sign in sign up
<<
Home , Aculeata, Andrena nuda, Andrenidae, Apoidea, Beaver (steamship), Blacini

c 3

Journal of Research

. .IV

6«**

Volume 15, Number 2 October 2006

ISSN #1070-9428 CONTENTS

BELOKOBYLSKIJ, S. A. and K. MAETO. A new of the Parachremylus Granger (Hymenoptera: ), a of Conopomorpha lychee pests (: Gracillariidae) in Thailand 181

GIBSON, G. A. P., M. W. GATES, and G. D. BUNTIN. (Hymenoptera: Chalcidoidea) of the cabbage seedpod (Coleoptera: ) in Georgia, USA 187

V. Forest GILES, and J. S. ASCHER. A survey of the of the Black Rock Preserve, New York (Hymenoptera: ) 208

GUMOVSKY, A. V. The and morphology of sylvestris (Hymenoptera: ), a larval endoparasitoid of Ceutorhynchus sisymbrii (Coleoptera: Curculionidae) 232

of KULA, R. R., G. ZOLNEROWICH, and C. J. FERGUSON. Phylogenetic analysis Chaenusa sensu lato (Hymenoptera: Braconidae) using mitochondrial NADH 1 dehydrogenase gene sequences 251

QUINTERO A., D. and R. A. CAMBRA T The genus Allotilla Schuster (Hymenoptera: Mutilli- dae): phylogenetic analysis of its relationships, first description of the female and new distribution records 270

RIZZO, M. C. and B. MASSA. Parasitism and sex ratio of the bedeguar gall Diplolqjis 277 rosae (L.) (Hymenoptera: Cynipidae) in Sicily (Italy)

VILHELMSEN, L. and L. KROGMANN. Skeletal anatomy of the mesosoma of Palaeomymar anomalum (Blood & Kryger, 1922) (Hymenoptera: ) 290

WHARTON, R. A. The species of Stenmulopius Fischer (Hymenoptera: Braconidae, Opiinae) and the braconid sternaulus 316

(Continued on back cover) INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991

OFFICERS FOR 2006 Michael E. Schauff, President James Woolley, President-Elect Michael W. Gates, Secretary Justin O. Schmidt, Treasurer Gavin R. Broad, Editor

Subject Editors Symphyta and

Biology: Mark Shaw Biology: Sydney Cameron Systematics: Donald Quicke Systematics: Wojciech Pulawski

All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, 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, Centre for Ecology & Hydrology, Monks Wood, Abbots Ripton, Hunting- don, Peterborough PE28 2LS, 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 , Smithsonian Institution, Washington, D.C. 20560- 0168, U.S.A. Members in good standing receive the Journal. Nonmember subscriptions are $60.00 (U.S. currency) per year.

The Society does not exchange its publications for those of other societies.

Please see inside back cover of this issue for information regarding preparation of manuscripts.

Statement of Ownership

Title of Publication: Journal of Hymenoptera Research. Frequency of Issue: Twice a year. Location of Office of Publication, Business Office of Publisher and Owner: International Society of Hymenopterists, Department of Entomology, Smithsonian Institution, 10th and Constitution NW, Washington, D.C. 20560-0168, U.S.A. Editor: Gavin R. Broad, Centre for Ecology & Hydrology, Monks Wood, Abbots Ripton, Hunting- don, Peterborough PE28 2LS, UK. Editor and Managing Known Bondholders or other Security Holders: none.

This issue was mailed 20 October 2006 J. HYM. RES. Vol. 15(2), 2006, pp. 181-186

A New Species of the Genus Parachremylns Granger (Hymenoptera: Braconidae), a Parasitoid of Conopomorpha Lychee Pests (Lepidoptera: Gracillariidae) in Thailand

Sergey A. Belokobylskij and Kaoru Maeto

(SAB) Zoological Institute Russian Academy of Sciences, St. Petersburg, 199034, Russia, and Museum and Institute of Zoology Polish Academy of Sciences, Wilcza 64, Warszawa 00-679, Poland; email: [email protected] (KM) Laboratory of Science, Faculty of Agriculture, Kobe University, Rokkodai-machi 1-1, Nada-ku, Kobe 657-8501, Japan; email: [email protected]

Abstract. — Parachremylns litchii Belokobylskij & Maeto, new species, from Thailand is the described as a parasitoid of larvae of Conopomorpha sinensis Bradley and C. litchiella Bradley, of and major pests of lychee and longan in South-East Asia. The taxonomic position Parachremylns the range of the hosts of related genera of parasitoids are discussed.

Several insect pests are seriously threat- The genus Parachremylns with type spe- de- ening lychee (Litchi chinensis Sonn.) and cies P. seyrigi Granger was originally longan (Dimocarpus longan Lour.) (Sapin- scribed from Madagascar (Granger 1949); daceae) growers. They are the fruit borer this genus— occurrs also in continental {Conopomorpha sinensis Bradley), leaf miner Nigeria and Niger (Wharton (Conopomorpha litchiella Bradley), longan 1993). Two additional species of this genus recorded from the sucking bug (Tessaratoma papillosa Drury), have already been fruit piercing (Othreis fnllonia Oriental region. Parachremylns oblongus India in the (Clerck)), and twig borer (Zenzera coffeae (Papp) was described from Nietner) (Menzel 2002). genus Avga Nixon (Papp 1990, 1997), and Conopomorpma sinensis, the lychee stem- P. temporalis Belokobylskij from Brunei end borer and the lychee fruit borer in (Belokobylskij 1999). A fourth species of China, Thailand and India, is the major this genus, similar to P. temporalis, is Thailand. The pest of lychee and longan in these coun- described below from sys- tries. Conopomorpha sinensis and the related tematic position of this genus is disputable. C. litchiella both attack lychee and longan, Parachremylns is included in the subfamily the latter preferring to mine leaves and Exothecinae (tribe Avgini: Belokobylskij shoots (Bradley 1986). There have been 1993), or conventionally in subfamily Hor- of the only tentative reports on braconid para- miinae (Wharton 1993). In spite sitoids of the pest Conopomorpiha borers: different understanding of the contents of the of this close Phanerotoma sp., Pholetesor (Apanteles) sp., subfamilies, position genus both and Colastes sp. (Menzel 2002, Anupunt to Avga Nixon is suggested by and Sukhvibul 2005), but possibly infor- authors. Belokobylskij (1993) discussed mation about Colastes is due to misdeter- the relationships of these genera with Tobias mination. Here we report a new braconid Parahormins Nixon, Pseudohormius of the genus Parachremylns Granger as & Alexeev and Allobracon Gahan (= Lcur- the loss of a larval parasitoid of C. sinensis and C. inion Muesebeck), which share litchiella. the prepectal (epicnemial) carina on the 182 Journal of Hymenoptera Research

mesosoma. Wharton (1993) provisionally vi, 1997, Supatra Dolsopon", "Host: Con- placed Avga near Parahormius and Pseudo- opomorpha sinensis— larvae" (NIAES, ZISP). hormius and showed the possible relation- Description. Female. Body length 2.6- ship of Avga and Parachremylus (shared 2.8 mm; fore wing length 2.5-2.6 mm. granulate mesonotal sculpture and the Antennae: thickened, almost filiform, 29- poorly developed propleural flange). How- 30-segmented, 1.1-1.2 times longer than ever, in his opinion, Allobracon does not body. Scapus 1.7-2.0 times longer than appear to be closely related to Parachremy- wide. First flagellar segment 2.5-2.8 times lus in spite of it sharing a number of longer than its apical width, 1.1-1.2 times features with Avga and Parahormius. longer than second segment. Penultimate The host of Parachremylus has not been segment 2.0-2.3 times longer than wide, 0.6-0.7 known till now. The new species described times as long as first flagellar 0.7-0.75 times as below as P. litchii sp. nov. was reared from segment, long as apical larvae of Conopomorpha sinensis and C. segment; the latter with distinct spine Head: litchiella (Gracillariidae), both important apically. width 1.8-2.0 times its median 1.25-1.4 pests of lychee and longan trees in South- length, times width of mesoscutum. East Asia. The members of related genera Temple very strongly and almost narrowed behind of the tribe Avgini {Parahormius, Avga, linearly eye (dor- sal Transverse diameter of Allobracon) are also recorded as parasitoids view). eye of the leaf-rollers or leaf-miners of the (dorsal view) 5.5-7.0 times longer than times if measured families Tortricidae, Gracillariidae, Lyone- temple length (7.0-7.7 on Ocelli in with tiidae, Cosmopterigidae, Coleophoridae, straight line). small, triangle base 1.1-1.15 times its sides. POL 0.7-1.0 and Gelechiidae (Belokobylskij 1993, Whar- times Od, 0.3-0.5 times OOL. Vertex with ton 1993) as well as rarely (recorded for narrow median furrow. Oc- Allobracon) of leaf-mining Coleoptera longitudinal carina (Wharton 1993). cipital dorsally distinctly curved towards rather The terms of wing venation are used as ocelli, widely interrupted not fused with cari- defined by Belokobylskij and Tobias (1998). medially; hypostomal na obliterated for a short The following abbreviations are used: ventrally being — distance. POL postocellar line; OOL—ocular-ocellar Eye large, sub-round, glabrous, 1.1-1.2 times as as broad. Malar line; Od—maximum diameter of lateral high space 0.25-0.3 times of 0.8-0.9 ocellus; NIAES—National Institute of height eye, times basal width of mandible. Face width 0.9 Agro-Environmental Sciences (Tsukuba, times height of and 1.2-1.25 times Japan); ZISP—Zoological Institute, Rus- eye sian of height of face and clypeus combined. Academy Sciences (St. Petersburg, Malar suture absent. suture rather Russia). Clypeal distinct and complete. Clypeus weakly Parachremylus litchii Belokobylskij & convex. Hypoclypeal depression sub- Maeto, new species round, its width 0.8-0.9 times distance (Figs 1-11) from edge of depression to eye, 0.35 times width of face. Head below eyes (front Holotype female.— "Horticultural Re- view) strongly and almost linearly nar- search Center, Rai, Thailand, viii, Chiang rowed. Mesosoma: length 1.5-1.55 times its 1996, Supatra "Host: Dolsopon", Conopo- height. Mesoscutum highly and almost morpha litchiella larvae on or Lon- Lychee perpendicularly raised above pronotum gan" (NIAES). 2 females, 1 Paratypes. (lateral view), with rather fine longitudinal male, with the same labels as holotype medioposterior keel (dorsal view). Notauli (NIAES, ZISP); 5 females, "Horticultural rather narrow, shallow anteriorly on verti- Research Center, Chiang Rai, Thailand, 6. cal surface and very shallow to almost Volume 15, Number 2, 2006 183

Six basal Figs 1-11. Parachremylus litchii sp. nov. 1, Head, frontal view. 2, Head, dorsal view. 3, . 4, segments of . 5, Mesosoma, lateral view. 6, Hind tibia and two basal segments of hind tarsus. 7, Metasoma, dorsal view. 8, Hind coxa. 9, Hind femur. 10, Fore wing. 11, Hind wing. 184 Journal of Hymenoptera Research

of hind tarsus absent on dorsal surface, finely sculptured. First-fourth segments ven- with wide and Prescutellar depression short, shallow, trally transparent flanges, 0.15-0.2 times which are on the of each finely crenulate-granulate, pointed tops Metnsomn: 1.7-2.0 times as long as scutellum. Scutellum almost flat. segment. longer Metanotum medially with small and ob- than its maximum width, 0.9-1.1 times as tuse tubercle. Subalar depression rather long as head and mesosoma combined. shallow, wide, densely and curvedly stri- First tergite strongly, uniformly and line- from base to with ate with fine granulation anteriorly. Ster- arly widened apex; small nauli shallow, rather wide, weakly curved, spiracular tubercles before its middle; with distinct and rather wide entirely smooth. Wings: Length of fore laterally high wing 2.8-3.0 times its maximum width. carinae; fine dorsal carinae fused in basal Radial cell not shortened, metacarpus 1.3 0.3 and then extending to apex as a single, times longer than pterostigma. Pterostigma elevated, median carina; dorsope absent. rather wide, 3.1-3.7 times longer than Apical width of first tergite 2.4-2.7 times its wide. Radial vein arising a little or rather basal width; its length 0.6-0.65 times apical rather distinct distinctly before middle of pterostigma. width. Second suture and Second radial abscissa 1.5-2.3 times longer convex. Second and third tergites with than first abscissa, 0.25-0.3 times as long as rather distinct and fine longitudinal medi- the straight third abscissa, 1.15-1.25 times an carina. Median length of second tergite longer than the weakly curved first radio- about half its basal width, equal to or 1.1- medial vein. Second radiomedial cell short, 1.2 times length of third tergite. Combined weakly narrowed towards apex, its length median length of second and third tergites 1.5-1.8 times maximum width, 0.9-1.1 nearly equal to basal width of second times length of brachial cell. First medial tergite, 0.7-0.75 times maximum width of abscissa rather distinctly S-shaped. Recur- tergites. sheath (visible part in rent vein 0.9-1.0 times as long as second lateral view) 1.1-1.3 times longer than first abscissa of medial vein. Discoidal cell 1.55- tergite, 1.0-1.2 times longer than hind 1.65 times longer than wide. Nervulus basitarsus, 0.25-0.4 times as long as meso- strongly postfurcal, distance from nervulus soma, 0.15-0.17 times as long as fore wing. to basal vein nearly twice nervulus length. Sculpture mid pubescence: Head very densely Parallel vein arising a little behind middle and minutely granulate, face additionally of distal margin of brachial cell. Hind wing with rather fine and irregular striation. 4.5-4.7 times longer than maximum width. Mesoscutum very densely and distinctly First abscissa of costal vein 0.85-0.9 times granulate, with rather narrow and long as long as second abscissa. First abscissa of rugulosity in medioposterior half. Scutel- mediocubital vein 1.15-1.2 times longer lum finely and densely granulate. Meso- than second abscissa. Recurrent vein short, pleuron almost smooth in lower half. unsclerotized, interstitial, curved toward Metapleuron coarsely, regularly and curv- base of wing. Legs: Hind coxa large, 1.5-1.6 edly striate for the most part, with fine times longer than wide, 0.7-0.75 times as granulation between striae and anteriorly. as hind femur. long Hind femur wide, 3.1- Propodeum almost entirely coarsely and 3.2 times longer than wide. Hind tibia rather sparsely striate, striae in areola more thickened towards apex. Hind tarsus 1.1 or less transverse and partly undulate or times than longer hind tibia; hind basitar- rugulose, with fine granulation partly; sus 0.6-0.65 times combined length of with distinctly delineated basolateral areas; second-fifth segments (without pretarsus). areola wide, its length 1.0-1.2 times Second tarsal segment 0.4-0.45 times as maximum width; dorsal carina 0.8-1.0 as long basitarsus, 1,2-1.3 times longer times as long as areola fork. Hind coxa than fifth segment (without pretarsus). smooth; hind femur finely punctulate with Volume 15, Number 2, 2006 185

very fine granulation dorsally, smooth median length. Transverse diameter of eye ventrally. Metasoma entirely densely (dorsal view) 8.8 times longer than temple granulate, granulation becoming finer to- length if measured on straight line. Anten- wards apex of metasoma. Mesoscutum na 28-segmented. Otherwise similar to entirely shortly and very densely setose. female. Hind tibia dorsally with rather short, Diagnosis. —The new species is very dense and semi-erect setae, its length similar to P. temporalis Belokobylskij from 0.35-0.55 times maximum width of tibia. Brunei (Belokobylskij 1999) and differs in Colour: Head and anterior half of meso- having the recurrent vein as long as second soma (including mesoscutum) yellow, abscissa of medial vein, the nervulus posterior part of mesosoma and metasoma strongly postfurcal, the pterostigma rather pale yellow, metasoma additionally often wide, the hind femur wide, the second with greenish tint. Antenna reddish tergite short, the face rather finely striate, brown or brown, scapus mostly yellow. and the propodeum almost entirely coarse- Palpi pale yellow. Legs yellow, all tarsi ly rugose-striate.— (especially posterior ones) more or less Host. Conopomorpha sinensis Bradley brown. Ovipositor sheath brown in basal and C. litchiella—Bradley (Gracillariidae). half and black in apical half. Fore wing Distribution.— Thailand. faintly infuscate. Pterostigma brownish Etymology. This species is named after the name of the fruit tree— yellow. — lychee {Litchi Male. Body length 2.0 mm; fore wing chinensis Sonn.)—on which their hosts length 2.4 mm. Head width 2.1 times its develop.

KEY TO SPECIES OF THE GENUS PARACHREMYLUS GRANGER

1. Temple longer; transverse diameter of eye (dorsal view) 4.0-5.0 times as long as temple length. Malar space larger than basal width of mandible. Mesopleuron smooth in upper st— th half, striation partly present in subalar depression only. 1 4 segments of hind tarsus with narrow and partly indistinct flanges 2

- Temple shorter; transverse diameter of eye (dorsal view) 5.5-7.0 times as long as temple length. Malar space less than basal width of mandible. Mesopleuron distinctly curvedly st striate in upper 0.4-0.5. l -4* segments of hind tarsus with wide flanges 3 2. Notauli complete, rather deep posteriorly. Metacarpus 1.2-1.3 times as long as pterostigma.

1 nd rd T flagellar segment 3.5-3.7 times as long as apical width. Median length of 2 and 3 nd metasomal combined a little than basal width of 2 tergites larger — tergite. Propodeum mostly coarsely and sparsely striate. Body length 2.2 mm. Madagascar P. seyrigi Granger

- Notauli incomplete, almost absent posteriorly. Metacarpus 1.5 times as long as pterostigma. st nd rd 1 flagellar segment 3.0 times as long as apical width. Median length of 2 and 3 nd metasomal tergites combined 1.3 times basal width of 2 tergite. Propodeum mostlv smooth. Body length 2.0 mm.—India P. oblougus (Papp) 3. Pterostigma narrow, 5.0 times as long as maximum width. Recurrent vein of fore wing about twice as long as second abscissa of medial vein. Nervulus not strongly postfurcal, distance from nervulus to basal vein 0.7 times nervulus length. Hind femur 3.5 times as nd long as wide. 2 tergite 0.6 times as long as its basal width. Face almost entirely distinctly transversely striate. Propodeum within background areolation sparsely striate, mostly smooth. Body length 2.3 mm.—Brunei P. temporalis Belokobylskij Journal of Hymenoptera Research

width. Pterostigma rather wide, 3.1-3.7 times as long as maximum Recurrent vein of fore wing almost as long as second abscissa of medial vein. Nervulus strongly postfurcal, distance from nervulus to basal vein nearly twice nervulus length. Hind femur 3.1-3.2 nd as its basal width. times as long as wide. 2 tergite about 0.5 times as long Face finely and partly indistinctly transversely striate and with dense fine granulation. Propodeum within background areolation almost entirely coarsely rugose-striate with fine granula- —Thailand P. litchii tion partly. Body length 2.0-2.8 mm. sp. nov.

ACKNOWLEDGMENTS Beiokobylskij, S. A. and V. I. Tobias. 1998. Fam. Braconidae. Introduction. Pp. 8-26. in: Lehr, P. A. We wish to thank Dr. S. Moriya (Tsukuba, Japan) ed. OpredeliteV nasekomykh Dal'nego Vostoka Rossii Dr. S. for us and Dolsopon (Bangkok, Thailand) giving [Keys to the of the Russian Far East]. 4(3). a chance to examine the reared material, and two Dal'nauka, Vladivostok. [In Russian.] reviewers for valuable comments of our manuscript. Bradley, J. D. 1986. Identity of the South-East Asian The work was the of present supported by grant Japan cocoa moth, Conopomorpha cramerella (Snellen) for the Promotion of Science, Invitation Society (Lepidoptera: Gracillariidae), with descriptions for Research in Fellowship Program Japan of three allied new species. Bulletin of Entomolog- (No. L05564) and grant 2 P04C 001 28 of the Ministry ical Research 76: 41-51. of Science and Information Society Technologies Granger, C. 1949. Braconides de Madagascar. Mem- (Poland) for the first author, and a Grant-in-Aid for oires de llnstitut Scientifique de Madagascar. Ser. A, Scientific Research (A) (No. 15208007) from the Japan Biologic Animate 2: 1-428. for the Promotion of Science for the second Society C. 2002. Menzel, The lychee crop in Asia and the Pacific. author. Food and Agriculture Organization of the United LITERATURE CITED Nations, Bangkok. 120 pp. 1990. Papp, J. New braconid (Hymenoptera, Braconidae) in the Natural Anupunt, P. and N. Sukhvibul. 2005. Lychee and Hungarian History Museum, 1. Annates historico-naturales Musei longan production in Thailand. Acta Horticultural' (ISHS) 665: 53-60. nationalis hungarici 82: 175-190. 1997. New braconid Beiokobylskij, S. A. 1993. [On the classification and Papp, J. wasps (Hymenoptera, in the Natural phylogeny of the braconid wasps of subfamilies Braconidae) Hungarian History 5. Annates historico-naturales Musei Doryctinae and Exothecinae (Hymenoptera, Bra- Museum, nationalis 89: 157-175. conidae). Part I. On the classification, 2.] Entomo- lunigarici R. D. logicheskoe Obozrenie 72: 143-164. [In Russian.] Shenefelt, 1975. Braconidae 8. Exothecinae, Beiokobylskij, S. A. 1999. [New taxa of the braconid Rogadinae. Hymoiopterorum Catalogus (nova editio) subfamily Exothecinae (Hymenoptera, Braconi- 12: 1115-1262. dae) from tropical and subtropical regions of the Wharton, R. A. 1993. Review of the Hormiini Old World. I.] Entomologicheskoe Obozrenie 78: (Hymenoptera: Braconidae) with description of 674-693. [In Russian.] new taxa. Journal of Natural History 27: 107-171. J. HYM. RES. Vol. 15(2), 2006, pp. 187-207

Parasitoids (Hymenoptera: Chalcidoidea) of the Cabbage Seedpod Weevil (Coleoptera: Curculionidae) in Georgia, USA

Gary A. P. Gibson, Michael W. Gates and G. David Buntin

(GAP) Agriculture and Agri-Food , and Integrated Pest Management, K. W. Neatby Bldg., 960 Carling Avenue, Ottawa, ON, Canada, K1A 0C6; email: [email protected] (MWG) Systematic Entomology Laboratory, PSI, Agricultural Research Service, U.S. Department of Agriculture, c/o National Museum of Natural History, Smithsonian Institution, Washington, DC 20560-0168, USA (GDB) Department of Entomology, University of Georgia, Georgia Station, Griffin, GA 30223, USA

— Abstract. Five families and 13 species of Chalcidoidea (Hymenoptera) were obtained from mass-reared seedpods of Brassica napus L. () as putative parasitoids of the cabbage seedpod weevil, Ceutorhynchus obstrictus (Marsham) (Coleoptera: Curculionidae), in Georgia, USA. The species are torvina () (), Euderus glaucus Yoshimoto and Necremnus tidius (Walker) (Eulophidae), Brasema allynii (French) n. comb, (from Eupelmus Dalman) and Eiipclimis cyaniceps Ashmead (), Eurytoma tylodermatis Ashmead (), and Lyrcus incertus (Ashmead), L. maculatus (Gahan), L. perdubius (Girault), Mesopolobus moryoides Gibson, Neocatolaccus tylodermae (Ashmead), Pteromalus cerealellac (Ashmead) and Pteromalus sp. (). An illustrated key is provided to differentiate the taxa. Lyrcus maculatus constituted about 96% of all reared Pteromalidae and 86% of the total parasitoid fauna. The associations of B. allynii, E. glaucus, E. cyaniceps, E. tylodermatis, L. incertus, N. tylodermae, Pteromalus sp. and P. cerealellac with C. obstrictus are new, but some of these species likely are hyperparasitoids or emerged from insect contaminants of the mass-reared seedpods. The only previous report of a parasitoid of C. obstrictus in eastern , Trichomalus perfectus (Walker) (Pteromalidae), is a misidentification. The parasitoid fauna of C. obstrictus in Georgia is discussed relative to that known for western North America.

The cabbage seedpod weevil, Ceuto- tera: Chalcidoidea) parasitoids of the cab- rhynchus obstrictus (Marsham) (Coleoptera: bage seedpod weevil in western North Curculionidae), was introduced from Eur- America, including Breakey et al. (1944), ope to western North America about Doucette (1944, 1948), Hanson et al. (1948), 70 years ago. Since then it has become the Carlson et al. (1951), McLeod (1953), Walz most important insect pest of canola and (1957), and Dosdall et al. (in press). rape, Brassica napus L. and B. rapa L. Murchie and Williams (1998) listed 7 (Brassicaceae), in most areas of the conti- identified and 4 unidentified species in 9 nent where these crops are grown (Car- genera and 5 families of Chalcidoidea as camo et al. 2001, Kuhlmann et al. 2002). It parasitoids of C. obstrictus in North Amer- was first reported from eastern North ica, but almost all of the species names America in North Carolina, USA (USDA either represent misidentifications or are 1960), and is now known to extend from now recognized as junior synonyms of Georgia to Quebec and Ontario, Canada older names (Gibson et al. 2005). Dosdall et (Brodeur et al. 2001, Mason et al. 2004). al. (in press) reported another six chalcid There have been several surveys of the species as reared from B. napus and B. rapa introduced and native chalcid (Hymenop- seedpods in Alberta. Consequently, the of Hymenoptera Research 188 Journal

with the Table 1. Chalcid parasitoids associated cabbage seedpod weevil in North America, including and for Georgia the number of specimens percentage total reared Buntin (in parenthesis) of parasitoids by (1998). Volume 15, Number 2, 2006 189

Aphididae) and larvae and pupae of the using Adobe Photoshop™ to enhance cla- diamondback moth, Plutella xylostella (L.) rity. The reared (Lepidoptera: Plutellidae). RESULTS parasitoids had been stored in ethanol, of Chalcidoi- but were critical-point dried, point- A total of 1,127 specimens mounted, and identified to genus by the dea were sufficiently intact that they could senior author using the relevant be identified accurately. Of these, there keys in Gibson et al. (1997). The senior was a single male Pachyneuron aphidis author is responsible for all species identi- (Bouche) and a female and male Asnphes fications except Eurytoma tylodennatis Ash- suspensus (Nees) (Pteromalidae). Members mead (Eurytomidae), which was identified of Pachyneuron and Asaphes are obligate by MWG. Information concerning the hyperparasitoids of (Gibson et al. further. The method of species identification within 1997) and are not dealt with each genus is provided under the relevant remaining 1,124 specimens included 5 and 13 of species discussion. Voucher specimens are families, 10 genera, species deposited in the Canadian National Col- Chalcidoidea that are possible parasitoids lection of Insects and Arachnids (CNC), of C. obstrictus. These taxa are keyed below Ottawa, Ontario, the University of Georgia and subsequently treated by family in Museum of Natural History (UGCA), alphabetical order. The key also segregates Gra- Athens, Georgia, and the United States Trichoinalus Thomson and Chlorocytus National Museum of Natural History ham, the only two genera reared from C. (USNM), Washington, District of Colum- obstrictus in western North America (Dos- bia. Terms used for parasitoid structure dall et al. in press) that were not recovered taxa follow Gibson (1997). Photographs are by Buntin (1998) (Table 1). The two composite serial images that were com- are included in the key because species of bined using Auto-Montage™. These both genera occur in eastern North Amer- images and the scanning electron micro- ica and may eventually be reared as part of fauna. photographs were digitally retouched the regional C. obstrictus parasitoid

KEY TO CHALCIDOIDEA PUTATIVELY PARASITIZING C. OBSTRICTUS IN GEORGIA

and 1 Hind leg with elongate coxa of similar length to conspicuously swollen ventrally

toothed femur, and with curved tibia (Fig. 1) . . . Conura torvina (Cresson) (Chalcididae) tibia ... 2 Hind leg with comparatively short coxa, slender femur, and straight (Figs 3-8) 2(1) Tarsi 4-segmented; flagellum with 3 or 4 funicular segments, the segments sometimes 3 branched (Figs 9, 10) (Eulophidae) ... 4 Tarsi 5-segmented; flagellum with 5 or 6 unbranched funicular segments (Figs 11-18) 3(2) Meso- and metatarsi with basal 4 segments white; forewing membrane bare dorsally setae behind marginal vein, but with clearly visible row of long admarginal of both sexes with 4 (Fig. 42, ams) on ventral surface near marginal vein; flagellum unmodified funicular segments Euderus gJaucus Yoshimoto Meso- and metatarsi with basal 1 or 2 segments white; forewing membrane uniformly setae setose dorsally behind marginal vein, the setae largely obscuring admarginal in and with 3 on ventral surface (Fig. 41); flagellum branched male (Fig. 10) only Necrcmmis tidius funicular segments in female (Fig. 9) (Walker) 4(2) Head and mesosoma with coarse piliferous punctures and non-metallic, dark brown in dorsal to black (Figs 3, 4); pronotal collar quadrangular view, only slightly shorter than mesoscutum; male with elongate petiole and flagellar segments having whorls of conspicuously long setae (Fig. 4) 190 Journal of Hymenoptera Research

Eiin/toma tylodermatis Ashmead (Eurytomidae) Head and mesosoma with finer mesh-like sculpture and often with metallic green to bluish luster; pronotum strongly transverse in dorsal view, conspicuously shorter than mesoscutum; male with short petiole and flagellar segments having short, inconspicuous setae 5 and 5(4) Mesopleuron elongate, convex or cushion-like, uniformly finely sculptured at of tibia and on ventral (Figs 5, 7); middle leg with strong black spines apex of surface of tarsal segments (Figs 5, 7, sp), the colour spines contrasting distinctly 6 with mostly yellowish leg (Eupelmidae: Eupelminae female) Mesopleuron about as high as long, usually with a smooth region dorsally below base of wings and often with an oblique femoral depression or groove, but at least not convex or uniformly sculptured (Figs 23-25); middle leg with slender spines at apex of tibia and on ventral surface of tarsal segments, the colour of spines not 7 contrasting with leg of 6(5) Ovipositor sheaths projecting only slightly beyond apex gaster and uniformly and coloured (Fig. 5); forewing completely setose behind parastigma marginal vein Brasema (Fig. 5) allynii (French) Ovipositor sheaths projecting beyond apex of gaster by about one-third its length and medially whitish between darker basal and apical bands (Fig. 7); forewing with slender, oblique bare band (Fig. 7, bb) below parastigma and base of marginal vein Eupelmus cyaniceps Ashmead 7(5) Flagellum with only basal segment conspicuously differentiated as strongly transverse ring segment and with 7 or 8 distinct funicular segments; head in frontal view with inner margin of eyes distinctly divergent over about ventral half; mesotibial spur much longer and thicker than metatibial spurs, as long as basal tarsal segment and about one-third length of tarsus (Eupelmidae: Eupelminae male) 8 Flagellum with 2 or 3 basal segments conspicuously differentiated as ring segments and then with 6 or 5 tubular funicular segments, respectively (Figs 11-18); head in frontal view with inner margin of eyes subparallel or slightly but uniformly incurved; mesotibial spur somewhat longer than, but otherwise similar to, metatibial spurs, the spur obviously shorter than basal tarsal segment and only about one-quarter length of tarsus (Pteromalidae) 9 8(7) Forewing completely setose behind parastigma and base of marginal vein (Fig. 6); hind leg with femur yellowish-white and tibia usually more or less distinctly brown (Fig. 6); flagellum clavate, the segments widening distinctly to clava and funicular apical segments transverse (Fig. 6) Brasema allynii (French) with Forewing large, oblique bare region (Fig. 8, bb) behind parastigma and base of hind with marginal vein; leg femur and tibia dark (Fig. 8); flagellum robust- the all filiform, segments about the same width and apical funicular segments quadrate (Fig. 8) Eupelmus cyaniceps Ashmead with 5 tubular 9(7) Flagellum funicular segments and 3 strongly transverse ring segments (Figs 11, 13, 14) 10 with 6 tubular funicular Flagellum segments and 2 ring segments (Figs 12, 15-18) ... 14 10(9) Female only: head and mesosoma dark with conspicuous, white, lanceolate setae costal cell (Figs 19, 20, 24, 25); ventrally with setae only within about apical half of cell (Figs 43, 46) 11 Female or male: head and mesosoma variable in colour, but with inconspicuous hairlike setae (Figs 21-23); costal cell ventrally with line of setae extending almost entire of cell or at least setae length present both basally and apically if line more or less interrupted medially (Figs 44, 45) 13 setose behind 11(10) Forewing dorsally marginal vein over about apical half of vein, the setae at least 3 rows of ventral partly obscuring admarginal setae apically (Fig. 43); with transverse propodeum ridge or carina within anterior half dividing it into anterior and sections on posterior either side of median carina (Fig. 30); Volume 15, Number 2, 2006 191

metapleuron completely sculptured and with anterior margin on same plane as and abutting mesopleuron (Fig. 25, am) Neocatolaccus tylodertnae (Ashmead) Forewing dorsally bare behind marginal vein to level at least equal with middle of stigmal vein, completely exposing 1 or sometimes 2 partial rows of ventral with transverse furrow admarginal setae (Fig. 46, ams); propodeum sometimes near middle, but without transverse ridge (Figs 27, 29); metapleuron partly smooth or anteriorly and with anterior margin either curved outward (Figs 24, 29) extending anteriorly above and over posterior margin of mesopleuron (Figs 27, 28) 12 carina 12(11) Propodeum with nucha (Fig. 29, nuc) delineated laterally by longitudinal within furrow along posterior margin; metapleuron with anterior margin (Figs 24, 29, am) curved outwards, extending as thin brown flange almost at right angle to posterior margin of mesopleuron; lower face without evident malar depression, evenly convex along oral margin between malar sulcus and clypeus Lyrcus perdubius (Girault) Propodeum with nucha not delineated laterally by carina, the furrow along posterior margin of callus continued uninterrupted mesally and anteriorly so as to delineate, more or less conspicuously, anterior limit of nucha (Fig. 27); metapleuron with anterior margin (Figs 27, 28, am) raised above and extending over posterior margin of mesopleuron; lower face with short but distinct, concave malar depression incertus (Fig. 20, md) between malar sulcus and clypeus Lyrcus (Ashmead) 13(10) Both sexes: mesonotum usually dark with conspicuous pattern of bluish-green spots, the spots usually most distinct on mesoscutum paramedially behind pronotum and laterally on lateral lobe adjacent to notaulus, though small specimens sometimes brown. Female: gaster lanceolate (Fig. 22); forewing dorsally bare behind marginal vein to level at least equal with middle of stigmal vein (Fig. 45). Male: flagellum brown with first funicular segment oblong and much longer than combined length as thick as width of of the 3 ring segments (Fig. 14); marginal vein strong, but only stigma and with posterior margin straight, parallel with anterior margin Lyrcus maculatus (Gahan) Both sexes: mesonotum metallic green. Female: gaster subcircular (Fig. 21); forewing dorsally bare behind marginal vein, but apically the setae extending to base of first funicular stigmal vein (Fig. 44). Male: flagellum yellowish with segment of the quadrate to slightly wider than long and at most as long as combined length et al. vein thickened 3 ring segments (Gibson 2005, fig. 8); marginal conspicuously relative to slender stigma and with posterior margin slightly convex (Gibson et al. Gibson 2005, fig. 31) Mesopiolobus moryoides to level about 14(9) Male only: forewing with bare band behind marginal vein extending equal with middle of stigmal vein, and with 1 or at most 2 partial rows of than setae on dorsal admarginal setae (Fig. 46, ams) that are obviously longer surface of disc; metapleuron partly smooth and with anterior margin (Figs 24, 28, am) curved outward or raised above mesopleuron 15 the Male or female: forewing with bare region behind marginal vein less extensive, discal setae extending to or almost to base of stigmal vein, and with more than 2 rows of admarginal setae of about same length as setae on dorsal surface of disc anterior (Figs 43, 47-50); metapleuron completely sculptured and with margin 16 (Fig. 25, am) on same plane as and abutting mesopleuron carina 15(14) Propodeum with nucha (Fig. 29, nuc) delineated laterally by longitudinal within furrow along posterior margin; flagellum with most funicular segments only than slightly longer than wide, the first segment subquadrate and shorter pedicel curved (Fig. 15); metapleuron with anterior margin (Figs 24, 29, am) outwards, extending as thin brown flange almost at right angle to posterior margin of mesopleuron Lyrcus perdubius (Girault) Journal of Hymenoptera Research

the furrow Propodeum with nucha not delineated laterally by carina, along posterior and so as to margin of callus extending uninterrupted mesally anteriorly delineate, with all more or less conspicuously, anterior limit of nucha (Fig. 27); flagellum as as funicular segments clearly oblong, the first segment long pedicel (Fig. 16); raised above and metapleuron with anterior margin (Figs 27, 28, am) extending incertus anteriorly over posterior margin of mesopleuron Lyrcus (Ashmead) lanceolate white setae 16(14) Male only: head and body dark with conspicuous, long, slightly with transverse or carina within anterior half it (Fig. 25); propodeum ridge dividing carina into anterior and posterior sections on either side of median Neocatolaccus (Fig. 30) tylodermae (Ashmead) with hairlike setae Male or female: head and body metallic green inconspicuous with or without median carina but without transverse (Fig. 26); propodeum ridge 17 (Figs 35-38) the reticulations 17(16) Pronotum anteriorly with collar rounded into neck, extending to inclined surface with uninterrupted from dorsal (cf. Figs 23, 24); forewing marginal vein comparatively short, less than 1.5 times as long as stigmal vein reticulate nucha reticulate (Figs 49, 50); propodeum with convex, (Figs 31, 32, nuc), distinct carina 18 panels (Figs 31, 32, pnl), and often without median Pronotum anteriorly with shiny, transverse carina differentiating collar from neck least 1.5 than (Fig. 26); forewing with marginal vein obviously (at times) longer flat lunate or stigmal vein (Figs 47, 48); propodeum with or slightly concave, inverted triangular adpetiolar strip (Fig. 35, aps) delineated by Y-shaped median carina anterior to petiolar foramen or, if with reticulate nucha (Fig. 37, nuc), then fine or with panels (Fig. 37, pnl) partly strigose (having oblique, irregular, carinae striae) 19

18(17) Female: costal cell with line of setae on ventral surface interrupted medially (Fig. 49); scutellum anterior to frenum with reticulations distinctly smaller medially than laterally (Fig. 33); propodeum with plical carina (Fig. 31, pc) directed obliquely toward inner margin of spiracle Pteromalus cereahilae (Ashmead) Female: costal cell with entire line of setae ventrally (Fig. 50); scutellum anterior to frenum with almost uniform meshlike reticulations (Fig. 34); propodeum with plical carina (Fig. 32, pc) less strongly angled, directed distinctly mesal of inner margin of spiracle toward outer margin of basal fovea (Fig. 32, bf) ... Pteromalus sp. 19(17) Metacoxa setose dorsally only over about apical half; forewing of female without setae on basal fold (Fig. 48); propodeum with inverted Y-shaped median carina delimiting adpetiolar strip (Fig. 35, aps); propodeum in male without complete plical carina (Fig. 36, pc) and in female largely bare posterior to spiracle, setose only 3 from callus to postspiracular sulcus (Fig. 35, pss) Chlorocytus Graham Metacoxa setose dorsally over at least apical two-thirds and often completely setose to of with at base; forewing female least a couple of setae on basal fold (Fig. 47, bf) differentiating basal cell from speculum; propodeum with or without median carina but with convex, reticulate nucha (Fig. 37, nuc); propodeum in male with complete plical carina (Fig. 38, pc) and in female extensively setose posterior to from callus to spiracle, complete, strong plical carina (Fig. 37) Trichomalus Thomson 3

Genus not yet reported parasitizing C. obstrictus in eastern North America. Volume 15, Number 2, 2006 193

SPECIES NOTES, ARRANGED species are more easily differentiated. BY FAMILY Males of C. torvina have the interantennal region and lower face yellow, whereas Chalcididae males of C. have the dark — albifrons clypeus One species of Chalcididae was reared so that they have a conspicuous, angulate Conura torvina (Cresson), which comprised (A-like), yellow band extending dorsally nine specimens (7 99, 2 33) or 0.8% of the between the antennal scrobes. parasitoid fauna. Delvare (1992) keyed the Eulophidae species-groups of Conura and differentiat- ed C. torvina as one of eight species of the Two genera and species of Eulophidae side-group in a key to the "common were reared—Euderus glaucus Yoshimoto (2 species" of that group in North America 99; 0.2% of the parasitoid fauna) and north of Mexico. Prior to Delvare (1992), C. Necremnus tidius (Walker) (1 9, 5 33; 0.5% torvina was consistently misidentified as of the parasitoid fauna). Although E. Conura side (Walker). Carlson et al. (1951) glaucus was known from Florida and Texas reported that a specimen of C. side was (Noyes 2002), its association with C. ob- reared from C. obstrictus in California. We strictus in Georgia represents a new state were unable to locate this specimen to distribution record and a possible new host confirm the identification, but it is possible record. The only other reported host for E. that it is conspecific with the Georgia glaucus is Epiblema obfuscana (Dyar) (Lepi- species because C. torvina is transcontinen- doptera: Tortricidae) (Yoshimoto 1971). tal in North America (Delvare 1992, Noyes Dosdall et al. (in press) reported a second 2002). Because of the confusion in names Euderus species, E. albitarsis (Zetterstedt), prior to Delvare (1992), the list of published as an incidental parasitoid of C. obstrictus distribution and host records given for C. in Alberta, but this association was also side by Noyes (2002) certainly contains obtained by mass-rearing seedpods. Eu- many records that actually refer to C. derus glaucus and E. albitarsis are differen- torvina. Based on previous name usage, of tiated in Yoshimoto (1971), though prob- those Curculionidae listed as hosts of C. lems remain in species recognition within side by Noyes (2002), the record of the the genus. cotton boll weevil, Anthonomus grandis Necremnus tidius is a comparatively com- (Boheman), probably does refer to C. side, mon parasitoid of C. obstrictus in western North but it misidentified as whereas the records of Rhynchaenus palli- America, was cornis (Say) and Hypera spp. likely refer to N. duplicatus Gahan prior to Gibson et al. C. torvina. Because of variability in the (2005), who differentiated and illustrated colour pattern features given by Delvare the species. The specimens from Georgia in (1992), females of C. torvina can be easily represent the first record of the species misidentified as Conura albifrons (Walsh), eastern North America. another transcontinental species that Dos- Eupelmidae dall et al. (in press) reported parasitizing C. obstrictus in Alberta. Females of both Two and of genera— species Eupelmidae species have paramedial yellow marks on were reared Eupelmus (Eupelmus) cyani- the first gastral tergum, but in female C. ceps Ashmead (2 99, 2 33; 0.4% of the torvina the distance between the marks is, parasitoid fauna) and Brasema allynii at most, only about equal to the length of (French) n. comb, (from Eupelmus Dalman) a mark (Fig. 2). In female C. albifrons the (1 9, 4 Jo; 0.5% of the parasitoid fauna). separation between the marks is at least Brasema Cameron is unrevised for the similar to the width of a mark, if not region, but there are about 25 described conspicuously greater. Males of the two species in North America north of Mexico. 194 Journal of Hymenoptera Research

Most of the species are currently misclassi- correctly classified to genus. Eupelmus is fied in Eupelmus (Gibson 1995). Gahan unrevised for the region, but there are (1933) described and partly illustrated both about 15 described species of E. (Eupelmus) of the sexes of B. allynii as a parasitoid in North America north of Mexico. Eu- destructor Hessian , Mayetiola (Say) (Dip- pelmus cyaniceps belongs to the urozonus tera: Cecidomyiidae). Phillips and Poos species-group sensu Gibson (1995). Hunter also both a dorsal and (1921) provided and Pierce (1912, pi. XVIII, f) provided lateral habitus of the female, and for both a dorsal habitus drawing of the female the colour of the sexes illustrated pattern when they recorded the species as a para- fea- legs, important species-recognition sitoid of A. grandis. A species revision of E. when described the immature tures, they (Eupelmus) that includes evaluation of size- B. of the stages of allynii as a parasitoid correlated and host-induced variation is wheat Tetramesia tritici jointworm, (Fitch) necessary to confidently characterize spe- The sexes of (Hymenoptera: Eurytomidae). cies limits within the subgenus. However, are (Gib- Eupelminae strongly dimorphic females of £. cyaniceps are differentiated son 1995), but the more di- important from most other regional species of the features of B. females in- agnostic allynii subgenus by the following combination of clude: head and mesosoma variably brown features: macropterous, the forewing hya- or dark with metallic green luster, scrobal line and with a linea calva (Fig. 7, bb), depression finely coriaceous and quite scape dark, mesosoma dark with metallic shiny, lower face with relatively sparse green luster, and ovipositor sheaths ex- and only inconspicuously lanceolate white tending for a distance at least equal to two- setae, mesonotum finely coriaceous, and thirds length of the metatibia and with middle legs entirely or largely yellow a medial white band (Fig. 7). Because of coxae (mesofemur and tibia often beyond extreme (cf. Figs 7, 8), with some light brown infusion but meso- species recognition in Eupelmus is based femur not extensively dark). Males of B. almost entirely on females; males are not allynii are in part diagnosed within Brasema characterized for most species. The fea- by a clavate flagellum with very short and tures provided in the key for males of E. inconspicuous setae (Fig. 6), head and cyaniceps are of family and generic level. mesosoma metallic green, head with only Eupelmus cyaniceps is a primary or faculta- very slightly lanceolate and comparatively tive of hosts in concealed sparse white setae, and legs with all femora hyperparasitoid situations. The from C. obstrictus re- yellow (Fig. 6). Brasema allynii is trans- rearing a new host record, but continental in North America and a po- presents putative listed 17 other in 11 lyphagous primary or facultative hyper- Noyes (2002) species different of Curculionidae as of parasitoid of hosts in concealed situations. genera part 65 host records in 20 families of Noyes (2002) listed 58 host species in 22 Coleoptera, and families of Coleoptera, Diptera, , Diptera, Hymenoptera Lepidoptera. The Hymenoptera, and Lepidoptera, though only eupelmid previously reported the putative host record of C. obstrictus is as a parasitoid of C. obstrictus is Eupelmus the first for Curculionidae. (Macroneura) vesicularis (Retzius) from Gibson (1995) recognized three subge- Washington state (Hanson et al. 1948) and nera in British Eupelmus, including £. (Episolinde- (McLeod 1953). This lia) Girault and E. (Macroneura) Walker in species likely represents one of the earliest addition to the nominate subgenus. Noyes accidental introductions from Europe to (2002) listed 45 valid species of Eupelmus in North America; only females are known in the Nearctic region, but this includes all North America and they are brachypterous three subgenera and several species in- (Gibson 1990). Volume 15, Number 2, 2006 195

Eurytomidae sitoids or as hyperparasitoids. It is beyond the scoP e of this stud to assess the One species of Eurytomidae was reared y of species-groups or —Eurytoma tylodermatis Ashmead, which monophyly species limits in but at least the four comprised 25 specimens (8 99, 17 f$) or Eurytoma, listed above have the 2.2% of the parasitoid fauna. The Georgia Eurytoma species setose lateral to the rearing represents a new host record for E. propodeum densely foramen, and the tylodermatis, but a previously unidentified propodeal petiole

Fi 40 ' t} has one dorsomedial and species of Eurytoma was also reared in most ( §- P ) two anterolateral Furthermore, surveys of the parasitoid complex of C. processes. the anterior of the first obstrictus in western North America (Douc- margin gastral is and ette 1948, Hanson et al. 1948, McLeod 1953, tergum (Fig. 40, Gtl) emarginate and the is Dosdall et al. in press). Examination of depressed medially, tergum voucher and additional reared material deeply depressed anterolateral^, to ac- the from Alberta, , , commodate processes of the petiole when the is raised. Oregon, and Washington indicates the gaster Bugbee's (1967) ke to differentiates E. western species is also £. tylodermatis. Y species tylodermatis, in the of the Noyes (2002) listed 19 other curculionid part, by stating sculpturing fourth extends over the species in 11 genera as part of 46 reported gastral tergum dorsal surface at least the host species in 14 families of Coleoptera, narrowly along anterior The Diptera, Hymenoptera and Lepidoptera. margin (couplet 20). species states that the Over 90 nominal species of Eurytoma are description, however, sculp- ture of the fourth is ventro- known from the Nearctic region (Noyes tergum heavy continues for about 2002). Bugbee (1967) revised the North laterally, dorsally one- half to two-thirds of the and then American species, but species recognition surface, fades out that is remains extremely difficult because of so the dorsal surface smooth and The extent of variability of the morphological features shiny. sculptur- on the fourth to he used to differentiate species, and be- ing gastral tergum appears cause sexual be variable in of and the dimorphism (cf. Figs 3, 4) species Eurytoma, presents difficulties in recognizing conspe- appearance is partly affected by telescop- cific sexes. Rearing is necessary to make ing of the terga. Features that can be used the association, and one sex of several in combination to differentiate £. tyloder- species of Eurytoma remains undescribed. matis from similar species include the malar Bugbee (1967) examined over 4,000 speci- space lacking an alveolate boss (a mens originally identified as E. tylodermatis slightly raised area), the ventrolateral in the USNM and stated that "even this margin of the scrobes (Fig. 39, vis) being number was not enough to give an produced anteriorly and reflexed poster- adequate picture of the geographical dis- iorly, and the median channel of the tribution, or the range of variation of propodeum being distinct and defined several species in the complex" (Bugbee laterally by carinae formed by longitudi- 1967, p. 492). He keyed E. tylodermatis as nally aligned crenulae (Fig. 40). one of 48 species of his "tylodermatis A single species of Eurytoma, E. curculio- complex" and considered the species to num Mayr, has also been reported as reared be most closely related to £. bolteri Riley, E. from C. obstrictus in Europe (Dmoch 1975). diastrophi Walsh, and E. pini Bugbee. He Individuals of E. curculionum have a meso- also stated that the four species were coxal lamella according to Claridge and associated with the larvae probably of Askew (1960, fig. 2), which is absent from and small that live in stems the North American specimens identified of various plants, either as primary para- as E. tylodermatis. Journal of Hymenoptera Research 196

(Wilson and Pteromalidae Rhinocyllus conicus (Frolich) Andres 1986) (Coleoptera: Curculionidae) of Walker, one Three species Lyrcus as a of the alfalfa gall one and, hyperparasitoid, of Mesopolobus Westwood, websteri Felt species midge, Asphondylia (Diptera: of Neocatolaccus Ashmead, and species Cecidomyiidae) (Gahan 1919). Among of Pteromalus what are two species L. maculatus is likely known species of Lyrcus, of the Swederus about 96% its distinctive comprised usually distinguished by in reared parasitoids (Table 1). mesoscutal colour pattern, as described is restricted to the New World. Lyrcus the key. Urbahns (1919, pi. 23A) provided for the Nearctic, The genus is unrevised a dorsal habitus of the female that illus- listed 16 from the in the but Noyes (2002) species trates this colour pattern. Included identifications in this study L. maculatus region. Species material we have identified as of material are based on examination type are about 25 unusually small, more or less American contained in have of the North species brown specimens that either quite excludes the four oldest the the USNM, which obscure blue spots or that lack spots to from the Nearc- coloured individ- names assigned Lyrcus entirely. The abnormally four tic. Walker (1847) described species uals also have much finer, coriaceous that are now classified the re- collected in Florida mesonotal sculpture rather than and material of in Lyrcus (Noyes 2002) type ticulate sculpture of typical specimens. The Natural intermedi- these species is in History However, some individuals are Burks so Museum, London. Although (1975) ate in both colour pattern and sculpture the names have to and examined the types, yet that a very fine mesonotal sculpture within a blue be placed adequately species brown body colour without regions of Until this is done within with small size. concept Lyrcus. likely is correlated body taxonomic revision, it is was the second a comprehensive Lyrcus perdubius (Girault) or more of the four com- possible that one most commonly reared parasitoid, names a senior 21 or 5.3% Walker represents syno- prising 60 specimens (39 99, >1) used in this nym of a name paper. of the parasitoid fauna. Georgia represents the most for the Lyrcus maculatus (Gahan) was a new state distribution record reared of all the et al. first reared commonly species para- species. Dosdall (in press) in sitoids, comprising 967 specimens (515 99, L. perdubius from canola seedpods of the fauna. The as a of C. 452 (J

whereas females of 1996, revealed the same five chalcid fam- plical carina (Fig. 31), America ilies that have been as reared from the species from western North reported obstrictus in western North America, have the region between the postspiracular C. The six shared in sulcus and plical carina bare. propo- including species apparently the shared deal structure of P. cerealellae is very similar common (Table 1). Of species, L. maculatus was far the most to that of the European species Pteromalus by commonly in semotus (Walker) (Graham 1969, fig. 385), reared parasitoid Georgia, comprising of the fauna. This contrasts to which was reported as reared from C. about 86% America where it to obstrictus in England and Poland (Murchie western North appears of C. and Williams 1998). Boucek (1977) once be only an incidental parasitoid considered the two names conspecific, but obstrictus. Furthermore, two common para- in at least some subsequently (Boucek 1988) re-established sitoids of C. obstrictus parts tidius and P. cerealellae. Among other features, P. of western North America, N. M. line reared as incidental semotus has an entire costal setal (cf. moryoid.es, were only cerealellae was de- in The latter Fig. 50). Pteromalus parasitoids Georgia. rearings scribed from, and until recently was represent the first distribution records of North thought to be a monophagous parasitoid the respective species in eastern If introduced to of, the Angoumois grain moth, Sitotroga America. C. obstrictus was cerealella (Olivier) (Noble 1932). Flanders Georgia from western North America, the (1932) stated that it would also oviposit two parasitoid species may have been into the tuberworm moth, Phthorimaea introduced accidentally at the same time. C. obstric- operculella (Zeller) (Lepidoptera: Gelechii- The most common parasitoid of dae). However, Brower (1991) showed that tus throughout most of western North it effectively parasitized 12 different spe- America, T. lucidus, was not reared in cies in 4 families of Coleoptera (including 3 Georgia despite the statement of Buntin species of Curculionidae) that are pests of (1998) that most of the reared specimens concluded that al- stored products. He consisted of T. perfectus (a misidentification though the species probably prefers S. of T. lucidus prior to Gibson et al. 2005). The cerealella, it is more habitat specific than second and third most commonly reared host specific. species in Georgia were L. perdubius and N. Pteromalus female reared and The second tylodermae, respectively. Because of their that of P. cerealellae are similar in having relative abundance and because neither behind the the forewing dorsally bare has been reported from hosts other than the vein entire length of marginal (Figs 49, Curculionidae (Noyes 2002), both species but differ in those features in the 50), given very likely are parasitoids of C. obstrictus. key. A comprehensive generic revision is However, it remains to be determined to establish the correct necessary species whether they are primary or hyperparasi- name of the unidentified female. The three toids. At least some of the other incidental unidentified males the may represent species, such as C. torvina, B. allynii and E. sex of the unnamed female, based opposite cyaniceps, likely are hyperparasitoids rather on the of a continuous line of presence than primary parasitoids. The rearing of £. setae on the costal cell, but char- species glaucus, B. allynii, E. cyaniceps, L. incertus, P. acteristics of male Pteromalus remain large- cerealellae and the unidentified species of unknown and the males are not in- ly Pteromalus from B. napus seedpods in cluded in the key. Georgia represent new rearing records, but these are at most incidental DISCUSSION parasitoids, if C. obstrictus was the actual host for all the The parasitoid fauna reared from B. species. Buntin (1998) obtained the para- napus seedpods in Georgia, 1994 through sitoids from mass-reared seedpods. The Volume 15, Number 2, 2006 199

Z. 1988. Australasian Chalcidoidea very few Asaphes and Pachyneuron that Boucek, (Hyme- A Revision of Genera of were reared, along with several Aphidiinae noptera). Biosystematic Fourteen Families, with a Reclassification of Species. (Braconidae) also preserved with the ma- C.A.B International Wallingford. 832 pp. terial, show that some mummies Boucek, Z. 1993. New taxa of North American contaminated the seedpods even though Pteromalidae and (Hymenoptera), with notes. Natural 27: an attempt was made to remove these prior Jour)ial of History 1239-1313. to rearing. Likewise, one or more of the Breakey, E. P., R. L. Webster, and E. C. Carlson. 1944. uncommon parasitoid taxa have may The cabbage seed pod weevil, Ceutorhynchus from other undetected insects emerged assimilis, in western Washington. Bulletin of the within or on the pods. For example, the Washington Agricultural Experiment Station 455: 118-119. only other host record for E. glaucus is L.-A. Brodeur, J., Leclerc, M. Fournier, and M. Roy. a lepidopteran. Individual rearing of para- 2001. Cabbage seedpod weevil (Coleoptera: Cur- sitoids dissected from seedpods is neces- culionidae): new pest of canola in northeastern to the host associa- sary definitively prove North America. Fhe Canadian Entomologist 133: tions listed in Table 1, which at present are 709-711. H. 1991. Potential host and only inferred. Brower, J. range perfor- mance of a reportedly monophagous parasitoid, Both L. incertus and L. perdubius have Pteromalus cerealellae (Hymenoptera: Pteromali- been as of reported previously parasitoids dae). Entomological News 102: 231-235. the cotton boll as has also E. weevil, Bugbee, R. E. 1967. Revision of chalcid wasps of genus E. N. in America north of Mexico. cyaniceps, tylodermatis, and tylodermae , Eurytoma Proceedi)igs the United States National Museum 118: 433-552. though not the most commonly reared of Buntin, G. D. 1998. Cabbage seedpod weevil (Ceutor- parasitoid of C. obstrictus in Georgia, L. hynchus assimilis, Paykull) management by trap maculatus. These results suggest that the cropping and its effect on parasitism by Tricho- chalcid fauna C. parasitoid acquired by malus perfectus (Walker) in oilseed rape. Crop obstrictus in any area where it is introduced Protection 17: 299-305. Burks, B. D. 1975. The of Chalcidoidea is partly influenced by what other curcu- species described from North America north of Mexico lionid species occur in the region. If so, the by Francis Walker (Hymenoptera). Bulletin of the parasitoid fauna from eastern Canada and British Museum (Natural History), Entomology 32: the southeastern USA might be expected to 139-170. differ as substantially as between eastern Carcamo, H. A., L. Dosdall, M. Dolinski, O. Olfert, and R. 2001. and western North America. J. Byers. The cabbage seedpod weevil, Ceutorhynchus— obstrictus (Coleoptera: Curculioni- ACKNOWLEDGMENTS dae) a review. Journal of the Entomological Society of British Columbia 98: 201-210. E. W. H. and R. H. Sclaroni. The senior author gratefully acknowledges Eric Carlson, C, Lange Jr, Grissell (USNM) for access to the USNM chalcid 1951. Distribution and control of the cabbage collection and loan of type and other material, without seedpod weevil in California. Journal of Economic which most species identifications would not have Entomology 44: 958-966. F. R. R. been possible. We thank Jennifer Read (CNC) for Claridge, M. and Askew. 1960. Sibling species in the rosae preparing the plates of illustrations used to clarify Eurytoma group (Hym., Eurytomidae). species differentiation, and John Huber and Peter Entomophaga 5: 141-153. Mason (CNC) as well as two anonymous reviewers Delvare, G. 1992. A reclassification of the Chalcidini with for helpful suggestions regarding improving this a checklist of the New World species. manuscript. Memoirs of the American Entomological Institute 53: 119-441.

LITERATURE CITED Dmoch, J. 1975. [Study on the parasites of the cabbage seed weevil (Ceuthorrhyuchus assimilis Payk.) (Co-

H. 1933. of the I. Bigger, J. Parasites sunflower weevil, leoptera, Curculionidae). Species composition Desmoris fulvus Lee, during 1931 and 1932. Journal and economic importance of the larval ectopar- of Economic Entomology 26: 652. asites.] Roczniki Nauk Rolmczych (E) 5: 99-112 Boucek, Z. 1977. A faunistic review of the Yugoslavian (Text in Polish.) Chalcidoidea L. B. P. H. A. (parasitic Hymenoptera). Acta En- Dosdall, M., J. Ulmer, G. A. Gibson, and tomologica Jugoslavia! 13 (suppl.): 3-145. Carcamo. In press. The spatio-temporal distribu- ZOO Journal of Hymenoptera Research

M. W. R. de V. 1969. The Pteromalidae of tion dynamics of the cabbage seedpod weevil, Graham, north-western Chalcidoi- Ceutorhynchus obstrictus (Marsham) (Coleoptera: Europe (Hymenoptera: the British Museum (Natural Curculionidae), and its larval ectoparasitoids in dea). Bulletin of canola in western Canada. Biocontrol Science and Histon/) (Entomology), supplement 16: 1-908. A. E. C. Carlson, E. P. and R. L. Technology. Hanson, J., Breakey, of the Doucette, C. F. 1944. The cabbage seedpod weevil, Webster. 1948. Biology cabbage seedpod weevil in northwestern State Ceutorhynchus assimilis (Payk). Bulletin of the Washington. College 455: Station Washington Agricultural Experiment Station of Washington Agricultural Experimental 123-125. Bulletin 498: 1-15. W. D. and W. D. Pierce. 1912. The Mexican Doucette, C. F. 1948. Field parasitizarion and larval Hunter, cotton-boll weevil: A of the results of mortality of the cabbage seedpod weevil. Journal summary the of this insect to December 31, of Economic Entomology 41: 765-767. investigation up Flanders, S. E. 1932. Recent developments in Tricho- 1911. United States Department of Agriculture, Bureau Bulletin 114: 1-188. gramma production (abstract). The review of of Entomology 19: L. M. and P. G. Mason. 2002. Applied Entomology. Series A: Agricultural Kuhlmann, U., Dosdall, 95-97. 11. Ceutorhynchus obstrictus (Marsham), cabbage weevil Gahan, A. B. 1914. Descriptions of new genera and seedpod (Coleoptera: Curculionidae). Pp. 52-58 in: P. G. and T. Huber, eds. species, with notes on parasitic Hymenoptera. Mason, J. Control in 1981- Proceeding of the United States National Museum 48: Biological Programmes Canada, 155-168. 2000. CABI Publishing, Wallingford. Gahan, A. B. 1919. Some chalcid-wasps reared from Mason, P. G., T. Baute, O. Olfert, and M. Roy. 2004 cecidomyid galls. Annals of the Entomological (2003). Cabbage seedpod weevil, Ceutorhynchus obstrictus Society of America 12: 159-170. (Marsham) (Coleoptera: Curculionidae) Gahan, A. B. 1933. The serphoid and chalcid parasites in Ontario and Quebec. Journal of the Entomological Ontario 134: 107-113. of the Hessian fly. United States Department of Society of H. 1953. Notes on the Agriculture Miscellaneous Publication 174: 1-147. McLeod, J. cabbage seedpod Gibson, G. A. P. 1990. Revision of the genus Macro- weevil, Ceutorhynchus assimilis (Payk.) (Coleop- neura Walker in America north of Mexico tera: Curculionidae), and its parasites. Proceedings (Hymenoptera: Eupelmidae). The Canadian Ento- of the Entomological Society of British Columbia 49: mologist 122: 837-873. 11-18. Gibson, G. A. P. 1995. Parasitic wasps of the subfamily Murchie, A. K. and I. H. Williams. 1998. A bibliogra- Eupelminae: classification and revision of world phy of the parasitoids of the cabbage seed weevil genera (Hymenoptera: Chalcidoidea, Eupelmi- (Ceutorhynchus assimilis Payk.). Bulletin. Section dae). Memoirs on Entomology, International 5: Regionale Quest Palaearctiquc, Organisation Inter- 1-421. nationale de Lutte Biologique 21: 163-169. Gibson, G. A. P. 1997. Chapter 2. Morphology and Noble, N. S. 1932. Studies of Habrocytus cerealellae in: A. T. terminology. Pp. 16-44 Gibson, G. P., J. (Ashmead), a pteromalid parasite of the Angou- B. Annotated to cerealella Huber, and J. Woolley, eds. Keys mois grain moth, Sitotroga (Olivier). the Genera of Nearctic Chalcidoidea (Hymenoptera). University of California Publications in Entomology NRC Research Press, Ottawa. 5: 311-354. G. A. H. B. L. F. Gibson, P., Baur, Ulmer, Dosdall, and Noyes, J. S. 2002. Interactive catalogue of world Muller. 2005. On the misidentification of chalcid Chalcidoidea. 2nd ed. [CD-ROM]. Taxapad, Van- (Hymenoptera: Chalcidoidea) parasitoids of the couver, British Columbia, and The Natural cabbage seedpod weevil (Coleoptera: Curculioni- History Museum, London. in North America. The Canadian dae) Entomologist Phillips, W. J. and F. W. Poos. 1921. Life-history 137: 381-403. studies of three joint worm parasites. Journal of G. A. D. R. and L. Dosdall. Gibson, P., Gillespie, 2006. Agricultural Research 21: 405-^26 + 6 pis. The of in- species Chalcidoidea (Hymenoptera) Pierce, W. D. 1909. Notes on the biology of certain troduced to North America for biological control weevils related to the cotton boll weevil. United of the cabbage seedpod weevil, and the first re- States Department of Agriculture, Bureau of Ento- of Stenomalina covery gracilis (Chalcidoidea: Pter- mology Bulletin 63: 39^14. omalidae). The Canadian Entomologist 138: 285-291. Pierce, W. D., R. A. Cushman, and C. E. Hood. 1912. G. A. T. B. Gibson, P., J. Huber, and J. Woolley eds. The insect enemies of the cotton boll weevil. 1997. Annotated to the Genera Keys of Nearctic United States Department of Agriculture, Bureau of Chalcidoidea (Hymenoptera). NRC Research Press, Entomology Bulletin 100: 1-99. Ottawa. 794 pp. Urbahns, T. D. 1919. Life-history observations on four A. A. 1917. The North American Girault, species of recently described parasites of Bruchophagus The Canadian Entomol- Habrocytus (Chalcid-). funebris. Journal of Agricultural Research 16: 49: 178-181. ogist 165-174 + 2 pis. Volume 15, Number 2, 2006 201

1960. Economic Insect 10: USDA. Cooperative Report 414. ern California. Pan-Pacific Entomologist 62: 329- Walker, F. 1847. Characters of undescribed Chalcidites 332. collected in North America by E. Doubleday, Yoshimoto, C. M. 1971. Revision of the genus Euderus Esq., and now in the British Museum. Annals and of America north of Mexico (Hymenoptera: Natural 19: Magazine of History 392-398. Eulophidae). The Canadian Entomologist 103: A. 1957. Walz, J. Observations on the biologies of some 541-578. hymenopterous parasites of the cabbage seedpod Yunus, C. M. and C. A. Johansen, C. A. 1967. weevil in northern Idaho. Annals Entomological Bionomics of the clover seed weevil, Miccotrogus America 50: 219-220. Society of picirostris (Fabricius) in southeastern Washington Wilson, R. C. and L. A. Andres. 1986. Larval and and adjacent Idaho. Technical Bulletin, Washington pupal parasites of Rhinocyllus conicus (Coleoptera: Agricultural Experiment Station, Washington State Curculionidae) in Carduus nutans in north- University 53: 1-16. of Hymenoptera Research 202 Journal

1-8. 1 and female: Figs 2, Conura torvina, 1, lateral habitus; 2, metasoma, dorsal. 3 and 4, Eurytoma tylodermatis, lateral habitus: male. 5 and Brasema lateral 3, female; 4, 6, allynii, habitus: 5, female; 6, male. 7 and 8, Eupelmus lateral habitus: female cyaniceps, 7, (insert: mesotarsus and apex of mesotibia showing spines); 8, male. = = (Abbreviations: bb bare band, sp spines.) Volume 15, Number 2, 2006 203

ilus cerealellae 9 = Figs 9-18. Antenna, male and female. (Abbreviations: fu funicular segment.) of Hymenoptera Research 204 Journal

L. 21 dorsal female: 21, Figs 19-26. 19, Lyrcus perdubius, head. 20, incertus, malar space. and 22, habitus, Mesopolobus moryoides, 22, L. maculatus. 23-25, mesosoma, lateral: 23, L. maculatus; 24, L. perdubius; 25, am = anterior Neocatolaccus tylodermae. 26, Trichomalus perfectus, pronotum and mesoscutum. (Abbreviations: = margin of metapleuron, md malar depression.) Volume 15, Number 2, 2006 205

Figs 27-34. 27 and 28, Lyrcus incertus: 27, propodeum; 28, metapleuron. 29-32, propodeum: 29, L. perdubins; 30, Neocatolaccus tylodermae; 31, Pteromalus cerealellae; 32, Pteromalus sp. 33 and 34, scutellar-axillar complex: 33, P. = = = cerealellae; 34, Pteromalus sp. (Abbreviations: am anterior margin of metapleuron, bf basal fovea, nuc = = nucha, pc plical carina, pnl propodeal panel.) Journal of Hymenoptera Research 206

WJffMfl'''

"'"f'M/ifibfiiiP'**'

36 37 and Figs 35^42. 35 and 36, Chlorocytus sp., propodeum: 35, female, posterior; male; posterolateral. 38, Trichomalus lucidus, propodeum: 37, female, posterior; 38, male, posterolateral. 39 and 40, Eurytoma tylodermatis female: 39, head, frontolateral; 40, propodeum and base of metasoma, dorsolateral. 41 and 42, forewing: 41, = = = first Necremnus tidius; 42, Euderus glaucus. (Abbreviations: ams admarginal setae, aps adpetiolar strip, Gtl = = = = = gastral tergite, nuc nucha, pc plical carina, pnl propodeal panel, pss postspiracular sulcus, ptl = of petiole, vis ventrolateral margin scrobes.) Volume 15, Number 2, 2006 207

Nebcatolaccus tylodermae (43

Sr\\< Mesopolobus moryoides v*3

J: Lvrcus maculatus (40-'

^-r* ^ bf Trichoma lus lucidus 4J

-^ s. »* v

m&c-'T, Chlorocytus sp. (48

"~ "* - > J?-.•v>" . » N

Pteromalus cerealellae (49.

= = = Figs 43-50. Forewing, female. (Abbreviations: ams admarginal setae, bf basal fold, cc costal cell, mv = marginal vein, sv stigmal vein.) J. HYM. RES. Vol. 15(2), 2006, pp. 208-231

New York A Survey of the Bees of the Black Rock Forest Preserve, (Hymenoptera: Apoidea)

Valerie Giles and John S. Ascher*

Division of Zoology American Museum of Natural History, ,h Central Park West at 79 Street, New York, NY 10024, USA ^Address for correspondence: [email protected]

the fauna of Black Rock Abstract—We present the results of a survey of Forest, Orange with more limited data for other County, New York, USA. The survey focused on bees, gathered in 2003 with nets and bowls incidentally collected groups such as apoid and vespid wasps. Surveys and 23 crabronid recorded 144 bee species (26 genera), 22 vespid species (9 genera) species (12 are detailed. checklist of the bee fauna of the BRF is genera). Noteworthy records A preliminary of York selected sites within the state, and of presented and discussed in relation to that New State, the northeastern USA as a whole. The cleptoparasitic species fattigi Mitchell, Sphecodes and the johnsonii Lovell, and () michiganense (Mitchell), oligolectic species Osmia (Melanosmia) inermis (Zetterstedt) are newly recorded from New York State. Ecological and are patterns pertaining to , type, specialization, parasitism, phenology, summarized and discussed, as are the efficacies of different collecting methods. The net collected vs. and in the number sample was richer than the bowl trapped sample in total bee species (117 113) of unique species (29, 20.4% vs. 25, 17.6%).

Key words. —native bees, Bombus, , , invertebrate survey, invertebrate biodiversity, bowl trap, pan trap, trap nest, Black Rock Forest, pollination

Bees (Hymenoptera: Apoidea) are the T. Griswold, unpublished; cf. Hurd 1979). single most important of Published data on the distribution of bees both native and cultivated vegetation in in New York State (NY) is limited (e.g., most habitats worldwide (Williams et al. Leonard 1928), but Ascher (unpublished 2001, Michener 2000). The mutualist re- information) has compiled a list of 423 lationship between bees and plants forms species known from New York, of which a key process in the maintenance of both 405 are native to North America, local biological diversity and agricultural Single-site inventories of poorly known productivity. As primary pollinators, bees invertebrate groups have the potential to provide a vital ecosystem service, affecting establish useful quantitative baseline esti- the integrity of ecological communities as mates of local biodiversity, as well as to a whole, including the health of humans help illuminate large-scale distributional (Williams et al. 2001, Nilsson 2000, Cane patterns within those groups. Such esti- and Tepedino 2001). Despite this ecological mates can prove useful in testing hypoth- importance, our understanding of some eses arising from practices as diverse as basic aspects of bee biology, including theoretical biogeography and conservation species level distributional patterns, re- planning. In addition, geo-referenced spec- mains are incomplete. There about 20,000 imen data are amenable to re-analysis and species of bees worldwide (Michener 2000) comparison with related data sets in the and approximately 3500 described species context of regional studies of biodiversity occur in America north of Mexico (JSA and across diverse groups. Finally, such in- Volume 15, Number 2, 2006 209

ventories help to address the need for (AMNH) and creation of a synoptic (taxo- natural history information that is crucial nomic reference) collection to be housed at for understanding community-level eco- BRF. Ecological data were also gathered, logical patterns (e.g. phenological patterns, such as abundance of bees across the host associations, habitat use, etc). season and on selected host plants, and Black Rock Forest (BRF) harbors a variety the efficacies of various collecting methods of distinct habitat types, many of which are (net collecting vs. trapping with bowls of typical of the larger Hudson Valley Region, three different colors vs. trap nesting) were and is managed in part as a long-term tested. research preserve. Because the landscape SITE matrix surrounding BRF is under increas- STUDY AND METHODS from land conversion and ing pressure The Black Rock Forest (BRF) is a 1520 habitat a of the bee degradation, survey hectare preserve and research facility lo- communities of BRF while surrounding cated in Orange County, New York (Lat. habitats are still intact should relatively 41.42267, Long. 74.03039), ca. 50 miles a valuable basis for future com- provide north of New York City (NYC). The BRF with a of other sites across parison variety is situated within the highest portion of the a of and scales. The range spatial temporal Hudson Highlands. The terrain comprising effects of environmental on bee change the preserve ranges in elevation from about communities remain insufficiently under- 135 m to 446 m. A network of closed stood. relevant studies have been Many canopy dirt roads permits access to within Matheson et al. published (see, e.g., 1996), 1 kilometer of any point within the BRF. but few of these are from eastern North The landscape, both within the forest pre- America. Cane (2005) notes that bees serve, and across several large adjacent a combination of salient "possess unique tracts (including West Point Military Acad- and traits that set foraging nesting together emy), is mostly forested (upland hardwood them from other taxa studied in the apart forests dominated by Quercus spp., Barrin- context of habitat fragmentation". Many ger and Clemants 2003). Other local habitat bees utilize areas for and open foraging types include successional hardwood and benefit nesting, may from forest stands, hemlock coves, chestnut- unlike fragmentation, forest-dwelling woods, red maple swamps, ponds, reser- bees are still songbirds. However, poten- voirs, and marshes. Important habitats for tially vulnerable to habitat change, partic- bees at BRF include small meadows, the loss of their host ularly plants. exposed road edges and reservoir edges, dams, and marshes, where flowering RESEARCH OBJECTIVES shrubs (such as Viburnum spp., Ilex verti- This cillata survey was undertaken with two (L.) A. Gray, Kalmia latifolia L., to primary goals; first, assemble a faunal Clethra alnifolia L., Spiraea spp., Rhododen- list of the bees (and selected aculeate dron spp.), and herbs (such as Veronica, wasps) of BRF to serve as a baseline Polygonum spp., Gnaphalium, spp., inventory of use to both ecologists and and Viola spp.) provide seasonal sources of conservation biologists, and second, to pollen and to bees. In addition, compare BRF data with other bee samples during early spring (April) prior to leaf- in order to shed light on larger-scale out, forest habitats hosted bees attracted to (regional) patterns of bee distributions flowering trees such as Acer rubrum L., and diversity. Second order objectives in- Salix spp. including S. discolor Muhl. and cluded enhanced representation of the Prunus spp. In May, other flowering trees regional bee fauna in the collection of the such as Craetaegus macrosperma Ashe, and American Museum of Natural History other Prunus spp. were important re- 210 Journal of Hymenoptera Research sources for many Andrena and other bee one of three florescent colors: yellow, white or blue. total of 150 were species. The early spring flower Erythro- A traps deployed nium americanum Ker-Gawler flowered on each of 17 survey visits and arrayed in visit. Each transect sparingly during our survey season and ten transects on each 15 of each was visited by relatively few bees. Vacci- consisted of traps (five color), in colors. con- nium species, especially highbush blueber- arrayed alternating Traps a solution of brand blue ry Vaccinium corymbosum L. and lowbush tained Dawn table to 1 blueberry V. august ifloiwn Ait. dominated dishwashing liquid (1 spoon in ten large areas of the forest understory at BRF, gallon tap water) and were placed including dry hillsides, damp forest areas, sites on the ground along transect lines, open forest gaps created by fires, and wet Traps were deployed over a period of marsh edges in association with herba- approximately 1 hour beginning at 0730 hr ceous communities. Beginning in May, and and were in place before 0900 hr on survey continuing into late June, Vaccinium stands visits during which they were used. In- composed of several species were visited dividual traps were placed at approximate- by large concentrations of nectaring and ly one meter apart. Transect sites were "buzz" pollen-collecting bees. Vaccinum chosen opportunistically and included: stamineum L. (Deerberry) was moderately open fields, roadsides, reservoir edges, common in hillside forests. Patches of dams, forest floors and stone outcroppings Lysimachia were noted. throughout the BRF property. At the close The survey season during the spring and of each survey visit the traps were re- summer of 2003 was generally wet and trieved during a two-hour period begin- cool in southern New York as confirmed ning at approx. 1600 hr. The contents were by weather data collected at BRF. Above poured through sieves and the recovered average rainfall and below average tern- specimens were transferred to plastic peratures would be expected to depress whirl-packs containing 75% ethyl alcohol, bee numbers and collecting success. Locality data and bowl trap color labels Sampling schedule. —We conducted bee were recorded, surveys at BRF during 24 days between 31 Hand netting of bees was conducted March 2003 and 16 October 2003. Each between 09:00 and 16:00 during 23 survey survey day began at approximately 0730 hr visits. Collecting by hand-net was under- and was completed generally between taken opportunistically at sites where bees 1800 hr and 1900 hr. Most fieldwork was were thought to be concentrated. Hand conducted on days with predominantly netting was pursued most intensively in sunny skies and warm temperatures. Col- exposed sunny habitats such as fields, road lection sites visited per sampling day and edges, reservoir and marsh edges, where the time at spent each site varied. In many shrubs and herbaceous perennials individual addition, collecting sites were bloom and where bees were most likely to chosen throughout the BRF property op- occur. When bees were captured they were in to the portunistically response presence transferred to cyanide killing jars before of bees or abundance of flowering plants, being stored in vials. Vials were labeled UTM coordinates were recorded for all and placed in a cooler for transfer to the sites where bees were collected. laboratory. Sampling methods.—We collected bees at Twenty wooden 'Binderboard' brand BRF 3 using principal methods: colored trap- were deployed for the duration plastic pan (or bowl) traps, hand-held of the survey beginning on 27 May 2003. insect and nets, wooden trap nests. Bowl Ten trap-nests consisted of a wooden block were made from 6 traps oz. plastic Solo bearing 39 holes, each measuring 5.5 mm brand bowls that were spray-painted with diameter, and a depth of 10 cm. The Volume 15, Number 2, 2006 211

10 remaining trap-nests were similar in tions from across NY and from nearby other respects, but each bore 21 holes states such as Connecticut were available, measuring 5 mm in diameter, drilled to including material collected by the authors, a depth of 16 cm. Each hole was lined with P. Gambino, S. W. T. Batra, K. N. Mag- a kraft paper tube to facilitate recovery of nacca, B. N. Danforth, D. L. Wagner, R. G. specimens. Trap-nests were mounted in Goelet and their associates. All discussion sets of at 10 sites across two, dispersed the of the past and present status and life BRF Each nest was from property. hung histories of bee species found at BRF is a tree limb 1.5 meters above approximately based, in part, on these historical and the ground with the holes oriented to face recent collections and the literature in south. nests were checked Trap periodical- addition to the sample obtained during to determine if ly any Hymenoptera had the survey of BRF. Totals cited for "south- inhabited the holes and to ensure that they ern New York" are for the area encom- were intact and undisturbed. nests Trap passing New York City (NYC), Long were retrieved from BRF on 20 March 2004, Island, and all counties adjacent to Orange and each was examined in the trap-nest County (i.e. the southeast portion of the for evidence of laboratory occupation by state north to Sullivan, Ulster, and Dutch- Hymenoptera. ess Counties). Specimens were sorted, mounted, and identified to the authors species by (initial- RESULTS AND DISCUSSION ly sorted by VG; species determinations The collected and then made or confirmed by JSA) except the survey individually databased 6,542 bee more difficult metallic Lasioglossum (Dialic- specimens represent- 144 of which 138 are native and tus) females, determined by S. Droege, ing species, six are exotic 26 and , determined by J. M. Carpenter, (Appendix 1), genera, five and the more difficult , de- families (Table 1). Of these, several records detailed below termined by P. Gambino. S. Droege made, represent notable confirmed, and revised identifications for range extensions, the only recent known numerous , and L. Day made and collection of a species in NY, or otherwise confirmed identifications of Bombus sauder- fill gaps in the known distributions of New York State soni and B. vagnns. Vouchers are deposited bees. Other aculeate specimens in the collection of the American Museum incidentally sampled included 22 vespid of Natural History (AMNH). A synoptic species (9 genera), 24 crabronid species (13 collection is housed at the BRF research genera), and 2 species of Isodontia (Spheci- dae sensu facility. Duplicate specimens were dis- stricto) (Appendix 2). Only a sin- an Osmia persed to various bee specialists. gle bee, cornifrons female, em- Comparative data on the North Ameri- erged from our trap nests. These were can bee fauna as a whole, and on the fauna occupied primarily by eumenine (Vespi- of New York State (NY), and of particular dae) and Trxfpoxylon {) (- areas within NY, were compiled by JSA nidae) wasps and were not examined in based on study of relevant taxonomic and detail due to the—lack of bees. faunistic literature and study of historical Species totals. Of the 144 bee species insect collections, especially those housed found at BRF we regard 138 as native to at: AMNH; Cornell University (CUIC); North America (Appendix 1). These are New York State Museum; National Muse- discussed first followed by the six species um of Natural History; University of known or suspected to have been intro- Connecticut Insect Collections, Storrs; and duced deliberately or accidentally to North Parker Gambino's personal collection (af- America from Europe or East Asia. filiated with the — AMNH). Recent collec- Native bee species. 212 Journal of Hymenoptera Research

Table 1. Summary of the number of described bee species for each genus known from New York State, with totals for the Eastern USA (sensu Mitchell 1960, 1962), New York State (NY), southern NY as defined in the text

(SNY), Black Rock Forest (BRF), New York City (NYC; i.e. the five boroughs), and Ithaca (within city limits; many additional species are known from the Ithaca vicinity in Tompkins Co.). The number of species not native to North America (i.e. adventive and introduced species) is given in parentheses following the total number of species. *No recent records.

Superfamily Apoidea: Clade Anthophila (Bees) EUSA NYS SNY BRF NYC Ithaca

Family : Subfamily Colletinae 35 17 Subfamily Hylaeinae: Tribe Hylaeini Hi/laeits 24(3) 14(2) 8(2) 8(3) 10(2)

Family

Subfamily : Tribe Rophitini Subfamily Tribe 3 Tribe Caenohalictini Tribe Subtribe Sphecodina Sphecodes 34 25 15 Subtribe Halictina 5 Lasioglossum 114(2)

Family Subfamily : Tribe Andrenini Andrcna 125(1) Subfamily Tribe Tribe Tribe Panurgini Subtribe Panurginina Subtribe Perditina

Family Subfamily Tribe Macropidini Mncropis 4 Tribe Melittini 3

Family Subfamily Tribe Volume 15, Number 2, 2006 213

Table 1. Continued.

Superfamily Apoidea: Clade Anthophila (Bees) of Hymenoptera Research 214 Journal

small both distributed across NY, Our sample of , including below), widely and two L. acuminatum and L. series of only two ubiquitous Hylaeus spp., by species, to eastern NY H. mesillae and H. modestus, is also im- fuscipenne, restricted (e.g., absent from the see poverished. At least four additional species Fingerlakes Region; are abundant in nearby Putnam County range maps in McGinley 1986). (JSA and P. Gambino, unpublished) and Two widely distributed, pollen-general- should occur at BRF. ist species of carinate Lasioglossum (Evy- laeus) (sensu Michener 2000) were collected Halictidae in addition to (L cinctipes and L. quebecense) the more localized L. nelumbonis. All three augochlorine species known (Evylaeus) associated from NY are numerous at BRF. The The latter seems to be strongly with flowers. In our abundance of in our aquatic emergent numerous L. nelumbonis were col- samples probably reflects the local avail- study, in which this lected in along a causeway ability of rotting logs species pan traps placed in which abun- excavates its nests. The most abundant bee bisecting Jim's Pond, grew dant odorata). in our sample, (1,222 Nymphaeaceae (Nymphaea data that L. nelum- individuals collected) is a eusocial, Museum label suggest bonis be a of ground-nesting species that is numerous may pollen-specialist Nym- across most of eastern North America, phaeaceae and /or Nelumbonaceae, but direct observations of Populations of this species from northeast- pollen collecting ern USA and southern Canada were behavior by this species have not yet been the of and known as A. striata (Provancher) prior to made due to difficulty observing recent synonymy with A. aurata in Coelho's collecting bees on aquatic vegetation. individuals (2004) revision of Augochlorella. Lasioglossum (Dialictus) were, abundant in our The two most common species of Aga- as expected, particularly These were found to postemon in NY (A. sericeus and A. virescens) bowl samples. belong were collected, but two species present to 22 identified species (additional, poorly be included our more locally in southern NY [A. texanus known species may among most Cresson and A. splendens (Lepeletier)] were undetermined metallic Dialictus; not found. Absence of A. splendens is not males of this subgenus were not deter- surprising, as this species seems to be mined) including two socially parasitic associated with sandy nesting substrates. species (Paralictus sensu Mitchell 1960) Dialictus The cleptoparasitic genus Sphecodes was and two black, non-metallic spe- S. S. cies = noncarinate see Michener represented by S. galerus, levis, fattigi, ( Evylaeus; of and S. johnsonii, the last two recorded for 2000). Among the identified species the first time in NY (JSA has also collected metallic, pollen-collecting Dialictus collect- S. johnsonii in Fairfield County, Connecti- ed in BRF (i.e., Dialictus sensu Mitchell cut, new state record). Two additional 1960) the wood-nesting species L. coeru- Sphecodes species, S. atlantis Mitchell and leum, L. cressonii, and L. oblongum were each S. dichrous Smith, not found at BRF were numerous. Other notable metallic, pollen- collected elsewhere in Orange County in collecting Dialictus species include two 1962 (Tuxedo Park vicinity; AMNH). species typical of northern forests (L. Three Halictus species ubiquitous in the nigroviride and L. versans), a distinctive eastern United States were found in good species often found in sand pits (L. hctero- numbers, but the more precinctive H. gnathum), and an infrequently recorded parallelus Say was not collected. species (due in part to identification diffi- Lasioglossum sensu stricto was represent- culties) previously known in NY from a few ed by L. coriaceum and L. leucozonium specimens collected in or near the lower for (regarded the first time as exotic, see Hudson River Valley (L. cattellae). Single 215 Volume 15, Number 2, 2006

of A. in females of the two socially parasitic Dia- The long malar space rufosignata, to its sister lictus species were collected, one of which, comparison likely species re- L. michiganense, has previously been (LaBerge 1980), may nectar from corded in the literature solely from Mitch- be an adaptation to collecting ell's (1960) unique holotype female, col- the bell-shaped corollas of Vaccinium spe- Andrena lected in Wayne County, Michigan, in 1940. cies. Another oligolectic collected, to be a Rhodo- Our single female specimen and another A. cornelli, is now thought female collected 30 June 2004 at the inlet to dendron specialist based on field observa- data for Lake Myosotis, Edmund Niles Huyck Pre- tions by JSA in Virginia, label serve, Rensslaerville, Albany County, NY, newly identified material in museum col- C. are the first records and the by JSA and J. Daley lections, widely spaced scopal outside of Michigan. Despite a lack of hairs of this species that can be considered published records, this species is probably an adaptation that holds Rhododendron widely distributed across the northeastern pollen connected by viscin threads USA. It was recently found in Maryland (S. (Ascher, unpublished; cf. LaBerge 1980). of Droege, pers. comm.; new state record) and Andrena violae, an oligolege Viola, pos- southern Ontario, Canada (L. Packer, pers. sesses elongate maxillary palpi used to comm; new Canadian record). The male of extract nectar from its host. Viola is other- this species remains unknown. The female wise most often visited by long-tongued of L. michiganense possesses a conspicuous, bees such as Osmia that are able to reach its inner, subapical mandibular tooth, where- concealed nectaries. Andrena violae is nu- as the mandibles of other parasitic female merous across much of the eastern United inner of L. (Dialictus) are simple (i.e., lack an States, excluding the colder areas the in tooth) with elongate slender tips. The other northeast, but was previously known socially parasitic Dialictns found at BRF, L. NY solely from a single male collected at in , has recently been found NYC Van Natta's Dam, Six Mile Creek, Ithaca, in Central Park, Prospect Park (JSA, new Tompkins County, 2 May 1936 (specimen records), and the Bronx (collected by P. examined, CUIC). This species was not Gambino). represented among collections made on Viola at this site and elsewhere in the Andrenidae Fingerlakes Region by JSA during 1997- so evidence of its in NY at Our BRF sample included 40 species of 2002, persistence occurrence is welcome, Andrena but is still far from complete, as an a new station of Other Andrena at BRF include A. additional 32 species known from southern oligolectic dan- NY were not recorded. Our sample was krigiana, a specialist of Krigia (dwarf and A. a of rich in vernal species characteristic of delion), fragilis, specialist northeastern forests such as A. imitatrix Cornus (Svida). Three Andrena specialists and Aster were and A. nivalis. Species associated with of Solidago (tribe Astereae) A. A. and A. blueberry were particularly well represent- found, hirticincta, nubecula, not its sister A. ed including the Vaccininm oligoleges simplex (but species, placata A. which has been collected (pollen specialists) A. hradleyi and Mitchell, recently in as was the Carolina, and the polylectic A. carlini and Putnam Co., NY), panurgine Astereae andre- A. rufosignata. The last species is abundant specialist Pseudopanurgus (but under-collected; cf. LaBerge 1980) in noides [we recognize genus Pseudopanurgus of northern blueberry bogs, and evidently in the broad sense Mitchell, 1960, reaches the southern limits of its range at including {Heterosarus) and P. or near Black Rock Forest, as it is unknown (Pterosarus) of Michener, 2000]. Andrena is a of Brassicaceae that from New York City, Long Island, and arabis specialist from of in- elsewhere along the mid-Atlantic coast, may actually benefit spread 216 Journal of Hymenoptera Research

O. vasive Garlic Mustard Alliaria petiolata ated O. bucephala and pumila. We of three (Bieb.) Cavara & Grande. Many species collected single specimens species absent in other recent that regularly collect pollen from rosaceous that are scarce or trees and shrubs, and are known or collections from New York State, O. collin- O. O. inermis. The last suspected to be important pollinators of siae, felti, and species, of Ericaceae S. apples, were collected in good numbers, a probable oligolege (M. including A. miserabilis, A. (Melandrena) Arduser, pers. comm.) previously unre- has also identified spp., and A. (Trachandrena) species. An- ported from NY, been drena (Trachandrena) nitda was numerous at among recent samples of bees from the BRF, which is near the northern edge of its Adirondacks (JSA and W. L. Romey, new range in NY (see map in LaBerge 1973). record). Another Osmia species, O. dis- tincta, has been found elsewhere in NY Melittidae Hill (e.g., South Swamp, Ithaca, Tompkins and the Hudson and Although a deliberate effort was ex- County; along River) pended to locate and collect from Vaccinum in Pennsylvania to visit Penstemon, includ- P. Nutt. The tuft of curved stamirieum, the host plant of Mclitta eick- ing digitalis hairs on the ocellar of this worti Snelling and Stage (1995), this re- region species would seem to be an for cently described species was not recorded adaptation from during our survey. However, it has been collecting pollen Penstemon, although collected nearby in Putnam County by P. O. distincta is apparently not a strict oligo- of this Gambino, as has O. virga Sandhouse, lege genus (M. Arduser, pers. another poorly known oligolege of Erica- comm.). ceae (see Cane et al. 1985; they recorded O. Our sample of native and associated is im- virga, as O. "felti", collecting "surprisingly cleptoparasites pure" loads of Deerberry pollen; this poverished, perhaps reflecting the ineffi- species also uses other ericaceous hosts, ciency of bowl traps for capturing these M. Arduser pers. comm.). No strong-flying species (although Megachi- were collected in this study although their lini can be trapped in numbers in bowls of host plant Lysimachia was present. appropriate color, S. Droege, pers. comm.). The species captured are widely distribut- Megachilidae ed and numerous across New York, ex- M. which is in Native megachilid species collected at cepting montivaga, known the state from a few collections in BRF included the cleptoparasites southern (Dolichostelis) louisae (one female) and Stelis NY (e.g., recently collected at Edmund Niles in (Stelis) nitida (one female). The former is Huyck Preserve Albany County). A of this a colorfully marked parasite of native resin report species from Ithaca bees in subgenus Megachile (Chelosto- (Leonard 1928) is based on a misidentified M. inermis. moides), including M. (C.) campamdae (the likely host in NY and New England), which reaches its northern distributional Apidae, excluding bumble bees limits in southern York. Stelis New nitida Large and small carpenter bees were was described in 1878 from specimens represented respectively by Xylocopa virgi- collected in Canada and but there NY, nica (locally very numerous at BRF, but have been few subsequent collections from most uncollected) and by two abundant eastern North America. It is most likely sister species of (Zadontomenis), C. a northern and montane that species calcaratn and C. dupla, that cannot be or parasitizes Osmia, possibly large distinguished in the females. The related Our species. sample of eight native Osmia C. strenna Smith is also common in NY but species includes series of the forest-associ- was not collected. Volume 15, Number 2, 2006 217

All 18 identified species belonging to the size and long flight season (JSA has observed males as late as November ruficornis group of Nomada (=Nomada s.str.) flying in our samples are known or suspected to 10 in Ithaca, NY, a colder locality than BRF) be cleptoparasites of Andrena species, characteristic of this species. Other bumble Commonly encountered species of this bee species encountered include B. per- group at BRF and other forested areas of plexus and B. vagans, both generally nu- the northeastern USA and southeastern merous in New York forests and bogs, and Canada include the large, conspicuous the widely distributed B. bimaculatus and B. B. the species N. luteolbides (a valid species griseocollis. We found few vagans, but distinct from N. sulphurata Smith; see extremely similar (and thus infrequently Schwarz and Gusenleitner 2004) and N. identified) B. Sanderson i was found in maculata, both cleptoparasites of large surprisingly large numbers, including se- Andrcna belonging to the subgenus Mclan- ries of queens, males, and workers. Two drena (Milickzy and Osgood 1995), and N. Bombus ternarius were found. This is bella, a cleptoparasite of A. imitatrix. A new a species of northern affinities found probable host association between N. bella commonly south to the Catskills. Leonard and A. imitatrix was inferred by JSA (new (1928:1031-1032) regarded it as, "Essential- information) based on repeated co-occur- ly a Canadian and northern transition rence of these species at several sites species...", and stated that "the species is of N. bella across several years. Females not found near NYC. (Beq) [indicating J. have been identified (M. Schwarz, pers. Bequaert as the source]." Long-tongued comm.) but remain undescribed. Further bumble bee species belonging to subgenera study of Nomada with bidentate mandibles Fervidobombus [B. fervidus (Fabricius) and B. = Gnathias sensu Mitchell is needed Subterraneo- ( 1962) pensylvanicus (Degeer)] and N. to clarify separation of N. bella from bombus (B. borealis Kirby) that frequently similar ovata, N. lepida, and other species, visit clovers (especially Trifolium) were not We collected a single male Nomada austra- collected. Absence of B. fervidus is surpris- is of the three in es, which one species ing, but B. pensylvanicus has been scarce to the = Cen- belonging erigeronis group ( NY in recent years and is no longer, "An are aestival trias) known from NY. These abundant southern species, common as cleptoparasites of Agapostemon. far north as central NY..." (Leonard was represented by the 1928:1032). Bombus borealis has always been wood-nesting species A. (Clisodon) termina- uncommon in New York State (Leonard is distributed and nu- lis, which widely 1928), and is generally absent from de- merous in northern and montane forests it is from the veloped areas (e.g., unknown Siberia to eastern Canada in from [Davydova city of Ithaca, NY, but occurs nearby these authors distin- and Pesenko 2002; countryside). guished the Holarctic A. terminalis from the Absence of Bombus (Bombus) affiuis in our Palearctic A. furcata (Panzer)]. sample of 1261+ bumble bee individuals is troubling because this species is well Bumble bees represented in historical collections from Black Rock Forest is a favorable habitat the northeastern United States, and is abundant for bumble bees, and certain species were expected to be "...moderately of the found in large numbers, especially Bombus in the eastern to southern parts [New 1928: How- (Pyrobombus) impatiens and its social para- York] State..." (Leonard 1031). has site B. () citrinus (also known to ever, this species recently disappeared from Ithaca and the attack other Bombus species). Large num- from New York (e.g., and else- bers of B. impatiens in our late season NYC area, JSA, unpublished) The samples reflect the unusually large colony where (L. Day, pers. comm.). regional Hymenoptera Research 218 Journal of

is coincident habitats in the eastern United States disappearance of B. affinis NYC) where it could be classified as with an abrupt decline in B. (Bombus) to the point We collected 66 from terricola Kirby at Ithaca NY (Ascher, un- invasive. specimens on or around native and in published), and elsewhere (L. Day, pers. vegetation and one female from a comm.), as well as the extirpation of the bowls, emerged trap is nest. to orchards should closely related B. (Bombus) occidental Non-specificity Greene from the San Francisco Bay Area not be surprising as Osmia (Osmia) species related and elsewhere in western North America, such as O. cornifrons and the closely native O. are not and the precipitous decline of the endan- species lignaria polylectic, of fruit In areas near gered B. (Bombus) franklini from its excep- specialists crops. Ore- where O. were re- tionally restricted range in southern cornifrons deliberately Patuxent National Wildlife gon and northern California (Thorp 2005). leased (e.g., B. and of all North see Cane 2003), a similar Populations of affinis, Refuge, very Asian Osmia taurus Smith American species of subgenus Bombus, and species, (Osmia) B. has been found to be established. This their obligate social parasites [e.g., has also been found in (Psitin/rus) ashtoni; a queen of this species species Huntingdon in south-central was collected at BRF on June 13 1988, by J. County Pennsylvania (VG, in G. Rozen], should be carefully monitored, new data), but not yet NY. 10 as parasitism by Nosema and other para- We collected , sites introduced and spread via the green- a species native to Europe and only re- house trade in Bombus colonies poses cently detected in North America (Hoebeke a potentially severe threat to their survival, and Wheeler 1999). This species is now — in- in the mid-Atlantic Introduced bee species. Our samples abundant States, New eluded numerous individuals of certain York, and southern New England, usually exotic bee species that have become estab- in association with favored host plants such lished and locally invasive in eastern North as Lotus corniculata, a weed generally America beginning in the 1990's. distributed in waste places such as road- Megacliile sculpturalis, a giant resin bee sides and abandoned lots, and Sedum. native to northeastern Asia, was first The halictine species Lasioglossum (L.) collected in New York State in 1997 leucozonium has long been present in North (Ascher 2001) and is now widely distrib- America and has therefore been generally uted and locally abundant in the Finger- regarded as native. However, its North lakes Region, and in southern NY, in- American range is restricted to northeast- eluding NYC. Outside of New York, M. ern USA and southeastern Canada and sculpturalis is now quite widely distributed does not include northwestern Canada or and has recently been found in additional (see maps in McGinley 1986). This northeastern states such as Massachusetts distributional pattern, and association of (Martha's Vinyard, P. Gambino pers. this species with introduced weeds such as comm.), Vermont, and New Hampshire Chicorium (), suggests that this (S. Droege, pers. comm.), as predicted by ground-nesting species is adventive from Hinojosa-Diaz et al. (2005). Europe, not native as has been assumed. The horn-faced Osmia comi- Molecular phylogenetic placement of L. frons, native to eastern Asia including leucozonium and L. zonulum (Smith) within Japan, was deliberately introduced by the otherwise exclusively Old World leuco- USDA scientists as a managed zonium species group, and lack of signifi- of apples. After wide distribution and cant genetic differences between Old and release, this has estab- species recently New World samples (see, e.g., Danforth lished in natural and large populations and Ji 2001), further support the idea that urban Manhattan and (e.g., Brooklyn, the occurrence of these species in North Volume 15, Number 2, 2006 219

America is adventive. It is possible that and 1,179 specimens (18.2%) belong to these species were introduced in soil parasitic species (Fig. la). Of the 144 bee carried in ships' ballast as has been species recorded in this study, 116 (80.5%) hypothesized for another ground-nesting are pollen-collecting species and 28 (19.4%) bee species native to Europe and found in are parasitic (Fig. lb). The abundance and our study, . Extensive diversity of parasites reflects a rich fauna sampling of variable molecular markers of vernal Nomada associated with Andrena such as COI is needed to test hypotheses of hosts. The preponderance of females (4321 native vs. adventive origin for bee species vs. 1977 males vs. 245 of unrecorded sex) with Holarctic distributions. Mcgachile cen- corresponds with the large number of tuncularis (L.) may be another early in- workers of eusocial species, including the traduction from Europe, as this species has two most numerous species at BRF. Of the not been recorded in Alaska as would be 6,543 bee specimens collected, 1,222 expected for a species with a naturally (18.7%) were Augochlorella aurata and 845 Holarctic range. (12.9%) were Bombus impatiens. The sample Workers of Apis mellifera (L.) were of 1,113 bumble bees collected was domi- abundant from mid-June and into October nated by B. impatiens (845, 75.9%) and its but were generally not collected. social parasite B. (Psithyrus) citrinus (154, Of the 144 bee species recorded in this 14.1%). study, six (4.2%) are exotic and 138 (95.8%) Seasonal patterns of occurrence can be are native. Of the 6,543 specimens collect- obtained from Appendix 1, which gives ed, 115 (1.7%) belong to exotic species, and extreme dates for BRF (by calendar date), 6,428 (98.2%) belong to native species. NY as a whole (by month), and the entire Wasps and other non-bees. —Our apoid North American range (by month) for each wasp samples include 23 crabronid species species. General patterns include an abun- of Andrena their (12 genera; Appendix 2). Some of these are dance and diversity and generally numerous in forest edge habitats Nomada parasites at forest floor sites prior in New York such as continuus, to leaf-out. At more open sites, seasonal which nests in holes in wood. Other turnover of the bee fauna was apparent, species collected such as leitthostromi with notable peaks of abundance and the and Bia/rtes quadrifasciata are ground-ne- species diversity corresponding with sters that favor more open, often sandy bloom of favored plants such as Vacciniinn Andrena and habitats. Our vespid wasp sample includes in late spring (visited by, e.g., long series of the native Osmia spp.) and Solidago in late summer fuscatus, both sexes of Vespula consobrina, (visited by, e.g., Colletes spp., Andrena a of northern (Canadian and simplex Pseiidopamirgus andrenoides, and Transition Zones) affinities, one individual the workers and males of the dominant of the rather scarce Zethus spinipes, and eusocial species Augochlorella aurata and a variety of eumenines including cavity- Bombus impatiens). Rather few oligolectic 4.5% nesting species found in our trap nests.— bee individuals were captured (292, Ecological and behavioral patterns. Eco- of the total) (Fig. 2a), but these represented of in our logical information (summarized in Ap- a significant number species pendix 1) was compiled for each of the 144 sample (19, 13.2%) (Fig. 2b). bee species from information found in Although soil nesting individuals and catalogs and revisions, primary literature, species predominated in our samples, hive and field observations, including those nesters, wood burrowers, and cavity- made during the BRF survey. nesters were also well represented (Fig. 3a, Of the 6,543 specimens collected, 5,364 b). Cavity-nesting species were numerous as (82.0%) belong to pollen collecting species, relative to the number of individuals, of Hymenoptera Research 220 Journal

1 B of 1A Sociality of Individuals Sociality Species

Parasitic Parasitic Rusocial 18% 21% 21% Subsocial Subsocial 1% Eusocial 54%

Solitary Solitary 27% 57%

2B vs. 2A Oligolectic vs. Polyectic Individuals Oligolcctic Polylectic Species Volume 15, Number 2, 2006 221

several cavity-nesters were represented by slow and low-flying species of, e.g., small singletons or doubletons. By contrast, Lasioglossum, Andrena, Osmia, and Nomada. wood burrowing species were relatively Our results generally support the currently few (Fig. 3b), although some of these accepted view that a combination of bowl species were captured in large numbers trapping using multiple colors and netting is the best to collect a (e.g., Augochlora pura). The large number way efficiently plural- of hive-nesting individuals relative to ity of species (S. Droege et al. protocol). species likely reflects their Only 89 of the 144 bee species collected (see above). (61.8%) were collected by both nets and A few species typical of more open and bowls, with 30 species (20.8%) unique to at nets 26 to sandy areas were found BRF (e.g., and (17.6%) unique bowls Lasioglossum heterognathum, , As- (Fig. 4b). The net collected sample was tata), but sand specialists such as L. vierecki richer than the bowl trapped sample in were not found. — total bee species (117 vs. 113) and in number Efficacy and Biases of trapping methods. of unique species (29, 20.4% vs. 25, 17.6%). The year 2003 was characterized by long Although wood and cavity-nesting bees periods of cold and cloudy weather and were numerous in this survey, only one pans may have been particularly useful individual bee (the introduced Osmia cor- under these conditions as these allow catch nifrons) used our trap-nests. The poor during brief windows of sun on days when performance of trap-nests might possibly net-collecting would be unrewarding. be explained by an abundance of natural Nearly twice as many individuals were nesting substrates (standing dead wood) at bowl trapped than netted (4,322 vs. 2,221) BRF. Alternatively, bees may have been (Fig. 4a), but the net sample was biased out-competed for the trap-nests by eume- against certain of the most common and nines and Trypoxylon, or else the nests may readily identified taxa (see above). Bowls have been placed in sites that ultimately were found to be particularly useful in proved to be too shady. — forest and at the forest edge where flowers Comparison to other bee faunas. In com- are few, dispersed, or in the case of trees parison to the bee fauna of NYS as a whole and shrubs, difficult to reach. Where (423 species) and to the fauna of some well- flowers are scarce, bowls may be particu- sampled localities within the state such as larly effective due to lack of competition Ithaca (274 species), the 144 species iden- from real flowers. Using bowls, we found tified in our BRF sample is relatively few certain inconspicuous forest-associated spe- (Table 1). However, several of these re- cies rarely taken in nets such as Stelis nitida. cords are of considerable biogeographic or Well known biases of bowl traps reinforced ecological interest (see above). The NY bee by our study include low catch rates for fauna includes many species which are certain groups, especially fast and high- regionally rare and/or have highly special- flying species of, e.g., Colletes, Megachile, and ized ecological requirements, and are perhaps Melissodes, and high catch rates for therefore unlikely to be found at BRF.

oligolectic, polylectic and solitary, or other; 2B, percentage of bee species that are oligolectic, polylectic and solitary, or other. 3, Nest substrates: 3A, percentage of individuals belonging to each nesting category: soil, cavity, wood burrowers, hive, or other (primarily cleptoparasites that live in the nests of their hosts); 3B, percentage of bee species known or inferred to use the nest substrate indicated. 4, Collecting method: net vs. white bowl, vs. blue bowl, vs. yellow bowl: 4A, percentage of bee individuals caught by each method; 4B, percentage of bee species caught by net only vs. bowl only vs. both net and bowl. of Hymenoptera Research 222 Journal

from the northern and montane of Nonetheless, it seems highly probable that portions the Adirondack Mountains and in at least 250 bee species could be present NY (e.g., but absent from at BRF based on totals of 274 species most cases Ithaca), NYC, and other warmer and coastal recorded from Ithaca, Tompkins County, Long Island, of northern occur- NY (Ascher, unpublished), in a colder areas. Examples species near their southern limits at BRF climate than BRF, and ca. 300 species ring at or unknown from NYC include Andrena recorded from the vicinity of Carlinville and A. Stelis Osmia in southern Illinois (Robertson 1929, Marlin rufosignata, algida, nitida, unre- O. Bombus ternarius, and and LaBerge 2001), in seemingly felti, inermis, B. sandersoni distribu- markable farm country. possibly (southern tional limits of this remain un- The high number and proportion of species of doubletons certain due to identification difficulties singletons (28 spp., 19.4%), from versus B. these northern (12 spp., 8.3%), of species known vagans). Although are absent from a single sex (ca. 31 spp., ca. 21.0%) species likely genuinely of the 42 known from excluding Lasioglossum, and of rarely col- NYC, many species individuals col- but not NYC be found in the lected species (i.e., 3-10 BRF, may that latter area when more lected: 36 species, 25.3%), indicate thorough samples been made of semi-natural habitats more prolonged and intensive surveying have such as Pelham Park. Southern ele- using the same methods would reveal Bay ments in the BRF fauna include the many additional species, likely resulting that are unknown from in taxonomically and biogeographically following species the well collected significant specimens. very Fingerlakes Region bni- Another indication of the incomplete- (which includes Ithaca): Lasioglossum A. ness of sampling of the total BRF fauna is neri, , A. eonfederata, hilaris, subillata. The ab- that only 57.8% of the 249 bee species and Melissodes apparent known from southern New York excluding sence of these species from Ithaca and elsewhere in central and northern New NYC and Long Island (i.e., Sullivan, Ulster, and Dutchess, Orange, Putnam, Rockland, York is probably genuine and likely reflects and Westchester, Counties) were found. a real faunal difference from BRF. Andrena These might be considered to represent violae is another species of southern affin- in Ithaca a regional pool of species from an area ities that is very rare (see above). = + b + relevant to BRF. The total of 249 species The Sorensen index [C s 2a /(2a c) known from an area relevant to BRF is only where a is the number of species shared 59% of the species total for New York State between two sites, b is the number of as a whole (423), which in turn is only 57% species found at only one site, and c is the of the 743 bee species known from the number of species found only at the other eastern USA. Twenty additional species site] was used to quantify similarity be- recorded in NY only from coastal NYC and tween various sites. The total for BRF vs. Long Island (e.g., the coastal dune special- Ithaca is 58.9% whereas the total for BRF ist Lasioglossum marinum) are less likely to vs. NYC is 58.2%. The similar Sorensen occur at BRF. values for comparisons involving these Most bee species at BRF are widely two areas (despite BRF's much greater distributed in NY and have been recorded geographical proximity to NYC) reflect from other well-collected sites such as many shared widespread and northern Ithaca (123 species shared with BRF, species with Ithaca, and significant differ- 85.4% of the BRF total) and NYC (103 ences between BRF and NYC due to the shared with of the species BRF, 71.5% BRF presence of northern forest elements (e.g., total). Northern elements of the fauna at blueberry associates) only at BRF and of BRF can be defined as those species known coastal /sand associates only in NYC. Volume 15, Number 2, 2006 223

ACKNOWLEDGMENTS nators: detection, evidence, and consequences. Conservation Ecology 5: art. 1 [online]. B. T. 2004. review of the We are grateful to the staff of Black Rock Forest for Coelho, W. A bee genus their invaluable advice and assistance with practical Augochlorella (Hymenoptera: Halictidae: Augo- matters, and to the Steifel Foundation for an award chlorini). Systematic Entomology 29: 282-323(42). from their Black Rock Forest Consortium Small Grants Danforth, B. N. and S. Ji. 2001. Australian Lasioglossum for Scientific Research and Education program + form a monophyletic group: Re- awarded to J.G. Rozen and V. Giles. We thank: S. solving the "Australian enigma". Systematic Bi- Droege, L. Day, and T. Griswold for identifying bees, ology 50: 268-283. and M. Arduser for sharing unpublished keys and Davydova, N. G. and Yu. A. Pesenko. 2002. Bees of the other taxonomic information and commenting on the subgenus Clisodon (Anthophora, Hymenoptera, manuscript. P. Gambino provided unpublished re- Apidae) in the fauna of Russia. Entomological cords of NY bees and identified many of our Review 82: 621-624. [translated from Zoologicheskii crabronid wasps. K. Magnacca provided bee records Zhumal 81(11)] and other assistance. The following people kindly Hinojosa-Diaz, I. A., O. Yaiiez-Ordonez, G. Chen, A. provided access to and information about historical T. Peterson, and M. S. Engel. 2005. The North collections of New York bees: B. N. Danforth, E. R. American invasion of the giant resin bee (Hyme- T. Hoebeke, J. K. Liebherr, T. McCabe, Schultz, D. noptera: Megachilidae). Journal of Hymenoptera Yanega. S. Kornbluth generously provided assistance Research 14: 69-77. with specimen labeling and data entry and manage- Hoebeke, E. R. and A. G. Wheeler, Jr. 1999. Anthidium ment. We thank K. Russell for her helpful discussons oblongatum (Illiger): an Old World bee (Hyme- of methods and data base management. C. Encarna- noptera: Megachilidae) new to North AMerica, cion, and E. Burbano prepared our specimens. T. and new North American records for another technical matters. G. adventive A. manicaium The Univer- Nguyen lent his assistance with J. species, (L.). Kansas Natural Museum Rozen, J. M. Carpenter, A. C. Vallely, and C. Dong sity of History Special Publication 24: 21-24. made valuable comments on this manuscript. J. M. Carpenter identified our vespid wasps. We thank R. Hurd, P. D., Jr. 1979. Superfamily Apoidea. Pp. 1741- in: D. R. G. Goelet for his support of bee research at the 2209 K. V. Krombein, P. D. Hurd, Jr., AMNH. We are especially grateful to J.G. Rozen for Smith, and B. D. Burks, eds. Catalog of Hymenop- his guidance and support. In addition to his help in tera of America north of Mexico, vol. 2. Smithsonian D. C. this project, we both owe our interest in bees to his Institution Press, Washington, generosity and encouragement. LaBerge. 1973. A revision of the bees of the genus Andrena of the Western Hemisphere, Part VI: LITERATURE CITED Subgenus Trachandrena. Transactions of the Amer- ican Entomological Society 99: 235-371. S. 2001. a Ascher, J. Smith, European LaBerge. 1980. A revision of the bees of the genus bee new to North America, with notes on other Andrena of the Western Hemisphere, Part X: adventive bees Proceed- (Hymenoptera: Apoidea). Subgenus Andrena. Transactions of the American the 103: ings of Entomological Society of Washington Entomological Society 106: 595-525. 184-190. Leonard, M. D. 1928. A list of the insects of New York, Barringer, K. and S. E. Clemants. 2003. The vascular with a list of the spiders and certain other allied York. flora of Black Rock Forest, Cornwall, New groups. Cornell University Agricultural Experiment journal of the Torrey Botanical Society 130: 292-308. Station Memoir 101: 5-1121. H. 2003. Exotic nonsocial bees E. 2001. native bee Cane, J. (Hymenoptera: Marlin, J. C. and W. LaBerge. The Apiformes) in North America: Ecological impli- fauna of Carlinville, Illinois, revisited after in: K. and H. cations. Pp. 113-126 Strickler, J. 75 years: a case for persistence. Conservation Cane, eds. For Non-native Crops, Whence Pollinators Ecology 5: art. 9. [online] of the Future. Thomas Say Publications in Ento- Matheson, A., S. L. Buchmann, C. O'Toole, P. mology: Proceedings. Westrich, and 1. Williams, eds. 1996. The Conserva- the Bees. Cane, J. H. 2005. Bees, pollination, and challenges tion of Linnaean Society Symposium Series, of sprawl. Pp. 109-124 in: E. A. Johnson and M. Number 11. Academic Press, London. eds. Nature in the R. 1986. Studies Halictinae W. Klemens, Fragments: Legacy of McGinley, J. of (Apoidea: Sprawl. Columbia University Press. Halictidae), I: Revision of New World Lasioglos- F. R. Curtis. Contributions to Cane, J. H., G. C. Eickwort, Wesley, and J. sum Smithsonian Zoology, Spielholz. 1985. Pollination ecology of Vaccinium no. 429: 294 pp. stamineum (Ericaceae: Vaccinioideae). American Michener, C. D. 2000. The Bees of the World. The Johns journal of Botany 72: 135-142. Hopkins University Press, Baltimore, Maryland. 2001. extent E. R. and E. A. 1995. Bionomics of Cane, J. H. and V. J. Tepedino. Causes and Miliczky, Osgood. vicina Smith in Maine and of declines among native North American polli- Andrena (Melandrena) 224 Journal of Hymenoptera Research

Washington, with new parasite records for A. Gattungen Coelioxys and Nomada (Hymeno- (M.) regularis Malloch and a review of Melandrena ptera, Apidae). Linzer biologische Beitrdge 62: biology, journal of the Kansas Entomological Society 1413-1485. 68: 51-66. Snelling, R. R. and G. I. Stage. 1995. A revision of the Mitchell, T. B. 1960. Bees of the eastern United States, Nearctic Melittidae: The subfamily Melittinae Vol. 1. North Carolina Agricultural Experimental (Hymenoptera: Apoidea). Natural History Museum Station Technical Bulletin. of Los Angeles County Contributions in Science Mitchell, T. B. 1962. Bees of the eastern United States, No. 451: 1-17.

Vol. 1. North Carolina Agricultural Experimental Thorp, R. W. 2005. Species profile: Bombus franklini. In: Station Technical Bulletin. Shepherd, M. D., D. M. Vaughan, and S. H. Black, Nilsson, L. A. 2000. Critical resource levels for viable eds. Red List of Pollinator Insects of North America. wild bee populations, www.cbm.slu.se/forskning/ CD-ROM Version 1 (May 2005). The Xerces naturvardskedjan/pdf/ wildbee.pdf Society for Invertebrate Conservation, , Robertson. 1928. Flowers and Insects. Lists of visitors of OR. four hundred and fifty-three flowers. Carlinville, Williams, N. M., R. L. Minckley, and F. A. Silviera. Illinois: Published by the author. 2001. Variation in native bee faunas and implica- Schwarz, M. and F. Gusenleirner. 2004. Beitrag zur tions for detecting community changes. Conserva- Klarung und Kenntnis parasitarer Bienen der tion Ecology 5: art. 7. [online] Volume 15, Number 2, 2006 225

O X;

01 226 Journal of Hymenoptera Research

eoeifli X X X X X X XX xxxx xxxxxxxx

A}Q y[iox Mafvj X XXX XXX X X X XXX X XX

Ajpijpads ua||oj I I I I I I &4 PLh PLh PLk I O P-tilllllllP-lQ-lllll I

Xiijepos fcPHfcwwmaiaiwmwaiuiBwwwwfcwwwinwwww cu

1 1 1 aiEHsqng )sa|Nj \u~)inininin{ntS)U~iininu~i

3140X3 SA 3AI4B[\I 22222ZZZ2W2Z2ZZ2ZZ2ZZ22Z2ZZ 2

aauepunqi? ^j\| «tt,D

O rJOOOtNOi—1 O OO OOO pouad jqSig 3ipieat\l ^-<(^CO^HO\<—'1—<*—li—1»—I ON i—< ON ON ON ON 1—I (^ ,_((—lt\<^<^'— ^H r-H <0\

I I I I I I I I I I I I I I I I I I I I I . I I I III I I I I 00 "+ LT) co ^fOcimfificnTftnintNinin^cn^O'* tJh ^ ro ^ no

o 0000I r- 00 (J> I i—< pouad jqgiij ^1 OpOOr-HONi— N IH(J\UU001Jl^rlOS<^OONt^t^ONi— ON ON I I I I I I I I I I I I I I I I I I I I „ _L I I I I I I »^in^rO'tin<*inin'tvom^t'*in^| iriiri't^ioniir4fn't no

00 i—l— ^ ri e-H 00 rl bC 3 CX ^ ex ^ QJ s a, ex < en ej U en ^ rN O.O pouad mSi0 jyg en a> 01 en ^ < =, I So 1 tn 1 O I I T rX _L 1 en en 1 -+ o 1 in •* X 00 CO vo ,>,>, >^ >, Oj £ >-. °C 'C £ 3CrerererereCX re C —< CX Q-, <° ex re re re re re 3 Ci. S ^ f^ at s 1 2 < a < 5 S 2 < < CA)

ON •<* ON fN H

NO 00 ON 00 ^ix ^^ O >;cn m 00 ON m >.- 00 00 c —o?m00 00 -rS 00 <— 1 J rH 1—1 t ON 3 IX 5 -3 01 o 5 & ON en 01 T3 f^ en c X; — "- 1 O re - ON O .O ml 1~~< &u s QJ -C K -t-> — en I s i Vi 3 en y en ON U T3 U -S O) 1— 1 o -5 , =a u .<50 3 2 fc fc s 5 J3 -3 _I1 « C u >¥ 33— w o o O 01 .-fcj U s

.2 c ex < Volume 15, Number 2, 2006 227

e:>t?mi X X X X X X X X X X X X X XXX XXXXXX X

X4Q ^JO^ M3(\l XX XXX xxxxx XXX XX xxxx

Aipypads uajjoj 1 1 1 1 1 1 1 0OOM^^fcO00COt-t.Oc-Ot-t-CL.c.

/Viijepog w w w u w w w inu*iw

a)BJ4sqng jsa^j cric/ii^uitnc/ia) a)cfi(/)a)tn(/itntntni/i(/)(/ii/)tntntnai(/ii/ic/ii/i

D140X3 SA 3AI4EN Z Z Z Z Z Z Z zzzzzzzzzzzzzzzzzzzzz

ajuepunqe XN UUU<

0000 no o o pouad 4L|8i[J Ji43iEa[\i

I I I I I I I 7 7 7 7 1 7 7 I I I I I I I I I I I I t -t n n ^f h -f

O OOOi nO 00 1— ON ^h ON ^ ON v£)iOirjvCN>£inNO»lJii-ifl\B\BNvOi£N'fl

4M§!U AJM I i pouad I I I I I I I I I I I I I I I I I I I I I I I I I

_ -t- ^ o CM ,— ON CM ^ H —' I 00 ON 1 T— O \D CM (N CM —1 rsl CM 01 Ol tN r-H , 01 C co s y 3 ^ H. Oh c S 3 oj I >. CS 01 01 2 c xL 2L « <_T) S 5; —, Si 41 ^~> T en en 1 1 pouad 4M§!U JJ19 I 3 3 -4H ^$ 1 o en en 4, 7 1 1 NO ** VnO rf ^2 > I I I A<— I I I I 1 T— 00 00 I cm (N T^ Si 00 -t I— ,— O t CM CN On ON CN in (N LO OJ I— I ON T— ON 00 _ __ ] 01 >, bJD M) >, >, 'C >.

cm *— i CD s|BnpiAipui jag

a. -c 3 C c Uo

X T3 C 01 a, < 228 Journal of Hymenoptera Research

tOElfl] xxxxxxxxxxxxxx xxxxxx xxxxxx

J Ma ^0 >t °A N X xxxxxxxxxxxxxxxxxxxx X X

Aipypads ua|[Oj cuc^cuaH&HCLHOPH(xcxciHCiHeHPHCiHfxcuci.cuciHO(x 1 1 a. c cu cu cu

A)i|Bpog (/it/i^wmt/i^i/it/ii/ii/iaiait/ic/i^aiaii/i^aimfcciHi/i^tfiwi/i

aie-Usqiig isa|\[ (/)c/i(/)^a)c/i^(^^(/)^a)i/)^i/)ui(/)(/it/)(/iLn(/) 1 lUUUUU

3140x3 sa aAi;i'N 2222222222wZZZZZZZZZZwZZZZ22Z

1 aouEpunqt? ^|nj Da< ii<<<;UUU<<

0000 pouad iqSiu djpibb^

-i -i I !. !. L i I. J. I J. .1 J !. (. I I I I. _L 1 1 i I. I. i ! ! co n to m n •t n to * m in *'*'*\OtO'*00vOfOvO'1, '1< 't-t-t

pouad iqSijj ,\[nj N^D*NNiON\OiO00>C^O\O\O\O>O00ifl00HrHl»NO\N00KvO I I I I I I I I I I I I I I I I I I I I I I I I I I I I Ln^ir)^inin^int^^^toiLo^^^^NO'*NocoNOLXNONOLrjNo^^t|

tN CN <* IN o tN N NO NO o SI 00 ^ § 5P ojd ^> 2P >, j>jg 3 pouad iqSi]j 3 1 i f I 1 3 Q X^Xi. T II NO ^ I < 5, I I < tx lx CO I I I CO 00 00 X A 00 1 eN CO 1 ~ 7 NO 1 CN CN LT) ON ^H ON LH tN LO r l CN __ CO CO ^ CN H CO On On >n E^ cu ^ >^'s :>, E^ ^. bJD x ^ >, >-, >^ (0 rS (U (C (U TO 03 C ta re {^ ta re re c X> rc ra 2 2 S 2 2 I 3 3 2 2

N^iBOigtOin^OONOOOrO^HaiNtO-f-l'rHOrHHU-iinNvOO spmpiAipui j>jg to n * ifi lonocn colo cn t—1 rj i—ico

NO ON IX NO 00 o LTj o <"< ON LT; ON CO CO «- ON CO CO ON „> IX ON CO CO 00 cu Cx ^ LX CO -* ON ^ 00 m Tf CO QO ,_l ON 00 no '- | ON 00 ST § ON cu ^r 22 -- —I ^ 00 1 CO —I CU t ON , 00 ~* I—I 00 NO * i-H 00 <^~l CO '""' 00 CO CU o u~> QJ in §•- C in > in ^ o _r CO X4= CO t/j rc O EC s -I-' HH 0050^ X P -tJ 1—( § (/) CU cu CO ir o J i-H "^ £ Cu « Xi CU & X cu H u cu X tjO Q J-t QJ .2 xi ^=£ t: 4 "§. X cu X c< X -= 01 " U x > o ei3 o ? .a I — xi _ ~ .- ? cu -^ u. o > c< S c u 2 C -2 -2 ~ E CO CJ Cj Cj "G CJ S- i>, -cs s S 2 X CU "S cs S £ S cu cs cs v a -is c k u G cj -~ £ £ 0-, _, p 55 W Ifl « Ul tSg c ~-^^3'

E3BUHI XXX xxxxxxx xxxxxxxx xxxxxxx

^D 1 J°A MaN XXXXXXXXX XXXXX X X X X

A^pypads uo||Oj fcfctO^fcDHfcfct.L^t.OCH I I I I I I I I I I I I

A)i|t?pog ai(/i(/i(/i(/)i/)aic/i«^an/)i/i^a)C.fcocDCQ c-c.c_o.c-o,a,&.

3;Ei)sqng isajsj UUUUUUUUUUUUUUUI 1 I U U 1 1 1 1 1 1 I

3140x3 sa aAjn'N Z Z Z Z Z 2 w Z w 2 2 2 Z Z Z Z Z Z Z Z ZZZZZZZZ

aauepunqp \f^ Dti,eiDP-UUUU(JUDUiiHU,Uu.U

^h o o o 00 pouad }q8tu 3tpJi'a|\j NO no IXnOIXnOnOOni-^i—> On i—i On 00 i—' 00 i—i i—'On no -?ix^cnolxsOnc

I I 1 I II.1 _ .1 I I iI I I I I I I I1 I1 I1 I I1 1 1 ill -tlnNomln''*''tcONOcNln^LnNOT^LnNOcocO'3, 3, cO"*inLn^t-r--t-

o 00 pouad iqgiy ^N I I I I I I I , llllllllllllllllllll| , •tlnlnlnln•fi ^o^o^Oln^u)^o^o^O l£fO^'l'!) in ro -t in in

-* 00 CN i—1 ° -. ^ ^ rN 1 (N a, « (U X c re bJD re g >, Oh CL, 3 o> 0) CU 0) -—^ 3 re |ji. d>jg en < qj en en pouad )q8iy en 3 I

1 ^ en 1 1 V 1 CX) * if

-* 1 1 I 00 CN CN 00 00 IX I IS I NO I IX h fN ON CN CN cn I-1 C/1 H rn t- ON CN on h in ,-H r-H ^ «N CO N ,i ^ >, 01 01 o 01 t>~. K*^l M " " r> >, >, !>N oJD ^ ^ ^ C re c. re C &, o, a, Kg J*£ ^ re re c a, re re re re re cv a 0> 3 3' = 2, s 1 < < < en 3, X en 22 2 S 2 <

+ ,

on in in * m in in rinHiHHKoornmNnN S[t>npiAipui jyg no cn no no co 1—1 co

m

t3

c Uo

>< -3 c O) ex < 230 Journal of Hymenoptera Research

EDeiHI XXXXXXXXX X X X X X X X XX X

b]D >F°A M3N X X X X XXX XXXXXXXX X X

/tipypads uanoj I I I I I I I OOp-iiiiiiii I

A)i|t'pog ^DHCHCHt^fcttn OlOK/lfcWWWHJUBW UJ

ojpjisqng jsa|\[ <-n^\XXXXXXX X

JI}OX3 SA 3Api?N ZZZZZ2ZZZ ZZZZZZZZZZZ w

aauepunqi' \f^ UUU^DftUB.B- tui-uUUUU

O OO ^OOOO (N —I —' < — — I I ' ' — — I < 1 — pouad iqSju aipjea[\[ ON ON Q\ I^Hi ll * i

1 I I I I I I I I I I [111 I I I I I I in \o in in ts n r-1 ^ -t •* in *— <

O i— i O O O O (N

1 —I —I pouad qqSiu ^N ON00ON^ONONr-(^Hl T t ^H I I I I I I I I I I I I I I I I I I I I I m-tTfm^NO^-F-F NN>cm't^| ^< -t-t, -tirj m

CN CN tN

pouad H|%1 jyg

sienpiAipui j>ig

73 QJ 3 C -3 C uO

T3 C QJ a, Volume 15, Number 2, 2006 231

Appendix 2. List of wasp species collected incidentally at F3RF in 2003.

Family Subfamily Specie!

Crabronidae Astata leuthstromi Ashmead, 1897 Crabronidae Bembecinae Biq/rtes quadrifasciata (Say, 1824) Crabronidae Bembecinae deceptor Krombein, 1958 Crabronidae Ectemnius (Clytochrysus) lapidarius (Panzer, 1804) Crabronidae Crabroninae Ectemnius (Ectemnius) atriceps (Cresson, 1865) Crabronidae Crabroninae Ectemnius (Ectemnius) borealis (Zetterstedt, 1838) Crabronidae Crabroninae Ectemnius (Ectemnius) dives (Lepeletier & Brulle, 1834) Crabronidae Crabroninae Ectemnius (Hypocrabro) continuus (Fabricius, 1804) Crabronidae Crabroninae Ectemnius (Hypocrabro) decemmaculatus (Say, 1823) Crabronidae Crabroninae Ectemnius (Hypocrabro) stirpicola (Packard, 1866) Crabronidae Crabroninae (Leptolarra) argentata (Beauvois, 1811) Crabronidae Crabroninae subita (Say, 1837) Crabronidae Crabroninae Tn/poxylon (Trypargilum) lactitarse Saussure, 1867 Crabronidae Crabroninae Trypoxylon (Trypoxylon) frigidum Smith, 1856 Crabronidae Crabroninae Tnjpoxylon (Tn/poxylon) pennsylvanicum Saussure, 1867 Crabronidae (Cemonus) inornata Say, 1824 Crabronidae Pemphredoninae Pemphredon (Cemonus) rugifera Dahlbom Crabronidae Pemphredoninae nigra (Packard, 1867) Crabronidae Pemphredoninae erythropoda Rohwer, 1910 Crabronidae Pemphredoninae Pseneo shnplicicornis (Fox, 1898) Crabronidae atramontensis Banks, 1913 Crabronidae Philanthinae Say, 1837 Crabronidae Philanthinae Cerceris halone Banks, 1912 Crabronidae Philanthinae (Fabricius, 1775) Isodontia (Isodontia) philadelphica (Lepeletier, 1845) Sphecidae Sphecinae Isodontia (Murrayella) mexicana (Saussure, 1867) Vespidae Eumeninae Parancistrocerus pedestris (Saussure, 1855) Vespidae Eumeninae Parancistrocerus pensylvanicus (Saussure, 1855) Vespidae Eumeninae Parancistrocerus perennis (Saussure, 1857) Vespidae Eumeninae Euodynerus foraminatus (Saussure, 1853) Vespidae Eumeninae Euodynerus hidalgo (Saussure, 1857) Vespidae Eumeninae Euodynerus leucomelas (Saussure, 1855) Vespidae Eumeninae adiabatus (Saussure, 1852) Vespidae Eumeninae Ancistrocents antilope (Panzer, 1798) Vespidae Eumeninae Ancistrocerus campestris (Saussure, 1852) Vespidae Eumeninae Ancistrocerus waldenii (Viereck, 1906) Vespidae Eumeninae Symmorphus (Symmorphus) canadensis (Saussure, 1855) Vespidae Eumeninae Eumenes (Eurnenes) fraternus Say, 1824 Vespidae Eumeninae Zethus (Zethus) spinipcs Say, 1837 Vespidae Polistinae Polistes dominulus (Christ, 1791) Vespidae Polistinae Polistes fuscatus (Fabricius, 1793) Vespidae Vespinae Dolichovespula arenaria (Fabricius, 1775) Vespidae Vespinae (Linnaeus, 1758) Vespidae Vespinae Vespula consobrina (Saussure, 1864) Vespidae Vespinae Vespula flavopilosa Jacobson, 1978 Vespidae Vespinae Vespula germanica (Fabricius, 1793) Vespidae Vespinae Vespula maculifrons (Buysson, 1905) Vespidae Vespinae Vespula vidua (Saussure, 1854) Scoliinae (Discolia) bicincta Fabricius, 1775 Pompilidae maculata (Fabricius, 1775) J. HYM. RES. Vol. 15(2), 2006, pp. 232-250

The Biology and Morphology of Entedon sylvestris (Hymenoptera: Eulophidae), a Larval Endoparasitoid of Ceutorhynchus sisymbrii (Coleoptera: Curculionidae)

Alex V. Gumovsky

Schmalhausen Institute of Zoology, 15 Bogdan Khmelnitsky St., 01601 Kiev-30, Ukraine; email: [email protected]

Entedon Abstract. — The biology and morphology of preimaginal stages of sylvestris Szelenyi first Entedon is a larval (Hymenoptera: Eulophidae), are described in detail for the time. sylvestris on the endoparasitoid of the seed-feeding larvae of the weevil Ceutorhynchus sisymbrii Dieckmann small tumbleweed mustard, Sisymbrium loeselii L. (Brassicaceae). In the Ukraine, females of £. until of sylvestris begin ovipositing in late May and continue to lay eggs the beginning July. Females within of E. sylvestris parasitize weevil larvae of various instars. The parasitoid remains the body of the host weevil larva until the emergence of the latter from the dried host-plant pods. The morphology of each of the three larval instars is described in detail. The moult of the parasitoid larva into the final instar, as well as pupation, takes place underground. Adults of E. sylvestris must therefore penetrate a soil layer to emerge the following spring.

Key words. —, larval endoparasitoids, parasitoid-host relationships, preimaginal morphology, Sisymbrium loeselii

Parasitic wasps of the genus Entedon ergias Walker, E. rumicis Graham, E. phar- Dalman (Eulophidae, Entedoninae) are nns Walker. Askew (1991) and Gumovsky endoparasitoids of immature stages of (1997) provided some information on per- . Curculionidae (including Scolyti- centage parasitism, the biology of the final nae), Brentidae (including Apioninae), An- instar larva and pupation procedure of E. obiidae, Chrysomelidae (including Bruchi- cioni Thomson, E. cionobius Thomson and nae), , Cerambycidae, Mordel- E. zanara Walker. lidae, and Nitidulidae are recorded as In general, despite some thorough re- hosts (Boucek and Askew 1968, Graham views (Parker 1924, Parker and Thomson 1971, Askew and Kopelke 1989, Rasplus 1925) and occasional detailed descriptions Entedon Walker has 1991). ergias been (e.g., Darling 1992, 1995) of the larvae of imported from Europe into North America Chalcidoidea, our knowledge of morphol- for the control biological of the smaller ogy of preimaginal stages of chalcid wasps European bark , Scolytus multi- is incomplete. Most discussions on larval striatus (Marsham) (Peck 1963). For some morphology and biology concern ectopar- species, parasitism rates and/or general asitoids, whereas endoparasitoid larvae of the larval descriptions morphology are traditionally attract less attention, mainly Erdos given (Ferriere 1939, 1944, Abedin due to the difficulties with their prepara- and Quayum 1972, Tiwari 1976). However, tion and identification. The larvae of these descriptions lack many morphologi- Eulophidae were classified by Parker cal for the first details, especially instar (1924) in group II (the ectoparasitoid larvae. Beaver and Fisher (1966) (1970) forms) and V (egg endoparasitoids), dif- the most and gave complete bioassays fering mostly in having spiracles (group II) for Entedon reported egg-larval parasitism or being apneustic (group V). Later, when Volume 15, Number 2, 2006 233 discussing the morphological peculiarities ical control agents. It is therefore of vital of the first instar larvae of Anastatus sp. importance to determine the parasitoid- (Eupelmidae) and Miscogaster sp. (Ptero- host associations of Ceutorhynchinae and malidae), Parker and Thomson (1925) host plant preferences in various geo- stated that some endoparasitoid larvae graphic regions. represent a transitional type between the This paper reports results obtained from groups corresponding to groups V and VI and the methodological approaches used sensu Parker (1924). during the study of the parasitoid-host The larval morphology of Entedon spe- relationships between Entedon sylvestris cies is even more vague. The papers of Szelenyi and its host, the weevil Ceuto- Beaver (1966) and Fisher (1970) are the only rhynchus sysimbrii Dieckmann on Sisym- sources of comprehensive descriptions of brium loeselii, which is the first host record preimaginal stages, but these concern for the parasitoid. the size, body proportions and mainly MATERIALS AND METHODS number of spiracles, and are illustrated of mostly by diagrammatic figures. Many Adults of Entedon sylvestris were collect- the structures minute morphological (e.g. ed in the field on plants of Sisymbrium un- sensorial organs) remain obscure and loeselii during late June and early July in described. 1995, 1997, 2001, 2002, 2004 in Kiev Weevils of the subfamily Ceutorhynchi- (50 28'N; 30°32'E) and Kherson oblast of nae have a wide range of host plants, but Ukraine (v. Lazurnoe) (46 04'N; 3229'E). many are restricted to Brassicaceae. Some Collections were made by both sweeping Ceutorhynchinae species have gained spe- and capture of individual adults in tubes, cial attention as pests of economically Females were kept in Petri dishes and fed important plants (e.g., Ceutorhynchus napi with diluted honey. Gyllenhal, C. pallidactylus (Marsham) and Mature pods of S. loeselii infested by the C. obstrictus (Marsham) damaging cabbage larvae of Ceutorhynchus sysimbrii (living and oilseed rape in North America). Other inside the pods and feeding on seeds) were merkli C. cardar- species (e.g. C. Korotyaev, exposed to females of E. sylvestris kept in iae Korotyaev, C. alliariae Brisout, C. roberti various reservoirs (Petri dishes, sealed Gyllenhal) are under investigation as po- plastic bags or boxes). Infestation by the tential biological control agents against weevil was indicated by the presence of some introduced weeds (i.e. whitetops a distinct hole in the pod made by the Cardaria spp., garlic mustard Alliaria petio- female's rostrum before oviposition. laria: Hinz et al. 2004, Hinz and Gerber, The sites where the parasitoid's oviposi- 1998). The small tumbleweed mustard, tion took place were marked with black Sisymbrium loeselii L., is a plant of European ink. To study the morphology of the differ- origin that was accidentally introduced into ent larval instars of E. sylvestris, weevil the New World, and is now recorded in 31 larvae were removed from the pods at states of the USA, and is regarded as an regular intervals and dissected. Endopar- invasive weed (Stubbendieck et al. 1994). asitoid larvae found were fixed in Bouin's 3 Some parasitoid species (e.g. Tersilochus fixing solution (15 cm picric acid (saturat- 3 3 spp., Microctonus spp.) have been released ed), 5 cm formaldehyde solution, 1 cm in North America to control Ceutorhynchus acetic acid) to keep their original shape, in spp. (e.g. C. obstrictus) that are pests of and further washed out 96-98% economic Brassica How- ethanol. plants —(e.g. spp.). ever, parasitoids— unless sufficiently host- Mature weevil larvae, leaving the pods specific could hamper the effectiveness of of their host plant, were put into plastic the Ceutorhynchus spp. released as biolog- tubes (50 X 16 mm) filled 3/4^1/5 with Journal of Hymenoptera Research

soil taken from field sites of S. loeselii. The blueyonder.co.uk/CombineZ/CombineZ3. behavior of the mature weevil larvae was zip). observed the walls of through transparent RESULTS the tubes. In September/October, when the were to be pupae of E. sylvestris expected all tubes were completely formed, carefully searched for earthen cells or dead weevil Entedon 1981 larvae. Dead weevil larvae found, were put sylvestris Szelenyi, Entedon Szelenyi, 1981: 277; Entedon into 5-10% solution of lactic acid to regain sylvestris Szelenyi; Askew, 1992: 119; Entedon their and soften. The soft- sylvestris original shape 1999: (Entedon) sylvestris Szelenyi; Gumovsky, ened larvae were then dissected in order to 142; Entedon sylvestris; Gumovsky, Boyadz- find the larvae, in the parasitoid particular, hiev, 2003: 23. final instar larvae. These were transferred examined.— female, into lactic acid solution of higher concen- Material Types: Holotype 10 females, 6 males, Horto- tration (40-70%) to obtain maximum swell- paratypes Hungary, National Park Mu- larvae were then bagy (Szelenyi) (Hungarian ing. The parasitoid put seum of Natural more than fix History, Budapest); into Bouin's fixing solution in order to 2,000 specimens from Kiev (June 1995, 1997, the and then washed out in regained shape 2003, 2002, 2004) and v. Lazurnoe (Kherson 96-98% ethanol. oblast), Ukraine, collected on plant shoots and All fixed larvae were in 100% kept fruits of S. loeselii; 7 females, 4 males, Kiev ethanol for one and then in 100% from day vicinity, Velyka Oleksandrivka, swept molecular sieved ethanol for maximal de- Berteroa incana and Capsella bursa-pastoris, of hydration. After absolute ethanol the spe- 31.V.2004, (Schmalhausen Institute Zoology, cimens were Critical Point Dried. The Kiev); 54 females, 1 male, Kyiv, Trukhaniv "Park Narodiv", 24.VI.1995, minute parasitoid larvae were put into Island, Druzhby sealed on swept from Sisymbrium loeselii (Natural History pipette tips of various diameters Museum, London); 1 female, 1 male, Giefien, ex the sides with cotton wool plugs, to avoid Capsella seeds V.1995 (Weiffenbach) (Zoolo- their loss during drying. The dried para- gische Staatssammlung Munchen). sitoid larvae were transferred to SEM stubs — on metallic pins, using static electrical Recent literature. Entedon sylvestris has charges to avoid damaging their extremely gained little attention since its description soft integuments. Finally the specimens (Szelenyi, 1981). This species belongs to the were coated with gold and observed using assemblage of the cyanellus and costalis a Scanning Electronic Microscope LEO species groups in having the anterior 1530VP in the Max-Planck Institute for margin of the clypeus produced (Figs 1A, Metal Research, Stuttgart (MPI). C, E, cly; 2C). Field and laboratory video recordings Askew (1992) reported it for Great were made using either 8 mm VP-A800 Pal Britain, presented a short corrected di- Samsung Video Camera or with digital agnosis and reported the possession of imaging of Leica IC A Videomodule in- traces of pale strips on the fore tibiae. He tegrated in the Leica MZ 125 stereomicro- also proposed to locate this species in the scope, using the video grabbing option cyanellus species group of Entedon based conducted in Adobe Photoshop 6.0 pro- on the presence of these hardly discern- gramme through the use of the Falcon \ Ea- ible foretibial strips and the 3-segmented gle Frame Grabber. The alignment of the funicle of the male. Gumovsky (1999), photos corresponding to different layers in when revising the cyanellus group of the focus was conducted using the Combine Z genus, argued against Askew's placement programme Version 3.9 (designed by Alan of E. sylvestris based on that only about Hadley, http://www. hadleyweb.pwp. 20% of specimens in a population possess Volume 15, Number 2, 2006 235

EHT * 5.00 kV Ma9 3 59 K X 2pm Deleclor = SE1 WO = 16 mm \—\

Fig. 1. Entedon sylvestris, female, details of morphology: A, C, head in frontal view; B, antenna; D, pedicel and 1st funicular segment; E, lower face; F, anelli; cly, clypeus.

the dim foretibial stripes, whereas the species for Bulgaria and also provided majority of the specimens have fore tibiae its comparative diagnoses in their key to largely darkened, and proposed instead species. to E. — accommodate sylvestris in the costalis Comparative notes. Within the European species group (decreasing the value of fauna, E. sylvestris can easily be confused the 3-segmented funicle of male as with £. cyanellus and E. fufius. From the a species-group character). Gumovsky former species, E. sylvestris is easily distin- and Boyadzhiev (2003) reported this guishable as female has a shorter 1st of Hymenoptera Research 236 Journal

male: head in frontal Fig. 2. Entedon sylvestris: A, B, female: A, posterior mesosoma; B, metasoma; C-F, C, view; D, antenna; E, posterior mesosoma and anterior metasoma (petiole); F, body.

funicular segment (1.5-1.6 times as long —as end of the tibia (the tibia is always with broad, Figs IB, D, whereas in E. cyanellus distal pale band in E. cyanellus). The distal 1.8-5.0 times) and by the narrower anten- pale bands on the tibiae also distinguish £. tibiae nal scape of males (3.8 times as long as sylvestris from £. fufius in which all broad, Fig. 2D, about 3.0 times in E. are completely dark. Also, the 1st funicular cyanellus). Also, the foretibial stripes (when segment of females is about 2.2 times as present) are much narrower and not long as broad and 1.4 times as long as the terminated a in £. by pale area on the distal 2nd segment in £. fufius, whereas Volume 15, Number 2, 2006 237

is 1.5-1.6 times as of the then eats the sylvestris the segment surface plant pod. She long as broad and just slightly (about 1.1 egg, as in other weevil species (Kozlowski times) longer than the 2nd (Figs IB, D). 2003). Moreover, all flagellar segments are free in The larvae feed on the seeds and remain E. funicle is males of fufius (the 4-segment- in pods until they are fully grown (Figs 3C, in E. the funicle is 3- ed), whereas sylvestris D, E; Figs HA, B). The pods then split and segmented and two last funicular segments the mature larvae fall to the ground are attached the closely (forming 2-seg- (Fig. 11D). Larvae leave the host plant's mented - clava, Fig. 2D). pods at the end of June beginning of July. Distribution. — Hungary (Szelenyi 1981), They quickly bury themselves into the soil, Britain (Askew 1992), Bulgaria (Gumovsky where they prepare an earthen cell, in and Ukraine (Gu- I Boyadzhiev 2003), which they pupate (Fig. HE). have found movsky 1999). adult beetles in the dissected earthen cells in October. However, I assume that adult Ecology weevils are already present before, but The host, Ceutorhynchus sisymbrii Dieck- leave their soil/earthen cocoons only the mann. —Ceutorhynchus sisymbrii is similar to following spring. C. pulvinatus Gyllenhal, 1837 in having Host searching and —oviposition of the dense coverage of dorsum and black parasitoid, E. sylvestris. The females of E. that in rostrum, but differs from species sylvestris can be found in the field from late tibiae not the that only the are red, May until early July (Figs 4A-G). The Des- entire legs (L. Behne, pers. comm.). parasitoid female searches along the pods pite C. sisymbrii (Dieckmann 1966) being of S. loeselii for weevil larvae, by drumming described its bio- nearly fifty years ago, the pod surface with her antennae remained unknown. Dieckmann logy (Fig. 4A). Once she has located a host, she (1972) mentioned that studies on the life walks back and forth several times before of this are of value. cycle species special starting to oviposit. She bends her gaster Below I a brief of the propose synopsis downwards and briefly hooks the ovipos- field and observations on this laboratory itor saw into the plant tissues. She then species. releases the gaster so that it strengthens in Adults of Ceutorhynchus sisymbrii a position perpendicular to the ovipositor 3A, B) feed on shoots, flowers and (Figs (Figs 4B, C). Thereafter she penetrates the fresh seeds of loeselii. Sisymbrium Oviposi- pod wall with her ovipositor, and carries tion takes in The female of place May-July. out rhythmic, twisting movements to find the weevil its into the of the lays eggs pods the weevil larva, as in other parasitoid host when the seeds are plant fully grown, species. Oviposition lasts about 30-90 but still soft and (mature seeds are green seconds. Quite often the ovipositor pene- to and At the yellow orange hard). first, tration causes the host's haemolymph to female makes an in the with opening pod exude, and then females of E. sylvestris feed her rostrum. It the deeply penetrates on these excretions. rostrum into the pod (up to the base of the and eats the seed below the rostrum) Immature stages of E. sylvestris hole. Then the female turns back and — white to presses its caudal end into the prepared Egg. Elongate, transparent, opening, and begins oviposition. This without discernible sculpture. No stalks behaviour is discernable by the rhythmic or terminal bulbous projections, reported for £. found. pulsing of her gaster. Occasionally, the by Beaver (1966) ergias, were female fails to aim into the prepared Size, about 260-300 urn long and about 80- hole and lays an egg directly onto the 100 urn wide. of Hymenoptera Research 238 Journal

Fig. 3. A-E, Ceutorhynchus sisymbrii. A, B, female, C-E, mature larva.

1st instar larva. —Habitus. The first instar different forms, which were found during larva of E. sylvestris is hymenopteriform, dissection of the host larvae. One is "slim" pale (nearly transparent), has 13 body (Figs 5A, 6A), about 300 jim long and segments and a cranium. There are some about 104 |im wide (max.), and another Volume 15, Number 2, 2006 239

4. the female of Fig. A-G, Entedon sylvestris oviposits into the larva of Ceutorhynchus sisymbrii in pods of Sisymbrium loeselii: A-C, in laboratory; D-G, in the field; hole, oviposition hole of weevil female. Scale bars: A-C, 2 mm; D-G, 5 mm. form is "swollen" (Figs 6E, F), about tions, and fixation and drying circum- 450 |im long and about 206 urn wide. stances. Occasionally, specimens of an intermediate Body segments. The last, XHIth, segment form were found (Fig. 6G), with a length of bears sharp triangular tubercles along its about 260 urn and a maximum width of margin, arranged in two or three rows in about 108 fim. Despite the differences in the shape of a crown (Figs 6C, D). These size, I regard all three forms as belonging tubercles are very distinct in freshly to the same larval instar, because of the emerged larvae, but occasionally are coat- of possession the peculiar shape of the ed with secretions (Fig. 6H). These rows of head capsule (cranium) and body surface. tubercles are also distinct in the "swollen" The different body proportions are proba- larvae (Fig. 6H), however, they are not so bly caused by different nutritional condi- clearly distinguishable in these larvae due of Hymenoptera Research 240 Journal

Fig. 5. £. sylvestris, first-instar larva: A, habitus; B, E, head close-up; C, D, cauda close-up.

to the thickness of the preceding segment which consist of small curved teeth. Seg- (Figs 6E, F, H). Segments IV-XII bear ments I—III bear no distinguishable teeth or distinct dorsal semicircular serrations serration, which reflects the subdivision their — (Fig. 6C, se) along anterior margins, into thoracic (I III) and abdominal (IV- Volume 15, Number 2, 2006 241

Fig. 6. E. sylvestris, first-instar larva: A, E-G, habitus; B, head; C, D, H, cauda; se, serration.

XIII) segments. No spiracles were found on trally, with a characteristic "beak"-shaped the body of the larvae. end, which is formed by the protruding Head capsule (cranium). The head cap- palpi of the labrum. In light microscopy, sule is weakly sclerotized, narrowing ven- the labrum adopts an active backward and of Hymenoptera Research 242 Journal

Photo No. = 74 EHT= 5.00 kV Mag= 3.53 KX 2um MPI Metallforschung Date :3 Aug 2004 Detector = SE1 WD= 20 mm HH Metallographie

Photo No. = 78 5.00 kV 2.56 KX EHT= Mag= 10(jm— MPI Metallforschung Date :3 Aug 2004 Detector = SE1 WD= 20 mm H Metallographie

7. text for Fig. E. sylvestris, first-instar larva, details of the head morphology (see abbreviations).

forward motion. Antennae are absent, in- developed and arranged in a relatively of dicated only as small swellings on the fixed position (Figs 7, 8). The lateral area upper part of the head capsule. The the cranium bears three pairs of cranial sensorial structures of the head are well- palpi: the upper (cpl), the lower (cp2) and Volume 15, Number 2, 2006 243

Fig. 8. £. sylvestris, first-instar larva, details of the head morphology (see text for abbreviations). the posterior (cp3) palpi. There is a pair of each side): the upper lateral labral palpi enlarged pleurostomal palpi (pip) just (ulrp), the lower lateral labral palpi (llrp) above the labrum. The labrum bears 3 and the inner labral palpi (ilrp). A large pairs of labral palpi (grouped by one from palpus is located near each maxilla (mxp) of Hymenoptera Research 244 Journal

= = Mag 496 X 20um MPI Metallforschung o No. = 113 EHT = 5.00 kV Mag 332 X MPI Metalirorschung JOOprr WD = 18 mm = 1 .5 Aug 2004 Detector SE1 WD- 18 Metallographie

PholoNo. = 119 EHT= 5.00W Mag = 293X 100pm = —— :5 Date Aug 2004 Detector SE1 WD 18 mm I

in frontal head Fig. 9. £. sylvestris, second-instar larva, details of morphology: A, E, habitus; B, head view; C, close-up; D, mouth area, close-up; F, mouth area, lateral view; pip, pleurostomal palpi; lb, labrum, md, mandible.

and behind them there is a pair of smaller 590 (im long. However, the proportions labial palpi (lbp). A pair of comparatively- and shape of the segments are different: long pharyngeal setae (phs) is situated the larva is more robust (Figs 9A, E) and behind the latter palpi. none of the segments has serrations. The 2nd instar larva. —The second instar larva last segment bears no "caudal crown". The of E. si/lvestris is also hymenopteriform and head is more spherical and the sensorial pale, with similar body segmentation (13 structures of the head are less clear body segments and the head capsule), 500- (Figs 9C, D). However, it is still possible Volume 15, Number 2, 2006 245 to discern the pleurostomal palpi, the about 3 cm), and the host larvae can delimited labrum and short mandibles probably pupate even deeper. (about 20 jam long). No spiracles are The is black, obtect, with distinct recognizable. — outlines of head, mesosoma, metasoma, Final instar larva. Dissections of the wings, legs and antennae. The average dead host larvae revealed two forms of length is 1.8-1.9 mm, the width of the head - the parasitoid larvae, which differ from is about 0.6-0.7 mm, of mesosoma - each other by their size and body shape. 0.8 mm, of metasoma 0.9 mm. The last The smaller form fits the size range and larval skin, covered by soil particles, is morphological peculiarities of the second often attached to the caudal end of the instar larva, described above. The larger pupa. form found within the body of the dead, Parasitoid-host relations. —Eggs of E. buried host larvae, were 0.96-1.02 mm sylvestris were found singly in the host long (Fig. 10A), which is nearly twice as larvae. When the host larvae were dis- long as the second instar, but still nearly sected, the parasitoid eggs and first instar half of the average length of the pupa larvae were found free-floating anywhere (1.8 mm). It has large (46 urn long) and in the host body cavity. No attempts to heavily sclerotized (dark brown in color) parasitize the host eggs were recorded. mandibles and discernible hypostome Parasitoid larvae hatch about one day (Figs 10D, F). I regard these peculiarities after being laid. Females of E. sylvestris to be specific to the final instar. The oviposited into weevil larvae of various sensory structures of the head are hardly instars, but only second instar larvae of the distinguishable, apart from large antennae parasitoid were found within the mature larvae. larva situated on broad swellings (Figs 10B, C). host Only one parasitoid per Spiracles were difficult to distinguish host larva was found in all studied because the skin of the larva was too samples. Parasitoid larvae could some- the host. shriveled. The large antennae are peculiar times be observed while inside to the final instar larvae of Chalcidoidea The larvae of £. sylvestris never emerged (Parker 1924). Fisher (1970) also draws the from the host's body within the pods of S. large round areas (equivalent to the large loeselli. Behaviour and pupation of parasit- antennae) on the upper part of the head of ized host weevil larvae did not differ from the final instar larva of E. rumicis, but does unparasitized. not mention them in the text. The posses- DISCUSSION sion of large antennae is one of the characters supporting the assumption that Life history of Entedon sylvestris this is the final instar larva of E. sylvestris. size of the larva is The smaller body pro- Females of Entedon sylvestris attack their artificial resulted bably and from incom- hosts, the weevils Ceutorhynchus sisymbrii, of the dead larvae. at their larval The larvae of C. plete swelling— dried, stage. Pupa. Generally, pupation takes place sisymbrii feed on the seeds of the small in the host's earthen cell (Fig. 11F). Some- tumbleweed mustard, Sisymbrium loeselii times, pupae could be found directly in soil L., and leave the host plant's pod when samples, which suggest that parasitoid mature. Unlike most other parasitoids of pupation can also take place without the seed-feeding weevils, which finish their successful creation of an earthen cell by the ontogenesis and kill the hosts inside the host. In the laboratory, the host's earthen host plant pods (e.g. Trichomalus spp., cells and the "free" pupae were found 2- Mesopolobus spp., Necremnus spp.), the 3 centimeters from the upper soil level (the larva of E. sylvestris is in its second instar overall height of the soil level in tubes was when the the host larva leaves the pod. The of Hymenoptera Research 246 Journal

lateral Fig. 10. A-E. E. sylvestris, final instar larva (isolated from the dead host larva): A, habitus, B, head, view; C, antenna enlarged; D, head, frontal view; E, mandibles; ant, antenna; md, mandible. Volume 15, Number 2, 2006 247

larva of E. isolated from the Fig. 11. A, B, the mature larva of Ceutorhynchus sisymbrii; C, first-instar sylvestris host's body; D, the dead weevil larvae isolated from the soil sample and treated by the lactic acid solution; E, the the of £. in the mature larva of C. sisymbrii creates an earthen cell (the larva is visible inside); F, pupa sylvestris

0.3 1 mm. host's earthen cell; par, parasitoid larva. Scale bars: A, 3 mm; B, 0.3 mm; C, mm; D, E, F, of Hymenoptera Research 248 Journal

into the final The anterior of the clypeus is less moult of the parasitoid larva margin in these than in E. instar and final consumption of the host produced species observa- the host's un- and E. cyanellus. Further take place within body, sylvestris takes outside tions on the emergence procedure may derground. Pupation place reveal other behavioral and of the remnants of the host's body, also peculiarities least of 2- functional devices for the species of En- underground, at a depth of at tedon an in 3 centimeters. The underground pupation having "underground span" their life means that adult parasitoids must pene- cycles. trate at least 2 cm of soil to emerge the , „ Larval morphologyr o; following spring. the first- Underground pupation is rather rare Beaver (1966) mentioned that to the has 12 but within Chalcidoidea, perhaps due instar larva of E. ergias segments, small body size of these insects and the last segment bears a dorsal plate with with of on its Fisher corresponding problems emergence a series sharp spines edge. the first instar from soil. Apart from Entedon sylvestris, (1970), when describing E. mentioned that it has the pupation and consequent emergence larva of rumicis, from the earthen cells of its host, is the head capsule and 13 body segments, a recorded for E. cyanellus Dalman, parasit- with "the last abdominal segment ventrally with oid of Tichius quinquepunctatus (L.) (Gu- with a sclerotized plate irregularly bi- The same author also movsky 1997). However, no detailed spinous edges". is available for E. that the first-instar larvae of E. ological information mentioned of its refer E. have a similar cyanellus, and records biology pharnus and philiscus body E. Since the first-instar larvae of other just to labels of museum specimens. So, shape. chalcid of have 13 Beaver's sylvestris is the only species species body segments, of 12 known biology, which begins its ontogen- statement of the possession only body esis above the ground and ends it un- segments by the first instar larva of E. is erroneous. derground. ergias likely of the Both known species for which unde- The morphological peculiarities include the ground pupation is known or suspected (E. first-instar larva of E. sylvestris the XIII sylvestris, E. cyanellus), have the produced notable indentation on (last) body This indenta- anterior margin of the clypeus (Fig. 1, cly). segment (Figs 6C, D, 8C, D). to the with The produced clypeus is diagnostic for the tion is likely equivalent plates of Entedon for the first-instar cyanellus and costalis groups spinous edges reported and E. (Gumovsky 1997), and occasionally present larva of E. ergias (Beaver 1966) also be in species of some other groups (E. kerteszi rumicis (Fisher 1970), and may Erdos, E. occidentalis Girault, E. diabolus assumed for the larvae of E. pharnus and as similar to the Rasplus). The function of the shape of the E. philiscus (mentioned The clypeus is unknown. The adult chalcids larva of E. rumicis by Fisher 1970). "caudal feed by sucking liquids through the use of tubercles of the indentation (the their labio-maxillary complex. Neither the crown") is likely homologous to the small clypeus nor the mandibles are actively curved teeth of the dorsal semicircular involved in this process. One of the serrations (Fig. 6C) along the anterior possible functions of the produced clypeus margins of the segments IV-XII. The for these may be associated with facilitating emer- function of the "caudal crown" gence from the ground. Entedon rumicis, E. larvae remains obscure, but there is no within pharnus and E. philiscus Walker also have evidence of its usage for movement larvae the anterior margin of the clypeus pro- the body cavity, since the parasitoid duced, but their adults emerge from the observed within the host's body are gen- host plant's stems, not soil (Fisher 1970). erally passive. Volume 15, Number 2, 2006 249

im Zentrum fur und Nutzungs- The first-instar larva of E. sylvestris is, to Agrarlandschafts- forschung (ZALF)) who kindly confirmed my identi- some extent, similar to the larva of Mis- fication of C. I am also grateful to Drs. described Parker and sisymbrii. cogaster sp. by Stanislav Gorb, Ulrike Wegst, Sabinne Kuhnemann larvae are Thomson (1925). Both apneustic, and Jiirgen Siewert (MPI) for their help in conducting

I like to Dr have the cuticular spines arranged in the SEM work. would also acknowledge Hariet Hinz (CABI-Bioscience, European Station, encircling lines (striations) on the body Delemont, Switzerland) for the suggestions on the segments (IV-XII) and peculiar indentation biology of Ceutorhynchus weevils. Special thanks to semi-circular or of the last body segment (a Drs Hariet Hinz, Andrew Polaszek (BMNH) and bilobed "caudal crown"). However, the Gavin Broad for careful reading of the manuscript tubercles of the "caudal crown" are sub- and suggested corrections. in the larva of E. equal in length sylvestris LITERATURE CITED (two inner spines are stouter than the of the "caudal crown" M. A. 1972. Some studies on surrounding spines Abedin, J. and Quayum. struc- the Entedon Chalcidoidea), an in Miscogaster sp.) and the sensorial sp. (Eulophidae: of the tures of the head have a somewhat differ- important parasite jute weevil, Apion corchori Marsh. Pakistan Journal of Science, Lahore ent arrangement. 23: 135-138. data is available for a few Although only Askew, R. R. 1991. Review of species of Entedon with species, the "caudal crown" is likely in- Dalman having a complete frontal fork herent to all first instar larvae of Entedon. redefinition of the species-group cioni Thomson (Hymenoptera: Eulophidae). Entomologica Scandi- This character is quite remarkable and rare navian 22: 219-229. in Chalcidoidea and Thomson (Parker Askew, R. R. 1992. Additions to the British list of thus be used later to 1925), and may Entedon Dalman (Hym., Eulophidae) with de- support monophyly of Entedon, if found scriptions of three new species. Entomologist's 128: 119-128. in other species of the genus. Also, the Monthly Magazine Askew, R. R. and J. -P. 1989. Entedoninae combination of a bare and the Kopelke. nearly body associated with Pontania and allied "caudal crown" facilitate may separating (Hymenoptera: Eulophidae and ). of this from other en- the larvae genus Entomologica Scandinavica 19: 431-434. doparasitoid larvae. Furthermore, the Beaver, R. A. 1966. The biology and immature stages of Entedon proper affiliation of the first-instar larva leucogramma (Ratzeburg) (Hymenop- tera: Eulophidae), a parasite of bark beetles. of E. sylvestris to a peculiar larval type Proceedings of the Royal Entomological Society of a revision of the current classifi- requires London (A) 41: 37-41. cation (Parker 1924) of these types in Boucek, Z. and R. R. Askew. 1968. Palaearctic Chalcidoidea. Eulophidae sine . Index of Entomoph- agous Insects 3: 260 pp. Delucchi, V., Remaudiere, ACKNOWLEDGMENTS G. Le Francois, Paris. Dieckmann, L. 1966. Die Mitteleuropaischen Arten der This work was in part supported by the grants of Gattung Neosirocalus Ner. et Wagn. (mit Beschrei- the following agencies and organizations: SRSF (State bung von drei neuen Arten). Entomologische Fund of Fundamental Research, Ukraine, grant Blatter 62: 82-110. No. 05.07/00078), SCOPES Institutional Partnership Dieckmann, L. 1972. Beitrage zur Insektenfauna der - project "Emphasising Classical and Conservation DDR: Coleoptera Curculionidae: Ceutorhynch- Biological Control in Research and Teaching" (7 IP inae. Beitrage zur Entomologie 22: 1-128. 65648), DAAD (Deutcher Akademischer Austausch- Darling, D. C. 1992. The life history and larval - dienst German Academic Exchange Service, grant morphology of Aperilampus (Hymenoptera: Chal- - 322 A/04/15867) and the SYNTHESYS Project (GB- cidoidea) with a discussion of the phylogenetic TAF-535). The help of Drs. Yakiv Didukh and affinities of the Philomidinae. Systematic Entomol- Antonina U'ins'ka (Kholodny Institute of Botany, ogy 17: 331-339. NUAS, Kiev) in identification of S. loeselii and other Darling, D. C. 1995. New species of Krombeinius Brassicaceae is also much appreciated, as well as the (Hymenoptera: Chalcidoidea: Perliampidae) from assistance of Dr. Stefan Schmidt during the author's Indonesia, and the first description of first-instar visit to Munchen. My sincere appreciation is to Dr larva for the genus. Zoologische Mededelingen 69: Lutz Behne (Deutsches Entomologisches Institut (DEI) 209-229. 250 Journal of Hymenoptera Research

Ferriere, C. 1939. Chalcid flies attacking noxious Lepidium draba and L. appelianum. Annual Report beetles in India and New Guinea. Bulletin of 2003. CABI Bioscience Centre Switzerland, Dele- Entomological Research 30: 163-168. mont, Switzerland. of M. W. 2003. of Fisher, J. P. 1970. The biology and taxonomy some Kozlowski, Consumption own eggs by chalcidoid parasites (Hymenoptera) of stem- curculionid females (Coleoptera: Curculionidae: living larvae of Apion (Col., Curculionidae). , Ceitorhynchinae). Weevil News 10: Transactions the 4, of Royal" Entomological Society of http://www.curci.de/WeevNews/KOZLOW_ London 122: 293-322. A.htm. Graham, M. W. R. de V. 1971. Revision of British Parker, H. L. 1924. Recherches sur les formes Entedon (Hymenoptera: Chalcidoidea) with de- postembryonaires de chalcidiens. Annates de la scriptions of four new species. Transactions of the Societe Entomologique de France 93: 261-379. Royal Entomological Society of London 123: 313-358. Parker, H. L. and W. R. Thompson. 1925. Notes on the Gumovsky, A. V. 1997. Review of Cionus weevils larvae of the Chalcidoidea. Annals of the Entomo- (Coleoptera, Curculionidae) with notes on their logical Society of America 18: 384-395. chalcidoid parasitoids (Hymenoptera, Chalcidoi- Peck, O. 1963. A catalogue of the Nearctic Chalcidoi- dea). Journal of the Ukrainian Entomological Society dea (Insecta; Hymenoptera). Canadian Entomolo- 3: 49-72. gist (Supplement) 30: 1-1092. Gumovsky, A. 1999. Review of the genus Entedon Rasplus, J.-Y. 1990. Nouvelles especes afrotropicales Dalman, 1820 (Hymenoptera, Eulophidae, Ente- du genre Entedon Dalman et notes sur leur doninae). 5. Revision of the cyanellus species biologic Bulletin de la Societe Entomologique de group. Annates historico-naturales Musei nationalis France 94: 223-245. 1999 hungarici (91): 141-176. Stubbendieck, J., G. Y. Friisoe, and M. R. Bolick. 1994. Gumovsky, A. and P. Boyadzhiev. 2003. Review of the Weeds of Nebraska and the Great Plains. Nebraska Bulgarian Entedon Dalman, 1820 (Hymenoptera: Department of Agriculture, Bureau of Plant Eulophidae, Entedoninae). Acta Entomologica Bul- Industry, Lincoln, Nebraska. 589 pp. garia 55 (3): 3-32. Szelenyi, G. 1981. On the chalcidoid flies of the H. L. E. Hinz, and Gerber. 1998. Investigations on Hortobagy, I. Eulophidae (Hymenoptera). The potential biological control agents of garlic Fauna of the Hortobagy National Park 1: 275-284. mustard, Alliaria petiolata (Bieb.) Cavara & Mahunka, S., ed. ISBN 963-05-2519-4. Grande. Annual 1998. CABI Bioscience K. Report Tiwari, N. 1976. Observations on Entedon sp. Centre Switzerland. Switzerland, Delemont, (Hymenoptera: Eulophidae) parasitising Frauen- H. M. B. L. Hinz, L., Cripps, J. Renteria, and A. feldiella jamboolii Grover. Indian journal of Ento- Winspurdy. 2004. Biological control of whitetops, mology 37: 197-198. J. HYM. RES. Vol. 15(2), 2006, pp. 251-265

Phylogenetic Analysis of Chaenusa sensu lato (Hymenoptera: Braconidae) using Mitochondrial NADH 1 Dehydrogenase Gene Sequences

Robert R. Kula,* Gregory Zolnerowich and Carolyn J. Ferguson

(RRK, GZ) Department of Entomology, Kansas State University, Manhattan, KS 66506, USA (CJF) Division of Biology, Kansas State University, Manhattan, KS 66506, USA f Current address and address for correspondence: Systematic Entomology Laboratory, USDA, c/o U.S. National Museum of Natural History, MRC-168, 10th & Constitution Ave, N.W., Washington, DC 20560-0168, USA.

Abstract.— Alysiinae currently contains over 1,500 described species and is divided into the tribes within Alysiini and Dacnusini. There is disagreement on how species should be grouped Dacnusini, and Chaenusa Haliday is a prime example. Chaenusa sensu lato is defined by the presence of setae on the compound eyes (Griffiths 1964). Alternatively, Riegel (1950, 1982) treated Chaenusa s.I. as three genera, Chaenusa sensu stricto, Chorebidea Viereck, and Chorebidella Riegel, and differentiated the genera primarily using forewing venation and shape of the forewing stigma. Phylogenetic analyses using molecular data have not been undertaken. Therefore, we assessed the and Chorebidella monophyly and interspecific relationships of Chaenusa s.I., Chaenusa s.s., Chorebidea, through maximum parsimony, maximum likelihood, and Bayesian analyses using mitochondrial NADH 1 dehydrogenase gene sequences. Chaenusa s.I. and Chorebidea were not monophyletic in any of the analyses, but four of five species of Chorebidea always formed a clade. Further, Chaenusa s.s. and Chorebidella were monophyletic in all analyses and were always sister taxa. The results of this s.I. and study largely support Riegel's (1950, 1982) treatment of Chaenusa as Chaenusa s.s., Chorebidea, Chorebidella. However, we suggest that Chaenusa s.I. be retained until additional phylogenetic In addition to the analyses have been undertaken to confirm the relationships inferred in this study. phylogenetic analyses, we discuss the morphological features relevant to Griffiths' definition of Chaenusa s.I. and Riegel's definition of Chaenusa s.s., Chorebidea, and Chorebidella.

Alysiinae currently contains over 1,500 (a plesiomorphy). Dacnusini is considered described species, and estimates of global monophyletic based on the absence of richness range from 2,900 to 5,300 species forewing vein r-m (an apomorphy) (Grif- (Dolphin and Quicke 2001). The mono- fiths 1964, Shaw and Huddleston 1991, phyly of Alysiinae is firmly established Wharton 1994) and has consistently been based on the possession of exodont mand- recognized, although at different hierar- ibles and the complete loss of the occipital chal levels, since Forster (1862). There is carina (Griffiths 1964, Shaw and Huddle- widespread disagreement on how species ston 1991, Wharton 1997). Host records should be grouped within Dacnusini, and suggest that all alysiines are koinobiont Chaenusa Haliday is a prime example. endoparasitoids of cyclorrhaphous Diptera Nixon (1943) divided dacnusines with (Shaw and Huddleston 1991, Wharton and setiferous compound eyes (Fig. 1) into Austin 1991, Wharton 1997). two genera, Chaenusa and Chorebidea Vier- Two tribes are currently recognized in eck, and differentiated the genera using of the fore- Alysiinae: Alysiini and Dacnusini. Alysiini forewing venation and shape is probably nonmonophyletic as it is de- wing stigma. Riegel (1950) established fined by the presence of forewing vein r-m Chorebidella Riegel, a third genus contain- of Hymenoptera Research 252 Journal

RS+M at least partially present and a "long" stigma were considered Chorebi- with RS+M absent and dea (Fig. 3); species a "short, wide" stigma were considered Chorebidella (Fig. 4). Griffiths (1964) hy- with setifer- pothesized that all dacnusines and ous eyes form a monophyletic group synonymized Chaemtsa sensu stricto, Clior- ebidea, and Chorebidella (i.e., Chaemtsa sensu

lato). However, Riegel (1982) disagreed with Griffiths' synonymies and continued to treat Chaemisa sensu Griffiths (1964) as the com- three genera. Riegel (1982), only of North American prehensive treatment included several species of Chaemisa sJ., s.s. and new species in Chaemisa Chorebidea, but Wharton (1997) followed Chaemisa sensu Griffiths (1964) rather than Chaemisa sensu Riegel (1982). With 29 described species worldwide, Chaemisa s.l. is small relative to other over 240 of dacnusine genera (e.g., species are Chorebus Haliday). Nearly all species Nearctic or Palaearctic, but three species are known from , and one species each is known from Madagascar and in the Argentina. As far as is known, flies ephydrid genus Hydrellia Robineau-Des- voidy are exclusively utilized as hosts (Griffiths 1964, Shaw and Huddleston - 1991, Wharton and Austin 1991, Wharton

sensu is an for Figs 1-4. Chaemisa sensu lato, Chaemtsa stricto, 1997). Hydrellia important group Chorebidea, and Chorebidella. 1, Chorebidea americana, classical biological control of aquatic setiferous 2, Cha. 1st compound eyes. quadriceps, weeds. For example, Hydrellia pakistanae subdiscal cell closed. 3, Chorebidea saxicola, 1st sub- have Deonier and Hydrellia balciunasi Bock discal cell open, RS+M partially present, and stigma been and released for control of "long". 4, Chaemisa sp. 3, 1st subdiscal cell open, imported = verticillata in the RS+M absent, and stigma "short, wide", a 1st Hydrilla (L.f.) Royle = = subdiscal cell, b RS+M, and c stigma. United States. However, Hydrellia also contains species that are rice pests, such and ing dacnusines with setiferous eyes. Like as Hydrellia griseola (Fallen) Hydrellia of s.l. Nixon (1943) Riegel (1950) differentiated philipvina Ferino. Species Chaemisa control the genera primarily using forewing vena- may hinder classical biological tion and shape of the forewing stigma. programs as contaminants in the quaran- Riegel (1950) regarded all dacnusines with tine phase (Wharton 1997) or through setiferous eyes and a closed 1st subdiscal parasitism (by endemics) of introduced of cell as Chaemisa (Fig. 2); he segregated natural enemies. Conversely, species dacnusines with setiferous eyes and an Chaemisa s.l. may be important natural open 1st subdiscal cell into Cliorebidea or enemies of pest flies (Natarajan and Chorebidclla. Species with forewing vein Mathur 1980). Volume 15, Number 2, 2006 253

source Table 1. Species analyzed in this study and their respective taxonomic placements, locality data, accession = Cimarron National KPBS = repositories or collectors, and GenBank numbers. CNG Grassland, = Konza Prairie Biological Station, and SFF Santuario de Fauna y Flora.

Species of Hymenoptera Research 254 Journal

Belshaw et al. 1998, Dowton cussed in this paper. The holotype of Quicke 1994, Brethes is housed et al. 1998, Shi et al. 2005). Griffiths (1964) Gyrocampa pallidinervis that of in the Museo Argentina de Ciencias Nat- suggested species Dapsilarthra (Aly- Dacnusini be urales (MACN). The first author made siini) and might closely did not related based on of leaf-mining multiple requests, but the MACN parasitism of almost loan the holotype. agromyzids. Species Dapsilarthra attack The ingroup was composed of either 13 exclusively leaf-mining agromyzids 13 of 1984, 1997), and dacnusines that species of Dacnusini or species (Wharton Griffiths considered Dacnusini and one species of Alysiini (1964) morphological- are of leaf- depending on the analysis. Nine species ly plesiomorphic parasitoids In with of Chaenusa s.l. were included, with Chae- mining agromyzids. analyses Opius 1 used to root trees, 1 nusa s.s., Chorebidea, and Chorebidella repre- sp. Dapsilarthra sp. was included in the to the sented by two, five, and two species, ingroup explore were of Dacnusini. 1 respectively. Undescribed species monophyly Dapsilarthra sp. to root trees in that considered Chaenusa s.s., Chorebidea, or was used analyses Chorebidella based on forewing configura- excluded Opius sp. 1. n. 1-3 will be described DNA isolation, tion. Chaenusa sp. — amplification, sequencing, in a taxonomic revision of New World and alignment. Genomic DNA was iso- individual a DNea- Chaenusa s.l. (Kula in preparation). Chae- lated from wasps using to undescribed Tissue Kit to the nusa sp. 1 and 2 appear be sy® (Qiagen) according and manufacturer's for insects. Most species but are only known from one protocol but two individuals, respectively. Thus, RRK specimens were ethanol-preserved, awaits the discovery of additional speci- several were dried, pinned specimens up chain reaction mens before describing them. Evaluation to 14 years old. Polymerase of the literature for Old World Chaenusa s.l. (PCR) amplifications and sequencing reac- an Re- suggests that Chaenusa sp. 3 is also unde- tions were performed using MJ scribed. search PTC-200 thermal cycler. A portion PCR Two species each from Chorebus and of the ND1 gene was amplified using Coelinius Nees were also treated as ingroup set up in 25 liI volume. Oligonucleotide taxa to test the monophyly of Chaenusa s.l. primers (ND1F: 5'-GATAAATCAAAW- Species of Chorebus and Coelinius possess GGKGT-3', ND1R: 5'-CAACCTTTTAGT- and the were as morphological features (i.e., eye setation, GATGC-3') PCR program number and position of mandibular teeth, in Smith et al. (1999) except the annealing metapleural setation, metasomal compres- temperature was optimized at 47 C. PCR sion) that suggest the potential for a close products were purified using a Qiaquick® to relationship with certain species of Chae- PCR Purification Kit (Qiagen) according nusa s.l. (Kula personal observation). Both the manufacturer's protocol. Both strands se- species of Chorebus fit in the affinis group of all purified PCR products were (Griffiths 1968), and both species of Coeli- quenced using the PCR primers as se- nius fit the concept of Lepton Zetterstedt quencing primers. Sequencing reactions = Coelinidea in Griffiths in 10 volume ( Viereck) (1964) (as were performed jal using a subgenus) and Riegel (1982) (as a genus). an ABI Prism® BigDye™ Terminator v3.0 A species of either Opius Wesmael or Ready Reaction Cycle Sequencing Kit Dapsilarthra Forster was specified as the (Applied Biosystems) according to the outgroup to root trees depending on the manufacturer's protocol. Sequencing reac- analysis. Previous phylogenetic analyses tion products were purified using spin support a sister group relationship be- columns filled with Sephadex® (Amer- tween Alysiinae and Opiinae (Quicke and sham Biosciences), dehydrated by vacuum van Achterberg 1990, Wharton et al. 1992, centrifuge, and sent to the DNA Sequenc- Volume 15, Number 2, 2006 255

ing & Synthesis Facility at Iowa State using the heuristic search option with University for gel runs on an ABI Prism® stepwise addition, 50 random addition 3700 DNA Analyzer (Applied Biosystems). sequence replicates, and TBR branch swap- Sequences generated from the forward and ping were used. reverse primers were aligned and edited in Bayesian analyses were performed using Sequencher™ 4.1.2 (Gene Codes Corpora- MrBayes 3.0b4 (Huelsenbeck and Ronquist tion) to acquire a consensus sequence for 2001). Modeltest 3.06 was used to deter- each species. Consensus sequences were mine the model of molecular evolution that manually aligned in SeqPup 0.6 (Gilbert best fit the data, and subsequently, Bayes- 1996) to produce a DNA sequence data ian analyses were performed for the DNA matrix. The DNA data matrix was trans- data matrix. The data matrix was parti- lated to construct an amino acid (AA) tioned by codon position (i.e., 1st, 2nd, sequence data matrix using the Drosophila 3rd), and among-site rate variation was set Fallen mtDNA genetic code in MacClade (as a prior) to allow variable rates across 4.06 (Maddison and Maddison 2003). partitions. The model of nucleotide sub- DNA and AA characteristics and stitution and rate variation was sequence— among-site phylogenetic analysis. The number of con- set as determined using Modeltest. The stant, variable parsimony uninformative, following model parameters were un- and parsimony informative characters linked across the partitions: substitution were determined using PAUP* 4.0M0 rates of GTR model, character state fre- (Swofford 2002), as were mean base fre- quencies, gamma shape parameter, and quencies. PAUP* 4.0M0 was also used to proportion of invariable sites. Each run test for significant heterogeneity of base consisted of 1,000,000 generations with frequencies across taxa; base frequencies a random starting tree and sample fre- were considered significantly heteroge- quency of every 100 generations. The < neous if P 0.05. burnin was determined by constructing X Maximum parsimony (MP) and maxi- an XY scatter plot (i.e., generation log mum likelihood (ML) analyses were per- likelihood value) using Microsoft® Excel to formed using PAUP* 4.0bl0. Maximum determine the number of generations until parsimony analyses were conducted for log likelihood values stabilized. Trees the DNA and AA data matrices using the sampled prior to the generation at which branch and bound algorithm. Modeltest log likelihood values stabilized were not 3.06 (Posada and Crandall 1998) was used included in the consensus tree. A 50% to determine the model of molecular majority-rule consensus of the retained evolution that best fit the data, and sub- trees, showing the frequency of all ob- sequently, ML analyses were conducted for served bipartitions (i.e., posterior probabil- the DNA data matrix using the heuristic ities), was constructed using PAUP* search option with stepwise addition, 100 4.0bl0. random addition sequence replicates, and Maximum parsimony analyses with tree bisection-reconnection (TBR) branch Chaenusa s.l. constrained as monophyletic swapping. If the Hierarchical Likelihood were also performed for the DNA data Ratio Test (hLRT) and the Akaike Informa- matrix. The search parameters were the tion Criterion (AIC) in Modeltest selected same as for unconstrained MP analyses as different models, ML analyses were per- discussed above. Most parsimonious trees formed using each model. Support for (MPTs) from unconstrained and con- individual clades was assessed via boot- strained analyses were compared statisti- strap analyses. For MP 1,000 pseudorepli- cally using the "Compare-2" permutation cates with the branch and bound algorithm test (Faith 1991) in PAUP* 4.0M0. Under were used. For ML 100 pseudoreplicates MP each of 10,000 random matrices (with of Hymenoptera Research 256 Journal

Table 2. Number of constant (C), variable and parsimony uninformative (VPU), parsimony informative (PI) characters for all nucleotide (nuc) and amino acid (AA) sites and positions (Pos) 1, 2, and and included 3 with Opius sp. 1 excluded (excl) (incl).

Dataset Volume 15, Number 2, 2006 257

n. 3 Table 4. Results of tests for significant hetero- the analysis. Further, Chaenusa sp. of base across taxa for all sites and geneity frequencies either formed a clade with Chorebus sp. 1 1 excluded positions (Pos) 1, 2, and 3 with Opius sp. and Chorebus sp. 2 or was sister to the rest (excl) and included (incl). of the ingroup. Chorebidea was not mono- Dataset P phyletic, although four of five species of Chorebidea included in the analysis formed All sites {Opius sp. 1 excl) 17.509269 0.99882241 0.99908575 a clade with 94% bootstrap support. Chor- All sites (Opius sp. 1 incl) 19.096701 ebus was in one but Pos 1 {Opius sp. 1 excl) 12.619144 0.99998081 monophyletic tree, 1 13.202722 0.99999385 Pos 1 (Opius sp. incl) bootstrap support was <50%. Chaenusa s.s., Pos 2 1 2.302238 1.00000000 (Opius sp. excl) Chorebidella, and Coelinius were monophy- 1.00000000 Pos 2 (Opius sp. 1 incl) 2.338632 letic with 97%, 100%, and 75% bootstrap Pos 3 (Opius sp. 1 excl) 53.386385 0.06219822 Pos 3 (Opius sp. 1 incl) 60.168328 0.03417160 support, respectively. Bootstrap support for the relationships among these clades for the sister phyletic. In both trees the Coelinius clade was <50% except group was sister to the clade formed by four of relationship between Chaenusa s.s. and the five species of Chorebidea included in Chorebidella (99% bootstrap support). of Hymenoptera Research 258 Journal

Table 5. Bootstrap support and posterior pro- recovered babilities for groups within Dacnusini and through maximum parsimony (MP) Bayesian 1 included. Maximum analyses with Opius sp. with parsimony analyses were conducted position nr = (Pos) 3 included (incl) and excluded (excl). groups not recovered. Volume 15, Number 2, 2006 259

Coeliniusferruginea 71 Coelinius

Coelinius hopkinsii

Chorebus sp. 1 80 Chorebus ^^ Chorebus sp. 2 • <50 ^™ Chaenusa n. sp. 3 1 Chorebidea <50

— 1 Dapsilarthra sp. 1 Alysiini <50 ^

Chaenusa n. sp. 1 75 Chaenusa s.s.

Chaenusa quadriceps 98

Chaenusa hergi 99 <50 Chorebidella

L Chaenusa sp. 3

Chaenusa sp. 2 98 I

79 Chaenusa n. sp. 2 Chorebidea

67 ( hacnusa sp. 1

Chorebidea americana

1 0/;///.v sp. 1 Opiinae (Outgroup) 0.05 substitutions/site

of the data matrix with 1 Fig. 6. Phylogram resulting from maximum likelihood analysis DNA Opius sp. included. Bootstrap values are above branches.

= unconstrained analysis are not significant- MPT from the constrained analysis (P ly shorter than either of the two MPTs from 0.031300). = the constrained analysis (P 0.125100, Maximum parsimony analysis of the AA 0.127600, 0.162700, 0.165300). Maximum data matrix with Opius sp. 1 included = 161 parsimony analysis with Opius sp. 1 in- resulted in two MPTs (tree length = = cluded and Chaenusa s.l. constrained as steps, CI 0.7087, RI 0.7176). Dacnusini monophyletic resulted in one MPT (tree was not monophyletic in the strict consen- = = length 444 steps, CI 0.5344, RI sus of the two MPTs (Fig. 7). Rather, 0.5499) (tree not shown) nine steps longer Dapsilarthra sp. 1 was sister to Chaenusa n. re- than the MPT from the unconstrained sp. 3, but bootstrap support for this s.l. not analysis. The "Compare-2" test revealed lationship was <50%. Chaenusa was that the MPT from the unconstrained monophyletic. Chorebidea was not mono- of analysis is significantly shorter than the phyletic, although four of five species of Hymenoptera Research 260 Journal

Coelinius ferruginea <50/53 Coelinins

Coelinins hopkinsii

Chorebus sp. 1 63/69 Chorebus

Chorebus sp. 2

Chaenusa n. sp. 1 98/96 <50/* Chaenusa s.s. Chaenusa quadriceps 94/95

Chaenusa bergi 100/100 Chorebidella <50/<50 Chaenusa sp. 3

n. 3 Chaenusa sp. 1 Chorebidea <50/<5<] <50/na

1 Dapsilarthra sp. 1 Alysiini

Chaenusa sp. 2

96/96 Chaenusa sp. 1 Chorebidea 93/94 Chaenusa n. sp. 2

( Chorebidea americana

Opius sp. 1 1 Opiinac (Outgroup)

trees from maximum of Fig. 7. Strict consensus of two most parsimonious resulting parsimony (MP) analysis are above branches the amino acid data matrix with Opius sp. 1 included. Maximum parsimony bootstrap values values are and left of slashes. Where clades were recovered in MP analysis with Opius sp. 1 excluded, bootstrap in with 1 above branches and right of slashes. The asterisk indicates a clade not recovered MP analysis Opius sp. = excluded, na not applicable.

of Chorebidea included in the analysis formed sus tree in Fig. 7 except the monophyly a clade with 93% bootstrap support. Chae- Chaenusa s.l. was unresolved (see asterisk in nusa s.s., Chorebidella, and Chorebus were in Fig. 7). Bootstrap support is reported monophyletic with bootstrap support of Fig. 7. 98%, 100%, and 63%, Coelinins respectively. DISCUSSION was monophyletic, but bootstrap support — was <50%. Bootstrap support for the DNA sequence characteristics. Character- relationships among these clades was istics of the ND1 DNA sequences in this <50% except for the sister group relation- study are consistent with ND1 DNA ship between Chaenusa s.s. and Chorebidella sequences of other braconids (e.g., Smith (94% bootstrap support). and Kambhampati 1999, Smith et al. 1999, Maximum parsimony analysis of the AA Michel-Salzat and Whitfield 2004). As in data matrix with Opius sp. 1 excluded the aforementioned studies, the sequenced = resulted in six MPTs (tree length 142 fragments in this study (including Opius = = steps, CI 0.7273, RI 0.7479). In terms of sp. 1) are biased towards adenine (35.7%) the relationships among dacnusines, the and thymine (46.2%), particularly in the strict consensus of the six MPTs (tree not first (82.7%) and third (93.5%) positions. shown) was identical to the strict consen- Significant heterogeneity of base frequen- Volume 15, Number 2, 2006 261

cies was detected for position 3 when Opins the AA data matrix with Chaenusa n. sp. 3 sp. 1 was included. As mentioned in excluded, Dacnusini was monophyletic in Michel-Salzat and Whitfield (2004), the two of six MPTs, but bootstrap support A+T nucleotide bias observed for insect was <50% (results not presented). Dacnu- mitochondrial DNA could influence the sini was not monophyletic in ML analysis level of homoplasy, particularly in the first of the DNA data matrix with Chaenusa n. and third positions. However, 27.2% and sp. 3 excluded (results not presented). 58.8% of the parsimony informative char- Thus, ND1 DNA sequences and the ab- acters in the DNA data matrix with Opius sence of forewing vein r-m largely, but not sp. 1 included are in the first and third conclusively, support the monophyly of positions, respectively. Therefore, for most Dacnusini. Exclusive utilization of plant- analyses all positions were considered and mining flies as hosts, particularly leaf- and were not differentially weighted. We per- stem-mining agromyzids, chloropids, and formed MP and bootstrap analyses with ephydrids (i.e., biological homogeneity), Opius sp. 1 included and position 3 provides further indication that Dacnusini excluded to examine the influence of is monophyletic. However, more extensive position 3 on tree topology and branch taxon sampling and the use of additional support. The exclusion of position 3 had no markers more conserved than ND1 are influence on tree topology but resulted in needed to confirm the monophyly of lower bootstrap support for several clades. Dacnusini and resolve the more ancient Conversely, there was a slight increase in divergences within the tribe. CI and RI values when position 3 was Genus Chaenusa sensu lato. —Chaenusa s.l. excluded. This suggests that position 3 was not monophyletic in any of the contains phylogenetic information that analyses. Rather, the results indicate that supports several clades but also increases certain species of Chaenusa s.l. are more the level of homoplasy in the data matrix. closely related to species of Chorebus and Tribe Dacnusini. —Griffiths (1964) and Coel in ius than they are to other species of Wharton (1994) suggested that Dacnusini Chaenusa s.l. This result is not surprising is monophyletic based on the absence of for several reasons. Certain species of forewing vein r-m. Further, Dacnusini is Chaenusa s.l. possess morphological fea- homogeneous in terms of host utilization; tures that suggest the potential for a close the tribe exclusively contains parasitoids of relationship with species of Chorebus and plant-mining flies, particularly parasitoids Coelinius. As is observed for species of the of leaf- and stem-mining agromyzids, Chorebus affinis group (Griffiths 1968), chloropids, and ephydrids (Wharton several species of Chaenusa s.l. have four- 1997). Maximum parsimony, ML, and toothed mandibles with three major teeth Bayesian analyses were conducted with and one small tooth along the ventral Dapsilarthra sp. 1 included in the ingroup margin of elongate tooth 2. In this study to explore the monophyly of Dacnusini. In Chaenusa n. sp. 3, Chaenusa n. sp. 1, and MP and Bayesian analyses of the DNA Cha. quadriceps (Ashmead) exhibit this data matrix, Dacnusini was monophyletic condition, as do four described (i.e., Chae- with 78% bootstrap support and a posterior nusa anticostae Riegel, Chaenusa californica probability of 0.99, respectively. However, Riegel, Chaenusa illinae Riegel, Chaenusa neither ML analysis of the DNA data rossi Riegel) and two undescribed Nearctic matrix nor MP analysis of the AA data species of Chaenusa s.l. not included in this matrix recovered Dacnusini. Rather, Dapsi- study (Kula unpublished). Further, the larthra sp. 1 was always sister to Chaenusa metapleural setation of Chaenusa n. sp. 3 n. sp. 3, but bootstrap support for this is nearly oriented in a rosette surrounding relationship was <50%. In MP analysis of a raised swelling, a character state used to 262 Journal of Hymenoptera Research

than a facet width define Chorebus. Chaenusa n. sp. 3 forms conspicuously longer a clade with Chorebus in certain MP and and are curved. However, 8.5% of the and 5.3% of the Coelinius mor- Bayesian analyses, and it is possible that Chorebus examined have curved setae on Chaenusa n. sp. 3 is a species of Cliorebus phospecies than a facet width. with setiferous eyes. the eyes longer Thus, A character state in females of Coelinius the mere presence of setae on the eyes is lateral compression of the metasoma. cannot be regarded as a synapomorphy Females of Chorebidea americana Riegel, that defines Chaenusa sd. sensu stricto. — Chorebidea bessae Riegel, Chorebidea mcclurei Genus Chaenusa Chaenusa s.s. in all and Riegel, Cha. rossi, Chorebidea saxicola Riegel, was monophyletic analyses, and one undescribed Nearctic species of branch support was moderate to strong. Chaenusa sd. have a laterally compressed Chaenusa s.s. should be more extensively metasoma (Kula unpublished). In this sampled in future phylogenetic analyses to study only Chorebidea americana clearly provide a more robust assessment of exhibits this condition. Further, Coelinius monophyly. Six of the 11 described New is partially defined on the possession of World species of Chaenusa sd. fit in Chae- four-toothed mandibles with three major nusa s.s. (i.e., Cha. anticostae, Cha. californica, teeth and one small tooth between tooth 1 Cha. illinae, Cha. pallidinervis, Cha. quadri- and 2. In this study Chaenusa sp. 2, ceps, Cha. rossi). However, all except Cha. Chaenusa n. sp. 2, and Chaenusa sp. 1 quadriceps are only known from the holo- exhibit this condition, and it also occurs type. Thus, a very small number of New in an undescribed Nearctic species of World specimens of Chaenusa s.s. are avail- Chaenusa sd. not included in this study able for DNA sequencing. Extensive col- (Kula unpublished). lecting will be needed to increase the Griffiths (1964) proposed that among representation of New World Chaenusa s.s. dacnusines setiferous eyes is unique to in future phylogenetic analyses. The most species of Chaenusa sd. and is a synapo- successful methods for collecting speci- morphy that defines Chaenusa sd. How- mens of Chaenusa sd. are yellow pan traps ever, dacnusines in genera other than placed along the shore of permanent Chaenusa sd. have setiferous eyes. New lakes, ponds, and streams and sweeping World species of Cliorebus (47 morphospe- within and along the edge of aquatic cies), Coelinius (19 morphospecies), Colo- habitats. neura Forster (two morphospecies), Dac- Riegel (1950, 1982) defined Chaenusa s.s. nusa Haliday (18 morphospecies), Epimicta using the following features: (1) 1st sub- Forster (two morphospecies), Exotela For- discal cell closed, (2) stigma "short, wide", ster (14 morphospecies), Laotris Nixon (six and (3) labial palpi four-segmented. Both specimens), and Synelix Forster (one mor- species of Chaenusa s.s. included in this phospecies) all have setiferous eyes. Only study have the 1st subdiscal cell closed, World New species of Symphya Forster (13 a relatively broad stigma, and three- or have morphospecies) glabrous eyes (Kula four-segmented labial palpi. The length of unpublished). Character states other than the distal palpomere in specimens with setiferous eyes clearly place the aforemen- three-segmented labial palpi is approxi- tioned species in their respective genera. In mately the combined length of palpomeres most cases the setae are straight and are so 3 and 4 in specimens with four-segmented minute that they could easily escape de- labial palpi. Further, examination with tection using a stereomicroscope at 120X a scanning electron microscope revealed magnification (i.e., usually snorter than that the distinction between palpomeres 3 a facet width). For species of Chaenusa sd., and 4 is extremely weak in some specimens at least some setae on the eyes are of Chaenusa n. sp. 1, Cha. quadriceps, and an Volume 15, Number 2, 2006 263

undescribed Nearctic species that fits Chae- Genus Chorebidella. —Chorebidella was nusa s.s. Thus, it appears that three-seg- monophyletic in all analyses, and branch in mented labial palpi Chaenusa n. sp. 1 and support was strong. Chorebidella should be Clin, quadriceps resulted from the fusion of more extensively sampled in future phylo- palpomeres 3 and 4 or the division of genetic analyses to provide a more robust palpomere 3 into two—palpomeres. assessment of monophyly. Only one of the Genus Chorebidea. Chorebidea was not 11 described New World species of Chae- in s.l. monophyletic any of the analyses. nusa fits in Chorebidella (i.e., Chaenusa However, four of five species of Chorebidea bergi (Riegel)). We acquired two Old World included in this study formed a clade in all species in addition to Cha. bergi but only analyses, and branch support was weak to had permission to use one for DNA strong. Riegel (1950, 1982) defined Chorebi- sequencing. As for Chaenusa s.s. extensive dea using the following features: (1) 1st collecting will be needed to increase the subdiscal cell open, (2) forewing vein representation of New World Chorebidella in RS+M at least partially present, (3) stigma future phylogenetic analyses. "long", (4) labial palpi three-segmented, Riegel (1950, 1982) defined Chorebidella and (5) gonoforceps "stocking-shaped in using the following features: (1) 1st sub- lateral view". All species of Chorebidea discal cell open, (2) forewing vein RS+M included in this study have an open 1st absent, (3) stigma "short, wide", (4) labial subdiscal cell through the partial or com- palpi three-segmented, and (5) gonofor- plete absence of forewing veins 2-1A and ceps "not stocking-shaped in lateral view". 2cu-a, and forewing vein RS+M is at least Both species of Chorebidella included in this partially present. Both features exhibit study have the 1st subdiscal cell open some degree of intraspecific variation. through the partial or complete absence of The 1st subdiscal cell is rarely (3.1%, one forewing veins 2-1 A and 2cu-a, forewing of 32 specimens examined) closed in vein RS+M absent, a relatively broad and that Chaenusa n. sp. 3, and although forewing stigma, gonoforceps gradually narrow to vein RS+M is present for all species, it may proximally distally and are vary from complete and tubular to minute- roughly triangular-shaped. Chaenusa bergi has labial but ly present posteriorly. Riegel (1950, 1982) three-segmented palpi, Chae- included a "long" stigma in his concept of nusa sp. 3 has two-segmented labial palpi. Chorebidea, but Chorebidea americana and Two-segmented labial palpi have not been recorded for of Chaenusa s.l. Chorebidea bessae have a relatively broad any species The is for stigma. stigma relatively long CONCLUSIONS Chaenusa sp. 2, Chaenusa n. sp. 2, Chaenusa and n. is sp. 1, Chaenusa sp. 3 but relatively The results of this study indicate that broad for Chorebidea americana. Chaenusa sp. Chaenusa s.l. is not monophyletic, but 2, Chaenusa n. sp. 2, Chaenusa sp. 1, and Chaenusa s.s. and Chorebidella are mono- Chorebidea americana have three-segmented phyletic groups with moderate to strong labial palpi, but the labial palpi are four- support. Chorebidea was not monophyletic segmented for Chaenusa n. sp. 3. Lastly, in any of the analyses, but four of five Chorebidea americana has "stocking- species of Chorebidea included in this study shaped" gonoforceps, but Chaenusa sp. 2, formed a clade in all analyses. The species Chaenusa n. sp. 2, and Chaenusa sp. 1 have of Chorebidea that did not form a clade with that gonoforceps gradually narrow proxi- the other species of Chorebidea (i.e., Chae- mally to distally and are roughly triangu- nusa n. sp. 3) exhibits morphological lar-shaped. Chaenusa n. sp. 3 has roughly character states observed for species of rectangular-shaped gonoforceps that are Chorebus. Further, Chaenusa n. sp. 3 forms truncate distally. a clade with Chorebus in certain MP and of Hymenoptera Research 264 Journal

Evenhuis, N. L. and G. A. Samuelson. 2005. The and this that analyses, suggests the World Website. Bayesian Insect and Spider Collections of n. 3 be a species Chaenusa sp. may actually Bishop Museum, Honolulu, Hawaii, http://hbs. on the of Chorebus with long curved setae bishopmuseum.org/codens/codens-r-us.html [accessed 25 February 2005]. eyes. tests for ND1 Faith, D. P. 1991. Cladistic permutation Phylogenetic analyses using gene monophyly and nonmonophyly. Systematic Zool- largely support Riegel's (1950, sequences ogy 40: 366-375. of Chaenusa s.l. as Chaenusa und Gattun- 1982) treatment Forster, A. 1862. Synopsis der Familien Chorebidella. However, der Braconen. des Naturhistor- s.s., Chorebidea, and gen Verhandlungen Rheinlande und West- s.l. be retained ischen Vereines we suggest that Chaenusa preussischen 19: 225-288. with nuclear phalens until phylogenetic analyses Gilbert, D. G. 1996. SeqPup, Version 0.6. Indiana markers, and greater taxon morphology, University, Bloomington, Indiana, http://iubio. to confirm sampling have been undertaken bio.indiana.edu/soft/molbio/seqpup/ [released this the relationships inferred in study. July 1996]. Griffiths, G. C. D. 1964. The Alysiinae (Hym. ACKNOWLEDGMENTS Braconidae) parasites of the Agromyzidae (Dip- tera). I. General questions of taxonomy, biology for We thank the following persons and repositories and evolution. Beitrdge zur Entonwlogie 14: used in this David B. providing specimens study: 823-914. Masner Wahl (AEIC), Henri Goulet (CNC), Lubomir Griffiths, G. C. D. 1968. The Alysiinae (Hym. Michael (CNC), E. Richard Hoebeke (CUIC), J. Braconidae) parasites of the Agromyzidae (Dip- Colin Favret KSUC, Cerodontha Rondani s. 1. Sharkey (HIC), IAVH, (INHS), tera). VI. The parasites of Robert A. Wharton 18: 63-152. Stefan Cover (MCZ), (TAMU), Beitriige zur Entonwlogie Steven L. F. 2001. MRBAYES: Matthew J. Yoder (TAMU), Heydon Huelsenbeck, J. P. and Ronquist. and David R. (UCDC), Mark F. O'Brien (UMMZ), Bayesian inference of phylogeny. Bioinformatics Smith (USNM). Srinivas Kambhampati provided 17:754-755. valuable technical advice and suggestions on the Maddison, D. R. and W. P. Maddison. 2003. MacClade C. Strakosh Character Evolution. manuscript. We are grateful to Suzanne 4. Analysis of Phylogeny and the course for providing technical support throughout Version 4.06. Sinauer Associates, Sunderland, of this research. We thank Matthew L. Buffington for Massachusetts. B. Whitfield. 2004. consultation on the use of MrBayes. Kent Hampton Michel-Salzat, A. and J. Preliminary within the (Scanning Electron Microscope Laboratory, KSU) evolutionary relationships parasitoid electron Cotesia Braconidae: captured the original scanning micrograph wasp genus (Hymenoptera: of four used to illustrate compound eye setation. This article Microgastrinae): combined analysis genes. 29: 371-382. is Contribution No. 05-271 -J from the Kansas Agricul- Systematic Entomology 1980. New record of tural Experiment Station and was supported by KAES Natarajan, K. and K. C. Mathur. and Na- as a Hatch Project No. 583, Insect Systematics, Chaenusa sp. (Braconidae-Hymenoptera) par- and Culture tional Science Foundation grant DEB-0089656 asite of the rice whorl maggot. Science awarded to CJF. 46: 337-338. of the Nixon, G. E. J. 1943. A revision European LITERATURE CITED Dacnusini (Hym., Braconidae, Dacnusinae). The 79: Eiitomologist's Monthly Magazine 20-34, Belshaw, R., M. Fitton, E. Herniou, C. Gimeno, and D. 159-168. L. 1998. reconstruction 1998. Modeltest: J. Quicke. A phylogenetic Posada, D. and K. A. Crandall. based on of the (Hymenoptera) testing the model of DNA substitution. Bioinfor- the D2 variable region of 28 S ribosomal RNA. matics 14: 817-818. 23: 109-123. and Systematic Entomology Quicke, D. L. J. 1994. Phylogenetics biological K. L. 2001. the in Braconidae Dolphin, and D. ]. Quicke. Estimating transitions the (Hymenoptera: global species richness of an incompletely de- Ichneumonoidea). Norwegian journal of Agricul- scribed taxon: an example using parasitoid wasps tural Sciences 16: 155-162. van 1990. (Hymenoptera: Braconidae). Biological Journal of Quicke, D. L. J. and C. Achterberg. the the Unman Society 73: 279-286. Phylogeny of the subfamilies of family Dowton, M., A. D. Austin, and M. F. Antolin. 1998. Braconidae (Hymenoptera: Ichneumonoidea). Zo- 1-95. Evolutionary relationships among the Braconidae ologische Verhandelingen 258: and of (Hymenoptera: Ichneumonoidea) inferred from Riegel, G. T. 1950. A new genus species partial 16 S rDNA gene sequences. Insect Molec- Dacnusini (Hym.: Braconidae). Entomological ular Biology 7: 129-150. News 61: 125-129. Volume 15, Number 2, 2006 265

Riegel, G. T. 1982. The American species of Dac- Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis nusinae, excluding certain Dacnusini (Hyme- Using Parsimony (*and Other Methods). Version noptera: Braconidae). Novitates Arthropropodae 1: 4.0bI0. Sinauer Associates, Sunderland, Massa- 1-185. chusetts. R. A. Wharton. 1997. R. 1977. Sharkey, M. J. and Morphology Wharton, A. New World Aphaereta species and terminology. Pp. 19-38 in: Wharton, R. A., P. (Hymenoptera: Braconidae: Alysiinae), with a dis- M. Marsh, and M. J. Sharkey, eds. Manual of the cussion of terminology used in the tribe Alysiini. New World Genera of the Family Braconidae (Hyme- Annals of the Entomological Society of America 70: noptera). Special Publication of the International 782-803. Society of Hymenopterists, Number 1. Wharton, R. A. 1984. Biology of the Alysiini (Hyme- Shaw, M. R. and T. Huddleston. 1991. Classification noptera: Braconidae), parasitoids of cyclorrha- and biology of braconid wasps (Hymenoptera: phous Diptera. Texas Agricultural Experiment Braconidae). Handbooks for the Identification of Station Technical Monograph 11: 1-39. British Insects 7 (11): 1-126. Wharton, R. A. 1994. New genera, species, and records Shi, M., X. X. Chen, and C. van Achterberg. 2005. of New World Alysiinae (Hymenoptera: Braconi- Phylogenetic relationships among the Braconi- dae). Proceedings of the Entomological Society of dae (Hymenoptera: Ichneumonoidea) inferred Washington 96: 630-664. from partial 16S rDNA, 28S rDNA D2, 18S rDNA Wharton, R. A. 1997. Subfamily Alysiinae. Pp. 84-1 17, in: R. P. gene sequences and morphological char- Wharton, A., M. Marsh, and M. j. acters. Molecular Phylogenetics and Evolution 37: Sharkey, eds. Manual of the New World Genera of 104-116. the Family Braconidae (Hymenoptera). Special Pub- Smith, P. T. and S. Kambhampati. 1999. Status of the lication of the International Society of Hymenop- Cotesia flavipes species complex (Braconidae: terists, Number 1. Microgastrinae) based on mitochondrial 16 S Wharton, R. A. and A. D. Austin. 1991. Revision of rRNA and NADH 1 dehydrogenase gene se- Australian Dacnusini (Hymenoptera: Braconidae: quence. Journal of the Kansas Entomological Society Alysiinae), parasitoids of cyclorrhaphous Diptera. 72: 306-314. Journal of the Australian Entomological Society 30: Smith, P. T., S. Kambhampati, W. Volkl, and M. 193-206. S. D. B. Mackauer. 1999. A phylogeny of aphid parasi- Wharton, R. A., R. Shaw, M. J. Sharkey, Wahl, B. B. P. toids (Hymenoptera: Braconidae: Aphidiinae) J. Woolley, J. Whitfield, M. Marsh, and W. inferred from mitochondrial NADH 1 dehydro- Johnson. 1992. Phylogeny of the subfamilies of genase gene sequence. Molecular Phylogenetics and the family Braconidae (Hymenoptera: Ichneumo- Evolution 11: 236-245. noidea): a reassessment. Cladistics 8: 199-235. I. HYM. RES.

Vol. 15(2), 2006, pp. 266-269

NOTE

Observations on Flower Association and Mating Behaviour of the in Greece Pollen Wasp Species Celonites abbreviates (Villers, 1789) (Hymenoptera: Vespidae, Masarinae)

VOLKER MAUSS

Staatliches Museum fur Naturkunde, Abt. Entomologie, Rosenstein 1, D-70191 Stuttgart, Germany, email: [email protected]

with Celonites abbreviatus (Villers) ranges from Satureja thymbra L. () neigh- Morocco across the entire North Mediter- bouring Phrygana vegetation (Fig. 1) (Ordo ranean Area to Turkey. In the North it Cisto-Micromeretalia Oberdorfer 54; cf. reaches the southern parts of Central Horvat et al 1974). The ground was hard, covered with Europe (Gusenleitner 1997). Most informa- stony and only sparsely tion on the bionomics stem from the vegetation. An open source of water was Central European range of the species present only in one of the localities. (Schremmer 1959, Bluthgen 1961, Bellmann Females of C. abbreviatus were frequently 1984, 1995, Miiller 1996, Amiet and Mauss observed to visit flowers of Satureja thym- 2003) whereas bionomical records from bra. During the flower visits the proboscis corolla in- other parts of its distribution are rare (cf. was protruded into the tube, Lichtenstein 1869, Ferton 1901, Fahringer dicating nectar uptake (Fig. 4). Simulta- 1922). Moreover, until now the mating neously, the facial part of the head was behaviour of the species has been com- rubbed over the nototribic anthers of the pletely unknown. flower, and pollen grains accumulated on On a field trip to the Maleas Peninsula of the frons of the female (Fig. 4), where the Peloponnese (Greece, Laconias) C. modified setae form a pollen-collecting th abbreviatus was observed between the 5 apparatus (Schremmer 1959, Miiller 1996). th and 10 of June 2005 at four localities in the At regular intervals the females interrupted transferred vicinity of Profitis Ilias near Neapoli (I. nectar and pollen uptake and 36 26.687' N 23 06.926' E, 55 m a.s.L; II. pollen from the head to the mouthparts by 36 26.719' N 23 08.482' E, 77 m a.s.L; III. alternating grooming movements of the 36 26.122' N 23 08.303' E, 10 m a.s.L; IV. forelegs and ingested the pollen (Fig. 5). 36' 26.252' N 23 07.012' E, 20 m a.s.L). All On two occasions females of C. abbreviatus sites were situated in old, open fallow collected pollen from Thymus capitatus (L.) areas characterised by large patches of Ho. and Li. (Lamiaceae) at localities where

Figs 1-5. Habitat and behaviour of Celonites abbreviatus in vicinity of Profitis Ilias (Peloponnese, Greece). 1, Fallow area at locality II with large patches of Satureja thymbra (middle and right foreground), at which male and female pollen wasps were observed; Phn/gana vegetation in background. 2, Female resting on sun-exposed stone. 3, Insertion phase of copulation; male clasping to the back of the female. 4, Female visiting flower of S. thymbra; her proboscis is protruded into the corolla tube, and the pollen collecting apparatus on the frons is Volume 15, Number 2, 2006 267

rubbed over the nototribic anthers. 5, Female transferring pollen from the fore-tarsal pollen comb of her left to foreleg her mouthparts, after grooming the pollen collecting hairs on the frons with the foreleg. 268 Journal of Hymenoptera Research flowers of S. thymbra were nearly or followed by the male which finally her. The fell to the completely withered. During visits to T. pounced on pair where a short of capitatus females stood with their middle ground, period grappling occurred of about one after which and hindlegs on the lower lip of the flower, second, on the back of the raised the anterior part of the body and the male was positioned rubbed the facial area of the head over the female with his middle and hindlegs nototribic anthers, while the proboscis wrapped around the female's metasoma into the remained retracted. A single flower visit and his genitalia inserted genital of a male was recorded at S. thymbra. The chamber of the female (Fig. 3). The pair in this observed flower visiting behaviour of C. remained motionless position for abbreviatus on the Maleas Peninsula adds about 10 seconds. Then the partners sepa- further evidence that C. abbreviatus is rated and flew away. The second copula- the first in that the specialised with regard to its pollen source tion differed from to flowers of Lamiaceae since in Central patrolling male pounced on a female vis- a flower of S. the Europe, Italy and Croatia pollen collecting iting thymbra. During of C. abbreviatus is also restricted to various following insertion phase the pair re- flowers of Lamiaceae (e.g. Acinos arvensis mained on the flower. The male released (Lam.) Dandy, Ballota nigra L., Ballota the female's body and fell over backwards in pseudodictamnus Bentham, Salvia officinalis but his genitalia remained the female's L., Stacln/s cretica L., Teucrium montanum L. genital chamber. After less than 10 sec- and Thymus spec; Bellmann 1984, 1995, onds, the pair separated and both partners Schremmer 1959, Miiller 1996). Although flew off. The female alighted on a nearby imagines have been recorded also from stone and cleaned her head with her flowers of other plants families such as forelegs, while the male flew off and Boraginaceae, Crassulaceae and Gerania- disappeared. This is the first record of ceae (Schremmer 1959, Schmiedeknecht mating behaviour in C. abbreviatus and also 1930, Bliithgen 1961), these visits were the first description of the copulation of probably for nectar uptake only (Schrem- a species of the genus Celonites, in general, mer 1959). However, it has been recorded that males Between flower visits the females often of Afrotropical species of Celonites also alighted briefly on the ground or on small search for females in the vicinity of forage stones close to the forage plants (Fig. 2). On plants (Gess 1996: 59) indicating that re- a single occasion a female defecated after source based mating systems may be more alighting on a stone. Similarly, females of widespread in this monophylum. The un- C. abbreviatus from Central Europe were usual position of the male in the second frequently observed to alight on sun- copulation looked similar to the hanging exposed stones or on the ground in the position in the first phase of the copulation of vicinity forage plants (Bluthgen 1961). A of vespine wasps (cf. Schulz-Langner comparable behaviour is shown by females 1954). However, it remains to be shown of the Afrotropical Celonites clypeatus whether this is a regular alternative mating Brauns (Gess 1993). position in C. abbreviatus or if the male Males repeatedly patrolled flowers of S. accidentally lost his hold on the female's thymbra flying at the level of the inflor- metasoma. escences. Copulations were observed in Imagines of C. abbreviatus were not two instances. The first was initiated by observed at water, which is in agreement a patrolling male that rapidly approached with the behaviour of C. abbreviatus in a female towards a of S. flying plant Central Europe (cf. Bellmann 1984, 1995) The female thymbra. responded, in that and of Afrotropical species of Celonites she flew about 0.2 m back from the plant, (Gess 1996: 107). Volume 15, Number 2, 2006 269

ACKNOWLEDGMENTS Gess, S. K. 1993. The Karoo Violet and a masarid wasp - a mutually beneficial association. The Naturalist John Plant (Vienna) kindly read the manuscript and 37: 32-36. S. K. — improved my English. Gess, 1996. The pollen wasps Ecologi/ and natural the Masarinae. Harvard LITERATURE CITED history of University Press, Cambridge, Massachusetts. 1997. Gusenleitner, J. Die europaischen Arten der Amiet, F. and V. Mauss. 2003. First report of male Gattung Celonites Latreille 1802 (Hymeno- sleeping aggregations in the pollen wasp Celonites ptera, Masaridae). Linzer Biologische Beitrdge 29: abbreviatus (Villers, 1789) (Hymenoptera: Vespi- 109-115. dae: Masarinae). Journal of Hymenoptera Research Horvat, I., V. Glavac, and H. 1974. 12: 355-357. Ellenberg. Vegetation Siidosteuropas. Geobotanica selecta 4. Bellmann, H. 1984. Beobachtungen zum Brutverhalten Gustav Fischer, Stuttgart. 768 pp. von Celonites abbreviatus Villers (Hymenoptera, Lichtenstein, J. 1869. No title. Annalcs de la Societe Masaridae). Zoologischer Anzeiger 212: 321-328. Entomologique de France (4) 9: XXIX. Bellmann, H. 1995. Bienen, Wespen, Ameisen: die Miiller, A. 1996. evolution of Hautfliigler Mitteleuropas. Kosmos Naturfuhrer. Convergent morpholog- ical in Central bee and Franckh-Kosmos, Stuttgart. specializations European as an to the Bliithgen, P. 1961. Die Faltenwespen Mitteleuropas honey wasp species adaptation of from nototribic flowers (Hymenoptera, Diploptera). Abhandlungen der uptake pollen (Hyme- and Deutschen Akademie der Wissenschaften zu Berlin noptera, Apoidea Masaridae). Biological Klasse Journal the Linnean 57: 235-252. fur Chemie, Geologie und Biologie 1961 (2): of Society 1-249. Schmiedeknecht, O. 1930. Handbuch der Hymenopteren 1922. 2. Aufl. Gustav Fische, Fahringer, J. Hymenopterologische Ergebnisse Mitteleuropas. Jena. einer wissenschaftlichen Studienreise nach der Schremmer, F. 1959. Der bisher unbekannte Pollen- Tiirkei und Kleinasien (mit Ausschluss des sammelapparat der Honigwespe Celonites abbre- viatus Vill. Amanusgebirges). Archiv fur Naturgeschichte 88: (Vespidae, Masarinae). Zeitschrift

149-222, 9 figs. fiir Morphologic und Okologie der Tiere 48: 424- Ferton, C. 1901. Notes detachees sur l'instinct des 438. Hymenopteres melliferes et ravisseurs avec la Schulz-Langner, E. 1954. Beobachtungen zur Kopula- de description quelques especes. Annalcs de la tion gefangen gehaltener Wespen. Zoologischer Societe Entomologique de France 70: 83-148. Anzeiger 152: 39^42. J. HYM. RES. Vol. 15(2), 2006, pp. 270-276

The Genus AUotilla Schuster (Hymenoptera: ): Phylogenetic Analysis of its Relationships, First Description of the Female and New Distribution Records

DlOMEDES QUINTERO A. AND ROBERTO A. CAMBRA T.

Museo de Invertebrados G.B. Fairchild, Estafeta Universitaria, Universidad de Panama, Panama, Panama; email: [email protected] [email protected]

Abstract. — The morphological characteristics of the monotypic genus AUotilla Schuster are discussed; new distribution records and a morphology-based phylogenetic analysis are presented, of the clarifying possible relationships with other sphaeropthalmine genera. Females genus are described and illustrated for the first time; male genitalia are illustrated.

Until now, AUotilla gibbosa Schuster, 1949 Natural History, New York (AMNH); was known only from the holotype male Museum of Comparative Zoology, Har- In a list of vard Mu- from Cordoba, Argentina. University , Cambridge (MCZ); Mutillidae from Argentina, Fritz (1998) seo de Invertebrados G. B. Fairchild, Uni-

included Buenos Aires as an additional versidad de Panama (MIUP); D. J. Broth- of collection locality for this species, without ers's personal collection, University providing any additional information such KwaZulu-Natal, South Africa (DJBC); and as the number of specimens examined, Institut Royal des Sciencies Naturelles de their sex, or the depository. We must Belgique (ISNB). consider the Buenos Aires record not to RESULTS be valid because it is incomplete and because we have been unable to locate AUotilla Schuster any of Fritz's AUotilla specimens in his AUotilla Schuster 1949: 89-93. reference collections at the AMNH, New Type species: AUotilla gibbosa Schuster, 1949, by original York (to which they were sold by his designation and monotypy. widow after his death) or at the Museo de Ciencias Naturales "B. Rivadavia," — Generic diameters of female. Body cov- Buenos Aires (Roig-Alsina, pers. comm.). ered with simple setae only. Head: almost The female of the mono- Neotropical, round, narrower than mesosoma; eye typic genus AUotilla is described and small, nearly circular and flat (Fig. 3); illustrated for the first time; new distribu- genal carina absent; scrobal carina absent; tion records and a morphology-based antenna 12-segmented; antennal tubercle phylogenetic analysis are presented. strongly projecting, with low lamellate on anteromedial surface MATERIALS AND METHODS ridge (Fig. 3); mandible slender, with preapical tooth We follow Brothers for the sub- (1999) nearly obsolete or totally absent (Fig. 3); family classification of the Mutillidae. For mandible evenly arcuate on ventral mar- SEM examination we used a JEOL model gin, without tooth or constriction; probos- 5600. The AUotilla JSM gibbosa specimens cidal fossa with carina nearly reaching examined are deposited in the follow- inner mandibular base; hypostoma without institutions: American ing Museum of tooth or tubercle; maxillary palp 6-seg- Volume 15, Number 2, 2006 271

merited; labial palp 4-segmented, the sec- sized punctures, except for mostly im- ond segment almost as long as wide. punctate anterior area near lateral area of Mesosoma: broader than long, pyriform pronotum; metapleura and lateral area of (Fig. 4); dorsum strongly convex, without propodeum impunctate, smooth; setation transverse carina along posterior margin of of pronotum, mesonotum and metanotum pronotum or lateral carina; dorso-lateral similar to that of vertex; mesopleura and margins without spines; scutellar scale dorsum of propodeum with pale setae; absent; mesopleuron swollen; leg with metapleura and lateral area of propodeum apex of tarsus simple, not produced above glabrous. Metasoma: terga I and II with claws. Metasoma: first metasomal tergum small, dense punctures, sparser in apical not constricted posteriorly, sessile with areas (Fig. 5); terga III and IV with small, tergum 2 (Fig. 5); tergum II evenly convex, sparse punctures; tergum V mostly without rows of longitudinal carinae; ter- smooth, except basal lateral area with gum II with felt lines; sternum II without a few punctures; tergum VI with scale-like felt lines; tergum VI with surface totally surface sculpture basally, scales diminish- sculptured and evenly merging with rest of ing in size toward apex, gradually turning tergum, pygidial area poorly defined by into granules (Fig. 6); sterna I and VI postero-lateral carina, only visible under smooth; sterna II and III with small, high magnification (>30x) (Fig. 6). somewhat sparse punctures; sterna IV and V mostly smooth, except apex with Allotilla Schuster, 1949 gibbosa small, dense punctures; tergum I mostly 3-8) (Figs with pale setae, a few dark setae at apex; II with dark lateral Allotilla gibbosa Schuster, 1949: 93-95, Holotype tergum setae, except male, Argentina: Cordoba, col. W. M. Davis, area and apex with pale setae; terga III and Harvard University, MCZ, type 30516, exam- IV mostly with pale setae; tergum V mostly ined. glabrous, with a few pale and dark setae laterally; tergum VI glabrous; sterna I to V Description of female. — col- Integumental with pale setae; sternum VI glabrous. or: all metasomal head, mesosoma, sterna, Additional male characters. —The external terga I, III and IV, reddish brown; antennae male genitalia and the penis valve (Figs 7- and mandible reddish legs yellow-red; 8) are illustrated here for the first time brown except apical third blackish; tergum (paramere, cuspis and digitus were de- II reddish brown except posteriorly with scribed previously but not illustrated). The two lateral, transverse, black inter- spots volsella has distinctive long setae on the rupted medially; terga V and VI black. cuspis apex (Fig. 7). The penis valve has Head: vertex and gena with sparse, medi- a subapical tooth, more distant from the um-sized punctures one or more diameters apical tooth than in males of Protophotopsis apart (Fig. 3); punctures of frons denser, (see Figs 11-14, Cambra and Quintero 1997). less than one diameter apart; vertex and Material examined 5 —All frons with long, sparse, erect and semi- (56 males, females). males were with Malaise (B. erect, dark setae; gena, clypeus laterally captured traps Garcete coll.) and females with pitfall traps (T. and hypostoma with long, sparse, pale Delsinne coll.). PARAGUAY: Boqueron setae. Mesosoma: pronotum and mesono- Depart- ment: Parque Nacional Teniente Enciso, Admin- tum with punctures as on vertex (Fig. 4); istracion, 239 m (21 12' S, 61 39' W) 16-19 Sep metanotum with transverse reticulate band 2003, 20 males [MIUP, DJBC]; same loc, 20-24 dorsum of (Fig. 4); propodeum mostly Mar 2004, 6 males [MIUP]; same loc, 23-26 Sep ex- densely micropunctate-rugose (Fig. 4) 2004, 3 males [MIUP]; Siracua, 275 m (21 02' S, cept a narrow impunctate area near meta- 61 45' W) 20-22 Sep 2003, 21 males [MIUP, notum; mesopleuron with dense, medium- MCZ, AMNH, ISNB, DJBC]; Estancia Maria 272 Journal of Hymenoptera Research

Vicenta, 235-244 m (20 55' S, 61 23' W) 26-30 sessile, evenly merging with second; and Sep 2004, 5 males [MIUP]; TransChaco, Mister seven Sphaeropthalmina (Nanotopsis, Pro- 62 02' 17 2003, 1 female Li- Long, (20°35' S, W), Sep tophotopsis, Reedomutilla, Scaptodactyla, Nacional Teniente Enciso, and females [MIUP]; Parque maytilla, Suareztilla Limaytilla), TransChaco, 23-25 2004, 3 females [MIUP, Sep with head nearly round, narrower than the ISBN, DTBC]; same data but 24-25 Sep 2003, 1 mesosoma, and genal carina absent. female [MCZ]. Presidente Hayes Department: Characters. —Twenty-three binary and Reserva Tinfunke, La Verde, 146 m (23 56' S, multistate characters of adult male (M) 69°29' W) 29 Nov-1 Dec 2003, 1 male [MIUP]. and female (F) external morphology and — frons dark reddish all Variations. Female male genitalia were coded for analysis; brown to black; tergum IV varying from were treated as unweighted and unor- black or the lateral totally reddish brown to dered. No autapomorphies were used. areas black with reddish brown in the The character matrix used is given in middle. Males from Paraguay are identical Table 1. The following characters were the to the holotype, except that propodeal employed for cladistic analysis: lateral area is rugose on the holotype, but Head: punctate with smooth areas to rugose- or in — punctate totally rugose specimens Head shape (F): small, almost—round, from Paraguay. We consider this variation not broader than mesosoma; 1 trans- to be size-related: male rugosity increases versely subquadrate, large, distinctly broader than mesosoma. with body length. In addition, larger males 2. Head —without conical have the notauli nearly obsolete (same as (M, F): large 1 —with conical the holotype), but notauli are absent in projection ventrally; large ventrally. smaller males. Total length, females: 3.5- projection 3. Scrobal carina — 1 —absent. 5 mm; males: 4-7 mm. (F): present; 4. Genal carina —absent; 1 — Distribution. —Paraguay and Argentina. (F): present. - 5. Mandible basal ventral margin (M, F): AUotilla gibbosa was previously known only with constriction; 1 —with broad lamellate from the holotype from Cordoba, Argen- — projection; 2 almost straight. tina. - Antennal tubercle (F): without lamellate

1 — on PHYLOGENETIC ANALYSIS projection; with lamellate projection anteromedial surface. T/7A77. —To test the subtribal of position 7. Antenna (F): 0-12-segmented; 1-13-seg- AUotilla (currently included in the subtribe mented. Brothers and to — Pseudomethocina, 1975), 8. Ocelli (M): —small (diurnally active); 1 recognize its phylogenetic affinities, we large (nocturnally active). selected as the outgroup the following two genera: Timulla (Mutillini) and Dasy- Mesosoma: labris (Dasylabrini, genus not present in — Dorsum of mesosoma (F): longer than America); as ingroup taxa, we selected the broad, sometimes as broad as or slightly following 17 Sphaeropthalmini genera, — broader than long; 1 distinctly broader with mainly South American distributions than long. and fully winged males: ten Pseudometho- — Shape of mesosoma (F): subrectangular; cina Tallium, Atillmu, Calomu- (Euspinolia, 1 —violin- shaped, strongly constricted at tilla, — Horcomutilla, Pseudomethoca, Hoplo- the propodeal spiracles; 2 pyriform. and AUo- crates, Pappognatha, Hoplomutilla 3. Axilla of mesonotum (M): —not expand- females with head tilla), transversely sub- ed; 1 —expanded posterolateral^ as a rect- quadrate, broader than the mesosoma, angular or acute protruberance. carina — 1 — genal present (except Euspinolia 4. Scutellar scale —(F): present;— absent. and Tallium), first metasomal segment 5. Notauli (M): present; 1 absent. Volume 15, Number 2, 2006 273

Table 1. Data matrix for the 23 characters used in the phylogenetic analysis

Characters 274 Journal of Hymenoptera Research sent autapomorphies in the present analy- Atlllum sis: tubercles on vertex, posterolateral Hoploc rates in females of present some Dasymutilla, Pseudomethoca HI Horcomulilla Traumatomutilla and Cephalomutilla; mand- Calomutilla ibles covered with dense, short pubes- Hoplomutilla of cence, an autapomorphy Pappognatha; Pappognatha humeral angle of pronotum produced as Eusplnolla H^ — Talhum a hook-like tooth, an of autapomorphy Nanotopsls

Gurisita females (males unknown); posteri- Protophotopsis or coxa with tooth apically on inner Allotilla an of Vianatilla Scaptodactyla margin, autapomorphy Us-[T Llmaylllla

an - females; tibia 2 with one apical spur, Xystromutilla - Reedomutilla autapomorphy of Acanthophotopsis males; - Dasylabrls males or apterous brachypterous, Morsyma - Suareztilla (apterous), Myrmilloides and Stethophotopsis — Atlllum Dasymutilla (rarely bra- (brachypterous), Psoudomcthoca I II arcuate trans- with ' chypterous); tergum Horcomutllla

i— verse band of dense, curled setae and Calomutilla anterior slight integumental ridge at margin of band, in most Dimorphomu- tilla females; female with felt line on sternum II, autapomorphy of Patquiatilla; sternum II with anteromedian seta-filled and pit, in some males of Dasymutilla Traumatomutilla; tegula elongated to or beyond the level of transscutal arti- culation, autapomorphy for Timulla males; eye inner margin deeply and abruptly notched, autapomorphy for 1-2. Preferred of Timulla males. Figs minimum-length cladograms four trees derived from heuristic analysis of the A heuristic search of trees derived from character data presented in Table 1 using ratchet = parsimony analysis was carried out using (WINCLADA) and mult* (NONA). Tree length 62; = = NONA version 2.0 using WinClada version consistency index 0.43; retention index 0.66. are show by the black circles. 1.00.08 (Nixon 2002), resulting in four Synapomorphies cladograms. We preferred two of these minimal-length cladograms (Figs 1-2), see the mesosoma, and because these were the Results and Discussion. The following only female and male Sphaeropthalminae were used: trees to options maximum keep collected from the same geographic area, = number of = 1000; replications (mult*N) Chaco, that we recognized as = Paraguay's 1000; starting trees per rep (hold/) 100; belonging to a genus whose female was not random seed = 1000; unconstrained known. search; search of TBR + strategy multiple The relationships between American TBR (mult* max*). mutillid genera (none is known to be outside of America, for RESULTS AND DISCUSSION present except Timulla) and those of Africa and Australia The female of Allolilla gibbosa was recog- are almost totally unknown, although nized based on morphological similarities different but closely related genera are to the known male, mainly the distinctive involved. Vicariant biogeographical hy- and peculiar inflation and broadening of potheses and their common ancestral geo- Volume 15, Number 2, 2006 275

Figs 7-8. Allotilla gibbosa, male genitalia. 7. Para- meres dorsal view. 8. Penis valve, lateral view. = = Abbreviations: P paramere; ST subapical tooth; VS = volsellar setae.

Figs 3-4. Allot ilia gibbosa, female. 3. Head, dorsal areas close view. 4. Mesosoma, postero-dorsal view. Abrevia- graphic suggest evolutionary = tions: LP lamellate projection of antennal tubercle, relationships with South American genera. = = = M metanorum, P pronotum, Pr propodeum, S Some genera widely distributed in the = scape. Americas, such as Sphaeropthalma and Traumatomutilla, cannot be included in the present cladistic analysis because they are .,-: •• as yet poorly defined. Wm——> — The heuristic analysis resulted in four cladograms, the two preferred minimal-

* length cladograms (Figs 1-2) postulate a sister relationship of Allotilla with Scapto- dactyla and Li nun/ 1 ilia. The other two cladograms postulate a less parsimonious, complex relationship for the selected South Amerian taxa and, following Occam's razor, were not accepted: Allotilla + [(Scap- todach/la + Limaytilla) + (((Xystromutilla + ((Reedomutilla + (Dasylabris + Suareztilla)))]. These preliminary cladograms suggest than Allotilla is more closely related to genera in the subtribe Sphaeropthalmina than to those in the Pseudomethocina. At present, we do not know of any unique morphological characters to separate these two subtribes.

The biology of Allotilla gibbosa is un- known, but the black integument, moder- sized and small ocelli of Figs 5-6. Allotilla gibbosa, female metasoma, dorsal ately eyes, very view. 5. Tergum 1 and basal part of tergum 2. 6. the males suggest that they are diurnal. Tergum 6. The following morphological characters of 276 Journal of Hymenoptera Research

AUotilla females indicate that they spend Spichiger et al. (2004), investigating the most of their lives underground: small, geographical zonation in of flattened eyes, relatively short legs with 32 common tree species encountered in a fore tarsal rake (used to excavate soil), Paraguay, found that the xeromorphic and a mostly reddish-brown integument. forests of the Chaco area act as an edaphic Extensive visual samplings carried out barrier to many species that are centered in during daylight hours in Teniente Enciso northern Argentina. The genus AUotilla is National Park did not yield any AUotilla reported only from northern Argentina females from the ground or on the sparse and Paraguay. vegetation; females were collected only If additional sphaeropthalmine genera data lend with pitfall traps. These capture eventually are included in the data matrix support to the postulated underground, presented here, the larger data matrix burrowing habits of the females. Probably might provide a better resolution in the they parasitize small, underground-nesting phylogenetic analysis. aculeates. Females of the here-recognized ACKNOWLEDGMENTS closely related sphaeropthalmine genera, and sister Scaptodactyla Limaytilla, genera Our sincere thanks to Thibaut Delsinne, ISBN, for to have a similar AUotilla, morphological sorting and sending us all the mutillids from his habitus, suggesting that they have similar pitfall traps from Paraguay. We are grateful to hypogeal lives and burrowing activities. Philip D. Perkins, MCZ, for the loan of the holo- type of A. gibbosa; Bolivar R. Garcete B., Museo A morphology-based phylogenetic anal- Nacional Historia Natural de Paraguay for col- ysis of 18 mostly South American mutillid lecting mutillid specimens; Annette Aiello, and one from outside America genera Smithsonian Tropical Research Institute, and Denis us to construct a of (Dasylabris) permits hypo- J. Brothers, University KwaZulu-Natal, South thetical scenario of biogeographic diver- Africa for reading the manuscript and for provid- us with to its gences. The vicariant event that divided ing very helpful suggestions improve contents; Alcides Muhoz, of Panama, for the of the common ancestor of University population SEM work. the taxa presented in Table 1 was the uplifting of the Andean high mountain LITERATURE CITED range. This uplifting event was followed D. 1999. Brothers, J. Phytogeny and evolution of by ecological divergence of the two An- wasps, and bees (Hymenoptera, Chrysidoi- dean regions: the elongated West costal dea, and Apoidea). Zoologica Scripta region, draining into the Pacific Ocean, 28: 233-249. isolated from a more extensive and ecolog- Cambra. T. R. A. and D. Quintero, A. 1997. A revi- sion of Schuster ically diverse Eastern region. The ancestor Protopmotopsis (Hymenoptera: Mutillidae). journal of Hymenoptera Research 6: population of the present-day Euspinolia, 263-272. an almost exclusively Chilean and Peru- Fritz, M. A. 1998. Mutillidae. Pp. 445-451 in: Morrone, vian taxon, was isolated on the Pacific J. J. and S. Coscaron, eds. Biodiversidad de Andes slopes. The common ancestor of Artropodos Argentinos, Ediciones Sur. Nixon, K. C. 2002. Winclada version 1.00.08. Published both Tallium and Limaytilla + Scaptodactyla + by the author, Ithaca, NY. AUotilla was relegated to the region east R. M. 1949. Contributions toward a Mono- of Schuster, the Andes. The climatic and vegetational graph of the Mutillidae of the Neotropical about the Andean changes brought by Region. III. A Key to Subfamilies Represented uplift, particularly the desertification of and the descriptions of several New Genera the 29: Chaco region, was the driving selective (Hymenoptera). Entomologica Americana (n.s.) force that caused the diversification and 59-140. Spichiger, R., C Calenge, and B. Bise. 2004. Geo- evolutionary split between Tallium and the graphical zonation in the Neotropics of tree sister genera of AUotilla + + Scaptodactyla species characteristic of the Paraguay-Parana Limaytilla. Basin. Journal of Biogeograpln/ 31: 1489-1501. J. HYM. RES. Vol. 15(2), 2006, pp. 277-285

Parasitism and Sex Ratio of the Bedeguar Diplolepis rosae (L.) (Hymenoptera: Cynipidae) in Sicily (Italy)

Maria Concetta Rizzo and Bruno Massa

Dipartimento SENFIMIZO, Universita di Palermo, viale delle Scienze 13, 1-90128 Palermo Italy; email: [email protected]

— Abstract. The Diplolepis rosae gall community is analysed in Sicily (Italy), based on collections totalling 82 galls from 12 sites from which 1,026 adult insects were obtained. The gall wasp exits in March-June from galls induced the previous year. On average 5.6 D. rosae individuals per gall were obtained, corresponding to 44.8% of all the emerged insects. We obtained 4.3% of D. rosae males overall, the highest figure found till now for the cynipid overall in Europe, where male D. rosae are visually rarer, and the first record of them for the Mediterranean area. No inquilines were found, and, consequently, none of their specific parasitoids. However, Eurytoma rosae, generally considered as a specific parasitoid of the inquiline Periclistus brandtii, was obtained, together with the polyphagous Exeristes roborator. In all, seven parasitoid species emerged from the galls: four of them, Orthopelma mediator, Torymus bedeguaris, Exeristes roborator, and Eupelmus urozonus, start to emerge together with D. rosae, while Glyphomerus stigma, Pteromalus bedeguaris, and E. rosae, have their maximum peaks later in the year. All the cited species, except for E. roborator, showed a second peak of emergence in September, when D. rosae is absent. Parasitization ranged from 12.5 to 100%, reaching more than 70% in 66.6% of the samples, but it was rather low (30.5%) when males were present, even though there was no overall correlation between parasitization and D. rosae sex ratio. Statistical analysis showed however that all the parasitoid species (except for males of T. bedeguaris and P. bedeguaris) are longer (which we take to signify larger) than D. rosae males, and neither the size nor the sex ratio of parasitoids differed statistically depending on the presence of male D. rosae.

Diplolepis rosae (Linnaeus, 1758) is a pa- ductively inactive (Stille and Davring 1980) laearctic species, introduced in the nearctic and of spasmodic occurrence probably due region (Shorthouse and Ritchie 1984). It is to female infection with Wolbacliia bacteria, widespread in peninsular Italy (Pagliano which is also common in other Diplolepis 1995), from where it has been known since species (Plantard et al. 1998, 1999). Larvae at least 1600 (Tuscany: Pagliano et al. 1997), grow inside the galls and overwinter in and in Sicily (De Stefani Perez 1887, as diapause as prepupae; they pupate the Rhodites rosae L.). Its multilocular galls have following spring and emerge soon after been recorded on 18 species of Rosa in Italy (Schroder 1967). Diplolepis rosae galls host (De Stefani Perez 1887, 1902, 1905, Pelliz- a large insect community that has been zari Scaltriti 1988, Pagliano et al. 1997), and extensively studied in many countries on 32 species in Europe and the Mediter- (Randolph 2005, and references therein). ranean basin (Houard 1908, Nieves-Aldrey Considering only Hymenoptera parasi- of 2001), though the gall wasp seems to prefer toids, Noyes (2003) lists 29 species those species that are taxonomically allied Chalcidoidea, and numbers increase when in the Section Caninae (Randolph 2005), parasitoids belonging to other superfami- and particularly R. canina L. (Schroder lies are included (Fulmek 1968). However, 1967). Although D. rosae reproduces itself several of the names listed by Fulmek by amphitokous parthenogenesis (Beauvi- (1968) are invalid or misidentifications or sage 1883, Callan 1940), males are repro- probably erroneously associated with D. 278 Journal of Hymenoptera Research

Table 1. Differences in the number of males and in the percentage of parasitization in different groups of gall samples. 419 280 Journal of Hymenoptera Research

emerged males

emerged females N o 400

v vi vn vm ix x xi xn Months

D. rosae in Fig. 1. Seasonal appearance of adults of Sicily. inquiline Periclistus brandtii (Ratzeburg) parasitoid of D. rosae (Askew 1960, Nieves- nor its parasitoids were found. Altough Aldrey 1981, Redfern and Askew 1992, sometimes a common inhabitant of D. rosae Randolph 2005, and references therein), galls, this species has a variable distribu- previously unrecorded from Sicily (Scara- tion in central and northern Europe and mozzino 1995, Noyes 2003). It was known seems rarer in Mediterranean countries as a parasitoid of D. rosae in Italy as early as (Nieves-Aldrey 1981, Pujade Villar 1983, 1600, since an unidentified , In it is Randolph 2005). Italy known only corresponding to it, is portrayed by Redi in from the north (Pagliano 1995). No other an unpublished plate on D. rosae galls inquiline species has been reared during collected in Tuscany (Pagliano et al. 1997). this study. — In this study it was the most common Parasitoid complex. Although a large (88.9% of the samples) and abundant number of parasitic wasps is associated parasitoid (66.1% of all parasitoids) with the bedeguar gall wasp, a typical list (Fig. 2). O. mediator emerged from March of parasitoids comprises about a dozen onwards (Fig. 3), so that its phenology species (Schroder 1967, Askew 1984, Ran- matches that of the gall inducer, as dolph 2005). During this study a total of Nieves-Aldrey (1981) already observed in 566 Hymenoptera, mainly belonging to the Spain, except in autumn when there was most common D. rosae parasitoid species, a late peak of emergences. was obtained from the Sicilian galls as Torymus bedeguaris is a holarctic species follows: 374 Orthopelma mediator (Thun- (Grissell 1995,' Noyes 2003), already re- O. luteolator berg) (= Gravenhorst) (Ich- corded in Italy (De Stefani Perez 1905, neumonidae), 98 Torymus bedeguaris (Lin- Pagliano 1995, Pagliano and Navone 1995, naeus) (), 52 Glyphomerus stigma Noyes 2003), mainly known as an ectopar- 23 Pteromalus be- (Fabricius) (Torymidae), asitoid of cynipid gall wasps belonging to deguaris (Thomson) (= Habrocytus bede- the genus Diplolepis or of their inquilines guaris Thomson) (Pteromalidae), 12 Exer- (De Stefani Perez 1905, Askew 1960, istes roborator (Fabricius) (Ichneumonidae), Nieves-Aldrey 1981, Noyes 2003); it is also 5 rosae Nees Eurytoma (Eurytomidae), and 2 occasionally reported attacking O. mediator urozonus Dalman Eupelmus (Eupelmidae). (Askew 1960, Schroder 1967). In Sicily T. 55.2% of all They represented insects and bedeguaris represented 17.3% of all emerged their relative abundance is reported in parasitoids and its phenology overlaps that 2. of D. rosae Fig. and O. mediator (cf. Figs 2 and 3). Orthopelma mediator is an endophagous This trend is similar to that recorded by known as a species, largely widespread Nieves-Aldrey (1981) in Spain. Volume 15, Number 2, 2006 281

Torymus bedeguaris Orthopelma 7.3% mediator 66.1%

Diplolepis rosae 44.8% '

Pteromalus bedeguaris 4.1%

Exeristes roborator 2.1% Eupelmus Eurytoma urozonus 0.3% rosae 0.9%

of of D. rosae in Fig. 2. Relative abundance parasitoids Sicily.

in it 9.2% of all Glyphomerus stigma is another holarctic Spain; Sicily reached only It later in the species typically associated with the genus parasitoids (Fig. 2). appeared as than the inducers and the Diplolepis (Noyes 2003), mainly known year previous an ectoparasitoid of D. rosae and its parasitoid species (Fig. 3), which agrees inquilines (Blair 1943, Askew 1960, with the findings of Nieves-Aldrey (1981) Nieves-Aldrey 1981), and occasionally as in Spain. a hyperparasitoid of E. rosae (Redfern and Pteromalus bedeguaris is one of the two Askew 1992). This is the first record for species of Pteromalidae typically associat- with the of D. rosae. It is a holarctic Sicily, although the species was previously ed galls in known from peninsular Italy (Pagliano species, hitherto unknown Italy (Pa- 1995, Pagliano and Navone 1995, Noyes gliano 1995, Pagliano and Navone 1995, 2003). Nieves-Aldrey (1981) reported it as Noyes 2003), although De Stefani Perez the second most abundant parasitoid in (1905) listed many other Pteromalidae as

N° 250

200

150

100 Orthopelma mediator Torymus bedeguaris "EZZEZZZZQ Glypitomerus stigma J~ | Pteromalus bedeguaris Exeristes roborator

Eurytoma rosae

Eupelmus urozonus v1 vn vra K » xn \ln ii l lis

of D. rosae in Fig. 3. Seasonal appearance of adults of parasitoids Sicily. 2g2 Journal of Hymenoptera Research

in Pteromalus C. were found this parasitoids of D. rosae Sicily. inflexa during study, a of and the five individuals of E. rosae, which bedeguaris is considered parasitoid but it is also in 2 and 3), several species of Diplolepis, emerged September (Figs may of D. known to attack O. mediator, T. bedeguaris have developed at the expense rosae, and in with what and G. stigma (Blair 1943, Redfern agreement Nieves-Aldrey in The Askew 1992, Noyes 2003), and occasionally (1981) recorded Spain. typical peak brandtii of in late summer has been parasitizes Periclistus (Nordlander emergences authors 1973, Noyes 2003). The phenology of P. considered by many (Blair 1945, two Niblett and Askew as bedeguaris in Sicily shows emergence 1951, Claridge 1960) and the a from of the periods, the first in late spring precocious emergence galls second in autumn (Fig. 3), similar to what same year. Nordlander (1973) in Sweden and Nieves- Finally, Eupelmus urozonus, a cosmopoli- and Aldrey (1981) in Spain recorded. In the tan generalist (Pagliano 1995, Pagliano occasional Sicilian samples, P. bedeguaris amounted to Navone 1995, Noyes 2003) and of D. rosae 4.1% of all emerged parasitoids (Fig. 2), parasitoid (Schroder 1967, while 22 individuals of this species (100% Noyes 2003) has been obtained during this individuals of this bivoltine of all parasitoids) emerged from the Hun- study: two is in March garian sample in June. This species species (Askew 1961) emerged recorded as the second most common and September. of central and with parasitoid in many countries Parasitism parasitoid— relationships Europe (Randolph 2005, and references the gall inducer. Parasitization ranged therein). The palaearctic Caenacis inflexa from 12.5 to 100%, reaching more than (Ratzeburg) (Pteromalidae) was not ob- 70% in 66.6% of the samples, in line with tained from any sample, being a specific other authors who found the bedeguar gall parasitoid of the inquiline P. brandtii wasp heavily parasitized (Schroder 1967, (Callan 1944), which itself was absent from Nordlander 1973, Stille 1984). In Sicily, the our samples. D. rosae parasitoid complex comprises two Exeristes roborator (Fabricius) is a pa- groups of species; namely, a first group, laearctic species, already known in Italy including O. mediator, T. bedeguaris, and E. (Scaramozzino 1995). It is a polyphagous roborator, whose phenology overlaps and and ectophagous parasitoid of Lepidoptera, strictly follows that of the gall inducer, Coleoptera and Hymenoptera larvae, a second group, with the prevailing species known only as a parasitoid of Biorhiza G. stigma and P. bedeguaris, which appears pallida (Olivier) among Cynipidae (Fulmek later in the year (Fig. 3). All the recorded 1968, Constantineanu and Pisica 1970). We parasitoid species, except for E. roborator, cannot definitely affirm here that it is showed a late peak of emergences in a primary parasitoid of D. rosae, even if autumn, when D. rosae is absent (Fig. 3). no inquiline species emerged from our This peak has often been reported by other samples. Twelve individuals of this species authors (see accounts above of single (2.1% of all parasitoids) exited from the parasitoid species) and explained, depend- in galls March (Figs 2 and 3). ing on species, as a precocious emergence Eurytoma rosae is another palaearctic (i.e. for E. rosae) or as a bivoltine cycle (i.e. species, already known in Italy (Pagliano E. urozonus). Our present data do not allow and Navone 1995, Noyes 2003), associated us to decide which of the two explanations with galls induced by Diplolepis and con- applies to O. mediator, T. bedeguaris, G. sidered as usually a specific parasitoid of stigma and P. bedeguaris. P. brandtii, less commonly of D. rosae (Blair Moreover, it seems that there is a nega- 1945, Claridge and Askew 1960). However, tive association between the presence of D. neither P. brandtii nor its specific parasitoid rosae males and parasitization (cf. Table 1), Volume 15, Number 2, 2006 283

results for the one Table 3. Summarized way the tested except for T. = = parasitoid species, ANOVA analysis (FD 273; F 49.67; p < 0.001) = bedeguaris and P. bedeguaris males (FD performed on the total length of male and female D. = 273; F 49.67; p < 0.001) (Table 3 and rosae and its parasitoids (small samples were excluded Given similar of the from analysis). D. rosae males are statistically shorter Fig. 4). shapes species than all the parasitoid species, except for T. bedeguaris concerned, we take "shorter" to be a good and P. bedeguaris males (in bold). indication of "smaller". From the ANOVA size did not differ D. rosae females D. rosae males analysis parasitoid between samples with or without males O. mediator both sexes longer* longer of D. rosae (cf. Fig. 4), and neither did their not different T. bedeguaris males shorter = sex ratio differ (Wilcoxon Test; Z 0.36; p T. bedeguaris females longer* longer = = n.s.; fd 4). However, our data are G. stigma females not different longer different D. rosae P. bedeguaris males shorter not insufficient to ascertain whether females not different P. bedeguaris longer males represent an inadequate food re- source for and /or * T. parasitoid development Even if statistically longer, O. mediator and of D. contain them are bedeguaris females overlap the maximum size whether galls which rosae females O. mediator males (Fig. 4), only being avoided by parasitoids. It would be in- Moreover the two parasitoid species clearly longer. to are much more slender than the massive D. rosae teresting compare parasitization per- females, the latter probably being yet a suitable food centages in other parts of Europe where for them. source good numbers of males of D. rosae are recorded. even if no correlation came out between = ACKNOWLEDGMENTS parasitization and D. rosae sex ratio (r = = thank much R.R. Askew who revised the -0.3; fd 16, p 0.21). Comparing total We very and an anonymous referee for his useful lengths of D. rosae adults (males and manuscript to Broad suggestions. We are also very grateful Gavin females) with total lengths of their com- and Mark Shaw, whose advice greatly improved the a one monest parasitoids through way last draft. For their kind collaboration in the identifi- that D. rosae ANOVA analysis, we found cation of specimens we thank: Jose Luis Nieves Klaus Horstmann males are statistically much shorter than all Aldrey (Cynipidae), (Ichneumoni-

— 5,5 'i i i r™

I •!• ! 5,0 It 4,5 X

4,0 ca XTlj 3,5 i hi

3,0

2,5 T^~ 1 | I [$#'] | ± Std. Dev 2,0 Mean I I ± Std Err ~1~

1,5 o o o co D CO LL. O O O a Q qor of Hymenoptera Research 284 Journal

confirmed the Houard, C. 1908. Les Zoocecidies des Plantes d'Enrope et dae), and Csaba Thuroczy, who du Bassin de la Mediterranee. Vol. 1. Libr. Sc. A. identifications of Chalcidoidea. We also thank Simon and Hermann, Paris. Randolph for his helpful information Virgilio Niblett, M. 1951. Some notes on the genus Rhodites Caleca, who collected some samples. (Hymenoptera, Cynipidae). Proceedings of the South London and Natural History LITERATURE CITED Entomological Society 1949-1950: 204-206. L. 1981. Datos sobre rosae Nieves-Aldrey, J. Diplolepis R. I960. Some observations on Askew, R. Diplolepis sus (L.) (Hym., Cynipidae) y himenopteros rosae and its parasites. (L.) (Hym. Cynipidae) en Salamanca. Bolctin de la Asociacion 95: 191-192. parasitos Entomologist's Monthly Magazine 107-113. Espanola de Entomologia 4 (1980): Askew, R. R. 1961. Eupelmus urozonus Dalman (Hym., J. L. 2001. Cynipidae. oak Nieves-Aldrey, Hymenoptera Chalcidoidea) as a parasite in cynipid galls. In: Ramos, M. A. et al., eds. Fauna Iberica, vol. 16. The 94: 196-201. Entomologist Museo Nacional de Ciencias Naturales, CSIC, 1984. The of the Gall In: Askew, R. R. Biology Wasps. Madrid. Ananthakrishnan, T. N., ed. Biology of Gall Insects. Nordlander, G. 1973. Parasitsteklar i galler au Diplo- and IBH New Oxford Publishing Company, D. Schlecttd. lepis rosae (L.) och mayri (Hym., Delhi. Cynipidae) (Hym., Ichneumonoidea, Chalcidoi- utiles. Octave Doin Beauvisage, G. E. C. 1883. Les galles 94 dea, ). Entomologisk Tidskrift (3-4): Paris. Editeur, 148-176. K. G. 1943. On the rosa and its Blair, bedeguar gall S. 2003. Universal Chalcidoidea Database. Noyes, J. inhabitants. Monthly Magazine 79: Entomologist's World Wide Web electronic publication, www. 231-233. nhm.ac.uk/entomology/chalcidoids/index.html the rose Blair, K. G. 1945. Notes on the economy of [accessed 19-Dec-2005]. formed Rhodites (Hymenoptera, Cynipi- 1-7 galls by Pagliano, G. 1995. Hymenoptera Cynipoidea. Pp. of the Society dae). Proceedings Royal Entomological in: Minelli, A., S. Ruffo, and S. La Posta, eds. London (A) 20: 26-31. 96. of Checklist delle specie della fauna italiana, Calder- Callan, E. M. 1940. On the occurrence of males of ini, Bologna. Rhodites rosae L. Cynipidae). Pro- (Hymenoptera: Pagliano, G. and P. Navone. 1995. Hymenoptera the Royal Society of London S. ceedings of Entomological Chalcidoidea. Pp. 1-40 in: Minelli, A., Ruffo, 15: 21-26. (A) and S. La Posta, eds. Checklist delle specie della E. M. 1944. Thomson and Callan, Habrocytus bedeguaris fauna italiana, 97. Calderini, Bologna. n. Pteromalidae) 1997. Habrocytns periclisti sp. (Hym., Pagliano, G., L. Santini, and F. Strumia. reared from of Rhodites rosae L. galls Proceedings of Descrizione delle carte del manoscritto. Pp. 95- the London (B) 13: Royal Entomological Society of 246 in: Bernardi, W., G. Pagliano, L. Santini, F. 90-93.' Strumia, L. Tongiorgi Tomasi, and P. Tongiorgi M. F. and R. R. Askew. 1960. Claridge, Sibling species Natura e Immagine. ETS ed., Pisa. in the rosae Eun/toma group (Hym., Eurytomidae). Pellizzari Scaltriti, G. 1988. Guida al riconoscimento delle 5: 141-153. della italiana. Patron ed., Entomophaga piit comuni galle flora Constantineanu, M. I. and C. Pisica. 1970. L'etude de Padova. la tribu des de la Pimplini (Hym. Ichneum.) Picard, F. 1926. Notes biologiques sur le Rhodites rosae Socialiste Romania Republique (insectes L. Feuillc des Jeunes Naturalistes 47: 113-116.

auxiliares a la et a Y. G. I. Le sylviculture l'agriculture). Plantard, O., J. Rasplus, Mondor, Clainche, Analelc ale Universitatii "Al. I. Cuza" din Stiintifice and M. Solignac. 1998. Wolbachia-mduced thely- 2: 1-106. lasi, Monografii toky in the rose gallwasp Diplolepis spinosissimae Sicilia. La De Stefani Perez, T. 1887. Cynipidae della (Giraud) (Hymenoptera: Cynipidae), and its Sicilia 5: 233-249. agricola consequences on the genetic structure of its host. Stefani T. 1902. Nuovi insetti e B 265: De Perez, galligeni Proceedings of the Royal Society of London cecidii vecchi e nuovi. Marcellia 5: 109-115. 1075-1080.

De Stefani T. 1905. Breve descrizione dei Y. I. Le Perez, Plantard, O., J. Rasplus, G. Mondor, Clainche, zoocecidii siciliani sino conosciuti. // ad oggi and M. Solignac. 1999. Distribution and phylog- Naturalista siciliano 18: 89-96, 104-116, 136-141, eny of Wolbachia inducing in Rhoditini 178-191. 160-168, and 'Aylacini' (Hymenoptera: Cynipidae). Insect Fulmek, L. 1968. Parasitinsekten der insektengallen Molecular Biology 8: 185-191. zur 18: 719-952. sobre Europas. Beitrage Entomologie Pujade Villar, J. 1983. Algunas consideraciones en Grissell, E. E. 1995. Toryminae (Hymenoptera: Chalci- Diplolepis rosae (L.) y D. mayri (Schlech) doidea: Torymidae): a redefinition, generic clas- Catalunya (Hym., Cynipidae, Cynipinae). Actas sification, and annotated world catalog of species. del Primer Congreso Iberico de Entomologia, Leon II: Memoirs on Entomology, International 2: 1-470. 613-622. Volume 15, Number 2, 2006 285

Randolph, S. 2005. The Rose Bedeguar Gall and its Insect croy (Hymenoptera: Cynipidae) inducing galls on community. The British Plant Gall Society, Bury St the stems of Rosa acicularis. Canadian Entomologist Edmunds, UK. 116: 1623-1636. R. Redfern, M. and R. Askew. 1992. Plant galls. StatSoft Italia s.r.l. 2003. STAT1STICA. Sistema soft- Naturalists' Handbooks No 17, Richmond Pub- ware di analisi dei dati, versione 6. www.statsoft.

lishing, Slough, UK. it. Scaramozzino, P. L. 1995. Hymenoptera Ichneumoni- Stille, B. 1984. The effect of hostplant and parasitoids dae. Pp. 1-62 in: Minelli, A., S. Ruffo, and S. La on the reproductive success of the parthenoge- Posta eds. Checklist delle specie delta fauna italiana, netic gall wasp Diplolepis rosae (Hymenoptera, 94. Calderini, Bologna. Cynipidae). Oecologia 63: 364-369. Schroder, D. 1967. Diplolepis (= Rhodites) rosae (L.) Stille, B. and L. Davring. 1980. Meiosis and reproduc- (Hymenoptera: Cynipidae) and a review of its tive strategy in the parthenogenetic gall wasp parasite complex in Europe. Commonwealth Institute Diplolepis rosae (L.) (Hymenoptera: Cynipidae). of Biological Control, Technical Bulletin 9: 93-131. Hereditas 92: 353-362. A. Ritchie. B. in Shorthouse, J. D. and J. 1984. Description Walsh, G. 1924. Proportion of sexes Rhodites rosae. and biology of a new species of Diplolepis Four- Entomologist's Monthly Magazine 60: 156. J. HYM. RES. Vol. 15(2), 2006, pp. 286-289

NOTE

Palaearctic Bee Hylaens punctatus (Brulle) (Colletidae), a Long Established in South America

Arturo Roig-Alsina

Rivadavia" Av. Gallardo Museo Argentino de Ciencias Naturales "Bernardino Angel 470, 1405 Buenos Aires, Argentina; email: [email protected]

in a The bee (Brulle) is sternum ends spoon-shaped expan- recorded for the first time for Argentina, sion, which protrudes from the apex of the where it was introduced over thirty years abdomen. exotic of bees have ago, according to museum records. The At least six species introduction of H. punctatus into the New become established in Argentina. Of these, World has already been reported for the only Hylaeus punctatus is deemed to have United States and Chile. Snelling (1983a) been introduced accidentally into this first recorded the species from southern country. Other species have been purpose- California in 1981, and later Toro et al. ly introduced, or have been introduced (1989) reported its first sighting in the into neighboring countries and later dis- central region of Chile in 1986. The study of persed to Argentina. Two species, Apis rotundata museum specimens reveals that H. puncta- mellifera L. and Megachile (Fabri- tus was present in southern South America cius), were introduced for economic pur- before these dates. Two specimens kept in poses at various times. A fourth species, the collection of the Museo Argentino de the Palaearctic Bombus ruderatus Fabricius, Ciencias Naturales in Buenos Aires were was introduced into Chile for pollination, collected by the entomologist Adriana and later expanded its range into southern Oliva in the city of Buenos Aires in May Argentina (Roig Alsina and Aizen 1996). A 1976. Today the species is a common and fifth species, huberi Ducke, is abundant member of the bee fauna of the widespread in Brazil, where it occurs from city of Buenos Aires and surrounding the northern state of Para to Minas Gerais areas. This bee is further recorded here and Sao Paulo (Silveira et al. 2002). It for two other, distant, localities in Argen- occurs in Argentina in the province of tina: the central west province of Mendoza Misiones (San Ignacio, Museo Argentino and the Patagonian province of Rio Negro, de Ciencias Naturales, new record). Ac- This small colletid bee is native to the cording to Snelling (1983b) this species Mediterranean area of the Palaearctic Re- belongs to the Indo-Australian group of gion (Dathe 1980, Snelling 1983a). It be- Lithurgus atratus (Smith); moreover, he longs to the subgenus Spatulariella Popov, found no morphological differences by all the species of which have a Palaearctic which L. atratus may be separated from L. distribution (Michener 2000), with the only huberi, considering that this species is exception being H. punctatus, adventive in adventive in Brazil, probably introduced the New World. This species is easy to by man in historical times. Another exotic distinguish from the many native South species present in Argentina is the Old American species of Hylaeus by the eighth World (L.). This inetasomal sternum of the male. This species has been recorded for the province Volume 15, Number 2, 2006 287

of Buenos Aires by Michener (2000). Facultad de Agronomia, XII-29-1994 (A. Roig Anthidium manicatum also occurs in the A.); Parque Centenario, X-31-1995, in balcony h on 7' floor on flowers of eastern and southern states of Brazil, Alyssum sp. (A. Roig A); 1-18-1996, in house on where it would have been introduced from Colegiales, garden flowers of Asclepias curassavica (A. Oliva); Europe (Silveira et al. 2002); according to Colegiales, 11-12-1996, in house garden on these authors the species, although broadly flowers of Mentha sp. (A. Oliva); Palermo, II-6- distributed in seems to be rare. Brazil, in 2003, park on flowers of Lavandula officinalis a candidate to become Finally, likely (A. Roig A.). Province of Buenos Aires: Partido de established in in the near future Argentina Vicente Lopez, La Lucila, 11-17-1985 (L. Moffatt); is terrestris in- Bombus (L.), which was Partido de Tigre, 12 km NW Tigre, 1-20-1990, on troduced into Chile for pollination in 1997 flowers of Eryngium sp.(A. Roig A.); Partido de on (Estay and Vitta 2004). This bee is expected Escobar, Maquinista Savio, 1-5-1997, flowers to follow the same route across the low of Baccharis pingraea (A. Roig A.); Partido de Vicente La Lucila, 1-1-2003 southern Andes into western Patagonia as Lopez, (A. Roig A.); Partido de Hurlingham, Estacion Experimental did Bombus ruderatus in the early 1990's. Inta Castelar, XI-6-2003, on flowers of Lithrea Hylaeus punctatus has been collected in molleoides (Veil.) Engl. (L. Compagnucci & A. the Buenos Aires area in urban and sub- Roig A.). Province of Mendoza: Mendoza city, II- urban, as well as in natural habitats. In the 15-1994, in house garden on flowers of Mentha of Buenos Aires it is common in city parks, aquatica (A. Roig A.). Province of Rio Negro: El and has been collected visiting flowers on Bolson, II-3-1994, in house garden (A. Roig A.). balconies of buildings up to the seventh Species of Hylaeus nest in pre-existing floor. Both males and females are common. holes and crevices, mainly in wood and The bee has been found in urban habitats twigs, but in other materials as well. This in the provinces of Mendoza and Rio nesting behavior has surely facilitated the Negro, in house gardens in the city of transport by man of these bees for long Mendoza, and in house gardens of the distances, carried along with their nesting town El Bolson in the latter province. Many substrates. The type of nesting behavior records of flower visitation correspond to among bees is tightly correlated with their introduced exotic plants, such as Eryobotria chance of becoming introduced species. jnponica (Thunb.) Lindl., Lavandula officina- There are two main types of nests (Maly- lis Chaix, Mentha aquatica L., and Alyssum shev 1935, Stephen et al. 1969, Michener sp. Specimens have also been collected on 2000): those of burrowing bees, which have curassavica an excavated tunnel in the Asclcpias L., Eryngium sp., and (usually soil) Baccharis pingraea DC, all of which are leading to the cells, and those of non- native plants. Toro et al. (1989) recorded burrowing bees, which are constructed in that the bee flies late in the season in Chile, the open or taking advantage of pre- formed cavities. there but collecting dates in the area of Buenos Although are certain Aires indicate that the bee is active from species of bees with intermediate behaviors spring (October) to late summer (May). (Stephen et al. 1969), and there are certain in which both of behavior The following is a list of the collection lineages types are this classification holds for records for H. punctatus in Argentina. Speci- present, mens studied and cited below are housed entire lineages of bees. Most species of bees that have been introduced into at the collection of the Museo Argentino de accidentally Ciencias Naturales "Bernardino Rivadavia" exotic areas are non-burrowers. For exam- of fourteen that have in Buenos Aires, Argentina. ple, species reported become established in the Buenos Aires City: Palermo, V-22-1976, in accidentally United States Eickwort house garden on flowers of Eryobotria japonica (Daly 1966, 1980, (A. Oliva); Villa Crespo, IV-14-1985, in balcony Snelling 1983a, Mangum and Brooks 1997), th on 7 floor (A. Roig A.); Botanical Garden, thirteen of them are non-burrowing bees. 288 Journal of Hymenoptera Research

that have from Ar- These bees belong in the genera Anthidium nopterans dispersed for the Fabricius, Latreille, Hoplitis gentina to Chile, example paper- Polistes Brethes D'An- Klug, Lithurgus Berthold, and Megachile wasp buyssoni (Perez the of Latreille of the family Megachilidae, Cer- gello 1970). Of course, dispersal to the atina Latreille of the family Apidae, and Hylaeus punctatus New World may have occurred several times Hylaeus Fabricius of the family Colletidae. independently World. various As another example, the only two exotic from the Old The possibil- ities of are difficult to assess species that occur in Brazil (Silveira et al. dispersal as 2002) which have been accidentally in- without appropriate methods, such a molecular be- troduced, belong to the genera Anthidium phylogeographic analysis, the of this contribution. and Lithurgus. All these groups nest in pre- yond scope holes or crevices. existing LITERATURE CITED Available records and distribution sug- gest that H. punctatus is mainly associated Daly, H. V. 1966. Biological studies on Ceratina an alien bee in California which today with human-modified habitats in dallatorreana, reproduces by parthenogenesis. Annals of the Argentina. The dispersal of this species Entomological Society of America 59: 1138-1154. within the country (city of Mendoza, and Dathe, H. H. 1980. Die Arten de Gattung Hylaeus F. in El Bolson have been mediated town) may Europa. Mitteilungen aits dem Zoologischen Museum by human transport, in a similar way in in Berlin 56: 207-294. G. C. 1980. Two of which the species first came to the country. Eickwort, European species Chelostoma established in New York State, USA There are no existing records in interme- (Hymenoptera: Megachilidae). Psyche 87: 315-323. diate places between Buenos Aires and Estay, P. and N. Vitta. 2004. Evaluacion de Bombus these two distant localities, sur- although terrestris reproducidos en Chile como polinizador veys have been conducted. In the province de flores de Palto, Persea americana. XXVI Congreso National de Sociedad de of Buenos Aires it has penetrated agricul- Entomologia, Entomologica Chile. Universidad de tural (Hurlingham, Escobar) and natural Concepcion, Concepcion, Chile. habitats (Tigre), where it has been recorded Malyshev, S. I. 1935. The nesting habits of solitary native On this the visiting plants. ground, bees. A comparative study. Eos 11: 201-309. progressive expansion of the bee to other Mangum, W. A. and R. W. Brooks. 1997. First records Smith areas in Argentina can be predicted. of Megachile () sculpturalis (Hymenoptera, Megachilidae) in the Continental The possible geographic origin of the United States. Journal of the Kansas Entomological invasive population in California was Society 70: 140-142. discussed He by Snelling (1983a). argues Michener, C. D. 2000. The Bees of the World. The John that this population matches more closely Hopkins University Press, Baltimore and London. the color variant found in southern Europe, Perez D'Angello, V. 1970. El genero Polistes Latreille en Chile. Museo Nacional de Historia than the darker forms found in central 1802, Natural, Santiago, Noticiario Mensual 169: 10. Europe or the more extensively pale Roig Alsina, A. and M. A. Aizen. 1996. Bombus marked of the H. specimens subspecies ruderatus Fabricius, una nueva especie de Bombus punctatus longimaculus Alfken. In this re- para la Argentina (Hymenoptera: Apidae). Physis, spect the populations from Argentina Buenos Aires, Sec. C 51: 49-50. Silveira, F., G. A. R. Melo, and E. A. B. Almeida. 2002. agree with the pattern described by Snel- Abelhas brasileiras. Sistemdtica e identificacao. Belo Since introduction in ante- ling. Argentina Horizonte. dates introduction in the U.S.A., the Cali- Snelling, R. R. 1983a. Studies on North American bees fornia population may have originated of the genus Hylaeus, 6. On adventive Palaearctic either from Europe or from Argentina. species in southern California. Bulletin of the Southern Sciences 82: 12-16. The same may be argued for the Chilean California Academy of Snelling, R. R. 1983b. The North American species of population. At least a dispersal from the bee genus Lithurge (Hymenoptera: Megachi- to Chile seems feasible, Argentina taking lidae). Contributions in Science, Natural History into account the invasion of other hyme- Museum of Los Angeles County 343: 1-11. 289 Volume 15, Number 2, 2006

P. E. Torchio. 1969. The Y. Frederick, and A. 1989. Stephen, W. P., G. E. Bohart, and Toro, H., Henry. Hylaeinae bees with a Colletidae), nueva subfamilia biology and external morphology of synopsis (Hymenoptera, America. Ex- la fauna Chilena. Acta Chilena ofthe genera of'Northwestern Agricultural para Entomological 15: 201-204. periment Station, Oregon State University, Corvallis. J. HYM. RES. Vol. 15(2), 2006, pp. 290-306

Skeletal Anatomy of the Mesosoma of Palaeomymar anomalum (Blood & Kryger, 1922) (Hymenoptera: Mymarommatidae)

Lars Vilhelmsen and Lars Krogmann

of of (LV) Zoological Museum, Natural History Museum Denmark, University Copenhagen, Universitetsparken 15, DK-2100; email: [email protected] of (LK) Zoological Institute and Zoological Museum, University Hamburg, email: Martin-Luther-King-Platz 3, D-20146 Hamburg, Germany; [email protected]

anomalum Abstract. — Detailed study of the skeletal anatomy of the mesosoma of Palaeomymar was for the undertaken, primarily with SEM. Four previously unrecognized putative autapomorphies anterior thoracic fusion Mymarommatidae were discovered: 1) absence of a functional spiracle; 2) a of rods on the anterior surface of of the propleural arm with the profurcal arm; 3) presence of pair of a concealed fused with the prophragma; and 4) absence of the metafurca. The presence prepectus the confirmed. Additional features of the posterolateral margin of pronotum was possible the of are described and significance for evaluating phylogenetic position Mymarommatidae is well discussed. A sister-group relationship with the Chalcidoidea the most supported hypothesis presently, but needs additional corroboration.

The small family Mymarommatidae is to the metapleuron-propodeum (Gibson taxon. the fore has a reticulate arguably the most enigmatic wasp 1986, fig. 10); wing to be confused Eight extant and ten extinct species have pattern on the surface (not been described (Grimaldi and Engel 2005). with true wing venation; Gibson 1986, fig. The biology of these minute wasps is 30); and the hind wing is reduced to that is bifurcate almost entirely unknown. They appear to a rodlike structure apically be associated with leaf litter and were to clasp and support the base of the Further reared once from a bracket fungus (Gibson forewing (Gibson 1986, fig. 33). et al. 1999). Because of the small size of the putative autapomorphies are discussed by adults (immatures are unknown), it has Gibson (1986) and Gibson et al. (1999) (see been suggested that they are egg parasi- also below). toids, but this is entirely conjectural. The While there has never been any serious absence of biological information for My- doubt about the monophyly of the Mymar- marommatidae is all the more intriguing ommatidae, their phylogenetic position because they display a number of highly within the Hymenoptera is less well unusual autapomorphic features, which in corroborated. Debauche (1948) was the the case of the first to in their own — head anatomy border on place them family bizarre the occipital region is separated (Mymarommidae [sic]), removing them from the remainder of the head capsule by from Mymaridae. This was reversed by an expanse of pleated membrane that some subsequent authors. The first corn- allows the head to expand and contract in prehensive discussion of mymarommatid an accordion-like manner (see Gibson 1986, affinities in a cladistic context was un- fig. 9). Also, the presternum and propleura dertaken by Gibson (1986), who placed are extensively fused, forming a continuous them as sister group to the Chalcidoidea. carapace in external view; the mesopleuron The main evidence for this is the presence is fused along most of its posterior margin of axillar phragmata, apodemes projecting Volume 15, Number 2, 2006 291 under the mesoscutum and accommodat- rinsed in demineralised water and trans- ethanol. were then ing the origins of the mesotergal-trochan- ferred to 70% They teral muscles (Gibson 1986, 1999, Gibson et transferred through a series of intermedi- al. 1999). This hypothesis was later sup- ate concentrations to absolute ethanol, ported by the phylogenetic analyses of which served as the transition medium Ronquist et al. (1999). for critical point drying. After critical point on Debauche (1948) provided an overview drying, the preparations were mounted of the anatomy of Palaeomymar anomalum. stubs with double-adhesive tape and coat- Gibson (1986) and Heraty et al. (1994) ed with platinum prior to examination in treated selected aspects of mesosomal a Jeol JSM-6335F field emission SEM unit, anatomy in considerable detail, but a com- RESULIb prehensive survey of this body region in Mymarommatidae is wanting. The present The mesosoma of Palaeomymar anomalum examines the external and internal It paper [ s a compact structure. can be subdivided skeletal anatomy of the mesosoma and into four major components that are of the anterior part metasoma (the petiole) comparatively easily separated by dissec- for P. will be discussed the anomalum. Findings tion (Fig. 1): 1) the pronotum; 2) on in a phylogenetic context, drawing propectus, comprising the propleura and information obtained from ongoing sur- prosternum and accommodating the head in veys of the mesosoma in Hymenoptera articulation and foreleg attachments; 3) the general and Chalcidoidea in particular, mesonotum and attached pro- and meso- L. Vilhelmsen and L. undertaken by phragma; 4) the mesopectus-metathorax- Krogmann, respectively. propodeum complex. The latter is by far the of the mesosoma and is MATERIALS AND METHODS largest part extensively fused to a degree where the Examined material. —Palaeomymar anoma- boundaries between the individual parts lum (Blood & Kryger, 1922). GERMANY, are not obvious. Niedersachsen, Luchow-Dannenberg, NSG In the following, each of these regions Forst Lucie. Car net. 15.viii.2001. Leg. H. are described in turn. In addition, we Meybohm. 5 females. Specimens and vou- describe in detail the anterior part of the chers deposited in Zoologisches Museum petiole comprising the mesosoma-meta- Hamburg. SWEDEN, Sm[aland], Almhults soma articulation. kommun, Stenbrohult, Djaknabygds bok- Pronotum. —Pronotum narrow medially into backe, N56 36.548', E14 11.583' (=Trap ID (Fig. 3), laterally expanded trapezoidal 24). Heath with old beeches, l-18.viii.2003 flange (Figs 2, 3); posteriorly articulating = coll. ID Swedish Malaise with with ( event 818). Leg. mesopectus, ventrally propectus striate Trap Project (Swedish Museum of Natural (Fig. 1); with sparse, sculpture, of History). 6 females. Specimens and vou- except for small median band dense, sub- chers deposited in the Zoological Museum, raised longitudinal striations (Fig. 3); dorsal University of Copenhagen. laterally with deep depression pos- — distinct vertical carina Pweed lire for slide mounting. Two speci- terior to extending mens were dissected and mounted on almost entire height of pronotum (Figs 2, in frontal a slide in Entellan. 3). Anteromedian margin view incurved Procedure for SEM-investigations.—Speci- distinctly (Fig. 3), forming slight- anterior of transverse mens were cleaned in a sonicator and ly upturned lip just dissected with small razor blade scalpels pronotal sulcus; internally, anteromedian of and minutien needles. The preparations margin inflected, the median part were macerated in KOH for a few hours at pronotum with concavity accommodating Dorsal in 40 C or overnight at room temperature, prophragma (Fig. 5). margin 292 Journal of Hymenoptera Research

of of anomalum. lines indicate boundaries Fig. 1. Lateral view mesosoma Palaeomymar Stippled approximate = fore hw = hind = between mesopectus, metathorax, and propodeum. Abbreviations: fw wing base; wing; pas = = propodeal antecostal suture; pes paracoxal sulcus; pl3p metapleural pit.

dorsal view deeply incurved, in lateral pleura in ventral view linearly separated fused view with long seta laterally close to dorsal anteromedially, posteromedially with 1 or 2 margin (Figs 2, 3). Posterolateral margin (Figs 6, 7); anteriorly elongate with almost straight, with small notch about 1/3 setae sublaterally (Fig. 6); posteriorly of total height from posterodorsal corner carinate margin slightly extended poster- with (Fig. 2); posteroventral corner extended omedially (Fig. 6). Propleura laterally ventrally between pro- and mesopectus, an oblique longitudinal carina separating slender (Fig. 2). Internally, pronotum pre- lateral, dorsal smooth part from ventrally dominantly smooth, posterolateral margin reticulate surface; posterior part of carina flanking low vertical carina (Figs 4, 5), the laterally delimiting distinct groove accom- carina (interpreted as the prepectus, see modating lateroventral margin of prono- discussion) extending for some distance, tum and proximal part of procoxa (Fig. 8). but neither reaching dorsal nor ventral Internally, propleural arm on posterodor- margin of pronotum, and deflected anteri- sal corner of propleuron slender, projecting orly prior to reaching dorsal margin; small posteriorly, posterior end fused with later- pit just anterior of carina (Fig. 4) associated al end of profurcal arm (Fig. 9, ppa). with narrow strands of tissue. Anterior Presternum (Fig. 8, stl) reduced in size, thoracic ('mesothoracic') spiracle not ob- inflected at an angle of about 90 degrees served and internally— or externally. relative to posterior part of propleura Propectus. Propleural cervical promi- not visible externally (Fig. 6); extensively nence (Fig. 7, cep) with head articulation fused with propleura ventrally, the bound- at anterodorsal corner of propectus, not ary of fusion indiscernible; externally retracted, separated from rest of propleura smooth, expanded dorsally of narrow by transverse carina (Fig. 7); two short procoxal cavities, separate from propleura setae situated anteriorly of carina, one laterally and with slit-like profurcal pit posteriorly. Cervical swellings with patch medially (Fig. 8, fulp). Internally, profur- of sensilla, otherwise not developed. Pro- cal base narrow, not extending far anteri- Volume 15, Number 2, 2006 293

£&WBf

Figs 2-5. Pronotum of P. anomalum. 2. Exterior lateral view. 3. Exterior anterior view. 4. Interior view of pronotal-mesopectal juncture showing prepectus (dorsal surface to left). 5. Interior, posterior view. = = = Abbreviations: n2 mesonotum; pll propleuron; p!2 mesopleuron.

orly (Fig. 9, ful), profureal arms slender, times with tonofibrillae (Fig. 14, tf) distally extending laterally (Fig. 9, fula); profurcal and apparently closely associated with bridge absent, anterior profurcal apodemes dorsomedian margin of pronotum. Mesos- not developed, only slender tendons pres- cutum with transverse groove just dorsal to ent sublaterally (Fig. 9, te). Procoxae very rods between its external surface and reduced proximally, without transverse prophragma accommodating dorsal mar- carina. Katepisterna and prosternal scler- gin of pronotum. Mesoscutum with medi- ites not observed.— an mesoscutal sulcus, notaulus, and para- Mesonotum. Externally, mesonotum psidal lines entirely absent externally and predominantly with reticulate sculpture internally (Figs 10, 12); with two elongate (Fig. 10). Internally, prophragma well de- setae posterolaterally just anterior to trans- veloped, projecting anteriorly under pro- scutal articulation (Fig. 10); transscutal notum, not subdivided medially nor ex- articulation complete, straight (Fig. 10, tended laterally (Figs 12, 14, phi); dorsal tsa); parascutal carina well developed surface of prophragma with paired, ante- laterally, overhanging posterodorsal mar- riorly projecting, slender cylindrical rods gin of pronotum and fore wing articulation submedially, the rods extending at an point (Fig. 16, psc) and extending posterior angle of approx. 90 degrees to each other to lateral end of transscutal articulation (Fig. 14) but continuous medially, some- (Fig. 10, psc), with two elongate setae 294 Journal of Hymenoptera Research

Figs 6-9. Propectus of P. anomalum. 6. Exterior ventral view; 7. Exterior anterior view. 8. Presternum, posterior = = view (pronotum removed); 9. Profurca, dorsal view. Abbreviations: cep cervical prominence; cxl procoxa; ======ful profurca; tula profurca! arm; fulp profurcal pit; pl2 mesopleuron; stl presternum; ppa = propleural arm; spl: prospinasternum; te tendon.

below parascutal carina just above fore extending to distal end of mesoscutellar of wing base (Fig. 16). Tegula absent. Mesos- arm (Fig. 11, sea); anteroventral branch cutellum with convex, reticulate anterome- axillar carina with seta. Mesoscutellar arm dian part between lateral depressed areas evident dorsolaterally on mesoscutellum having more weakly developed sculpture, just anterior of frenum (Fig. 11, sea), and and with depressed, longitudinally striate with anterolateral part of arm slightly posterior frenum (Figs 10, 11) anteriorly overhanging fore wing base and terminat- delimited by external frenal line (Fig. 10, ing in slender vertical rod just anteriorly frle); convex median part with elongate of hind wing base (Fig. 11, hw). Internally, seta laterally, without evident scutoscutel- mesoscutellum delimited by low mesos- lar sulcus (Figs 10, 11); frenum with sub- cutellar carina anterolaterally, the carina lateral triangular extension overlapping not continuous medially (Fig. 13, msc); metanotum dorsally (Fig. 11, n3). Mesos- surface spanned by extended septum cutellum with axillar carina distinct later- penetrated by small oval fenestrum in ally (Figs 11, 16, axe), the carina subdi- middle (Fig. 13, fen) and with a convex vided posteriorly into posterodorsal line (Fig. 13, frli) corresponding to exter- branch extending across mesoscutellar nal frenal line delimiting oblong trans- arm proximally and anteroventral branch verse impression posteriorly of fenestrum; Volume 15, Number 2, 2006 295

Figs 10-15. Mesonotum of P. anomalum. 10. Mesonotum, dorsal view. 11. Mesoscutellum and axilla, posterolateral view. 12. Mesoscutum, ventral view. 13. Mesoscutellum, ventral view. 14. Prophragma, = = anterolateral view. 15. Mesophragma, ventral view. Abbreviations: axe axillar carina; axp axillar phragma; = = = = = fen fenestrum; frle external frenal line; frli internal frenal line; hw hind wing; msc mesoscutellar = = = = = carina; n3 metanotum; phi prophragma; prf propodeal foramen; psc parascutal carina; sea mesoscutellar arm; tf = tonofibrillae; tsa = transscutal articulation; tsc = transscutal carina.

anteriorly with rod-like axillar phragmata (Fig. 13, tsc) and projected anterolaterally, (Figs 12, 13, axp) arising just posterior of the projected parts inflected rather than transscutal articulation, the rods continu- flush with inner surface of mesoscutum ous medially along transscutal carina (see discussion). 296 Journal of Hymenoptera Research

Figs 16-20. Mesopectus of P. anomalwn. 16. Dorsal part of mesothorax, lateral view. 17. Ventral part of mesopleuron, ventral view. 18. Dorsal part of mesopleuron, internal view; 19. Posterior part of mesopleuron, = = posterior view. 20. Mesofurca, anterior view. Abbreviations: axe axillar carina; cx2f mesocoxal foramen; = = cx2ma mesocoxal median articulation; di2 = mesodiscrimen; di21 = mesodiscrimenal line; fu2a mesofurcal = = = = nl = arm; fu2p mesofurcal pit; fw forewing base; hw hind wing; mpa mesopleural arm; pronotum; = = = ph21 mesolaterophragmal lobe; pl3a metapleural apodeme; psc parascutal carina.

Mesophragma (Fig. 15) large, rectangu- tendon medially (Fig. 21). Laterophragmal lar, extending far into propodeum almost lobes developed as slender, ventromedially to propodeal foramen; pseudophragmal projected rods on anterolateral extensions ventral lobes, median longitudinal carina, of mesophragma (Fig. 18, ph21). and posterior median slit not developed, Mesopectus-metathorax-propodeitni com- but with broad — low, transverse swelling plex. Mesopectus in anterior view with just posterior to anterior margin; anterior mesopleural margins forming U-shape, with margin tiny apodeme and associated lateroventrally with triangular flange op- Volume 15, Number 2, 2006 297

Figs 21-23. Metathorax of P. anomalum. 21. Metanotum, anterodorsal view (dissected from mesothorax). 22. = Ventral metapleuron, internal view. 23. Ventral metapleuron, external view. Abbreviations: di3 metadiscri- = = = = men; di31 metadiscrimenal line; n3 metanotum; pas propodeal antecostal sulcus; ph2 mesophragma; = pl3c metapleural paracoxal carina.

posite basal part of procoxa, the flange with distinct spine anterolaterally laterally articulating with posteroventral (Fig. 17), and posteriorly with transverse corner of pronotum (Fig. 17, nl). Meso- carina delimiting smooth area around pleura with transverse sulcus extending mesocoxal foramina (Figs 17, 19). Internal- entire anterior along margin, internally the ly, mesopleural apodeme and mesepister- sulcus corresponding to low transverse nal ridge absent. Mesocoxal foramen carina extending onto mesopleural arms (Fig. 19, cx2f) narrow, laterally overlapped (Fig. 18). Prospinasternum present as very by flange (Fig. 1) continuous with trans- small area anterior to transverse carina, verse carina; median mesocoxal articula- visible in hardly external view; internally, tion developed as short lobe (Fig. 17, prospinasternal apodeme absent (Fig. 9, cx2ma). Mesocoxa with proximal part spl). Mesopleural arm (Fig. 16, mpa) dor- (basicoxite) reduced. Mesodiscrimenal line sally accommodating fore wing articula- (Fig. 17, di21) a longitudinal smooth groove tion, posteriorly delimited by distinct extending from anterior margin of meso- notch; internally, mesopleural arm sup- pectus and terminating in mesopleural pit ported by short, almost vertical carina, (Fig. 19, fu2p) between mesocoxal forami- ventrally delimited by short horizontal na. Internally, mesodiscrimen (Fig. 20, di2) ledge connected to anterior transverse anteriorly continuous with transverse cari- carina (Fig. 18). Mesobasalare and posteri- na; discrimen rising gradually to form high or thoracic (meta thoracic) spiracle neither septum at point of origin of mesofurcal observed internally nor externally. Meso- arms (Fig. 20, fu2a), then descending steep- pleuron with lateral surface reticulate ly to terminate in mesofurcal pit. Proximal anteroventrally, smooth posterodorsally part of mesofurcal arm slender, extending (Fig. 1); ventral surface reticulate/scaly, dorsolaterally, with shallow depression 298 Journal of Hymenoptera Research

Figs 24-26. Propodeum of P. anomalum. 24. Posterior view; 25. Dorsolateral view; 26. Propodeal spiracle. = = = Abbreviations: cx3f metacoxal foramen; cx31a metacoxal lateral articulation; fu3p? possible metafurcal = pit; prf propodeal foramen. dorsally; anterior mesofurcal apodemes, extending from metapleural pit to notch mesofurcal bridge, and mesospinasternal just dorsal to metacoxal foramen (Fig. 1). apodeme absent. Metapleuron with lateral surface smooth Metanotum (Figs 11, 21, n3) a narrow except for striate sculpture just anterior of in transverse strip between anterior margin of metacoxal foramina and with small pit mesophragma and antecostal sulcus; with- middle (Fig. 1, pl3p) continuous with out developed metascutellum but poster- shallow paracoxal sulcus extending ven- iorly with slight median incurvation de- trally (Fig. 1, pes); internally, metapleural limited by minute posterior projections pit corresponding to low apodeme (Fig. 20, and laterally with depression having two pl3a) situated at dorsal end of low para- short setae (Figs 11, 16, 21). Metapleuron coxal carina extending across ventral part broad dorsally, accommodating hind wing of metapleuron (Fig. 22, pl3c). Meta- in incurvation in thickened dorsal margin pleuron with ventral surface smooth ex- (Figs 1, 16). Small, triangular projection cept for low longitudinal ridge of scaly posteriorly of hind wing separated from projections formed along metadiscrimenal bulk of metapleuron by narrow sulcus line (Fig. 23, di31), the discrimenal line (Figs 11, 16); projection with small bilobed, extending to circular depression (Fig. 24, possibly membraneous extensions poster- fu3p?) between metacoxal foramina. In- odorsally. Metapleuron fused with meso- ternally, circular depression corresponding pleuron for almost entire length laterally to slight swelling, and metadiscrimen and ventrally (Fig. 1); separated from (Fig. 22, di3) developed as low longitudi- propodeum posterodorsally by antecostal nal swelling reaching paracoxal carina; suture (Fig. 1, pas), but fused with propo- metafurca absent. Externally, metepisternal deum below propodeal spiracle, the depression well developed laterally of boundary indicated by shallow groove metadiscrimenal line, laterally delimited Volume 15, Number 2, 2006 299

27-29. First of P. Figs petiolar segment anomalum. 27. Dorsal view. 28. Ventral view of anterior part. 29. Anterior view.

low carina Meta- by longitudinal (Fig. 23). sophragma (Fig. 15; see above). Propodeal coxal foramen (Fig. 24, cx3f) constricted, foramen (Fig. 24, prf) pear-shaped in out- circular, surrounded by raised rim, sepa- line, encircled by raised rim and sur- rated from propodeal foramen by cuticle; rounded by smooth cuticle; without dis- median metacoxal articulation not devel- tinct articulation process with metasoma. oped, lateral articulation (Fig. 24, cx31a) Mesosoma-metasoma articulation. —Petiole concealed in lateral view. two-segmented. First petiolar segment Propodeum dorsally with scaly-reticu- (predominantly abdominal T2, see discus- late sculpture anterior to transverse, semi- sion) slender, cylindrical, and dorsally circular carina above propodeal foramen reticulate (Fig. 27); anteriorly with portion (Figs 24, 25); bare except for long seta articulating with propodeum constricted posterior to lateral end of antecostal suture relative to rest of petiole (Figs 27, 28), its (Fig. 1, pas); dorsally with minute subme- anterior margin swollen and circular in dian projection extending from anterior anterior view (Fig. 29); cuticle surrounding margin across antecostal suture to abut constricted portion smooth and posteriorly corresponding structure on metanotum delimited by transverse carina encircling (Fig. 25). Propodeal antecostal suture entire petiole (Figs 28, 29); anteroventrally with (Fig. 1, pas) a narrow transverse groove two sublateral setae on smooth separating metanotum and propodeum surface posterior to anterior swollen mar- with (Figs 1, 21), the suture terminating laterally gin (Figs 28, 29); median longitudinal carina in small oval depression concealing pro- posterior to anterior transverse but otherwise without podeal spiracle (Figs 25, 26). Internally, carina, conspicuous propodeum with apodemes for spiracular sculpture ventrally. muscles; metaphragma only observed lat- DISCUSSION erally above propodeal spiracle, but medi- an part might be indicated by presence of Debauche (1948) confused the lateral low transverse swelling anteriorly on me- parts of the pronotum in P. anomalum with 300 Journal of Hymenoptera Research

of the and the the propleuron, stating that 'les propleures gin pronotum anteriorly insertions alaires' An anterior atteignent, en arriere, les prepectus posteroventrally. reach the thoracic cannot [p. 31: 'the propleura wing ('mesothoracic') spiracle be observed in P. anomalum. articulations posteriorly']. Gibson (1986) externally clarified the configuration of the pronotum Gibson (1986) tentatively identified the and its association with surrounding scler- spiracle in Mymarommatidae as a structure ites. The condition in P. anomalum differs close to the posterolateral margin of the from Chalcidoidea in that the pronotum is pronotum observed in a slide preparation, but unable to confirm this with SEM rigidly associated with the mesopleuron, was Neither could the a large exposed prepectus is absent (there- investigations. we; only fore the posterodorsal corner of the prono- internal structure observed close to the turn and the fore wing base are not widely prepectus is a small pit just anterior to it it is this that separated), and there is no exposed spira- (Fig. 4), and likely pit Gibson cle situated dorsally between the pronotum (1986) observed in the slide preparation, and the mesoscutum. However, the pit is not associated with The above statements require some a tracheal trunk (unless the narrow tissue strands observed to issue from qualification. A more or less rigid associ- sometimes ation of the pronotum with the meso- the pit are remains of this) or with an pleuron and concurrent absence of an apodeme for a spiracle occlusor muscle, independent prepectus is widespread The plesiomorphic condition in Hymenop- within the and has probably tera is that the spiracle occlusor muscle evolved independently a number of times arises from the prepectus (Gibson 1985, (Gibson 1985, 1999). A number of taxa Vilhelmsen 2000a). Consequently, P. anom- (Cynipoidea, , most Procto- alum apparently lacks a functional anterior trupoidea) have the prepectus fused to the thoracic spiracle. posterior margin of the pronotum as the Mymarommatidae also lacks a fore wing posterolateral pronotal inflection, forming tegula according to Gibson (1986). Never- an articulation with the mesopleuron. This theless, the condition in P. anomalum is not condition is apparently not present in P. similar to that in most Chalcidoidea, in anomalum. Gibson (1986) could not estab- which there is broad contact between the lish the identity of the prepectus in mesoscutum and the exposed and dorsally Mymarommatidae with certainty, but sug- expanded prepectus, which separates the gested it to be homologous with the tegula from the pronotum (a notable narrow internal carina extending vertically exception to this is the , which close to the posterolateral margin of the have a slender, partially concealed prepec- pronotum. Our findings support this; the tus; Gibson & Huber 2000). Even though carina survived treatment with KOH and the tegula is absent, the ventral part of the hence must be cuticular, and a sclerite in parascutal carina (Fig. 16, psc), which this position would be the prepectus by anteriorly delimits the concavity accom- default. Furthermore, the carina is appar- modating the tegula, extends anteriorly of ently continuous with the pronotum, there the posterodorsal corner of the pronotum indication being no of displacement of the in P. anomalum (Fig. 16), like in virtually all structure due to shrinkage of membra- Hymenoptera except Chalcidoidea, where neous cuticle (Fig. 4). the carina terminates well posteriorly of To further corroborate the identity of the the corner. The configuration of the tegula prepectus, it would be desirable to estab- and parascutal carina relative to the poster- lish the position of the anterior thoracic odorsal corner of the pronotum is correlat- spiracle, which in most Hymenoptera is ed with the occurrence of a dorsally en- closely associated with the posterior mar- larged prepectus. In conclusion, the pres- Volume 15, Number 2, 2006 301 ence of a concealed prepectus not in their length from just posteriorly of the contact with the in mesoscutum Mymar- cervical prominences to the posterior mar- ommatidae is plesiomorphic relative to gin of the propleura level with the procoxal the Chalcidoidea, whereas fusion of the bases, and cover the anteroventral part of prepectus to the pronotum is apomorphic, the prosternum if the latter is developed, but paralleled in many other Apocrita Given that the is the (Gibson 1985, 1999). The absence of both putative sistergroup of Chalcidoidea, it tegulae and anterior thoracic spiracles are would be highly informative to know what possibly autapomorphic for Mymaromma- the configuration of the prosternum was tidae. prior to its fusion with the propleura. Another putative autapomorphy is the Because of the difficulties with establishing fusion of the medioventral margins of the the anterior boundary of the prosternum in propleura and the fusion of the proster- P. anomalum (see above) it is not possible to num to the propleura (Fig. 6). The bound- determine this condition, ary between the propleura and the pro- The fusion of the propleural and profur- sternum is difficult to establish. The pro- cal arms in P. anomalum (Fig. 9) is another furcal pit (Fig. 8, fulp) is situated dorsally putative mymarommatid autapomorphy of the procoxal bases (Fig. 8, cxl); however, and possibly correlated with the fusion of in most Hymenoptera the prosternum the prosternum and propleura. In other extends well ventrally of the pit before Hymenoptera, the lateral end of the pro- bending anteriorly above the propleura. furcal arm articulates with the posterodor- We consider it most likely that the smooth sal margin of the propleuron, an autapo- areas medioventral to the procoxal bases morphy of the Hymenoptera (Vilhelmsen are of the part prosternum (Fig. 8, stl) and 2000a). The propleural-profurcal articula- the fusion with the propleura took place tion perhaps forms a suspension for the ventrally of these, just dorsally of the fore legs. The fore legs articulate with the extended posteroventral margins of the propleura and abut the prosternum. The propleura. Anteriorly, the boundary be- propleural-profurcal articulation presum- tween the propleura and prosternum is ably allows the prosternum to become even less obvious. The base of the profurca slightly displaced relative to the propleura, does not extend far anteriorly, indicating thus softening the impact of the legs on the that the anterior part of the prosternum substrate and dampening it before it is was short prior to fusion. However, those transmitted to the head. The fusion of the Chalcidoidea with an exposed, anteriorly propleura and the prosternum in P. anom- extended prosternum also do not have the alum could have made the articulation profurcal base extending far anteriorly redundant because the prosternum and (Vilhelmsen unpubl.; Krogmann unpubl.), propleura are immovable relative to one so this feature is apparently of limited another. The profurcal bridge, a sclerotised value in deducing the anterior limits of the connection between the opposite dorsal prosternum. parts of the profurcal arms extending In virtually all Chalcidoidea the medio- above the ventral nerve cord, is also ventral margins of the propleura diverge missing from the majority of Chalcidoidea posterolaterally to broadly expose an ante- (Krogmann 2005). However, the occur- riorly extended, diamond-shaped proster- rence of this feature is highly variable the num (Krogmann 2005, Bucher 1948, fig. throughout Apocrita (Vilhelmsen 18). In most other Hymenoptera, except for 2000a, unpublished observations), the basalmost lineages (Xyeloidea and The exposed cervical prominence at the ; Vilhelmsen 2000a), the anterolateral corner of the propleuron medioventral margins abut for most of (Fig. 7, cep) is a feature shared by P. 302 Journal of Hymenoptera Research anomalum and at least some Chalcidoidea the axillar phragmata, the prophragmal (unpublished observations). In most other rods probably do not accommodate noto- Apocrita, the cervical prominence is some- trochanteral muscles because pronoto/ what retracted and below the anterodorsal phragmal-protrochanteral muscles are ab- corner of the propleuron (Vilhelmsen sent from the Hymenoptera that have been 2000a). The condition in P. anomalum is per- examined for musculature, although pro- haps correlated with miniaturization. The notal-procoxal muscles do occur (Vilhelm- small number of sensilla (3) at the cervical sen 2000a). The presence of axillar phrag- prominence probably also is a consequence mata remains the most convincing syna- of small size; other, larger Hymenoptera, pomorphy for Chalcidoidea and Mymar- have up to 20-30 sensilla in this position on ommatoidea (Gibson et al. 1999). The a prominent cervical swelling (Vilhelmsen homology of the structures in the two taxa 2000a, unpublished observations). is based on the topology regarding muscle In P. anomalum, the dorsal posterolateral attachments and position relative to the corner of the propleuron accommodates surrounding structures. However, the con- the posteroventral corner of the pronotum figuration of the axillar phragmata is quite in a distinct groove (Figs 2, 8). This is not different from that of Chalcidoidea. In observed in any Chalcidoidea, where the Chalcidoidea, the phragmata are flat, pronotum is much less integrated with broad structures that usually lie adjacent adjacent sclerites (see above). A similar to the internal part of the mesoscutum condition to that in P. anomalum is pos- (Krogmann 2005). In P. anomalum, the sessed by most Ichneumonoidea (unpub- phragmata are slender, cylindrical rods lished observations). No other features that are continuous medially through the appear to link Ichneumonoidea with My- transscutal carina (Fig. 13, tsc) and project marommatidae, so the condition in P. away from the mesoscutum into the lumen anomalum is probably independently de- of the mesosoma. The difference in struc- rived and a further putative autapomorphy ture and the possibility that in Mymarom- of Mymarommatidae. matidae the axillar phragmata are serial The prophragma of P. anomalum differs homologues of the prophragmal rods, from that of Chalcidoidea in not being might indicate that they were indepen- subdivided medially by a slit and not dently derived from those of Chalcidoidea. extending laterally. The presence of a slit Alternatively, the difference in structure of and the absence of a lateral extension of the the axillar phragmata between Mymarom- prophragma are the plesiomorphic condi- matidae and Chalcidoidea may reflect tions, being common within Hymenoptera. simple size differences in the tergo-tro- Furthermore, the prophragma has rods chanteral muscles or some other functional extending from it dorsally (Fig. 14), which requirement. have not been observed in any other The mesoscutum anteriorly of the trans- Hymenoptera. The function of these rods scutal articulation in P. anomalum is devoid could not be ascertained, but it is likely of any external and internal features are they apodemes that accommodate (mesoscutal sulcus, notauli, parapsides). prophragmo-pronotal muscles; indeed, Most apocritan wasps, including Chalci- what appears to be tonofibrillae were in doidea, have at least the notauli developed, some instances associated with the rods The posterior border of the axilla in P. The (Fig. 14, tf). position and configuration anomalum is delimited internally by the of the prophragmal rods might indicate posterior margin of the axillar phragmata that they are serial homologues of the and the lateral mesoscutellar carina, axillar phragmata (see below), which arise The transverse subdivision of the me- of the just anteriorly mesophragma. Unlike soscutellum into an anterior raised part Volume 15, Number 2, 2006 303 and a in Vilhelmsen posterior depressed part (further Xiphi/dria" (; emphasized by the difference in sculpture 2000b). in P. anomalum, Figs 10, 11) is also ob- The extensive fusion in the mesosoma served in some Chalcidoidea (Krogmann forming the mesopectus-metathorax-pro- 2005). In the Chalcidoidea, the posterior podeum complex is not unique for My- part is known as the frenum, and the marommatidae. Indeed, the fusion be- dividing line as the frenal line. If the tween the propodeum and posterodorsal presence of a frenum is a ground plan part of the metapleuron is an autapomor- feature of the Chalcidoidea, it might be an phy for the Apocrita (Vilhelmsen 2000b). additional putative synapomorphy of My- Within the Apocrita several taxa have the marommatidae and Chalcidoidea (Gibson meso- and metapleura fused to varying et al. 1999). The identity of the frenum in degrees, but only in Mymarommatidae Mymarommatoidea is corroborated by its and are they fused along relation to the mesoscutellar arms, the almost their entire common boundary latter arising just anteriorly of the external (Vilhelmsen unpublished). The integration frenal line (Fig. 10, frle) as in those Chalci- in Ceraphronoidea has gone even further, doidea having a frenum. Of the two because the dorsal and ventral parts of the branches of the axillar carina in P. anom- metathorax have become almost entirely alum, the anterior is probably homologous reduced. In Chalcidoidea, the meso- and with the single carina in most other metapleura closely abut, as in all Apocrita, Hymenoptera, because it terminates at the but usually are not fused. anterior end of the mesoscutellar arm, the The pronounced integration and general normal condition in taxa with only one scarcity of anatomical reference points in branch. the mesopectus-metathorax-propodeum The large, rectangular mesophragma complex means that the boundaries be- with a straight posterior margin extending tween the constituent parts can only be through most of the propodeum in P. loosely established. The following anatom- anomalum is a feature also observed in ical features are relevant for this exercise: un- the fore and the lateral some Mymaridae (e.g., Gonatocerus; hing wing bases, published observations). If Mymaridae are ends of the metathorax, the metapleural indeed basal within the Chalcidoidea pit, the propodeal antecosta and spiracle, (Gibson et al. 1999), this configuration of the meso- and metacoxal cavities and the mesophragma is putatively the ground lateral articulations, the meso- and meta- plan condition for Chalcidoidea. Many furcal pits, and the propodeal foramen. Chalcidoidea have the mesophragma ta- The lateral boundaries between the pered posteriorly and with an incurvation meso- and metapleuron and the propo- in 1. dorsal in the posterior margin (e.g., Bucher figs deum are indicated Fig. The 28-29), as do most Hymenoptera. The endpoint of the boundary between the muscle arising from the apodeme medially meso- and metapleuron can be identified on the anterior margin of the meso- at the bottom of the slit that separates the phragma (Fig. 21) is probably homologous fore- and hind wing bases (Fig. 16). The with the mesoscutellar-metanotal muscle posterior thoracic spiracle would have arising from the anterior margin of the provided a useful landmark, but is appar- metanotum in other Hymenoptera. Heraty ently absent, as in many apocritan wasps, et al. (1994) did not detail the condition in including Chalcidoidea (unpublished ob- Mymarommatidae or report a similar con- servations). Posteroventrally, the boundary figuration for any other taxa they exam- between the meso- and metapleuron prob- ined; the muscle also inserts on the ably lies just anteriorly of the metapleural anterior margin of the mesophragma paracoxal sulcus, which extends as a verti- 304 Journal of Hymenoptera Research

distinct features cal groove from the metapleural pit to the identifiable. No (e.g., notch above the mesocoxal foramen metascutellum) are evident (Figs 11, 21) the between other than the two small submedian (Fig. 1). Ventrally, boundary that across the ante- the two regions is more easily identified by projections extends the groove that extends between the costal suture to abut similar, anteriorly notches above the mesocoxal foramina projecting structures on the propodeum. are absent Chalcidoi- and the mesofureal pit (Figs 19, 23). The Such structures from but mesofurcal pit lies between the coxal dea (unpublished observations), they foramina, not anteriorly of them as in might be homologous with those observed almost all Chalcidoidea; the anterior posi- in many , Cynipoidea, and tion of the mesofurcal pit is a putative Ichneumonoidea. autapomorphy of the Chalcidoidea. A The function of the triangular projection transverse carina is present just anteriorly posteriorly of the hind wing and the of the mesocoxal foramina and pit as in bilobed extensions it accommodates Chalcidoidea, but also in some other (Figs 11, 16) is unknown; the extensions Apocrita. may possibly serve as sensillae monitoring The mesofurca in P. anomalum is com- the movements of the hind wing base. It is paratively simple, lacking a mesofurcal also unclear whether the extensions are bridge or any conspicuous apodemes membraneous; if they are, their bilobed is is artifact. (Fig. 20). The mesofurcal bridge also configuration possibly an absent from some Chalcidoidea, notably Internally, the metapleuron has a low putatively basal taxa such as Mymaridae internal swelling (Fig. 20, pl3a) that corre- and Rotoitidae. This was noted by Heraty sponds to the metapleural pit (Fig. 1, pl3p) et al. (1997), but because of its widespread and which is apparently the reduced occurrence in Apocrita they considered the metapleural apodeme. The low carina presence of a bridge to be the ground plan extending between the apodemes is the state for Chalcidoidea. The proximal part paracoxal carina, usually situated along of the mesofurcal arm is only slightly the anterior margin of the metapleura in P. In depressed anomalum. most Chalci- (Fig. 22, pl3c); this configuration of the doidea, a distinct dorsal concavity is de- apodeme and carina is widespread in veloped to various extents in this region, Apocrita. The metafurca is usually closely a putative autapomorphy for the super- associated with the paracoxal carina, the family. metafurcal arms arising medially at the The boundary between the metathorax anterior end of the metafurcal discrimen and propodeum is comparatively easy to (Vilhelmsen 2000b). In P. anomalum, the establish, at least dorsally, being marked only remnant of the metafurca is the by the antecostal suture (Fig. 1). The median depression between the metacoxal putative boundary then extends between foramina (Fig. 24, fu3p?), which probably the lateral of the end suture and the corresponds to the metafurcal pit, and metapleural pit towards the metacoxal a low longitudinal swelling (Fig. 22, di3) lateral articulation along a shallow groove that corresponds to the metafurcal discri- (Fig. 1). Posteriorly, the boundary extends men. The entire absence of metafurcal between the metacoxal and propodeal arms is a putative autapomorphy for the foramina (Fig. 24). Apparently, the lateral Mymarommatidae. In Chalcidoidea, there part of the metapleuron is expanded is substantial variation in the configuration dorsally and ventrally, but constricted near of the metafurcal arms and the number the metapleural pit (Fig. 1). and configuration of the metafurcal pits The metanotum is reduced in compari- (Krogmann 2005). Unfortunately, the ab- son to most other Hymenoptera, but is still sence of the former structure in Mymar- Volume 15, Number 2, 2006 305

ommatidae makes it difficult to interpret CONCLUDING REMARKS

the evolution of this structure within T , . . , , ,. ,..,•,, I he underlines the , present study highly • N_naiciQoiclea. , 1 , , r .1 » »

. . status of _. . the , , autapomorphic Mymaromma- The accommodation of the propodeal tidae However, we have only examined P. within the lateral ends of the spiracle anomalum> and the features we have stud . antecostal is suture 26) also observed , (Fig. ied need to be surveyed for a le in Tnchogramma (Krogmann 2005), but not of the family According to Gibson et al. outside Chalcidoidea. A more comprehen- the (1999)/ diversity of Mymarommatidae sive survey of the occurrence of this trait is not sufficiently reflected by the current within Chalcidoidea is to estab- required classification that has all extant species lish its possible phylogenetic significance, lumped in Palaeomymar, a situation that The of well presence developed apodemes will probably change once the family has of the indicates that it is internally spiracle been comprehensively revised. Given the the only functional spiracle in the meso- small size of all Mymarommatidae and soma of P. anomalum. that it possibly occupies a key phylogenetic The is slender, two-segmented petiole position within Apocrita, detailed morpho- observed in the extinct taxon only Serphi- logical studies of additional species are tidae apart from in Mymarommatidae; recommended to test our inferences, based it on this, has been hypothesized Many of the putative autapomorphies of that they form a monophylum (cf. Gibson Mymarommatidae are reductional and 1986). The first petiolar segment of P. probably correlated with miniaturization anomalum essentially is a narrow, sclero- (e.g., the absence of the anterior thoracic tised tube with little indication of bound- spiracle and the metafurca). Furthermore, aries between individual sclerites. This being morphologically highly derived and isolated condition occurs in many Apocrita, in- confounds attempts to homologize features with those in other eluding some Chalcidoidea, and is difficult Apocrita (e.g., tne extensive fusion in the the to interpret. The ventral longitudinal carina propectus, difficulties in the and extending along most of the first petiolar establishing identity of the The sis- segment in P. anomalum is perhaps formed configuration prepectus). to the Chalcidoidea by the fusion of the lateral margins of the tergroup relationship a ( also taxon) seems second abdominal tergum (T2), indicating highly autapomorphic the best at that T2 is expanded ventrally and the supported hypothesis present, but ma rove difficult to corroborate sternum of this segment is reduced or y P further from evidence absent (Gibson in litt.). However, the setae morphological alone ' Mymarommatidae has not been situated anteroventrally just posterior to represented in any phylogenetic analyses the anterior margin of the first petiolar of Hymenoptera including molecular data, segment (Figs 28, 29) are probably homol- not even the most recent comprehensive ogous with the sensillar patches situated °nes Cam b et aL 2000 » ' Dowto and on S2 in that have ( P sublaterally Apocrita A f Austin 2001). Sequence data might confirm this sclerite separate from T2. This indi- °r chaUen C°nCeP tS ° f the P^" cates that at least the anteriormost part of ^I™"™* u • , of these ani- C t r> 1 -ri , logenetic affinity intriguing S2 is in P. & b present anomalum. The continu- ,

. , , mals. ous transverse carina anteriorly on the first petiolar segment (Fig. 29) is also observed ACKNOWLEDGMENTS in many Proctotrupoidea, Cynipoidea, Pla- Jens-Willu-lm Janzen, Hamburg, Germany, and tygastroidea, Ceraphronoidea as well as in , . . . Kiel] Arne Naturhistoriska >-,, , , , ,. , , , Johansson, Riksmuseet, some Chalcidoidea observa- (unpublished Stockholm . , Sweden/ kindly provided access to mate ti°ns )- rial employed in this study. John Huber, CNCI, 306 Journal of Hymenoptera Research

G. A. M. and B. 1999. Ottawa, Canada provided useful comments on earlier Gibson, P., J. Heraty, ]. Woolley. and classification of Chalcidoidea versions of the paper. Gary Gibson, CNCI, Ottawa, Phylogenetics — and an anonymous referee commented extensively on and Mymarommatoidea a review of current the submitted version. concepts (Hymenoptera, Apocrita). Zoologica Scripta 28: 87-124. P. T. 2000. of the LITERATURE CITED Gibson, G. A. and J. Huber. Review family Rotoitidae (Hymenoptera: Chalcidoidea), Monodontomerus Bucher, G. E. 1948. The anatomy of with description of a new genus from Chile. chalcid. Canadi- dentipes Boh., an entomophagous Journal of Natural History 34: 2293-2314. 230-281. an Journal of Research Section D 26: Grimaldi, D. and M. S. Engel. 2005. Evolution of the

M. K. Chan, J. insects. York. Campbell, B., J. Heraty, J.-Y. Rasplus, Cambridge University Press, New and C. Babcock. 2000. Molecu- B. D. C. 1994. Steffen-Campbell, Heraty, J. M., J. Woolley, and Darling. lar systematics of the Chalcidoidea using 28S-D2 Phylogenetic implications of the mesofurca and rDNA. 59-73 in: Austin, A. D. and M. in Pp. — mesopostnotum Hymenoptera. Journal of Hy- Dowton, eds. Hymenoptera evolution, biodiversity menoptera Research 3: 241-277. and control. CSIRO B. D. 1997. biological publishing, Colling- Heraty, J. M., J. Woolley, and C. Darling. wood. Phylogenetic implications of the mesofurca in Debauche, H. R. 1948. Etude sur les Mymarommidae Chalcidoidea (Hymenoptera), with emphasis on et les Mymaridae de la Belgique (Hymenoptera, . Systematic Entomology 22: 45-65. Chalcidoidea). Memoires du Musee Royal d'Histoire Krogmann, L. 2005. Molekulargenetische und mor- Naturelle Belgique 108: 1-248. phologische Untersuchungen zur systematischen Dowton, M. and A. D. Austin. 2001. Simultaneous Stellung der Pteromalidae innerhalb der Chalci- of COl and in the analysis 16S, 28S, — morphology doidea (Hymenoptera: Apocrita). Ph.D.-thesis, Hymenoptera: Apocrita evolutionary transi- University of Hamburg, 204 pp. http://www. tions among parasitic wasps. Biological Journal of sub.uni-hamburg.de/opus/volltexte/2005/ the Linnean Society 74: 87-111. 2652/ Gibson, G. A. P. 1985. Some pro- and mesothoracic Ronquist, F., A. P. Rasnitsyn, A. Roy, K. Eriksson, and structures important for phylogenetic analysis of M. Lindgren. 1999. Phylogeny of Hymenoptera: A Hymenoptera, with a review of terms used for cladistic reanalysis of Rasnitsyn's (1988) data. the structures. The Canadian Entomologist 117: Zoologica Scripta 28: 13-50. 1395-1443. Vilhelmsen, L. 2000a. Cervical and prothoracic ske- Gibson, G. A. P. 1986. Evidence for monophyly and leto-musculature in the basal Hymenoptera (In- relationships of Chalcidoidea, Mymaridae, and secta): comparative anatomy and phylogenetic Mymarommatidae (Hymenoptera: Terebrantes). implications. Zoologischer Anzeiger 239: 105-138. The Canadian Entomologist 118: 205-240. Vilhelmsen, L. 2000b. Before the wasp-waist: compar- Gibson, G. A. P. 1999. Sister-group relationships of the ative anatomy and phylogenetic implications of Platygastroidea and Chalcidoidea (Hymenop- the skeleto-musculature of the thoraco-abdominal tera) —an alternate hypothesis to Rasnitsyn boundary region in basal Hymenoptera (Insecta). (1988). Zoologica Scripta 28: 125-138. Zoomorphology 119: 185-221. J. HYM. RES. Vol. 15(2), 2006, pp. 307-316

Nesting Biology of a Tropical Myrmicine Ant, Myrmicaria arachnoides (Formicidae), in West Java, Indonesia

Bakhtiar Effendi Yahya and Seiki Yamane

of Earth (BEY, SKY) Department and Environmental Sciences, Faculty of Science, Kagoshima University, Korimoto-1, Kagoshima, 890-0065 Japan, BEY email: [email protected], SKY email: [email protected] (BEY) Institute for Tropical Biology and Conservation, Universiti Malaysia Sabah, Kota Kinabalu, Sabah, Malaysia

— Abstract. Nesting biology of a myrmicine ant, Myrmicaria arachnoides, was studied on Java island, Indonesia. The colonies were polydomous and polygynous. Nests of a colony were located close to each other on different leaves of a tree. The number of dealated queens was positively correlated with the numbers of nests, adult workers and immatures, but not with the numbers of males and new queens present. Adaptive aspects of nesting site, polygyny and polydomy are discussed.

The genus Myrmicaria is widely distrib- and polygynous colonies but on vegetation uted in the Old World tropics, i.e., South- (Karawajew 1935, this study, Bakhtiar and east Asia, South Asia and tropical Africa. Yamane unpubl.). This may allow them to Emery (1922) sorted Myrmicaria species develop relatively large colonies similar to into two principal groups based on mor- Polyrhachis species nesting in similar situa- phology, i.e., the M. brunnea group and M. tions (R. Kohout pers. com.). arachnoides group, and briefly mentioned Research on colony growth in eusocial that species of the former construct huge insects has focused on the relationship nests underground while those of the latter between colony size and productivity. At construct carton nests on trees. Our ob- each growth period, colonies make invest- servations have confirmed this for most of ment decisions about whether to produce the Southeast Asian forms (also see Kar- more workers and grow larger, or rather to avaiev 1935). However, there are few invest that energy in reproductive output detailed studies on the nesting habits of (Billick 2001). Because colony size is often Myrmicaria. In Cameroon, Africa, one the most important factor determining species, M. opaciventris Emery, belonging reproductive output (Odum and Pontin to the M. brunnea group, has been in- 1961, Michener 1964, Fowler 1986, Tschin- tensively studied for its ecology/bionomics kel 1993, Savolainen et al. 1996), maximiz- including nesting habits (Kenne and De- ing the long-term size of the colony is an jean 1999, Kenne et al. 2000, 2001). This important component of colony fitness species has polydomous and polygynous (Oster and Wilson 1978). Colony size is colonies in the soil. Interconnected nests known to be related with queen number are built and trenches and tunnels are and also to affect caste/sex composition in constructed as underground trails connect- ant nests. In this respect nesting behaviour ing nests. Species of the M. arachnoides of the M. arachnoides group is oi special group also often construct polydomous interest. 308 Journal of Hymenoptera Research

Little has been studied on the colony RESULTS structure of the Southeast Asian Myrmi- Nesting site and structure. —In Salak and arachnoides F. Smith see Kara- caria (but Salak-Halimun Corridor, colonies were vaiev 1935). our study on the During found in plantation areas with sparse trees of Oriental we ob- taxonomy Myrmicaria and bushes, while in the botanical garden tained relatively good samples of M. they were found in a forested area with arachnoides colonies on Java island, Indo- relatively high trees. Nests were located on nesia. We report here the nesting sites, the underside of leaves of various plant colony size, and reproductive output of species at 0.5-1.5 m above the ground in these colonies. the former two sites (Fig 1 a,b), but in the MATERIALS AND METHODS botanical garden they were positioned higher at around 3-4 m above the ground. The of is still taxonomy Myrmicaria In the case of polydomous colonies com- unresolved, particularly the status of in- ponent nests were generally constructed forms of some species. Myrmi- fraspecific separately on different leaves of the same caria arachnoides was originally described plant, but in one case, two nests were built from Borneo and consists of several 'sub- narrowly connected on one and the same species' (Bolton 1995), of which at least leaf (BOG25-2 & -4) (Appendix 1). some would be The form good species. Nests were made of carton-like material studied here is in coloration most similar to (probably chewed plant tissues), flattened 'M. arachnoides arachnoides'. domes in shape, and greyish brown in were collected from three dis- Samples colour. Various sizes of nests were built on turbed sites in West Java in September - the underside of various sizes of leaves 2004: foot of G. Salak Site 1 (6 39' S, 106 (Appendix 1). Within the nest, there were 46' E, 560 m) (BOG3, 10 & 18), Salak- - numerous chambers for adults and imma- Halimun Corridor Site 2 ( 6 45'S, tures; some of these would be used as 106 37'E, 710 m) (BOG24, 25 & 26), and - galleries for the movements of workers Bogor Botanical Garden Site 3 (6 36'S, (Fig. 1 b,c). 106 48'E, 220 m) (BOG 38). The distance The number of nests per colony varied between each plant where these colonies from 1 to as many as 12; nests of a colony were found was approx. 5-50 m in Site 1, were built close to each other on and 5-10 m in Site 2. These habitats have usually one plant (approx. 15-30 cm apart). been infringed by plantation or agricultur- —Dealated al activities or surrounded by residential Colony composition. queens (Q) were considered to be foundresses or those areas. Nests constructed on a same plant that have later called were thought to constitute a single colony joined (simply 'queens' and were found in all colonies and as use same foraging trails. Nests hereafter) all nests in BOG18-3 and BOG25-11. of each colony from different plants were except The two nests had immatures, collected intact and put into plastic bags queenless the of them from separately. In total, seven colonies of suggesting transportation different sizes were collected. other nests where a laying queen(s) existed. Nests were measured for their maximum The number of queens differed from colony to and nest to nest Number of width and length, and then dissected colony (Fig. 2). is correlated with the number carefully. Workers, reproductives (dealat- queens highly of nests The number ed queens, young winged queens and (Fig. 3a). highest mean of 1 1 in a 3- males), immatures (eggs, larvae and pu- queens per nest, .67, was found ) were counted and preserved in 80% nest colony (BOG18) (range: 0-22) (Fig. 2). alcohol. Pupae were sorted into sexes and However, BOG38 which also had 3 nests had castes as much as possible. the smallest number of queens in each of the Volume 15, Number 2, 2006 309

Fig. la-d. a, Nests constructed on underside of leaves, b, Nest on the underside a leaf, c, Structure of a nest; top of envelop removed to show the interior, d, Compartments within nest; outer walls removed. nests (mean=1.33 with a range of 1 to 2). constitute relatively lower ratios, while BOG24 (1-nest colony) and BOG3 (2-nest other life forms (stages) were much fewer. colony) contained relatively small numbers Most of the nests had queen(s), workers of (1-3) queens per nest. In larger colonies, and immatures, and there was no striking BOG10 (4-nest colony) and BOG25 (1 2-nest specialization for a certain nest(s) in a sin- colony), the mean number of queens per nest gle colony with respect to worker /repro- was larger and relatively stable (4.00 ± 2.16 ductive production. However, BOG18-3, & 4.00 ± 3.34 respectively). BOG38-1, BOG26-3 and BOG 10-2 con- Within each colony, workers represent tained higher percentages of workers com- the greatest number among inhabitants pared with other nests composing these (approx. 40%) except in BOG 25 and colonies (Appendix 1). BOG26 (approx. 30.3%), followed by eggs In the single 1-nest colony (BOG24), all (approx. 20-40%) and larvae (approx. 10- the adults were workers (approx. 40% of all 30%). Worker pupae and male adults the inhabitants) except for three queens 310 Journal of Hymenoptera Research

BOG10 Nests/colony

Fig. 2. Number of queens in each nest and colony. that were possibly egg layers. There were all colonies), and winged adult queens in numerous eggs (approx. 45%), while larvae colony BOG26 (Appendix 1). As in smaller and pupae were relatively few, being colonies mentioned earlier, workers again approx. 10% and 5%, respectively. Further- constitute between 30 and 40% of all more, all the pupae were workers, all this inhabitants. Following the workers, eggs that the suggesting colony was in its and larvae also occupied large proportions ergonomic stage (reproduction and dis- except in BOG18-3, just as in the smaller persal are not its immediate concern) (cf. colonies. In BOG18-3, more adult males Oster and Wilson 1978). were observed. Worker pupae were ob- In the single 2-nest colony (BOG3) all life served to constitute approximately 5% in stages were present except the pupae of the colonies BOG38 and BOG26. new queens (Appendix 1). Males had In the 12-nest colony (BOG25; Appen- started to be produced, representing ap- dix 1), winged queens were seen only in proximately 5% of the total adults for the two nests in small numbers, while males colony (Appendix 1). Male pupae also were distributed more evenly; the pupae of existed in BOG3-1 (Appendix 1). Adult new queens were absent in this colony. For and workers, eggs larvae each had similar the whole colony, workers, eggs and larvae percentages for the whole colony and also had approximately the same numbers for each component nest. (around 30%) while approximately 10% In the three 3-nest and 4-nest colonies were worker pupae. 26 and all life — (BOG18, 38), stages were Reproductive output. There were strong the present except pupae of new queens (in relationships between the number of Volume 15, Number 2, 2006 311

12000 - 312 Journal of Hymenoptera Research

Table 1. Egg:worker (E:W) ratio. Volume 15, Number 2, 2006 313 or trees with a group of workers. However, reproductives. Furthermore we cannot we do not have any evidence supporting know how many reproductives have been this view. produced in a colony at the time of — Coloni/ size and reproductive production collection because dispersed individuals Smaller colonies (eg. BOG24) obviously do not leave any indication of their pre- had lower numbers of inhabitants (Appen- vious presence in the colony unlike the dix 1). These may have been at stages just case for social vespids where reproductive after the foundation and colony (cf. Sudd production can be measured rather pre- Data for other colonies show Franks 1987). cisely by observing pupal remnants. that as a colony grows in terms of the number of nests per colony, the number of ACKNOWLEDGMENTS colony members increases dramatically (Appendix 1), as suggested by Oster and We would like to express our appreciation to the Indonesian Institute of Sciences for us Wilson (1978). In this subsequent stage, (LIPI) granting the permission to carry out this survey in West Java. profits are mainly re-invested in workers We also would like to express our great appreciation and infrastructure such as the nest (Sudd to our research counterparts in Indonesia, particularly and Franks 1987). Mr. Rosichon Uhaidillah and his staff from Museum At some critical size, a colony begins to Zoologicum Bogoriense (Cibinong); and also Mr. Akhmad Rizali for us an immense amount of produce sexual offspring in order to realize giving the Our sincere thanks are its inclusive fitness. In our case even the 2- help during survey. extended to Dr. Rudy Kohout of the Queensland nest colony had already produced some Museum for giving us valuable information on males. But there was no relation between Polyrhachis nesting habits. the colony size and number of males present. For example, in the 12-nest colony LITERATURE CITED (BOG25), the number of males was rela- tively small as compared with BOG38 and Aron, S., L. Keller, and L. Passera. 2001. Role of BOG10 with fewer nests. Several factors resource availability on sex, caste and reproduc- tive allocation ratios in the Argentine ant Eine- may be responsible for this. Some males pithema humile. journal of Animal Ecology 70: have left the nest when it might already 831-839. was collected. Furthermore we do not Billick, I. 2001. Density dependence and colony males are know whether produced growth in the ant species Formica neorufibarbis. throughout the year or during restricted journal of Animal Ecology 70: 895-905. seasons. Bolton, B. 1995. A new general catalogue of the ants of the zvorld. Harvard and The increase in number University Press, Cambridge queen may London. increase the size of the and finally colony, Djieto-Lordon, C. and A. Dejean. 1999. Tropical the number of reproductives. Although in arboreal ant mosaics: innate attraction and im- this study queen number positively affect- printing determine nest site selection in dominant 219-225. ed the number of workers and immatures, ants. Behavioral Ecologi/ 45: Elias, M., R. Rogengren, and L. Sundstrom. 2005. the relationship between queen number Seasonal polydomy and unicoloniality in a polyg- and new queen (also male) number was ynous population of the red wood ant Formica not This shows that other factors positive. truncorum. Behavioral Ecology Sociobiology 57: may have operated in determining the 339-349. development of queens and males as Emery, C. 1922. Hymenoptera, Fam. Formicidae, subfam. 95-206 in: reported for many other genera (for the . Pp. Wytsman, P., Genera insectorum. Fasc. 174B. Argentine ant, see Aron et al. 2001). In M. ants Hansen, L. D. and J. H. Klotz. 2005. Carpenter of araclinoides maintaining a large worker the United States and Canada. Comstock Publishing force on one tree itself be may important Associates, Ithaca and London. under certain conditions (e.g., presence of Holldobler, B. and E. O. Wilson. 1990. The ants. competitors) at the cost of producing more Harvard University Press, Cambridge. 314 Journal of Hymenoptera Research

1993. of Karavaiev, W. 1935. Neue Ameisen aus dem Indo- Majer, J. Comparison the arboreal ant mosaic Australischen Gebiet, nebst Revision einiger For- in Ghana, brasil, papua New Guinea and Aus- men. Treubia 15: 57-117. tralia - its structure and influence on

life: the of in: I. Keller, L. 1995. Social paradox multiple- diversity. Pp. 115-141 LaSalle, J. and D. queen colonies. Trends in Ecologx/ and Evolution 10: Gauld eds. Hymenoptera and Biodiversity. CAB 353-360. International, Wallingford. Kenne, M. and A. Dejean. 1999. Spatial distribution, Oster, G. F. and E. O. Wilson. 1978. Caste and ecology in size and density of nests of Myrmicaria opaciven- the social insects. Princeton University Press, tris Emery (Formicidae, Myrmicinae). Insectes Princeton. K. R. Sociaux 46: 179-185. Savolainen, R., Vepsalainen, and J. Deslippe. L. Kenne, M., B. Schatz, R. Feneron, and J. Durrand. 1996. Reproductive strategy of the slave ant 2000. Changes in worker polymorphism in Formica podzolica relative to raiding efficiency of Myrmicaria opaciventris Emery (Formicidae, Myr- enslaver species. Insectes Sociaux 43: 201-210.

micinae). Insectes Sociaux 47: 50-55. Sudd, J. H. and N. R. Franks. 1987. The behavioral Kenne, M., B. Schatz, R. Feneron, and A. Dejean. 2001. ecology of ants. Chapman and Hall, New York. Hunting efficacy of workers from incipient Tschinkel, W. R. 1993. Sociomerry and sociogenesis of colonies in the myrmicine ant Myrmicaria opaci- colonies of the fire ants Solenopsis invicta during ventris (Formicidae: Myrmicinae). Sociobiology 37: one annual cycle. Ecological Monographs 63: 121-134. 425-457. Volume 15, Number 2, 2006 315

Appendix 1. Composition of life forms (stages) in colonies and nests. 316 Journal of Hymenoptera Research

Appendix 1. Continued.

Nest J. HYM. RES. Vol. 15(2), 2006, pp. 317-347

The Species of Sternanlopius Fischer (Hymenoptera: Braconidae, Opiinae) and the Braconid Sternaulus

Robert A. Wharton

Department of Entomology, Texas A&M University, College Station, TX 77843, USA email: [email protected]

Abstract. — The opiine braconid genus Sternaulopius Fischer is recognized as valid, redefined, and one new species, Sternaulopius duplicatus, from Madagascar, is described. The type species, Sternaulopius bisternaulicus Fischer, is fully redescribed, with new host and distribution records. The only other previously included species, Sternaulopius beieri Fischer, is placed as a junior subjective is also redescribed and the synonym of Opius bajulus Haliday, new synonym. Opius bajulus genus group name Biophthora Foerster is resurrected for this species. Opius rossicus Szepligeti, is transferred to the genus Biophthora (new combination) and Opius castaneus Granger, type species of Frekius Fischer, is transferred to Utetes Foerster (new combination). Thus, Frekius is a junior is a valid and subjective synonym of Utetes, new synonym, but retained as subgenus. Biophthora Sternaulopius are compared to Xynobius Foerster (where Opius bajulus has frequently been placed), and Xynobius is redefined and treated as a subgenus of Eurytenes Foerster. Stigmatopoea Fischer is also recognized as a valid subgenus of Eurytenes. Characters used to define these genus-group taxa are discussed in detail, with emphasis on venation, placement of metasomal spiracles, and the ventral sculptural details of the body. Use of the term sternaulus for a longitudinal groove on mesopleuron in Ichneumonoidea is reviewed, and it is shown that the sternaulus in cryptine and mesochorine Ichneumonidae is not homologous to the precoxal sulcus in Braconidae based on dissections of associated thoracic musculature. A true sternaulus, defined internally as the ridge in Braconidae. supporting the origin of the mesopleural-basalare muscle, is rarely present

is still Wharton The genus Sternaulopius Fischer, 1965 Opiinae unsettled. (1988) was described to accomodate a single observed that, except for the double ster- described more species from the Democratic Republic of naulus, the two species within the Congo. Subsequently, Fischer (1968) closely resembled other species

s. /. than did each described a second species from Germany. Opius Wesmael they limited No additional species have been described, other. On this relatively evidence, both to and only five specimens have been re- Wharton (1988) transferred species corded (Fischer 1965, 1972, Quicke et al. Opius, thus treating Sternaulopius as a syn- while at the same time 1997). The two species that have been onym of Opius

s. I. was not demonstra- included in Sternaulopius (Figs 1-20) have noting that Opius two distinct grooves (sternauli) on each bly monophyletic. Quicke et al. (1997) side of the mesopleuron but in nearly all recognized six genera that were formerly Fischer as other braconid wasps only a single groove treated by (1972, 1977, 1987) of and that the is present, or the groove is completely lost subgenera Opius, reported of a of Sternau- (exceptions include Trigastrotheca laikipien- apparatus specimen beieri Fischer resembled that of sis Quicke and some species of Pambolus lopius some, all the that included Haliday). These two species of Sternnulo- but not of species they of these Foerster. pius are thus distinctive, though their in one genera, Xynobius on Ster- placement in the classification of the Nothing else has been published Research 318 Journal of Hymenoptera

face. lateral view of head. dorsal Figs 1-4. Sternaulopius bisternaulicus reared from Strombosia fruits: 1, 2, 3, and labrum. view of head. 4, right mandible, with oblique views of clypeus

the of while naulopius, undoubtedly due to the paucity mesopleuron Braconidae, the of individuals available for study. Van other braconid specialists have used van Achterberg (2004) recently re-characterized name precoxal suture/sulcus (notably and Hud- Xynobius, but made no mention of Sternau- Achterberg 1975, 1993, Shaw or fur- lopius. dleston 1991) simply longitudinal The term sternaulus has long been used row of mesopleuron (Muesebeck 1970). by students of Ichneumonidae to describe The term precoxal sulcus has often been a longitudinal groove on the lower part of applied more widely in the Hymenoptera, the mesopleuron extending posteriorly though not necessary consistently (com- from the ventral-lateral region of the pare Richards 1956 with Bohart and al. treat epicnemial (= prepectal) carina towards Menke 1976). Gibson et (1998) the the coxifer (or pleural coxal process) the sternaulus as a synonym of (Viereck 1916, Richards 1956, Townes transepisternal line/sulcus (though their Fischer 1969). Several workers (e.g. Granger focus is on Chalcidoidea). (1972) 1949, Fischer 1958, Mason 1964, Marsh noted that it was probably incorrect to 1971, Wharton et al. 1997) have also equate the sternaulus with the precoxal applied this term to a similar groove on sulcus in Braconidae, and van Achterberg Volume 15, Number 2, 2006 319

fruits: left side of left arrow = Figs 5-10. Stemaulopius bisternaulicus reared from Strombosia 5, mesosoma, = arrow = sternaulus. 6, ventral-lateral view propleural flange, upper right arrow precoxal sulcus, lower right = = dorsal view of mesosoma. of mesopleuron, upper arrow precoxal sulcus, lower arrow sternaulus. 7, 8, = = left dorsal dorsal view of pronotum, upper arrow oblique groove on propleuron, lower arrow notaulus. 9, view of metanotum and propodeum. 10, dorsal view of petiole.

(1993) illustrated the two as separate be observed in some Opiinae (undoubt- structures for braconids. In a newsletter, edly referring to Stemaulopius), and Pam- "the sternaulus is van Achterberg (1994) stated more specif- bolinae where clearly the bent and situated ically that the difference between downwards anteriorly sternaulus and the precoxal sulcus can more ventrad." 320 Journal of Hymenoptera Research

Figs 11-12. Sternaulopius bisternaulicus reared from Strombosia fruits: 11, lateral view of metasoma. 12, same showing dorsope (left arrow) and spiracle on second tergum (right arrow).

There are no published host records for MATERIALS AND METHODS Sternaulopius (with the exception of in- of formation I have recently included on Specimens Sternaulopius (sensu Fischer were either reared from a website: hymenoptera.tamu.edu/parof- 1972, 1987) fruits infested with or fit). However, examination of material Tephritidae bor- reared from fruit in both Cameroon and rowed from museums. Rearing methods, and methods of identification of Kenya (Steck et al. 1983, Copeland et al. localities, flies and are described in Steck et al. 2002), indicates that tephritid fruit flies plants (Diptera) are the hosts of at least some of (1983) and Copeland et al. (2002). Material the species of Sternaulopius. This reared was borrowed from the Koninklijk Muse- material forms the basis for the present um voor Midden-Afrika, Tervuren, Bel- treatment, including a preliminary exami- gium (MRAC), Museum fur Naturkunde nation of the nature of the sternaulus in der Humboldt-Universitaet, Berlin, Ger- Ichneumonoidea. many (ZMHB), Hungarian Natural History Volume 15, Number 2, 2006 321

Fig. 13. Sternaulopius bisternaulicus reared from Strombosia fruits, ovipositor and ovipositor sheaths.

Museum, Budapest (HNHM), Museum Texas A&M University were dissected to National d'Histoire Naturelle, Paris examine the internal musculature of the (MNHN), and the U. S. National Museum mesothorax. All dissected material was of Natural History (USNM). Additional initially stored in 70-80% ethanol, then specimens examined are in the Naturhis- either air dried following transfer through torisches Museum Wien, Austria (NHMVV) 95% ethanol and 99% amyl acetate or and the National Museum of Ireland, dissected while still in alcohol. Specimens Dublin (NMID). of Apis mellifera L. and several other Fly puparia were individually isolated in Apocrita collected in central Texas with Kenya by R. Copeland from one heavily- Malaise traps were also dissected to ascer- tain the of thoracic muscu- infested sample of Strombosia scheffleri Engl. pattern general (Olacaceae) fruit, collected in Kakamega lature, using the works of Daly (1964) and Forest, Kenya on 30th April 2000. Associ- Gibson (1985) for orientation and identifi- ation of wasps with hosts was made by the cation of major muscles. All identifica- author, based on characteristics of the tions were made by the author; voucher puparia. All other host records listed below specimens are deposited as Texas A&M are based on wasps that emerged from University voucher number 656 in the Collection tephritid puparia that were not individu- Texas A&M University Insect ally isolated. (TAMU). Numerous specimens of Braconidae and Measurements are as in Wharton (1977, Ichneumonidae from material housed at 1986); terminology for descriptions of 322 Journal of Hymenoptera Research

V

14

15

Figs 14-16. 14, 15, Fore and hind wings of Sternaulopius bisternaulicus. 16, Hind wing of Biophthora bajulus, arrow = 2-1A.

venation and external features of the body precoxal sulcus (following van Achterberg follows Sharkey and Wharton (1997), ex- 1993). Details of the mesothorax associated cept that in the present paper a distinction with the musculature are provided in the is made between the sternaulus and the results and discussion where additional Volume 15, Number 2, 2006 323

habitus. face. Figs 17-20. Sternaulopius beieri holotype female (= B. bajulus): 17, lateral 18, 19, dorsal-posterior view of posterior part of mesosoma showing mesonotal midpit and sculpture of scutellum and propodeum. 20, = = lateral oblique view of mesosoma, upper arrow precoxal sulcus, lower arrow sternaulus.

within terms, when used, are specifically defined. also apply to many other taxa both Thoracic muscles relevant to discussions of and outside apocritan Hymenoptera, as the sternaulus are noted in Figs 21-32. Of evidenced by the detailed studies of Daly particular focus are various axillary mus- (1964) and Gibson (1985, 1986a, b, 1993). cles and especially the mesopleural-basa- The major muscles discussed below for the lare muscle (sensu Gibson 1985). The latter Ichneumonoidea were roughly similar in originates on the wall of the mesopleuron their points of origin and insertion in the and inserts on the basalare sclerite near the few aculeates and non-aculeate apocritans at least in all the base of the wing, that were dissected (namely A. mellifera, Aculeata Ichneumonoidea (22 genera) and Khopalosoma nearcticum Brues, and unde- (4 genera) examined (Table 1). termined species of Latreille and Polities Latreille). Many of the details RESULTS AND DISCUSSION of thoracic musculature found in such excellent works as those of Gibson (1985, Musculature and associated 1993) have been omitted here since the external features major interest is to provide a better char- acterization of the sternaulus. The observations made here refer specif- the the sternaulus and ically to the Ichneumonoidea (Ichneumo- On mesopleuron, or other nidae + Braconidae). However, they will various pits, depressions, changes 324 Journal of Hymenoptera Research Volume 15, Number 2, 2006 325 in elevation are, to a greater or lesser recent taxonomic treatments), rather than extent, the external representation of the parapsidal furrows. attachment of the thoracic muscles. In- Immediately laterad the dorsoventral in ternally, the mesothorax, at least ichneu- indirect flight muscle is the smaller monoids and aculeates, is dominated by (though still fairly large) mesopleural- the massive dorsolongitudinal (Fig. 21) basalare muscle (Figs 23, 25-29, 31). This and dorsoventral (Fig. 22) indirect flight is a distinctive band of muscle fibers muscles. The former originates largely readily identified by the dorsal, sclerotized from the median mesoscutal lobe (and cap that is strongly constricted to form partly from the phragma along the anterior a tendon-like attachment connected to the margin of the lobe) and extends posteriorly basalare (Figs 27, 29, 31). In Ichneumonoi- through the middle of the mesothorax. On dea, the mesopleural-basalare originates either side of this median muscle mass is ventrally in the trough of the mesothorax a broad band of dorsoventral muscle fibers just laterad the dorsoventral muscle or it the lateral lobes arise the vertical (Fig. 22), which insert on may somewhat higher on of the mesoscutum and originate ventrally wall of the mesepisternum. The subtegular on the mesothorax. The floor of the (= subalar) ridge, immediately ventrad the mesothorax, where the dorsoventral mus- anterior subalar depression of van Achter- is the external cles originate, is a trough interrupted berg (1988), representation medially by a phragma representing the of the internal pocket in which resides the midventral invagination of the mesoster- sclerotized cap of the mesopleural-basalare num (partially shown in Figs 25 and 29). muscle. The epicnemial carina, when pres- The lateral and ventral sides of the trough ent, delineates the anterior border of this are formed on each side of the body by the muscle mass ventrally, and dorsally it is relatively sharp vertical to horizontal tran- delineated by the pronotal-mesopleural sition in the mesepisternal wall ventrally as suture. it curves towards the ventral midline. The mesopleural-basalare muscle origi- Notauli are perhaps best defined as nates on the sternaulus in those members external grooves on the mesoscutum in- of the Ichneumonidae that possess a ster- but not dicating the position of the internal ridge naulus (Figs 25, 28), does originate has or phragma that separates the attachment on the structure that frequently (e.g. been called the point of the dorsolongitudinal muscle mass Sharkey and Wharton 1997) In from the attachment points of the right and sternaulus in Braconidae (Figs 29-30). left dorsoventral muscle bundles (see the the ichneumonid taxa Cryptini and Meso- excellent discussion in Gibson (1985) for chorinae, both of which have a well-de- on the terminology of mesonotal grooves). In veloped sternaulus ventrolateral^ Ichneumonoidea, the position of these mesepisternum, the mesopleural-basalare on the internal grooves, when present, relative to the muscle originates ledge of the attachment points of these two indirect formed by the anterior portion as the latter extends flight muscles confirms that they should be sternaulus posteriorly In those called notauli (as they are in nearly all from the epicnemial carina. mem-

arrow internal mesopleuron, dried specimen, left arrow showing mesopleural-basalare muscle, right showing ledge of short sternaulus on which anterior end of mesopleural-basalare originates. 26, Enicospilus sp., of in Ophioninae, right side of body in ethanol, showing origin elongate mesopleural-basalare trough of mesopleuron. 326 Journal of Hymenoptera Research

on Figs 27-32. Longitudinal sections through mesosomas of Ichneumonoidea, showing muscles originating mesopleuron: 27-28, Ichneumonidae, Cryptinae, right side of body: 27, in ethanol showing mesopleural-basalare muscle (arrow) and its dorsal cap; 28, same specimen, dried (resulting in gaps in muscle bundle), showing origin of mesopleural-basalare muscle on internal ledge formed by sternaulus (arrow). 29-30, Fopius vandenboschi = (Braconidae, Opiinae), dried: 29, ventral and left portions of body, bottom arrow midventral phragma = (partially broken in dissection), top arrow mesopleural-basalare muscle; 30, right side of body, mesopleural- = == basalare removed, bottom arrow region where mesopleural-basalare originates, middle arrow precoxal = sulcus visible through cuticle, top two arrows muscles inserting on axillary sclerites. 31, Wroughtonia ferruginea Volume 15, Number 2, 2006 327 bers of the subfamily Ichneumoninae in inside, the ledge along the dorsal side of which the sternaulus is distinct as a short this braconid "sternaulus" serves as a gen- anterior crease (e.g. Fig. 25), the meso- eral attachment point for three sets of pleural-basalare also originates there, pro- muscles located laterad the mesopleural- viding evidence that the sternaulus in these basalare muscle (the ledge best seen in three subfamilies is homologous. In the Fig. 30 where this "sternaulus" is visible pimplines, ichneumonines, and campople- through the semi-transparent cuticle). All gines that lack a visible sternaulus, the three of these muscles are fan-shaped, with mesopleural-basalare muscle originates at a broad origin on the wall of the meso- the lateral margin of the ventral curve of pleuron. They are strongly tapered dorsal- laterad the the in fan the mesepisternum just origin ly, where muscle bundles each of the dorsoventral indirect flight muscle unite to form tendons (Figs 24, 30, 32). Two (Figs 21-23, 26). This is also where the of these muscles appear to correspond to mesopleural-basalare originates in all of the muscles numbered 8 and 9 in Gibson the braconids that were dissected. The (1986b) and are clearly axillary muscles, sternaulus is long in nearly all cryptines inserting on the third axillary sclerite at the and tends to curve dorsally as it extends base of the fore wing. The third is similar in posteriorly. However, in the species dis- position to Gibson's (1986b) muscle num- sected, the mesopleural-basalare is at- ber 4, and inserts on the mesoscutum near tached to the anterior end of the sternaulus, the posterior notal wing process. This same which is more ventrally displaced. In those pattern of three muscles was found in all braconids with a well-developed "sternau- ichneumonoids examined. However, the lus," such as members of the Agathidinae, position of the muscles and in particular Doryctinae, Helconinae, and Rogadinae, their orientation (nearly longitudinal vs. the mesopleural-basalare passes over this dorsoventral) varies with the shape of "sternaulus" from its more ventral origin and, in braconids, the angle of the on the floor of the mesopleuron. In both "sternaulus." When the braconid "sternau- braconids and ichneumonids, the muscle lus" is strongly oblique, for example, the be varies in length and width (compare dorsal-most of the axillary muscles may rather than dorsoven- Fig. 29 with Fig. 31), as might be expected nearly longitudinal given the differences among species in the tral in orientation. shape of the mesothorax. Although the origin of a muscle may be In the opiine braconid Sternaulopius, less stable than the insertion for establish- which has a second "sternaulus" below ing homologies (see especially Daly 1963, the observations the the first, the mesopleural-basalare muscle 1964), above support of van originates on the internal ledge formed by conclusions Achterberg (1993, 1994) the ventral-most sternaulus, which, not that the ventral sternaulus in Sternaulopius to the sternaulus of ichneu- coincidentally, is in the same position as is homologous the dorsal "sternaulus" the attachment point of this muscle in other monids. Similarly, braconids. The dorsal-most "sternaulus" in of Sternaulopius is homologous to the single found in other which Sternaulopius is homologous to the "ster- groove braconids, naulus" of other braconids, and on the van Achterberg has termed the precoxal

muscle, arrow = (Braconidae, Helconinae), left side of body in ethanol, showing broad mesopleural-basalare side of dried, dorsal cap of muscle. 32, Alabagrus sp. (Braconidae, Agathidinae), right body, mesopleural- = basalare muscle removed, arrows muscles inserted on axillary sclerite. 328 Journal of Hymenoptera Research

and other mesothoracic muscles. Table 1. List of taxa dissected for examination of mesopleural-basalare

BRACONIDAE Agathidinae USA Alabagrus sp., College Station, TX, Braconinae Station, TX, USA Atanycolus simplex (Cresson) and Digonogastra sp., College

Cenocoelius saperdae (Ashmead), College Station, TX, USA Doryctinae Heterospilus sp., Gainesville, FL, USA Helconinae National Forest, TX, USA Helconidea ferruginea (Brues), College Station & Angelina Homolobinae

Homolobus truncator (Say), Conroe, TX, USA Macrocentrinae

Austrozele sp., Gainesville, FL, USA Opiinae Fopius arisanus (Sonan), Honolulu, HI, USA W. Indonesia Fopius vandenboschi (Fullaway), Tangerang, Java, Psyttalia incisi (Silvestri), Honolulu, HI, USA coastal Sternaulopius bisternaulicus Fischer, Kenya Rogadinae Aleiodes aciculdtus Cresson, College Station, TX, USA ICHNEUMONIDAE Anomaloninae

Anomalon sp., Conroe, TX, USA

Syzeuctus sp., Conroe, TX, USA Cryptinae Dominica Cryptini spp., College Station, TX, USA and Springfield, Ichneumoninae

Barichneumon sp., Bastrop, TX, USA

Cratichneumon sp., Huntsville, TX, USA

Limonethe sp., Gainesville, FL, USA Mesochorinae

Mesochorus sp., College Station, TX, USA Ophioninae

Enicospilus sp., Gainesville, FL, USA Pimplinae Pimpla sp., College Station, TX, USA Rhyssinae macrurus Station, TX, USA Megarhyssa' (L.), College APIDAE

Apis mellifera L., College Station, TX, USA

Gasteruption sp. (Rhydinofoenus sensu Townes), College Station, TX, USA Rhopalosoma nearcticuni Brues, College Station, TX, USA VESPIDAE

Polistes sp., College Station, TX, USA

sulcus. The sternaulus of ichneumonids likely to be homologous. It therefore seems has the same function as the "sternaulus" appropriate to use the term precoxal sulcus (= precoxal sulcus) of most braconids, but for the more dorsally displaced groove the the muscles whose origin they support are found in most braconids, despite different and thus the structures are un- probability that this term is used for non- Volume 15, Number 2, 2006 329 homologus features in other Apocrita. conceal ventral corner of pronotum above Coining another term would simply add base of fore coxa; oblique carina and to the confusion, and seems unnecessary associated, shallow groove (Fig. 8) weakly since precoxal sulcus has become fairly to distinctly developed, separating flange well established at least in the European from rest of propleuron. Pronotum dorsal- braconid literature. ly with transverse, crenulate groove along posterior margin; pronope absent. Notauli Taxonomy deep anteriorly, shallow to barely indicat- ed into dis- Sternaulopius Fischer, 1965 posteriorly, weakly curving tinct, rounded midpit; unsculptured pos- Stemaulopius Fischer, 1965: 311. teriorl Mesoscutum carinate i y margined bisternauhcus . , , Type species: Stemaulopius ,, , i . -n • from base of notaulus to axillarv £. -.nsr- *„ i j i j laterallyJ 1965. and _ _. Fischer, Monobasic original desig- tT^. _ , ... flange (Fig. 5). Iransscutal articulation nation. ? , j represented by a distinct sulcus lmmedi- carina Head (Figs 1-4). Occipital widely ately anteriorad crenulate scutoscutellar carina at base separated from hypostomal sulcus. Scutellum (Fig. 7) polished, un- than of mandible, extending dorsally more sculptured to weakly punctate (except not or half height of head, only weakly along posterior margin). Precoxal sulcus at dorsal curved medially end, completely (Fig. 6) broad, crenulate anteriorly, absent carina absent dorsal-medially; hypostomal or apparently so posteriorly; distinct from protruding as a flange below base of the more ventrally placed sternaulus; ster- mandible. Gena (including temple behind naulus well-developed as a broad, crenu- eye) and vertex smooth, polished. Frons late groove extending more than half bare, polished and impunctate but usually length of mesopleuron; both precoxal with weak, irregular sculpture along mid- sulcus and sternaulus straight over poste- line. Mandible (Fig. 4) evenly curved along rior half, never curving ventromedially. dorsal margin; ventral margin distinctly Hind margin of mesopleuron crenulate carinate, without basal tooth or lobe; above (though sometimes weakly) and bidentate, dorsal tooth longer than ventral below mesopleural fovea (Figs 5, 6). Ven- tooth; outer surface of mandible weakly to tral midline crenulate, with carinations stongly curved from base to apex, species distinct posteriorly but not developed as with weaker curvature have a completely a postpectal carina. Metapleural/propo- concealed labrum. Labrum varying from deal junction either obscured by sculpture exposed to completely concealed; clypeus or represented by a weak impression. to (Figs 1, 2, 4) with ventral margin truncate, Propodeum rugose carinate-rugose. varying from sharp (when labrum ex- Hind tibia lacking basal carina typical of posed) to slightly thicker (when labrum Utetes Foerster, 1862. concealed), in profile clypeus evenly con- Wings (Figs 14, 15). Stigma thickened, vex, moderately protruding, distinctly wedge-shaped; r arising from basal 0.3-0.4; overhanging labrum when the latter ex- second submarginal cell long (2RS<3RSa); posed. Epistomal sulcus generally more m-cu arising distad base of 2RS, (RS+M)b deeply impressed laterally than dorsally. thus absent; (RS+M)a arising distinctly 1-1 A Malar sulcus a distinctly impressed line, below parastigma, 1RS>0.3X1M; but re- Maxillary palp about as long as height of curved towards posterior margin head, 6 segmented; labial palp 4 segment- maining separate from margin by at least subdiscal cell closed, ed. First flagellomere longer than second; width of 1-1A; 1st at Hind RS absent, apical flagellomere spinose tip. wing nearly represented a m-cu Mesosoma (Figs 5-9). Propleural flange by weak, unpigmented crease; at least half distance to large, strongly curved laterally to partially extending wing 330 Journal of Hymenoptera Research

1837 see affords an margin as a curved, distinctly impressed, (new synonym, below) weakly pigmented line. opportunity to compare Sternaulopius to 1862 as well as to Metasoma (Figs 10-12). Petiole with spi- Xynobius Foerster, discuss the of racle located near middle of segment; pit- placement Sternaulopius within the like dorsope present basally at junction of Opiinae. has been in dorsal and lateral carinae; laterope present Although O. bajulus placed at extreme base. Second and following Xynobius in all recent treatments of Opiinae Fischer Tobias and visible terga neither sculptured nor un- (e. g. 1972, Jakamavi- van usually shortened. Second metasomal ter- cius 1986, Achterberg 1997, 2004; et al. there are distinct gite with spiracle placed dorsally, just Quicke 1997), mesad the rounded lateral fold. Hypopy- differences between bajulus and the type gium weakly protruding at midline, never species of Xynobius {viz, X. pallipes Foerster, the current valid name is strongly attenuate nor extending beyond 1862, for which caclatus new metasomal tergites. Ovipositor lacking Eurytenes (Haliday, 1837), of the more sharp subapical notch. combination, see below). One — infest- of these differences is the ar- Biology. Reared from tephritids important of metasomal In £. ing fruits belonging to several plant fam- rangement spiracles. abnormis ilies. Known only from Subsaharan Africa caelatus (as well as in Eurytenes the of Eur- and Madagascar. Included species: Ster- (Wesmael, 1835), type species the of the naulopius bistemaulicus Fischer, 1965 and ytenes Foerster, 1862), spiracle is situated lat- the new species from Madagascar de- second metasomal tergum scribed below. erally, well below the crease that separates — of the from the Excluded species. Sternaulopius beieri the dorsal portion tergite of the In both Fischer, 1968 (= Opius bajulus Haliday, lateral portion tergite. beieri = the 1837). bistemaulicus and ( bajulus), is located in Diagnosis. —Sternaulopius is distin- spiracle (Fig. 12) dorsally of most other guished from nearly all other supraspecific a position typical opiines: taxa in the Opiinae by the presence of both either adjacent to the crease (when a dis- more a sculptured precoxal sulcus and a true tinct crease is present) or situated was de- sternaulus (Fig. 6), and from Biophthora by medially. Xynobius originally the absence of rugose sculpture on the scribed as a cyclostome braconid (Foerster scutellum dorsomedially. Sternaulopius is 1862), but Foerster obviously had difficulty further recognizable by the presence of placing this taxon since in the same paper other a dorsope, a distinct median pit on the he described two genera (Biophthora mesoscutum, and a five-sided second sub- Foerster, 1862 and Aclisis Foerster, 1862) in taxa that are marginal cell, as well as the absence of both two different family-group a basal tooth or lobe on the mandible and now (Fischer 1972, van Achterberg 2004) a basal carina on the hind tibia. included in Xynobius. Muesebeck and — the first Remarks. The type species of Stemaulo- Walkley (1951) were apparently with plus was described on the basis of one to associate the name Xynobius Opii- female and two males from the Democratic nae when they placed all three of these Republic of the Congo (Fischer 1965). A Foerster genera as synonyms of Opius established second species, Sternaulopius beieri Fischer, Wesmael, 1835. Fischer (1972) was subsequently described from Ger- a comprehensive classification for the as one of many on the basis of a single female Opiinae that included Xynobius (Fischer 1968). Only one additional speci- 36 subgenera of Opius. Fischer (1972) men has been reported to date (Quicke et presented the first clear delineation of al. 1997). The discovery that S. beieri is Xynobius, which he characterized based a junior synonym of Opius bajulus Haliday, on its exposed labrum and sculptural Volume 15, Number 2, 2006 331 features of the mesothorax. He noted separated throughout. Despite these differ- especially the heavily sculptured scutellum ences, removal of bajulus from Xynobius is in the the fact that both (Fig. 19), an unusual feature Opiinae, complicated by bajulus and this is shared by both bajulus and and caelatus can be characterized by the caelatus. Both species have been character- shared possession of strongly elevated and ized as having an exposed labrum and also heavily sculptured scutellum (Fischer in rela- a deep, pit-like dorsope on the petiole, 1972), an unusual feature found features emphasized by Tobias (1977) and tively few Opiinae. In one of the other van Achterberg (2004) respectively. Thus, species with a similarly sculptured scutel- both of these species have been included in lum, Opius severini Fischer, 1964, the Xynobius since 1972 regardless of the scutellum, fore wing venation, clypeus differences in how Xynobius has been and mandible are as in caelatus, but the characterized by Fischer (1972) and sub- spiracle of the second metasomal segment sequent authors, and regardless of whether is positioned as in bajulus. One is therefore Xynobius has been accorded subgeneric left with the choice of a very broad (Fischer 1972, Tobias 1977) or generic (van definition for Xynobius, such as that offered Achterberg 1997, 2004) rank. Thus, if by Tobias (1977), Tobias and Jakimavicius van bajulus is retained in Sternaulopius, follow- (1986) and, apparently, Achterberg accomodate ing the discovery here of the synonymy (2004), which could bajulus, restricted one Fischer between bajulus and beieri, then Sternaulo- or a much more (e.g. relative merits of each are phis becomes a synonym of either Xynobius 1972). The below. or Biophthora (bajulus is the type species of outlined of the most Biophthora). Throughout much 1900s, The difference between bajulus and cac- Opiinae were placed in the genus Opius, over 1000 latus in placement of the spiracle of the which eventually encompassed facilitated work on second metasomal tergum argues against species. Fischer (1972) a classification retention of bajulus in Xynobius. Differences the Opiinae by presenting in the details of fore wing venation and in which several distinctive genera were an extensive shape of the clypeus, mandibles, and recognized and subgeneric classification was for petiole provide further evidence that baju- proposed Opius. been reduced lus and caelatus may not be closely related, Opius has subsequently by taxa and bring into question the homology of removal of putatively monophyletic 1860 some of the character states previously such as Utetes and Psyttalia Walker, and an to used to place these two species together in (Wharton 1987, 1988), by attempt Xynobius. The second submarginal cell is restrict the definition of Opius to species a basal tooth on the mandible parallel-sided in caelatus, m-cu is antefur- with (van and Salvo Further deli- cal, the stigma is more nearly parallel- Achterberg 1997). is essen- sided than wedge-shaped, the clypeus is mitation of monophyletic groups in the short and strongly protruding, the mandi- tial for progress understanding of ble is more massive basally and strongly relationships and evolutionary biology of this and of tapered distally, and the dorsal carinae large important group dipteran in the sense the petiole are strongly convergent basally. endoparasitoids. Xynobius sub- used Tobias and van In bajulus (Figs 16-20), the second by (1977) Achterberg is is well delineated, but difficult marginal cell is narrowed distally, m-cu (2004) fairly to define as because all of the postfurcal, the clypeus is truncate and monophyletic for of the nearly flat (and the labrum consequently characters given recognition can be considered more weakly exposed), the mandible is genus plesiomorphic. noted van slenderer and nearly parallel-sided, and As correctly by Achterberg for has a number the dorsal carinae of the petiole are widely (2004), example, Xynobius 332 Journal of Hymenoptera Research of features suggestive of a basal placement argument for the monophyly of Xynobius within the Opiinae. Features such as using characters emphasized by previous a distinct dorsope, distinct hypoclypeal workers, I prefer to retain Xynobius as depression (labrum exposed by a gap a subgenus at this time rather than elevate between the clypeus and mandibles), ab- it to generic rank. If Opius is defined in sence of a transverse carina on the pro- a restricted sense to include only species pleuron and absence of a true sternaulus with a basal tooth on the mandible (van may be useful for recognition of Xynobius Achterberg and Salvo 1997), then Xynobius within the Opiinae (van Achterberg 2004), is better placed in Eurytenes. but they are also indicative of the basal Placement of Xynobius in Eurytenes by no placement of this group since these fea- means solves the problem of monophyly of tures also characterize the Exothecini, the Xynobius, but does bring together several putative outgroup taxon and are thus groups of opiines that share characters plesiomorphic in this context. Some suggestive of a close relationship (though exothecines also have a short (relative to it is admittedly tricky to define relation- 2RS) cross-vein r in the fore wing, the only ships among those putatively basal taxa other recognition feature noted by van that retain both a deep dorsope and an Achterberg (2004). The latter is significant exposed labrum). Eurytenes has long been because it serves to separate Xynobius characterized by the extremely basal origin sensu van Achterberg (2004) from Eury- of r on the stigma, and only two species tenes. have been described with this characteris-

The sculptural patterns noted by Fischer tic. As pointed out by Wharton (1988), (1972) in his more restricted definition of however, the shape of the pterostigma is Xynobius provide a seemingly better op- also unusual, and the shape of both the portunity for defining this taxon as mono- stigma and the petiole suggest a relation- phyletic, but there are problems of homo- ship between the type species of Eurytenes, logy. The species included in Xynobius by £. abnormis, and species such as Opius Fischer all have a sculptured scutellum, but macrocerus Thomson, 1895 (type species of the pattern differs markedly among the Stigmatopoea Fischer, 1986). Critical exam- species. For example, Opius rudis Wesmael, ination of the fore wing of macrocerus 1835, included in Xynobius by Fischer demonstrates that the Eurytenes condition (1972), has a flattened scutellum with has been achieved by increasing the angle coriaceous sculpture lacking rugosities. In between r and 3RSa and increasing the species such as caelatus, bajulus, and sever- length and curvature of r so that the latter ini, the scutellum is both elevated and extends to the base of the stigma. The close rugose, providing a putative synapomorph relationship between macrocerus and abnor- in terms of a complex character. However, mis is further supported by other shared this same complex character state is found features of the fore wing (shape of second in some of the species of Biosteres Foerster, submarginal cell and placement of m-cu 1862. Further, as detailed above, these and 2CUb) as well as the fact that both three species (caelatus, bajulus, and severini) have the spiracle of the second metasomal differ markedly in wing venation, shape of segment displaced laterally. Fischer (1998) the clypeus and mandibles, and placement accepted the relationship between macro- of the metasomal spiracles. The available cents and Eurytenes based on fore wing evidence suggests, then, that this scutellar venation, and in particular the shape of the feature has either developed independent- stigma, and tranferred several additional ly in at least a few opiine lineages or is species to Eurytenes. Van Achterberg a primitive feature retained in these (2004), on the other hand, placed Stigmato- lineages. As I am unable to provide a strong poea as a synonym of Xynobius, but did not Volume 15, Number 2, 2006 333 include or discuss Eurytenes despite the for monophyly of the included species. fact that Quicke et al. (1997) included Additionally, based on the above hypoth- a of Eurytenes as subgenus Xynobius, citing esized relationships, recognition of Eury- van Achterberg and van Zuijlen (in press). tenes separate from Xynobius would render In either case, macrocerus is nicely posi- the latter paraphyletic if Stigmatopoea is tioned as morphologically intermediate treated as a synonym of Xynobius (as in van between caelatus and abnormis, with suffi- Achterberg 2004). Eurytenes and Xynobius in cient similarities the shape of the stigma, also compete for priority since both were the fore wing venation, and the placement described in the same publication (Foerster of T2 spiracles to justify their grouping as 1862). Eurytenes has been consistently a single genus. Considering only these recognized as a genus since it's initial three taxa, the hypothesized relationships description; hence my preference for use are Xynobius + (Stigmatopoea + Eurytenes of this name over Xynobius. This still leaves s. If the of the distal s.). shape elongate unresolved the question of the most ap- portion of the stigma is used to support propriate genus group name(s) for bajulus the monophyly of Eurytenes s. /., then the and bisternaulicus. narrowed basal portion in conjunction If Opius is temporarily maintained as with a relatively long vein r can be used a paraphyletic assemblage from which to support a Stigmatopoea + Eurytenes s. s. monophyletic groups are extracted as they relationship. However, this would be an become recognized and delineated over oversimplified picture of relationships time, following Wharton (1988), then both since Eurytenes, as defined here, embodies bajulus and bisternaulicus can be retained in a fair number of species, including most of Opius until their relationships are better Fischer's (1998) Eurytenes and van Achter- understood. A restricted definition for berg's (2004) Xynobius. Here, I propose Opius is ultimately desirable, however, restricting the name Xynobius to those and since van Achterberg and Salvo species with the spiracle of the second (1997) have provided one, this leaves metasomal segment located more laterally, bajulus and bisternaulicus excluded from though this will need to be re-evaluated in Opius, but with Biophthora and Sternaulo- a more comprehensive treatment of this pius, respectively, as available genus-group group. Therefore, I reject the inclusion of names. Aulonotus Ashmead, 1900 is also bajulus in Xynobius, though an argument available for species with the second could be made that bajulus is simply metasomal spiracle placed dorsally and a highly derived species of Xynobius. possessing a distinct dorsope, a sculptured Additional differences in the stigma, vena- precoxal sulcus, and a mesonotal midpit. tion, clypeus, and mandibles all suggest Fischer (1972) and Fischer and Koponen that bajulus should be excluded not only (1999) treated Aulonotus as a subgenus of from Xynobius but also from Eurytenes s. I. Opius, Fischer (1998) elevated it to generic Similarly, rudis is also excluded from both rank, and van Achterberg (2004) listed it as Eurytenes and Xynobius since it lacks both a synonym of Xynobius. Aulonotus is a con- the ventrally displaced T2 spiracle and venient place to put species formerly a dorsope. placed in either Xynobius or Eurytenes, but Rank is subjective, and thus an argument which do not fit the restricted definition could be made for retaining both Eurytenes presented above. Unfortunately, the name and Xynobius as separate genera. The Biophthora, which has remained in obscu- Foerster shared placement of the spiracle laterally rity since originally proposed by on the second metasomal segment, which (1862) and has been treated as a synonym differs from that in Exothecini, provides an since the works of Marshall (1891) and argument against this course of action and Dalla Torre (1898), has priority over the 334 Journal of Hymenoptera Research more widely used name Aulonotus. Apodes- than high; eye in lateral view large, 2.9-3.9 mia and Utctes also date from Foerster X longer than temple. Frons, vertex, and (1862), and also have been recently treated occiput highly polished, face appearing as genera, but their definitions would have less polished due to punctation; low, to be expanded considerably to accomo- median ridge present from epistomal sul- to level of date bajulus and/or bisternaulicus since the cus antennal bases, replaced on type species of both Apodesmia and Utetes frons by shallow, median, crenulate line lack a dorsope. Biophthom Foerster, new extending to or nearly to median ocellus; status, is thus the most suitable genus area between and immediately posteriorad antenna often face dis- group name for bajulus and Sternaulopius is weakly rugulose; the most suitable name for bisternaulicus. tinctly punctate throughout, punctures about their Following the discovery of a species separated by own diameter; from Madagascar (described below) with frons with 3-5 setae near eye margin, vertex and frons otherwise clypeus and mandibles resembling bajulus bare; occipital with a row of setae but otherwise more closely resembling margin single dorsally. of ocellar field X bisternaulicus, it has become much more Width 1.1-1.5 distance from lateral ocellus to width of ocellar challenging to find morphological differ- eye; field 2.2-2.8 X width of lateral ocellus. ences useful for retaining Sternaulopius and carina as a short Biophthora as separate genera. In addition Hypostomal protruding but distinct beneath mandible when to similar sculptural features already not- flange mandible carina ed, the venation of fore and hind wing is closed; occipital widely from carina ventral- nearly identical (and similar to many other separated hypostomal about to basal width of opiines). Nevertheless, I retain them as ly (separation equal and elevated separate largely because Sternaulopius is an mandible), sharp distinctly below Afrotropical group of tephritid parasitoids throughout, extending dorsally just of in lateral view, not reflected and bajulus is a temperate species with the top eye at dorsal terminus; shallow- general habitus and short ovipositor sug- medially very curved in lateral view. Malar gestive of it being a leaf-miner parasitoid. ly space distinct, 0.15-0.25 X a little In addition to differences in the ovipositor eye height, shorter than basal width of mandible; and shape of the metasoma, bajulus has to between distinct a rugose, elevated scutellum. Other, seem- punctate (similar face) malar sulcus and of ingly more trivial, differences are noted in margin clypeus, smooth and laterad sulcus. the descriptions below. polished Clyp- eus about twice wider than high; truncate Sternaulopius bisternaulicus Fischer, 1965 ventrally in frontal view, slightly protrud- in in (Figs 1-15) ing ventrally profile; nearly triangular outline, the epistomal sulcus only weakly Sternaulopius bisternaulicus Fischer, 1965: 312- curved; uniformly punctate, pattern of 314, holotype female in MRAC; Fischer 1968: punctation as on face; epistomal sulcus 105, key. Fischer 1971a: 125, catalog; Fischer distinctly impressed laterally, more weakly 1987: 562-564, redescription, figures; van so anterior tentorial small, Achterberg 1993: 121, figures; Yu and van medially; pits round. Mandibles over Achterberg 2005, electronic catalog. abruptly narrowing Opius bisternaulicus: Wharton 1988: 355-357. basal half, more gradually and evenly — narrowing from midpoint to apex; outer Redescription. Head (Figs 1-4) in dorsal surface strongly convex; mandibles strong- view 1.8-2.0 X broader than long, 1.40-1.55 ly deflected ventrally, broadly exposing X broader than length in lateral view, eyes labrum. Antenna with 23-30 flagellomeres width at distinctly bulging, eyes 1.05-1.10 (27 in holotype), clearly varying with body X width at face temples; 1.45-1.65 X wider length; first flagellomere slightly longer Volume 15, Number 2, 2006 335

than second to second, subequal third; first posteromedially, transverse carina varying 3 X than from flagellomere roughly longer wide, distinct to indistinct; propodeal spi- tenth flagellomere roughly 2 X longer than racle minute, situated about midway be- wide; apical flagellomere sharply pointed, tween anterior and posterior margins; but the not attenuate. tip Maxillary palps propodeum separated from metapleuron about equal in length to height of head. by a shallow to deep groove. Metapleuron 1.25-1.35 X Mesosoma (Figs 5-9) longer broadly impressed and carinately rugose than 1.65-1.80 X high, longer than wide, around margins; median plate varying Pronotum dorsally without median pit, from polished, punctate, and largely un- crenulate along posterior margin, other- sculptured to uniformly rugulose. Hind wise polished, unsculptured; pronotum margin of mesopleuron crenulate through- laterally with dorsal, crenulate line extend- out, though sometimes more weakly so ing as a broad, median groove to ventral dorsally, the crenulate impression forming corner, the groove crenulate for most of its nearly a straight line. Precoxal sulcus length, though often more weakly so incomplete posteriorly, but extending most medially, posterior margin crenulate at of way to mid coxa; gradually widening least over ventral 0.5, area between poste- anteriorly, roughly twice as broad anteri- rior crenulate margin and median crenu- orly as posteriorly; crenulate throughout; late groove varying from weakly to mod- precoxal sulcus anteriorly joining broadly erately rugulose, pronotum laterally other- crenulate groove along anterior margin of wise smooth and polished dorsoposteriorly mesopleuron and extending dorsally then and anteriorad median groove. Propleural posteriorly ventrad the setose subtegular flange large, distinct, sharply bent poster- ridge. Sternaulus crenulate throughout, oventrally; separated from remainder of nearly parallel to but distinctly separated propleuron by oblique, weakly to strongly from more dorsally positioned precoxal sculptured groove. Anterior declivity of sulcus, slightly longer than the latter and of mesoscutum densely covered with decum- approximately uniform width throughout, bent setae, setae extending onto base of Fore wing (Fig. 14) with stigma broad, in disc, 2-3 rows along notauli, and as wedge-shaped: widest at origin of r, a scattered patch of somewhat more erect tapered into metacarpus distally; r arising setae between midpit and scutellar sulcus; from about basal 0.4, length of r 0.75-0.90 notauli deeply impressed and sculptured X width of stigma; second submarginal on basal 0.4-0.5 of mesoscutum, abruptly cell large, weakly converging distally, 5- to gradually transforming to very shallow, sided with m-cu distinctly postfurcal, 2RS unsculptured depressions extending pos- strongly reclivous, r-m slightly inclivous, teriorly and sometimes almost impercepti- completely depigmented and descler- bly to oval or tear-drop shaped midpit; otized; 3RSa 1.35-1.70 X longer than 2RS; midpit covering apical 0.2 of disc, well 3RSb 1.45-1.60 X longer than 3RSa; 3RSb separated from distinct transscutal articu- extending to wing margin very close to lation; carinate lateral margin of disc wing apex; RS+M weakly sinuate, nearly crenulate, deeply impressed between te- straight, arising low on almost evenly gula and rugose base of notaulus. Scutellar bowed 1M (the curvature slightly stronger sulcus about 4 X wider than length along posteriorly), IRS 0.35-0.45 X length of 1M; midline; smooth, with 6-8 distinct ridges, large median bulla covering all of 2 Ma, Scutellum evenly convex, not strongly posterior extremities of RS+M and 2RS, elevated; polished throughout, with scat- and anterior portion of m-cu; 3M tubular tered setae. Metanotum with small, low and distinctly pigmented for more than median tubercle. Propodeum densely ru- half its length; lcu-a inclivous, separated gose, the sculpture often somewhat weaker from 1M by nearly its own length; 1st 336 Journal of Hymenoptera Research subdiscal cell closed, 2CUa strongly incli- often parts of metanotum reddish brown; vous, nearly twice length of tubular 2cu-a; metasomal terga and sterna otherwise dark 1-1A weakly bowed towards wing margin, brown in specimens from Rutshuru and separated near mid-length from the latter Psychotria fruits in Kenya, the laterotergites by about twice its width. Hind wing lighter brown with pale margins, terga and sterna in (Fig. 15) with RS represented by unpig- yellow brown pale specimens mented, spectral crease; 2M and m-cu from Cameroon, terga 2+3 with large, dark nearly always weakly but distinctly pig- brown spot medially, remainder of meta- mented, m-cu strongly bowed with poste- soma dorsally and ventrally yellow; wings rior end directed towards wing base; 2-1A hyaline. absent. Length of body (exclusive of antenna Metasoma (Figs 10-13) with gaster in and ovipositor) 2.5-3.6 mm; of wing 2.6- dorsal view broadly oval, distinctly taper- 3.6 mm; of antenna about 3.1-4.3 mm. ing anteriorly and posteriorly. Petiole 1.0- Male as in female except as follows: 1.3 X longer than apical width, apex 1.8- antenna with 23-30 flagellomeres; petiole 2.2 X wider than base; largely smooth, narrower apically, thus appearing more polished basal-medially, striate to strigose parallel-sided, 1.25-1.40 X longer than laterally and over entire apical half, dorsal apical width, apex 1.5-1.8 X wider than carinae converging near mid-length but base; body and wing length more variable, not quite touching, replaced by a low, 1.8-3.4 and 2.0-3.3 mm respectively; speci- median ridge over apical half; dorsope mens from Cameroon with antenna entire- distinct, deep. Hypopygium weakly scler- ly yellow and metasomal terga 2+3 yellow otized and folded along midline; short, brown with remaining terga brown; speci- with posterior margin moderately protrud- mens reared from Strombosia fruits in ing medially but apex not extending to tip Kenya also with antenna entirely yellow, of metasoma. Ovipositor protruding dis- but metasomal terga entirely dark brown. tinctly beyond metasoma, 1.4-1.5 X longer Fischer (1965) noted that in males, r was than mesosoma, upper valve with distinct half the length of 2RS in the male paratypes subapical node, valve slightly narrowed and 2/3rds the length of 2RS in the female immediately basad node; ovipositor sheath holotype; in the longer series examined 0.85-0.95 X length of mesosoma, with tuft here, most males had a slightly shorter r of long setae over apical third, 2 ventral cross-vein than females, but the difference and 1 lateral row of more widely spaced was smaller— than in Fischer's material. setae basally. Biology. Reared from larvae of the Color mostly dark brown; legs yellow, following Tephritidae (Diptera) infesting fifth tarsomere brown, apical 0.15-0.20 of fruit of Moraceae (from Ceratitis MacLeay posterior face of hind tibia and hind tarsi spp. infesting Antiaris toxicaria (Pers.) dorsally light brown in specimens from Lesch. in Kakamega forest, Kenya), Olaca- Rutshuru and those reared from Psychotria ceae (from Ceratitis anonae Graham and /or fruits in Kenya, brown infumation on tibia C. fasciventris (Bezzi) infesting Strombosia absent or very weakly indicated in speci- scheffleri Engl, in Kakamega), Sapotaceae mens from Cameroon and those reared (from C. anonae infesting Engleropmytum from Strombosia fruits in Kenya; flagellum oblanceolatum (S. Moore) T. D. Penn. in brown to dark brown; scape, occasionally Kakamega) and Rubiaceae (from Trirhi- pedicel, most of mandible, labrum, palps, thrum Bezzi and/or Ceratitis infesting tegula, and pleural areas adjacent tegula Coffea spp. for all Cameroon collections; yellow; apical margin of clypeus, posterior from Trirhithrum meladiscum Munro infest- traces of notauli, extreme base of metaso- ing Psychotria fractinervata Petit in Gata- mal tergum 2, scutellum posteriorly, and mayu forest, Kenya and P. mahonii C. Volume 15, Number 2, 2006 337

in Mt. and from T. Wright Kenya forest; from these species by its more broadly senex Munro P. K. infesting alsophila exposed labrum and long ovipositor with Schum. in Taita Hills, Kenya). See material distinct subapical node. examined section for details of localities. — Remarks. In overall appearance (body Bactrocera amplexa (Munro) was also pres- shape, size, and fore wing venation), S. ent in the of but samples Strombosia, the bisternaulicus resembles species of Utetes, isolated puparia from which wasps but has a distinct dorsope and lacks the emerged were those of Ceratitis. Emer- sharp carina on the base of the inner side of gence, as in all known opiines, was from the hind tibia that characterizes members the host puparium. Host stage attacked is of the genus Utetes. Since species of both unknown. genera have occasionally been reared from tephritids infesting the same host fruits, Material examined.—Holotype female, DEMO- and both lack the attenuate CRATIC REPUBLIC OF THE CONGO Rut- hypopygium of many of the other parasitoids of fruit- shuru, 29.V.1936 (L. Lippens) (MRAC). careful attention needs Additional specimens: BURUNDI 1 male(?), infesting tephritids, to be to these charac- Bururi, Urundi, 2200 m, 4.ix.l948 (Francois) paid distinguishing teristics in order to avoid (MRAC). CAMEROON 2 males, Akonolinga, misidentifica- viii.1982 (G. Steck), TAMU Quarantine # 82-31, tions. TAMU Voucher # 82; 2 females, 3 males, The most obvious difference between x.1982 Nkolbisson, (G. Steck), TAMU Quaran- bisternaulicus and the two other species tine #s 82-39 to 82-41, TAMU Voucher 87 to #s treated in detail here (duplicatus and 89; 1 male, same 1980 (Perkins) locality, bajulus) is the more broadly exposed (TAMU). DEMOCRATIC REPUBLIC OF THE labrum formed by the ventrally deflected CONGO 2 females, 2 males, same locality as mandibles. Compared to bajulus, both holotype, ix-x.1936 & 7.iv.l937 (MRAC). bisternaulicus and duplicatus have an KENYA 4 females, 4 males, Central Province, evenly : and convex, Gatamayu forest, 58.45'S, 36°41.83'E, weakly dorsally unsculptured scutellum, a more an- 17.iv.2001 & 19.X.2001 (R. Copeland); 1 male, broadly impressed terior of the and Coast Province, Ngangao forest, Taita Hills, margin metapleuron a fore cross-vein r. 3 21.255'S, 38 20.579'E, 12.viii.2002 (R. Cope- distinctly longer wing land & R. Wharton); 2 females, 2 males, Eastern Minor differences between these two and Province, Mt. Kenya forest, 12.568'S, bajulus include a straighter carinate groove 37 30.218'E, 6.xii.200l'(R. Copeland); Western on the posterior margin of the meso- 4 3 Province, Kakamega forest, females, males, pleuron and somewhat coarser propodeal 14.16'N, 34 51.82'E, reared from Ceratitis and mesopleural sculpture. infesting fruits of Strombosia scheffleri, There is considerable morphological 30.iv.2000 (R. Copeland); 1 male, 13.14'N, variation amongst the material available 34 53.76'E, reared from Trirhitlirum senex infest- for examination, but I am unable to de- ing fruits of Englerophytum oblanceolatum, lineate more than a 7.vi.2001 (R. Copeland);l female, 1 male, single species. Though not I consider all of 14.16'N, 34 51.87'E, reared from Ceratitis entirely satisfactory, the in the material examined infesting fruits of Antiaris toxicaria, 3.iii.2000 specimens a (R. Copeland); 1 male, same but 13.66'N, section as representing single, variable 34°53.12'E, 19.ii.2001 (TAMU, National Mu- species because the series reared from seums of Kenya).— Trirhithrum meladiscum infesting Psi/chotria Diagnosis. This species is similar to fruits in central Kenya encompasses most Sternaulopius duplicatus new species, of the range of variation observed in the Biophthora rossicus (Szepligeti) new combi- rest. More specifically, the larger speci- nation, and B. bajulus new status, in the mens reared from Strombosia fruits in possession of both a sculptured sternaulus western Kenya have slightly larger ocelli, and a sculptured precoxal sulcus. It differs and consequently a broader ocellar field. 338 Journal of Hymenoptera Research

Individuals reared from Psychotria fruits in measuring this structure on intact speci- various Kenyan localities, as well as indi- mens and the slightly longer, narrower viduals from Cameroon, have somewhat petiole of the Rutshuru specimens. Perhaps smaller ocelli as in the specimens from the most distinctive difference among the Rutshuru. Specimens from Cameroon tend material examined is in coloration, with to have more extensive, better-developed specimens from Rutshuru and from Psy- sculpture on the pronotum laterally and chotria and Engleraphytum fruits in Kenya a more consistently deep and heavily darker than specimens from Cameroon sculptured oblique groove on the pro- and from Strombosia and Antiaris fruits in pleuron than the larger individuals reared Kenya. Thus, both lighter and darker forms from Strombosia fruits in western Kenya, can be found in Kakamega forest in Individuals reared from Psychotria fruits in western Kenya. central are variable in this Kenya highly The species is distributed across equato- regard, covering the extent of variation rial Africa from coastal Kenya to Camer- seen in all other setae populations. The on oon. The type material of bisternaulicus is the median mesoscutal lobe extend further from Rutshuru, in the eastern portion of in from posteriorly specimens Cameroon present day Democratic Republic of the than in the others, but in these the Congo, near the Uganda border, mesoscutum is still less setose distinctly The medially desclerotized hypopygium than in the described below. new species i s reminiscent of some Cardiochilinae and The is much weaker propodeal sculpture Microgastrinae and similar in this respect in most of the from posteriorly specimens to the equally short hypopygium found in Cameroon and the material reared Kenyan fruit-infesting tephritid parasitoids of the from Strombosia fruits. The transverse genus Utetes. The functional significance is carina is also distinct and readily visible unknown, in these specimens. The material from Rutshuru and most of the specimens Sternaulopius duplicatus Wharton, reared from Psychotria fruits in Kenya are new species and as densely sculptured throughout, (Figs 33-35) a consequence, the transverse carina is usually indistinct. However, there is some Head in dorsal view 1.85-2.00 X broader than X variation within localities and in particular long, 1.50-1.55 broader than length the large series from Psychotria fruits in lateral view, eyes weakly bulging be- contains some individuals with weaker yond temples; face 1.30-1.45 X wider than propodeal sculpture and a couple of the high; eye in lateral view large, 3.35 (female) X males from Cameroon have denser sculp- and 2.90 (male) longer than temple, ture posteriorly. The sternaulus is a little Frons, vertex, and occiput as in bisternau- broader throughout in Cameroon and licus except median sculpture on frons Strombosia material than in Rutshuru and confined to area immediately posteriad Psychotria material, but is of the same antennal bases; face (Fig. 33) distinctly length and very well developed in both punctate on either side of polished, narrow groups. The second submarginal cell is midridge about as in bisternaulicus but with similarly shaped in all populations, but 1-2 distinct vertical rows of more closely- extreme within-population variation in set punctures on either side of midridge in of length 2RS resulted in the broad range addition to the irregular pattern of punc- for the 3RSa/2RS ratio noted in the above tures. Ocelli smaller than in most individ- description. Variation in the shape of the uals of bisternaulicus; width of ocellar field petiole, on the other hand, is partly 1.05-1.10 X distance from lateral ocellus to a function of the difficulty in accurately eye; width of ocellar field 2.2-2.5 X width Volume 15, Number 2, 2006 339

of Figs 33-35. Sternatdopius duplicatus, n. sp., holotype female: 33, face. 34, anterior-dorsal view mesonotum. = 35, lateral habitus, arrow precoxal sulcus. 340 Journal of Hymenoptera Research of lateral ocellus. Hypostomal and occipital scutal articulation; carinate lateral margin carinae as in bisternaulicus. Malar space of disc crenulate, deeply impressed be- short but distinct, 0.15-0.20 X eye height, tween tegula and rugose base of notaulus. a little shorter than basal width of mandi- Scutellar sulcus a bit narrower medially ble; region of malar space sculptured as in than in bisternaulicus, about 5.6 X wider bisternaulicus. Clypeus (Fig. 33) a little tal- than length along midline; smooth, with 6- ler and more oval in male than female, 8 distinct ridges. Scutellum and metano- appearing truncate below in both depend- tum as in bisternaulicus. Propodeum dense- ing on angle of view; nearly flat in profile; ly rugose; propodeal spiracle minute, uniformly setose and punctate, distinctly situated about midway between anterior but more sparsely punctate than in bister- and posterior margins; propodeum sepa- naulicus; epistomal sulcus impressed later- rated from metapleuron by a shallow ally, barely indicated medially; anterior groove. Metapleuron broadly impressed tentorial pits round, a bit larger than in and carinately rugose around margins; bisternaulicus. Mandibles gradually and median plate polished and largely un- evenly narrowing from base to apex; about sculptured. Hind margin of mesopleuron 3 X longer than median width; outer crenulate throughout, the crenulate im- surface very weakly convex, nearly flat; pression forming nearly a straight line. mandibles barely deflected ventrally, com- Precoxal sulcus incomplete posteriorly, but pletely concealing labrum when closed extending most of distance to mid coxa; (Fig. 33). Antennae broken in both speci- crenulate and of uniform width through- mens, male with 25 flagellomeres remain- out; precoxal sulcus anteriorly as in bister- ing, female with 18; first flagellomere naulicus. Sternaulus crenulate throughout, slightly longer than second, second slightly nearly parallel to but distinctly separated longer than third; first flagellomere rough- from more dorsally positioned precoxal ly 3.5 X longer than wide, tenth flagello- sulcus (the two grooves slightly converg- mere more than twice longer than wide ing posteriorly), slightly longer than pre- and longer in male than female. Maxillary coxal sulcus and of approximately uniform palps curled and length difficult to esti- width throughout. mate but appear slightly shorter than in Fore wing stigma about as in bisternau- bisternaulicus. licus but a little more gradually tapered Mesosoma (Figs 34, 35) 1.35 X longer distally; r appearing to arise more proxi- than high, 1.75 X longer than wide. mally, from basal 0.30-0.35; second sub- Pronotum not visible dorsally; laterally as marginal cell, m-cu, r-m, 1M, IRS, 2RS, 3M, in bisternaulicus. Propleural flange distinct, 1-1A, and median bulla as in bisternaulicus; bent posteroventrally; separated from re- 3RSa about 1.70 X longer than 2RS; 3RSb mainder of propleuron by oblique, weakly 1.55-1.60 X longer than 3RSa; 3RSb ex- to strongly sculptured groove. Anterior tending to wing margin very close to wing declivity of mesoscutum densely covered apex; RS+M straight to nearly so; lcu- with decumbent setae, these extending a vertical to nearly vertical, separated from onto disc, covering entire median mesos- 1M by half its length or slightly less; 1st cutal lobe and anterior and median por- subdiscal cell about as in bisternaulicus but tions of lateral lobes (Fig. 34), though not 2cu-a longer, distinctly more than half as as densely on anterior declivity; notauli length of 2CUa. Hind wing about as in weaker than in bisternaulicus, impressed bisternaulicus but m-cu shorter, straighter, and weakly sculptured on basal 0.2-0.3 of and more weakly pigmented than in most mesoscutum, faintly indicated medially, individuals of the latter; 2-1A absent. imperceptible posteriorly; midpit small, Metasoma with gaster in dorsal view round, deep, separated from distinct trans- broadly oval, distinctly tapering anteriorly Volume 15, Number 2, 2006 341 and posteriorly. Petiole 1.35 (female) and densely setose mesoscutum, and concealed 1.60 (male) X longer than apical width, labrum. apex 1.6 X wider than base in male, base Remarks. —The concealed labrum of this not completely visible in female; striate to species makes it more challenging to argue in strigose except basal depression, dorsal for the exclusion of bajulus from Sternaulo- carinae extending to apex though weaker pius since the differences in appearance of posteriorly, weakly converging throughout the clypeus and mandibles between dupli- in bisternaulicus. length; dorsope as Hypo- catus and bajulus are relatively minor. As pygium short, not extending to tip of noted above in the general discussion metasoma; shape about as in bisternaulicus under Sternaulopius, recognition of Bio- but difficult to discern because of deflected phthora as a separate genus is based ovipositor. Ovipositor about 0.75 X length primarily on the nature of the scutellum of mesosoma, upper valve without distinct and in this duplicatus and bisternaulicus are subapical node; ovipositor sheath about markedly different from bajulus. Other 0.35 X length of mesosoma, with tuft of essential features that assist in the place- long setae apically, and more widely ment of duplicatus in Sternaulopius rather spaced setae basally. than Biophthora are found in the details of Color mostly dark brown to black; scape, sculpturing of the malar region, meso- clypeus, palps, tegula, base of metasomal pleuron, and propodeum as well as the tergum 2 and legs yellow, the coxae and punctation and shape of the clypeus. See trochanters more pale in male; mesosoma further remarks under bisternaulicus. dark brown to black dorsally and ventrally, The length of the ovipositor and sheath with small reddish patches laterally in given in the above description are esti- female, male more extensively reddish to mates since the holotype was not dissected, reddish-yellow; petiole reddish brown, However, since the base of the ovipositor is terga 2+3 yellow-brown to reddish brown, partially exposed in the holotype, the remaining metasomal terga dark brown; estimate is fairly accurate. The ovipositor wings hyaline or nearly so. of this species is thus considerably shorter Length of body (exclusive of antenna than in bisternaulicus, and at rest probably and ovipositor) about 2.3 mm; of wing 2.6- protrudes only slightly beyond the apex of 2.7 mm. the metasoma. As in bajulus, there is no the dorsal . discernible node on , , , , A subapical — , Material examined. female MADA- ., , Holotype , ,. , ._. ,. Valve in du tus ^ggeshng that duphca- GASCAR. Label with 4 lines as follows: "COLL. ?™ tus ma ovl oslt mto an earher host sta 8e MUS. CONGO", "Madagascar: Ankaratra", V P than bisternaulicus. "IV-1944" and "A. Seyrig" (deposited in MRAC). ^ rne presence of a sternaulus (in the is over- Paratype male MADAGASCAR same label addition to precoxal sulcus) data as holotype (MRAC). looked, duplicatus runs to Opius (Firkins) in the to Diagnosis. —This species is characterized castaneus Granger keys Opiinae by the presence of both a sculptured published by Granger (1949) and Fischer sternaulus and sculptured precoxal sulcus, (1972, 1987). Opius castaneus is the type 1971b. The a combination known only in Sternaulopius species of Frekius Fischer, lacks a sternaulus and Biophthora among the Opiinae. Ster- holotype of castaneus tibia is carinate naulopius duplicatus lacks the elevated, but surprisingly the hind

1 therefore transfer dorsally sculptured scutellum of the spe- basally as in Utetes. to Utetes the new cies placed in Biophthora below, and is most Granger's species where castaneus readily differentiated from the only other combination is Utetes [Frekius) the valid species of Sternaulopius, S. bisternau- (Granger). I retain, at least temporarily, licus, by the shorter ovipositor, more name Frekius as a valid subgenus pending 342 Journal of Hymenoptera Research a badly needed revision of the genus X longer than temple. Face (medially), Utetes. Typical Utetes have a broadly ex- vertex, frons and occiput polished; low posed labrum and thus the inclusion of median ridge present from epistomal sul- Frekius in Utetes is not without problems cus to level of antennal bases, replaced by since castaneus does not have an exposed shallow groove on frons; face sparsely and labrum. The parallels between Utetes and finely punctate medially, punctures sepa- Sternaulopius in the unusually variable rated by at least their own diameter, nature of the labral exposure are also weakly rugulose and dull (unpolished) deserving of closer attention. adjacent eye margin; frons with 3-5 setae Sternaulopius duplicatus is known from along eye margin, otherwise bare and only two specimens; the species name unsculptured; vertex sparsely setose. Ocelli ocellar field refers to the two parallel grooves on the and small, width of ocellar X mesepisternum (precoxal sulcus and ster- field 0.70-0.75 distance between lateral naulus). ocellus and eye; width of ocellar field 3.0- 3.4 X width of lateral ocellus. Hypostomal 1837) Biophthora bajulus (Haliday, carina protruding as a short but distinct (Figs 16-20) flange beneath mandible when mandible closed; occipital carina very widely sepa- Opius bajulus Haliday, 1837: 214; Marshall 1891: rated from carina 43, redescription, key, English; Marshall 1894: hypostomal ventrally (by a distance than basal width of 327-328, redescription, key, French; Dalla greater Torre 1898: 59, catalog; Szepligeti 1904: 164, mandible), relatively short, extending dor- checklist; Fischer 1958: 58-60 redescription, sally just above middle of eye in lateral keys; Fischer 1971a: 46, catalog; Papp 1981b: view, not reflected medially at dorsal Tobias and 1986: 35, key; Jakimavicius 96-97, terminus, broadly curved in lateral view et al. 2003: 393, key, figures; Belokobylskij due to prominantly swollen gena. Malar checklist, clarification of date of original space broad, 0.3-0.4 X eye height, about description in this and other 19th century equal to basal width of mandible; rugu- publications relevant to Opiinae. lose-strigose (Fig. 18), with sculpture ex- Biophthora bajulus: Foerster 1862: 260. tending from malar sulcus to ventral- Sternaulopius beieri Fischer, 1968: 103-105, new lateral of clypeus; malar sulcus synonym; Fischer 1971a: 125, catalog; Fischer margin from to but 1972: 478-479, redescription; Papp 1981a: complete eye subgenal margin from 273-274, distribution, diagnosis; Papp 1981b: barely distinguishable adjacent sculp- 135, redescription, key; Quicke et al. 1997: 25- ture. Clypeus broadly hemi-elliptical, trun- 36, venom glands, relationships; Tobias 1998: cate and sharpy margined ventrally, punc- 559-561, key; Belokobylskij et al. 2003: 396, tate and setose along dorsal margin, other- Yu and van checklist; Achterberg 2005, wise polished and bare; epistomal sulcus electronic catalog. distinctly impressed throughout, though Opius (Xynobius) bajulus: Fischer 1972: 88-90, slightly less so medially; anterior tentorial redescription, key, figures; Tobias and Jaki- pits narrow, slit-like. Mandible gradually mavicius 1986: 29, key, figures. and evenly narrowing from base to apex; Xynobius bajulus: van Achterberg, 1997: 18, slightly more than twice longer than basal original description listed as published in width but less than 3 X than median 1836, following Horn and Schenkling (1928); longer deflected van Achterberg 2004: 315. width; weakly ventrally, expos- — ing labrum in a narrow but distinct gap Redescription. Head in dorsal view 1.8- between clypeus and mandibles when the 1.9 X broader than long, 1.3-1.4 X broader latter are closed (Fig. 18). Antenna with 24- than in length lateral view, as wide at 26 flagellomeres; first flagellomere slightly as at temples eyes; face (Fig. 18) 1.7-1.8 X longer than second; apical flagellomere wider than in lateral high; eye view 1.2-1.3 sharply pointed, but tip not attenuate. Volume 15, Number 2, 2006 343

Maxillary palps shorter than height of sculptured. Hind margin of mesopleuron head. finely crenulate throughout, the crenulate Mesosoma (Figs 17, 19, 20) 1.2-1.3 X impression emarginate near middle. Pre- 1.8-1.9 X longer than high, longer than coxal sulcus (Figs 17, 19) incomplete pos- wide. Pronotum dorsally without median teriorly, extending slightly more than half pit, crenulate along posterior margin, oth- distance from anterior margin to mid coxa; erwise polished, unsculptured; pronotum finely crenulate throughout, weakly ta- laterally finely sculptured throughout ex- pered posteriorly; precoxal sulcus weakly cept largely smooth anteriorad median separated from crenulate groove along groove. Propleural flange large, distinct, anterior margin of mesopleuron or con- sharply bent posteroventrally; separated nected to groove only as a faint trace; from remainder of propleuron by oblique, anterior margin of mesopleuron finely strongly sculptured groove. Anterior de- crenulate, the sculpture extending poster- clivity of mesoscutum densely covered iorly ventrad subtegular ridge, with a few with decumbent setae; disc bare except weak striae extending ventrally towards for scattered row of setae along each middle of mesopleuron from subtegular notaulus and on either side of midpit; ridge. Sternaulus (Figs 17, 19) finely cren- notauli extending onto anterior 0.4 of disc ulate throughout, parallel to but distinctly as crenulate grooves, abruptly transform- separated from more dorsally positioned ing to shallow, unsculptured depressions precoxal sulcus; of uniform width through- extending posteriorly to narrow, slit-like out. Midventral groove crenulate; postpec- midpit, largely to completely obscured at tal carina completely absent, base by rugose sculpture; midpit (Fig. 19) Fore wing stigma broad, wedge-shaped: weakly sculptured, covering apical 0.25 of widest at origin of r, tapered into meta- disc, extending to barely perceptible trans- carpus distally; r arising from basal 0.35, r scutal articulation; lateral margin of me- short, at most half as long as width of soscutum crenulate, deeply impressed be- stigma; size and shape of second sub- tween tegula and rugose base of notaulus. marginal cell as in bisternaulicus; m-cu Scutellar sulcus about 4-5 X wider than distinctly postfurcal, r-m and 2Ma slightly length along midline, with 6-8 ridges more distinct than in most specimens of partially obscured by rugulose sculpture, bisternaulicus, due to trace of pigmentation; Scutellum (Fig. 19) arising vertically from 3RSa 1.65-1.75 X longer than 2RS; 3RSb posterior margin of scutellar sulcus, dis- 1.2-1.3 X longer than 3RSa; 3RSb extend- tinctly elevated above plane of mesoscu- ing to wing margin near tip but not as close turn; dorsal surface flattened, densely to apex as in bisternaulicus and duplicatus; rugose throughout, with sculpture extend- RS+M weakly but distinctly sinuate, aris- ing to metanotum. Metanotum with small, ing low on almost evenly bowed 1M (the low median tubercle. Propodeum (Fig. 19) curvature slightly stronger posteriorly), densely and finely rugose, the sculpture IRS 0.30-0.35 X length of 1M; 3M tubular somewhat weaker posterior-medially, and distinctly pigmented for slightly less transverse carina weak, barely discernible; than half its length; lcu-a vertical to very propodeal spiracle minute, situated about weakly inclivous, separated from 1M by midway between anterior and posterior about 0.5-0.7 X its length; 1st subdiscal cell margins; propodeum separated from me- closed, 2CUa strongly inclivous, distinctly tapleuron by a shallow groove margined longer than tubular 2cu-a; 1-1 A weakly medially by a low carina posteriorly, bowed towards wing margin, separated Metapleuron narrowly impressed and cari- near mid-length from the latter by nearly 3 nately rugose around margins; median X its width. Hind wing (Fig. 16) as in plate polished, punctate, and largely un- bisternaulicus except m-cu not so strongly 344 Journal of Hymenoptera Research

curved, nearly reaching wing margin and 1997). The specimen noted above from 2-1A short but distinct. Thiiringen in NHMW is assumed to be Metasoma (Fig. 17) with gaster in dorsal from Austria but this new record needs view nearly parallel-sided, gradually ta- verification. The holotype female of beieri pering posteriorly. Petiole short, length has the date hand-written sideways on the 0.85-0.95 X apical width; 2.35-2.45 X label, "Sachen" and "Dr. R. Krieger" wider at apex than at base; sparsely striate printed, and the specific locale hand- to strigose (Fig. 19), dorsal carinae distinct written and difficult to read. Fischer basally, difficult to distinguish from sur- (1968) interpreted the specific locality as rounding sculpture apically, apparently Altenberg, which was a well-known local- extending to apex, not converging; dorsope ity in Saxony of that time period. However, small, round, deep. Spiracle of second there was also a small village called metasoma] tergum positioned as in bister- Altenbach just east of Leipzig near Wur- naulicus, dorsad and nearly adjacent lateral zen, and I interpret the label to read Altenbach rather than Data crease of median tergite. Hypopygium Altenberg. on the of are short, apex not extending to tip of meta- holotype bajulus provided by Marshall and van soma. Ovipositor very short, not extending (1891)— Achterberg (1997). beyond apex of metasoma, upper valve Diagnosis. Biophthora is readily recog- nized the without subapical node; ovipositor sheath by presence of both a sculptured sternaulus and a about 0.25 X length of mesosoma, with tuft sculptured precoxal sul- cus in addition to the dorsal of long setae apically. rugose surface Head and mesosoma dark brown to of the elevated scutellum and a deep The scutellar feature black; tegula, metasoma and coxae dark dorsope. separates reddish brown, fore coxa sometimes more bajulus from members of the genus Ster- but not from yellowish; tarsomeres 5 and sometimes 4 naulopius, Biophthora rossicus dark 1901), new which brown, remainder of legs dark yel- (Szepligeti, combination, is identical to the low; clypeus and mandibles yellow; palps nearly bajulus (including of a true The holo- light brown; wings hyaline or nearly so. presence sternaulus). of rossicus has the notauli Length of body (exclusive of antenna type (HNHM) better than in I have and ovipositor) about 2.6-2.7 mm; of wing developed specimens seen of 2.8-2.9 mm; of antenna 3.0-3.1 mm. bajulus (including beieri).

Material examined. — Remarks Holotype female of baju- lus, ENGLAND (Walker) (NMID). Van Achter- I exclude bajulus from Sternaulopius, berg (1997) noted that the holotype of bajulus is despite the presence in bajulus of both a female but was listed in the original de- a crenuate sternaulus and a sculptured scription as a male probably because the precoxal sulcus; bajulus differs from the genitalia are not readily visible. species of Sternaulopius by the more rugose Holotype female of beieri, GERMANY "Sach- sen malar region, the more parallel-sided me- Altenbach bei Wurzen Coll. Dr. R. Krieger 31.5.93" (ZMHB). tasoma, and in particular the sculpture and of the scutellum. of Additional specimens: 1?, AUSLRIA, Thurin- shape The presence a in gen (NHMW); CZECH REPUBLIC, 1?, Hradek true sternaulus bajulus is hypothesized (NHMW); 1?, Moravia, Brno (NHMW); 1 as a case of remarkable convergence with female, IRELAND, County Kildare, R. Canal, the condition in Sternaulopius. The sternau- 18.vi.1944 (Stelfox) (USNM). lus of bajulus, though distinct when not Distribution. — recorded from Previously obscured by glue or the position of the legs Czech Republic, Denmark, Germany, (as is often the case), is decidedly finer and and Hungary, England, Ireland, Turkey weaker than in bisternaulicus and duplicatus. (Fischer 1971a, Papp 1981a, Quicke et al. See remarks section under the genus Ster- Volume 15, Number 2, 2006 345 naulopius for rationale regarding recogni- more detailed study as is the possible tion of Biophthora and placement of bajulus relationship between Sternaulopius and therein. Utetes. The synonymy between beieri and bajulus The differences noted by Wharton (1988) has undoubtedly been overlooked because between bisternaulicus and bajulus relative few specimens have been available for to hind wing RS and m-cu have not held study and the sternaulus in bajulus is up upon examination of more material, difficult to see relative to the better de- though another potentially useful character veloped sternaulus in bisternaulicus. Thus, (presence vs. absence of 2-1A) has been both Fischer (1972) and van Achterberg discovered. The ovipositor and sheath of (2004) placed bajulus in Xynobius on the bajulus are distinctly shorter than in bister- basis of more readily visible characters naulicus, and also appear to be shorter than the such as the sculptured scutellum and in duplicatus. Although the ovipositor is dorsope. Compared to other species placed very short in bajulus, the exact length could in Xynobius, bajulus has a more completely not be measured accurately on the avail- concealed labrum, with the clypeus able material. broader, flatter, truncate and not as sharply Papp (1981a) has provided some addi- margined ventrally and the mandibles are tional features for separation of bajulus (as less from flatter and strongly narrowed beieri) from rossicus. Since the holotype and venation is base to apex. The fore wing only known specimen of rossicus is male different from that of very typical Xynobius and bajulus is rare, the separation of the the of the the due to shape stigma, two species will eventually need to be postfurcal m-cu, the relatively short 1M, examined in more detail. the strongly reclivous 2RS and less strong- the ACKNOWLEDGEMENTS ly bowed 1-1A. Except for notably shorter cross-vein r, the fore of wing bajulus 1 to for am particularly grateful J. O'Connor (NMID) is identical in all to that important respects examining Haliday's type of bajulus in Dublin, as well of bisternaulicus and duplicatus. as the following curators and their assistants for extended loans and /or access to material in their care: The present study focuses on the mor- E. De Coninck, S. Hanot, and M. De Meyer (MRAC), phological differences and similarities H. Zettel and M. Fischer (NHMW), F. Koch (ZMHB), s. and amongst Biophthora, Eurytenes /., and P. C. Villemant (MNHN), J. Papp (HNHM), little to resolve their Sternaulopius, but does Marsh and D. Smith (USNM). Both R. Copeland and phylogenetic relationships, either relative G. Steck were instrumental in collection of fruits that and thanks to one another or within the Opiinae as produced the reared Sternaulopius, special to R. Copeland for isolating the puparia from a whole. Quicke et al. (1997), using venom Strombosia. This work has also benefitted from to assess relationships gland morphology discussions with C. van Achterberg about the nature a of and amongst wide variety opiine of the braconid sternaulus and G. Gibson regarding alysiine braconids, suggested a relationship thoracic musculature (though all interpretations of and associated errors are own). between beieri and various species of homology any my thanks to Jim Ehrman at the Digital Micros- Xynobius based on venom gland apparatus Special copy Facility, Mount Allison University, Sackville, of a from Denmark. The utility of specimen NB, Canada for doing all of the SEM work and to Matt the reservoir and small tripartite notably Yoder and Karl Roeder who provided much-needed glands will need to be reassessed now that assistance in acquiring and assembling the other was the classification has been considerably figures. This project supported primarily by NSF/PEET no. DEB 0328922 and in part by altered, especially since the glands of grant Initiative for Future Agriculture and Food Systems caelatus were not examined. Nevertheless, grant no. 00-52103-9651 from the USDA/ CSREFS, the of a close hypothesis relationship USAID grant no. PCE-G-00-98-0048-00 (the latter two between Biophthora and Eurytenes s. I. in collaboration with ICIPE in Kenya), and an NSF/ to the PEET cannot be rejected and is deserving of REU supplement #0616851 grant. 346 Journal of Hymenoptera Research

LITERATURE CITED derer Beruecksichtigung der Gattungen Eurytenes Foerster, Aulonotus Ashmead, Biosteres Foerster Ashmead, W. H. 1900. Some changes in generic names und der Untergattung Gastrosema Fischer (Hyme- in the Hymenoptera. The Canadian Entomologist noptera, Braconidae: Opiinae). Linzer Biologische 32: 368. Beitriige 30: 21-51. E. 1999. Belokobylskij, S. A., A. Taeger, C. van Achterberg, Fischer, M. and M. Koponen. A survey of Haeselbarth, and M. Riedel. 2003. Checklist of the Opiinae (Hymenoptera, Braconidae) of Finland, Braconidae of Germany (Hymenoptera). Beitrdge part 1. Entomologica Fennica 10: 65-93. zur Entomologie 53: 341^35. Foerster, A. 1862. Synopsis der Familien und Gattun- Bohart, R. M. and A. S. Menke. 1976. Sphecid wasps of gen der Braconen. Verhandlungen des Naturhistor- the world, a generic revision. University of Califor- ischen Vereines Preussischen der Rheinlande und nia Press, Berkeley. 695 pp. Westphalens 19: 225-288. M. G. A. P. 1985. Copeland, R. S., R. A. Wharton, Q. Luke, and De Gibson, Some pro- and mesothoracic Meyer. 2002. Indigenous hosts of Ceratitis capitata structures important for phylogenetic analysis of (Diptera: Tephritidae) in Kenya. Annals of the Hymenoptera, with a review of terms used for the Entomological Society of America 95: 672-694. structures. The Canadian Entomologist 117: Dalla Torre, C. G. 1898. 5. Subfam. Opiinae. Pp. 54-70. 1395-1443. in: Catalogus Hymenopterorum hucusque descrip- Gibson, G. A. P. 1986a. Evidence for monophyly and torum systematicus et synonymicus IV: Braconidae. relationships of Chalcidoidea, Mymaridae and G. Engelmann, Leipzig. 323 pp. Mymarommatidae (Hymenoptera: Terebrantes). Daly, H. V. 1963. Close-packed and fibrillar muscles of The Canadian Entomologist 118: 205-240. the Hymenoptera. Annals of the Entomological Gibson, G. A. P. 1986b. Mesothoracic skeletomuscu- Society of America 56: 295-306. lature and mechanics of flight and jumping in Daly, H. V. 1964. Skeleto-muscular morphogenesis of Eupelminae (Hymenoptera, Chalcidoidea: Eupel- the thorax and wings of the Apis midae). The Canadian Entomologist 118: 691-728. mellifera (Hymenoptera: Apidae). University of Gibson, G. A. P. 1993. Groundplan structure and California Publications in Entomology 39: 1-77. homology of the pleuron in Hymenoptera based Fischer, M. 1958. Die europaeischen Arten der on a comparison of the skeletomusculature of Gattung Opius Wesm. Tiel la. (Hymenoptera, (Hymenoptera) and Raphidiidae (Neu- Braconidae). Annali del Museo Civico di Storia roptera). Memoirs of the Entomological Society of Naturale di Genova 70: 33-70. Canada 165: 165-187. A. R. 1998. Fischer, M. 1965. Zwei neue Opiinen-Gattungen aus Gibson, G. P., J. D. Read, and Fairchild. dem Kongo (1. Beitrag ueber die Opiinae des Chalcid wasps (Chalcidoidea): illustrated glossa- - Musee Royal de l'Afrique Centrale) (Hymenop- ry of positional and morphological terms. URL tera, Braconidae). Revue de Zoologie et de Botanique http://res2.agr.ca/ecorc/apss/chglintr.htm. Africaines 71: 309-323. June 15, 1998. Fischer, M. 1968. Sternaulopius in Deutschland (Hy- Granger, C. 1949. Braconides de Madagascar. Mem- menoptera, Braconidae, Opiinae). Aivialen des oircs de L'Institut de Recherche Scientifique de Naturhistorischen Museum in Wien 72: 103-106. Madagascar Serie A. Biologic Animate 2: 1-428. Fischer, M. 1971a. Index of Entomophagous Insects. Haliday, A. H. 1837. Essay on parasitic Hymenoptera. World Opiinae. Le Francois, Paris. 189 pp. Entomologist's Monthly Magazine 4: 203-221. Fischer, M. 1971b. Die aethiopischen Opius-Arten der Horn, W. and S. Schenkling. 1928-1929. Index Litter- Sektion A, aufgeteilt auf die Untergattungen aturae Entomologicae. Serie 1. Die Welt-Literatur (Hymenoptera, Braconidae, Opiinae). Annalen ueber die gesamte Entomologie bis inklusive 1863. des Naturhistorischen Museum in Wien 75: 387-433. Berlin, Dahlem. 1426 pp. Fischer, M. 1972. Hymenoptera Braconidae (Opiinae Marsh, P. M. 1971. Keys to genera of the families I). Das Tierreich 91: 1-620. Braconidae, Aphidiidae, and Hybrizontidae (Hy- Fischer, M. 1977. Hymenoptera Braconidae (Opiinae menoptera). Annals of the Entomological Society of II-Amerika). Das Tierreich 96: 1-1001. America 64: 841-850. Fischer, M. 1986. Neue Bestimmungsschlussel fuer Marshall, T. 1891. A monograph of British Braconidae. palaearktische Opiinae, neue Subgenera, Rede- Part IV. Transactions of the Entomological Society of skriptionen und eine neue Art (Hymenoptera, London 1891: 7-61. - Braconidae). Annalen des Naturhistorischen Muse- Marshall, T. 1894. Les braconides (suite), 10 Tribu um in Wien 88/89: 607-662. Opiidae. Pp. 280-359 in: E. Andre, ed. Species des M. 1987. Fischer, Hymenoptera Braconidae Opiinae III Hymenopteres d'Europe et d'Algerie 5. Bouffaut - aethiopische, orientalische, australische und freres, Gray, 635 pp. oceanische Region. Das Tierreich 104: 1-734. Mason, W. R. M. 1964. The genus Chrysopophthorus Fischer, M. 1998. Neue taxonomische Untersuchungen Goidanich (Hymenoptera: Braconidae). The Cana- ueber Madenwespen der Alten Welt mit beson- dian Entomologist 96: 1005-1017. Volume 15, Number 2, 2006 347

Muesebeck, C. F. W. 1970. The Nearctic of H. 1969. species Townes, The genera of Ichneumonidae, part 1. Memoirs the Orgilus Haliday (Hymenoptera: Braconidae). of American Entomological Institute Smithsonian Contributions to Zoology 30: 1-104. 11: 1-300. Muesebeck, C. F. W. and L. Walkley. 1951. Braconidae. Van Achterberg, C. 1975. A revision of the tribus Pp. 90-184 in: C. F. W. Muesebeck, K. V. Blacini (Hymenoptera, Braconidae, Helconinae). Krombein, and H. K. Townes, eds. Hymenoptera Tijdschrift voor Entomologie 118: 159-322. of America north of Mexico. Synoptic catalog. Mono- Van Achterberg, C. 1988. Revision of the subfamilv graph No. 2. U.S. Government Printing Office, Blacinae Foerster (Hymenoptera, Braconidae). Washington D. C. 1420 pp. Zoologische Verhandelingen 249: 1-324. Van C. 1993. Papp, J. 1981a. A survey of the braconid fauna of the Achterberg, Illustrated key to the sub- Hortobagy National Park (Hymenoptera, Braco- families of the Braconidae (Hymenoptera: Ichneu- nidae), I. The Fauna of the Hortobdgx/ National Park, monoidea). Zoologische Verhandelingen 283: 1-189. 1981: 255-274. Van Achterberg, C. 1994. New morphological terms. - Ichnews 14: 5. Papp, J. 1981b. Fuerkeszdarazs-Alkatuak IX/A. Ich- Van C. 1997. neumonoidea IX/A. Fauna Hungariae 144: 1-161+. Achterberg, Revision of the Haliday D. L. K. van collection of Braconidae Zoolo- Quicke, J., Achterberg, and H. C. J. (Hymenoptera). 314: 1-115. Godfray. 1997. Comparative morphology of the gische Verhandelingen Van C. 2004. New Indo-Australian sub- venom gland and reservoir in opiine and alysiine Achterberg, and of the braconid wasps (Insecta, Hymenoptera, Braconi- genera species genera Xynobius Foerster and Ademoneuron Fischer dae). Zoologica Scripta 26: 23-50. (Hymenoptera: Braconidae: 78: Richards, O. W. 1956. Hymenoptera. Introduction and Opiinae). Zoologische Mededelingen 313-329. keys to families. Handbooks for the Identification of Van C. and A. Salvo. 1997. Reared British Insects 6 (1): 1-94. Achterberg, from M. and R. A. Opiinae (Hymenoptera: Braconidae) Argen- Sharkey, J. Wharton. 1997. Morphology tina. Zoologische Mededelingen 71: 189-214. and terminology. Pp. 19-37 in: R. A. Wharton, P. Viereck, H. L. 1916. Guide to the insects of Connecti- M. Marsh, and M. I. Sharkey, eds. Manual of the cut part III. The or Neiv World Genera of the Family Braconidae (Hyme- Hymenoptera, wasp-like insects, of Connecticut. Connecticut State Geology noptera). The International Society of Hymenop- and Natural History Survey Bulletin 22: 1-824. terists, Washington, D. C. 439 pp. Walker, F. 1860. Characters of some Shaw, M. R. and T. Huddleston. 1991. Classification apparently undescribed Ceylon insects. Annals and Magazine and biology of braconid wasps. Handbooks for the of Natural History 3 (5): 304-311. identification of British Insects 7 (11): 1-126. Wesmael, C. 1835. Monographic des Braconides de Steck, G. J., F. E. Gilstrap, R. A. Wharton, and W. G. Belgique. Nouveaux Memoires de VAcademic Royale Hart. 1986. Braconid parasitoids of Tephritidae des Sciences el Belles-lettres de Bruxelles 9: 1-252. (Diptera) infesting coffee and other fruits in west- Wharton, R. A. 1977. New World Aphaereta species central Africa. Entomophaga 31: 59-67. (Hymenoptera: Braconidae: Alysiinae), with dis- Szepligeti, G. 1901. Hymenoptera. Pp. 121-169 in: cussion of terminology used in the tribe Alysiini. Zoologische Ergebnisse der dritten Asiatischen For- Annals of the Entomological Society of America 70: des Vol. 2. schungsreise Grafen Eugen Zichy. 782-803. Szepligeti, G. 1904. Hymenoptera. Braconidae. 24. Wharton, R. A. 1986. The braconid genus Alysia Subfamily Opiinae. Genera Insectorum 22: 158-167. (Hymenoptera): a description of the subgenera Thomson, C. G. 1895. LII. Bidrag till Braconidernas and a revision of the subgenus Alysia. Systematic kannedom. (Lund) 20: Opuscula Entomologica Entomology 11: 453-504. 2141-2339. Wharton, R. A. 1987. Changes in nomenclaure and Tobias, V. I. 1977. The Wesm. genus Opius (Hyme- classification of some opiine Braconidae (Hyme- as of fruit flies noptera, Braconidae) parasites noptera). Proceedings of the Entomological Society of Obozre- (Diptera, Tephritidae). Entomologicheskoe Washington 89: 61-73. nie 56: 420-430. Wharton, R. A. 1988. Classification of the braconid V. I. 1998. 558-655 in: Tobias, Subfamily Opiinae. Pp. subfamily Opiinae (Hymenoptera). The Canadian A. P. ed. to the insects Russian Far East. Ler, Key of Entomologist 120: 333-360. 4. 3. R. P. Neuropteroidea, , Hymenoptera. Hy- Wharton, A., M. Marsh, and M. J. Sharkey, eds. Vladivostok. menoptera (part). Dal'nauka, 706 pp. 1997 Manual of the New World Genera of the Family V. I. Tobias, and A. B. Jakimavicius. 1986. Subfamily Braconidae (Hymenoptera). The International Soci- Opiinae. Pp. 7-100 in: G. S. Medvedev, ed. Keys to ety of Hymenopterists, Washington, D C. 434 pp. the insects of the European part of the USSR. 3. Yu, D. S. and C. van Achterberg. 2005. Interactive Hymenoptera. 5. Braconidae. Academia Nauk, catalogue of World Ichneumonoidea. Compact Disc. Leningrad. Taxapad, .

INSTRUCTIONS FOR AUTHORS

General The Research invites of Policy. Journal of Hymenoptera papers high scientific quality reporting research on all of comprehensive aspects Hymenoptera, including biology, behavior, ecology, systematics, and Taxonomic taxonomy, genetics, morphology papers describing single species are acceptable if the species has economic or new data on the importance provides biology or evolution of the genus or higher taxon. Manuscript should not exceed 50 length generally typed pages; however, no upper limit on length has been set for papers of and exceptional quality importance, including taxonomic monographs at generic or higher level. All papers will be reviewed at least two referees. The referees will by be chosen by the appropriate subject editor. However, it would be if authors would submit the of two helpful names persons who are competent to review the manuscript. of is The language publication English. Summaries in other languages are acceptable. The deadline for receipt of manuscripts is 1 September (for the April issue) and 1 March (for the October issue). Format and Authors are Preparation. strongly encouraged to submit manuscripts electronically to the editor at the email address and in the format below, specified below. On the upper left of the title page give name, address, telephone and fax numbers, and email address of the author to whom all correspondence is to be sent. The paper should have a concise and informative title, followed by the names and addresses of all authors. The of material should be: sequence title, author(s), abstract, text, acknowledgments, literature cited, appendix, figure legends, figure copies (each numbered and identified), tables (each numbered and with heading). Each of the should start a new title following page: (1) page, (2) abstract, (3) text, (4) literature cited, (5) figure legends, (6) footnotes. Upon final acceptance of a manuscript, the author should provide the editor with an emailed IBM formatted electronic version. CD-ROMs or 3.5 inch floppy disks are acceptable. Because symbols and tables are not always correctly translated it is best to also send a printed copy of the manuscript. Preferred word processing programs are Microsoft Word and WordPerfect. If possible, all words that must be italicized should be done so, not underscored. Tables may be formatted in a spread sheet program such as MS Works or MS Excel. Text should be double-spaced typing, with 25 mm left and right margins. Tables should be put in a separate file. CDs and Diskettes should be the accompanied by name of the software program used (e.g., WordPerfect, Microsoft Word). Authors should keep backup copies of all material sent to the Editor. The Society cannot be responsible for diskettes or text mislaid or destroyed in transit or during editing. Illustrations should be planned for reduction to the dimension of the printed page (14 x 20.5 cm, column width 6.7 mm) and allow room for legends to at the top and bottom. Do not make plates larger than 14 x 18 in. (35.5 x 46 cm). Individual figures should be mounted on a suitable drawing board or similar heavy stock. Photographs should be trimmed, grouped together and abutted when mounted. Figure numbers should be on the plate. Include title, author(s) and address(es), and illustration numbers on back of each plate. Original figures need not be sent until requested by the editor, usually after the manuscript has been accepted. Reference to figures/tables in the text should be in the style "(Fig.l)" "(Table 1)". Measurements should be in the metric system. Electronic plates may be submitted on disc, via email or uploaded to an ftp site (instructions will be given). They must be fully composited, labeled, and sized to fit the proportions of the journal page. Line art should be scanned at 1200 dpi (minimum input resolution is 600 dpi). Color or grayscale (halftone) images should have a dpi of 300-350. Color files should be in CMYK and not RGB. Graphics should be submitted as TIFF, Adobe Illustrator or EPS files. No PowerPoint or Word /Word Perfect files with images embedded in them are acceptable. All papers must conform to the International Code of Zoological Nomenclature. The first mention of a plant or animal name should include the full scientific name including the authority. Genus names should not be abbreviated at the beginning of a sentence. In taxonomic papers type specimens must be clearly designated, type depositories must be clearly indicated, and new taxa must be clearly differentiated from existing taxa by means of keys or differential diagnoses. Authors are required to deposit all type material in Recognized institutions (not private collections). Voucher specimens should be designated for specimens used in behavioral or autecological studies, and they should be deposited similarly. DNA sequences must be deposited in GenBank/EMBL/DNA Databank of Japan. Acceptance of taxonomic papers will not require use of cladistic methods; however, authors using them will be expected to specify the phylogenetic program used, including discussion of program options used. A data matrix should be provided for morphological characters. Cladograms must be hung with characters and these should include descriptors (not numbers alone) when feasible. The number of parsimonious cladograms generated should be stated and reasons given for the one adopted. Lengths and consistency indices should be provided. Adequate discussions should be given for characters, plesiomorphic conditions, and distributions of characters among outgroups when problematical. References in the text should be (Smith 1999), without a comma, or Smith (1999). Two articles by a single author should be (Smith 1999a, 1999b) or Smith (1999a, 1999b). For multiple authors, use the word "and," not the symbol "&" (Smith and Jones 1999). For papers in press, use "in press," not the expected publication date. The Literature Cited section should include all papers referred to in the paper. Journal names should be spelled out completely and in italics. Charges. Publication charges are $10.00 per printed page. At least one author of the paper must be a member of the International Society of Hymenopterists. Reprints are charged to the author and must be ordered when in are also to the returning the proofs; there are no free reprints. Author's corrections and changes proof charged author. Color plates will be billed at full cost to the author.

All manuscripts and correspondence should be addressed to:

Dr Gavin Broad Centre for Ecology & Hydrology Monks Wood, Abbots Ripton Huntingdon PE28 2LS, UK Phone: +44(0)1487 772406; Fax: +44(0)1487 773467; Email: [email protected] INSTrTUTION L SNirTHSONIAN »"*"J"

3 9088 01283 7498

CONTENTS (Continued from front cover)

YAHYA, B. E. and S. YAMANE. Nesting biology of a tropical myrmicine ant, Myrmicaria arachnoides (Formicidae), in West Java, Indonesia 307 NOTES:

MAUSS, V. Observations on flower association and mating behaviour of the pollen wasp species Celonites abbreviatus (Villers, 1789) in Greece (Hymenoptera: Vespidae, Mas- arinae) 266

ROIG-ALSINA, A. Hylaeus punctatus (Brulle) (Colletidae), a Palaearctic bee long estab- lished in South America 286