fcXL- <<* Journal of Hymenoptera Research Volume 12, Number 2 October 2003 ISSN #1070-9428 CONTENTS BENNETT, A. M. R. A new genus and five new species of Neotropical Tryphoninae (Hymenoptera: Ichneumonidae) 209 BISCHOFF, I., K. FELTGEN, and D. BRECKNER. Foraging strategy and pollen preferences of Andrena vaga (Panzer) and Colletes cunicularius (L.) (Hymenoptera: Apidae) . . . 220 BRICENO G., R. A. Taxonomic revision of the genus Sesioctonus Viereck (Hymenoptera: Braconidae: Agathidinae) 238 HARTLEY, C. S. and R. W. MATTHEWS. The effect of body size on male-male combat in the parasitoid wasp Melittobia digitata Dahms (Hymenoptera: Eulophidae) 272 KULA, R. R. Morphological variation in Opius Wesmael (Hymenoptera: Braconidae) with an emphasis on nearctic species in the subgenus Gastrosema Fischer 278 PULAWSKI, W. J. Prosopigastra morogoro, a new species from Tanzania (Hymenoptera: Apoidea: Crabronidae: Larrini) 303 SHARKOV, A., T E. KATZNER, and T. BRAGINA. A new species of Copidosoma Ratzeburg (Hymenoptera: Encyrtidae) from eagle nests in Kazakhstan 308 SMITH, D. R. and D. H. JANZEN. Food plants and life histories of sawflies of the families Tenthredinidae and Pergidae (Hymenoptera) in Costa Rica, with descriptions of four new species 312 SMITH, D. R. and M. S. SAINI. Review of the southeastern Asian sawfly genus Eusunoxa Enslin (Hymenoptera: Tenthredinidae) 333 VILHELMSEN, L. Larval anatomy of Orussidae (Hymenoptera) 346 NOTE: AMIET, F and V. MAUSS. First report of male sleeping aggregations in the pollen wasp Celonites abbreviatus (Villers, 1789) (Hymenoptera: Vespidae: Masarinae) 355 INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991 OFFICERS FOR 2003 Lynn Kimsey, President Denis Brothers, President-Elect James B. Woolley Secretary John T. Huber, Treasurer E. Eric Grissell, Editor Subject Editors Symphyta and Parasitica Aculeata Biology: Mark Shaw Biology: Sydney Cameron Systematics: Donald Quicke Systematics: Wojciech Pulawski All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, Bohart Museum of Entomology, Department of Entomology, University of California, Davis, CA 95616; Secretary, Department of Entomology, Texas A&M University, College Station, Texas 77843; Treasurer, Eastern Cereal & Oilseed Research Centre, Agriculture Canada, K. W. Neatby Building, Ottawa, Ontario, Canada K1A 0C6; Editor, Systematic Entomology Laboratory, USDA, P.O. Box 37012, % National Museum of Natural History CE 520, MRC 168, Washington, D.C. 20013-7012. Membership. Members shall be persons who have demonstrated interest in the science of ento- mology. Annual dues for members are US$40.00 per year (US$35.00 if paid before 1 February), payable to The International Society of Hymenopterists. Requests for membership should be sent to the Treasurer (address above). Information on membership and other details of the Society may be found on the World Wide Web at http://IRIS.biosci.ohio-state.edu/ish. Journal. The Journal ofHymenoptera Research is published twice a year by the International Society of Hymenopterists, % Department of Entomology, Smithsonian Institution, Washington, D.C. 20560-0168, U.S.A. Members in good standing receive the Journal. Nonmember subscriptions are $60.00 (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: Twice a year. Location of Office of Publication, Business Office of Publisher and Owner: International Society of Hymenopterists, % Department of Entomology, Smithsonian Institution, 10th and Constitution NW, Washington, D.C. 20560-0168, U.S.A. Editor: E. Eric Grissell, Systematic Entomology Laboratory USDA, % National Museum of Nat- ural History, 10th and Constitution NW, Washington, D.C. 20560-0168. U.S.A. Managing Editor and Known Bondholders or other Security Holders: none. This issue was mailed 23 October 2003 j. HYM. RES. Vol. 12(2), 2003, pp. 209-219 A New Genus and Five New Species of Neotropical Tryphoninae (Hymenoptera: Ichneumonidae) Andrew M. R. Bennett Centre for Biodiversity and Conservation Biology, Royal Ontario Museum, 100 Queen's Park, Toronto, Ontario, Canada, M5S 2C6 [current address: Agriculture and Agri-Food Canada, K.W. Neatby Building, 960 Carling Avenue, Ottawa, Ontario Canada, K1A 0C6]; email: bennetta@agr.gc.ca Abstract. — A new genus of tryphonine ichneumonid, Boethella Bennett, n. gen. is described from the Neotropics. Synapomorphies are provided which support the sister group relationship of Boethella and Boethus Foerster (Tryphonini) (the latter known from the Holarctic, Neotropical and Ethiopian regions). Boethella darlingi Bennett, n. sp. is described from eastern Brazil, B. canilae Bennett, n. sp. (type species) is described from southern Mexico to eastern Brazil, B. hubleyi Bennett, n. sp. is described from southeastern Brazil, B. guidottiae Bennett, n. sp. is described from western and central Brazil and B. curriei Bennett, n. sp. is described from Peru. The latest survey of the family Ichneu- monidae (Yu and Horstmann 1997) listed 21,805 described extant species classified into 1485 genera. These genera are as- signed to 36 or 37 extant subfamilies de- pending on opinion (Wahl 1990, Yu and Horstmann 1997, Gauld 2000, Gauld and Wahl 2000). Some studies have attempted to elucidate the subfamily relationships (Wahl 1991, Wahl and Gauld 1998, Quicke et til. 2000); however, to date, no complete subfamily phylogeny is available for the Ichneumonidae. A recent study of one subfamily, the Tryphoninae, was under- taken in order to ascertain its placement within the family (Bennett 2002). In the process of this study, two new genera were discovered, one of which is de- scribed below. The other new genus from the tribe Oedemopsini will be described elsewhere allowing a full discussion of the characters and relationships of the 12 ex- tant oedemopsine genera. The subfamily Tryphoninae comprises 1170 described species (Yu and Horst- mann 1997) assigned to 53 described gen- era in seven tribes (Bennett 2002). The sub- family is cosmopolitan with its centre of diversity in the Holarctic Region (Kaspar- yan 1973). Tryphonines are koinobiont ec- toparasitoids of lepidopterans and saw- flies of the families Xyelidae, Tenthredi- nidae, Cimbicidae, Diprionidae and Argi- dae. They exhibit the uniquely derived trait of bearing eggs that travel down the outside of the ovipositor (Kasparyan 1973) (although this trait is unknown in the Idi- ogrammatini). The apical end of the stalk (which bears an anchor in most genera) travels down the inside of the ovipositor so that the body of the egg is suspended by the stalk ventral to the ovipositor. This anchor is pushed through the host integ- ument during oviposition, thereby hold- ing the egg in place until hatching. Fe- males of most genera of tryphonines have the habit of bearing an egg on their ovi- positor while searching for their hosts (Kasparyan 1973) which provides an in- controvertible character to identify them as tryphonines. Male tryphonines and fe- males that do not carry eggs externally during host searching (e.g. the Phytodie- tini) are more difficult to recognize be- cause the subfamily is relatively heter- ogenous in structure with some taxa re- 210 Journal of Hymenoptera Research sembling ctenopelmatines, whereas others resemble phygadeuontine cryptines, ban- chines or ophionines. During studies at the American Ento- mological Institute (AEIC), I found a series of Neotropical specimens set aside by Henry Townes labeled "New Genus near Bocthus". Two of these putative new spe- cies were included in a cladistic analysis of tryphonine generic relationships (Ben- nett 2002) to examine their status and their hypothesized relatedness to BoetJius. The cladistic analysis verified the generic sta- tus of this new genus as well as its sister group relationship with Boethus (see Ben- nett 2002, Bennett in prep, and below). The present paper describes this new tax- on as Boethella Bennett, n. gen., the 54th genus of the Tryphoninae (37th genus of the tribe Tryphonini). MATERIALS AND METHODS Specimens were borrowed from and are deposited in the American Entomological Institute, Gainesville, FL, USA (AEIC) (D. Wahl) and the Canadian National Collec- tion, Ottawa, Canada (CNCI) (J. Huber). Examination of other major ichneumonid collections (e.g. INBio Costa Rica and The Natural History Museum, London) did not reveal more specimens of this genus. Morphological terms follow Townes (1969) with some modifications: supra-an- tennal area for 'frons', supraclypeal area for 'face', malar space for 'cheek', epicne- mial carina for 'prepectal carina' and la- terotergites for 'epipleura'. MSI, MS2 re- fers to metasomal segments 1, 2 etc. Tl etc. refers to the tergites of metasomal seg- ments 1 etc. and SI etc. refers to the ster- nites of metasomal segment 1, etc. Wing venation terms follow the Comstock- Needham system as updated by Ross (1936) and incorporates the recommenda- tions of Goulet and Huber (1993) except for naming of the vein that forms the dis- tal edge of fore wing cell l + 2Rs (the 'ar- eolet' of Townes 1969) which is referred to as vein 3rs-m in accordance with Gauld (1997). TAXONOMIC PLACEMENT The strongest evidence that a taxon be- longs to the Tryphoninae is that its fe- males bear stalked eggs that travel down the outside of the ovipositor. The egg of Boethella is not known, therefore the place- ment of Boetliella in the Tryphoninae is not certain. Most adult tryphonines exhibit the following characters: clypeus with an api- cal fringe of hairs; cell 1 + 2Rs (areolet) of fore wing triangular to subrectangular; spiracle of Tl anterior to middle; Tl with a glymma present; dorsal valve of ovipos- itor high (not strongly tapered) and un- notched subapically. Of these characters, Boethella possesses the clypeal fringe of hairs (albeit sparse) and the high, un- notched ovipositor (Fig. 7). The latter character precludes the placement of Boe- thella in the Ctenopelmatinae. In addition, the areolet of the fore wing of Boethella (Fig. 1) resembles other tryphonines which have an open areolet (i.e., if the areolet were closed, it would be triangular, not pentagonal as in cryptines or ichneumon- ines or rhombic as in mesochorines). On the basis of these three characters, Boethella fits best in the subfamily Tryphoninae (compared to all other extant subfamilies), despite its apomorphic structure of Tl (i.e., glymma absent and spiracle posi- tioned posteriorly which is the condition found in cryptines and ichneumonines). It should be noted that Boethella is not the only genus of tryphonine with the latter two character states (e.g., Sphiuetus Grav- enhorst and Ankylophon Gauld). Boethella can further be assigned to the tribe Tryphonini because of its apomorph- ic pectinate tarsal claws that within the Tryphoninae are known only in the tribes Tryphonini, Exenterini (= the Exenterus group of genera, see Bennett 2002), Sphinctini and Phytodietini. Boethella does not possess any apomorphies that would place it in either of the latter two tribes Volume 12, Number 2, 2003 211 Fig. 1. Boethella canilae, holotype female, habitus. (e.g., the strongly pointed apical edge of the clypeus in the Sphinctini or the loss of propodeal carinae in the Phytodietini). Boethella also cannot be assigned to the Ex- enterini because it has paired tibial spurs on both the middle and hind legs (the Ex- enterini have the autapomorphies of only one spur on the middle leg and no spurs on the hind leg). The sister group relationship of Boethella and Boethus is supported by the following synapomorphies: occipital carina absent; epomia absent; notauli absent; fore wing vein 2m-cu with one bulla; fore wing vein 3rs-m absent (areolet open); Tl petiolate; spiracle of Tl in posterior 0.4. A complete cladistic analysis describing character po- larities and the relationships of all try- phonine genera including Boethus and Boe- thella is given in Bennett (2002) and Ben- nett (in prep.). Boethella can be distin- guished from Boethus by the former's possession of propodeal, epicnemial and submetapleural carinae (all of which are absent in Boethus). In addition, the glym- ma is absent in Boethella (present in Boe- 212 Journal of Hymenoptera Research thus). Boethella is exclusively Neotropical and the majority of species of Boethus are known from the Neotropical and southern Nearctic regions (Townes et ah 1992); how- ever, several Ethiopian species of Boethus are known (Scaramozzino 1991) as well as one Palaearctic species (Kasparyan 1973). Boethella Bennett, n. gen. Type species. — Boethella canilae Bennett, by original designation. Diagnosis. — Distinguished from other genera of tryphonines by the combination of: 1) occipital carina absent; 2) propodeal carinae present (Fig. 6). In addition, the ovipositor of Boethella is distinctive within the Tryphoninae (slightly upcurved with a high, wide, apically rounded dorsal valve that strongly overlaps the ventral valve medially) (Fig. 7). Description. — Fore wing length 2.9 to 4.8 mm; clypeus slightly rounded in profile, without a transverse line separating it into dorsal and ventral faces, apical margin truncate to slightly rounded in anterior view, strongly impressed laterally (Figs. 2 and 3) without medial paired tubercles (medial notch absent), clypeal fringe of hairs present, but sparse; malar space obliterated (mandibular socket contiguous with ventral edge of eye) (Fig. 2) except in B. darlingi space is 0.5 times basal width of mandible (Fig. 3); lower mandibular condyles separated by distance greater than distance of inner eye margins at level of clypeal foveae; mandible with teeth subequal in width and height, moderately convex in cross-section near base; labio- maxillary complex moderately elongate, glossae visible in anterior view in most specimens (Fig. 1); occipital carina absent; postgena without a tooth; supra-antennal area without a horn or carina; antennal sockets separated by distance greater than 0.5 diameter of socket; eyes without prom- inent setae; epomia absent (Fig. 1); dorso- posterior region of pronotum not strongly thickened in dorsal view; epicnemial ca- rina present, not dorsally curving toward anterior edge of mesopleuron (Figs. 1, 4- 5); auxiliary carina of mesopleuron either long and joining epicnemial carina (Figs. 4-5) or short and not joining (Fig. 1); ster- naulus present (Figs. 1, 4-5); subtegular ridge slightly curving out laterally, not produced into a vertical lamella that near- ly reaches tegula when tegula is down; no- tauli absent; projection on posterolateral edge of mesoscutum absent; carinae of scutellum present at base only; propodeal carinae all present except lateral longitu- dinal carinae absent (Fig. 6), medial lon- gitudinal carinae strongly raised, medial portion of posterior transverse carina weak; submetapleural carina present; fore tibia without an anterior, apical tooth; fore tibial spur evenly curved; middle and hind trochanters two segmented; middle and hind tibiae each with two spurs; tarsal claws pectinate to apex or nearly to apex (Fig. 1); fore wing vein 3rs-m absent (Fig. 1); fore wing vein 2m-cu weakly to strong- ly inclivous with one bulla (Fig. 1); wings hyaline to moderately infumate; Tl petio- late (Fig. 6) with spiracle at 0.6 to 0.75, dorsal longitudinal carinae absent (Fig. 6), dorsolateral longitudinal carinae present but not extending to spiracle (Fig. 1); glymma of Tl absent (Fig. 1) (slight de- pression present ventral to dorsolateral longitudinal carina in some specimens, but not a glymma); Tl and T2 not fused, their sculpture impunctate; SI not fused to Tl, membranous portion of SI not or only slightly projecting lateral to sclerotized portion of Tl; T2 without a transverse postmedian groove or oblique grooves de- lineating the anterolateral corners; latero- tergites of MS2 to MS4 separated from ter- gites by a complete crease; T6 to T8 not strongly turned anteriorly under T5; ovi- positor (only known in two species) short- er than apical depth of metasoma, mod- erately upcurved, dorsal valve thick and rounded apically, overlapping ventral valve laterally (Fig. 7). Mature larva. — Unknown. Egg. — Unknown. Volume 12, Number 2, 2003 213 Hosts. — Unknown. The sister genus Boe- Species included. — Five species (see key thus has been reared from argid sawflies and descriptions below). (Townes et al. 1992, Gauld 1997). Etymology. — Boethella is a modification Distribution. — Southern Mexico, Peru of Boethus (which means "helper" in and Brazil. Greek) indicating its close relationship with this genus. Its gender is feminine. KEY TO THE SPECIES OF BOETHELLA BENNETT 1 Malar space 0.5 times basal width of mandible (Fig. 3) (eastern Brazil) darlingi Bennett, n. sp. Malar space obliterated: dorsal edge of socket of mandible contiguous with ventral edge of eye (Fig. 2) 2 2(1) Mesopleuron with auxiliary carina short, not extending to epicnemial carina (Fig. 1) ... 3 Mesopleuron with auxiliary carina extending from anterior edge and joining epicnemial carina slightly ventral to ventral edge of pronotum (Figs. 4 and 5) 4 3(2) Hind tibia brown. (Brazil to southern Mexico) canilae Bennett, n. sp. Hind tibia with basal 0.7 yellow, apical 0.3 brown (Brazil — Santa Catarina) hubleyi Bennett, n. sp. 4(2) Epicnemial carina extending dorsal to point of union of auxiliary carina by at least the length of auxiliary carina (Fig. 4). T4 predominantly brown, yellow laterally and with a yellow medial longitudinal stripe or spot in some specimens (Peru) curriei Bennett, n. sp. Epicnemial carina extending dorsal to point of union of the auxiliary carina by much less than the length of auxiliary carina (Fig. 5), or not extending at all. T4 entirely yellow or yellow with a trace of brown in apical 0.2 and with a longitudinal brown or brown and white region in medial 0.3 (Brazil: Mato Grosso and Amazonas) .... guidottiae Bennett, n. sp. Boethella darlingi Bennett, n. sp. Fig. 3 Diagnosis. — Distinguished from other species of Boethella by having the malar space 0.5 times basal width of mandible (Fig. 3) (not zero times basal width). Female. — Unknown. Male (holotype). — Fore wing length 3.5 mm; medial part of apical edge of clypeus slightly convex in anterior view, without emargination; groove between clypeus and supraclypeal area weak laterally so that base of clypeus is relatively flat; malar space 0.5 times basal width of mandible; antenna with fourteen flagellomeres; aux- iliary carina of mesopleuron short, not joining epicnemial carina; abscissa of fore wing vein M between 3rs-m and 2m-cu greater than 0.75 length of 2m-cu; spiracle of Tl posterior to 0.7. Orange; antenna ex- cept anterior side of apical two to three flagellomeres, apical 0.2 of mandible, pos- terior of occiput medial to inner margin of eyes, area between ocelli, pronotum along dorsal edge, lateral lobes of mesoscutum and anterior 0.3 of medial lobe, scuto-scu- tellar groove, ventral edge of scutellum, mesopleuron except dorsoanterior quar- ter, mesosternum except small region me- dioposteriorly; dorsal and ventral edges of metapostnotum, anterior groove of pro- podeum including base of medial longi- tudinal carinae, ventroanterior corner of metapleuron; apical 0.8 of hind tibia, hind tarsus, wing veins and stigma, T2 except anterior 0.2, lateral 0.2 and posterior 0.2, 214 Journal of Hymenoptera Research 2. canilae 3. darlingi Figs. 2-3. Boethella spp. male, anterior view of head. 2, B. canilae, paratype. 3, B. darlingi, holotype. T3 to T7 except posterior 0.2 brown; clyp- eus, ventral 0.6 of supraclypeal area, ven- tral 0.5 of gena, propleuron, ventral 0.2 of pronotum, fore leg except coxa, entire middle leg, coxa, trochanter, femur and basal 0.2 of tibia of hind leg, Tl, anterior 0.2 and lateral 0.2 of T2 yellow; dorsal 0.8 of pronotum except dorsal ridge and me- dial region of petiolar region of propo- deum brownish orange; mouthparts ex- cept apical 0.5 of mandible, anterior side of apical two to three flagellomeres, fore coxa and metasomal sternites white; pos- terior 0.2 of T2 to T6 translucent white; gonoforceps light brown; wings strongly infumate basally, fading to hyaline apical- Material— Holotype 6 : BRAZIL, Rio de Janeiro State, Guanabara, Rio Grande, Re- presa, 1-31. iii. 1972 (Alvarenga) (AEIC). Etymology. — This species is named in honour of Dr. D. C. Darling, senior curator at the Royal Ontario Museum, in recog- nition of his long-standing appreciation of the magnificence of the family Ichneu- monidae. Comments. — Known only from the ho- lotype. Boethella darlingi may be the sister species of the other four species of Boe- thella because it lacks the obliterated malar space that is synapomorphic of these four species. Boethella canilae Bennett n. sp. Figs. 1-2, 6-7 Diagnosis. — Distinguished from other species of Boethella by the combination of all: 1) malar space obliterated (mandibular socket contiguous with ventral edge of eye) (Fig. 2); 2) auxiliary carina of meso- pleuron short, not joining epicnemial ca- rina (Fig. 1); 3) hind tibia brown. Female (holotype). — Fore wing length 4.1 mm; medial part of apical edge of clypeus slightly and broadly emarginate; groove between clypeus and supraclypeal area moderately strong laterally; malar space obliterated (dorsal edge of mandibular socket contiguous with ventral edge of eye) (Fig. 2); antenna with sixteen flagel- lomeres; auxiliary carina of mesopleuron short, not joining epicnemial carina (Fig. 1); abscissa of fore wing vein M between 3rs-m and 2m-cu less than 0.5 length of 2m-cu; spiracle of Tl anterior to 0.7. Yel- lowish orange; apical 0.2 of mandible, an- tenna except apical flagellomere, occiput posterolaterally, posteriorly and in a lon- gitudinal stripe extending posteriorly from between lateral ocelli, dorsal 0.8 of pronotum, tegula, anterior 0.5 of medial lobe mesoscutum and all of lateral lobes, ventral 0.7 of mesopleuron except anterior to epicnemial carina and in ventroposter- ior corner, mesosternum except medial 0.3, metanotum, hind tibia and tarsus, Volume 12, Number 2, 2003 215 4. curriei 5. guidottiae 6. canilae Figs. 4-6. Boethella spp. 4-5, Male, lateral view of mesopleuron: aux. = auxiliary carina, epic. = epic- nemial carina. 4. B. curriei, paratype. 5, B. guidottiae, paratype male. 6, B. canilae, holotype female, dorso- posterior view of propodeum and first metasomal segment. wing veins and stigma, spot occupying posterior 0.5 and lateral 0.3 of T2, T3 to T7 except posterior 0.2 and triangular-shaped medial portions of T4 to T7 brown; head (except antenna, mouthparts and occiput as noted above), ventral 0.2 of pronotum, posterior 0.5 of medial lobe of mesoscu- tum, scutellum, dorsal 0.3 of mesopleuron, anterior to epicnemial carina and in ven- troposterior corner orange; posterior 0.2 of T2, posterior 0.2 and lateral 0.2 of T3 to T7, medial triangular portions of T4 to T7, metasomal epipleura and sterna including vent. / dors./ hyp- sheath Fig. 7. Boethella canilae, holotype female, lateral view of posterior of metasoma showing ovipositor: vent. = ventral valve of ovipositor, dors. = dorsal valve of ovipositor, sheath = ovipositor sheath, hyp = hypo- pygium. hypopygium whitish yellow; dorsal valve of ovipositor, ovipositor sheath and mem- branes at base of ovipositor white; apical flagellomere light brown; membrane of wings hyaline with a trace of infumation apically in fore wing. Male. — Same as female except fore wing length 2.9 to 4.8 mm and antenna with fif- teen to eighteen flagellomeres. Colour var- iations: Dark morph as female, except all flagellomeres brown, occiput with less brown posterolaterally, apical 0.2 of hind tibia brown in some specimens and T2 completely brown except orange in a tri- angular region basomedially and yellow- ish white in posterior 0.2; gonoforceps light brown. Light morph as female except clypeus, supraclypeal area and orbits yel- lowish orange; occiput, pronotum, meso- scutum, scutellum and mesopleuron com- pletely orange; T2 entirely brown except anterolateral corners (or entire anterior 0.1) orange, posterior 0.2 yellowish-white; gonoforceps light brown. Inter morph as 216 Journal of Hymenoptera Research light morph except apical flagellomere only light brown at apex; lateral lobes of mesoscutum completely brown; dorsal 0.5 of pronotum orange brown. Material— Holotype 9, BRAZIL, Espir- ito Santo, Castelo, l-30.xi.1976 (M. Alvar- enga) (AEIC). Paratypes. 3 6, same data as holotype; 1 6, Sao Paulo State, Sao Jose do Barreiro, Serra da Bocaina, 22°37'59"S, 44°34'59"W, 1650 m, l-30.xi.1969, (Alvar- enga and Seabra) (AEIC); 1 6, MEXICO, Chiapas, 10 km south of Ocozocoautla, 1220 m, 2.viii.l962. (H. Milliron) (CNCI). Etymology. — This species is named in honour of Dr. C. Canil in recognition of her exemplary volunteer work at the Roy- al Ontario Museum. Comments. — Boethella canilae is quite var- iable in colour with the pronotum, meso- pleuron and mesoscutum ranging from predominantly brown to completely or- ange (in same collection site). The male specimen from Mexico is smaller than the other males (fore wing = 2.9 mm) but ap- pears to be conspecific (similar to the light morph males except that T2 and T3 are orange brown instead of brown). The specimen is in relatively poor condition and newer material from Mexico may re- veal structural differences that distinguish this population as a distinct species. Boethella guidottiae Bennett, n. sp. Fig. 5 Diagnosis. — Distinguished from other species of Boethella, by the combination of: 1) auxiliary carina of mesopleuron long, joining epicnemial carina, the latter ex- tending only slightly dorsal to point of union with auxiliary carina or not extend- ing (Fig. 5); 2) T4 predominantly yellow. Female (holotype). — Fore wing length 4.0 mm; medial part of apical edge of clypeus with a slight, narrow emargination; groove between clypeus and supraclypeal area moderately strong laterally; malar space obliterated (dorsal edge of mandib- ular socket contiguous with ventral edge of eye); antenna with seventeen flagello- meres; auxiliary carina of mesopleuron long, joining epicnemial carina, the latter extending only slightly dorsal to point of union with auxiliary carina (Fig. 5); ab- scissa of fore wing vein M between 3rs-m and 2m-cu less than 0.5 length of 2m-cu; spiracle of Tl anterior to 0.7. Yellow; basal 0.5 of mandible, apical five flagellomeres of antenna, medial 0.5 of propleuron, me- dial 0.2 of mesosternum, entire metapleu- ron, coxa and trochanter of fore and mid- dle legs, tarsus of fore leg whitish yellow; posterior 0.2 of T2 to T5 as well as lateral edges ventral to spiracle, medial triangu- lar region (widest in posterior) on T3 to T7, metasomal sternites, hypopygium ex- cept for a triangular stripe just ventral to dorsal edge, membranes around oviposi- tor white; scape, pedicel, apical 0.3 of hind tibia, hind tarsus, stigma, apical wing veins and narrow border around medial triangular regions of T3 to T7 yellowish brown; supra-antennal area, vertex, occi- put, dorsal 0.3 of gena, dorsal 0.5 of pro- notum and mesoscutum yellowish orange; apical 0.2 of mandibles, flagellum except apical five flagellomeres, basal wing veins brown; wing membrane weakly to mod- erately infumate anteriorly and apically, hyaline posteriorly and subapically. Male. — Similar to female except apical edge of clypeus slightly convex to truncate medially, (emargination present in only some male specimens); fore wing length 3.2 to 4.8 mm; antenna with sixteen to nineteen flagellomeres. Colour variations: Light morph similar to female except api- cal two to seven flagellomeres whitish yel- low and in most specimens, medial 0.5 of T6 and all of T7 yellowish brown; gono- forceps whitish yellow. Dark morph or- ange; clypeus, mouthparts except apical 0.6 of mandible, ventral 0.5 of pronotum, propleuron, metapleuron, propodeum, fore leg, middle leg, coxa, trochanter and femur of hind leg yellow; apical two to four flagellomeres and sternites of meta- soma whitish yellow; all structures that are brown in female are also brown in Volume 12, Number 2, 2003 217 dark morph male as are all wing veins and stigma; T2 to T7 vary from yellow to light brown, tending to be more brown medially and posteriorly, apical 0.1 to 0.2 of each tergite may be white or uniform with rest of tergite, some specimens are also similar to holotype with medial 0.3 to 0.5 of T3 to T7 with unpigmented, trian- gular regions which may indicate incom- plete sclerotization of these segments (a longitudinal, medial suture line is also present on the posterior tergites in these specimens); basal 0.7 to 0.8 of hind tibia may be yellow to brownish yellow (but base always lighter than apical 0.2 which is brown); gonoforceps yellowish orange to whitish brown. Material— Holotype 9, BRAZIL, Mato Grosso, Sinop, 12°31'S, 55°37'W, Malaise Trap, 1-31. x. 1974 (M. Alvarenga) (CNCI). Paratypes. 5 8 , same data as holotype ex- cept one from 1-31. x. 1976 and three 1- 31. xi. 1975; 3 8 , same data as holotype ex- cept 1-31. x. 1976 (AEIC); 2 8 , Amazonas, 4°33'S, 71°38'W, l-30.ix.1979 {Alvarenga) (AEIC). Etymology. — This species is named in honour of Ms. A. Guidotti in recognition of her dedicated work as technician of the entomology collection of the Royal Ontar- io Museum. Boethella hubleyi Bennett, n. sp. Diagnosis. — Distinguished from other species of Boethella by the combination of rt//: 1) malar space obliterated (mandibular socket contiguous with ventral edge of eye); 2) auxiliary carina of mesopleuron short, not joining epicnemial carina; 3) hind tibia yellowish orange in basal 0.7, brown apically. Female (holotype). — Fore wing length 4.5 mm; medial part of apical edge of clypeus slightly and broadly emarginate; groove between clypeus and supraclypeal area moderately strong laterally; malar space obliterated (dorsal edge of mandibular socket contiguous with ventral edge of eye); antenna with seventeen flagello- meres; auxiliary carina of mesopleuron short, not joining epicnemial carina; ab- scissa of fore wing vein M between 3rs-m and 2m-cu less than 0.5 length of 2m-cu; spiracle of Tl anterior to 0.7. Yellowish or- ange; apical 0.2 of mandible, antennae ex- cept apical two flagellomeres, anterior 0.5 of medial lobe of mesoscutum, lateral lobe of mesoscutum except anterior, lateral and medial edges, mesopleuron, mesosternum except medial 0.2, apical 0.3 of hind tibia, hind tarsomeres, wing veins, stigma and T3 to T7 except posterior 0.2 and lateral 0.1 to 0.2 brown; occiput medioposterior- ly, pronotum, tegula, posterior 0.5 of me- dial lobe of mesoscutum, anterior, lateral and medial edges of lateral lobes of me- soscutum, scutellum, T2, lateral 0.2 of T3, lateral 0.1 of T4 to T7 orange; palpi, basal 0.8 of mandibles, fore and middle coxa and trochanter and anterior 0.7 of Tl whit- ish yellow; apical two flagellomeres light brown; membrane of wings hyaline with a trace of inhumation apically in fore wing. Male. — Unknown. Material— Holotype 9, BRAZIL, Santa Catarina, 27°11'S, 52°23'W, 300-500m, 25.viii.1962 (F. Plaumann) (CNCI). Etymology. — This species is named in honour of Mr. B. Hubley in recognition of his tireless work as collection manager of entomology at the Royal Ontario Muse- um. Comments. — Known only from the ho- lotype. Colours in fresh material may be more contrasting because of the age and condition of the holotype at time of de- scription. Boethella hubleyi is similar to B. canilae but the former can be distinguished by the bi-coloured hind tibia. Similar col- ouration of the hind tibia is only known in B. guidottiae, however this can be distin- guished from B. hubleyi by the structure of the epicnemial carina. Boethella curriei Bennett, n. sp. Fig. 4 Diagnosis. — Distinguished from other species of Boethella by the combination of: 218 Journal of Hymenoptera Research 1) the auxiliary carina of mesopleuron ange are more yellowish orange and yel- long, joining epicnemial carina, the latter lowish orange structures more yellow; extending dorsal to point of union of aux- hind tibia yellow in basal 0.2, yellowish iliary carina by at least the length of aux- brown medially, brown apically; medial iliary carina (Fig. 4); 2) T4 predominantly longitudinal stripe on T4 to T7 less prom- brown, inent; T4 completely yellowish brown ex- Female. — Unknown. cept for anterolateral corners and posteri- Male. — Fore wing length 3.2 to 3.8 mm; or 0.2. medial part of apical edge of clypeus Material. — Holotype 6, PERU, Cusco, slightly and broadly emarginate; groove near Marcapata, Avispas, 1-30. ix. 1962 (L. between clypeus and supraclypeal area Vena) (AEIC). Paratypes. 1 6, same data moderately strong laterally; malar space as holotype except 1-15. x. 1962; 1 6 , same obliterated (dorsal edge of mandibular as holotype except 20-30. ix.1962. socket contiguous with ventral edge of Etymology. — This species is named in eye); antenna with sixteen flagellomeres; honour of Dr. D. C. Currie, curator and auxiliary carina of mesopleuron long, join- keeper of black flies at the Royal Ontario ing epicnemial carina, the latter extending Museum, in recognition of his unfathom- dorsal to point of union of auxiliary carina able appreciation of fried spam sandwich- by at least the length of auxiliary carina; es. abscissa of fore wing vein M between 3rs- Comments. — The specimen caught 1- m and 2m-cu less than 0.5 length of 2m- 15.X.1962 is not only lighter in colour than cu; spiracle of Tl anterior to 0.7. Yellowish the other two specimens, but also has a orange; apical 0.2 of mandible, antenna ex- much less pronounced sternaulus and a cept for apical one to two flagellomeres, weaker, shorter epicnemial carina. Addi- hind tibia (except in basal 0.2 in some tional material is necessary to determine if specimens), hind tarsus, wing veins and these differences are intra- or interspecific, stigma brown (ventral part of stigma light brown in some specimens); dorsal 0.5 of ACKNOWLEDGMENTS head, dorsal 0.5 of pronotum and meso- Funding for this study was provided by an NSERC SCUtum orange (head and pronotum blend operating grant to Dr. D.C. Darling and by the amaz- uniformly from orange dorsally to yellow- ing and orten under-appreciated efforts of my gain- ish orange ventrally); basal 0.8 of mandi- fully emPloyed wife Dr. C. Canil. Additional funding ii „ ^„i • c a ■jji l was provided bv the University of Toronto, Depart- bles, palpi, fore and middle legs, propo- , v c~ , „ , * „ . ' / , i i n ^ <- rnn t-, & ' r r ment of Zoology as well as by the Board of the Amer- deum, lateral 0.2 of T2 to T4, posterior 0.1 ican Entomological Institute which funded travel to to 0.2 of T3 to T4 and thin, medial longi- the AEIC. The hospitality of Dr. D. Wahl is much tudinal Stripe on T4 to T7 yellow; tegula, appreciated during visits to the AEIC as well as his medial 0.6 of T4, (except for posterior 0 1 permission to include undescribed Townes material to 0.2 and medial stripe), medial 0.8 of T5 !n ,my, studies.In addition, I*. J. Huber was of great , . , ... F , ' „ , rri, , m help during visits to the CNCI and Dr. N. Johnson (except for medial stripe), all of T6 and T7 helped pinpoint Bra2llian locallties Two anonymous (except for medial Stripe) yellowish reviewers made valuable comments to the manu- brown; apical one to two flagellomeres script. Study space and equipment was provided by and gonoforceps whitish brown; sternites the R°yal °ntario Museum. of metasoma yellowish white; glossa white; wings strongly infumate basally and dorsally, hyaline subapically and pos- Bennett, A. M. R. 2002. Cladistics of the Tryphoninae teriorly and weakly infumate apically. (Hymenoptera: Ichneumonidae) with a discus- Colour variation: specimen caught 1- sion of host use and the evolution of parasitism iCwir>£'-> ■ i- Ui. . in the Ichneumonidae, Ph. D thesis. University of lb.x.1962 is lighter than other two speci- Toronto/ 366 pp mens— structures described above as or- Bennett, A. M. R. in prep. Cladistics of the Tryphon- LITERATURE CITED Volume 12, Number 2, 2003 219 inae (Hymenoptera: Ichneumonidae). Memoirs of the American Entomological Institute. Gauld, I. D. 1997. The Ichneumonidae of Costa Rica, 2. Memoirs of the American Entomological Institute, 57: 1-485. Gauld, 1. D. 2000. The Ichneumonidae of Costa Rica, 3. Memoirs of the American Entomological Institute, 63: 1-453. Gauld, I. D. and D. B. Wahl. 2000. The Townesioni- nae: a distinct subfamily of Ichneumonidae (Hy- menoptera) or a clade of the Banchinae? Trans- actions of the American Entomological Society, 126: 279-292^ Goulet, H and J. T. Huber. 1993. Hymenoptera of the World: An Identification Guide to Families. Research Branch of Agriculture Canada Publication 1894/ E, 668 pp. Kasparyan, D. R. 1973. Fauna of the USSR Hymenop- tera Vol. Ill Number 1. Ichneumonidae (Subfam- ily Tryphoninae) Tribe Tryphonini. Nauka Pub- lishers, Leningrad (In Russian). Translated into English (1981) by Amerind Publishing Co, Ltd. New Delhi, 414 pp. Quicke, D. L. J., M. G. Fitton, D. G. Notton, G. R. Broad, and K. Dolphin. 2000. Phylogeny of the subfamilies of Ichneumonidae (Hymenoptera): a simultaneous molecular and morphological anal- ysis. In: Austin, A. D. & Dowton, M. (eds) Hy- menoptera: Evolution, Biodiversity and Biological Control. CSIRO, Collingwood, Victoria, pp. 74- 83. Ross, H. H. 1936. The ancestry and wing venation of the Hymenoptera. Annals of the Entomological So- ciety of America, 29: 99-111. Scaramozzino, P. L. 1991. Two new species of the ge- nus Boethus Foerster, 1869 from Africa (Hyme- noptera, Ichneumonidae, Tryphoninae). Bollettino delta Soeieta Entomologica Italiana. 123: 55-61. Townes, H. K. 1969. Genera of Ichneumonidae Part I. Memoirs of the American Entomological Institute, 11: 1-300. Townes, H. K., V. K. Gupta, and M. J. Townes. 1992. Nearctic Tryphoninae. Memoirs of the American Entomological Institute, 50: 1-296. Wahl, D. B. 1990. A review of the mature larvae of Diplazontinae, with notes on larvae of Acaeniti- nae and Orthocentrinae and proposal of two new subfamilies (Insecta: Hymenoptera, Ichneumon- idae). Journal of Natural History, 24: 27-52. Wahl, D. B. 1991. The status of Rhimphoctona with special reference to the higher categories within Campopleginae and the relationships of the sub- family (Hymenoptera: Ichneumondae). Transac- tions of the American Entomological Society, 117: 193-213. Wahl, D. B. and I. D. Gauld. 1998. The cladistics and higher classification of the Pimpliformes (Hy- menoptera: Ichneumonidae). Systematic Entomol- ogy, 23: 265-298. Yu, D. and K. Horstmann. 1997. Catalogue of world Ichneumonidae (Hymenoptera). Memoirs of the American Entomological Institute, 58: 1-1558. J. HYM. RES. Vol. 12(2), 2003, pp. 220-237 Foraging Strategy and Pollen Preferences of Andrena vaga (Panzer) and Colletes ciinicularius (L.) (Hymenoptera: Apidae) Inge Bischoff, Kerstin Feltgen, and Doris Breckner (IB) Zoologisches Forschungsinstitut und Museum Alexander Koenig, Adenauerallee 160, D-53113 Bonn, Germany, email: i.bischoff.zfmk@uni-bomi.de; (KF) Dottendorferstr. 29, D-53123 Bonn, Germany, email: kerstin.feltgen@web.de; (DB) Johann-Sebastian-Bach-Str. 4, D-77654 Offenbach, Germany Abstract.— Andrena vaga (Panzer) and Colletes ciinicularius (L.), both vernal ground nesting bees, were studied in the years of 1996-1999 in a lowbush heath near Cologne, western Germany. Both species are solitarily but nest gregariously and sometimes form large aggregations with thousands of nests. They are reported to feed strictly oligolectic on the genus Salix. We observed the daily foraging rhythms of both species and compared their foraging strategies. Colletes ciinicularius start- ed provisioning trips earlier in the morning, made more trips per day than A. vaga, and finished nest provisioning later in the evening. Colletes ciinicularius burrowed even in the dark after 08.00 p.m. Andrena vaga collected pollen and nectar on different days each. One pollen day included 1 to 5 pollen trips. There was no clear correlation between the number of pollen trips and the occurrence of a subsequent nectar day. We found also no correlation between the occurrence of nectar-provisioning trips and weather conditions. Pollen loads of both species were analyzed qualitatively and quantitatively with a cell counter and two different hand-counting systems. Andrena vaga collected nearly twice as much pollen as C. cuuiciilarius during one foraging trip. Cells and pollen loads of C. ciinicularius contained large portions of other pollen types, mostly Rosaceae such as Prunus, Sorbus and Pyrus or Acer, Quercus and Ilex. Thus, C. ciinicularius is not oligolectic as described in the literature. The percentage of pollen types other than Salix increased at the end of the flowering period of Salix, which indicates a resource restriction at the end of the season. The reproductive success of C. ciinicularius measured by nest provisionment exceeded that of A. vaga, because of longer activity per day and digging activity in the evening. Andrena vaga (Panzer 1799) and Colletes ciinicularius (L.) are univoltine, vernal sol- itary bees which are distributed through- out the entire Paleartic Region. They pre- fer to nest in sandy soils and often form large aggregations (Friese 1923, Moeschler 1938, Vleugel 1947, Westrich 1990). Both species are reported to be specialized on Salix as a pollen resource for their larvae (Westrich and Schmidt 1987). Being spe- cialized on the same host-plant, their sea- sonal activity strongly overlaps (Westrich 1990). Therefore, we investigated the di- urnal activity pattern and foraging strate- gy of both species. Several studies have dealt with diurnal activity patterns and the impact of weather conditions on a number of mostly Nearctic bee species (cf. Batra 1999, Lind 1968, Michener and Ret- tenmeyer 1956, Schonitzer and Klinksik 1990, Stephen 1966), but precise data on the life cycle of European species of An- drena and Colletes are very scarce (Geb- hardt and Rohr 1987, Malyshev 1927, Witt 1992). By analyzing thoroughly the daily activ- ity patterns, niche differences between two species may be shown. Levermann et al. (2000) investigated the diurnal cycle and niche differentiation of Dasypoda hirtipes (Fabricius) and Panurgus calcaratus (Scopo- li). They demonstrated that apart from weather conditions, body mass and pollen- collecting apparatus are important factors METHODS Volume 12, Number 2, 2003 221 determining the diurnal activity cycle. Re- and overlaying quicksand, soils are most- garding solitary Hymenoptera, female size ly sandy, permeable, and poor in nutri- appears to have a great influence on pro- ents. The climate is humid-oceanic with an visioning- and reproductive success (Al- annual mean temperature of 9.5 °C and an cock 1979). Colletes cunicularius is larger annual mean precipitation of 804 mm. than A. vaga and has more hairs, especially Due to drainage and loss of traditional ag- on the thorax. Therefore, we hypothesized ricultural use, grass and bushland domi- that better thermoregulatory abilities allow nate great areas of the heathland (Inter- C. cunicularius an earlier start to provision- kommunaler Arbeitskreis Wahner Heide ing activity in the morning. 1989). The investigated aggregations of C. Apart from the different sizes of the two cunicularius and A. vaga lie on sandy in- species, the specific pollen-collecting ap- land dunes, paratus suggests differences in their pol- len collecting efficiency. Although both species have trochanter-femur-baskets, Field work. — To study their diurnal ac- floccus and thoracic pollen basket of A. tivity, individual female bees were vaga is more strongly developed than in marked with opalith-plates. A total of 238 C. ciniicularius. In a number of bee genera, females of C. cunicularius in 1996 and 174 the size and the type of pollen-collecting females of A. vaga in 1997 were marked, apparatus results in different amounts of Corresponding nests were identified by a pollen transported (Braue 1916, Friese colored nail of the same number as on the 1923, Michener et al. 1978, Pasteels and opalith-plate of the female. To record the Pasteels 1979, Westerkamp 1987, 1996, exact departure and return times of the fe- Westrich 1990). Thus, we examined the males, several nests were covered with number of collected pollen grains in both preserving jars (cf. Witt 1992). species. Climatic parameters were measured Although both species are regarded as with data loggers (Orion Tiny Logger strictly oligolectic on Salix (Vleugel 1947, Manager OTLM Tiny talk©). Soil temper- Westrich and Schmidt 1987, Westrich atures of all years were recorded 20 cm 1990), anecdotal observations reviewed by below the surface, and air temperature Mader (1999) suggest that C. cunicularius and atmospheric humidity in a portable visits also flowers other than Salix. A clear weather station were measured at a height evaluation about oligolecty in a bee spe- of 2 m in 1997-1999. Data from the Co- des can only be made by a quantitative logne/Bonn airport weather station were analysis of pollen loads or cell provisions, also used. The data on the activity and the The aims of this study were to analyze nectar trip frequency of the bees were test- niche differentiation of two synchronous ed for possible correlation against various and syntopic bee species on the basis of climatic parameters (daily hours of sun- activity patterns and the use of host shine, mean daily temperature, mean dai- plants, to asses the degree of oligolecty, ly atmospheric humidity, humidity of soil and to compare their pollen collecting ef- and minimal daily soil temperature), ficiency. Since the activity of most of the bees is The study area is part of the so-called extremely influenced by weather condi- "Wahner Heide", a large 5000 ha heath- tions (Larsson 1991, Lind 1968) we created land, east of the river Rhine near Cologne, a measure for the bees activity indepen- Since 1961 the heathland has been a mili- dent from weather conditions (cf. Midler tary training area for NATO and was des- 1994), which we referred to a so-called ignated a Nature Reserve in 1968. Due to "bee day" (= BD). Such a measure makes the presence of Quaternary gravel, sand it also easier to compare different years. 222 = bee in the nest C. cunicularius n10 n423 n465 first departure 8:35 (n35) A.vaga ("nectar-day") A.vaga ("pollen-day") = provisioning trip Journal of Hymenoptera Research waiting at the entrance yyfa = digging l l l I I I last return 17:31 (n.20) n19 n39 tirst departure 11:13 (n25) last return 14:29 (n39) n19 n90 n82 H 1 1 h first departure 10:18 (n20) H 1 1 1 1 h H 1 1 1 1 1 1 1 h last return 16:40 (n29) H 1 1 1 h H 1 1 1 1 Time (MET) Fig. 1. "Ideal ethogram". Time schedule of provisioning behavior of A. vaga and C. cunicularius (dotted line = end of burrowing unknown). Further information's see text. One bee day corresponds to a day with optimal weather conditions, which fe- males could use completely for provision- ing activities. A bee day for C. cunicularius had between 8-9 hours. The measure for these 8-9 hours was the observed flying activity at such an optimal day. Other days with less optimal flying conditions were defined as follows: 1.0 BD = flying activity of 8-9 hours 0.75 BD = flying activity more than Vi of one BD 0.25 BD = flying activity less than Vi of one BD 0 BD = no flying conditions the whole day The classification of the bee day of A. vaga was more difficult because of the greater variance in activity time, in spite of good weather conditions: 1.0 BD = flying activity of 8-11 hours 0.75 BD = flying activity more than Vi of one BD 0.25 BD = flying activity less than Vi of one BD 0 BD = no flying conditions the whole day Nectar- and pollen days of A. vaga. — To assess a pattern between A. vaga's nectar- and pollen-provisioning trips, we ana- lyzed the number of pollen days as well as the number of pollen trips between nec- tar trips. One nectar trip corresponds to one nectar day and the patterns of pollen trips between nectar trips occurs over a period days since pollen and nectar are collected on different days. We obtained complete observations of pollen trips be- tween nectar trips of 7 different females. Pollen analysis. — For pollen analysis we excavated 10 cells of C. cunicularius in 1996 and 6 cells in 1998. The cells could not be attributed to a specific nest or time of com- Volume 12, Number 2, 2003 223 pletion. Additionally, females were cap- tured to analyze pollen loads. This was necessary for A. vaga because cells could not be excavated without destruction and partial loss of the pollen mass. A total of 38 pollen loads of A. vaga and 28 pollen loads of C. cunicularius were analyzed (see Table 1). To remove the complete pollen load from the bees, all body parts (legs and sometimes thorax without wings) were sonicated in vials filled with a liquid medium (cf. Buchmann and Shipman 1990). For a comparison of grain numbers, only pollen loads with nearly 100% Salix were used because the number of pollen grains varies with their size (see also Sil- veira 1991, Tasei 1973). For 10 pollen sam- ples of C. cunicularius also volumetric per- centages of the different pollen types were considered (cf. Buchmann and O'Rourke 1991). Pollen grain dimensions of Salix, Quercus robur and Prunus padus were mea- sured for 10 grains of each species under a scanning electron microscope (SEM). The average size of the grains of Acer pseudo-platanus were taken from Cromp- ton and Wojtas (1993). The number of pollen grains in cells and pollen loads was counted by different methods: a) Ratio-counting with Lycopodium spores: For this quantification, pollen was acetolyzed (Erdtmann 1960, Moore et al. 1991). During acetolysis a tablet with a known number of Ly- copodium spores was added (Stock- marr 1971). In a subsample all pollen grains and spores were counted on a slide under a microscope. The total number of pollen grains in the sample was estimated from the equation: total number of grains added Lycopodium spores counted Lycopodium spores X counted grains b) Cell counter: Most of the samples were additionally analyzed using a cell counter (Casy® 1 Cell counter and An- alyser) and then checked with the SEM. Eight electronically counted samples for the years 1996 and 1998 were checked with the Lycopodium spores method de- scribed above. Between 500 and 1000 Salix grains were counted, respectively. c) Hand-counting with a counting chamber: Samples from 1999 were counted with a Buerker counting chamber, a special slide with a cavity of a defined volume. To achieve an ac- curacy of 10 or 20 grains per |jl1, we de- termined the number of subsamples needed using the following formula (/? = random sample, s = standard error, d = accuracy): // 1.96 X s d In this study, 6 to 14 subsamples had to be counted. To calculate the number of provisioning trips for one cell, the average weight of food stored in the cell has to be divided by the average weight of the pollen load carried by the female (cf. Mohamed 1973). Therefore we determined the dry weight of the pollen load samples of the year 1996. The females were dried in a drying chamber, head and wings removed, and the rest weighted. Then pollen was re- moved from the body hairs and scopae with a sonicator. After removing the pol- len, the thorax and abdomen were dried and weighed again. The difference corre- sponds to the weight of the pollen load. Statistics. — Mean values of all departures and returns or other activities were used to construct an "ideal ethogram". Except in cases when data were not normally distrib- uted, the median was used. The compari- 224 Journal of Hymenoptera Research 26- 24- 22- 20- 18- 16" 14- 12- 10- 8- u 6- ? . 4- 0) 2- 3 0 Andrena vaga First departure Last return -i r i i 07:00 08:00 09:00 10:00 11:00 12:00 13!00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 0) a E 26 n H 24- 22- 20- 18" 16- 14- 12- 10- 8- 6- 4- 2- Colletes cunicularius First departure ♦ • 2m O O -20 cm 0 07:00 08:00 09:00 10:00 1:00 12:00 13:00 — i 1 1 1 i 14:00 15:00 16:00 17:00 18:00 — i 1 19:00 20:00 Time [MET] Fig. 2. Correlation of time of first departure and last return of A. vaga and C. cunicularius with temperature at 2 m height (BC = black circle; BS = black square) and 20 cm below soil surface (WC = white circle; WS = white square). (First departure: BS y = 26.49x + 0.21, r = 0.48, n = 36; WS y = 30.92x-6.38, r = 0.82, n = 38; BC y = - 21. 57x+ 21 .45, r = 0.36, n = 78; WC y = -24.56x+22.37, r = 0.71, n = 78; Last return: BS y = 29.94x-4.31, r = 0.68, n = 42; WS y = 29.35x-6.16, r = 0.81, n = 42; BC y = 0.34x + 19.25, r = 0.01, n = 20; WC y = 30.12x-3.88, r = 0.89, n = 20). son of mean values of activity data was presented in boxplots. These independent samples were analyzed with the t-test. Data of pollen samples were treated like- wise. Mean values of non normally distrib- uted data (number of pollen grains per pol- len load and soil temperatures) were com- pared using the Man-Whitney U-test. The relationship between activity data and cli- mate and mixed pollen cells and the blooming time of Salix were tested by Pear- sons's correlation analysis. RESULTS Foraging strategy — diurnal cycle. — A comparison of the "ideal ethogram" (Fig. Volume 12, Number 2, 2003 225 18.00 17:00 16.00 15:00 14:00 lu 13.00 — 12:00 CD .1 11:00 *" 10:00 09:00 08:00 First departure Last return C. cunicularius A. vaga (pollen) A. vaga (nectar) Provisioning flights Daily flight activity Fig. 3. Variance and difference in time of first departure, last return, time of provisioning trips and hours of daily activity of A. vaga and C. cunicularius. 1) shows a different diurnal cycle of A. vaga and C. cunicularius. For A. vaga two separate cycles are shown since it collects pollen and nectar on different days and this is represented by different flight pat- terns. Colletes cunicularius started its first trip at 8:37 in the morning, after waiting about 20 minutes at the entrance. It made seven provisioning trips and returned from its last trip at 17:34. It did not interrupt its foraging cycle by digging or other activi- ties in the nest, as indicated by short du- rations of presence in the nest (six minutes on average). In the evening, after the last return to the nest, many females began burrowing and continued even in the dark (Fig. 1). Colletes cunicularius showed lower variation in the number of trips per day than A. vaga. The first departure and the last return of C. cunicularius correlated with the soil temperature (Fig. 2). The fe- males started earlier when soil tempera- tures were higher and returned later from their last trip when temperature was still high. The correlation of the last return with the air temperature was not signifi- cant. The soil temperature at the aggre- Cells C.c. Pollen loads Year C.c. A.v. 1996 10 9 7 1998 6 1 14 1999 18 17 22(_ Journal of Hymenoptera Research Table 1. Number of excavated cells and collected for the first trip of A. VClga was 8 °C, the pollen loads (C. c. = C. cunkularius, A. v. = A. vaga). mean temperature of first departure was 12 °C. To summarize, the females of C. cuni- cularius started their provisioning behav- ior earlier in the morning, made more but shorter trips a day, remained for shorter periods in the nest, returned later in the evening (Fig. 3) and burrowed after dark, gation of C. cunkularius was significantly Andrena vaga started its first trip later in higher (U = 88698, p < 0.001) than the soil the morning but the temperature thresh- temperature of the location of A. vaga's ag- old for the first trip was lower than that gregation, though the year 1996 was much of C. cunkularius. colder than 1997 (Bischoff 2000). Thus, Nectar and pollen days of A. vaga.— A. lowest soil temperature during first de- vaga made 1 to 9 pollen trips between two parture of C. cunkularius (11 °C) differed nectar trips (mean: 4 trips, Table 2). Two highly from soil temperature in the aggre- groups of flight patterns seem to exist: the gation of A. vaga (5 °C). The temperature nrst group of bees makes 1 to 4 pollen threshold for the first trip of C. cunkularius trjpS between two nectar trips, and the was 9.5 °C; the mean temperature of first second group makes 7 to 9 pollen trips be- departure was 12 °C. tween each nectar trip. On pollen days A. vaga made three trips The activities 0f A mga females at dif- on average. After remaining a while at the ferent bee days are shown in Rg 4 There entrance, it started its first trip at 10:32. wag fl Wgh percentage of mactive females The females stayed on average half an qr Q25 bd% [e dayg ^ legs ^ 5 5 our ^ ' , ,. . hours of good flying conditions. On days two provisioning trips and no digging ac- .Ll_ f a ,.t. ,., n n^, . . r , , ■ i ■ , with good flying conditions (1.0 BD), more tivitv was observed during these periods. , , , ,, , . ,i .,/,.„., , , , ., , , , . j females made pollen trips than nectar At 16:17 it came back from its last trip and . . , , . , . , c , , j . ^ j trips. Yet, this relation was also found on closed its nest entrance. On nectar days, , . , females emerged not before 11:30 and re- °'25 BD s" Furthermore we compared the turned at 14:29. In the evening no intense Percentage of nectar trips with the bee day digging activity as observed for C. cum- status of the preceding day. After a 0 BD cularius, occurred. The last return of the 90% Salix) and 9 cells were mixtures of Salix and other grain types. The remaining pollen types in the cells of 1996 were mainly composed of various Rosaceae (Table 4). Apart from Rosaceae, only Quercus and Sambucus occurred in higher percentages. Ilex pollen dominated one pollen load sample of C. cunicularius of 1996. Percentages of other pollen types were also found in pollen loads of C. cunicular- ius of the years 1998 and 1999. The loads of two females (captured on 29.4. and 7.5.1999) contained only grains of Acer sp., one load (also from 7.5.1999) contained Acer and Ilex grains in a ratio of 1:1, and two pollen loads contained exclusively Quercus pollen. Regarding also the vol- ume of the different pollen types (10 sam- ples of 1999), Salix grains represent even a smaller proportion of the diet of C. cuni- cularius, since Acer pseudo-platanus, Primus padus and Quercus robur grains are much larger than Salix (Fig. 5). Percentages of other pollen types in pollen loads of C. cunicularius (in the year 1998) increased significantly with time, i.e. with the end of the blooming of Salix (r = 0.731, p = 0.01, n = 14). In the graphs of the cell counter (Fig. 6), mixed loads of mainly two pollen types {Salix and various Rosaceae) could be rec- ognized as two separated peaks. These graphs are counts from C. cunicularius cells of 1999. They show three clearly sep- arated peaks. The first peak represents particles smaller than 10 (xm and can be interpreted as pollution. The second peak (15-22 fxm) represents the Salix grains. The third peak (25-35 |x,m) shows bigger grains, e.g. Rosaceae. The broad distribu- tion of grain sizes as displayed in the counter graph corresponded to different pollen types detected under the light mi- croscope. However, results of the hand countings differed often from percentages given by the counter. All results of the counter had to be checked at least quali- tatively by microscope, because one peak could represent pollen types other than Salix. The cell counter calculated a mean num- ber of 1512901 (± 720715 SD) pollen grains per pollen load for C. cunicularius. Only 7 out of 18 pollen loads counted by the cell counter were pure Salix loads. In A. vaga, electronic counting determined 2058692 (± 737197 SD) grains per pollen load. An- il rena vaga collected on average one and a half more pollen on one provisioning trip than C. cunicularius. This difference is highly significant (t-test, p < 0.001). A comparison of the results using dif- ferent counting methods is displayed in 228 Journal of Hymenoptera Research 0,25 BD □ 0,75 BD 1.00BD 77 Nectar trip Pollen trip Resident (in nest) Fig. 4. Activities of A. vaga females at a bee day of the category 0,25 (< 5,5 hours good flying-conditions), a bee day of the category 0,75 (> 5,5 hours good flying conditions) and the category 1 (8-11 hours good flying-conditions) (number of total observations = 360). Table 5. Ratio counting with Lycopodium spores showed greater divergence from cell counter results than counting with the counting chamber. Counting chamber re- sults revealed a mean difference of 255728 (± 192194) grains in comparison to the cell counter. Regarding Lycopodiiun ratio counting, one half of the results exceeded the cell counter calculations and the other half was below the cell counter calcula- tions. The mean difference between Lyco- podium spores ratio counting and cell counter was 498713 (± 675670) grains. The pollen loads of A. vaga contained fewer foreign grains than those of C. cun- icularius. In five (13%) of 38 loads, we found percentages of other pollen types ranging from 1 to 7% and consisting most- ly of Rosaceae, Quercus and Betula grains. DISCUSSION Diurnal cycle — foraging strategies. — In this study, C. cunicitlarius started its pro- visioning cycle much earlier than A. vaga. The specified activity times are probably dependent on weather conditions. This earlier departure may have been caused by higher soil temperatures. However, the departure time is not known from the oth- er study site at the Fliegenberg. The in- vestigated aggregation of C. cunicularius is exposed southward, has a strong slope and the soil is only sparsely covered with vegetation. The exposition of A. vaga's ag- Table 4. Classification of pollen types other than Salix in the cells of C. cunicularius in 1996. Family Genus or species Percentage [%] Rosaceae Fagaceae Caprifoliceae Ranunculaceae Celastraceae Aceraceae Sorbus aucuparia, Prunus padus, Primus lau- 5-92 rocerasus, Primus sp., Pyrus sp., Mains sp. Pilipendula sp. <1 Quercus sp. 2-10 Sambucus niger 7 Ranunculus sp. 0.2-3.3 Euonymus europaeus 2 Acer sp. <1 Volume 12, Number 2, 2003 229 Salix □ Acer 0 Prunus □ Quercus 100 -90 n | 80 S 70 o o- 60H § 50 i 40 a 30 D) 1 20 | 10 I a 5 Pollen load Fig. 5. Numerical (a) and volumetric (b) percentage of Salix pollen and other pollen types in excavated cells of C. cwiicularius in the year 1996 and 1998. gregation is south-eastward, slope is low- er than in the C. cuniciilarius aggregation and the sand path is covered with grass. The surrounding site of A. vaga's aggre- gation has also more and higher vegeta- tion cover. Thus, higher soil temperature at the C. cuniciilarius aggregation may be caused by these local differences of expo- sure, slope and vegetation cover (cf. Bis- choff 2000). Ideally, the diurnal cycle of C. cuniciilarius should be investigated at ag- gregations close to A. vaga's aggregation, i.e. with similar conditions of soil, expo- sure, slope and vegetation cover. Several authors have found correlations between the behavior of bees and weather conditions (e.g. Linsley 1958, Michener and Rettenmeyer 1956, Willmer 1983). Temperature thresholds for the bee's ac- tivities depend on weather conditions and the season in which the species occurs. Many vernal bees begin flight activity at 10 °C and they are less influenced by cloud cover or wind. Flight activity tem- peratures reported for other European ear- ly spring species of Andrena, for instance A. barbilabris (Kirby), A. cineraria (Linnae- us), A. clarkclla (Kirby), and the Nearctic species A. erythronii Robertson, A. vibur- nella Graenicher, and A. vicina Smith range from 10 to 16 °C (Gebhardt and Rohr 1987, Johnson 1981, Michener and Rettenmeyer 1956, Miliczky and Osgood 1995, Stephen 1966, Witt 1992). For vernal species of Col- letes like C. inaequalis Say and C. validus Cresson, similar temperature thresholds are known (Batra 1980). In our study, the two species began flight activity at 8-9.5 °C air temperature. Similarly, Schonitzer and Klinksik (1990) recorded flight activi- ty at a temperature of 8 °C for A. nyc- theincra Imhoff. Due to unstable weather conditions in spring, vernal species have to use days with good weather conditions very efficiently. To illustrate, on days with optimal weather conditions provisioning activity of A. clarkclla is completed in 4 or 5 days (Friese 1923). Late summer species such as Panurgus banksianus (Kirby) and Dasypoda hirtipes (Fabricius) often need temperatures > 20 °C to start their first trip (Lind 1968, Munster-Swendsen 1968). Though the flight of A. vaga may already start at 8 °C, time of its first departure is later than that of many of the other species of Andrena mentioned above. This may be caused by local and seasonal differences in temperature compared to the other stud- ies mentioned above. In the present study the required temperature threshold of 8 °C 230 Journal of Hymenoptera Research at the investigated aggregation of A. vaga Miliczky and Osgood 1995, Schonitzer was not reached before 09.00 a.m. This fact and Klinksik 1990). may explain the strange correlation of the A second reason for the marked differ- first departure of A. vaga with the temper- ence in daily activity of the two investi- ature, which represents in fact no corre- gated species could result from A. vaga's' lation with the temperature. The bees can prolonged stay in the nest. It could not be start their first trip at a temperature clarified whether A. vaga uses these peri- threshold of 8 °C and the regression rep- ods in the nest for digging, since no new resents only the increasing number of sand was pushed to the surface. The oc- starting bees with time. currence of sand output depends of the se- Apart from differences in soil tempera- quence of nest construction. When the bee ture, the beginning of flight activity of the first digs the main burrow and constructs two species may be influenced by their re- the cells regressively (i.e., the lowest one spective thermoregulatory abilities. Larger is built first and each subsequent one is at bees are more likely to achieve flight tern- a higher level), it can fill the inferior main peratures at low ambient temperatures burrow with the material of the side bur- (Michener and Rettenmeyer 1956, Stone et rows. Thus no new sand needs to be al. 1988, Stone and Willmer 1989, Stone pushed out. This has been described by 1993a, b, Stone 1994, Stone et al. 1995, Malyshev (1927) for C. cunicularius and Wolda and Roubik 1986). Colletes cunicu- also by Rajotte (1979) for C. validus. Yet, A. larius is one of the largest bees in Germa- vaga constructs its nest conversely, subse- ny, having a mean heating rate of 7.35 °C quent cells lie deeper and the oldest cell is per minute (Stone and Willmer 1989). nearest to the surface. Side burrows of A. These authors investigated the heating- vaga are also filled with sand and in order rate among A. clarkella and A. fulva (Mull- to fill the first side burrow of a completed er). These two species are comparable to cell, the bee may use the material of a sec- A. vaga in body size, hairiness and flight ond side burrow. This was also assumed season, and differ only in color from A. by Michener and Rettenmeyer (1956) for vaga. Mean heating rate of these two spe- A. erythronii. Our (Bischoff 2001) and Ma- cies of Andrena is about 4 to 6.2 °C per lyshev's (1926) description of A. vaga's minute, respectively. Although C. cunicu- nest architecture are contradictory to the larius is larger and more hairy than A. descriptions and figures of Friese (1882, vaga, the abdomen of the latter is deep 1923), in which the last cell is located at black and passive heat absorbency may be the lowest level. Michener and Rettenmey- increased. Nevertheless, body size is prob- er (1956) suggested that Friese's nest fig- ably more important for warming up in ure with cells close along the main burrow the morning and may enable C. cunicular- like a cluster of grapes resulted from a ins to begin earlier with daily activity. Af- mixture of different nests lying very close ter sunset, this species may also benefit together. from its larger size. In many species, digging of side bur- Most of the flights of A. vaga took be- rows and new cells has been described to tween 1 and 2.5 hours on pollen days, and take place in the afternoon (Lind 1968, between 2 and 3.5 hours on nectar days. Munster-Swendsen 1968, Gebhardt and The known duration of provisioning trips Rohr 1987, Michener and Rettenmeyer of other species of Andrena ranged from 20 1956, Witt 1992). However, our results do minutes to 4 hours, and the number of not confirm these findings for A. vaga, provisioning trips per day showed a tran- where even after the last provisioning trip, sition from 1 to 5 (Gebhardt and Rohr females did no intensive digging as ob- 1987, Michener and Rettenmeyer 1956, served in C. cunicularius. Yet, at the begin- Volume 12, Number 2, 2003 231 Table 5. Results of pollen counts with the electronic counter, the counting chamber and the Lycopodium- ratio-method (means of all counts of both species). Grain number/pollen load Sta ndard deviation Mean difference to electronic counter Electronic 1053796 722183 counter 1777806 626778 Counting 798069 711542 255728 chamber Lycopodium- 1850200 515579 498713 ratio-counting ning of the season, when the aggregation of A. vaga develops, new tumuli could be found early in the morning. Thus, we as- sume that the construction of these tumuli took place at night or early in the morn- ing, because sand often was still moisted and no sand output was observed in the late evening of the previous day. Andrena erythronii digs in the late afternoon and even in the dark (Michener and Retten- meyer 1956). Nocturnal digging activity has been reported from other Nearctic species of Colletes (Batra 1980, Rajotte 1979). Since C. cunicularius constructs its cells regressively as described above, no large sand output should occur after the construction of the main burrow. Digging activities in the evening can be interpreted as constructions of new nests, since C. cun- icularius makes 2 or 3 nests in its life. Col- letes cunicularius defers the construction of new nests to the evening time, thus it can use the whole day for provisioning trips and can increase the number of construct- ed nests. The third explanation for the difference in daily activity may be the different pol- len carrying capacities of both species. Braue (1916) and Friese (1923) described the different pollen collecting apparatus of bees and inferred from these differences the systematic order of bee genera. Since these early studies, many authors worked on different pollen collecting apparatus (Grinfeld 1962, Michener et al. 1978, Pas- teels and Pasteels 1979, Proctor et al. 1996, Thorp 1969, Westerkamp 1987). According to Braue (1916), Andrena is the genus that can carry home the largest amount of pol- len with its hind leg brushes and parts of the thorax. Although both species have trochanter-femur baskets, the floccus and thoracic pollen baskets of A. vaga seem to be more strongly developed than in C. cunicularius. In the present investigation, A. vaga collected nearly twice as much pollen per load on average than C. cuni- cularius. Exact quantitative data on num- ber of pollen grains per pollen load are scarce in the literature. In most cases, only percentages of different pollen types are presented, e.g. pollen loads of Andrena (Chambers 1946). Parker (1981) analyzed pollen loads of polylectic and oligolectic bee species quantitatively in order to de- termine the effectiveness of these species in pollinating sunflowers. He demonstrat- ed that female oligoleges carried more pollen than did any other group of bees studied. In our study, species collected much more grains per load than in Park- er's example; however, the number of car- ried pollen grains depends highly on the mean size of the grain type. Accordingly, our results can only be compared quanti- tatively to data of the same bee species, collecting the same pollen species. One possible factor as to why C. cunicularius pollen loads were smaller than A. vaga's is that C. cunicularius collects nectar along with pollen in each trip. From honey bees and bumble bees it is known that they can carry an amount of nectar from 50 to 90% of their body weight (Heinrich 1979). We 232 Journal of Hymenoptera Research 500- 250- (0 o o t: (0 o 500 x Cap 150 Ljm Range: 13.6 to 80.0 pm Sample 3 x 400 lI Counts /ml 5136 25 50 75 pm 250 Cap: 150 urn Range 13,6 to 80,0 urn Sample 3x400 pi Counts / ml 6252 '■ «*-■■*-■- 50 Particle size 75 pm Fig. 6. Quantitative analysis of the pollen loads of C. cunicularius with the cell counter. Upper graph: 1. female, 23.04.99, below: 2. female, 27.94.99 (Cap = capillary, particles beneath dotted line = smaller than 10 (xm, no pollen). were unable to find precise data in the lit- though contains a lot of nectar (Malyshev erature on the amount of nectar collected 1936). Other bees like Osmia ritfa (Linnae- by solitary bees on one single trip, but for us) make more dry provisions (Westrich C. cunicularius it is reported that the pro- 1990). vision in the cell is extremely liquid and Dividing the grain number of the pure Volume 12, Number 2, 2003 233 Salix cells for the year 1996 by the grain number of pure Salix pollen loads (same year), C. cunicularius had to make approx- imately nine provisioning trips per cell. In 1998, we had only one cell of C. cunicular- ius, containing only Salix grains. In this case, C. cunicularius had to collect seven pollen loads for completing one cell. If A. vaga had to gather approximately the same amount of pollen per cell, it would have to make only four trips per cells, be- cause of its bigger carrying capacity per load. These assumptions agree well with the observed trips of both species during one day. Vleugel (1947) observed for A. vaga only 1 or 2 trips on days with good weather conditions. The foraging statistics for Andrena complexa visiting Ranunculus for pollen show a time of Wi hours to complete a load, and a pollen foraging rate of three loads per day (Linsley and MacSwain 1959). Futhermore the amount of pollen per cell may depend on the sex of the offspring. Gerber and Klostermeyer (1970) provided evidence that females de- termine the sex of their offspring by fer- tilizing the egg or not. Males are often smaller than females and therefore the stored pollen mass for males is smaller (Helms 1994, Strickler 1982, Maddocks and Paulus 1987, Johnson 1988). Regard- less of the sex of the offspring, it seems likely that C. cunicularius has to make more provisioning trips per cell, because of its smaller pollen carrying capacity. In conclusion, A. vaga can carry more pollen per collecting trip; however, due to its body size C. cunicularius is more indepen- dent of weather conditions and can be ac- tive for longer periods per day. Indeed, the last point is of considerable impor- tance for vernal bee species, because weather conditions are often quite unfa- vorable during spring. Additionally, C. cunicularius uses the evening and perhaps the night for digging activity. Our inves- tigations of daily collecting capacity as well as those of nest excavations (cf. Bis- choff 2001) indicate a higher reproduction rate for C. cunicularius in comparison to A. vaga. Nectar- and pollen provisioning trips. — We observed a rhythm of nectar and pollen provisioning trips and assumed that A. vaga collected first all the pollen for one cell, then added the nectar. Friese (1923) described exactly this type of behavior for A. vaga. Dasypoda hirtipes, Andrena ery- thronii and various species of Anthophora are also known to add nectar only after several pollen loads have been carried into the cell (Lind 1968, Michener and Retten- meyer 1956, Muller 1884, Westrich 1990). Malyshev (1936) states that pollen pre- dominates in the first load or even makes up the whole load and that the last load deposited in the cell usually consists mainly of honey. Other species clearly al- ternate nectar- and pollen provisioning trips, e.g. Osinia adunca (Panzer), Osmia fulviventris Panzer or Chelostoma florisomne (Linnaeus) (Brechtel 1986, Kapyla 1978, Westerkamp 1978, Westrich 1990). It is possible that digger bees and carpenter or mason bees differ with respect to this be- havior. Miliczky and Osgood (1995) de- scribed four trips for A. vicina during which no pollen was collected (in com- parison with 64 pollen-collecting trips) and interpreted them as adult feeding trips. Since we did not analyze quantity of nectar in A. vaga's cells, it cannot be defi- nitely clarified whether nectar trips are adult feeding trips or nectar provisioning trips for the offspring. Assuming that A. vaga cells contain nectar, then there must be a rhythm between nectar and pollen provisioning trips. The two observed flight patterns may represent the different provisioning behavior for female and male cells. However, in order to prove this hypothesis, a longer series of provisioning trips of a greater number of females have to be documented and the mass of provi- sioned pollen has to be analyzed for sex specifically. Another reason for the observed pattern of pollen and nectar trips may be the in- 234 Journal of Hymenoptera Research fluence of weather conditions. Our under- flowers than Salix, but their reports con- lying hypothesis was that females make a tain no precise information whether these nectar trip after a particular hot and dry flower visits were for collecting pollen or day to increase humidity inside the cell. In nectar. In fact it is not proved at all that fact, Stephen (1966) noted that the tern- C. cunicularius is really oligolectic on Salix. perature at which flight activity was ini- Therefore it is not clear whether the col- lated in A. viburnella was a function of lection of other pollen, as observed in this weather conditions of the previous day. study, is a result of resource restriction. Yet, we did not find any correlation of oc- The correlation of increasing percentages currence of nectar trips to the climatic pa- of other pollen types in the pollen loads rameters of the previous day. On the con- of C. cunicularius with the end of flowering trary, after a bad weather day (no flying time of Salix may be an indication for a conditions the whole day) more females resource limitation. We registered all Salix made a nectar trip. This may be caused by trees within a radius of 3 km, most of an mcreased energy consumption after which were S. caprea. In the years 1996 to one day in the nest. Probably females pro- 1998, these trees were blooming very early vide themselves with nectar during their and had ceased to flower before females pollen collecting trips (male Salix plants of C. cunicularius and A. vaga began to col- also produce nectar). lect pollen. Only several bushes of S. au- Oligolecty. — Qualitative comparison of ricula were available during nest provi- collected pollen of both bee species indi- sioning time of both species. To prove cated important differences in diet whether C. cunicularius collects only other breadth between the two species: A. vaga pollen when Salix is not available, corn- collected almost pure Salix pollen, where- parative studies with resource quantifica- as C. cunicularius collected also a high per- tions at other locations from different centage of other pollen types. The fact that years are needed. whole cells contained exclusively other In conclusion, C. cunicularius can not be pollen types indicates that females of C. regarded as an oligolectic species. The use cunicularius systematically collect pollen of other, longer blooming host plants, from other host plants. Early flowering, which are more abundant in the study tree-like Rosaceae in particular seem to be area, may increase the reproductive suc- of great importance to this bee species, cess of this species. In contrast to C. cuni- Mader (1999) listed a number of species of cularius, A. vaga seems not to be affected Colletes having a relationship to Rosaceae. by the problem of long searching times for In fact, the Nearctic C. thoracicus Smith pollen sources, since it collected only Salix and C. nigrifrons Titus are even specialized pollen. However, the activity time of A. on Rosaceae. In our study, we found also vaga ceased approximately 4 weeks before pollen loads containing only Quercus, Acer that of C. cunicularius and the problem of or Ilex pollen, indicating that not only Ro- pollen availability probably did not yet oc- saceae can replace missing Salix plants. On cur. the Turkish coast, C. cunicularius females Niche differentiation.— A. vaga and C. were observed foraging on Pistacia; Salix cunicularius use the same host plant. This did not occur at this location. In Italy, C. overlap may result in interspecific com- cunicularius females were observed exclu- petition, if resources are limited. Since sively on Fabaceae (Kuhlmann in litt). availability of specific pollen is one of the The whole complex of species, subspecies most relevant niche parameters for bees and their host-plants seems not yet clear. (Eickwort 1973, Strickler 1979), interspe- Mader (1999) cited several authors which cific competition in case of a resource re- observed C. cunicularius on many other striction seems very likely. Niche differ- Volume 12, Number 2, 2003 235 entiation is often the basis for the coexis- tence of competitors. There are a number of ways in which niches can be differen- tiated. In this case the niches of the two species seemed to be differentiated on the basis of conditions. This means that they use the same resource but their ability to do so is influenced by environmental con- ditions and they respond differently to these conditions (Begon et al. 1990). The two species show diurnal differences in their foraging behavior. This temporal separation is influenced by climatic pa- rameters such as temperature. Whether C. cunicularius uses other host plants because of resource restriction and/ or competition with other species (besides A. vaga two other Andrena species special- ized on Salix occur in the study area) or whether it is not oligolectic at all can only be proved with removal experiments and manipulation of the resource availiability. ACKNOWLEDGMENTS We thank Annemarie Gossmann, Dr. Beate Kubitz and Dr. Martina Stebich (Institute for Palaeontology of the University of Bonn), for introducing us to the methods of pollen acetolysis and pollen quantifica- tion with Lycopodium spores. We are grateful to Dirk Schiffler and Andre Hamm (Institute for Agricultural Zoology and Bee-Research), for providing a cell coun- ter and an introduction to this method. We would like to thank Prof. Dr. G. Vorwohl for determination of pollen types other than Salix. Special thanks also to Dr. Rainer Hutterer, Dr. Anrje Bischoff and Dr. Bradley Sinclair, who reviewed the manuscript. We would like to thank dozens of students who assisted during the field work. We are grateful to Prof. Dr. C. M. Naumann for supporting this study. This work was funded by a Ph.D. grant from the Deutsche Bun- desstiftung Umwelt to the senior author. LITERATURE CITED Alcock, J. 1979. The relation between body size and provisioning behavior in the bee Centris pallida (Hymenoptera: Anthophoridae). Journal of the Kansas Entomological Society 52: 623-632. Batra, S. W. T. 1980. Ecology, behavior, pheromones, parasites and management of sympatric vernal bee Colletes inaequalis, C. thoracicus and C. validus. Journal of the Kansas Entomological Society 53: 509- 538. Batra, S. W. T. 1999. Biology of Andrena (Scapteropsis) fenningeri Viereck (Hymenoptera: Andrenidae), harbinger of spring. Proceedings of the Entomolog- ical Society of Washington 101: 106-122. Begon, M, Harper, J. L. and C. R. Townsend 1990. Individuals, populations and communities. Blackwell Scientific Publications, Cambridge. Bischoff, I. 2000. Untersuchungen zur Nisthabi- tatwahl von Andrena vaga und Colletes cunicular- ius (Apidae) in der Wanner Heide (Rheinland). Beitrage Hymenopterologen-Tagung Stuttgart 2000: 31-36. Bischoff, I. 2001. Populationsdynamik, Sammelstrategie und Nisthdbitatwahl ausgewahlter Wildbienen (Hy- menoptera, Apidae) in der Wahner Heide (Rhein- land). Shaker, Aachen. Braue, A. 1916. Die Pollensammelapparate der bein- sammelnden Bienen. Jenaische Zeitschrift fur Na- turwissenschaft 50 (N.F. 43) (1913): 1-96. Brechtel, F. 1986. Die Stechimmenfauna des Bienwal- des und seiner Randbereiche (Siidpfalz) unter besonderer Berucksichtigung der Okologie kunstnestbewohnender Arten. Pollichia-Buch Nr. 9: 282 pp., Bad Durkheim. Buchmann, S. L. and M. K. O'Rourke 1991. Impor- tance of pollen grain volumes for calculation bees diets. Grana 30: 591-595. Buchmann, S. L. and C. W. Shipman 1990. Pollen har- vesting rate for Apis mellifera L. on Gossx/pium (Malvaceae) flowers. Journal of the Kansas Ento- mological Society 63: 92-100. Chambers, V. H. 1946. An examination of the pollen loads of Andrena: the species that visit fruit trees. Journal of Animal Ecologi/ 15: 9-21. Crompton, W. C. and W. A. Wojtas 1993. Pollen grains of Canadian honey plants. Agriculture Canada. Re- search Branch. Publication 1892/E. Eickwort, G. C. 1973. Biology of the European mason bee, Hoplitis anthocopoidcs (Hymenoptera: Me- gachilidae), in New York state. Search 3: 1-31. Erdtmann, G. 1960. The acetolysis method. A revised description. Svoisk Botanisk Tidskrift 54: 561-564. Friese, H. 1882. Beitrag zur Biologie der Andrena pra- tcnsis Nyl. (ovuia Kl.). Entomologische Nachrichten 23: 317-319. Friese, H. 1923 Die europdischen Bienen (Apidae). — Das Leben und Wirken unserer Blumenwespen. — Walter de Gruyter, Berlin and Leipzig 1923. Gebhardt, M. and G. Rohr 1987. Zur Bionomie der Sandbienen Andrena clarkella (Kirby), A. cineraria (Linnaeus), A. fuscipes (Kirby) und ihrer Kuck- ucksbienen (Hymenoptera: Apoidea). Drosera 11: 89-114. Gerber, H. S. and E. C. Klostermeyer 1970. Sex control by bees: a voluntary act of egg fertilization dur- ing oviposition. Science 167: 82-84. Grinfeld, E. K. 1962. Origin and development of the apparatus for pollen collection in bees (Hyme- 236 Journal of Hymenoptera Research noptera: Apoidea). Entomological Review 41: 37- 42. Heinrich, B. 1979. Bumblebee economics. Havard Uni- versity Press. Helms, K. R. 1994. Sexual size dimorphism and sex ratios in bees and wasps. American Naturalist 143: 418-434. Interkommunaler Arbeitskreis Wahner Heide 1989. Die Wahner Heide. — Eine rheinische Landschaft im Spannungsfeld der Interessen. Rheinland, Koln. Johnson, M. D. 1981. Observations on the biology of Andrena (Melandrena) dunningi Cockerell (Hy- menoptera: Andrenidae). Journal of the Kansas En- tomological Society 54: 32-30. Johnson, M. D. 1988. The relationship of provision weight to adult weight and sex ratio in the soli- tary bee Ceratina calcerata. Ecological Entomology 13: 165-170. Kapyla, M. 1978. Bionomics of five wood-nesting sol- itary species of bees (Hym., Megachilidae), with emphasis on flower relationships. Biological re- search reports of the University Jyvaskyla 5: 3-89. Larsson, F. K. 1991. Some take it cool, some like it hot — a comparative study of male mate search- ing tactics in two species of Hymenoptera (Col- letidae and Sphecidae). journal of therm. Biology 16: 45-51. Levermann, E.-M., I. Bischoff, and T. Wagner. 2000. Species-specific foraging strategies of the syntop- ical and synchronous bees Panurgus calcaratus (Scopoli, 1763) and Dasypoda hirtipes (Fabricius, 1793). Beitriige zur Entomologic 50: 179-191. Lind, H. 1968. Nest provisioning cycle and daily rou- tine of behaviour in Dasypoda plumipes (Hym. Apidae). Entomologiske Meddeleser 36: 343-372. Linsley, E. G. 1958. The ecology of solitary bees. Hil- gardia 27: 543-599. Linsley, E. G. and J. W. MacSwain 1959. Ethology of some Ranunculus insects with emphasis on com- petition for pollen. University of California Publi- cations in Entomology 16: 1-46. Maddocks R. and H. F. Paulus 1987. Quantitative As- pekte der Brutbiologie von Osmia rufa L. und Os- mia comuta Latr. (Hymenoptera, Megachilidae): Eine vergleichende Untersuchung zu Mechanis- men der Konkurrenzverminderung zweier nah- verwandter Bienenarten. Zoologische Jahrbiichcr fur Systematik 144: 15-44. Mader, D. 1999. Gcologischc und biologische Entomook- ologie der rezenten Seidenbiene Colletes. Band 1. Lo- gabook Koln. Malyshev, S. I. 1926. The nesting habits of Andrena F. (Hymenoptera, Apoidea). Travaux de la Societedes Naturalistcs de Leningrad, Section Zoologie et Phy- siologic (Leningradskoe Obscestue Estestuoispytatetej Trudy) 56: 25-78 (in Russian with English sum- mary). Malyshev, S. I. 1927. Lebensgeschichte des Colletes cunicularius L. Zeitschrift fur Morphologic und Oko- logie der Tiere 9: 390-409. Malyshev, S. I. 1936. The nesting habits of solitary bees. A comparative study. Revista espahola de en- tomologfa (EOS) 11: 201-309. Michener, C. D. and C. W. Rettenmeyer 1956. The ethology of Andrena erythronii with comparative data on other species (Hymenoptera, Andreni- dae). University of Kansas Science Bulletin 37: 645- 684. Michener, C. D., M. L. Winston, and R. Jander 1978. Pollen manipulation and related activities and structures in bees of the family Apidae. Univer- sity of Kansas Science Bulletin 51: 575-601. Miliczky E. R. and E. A. Osgood 1995. Bionomics of Andrena (Melandrena) vicina Smith in Maine and Washington, with new parasite records for A. (M.) regularis Malloch and a review of Melandrena biology, journal of the Kansas Entomological Society 68: 51-66. Moeschler, A. 1938. Ein Beitrag zur Bienenfauna in Ostpreussen, insbesondere der Kurischen Neh- rung. Schriftcn der Physikalisch-Okonomischen Ge- scllschaft zu Konigsberg 70: 243-288. Mohamed, M. I. 1973. A new method for determining the number of journeys needed by solitary bees to provision a cell. Bulletin Societe Entomologic d' Egypt e 57: 435-440. Moore, P. D., J. A. Webb, and M. E. Collinson 1991. Pollen Analysis. Blackwell Scientific Publications, Oxford. Miiller, A. 1994. Die Bionomie der in leeren Schneck- enhausern nistenden Biene Osmia spinulosa (Kir- by 1802) (Hymenoptera, Megachilidae). Verof- fcntlichuugcu fur Naturschutz und Landschaftspflege in Badcn-Wiirttemberg 68/69: 91-334. Miiller, H. 1884 Ein Beitrag zur Lebensgeschichte der Dasypoda hirtipes. Vcrhandluugen des Naturhisto- rischen Vereins Preussen Rheinland und Westfalen 41: 1-52. Miinster-Swendsen, M. 1968. On the biology of the solitary bee Panurgus banksianus (Hymenoptera, Apidae), including some ecological aspects. Year- book Royal Veterinary Agricultural College Copen- hagen 1968: 215-241.' Parker, F. D. 1981. How efficient are bees in pollinat- ing sunflowers, journal of the Kansas Entomological Society 54: 61-67. Pasteels, J. J. and J. M. Pasteels 1979. Etude au micro- scope electronique a balayage des scopas collec- trices de pollen chez les Andrenidae (Hymenop- tera: Apoidea: Andrenidae). Archives de Biologic Paris 90: 113-130. Proctor, M., P. Yeo, and A. Lack 1996. The Natural History of Pollination. Timber Press, Portland. Rajotte, E. G. 1979. Nesting, foraging and pheromone response of the bee Colletes validus Cresson and its association with lowbush blueberries (Hv- Volume 12, Number 2, 2003 237 menoptera: Colletidae) (Ericaceae: Vaccinium). journal of the Kansas Entomological Society 52: 349- 361. Schonitzer K., and C. Klinksik 1990. The ethology of the solitary bee Andrena nycthemera Imhoff, 1866 (Hymenoptera, Apoidea). Entomofauna 11: 377- 427. Stephen, W. P. 1966. Andrena (Cryptandrena) viburnel- la. I. Bionomics. Journal of the Kansas Entomological Society 39: 42-51. Stockmarr, J. 1971. Tablets with spores used in ab- solute pollen analysis. Pollen and Spores 13: 614- 621. Stone, G. N. 1993a. Endothermy in the solitary bee Anthophora plumipes: Independent measures of thermoregulatory ability, costs of warm-up and the role of body size. Journal of Experimental Bi- ology 174: 299-320. Stone, G. N. 1993b. Thermoregulation in four species of tropical bees: the role of size, sex and altitude. journal of Comparative Physiology B 163: 317-326. Stone, G. N. 1994. Activity patterns of females of the solitary bee Anthophora plumipes in relation to temperature, nectar supplies and body size. Eco- logical Entomology 19: 177-189. Stone, G. N., J. N. Amos, T. F. Stone, R. L. Knight, H. Gay and F. Parrot 1988. Thermal effects on activ- ity patterns and behavioural switching in a con- cours of foragers on Stachytarpheta mutabilis (Ver- benaceae) in Papua New Guinea. Oecologia 77: 56-63. Stone, G. N., P. M. J. Loder, and T. M. Blackburn 1995. Foraging and courtship behaviour in males of the solitary bee Anthophora plumipes (Hymenop- tera: Anthophoridae): Thermal physiology and roles of body size. Ecological Entomology 20: 169- 183. Stone, G. N. and P. G. Willmer 1989. Warm-up rates and body temperatures in bees: The importance of body size, thermal regime and phylogeny. journal of Experimental Biology 147: 303-328. Strickler, K. 1979. Specialization and foraging effi- ciency of solitary bees. Ecology 60: 998-1009. Strickler, K. 1982. Parental investment per offspring by a specialist bee: does it change seasonally? Evolution 36: 1098-1100. Tasei, J. 1973. Le comportement de nidification chez Osmia (Osmia) cornuta Latr. et Osmia (Osmia) rufa L. (Hymenoptera, Megachilidae). Apidologie 4: 195-225. Thorp, R. W. 1969. Systematics and ecology of bees of the subgenus Diandrena (Hymenoptera: An- drenidae). University of California Publications m Entomology 52: 1-146. Vleugel, D. A. 1947. Waarnemingen aan het gedrag van de Grijze Graafbij (Andrena vaga Panz.) (Hym.). Entomologische Berichten 278: 185-192. Westerkamp, C. 1978. Zum Pollensammeln bauch- sammelnder Bienen (Hymenoptera: Megachili- dae) an Compositen. Sitzungsberichte der Gesell- schaft Naturforschender Freunde zw Berlin 18: 71- 79. ' Westerkamp, C. 1987. Das Pollensammeln der sozialen Bienen in Bezug auf die Anspassungen der Bliiten. Dissertation, Universitat Mainz. Westerkamp, C. 1996. Pollen in bee-flower relations. Some considerations on melittophily. Botanica Acta 109: 325-332. Westrich, P. 1990. Die Wildbienen Baden-Wiirttembergs. Bd I/II. Eugen Ulmer, Stuttgart. Westrich, P. and K. Schmidt 1987. Pollenanalyse, ein Hilfsmittel beim Studium des Sammelverhaltens von Wildbienen (Hymenoptera: Apoidea). Api- dologie 18: 199-214. Willmer, P. G. 1983. Thermal constraints on activity patterns in nectar-feeding insects. Ecological En- tomology 8: 455-469. Witt, R. 1992. Zur Bionomie der Sandbiene Andrena barbilabris (Kirby 1802) und ihrer Kuckucksbi- enen Nomada alboguttata Herrich-Schaffer 1839 und Sphecodes pellucidus Smith 1845. Drosera 6: 47-81. Wolda, H. and D. W. Roubik 1986. Nocturnal bee abundance and seasonal activity in a Panamani- an forest. Ecology 67: 426-433. J. HYM. RES. Vol. 12(2), 2003, pp. 238-271 Taxonomic Revision of the Genus Sesioctonus Viereck (Hymenoptera: Braconidae: Agathidinae) Rosa A. Briceno G. Universidad Centroccidental "Lisandro Alvarado", Decanato de Agronomia, Dpto. de Ciencias Biologicas-Seccion Entomologia, Tarabana, Cabudare, Estado Lara, Venezuela, e-mail: rabricen@telcel.net.ve Abstract. — The Neotropical genus Sesioctonus Viereck and its type species S. parathyridis Viereck, are redescribed. Descriptions for twenty-six new species are presented: S. acrolophus, S. amazo- nensis, S. ammosakron, S. analogus, S. areolatus, S. ariasi, S. armandoi, S. biospleres, S. boli- viensis, S. brasiliensis, S. chaconi, S. chrestos, S. clavijoi, S. diazi, S. dichromus, S. dominions, S. eumenetes, S. galeos, S. garciai, S. grandis, S. kompsos, S. miyayensis, S. peruviensis, S. qui, S. theskelos, and S. venezuelensis. Seventy-six morphological and color characters are discussed. A key to species is presented. Sesioctonus Viereck 1912 is a Neotropical genus of the subfamily Agathidinae that, together with the genus Earinus Wesmael, comprise the tribe Earinini (Sharkey 1992). Viereck proposed Sesioctonus for the unique species described so far: Sesiocton- us parathyridis Viereck. The main characters that diagnose Se- sioctonus are: tarsal claws simple, and no- tauli absent (Viereck 1912; Sharkey 1997). Many members of Sesioctonus have showy color patterns with body size varying be- tween 3.0-12.0 mm. Despite the vivid col- or patterns and relatively large size, this genus is poorly represented in insect col- lections, and there is little biological infor- mation. Most species of Sesioctonus have been collected from November until March and between 100-2000 meters above sea level. Almost all specimens were collected in Malaise traps. All known agathidines are parasitoids of Lepidoptera larvae (Sharkey 1988, 1997). The only previously known species of Sesioctonus, S. parathyridis, is recorded as a larval parasitoid of Parathyris perspi- cilla Stall (Lepidoptera: Arctiidae), al- though precise biology data (life cycle) are not known (Viereck 1912, 1914). Sharkey (1997) estimated that this genus might in- clude about thirty species and this revi- sion includes a total of twenty-seven spe- cies. MATERIALS AND METHODS Specimens are deposited in the follow- ing collections: AEI: American Entomological Insti- tute, Gainesville, FL. USA. BMNH: The Natural History Museum, London. England. CNC: Canadian National Collection, Biosystematics Research Centre, Agriculture Canada, Ottawa, On- tario. Canada. CUIC: Cornell University Insect Collec- tion, Department of Entomology, Cornell University, Ithaca, NY. USA. FML: Fundacion Miguel Lillo, Instituto de Zoologia Miguel Lillo, Tucu- man. Argentina. INBio: Instituto de Biodiversidad, Santo Domingo, Heredia. Costa Rica. INPA: Instituto Nacional de Pesquisas da Amazonia. Depto. de Ecologia e Biologia Evolutiva, Universida- Volume 12, Number 2, 2003 239 Figs. 1-4. 1, Subpronope triangular in dorso-lateral view. 2, Subpronope oval-shaped in dorso-lateral view. 3-4, Epicnemial carinae. 3, Complete and bilobed. 4, Incomplete and straight. de Federal de Sao Carlos, Sao Carlos, SP. Brasil. MIZA: Museo del Instituto de Zoologia Agricola "Francisco Fernandez Yepez", Universidad Central de Venezuela, Maracay, Aragua. Ve- nezuela. MZLU: Museum of Zoology, Lund Uni- versity, Sweden. TMB: Termeszettudomanyi Miizeum, Budapest. Hungary. UCOB: Museo Entomologico "Jose M. Osorio", Universidad Centrocci- dental "Lisandro Alvarado", Tar- abana, Lara. Venezuela. UCR: Universidad de Costa Rica, San Jose. Costa Rica. UK: Department of Entomology, Uni- versity of Kentucky, KY. USA. USNM: United States National Museum, Smithsonian Institution, Wash- ington, D. C. USA. UWY: U. W. Insect Museum, University of Wyoming, Laramie, WY. USA. ZSBS: Zoologische Sammlungen des Bayerischen Staates, Munich, Ger- many. The revision was carried out at MIZA at the Universidad Central de Venezuela in Maracay, Venezuela. Several morphometric characters were used in the descriptions and key. Mea- surements of Body length (mm) do not in- clude the ovipositor. Hind femur length and width (Fig. 5) were measured in lateral view, with the length taken from the distal end of the trocantellus to the apex of the femur. The maximum width was mea- sured, which is always near the midpoint. First metasomal tergite length and width re- 240 Journal of Hymenoptera Research lationship (Fig. 7) refers to the median ter- gite with the length taken along the mid- line and the width measured at the apex. Ovipositor length (mm) was measured from the base of the hypopygium. In the de- scription a measurement in parentheses following a range belongs to the holotype. The terminology used for the carinae and areolae of the metanotum and pro- podeum follows Sharkey (1988), and the rest of the terminology, including wing venation follows Sharkey and Wharton et al. (1997). Additional terms used in this revision include the occipital tubercles, a pair of projections, one at each side of the occipital medial line (Figs. 16-18); and the median areola of the metanotum, which is the central area of metanotum often bordered with carinae (Figs. 25-29). The key to Sesioctonus species was gen- erated with the software DELTA version 4.07 (Dallwitz et al. 1997). Seventy six me- ristic, morphological and color characters were used for the matrix. The species de- scriptions also were generated with this software. The biology for all the species described is unknown. RESULTS AND DISCUSSION Diversity and distribution of Sesiocton- us.— The genus is distributed from South- ern Mexico to Southern Brazil. Members of Sesioctonus are poorly represented in entomological collections but most of these have been collected in Malaise traps in tropical localities between 0 and 1500 meters above sea level. Only S. areolatus, has been collected above 2000 meters, spe- cifically at the Estacion Biologica Las Al- turas, in Costa Rica. The poor collecting in many regions of the Neotropics may explain the scarcity of this genus in collections. This is clearly shown by the numerous specimens from Costa Rica and Panama where collecting has been extensive in recent years. Color patterns. — Color patterns (yellow and black) shown by many braconids, in- cluding Sesioctonus, are frequently shared with other orders of insects, specially Co- leoptera and Hemiptera. However, some Lepidoptera, Neuroptera and Diptera also exhibit (Quicke 1997) these color patterns. Certainly, specimens of Sesioctonus could be confused with specimens of some of these groups when they are observed in nature, but the existence of a large mi- metic complex has not yet been conclu- sively demostrated. At taxonomic level, the various combination of yellow, or- ange, and black colors on the body of wasps are important characters for the separation of species. Wing color pattern. — The color pattern of the wings seems to be related to geo- graphical distribution. Most species from South America have the wings banded from the base: yellow, infuscate, yellow, infuscate; while most species from Central America have the wings entirely infuscate. Another curious fact is that the only spe- cies from an island (S. dominions) has hy- aline wings. These observations could suggest that there are mimetic relation- ships with other insects on the continent, or perhaps the presence of a sham apose- matic behavior, which was not present in the island. Genus Sesioctonus Viereck Sesioctonus Viereck 1912:1. Type species: Sesioc- tonus parathyridis Viereck. (Monobasic and original designation). Viereck 1914: 133. Diagnosis. — Sesioctonus species may be distinguished from other agathidines us- ing the following combination of charac- ters: Body smooth and bright, lacking sculpture, scutellar depression smooth, notauli absent, tarsal claws simple (Fig. 32), epicnemial carina bilobed between fore coxae; rarely (8%) straight, ovipositor about as long as body length. Description. — Head: Antenna with 23-48 flagellomeres, usually with 28-35; inter- antennal space with distinct keel (14%), or flat without distinct keel; antennal sockets Volume 12, Number 2, 2003 241 ^ Foretibial spines 7^ first metasomal median tergite width Figs. 5-8. 5, Hing legs of Braconidae (Wharton et al. 1997) showing measurements done. 6, Apex of foretibia with spines. 7, First metasomal median tergite showing measurements done. 8, Propodeum, a, with central areola; b, without areola. 242 Journal of Hymenoptera Research Figs. 9-10. Venation in Sesioctonus. 9, Forewing, a, (RS+M)a vein incomplete; b, 3RSa vein, hind wing; c, Cub vein not tubular. 10, Forewing, a, (RS+M)a vein complete. excavated (51%); face rarely with longitu- dinal median carina; mandibles with two teeth, usually the outer tooth as long as the inner; maxillary palpus with four or five segments; labial palpus with three or four segments; third segment of labial pal- pus, when present, shorter and partially fused to fourth segment; gena usually moderately expanded posteroventrally (37%), sometimes not expanded; occipital tubercles often present (51%); occiput usu- ally excavated. Mesosoma: Smooth and bright, without sculpture; subpronope tri- angle-shape or oval-shape; notauli always absent; scutellum generally convex in lat- eral view, rarely flat; lateral carina of the scutellar depression usually absent; me- dian areola of the metanotum often with longitudinal carina and lateral carinae that join posteriorly or not, generally smooth, sometimes with small rugosities; propo- deum convex (49%) or flat (51%) in lateral view; longitudinal carina of propodeum usually absent, sometimes present; epic- nemial carina absent (6%) or present (94%), complete (70%) or incomplete lat- Volume 12, Number 2, 2003 243 erally (24%), generally bilobed medially between fore coxae, sometimes straight; hind coxal cavities open, forming a com- mon foramen with the metasoma; fore wing (RS+M)a vein present, complete (57%) or incomplete (43%); fore wing 3RSa usually present; hind wing with 3-10 hamuli, generally 4-6; hind wing 2-1A vein usually tubular; hind wing CUb vein not tubular; fore tibia sometimes with spines; medial tibia usually with 2-16 spines; hind tibia with 5-25 spines, usu- ally 12-17; tarsal claws simple on all legs. Metasoma: Smooth and bright, without sculpture; median tergite of the first me- tasomal segment usually with two longi- tudinal dorsal carinae, rarely with pits posterad spiracle; ovipositor as long as length of the body. Length. 2.0-12. 0mm; excluding ovipositor. KEY TO SPECIES OF SESIOCTONUS 1 Occipital tubercles present (Figs. 16-18) 2 Occipital tubercles absent (Figs. 19) 13 2(1) Epicnemial carina straight medially or absent (indented at midline, between forecox- ae), sometimes difficult to see (Figs. 4, 23) 3 Epicnemial carina bilobed medially, (indented at midline, between the forecoxae) (Figs. 3, 22) 5 3(2) Epicnemial carina complete in lateral view (Figs. 3, 22) garciai sp. n. Epicnemial carina incomplete or absent in lateral view (Fig. 23) 4 4(3) Face with median longitudinal carina (Fig. 13) acrolopilms sp. n. Face without median longitudinal carina (similar to Figs. 12, 14) analogns sp. n. 5(2) Midcoxa not completely melanic, color variable 6 Midcoxa completely melanic 7 6(5) Fore tibia with spines (Figs. 6, 30), midcoxa yellowish orange, forewing (RS+M)a vein complete (Fig. 10a) and 3RSa vein present (Fig. 9b) peruviensis sp. n. Fore tibia without spines, midcoxa melanic dorsally and yellowish orange ventrally, forewing (RS+M)a vein complete (Fig. 10a) and 3RSa vein absent (Fig. 10) chaconi sp. n. 7(5) Longitudinal carina of scutellar depression present venezuelensis sp. n. Longitudinal carina of scutellar depression absent 8 8(7) Median areola of metanotum with longitudinal rugosities (Fig. 29), median tergite of first metasomal segment without pair of lateral longitudinal carinae (similar to Fig. 34), fore wing (RS+M)a vein complete (Fig. 10a) kompsos sp. n. Median areola of metanotum smooth (Figs. 25-28) 9 9(8) Mesoscutum melanic 10 Mesoscutum yellowish orange 11 10(9) Fore wing infuscate with large hyaline spot, metasoma reddish brown brasiliensis sp. n. Fore wing either infuscate without hyaline spot or hyaline basally, infuscate apically, mesosoma melanic and metasoma yellowish orange dichromus sp. n. 11(9) Median longitudinal carina of propodeum present and complete ariasi sp. n. Median longitudinal carina of propodeum absent or incomplete 12 12(11) Subpronope triangular (Fig. 1), fore wing 3RSa vein absent (Fig. 10) . . boliviensis sp. n. Subpronope oval-shape (Fig. 2), fore wing 3RSa vein present (Fig. 9) diazi sp. n. 244 Journal of Hymenoptera Research 13(1) Occiput excavated (similar to Figs. 16-18) eumenetes sp. n. Occiput not excavated (Fig. 19) 14 14(13) Median areola of metanotum with lateral carinae (Figs. 25-28), flagellum with less than 40 flagellomeres, interantennal space with a rounded longitudinal keel or keel absent (Fig. 12), never sharp; specimens with less than 8mm in body length 15 Median areola of metanotum without sharp lateral carinae (Fig. 29), flagellum with 40 flagellomeres or more, interantennal space with a sharp longitudinal keel (Fig. 11), specimens greater than 8mm in body length 25 15(14) Median areola of metanotum with lateral carinae present and meeting posteriorly (Figs. 25, 26) 16 Median areola of metanotum with lateral carinae present and not meeting posteriorly (Figs. 27, 28) 23 16(15) Epicnemial carina present (Figs. 3, 4) 17 Epicnemial carina completely absent ammosakron sp. n. 17(16) Epicnemial carina complete laterally (Fig. 3) 18 Epicnemial carina incomplete laterally (Fig. 4) 21 18(17) Hind tibia melanic amazonensis sp. n. Hind tibia mostly yellowish orange 19 19(18) Propodeum with central areola absent 20 Propodeum with central areola present (Fig. 8a) areolatus sp. n. 20(19) Flagellum with 32 flagellomeres, interantennal space with rounded longitudinal keel (similar to Fig. 12), hind tibia yellowish orange in basal half, melanic apically . . . miyayensis sp. n. Flagellum with 25 flagellomeres, interantennal space without longitudinal keel, hind tibia mostly yellowish orange, melanic apically clavijoi sp. n. 21(17) Epicnemial carina straight medially (between forecoxae) (Fig. 4), body length less than 3mm dominions sp. n. Epicnemial carina bilobed medially (indented at midline, between forecoxae) (Fig. 3), body length more than 3mm 22 22(21) Forewing (RS + M)a vein complete (Fig. 10a) armandoi sp. n. Forewing (RS + M)a vein incomplete (Fig. 9a) biospleres sp. n. 23(15) Epicnemial carina present, complete or incomplete laterally (Figs. 3, 4) 24 Epicnemial carina completely absent chrestos sp. n. 24(23) Epicnemial carina straight medially (indented at midline, between forecoxae) (Fig. 4) galeos sp. n. Epicnemial carina bilobed medially (indented at midline, between forecoxae) (Fig. 3) theskelos sp. n. 25(14) Third and fourth labial palpomeres not fused, first metasomal median tergite with depression posterad spiracle (Figs. 36, 37) grandis sp. n. Third and fourth labial palpomeres fused, first metasomal median tergite with or with- out depression posterad spiracle 26 26(25) First metasomal median tergite with depression posterad spiracle (similar to Figs. 36, 37) qui sp. n. First metasomal median tergite without depression posterad spiracle parathyridis Viereck Volume 12, Number 2, 2003 245 Sesioctomis acrolophns Briceno, sp. n. Figs. 11, 13, 15, 21, 23, 24, 25, 30, 32, 35 Diagnosis. — Face with median longitu- dinal carina, interantennal space with a sharp longitudinal keel and median areola of metanotum with median longitudinal carina. S. analogous is similar but can be distinguished by this combination of char- acters. Description. — 9. Length. Body, 9-10 mm (9.5). Head: Flagellum with 32 flagello- meres. Interantennal space with sharp lon- gitudinal keel. Antennal sockets deeply excavated. Face with median longitudinal carina. Genae strongly expanded poster- oventrally. Occipital tubercles present. Oc- ciput excavated. Mandible concave, outer tooth longer than inner tooth. Maxilla with 5 palpomeres. Third and fourth labial palpomeres not fused. Mesosoma: Sub- pronope oval. Longitudinal carinae of scu- tellar depression absent. Scutellum con- vex. Median areola of metanotum smooth; with median longitudinal carina; and with lateral carinae present and meeting pos- teriorly. Propodeum convex, median lon- gitudinal carina present. Epicnemial cari- na blunt, incomplete dorsally, straight me- dially (between forecoxae). Foretibial spines present. Midtibia with 8 spines. Hind tibia with 11 spines. Hind femur 3.17 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of fore- wing present. 2-1A vein of hind wing not tubular. CUb vein of hind wing not tu- bular. Hind wing with 4-5 hamuli (4). Me- tasotna: Median tergite of first metasomal segment with pair of lateral longitudinal carinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 1.09. Ovipositor 9-10 mm (9.2). Color: Head melanic except maxil- lary and labial palpomeres sometimes yel- lowish orange. Antenna melanic. Meso- soma yellowish orange. Forelegs melanic except coxae yellowish orange. Midlegs melanic except coxae sometimes yellowish orange. Hindleg melanic except coxa and femur mostly yellowish orange, but me- lanic distally. Forewing yellow basally and infuscate apically. Stigma melanic, or yellowish orange. Hind wing yellow ba- sally infuscate apically. Metasoma yellow- ish orange. Ovipositor yellowish orange. 6 . — Unknown. Material examined. — Holotype: Costa Rica: 9, CRI002 492066, Prov. Alajuela, sector Colonia Palmarena, 700m, oct.1996, G. Carballo (INBio). Paratopes: Costa Rica: 9, Heredia, Est. Biol. La Selva, 50-1 50m, 10° 26'N 84° 01 'W, 02 de Marzo 1993, bosque primario (INBio); 9, R. San Lor- encito, 900m, R. F. San Ramon, 5km N de Colonia Palmarena, Prov. Alajuela, 13-18 Junio.1993, (without abdomen) (INBio); 9, Heredia, 3km S. Puerto Viejo, OTS, La Sel- va, 100m, xii.1992, P Hanson, (UWY); 9, Heredia, 3km S. Puerto Viejo OTS, La Sel- va, 100m. l-15.ix.1992, P. Hanson, huer- tos, set de trampas malaise de G. Wright (UCR); 9, La Selva, 15.xii.1993, J Longino (M/04/272) (UK); 29, Heredia, Est. Biol. La Selva. 50-150m, 10° 26'N 84° 01' W, x.1992, P Hanson, C Godoy (UCR) (UCOB); 9, Limon. 16km W. Guapiles, 400m, i-iv. 1992, col. Paul Hanson (UCR). Distribution. — Known only from low- land Atlantic rain forests in Costa Rica, up to 900m. Etymology. — From Greek acrolophus that means keel, in reference to the longitudi- nal carina on the face in this species. Sesioctomis amazonensis Briceno, sp. n. Diagnosis. — S. amazonensis share charac- ters with S. armandoi from which it is sep- arated by the lack of longitudinal rugosi- ties on the median areola of metanotum and the presence of a complete epicnemial carina. Description. — 9. Length. Body 5-7 mm (7.0). Head: Flagellum with 35-38 flagel- lomeres (38). Interantennal space with rounded longitudinal keel. Antennal sock- ets moderately excavated. Face without median longitudinal carina. Genae not ex- 246 Journal of Hymenoptera Research Figs. 11-15. Head. 11, S. acrolophus, arrow shows interantennal space with a sharp longitudinal keel. 12, S. dichromus in dorsal view showing interantennal space with a rounded longitudinal keel or absent. 13, S. acrolophus, arrow shows face with median longitudinal carina. 14, S. brasiliensis. 15, S. acrolophus, arrow shows expanded gena. panded posteroventrally. Occipital tuber- cles absent. Occiput not excavated. Man- dible flat, outer tooth of mandible not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial pal- pomeres completely fused. Mesosoma: Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum smooth; without median longitudinal Volume 12, Number 2, 2003 247 carina; and with lateral carinae present and meeting posteriorly. Propodeum con- vex, median longitudinal carina absent. Epicnemial carina, sharp, complete, bi- lobed medially (between forecoxae). Fore- tibial spines absent. Midtibia with 5-9 spines (9). Hind tibia with 10 spines. Hind femur 3.51 times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing absent. 2-1 A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 4-5 hamuli (5). Metasoma: Median tergite of first meta- somal segment without pair of lateral lon- gitudinal carinae. First metasomal median tergite without depression posterad spi- racle. Length width ratio of first metaso- mal median tergite 1.01. Ovipositor 7 mm. Color: Head melanic. Antenna melanic. Maxillary palpomeres yellowish orange except basal two palpomeres melanic. La- bial palpomeres yellowish orange. Meso- soma yellowish orange. Forelegs yellow- ish orange except femur sometimes me- lanic ventrally. Midlegs yellowish orange except tibia and tarsus melanic. Hindleg melanic with coxa yellowish orange. Fore- wing banded from base, yellow, infuscate, yellow, infuscate. Stigma melanic. Hind wing banded from base, yellow, infuscate, yellow, infuscate. Metasoma yellowish or- ange. Ovipositor yellowish orange. 6 . — Unknown. Material examined. — Holotype. 9 . Brasil: Amazonas, Res. Ducke, 26km NE Manaus, 22.07.1981, J.A. Rafael, trampa malaise (CNC). Paratypcs. BRASIL: 9, same data as holotype (CNC); 9, Manaus, ZF3, Km 23, Faz.Esteio, Res 1112; B. Klein col. 21.1.1986 (INPA). Distribution. — This species is known only from the Amazonas region of Brasil. Etymology. — This species is named after the locality of the holotype specimen. Sesioctonus ammosakron Briceno, sp. n. Diagnosis. — Epicnemial carina absent, a characteristic shared with S. chrestos and S. grandis. However, fore tibial with spines are present the longitudinal carina of the median metasomal tergite are absent in S. chrestos. S. grandis can be separated by its larger size and the presence of 4 maxillary palpomeres. Description. — 9. Length. Body 3.5 mm. Head: Flagellum with 24 flagellomeres. In- terantennal space without longitudinal keel. Antennal sockets not excavated. Face without median longitudinal carina. Ge- nae not expanded posteroventrally. Occip- ital tubercles absent. Occiput not excavat- ed. Mandible concave, outer tooth of man- dible not longer than inner tooth. Maxilla with 5 palpomeres. Third and fourth labial palpomeres not fused. Mesosoma: Sub- pronope oval. Longitudinal carinae of scu- tellar depression absent. Scutellum con- vex. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and meeting posteriorly. Propodeum convex, median longitudinal carina absent. Epicnemial ca- rina completely absent. Foretibial spines absent. Midtibia with 9 spines. Hind tibia with 15 spines. Hind femur 3.14 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of forewing absent. 2-1A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 3 hamuli. Metasoma: Median tergite of first metasomal segment with pair of lat- eral longitudinal carinae. First metasomal median tergite without depression poster- ad spiracle. Length width ratio of first me- tasomal median tergite 1.32. Ovipositor 3.5 mm. Color: Head yellowish orange ex- cept vertex and occiput melanic. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Mesosoma yellowish orange sometimes melanic. Forelegs yel- lowish orange. Midlegs yellowish orange. Hindleg yellowish orange except tibia and tarsus melanic distally. Forewing entirely infuscate. Stigma melanic. Hind wing en- tirely infuscate. Metasoma yellowish or- ange but third tergum with median tergite melanic in posterior quarter; fourth ter- gum with median tergum melanic and 248 Journal of Hymenoptera Research fifth to eighth metasomal terga mostly yel- lowish orange but median tergites melanic centrally. Ovipositor yellowish orange. 8 . — Unknown. Material examined. — Holotype. Costa Rica: 9, Puntarenas, San Vito, Est. Biol. Las Alturas, 1500m. iii.1992, Paul Hanson. (UWY). Distribution. — This species is known only from the Puntarenas region of Costa Rica. Etymology. — From Greek aminos that means sand and akron that means top, in reference to holotype locality, Puntarenas. Sesioctomts analogus Briceno, sp. n. Diagnosis. — S. analogus can be distin- guished from most Sesioctonus species by the presence of 5 maxillary palpomeres and 4 labial palpomeres. However, this character occurs in 5 other species of the genus, of which S. acrolophus is closest to S. analogus. These two species are separat- ed by the presence of a longitudinal carina on the face of S. acrolophus, which is absent in S. analogus. Description. — 9. Length. Body 10 mm. Head: Flagellum with broken after flagel- lomere 20. Interantennal space with rounded longitudinal keel. Antennal sock- ets not excavated. Face without median longitudinal carina. Genae strongly ex- panded posteroventrally. Occipital tuber- cles present. Occiput excavated. Maxilla with 5 palpomeres. Third and fourth labial palpomeres not fused. Mesosoma: Sub- pronope oval. Longitudinal carinae of scu- tellar depression absent. Scutellum con- vex. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and meeting posteriorly. Propodeum convex, median longitudinal carina present. Epicnemial carina blunt, incomplete dorsally, straight medially (between forecoxae). Foretibial spines present. Midtibia with 18 spines. Hind tibia with 21 spines. Hind femur 2.88 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of fore- wing present. 2-1 A vein of hind wing tu- bular. CUb vein of hind wing not tubular. Hind wing with 4 hamuli. Metasoma: Me- dian tergite of first metasomal segment with pair of lateral longitudinal carinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal median ter- gite 0.99. Ovipositor 8 mm. Color: Head melanic. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Me- sosoma yellowish orange. Forelegs melan- ic sometimes coxa and tarsus yellowish orange. Midlegs melanic sometimes coxa, femur and basitarsus yellowish orange. Hindleg melanic, sometimes coxa, femur and basitarsus yellowish orange basally. Forewing yellow basally and infuscate apically. Stigma melanic. Hind wing yel- low basally infuscate apically. Metasoma yellowish orange. Ovipositor yellowish orange. 6 . — Unknown. Material examined. — Holotype: Costa Rica: 9, Est. Pitilla, 9km S, Sta. Cecilia, A.C., Guanacaste, Prov. Guana, 700m, v.1994, P. Rios, Malaise (INBio). Distribution. — This species is known only from lowlands of Costa Rica up to 700m. Etymology. — This species name refers to a similarity with S. acrolophus. Sesioctonus areolatus Briceno, sp. n. Diagnosis. — Presence of central areola on the propodeum, median longitudinal carina of the scutellar depression, foretib- ial with spines and epicnemial carina com- plete and straight medially. Also, the body size is small. Description. — 9. Length. Body, 4-5 mm (4.2). Head: Flagellum with 26-28 flagel- lomeres (26). Interantennal space without longitudinal keel. Antennal sockets not ex- cavated. Face without median longitudi- nal carina. Genae not expanded poster- oventrally. Occipital tubercles absent. Oc- ciput not excavated. Mandible concave, outer tooth of mandible not longer than Volume 12, Number 2, 2003 249 Rk> ^™$»iv fi ?^^^ |Pj '^^B ^■Rv j^^H Par ■tttt^ w£*' L Figs. 16-21. Head. 16-19, Occipital tubercles. 16, 18, S. dichromus. 17, S. diazi. 19, S. grandis without tubercles. 20-21, Mandible teeth. 20, S. dichromus with outer tooth not longer than inner tooth. 21, S. acrolophus with outer tooth longer than inner tooth. inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres com- pletely fused. Mesosoma: Subpronope tri- angular. Longitudinal carinae of scutellar depression present. Scutellum convex. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and meeting posteriorly. Propodeum flat, median lon- gitudinal carina present. Epicnemial cari- na sharp, complete, straight medially (be- tween forecoxae), sometimes bilobed. Foretibial spines present. Midtibia with 7- 8 spines (7). Hind tibia with 10-12 spines 250 Journal of Hymenoptera Research (10). Hind femur 3.35 times as long as acters: presence of occipital tubercles, me- wide. (RS+M)a vein of forewing incom- dian areola of metanotum with lateral ca- plete. 3RSa vein of forewing present. 2-1A rinae not meeting posteriorly, and vein of hind wing tubular. CUb vein of presence of the longitudinal carinae in the hind wing not tubular. Hind wing with 3 propodeum. This combination could con- hamuli. Metasoma: Median tergite of first fuse S. ariasi with S. acrolophus, however metasomal segment with pair of lateral these species can be separate easily be- longitudinal carinae. First metasomal me- cause the number of maxillary and labial dian tergite without depression posterad palpomere is fewer in S. ariasi. spiracle. Length width ratio of first meta- Description. — 9. Length. Body, 6-8.5 mm somal median tergite 1.04. Ovipositor 3- (8.0). Head: Flagellum with 30-34 flagel- 4.3 mm (4.0). Color: Head yellowish or- lomeres (33). Interantennal space with ange. Antenna melanic. Maxillary and la- rounded longitudinal keel. Antennal sock- bial palpomeres yellowish orange, some- ets not excavated. Face without median times melanic. Mesosoma entirely longitudinal carina. Genae not expanded yellowish orange. Forelegs yellowish or- posteroventrally. Occipital tubercles pre- ange. Midlegs yellowish orange some- sent. Occiput excavated. Mandible con- times tibia melanic. Hindleg yellowish or- cave, outer tooth of mandible not longer ange with tibia yellowish orange but me- than inner tooth. Maxilla with 4 palpo- lanic distally; and tarsus melanic, or yel- meres. Third and fourth labial palpomeres lowish orange. Forewing entirely infuscate completely fused. Mesosoma: Subpronope or infuscate with hyaline spots. Stigma oval. Longitudinal carina of scutellar de- melanic. Hind wing entirely infuscate. pression absent. Scutellum convex. Medi- Metasoma yellowish orange. Ovipositor an areola of metanotum smooth; without yellowish orange. median longitudinal carina; and with lat- 6 . — Unknown. eral carinae present and not meeting pos- Material examined. — Holotype. Costa teriorly. Propodeum convex, median Ion- Rica: 9, Puntarenas, San Vito, Est. Biol, gitudinal carina present. Epicnemial cari- Las Alturas, 2100m, iii-v.1995, Paul Han- na sharp, complete, bilobed medially (be- son (UWY). Paratypes. Costa Rica: 39, tween forecoxae). Foretibial spines absent. Puntarenas, San Vito, Est. Biol. Las Altu- Midtibia with 8 spines. Hind tibia with ras, 1500-2100m, v. 1992, iii-v.1995, Paul 14-16 spines (15). Hind femur 3.32 times Hanson (UWY); 9, Guanacaste Prov., as long as wide. (RS+M)a vein of fore- 300m, decidious woods (25-30 yrs), 4.ix- wing incomplete. 3RSa vein of forewing 5.x. 1985, Gauld & Janzen (UK). Honduras: present. 2-1A vein of hind wing tubular. 9 , Cortes, Parque Nacional Cusuco 5km CUb vein of hind wing not tubular. Hind N de Buenos Aires, 15°29'N 88°13'W, wing with 5 hamuli. Metasoma: Median 15.X.1995, malaise trap, R. Cave (MZLU). tergite of first metasomal segment without Distribution. — This species is known pair of lateral longitudinal carinae. First only from Costa Rica and Honduras. metasomal median tergite without depres- Etymology. — From Latin areolatus that sion posterad spiracle. Length width ratio means areolated, in reference to the cen- of first metasomal median tergite 1.18. tral areola on the propodeum in this spe- Ovipositor 5-8 mm (7.5). Color: Head me- cies- lanic. Antenna melanic. Maxillary pal- c • x ■ n • ~ pomeres melanic, sometimes yellowish or- beswctonus ariasi Bnceno, sp. n. . , , J , r ange except basal two palpomeres melan- Diagnosis.—S. ariasi can be distin- ic. Labial palpomeres melanic. Mesosoma guished from the rest of Sesioctonus spe- mostly yellowish orange with pronotum cies by the following combination of char- melanic anteriorly, propleuron and meta- Volume 12, Number 2, 2003 251 pleuron melanic, or yellowish orange, latter species is straight and not bilobed as sometimes melanic posteriorly. Forelegs S. armandoi. melanic except tibia melanic, or yellowish Description. — 9. Length. Body, 6 mm. orange and tarsus yellowish orange. Mid- Head: Flagellum with 34 flagellomeres. In- legs melanic except tarsus with basitarsus terantennal space with rounded longitu- sometimes yellowish orange in basal half, dinal keel. Antennal sockets moderately Hindleg melanic. Forewing entirely infus- excavated. Face without median longitu- cate. Stigma melanic. Hind wing entirely dinal carina. Genae not expanded poster- infuscate. Metasoma mostly yellowish or- oventrally. Occipital tubercles absent. Oc- ange except fifth to eighth metasomal ter- ciput not excavated. Mandible concave, ga melanic. Ovipositor yellowish orange Maxilla with 4 palpomeres. Third and except apical eighth melanic. fourth labial palpomeres completely 6\ — Essentially as female. fused. Mesosoma: Subpronope triangular. Material examined. — Holotype: Brazil: 9, Longitudinal carinae of scutellar depres- Matogrosso, 12° 31 S, 55° 37 W, ii. 1976, sion absent. Scutellum convex. Median M. Alvarenga (CNC). Parah/pes: Colom- areola of metanotum with longitudinal ru- bia: 8, Cundinamarca, Monterredondo, gosities; without median longitudinal ca- 14.xii.1958, J. Foerster (CNC), 9, Antio- rina; and with lateral carinae present and quia, Mun. San Luis R.N. Rio Claro, El Re- meeting posteriorly. Propodeum convex, fugio 5° 47'N, 75° O'W, 500m, malaise, median longitudinal carina absent. Epic- 13.i.98. Diego Campos (UK); 9, Amazon- nemial carina, sharp, incomplete dorsally, as, PNN. Amacayacu, Mocagua, 3° 23'S, bilobed medially (between forecoxae). 70° 06'W, 150m, malaise, 26.ii-12.iii.2001, Foretibial spines absent. Midtibia with 5 B. Amado (UK); 9, Cauca, PNN. Gorgona, spines. Hind tibia with 10 spines. Hind fe- Mancora, 2° 58' N, 78° 11' W, 60m, mal- mur 3.27 times as long as wide. (RS+M)a aise, 26.vi-18.vii.2000, H. Torres (UK), vein of forewing complete. 3RSa vein of Costa Rica: 8 , Est. Hitoy Cerere, 100m, R. forewing absent. 2-1 A vein of hind wing Cerere, Res. Biol. Hitoy Cerere, Prov. Li- tubular. CUb vein of hind wing not tu- mon, vii.1992, G. Carballo (INBio). Bolivia: bular. Hind wing with 4 hamuli. Metaso- 9 , Staudinger K (ZSBS); Ecuador: 8 , ma: Median tergite of first metasomal seg- Napo, Prov. Sacha, 9.iii.l983, L. Huggert ment without pair of lateral longitudinal (CNC). carinae. First metasomal median tergite Distribution. — From Costa Rica in Cen- without depression posterad spiracle, tral America to Brazi, Bolivia and Colom- Length width ratio of first metasomal me- bia in South America. dian tergite 0.94. Ovipositor 6 mm. Color: Etymology. — This species is named in Head melanic. Antenna melanic. Maxil- honor of Quintin Arias, for his friendship lary and labial palpomeres yellowish or- and advise regarding computer software. ange. Mesosoma melanic except metano- tum, propodeum and metapleuron yel- Sesioctonus armandoi Briceno, sp. n. }owish omnge Forelegs yeliowish orange. Diagnosis. — S. armandoi can be distin- Midlegs yellowish orange. Hindleg me- guished from the rest of Sesioctonus spe- lanic except trochanter and trocantellus cies by the following combination of char- yellowish orange. Forewing banded from acters: occipital tubercles absent, epicne- base, yellow, infuscate, yellow, infuscate. mial carina incomplete laterally, median Stigma melanic. Hind wing banded from areola of metanotum with longitudinal ru- base, yellow, infuscate, yellow, infuscate. gosities and with lateral carinae meeting Metasoma yellowish orange until third posteriorly. This combination is present in tergum, rest melanic. Ovipositor yellowish S. galeos but the epicnemial carina in this orange. 252 Journal of Hymenoptera Research Figs. 22-27. 22-23, Epicnemial carina. 22, Complete and bilobed in S. brasiliensis, 23, incomplete and straight in S. acrolophus. 24, S. acrolophus in lateral view, right arrow shows longitudinal carina of propodeum and left arrow first metasomal segment with pair of lateral longitudinal carinae. 25-27, Median areola of metanotum. 25, Smooth with longitudinal carinae and lateral carinae meeting posteriorly in S. acrolophus. 26, Smooth with lateral carinae meeting posteriorly in S. clavijoi. 27, Smooth with lateral carinae not meeting posteriorly in S. brasiliensis. 6\— Essentially as female. N, 70° 06' W, 150m, 7-19.vii.2000, A Par- Material examined. — Holotype: Ecuador: ente (UK); S , Amazonas, PNN Amacaya- 9, Napo & Coca Rivers, 2-10.V.1965, Luis cu Matamata, 8-12.iii.2000, Sharkey (UK). Pena (AEI). Paratypes: Colombia: $,Ama- Distribution. — This species is known zonas, PNN Amacayacu Mocagua, 3° 23' only from Ecuador and Colombia. Volume 12, Number 2, 2003 253 Etymology. — This species is named in honor of Armando Briceno, Venezuelan entomologist, and my uncle. Sesioctonus biospleres Briceno, sp. n. Diagnosis. — S. biospleres can be distin- guished by the following combination of characters: genae expanded posteriorly, occipital tubercles absent, epicnemial ca- rina incomplete laterally, foretibia with spines and median tergite of first metaso- mal tergum with pair of longitudinal ca- rinae. Description. — 9. Length. Body, 6-8 mm (6.5). Head: Flagellum with 30-35 flagel- lomeres (34). Interantennal space with rounded longitudinal keel, or keel lacking. Antennal sockets not excavated. Face without median longitudinal carina. Ge- nae expanded posteroventrally. Occipital tubercles absent. Occiput not excavated. Mandible concave, outer tooth of mandi- ble not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres completely fused. Mesosoma: Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum smooth; without median longitudinal carina; and with lateral carinae present and meeting posteriorly. Propodeum con- vex, median longitudinal carina present. Epicnemial carina, sharp, incomplete lat- erally, bilobed medially (between forecox- ae). Foretibial spines present. Midtibia with 6-10 spines (9). Hind tibia with 10- 14 spines (14). Hind femur 3.11 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of forewing absent. 2-1 A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 3 hamuli. Metasoma: Median tergite of first metasomal segment with pair of lat- eral longitudinal carinae. First metasomal median tergite without depression poster- ad spiracle. Length width ratio of first me- tasomal median tergite 0.98. Ovipositor 5.5-6.5 mm (6.3). Color: Head yellowish orange. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Me- sosoma yellowish orange. Forelegs yel- lowish orange except femur and tibia me- lanic, or melanic with yellowish orange apically. Midlegs yellowish orange except tibia melanic apically. Hindleg yellowish orange except tibia yellowish orange with apical third melanic and tarsus melanic with basitarsus yellowish orange in basal third. Forewing yellow basally and infus- cate apically. Stigma melanic. Hind wing yellow basally infuscate apically. Metaso- ma yellowish orange. Ovipositor yellow- ish orange. 6 . — Unknown. Material examined. — Holotype: Costa Rica: 9, Prov. Puntarenas, Est. Agujas, Rio Agujas, sendero Samia, 300m, 1-3. vi. 1997, A. Azofeifa (INBio). Paratypes: Costa Rica: 9, Rancho Quemado, 200m, Peninsula de Osa, Prov. Puntarenas, vi.1992, F. Quesa- da y M. Segura (INBio); 9, Rancho Que- mado, Peninsula de Osa, Prov. Puntaren- as, 200m, 01.xi-01.xii.1992, A. L. Marin (INBio). Panama: 9, Barro Colorado Is, 9°9'N 79°51'W, 11-18.V.1994, J. Pickering (UK). Distribution. — This species is known only from lowlands of Prov. Puntarenas, Costa Rica and Barro Colorado in Panama. Etymology. — From Greek bios meaning life and pleres that means abundance. Sesioctonus boliviensis Briceno, sp. n. Diagnosis. — S. boliviensis can be separat- ed by the following combination of char- acters: interantennal space with a longi- tudinal rounded keel and the presence of the occipital tubercles. Description. — 9. Length. Body, 10 mm. Head: Interantennal space with rounded longitudinal keel. Antennal sockets mod- erately excavated. Face without median longitudinal carina. Genae not expanded posteroventrally. Occipital tubercles pre- sent. Occiput excavated. Mandible con- cave, outer tooth of mandible not longer than inner tooth. Third and fourth labial palpomeres completely fused. Mesosoma: 254 Journal of Hymenoptera Research Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum smooth; without median longitudinal carina; and with lateral carinae present and not meeting posteriorly. Propodeum flat, median longitudinal carina absent. Epicnemial carina sharp, complete, bi- lobed medially (between forecoxae). Fore- tibial spines absent. Midtibia with 6 spines. Hind tibia with 19 spines. Hind fe- mur 3.21 times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing absent. 2-1A vein of hind wing tubular. CUb vein of hind wing not tu- bular. Hind wing with 5 hamuli. Metaso- ma: Median tergite of first metasomal seg- ment without pair of lateral longitudinal carinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 0.79. Ovipositor 10 mm. Col- or: Head melanic except labial palpomeres melanic. Antenna melanic. Mesosoma yel- lowish orange sometimes pronotum and propleuron melanic. Forelegs melanic. Midlegs melanic. Hindleg melanic. Fore- wing banded from base infuscate, yellow, infuscate. Stigma melanic. Hind wing en- tirely infuscate. Metasoma yellowish or- ange. Ovipositor yellowish orange. S . — Unknown. Material examined. — Holotype. Bolivia: 9, Staudinger K (ZSBS). Distribution. — This species is known only from Bolivia, South America. Etymology. — The name boliviensis refers to the country of origin of the holotype. Sesioctomts brasiliensis Briceno, sp. n. Figs. 14, 22, 27, 34 Diagnosis. — S. brasiliensis is the only spe- cies with metasoma reddish brown and forewing with a large hyaline spot. Also, the median areola of metanotum does not have longitudinal and lateral carinae, nor longitudinal rugosities. Occipital tubercles are present in this species. Description.— 9 . Length. Body, 9-10 mm (9.0). Head: Flagellum with 30-35 flagel- lomeres (33). Interantennal space with rounded longitudinal keel. Antennal sock- ets deeply excavated, sometimes moder- ately excavated. Face without median lon- gitudinal carina. Genae moderately ex- panded posteroventrally. Occipital tuber- cles present. Occiput excavated. Mandible concave, outer tooth of mandible not lon- ger than inner tooth. Maxilla with 4 pal- pomeres. Third and fourth labial palpo- meres completely fused. Mesosoma: Sub- pronope triangular. Longitudinal carinae of scutellar depression absent. Scutellum convex. Median areola of metanotum smooth; without median longitudinal ca- rina; and lacking lateral carinae. Propo- deum convex, median longitudinal carina absent. Epicnemial carina sharp, complete, bilobed medially (between forecoxae). Foretibial spines absent. Midtibia with 8- 11 spines (8). Hind tibia with 14-17 spines (14). Hind femur 3.64 times as long as wide. (RS+M)a vein of forewing com- plete. 3RSa vein of forewing present. 2-1A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 5- 6 hamuli (6). Metasoma: Median tergite of first metasomal segment without pair of lateral longitudinal carinae. First metaso- mal median tergite without depression posterad spiracle. Length width ratio of first metasomal median tergite 1.09. Ovi- positor 8 mm. Color: Head melanic. An- tenna melanic. Maxillary palpomeres yel- lowish orange. Labial palpomeres melan- ic. Mesosoma melanic. Forelegs mostly yellowish orange except coxae melanic, tibia yellowish orange, or melanic; and tarsus mostly yellowish orange, but apical tarsomere melanic. Midlegs mostly melan- ic except trochanter and trocantellus yel- lowish orange, femur yellowish orange in basal half, melanic apically. Hindlegs mostly melanic except trochantellus me- lanic, or yellowish orange. Forewing in- fuscate with large hyaline spot. Stigma melanic. Hind wing entirely infuscate. Metasoma reddish brown with the last Volume 12, Number 2, 2003 255 Figs. 28-32. Median areola of metanotum. 28, Smooth with lateral carinae not meeting posteriorly in S. diazi. 29, With longitudinal rugosities and lacking lateral carinae in S. grandis. 30-32, 30, Foretibia of S. acrolophus showing spines. 31, Midtibia of S. dichromus showing spines. 32, Simple tarsal claws in S. acrolophus. four segments melanic. Ovipositor yellow- ish orange. 6 . — Essentially as the female. Material examined. — Holotype. Brazil: 9, Nova Teutonia, 27°11'S 52°23'W, 300- 500m, 24.L1939, Fritz Plaumann (CNC). Paratypes. Brazil: 9, Nova Teutonia, 27°11'S 52°23'W, 300-500m, Fritz Plau- mann; 29, same data except 30. i. 1939, 26.U939 (BMNH); 9, same data except vii.1940; 6, same data except 18.V.1954 (AEI); 59, same data except ii.1967, iii.1965, xi.1968, ii.1966; 6, same data ex- cept 21.x. 1940 (CNC); 29, same data ex- 256 Journal of Hymenoptera Research cept 6-10.iii.1967 (UK); 9, Represa Rio ange. Antenna melanic. Mesosoma mostly Grande, Guanabara, xii.1967. M. Alvaren- yellowish orange except pronotum melan- ga (AEI). ic anteriorly; metanotum, propodeum and Distribution. — This species is known metapleuron yellowish orange or melanic; only from Nova Teutonia region of Brazil, and propleuron always melanic. Forelegs Etymology. — This species is named after with coxa, trochanter and trocantellus me- the country of the holotype specimen. lanic, or yellowish orange; femur yellow- ish orange, melanic basally; tibia and tar- Sesioctonus chaconi Briceno, sp. n. sus yenowish orange. Midlegs with coxa Diagnosis. — This species may be recog- yellowish orange ventrally, melanic dor- nized by the combination of a long outer sally; trochanter melanic, trochantellus tooth of the mandible and the absent of yellowish orange, femur melanic in basal the 3RSa vein in the forewing. half, yellowish orange apically, or yellow- Description. — 9. Length. Body, 6-9 mm ish orange, tibia yellowish orange, or me- (6.5). Head: Flagellum with 29-33 flagel- lanic, and tarsus yellowish orange but api- lomeres (29). Interantennal space with cal tarsomere melanic. Hindleg melanic rounded longitudinal keel. Antennal sock- except coxa yellowish orange in basal half, ets moderately excavated, or not excavat- melanic apically; femur melanic, or yel- ed. Face without median longitudinal ca- lowish orange, melanic basally. Forewing rina. Genae not expanded posteroventral- banded from base, yellow, infuscate, yel- ly. Occipital tubercles present. Occiput ex- low, infuscate. Stigma melanic. Hind wing cavated. Mandible concave, outer tooth of banded from base, yellow, infuscate, yel- mandible longer than inner tooth. Maxilla low, infuscate. Metasoma yellowish or- with 4 palpomeres. Third and fourth labial ange except first metasomal tergum yel- palpomeres completely fused. Mesosoma: lowish orange, but median tergite melanic Subpronope triangular. Longitudinal ca- centrally and fifth to eighth metasomal rinae of scutellar depression absent. Scu- terga melanic. Ovipositor yellowish or- tellum convex. Median areola of metano- ange. turn smooth; without median longitudinal 6 . Essentially as female, carina; and with lateral carinae present Material examined. — Holotype. Ecuador: and not meeting posteriorly. Propodeum 9, Napo, Tena, 23.V.1977, DL& SS Vincent flat, median longitudinal carina absent. (CNC). Paratypes. Ecuador: 9, Banos, Epicnemial carina, sharp, complete, bi- 700m, 20.iii.1939, W. Clarke-Maclntyre lobed medially (between forecoxae). Fore- (AEI); 9, Coca, v. 1965, Luis Pena (AEI); 9, tibial spines absent. Midtibia with 8-13 Prov. Santa Clara, 30.vi.1976, P. M. Turner spines (9). Hind tibia with 13-16 spines (CNC). Peru: 6, Yahuarmayo, 8.ii.l910, (15). Hind femur 3.39 times as long as CHTTownsend col (CNC); 9, Loreto, Pu- wide. (RS + M)a vein of forewing com- callpa, 12.vi.1951, J. M. Schuncke (CNC); plete. 3RSa vein of forewing absent. 2-1 A 9, Loreto, Boqueron Abad, 27.xii.1961, J. vein of hind wing tubular. CUb vein of M. Shuncke (CNC); 9, Avispas, 30m nr. hind wing not tubular. Hind wing with 4- Marcapata, 1-15.X.1962, Luis Peha (AEI); 5 hamuli (4). Metasoma: Median tergite of 9, Loreto, Boqueron, 500m, 7-14.vii.1965, first metasomal segment without pair of J. Schuncke (UK). Colombia: 9, Putuma- lateral longitudinal carinae. First metaso- yo, Villa Garzon, 8mi. s. Mocoa, mal median tergite without depression 3.viii.l978, M. Cooper (CNC). postered spiracle. Length width ratio of Distribution.— -This species is distributed first metasomal median tergite 1.19. Ovi- in northwestern of South America, from positor 6-8 mm (6.1). Color: Head melanic Colombia to Peru and Ecuador. vpt maxillary palpomeres yellowish or- Etymology.— This species is named in Volume 12, Number 2, 2003 257 honor of Mr. Anibal Chacon, who during many years has shared with Venezuelan entomologists his passion for entomology and insects collecting in Venezuela. Sesioctonus chrestos Briceno, sp. n. Diagnosis. — S. chrestos is known only from one male specimen, however can be distinguished from all other species by the following combination of characters: max- illa with 5 palpomeres, labium with 4, ep- icnemial carina absent and foretibia with spines. Description. — 6. Length. Body, 5.5 mm. Head: Flagellum with broken after flagel- lomere 12. Interantennal space with rounded longitudinal keel. Antennal sock- ets not excavated. Face without median longitudinal carina. Genae not expanded posteroventrally. Occipital tubercles ab- sent. Occiput not excavated. Maxilla with 5 palpomeres. Third and fourth labial pal- pomeres partly fused. Mesosoma: Sub- pronope oval. Longitudinal carinae of scu- tellar depression absent. Scutellum con- vex. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and not meet- ing posteriorly. Propodeum convex, me- dian longitudinal carina absent. Epicne- mial carina completely absent. Foretibial spines present. Midtibia with 2 spines. Hind tibia with 5 spines. Hind femur 3.06 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of fore- wing present. 2-1 A vein of hind wing tu- bular. CUb vein of hind wing not tubular. Hind wing with 3 hamuli. Metasoma: Me- dian tergite of first metasomal segment without pair of lateral longitudinal cari- nae. First metasomal median tergite with- out depression posterad spiracle. Length width ratio of first metasomal median ter- gite 1.04. Color: Head yellowish orange. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Mesosoma yellowish orange except scutellum, meta- notum, propodeum, mesopleuron and metapleuron melanic. Forelegs yellowish orange. Midlegs mostly melanic except fe- mur yellowish orange apically; and tibia yellowish orange. Hindleg melanic. Fore- wing entirely infuscate. Stigma melanic. Hind wing entirely infuscate. Metasoma with first and second terga yellowish or- ange, but median tergite melanic centrally; rest of metasomal terga melanic. 9 . — Unknown. Material examined. — Holotype. Peru: 6, Marcapata (TMB). Distribution. — This species is known only from Peru. Etymology. — From Greek chrestos that means good, useful. Sesioctonus clavijoi Briceno, sp. n. Fig. 26 Diagnosis. — S. clavijoi shows color pat- terns similar to S. areolatus, however can be separated from the latter by the pres- ence of the bilobed epicnemial carina. S. areolatus has this carina straight and also has a central areola on the propodeum and a longitudinal carina on the scutellar depression, both of which are absent in S. clavijoi. Description. — 9. Length. Body, 4-5.5 mm (5.1). Head: Flagellum with 25 flagello- meres. Interantennal space lacking longi- tudinal keel. Antennal sockets not exca- vated. Face without median longitudinal carina. Genae not expanded posteroven- trally. Occipital tubercles absent. Occiput not excavated. Mandible concave, outer tooth of mandible not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres completely fused. Mesosoma: Longitudinal carinae of scutellar depression absent. Scutellum flat. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and meeting posteriorly. Propodeum convex, median longitudinal carina absent. Epicnemial ca- rina sharp, complete, bilobed medially (between forecoxae). Foretibial spines pre- sent. Midtibia with 5-9 spines (8). Hind tibia with 10-12 spines (12). Hind femur 258 Journal of Hymenoptera Research 3.1 times as long as wide. (RS+M)a vein of forewing incomplete. 3RSa vein of fore- wing absent. 2-1 A vein of hind wing not tubular. CUb vein of hind wing not tu- bular. Hind wing with 3 hamuli. Metaso- ma: Median tergite of first metasomal seg- ment with pair of lateral longitudinal ca- rinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 1.04. Ovipositor 4 mm. Color: Head yellowish orange. Antenna melanic. Maxillary and labial palpomeres yellow- ish orange. Mesosoma yellowish orange. Forelegs yellowish orange. Midlegs yel- lowish orange. Hindleg yellowish orange except tibia mostly yellowish orange, but melanic apically and tarsus melanic, or melanic with basitarsus yellowish orange basally, or melanic with basitarsus yellow- ish orange in basal third. Forewing entire- ly infuscate. Stigma melanic. Hind wing entirely infuscate. Metasoma yellowish or- ange. Ovipositor yellowish orange. 6 . — Essentially as female. Material examined. — Holotype. Costa Rica: 9, Puntarenas, Golfo Dulce, 24km W. Piedras Blancas, 200m, iv.1993, Paul Hanson (UWY). Paratypes. Costa Rica: 39, same data as holotype, except 10-200m, xii.1992, ii.1993, Paul Hanson (UWY); 6, Puntarenas, San Vito, Est. Biol. Las Altu- ras, 1500m, iii.1992, Paul Hanson (UWY). Peru: 9, Quincemil, 750m nr. Marcapata, 10-15.xi.1962, Luis Pena (AEI). Mexico: 9, Vista Hermosa, Oaxaca, 96.5km SW Tux- tepec, 19.X.1962, H. & M. Townes (AEI). Distribution. — Southern Mexico, Costa Rica and Peru. Etymology. — This species is named in honor of Jose Clavijo A., Venezuelan tax- onomist, my professor and my friend, who has shared with me many of my en- tomologist dreams and has been an im- portant part of my life. Thanks for all that you have done for me. Sesioctonus diazi Briceno, sp. n. Figs. 17, 28 —S. diazi is not an easy spe- distinguish because of intraspecific variation, especially in color pattern. It can be confused with specimens of S. chaconi, and separation of these species is compli- cated because the differences are primarily in the relative sizes of the mandible teeth. Description. — 9. Length. Body, 7-8 mm (7.5). Head: Flagellum with 28-34 flagel- lomeres (33). Interantennal space with rounded longitudinal keel. Antennal sock- ets moderately excavated. Face without median longitudinal carina. Genae not ex- panded posteroventrally. Occipital tuber- cles present. Occiput excavated. Mandible concave, outer tooth of mandible not lon- ger than inner tooth. Maxilla with 4 pal- pomeres. Third and fourth labial palpo- meres completely fused. Mesosoma: Sub- pronope triangular or oval-shape. Longi- tudinal carinae of scutellar depression absent. Scutellum convex. Median areola of metanotum smooth; without median longitudinal carina; and with lateral cari- nae present and not meeting posteriorly. Propodeum flat, median longitudinal ca- rina absent. Epicnemial carina, sharp, complete, bilobed medially (between fore- coxae). Foretibial spines absent. Midtibia with 8-11 spines (10). Hind tibia with 15- 23 spines (17). Hind femur 3.01-3.22 (3.08) times as long as wide. (RS + M)a vein of forewing complete or incomplete. 3RSa vein of forewing present. 2-1 A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 4-5 (5) ham- uli. Metasoma: Median tergite of first me- tasomal segment without pair of lateral longitudinal carinae. First metasomal me- dian tergite without depression posterad spiracle. Length width ratio of first meta- somal median tergite 0.84-1.03 (0.84). Ovi- positor 5-8 mm (7.1). Color: Head melan- ic. Antenna melanic. Maxillary and labial palpomeres yellowish orange except basal two palpomeres melanic. Mesosoma most- ly yellowish orange sometimes pronotum and propleuron melanic. Forelegs melanic, or tarsus mostly yellowish orange, but apical tarsomere melanic. Midlegs vari- able, tarsus melanic, or yellowish orange Volume 12, Number 2, 2003 259 Figs. 33-37. First metasomal tergite. 33,35, With lateral longitudinal carinae. 33, S. dichromus. 35, S. acrolophus. 34, S. brasiliensis without lateral longitudinal carinae. 36-37, S. grandis showing depression posterad spiracle. but apical tarsomere melanic. Hindlegs melanic but coxa can be yellowish orange in basal half, melanic apically, or yellow- ish orange with melanic spots laterally, or melanic, or yellowish orange basally, oth- erwise melanic. Forewing yellow basally and infuscate apically, or banded from base, yellow, infuscate, yellow, infuscate. Stigma melanic. Hind wing yellow basally infuscate apically. Metasoma yellowish or- ange. Ovipositor yellowish orange, or yel- lowish orange except apical eighth melan- ic. 6 . — The color pattern in males is essen- 260 Journal of Hymenoptera Research tially as females except fifth to eighth me- tasomal tergite which sometimes can be yellowish orange with central portion of median tergite melanic. Material examined. — Holotype. Costa Rica: 9, Heredia, Est. Biol. La Selva, 50- 150m, 10°26'N 84°01'W, viii.1998 (INBio). Paratypes. Costa Rica: 9, Nara NE Quepos, 16.vii.1975, W.J.Hanson (CNC); 9, Prov. Alajuela, 28.V.1972, J.F.Alvarez (UCR); 6, Alajuela, Upala, Colonia Libertad, 450m, 1-6. v. 1988, Gonzalez & Soto (UCR); 9, Prov. Golfo Dulce, 24km W Pan-Am high- way, 200m, iii-v.1989, Gauld (BMNH); 39, 2c?, same data as holotype except, 02.V.1993, vi.1993, vii.1993, 03.viii.1993, v.1996 (INBio)(UCOB)(UK); 9, Prov. Pun- tarenas, Est. Agujas, sendero Ajo., 300m, 14-24.viii.1996, A. Azofeifa (INBio). Gua- temala: 9, Concepcion, 1400m (UK). Pan- ama: 49, Barro Colorado Is, 9°9'N 79°51'W, v.1939, Jas Zetek (USNM), 5- 12.V.1993, 23-30.iii.1994, 30.iii-6.iv.1994. J. Pickering (UK); 19, 16, Portobello, 24.ii.1911, 13.iii.1911, A. Busk (CNC); 9, San Bias Nusagandi Reserve, 9°20'N 79°0'W, 20-27.xi.1993. J. Pickering (UK). Distribution. — Sesioctonus diazi is distrib- uted from Guatemala to Panama in Cen- tral America. Etymology. — This species is named in honor of Francisco Diaz, Venezuelan en- tomologist, for his contribution to the knowledge of Venezuela ichneumonids. Sesioctonus dichromus Briceno, sp. n. Figs. 12, 16, 18, 20, 31, 33 Diagnosis. — S. dichromus can be distin- guished from other Sesioctonus species by the following combination of characters: occipital tubercles present, occiput exca- vated, median areola of metanotum with lateral carinae present and meeting pos- teriorly, the median tergite of first meta- somal segment with well defined pair of lateral longitudinal carinae. Description.— 9. Length. Body, 6-10 mm (9.5). Head: Flagellum with 30-35 flagel- lomeres (33). Interantennal space with rounded longitudinal keel. Antennal sock- ets moderately excavated. Face without median longitudinal carina. Genae not ex- panded posteroventrally. Occipital tuber- cles present. Occiput excavated. Mandible concave, outer tooth of mandible longer than inner tooth. Maxilla with 4 palpo- meres. Third and fourth labial palpomeres completely fused. Mesosoma: Subpronope triangular or oval. Longitudinal carinae of scutellar depression absent. Scutellum convex. Median areola of metanotum smooth; without median longitudinal ca- rina; and with lateral carinae present and meeting posteriorly, or lacking lateral ca- rinae. Propodeum flat, median longitudi- nal carina present or absent. Epicnemial carina sharp, complete, bilobed medially (between forecoxae). Foretibia spines ab- sent. Midtibia with 7-13 spines (12). Hind tibia with 18-25 spines (19). Hind femur 3.28-3.6 times as long as wide (3.54). (RS+M)a vein of forewing complete or in- complete. 3RSa vein of forewing present or absent. 2-1 A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 5-7 hamuli (7). Metasoma: Me- dian tergite of first metasomal segment with pair of lateral longitudinal carinae. Median tergite of first metasomal segment without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 0.97. Ovipositor 6-9 mm (9.0). Color: Head melanic including maxillary and labial palpomeres. Antenna melanic. Mesosoma melanic. Forelegs melanic. Midlegs melanic. Hindleg melanic. Fore- wing entirely infuscate, or hyaline basally, infuscate apically. Stigma melanic. Hind wing entirely infuscate, or infuscate with large hyaline spots, or hyaline basally, in- fuscate apically. Metasoma yellowish or- ange. Ovipositor yellowish orange. 6 . — Essentially as the female. Material examined. — Holotype. Costa Rica: 9, Alajuela, San Ramon/800m, 29.i- 03.ii.1995, G. Carballo (INBio). Paratypes. Costa Rica: 29 9, Heredia, Est. Biol. La Sel- va, 50-150m, 10°26'N 84°01'W, 11- Volume 12, Number 2, 2003 261 17. vi. 1986, 2. v. 1993, vi.1993, viii.1993, ix.1993, 01. ix. 1993, 16. ix. 1993, xii.1993, 22.iii.1994, 29.ix.1995, ix.1995, 15.xii.1995, 14.xii.1995, ii.1996, 01.iii.1996, iii.1996, vi.1996, 2. v. 1996, 15. v. 1996, 02.x. 1997, xi.1997, ix.1998, 22.L1999 (INBio) (UK) (UCR) (UWY) (UCOB); 6, Heredia, Est. El Ceibo, Braulio Carrillo, N.P. 400-600m, iii.1990, C. Chavez (INBio); 29, Est. Pitilla, 9km Sur Santa Cecilia, Guanacaste, 700m, xi.1988, C. Chavez & M. Espinoza, ii.1990, P. Rios, C. Moraga & R. Blanco (INBio); 4 9, El Limon, Sector Cerro Cocori, Finca de E. Rojas, 150m, v.1991, . 5.vii-12.viii. 1992, ll.i-12.ii.1993, Trampa malaise (IN- Bio); 9, El Limon, 16km W Guapiles, Parque Nacional Braulio Carrillo, 400m, iv-v.1989, Gauld (BMHN); 9, Limon, P N Tortuguero, Est. 4-esquinas, 0m, vi- viii.1989, Solano col (UWY); 9, Puntaren- as, Rancho Quemado, Peninsula de Osa, 200m, iv.1992, L.Brenes (INBio); 9, Ala- juela, Penas Blancas, 700m, viii.1987, E. Cruz (CNC); 9, Alajuela, Sector Colonia Palmareha, 9km SO de Bajo Rodriguez, 700m, ix.1996, G. Carballo (INBio); 9, San Jose, P.N. Braulio Carrillo, 9.5km E tunel, 1000m, viii-ix.1989 (UWY); 6\ Alajuela, Sector San Ramon, 800m, ll-15.iv.1994, M. Zumbado (INBio); 9, Est. Biol. La Sel- va, 30.vi.1995 (UK). Mexico: 9, Chiapas, Muste, 440m near Huixtla. 1970. Mai. trap. Welling (CNC). Distribution. — This species is known from Mexico to Costa Rica. Etymology. — From Greek di that means two and chromos that means color, in ref- erence to the color pattern of the species, half melanic, half yellowish orange. Sesioctomis dominions Briceno, sp. n. Diagnosis. — S. dominicus is known only from males specimens. However, it can be separated from all other Sesioctomis spe- cies by its small size (2.0-3. 5mm) and to- tally hyaline wings. Description. — 6 . Length. Body, 2-3.5 mm (2.5). Head: Flagellum with 23-27 flagel- lomeres (25). Interantennal space lacking longitudinal keel. Antennal sockets not ex- cavated. Face without median longitudi- nal carina. Genae not expanded poster- oventrally. Occipital tubercles absent. Oc- ciput not excavated. Mandible flat, outer tooth of mandible not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres completely fused. Mesosoma: Subpronope elongate- oval. Longitudinal carinae of scutellar de- pression absent. Scutellum convex. Medi- an areola of metanotum smooth; without median longitudinal carina; and with lat- eral carinae present and meeting posteri- orly. Propodeum convex, median longi- tudinal carina absent. Epicnemial carina (difficult to see), sharp, incomplete dorsal- ly, straight medially (between forecoxae). Foretibial spines absent. Midtibia with 7- 10 spines (8). Hind tibia with 13-16 spines (15). Hind femur 1.14 times as long as wide. (RS + M)a vein of forewing incom- plete. 3RSa vein of forewing absent. 2-1A vein of hind wing not tubular. CUb vein of hind wing not tubular. Hind wing with 3 hamuli. Metasoma: Median tergite of first metasomal segment without pair of lateral longitudinal carinae. First metaso- mal median tergite without depression posterad spiracle. Length width ratio of first metasomal median tergite 1.10. Color: Head black except face yellowish orange. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Mesosoma melanic except propodeum, propleuron and metapleuron yellowish orange and pronotum yellowish orange, or melanic. Forelegs yellowish orange. Midlegs yel- lowish orange except tibia melanic, or yel- lowish orange. Hindleg yellowish orange except tibia melanic and tarsus melanic. Forewing hyaline. Stigma melanic. Hind wing hyaline. Metasoma with first and second terga yellowish orange, the rest melanic. 9 . — Unknown. Material examined. — Holotype. Dominica W.I.: 6, Springfield, xi.1967, N.L.H Krauss (UK). Paratypes. Dominica W.I: 2<5, Clarke 262 Journal of Hymenoptera Research Hall, Layou Vail, 10-17.ii.1965, 20- 28.ii.1965, H.E.Evans (UK) (USNM); 6\ Mth. Layou R, 13.iii.1965, H.E.Evans (USNM); 6, Hillsborough, 15.iii.1965 (USNM). Distribution. — This species is known only from the Dominica W. I. in the Ca- ribbean. Etymology. — This species is named after the locality of the holotype specimen. Sesioctomis enmenetes Briceno, sp. n. Diagnosis. — S. enmenetes is known only from one specimen, however, can be dis- tinguised from all other Sesioctomis species by the following combination of charac- ters: occipital tubercles absent, presence of median longitudinal carina on the propo- deum, epicnemial carina complete, medi- an areola of metanotum smooth and a pair of longitudinal carinae on the first meta- somal tergite. Description. — 9. Length. Body, 6 mm. Head: Flagellum broken after flagellomere 11. Interantennal space with rounded lon- gitudinal keel. Antennal sockets not exca- vated. Face without median longitudinal carina. Genae not expanded posteroven- trally. Occipital tubercles absent. Occiput excavated. Mandible concave, outer tooth of mandible longer than inner tooth. Max- illa with 4 palpomeres. Third and fourth labial palpomeres completely fused. Me- sosoma: Subpronope triangular. Longitu- dinal carinae of scutellar depression ab- sent. Scutellum convex. Median areola of metanotum smooth; without median lon- gitudinal carina; and with lateral carinae present and not meeting posteriorly. Pro- podeum flat, median longitudinal carina absent. Epicnemial carina sharp, complete, bilobed medially (between forecoxae). Foretibia spines absent. Midtibia with 10 spines. Hind tibia with 15 spines. Hind fe- mur 3.17 times as long as wide. (RS+M)a vein of forewing incomplete. 3RSa vein of forewing present. 2-1A vein of hind wing eir. CUb vein of hind wing not fu- nd wing with 5 hamuli. Metaso- ma: Median tergite of first metasomal seg- ment with pair of lateral longitudinal ca- rinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 1.09. Ovipositor 6.5 mm. Col- or: Head melanic. Antenna melanic. Max- illary and labial palpomeres melanic. Me- sosoma yellowish orange except prono- tum mostly yellowish orange, melanic anteriorly and propleuron melanic. Fore- legs melanic except coxa mostly yellowish orange and tarsus mostly yellowish or- ange, but apical tarsomere melanic. Mid- legs melanic except coxa yellowish or- ange. Hindleg melanic except coxa yel- lowish orange and femur melanic in basal third, otherwise yellowish orange. Fore- wing entirely infuscate. Stigma melanic. Hind wing entirely infuscate. Metasoma yellowish orange. Ovipositor yellowish orange except apical eighth melanic. (5 . — Unknown. Material examined. — Holotype. Costa Rica: 9, San Vito Las Cruces, 1200m, 9.vii- 7.viii.l982, B. Gill (CNC). Distribution. — This species is known only from the Las Cruces region of Costa Rica. Etymology. — From Greek enmenetes that means friend, in honor of my colleagues and friends Ana, Carmen Liceth, Carlos, Dorys and Maria del Carmen. Sesioctomis galeos Briceno, sp. n. Diagnosis. — S. galeos shows the follow- ing combination of characters: occipital tu- bercles absent, occiput not excavated, maxillary palpi with 4 palpomeres and la- bial palpi with 3, median areola of meta- notum with longitudinal rugosities, epic- nemial carina incomplete laterally and straight medially. Description. — 9. Length. Body, 8.5 mm. Head: Flagellum with 33 flagellomeres. In- terantennal space with rounded longitu- dinal keel, or lacking of longitudinal keel. Antennal sockets deeply excavated, or moderately excavated. Face without me- Volume 12, Number 2, 2003 263 dian longitudinal carina. Genae moderate- ly expanded posteroventrally. Occipital tubercles absent. Occiput not excavated. Mandible concave, outer tooth of mandi- ble not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres completely fused. Mesosoma: Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum with longitudinal rugosities; without median longitudinal carina; and with lat- eral carinae present and not meeting pos- teriorly. Propodeum flat, median longitu- dinal carina of propodeum absent. Epic- nemial carina sharp, incomplete laterally, straight medially (between forecoxae). Foretibia spines absent. Midtibia with 2-7 (2) spines. Hind tibia with 8-12 (12) spines. Hind femur 3.6 times as long as wide. (RS + M)a vein of forewing com- plete. 3RSa vein of forewing present or ab- sent. 2-1 A vein of hind wing tubular or not tubular. CUb vein of hind wing not tubular. Hind wing with 4-5 (5) hamuli. Metasonta: Median tergite of first meta- somal segment without pair of lateral lon- gitudinal carinae. First metasomal median tergite without depression posterad spi- racle. Length width ratio of first metaso- mal median tergite 0.9. Ovipositor 7.2mm. Color: Head melanic. Antenna melanic. Maxillary and labial palpomeres yellow- ish orange. Mesosoma with pronotum yel- lowish orange, sometimes melanic anteri- orly; mesoscutum, scutellum and meta- notum yellowish orange; propodeum mostly yellowish orange with melanic spots; propleuron yellowish orange or me- lanic; mesopleuron mostly yellowish or- ange, melanic basally and metapleuron melanic, or yellowish orange. Forelegs yel- lowish orange. Midlegs yellowish orange. Hindleg melanic except trochanter and trochantellus yellowish orange. Forewing banded from base, yellow, infuscate, yel- low, infuscate. Stigma melanic. Hind wing banded from base, yellow, infuscate, yel- low, infuscate. Metasoma mostly yellow- ish orange but fourth tergum with median tergum melanic and fifth to eighth terga melanic. 8 . — Essentially as female. Material examined. — Holotype. Brasil: 9, Manaus, Reserva Dulce, 31.viii.1990, Trampa malaise, Vidal col. (INPA). Para- types. Peru: 6, Puerto Bermudez, 12- 19.vii.1920, Cornell Univ. Expedition (CUIC); 9, Avispas, 30km nr Marcapata, ix.1992, Luis Pena (AEI). Distribution. — Sesioctonus galeos is known only from Brazil and Peru. Etymology. — From Greek galeos that means shark in honor of Michael Sharkey, in acknowledgment for his advising in my formation as braconid taxonomist and for his contribution to the knowledge of the Agathidinae of the World. Sesioctonus garciai Briceno, sp. n. Diagnosis. — S. garciai is known only from one specimen and can be distin- guished from all other species by the fol- lowing combination of characters: occipi- tal tubercles present, occiput excavated, epicnemial carina complete and straight medially, although the presence of spines on the fore tibia. Description. — 9. Length. Body, excluding ovipositor, 6 mm. Head: Flagellum with 30 flagellomeres. Interantennal space with rounded longitudinal keel. Antennal sock- ets not excavated. Face without median longitudinal carina. Genae not expanded posteroventrally. Occipital tubercles pre- sent. Occiput excavated. Mandible con- cave, outer tooth of mandible longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres com- pletely fused. Mesosoma: Subpronope tri- angular. Longitudinal carinae of scutellar depression absent. Scutellum convex. Me- dian areola of metanotum smooth; with- out median longitudinal carina; and with lateral carinae present and not meeting posteriorly. Propodeum flat, median lon- gitudinal carina of propodeum absent. Ep- icnemial carina sharp, complete, straight 264 Journal of Hymenoptera Research medially (between forecoxae). Foretibia spines present. Midtibia with 9 spines. Hind tibia with 15 spines. Hind femur 3.64 times as long as wide. (RS + M)a vein of forewing incomplete. 3RSa vein of fore- wing absent. 2-1 A vein of hind wing tu- bular. CUb vein of hind wing not tubular. Hind wing with 4 hamuli. Metasoma: Me- dian tergite of first metasomal segment without pair of lateral longitudinal cari- nae. First metasomal median tergite with- out depression posterad spiracle. Length width ratio of first metasomal median ter- gite 0.9. Ovipositor 5 mm. Color: Head melanic. Antenna melanic. Maxillary and labial palpomeres yellowish orange except two basal palpomeres melanic. Mesosoma yellowish orange. Forelegs yellowish or- ange. Midlegs yellowish orange except tibia and tarsus melanic. Hindleg melanic with coxa yellowish orange femur mostly yellowish orange, but melanic distally. Forewing entirely infuscate. Stigma me- lanic. Hind wing entirely infuscate. Meta- soma tergum yellowish orange except fifth to eighth metasomal terga mostly yellow- ish orange but median tergites melanic posteriorly. Ovipositor yellowish orange except apical eighth melanic. 6 . — Unknown. Materia] examined. — Holotype. Brasil: 9, Nova Teutonia, 27°11'N 52°23'L, 2.ii.l939, Fritz Plaumann (CNC). Distribution. — This species is known only from the Nova Teutonia region of Brazil. Etymology. — This species is named in honor of Jose Luis Garcia, Venezuelan en- tomologist, for his contribution to the knowledge of Venezuela proctotrupoids. Sesioctonus grandis Briceno, sp. n. Figs. 19, 29, 36, 37 Diagnosis. — S. grandis is the larger spe- cies of Genus Sesioctonus and it has a high- er number of flagellomeres (48), maxilla and labium with four palpomeres, and the first metasomal median tergite with de- pression posterad spiracle. This latter character is shared with S. qui from which it is separated by the presence of an oval- shape subpronope, epicnemial carina ab- sent and the presence of a pair of lateral longitudinal carinae on the first metaso- mal median tergite. Description. — 9. Length. Body, excluding ovipositor, 10-13 mm (10.0). Head: Flagel- lum with 44-48 flagellomeres (46). Inter- antennal space with sharp longitudinal keel. Antennal sockets deeply excavated. Face without median longitudinal carina. Genae strongly expanded posteroventral- ly. Occipital tubercles absent. Occiput not excavated. Mandible flat, outer tooth of mandible not longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres partly fused. Me- sosoma: Subpronope oval. Longitudinal carinae of scutellar depression absent. Scutellum convex. Median areola of me- tanotum with longitudinal rugosities; without median longitudinal carina; and lacking lateral carinae. Propodeum con- vex, median longitudinal carina of pro- podeum absent. Epicnemial carina com- pletely absent. Foretibia spines absent. Midtibia with 5-8 spines (8). Hind tibia with 14-21 spines (21). Hind femur 4.4 times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing absent. 2-1 A vein of hind wing not tubu- lar. CUb vein of hind wing not tubular. Hind wing with 8-10 hamuli (9). Meta- soma: Median tergite of first metasomal segment with pair of lateral longitudinal carinae. First metasomal median tergite with depression posterad spiracle. Length width ratio of first metasomal median ter- gite 0.71. Ovipositor 10 mm. Color: Head melanic. Antenna melanic, sometimes yel- lowish orange. Maxillary and labial pal- pomeres yellowish orange except two bas- al palpomeres melanic. Mesosoma melan- ic. Forelegs yellowish orange except coxa melanic, trochantellus, tibia and tarsus yellowish orange, or melanic. Midlegs yel- lowish orange except coxae and tarsus melanic. Hindleg melanic. Forewing Volume 12, Number 2, 2003 265 banded from base, yellow, infuscate, yel- low, infuscate. Stigma melanic, or melanic and yellowish orange. Hind wing yellow basally infuscate apically. Metasoma me- lanic. Ovipositor yellowish orange. 6 . — Essentially as the female. Material examinated. — Holotype. Brazil: 9, Rio Grande do Sul, Staudinger K col. (ZSBS). Paratypes. Brazil: 49, 56, Rio Grande do Sul, Staudinger K col. (ZSBS) (UCOB). Distribution. — This species is known only from the Rio Grande do Sul region of Brazil. Etymology. — grandis refers the big size of these specimens. Sesioctonus kompsos Briceno, sp. n. Diagnosis. — S. kompsos can be distin- guished for the following combination of characters: occipital tubercles present, oc- ciput excavated, median areola of meta- notum with longitudinal rugosities. Description. — 9. Length. Body, excluding ovipositor, 8.0-10. 0mm (10). Head: Flagel- lum with 33-34 (34) flagellomeres. Inter- antennal space with rounded longitudinal keel. Antennal sockets not excavated. Face without median longitudinal carina. Ge- rtae not expanded posteroventrally. Occip- ital tubercles present. Occiput excavated. Mandible concave, outer tooth of mandi- ble longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial pal- pomeres completely fused. Mesosoma: Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum with longitudinal rugosities; without median longitudinal carina; and with lat- eral carinae present and not meeting pos- teriorly. Propodeum flat, median longitu- dinal carina absent. Epicnemial carina sharp, complete, bilobed medially (be- tween forecoxae). Foretibia spines absent. Midtibia with 9-13 (13) spines. Hind tibia with 21-25 (25) spines. Hind femur 3.2- 3.43 (3.2) times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing absent. 2-1 A vein of hind wing tubular. CUb vein of hind wing not tu- bular. Hind wing with 6 hamuli. Metaso- ma: Median tergite of first metasomal seg- ment without pair of lateral longitudinal carinae. First metasomal median tergite without depression posterad spiracle. Length width ratio of first metasomal me- dian tergite 0.94-1.20 (0.94). Ovipositor 8 mm. Color: Head melanic. Antenna me- lanic. Maxillary palpomeres yellowish or- ange except two basal palpomeres melan- ic. Labial palpomeres melanic. Mesosoma mostly melanic except metanotum, pro- podeum and metapleuron yellowish or- ange. Forelegs melanic. Midlegs melanic. Hindleg melanic. Forewing entirely infus- cate. Stigma melanic. Hind wing entirely infuscate. Metasoma yellowish orange. Ovipositor yellowish orange except apical eighth melanic. 6 . — Unknown. Material examined. — Holotype. Costa Rica: 9, Rancho Quemado, 2km N. cami- no Drake, 275m, i.1991, P. Hanson (UCR). Paratype. Brasil: 9, Guanabara, Represa Rio Grande, viii.1966, M.Alvarenga (AEI). Distribution. — This species is known only from Costa Rica and Brazil. Etymology. — From Greek kompsos that means elegance in reference to the beauty of this species. Sesioctonus miyayensis Briceno, sp. n. Diagnosis. — S. miyayensis shows the fol- lowing combination of characters: occipi- tal tubercles absent, subpronope elongate- oval-shaped, median areola of metanotum with lateral carinae meeting posteriorly, epicnemial carinae complete and straight medially, foretibia with spines, first me- tasomal tergite with pair of lateral longi- tudinal carinae. Description.— 9 . Length. Body, excluding ovipositor, 5-7 mm (6.5). Head: Flagellum with 32 flagellomeres. Interantennal space with rounded longitudinal keel. Antennal sockets not excavated. Face without me- dian longitudinal carina. Genae moderate- Journal of Hymenoptera Research ly expanded posteroventrally. Occipital (AEI). Paratopes. Costa Rica: 89, Heredia, tubercles absent. Occiput not excavated. Est. Biol. La Selva, 10°26'N 84°01'W, 50- Mandible flat, outer tooth of mandible not 150m, xii.1992, P. Hanson (UWY), longer than inner tooth. Maxilla with 4 l.xi.1993, J. Longino (UK), 01.viii.1995, palpomeres. Third and fourth labial pal- 15.1.1996, iii.1996, ii.1996, 31. v. 1996, pomeres completely fused. Mesosoma: 19.ii.1998 (UK); 9, Prov. Guanacaste, Est. Subpronope elongate-oval. Longitudinal Pitilla 9km S. Sta. Cecilia, 700m, 4- carinae of scutellar depression absent. 14.xi.1991, D.Garcia (INBio); 9, Prov. Pun- Scutellum convex. Median areola of me- tarenas, Rancho Quemado, Peninsula de tanotum smooth; without median longi- Osa, 200m, vi.1992. F.Quesada y M.Segura tudinal carina; and with lateral carinae (INBio). present and meeting posteriorly. Propo- Distribution.— This species is known deum flat, median longitudinal carina ab- only from Costa Rica, sent. Epicnemial carina sharp, complete, Etymology.— This species is named in straight medially (between forecoxae). honor of my father's birthplace, Miyayi. Foretibia spines present. Midtibia with 5- 7 spines (7). Hind tibia with 14 spines. Sesioctonus parathyridis Viereck Hind femur 3.42 times as long as wide. ^^ tln/ndis Viereck 1912: h 9> »Par. (RS + M)a vein of forewing incomplete. ^ Canal de panama„ (Washington/ 9# 3RSa vein of forewing absent. 2-1A vein i4552)._Busk 1912: 10, fig. (host a). Host a: of hind wing tubular. CUb vein of hind Parathyridis perspicilla Stoll. wing not tubular. Hind wing with 3-4 hamuli (3). Metasoma: Median tergite of Diagnosis. — S. parathyridis shares char- first metasomal segment with pair of lat- acters with S. grandis and S. qui. They con- eral longitudinal carinae. First metasomal form the group of species with the larger median tergite without depression poster- size and the number of flagellomeres more ad spiracle. Length width ratio of first me- than 40. Also, they have the interantennal tasomal median tergite 1.1. Ovipositor 5 space with a longitudinal sharp keel, mm. Color: Head yellowish orange. An- which is present in S. acrolophus. However, tenna melanic. Maxillary and labial pal- S. parathyridis can be separated from these pomeres melanic, or yellowish orange ex- species by the presence of a pair of lateral cept two basal palpomeres melanic. Me- longitudinal carinae on the median tergite sosoma yellowish orange. Forelegs yel- of first metasomal segment. This character lowish orange except femur and tibia is shared with S. grandis. However, this melanic. Midlegs yellowish orange except latter species has the first metasomal seg- tibia yellowish orange basally, otherwise ment with a depression posterad spiracle, melanic and tarsus melanic with basitar- which are absent in parathyridis. sus yellowish orange in basal half. Hind- Description. — 9. Length. Body, excluding leg yellowish orange except tibia yellow- ovipositor, 9-12 mm (12.0). Head: Flagel- ish orange in basal half, melanic apically lum with 44-45 flagellomeres (42). Inter- and tarsus melanic. Forewing yellow ba- antennal space with sharp longitudinal sally and infuscate apically. Stigma melan- keel. Antennal sockets deeply excavated, ic, or yellowish orange. Hind wing yellow Face without median longitudinal carina, basally infuscate apically. Metasoma yel- Genae strongly expanded posteroventral- lowish orange. Ovipositor yellowish or- ly. Occipital tubercles absent. Occiput not ange. excavated. Mandible flat, outer tooth not (J— Unknown. longer than inner tooth. Maxilla with 4 Material examined.— Holotype. Costa palpomeres. Third and fourth labial pal- Rica: 9, Pto. Viejo, 50m, ii. 1980, W. Mason pomeres completely fused. Mesosoma: Volume 12, Number 2, 2003 267 Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum with longitudinal rugosities; without median longitudinal carina; and lacking lateral carinae. Propodeum flat, median longitudinal carina absent. Epicnemial ca- rina blunt, incomplete laterally, bilobed medially (between forecoxae). Foretibia spines absent. Midtibia with 5-8 spines (6). Hind tibia with 14-16 spines (14). Hind femur 4.88 times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing present. 2-1A vein of hind wing not tubular. CUb vein of hind wing not tubular. Hind wing with 7-8 hamuli (7). Metasoma: Median tergite of first metasomal segment with pair of lat- eral longitudinal carinae, or without pair of lateral longitudinal carinae. First meta- somal median tergite without depression posterad spiracle. Length width ratio of first metasomal median tergite 0.8. Ovi- positor 10-12 mm (12.0). Color: Head me- lanic. Antenna melanic. Maxillary and la- bial palpomeres yellowish orange except two basal palpomeres melanic. Pronotum mostly melanic with yellowish orange ar- eas. Mesoscutum yellowish orange, or mostly melanic, yellowish orange dorsal- ly. Scutellum yellowish orange. Metano- tum yellowish orange. Propodeum yel- lowish orange, or mostly yellowish orange with melanic spots. Propleuron mostly melanic with yellowish orange areas, or yellowish orange. Mesopleuron yellowish orange. Metapleuron yellowish orange. Forelegs mostly yellowish orange except femur melanic; tibia yellowish orange, or melanic but yellowish orange distally and tarsus mostly yellowish orange, but apical tarsomere melanic. Midlegs mostly yel- lowish orange with variations on coxa sometimes melanic apically; femur yel- lowish orange in basal half, melanic api- cally; tibia yellowish orange in basal half, melanic apically, or yellowish orange and tarsus melanic. Hindleg melanic except coxa yellowish orange in basal half, me- lanic apically. Forewing banded from base, yellow, infuscate, yellow, infuscate. Stigma melanic. Hind wing yellow basally infuscate apically. Metasoma yellowish or- ange except the last four segments yellow- ish orange or melanic. Ovipositor yellow- ish orange except apical eighth melanic. 6 . — Essentially as the female. Material examined. — Holotype. Panama: 9, Paraiso, Canal de Panama, host: Para- thyris perspicilla Stoll (USNM). Another specimens reviewed. Panama: 9 , Barro Col- orado Is., 9°9'N 79°51'W, 2-9.X.1996, J. Pickering (UK); Costa Rica: 29, 6, Prov. Limon, Sector Cerro Cocori, Finca de E. Rojas, 150m, 26.vi-16.vii. 1992, 12- 31.viii.1992, ii.1993, E. Rojas (INBio); 9, Prov. Puntarenas, Est. Sirena, P.N. Corco- vado, 0-1 00m, ii.1992, G. Rodriguez (IN- Bio); 9, Prov. Puntarenas, Vuelta Cam- pana, R. Terraba, 100-150m, 10- 31.viii.1992, S. Rojas (INBio); Peru: Loreto, Pucallpa, 24. vi. 1963, J. M. Schunke (BMHN). Distribution. — This species is distributed from Costa Rica and Panama in Central America to Peru in South America. Biology. — Larvae of Sesioctonus parathyr- idis were reported as parasitoids in larvae of the arctiid Parathyris perspicilla (Viereck 1914). Sesioctonus peruviensis Briceno, sp. n. Diagnosis. — S. peruviensis is known only from one specimen, however, can be dis- tinguished from all other species by the following combination of characters: max- illa with four palpomeres and labium with three, occipital tubercles present, occiput excavated, foretibia with spines. Speci- mens of peruviensis could be confused with S. garciai specimens, however they are separated for the presence of epicne- mial carinae bilobed in S. peruviensis, which is straight in S. garciai. Description. — 9. Length. Body, excluding ovipositor, 5.5 mm. Head: Flagellum with 31 flagellomeres. Interantennal space with rounded longitudinal keel. Antennal sock- 268 Journal of Hymenoptera Research ets moderately excavated. Face without median longitudinal carina. Genae mod- erately expanded posteroventrally. Occip- ital tubercles present. Occiput excavated. Mandible concave, outer tooth longer than inner tooth. Maxilla with 4 palpomeres. Third and fourth labial palpomeres com- pletely fused. Mesosoma: Subpronope tri- angular. Longitudinal carinae of scutellar depression absent. Scutellum convex. Me- dian areola of metanotum smooth; with- out median longitudinal carina; and with lateral carinae present and not meeting posteriorly. Propodeum convex, median longitudinal carina absent. Epicnemial ca- rina sharp, complete, bilobed medially (between forecoxae). Foretibia spines pre- sent. Midtibia with 10 spines. Hind tibia with 18 spines. Hind femur 3.27 times as long as wide. (RS + M)a vein of forewing complete. 3RSa vein of forewing present. 2-1 A vein of hind wing not tubular. CUb vein of hind wing not tubular. Hind wing with 4 hamuli. Metasotna: Median tergite of first metasomal segment without pair of lateral longitudinal carinae. First metaso- mal median tergite without depression posterad spiracle. Ovipositor 5 mm. Color: Head melanic. Antenna melanic. Maxil- lary palpomeres yellowish orange except basal two palpomeres melanic. Labial pal- pomeres melanic. Mesosoma yellowish or- ange. Forelegs yellowish orange except trochantellus melanic, femur yellowish or- ange, melanic basally and tarsus mostly yellowish orange, but apical tarsomere melanic. Midlegs yellowish orange except trochanter, tibia and tarsus melanic Hind- leg mostly melanic except coxa yellowish orange and femur yellowish orange, me- lanic basally. Forewing infuscate with hy- aline spots. Stigma melanic. Hind wing entirely infuscate. Metasoma yellowish or- ange. Ovipositor yellowish orange except apical eighth melanic. 6 . — Unknown. Material examined. — Holotype. Peru: 9, Quincemil, 750m near Marcapata, 20- 30.X.1962, Luis Pena (AEI). Distribution. — Known only from Marca- pata, region of Peru. Etymology. — This species is named after the country of the holotype specimen. Sesioctonus qui Briceno, sp. n. Diagnosis. — S. qui is distinguised from all other species for the following combi- nation of characters: occipital tubercles ab- sent, interantennal space with a sharp lon- gitudinal keel, gena strongly expanded posteroventrally, median areola of meta- notum with longitudinal rugosities and first metasomal median tergite with de- pression posterad spiracle. Description. — 9. Length. Body, excluding ovipositor, 10-15 mm (10.0). Head: Flagel- lum with 45 flagellomeres. Interantennal space with sharp longitudinal keel. Anten- nal sockets deeply excavated. Face with- out median longitudinal carina. Genae strongly expanded posteroventrally. Oc- cipital tubercles absent. Occiput not exca- vated. Mandible flat, outer tooth not lon- ger than inner tooth. Maxilla with 4 pal- pomeres. Third and fourth labial palpo- meres completely fused. Mesosoma: Subpronope triangular. Longitudinal ca- rinae of scutellar depression absent. Scu- tellum convex. Median areola of metano- tum with longitudinal rugosities; without median longitudinal carina; and lacking lateral carinae. Propodeum convex, medi- an longitudinal carina absent. Epicnemial carina, blunt, incomplete laterally, or com- pletely absent, bilobed medially (between forecoxae). Foretibial spines absent. Mid- tibia with 5-8 spines (5). Hind tibia with 14-16 spines (14). Hind femur 4.27 times as long as wide. (RS + M)a vein of fore- wing complete. 3RSa vein of forewing ab- sent. 2-1A vein of hind wing tubular. CUb vein of hind wing not tubular. Hind wing with 8 hamuli. Metasoma: Median tergite of first metasomal segment without pair of lateral longitudinal carinae. First metaso- mal median tergite with depression pos- terad spiracle. Length width ratio of first metasomal median tergite 0.84. Ovipositor Volume 12, Number 2, 2003 269 10-12 mm (10.0). Color: Head melanic. Antenna melanic. Maxillary and labial palpomeres yellowish orange except basal two palpomeres melanic. Mesosoma yel- lowish orange. Forelegs yellowish orange. Midlegs yellowish orange. Hind leg me- lanic except coxa yellowish orange but melanic laterally, hind femur yellowish orange, hind tibia yellowish orange in bas- al half, melanic apically. Forewing banded from base, yellow, infuscate, yellow, in- fuscate. Stigma yellowish orange. Hind wing yellow basally infuscate apically. Metasoma yellowish orange with the last four tergites melanic. Ovipositor yellow- ish orange. 6. — Male exhibits a color darker than females, showing the body brownish. However, only one male of this specie was examined for this revision and another fu- ture observations are necessary. Material examined. — Holotype: Venezuela: 9, Aragua, El Limon, 450m, 25.vi.1978, luz de mercurio, Francisco Fernandez Yepez col (MIZA). Paratypes. Brasil: 2$, Sao Pau- lo, Teodoro Sampaio, xii.1977, F. M. Oliv- eira (CNC) (BMNH); Peru: 9, 6, Loreto, Pucallpa, 19.iv.1962, vi.1965, J. M. Schunke (BMNH). Distribution. — This species is distributed from Venezuela until Brazil and Peru in South America. Its presence in Colombia, Ecuador and Bolivia is probable. Etymology. — The species name qui is an arbitrary combination of letters. Sesioctonns theskelos Briceno, sp. n. Diagnosis. — S. theskelos can be distin- guished for the following characters com- bination: lacking of occipital tubercles, maxilla with 5 palpomeres and labial with 4, median longitudinal carina of propo- deum present, and median tergite of first metasomal segment with pair of lateral longitudinal carinae. Description. — 9. Length. Body, excluding ovipositor, 7 mm. Head: Flagellum with 35 flagellomeres. Interantennal space with rounded longitudinal keel. Antennal sock- ets moderately excavated. Face without median longitudinal carina. Genae mod- erately expanded posteroventrally. Occip- ital tubercles absent. Occiput not excavat- ed. Mandible concave, outer tooth of man- dible not longer than inner tooth. Maxilla with 5 palpomeres. Third and fourth labial palpomeres not fused. Mesosoma: Sub- pronope triangular. Longitudinal carinae of scutellar depression absent. Scutellum convex. Median areola of metanotum smooth; without median longitudinal ca- rina; and with lateral carinae present and not meeting posteriorly. Propodeum con- vex, median longitudinal carina present. Epicnemial carina sharp, complete later- ally, bilobed medially (between forecox- ae). Foretibia spines absent. Midtibia with 6-9 spines (9). Hind tibia with 12-15 spines (12). Hind femur 3.5 times as long as wide. (RS+M)a vein of forewing com- plete. 3RSa vein of forewing present or ab- sent. 2-1A vein of hind wing tubular or not tubular. CUb vein of hind wing not tubular. Hind wing with 3-A (4) hamuli. Metasoma: Median tergite of first meta- somal segment with pair of lateral longi- tudinal carinae. First metasomal median tergite without depression posterad spi- racle. Length width ratio of first metaso- mal median tergite 0.94-1.12 (0.94). Ovi- positor 5-6 mm (5.5). Color: Head melanic sometimes with a spot yellowish orange on the front. Antenna melanic. Maxillary and labial palpomeres yellowish orange. Mesosoma yellowish orange. Forelegs yel- lowish orange. Midlegs yellowish orange except tarsus melanic. Hindleg melanic except coxa yellowish orange, or yellow- ish orange but melanic laterally; and fe- mur yellowish orange, sometimes melanic distally. Forewing entirely infuscate. Stig- ma melanic. Hind wing entirely infuscate. Metasoma entirely yellowish orange or yellowish orange with the last four ter- gites melanic posteriorly. Ovipositor yel- lowish orange. 6 . — Unknown. Material examined. — Holotype. Brasil: 9, 270 Journal of Hymenoptera Research Matogrosso, Sinop, x.1975, Trampa mal- aise, M. Alvarenga (CNC). Paratypes. Brasil: 9, Matogrosso, Sinop, x.1975, Trampa malaise, M. Alvarenga (CNC); Ec- uador: 9, Coca, v.1992, Luis Pena (AEI). Distribution. — This species is known only from Brazil and Ecuador. Etymology. — The species name tJieskelos means wonderful. Sesioctomis venezuelensis Briceno, sp. n. Diagnosis. — S. venezuelensis is the only one species that shows five longitudinal carinae on the scutellar depression, in combination with the presence of occipital tubercles and occiput excavate. Description. — 8. Length. Body, excluding ovipositor, 8 mm. Head: Flagellum with broken after flagellomere 28. Interantennal space with rounded longitudinal keel. An- tennal sockets moderately excavated. Face without median longitudinal carina. Ge- nae not expanded posteroventrally. Occip- ital tubercles present. Occiput excavated. Mandible concave. Maxilla with 4 pal- pomeres. Third and fourth labial palpo- meres completely fused. Mesosoma: Sub- pronope oval. Longitudinal carinae of scu- tellar depression present. Scutellum flat. Median areola of metanotum smooth; without median longitudinal carina; and with lateral carinae present and not meet- ing posteriorly. Propodeum flat, median longitudinal carina absent. Epicnemial ca- rina sharp, complete, bilobed medially (between forecoxae). Foretibia spines ab- sent. Midtibia with 7 spines. Hind tibia with 16 spines. Hind femur 3.5 times as long as wide. (RS+M)a vein of forewing complete. 3RSa vein of forewing present. 2-1 A vein of hind wing not tubular. CUb vein of hind wing not tubular. Hind wing with 5 hamuli. Metasoma: Median tergite of first metasomal segment without pair of lateral longitudinal carinae. First metaso- mal median tergite without depression posterad spiracle. Length width ratio of first metasomal median tergite 1.28. Color: Head melanic. Antenna melanic. Maxil- lary and labial palpomeres yellowish or- ange. Mesosoma melanic. Forelegs yellow- ish orange except coxa melanic. Midlegs yellowish orange except coxa, trochanter and trocantellus melanic. Hindleg melanic except femur melanic but yellowish or- ange apically. Forewing banded from base, yellow, infuscate, yellow, infuscate. Stigma melanic. Hind wing banded from base, yellow, infuscate, yellow, infuscate. Metasoma yellowish orange with third tergum yellowish orange but median ter- gite melanic in posterior quarter and fourth and fifth tergum melanic. 9 .—Unknown. Material examined. — Holotype. Venezuela: 8 , Aragua Est. Exp. Cataurito, 32km E Vil- la de Cura, 1100m, 28.vi.1983, O. S. Flint, Jr. (CNC). Distribution. — This species is known only from the Aragua region of Venezue- la. Etymology. — This species is named after the country of the holotype specimen. ACKNOWLEDGMENTS I am sincerely grateful to the entomological collec- tions that lent the specimens for this research. Special acknowledgment to Dr. Michael Sharkey for his im- portant advising and to Ing. Quintin Arias for assis- tance with graphics software. Jose Clavijo, Jose Luis Garcia and Jurg De Marmels did important correc- tions for the first manuscript. The "Centro de Mi- croscopia Electronica de la Facultad de Agronomia- UCV" (CENMEFA- Venezuela) prepared specimens for SEM and took the photomicrografhs. Support for this investigation was provided by FONACIT (Project SI-2000000479), by the Consejo de Desarrollo Cientf- fico, Humanistico y Tecnologico (CDCHT-028-AG- 2001) of the Universidad Centroccidental "Lisandro Alvarado" (Barquisimeto- Venezuela) and by the Fun- dacion Cientifica y Tecnologica of the Aragua state (FUNDACITE- Aragua- Proyecto 2000-FHR-02-05-02- 1). Specimens from Colombia were provided by M.Sharkey through with the support of NSF grant DEB9972024. LITERATURE CITED Dallwitz, M. J., T. A. Paine and E. J. Zurcher. 1997. User's guide for DELTA System: A general system for processing taxonomic descriptions. Australia. 104 p. Volume 12, Number 2, 2003 271 Quicke, D. 1997. Parasitic wasps. Chapman & Hall publications. 470 p. Sharkey, M. J. 1988. A taxonomic revision of Alaba- grus (Hymenoptera: Braconidae). Bulletin of the British Museum (Natural History). 57 (2): 311-437. Sharkey, M. J. 1992. Cladistic and tribal classification of the Agathidinae (Hymenoptera: Braconidae). Journal of Natural History. 26 425-447. Sharkey, M. J. 1997. Subfamily Agathidinae. In Whar- ton, R. A.; P. M. Marsh and M. J. Sharkey. Manual of the New World genera of the family Braconidae (Hymenoptera). Special publication of the Inter- national Society of Hymenopterists. Number 1. Washington D. C. 439 p. Viereck, H. L. 1912. New genus and species of Hy- menoptera of the family Braconidae from Pana- ma. Smithsonian Miscellaneous Collections. 59: 1-2. Viereck, H. L. 1914. Type species of Ichneumon-flies. Bulletin of the United States National Museum. 83: 1-186. Wharton, R. A., P. M. Marsh, and M. J. Sharkey. 1997. Manual of the Neiv World genera of the family Bra- conidae (Hymenoptera). Special Publication of the International Society of Hymenopterists. Num- ber 1. Washington, D. C. 439 p. J. HYM. RES. Vol. 12(2), 2003, pp. 272-277 The Effect of Body Size on Male-Male Combat in the Parasitoid Wasp Melittobia digitata Dahms (Hymenoptera: Eulophidae) Christopher S. Hartley and Robert W. Matthews University of Georgia, Department of Entomology, Athens, GA 30602-2603, USA, CSH email: chartley@uga.edu; RWM email: rmatthew@uga.edu Abstract.— The parasitic wasp Melittobia digitata Dahms (Hymenoptera: Eulophidae) is a gregar- ious external parasitoid of various insects, primarily solitary wasps and bees. Males of M. digitata commonly engage in fierce, often fatal, fights with other males. The mandibles are the main weapons used, and injuries inflicted vary greatly in severity from loss of appendages to death. We investigated the effect of size differences between winning, losing, and non-fighting males and whether body size was related to being a fighter or non-fighter. The head width and tibia length of fighting and non-fighting males were measured. Winning males (21 of 29 pairs) were found to be larger than losing males, and fighting males (winners and losers combined) were found to be larger than non-fighting males (8 pairs). Mandible lengths of a subset of all males (15 fighters, 12 non-fighters) were subsequently measured; only fighter's left mandible length was significantly correlated with head width. The possibility that two behavioral forms (fighters and non-fighters) exist is considered but will require further experiments to resolve. Ritualized fighting over food, territory, the swollen inflorescences of various fig mates and other resources occurs between species, and fights occur between well-ar- males in the majority of animal species, mored flightless males that pursue newly Many studies have emphasized the re- emerged females. The majority of males strained nature of male fights by showing do not emerge from the fig in which they that the majority of fights end peacefully are born though emergence is more com- with neither combatant being injured, mon in some species (Bean and Cook even in species where the males possess 2001). Males have also been shown to large and dangerous weapons (Maynard have a strong attraction to the fig in which Smith and Price 1973; Maynard Smith they were reared (Frank 1985), which 1982). In addition, fighting males often makes dispersal unlikely. Since potential display striking dimorphisms such as mating opportunities are limited, fights those found in certain beetles (Forsyth and between males are fierce and result in Alcock 1990; Zeh et al. 1992; Goldsmith many fatalities. and Alcock 1993) and mites (Saito 1990, Melittobia digitata Dahms (Hymenop- 1995). Instances of fatal fighting have been tera: Eulophidae) is a gregarious external observed in some animal species. Where parasitoid of many different insects, but they do occur, fatal fights usually involve its principal hosts are solitary wasps and opportunities to mate, and they are limit- bees. In the southeastern United States, M. ed to species where males have limited digitata is most commonly found attacking opportunities to mate (Enquist and Leimar the mud dauber wasp, Trypoxylon politum 1990). Fatal fighting has been well docu- Say (Hymenoptera: Sphecidae). A female mented in many fig wasps (Hamilton Melittobia enters a Trypoxylon cocoon be- 1979; Murray 1987; Bean and Cook 2001). fore it is sealed and waits until the host These wasps spend their entire life within transforms into a prepupa before ovipos- Volume 12, Number 2, 2003 273 iting hundreds of eggs directly onto the culturing in the laboratory does not alter prepupa's cuticle (Dahms 1984; Gonzalez Melittobia behavior (Assem and Jachmann and Teran 2001). Upon hatching, the gre- 1999), so intense fighting is not likely to gariously developing Melittobia larvae be an artifact of mass rearing, consume the host, complete their devel- The occurrence of fatal fighting in both opment and begin to emerge in approxi- fig wasps and Melittobia is unexpected be- mately 20 days, depending on tempera- cause males are normally fighting their ture. The average brood size for M. digitata brothers. Hamilton (1979) suggested that resulting from one female foundress is 522 fighting behavior would not exist where a young. Males and females emerge togeth- male's rival has a high chance of being a er, but the sex ratio is extremely female brother. Recent work with fig wasps, how- biased — about 98% female (J.M. Gonzalez, ever, found no relationship between relat- personal communication). Females char- edness of males and fighting behavior. It acteristically mate once, usually soon after was found instead that the level of fatal emergence, and one male may mate with fighting was negatively correlated with fu- numerous females in his lifetime. Mated ture mating opportunities (West et al. females then chew their way out of the 2001). A similar situation exists in Melit- host's cell and disperse to search for new tobia where males have little chance of fu- hosts (Dahms 1984). ture mating opportunities since they are Sexual dimorphism is extreme in Melit- not likely to disperse. tobia. Adult males possess vestigial eyes, The objective of this study was to de- short non-functional wings, enlarged an- termine if size differences exist in M. dig- tennal scapes, and mandibles with well- itata between winners and losers of fights developed teeth. The blind males wander and between fighting and non-fighting freely inside the host's cocoon until they males. We hypothesized that winners encounter a female or another male. En- would be larger than losers and that fight- counters with females instigate courting ers would be larger than non-fighters, and mating behaviors, and virgin females often gather in groups around males to MATERIALS AND METHODS await mating (Gonzalez et al. 1985; Con- Melittobia digitata cultures were reared soli et al. 2002). Encounters with other in an incubator at 25°C on T. politum pre- males quickly escalate into a grappling pupae in small plastic boxes (5cm X 2.5cm contest where the males interlock their x 1.8cm) with tightly fitting lids. Males legs and struggle briefly with each other, were removed from cultures as pupae and Following these bouts, the males will ei- isolated in Carolina^ clear Deep Well Pro- ther separate or attempt to use their man- jection Slides (25 mm diameter, 2 mm dibles to tear at the body of the opponent, deep). This isolation ensured that a male's These fights often lead to loss of append- age and prior mating and fighting expe- ages and death in one or both fighters rience could be controlled. No data were (Dahms 1984). Inside naturally parasitized recorded on male emergence time relative mud dauber cocoons, one routinely finds to other males from a particular culture the remains of several males, many dis- nor from which culture a given male membered. In most laboratory cultures of emerged. Thus, males used in the experi- M. digitata, males grapple and fight with ments can be regarded as arbitrarily se- little provocation, and these contests fre- lected from among a range of males avail- quently end with the death of one or both able. combatants. In other cultures, we have When the males isolated in the depres- found many males alive with no injuries sion slides emerged, the date of their and no evidence of fighting. Long term emergence was recorded. Eighty-seven 274 Journal of Hymenoptera Research Table 1. Morphometric measurements of M. digitata males. Total number Winners Losers Fighters Non-Fighters 21 21 42 16 * Values in mm ± Standard Deviation. Mean tibia length* 0.25 ± 0.03 0.23 ± 0.03 0.24 ± 0.03 0.20 ± 0.03 Mean head width* 0.35 ± 0.04 0.34 ± 0.04 0.35 ± 0.04 0.29 ± 0.03 males were kept isolated and observed daily, and their date of death was record- ed. These males served as controls for the following experiment. Twenty-nine newly emerged male pairs were formed in the depression slides by opening the two individual slides and us- ing a paintbrush to move one of the males into the other male's slide. Because of the difficulty of marking individuals, it was not possible to track which male was res- ident versus intruder in the pairings. Once each pair had been formed, the slides were not opened again until after both males had died. Each pair was observed daily, and the date of each male's death was recorded. After death, each male was examined, and any obvious injury (e.g., loss of ap- pendages, body wounds) was recorded. The head and right front tibia of each male were then mounted temporarily in glyc- erol on standard glass microscope slides and measured under 50 X magnification using an ocular micrometer. Both males of one pair were mounted on the same slide so that winners and losers could be asso- ciated. To avoid crushing the specimens, pieces of 61b. test nylon fishing line were placed around them to elevate the cover slip. To test whether head width is correlated with mandible length, 15 fighter heads and 12 non-fighter heads were arbitrarily selected, and their mandibles were dis- sected. Both left and right mandibles were mounted on microscope slides and their maximum length was measured. Sign tests were used to analyze winner versus loser data so that both males of one fighting pair could be compared against each other. Mann-Whitney U tests were used to analyze fighter versus non-fighter data. Spearman R Correlations were used for all correlations. A P value of 0.05 was taken as the critical value for establishing significance. Analyses were done using STATISTICA 6 © StatSoft, Inc. RESULTS In 21 of the pairs, one male killed the other in a fight. These males were termed fighters. In eight of the pairs, the males were never observed to come into contact with each other, and after death neither male was found to have lost appendages or incurred wounds to the body. We con- cluded in these cases that no fighting had occurred, and these males were recorded as non-fighters. Table 1 shows the measurement data for all males. Winners had significantly longer tibia (Z = 2.29; P = 0.022), but head widths of winners and losers did not differ signifi- cantly (Z = 1.21; P = 0.228). The tibia length of all fighters (winners and losers combined) was significantly longer than the non-fighters' tibia length (U = 134.0; Z - 3.58; P < 0.001), and fighters' heads were significantly wider than non-fight- ers' (U - 88.5; Z = 4.34; P < 0.001). Head width and tibia length for all males (fight- er and non-fighter) were significantly cor- related (p = 0.665; P < 0.001). The only significant correlation between head width and mandible length was found for fighters' left mandibles (p = Volume 12, Number 2, 2003 275 14 12 10 3 £ 6 ■ Fighters ■ Non-fighters 1 «- CM o d CO o d d CD d d CM t- ^ CM cm co CM CM odd CM CM CD CM CM co CM Tibia Length (mm) Fig. 1. Frequency distribution for tibia length comparing all fighters and non-fighters (n = 58). 0.692; P = 0.004). Fighters' right mandi- bles were not significantly correlated with head width (p = 0.351; P = 0.200). Non- fighters' left mandibles were not signifi- cantly correlated with head width (p = 0.507; P = 0.092), and non-fighters' right mandibles were not significantly correlat- ed with head width (p = 0.452; P = 0.140). Differences in size between fighters and non-fighters and the obvious behavioral differences suggested the possibility that the Melittobia male population could be di- morphic. To explore this possibility, fre- quency distributions of the measurements of tibia length and head width were pre- pared (Figs. 1 and 2). If a dimorphism ex- ists, a bimodal curve is expected. The graph of tibia length frequency reveals only one peak for both fighters and non- fighters, and this peak occurs at 0.23mm. The graph of head width frequencies sug- gests the presence of two peaks — one for non-fighters at 0.28mm and one for fight- ers at 0.37mm. There is, however, consid- erable overlap, and values for fighters and non-fighters occur at both ends of the scale. DISCUSSION We predicted that winners would be the larger males. The results revealed that winners were larger than losers based on their tibia length measurements, although their head widths did not differ. The cor- relation between head width and tibia length suggests that relative size of either is likely to be a good predictor of overall body size, which in turn is related to fight- ing success. Mandible size has been often discussed in relation to fighting fig wasps (Bean and Cook 2001). The frequent asym- metry between right and left mandible lengths in our study was unexpected, and may be interesting to pursue. The existence of non-fighter males, in which paired males never engaged in a fight even though they were isolated to- gether for their whole life, raises the pos- 276 Journal of Hymenoptera Research Fig. 2. Frequency distribution for head width comparing all fighters and non-fighters (n = 58). sibility that males exist in two behavioral morphs. Freeman and Ittyeipe (1982) de- scribed two morphologically distinct male morphs in Melittobia hawaiiensis Perkins ( = M. australica Girault): a larger morph with ocelli and a smaller morph without ocelli. However, we found that all of our M. dig- itata males, large and small, had fully pig- mented ocelli. The frequency graphs of head width and tibia length (Figs. 1 and 2), while showing a trend towards a bi- modal distribution for each trait, also re- veal that a wide range of sizes exists in both fighting and non-fighting males. Overall, most of the larger males became fighters while most of the smaller males never engaged in fights, but there were obvious exceptions. Perhaps fighting and non-fighting are conditional rather than fixed traits. Alternatively, there could be culture or lineage specific effects on the likelihood of a male becoming a fighter or non-fighter. These questions will require further investigation. Abe et al. (2003) found that when an emerged male and a pupal male were placed together, the already emerged male usually killed the pupal male at or im- mediately after eclosion, but they did not record sizes of any of the males in their experiment. We have noted that the first males to emerge are generally larger most likely due to better food quality and quan- tity, and small males emerge later in the culture's life, when the host is covered with developing pupae (unpublished data). Males of different fig wasp species are known to exist in a wide variety of body shapes, each with a different fighting propensity, and each is adapted to court- ing females in a different way (Murray 1990). If fighting behavior in Melittobia is linked to size, then perhaps a similar sit- uation occurs with small, non-fighting males being better adapted to maneuver- ing amongst the developing pupae in the tightly packed confines of a Trypoxylon co- coon. They could avoid the stress of fight- Volume 12, Number 2, 2003 277 ing by staying hidden, but they could still court females. Repeating these experi- ments using males from one culture and tracking relative emergence times could help to determine if males that emerge early tend to be larger and become fight- ers and males that emerge late tend to be smaller and become non-fighters. ACKNOWLEDGMENTS Jorge M. Gonzalez and Leif Deyrup provided in- valuable comments on the experimental design and the manuscript. We also thank Jan Matthews, David Jenkins, Christian Torres, and LuAnn Brown for all of their help and advice. This study was conducted as a senior Honor's Thesis at the University of Geor- gia, and was supported in part by NSF Grant 0088021, R. W. Matthews, Principal Investigator. LITERATURE CITED Abe, J., Y. Kamimura, N. Kondo, and M. Shimada. 2003. Extremely female-biased sex ratio and le- thal male-male combat in a parasitoid wasp, Mel- ittobia australica (Eulophidae). Behavioral Ecology 14: 6-11. Assem, J. van den, and F. Jachmann. 1999. Changes in male perseverance in courtship and female readiness to mate in a strain of the parasitic wasp Nasonia vitripennis over a period of 20+ years. Netherlands Journal of Zoology 49: 125-137. Bean, D., and J. M. Cook. 2001. Male mating tactics and lethal combat in the nonpollinating fig wasp Sycoscapter australis. Animal Behaviour 62: 535- 542. Consoli, F. L., H. J. Williams, S. B. Vinson, R. W. Mat- thews, and M. F. Cooperband. 2002. trans-Ber- gamotenes — Male pheromone of the ectoparasi- toid Melittobia digitata. journal of Chemical Ecology 28: 1675-1689. Dahms, E. C. 1984. A review of the biology of species in the genus Melittobia (Hymenoptera: Eulophi- dae) with interpretations and additions using ob- servations on Melittobia australica. Memoirs of the Queensland Museum 21: 337-360. Enquist, M., and O. Leimar. 1990. The evolution of fatal fighting. Animal Behaviour 39: 1-9. Forsyth, A., and J. Alcock. 1990. Female mimicry and resource defense polygyny by males of a tropical rove beetle Leistotrophus versicolor Coleoptera Sta- phylinidae. Behavioral Ecology and Sociobiology 26: 325-330. Frank, S. A. 1985. Are mating and mate competition by the fig wasp Pegoscapus assuetus (Agaonidae) random within a fig? Biotropica 17: 170-172. Freeman, B. E., and K. Ittyeipe. 1982. Morph deter- mination in Melittobia, a eulophid wasp. Ecologi- cal Entomology 7: 355-363. Goldsmith, S. K., and J. Alcock. 1993. The mating chances of small males of the cerambycid Trach- yderes mandibularis differ in different environ- ments (Coleoptera: Cerambycidae). journal of In- sect Behavior 6: 351-360. Gonzalez, J. M., and J. B. Teran. 2001. Dispersion, busqueda y acceso al hospedador por Melittobia acasta (Hymenoptera: Eulophidae). Boletin del Centro de Investigations Biologicas 35: 52-64. Gonzalez, J. M., R. W. Matthews, and J. R. Matthews. 1985. A sex pheromone in males of Melittobia aus- tralica and Melittobia femorata (Hymenoptera: Eu- lophidae). Florida Entomologist 68: 279-286. Hamilton, W. D. 1979. Wingless and fighting males in fig wasps and other insects, pp. 167-220. In M.S. Blum and N.A. Blum (eds.), Sexual Selection and Reproductive Competition in Insects. Academic Press, New York. Maynard Smith, J. 1982. Evolution and the Theory of Games. Cambridge University Press, Cambridge. Maynard Smith, J., and G. R. Price. 1973. The logic of animal conflict. Nature 246: 15-18. Murray, M. G. 1987. The closed environment of the fig receptacle and its influence on male conflict in the Old World fig wasp, Philotrypesis pilosa. Animal Behaviour 35: 488-506. Murray, M. G. 1990. Comparative morphology and mate competition of flightless male fig wasps. Animal Behaviour 39: 434-443. Saito, Y. 1990. Factors determining harem ownership in a subsocial spider-mite (Acari, Tetranychidae). Journal of Ethology 8: 37-43. Saito, Y. 1995. Clinal variation in male-to-male antag- onism and weaponry in a subsocial mite. Evolu- tion 49: 413-417. West, S. A., M. G. Murray, C. A. Machado, A. S. Grif- fin, and E. A. Herre. 2001. Testing Hamilton's rule with competition between relatives. Nature 409: 510-512. Zeh, D. W., J. A. Zeh, and G. Tavakilian. 1992. Sexual selection and sexual dimorphism in the harle- quin beetle Acrocinus longimanus. Biotropica 24: 86-96. J. HYM. RES. Vol. 12(2), 2003, pp. 278-302 Morphological Variation in Opius Wesmael (Hymenoptera: Braconidae) with an Emphasis on Nearctic Species in the Subgenus Gastrosema Fischer Robert R. Kula Department of Entomology, Texas A&M University, College Station, TX, USA 77843, email: rkula@oznet.ksu.edu Abstract— The variability of morphological characters and their potential phylogenetic utility in Opiinae are evaluated using Nearctic species in the subgenus Gastrosema Fischer. Diagnoses for three species groups within Nearctic Gastrosema are provided, and 43 Nearctic species in Opius Wesmael are placed in the species groups. Two new Nearctic species in Gastrosema are described: Opius abbyae Kula and O. lacopitaensis Kula. Five previously described Nearctic species in Gastrosema are redescribed: O. castaneigaster Fischer, O. intermissus Fischer, O. oscinidis (Ashmead), O. salmonensis Fischer, and O. striativentris Gahan. Opius gracillariae Gahan is designated a junior synonym of O. striativentris. Opiinae is a very large subfamily in Bra- ical limits of these subgenera need to be conidae with approximately 1,500 de- evaluated. This research begins to address scribed species worldwide (Wharton these problems by performing a detailed 1997b). All valid rearing data suggest that assessment of intraspecific morphological species in Opiinae are strict koinobiont en- variation for Nearctic species in Gastrose- doparasitoids of cyclorrhaphous Diptera ma. (Wharton 1999). Opius Wesmael is the Other than Fischer's (1977) revision of largest genus in Opiinae, and over 1,000 New World Opiinae, there has been little species are currently placed in Opius sensu work on Nearctic species in Gastrosema. Fischer (1972). In a series of regional re- Gahan (1915) described two Nearctic spe- visions, Fischer (1972, 1977, 1987) segre- cies currently included in Gastrosema and gated species in Opius into 27, 30, and 34 provided a key to the North American subgenera, respectively. Most of the sub- species in Oprius. Fischer (1964, 1970) de- genera are broadly defined and lack fea- scribed several Nearctic species in Opius tures that clearly characterize them as that were subsequently placed in Gastro- monophyletic. Further, the defining fea- sema by Fischer (1977). Marsh (1974) syn- ture of several subgenera, including Gas- onymized two Nearctic species in Gastro- trosema Fischer, Merotrachys Fischer, Phae- sema, but insufficient material was exam- drotoma Forster sensu Fischer (1972), and ined to assess intraspecific morphological Tolbia Cameron sensu Fischer (1972), is the variation. Additional specimens must be presence or absence of sculpture associat- acquired and compared with the holo- ed with specific anatomical features. Intra- types to confirm Marsh's synonymies. Van specific variation of sculpture has never Achterberg (1997) transferred O. pumilio been assessed for species in these subgen- Wesmael (the type species for Gastrosema) era, and preliminary sorting indicates that to Phaedrotoma sensu van Achterberg and certain species fit the concept of more than Salvo (1997). This action suggests that van one subgenus. Therefore, the morpholog- Achterberg (1997) treated Gastrosema Fi- Volume 12, Number 2, 2003 279 scher 1972 as a junior synonym of Phae- drotoma Forster 1862, although this was not explicitly stated. Since only some spe- cies currently classified as Gastrosema pos- sess the characters used to define Phaedro- toma sensu van Achterberg and Salvo (1997), synonymy seems inappropriate. Additionally, Phaedrotoma sensu van Ach- terberg and Salvo (1997) is partially de- fined by the absence of a basal mandibular tooth which the authors regarded as ple- siomorphic. It appears that Phaedrotoma sensu van Achterberg and Salvo (1997) has not been defined as a monophyletic group, and I see no advantage in accept- ing the changes proposed by van Achter- berg (1997) over Fischer's (1972) concept of Gastrosema. MATERIALS AND METHODS Specimens used in this study were bor- rowed from several North American mu- seums and collections. The American En- tomological Institute (AEI), the California Academy of Sciences (CAS), the Canadian National Collection of Insects (CNCI), the Insect Research Collection at the Univer- sity of Wisconsin-Madison (IRCW), the Museum of Comparative Zoology at Har- vard University (MCZ), the insect collec- tion at Texas A&M University (TAMU), and the United States National Museum of Natural History (USNM) provided de- termined and undetermined opiines. In addition to the borrowed material, two opiine species treated in this study were reared from isolated puparia. Dr. Sonja J. Scheffer provided several hundred speci- mens of O. striativentris Gahan reared from five species of Phytomyza Fallen min- ing the leaves of seven species of Ilex L. (hollies). Eighteen specimens of an unde- scribed opiine were reared by the author from a species of Calycomyza Hendel min- ing the leaves of Helianthus annuus L. (sun- flower). Species treated in this study can be identified to Opiinae using Sharkey (1997) and to Opius using Wharton (1997a). Sev- eral hundred Nearctic species in at least 11 subgenera have not yet been examined but potentially fall into the species groups treated in this paper. As a result a com- prehensive key to the Nearctic species in each species group cannot be provided at this time. Therefore, a detailed diagnosis is provided for each species group and species to facilitate their identification. Descriptive terminology for anatomical features, including wing cells and veins, is based on Sharkey and Wharton (1997). Terminology used to describe surface sculpture follows Harris (1979) and Shar- key and Wharton (1997). Two anatomical terms used throughout this work are not found in Harris (1979) or Sharkey and Wharton (1997). Both terms must be de- fined and illustrated to provide maximum clarification. The anterior pronotal furrow is a groove located at the antero-lateral edge of the pronotum just posterior to the propleural flange, and the posterior me- sopleural furrow is a groove located at the posterior edge of the mesopleuron just an- terior to the metapleuron (Fig. 1). Several measurements were taken to quantitatively characterize each species described or redescribed in this study (Figs. 2-7). Two criteria were used to se- lect measurements: 1) measurements his- torically used in opiine species descrip- tions and 2) measurements that describe the general size and shape of each species. MORPHOLOGICAL VARIATION Results from the analysis of intraspecific morphological variation provide a basis for discussing character variability and their potential phylogenetic utility in Opi- inae. The following section is an assess- ment of characters potentially useful for supporting monophyletic groups, as well as characters useful for defining the limits of species. Size and shape ofclypeus. — Fischer (1972) used the exposure of the lab rum in frontal view (a result of clypeus length) to par- tially define certain subgenera. For exam- 280 Figs. 1-7. 1, Anterior pronotal furrow (arrow a) and posterior mesopleural furrow (arrow b). 2, Head width (HW). 3, Mesosoma width (MW) and tergite 1 width (tlW). 4, Head length (HL), eye length (EL), eye width (EW), mesosoma length (ML), and meso- soma depth (MD). 5, Tergite 1 length (tlL) and ex- posed ovipositor length (EOL). 6, Hind tibia length (HTL). 7, 2RS length (2RSL), 3RSa length (3RSaL), and 2M length (2ML). Figs. 1 and 3-5 modified from Fischer (1972). pie, species that Fischer included in Sto- mosema Fischer possess a broad clypeus completely concealing the labrum (Fig. 8). Alternatively, species that Fischer includ- ed in Gastrosema have a reduced clypeus Journal of Hymenoptera Research exposing the labrum in frontal view (Fig. 9). Three basic clypeal morphologies are observed among Nearctic species in Gas- trosema. Both O. oscinidis (Ashmead) and O. walleyi Fischer possess a flattened or convex clypeus with the ventral margin lobed mesally (Fig. 10). This condition is also commonly observed for Nearctic spe- cies in RJwgadopsis Brethes sensu Wharton (1987) and Thoracosema Fischer. Opius striativentris possesses a convex, hemi- spherical to narrowly hemielliptical shaped clypeus with a straight to slightly arched ventral margin (Fig. 9). This is the most commonly observed clypeal shape for Nearctic species in Gastrosema and is also commonly observed for Nearctic spe- cies in Phaedrotoma sensu Fischer (1972) and Tolbia sensu Fischer (1972). Both O. flaviceps Gahan and O. sahuonensis Fischer possess a flattened, broadly hemielliptical shaped clypeus with a straight to slightly arched ventral margin (Fig. 11). This clyp- eal shape is also commonly observed for Nearctic species in Pleurosema Fischer. The consistency of clypeus size and shape within a species, coupled with the vari- ability of clypeus size and shape among groups of species, suggests that this char- acter is potentially phylogenetically infor- mative and merits consideration in future phylogenetic analyses. Shape of mandibles. — Mandible shape is another feature Fischer (1972) used to par- tially define certain subgenera. For exam- ple, Opiognathus Fischer and Opiostomus Fischer contain species with mandibles that are abruptly widened distally to prox- imally (basal tooth present) (Fig. 12). In more recent revisionary work, van Ach- terberg and Salvo (1997) used mandible shape to split Opius sensu Into into two genera: Opius and Phaedrotoma. Opius sen- su van Achterberg and Salvo (1997) in- cludes species that possess a distinct basal tooth, while Phnedrotomn sensu van Ach- terberg and Salvo (1997) includes species that lack a distinct basal tooth (Fig. 13). Volume 12, Number 2, 2003 281 VR9V ^^^^r fee - %^-' """ m '■■ s^fuwSHi3^ ■ M WfflH 1 Ifiu^?i .. Figs. 8-11. Size and shape of clypeus. 8, Broad and completely concealing labrum. 9, Reduced (arrow a) and exposing labrum (arrow b). 10, Ventral margin lobed mesally. 11, Broadly hemielliptical with slightly arched ventral margin. Van Achterberg and Salvo (1997) explicit- ly regarded the presence of a basal tooth as apomorphic. Thus, if the presence of a basal tooth is apomorphic, Phaedrotoma sensu van Achterberg and Salvo (1997) is partially defined by a symplesiomorphy. The nomenclatural changes in van Achter- berg and Salvo (1997) collapse the subge- neric classification established by Fischer (1972, 1977, 1987). The species currently included in Opius sensu lato are segregated into the putatively monophyletic genus Opius sensu stricto or the seemingly para- phyletic genus Phaedrotoma sensu lato. Spe- cies with mandibles that are gradually widened distally to proximally (Fig. 14) cannot be unequivocally placed in either genus. Additionally, species that exhibit distinct similarities in clypeal shape, body sculpture, and wing venation (e.g. O. rc- lativus Fischer and O. salmonensis) fall into different genera. Because of these short- comings I see no reason to adopt the clas- sification proposed by van Achterberg and Salvo (1997) over Fischer (1972, 1977, 1987). The Nearctic species of Opius sensu lato that fall within Gastrosema vary extensive- ly in mandibular shape. In O. lacopitaensis Kula, new species (described below) the basal tooth is obviously absent in most in- dividuals but gradually widened distally to proximally in some individuals. The ab- sence of the basal tooth is the most com- monly observed condition for Nearctic species in Gastrosema. The continuous na- ture of this character makes it verv diffi- J cult to code into discrete character states - Journal of Hymenoptera Research Figs. 12-14. Mandibles. 12, Basal tooth distinctly present. 13, Basal tooth completely absent. 14, Gradually widened distally to proximally. for certain species. However, species such as O. salmonensis consistently lack a basal tooth, and species such as O. relativus con- sistently possess a distinct basal tooth. Op- ius tangens Fischer is another Nearctic spe- cies in Gastrosema that possesses a distinct basal tooth. Unfortunately, only one indi- vidual was available for examination, so intraspecific variation of mandibular shape is not known for O. tangens. Nearc- tic species in several subgenera exhibit variation in mandibular shape similar to Nearctic species in Gastrosema. PJiaedroto- ma sensu Fischer (1972) and Pleurosema contain species that either obviously lack or distinctly possess a basal tooth, as well as species with mandibular shapes that cannot be reliably determined as basal tooth present or absent. The results from this study suggest that mandible shape is useful for distinguishing between certain species but may not be possible to code unambiguously for phylogenetic analyses. Sculpture associated with posterior meso- pleural furrow. — Fischer (1972) segregated species in Opius into different subgenera based on the presence or absence of cren- ulations within the posterior mesopleural furrow. This interpretation is inadequate because within a species the posterior me- sopleural furrow can be either entirely smooth or crenulate below but smooth above the episternal scrobe. Nearctic spe- cies in Gastrosema that exhibit this condi- tion include O. abbyae Kula, new species (described below) (Figs. 15-16). An entirely smooth posterior meso- pleural furrow is found in the majority of Nearctic species in Gastrosema. Yet several species in Gastrosema consistently possess Volume 12, Number 2, 2003 283 Figs. 15-16. Posterior mesopleural furrow. 15, Entirely smooth. 16, Crenulate below (arrow a) but smooth above (arrow b) episternal scrobe. Figs, modified from Fischer (1972). an entirely crenulate posterior mesopleur- al furrow (Fig. 1). Opius flaviceps, O. rela- tivus, O. salmonensis, and O. tangens are Nearctic species in Gastrosema that possess an entirely crenulate posterior mesopleur- al furrow. Additionally, several Nearctic species in Pleurosema consistently possess an entirely crenulate posterior mesopleur- al furrow. When interpreting this charac- ter in phylogenetic analyses, the location of crenulations in the posterior mesopleur- al furrow is more informative than the mere presence or absence of crenulations and may be important for establishing ho- mology. Presence or absence of mesoscntal midpit. — Fischer (1972) used the presence or ab- sence of the mesoscutal midpit (Fig. 17) to divide the subgenera of Opius into two roughly equal sized groups. The assess- ment of variation conducted in this study confirms the potential phylogenetic utility of this character. Midpit shape and depth vary intraspecifically in Opiinae, but the midpit is consistently present or absent for a particular species. All Nearctic species in Gastrosema consistently lack a midpit, but there are several Nearctic species in other subgenera that consistently possess a mid- pit. Because the midpit can be quite re- duced within a species, specimens should be examined at the appropriate angle and with light dispersing plastic (mylar). Presence, absence, and sculpture associated with notauli. — Species in Opiinae exhibit several character states associated with the notauli. Historically, the length of the no- tauli has been used to group putatively Figs. 17-18. 17, Mesoscutal midpit. 18, Notauli represented by shallow anterior depressions. 284 Journal of Hymenoptera Research Figs. 19-20. Sternaulus. 19, Short, narrow, and crenulate. 20, Represented by a smooth depression. closely related species (Fischer 1972). However, the presence or absence of cren- ulations within the notauli, as well as the termination point of the notauli, may pro- vide additional useful information for phylogenetic inference. Among Nearctic species in Gastrosema, the notauli are shal- low or deep anterior depressions that can be sculptured or unsculptured (Fig. 18). Presence, absence, and sculpture associated with sternaulus. — Fischer (1972) used the following three attributes to partially de- fine certain subgenera: sternaulus absent, sternaulus present and smooth, and ster- naulus present and sculptured (e.g. cren- ulate). Gastroscma is partially defined by the presence of sculpture within the ster- naulus. The most commonly observed condition for Nearctic species in Gastrose- ma is a relatively short and narrow ster- naulus with crenulate sculpture (Fig. 19). However, examination of multiple indi- viduals for several Nearctic species in Gas- troscma revealed that the sternaulus can vary from a smooth depression (Fig. 20) to distinctly crenulate within a species. Spe- cies that exhibit this condition include O. lacopitaensis and O. striativentris. Further, species such as O. castaneigaster Fischer of- ten possess a very weakly sculptured ster- naulus that can appear unsculptured us- ing a stereomicroscope with a fiber optic illuminator. Despite this large breadth of variation, Fischer (1972) used the presence and absence of sculpture within the ster- naulus to place species in Gastrosema and Phaedrotoma, respectively. Because the presence and absence of sculpture within the sternaulus can vary intraspecifically, certain Nearctic species in Phaedrotoma sensu Fischer (1972) potentially share a most recent common ancestor with certain Nearctic species in Gastrosema. The high level of intraspecific variation in sternau- lus sculpture limits the use of this char- acter in phylogenetic analyses. If sculpture within the sternaulus is utilized in future phylogenetic analyses, the location of the sternaulus and the type of sculpture with- in the sternaulus are important for estab- lishing homology. Sculpture associated with propodeum. — The presence or absence of propodeal sculpture is a major defining feature for several subgenera in Opius. In Fischer (1972) the presence or absence of propo- deal sculpture is the only attribute differ- entiating species in Merotrachys and Phae- drotoma, respectively. The presence, ab- sence, and type of propodeal sculpture are quite variable among Nearctic species in Gastrosema, but these attributes are only slightly variable within a species. For ex- ample, in O. striativentris the propodeum is rugulose to rugose (Fig. 21), in O. inter- missus Fischer the propodeum is always rugose, and in O. salmonensis the propo- deum is always areolate-rugose (Fig. 22). Volume 12, Number 2, 2003 285 Figs. 21-23. Propodeal sculpture. 21, Rugulose. 22, Areolate-rugose. 23, Smooth with rugulose sculpture surrounding spiracles. The majority of Nearctic species in Gastro- sema possess a rugulose to rugose propo- deum. However, species that possess an areolate-rugose propodeum are also com- monly encountered. Further, species with an areolate-rugose propodeum usually have an entirely crenulate posterior me- sopleural furrow. Nearctic species that ex- hibit this condition include O. flaviceps, O. relativus, O. salmonensis, and O. tangens. The majority of Nearctic species in Pleu- rosema also possess an areolate-rugose propodeum and an entirely crenulate pos- terior mesopleural furrow. Thus, certain Nearctic species in Pleurosema potentially share a most recent common ancestor with certain Nearctic species in Gastrosema. Of the described Nearctic species in Gastro- sema, only O. haucockamis Fischer has a nearly entirely smooth propodeum (Fig. 23). A trace of rugulose sculpturing sur- rounds the propodeal spiracles, but the rest of the propodeum is completely smooth. This condition is also commonly observed for Nearctic species in Phaedro- toma sensu Fischer (1972). Propodeal sculpture can be assigned distinct character states, and different types of propodeal sculpture often corre- late with sculpture in the posterior meso- pleural furrow. Species with an areolate- rugose propodeum frequently have an en- tirely crenulate posterior mesopleural fur- row, and species with a smooth or rugulose to rugose propodeum typically do not have an entirely crenulate posterior mesopleural furrow. Consideration of pro- podeal sculpture in phylogenetic analyses must go beyond the presence or absence of sculpture. Different types of propodeal 286 Journal of Hymenoptera Research sculpture, and the correlation between propodeal sculpture and sculpture asso- ciated with other anatomical features, should be considered. Sculpture associated with median abdomi- nal tergites. — In Fischer's original diagno- sis for Gastrosema, a major defining feature was the presence of "shagreened" or co- riaceous sculpture on abdominal tergite two (t2) (Fig. 24) (Fischer 1972). Exami- nation of the Nearctic species in Gastrose- ma treated by Fischer (1977) revealed that Gastrosema was not limited to species with a coriaceous t2. Rather, species with a car- inate or costate t2 were included as well (Figs. 25-26). The vast majority of Nearctic species included in Gastrosema have a co- riaceous t2. Examples include O. abbyae, O. castaneigaster, O. intermissus, O. lacopitaen- sis, and O. striativentris. Opius flaviceps and O. salmonensis are two species that possess a costate t2, and O. oscinidis possesses a carinate t2. Neither costate nor carinate sculpture on t2 is common among Nearc- tic species in Gastrosema, but a carinate t2 is commonly observed for several Nearctic species in Rhogadopsis sensu Wharton (1987) and Thoracosema. In terms of intra- specific variation, species with a coria- ceous t2 show little variation in sculpture density. Conversely, species with a cari- nate or costate t2 exhibit substantial intra- specific variation in sculpture density. In O. oscinidis the carinations are extremely reduced in some individuals, and t2 ap- pears smooth unless the specimens are ex- amined at several angles using mylar and a variety of light intensities. Yet the cari- nations on t2 are unmistakably present in other individuals. Thus, sculpture on t2 has potential phylogenetic importance but may be difficult to code unambiguously for certain species. Configuration of forewing venation. — Fi- scher (1972) used the position of vein lm- cu relative to vein 2RS to segregate species into certain subgenera. For example, spe- cies that Fischer placed in Phlebosema Fi- scher possess vein lm-cu basad or directly in line with vein 2RS (Figs. 27-28). Phle- bosema is currently a junior synonym of Baeocentrum Schulz, but this nomenclatur- al change does not alter Fischer's concept of Phlebosema (Wharton 1987). Alternative- ly, species that Fischer placed in Gastrose- ma possess vein lm-cu distad vein 2RS (Fig. 29). The results of this study confirm that species with vein lm-cu basad vein 2RS occasionally display vein lm-cu di- rectly in line with vein 2RS. Further, spe- cies with vein lm-cu distad vein 2RS nev- er exhibit vein lm-cu directly in line with or basad vein 2RS. All described Nearctic species in Gastrosema have vein lm-cu dis- tad vein 2RS. However, the Nearctic spe- cies in Baeocentrum sensu Wharton (1987) and certain Nearctic species in Merotrachys and Tolbia sensu Fischer (1972) have vein lm-cu basad or directly in line with vein 2RS. Thus, the position of vein lm-cu rel- ative to vein 2RS should be evaluated in future phylogenetic analyses. The size and shape of the forewing stig- ma is another character of potential phy- logenetic importance. The Nearctic species in Gastrosema exhibit minimal intraspecific variation in stigma size and shape, but stigma size and shape can vary substan- tially from species to species. Several Ne- arctic species in Gastrosema exhibit con- spicuous similarities in stigma size and shape. Opius abbyae, O. castaneigaster, O. intermissus, O. lacopitaensis, and O. striativ- entris possess a narrow and elongate stig- ma (Fig. 28), while O. flaviceps, O. relativus, O. salmonensis, and O. tangens possess a wedge shaped stigma (Fig. 29). In future studies stigma size and shape should be thoroughly examined for species in other opiine subgenera. Color. — In Opiinae color is not of any apparent phylogenetic importance but is useful for distinguishing between mor- phologically similar species. For example, O. intermissus and O. niobe Fischer are sim- ilar in size and shape but have conspicu- ously different coloration on the head and mesosoma. Species such as O. striativentris Volume 12, Number 2, 2003 287 Figs. 24-26. Sculpture on t2. 24, Coriaceous. 25, Carinate. 26, Costate. are variable in color, while color is more conserved in species such as O. abbyae. Thus, it is important to examine several individuals to determine the entire spec- trum of color within a species. Proper specimen preparation is essential for pre- serving natural coloration. Specimens ex- posed to light for prolonged periods of time become lighter, and air-dried speci- mens tend to darken. Material preserved in alcohol for extended periods of time should be stored at low temperatures and protected from direct exposure to light. Specimens should be dehydrated using a critical-point-dryer or chemicals such as amyl acetate and Hexamethyl-Disilazane. Setation. — Setation is another useful character for distinguishing between mor- phologically similar species. For example, a species of Opius near extiratus Fischer possesses a consistently setose propo- deum, while setation on the propodeum of O. striativentris is usually setiferous and rarely glabrous. Opius near extiratus and O. striativentris are sympatric and have been reared from the same host species on the same host plant. The density of setae on the propodeum is one character that can be used to distinguish these species. Flagellomere number. — Godfray (1984) demonstrated that in the alysiine Exotela cyclogaster Forster flagellomere number in- creases with body size. Alysiinae and Opi- inae form a monophyletic group (Wharton 1988), and it is possible that species in Opiinae also exhibit this trend. Flagello- mere number is useful when morpholog- ically similar species have non-overlap- ping ranges of flagellomere number. For example, O. abbyae and O. intennissus are morphologically similar, but O. abbyae has 19-22 flagellomeres compared to 26-29 in 288 Journal of Hymenoptera Research 27 28 29 Figs. 27-29. Position of forewing vein lm-cu (arrow a) relative to vein 2RS (arrow b). 27, lm-cu basad 2RS. 28, lm-cu directly in line with 2RS. 29, lm-cu distad 2RS. O. intermissus. When using flagellomere number to distinguish between species, it is important to examine several individu- als to establish minimum and maximum flagellomere numbers for each species. Quantitative characters. — Fischer (1972, 1977, 1987) extensively used quantitative ratios in keys to subgenera, species groups, and species in Opius, but the ra- tios were based on measurements of one or a few individuals. As a result most taxa treated in the keys cannot be reliably iden- tified. A prime example is the treatment of species groups in Fischer's (1977) key to New World Gastrosema. The defining char- acter of the lissopleurum-group of species is "thorax 1.25-1. 40 X as long as high." Opius intermissus is included in the lisso- pleurum-group, but the results of this study indicate that the thorax is 1.12- 1.30X as long as high in O. intermissus. Only 33.0% of the specimens examined for O. intermissus fit within the morphological limits of the lissopleurum-group as defined by Fischer (1977). Another problem with the use of quan- titative ratios is that measurement error as high as 15.7% can occur when the same specimen is measured on different dates. Factors that contribute to measurement er- ror may be found in Wharton (1980). Giv- en the limitations of quantitative char- acters, it is preferable to use qualitative characters in phylogenetic analyses and identification systems. However, measure- ments can be useful for distinguishing cer- tain species. Opius flaviceps and O. salmo- nensis are similar in terms of coloration and sculpture, but the two species can be differentiated using ovipositor length, me- sosoma length, and mesosoma width. When using measurements to characterize a species, it is optimal to measure several individuals to establish the entire range of variation for each measurement. In cases where only one or a few individuals are available for examination, the taxonomist should consider the usefulness of quanti- tatively describing the species. NEARCTIC SPECIES IN THE SUBGENUS GASTROSEMA Species group diagnoses. — With the excep- tion of O. walleyi, all described Nearctic species in Gastrosema fall into three easily characterized groups that have not been recognized in previous studies (Table 1). A diagnosis of each group is provided to facilitate future studies examining the monophyly and relationships of species in Opius. Each group should be viewed as a hypothesis supported by potentially phy- logenetically informative characters, and the monophyly of each group will be test- ed in future phylogenetic analyses. The first three characters presented in each di- Volume 12, Number 2, 2003 289 Table 1. Placement of described Nearctic species in Gastrosema into three previously unrecognized spe- cies groups. flatriceps- group oscinidis- group striativentris- group flaviceps relativus salmonensis oscinidis abbyae adductus alconanus Fischer tan gens castaneigaster clevelandeiisis cordobensis deiphobe Fischer liaiicockaiius intermissus lacopitaensis niobe pallas porteri Fischer striativciitris agnosis differentiate each group from the remaining Opius sensu lato. flaviceps-group: midpit absent; clypeus reduced, labrum exposed when mandibles closed; lm-cu distad 2RS; posterior me- sopleural furrow crenulate above epister- nal scrobe; propodeum rugose to areolate- rugose; abdominal tergites posterior to tl smooth or variously sculptured; forewing stigma usually broad and wedge shaped, second submarginal cell size and shape variable but often shaped as in Fig. 29. oscinidis-group: midpit absent; clypeus reduced, labrum exposed when mandibles closed; lm-cu distad 2RS; posterior me- sopleural furrow smooth above episternal scrobe; propodeum smooth to rugose; ab- dominal tergites posterior to tl smooth or longitudinally carinate; forewing stigma usually narrow and elongate, second sub- marginal cell long and usually shaped as in Fig. 30. striativentris-group: midpit absent; clyp- eus reduced, labrum exposed when man- dibles closed; lm-cu distad 2RS; posterior mesopleural furrow smooth above epister- nal scrobe; propodeum smooth to rugose; at least one abdominal tergite posterior to tl coriaceous; forewing stigma usually Figs. 30-31. Shape of forewing stigma and second submarginal cell. 30, oscinidis-group. 31, striativentris- group. narrow and elongate, second submarginal cell short and usually shaped as in Fig. 31. Discussion. — The posterior mesopleural furrow in O. walleyi is smooth above the episternal scrobe, and thus, O. walleyi does not fall within the flaviceps-group. The shape of the second submarginal cell in O. walleyi fits the diagnosis for the oscinidis- group. However, t2 and t3 are coriaceous which fits the diagnosis for the striativen- tris-group. Thus, Opius walleyi possesses characteristics of both the oscinid /s-group and the striativentris-group and is left un- placed at this time. Opius pumilio (the type species for Gas- trosema) belongs in the striativentris-group. Additionally, several Nearctic species treated in Fischer (1977) were not included in Gastrosema but belong in one of the three species groups (Table 2). Thus, if the species groups proposed in this study are monophyletic, subgenera with species that fall within the species groups are obvious- ly not monophyletic. In future studies of Opius, species in Al- lophlebus Fischer, Gastrosema, Hypocynodus Forster sensu Fischer (1972), Merotrachys, Opiothorax Fischer, Peudopius Fischer, Phaedrotoma sensu Fischer (1972), Pleurose- 290 Journal of Hymenoptera Research Table 2. Placement of described Nearctic species not included in Gastrosema into three previously unrec- ognized species groups. flaviceps-group O. (Pleurosema) euwattacooanus Fischer hermosanus Fischer metatensis Fischer orizabensis Fischer paratakomanus Fischer pilosinotum Fischer sybile Fischer thalis Fischer O. (Rhogadopis) northcarolinensis Fischer oscinidis-gioxip O. (Rhogadopis) stenopectus Fischer O. (Thoracosema ) crabtreeanus Fischer extiratus parkercrcekensis Fischer prolongatus Fischer pseudocolumbiacus Fischer schuleri Fischer southcarolinensis Fischer striativentris-group O. (Merotrachys) brownsvillensis Fischer lansingensis Fischer paulior Fischer paulus Fischer virentis Fischer O. (Phaedrotoma) complicans Fischer O. (Tolbia) heroicus Fischer ma, Rhogadopisis sensu Wharton (1987), Thoracosema, and Tolbia sensu Fischer (1972) should be examined to determine which species group, if any, they fit into. Examination of species in the aforemen- tioned subgenera will almost certainly re- sult in the discovery of species groups in addition to the three discovered during this research. Placement of all described species in these subgenera will be a mon- umental task, as several hundred holo- types must be examined. The holotypes are housed in repositories found through- out the world, although most are in North American and European museums and collections. Species currently included in subgenera not listed above lack at least one of the three characters used to differ- entiate the three species groups from the remaining Opius sensu lato (see species group diagnoses). Opius (Gastrosema) abbyae Kula, new species Fig. 32 Female. — Head: 1. 23-1. 36 X as wide as long; eyes 0.67-0.77X as wide as long, gla- brous; face smooth with rugulose sculp- ture between antennal sockets and com- pound eyes, slightly raised mesally, seti- ferous; clypeus hemielliptical, convex, ventral margin straight, setiferous; man- dibles simple, with two apical teeth per mandible, setiferous; antennae with 19-21 flagellomeres. Mesosoma: 0.30-0.40 mm wide; 0.51-0.64 mm long; 0.58-0.66 X as Figs. 32-33. Opius spp. 32, O. abbyae. 33, O. lacopitaensis. Volume 12, Number 2, 2003 291 wide as long; 0.71-0.81 X as wide as deep; 1.10-1.31 X as long as deep; pronotum co- riaceous laterally, crenulate in anterior furrow; notauli represented by anterior depressions; mesoscutum smooth, setifer- ous (78.6%) or sparsely setiferous anteri- orly and along lines where notauli would run if present (21.4%); mesopleuron nearly entirely smooth, usually rugulose or cren- ulate near tegulae (85.8%), rarely smooth near tegulae (14.2%), sternaulus distinctly crenulate; scutellar disc smooth; propo- deum rugose, setiferous; metapleuron se- tiferous. Metasoma: tl 0.93-1.25 X as wide as long, rugulose to rugose; t2 coriaceous; t3 coriaceous; t4 always at least partially coriaceous; remaining tergites smooth; ex- posed ovipositor 0.11-0.26 mm long. Legs: smooth, setiferous, hind tibia 0.95-1. 09 X as long as mesosoma length. Wings: hya- line; forewing stigma narrow and elon- gate; 2RS 0.53-0.71 X as long as 3RSa; 2RS 0.41-0.50X as long as 2M; 3RSa 0.70-0.77X as long as 2M. Color: head entirely very dark brown to black; mesosoma usually black (85.7%), rarely very dark brown (14.3%); tl entirely dark brown to nearly entirely dark brown with slight brownish orange coloration laterally; t2 light brown to dark brown medially, orange to yellow laterally; t3 usually brown to dark brown with orange to yellow coloration at an- terolateral edges (85.7%), rarely entirely dark brown (14.3%); t4 usually entirely dark brown (92.9%), rarely brownish yel- low (7.1%); remaining tergites dark brown; legs yellow to yellowish brown, prothoracic legs lighter than mesothoracic legs, mesothoracic legs lighter than meta- thoracic legs, metatibia and metatarsus al- ways yellowish brown. Male.— As in 9 except: Head: 1.21-1. 30 X as wide as long; eyes 0.64-0.71 X as wide as long; antennae with 20-22 flagello- meres. Mesosoma: 0.57-0.61 X as wide as long; 1.24-1. 32 X as long as deep. Wings: 2RS 0.50-0.53 X as long as 2M. Color: me- sosoma always black. Host. — Calycomyza sp. mining the leaves of Helianthus annuus (TAMU). Material examined. — Holotype 9: USA, Nebraska, Arthur County, Highway 61 south of Arthur near Arapaho Prairie, 8.VU.1999, R. R. Kula and A. A. Rogers (TAMU). Paratypes: 69, same data as ho- lotype; 29, Id, same data as holotype ex- cept ll.vii.1998; 59, 3d, same data as ho- lotype except 10.vii.1999 (TAMU). Diagnosis. — Within the striativentris- group O. intermissus, O. lacopitaensis, and O. weemsi Fischer are morphologically similar to O. abbyae. Head coloration can be used to distinguish abbyae from both in- termissus and weemsi. In abbyae the head is entirely very dark brown to black. Alter- natively, in intermissus the head is yellow with the vertex very dark brown to black, and in weemsi the head is mostly orange with the vertex very dark brown and the ventral half of the face brown. The num- ber of antennal flagellomeres can also be used to distinguish abbyae from intermissus and weemsi. Opius abbyae possesses 19-22 flagellomeres, while intermissus possesses 26-29 flagellomeres and weemsi possesses 32 flagellomeres. Another character that can be used to differentiate abbyae from in- termissus and weemsi is tergite coloration. Tergite 3 in abbyae is usually brown to dark brown with orange to yellow color- ation at the antero-lateral edges. In inter- missus t3 is yellow with the posterior edge dark brown, and in weemsi t3 is orangish yellow with the posterior edge dark brown. Sculpture on the abdominal ter- gites can be used to distinguish abbyae from lacopitaensis. In abbyae t4 is always co- riaceous, while in lacopitaensis t4 is never coriaceous. Discussion. — Opius abbyae falls within the striativentris-group. This species exhib- its less intraspecific variation in qualitative characters than observed for intermissus, lacopitaensis, oscinidis, and striativentris. For example, the sternaulus is consistently crenulate in abbyae, while the sternaulus varies from a smooth depression to cren- 292 Journal of Hymenoptera Research ulate in lacopitaensis and striativentris. Col- Wings: hyaline; forewing stigma narrow oration of the head and mesosoma is and elongate; 2RS 0.56-0.64X as long as much less variable in abbyae than in oscin- 3RSa; 2RS 0.41-0.48X as long as 2M; 3RSa idis or striativentris. No substantial differ- 0.67-0.77X as long as 2M. Color: head en- ences were observed between female and tirely dark brown to very dark brown; me- male specimens in abbyae. sosoma entirely dark brown to very dark Etymology. — This species is named in brown; tl dark brown; t2 brownish orange honor of Abigail Rogers Kula who assist- to yellow; t3 anterior half to two thirds ed in collection of the host. brownish orange to yellow, posterior half to one third brown; remaining tergites Opins (Gastrosema) lacopitaensis Kula, brown to dark brown; legs yellow, poste- new species rjor p0rnon of metatibia and entire meta- Fig- 33 tarsus yellowish brown. Female.— Head: 1.25-1. 36 X as wide as Male.— As in 9 except: Head: eyes 0.67- long; eyes 0.67-0.73 X as wide as long, gla- 0.79 X as wide as long; face usually brous; face usually smooth with rugulose smooth with rugulose sculpture between sculpture between antennal sockets and antennal sockets and compound eyes compound eyes (92.9%), rarely minutely (90.9%), rarely entirely smooth (9.1%); an- coriaceous with rugulose sculpture be- tennae with 19-22 flagellomeres. Mesoso- tween antennal sockets and compound ma: 0.54-0.67 mm long; 0.68-0.73 X as eyes (7.1%), slightly raised mesally, setifer- wide as deep; 1.17-1.38X as long as deep; ous; clypeus narrowly hemielliptical to pronotum often coriaceous laterally crescent shaped, convex, ventral margin (63.6%), occasionally smooth (36.4%); me- straight to slightly arched, setiferous; man- sopleuron nearly entirely smooth, usually dibles simple to gradually widened distally smooth near tegulae (90.9%), rarely cren- to proximally, with two apical teeth per ulate near tegulae (9.1%), sternaulus often mandible, setiferous; antennae with 19-20 a smooth depression (54.6%), occasionally flagellomeres. Mesosoma: 0.29-0.35 mm crenulate (45.5%). Wings: 2RS 0.56-0.67X wide; 0.51-0.61 mm long; 0.55-0.64X as as long as 3RSa. wide as long; 0.69-0.75 X as wide as deep; Host. — Unknown. 1.14-1.27X as long as deep; pronotum of- Material examined. — Holotype 9: USA, ten entirely coriaceous laterally (78.6%), oc- Texas, Jim Wells County, 8 miles west of casionally smooth dorsally and coriaceous Ben Bolt, La Copita Research Station, ventrally (21.4%), crenulate in anterior fur- 20.V.1987, J. B. Woolley (TAMU). Para- row; notauli represented by anterior de- types: 11$, 11<3, same data as holotype; pressions; mesoscutum smooth, setiferous 2$, same data as holotype except north- anteriorly and along lines where notauli west of Ben Bolt, 21. v. 1987, G. Zolnerow- would run if present; mesopleuron nearly ich (TAMU). entirely smooth, often smooth near tegulae Diagnosis. — Within the striativentris- (78.6%), occasionally crenulate near tegulae group O. abbyae, O. castaneigaster, O. cor- (21.4%), sternaulus often crenulate (78.6%), dobensis Fischer, and O. pallas Fischer are occasionally a smooth depression (21.4%); morphologically similar to O. lacopitaensis. scutellar disc smooth; propodeum rugose, Characters used to differentiate lacopitaen- setiferous; metapleuron setiferous. Metaso- sis from abbyae and castaneigaster can be ma: tl 0.71-1.00X as wide as long, rugose; found in the diagnoses for abbyae and cas- t2 coriaceous; t3 coriaceous; remaining ter- taneigaster. Optus lacopitaensis can be dif- gites smooth; exposed ovipositor 0.16-0.32 ferentiated from cordobensis using propo- mm long. Legs: smooth, setiferous, hind tib- deal sculpture. In lacopitaensis the propo- ia 0.88-0.97X as long as mesosoma length, deum is rugulose to rugose, while in cor- Volume 12, Number 2, 2003 293 dobensis the propodeum is very heavily rugose with conspicuous carinae through- out. The number of antennal flagellomeres can be used to distinguish lacopitaensis from pallas. Optus lacopitaensis possesses 19-22 flagellomeres, while pallas possesses 28 flagellomeres. Discussion. — O. lacopitaensis falls within the striativentris-group. In Fischer's sub- generic classification individuals with a crenulate sternaulus fall within Gastrose- ma, but individuals with a smooth ster- naulus fall within Merotrachys. Opius lacop- itaensis is provisionally placed in Gastro- sema because the majority of individuals (i.e. 16 of 25) fall within Gastrosema. Opius lacopitaensis exhibits variation in color and sculpture similar to that ob- served in castaneigaster. No substantial dif- ferences were observed between female and male specimens in lacopitaensis. No bi- ological data are associated with lacopi- taensis, but other species in the striativen- t r/s-group have been reared from leaf min- ing Agromyzidae (e.g. O. abbyae, O. adduc- tus Fischer, O. striativentris). Etymology. — This species is named after the collection site. Opius (Gastrosema) castaneigaster Fischer Opius castaneigaster Fischer 1964: 413, 419. Opius (Gastrosema) castaneigaster: Fischer 1977: 311, 324. Female. — Head: 1.19-1. 33 X as wide as long; eyes 0.62-0.75 X as wide as long, gla- brous; face usually smooth with rugulose sculpture between antennal sockets and compound eyes (91.7%), rarely entirely smooth (8.3%), slightly raised mesally, se- tiferous; clypeus hemispherical, convex, ventral margin straight, setiferous; man- dibles simple, with two apical teeth per mandible, setiferous; antennae with 21-24 flagellomeres. Mesosoma: 0.30-0.45 mm wide; 0.54-0.82 mm long; 0.53-0.64 X as wide as long; 0.79-0.93 X as wide as deep; 1.29-1. 70 X as long as deep; pronotum co- riaceous laterally, crenulate in anterior furrow; notauli represented by anterior depressions; mesoscutum smooth, setifer- ous anteriorly and along lines where no- tauli would run if present; mesopleuron nearly entirely smooth, usually rugulose or crenulate near tegulae (91.7%), rarely smooth near tegulae (8.3%), sternaulus usually crenulate (91.7%), rarely reduced to a sculptured depression (8.3%); scutel- lar disc smooth; propodeum rugulose to rugose, setiferous; metapleuron sparsely setiferous to setiferous. Metasoma: tl 1.00- 1.33 X as wide as long, rugulose to rugose; t2 usually coriaceous (91.7%), rarely smooth (8.3%); t3 usually at least partially coriaceous (91.7%), rarely entirely smooth (8.3%); remaining tergites smooth; ex- posed ovipositor 0.24-0.49 mm long. Legs: smooth, setiferous, hind tibia 0.83-0.95 X as long as mesosoma length. Wings: hya- line; forewing stigma narrow and elon- gate; 2RS 0.48-0.67X as long as 3RSa; 2RS 0.39-0.50X as long as 2M; 3RSa 0.73-0.84X as long as 2M. Color: head entirely brown to dark brown; mesosoma entirely brown to dark brown; tl brown to dark brown; t2 orange to yellow; t3 anterior half orange to yellow, posterior half brown to dark brown or t3 anterior two thirds orange to yellow, posterior one third brown to dark brown; remaining tergites dark brown; legs orangish yellow to yellow, metatibia and metatarsus usually yellowish brown (83.3%). Male.— As in 9 except: Head: 1.1 1-1. 26 X as wide as long. Mesosoma: 0.51-0.60X as wide as long. Metasoma: tl 0.86-1. 17X as wide as long. Host. — Unknown. Material examined. — Holotype 9: USA, New York, Tompkins County, Ithaca, 7.ix.l935, H. K. Townes (AEI). Allotype 6: same data as holotype except l.vi.1935 (AEI). Paratypes: all USA; 19, same data as holotype except 30. v. 1934; 16, New York, Oneida County, Rome, 24.vi.1934, H. K. Townes; 19, New York, Delaware County, Hancock, 31.vii.1935, H. K. 294 Journal of Hymenoptera Research Townes; 1 9 , 1 6, New York, Otsego Coun- ty, Oneonta, 17.viii.1935, H. K. Townes; 19, New York, Otsego County, Oneonta swamp, 1,900 feet elevation, 18.viii.1935, H. K. Townes; 1 9 , Ohio, Summit County, Akron, 18.V.1941, H. K. Townes; 1 9, South Carolina, Greenville County, Greenville, 22.iv.1952, G. and L. Townes; 19, South Carolina, Greenville County, Greenville, 31.V.1952, G. and L. Townes; 19, South Carolina, Pickens County, Wattacoo, lO.v.1959, G. F. Townes; 16, South Caro- lina, Pickens County, Wattacoo, 6. v. 1961, G. F. Townes; 1 9 , South Carolina, Pickens County, Wattacoo, 19. v. 1961, G. F. Townes (AEI); 1 9 , Wisconsin, Polk Coun- ty, July, collection C. F. Baker (USNM). Other determined material: 1 9 , same data as holotype (det. Fischer) (AEI). Diagnosis. — Within the striativentris- group, O. cordobensis, O. lacopitaensis, O. niobe, O. pallas, and O. striativentris are morphologically similar to O. castaneigas- ter. Propodeal sculpture can be used to distinguish castaneigaster from cordobensis. In castaneigaster the propodeum is rugu- lose to rugose, while in cordobensis the pro- podeum is very heavily rugose with con- spicuous carinae throughout. Opius casta- neigaster can be differentiated from niobe on the basis of mesosoma shape and the number of antennal flagellomeres. In cas- taneigaster the mesosoma is subelliptical (1.29-1. 70 X as long as deep), and the number of flagellomeres ranges from 21- 24. In niobe the mesosoma is subspherical (1.21-1.23 X as long as deep), and the number of flagellomeres ranges from 25- 31. Opius pallets is very morphologically similar to castaneigaster, but the two spe- cies can be differentiated using flagello- mere number. Opius pallas possesses 28 flagellomeres, while the maximum num- ber of flagellomeres for castaneigaster is 24. Several characters can be used to distin- guish castaneigaster from striativentris. Ter- gite 1 is brown to dark brown in castanei- gaster, while tl is usually yellow to brown- ish orange in striativentris. Tergite 3 is only partially coriaceous in castaneigaster, as op- posed to entirely coriaceous in striativen- tris. Tergite 4 is always smooth in casta- neigaster, while t4 is smooth or coriaceous in striativentris. The mesosoma is uniform- ly brown or dark brown in castaneigaster, while the mesonotum is often conspicu- ously darker than the rest of the mesoso- ma in striativentris. Opiits castaneigaster can be differentiated from lacopitaensis using tl width to length ratios, clypeus shape, and the number of flagellomeres. Tergite 1 width in castaneigaster is usually greater than or equal to tl length (91.7%) and is rarely less than tl length (8.3%). Tergite 1 width in lacopitaensis is always less than or equal to tl length. The shape of the clyp- eus is hemispherical in castaneigaster, while the shape of the clypeus is narrowly hemielliptical to crescent shaped in lacop- itaensis. Opiits castaneigaster possesses 21- 24 flagellomeres, as opposed 19-22 in la- copitaensis. Discussion. — Opius castaneigaster falls within the striativentris-group. Opius pmllas may eventually be determined as a syno- nym of castaneigaster. The two species are very morphologically similar and exhibit overlapping geographic distributions. Op- ius pallas is only known from the holotype, and acquisition of additional specimens may reveal a broader range of flagello- mere number. One specimen of castaneigaster exam- ined in this study lacks sculpture on t2 and t3. However, no additional attributes were observed that indicated the speci- men was not castaneigaster. Single speci- mens with reduced sculpture are very dif- ficult to identify, and comparison of these individuals with a long series of deter- mined specimens will facilitate their iden- tification. No biological data are associat- ed with castaneigaster, but other species in the striativentris-group have been reared from leaf mining Agromyzidae (e.g. O. ab- byae, O. adductus, O. striativentris). Volume 12, Number 2, 2003 295 Opius (Gastrosema) intermissus Fischer Opius intermissus Fischer 1964: 413, 428. Opius {Gastrosema) intermissus: Fischer 1977: 314, 357. Female. — Head: 1.23-1.29 X as wide as long; eyes 0.62-0.67X as wide as long, gla- brous; face minutely coriaceous with ru- gulose sculpture between antennal sockets and compound eyes, slightly raised mes- ally, usually setose (80.0%), rarely setifer- ous (20.0%); clypeus hemispherical, con- vex, ventral margin straight, usually se- tose (80.0%), rarely setiferous (20.0%); mandibles simple, with two apical teeth per mandible, usually setose (80.0%), rare- ly setiferous (20.0%); antennae with 27-29 flagellomeres. Mesosoma: 0.38-0.46 mm wide; 0.64-0.75 mm long; 0.59-0.63 X as wide as long; 0.71-0.74 X as wide as deep; 1.14-1.21 X as long as deep; pronotum co- riaceous laterally, crenulate in anterior furrow; notauli represented by anterior depressions; mesoscutum smooth, setifer- ous anteriorly and along lines where no- tauli would run if present; mesopleuron nearly entirely smooth, rugulose to cren- ulate near tegulae, sternaulus distinctly crenulate; scutellar disc smooth; propo- deum rugose, setiferous; metapleuron usually setiferous (80.0%), rarely setose (20.0%). Metasoma: tl 0.70-1. 20 X as wide as long, longitudinally rugulose with two prominent dorsal carinae converging pos- teriorly; t2 coriaceous; t3 coriaceous; re- maining tergites smooth; exposed ovipos- itor 0.18-0.32 mm long. Legs: smooth, se- tiferous, hind tibia 0.96-1. 02 X as long as mesosoma length. Wings: hyaline; fore- wing stigma narrow and elongate; 2RS 0.52-0.62 X as long as 3RSa; 2RS 0.41- 0.47X as long as 2M; 3RSa 0.75-0.82X as long as 2M. Color: head mostly yellow, vertex very dark brown to black; meso- soma very dark brown to black, meso- pleuron occasionally yellowish orange ventrally and near tegulae; tl brown to dark brown medially, yellow laterally; t2 yellow; t3 yellow with posterior edge dark brown; t4 dark brown medially, yellow laterally, and posterior edge dark brown (60.0%) or t4 mostly yellow with posterior edge dark brown (40.0%); remaining ter- gites dark brown medially, yellow later- ally, and posterior edge dark brown; legs orangish yellow to yellow, posterior por- tion of metatibia and entire metatarsus yellowish brown. Male.— As in 9 except: Head: 1.19-1.27X as wide as long; eyes 0.63-0.71 X as wide as long; antennae with 26-27 flagello- meres. Mesosoma: 0.35-0.42 mm wide; 0.59-0.70 mm long; 0.57-0.65 X as wide as long; 0.67-0.77X as wide as deep; 1.12- 1.30X as long as deep; mesoscutum usu- ally smooth (80.0%), occasionally minutely coriaceous (20.0%). Legs: hind tibia 0.90- 1.03 X as long as mesosoma length. Wings: 2RS 0.42-0.48 X as long as 2M. Color: me- sosoma usually very dark brown to black (80.0%), rarely dark reddish brown (20.0%); tl entirely dark brown (40.0%) or tl dark brown medially and yellow later- ally (40.0%) or tl reddish brown medially and yellow laterally (20.0%); t4 usually dark brown (80.0%), rarely brown medi- ally, yellow laterally, and posterior edge dark brown (20.0%); remaining tergites dark brown. Host. — Unknown. Material examined. — Holotype 9: USA, South Carolina, Greenville County, Green- ville, 18.ix.1955, G. and L. Townes (AEI). Allotype 6 : same data as holotype except 6.vii.l952 (AEI). Other determined mate- rial: all USA, North Carolina, Transylva- nia County, Pisgah National Forest except 19, South Carolina, Greenville County, Paris Mountain State Park, 16.-21. v. 1999, R. R. Kula, Malaise trap; 19, Coontree Creek Trail, 17.-21 .v. 1999, R. R. Kula, Mal- aise trap; 19, Pink Beds, 17.V.1999, R. R. Kula, sweep net; 19, 36, Coontree Creek Trail, 18.V.1999, R. R. Kula, sweep net; 16, Pink Beds, 19.V.1999, R. R. Kula, sweep net (det. Kula) (TAMU). Diagnosis. — Within the striativentris- group O. clevelandensis Fischer, O. niobe, 296 Journal of Hymenoptera Research and O. weemsi are morphologically similar to O. intermissus. Head coloration can be used to distinguish intermissus from all three species. The head is yellow with a very dark brown to black vertex in inter- missus. The head is yellowish brown with a brown vertex in clevelandensis, entirely brown to yellowish brown in uiobc, and mostly orange with a very dark brown vertex and a partially brown face in weem- si. Mesosomal coloration can also be used to distinguish intermissus from clevelanden- sis and niobe. In intermissus the mesosoma is usually very dark brown to black, while in clevelandensis and niobe the mesosoma is brown to dark brown. Additionally, inter- missus can be distinguished from clevelan- densis and weemsi using exposed oviposi- tor length and the number of antennal fla- gellomeres, respectively. In intermissus the ovipositor is relatively short (exposed ovi- positor 0.18-0.32 mm long) and barely ex- serted from the abdomen, while in cleve- landensis the ovipositor is long (exposed ovipositor 1.12 mm long) and conspicu- ously exserted from the abdomen. The number of flagellomeres in intermissus is 26-29, as opposed to 32 in weemsi. Discussio)i. — Opius intermissus falls with- in the striativentris-group. Dark brown me- sosomal coloration was observed in only one male specimen, and all other speci- mens have a very dark brown to black me- sosoma. Distinct color differences were observed between female and male spec- imens. In females t4 is dark brown medi- ally and yellow laterally or mostly yellow with the posterior edge dark brown. In males t4 is usually entirely dark brown to black. Yellow coloration on t4 was ob- served in only one male specimen. No bi- ological data are associated with intermis- sus, but other species in the striativentris- group have been reared from leaf mining Agromyzidae (e.g. O. abbyae, O. adductus, O. striativentris). Opius (Gastrosema) oscinidis (Ashmead) Rhi/ssalus oscinidis Ashmead 1889 (1888): 630. Eutrichopsis oscinidis: Viereck 1913: 559. Opius oscinidis: Gahan 1915: 72, 90. Opius (Aulonotus) oscinidis: Fischer 1977: 78, 87. Opius pusilloides Fischer 1964: 413, 433. Syno- nym Marsh (1974). Female. — Head: 1.21-1. 35 X as wide as long; eyes 0.65-0.72 X as wide as long, gla- brous; face entirely smooth to entirely ru- gulose, slightly raised mesally, usually se- tose (81.8%), rarely setiferous (18.2%); clypeus flattened to convex, ventral mar- gin lobed mesally, usually setose (81.8%), rarely setiferous (18.2%); mandibles sim- ple, with two apical teeth per mandible, setiferous; antennae with 21-22 flagello- meres. Mesosoma: 0.38-0.43 mm wide; 0.66-0.75 mm long; 0.53-0.60 X as wide as long; 0.73-0.79 X as wide as deep; 1.34- 1.42X as long as deep; pronotum smooth laterally, crenulate in anterior furrow; no- tauli represented by anterior depressions; mesoscutum smooth, setiferous anteriorly and along lines where notauli would run if present; mesopleuron nearly entirely smooth, slightly to distinctly crenulate near tegulae, sternaulus distinctly crenu- late; scutellar disc smooth; propodeum ru- gose, especially medially, setose; meta- pleuron setose. Metasoma: tl 0.69-1. 00 X as wide as long, longitudinally rugulose with two prominent dorsal carinae delimiting a raised median area; t2 often longitudinally carinate at posterior edge (63.6%), occa- sionally longitudinally carinate medially (36.4%); t3 often smooth (63.6%), occasion- ally longitudinally carinate at anterior edge (36.4%); remaining tergites smooth; exposed ovipositor 0.16-0.46 mm long. Legs: smooth, setiferous, hind tibia 0.91- 1.00X as long as mesosoma length. Wings: hyaline; forewing stigma narrow and elongate; 2RS 0.42-0.59 X as long as 3RSa; 2RS 0.33-0.43 X as long as 2M; 3RSa 0.73- 0.80 X as long as 2M. Color: face and clyp- eus brownish yellow, remainder of head brown (54.6%) or head entirely brown (27.3%) or face and clypeus yellow, re- mainder of head brown (18.2%); mesoso- ma brown to dark brown; tl dark brown; Volume 12, Number 2, 2003 297 t2 yellow (54.6%) or yellowish brown (45.5%); t3 often yellowish brown (63.6%), occasionally anterior half yellow, posterior half brown (27.3%), rarely anterior two thirds yellow, posterior one third brown (9.1%); t4 usually brown to dark brown (90.9%), rarely yellowish brown (9.1%); re- maining tergites brown to dark brown; legs orangish yellow to yellow. Male. — As in 9 except: Head: 1.19 X as wide as long. Mesosoma: 0.37 mm wide. Hosts. — An undetermined species of CJilorops Meigen on Plantago major L. (Ash- mead 1889); Phytomyza orobanchia Kalten- bach (Hennig 1953); Ph. plantaginis Robi- neau-Desvoidy (Muesebeck and Walkley 1951). All records need verification. Material examined. — Holotype 9, O. os- cinidis: no locality data (USNM). Holo- type 9, O. pusilloides: USA, Allegany County, Thomas Road near Cumberland, 28. vi. 1953, L. M. Walkley (USNM). Allo- type 6, O. pusilloides: USA, Maryland, Montgomery County, Cabin John, 21.viii.1917, R. M. Fouts (USNM). Para- types O. pusilloides: all USA; 19, New York, Otsego County, Oneonta, 24.viii.1935, H. K. Townes; 1 9 , New York, Otsego Coun- ty, Oneonta, 2.ix.l935, H. K. Townes; 19, New York, Tompkins County, Ithaca, 7.ix.l935, H. K. Townes; 19, New York, Tompkins County, Ithaca, ll.bc.1935, H. K. Townes; 1 9 , North Carolina, Transylvania County, Pink Beds, 22.vii.1952, G. and L. Townes; 1 9 , North Carolina, Henderson County, Flat Rock, 6.ix.l952, G. and L. Townes; 1 9 , South Carolina, Greenville County, Greenville, 24. v. 1952, G. and L. Townes; 1 9 , South Carolina, Greenville County, Greenville, l.ix.1952, L. and G. Townes; 19, South Carolina, Greenville County, Greenville, 7.ix.l952, L. and G. Townes (AEI). Other determined material: 19, South Carolina, Greenville County, Greenville, 1.x. 1954, G. and L. Townes (det. Fischer as pusilloides) (AEI). Diagnosis. — Opius oscinidis is extremely different morphologically from all de- scribed Nearctic species in Gastrosema. In oscinidis t2 is at least partially longitudi- nally carinate, although it is nearly smooth in some individuals. Thus, oscini- dis can be differentiated from all species in the striativentris-group based on tergite sculpture. Opius oscinidis lacks crenula- tions above the episternal scrobe in the posterior mesopleural furrow. Thus, oscin- idis can be differentiated from all species in the flaviceps-group based on posterior mesopleural furrow sculpture. Addition- ally, oscinidis has a narrowly elongate sec- ond submarginal cell, while species in the striativentris-group often have a narrow but relatively short second submarginal cell. The second submarginal cell is shape variable for species in the flaviceps-group, but a narrowly elongate second submar- ginal cell, as exemplified by oscinidis, has not been observed for species in the flavi- ceps-group. Discussion. — Opius oscinidis falls within the oscinidis-group. Marsh (1974) synony- mized O. pusilloides with oscinidis. Com- parison of the holotypes for oscinidis and pusilloides with a long series of determined pusilloides from the AEI and USNM veri- fied the synonymy. Of the 21 described Nearctic species in Gastrosema, O. oscinidis is the only species that falls within the os- cinidis-group. However, several species in other subgenera fit within the oscinidis- group (as noted above). All biological data associated with oscinidis needs verifica- tion. Opius oscinidis is morphologically similar to a species of Opius near extiratus that has been reared from Phytomyza spp. mining the leaves of hollies. Opius (Gastrosema) salmonensis Fischer Opius salmonensis Fischer 1964: 412, 438. Opius (Gastrosema) salmonensis: Fischer 1977: 313, 385. Female. — Head: 1. 36-1. 47 X as wide as long; eyes 0.77-0.88 X as wide as long, gla- brous; face punctate, slightly raised mes- ally, setose; clypeus broadly hemiellipti- cal, usually flattened (80.0%), rarely weak- 298 Journal of Hymenoptera Research ly convex (20.0%), ventral margin straight posterior half dark brown (20.0%) or t3 an- to slightly arched, often setose (60.0%), oc- terior one third orangish yellow, posterior casionally setiferous (40.0%); mandibles two thirds dark brown (20.0%) or t3 en- simple, with two apical teeth per mandi- tirely reddish brown (20.0%) or t3 entirely ble, often setose (60.0%), occasionally se- dark brown (20.0%); remaining tergites tiferous (40.0%); antennae with 21-28 fla- dark brown; legs yellow, metatibia and gellomeres. Mesosoma: 0.61-0.86 mm wide; metatarsus usually yellow (80.0%), rarely 1.02-1.36 mm long; 0.57-0.64 X as wide as yellowish brown (20.0%). long; 0.71-0.79 X as wide as deep; 1.25- Male. — As in 9 except: Head: face punc- 1.31 X as long as deep; pronotum smooth tate (50.0%) or rugulose (50.0%); clypeus to rugose laterally, rugose to crenulate in flattened, setiferous (50.0%) or setose anterior and posterior furrows; notauli (50.0%); mandibles setose; antennae with represented by deep anterior depressions; 23-30 flagellomeres. Mesosoma: 0.56-0.67 mesoscutum often smooth (60.0%), occa- mm wide; 0.55-0.59 X as wide as long; sionally minutely coriaceous (40.0%), usu- 0.70-0.77X as wide as deep; pronotum of- ally setose anteriorly and along lines ten minutely coriaceous laterally (75.0%), where notauli would run if present occasionally smooth laterally (25.0%); me- (80.0%), rarely entirely setose (20.0%); me- soscutum often minutely coriaceous sopleuron minutely coriaceous, deeply (75.0%), occasionally smooth (25.0%), se- crenulate near tegulae, sternaulus broadly tose anteriorly and along lines where no- crenulate to broadly lacunose; scutellar tauli wouM mn if present; scutenar disc disc often smooth (60.0%), occasionally often minutel coriaceous (75.0%), occa- minutely coriaceous (40.0%); propodeum sionall smooth (25M). metapleuron of- areolate-rueose, setiferous; metapleuron . . Inr- no/ \ n L-c r ,/^ on/ x 11 r ten setose (75.0%), occasionally setiferous often setose (60.0%), occasionally setifer- nrno/\ a* 1 t1 n ™ 1 nn^, -j ,,nnnn \, , ' „ _ _. { „., (25.0%). Metasoma: tl 0.79-1.00X as wide ous (40.0%). Metasoma: tl 0.84-1. llx as , ,cnno/, , ,L ,. n ., , n /onno/N as long, rugose (50.0%), longitudinally ru- wide as long, usually costate (80.0%), rare- ,^_ nn,. /nr. °n, . „ J 1 nn no/ ^ ^ , , \o u 1 8ose (25.0%), or costate 25.0%); t2 costate ly rugose (20.0%); t2 costate; t3 often Ion- ?__ __, , . ,nr-\n,^ .. .6 ,, • . • . . u ,r 4 50.0% , coriaceous (25.0% , or smooth gitudinally carmate in anterior half and . ' smooth in posterior half (60.0%), rarely en- Wlth antenor ed§e longitudinally carinate tirely minutely coriaceous (20.0%) or en- (25-0%); t3 smooth (50-0%)' minutelY en- tirely smooth (20.0%); remaining tergites naceous (25.0%), or minutely coriaceous smooth; exposed ovipositor 0.64-1.52 mm Wlth antenor edge longitudinally carinate long. Legs: smooth, setiferous, hind tibia (250%)' remaining tergites smooth. Wings: 0.73-0.84 X as long as mesosoma length. 2RS 0-61-0.75 X as long as 3RSa; 2RS 0.43- Wings: hyaline; forewing stigma broad 0-50x as lon§ as 2M' 3RSa 0.63-0.70 X as and wedge shaped; 2RS 0.57-0.67X as lon§ as 2M- Color: head mostly yellow, long as 3RSa; 2RS 0.42-0.45 X as long as vertex dark brown; mesosoma very dark 2M; 3RSa 0.67-0.75 X as long as 2M. Color, brown; tl dark brown; t2 yellow (50.0%), head mostly yellow (80.0%) or orangish orangish yellow (25.0%), or brownish yel- yellow (20.0%), vertex dark brown, at least l°w (25.0%); t3 anterior one third brown- around ocelli; mesosoma usually very ish yellow, posterior two thirds dark dark brown (80.0%), rarely reddish brown brown (25.0%) or t3 anterior half orangish (20.0%); tl usually dark brown (80.0%), yellow, posterior half dark brown (25.0%) rarely reddish brown (20.0%); t2 usually or t3 anterior half yellow, posterior half yellow (40.0%) or orangish yellow (40.0%), dark brown (25.0%) or t3 anterior two rarely reddish brown (20.0%); t3 anterior thirds yellow, posterior one third dark half yellow, posterior half dark brown brown (25.0%); remaining tergites dark £0.0%) or t3 anterior half orangish yellow, brown; legs yellow, metatarsus often yel- Volume 12, Number 2, 2003 299 low (75.0%), occasionally yellowish brown (25.0%). Host. — Unknown. Material examined. — Holotype 9: CAN- ADA, British Columbia, Salmon Arm, shore of Shuswap Lake, 13.vii.1949, H. B. Leech (CAS). Other determined material: 26, CANADA, Province Quebec, 50°03'N 77°07'W, 12.vi.-8.viii.1987, Leblanc (AEI); 1 9 USA, New Hampshire, Grafton Coun- ty, Mount Cardigan (USNM); 19, USA, North Carolina, Macon County, High- lands, 22.vi.1977, H. and M. Townes; 19, USA, North Carolina, Macon County, Highlands, 26.vi.1977, H. and M. Townes; 1 9 , USA, Wisconsin, Grant County, T6N R6W SI 7, 29-v.-3.vi.1975, gypsy moth Malaise trap; 16, USA, Wisconsin, Fond du Lac County, T13N R19E S23, 11.- 18.vi.1975, gypsy moth Malaise trap; 16, USA, Wisconsin, Jackson County, T21N R4W S27, 1 6.-23. vi. 1975, gypsy moth Mal- aise trap (det. Kula) (AEI). Diagnosis. — Within the flaviceps-group O. flaviceps is morphologically similar to O. salmonensis. Opius salmonensis is con- spicuously smaller than flaviceps. In sal- monensis the mesosoma is 1.02-1.36 mm long, while in flaviceps the mesosoma is 1.84 mm long. In salmonensis the mesoso- ma is 0.56-0.86 mm wide, while in flavi- ceps the mesosoma is 0.98 mm wide. Ad- ditionally, exposed ovipositor length in salmonensis is 0.64-1.52 mm, as opposed to 1.78 mm in flaviceps. Discussion. — Opius sahnonensis falls within the flaviceps-group. One specimen of salmonensis in the USNM and five spec- imens of salmonensis in the AEI were mis- identified as O. flaviceps. All specimens differ from the holotype of flaviceps in size and exposed ovipositor length (as detailed in the diagnosis for salmonensis). Size differences were observed between female and male specimens of salmonensis. In females the mesosoma is 0.61-0.86 mm wide, while in males the mesosoma is 0.56-0.67 mm wide. In females the maxi- mum mesosoma length is 1.36 mm, and in males the maximum mesosoma length is 1.15 mm. Differences in sculpture on t2 were also observed between the sexes. In females t2 is always costate, while in males t2 may be costate, coriaceous, or smooth with the anterior edge longitudi- nally carinate. Sculpture on t2 is generally reduced in males relative to females. Unfortunately, no biological data are as- sociated with salmonensis. However, the ovipositor of salmonensis is long relative to species that have been reared from leaf mining Agromyzidae (e.g. O. abln/ae, O. adductus, O. striativentris). This suggests that salmonensis may attack a non-leaf miner host in a concealed situation. Opius (Gastrosema) striativentris Gahan Opius striativentris Gahan 1915: 72, 89. Opius (Gastrosema) striativentris: Fischer 1977: 312, 393. Opius gracillariae Gahan 1915: 72, 90. New syn- onym. Female. — Head: 1.24-1.48 X as wide as long; eyes 0.63-0.79 X as wide as long, gla- brous; face usually smooth with rugulose sculpture between antennal sockets and compound eyes (86.4%), rarely entirely smooth (13.6%), slightly raised (60.9%) or flattened (39.1%) mesally, setiferous; clyp- eus hemispherical to narrowly hemiellip- tical, convex, ventral margin straight to slightly arched, setiferous; mandibles sim- ple, with two apical teeth per mandible, setiferous; antennae with 18-22 flagello- meres. Mesosoma: 0.35-0.50 mm wide; 0.56-0.82 mm long; 0.56-0.65 X as wide as long; 0.78-1.00 X as wide as deep; 1.31- 1.61 X as long as deep; pronotum usually coriaceous laterally (92.1%), rarely entirely smooth (4.6%) or dorsal half smooth ven- tral half coriaceous (3.4%), usually crenu- late in anterior furrow (96.6%), rarely smooth (3.4%); notauli represented by an- terior depressions; mesoscutum usually smooth (94.3%), rarely rugulose medially near transscutual articulation (5.7%), usu- ally setiferous anteriorly and along lines where notauli would run if present 300 Journal of Hymenoptera Research (86.4%), rarely setiferous anteriorly only (9.1%) or glabrous (4.6%); mesopleuron nearly entirely smooth, often smooth near tegulae (55.5%), occasionally rugulose to crenulate near tegulae (44.5%), sternaulus usually crenulate (86.4%), rarely rugulose (10.2%) or reduced to a smooth depression (3.41%); scutellar disc smooth; propodeum rugulose to rugose, usually setiferous (98.9%), rarely glabrous (1.1%); metapleu- ron usually setiferous (98.9%), rarely gla- brous (1.1%). Metasoma: tl 1 .00-1.56 X as wide as long, usually rugulose (87.4%), rarely rugose (11.5%) or smooth (1.2%); t2 coriaceous; t3 coriaceous; t4 often smooth (59.8%), occasionally coriaceous (33.3%); remaining tergites smooth; ovipositor 0.37-0.48 mm long. Legs: smooth, setifer- ous, hind tibia 0.82-0.95 X as long as me- sosoma length. Wings: hyaline; forewing stigma narrow and elongate; 2RS 0.50- 0.69 X as long as 3RSa; 2RS 0.40-0.52 X as long as 2M; 3RSa 0.73-0.87 X as long as 2M. Color: head entirely dark brown to or- angish yellow with vertex dark brown to black; mesonotum often brown to dark brown and conspicuously darker than rest of mesosoma (63.6%), occasionally entirely brownish orange to dark brown (36.4%); tl usually yellow to brownish orange (94.3%), rarely light orangish brown (4.6%) or yellowish brown (1.2%); t2 brownish orange to yellow; t3 brownish orange to yellow; remaining tergites dark brown to orangish yellow; legs yellow, metatibia and metatarsus yellow to yel- lowish brown. Male— As in 9 except: Head: 1.23-1.39 X as wide as long; eyes 0.67-0.80 X as wide as long; face often with rugulose sculpture between antennal sockets and compound eyes (73.3%), occasionally entirely smooth (26.7%). Mesosoma: 0.34-0.43 mm wide; 0.55-0.65 X as wide as long; 0.77-0.90 X as wide as deep; pronotum usually coria- ceous laterally (86.7%), rarely smooth (13.3%), usually crenulate in anterior fur- row (86.7%), rarely smooth (13.3%); me- soscutum usually smooth (96.7%), rarely rugulose medially near transscutual artic- ulation (3.33%), usually setiferous anteri- orly and along lines where notauli would run if present (93.3%), rarely setiferous an- teriorly only (6.7%); mesopleuron nearly entirely smooth, often crenulate near te- gulae (70.0%), occasionally smooth near tegulae (30.0%), sternaulus usually crenu- late (86.7%), rarely reduced to a smooth depression (13.3%). Wings: 2RS 0.55-0.71 X as long as 3RSa. Hosts. — Phytomyza glabricola Kulp min- ing the leaves of Ilex coriacea (Pursh) and 7. glabra (L.); P. ilicicola Loew (as P. ilicis Curtis) mining the leaves of /. opaca So- land. in Ait. (Langford and Cory 1937); P. vomitoriae Kulp mining the leaves of I. vomit or ia Soland. in Ait. Material examined. — Holotype 9, O. striativentris: no locality data (USNM). Ho- lotype 9, O. graeillariae: no locality data (USNM). Other determined material: all USA; 1 9 , Louisiana, Cameron Parish, Route 27, 2.1.1998, S. Scheffer; 19, Missis- sippi, Forrest County, De Soto National Forest, Route 13, 7.L1998, S. Scheffer; 39, North Carolina, Montgomery County, Uwharrie National Forest, i.1996, S. Schef- fer; 149, North Carolina, Moore County, Ebersole Holly Garden, 21. ii. 1996, S. Scheffer; 6 9 , North Carolina, Wake Coun- ty, North Carolina State University Arbo- retum, 23.ii.1996, S. Scheffer; 19, North Carolina, New Hanover County, Carolina Beach State Park, 24.ii.1996, S. Scheffer; 19, North Carolina, iv.1996; 8 9, North Carolina, New Hanover County, Carolina Beach State Park, marina, 21.ii.1997, S. Scheffer; 3 9, North Carolina, New Hano- ver County, Carolina Beach State Park, Flytrap Trail, 26.L1998, S. Scheffer; 26 9, 30 6, South Carolina, Berkeley County, Francis Marion National Forest, Big Ocean Bay, 18.ii.1997, S. Scheffer; 89, South Car- olina, Charleston County, Francis Marion National Forest, Buck Hall, 19.ii.1997, S. Scheffer; 5 9, South Carolina, Berkeley County, Francis Marion National Forest, North Honey Hill Road, 19.ii.1997, S. Volume 12, Number 2, 2003 301 Scheffer; 19, South Carolina, Berkeley County, Francis Marion National Forest, Road 188, 19.ii.1997, S. Scheffer; 7$, South Carolina, Berkeley County, Francis Mari- on National Forest, Big Ocean Bay, 21.U998, S. Scheffer; 29, Tennessee, Shel- by County, Memphis Botanical Garden, 29.viii.1997, S. Scheffer; 19, Texas, Jasper County, Angelina National Forest, Boykin Springs Trail, 2.U998, S. Scheffer (det. Kula as striativentris) (TAMU). Diagnosis. — Within the striativentris- group O. castaneigaster and O. hancockanus are morphologically similar to O. striativ- entris. Characters used to differentiate striativentris from castaneigaster can be found in the diagnosis for castaneigaster. The shape of the mesosoma and tl width to length ratios can be used to distinguish striativentris from hancockanus. In striativ- entris the mesosoma is 1.31-1.61 X as long as deep and similar in shape to castanei- gaster. In hancockanus the mesosoma is 1.26X as long as deep and similar in shape to niobe. Tergite 1 is 1. 00-1. 56 X as wide as long in striativentris, while tl is 0.65 X as wide as long in hancockanus. Discussion. — Opius striativentris falls within the striativentris-group. In Fischer's subgeneric classification individuals with a crenulate sternaulus fall within Gastro- sema, but individuals with a smooth ster- naulus fall within Merotrachys. Opius stria- tiventris is retained in Gastrosema because the majority of individuals (i.e. 91.7%) fall within Gastrosema. Comparison of the holotype of O. gra- cillariae with hundreds of striativentris specimens revealed that gracillariae falls within the morphological limits of striativ- entris as defined in this study. Thus, O. gracillariae Gahan 1915 is a new synonym of O. striativentris Gahan 1915. The holo- type of gracillariae is morphologically sim- ilar to striativentris reared from P. glabri- cola infesting /. coriacea and I. glabra. Opius gracillariae and striativentris were both originally described in Gahan (1915). Op- ius striativentris is designated the senior synonym because the name striativentris has been associated with a greater number of studies and appears more frequently in the literature than gracillariae. Gahan (1915) reported that gracillariae was reared from Porphyrosela desmiodella (Clement). This record is clearly invalid because P. desmiodella is a lepidopteran. Opiinae, as currently defined, is limited to braconids that are endoparasitoids of cyclorrha- phous Diptera. However, Porphyrosela des- miodella feeds on Desmodium Desv., and ]a- panagromyza desmodivora Spencer is an agromyzid that mines the leaves of D. tor- tuosum DC. Thus, Gahan's gracillariae may have actually been reared from an agro- myzid. ACKNOWLEDGMENTS This work was supported by National Science Foundation Partnerships for Enhancing Expertise in Taxonomy (PEET) grant number DEB9712543 award- ed to R. A. Wharton and J. B. Woolley (TAMU). I am especially grateful for the guidance and advice pro- vided by Bob Wharton. My sincere thanks to Jim Woolley for contributing several useful suggestions during the course of this research. Many thanks to Jim Ehrman (Digital Microscopy Facility, Mount Al- lison University) for original SEMs and Greg Zolne- rowich (Kansas State University) for advice on the construction of plates. Special thanks to Sonja Schef- fer (Systematic Entomology Laboratory) for provid- ing several hundred reared specimens of O. striativ- entris. Material supplied by the following curators and staff members made this research possible: David Wahl (AEI), Robert Zuparko (CAS), Henri Goulet (CNCI), Steven Krauth (IRCW), Philip Perkins (MCZ), Stefan Cover (MCZ), David Smith (USNM), and Cathy Anderson (USNM). LITERATURE CITED Ashmead, W. H. 1889 (1888). Descriptions of new Braconidae in the collection of the U. S. National Museum. Proceedings of the United State* National Museum 11: 611-671. Fischer, M. 1964. Die Opiinae der nearktischen Re- gion (Hymenoptera, Braconidae). I. Teil. Polskie Pismo Entomologiczne 34: 197-530. Fischer, M. 1970. Nearktische Opiinae aus der Samm- lung Townes (Hymenoptera, Braconidae). Polskie Pisino Entomologiczne 40: 763-827. Fischer, M. 1972. Hymenoptera: Braconidae (Opiinae I). Das Tierreich 91: 1-620. 302 Journal of Hymenoptera Research Fischer, M. 1977. Hymenoptera: Braconidae (Opiinae II— Amerika). Das Tierreich 97: 1-1001. Fischer, M. 1987. Hymenoptera: Opiinae III — athiop- ische, orientalische, australische und ozeanische Region. Das Tierreich 105: 1-734. Forster, A. 1862. Synopsis der Familien und Gattun- gen der Braconen. Verhandlungen des Naturhisto- rischen Vereines preussischen Rheinlande und West- phalens 19: 225-288. Gahan, A. B. 1915. A revision of the North American ichneumon-flies of the subfamily Opiinae. Pro- ceedings of the United States National Museum 49: 63-95. Godfray, H. C. J. 1984. Intraspecific variation in the leaf-miner parasite Exotela cyclogaster Forster (Hymenoptera: Braconidae). Proceedings and Transactions of the British Entomological and Natu- ral History Society 17: 47-50. Harris, R. A. 1979. A glossary of surface sculpturing. Occasional Papers in Entomology 28: 1-31. Hennig, W. 1953. Diptera, Zweiflugler, p. 1-58, 66- 166. In: Blunck, H. (ed.). Handbuch der Pflanzen- krankheiten, Fiinfte Band, Tierisehe Schadlinge an Nutzpflanzen, 2. Teil, Fiinfte neubearbeitete Auflage, Erste Lieferung. P. Parey, Berlin. Langford, G. S. and E. N. Cory. 1937. The holly leaf miner and its control. Proceedings of the 13th Na- tional Shade Tree Conference: 109-112. Marsh, P. M. 1974. New combinations and new syn- onyms in North American Braconidae (Hyme- noptera). Proceedings of the Entomological Society of Washington 76: 285-289. Muesebeck, C. F. W. and L. M. Walkley. 1951. Family Braconidae, p. 90-184. In: Muesebeck, C. F. W., K. V. Krombein, and H. K. Townes (eds.). Hy- menoptera of America North of Mexico Synoptic Cat- alog. United States Government Printing Office, Washington D.C. Sharkey, M. J. 1997. Key to the New World subfam- ilies of the family Braconidae, p. 39-45. In: Whar- ton, R. A., P. M. Marsh, and M. J. Sharkey (eds.). Manual of the New World Genera of the Family Bra- conidae (Hymenoptera). Special Publication of the International Society of Hymenopterists 1, Wash- ington D.C. Sharkey, M. J. and R. A. Wharton. 1997. Morphology and Terminology, p. 19-37. In: Wharton, R. A., P. M. Marsh, and M. J. Sharkey (eds.). Manual of the Neiv World Genera of the Family Braconidae (Hy- menoptera). Special Publication of the Internation- al Society of Hymenopterists 1, Washington D.C. Van Achterberg, C. 1997. Revision of the Haliday col- lection of Braconidae (Hymenoptera). Zoologische Verhandelingen 314: 1-115. Van Achterberg, C. and A. Salvo. 1997. Reared Opi- inae (Hymenoptera: Braconidae) from Argentina. Zoologische Mededelingen 71: 189-214. Viereck, H. L. 1913. Descriptions of ten new genera and twenty-three new species of ichneumon- flies. Proceedings of the United States National Mu- seum 44: 555-568. Wharton, R. 1980. Review of the Nearctic Alysiini (Hymenoptera: Braconidae) with discussion of generic relationships within the tribe. University of California Publications in Entomology 88: 1-112. Wharton, R. A. 1987. Changes in nomenclature and classification of some opiine Braconidae (Hyme- noptera). Proceedings of the Entomological Society of Washington 89: 61-73. Wharton, R. A. 1988. Classification of the braconid subfamily Opiinae (Hymenoptera). The Canadian Entomologist 120: 333-360. Wharton, R. A. 1997a. Subfamily Opiinae, p. 379-395. In: Wharton, R. A., P. M. Marsh, and M. J. Shar- key (eds.). Manual of the New World Genera of the Family Braconidae (Hymenoptera). Special Publi- cation of the International Society of Hymenop- terists 1, Washington D.C. Wharton, R. A. 1997b. Generic relationships of opiine Braconidae (Hymenoptera) parasitic on fruit-in- festing Tephritidae (Diptera). Contributions of the American Entomological Institute 30: 1-53. Wharton, R. A. 1999. A review of the Old World Ge- nus Fopius Wharton (Hymenoptera: Braconidae: Opiinae), with description of two new species reared from fruit-infesting Tephritidae (Diptera). Journal of Hymenoptera Research 8: 48-64. J. HYM. RES. Vol. 12(2), 2003, pp. 303-307 Prosopigastra morogoro, a New Species from Tanzania (Hymenoptera: Apoidea: Crabronidae: Larrini) WOJCIECH J. PULAWSKI Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA; email: wpulawski@calacademy.org Abstract. — The new species Prosopigastra morogoro Pulawski, from Tanzania, is characterized by a densely punctate frontal protuberance and gena, an unusual female pygidial plate, presence of a male pygidial plate, and unique male sternum VII. Its closest congener is the southern African capensis Brauns. Several corrections are made to an earlier diagnosis of the genus by Pulawski, 1979. I revised the world species of Prosopi- gastra more than twenty years ago (Pu- lawski 1979). During a collecting trip to Tanzania in 2001, I discovered a spectac- ular undescribed species apparently never collected before. Its discovery requires three corrections to my earlier diagnosis of the genus: 1) The marginal cell is longer than in other members of the genus and is not broadly truncate, its length being 2.6-2.9 X maximum width of the cell (in- ner dimensions) rather than 1.6-2.3, and the distance between its posteroapical cor- ner equaling 1.1-1.4X its maximum width. Contrary to my original statement, the cell length of the new species overlaps with those of Holotachysphex, Parapiagetia, and Tachysphex and therefore is not diag- nostic for the entire genus. 2) The pygidial plate of the female has a number of large, ill-defined punctures on its entire surface. The presence of an adlateral row of punc- tures, therefore, is not diagnostic for the entire genus. 3) Similarly, male tergum VII has a well-defined pygidial plate and lacks a translucent, impunctate apical depres- sion. The presence of the depression and lack of a pygidial plate are not character- istics of the entire genus. The terminology in the following de- scription is as in Pulawski (1979). Prosopigastra morogoro Pulawski, new species Name derivation. — Morogoro, a town in Tan- zania in whose vicinity the species was first dis- covered; a noun in apposition. Recognition. — Prosopigastra morogoro is unique in having a conspicuous, densely punctate fron- tal protuberance and a uniformly, densely punctate gena, with punctures one diameter apart or less. In other Prosopigastra, the frontal protuberance is either prominent and impunc- tate or punctate and inconspicuous, and the ge- nal punctures are several to many diameter apart, at least near the hypostomal carina. Also, the marginal cell of morogoro is longer than in any other species, its anterior margin being 2.6- 2.9 X maximum cell width (inner dimensions) rather than 1.6-2.3. The female has irregular, large punctures on the entire pygidial plate (Fig. Id). In the male, tergum VII has a well- defined pygidial plate (unlike any other Proso- pigastra), and sternum VIII is thickened near the apex, the thickening having an apical concavity (Fig. 2d, f), possibly a unique feature among Apoidea. Description. — Frons microridged between an- tennal socket and protuberance; protuberance prominent, punctate throughout (punctures less than one diameter apart). Middle clypeal section convex, with minute carina emerging from corner of clypeal lobe and nearly parallel to clypeal free margin; lip slightly, obtusely pointed mesally, not incised laterally. Gena densely, uniformly punctate throughout, punc- 304 Journal of Hymenoptera Research -.HO 0.25 mm ••II ?'W Fig. 1. Prosopigastra morogoro, female: a — head in frontal view, b — clypeus, c — head in lateral view, d — pygidial plate. tures no more than one diameter apart. Ventral mandibular margin step-like, without preapical expansion at distal end of notch (hence notch open distally). Mesopleuron either punctate (punctures less than one diameter apart) or punctatorugose, with small unsculptured area above scrobe. Precoxal mesopleural carina sharp, expanded into spine in male. Propodeal dorsum with longitudinal, anastomosed ridges, in some specimens irregularly rugose mesally. Marginal cell with dense microtrichia, its ante- rior margin longer than pterostigma, with length 2.6-2.9 X maximum cell's width (inner dimensions), apical truncation oblique; distance Volume 12, Number 2, 2003 305 0.25 mm 0.1 mm 0.2 mm 0.3 mm 0.1 mm Fig. 2. Prosopigastra morogoro, male: a — clypeus, b — thorax in lateral view, c — pygidiaJ plate, d — gastral ster- na, e — sternum VII, f — sternum VIII in oblique lateral view. 306 Journal of Hymenoptera Research 0.2 mm Fig. 3. Prosopigastra morogoro, male genitalia: a — dorsal view, b — gonoforceps in lateral view, c — volsella, d- penis valve. between cell's posteroapical corner and anterior margin equal to 1.1-1.4X cell's maximum width. Punctures less than one diameter apart between midocellus and orbit and on gena (ex- cept dorsally), no more than one diameter apart on interocellar area and scutum; less than one diameter apart on hindfemoral outer surface except more than one diameter apart along ven- tral margin in female. Gastral terga coarsely punctate, terga 1 and II each with lateral line (lateral carina of Pulawski, 1979). Setae inclined anterad on vertex, about 1.5 X as long as mi- Volume 12, Number 2, 2003 307 docellar diameter, on scutum inclined posterad, shorter than midocellar diameter (longest near anterior margin). Mesopleural vestiture not ob- scuring integument. Upper metapleuron large- ly glabrous. Tergum I without tomentum in fe- male, with rudimentary tomentum just poster- ad of basal declivity in many males. Body black except mandible reddish preapically in female and yellowish white in basal two thirds in male and tarsal apex brown. Wing membrane slight- ly infumate, veins dark brown. Female. — Clypeus (Fig. la): lip slightly, ob- tusely pointed mesally, not incised laterally. Width of postocellar area about 2.5 x length. Precoxal mesopleural carina sharp. Pygidial plate with irregular, large punctures, without setae in unique specimen in unique specimen, probably due to abrasion (Fig. Id). Length 7.5 mm. Male. — Flagellum cylindrical. Width of pos- tocellar area 1.7-2.1 x length. Mesothoracic ven- ter deeply concave; precoxal mesopleural cari- na expanded into prominent spine (which is disproportionately larger in large specimens); spine connected by carina to apophysis-like sig- num. Tergum VII with well defined pygidial plate (Fig. 2c). Sterna III— VI shallowly concave mesally; sterna V and VI posterolaterally each with conspicuous tuft of setae; sternum VII with apical emargination that is partly covered by membrane (Fig. 2e); sternum VIII conspicu- ously thickened apically, with apical surface of thickening characteristically concave (2f). Length 6.1-8.5 mm. Genitalia, gonoforceps, vol- sella, and penis valve: Fig. 3a-d. Relationships. — Within Prosopigastra, the lack of a ventral preapical expansion in the mandible is shared only by morogoro and capensis Brauns, and is clearly derived within the genus. The markedly modified male mesopleuron of morogoro, another conspicuous apomorphy, is also found in capensis and creon (Nurse). Most likely it is an independent development in the latter species, which belongs to the globiceps group of Pulawski (1979). The group is characterized by the yellow legs markings and holoptic eyes in most males, two de- rived features that are absent in both ca- pensis and morogoro. Thus, capensis and morogoro appear to be the closest relatives within Prosopigastra. Type material. — Holotype 6: Tanzania: Morogoro Region: 48 km W Morogoro at 6°56.9'S 37°20.2'E, M.H. Bourbin and WJ. Pulawski (California Academy of Scienc- es). Paratypes (all in California Academy of Sciences): TANZANIA: Iringa Region: 18 km W Iringa at 7°53.8'S 35°35.7'E, M.H. Bourbin and WJ. Pulawski, 9 June 2001 (IS), 20 June 2001 (19, 56). Morogoro Region: same data as holotype: 3 June 2001 (2 J), 6 June 2001 (2d), 11 June 2001 (26), 18 June 2001 (1 6 ); Omary S Haji and W.J. Pulawski, 2-3 July 2001 (16), 7 July 2001 (16), and 23 July 2001 (16). Most of the paratypes are deposited at the Califor- nia Academy of Sciences, and one each in the Museum fur Naturkunde, Berlin, The Natural History Museum, London, and the United States National Museum of Natural History, Washington, D.C. ACKNOWLEDGMENTS I sincerely thank Professor Marcelian Njau (Uni- versity of Dar es Salaam, Tanzania) for his help in organizing my expedition, as well as my travel com- panions in Tanzania, Mrs. Maureen H. Bourbin (Cal- ifornia Academy of Sciences, San Francisco, Califor- nia) and Mr. Omary S. Haji (Dar es Salaam, Tanzania) for their help in collecting specimens; and Ms. Vir- ginia Kirsch (San Francisco, California) for generating the illustrations. Upon my request, Arnold S. Menke and Michael A. Prentice reviewed earlier versions of the manuscript and significantly improved it. Mi- chael Ohl, the official reviewer, also contributed to the quality of the paper. LITERATURE CITED Pulawski, W. J. 1979. A revision of the World Proso- pigastra Costa (Hymenoptera, Sphecidae). Polskie Pismo Entomologiczne 49: 3-134. J. HYM. RES. Vol. 12(2), 2003, pp. 308-311 A New Species of Copidosoma Ratzeburg (Hymenoptera: Encyrtidae) from Eagle Nests in Kazakhstan Andrey Sharkov, Todd E. Katzner, and Tatyana Bragina (AS) Department of Biochemistry, Chemical Abstracts Service, 2540 Olentangy River Road, Columbus, OH 43202, USA, email: asharkov2@cas.org; (TEK) Department of Biology, Arizona State University, P.O. Box 871501, Tempe, AZ 85287-1501, USA; (TB) Department of Biology, Kostanay State University, 118 Taran St., Kostanay, 458000, Kazakhstan Abstract. — A new species of polyembryonic encyrtid of the genus Copidosoma is described from north-central Kazakhstan. Copidosoma naurzumense, n. sp., was reared from tineid moth larvae collected from regurgitated pellets collected near eagle nests in the Naurzum nature reserve. This is the third species of the genus Copidosoma reported from Kazakhstan. This species is similar to C. longiventre Myartseva form Turkmenistan, from which it differs by having dark tegulae and front coxae, shorter antennal segments and clava, and smaller body size. The natural history and ecology of the parasitoid and its host are discussed. Pe3K)Me. OnucaH hobwh bhjx 3HiiHpTH,a poaa Copidosoma H3 Ka3axcraHa (Hayp3yMCKHH rocy^apcTBeHHbiH npHpoAHbiH 3anoBe^HHK). Copidosoma naurzumense, sp. n. Bbme^eH H3 ryceHnu MOJieii ceMeHCTBa Tineidae (Lepidoptera), HafmeHHbix b nora/iKax, co6paHHbix okojio rHe3^ opnoB. 3to TpeTHH npcacraBHTejib po^a Copidosoma, o6Hapy>KeHHbiH b Ka3axcraHe. C. naurzumense sp. n. cxoneH c C longiventre Myartseva H3 TypKMeHHCTaHa, ot KOToporo oh OTJiHHaeTca TeMHbiMH TerynaMH h nepe^HHMH Ta3HKaMH, 6ojiee KOponcHMH HjreHHKaMH h GyjiaBoii ycHKOB, a TaioKe MeHbiiiHMH pa3MepaMH Tcna. npuBe/teHbi KpaTKne CBeAeHHH IIO GnOJlOrHH H 3KOJIOrHH napa3HTOH,Zia H X03MHHa. Parasitic wasps of the encyrtid genus Copidosoma are polyembryonic parasitoids of lepidopteran caterpillars. The genus has a worldwide distribution and, according to Noyes et al. (1997), includes about 150 described species. However, Trjapitzin (1989) suggests that there are 184 de- scribed species of Copidosoma in the world, 133 of which occur in the Palearctic. The fauna of the family Encyrtidae of Central Asia was revised most recently by Myart- seva (1984), and includes 31 species of mdosoma (4 of them as Litomastix), with ecies, C. filicorne (Dalman) and Ratzeburg, found in Ka- zakhstan. The new species described below was reared by T.Katzner from caterpillars of clothes moths (Lepidoptera: Tineidae) col- lected from regurgitated pellets from ea- gles in the Naurzumskiy Zapovednik (Naur- zum National Nature Reserve) in the Naurzumskiy region of the Kostanay Ob- last' of north-central Kazakhstan (51° N, 64° E). Copidosoma naurzumense Sharkov, Katzner and Bragina Female. — Body length 1.2-1.6 mm (ho- lotype — 1.44 mm). Head: Width approxi- mately twice its length and equal to its height (35:18:35). Frontovertex width at Volume 12, Number 2, 2003 309 Figs. 1-2. Copidosoma naurzumense Sharkov, Katzner and Bragina, n. sp. 1 — female antenna; 2 — male antenna. the level of anterior ocellus approximately Vi head width (18:35). Distance between posterior margin of eye and occipital mar- gin approximately 1/9 eye length from above (1.5:13.5). Ocelli in obtuse triangle with the angle at the anterior ocellus of 103°. POL:OOL:LOL:OCL = 11:1.3:5.5:1. Maximum diameter of eye 1.2 times its minimum diameter (17:14). Distance be- tween antennal toruli twice the distance between antennal torulus and mouth mar- gin and approximately lA the distance be- tween antennal torulus and eye margin (4: 2:13). Antenna as in Fig. 1. Mouth width slightly greater than malar space (16:15). Mesosoma: length 1.4 times its width (49: 35). Scutum transverse, its length about % its width (22:35). Scutellum of equal length and width (22:22). Mid tibial spur length equal to length of first tarsomere (9: 9) and approximately V4 length of middle tibia (9:35). Fore wing length 2.3 times its width (93:41) (Fig. 3). Metasoma: Longer than head and mesosoma combined. In dry specimens its length varies depending on degree of extension of sclerites, and ranges from 1.1 to 2.0 times combined length of head and mesosoma (85:60 in the holotype). Ovipositor not exserted, with gonostyli fused to second valvifers, their length about Vi length of mid tibia. Color: Head and body generally dark, almost black. Face very dark brown, almost black, frons and vertex black, with very slight dark blue-green shine. Antenna dark brown, with slightly lighter apex of the pedicel. Scutum black, with dark blue- green shine, which is slightly more strong- ly expressed than on vertex; scutellum with slight dark purple reflection. All cox- ae dark brown. Front femur and tibia dark brown, with very light brown apex of fe- mur and base of tibia, tibia very slightly lighter towards the apex; tarsus brown. Middle femur dark brown, with light brown apex; tibia light brown, with white translucent base, and gradually becoming brownish yellowish white toward apex; spur and tarsus almost white, with last tarsomere brownish. Hind femur dark brown, with light apex, tibia with whitish translucent base, dark brown in middle, 310 Journal of Hymenoptera Research •^ •"' ^ , C ■J'^ ^ / , /t/ • ^ ** «** ,^ U" *" '— *""^ «f^ c— v~ l — t t— tr~ t— ^ °— e— t— ^0-~ *t- n* III / v * ■•" "^r" *■ ^ **■« _ tfc *-- ***- Fig. 3. Copidosoma naurzumense Sharkov, Katzner and Bragina, n. sp. female forewing. and gradually becoming brownish yellow in apical Vr, tarsus yellowish white, with last tarsomere brownish. Metasoma black, with very slight metallic reflection. Sculp- ture: Head and body reticulate, with cells rounded on frons, vertex, and dorsal side of mesosoma, and elongate on gena, me- sopleuron and metasoma. Cell diameter on dorsal side of head and mesosoma about Vi diameter of posterior ocellus. Male. — Body length 1.2-1.4 mm. Gen- eral appearance as in female, except head slightly wider (width 2.1 times length and 1.1 times height), ocelli in more obtuse tri- angle (angle at anterior ocellus 115°), mid tibial spur slightly shorter than first tar- somere, and metasoma equal in length or shorter than head and mesosoma com- bined. Length of aedeagus, when exserted, V3 to Vz length of mid tibia. Digiti with two teeth. Antenna as in Fig. 2. Color and sculpture as in female, although sculpture patterns somewhat more pronounced. Types.— Holotype female: 51°33.3'N 064°07.9'E KAZAKHSTAN, Kostanay re- miskiy zapovednik, near Kar- nerly Dokuchaevka), eagle #10, 6.V.1998, emerged itzner. Para types: 2 fe- "i males, same data ex- cept, 16.vii.1998; 30 males, same data ex- cept, eagle pellet from nest #21, 26.V.1999; 8 females, same data except eagle pellet from nest #13, 26.V.1999; 3 females, 4 males, same data except, eagle pellet from nest #12, 19.vi.1999; 41 females, 43 males, same locality, summer 1998 (deposited at the OSU Insect Collection, Columbus, OH). Diagnosis. — From C. filicorne, which also occurs in Kazakhstan, differs in having the metasoma longer than the head and me- sosoma combined, and dark brown front coxa. The second species occuring in Ka- zakhstan, C. bouchenum, has white tegulae, the clava shorter than three preceding fu- nicular segments, and the body length of 3.0-3.2 mm. In Myartseva's (1984) key runs to C. longiventre Myartseva from Turkmenistan, from which it differs by having dark tegulae and front coxae, shorter funicular segments and clava, and a smaller body size. In C. longiventre fu- nicular segments are more than twice lon- ger than wide, clava is equal in length to four preceding funicular segments com- bined, and the body length is 2.4 mm. In Trjapitzin's (1989) key C. naurzumense runs to C. clavatutn Myartseva from Turk- menistan, from which, according to Volume 12, Number 2, 2003 311 Myartseva's (1984) key, it differs in having the metasoma longer than the head and mesosoma combined. In Kazmi and Hay- at's (1998) key to Indian Copidosoma, runs to C. koehleri, an introduced South Amer- ican species, which is a parasitoid of the potato tuber moth. Natural history and ecology. — The climate in the region of the Naurzumskiy Zapov- ednik is harsh continental. Minimum win- ter temperatures are —45 °C, and summer maximums reach 41 °C, with average yearly temperature being 2.4 °C. Precipi- tation is highly variable, but averages 233 mm per year. Parasitized and unparasitized host lar- vae were found in summers 1997-2000 in regurgitated pellets collected from nests and roosts of several species of eagles Aq- uila and Haliaeetus. Although hosts and parasitoids were collected from pellets during each month of the summer and late spring, host larvae were most fre- quently observed during the colder and wetter months of April and May. Host lar- vae were found in pellets composed of mammal fur, bird feathers, or mixtures of both. Of more than 9500 pellets evaluated, less than 500 contained the host larvae, with the number of larvae per pellet being generally less than ten, but occasionally more than 100. Current estimates of the parasitism rate are that it is less than 40%. Sterile defender (precocious) larvae occur in this species, as they do in several other polyembryonic encyrtids (Cruz 1981, 1986). Similar numbers of tineid moths and hymenopterans were found in pellets of raptors in North America (Philips and Dindal 1979). Pellets represent a large con- centration of potential food that can sup- port diverse invertebrate communities. ACKNOWLEDGMENTS The species description and drawings were made using the facilities and equipment of the Insect Col- lection of the Department of Entomology of the Ohio State University. The authors are grateful to Dr. Nor- man F. Johnson, Dr. Peter W. Kovarik, and Dr. John W. Wenzel for their assistance. Dr. Yolanda Cruz kindly reviewed the manuscript. Dr. John S. Noyes (NHM) made several important suggestions. Dr. Ev- geny Bragin, Anatoly Taran, Fergus Crystal, Seth Layman and Doug Grier assisted with collection and analysis of eagle pellets. This research was partially funded by a US-EPA STAR research fellowship, USGS Biological Resources Division, Wildlife Conser- vation Society, Arizona State University Department of Biology, Frank M. Chapman Memorial Fund, Hawk Mountain-Zeiss Raptor Research Award, Ari- zona State University Graduate Research Support Program, The International Osprey Foundation, World Nature Association, Arizona State University Russian and East European Studies Consortium, and Northwest Airlines. We acknowledge assistance of the Naurzumskiy Zapovednik and the government of Kazakhstan. LITERATURE CITED Cruz, Y. P. 1981. A sterile defender morph in a poly- embryonic hymenopterous parasite. Nature (Lon- don) 294: 446-447. Cruz, Y. P. 1986. The defender role of the precocious larvae of the polyembryonic encyrtid wasp Cop- idosomopsis tanytmemus Caltagirone (Encyrtidae, Hymenoptera). journal of Experimental Zoology 237: 309-318. Kazmi, S. I. and M. Hayat. 1998. Revision of the In- dian Copidosomatini (Hymenoptera, Chalcidoi- dea: Encyrtidae). Oriental Insects 32: 287-362. Myartseva, S. N. 1984. Parasitic hymenopterans of the family Encyrtidae (Hymenoptera: Encyrtidae) of Turkmenistan and adjacent regions of Central Asia. Ashkhabad, Ylym, 305 p. [in Russian] Noyes, J. S., J. B. Woolley, and G. Zolnerowich. 1997. Encyrtidae, pp. 170-320. In: Annotated keys to the genera of Nearctic Chalcidoidea (Hymenoptera). Ot- tawa, NRC Research Press. Philips, J. R. and D. L. Dindal. 1979. Decomposition of raptor pellets, journal of Raptor Research 13: 102-111. Trjapitzin, V. A. 1989. Parasitic Hymenoptera of the fam- ily Encyrtidae from the Palacarctic region. Lenin- grad, Nauka, 488 p. [in Russian] J. HYM. RES. Vol. 12(2), 2003, pp. 312-332 Food Plants and Life Histories of Sawflies of the Families Tenthredinidae and Pergidae (Hymenoptera) in Costa Rica, with Descriptions of Four New Species David R. Smith and Daniel H. Janzen (DRS) Systematic Entomology Laboratory, PSI, Agricultural Research Service, U.S. Department of Agriculture, c/o National Museum of Natural History, Smithsonian Institution, Washington, DC 20560-0168, USA, email: dsmith@sel.barc.usda.gov; (DHJ) Department of Biology, University of Pennsylvania, Philadelphia, PA 19104, USA, email: djanzen@sas.upenn.edu Abstract. — Food plants and biological information are given for five species of Tenthredinidae and six species of Pergidae reared in the Area de Conservacion Guanacaste, Guanacaste Province, northwestern Costa Rica. The Tenthredinidae are Adiaclema chigiyae Smith, n. sp. on Entodon- topsis leucostega (Stereophyllaceae), Waldheimia fascipennis (Norton) on Cissus pseudosicyoides (Vi- taceae), Waldheimia suturalis (Cameron) on Cissus rhombifolia (Vitaceae), Waldheimia interstitialis (Cameron), n. comb., on Hamelia patens (Rubiaceae), and Pristiphora auricauda Smith on Primus annularis (Rosaceae). The Pergidae are Aulacomerus delictus Smith on Mesechites trifida (Apocyna- ceae), Anatlntlea bimaculata (Cameron) on Hippocratea volubilis (Hippocrateaceae), Suwatnus nigriceps (Cameron) on Psidium guajava (Myrtaceae), Acordulecera binelli Smith, n. sp., on Posoqueria latifolia (Rubiaceae), Acordulecera liami Smith, n. sp., on Erythroxylwn havanense (Erythroxylaceae), and Acordulecera dashielli Smith, n. sp., on Arrabidaea patellifera (Bigoniaceae). Several hundred wild- caught larvae of these 11 species produced no parasitoids. Redescriptions are given for Aulacom- erus daktus, Anathulea bimaculata, and Suwatnus nigriceps. This is the second of two treatments of the host plants and life histories of saw- flies reared by DHJ during the Lepidop- tera caterpillar inventory of the Area de Conservacion Guanacaste (ACG), which lies primarily in Guanacaste Province in northwestern Costa Rica. The first covered the family Argidae (Smith and Janzen 2003). Here we consider the families Ten- thredinidae and Pergidae, the adults of which may be distinguished in the key to families by Smith (1988, 1995). Symphyta larvae are keyed to family and to subfam- ilies within the Tenthredinidae by Smith and Middlekauff (1987). Though based on the Nearctic fauna, this larval key will be helpful for larvae collected in Costa Rica. Details of the methods and rearing records may be found at http://janzen.sas.upenn. edu and in Janzen (2000, in press), Schauff and Janzen (2001), Janzen et al. (2003), and Burns and Janzen (2001). Acronyms used are: INBio = Instituto Nacional de Biodiversidad, Santo Domin- go de Heredia, Costa Rica; USNM = Na- tional Museum of Natural History, Smith- sonian Institution, Washington, DC, USA; BMNH = The Natural History Museum, London, UK. Voucher numbers associated with each reared adult are expressed as, for example, "99-SRNP-4547"; full details of the voucher record and associated im- ages may be obtained at http:/ /janzen. sas.upenn.edu. TENTHREDINIDAE This is a large family in the Neotropics, with about 32 genera and over 300 species. Four of the six Neotropical subfamilies are known from Costa Rica, the largest being Volume 12, Number 2, 2003 313 the Selandriinae and Blennocampinae. Only a few species of the other two, Ne- matinae (Smith 2003a) and Allantinae (Smith 2003b), occur there. The subfami- lies may be distinguished with the keys in Smith (1995, 2003a). All Tenthredinidae in Costa Rica have 9-segmented, filiform or slightly compressed antennae. SELANDRIINAE This subfamily is recognized by its dis- tinctive wing venation; most have vein 2A+3A complete, lack the anal crossvein, and have vein Rs + M distinctly curved near Sc + R in the forewing. It is the largest subfamily of Tenthredinidae in the Neo- tropics, and probably 60-70 species in nine genera occur in Costa Rica. Food plants are known for very few of them. Most extra-tropical members of this sub- family have ferns and grasses as food plants. Adiaclema Enderlein About 30 species of Adiaclema occur from southern Mexico to northern Argen- tina. The genus is distinguished by the ab- sence of the anal crossvein in the forewing (a characteristic of the subfamily) and sim- ple mandibles, a trait not possessed by other Neotropical Selandriinae. All other Neotropical selandriines have one or more subapical teeth on one or both mandibles. Types of all described species have been examined by DRS except for several that cannot be located, and it was determined that the following species reared in Costa Rica represents a new species. This is the first food plant record for the genus. Adiaclema chigiyae Smith, new species (Figs. 1, 8, 15, 16, 27, 32, 33, 37-39) Female. — Length, 8.0-8.5 mm. Antenna black. Head black with clypeus, labrum, and maxillary and labial palpi white. Tho- rax orange. Abdomen orange with apical three segments and sheath black. Legs or- ange with foretarsal segments 3-4 slightly infuscate; midfemur with a narrow black ring at apex; midtarsus black; apical 2A of hind tibia black with basal V3 white; hind tarsus black. Forewing yellow with apex beyond stigma black; costa, subcosta, stig- ma, and veins yellow in yellow area; veins black in black apex. Antennal length 2.1 X head width; length of 3rd segment 1.3X length of 4th segment; segments beyond third gradual- ly decreasing in length (Fig. 8). Eyes large and converging below; lower interocular distance 0.8 X eye length; upper interocu- lar distance slightly greater than eye length (Fig. 15). Head from above strongly narrowing behind eyes; postocellar area 1.2X broader than long; distances between hind ocelli, hind ocellus and eye, and hind ocellus and posterior margin of head as 6: 12:10 (Fig. 16). Malar space linear; clypeus with anterior margin slightly convex; la- brum about 2X broader than long with anterior margin truncate. Hind basitarsus subequal in length to length of remaining tarsal segments combined. Hind wing with anal cell sessile. Tarsal claw with long inner tooth slightly shorter than out- er tooth and without basal lobe (Fig. 37). Sheath short and rounded at apex in lat- eral view; in dorsal view slightly broader at center (Fig. 32). Lancet (Fig. 27) short, triangular, with about 9 serrulae; protu- berances laterally on annuli 2-10, becom- ing more spinelike toward apex. Male. — Length, 7.1-7.5 mm. Color simi- lar to that of female except mesoprescu- tum blackish at center and nearly all hind tibia black with white only at extreme base. Tarsal claw with long, slender outer tooth and minute inner tooth (Fig. 38). Genitalia (Fig. 39) with harpe elongate and rounded at apex; parapenis narrow, taper- ing to small rounded apex; apex of penis valve about as long as broad, rounded at apex, with long dorsal lobe. Holotype. — Female, labeled "Voucher: D. H. Janzen & W. Hallwachs, caterpillar (Lepidoptera) database, Area de Conser- vacion Guanacaste, Costa Rica, http:// 314 Journal of Hymenoptera Research janzen.sas.upenn.edu, 99-SRNP.9777." De- posited at INBio. Paratypes.— COSTA RICA: All labeled as holotype except for voucher codes: 99- SRNP-9757 (19); 99-SRNP-10267 (16); 99- SRNP-10269 (19); 99-SRNP-10273 (19); 99-SRNP-10285 (1 9 ); 99-SRNP-10288 (16); 99-SRNP-10297 (1 9 ); 99-SRNP-10305 (1 9 ); 99-SRNP-10314 (1 9 ); 99-SRNP-10334 (1 6 ); 99-SRNP-10348 (19); Guanacaste Prov- ince, Santa Rosa National Park, D. H. Jan- zen, 24-VIII-14-IX-1985, Malaise trap SE- 6-C (19). Deposited at INBio and USNM. Etymology. — This species is named in honor of Chigiy Binell in recognition of her great enthusiasm for the ACG and support for the Rincon Rainforest. Food plant and biology. — Larvae (Fig. 1) feed on moss, Entodontopsis Icncostega (Brid.) W. R. Buck & Ireland (Stereophyl- laceae). In the second month of the rainy season (late June to early July), the pen- ultimate and last instar larvae were en- countered feeding solitarily day and night while fully exposed on the rough surface of moss patches on logs and stumps in the ACG dry forest. They were not encoun- tered in the first two decades of the cat- erpillar inventory because it did not occur to us to search shady, wet patches of moss. The prepupal larva chews into the sur- face of rotten bark or wood, hollows out a smooth-walled ovoid chamber in which to pupate, and closes the entrance hole with wood or bark chips glued together. There is no sign of a silken cocoon in the chamber. The adult emerges 40-60 days after the cocooning chamber is construct- ed (average 47 days, n = 12) in the ACG rainy season climate. There was no evi- dence of pupal dormancy, but this does not exclude the possibility that the pre- pupae or pupae late in the rainy season pass the six month dry season dormant in the cocoon chamber. No parasitoids were produced from 135 wild-caught penulti- mate and last-instar larvae. Remarks.— -Of the Neotropical Adiackma species, only a few have the combination of yellow wings with a black apex, black head, entirely orange thorax, and orange abdomen with the apical segments black. This color is similar to Adiackma tetricum (Konow) described from "Peru (Pozuzo)" (type examined), but A. tetricum differs by having the supraclypeal area and area around antennal bases white, has much longer antennae (length nearly three times the head width), has a long, narrow ovi- positor sheath (with valvula 3 about two times longer than broad), has the lower in- terocular distance about 0.9 X the eye length, and is generally larger, being about 10 mm in length. Adiackma blandu- lum (Enderlein) (type not located) de- scribed from Ecuador is also similar, but has most of the face above the antennae white. In all other Adiackma species examined, the tarsal claws of the female and male are similar with a long inner tooth and with- out a basal lobe (Fig. 37). It is most un- usual for the male tarsal claws to differ so much from those of the female (Figs. 37, 38). BLENNOCAMPINAE There are about 150 species in 14 genera of Blennocampinae in the Neotropical Re- gion and 25-30 species in five genera in Costa Rica, most of which are in the large genus Waldheimia. The subfamily is distin- guished by its wing venation, with veins M and lm-cu parallel, veins M and Rs + M meeting Sc + R at the same point, and the anal cell petiolate with the basal section of vein 2A+3A absent. One species, Metape- dias siibcoeruka (Cameron) has been reared from Conostegia xalapensis D. Don (Melas- tomataceae) (Smith 1995). Species of the genus Periclista Konow are found at high elevations in Costa Rica and may feed on Quercus sp. (Fagaceae), as do the North American counterparts. Waldheimia Brulle Nearly 100 species of Waldheimia are known with about ten in Costa Rica, and Volume 12, Number 2, 2003 315 the genus is widely distributed from southwestern United States to northern Argentina. Characteristics of the genus in- clude the reduced apical four antennal segments, straight forewing vein 2A + 3A, and the bifid tarsal claws with the inner tooth usually broader and longer than the outer tooth and with a basal lobe. Three species have been reared in ACG. Waldheimia fascipennis (Norton) (Fig. 2) Discussion. — This is one of the few spe- cies of Waldheimia that have the forewings broadly black at both the base and apex and yellow in the center. The head, thorax, and abdomen are orange with the intero- cellar area, apical three abdominal seg- ments, and sheath black. The coxae, tro- chanters, and femora are orange, the tibiae are white at the bases with the apical half to two-thirds black, the basitarsi are white with a narrow black apical ring, and the remaining tarsal segments are black. The male is unknown. Distribution. — Costa Rica (Guanacaste); Mexico (Campeche, Chiapas); El Salvador. In addition to the reared specimens, the following from Costa Rica also have been examined: Santa Rosa National Park, 14- IX-5-X-1985, Malaise trap H-3-0; Prov. Guanacaste, OTS Palo Verde Sta., 29 km W.S.W. Canas, 10°21'N, 85°21'W, 14- VII- 1976. Food plant and biology. — Waldheimia fas- cipennis larvae feed on leaves of Cissus pseudosicyoides Croat (Vitaceae) during the rainy season in ACG dry forest (two fe- males: 90-SRNP-1141, 83-SRNP-1144). One record from the last month of the rainy season represents either a second or third rainy season generation. The bluish and pale yellow larvae feed side-by-side on the upper sides of the relatively horizontal mature leaves in groups of 1-5 (Fig. 2). They are currently indistinguishable in color pattern, shape, and behavior to those of Waldheimia suturalis, which have been reared from Cissus rhombifolia in the same habitat. Numerous Waldheimia larvae feed- ing on both species of Cissus have not pro- duced adults, and it may be that both spe- cies of Waldheimia feed on both species of Cissus. The prepupal larva burrows down into the litter and pupates naked in a chamber with no silk cocoon. The adults emerged 13-14 days after the prepupa entered the soil (n = 2). While no dormancy was re- corded in rainy season pupae, the prepu- pae may well pass the dry season dormant in the litter. A total of 36 Waldheimia (W. fascipennis and W. suturalis) wild-caught penultimate and last-instar larvae pro- duced no parasitoids. Waldheimia interstitialis (Cameron), new combination (Fig. 3) Discussion. — This species was described in the genus Blennocampa Hartig by Cam- eron (1883) and transferred to Erythraspi- des Ashmead by Smith (in Kimsey and Smith 1985), but Erythraspides and Wald- heimia cannot be distinguished in the Neo- tropics. Formal generic synonymy will be presented later in another paper, and here we give the new combination. The coloration is distinctive for both sexes of this species. The antennae are black with the scape and pedicel orange. The head, thorax, and abdomen are or- ange with the interocellar area and sheath black. The legs are orange, with the apex of the midfemur, the entire hind femur, the apical third to half of the mid- and hind tibiae, and the apical three tarsal seg- ments of all legs black. The wings are yel- low with the apex beyond the stigma black. Cell M is absent in the hind wing. Distribution. — Costa Rica (Alajueia, Guanacaste, Heredia, San Jose), Mexico (Veracruz); Panama; Venezuela. Costa Ri- can specimens examined are as follows: Prov. Heredia, La Selva Biol. Sta., 3 km S Pto. Viejo, 10°26'N, 84°01'W, 6-VI-83, 27- IV-90, 17-1-91; Escazu, May 21, 24, 26, 27, 316 Journal of Hymenoptera Research Figs. 1-4. Larvae. 1, Adiaclema chigiyae. 2, Waldhtimia fascipennis. 3, W. inter stitialis. 4, Aulacomerus daktus. 1987; Alajueia, 700 m, Penas Blancas, IV- 1987. Food plant and biology. — Hamelia patens Jacq. (Rubiaceae) was recorded as the host plant by Kimsey and Smith (1985) who also described and illustrated the larva and gave the life history in Panama. One female was reared from the same species of food plant in Costa Rica under number 00-SRNP-9451 (three others on the same food plant died of disease) in the lower margin of ACG cloud forest at about 1000 m elevation (April, July). Waldheimia inter- stitialis spun no cocoon, pupated naked in the bottom of its rearing container, and used eight days from prepupa until eclo- sion. The red-headed, dark gray-blue- black solitary larvae with the underside yellow (Fig. 3) are quite similar to those of the other two species of Waldheimia de- scribed here, and, like the others, feed in the daytime on the upper surface of the leaf. In laboratory rearings, Kimsey and Smith (1985), reported that the larvae go into the soil and form a smooth-walled, silk-lined cocoon. There were ten days from cocoon formation to eclosion. Waldheimia suturalis (Cameron) Discussion. — Both sexes of this species are entirely black with the following parts of the thorax red: pronotum, tegulae, me- sonotum, mesopleuron, and upper half of the metapleuron. The wings are uniform- ly, darkly infuscated. The female has cell M present in the hind wing, and the male has a peripheral vein in the hind wing. Distribution. — Costa Rica (Guanacaste, Heredia); Guatemala; Honduras; Mexico (Morelos, Veracruz). Specimens examined from Costa Rica other than those reared Volume 12, Number 2, 2003 317 Figs. 5-7. Larvae. 5, 6, Anathulea bimaculata. 7, Acordulecera dashielli. 318 Journal of Hymenoptera Research 8 * 13 14 Figs. 8-14. Antennae. 8, Adiaclema chigiyae. 9, Aulacomerus daktus, flagellum. 10, Anathulea bimaculata. 11, Suwatnus nigriceps. 12, Acordulecera binelli, pedicel and flagellum. 13, A. liami, pedicel and flagellum. 14, A. dashielli, pedicel and flagellum. Volume 12, Number 2, 2003 319 Figs. 15-18. Head. 15, Adiacleum chigiyae, front view. 16, A. chigiyae, dorsal view. 17, Anathulea bimaculata, front view. 18, A. bimaculata, dorsal view. are as follows: Guanacaste Prov., OTS Palo Verde Sta., 29 km W.S.W. Canas, 10°21'N, 85°21'W, 5-VII-1976; Heredia Prov., La Selva Biol. Sta., 3 km S Pto. Vie- jo, 10°26'N, 84°01'W, 17-IV-1988. Food plant and biology. — Three females were reared from larvae eating mature leaves of Cissus rhombifolia Vahl (Vitaceae) during the rainy season in ACG dry forest (83-SRNP-727, 98-SRNP-1831, and 98- SRNP-1833); two males were reared under numbers 98-SRNP-l 832.1 and 98-SRNP- 1832 from the same food plant species at the same time. The bluish and pale yellow larvae feed side-by-side on the upper sides of the relatively horizontal mature leaves in groups of one to five. They are currently indistinguishable in color pat- tern, shape, and behavior to those of Wald- heimia fascipennis, which have been reared from Cissus pseudosici/oides in the same habitat. Numerous Waldheimia larvae feed- ing on both species of Cissus have not pro- duced adults, and it may be that both spe- cies of Waldheimia feed on both species of Cissus. The prepupal larva burrows down into the litter and pupates in a chamber with no silk cocoon. The adults emerged 56-102 days (average 74 days, n = 5) after the prepupa entered the soil. This long pu- pation period is strikingly different from that of W. fascipennis, and probably rep- resents some kind of rainy season dor- mancy. As mentioned above, a total of 36 Waldheimia (W. fascipennis and W. suturalis) wild-caught penultimate and last instar larvae produced no parasitoids. NEMATINAE This is a large subfamily in the arctic, subarctic, and temperate regions of the Northern Hemisphere. The number of species decreases sharply to the south, and very few are known from the Neotropics (Smith 2003a). Pristiphora, the only genus 320 Journal of Hymenoptera Research known in Costa Rica, is the largest tropi- cal-extending genus, and occurs from Mexico south to southeastern Brazil. The subfamily is recognized by its distinctive forewing venation, with veins M and 1 ni- di markedly divergent, vein M meeting Sc + R far basal to the point where Rs + M meets Sc + R, and the anal cell petiolate with the base of vein 2A+3A absent. Pristiphora Latreille More than 50 species are known in the Nearctic, but only nine are known from Mexico to southern Brazil, and three in Costa Rica, mostly from elevations above 1000 in. The Neotropical species are keyed and described in Smith (2003a). Pristiphora auricauda Smith Discussion. — The distinctive coloration of the female (antennae, thorax, and legs black, and abdomen a contrasting bright orange, except for black basal plates and black anterior margin of the second seg- ment) distinguishes this species from oth- er Pristiphora in Costa Rica. The male has the head, thorax, and legs mostly yellow orange, with most of the head and thorax dorsally and the tibiae and tarsi black. The abdomen is orange, as is that of the fe- male. The female ovipositor and male gen- italia are illustrated in Smith (2003a). Distribution. — Costa Rica (Guanacaste). Other than the reared specimens from ACG, an additional Costa Rican record is from Est. Cacao, 1000-1400 m, Lado Sur- oeste del Volcan Cacao, Prov. Guam, II curso parataxon., Jun 1990. Food plant and biology. — The larvae live solitarily in leaf rolls of mature leaves of Prunus annularis Koehne (Rosaceae). Four- teen females and seven males were reared from 60 leaf rolls collected from two trees in the lower edge of ACG cloud forest at about 1000 m elevation (00-SRNP-9074, -9075, -9077, -9100, -9107, -9112, -9114, -9121, -9122, -9123, -9124, -9125, -9126, -9127, -9128, -9130, -9131, -9132, -9133, -9134, -9135). Eleven to 30 days were used between spinning the cocoon and eclosing (average 17 days, n = 14) during the rel- ative warm weather of the dry season. There was no suggestion of dormancy. The cocoon is a rough dark brown shaggy cylinder spun directly on the surface of the green leaf in the rearing container, though in nature the larvae probably de- scend to the litter to spin their cocoons. No parasitoids were reared from 60 wild- caught larvae. PERGIDAE Neotropical Pergidae were treated by Smith (1990). Eight subfamilies, 32 genera, and 256 species were recognized. Some additional species have been described since, and it would not be surprising if the number of described species is eventually doubled. Five subfamilies, ten genera, and 40 or more species occur in Costa Rica (Smith 1995). Representatives of two sub- families, Loboceratinae and Acorduleceri- nae, have been reared by the ACG cater- pillar inventory project. Food plants for other species include jel- ly fungus (Auricularia sp.) growing on rot- ting wood for Decameria rufiventris (Cam- eron) (Perreyiinae) (Smith 1995) and dried leaves for Perreyia tropica (Norton) (Perre- yiinae), the larvae of which travel in groups on the ground (Flores et al. 2000). LOBOCERATINAE Three genera were treated by Smith (1990), but most species are in the genus Aulacomerus, which is the only Costa Rican genus. Two species were recorded from Costa Rica by Smith (1990), but more oc- cur in the country. In Costa Rica, the sub- family is recognized by the 7-segmented, slightly clavate antennae with a large sen- sory pit on the apical segment, the midtib- iae with a preapical spine, and the hind tibiae lacking a preapical spine. Aulacomerus Spinola This genus contains 22 species and oc- curs from Mexico to northern Argentina Volume 12, Number 2, 2003 321 Figs. 19-22. Head. 19, Suwatnus nigriceps, front view. 20, S. nigriceps, dorsal view. 21, Acordulecera binelli, front view. 22, A. binelli, dorsal view. (Smith 1990). This is the first food plant record for a member of this genus. Aulacomerus daktus Smith (Figs. 4, 9) Discussion. — We redescribe this species since it was originally known from only a few specimens. It is distinguished from other Aulacomerus species by the key in Smith (1990). Description. — Length of female, 8.5-9.5 mm; male 7.0-8.0 mm. Antenna black with scape and pedicel orange yellow. Head orange with ocellar area and posto- cellar area black and with light black markings toward antenna and sometimes from lateral ocellus to eye. Antennal tu- bercles, supraclypeal area, clypeus, la- brum, and mandible whitish. Thorax or- ange. Legs orange with about apical half of hind tibia and all hind tarsus black; fore- and midtarsi black except bases of basitarsi usually orange yellow. Abdomen orange with center of 6th tergum and seg- ments 7 to apex and sheath entirely black. Wings yellow with apex beyond apex of stigma black; veins and stigma yellow, veins black in black apices. Antennal length 1.6X head width; large sensory area on apical segment; 3rd seg- ment slightly longer than 4th segment (Fig. 9). Clypeus with slight central circu- lar emargination. Eyes slightly converging below, lower interocular distance sube- qual to eye length, upper interocular dis- tance 1.1 X eye length. Head from above narrowing behind eyes. Postocellar area 1.6X broader than long. Distances be- tween hind ocelli, hind ocellus and eye, and from hind ocellus to posterior margin of head as 10:11:16. Hind basitarsus 1.3X length of remaining tarsal segments com- bined. Inner hind tibial spur 0.8X length of hind basitarsus. Female lancet and sheath and male genitalia illustrated by Smith (1990, figs. 338, 345, 368). 322 Distribution.— Costa Rica (Guanacaste), Guatemala, Mexico (Chiapas). This is the first record for Costa Rica. An additional record, other than the ACG specimens is: Vicinity Estac Murcielago, 8 km SW Cu- ajniquil, Guanacaste Prov., 100 m, Jun 1989, GNP Biodiversity Survey 320300, 380200. Food plant and biology.— Adults were reared from larvae eating mature leaves of Mesechites trifida (Jacq.) Mull. Arg. (Apo- cynaceae) [89-SRNP-500 (1$); 92-SRNP- 2490 (29, 16); 92-SRNP-3901 (39); 94- SRNP-9444 (59, 16); 94-SRNP-9529 (16)]. This species is occasionally encountered in groups of 3-7 greenish-black larvae (Fig. 4) feeding side-by-side on the upper or lower side of a single leaf of its herbaceous vine food plant during the mid to late rainy season in ACG dry forest. After con- suming the entire leaf, the group moves up the stem to then consume the next leaf; in larval Lepidoptera, this has been inter- preted a a strategy to remove the visual evidence of leaf damage from the view of avian predators (Heinrich 1993). As is commonplace with species of caterpillars feeding on latex-rich plants, the larvae cut the petiole of the leaf partly through be- fore feeding on the blade, a behavior that reduces the flow of fresh latex from the bitten leaf blade (Dussourd and Eisner 1987). If the feeding group of penultimate or last-instar larvae is molested, they walk off in different directions on the food plant, but within an hour they regroup into the same feeding groups as before. The penultimate instar larva molts into a non-feeding orange-purple morph that, in captivity, wanders on the foliage and litter for about 24 hours before spinning its smooth-walled, dark brown, ovoid co- coon in the litter. The adult emerges 12- 16 days after cocoon spinning (average 14 days, n = 14). A total of 28 wild-caught late instar larvae produced no parasitoids. ACORDULECERINAE This is a large subfamily with many un- described species. Smith's (1990) key cov- JOURNAL OF HYMENOPTERA RESEARCH ers ten genera, four of which occur in Cos- ta Rica (Smith 1995). Many are very small, no more than 3-4 mm in length. The an- tennae are 6-9 segmented, mostly 6-seg- mented and filiform, the eyes are large, oc- cupying much of the head, the mid- and hind tibiae each have a preapical spine, and the hind wing usually has the veins forming the base cells RS and M in a straight or almost straight line. Acordule- cera, as defined by Smith (1990), is by far the largest genus in this subfamily. Anathulea Malaise Anathulca is characterized by the 6-seg- mented antenna, truncate clypeus which is two times or more broader than long, and the long pedicel which is one and one-half times or more longer than broad and nearly as long as the first flagellar seg- ment (Fig. 10). Four species were listed by Smith (1990) from Guatemala and Brazil, but 10 to 20 species probably occur in the Neotropics. This is the first food plant re- cord for a member of this genus. Anathulca bimaculata (Cameron) (Figs. 5, 6, 10, 17, 18, 28, 33, 40) Discussion. — This species has not been treated since its original description. It was described from Guatemala (Cameron 1883). Female. — Length, 6.3-7.5 mm. Antenna black with scape and base of pedicel or- ange yellow. Head black, orange yellow from halfway between ocelli and antennae to clypeus and labrum. Thorax orange yel- low with prescutum (except for sides) and lateral lobes (except for depressed lateral areas) black. Legs orange with apical 4 tar- sal segments black. Abdomen orange with segments 7 or 8 to apex black. Forewing yellow, apex beyond apex of stigma black; veins and stigma yellow, veins black in black apex. Antennal length 1.4X head width; ped- icel 2.7 X longer than broad and 0.6 X length of 3rd segment, 3rd segment 1.5X length of 4th segment; long seta at apex of Volume 12, Number 2, 2003 323 W- JJf 3 M " Xm y 3* ■m p] Pg-v A ' - * • V . ■ wtllF . it J EjMU^r^SiL^ ^* Tf v :JF 11© ■ 23 Figs. 23-26. Head. 23, A liami, front view. 24, A. liamiA dorsal view. 25, A. dashielli, front view. 26, A dashielli, dorsal view. apical segment (Fig. 10). Eyes large and strongly converging below, lower intero- cular distance 0.6 X eye length, upper in- terocular distance 0.9 X eye length (Fig. 17). Head from above strongly narrowing behind eyes; postocellar area 2.1 X broader than long; distances between hind ocelli, hind ocellus to eye, and hind ocellus to hind margin of head as 7:6:8 (Fig. 18). Hind basitarsus 1.2 X longer than length of remaining tarsal segments combined; in- ner hind tibial spur 0.6 X length of hind basitarsus; hind tarsal segments 2-4 each as broad as long. Sheath (Fig. 33) long, length subequal to length of hind tibia; without scopae, in dorsal view broad, rounded to acute apex. Lancet (Fig. 28) with about 32 serrulae, broad at base and tapering to apex, with numerous, closely set annuli slanted toward apex dorsally. Male. — Length, 4.8-6.0 mm. Similar to female except mesoprescutum mostly or- ange and costa and subcosta of forewing more brownish. Genitalia (Fig. 40) with harpe round, slightly broader than long, with long hairs; parapenis acute on me- son; penis valve rounded ventrally, slight- ly concave dorsally, without spines. Distribution. — Costa Rica (Guanacaste), Guatemala, Panama. These are the first re- cords for Costa Rica and Panama. In ad- dition to the rearings from ACG, speci- mens from Costa Rica have been exam- ined from the following: Est. Maritza, 600 m, Lado oeste Volcan Orosi, Prov. Guan., II curso Parataxonomos, Ago 1990, L-N- 326900, 373000; Guanacaste, W side Vol- can Orosi, Est. Maritza, 600 m, 1988; Guanacaste NP, riparian, Oct. 20, 1977; Guanacaste Prov., Santa Rosa National Park, D. H. Janzen, ll-V-l-VI-85, Malaise trap BH-9-0, 13-VII-3-VIII-85, Malaise trap BH-12-C, 16-XI-7-XII-85, Malaise trap BH- 10-C, 24-VIII-14-IX-84, Malaise trap SE- 5-0. Food plant and biology. — The larvae feed on the new shoot tips and shoot epidermis of Hippocratea volubilis L. (Hippocratea- ceae) in the ACG dry forest (209 and 10c? : 01-SRNP-15925 to -15983, 01-SRNP-15940 324 Journal of Hymenoptera Research to -15945, 01-SRNP-15947 to -15949, 01- SRNP-15951 to -15953, and 01-SRNP- 16144 to -16146). The gray-brown early in- star larvae occur in a tight cluster of 8-12 larvae arrayed on the lengthening shoot tips (Figs. 5, 6) of vine shoots growing hor- izontally (in search of insulated areas) across the forest floor in the deep shade of old-growth forest in the second month (July) of the rainy season. In this position, the larvae are extremely cryptic, appear- ing to be a cluster of small leaf buds at the shoot tip end (possibly an example of col- lective mimicry; Pasteur 1982). The larvae collectively eat the very tender shoot tip back down the stem until reaching the woody stem, and then continue back down the stem eating off the green tender outer epidermis, leaving the dead beige woody branch end (Fig. 6). These later in- star larvae, arranged 3-6 around the stem look like dead tissue peeling back from the stem. There is no sign of larvae in the canopy tens of meters above where the mature woody vines have their crowns in the full sun. The prepupal larvae drop or crawl off into the litter and spin solitary beige- brown ovoid cocoons among the dead leaves. There are 14-20 days between co- coon spinning and adult eclosion (average 16 days, n = 31). About 100 wild-caught larvae of all ages produced no parasitoids. Suwatnus Smith This genus is very similar to Acordule- cern, but has 7-segmented antennae (Fig. 11). One species was treated by Smith (1990). Suwatnus nigriceps (Cameron) (Figs. 11, 19, 20) Discussion. — We redescribe this species because the original description is not ad- equate. It was described from "Mexico" by Cameron (1883). Female. — Length, 4.2-4.7 mm. Antenna black. Head black with clypeus and mouthparts yellow orange and apical maxillary palpal segment blackish; apex of mandible reddish brown. Thorax orange; tegula black; upper part of mesopleuron may be blackish. Legs orange white with tarsi and outer surfaces of tibiae black. Abdomen orange ventrally and laterally, black above with central longitudinal or- ange stripe; sheath black. Wings lightly, uniformly infuscated; veins and stigma black. Antennal length 1.1 X head width; 3rd segment 0.8 X length of 4th segment, fla- gellar segments each longer than broad; apical segment without a long seta (Fig. 11). Eyes strongly converging below, lower interocular distance 0.8 X eye length; upper interocular distance subequal to eye length (Fig. 19). Head from above strongly nar- rowing behind eyes; postocellar area 2.0 X broader than long; distances between hind ocelli, hind ocellus and eye, and hind ocel- lus and posterior margin of head as 11:8:19 (Fig. 20). Hind basitarsus 1.3X longer than length of remaining tarsal segments com- bined. Sheath with slender, posteriorly pro- jecting scopae; sheath and lancet illustrated by Smith (1990, figs. 415, 420). Male. — Length, 4.3 mm. Black, with head similar to that of female; thorax black with pronotum brownish; legs white with bases of coxae, tarsi, and stripe on outer surface of mid- and hind femora black; abdomen black with paired brown- ish spots on terga 2-4. Genitalia illustrated by Smith (1988, fig. 418). Food plant and biology. — Three females and one male were reared from larvae feeding on mature leaves of Psidium gua- java L. (Myrtaceae) (99-SRNP-3146) on the ACG interface of dry forest with rainfo- rest. The white, solitary, globular silk co- coons were spun among the litter in the rearing container. Six wild-caught larvae produced no parasitoids. Acordulecera Say This large genus occurs from southern Canada to Argentina. Smith (1990) did not give a species key but listed the 45 de- Volume 12, Number 2, 2003 325 27 9 v ^^P^^^^^^ 28 Figs. 27-31. Female lancets. 27, Adiaclema chigiyae. 28, Anathulea bimaculata. 29, Acordulecera binelli. 30, A Iwmi 31, A. dashielli. scribed species from south of the United States. There are numerous undescribed species, and the size of the genus may ex- ceed 200 species. Most species are small, 4-6 mm long, and all have 6-segmented antennae (Figs. 12-14). Probably more than 20 species occur in Costa Rica, most of which are undescribed. All types of the Neotropical species have been examined by DRS; the following three species do not agree with the three described species from Mexico and Central America, and they are not the same as any of the species described from South America. Because of the new host plant and life historv infor- mation, they are described here. 326 Journal of Hymenoptera Research Acordulecera binelli Smith, new species (Figs. 12, 21, 22, 29, 34) Female.— Length, 5.5 mm. Antenna black with scape orange yellow. Head black, yellow orange below line just in front of front ocellus (Fig. 21); mouthparts yellow orange; apex of mandible reddish brown. Thorax yellow orange with pro- notum white and mesonotal front and lat- eral lobes, tegula, and metanotum black. Legs orange with apical three tarsal seg- ments black. Abdomen and sheath orange. Wings uniformly hyaline; forewing with apical half of costa and stigma, except for black extreme apex, yellow orange, other veins black. Antennal length 1.1 X head width and 2.1 X distance between eyes above; 3rd segment 1.4X length of 4th segment; api- cal segment slightly longer than 4th seg- ment and 4.1 X longer than broad; apical segment with long apical seta; hairs long, mostly longer than width of segments (Fig. 12). Head with scattered punctures. Eyes converging below, lower interocular distance about 0.7X eye length; upper in- terocular distance about 0.9 X eye length (Fig. 21). Head from above strongly nar- rowing behind eyes; postocellar area about 2.0 X broader than long; distances between hind ocelli, hind ocellus and eye, and hind ocellus and posterior margin of head as 7:7:10 (Fig. 22). Distance between antennae subequal to distance between antenna and eye. Hind basitarsus 1.4X length of remaining tarsal segments com- bined. Sheath (Fig. 34) with stout, short lateral scopae, in lateral view scopae rounded and about equal to length of in- ner portion of sheath, in dorsal view with long hairs slightly curved inward. Lancet (Fig. 29) long, with about 24 rounded ser- rulae, serrulae at apex flatter; annuli on apical half with hairs, annuli on basal half with no or only short, indistinct hairs. Male. — Unknown. Holotype. — Female, labeled "Voucher: D. H. Janzen & W. Hallwachs caterpillar (Lepidoptera) database, Area de Conser- vacion Guanacaste, Costa Rica, http:// janzen.sas.upenn.edu," "00-SRNP-9024." Deposited at INBio. Etymology. — This species is named in honor of Rich Binell in recognition of his great enthusiasm for the ACG, love of its roads, and support for the Rincon Rain- forest. Food plant and biology. — Three larvae were found feeding side-by-side on ma- ture leaves of Posoqueria latifolia (Rudge) Roem. & Schult. (Rubiaceae) in lower ACG cloud forest at about 1000 m eleva- tion. The larvae are yellowish at the rear and thorax, with a black head. Two died of disease and one spun a broadly ovoid, pinkish-beige cocoon directly on the sur- face of a leaf in the rearing container. The adult emerged 26 days later. Remarks. — The color and presence of a seta on the apical antennal segment is quite similar to Acordulecera lituratus (Ko- now) described from Amapa, Brazil. In A. lituratus, the abdomen is black above, and the apical seta of the apical antennal seg- ment is extremely long, nearly as long as the apical antennal segment. Characters of the antenna, head, sheath, and lancet should be examined and compared with the illustrations for identification of this species. Acordulecera liami Smith, new species (Figs. 13, 23, 24, 30, 35, 41) Female. — Length, 4.1-4 .4 mm. Antenna black, scape and pedicel more whitish. Head black with clypeus and mouthparts yellow orange; apex of mandible reddish brown. Thorax orange. Legs yellow or- ange with apical 2-3 segments of fore- and midtarsi and entire hind tarsus black. Ab- domen orange, dorsally black with medi- an, longitudinal orange stripe; sheath black. Wings hyaline; veins and stigma black. Antennal length 0.6 X head width and 1.2 X distance between eyes from above; 3rd segment 1.4X length of 4th segment; Volume 12, Number 2, 2003 327 apical segment subequal in length to 4th Guanacaste Province, Santa Rosa National segment and 2.8X longer than broad at its Park, D. H. Janzen, ll-V-l-VI-1985, Mal- greatest width; long seta at apex of apical aise trap SE-8-C (59), Malaise trap H-1-0 segment; hairs equal to or longer than (19), Malaise trap SE-5-0 (19), Malaise width of segments (Fig. 13). Head shining trap H-2-C (1 9 ), Malaise trap SE-7-0 (3 9 ), with scattered punctures. Eyes converging Malaise trap SE-6-C (119), Malaise trap below, lower in terocular distance 0.8 X eye BH-9-0 (19); 1-22-VI-1985, Malaise trap length; upper interocular distance sub- SE-6-C (19). Deposited in INBio, USNM, equal to eye length (Fig. 23). Head from BMNH. above strongly narrowing behind eyes; Etymology. — This species is named in postocellar area 2.0 X broader than long; honor of Liam Binell in recognition of his distances between hind ocelli, hind ocellus great enthusiasm for the ACG, potential as and eye, and hind ocellus and posterior crocodile bait, and support for the Rincon margin of head as 10:7:10 (Fig. 24). Dis- Rainforest. tance between antennae 1.7X distance be- Food plant and biology. — The light green tween antenna and eye. Hind basitarsus larvae feed on the new and expanding subequal in length of length of remaining leaves of the understory shrub Erythroxy- tarsal segments combined. Sheath (Fig. 35) him havanense Jacq. (Erythroxylaceae) in with short, stout scopae, in lateral view the first month of the rainy season in ACG scopa narrow and rounded, shorter than dry forest. They feed solitarily, one on inner portion of sheath, in dorsal view each of the small ovoid leaves, and perch with slightly incurved hairs. Lancet (Fig. -along the margin so as to appear to be part 30) with about 14 pointed serrulae; annuli of the leaf. In some years, such as 1992 and strongly slanted apically toward dorsum 1998, they were extremely abundant, near- and lacking hairs or armature. ly defoliating many of the food plant Male. — Length, 3.1 mm. Similar in color shrubs. The solitary beige to brown ovoid and structure to female except antenna cocoons are spun among the litter. Only with only scape whitish; thorax black with five days transpire between spinning and posterior margin of pronotum yellow or- adult eclosion. No parasitoids were reared ange with small spot at center of meso- from hundreds of wild-caught larvae, scutellum orange; abdomen black with Remarks. — This species is close in color basal sterna pale orange and median Ion- to Acordulecera calceolatus (Konow) de- gitudinal orange stripe only on anterior scribed from Oyapock, Brazil, and A. cer- half of dorsum. Genitalia (Fig. 41) with vicatus (Konow) described from Itaituba, harpes nearly rectangular, parapenis Brazil, all sharing the presence of a long rounded mesally, and penis valve nearly seta on the apical antenna] segment. In A. straight ventrally, slightly concave dorsal- calceolatus, the mesoprescutum and meso- ly, narrowly rounded at apex, and with- scutellum are dark orange, the abdominal out spines. dorsum is black, the apical 2-3 tarsal seg- Holotype. — Female, labeled "Voucher: ments are black, and the sheath has more D. H. Janzen & W. Hallwachs caterpillar slender, projecting lateral scopae. In A. cer- (Lepidoptera) database, Area de Conseer- vicatus, the anterior half of the mesopresc- vacion Guanacaste, Costa Rica, http:// turn, apical half of the mesoscu tell urn, and janzen.sas.upenn.edu, 98-SRNP-1825." the metascutellum are black, the abdomen Deposited at INBio. is orange with the apical two segments Paratypes. — COSTA RICA: Same labels black, and the tibiae and tarsi are black, as holotype, except numbers, 98-SRNP- Identification of A. liami should be based 1823 (19), 98-SRNP-1826 (19), 98-SRNP- on comparison of the specimens with the 1827 (19), 98-SRNP-1824 (1 6 ); Costa Rica, illustrations. 328 Journal of Hymenoptera Research ^■■M\\\\^%\ Figs. 32-36. Female sheaths, lateral and dorsal views. 32, Adiaclema chigiyae. 33, Anathulea bimaculata. 34, Acordulecera binelli. 35, A. liami. 36, A. dashielli. Acordulecera dashielli Smith, new species (Figs. 7, 14, 25, 26, 31, 36, 42) Female. — Length, 4.8-5.2 mm. Antenna black with scape yellow orange. Head black with area below line just above an- tennae and mouthparts yellow orange (Fig. 25); apex of mandible reddish brown. Thorax orange with tegula black; blackish area laterally on each mesonotal lateral lobe. Legs orange with tibiae and tarsi black. Abdomen orange with apical two terga and sheath black. Wings moderately and uniformly infuscated; veins and stig- ma black. Antennal length subequal to head width and 1.9X distance between eyes from above; 3rd segment slightly shorter than 4th segment; apical segment 0.8 X length of 4th segment and 2.5 X longer than broad at its greatest width; apical segment without long seta at apex; hairs equal to or shorter than width of segments (Fig. 14). Head shining, nearly impunctate. In- ner margin of eyes subparallel, hardly converging below, lower and upper inter- Volume 12, Number 2, 2003 329 ocular distances subequal to slightly shorter than eye length (Fig. 25). Head from above strongly narrowing behind eyes; postocellar area 2.1 X broader than long; distances between hind ocelli, hind ocellus and eye, and hind ocellus and pos- terior margin of head as 10:8:11 (Fig. 26). Distance between antennae 2.8 X distance between antenna and eye. Hind basitarsus 1.3X longer than length of remaining tar- sal segments combined. Sheath (Fig. 36) with long, narrow projecting scopae, in lateral view slender and rounded, much longer than inner portion of sheath; in dorsal view, scopae forcepslike, curving inward at apices. Lancet (Fig. 31) short, with about 14 serrulae, with slight dorso- apical notch at apex. Basal 8 serrulae broad and rounded and separated by nar- row notch; serrulae beyond 8 small and becoming indistinct toward apex. Annuli strongly curved, ventral half without hairs, dorsal half with fine hairs. Male. — Length, 3.5-4.2 mm. Similar to female except bases of tibiae paler and apical maxillary palpal segments blackish. Genitalia (Fig. 42) with harpe nearly tri- angular, lateral and apical margins round- ed, inner margin straight; parapenis rounded mesally; penis valve elongate, rounded at apex, with ventral and lateral spines. Holotypc. — Female, labeled "Voucher: D. H. Janzen & W. Hallwachs caterpillar (Lepidoptera) database, Area de Conser- vacion Guanacaste, Costa Rica, http:// janzen.sas.upenn.edu, 96-SRNP-6815." Deposited at INBio. Paratypes. — COSTA RICA: Specimens labeled as above, all beginning with 96- SRNP-: 6811 (19); 6815 (19); 6821 (16); 6822 (let); 6824 (19); 6824.1 (19); 6825 (19); 6826 (19); 6829 (19); 6948 (19); 6948.1 (19); 6948.2 (19); 6948.3 (19); 6948.4 (19); 6948.5 (19); 6948.7 (19); 6948.8 (19); 6948.10 (19); 6948.11 (19); 6948.12 (19); 6948.13 (19); 6948.14 (19); 6948.15 (19); 6948.16 (19); 6948.17 (19); 6948.18 (19); 6948.19 (19); 6948.20 (19; 6948.21 (lc5); 6948.22 (16); 6948.23 (16); 6948.24 (19); 6948.25 (19); 6948.26 (19); 6948.27 (19); 6984.6 (19); 6984.9 (19); Guanacaste Province, Santa Rosa National Park, D. H. Janzen, ll-V-l-VII-1985, Mal- aise trap H-2-C (1 9 ), ll-V-l-VI-1985, Mal- aise trap SE-8-C (16), Malaise trap H-2-C (16), Malaise trap SE-7-0 (lc?), Malaise trap SE-6-C (36), 1-22-VI-1988, Malaise trap SE-8-C (1 9 ), 14-IX-5-X-1985, Malaise trap SE-7-0 (19), 22-VI-13-VII-1985, Mal- aise trap H-4-C (19). Deposited in INBio, USNM, BMNH. Etymology. — This species is named in honor of Dashiell Binell in recognition of his great enthusiasm for the ACG, love of its beaches, and support for the Rincon Rainforest. Food pilant and biology. — The larvae feed on the newly expanded leaves of the wood vine Arrabidaca patellifera (Schltdl.) Sandwith (Bigoniaceae) in the ACG dry forest in the first month of the rainy sea- son. The green and yellow larvae (Fig. 7) feed on the upper side of the nearly hori- zontal leaves. The early instars feed as a group, but, by the penultimate instars, they have separated to feed solitarily. The larvae drop to the ground to spin a very tough brown ovoid cocoon in the leaf lit- ter. Between seven and 17 days lapsed be- tween spinning and adult eclosion, but some larvae remained as dormant prepu- pae in their cocoons throughout the five remaining months of the rainy season and the entire six month dry season before dy- ing, apparently because they did not get the right cue to eclose under laboratory conditions. This species has been found feeding on its common food plant only at the beginning of the rainy season, and it may well be univoltine in nature. No par- asitoids were produced from 50 wild- caught larvae of all instars. Remarks. — This species is very similar in coloration to Acordulecera ricatus (Konow) described from Peru, and both lack the seta on the apical antennal segment. How- ever, A. ricatus has a hyaline forewing 330 Journal of Hymenoptera Research Figs. 37-42. Tarsal claws and male genitalia, ventral view of left half of genital capsule on left, lateral view of penis valve on right. 37, Female tarsal claw of Adiaclema chigiyae. 38, Male tarsal claw of A. chigiyae. 39, Male genitalia of A. chigiyae. 40, Male genitalia of Aimtluilca bimaculata. 41, Male genitalia of Acordulecera liami. 42, Male genitalia of A. dashielli. Volume 12, Number 2, 2003 331 with the base and apex of the wings black, whereas A. dashielli has uniformly lightly blackish wings. The sheath is distinctive for A. dashielli, but other characters need to be compared with the illustrations for determination. ACKNOWLEDGMENTS This part of the ACG caterpillar inventory has been supported by NSF grants BSR 90-24770, DEB 93- 06296, DEB-94-00829, DEB-97-05072, and DEB- 0072730, by taxonomists of the Smithsonian Institu- tion and the Systematic Entomology Laboratory of the U. S. Department of Agriculture, and by financial, administrative, and logistic support from Costa Rica's INBio, the government of Costa Rica, the Area de Conservacion Guanacaste, and CONICYT of Costa Rica. Many individuals have supported the devel- opment of all stages of the project in a multitude of ways. We specially thank the following people for caterpillar hunting and husbandry: W. Hallwachs, R. Moraga, G. Sihezar, G. Pereira, L. Rios, M. Pereira, O. Espinosa, E. Cantillano, M. Pereira, R. Franco, H. Ra- mirez, F. Chavarria, M. M. Chavarria, E.Olson, C. Moraga, P. Rios, C. Cano, D. Garcia, F. Quesada, E. Araya, E. Guadamuz, R. Espinosa, R. Blanco, A. Gua- damuz, D. Perez, R. Blanco, F. Chavarria, C. Camar- go, H. Kidono, A. Masis, and W. Haber. We thank the following for allowing examination of Costa Rican collections: J. Ugalde Gomez, Instituto Nacional de Biodiversidad, Santo Domingo de He- redia, Costa Rica; I. Gauld, The Natural History Mu- seum, London, U.K.; P. Hanson, Universidad de Cos- ta Rica, San Jose; J. Longino, The Evergreen College, Olympia, Washington, U.S.A.; and H. Hespenheide, University of California, Los Angeles, California, U.S.A. Cathy Apgar, Systematic Entomology Labo- ratory, U. S. Department of Agriculture, took the pho- tographs of the ovipositors, heads, and antennae, and arranged and labeled the plates. We appreciate the reviews by N. Schiff, U.S. Forest Service, Stoneville, MS, and S. Scheffer and E. E. Grissell, Systematic En- tomology Laboratory, U. S. Department of Agricul- ture, Beltsville, MD, and Washington, DC, respective- ly- LITERATURE CITED Burns, J. M. and D. H. Janzen. 2001. Biodiversity of pyrrhopygine skipper butterflies (Hesperidae) in the Area de Conservacion Guanacaste, Costa Rica, journal of the Lepidopterists' Society 55: 15- 43. Cameron, P. 1883. Hymenoptera, Tenthredinidae— Chrysididae. In Godman and Salvin, Biologia Centrali-Americana, Vol. 1, 486 pp. Dussourd, D. E. and T. Eisner. 1987. Vein-cutting be- havior: Insect counterploy to the latex defense of plants. Science 237: 898-901. Flores, C, J. Ugalde, P. Hanson, and I. Gauld. 2000. The biology of perreyiine sawflies (Hymenop- tera: Pergidae) of the Perreyia genus-group, pp. 258-266. In Austin, A. D„ and M. Dowton, eds. Hymenoptera Evolution, Biodiversity and Biological Control. CS1RO Publishing, Collingwood, Austra- lia. 468 pp. Heinrich, B. 1993. How avian predators constrain cat- erpillar foraging, pp. 224-247. In Stamp, N. E. and T. M. Casey, eds. Caterpillars, Ecological and Evolutionary Constraints on Foraging. Chapman & Hall, New York. 587 pp. Janzen, D. H. 2000. Costa Rica's Area de Conserva- cion Guanacaste: a long march to survival through non-damaging biodevelopment. Biodi- versity 1: 7-20. Janzen, D. H. In press. Ecology of dry forest wildland insects in the Area de Conservacion Guanacaste, northwestern Costa Rica, pp. 1-44. In Frankie, G. W., A. Mata, and S. B. Vinson, eds. Biodiversity Conservation in Costa Rica: Learning the Lessons in Seasonal Dry Forest. University of California Press, Berkeley. Janzen, D. H., A. K. Walker, J. B. Whitfield, G. Del- vare, and I. D. Gauld. 2003. Host-specificity and hyperparasitoids of three new Costa Rican spe- cies of Microplitis Foerster (Hymenoptera: Bra- conidae: Microgastrinae), parasitoids of sphingid caterpillars. Journal of Hymenoptera Research 12: 42-76. Kimsey, L. S. and D. R. Smith. 1985. Two new species, larval descriptions and life history notes of some Panamanian sawflies (Hymenoptera: Argidae, Tenthredinidae). Proceedings of the Entomological Society of Washington 87: 191-201. Pasteur, G. 1982. A classificatory review of mimicry- systems. Annual Review of Ecology and Systematica 13: 169-199. Schauff, M. E. and D. H. Janzen. 2001. Taxonomy and ecology of Costa Rican Euplectrus (Hymenoptera: Eulophidae), parasitoids of caterpillars (Lepidop- ter a). Journal of Hymenoptera Research 10: 181-230. Smith, D. R. 1988. A synopsis of the sawflies (Hy- menoptera: Symphyta) of America south of the United States: introduction, Xyelidae, Pamphili- idae, Cimbicidae, Diprionidae, Xiphydriidae, Sir- icidae, Orussidae, Cephidae. Systematic Entomol- ogy 13: 205-261. Smith, D. R. 1990. A synopsis of the sawflies (Hy- menoptera, Symphyta) of America south of the United States: Pergidae. Revista Brasileira Ento- mologia 34 ( 1 ): 7-200. Smith, D. R. 1995. 6, The sawflies ^nd woodwasps, pp. 157-177. In Hanson, P. F. and I. D. Gauld, eds. The Hymenoptera of Costa Rica. Oxford Uni- versity Press, Oxford, U.K. 893 pp. 332 Journal of Hymenoptera Research Smith, D. R. 2003a. A synopsis of the sawflies (Hy- menoptera: Symphyta) of America south of the United States: Tenthredinidae (Nematinae, Het- erarthrinae, Tenthredininae). Transactions of the American Entomological Society 129: 1-45. Smith, D. R. 2003b. A synopsis of the sawflies (Hy- menoptera: Symphyta) of America south of the United States: Tenthredinidae (Allantinae). Jour- nal of Hymenoptera Research 12: 148-192. Smith, D. R. and D. H. Janzen. 2003. Food plants and life histories of sawflies of the family Argidae (Hymenoptera) in Costa Rica, with descriptions of two new species. Journal of Hymenoptera Re- search 12: 193-208. Smith, D. R. and W. W. Middlekauff. 1987. Suborder Symphyta, pp. 618-649. /;; Stehr, F. W., ed. Im- mature Insects. Kendall Hunt Publishing Com- pany, Dubuque, Iowa. 754 pp. J. HYM. RES. Vol. 12(2), 2003, pp. 333-345 Review of the Southeastern Asian Sawfly Genus Eusunoxa Enslin (Hymenoptera: Tenthredinidae) David R. Smith and M. S. Saini (DRS) David R. Smith, Systematic Entomology Laboratory, PSI, Agricultural Research Service, U.S. Department of Agriculture, % National Museum of Natural History, Smithsonian Institution, Washington, DC 20560-0168, USA, email: dsmith@sel.barc.usda.gov; (MSS) Department of Zoology, Punjabi University, Patiala, 147002 India email: saini20@glide.net. in Abstract. — Nine species of Eusunoxa Enslin are keyed, described, and illustrated: E. buchi Togashi from the Philippines; E. ceylonica Malaise from Sri Lanka and southern India; E. formosana Enslin from Taiwan; E. ebena, n. sp., from Indonesia (Kalimantan); and E. nigriceps (Rohwer), E. auri- cauda, n. sp., E. lissofrons, n. sp., E. punctata, n. sp., and E. semipunctata, n. sp., from southern India. Eusunoxa subg. Asunoxa Wei is a new synonym of Eusunoxa Enslin, and Eusunoxa In- diana Haris is a new synonym of E. ceylonica Malaise. Eusunoxa Enslin, a small genus of the subfamily Allantinae, is known from In- dia, Sri Lanka, Taiwan, Indonesia, and the Philippines. Malaise (1963) also men- tioned Burma in the distribution of the ge- nus, but we have not seen specimens from Burma, although one species was collected during extensive surveys of northeastern India by the junior author. Specimens are not common, but significant collections from southern India and Sri Lanka have prompted this review. We treat nine spe- cies, five of which are described as new. Food plants are not known. Acronyms used are: BMNH = The Nat- ural History Museum, London, UK; DEI = Deutsches Entomologisches Institut, Eberswalde, Germany; PUNJ = Punjabi University, Patiala, India; USNM = Na- tional Museum of Natural History, Smith- sonian Institution, Washington, DC, USA. Abbreviations used are: OOL = distance between eye and hind ocellus; POL == dis- tance between hind ocelli; OCL = distance between hind ocellus and posterior mar- gin of head. Eusunoxa Enslin Eusunoxa Enslin 1911: 99. Type species: Eusunoxa formosana Enslin, by original designation. Eusunoxa subg. Asunoxa Wei 1997: 88. Type spe- cies: Eusunoxa ceylonica Malaise, by original designation. New synonymy. Description. — Antenna filiform, some- times slightly incrassinate in middle and flagellomeres slightly serrate, length 2X or less head width; 1st and 2nd segments each longer than broad; 3rd segment sub- equal to or slightly longer than 4th; apical segments without ventral membranous areas. Head from above strongly narrow- ing behind eyes; postgenal carina absent; inner margins of eyes slightly converging downwards; clypeus subtruncate to very shallowly circularly emarginate (Fig. 2); labrum short, about 2x broader than long; malar space linear or very narrow to slightly more than half diameter of front ocellus. Head with frontal area about same height as eyes; supraantennal tubercles and frontal ridges indistinct; median fovea a shallow groove; supraan- tennal pits distinct, about 2x their di- 334 Journal of Hymenoptera Research Figs. 1-4. Front view of head. 1, Eusunoxa punctata. 2, E. semipunctata. 3, Eusunoxa ceylonica. 4, E. lissofrons. ameter from antennal socket and each with a small conical projection at center; lateral furrows deep and slightly diverg- ing posteriorly; post-, inter-, and circu- mocellar furrows indistinct to absent. Ep- icnemium present as a flat sclerite sepa- rated from mesepisternum by a suture. Hind coxae lengthened, end of hind bas- itarsus reaching to and beyond apex of abdomen; hind basitarsus (Figs. 5-8) lon- ger than length of remaining tarsal seg- ments combined, broad and laterally flat- tened, with outer surface concave; tarsal pulvilli present only on segments 3 and 4. Tarsal claws with single inner tooth, slightly shorter than outer tooth, and po- sitioned lateral to outer tooth; basal lobe present or absent. Hind wing with cell Rs absent, cell M present. Thorax and ab- domen impunctate, shining. Discussion. — The presence of an epicne- mium, long hind legs with the hind basi- tarsus reaching to or beyond the apex of the abdomen, the large, laterally com- pressed hind basitarsus, and presence of one closed cell in the hind wing are dis- tinctive for Eusunoxa. Eusunoxa may be separated from other Allantinae by Mal- aise's (1963) key. It is closest to Megabeleses Takeuchi, Beleses Cameron, and Nesotax- onus Rohwer, but Megabeleses has the inner tooth of the tarsal claws much longer than the apical one, much longer antennae (more than twice the head width), and the head from above slightly dilated behind the eyes. The other genera have a normal (cylindrical) hind basitarsus and have ven- tral membranous areas in the apical four antennal segments. The broad, laterally compressed hind basitarsus is reminiscent of the same in the Holarctic genus Craesus Leach (Nematinae). Volume 12, Number 2, 2003 335 Wei (1997) proposed the subgenus Asu- noxa for those species lacking a basal lobe to the tarsal claws (Fig. 22). This included all known species except E. fonnosana En- slin, which has a basal lobe (Fig. 21) and was the only described species in the typ- ical subgenus (Wei [1997] separated some unnamed species, designated as "£. spp.," and we add one, E. ebena, n. sp., here). Other than the basal lobe, £. formosana shares all other characters with other spe- cies, including similarities in the genitalia, and shares the presence of punctures on the frons with some other species. Because the presence or absence of a basal lobe is of suspicious importance, we do not see the necessity of recognizing subgenera at present. KEY TO SPECIES OF EUSUNOXA 1 Tarsal claws with basal lobe (Fig. 21); front of head with large, closely set punctures separated by narrow ridges, as in Fig. 1 2 - Tarsal claws without basal lobe (Fig. 22); punctures on front of head as above, widely spaced with broad, flat, shining interspaces (Fig. 2) or absent (Figs. 3, 4) 3 2 Head, thorax, and abdomen orange yellow; antenna black with basal segments orange; wings yellow hyaline, costa and subcosta yellowish, stigma brownish with margins yellowish; hind basitarsus 4.8 X longer than maximum width (lancet in Fig. 9; male genitalia in Figs. 13, 14) fonnosana Enslin - Head (except for reddish-brown clypeus), thorax and abdomen black; wings lightly black- ish with costa, subcosta and stigma black; hind basitarsus 3.8x longer than maximum width ebena, n. sp. 3 Thorax black or black with pronotum, tegula, and part of meso- and metanotum dark orange 4 - Thorax entirely orange 4 Pronotum, tegula, V-shaped mark on mesoprescutum, mesoscutellum, and metanotum orange (head shining, impunctate, as in Fig. 4; male genitalia with penis valve curved, in Fig. 20) nigriceps (Rohwer) (6) - Thorax entirely black 5 5 Abdomen orange except for blackish basal plates (frons sparsely punctured, with shining interspaces, as in Fig. 2; male genitalia in Fig. 17) auricauda, n. sp. - Abdomen black 6 6 Legs entirely black; head shining, impunctate (Fig. 4); penis valve oval (Fig. 19) lissofrons, n. sp. - Legs with fore- and midtibiae and tarsi white; head with sparse, widely separated punc- tures with shining interspaces on frons (Fig. 2); penis valve elongate, constricted at center (Fig. 18) semipunctata, n. sp. (6) 7 Clypeus and supraclypeal area dark orange; first antennal segment or basal 2 antennal segments and base of 3rd segment may be orange; head with Frons densely punctate, punctures close together, separated bv narrow ridges, without shining interspaces (Fig. 1) 8 - Head and antenna black; frons impunctate, shining, or with widely spaced punctures separated by broad, shining interspaces (Figs. 2—4) 9 8 Hind femur basally orange, apical half or more black; basal 2 antennal segments and base of 3rd segment orange; hind basitarsus about 1.4X length of remaining tarsal segments combined; Philippines buchi Togashi - Hind femur entirely black; only 1st antennal segment dark orange; hind basitarsus about 336 Journal of Hymenoptera Research 1.7X length of remaining tarsal segments combined (Fig. 8) (lancet in Fig. 11); India punctata, n. sp. 9 Femora blackish, only extreme bases may be orange; frons with distinct, widely spaced punctures, separated by broad, shiny interspaces (Fig. 2) (lancet in Fig. 12) semipunctata, n. sp. (?) - Fore- and midfemora orange, hind femur entirely or partly black; frons with few minute punctures or impunctate, shining (Figs. 3, 4) 10 10 Most of hind femur black, only extreme base orange; extreme apices of fore- and midtibiae black (lancet in Fig. 10; male genitalia in Figs. 15, 16) ceylonica Malaise - Apical half of hind femur black; tibiae orange (after Rohwer 1915) . . . nigriceps (Rohwer) (9) Ensiinoxa auricauda Smith and Saini, new species (Fig. 17) Female. — Unknown . Male. — Length, 6.5 mm. Antenna, head, and thorax black, abdomen orange with black basal plates, basal plates darkest black laterally. Legs with apices of coxae and trochanters on upper surfaces white; fore- and midlegs with apices of femora and tibiae and tarsi entirely white; hind tibia black with white streak on inner sur- face of basal half. Wings lightly infumated towards apex; stigma, costa, and rest of venation dark brown to black. Antenna slightly compressed and flagellomeres ser- rate on underside with blunt projections on segments at apex below; length 2X head width; pedicel 2x its apical width; scape as long as its apical width; segment 3 slightly longer than segment 4. Clypeus with shallow V-shaped emargination; lower interocular distance 0.8X eye length; OOL:POL:OCL 1.00:0.92:0.80; postocellar area subconvex, 3X broader than long at its maximum width; head with sparse, shallow punctures with flat shining interspaces on and around the frontal area (similar to Fig. 2), posterior or- bits and postocellar area impunctate. Hind basitarsus 1.3X length of remaining tarsal segments combined, about 4.0 X longer than maximum width. Tarsal claws with- out basal lobe. Genital capsule similar to Fig. 15; penis valve in Fig. 17. Holotype. — 6, India, Kerala: Munar, 4000 ft., 22.9.1995, collection M.S. Saini (PUNJ). Distribution. — India (Kerala). Etymology. — The species name is based on the orange abdomen. Remarks. — The black head and thorax with the contrasting orange abdomen are unlike other species of Eusunoxa. This col- oration and the sparsely punctured head (similar to Fig. 2) will distinguish this spe- cies from other species in the genus. Eu- sunoxa auricauda is most similar to the male of E. semipunctata, with which it shares the punctate head, similar hind basitarsus, and very similar male genitalia (Figs. 17, 18). However, the absence of col- or variation in the large series of males of E. semipunctata and the slight difference in shape of the penis valve (less constricted at its center than that of E. semipunctata), provide evidence that E. auricauda is a sep- arate species. Eusunoxa buchi Togashi Eusunoxa buchi Togashi 1981: 419, figs. 21-24, 26-28.— Wei 1997: 89 (in subg. Asunoxa; in key). Female. — Unknown. Male. — Length, 6.0 mm. Antenna black with segments 1, 2, and basal half of 3rd segment orange. Head black with clypeus, mouthparts, and supraclypeal area or- ange. Thorax and abdomen orange. Fore- and midlegs orange with apical 4 tarsal segments infuscate; hind leg with coxa and trochanter orange, femur black except Volume 12, Number 2, 2003 337 8 Figs. 5-8. Apex of hind tibia and hind tarsus. 5, Eusunoxa ceylonica. 6, E. lissofrons. 7, E. semipunctata. 8, E. punctata. basal quarter orange, tibia black with bas- al third white, and tarsus black. Wings very lightly, uniformly infuscated; veins and stigma black. Antennal length 1.7X head width; 3rd segment subequal in length to 4th. Lower interocular distance 0.7X eye length. Postocellar area 2.5 X broader than long; clypeus subtruncate; malar space linear; OOL:POL:OCL == 1.00: 0.80:0.80. Front of head densely punctate, punctures large and close together, sepa- rated by narrow ridges (similar to Fig. 1); postocellar and postocular areas nearly impunctate, shining. Hind basitarsus 1.4X length of remaining tarsal segments com- bined, 4.2X longer than maximum width. Tarsal claws without basal lobe. Types.— Holotype 6, "Sept. 7, 1961, Mantalingajan, Pinigisan, 600 m, Palawan Is., Noona Dan Expedition," in Universi- tets Zoologiske Museum, Copenhagen, Denmark. Ten paratype males, dated Sept. 6-24, 1961, also in type series. None ex- amined. Specimen examined.— PHILIPPINES: Pa- lawan, Mantalingajan, Pinigisan, 600 m, 23 338 Journal of Hymenoptera Research Sept. 1961, Noona Dan Exp. 61-62 (1 6\ BMNH). From same locality as type series. Distribution. — Philippines (Palawan). Remarks. — Eusunoxa buchi is the only species in the genus known from the Phil- ippines. It shares the densely punctate head with E. formosana, E. ebena, and £. punctata. The mostly black coloration and lack of a basal lobe on the tarsal claws sep- arate it from E. formosana and E. ebena, and the dark orange basal antennal segments and only the apical half of the hind femur black separate it from E. punctata from In- dia, which has the hind femur black and only the first antennal segment dark or- ange. Eusunoxa ceylonica Malaise (Figs. 3, 5/l0, 15, 16, 22) Netrocerus nigriceps Enderlein 1920: 371. Preoc- cupied in Eusunoxa by Eusunoxa nigriceps (Rohwer 1915). Eusunoxa ceylonica Malaise 1932: 147. — Smith 1982: 120,' figs. 2, 8, 14 (Sri Lankan records; female lancet, male genitalia, tarsal claw fig- ured; syn.: N. nigriceps Enderlein). — Saini and Deep 1994: 50 (India, Tamil Nadu; as "cevlonica").—Wei 1997: 89 (in subg. Asu- noxa; in key). Eusunoxa indiana Haris 2000: 299. New synon- ymy. Female. — Length, 6.5-7.5 mm. Antenna and head black; clypeus black to partially orange on side or anterior margins. Tho- rax and abdomen orange. Foreleg orange with extreme apex of tibia and apical 4 tarsal segments blackish; midleg orange with apical half of tibia and entire tarsus black; hind leg with coxa, trochanters, and basal third of femur orange, apical two- thirds of femur and entire tibia and tarsus black. Wings uniformly infuscated; veins and stigma black. Antennal length 1.7X head width, 3rd segment slightly longer than 4th. Lower interocular distance sub- equal to eye length; clypeus subtruncate; postocellar area 4.5 X broader than long; malar space nearly linear; OOL:POL:OCL = 1.00:0.90:0.35. Head shining, impunctate except for small widely scattered punc- tures on frontal area (Fig. 3). Hind basi- tarsus (Fig. 5) 1.5X length of remaining tarsal segments combined; about 3.6 X lon- ger than maximum width. Tarsal claws without basal lobe. Lancet in Fig. 10, with about 17 serrulae; apex roundly acute. Male. — Length, 5.5-6.0 mm. Similar to female but with terga 2 to apex mostly in- fuscate to blackish and clypeus black or partly to almost entirely orange. Genitalia in Figs. 15, 16. Types. — The holotype 9 of Netrocerus ni- griceps Enderlein, is in the Polish Academy of Sciences, Warsaw (examined). The ho- lotype 9 of £. ceylonica Malaise is in the BMNH, labeled "Colombo, Ceylon"; a paratype from the Swedish Museum of Natural History was examined. The holo- type of £. indiana Haris is at the BMNH (examined), a 6 labeled: "Holotype," "In- dia, Tamil N., Madras, 2.XI.79. Boucek," "Holotype Eusunoxa indiana sp. n. det. A. Haris 1999," "B.M. type Hym. 1.851." Specimens examined. — INDIA: Poonmu- di, 18-9-95 (1 9); South India, Anamalai Hills, Cinchona, 3500 ft., V-1964, P.S. Na- than (2 9), V-1966 (2 9), X-1966 (1 9), V- 1967 (1 9), V-1968 (1 9); Nilgiri Hills, Sin- gara, S. India, 3400 ft., June 1954, col: P.S. Nathan (1 9); South India, Kerala State, Trivandrum Dt., Poonmudi Range, May 1971, 3000 ft. elevation, T.R.S. Nathan (2 9); South India, Pondicherry State, Kari- kal, P.S. Nathan (1 9 ), same but with date, X-'62 (1 9). SRI LANKA: Western Prov- ince, Central Province, Southern Province, Eastern Province, North Western Prov- ince, Northern Province (see Smith 1982 for records); Ceylon, Trincomalee (1 9, BMNH). Distribution. — India (Kerala; Tamil Nadu); Sri Lanka. Remarks. — Eusunoxa indiana Haris is the male of £. ceylonica. It is typical except for the orange clypeus. We have not seen oth- er specimens with an entirely orange clyp- eus; however, among the material we ex- amined, coloration varies from black to Volume 12, Number 2, 2003 339 %^^ l- f^ ■I ' 12 Figs. 9-12. Female lancets. 9, Eusunoxa formosana. 10, E. ceylonica. 11, E. punctata. 12, E. semipunctata. brownish, to some partially orange at the imen and glued on cardboard on the same base and laterally. pin) is more punctate than most spuci- The specimen from Ceylon at BMNH is mens, like E. ceylonica, but the head (off the spec- We have not seen variation in color of 340 Journal of Hymenoptera Research the thorax and abdomen, both are mostly orange, with the abdomen of the male more blackish dorsally. Eusunoxa ceylonica appears to be most similar to E. nigriceps, both sharing the shining head with few widely spaced punctures, and the female is difficult to distinguish from that of E. nigriceps based on Rohwer's (1915) de- scription of the latter. The male paratype of E. nigriceps, however, is mostly black, and the genitalia differ (Figs. 16, 20) from that of E. ceylonica. In Sri Lanka, this species occurs in both open and wooded areas in dry and wet zones with annual rainfall ranging from 660 to 1952 mm (Smith 1982). Eusunoxa ebena Smith and Saini, new species Female. — Length, 8.0 mm. Antenna black with segments 1, 2, and basal half of 3 dark orange; head black with clypeus reddish brown. Thorax and abdomen black. Foreleg dark brown with coxa and apical 3 tarsal segments black; midleg with coxa black, femur and tibia dark brown (tarsus missing); hind leg black with basal half of tibia white (apical two tarsal segments missing). Wings very lightly, uniformly infuscated; veins and stigma black. Antennal length 1.6x head width, 3rd segment 1.1 X longer than 4th. Lower interocular distance 0.8 X eye length; clypeus very shallowly circularly emarginated; postocellar area 2.7X broad- er than long; malar space distinct, slightly more than half diameter of front ocellus; OOL:POL:OCL = 1.0:1.2:1.0. Head with frontal area densely punctured, punctures deep, close together, separated by narrow ridges (similar to Fig. 1); clypeus impunc- tate, shining; postocellar area and head be- hind eyes sparsely punctate, shining. Hind basitarsus extremely flat and concave; 3.8 X longer than maximum width (length to remaining tarsal segments unknown since apical tarsal segments missing). Tar- sal claws with basal lobe (Fig. 21). Male. — Unknown. Holotype. — $, "Borneo, Pontianak, F. Muir" (BMNH). Distribution. — In present-day Kaliman- tan, Indonesia. Etymology. — The name is based on the almost entirely black coloration of this species. Remarks. — Eusunoxa ebena and E. formo- sana are the only species that have a basal lobe on the tarsal claws. Also, both share the densely punctate frons. The coloration of the two are, however, entirely different. Eusunoxa ebena is almost entirely black and £. formosana almost entirely orange yellow. The hindbasitarsus is extremely flat, thin, and concave on its outer surface, similar to that of E. formosana. Because of the fragility of the specimen, the lancet was not examained; however, from the small apical part visible the serrulae ap- pear to be rather deep similar to E. for- mosana (Fig. 9). Eusunoxa formosana Enslin (Figs. 9, 13, 14, 21) Eusunoxa formosana Enslin 1911: 99. — Wei 1997: 88 (in subg. Eusunoxa; in key). Female. — Length, 7.5-8.0 mm. Antenna, head, thorax, and abdomen yellowish or- ange, with antennal segments 5-9, apical third to half of hind femur, and most of hind tarsus black; apical segments of hind tarsus paler, more dark orange. Wings yel- low hyaline, costa and subcosta yellowish, stigma brownish with margins yellowish, rest of venation dark brown. Antenna fi- liform, very slightly incrassinate in mid- dle; antennal length 1.7X head width; 3rd segment 1.1 X length of 4th; pedicel as long as its apical width; scape slightly lon- ger than its apical width. Clypeus sub- truncate; malar space distinct but much less than half diameter of front ocellus; lower interocular distance 0.8X eye length. OOL:POL:OCL = 1.00:1.00:0.83, postocellar area convex, 2X broader than long at its maximum breadth. Head with frons densely punctured, punctures deep, Volume 12, Number 2, 2003 341 close together, separated by narrow ridges (similar to Fig. 1); postocellar area and head behind eyes sparsely punctate, shin- ing. Tarsal claws with inner tooth shorter than apical one, basal lobe distinct (Fig. 21). Hind basitarsus extremely flat and concave on its outer surface, 1.7X length of remaining tarsal segments combined; 4.8 X longer than maximum width. Lancet in Fig. 9, serrulae deep with almost per- pendicular anterior margin. Male. — Length, 7.5 mm. Similar in color and structure to female. Genitalia in Figs. 13, 14. Type. — Described from a female from "Formosa," in the Zoologisches Staats- sammlung, Miinchen, Germany; not ex- amined. Specimens examined. — TAIWAN: Taihor- in, Formosa, H. Sauter, 7.VI.1911 (1 9, USNM); Kankau (Koshun) Formosa, H. Sauter, VI-1912 (1 9, USNM), same except VII-1912 (1 9, DEI); PingTung Co., Rent- ing Bot. Garden, subtropical forest, July 10-14, 1980, 260 m, D.R. Davis (1 8, USNM); Formosa, Sauter, Taihorinaho, 1090.VIII (1 9, BMNH). Distribution. — Taiwan. Remarks. — We did not examine the type since the description and uniqueness of this species is sufficient for its identity. This and £. ebena are the only species of Eusunoxa with a basal lobe on the tarsal claws and one of the few species (other than £. ebena, E. punctata and £. buchi) with a densely punctate frons. The almost en- tirely orange coloration is also distinctive. The female ovipositor, especially the shape of the serrulae (Fig. 9), differs from other species of Eusunoxa. Male genitalia (Figs. 13, 14) differ from other species by the longer parapenis of the genital cap- sule. Eusunoxa lissofrons Smith and Saini, new species (Figs. 4, 6, 19) Female. — Unknown. Male. — Length, 5.5-6.0 mm. Head and thorax black, abdomen and fore- and mid- legs brownish, hind leg black. Wings in- fumated apically from base of stigma, bas- al half clear; venation dark brown to black. Antennal length 1.7X head width, slightly compressed and only feeblv in- crassinate in middle; pedicel 2X longer than its apical width; scape as long as broad at apex; 3rd segment 1.1 X longer than 4th. Clypeus subtruncate; lower in- terocular distance 0.9X eye length; malar space distinct but much less than half di- ameter of front ocellus; OOL:OCL:POL = 1.00:1.00:0.70; postocellar area convex, 2.8 X broader than long. Head (Fig. 4), tho- rax, and abdomen smooth, impunctate, with polished surface. Hind basitarsus about 1.2X length of remaining tarsal seg- ments combined, about 4.0 X longer than maximum width. Tarsal claws without basal lobe. Genital capsule similar to Fig. 15, penis valve in Fig. 19. Holotype. — 8, "South India, Anamalia [Anamalai ?] Hills, Cinchona, 3500 ft., V- 1965, P.S. Nathan" (USNM). Paratypes. — INDIA: Same data as holo- type (18, USNM), same except V-1964 (1 8, USNM), X-1967 (2 8, USNM); Coim- batore, S. India, 9.7.1949, P. Susai Nathan (1 8 , USNM); Kerala: Permade, Parryar, 3500 ft., 21-9-1995, coll. M.S. Saini (1 8, PUNJ). Distribution. — India (Kerala, Tamil Nadu). Etymology. — The species name is based on the shining, impunctate frons. Remarks. — The impunctate, shining frons is similar to that of E. ceylonica and £. nigriceps, but the entirely black colora- tion precludes the association of E. lissof- rons because color variation has not been observed in males of those species. The male of £. nigriceps has the thorax partly orange, and the male of £. ceylonica has the thorax and abdomen mostly orange, un- like the entirely black thorax and abdo- men of £. lissofrons. The male penis valve of £. lissofrons (Fig. 19) is oval and similar to that of £. ceylonica (Fig. 16); it is less 342 Journal of Hymenoptera Research similar to the curved valve of £. nigriceps (Fig. 20). There is no indication that E. lis- sofrons is a color variant, but, if so, it could be one of E. ceylonica. Eusunoxa nigriceps (Rohwer) (Fig. 20) Beleses nigriceps Rohwer 1915: 51. Eusunoxa nigriceps: Smith 1982: 120.— Wei 1997: 89 (in subg. Asunoxa; in key). Female. — From Rohwer (1915). "Length, 6 mm." "Head and posterior femora be- yond middle, four posterior tarsi and the antennae black, the rest of the insect ru- fous." "Wings distinctly hyaline, venation dark brown." "Clypeus truncate, surface coarsely, irregularly punctured;" "front and posterior orbits shining, impunctate;" "flagellum gradually thickened until it reaches the apex of the second joint;" "the fourth and following joints compressed; thorax shining, impunctate." Male (paratype). — Length, 6.2 mm. An- tenna and head black. Thorax blackish with tegula, pronotum, V-shaped mark on mesoprescutum, mesoscutellum, and me- tanotum dark orange. Abdomen mostly black. Foreleg orange with tarsus black; midleg with coxa, trochanter, and femur orange, tibia and tarsus black; hind leg with coxa, trochanter, and basal third of femur orange, apical two-thirds of femur and entire tibia and tarsus black. Wings lightly,uniformly infuscated; veins and stigma black. Antennal length 1.7X head width; 3rd segment about 1.1 X length of 4th. Lower interocular distance 0.8 X eye length; postocellar area 2.6 X broader than long; clypeus subtruncate; malar space lin- ear; OOL:POL:OCL = 1.00:1.00:0.74. Head shining, impunctate with very few punc- tures between antennae and eyes (similar to Fig. 4). Hind basitarsus 1.5X length of remaining tarsal segments combined, about 2.5 X longer than maximum width. Tarsal claws without basal lobe. Genital capsule similar to Fig. 15; penis valve in Fig. 20, markedly curved in lateral view. Types. — "Described from one female from Marikuppam, . . . 3,500 feet" and from "two males, one allotype, from Ban- galore, . . . 3,000 feet" (South India). Type and allotype in the Indian Museum, as stated by Rohwer (1915), but we could not locate them and they are apparently no longer in existence. One paratype male is in the USNM, labeled "Ind. Mus., Banga- lore, S. India, ca. 3000 ft., 12-X-10, Annan- dale," "U.S.N.M. Paratype No. 18909." Distribution. — India (Karnataka). Remarks. — Rohwer mentioned claws cleft, with the inner teeth exceeding the outer. Because the teeth are lateral and the paratype examined has the inner tooth shorter, Rohwer probably meant the outer tooth is longer. Because we could not examine the ho- lotype female, we assume Rohwer's asso- ciation of sexes was correct. The descrip- tion of the female is based on Rohwer's (1915) original description. It is very sim- ilar to the female of £. ceylonica and, so far as we can determine, can only be separat- ed by the orange coloration of the tibiae and hind femur (see preceding key). Sep- aration of E. nigriceps is based on the para- type male which, unlike most species and shared only by E. lissofrons and E. ceylon- ica, has the head shining and impunctate. However, the male of E. lissofrons is en- tirely black, and the male of E. ceylonica has the thorax and abdomen almost en- tirely orange. The shape of the male penis valve (Fig. 20) is markedly different from those species, and it is based on this fea- ture that we believe £. nigriceps is a valid species and not a color variation of the others. We have not seen specimens that match Rohwer's description of the female or the paratype male examined. Thus, we keep £. nigriceps a separate species until more material is available for evaluation of its status. Eusunoxa punctata Smith and Saini, new species (Figs. 1, 8, 11) Female. — Length, 8.2 mm. Head and an- tenna black with 1st antennal segment, Volume 12, Number 2, 2003 343 /> 14 21 Figs. 13-22. 13-20, Male genitalia. 13, Genital capsule, ventral view of left half, ol I usunoxa formosana. 14, Penis valve, lateral, of E. formosana. 15, Genital capsule, ventral view of left half, of £. ceylonica. 16, Penis valve of E. ceylonica. 17, Penis valve of E. auricauda. 18, Penis valve of E. semipunctata. 19, Penis valve ol E. lissofrons. 20, Penis valve of E. nigriceps. 21-22, Tarsal claws. 21, E. formosana. 11, E. ceylonica. area below antenna, clypeus, labrum, mouthparts, mandible except tip, and very dim spots lateral to lateral furrows dark orange. Thorax and abdomen orange, ex- cept tergites 2-7 dark brownish at center. Legs orange with midtarsus, apical two- thirds of hind femur, apical half o\ hind tibia, and entire hind tarsus black. Wings hyaline, infumated on apical half; stigma and costa brownish, rest of venation dark 344 Journal of Hymenoptera Research brown to black. Antenna subincrassinated in middle, antennal length 1.7X head width; pedicel 2X its apical width; scape as long as its apical width; 3rd and 4th antennal segments subequal in length. Clypeus truncate; lower interocular dis- tance subequal to eye length; malar space distinct but much less than half diameter of front ocellus; OOL:POL:OCL = 1.00: 0.96:0.96; postocellar area subconvex, 2.4X broader than long. Frons densely punc- tate, punctures large and close together, separated by narrow ridges (Fig. 1); hind orbits and postocellar area less punctured, shining. Hind basitarsus (Fig. 8) 1.7X length of remaining tarsal segments com- bined; 3.7X longer than its maximum width. Tarsal claws without basal lobe. Lancet as in Fig. 11, with about 21 serru- lae. Male. — Unknown. Holotype. — 9, Nagaland: Chuchuyin- long, 2800 ft., 9.5.1994, collection M.S. Sai- ni (PUNJ). Distribution. — India (Nagaland). Etymology. — The species name is based on the densely punctate frons. Remarks. — In E. punctata, the hind basi- tarsus is unusually long in comparison to the length of the remaining tarsal seg- ments, and the frons is densely punctate, both character states of which are shared only with E. formosana and E. ebena. Eu- sunoxa formosana, however, is almost en- tirely orange yellow and both £. formosana and £. ebena have a basal lobe on the tarsal claws. The shape of the serrulae of the lan- cets of £. punctata and E. formosana also differ (Fig. 9, 11). The only other species with a densely punctate frons and absence of a basal lobe on the tarsal claws is E. buchi from the Philippines. However, £. buchi has a shorter hind basitarsus, only about 1.4X the length of the remaining tarsal segments combined, and the basal two and basal half of the third antennal segments are dark orange. Eusunoxa semipunctata Smith and Saini, new species (Figs. 2, 7, 12, 18) Female. — Length, 7.2-8.0 mm. Antenna and head black. Thorax and abdomen or- ange, upper half of mesepisternum some- what more whitish than orange. Legs with coxae blackish except apices, trochanters white, femora blackish except extreme ba- ses orange, fore- and midtibiae and fore- and midtarsi white with apical 3-4 tarsal segments infuscated to black. Wings light- ly infumated apically from base of stigma, basal halves clear, venation dark brown to black. Antenna subincrassinated in mid- dle, antennal length 1.8X head width; pedicel and scape each 2X longer than their apical widths; 3rd segment very slightly longer than 4th. Clypeus subtrun- cate; lower interocular distance subequal to eye length; malar space nearly linear; OOL:POL:OCL = 1.00:0.80:0.68; postocel- lar area subconvex, 2.9 X broader than long. Frons with distinct but sparse, wide- ly separated punctures with broad, shiny interspaces between them (Fig. 2). Hind basitarsus (Fig. 7) 1.3X length of remain- ing tarsal segments combined; 3.9 X longer than maximum width. Tarsal claws with- out basal lobe. Lancet as in Fig. 12, with about 20-21 serrulae. Male. — Length, 6.5-7.0 mm. Antenna, head, and thorax black. Color of legs sim- ilar to that of female. Structure similar to that of female. Genital capsule similar to Fig. 15, penis valve in Fig. 18. Holotype. — 9, "South India, Madras State. Anamalai Hills, Kadamparai, 3500', V-'63, P.S. Nathan" (USNM). Paratypes. — INDIA: Same data as holo- type (12 9, 43 cT); Nilgiri Hills, Naduva- tam, S. India, 6000 ft., May 1958, P.S. Na- than (1 6); South India, Anamalai Hills, Cinchona, 3500 ft., IV-1969, P.S. Nathan (6 6 ); V-1964 (4 6 ); V-1965 (4 9, 1 6); V-1966 (3 6); V-1967 (4 9,2 c?); IV-1969 (1 6); V- 1968 (2 9, 2 cT); V-1969 (2 9, 2 6); IX-1969, T.R.S. Nathan (1 6); Poomundi, 1550 m, Volume 12, Number 2, 2003 345 Kerala, India, 19-9-95 (1 6) (all USNM); Kerala: Poonmundi Hills, 4800 ft., 18.5.1995, collection M.S. Saini (1 6, PUNJ). Distribution. — India (Kerala). Etymology. — The name is based on the presence of widely spaced punctures on the frons. Remarks. — Association of sexes is based on series taken at the same localities and at the same time. Only the coloration of the thorax and abdomen differs; the males black and the females orange. Such color dimorphism is not unusual in sawflies, and E. nigriceps apparently also shows col- or dimorphism. However, males of E. cey- lonica and E. formosana are similar in color to the females. Both sexes of E. semipunc- tata have the distinct but widely spaced punctures on the frons with shining inter- spaces (as in Fig. 2). The female is sepa- rated from those of other species of Eux- unoxa by the mostly black femora, and the male by the mostly black legs with the fore- and midtibiae and tarsi white. The punctation of the head also separates the female; no other species with an orange thorax and abdomen have such dense punctures. The male is similar to E. auri- cauda, but the abdomen of E. auricauda is orange (see remarks under that species). ACKNOWLEDGMENTS We thank the following for loan of specimens in their care: A. Taeger and S. Blank, Deutsches Ento- mologisches Institut, Eberswalde, Germany; C. Tay- lor, The Natural History Museum, London, U.K.; I. Persson, Naturhistoriska Riksmuseet, Stockholm, Sweden; and E. Kierych, Polska Akademia Nauk, Warsaw, Poland. Cathy Apgar, Systematic Entomol- ogy Laboratory, USDA, prepared the photographs. We thank the following for review of the manuscript: H. Goulet, Agriculture and Agri-Food Canada, Otta- wa, and J. W. Brown and E. E. Grissell, Systematic Entomology Laboratory, USDA, Washington, DC. Fi- nancial assistance provided by the U.S. Department of Agriculture (PL-480, Grant No. FG-In-753, Project No. IN-ARS-418) in collaboration with ICAR, New Delhi, is also thankfully acknowledged. LITERATURE CITED Enderlein, G. 1920 [1919]. Symphytologica II. Zur Kenntnis der Tenthredininen. Sitzungsberichl der Gesellschaft Naturforschendcr Freunde zu Berlin, pp. 347-374. Enslin, E. 1911. Ein Beitrag zur Tenthrediniden-Fauna Formosas. Societas Entomologica 24: 93-104. Haris, A. 2000. New Oriental sawflies (Hymenoptera: Tenthredinidae). Somogyi Muzeumok Kozlemenyei 14: 297-305. Malaise, R. 1932. A new sawfly from Ceylon (Hym. Tenthredinidae). Ceylon journal of Science, Section B 17: 147-148. Malaise, R. 1963. Hymenoptera Tenthredinoidea, sub- family Selandriinae, key to the genera of the world. Entomologisk Tijdskrift 84: 159-315. Rohwer, S. A. 1915. Some Oriental sawflies in the In- dian Museum. Records of the Indian Museum 1 1 : 39-53. Saini, M. S. and J. S. Deep. 1994. First record of Al- lantinae (Tenthredinidae: Hymenoptera) from India, journal of the Bombay Natural History Society 91: 47-50. Smith, D. R. 1982. Symphyta (Hymenoptera) of Sri Lanka. Proceedings of the Entomological Society of Washington 84: 117-127. Togashi, I. 1981. Some sawflies (Hymenoptera, Per- gidae and Tenthredinidae) from New Britain, Papua New Guinea, and Palawan Is., Philip- pines. Kontyti 49: 414-421. Wei, M. 1997. Review of the genus Eusunoxa with erection of a new subgenus (Hymenoptera: Blen- nocampidae). journal of Central South Forestry University 17: 88-89. J. HYM. RES. Vol. 12(2), 2003, pp. 346-354 Larval Anatomy of Orussidae (Hymenoptera) Lars Vilhelmsen Zoological Museum, University of Copenhagen, Universitetsparken 15, DK-2100, Denmark, e-mail: lbvilhelmsen@zmuc.ku.dk Abstract. — The external and selected parts of the internal anatomy of the larva of Orussus abie- tinus (Orussidae) are examined The external anatomy is similar to that reported for other Orus- sidae. The eyes, laterocervicalia, and thoracic legs are absent, and the antennae, maxillae, and labium are reduced. Each thoracic and abdominal segment has a pair of short transverse rows of recurved cuticular spines laterodorsally, probably for locomotory purposes. A previously unre- ported feature is the configuration of the hindgut, which has a set of transverse interlocking cuticular folds. The mid- and hindgut are anatomically continuous. Overall, the larval anatomy of Orussidae is highly reduced compared with that of other basal hymenopteran lineages and resembles that of apocritan larvae. The modifications of the larval anatomy are probably correlated with the shift in lifestyle from herbivorous to carnivorous, which is less demanding of the sensory, feeding, and locomotory apparatus. Orussidae is a small wasp family, com- termediate in Hymenoptera (Hanson and prising 75 species worldwide (Vilhelmsen Gauld 1995; Vilhelmsen 1997). in press). Traditionally, the Hymenoptera Nutall (1980) provided a very brief de- have been subdivided in two suborders, scription of the larva of Guiglia schauins- the almost exclusively herbivorous 'Sym- landi (Ashmead, 1903), a New Zealand phyta' and the predominantly entomoph- species. The only detailed descriptions of agous Apocrita; Orussidae were usually the immature stages of an orussid species placed in the 'Symphyta' because of the are by Rohwer and Cushman (1917) and absence of the wasp-waist in the adults, Parker (1935), who dealt with Orussus oc- the defining feature of the Apocrita. How- cidentalis Cresson, 1879, a Nearctic species, ever, recent phylogenetic treatments of the The section on Orussidae in Yuasa's (1922) Hymenoptera unequivocally have re- treatment of non-apocritan larvae is based trieved Orussidae + Apocrita as an ex- entirely on Cushman and Rohwer (1917). tremely well supported clade (Ronquist et They noted that the larva of Orussus share al. 1999, Vilhelmsen 2001). Information on many traits with those of apocritan Hy- the biology of most species of Orussidae menoptera (eyes, most head appendages, is scarce or non-existent; what evidence and thoracic legs very reduced), whereas there is (see Vilhelmsen et al. 2001) indi- the anatomy of adult Orussidae apparent- cates that the larvae are ectoparasitoids of ly resembles that of 'Symphyta'. This mix woodboring insect larvae, usually Bupres- of features and the parasitoid lifestyle led tidae (Coleoptera). This was probably the Rohwer and Cushman (1917) to erect a lifestyle of the common ancestor of all par- new suborder, the Idiogastra, comprising asitoid Hymenoptera. The position of the just the Orussidae. woodboring 'siricoid' superfamilies as Parker (1935) observed a few internal successive outgroups to the Orussidae- features of the head anatomy of O. occi- Apocrita clade indicates that the parasit- dentalis, but otherwise information about oid lifestyle evolved via a woodboring in- the internal larval anatomy of the family Volume 12, Number 2, 2003 347 is wanting. Orussidae were not included lected in a kerosene-acetic acid-dioxane in the survey of 'symphytan' larvae by solution (KAAD). Only a limited number Maxwell (1955). Specifically, it is not of observations of the external anatomy known whether there is any connection could be carried out on these specimens. between the mid- and hindgut. The sepa- External features.— Initial examination of ration of these gut sections and the post- external features on specimens kept in eth- ponement of voiding of the gut contents anol was carried out with a Leica MZ Apo (as a meconium) until just prior to pupa- dissection microscope. tion is a near universal feature of the Apo- SEM.— Specimens preserved in ethanol crita (Hanson and Gauld 1995; Quicke were critically point dried and coated with 1997); it usually is interpreted as an im- Pt/Pd in a Jeol JFC-2300HR high resolu- portant adaptation to the parasitoid life- tion fine coater. They were examined with style, especially endoparasitism, apparent- a Jeol JSM-6335F field emission scanning ly serving to prevent contamination of the electron microscope. host. However, the condition in Orussidae Histological sections. — Specimens pre- and some of the putatively basal apocritan served in ethanol or Pampels fluid were taxa (Stephanidae, Megalyridae) has not dehydrated in butanol and embedded in been investigated, making it impossible to paraplast. Sagittal sections of 8-12 fxm deduce exactly when the feature arose in thickness, depending on the size of the the evolutionary history of parasitic Hy- specimen, were cut with a microtome. The menoptera. sections were treated in a trichrome stain In the present study, I investigate the (Weigert's haematoxylin-bluish erythro- larval anatomy of two species of Orussi- sin-fast green, preceded by phosphomo- dae in some detail. In addition to the ex- lybdic acid); permanent preparations were ternal features already described by Roh- made in entellan. wer and Cushman (1917), selected internal features are examined. The findings are KhSULIS discussed in relation to features observed The descriptions are based entirely on in other hymenopteran larvae. the prepupae and early instar of O. abie- timts. The observations that could be MATERIALS AND METHODS made on the G sd„mm.ll„ll1i specimens Larvae of Orussus abietimis (Scopoli, did not reveal any significant differences 1763) and Guiglia schauinslandi (Ashmead, between the two species. 1903) were available for study. Unfortu- Overall. — Body elongate, dorsoventrally nately, only the material of the former was flattened (less conspicuous in early instar). in good condition, having been collected Body outline in cross-section with distinct recently (1998 or later) and preserved in bend laterally between upper and lower 80% ethanol or Pampels fluid and subse- surface. Head hypognathous, boundary quently transferred to ethanol. In total, between head and prothorax indistinct in five larvae of O. abietinus were examined, early instar, especially laterally. Segment Of these, four were final instars or pre- boundaries otherwise distinctly demarcat- pupae, collected in early spring prior to ed by deep furrows dorsally, especially on pupation, and one was an early instar col- abdomen. Body cuticle wrinkled, but lected later in spring on a moldy pupa of without regular transverse annuli or other Buprestis sp. (Coleoptera, Buprestidae). regular intrasegmental subdivisions. Four larvae of G. scliaiiiiislandi were ex- Small trichoid sensilla scattered over body amined. All were in rather poor condition, surface. All thoracic and abdominal seg- having been preserved in 95% ethanol for ments dorsally with paired, transverse 40+ years after having initially been col- rows of 2—1 (early instar) or 3-5 (prepupa) 348 Journal of Hymenoptera Research Figs. 1-6. Orussus abietinus, head anatomy. 1, 3-6. Prepupa. 1, Head, ventral view, arrow indicates position of anterior tentorial pit: an = antenna; la = labium; lb = labrum; mx = maxilla. 2, Early instar larva, mouth- parts, lateral view: md = mandible (only base visible). 3, Antenna. 4, Tips of mandibles, ventral view. 5, Maxilla, arrows indicate trichoid sensilla (two leftmost broken). 6, Labium, arrows indicate trichoid sensilla: so = salivarial orifice. sligthly recurved cuticular spines subme- dially (Figs. 7-10). Head. — Head capsule weakly sclero- tized. Eye absent. Antenna one-segment- ed, short, rounded, situated lateroventral- ly on head capsule (Figs 1-2), distally with two short peglike sensilla in shallow de- pression (Fig. 3). Clypeus weakly delim- ited dorsally; anterior tentorial pits faintly visible as shallow, elongate depressions (Fig. 1). Circular cuticular structure pre- sent medially of anterior tentorial pits, not visible externally. Small, transversely elongate sclerite present between clypeus and labrum. Labrum fleshy, slightly bifid apically, with 7-8 trichoid sensilla antero- lateral^. Mandibular base broad, fleshy (Fig. 2), mandibular articulations dis- placed medially, not visible externally on head capsule. Distal part of mandible slen- Volume 12, Number 2, 2003 349 der and heavily sclerotized, arising medi- surrounded by muscular sheath, except ally from base, partly covered by labrum for part posteriorly of cuticular folds (Fig. anteriorly and maxilla posteriorly (Fig. 2), 12); sheath with transverse and outer Ion- terminating in three distinct cuticular gitudinal fibers. Prepupa similar to early teeth visible ventrally (Fig. 4). Maxilla un- instar except for following: Midgut with differentiated, transversely elongate fleshy straight walls, apparently empty except lobe, anteriorly abutting labrum; maxil- for peritrophic membrane. Boundary be- lary palp not developed, three trichoid se- tween mid- and hindgut marked by one tae situated in middle of maxillar lobe pair of malphigian tubules opening into (Fig. 5). Labium narrow anteriorly, broad- lateral part of gut. Epidermis retracted ening posteriorly (Fig. 6), weakly delimit- from cuticle in posteriormost abdominal ed from maxilla by shallow oblique sul- segment (Fig. 13), interspace between epi- cus, retracted relative to maxillae. Salivar- dermis and cuticle occupied by amor- ial orifice transverse slit surrounded by phous material traversed by elongate tis- sclerotized cuticle (barely discernible in sue strands. Epidermis of anterior part of early instar larva), situated subapically on hindgut not thickened. Muscle sheath not labium. Labial palp not developed, paired extending very far along hindgut, not short trichoid sensilla present posterolat- reaching region of cuticular folds, com- erally. paratively less developed than in early in- Thorax. — Laterocervicalia absent. Ante- star larva. Hindgut walls with 18-20 close- rior thoracic spiracle well developed, ap- ly appressed cuticular folds (Fig. 14); folds parently situated slightly anterior to transversely oriented medially, obliquely boundary between pro- and mesothorax. posteromedially oriented laterally. Posterior thoracic spiracle vestigial, rep- resented by small, elongate sclerotisation ui^u^iui\ just posterior to meso/metathoracic The larva of O. abietinus closely resem- boundary. Thoracic legs entirely absent. bles O. occidental is (see Rohwer and Cush- Abdomen. — Ten segments present. Seg- man 1917) in all external features; this is ments 1-8 with well developed spiracles hardly surprizing, since the two species situated laterodorsally (Fig. 7); spiracles are probably sister taxa (Vilhelmsen in circular, rim well sclerotized. Abdominal press). The more distantly related G. legs, suranal process, suranal hook, and schauinslandi does not differ significantly subanal appendages absent. Anal slit sit- from Orussus spp. in the features that uated posteriorly on segment 10, trans- could be observed. However, all three spe- versely elongate (Figs 9, 10), shallow (can- cies are comparatively derived within not be pried open with a fine needle). Orussidae, making inferences about the Hindgut. — Posterior part of midgut of ground plan states for the family tenuous. early instar larva with convoluted walls. The larval head anatomy of Orussidae Gut contents amorphous, with dark is highly reduced compared with other grains. Malphigian tubules not observed, non-apocritan Hymenoptera. The eye is Mid- and hindgut continuous (Fig. 11). entirely absent (a condition also observed Hindgut lined with unsclerotized cuticle in 'siricoid' superfamilies), the antenna is becoming progressively thicker posterior- one-segmented and of the mouthparts, ly. Anterior part of hindgut with some- only the mandibles appear to be function- what thickened epidermis. Hindgut with al, both the maxilla and labium having walls straight, except for 10-12 narrow lost all traces of endites and palps and be- transverse cuticular folds posteriorly (Figs ing immovable relative to the head cap- 11, 12); cuticular folds correspond to con- sule. In contrast, most 'Symphyta' have cavities in opposite wall. Most of hindgut multi-segmented antenna as well as dif- 350 Journal of Hymenoptera Research ferentiated endites and palps on the max- illa and labium (see Yuasa 1922). The mandibles of herbivorous 'Symphyta' are usually well sclerotized throughout with external articulations and broad shearing teeth distally. The condition of the man- dibles (partly concealed by labrum and maxilla, proximal part unsclerotized, dis- tal part narrowed, see also Rohwer and Cushman 1917, fig. 2a) observed in Orus- sidae is apparently similar to that of Schlettererius cinctipes (Cresson) (Stephan- idae; Short 1978, fig. 20; Taylor 1967). Gen- erally, the head anatomy of orussid larva is much closer to that of Apocrita than of other 'Symphyta'. With regard to maxillar and labial palps, those of Orussidae are even more reduced than some Apocrita (e.g., some Ichneumonoidea (Finlayson 1987) and Aculeata (Evans 1987; McGinley 1987)), which have papilliform, one-seg- mented palps. The thoracic and abdominal anatomy of the larvae of Orussidae and Apocrita are also reduced relative to more basal Hy- menoptera. The laterocervicalia, which in representatives of most 'symphytan' su- perfamilies articulates with the head cap- sule anteriorly and connects to the pro- notum posteriorly (Vilhelmsen unpubl.), is entirely absent in Orussidae (see also Parker 1935) and apparently also in Apo- crita, providing a closer integration of the head and thorax in the two latter taxa. The thoracic legs are entirely absent, having become progressively reduced within the basal lineages of the Hymenoptera (Vil- helmsen 2001). A suranal process, a heavi- ly sclerotized projection on the 10th ab- dominal segment dorsally of the anal slit prominent in Cephidae, Anaxyelidae, Sir- icidae, and Xiphydriidae (see Yuasa 1922, pi. xii) is absent from Orussidae and Apo- crita. The transverse rows of cuticular spines dorsally have been reported only from Schlettererius cinctipes (Stephanidae; Taylor 1967) outside the Orussidae, al- though in this species, they are only pre- sent in the prepupa. Their function in orussid larvae is perhaps to facilitate movement along wood galleries, in- creased leverage being provided by the deep dorsal furrows between the body segments. Cooper (1953) observed ovipo- sition by female Orussus in wood galleries some distance from potential hosts; he in- terpreted this as indicating that the larvae might be frass feeders. Alternatively, the early instar larva might actively seek out the host within its galleries, if the female was unable to reach the host with its ovi- positor; this possibility was considered but rejected by Cooper. Given the well corroborated monophyly of Orussidae + Apocrita, the interpreta- tion of the phylogenetic significance of the shared reductional features in the larval anatomy is fairly straightforward: they represent synapomorphies for these two taxa. Some of these characters have al- ready been included in recent phylogenet- ic treatments of the Hymenoptera (e.g., number of larval antennal segments, re- duction of larval thoracic legs, characters 224 and 228 in Vilhelmsen 2001), whereas as others (reduced maxillar and labial ap- pendages, absence of laterocervicalia) have been ignored until now. The functional significance of the ana- tomical modifications can be interpreted in the context of the shift in larval lifestyle from herbivorous/woodboring to carniv- orous/parasitoid having taken place in the common ancestor of Orussidae and Apocrita. The reduced sensory capabilities (loss of eyes, reduction of antennae) reflect the confined habitat (galleries in wood) of the larvae of the earliest parasitoid wasps and their closest relatives, the woodwasps. The reduction of the mouthparts is per- haps a consequence of the shift in food source from particulate plant matter that requires considerable handling and chew- ing to a much less demanding diet of in- sect body fluids. Once an ectoparasitoid wasp larva is in contact with its host, it is required to do little more than puncture the integument of the latter to gain access Volume 12, Number 2, 2003 351 Figs. 7-10. Orussus abietinus, abdominal anatomy. 7-8, Early instar larva, posterior to the left. 7, Dorsal part of abdominal segment: arrows indicate cuticular spines; sp = spiracle. 8, Recurved cuticular spines. 9-10, Prepupa. 9, 10th abdominal segment, posterior view, arrows indicate cuticular spines. 10, Same, lateral view, arrow indicates anal slit. 352 Journal of Hymenoptera Research Figs. 11-14. I Orussus abietinus, internal anatomy of hindgut (posterior to left). 11-12. Early instar larva. 11, Overview of hindgut region, arrow indicates transition between mid- and hindgut: mus = muscle sheath. 12, Details of cuticular folds in hindgut. 13-14, Prepupa. 13, Overview of hindgut region, arrow indicates tran- sition between mid- and hindgut, rectangle indicates approximate extent of Fig. 14: epi = retracted epidermis. 14, Details of cuticular folds in hindgut. to its contents, a purpose for which the narrow, pointed mandibles of the orussid larva probably is admirably suited. The reduction of the locomotory apparatus again reflects the confined larval habitat, being taken even further than in the Ce- phidae and woodwasp families, which re- tain vestiges of thoracic legs. The suranal process present in these taxa is also absent in Orussidae and Apocrita, a change that perhaps is correlated with the shift in food source. This feature is needed as a brace by the endophytophagous taxa when chewing a tunnel through tough plant ma- terial; the larvae of Orussidae and wood- living Apocrita do not chew their own tunnels and hence do not need a posterior brace. The configuration of the hindgut in the early instar larva and prepupa of Orussus is unlike anything else reported from Hy- menoptera. The cuticular folds matching concavities in the opposite walls and the extension of the muscular sheath sur- rounding the hindgut to include the part with the folds indicate that even though the mid- and hindgut are not anatomically separated, the larva is able to close the hindgut by muscular action. In the pre- pupa, the opposite walls of the hindgut abut in the region with the cuticular folds even though the muscular sheath does not extend this far back; this may be caused by the loosening of the prepupal cuticle and the retraction of the epidermis and as- sociated musculature prior to pupation. Volume 12, Number 2, 2003 353 The connection between mid- and hind- gut in 'Symphyta' whose embryology have been examined (e.g., Pontania caprae Linnaeus (Tenthredinidae), see Ivanova- Kasas 1959) is established already at the time of hatching from the egg. The ana- tomical separation caused by the failure of the two gut sections to join during the em- bryological development (Hanson and Gauld 1995) has been reported from a wide range of apocritan superfamilies: Ceraphronoidea (Megaspilidae: Dendroce- rus (= Lygocerus) spp., see Haviland 1920a, fig. 14), Chalcidoidea (Pteromalidae: Spa- langia muscidarum Richardson, see Rich- ardson 1922, fig. 7; Asaphes vulgaris Walk- er, Pachycrepis clavata Walker, see Havi- land 1922), Cynipoidea (Figitidae: Charips spp., see Haviland 1920b, fig. 11a), Ichneu- monoidea (Ichneumonidae: Pimpla turi- onellae (Linnaeus), see Fiihrer and Willers 1986, fig. 3), Platygastroidea (Platygastri- dae spp., see Marchal 1906, pi. xviii: 24), Proctotrupoidea (Proctotrupidae: Phaenos- erphus viator (Haliday), see Eastham 1929, fig. 12). It is often accompanied by consid- erable differentiation between hindgut re- gions (epidermis and muscle layer of var- iable thickness, formation of valves; e.g., Eastham 1929, Fiihrer and Willers 1986, Haviland 1920b) that is less conspicuous in O. abietinus. However, many 'Symphy- ta' also have the hindgut differentiated into several regions (Maxwell 1955). The functional and phylogenetic signif- icance of the larval hindgut anatomy of Orussus is difficult to interpret without further information about the orussid life- style and investigation of larvae of some of the basalmost apocritan ectoparasitoid taxa (e.g., Stephanidae and Megalyridae). It is possible, but entirely conjectural, that the cuticular folds help the orussid larva to clamp its hindgut shut for most of its development, thus preventing contami- nation of its host. However, the expulsion of a meconium prior to pupation as seen in most apocritans examined was not ob- served in a couple of hatchings of O. abie- tinus (Vilhelmsen unpubl.). The anatomi- cal position (posteriorly in the hindgut) of the cuticular folds in Orussus is not ho- mologous with the position (boundary be- tween mid- and hindgut) of the separation of the gut sections in Apocrita, making it unlikely that the latter evolved from the former. Thus, this study has revealed an- other intriguing feature occurring in par- asitic Hymenoptera, rather than elucidat- ing the evolution of an already known trait. ACKNOWLEDGMENTS Hans Ahnlund, Gnesta, Sweden, provided invalu- able assistance with collecting material of O. abietinus. Toni Withers, Forest Research Institute, Rotorua, New Zealand, graciously made the material of G. schauinslandi available for study. Dave Smith and Eric Grissell, both in Systematic Entomology Laboratory, Washington, USA commented on an earlier version of the manuscript. LITERATURE CITED Cooper, K. W. 1953. Egg gigantism, oviposition, and genital anatomy: their bearing on the biology and phylogenetic position of Orussus (Hymenop- tera: Siricoidea). Proceedings of the Rochester Acad- emy of Sciences 10: 38-68. Eastham, L. E. S. 1929. The post-embryonic develop- ment of Phaenoserphus viator Hal. (Prototrypoi- dea), a parasite of the larva of Pterostichus niger (Carabidae), with notes on the anatomy of the larva. Parasitology 21: 1-21 + 3 plates. Evans, H. E. 1987. Families: Bethylidae, Dryinidae, Chrysididae, Scoliidae, Tiphiidae, Mutillidae, Sapygidae, Formicidae, Rhopalosomatidae, Pom- pilidae, Vespidae, Sphecidae. Pp. 670-688 //;: Stehr, F.W. Immature Insects. Kendall/Hunt Pub- lishing Company, Dubuque, Iowa, USA, 754 pp. Finlayson, T. 1987. Ichneumonoidea. I'p. 649-663 /;;: Stehr, F.W. Immature Insects. Kendall/Hunt Pub- lishing Company, Dubuque, Iowa, USA, 7^4 pp. Fiihrer, E. and D. Willers. 1986. The anal secretion oi the endoparasitic larva Pimpla lurionellae: Sites of production and effects, lourual of Inset I Physiology 32: 361-367. Hanson, P. E and I. D. Gauld. L995. The Hymenoptera of Costa Rica. Oxford University Press, Oxford, 893 pp. Haviland, M. D. 1920a. On the bionomics and devel- opment of Lygocerus testaceimanus and Lygocerus cameroni Kieffer (Proctotrypoidae-Ceraphroni- dae), parasites of Aphidius, Braconidae. Quarterly Journal of microscopical Science 65: 101 127. 354 Journal of Hymenoptera Research Haviland, M. D. 1920b. On the bionomics and post- embryonic development of certain cynipid hy- perparasites of aphides. Quarterly Journal of mi- croscopical Science 65: 451^478. Haviland, M. D. 1922. On the post-embryonic devel- opment of certain chaicids, hyperparasites of aphides. Quarterly Journal of microscopical Science 66: 321-338. Ivanova-Kasas, O. M. 1959. Die embryonale Entwick- lung der Blattwespe Pontania caprae L. (Hyme- noptera, Tenthredinidae). Zoologische Jahrbucher, Abteilung fiir Anatomie und Ontogenie der Tiere 77: 193-228. Marchal, P. 1906. Recherces sur la biologie et le de- velopement des hymenopteres parasites. Les pla- tygasters. Archives de zoologie experimental et ge- nerate Serie 4, 4: 475-640 + 8 plates. Maxwell, D. E. 1955. The comparative internal larval anatomy of sawflies (Hymenoptera: Symphyta). The Canadian Entomologist 87, supplement 1, 132 PP- McGinley, R. J. 1987. Families: Colletidae, Oxaeidae, Halictidae, Andrenidae, Mellitidae, Megachili- dae, Anthophoridae, Apidae. Pp. 689-704 In: Stehr, F.W. Immature Insects. Kendall/Hunt Pub- lishing Company, Dubuque, Iowa, USA, 754 pp. Nutall, M. J. 1980. Insect Parasites of Sirex. Forest and Timber Insects in New Zealand 47: 12 pp. (unpa- ginated). Parker, H. L. 1935. Note on the anatomy of tenthre- dinid larvae with special reference to the head. Bollcttino del Laboratorio di Zoologia Generale e Agraria delta R. Scuola Superiore d'Agricoltura in Portici 28: 159-191. Quicke, D. L. J. 1997. Parasitic Wasps. Chapman and Hall, London, 470 pp. Richardson, E. 1922. Studies on habits and the devel- opment of a hymenopterous parasite, Spalangia muscidarum Richardson. Journal of Morphology 24: 513-557. Rohwer, S. A. and R. A. Cushman. 1917. Idiogastra, a new suborder of Hymenoptera with notes on the immature stages of Oryssiis. Proceedings of the Entomological Society of Washington 19: 89-98. Ronquist, F., A. P. Rasnitsyn, A. Roy, K. Erikson, and M. Lindgren. 1999. Phylogeny of the Hymenop- tera: A cladistic reanalysis of Rasnitsyn's (1988) data. Zoologica Scripta 28: 13-50. Short, J. R. T. 1978. The final larval instars of the Ich- neumonidae. Memoirs of the American Entomolog- ical Institute 25, 508 pp. Taylor, K. L. 1967. Parasitism of Sirex noctilio F. by Schlettererius cinctipes (Cresson) (Hymenoptera: Stephanidae). Journal of the Australian Entomolog- ical Society 6: 13-19. Vilhelmsen, L. 1997. The phylogeny of lower Hyme- noptera (Insecta), with a summary of the early evolutionary history of the order. Journal of Zoo- logical Si/stematics and evolutionary Research 35: 49-70. Vilhelmsen, L. 2001. Phylogeny and classification of the extant basal lineages of the Hymenoptera (In- secta). Zoological Journal of the Linnean Society 131: 393-442. Vilhelmsen, L. in press. Phylogeny and classification of the Orussidae (Insecta: Hymenoptera), a basal parasitic wasp taxon. Zoological Journal of the Lin- nean Society. Vilhelmsen, L., N. Isidoro, R. Romani, H. H. Basibu- yuk, and D. L. J. Quicke. 2001. Host location and oviposition in a basal group of parasitic wasps: the subgenual organ, ovipositor apparatus, and associated structures in the Orussidae (Hyme- noptera, Insecta). Zoomorphology 121: 63-84. Yuasa, H. 1922. A classification of the larvae of Ten- thredinoidea. Illi>wis Biological Mo)iographs 7(4), 172 pp. J. HYM. RES. Vol. 12(2), 2003, pp. 355-357 NOTE First Report of Male Sleeping Aggregations in the Pollen Wasp Celonites abbreviatus (Villers, 1789) (Hymenoptera: Vespidae: Masarinae) F. Amiet and V. Mauss (FA) Forststr. 7, CH-4500 Solothurn, Switzerland; (VM) Staatliches Museum fur Naturkunde, Abt. Entomologie, Rosenstein 1, D-70191 Stuttgart, Germany, email: volker.mauss@stechimmenschutz.de The nightly resting behaviour of solitary and social bees, and probably also of many other Aculeata, constitutes a sleep- like state with many neurophysiological parallels to mammalian sleep (Kaiser and Steiner-Kaiser 1983, Kaiser 1988, 1995). Sleeping behaviour has been well docu- mented in many species of solitary wasps (O'Neill 2001: 294). Interspecific variation of sleep is evident in (1) the location of the sleeping site, (2) the postures adopted during sleep, and (3) whether the wasp sleeps in the company of other members of its own species, members of the oppo- site sex, and members of other species (O'Neill 2001: 294). In contrast, present knowledge of sleeping behaviour of male pollen wasps is poor. Males of some spe- cies of the genus Ceramius seek overnight shelter in conspecific nests that may or may not contain females (Brauns 1910, Gess 1996: 63, Mauss 1996). Celonites andrei Brauns was observed to spend the night sleeping on vegetation (Brauns 1910), and a male of Celonites abbreviatus slept curled up around a blade of grass (Bischoff 1927: Fig. 29). Males of Masarina mixta Richards were commonly observed to sleep in bell- shaped flowers of Wahlenbergia (Campan- ulaceae) on which they and the females forage during the daytime (Gess 1996: 63). Male sleeping aggregations have not previously been recorded for any species of the Masarinae (Gess 1996: 63), although they have been observed for numerous other wasps and bees (summarised by Bis- choff 1927: 62-64, Westrich 1989: 124, O'Neill 2001, Wcislo 2003). We describe here for the first time, male sleeping ag- gregations in C. abbreviatus which were discovered at two different locations in Switzerland. (1) Pfynwald (07°35'E 46°38'N), Rhone Valley, Wallis, Switzerland, July 1977, obs. F. Amiet, A. Krebs: In the evening four males were sleeping in an aggregation on a withered, branched stem of an herba- ceous perennial plant about 0.2 m above the ground. A photograph of this aggre- gation was reproduced in Witt (1998: 192 bottom right), however it was not further commented on, and it was erroneously shown in the vertical position. Distance between the males varied from 0 (i.e., in physical contact with each other) to about 8 mm. All males adopted the same typical posture: They curled their bodies around the stem so that the tip of their metasoma covered the ventral part of the clypeus. Antennae and legs were pulled up under the mesosoma, and the wings were folded underneath the metasoma (Figs. 2, 3, the identical posture as in Bischoff 1927, fig. 29). The males were observed to aggregate and sleep on the same stem on several consecutive nights. (2) Berner Oberaland near Boltigen (07°22'E 46°38'N), Simmental, Switzer- 356 Journal of Hymenoptera Research Figs. 1-3. Celonites abbreviates, male sleeping aggregation. 1, Berner Oberaland, August 1997 (actual body length of males about 8 mm). 2, Sleeping posture (viewed from left), Pfynwald, July 1977 (Photo: A. Krebs). 3, Sleeping posture (viewed from ventral), Pfynwald, July 1977. Volume 12, Number 2, 2003 357 land, at the foot of a scree slope, August 1997, obs. F. Amiet: The sky was cloudy and whenever the sun became obscured by a larger cloud some males of C. abbrev- iatus alighted on two withered stems of herbaceous perennial plants about 0.25 m above the ground (Fig. 1) and adopted the typical sleeping posture. Moments later when the sun emerged again some of these males became active and flew away. From three o'clock p.m. (Central Europe- an Time) onward all the males remained in the sleeping posture on the stems al- though the sun still shined intermittently. At four o'clock p.m. it commenced to rain. A maximum of 14 males were observed on both stems. Distance between the sleeping males in the aggregation ranged from 0 to about 8 mm. Other clades of the Vespidae in which male sleeping aggregations occur are the Euparagiinae (Euparagia scutellaris Cres- son, Moore 1975) and at least six genera of the Eumeninae (Labus, Bischoff 1927: 62, and Ancistrocerus, Eumenes, Pterocheihis, Rhynchium, Stenodynerus, Linsley 1962). The sleeping postures of the eumenine wasps studied differ distinctly from that in Celonites abbreviatus: Males of the Eu- meninae attach themselves to the sub- strate with their legs and mandibles (Lin- sley 1962). Their wings are folded but ex- tend outward at an angle of approximate- ly 45 degrees. Unfortunately, the sleeping posture of Euparagia cannot be brought into context since it was not described in sufficient detail by Moore (1975). The evo- lutionary significance of aggregated sleep- ing is uncertain (Evans et al. 1986), but it has been suggested that it offers protec- tion against predators, may influence ther- moregulation or may be associated with mating behaviour (cf. Freeman and John- ston 1978). LITERATURE CITED Bischoff, H. 1927. Biologic der Hymenopteren. Springer Verlag, Berlin. 598 pp. Brauns, H. 1910. Biologisches iiber siidafrikanische Hymenopteren. Zeitschrift fiir wissenschaftliche In- sekten Biologic 6: 384-387, 445-447. Evans, H.E., O'Neill, K.M. and O'Neill, R.P. 1986. Nesting site changes and nocturnal clustering in the sand wasp Bembecinus quinquespinosus (Hy- menoptera: Sphecidae). journal of the Kansas En- tomological Society 59: 280-286. Freeman, B.E. and Johnston, B. 1978. Gregarious roosting in the sphecid wasp Sceliphron assimile. Annals of the Entomological Society of America 71: 435-441.' Gess, S.K. 1996. The pollen wasps — Ecology and natural history of the Masarinae. Harvard University Press, Cambridge, Massachusetts. 340 pp. Kaiser, W. 1988. Busy bees need rest, too. Behavioural and electromyographical sleep signs in honey- bees. Journal of comparative physiology A 163: 565- 584. Kaiser, W. 1995. Rest at night in some solitary bees — a comparison with the sleep-like state of honey bees. Apidologie 26: 213-230. Kaiser, W. & Steiner-Kaiser, J. 1983. Neuronal corre- lates of sleep, wakefulness and arousal in a di- urnal insect. Nature 301: 707-709. Linsley, E.G. 1962. Sleeping aggregations of aculeate Hymenoptera. Annals of the Entomological Society of America 55: 148-164. Mauss, V. 1996. Contribution to the bionomics of Cer- amius tuberculifer Saussure (Hymenoptera, Ves- pidae, Masarinae). journal of Hymenoptera Re- search 5: 22-37. Moore, W.S. 1975. Observations on the egg laying and sleeping habits of Euparagia scutellaris Cres- son (Hymenoptera, Masaridae). The Pan-Pacific Entomologist 51: 286. O'Neill, K.M. 2001. Solitary wasps. Cornell LTniversir) Press, Ithaca, London. 406 pp. Wcislo, W.T. 2003. A male sleeping roost of a sweat bee, Augochlorella neglectula (Ckll.) (Hymenop- tera: Halietidae), in Panama, journal of the Kansas Entomological Society 7b: 55-59. Westrich, P. 1989. Die Wildbienen Baden-Wurttembergs. Eugen Ulmer Verlag, Stuttgart. 972 pp. Witt, R. 1998. Wespen beobachten, bestimmen. Natur- buch Verlag, Augsburg. 360 pp. INSTRUCTIONS FOR AUTHORS General Policy. The Journal of Hymenoptera Research invites papers of high scientific quality reporting comprehensive research on all aspects of Hymenoptera, including biology behavior, ecology systematics, taxonomy genetics, and morphology Taxonomic papers describing single species are unlikely to be accepted unless a strong case is evident, such as importance in economic entomology or with concurrent biology or ecology. Manuscript length generally should not exceed 50 typed pages; however, no upper limit on length has been set for papers of exceptional quality and importance, including taxonomic monographs at generic or higher level. All papers will be reviewed by at least two referees. The referees will be chosen by the appro- priate subject editor. However, it would be helpful if authors would submit the names of two persons who are competent to review the manuscript. The language of publication is English. Summaries in other lan- guages are acceptable. The deadline for receipt of manuscripts is 1 September (for the April issue) and 1 March (for the October issue). Format and Preparation. Three copies of each manuscript, including copies of illustrations, should be submitted on letter size or A4 paper, double spaced, with at least 25 mm margins on all sides. On the upper left of the title page give name, address, telephone and fax numbers, and e-mail address of the author to whom all correspondence is to be sent. The paper should have a concise and informative title, followed by the names and addresses of all authors. The sequence of material should be: title, author(s), abstract, text, acknowledgments, literature cited, appendix, figure legends, figure copies (each numbered and identified), tables (each numbered and with heading). Each of the following should start a new page: (1) title page, (2) abstract, (3) text, (4) literature cited, (5) figure legends, (6) footnotes. Upon final acceptance of a manuscript, the author should provide the editor with one copy accompanied by either an IBM or Macintosh formatted electronic version. ZIP discs, CD-ROMS, or 3.5 inch floppy discs are acceptable. Final manuscripts and figures may also be sent via email, but because symbols and tables are not always correctly translated it is still best to send a printed copy of the manuscript. Preferred word processing programs are Microsoft Word, WordPerfect, and MacWrite Pro. If possible, all words that must be italicized should be done so, not underscored. Tables may be formatted in a spread sheet program such as MS Works or MS Excel. Text should be double-spaced typing, with 25 mm left and right margins. Tables should be put in a separate file. Diskettes should be accompanied by the name of the software program used (e.g., WordPerfect, Microsoft Word). Authors should keep backup copies of all material sent to the Editor. The Society cannot be responsible for diskettes or text mislaid or destroyed in transit or during editing. Illustrations should be planned for reduction to the dimension of the printed page (14 X 20.5 cm, column width 6.7 cm) and allow room for legends at the top and bottom. Do not make plates larger than 14 X 18 in. (35.5 X 46 cm). Individual figures should be mounted on a suitable drawing board or similar heavy stock. Photographs should be trimmed, grouped together and abutted when mounted. Figure numbers should be on the plate. Include title, author(s) and address(es), and illustration numbers on back of each plate. Original figures need not be sent until requested by the editor, usually after the manuscript has been accepted. Reference to figures /tables in the text should be in the style "(Fig. 1)" "(Table 1)". Measurements should be in the metric system. Electronic plates may be submitted on disc or via email. They must be fully composited, labeled, and sized to fit the proportions of the journal page. Line art should be scanned at 1200 dpi (minimum input reso- lution is 600 dpi). Color or grayscale (halftone) images should have a dpi of 300-350. Color files should be in CMYK and not RGB. Graphics should be submitted as TIFF or EPS files. No PowerPoint or Word/Word Perfect files with images embedded in them are acceptable. All papers must conform to the International Code of Zoological Nomenclature. The first mention of a plant or animal should include the full scientific name including the authority. Genus names should not be abbre- viated at the beginning of a sentence. In taxonomic papers type specimens must be clearly designated, type depositories must be clearly indicated, and new taxa must be clearly differentiated from existing taxa by means of keys or differential diagnoses. Authors are required to deposit all type material in internationally recognized institutions (not private collections). Voucher specimens should be designated for specimens used in behavioral or autecological studies, and they should be deposited similarly. Acceptance of taxonomic papers will not require use of cladistic methods; however, authors using them will be expected to specify the phylogenetic program used (if any), including discussion of program options used. A data matrix should be provided if the subject is complex. Cladograms must be hung with characters and these should include descriptors (not numbers alone) when feasible. The number of parsimonious cladograms generated should be stated and reasons given for the one adopted. Lengths and consistency indices should be provided. Adequate discussions should be given for characters, plesiomorphic conditions, and distributions of characters among outgroups when problematical. References in the text should be (Smith 1999), without a comma, or Smith (1999). Two articles by a single author should be (Smith 1999a, 1999b) or Smith (1999a, 1999b). For multiple authors, use the word "and," not the symbol "&" (Smith and Jones 1999). For papers in press, use "in press," not the expected publication date. The Literature Cited section should include all papers referred to in the paper. Journal names should be spelled out completely and in italics. Charges. Publication charges are $10.00 per printed page. At least one author of the paper must be a member of the International Society of Hymenopterists. Reprints are charged to the author and must be ordered when returning the proofs; there are no free reprints. Author's corrections and changes in proof are also charged to the author. Color plates will be billed at full cost to the author. All manuscripts and correspondence should be sent to: Dr. E. Eric Grissell Systematic Entomology Laboratory, USDA Smithsonian Institution P.O. Box 37012 National Museum of Natural History CE 520, MRC168 Washington, DC 20013-7012 Phone: (202) 382-1781 Fax: (202) 786-9422 E-mail: egrissel@sel.barc.usda.gov 9088 01058 9604