis. Even though regidaris workers lack the small la- terobasal costulate area on the mandibles, females have it; there is a slight depression of the posterior mesosomal dorsum, ves- tigial propodeal lobes and totally longi- tudinal costulae on the node. Specimens Examined. — BRAZIL, Santa Catarina: Nova Teutonia: 27°11'S 52°23'W, 300-500m. Gnamptogenys schmitti (Forel) Emeryella schmitti Forel 1901:334; Wheeler & Wheeler 1952:127. Emeryella schmitti minor Wheeler 1936:195; Brown 1958:229. Gnamptogenys schmitti (Forel); Brown 1958:330; Kempf 1972:114. Diagnosis. — Very broad head, in frontal view, and striking elongate mandibles without denticles; striae on anterior nodal face transverse, slightly effaced laterally; striae on metanotum and propodeum transverse; gastric sternum II smooth and shining. Ecology. — A predator of millipedes whose range has shrunken considerably this century due to massive deforestation on Hispaniola. Comments. — This species represents an extreme in the development of falcate mandibles within the genus (Brown 1958: 215), but probably represents a develop- ment independent of the banksi subgroup species. The type series in the MCZC was examined. Specimens Examined. — HAITI, Diquini. Gnamptogenys sentiferox Brown Gnamptogenys semiferox Brown 1958:324; Kempf 1972:114. Volume 4, 1995 183 Diagnosis. — Body with abundant coarse eral view subquadrate; subpetiolar pro- punctation, especially on head; mandibles cess subquadrate, with angulate ventral with about six denticles and projecting corners; postpetiolar sternum medially basal lobe; anterior clypeal margin con- smooth and shining and with lateral Ion- vex; declivitous propodeal face longitudi- gitudinal costulae; tibiae and femora nally costulate. smooth and shining; metacoxae with basal Ecology. — A millipede hunter known triangular lobe, dorsally transversely cos- from rain forests between 900-1200 m on tulate and laterally mostly smooth and Hispaniola. Also suffering from habitat shining; abundant long suberect and sub- shrinkage due to deforestation. decumbent hairs on body. Color ferrugi- Comments. — Even though it shares the nous, lack of metacoxal spines and elongate Queen, Male. — Unknown, mandibles with schmitti, it does not seem Discussion. — This species is quite near to to be closely related to it. Its mandibles are lucaris and the differences between the different, including in cross section: shin- two species may be infraspecific: in lucaris ing and longitudinally rugulose, as op- the petiolar node dorsum is evenly con- posed to opaque and striate in schmitti. vex, the metacoxal tooth is broadly trian- Specimens Examined. — DOMINICAN gular and the subpetiolar process has REPUBLIC, Mt. Diego de Ocampo, 3^000 acutely pointed angles. Due to the great ft (Paratype). differences that separate the range of the two species, the apparent endemicity of Gnamptogenys siapensis Lattke, g^^^^ Gnamptogem/s to southern Brazil, new species ^^^ ^^xe scarcity of material, it seemed (Figs. 62,63) prudent to separate the forms until more Type Material. — Holotype worker. VEN- specimens are available. EZUELA, Amazonas: Alto Rio Siapa, Etymology. — The species name alludes 1°40'N 64°35'W, 600 m, 4-II-89, J. Lattke, to the type locality. leg. Sifted leaf litter sample. The specimen 1 . . • 11 fi ^^ Gnamptogenys stellae Lattke, was caught in an area occasionally flood- ^ f & y ed during the rainy season. Deposited in ^^ ^P \ MIZA. (P^§^- ^^' ^^^ Worker.— UL 1.13; ML 0.44; HW 1.00; Type Material— Holotype worker. COS- ED 0.25; SL 0.97; WL 1.60 mm; CI 0.89; SI TA RICA, Estrella Valley, April, 1924, 0.97; Ol 0.25. W.M. Mann, leg. Deposited in USNM. Cephalic dorsum with fine longitudinal Worker. — Holotype measurements: HL costulae; eyes placed behind midlength; 1.18; ML 0.57; HW 1.06; SL 0.65; ED 0.13; clypeal lamella anteriorly straight and lat- WL 1.55 mm; CI 0.90; SI 0.61; OI 0.12. erally bluntly angulose; mandibles smooth Head elongate, in frontal view with con- and shining, subtriangular with apical and cave vertexal margin, more or less parallel basal margins joined through broad con- sided with eyes at about midlength; clyp- vexity; scapes smooth and shining, sur- eal lamella gradually projects forward passing vertexal margin, antennal bullae from sides and is medially weakly emar- partially visible beneath frontal lobes; cos- ginate; mandibles almost linear and with tulae longitudinal throughout mesosomal a double row of low tubercles on apical dorsum, propodeal declivity and petiolar border, dorsally and laterally smooth and node; promesonotal and mesometanotal shining with sparse punctae, basally with sutures weakly impressed; mesometa- small area of longitudinal costulae; scapes pleural suture well impressed; petiolar smooth and shining, dorso-ventrally corn- node more or less as long as wide, in lat- pressed, bent at basal one-fourth and 184 Journal of Hymenoptera Research Figs. 60-61. Micrographs of G. stellae. 60, head, scale bar = 0.5 mm; 61, body, scale bar = 1.0 mm. Volume 4, 1995 185 Figs. 62-65. Frontal view of head and lateral view of mesosoma and petiole. 62-63, G. siapensis; 64-65, G. volcano. thickest just apicad of midlength; cepahlic dorsum with longitudinal striae, diverg- ing posterad at vertex, gula mostly longi- tudinally striate, weakly diverging poster- ad; occipital margin glabrous. Anterior pronotal face and collar with transverse striae; longitudinal striae on mesosomal dorsum and propodeal decliv- ity, slightly shallower on metanotum and dorsal propodeal face; promesonotal su- ture very lightly impressed, metanotal groove markedly impressed and inter- rupting longitudinal striae. Mesosomal sides longitudinally striate, except for oblique striae on anepisternum; propodeal spiracle round and slightly raised above rest of integument; procoxae anteriorly and anterolaterally smooth and shining, pc)Sterolaterally striate; mesocoxae with dorsal transverse striae that are effaced posteriorly; metacoxae with transverse striae and low dorsobasal swelling; tibia and femora smooth and shining; anterior petiolar node face smooth and shining, laterally with weakly effaced longitudinal striae, dorsally diverging posterad; node wider posterad than anterad; petiole in lateral view slightly pedunculate, anteri- orly slightly concave, dorsally slightly convex, posterior face dropping sharply; subpetiolar process anteriorly projecting with convex anterior border and posteri- orly sinuous; gastric terga I and II with longitudinal striae, effaced towards pos- terior margin of tergum II. Gastric sterna I and II smooth and shining; pilosity sparse, a row of stout hairs present along apical mandibluar order, few standing hairs on posterior cephalic dorsum, pro- notum and gaster. Body reddish brown; mesosomal dorsum and head darker brown, leg and antennae ferruginous. Queen, Male. — Unknown. Discussion. — The median anterior pro- jection of the clypeal lamella is unique among the extant species of the mordax 186 Journal, of Hymenoptera Research group. This species seems to occupy an in- termediate position between G. continua and G. horni. The combination of clypeal configuration, small eyes, subfalcate man- dibles, glabrous occiput, mesometanotal suture and very brief petiolar peduncle point to this placement. Etymology. — The species epithet is de- rived from the Latin word for star, stella, and alludes to the type locality. Gnamptogenys striatula Mayr Gnamptogenys striatula Mayr 1883:32; Mann 1916:404; Brown 1959:327; Kempf 1970:325; Kempf 1972:115; Kempf 1976:52; Lattke 1990: 23. Holcoponera curtulum Emery 1896:47; Forel 1899: 7. NEW SYNONYMY. Holcoponera obscurum Emery 1896:48; Lueder- wait 1926:238; Santschi 1929:442. Holcoponera curtula var. stolli Forel 1899:7; Brown 1958:228. Holcoponera brasiliensis Emery, 1902:181; Brown, 1958:229. Holcoponera striatula obscura var. angustiloba For- el 1908:341; Luederwalt 1926:237; Brown 1958:229. Unavailable. Holcoponera striatula obscura var. simplicoides Forel 1908:341; Luederwalt 1926:328. NEW SYNONYMY. Unavailable. Holcoponera striatula obscura var. angustipleura Forel 1908:342; Brown 1958:229. Unavailable. Holcoponera curtula var. paulina Forel 1908:342; Brown 1958:229. Holcoponera curtula var. vollemoeideri Forel 1912: 33; Santschi 1929:453; Brown 1958:229. Holcoponera striatula (Mayr); Luederwalt 1926: 237; Santschi 1929:442; Wheeler and Wheeler 1952:123. Holcoponera striatula var. antillana Santschi 1929: 444; Brown 1957:490. Holcoponera rustica Santschi 1929:446. NEW SYNONYMY. Holcoponera wheeleri Santschi 1929:448. NEW SYNONYMY. Holcopwnera brasiliensis var. pernambucana Sants- chi 1929:452; Brown 1958:229. Holcoponera brasiliensis var. calcarata Santschi 1929:452; Brown 1958:229. Holcoponera brasiliensis var. mayri Santschi 1929: 453; Brown 1958:230. Holcoponera brasiliensis simplicoides Santschi 1929:45. Holcoponera brasiliensis simplicoides var. hybrida Santschi 1929:455; Brown 1958:229. Unavail- able. Holcoponera emeryi var. recta Santschi, 1929:465; Brown, 1958:229. NEW SYNONYMY. Holcoponera regularis arcuata Santschi 1929:457; Brown 1958:227; Kempf 1972:112. NEW SYN- ONYMY. Holcoponera regularis Santschi 1929:457 (nee Mayr 1970); Brown 1958:227. Holcoponera wasmanni Santschi 1929:466. NEW SYNONYMY. Holcoponera wasmanni var. isthmica Santschi 1929:467. Gnamptogenys arcuata (Santschi); Brown 1958: 227, 237; kempf 1972:112. Gnamptogenys curtula (Emery); Brown 1958:227, 327; Kempf 1972:112. Gnamptogenys simplicoides (Santschi); Brown, 1958:229,327; Kempf, 1972:114. Gnamptogenys rustica (Santschi); Brown, 1958: 229, 327; kempf, 1972:114. Gnamptogenys wasmanni (Santschi); Brown 1958: 229, 238; Kempf 1972:116. Gnamptogenys wheeleri (Santschi); Brown 1958: 230, 238; Kempf 1972:116. Diagnosis. — Pronotal costulae semicir- cular; dorsoposterior mesosomal costulae longitudinal and slightly diverging cau- dad; mesometanotum sometimes with semiciruclar costulae; triangular metacox- al teeth; body dark brown. Ecology. — A generalist epigaeic forager of humid forests. See Lattke (1990). A mite was found on the anterolateral petiolar side of one specimen from Beni, Bolivia. Comments. — This widespread, common, and fairly variable ant has been the object of a lengthy synonymic list. Most of the names were based upon differences in size, gauge of costulation, and the shape of the subpetiolar process and the meso- pisternal lobe. These size differences are all normal for the species. One can recog- nize coarsely costulate and finely costulate specimens, as well as intermediate forms. The finely costulate ants can be found in several places: s. Brazil, Argentina, Guia- Volume 4, 1995 187 na, Costa Rica, Jamaica and Hispaniola. Cundinamarca: 81 km de Bogota; Fusa- Occasional series can be found that have gasuga; Valle: Sevilla; Cali, 90m. COSTA transverse costulae on part or all of the RICA, Hacienda La Pacifica, 50m; Limon: propodeal declivity; this condition is not Finca La Lola, Siquerres; Heredia: Finca frequent on the finely costulate forms. La Selva; Puntarenas: 6 km SW Monte- Several series from Bolivia have a long an- verde, 10°16'N 84°50'W, 900m. DOMINI- terior slope of the node, which approaches CAN REPUBLIC, Clarke Hall. ECUA- the condition in pleurodon. The use of the DOR, Napo: Limoncocha, 250m; Sushijin- subpetiolar process in the separation of di: Sucua: 1500m. JAMAICA, St. Ann, Ce- species is of limited value as a variation dar V., 760m; Westmor. bog, 460m. EL from subquadrate to anteriorly projecting SALVADOR, 6,4 Km N Quetaltepec. lobe can be found in series from the same MARTINIQUE, (no other data). MEXICO, population or even nest. Brown (1957:489) Maria Madre; Michoacan: 8 km SW Tiqui- already discussed the futility of using the cheo, 430m; Consequina slope; Sinaloa: mesepisternal lobe. Mazatlan; Veracruz: Lago Catemaco, Specimens determined as G. curtula by 250m; 5 km S Cuernavaca; Chiapas: 8 km Brown for the 1958 revision are G. stria- NE Huixtla, 225m. PERU, Panguana, tula, and so are several specimens deter- 9°37'S 74°56'W, 220 m; Puerto Maldonado, mined as H. wasmanni var. isthmica. Spec- 260m. PUERTO RICO, 49 Carr. 186, km imens labelled as type series of H. emenji 13.2, Rio Grande, Humacao. VENEZUE- recta were studied in the USNM and LA, Amazonas: Alto Mavaca, 2°01'N found to be conspecific with striatula and 65°07'W, 200m; Tachira: Las Cuevas, not with G. pleurodon. I was able to ex- 7°48'N 71°46'W, 200 m. amine two G. striatula syntypes in the BMNH. Bill Brown generously shared the Gnamptgenys strigata (Norton) notes he took during his study of the fol- ,, , . • x m . lorr, a .. ^ . ^r^^^ TT I • Holcoponera stn^ata Norton 1871A. lowmg Santschi types m 1963: H. regularis, ^,i,,^,,,,,, conceutnca Mayr 1870:964; Emery H. regularis v. arcuata, H. rustica, and H. 1891-167 wheeleri. He concluded that they are all the Holcoponera simplex Emery 1896:46; Kempf and same species. He also wrote that these Brown 1968:90. specimens plus additional types and spec- Holcoponera satzgeri Forel 1908:39; Brown 1957: imens he studied (H. striatula v. antillana, 490. v. angustipleura, H. wasmanni, H. ivasmanni Holcoponera simplex spuria Forel 1908:39; Brown V. isthmica. H. brasiliensis v. calcarata, v. 1957:490. mayri, H. curtula v. paulina (det. Santschi)) Holcoponera simplex foreli Santschi 1929:460; perhaps could be separated into two spe- ^ Brown 1957:490. '^ . • ,1 1 \- 1 r> 1. Gnampto?enus sniiplex CEmery); Brown 195S:229. cies usmg the subpetiolar process. But as „ , x x /m / ^ u mco , , , , . ^, , . Gnampto^enys stri'^ata (Norton); Brown 1958: already has been menhoned, this process 339. ^^^^^ 1958:115; Kempf and Brown has proven to be an unreliable character i968:90; Kempf 1972:115; Billen 1986:168. for species determination in this group. Material Examined. — ARGENTINA, Sal- Diagnosis. — Roughly costate; anterior ta: El Rey; Misiones: Puerto Iguazu. BO- postpetiolar surface fairly flat and with 3- LIVIA, Beni: Blancaflor. BRAZIL: Para: 4 transverse costae; abundant long, sub- Tucurul; Sao Paulo: Reserva Caraguata- decumbent and suberect hairs on body, tuba, 40-80 m; Rio Grande do Sul: Caixa Ecology. — A dweller of wet montane do Sul, Nossa Senhora da Saude; Ceara: transandean forests, commonly taken in Maranganape Mts. COLOMBIA, Meta: Re- leaf litter samples. serva La Macarena, Rio Guayabero, 270m; Comments. — Aside from aspects covered Quebrada Chirijara, Villavicencio, 1400m; by Kempf & Brown (1968), other charac- 188 Journal of Hymenoptera Research ters that a vary are the size of the propo- deal spiracle, costulation of the propodeal declivity and the shape of the node. Specimens Examined. — COLOMBIA, Val- le: Pance, 1700m; Reserva Forestal de Yo- toco. COSTA RICA, Monteverde: 1520m; Puntarenas: 1300-1580m; Alajeula: Rio Penas Blancas, 800-880m; Heredia: 17-12 km N Volcan Barba, 880-1420m; San Jose: 2 km E San Gerardo, 9°27'N 83°43'W, 1440m. GUATEMALA, Sepacuite. MEXI- CO, Chiapas: Ocosinga; 19 km NW Oco- zocoautla, 975m; Veracruz: 5 km N Cuil- huac. Gnamptogenys sulcata (Smith) Edatomma sulcatum F. Smith, 1858:99. Ponera tomata Roger 1861:15. NEW SYNONY- MY. Gnamptogenys lineata Mayr 1870:965; Brown 1958:229.' Gnamptogenys sulcata var. cearensis Forel 1912: 33; Brown 1958:229 Gnamptogenys sulcata var. nitens Mann 1916:407; Brown 1958:229. Gnamptogenys ypirangensis Borgmeier 1928:229. Gnamptogenys sulcata bufonum Weber 1938:208; Brown 1958:229. Gnamptogenys sulcata (Smith); Brown 1958:329; Kempf 192:115; Lattke 1990:24. Gnamptogenys tomata (Roger); Mann 1922:3; Brown 1958:319, 329; Kempf 1968:378; Kempf 1972:116; Lattke 1990:25. Diagnosis. — Mandibles subtriangular; dorsal mesosomal costulae convergent caudad and transverse on declivitous pro- podeal face; metacoxal tooth absent, at most present as very small tubercle or short lobe; very variable color: mesosoma black to light brown, frequently head, gas- tric apex and posterior margin of gastric tergum II darker than rest of body. Legs antennae, and mandibles light to dark brown. Ecology. — Used as prey by Dendrohates histrionicus. See Lattke (1990). Comments. — Color is of no help in sep- arating the aforementioned forms, except at an occasional local level. Brown (1958) recognized these difficulties and made it clear the situation would need further study. Variation has been observed in size and shape of the subpetiolar process, mandibular width, and relative size of the eyes. The node can project posteriorly and approach the condition in acuminata but does not become acutely pointed. Even though the propodeal declivity is com- monly transversely costulate, it can some- times have oblique or even longitudinal costulae. I examined the sulcata type in the BMNH found it to have transverse costu- lae on the posterior nodal face. Roger's de- scription includes a key character that points to tornata's synonymy with sulcata: "The petiole, as in rimulosa, is concentri- cally costulate, with the smallest circle or oval in the middle of the dorsum, gradu- ally becoming larger." This I have inter- preted as meaning that the costulation on the posterior face of the petiole is trans- verse. My examination of Mann's nitens type reveals tansverse costulation on the posterior petiolar face, longitudinal cos- tulae on the declivitous propodeal face and very low triangular metacoxal teeth. Borgmeier's ypirangensis is described with transverse costulae on the posterior nodal face. Dr. Ivan Lobl of the MHNG kindly examined the cearensis type and reported transverse costulae on the posterior nodal face. Dr. Max Fischer of the NHMV gra- ciously examined the types of lineata and reported transverse costulation. What could constitute a separate species are specimens from the Amazonas Basin (MUSP) that are black with yellow man- dibles and are considerably larger than the average sulcata. But in the midst of so much variability it is prudent to await the accumulation of additional evidence be- fore coining a name for a few specimens. Specimens Examined. — BELIZE, Caves Branch. BOLIVIA, Rosario. BRAZIL, Ama- zonas: 24 km NE Manaus, Ig. Marianil, Rio Branco Rd.; Independencia: Parahyba; Para: Rio Xingu (Cachoiera do Espelho); 10 km N Tucurul; Porto Velho: Rio Ma- Volume 4, 1995 189 67 Figs. 66-67. G. transversa. 66, frontal view of head; 67, lateral view of mesosoma and petiole. deira. COLOMBIA, Casanare: (no other data); Magdalena: 12 km ESE Minca, 11°08'N 74°06'W, 780m; Parque Tayrona, Pueblito, 300m; Meta: Vista Hermosa; S Villavicencio, Cano El Buque, 480m; Na- rino: La Guayacana; Sucre, Est. Primates Inderena; Valle: 64 km E Buanventura, 570m. COSTA RICA, Sirena: Peninsula de Osa, 8°28'N 83°35'W, 50m; Pq. Nac. Brau- lio Carrillo, 500m; Peninsula de Osa, Pq. Nac. Corcovado, San Pedrillo, 0-lOOm; Puntarenas: 5 km N Ciudad Neily, 780m; Heredia, 10°20'N 84°04'W, 500m. ECUA- DOR, Estacion Rio Palenque; Pichincha: Tinalandia, 16 km SE Santo Domingo de los Colorados. GUATEMALA, Escuintla. HONDURAS, Ola, 18 km NE Catacamas, 370m. MEXICO, Veracruz: Lago Catema- co, 350m; Los Tuxtlas, 10 m NNW Sonte- comapan, 200m; trail above Presidio, 305- 490m; Cordoba. PANAMA, Barro Colora- do. PERU, Tingo Maria: Yurac, 107 km E Tingo Maria; Madre de Dios: Est. Biol. Co- cha Cashu, 400m; Rio Tambopata Res., 20°50'S 69°20'W, 290m; Huanuco: Cueva de la Lechuzas, 6 km W Tingo Maria Mar- ia; Cueva de la Boca del Lobo, 710m, 69 km W Tingo Maria. VENEZUELA, Ama- zonas: Alto Mavaca, 2°02'N 65°06'W, 250; Rio Baria, 0°50'N 66°10'W, 140m. Gnamptogenys tortuolosa (Smith) Poncra tortuolosa F. Smith 1858:99. Gnamptogeui/s tortuolosa var. quitensis Forel 1921:133; Brown 1958:230. Ectatoinma {Gnamptogeriys) tortuoloswn (Smith); Emery 1896:51; Mann 1916:406; Wheeler and Wheeler 1952:134. Gnamptogenys tortuolosa (Smith); Brown 1958: 230; Kempf 1961:492; Kempf 1970:325; Kempf 1972:116; Lattke 1990:25. Diagnosis. — Mandibles subtriangular; clypeal lamella laterally angular and me- dianly straight to softly concave; propo- deum with transverse costulae; coxal tooth small sometimes shaped as a short lobe; tibiae, femora and scapes smooth and shining. Piceous body; legs and an- tennae ferruginous. Ecology. — Found nesting in rotten wood in humid forests and foraging in grassy or weedy areas. Comments. — Very constant in sculpture and color. Specimens Examined. — BRAZIL, Rorai- ma: 64 km S Boa Vista. COLOMBIA, Ama- zonas: Trocha Buenos Aires; Meta: 8 km w Villavicencio; Mesetas, La Uribe, 720m; Serrania La Macarena. ECUADOR, Sucua. GUIANA, Kartabo; Bartica; Rupupuni: Apoteri, 4°05'N 58°35'W, 100m. VENE- ZUELA, Amazonas: Alto Rio Siapa, 1°05'N 58°35'W, 600m; Alto Rio Mavaca, 2°01'N 65°07'W, 200. Gnamptogenys transversa Lattke, new species (Figs. 66, 67) Type Material. — Holotype worker: PAN- AMA, Bocas del Toro, Fortuna-Chirigui 190 Journal of Hymenoptera Research Grande rd., 8°47'N 82°12'W, 12/14-VII-78, 1050m, D.M. Olson (523), leg. Premontane rainforest sifted leaflitter. Deposited in MCZC. Worker. — Holotype measurements: HL 1.30; HW 1.10; ML 0.70; ED 0.28; SL 1.04; WL 1.92 mm; CI 1.18; SI 0.95; OI 0.25. Head in frontal view with rectangular, elongate head: sides broadly convex, pos- terior margin slightly concave; clypeal la- mella relatively long, medianly slightly concave and laterally rounded; eyes mod- erately prominent. Costulae on head prin- cipally longitudinal, slightly convergent anterad, costulae between eyes and frontal lobes curving into depression laterad of antennal sclerite. Mandibles triangular with blunt denticles and relatively long basal broader, rugulose with piligerous punctae. Scapes shining and slightly ru- gulose, barely passing posterior edge. Me- sosoma dorsally with longitudinal costu- lae from pronotum to dorsum of propo- deum, declivity with transverse costulae; promesonotal suture visible as a brief transverse depression. Laterally with lon- gitudinal costulae throughout. Anepister- nal flange well-developed along anterior third; katepisternum well-defined, anepis- ternum not as well-defined. Petiole later- ally subcylindrical, costulae longitudinal/ oblique, ventral process is an anteriorly projecting lobe. Petiole in dorsal view wider posteriorly than anteriorly, costulae form concentric arches: transverse in oblique-anterior view and longitudinal in posterior view. Gastric dorsum with lon- gitudinal costulation, postpetiolar sternal disc smooth and shining. Transverse cos- tulae on forecoxa, metacoxa with a paral- lel-sided denticle with a rounded apex. Antennae, tibiae, femora with no pilos- ity, only decumbent to suberect hairs and sparse punctulae. Body brown, legs fer- ruginous brown. Female, Male. — Unknown. Discussion. — This species is very close to hartmani but hartmani has longitudinal cos- tulae on the propodeal declivity; mostly smooth and shining mandibles; dorsum of petiole with mostly longitudinal striae; longitudinal striae on the postpetiolar sternum; eyes not as prominent, and scapes with more defined striae and ru- gulae. Etxjmology. — The name alludes to the transverse costulae on the propodeal de- clivity. Gnamptogenys triangularis (Mayr) Ectatomma [Gnamptogenys) triangularis Mayr 1887:544. Gnamptogenys triangularis (Mayr); Emery 1905: 113; Kusnezov 1954:34; Brown 1958:323; Kus- nezov 1962:236; Brown 1958:230 321. Ectatomma (Parectatomma) triangularis richteri Forel 1913:203; Luederwalt 1926:236; Brown 1958:230 321. Ectatomma (Parectatomma) aculeaticoxae Santschi 1921:82; Wheeler and Wheeler 1952:133. NEW SYNONYMY. Gnamptogenys aculeaticoxae (Santschi); Brown 1958:227,^330; Kempf 1960:390; Kempf 1961: 491; Kempf 1972:116; Deyrup, et. al. 1989:93; Lattke 1990:8. Diagnosis. — Promesonotal suture weak- ly impressed; node dorsum with trans- verse costulae and subquadrate costulae and subquadrate subpetiolar process; first gastric sternum with transverse costulae; metacoxal tooth long and thin. Piceous body. Ecology. — Millipede feeder found in hu- mid forests. The USA record (Deyrup, et. al. 1989:93) is undoubtably a recent intro- duction and they have apparently found prey in the local species of millipedes, as they are well established. Comments. — The number of transverse costulae on the petiolar dorsum can vary from 8 to 14, and those on the pronotum from 13 to 23. Specimens from Argentina tend to have a higher count but there is no gap separating the values. The length of the coxal teeth is variable and bears no relation to the number of petiolar costulae. Propodeal teeth also show variation form a low mound to the usual low, sharp Volume 4, 1995 191 teeth. Occasional specimens can have up to 4 transverse costulae on the anterior pronotal face, and rarely longitudinal cos- tulae on the petiolar node. Other traits used by Santschi to separate aciilenticoxae, such as degree of impression of the pro- mesonotal suture, gastric constriction and gauge of hairs, length vs. width of petiolar node and the mandibular costulation all show continuous variation that is best de- scribed as infraspecific. Specimens Examined. — ARGENTINA, Tu- cuman: vie. Horco Molle, 750-900m; Tafe- cillo; La Cavera; 48 km S Jujuy. BOLIVIA, Huachi: Rio Beni. BRAZIL, Espiritu Santo: 4 km W Santa Tereza, Res. Nova Lombar- dia, 900m; Parana: Foz do Iguaqu; Sao Pau- lo: Agudos. COLOMBIA, Magdalena: 3 km SE Minca, ir08 74°06'W, 1050m; Pq. Tay- rona, Pueblito, 360m. COSTA RICA, He- redia: La Selva; Peninsula de Osa, Pq. Nac. Corcovado, 8°28'N 83°35'W, 0-lOOm. EC- UADOR, Pichincha: 47 km S Santo Domin- go, Rio Palenque, 215m. GUIANA, Kamak- usa. PANAMA, Barro Colorado. PERU, Tingo Maria: Yurac, 108 km E Tingo Maria; Puerto Maldonado, 260 m; Panguana, 9°37'S 74°56'W, 220m. USA, Florida: S. Mi- ami. VENEZUELA, Amazonas: Alto Rio Mavaca, 2°01'N 65°07'W, 200m. Gnamptogenys volcano Lattke, new species (Fig. 64, 65) Type Material. — Holotype worker. COS- TA RICA, Heredia: 18 km N Volcan Barba, 10°17'N 84°05'W, 800m, 4/14-VII-1986, J. Longino no. 1383-5. Wet forest litter sam- ple. Deposited in LACM. One antenna lacking. Worker. — Holotype measurements: HL 1.38; ML 0.55; HW 1.18; ED 0.25; SL 1.18; WL 1.93 mm; CI 0.86; SI 1.00; OI 0.21. Head in frontal view elongate, vertexal margin concave; eyes prominent and situ- ated at mid-length; clypeal lamella relative- ly long, laterally rounded and medianly concave; mandibles smooth and shining, with prominent piligerous fossae; subtrian- gular basal angle preceded by a tooth and concavity (or notch); scape when laid back barely surpasses vertexal margin; mesoso- mal dorsum with longitudinal costulae and propodeal declivity with transverse costu- lae, in dorsal view the costulae arch pos- terad and become longitudinal; gaster with longitudinal costulae, including postpetio- lar sternum, weakly effaced anteromedi- anly; metacoxae with dorsobasal tooth, dorsally with transverse costulae and lat- erally mostly smooth and shining; subpe- tiolar process with no posterior heel; tibiae and femora smooth and shining. Body with numerous standing golden hairs; body brown, legs and antennae reddish brown. Queen, Male. — Unknown. Discussion. — Gnamptogenys volcano is close to sulcata but the latter can be recog- nized by the longitudinal costulae on the propdeal declivity, broader head, rounded mandibular basal angle with no tooth and the straight clypeal lamella with acutely angulate sides. The structure of the clypeal lamella, mesepisternum, and mandible make this a striking species within the sul- cata group. Etymologi/. — The name of the species comes from the latin word for volcano, as the type locality is near the Barbas volcano. ACKNOWLEDGEMENTS I gratefully acknowledge the following indi\'iduals for providing specimens, logistical and moral support in field, lab, and museum: M.L. Baena, B. Bolton, C. Bordon, C.R. Brandao, W.L. Brown, F. Castano (R.I.P.), S. Cover, D. Davis, E. de Bellard, F. Fernandez, M. Fischer, A. Gentry, M. Harrison, K. Jaffe, D. Kistner, Lynda Lattke, I. Lobl, J. Longino, W.P. MacKay, A. Nunes, S. Peck, R. Robbins, E. Ross, D. Smith, R. Snell- ing, P. Ulloa-Chacon, G. Umphrey, and P.S. Ward. The following institutions are acknowledged for providing funds and logistical support: British Coun- cil, California Academy of Sciences, Fundacion para el Desarrollo de las Ciencias Fisicas, Matematicas y Na- turales, Fundacion Polar, Fimdacion Terramar, Insti- tuto de Zoologia Agricola (Universidad Central de Venezuela), Laboratorio de Comportamiento {Univer- sidad Simon Bolivar) and (especially) the Smithsonian Institution. 192 Journal of Hymenoptera Research I LITERATURE CITED Baroni Urbani, C. 1980. The ant genus Gnamptogenys in Dominican Amber. Stuttgarter Beitrcige ziir Na- turkunde Serie B 67:1-10. Billen, J. P. 1986. Comparative morphology and ul- trastructure of the Dufour Gland in ants. Ento- mologia Generalis 11:165-181. Borgmeier, T. 1928. Algumas formigas do Museo Paulista. Boletim Biologico, Sao Paulo 12:5-70. Borgmeier, T. 1929. Zur Kenntnis der brasilianische Ameisen. Eco 5:195-214. Borgmeier, T. 1937. Formigas novas ou pouco con- hecidas da America do sul e Central, principal- mente do Brasil. Arcliivos do Instituto de Biologia Vegetal, Rio de Janeiro 3:217-255. Borgmeier, T. 1948. Die Geschlechtstiere zweier Eci- to»-arten und einige andere Ameisen aus Mittel- und Siidamerika. Revista de Entomologia, Rio de Janeiro 19:191-206. Borgmeier, T. 1957. Myrmekologische Studien. I. An- ais do Academia Brasileira de Ciencias 29(1):103- 128. Brandao, C.R., J. Lattke. 1990. Description of a new Ecuadorean Gnamptogenys species with a discus- sion on the status of the Alfaria group. Journal of the New York Entomological Society 98:489-494. Brown, W.L., Jr. 1957. Notes on the ant genus Hol- coponera Mayr, with descriptions of two new spe- cies. Insectes Sociaux 3:489^97. Brown, W.L., Jr. 1958. Contributions toward a re- classification of the Formicidae. II. Tribe Ecta- tommini. Bulletin of the Museum of Comparative Zoologxi 118:(5):175-362. Brown, W.L., Jr. 1961. A note on the ant Gnampto- genys hartmani Wheeler. Psi/che 68(2-3):69. Brown, W.L., Jr. 1988. Data on Malpighian Tubule numbers in ants. pp. 17-27, in Trager, J., ed.. Ad- vances in Myrmecology. Brown, W.L., Jr. 1993. Two new species of Gnamp- togenys, and an account of millipedes predation by one of them. Psyche 99(4):295-289. Clapperton, C. 1993. Nature of the environmental changes in South America at the Last Glacial Maximum. Palaeogeography, Palaeoclimatology, and Palaeoecology 101:189-208. Deyrup, M., et. al. 1989. A preliminary list of the ants of Florida. Florida Entomologist 72(1):91-101. Donnelly, T.W. 1988. Geologic constraints on Carib- bean biogeography. pp. 15-37 in J. Liebherr, ed. Zooegeography of Caribbean Insects. Cornell Uni- versity Press, 285p. Echols, H. 1964. Gnamptogenys hartmani discovered in Louisiana. Annals of the E)itomological Society of America 57:137. Emery, C. 1890. Studi suUe formiche della fauna neo- tropica. Bolletino della Societa Entoniologica Italiana 22:38-80. Emery, C. 1894. Studi sulle formiche della fauna neo- tropica. VI-XVI. Bolletino della Societa Entoniolo- gica Italiana 26:137-243. Emery, C. 1896. Studi sulle formiche della fauna neo- tropica. XVII-XXV. Bolletino della Societa Entonio- logica Italiana 28:33-107. Emery, C. 1901. Notes sur les sous- families des do- rylines et ponerines. Annates de la Societe Ento- mologique de Belgique 45:32-54. Emery, C. 1902. Note mirmecologische. Rendito Ac- cademia di Scienza Istituto Bologna (n.s.) 6:22-34. Emery, C. 1905. Studi sulle formiche della fauna neo- tropica. XXVI. Bolletino della Societa Entoniologica Italiana 37:107-194. Forel, A. 1899. Formicidae. Biologia Centrali-Ameri- cana. Hymenoptera, 3:1-169. Forel, A. 1901. Nouvelles especes de Ponerine. Revue Suisse de Zoologie 9:325-353. Forel, A. 1908. Ameisen aus Sao Paulo (Brasilien), Paraguay, etc. Verliandhingen der zoologische-bo- tanische Gesellschaft , Wien 58:340-418. Forel, A. 1909. Ameisen aus Guatemala usw., Para- guay und Argentinien. Deutsche entomologische Zeitschrift pp. 239-264. Forel, A. 1912. Formicides neotropiques. I. Poneri- nae, Dorylinae. Annals de la Societe Entomologique de Belgique 56:28-49. Forel, A. 1913. Fourmis d'Argentine, du Bresil, du Guatemala et de Cuba. Bulletin de la Societe Vau- doise de Sciences Naturelles 49:203-135. Forel, A. 1921. Quelques fourmis des environs de Quito (Ecuador). Bulletin de la Societe Vaudoise de Sciences Naturelles 54:131-135. Kempf, W.W. 1960a. Insecta Amapaensia-Hymenop- tera: Formicidae. Studia Entoniologica 3:385-466. Kempf, W.W. 1960b. Miscellaneous studies on neo- tropical ants. Studia Entoniologica 3:417-466. Kempf, W.W. 1961. A survey of the ants of the soil fauna in Surinam. Studia Entoniologica 4:481-554. Kempf, W.W. 1966. New ants from southeastern and central Brazil. Studia Entoniologica 9:121-128. Kempf, W.W. 1967. Three new South American ants. Studia Entoniologica 10:353-360. Kempf, W.W. 1968. Miscellaneous studies on neo- tropical ants. IV. Studia Entoniologica 11:369^15. Kempf, W.W. 1969. Miscellaneous studies on neo- tropical ants. V. Studia Entoniologica 12:273-296. Kempf, W.W. 1972. Catalogo abreviado das formigas da regiao neotropical. Studia Entomologica, 15:1- 344. Kempf, W.W. 1976. Levantamento da formicifauna no Litoral Norte e ilhas adjacentes de Estado de Sao Paulo, Brasil. I. Subfamilias Dorylinae, Po- nerinae e Pseudomyrmecinae. Studia Entomologi- ca 19:45-66. Kempf, W.W. 1978. Five new synonyms for the Ar- gentine ant fauna. Studia Entoniologica 20:35-38. Kempf, W.W. and W.L. Brown. 1968. Report on I Volume 4, 1995 193 some neotropical ant studies. Papieis Avulsos de Zoologia 22(10):89-102. Kempf, W.W. and W.L. Brown. 1970. Two new ants of the tribe Ectatommini from Colombia. Stiidia Entonwiogica 13:311-320. Kugler, C. 1991. Stings of the ant tribe Ectatommini. hisecta Mundi 5:153-166. Kusnezov, N. 1954. Phyletische bedeutung der Max- ilar-und Labialtaster der Ameisen. Zoologische Anzeiger 153(l-2):23-38. Kusnezov, N. 1955. Evolucion de la hormigas. Dii- seuia 5(l-2):l-34. Kusnezov, N. 1962. El ala posterior de las hormigas. Acta Zoologica Lilloana 18:367-378. Kusnezov, N. 1969. Nuevas especies de hormigas. Acta Zoologica Lilloana, 24:33-38. Kusnezov, N. Hormigas Argentinas. Clave para su Identificacion. Fundacion Miguel Lillo. Publica- ciones Miscelaneas, 61:1-147. Latreille, P. 1802. Histoire naturelle des fourmis. Par- is, (not seen). Lattke, J. 1990. Revision del genero Gnatnptogemjs Mayr para Venezuela. Acta Terramaris 2:1-47. Lattke, J. 1991. Revision of the minuta group of the genus Gnamptogenys. Deutsche entomologische Zeitschrift 39:139-129. Lattke, J. 1994. Phylogenetic relationships and clas- sification of the Ectatommini. Entomologica Scan- dinavica 25:105-119. Lenko, K. 1964. Uma nova especie de Gnamptogenys de Mato Grosso. Papeis Avulsos de Zoologia 16(25): 257-261. Luederwalt, H. 1926. Observagoes biologicas sobre formigas brasileiras. Revista de Museo Paulista 14: 185-302. MacKay, W.P., W.P. Bradleigh Vinson. 1988. Redis- covery of the ant Gnamptogenys hartmani in East- ern Texas. Proceedings of the Entomological Society of Washington 9'[(\y.U7. MacKay, W.P., W.P. Bradleigh Vinson. 1989. A guide to the species identification of New World ants. Sociobiology 16(l):3-47. Mann, W. 1916. The ants oi Brazil Bulletin of the Mu- seum of Comparative Zoology 60:397-490. Mann, W. 1916. 1922. Ants from Honduras and Gua- temala. Proceedifigs of the Ihiited States National Museum 61(15):l-54. Mann, W. 1916. 1926. Some new neotropical ants. Psyche, 33:97-107. Mayr, G. 1866. Diagnosen neuer und wenig bekann- te Formiciden. Verhandlungen der zoologische-bo- tanische Gesellschaft 16:885-908. Mayr, G. 1870. Neue Formiciden. Verhandlungen der zoologische-botanische Gesellschaft 20:939-996. Mayr, G. 1883. Fourmis de Cayenne Franqaise par O. Radoszkowsky. Horae Societe Entomologique Ros- sicae 18:30-39. Mayr, G. 1887. Siidamerikanische Formiciden. Ver- handlungen der zoologische-botaitische Gesellschaft 37:511-632. Norton, E. 1868. Description of Mexican ants noticed in the American Naturalist, April 1868. Commu- nications of the Essex Institute 6:1-10. Pergande, T. 1895. Mexican Formicidae. Proceedings of the California Academy of Sciences 5:858-896. Perrault, G. 1986. Gnamptogenys falcifera Kempf, 1967, description de I'ouviere et levee d'un doute. Re- vue frangaise d'Entomologic 8:157-159. Roger, J. 1861. Die Ponera-artigen Ameisen. Berliner entomologische Zeitschrift 4:1-54. RuU, v., C. Schubert. 1989. Evolucion de las hipotesis sobre el origen del Caribe. Interciencia 14(2):74-85. Santschi, F. 1921. Ponerines, Dorylines et quelques autres formicides neotropiques. Bulletin de la So- ciete Vaudoise de Sciences Naturelles 54:81-103. Santschi, F. 1922. Descriptions de nouvelles fourmis de I'Argentine et pays limitrophes. Anales de la Sociedad Cientifica Argentina 94:241-262. Santschi, F. 1929. Revision de genre Holcoponera Mayr. Zoologische Anzeiger S2-A37-A77. Santschi, F. 1931. Fourmis de Cuba et de Panama. Revista de Entomologia l(3):265-282. Schubert, C. 1988. Climatic changes during the last glacial maximum in northern South America and the Caribbean: a review. Interciencia 13:128-137. Smith, F. 1858. Catalogue of Hymenopterous insects in the British Museum. VI. Formicidae. London. 216 p. Weber, N. 1938. New ants from stomachs of Bufo marinus L. and Typhlops reticidatus (L). Annals of the Entomological Society of America 31:207-210. Weber, N. 1938. 1940. Rare ponerine genera in Pan- ama and British Guiana. Psyche 47(2-3):75-84. Wheeler, G.C. and J. Wheeler. 1952. The ant larvae of the subfamily Ponerinae — Part L American Midland Naturalist 48(1):111-144. Wheeler, G.C. and J. Wheeler. 1971. Ant larvae of the subfamily Ponerinae: Second supplement. Annals of the Entomological Society of America 64(6):1197-1217. Wheeler, W.M. 1915. Some additions to the North American ant fauna. Bulletiti of the American Mu- seum of Natural History 34:389^21. Wheeler, W.M. 1922. The ants of Trinidad. y4»;t'n'a7/; Museum Novitates 45:1-16. Wheeler, W.M. 1923. Wissenschaftliche Ergebnisse der schwedischen entomologische Reise des Herrn Dr. A. Roman in Amazonas 1914-1915. Arkiv for Zoologie 15(7):l-6. Wheeler, W.M. 1930. A new Emeryella from Panama. Proceedings of the New England Zoological Club 12: 9-13. Wheeler, W.M. 1936. Ants from Hispaniola and Mona Island. Bulletin of the Muscu})i of Compara- tive Zoology 80:1 93-21 L Wheeler, W.M.' & W.M. Mann. 1914. The ants of Hai- ti. Bulletin of the American Museum of Natural His- tory 33:1-61. J. HYM. RES. Vol. 4, 1995, pp. 194-203 Latitudinal Gradients in North American Braconid Wasp Species Richness and Biology Donald L. J. Quicke and Rachel A. Kruft Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, U.K. Abstract. — Latitudinal trends in the numbers of species and genera of all subfamilies of Bracon- idae in America north of Mexico were examined by compiling lists of published records for states (USA) and provinces (Canada) and grouping these into approximate latitudinal belts. The data obtained agree well with past studies on the species richness of Ichneumonidae in North America (Janzen 1981) with peaks in numbers of taxa occurring in the belt lying approximately between 37° and 42° North. The ratio of idiobiont to koinobiont genera and species increased monotonically with decreasing latitude. Most insect groups increase in species using the catalogues of Townes. These cat- richness with decreasing latitude (see Ste- alogues were based on an enormous phens 1989). However, there is a body of amount of work and were therefore taken evidence that the species richness and di- to be relatively geographically unbiased, versity of some groups of insect parasit- though the possibility does exist that since oids, whilst increasing towards the trop- Townes himself did much of his collecting ics, does so at a lesser rate than does the around Michigan, the taxonomic coverage species richness and diversity of herbivo- oi that part of the United States may have rous insects (Owen & Owen 1974; Gauld been relatively more complete. However, 1986; Noyes 1989). Whilst there is good Janzen's approach does overcome some of evidence that the species richness and per- the difficulties encountered when working haps also diversity of some groups of par- ^^ tropical versus temperate comparisons asitic Hymenoptera, such as the Chalci- ^^ ^^at sampling and taxonomic effort doidea, increases towards the tropics (e.g. ^^*^^^" *^^ temperate region has been Hespenheide 1979), for others it has been comparatively evenly distributed and suggested that species richness may even "^°^^ thorough because of the long history , ... .. 1.U • .. • 1 of collectme and revisionary work that be greater m temperate than m tropical , , ^ . , , ^^ . . , . ,. J r J .^ 1 has been carried out there, burprismely, regions, a trend referred to as anomalous , r , i i i • x t / Janzen found that the peak in North diversity (Owen & Owen 1974; Janzen & Pond 1975; Janzen et al. 1976). However, American ichneumonid species richness , , . 1 , , ,, , occurs between 38° and 42° North, i.e. spe- these conclusions have been challenged on • • , u c ^ u ■ ^ cies richness was not found to mcrease the grounds of sampling methodology consistently with decreasing latitude. The (Morrison et al. 1978; Hespenheide 1979), question remained, however, whether this and also because they mostly deal with ^^^^^^ ^^^ ^^^^ generally applicable to just one family, the Ichneumonidae, and ^t^er groups of parasitic Hymenoptera. the generality of these findings has yet to indeed there is a growing body of evi- be demonstrated. dence to suggest that different groups of Janzen (1981) examined the latitudinal parasitoids show quite different trends in distributions of members of eight subfam- their species richness with respect to lati- ilies of Ichneumonidae in North America tude (Gauld 1986; Askew 1990). Volume 4, 1995 195 The present work attempts to carry out for the Braconidae, a similar exercise to Janzen's (1981) ichneumonid study. The Braconidae are generally accepted as be- ing the extant sister group of the Ichneu- monidae (Sharkey & Wahl 1992), and col- lectively, the two families constitute the superfamily Ichneumonoidea. Juillet (1964) obtained data from samples of ich- neumonoids caught in a rotary trap at a single site in Canada, which suggested that members of the Ichneumonidae 'pre- fer', and hence are more diverse in, cooler and more humid habitats than members of the Braconidae. Thus, it might be ex- pected that braconids will, as a whole, be more species rich at lower latitudes than ichneumonids, and therefore might not show the same peak in species richness that Janzen found for Nearctic Ichneu- monidae. Thus, as pointed out by Whar- ton (1993a), the distribution of species richness in the Braconidae ought to make an interesting comparison with the Ich- neumonidae. Catalogues and descriptions of Nearctic Braconidae usually provide data on the state or province from which material has been recorded or described. We therefore divided the states of the USA and the Ca- nadian provinces collectively into 5 groups such that each group occupied a reasonably constant latitudinal zone with minimal latitudinal overlap between groups (Fig. 1). The areas of the five zones are given in Table 2. Whilst it was not pos- sible to completely avoid overlap, the ef- fect of such overlaps will be to reduce rather than enhance the likelihood of de- tecting latitudinal gradients, and conse- quently our results will be conservative. The resolution of latitudinal trends in the present study is necessarily somewhat coarser than that obtained by Janzen (1981) as information on distributions is only readily available in the form of state or province records. The primary sources of distributional data were the catalogues of Shenefelt (1969, 1970a, b, 1972, 1973a, b, 1974 1975, 1978), Shenefelt & Marsh (1976) and Marsh (1979). We then attempted to take into account many c^f the subsequently published taxonomic changes so as to al- low for new synonymies, new generic placements, newly described taxa and new distribution records (van Achterberg 1977, 1979, 1983, 1985, 1986, 1987, 1988a, b; Capek & Achterberg 1992; Deyrup 1981; Haeselbarth & Loan 1985; Huddleston 1976; Johnson 1987; Loan 1979; Loan & HoUiday 1979; Marsh 1988, 1989, 1991, 1993; Mason 1975, 1976a, b, 1978, 1979, 1981, 1987, 1991; Quicke & Kruft 1995; Quicke & Sharkey 1985; Riegel 1987; Ries- ke et al. 1989; Saffer 1982; Sharkey 1985, 1988; Sharkey & Mason 1986; Sharkey & Wharton 1985, 1994; Shaw 1983, 1985, 1992, 1993; Stary & Marsh 1982; Wharton 1977a, b, 1980, 1983, 1986, 1988, 1993b, 1994; Wharton & Quicke 1988; Wharton et al. 1989; Wheeler & Loan 1984; Whitfield 1985, 1988a, b; Whitfield & van Achter- berg 1987; Whitfield & Mason 1994; Wil- liams 1985, 1988). In addition, a few un- published data were included for the fol- lowing taxa: Vipio (Braconinae) (Inayatul- lah 1992); Pholetesor (Microgastrinae) (J. Whitfield, in press); undescribed genus of Euphorinae (S. R. Shaw, pers. comm.). The genus Celerion is excluded as its subfamil- ial status is questionable (Marsh, pers. comm.). No data are included for intro- duced species. The treatment of subfamilies largely fol- lows that of Quicke & Achterberg (1990) except that the Hormiini and Meteorini are treated here as separate subfamilies. No pretence is made that this taxonomic treatment is complete, but it is hoped that it is unbiased and, therefore, that any er- rors that may have been included and any omissions there may be will not affect the results substantially. RESULTS AND DISCUSSION The numbers of genera and species of each subfamily recorded as occurring in 196 Journal of Hymenoptera Research Fig. 1 . Map of North America showing boudaries of latihidinal zones employed in the present study. each of the five latitudinal zones are pre- sented in Tables 1 and 2. The subfamilies are arranged in two groups, the first being idiobionts (Table 1) and the second koino- bionts (Table 2). As expected, different subfamilies showed markedly different latitudinal trends, though nearly all showed maximum generic and specific representation in our latitudinal zone IV which corresponds approximately to the latitudinal range from 37° and 42° North. None of the larger subfamilies was found to be most speciose in zone V. Four of these, the Braconinae, Microgastrinae, Ma- crocentrinae and Opiinae were most di- verse in zone III. Our data for the Alysi- inae, usually considered a more temperate and northern group, suggest that within Volume 4, 1995 197 Table 1. Distributions of genera and species of braconid subfamilies with respect to latitudinal zones (see Fig. 1). Idiobiont taxa Sublamilies Histcro- Rhyssalinao iTiLTinao Doryctin.ie Braconinae l-ADthcLinaL' Hormimae Total Genera Zone I 0 0 2 3 2 1 8 Zone II 2 1 8 9 3 2 25 Zone III 1 1 16 8 4 4 34 Zone IV 2 1 20 7 3 6 39 Zone V 1 1 28 10 2 8 50 Species Zone I 0 0 2 4 3 1 10 Zone II 2 1 35 67 4 3 112 Zone III 5 1 72 108 5 5 196 Zone IV 4 1 111 104 3 12 235 Zone V 1 1 105 86 2 11 206 North America, it is at its richest in the middle latitudes of the U.S.A. (zone IV) as with most other subfamilies. Table 3 provides a comparison of the relative latitudinal distributions of braco- nid genera and species according to zone and in terms of taxa per unit area. Addi- tionally, it shows the total numbers of idiobiont and koinobiont taxa and the ra- tios of numbers of koinobiont to idiobiont taxa in each latitudinal zone. The latter show consistent trends for both genera and species, with the ratio of koinobionts to idiobionts decreasing with decreasing latitude. It has long been recognised that some subfamilies of Braconidae are most species rich in the temperate region whilst others are tropicocentric. For example, the Alysi- inae, which are entirely koinobiont endo- parasitoids of Diptera, are a principally northern group, and whilst there are many undescribed tropical members of the genus Asobam, a group that attacks Drosophili- dae, more detailed knowledge of these is not likely to change our concept of the sub- families overall distribution to a major de- gree. In contrast, the Agathidinae and Car- diochilinae, two groups of koinobiont en- doparasitoids of Lepidoptera larvae, are both most speciose in the tropical region (Sharkey 1992; Huddleston & Walker 1988), and the latter in particular is rather species poor in temperate latitudes. Some groups are restricted in latitudinal range by the distributions of their hosts. For exam- ple, members of the ichneumonid subfam- ily Ctenopelmatinae, most tryphonine ich- neumonids, and the braconid tribes Ich- neutini and Proteropini (both usually treat- ed as belonging to the Ichneutinae; see Sharkey & Wharton 1994) are all endopar- asitoids of sawflies, and since sawflies are most abundant in temperate regions they are likely to support a higher species rich- ness of parasitoids there. In the case of the Braconidae, life history strategy, idiobioncy and koinobioncy (Askew & Shaw 1986), is very tightly linked with whether the wasps develop as ecto- or endoparasites respectively (Quicke & Achterberg 1990; Shaw & Huddleston 1991). It is not possible therefore, to dis- cern which of these life history features is most responsible for the observed trends. Further, as biology is usually consistent within subfamilies within the Braconidae (Shaw & Huddleston 1991), any latitudi- nal trends in the ratios of idiobiont to ko- inobiont taxa will necessarily involve the differential occurrence or representation of subfamilies. 198 Journal of Hymenoptera Research Table 2. Distributions of genera and species of braconid subfamilies with respect to latitudinal zones (see Fig. 1). Koinobiont subfamilies Subfamilies ial of Tropical Ecology 9: 491-499. Gauld, I.D. 1986. Latitudinal gradients in ichneu- monid species-richness in Australia. Ecological Entomology 11: 155-161. Volume 4, 1995 201 Gauld, I. D. 1987. Some factors affecting the com- position of tropical ichneumonid faunas. Biolog- ical Journal of the Linnean Society 30: 299-312. Gauld, I. D., K. j. Gaston and D. Janzen. 1992. Plant allelochemicals, tritrophic interactions and the anomalous diversity of tropical parasitoids: the "nasty" host hypothesis. Oikos 65: 353-357. Haeselbarth, E. and C. Loan. 1983. Toumesilitiis, a new genus for a species group in Microctonus (Hymenoptera: Braconidae, Euphorinae). Contri- butions of the American Entomological Institute 20: 384-387.' Hawkins, B. A. 1994. Pattern and Process in Host-Par- asitoid Interactions. Cambridge University Press, Cambridge. 190pp. Hawkins, B.A. 1990. Global patterns of parasitoid as- semblage size. Journal of Animal Ecology 59: 57- 72. Hawkins, B.A., M.R. Shaw and R.R. Askew. 1992. Relations among assemblage size, host special- ization and climate variability in North Ameri- can parasitoid communities. American Naturalist 139: 58-79. Hespenheide, H.A. 1979. Are there fewer parasitoids in the tropics? American Naturalist 113: 766-769. Huddleston, T. 1976. A revision of Elasmosoma Ruthe (Hymenoptera: Braconidae) with two new spe- cies from Mongolia. Annales Historico-Naturales Musei Nationalis Hungarici 68: 215-225. Huddleston, T. and A. K. Walker. 1988. Carciiochiles (Hymenoptera: Braconinae), a parasitoid of lep- idopterous larvae, in the Sahel of Africa, with a review of the biology and host relationships of the genus. Bulletin of Entomological Research 78: 435-461. Inayatullah, M. 1992. A systematic study of the ge- nus Vipio Latreille (Hymenoptera: Braconidae) of the Nearctic and Neotropical Regions. Unpub- lished Ph.D. thesis. University of Wyoming, Lar- amie. Janzen, D.H. 1981. The peak in North American ich- neumonid species richness lies between 38° and 42°N. Ecology 62: 532-537. Janzen, D. H., I. M. Ataroff, M. Farinas, S. Reyes, N. Rincon, A. Soler, P. Soriano and M. Vera. 1976. Changes in the arthropod community along an elevational transect in the Venezuelan Andes. Biotropica 8:193-203. Janzen, D.H. and CM. Pond. 1975. A comparison, by sweep sampling, of the arthropod fauna of secondary vegetation in Michigan, England and Costa Rica. Transactions of the Royal Entomological Society of London 127: 33-50. Johnson, J. W. 1987. A revision of the species of Praon Haliday in North America north of Mexico (Hymenoptera: Aphidiidae). Canadian Entomolo- gist 119: 999-1025. Juillet, J. A. 1964. Influence of weather on flight ac- tivity of parasitic Hymenoptera. Canadian Journal of Zoology 42: 1133-1141. Loan, C.C. 1979. Three new species of Peristenus Foerster from Canada and Western Europe (Hy- menoptera: Braconidae: Euphorinae). Naturaliste Canadiene 106: 387-391. Loan, C.C. and N.J. Holliday. 1979. Euphorinae par- asitic on ground beetles with descriptions of Mi- croctonus Wesmael (Hymenoptera: Braconidae, and Coleoptera: Carabidae). Naturaliste Canadiene 106: 393-397. Marsh, P. M. 1979. Family Braconidae. Pages 144- 293. in K. V. Krombein et al. eds Catalogue of the Hymowfjtera in America Nortli of Mexico. Smith- sonian Institute Press, Washington D. C. Marsh, P.M. 1988. Revision of the tribe Odontobra- conini in the western hemisphere (Hymenoptera: Braconidae: Doryctinae). Systematic Entomology 13: 443-464. Marsh, P. M. 1989. Notes on Braconidae (Hymenop- tera) associated with Jojoba (Simmondsia clnnen- sis) and descriptions of new species. Pan-Pacific Entomologist 65: 58-67. Marsh, P. M. 1991. A new species of Pauesia (Hy- menoptera: Braconidae, Aphidiinae) from Geor- gia and introduced into South Africa against the black pine aphid (Homoptera: Aphididae). ]our- nal of Entomological Science 26:81-84. Marsh, P. M. 1993. Descriptions of new western hemisphere genera of the subfamily Doryctinae (Hymenoptera: Braconidae). Contributions of the Amercan Entomological bistitute 28: 1-58. Mason, W.R.M. 1975. A new Nearctic y4pfl«fe/es (Hy- menoptera: Braconidae) from Oregon grape (Ber- beridaceae). Canadian Entomologist 107: 1133- 1135. Mason, W.R.M. 1976a. A revision of Dyscoletes Hal- iday (Hymenoptera: Braconidae). Canadian Ento- mologist 108: 855-858. Mason, W.R.M. 1976b. The identity of Macrocentrus uniformis Provancher (nee Cresson), description of a sibling species and a possible grooming or- gan (Hym.: Braconidae). Naturaliste Canadiene 103: 513-515. Mason, W.R.M. 1978. A synopsis of the Nearctic Bra- conini, with revisions of Nearctic species of Coe- loides and Myosoma (Hymenoptera: Braconidae). Canadian Eiitomologist 110: 721-768. Mason, W. R. M. 1979. A new Rogas (Hymenoptera: Braconinae) parasite of tent caterpillars (Malaco- soma spp. Lepidoptera: Lasiocampidae) in Can- ada. Ca)iadian Entomologist 111: 783-786. Mason, W.R.M. 1981. The polyphyletic nature of Apanteles Foerster (Hymenoptera: Braconidae): A phylogeny and reclassification of Microgasteri- nae. Memoirs of the Entomological Society of Canada 115: 1-147. Mason, W.R.M. 1987. Vadum, a new genus of Nearc- 202 Journal of Hymenoptera Research tic Braconidae (Hymenoptera). Proceedings of the Entomological Society of Washington 89: 325-328. Mason, W.R.M. 1991. A new genus of exodont Ich- neutinae (Hymenoptera: Braconidae). Insecta Mundi 5: 95-99. Morrison, G., M. Auerbach and E.D. McCoy. 1978. Anomalous diversity of tropical parasitoids: A general phenomenon? American Naturalist 114: 303-307. Noyes, J.S. 1989. The diversity of Hymenoptera in the tropics with special reference to Parasitica in Sulawesi. Ecological Entomology 14: 197-207. Owen, D.F. and J. Owen. 1974. Species diversity in temporate and tropical Ichneumonidae. Nature 249: 583-584. Quicke, D.L.J, and Achterberg, C. van. 1990. Phylog- eny of the subfamilies of the family Braconidae (Hymenoptera: Ichneumonoidea). Zoologische Verhandelingen 258: 1-95. Quicke, D. L. J. and R. A. Kruft. 1995. Species of Yelicones (Hymenoptera: Braconidae: Rogadinae) in North America with descriptions of two new species. Annals of the Entomological Society of America 88: 129-138. Quicke, D.L.J, and M.J. Sharkey. 1989. A key to and notes on the genera of Braconinae (Hymenop- tera: Braconidae) from America north of Mexico with descriptions of two new genera and three new species. Canadian Entomologist 121: 337-361. Rathcke, B.J. and P.W. Price. 1976. Anomalous di- versity of tropical ichneumonid parasitoids: a predation hypothesis. American Naturalist 110: 889-893. Riegel, 1982. The American species of Dacnusinae, excluding certain Dacnusini (Hymenoptera: Bra- coninae). Novitates ArtJiropodae 1(13): 1-185. Rieske, L. K., D. W. A. Hunt and K. F. Raffa. 1989. Microctonus pachylobii (Hymenoptera: Braconi- dae) parasitizes Hylobius weevils in Wisconsin: New host genus and geographic records. Ento- mological News 100: 153-154. Saffer, B. 1982. A systematic revision of the genus Cenocoelius (Hymenoptera: Braconidae) in North America including Mexico. Polskie Pismo Ento- molgiczne — Bulletin Entomologique de Pologne 52: 73-167. Sharkey, M. J. 1985. Notes on the genera Bassus Fa- bricius and Agathis Latreille, with a description of Bassus arthurellus n. sp. (Hymenoptera: Bra- coninae). Canadian Entomologist 117: 1497-1502. Sharkey, M. J. 1988. A taxonomic revision of Alaba- grus (Hymenoptera: Braconinae). Bulletin of the British Museum (Natural History) Entomology, 57: 311-437. Sharkey, M. J. 1992. Cladistics and tribal classifica- tion of the Agathidinae (Hymenoptera: Braconi- nae). Journal of Natural History 26: 425-447. Sharkey, M. J. and W. R. M. Mason. 1986. The ge- neric validity of Aenigmostomus and Asiacardi- ochiles (Hymenoptera: Braconidae). Proceedings of the Entomological Society of Washington 88: 300- 302. Sharkey, M. J. and D. Wahl. 1992. Cladistics of the Ichneumonoidea (Hymenoptera). Journal of Hy- menoptera Research 1: 15-24. Sharkey, M. J. and R. A. Wharton. 1985. Redefinition of Megagathis Kreichbaumer, and reassignment of New World species to Zacremnops new genus (Hymenoptera: Braconinae: Agathidinae). Cafia- dian Entomologist 177: 599-603. Sharkey, M. J. and R. A. Wharton. 1994. A revision of the genera of the world Ichneutinae (Hyme- noptera: Braconidae). Journal of Natural Histon/ 28: 873-912. Shaw, M. R. 1994. Parasitoid host ranges, in B. A. Hawkins and W. Sheehan eds Parasitoid Com- munity Ecology. Oxford University Press, pp. 111- 144. Shaw, M. R. and T. Huddleston. 1991. Classification and biology of braconid wasps. Handbooks for the Identification of British Insects 7: 1-126. Shaw, S.R. 1983. A taxonomic study of Nearctic As- cogaster and a description of a new genus Lepto- drepana (Hymenoptera: Braconidae). Entomogra- phy 2: 1-54. Shaw, S.R. 1985. A phylogenetic study of the sub- families Meteorinae and Euphorinae (Hymenop- tera: Braconidae). Entomography 3: 277-370. Shaw, S.R. 1992. Seven new North American species of Neoneunis (Hymenoptera: Braconidae). Pro- ceedings of the Entomological Society of Washington 94: 26-47. Shaw, S. R. 1993. Systematic status of Eucystomastax Brues and characterization of the Neotropical species (Hymenoptera: Braconidae: Rogadinae). Journal of Hymenoptera Research 2: 1-11. Shenefelt, R. D. 1969. Hymenopterorum Catalogus (nova editio). Pars 4, Braconidae 1. Hybrizoni- nae, Euphorinae, Cosmophorinae, Neoneurinae, Macrocentrinae. The Hague, Junk. 176 pp. Shenefelt, R. D. 1970a. Hymenopterorum Catalogus (nova editio). Pars 5, Braconidae 2. Helconinae, Calyptinae, Mimagathidinae, Triaspinae. The Hague, Junk. 130 pp. Shenefelt, R. D. 1970b. Hymenopterorum Catalogus (nova editio). Pars 6, Braconidae 3. Agathidinae. The Hague, Junk. 124 pp. Shenefelt, R. D. 1972. Hymenopterorum Catalogus (nova editio). Pars 7, Braconidae 4. Microgaster- inae, Apanteles. The Hague, Junk.243 pp. Shenefelt, R. D. 1973a. Hymenopterorum Catalogus (nova editio). Pars 9, Braconidae 5. Microgaster- inae & Ichneutinae. The Hague, Junk. 144 pp. Shenefelt, R. D. 1973b. Hymenopterorum Catalogus (nova editio). Pars 10, Braconidae 6. Cheloninae. The Hague, Junk. 124 pp. I Volume 4, 1995 203 Shenefelt, R. D. 1974. Hi/menoptewriim Catalogus (nova editio). Pars 11, Braconidae 7. Alysiinae. The Hague, Junk. 180 pp. Shenefelt, R. D. 1975. Hymoioptcronun Catalogus (nova editio). Pars 12, Braconidae 8. Exothecinae, Rogadinae. The Hague, Junk. 150 pp. Shenefelt, R. D. 1978. Hi/nwnoptcronini Catalogus (nova editio). Part 15. Braconidae, vol. 10. Bra- coninae, Gnathobraconinae, Mesostoinae, Pseu- dodicrogeniinae, Telengainae, Ypsistocerinae plus Braconidae in general, major groups, un- placed genera and species. The Hague, Junk, 440 pp. Shenefelt, R. D. and P. M. Marsh. 1976. Hymenopter- orum Catalogus (nova editio). Pars 13, Braconidae 9. Doryctinae. The Hague, Junk. 162 pp. Stary, P. and P. M. Marsh. 1982. A new species of Trioxi/s (Nymenoptera: Aphidiidae) parasitic on a pecan aphid. Proceedings of the Entomological So- ciety of Washington 84: 726-728. Stephens, G. S. 1989. The latitudinal gradient in geo- graphic range: How so many species coexist in the tropics. American Naturalist 133: 240-256. Townes, H. 1971. Ichneumonids as biological control agents. Proceedings of the Tall Timbers Conference oil Ecological Animal Control by Habitat Manage- ment, pp. 235-248. Wharton, R. A. 1977a. New World Aplmereta species (Hymenoptera: Braconidae: Alysiinae), with a discussion of terminology used in the tribe Aly- siini. Annals of the Entomological Society of America 70: 782-803. Wharton, R. A. 1977b. Exodontiellini, a new tribe of Opiinae with exodont mandibles (Hymenoptera: Braconidae). Pan-Pacific Entomologist 53: 297-303. Wharton, R.A. 1980. Review of the Nearctic Alysiini (Hymenoptera: Braconidae) with discussion of the generic relationships within the tribe. Ento- mology 88: 1-112. Wharton, R.A. 1983. New species of lllidops and Bra- con (Hymenoptera: Braconidae) of potential use in biological control. Canadian Entomologist 115: 667-672. Wharton, R.A. 1986. The braconid genus Alysia (Hy- menoptera): a description of the subgenera and a revision of the subgenus Alysia. Systematic En- tomology 11: 453-504. Wharton, R.A. 1988. The braconid genus Alysia (Hy- menoptera): A revision of the subgenus Anarcha. Contributions of the American E)itomological Insti- tute 25: 1-69. Wharton, R. A. 1993a. Bionomics of the Braconidae. Annual Revieio of Entomology 38: 121-143. Wharton, R. A. 1993b. Review of the Hormiini (Hy- menoptera: Braconidae) with a description of new taxa. Journal of Natural History 27: 107-171. Wharton, R. A. 1994. New genera, species and re- cords of New World Alysiinae (Hymenoptera: Braconidae). Proceedings of the Entomological So- ciety of Washington 96: 630-664. Wharton, R.A., and D.L.J. Quicke. 1988. A new spe- cies of Bracon (Hymenoptera: Braconidae) para- sitic on Eoreuma loftini (Dyar) (Lepidoptera: Py- ralidae). Proceedings of the Entomological Society of Washington 90: 288-293. Wharton, R.A., J. W. Jr. Smith, D.L.J. Quicke and H. W. Browning. 1989. Two new species of Digon- ogastra Viereck (Hymenoptera: Braconidae) par- asitic on Neotropical pyralid borers (Lepidop- tera) in maize, sorghum and sugarcane. Bulletin of Entomological Research 79: 401-410. Wheeler, A. G.'jr, and C. C. Loan, 1984. Peristenus henryi (Hymenoptera: Braconidae, Euphorinae) a new species parasitic on the honeylocust plant bug, Diaphnocoris chlorionis (Hemiptera: Miridae). Proceedings of the Entomological Society of Washing- ton 86: 669-672. Whitfield, J.B. 1985. The Nearctic species of Deuterix- ys Mason (Hymenoptera: Braconidae). Pan-Pacific Entomologist 61: 60-67. Whitfield, J.B. 1988a. Revision of the Neactic species of the genus Stiropius Cameron { = Bucculatriplex Auct.) with the description of a new related ge- nus (Hymenoptera: Braconidae). Systematic En- tomology 13: 373-385. Whitfield, J.B. 1988b. Two new species of Paradelius (Hymenoptera: Braconidae) from North America with biological notes. Pan-Pacific Entomologist 64: 313-319. Whitfield, J.B. and C. van Achterberg. 1987. Clarifi- cation of the taxonomic status of the genera Can- tJiaroctonus Viereck, Noserus Foerster and Pseu- davga Tobias (Hymenoptera: Braconidae). Sys- tematic Entomology 12: 509-518. Whitfield, J. B. and W. R. M. Mason. 1994. Mende- sellinae, a new subfamily of braconid wasps (Hy- menoptera, Braconidae) with a review of rela- tionships within the microgastroid assemblage. Systematic Etitomology 19: 61-76. Williams, D. J. M. 1985.' The New World genus Lath- rapanteles n. gen.: Phylogeny and placement in Microgastrinae (Hymenoptera: Braconinae: Co- tesiini). Canadian Journal of Zoology 63: 1962-1981. Williams, D. J. M. 1988. Classification, phylogeny and zoogeograhic studies of species of Sathon Mason (Hymenoptera: Braconidae). Quaestiones Entomologicae 24: 529-639. Williamson, M. 1988. Relationship of species number to area, distance and other variables. Pages 91- 115. in A. A. Myers and P.S. Giller eds Analytical Biogeography. Chapman and Hall, London. J. HYM. RES. Vol. 4, 1995, pp. 204-284 Revision of the World Genera of the Subtribe Stigmina (Hymenoptera: Apoidea: Sphecidae: Pemphredoninae), Part 1 Albert T. Finnamore Curator, Invertebrate Zoology, Provincial Museum of Alberta, 12845 — 102 Avenue, Edmonton, Alberta CANADA T5N 0M6 Abstract. — The following new genera are described: Araucastigmus, Aykhustigmus, Incastigmus, Llaqhastigmus, and Tziistigmus. The following new species are described: Araucastigmus masneri, yanillus; Aykhustigmus fritzi, patanaivi, phasti, luarawa; Incastigmus inti; Llaqhastigmus ambiguus, aus- tralis, chutiyana, colomhianus, ecuatorialis, jatunkirus, llutanis, mantanti, muthus, nigricollaris, santan- deranus, sapanis, shachus, sharkeyi; Parastigmus huecuvus, nina; Tzustigmus khmer, syam, veda, and wuming. New synonymy: Stigmus smithii Ashmead is placed under Incastigmus tlioracicus (Ash- mead). A key to world genera is presented, and history, biology, generic relationships, and bio- geography are discussed. Species of Araucastigmus, Aykhustigmus, Llaqhastigmus and Parastigmus are keyed and described, and their distributions are presented. INTRODUCTION This contribution is intended as the first in a series of papers dealing with the world species of the Stigmina. Members of the Stigmina are small, generally 3 to 5mm in length, and black or red and black in coloration. The group contains an estimat- ed 150 species distributed in all areas ex- cept the polar regions but for the most part is found throughout the southern hemisphere tropics. This first paper de- scribes five new genera, Tzustigmus, Ar- aucastigmus, Aykhustigmus, Incastigmus and Llaqhastigmus, treats the species of the smaller genera, reviews biology, taxonom- ic history, and presents an analysis of ge- neric phylogeny before considering the biogeography of the Stigmina. Future pa- pers will deal with the species of Incastig- mus, Stigmus, and Carinostigmus. Morphological Terms And Symbols. — Ter- minology generally follows Bohart and Menke (1976) but in some cases terms need clarification and some new terms are introduced. They are listed below: Appressed setae: setae forming an angle close to 0° with the body surface. Carinate: a series of relatively large ridges (see microcarinate below) that do not impart a dull appearance to the body. Lateral sphere of propodeum: area of pro- podeum curving between propodeal en- closure and side. LCD: maximum diameter of lateral ocellus. Mesosoma: the thorax plus the propo- deum. Metasoma: the apparent abdomen con- sisting of the abdomen excluding the first segment or propodeum. Micropore field (Fig. 32): a grouping of very small pores usually visible only by scanning electron microscope, but by stereomicroscope apparent as a discrete microsculpture patch. Microcarinate: a series of parallel ridges so fine as to impart a dull appearance to the body. Microsculpture: minute sculpture impart- ing a dull appearance to the body. GOD: least distance between lateral ocel- lus and eye. Preomaular area: = preomaulal area of Bohart and Menke (1976). Sternum: metasomal sternum. Tergum: metasomal tergum. Volume 4, 1995 205 CARN CAS FSCA IIES Transscutellar sulcus: the anterior trans- BMNH verse sulcus of the scutellum, immedi- ately posterior to the transscutal articu- lation. BRD Transverse sulcus: on pronotal dorsum, the transverse sulcus immediately pos- terior to the transverse carina. *: holotype examined. Species Treatments. — Descriptions of all included species are provided. Descrip- tions of new species are based on all the material examined. In species demonstrat- ing variability descriptions are based on representatives of the most prevalent phe- notype with variation noted throughout the description. Collection data for the pri- mary type material are presented exactly as they appear on the label, thus several spellings for the same locality and collec- tor, and several formats for date of collec- tion will be encountered. Collection data for secondary type specimens are present- ed in a standardized format organized by country and state or province. In previ- ously described species the Material Ex- amined section lists only localities of col- lection. Sources Of Material. — Over 10,000 speci- mens from 47 institutions were amassed for this study. The following institutions have provided material used in this paper or loaned types. The abbreviation preced- ing the institution is that used in the text to designate type repositories. AEI American Entomological Insti- tute, 3005 sw 56th Ave. Gaines- ville, FLORIDA 32608 USA. V.K. Gupta, the late H.K. Townes, D. MCSN Wahl. ANIC Australian National Insect Col- lection, CSIRO, Division of En- tomology, GPO Box 1700 Can- MCZ berra, ACT 2601, AUSTRALIA. I.D. Naumann. BISH Bishop Museum, 1525 Bernice Street, P.O. Box 19000-A Hono- lulu, HAWAII 96819 USA. G.M. MPEG Nishida. LILLO MACN The Natural History Museum, Cromwell Road, London, SW7 5BD ENGLAND. C.R. Vardy. Biosystematics Research Divi- sion, CLBRR, Agriculture Cana- da, Research Branch, K.W. Neat- by Building, Ottawa, Ontario, CANADA KIA 0C6. L. Masner. Carnegie Museum, 4400 Forbes Avenue, Pittsburgh, PENNSYL- VANIA 15213 USA. J.E. Raw- lings. California Academy of Sciences, Golden Gate Park, San Francis- co, CALIFORNIA 94118 USA. W.J. Pulawski. Florida State Collection of Ar- thropods, Division of Plant In- dustry, Department of Agricul- ture, Gainesville, FLORIDA 32601 USA. L.A. Stange. Consejo Nacional de Investiga- ciones Cientificas y Tecnicas, In- stitute de Investigaciones Ento- mologicas Salta (INESALT), 9 de Julio 14-Casilla de Correo 3, 4405 Rosario de Lerma, Salta, AR- GENTINA. M.A. Fritz. Fundacion Miguel Lillo, Institu- te de Zoologia Miguel Lillo 251, 4000 San Miguel de Tucuman, ARGENTINA. A. Willink. Museo Argentine de Ciencias Naturales, Seccion Entomologia, Angel Gallardo 470, Casilla de Correo 220, Sucursal 5, 1405 Buenos Aires, ARGENTINA. A. Roig Alsina. Museo Civico Di Storia Naturale "Giacomo Doria" 1-16121 Gene- va, via Brigata Liguria, N. 9, IT- ALY. V. Raineri. Museum of Comparative Zoolo- gy, The Agassiz Museum, Har- vard University, Cambridge, MASSACHUSETTS 02138 USA. C. Vogt. Museu Paraense Emilio Goeldi, Av. Magalhaes Barata, 376 — C.P. 206 Journal of Hymenoptera Research 399, CEP 66000-Belem, Para, BRASIL. W.L. Overal. NMW Naturhistorisches Museum Wien, 2. Zoologische Abteilung (Insek- ten), A-1014 Wien, Burgring 7, oSTERREICH (Austria). M. Fi- scher. OSU Oregon State University, Depart- ment of Entomology, Corvallis, OREGON 97331 USA. G.R. Fer- guson. PMA Provincial Museum of Alberta, 12845-102 Avenue, Edmonton, Alberta, CANADA T5N 0M6. RNH Rijksmuseum van Natuurlijke Historie, Raamsteeg 2, Leiden, NETHERLANDS. K. van Ach- terberg. SEM Snow Entomological Museum, University of Kansas, Lawrence, KANSAS 66045 USA. R.W. Brooks. TMA Termeszettudomanyi Muzeum Allattara, H-1088 Budapest, Bar- oss u. 13 HUNGARY. J. Papp. UCV Instituto de Zoologia Agricola, Universidad Central de Vene- zuela, Maracay 2101-A, VENE- ZUELA. J. Luis Garcia. USNM Systematic Entomology Labora- tory, USDA % U.S. National Museum, NHB 168, Washing- ton, D.C 20560 U.S.A. A.S. Men- ke. ZIN Insect Systematics Laboratory, Zoological Institute, Russian Academy of Sciences, Sankt-Pe- terburg 199034 RUSSIAN FED- ERATION. ZMD Zoologisk Museum, Department of Entomology, Universitetspar- ken 15, DK 2100 Kobenhavn, DANMARK (Denmark). O. Lomholdt. ZMMU Zoological Museum of the Mos- cow State University, Herzen Street 6, Moscow 103009 RUS- SIAN FEDERATION. A.V. An- tropov. HISTORY G.W.F. Panzer (1804) described the ge- nus Stigmus for a single species, Stigmus pendiilus from Europe. Species description in the Stigmina accumulated more or less ad hoc until Kohl (1890) produced the first world synthesis of Stigmus and related genera. His revision of the Pempliredon ge- nus group provided a sound generic grouping that is still, a century later, the basis of modern classification. Over the period of 1890-1911, 24 species were de- scribed by such notable authorities as Kohl (1892, 1905), Cameron (1891), Fox (1892, 1897), Ashmead (1900), Mantero (1901), Rohwer (1911) and Strand (1911). During this period Turner (1907) de- scribed the genus Paracrabro from Austra- lia. By 1920 most of the descriptive work had shifted to fauna of the southern hemi- sphere culminating with Tsuneki's (1954) description of a new subgenus, Carinostig- miis, accompanying a revision of the Asian and European Stigmus. In subsequent pa- pers Tsuneki (1963, 1966, 1971) went on to describe numerous species and subspecies in the Oriental Region while Leclercq (1959) described species from the Afro- tropical Region. The Nearctic fauna was revised by Krombein (1973) at which time he described a new subgenus Atopostig- mus. Work on all groups of sphecoid wasps reached a high point when Bohart and Menke (1976) published a world ge- neric revision, a landmark synthesis un- precedented in the history of the group. That work delineated the subtribe Stig- mina to which the following 7 genera were assigned: Arpnctophilus, Paracrabro, Stigmus, Carinostigmus, Spilomena, Micro- stigmus, and Xysma. Subsequently Krom- bein (1984) described 4 new species of Car- inostigmus and a new subgenus Perissostig- mus from Sri Lanka. Budrys (1987) de- scribed new species from the former Soviet far east. Antropov (1992) described Volume 4, 1995 207 the genus Parastigmus from the Neotropi- pith. Cells are mass provisioned or possi- cal Region. bly progressively provisioned with 12-30 Among the most important publications aphids depending on prey size and spe- influencing the classification of the Stig- cies of wasp. Progressive provisioning cre- mina was Menke's (1989) paper on new ates a subsocial situation where mothers species of Arpnctophilus from New Guinea, and daughters are present in the same The author indicated that the sub tribal nest. Iwata (1964) observed progressive limits proposed by Bohart and Menke in provisioning in nests of Cnriiiostignius ium- the 1976 world revision should be recon- tai (Tsuneki) and found larvae and adults sidered. He established a new subtribe, of different ages in nests of C. monstrosus Spilomenina, containing ArpactopJiilus, (Tsuneki). The only reported cleptoparas- Spilornena, Microstigmus and Xi/sma. Fur- ites are chrysidid wasps of the genus Otn- thermore Menke was able to unite these alus. genera using a newly discovered apotypic Unpublished or anecdotal records indi- character state, that of reduced papal seg- cate some species of Stigmina are sand mentation. Menke's work first of all re- nesting, a presumed plesiotypic character duced the Stigmina to 3 genera: Pnracrabro state. Krombein (1973, 1984) in his revi- (1 Australian species), Stigmus (30 species, sions of Nearctic Stigmiis and Sri Lankan Holarctic, Neotropical and Oriental), and Cnrinostignnis, lent some credibility to Carinostigmiis (25 species, Afrotropical and these observations when he noted nesting Oriental). Secondly, no synapotypic char- differences in females with 2 or 3 mandib- acter states supporting the 3 genera re- ular teeth. Females with 3 mandibular maining in the Stigmina were known, teeth excavated nests in pithy twigs or which meant that the Stigmina could be a stems while those with 2 teeth used pre- plesiotypic assemblage representing an ar- existing cavities in the ground. The impli- tificial rather than a natural grouping, cation is that those species constructing This paper provides ample evidence of the nests evolve more complex mandibular monophyly of the Stigmina and indicates dentition presumably to perform more that 9 genera and about 150 species are specialized tasks. Microstigmus (Spilomen- included. ina) have a bidentate mandible that is not used in construction of a complex nest suspended beneath a leaf and housing Published records (Arnold 1924; Eick- multiple generations. The foregoing indi- wort 1967; Iwata 1964; Janvier 1962; cates that mandibular dentition likely Krombein 1956, 1958a, 1958b, 1961, 1963, bears a straightforward relationship to 1984; Rau 1928; Richardson 1915; Smith nest construction when mandibles are em- 1923; Tsuneki 1970; Wasbauer and Si- ployed for that purpose, monds 1964; Yasumatsu and Watanabe More reliable evidence of sand nesting 1964) indicate that Stigmus and Carinostig- can be inferred from the presence of a fore mus nest in twigs or pre-existing cavities tarsal rake and a broad pygidial plate in and prey on aphids (Homoptera: Aphi- the female or a remnant of that plate in doidea). Twig nests are usually construct- the male. The fore tarsal rake is a series of ed by excavating pithy twigs like Ruhrus, elongate setae which are used to move Sambucus, or Sassafras among others. Pre- sand. Members of Stigmina lack a fore tar- existing cavities include straws of thatch, sal rake but have spatulate setae on the holes or borings in timber and galls. Nests rake area, possibly a secondary adaptation vary in length from 8cm to half a metre, to manipulation c^f nest material. The py- the cells are constructed in linear series gidial plate is a flat specialized area de- separated by masticated wood or plugs of fined by a carina or groove on the 6th BIOLOGY 208 Journal of Hymenoptera Research metasomal tergum in the female and the 7th in the male. In sand nesting species this plate occupies a much greater pro- portion of the tergum than in twig or cav- ity nesting species. Almost all Stigmina have a narrow pygidial plate occupying a small fraction of the tergum indicating twig or cavity nesting behaviour. A few species have an intermediate pygidial plate and several Argentine species have a large pygidial plate with a remnant of this plate in the male. Manfredo Fritz (per- sonal communication) has observed sand nesting Stigmus in Argentina, he was un- able to capture any specimens. Specimens from his collection, used in this study, contained females with relatively large pygidial plates and with remnants in sev- eral males, possibly representing the only true sand nesting Stigmina. In summary, biological observations, al- though detailed in several instances, are available for few species only. More ob- servations are necessary to document nest construction and the existence of social be- havior in the group. The Stigmina appear to be twig or cavity nesting and restrict prey to aphids. There is some anecdotal evidence of sand nesting behaviour in southern Neotropical species. A number of new genera have no behavioral data whatever. Observations on these genera are often impeded by the small size of the individuals but would offer a fertile field of study to a dedicated biologist. SUBTRIBE STIGMINA Diagnosis. — The Stigmina, as here de- fined, includes 9 genera with a microsetal or micropore fields laterally on metaso- mal sternum II (Figs. 44, 45), more or less centrally on the fore wing stigma (Figs. 46, 47) and on the vertex between the lat- eral ocellus and the compound eye (Figs. 3, 8, 20, 26, 32). The latter character is not fully developed in several genera but consists of several pits in a small depres- sion. The stigmal pit field is diffuse in 11 species from Chile and Argentina and may not be readily apparent in these in- stances. These characters are found in all members of the Stigmina and nowhere else in the Pemphredoninae. They are thus autapotypies that unambiguously delineate the Stigmina. Described genera include Carinostigmus Tsuneki, Paracrabro Turner, Parastigmus Antropov, and Stig- mus Panzer. In addition to Paracrabro and Parastigmus, five new genera are pro- posed and 33 of the 58 species are treated in the present paper: Tzustigmus (5 spe- cies. Oriental, far eastern Palearctic), Par- astigmus (4 species, southern Neotropi- cal), Araucastigmus (3 species, Chile), Par- acrabro (1 species, Australian), Aykhustig- mus (4 species. Neotropical), Incastigmus (25 species. Neotropical), and Llaqhastig- mus (14 species. Neotropical). Description. — Head. Palpal formula 6-4; mandibular socket closed; antennal sock- ets placed low on frons near or touching clypeus (Fig. 1); frons unmodified by grooves, except occasionally along inner eye margin; vertex with a micropore field or at least a few associated pits between lateral ocellus and eye (Figs. 3, 8, 20, 26, 32); genal setae usually normal, not elon- gate setae scattered among short setae (ex- ceptions among some males of Stigmus and Carinostigmus). Mesosoma. Pronotum with transverse carina present, at least laterally (Fig. 2); omaulus present; episternal sulcus unde- veloped; hypersternaulus present, obli- quely oriented across mesopleuron (Fig. 4); fore tarsus of female without rake; mid coxa with oblique dorsal carina; inner hind coxal carina absent; hind coxa with- out elongate bristle; metasternum with shallow posterior emargination; propo- deal enclosure present (Fig. 10); stigma of fore wing enlarged, lenticular with dis- crete micropore field (Figs. 46, 47); mar- ginal cell normal size, subequal to or larg- er than stigma, closed and bordering an- terior wing margin; 2 submarginal cells; submarginal cell II quadrate; submarginal cell I out of line with respect to marginal Volume 4, 1995 209 Figs. 1-5. Carinostigmiis sp. 9. 1, Head, frontal view. 2, head and prothorax, dorsal view. 3, micropore field between lateral ocellus and compound eye. 4, mesosoma, lateral view. 5, mesosoma, dorsal view. and discoidal cells; submarginal cell I without spur, closed; recurrent vein re- ceived by submarginal cell I; 2 discoidal cells; hind wing cells closed. Metasoma. Petiole present, longer than wide, setae short, scattered over surface; tergum I without lateral carina; male ster- na without elongate specializeci setae; sterna II and often III with lateral micro- setal patches (Figs. 44, 45); sternum VI without apical pegs; pygidial plate present in female, almost always absent in male; sternum VIII of male elongate, volsella di- vided into digitus and cuspis. 210 Journal of Hymenoptera Research KEY TO GENERA OF STIGMINA 1. Hind wing submedian cell reduced, cu-a positioned about halfway from wing base to origin of media (Fig. 50); Old World, except Australian Region .... Carinostigmus Tsuneki 1'. Hind wing submedian cell normal size, cu-a positioned next to origin of media (Fig. 49) 2 2. Hind wing media diverging beyond cu-a (Fig. 49) or mesopleuron coarsely areolate to sternopleural region (Figs. 22, 28) 3 2'. Hind wing media diverging before cu-a (Fig. 48); mesopleuron not entirely coarsely sculp- tured (Fig. 34) 5 3. Mesopleuron coarsely sculptured to sternopleural region (Figs. 22, 28); male with prom- inent genal carina separated from mid outer orbit by a distance subequal to basal width of mandible (Fig. 21); Neotropical Aykhustigmus Finnamore, new genus 3'. Mesopleuron mostly unsculptured, shiny; genal carina, if present, close to eye margin (Fig. 9) 4 4. Interantennal tubercle absent (Fig. 40); eyes not margined by a carina: petiole carinate; Japan Stigtnus Panzer, in part 4'. Interantennal tubercle present (Fig. 7); eyes margined by a carina (Fig. 9); petiole smooth, round, at most microcarinate; Oriental, far eastern Palearctic Tztistigmus Finnamore, new genus 5. Acetabular carina absent (Fig. 11); southern Neotropical 6 5'. Acetabular carina present (Fig. 43); widespread 7 6. Pygidial plate small, narrow, present in female only; mandible in male bidentate and in female tridentate at apex; body shiny, without microsculpture (Figs. 14, 15); Chile .... Araucastigmus Finnamore, new genus 6'. Pygidial plate broad, present in both sexes, evanescent in male; mandible of both sexes bidentate at apex; microsculpture present on head and often most of body; Argentina, Chile Parastigmus Antropov 7. Vertex micropore field absent; Australia Paracrabro Turner 7'. Vertex micropore field present (Figs. 20, 26, 32) 8 8. Mandibles in male bidentate, female uni-, bi-, or usually tridentate; male clypeal apex usually with lateral bevel (Fig. 42); scutum without trace of median groove or postero- median pit but may be ridged (Fig. 41); apicoventral mandibular tooth in female, acute (Fig. 39); widespread, except Afrotropical and Australian Regions . . Stigtnus Panzer, in part 8'. Mandibles tridentate in both sexes; male clypeal apex without bevel; scutum usually with median groove or posteromedian pit (Figs. 33, 35); females lacking median scutal groove have apicoventral mandibular tooth enlarged and broadly truncate (Fig. 37); Neotropical 9 9. Scutum with median groove or posteromedian pit (Figs. 33, 35); apicoventral mandibular tooth in female, acute (Fig. 39); labrum quadrilobed; throughout Neotropical Region Incastigvtus Finnamore, new genus 9'. Scutum without trace of median groove or posteromedian pit (Fig. 38); female with api- coventral mandibular tooth enlarged, broadly truncate (Fig. 37); labrum usually bilobed, rarely 4-lobed; Neotropical: restricted to continental South America Llaqhastigmus Finnamore, new genus Volume 4, 1995 211 Tzustigmus Finnamore, new genus (Figs. 6-11, 44, 45, 47) Derivation of Name. — Tzu, a Chinese term meaning master or teacher, in this case referring to the relatively basal posi- tion of the genus within the Stigmina and its subsequent value in the inference of polarization of character states in the sub- tribe. Diagnosis. — The combination of an in- terantennal tubercle, the hind wing me- dia diverging just beyond cu-a, and a mostly unsculptured mesopleuron will separate this genus from all others in the Stigmina. Description. — Head. Labrum quadri- lobed with median emargination narrow and deep; mandibular apex bidentate in male, tridentate in female; apicoventral mandibular tooth of female acute; man- dible without inner basal tooth; male clypeal apex unmodified, not bevelled; 3 clypeal teeth in female; interantennal tu- bercle developed; frontal carina absent; vertex with micropore field undeveloped consisting of only a few associated pits in a puncture-like depression; inner orbits converging below; eyes margined by a carina; occipital carina present, simple in female, foveolate in male; occipital carina complete, not intersecting hypostomal ca- rina. Mesosoma. Scutum without median groove; notaular grooves normal, not elongate. Acetabular carina absent, omau- lus curving forward toward prothorax. Scrobal sulcus absent to weakly present. Hypoepimeral area without coarse sculp- ture. Hind tibia with 2 or 3 spines on pos- terior margin. Mid basitarsus of male as long or longer than next 3 tarsomeres combined. Fore wing setae absent in cel- lular area. Hind wing media diverging just after cu-a; submedian cell of normal size, not reduced. Metasoma. Petiole smooth and round, occasionally finely microcarinate. Pygidial plate narrow, present in female only. Dig- itus elongate, toothed; cuspis not lobed but multitoothed. Type Species. — Tzustigmus syam new species. Distribution. — Oriental, far eastern Pale- arctic. Species Transferred. — Carinostigmus rhin- ocerus Budrys. KEY TO SPECIES OF TZUSTIGMUS (Males of T. rhinoccnis and T. veda are unknown) 1. Hypoepimeral area with a single ventral carina forming upper margin of scrobal sulcus (Fig. 9); scrobal sulcus foveolate; male fore basitarsus unmodified; female with area ad- jacent to propodeal enclosure unsculpted at least posterolaterally; south India to Vietnam 3. syam Finnamore, new species 1'. Hypoepimeral area carinate on ventral half; scrobal sulcus not evident; male fore basitar- sus angulate medially; female with area adjacent to propodeal enclosure finely carinate ... 2 2. At least dorsal surface of petiole microcarinate; Thailand to Vietnam 1. khmer Finnamore, new species 2'. Dorsal surface of petiole unsculptured, shiny, dorsolateral sulci may be present 3 3. Pronotal transverse sulcus unsculptured; south India 4. veda Finnamore, new species 3'. Pronotum carinate on transverse sulcus 4 4. Stemopleural region closely and finely striatopunctate; Russian Federation: far east 2. rhinocerus (Budrys) 4'. Stemopleural region shiny, punctate only; Taiwan 5. iviiming Finnamore, new species 212 Journal of Hymenoptera Research 1. Tzustigmus khmer Finnamore, new species Derivation of Name. — The species epithet, khmer, refers to the 11th century invaders of the Dvaravati Kingdom then occupying present day Thailand which is within the distribution range of this species. Diagnosis. — This species is readily dis- tinguished from its relatives by the micro- carinate petiole (at least dorsally). Description. — Male. Length 4.5 mm. Head. Flagellomeres without visible tyli. Head shiny, without microsculpture. Clypeus obscured by dense appressed se- tae that extend along inner orbits to half height of scape. Inner and outer orbit mar- gined by coarsely foveolate sulcus that is absent ventrally and anterodorsally. Frons lower two thirds with irregular, coarse, transverse carinae. Vertex with sparse, ev- anescent punctures. Ocelli closer to each other than to eye, OOD 1.4 X LOD. Gena with sparse punctures and sparse setae that increase in density ventrally. Occipi- tal carina foveolate, more coarsely so ven- trally. Mesosoma. Pronotum with transverse sulcus longitudinally carinate; side longi- tudinally carinate. Scutum shiny, sparsely punctate, with anterior transverse micro- carinae becoming longitudinal adjacent to admedian lines, posterior edge margined by a transverse row of short carinae. Hy- poepimeral area microcarinate on ventral half. Scrobal sulcus not or at most only slightly evident. Sternopleural region shiny, without microsculpture, sparsely punctate. Omaulus and hypersternaulus coarsely foveolate. Scutellum entirely or mostly shiny, without microsculpture an- teriorly, but sometimes present posterior- ly. Metapleuron carinate. Propodeum with side entirely finely longitudinally carinate, dorsolateral carina evanescent, lateral spheres shiny with microcarinae, enclosure with irregularly areolate sculp- ture. Fore basitarsus angulate at basal third. Metasoma. Petiole round, longitudinal- ly microcarinate dorsally and laterally, ventrolateral carinae faintly indicated pos- teriorly. Terga II-V usually with large sparse punctures in a more or less trans- verse median band. Sterna IV-VI with api- cal band of dense punctures up to one puncture diameter apart, apical margins impunctate. Metasoma with microsculp- ture so fine as to impart an oily sheen that is most evident on the terga. Color. Ground color black. White: pron- otal lobe. Brown to yellow-brown: man- dible, palpi, antenna, legs, except coxae, tegula, stigma. Female. Length 5 mm. Similar to male except as follows: clypeus sparsely setose, underlying sculpture visible, punctures sparse and up to several diameters apart on median area which is slightly bulging. OOD 2.3 X LOD. Body sculpture stronger than in male, punctation closer than in male particularly on sternopleural region where fine punctures are evenly distrib- uted up to 2 diameters apart. Mesopleu- ron longitudinally carinate from dorsal hypoepimeral area to hypersternaulus. Fore basitarsus straight, not angled at bas- al third. Petiole round, entirely microcar- inate. Abdominal tergum VI with narrow, shiny, impunctate pygidial plate. Material Examined {26, 29). — Holotype: 6, VIET NAM: 15 km nw. of Dalat [Da Lat], 1850 m, 5-V-1960. L.W. Quate (BISH). Paratypes: THAILAND: Chiangmai: Doi pui, 2-IV-1958. T.C. Maa (d: 1 BISH). VI- ETNAM: Fyan, 1200 m. 11-VII-9-VIII- 1961. N.R. Spencer (9: 1 BISH). VIET- NAM: Mt. Lang Bian, 1500-2000 m. 19- V- 8-VI-1961. N.R. Spencer (9:1 BISH). Distribution (Map 1). — Thailand to Vi- etnam. 2. Tzustigmus rhinocerus (Budrys), new combination Carinostigmus rhinocerus Budrys 1987:54, 9, c?. Holotype: 9 (not examined), Russian Feder- ation: Khabarovsk, Khekhtsir Range, Levaya River (ZIN). Volume 4, 1995 213 Diagnosis. — The combination of a hy- late. Fore basitarsus straight, sub- poepimeral area carinate on ventral half, cylindrical, not bent or otherwise modi- sternopleural region closely and finely fied. striatopunctate, propodeum with carinate Metasoma. Petiole round, shiny with lateral spheres and petiole shiny without dorsolateral carina partially developed an- microcarinae (dorsolateral carinae are teriorly. Terga II-V with medium size, faintly indicated) separate this species widely scattered sparse punctures; micro- from others in the genus. sculpture on anterior terga imparts an oily Description. — Male. Not seen. sheen to metasoma. Sterna IV and V with Female. Length 6.5 mm. apical bands of punctures, punctures at Head. Clypeus with relatively long de- least 2 diameters apart; sternum VI with curved setae, not obscuring underlying close nearly contiguous punctures, sculpture, punctures coarse and contigu- Color. Ground color black. White: pron- ous on median area which is strongly otal lobe. Brown to yellow-brown: man- bulging. Inner and outer orbit margined dibles except apex, palpi, scape, underside by a coarsely foveolate sulcus which is ab- of flagellum, legs except coxae and hind sent ventrally and anterodorsally. Frons femur, tegula, pygidial plate, irregularly, transversely carinate adjacent Material Examined (19 paratype). — to lower inner orbit becoming striatopunc- RUSSIAN FEDERATION: Khabarovsk: tate toward vertex. Scapal basin dull, Khekhtsir, 18th km. densely microsculptured. Vertex shiny an- Distribution (Map 2). — Russian Federa- teriorly, obscurely striatopunctate, punc- tion far east, tures sparse with most over 3 diameters apart; posterior vertex with transverse mi- 3. Tzustigmus syam Fmnamore, crosculpture. Lateral ocelli closer to each ^^^ species other than to eyes, OOD 2.6 X LOD. Gena Derivation of Name. — The species epithet, striatopunctate, more densely so ventrally. syam, is a Sanskrit term for dark used by Occipital carina simple, not foveolate. 12th century Khmers to refer to the dark Mesosoma. Pronotum with transverse skin of the people migrating south from sulcus longitudinally carinate; side longi- southern China at that time. It is used here tudinally carinate. Scutum shiny, sparsely to refer to the generally dark appearance punctate, with anterior transverse micro- of this species. carinae that become longitudinal adjacent Diagnosis. — The unsculpted (non-cari- to admedian lines, posterior edge mar- nate) hypoepimeral area will distinguish gined by transverse row of short carinae. this species from others in the genus. Mesopleuron longitudinally carinate from Description. — Male. Length 4 mm. dorsal part of hypoepimeral area to hy- Head. Flagellomeres without tyli. Head persternaulus (which is coarsely foveola- shiny with microsculpture on frons, gena te). Sternopleural region shiny with dense, and posterior vertex. Clypeus obscured by fine punctures (that are less then 1 diam- dense appressed setae which extend along eter apart), somewhat finely striatopunc- inner orbits to a point Va height of scape, tate toward posterior. Omaulus foveolate. Inner and outer orbit margined by a Scutellum shiny anteriorly, sparsely punc- coarsely foveolate sulcus except ventrally tured, microsculptured on posterior third, and dorsally. Frons microsculptured, Metapleuron longitudinally carinate. Pro- without carinae except along inner orbital podeal side longitudinally carinate margin where they are evanescent. Vertex throughout; dorsolateral carina present, ir- sparsely punctate, with minute punctures, regular; lateral spheres carinate to propo- appearing impunctate. Ocelli closer to deal enclosure which is irregularly areo- each other then to eyes; OOD 1.4 X LOD. 214 Journal of Hymenoptera Research Figs. 6-11. Tziistigmus syam 9. 6, head, frontal view. 7, head and prothorax, dorsal view. 8, micropore field between lateral ocellus and compound eye. 9, mesosoma, lateral view. 10, mesosoma, dorsal view. 11, me- sosoma, ventrolateral view. Gena almost impunctate, with microsculp- ture present to absent. Occipital carina foveolate, areolae becoming evanescent dorsally. Mesosoma. Pronotum with transverse sulcus longitudinally carinate; side longi- tudinally carinate. Scutum shiny, sparsely punctate, without anterior microcarinae; with transverse row of short longitudinal carinae next to posterior margin. Hypoe- pimeral area microsculptured, not cari- nate; scrobal sulcus distinct, carinate and foveolate. Sternopleural region shiny, without microsculpture, sparsely micro- punctate. Omaulus and hypersternaulus coarsely foveolate. Scutellum shiny, with several punctures mediolaterally. Meta- pleuron not carinate. Propodeal side with Volume 4, 1995 215 evanescent carinae, less so in the discal area; dorsolateral area a series of areolae, lateral spheres shiny, weakly carinate ba- sally becoming evanescent and absent to- ward midline; propodeal enclosure areo- late. Fore basitarsus straight, cylindrical. Metasoma. Petiole round, shiny, with- out microsculpture or dorsolateral carina. Terga impunctate (this character is vari- able in other species and as more speci- mens are encountered may prove to be punctate in some specimens). Sterna IV to VI with apical bands of small punctures less dense than in previous species. Terga with fine microsculpture so as to impart an oily sheen. Color. Ground color black. White: pron- otal lobe. Brown to yellow-brown: man- dibles except apically, palpi, antenna, teg- ula, legs. Female. Length 4.5 mm. Similar to male except as follows: clypeus with 3 median teeth, sparsely setose, underlying sculp- ture visible medially; median area bulging but not strongly so and shiny with punc- tures 1.5 to 3 diameters apart. OOD 2.6 X LOD. Occipital carina not as strongly are- olate. Gena striatopunctate ventrally. Pro- podeal side more strongly carinate than in male. Tergum VI with a narrow shiny py- gidial plate. Materinl Exatnined {26 , 89). — Holotype: 9, THAILAND: Chiengmai Prov., 10 km w. Fang, V-5-69, G.R. Ballmer, on tree trunk (PMA). Paratypes: INDIA: Karna- taka: Bangalore, 916 m, 9-XII-1973, Ghor- pade collection, Bangalore (9: 1 ZMD). Nilgiri Hills, Naduvatam, 6000 ft. May 1958, P.S. Nathan. George R. Ferguson Collection, 1972 {6:1 OSU). THAILAND: Chiengmai: 10 km w. Fang, 5-V-69, G.R. Ballmer, on tree trunk (9:2 PMA). Doi Suthep-Pui natn. Park, Konthathan water- fall area, 600 m, 20-27-X-1979. Zool. Mus. Copenhagen Exped. (9:1 ZMD). VIET- NAM: Dalat [Da Lat] 6 km s., 1400—1500 m, 9-VI-7-VII-1961, N.R. Spencer, 9: 1 BISH. Fyan, 900 -1000 m. ll-VII-9-VIII-'61, N.R. Spencer (9:1 BISH). Fyan, 1200 m. ll-VII-9-VIII-'61, N.R. Spencer (9:1 BISH, 6: 1 BISH). The male was collected at a light trap. Distribution (Map 3). — India to Vietnam. 4. Tzustigmus veda Finnamore, new species Derivation of Name. — The species epithet was arbitrarily selected for the species. Veda, in Hindu perspective is the eternal wisdom realized by the ancient seers or sages preserved over thousands of years by generations of professional reciters in oral transmission. Diagnosis. — The non-carinate, smooth, transverse pronotal sulcus is diagnostic for this species. Description. — Male unknown. Female. Length 4.5 mm. Head. Head shiny, faint microsculpture on frons and posterior vertex. Clypeus with 3 median teeth, sparsely setose, un- derlying sculpture visible medially, shiny, closely punctate, punctures 1-2 diameters apart. Inner and outer orbit margined by a coarsely foveolate sulcus except ventral- ly and dorsally. Frons with fine transverse microcarinae and some coarse transverse carinae near inner orbital sulcus. Vertex shiny with fine microsculpture posterior- ly, punctures fine, sparse. Lateral ocelli closer to each other than to eyes, OOD 1.9 X LOD. Gena shiny, sparsely punctate dorsally, closely punctate to striatopunc- tate ventrally. Occipital carina simple, ob- scurely foveolate ventrally. Mesosoma. Pronotum with transverse sulcus unsculptured although longitudi- nal carinae are present at extreme dorso- lateral end, side longitudinally carinate. Scutum shiny, sparsely punctate, with fine anterior transverse microcarinae that turn longitudinally adjacent to admedian lines; with transverse row of short carinae next to posterior margin. Hypoepimeral area carinate over most of its area. Scrobal sul- cus evanescent. Omaulus and hyperster- naulus foveolate. Sternopleural region shiny, without microsculpture, punctures 216 Journal of Hymenoptera Research fine 1-2 diameters apart. Scutellum shiny anteriorly, posterior Vz microsculptured and sparsely punctate. Metapleuron cari- nate. Propodeum carinate laterally, dor- solateral carina present, lateral spheres finely carinate to propodeal enclosure which is areolate. Metasoma. Petiole round, shiny, with- out microsculpture, dorsolateral carina faintly indicated anteriorly. Terga with few fine evanescent punctures and with microsculpture imparting oily sheen. Ster- na III-V with sparse bands of punctures, VI closely punctate, punctures less than 1 diameter apart near apex. Color. Ground color black. White: pro- notal lobe. Brown to yellow-brown: man- dibles except apex, palpi, scape, tegula, legs. Material Examined (1$). — Holotype: 9, S. INDIA: Cinchona, Anamalai Hills [An- aimalai], 3500 ft. May 1959, P.S. Nathan. George R. Ferguson Collection (OSU). Distribution (Map 4). — Southern India. 5. Tzustigmus wuming Finnamore, new species Derivation of Name. — The species epithet, wuming, is a Chinese term derived from Taoist philosophy referring to that which is unnameable. Diagnosis. — Resembling T. rhinocerus in having the hypoepimeral area carinate on ventral half, the propodeum with closely carinate lateral spheres, and the petiole shiny without microcarinae. Unlike rhino- cerus, wuming has a smooth, shiny, sparse- ly punctate sternopleural region. Addi- tionally the pleuron of wuming is unsculp- tured immediately above hypersternaulus, while in rhinocerus it is sculptured from the mid hypoepimeral area to hyperster- naulus and sternopleural region. Description. — Male. Length 4 mm. Head. Flagellomeres without tyli. Head shiny with microsculpture on frons, vertex and occasionally gena. Clypeus obscured by dense, appressed setae which extend along inner orbits to about half height of scape. Inner and outer orbits margined by a coarsely foveolate sulcus except ventral- ly and dorsally. Frons microsculptured on scapal basin, elsewhere with relatively coarse, irregularly transverse carinae which meet dorsally. Vertex sparsely punctate with minute punctures. Ocelli closer to each other than to eyes; OOD 1.8 X LOD. Gena sparsely punctate dorsally, more densely punctate ventrally; micro- sculpture present to absent. Occipital ca- rina foveolate. Mesosoma. Pronotum with transverse sulcus longitudinally carinate; side longi- tudinally carinate. Scutum shiny, sparsely punctate; with transverse anterior micro- carinae that turn longitudinally adjacent to admedian lines; posterior edge mar- gined by a transverse row of short carinae. Hypoepimeral area carinate. Scrobal sul- cus indistinct, carinae not extending ven- trally to hypersternaulus. Sternopleural region shiny, without microsculpture, sparsely and obscurely punctate. Omaulus and hypersternaulus coarsely foveolate. Scutellum shiny, without microsculpture except faintly microsculptured on poste- rior third; sparsely punctate. Metapleuron carinate on ventral half. Propodeal side carinate, dorsolateral area with slightly more coarse irregular sculpture, lateral spheres carinate; propodeal enclosure are- olate. Fore basitarsus angulate at basal third. Metasoma. Petiole round, shiny, with- out microsculpture or dorsolateral carina except remnant anteriorly. Terga sparsely, obscurely punctate. Sterna III-VI with broad bands of punctures most of which are less than 1 diameter apart. Terga with fine microsculpture imparting oily sheen. Color. Ground color black. White: pro- notal lobe. Brown to yellow-brown: palpi, antenna, legs except coxae, tegula, stigma. Female. Length 5-6 mm. Similar to male except as follows: clypeus sparsely setose, underlying sculpture visible on median area, punctures sparse (1-2 diameters apart) on median area which is slightly Volume 4, 1995 217 bulging, with 3 weakly developed apical in both sexes; apicoventral tooth of female teeth. Frons with irregular transverse ca- acute; inner basal tooth in at least females rinae not as strong as in male and some- of some species. Clypeal apex in male times evanescent dorsally on midline, without bevelled modifications; 4 clypeal OOD 2.2-2.3 X LOD. Sternopleural region teeth present in female, although median slightly more densely punctate than in teeth are often fully reduced imparting a male, punctures pin prick fine. Fore basi- deep, broad, median emargination. Inter- tarsus straight, not angulate. Abdominal antennal tubercle absent. Frontal carina tergum VI with narrow shiny pygidial absent in male, present but simple in fe- plate. Mandibles yellow-brown. male. Vertex without micropore field. In- Material Examined 26, 4 9). — Holotype: ner orbits parallel in female, converging 9, TAIWAN: Wushe, 1150 m, III-16-'83, below in male. Eye not margined by a ca- H. & M. Townes (AEI). Paratypes: TAI- rina. Occipital carina simple, complete, WAN: Wushe, 1150 m, III-23-'83, H. & M. not intersecting hypostomal carina. Townes. 1150 m (d: 1 AEI). Wushe, IV-2- Mesosoma. Scutum without median 83, Henry Townes (9:1 AEI). Wushe, 1150 groove; notaular groove normal, short, not m, IV-19-83, Henry Townes (9: 1 AEI). elongate. Acetabular carina absent; omau- Wushe, 1150 m, IV-26-83, Henry Townes lus curving forward toward prothorax. {6: 1 AEI). Wushe, 1150 m, V-3-83, Henry Scrobal sulcus absent in male, evanescent Townes (9: 1 AEI). in female. Hypoepimeral area without Distribution. — (Map 5). Taiwan. coarse sculpture. Mid basitarsus in male n A r> A cTir-ii/fi tc a >. Unmodified, as long as or longer than next PARASTIGMUS Antropov _ ^ u- j u- j i.u- -..u ^ ^ 3 tarsomeres combmed. Hmd tibia with 2 Parnsti;^miis Antropov 1992:97. Type species or 3 spines on posterior margin. Fore wing Parastigmus argentinus Antropov, 1992, by without setae in cellular area. Hind wing original designation and monotypy. ^^^^^ diverging before cu-a; submedian Diagnosis. — The combination of a mi- cell of normal size, not reduced. crosculptured body, the hind wing media Metasoma. Petiole carinate. Pygidial diverging just before cu-a, a broad pygid- plate broad, present in both sexes. Digitus ial plate, and the lack of an acetabular ca- elongate, toothed; cuspis not lobed, rina will separate this genus from others toothed. in the Stigmina. Distribution (Map 34). — Southern Neo- Description. — Head. Labrum quadri- tropical. lobed, with median emargination narrow Species Transferred. — Stigwus patagonicus and deep. Mandible with apex bidentate Mantero. KEY TO SPECIES OF PARASTIGMUS 1 Males 2 1' Females 5 2 Mesosoma entirely dull, microsculptured; Chile 2. hiiectanis Finnamore, new species 2' Scutellum or sternopleural region shiny, without microsculpture; Argentina 3 3 Sternum II and apices of terga I and II, red 1. argcntinus Antropov 3' Metasoma all black 4 4 Pedicel subequal to length of first flagellomere (1:1) 4. patagonicus (Mantero) 4' Pedicel longer than first flagellomere (1:0.7) 3. nina Finnamore, new species 5 Apex of clypeus transverse with 4 small equally developed teeth; labrum not exposed; Chile 2. htieciivus Finnamore, new species 218 Journal of Hymenoptera Research 5' Apex of clypeus with a broad, deep median emargination or biemargination exposing labrum; lateral teeth prominent; median lobe present or absent; Argentina 6 6 Pedicel longer than first flagellomere (1:0.6-0.7); mesosoma red or black 3. nina Finnamore, new species 6' Pedicel subequal to length of first flagellomere (1:0.9-1.1); mesosoma black 7 7 Clypeus biemarginate, median lobe developed; clypeus and inner orbits bare, without appressed setae 1. argentinus Antropov 7' Clypeus with single broad median emargination, median lobe evanescent or absent; clyp- eus and inner orbits with appressed setae that partially obscure surface 4. patagonicus (Mantero) 1. Parastigmus argentinus Antropov somewhat shiny, weakly microsculptured. Scutum without appressed setae; punc- Parastigmus argentinus Antropov 1992:99 $ . Ho- . i. i. i lotype: 9 (not examined), ARGENTINA: Ne- "'"^ f?"''^"' ^^^^^^^^^'^ ^^'^^^^^^ g^^^^es uquen, Collon Cura, 19-1-1980. WiUink, Fi- fo^^olate. Scutellum sparsely punctate, dalgo, Dominguez, Claps (LILLO). punctures evanescent. Scrobal sulcus ab- sent. Mesopleuron impunctate. Hyperster- Diagnosis. — The male is recognized by naulus and omaulus weakly foveolate. its red metasomal sternum II, longitudi- Metapleuron shiny, weakly microsculp- nally microsculptured scutum, and length tured. Propodeum with areolate sculpture of flagellomere I which is 0.75 X length of over most of surface, sculpture finer to- pedicel. Males of other species have a first ward metapleuron; dorsal enclosure not flagellomere length subequal to pedicel defined by a carina. and /or microsculpture of scutum dem- Metasoma. Microsculptured, punctate onstrating no particular orientation. The at apices of terga II-VI, more coarsely so metasoma of other species is black and on posterior terga; terga V and VI sparsely may also prove to be black in other spec- punctate over entire dorsal area. Tergum imens of this species. The female is rec- VII with ill-defined pygidial plate, punc- ognized by its biemarginate clypeal edge tured, shiny, impunctate apically. Setal with median lobe developed. density abruptly increasing on tergum IV Description. — Male. Length 3.5 mm. to posterior metasoma. Head. Flagellomeres without tyli. Ratio Color. Ground color black. White: palpi, of pedicel to flagellomere I is 1:0.75. Head labrum, mandible except apex, scape ven- coarsely microsculptured, without shiny trally, pronotal lobe, anterior spot on teg- areas. Clypeus and subantennal area ob- ula. Brown to yellow-brown: antenna ex- scured by dense appressed setae which cept ventral scape, fore leg beyond coxa, extend broadly up the inner orbits to a mid leg beyond coxa, hind leg except coxa point 1.6 X scapal length. Frons and ver- and femur, tegula except anterior spot, tex with sparse obscure punctures, ap- stigma. Red: apices of metasomal terga I pearing impunctate at low magnification, and II, sterna II, base of III and VII. Ocelli closer to each other than to eyes; Female. Length 4 mm. Similar to male OOD 1.6 X LOD. Genal punctures eva- except as follows: ratio of pedicel to fla- nescent. gellomere I is 1:1.1. Clypeus almost de- Mesosoma. Pronotum with transverse void of setae and punctures, shiny, with- sulcus longitudinally carinate. Scutum out microsculpture; apex biemarginate longitudinally microsculptured; scutellum with lateral teeth prominent and median and mesopleuron above hypers ternaulus lobe developed. Inner orbits with several microsculptured; sternopleural region setae that do not obscure underlying Volume 4, 1995 219 sculpture. Subantennal area without setae. Punctures of frons and vertex sparse, ev- Frons somewhat shiny, with weak micro- anescent. Ocelli closer to each other than sculpture. Vertex posterior to ocelli, trans- to eyes; OOD 1.4-1.6 X LOD. Genal punc- versely microsculptured. OOD 2.2 X LOD. tures evanescent. Upper frons and vertex sparsely punctate. Mesosoma. Entirely microsculptured. Lower gena more densely punctate than Pronotum with transverse sulcus longitu- upper gena, punctures 1-2 diameters dinally carinate; pronotal side longitudi- apart near mandibular base. Scutum nally carinate. Scutum and scutellum sparsely punctate. Mesopleuron shiny, sparsely punctate, with many large im- without microsculpture; scrobal sulcus punctate areas; microsculpture simple, not slightly impressed; sternopleural region oriented in any direction. Notaular closely punctate, punctures minute, 1-3 grooves deeply impressed, coarsely fov- diameters apart. Pygidial plate broad, eolate. Scrobal sulcus absent. Mesopleu- shiny, punctate on midline, more broadly ron impunctate, without carinae except punctate basally. Colored as in male but occasionally at lower posterior margin, without red. Omaulus and hypersternaulus foveolate. Material Examined {16, 19). — ARGEN- Metapleuron microsculptured, without ca- TINA: Neuquen: Collon Cura, 19-1-1980. rinae, impunctate. Propodeal sculpture ir- Willink, Fidalgo, Dominguez, Claps. regularly areolate on dorsal and posterior Distribution (Map 6). — Argentina. surface. Propodeal enclosure not evident. Propodeal side with sculpture evanescent 2. Parastigmus huecuvus Finnamore, toward metapleuron. new species Metasoma. Microsculptured except ter- Derivation of Name. — Huecuvue, from gum VII. Punctures of terga and sterna Araucanian mythology (Chile), are spirits small, sparse, evanescent. Pygidial plate able to change into any form they wished shiny medially, without microsculpture, 2 for the purpose of doing evil, in this case or 3 punctures laterally. Setal density in- creating confusion among taxonomists creasing toward posterior metasoma. due to aberrant characters. Color. Ground color black. White: Man- Diagnosis. — Males are distinguished dible except apex, scape ventrally, prono- from other Parastigmus by the entirely mi- tal lobe, anterior spot on tegula. Brown to crosculptured mesosoma and black abdo- yellow-brown: palpi, antenna except ven- men. Other species have a shiny sterno- tral scape, stigma, legs but coxae and fem- pleural region, and the metasoma is in ora tend to be darker brown and fore leg part red in patagonicus. The female is tends to be lighter yellow-brown, unique in the genus in having 4 sub- Female. Length 4 mm. Similar to male equally developed clypeal teeth and with- except as follows: clypeus entirely, evenly out a broad median emargination. The setose although setae not obscuring un- clypeus in females of other species has a derlying surface, shiny, microsculpture broad median emargination in which the absent, with several relatively large punc- median teeth or lobe is recessed or absent tures subapically along apex. Clypeal apex resulting in prominent lateral teeth. with 4 subequally developed teeth, with- Description. — Male. Length 3.5 mm. out a broad median emargination. Suban- Head. Flagellomeres without tyli. Ratio tennal area and inner orbits without setae, of pedicel to flagellomere I is 1:0.9. Head OOD 2 X LOD. Pygidial plate shiny, im- entirely microsculptured. Clypeus and punctate over lateral third; median third subantennal area obscured by dense ap- slightly raised, shiny, without microsculp- pressed setae which extend along inner ture, irregularly punctate and irregularly, orbits to a point 1.5 times length of scape, finely rugose. 220 Journal of Hymenoptera Research Material Examined (36, 109). — Holo- type: 9 , CHILE: Santiago — Cordillera, Re- serva Rio Clarillos 1-20-II-1989. R. Miller 6 L. Stange. Malaise Trap (FSCA). Para- types: CHILE: Concepcion: Concepc. 19/ 12 1907. P. Herbst (9:1 NMW). The 19/ 12 and the 7 in 1907 are inscribed in red ink on the specimen label, in the latter case the 7 is written over the 4 in 1904. Specimen bears a second label with the numbers 17 and 84. Concepc. 27/12 1907. P. Herbst (6:2 NMW). The 27/12 and the 7 are inscribed in red ink on the specimen label in the latter case the 7 in 1907 is writ- ten over the 3 in 1903 or the 4 in 1904. Concepc. 3-1908. P. Herbst (6: 1 NMW). The 3 and 8 are inscribed in red ink, the latter over the 4 in 1904. Santiago: Cordil- lera, Reserva Rio Clarillos 1-20-II-1989. R. Miller & L. Stange. Malaise Trap (9: 5 FSCA, 2 PMA). Cordillera, Reserva Rio Clarillos 21-28-11-1989. R. Miller & L. Stange. Malaise Trap (9:1 FSCA). Distribution. — (Map 7). Chile. 3. Parastigmus nina Finnamore, new species Derivation of Name. — Nina is a Quechua term meaning fire, referring to the red me- sosoma of the holotype. Diagnosis. — The short flagellomere I (0.6-0.7 X length of pedicel) distinguishes this species from others in the genus ex- cept males of argentinus. In all other spe- cies the length of flagellomere I equals 0.9-1.1 of the pedicel length, except male of argentinus which differs from nina in its red second metasomal sternum. Parastig- mus nina has a pedicekflagellomere I ratio of 1:0.6-0.7, black metasoma, and the me- sosoma may be red or black, that of the female holotype is extensively red. Description. — Male. Length 4 mm. Head. Flagellomeres without tyli. Ratio of pedicel to flagellomere I is 1:0.7. Head microsculptured except near frontal carina where microsculpture is evanescent. Clyp- eus and subantennal area obscured by dense appressed setae which extend broadly up the inner orbits to a point al- most twice the length of the scape. Frons and vertex irregularly punctate (1-5 di- ameters apart). Ocelli slightly closer to each other than to eyes; OOD 1.5 X LOD. Genal punctures evanescent. Mesosoma. Scutum, scutellum, meso- pleuron, and upper metapleuron exten- sively shiny, microsculptured in part. Pro- notum with transverse sulcus longitudi- nally carinate; side weakly carinate. An- terior quarter of scutum partially ob- scured by dense appressed setae, micro- sculptured; median and posterior areas of scutum shiny with evanescent microsculp- ture, punctures sparse, notaular grooves foveolate. Scutellum shiny, sparsely punc- tate anteriorly, microsculptured posteri- orly. Scrobal sulcus absent. Mesopleuron immediately ventral to hypersternaulus, minutely punctured; elsewhere punctures sparse, evanescent. Hypersternaulus and omaulus weakly foveolate. Mesopleuron microsculptured posteriorly, mostly shiny anteriorly. Metapleuron microsculptured over most of surface except dorsal shiny patch. Propodeum coarsely microsculptu- red on side, finely areolate dorsally, pos- teriorly, and on lateral sphere. Propodeal enclosure not defined by a carina. Metasoma. Microsculptured, impuncta- te except tergum VII. Sterna II-IV minute- ly punctate posteriorly. Pygidial plate ill- defined, shiny medially, without micro- sculpture, several punctures present lat- erally. Setal density gradually increasing toward posterior metasoma. Color. Ground color black. White: man- dible except apex, labrum, scape ventrally, pronotal lobe, anterior spot on tegula. Brown to yellow-brown: palpi, antenna except ventral scape, stigma, legs except coxae, hind leg is darker than other legs. Female. Length 4.0-4.5 mm. Similar to male except as follows: ratio of pedicel to flagellomere I is 1:0.6-0.7. Clypeus sparse- ly setose, shiny, sparsely punctate with median bulge; apex broadly and deeply emarginate over median 0.45; lateral teeth Volume 4, 1995 221 acute, prominent; median teeth absent. In- the pedicel subequal in length to flagello- ner orbits and subantennal area not setose, mere I, and the broadly emarginate clyp- OOD 2.3 X LOD. Mesosoma red or black, eus with prominent lateral teeth but with- the holotype has a red mesosoma with out a median lobe. black in the following areas: propodeal en- Description. — Male. Length 4.5-5.0 mm. closure, mid ventral line, coxae, mid tro- Head. Flagellomeres without tyli. Ratio chanter, mid femur, hind leg; otherwise of pedicel to flagellomere I is 1:1. Head colored as in the male. The 9 paratype has coarsely microsculptured, without shiny the same color pattern as the male. Pygid- areas. Clypeus and subantennal area ob- ial plate broad, shiny, sparsely punctate scured by dense appressed setae which medially. extend broadly up inner orbit to a point Material Examined {16, 2$). — Holotype: 1.5 X length of scape. Upper frons and 9, ARGENTINA: Neuquen, Collon Cura lower gena sparsely punctate. Ocelli closer 19-1-1980. Willink, Fidalgo, Dominguez, to each other than to eye. OOD 1.6-1.7 X Claps (LILLO). Paratypes: ARGENTINA: LOD. Neuquen: Collon Cura 19-1-1980. Willink, Mesosoma. Pronotum with transverse Fidalgo, Dominguez, Claps {6: 1 LILLO). sulcus longitudinally carinate. Scutum Santa Cruz: 3 km s. Rio Chico at Hwy. 3; and scutellum microsculptured, without XII-1 1-1966, 90m, E.I. Schlinger & M.E. Ir- appressed setae, sparsely punctate; notau- win (9:1 CAS). lar grooves foveolate. Mesopleuron shiny. Distribution (Map 8). — Argentina. without microsculpture ventral to hypo- epimeral area which has evanescent mi- 4. Parastigmus patagonicus (Mantero) crosculpture (may be longitudinally ori- new combination ented). Punctures on sternopleural region ^,. , . », , ir>ni mr, ^ TT 1 minute, 3-4 diameters apart. Scrobal sul- Sti';^miis imtamnicus Maniero 1901:199 d. Holo- . , , , lype- 6, Rio Sta. Cruz, Silvestri (MCSN). ^^^ absent, mdicated only by change m microsculpture. Hypersternaulus and Diagnosis. — The male can be distin- omaulus weakly foveolate. Metapleural guished from all other species in the genus microsculpture evanescent. Propodeum by the combination of a pedicel subequal microsculptured, areolate over most of to length of flagellomere I, and a shiny surface except near metapleuron where it mesopleuron without microsculpture ven- is carinate; propodeal enclosure not indi- tral to hypoepimeral area. The male of this cated by a carina. species is superficially similar to huecuvus Metasoma. Metasoma microsculptured with which it shares a long first flagello- except tergum VII, punctures evanescent mere. In addition to the mesopleural mi- but more evident on apices of apical terga. crosculpture the length of the petiole may Pygidial plate ill-defined, shiny, sparsely be used to separate the two species; the punctate. Setal density gradually increas- petiole slightly shorter than tergum I in ing toward posterior of metasoma. huecuvus and slightly longer than tergum Color. Ground color black. White: palpi, I in patagonicus. The petiole length should labrum, mandible except apex, scape ven- be used with caution since it is subject to trally, pronotal k^be, anterior spot on teg- variation in closely related genera and has ula. Brown to yellow-brown: antenna ex- been excluded from descriptions for that cept scape ventrally, fore leg, mid leg, reason. The number of specimens avail- hind leg except coxa and femur, tegula ex- able for this genus is insufficient to deter- cept anterior spot, mine variation in petiole length. Female. Length 6.0-6.5 mm. Similar to Females can be separated form other male except as follows: clypeus evenly species in the genus by the combination of covered with dense setae that partially ob- 222 Journal of Hymenoptera Research scure sculpture beneath and extend up in- ner orbit to a point subequal to length of scape. Clypeus without microsculpture, shiny, punctures minute and sparse. Clyp- eal apex broadly emarginate, lateral teeth prominent, median lobe (teeth) reduced to a small mid point. Subantennal area with- out setae. OOD 1.8-1.9 X LOD. Mesosoma with scrobal sulcus slightly more im- pressed than in male. Propodeum more coarsely areolate than male. Pygidial plate broad, shiny with submedian puncture row that converges apically with the me- dian. Material Examined {56, 2 9).— ARGEN- TINA: Catamarca: Andalhuallas 2000m. Santa Cruz: Rio Santa Cruz (Type locali- ty). Tucuman: Amaicha. Amaicha del Val- le 1900m. Distribution (Map 9). — Argentina. ARAUCASTIGMUS Finnamore, new genus (Figs. 12-15) Derivation of Name. — The genus is named after the Araucanian people of Chile the country of origin of most known specimens. Diagnosis. — The combination of an ace- tabular carina present, hind wing media diverging just before cu-a and the meso- soma (except propodeum) without mi- crosculpture will separate this genus from all others in the Stigmina. Description. — Head. Labrum quadri- lobed with a narrow deep median emar- gination. Mandible with apex bidentate in male, tridentate in female; apicoventral tooth acute in female; inner basal tooth ab- sent. Clypeal apex in male without bev- elled modification, in female with 4 teeth. Interantennal tubercle absent. Frontal ca- rina present, simple. Vertex with micro- pore field present, diffuse. Inner orbits converging below. Eye not margined by a carina. Occipital carina present, simple, complete, and separated from hypostomal carina. Mesosoma. Scutum without median groove, notaular grooves normal, short, not elongate. Acetabular carina absent, omaulus curving forward toward protho- rax. Scrobal sulcus absent. Hypoepimeral area without coarse sculpture. Mid basi- tarsus of male unmodified, as long or lon- ger than next 3 tarsomeres combined. Hind tibia with 2 or 3 spines on posterior margin. Fore wing with setae in cellular areas. Hind wing media diverging before cu-a; submedian cell of normal size, not reduced. Metasoma. Petiole carinate. Pygidial plate narrow, absent in male. Digitus somewhat elongate; cuspis with a large lobe bearing several teeth. Type Species. — Araiicastigmiis glabratus (Kohl), new combination. Distribution (Map 35). — Southern Neo- tropical: Chile. Species Transferred. — Stigmus glabratus Kohl. KEY TO SPECIES OF ARAUCASTIGMUS 1 Propodeum sculptured throughout, area adjacent to metapleuron microsculptured (Figs. 14, 15) 2. masneri Finnamore, new species 1' Propodeal side with basal shiny area adjacent to metapleuron 2 2 Propodeum microsculptured, shiny without microsculpture along mid dorsal line (outside enclosure) in male and also on mid posterior line in female 3. yanillus Finnamore, new species 2' Propodeum with sculpture in addition to microsculpture at least on dorsolateral areas in male; female without shiny areas on propodeum other than adjacent to metapleuron . . 1. glabratus (Kohl) Volume 4, 1995 223 Figs. 12-17. Araucastigmus and Paracrabro. 12, Amucastigmus masneri 3, head, frontal view. 13, A. masneri 6, head, dorsal view. 14, A. masneri 6, mesosoma, lateral view. 15, A. masneri 6 , mesosoma, dorsal view. 16, Paracrabro froggatti 9, mandible, outer view. 17, Paracrabro froggatti 9, mandible, inner view. 1. Araucastigmus glabratus (Kohl), new combination Stigmus glabratus Kohl 1905:358 9. Holotype*: 9, CHILE: Corral [Coual?], 3-1904, P. Herbst (NMW). Ding)iosis. — Propodeum with shiny area adjacent to metapleuron and with both microsculpture and sculpture at least on dorsolateral areas. Other species are either microsculptured throughout or macro- sculpture is absent or shiny areas are much more extensive. Description. — Male. Length 3.2-5 mm. Head. Flagellomeres without tyli. Head shiny, mostly without microsculpture, sparsely punctate. Clypeus, subantennal area and inner orbit obscured by dense appressed setae which extend broadly up inner orbit to a point 1.5 X length of scape. 224 Journal of Hymenoptera Research Frons impunctate, weakly microsculp- tured. Vertex transversely microsculp- tured. Ocelli closer to each other than to eyes. OOD 1.2-1.3 X LOD. Mesosoma. Pronotal transverse carina foveolate laterally; transverse sulcus un- sculptured; side irregularly carinate. Scu- tum and scutellum microsculptured. Scu- tum usually with a transverse row of mi- crocarinae next to posterior margin; punc- tation variable, ranging from small sparse pinpricks to larger more dense (3^ diam- eters apart) punctures more or less concen- trated in the notaular region; anterior scu- tal margin nearly contiguously punctate between notaular grooves. Scutellum sparsely punctate on disc. Mesopleuron shiny, microsculpture at most weakly evi- dent, nearly impunctate; scrobal sulcus ab- sent; hyperstemaulus and omaulus weakly foveolate. Metapleuron impunctate, weak- ly microsculptured. Propodeal side shiny adjacent to metapleuron, otherwise micros- culptured and finely areolate. Propodeal enclosure weakly defined, not carinate. Metasoma. Weakly microsculptured. Punctures evanescent on terga I-III; poste- rior band of punctures on terga IV- VII that increases in width and puncture density to- wards posterior metasoma. Sterna minute- ly punctured. Color. Ground color black. White to cream: palpi, mandibles except apically, antenna ventrally except apical flagello- meres, pronotal lobe, tegula, fore leg, mid leg except base of coxa and median area of femur, hind leg except base of coxa and basal fj of femur. Yellow-brown: apical flagellomeres of antenna, base of mid and hind coxae, median area of mid femur, basal % of hind femur. Orange: apex of ter- gum VII of metasoma. Female. Length 4-6 mm. Similar to male except as follows: clypeus without micro- sculpture, evenly punctured and setose, se- tae not obscuring underlying sculpture; punctures small about 1-2 diameters apart; apex with 2 median teeth separated by a shallow emargination; setae extend from clypeus over subantennal area and up in- ner orbit to a point slightly greater than that of scapal apex. OOD 1.8-2.0 X LOD. Pygidial plate present, microsculptured, with several punctures. Antenna beyond pedicel brown; coxae, femora, and hind tib- ia suffused with brown; remaining area of legs more yellowed than in male. Apex of metasomal tergum VI orange. Material Examined {266, 43 9 ).— ARGEN- TINA: Neuquen: Pucara, Lago Lacar. Pu- cara, P. Nac. Lanin. San Martin Andes. CHILE: Arauco: Contulmo, Palo Botado. Aysen: Lago Frio. Puerto Cisnes. Cautin: Afunalhue Villarrica. Chiloe: Chiloe Is- land, Dalcahue. Curico: Cord. Curico, El Coigo. Cubillo, C. Curico. Estero La Juala, Los Queues, Cord. Curico. Los Lagos: Cor- ral (type locality). Malleco: P. N. Contul- mo. Victoria, 4km s., 300m. Nuble: Las Trancas, Cord. Nuble (Chilian). O'Higgins: Graneros. Santiago: El Canelo. Lo Canas. Talca: Altos de Vilches, 1200m. Valparaiso: Los Andes. Vina del Mar. Valparaiso. Distribution (Map 10). — Argentina, Chile. 2. Araucastigmus masneri Finnamore, new species Derivation of Name. — The species is named in honour of Lubomir Masner, col- lector of half the known specimens. Diagnosis. — Propodeum without shiny areas. Other species in the genus have a shiny area on the propodeum adjacent to the metapleuron. Description. — Male. Length 4.0-5.5 mm. Head. Flagellomeres without tyli. Head sparsely punctate, shiny, usually without microsculpture except on frons but larger specimens have evanescent microsculpture on vertex and gena. Clypeus and subanten- nal area obscured by dense appressed setae which extend up inner orbit to a point slightly greater than height of scape. Frons impunctate, microsculptured. Vertex mi- nutely and sparsely punctate. Ocelli closer to each other than to eyes. OOD 1.8-1.9 X LOD. Mesosoma. Pronotal transverse carina Volume 4, 1995 225 foveolate laterally, transverse sulcus un- type: (5, CHILE: Cautin, Conguillio Nation- sculpted; pronotal side with a few short ca- al Park, 1150m 4-II-1988 L. Masner, Aurau- rinae. Scutum with contiguous punctures carin [Araucarial , Araucariaceae], NotJiofa- between notaular grooves near anterior gus [Fagaceae] (BRD). Paratypes: ARGEN- margin, otherwise impunctate or nearly so; TINA: Neuquen: Lanin, Pucara 15-III-1971 shiny, without microsculpture; posterior Schajovskoy (9: 4 IIES). CHILE: Arauco: margin weakly longitudinally microcari- Pichinahuel, Cord. Nahuelbuta 20-28-1- nate; notaular groove deeply impressed, 1959 L. Pefia {6:1 BRD). Aysen: Lago Frio foveolate. Scutellum shiny on anterior half, 20/23-1-61 L. Peiia {6: 1 MCZ). Cautin: microsculptured on posterior half; with same data as holotype ( c5 : 2 BRD, 2 PM A; several scattered punctures on transverse 9: 4 BRD, 5 PMA). Los Lagos: Osomo, midline. Mesopleuron shiny, without mi- Puyehue Nat. Pk., Antillanca, 1200m 16-11- crosculpture, punctures sparse, obscure; 1988 L. Masner, NotJwfngus [Fagaceae] near scrobal sulcus absent. Hyperstemaulus and treeline {6: 1 BRD). Magallanes: Laguna omaulus foveolate. Metapleuron weakly Amarga, 12-60 Pena (9: 1 IIES). Laguna microsculptured, impunctate. Propodeum Amarga, Natales XII-14-21-60 Luis Pena without shiny areas, entirely microsculp- (9:1 OSU). Malleco: Cord, de las Raices, tured; evanescent, irregular, areolae on lat- 40km e. Curacautin 6-7-II-1979, 1200m D. eral spheres; propodeal enclosure not de- & M. Davis & B. Akerbergs (9:1 USNM). fined by a carina, with irregular carinae Curacautin, R. Blanco 11-64 Luis E. Pefia (d: medially and on anterior margin. 1 AEI). Icalma, 31-XII-1958 Luis E. Pena (9: Metasoma. Terga dull, microsculptured, 1 BRD). Liucura 1800m, Jan. 1959 L.E. Pefia punctures evanescent; posterior metasoma (9:1 OSU). Nuble: Las Trancas 1600m 19- more shiny with punctures forming more 22-1979 (9:1 PMA). Las Trancas, Chilian or less discrete apical bands on terga. Ster- 1-19-22-1979, 1600m L. Pena (9: 1 AEI). na microsculptured, impunctate. O'Higgins: Graneros 1100m 4-III-62 Pena Color. Ground color black. White: palpi, (9:1 MCZ). Valdivia: 30km w. La Union, mandible except apex, pronotal lobe, ante- Las Traucas 500m 8-11-II-1988 PT, L. Mas- rior spot on tegula. Brown to yellow- ner {6:1 BRD). brown: scape, pedicel, at least flagellomer- Distribution (Map 11). — Argentina: Neu- es I-V and occasionally I-XI, tegula except quen. Chile, anterior spot, trochanters, fore and mid tib- iae and tarsi, occasionally bases of sterna 3. Araucastigmus yanillus Finnamore, IV, V and apices of VI and VII. ^^^ species Female. Length 4.5-5.5 mm. Similar to Derivation of Name. — The name yanillus male except as follows: clypeus shiny, is derived from two Quechua terms, yana evenly punctured and setose, setae not ob- meaning black and illay meaning bright or scuring underlying sculpture; punctures shiny, in reference to the shiny black ap- minute, 2-3 diameters apart. Clypeal apex pearance of this species, with 2 teeth separated by shallow median Diagnosis. — The extensive shiny areas in emargination. Subantennal area less dense- the female propodeum and in the male ly setose than clypeus. OOD 2.0-2.1 X propodeum, the lack of sculpture overlying LOD. Scutum and mesopleurc^n more microsculpture, and the lateral shiny area coarsely punctate than in male. Pygidial next to the metapleuron will separate this plate weakly microsculptured, with several species from all others in the genus. Other punctures. Flagellum of antenna brown to species have the propodeum entirely mi- black, mid tibia occasionally black, sterna crosculptured or they have sculpture on at black. least lateral spheres. Material Examined {106, 209). — Holo- Description. — Male. Length 4 mm. 226 Journal of Hymenoptera Research Head. Flagellomeres without tyli. Head anterior spot on tegula, trochanters, fore sparsely punctate, shiny, without micro- and mid tarsi, coxae ventrally. Yellow- sculpture except sometimes evanescent mi- brown: scape, pedicel, flagellomeres I-VII crosculpture on gena. Clypeus and suban- ventrally, coxae dorsally, femora, tibiae, tennal area obscured by dense appressed hind tarsus, and sterna beyond II more or setae which extend up inner orbit to a less suffused. point 1.1 X length of scape. Lower frons Female. Length 4.5-5.5 mm. Similar to impunctate, shiny; upper frons with sev- male except as follows: clypeus sparsely eral scattered punctures. Vertex sparsely and evenly setose, more sparsely setose punctate. Ocelli closer to each other than to than in other species, punctures minute eyes. OOD 1.5-1.6 X LOD. with several larger punctures medially. Mesosoma. Pronotal transverse carina Clypeal apex with 2 teeth separated by foveolate laterally, several longitudinal ca- shallow median emargination. Subanten- rinae sometimes present near median re- nal area without setae. OOD 2.1 X LOD. gion of transverse sulcus. Scutum with Propodeum with sculpture on lateral variable, non-contiguous punctures and a spheres variable from nearly completely few fine, transverse microcarinae on ante- shiny (most specimens) to almost entirely rior margin between notaular grooves; mi- microsculptured, without sculpture over- crosculpture present between admedian lying microsculpture; area adjacent to me- lines. Scutum otherwise shiny, without mi- tapleuron shiny. Pygidial plate narrow crosculpture, mostly impunctate, few with median area minutely punctate. Mid punctures clustered posterior to notaular and hind legs brown except trochanters, groove; notaular groove deeply impressed. Material Examined (8(?, 109). — Holotype: foveolate; posterior margin of scutum 9, CHILE: Cautin: Conguillio National smooth, without transverse row of longi- Park 1150m 4-II-1988 L. Masner, A?/rni/cann tudinal carinae. Scutellum shiny on ante- [Araiicaria? , Araucariaceae], Nothofagiis [Fa- rior half, microsculptured on posterior half; gaceae] (BRD). Para types: ARGENTINA: a few punctures along transverse median Neuquen: Pucara (S.M. Andes) XII-6-16- discal area. Mesopleuron shiny, impuncta- 1968 C.C. Porter {6:1 MCZ). CHILE: Ar- te above stemopleural region; scrobal sul- auco: Pichinahuel, Cord. Nahuelbuta 10- cus absent. Hypersternaulus foveolate; 20-1-1959 L. Pena (9:1 BRD). Cautin: same omaulus not foveolate, usually evanescent data as holotype (6:5 BRD, 2 PMA; 9: 1 to absent toward hypersternaulus but con- BRD, 5 PMA). Curico: El Coigo Jan-1961 J. tinned ventrally. Metapleuron shiny, with- Foerster (9:1 BRD). Nuble: Las Trancas, out microsculpture. Propodeum shiny lat- 1400m Chilian XII-19-21-1983 Luis Pena erally adjacent to metapleuron, otherwise (9:1 AEI). microsculptured without raised lines Distribution (Map 12). — Argentina: Ne- (sculpture) except on median area of pro- uquen. Chile, podeal enclosure where several more or PARACRABRO Turner (Figs. 16, 17) less transverse carinae are usually evident. Propodeal enclosure not defined by a ca- rina. Paracrabro Turner 1907:274. Type species: Para-' Metasoma. Weakly microsculptured, cratro /ro^^am' Turner, 1907 by original des- somewhat shiny compared to other spe- ignation and monotypy. cies; punctures minute, sparse, evanescent. Diagnosis. — The presence of an acetab- not forming apical bands on posterior ter- ular carina, hind wing media diverging ga. Sterna microsculptured, impunctate. just before cu-a, and the lack of a discrete Color. Ground color black. White: palpi, micropore field on the vertex will separate mandible except apically, pronotal lobe, this genus from all others in the Stigmina. Volume 4, 1995 227 Description. — Head. Labrum weakly quadrilobed (appears bilobed), median lobes evanescent. Mandible with apex bi- dentate in male, pentadentate in female; apicoventral tooth in female acute; inner basal tooth absent. Clypeal apex in male without bevelled modifications; female clypeus with 4 apical teeth although me- dian teeth are greatly reduced imparting a bidentate appearance. Interantennal tuber- cle present, protruding. Frontal carina present as a simple raised line immediate- ly above interantennal tubercle. Vertex with loose association of pits rather than micropore field; pits shifted anteriorly in female. Inner orbits parallel in female, converging below in male. Eye margined by a carina on inner orbit in female, en- tirely margined in male. Occipital carina foveolate, more or less raised ventrally, not intersecting hypostomal carina. Mesosoma. Scutum without median groove, notaular groove not elongate. Ac- etabular carina present, omaulus intersect- ing acetabular carina, subomaulus absent. Scrobal sulcus absent in female, evanes- cent in male. Hypoepimeral area without coarse sculpture except in area of scrobal sulcus in male, microsculptured only. Mid basitarsus of male flattened, curved with posterior subapical expansion. Hind tibia excluding apex, with 4 posterior spines in female, 2 in male. Fore wing cellular area without setae in male, setae present in fe- male. Hind wing media diverging at cu-a; hind wing submedian cell not reduced, of normal size. Metasoma. Petiole rounded, carinae present apically and basally. Pygidial plate narrow, trough-like, absent in male. Male genitalia unusually elongate relative to other genera in the subtribe, digitus elongate relative to cuspis. Distribution (Map 13). — Australia. Paracrabro froggatti Turner Paracrabro froggatti Turner 1907:275 9. Holo- type 9 (not examined), Australia: Victoria. Description. — See generic description above. Material Examined {IS, 1 9).— AUSTRA- LIA: New South Wales: Elizabeth Beach, 21km s. Forster. Western Australia: Gas- coyne Exp. Sta., Carnarvon. Distribution (Map 13). — Australia. AYKHUSTIGMUS Finnamore, new genus (Figs. 18-29) Derivation of Name. — Aykhu is a Quechua term meaning extravagant or fancy in reference to the unusually com- plex sculpturing found in members of this genus. Diagnosis. — The coarsely sculptured mesopleuron will separate this genus from all others in the Stigmina. Description. — Head. Labrum quadri- lobed, with a shallow broad median emar- gination. Mandible with apex in male bi- dentate, in female tridentate; apicoventral tooth in female acute; inner basal tooth ab- sent. Clypeal apex in male without bev- elled modification; female clypeal apex with 4 teeth. Interantennal tubercle absent. Frontal carina absent. Vertex with micro- pore field between lateral ocellus and eye margin. Inner orbits converging below. Eye margined by a strong carina. Occipital carina present, complete, not joining hy- postomal carina, simple in female, raised and foveolate in male. Mesosoma. Scutum with median groove usually absent but occasionally evanescent posteriorly; notaular groove normal, not elongate. Acetabular carina present, inter- cepted by omaulus. Scrobal sulcus pres- ent, evanescent. Hypoepimeral area coarsely sculptured. Mid basitarsus in male unmodified. Posterior margin of hind tibia with 2 or 3 spines. Fore wing with setae absent from cellular area. Hind wing with media diverging before or after cu-a; submedian cell of normal size, not reduced. Metasoma. Petiole carinate. Pygidial 228 Journal of Hymenoptera Research plate narrow, absent in male. Genitalia Distribution (Map 36). — Neotropical, with digitus elongate, acute, cuspis south of the Brazilian Highlands although (lobed?) with tooth. a single record exists for Venezuela, pos- Type Species. — Ai/khiistigmus patanawi sibly an introduced species or mislabelled new species. specimen. KEY TO SPECIES OF AYKHUSTIGMUS (Male of A. phasti is unknown) 1 Males 2 r Females 4 2 Hind wing media diverging before cu-a (Fig. 48); vertex and gena striatopunctate (Figs. 19, 21); frons irregularly carinate; scutum closely and coarsely punctate to striatopunctate posteriorly (Fig. 23) 2. patanaivi Finnamore, new species 2' Hind wing media diverging after cu-a (Fig, 49); head at most with microsculpture; scutum with smaller more widely separated punctures 3 3 Frons with pair of small swellings anterior to mid ocellus on each side of frontal carina 1. fritzi Finnamore, new species 3' Frons smoothly rounded, without swellings 4. waraiva Finnamore, new species 4 Hind wing media diverging before cu-a (Fig. 48); vertex transversely striatopunctate; upper frons striatopunctate 2. patanaivi Finnamore, new species 4' Hind wing media diverging after cu-a (Fig. 49); frons and vertex at most with microsculp- ture 5 5 Frons with a pair of broad swellings anterior to mid ocellus that are visible in lateral profile 1. fritzi Finnamore, new species 5' Frons evenly rounded, without swellings 6 6 Clypeus with 4 apical teeth (Fig. 24); scutum microsculptured on anterior Va to Vi, shiny and punctate to striatopunctate on posterior half (Fig. 29) 4. ivarawa Finnamore, new species 6' Clypeus with 2 apical teeth; scutum shiny, without microsculpture, carinate on median third 3. phasti Finnamore, new species 1. Aykhustigmus fritzi Finnamore, new species Derivation of Name. — The species is named in honour of Manfredo A. Fritz, collector of most of the specimens I have seen of the genus. Diagnosis. — The hind wing media di- verging after cu-a, swellings on the upper frons, and microsculpture on the anterior scutum will distinguish this species from all others in the genus. Description. — Male. Length 4 mm. Head. Flagellomeres without tyli. Clyp- eus and subantennal area obscured by dense appressed setae that extend up in- ner orbits to a point about % length of scape. Eye margined by a coarse carina that is interrupted ventrally near base of mandible. Carina foveolate, except on ver- tex and separated from outer orbit by a distance subequal to basal width of man- dible. Frons microsculptured and impunc- tate on lower face; shiny, with evanescent microsculpture on upper surface to lateral ocellus where it is sparsely punctate with punctures at least 3 diameters apart. Up- per frons with a swelling on each side of frontal line that is just visible in lateral profile. Vertex microsculptured, with a circular depression on midline posterior to lateral ocelli. Posterior vertex with an ev- anescent transverse ridge (posterior to cir- cular depression) connecting outer orbital carinae. Ocelli closer to each other than to Volume 4, 1995 229 eyes. OOD 1.8 x LOD. Gena microsculp- tured, punctures evanescent; irregular ev- anescent carinae radiate from occipital ca- rina on lower genal area. Occipital carina considerably raised, coarsely foveolate to mid ventral line. Mesosoma. Pronotum with transverse carina greatly raised, toothed at lateral end; carina continued on side to antero- ventral corner which is produced anteri- orly as a tooth that surpasses dorsolateral tooth in size. Pronotum with transverse sulcus longitudinally carinate; setae sparse, not obscuring sculpture. Pronotal side longitudinally carinate. Scutum mi- crosculptured anteriorly, shiny evanescent microsculpture posteriorly. Scutum be- tween notaular grooves, striatopunctate, elsewhere sparsely punctate. Scutellum shiny anteriorly, microsculptured posteri- orly, with a foveolate transverse sulcus on lateral disc. Mesopleuron coarsely areolate from subalar fossa to sternopleural region. Scrobal sulcus indicated by slight depres- sion. Hypersternaulus and omaulus coarsely foveolate. Mesosternum micro- sculptured, punctate next to areas of are- olate sculpture on sternopleural region. Metapleuron microsculpture evanescent with a few irregular longitudinal carinae. Hind wing media diverging after cu-a. Propodeum coarsely areolate, except area next to metapleuron which is weakly car- inate. Metasoma. Terga shiny, without mi- crosculpture, impunctate. Sterna shiny, without microsculpture, with several punctures on lateral margins. Color. Ground color black. White: pro- notal lobe. Yellow-brown: palpi, mandible except apex, antenna, tegula, wing veins, stigma, legs. Female. Length 3.5 mm. Similar to male except as follows: clypeus with 2 median teeth separated by a deep semicircular emargination; setose, although less so than male; setae partially obscuring densely punctate surface. Subantennal area and inner orbits sparsely setose. Inner orbits margined by an evanescent, non- foveolate carina. Eye elsewhere not mar- gined by carina. Frons between swellings and lateral ocelli, shiny, weakly micro- sculptured with coarse sparse punctures (3 or more diameters apart). Vertex with- out circular depression or transverse ridge but with frontal line continued as a sulcus on midline posterior to ocelli. OOD 1.7 X LOD. Gena with coarse sparse punctures. Occipital carina simple, not raised or fo- veolate. Pronotum with teeth less promi- nent than in male. Metasomal terga and sterna shiny with sparse punctures; ster- num VI densely punctate. Pygidial plate narrow, shiny, bisected by a longitudinal ridge, with several coarse punctures. Material Examined {16, 2 9). — Holotype: 6, PARAGUAY: S. Bernardino, Mitte Decbr., Fiebrig (NMW). Paratypes: BO- LIVIA: Km 335 de Cochabamba a Santa Cruz, Fritz -Martinez Col. (9: 1 IIES). BRAZIL: Estado do Ceara: Ceara, Baturite 17-6-1908, Ducke (9:1 MPEG). Distribution (Map 14). — Brazil: Ceara. Bolivia, Paraguay. 2. Aykhustigmus patanawi Finnamore, new species Derivation of Name. — The species epithet is derived from two Quechua terms, "pata" meaning margin and "nawi" meaning eye, in reference to the margined eye of this species. Diagnosis. — Hind wing media diverging before cu-a and vertex with transverse car- inate sculpture. Description. — Male. Length 4.2 mm. Head. Flagellomeres without tyli. Clyp- eus and subantennal area obscured by dense appressed setae that extend narrow- ly up inner orbit to a point about IV3 height of scape. Eye margined by a coarse foveolate carina that is interrupted ven- trally at base of mandible. Carina separat- ed from mid point of outer orbit by a dis- tance subequal to basal width of mandi- ble. Frons with frontal surface impunctate, microsculptured; irregular transverse ca- 230 Journal of Hymenoptera Research Figs. 18-23. Aykhustigmus patanawi 6. 18, head, frontal view. 19, head and prothorax, dorsal view. 20, mi- cropore field between lateral ocellus and compound eye. 21, head, lateral view. 22, mesosoma, lateral view. 23, mesosoma, dorsal view. rinae laterally next to inner orbital carina. Upper frons with coarse more or less ir- regular striatopunctate sculpture radiating from mid ocellus, punctures less than 1 di- ameter apart; small swelling on either side of mid line just anterior to mid ocellus. Vertex densely microsculptured with large, median, circular depression poste- rior to lateral ocelli; several short carinae radiating from lateral ocellus. Posterior vertex with a series of 3 or 4 transverse, striatopunctate ridges extending between outer orbits posterior to orbital carinae. OOD 0.9-1.3 X LOD. Gena entirely cari- nate or lower area areolate, sculpture in either case continuous from occipital ca- Volume 4, 1995 231 rina to hypostomal carina and outer orbit- notch; setae dense, obscuring underlying al margin. Occipital carina considerably sculpture. Setae less dense on subantennal raised, lamella-like, coarsely foveolate to area and inner orbit. Inner orbit margined mid ventral line. by a non-foveolate carina that extends to Mesosoma. Pronotum with transverse vertex. Outer orbit without carina. Frons carina considerably raised, toothed at lat- with lower area impunctate, densely mi- eral end; carina continued ventrally on crosculptured; upper area with evanescent propodeal side to anteroventral corner microsculpture, shiny, obscurely striato- which is produced anteriorly as tooth sub- punctate; carinae, if present, radiating equal in size to dorsolateral tooth. Prono- from mid ocellus anteriorly. Frons with turn with transverse sulcus longitudinally small swelling on either side of mid line, carinate; setae sparse, not obscuring sculp- Vertex with reduced circular depression ture. Pronotal side with irregular longitu- posterior to ocelli (not evident in 1 speci- dinal carinae. Scutum microsculptured an- men). Vertex densely microsculptured, teriorly and in depressions medially and with a series of irregular, transverse fine posteriorly; striatopunctate, punctures carinae between and posterior to outer or- coarse, variable from relatively sparse, bits. Gena carinate and punctate near about 1 diameter apart on disc to nearly mandible, elsewhere microsculptured. contiguous; notaular groove deeply im- OOD 1.9-2.1 X LOD. Occipital carina sim- pressed, foveolate, continued over half pie, complete, not raised or foveolate. length of scutum. Scutellum microsculp- Punctation of metasoma as in male except tured, with foveolate transverse sulcus on sternum VI densely punctate. Pygidial mid lateral disc. Mesopleuron coarsely plate narrow, shiny, bearing a lateral row areolate from subalar fossa to sternopleu- of punctures. ral region. Scrobal sulcus indicated by Material Examined {16, 3 9). — Holotype: slight depression. Hypersternaulus and 6, ARGENTINA: B. Aires, La Plata (Fac. omaulus coarsely foveolate. Mesosternum Agronomia) X-XI-68 C. Porter (MCZ). microsculptured, sparsely punctate. Hind Paratypes: ARGENTINA : Buenos Aires: wing media diverging before cu-a. Meta- Moreno, Fritz (9:1 IIES). Moreno, 12-73 pleuron microsculptured, few short cari- Fritz (9:1 IIES). Tigre XI-39 Viana Leg. nae on posterior margin. Propodeum mi- (9:1 IIES). crosculptured and coarsely areolate, ex- Distribution Map 15. — Argentina: Bue- cept for side next to metapleuron which is nos Aires, without areolae and partially without mi- crosculpture. ^- Aykhustigmus phasti Finnamore, Metasoma. Terga shiny with slight oily ^^^ species sheen, appearing impunctate on anterior Derivation of Name. — Phasti is a Quech- terga (punctures minute, sparse, evanes- ua term meaning naked or bare in refer- cent) but punctures increasing in size and ence to the lack of microsculpture on the density on posterior terga. Sterna shiny scutum and the resulting naked appear- with slight oily sheen, punctures sparse ance. throughout. Diagnosis. — Male unknown. Female: the Color. Ground color back. White: pro- absence of microsculpture on the scutum notal lobe. Yellow-brown to brown: palpi, will distinguish this species from others in mandible except apex, antenna, legs, teg- the genus. In addition the hind wing me- ula, wing veins, stigma. dia diverges after cu-a, the vertex is with- Female. Length 4.5 mm. Similar to male out a transverse carinae, and the clypeal except as follows: clypeus with 2 median apex has 2 teeth, teeth separated by a deep semicircular Descripition. — Female. Length 4 mm. 232 Journal of Hymenoptera Research Head. Clypeus with 2 teeth on apical sparse, evanescent on anterior terga, more margin separated by a deep U-shaped apparent on posterior terga. Pygidial plate notch; surface partially obscured by dense with slight raised median basal ridge, appressed setae which extend up suban- Sterna shiny, sparsely punctate except tennal area and narrowly up inner orbit to sternum VI which is densely punctate, about % height of scape. Inner orbit with Color. Ground color black. White: man- non-foveolate carina that is evanescent to- dible on basal third, pronotal lobe. Yel- ward vertex. Outer orbit not margined, low-brown to brown: palpi, mandible ex- Frons microsculptured on scapal basin, cept apex and basal third, antenna, legs, impunctate; upper frons more or less tegula, wing veins, stigma, shiny with evanescent microsculpture. Material Examined (39). — Holotype: 9, punctures sparse. Vertex sometimes with ARGENTINA: San Isidro 1-1976 B.A. poorly defined depression posterior to lat- (PMA). Paratypes: ARGENTINA : Entre eral ocelli on mid line or depression ab- Rios: Palmar Colon 1-74 Fritz (9:1 IIES). sent; densely microsculptured in contrast VENEZUELA : Zulia: 6km w. La concep- to upper frons. OOD 1.8-1.9 X LOD. Gena cion, 18-VI-1976 A.S. Menke & D. Vincent microsculptured, punctate toward man- (9:1 USNM). dibular base. Occipital carina simple, com- Distribution (Map 16). — Argentina, Ven- plete, not raised or foveolate. ezuela. The Venezuelan record represents Mesosoma. Pronotum with transverse the only collection locality north of the carina considerably raised, lateral end Brazilian Highlands for the genus and forming right angle that continues ven- could represent an introduction of this trally on carinate pronotal side to antero- species to that area, ventral corner which is produced as a small tooth. Pronotum with transverse 4. Aykhustigmus warawa Finnamore, sulcus longitudinally carinate. Scutum ^^^ species shiny, without microsculpture except Derivation of Name. — The species epithet, weakly at extreme anterolateral corner, warawa, is a Quechua term meaning median third (between notaular grooves) adorned in reference to the elaborate striatopunctate; lateral area punctate, usu- sculpture of this species, ally more densely punctate next to notau- Diagnosis. — The combination of the hind lar groove. Scutellum shiny or weakly mi- wing media diverging after cu-a; and crosculptured anteriorly, microsculptured frons without swellings, smoothly round- posteriorly; a foveolate, transverse sulcus ed in both sexes; and the female clypeal on mid lateral disc; punctures few, present edge with 4 teeth, and scutum microsculp- on anterolateral disc. Mesopleuron coarse- tured on anterior V4 or more will separate ly areolate from subalar fossa to sterno- this species from all others in the genus, pleural region. Scrobal sulcus indicated by Description. — Male. Length 3.0-4.5 mm. a slight depression. Hypersternaulus and Head. Flagellomeres without tyli. Clyp- omaulus coarsely foveolate. Mesosternum eus and subantennal area obscured by microsculptured, sparsely punctate. Meta- dense appressed setae that extend up in- pleuron weakly microsculptured, a few ner orbit about ll^ length of scape. Inner short carinae on posterior margin. Hind orbit margined by non-foveolate carina wing media diverging after cu-a. Propo- extending % up frons. Outer orbit mar- deum with microsculpture evanescent or gined from base of mandible to vertex by absent over most of its surface; coarsely foveolate carina or evanescent non-foveo- areolate, except next to metapleuron late ridge in smaller specimens; at mid which is irregularly carinate. point the carina is separated from orbit by Metasoma. Terga shiny; punctures distance subequal to basal width of man- Volume 4, 1995 233 dible. Lower frons microsculptured; upper Color. Ground color black. White: pron- frons shiny, weakly microsculptured, otal lobe. Yellow-brown to brown: palpi, punctate. Vertex microsculptured with mandibles except apically, antenna, legs, puncture-like depression on mid line pos- tegula, wing veins, stigma, terior to lateral ocelli; transverse evanes- Female. Length 4.0-4.5 mm. Similar to cent ridge connects laterally with outer or- male except as follows: clypeus with 4 bital carina. Lateral ocelli closer to each teeth on apical margin, lateral teeth locat- other than to eye. OOD 1.3-1.4 X LOD. ed below inner antennal socket margin Gena microsculptured, punctate near base and often obscured by overhanging clyp- of mandible. Occipital carina complete, eal setae. Surface of clypeus and subanten- foveolate, not raised. nal area obscured by dense appressed se- Mesosoma. Pronotum with transverse tae which extend up inner orbit 3/4 length carina raised, lateral end with small tooth; of scape. Outer orbit unmargined. Lower carina continued on pronotal side to an- ggna striatopunctate. OOD 1.4-1.6 X LOD. teroventral corner which is produced as Occipital carina simple, complete, not tooth surpassing in size that of dorsolat- raised or foveolate. Pygidial plate narrow, eral tooth. Pronotum with transverse sul- shiny, sparsely punctate. Sternum VI cus longitudinally carinate; setae sparse, densely punctate. not obscuring sculpture. Pronotal side Material Examined (13(?, 13 9).— Holo- with irregular, longitudinal carinae. Scu- ^yp^. ^^ ^rq [ARGENTINA]: Salta: Ro- tum with microsculpture variable from ^^^^^ ^^^j Lerma 10-84 Fritz (IIES). Para- densely microsculptured to microsculp- ^^^^^. ARGENTINA: Buenos Aires: 48- tured on anterior margin only; punctures 889 (9: 1 MACN). Burzaco Sea L. 12-13- present on median third (between notau- iji.2974 c R. Vardy B.M. 1974-204 (9:1 lar grooves which extend posteriorly as pMNH). Moreno, M.A. Fritz (d: 1 IIES). somewhat polished low ridges); lateral ar- Moreno 1-73, Fritz (9:2 IIES). La Plata eas impunctate to sparsely punctate; pos- ^^^^ Agronomia) X-XI-68, C. Porter {6:1 terior margin and often median area stria- j^^^). Jujuy: Palpala 11-1968, M. Aczel topunctate. Scutellum shmy, weakly mi- ^^^ ^ ^^^^ ^^^^^, p^^.^^^ ^2-56, Fritz {6: crosculptured anteriorly, punctured; pos- ^ ^^^^^.^ ^^^^ P^^^^ ^^_^^^ P^.^^ terior microsculptured, jn^P^^nctate; (^, 4 HES). Urundel 26-1-950, R. Golbach without foveolate sulcus on disc laterally. p^pp^ ^^ ^^^ ^^^^^_ ^^^^ ^_ Mesopleuron coarsely areolate from sub- 1^ ^, . .n 1 ?■,T-c-^ t t , / , , 1 1 • c u 1 74, Martinez (9:1 IIES). Tucuman: Las alar fossa to sternopleural region. Scrobal ' v.t c^ r^ 1 ur / j^ 1 ttttcx c sulcus indicated by slight depression. ^^S^^^^^/J'^^l.^f^f ^" ^f /i "^^ \^ Hypersternaulus and omaulus coarsely Janvier 21-X-1950, M. Aczel (d: 1 LILLO). foveolate. Mesosternum microsculptured, ^an Miguel, Tucuman 18-XI-1979, L.A. sparsely punctate. Metapleuron shiny, Stange (d: 1 FSCA). San Pedro de Cola- without or with evanescent microsculp- 1^«' Foerster (9:1 IIES). Trancas, Fritz {6: ture, several short carinae on posterior 1 "ES, 9: 1 IIES). BOLIVIA: Beni: Ichilo margin. Hind wing media diverging after Yapacani 2-71, Fritz ((5:1 IIES, 9: 1 IIES). cu-a. Propodeum microsculptured, BRAZIL: Santa Catarina: Nova Teutonia coarsely areolate, except for side next to Feb. 1957, F. Plaumann (9:2 OSU). Sao metapleuron which is shiny, unsculp- Paulo: Campinas Mar. 1924, F.X. Wil- tuj-ed. Hams {6:1 BISH). Unplaced: Minas, Ja- Metasoma. Terga shiny with slight oily care Dec. 1-1907 Carn. Mus. Ace. 3533 ( 9 : sheen, punctures sparse and evanescent. 1 CARN). Sterna similar to terga except punctures Distribution (Map 17).— Argentina, Bo- more dense. livia, southern Brazil. 234 Journal of Hymenoptera Research Figs. 24-29. Aykhustigmus warawa 9 . 24, head, frontal view. 25, head and prothorax, dorsal view. 26, micro- pore field between lateral ocellus and compound eye. 27, head, lateral view. 28, mesosoma, lateral view. 29, mesosoma, dorsal view. INCASTIGMUS Finnamore, new genus (Figs. 30-35) Derivation of Name. — Incastigmus is named after the indigenous Inca people of South America. Diagnosis. — The presence of a median groove or posteromedian pit on the scu- tum will separate this genus from all oth- ers in the Stigmina. Description. — Head. Labrum quadri- lobed with narrow, deep median emargi- nation. Mandible with apex in both sexes tridentate; apicoventral tooth in female acute; inner basal tooth absent. Clypeal apex in male without bevelled modifica- tions; with 4 teeth in female. Interantennal tubercle absent. Frontal line absent. Micro- pore field present. Inner orbits converging I Volume 4, 1995 235 below. Eye at most partially margined by a carina. Occipital carina complete, not joining hypostomal carina, simple in fe- male; raised, foveolate in male. Mesosoma. Scutum with a median groove or at least a posteromedian pit; no- taular groove usually elongate. Acetabular carina intercepting omaulus. Scrobal sul- cus present. Hypoepimeral area without coarse sculpture. Mid basitarsus of male elongate, as long as next 3 tarsomeres combined. Posterior margin of hind tibia with 2 or 3 spines. Fore wing without se- tae in cellular area. Hind wing media di- verging before cu-a; submedian cell of normal size, not reduced. Metasoma. Petiole carinate. Pygidial plate narrow, absent in male. Digitus elon- gate relative to cuspis, clubbed. Type Species. — Incastigmus inti new spe- cies. Distribution. — Neotropical. Species Transferred. — Stigmus hexagonalis Fox, S. neotropiciis Kohl, S. smithii Ash- mead, and S. thorncicus Ashmead. The 25 species of Incastigmus are the subject of the forthcoming Part 2 of this revision. Incastigmus inti Finnamore, new species Derivation of Name. — Inti is a Quechua term meaning sun in reference to the cen- tral position of this species as type for the genus. Diagnosis. — The combination of the fol- lowing characters will separate this spe- cies from all others in the genus. Micro- pore field on vertex oval. Scutum with median and notaular grooves complete from anterior to posterior margin, without regular longitudinal carinae between grooves. Pronotal lobe white, rounded conical, not flattened or toothed or dark. Male antenna with tylus on apical flagel- lomere. Female clypeus with 2 elongate setae issuing from 2 narrowly separated pits on median lobe. Female frons shiny, without microsculpture anterolaterally to mid ocellus. Description. — Male. Length 2.5-4.0 mm. Head. Flagellomeres without special- ized setae; II-XI with tyli, that on flagel- lomere XI imparting asymmetrical shape; basal flagelk^mere length 2.2 X apical width; penultimate flagellomere length 1.4 X apical width. Head microsculptured on lower frons, mid frons and vertex but shiny on upper lateral frons, punctures sparse. Clypeus obscured by dense ap- pressed setae which extend up inner orbits little more than height of antennal socket. Micropore field present as discrete patch. Ocelli closer to each tether than to com- pound eyes. OOD 1.7 X LOD. Gena mi- crosculptured, sparsely punctate, non-car- inate, without ventral tooth or swelling. Occipital carina foveolate. Mesosoma. Transverse carina of prono- tum toothed laterally, continued ventrally to pronotal side and terminating in a ven- trolateral tooth; transverse sulcus longitu- dinally carinate. Pronotal side carinate, lobe conical with rounded apex. Scutum entirely microsculptured to weakly mi- crosculptured with shiny patches; notau- lar groove complete to posterior margin; median groove complete to anterior mar- gin; punctures sparse. Scutellum micro- sculptured. Mesopleuron variable from entirely microsculptured to weakly shiny with reduced microsculpture, sparsely punctate to impunctate. Preomaular area with sparse setae that do not obscure un- derlying sculpture. Hypersternaulus, omaulus and scrobal sulcus foveolate. Me- tapleuron usually microsculptured. Pro- podeum shiny, generally without micro- sculpture, uniformly coarsely areolate ex- cept anterolaterally without sculpture. Metasoma. Terga shiny, microsculpture and punctures evanescent (appearing im- punctate). Sterna weakly microsculptured, punctures sparse. Color. Ground color black. White: man- dible medially, pronotal lobe. Yellow- brown: palpi, mandible subapically, an- tenna, fore leg except coxa, mid leg except coxae, hind trochanter and tarsus, tegula, metasomal sternum VIII. 236 Journal of Hymenoptera Research Figs. 30-35. Incastigmiis inti. 30, 9 head, frontal view. 31, S head and prothorax, dorsal view. 32, 6 micropore field between lateral ocellus and compound eye. 33, i mesosoma, dorsal view. 34, 9 mesosoma, lateral view. 35, 9 mesosoma, dorsal view. Female. Length 3.0-4.25 mm. Similar to male except as follows: antenna without tyli, penultimate flagellomere length 1.25 X apical width. Clypeus shiny, usually sparsely punctate, occasionally moderate- ly dense punctation grouped in median area; median lobe with 2 elongate setae is- suing from 2 narrowly separated pits; apex with truncate median lobe. Scutum with several irregular carinae between no- taular and median grooves. Color as above but variable to dark antennal fla- gellum and dark femora. Material Examined (258c5 109 9).— Holo- type: S, ECUADOR: Napo Prov. Tena, sweep 15-11-1986 A.T. Finnamore (PMA). Paratypes: BOLIVIA: La Paz: Chulumani 1,700m 25-III-1979 M. Cooper B.M. 1979- 216 (9: 2 BMNH). Chulumani 1,700m 30- III-1979 M. Cooper B.M. 1979-216 (9:1 Volume 4, 1995 237 BMNH). Chulumani 1,700m 31-III-1979 M. Cooper B.M. 1979-216 {6: 1 BMNH). Chu- lumani 1,700m 3-IV-1979 M. Cooper B.M. 1979-216 {6:1 BMNH. 9: 2 BMNH). Co- roico-Chulomani III-7-1968 Garcia & Por- ter (9: 1 MCZ). Coroico, Sta. Barbara 1100m 4-1-1975 (9: 1 PMA). Tumupasa Dec. W.M. Mann, Mulford Biol. Expd. 1921-1922 {6:1 USNM). Yungas, 13km s. Caranavi 850m 27-III-1973 J. Helava (9:2 PMA). Unplaced: Las Juntas Dec. 1913, Steinbach (9: 1 CARN). BRAZIL: Mato Grosso: Itaum 111-1974 M. Alvarenga (9: 3 AEl). Para: Belem IPEAN Xll-1-4-1969 J.M. & B.A. Campbell (9: 1 BRD). CO- LOMBIA: Caqueta: Florencia 480m 31-X- 1971 M. Cooper B.M. 1972-275 (9: 1 BMNH). Yuruyaco, 73k sw. Florencia 3-II- 1979 M. Cooper B.M. 1979-106 (9: 1 BMNH). Yuruyaco, 73k sw. Florencia 12- 11-1979 M. Cooper B.M. 1979-106 {6: I BMNH). Yuruyaco, 73k sw. Florencia 13- 11-1979 M. Cooper B.M. 1979-106 {6: 1 BMNH). Putumayo: Mocoa 13-IV-1974 M. Cooper B.M. 1974-327 {6:1 BMNH). Mo- coa 16-1V-1974 M. Cooper B.M. 1974-327 (9:5 BMNH). Mocoa 18-IV-1974 M. Coo- per B.M. 1974-327 {6:1 BMNH). Mocoa 19-rV-1974 M. Cooper B.M. 1974-327 {6:1 BMNH). Mocoa 14-VI-1974 M. Cooper B.M. 1974-548 {6:1 BMNH). Mocoa 24- Vl-1974 M. Cooper B.M. 1974-548 {6: 2 BMNH). Mocoa 26-V1-1974 M. Cooper B.M. 1974-548 (9:2 BMNH). Mocoa 16- VII-1978 M. Cooper B.M. 1978-431 (9:1 BMNH). Mocoa 7-IX-1974 M. Cooper B.M. 1975-33 {6: 1 BMNH). Mocoa 27-IX-1974 M. Cooper B.M. 1975-33 {6: 1 BMNH). Mocoa 600m 26-111-1976 M. Cooper B.M. 1976-290 {6:1 BMNH). Mocoa 18-111-1978 M. Cooper B.M. 1978^31 {6: 1 BMNH). Villa Garzon, 8 mi s. Mocoa 3-V1II-1978 M. Cooper B.M. 1978^31 {6:1 BMNH). Vau- pes: Mitu ll-V-1974 M. Cooper B.M. 1974-327 {6:1 BMNH). ECUADOR: Car- chi: Chical 1250m 0-56N 78-1 IW, 1-VlII- 1983 J. Rawlings, M. Smyers (9:1 CARN). Napo: Coca, May 1965 Luis Pefia {6: 1 AEl). Coca & Napo Rivers, V-1-12-65, Luis Pefia {6:1 AEl). Limoncocha 250m, 15-28- VI-1976 S. & J. Peck {6: 2 BRD, 9: 2 BRD). Misahualli, down Rio Napo 19-11- 1983 L. Huggert {6:15 PMA, 9: 2 PMA). Muyuna, 5km w. Tena 29-IX-1978 M. Coo- per B.M. 1979-20 (9: 1 BMNH). Puerto Misahualli 350m 11-1983 Sharkey {6: 2 PMA, 9: 3 PMA). Puerto Misahualli, 30km e. 350m 11-1983 Sharkey {6: 9 PMA). Santa Cecilia III-25-31-1969 P. & P. Span- gler ((5:1 USNM). Tena 9-14-II-1971 M. Cooper B.M. 1972-275 {6:1 BMNH). Tena 8-VII-1976 S. & J. Peck (9:3 PMA). Tena 400m 11-1983 M. Sharkey {6:6 PMA, 9: 1 PMA). Tena, sweep 15-11-1986 A.T. Fin- namore {6: 12 PMA, 9: 3 PMA). Tena, sweep 18-11-1986 A.T. Finnamore (9: 1 PMA). Tena, 12km sw. 500m 8-11-VII- 1976 S. & J. Peck (9:1 BRD). Tena-Puyo Hwy. 5km n. Santa Clara 21-11-1986 sweep, T. Thormin & J Wojcicki {6: 1 PMA). Pastaza: Puyo 960m 1-8-X-1970 ]. & M. Sedlacek {6: 8 BISH). Puyo 900- 960m 1-8-X-1970 J. & M. Sedlacek {6:7 BISH). Puyo 22km sw. 900m 14-16-VII- 1976 S. & J. Peck, forest {6: 1 BRD, 9: 2 BRD). Puyo 23km se 19-V-1977 P.J. Span- gler & D.R. Givens #58 (9: 1 USNM). Puyo 44km s. 21 May 1977 DL & SS Vin- cent {6: 5 USNM, 9: 2 USNM). Puyo 18km n. 1100m 14-8-1982 R. Hensen & \. Aptroot {6:1 RNH). Pichincha: Tinalan- dia 800m 111-1983 L. Masner & M. Sharkey {6:2 PMA). Zamora: Rio Jumboe 1-IV- 1965 Pefia {6:5 MCZ, 9: 2 MCZ). Zamora lV-4-1965 Pena {6: 1 AEl). Unplaced: Cumbaratza (E) Xl-21-1970 Luis E. Peha {6: 1 AEl). Mera 26-1-1923 F.X. Williams {6:1 BISH). PARAGUAY: Guaira: w. Vil- larica, Caballero, 1-72 Pena {6:1 IlES). Un- placed: Piareta 12-71 Pena {6: 1 IIES). PERU: Cuzco: Agua Calliente 21- 28-X1I-1983 L. Huggert {6: 8 PMA, 9: IPMA). Machu Picchu l-XII-1965 H. & M. Townes (9:1 AEl). Qunicemil, 750m nr. Marcapata September 1962 Luis Pena {6: 8 AEl, 9: 3 AEl). Huanuco: Cayumba, 35km s. Tingo Maria 800m l-XI-1973 J.M. Schunke B.M. 1974-37 (9:1 BMNH). Las 238 Journal of Hymenoptera Research Palmas 5km sw. 1000m X-16-1954 E.I. Schlinger & E.S. Ross {6:2 CAS). Las Pal- mas lOmi sw. 1000m IX-26-1954 E.I. Schlinger & E.S. Ross (9:4 CAS). Monson Valley, Tingo Maria IX-21-1954 E.I. Schlinger & E.S. Ross ((5:1 CAS). Monson Valley, Tingo Maria 23-IX-1954 E.I. Schlin- ger & E.S. Ross (c?: 6 CAS, 9: 3 CAS). Monson Valley, Tingo Maria X-10-1954 E.I. Schlinger & E.S. Ross {6:4: CAS, 9: 3 CAS). Monson Valley, Tingo Maria X-19- 1954 E.I. Schlinger & E.S. Ross (9:1 CAS). Monson Valley, Tingo Maria X-21-1954 E.I. Schlinger & E.S. Ross {6:1 CAS, 9: 1 CAS). Monson Valley, Tingo Maria X-26- 1954 E.I. Schlinger & E.S. Ross (9: 1 CAS). Monson Valley, Tingo Maria XI-21-1954 E.I. Schlinger & E.S. Ross {6: 1 CAS). Monson Valley, Tingo Maria XI-29-1954 E.I. Schlinger & E.S. Ross {6:1 CAS, 9: 1 CAS). Monson Valley, Tingo Maria XII-2- 1954 E.I. Schlinger & E.S. Ross (9:1 CAS). Tingo Maria 26-29-1-1984 L. Huggert {6: 14 PMA, 9: 3 PMA). Tingo Maria 30-1- 1984 L. Huggert {6: 6 PMA, 9: 1 PMA). Tingo Maria, 1km e. 15-VIII-1971 P.S. & H.L. Broomfield. MT dense woodland B.M. 1971-486 (9:1 BMNH). Tingo Maria, 26 mi. e. XII-10-54 1100m E.I. Schlinger & E.S. Ross (9:1 CAS). Tingo Maria, 67 mi. e. X-4-54 350m E.I. Schlinger & E.S. Ross (9:1 CAS). Tocache 2-II-1984 L. Huggert {6: 4 PMA, 9: 1 PMA). Junin: Colonia Perene, Rio Perene 18 mi ne La Merced I- 3-55 E.I. Schlinger & E.S. Ross {6:2 CAS, 9: 1 CAS). Paratuchali 22-1-1984 L. Hug- gert {6:10 PMA, 9: 4 PMA). Satipo 18-1- 1984 L. Huggert {6:10 PMA, 9: 1 PMA). Satipo 19-24-1-1984 L. Huggert {6: 11 PMA, 9: 6 PMA). Lima: Magdalena [del Mar?] [Lima?] Mar 28-10 CHT Townsend (9:1 USNM). Loreto: Iquitos, ne. Rio Na- nay 6-n-1984 L. Huggert {6: 10 PMA). Iquitos, Gransa UNAP 9-II-1984 L. Hug- gert {6:2 PMA). Iquitos, Quisto Cocha 5- 11-1984 L. Huggert {6: 1 PMA). Iquitos, Barilla 10-11-1984 L. Huggert {6:1 PMA). Madre de Dios: Laberinto, 70km w. Pto. Maldonado on Rio Madre de Dios 1-2-1- 1984 A.T. Finnamore (9: 1 PMA). Pto. Maldonado 1-11-1-1984 L. Huggert {6:50 PMA, 9: 14 PMA). Tambopata Reserve, 50km s. Pto. Maldonado on Rio Tambo- pata 3-8-1-1984 A.T. Finnamore {6: 1 PMA, 9: 1 PMA). Ucayali: Tacshitea, 88km n. Pucallpa jet. Rio Callaria & Rio Ucayali 22-25-1-1984 A.T. Finnamore {6:1 PMA). VENEZUELA: Zulia: El Tucuco, 45km sw. Machiques, 5-6-VI-1976 A.S. Menke & D. Vincent (9:1 USNM). Distribution (Map 18). — Bolivia, Brazil, Ecuador, Paraguay, Peru, Venezuela. Incastigmus thoracicus (Ashmead), new combination Stigmus thoracicus Ashmead 1900:223 9. Holo- type*: 9, St. Vincent W.I., H. Smith 238/ W. Indies 99-331 (BMNH type Hym 21.885). Stigmus smithii Ashmead 1900:223 2 . New syn- onymy. Holotype*: S, W. Indies (BMNH type Hym 21.886). 99-331. Ashmead incor- rectly sexed the type specimen when he de- scribed sniitliii based on a single female. The holotype is a male. Diagnosis. — The red petiole is sufficient to distinguish this species in lighter col- ored specimens. Darker (and lighter) spec- imens are distinguished by the extreme re- duction of the median scutal groove, shiny hypoepimeral area, and rounded corners of the transverse pronotal carina. Description. — Male. Length 3.0-3.8 mm. Head. Flagellomeres without tyli or spe- cialized setae; basal flagellomere length 2.0 X apical width; penultimate flagello- mere length 1.1 X apical width. Head al- most uniformly microsculptured; upper frons and vertex with slightly less micro- sculpture and with sparse obscure punc- tures. Clypeus obscured by dense ap- pressed setae which extend broadly up in- ner orbits to height of scape. Micropore field present as discrete patch. Ocelli clos- er to each other than to eyes. OOD 2.0 X LOD. Gena microsculptured, sparsely punctate, non-carinate, without ventral swelling or tooth. Occipital carina foveo- late. Volume 4, 1995 239 Mesosoma. Pronotum with transverse Female. Length 3.9 mm. Similar to male carina rounded laterally, not toothed or except as follows: basal flagellomere produced; transverse sulcus with evanes- length 3.0 X apical width. Clypeus shiny, cent longitudinal carinae; pronotal lobe with several punctures, setae sparse; clyp- normal, not toothed; side with several ev- eal apex with median teeth separated by anescent carinae. Scutum shiny, weakly shallow emargination; median lobe with 2 microsculptured; notaular groove present elongate setae issuing from 2 narrowly anteriorly, not elongate; median groove separated pits. Inner orbits sparsely se- absent, pit evident posteriorly as one of tose, not obscuring sculpture. OOD 2.5 X many evanescent fovea in transverse row LOD. Color, as above for light form except next to posterior margin; punctures min- white on scape and orange on clypeus and ute, sparse. Scutellum and mesopleuron propodeal dorsum. shiny, without microsculpture; impunc- Material Examined {66, 59). — DOMIN- tate except sternopleural region with min- ICA: Pont Casse; Springfield; Springfield ute sparse punctures; preomaular area Plantation. GRENADA: 2500 feet. Baltha- with sparse setae; hypersternaulus with- zar, windward side; Botanical Garden. ST. out foveae; scrobal sulcus and omaulus VINCENT: (no other data), weakly foveolate. Metapleuron micros- Distribution (Map 19). — Dominica, Gre- culptured on ventral half, otherwise shiny, nada, St. Vincent, impunctate. Propodeum shiny, without microsculpture over most of basolateral LLAQHASTIGMUS Finnamore, area and dorsolateral spheres; basolateral new genus area adjacent to metapleuron without ^ ^§^' ~ -' sculpture; dorsal enclosure areolae eva- Derivation of Generic Name. — Llaqha is a nescent, areolae small relative to larger ev- Quechua term meaning dark, in reference anescent areolae of dorsolateral spheres, to the generally black coloration of species the 2 groups of areolae separated by a in this genus, smooth unsculptured area. Diagnosis. — Females of this genus can be Metasoma. Terga shiny, without mi- separated from all others in the Stigmina crosculpture; punctures sparse, evanes- by the enlarged, truncate, apicoventral cent. Sterna weakly microsculptured, mandibular tooth. Females in other genera punctures sparse but increasing in density have an acute apicoventral tooth. Males on posterior metasoma. can be separated with a combination of a Color. Light form: ground color black, tridentate mandible and a scutum without White: palpi, mandible except apex, spot a trace of a posteromedian pit or groove, on pronotal lobe, fore tibia and tarsus. Males in other genera have a bidentate mid tibia and tarsus, hind tibia on basal mandible or at least a posteromedian pit half or more and tarsus. Yellow-orange: on the scutum. antenna, mesosoma except propodeal dor- Description. — Head. Labrum bilobed sum, fore leg basal to tibia, mid leg basal with a small median notch (labrum quad- to tibia, hind leg basal to tibia and apex of rilobed in 2 species both of which lack a tibia, petiole except ventral apex, apex of subomaulus). Mandible with apex triden- 1 or more sterna. Dark form: ground color tate; apicoventral tooth in female en- black. White: palpi, mandible except apex, larged, apex truncate; without inner basal spot on pronotal lobe, basal ring of hind tooth. Clypeal apex unmodified in male, tibia. Yellow-brown: antenna, pronotum without bevelled edge; in female with 4 except dorsally and spot on pronotal lobe, teeth. Interantennal tubercle absent. Fron- mesopleuron anteriorly, fore leg, mid leg, tal carina absent. Vertex with micropore hind trochanter and tarsus, tegula. field between lateral ocellus and com- 240 Journal of Hymenoptera Research pound eye. Inner orbits converging below. Eye not margined by carina. Occipital ca- rina complete, not joining hypostomal ca- rina, simple in female, foveolate in male, otherwise unmodified. Mesosoma. Scutum without trace of me- dian groove or posteromedian pit; notau- lar groove normal, not elongate. Acetab- ular carina intercepting omaulus. Scrobal sulcus often evanescent. Hypoepimeral area without coarse sculpture. Mid basi- tarsus of male elongate, as long or longer than next 3 tarsomeres combined. Poste- rior margin of hind tibia with 2 or 3 spines. Fore wing without setae in cellular area. Hind wing media diverging before cu-a; submedian cell of normal size, not reduced. Metasoma. Petiole carinate. Pygidial plate narrow, absent in male. Digitus elon- gate. Ti/pe Species. — Llaqhastig?nus jatunkirus Finnamore, new species. Distribution (Map 37). — Neotropical, re- stricted to South America (subtropical highland or montane forest?). KEY TO SPECIES OF LLAQHASTIGMUS 1 Male 2 1' Female 15 2 Gena ventrolaterally and sternopleural region with relatively dense, elongate setae; Peru 13. shachus Finnamore, new species 2' Gena and sternopleural region sparsely pubescent with short inconspicuous setae .... 3 3 Preomaular area densely setose, obscuring sculpture; mesosoma often extensively red; se. Brazil 8. mantanti Finnamore, new species 3' Preomaular area sparsely setose; mesosoma black except pronotal lobe which is usually white 4 4 Scape, fore femur, and pronotal lobe, black; mesosoma with scrobal sulcus foveolate; clypeus sparsely setose, underlying punctures visible; Ecuador, Colombia 3. chiitiyana Finnamore, new species 4' One or more of scape, fore femur, or pronotal lobe, yellow-brown or white; mesosoma with scrobal sulcus usually smooth, rarely foveolate; clypeus usually obscured by dense appressed setae 5 5 Subomaulus absent 6 5' Subomaulus present 7 6 Petiole with 3 or 4 coarse lateral carinae; Bolivia 2. australis Finnamore, new species 6' Petiole microcarinate laterally; Ecuador, Colombia 11. santanderamis Finnamore, new species 7 Scape brown with yellow-brown at base and apex only; flagellum and femora black; propodeum with area adjacent to enclosure finely carinate, lateral spheres without large areolae; transscutellar sulcus with foveae small or absent; scrobal sulcus not foveolate, only slightly impressed; Ecuador 5. ecuatorialis Finnamore, new species 7' Scape yellow-brown at least ventrally; one or more femora usually yellow to yellow- brown; propodeum with area adjacent to enclosure irregularly sculptured, lateral spheres coarsely areolate; transscutellar sulcus usually with prominent foveae; scrobal sulcus of- ten foveolate 8 8 Head and mesosoma coarsely, uniformly microsculptured, dull; microsculpture of frons not distinct from that of vertex; scutum with small evanescent punctures; Peru, Ecuador, Venezuela 9. tmithiis Finnamore, new species 8' Vertex, anterior to mid ocellus, more shiny than frons (less microsculpture) and/or me- sosoma with mid and lower mesopleuron shiny, without or with little microsculpture; punctation of scutum distinct or absent 9 9 Scutum shiny, without microsculpture on posterior %; scutellum punctures sparse, min- Volume 4, 1995 241 ute, appearing impunctate; scrobal sulcus coarsely foveolate; Ecuador, Colombia, Vene- zuela 14. sharkeyi Finnamore, new species 9' Posterior half of scutum microsculptured and /or punctate or scrobal sulcus smooth, with- out foveae 10 10 Pronotal lobe brown to black; posterior half of scutum shiny, punctures sparse; scrobal sulcus only slightly evident, not foveolate; Ecuador, Colombia 10. nigricollaris Finnamore, new species 10' Pronotal lobe white; posterior half of scutum punctate and /or microsculptured; scrobal sulcus usually foveolate 11 11 Transverse sulcus of pronotum without longitudinal carinae, sulcus smooth, usually with- out sculpture 12 ir Transverse sulcus of pronotum with longitudinal carinae 13 12 Propodeum with few carinae on lateral sphere, shiny toward metapleuron; petiole (mea- sured dorsally) longer than first tergum; Peru 7. Uiitanis Finnamore, new species 12' Propodeum with dense irregular carinae on lateral sphere and with microsculpture ad- jacent to metapleuron; petiole (measured dorsally) shorter than first tergum; Bolivia, Peru 12. sapanis Finnamore, new species 13 Occipital carina simple, not foveolate or raised ventrally; tarsi white; Ecuador 1. atnbigutis Finnamore, new species 13' Occipital carina at least finely foveolate and usually slightly raised ventrally; tarsi white to yellow brown 14 14 Scrobal sulcus coarsely foveolate; scutum coarsely but sparsely punctured in single known specimen; Colombia 4. colombiamis Finnamore, new species 14' Scrobal sulcus finely, if at all, foveolate; scutum finely to coarsely punctured; Venezuela to Bolivia 6. jatunkirus Finnamore, new species 15 Gena ventrally and sternopleural region with relatively dense, elongate setae; Peru 13. shachus Finnamore, new species 15' Gena ventrally and sternopleural region sparsely pubescent with short inconspicuous setae 16 16 Fore coxa and usually some of prothorax red; se. Brazil 8. mantanti Finnamore, new species 16' Prothorax and fore coxa black 17 17 Microsculpture of head and scutum uniform throughout; scutum with small obscure punctures; Peru, Ecuador, Venezuela 9. muthiis Finnamore, new species 17' Microsculpture of vertex not as dense as frons (vertex more shiny); scutum often shiny or with large punctures 18 18 Pronotal lobe dark, brown to black 19 18' Pronotal lobe white 20 19 Scrobal sulcus coarsely foveolate; dorsal mandibular tooth acute; scape black; Ecuador, Colombia 3. chutiyana Finnamore, new species 19' Scrobal sulcus finely foveolate; dorsal mandibular tooth broad; scape yellow; Ecuador, Colombia 10. nigricollaris Finnamore, new species 20 Propodeum closely and finely carinate on area adjacent to enclosure; Ecuador 5. ecuatorialis Finnamore, new species 20' Propodeum coarsely areolate on area adjacent to enclosure 21 21 Transverse sulcus of pronotum without longitudinal carinae, smooth, usually without sculpture 22 21 ' Transverse sulcus of pronotum with at least a few longitudinal carinae 23 22 Clypeus with elongate specialized setae on apical edge except for narrow median emar- gination; Peru 7. llutanis Finnamore, new species 22' Clypeus with elongate specialized setae on median and lateral teeth only; Peru, Bolivia 12. sapanis Finnamore, new species 23 Scutum shiny, without microsculpture on posterior %; punctures minute, sparse, appear- 242 Journal of Hymenoptera Research ing impunctate posteriorly; scrobal sulcus foveolate; Ecuador, Venezuela 14. sharkeyi Finnamore, new species 23' Scutum with microsculpture and/or punctures posteriorly; scrobal sulcus foveolate or not 24 24 Scrobal sulcus entirely foveolate; clypeal punctures minute, relatively dense and more or less equally distributed on raised area above median teeth; Colombia 4. colotnbiamis Finnamore, new species 24' Scrobal sulcus usually foveolate anteriorly or not at all, occasionally entirely foveolate; clypeal punctures sparse with relatively large more or less impunctate area above median teeth 25 25 Clypeal teeth black; Venezuela to Bolivia 6. jatiinkiriis Finnamore, new species 25' Clypeal teeth yellow-brown; Ecuador 1. ambiguus Finnamore, new species 1. Llaqhastigmus ambiguus Finnamore, ocellus. Occipital carina simple, not raised new^ species ventrally or foveolate. Derivation ofName.-The species epithet, M^sosonia. Transverse sulcus of pro- ambiguus, is from the Latin "ambiguus" "°^^"^ ^/^^ longitudinal carinae. Scutum meaning of doubtful nature, in reference nucrosculptured, with fine sparse punc- to the single known locality and lack of ^"^^' "^^'"^ *^^^ ^ diameters apart (some- material to adequately judge variation. ^'"^^^ mterspersed with more coarse, al- Diagnosis.-The simple occipital carina ^^^^S^ ^^^^^^y ^P^^^^ punctures) and that is not raised ventrally or foveolate concentrated on centre of disc. Transscu- will distinguish males of this species from *^"^^ ^^^^"^ foveolate. Preomaular area all others and females from most other sparsely setose, setae not obscuring un- species in the genus. Males of all other derlying sculpture. Subomaulus present, species and females of most species have Scrobal sulcus usually with irregular fo- an occipital carina at least slightly raised ^^ae on at least anterior third; foveae vari- ventrally bearing at least several foveae. ^^^^ f^om almost absent to present on en- Additionally, female fl/7?&/g» us are without ti^e scrobal sulcus. Sternopleural region red coloration, have yellow-brown clypeal microsculptured with sparse (apparently teeth, normal genal setae, differing mi- absent) setae. Propodeum shiny, area ad- crosculpture density between frons and jacent to enclosure and lateral sphere with vertex, and scutum with microsculpture several large areolae composed of relative- and sparse punctures. Females likely to be ly low carinae. confused with this species have one or Metasoma. Petiole with 3 or 4 coarse ca- more of the following: clypeal teeth black, rinae laterally. mesosoma red in part, setae of lower gena Color. Ground color black. White: man- elongate and relatively dense, frons and dible except apically, palpi, pronotal lobe, vertex with uniform microsculpture den- tarsi. Yellow to yellow-brown: antenna sity, and /or scutum shiny apparently (generally yellow ventrally to brown dor- without punctures. sally on apical flagellomeres), tegula, fore Description. — Male. Length 3.4^.6 mm. and mid legs except tarsi, wing veins, api- Head. Sculpture of clypeus obscured by cal metasomal sterna. Brown: mandibular dense appressed setae. Gena ventrally, apex, stigma of fore wing, hind leg except with sparse straight setae that are sub- tarsus. equal to or less than greatest width of fore Female. Length 3.9 mm. Similar to male basitarsus. Microsculpture of frons more except as follows: apicoventral mandibu- dense than that of vertex adjacent to mid lar tooth enlarged, truncate, much larger Volume 4, 1995 243 Figs. 36-41. Llaqhastigmus and Stigmus. 36, Llaqhastigmiis jatunkirus 9 , head, frontal view. 37, L. jatiinkinis 9, mandible, outer view. 38, L. jatunkirus 9, mesosoma, dorsal view. 39, Stigmus sp. 9, mandible, outer view. 40, Stigmus sp. 9, head, frontal view. 41, Stigmus sp. 9, mesosoma, dorsal view. than other teeth. Clypeal apex quadriden- tate as follows: a lateral lobe flanking lat- eral edge of labrum, a pair of smaller me- dian teeth separated from lateral lobes by broad emarginations and separated from each other by a narrow U-shaped emar- gination. Clypeal apex with specialized flattened, elongate, setae on teeth but ab- sent from emarginate interspaces; 2 setae per median tooth and 6-7 on lateral lobe. Clypeus shiny, microsculpture absent, punctures 3^ diameters apart on disc. Apical margin of clypeus yellow-brown. Material Examined {27 S, 19). — Holo- type: 9, ECUADOR: Pich. Prov. Guay- llabamba, 10km n. on Rio Pisque, 2500m 11-1983. M. Sharkey, L. Masner (BRD). Paratypes: ECUADOR: Pichincha: Guay- 244 Journal of Hymenoptera Research llabamba, 10km n. on Rio Pisque, 2500m 11-1983. L. Masner, M. Sharkey {6:2 BRD, 2 PMA). 10km n. Guayllabamba 26-11- 1983. Masner & Sharkey (cJ: 4 BRD, 19 PMA). Distribution. — (Map 20). Ecuador. 2. Llaqhastigmus australis Finnamore, new species Derivation of Name. — The species epithet is derived from the Latin "austral" mean- ing southern in reference to the southern locality of the only known specimen. Diagnosis. — Female unknown. Male: la- brum quadrilobed with a deep median notch; subomaulus absent, petiole coarse- ly carinate (carinae normal) laterally. L. australis and santanderanus are the only species in the genus lacking a subomaulus and having a quadrilobed labrum (at least in males, females for both species are un- known). In all other species in the genus the subomaulus is present and the labrum is bilobed with a slight median emargi- nation. Description. — Male. Length 5.0 mm. Head. Sculpture of clypeus obscured by dense appressed setae. Labrum quadri- lobed with a deep median emargination. Gena ventrally, with sparse straight setae that are subequal to or less than greatest width of fore basitarsus. Microsculpture of frons more dense than that of vertex. Oc- cipital carina raised ventrally and finely foveolate at least ventrally. Mesosoma. Transverse sulcus of pro- notum with longitudinal carinae. Scutum microsculptured, punctures sparse more or less evenly distributed and separated by 5-10 puncture diameters. Transscutel- lar sulcus not foveolate. Preomaular area sparsely setose, setae not obscuring un- derlying sculpture. Subomaulus absent, represented as an evanescent ridge near posteroventral apex of pronotum, not car- inate or intersecting omaulus. Scrobal sul- cus weakly impressed anteriorly, attenu- ated posteriorly, not foveolate. Sterno- pleural region shiny without microsculp- ture, sparsely punctate, sparsely setose. Propodeum shiny, area adjacent to enclo- sure without microsculpture, with several large areolae composed of low carinae; lat- eral sphere shiny, with evanescent micro- sculpture, areolae smaller. Metasoma. Petiole with 3 coarse lateral carinae. Color. Ground color black. White: man- dible except apically, palpi, pronotal lobe. Yellow-brown: scape, pedicel, fore and mid legs except coxae, hind tibia and tar- sus, tegula, apical metasomal sterna. Female. Unknown. Material Examined (IS). — Holotype: 6, BOLIVIA: Cochabamba, Prov. Carrasco, Empalme 3000m 11-1971. Fritz y Martinez (IIES). Distribution (Map 21). — Bolivia. 3. Llaqhastigmus chutiyana Finnamore, new species Derivation of Name. — The species epithet is derived from the Quechua terms "ch'uti", meaning naked and "yana", meaning black, in reference to the rela- tively hairless condition of this black spe- cies. Diagnosis. — The combination of the black scape and pronotal lobe, scrobal sul- cus foveolate, and the apicodorsal man- dibular tooth of female acute (normal) will separate this species from all others in the genus. All other species except nigricollaris have a white pronotal lobe. L. chutiyana is distinguished from nigricollaris by the black scape, foveolate scrobal sulcus and acute apicodorsal mandibular tooth in the female; nigricollaris has a yellow-brown scape, partially foveolate scrobal sulcus and enlarged apicodorsal mandibular tooth in the female. Description. — Male. Length 3.9-5.0 mm. Head. Sculpture of clypeus not ob- scured by setae. Clypeus shiny without microsculpture, setae and punctures sparse and at least 1 diameter apart on disc. Gena ventrally with evanescent mi- crosculpture, punctures and setae sparse, Volume 4, 1995 245 setae less than greatest width of fore ba- sitarsus. Microsculpture of frons more dense than that of vertex. Occipital carina raised ventrally and foveolate on at least that area. Mesosoma. Transverse sulcus of pro- notum with longitudinal carinae. Scutum microsculptured, punctures small and sparse, punctures more or less evenly dis- tributed although somewhat less dense on posterior disc. Transscutellar sulcus fo- veolate. Preomaular area asetose. Subo- maulus present. Scrobal sulcus coarsely foveolate. Sternopleural region shiny often with evanescent microsculpture, punc- tures sparse and minute. Propodeum shiny without microsculpture, area adja- cent to enclosure and lateral sphere with large areolae composed of relatively high carinae. Metasoma. Petiole with 2 or 3 lateral ca- rinae. Color. Ground color black. Yellow- brown to brown: mandibles except apical- ly, palpi, fore and mid tibiae and tarsi, hind tibia on inner side and tarsus, tegula, apical metasomal sterna. Female. Length 4.1-4.4 mm. Similar to male except as follows: apicoventral man- dibular tooth enlarged, truncate and much larger than other teeth. Clypeal apex quadridentate as follows: lateral lobe flanking edge of labrum, pair of smaller median teeth separated from lateral lobe by deep emargination and from each other by U-shaped notch. Clypeal apex with 4 specialized, flattened, elongate setae on lateral lobe and 2 per median tooth, setae absent from emarginate interspaces. Clyp- eus shiny, microsculpture absent; punc- tures sparse, 3 or more diameters apart. Apical margin of clypeus black. Material Examined (8(5, 3 9). — Holotype: (5, ECUADOR: Tungurahua, Banos 2,000m. 21-X-1974. M. Cooper B.M.1975- 33 (BMNH). Paratypes: COLOMBIA: Val- le: Penas Blancas 22-Xi-1974-L R. Wilk- erson. Malaise Trap (9:1 FSCA). ECUA- DOR: Tungurahua: Banos 2,000m. 17-X- 1974. M. Cooper B.M. 1975-33 {6: 1 BMNH). Banos 2,000m. 21-X-1974. M. Cooper B.M.1975-33 {6: 2 BMNH, 1 PMA). Banos. c.2,000m. 7-X-1978. M. Coo- per B.M. 1979-20 (9: 2 BMNH). Banos. c.2000m 19-X1-1978 M. Cooper B.M. 1979- 20 {6: 2 BMNH). Banos;-Tungurahua Norths.-:2300m; Leg. R. Hensen et A. Ap- troot; 11-8-1982 {S:\ RNH). Distribution. — (Map 22). Colombia, Ec- uador. 4. Llaqhastigmus colombianus Finnamore, new species Derivation of Name. — The species epithet, colombianus, is derived from the country Colombia, where all specimens of this spe- cies have been collected. Diagnosis. — The following combination will separate this species from others in the genus. Occipital carina foveolate ven- trally, pronotum black except white pro- notal lobe, coarsely foveolate scrobal sul- cus usually with 3 or 4 large foveae and sulcus usually diminishing in width pos- teriorly, and female clypeus with fine punctures about 2 diameters apart and equally distributed on median third. This species is similar to jatiinkirus and sharkeyi from which it differs with the coarsely foveolate scrobal sulcus and in the female with the homogeneously punctate median clypeal region. Description. — Male. Length 3.9 mm. Head. Sculpture of clypeus obscured by dense appressed setae. Gena ventrally with sparse straight setae that are sub- equal to or less than greatest width of fore basitarsus. Microsculpture of frons more dense than that of vertex. Occipital carina raised ventrally and foveolate. Mesosoma. Transverse sulcus of pro- notum with longitudinal carinae. Scutum microsculptured, punctures sparse and relatively coarse. Transscutellar sulcus foveolate. Preomaular area sparsely se- tose, setae not obscuring underlying sculpture. Subomaulus present. Scrobal sulcus with 3 or 4 coarse foveae, sulcus 246 Journal of Hymenoptera Research Figs. 42^7. Stigmus, Tzustigmus and Incastigmiis 42, Stigmus sp. S , 19). — Holotype: dense appressed setae. Gena ventrally, 9, PERU: Machu Picchu XI-30-1965 H. & with sparse straight setae that are sub- M. Townes (AEl). Paratypes: PERU: Cuz- equal to or less than greatest width of fore co: Machu Picchu XI-29-1965 H. & M. basitarsus. Microsculpture of frons more Townes {6:1 AEI). Machu Picchu Xl-30- dense than that of vertex adjacent to mi- 1965 H. & M. Townes {6:1 AEI, 1 PMA). docellus. Occipital carina ventrally, raised Distribution (Map 26). — Peru. and foveolate. Mesosoma. Transverse sulcus of pro- 8. Llaqhastigmus mantanti Finnamore, ^^^^^ carinate. Scutum microsculptured, new species sometimes shiny medially, punctures Derivation of Name. — The species epithet sparse, posterior margin with transverse mantanti is derived from two Quechua row of short carinae. Transscutellar sulcus words, "manta" a preposition meaning foveolate. Sculpture of preomaular area "from" and "anti" meaning "east". The obscured by dense appressed setae that name refers to the distribution of this spe- are usually continued along midventral cies in eastern Brazil. line to mid coxae. Subomaulus present. Diagnosis. — The preomaular area sculp- Scrobal sulcus impressed, foveolate. Ster- ture that is obscured by setae in male and nopleural region shiny, weakly micro- partially obscured by setae in female will sculptured, sparsely punctate, with in- separate this species from all others in the creasing setal density ventrally. Propo- genus. Although females of this species deum usually microsculptured and usu- 252 Journal of Hymenoptera Research ally with relatively small areoleae adjacent eira (9:1 PMA). Parana: Campina Grande to enclosure that are composed of relative- nr. Curitiba 11-23-1966 H. & M. Townes ly coarse carinae. Some specimens have (6:1 AEI). Pernambuco: Caruaru IV-1972 larger more typical areolae adjacent to en- M. Alvarenga (9: 2 PMA). Caruaru V- closure. 1972. J. Lima (9:1 PMA). Curuaru [sic] Metasoma. Petiole, laterally smooth or 900m IV-1972 (9:1 BRD). Rio de Janeiro: with several fine carinae. III-5-1966 H. & M. Townes (9: 1 AEI). Color. Ground color black. White: palpi, Guanabara 11-1972. M. Alvarenga ( 9 : 1 base of mandible, pronotal lobe, at least PMA). Mangaratiba, Muriqui VII-1969 M. fore tarsus, sometimes all tarsi. Yellow to Alvarenga (9:1 AEI). Teresopolis III-12- yellow-brown: antenna, mandible medi- 1966 H. & M. Townes (d: 1 AEI). Santa ally, tegula, legs except at least fore tarsus Catarina: Nova Teutonia 27'^'11'B.52"23'L. and sometimes hind femur and tibia. Red: 8-X-1937 Plaumann. B.M.1938-312 {6: 1 at least pronotal side to entire mesosoma BMNH). Sao Paulo: Sao Paulo XII-31-1968 except legs, pronotal lobe, tegula and pro- V.N. Alin {6:1 USNM). San Paulo 20-1- podeum; all males have some mesosomal 1978 [label in Russian] {6:2 ZMMU). red coloration. Brown: occasionally hind Distribution (Map 27). — Southeastern femur and tibia. Brazil. Female. Length 4.8-5.0 mm. Similar to male except as follows: apicoventral man- ^- Llaqhastigmus muthus Finnamore, dibular tooth enlarged, but only slightly ^^^ species larger than other teeth. Clypeal apex Derivation of Name. — The species epithet quadridentate with lateral lobe flanking is derived from the Quechua "muthu" lateral edge of labrum, pair of small me- which means dull, in reference to the de- dian teeth separated from lateral lobe by gree of microsculpture found on members broad emargination and separated from of this species. each other by U-shaped notch. Apical Diagnosis. — The dense uniform micro- clypeal edge with specialized, elongate, sculpture on the head (except clypeus) flattened setae on teeth but not on emar- and mesosoma (except sometimes sterno- ginate interspaces (2 setae per median pleural region and propodeum) will sep- tooth and 4 per lateral lobe). Clypeus arate this species from all others in the ge- shiny, microsculpture absent, punctures nus. In all other species the microsculp- variable from almost absent to sparsely ture of the frons is of greater density than distributed over entire surface. that of the vertex. Material Examined {66, 19 9). — Holo- Description. — Male. Length 3.5^.0 mm. type: 9 , BRAZIL: Caruaru, Pernambuco Head. Sculpture of clypeus obscured by 900m IV-1972 M. Alvarenga (BRD). Para- dense appressed setae. Gena ventrally, types: BRAZIL: Bahia: 7-VIII-1983 Cepec with sparse straight setae that are sub- Itabuna P.P. Benton (9: 1 BMNH). Gua- equal to or less than greatest width of fore nabara: Floresta da Tijuca IV-1966 Alvar- basitarsus. Microsculpture of frons and enga & Seabra (9: 1 AEI). Represa Rio vertex of uniform (subequal) density. Oc- Grande, Guanabara XII-1967 F.H. Oliveira cipital carina ventrally, slightly raised and (9:1 BRD). Represa Rio Grande 1-1968 M. finely foveolate. Alvarenga (9: 2 PMA). Represa Rio Gran- Mesosoma. Transverse sulcus of pro- de 11-1972 P.M. Oliveira (9:2 PMA). Re- notum with longitudinal carinae. Scutum presa Rio Grande VII-1972 M. Alvarenga densely microsculptured, impunctate. (9:2 BRD). Minas Gerais: Serra do Car- Transscutellar sulcus foveolate. Preomau- aca, Sta. Barbara 1-1970 (9:1 PMA). Serra lar area sparsely setose, setae not obscur- do Caraca, S. Barbara III-1971 P.M. Oliv- ing underlying sculpture. Subomaulus Volume 4, 1995 253 present. Scrobal sulcus weakly impressed FSCA). Lima: Chancay shrubs nr. river 40 with evanescent usually irregular foveae miles n. of Lima 29-VII-1971 P.S. & H.L. best developed anteriorly. Sternopleural Broomfield B.M. 1971^86 fertile irrigated region densely microsculptured, impunc- region in arid coastal desert {S : I BMNH). tate. Propodeum weakly microsculptured Chosica XI-1961 N.L.H. Krauss {9 1 to shiny; area adjacent to enclosure and USNM). Cupiche, 10km n. Chosica 25-VI- lateral area carinate; carinae converging 2-VII-1974 C. Porter & L. Stange (J: 1 LIL- dorsolaterally to form small irregular are- LO, 9 : 2 LILLO). Cupiche 26-VI-1976 C. olae. Porter, C. Calmbacher {S: I FSCA). Lima Metasoma. Petiole with several coarse 13-XI-1950 E.S. Ross (9: 2 CAS). Lima carinae laterally. City, S. Marcos Univ. campus 26-27-IV- Color. Ground color black. White: palpi, 1983 Mai. trap C.& M. Vardy B.M.1983- mandible except apically, pronotal lobe. 217 (9:4 BMNH, 1 PMA). San Geronimo Yellow to yellow-brown: scape, pedicel ca. Chosica 1-5-VII-1976 C. Porter, C. and basal flagellomeres ventrally, tegula, Calmbacher (9:3 FSCA). San Geronimo fore and mid legs except coxae, hind tro- 30-VII-1982 R.B. Miller, L.A. Stange (9:2 chanter, hind tarsus. Brown: flagellomeres FSCA). VENEZUELA: Distrito Federal: except as indicated above, hind leg except nr. Caracas 28-VIII-1943 D.G. Hall (9: 1 trochanter and tarsus. USNM). Female. Length 3.7-4.6 mm. Similar to Distribution (Map 28). — Venezuela, Fe- male except as follows: apicoventral man- uador, Peru, dibular tooth enlarged, truncate, much larger than other teeth. Clypeal apex ^^- Llaqhastigmus nigricollaris quadridentate as follows: lateral lobe Finnamore, new species flanking lateral edge of labrum, pair of Derivation of Name. — The species epithet smaller median teeth separated from lat- is derived from two latin words, niger and eral lobe by broad emargination and from collaris, in reference to the dark pronotal each other by narrow U-shaped notch, collar characteristic of this species. Clypeal apex with specialized, elongate. Diagnosis. — The combination of a brown flattened setae present on median teeth (2 or black pronotal lobe, antennal scape yel- per tooth) and on lateral lobe (6-7 per low-brown and in the female the apico- lobe) with occasionally single seta on dorsal mandibular tooth (in addition to emarginate interspace. Clypeus shiny, apicoventral tooth) enlarged and truncate, sparsely punctate, area immediately above will separate this species from all others median teeth impunctate and area be- in the genus. tween antennal sockets most densely Description. — Male. Length 3.7-3.9 mm. punctate with punctures 1 diameter or less Head. Sculpture of clypeus obscured by apart (elsewhere punctures 3 or more di- dense appressed setae. Gena ventrally, ameters apart). Lateral lobe yellow-brown with sparse straight setae that are sub- to red-brown. equal to or less than greatest width of fore Material Examined {86, 16 9). — Holo- basitarsus. Microsculpture of frons more type: 6 , PERU: Samne, ca. 40km ne. Tru- dense than that of vertex adjacent to mid jillo, Prov. La Libertad 7°59's 78°4rw elev. ocellus. Occipital carina ventrally, raised ca. 1500m 12-17-VII-1975 C. Porter, L. and foveolate. Stange (LILLO). Paratypes: PERU: San Mesosoma. Transverse sulcus of pro- Bartolome VII-1913 C.T. Brues {6: 1 notum with longitudinal carinae. Scutum MCZ). La Libertad: same data as holotype shiny, weakly microsculptured especially {6: 1 LILLO, 1 PMA). 15km e. Laredo 19- on disc, punctures sparse. Transscutellar VII-1982 R.B. Miller, L. A. Stange {6: 1 sulcus foveolate. Preomaular area sparsely 254 Journal of Hymenoptera Research setose, setae not obscuring underlying XII-1974 Malaise trap (?: 1 FSCA). Penas sculpture. Subomaulus present. Scrobal Blancas 1750m 10km w. Call, very wet sulcus weakly impressed, irregularly fo- premontane forest, R.C. Wilkerson 20-22- veolate anteriorly, occasionally entirely fo- 1-1975 Malaise trap ( $ : 3 FSCA). Penas veolate. Sternopleural region weakly mi- Blancas 1750m 10km w. Cali, very wet crosculptured with sparse punctures and premontane forest, R.C. Wilkerson 31-1- setae. Propodeum shiny, with several 1975 Malaise trap (9:3 FSCA, 1 PMA). coarse areolae adjacent to enclosure com- Penas Blancas 12-11-1975 R. Wilkerson posed of relatively high carinae. Malaise trap (9:3 FSCA, 1 PMA). Penas Metasoma. Petiole with 2 to 4 coarse ca- Blancas 17-19-III-1975 R. Wilkerson Mal- rinae laterally. aise trap (9:1 FSCA). ECUADOR: Napo: Color. Ground color black. Yellow to El Chaco 2000m 11-1983 M. Sharkey, L. yellow-brown: palpi, mandibles medially, Masner {6:1 BRD). PERU: Huanuco: Tin- scape, pedicel and basal flagellomeres go Maria, 40 mi. s. Carpish Mts. XII-28- ventrally, tegula, fore and mid legs except 1954 E.I. Schlinger & E.S. Ross (9:1 CAS), coxae and usually except femora, hind tro- Distribution (Map 29). — Colombia, Ec- chanter and tarsus. Brown: mandible ba- uador, Peru, sally, antenna other than noted above, pronotal lobe, femora, hind tibia. l^- Llaqhastigmus santanderanus Female. Length 4.2-5.0 mm. Similar to Finnamore, new species male, except as follows: apicoventral and Derivation of Name. — The species epithet apicodorsal mandibular teeth enlarged, is derived from the Colombian state of truncate. Clypeal apex quadridentate as Santander Del Norte, the northernmost follows: lateral lobe flanking lateral edge record of the species. of labrum, pair of small median teeth sep- Diagnosis. — The species is distinguished arated from lateral lobe by deep, broad, by the lack of a subomaulus, and a petiole emargination and from each other by shal- that is finely carinate to microcarinate lat- low notch. Clypeal apex with specialized, erally. This species may prove to be con- flattened, elongate setae on lateral lobe (4 specific with australis, at present there are or 5) and on median teeth (2 per tooth) but too few specimens to assess variation generally absent on emarginate interspac- within either species, es. Clypeus shiny, microsculpture absent. Description. — Male. Length 4.8-5.2 mm. punctures sparse (3 or more diameters Head. Sculpture of clypeus obscured by apart laterally, slightly more dense medi- dense appressed setae. Gena ventrally, ally), lateral lobe red-brown. with sparse straight setae that are sub- Material Examined {26, 16 9 ). — Holo- equal to or less than greatest width of fore type: 9, COLOMBIA: Valle Dept.: Penas basitarsus. Microsculpture of frons only Blancas 1750m 10km w. Cali, very wet slightly more dense than that of vertex, premontane forest, R.C. Wilkerson 15-1- Occipital carina ventrally, raised and fov- 1975 Malaise trap (FSCA). Paratypes: CO- eolate. LOMBIA: Cauca: Popayan 1,800m 11-X- Mesosoma. Transverse sulcus of pro- 1974 M. Cooper B.M. 1975-33 {16: 1 notum with longitudinal carinae. Scutum BMNH, 9: 1 BMNH). Valle: Cali 900m 17- microsculptured, punctures sparse, ob- 1-1972 M. Cooper B.M.1972-275 (9: 1 scure. Transscutellar sulcus not, or at most BMNH). Pance CVC 1700m 15km w. Cali, obscurely foveolate. Preomaular area very wet premontane forest, R.C. Wilker- sparsely setose, setae not obscuring un- son 23-XII-1974 Malaise trap (9: 1 FSCA). derlying sculpture. Subomaulus absent. Penas Blancas 1750m 10km w. Cali, very Scrobal sulcus slightly impressed, foveae wet premontane forest, R.C. Wilkerson 23- absent or slightly evident anteriorly. Ster- Volume 4, 1995 255 nopleural region microsculptured, sparse- ly punctate. Propodeum microsculptured, area adjacent to enclosure with fine irreg- ular carinae, some small areolae dorsola- terally that are composed of fine, low ca- rinae. Metasoma. Petiole microcarinate later- ally or with fine carinae laterally. Color. Ground color black. White: man- dible except apically, pronotal lobe. Yel- low-brown: palpi, antenna, tegula, fore and mid legs except coxae, hind trochan- ter and tarsus. Brown: hind femur and tib- ia. Female. Unknown. Material Examined (2c?). — Holotype 6, ECUADOR: Napo: Pastaza, Sebundoi, 2600m 11-15-IX-1977 L. Peha B.M.1978- 293 (BMNH). Paratype: COLOMBIA: Santander Del Norte: Oroque lO-Vl-1965 J.&B. Bechyne {6:1 UCV). Distribution (Map 30). — Colombia, Ec- uador. 12. Llaqhastigmus sapanis Finnamore, new species Derivation of Name. — The species epithet is derived from the Quechua "sapan" meaning lone, in reference to the single known male specimen. Diagnosis. — This species can be separat- ed from all others in the genus with the following combination of characters: transverse sulcus of pronotum smooth and propodeum areolate dorsolaterally. The only other species known to have a smooth pronotal sulcus is Ihitanis which differs from sapmnis in its non-areolate pro- podeum. Description. — Male. Length 3.8 mm. Head. Sculpture of clypeus obscured by dense appressed setae. Gena ventrally, with sparse straight setae that are sub- equal to or less than greatest width of fore basitarsus. Microsculpture of frons more dense than that of vertex. Occipital carina ventrally, raised and foveolate. Mesosoma. Transverse sulcus of pro- notum smooth, without carinae. Scutum microsculptured, punctures relatively coarse and sparse (usually 3 or more di- ameters apart on disc). Transscutellar sul- cus foveolate. Preomaular area sparsely setose, setae not obscuring underlying sculpture. Subomaulus present. Scrobal sulcus weakly impressed, not foveolate. Sternopleural region microsculptured, punctures sparse and evanescent. Propo- deum microsculptured, area adjacent to enclosure and dorsolateral region areolate, areolae composed of relatively low cari- nae; side irregularly carinate. Metasoma. Petiole carinate laterally. Color. Ground color black. White: man- dible except apically, pronotal lobe. Yel- low-brown: palpi, scape, ventrally on ped- icel and basal flagellomeres, tegula, fore and mid legs except coxae and femora, hind tarsus. Brown: antenna except as not- ed, fore and mid femora, hind leg except tarsus and coxa. Female. Length 4.0-4.1 mm. Similar to male except as follows: apicoventral man- dibular tooth enlarged, truncate, much larger than other teeth. Clypeal apex quadridentate as follows: lateral tooth flanking lateral edge of labrum, pair of smaller median teeth separated from lat- eral tooth by broad emargination and sep- arated from each other by narrow U- shaped notch. Clypeal apex with special- ized, flattened, elongate setae confined to teeth (2 per median tooth and 3 or 4 per lateral tooth). Clypeus shiny, without mi- crosculpture, punctures sparse and rela- tively more dense between antennal sock- ets. Apical margin of clypeus black. Material Examined {IS, 49).— Holotype: 9, PERU: Machu Picchu XI-28-1965 H.& M. Townes (AEl). Paratypes: BOLIVIA: Cochabamba: Carrasco, Empalme 3000m 11-1971 Fritz y Martinez {6: 1 IIES). PERU: Cuzco: Machu Picchu XIl-1-1965 H. & M. Townes (9:1 AEI). Machu Picchu 11-24- 27-1968 A. Garcia & C. Porter (9 : 2 MCZ). Distribution (Map 31).— Bolivia, Peru. 256 Journal of Hymenoptera Research 13. Llaqhastigmus shachus Finnamore, new species Derivation of Name. — The species epithet is derived from the Quechua "sh'achu", meaning hairy in reference to the dense pilosity of this species. Diagnosis. — This species can be distin- guished by the presence of dense, elongate setae on the lower gena and sternopleural region. In the male, setae of lower gena are bent apically; in the female only a few near the hypostomal carina are bent api- cally. In both sexes the lower genal setae are longer than the greatest width of the fore basitarsus. Females of this species have a quadridentate mandible due to a deep U-shaped notch in the apicodorsal mandibular tooth (visible only by spread- ing the mandibles). All other species have tridentate mandibles and short sparse se- tae on the lower genal region. Setae in oth- er species are never longer than greatest width of fore basitarsus. Other species have little or no setae on the sternopleural region. Description. — Male. Length 4.4 mm. Head. Sculpture of clypeus obscured by dense appressed setae. Gena ventrally, with dense elongate setae that are about twice the greatest width of fore basitarsus and bent 90' inward toward oral cavity. Microsculpture of frons slightly more dense than that of vertex. Occipital carina ventrally, raised and foveolate. Mesosoma. Transverse pronotal sulcus carinate. Scutum microsculptured, punc- tures small and sparse. Transscutellar sul- cus foveolate. Subomaulus present. Preo- maular area densely setose, setae erect not obscuring underlying sculpture. Scrobal sulcus weakly impressed, a few evanes- cent irregular foveae anteriorly. Sterno- pleural region weakly microsculptured, densely and minutely punctate (punctures about 2 diameters apart ventrally); setae dense ventrally and elongate, slightly shorter than genal setae and with many tips bent 90° toward posterior. Propodeum weakly microsculptured, area adjacent to enclosure and dorsolateral sphere with numerous areolae composed of relatively low carinae, side more or less vertically carinate with carinae merging into areolae of dorsolateral sphere. Metasoma. Petiole laterally, with 4 lon- gitudinal coarse carinae. Color. Ground color black. White: man- dibles basally, pronotal lobe. Yellow- brown: Palpi, mandible medially, scape ventrally, tegula, fore and mid legs except coxae and femora, hind tarsus. Female. Length 4.8-5.4 mm. Similar to male except as follows: mandible quadri- dentate, apicodorsal tooth with deep U- shaped notch thereby imparting quadri- dentate appearance to the mandible; api- coventral mandibular tooth enlarged, truncate, much larger than other teeth. Clypeal apex quadridentate as follows: lateral lobe flanking lateral edge of la- brum, pair of median teeth separated from lateral lobe by broad emargination and from each other by U-shaped notch. Clyp- eal apex with specialized, flattened, elon- gate setae distributed over most of apex. Clypeus shiny, without microsculpture, punctures sparse (5-10 diameters apart). Apical clypeal margin black. Genal and sternopleural setae not as dense or as long as those in male. Omaulus ending ven- trally, not continued forward to acetabular carina. Material Examined (16, 2 9). — Holotype: 9, PERU: Huanuco IX-16-1954 E.I. Schlin- ger & E.S. Ross (CAS). Paratypes: ECUA- DOR: Loja: 14km w. Catamayo 1770m 9- VII-1989 L. Stange & R. Miller (9: 1 FSCA). PERU: Huanuco: Huanuco IX:16- 1954 E.I. Schlinger & E.S. Ross {6: 1 CAS). Distribution (Map 32). — Ecuador, Peru. 14. Llaqhastigmus sharkeyi Finnamore, new species Derivation of Name. — This species is named in honour of Michael J. Sharkey, co-collector of the primary type material. Diagnosis. — The combination of a white Volume 4, 1995 257 pronotal lobe; shiny scutum with punc- along edge except for median teeth which tures minute and sparse, without micro- bear 2 additional setae each. Clypeus sculpture on posterior % except narrow shiny, punctures minute but dense later- posterior margin; and a coarsely foveolate ally and larger and more sparse medially scrobal sulcus will separate this species (1 to 2 diameters apart). Apical margin of from all others in the genus. clypeus black. Description. — Male. Length 3.6-4.7 mm. Material Examined {46 , 4 9). — Holotype: Head. Sculpture of clypeus obscured by S , ECUADOR: Napo Prov.: Baeza, 5km s. dense appressed setae. Gena ventrally, 11-1983 2000m Sharkey, Masner (BRD). with sparse straight setae that are sub- Paratypes: COLOMBIA: Santander del equal to or less than greatest width of fore Norte: Prima 1700m 27-V-1965 J. & B. Be- basitarsus. Microsculpture of frons more chyne (9: 1 UCV). ECUADOR: Napo: dense than that of vertex. Occipital carina Baeza, 5km s. 11-1983 2000m Sharkey, ventrally, raised and foveolate. Masner (9:1 BRD). El Chaco 2000m II- Mesosoma. Transverse pronotal sulcus 1983 M. Sharkey, L. Masner {6:1 PMA). with longitudinal carinae. Scutum shiny Pastaza, Reventador 1750m 3-5-X-1977 on most of posterior %, microsculptured L.E. Pena B.M.1978-293 {6: 1 BMNH). anteriorly and narrowly on posterior mar- VENEZUELA: Aragua: Hac. Portapan gin; punctures minute, sparse, 10 or more 1700m 15-VIII-1969 J. & B. Bechyne (9:1 diameters apart. Transscutellar sulcus fo- UCV). Rancho Grande 1100m 16-X1-1967 J. veolate. Preomaular area sparsely setose, & b. Bechyne (9: 1 UCV). Rancho Grande, setae not obscuring underlying sculpture. Portachuelo 1100m 22-V-1981 J.A. Clavijo, Subomaulus present. Scrobal sulcus im- J. L. Garcia ((5: 1 UCV). Falcon: Curimagua pressed, foveolate. Sternopleural region uoOm l-XII-1971 J. & B. Bechyne (9:1 shiny, without microsculpture, punctures UCV). minute and sparse. Propodeum shiny, Distribution (Map 33).— Ecuador, Co- without microsculpture, area adjacent to lombia Venezuela, enclosure and lateral sphere with areolae composed of relatively high carinae; side GENERIC RELATIONSHIPS unsculptured toward metapleuron. Metasoma. Petiole carinate laterally. The following hypothesis of relation- Color. Ground color black. White: pro- ships of genera among the Shgmma is notal lobe. Yellow to yellow-brown: palpi, based on outgroup analysis of character mandible medially, scape, pedicel, ven- state distribution. The characters listed be- trally on basal flagellomeres or more, teg- ^^w were polarized by outgroup analysis ula, fore and mid legs except coxae, hind with the Spilomenina. Morphology is after tarsus. Brown: antennal flagellum except Bohart and Menke (1976) unless otherwise as noted above, hind leg except tarsus. indicated in introduction, only characters Female. Length 4.2^.6 Similar to male not sufficiently treated by those authors except as follows: apicoventral mandibu- are explained in detail below. In the fol- iar tooth enlarged, truncate, much larger lowing analysis 0 denotes the plesiotypic than other teeth. Clypeal apex quadriden- state and 1 to 5 denote various apotypic tate as follows: lateral tooth flanking lat- expressions of a character and do not nec- eral edge of labrum, pair of smaller me- essarily represent transition series. Data dian teeth separated from lateral tooth by matrices for the Stigmina, Pemphredoni- broad emargination and separated from na, Spilomenina, and Ammoplanina are each other by narrow U-shaped notch, presented in Tables I-IV respectively. A Clypeal apex with 5 or 6 specialized, elon- cladogram illustrating relationships is pre- gate, flattened setae equally distributed sented in Fig. 51. The cladogram is based 258 Journal of Hymenoptera Research o c •§ ••3 '-a 3 2 o JQ b I S I 3 •I .1 -I I *• ^^ ■*" ••■ CA TSs CA ■'irf 3 ^ Fig. 51. Cladogram depicting relationships among genera of Stigmina. on the following characters and was de- Stigmina and the Pemphredonina, palpal veloped without electronic assistance. segment length generally exceeds twice its Character Polarization ^^^t^^' P^^P^ greatly exceed hypostomal carma. 1. Palpal formula. 1— Palpal formula 6-A, palpal segments 0 — Palpal formula 6-4, palpal segments short, palpi inconspicuous. Found in the long, palpi conspicuous. Found in the Ammoplanina (genera examined: Piilver- Volume 4, 1995 259 ro, Ammoplanus, Ammoplanops, Parammo- planus, Ammoplanellus, Timberlakena), pal- pal segment length generally about twice its width or less, palpi at most slightly ex- ceed hypostomal carina. 2 — Palpal formula 5-4, palpal segments short, palpi inconspicuous. This character is synapotopic for the Spilomenina and was the justification used by Menke (1989) to separate the group from the Stigmina. Palpal segments are short, particularly basal segments which may be no longer than wide, palpi at most slightly exceed hypostomal carina. 2. Labrum shape. 0 — Labrum truncate to broadly round- ed, wider than long. This character state is found in Arpactophilus, Microstigmus, Xys- ma, Pulverro, Ammoplanops, and many oth- er apoid and vespoid wasps. This charac- ter state in the genera listed above may represent a reversal from 2.3 or 2.4. Only a few species and specimens were avail- able for dissection in these groups, fur- thermore the difference between truncate and broadly rounded is sometimes slight, rendering the polarity of this character es- pecially difficult to establish. 1 — Labrum triangular, longer than wide. Present in Pemphredon, Polemistus, and Carinostigmus, as well as most Am- pulicinae (Ampulex) and Ammophilini (Sphecinae). Sides of triangle are out- curved, basal and apical regions are not distinguished. This state may represent the correct plesiotypic expression of the la- brum in the Pemphredonini. 2 — Labial apex distinguished from base. In Passnloecus the lateral margins of the la- brum are incurved resulting in a state in which the base is abruptly broadened and thereby distinguished from the apex. The abrupt shoulder is present, although mod- ified, in all genera listed under 2.3, 2.4, and presumably 2.5 although I have not seen specimens of the latter. 3 — Labrum emarginate to bilobed. This development ranging from a slight emar- gination found in Paracrabro to the deeply bilobed expression found in Diodontus, Llaqhastigmus, Arpactophilus, Ammop)lamts, Parammop)lanus, Ammoplanellus, Timberlak- ena and as well is present in those genera listed under 2.4. The labrum of Llaqhastig- mus is considered a character reversal from 2.4 caused by a reduction of the me- dian lobes. 4 — Labrum 4-lobed. This character state results from an apical shift of the lateral shoulders (see character state 2.2 above) imparting a 4-lobed appearance to the la- brum. Found in Stigmus, Tzustigmus, Ar- aucastigmus, Parastigmus, Aykhustigmus, In- castigmus, Arpactophilus, and Spilomena. 5 — Labrum 6-lobed. I have not seen specimens bearing this character. Menke (1989) noted undescribed Arpactophilus having a "six toothed labrum". 3. Mandibular socket. 0— Open. 1 — Closed. 4. Apical mandibular teeth. 0 — Mandible with 2 apical teeth. This character state is found in males of all gen- era of Stigmina, except Incastigmus and Llaqhastigmus, it also occurs in both sexes of all genera in the Spilomenina and spo- radically in all genera of the Pemphredon- ina. 1 — Mandible unidentate, apically ter- minating in a single point or truncate. 2 — Mandible tridentate apically (Figs. 37, 39). Females of most genera of Stig- mina (except Parastigmus, Paracrabro and some Stigmus) and males of Incastigmus and Llaqhastigmus. The character state also occurs in some species of Passaloecus and Pemphredon. 3 — Mandible with 4 apical teeth. 4 — Mandible with 5 apical teeth. 5 — Manciible with 6 apical teeth. 5. Apicoventral mandibular tooth in females having a tridentate mandible (character 4.2). 0 — Apicoventral tooth acute, similar in size to other teeth (Fig. 39). 260 Journal of Hymenoptera Research 1 — Apicoventral tooth enlarged and truncate (Fig. 37). This character state oc- curs in females of Llaqhastigmus and in- dependently in females of some Pemphre- don. 6. Clypeal apicolateral bevel. 0 — Clypeus unmodified apicolaterally. 1 — Clypeal apex bevelled apicolaterally (Fig. 42). The lateral clypeal bevel is an au- tapotypic character state occurring in Stig- miis. It occurs only in males and is most developed in species from the Neotropical Region. 7. Clypeus size. 0 — Clypeus not reduced, antennal sock- et usually more than its diameter from nearest clypeal edge; labrum generally flat, lightly sclerotized, mostly or entirely obscured. Found in males and most fe- males. 1 — Clypeus reduced, antennal socket about its diameter from nearest clypeal edge, labrum convex, heavily sclerotized, entirely exposed, or almost so. Some fe- males of a few genera. 8. Clypeal teeth. 0 — Clypeus without teeth or lobes on apical margin. 1 — Clypeus with single median tooth or lobe. 2 — Clypeus with 2 median teeth. 3 — Clypeus with 3 teeth. 4 — Clypeus with 4 teeth. 5 — Clypeus with 2 lateral teeth, median teeth lost. 9. Interantennal tubercle. 0 — Reduced or absent, usually repre- sented by a small, raised conical point, sel- dom exceeding half width of antennal scape (Fig. 36). 1 — Protruberant, often exceeding width of antennal scape (Figs. 6, 7). 10. Frontal carina. 0 — Frontal carina absent, indicated by a groove, complete or in part, from mido- cellus to clypeus. 1 — Frontal carina absent, frons flat, without indication of groove or carina. 2 — Frontal carina present as a simple raised line, complete or in part, from mi- docellus to ocellus. 3 — Frontal carina strongly raised, la- mellate, usually between antennal sockets. 11. Vertex micropore field. 0 — Vertex, between lateral ocellus and eye, of similar microsculpture and punc- tation to adjacent areas; without group of loosely associated punctures or discrete pore field. 1 — Vertex, between lateral ocellus and eye, with a loosely associated group of punctures (Figs. 3, 8). 2 — Vertex, between lateral ocellus and compound eye, with a micropore field, ap- pearing as a discrete patch of fine micro- sculpture relative to adjacent microsculp- ture (Figs. 19, 20, 25, 26, 31, 32). 12. Eye margins. 0 — Eye not margined by a carina. 1 — Eye at least partially margined by a carina. 13. Occipital Carina. 0 — Occipital carina present, simple. 1 — Occipital carina raised, foveolate. 2 — Occipital carina incomplete dorsally, present ventrolaterally. 3 — Occipital carina absent. 14. Occipital-hypostomal carinae. 0 — Occipital carina intersecting hypo- stomal carina. 1 — Occipital carina ending ventrally not intersecting hypostomal carina. 2 — Occipital carina forming a complete circle, not intersecting hypostomal carina. Volume 4, 1995 261 15. Pronotal collar, 0 — Pronotal collar of normal length, transverse sulcus without median longi- tudinal raised area (Figs. 7, 10). 1 — pronotal collar elongate, transverse sulscu with median longitudinal area raised (Figs. 2, 5). 16. Transverse pronotal carina. 0 — Present, at least laterally. 1 — Absent. 17. Median scutal groove. 0 — Scutum without median groove or pit (Figs. 38, 41). 1 — Scutum with a median groove or at least a median posterior pit (Figs. 33, 35). 18. Notaular grooves. 0 — Notaular grot^ves unmodified, pres- ent anterolaterally (Figs. 38, 41). 1 — Notaular grooves elongate, often to posterior margin of scutum (Figs. 33, 35). 19. Acetabular carina. 0 — Acetabular carina absent (Fig. 11). 1 — Acetabular carina present (Fig. 43). 20. Omaulus. 0 — Omaulus present. 1 — Omaulus absent. 21. Episternal sulcus. 0 — Episternal sulcus present. 1 — Episternal sulcus absent. 22. Hypersternaulus. 0 — Hypersternaulus present. 1 — Hypersternaulus absent. 23. Orientation of hypersternaulus. 0 — Obliquely oriented. 1 — Horizontally oriented. 24. Scrobal sulcus. 0— Present (Fig. 34). 1— Absent (Fig. 14). 25. 6 midbasitarsus. 0 — Midbasitarsus elongate, longer than remaining tarsomeres combined. 1 — Midbasitarsus, straight, cylindrical, shorter than remaining tarsomeres com- bined. 2 — Midbasitarsus short, modified — curved, excavated, expanded or combina- tions thereof. 26. Hindtibial spines. 0 — One or more rows of spines on pos- terior surface. 1 — 2-4 widely separated spines on pos- terior surface. 2 — Without spines on posterior surface. 27. Propodeal enclosure. 0 — Present. 1 — Absent. 28. Stigma size. 0 — Stigma of fore wing of normal size, smaller than marginal cell. 1 — Stigma of fore wing enlarged about size of marginal cell or larger. 29. Stigma shape. 0 — Stigma of fore wing broadly lentic- ular. 1 — Stigma of fore wing subglobular. 30. Stigma micropore field. 0 — Stigma of fore wing without micro- pore field, dorsal surface uninterrupted. 1 — Stigma of fore wing with micropore field, appearing as a discrete circular or elliptical microsculpture patch interrupt- ing dorsal surface (Figs. 46, 47). 31. Marginal cell size. 0 — Marginal cell not reduced, larger than stigma. 1 — Marginal cell reduced, smaller than stigma. 32. Marginal cell veins. 0 — Marginal cell closed. 1 — Marginal cell open. 262 Journal of Hymenoptera Research 33. Submarginal cells. 0 — Two submarginal cells. 1 — One submarginal cell. 2 — Submarginal cells absent. 34. Submarginal cell position. 0 — Outer veinlet of submarginal cell 1 out of line with outer veinlet of marginal and discoidal cells. 1 — Outer veinlet of submarginal cell 1 in line with outer veinlet of marginal and discoidal cells. 35. Submarginal cell I veins. 0 — Submarginal cell 1 closed. 1 — Submarginal cell 1 open. 36. Discoidal cell number. 0 — Fore wing with three discoidal cells. 1 — Fore wing with two discoidal cells. 2 — Fore wing with one discoidal cell. 37. Hind wing media divergence. 0 — Hind wing media diverging before or at cu-a. 1 — Hind wing media diverging just af- ter cu-a. 2 — Hind wing media diverging far after cu-a. 38. Hind wing submedian cell. 0 — Hind wing submedian cell about half or more length of median cell. 1 — Hind wing submedian cell about one third length of median cell. 39. Hind wing cells. 0 — Closed cells present in hind wing. 1 — Hind wing without closed cells. 40. Petiole length. 0 — Petiole undeveloped, abdomen ses- sile. 1 — Petiole developed, composed of ster- num only, wider than long. Some Micros- tigmus approximate this character state but the apparent petiole is composed of both tergum and sternum. 2 — Petiole developed, composed of ster- num only, longer than wide. 41. Petiolar sculpture. 0 — Petiole carinate — 2 or more longitu- dinal carinae. 1 — Petiole punctate. 2 — Petiole smooth and round. 42. Sternum I carinae. 0 — Metasomal sternum 1 without basal carinae. 1 — Metasomal sternum 1 with single basal medial carina. 2 — Metasomal sternum 1 with two basal median carinae. 43. Sternum 1 basal groove. 0 — Metasomal sternum 1 basal trans- verse groove undeveloped. 1 — Metasomal sternum 1 with a trans- verse groove or constriction just posterior to base. 44. Tergum 1 lateral carina. 0 — Metasomal tergum 1 with a lateral carina. 1 — Metasomal tergum 1 lateral carina absent. 45. Sternum 2 microsetal patches. 0 — Metasomal sternum 2 without mi- crosetal patches. 1 — Metasomal sternum 2 with discrete microsetal patch laterally (Figs. 44, 45). 46. Pygidial Plate. 0 — Pygidial plate present, broad. 1 — Pygidial plate present, narrow, trough-like. 2 — Pygidial plate absent. 47. Male Sternum 8. 0 — Metasomal sternum 8 broad, not narrowed apically. 1 — Metasomal sternum 8 triangular, gradually narrowed apically. 2 — Metasomal sternum 8 narrowly tri- angular, base differentiated from apex. Volume 4, 1995 263 Fig. 52. World distribution and post continental drift dispersal routes for Carinostigmus (light shading) and other genera in the Stigmina (dark shading). Wide lines indicate generic level dispersals, narrow lines indicate species group dispersal. 3 — Metasomal sternum 8 abruptly nar- rowed, apical portion forming a pseudo- sting, elongate and parallel-sided. 4 — Metasomal sternum 8 spatulate api- cally. 48. Genitalia. 0 — Digitus and cuspis lobular, more or less equal in length. 1 — Digitus, elongate-triangular, longer than cuspis. 2 — Digitus elongate, sides parallel, greatly exceeding cuspis. 49. Microsculpture. 0 — Present on most of body, dull. 1 — Absent, body shiny. DISCUSSION The relationship of subtribes within the Pemphredoninae remains the same as that presented in Bohart and Menke (1976). The sister subtribe to the Stigmina is pre- sumed to be the Spilomenina, based on the shared presence of a 4-lobed labrum (character 2.4 above). The Ammoplanina (synapotypies: loss of transverse pronotal carina 16.1 and loss of omaulus 20.1) is considered immediate outgroup to the Stigmina + Spilomenina based on the syn- apotypic presence of an enlarged fore wing stigma. The immediate outgroup to the Stigmina + Spilomenina + Ammo- planina is the Pemphredonina based on the synapotypic loss of the third submar- ginal cell and the presence of specialized setae on the apicomedial clypeal apex in females. The Pemphredoninae are in need of critical cladistic analysis. In the Stigmina the analysis of charac- ters resulted in three equally parsimoni- ous solutions. Electronic analysis of the data would have undoubtedly resulted in many more. The cladogram (Figure 51) was selected on the basis of the best fit with plate tectonics. The recognition of new genera was based on the cladistic analysis coupled with biogeography and morphology. Morphological distinctions for genera are consistent with those for genera in Bohart and Menke (1976). A dendrogram placing Pamcrabro in a posi- 264 Journal of Hymenoptera Research Fig. 53. Distribution of 27 species of Stigmina in the Caribbean. tion basal to Carinostigmus and Stigmus was presented by Bohart and Menke (1976). The presence of a number of apo- typic character states suggest that the ar- rangement presented here with Carinostig- mus in the basal position, is correct. These character states include the presence of an acetabular carina, a clypeus with four teeth on the apical margin, two to four spines on the posterior surface of the hind tibia, and an elongate digitus that greatly exceeds the length of the cuspis. A composite distribution map of the Stigmina is presented in Fig. 52, Carinos- tigmus is plotted against all other genera in the Stigmina. The presence of genera on two southern hemisphere continents and their absence on other continents is indic- ative of a group radiating at the time of continental separation. Carinostigmus is re- stricted to the Old World, generally south- ern hemisphere (based on large numbers of undescribed species from southern regions) but is absent from Australia (an introduced species is present in New Guinea). The other genera occur in the western hemisphere {Araucastigmus, Ay- khustigmus, Incastigmus, Llaqhastigmus, Par- astigmus, Stigmus), across the Palaearctic Region with several outliers on the east asian islands and the orient (Stigmus, Tzustigmus), and Australia (Paracrabro). The only major zone of contact between Carinostigmus and the other genera is in the Oriental Region; a minor contact zone occurs in southern Spain. Elsewhere Cari- nostigmus is separated from the other gen- era by the Himalayas, the Gobi and Sahara Deserts, and the Atlantic Ocean. Outliers in Chile, India, Thailand, Japan, Taiwan, Vietnam, Sumatra and Australia represent distributions of relict genera and species. The only anomaly is the presence of Carinostigmus in New Guinea which probably represents a recent introduction. The Stigmina are restricted in distribution by 2 primary barriers, those of xeric con- ditions and water. The effect of xeric con- Volume 4, 1995 265 ditions is evident in Fig. 52 with the ab- sence of genera from major desert regions (some exceptions in the New World). The effect of water is evident in the Caribbean fauna (Fig. 53). This composite distribu- tion map of 27 species shows only 7 spe- cies on the Caribbean Islands, of those six occur within the area influenced by the Orinoco River current where 3 are endem- ic (including Incastigmiis thoraciciis). A fourth endemic species (an undescribed Stigmus) occurs on Hispaniola. Many of the islands of the Lesser Antilles are with- in sight of neighbouring islands. The fail- ure of the Stigmina to traverse these short distances is indicative of the effectiveness of water as a barrier to dispersal and also to the lack of long distance dispersal mechanisms within this group. This point must be born in mind when attempting to interpret the present distribution patterns of genera. If there was not, at some time in the past, an overland route between points presently separated by water, then it is unlikely that species of Stigmina would disperse between those points. In South America Stigmus has a C- shaped distribution restricted to the high- lands surrounding the Amazon basin and restricted by the lowlands of the Rio Tir- ana in the south. Araucnstigmus (Map 35) is restricted to Chile, for the most part to the Pacific side of the continental divide and to the north restricted by extreme des- ert conditions. There are 3 species, and thanks to Lubomir Masner, an abundance of specimens. Parastigmus (Map 34) occurs on the east side of the divide although 1 species, the most plesiotypic in the genus occurs in Chile, possibly indicating a com- mon regional origin. In Argentina Paras- tigmus appears to follow the distribution of grassy steppe vegetation surrounding the xeric vegetation of Patagonia. The ge- nus is represented by 4 species known from 16 specimens. Aykhustigmus (Map 36), represented by 4 species known from 37 specimens, occurs in the Brazilian Highlands south to the Rio Pirana and the 15.5°C mean July isotherm. The only ex- ception is a single specimen recently col- lected from the Venezuelan state of Zulia believed to be an introduction. Llaqhastig- mus (Map 37) follows a distribution simi- lar to but not as extensive as Stigmus. Thir- teen species are confined to the Andes while a single species occurs across the Brazilian Highlands. Llaqhastigmus is not known from the Guyana Shield but could well occur there. Incastigmus with 25 spe- cies is not only the largest Neotropical ge- nus but has the broadest distribution. It is limited in the south by the 15.5°C mean July isotherm, on the west by xeric con- ditions and to the north by the 10°C mean January isotherm which is the approxi- mate boundary of the Nearctic and Neo- tropical Regions. The composite map (Fig. 54) demonstrates the limiting influences of winter temperatures and rainfall on the distribution of Stigmina in the Neotropics. About 200 million years ago a single landmass stretched from pole to pole, Laurasia to the north, Gondwanaland to the south. The central part of this great landmass was dominated by xeric condi- tions. Areas of rainforest existed in the ex- treme south (southern South America, South Africa, Antarctica and Australia), seasonal or monsoon rains existed in the mid latitudes (southeast Asia, Arabian Peninsula) (Lamb 1977). Superimposed on these extremes of wet-dry conditions were areas of exposed basement layer or granite shield characterized by localized or edaphic aridity (Axelrod 1972). Similar conditions occur today in the Tipui system of the Guiana Shield. The predrift uplift- ing of the continents along rift lines al- lowed water to flow east and west from the South American-African divide there- by slowly exposing the basement layer through erosion. The exposure of the base- ment layer in this region has probably had more consequences for floral and faunal radiation than any other single post drift event. The Stigmina belong to the apoid lin- 266 Journal of Hymenoptera Research ,.i_--=c:s;^ ..■^> 10°C January mean ^^F'^'-^^- >•• • >•• .^, %. NORTH, CENTRAL, AND SOUTH AMERICA 500 I 15.5°C July mean ^ 10mra rainfal 1 MILES ^^ <=^ Fig. 54. Composite distribution of Incastigmus, Araucastigmus, Parastigmus and Aykhustigmus in the Neotrop- ical region. eage (Apoidea) of Hymenoptera, of which bees form the dominant component. There is a well known coevolutionary relation- ship between bees and angiosperms. The earliest flowers were probably beetle pol- linated (Michener 1979). Bees were prob- ably an already diversified west gond- wanan group of sandnesting lineages (Michener 1979) by the time the first flow- ering plants (Magnoliaceae) spread south- ward from Laurasia (Raven and Axelrod 1974). Michener (1979) for bees, and Ax- elrod (1972) and Raven and Axelrod (1974) for angiosperms hypothesize a xeric west Gondwanaland radiation of these groups. The influences of interior xeric conditions, thousands of island-like patch- es of arid exposed basement layer, along with a group of wasps (bees) preadapted for pollination contributed to an explosive radiation of bees and angiosperms. Conversely the southern rainforests pre- Volume 4, 1995 267 sented a habitat unsuitable for sand nest- ney and likely introduced Gondwanaland ing organisms perhaps because of high flora and fauna into Southeast Asia. If this water tables, high humidity and the likely tectonic event occurred, then gondwana- persistence of fungi in brood cells. I hy- land flora and fauna (Stigmina) could be pothesize that twig-nesting behaviour de- introduced to the Laurasian continent at a veloped early in the apoid lineages as a very early period. This would allow the response to increasingly wet conditions to lineages of Stigmina to be in a position to the south. It is also likely that twig-nesting colonize the Nearctic Region via Europe developed more than once since there are and Beringia at 65 million B.P. several apparently unrelated types of ae- The rifting of South America from Af- rial nest architecture. It is the twig nesting rica commenced during the same period groups that reach the greatest diversity in (180 million B.P.) and resulted in the sep- the rainforests today. Lineages of bees, aration of southern Africa from southern and the Stigmina likely radiated from one South America by a narrow body of wa- or the other of the two extreme gondwan- ter. The inability of the Stigmina to cross aland ecosystems. Modern distributions a water barrier is instrumental in my ex- show presumed relict bees inhabiting planation of the absence of genera other south temperate xeric areas while pre- than Carinostigmus from Africa. Although sumed relict Stigmina inhabit south wet India likely once held more genera of Stig- temperate or tropical rainforest. Based on mina than it does today, that fauna was present-day distribution and on the hy- almost certainly lost when the craton trav- pothesized relationships between genera, elled over a hot spot in the crust on its the Stigmina probably existed prior to northward journey. Severe volcanism on continental breakup, over 200 million the Indian plate likely caused extinction of years ago. Some support is lent to this biota in virtually all its ecosystems. What statement by the existence of specialized fauna the plate brought to Asia is proba- pemphredonine fossils in the Upper Cre- bly the result of a brief encounter with taceous, about 135 million B.P. (Evans north Africa before colliding with Asia 1969, 1973). Stigmina at the time of conti- about 45 million years B.P. By the late Cre- nental breakup would likely have oc- taceous Period (135 million years B.P.) curred in the southern rainforests (south- Stigmus and Tzustigmiis probably colo- ern South America, South Africa, Antarc- nized Laurasia. The Nearctic fauna con- tica, Australia) and in the southern savan- sists of eastern and western species na surrounding the midcontinental xeric groups likely the result of colonization by area. Stigmus via both Europe and Beringia Of pivotal importance in this analysis is about 65 million years B.P. a report on the tectonics of the Indonesian Based on relationships presented in the region (Hamilton 1979). Hamilton pre- cladogram and present distributions it is sents evidence of Jurassic (180 million likely that the Stigmina originated in the B.P.) rifting of eastern Gondwanaland. southern hemisphere continents, especial- Present-day land masses of Thailand, Bur- ly Antarctica (Fig. 52) over 200 million ma, Malaysian Peninsula, Sumatra and years B.P. Australia was colonized once Borneo rifted from the northern Australia- (Paracrabro). South America was colonized New Guinea-Antarctica continental mar- at a very early stage (180 million B.P.) on gin and moved north to collide with the at least 3 separate occasions, and more re- Eurasian plate. Furthermore Audley- cently by Stigmus from the Nearctic Re- Charles (1987) found an uninterrupted gion about 5 million B.P. North America pollen record indicating the craton was was colonized at least twice, 3 times if one not submerged during its northward jour- includes the incursion of Neotropical fau- 268 Journal of Hymenoptera Research na (Incastigmus) into Central America about 5 million B.P. The Palaearctic was colonized at least twice; finally the Ori- ental Region was colonized on at least 2 occasions. The great diversity of the Neotropical fauna is due to 3 major factors. The wet adapted fauna that colonized it, periodic aridity that likely restricted that fauna to islands of forest, and mountains that al- lowed extremes in climatic conditions to exist in close proximity. The xeric areas of southern Argentina afford the best oppor- tunity to collect relict stigmine species in the New World. By contrast the relatively depauperate nature of the Afrotropical fauna may be the result of a xerically adapted original fauna succumbing to the spread of mesic conditions over the entire continent, in- cluding the Sahara (Axelrod and Raven 1978). There was no, or very little, of the faunal island effect that occurred in South America. An increasingly wet Africa prob- ably afforded little opportunity for the xerically adapted fauna to survive. Africa is again in a dry phase with rainforest re- stricted to very limited areas. The more equitable continental outliers such as the mesic islands of the Comoros or Mada- gascar probably still harbour remnants of relict faunas. The interactions in the Oriental Region are complex. Most of the relict taxa are found there. Species of relict groups have highly localized distributions thus ac- counting for their relative absence in most collections. There is a possibility that their distributions are tied to forest over lime- stone terrain. Whitmore (1984) reported that limestone terrain is not common in Malaysia, but the flora of a mainland for- est over limestone formation had 23% of its species endemic to the country with 11% endemic to limestone. Islands like Borneo or Sulawesi are melange forma- tions (Hamilton 1979) resulting from near- by subduction zones indicating that lime- stone has likely been present throughout the islands history. Although no habitat information is available for the relict fau- na, it is likely, because of the high degree of endemic flora that limestone forest may offer fertile grounds for collecting speci- mens of these ancient apoid wasp lin- eages. Little mention has been made of the or- igins of Cnrinostigmus. Two scenarios exist based on the cladogram, present distri- bution, and the number of undescribed species. In the first scenario I hypothesize an African origin with the genus in place at the time of rifting from South America or shortly thereafter. In the second scenar- io I hypothesize an Oriental origin with more recent colonization of Africa. The Oriental Region is a mix of land masses of Laurasian and Gondwanaland origin, and most species of Carinostigmus occur there (mostly undescribed). If a cladistic analy- sis of species within Carinostigmus were to demonstrate a Laurasian origin then the age of lineages in the Stigmina could be moved back to a Pangean origin about 230 million B.P. It would also imply a north- south Pangean split of lineages within the Stigmina with Carinostigmus of northern (Laurasian) origin and all other genera of southern origin (Gondwanaland). Such a scenario is compatible with the cladogram presented here and with ecosystem recon- structions of that period. ACKNOWLEDGEMENTS I wish to thank the reviewers of the manuscript, R.M. Bohart, A.S. Menke, and W.J. Pulawski, for their time and the numerous suggestions for improve- ments to the manuscript. LITERATURE CITED Antropov, A.V. 1992. Pamstigmus, a new genus of the subtribe Stigmina from South America (Hy- menoptera: Sphecidae: Pemphredoninae). Linzer biologische Beitriige 24:97-101. Arnold, G. 1924. The Sphegidae of South Africa, part V. Annals of tlic Transvaal Museum 11:1-73. Ashmead, W.H. 1900. Report upon the aculeate Hy- menoptera of the islands of St. Vincent and Gre- nada, with additions to the parasitic Hymenop- tera and a list of the described Hymenoptera of Volume 4, 1995 269 the West Indies. Transactions of the Entomological Society of London (1900):207-367. Audley-Charles, M.G. 1987. Dispersal of Gondwa- naland: relevance to evolution of the Angio- sperms. pp. 5-25. In: Whitmore, T.C. (editor). Bio- geographical Evolution of the Malay Archipelago. Clarendon Press, Oxford, 147 pp. Axelrod, D.I. 1972. Edaphic aridity as a factor in an- giosperm evolution. American Naturalist 106:311- 320. Axelrod, D.I. and P.H. Raven. 1978. Late Cretaceous and Tertiary vegetation history of Africa. Pp. 77- 130. In: M.J. A. Werger (editor) Biogeography and Ecologi/ of Southern Africa. Dr. W. Junk, The Hague. Bohart, R.M. and A.S. Menke. 1976. Sphecid wasps of the world, a generic revision. University of California Press, Berkeley, 695 pp. Budrys, E.R. 1987. [Roiushchie osy rodov Stigmus Panzer and Carinostigmus Tsuneki (Hymenop- tera, Sphecidae) dal'nego Vostoka SSSR] pp 49- 56. In: P.A. Lehr and N.A. Storoshev (editors). Novye Dannye pa Sistematike Nasekomykh Dal'nego Vostohi. Biological Pedological Institute, Far East Section, Academy of Science. Cameron, P. 1891. Hymenoptera, Vol. 2:129-176, pis. viii-x: Cerceris to Priocnemis. In: Godman, F.D. and O. Salvin; Biologia Centrali-Americana pts. xci-xcviii. Eickwort, G.C. 1967. Aspects of the biology of Oiil- icola ashmeadi in Costa Rica, journal of the Kansas Entomological Society 40:42-73. Evans, H.E. 1969. Three new Cretaceous aculeate wasps. Psyche 76:251-261. Evans, H.E. 1973. Cretaceous aculeate wasps from Taimyr, Siberia. Psyche 80:166-178. Fox, W.J. 1892. The North American Pemphredoni- dae. Transactions of the American Entomological So- ciety 19:307-326. Fox, W.J. 1897. Contributions to the knowledge of the Hymenoptera of Brazil. No. 3. Sphegidae (sens. lat.). Proceedings of the Academy of Natural Sciences of Philadelphia (1897):373-388. Hamilton, W. 1979. Tectonics of the Indonesian Re- gion. United States Geological Survey Professional Paper 1078, 345 pp. Iwata, K. 1964. Bionomics of non-social wasps in Thailand. Nature and Life in Southeast Asia 3:323- 383. Janvier, H. 1962. Recherches sur les Hymenopteres nidifiants aphidivores. Annates des Scieiu^es Na- turelles. Zoologie (12) 4:489-516. Kohl, F.F. 1890. Zur Kenntniss der Pemphredonen. Annalen des K.K. Naturhistorischc}! Hofnnisetims 5: 49-65. Kohl, F.F. 1892. Neue Hymenopterenformen. A)uia- len des K.K. Naturhistorischen Hofmuseums 7:197- 234, pis. xiii-xv. Kohl, F.F. 1905. Hymenopterentypen aus der neotro- pischen Fauna. Verliandlungen der k.-k. zoologisch- botanischen Gesellschaft in Wien :338-366. Krombein, K.V. 1956. Miscellaneous prey records of solitary wasps. II. Bulletin of the Brooklyn Ento- mological Society 51:42-44. Krombein, K.V. 1958a. Miscellaneous prey records of solitary wasps. III. Proceedings of the Biological So- ciety of Washington 71:21-26. Krombein, K.V. 1958b. Additions during 1956 and 1957 to the wasp fauna of Lost River State Park, West Virginia, with biological notes and descrip- tions of new species. Proceedings of the Entomolog- ical Society of Washington 60:49-64. Krombein, K.V. 1961. Miscellaneous prey records of solitary wasps. IV. Bullethi of the Brooklyn Ento- mological Society 56:62-65. Krombein, K.V. 1963. Natural history of Plummers Island, Maryland. XVII. Annotated list of the wasps. Proceedings of the Biological Society of Wash- ington 76:255-280. Krombein, K.V. 1973. Notes on North American Stig- mus Panzer (Hymenoptera: Sphecoidea). Proceed- ings of the Biological Society of Washington 86:211- 230. Krombein, K.V. 1984. Biosystematic studies of Cey- lonese wasps, XIV: A revision of Carinostignnis Tsuneki (Hymenoptera: Sphecoidea: Pemphre- donidae). Smithsonian Contributions to Zoology No. 396, 37 pp. Lamb, H.H. 1977. Clinuite Present, Past and Future. Vol. 2 Climatic History and the Future. Methuen & Co. Ltd. London, Barnes & Noble Books, New York 835 pp. Leclercq, J. 1959. Pemphredoninae. Pare. Natl. Upemba, I. Mission G.F. Witte, fasc. 53(2):17-62. Mantero, G. 1901. Descrizione di alcune specie nu- ove di Imenotteri scavatori provenienti dal Ri'o Santa Cruz in Patagonia. Bullet tino delta Societa entomologica italiana 33:197-203. Menke, A.S. 1989. Arpactophilus reassessed, with three bizarre new species from New Guinea (Hy- menoptera: Sphecidae: Pemphredoninae). biver- tebrate Taxonomy 2:737-747. Michener, CD. 1979. Biogeography of the bees. An- nals of the Missouri Botanical Garden 66:277-347. Panzer, G.W.F. 1804. Faunae Insectorum Germanicae initiae oder Deutschlands Insekten. Heft 86, 24 plates (dating after Sherborn 1923). Rau, P. 1928. Field studies in the behaviour of the non-social wasps. Transactions of the Academy o} Sciences of St. Louis 25:325^89. Raven, P.H. and D.I. Axelrod. 1974. Angiosperm bio- geography and past continental movements. An- nals of the Missouri Botanical Garden 61:539-673. Richardson, C.H. 1915. An observation on the breed- ing habits of Stigiinis conestogorum Rohwer. Psy- che 22:104-105. ' 270 Journal of Hymenoptera Research Rohwer, S.A. 1911. Descriptions of new species of wasps with notes on described species. Proceed- ings of Die United States National Museum 40:551- 587. Sherborn, CD. 1923. On the dates of G.W.F Panzer's "Fauna Insect. German.", 1792-1844. The Annals and Magazine of Natural History (9) 11:566-568. Smith, M.R. 1923. Unusual damage to the floors of a house by a species of pemphredonid wasp, Stigmus fidvicornis Rohwer. Journal of Economic Entomology 16:553-554. Strand, E. 1911. Hymenoptera aus Peru und Equa- dor. Archiv fiir Naturgeschichte 77 Bd 1 suppl. H. 2:141-157. Tsuneki, K. 1954. The genus Stigmus Panzer of Eu- rope and Asia with descriptions of eight new species (Hymenoptera, Sphecidae). Memoirs of the Faculty of Liberal Arts, Fukui University Series 2, Natural Science 3:1-38. Tsuneki, K. 1963. Chrysididae and Sphecidae from Thailand (Hymenoptera). Etizenia 4:1-50. Tsuneki, K. 1966. Contribution to the knowledge of the Pemphredoninae fauna of Formosa and the Ryukyus (Hymenoptera, Sphecidae). Etizenia 14: 1-21. Tsuneki, K. 1970. Gleanings on the bionomics of the east- Asiatic non-social wasps. V. Some species of Pemphredoninae. Etizenia (42):l-20. Tsuneki, K. 1971. Studies on the Formosan Spheci- dae (XIII). A supplement to the subfamily Pem- phredoninae (Hym.) with a key to the Formosan species. Etizenia 57:1-21. Turner, R.E. 1907. New species of Sphegidae from Australia. The Annals and Magazine of Natural His- tory (7) 19:268-276. Wasbauer, M.S. and W.E. Simonds. 1964. A note on the prey and nest structure of Stigmus inordinatus inordinatus. The Pan-Pacific Entomologist 40:114- 116. Whitmore, T.C. 1984. Tropical Rainforests of the Far East. Second Edition. Clarendon Press, Oxford, 352 pp. Yasumatsu, K. and C. Watanabe. 1964. A tentative catalogue of insect natural enemies of injurious insects in Japan. Part I. Parasite-predator host catalogue. 166pp. Entomology Laboratory of the Faculty of Agriculture, Kyushu University, Fu- kuoka, Japan. Volume 4, 1995 271 Table 1. Data matrix for the Stigmina Carino Ti-ustig Arjuca Parac Parastig Aykhus Stignitis Llaqh Incast 1. Palpal formula 2. Labrum shape 3. Mandibular socket 0 1 1 0 4 1 0 4 1 0 4 1 0 4 1 0 4 1 0 4 1 0 3 1 0 4 1 4. Apical manibular teeth 03 29 06 29 06 2? 06 49 06 09 06 29 06 0,29 26 29 26 29 5. Apicoventral tooth 6. Clypeal bevel 7. Clypeus size 8. Clypeal teeth 9. Interantennal tubercle 0 0 0 2 1 0 0 0 3 1 0 0 0 4 0 na 0 1 5 1 na 0 1 4,5 0 0 0 0 4 0 0 1 0 2,4 0 1 0 0 4 0 0 0 0 4 0 10. Frontal carina 2 0 2 2 0,2 0 0 0 0 11. Vertex micropore field 12. Eye margins 13. Occipital carina 14. Occi.-hypo. carina 15. Pronotal collar 1 1 0,1 2 1 1 1 0,1 2 0 1 0 0 2 0 1 1 1 2 0 1 0 0 2 0 2 1 0,1 2 0 2 1 0,1 1,2 0 2 1 0,1 2 0 2 1 0,1 2 0 16. Transverse pronotal carina 0 0 0 0 0 0 0 0 0 17. Median scutal groove 18. Notaular grooves 19. Acetabular carina 0,1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0,1 0 1 0 0 1 0 0 1 1 1 1 20. Omaulus 0 0 0 0 0 0 0 0 0 21. Epistemal sulcus 22. Hypersternaulus 23. Orientation of hypersternaulus 24. Scrobal sulcus 1 0 0 0,1 1 0 0 0,1 1 0 0 1 1 0 0 0 1 0 0 0,1 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 25. 6 midbasitarsus 0 0 0 2 1 1 2 0 0 26. Hindtibial spines 27. Propodeal enclosure 28. Stigma size 2 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 29. Stigma shape 30. Stigma pore field 31. Marginal cell size 32. Marginal cell veins 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 33. Submarginal cells 34. Submarginal position 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 35. Submarginal veins 0 0 0 0 0 0 0 0 0 36. Discoidal cell number 1 1 1 1 1 1 1 1 1 37. Media divergence 2 1 0 0 0 0,1 0 0 0 38. Submedian cell 1 0 0 0 0 0 0 0 0 39. Hind wing cells 0 0 0 0 0 0 0 0 0 40. Petiole length 41. Petiolar sculpture 2 2 2 0,2 2 0 2 0 2 0 2 0 2 0 2 0 2 0 42. Sternum I carinae 0 0 0 0 0 0 0 0 0 43. SI basal groove 0 0 0 0 0 0 0 0 0 44. Tl lateral carina 1 1 1 1 1 1 1 1 1 45. S2 microsetae 1 1 1 1 1 1 1 1 1 46. Pygidial Plate 47. Male Sternum 8 1 3 1 3 1 3 0 3 1 3 1 3 1 3 1 3 1 3 48. Genitalia 0 2 0 2 2 2 2 2 2 49. Microsculpture 1 0,1 1 0 0 0 0 0 0 0 = plesiotypic character state, 1-6 = apotypic character states, na = not applicable, Carino = Carinostigmus, Tzustig = Tziistigtmis, Arauca = Araucastigiiiiis, Parac = Paracrabro, Parastig = Parastigmus, Aykhus = Ay- kliustigmiis, Llaqh = Uaqluistigniiis, Incast = Incastigmns. 272 Journal of Hymenoptera Research Table 2. Data matrix for the Pemphredonina Diodontus Pemphredon Passalcecus Polemistus 1. Palpal formula 0 0 0 0 2. Labrum shape 3 1 2 1 3. Mandibular socket 1 1 1 1 4. Apical manibular teeth 0,1 2, 3, 4, 5 0,2 0,1 5. Apicoventral tooth na 0,1 0 na 6. Clypeal bevel 0 0 0 0 7. Clypeus size 0 0 0 0 8. Clypeal teeth 3 1, 2, 3 1,2,3 1 9. Interantennal tubercle 0 0 0 1 10. Frontal carina 2 2 0 2 11. Vertex micropore field 0 0 0 0 12. Eye margins 0 0 0 1 13. Occipital carina 0 0 0 1 14. Occi.-hypo. carina 1 1 2 2 15. Pronotal collar 0 0 0 0 16. Transverse pronotal carina 0 0 0 0 17. Median scutal groove 0 0 0 0 18. Notaular grooves 0 0 0 1 19. Acetabular carina 0 0 0 0 20. Omaulus 0 0 1 0 21. Episternal sulcus 0 0 0 0 22. Hypersternaulus 0 0 0 0 23. Orientation of hypersternaulus 0 0 1 1 24. Scrobal sulcus 0 0 0 0 25. 6 midbasitarsus 1 1 1 1 26. Hindtibial spines 0 0 2 2 27. Propodeal enclosure 0 0 1 0 28. Stigma size 0 0 0 0 29. Stigma shape 0 0 0 0 30. Stigma pore field 0 0 0 0 31. Marginal cell size 0 0 0 0 32. Marginal cell veins 0 0 0 0 33. Submarginal cells 0 0 0 0 34. Submarginal position 0 0 0 0 35. Submarginal veins 0 0 0 0 36. Discoidal cell nvimber 0 0 0 0 37. Media divergence 0 0 0 0 38. Submedian cell 0 0 0 0 39. Hind wing cells 0 0 0 0 40. Petiole length 1 2 1 2 41. Petiolar sculpture 0 1 0 0 42. Sternum I carinae 1 1 1 1 43. SI basal groove 0 0 0 0 44. Tl lateral carina 0 0,1 0 0 45. S2 microsetae 0 0 0 0 46. Pygidial Plate 0 1 2 2 47. Male Sternum 8 1 1 3 3 48. Genitalia 0 0 0 0 49. Microsculpture 0 0,1 0 0 0 = plesiotypic character state, 1-6 = = apotypic character states, na = not applicable. Volume 4, 1995 273 Table 3. Data matrix for the Spilomenina Arpnclop\u\iiy- Spilonicna Microstigmu^ Xxi^-nia 1. Palpal formula 2 1 2 2 2. Labrum shape 3,4,5 4 0? 0? 3. Mandibular socket 1 1 1 1 4. Apical manibular teeth 0 0 0 0 5. Apicoventral tooth na na na na 6. Clypeal bevel 0 0 0 0 7. Clypeus size 1 1 1 1 8. Clypeal teeth 2 2 2 1 9. Interantennal tubercle 0 0 0 0 10. Frontal carina 2,3 0,1 2 2 11. Vertex micropore field 0 0 0 0 12. Eye margins 1 0 0 0 13. Occipital carina 2 2,3 3 3 14. Occi.-hypo. carina 2 na na na 15. Pronotal collar 0 0 0 0 16. Transverse pronotal carina 0 0 0 0 17. Median scutal groove 0 0 0 0 18. Notaular grooves 0 0 0 0 19. Acetabular carina 0 0 0 0 20. Omaulus 0 0 0 0 21. Episternal sulcus 1 1 1 1 22. Hypersternaulus 0 0,1 0 1 23. Orientation of hypersternaulus 1 1 1 na 24. Scrobal sulcus 0,1 1 1 1 25. 6 midbasitarsus -- 1 26. Hindtibial spines 2 2 2 2 27. Propodeal enclosure 1 0 1 1 28. Stigma size 1 1 1 1 29. Stigma shape 0 0 0 0 30. Stigma pore field 0 0 0 0 31. Marginal cell size 0 0 0 0 32. Marginal cell veins 0 0 0 1 33. Submarginal cells 0,1 0 1 2 34. Submarginal position 0 0 0 0 35. Submarginal veins 0 0 0 1 36. Discoidal cell number 1 1 1 2 37. Media divergence 0 0 0 na 38. Submedian cell 0 0 0 na 39. Hind wing cells 0 0 0 1 40. Petiole length 0 0 0 0 41. Petiolar sculpture na na na na 42. Sternum I carinae 2 2 2 2 43. SI basal groove 0 0 0 0 44. Tl lateral carina 0 0 0 0 45. S2 microsetae 0 0 0 0 46. Pygidial Plate 2 2 2 2 47. Male Sternum 8 2 48. Genitalia 1 49. Microsculpture 0 0 0 0 0 = plesiotypic character state, 1-6 = apotypic character states, na = not applicable, blank cells indicate specimens were unavailable and character state could not be determined from literature. 274 Journal of Hymenoptera Research Table 4. Data matrix for the Ammoplanina Pulv Ammops Ammous Paramo Ammoplus Timber 1. Palpal formula 1 1 1 1 1 1 2. Labrum shape 0? 0? 3 3 3 3 3. Mandibular socket 1 1 1 1 1 1 4. Apical manibular teeth 0 0 0 0 0 0 5. Apicoventral tooth na na na na na na 6. Clypeal bevel 0 0 0 0 0 0 7. Clypeus size 0 0 1 1 1 1 8. Clypeal teeth 2 2 2 2 2 2 9. Interantennal tubercle 0 0 0 0 0 0 10. Frontal carina 0,1 1 0 1 0 1 11. Vertex micropore field 0 0 0 0 0 0 12. Eye margins 0 0 0 0 0 0 13. Occipital carina 2 2 2 2 0 3 14. Occi.-hypo. carina 2 1 1 1 1 na 15. Pronotal collar 0 0 0 0 0 0 16. Transverse pronotal carina 0 0,1 1 1 1 1 17. Median scutal groove 0 0 0 0 0 0 18. Notaular grooves 0 0 0 0 0 0 19. Acetabular carina 0 0 0 0 0 0 20. Omaulus 1 1 1 1 1 1 21. Episternal sulcus 0 0 0 0 0 0 22. Hypersternaulus 1 1 1 1 1 1 23. Orientation of hypersternaulus na na na na na na 24. Scrobal sulcus 1 1 1 1 1 1 25. 3 midbasitarsus 1 1 1 0 0 26. Hindtibial spines 1 2 2 2 2 2 27. Propodeal enclosure 1 1 1 1 1 1 28. Stigma size 1 1 1 1 1 1 29. Stigma shape 0 0 1 1 1 1 30. Stigma pore field 0 0 0 0 0 0 31. Marginal cell size 1 1 1 1 1 1 32. Marginal cell veins 0 0 0 1 1 1 33. Submarginal cells 1 1 1 1 1 0,1,2 34. Submarginal position 0 1 0 0 0 0 35. Submarginal veins 0 0 0 0 0 0,1 36. Discoidal cell number 1 1 1 1 1 1 37. Media divergence 0 0 2 2 2 na 38. Submedian cell 0 0 0 0 0 na 39. Hind wing cells 0 0 0 0 0 1 40. Petiole length 0 0 0 0 0 0 41. Petiolar sculpture na na na na na na 42. Sternum I carinae 0 0 0 0 0 0 43. SI basal groove 1 1 1 1 1 1 44. Tl lateral carina 1 1 1 1 1 1 45. S2 microsetae 0 0 0 0 0 0 46. Pygidial Plate 0 0 0 0 2 2 47. Male Sternum 8 4 4 2 2 2 2 48. Genitalia 0 0 1 1 1 49. Microsculpture 0 0 0 0 0 1 0 = plesiotypic character state, 1-6 = apotypic character states, na = not applicable, blank cells indicate specimens were unavailable and character state could not be determined from literature. Pulv = Pulverro, Ammops = Ammoplmiops, Ammous = Ammoplanus, Paramo = Parammoplanus, Ammoplus = Ammoplanellus, Timber = Tunherlakena. Volume 4, 1995 275 276 Journal of Hymenoptera Research Map 5. Tzustigmus wuming I t ^ Map 7. Parastigmus huecuvus Map 8, Parastigmus n i na Volume 4, 1995 277 ^2^^=^=D^^ Map 9. Parast igmus patagonicus ^^2^"=^=^^ Map 11. Araucas t igmus masneri ^^F^=^^ Map 10, Araucast igmus glabratus ^2^=^=I>^ Map 12. Araucas t i gmus yani J lus 278 Journal of Hymenoptera Research Map 13. Paracrabro froggatti il?2^'=^^^ Map 15. Aykhustigmus patanawi Map \k. Aykhustigmus f ri tzi 0^"^=^^ %«^ a \ ly ' P Map 16. Aykhustigmus phasti %K ^ •» Volume 4, 1995 279 /?7=^^^^ Map 17. Aykhustigmus wa rawa ^^2^""^=^^^ Map. 19. Incastigmus thoraci cus ^F=^^^ Map 18. I ncast igmus i nt i ^^'^^^^ „. Map 20. Llaqhastigmus ambi guus 280 Journal of Hymenoptera Research ^^^2==^=^^ Map 23. Llaqhastigmus col omb! anus ^^^'^^^^ Map 22. Llaqhastigmus chut iyana ii'^^^^^^ Map Ik. Llaqhastigmus ecuatori al is Volume 4, 1995 281 Z!^^'^'^^^ Map 25. Ll aqhas t igmus jatunki rus ^PT^""^^^^ Map 27. Ll aqhas t igmus mantant i iJf^'^^i^J^ Map 26. Llaqhas ti gmus 1 1 utani s ^^^^^^^^ Map 28. Llaqhas tigmus mu thus 282 Journal of Hymenoptera Research /:^^'=^^^ Map 29. LI aqhast igmus nigricollaris ^ J /^^"^^^^ Map 31 . L laqhas t i gmus sapani s /^2='^=^^ Map 30. LI aqhast igmus santanderana ^'T^''^^^ Map 32. LI aqhast igmus shachus Volume 4, 1995 283 /?2'^^=D^^ Map 33. Llaqhast igmus sharkeyi ^=^=^^ Map Zk. Parast igmus ^^^'^^^ Map 36. Aykhust igmus 284 Journal of Hymenoptera Research iJ^=?^=^>5