Journal of © Hymenoptera Research April 2008 ISSN #1070-9428 Volume 17, Number 1 CONTENTS BENNETT, DANIEL J. The ophionine wasps of Hawaii (Hymenoptera: Ichneumonidae) ..... 1 BUFFINGTON, M. L. and J. LILJEBLAD. The description of Euceroptrinae, a new subfamily of Figitidae (Hymenoptera), including a revision of Euceroptres Ashmead, 1896 and the ERD IDGIN GDL GIME ISU 0S TS SS I Le a ia ope ne to COELHO, J. R., J. M. HASTINGS, C. W. HOLLIDAY, and A. MENDELL. Load carriage during nen EMIS DECIES OF SOMLATY WASPS... 262.3 < sss coe ne eee ee ee de ete eee 57. FEITOSA, R. M., C. R. F BRANDAO, and J. L. M. DINIZ. Revisionary studies on the enigmatic neotropical ant genus Stegomyrmex Emery, 1912 (Hymenoptera: Formicidae: Myrmi- oie) hyirer Le GESClIPLION OL LWO MEW SPECIES 222... e ee ee ce ee ee ee 64 GESS, F. W. The genus Quartinia Ed. André, 1884 (Hymenoptera: Vespidae: Masarinae) in southern Africa. Part Il. A new species with complete venation and with a deeply SDSS. SUS SLT ONS UN a0 e ee e ae n ee 83 GESS, S. K. and F. W. GESS. Patterns of usage of snail shells for nesting by wasps (Vespidae: Masarinae and Eumeninae) and bees (Megachilidae: Megachilinae) in southern EE ec ch Lise hos ER ile ARE 86 GONZALEZ, V. H. and M. OSPINA. Nest structure, seasonality, and host plants of Thygater aethiops (iymenoptera: Apidae, Eucerini) in the Andes ...:.........0.......0045- 110 PUCCI, T. A comparison of the parasitic wasps (Hymenoptera) at elevated versus ground vena pan beast beeen-Widple £OfeSE . 26s es. oe ee ee ee cee ee 116 CD REVIEW POLASZEK, A.—N. B. Stevens, C. J. Stevens, M. Iqbal, J. T. Jennings, J. La Salle, and A. D. Aus-. tin. What wasp is that? An interactive identification guide to the Australasian MN ETEE Pe ee sc 8 oe OR ye i SS eR yee we See Me ee 8s 124 INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991 OFFICERS FOR 2007 Michael E. Schauff, President James Woolley, President-Elect Michael W. Gates, Secretary Justin O. Schmidt, Treasurer Gavin R. Broad, Editor : Subject Editors SYMPHYTA AND PARASITICA ACULEATA Biology: Mark Shaw Biology: Jack Neff Systematics: Andrew Deans Systematics: Wojciech Pulawski All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, Plant Sciences Institute, Bldg. 003, Rm. 231 BARC-West, Beltsville, MD 20705, USA; Secretary, Southwestern Biological Institute, 1961 W. Brichta Dr., Tucson, AZ 85745, USA; Treasurer, PO Box 37012, c/o Smithsonian Institution, MNMH, MRC168, Washington, DC 20013-7012, USA; Editor, Dept. of Entomology, The Natural History Museum, Cromwell Road, Lon- don SW7 5BD, UK. Membership. Members shall be persons who have demonstrated interest in the science of entomol- ogy. Annual dues for members are US$45.00 per year (US$40.00 if paid before 1 February), payable to The International Society of Hymenopterists. Requests for membership should be sent to the Trea- surer (address above). Information on membership and other details of the Society may be found on the World Wide Web at http:/ /hymenoptera.tamn.edu/ish/. Journal. The journal of Hymenoptera 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, 0 Department of Entomology, Smithsonian Institution, 10th and Constitution NW, Washington, D.C. 20560-0168, U.S.A. Editor: Gavin R. Broad, Department of Entomology, The Natural History Museum, Cromwell Road, London SW7 5BD, UK. Managing Editor and Known Bondholders or other Security Holders: none. This issue was mailed 31 March 2008 J. HYM. RES. Vol. 17(1), 2008, pp. 1-43 The Ophionine Wasps of Hawaii (Hymenoptera: Ichneumonidae) DANIEL J. BENNETT Division of Entomology, Natural History Museum, and Department of Ecology and Evolutionary Biology, 1501 Crestline Drive-Suite #140, PSB, University of Kansas, Lawrence, Kansas 66049-2811, USA Abstract.— Hawaii's largest group of Ichneumonidae, the Ophioninae, is reviewed. Thirty species are recognized in one genus, Enicospilus Stephens. A key to species and a table of distributions indicating 26 new island records are provided. The following seven species are described as new: Enicospilus ashei, Enicospilus dorsolineatus, Enicospilus elekino, Enicospilus hainesi, Enicospilus gladiator, Enicospilus minimus, and Enicospilus petilus. Enicospilus tyrannus Perkins 1910 is newly synonymized with Enicospilus longicornis Ashmead 1901. The following genera are synonymized with Enicospilus: Abanchogastra Perkins 1902, Banchogastra Ashmead 1900, and Pycnophion Ashmead 1900. Replacement names are Enicospilus blackburni (= Enicospilus molokaiensis Ashmead 1901) and Enicospilus swezeyi (= Pycnophion fuscipennis Perkins 1910). “The variability of many of the Hawaiian Ophionini is so excessive, that if similar variation occurs in other tropical countries, the group may well prove one of the most difficult of entomological studies.”’ ee Ca Lherdiis toils Perkins’ prescience indeed foretold a challenge to systematic entomologists. Yet, at the time he couldn’t have fully known the magnitude of the problem. The ophionine genus Enicospilus alone is now represented by an excess of 650 described species (Yu and Horstmann 1997), with an untold diversity concentrated in tropical areas and particularly large radiations occurring in Madagascar and New Guinea (Gauld and Mitchell 1981). Likewise, in the most remote of tropical areas, the Hawai- ian Islands, Enicospilus has flourished and given rise to an array of species that comprise the majority of Hawaii’s native ichneumonids. Many of these are notable for morphologies and habits that differ strikingly from an otherwise homogeneous Enicospilus outside the islands (Gauld 1985, Bennett 2004). Such features include a variety of ovipositor lengths and shapes, drastic reductions in body size, stout body forms, and, concomitant with diurnal behavior, smaller eyes and dark coloration (e.g. Figs 1, 4E). Woefully little is known about the biology of most Hawaiian En- icospilus species, yet from the variety of Ovipositor types exhibited, and from one host record, it is evident that the evolution of this morphological exuberance is at least in part related to the attack of novel hosts. As koinobiont endoparasitoids, ophionine species are generally known to parasitize large, exposed caterpillars, particularly of the families Noctuidae, Lasiocampidae, Lymantriidae, Saturniidae, Geometridae, Arctiidae, and Sphingidae (Gauld 1988). The habits of several Hawaiian species are indeed consistent with this. Swezey (1931, 1954), however, reared Enicospilus swezeyt (Fig. 1), a species with a long, straight Ovipositor from the cosmopterigid Hypos- mocoma chilonella Walsingham concealed within Rubus stems. There are as of yet no host data for additional species with long, straight ovipositors or for those with long, curved ovipositors. As is the case for many Hawaiian insects, the first ophionine wasps were collected in Nm JOURNAL OF HYMENOPTERA RESEARCH Bie. 1, Enicospilus swezeyi. Hawaii by the minister naturalist Thomas Blackburn. During the years 1877-1883, he sent many insects to specialists in London including four ophionine wasps to Ca- meron who described them as male-female pairs of two species in the genus Ophion (Cameron 1883), though the original series actually contained four distinct species (Perkins 1915). These were appropriately transferred to Enicospilus (or the unjustified form Henicospilus) in subsequent catalogs (Szépligeti 1905, Morley 1912). Meanwhile, Ashmead was describing new species and genera of Hawaiian Ophioninae (Ashmead 1900, 1901) sent to him by Perkins who later complained bitterly about Ashmead’s “extraordinary’’ treatment of conspecific individuals, his mixtures of species under single names, and his habit of designating as types, individuals from locations other than those for which such species were named (the latter can be explained given that Ashmead didn’t designate holotypes per se, but rather often wrote “type” on each individual of his syntype set, which in some cases represented multiple islands). Perkins’ revision (1915) recognized six genera of Hawaiian Ophioninae and fully treated the species of Enicospilus, providing a key and noting many important charac- ters. Cushman (1944) attempted to use subgenera for a number of taxa (including VOLUME 17, NUMBER 1, 2008 two of Perkins’ genera) as a means of recognizing the increasing number of aberrant derivatives of Enicospilus in Ha- waii. His use of subgenus was not fol- lowed, but his key works well, and his review was important in showing that a good number of names were confused for widespread, polymorphic species. Since Cushman’s work, species-level taxa have generally remained stable, but the generic classification of these species has fluctuat- ed between the opinions of several authors. Townes (1945) further reduced the number of genera by synonymizing Abanchogastra and Banchogastra under Enicospilus. Cush- man (1947) recognized Pycnophion and Banchogastra as genera, but not Abanchogas- tra. Townes et al. (1961) took the same position as Townes (1945), but later Townes (1971) also raised Banchogastra to genus. That such confusion would reign regarding the genus-group status of these taxa is a result of the evolution of highly apomorphic morphologies and the subjec- tivity inherent in deciding which deriva- tives are sufficiently different to warrant removal from Enicospilus. The phylogenetic analysis of Gauld (1985) provided the first congruence test to indicate that Pycnophion, Banchogastra, and Abanchogastra were in- deed apomorphic, insular lineages derived from within Enicospilus. Recent and forth- coming cladistic analyses have upheld this view (Bennett 2004, in prep.). Gauld (1985), however, maintained the genus- rank status of these groups owing to his view that it was impractical to include highly aberrant derivatives within an oth- erwise morphologically and behaviorally homogeneous Enicospilus; this arrangement was upheld in a recent catalogue of Ichneumonidae (Yu and Horstmann 1997). Herein is proposed a classification that, for the first time, reflects the Enicos- pilus ancestry of all Hawaiian Ophioninae. Descriptions of new taxa, a summary of species and their distributions (Table 1), and an updated key to species are also provided. METHODS Morphological terminology, indices, and species description format generally follow Gauld and Mitchell (1981) and Gauld (1988); select additional terms are de- scribed by Townes (1969). Integumental sculpture terminology follows Harris (1979). Mandibles are described in refer- ence to a horizontal position as opposed to projecting ventrally. Malar space is mea- sured as the shortest distance between a point just above the anterior dorsal margin of the mandible and the eye. Fore wing length does not include the tegula. The cubital index (CI) of the hind wing is newly defined as the distance between the junc- tion of Cul (second abscissa) and cu-a and the junction of cu-a and 1A along an imaginary line between the junction of M+Cu and Cul (first abscissa) and the junction of cu-a and 1A (Fig. 5D-b) divided by the latter imaginary line (Fig. 5D-a). The ventral face of the mesopleuron is described as the ““mesosternum.” The “lower meta- pleuron” is used to mean that part of the metapleuron ventral and posterior to the propodeal spiracle. The angle of the anterior mesoscutum and the posterior declivity of the scutellum are estimated with reference to a horizontal line taken as a line between the cervix and the posterior foramen of the propodeum. Hind coxa length is measured from the basal constriction to the dorsal apical-most point in lateral view. Tergal numbers are in reference to the metasoma and not the true abdomen. The length of T2 is measured in lateral view between anteri- or and posterior dorsal midpoints. Many of the characters previously point- ed out as critical to the delineation of ophionine species (Gauld and Mitchell 1981, Gauld 1988) are likewise important in Hawaiian Enicospilus. Paramount among these is the form of the hairless region of the discosubmarginal cell, or fenestra, and the sclerites often accompanying it. On this basis alone, many species can be identified. Also important are the mandible shape, de Table 1. unverified literature records. JOURNAL OF HYMENOPTERA RESEARCH Distributions of Hawaiian Ophioninae. New island records indicated by *. Lower case “’x’”” denotes Species Kauai Oahu Molokai Lanai Maui Hawaii Enicospilus ashei Bennett A ip Enicospilus bellator Perkins 1915 3 x x X X Enicospilus blackburni Bennett X X X x X X Enicospilus castaneus Ashmead 1901 X X X X* X Enicospilus debilis (Perkins 1902) X X X Enicospilus dispilus Perkins 1902 X X X X* X X Enicospilus dorsolineatus Bennett X* Enicospilus elekino Bennett X* Enicospilus ferrugineus (Perkins 1915) X* X Enicospilus fullawayi Cushman 1944 X X* X* X* Enicospilus gladiator Bennett X* Enicospilus hainesi Bennett AG Enicospilus hawaiiensis (Ashmead 1900) X Enicospilus kaalae Ashmead 1901 X X X X X Enicospilus kauaiensis (Ashmead 1901) X D Enicospilus lineatus (Cameron 1883) X X X x X X Enicospilus longicornis Ashmead 1901 X X X* X X Enicospilus melanochromus Perkins 1915 x X X* x X* Enicospilus minimus Bennett X* X* Enicospilus molokaiensis (Ashmead 1900) X X X Enicospilus niger (Ashmead 1900) X Enicospilus nigrolineatus Ashmead 1901 X X X X X X Enicospilus orbitalis (Ashmead 1901) X X X X X Enicospilus perkinsi Cushman 1944 X Enicospilus petilus Bennett X* 2 Ses X* Enicospilus pseudonymus Perkins 1915 he X Enicospilus swezeyi Bennett X X* X Enicospilus variegatus Ashmead 1901 X Enicospilus vitreipennis (Perkins 1910) X* X* X X* Enicospilus waimeae Ashmead 1901 X x Total 30 18 17 13 5 19 20 upper tooth shape, malar space, size and shape of the compound eye and ocelli, shape and sculpture of the scutellum, hind wing venation, pretarsal claws, propodeal sculpture, metasoma shape, ovipositor shape, and color. The posterior propodeal carina is not known to occur in any Hawaiian Enicospilus; its absence is not repeated in the descriptions. Institutions and their acronyms are as follows: American Entomological Institute, Gainesville, FL, USA (AEIC); Bernice Pauahi Bishop Museum, Honolulu, HI, USA (BPBM); Canadian National Collection, Ottawa, On- tario, Canada (CNCI); The Natural History Museum, London, UK (BMNH); The United States National Museum (USNM); University of Hawaii, Manoa, HI, USA (UHM). KEY TO SPECIES OF HAWAIIAN OPHIONINAE —" . Hind wing with first abscissa of Rs < 2.0 as long as rs-m (Fig. 7E), 1A absent and second abscissa of Cul present only as a short stub; very small, fore wing length about 5 mm or less 28. oe 2 ee ee Oe ee ewe Oe Wee eee ee Oe we ee ee ee ne Ore ee ee E. minimus VOLUME 17, NUMBER 1, 2008 5 Co ! © 1. WG 12. Hind wing with first abscissa of Rs > 2.0X as long as rs-m, bothlA and second abscissa of Cul distinct (Fig. 5D); larger, fore wing at least 6 mm, usually much eae iy PRR At EOD. 39). NE ORE HOLA OR ORG RE DONT BOE. Ds 2 Fore wing discosubmarginal cell with or without sclerites, with distinct fenestra present; posterior transverse carina of mesosternum present medially ............ s) Fore wing discosubmarginal cell without sclerites, fenestra absent (Figs 23, 29), or if present, then only as a poorly defined region of reduced pubescence without distinct lower margin (Figs 17, 22, 27); posterior transverse carina of mesosternum absent mecialilypionsispresent them Weak ater SIE OUR hE te D9. Fore wing with 1lm-cu usually evenly curved or arched, if angulate and swollen medially, then only slightly so (Figs 10, 21, 26); ovipositor short and straight or curved, sper Henetnnoml2vorless: MI GUn Je60) 18. EUG, StS IY, SERVI SION. +. Fore wing with 1m-cu medially strongly angulate, swollen (Fig. 6), or with short stub (Figs 5C, 15); ovipositor long and straight, about 1.8< length of T2 or more ...... 29 Metasoma extremely slender, dorsomedial length of exposed portion of T5 in female, Hein male/-ereatemiinanvlateraldep thy te 4 HERG al) Vl eh RED ITM oo 5 Metasoma not as slender, dorsomedial length of exposed portion of T5 in female, T4 in impeller eschtirantlateraladepth Geni oA ew. eee HY SE ot EC WN ON ae « 8 Fore wing discosubmarginal cell with 2 sclerites, proximal one very large GiiemOID ays Mies. 38 OD Ree oe oe ee A ee i oe Pe eRe E. petilus Fore wing discosubmarginal cell without sclerites, or if one present, then not BpORGachinamnNusiZeitnak OmabOverts hata ie bint ESP EA QR ots 6 Fore wing discosubmarginal cell with fenestra broad, posterior margin extending beyond midpoint between Rs+2r and 1m-cu, with a single, linear sclerite at proximal, Men tidlenmankcimohtencstiran (hie 214) ge Wise Pl Le ee E. fullawayi Fore wing discosubmarginal cell with fenestra round and smaller, ventral margin not extending beyond midpoint between Rs+2r and im-cu, if sclerite present, then Sp WeHiCalhuns saat: Bike wae» Of Anite aT ay ARR fey NS MPD RAR AE 5 . Ip Fore wing discosubmarginal cell without a sclerite (Fig. 25) ............ E. orbitalis Fore wing discosubmarginal cell with a single sclerite (Fig. 26) .......... E. perkinsi itesosoma, yellow. or, yellow-and, black 255. hs... 410.I0IPR. WEAR EI, #: 9 Niescsonia brown pred) oraneevori black yi s6 2% 2 £2 GNAOO RL AI). A hs 12 Mandible with a heavily setose, diagonal groove (Fig. 36); fore wing discosubmarginal cell with 2 or 3 sclerites, central sclerite oval or triangular and medially placed in hemestnan Eiee20) Jovipositor downeurved (4x Bae. nine. Lee) 22s E. longicornis Mandible with at most a weakly to moderately setose, diagonal line (Fig. 37); fore wing discosubmarginal cell with 1 or 2 sclerites, if 2, then the second positioned along wemralimarom of fenestra;:ovipositor straight oo.. ses Yee be eels ov aeT lee. 10 Fore wing discosubmarginal cell with 2 sclerites (the second may be translucent; showing weakly in figure) (Fig. 24); hind femur yellow or yellowish-brown throughout; propodeum in large part black; metasoma yellow or yellowish-brown Sreepiron lateralndark hneaiiycnsai's biecs. Ye, BO28ES, Ue 8 E. nigrolineatus Fore wing discosubmarginal cell with 1 or 2 sclerites (Figs 3C, 28); hind femur as described above or with apex black; propodeum as above or yellow to yellowish brown throughout; metasoma yellow, yellowish-brown, or black, never with lateral Sache lice een ey ee Mae) Ae Pen tech ROR ENED ea Tae es 11 Malar space 0.4-0.5X basal mandibular width; metasoma mostly yellow or yellowish- brownyexcept forardorsomedial darkiline i Sits 2G We BEN E. dorsolineatus Malar space 0.3-0.4 basal mandibular width; metasoma mostly black, not forming a Glonsomiedial Aimer rae sae: va a VOLS LET SID EI TS Ole 2) E. variegatus Fore wing discosubmarginal cell with fenestra large, extending posteriorly 3/4 or more the distance between Rs+2r and 1m-cu, and apically to about midpoint of Rs+2r or further; with 1 small, oval or attenuated sclerite, positioned at proximal posterior 14. 16. 17. 18. 19. JOURNAL OF HYMENOPTERA RESEARCH margin of fenestra, nearer 1m-cu than Rs+2r (Figs 10, 16, 21); female S7 enlarged (Fig.32); ovipositor straight) ;2).4 102. .onet. oo. eee Paes Eee meet 13 Fore wing discosubmarginal cell with fenestra smaller, posterior margin extending about 2/3-1/2 or less the distance between Rs+2r and 1m-cu, apically to about midpoint of Rs+2r or less; if alar sclerite(s) present, then variously shaped and positioned; female S7 not enlarged (Fig. 31), or if enlarged, then ovipositor downcurved. (Fig..33), > s.!os:s.2lw sai -rat tues Areata ee Se Sie eee Ce 15 Fore wing with 1lm-cu not thickened or angled medially (Fig. 16), alar sclerite spherical, not attenuated apically; middle segments of metasoma deep reddish or orangish-brown, usually strongly contrasted with much darker petiole and apical seomeniss oui. bod feviawdmeny es Gas AO. ee ae eee eels ects eee E. kaalae Fore wing with 1m-cu usually at least slightly thickened and/or angled medially (Figs 10, 21), alar sclerite often attenuated apically; middle segments variable in color but usually not strongly contrasted with petiole and apical segments ........... 14 Fore wing with 1m-cu usually slightly angled medially; fenestra not or poorly defined proximal of sclerite (Fig. 10); aedeagus slender apically; light reddish-brown to orange in colet.. ..5% desrsttan baccdee te tit be heeete toler a eee E. blackburni Fore wing with lm-cu not angled medially; fenestra broad and well defined proximal of sclerite; aedeagus bulbous apically; usually dark in color................... ee a ar 2 a ee, Meng ne ee eA Eo at E. melanochromus Dorsal surface of scutellum more or less flat, rugulose and/or coarsely pitted, scutellar carinae strong, often produced above medial part of scutellum; malar space 0.3—-0.7X basal mandibular width; male apical tarsomere parallel-sided or swollen basally in dorsal view, strongly curved: in Jateral view |) /o5 sce) «2 ee Wsiee! helt) ea 16 Dorsal surface of scutellum convex and lightly punctate, scutellar carinae weak or moderate, not produced above medial part of scutellum; malar space 0.1—-0.4x basal mandibular width; apical tarsomere of male evenly broadened apically in dorsal view, straight to moderately curved in lateral view ...°..........=:. =... 233 18 Fore wing with fenestra very small, without a distinct sclerite but with a faint sclerotization or pigmentation at posterior margin (Fig. 9); ratio of head height to width in frontal:view about 1.1 <. .... bait hadveeiss a6 ere See E. bellator Fore wing with fenestra at least slightly larger, with 1 or 2 sclerites, or rarely with none or a vestigial proximal sclerite; ratio of head height to width in frontal view = O94 vdnitions due. sa ewecizk «6 leu eeitiee Hebige: Sete tp: © oe ibs Malar space long, 0.4-0.7 basal mandibular width; flagellum of female short, length equal to or less than that of fore wing; fore wing fenestra usually with 1 sclerite (Fig. 11), occasionally with none or with a second vestigial, medially placed scletiie 2068) Oe. Lal. 4 a ee eee ee eee E. castaneus Malar space 0.3-0.5 basal mandibular width; flagellum of female longer, length equal to or greater than that of fore wing; fore wing fenestra usually with 2 distinct sclerites (Fig. 13), occasionally the second, medially placed sclerite is weak, or a third sclerite at distal margin of fenestra is apparentt.«:... +426). «...:3eiP eee eee eee E. dispilus Fore wing discosubmarginal cell without an alar sclerite ..................... 19 Fore wing discosubmarginal cell with at least 1 alar sclerite .................. 20 Orange except apical segments of metasoma black; mandible slender, with upper tooth medially swollen and long, 1.4-1.6* length of lower tooth (Fig. 2D) ... E. ashet Usually brown or reddish-brown, apex of metasoma at most slightly darker; mandible moderately stout, upper tooth not conspicuously swollen medially, 1.1-1.6< length of lower toothy 0.4) clio ow 42elae re oR Ries Mahe eo Aa RE otnees E. lineatus Mandible with a heavily setose, diagonal groove (Fig. 36); fore wing discosubmarginal cell with a large triangular proximal sclerite, a distinct central sclerite, and often a third pale sclerite outlining distal ventral margin of fenestra (Fig. 20); ovipositor downcurved.(Fig. 33) »0.: 5-22. ¢ews! a. petmaeetinee: lees iat £ oe. ee E. longicornis VOLUME 17, NUMBER 1, 2008 7 24. 20: 26. a 28. ee) Mandible with at most a moderately setose, diagonal line (Fig. 37); fore wing discosubmarginal cell usually with a single variously sized sclerite, rarely with a second medial sclerite, never with a third apical sclerite; ovipositor pote Nita eee ees 2,4 ieee Ye a, ol area, RL RIC ers ANA mee es ait te. Zl Fore wing discosubmarginal cell with a single, extremely large proximal sclerite (EGO) ieee Shel as Sake Sets ee a err eee E. waimeae Fore wing discosubmarginal cell with proximal sclerite smaller, at most as in Fig. 19, pouclivemuichestealle: (iC S) tai ts ses ee ass ET ARE eer... E. lineatus Metasoma more or less black, rarely with a deep reddish tint; fore wing cHiscOsuomancimalieell without atseletite .ANMIents 2 eo. My is SR. oe. 29 Metasoma of typical brown, red, or orange coloration; fore wing discosubmarginal cell aOR SCleCiee ye Mere. -eetT- Soreieywes. Hee es 8. claws 28 Head entirely black; compound eye highly reduced (Figs 4A, B); propodeum evenly colliculate, without rugae; mid coxa with ridges dorsomedially ........... E. elekino Head entirely black or with pale areas; if compound eye reduced, then propodeum coarsely rugose, areolate, or rugostriate; mid coxa at most slightly wrinkled dessomed talline. etaaniae ace fanys ss Aaa tie tare oa eye ieee» 24 Ovipositor shorter than petiole; compound eye highly reduced (cf. Figs 4A, B); fore wing discosubmarginal cell usually densely setose throughout, usually with no trace of a fenestra (Figs 23, 29); propodeum coarsely rugose, areolate, or rugostriate, with SOMO LeELIOr-Lransverse Calildy nf 2, +. a kt. RO AGM oe Mlle ibs. be eho} a5) Ovipositor about twice petiole length or more; compound eye not reduced; fore wing discosubmarginal cell markedly less setose, especially proximally, often with a vestigial fenestra or poorly defined area of reduced pubescence below Rs+2r (Figs 17, 22); propodeum evenly colliculate or moderately rugose, if the later, then anterior NCS LTEV CUSS CET BIO SSITIE TER “stow want 8 oe Reg a ean ne re meee 26 Petiole very compact, bulbous apically, ventral posterior midpoint positioned far anterior such that the ratio of ventral to dorsal length = about 0.4 or less (Fig. 34) (measured in lateral view from sub-basal narrowing); T2 wider than long in dorsal view; fore wing dark brown anteriorly, highly contrasted with lighter posterior apical EIUSAL | Tea Ss Me oat ala gl Elec ae aanaemmete Ur war. A. Woe Ie eae oA aS OSES SEE EI or E. niger Petiole not as compact, flatter apically, ventral posterior margin usually positioned further posterior such that the ratio of ventral to dorsal length = about 0.5 (Fig. 35); T2 usually longer than wide in dorsal view; fore wing variously light or dark, anterior and posterior apical area usually of similar hue or only slightly CSITPES PEOUOMEE Os, 8) 5 Bes cae deans, 8 dandy en er er oe rr E. vitreipennis Ovipositor straight; anterior transverse carina of propodeum absent; setae of dorsomedial propodeum posteriorly directed; propodeum moderately rugose, at least pestenory, mesosomausually mostly ted ys eile g vena s Ske ee es E. swezeyi Ovipositor upcurved; anterior transverse carina of propodeum present or absent; setae of dorsomedial propodeum erect or curved anteriorly; propodeum evenly colliculate MEOUSHOUL, MesOsoOMma Usually mostly black4 i222 Het S Pelee oP oe oe 27 micaaaniG MmleSesoma-allDlack . sot; . at Boe Pee, eee E. kauaiensis Head and/or mesosoma with pale or red areas .................. E. molokaiensis Mandible with teeth stout, upper tooth shorter than lower tooth; posterior mesonotum and scutellum with lateral, longitudinal depressions; relatively large, fore wing length Lia as iavlily Oystin Ulf ope etmnrerel anit « - a ene ‘" Pose ty ae ot —— +. wa 3 ; , ee fe Fig. 5. Enicospilus gladiator: A, frontal aspect of head; B, dorsal aspect of head; C, discosubmarginal cell of fore wing; D, hind wing, cubital index = ratio of b to a. female in dorsal view evenly broadened apically, 3.4-3.6X as long as broad, in lateral distally, 3.0-3.1X as long as broad, in lateral view nearly straight or slightly curved; view nearly straight; 5" tarsomere of male pretarsal claw of female evenly curved, in dorsal view somewhat abruptly widened with 7-10 pre-apical teeth (cf. Fig. 6C); 20 pretarsal claws of male longer and densely pectinate (cf. Fig. 6D). Metasoma: Fairly compact and apically deep in female; T2 in female 4.8-5.7X as long as lateral height, 2.7-3.6X as long as dorsal width; T2 in male 5.2-8.9X as long as lateral height, 4.3-6.2x as long as dorsal width; thyridium tear- shaped to elliptical, positioned posterior of anterior margin of T2 by 0.3-0.4X length of T2. Ovipositor long and straight, about 2.0X length of T2, with slight swelling distal of midpoint. Color: Generally evenly brown to brown- ish-red throughout, paler on face, behind compound eyes and, in some individuals, variously light brown on anterior prono- tum, notaulus, subalar prominence, mese- pimeron, and distal leg podites (other than apical tarsomeres and pretarsus which are dark); wings hyaline to slightly infumate; setae white to pale brown. Material examined.—Holotype: female, Ha- waii, Kauai: Kokee State Park, near main entrance, 22.1194°, —159.6670°, 1084 m eleva- tion, 27-28 May 2006, UV light trap (D. J. Bennett) (BPBM). Paratypes (13, all from Kauai): 1 male, Kokee Camp, 3600 ft elevation, 29 March 1961 (D. F. Hardwick) (CNCI); 1 female, Kokee, 3400 ft elevation, 16 August 1961 (G. and J. Holland) (CNCI); 1 male, Na Pali- Kona Forest Reserve, Milolii Ridge, 3000 ft elevation, 26 June 1982 (K. and E. Sattler) (BMNH); 1 female, Kokee State Park, Kaluapuhi Trail, about 5 miles from lower trailhead, approximately 22.1432°, —159.6421°, 1150 m elevation, 28-29 May 2006, UV light trap (D. J. Bennett) (BPBM); 2 females, Kokee Road, 4000 ft elevation, 19 May 1982 (J. W. Beardsley) (BPBM); 1 female, Kumuwela, 8 August 1921, (Swezey) (BPBM); 2 males, Nualolo Valley, 3400 ft elevation, August 1952 (D. E. Hardy) 5? BM); 1 male, Kokee, 13-17 September 1965 (Bi 2 females, Kokee State Park, Discovery Ceri May 1998 (M. J. and C. A. Tauber) (BPB Kokee, 4—6 August 1961 (Maa, Miyat himoto) (BPBM). Etymc species epithet, a Latin noun for irer,’” is in reference to the long, str. of this species. ‘positor characteristic JOURNAL OF HYMENOPTERA RESEARCH Enicospilus hainesi, new species Fig. 6A—-E Diagnosis.—The long, broad fenestra lacking alar sclerites is unique among Hawaiian Ophioninae (Fig. 6E). Also help- ful in recognizing this species is the shape of 1m-cu and the long, straight ovipositor. Description.—Length of fore wing 7.4— 9.4mm in female, 6.6-8.7 mm in male. Head: Mandible moderately slender to moderately stout, evenly or somewhat abruptly tapered proximally, slightly twist- ed; basal ventral margin at least slightly concave; outer surface with setae scattered or loosely aggregated medially, without diagonal groove, basal concavity shallow; upper tooth long, 1.2-1.5x as long as lower tooth, about equal in width to lower tooth or the latter slightly wider at base. Labrum 0.2-0.3X as long as broad, apical margin broadly rounded. Malar space 0.3-0.5X as long as basal mandibular width. Clypeus in profile weakly to moderately convex, weakly to moderately distinct from lower face; in frontal view 1.6-2.0X as broad as long, coriaceous to colliculate, sparsely and lightly punctate, apical margin sharp, broadly flat, weakly to distinctly impressed medially. Lower face 0.7-0.8X as broad as long, lightly punctate and coriaceous to colliculate. Compound eye large and strongly convex, head width in frontal view 1.2-1.3X length (Fig. 6A). Gena with setae short, inconspicuous and declined forward; in dorsal view moderately round- ed behind compound eye (Fig. 6B); GOI = 2.2-3.1. Ocelli large, lateral ocellus re- moved from compound eye by 0.1-0.2x its diameter; FI = 0.4—0.6. Occipital carina dorsally flat or broadly rounded, ventrally joining hypostomal carina. Flagellum in female 1.3-1.5x length of fore wing, with 44-45 segments, mid segment 1.9-2.3X as long as broad; in male 1.5-1.6x length of fore wing, with 45-51 segments, mid segment 2.0-2.2 as long as broad. Meso- soma: Mesoscutum strongly rounded in profile, anterior angle 70°-80°; notauli VOLUME 17, NUMBER 1, 2008 weak to distinct. Scutellum in dorsal view 1.2-1.4X as long as anteriorly broad, with upper surface more or less flat to weakly convex, colliculate; lateral carinae moder- ately weak to moderately strong, moder- ately convergent, extending near entire scutellar length, albeit weakly so posteri- orly; posterior declivity angled by 30°-45°, striate, nearly smooth or colliculate. Meso- pleuron colliculate and transversely rugu- lostriate; scrobe distinct or indistinct; spec- ulum distinct or indistinct; mesopleural sulcus with weak transverse ridges; epic- nemial carina strong, complete medioven- trally. Mesosternum with or without shal- low depression behind lateroventral corner of epicnemial carina; with posterior trans- verse carina complete medioventrally. Lower metapleuron moderately convex, colliculate and/or rugulose. Propodeum in profile weakly convex; with setae low- lying and posteriorly declined; anterior furrow strong, rugostriate, anterior area about 0.1 total propodeal length; anterior transverse carina in female strong, extend- ing laterally almost to lower metapleuron, in male narrowly present as a weak medial vestige; posterior transverse carina absent; spiracular area smoothly colliculate, 0.2- 0.3X total propodeal length; posterior area rugose becoming areolate in part posteri- orly in some individuals. Separation be- tween propodeum and metapleuron indi- cated by a weak furrow posteriorly and a distinct, evanescent, or irregular carina anteriorly. Fore wing (Fig. 6E) with pter- ostigma long, extending about 3/4 the length of Rs+2r, fairly abruptly narrowed distally; discosubmarginal cell without sclerites (vestiges apparent in some indi- viduals), fenestra moderately long and broad, extending apically to 2/3-3/4 the length of Rs+2r and posteriorly to near 1m- cu; Rs+2r thickened medially, at least slightly arched (rather than sinuous); Rs+M slightly to distinctly arched, at least basally; 1m-cu strongly angulate medially with a distinct swelling or short stub projected anteroproximally; AI = 1.5-2.6; 21 CI = 0.3-0.5; ICI = 0.2-0.3; SDI = 1.0-1.1; cu-a anterior of Rs+M by 0.3-0.5 length of cu-a; 1* subdiscal cell sparsely and evenly pubescent or nearly devoid of setae throughout. Hind wing with 5-6 hamuli in distal set; 1* abscissa of Rs slightly concave basally or sinuous, 2°* abscissa more or less straight; 2"* abscissa of Cul positioned well below midpoint between M and 1A, CI = 0.2-0.4. Fore leg with tibia 7.4-9.1X as long as wide, subapical spines absent or present as few and weak. Mid leg with coxa colliculate, becoming slightly rugulose dorsoposteriorly; inner tibial spur 1.2-1.4X as long as outer spur. Hind leg with coxa in lateral view 1.5-1.8X as long as deep, colliculate and at least slightly rugulose dorsoposteriorly; trochantellus 0.4-0.6x as dorsally long as broad; 4" tarsomere of female 2.2-2.7X as long as broad, 2.4—2.9X in male; 5" tarsomere of female 2.4-2.9X as long as broad, evenly broadened apically in dorsal view, nearly straight to slightly curved in lateral view; 5 tarsomere of male 3.1-3.3X as long as broad, evenly broadened apically in dorsal view, in lateral view moderately curved; pretarsal claw of female and male approx- imately as figured (Fig. 6C, D). Metasoma: Apically deep in female; T2 4.3-5.3 as long as lateral height, 2.8-4.2 as long as dorsal width; thyridium narrowly oval to tear-shaped, midpoint positioned posterior of anterior margin of T2 by 0.3-0.4 length of T2. Ovipositor long and straight, about 2.1X length of T2. Color: Generally brown or reddish-brown; head lighter brown except variously darker on clypeus, face medially, frons, and poste- rior gena; wings hyaline to slightly infumate; legs slightly lighter than body except coxae (and in some individuals additional basal podites); setae white to pale brown. Remarks.—This species is only known from rather recently collected material taken from within and near Honolulu. Despite the suspicious locality data (one such location is a Honolulu Pier), it is an obvious native Hawaiian insect, sharing i 9) “—_ Fig. 6. JOURNAL OF HYMENOPTERA RESEARCH Enicospilus hainesi: A, frontal aspect of head; B, dorsal aspect of head; C, male hind outer claw; D, female hind outer claw; E, discosubmarginal cell of fore wing. typical general features of Hawaiian En- icospilus (colliculate punctation and lack of fore-tibial spines) and further resembles particular Hawaiian species such as E. kaalae (fenestra shape) and E. hawaiiensis (long, straight ovipositor). It is peculiar that it was never found by the early, prolific Hawaiian entomologists such as Perkins and Swezey who collected heavily in areas such as Tantalus where it occurs. The late Hyme- nopterist Dr. John Beardsley realized this conundrum and left a note on a specimen in the BPBM postulating a switch to a non- native host as a possible explanation (one label even reads “Manoa vegetable gar- den’’). Presumably under this scenario it was formerly rare and escaped capture but is more common now owing to this change. Discovery of the host range of this species both within Honolulu and its surrounding forests could be of potential general interest. Material examined.—Holotype: female, Ha- waii, Oahu, Mt. Tantalus, elevation 1900 ft, 20 VOLUME 17, NUMBER 1, 2008 September 1985 (W. E. Perreira) (BPBM). Para- types (12 all from Oahu): 1 female, Poamoho Trail summit, elevation 2500 ft, 5 May 1995 (A. Asquith) (BPBM); 1 female, 1 male, Manoa, 4 October 1984, “vegetable garden,” (K. Rhoads) (BPBM); 1 female, Ko’olau Mts, Wiliwilinui Trail, 6 June 2006, UV light trap (J. Eiben and W. Haines) (Manoa); 2 females, 3 males, Waianae Mts, Honouliuli Preserve, Palikea Trail, elevation 2700 ft, 15-16 May 2006, UV light trap (W. Haines) (Manoa); 1 male, Hono- lulu, Pier 32, 28 November 1978, light trap (Beardsley) (BPBM); 1 female, 1 male, Kaluaa Gulch, 18 November 1984, (Perlman) (BMBP). Etymology.—This species is dedicated to the Lepidopterist Will Haines who collect- ed the majority of specimens known of this species. Enicospilus hawatiensis (Ashmead) ies 15 Pleuroneurophion hawatiensis Ashmead 1900: 86. Holotype (by monotypy) female, Hawaii [presumably Hawaii Isand], Koebele [Coll.] (USNM, #5553); transferred to Enicospilus by Cushman 1944. Ashmead 1901: 342. Enicospilus (Pleuroneurophion) hawatiensis (Ash- mead); Cushman 1944: 46. Enicospilus hawaiensis [!] (Ashmead); Townes 1945: 737. Townes 1957: 116. Yu and Horst- mann 1997: 741. Enicospilus hawatiensis (Ashmead); Cushman 1947: 466. Townes et al. 1961: 277. Gupta W872 539) Remarks.—This small to medium-sized species (fore wing length 8.5-13.0 mm) can be recognized by the combination of its long, straight ovipositor; Im-cu of fore wing with a medial stub (Fig. 15); disco- submarginal cell without a sclerite; and brown to slightly reddish-brown coloration (with typical exception of yellowish areas of the face, clypeus and gena; wings are hyaline or very slightly infumate). Enicospilus kaalae Ashmead Fig. 16 Enicospilus kaalae Ashmead 1901: 347. Lectotype (designated by Townes et al. 1961: 278 [Perkins’ (1910: 678) claim that the type of 23 this species is from Kauai is not herein regarded as a valid lectotype designation as it does not refer to an individual from Ashmead’s syntype set as published.]) male, Oahu, Ka’ala Mts, 6500 ft (USNM). Perkins 1910: 278. Perkins 1913: cix. Perkins 1915: 524. Anonymous 1955: 386. Townes et al. 1961: 278. Gupta 1987: 545. Yu and Horstmann 1997: 742. Enicospilus semirufus Perkins 1902: 142. Lecto- type (designated by Townes et al. 1961: 278 [Their usage of “type” is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) female, Oahu, Honolulu Mts [published as Ko’olau range] (BPBM) [examined]; synonymized by Perkins 1910. Henicospilus kaalae (Ashmead); Szépligeti 1905: DT Henicospilus semirufus (Perkins); 5zépligeti 1905: 27. Morley 1912: 52. Enicospilus (Enicospilus) kaalae Ashmead; Cush- man 1944: 47. Remarks.—This large (fore wing length 12.5-17.5 mm) and common species (par- ticularly on Kauai) is consistently colored dark brown to black on the mesosoma, petiole, and apically on the metasoma; the medial part of the metasoma is a charac- teristic deep reddish to orangish-brown, and the wings are hyaline to slightly infumate. It is further recognized by the broad fenestra with a single, posteriorly positioned, oval sclerite (Fig. 16). Enicospilus kauaiensis (Ashmead), new combination Fig. 17 Pycnophion kauaiensis Ashmead 1901: 344. Lec- totype (designated by Townes et al. 1961: 295) female, Kauai, 4000 ft, VII.[18]96 (BMNH). Gupta 1987: 505. Yu and Horstmann 1997: 761. Remarks.—This rare species is easily recognized by the combination of its small size (fore wing length about 8.0 mm); dark brown to black coloration (including face, clypeus and gena), weakly to distinctly infumate wings (particularly apically); long, upcurved ovipositor; evenly collicu- 24 late propodeum; and discosubmarginal cell without a fenestra or sclerite (Fig. 17). Enicospilus lineatus (Cameron) Figs 13,.19,37 Ophion lineatus Cameron 1883: 192. Lectotype (designated by Townes et al. 1961: 279 [Their usage of “type” is herein regarded as equiv- alent to a lectotype designation (ICZN 1999: Art. 74.5).]) female [metasoma absent], Lanai, (BMNH); transferred to Henicospilus by Mor- ley 1912 and to Enicospilus by Perkins 1915. Blackburn and Cameron 1886: 179. Blackburn and Cameron 1987: 240. Ashmead 1901: 341. Dalla Torre 1901: 192. Szépligeti 1905: 31. Enicospilus mauicola Ashmead 1901: 347. Lecto- type (designated by Townes et al. 1961: 279) female, Molokai, Mts, 4500 ft (USNM); syn- onymized with E. lineatus (Cameron) by Cushman 1944. Perkins 1907a: 44. Perkins 1915: 526. Enicospilus henshawi Ashmead 1901: 349. Lecto- type (designated by Townes et al. 1961: 279 [Their usage of “type” is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) female, Hawaii [Is.], Hilo, May (lost); synonymized with E. lineatus (Cameron) by Perkins 1915. Enicospilus dimidiatus Perkins 1902: 143. Lecto- type (designated by Townes et al. 1961: 279 [Their usage of “type” is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) female, Oahu, Wailua [published as Ko’olau range], 1500 ft, R.C.L.P. [Perkins] (BPBM) [examined]; syn- onymized with E. mauicola Ashmead by Perkins 1915 and with E. lineatus by Cushman 1944. Perkins 1907a: 44. Perkins 1910: 679. Henicospilus mauicola (Ashmead); Szépligeti 1905: 27. Henticospilus dimidiatus (Perkins); Szépligeti 1905: 27. Morley 1912: 52. Henicospilus henshawi (Ashmead); Szépligeti 1905: 27. Enicospilus capnodes Perkins 1910: 679. Lectotype (designated by Townes et al. 1961: 279 [Their usage of “‘type’’ is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) male, Hawaii [Is.], Kona, (00 ft, (BPBM) [examined]; synonymized with —. mauicola Ashmead by Perkins 1915 an th E. lineatus by Cushman 1944. JOURNAL OF HYMENOPTERA RESEARCH Henicospilus lineatus (Cameron); Morley 1912: 47,52. Uchida 1928: 219. Chu 1935:14. Uchida 1937: 11. [All misidentifications of Enicospilus lineolatus (Roman) (according to Gauld and Mitchell 1981, and Gupta 1987) except second reference by Morley (page 52).] Enicospilus funereus Perkins 1915: 525. Lectotype (here designated) female, w. [West] Maui, 1500 ft, IIl.02, R.C.L.P. [Perkins] (BPBM) [examined] [Townes et al. 1961: 279 errone- ously described syntypes from other is- lands.]; synonymized with E. lineatus (Ca- meron) by Townes et al. 1961. Swezey and Williams 1932: 182. Cushman 1944: 51. Enicospilus lineatus (Cameron); Perkins 1915: 526. Cushman 1944: 50. Iwata 1950 [misiden- tification, likely of Enicospilus lineolatus (Ro- man)]. Townes et al. 1961: 279. Lee and Kim: 1980: 11 [misidentification, likely of Enicospi- lus lineolatus (Roman)]. Gauld and Mitchell 1981: 8. Gupta 1987: 548. Yu and Horstmann 1997: 743. Enicospilus ashmeadi Perkins 1915: 527. Lectotype (designated by Townes et al. 1961: 279) female, Hawaii [Is.], Kilauea, VII.03, (BPBM) [exam- ined]; synonymized with E. lineatus (Cameron) by Cushman 1944. Anonymous 1925: 11. Remarks.—This species displays an im- pressive amount of variation in features including size (fore wing length 8.5— 16.4 mm), color (monochrome or mixed, ranging from dark brown or almost black to reddish-brown or orange, with frequent exception of yellowish areas of the face and gena; wings vary from slightly to strongly infumate), and the form of the alar sclerite (Figs 18, 19). This was pointed out by Cushman (1944) who synonymized a num- ber of names under E. lineatus. Additional specimen material and a focused study of the variation and distribution of forms could potentially reveal cryptic species and/or forms representing early phases of divergence. Enicospilus longicornis Ashmead Figs 20, 36 Enicospilus longicornis Ashmead 1901: 350. Lec- totype (designated by Townes et al. 1961: 280 [Their usage of “type” is herein regarded as VOLUME 17, NUMBER 1, 2008 equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) male [labeled as female], Hawaii [Is.], Kilauea (BMNH). Perkins 1915: 524. Swezey and Williams 1932: 182. Townes et al. 1961: 280. Gupta 1987: 551. Yu and Horstmann 1997: 743. Henicospilus longicornis (Ashmead); Szépligeti 1905; 27. Enicospilus tyrannus Perkins 1910: 678. Lectotype (designated by Townes et al. 1961: 292 [Their usage of “type’’ is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).|) female, Molokai, 4000 ft, 11.02 (BPBM) [examined]; new synonymy. Perkins 1915: 524. Anonymous 1925: 11. Swezey and Williams 1932: 182. Cushman 1944: 53. Townes et al. 1961: 293. Yu and Horstmann 1997: 752. Enicospilus (Enicospilus) longicornis Ashmead; Cushman 1944: 52. Remarks.—This medium-sized to large species (forewing length 10.1-19.5 mm) is easily recognized by the distinctly down- curved ovipositor, heavily setose, diagonal groove of the mandible (Fig. 36), and at least two prominent alar sclerites (Fig. 20). The gena (widely), face below the toruli, and often the clypeus are largely yellow with the possible exception of the medial area; the mesosoma varies from orange to red- dish-brown throughout (typically in the smaller individuals) to a patchwork of brownish-yellow and dark brown areas; the metasoma varies from orange or reddish-brown to dark brown, either sim- ilar throughout or with the petiole notice- ably darker; and the wings vary from slightly to distinctly infumate, often with a yellowish tint. As suspected by Cushman (1944), E. tyrannus represents the larger individuals among a continuum of varia- tion in size and color. Enicospilus melanochromus Perkins Fig. 21 Enicospilus melanochromus Perkins 1915: 523. Lectotype (designated by Townes et al. 1961: 281 [Their usage of “type’’ is herein regarded as equivalent to a lectotype desig- 25 nation (ICZN 1999: Art. 74.5).]) female, Maui, Haleakala, 2500 ft, I.02, R.C.L.P [Perkins] (BPBM) [examined]. Townes et al. 1961: 281. Gupta 1987: 554. Yu and Horstmann 1997: 744. Enicospilus (Enicospilus) melanochromus Perkins; Cushman 1944: 47. Remarks.—This medium-sized (fore wing length 9.1-10.5 mm) and rarely collected species is similar to E. blackburni but can be distinguished from the latter by its usual dark brown coloration (with possible ex- ception of a yellowish face, clypeus and gena; one individual examined is orangish- brown throughout as in E. blackburni), moderately infumate wings, fore wing with a basally expanded fenestra (Fig. 21) and 1m-cu without an angle medially (though often thickened), and the apically swollen aedeagus. Enicospilus minimus, new species ica Diagnosis.—Owing to its unusual habi- tus, the difficulty in diagnosing this species for the uninitiated will be in recognizing it as an ophionine, and not one of the many small, introduced ichneumonoids among the Hawaiian fauna. Its small size, yellow and brown coloration, reduced ocelli, and vestigial fenestra allow for its identification to species. Description.—Length of fore wing 3.7— 4.3 mm in female. Head: Mandible moder- ately stout, more or less parallel-sided medially, weakly twisted; basal, ventral margin strongly concave; outer surface without strong basal concavity, sparsely setose with hairs scattered or aggregated medially but not along a distinct diagonal groove; upper tooth long, 1.7-1.8X as long as lower tooth, about as wide as lower tooth at base. Labrum concealed in exam- ined specimens. Malar space 0.5X as long as basal mandibular width. Clypeus in profile weakly to strongly convex, proxi- mal margin distinct from lower face; in frontal view 2.3-2.4X as broad as long, 26 colliculate, apical margin broadly flat, sharp, impressed medially. Lower face as broad as long, evenly colliculate. Com- pound eye reduced, head width in frontal view 1.3-1.4X length (Fig. 7A). Gena with setae short, inconspicuous and declined forward; in dorsal view very broadly rounded behind compound eye (Fig. 7B), GO = 1.1-1.4. Ocelli small, posterior ocellus removed from compound eye by 0.8-0.9X its diameter, FI = 0.2-0.3. Occip- ital carina dorsally rounded, ventrally ending well short of hypostomal carina. Flagellum in female 1.5-1.6 length of fore wing, with 30-33 segments, mid segment 2.3-2.5xX as long as broad. Mesosoma: Mesoscutum strongly rounded anteriorly in profile, anterior angle about 90°; notauli not impressed (though marked by darker color). Scutellum compact, in dorsal view 1.1-1.2X as long as anterior width; upper surface strongly convex, evenly colliculate, lateral carinae present only anteriorly, extending about 1/10 scutellar length; posterior declivity smooth or weakly stri- ate, angled by about 45° in profile. Meso- pleuron evenly colliculate throughout, evenly rounded with little variation in relief; scrobe small, distinct or indistinct, not set in shallow groove, speculum not apparent; mesopleural sulcus with weak transverse marks; epicnemial carina strong, complete medioventrally. Mesosternum without lateral longitudinal depression; with posterior transverse carina present medioventrally. Lower metapleuron mod- erately convex, evenly colliculate. Propo- deum in profile weakly to moderately rounded anteriorly, flat medially and pos- teriorly; sparsely setose with setae low, inconspicuous, posteriorly declined; spira- cle small, oval; anterior furrow shallow, rugostriate, 0.1-0.2x total propodeal length; anterior transverse carina absent or present as a vestigial medial remnant, posterior transverse carina absent; spirac- ular area evenly colliculate, 0.2-0.3 total propodeal length; posterior area coarsely or evenly colliculate, becoming rugulose or JOURNAL OF HYMENOPTERA RESEARCH rugulostriate posteriorly and posterolater- ally. Separation between propodeum and lower metapleuron indicated by a com- plete furrow, not accompanied by a carina. Fore wing (Fig. 7C) with pterostigma short, wide, abruptly narrowed; discosubmargi- nal cell without sclerites, fenestra fairly small, round (ill-defined in one specimen examined), apical margin extending be- yond midpoint of Rs+2r, posterior margin extending to about midway between Rs+2r and 1im-cu or to nearer 1Im-cu; Rs+2r straight or slightly arched, thickened in basal half 1/2 to 3/4; Rs+M slightly arched in basal half; 1m-cu evenly arched; 3r-m absent or reduced to such extent that Al is about 3.2; CI = 0.4-0.6; ICI = 0-0.4; SDI = 0.8-0.9; cu-a anterior of Rs+M by 0.5-0.7 length of cu-a; 1st subdiscal cell with a setae sparse and even in ventral half, lacking in anterior half. Hind wing (Fig. 7E) with 1° abscissa of Rs straight, 2"* abscissa entirely nebulous, nearly straight, continuing to wing margin; 2"* abscissa of Cul emerging much nearer 1A than M, CI = 0.1-0.2; 1A absent distal of cu-a. Fore leg tibia 7.8-8.2 x as long as wide, without subapical spines on outer surface. Mid leg with coxa evenly colliculate; inner tibial spur about 1.3 as long as outer spur. Hind leg with coxa in lateral view 1.7X as long as deep, evenly colliculate; trochantellus dorsally about 0.4x as long as broad; 4" tarsomere in female 2.2-2.5x as long as broad in dorsal view; 5" tarsomere of female in dorsal view evenly broadened distally, 4.5-6.0X as long as broad, in lateral view nearly straight; pretarsal claw of female approximately as figured (Fig. 7D). Metasoma: Fairly com- pact and apically deep in female; T2 in female 2.4-3.3X as long as lateral height, 2.5-3.0 as long as dorsal width; thyridium tear-shaped to elliptical, positioned posteri- or of anterior margin of T2 by 0.2 length of T2. Ovipositor short and straight, about 0.8X length of T2. Color: Head yellowish-brown, darker near ocelli, on frons, and gena; mesosoma trunk and legs brown and yellow; wings hyaline. VOLUME 17, NUMBER 1, 2008 Fig. 7. wing D, female hind outer claw; E, hind wing. Remarks.—The minute size of E. minimus is unique among Enicospilus and probably Ophioninae in general. Its reduction has resulted in the extreme contraction of vein 3r-m of the fore wing, the presence of which is a synapomorphy for the subfam- ily (Gauld 1985). It is, however, recogniz- able as an Enicospilus owing to the slight twist of the mandible, vestigial fenestra, and the extension of the posterolateral area of the pronotum over the pronotal spirac- 27. Enicospilus minimus: A, frontal aspect of head; B, dorsal aspect of head; C, discosubmarginal cell of fore ular sclerite. It seems to be allied to the other Hawaiian Enicospilus species and particularly resembles E. petilus in head shape and color pattern. Furthermore, it possesses the general apomorphic features for Hawaiian Enicospilus (colliculate punc- tation and loss of fore-tibial spines). Material examined.—Holotype: female, Ha- waii, Hawaii Is.: Near Wauhaula Heiau, 7 December 2006, (W. C. Gagné) (BPBM). Para- 28 type: female, Oahu, Waimano Trail, 1900 ft elevation, 1 February 1970 (W. C. Gagné) (BPBM). Etymology.—The species epithet is Latin for ‘‘small’’, in reference to the minute size of this species as compared to all other Enicospilus species. Enicospilus molokaiensis (Ashmead), new combination Fig. 22 Pycnophion molokaiensis Ashmead 1900: 87. Lectotype (herein designated) female, Molo- kai, Mts, 4500 ft, 7[?] IX 1893, Perkins (AEIC) [examined]. Ashmead 1901: 344. Dalla Torre 1901: 185. Szépligeti 1905: 71. Anonymous 1925: 11. Cushman 1947: 461. Townes 1957: 117. Townes et al. 1961: 295. Townes 1971: 80. Gauld 1985: 168. Gupta 1987: 505. Yu and Horstmann 1997: 761. Remarks.—This species can be easily recognized by the combination of its stout form; moderately large size (fore wing 9.8— 11.8 mm); large compound eyes; long, upcurved ovipositor; weakly setose fore wing (at least proximally); discosubmargi- nal cell without a distinct fenestra or sclerites (Fig. 22); evenly colliculate propo- deum; and weak or absent medial part of the posterior carina of the mesosternum. It is largely black with possible exceptions of the face, gena, and clypeus which usually contain yellow or whitish areas; the fore leg, mid leg, and anterolateral and ventral areas of the mesosoma are usually red to orange in part, and the wings are distinctly infumate. This combination is not to be confused with its junior homonym Enicospilus molokaiensis Ashmead 1901 (= E. black- burni Bennett). Townes et al. (1961) and Gupta (1987) include a record from Hawaii Island as part of Ashmead’s syntype set, but I believe this is not correct. Ashmead (1901) did not list such a record in his publication, and I have seen no evidence indicating that this species occu. there. JOURNAL OF HYMENOPTERA RESEARCH Enicospilus niger (Ashmead), reinstated combination Figs 23, 24 Banchogastra nigra Ashmead 1900: 87. Holotype (by monotypy) female, Hawaii [Is.], Kilauea, IX.[18]95 (BMNH); transferred to Enicospilus by Townes 1945, and Townes et al. 1961. Ashmead 1901: 343. Dalla Torre 1901: 185. Szépligeti 1905: 71. Anonymous 1913: 203. Anonymous 1925: 11. Cushman 1947: 460. Townes 1971: 79. Gauld 1985: 169. Gupta 1987: 505. Yu and Horstmann 1997: 730. Banchogastra nigri [!] Ashmead; Perkins 1907b: OZ. Enicospilus niger (Ashmead); Townes 1945: 737. Townes 1957: 102. Townes et al. 1961: 283. Remarks.—This species is known from only a handful of specimens, and to my knowledge, it has not been collected since 1922. It can be recognized by the combina- tion of its moderate size (fore wing length about 11.5 mm); highly stout form; com- pact and apically bulbous petiole that is further described in the key (Fig. 34); T2 wider than long in dorsal view; reduced compound eye; discosubmarginal cell of fore wing densely setose throughout with- out a fenestra or sclerite; coarsely rugose, areolate, or rugostriate propodeum with a strong anterior transverse carina; mid coxa without strong ridges dorsomedially; short, straight ovipositor; and coloration (head and mesosoma black, metasoma black to deep reddish-brown throughout, fore wing dark brown anteriorly, lighter in posterior, apical area). Enicospilus nigrolineatus Ashmead Fig. 24 Enicospilus nigrolineatus Ashmead 1901: 348. Lectotype (designated by Townes et al. 1961: 284) male, Lanai, 2000 ft, 1.1894, Perkins (BMNH); Perkins 1915: 524. Townes et al. 1961: 284. Gupta 1987: 559. Yu and Horst- mann 1997: 746. Henicospilus nigrolineatus (Ashmead); Szépligeti {OORT 27: Enicospilus (Enicospilus) nigrolineatus Ashmead; Cushman 1944: 52. VOLUME 17, NUMBER 1, 2008 Remarks.—This large species (fore wing length 12.5-16.5 mm) can be easily recog- nized by its light brown to yellow colora- tion with the exception of a dark brown or black line laterally on the metasoma and the following black areas: scutum medially, mesosternum, and propodeum dorsome- dially; wings are more or less hyaline. The second alar sclerite is linear and lies along the posterior, apical margin of the fore wing fenestra (Fig. 24). Enicospilus variega- tus, which can be similar in this respect, is always more extensively covered in dark brown or black areas (see below). Enicospilus orbitalis (Ashmead) Fig. 25 Eremotylus orbitalis Ashmead 1901: 345. Lecto- type (designated by Townes et al. 1961: 285) female, Kauai, 2000-3000 ft, I.II.[18]97 (BMNH); transferred to Eremotyloides by Perkins 1915 [Perkins didn’t state a type species for his Eremotyloides but he apparent- ly intended it to be E. orbitalis Ashmead 1901 (see also discussion in Cushman 1947: 472).]. Szépligeti 1905: 36. Swezey and Bryan 1927: 412. Eremotyloides orbitalis (Ashmead); Perkins 1915: 532. Anonymous 1925: 11. Swezey and Williams 1932: 182. Enicospilus (Eremotyloides) orbitalis (Ashmead); Cushman 1944: 44. Enicospilus orbitalis (Ashmead); Townes 1945: 737. Cushman 1947: 472. Townes et al. 1961: 285. Gupta 1987: 562. Yu and Horstmann 1997: 746. Remarks.—This small to medium-sized species (fore wing length 6.5-11.5 mm) is relatively common in areas such as Kauai’s Alakai swamp. It can be easily recognized by the combination of an extremely slender metasoma (dorsomedial length of exposed portion of T5 in female, T4 in male, greater than lateral depth) and the forewing dis- cosubmarginal cell with a rather small, round fenestra lacking a sclerite (Fig. 25). Additionally, the ovipositor is short and slightly or distinctly upcurved; coloration is more or less evenly brown throughout, 29 with the possible exception of the face, clypeus, and gena, which are usually narrowly or broadly yellowish; and the wings are hyaline to slightly infumate. Enicospilus perkinsi Cushman Fig. 26 Enicospilus (Eremotyloides) perkinst Cushman 1944: 44. Holotype (by original designation) female, July 6, 1937, E. C. Zimmerman Oahu, at light (BPBM). Enicospilus perkinsi Cushman; Townes et al. 1961: 286. Gupta 1987: 564. Yu and Horst- mann 1997: 747. Remarks.—This rarely collected, medi- um-sized species (fore wing length 9.0- 11.0 mm) is easily recognized by the combination of an extremely slender me- tasoma (dorsomedial length of exposed portion of T5 in female, T4 in male, greater than lateral depth) and the distinctly orange head and metasoma which contrast with the darker metasoma (at least apical- ly). Additionally, the ovipositor is up- curved; the fore wing discosubmarginal cell contains a round fenestra and a distinct, oval sclerite (Fig. 26); and the wings are slightly infumate, with a yellow- ish tint. Enicospilus petilus, new species Fig. 8A—-D Diagnosis.—This species can be easily recognized by its greatly attenuated meta- soma and very large proximal alar sclerite (Fig. 8D). Description.—Length of fore wing 7.0- 8.3mm in female, 6.3=7.7 mm in male. Head: Mandible moderately stout, moder- ately twisted; basal ventral margin moder- ately to rather strongly concave; outer surface with a moderate basal concavity, sparsely to moderately setose along a weak diagonal groove; upper tooth 1.0-1.3x as long as lower tooth, about as wide or slighty narrower than lower tooth at base. Labrum 0.2-0.3X as long as broad, apical margin broadly rounded or flat medially. 30 Malar space 0.5-0.6* as long as _ basal mandibular width. Clypeus in profile nearly flat to weakly convex, proximal margin weakly to moderately distinct from lower face; in frontal view 1.6-2.0X as broad as long, lightly punctate and finely colliculate, apical margin broadly rounded or broadly flat, sharp, not impressed medially. Lower face 0.9-1.2 as broad as long. Compound eye reduced, head width in frontal view 1.2-1.3x length (Fig. 8A). Gena with setae inconspicuous, short, pale, and declined forward; in dorsal view broadly rounded behind compound eye (Fis. 8B), GOI — 15-23) Ocellt reduced: posterior ocellus separated from com- pound eye by 0.4—0.6x its diameter, FI = 0.3-0.4. Occipital carina dorsally flat or rounded, ventrally joining or ending short of hypostomal carina. Flagellum in female 1.3X length of fore wing, with 43-45 segments, mid segment 2.3-2.5x as long as broad; in male 1.3-1.5x length of fore wing, with 41-44 segments, mid segment 2.3-2.4X as long as broad. Mesosoma: Mesoscutum strongly rounded anteriorly in profile; notauli weak or not apparent. Scutellum short, rounded, in dorsal view 1.1-1.3X as long as anterior width; upper surface strongly convex, evenly and smoothly colliculate; lateral carinae more or less absent to weakly present through about 1/2 scutellar length; posterior de- clined by 30°-40° in profile. Mesopleuron evenly colliculate throughout, rather flat and evenly rounded; scrobe small but clearly apparent, set in very shallow depression; speculum weakly apparent to absent; mesopleural sulcus with weak transverse ridges; epicnemial carina strong, medioventrally complete or narrowly ab- sent. Mesosternum without lateral longitu- dinal depression behind epicnemial carina; with posterior transverse carina present medially. Lower metapleuron weakly to moderately convex, evenly colliculate to rugulose. Propodeum in profile weakly < anteriorly and flat posteriorly to moderately convex throughout; exceeding- con \ JOURNAL OF HYMENOPTERA RESEARCH ly sparsely setose (except one examined specimen moderate in this regard) with setae lying low and posteriorly declined; spiracle narrow; anterior furrow shallow, forming a broad concavity rather than a sharp groove, coarsely rugostriate, anterior area 0.1-0.2 total propodeal length; ante- rior transverse carina present or absent, posterior transverse carina absent or pre- sent; spiracular area finely colliculate, about 0.3 total propodeal length; posteri- or area anteriorly rugulose becoming ru- gostriate posteriorly. Separation between propodeum and lower metapleuron indi- cated by a weak furrow and weak, irreg- ular carina. Fore wing (Fig. 8D) with pterostigma short and triangular, distal end narrowed abruptly; discosubmarginal cell with 2 sclerites, the basal one very large, semicircular or roughly triangular, the distal one linear, partially outlining distal ventral margin of fenestra; fenestra semicircular, extending apically to at least midpoint of Rs+2r, posterior margin ex- tending to about 2/3 the distance between Rs+2r and 1m-cu; Rs+2r thickened medial- ly and sinuous; Rs+M nearly straight or slightly arched in basal half; Im-cu evenly arched; Al =°1.5—2.1; Cl = 03-502 te 0.2-0.3; SDI = 0.9-1.3; cu-a positioned directly opposite base of Rs+M; 1st sub- discal cell sparsely setose throughout or only in posterior part. Hind wing with 4-6 hamuli in distal set; 1°‘ abscissa of Rs nearly straight or slightly concave basally, 2"° abscissa straight; 2" abscissa of Cul emerging much nearer 1A than M, CI = 0.2. Fore leg tibia 7.7-8.7X as long as wide, without an array of subapical spines on outer surface. Mid leg with coxa evenly colliculate; with inner tibial spur 1.3-1.5x as long as outer spur. Hind leg with coxa in lateral view 1.6-1.7X as long as deep; trochantellus dorsally 0.3-0.5x as long as broad; 4 tarsomere in female 2.0-2.1 as long as broad in dorsal view, about 2.2 in male; 5" tarsomere of female in dorsal view evenly broadened distally, 2.8-3.1 as long as broad, in lateral view weakly curved; 5™ VOLUME 17, NUMBER 1, 2008 Big. 8. discosubmarginal cell of fore wing. tarsomere of male in dorsal view some- what abruptly widened apically, about 3.5X as long as broad, in lateral view weakly curved; pretarsal claw approxi- mately as in Fig. 8C, apparently with little or no sexual dimorphism. Metasoma: Very elongate (especially in female), narrowed and laterally flattened apically in female; T2 in female 4.3-5.3x as long as lateral height, 3.0-3.5 x as long as dorsal width; T2 in male about 4.5x as long as lateral height, 2.4-4.9 as long as dorsal width; ol Enicospilus petilus: A, frontal aspect of head; B, dorsal aspect of head; C, female hind outer claw; D, thyridium tear-shaped, positioned posteri- or of anterior margin of T2 by 0.3-0.4x length of T2. Ovipositor short and straight or slightly upcurved. Color: Generally yellow and brown; head yellowish-brown, darker on dorsal gena and antenna; mesosoma brown with yellow patches on mesoscutum, medially on scutellum, and variously on meso- pleuron and propodeum; wings hyaline; legs yellow to yellowish-brown with apical tarsomeres, hind coxa and femur (at least 32 in part) darker; metasoma with basal half or more of petiole pale yellow or yellow- ish-brown, otherwise brown except for various ill-defined lighter intersegmental areas which are lighter. Material examined.—Holotype: female, Ha- waii, Maui: Haleakala National Park, upper Kipahulu Valley, ‘‘Charlie Camp’; 1450 m elevation, 28 February-4 March 1984, UV light trap in forest (W. C. Gagné, S. Gon III) (BPBM). Paratypes (4): 1 female, Hawaii: Kilauea, “29 mi,” August 1912 (W. M. Giffard) (AEIC); 1 female, Molokai: Pepeopae, 4000 ft elevation, 30 July 1959 (D. E. Hardy) (BPBM); 1 male, Molokai: West end of Hanalilolilo Trail, 1070 m, 7 January 1981, M. V. light, (W. C. Gagné) (BPBM); 1 male, Hawaii: Manuka Forest Reserve, South Kona, Kopua T. [Trail ?], 3600 ft elevation, 22 June 1977, night, (R. S. Villegas and S. M. Gon III) (BPBM). Etymology.—The species epithet, a Latin adjective for ‘’slender,’”’ is in reference to the extremely elongate metasoma of the female. Enicospilus pseudonymus Perkins Fig. 27 Enicospilus pseudonymus Perkins 1915: 529. Lec- totype (designated by Townes et al. 1961: 286 [Their usage of “type” is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) male, Maui, Haleakala, 4000 ft (BPBM) [examined]. Anonymous 1925: 10. Cushman 1944: 53. Townes et al. 1961: 286. Gupta 1987: 568. Yu and Horst- mann 1997: 748. Remarks.—This rarely collected, fairly large species (fore wing length 12.0- 13.2 mm) can be recognized by the unique lateral, longitudinal depressions of the posterior mesonotum and scutellum. Ad- ditionally, the upper mandibular tooth is shorter than the lower tooth; the fore wing discosubmarginal cell lacks a sclerite and contains a fenestra which is at most a rather narrow, poorly defined region of reduced pubescence (Fig. 27); and the posterior transverse carina of meso- sternum is absent or weak medially JOURNAL OF HYMENOPTERA RESEARCH (often weak medially and absent sub- medially). It is more or less orangish- brown with exception of the face, clypeus, and gena, which are largely yellow, as well as parts of the mesonotum, mesopleuron, propodeum, and petiole, which are often slightly or distinctly darker; the wings are hyaline. Enicospilus swezeyi, new name Fig. 1 Pycnophion fuscipennis Perkins 1910: 680. Lecto- type (designated by Townes et al. 1961: 295 [Their usage of “type” is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) female, Kauai, 3000 ft, winter 1901 (BPBM) [examined]; preoccupied in Enicospilus by E. fuscipennis (Szépligeti 1906). Anonymous 1925: 11. Swezey 1931: 502. Townes et al. 1961: 295. Gupta 1987: 505. Yu and Horstmann 1997: 761. Pycnophion fumipennis [!] Perkins; Cushman 1947: 462. Remarks.—This distinctive, small to me- dium-sized species (fore wing length 8.2- 10.7 mm) can be recognized by its highly contrasting red and black coloration (Fig. 1); a long, straight ovipositor; propo- deum posteriorly rugose, without an ante- rior transverse carina, with posteriorly projecting setae dorsomedially; lack of both a sclerite and a clearly defined fenestra in the fore wing discosubmarginal cell; distinctly infumate wings; and a weak or absent posterior transverse carina of the mesosternum. Etymology.—The new name is dedicated to the prolific Hawaiian entomologist, Otto Swezey, whose efforts in the rearing of Hawaiian insects over many years (Swezey 1954) resulted in many discoveries includ- ing the host for this species, as elaborated on above. Enicospilus variegatus Ashmead Fig. 28 Enicospilus variegatus Ashmead 1901: 348. Lec- totype (designated by Townes et al. 1961: 293) male, Hawaii [Is.], Ola’a, 11.1896 (BMNH). VOLUME 17, NUMBER 1, 2008 - & = “ Tt 9. Enicospilus bellator 11.Enicospilus castaneus Figs 9-11. Fore wing discosubmarginal cells. 33 JOURNAL OF HYMENOPTERA RESEARCH 12. Enicospilus debilis ~ 13. Enicospilus dispilus 14. Enicospilus fullawayi Figs 12-14. Fore wing discosubmarginal cells. VOLUME 17, NUMBER 1, 2008 eee ALES ae . ws Z - . = = on i$ e « < ~ o . a ~ . . ‘ . . ‘ i Ae ‘ ~- “Sp “ - ‘ 2 \ < . ‘ . — 15. Enicospilus hawailensis Le 17. Enicospilus kauaiensis Figs 15-17. Fore wing discosubmarginal cells. a2 — 18. Enicospilus lineatus 20. Enicospilus longicornis 18-20. Fore wing discosubmarginal cells. JOURNAL OF HYMENOPTERA RESEARCH oF VOLUME 17, NUMBER 1, 2008 l Y 5 = © <& S 9 = & o = w = a 12) 2 S Uy N “ ; Kj ti 4 Ce q “as t Ly a, \/ «4 ¢ 4 . \ i 4 yd Felis < Gary } \ 9 ‘ gpk “| iensis 22. Enicospilus moloka 23.Enicospilus niger Fore wing discosubmarginal cells. Figs 21-23. YJ (oe ) 25.Enicospilus orbitalis 26. Enicospilus perkinsi Figs 24-26. Fore wing discosubmarginal cells. JOURNAL OF HYMENOPTERA RESEARCH VOLUME 17, NUMBER 1, 2008 27.Enicospilus pseudonymus \ 28. Enicospilus variegatus 29. Enicospilus vitreipennis Figs 27-29. Fore wing discosubmarginal cells. 2 | 39 40 JOURNAL OF HYMENOPTERA RESEARCH oe Fae SLE EE Ps] a hi a ee —_ P, A Cee va ,, a A « —," -“ 3 AT LEA ee Sr Oe ae PO AP Le LT oe a ye =e —- is SP Gene : - -_ eo Pn gee “ fore or —= a en C7, 3 SAT pip Poel — - owe ~— " — — -_ : . - - > ‘ > » > . ~ —Z ' . .* . = s ~ “A = = Z ~ _ - el - a - = . > a? - , ~ ee ee ~ en £4. — —— > ~ 4 7 ‘ 4 ‘ pa a — “« ~% , = map eB a beni ~- . - a + . — . . « ~ ; a . os oe. . ; _ es —T —_ . = ~ ‘ . - : * = ° . . whi .* ~\. ° ° " . 30. Enicospilus waimeae Fig. 30. Fore wing discosubmarginal cell. Figs 31-33. metasomal apices, arrows indicate S7. Fig. 34. Enicospilus niger petiole in lateral view, arrow indicates ventral posterior margin, brackets indicate dorsal and ventral lengths. Fig. 35. Enicospilus vitreipennis petiole in lateral view. Fig. 36. Enicospilus longicornis mandible. Fig. 37. Enicospilus lineatus mandible. Perkins 1915: 525. Townes et al. 1961: 293. Enicospilus (Enicospilus) variegatus Ashmead; — Gupta 1987: 583. Yu and Horstmann 1997: Cushman 1944: 51. _ Henicospilus variegatus (Ashmead); Szépligeti Remarks.—This large species (fore wing 905: 27 length 14.5-16.0 mm) can be recognized by VOLUME 17, NUMBER 1, 2008 its highly contrasting color pattern. It is largely dark brown to black except for the following yellow or yellowish-brown areas: head, pronotum (in total or in part), mesonotum and scutellum in part, upper mesopleuron, metapleuron, anterodorsal part of propodeum, and legs (except for the femora apically, and with the possible exception of the apical tarsomeres). Addi- tionally, the fore wing discosubmarginal cell contains a large, triangular, proximal sclerite and often a second, vestigial scler- ite at the apical, posterior fenestral margin (Fig. 28); the wings are more or less hyaline. Enicospilus vitreipennis (Perkins), reinstated combination Figs 29, 35 Banchogastra vitreipennis Perkins 1910: 680. Lec- totype (designated by Townes et al. 1961: 293 [Their usage of “‘type’’ is herein regarded as equivalent to a lectotype designation (ICZN 1999: Art. 74.5).]) female, Maui, Ha- leakala, 5000 ft (BPBM) [examined]; trans- ferred to Enicospilus by Townes et-al. 1961. Gupta 1987: 506. Yu and Horstmann 1997: Zou: Enicospilus vitretpennis (Perkins); Townes et al. 1961, 293. Remarks.—This small species (fore wing length 7.5-9.5 mm) can be recognized by the combination of its coloration (darkest brown to black throughout with the possi- ble exception of the wings which vary from nearly hyaline to dark brown); short ovipositor; highly reduced compound eye; discosubmarginal cell usually without a trace of a fenestra and densely setose throughout; propodeum coarsely rugose, areolate, or rugostriate with a strong anterior transverse carina; mid coxa with- out strong dorsomedial ridges; posterior transverse carina of the mesosternum absent; T2 usually longer than wide in dorsal view; and, with respect to E. niger, a less compact, flatter petiole that is further described in the key (Fig. 35). 4] Enicospilus waimeae Ashmead Fig. 30 Enicospilus waimeae Ashmead 1901: 348. Lecto- type (designated by Townes et al. 1961: 293) female, Kauai, Mts Waimea, 4000 ft, VI.1894, Perkins (BMNH); Perkins 1915: 525. Townes et al. 1961: 293. Gupta 1987: 586. Henicospilus waimae ['!] (Ashmead); Szépligeti 1905: 27: Enicospilus (Enicospilus) waimeae Ashmead; Cushman 1944: 51. Remarks.—This rarely collected, moder- ately large species (fore wing length 11.0- 13.4 mm) is principally recognized by its single, extremely large sclerite of the fore wing discosubmarginal cell (Fig. 30). Ad- ditionally, it is generally slenderer than E. lineatus (to which it otherwise most closely resembles); is more or less brown through- out (becoming lighter brown on the face laterally, gena, tibiae, tarsi, and metasoma apically); has a short, straight ovipositor; and has slightly infumate wings. ACKNOWLEDGMENTS I am grateful to a number of persons who have assisted me in various ways with this study. Michael Engel and Charles Michener read an early version of this paper and made helpful suggestions. The late James S. Ashe provided valuable guidance early on. Gavin Broad and an anonymous reviewer provided helpful comments. Personnel at several institutions assisted with specimen loans and information re- quests including Ian Gauld, Gavin Broad, Stuart Hine, David Wahl, David Furth, Robert Carlson, Robert Kula, Brian Harris, Frank Howarth, Tino Gonsalves, Shepherd Myers, Robert Zuparko, John Huber, James Liebherr, Daniel Rubinoff, and Will Haines. Betsy Gagné and Wayne Souza assisted in acquiring collecting permits. Support was provided by National Science Foundation grant EF-0341724 (M. S. Engel P. I.) and NSF/Kansas EPSCoR grant KAN29503 (M. S. Engever tt): LITERATURE CITED Alfken, J. D. 1904. Beitrag zur insectenfauna der Hawaiischen und Neuseelandischen Inseln. (Er- gebnisse einer Reise nach dem Pacific.) Schauins- land 1896-97. Zoologische Jahrbiicher 19: 561-628. Anonymous. 1913. Notes and exhibitions. Proceedings of the Hawaiian Entomological Society 2: 202-203. 42 Anonymous. 1917. Notes and exhibitions. Proceedings of the Hawaiian Entomological Society 3: 202-286. Anonymous. 1924. Notes and exhibitions. Proceedings of the Hawaiian Entomological Society 5: 345. Anonymous. 1925. Notes and exhibitions. Proceedings of the Hawaiian Entomological Society 6: 10-11. Anonymous. 1955. Notes and exhibitions. Proceedings of the Hawaiian Entomological Society 15: 385-387. Ashmead, W. H. 1900. Classification of the ichneu- mon-flies, or the superfamily Ichneumonoidea. Proceedings of the United States National Museum 23: 1-220. Ashmead, W. H. 1901. Hymenoptera Parasitica. Pp. 277-364. in: D. Sharp ed. Fauna Hawatiensis, [Vol. 1, Part 3]. Cambridge University Press, Cambridge, UK. Bennett, D. J. 2004. A cladistic analysis of Hawatian ophionine wasps (Hymenoptera: Ichneumonidae). Master’s thesis, University of Kansas, Lawrence; v + 45 pp. Blackburn, J. and P. Cameron. 1886. On the Hyme- noptera of the Hawaiian Islands. Proceedings of the Manchester Literary and Philosophical Society 25: 134-183. Blackburn, J. and P. Cameron. 1887. On the Hyme- noptera of the Hawaiian Islands. Memoirs of the Manchester Literary and Philosophical Society 3 (10) [= third series, tenth volume]: 194-244. Cameron, P. 1883. Descriptions of new genera and species of Hymenoptera. Transactions of the Entomological Society of London 1883 (2): 187-197. Chu, J. T. 1935. Preliminary notes on the Ichneumon- flies in Kiangsu and Chekiang Provinces, China. Yearbook of the Bureau of Entomology of Chekiang Province 1934: 7-32. Cushman, R. A. 1944. The Hawaiian species of Enicospilus and Abanchogastra (Hymenoptera: Ich- neumonidae). Proceedings of the Hawatian Entomo- logical Society 9 (12): 39-56. Cushman, R. A. 1947. A generic revision of the ichneumon-flies of the tribe Ophionini. Proceed- ings of the United States National Museum 96: 417-482. Dalla Torre, C. G. 1901. Catalogus Hymenopterorum. Volumen III: Trigonalidae, Megalyridae, Stepha- nidae, Ichneumonidae, Agriotypidae, Evaniidae, Pelecinidae. Guilelmi Engelmann, Lipsiae. [1901: 1-544; 1902: 545-1141]. Fullaway, D. T. 1957. Checklist of the Hymenoptera of Fiji. Proceedings of the Hawaiian Entomological Society 16: 269-290. Fullaway, D. T. and W. M. Giffard. 1919. Notes on Collection of Hawaiian Insects on Island of Maui. Proceedings of the Hawatian Entomological Society 4: 50-52. Gauld, |. D. 1985. The phylogeny, classification and evolution of parasitic wasps of the subfamily Ophioninae (Ichneumonidae). Bulletin of the Brit- JOURNAL OF HYMENOPTERA RESEARCH ish Museum of Natural History (Entomology) 51: 61-185. Gauld, I. D. 1988. A survey of the Ophioninae (Hymenoptera: Ichneumonidae) of tropical Me- soamerica with special reference to the fauna of Costa Rica. Bulletin of the British Museum of Natural History (Entomology) 57: 1-309. Gauld, I. D. and P. A. Mitchell. 1981. The taxonomy, distribution and host preferences of Indo-Papuan parasitic wasps of the subfamily Ophioninae. Com- monwealth Agricultural Bureaux, Slough, UK; 611 pp. Gupta, V. K. 1987. The Ichneumonidae of the Indo- Australian area (Hymenoptera). Memoirs of the American Entomological Institute 41: 1-597. Harris, R. A. 1979. A glossary of surface sculpturing. Occasional Papers in Entomology. State of Califor- nia, Department of Food and Agriculture, Divi- sion of Plant Industry-Laboratory Services 28: 1-31. Iwata, K. 1950. Biology of Ichneumon parasites on bag-worms in Japan, I. Transactions of the Kansai Entomological Society. Osaka 15: 35-47. International Commission on Zoological Nomencla- ture. 1999. International Code of Zoological Nomen- clature, 4 ed. International Trust for Zoological Nomenclature, London, UK; xxix + 306 pp. Lee, J. W. and C. W. Kim. 1980. A taxonomical study on the Korean Ophioninae (Hym., Ichneumoni- dae). Korean Journal of Entomology 10: 9-18. Morley, C. 1912. A revision of the Ichneumonidae 1. London, UK; vi + 88 pp. Perkins, R. C. L. 1902. Four new species and a new genus of parasitic Hymenoptera (Ichneumonidae: Ophioninae) from the Hawaiian Islands. Transac- tions of the Entomological Society of London 1902: 141-143. Perkins, R. C. L. 1907a. The Insects of Tantalus. Proceedings of the Hawaiian Entomological Society 1: 38-51. Perkins, R. C. L. 1907b. Insects at Kilauea, Hawaii. Proceedings of the Hawaiian Entomological Society 1: 89-99. Perkins, R. C. L. 1910. Supplement to Hymenoptera. Pp. 600-686. in: D. Sharp ed. Fauna Hawatiensis [Vol. 2]. Cambridge University Press, Cambridge, UK. Perkins, R. C. L. 1913. Introduction. Pp. xv—cexxvii. in: D. Sharp ed. Fauna Hawatiensis [Vol. 1, part 6]. Cam- bridge University Press, Cambridge, UK. Perkins, R. C. L. 1915. On Hawaiian Ophioninae (Hymenoptera, Fam. Ichneumonidae). Transac- tions of the Entomological Society of London 1914: 521-555, Ruthe, J. F. 1859. Verzeichnis der von Dr. Staudin- ger im Jahre 1856 auf Island gesammelten Hymenopteren. Stettiner Entomologische Zeitung 20: 362-379. VOLUME 17, NUMBER 1, 2008 Swezey, O. H. 1915. A preliminary list of the hymenopterous parasites of Lepidoptera in Ha- waii. Proceedings of the Hawaiian Entomological Society 3: 99-109. Swezey, O. H. 1931. Some observations on the insect faunas of native forest trees in the Olinda forest on Maui. Proceedings of the Hawaiian Entomological Society 7: 493-504. Swezey, O. H. 1954. Forest entomology in Hawaii. An annotated check-list of the insect faunas of the various components of the Hawaiian forests. Special Publication of the Bernice P. Bishop Museum 44: 1-266. Swezey, O. H. and E. H. Bryan. 1927. Notes on some forest insects of Molokai. Proceedings of the Hawatian Entomological Society 6: 411-422. Swezey, O. H. and F. X. Williams. 1932. Some observations on forest insects at the Nauhi Nursery and vicinity in Hawaii. Proceedings of the Hawatian Entomological Society 8: 179-190. Szépligeti, G. V. 1905. Hymenoptera Ichneumonidae. In: Wytsman, P. ed. Genera Insectorum 34: 1-68. Bruxelles. Szépligeti, G. V. 1906. Neue exotische Ichneumoniden aus der Sammlung des Ungarischen national Museums. Annales Musei Nationalis Hungarici 4: 119-156. Townes, H. 1945. A catalogue and reclassification of the Nearctic Ichneumonidae (Hymenoptera). Part II. The subfamilies Mesoleiinae, Plectiscinae, 43 Orthocentrinae, Diplazontinae, Metopiinae, Ophioninae, Mesochorinae. Memoirs of the Amer- ican Entomological Society 11: 478-925. Townes, H. 1957. A review of the generic names proposed for old world ichneumonids, the types of whose genotypes are in Japan, Formosa, or North America. Proceedings of the Entomological Society of Washington 59: 100-120. Townes, H. 1969. Genera of Ichneumonidae 1. Memoirs of the American Entomological Institute 11: 1-300. Townes, H. 1971. Genera of Ichneumonidae 4. Memoirs of the American Entomological Institute 17: 1-372. Townes, H. and M. Townes. 1973. Ethiopian Ichneu- monidae. Memoirs of the American Entomological Institute 19: 1-416. Townes, H., M. Townes, and V. K. Gupta. 1961. A catalogue and reclassification of the Indo-Austra- lian Ichneumonidae. Memoirs of the American Entomological Institute 1: 1-522. Uchida, T. 1928. Zweiter Beitrage zur Ichneumoniden- Fauna Japans. Journal of the Faculty of Agriculture of Hokkaido Imperial University 21: 177-297. Yu, D. S. and K. Horstmann. 1997. A Catalogue of World Ichneumonidae (Hymenoptera). Part 1. Subfamilies Acaenitinae to Ophioninae. Mem- oirs of the American Entomological Institute 58: if Jos: J. HYM. RES. Vol. 17(1), 2008, pp. 44-56 The Description of Euceroptrinae, a New Subfamily of Figitidae (Hymenoptera), including a Revision of Euceroptres Ashmead, 1896 and the Description of a New Species MATTHEW L. BUFFINGTON* AND JOHAN LILJEBLAD (MLB) Systematic Entomology Lab, USDA, c/o USNM, Smithsonian Institution, 10th & Constitution Ave NW, PO Box 37012 MRC-168, Washington DC 20013, 202-382-1784, USA; email: matt.buffington@ars.usda.gov (JL) Department of Entomology, Texas A&M University, current address: Kopparv. 57 B, SE-791 41 Falun, Sweden +46236614644; email: cynips@gmail.com Abstract.—The figitid genus Euceroptres Ashmead has recently been determined to render the Thrasorinae, a subfamily where the genus is currently classified, paraphyletic. To maintain the monophyly of Thrasorinae, Euceroptres is here redescribed and placed in its own subfamily, Euceroptrinae. All known species are redescribed and a new species is described; a lectotype is designated for E. primus Ashmead, 1896. The phylogenetics of Euceroptrinae, Parnipinae, Plectocynipinae and Thrasorinae are discussed, and the hypothesis that ancestral lineages of figitids attacked gall-inducing Hymenoptera is supported. Agastoparasitism among these lineages appears to be plesiomorphic. Though branch support is relatively low for inferring the precise branching order of the gall-inducer parasite lineages, the classification of problematic species is much improved. Resolving the early branching events separating the phytophagous Cynipidae from the entomophagous Figitidae (Hyme- noptera: Cynipoidea) continues to chal- lenge hymenopterists. Although cynipoids resolve phylogenetically within the ento- mophagous parasitoid Hymenoptera (Ron- quist et al. 1999, Dowton and Austin 2001), the majority of Cynipidae are obligate phytophages, inducing spectacular galls on host plants in 17 families of angio- sperms (Weld 1952, Liljeblad and Ronquist 1998, Ronquist 1999). Figitids associated with the gall community are important for understanding the early evolution of these lineages; Ronquist (1995a, 1999) and Buf- rington et al. (2007) discussed two ancestral groups of cynipoids, i.e., Parnipinae and Thrasorinae, that represent lineages whose biology lie somewhere between entomo- phagy and phytophagy. Further, these * Author for c espondence lineages also represent figitid agasotopar- asites (parasitoids whose primary hosts are themselves close relatives (Ronquist 1994)), a life-history strategy rare among Figitidae. Hence, understanding the taxonomy, biol- ogy, and phylogenetics of these groups will elucidate the evolutionary origins of the phytophagous cynipid lineage. The Thrasorinae have been the subject of a few recent studies attempting to clarify our accumulated knowledge on this group. Ronquist (1994) grouped several cynipoid genera together in what he called the ‘figitoid inquilines’; later, in Ronquist (1999), these genera (Euceroptres Ashmead, Thrasorus Weld, Myrtopsen Dettmer, Pega- cynips Bréthes and Plectocynips Diaz) were placed within Thrasorinae, a group that previously only contained Thrasorus (Ko- valev 1994). Ros-Farré and Pujade-Villar (2007) removed Pegacynips and Plectocynips from Thrasorinae and placed them into the newly described Plectocynipinae, and de- VOLUME 17, NUMBER 1, 2008 scribed a new thrasorine genus, Scutimica Ros-Farré. ‘Buffington (submitted) revises the Australian Thrasorinae and describes a new genus of thrasorines associated with galls on Eucalyptus spp.’ Ronquist (1999) suggested the placement of Euceroptres within Thrasorinae as tenta- tive at best, given that the taxon lacks a number of synapomorphies the remaining taxa possess. In both Ros-Farré & Pujade- Villar (2007) and Buffington (submitted), Euceroptres Ashmead was determined not to be a thrasorine, leading Buffington (submitted) to render the taxon incertae sedis. Buffington et al. (2007) found weak support for Euceroptres to be included within Thrasorinae. Further, Euceroptres rendered Thrasorinae paraphyletic if mor- phological data were excluded. Based on the total evidence phylogeny of Buffington et al. (2007), if Plectocynipinae is recog- nized, Thrasorinae is rendered paraphy- letic unless Euceroptres is excluded. MATERIALS AND METHODS Rearing methods.—Fully developed galls were collected by us from Quercus agrifolia NZe (Fagaceae) in Eaton Canyon State Park, Pasadena, CA. All leaves were removed and bare galls were placed together with tissue paper in plastic zipper bags to collect emerging wasps. Non- reared material for examination was bor- rowed from institutions listed below. Descriptions.—Morphological terminolo- gy follows that of Ronquist and Nordlan- der (1989), Fontal-Cazalla et al. (2002) and Buffington et al. (2007); cuticular surface terminology follows that of Harris (1979). Specimens were examined using a Leica Wild M10 with fluorescent lighting. Images for figures were obtained using an EntoVi- sion Imaging Suite, which included a firewire JVC KY-75 3CCD digital camera mounted to a Leica M16 zoom lens via a Leica z-step microscope stand. This system fed image data to a desktop computer Where Cartograph 5.6.0 (Microvision In- struments, France) was used to capture a 45 fixed number of focal planes (based on magnification); the resulting focal planes were merged into a single, in-focus com- posite image. Lighting was achieved using an LED illumination dome with all four quadrants set to 99.6% intensity. Scanning electron micrographs of Euceroptres monta- nus Weld were made by the second author and were downloaded for this study from Morphbank (http://morphbank.net). All images generated during this study are available under the Morphbank ID 195606. Phylogenetic analysis — matrix.—The ma- trix used in Buffington et al. (2007) was complemented with the addition of [balia anceps Say (Ibaliidae). This taxon was in- cluded to serve as an additional outgroup to the liopterids (see Ronquist 1999). Sequence data for the ibaliid was downloaded from GenBank (DQ012642, DQ012641, DQ012599, AY621150, EF032242, EF0O32274). Precisely the same gene regions were used as they were in Buffington et al. (2007); alignments used herein were based entirely on the structural model (Gillespie 2004, Gillespie et al. 2005) proposed for Cynipoidea (Buf- fington et al. 2007) and included all genetic data; regions of ambiguous alignment were aligned by eye. The model as proposed in Buffington et al. (2007) was not altered by the inclusion of the ibaliid. Morphological and biological characters described in Buf- fington et al. (2007) were included and used to code Ibalia anceps. The final molecular and morphological matrix is available from Treebase (ID SN3726). It should also be noted that Buffington et al. (2007) included a taxon identified as ‘Myrtopsen sp.’ from Colombia; this taxon is, in fact, Scutimica flava Ros-Farré & Pujade-Villar. Phylogenetic analysis — parsimony and Bayesian inference.-—The _ structurally aligned total evidence matrix was analyzed based on Buffington et al. (2007); the differences are the addition of [balia anceps in all analyses, and the Bayesian analyses were run for 5 million generations, sam- pling every 100 generations and burn-in set to 350 (2.5 million generations and burn-in 46 set to 250 generations in Buffington et al. (2007)). List of Depositories. AMNH American Museum of Natural History, New York, NY, USA. USNM_ National Museum of Natural History, Washington DC, USA. UCRM_ Entomology Research Museum, UC Riverside, Riverside, CA, USA. DESCRIPTIONS Euceroptrinae Buffington and Liljeblad, new subfamily Type genus: Euceroptres Ashmead, 1896 Diagnosis.—The areolet in the fore wing (ARE, Fig. 1D), lack of a hairy ring at the base of the metasoma, lack of a circumtor- ular impression and well-developed lateral pronotal carina (LPC, Fig. 1A) differentiate this group from Thrasorinae and Plectocy- nipinae. Nearly all other species of Figiti- dae have a smooth mesoscutum (save for notauli) whereas the mesoscutum in Eu- ceroptrinae is transversly carinate to rugu- lose; the only other figitid groups with a transversly carinate mesoscutum are some Aspicerinae (e.g. Anacharoides Cameron, Callaspidia Dalhbom, Omalaspis Giraud and Pujadella Ros-Farré) but these species have a sinuate posterior margin of tergum 2 of the metasoma, and are parasites of Syrphidae (Diptera). Parnipinae bears the closest resemblance to Euceroptrinae, but there are several key differences, including the lack of a mesopleural furrow in Parnipinae (complete in Euceroptrinae (F, Fig. 1B)). Parnipinae are Palearctic parasit- oids of Barbotinia (Cynipidae: Aylacini) on Papaver (Papaveraceae) and Euceroptrinae are Nearctic parasitoids of Cynipini (An- dricus gall inducers) on Quercus spp. (Fagaceae) (Summarized in Table 1). Description.—Body color black to pale orange; legs orange proximally, darker distally. Female with 12-14 antennal seg- ments; male with 15 segments, first flagel- iomere laterally excavated. Lateral prono- JOURNAL OF HYMENOPTERA RESEARCH tal carina (LPC, Fig. 1A) well developed. Mesoscutum ranging from transversly car- inate to rugulose; notauli present, well developed; median mesoscutal impression (MMI, Fig. 1C) present, extending up to % length of mesoscutum; scutellum rugulose, posteriorly rounded (Figs 2A—C). Fore wings hyaline, areolet present (ARE, Fig. 1D). Anterior margin of tergum 3 (T3) of meta- soma glabrous; T4—T7 with micropores (MP, Fig. 1F) (reduced in some species). Euceroptres Ashmead, 1896: Trans. Am. Ento- mol. Soc, v..23,,p 187 Type species: Euceroptres primus Ashmead, 1896 (by monotypy). Included species: Euceroptres primus Ashmead. Euceroptres maritimus Weld, 1926. Euceroptres montanus Weld, 1926. Euceroptres whartoni Buffington & Liljeblad, new species. Diagnosis.—Differs from nearly all other Figitidae by the presence of the areolet in the fore wing (Fig. 1D); the only other group of Figitidae with an areolet is Parnips (Parnipinae), but this group lacks a meso- pleural furrow. The presence of a well- developed lateral pronotal carina is a plesiomorphic trait within Cynipoidea (Ronquist 1995b, 1999 Ronquist and Nieves-Aldrey 2001, Fontal-Cazalla et al. 2002, Buffington et al. 2007, Liu et al. 2007), and this trait is useful for separating Euceroptres from other figitids; among the figitids with this character is Parnips (Ronquist and Nieves-Aldrey 2001), all Aspicerinae except Melanips (Buffington et al. 2007) and some members of the Grono- toma group of Eucoilinae (Fontal-Cazalla et al. 2002, Buffington et al. 2007). Aside from Parnips, none of these aforementioned taxa are reared from cynipid galls but are instead reared from cyclorrhaphous Dip- tera (Ronquist 1999, Buffington et al. 2007). Further, Eucoilinae all possess a scutellar plate with a glandular release pit and nearly all Aspicerinae (except Melanips) have a sinuate posterior margin of T-3 of the metasoma (Ronquist 1999, Buffington et al. 2007). VGLUME 17, NUMBER 1, 2008 Bie. te. anterodorsal view; B, mesosoma, lateral view; C, mesosoma, dorsal view; D, forewing, dorsal view; E—M, female metasoma, posterolateral view. Abbreviations: LPC, lateral pronotal carina; F, mesopleural furrow; MMI, median mesoscutal impression; ARE, areolet; MP, micropore. Redescription—Female. Head. Black to rusty orange; frons rugulose, densely setose; malar space costulate ventral of eye, rugulose approaching mandibular base; gena and vertex costulate, covered in short appressed setae (Fig. 2A—D); gena 47 A-C, F: Euceroptres montanus Weld; D: E. maritimus Weld; E: E. whartoni n. sp. A, mesosoma, broadly rounded (Fig. 2A—C). Antenna basally orange, distally ranging from orange to dark brown, non-clavate; scape 2.25-3 length of radicle, short appressed setae on all flagellomeres, 10-12 flagello- meres present, moniliform (Fig. 2E-F); JOURNAL OF HYMENOPTERA RESEARCH Table 1. Species of Euceroptres, their gall wasp hosts and oak hosts. Euceroptres Andricus gall wasp host Oak host species and section E. maritimus A. quercussuttoni . agrifolia, Erythrobalanus . montanus A. truckeensis . chrysolepis, Protobalanus . primus A. quercusflocci, A. quercusfutilis . alba (Q. stellata), Quercus . whartoni A. quercusoperator . nigra, Erythrobalanus (A. quercuspetiolicola)' (Q. alba, Quercus)! 1. Dubious host record and E: Euceroptres primus Ashmead; B and F: E. maritimus Weld; C and D: Euceroptres whartoni n. sp. -oma, dorsal view; D, mesosoma, lateral view; E-F, habitus, female. VOLUME 17, NUMBER 1, 2008 apical segment 2x length of subapical segment. Mesosoma. Lateral surface of pronotum deeply rugulose, densely covered in stiff, moderately long setae (Fig. 1A—B); lateral pronotal carina (LPC, Fig. 1A) well devel- oped, extending from lateral margin of pronotal plate to ventral margin of ante- roventral inflection of pronotum (Fig. 1A); lateral margins of pronotal plate indistinct; submedial pronotal depressions deep, open laterally (Fig. 1A). Mesopleuron cos- tulate to rugulose anteriorly, setose; meso- pleural furrow composed of rugae (F, Fig. 1B); mesopleural triangle deeply im- pressed, setose, clearly defined along all edges (Figs 1B & 2D); area posterior of mesopleural triangle and dorsal of meso- pleural furrow highly polished, glabrous (Figs 1B & 2D). Mesoscutum transversely carinate to rugulose, moderate to densely setose; anteroadmedian signum present; median mesoscutal impression present, ranging from short, notch-like to % length of mesocutum; notauli complete, originat- ing at anterior end of parascutal impres- sion, gradually becoming wider posteriorly (Figs 1C & 2 A-C). Disk of scutellum heavily rugulose, evenly setose (Fig. 2 A- F); scutellar ridge separating scutellar fovea narrow, short; scutellar fovea oval, obliquely angled relative to midline, pos- terior rim present, center gently rugulose, sparsely setose (Figs 1C & 2A-C). Metapleural-propodeal complex. Meta- pleuron and propodeum ranging from glabrous to completely covered in long setae; anterior margin of upper meta- pleural area jutting-out laterally, glabrous (Fig. 1B); setal pit at ventral margin of 49 metapleuron present; posterior aspects of propodeum smooth to gently rugulose, flat; propodeal carinae thin, complete, parallel; area between propodeal carinae glabrous to setose, with dense, felt-like setae under long, thin setae. Nucha short, glabrous, deeply striate. Fore wing. Marginal cell closed along anterior margin (Fig. 1D); distinct break present in vein proximal to marginal cell (Pies 1»): sareoletn present, (ARE Fie. 1D); marginal and cubital veins represented by trace veins; short setae present on wing surface and along margins. Legs. Femora orange, tibiae orange to dark brown; sparse, appressed setae pres- ent on all femorae and tibiae. Tarsomeres orange-yellow to brown, covered in short, appressed setae (Fig. 2E-F). Metasoma. Ranging from black or brown to orange; petiole frequently ob- scured by anterior margin of T3. Posterior margins of T3 and T4 parallel, angled obliquely at 45 degrees relative to horizon- tal, subequal in length; remaining terga short, telescoped within T4; T4-T9 with micropores (MP, Fig. 1F) though signifi- cantly reduced to absent in some species; setae frequently present on T8. Male. As in female but with 13 flagello- meres; flagellomere 1 as long as fourth antennal segment, laterally excavated, ex- panded slightly on distal end. Distribution. Nearctic Region: United States of America: AZ, CA, DC, FL, MA, MD, OR, TX, VA. Biology. Parasitoids of species of Andri- cus (Cynipidae: Cynipini), which are gall inducers on various species of oak (Quercus Spp.). KEY TO SPECIES OF EUCEROPTRES (FEMALES AND MALES UNLESS OTHERWISE NOTED) i Females with 11 flagellomeres. Metasomal terga 3 — 8 with significantly reduced to absent micropores (Fig. 1) (collected East of the Rocky Mountains) = Females 10 or 12 flagellomeres. Metasomal terga 3 — 8 with well developed micropores (MP, Fig. 1F) (collected West of the Rocky Mountains) 2. Median mesoscutal impression short, notch-like e © © © © © © «© 50 JOURNAL OF HYMENOPTERA RESEARCH . Median mesoscutal impression elongate, often 1/4 to 1/3 length of mesoscutum noi tx. ene ee pee ee rete ees Euceroptres whartoni, new species 3. Females with 10 flagellomeres. Dense, felt-like setae present in mesopleural triangle and setal pits between propodeal carinae (Fig. 1B). Area of episternum posterior to anterodorsal margin of metepisternum gently crenulated .. Euceroptres maritimus Weld - Females with 12 flagellomeres. No felt-like setae present. Area of episternum posterior to anterodorsal margin of metepisternum smooth and glabrous Dees Re hg oa ee Euceroptres montanus Weld ah Was wee Je ee eee eee ee eee ee ee ae Euceroptres maritimus Weld, 1926 Figs 1D & 2B, F Diagnosis.—Females_ readily distin- guished by having 10 flagellomeres (all other species with either 11 or 12). Sepa- rated from E. primus and E. whartoni by the presence of well-developed micropores on the metasoma (cf. MP, Fig. 1F); this trait is also shared with E. montanus. Males and females have dense felt-like setae present in the mesopleural triangle, the setal pits of the metapleuron and between the propo- deal carinae; all other species have setae in these areas, but they are not dense and felt- like. Redescription.—Female. As in description of genus, with antennae of female 12 segmented (10 flagellomeres); median me- soscutal impression deep, short, notch-like (Fig. 2B); mesoscutum distinctly trans- versely rugulose across entire surface; dorsal-anterior margin of mesoplueron umbilicate-rugulose, transitioning to rugu- lose ventrally; mesopleural triangle with dense, felt-like setae; metapleuron evenly covered with long setae except episternal area posterior to anterior impression of metepisternum, smooth, glabrous; felt-like setae in ventral setal pit; metasomal T4—T9 with distinctly visible micropores (cf. Fig. 1F), dense setal band present along posterior margin of T8; metafemora orange; metatibia medially orange, lateral- ly dark brown; pro- and mesotarsomeres dark orange to brown; metatarsomeres dark brown or black. Male. As in female but with 13 flagello- meres; flagellomere 1 as long as fourth antennal segment, laterally excavated, ex- panded slightly on distal end. Material examined.—Holotype. [first label] “Berkeley, Calif 4/20/12” [20 April 1912], [second label] ‘““Mrs. G.D. Louderbeck’”’, [third label] “1601”, [fourth label] ‘Quercus agrifolia’’, [fifth label] ‘“Type 27299, U.S.N.M.”, [sixth label] ‘’Euceroptres maritimus Weld’. The holo- type is a female in good condition, deposited in the USNM. Additional material. Allotype. Same data as holotype, 1 male (USNM). Paratypes. USA: CALIFORNIA. Alameda Co. Same data as holotype, 1 female; Oakland. Bred by Bassett from galls collected by W.M. Beutenmueller, Beutenmueller Coll., received 1935 [no other data available], 1 male (USNM). Los Angeles Co., Santa Anita, Hopkins File number 156055, June 17-18 1918, reared from Quercus agrifolia, collected and bred by L. Weld, 1 female (USNM). Alameda Co., ‘through C.V. Riley’, [no other data available], 4 males, 3 females (USNM). Non-types. USA: CALIFORNIA. Ala- meda Co., Berkeley, gall collected 10 Mar 1928 from Quercus agrifolia, emerged 15 May 1928, W. Ebeling, coll., 3 males, 3 females (AMNH). Los Angeles Co. Pasadena, gall collected 22 Feb 1920 from Quercus agrifolia, Kinsey Coll., ex gall of Callirhytis polythyra Bassett, 23 males, 46 females (AMNH); Claremont, Metz Coll., acc 5635 [no other data available], 1 male (AMNH); Eaton Canyon State Park.,19.IV.2004, ex gall on Quercus agrifolia, M. Buffington & J. Liljeblad, 12 males and 15 females (1 male, 1 female under voucher 56737, UCRM; remaining specimens in USNM). OREGON. Josephine Co., Grants Pass, coll #8536, taken from Quercus spp., [no other data available], 5 males, 13 females (AMNH). Note: Weld (1926) recorded that part of the paratype series was sent to the USNM under the Hopkin’s number 156055, which was supposedly collected in Montecito, CA VOLUME 17, NUMBER 1, 2008 (near Santa Barbara); the specimen in USNM bearing this Hopkin’s number is labeled as being collected in Santa Anita, CA (near Pasadena), not Montecito. Distribution.—Western United States, from southern Oregon in the north to southern California in the south. Biology.—Reared from the cynipid An- dricus quercussuttoni (Bassett) on Quercus agrifolia Née. Euceroptres montanus Weld, 1926 Fis. TA-C, F Diagnosis.—Females_ readily distin- guished from other Euceroptres species by the possession of 12 flagellomeres. Males can be distinguished from E. maritimus and E. primus by having the area of the episternum posterior of the anterodorsal margin of the metepisternum smooth and glabrous (gently crenulate and setose in E. maritimus and E. primus); from males of E. whartont by the umbilicate anterodorsal margin of the mesopleuron (rugulose in E. whartoni), and the presence of micro- pores on T4 —T9 (present but barely visible in E. whartont). Redescription.—Female. As in descrip- tion of genus, with antennae of female with 12 flagellomeres; median mesoscutal impression deep, short, notch-like (Fig. 1C); mesoscutum distinctly transversely rugulose across entire surface; dorsal- anterior margin of mesopleuron umbili- cate, transitioning to rugulose ventrally; mesopleural triangle setose; metapleuron evenly covered with long setae, with denser setae in ventral setal pit; metaso- mal T4 -T9 with distinctly visible micro- pores (cf. Fig. 1F), dense setal band present along posterior margin of T8; metafemora and metatibiae orange; pro-, meso- and metatarsomeres dark orange to brown. Male. As in female but with 13 flagello- meres; flagellomere 1 as long as fourth antennal segment, laterally excavated, ex- panded slightly on distal end. 51 Material examined.—Holotype. [first label] “Idyllwild, Cal.’’, [second label] ‘1622’, [third label] ‘““Type No. 27228 U.S.N.M.”, [fourth label] “Euceroptres montanus Weld”. The type is a female in good condition, deposited in USNM. Additional material. Allotype. USA: CALIFOR- NIA. Riverside Co. Idyllwild, April [19]23 (remaining label data as in holotype labels), 1 male (USNM). Paratypes. USA: CALIFORNIA. Riverside Co. Idyllwild, April [19]23 (remain- ing label data as in holotype labels), 17 males, 19 females (USNM). Trinity Co. Big Bar, cut out Dec [19]23, code 1622, 1 female (USNM). El Dorado Co. Kyburz, code 1622 [no other data available], 1 female (USNM). OREGON. Dou- glas Co. Canyonville, June 3 [no year recorded], code 1622, Beut.[enmueller] Coll., rec’d 1935, 2 males, 4 females (USNM). Josephine Co. Hol- land, April [19]23, code 1622, 1 male (USNM). Non-types. USA: CALIFORNIA. Santa Clara Co. Los Gatos, Hopkin’s number 159225, reared various dates between 7-21 May 1919 from Quercus chrysolepis Liebmann, R.D. Hartman, collector, 5 males, 4 females (USNM). Note: Weld (1926) recorded two speci- mens of this species recorded from Kern Co., CA, from the ‘Museum’, presumably the USNM; these specimens were not located. Distribution. —Western United States, from southern Oregon in the north to southern California in the south. Biology.—Reared from the cynipid gall inducer Andricus truckeenis (Ashmead) on Quercus chrysolepis Liebm. in CA. This data taken from Weld (1926); no specimens were found in the USNM with associated host data. Euceroptres primus Ashmead, 1896 Fig. 2A & 2E Diagnosis.—Females distinguished from E. montanus and E. maritimus by possessing 11 flagellomeres (12 in E. montanus, 10 in E. maritimus) and lacking micropores on metasomal T4 through T9 (males and females). Euceroptres whartoni also has 13 antennal segments in the female, but E. primus can be separated from E. whartoni by the short median mesoscutal impression 52 (elongate,1/4-1/3 length of mesoscutum in E. whartoni; compare Fig. 2A with 2C) in both males and females; FE. primus also lacks micropores on metasomal T4-T8 (faintly visible in E. whartoni). _ Redescription.—Female. As in description of genus, with antennae of female with 11 flagellomeres); median mesoscutal impres- sion deep, short, notch-like (Fig. 2A); me- soscutum distinctly transversely rugulose across entire surface; dorsal-anterior mar- gin of mesopleuron gently rugulose, setose; mesopleural triangle setose; metapleuron evenly covered with long setae, anterior margin of episternum gently rugulose, occasionally glabrous; dense setae in ven- tral setal pit; metasomal T4-T9 without visible micropores (cf. Fig. 1E), dense setal band lacking along posterior margin of T-8 (cf. Fig. 1E); metafemora and metatibiae orange; pro-, meso- and metatarsomeres dark orange to brown. Male. As in female but with 13 flagello- meres; flagellomere 1 as long as fourth antennal segment, laterally excavated, ex- panded slightly on distal end. Material examined.—For the purposes of no- menclatural stability, the female specimen in the USNM, currently labeled ‘type #3286’ is desig- nated as lectotype. Lectotype. [first label] “Through C.V. Riley”, [second label] ‘2640, scrub oak, Whitfelt, Georgiana, Fla., Mar 24 — [18]82, [third label] ‘Type No. 3286, U.S.N.M., [fourth label] ‘Euceroptres primus Ashm.”, [fifth label] Lectotype designation. The lectotype is a female in good condition, deposited in USNM. Additional material. Paralectotypes. USA: FLOR- IDA. Brevard Co. Georgiana, through C.V. Riley, 2640, scrub oak, Whitfelt, Georgiana, Fla., Mar 24 -{18]82, 1 male (USNM) [this specimen was included in Ashmead’s original description]; Georgiana, through C.V. Riley, 2640, scrub oak, Whitfelt, Georgiana, Fla., Mar 7 -[(18]82, 1 female (USNM). MASSACHUSETS. Merrimac River, “780P”, through C.V. Riley, 3 July, 1883, ex Quercus alba L., 2 females (USNM). Non-types. USA: VIRGINIA. Fairfax Co. Falls Church, Minor’s Hill, Hopkin’s number 84892, reared 24 Jun — 11 Jul 1912, ex Quercus alba L. Wm. Middleton, collector, 5 males, 7 females JOURNAL OF HYMENOPTERA RESEARCH (USNM); Falls Church, Hopkin’s number 84895, reared 24 Jun 1912, ex Quercus alba L., Wm. Middleton, collector, 4 males (USNM); Falls Church, Hopkin’s number 84912, reared 29 Jun 1912, ex galls of Callirhytis papillatus, Wm. Middleton, collector, 4 males, 4 females (USNM); Falls Church, Hopkin’s number 12059, reared 25 Jun 1914, ex Quercus alba L., Wm. Middleton, collector, 1 male (USNM); Falls Church, Kearney, Hopkin’s number 120692, reared 27 Jul 1914, ex Quercus minor (Marshall) Sarg., Wm. Middleton, collector, 1 male (USNM); Falls Church, Hopkin’s number 13600, reared 26 Apr 1915, ex galls of Andricus flocci, Wm. Middleton, collector, 4 males (USNM); Falls Church, Kearney, Hopkin’s number 120692, reared 27 Jul 1914, ex Quercus minor (Marshall) Sarg., Wm. Middleton, collec- tor, 1 male (USNM). Distribution.—Eastern United States, from Maryland in the north to Florida in the south. Biology.—Reared from Andricus quercus- futilis (Osten-Sacken) and Andricus quercus- flocci (Walsh) on Quercus alba L.; also reared from an unknown cynipid host on Quercus stellata Wangenheim. Weld (1926) also records this species from petiole galls on Quercus stellata from Rosslyn, VA but these specimens could not be located in the USNM. Euceroptres whartoni Buffington & Liljeblad, new species Figs 1E & 2C-& Diagnosis.—Females can be _ distin- guished from E. montanus and E. maritimus by the possession of 11 flagellomeres (12 in E. montanus, 10 in E. maritimus) and the lack of distinct micropores on metasomal T4 through T9 (Fig. 1E) (males and fe- males); distinguished from E. primus by the presence of an elongate median mesoscutal impression (short and notch-like in E. primus; viz. Fig. 2A & 2C). Description.—Female. As in description of genus, with antennae of female with 11 flagellomeres; median mesoscutal impres- sion deep, elongate, reaching 1/4 to 1/3 length of mesoscutum (Fig. 2C); mesoscu- VOLUME 17, NUMBER 1, 2008 tum distinctly transversely rugulose ante- riorly, less striate posteriorly; dorsal-anteri- or margin of mesopleuron shagreen to gently rugulose, setose; mesopleural trian- gle setose; metapleuron evenly covered with long setae, anterior margin of epister- num gently rugulose, frequently glabrous; dense setae in ventral setal pit; metasomal T4 -T9 without visible micropores (Fig. 1E), dense setal band lacking along posterior margin of T8 (cf. Fig. 1E); metafemora and metatibiae orange; pro-, meso- and metatar- someres dark orange to brown. Male. As in female but with 13 flagello- meres; flagellomere 1 as long as fourth antennal segment, laterally excavated, ex- panded slightly on distal end. Etymology.—Named in honor of our mentor and friend, Robert Wharton. Material examined.—Holotype. [first label] Hopk. U.S. 107675, [second label] reared Mar. 26.21 [26 Mar 1921], Quercus minor, [fourth label] Denton, TX, [fifth label] Marquis, R.L., coll., [sixth label] holotype, Euceroptres whartoni Buffington & Liljeblad. The holotype is a male in good condition, deposited in the USNM. Additional material. Paratypes. Same data as holotype: USA: TEXAS. Denton Co. Denton, ex triangular galls on Quercus minor, collected 26 Jan 1920 [emergence date not recorded], R.L. Marquis, collector, Hopkin’s number 10767 (2 males and 4 females, NHMN). Non-types. USA. FLORIDA. Volusia Co. Hopkin’s number 15634¢, reared 15 Oct 1922 from woolly midrib cluster galls collected 8 Dec 1919 from Quercus laurifolia Michx. (1 female, NMHH). MARY- LAND. Montgomery Co. Plummers Island, 12 Apr 1914, W.L. McAtee, coll. (1 male, USNM). MISSOURI. Stoddard Co. 30 Mar 1938, T-10242, on peach (1 female, USNM). VIRGINIA. Fairfax Co. East Falls Church, Hopkin’s number 13651i, 23 Apr 1917, reared from Callirhytis operator sexual generation [in Weld’s hand] collected from Quercus marylandica Du Roi, Wm. Mid- dleton, coll., 1 Jun 1916 (2 females, USNM). [collection data unknown] No. 2640, Apr 19.82 [(19 Apr 1882(?)], (1 male USNM); No. 2640, Apr 2182 (Qi Apr 1882(?)), (i male USNM); [collection data unknown] “with A. cicatricula Bass.” [this specimen corresponds to a speci- men mentioned in Weld (1926), originally 58 determined as E. primus, that was originally found among the cotype material of Cynips cicatricula Bassett, collected from Quercus alba in Waterbury, CT.] (1 male, USNM). Distribution.—Eastern and Southeastern United States, from Texas in the West to Connecticut and Maryland in the East. Biology.—Reared from Andricus quercus- operator (Osten-Sacken) galls on Quercus nigra L. in Virginia. A second rearing record is circumstantial at best: specimens mentioned above that were associated with Cynips cicatricula Bassett (=Andricus quer- cuspetiolicola (Bassett)) were bred from Quercus alba L. Quercus nigra and Q. alba belong to different sections of Quercus subg. Quercus; the former is a red oak wheras the latter is a white oak. Not much is known about the host-specificity of Euceroptres. If they are anything like the cynipid inquilines or even true parasitoids, these host records could very well be correct. On the other hand, no Nearctic species of oak gall wasps is known to attack hosts from more than one oak section (Stone et al. 2002). DISCUSSION Two distinct lineages of Euceroptres emerged from this study. One, composed of E. maritimus and E. montanus, appears to be restricted to the western Nearctic Region, chiefly collected in CA and OR. The second lineage comprises E. primus and E. whartoni and occurs in the eastern and southeastern Nearctic Region. In fact, this latter lineage may be rather widespread throughout the southeast. Within Euceroptres, based on characters described herein, the following set of relationships is proposed: ((E. mon- ianuses e. mariimus, \(E primus 4. I. whartoni)). The first clade (E. montanus + E. maritimus) is united by the shared presence of micropores on the metasoma (Fig. 1F) and a distribution restricted to the western Nearctic. The second clade, composed of (E. primus + E. whartoni) are united by the shared presence of 13 antennal segments in the female, the overall reduction in micro- 54 JOURNAL OF HYMENOPTERA RESEARCH Ibaliidae P Liopteridae P 50/N Euceroptrinae A 50/72 — —= Plectocynipinae A — Thrasorinae P 63/N ri Parnipinae A ao Diplolepidini G 7oN} oe Cynipini G /9 ant m— Anacharitinae P Lonchidia ? 0.62/N 100/79 tet Shae — cos? Figitinae P 00/8 ° = =< Aspicerinae P 100/100 Same at Charipinae P 100/100 ae Emargininae ? 0.85/N 100/84 — Gronotoma Group P 100/100 94/61 . Zaeucoila Group* P Eucoilinae se a Core Eucoilinae P 10 changes Fig. 3. Phylogram of Figitidae resulting from Bayesian analysis of 28S D2&D3 and 18S rRNA, COI and morphology (see Materials and Methods). Analyses of the same data using parsimony result in nearly the same tree (see below); ‘*’ at the Zaeucoila Group indicates this clade is sister-group to the Gronotoma group in the parsimony analysis. Letters after terminal names refer to: A, agastoparasite; G, gall inducer; P, non- agastoparasite. Numbers on branches indicate branch support in the form of Bayes posterior probability/ parsimony bootstrap support; ‘N’ indicates less than 50% support was recovered for that node; Bayesian posteriors calculated using 50% majority rules consensus. Parsimony analysis resulted in 375 trees distributed across 46 islands; CI=.18, RI=.62. pores on the metasoma, and a distribution encompassing the eastern/southeastern Nearctic Region. A striking symplesiomorphy of Eucerop- tres is the presence of a very well devel- oped lateral pronotal carina (Fig. 1A). As stated earlier, this character is shared with Liopteridae (Ronquist 1995b) and Stolamis- sidae (Liu et al. 2007), the figitids Aspic- erinae, Parnipinae, the Gronotoma group of Eucoilinae (Buffington et al. 2007) and a few Cynipidae (Ronquist 1995b). Conclusive evidence for the exclusion of Euceroptres from Thrasorinae was provided in two independent studies (Ros-Farre and Pujade-Villar 2007, Buffington et al. 2007) VOLUME 17, NUMBER 1, 2008 as well as this study (Fig. 3). With the erection of Euceroptrinae, a system of small subfamilies, sister to the remaining higher taxa within Figitidae, reflects a rather complicated evolutionary history (Fig. 3). It should be noted, however, that the branch support for (Euceroptrinae (Plectocynipinae+Thrasorinae)) is weak in both parsimony and Bayesian analyses (Fig. 3). As stated by Buffington et al. (2007), additional data are required to definitively resolve these relationships; data collection for additional species of Plectocynipinae and Thrasorinae is cur- rently underway (Buffington and Scheffer unpublished). The association of these lineages with gall-inhabiting hymenopterous hosts begs the intriguing possibility that phytopha- gous cynipids arose from a rather diverse range of proto-figitids attacking various gall-inducing hosts (Ronquist and Nieves- Aldrey 2001). These lineages also may represent the origins of agastoparasitism within Cynipoidea, though Nylander (2004) and Melika (2006) suggest inquilin- ism arose independently in Cynipidae numerous times. Certainly within Figiti- dae, agastoparasitism is the plesiomorphic life-history strategy (Fig. 3, terminals let- tered ‘A’, agastoparasites; ‘G’, gallers; ‘P’, non-agastoparasites), with the more de- rived Thrasorinae shifting to chalcidoid hosts (Ros-Farré and Pujade-Villar 2007, Buffington et al. 2007, Buffington submit- ted). Although the cynipid hosts of Euceroptres gall only 5 or 6 species of Quercus, their oak hosts could hardly be a more diverse sample coming from the Nearctic (Table 1). All three sections of Quercus subgenus Quercus are represented, the missing fourth being the exclusively Palearctic section Cerris (Manos et al. 1999). This could be just a random sample from a few species attacking a number of Andricus gall wasps. If, however, species of Euceroptres are more host-specific, it lends further support to the idea that this genus was once a more 3D species rich group of which only a few scattered lineages have survived to date. Liu et al. (2007) date the split of cynipids and figitids to at least the early Cretaceous, providing evidence that even ‘neo-eucoi- lines’ (‘core’ Eucoilinae of Fontal-Cazalla et al. 2002; Zamischus group of Buffington et al. 2007) were present in the mid-Creta- ceous. These data suggest the Figitidae are indeed an old lineage, and members of these depauperate ancestral lineages may represent so-called ‘living fossil’ taxa, giving us a tantalizing opportunity to look into the evolutionary history of this diverse group of parasitoid Hymenoptera. ACKNOWLEDGMENTS We wish to thank Fredrik Ronquist and Sean Brady for assistance in setting up MrBayes analyses; David Furth provided critical assistance in obtaining Hop- kin’s records; John Heraty supported the fieldwork that resulted in fresh material of Euceroptres maritimus under NSF PEET BSR 9978150. Juli Pujade-Villar and Palmira Ros-Farré provided fruitful discussion on thrasorines and plectocynipines. Reviews provided by Michael Pogue and Thomas Henry (Systematic Ento- mology Lab, Washington DC), Molly Rightmyer (Department of Entomology, Smithsonian Institute), George Melika (Systematic Parasitoid Laboratory, Tanakajd, Hungary), Juli Pujade-Villar (University of Barcelona, Spain) and Andy Deans (NCSU, Raleigh, NC) improved the quality of this paper considerably. LITERATURE CITED Ashmead, W. H. 1896. Description of new parasitic Hymenoptera. Transactions of the American Ento- mological Society 23: 179-234. Buffington, M. L., J. A. A. Nylander, and J. M. Heraty. 2007. The phylogeny and evolution of Figitidae (Hymenoptera: Cynipoidea). Cladistics 23: 403-431. ———.. in press. A revision of Australian Thrasorinae (Hymenoptera: Figitidae) with a description of a new genus and six new species. Australian Journal of Entomology 00: 000-000. Dowton, M. and A. Austin. 2001. Simultaneous analysis of 16S, 285, COI and morphology in the Hymenoptera: Apocrita — evolutionary transi- tions among parasitic wasps. Biological Journal of the Linnean Society 74: 87-111. Fontal-Cazalla, F. M., M. Buffington, G. Nordlander, J. Liljeblad, P. Ros-Farré, J. L. Nieves-Aldrey, J. Pujade-Villar, and F. Ronquist. 2002. Phylogeny of the Eucoilinae (Hymenoptera: Cynipoidea: Figitidae). Cladistics 18: 154-199. 56 Gillespie, J. J. 2004. Characterizing regions of ambig- uous alignment caused by the expansion and contraction of hairpin stem-loops in ribosomal RNA molecules. Molecular Phylogenetics and Evo- lution 33: 936-943. , J. B. Munro, J. M. Heraty, M. J. Yoder, A. K. Owen, and A. E. Carmichael. 2005. A secondary structural model of the 285 rRNA expansion segments D2 and D3 for chalcidoid wasps (Hymenoptera: Chalcidoidea). Molecular Biology and Evolution 22: 1593-1608. Harris, R. 1979. A glossary of surface sculpturing. State of California, Department of Food and Agricul- ture. Occasional Papers in Entomology 28: 1-31. Kovalev, O. V. 1994. Paleontological history, phylog- eny and the systematics of brachycleistogastro- morphs and cynipomorphs (Hymenoptera, Bra- chycleistogastromorpha infraorder n., Cynipo- morpha infraorder n.) with description of new fossil and recent families, subfamilies and genera. Entomologicheskoe Obozrenie 73 (2): 385-426; 495 [Translated to English in Entomological Review 74: 105-147]. Liljeblad, J. and F. Ronquist. 1998. A phylogenetic analysis of higher-level gall wasp relationships (Hymenoptera: Cynipidae). Systematic Entomology 23: 229-252. Liu, Z., M. Engel, and D. Grimaldi. 2007. Phylogeny and geological history of the cynipoid wasps (Hymenoptera: Cynipoidea). American Museum Novitates 3583: 1-48. Manos, P. S., J. J. Doyle, and K. C. Nixon. 1999. Phylogeny, biogeography, and processes of mo- lecular differentiation in Quercus subgenus Quer- cus (Fagaceae). Molecular Phylogenetics and Evolu- tion 12: 333-349. Melika, G. 2006. Gall wasps of Ukraine. Vol. 1. Vestnik Zoologiii, Journal of Schmalhausen Institute of Zoology, Supplement 21 (1): 1-300. Nylander, J. A. A. 2004. Bayesian phylogenetics and the evolution of gall wasps. [Doctoral Dissertation]. Acta Universitatis Uppsaliensis. Uppsala. JOURNAL OF HYMENOPTERA RESEARCH Ronquist, F. 1994. Evolution of parasitism among closely related species: Phylogenetic relationships and the origin of inquilinism in gall wasps (Hymenoptera, Cynipidae). Evolution 48: 241-266. . 1995a. Phylogeny and early evolution of the Cynipoidea (Hymenoptera). Systematic Entomolo- gy 20: 309-335. . 1995b. Phylogeny and classification of the Liopteridae, an archaic group of cynipioid wasps (Hymenoptera). Entomologica scandinavica, Sup- plement No. 46: 1-74. . 1999. Phylogeny, classification and evolution of the Cynipoidea. Zoologica Scripta 28: 139-164. and J. L. Nieves-Aldrey. 2001. A new subfamily of Figitidae (Hymenoptera, Cynipoidea). Zoological Journal of the Linnean Society 133: 483-494. and G. Nordlander. 1989. Skeletal morphology of an archaic cynipoid (Hymenoptera: Ibaliidae). Entomologica Scandinavica Supplements 33: 10. , A. P. Rasnitsyn, A. Roy, K. Eriksson, and M. Lindgren. 1999. Phylogeny of the Hymenoptera: a cladistic reanalysis of Rasnitsyn’s (1988) data. Zoologica Scripta 28: 13-50. Ros-Farré, P. and J. Pujade-Villar. 2007. Plectocynipi- nae, a new subfamily of Figitidae and description of a new Neotropical genus of Thrasorinae (Hymenoptera: Cynipoidea). Zootaxa 1583: 1-13. Stone, G. N., K. Sch6nrogge, R. J. Atkinson, D. Bellido, and J. Pujade-Villar. 2002. The population biology of oak gall wasps (Hymenoptera: Cynipidae). Annual Review of Entomology 47: 633-668. Weld, L. 1926. Field notes on gall-inhabiting cynipid wasps with descriptions of new species. Proceed- ings of the United States National Museum 68: 1-131. . 1952. Cynipoidea (Hym.) 1905-1950 being a Supplement to the Dalla Torre and Kieffer monograph - the Cynipidae in Das Tierreich, Lieferung 24, 1910 and bringing the systematic literature of the world up to date, including keys to families and subfamilies and lists of new generic, specific and variety names. Privately Published, Ann Arbor, Michigan. 351 pp. J. HYM. RES. Vol. 17(1), 2008, pp. 57-63 Load Carriage during Foraging in Two Species of Solitary Wasps JOSEPH R. COELHO, JON M. HASTINGS, CHARLES W. HOLLIDAY, AND ANGELA MENDELL (JRC) Institute for Franciscan Environmental Studies, Biology Program, Quincy University, 1800 College Ave, Quincy, IL 62301, U.S.A. (MH, AM) Department of Biological Sciences, Northern Kentucky University, Highland Heights, KY 41099, U.S.A. (CWH) Department of Biology, Lafayette College, Easton, PA 18042, U.S.A. Abstract.— Foraging strategies of two species of solitary digger wasps (Hymenoptera: Crabronidae) were examined. Both species capture insects and return with them in flight to a burrow where they are used as food for larvae. The actual loading of the wasps was compared to a theoretical ideal, the maximum that they should be able to carry in flight. Bembix troglodytes Handlirsch is a relatively large wasp that carries many small flies to its burrows. The flies are much smaller than the ideal size, but the choice of small prey appears to be adaptive in that it should reduce the rate of stealing of flies by conspecific females, which was a common event. Tachytes chrysopyga Cresson is a relatively small wasp that carries relatively large prey, grasshoppers and crickets, to its burrow. The average size of prey appears to be ideal; however, the distribution of prey size is so great that many wasps were underloaded, while others were overloaded. Prey theft was not observed in T. chrysopyga, and flexible flight behaviors (e.g. short, hopping flights) allow it to carry a broad range of prey sizes. These two wasp species may represent near extremes of a continuum of behavior among predaceous wasps. Although much is now known about the foraging and nest provisioning behaviors of solitary wasps (O’Neill 2001, Evans and O'Neill 2007), their foraging strategies have not been extensively studied using the conceptual framework offered by optimal foraging theory. One approach to examin- ing foraging decisions in such species is to develop a prediction of the optimal load size of the prey, and to test this prediction against the size of actual loads. Fortunate- ly, the maximal load size is simple to calculate for species that carry the prey in flight. Marden (1987) measured the maxi- mum lift force of a variety of flying animals, including 33 bees and wasps, by progressively loading individuals with weights until they could not take off. The force production of flying animals is primarily dependent on their flight muscle mass. In Hymenoptera, this relationship is quive* strong. (> 0=> 0:99), sand-at» the maximum load mass, the ratio of flight muscle mass to body mass (or flight muscle ratio, FMR) is 0.179 (Marden 1987). Use of flight muscle ratio as a metric for flight capability has several advantages, primarily its independence of the size of the wasp. For wasps that carry their prey in flight, the maximum force produced must equal or exceed the combined weight of the wasp and prey. Assuming it is optimal for a wasp to carry the largest prey possible, the FMR (now flight muscle mass divided by combined body and prey mass) should approach but not exceed 0.179. We have tested this prediction in yellowjackets (Vespula spp, Coelho and Hoagland 1995), the eastern cicada killer (Sphecius speciosus Drury, Coelho 1997), the great golden digger wasp (Sphex ichneumoneus L., Coelho and LaDage 1999), and the carpen- ter wasp (Monobia quadridens L., Edgar and Coelho 2000). In no case yet examined has 58 the FMR of foraging wasps fallen precisely at the predicted value; however, investiga- tion of the causes for the deviation from “optimality’’ has always revealed interest- ing insights into foraging behavior. In this study we apply this method to examine the foraging strategies of two rather different crabronid wasps facing different selection pressures. Bembix The general biology of Bembix and specific details of B. troglodytes Handlirsch, the subject of this study, are described by Evans (1957; 1963). Only relevant aspects of their behavior are summarized here. Bem- bix is a genus of crabronid wasps that hunt flies (Diptera) and carry them to their burrow in flight. The female digs a single burrow in the ground and excavates a nest cell. In progressive provisioners, an egg is laid and attached to the first prey in the cell or to the substrate in the center of the cell. In B. troglodytes, oviposition occurs in an empty cell, and the first fly is captured, paralyzed and placed in the cell later (Evans 1957, Evans and O’Neill 2007). The female continues to capture flies and feed them to its developing larva. As the larva increases in size, the rate of provisioning increases (Tengo et al. 1996). A final flurry of foraging provides the larva with all the flies it will require and the female then seals the burrow and digs another. Bembix burrows often occur in high-density aggre- gations, perhaps as a response to parasite and predator pressure, as the relative incidence of parasites per nest decreases with increasing nest density in at least one species (Larsson 1986). Bembix troglodytes females close the burrow when leaving it to hunt, but only when the larva is young. The burrow is left open during the inten- sive foraging phase and closed only at night. Bembix exploit the most abundant flies available, apparently learning the richest sources of flies and repeatedly exploiting them (Evans 1957, 1963). Bembix troglodytes preys upon a large variety of JOURNAL OF HYMENOPTERA RESEARCH flower-visiting flies, though most are rela- tively small in size. Exceptionally small flies are provided to young larvae, while older larvae are provisioned with larger flies. Digging and provisioning a single nest requires only about six days and from 21 to 26 flies are required to fully provision a larva (Evans 1957). Tachytes Tachytes is a genus of digger wasps that stocks its underground burrows with Orthoptera carried in flight from foraging areas. Tachytes digs complex burrows with multiple cells and packs each cell with up to 10 prey items. The egg is not laid until the cell is fully provisioned (Evans and Kurczewski 1966, Elliot and Salbert 1981), a case of mass provisioning. Tachytes nest aggregations are sometimes associated with those of Sphecius, as they were in this study. In one species, T. distinctus F. Smith, males establish perches near the nest entrance of a female and chase any insect that flies near, including brood parasites such as Zanysson texanus (Cresson) (Lin and Michener, 1972). MATERIALS AND METHODS We conducted our observations and measurements on a large nesting aggrega- tion of Bembix troglodytes Handlirsch at the Hot Springs area of Big Bend National Park (N 29°- 10' 39:57” latitude, Wize 52.73'' longitude, Brewer Co., Texas) from 22 to 23 May 2006. We noted other ageregations, apparently of the same spe- cies, at Santa Elena Canyon and Boquillas Canyon, all in sandy areas immediately adjacent to the Rio Grande River. These ageregations were in approximately the same locations as those we observed in previous years. Female wasps carrying prey were netted and weighed on an Ohaus Adventurer-Pro electronic balance to the nearest mg. The head, abdomen, legs and wings of the wasps were removed with scissors and the thorax mass deter- mined. Flight muscle mass was estimated VOLUME 17, NUMBER 1, 2008 as 95% of thorax mass (Marden 1987). The mass of the prey fly and its thorax mass were similarly determined. Several of the wasps were collected as voucher speci- mens and deposited in the Lafayette College Insect Collection and the Quincy University Life Sciences Museum. We discovered a large colony of Tachytes chrysopyga obscurus Cresson nesting on an earthen berm within an even larger aggre- gation of the eastern cicada killer (Sphecius speciosus Drury). This berm was located within a large chemical production facility (Flint Hills Resources) in Will County, IL, deena 26 39.31" latitude, W 88° 10’ 22.16"’ longitude. From 27 to 28 July, 2006, female wasps were collected as they returned to their burrows with prey. Body, thorax and prey mass were determined as described above. Several wasps were col- lected as voucher specimens and deposited with the California Academy of Sciences entomology collection. All prey were de- posited in the Quincy University Life Sciences Museum. All data are reported as mean + standard error unless otherwise indicated. Descriptive statistics were calcu- lated using Microsoft® Excel 2003, while T tests were performed using VassarStats (Lowry 2007). RESULTS Bembix troglodytes Bembix troglodytes females averaged 99 + 2(N = 48) mg in body mass and 37 + 0.003(N = 48) mg in thorax mass, resulting in an unladen FMR of 0.36 + 0.004(N = 48). Fly prey of B. troglodytes averaged 45 + 3(N = 33) mg in body mass and 16 + 1.1(N = 33) mg in thorax mass, resulting in an FMR of, 0.39 = 0.01(N = 32). Carrying flies resulted in a loaded FMR of 0.29 + 0.006(N = 27) for female B. troglodytes and loaded FMR ranged from 0.23 to 0.34. Hence, all fly-carrying females had FMRs well above the marginal FMR of 0.179 (Fig. 1A). There was no significant relationship between wasp body mass and prey body 59 mass in B. troglodytes. However, small wasps were restricted to the smallest flies, while larger wasps carried a greater range of sizes of flies (Fig. 2A). There were usually many wasps flying about the nest aggregation, some of which were doubtlessly males engaged in the sun dance (Evans 1957). Many, however, were females. As successful hunters returned to the aggregation, they were nearly always pounced upon by conspecifics before they had the opportunity to land. Often the prey was dropped, and then picked up by the same or another female. Prey-laden fe- males were fast, maneuverable, surprising- ly difficult to distinguish from unladen females, and difficult to catch. The fly was tucked under the body and held tightly with all legs. A slightly larger profile normally provided the investigator with the cue that a wasp was carrying prey. Bembix troglodytes with prey generally landed and entered their burrows very rapidly if they escaped attempts at prey theft. Digging activity was frequent. The this- tledown velvet ant (Dasymutilla gloriosa (Sauss.)), a brood parasite, was common in the area, and often observed digging in the sand as well as entering open burrows. Brood parasitic satellite flies were fre- quently observed perched at nest entrances and occasionally entering burrows, some- times closely following prey-laden B. trog- lodytes down their burrows. Prey flies collected for this study were dismembered to determine their FMR, and could not be subsequently identified to any great degree. Tachytes chrysopyga Tachytes chrysopyga females averaged 52 + 2.6(N = 31) mg in body mass and 18 + 1.0(N = 31) mg in thorax mass, yielding an unladen FMR of 0.33 + 0.007(N = 30). Prey items of T. chrysopyga averaged 50 + 4.8(31) mg, producing a mean loaded FMR of 0.18 + 0.007(N = 30) in the wasps. The latter value is very close to the 60 JOURNAL OF HYMENOPTERA RESEARCH 81 A. Bembix troglodytes Number of wasps 0 +——— Te T AG i T i 0:1 O12 20.14 018 O18) O27 0.220 0.2400 2er 028) 0a 02 0 Sa Se © J Tachytes chrysopyga a OT, O12 014 “ONS O16 Of 022-004 0.25 (026. 03 032 Usui Loaded FMR Fig. 1. The distribution of flight muscle ratios (FMR) in wasps carrying prey. Arrows indicate the marginal flight muscle ratio, below which take-off is not possible. a. Bembix troglodytes from south Texas. b. Tachytes chrysopyga obscurus from northern Illinois. CO | 7) Number of wasps =— Mo © Bh OW 0 70 a al & 6 , — aie mM 50 po s 30 Se . ee ° s oO . 2 a PS * Q 20 r ‘ : > 2 Fat faut , *e Bembix troglodytes 0 ait T I T = 1 i = las Sally = = Sy lo al 140 5 sn + dD) ; = 120 + B w 100 5 ve 7) = 80 + ” . . > 60 4 > a cali ha a 40 ‘ « oe .? ‘4 ° > > = . Ki > @ 207 * 4 = : Tachytes chrysopyga Q. 0 T SF 20 30 40 50 60 70 80 90 100) eetiG! 120 iostst Wasp body mass (mg) Fig. 2. The effect of wasp size on prey size in two solitary wasps. a. Bembix troglodytes from south Texas. b. lachytes chrysopyga obscurus from northern Illinois. The lines were fitted by eye to each graph to demonstrate the maximum prey size for wasps of a given body mass. VOLUME 17, NUMBER 1, 2008 marginal FMR of 0.179, which lies within the 95% confidence interval of loaded FMR (0.16 to 0.19). However, the range of prey mass (11 to 132 mg) resulted in a very broad distribu- tion of FMRs, with many individuals far above (maximum = 0.26) and many far below (minimum = 0.099) the predicted value (Fig. 1B). At times, prey-loaded T. chrysopyga were fast, maneuverable and difficult to capture. However, some indi- viduals performed short, hopping flights along the ground and were easily caught. It was very common for the females with prey to land, perhaps even releasing their prey for a few seconds, en route to their burrows. We never observed prey stealing, nor did we ever see conspecific females lurking around the burrows of others. In fact, we found small numbers of aban- doned prey in the area of the nest ageregation. Although satellite flies were often observed closely trailing prey-laden S. speciosus, perching near and entering their burrows, we never saw evidence of such flies similarly harassing female T. chrysopyga in the same area. During much of the day, many T. chrysopyga were flying about low to the ground, though some were certainly males. The burrows were small and relatively inconspicuous, with entranceways some- times near that of an eastern cicada killer. There was no obvious tumulus near the T. chrysopyga entranceway, though the small amount of dirt could easily have been displaced by weather or other disturbanc- es. Entranceways appeared to be open, although they could have been sealed deeper in the burrow. Prey of T. chrysopyga were diverse Orthoptera, including Gryllidae (16 Enop- terinae, 7 Oecanthinae), Tettigoniidae (1 Conocephalinae, 3 Copiphorinae) and Ac- rididae (8 Cyrtacanthacridinae). The sole adult among the prey was a meadow grasshopper (Conocephalinae). There was no significant relationship between wasp body mass and prey body 61 mass in T. chrysopyga. But, as in B. troglodytes, small wasps were restricted to the smallest prey, while larger wasps carried a greater range of sizes of orthop- terans (Fig. 2B). B. troglodytes was significantly larger than T. chrysopyga in both body mass (P < 0.0001, T = —10.85, df = 77) and thorax mass (P < 0.0001, T = —12, df = 76). Unladen FMR was significantly higher in B. troglodytes (P < 0.0001, T = —10.25, df = 76), as was laden FMR (P < 0.0001, T = —10.25, df = 55). However, prey mass was significantly higher in T. chrysopyga than B. troglodytes (P = 0.0025, T = 3.15, df = 62). DISCUSSION Bembix troglodytes At first glance, B. troglodytes appears to be a suboptimal forager, at least in the context of our FMR-based model. The wasps could in theory decrease the time spent hunting and the number of hunting trips by simply capturing and carrying larger flies. This strategy would doubtless increase the number of onepaae that could be reared during the wasps’ short life spans, but there may be mitigating factors involved. Larger flies may be less available or more difficult, energetically expensive and time-consuming to capture. The flies that were captured had lower FMR than unladen B. troglodytes. FMR is a useful indicator of maneuverability (see Marden 1989, Marden and Chai 1991). Hence, B. troglodytes should have been able to cap- ture the flies on the wing. Prey records for Bembix include fast-flying tabanids (Evans 1957). It is possible that the foraging strategy of B. troglodytes minimizes prey stealing. Few wasps returning with prey made it to their burrows unchallenged, and though some of the attacking conspecifics could have been males attempting to mate, theft by females was frequent. Evans (1957) did not report prey stealing in B. troglodytes, but the aggregations he studied were relatively 62 diffuse. Ours was rather dense, which increases the likelihood of thievery or harassment by males. Evans (1957) did report prey stealing in five of ten species of Bembix examined. In the related wasp Stictia heros (Fabr.), in which prey theft is common, the probability of being attacked is directly related to the size of prey (Villalobos and Shelly 1996). These authors suggest that female Stictia carrying large prey were more vulnerable because of their lower flight speed and maneuverability. Therefore, if B. troglodytes females took larger flies, they would likely suffer greater rates of conspecific attack. Small prey help them maintain a high level of maneuver- ability, as we observed, and probably improve the likelihood of successful trans- port of the prey all the way to the burrow. Bembix females can compensate for small prey size by increasing the number of prey. Tachytes chrysopyga On first appraisal, T. chrysopyga appears to be an optimal forager (using our simple FMkR-based model), its loaded FMR being indistinguishable from the predicted value that would maximize load carriage. On average, I. chrysopyga takes prey that make full use of its load-lifting capacity. Howev- er, upon closer inspection, T. chrysopyga is, in fact, highly variable in the size of prey it takes, and consequently, the magnitude of loaded FMR it experiences. These results are comparable to those of Elliot and Salbert (1981), who found that T. tricinctus (F.) prey varied from 36.3 to 214 mg, averaging 93.8 mg. These prey are about twice the size of those of T. chrysopyga, which is not surprising, as T. tricinctus females, averaging 126.5 mg, are over twice the size of T. chrysopyga. However, the approximately tenfold range in prey mass in both species may reflect a similarly opportunistic foraging behavior. Prey need not be exceedingly small, as prey stealing is not apparent in these species. Overloading is compensated by behavior — short, hop- ping flights being adequate to return some JOURNAL OF HYMENOPTERA RESEARCH prey to the burrow. Prey need not be exceedingly large either, as Tachytes provi- sions with a variable number of prey (Evans and Kurczewski 1966), and greater numbers could compensate for smaller size. In spite of being ground-nesting digger wasps with many behaviors in common, B. troglodytes and T. chrysopyga provide inter- esting contrasts in foraging strategy. As the vast majority of its prey are flightless, T. chrysopyga does not require great maneu- verability (bestowed by high unladen FMR) to capture them, as does B. troglodytes. Bembix troglodytes is a large wasp that takes many, small prey, while T. chrysopyga is a small wasp that takes fewer, larger prey (Fig. 2). Both revealed their maximum prey size, as demonstrated by the nearly straight line that can be drawn through the highest points (at a give body size) in Figs 2 A and B. For B. troglodytes, the maximum is somewhat below the wasp’s own body mass, while for T. chrysopyga it is well above the wasp’s body mass. In each species of wasp, the same predictions were applied and essentially the same methods used. Neither species conformed to these predic- tions; however, much was learned about the biology and behavior of each wasp. ACKNOWLEDGMENTS We thank Wojciech Pulawski and Allan Hook for species identifications. We are indebted to the National Park Service for permission to conduct research in Big Bend National Park under NPS permit # BIBE-2005-SCI-003, and to Flint Hills Resources, Inc. for permission to conduct research on their premises, and especially to the Wildlife Committee for provid- ing the required supervision. This work was support- ed in part by grants from the Quincy University Faculty Development and Welfare Committee (JRC), the Northern Kentucky University Center for Integra- tive Natural Sciences and Mathematics (JMH), and the Lafayette College Academic Research Committee (CWH). LITERATURE CITED Coelho, J. R. 1997. Sexual size dimorphism and flight behavior in cicada killers (Sphecius speciosus). Oikos 79: 371-375. VOLUME 17, NUMBER 1, 2008 — and J. Hoagland. 1995. Load-lifting capacity and foraging of three species of yellowjackets (Vespula) on honey-bee corpses. Functional Ecology 92171-174. and L. D. LaDage. 1999. Foraging capacity of the great golden digger wasp, Sphex ichneumoneus L. Ecological Entomology 24: 480-483. Edgar, P. K. and J. R. Coelho. 2000. Load-lifting constraints on provisioning and nest building in the carpenter wasp, Monobia quadridens L. (Hy- menoptera: Eumenidae). Journal of Hymenoptera Research 9: 370-376. Elliott, N. B. and P. Salbert. 1981. Observations on the nesting behavior of Tachytes tricinctus (F.) (Hy- menoptera: Sphecidae, Larrinae). Journal of the New York Entomological Society 88: 170-173. Evans, H. E. 1957. Studies of the comparative ethology of digger wasps of the genus Bembix. Cornell Univer- sity, New York. . 1962, The evolution of prey-carrying mecha- nisms in wasps. Evolution 16: 468-483. and F. E. Kurczewski. 1966. Observations on the nesting behavior of some species of Tachytes (Hymenoptera: Sphecidae, Larrinae). Journal of the Kansas Entomological Society 39: 323-332. and K. M. O’Neill. 2007. The sand wasps: natural history and behavior. Harvard University, Cam- bridge. Field, J. 2005. The evolution of progressive provision- ing. Behavioural Ecology 16: 770-778. 63 and S. Brace. 2004. Pre-social benefits of extended parental care. Nature 428: 650-652. Larsson, F. K. 1986. Increased nest density of the digger wasp Bembix rostrata as a response to parasites and predators (Hymenoptera: Spheci- dae). Entomologia Generalis 12: 71-75. Lin, N. and C. D. Michener. 1972. Evolution of sociality in insects. Quarterly Review of Biology 47: 131-159. Marden, J. H. 1987. Maximum lift production during takeoff in flying animals. Journal of Experimental Biology 130: 235-248. . 1989, Bodybuilding dragonflies: costs and benefits of maximizing flight muscle. Physiological Zoology 62: 505-521. and P. Chai. 1991. Aerial predation and butterfly design: how palatability, mimicry and the need for evasive flight constrain mass allocation. American Naturalist 138: 15-36. O’Neill, K. M. 1985. Egg size, prey size, and sexual size dimorphism in digger wasps (Hymenoptera: Sphecidae). Canadian Journal of Zoology 63: 2187- 2193. Tengo, J., H. Schéne, W. D. Kithme, H. Schone, and L. Kihme. 1996. Nesting cycle and homing in the digger wasp Bembix rostrata (Hymenoptera Sphe- cidae). Ethology, Ecology and Evolution 8: 207-211. Villalobos, E. M. and T. Shelly. 1996. Intraspecific nest parasitism in the sand wasp Stictia heros (Fabr.) (Hymenoptera: Sphecidae). Journal of Insect Be- havior 9: 105-119. J. HYM. RES. Vol. 17(1), 2008, pp. 64-82 Revisionary Studies on the Enigmatic Neotropical Ant Genus Stegomyrmex Emery, 1912 (Hymenoptera: Formicidae: Myrmicinae), With the Description of Two New Species RODRIGO M. FEITOSA, CARLOS R. F. BRANDAO AND JORGE L. M. DINIZz (RMF, CRFB) Museu de Zoologia da Universidade de Sao Paulo, Av. Nazaré 481, Sao Paulo, SP, 04263-000, Brazil; RMF email: rfeitosa@usp.br; CRFB email: crfbrand@usp.br (JLMD) Campus Jatai, Unidade Jatoba, Universidade Federal de Goidas, BR 364, km 192, Jatai, GO, 75800-000, Brazil; email: jlmdiniz05@hotmail.com Abstract.—The recent increase in leaf litter ants sampling effort in Neotropical wet forests has revealed new and interesting records of the highly specialized myrmicine ant genus Stegomyrmex Emery, previously considered as extremely rare. We present a modified diagnosis for the genus and describe Stegomyrmex bensoni n. sp. and S. olindae n. sp., based on, respectively, workers, males, and gyne (central-north Brazil) and on a single worker (northern Brazil). Stegomyrmex vizottoi Diniz (southeastern Brazil) is redescribed and compared with S. olindae n. sp.; these species present significant differences in size, sculpturation, and sting apparatus morphology. The males of S. vizottoi are described for the first time. A key for workers and queens and a distribution map for the five know Stegomyrmex species are provided. Stegomyrmex is the sole representative of the peculiar and exclusively Neotropical myrmicine tribe Stegomyrmecini (Bolton 2003). These ants have been considered extremely rare by many authors, perhaps due to their cryptobiotic habits enhanced by peculiar soil-binding pilosity (H6élldo- bler and Wilson 1986), and by the foraging technique they employ (Diniz and Brandao 1993). However, recent collections employ- ing large-scale sampling (e.g. Agosti et al. 2000) have revealed that they are relatively common inhabitants of the dense leaf litter of Neotropical forests. Emery (1912) described Stegomyrmex with a single species, S. connectens, based on a gyne and a male from Peru and Bolivia respectively. Emery included Ste- gomyrmex in the Dacetini based on gyne characters. Wheeler (1922) established a new tribe, Stegomyrmicini (sic), with Ste- gomyrmex as its only member. He separat- ed it from the Dacetini mainly by the shape of mandibles and wing venation. Smith (1946) described the second Stegomyrmex species, S. manni, from Barro Colorado Island, Panama, and agreed with Wheeler’s placement of the genus in an individual tribe. Bernard (1951) and Lenko (1965) com- mented on the morphological resemblance of Stegomyrmex to some Attini. However, Brown (1949), Brown and Kempf (1960), and Holldobler and Wilson (1986) consid- ered stegomyrmecine ants more closely related to Basicerotini than to Attini or Dacetini, mainly by the presence of deep antennal scrobes and the soil-binding pilosity. Dlussky and Fedoseeva (1988) considered Stegomyrmex as incertae sedis in Myrmicinae, without further discus- sion. However, in the last proposals of Bolton (1994, 2003, 2006 et al.), Stegomyr- mex is placed in its own tribe within the Myrmicinae. Bolton (2003) commented that the structure of the promesonotum may suggest a relationship between Ste- gomyrmecini and Pheidolini, but that there is no undisputed evidence for this yet. VOLUME 17, NUMBER 1, 2008 Lenko (1965) found a worker of S. vizottoi (identified by him as S. manni) in the gizzard of a Conopophaga lineata Wied (Aves, Conopophagidae). Hélldobler & Wilson (1986) commented on the pre- sumed role of the soil-binding hairs of basicerotine and stegomyrmecine ants in enhancing their camouflage to predators. Diniz (1990) was the first to revise the taxonomy of Stegomyrmecini, describing the third species of the genus, Stegomyrmex vizottoi, based on workers and a gyne from Brazil and Paraguay. In the same work, Diniz commented on the relatively slow movements of stegomyrmecine ants. Diniz and Brandao (1993) were the first to describe the nesting habits of Stegomyrmex, based on observations on colonies of S. vizottoi from Mirassol, state of Sao Paulo, Brazil, describing nest architecture, population dis- tribution among nest chambers, different worker behaviors at each part of the nest, and the foraging habits of the workers, which exploit the environment surrounding their nests singly, searching for myriapod eggs. Recent surveys of leaf litter ants in the Brazilian Atlantic forest and in sparse localities of central and northern Brazil revealed several Stegomyrmex specimens, including a remarkable new species de- scribed here, and extending considerably the known distribution range of S. vizottoi. Our analysis of S. vizottoi along its distri- bution shows, however, that, as presently accepted, it includes two distinct species, recognizable by the surface sculpture, by morphometry, and by differences in the sting apparatus, as we fully describe and comment on below. In this paper we offer taxonomic notes on the peculiar ant genus Stegomyrmex, based on the study of the specimens deposited in the Museu de Zoologia da Universidade de Sao Paulo ant collection, literature information, and enriched by unpublished observations. We also de- scribe two new species and comment on new records and information regarding these seldom collected ants. METHODS This study was based on the available specimens in the collection of the Museu de Zoologia da Universidade de Sao Paulo, Sao Paulo, Brazil, which is believed to hold most of the known Stegomyrmex specimens. Depository collections are referred to by the following acronyms: BMNH - The Natural History Museum, London, UK. CASC - California Academy of Sciences, San Francisco, California, USA. CPDC —- Centro de Pesquisas do Cacau, Itabuna, Bahia, Brazil. JLMD - Laboratério de Zoologia, Cam- pus Jatai, Universidade Federal de Goias, Brazil. LACM - Los Angeles County Museum of Natural History, Los Angeles, Califor- nia, USA. MCSN - Museo Civico di Storia Natur- ale ““Giacomo Doria’’, Genoa, Italy. MPEG - Museu Paraense Emilio Goeldi, Belém, Para, Brazil. MZSP —- Museu de Zoologia da Uni- versidade de Sao Paulo, Sao Paulo, Brazil. USNM - National Museum of Natural History — Smithsonian Institution, Wash- ington, DC, USA. The terms for external morphology and surface sculpturing follow, respectively, Bolton (1994, 2000) and Harris (1979). The terms for wing venation follow Brown and Nutting (1950). The reproductive females are here called ‘‘gynes’’, as suggested by De Andrade and Baroni Urbani (1999). Measurements were obtained with a micrometric reticule and using the scale of a scanning electron microscope (SEM). All measurements are given in mm, and the abbreviations used are: HW: head width; the maximum width of the head capsule, measured in full face view, at a median transverse line that touches the superior margins of the com- pound eyes. HL: head length; the maximum measur- able length of head capsule excluding the 66 mandible, measured in full face view, in a straight line from the midpoint of the anterior clypeal margin to the midpoint of the vertexal margin. SL: antennal scape length; the chord length of the antennal scape, excluding the basal condyle and its peduncle. WL: mesosoma length (Weber’s length); the diagonal length of mesosoma in profile, from the midpoint of the anterior pronotal declivity to the posterior basal angle of the metapleuron. PL: petiole length; the longitudinal axis of petiole in lateral view. PPL: postpetiole length; the longitudinal axis of postpetiole in lateral view. GL: gaster length; the maximum length of gaster in lateral view, excluding sting. TL: total length; the summed length of HL (plus the closed mandibles), WL, PL, PPL, and GL. CI: cephalic index. HW x 100/HL. SI: scape index. SL x 100/HW. The SEM images of Stegomyrmex speci- mens were obtained from a single speci- men of each species. The specimens were previously cleaned in acetone, critical- point dried in a Balzer (Bal-Tec® CPD 030), and sputtered over with gold (Bal- Tec® SCD 050). After that, the specimens were mounted on the tip of metallic triangles using silver glue and then affixed to stubs for the electron microscopy. The images were obtained under several mag- nifications (40 to 300x), according to the size of the specimen and/or structure observed. Finally, the images were edited (Adobe PhotoShop 7.0®) to enhance some brightness and contrast details. We studied also the sting apparatus of the species from which we had enough individuals. The sting was obtained by rehydrating ants in 70% ethanol, extracting the terminal segments from the gaster, clearing them in 55-60°C lactophenol for five minutes (or longer if necessary), rinsing twice in 70% ethanol, and twice in 95% ethanol. After the clearing process, the sting apparatus was dismembered, soaked JOURNAL OF HYMENOPTERA RESEARCH in xylene, and then mounted in Canada balsam for observation and _ illustration under optical microscope. The terms for sting apparatus morphology follow Kugler (1978). Coordinates of localities were obtained from the information on the specimens labels and after consulting the ENCARTA World Atlas® (Microsoft); they were plot- ted on the distribution map generated by the software ArcView 3.2 GIS®. When citing label data, we present additional information between brackets, explanation of codes on the labels, eventual corrections to the misprints, and reference to the notebooks from which we took information regarding the localities and/ or the biology of the species. RESULTS Stegomyrmex Emery, 1912 Stegomyrmex Emery, 1912: 99. “Gynes type species: Stegomyrmex connectens, by mono- typy. Emery, 1912: 101 (placement in Dace- tini); Emery, 1914: 42 (placement in Dace- tini); Forel, 1917: 246 (placement in Dacetini); Emery, 1924: 314 (placement in Dacetini, diagnosis, catalogue); Wheeler, 1922: 668 (establishment of Stegomyrmecini [as Stego- myrmicini]); Donisthorpe, 1943: 727 (place- ment in Dacetini, list of type specimens); Smith, 1946: 286 (revision); Brown and Kempf, 1960: 162 (systematic notes); Lenko, 1965: 201 (distribution and biology); Kempf, 1972: 242 (catalogue); Wheeler and Wheeler, 1985: 258 (tribal classification); Hd6lldobler and Wilson, 1986: 16 (pilosity); Dlussky & Fedoseeva, 1988: 81 (incertae sedis in Myrmi- cinae); Diniz, 1990: 277 (revision, species key); Holldobler and Wilson, 1990: 15 (tribal classification); Brandao, 1991: 379 (cata- logue); Diniz and Brandao, 1993: 301 (biol- ogy); Bolton, 1994: 106 (catalogue); Bolton, 1995a: 1052 (census); Bolton, 1995b: 392 (catalogue); Serna, 2002: 217 (first record for Colombia); Bolton, 2003: 255 (diagnosis, synoptic classification); Fernandez and Os- pina, 2003: 49 (census); Fernandez, 2003: 325 (genera list for Neotropics); Bolton et al., 2006 (catalogue). VOLUME 17, NUMBER 1, 2008 Worker. Monomorphic. 5 to 6.5 mm in length. Reddish brown to black. Integu- ment thick, shining and in general densely areolate, except for S. bensoni. Pilosity conspicuous and bizarre; hairs varying from short, subdecumbent and filiform to long, suberect, and variably branched; mandibles sparsely covered by long fili- form hairs; anterior margin of clypeus bearing one or two pairs of very long setae, reaching half the length of mandi- bles, but without an isolated median seta; appendages covered by short decumbent hairs and by a fine and dense appressed pubescence; inferior corners of pronotum with a dense row of plumose hairs. Head subtrapezoidal with vertexal mar- gin slightly depressed to slightly convex; occipital corners angulate; broader pos- terad. Palpal formula 2:2. Labrum bilobed. Mandible triangular, long, strongly curved down apically and with the blades crossing apically when mandibles are closed; mas- ticatory border multidenticulate (total den- tal count 12-15), with the apical tooth longer than the preceding ones. Median portion of clypeus narrow, flat and vertical, not bicarinate, quite narrowly inserted between the frontal lobes. Frontal lobes enormously expanded anterolaterally and projected far out over the lateral portions of clypeus and mandibles. Each frontal lobe covering dorsally a very deep anten- nal scrobe; in full face view, space between frontal lobes narrowest near the middle of head, revealing the compound eyes; clyp- eus and basal portion of mandible entirely concealed by the frontal lobes. Frontal area impressed, glabrous and smooth, the ante- rior suture obsolete. A shallow groove, almost devoid of any sculpture, present on each side of the head dorsum, extending from the frontal area to the occipital corner of head, the two grooves meeting anterior- ly, forming a noticeable V. Antenna with 12 segments, with a three-segmented club; antenal scape slender, curved basally and broader at apex. Compound eye exceed- ingly small, oval in shape, almost indistinct 67 from the integument sculpture; placed on the sides of head immediately beneath the antennal scrobe, but visible when head is in full face view. Mesosoma, in dorsal view, slender, widest at the level of the anterior area of pronotum. Promesonotum evenly rounded in profile, dome-like; anterosuperior corner of anepisternum set much lower than the adjacent surface, forming a deep fovea; promesonotal suture almost obsolete in some individuals. Mesonotum elongate with posterior portion sloping down; me- tapropodeal impression relatively broad and usually shallow, except for S. bensoni. Propodeum, in side view, variably convex dorsally, and with the declivity sinuous; propodeal spiracles low on side and raised in prominent, subcylindrical protuberanc- es; propodeal spines short and more or less acute; propodeal lobes large and usually projected over the petiolar peduncle. Legs relatively long; femora and tibiae moder- ately incrassated; tarsal claws simple; metatibial spurs absent. Petiole long and pedunculate, with about twice the length of the postpetiole; petiolar node variably convex in profile; ventral carina present and bearing 0-2 blunt anterior projections. Postpetiole approxi- mately as long as broad, globose, without ventral process. Gaster oval, without basal shoulder; tergite of abdominal segment IV (first gastral) not broadly overlapping sternite on gaster ventral surface. Sting apparatus. Spiracular plate with spiracle placed ventrally; anal plate with several sensillae; lancet with a pair of functional valves; furcula with indistinct dorsal arms. Gyne. Like conspecific worker, with the modifications expected for myrmicine gynes. Anterior ocellus slightly larger than posterior ones. Notauli and parapsidial lines usually indistinct from surrounding sculpture; prescutellum with central area indistinct, scutoscutellar sulcus shallowly impressed, with transversal rugulae vary- ing in number and forming distinct cells; 68 lateral wing of prescutellum not projecting laterally; scutellum semicircular, with its posterior half always sloping down and with posterior border concave; propodeal spines shorter than in conspecific workers. Forewing with distinct and strongly colored stigma; longitudinal veins Sc+R, SR, M+Cu, and A present. Cells R, Cu and 1M closed. Hind wing with Sc+R extending shortly beyond point where they connect to M, which extends as a tubular vein up to the wing distal border; Cu cell closed and very short; six to eight submedian hamuli. Male. Dark brown to black, with ap- pendages and gaster usually lighter. Integ- ument densely sculptured, opaque or nearly so, except for the postpetiole and gaster which are smooth and_ shining; appendages very finely punctate. Pilosity composed of fine hairs, whitish to golden, mostly curved or suberect on body, sparser on metasoma. Apressed pubescence on antennae and legs. Head broadest across compound eyes, narrowed anteriorly; median portion of vertexal margin usually weakly convex; occipital corners rounded; ocelli promi- nent. Mandible relatively developed and subtriangular; masticatory border multi- denticulate, with the apical tooth much more developed than the others. Clypeus broad. Frontal lobes not so developed as in the conspecific gynes and workers, but concealing the antennal insertions, forming a short and shallow antennal scrobe. Antennae long and slender with 13 seg- ments; scapes relatively short. JOURNAL OF HYMENOPTERA RESEARCH Mesosoma robust; prescutellum separat- ed from scutellum by an impression with short longitudinal rugae. Scutellum narrow posterad. Metanotum narrow, with blunt median tumosity. Propodeum dorsal face flat, steeply sloping posterad, unarmed. Legs slender, middle and hind tibiae without apical spurs; tarsal claws slender and simple. Wing venation as in the gynes. Petiole clavate, pedunculate, and with a long, low, rounded node. Postpetiole as broad as long, attached to the gaster by almost its full width. Gaster elongate, with first segment occupying most of its length; visible apical segments subequal in length. Comments.—We revise the Stegomyrmex diagnosis presented by Diniz (1990) in order to include information on the shape of the head and on the structure of the alate’s mesosoma, besides features present in S. bensoni n. sp and S. olindae n. sp. Apomorphies for Stegomyrmecini defined by Bolton (2003) hold true for the new species. Despite the recent information regarding these seldom collected ants, the phyloge- netic position of Stegomyrmex remains truly enigmatic. The affinities with Dacetini and Attini, proposed in the past, seem improb- able by the significant differences in habits and morphology. Despite the body sculp- turation patterns and the presence of specialized pilosity approximating Stego- myrmex and Basicerotini (H6lidobler and Wilson 1986), the possibility of homoplasy can not be presently discarded. REVISED KEY TO THE STEGOMYRMEX SPECIES (WORKERS AND GYNES) ih Integument of mesosoma predominantly smooth and shining; body covered mainly by sparse aggregations of somewhat curved and multibranched hairs; metapropo- deal groove deeply impressed; petiole without anteroventral spines (state of Para, Brazil) ®) © fe oe ne #8 ei es 66nd) oO eye tee. dice. vaio ye eee, sie, eo eel cee 60) A ee ee cee ere & S. bensoni sp. n. - Integument of mesosoma predominantly sculptured, areolate; body covered mainly by sparse and erect clavate setae; metapropodeal groove only moderately impressed; petiole with at least one anteroventral spine VOLUME 17, NUMBER 1, 2008 69 Ze filiform hairs; known only from the gyne (Peru and Bolivia) .. . Se laterally (Costa Rica, Panama and Colombia) strongly projected laterally 4. Petiole with two anteroventral spines; inferior margin of pronotum with a row of S. connectens Emery Petiole with a single anteroventral spine; inferior margin of pronotum with a row of ollie Imaiines eLearn | 3 Promesonotum much higher than propodeum, in lateral view; propodeal spines blunt and directed posteriorly; in dorsal view propodeal spiracles strongly projected S. manni Smith Promesonotum slightly higher than propodeum, in lateral view; propodeal spines subtriangular, acute and directed upwards; in dorsal view propodeal spiracles not ‘cigte “ie hin je), \e" te) Ye) 0) wa’s: fe. tale fe! <8) es Mesosoma length = 1.59 mm; mesosoma partially sculptured, with foveae sparsely set on the polished integument; metapropodeal impression without a projecting tubercle; nucal area predominantly smooth; in dorsal view basal face of propodeum relatively narrow (northern Argentina, Paraguay and southeastern Brazil)...... wi Lelme sie! ‘ef ce) (e" 10) mite)”.0) (eo) 6 el) s) “ene (e./'e| (ee) je, je «Ker \e. « « ©) «) fetal le jelgie) «: (o («, (a) ve: ‘eo (a) le) a: Xo).¢p) ie af 0! (ce! S. vizottoi Diniz - Mesosoma length < 1.59 mm); mesosoma strongly sculptured, with the integument completely areolate; metapropodeal impression with a projecting tubercle; nucal area predominantly sculptured; in dorsal view basal face of propodeum relatively broad (central-north Brazil) ....... Stegomyrmex bensoni n. sp. (Figs 1, 7) Holotype worker—BRAZIL: Para, Canaa dos Carajas (06°44'49’S, 50°21'05"W) (Gruta NV06) 22-28.11.2005 (Andrade & Arnoni) [MZSP]. Worker description—HW 1.26; HL 1.09; ML 0.61; SL 0.84; WL 1.77; PL 0.78; PPL 0.45; lee70; 1176.40; CT 115.56;-S1 66:35. ‘Color reddish brown. Basal portion of mandible finely and densely striate, with large and sparse piligerous punctures, apical portion and masticatory border mostly smooth and shining; inner surface of antennal scrobes with fine, dense, transversal and concentric striation; dorsal surface of head predomi- nantly smooth and shining, with scattered punctures near the vertexal border; margin of frontal lobes finely areolate-rugose; central portion of each frontal lobe virtually trans- lucent, so that is possible to observe the inner surface of the antennal scrobes near the insertions of antennae; antennae opaque and finely punctate; lateral and ventral surface of head deeply areolate; occipital face of head smooth and shining except for the nucal collar which is regularly and deeply scrobi- culate; mesosoma almost entirely smooth oe im te) lay iol re; Ke, [ej (o, Busou avuLIeseyA] CII Tus aS ‘xtpueddy © LV 8 WIT € WV 6L CE %8'°8S OOL %VEL OZT £7/¥0/€0 6 A6E9T SOO'SZ sisuaqiqo snaydauosit4y, ays NY qiqO Yl L YoleaL Cl Cl VoL. & OOL ce/F0/€0 Hoe 9L Soo°ZZ snaydauosiry, qiqg ‘eunq payy acral wel Hoe aGOe %CSS SOT €¢/F0/€0 HSE9L S6r°ZZ Z oUuTIpTy}ue sIsuaqiqo sniydauostiy, aul uotds09S VSL 8 %L VL ST WOT € %6'TS PS %8'8S COL G¢/P0/€0 Hce OL SSP Zc Z oUTIpTy}ue sIsuaqiqo snaydauosisy, aul uotdsz03S AN GE L WOVE 9 Yoel 74 %0L OC ZL/€0/720 CV) ‘Hé ASO'OL Ser ZZ (D>) Me snaydauosi4y, qneuurey> %09 TCL SV 6 %S8 OT 8/€0/7Z0 HSO'9L SPL snaydauosi4y, qneuurey) Vere |b 6 VOLiG wae SCE ES 6¢/F0/€0 Hog OL SZE-4¢ (Y6F 8 -S|yeM pnut) snaydouosi4y, shiny J %9I6L 6 %1V'8S 99 %8'C9 CLL 8¢/F0/€0 NEGO Svea sniydouosiry, shiny ¥ %VE SG Voi Ona: LOC CV ie IIE EG %CLG SPL Z€/¥0/€0 PEOUSOC HcC OL SOLE uL prprisAryo sniydoauosity, sniny %8'0 C %8'l V VSL GE Holas, VSP COL %VS LCC €F/F0/€0 Hcl 9L SLEZc 00 efoorsnjol snaydouosiry, peq JaAry qeyonsA S VE € WGV V %8 88 €0797C/90/S0 a xx A PBST SOELZ ce sIsuaqiqo snaydouosiy, dures s tezrey ru 2 %S'0 1 WIE 8 eo 0 1 %VSE 6L 7898 GC L¥/¥0/€0 Z HOL9L S6c°Zc < %G'0 1 — SffeM pnut) snaydoauosity, _ASAI0J,, PUOYIO 2 S]UepIsar sjjoys reus Aydura sjjous [reus Aydura ,soysered,, —_, Sayisered_, /sousered ,, S][ous PeT[y-pues ut pasn % 29 ou ‘OU ,S10}99][OO S siopidg ut 8urjsou seurueunY url 3uysou sepriypesayy aepryyunyy, aeprmyAquiog aeprpisAry) Suljsou oeulreseyy] iI 1194s aS 6 ‘panuyuoy ~xipueddy > JOURNAL OF HYMENOPTERA RESEARCH 104 VIG (V) Hé ES en 574 %SEV EL 9€/86/L6 (CD) Mé HIE OL SOE'8Z UY) 'V POoISNjor snaydoauostiiy, punwoluei9 J 5 € s|[e eIUuIeENG Xq snuas3 Mau L3 TC SVE 8 6 T [9 %S9G EL ESL €T 79T-091 / 26/96 elunAeNC) HOP 9L S67’ 87 vids (D>) "mM Keno snaydauosi1y, soreypueig MS IOL GS Colic. %6VL ZL %99 TE P68 LV CSI-LEL/ 26/96 T ds CY) H PTOOTYIUOD SIsueqiqo *xdOP 9L S67 87 v ds (D>) 'M e[OIIBNjal sniydoauosisy, soreypurig MS TS I Te I LIL GS %68E FL %GL IE V801-66/26/96 T 8 Cy) AOP'9L S6z7'8Z u(YV) ‘Vv eOoIsnjor sniydoauosiy, soreypueig MS VSL T ey Ay Ama %9IT8 IL %9'P8 EL L9/86/ 26 (V) Hé ACP IL $97'8z (D)'Mé ejOoIsNjol sniydauosi1y, punwoluel9 | YP GS OT € IVE YE GLE 6E %EL9 POL 0¢/F0/€0 Acer OL S80°8z yds (>) NM sniydouosiy, qiqg ‘eunq pey WIEL IL VIE EP %I9P SIT 6c/F0/€0 H7Pr9L S80°8z POoIsnjor snaydouosiy, qiqg ‘ound poy € i 9 $8 ‘18-8Z ‘79 ‘09/26/96 }sou 9}IIO[PI HLS'9L S087 PIULALNC) Ul “VW R[OSIBNJat Ul SATJIALD poT[y pues yeulg Yysoy Ss L € %OOL ZL %00L ZI ZL-€9 ‘19’6S/26/96 (SD) “Mé sou sIsuaqiqo ALS‘9OL SEO'SZ L ‘ds (Vv) ‘Hé eIUAeNG) Ul “VW + epoorsnyjad snaydouosisy, yeulgy Ysoy Ss %O0C LL “os LAL SPL 8 CW) %60E LL %LTL GS 91/€0/70 (CD) Mé ASST SO0'SZ uUCY) ‘VW snaydoauosiy, S1aqa0s800g MASS s]uapisa. sjjoys jreus Ayduo sjjoys jreus Ayduio _soysered,, soyisered,, _soyisered,, s][oys payy-pues ur pasn % 29 Ou ‘OU ,S10}D9]]09 siapidg ut Suysou seurusung ur Suljsou oevpryiupesayy eepyyuny, aepuyAquiog oevpripisAry Sulysou ovuLteseyl al us OS ‘ponuguoy ‘xipueddy 105 VOLUME 17, NUMBER 1, 2008 E656 8 LIC EC MV 8E 98 — $8/90/S0 HPS 9L SIL'6C snaydoauosis q, YIOTION Og FW OV V %OV PV %OL TL %08 OL 68/90/50 (OW) WW APpS-€G OL S9L'6Z vds(>)°M snaydouost. q, YOTION 0g V6 YI WIV € %*LVL V9 €8/90/50 HES 9L SOT'67 euesid eqouy, YIOTION 0g MES YO REC Y MWEC Y SEG IT €8/90/50 HES 9L SIT'6C é snaydouosity, UYIOTION 10g %OOL S OV C GL L WOOL S 78/90/50 HES 9L SSL'6C B snaydouosi.4y, UIOTION Og F Uns %I6 CL VOGT 61 3 VC GCL 18/90/S0 HES 9L SSL6C euesid eqouy, YIOTION Mog q ways 8 TC W8C L OV GC 691/26/96 HOV 9L SOS'8Z sniydouosi4y, UIOTION MOg N Ve vl G vole LG 2c WLGE G ISL VI €ZZ/86/26 PB[OOTYIUOD H8e91 SZV'87é yu CY)‘V eOOTsNyjor snaydouosizy, UIOTION Og N I 6 T S|[e € i 86-S6/ 26/96 eulytend xe H87 91 S8E°8c ‘Uu (CV) “Vv snue3 Mou snaydoauosi4], Aeg iapuexealVy 4 posite Z pos1sula g posisu | EL VS-VLOI HOV 9L SZE87c Aeg Japuexely AN ‘ds syytdozy 43} 25) sniydauosi4y, Weg }JOOID I oT S|[90 G ig 86-56/26/96 euljzengd xq HOE IL SSE8c U (V) ‘V snue3s Mou eTOOTYIUOD snaydoauosi4y, punuteluer9 WS syUopIsel s]joys Jreus Aydura s]jeys [reus Aydura _soysered, _ saytsvaed,, /soyseied,, S[[oUs peT[y-pues ut pasn % 29 ou ‘OU S10}D9][OO siopidg ur 8uysou seutusun | ul Suysou sepryiyoesayy seplyyyNnyY oeprijAquiog aeprpisAry) SUT|SOU deUTIESLI\y aimicrs OHS ‘penuyuod ‘xtpueaddy JOURNAL OF HYMENOPTERA RESEARCH 106 LEC PL COL 99 EI 8 INC L WEED IC %LOL 8 E83 OL %ULS IL LTE C %0C TL ouTypryyue VE V OUITUISO 3 wT 1 (Co) %09 9E v ‘ds (>) "M %L'SS PP vids (>) “M U(CV) ‘VV ris ouTIpryyue MGT € OUITUSO ¢ WELE € OUITUSO 3 SC ZL IS LT jsou erurend) ul We %69L OL eOoTYyDUOD %COL 99 %O6C BE efooryouoD 8 1L V %0C TL enbeureu %OSL CL SI v[OoryDUoD stsuonbewreu VL 6 enbeureu SGP S enbewreu %6'CP CL %C CC PV %L9 6S snaydouosiy, %69L 6S ews snaydouosisy, %V6E LCL snaydouosiiy, %VIL VE snaydouosiy, %e C6 09 snaydouosiy, %18 62 snaydouosisy, %S V8 8S snaydouosiy, %V LL ICL vuesid eqou], Yoo to” CL snaydouosiy, %UTB 8 snaydouosisy, %0S 81 snaydouo8isy, 86/90/50 He?’ sl SIE LE OT[LAZIN’'] AS 06/90/50 HePr’st STE LE diopsudysue A 06/90/50 HePr’st SLE le diopsuAysueA 8Z1/26/96 HePr’st SOE TE diopsudysueA XN 88/90/50 H87 ZL S77 0€ [eeIyoT[PM M CPL /86/26 AZo LL S72? '0E [eRDOTTPEM M 9F1/86/26 HOC LL SIZ OE [eRD9TTPM M Z8/90/S0 HZUZL S6L'0€ Aeg diyapuopyy 4 Z8/90/S0 AZUL S6l'0€ Avg diyapuopyy 4 82/90/50 HSS 9L S9OL’67 UYOTION Og F wy9 62/86/26 HSS 9L SYL'67 YIOTION WOd 4 sjuapIsal siapidg sjjoys [reus Aydura ul Suysou seurueung| sjjoys reus Ayduue ul Suysau oeprryoesayy aepuyAquiog _souseied ,, ovprpisAryy S][OYs pel]y-pues ut Suljsou oveulieseyy pasn % 2 Ou CU Iles ‘OU S10] DaT]OO aus ‘ponuyuoy *xipueddy 107 VOLUME 17, NUMBER 1, 2008 MAG. seb YoGG -& %61LL PEL TLL/90/S0 HZ78L S9P'CE tadsaeuoq euesid eqouy, puesyssieqnolg N %C 6 CL 69 6 %76 CL %8'ET OFT OLL/90/S0 HZ7 S81 S9P'SE euesid eqoyy, puessieqnolg N EPL LT MCV € MULG Z CTLI+OL1L/90/S0 HZ78L S9P'SE tedsaeuoq sniydouosi1y, pue.yssieqnolg N I ll 611/90/S50 HIT SL SCP'EE esiodse xI[apy pueyssogyea SPL TC OD ZL IOC SST 611/90/S0 HIT SL SCPE’ 1adsaeuoq euesid eqouy, pueyssoqyaW %V60L 87d SoZ LS IT CL WS 9L IGT P1/Z0/90 HOLS. SO? SE 5 euesid eqayy, UT9}UOJ.19Z %9 CL G0 LT ME YO %G'6 O07 enbeureu ssa ID 2 LD ‘MG sTSusqIqo HOLS. SOC EE é tadsaeuoq euesid eqoyy, UTO}JUOJ1IZ JX YL Coiy TPUL 2p Jommoj H81'8L SIE CE tadsaeuoq snaydauosii{, "Sdy ‘JEN uedsayosoyy LLL IC %L0 LT MLiG ¥. TON WG /4iv/\t 601/90/50 HO? SL S61LZE tadsaeuoq euesid eqauy, Aeg spueyq ECL EL LD GS %6LZ1 IO €6/90/S50 H6L'8L SSO'CE euesid eqoul, Aeg sjrloquiey] VGAG Hexetore, 12 IGG 6 €6/90/S0 H6L'8L SSO'CE v ds (D>) "mM Tadsaeuoq sniydoauosiry, Aeg syzoqure’] 6 LT LG TC %L'S9 CC MIL GE 6/90/50 HZUSL SPS TE (D) M tadsaeuoqg sniydeuosiry, Teeqsulyuoq a a a a EN eee eS a Oe ES) ee AR ie ki ee ee) 2 Og A a ee sjuapIso1 s]joys [reus Aydure sjjeys [reus Aydure ,saysered,, |, Saytsered,, _soysered,, S][oUS PeT[y-pues ur pesn % 2p ou ‘OU ,SI0}D9][OO siapidg ut 3uljseu seutuewng ul Suysou seprynpesayy seplyyny, sepmpAquiog aepriprisAmy5 SuljSou seuTIesey\y CI Ileus aS ‘panuyuoy ‘xipueddy JOURNAL OF HYMENOPTERA RESEARCH 108 8ST 9 8ST O8E 01/20/90 eye3T ASE 97 SGF'EE eroydopido.ry, sauyog %0 OOL 6/20/90 HS¢97 SGP'Se eroydopridory, souyog %0 OOE 8/20/90 eye3I| age'cz SIS’ eroydopridory, Aeg Jayemon{g %0 O0T 8/20/90 H8e'Sc SIGE euesid eqouy, Aeg 1oyeMmony{g %6VT € 6V'L COT 9/20/90 eyeST] AZE'ST SZO'PE eroydoprido.y, Aeg erurpies %F0'S LL i PEG LEC ¢/Z20/90 ACE'ST SCOPE » vuesid eqoyy, Aeg erurpies 0 € PUeYSY %0 © euesid eqoyy, 91/20/90 %P6E OL %P6E PSC HZIU'ST S6G°EE eyes, eroydoprdo.y, YNo|py IAT pueypyreyy YLT SL MGE € %IT GB €OL/€0/7Z0 46°77 SOO'FE (D>) Mé euesid eqou SSOUTOP|IM IFC LL EET IT %8P 69 LOL/€0/Z0 HcS'07 SET PE siyeaysne vuesid eqouyy, puers}iM %OSG OC MEL G %L9 88E FOL/€0/70 HL 1? SCC PE 9(D)'M euesid eqay Aeg [4S %VC 9 %8 TC %CE GT 66/€0/T70 HZ7 Sl S9F'SE vuesid eqoy |, puryss.ieqnojlg € TLL /90/S0 AZe Sl S9F'SE esiodse xI[o}] pur.yss1aqnolg \) sjudpIsas sjjays [reus Ayduro sjjoys [reus Aydute s]Joys paT[y-pues ut pasn % 29 Ou ‘OU SIO} DaTJOO siaptdg ut Suysou seurueung ul Suysou oepriypesoypy aepryjAquiog aeprpisAry Suljsou ovuliesey| CI Tes OS ‘panuyuoy ‘xtpueddy 109 VOLUME 17, NUMBER 1, 2008 %e 6 %l € %OV C 94 (SD) M posioula G 39 (D)'M Lb S %E'TL EL 92 (D) M Ss9eq pur $}SOU 90d jo SUTPULOY] %oC C jsoU 9dq JO SUTLUTOY] “oP OOF vuesid eqouy, EL LP eyes, eroydoprdo.ry %0 C esiodse xI[op] %0 VP ‘eyesiy eroydoprdo.y, vyest] eroydoprido.ry, %6L1 IOL eye31] eroydopidoiy, %S OOT eyesty eroydoprido.y, yjoduayyey HLO'LC SPE EE YINOP AAT JOP] €/20/90 HLO'LZZ SPE'EE YINOPY JOATY JOR] ¢/Z0/90 Heo 9~ SLE'Se yuawdoyaAop Buisnoy Aq padoasep yeyIqey TOLA HPS “PHTY Od 886L SSO) pue SSO) Ul 6/6] URL ‘D OW “A payty Mod 7/40/90 HOP 9¢ Scr et 2aG-UO-UO}Udy] 1/20/90 HOP 9C SCP ct PIS-UO0-UO}USY| sjuaptsal s}foys [reus Aydura sjfeys [reus Ayduo stopidg ut 3uyjsou seurusumn | ul Busou seprryoesoyy s[ys pay[y-purs ut oepryAquiog aeprpisAry) SuIJsou ovuTIesey| posn % 29 ou CI 1194S ‘OU S10}99][OO OHS ‘penuyuoy ‘“xtpueddy J. HYM. RES. Vol. 17(1), 2008, pp. 110-115 Nest Structure, Seasonality, and Host Plants of Thygater aethiops (Hymenoptera: Apidae, Eucerini) in the Andes VICTOR H. GONZALEZ AND MONICA OSPINA (VHG) Department of Ecology and Evolutionary Biology, Haworth Hall, 1200 Sunnyside Avenue, University of Kansas, Lawrence, Kansas 66045-7523, USA; email: vhgonza@ku.edu (MO) Instituto de Investigacién de Recursos Bioldgicos Alexander von Humboldt, Villa de Leyva, Boyaca, Colombia; email: monikaospina@gmail.com Abstract.— Thygater is a Neotropical genus of 30 species of solitary bees occurring from Mexico to Argentina. Information on the nesting biology is only available for a single species, T. analis, from southern Brazil. We studied the nest architecture, seasonality, and host plants of T. aethiops in two localities in the Eastern Andes of Colombia (~ 2700-2900 m). Bees nested singly or forming aggregations in flat ground or sloping surfaces. Nests were deep (39-59 cm) and consisted of an unlined but smooth long main tunnel meandering downward from the surface. Nests had three to seven nearly vertical cells, which were located singly at the ends of short vertical tunnels that descended from the main tunnel. The older brood was located closer to the nest entrance, indicating a progressive nest development. Monthly sampling and appraisal of museum specimens showed that, although females were more commonly collected than males, both sexes were present year- round. We present a list of 32 plant species (27 genera in 18 families), including exotic, native, and cultivated plants used by T. aethiops in Colombia. The most noticeable differences from T. analis are in the pattern of nest development and seasonality. Thygater aethiops builds nests in a progressive manner and it is likely to be multivoltine, whereas T. analis has a regressive nest development (.e., older brood located far from the entrance) and a single generation per year. Long-term studies on different populations are required to determine if T. aethiops has continuous brood production or if larval diapause occurs at some point during the year. The purpose of this paper is to provide information on the nest architecture, sea- sonality and host plants of Thygater aethiops (Smith) at higher elevations in the Colom- bian Andes. Thygater Holmberg is a Neo- tropical genus of 30 species of solitary bees occurring from Mexico to Argentina (Ur- ban 1967, 1999). The biology of Thygater is still poorly known. Information on the nesting biology is only available for a single species, T. analis (Lepeletier), from southern Brazil (Michener and Lange 1958, Rozen 1974). Thygater aethiops is widely distributed in the Neotropical region, occurring from Costa Rica to Argentina, and is largely sympatric with T. analis in most of its range (Urban 1967). In Colom- bia, T. aethiops is found between 1400 and 35°90 m of elevation and is the most common species of the genus. Nests of this species are frequently found in grazing pastures, city parks, gardens, and along sidewalks in towns and densely populated cities such as Bogota (Gonzalez and Engel 2004, Gonzalez et al. 2005). We studied the nesting biology of T. aethiops in two localities in the Eastern Andes of Colombia (~ 2700-2900 m), and compared it with that of T. analis. We describe the nest and, based on monthly samplings and examination of museum specimens, we show that both sexes of T. aethiops are present year-round, visiting flowers of exotic, native, and cultivated plants. The nesting biologies of both species are very similar, except for the pattern of nest development and seasonal- ity. Thygater aethiops builds nests in a VOLUME 17, NUMBER 1, 2008 100% 90% 80% 70% 60% 50% 40% Percentage of individuals 30% 20% 10% Jan Feb Mar Apr May Jun Month Fig. 1. a 200 150 0 ei | ) Jul Aug Sep 1h! 29 a5 64 62 18 ry] 300 is 19) - Males [-_—] Females 250 =< Rainfall on (Umut) []esUTes [e}0], mahi jo) oO Oct Nov Dec Seasonal collections of Thygater aethiops at high altitudes (2700-2900 m) in the Andes of Colombia. Monthly surveys were done in 1999 in La Calera (Departamento of Cundinamarca), except in January, April, May, June and December. To complete the seasonal cycle, we used the number of museum specimens collected in La Calera and contiguous areas during those months. Females were more commonly collected than males (X? 001 [22] = 534.1, P < 0.001) every month, and the female/male ratios were not homogeneous among months (X? 05 [11] = 28.9, P < 0.001). The total number of individuals is indicated at the top of each column. progressive manner (i.e., older brood lo- cated near the entrance, thus first to be completed), and is likely to be multivoltine, whereas T. analis has regressive nest development and a single generation per year. However, long-term studies on dif- ferent populations are required to deter- mine if T. aethiops has continuous brood production or if larval diapause occurs at some point during the year, and to detect any variation in nest development depend- ing on soil conditions. MATERIAL AND METHODS Study sites and nest excavations The nesting biology of T. aethiops was studied by VG during December 2004 in Mondonedo (2720m, 4°39'52.9"N, 74°17'2"W), whereas the seasonality and host plants were studied in 1999 by MO in La Calera (2900 m, 4°43'22”N, 73°58'18’"W). Mondonedo is a semiarid area highly disturbed by cattle ranching, gravel extrac- tion, and waste dumping whose soils are very compacted, shallow, and strongly eroded (details in Gonzalez and Chavez 2004). The soils in La Calera are looser and less eroded than in Mondonedo, and primarily used to grow potato, corn, beans, and wheat (IGAC 1996). Annual rainfall in La Calera is bimodal (Fig. 1), with higher monthly precipitation than in Mondonedo. We excavated the nests using a geological pick, hand shovel, wood chisels and a pocketknife. Prior to excavation, and in order to trace the nest structure, we used a syringe to inject into the tunnel a slurry of plaster-of- Paris and water, which was allowed to harden for about 2—5 minutes; we then excavated the nests. We measured nest features in the field using a caliper. Maxi- mum nest depth was measured from the nest entrance to the bottom of the deepest cell. 112 Seasonality We collected bees on flowers and exam- ined museum specimens to determine the seasonal cycle of T. aethiops in La Calera. Bees were collected when they appeared to be more active, frequently between 8:00 and 11:00 am, and from one to three days each month depending on weather condi- tions. Once we identified and sexed them, bees were released. Thygater aethiops is the only species of the genus occurring in La Calera. However, we decided to capture them because sometimes it was difficult for us to distinguish T. aethiops females from small, black workers of the bumble bee Bombus atratus Franklin (Apidae, Bombini). Males and females of T. aethiops were easy to distinguish in the field because, as in other eucerines, males have longer anten- nae that surpass the base of the abdomen. The monthly surveys in La Calera were done during the year of 1999, except in January, April, May, June, and December. To complete the seasonal cycle, we used the number of museum specimens collect- ed in La Calera and contiguous areas during those months. We used a G-test (Sokal and Rohlf 2000) to compare percent- ages of males and females of T. aethiops, and to determine if female/male (F/M) ratios were homogeneous among months. Host plants We recorded the plants visited by T. aethiops in La Calera, and also extracted floral records from data from specimen labels. We examined about 300 specimens of T. aethiops deposited in the Snow Entomological Museum, University of Kansas, USA (SEM), National Museum of Natural History, Washington, D.C. (USNM), and the following Colombian institutions: Laboratorio de Investigaciones en Abejas, Departamento de Biologia (LA- BUN), and the Museo de Historia Natural, Instituto de Ciencias Naturales (ICN) in Bogota, Instituto de Investigaci6n de Re- cursos Biolégicos Alexander von Hum- JOURNAL OF HYMENOPTERA RESEARCH Table 1. Measurements (cm) of some structures of eight nests of Thygater aethiops. X: mean value, + : standard error, R : range, N : sample size. Samples varied in number because, to increase variation, we measured remains of nest structures (e.g., cells, tunnels) that we found during excavations. Nest structure pi R N Nest density (nests/m’) 3.3-+:0.6 , DOA 6 Inter-nest distance 27.5+7.5 4.0-108 13 Maximum nest depth 43.8 + 3.1 36-59 | Tumulus Length 8.3 +10 ~G0-10 4 Width 9340.9 8.0-12 4 Nest entrance 0.75 + 0.1 0.68-0.80 17 Main tunnel Length 64 + 10 53-94 4 Diameter 0.85 + 0.1 0.80-0.93 16 Laterals Length 2.20 + 0.2 1.08-3.30 13 Diameter 0.76 + 0.2 0.67-0.80 6 Cell depth 36 = 2.6 “DIES ts Number of cells pernest 5.0 + 0.7 3-7 8 Cell dimensions Length 189 = 0.6 “= 0s Diameter 0.88 + 0.2 0.83-0.95 5 boldt ([AvH), Villa de Leyva, Boyaca, and the Museo Entomoldgico Francisco Luis Gallego, Universidad Nacional de Colom- bia in Medellin (MEFLG). RESULTS Nest architecture Nest measurements are given in Table 1. Nests were found singly or forming aggre- gations in flat ground or sloping surfaces. The nest sites were either barren or sparsely covered with grasses in areas frequently well exposed to the sun (Figs 2, 3). Nest entrances were circular and when nests were active, irregular tumuli extended downward from the nest entrances. Nests were deep, and consisted of an unlined but smooth long main tunnel meandering downward from the surface (Fig. 4). Cells were nearly vertical and placed singly at the ends of short laterals, or vertical tunnels that descended from the main tunnel. Once the cells were completed, these vertical tunnels were filled with loose, coarse soil (Fig. 5). VOLUME 17, NUMBER 1, 2008 igs 2). Nests of Thygater aethiops. 2, single nest in flat ground; note the tumulus around nest entrance. 3, cluster of nests in a barren, sloping surface; all nest entrances are encircled and one of them is indicated by an arrow. 4, an excavated nest showing the meandering main tunnel, indicated by a white solid line. 5, detailed view of a cell in saggital section; A, main tunnel; B, short laterals filled with loose soil after cell completion; C, whitish spiderweb- like layer, likely a mass of mold hyphae, found in most of excavated cells (nest measurements in Table 1). The cells were excavated in the soil and were not separable from the substrate. The three to seven cells per nest were con- structed at different depths. The cells were elongated, with a concave spiral closure inside, slightly wider at the bottom than top, and lined with a ‘‘wax-like”’ secreted material resembling those of other eucerine bees. In most cells, the upper one third of the inner wall was covered with a thin, whitish spiderweb-like layer (Fig. 5). We did not examine it under the microscope, but it might be a mass of mold hyphae. As described for Thygater (Rozen 1974, Packer 1987), larval feces were incorporated with- in the dark brown, paper-like cocoon. The basal one third of the cell contained liquid provisions with the egg floating on the surface. Cell contents ranged from eggs to small larvae and pollen. Older brood was located closer to the nest entrance, indicat- ing progressive nest development. Numer- ous old burrows and remains of cocoons from older generations were commonly found during excavations. Associated organisms We found empty puparia of an unknown fly inside three sealed cells that had a perforation (2.0-2.6 mm in diameter) in the cell closure. Each cell had from two to six puparia, about 3.8mm long and 1.5 mm wide, among debris. Seasonal cycle The monthly sampling in La Calera and appraisal of museum specimens revealed that both sexes are present on every month 114 of the year (Fig. 1). Moreover, females were more commonly collected than males (X? 001 [pT 534.1, P= 0.001), and the F/ M ratios were not homogeneous among months (X* 95 [11] = 28.9, P < 0.001). F/M ranged from about 2:1 in January to 31:1 in October (Fig. 1). Host plants Both males and females were recorded from 32 plants species (27 genera in 18 families), including exotic, native, and cultivated plants. Females were recorded from the most plants. Records for males are indicated with an asterisk. AGAPANTHA- CEAE: Crinum africanum L’Heéritier de Bru- telle*. ASTERACEAE: Baccharis sp., Senecio sp., Vernonia canescens Kunth. BALSAMINA- CEAE: Impatiens balsamina Linnaeus. BRASSI- CACEAE: Brassica napus Linnaeus’, B. nigra (Linnaeus)*, Raphanus raphanistrum Lin- naeus*. CAESALPINACEAE: Cassia tomentosa Linnaeus. CONVOLVULACEAE: Convolovulus sp., Ipomoea congesta R. Br. CUCURBITACEAE: Cyclanthera pedata Scrad, Sechium edule Swartz. FABACEAE: Desmodium uncinatum (Jacquin), Phaseolus vulgaris Linnaeus*, T7i- folium pratense Linnaeus*, T. repens Lin- naeus*. LAMIACEAE: Salvia cuatrecasana Epling*, S. bogotensis Benth*. MALVACEAE: Hibiscus grandiflorus Juss. Ex Candolle. MELASTOMATACEAE: Tibouchina sp. PAPA- VERACEAE: Papaver somniferum Linnaeus*. PASSIFLORACEAE: Passiflora mollissima L.H. Bailey. ROSACEAE: Rubus sp*. RUBIACEAE: Coffea arabica Benth, Palicourea sp. RUTA- CEAE: Citrus aurantium Linnaeus. SOLANA- CEAE: Cyphomandra betacea (Cav.) Sendtn, Solanum tuberosum Poepp ex Walp%, S. lycioides Linneaus. Solanum sp. TROPAEO- LACEAE: Tropaeolum majus Linnaeus*. DISCUSSION The nest architecture of T. aethiops is very similar to that described for T. analis, the only species of Thygater whose nests have been studied (Michener and Lange 1958, Rozen 1974). Both species nest solitarily or in aggregations in flat or sloping ground. JOURNAL OF HYMENOPTERA RESEARCH Their nests consist of a long, unlined, main tunnel from which short laterals, ending in a single vertical elongated cell, branch off; these laterals are filled with soil once cells are completed. The most noticeable differ- ence from T. analis is the pattern of nest development. We inferred from the distri- bution of young and old brood in the excavated nests, that T. aethiops builds nests in a progressive manner (i.e., older brood located near the entrance, thus, first to be completed) whereas it is regressive in T. analis (older brood located far from the entrance). However, we examined nests of T. aethiops from a single population and only during the dry season, so we do not know if the nest development pattern varies among localities, with time of year, or with soil hardness. The monthly surveys in La Calera and the appraisal of museum specimens showed that males and females are present throughout the year. In La Calera, females were more frequently collected than males, and the F/M ratios varied significantly among months (Fig. 1). We did not deter- mine adult longevity but unless adults have an extraordinary long life, the pres- ence of both sexes indicates that T. aethiops must be multivoltine. In contrast, due to seasonality in southern Brazil, T. analis over-winters as postdefecating larvae and emerge as adults during the summer. Thygater analis is the most widespread species of the genus, occurring from tropical to subtropical lowlands to high altitudes in the Andes (Urban 1967) and thus could have multiple generations per year in more tropical environments as does T. aethiops. However, we do not know if T. aethiops has continuous brood production or if larval diapause occurs at some point during the year, as it has been observed in other Andean solitary bees, such as Antho- phora walteri Gonzalez (Apidae, Antho- phorini) (Gonzalez and Chavez 2004). Thygater aethiops seems to use a wide range of exotic, native, and cultivated plants for pollen and nectar (e.g., Impatiens, VOLUME 17, NUMBER 1, 2008 Salvia spp, and Solanum tuberosum, respec- tively). Our records agree with the appar- ent polylectic diet of Thygater, as inferred by Urban (1967) based on scarce floral records. Also, as previously noted for some Thygater species (Urban 1967), females of T. aethiops buzz pollinate flowers with por- icidal anthers (e.g., Solanum). It might be interesting to explore the use of T. aethiops as a crop pollinator of tomatoes (Solanum lycopersicum Linnaeus) or Inca berries (Physalis peruviana Linnaeus), both Solana- ceae species with poricidal anthers that are extensively cultivated in the Andean re- gion. Undoubtedly, research on adult longevity, brood production, and other biological aspects of T. aethiops are needed. ACKNOWLEDGMENTS We are greatly indebted to each of the individuals who allowed us to visit their insect collections; M. S. Engel, Z. Falin (SEM), D. Furth (USNM), G. Nates (LABUN), F. Fernandez, G. Amat, E. Florez (ICN), J. E. Castillo (IAvH), and A. Smith, J. Quiroz (MEFLG). We also thank E. Palacio and B. Mantilla for their help in the field; P. Sepulveda, B. Alexander, C. Rasmussen, G. Broad, J. Neff, and two anonymous reviewers provided valuable comments on the manuscript. The University of Kansas (KU), Undergraduate Program in Biology, Department of Ecology and Evolutionary Biology, KU General Research Fund and US-Israel Binational Science Foundation grant 2000-259 (to D. Smith and Y. Lubin) provided financial support for VG through teaching assistantships and laboratory facilities. NS LITERATURE CITED Gonzalez, V. H. and F. Chavez. 2004. Nesting biology of a new high Andean bee, Anthophora walteri Gonza- lez (Hymenoptera: Apidae: Anthophorini). Journal of the Kansas Entomological Society 77: 584-592. and M. S. Engel. 2004. The Tropical Andean bee fauna (Insecta: Hymenoptera: Apoidea), with examples from Colombia. Entomologische Abhan- dlungen 62: 65-75. , M. Ospina, and D. Bennett. 2005. Abejas altoandinas de Colombia: Guia de campo. Instituto de Investigacion de Recursos Bioldgicos Alexan- der von Humboldt, Bogota, D.C., Colombia. 80 pp. IGAC. 1996, Diccionario Geografico de Colombia. Insti- tuto Geografico Agustin Codazzi, Bogota, Colom- bia, 2504 pp. Michener, C. D. and R. B. Lange. 1958. Observations on the ethology of Neotropical Anthophorine bees (Hymenoptera: Apoidea). The University of Kansas Science Bulletin 39: 69-96. Packer, L. 1987. A description of the mature larvae and cocoon of the bee Thygater (Hymenoptera: Anthophoridae). Journal of the New York Entomo- logical Society 95: 23-27. Rozen, J. G. 1974. Nesting biology of the Eucerini bee Thygater analis (Hymenoptera: Anthophoridae). Journal of the New York Entomological Society 82: 230-234. Sokal, R. R. and F. J. Rohlf. 2000. Biometry. The principles and practice of statistics in biological research. Third Edition. W. H. Freeman and Company, New York, USA, xix + 887 pp. Urban, D. 1967. As espécies do género Thygater Holmberg, 1884. Boletim da Universidade Federal do Parana (Zoologia) 2: 177-307. . 1999, Espécies novas de Thygater Holmberg (Hymenoptera, Apoidea, Anthophoridae). Revista Brasileira de Zoologia 16: 213-220. J. HYM. RES. Vol. 17(1), 2008, pp. 116-123 A Comparison of the Parasitic Wasps (Hymenoptera) at Elevated Versus Ground Yellow Pan Traps in a Beech-Maple Forest THOMAS PUCCI Department of Invertebrate Zoology, Cleveland Museum of Natural History, 1 Wade Oval Dr. Cleveland, Ohio, 44106, USA; email: tpucci@cmnh.org Abstract.— The abundance, diversity, and morphospecies composition of the parasitic wasp fauna is compared at two levels in an Ohio (USA) temperate forest. The ground layer and the elevated layer (~10m) exhibited similar abundance but a distinctly different composition. The diversity at the ground layer was greater. Encyrtidae were captured more often in elevated traps while Pteromalidae, Ichneumonidae, and alysiine Braconidae were more prevalent at ground level. A retrospective analysis of the edge effect showed no difference in the composition relative to distance from forest edge but elevated samples had a higher diversity near the edge than in the interior. A key goal in ecology is to describe the distribution of species within the environ- ment. However, difficulty of canopy access has limited our knowledge of forest com- munities (Ozanne et al. 2003). A better understanding of arthropod forest stratifi- cation is needed to identify which taxa are dependent on elevated strata and under what conditions do distinct stratum com- munities develop. This will direct more efficient and thorough surveys of forest life. There are many accounts of arthropod stratification in tropical forests. Many document increased abundance (e.g. Kato et al. 1995, Barrios 2003, Sutton and Hudson 1980) and diversity (e.g. Basset 2001a, Basset et al. 2001, Rees 1983, Yanouvial and Kaspari 2000) in elevated layers. Contradictory results however are evident. For example, Stork and Grimba- cher (2006) found a similar abundance and richness of beetles between ground and elevated layers. Abundant arthropod stra- tum specialists have also been observed (Schulze et al. 2001, Hammond et al. 1997, Sorensen 2003). Temperate forests are also stratified but patterns are similarly not well established. The stratum of peak abundance has been shown to be variable: elevated (Hollier and Belshaw 1993); near ground (Nielsen 1987), or comparable (Le Corff and Marquis 1999, Preisser et al. 1998, Ulyshen and Hanula 2007). In several studies diversity has been shown to be greater near ground level (Lowman et al. 1993, Le Corff and Marquis 1999, Ulyshen and Hanula 2007). Distinct stratum communities were revealed in the preceding and by Winchester and Ring (1996), Gibson (1947) and Hollier and Belshaw (1993). But Fowler (1985) ob- served little stratification in richness or species composition of herbivores on birch branches. Examples of stratification involv- ing a limited number of species include Munster-Swendsen (1980) and Henry and Adkins (1975). Although there has been pronounced growth in canopy research in recent de- cades (Basset 2001b), basic questions re- garding forest stratification remain. I am not aware of another temperate zone species-level survey that addresses the complete parasitic wasp fauna at ground versus elevated strata. The purpose of this study is to compare the composition of adult parasitic wasps at two levels of a temperate forest. VOLUME 17, NUMBER 1, 2008 METHODS The study site is located within the Soubusta Sugarbush Preserve (041°34'24” N 081°14'04” W) in northeastern Ohio, USA. Canopy trees are dominated by sugar maple (Acer saccharum Marshall) and American beech (Fagus grandifolia Ehrh.). Tulip tree (Liriodendron tulipifera L.) is also common. Ground cover is dominated by sugar maple seedlings, mayapple (Podo- phyllum peltatum L.), white trillium (Trilli- um grandiflorum Salisb.), and white ash (Fraxinus americana L.) seedlings. The site was deforested in the 19" century and has been harvested for maple syrup since the 1940s but apparently no tree thinning has been practiced (K. Vouk, pers. comm.). The study site is within a forested lot which is approximately 85 hectares and connected to other lots by a corridor. Traps were made from plastic containers spray-painted fluorescent yellow. The di- mensions (L-W-D) were 201510 cm. A bow and arrow were used to string a line for eight traps hung over sugar maple limbs. Trap placement was dictated by the presence of an accessible tree limb under the closed canopy. Elevated trap placement ranged from 6.7 to 11.9 meters (mean 9.8m) high. A ground level trap was placed directly under each elevated trap. Rope was attached to the trap and fastened to a nearby tree to mimic the elevated trap set-up. All traps were filled half way with water and a few drops of unscented detergent. Traps were serviced every two days from June 3° to July 1* 2005. Specimens of parasitic wasps (including all Cynipoi- dea; excluding Dryinidae) were pointed and sorted to morphospecies. Occasionally, within a single pan sample, only represen- tative specimens of a common morphospe- cies were pointed. Inherent in the concept of morphospecies is the likelihood of misclassification. Approximately 60 hours were spent sorting specimens in an effort to minimize this occurrence. Seventeen ny specimens are unassigned to morphospe- cles OWing to damage or uncertainties regarding sexual dimorphism. Specimens were taken at least to family level using Goulet and Huber (1993). Braconid speci- mens were taken to subfamily using Wharton et al. (1997). Representative spec- imens are housed at the Cleveland Muse- um of Natural History. Paired ¢ tests were used to compare various components of each stratum. The Mann-Whitney Rank Sum Test was used when the data were not normally distrib- uted. The Simpson Index of diversity was used as recommended by Magurran (2004). The Morisita-Horn index was used to measure the faunal similarity between groups. It uses abundance data and is preferable because it is not dependent on sample size (Krebs 1999). All traps were within 250 meters of each other. Two pairs of traps (sites #1 and #2) were placed approximately 25 m from an old field while the remaining were situated 100 + 10 meters from the forest edge. Although this set-up was not designed to address any edge effects, a retrospective comparison of these two groups was carried out. To examine the faunal compo- sition of the sites, a cluster analysis using MVSP version 3.130 (Kovach 2005), the UPGMA clustering method, and the mod- ified Morisita’s coefficient of similarity was executed. Also, the Morisita-Horn index was compiled for each pair of traps per stratum. Trap pairs sited a similar distance from the edge were compared with those sites at differing distances by means of a t test. RESULTS A total of 2,541 specimens and 269 morphospecies were collected. The braco- nid, Eubazus pallipes Nees, represents 36.4 % of the total catch. Ninety-five percent of the specimens (878 specimens) were from a single elevated trap taken during five consecutive collecting dates. Therefore, E. pallipes is not included in the following 118 Table 1. Abundance and diversity of each stratum. Abundance Elevated * 703 Ground * 837 Mean (SD) ** elev. 6.38 (2.84) gr. 7.61 (3.21) t test (N = 30) ** t = —1.114 P = 0.275 *data deleted to yield equal collection effort (216 trap **based on raw data analysis where it would skew the results. Traps were occasionally disturbed, result- ing in unequal collection effort between the strata. This was corrected for by deleting the data from each pair of traps when one was disturbed in comparisons that do not use the mean. The abundance of parasitic wasps was similar between strata (Table 1). The .di- versity was significantly higher at ground level based on the Simpson Index but marginally insignificant based on species richness (Table 1). Based on equal collec- tion effort, there were 130 singletons: 54 from elevated traps and 76 at ground level. This does not represent a significant difference ({ = —1.25, P = 0.223). The inter-strata similarity is low, 0.353, compared to the mean (SD) of the within stratum similarity: elevated 0.708 (0.088), UPGMA GROUND ELEVATED JOURNAL OF HYMENOPTERA RESEARCH Number of species Simpson index (1/D) 147 20.18 183 41.79 elev. 4.92 (1.34) elev. 0.915 (0.041) gr. 6.35 (2.47) gr. 0.941 (0.025) t = —1.964 P = 0.059 t = —2.164 P = 0.039 days per stratum) ground 0.586 (0.084). Similarly, Figure 1 shows the composition of the strata were distinct. Table 2 lists the taxa with at least ten specimens. Six of the 14 families and 21 of the 45 morphospecies were significantly more abundant in a particular stratum. Excluding Eubazus pallipes (from the trap with the exceptional catch), the abundance of the 21 “stratum specialists” represented 67% of the ‘““common morphospecies”’ listed in Table 2. This proportion rises to 71% if the morphospecies at P = 0.08 are considered stratum specialists. The influence of the edge effect on diversity is shown in Table 3. There was a significantly higher diversity in the elevat- ed traps near the edge but no significant difference at ground level. Table 4 and Figure 1 show no edge effect in the species composition for either stratum. S—H-NONWOAKhRADHAS|HNWOADAN YN Modified Morisita’s Similarity Fig. 1. Cluster analysis of the abundance based species composition for each trap. Sites one and two were approximately 25 meters from the forest edge while the remaining were approximately 100 meters away. VOLUME 17, NUMBER 1, 2008 DISCUSSION The parasitic wasp fauna differed be- tween strata at the family level (individuals per family) and the morphospecies level. Although elevated traps were in the lower reaches of the sugar maple canopy and the forest floor contained abundant sugar maple seedlings, numerous taxa were found in significantly higher numbers in the elevated traps. Likewise, some mor- phospecies exhibited a preference for the ground level. My results are consistent with Le Corff and Marquis (1999) and Ulyshen and Hanula (2007) who also found distinctly different insect communi- ties at elevated and near ground strata in temperate habitats. Additionally, they also found the near ground stratum was more diverse. These results are consistent with Lowman et al. (1993) who hypothesized that the stratum of peak diversity will generally be near the ground in temperate forests and in the canopy in tropical forests, coinciding with the stratum with the most niches. Further evidence includes Leksono et al. (2005) who used both yellow and blue pan traps to survey attelabid and cantharid (Coleoptera) stratification in a mixed decid- uous forest in Japan. Although common species were more abundant at the highest (20m) layer, rare species were found only at the lower layers (0.5m and 10m). In contrast, the canopy is often shown to be more diverse in tropical forests (e.g. Basset 2001a, Rees 1983, Yanouvial and Kaspari 2000). The particular sampling regime in this study precludes examination of the influ- ence of some known factors that affect insect collections. Seasonality has been shown to influence guild structure (Askew and Shaw 1979, Sheehan 1994). Ulyshen and Hanula (2007) found the ground: canopy ratio of abundance and richness is liable to change significantly throughout the collecting season. Ozanne (1999) found guild structure changes with tree species. 119 Davis and Sutton (1998) found forest type influences the vertical distribution of cer- tain dung beetles. Also, it is well known that collecting method influences the com- position of parasitic wasp samples (Noyes 1989, Idris et al. 2001). Perhaps because yellow pan traps are attractive the focus on a single tree species is not necessary to collect comparable samples in the canopy but this has not been established. The edge effect has been shown to change the species composition and in- crease the abundance of insects (e.g. Noyes 1989, Foggo et al. 2001). The lack of a decipherable edge community in this study could be due to the rather distant (25 m) location of the “edge” sites. Alternatively, the gradient may be much weaker. For example, Dangerfield et al. (2003) found habitat specialists can be prevalent for hundreds of meters beyond a discrete riparian/treeless saltbush edge. In this study, elevated edge sites displayed a greater diversity than the elevated interior sites but there was no significant difference at ground level. These results are prelim- inary due to the lack of replication of edge sites, but suggest that further work into the edge/canopy effect on parasitic wasps may be fruitful for building our knowledge of wasp biology and forest ecology. The stratification displayed in Fig. 1 and Table 2 probably largely reflects host distri- bution. For example, alysiine braconids and the majority of diapriids attack immature Diptera which predominate in the soil layers of temperate forests (Schaefer 1991). Additionally, others have found certain groups of egg parasitoid abundance to be greater in elevated layers (Compton et al. 2000, Basset et al. 2001, Noyes 1989). Mymaridae however has been shown to be low fliers (Compton et al. 2000, Noyes 1989). Compton et al. (2000) demonstrated various chalcid families (esp. fig wasps) can be found above the canopy where there is stronger wind. A strong case is made that their presence at that height is associated with dispersal. Perhaps some parasitic 120 JOURNAL OF HYMENOPTERA RESEARCH Table 2. List of taxa with at least ten specimens and a comparison of the vertical distribution. Taxa Ceraphronoidea Ceraphronidae Ht] #2 #3 Chalcidoidea Encyrtidae +] #2 #3 #4 #5 #6 Eulophidae 4] #2 5 Mymaridae Stephanodes sp. #1 #2 Pteromalidae #1 Cynipoidea Charipidae #1 Cynipidae #1 Eucoilidae Evanioidea Aulacidae Ichneumonoidea Braconidae Alysiinae Aphaereta pallipes (Say) Dinotrema sp. Aphidiinae Trioxys sp. Doryctinae Spathius elegans Matt. Euphorinae Helconinae Eubazus pallipes Nees Ichneumonidae #1 #2 #3 #4 #5 Platygastroidea Platygastridae 4] #2 #3 Elevated Abundance 238 trap days Ground Abundance 218 trap days he mB woOooworooqo ao nN N 79 ft test (t) or Mann-Whitney Rank Sum test (T) t = —0.305 P = 0.763 T = 269 P = 0.134 T = 249 P = 0.520 T = 173 P =0.015 = —5.30 P = < 0.001 T = 208 P = 0.316 = 199 P = 0.170 T = 238 P = 0.835 T = 158 P = 0.002 T = 150 P = < 0.001 T =177 P=0.021 t = —1.75 P = 0.092 T = 173 P = 0.130 T = 280 P = 0.05 T = 202 P = 0.202 T = 209 P = 0.330 T = 287 P = 0.025 t = —2.98 P = 0.006 T = 220 P = 0.602 t = 2.35 P = 0.026 T = 281 P = 0.047 T = 208 P = 0.316 T = 208 P = 0.316 T = 275 P = 0.081 T = 259 P = 0.288 T = 288 P = 0.022 T = 188 P = 0.063 T = 232 P = 0.983 T = 326 P = < 0.001 T = 272 P = 0.103 T = 276 P = 0.073 T = 214 P = 0.441 T = 214 P = 0.453 T = 277 P = 0.071 T = 303 P = 0.004 T = 202 P = 0.203 T = 154 P = 0.001 T = 169 P = 0.009 T = 321 P=< 0.001 T = 257 P = 0.315 T = 255 P = 0.357 T = 293 P = 0.013 T = 272 P = 0.109 T = 275 P = 0.080 T = 216 P = 0.507 T =172 P= 0.012 T = 240 P = 0.787 T = 241 P = 0.738 VOLUME 17, NUMBER 1, 2008 Table 2. Continued. Elevated Abundance Taxa 238 trap days #4 7 #5 1? #6 Ds #7 3 Scelionidae . 40 Telenomus sp. 20 Trissolcus sp. 11 Proctotrupoidea Diapriidae 22. Basalys sp. 5 Paramesius sp. 0 Spilomicrus sp. 1 Trichopria sp. 2 #1 2 #2 5 #3 0 #4 0 wasps in temperate forests use elevated layers for dispersal. Evidence for this include Karem et al. (2006) who found the abundance of small wasps (many Chalci- doidea, Cynipoidea, and Proctotrupoidea) was similarly stratified along forest/field transects so that the wasps were flying in the canopy in addition to well above field vegetation. Nielsen (1987) found the great- est wind speeds just below the canopy (10m) in a fully leaved beech stand. Using light traps, it was shown that insects in general (including Hymenoptera) avoided this layer compared to traps at 0.6 m and 21m. Unfortunately, the Hymenoptera composition per level was not documented. Similar to my results, other studies that have compared canopy versus near ground Table 3. Edge effect and diversity for each stratum. Simpson Index Distance from edge N Mean (SD) t test Elevated ~25 m Bae MOA ESO) aks == 209 1 ~100 m 6 10.81 (2.74) P=0.027 Ground — 25m yD SOU 2on(t22 53) oe — 122 ~100 m 6 37.50 (5.72) P=0.270 jb | Ground Abundance 218 trap days t test (t) or Mann-Whitney Rank Sum test (T) 15 T = 263 P = 0.219 13 T = 233 P = 1.0 10 T = 275 P = 0.080 7, T = 299 P = 0.006 8 T = 164 P = 0.004 3 T = 171 P = 0.010 0 T = 195 P = 0.121 244 T = 341 P= < 0.001 22 T = 293 P = 0.013 17 T = 285 P = 0.030 10 T = 249 P = 0.502 38 T = 326 P = < 0.001 87 T = 335 P = < 0.001 is) T = 283 P = 0.039 11 T = 285 P = 0.030 14 T = 300 P = 0.005 level insects have found approximately half of the common species to be associated with a particular stratum, about a quarter at each level (Stork and Grimbacher 2006, Broadhead 1983). Likewise, Ulyshen and Hanula (2007) report that the proportion of beetle species captured exclusively at a particular stratum to be approximately 30 % at each layer. Although most of the taxa with at least ten specimens are found at both strata in my study, it has been shown that the fauna collected at only one layer gives a biased representation of the com- position of parasitic wasps. Table 4. Comparison of faunal similarity and edge effect for each stratum. Morisita-Horm Trap Location N Mean (SD) t Test Elevated similar distance from edge 16 .710 (.092) t = .110 P = 918 dissimilar distance from edge 12 .706 (.090) Ground similar distance from edge 16 .587 (.085) f= 22 P = .904 dissimilar distance from edge 12 .583 (.087) 122 ACKNOWLEDGEMENTS I would like to thank Michael Ulyshen and James Hanula for sharing their unpublished results, Joe Keiper for his review of a draft and assistance with the statistical analysis, James Bissell and Keith Vouk for access to the study site, [lari Saaksjarvi, Mark Shaw, and an anonymous reviewer for their helpful evaluations, and Hans Clebsch for identification of the Diapriidae. LITERATURE CITED Askew, R. R. and M. R. Shaw. 1979. Mortality factors affecting the leaf-mining stages of Phyllonorycter (Lepidoptera: Gracillariidae) on oak and birch. 2. Biology of the parasite species. Zoological Journal of the Linnean Society 67: 51-64. Basset, Y. 2001a. Communities of insect herbivores foraging on saplings versus mature trees of Pourouma bicolor (Cecropiaceae) in Panama. Oeco- logia 129: 253-260. . 2001b. Invertebrates in the canopy of tropical rain forests. How much do we really know? Plant Ecology 153: 87-107. , H. P. Aberlenc, H. Barrios, G. Curletti, J. M. Berenger, J. P. Vesco, P. Causse, A. Haug, A. S. Hennion, L. Lesobre, F. Marques, and R. O’Meara. 2001. Stratification and diel activity of arthropods in a lowland rainforest in Gabon. Biological Journal of the Linnean Society 72: 585-607. Barrios, H. 2003. Insect herbivores feeding on conspe- cific seedlings and trees. Pp. 282-290 in: Basset, Y., V. Novotny, S. E. Miller, and R. L. Kitching, eds. Arthropods of Tropical Forests: Spatio-temporal Dynamics and resource use in the Canopy. Cam- bridge University Press, Cambridge. Broadhead, E. 1983. The assessment of faunal diversity and guild size in tropical forests with particular reference to the Psocoptera. Pp. 107-119 in: Sutton, S. L., T. C. Whitmore, and A. C. Chadwick, eds. Tropical Rain Forests: Ecology and Management. Blackwell Scientific Publications, Oxford. Compton, S. G., M. D. F. Ellwood, A. J. Davis, and K. Welch. 2000. The flight heights of chacid wasps (Hymenoptera, Chalcidoidea) in a lowland Bor- nean rain forest: Fig wasps are the high fliers. Biotropica 32: 515-522. Dangerfield, J. M., A. J. Pik, D. Britton, A. Holmes, M. Gillings, I. Oliver, D. Briscoe, and A. J. Beattie. 2003. Patterns of invertebrate biodiversity across a natural edge. Austral Ecology 28: 227-236. Davis, A. J. and S. L. Sutton. 1998. The effects of rainforest canopy loss on arboreal dung beetles in Borneo: implications for the measurement of biodiversity in derived tropical ecosystems. Di- versity and Distributions 4: 167-173. Gibson, W. W. 1947. An ecological study of the spiders of a river-terrace forest in western Tennessee. Ohio Journal of Science 47: 38-44. JOURNAL OF HYMENOPTERA RESEARCH Goulet, H. and J. T. Huber, eds. 1993 Hymenoptera of the world: an identification guide to families. Re- search Branch, Agriculture Canada, 668 pp. Foggo, A., C. M. P. Ozanne, M. R. Spleight, and C. Hambler. 2001. Edge effects and tropical forest canopy invertebrates. Plant Ecology 153: 347-359. Fowler, S. V. 1985. Differences in insect species richness and faunal composition of birch seed- lings, saplings and trees: the importance of plant architecture. Ecological Entomology 10: 159-169. Hammond, P. M., N. E. Stork, and M. J. D. Brendell. 1997. Tree-crown beetles in context: a comparison of canopy and other ecotone assemblages in a lowland tropical forest in Sulawesi. Pp. 184-223 in: Stork, N. E., J. Adis, and R. K. Didham, eds. Canopy Arthropods. Chapman & Hall, London. Henry, L. G. and T. R. Adkins’ 1975." Vermear distribution of biting midges in coastal South Carolina. Annals of the Entomological Society of America 68: 321-324. Hollier, J. A. and R. D. Belshaw. 1993. Stratification and phenology of a woodland Neuropteran assemblage. The Entomologist 112: 169-175. Idris, A. B., N. N. Zaneedarwaty, A. D. Gonzaga, M. I. Zaidi, S. Azman, and Y. Salmah. 2001. A study on four methods of sampling Ichneumonidae and Braconidae at two different habitats of Fraser’s Hill, Malaysia. Pakistan Journal of Biological Sci- ences 4: 1515-1517. Karem, J., S. A. Woods, F. Drummond, and C. Stubbs. 2006. Sampling native wasps along both vertical and horizontal gradients in the Maine Lowbrush Blueberry landscape. Environmental Entomology 35: 1083-1093. Kato, M., T. Inoue, A. A. Hamid, T. Nagamitsu, M. B. Merdek, A. R. Nona, T. Itino, S. Yamane, and T. Yumoto. 1995. Seasonality and vertical structure of light-attracted insect communities in a dip- terocarp forest in Sarawak. Research in Population Ecology 37: 59-79. Kovach, W. L. 2005. MVSP- A Multivariate Statistical Package for Windows, ver. 3.1. Kovach Computing Services, Pentraeth, Wales, U.K. Krebs, C. J. 1999. Ecological Methodology. Addison- Welsey Educational Publishers Inc. Menlo Park, California. 581 pp. Le Corff, J. and R. J. Marquis. 1999. Difference between understory and canopy in herbivore community composition and leaf quality for two oak species in Missouri. Ecological Entomology 24: 46-58. Leksono, A. S., N. Nakagoshi, K. Takada, and K. Nakamura. 2005. Vertical and seasonal variation in the abundance and the species richness of Attelabidae and Cantharidae (Coleoptera) in a suburban mixed forest. Entomological Science 8: 235-243. Lowman, M., P. Taylor, and N. Block. 1993. Vertical stratification of small mammals and insects in the VOLUME 17, NUMBER 1, 2008 canopy of a temperate deciduous forest: A reversal of tropical forest distribution? Selbyana 14: 25. Magurran, A. E. 2004. Measuring Biological Diversity. Blackwell Publishing, Malden, Massachusetts. 256 pp. Munster-Swendsen, M. 1980. The distribution in time and space of parasitism in Epinotia tedella (Cl.) (Lepidoptera: Tortricidae) Ecological Entomology 5: S703): Nielsen, B. O. 1987. Vertical distribution of insect populations in the free air space of beech woodland. Entomologiske Meddelelser 54: 169-178. Noyes, J. S. 1989. A study of five methods of sampling Hymenoptera (Insecta) in a tropical rainforest, with special reference to the Parasitica. Journal of Natural History 23: 285-298. Ozanne, C. M. P. 1999. A comparison of the canopy arthropod communities of coniferous and broad- leaved trees in the United Kingdom. Selbyana 20: 290-298. 2a wAnhuts S. I. Boulter, M» Keller;-R. L: Kiichine, ©. Korner, F. C..Meinzer, A. W. Mitchell, T. Nakashizuka, P. L. Silva Dias, N. E. Stork, S. J. Wright, and M. Yoshimura. 2003. Biodiversity meets the atmosphere: A global view of forest canopies. Science 301: 183-186. Preisser, E., D. C. Smith, and M. D. Lowman. 1998. Canopy and ground level insect distribution in a temperate forest. Selbyana 19: 141-146. Rees, C. J. C. 1983. Microclimate and the flying Hemiptera fauna of a primary lowland rainforest in Sulawesi. Pp. 121-136 in: Sutton, S. L., T. C. Whitmore, and A. C. Chadwick, eds. Tropical Rain Forests: Ecology and Management. Blackwell Scien- tific Publications, Oxford. Schaefer, M. 1991. The animal community: diversity and resources. Pp. 51-120 in: Rohrig, E., and B. Ulrich, eds. Ecosystems of the World, Temperate Deciduous Forests. Elsevier, Amsterdam. Schulze, C. H., K. E. Linsenmair, and K. Fiedler. 2001. Understory versus canopy: patterns of vertical 123 stratification and diversity among Lepidoptera in a Bornean rain forest. Plant Ecology 153: 133-152. Sheehan, W. 1994. Parasitoid community structure: effects of host abundance, phylogeny, and ecol- ogy. Pp. 90-107 in: Hawkins, B., and W. Sheehan, eds. Parasitoid Community Ecology. Oxford Uni- versity Press, Oxford, UK. Sorensen, L. L. 2003. Stratification of the spider fauna in a Tanzanian forest. Pp. 92-101 in: Basset, Y., V. Novotny, S. E. Miller, and R. L. Kitching, eds. Arthropods of Tropical Forests: Spatio-Temporal Dynamics and Resource use in the Canopy. Cam- bridge University Press, Cambridge. Stork, N. E. and P. S. Grimbacher. 2006. Beetle assemblages from an Australian tropical rain- forest show that the canopy and the ground strata contribute equally to biodiversity. Proceedings of the Royal Society B. 273: 1969-1975. Sutton, S. L. and P. J. Hudson. 1980. The vertical distribution of small flying insects in the lowland rain forest of Zaire. Zoological Journal of the Linnean Society 68: 111-123. Ulyshen, M. D. and J. L. Hanula. 2007. A comparison of the beetle (Coleoptera) fauna captured at two heights above the ground in a North American temperate deciduous forest. American Midland Naturalist 158: 260-278. Wharton, R. A., P. M. Marsh, and M. J. Sharkey, eds. 1997 Manual of the New World genera of the family Braconidae (Hymenoptera). Special Publication of the International Society of Hymenopterists, 439 RP. Winchester, N. N. and R. A. Ring. 1996. Northern temperate coastal Sitka Spruce forestswith special emphasis on canopies: Studying arthropods in an unexplored frontier. Northwest Science 70: 94— 103. Yanoviak, S. P. and M. Kaspari. 2000. Community structure and the habitat template: ants in the tropical forest canopy and litter. Oikos 89: 259-266. J. HYM. RES. Vol. 17(1), 2008, pp. 124-125 CD REVIEW What wasp is that? An interactive identi- fication guide to the Australasian families of Hymenoptera. N.B. Stevens, C.J. Ste- vens, M. Iqbal, J.T. Jennings, J. La Salle & A.D. Austin. Australian Biological Re- sources Study / Centre for Biological Information Technology (CBIT), 2007. Price: Aus$64.90. ISBN-13: 978 0 642 56851 9; ISBN-10: 0 642 56851 0 This CD represents a first attempt at an interactive key to the 67 families of Hyme- noptera present in Australasia — i.e. Aus- tralia, New Zealand and the islands south- east of Wallace’s line. In fact, because most of the families treated are cosmopolitan, the key is useful for the majority of Hymenop- tera specimens collected anywhere in the world, and this represents a major bonus. The CD is easy to run, compatible with Windows, Mac OSX, Linux or Solaris, and is based on two Lucid multiple-entry keys with attached html pages. The opening page provides links to the following sections: Introduction, Morphology, Fami- lies, Biology & Ecology, Systematics, Col- lecting, Bibliography and Acknowledge- ments, as well as the option to immediately enter the main key, or the linked subkey to chalcidoid families. The introduction is succinct and ex- tremely useful, especially because of the links throughout the text to related infor- mation and images. Abundant links are a characteristic of the CD as a whole, both to further information within the CD, but also to other websites, and this is a very helpful feature. The morphology section provides an opportunity to standardise terminology in an area that has historically been prone to multiple terms for the same feature. The section is brief, and could perhaps have been expanded (or can in future editions) to include more information on the female and male genitalia, though admittedly these are of less importance at the family level. Within this section the chalcidoid family Mymaridae is described as having wings that are “completely veinless”, whereas all Mymaridae have a single vein in the fore wing. Only certain platygastrids can correctly be described as having completely veinless wings. The next sec- tion is an A-Z listing of the 67 families treated, with just a couple of paragraphs on each family, together with key references and links. The systematics section provides an overview by superfamily of all the families of Hymenoptera, with links back to the family pages in the previous section, where appropriate. The collecting section is comprehensive and excellent, and even includes information on obtaining collect- ing permits in Australia, with links to the appropriate websites. There are just a few typos among these pages, listed here to aid preparation of any future edition: Crabrio- nidae, Mymarommatiodea, Mutilidae, Plu- maridae, Trigonalidae. Megalodontidae should be Megalodontesidae. Starting the main key, the user has little guidance or advice, and I suspect this could present initial problems for someone unfamiliar with Lucid, or Hymenoptera families, or both. However, a little persis- tence and experience should enable even a complete novice to successfully identify most families fairly quickly. Certain fami- lies are less straightforward to key out, but this can be due to the morphological heterogeneity of their constituent genera (e.g. the chalcid family Aphelinidae). The “magic wand” option within Lucid selects the most discriminatory character, and is therefore recommended. In the case of the main key, following geographical distribu- tion, this is apparently the length of the first discal cell in the fore wing. This choice of character as the most effective one to VOLUME 17, NUMBER 1, 2008 start with (after distribution) could be off- putting for a beginner. Following this, the user is then asked whether “‘hind wing vein rm joins RS after RS has diverged from SC+R”. The trouble with characters of this degree of complexity is that firstly they discourage the novice, and secondly that by the time they have been mastered, the user is anyway likely to have no difficulty recognising any hymenopteran to family level. It would be an interesting exercise to attempt a key that excludes these kinds of characters, but still works at least 95% of the time. Such a key is far more likely to be widely adopted by non-specialists. The chalcidoid and mymarommatoid families are in a separate key that can be entered directly, as well as via the general key. An attempt has been made here to present a number of “character suites” that 125 characterise several families. Again, I would suggest that the beginner faced with this barrage of information required to go one more step could be rather easily put off the whole enterprise. In summary, the authors have made a first attempt at an extremely demanding task, and have successfully produced a useful and workable product that has relevance globally, and not just to Austra- lasia. The background information and illustrations are well-researched and pre- sented, and the links are invaluable. If the keys were more straightforward to use by non-specialists, this CD could rapidly become a regularly used resource for a far greater audience than is likely to be the case while in its present form. ANDREW POLASZEK ie FY sek neal ds, ! based L te cok ry a vr ? ninollont 41% iF tie Sti ¢ ce time ; a a e-eore ond begreavir Len en sate 2 ly ! Tete Pl Mee as . ai AeXAYy ye Hy tLe, Ph « ¥ ui’ Asethip sites ¢ fe wie Gal 4 a fee Spe ne, Dap pe iid ae | we a *¢ ; i?" re ‘ip! ee aha fm: epee AEE Sa ath: iy oe } a Liane woes ! ath sit ae ae rs AgD ¥, LaGt ap: ee rgeey hey = | © 2 4 - » ee ne 2 é J a, ha. ‘ oy | ghee ict re are t ey ’ oi » * 4, : 4 ’ . a . Ss ad : & : 4 4 > 9 ‘ ie I ‘ 4 , ? { —-— s fr Pas)- 7 peal Lita! Te i - J at iy shiv 7 } } iT ’ A 38. v7 S 5 | i iY ba | Ar 1? ve, ‘ —.. ' nm) 4 * = J sad i ‘ imi * ¢ on P i P ei | ' = ? > N Pat . ‘7 My ‘ . 4 = > ¢ '