3(1 GLL- sOCIETy o '/> *, Journal of Hymenopter Research •SW\ITHS0K 1 M 0 6 twuu Volume 9, Number 1 V /to« « y April 2000 ■"" ISSN #1070-9428 CONTENTS BELOKOBYLSKIJ, S. A. and D. L. J. QU1CKE. Seven new genera of the subfamily Doryctinae (Hymenoptera: Braconidae) from the Old World Ill BRICENO G., R. A. and M. J. SHARKEY. Bassus macadamiae sp. n. (Hymenoptera: Bracon- idae: Agathidinae), parasitoid of Ecdytolopha torticornis and E. aurantianum (Lepi- doptera: Tortricidae) in macadamia nut crops in Central and South America .... 99 ENGEL, M. S. A revision of the panurgine bee genus Arhysosage (Hymenoptera: Andren- idae) 182 HERATY, J. M. and A. POLASZEK. Morphometric analysis and descriptions of selected species in the Encarsia strenua group (Hymenoptera: Aphelinidae) 142 KAZENAS, V. L. New synonyms in Central and South Asian Sphecidae (Hymenoptera) ... 48 KTMSEY, L. S. Revision of the Australian tiphiid genus Leiothynnus (Hymenoptera: Tiphiidae: Thynninae) 18 MATTHEWS, R. W. A new species of Nitela (Hymenoptera: Sphecidae: Larrinae) from Australia with notes on the nests and prey of two species 41 MAUSS, V. and A. MULLER. A study of the bionomy of the Spanish pollen wasp Ceramius hispanicus Dusmet (Hymenoptera, Vespidae, Masarinae): nesting, mating, and flower associations 1 NEWMAN, T. M. and D. L. J. QUICKE. Sperm development and ultrastructure of mature spermatozoa of Megalyra (Hymenoptera: Megalyroidea) 62 PACKER, L. The biology of Thrincohalictus prognathus (Perez) (Hymenoptera: Halictidae: Halictini) 53 PITTS, J. P. and J. V. McHUGH. Stethophotopsis, a new genus of Sphaeropthalmini (Mutillidae: Sphaeropthalminae) with a brachypterous male from Arizona (Continued on back cover) INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991 OFFICERS FOR 2000 Andrew D. Austin, President John LaSalle, President-Elect James B. Woolley, Secretary John T. Huber, Treasurer E. Eric Grissell, Editor Subject Editors Symphyta and Parasitica Aculeata Biology: Mark Shaw Biology. Sydney Cameron Systematics: Donald Quicke Systematics: Wojciech Pulawski All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, Department of Crop Protection, University of Adelaide, Glen Os- mond, Australia 5064; Secretary, Department of Entomology, Texas A&M University, College Station, Texas 77843; Treasurer, Eastern Cereal & Oilseed Research Centre, Agriculture Canada, K. W. Neatby Building, Ottawa, Ontario, Canada Kl A 0C6; Editor, Systematic Entomology Labo- ratory, USDA, % National Museum of Natural History, Washington, D.C 20560-0168. Membership. Members shall be persons who have demonstrated interest in the science of ento- mology. Annual dues for members are US$40.00 per year (US$35.00 if paid before 1 February), payable to The International Society of Hymenopterists. Requests for membership should be sent to the Treasurer (address above). Information on membership and other details of the Society may be found on the World Wide Web at http://IRIS.biosci.ohio-state.edu/ish. Journal. The journal of Hymenoptera Research is published twice a year by the International Society of Hymenopterists, % Department of Entomology, Smithsonian Institution, Washington, D.C. 20560-0168, U.S.A. Members in good standing receive the Journal. Nonmember subscriptions are $60.00 (U.S. currency) per year. The Society does not exchange its publications for those of other societies. Please see inside back cover of this issue for information regarding preparation of manuscripts. Statement of Ownership Title of Publication: Journal of Hymenoptera Research. Frequency of Issue: Twice a year. Location of Office of Publication, Business Office of Publisher and Owner: International Society of Hymenopterists, % Department of Entomology, NHB 168, Smithsonian Institution, Wash- ington, D.C. 20560, U.S.A. Editor: E. Eric Grissell, Systematic Entomology Laboratory, USDA, % National Museum of Nat- ural History, Washington, D.C. 20560-0168. Managing Editor and Known Bondholders or other Security Holders: none. This issue was mailed 5 May 2000 J. HYM. RES. Vol. 9(1), 2000, pp. 1-17 A Study of the Bionomy of the Spanish Pollen Wasp Ceramius hispanicus Dusmet (Hymenoptera, Vespidae, Masarinae): Nesting, Mating, and Flower Associations Volker Mauss and Andreas Muller (VM) Institut fur Landwirtschaftliche Zoologie und Bienenkunde, Rheinische Friedrich- Wilhelms-Universitat, Melbweg 42, D-53127 Bonn, Germany, E-mail: ULT402@uni-bonn.de; (AM) Institut fur Pflanzenwissenschaften, Angewandte Entomologie, ETH, Clausiusstrasse 25 /NW, CH-8092 Zurich, Switzerland, E-mail: andreas.mueller@ipw.agrl.ethz.ch Abstract. — Data about the bionomy of the Spanish pollen wasp species Ceramius hispanicus Dus- met are presented for the first time. Following the concept of Gess and Gess (1988) C. hispanicus can be characterized by the following ethological elements: a. Nest excavated in non-friable soil; b. Burrow surmounted by a turret from earth extracted from within the burrow; c. Nest possibly annual; d. Nest with relatively short main shaft, main shaft vertical to sub-vertical, with an ex- pansion at the bottom of the shaft; e. Main shaft not terminated by a cell; f. Secondary shafts absent; g. Cells sub-horizontal, in a group to one side, all at different depths; h. A constructed mud-cell within an excavated-cell, formed from earth excavated within the burrow; i. Main forage plants are representatives of Cistaceae, Primulaceae, Lamiaceae, and Fabaceae. The entrance turret of the nest is unique in that it is reduced to three distally converging processes. During nest excavation the females perform pellet dropping flights and use a defined pellet-dropping area. Water is used to soften the soil. Females alight on the edge of a water source to collect water. Cell construction rate during this investigation was about 0.5 cells per day. Analysis of brood cell contents revealed Ceramius hispanicus to be polylectic. Cells were found to be provisioned with pollen originating from six different plant families. The most important pollen sources were He- lianthemum (Cistaceae), Coris (Primulaceae), several species of Lamiaceae, and Lotus (Fabaceae). Pollen grains of the same plant taxa were found in crop and rectum of both male and female wasps. Males patrol and perch at water and to a lesser extent at flowers. Pairings were frequently observed at water. The daily period of activity lasted about 10 hours. Activity of males at water was high in the morning and declined during the day, while female activity increased towards the afternoon. The genus Ceramius shows a disjunct distribution with six species-groups occur- ring in the Afrotropical Region and two species-groups in the Palaearctic (Richards 1962, Gess 1992). For the Afrotropical taxa it has been shown that there are distinct differences between the species-groups in regard to nest construction and flower as- sociation (Gess 1996, Gess and Gess 1980, 1986, 1988, 1990). By contrast, the biono- my of the Palaearctic species is insuffi- ciently known (cf. Gess 1996) to make it possible at present to draw conclusions about the evolution of life history traits within the whole genus. For the Palaearc- tic Region comprehensive information is only available for Ceramius tuberculifcr Saussure (Ferton 1901, Giraud 1863, 1871, Mauss 1996a). The knowledge of the bion- omy of the remaining 13 species is very poor (cf. Fonscolombe 1835, Mauss 1998, Richards 1963), bionomical data concern- ing Ceramius hispanicus Dusmet are com- pletely lacking. Like Ceramius tuberculifcr, C. hispanicus belongs to Ceramius species- group 7 of Richards (cf. Mauss 1996b). However, these species are not very close- ly related, i.e. C. hispanicus is not a mem- Journal of Hymenoptera Research ber of the C. lusitanicus-complex (Mauss 1996b). Biogeographically C. hispanicus seems to be restricted to Central, Eastern and Southern Spain (cf. Ceballos 1956: 342, Richards 1962: 107, Mauss unpubl.). Data concerning habitat, nest architec- ture, flower visiting, male activity and mating of C. hispanicus are presented here for the first time and are compared with ethological accounts for other species of Ceramius. METHODS Investigations were carried out from 19 to 26 June 1998. Weather conditions were good throughout the whole period. Max- imum air temperatures were circa 30 °C. Sunrise took place at 4h24, sun-transit at llh56 and sunset at 19h28 (Bien in lit.). Time used is Greenwich Time. Observa- tions were made with the aid of close-up binoculars (Eschenbach Binoskop) and documented using a 35 mm camera with a 100 mm lens (scale up to 1:1) or a 300 mm lens. Activity of males and females at water was measured on 21 June (from 8h30 to 18h30) and 24 June (from 7h00 to 18h30). A rectangular sample area of 2m X lm was marked out with a string. The area was completely shaded by the ridge of a mountain after 17h30. Accuracy of count- ing was improved by use of mechanical hand-counters. Every half hour a measur- ing cycle lasting 20 minutes was carried out following the sequence: 1. Activity of females measured by counting the number of females sitting on the ground of the sample area every 30 s during a period of 300 s, leading to 10 counts per period. 2. Activity of males measured by count- ing crossings of the string into the sample area (category "males flying") during a period of 600 s and counting landings on a perch within the sample area (category "males perching"). 3. Activity of females measured again uring a period of 300s as described in I.; female activity per half hour is the sum of the 20 counts of a complete measuring cy- cle (category "females at water"). Copulations were counted during the complete 1200 s of the measuring cycle ir- respective whether they occurred inside or outside of the sample area. A situation was rated as a "copulatory attempt" when at least a short struggle on the ground could be observed after a male ap- proached a female and pounced on her. Finally, two thermometers (precision: 0.5°C) were read one of which was situ- ated 0.5 m above the ground in the shade within a juniper tree (Tair), the other one was placed on the wet ground with its point in the shadow of plants (TBmund). Short notes were made about the weather. Radiation conditions were noted applying the categories "cloudless" (sun not cov- ered during the whole period), "hazy" (sun covered by hazy clouds at least for a short period, resulting in half shade) and "cloudy" (sun covered by clouds at least for a short period, resulting in shade). The observed frequencies of the measured be- havioural categories were summed up for every hour. Then the proportion of activ- ity for each hour to total activity for the day was calculated separately for each cat- egory and expressed as a percentage. For statistical analysis, a Chi square test was performed which was calculated by Aba- cus Concepts, StatView® Student for Mac- intosh. For 21 June the period from 8h30 to 12h30 was compared with the period from 12h30 to 16h30. For 24 June the pe- riod from 6h30 to 12h30 was compared with the period from 12h30 to 18h30. For each category the observed frequencies were tested against the expected equal distribution. For nest excavation on 26 June the shaft of each nest was completely filled with Maizena® (fine maize flour) which was in- jected with a squeezing bottle. Nests were carefully excavated afterwards. Measure- ments were taken by use of small strips of graph paper, orientation of the cells was Volume 9, Number 1, 2000 measured with a bearing-compass. All nest cells were collected and stored in a refrigerator for three days. Afterwards, all cells were measured (external maximal di- mensions) using a stereo-microscope (Wild M3) with a calibrated ocular-micro- meter; then they were opened and the contents were recorded. Flowering plants in the neighbourhood of the nesting site were collected and pre- served both dried and in 70% ethanol. They are named according to Tutin et al. (1964-1980). Pollen samples from the nest and the alimentary tract of imagines fixed in Bouin's solution were prepared using the method outlined by Westrich and Schmidt (1986). The different pollen types were ascertained under a light microscope at a magnification of 400 X or 1000X and determined to the family or genus level with the aid of a reference collection con- sisting of pollen samples of 500 mainly Mediterranean plant species including those growing at the nesting site. Exact knowledge of the plant species flowering at the study site during nest provisioning in some cases allowed pollen determina- tion down to species level. The percentage of the different pollen types per brood cell was estimated by counting 50 grains at each of 30 loci distributed randomly over the cover slip. For each gut sample be- tween 100 and 250 pollen grains were counted. RESULTS Description of the Imbitnt. — A large pop- ulation of Ceramius hispanicus was local- ized in the Barranco de Zorita (GPS: 01°26.402 W 40°27.334 N), a small valley in the Sierra de Albarracin about 6 km north of Albarracin in Teruel province, sit- uated on the north-east slope of the Val- lejo Largo at an altitude of 1200 m. The narrow, steeper part of the valley was ori- entated from the east-south-east (down- stream) to the west-north-west (up- stream); at the upper end it got wider, sloped only gently and changed its direc- tion towards the southwest. A water trough which was supplied by a perpetual spring non-seasonally was situated at the upstream end of the narrow part of the valley. The water ran away from the trough into a little stream (Fig. 1) that dried out after about 100 m. Further downstream a few puddles remained at first but they dried out during the obser- vation period. Two small ponds were sit- uated about 100 m upstream of the trough, the border of which was complete- ly overgrown with rushes (J uncus sp., Jun- caceae). The whole area was covered by sparse montane forest, on calcareous soil, dominated by trees and shrubs of differ- ent junipers (Juniperus sp., Cupressaceae) (cf. Polunin and Smythies 1973) forming a Junipereto hemisphaerico-thuriferae sig- mentum (Rivas-Martinez 1986). The ground cover was about 70% in the valley but decreased markedly uphill where it became more rocky and much drier (Fig. 1). The following plant species were in flower: the Lamiaceae Nepeta nepetella L., Marrubium supinum L., Sideritis spinosa Lam. and Thymus zygis L., the Cistaceae Helianthemum apenninum (L.) Mill, and H. cinereum (Cav.) Pers., the Fabaceae Lotus corniculatus L., Coronilla minima L. and species of Ononis, Hippocrepis, Onobrychis, Medicago and Vicia, the Primulaceae Coris monspeliensis L., the Asteraceae Anacylus clavatus Pers. and Achillea sp., the Rese- daceae Reseda lutea L., the Boraginaceae Echium vulgare L., the Rosaceae Potentilla reptans L. and unidentified species of Bras- sicaceae, Cichorioideae, Convolvulaceae and Crassulaceae. The area was grazed by sheep and goats. A small cornfield ad- joined in the upper, widened part of the valley. Nest site. — An aggregation of five nests was located on a bank of hard, clayey soil mixed with some gravel. The bank was about 20 m long, 2.5 m wide and rose above the adjacent terrace by 0.4 m. It was situated about 50 m upstream of the ponds and ran from the southwest to the Journal of Hymenoptera Research Fig. 1-5. 1, Habitat <>t Ceramius hispanicus at Barranco de Zorita (Prov. Teruel, Spain) covered by sparse montane forest dominated b\ different junipers. Males patrolled along the stream in the centre where females collected water. 2, Male ol ( eramius hispanicus perching on a stone close to the water (glossa visible between slightly opened mandibles) » Female ol ( '.eramius hispanicus Standing on the wet ground at the stream during water uptake (note extended glossa). 4, Nest No. 1 oi ('animus hispanicus alter excavation on 2h June, 1998 (turret removed/ shaft filled with Maizena" ; s mam shaft; c constructed mud cell). 5, Constructed mud cells of nest No. 3 of Ccrauiius hispanicus on 29 I line, 1998 (length ol scale bar 5 mm). Volume 9, Number 1, 2000 Fig. 6-10. 6, Ceramius hispanicus female during nest excavation. The female had backed out of the entrance with the soil-pellet held in her mandibles and was about to turn round and start the pellet discard flight. The nest entrance is surmounted bv three elongated, distallv converging processes (= p; third process mainly hidden by the middle one) (see also [ig. 11)7. female of Ceramius hispanit us initiating the turret at the entrance of the main shaft (mud pellet supported bv mid-legs on the outside i. B, Female of Ceramius hispanicus placing a mud pellet on the distal end of one of the processes of the turret. 9, Copulation of Ceramius hispanicus, male and female grappling on the ground. 10, Copulation of Ceramius hispanicus; male still connected to the female by its genitalia after it had lost its hold on the thorax of the female that had tried to escape. The situation lasted for about 180 s. Journal of Hymenoptera Research Fig. 11. Schematic representation of the nest en- trance of Ceramius hispanicus (see also Fig. 6) (weakly dotted area = parts of the nest entrance which are made out of mud by the female; p = elongated, dis- tally converging processes probably homologous with the turret; length of scale bar 1 mm). northeast, gently sloping towards the lat- ter. The nesting area measured about 2 m2, its ground cover was 40-50%. The mini- mum distances between the nests varied from 0.4 to 1.2 m (median 0.7 m). Four nests were situated on the face exposed to the south-east and one was on top of the bank. Nest architecture. — All nest entrances were to some extent hidden under leaves of plants. The entrance was surmounted by a low turret which was built of mud pellets cemented together. It consisted of a short basal ring which continued into the lining of the shaft towards the base. Distally, three elongated, converging, col- umn-like processes arose obliquely from the basal ring (Figs 6, 11). The processes were about 3 mm long and were situated around that half of the entrance which was situated above the outside bend of the oblique outset of the shaft forming a three- pronged hood. Basally the processes were separated from each other by gaps of ca. 1 mm. The shaft ran obliquely downwards at an angle of ca. 45° for the first 2-3 mm (Fig. 12). Below this it continued down- wards more or less vertically for 40-60 mm, except in nest No. 2 in which the shaft descended obliquely to avoid stones (Fig. 12). This section of the shaft was 5-6 mm in diameter. The remaining part of the shaft was obliquely or vertically ori- entated and was not terminated by a cell. It widened to 9-10 mm in diameter in the region of the brood cells which lay hori- zontally to sub-horizontally and radiated out from the main shaft. Secondary shafts were lacking so that the cell openings were directly integrated into the wall of the shaft. The constructed mud cells could be easily separated from the adhering soil (Fig. 4). The cells were elongate, more or less straight and noticeably wider at mid- length than before and after it; the inner end was markedly rounded (Fig. 5). Their outer surface was irregular but more or less homogenous (Fig. 5); the inner surface was smooth but dull. Measurements of each cell and details of its contents are list- ed in Table 1. The cell provision was a firm and relatively dry pollen and nectar loaf which did not adhere to the wall. Nest-building behaviour. — Initiation of a nest by a female was observed three times. The females were flying slowly low over the ground. They interrupted their flight several times to alight on the ground which in some instances they scratched with their mandibles. At the future site of its nest, each female flew up from the ground and performed a circular orienta- tion flight, the diameter of which was about lm. Then the female alighted on the same spot again and directly started ex- cavation. Excavation was initiated by softening the soil with a liquid which was appar- ently regurgitated. A pellet of mud was formed by the mandibles; scratching movements of the fore- and mid-legs were performed in addition. When a pellet had attained about half the size of the head the female flew up with the pellet held be- tween her mandibles. She flew very rap- idly about 0.1 m above the ground to- wards an area situated between 0.3 to 0.7 m away from the nest in the immediate vicinity of a plant. At the end she hovered for a short moment, dipped down a few centimetres while dropping the pellet, re- Volume 9, Number 1, 2000 c E C c = - u 1 o — £ 0j O ■x e - °u o 0] ■5 ii a T3 E 73 4) g re g £ ll en 4-1 C £ sc u (J 73 C 5 >- £> 01 -E ^-^ ■-*- ,__ 0 o (/) s ai H > (U E r^ £ 73 ■p C ia m £ If, (/I 0 (/I O c re D. p C ra u T3 l. E ■ > C tt rs — o ^ -x c (> A o 8 Journal of Hymenoptera Research Table 1. Details pertaining to the five nests of Ceramius hispanicus excavated on 26 June, 1998 at Barranco de Zorita (measurements of cells and investigation of cell content were made on 29 June, 1998). Nest data Details of cells Cell no. Orien- tation Depth heli '\n the sur- face of the ground (mm) Exter- nal length (mm) Exter- nal width (mm) Inner diamet- er of cell open- ing (mm) Condition Nest no. Date of founding i fe- males V cells Content 1 22.06. ilh.25 1 2 I 230° 73 7 ? ? sealed small larva on pollen loaf 2 275° 77 19.4 8.4 4.0 open egg underneath pollen loaf 2 22.06. I2h47 1 2 1 30° 58 22.4 8.6 4.1 sealed small larva laterally at the bottom of pollen loaf 2 50° 64 21.0 8.2 4.3 open egg underneath pollen loaf 3 22.06. 13h45 1 4 1 245° 69 17.3 8.4 4.1 sealed larger larva below the pollen loaf 2 245° 63 20.4 8.8 3.8 sealed small larva laterally at the bottom of pollen loaf 3 300° 66 17.5 8.8 3.8 sealed small larva laterally at the bottom of pollen loaf 4 320° 72 19.8 8.2 3.8 open egg ± laterally from pollen loaf 4 23.06. 1 1 1 110° 63 20.4 8.8 4.2 open small larva laterally at the bottom of pollen loaf 5 26.06. 1 0 turned to the previous flying height again and flew rapidly back to the nest-initial. The female repeated the whole sequence of pellet formation and pellet dropping several times, in which always the same Table 2. Flower-visiting records tor males and fe- males of Ceramius hispanicus during 11 hours of ob- servation on three days at Barranco de Zorita (the number of flowers visited per plant is not taken into account). Nu mber of plants visited by Plant taxori Mali 'S 1. males Lamiaceae Nepeta nepetella 5 Marrubium supinum 4 7 Sideritis spinosa ///l/»//(s Zygis Fabaceae 1 1 Lotus cornit ulatus 7 ( iistaceae / lelianthemum apenninum 2 Helianthemum i inereutn 1 Astri.u eae ellow < n horioideae sp. 1 area was used for pellet dropping. All fe- males continued to use their individual pellet-dropping areas over the whole ob- servation period until 26 June. Once it was observed that a stone which was obvious- ly too heavy to be removed by a pellet dropping flight was carried away from the nest-initial on foot. Formation of a pellet took about 30 seconds. After a female had discarded three to five pellets she flew away from the nesting site, probably to collect water. On average females re- turned after 205 seconds (n = 6) and con- tinued to excavate the entrance of the nest. After 25 to 60 minutes of excavation shafts were sufficiently deep that an ex- cavating female was no longer visible when she had entered the nest. An exca- vating female now had to back up the shaft with a pellet held between her man- dibles. As soon as such a female had left the nest entrance she turned about 90° on her vertical axis and performed a pellet dropping flight as described before. About 60 minutes after initiation of the Volume 9, Number 1, 2000 Table 3. Pollen composition of provision from nine brood cells from four nests of Ceramius hispanicus collected on 26 June, 1998 at Barranco de Zorita (n = 1500 pollen grains/ provision). Nest l Nest ? Nesl 3 Nest 4 Cell 1 T 1 ~y l j 3 4 1 Cistaceae Helianthemum 94.3% 97.5% 61.9% 75.9% 97.7% 83.8% 64.8% 61.0% 94.9% Primulaceae Cor is 1.2% 1 .3% 12.1% 6.1% 1.6% 15.2% 33.8% 34.4% 1.0% Lamiaceae four species 0.7% 0.9% 15.0% 1 1 .8% 0.6% 1.0% 1.0% 2.4% 3.0% Fabaceae Lotus 3.8% 0.3% 9.7% 6.0% 0.7% 0.4% Convolvulaceae 0.2% 0.1% 0.1% 0.3% 0.5% 0.1% Crassulaceae 0.8% 0.1% unknown pollen 1.1% 0.1% 0.1% 0.2% 0.5% 1 100% 100% 100% 100% 100% 100% 100% 100% 100% nest a female started to build a turret. At this stage when she backed up the shaft with a soil pellet held between her man- dibles she held on to the ground around the rim of the entrance with her hind legs with her venter positioned outwards (Fig. 7). The wet soil pellet was placed on the rim of the entrance of the shaft and worked with the mouthparts from the in- side while it was supported on the outer surface with the mid-legs (Fig. 8). Finally the female entered the shaft again and the whole sequence started anew. On a few occasions the female curved the metasoma around while placing a pellet so that the pellet was obviously supported on its out- er surface by the ventral surface of the tip of the metasoma and the mid-legs. After about 15 minutes (interrupted by water collecting flights) the turret was complet- ed and the female resumed excavating the nest as before (Fig. 6). The turret of nest No. 1 was immediately rebuilt to similar design by the female after it had been ex- perimentally destroyed on 23 June. The females always entered the nest Table 4. Pollen composition of gut content of five males and five females of Ceramius hispanicus collected at Barranco de Zorita (n = 100-250 pollen grains/ individual; dbM No. = serial number in the database of V.Mauss, repeated on the determination labels). Fern. lies Males dbM No. 1419 1421 1424 1430 1432 1418 142h 1 135 1441 1442 Cistaceae Helianthemum 23% 97% 18% 65% 69% 74% 27% 40% 52% 56% Primulaceae Con's 7% 5% 25% 1% 2% Lamiaceae Sideritis 28% 3% 15% 10% 3% 16% 38% 2% 15% 3% Marrubium 1% 6% 1% 34% other taxa 26% 56% IS",, 2% 56% 2g..(l 4% Fabaceae Lotus 14% 1% Asteraceae Anthemis-type 5% IS"., 1% Coniferopsida 4% 1% 2% 4% 7% 1% 2% 2% unknown pollen 1% t • 1% 3% 2",. 1"., 1% 10 Journal of Hymenoptera Research head first. They were able to turn around ternately underneath their body. The for- inside the nest, but during excavation they etarsi were brought to the mouthparts a left the entrance always backwards. few times during the process, indicating Water collection. — Females alighted fre- pollen uptake. When on flowers of this quently at the edge of the little stream or plant species, the females were never ob- on the damp soil in its immediate vicinity served to insert their heads into the corolla to collect water. They were never ob- base. served to settle on the water surface. After The brood cells were provisioned with landing the females often walked a few a firm and relatively dry loaf composed of steps forward, stopped, extended the glos- nectar and pollen. The pollen composition sa and started to take up water, this being of all provision sampled was remarkably accompanied by vigorous pumping move- similar. All provision contained high per- ments of the metasoma (Fig. 3). During centages of pollen of Helianthemum apen- water-collection females often chose spots ninum and H. cinereum (Tab. 3); less im- on the damp soil or at the edge of the wa- portant pollen sources though well repre- ter, which were hidden by vegetation. Fe- sented in some brood cells were Coris mon- males were observed much less frequently speliensis, four different species of at the two ponds and the trough than at Lamiaceae, and Lotus corniculatiis. Pollen the stream. Females visiting the ponds of Convolvulaceae and Crassulaceae oc- stood on the blades of the rushes during curred in small amounts in some cells, water uptake; at the trough they held on Likewise, the alimentary tract of the ima- to the vertical walls a short distance above gines contained pollen grains of Helianthe- the surface of the water. mum, Coris, several species of Lamiaceae, Forage plants. — Flower-visiting records and Lotus; pollen of Asteraceae occurred for imagines are summarized in Table 2. in addition (Tab. 4). Males visited the flowers of the Lamiaceae Mating behaviour. — Males were most fre- extensively while the observed single vis- quently observed at water. They flew in its to Helianthemum cinereum and the yel- elliptic flight paths along the stream banks low Asteraceae were very short. The fe- in a slow, constant flight about 0.1 m males were observed to visit mainly white above the ground. The most striking fea- flowering Lamiaceae and Lotus comicula- ture of the flying males was the white col- tus. Three times females were observed to oration of the clypeus and the mandibles, change from one plant taxon to another which strongly contrasted with the dark during a single foraging trip, indicating coloration of the body. In addition, the an- low flower fidelity. The behavioural pat- tennae, raised at about 45° to the median tern exhibited on flowers differed remark- axis of the body, showed their conspicu- ably with the plant taxon. While visiting ously orange-marked curved distal ends, flowers of Lamiaceae the imagines insert- The patrolling males sometimes interrupt- ed the mouthparts and the distal parts of ed their flight and alighted on sun-ex- the head deeply into the corolla. On one posed stones which were situated 0.1 to 1 occasion it was seen with certainty that m (exceptionally 3 m) away from the the glossa was extended when the head stream. On the perch the males main- was removed from the flower, indicating tained a characteristic posture. Antennae nectar uptake. When females alighted on and wings were raised at about 45° to the flowers of Lotus corniculatiis the alae of the median axis of the body; the head was of- flower were pressed ventro-laterally. Si- ten slightly raised; and the mandibles multaneously, the females performed lat- were usually closed, although it was ob- I movements with the gaster and served a few times that the glossa was moved the distal parts of the forelegs al- stretched forward slightly (Fig. 2). Perch- nm Volume 9, Number 1, 2000 11 ing males occasionally rubbed the meta- soma ventrally and laterally by alternate movements of the hind-legs, or they groomed the head, the thorax and the an- tennae by alternate movements of the fore-legs. The frequency of perching and the time spent on a perch decreased dur- ing the day. In the morning, males stayed for up to 60 s on a perch whereas males alighted only for a few seconds later in the day. Interactions between males were ob- served occasionally. Two incidents were observed of two patrolling males rapidly approaching each other, falling to the ground, grappling there for a short time and finally separating and flying away. Flying males were also observed to ap- proach perching males resulting in the de- parture of the latter from the perch, fol- lowed by contact in the air, grappling on the ground, and finally separation. Copulatory attempts were frequently observed at the edge of the stream. Pa- trolling males approached females which were on the ground collecting water. Males were often observed to turn away after they had nearly reached the females but before coming into contact with them. However, they also frequently pounced on sitting females, vigorous grappling on the ground following (Fig. 9). Insertion of the male genitalia was observed with cer- tainty on three occasions although it prob- ably occurred more often. On one occasion the male lost its hold on the thorax of the female during insertion and held on to a plant while the female tried to escape. The couple was still connected by the genitalia and remained in this position for a further 180 s (Fig. 10). Often the pairs separated after a short spell (1-5 s) of grappling on the ground but some copulations lasted a few minutes, at most six. Pairs never flew off together during copulation but always separated on the ground before they de- parted independently. Males also patrolled along plants in a slow, constant flight. Patrolling males were mainly observed in the afternoon at patches of Marrubium supinum about 50 m away from the nesting aggregation. Be- tween 8h00 and 9h00 on 22 June two searching males were observed; between 14h00 and 16h00 eight records of at least six different patrolling males (marked or collected) were made and five females were observed visiting the flowers. No re- sightings occurred. Copulations were not observed but twice a male briefly ap- proached and followed a honeybee work- er (Apis mellifera L.). On 23 June one or several males were repeatedly observed patrolling over the nesting aggregation and the adjacent vegetation at 12h40. Activity pattern of males and females at wa- ter.— The results of the activity measure- ments are summarized in Figure 13. On both days females collected water after 12h30 more frequently than expected (Chi- Square test; p < 0.001). In contrast the ac- tivity of males and the frequency of cop- ulatory attempts was significantly higher before 12h30 (Chi-Square test; p < 0.001 and p s 0.01 respectively) and declined in the afternoon. Males were observed to perch more often in the morning than later in the day but this was only significant on 21 June (Chi square test; p < 0.001; p = 0.06 for 24 June). The first male appeared at the stream at 7h36, the first female at 7h43. Males were not observed after 17h20 whereas females collected water until 18h03. Associated organisms. — A female of a bee (probably Lasioglossum sp., Halictidae) was hiding in cell No. 2 of nest No. 1 on June 26. It escaped during excavation. DISCUSSION Nest construction. — All species of Cer- amius for which nesting is known con- struct a cylindrical turret surmounting the nest entrance (Gess and Gess 1988, 1992, Mauss 1996a). The presence of a turret-like structure in C. hispanicus is therefore con- sidered to be a plesiomorphic trait, al- though the shape of the turret is strongly derived. The homology of the structures is 12 Journal of Hymenoptera Research supported by their identical position at the nest entrance and the strong similarities in the behaviour of turret-construction (cf. Gess and Gess 1980). A possible function of the three converging processes of the turret of C. hispanicus may be to camou- flage the nest by disguising the contour of the entrance hole. Thereby the nest is nearly invisible to potential vertebrate predators, made more so by hiding the en- trance under leaves. The burrow of C. hispanicus differs from that of C. tuberculifer (cf. Giraud 1871, cf. Mauss 1996a) and the majority of the Af- rotropical Ceramius (Gess and Gess 1986, 1988, 1990, 1992) in that the main shaft is not terminated by a cell, a situation which is probably apomorphic. Within the ground-nesting Masarinae lack of a ter- minal cell at the end of the main shaft is only known for Ceramius lichtensteinii (Klug) (Gess and Gess 1980), Paragia tri- color Smith (Houston 1984) and Jugurtia confusa Richards (Gess and Gess 1980). A further derived character of the nest of C. hispanicus is the absence of secondary shafts which are reported to occur in all ground-nesting Masarinae for which nests with more than one cell have been found (cf. Gess 1996: 66 ff., 1999, Gess et al. 1995, Mauss 1996a). As in members of the Af- rotropical species-groups 3 and 6 the main shaft of the nest of C. hispanicus is en- larged at or near its base. In the remaining taxa of Ceramius the main shaft shows a short bulbous enlargement at mid-length (Gess and Gess 1980, 1986, 1988, 1990, 1992) or is not enlarged (Gess 1999, Mauss 1996a). The "bulb" probably allows the imagines to turn around in the shaft (Gess and Gess 1988). Lack of a defined bulb in C. hispanicus and C. tuberculifer (Mauss 1996a) may be functionally correlated with the comparatively short length of the main shaft which causes the basal turning area to be situated in a tolerable distance to the entrance. The dimensions of the burrow of C. his- panicus and C. tuberculifer (cf. Mauss 1996a) are quite similar, but they differ in that the main shaft normally descends more or less vertically in C. hispanicus whereas it descends vertically (Giraud 1871) or obliquely to sub-horizontally (Mauss 1996a) in C. tuberculifer. The dif- ferences may only be modifications relat- ed to the nature of the substrate at the nest site. In contrast to C. tuberculifer (Mauss 1996a), the main shaft of the nest of C. his- pmnicus is not terminated by a cell. Within Ceramius this condition is merely known from C. lichtensteinii (Gess and Gess 1988, Gess 1996, Gess 1999) the only member of species-group 5. The existence of con- structed mud-cells (sensu Gess and Gess 1986) which are presumably built within an excavated cell and the sub-horizontal orientation of these cells can be assumed to be plesiomorphic traits of C. hispanicus which are adopted from the ground-pat- tern (sensu Ax 1984: 156) of Ceramius. They exist in the majority of species of Ceramius and are also present in some Par- agia (cf. Houston 1984, 1986) and Jugurtia (Gess and Gess 1980, Gess 1996: 95).' Digging females of C. Jiispanicus use a clearly defined pellet-dropping area over successive days. A set pellet-dropping area is also used by C. tuberculifer (Mauss 1996a), C. rex Saussure, C. metanotalis Rich- ards, C. bicolor (Thunberg), C. capicola Brauns and C. socius Turner (Gess 1996, Gess and Gess 1980, 1988), whilst females of other Ce/w/m/s-species spread out the pellets over a larger area (Gess and Gess 1980, 1988). All species of Ceramius discard pellets in flight, with the exception of C. tuberculifer, the females of which move to the pellet-dropping area on foot (Mauss 1996a). The pellet-dropping area of C. his- panicus is situated farther away from the nest entrance than in the remaining five species using a defined pellet-dropping area (cf. Mauss 1996a, cf. Gess and Gess 1980, 1988). C. hispanicus utilizes a liquid to soften the soil in nest construction. This liquid is probably water since water is fre- quently collected by the females. Usage of Volume 9, Number 1, 2000 13 percentage of total activity 40i transit of sun : 1 lh56 y. temperature cloud cover <°q 30- 20 10 21.06. time of day 07h30 08h30 09h30 10h30 llh30 12h30 13h30 14h30 15h30 16h30 17h30 (Greenwich) females at water (n = 389) 0 males flying (n = 454) LJ males perching (n = 29) LJ copulatory attemps (n = 18) percentage of total activity 40 transit of sun : 1 lh57 y temperature cloud cover (°C) 30" t time of day ( Grt?t?nwich ) 07h30 08h30 09h30 10h30 llh30 12h30 13h30 14h30 15h30 16h30 17h30 .^ females at water (n = 2629) £2 males flying (n = 685) O males perching (n = 42) CH copulatory attemps (n = 38) Fig. 13. Activity pattern of males and females of Ccramius hispanicus on 21 |une and 24 |unc, ltH|S .it a little stream in the Barranco de Zorita (prov. Teruel, Spain). The proportion of the summed activity ot each hour to the amount of activity of the whole day (= n) is plotted against the time of the d^\ tor each category. The line chart shows the temperature changes over the day; covering of clouds is expressed by symbols (for further details see text). 14 Journal of Hymenoptera Research water in nest building has been reported are Helianthemum (Cistaceae) and Coris for all Afrotropical Ceramius and C. tuber- (Primulaceae), to a lesser extent also Lotus culifer (Gess and Gess 1980, 1986, 1988, (Fabaceae) and four species of Lamiaceae. 1990, Gess 1996, Ferton 1901), although Furthermore, pollen of Crassulaceae and the latter species may use nectar instead Convolvulaceae is used in small amounts (Mauss 1996a). During water collection and the occurrence of pollen of Asteraceae the females of C. hispanicus land on very in the alimentary tract of males and fe- wet ground or at the edge of a water males indicates that pollen of this family source. Standing on the ground during might be used for larval nourishment as water uptake occurs in species-group 2 as well. Flowers of Helianthemum exclusively well, while members of species-groups 3, provide pollen (Kugler 1970: 206), so that 4, 5, and 6 stand on the surface of the wa- the females have to take nectar from other ter (Brauns 1910, Gess and Gess 1988, plants. Nectar seems to be mainly collect- 1990, Gess 1996: 76, 1999). Group 8 is ex- ed from flowers of various Lamiaceae ceptional in showing inter- and intraspe- which is indicated by a high proportion of cific variation in water-collecting behav- flower-visiting records being for Lami- iour (Gess 1996: 76). Nothing is known aceae despite the comparatively low quan- about the functional significance of the tity of pollen of Lamiaceae present in the different water-collecting strategies. The provisions of the brood cells or in the al- observed vigorous pumping movements imentary tract. Males also consume large of the metasoma of the females during amounts of Helianthemum pollen indicat- water uptake may serve to eliminate air ing that they do not incidentally ingest from the anterior air-sacs to make some pollen during nectar uptake but actively space for the dilatation of the crop. How- feed on it. As already established by Gess ever, they may also turn out to be normal & Gess (1988) pollen analysis is the only respiratory movements after a flight. reliable method to elucidate pollen-plant Perennial re-use of nests has been re- preferences of masarine wasps, corded for most Afrotropical species Some of the remaining members of Cer- which construct mud-cells and C. tuber- amius species-group 7 are also polylectic. culifer (Gess and Gess 1988, Gess 1996, Ceramius tuberculifer consumes pollen of Mauss 1996a). Reuse of nests by C. hispan- Lamiaceae, Cistaceae and Fabaceae icus can not be excluded, but it is remark- (Mauss 1996a) and other members of the able that all the nests, examined were new- C. lusitanicus-complex have been recorded ly founded. The cell-construction rate of C. visiting flowers of Lamiaceae, Fabaceae hispanicus can be roughly estimated from and Apiaceae (Richards 1963). The main the field data. Based on the present sam- pollen source of C. tuberculifer is Teucrium pie, a female starting a new nest can be montanum L. (Lamiaceae) and the imagi- expected to construct, provision and seal nes exhibit behavioural adaptations to the 0.5 (range 0.3 to 0.9) cells per day (open nototribic pollen presentation of Lami- cells were calculated as 0.8 cells). Com- aceae (Mauss 1996a). Such behavioural ad- parable data are not available for other aptations seem to be lacking in C. hispan- ground-nesting masarines. The aerial nest- icus which uses flowers of Lamiaceae ing Celonites abbreviatus (Villers) was ob- mainly as a nectar source. Afrotropical served also to finish about 0.5 cells per species of Ceramius have not been record- day (calculated from Bellmann 1984). ed even as casual visitors of this family Forage plants.— Ceramius hispanicus is (Gess 1989, 1996) with the exception of broadly polylectic and capable of dealing Ceramius ilamarinus Turner the imagines of ith flowers of very different architecture, which use flowers of Lamiaceae as a nec- i the study area the main pollen sources tar source (Gess 1999). It still remains un- Volume 9, Number 1, 2000 15 certain whether polylecty is a character of the ground pattern of species-group 7 or if it evolved within the group, as Ceramius ductus (F.) which diverged comparatively early may be restricted to Asteraceae (Mauss 1998). In contrast, the Afrotropical species of Ceramius, like the majority of the Masarinae (Gess 1996: 46-47), are mark- edly oligolectic (Gess 1989, 1996: 41) using only pollen of Asteraceae, Aizoaceae or Fabaceae. Furthermore, preferences for single plant families are characteristic at the species-group level in Afrotropical Ceramius. Mating system. — Mate location behav- iour of males of C. hispanicus includes perching and patrolling at water-collect- ing sites and also patrolling along flowers of Lamiaceae at times when these are vis- ited by the females. This is confirmed by observations made at Valdelobus (Teruel) where males also patrolled along the edge of a stream and along Marrubium supinum (Mauss unpubl.). Multiple encounter sites are common for various aculeate Hyme- noptera (Alcock et al. 1978, Eickwort and Ginsberg 1980), however, most cases in- volve only nesting areas and flowers. Cer- amius like Paragia utilizes three potential encounter sites: nesting areas, flowers and water collecting sites (cf. Gess and Gess 1990, cf. Gess 1996: 59 ff., cf. Houston 1984, cf. Naumann and Cardale 1987). Mate-seeking both at flowers and at water has only been recorded for five Afrotrop- ical species of Ceramius (Gess and Gess 1990, Gess 1996: 61, Gess 1999). In C. his- panicus, preference for one of the encoun- ter sites seems to be correlated with time of the day. Males are most active at water in the morning, where their activity de- clines during the day, but they patrol more frequently at flowers in the after- noon. However, total activity of males is much lower at flowers than at water and copulatory attempts were only observed at the latter. Interestingly enough, the males of C. liispauicus encountered females at their main nectar sources and not at the pollen plants. Possibly the probability of a male encountering a female was higher at the nectar-plants, species of Lamiaceae, since these were more aggregated in the study area then were the major pollen- plants which were scattered. Within the patches of Lamiaceae the density of flow- ers was comparatively high, so that a male could patrol along higher numbers of flowers per unit time than in the more evenly distributed pollen plants, like e.g. Helianthemum. At water, mating occurs more frequent- ly in the morning than in the afternoon which corresponds well with the obser- vation that males appear earlier than fe- males. The decline of male activity in the afternoon is probably not the result of changes of the abiotic conditions, since fe- male activity increases during the day and the fewer males observed still behaved as before. Two reasons should be considered. First, the males may require nectar since they have depleted their energy reserves and secondly, the number of virgin fe- males may decline during the day due to the mating effort of the males. A compa- rable activity pattern is exhibited by males of the Australian masarine Paragia tricolor, which Houston (1984) observed flying around shrubs between 8h30 and 14h00 diminishing in numbers after midday. Males of C. tuberculifer were observed to patrol along flowers between 10h30 and 14h00 (Mauss 1996a). At water, the males of C. hispanicus alternate between perching and patrolling. The proportion of perching to patrolling is highest at the onset of the daily flight period and declines rapid Iv thereafter. This may be correlated with ris- ing temperature (cf. Alcock et al. 1978), but as the males are active ahead of the females it could be the arrival of the fe- males which prompts this behavioural change as well. The few incidents of grap- pling between males indicate that some kind of territoriality is involved in the mating system (cf. Eickwort and Ginsberg 1980). It is important to emphasize that the 16 Journal of Hymf.noptera Research males probably did not mistake each other for a female, since all combats started in the air. In contrast, patrolling males were never observed to approach flying fe- males, but frequently pounced on females sitting on the ground. Important visual signals which enable the males to recog- nize each other possibly come from their distinct coloration of head and antennae. During mating, the males grappled vig- orously with the females giving the im- pression of a forced copulation. This could indicate that the majority of the females had been inseminated already. On the oth- er hand it may be possible that the females test the potential fitness of a male by of- fering resistance. Knowledge about the daily period of flight activity of Ceramius is very fragmen- tary. C. hispanicus seems to be unusual in that its activity lasts for about 10 hours being from 8h00 to 18h00 and imagines are not much affected by some cloud cov- er. Females even collected water in the evening after the stream was completely shaded by the ridge of the mountains. In contrast the active period of the Afrotrop- ical species and of C. tuberculifer seems to be shorter (Gess and Gess 1980, 1990, Mauss 1996a) and their imagines are very sensitive to less favourable weather con- ditions in general and disappear quickly when the sun is obscured by cloud or a breeze gets up (Gess and Gess 1980, Mauss 1996a). Associated organisms. — Presence of fe- male halictine bees in nests of Ceramius has also been recorded for C. lusitanicus (Klug) (Mauss, unpubl.), the nests of which they were observed to appropriate. ACKNOWLEDGEMENTS We are especially grateful to L. Castro (Teruel) for his extraordinary hospitality and indispensable sup- port of V. Mauss during the field studies and useful comments on the manuscript. M. I.opiv (Diputacion Genera] de Aragon) kindly issued the required col- lection permits. R. Bien ( Astronomisehes Rechenin hint 1 leidelberg) provided accurate limes ol sunrise, sun-transit and sunset. Finally we are also very much indebted to F.W. Gess and S.K. Gess (Albany Muse- um Grahamstown) tor their critical reading of the manuscript and improvement of our English. LITERATURE CITED Alcock, J., E. M. Barrows, G. Gordh, L. J. Hubbard, L. Kirkendall, D. W. Pyle, T. L. Ponder, and F. G. Zalom. 1978. The ecology and evolution of male reproductive behavior in the bees and wasps. Zoological Journal of the Linnean Society 64: 293-326. Ax, P. 1984. Das Phylogenetische System — Systematisi- erung der lebenden Natur aufgrund Hirer Phyloge- nese. Gustav Fischer, Stuttgart. Bellmann, H. 1984. Beobachtungen zum Brutverhal- ten von Celouites abbreviates Villers (Hymenop- tera, Masaridae). Zoologischer Anzeigcr 212: 321- 328. Brauns, H. 1910. Biologisches iiber sudafrikanische Hymenopteren. Zeitschrift fiir wissenschaftliche In- sekten Biologic 6: 384-387, 445^147. Ceballos, G. 1956. Catalogo de los Himenopteros de Espaiia. Trabajos del Instituto Espanol de Entomo- logia (Ins. Esp. Entomologia C.S.I.C.): 1-553. Eickwort, G. C. and H. S. Ginsberg. 1980. Foraging and mating behavior in Apoidea. Annual Review of Entomology 25: 421-446. Ferton, C. 1901. Notes detachees sur 1'instinct des Hy- menopteres melliferes et ravisseurs avec la de- scription de quelques especes. Annales de la So- ciete Entomologique de France 70: 83-148. Fonscolombe, H. B. de 1835. Description de Ceramius fonscolombei (Latr.). Annales de la Societe Entomo- logique de France 4: 421-427, plate 10A. Gess, F. W. and S. K. Gess. 1980. Etiological studies of jugurtia confusa Richards, Ceramius capicola Brauns, C. linearis Klug and C. lichtensteinii (Klug) (Hymenoptera: Masaridae) in the eastern Cape Province of South Africa. Annals of the Cape Provincial Museums Natural History 13 (6): 63-83. Gess, F. W. and S. K. Gess. 1986. Ethological notes on Ceramius bicolor (Thunberg), C. clypeatus Rich- ards, C. nigripennis Saussure and C. socius Turner (Hymenoptera: Masaridae) in the Western Cape Province of South Africa. Annals of the Cape Pro vincial Museums Natural History 16 (7): 161-178. Gess, F. W. and S. K. Gess. 1988. A further contri- bution of the ethology of the genus Ceramius La- treille (1 lymenoptera: Masaridae) in the southern and western Cape Province of South Africa. An- nals of the Cape Provincial Museums Natural His- tory 18 (1): 1-30. Cess, F. W. and S. K. Gess. 1990. A fourth contribu- tion to the knowledge ol the ethology of the ge- nus Ceramius Latreille (1 lymenoptera: Vespoidea: Masaridae) in southern Africa. Annals of the Cape Provincial Museums Natural History IS (9): 183- 202. Volume 9, Number 1, 2000 17 Gess, F. W. and S. K. Gess. 1992. Ethology of three Southern African ground nesting Masarinae, two Celonites species and a silk-spinning Quartinia species, with a discussion of nesting by the sub- family as a whole (Hymenoptera: Vespidae). Journal of Hymenoptera Research 1: 145-155. Gess, F. W., S. K. Gess, and R. W. Gess. 1995. An Australian Masarine, Rolandia angulata (Richards) (Hymenoptera: Vespidae): Nesting and evalua- tion of association with Goodenia (Goodeniaceae). journal of Hymenoptera Research 4: 25-32. Gess, S. K. 1992. Biogeography of the masarine wasps (Hymenoptera: Vespidae: Masarinae), with par- ticular emphasis on the southern African taxa and on correlations between masarine and forage plant distributions, journal of Biogeography 19: 491-503. Gess, S. K. 1996. The pollen zvasps — Ecology and natural history of the Masarinae. Harvard University Press. Cambridge, Massachusetts. Gess, S. K. 1999. Distribution and ethology of Cer- ainius dainariuus Turner (Hymenoptera: Vespi- dae: Masarinae) in Namibia. University of Kansas Natural History Museum Special Publications 24: 18-25. Gess, S. K. and F. W. Gess. 1989. Flower visiting by masarid wasps in southern Africa (Hymenop- tera: Vespoidea: Masaridae). Annals of the Cape Provincial Museums Natural History 18 (5): 95-134. Giraud, J. 1863. Hymenopteres recueillis aux environs de a Suse, en Piemont, et dans le departement des Hautes-Alpes, en France et description de quinze especes nouvelles. Verhandlungen der kais- erlich-koniglichen zoologisch-botanischen Gesellschaft in Wien 13: 11-46. Giraud, J. 1871. Note sur les moeurs du Ceramius lu- sitanicus Klug (Division Paraceramius Saussure). Annates de la Societe Entomologique de France: 375- 379. Houston, T. F. 1984. Bionomics of a pollen-collecting wasp, Paragia tricolor (Hymenoptera: Masarinae), in Western Australia. Records of the Western Aus- tralia)! Museum 11: 141-151. Houston, T. F. 1986. Biological notes on the pollen wasp Paragia (Cygnaea) vespiformis (Hymenop- tera: Vespidae: Masarinae) with description of a nest. Australian Entomological Magazine 12 (6): 115-118. Kugler, H. 1970. Blutenokologie. Gustav Fischer, Stutt- gart. Mauss, V. 1996a. Contribution to the bionomics of Ceramius tuberculifer Saussure (Hymenoptera. Vespidae, Masarinae). journal of Hymenoptera Re- search 5: 22-37. Mauss, V. 1996b. Morphological characters of Cer- amius palaestinensis (Giordani Soika 1957) Gus- enleitner 1992 (Hymenoptera, Vespidae, Masari- nae) and considerations about its phvlogenetic position. Einzer biologische Beitrage 28 (2): 953- 966. Mauss, V. 1998. The identity and distribution of Cer- amius auctus (Fabricius, 1804) Vecht, 1970 (Cer- amius spiricornis Saussure, 1854 syn. now) and Ceramius heaumonti (Giordani Soika, 1957) Rich- ards, 1962 (Hymenoptera, Vespidae, Masarinae). Annates de la Societe Entomologique de France 34 (2): 163-183. Naumann, I. D. and J. C. Cardale. 1987. Notes on the behaviour and nests of an Australian masarid wasp Paragia (Paragia) decipiens decipiens Shuck- ard (Hymenoptera: Vespoidea: Masaridae). Aus- tralian Entomological Magazine 13 (5,6): 59-65. Polunin, O. and B. E. Smythies. 1973. Flowers of South- west Europe. Oxford University, London. Richards, O. W. 1962. A revisional study of the masarid zvasps (Hymenoptera, Vespoidea). British Museum (Natural History), London. Richards, O. W. 1963. New species of Ceramius La- treille (Hymenoptera, Vespoidea) allied to Cer- amius lusitanicus Klug. Zoologische Mededelingen 38 (13): 213-220. Rivas-Martinez, S. 1986. Mapa de las series de vegetacisn de Espaqa No. 14. Servicio de Publicaciones del Ministerio de Agricultura, Pesca y AlimentaciMi. Madrid. Tutin, T. G., V. H. Hevwood, N. A. Burges, D. H. Valentine, S. M. Walters and D. A. Webb. 1964- 1980. Flora Europaea. Volumes 1-5. Cambridge University, Cambridge, England. Westrich, P. and K. Schmidt. 1986. Methoden und Anwendungsgebiete der Pollenanalyse bei Wild- bienen (Hymenoptera, Apoidea). Linzer biologis- che Beitrage 18 (2): 341-360. J. HYM. RES. Vol. 9(1), 2000, pp. 18-28 Revision of the Australian Tiphiid Genus Leiothynnus (Hymenoptera: Tiphiidae: Thynninae) Lynn S. Kimsey Bohart Museum of Entomology, Department of Entomology, University of California, Davis, CA 95616, USA Abstract. — The Australian thyronine genus Leiothynnus Turner is revised, describing five new species from Queensland, including cardalae, ferricolus, linnis, multimaculatus and ochrotarsus, and two previously described species, mackayensis (Turner) and spinigerus Turner, also from Queensland. Distribution maps of species and a key to males are given. Most Australian genera of Thynninae with Agriomyia Guerin de Meneville, in- are widespread, occurring in many Aus- eluding the flat male face (Fig. 5), with ex- tralian states, with some even extending treme reduction of the antennal lobes, and up into New Guinea, New Caledonia and simple male epipygium, lacking the well- adjacent islands. However, there are some developed transverse carina or ridge typ- small genera with relatively restricted dis- ical of Epactiothynnus and related genera, tributions, such as Oncorhinothynnus, However, remnants of this ridge can be which is only found in west central West- seen, particularly in cardalae and ochrotar- ern Australia, or Gymnothynnus Turner, sus. Leiothynnus can be distinguished from from Northern Territory and western these and other genera by the long brush Queensland. The genus Leiothynnus Turn- of setae on the base of the male premen- er is one of these relatively localized turn, the darkly stained marginal cell in groups. All Leiothynnus species have been the forewing, and slender, petiolate male collected from eastern Queensland. In ad- abdomen. dition to the two described species, mack- Leiothynnus species differ from one an- ayensis (Turner) and spinigerus Turner, other in modifications of the male legs, ab- there are five new ones, which are de- dominal apex and to some extent color, scribed below. All of the species treated below have the The specific relationships between male thorax with dense, nearly contiguous Leiothynnus and other thynnine genera are punctation, and the propodeal punctation not fully resolved. Leiothynnus shares a obscured by fine shagreening. These fea- number of features with Iswaroides Ash- tures may or may not be significant at the mead and Epactiothynnus Turner, includ- species level. Collecting seems to be too ing in the male the well-developed and patchy to say whether or not there are ad- somewhat arcuate penis valves, volsella ditional undescribed species. Too few fe- U-shaped in cross-section, stipes arcuate males have been collected to generalize with long marginal fringe of setae, vertex about diagnostic features among the spe- with small reddish spot behind the dorsal cies in females, although there appear to eye margin, and a large oral plate. In the be differences in the overall shape of the female, the pygidium has two submedial head and development of carinae or lobes longitudinal carinae, subtended by a long on abdominal segment V. tuft of setae. A few characters are shared Specimens were obtained from the fol- yHWHUmi^HHHIIIIIflBDHHIBIMUaMHHWBnHiHaHl Volume 9, Number 1, 2000 19 2. linnis 3. cardalae 6. ochrotarsus 7. mackayensis 8. spinigerus 9 cardalae Figs. 1-9. Leiothynnus species. 1-3, Front view ol female face, antennae removed. 4, Front view of male face, right antenna removed. 5, lateral view of male face. 6-9, Dorsal view of male body showing color pattern, white = yellow, black = black, stippled = orange to red; wings removed. 20 Journal of Hymenoptera Research lowing institutions and individuals: The Australian National Insect Collection, CSI- RO, Canberra, ACT, J. Cardale (CANBER- RA); the Natural History Museum, Lon- don, England, S. Lewis; Queensland Mu- seum, Brisbane, Australia, C. Burwell (BRISBANE-QM), and University of Queensland, Insect Collection, Brisbane, Australia, G. Daniels (BRISBANE-UQIC). Some paratypes will be deposited in the Bohart Museum of Entomology, Univer- sity of California, Davis (DAVIS). The types of both previously described species were studied. Leiothynnus cardalae Kimsey, new species (Figs. 3, 4, 9, 16, 18, 21, 27) Male. — Body length 7.5 mm. Head: face (Fig. 4) with dense small, nearly contigu- ous punctures; flagellomere I length 1.5X breadth; flagellomeres II-III 2.5 X as long as broad. Thorax: punctures dense and nearly contiguous; propodeal punctures obscured by fine dense shagreening; scro- bal sulcus strongly U-shaped, ventral loop sharply defined; foretrochanter convex in cross section; forefemur evenly convex ba- sally, without carina, knob or other mod- ification; midtrochanter and femur un- modified; midcoxal inner margin narrow- ly rounded, not angulate. Abdomen: hy- popygial apex broadly rounded with short medial projection (projection often weakly sclerotized or translucent in some individ- uals) (Fig. 21). Genitalia: paramere arcuate, broadly rounded apically, broadest sub- apically (as in Figs. 25, 26). Color (Fig. 9): body black, with yellow and orange mark- ings: head black with yellow clypeus, yel- low along inner eye margin, between and across antenna 1 lobes; scape pedicel and flagellomeres I-V red; rest of flagellum dark brown; pronotum black with yellow transverse anterior and marginal posterior bands; mesopleuron with subalar yellow spot and large yellow spot above midcoxa; one specimen with small yellow scutal spot; scutellum and metanotum with large medial yellow spot; propodeum with large lateral yellow marks, narrowly sep- arated medially; trochanter and femora reddish brown, femoral apices sometimes paler, tibiae red, tarsi yellow to cream-col- ored; tergum I orange with transverse subapical yellow band, narrowed medial- ly; tergum II yellow with narrow dark brown bands along anterior and posterior margins; tergum III-IV entirely black; ter- gum V-VI yellow anteriorly with narrow blackish posterior band, or black with yel- low lateral spot; tergum VII dark brown to black; sternum I blackish basally, or- ange apically with yellowish middle; ster- num II orange; sterna III- VI black. Female. — Body length 4 mm. Head (Fig. 3): broader than long, vertex somewhat concave, with long seta above each eye; genal area above mandible with carina ex- tending the length of eye. Thorax (Fig. 16): pronotal disk abruptly elevated above col- lar, disk with longitudinal medial sulcus, anterior margin with six long hairs; scu- tellum about twice as broad as long, pro- podeum strongly convex dorsally and el- evated above scutellum, bulging medially and flattened sublaterally, medial bulge with long erect hairs. Abdomen: tergum I with broadly W-shaped transverse sulcus; tergum II with four large transverse ridg- es; tergum V apicomedially emarginate; tergum VI with narrow parallel-sided ca- rina-edged medial plate, apically subtend- ed by short dense tuft of setae (Fig. 18); sternum VI apicomedially emarginate and thin-edged, with long brush of setae lat- erally and shorter brush apicomedially. Color: dark reddish brown. Type material. — Holotype 6: Queens- land, Luster Creek, 8 km nw Mt. Molloy, 21-22 May 1980, I. Naumann and J. Car- dale (CANBERRA). Seven paratypes— 3 6 6: Mt. Webb National Park, 15.04°S 145.07°E, 20-27 April 1981, I. Naumann; 1 6, one 9: Shipton's Flat, 15.47°S 145.07°E, 16-18 May 1981, I Naumann; 1 6: Coen, 13.57°S 143.12°E, 13 Jan.-25 Feb. 1994, mal- aise trap, Zborowski and McKay; 1 6 : Tol- Volume 9, Number 1, 2000 21 21. cardalae 22 //'nn/'s 24 //'nn/'s 25 //nn/'s 26. spinigerus Figs. 10-26. Leiothynnus Figs. 10, 11, Lateral view of male forefemur and trochanter. 12-14, Lateral view ot inner surface of male forefemur and trochanter. 15-17, Lateral view of female thorax, legs removed. 18-20, Posterior view of female pygidium. 21-23, Dorsal view of male epipygium and hypopygium. 24. Ventral view of female apical abdominal sternum. 25, Dorsal view of male genital capsule. 26, Lateral \ iew of male genital capsule. 22 Journal of Hymenoptera Research ▲ cardalae • linnis Ik multimaculatus □ mackayensis O spinigerus 0 ferricolus I ig. 27. Distribution map ol six species of Leiothynnus in Australia. Volume 9, Number 1, 2000 23 ga, 3 Mar. 1964, R. Elder (BRISBANE- UQIC CANBERRA, DAVIS). Distribution. — Fig. 27. Etymology. — This species is named in honor of Jo Cardale, who was one of the collectors of the holotype specimen. She also made much of this study possible, with collections support and encourage- ment overall. Discussion. — Leiothynnus cardalae is a brightly colored species in the male, char- acterized by having a reddish brown to orange petiole. This coloration is shared with ferricolus and multimaculatus. Male cardalae can be distinguished by the un- modified fore- and midfemur (a character- istic shared with ochrotarsus), red legs and bicolored antenna. The female of cardalae has the least modified head of species where females are known. The female tho- rax is distinctive with a sparse row of long hairs along the anterior margin of the pronotum, strongly dorsally arched pro- podeum, and pygidium with lateral cari- nae parallel or converging slightly sub- medially, with a small lateral lobe on ter- gum VI. Leiothynnus ferricolus Kimsey, new species (Figs. 10, 12, 27) Male. — Body length 7.5-8.0 mm; punc- tation as in cardalae. Head: flagellomere II 2.5 X as long as broad; flagellomere III length 3x breadth. Thorax: scrobal sulcus with ventral loop poorly defined; foretro- chanter strongly flattened and concave (Fig. 12); forefemur with short basoventral longitudinal carina, basolateral angle ex- panded and flattened ventrally (Fig. 12); midfemur with long basoventral tooth, tooth nearly as long as tarsal claw (Fig. 10), adjacent surface of trochanter flat- tened; midcoxal inner margin sharp- edged and angulate. Abdomen: hypopygial apex angulate laterally with long medial projection. Genitalia: as in Figs. 25, 26. Col- or: black with yellow and orange mark- ings; clypeus yellow, interantennal area yellow; inner eye margin with broad yel- low stripe; mandible red with basal yel- low spot; pronotum with broad transverse yellow anterior and posterior stripes; me- sopleuron with large yellow subalar and supracoxal spots; scutellum and metano- tum with large yellow medial spot; pro- podeum with large lateral yellow spots, narrowly separated medially; hindcoxa with yellow dorsal spot; trochanter and femora dark brown, femoral apices some- times paler, orange; fore and midtibiae or- ange; hindtibia dark brown or orange; tar- si yellowish brown; wing membrane yel- low stained; abdominal segment I black basally, becoming orange for most of length, with transverse yellow band or yellow spots; tergum II brown basally and apically with broad medially yellow band; sternum II brown with small lateral vel- J low spot; terga III, and in most specimens IV, black with small lateral yellow spot; terga V-VI black with large lateral yellow spot, spots sometimes convergent medi- ally; tergum VII black to dark brown; ster- na III- VII black. Female. — Unknown. Type material. — Holotype 6: Queens- land, Cape York Peninsula, Iron Range, Apr. 7-May 4, 1973, S. R. Monteith (CAN- BERRA). Nine paratype 6 6: same data as holotype; 1 6: Mcllwrath Range, 8 km ne Coen, 13°53.30S 143°15.21E, 13 Jan. 1994, G. & A. Daniels and R. Eastwood (BRIS- BANE-UQIC, CANBERRA, DAVIS). Distribution. — Fig. 28. Etymology. — This species is named after its collection locality the Iron Range; ferrus = iron, icolus = loving, Latin, masculine. Discussion. — The most distinctive fea- tures of this species in the male are the cupped foretrochanter, basally carinate forefemur and long midfemoral tooth. Male coloration is very similar to that of cardalae, as discussed under that species. Leiothynnus linnis Kimsey, new species (Figs. 2, 17, 20, 22, 24, 25, 27) Male. — Body length 8-10 mm; puncta- tion as in cardalae. Head: flagellomere I 24 Journal of Hymenoptera Research length 1.5 X as long as broad; flagello- meres II-III 2.2-2.4 X as long as broad. Tho- rax: scrobal sulcus U-shaped, ventral part weakly defined; foretrochanter convex in cross section; forefemur evenly convex ba- sally, without carina, knob or other mod- ification; midtrochanter unmodified, mid- femur projecting basoventrally in right an- gle; mid- and hindcoxal inner margins broadly rounded, not angulate. Abdomen: hypopygial apex subtruncate, with strong medial projection (Fig. 22). Genitalia (Fig. 25): paramere arcuate, broadly rounded apically, broadest subapically. Color: body black with yellow markings: head black with yellow clypeus, yellow along inner eye margin and between and across anten- nal lobes; scape, pedicel and basal flagel- lomeres red; rest of flagellum dark brown; pronotum black, with yellow transverse anterior band and band along posterior margin; scutum with short yellow band adjacent to tegula; mesopleuron with sub- alar yellow spot and small yellow spot above midcoxa; mesopleural lamellae translucent with yellow margins; scutel- lum and metanotum with large medial yellow spot; propodeum with large lateral yellow marks, narrowly separated medi- ally; trochanter and femoral base dark brown, femoral apices and rest of legs red- dish orange; tergum I orange with trans- verse subapical yellow marks becoming darker basomedially, tergum II brown with broad transverse medial yellow band; tergum III-IV black without small yellow spot laterally; tergum V-VI yellow anteriorly, with narrow blackish posterior band, or black with yellow lateral spot (tergum V may also be entirely black); ter- gum VII brown becoming paler apically. Female. — Body length 5-7 mm. Head (Fig. 2): slightly broader than long, strong- ly indented laterally above eye; vertex somewhat concave medially; genal area between mandible and oral fossa deeply longitudinally sulcate. Thorax (Fig. 17): pronotal disk abruptly elevated above col- lar, concave anteriorly; scutellum about twice as broad as long, propodeum broad, slightly convex dorsally, nearly planar be- tween petiolar socket and scutellum, strongly rounded laterally in dorsal view. Abdomen: tergum I with broadly W- shaped transverse sulcus; tergum II with four large transverse ridges; tergum V ap- icomedially emarginate; tergum VI with narrow parallel-sided carina-edged medi- al plate with lateral upturned flanges, api- cal part subtended by long dense tuft of setae (Fig. 20); sternum VI apicomedially notched, separated into two halves, by long ventral groove, with long brush of se- tae laterally, and shorter brush apicome- dially (Fig. 24). Color: dark reddish brown. Type material. — Holotype 6: Queens- land, Brisbane, March (BRISBANE). Para- types: 29 6 6, 8 9 9, same data as holo- type; 1 6, 8.5 km sse Dayboro, 27°16S 152°52E (BRISBANE, DAVIS). Distribution. — Fig. 27. Etymology. — The species name, linnis, is a nonsense combination of letters and is assumed to be masculine. Discussion. — The color and general ap- pearance of this species are similar to car- dalae. However, linnis can be immediately distinguished in the male by the unmod- ified forefemur and small basoventral an- gle on the midtibia, and in the female by the peculiarly modified head and ventral- ly divided apical abdominal sternum. Leiothynnus mackayensis (Turner) (Figs. 7, 14, 27) Thynnus mackayensis Turner 1908:123. Lectotype male (designated by Kimsey & Brown 1993); Australia: Qld., MacKay (LONDON). Male. — Body length 9-10 mm; puncta- tion as in cardalae. Head: flagellomere I 1.5X as long as broad; flagellomere II 2.5 X as long as broad; flagellomere III 3x as long as broad. Thorax: scrobal sulcus U- shaped, ventral segment well-defined; foretrochanter convex in cross section; forefemur deeply cupped ventrobasally (Fig. 14); midtrochanter unmodified, mid- Volume 9, Number 1, 2000 25 femur with basoventral tooth, nearly as long as tarsal claw; midtrochanter unmod- ified; mid- and hindcoxal inner margins broadly rounded, not angulate. Abdomen: hypopygium strongly exserted, apex api- cally rounded, with strongly sclerotized medial tooth. Genitalia: paramere arcuate, rounded apically (as in Figs. 25, 26). Color (Fig. 7): body black with yellow markings: head black with yellow clypeus, yellow along inner eye margin and between and across antennal lobes; mandible yellow, becoming reddish brown apically; scape, pedicel and flagellomere I red; rest of fla- gellum dark brown; pronotum with yel- low transverse anterior band and posteri- or lobe adjacent to tegula yellow; meso- pleuron with subalar yellow spot and large yellow spot above midcoxa; scutel- lum and metanotum with large medial and smaller lateral yellow marks; propo- deum with large lateral yellow marks, narrowly separated medially; trochanters and most of femora dark brown, femoral apices and rest of legs reddish brown; ter- ga I and III-IV black with small apicola- teral yellow spot; tergum II with large lat- eral yellow spots nearly meeting medially; terga V-VII black; sterna black, although sternum II often with small yellow apico- late-ral spot; wing membrane yellow-tint- ed. Female. — Unknown. Material examined. — Australia: Qld, Dunk Is. and Mackay; 5 6 6 were seen in- cluding the lectotype. Distribution. — Fig. 27. Discussion. — Male mackayensis share their dark coloration with spinigerus, al- though unlike spinigerus this species has a yellow band across the propodeum and tergum II. L. mackayensis males can be readily distinguished from spinigerus and other species by the combination of the basally cupped forefemur (shared with multimaculatus), and strongly dentate mid- femur (shared with ferricolus and spinige- rus). Leiothynnus multimaculatus Kimsey, new species (Figs. 11, 27) Male. — Body length 11 mm; punctation as in cardalae. Head: flagellomere I 1.5X as long as broad; flagellomere II 2.2X as long as broad; flagellomere III 2.5 X as long as broad. Thorax: scrobal sulcus U-shaped with ventral part weakly defined; forefe- mur with deep U-shaped basoventral im- pression, with associated longitudinal ca- rina (similar to Fig. 12); foretrochanter un- modified; midfemur with short basoven- tral tooth, one half or less as long as tarsal claw (Fig. 11), adjacent area on trochanter flattened; mid and hindcoxae without sharp inner margin, evenly rounded not angulate. Abdomen: hypopygium slightly flattened apically on either side of medial spine. Genitalia: as in Figs. 25, 26. Color: black with yellow and orange markings; clypeus mostly yellow; antennal lobes and subantennal sclerite yellow; inner eye margin with broad yellow band; mandible red with yellow basomedially; postocular margin with short yellow band; scape, pedicel and ventral surface of flagello- meres I-IV paler red; rest of flagellum black; pronotum with yellow band along posterior margin; mesopleuron with large yellow subalar spot; mesopleural lamella with whitish stripe along inner margin; scutellum and metanotum with large me- dial yellow spot; propodeum with large lateral spots, broadly separated medially; coxae black becoming reddish dorsally on mid and hindlegs; femora dark brown to black, becoming red apically; tibiae and tarsi red (except hindtibia darkened sub- apically in holotype); terga III-IV black without lateral yellow spots; tergum VI black with small irregular medial yellow spot; tergum VII black; sternum I orange; sternum II orange becoming darker api- cally with yellowish lateral spot; sterna III- VI black Female. — Unknown. Type material. — Holotype 6: Queens- 26 Journal of Hymenoptera Research land, Rockpool Gorge, Bluff Range, near Biggenden, 4 Oct. 1976, H. Frauca (CAN- BERRA). Paratype o\ Bluff Range, Biggen- den, 9-20 Dec. 1972, H. Frauca (CANBER- RA). Distribution. — Fig. 27. Etymology. — The species name is de- rived from the multicolored male; multus = many, maculatus = marks, Latin, mas- culine. Discussion. — The most distinctive fea- tures of this species in the male are the ventrally cupped forefemur, small basal midfemoral tooth and associated indenta- tion on the midcoxa, red legs and red bas- al abdominal segments. L. multimaculatus most closely resembles mackayensis but can be distinguished by the more extensive yellow and orange coloration and smaller midfemoral angle (as in Fig. 11). Leiothynnus ochrotarsus Kimsey, new species (Figs. 5, 6, 23) Male. — Body length 11 mm; punctation as in cardalae. Head (Fig. 5): flagellomere I 1.5X as long as broad; flagellomere II length twice breadth; flagellomere III 2.3 X as long as broad. Thorax: foretrochanter convex in cross-section; forefemur basally indented with short longitudinal carina; mesopleuron with ventral part of scrobal sulcus obsolescent; midfemur basoventral- ly with slight angle or unmodified; mid- trochanter unmodified; mid- and hindcox- ae inner margins broadly rounded. Abdo- men: epipygium with well-developed sub- apical transverse ridge; hypopygium broadly triangular apically (Fig. 23). Gen- italia: as in Figs. 25, 26. Color (Fig. 6): head yellow, except frons medially, mandibular apex and occiput black; thorax yellowish orange, except anterior face of pronotum medially black, scutum black between no- tauli surrounding large medial yellow spot, mesopleural venter and propodeal base black; legs orange to yellow except base of coxae blackish; mesopleural la- mellae translucent with yellow or whitish margins; abdominal segment I orange be- coming yellow laterally; abdominal seg- ment II yellow with narrow dark brown posterior band; abdominal segment III black; tergum IV yellow, basally and api- cally with narrow black band; terga V-VII yellow with narrow basal black band; ster- na IV-VII black with yellow lateral spot; parameres yellow; wing membrane yel- low-tinted, except marginal cell darker, brownish. Female. — Unknown. Type material. — Holotype 6: Queens- land, 30 km w Collinsville, 12 Sept. 1950, E. F. Riek (CANBERRA). Etymology. — orchros = yellow; tarsus = legs, Greek, masculine. Discussion. — Male ochrotarsus can be dis- tinguished by the bright coloration, fore- femur with basal depression and associ- ated longitudinal carina and unmodified midfemur. The apically triangular hypo- pygium may or may not be diagnostic as the holotype is an old individual with highly worn mandibles and the hypo- pygial apex may also be worn. Leiothynnus spinigerus Turner (Figs. 1, 8, 13, 15, 19, 26, 27) Leiothynnus spinigerus Turner 1912:534. Lecto- type male (designated by Kimsey & Brown 1993); Australia: Stradbroke Is., Moreton Bay (LONDON). Male. — Body length 9-12 mm; puncta- tion as in cardalae. Head: flagellomere 1 1.3- 1.4X as long as broad; flagellomere II twice as long as broad; flagellomere III 2.4 X as long as broad; punctation as in cardalae. Thorax: mesopleuron with ventral part of scrobal sulcus obsolescent; forefe- mur with basoventral knob or swelling, separated from longitudinal carina by in- dentation (Fig. 13); forecoxa convex in cross section; midfemur with long baso- ventral tooth, nearly as long as tarsal claw; midtrochanter flattened adjacent to femo- ral tooth; mid and hindcoxae inner mar- gins rounded. Abdomen: hypopygium Volume 9, Number 1, 2000 27 strongly exserted, parallel-sided, apex rounded with long medial tooth. Genitalia: as in Figs. 25, 26. Color (Fig. 8): black, with yellow and red markings; head black with yellow band along inner eye margin, clyp- eal apical margin yellow, antennal lobes yellow; scape and pedicel reddish; flagel- lum dark brown to black; pronotum with short transverse anterior yellow band, band sometimes ending at lateral notch; mesopleuron may have subalar yellow spot; scutellum in some specimens with small medial yellow spot; mesopleural la- mella translucent with white or yellow margins; metanotum with yellow medial spot; meso- and metapleuron with small pale spot above coxae; coxae, trochanters and midfemoral base and hindfemur and hindtibia brown, rest of legs red; abdomen black except small lateral spot on tergum I-II or I-IV; wing membrane yellow-tinted, becoming brownish in marginal cell. Female. — Head (Fig. 1): vertex dorsally convergent and angulate, not evenly rounded; gena evenly rounded, without sulci or grooves; posterior margin behind eyes strongly convex in front view; clyp- eus narrowly truncate apicomedially; mandible slender, broadest basally, eden- tate. Thorax (Fig. 15): pronotal disk broad- ly quadrate, with scattered erect setae of irregular lengths, particularly along ante- rior margin; propleuron with ventral tuft of long setae on either side; scutellum about as long as broad; propodeum with broadly convex dorsal surface, planar with scutellum, flattened posteriorly, par- allel-sided in posterior view, with erect se- tae particularly laterally. Abdomen: tergum V apicomedially emarginate; pygidium with lateral carinae parallel-sided or di- verging medially, subtended laterally by short flange and long tuft of setae; ster- num VI apex hoof-like (Fig. 19). Material examined. — Australia: Qld, Stradbroke Is., Brisbane, Bundaberg, and Bribie Is.; 17 6 6 and 3 9 9 were examined including the lectotype. Distribution. — Fig. 27. Discussion. — The coloration of male spi- nigerus is similar to that of mackayensis, but without the broad yellow bands or stripes seen in that species. Other diagnostic fea- tures of male spinigerus are the unmodi- fied foretrochanter, forefemur with basal knob and longitudinal carina, and long midfemoral tooth. Females have a distinc- tively narrowed vertex, tufted propleuron, and pygidium subtended by a short flange. KEY TO MALES OF THE SPECIES LEIOTHYNNUS 1 Forefemur basoventrally convex, without depression or ridge 2 - Forefemur with basoventral depression, often accompanied by short longitudinal ridge (as in Figs. 12-14) 3 2 Midfemur with small basoventral angle (as in Fig. 11); flagellum monochrome, brown or red littnis Kimsey, new species - Midfemur without basoventral angle; flagellum bicolored red and black (or dark brown) cardalae Kimsey, new species 3 Forefemur with distinct longitudinal basoventral ridge and associated depression (as in Fig. 12) 4 - Forefemur with basoventral cuplike depression without longitudinal ridge (as in Fig. 14) 5 4 Foretrochanter strongly concave or cuplike in cross-section (Fig. 12); forefemur without knob or swelling adjacent to longitudinal basal ridge (Fig. 12); midcoxa inner margin an- gulate and sharp-edged ferricolus Kimsey, new species - Foretrochanter convex in cross-section, unmodified; forefemur with knob or swelling ad- jacent to longitudinal basal ridge (Fig. 13); midcoxa unmodified spinigerus Turner 28 Journal of Hymenoptera Research 5 Forefemur with large basoventral tooth, tooth nearly as long as tarsal claw; petiole (basal two abdominal segments) black; tarsi red mackayensis (Turner) - Forefemur with short basoventral tooth, less than half as long as tarsal claw, or basoven- trally rounded without tooth or angle; petiole primarily yellow or orange; tarsi pale yellow, red or brown 6 6 Midfemur unmodified; tarsi pale yellow ochrotarsus Kimsey, new species - Midfemur with small basoventral tooth or angle (Fig. 11); tarsi red to brown multitnaculatus Kimsey, new species ACKNOWLEDGEMENTS nal °f the Australian Entomological Society. 32: 317-326. Many thanks to Jo Cardale, Greg Daniels and Chris Turner, R. E. 1908. Revision of the Thynnidae of Aus- Burwell for their assistance with this project. tralia. Part 2. Proceedings of the Linnaean Society of New South Wales 33: 70-256. LITERATURE CITED Turner, R. E. 1912. Notes on fossorial Hymenoptera. 1 1 . On some new Australian and Austro-Malay- Kimsey, L. S. and G. R. Brown. 1993. Lectotype des- an Thynnidae. Annals and Magazine of Natural ignations within the subfamily Thynninae. Jour- History 10: 533-546. J. HYM. RES. Vol. 9(1), 2000, pp. 29-33 Stethophotopsis, a New Genus of Sphaeropthalmini (Mutillidae: Sphaeropthalminae) with a Brachypterous Male from Arizona James P. Pitts and Joseph V. McHugh Department of Entomology, University of Georgia, Athens, GA 30602, USA Abstract. — A new genus of Mutillidae from southern Arizona with an bachypterous male, Steth- ophotopsis Pitts, is described and illustrated, including the new species Stethophotopsis maculata Pitts. The genus can be distinguished from males of other sphaeropthalmine genera by the pos- terior position of the mesosternal processes, the dilated and elongated condition of the cuspis and the absence of plumose pubescence on the cuspis. The subfamily Sphaeropthalminae in- cludes approximately 71 genera in two tribes, Sphaeropthalmini and Dasylabrini. Dasylabrini are restricted to the Old World while Sphaeropthalmini occur in the New World, Japan, and in the Medi- terranean and Australian regions (Broth- ers 1975). Of the 60 genera of Sphaerop- thalmini, 55 occur in the New World. This tribe is distinguished by two synapomor- phies apparent in both sexes: the approx- imately hemispherical, smooth and pol- ished condition of the eye and the pres- ence of plumose pubescence (Brothers 1975). In a study of Mutillidae from the south- western United States, two male speci- mens of an undescribed brachypterous species were found. Although no phylo- genetic hypothesis is available for genera of Sphaeropthalminae, this new species is unique in several features considered to be of generic-level importance for the sub- family. This new genus and species are described, illustrated and discussed be- low. MATERIALS AND METHODS We follow the terminology suggested by Menke (1993) for the scutum rather than that of Schuster (1958). The term "tib- ial spurs" is used instead of "calcaria." We denote the second, third, etc. metaso- mal tergites as T2, T3, etc. and the second, third, etc. metasomal sternites as S2, S3, etc. Stethophotopsis Pitts, new genus (Figs. 1-6) Male. — Head: As wide as thorax. Ocel- locular distance (Fig. 5) 2X width of lat- eral ocellus. Clypeus forming a trapezoi- dal, truncated anterior lobe, slightly to moderately depressed below dorsal man- dibular rim; clypeal base tuberculate. Ma- lar space (Fig. 4) 0.5X maximum eye width. Gena well developed, width ap- proximately equal to width of mandibu- lar base. Mandible tridentate apically, ventral margin with slight excision, not subtended by distinct sub-basal tooth. Antennal scrobes carinate above, with tu- bercle. First flagellomere 1.6X length of pedicel; second flagellomere 1.3x length of first flagellomere. Maxillary palp 6- segmented, labial palp 4-segmented. Me- sosoma: Mesoscutum with notali present posteriorly, absent or obscure on anterior third of mesoscutum. Tegula glabrous. Wings brachypterous, reduced to 0.5 X length of tegula. Mesosternum (Fig. 3) armed with pair of densely pubescent, triangular tapering processes, originating near midline immediately anterior to me- 30 Journal of Hymenoptera Research 0.40 mm 0.65 mm 0.75 mm 0.35 mm Figs. 1-6. Stethophotopsis maculata, Holotype. 1, Genitalia lateral view. 2, Genitalia, dorsal view on right, ventral view on left. 3, Sternum (legs except coxae, sculpture and pilosity omitted). 4, Body, lateral view (wings, legs except coxae, sculpture and pilosity omitted). 5, Head, frontal view (sculpture and pilosity omit- ted). 6, Left mandible. socoxae, appearing to cup anterior mar- gin of mesocoxae. Tibial spurs 1-2-2; tib- iae slender, not flattened. Metasoma: First segment (Fig. 4) petiolate, slender, not nodose, moderately constricted dorsally and laterally at apex, distal width much less than that of base of segment 2. Seg- ment 2 with both tergal and sternal felt lines. Apical margins of segments 1 and 2 with plumose pubescence. Pygidium short, subtruncate at apex. Hypopygium transverse, broader than long, laterally undefined. Faramere arcuate, stout at base and weakly dorsoventrally flattened. Cuspis (Figs. 1, 2) elongate, about equal to free length of paramere, curved ven- trally, basal portion cylindrical, distal portion dilated and weakly concave on ventral surface, ventral surface with dense simple pubescence distally. Digitus devoid of pubescence. Female. — Unknown Etymology. — From the Greek stetho "chest" + photopsis, a commonly used sphaeropthalmine suffix, refering to the characteristic sternal processes of this ge- nus. Gender feminine. Distribution. — USA, Southern Arizona. Type species. — Stethophotopsis maculata sp. nov. Volume 9, Number 1, 2000 31 Stethophotopsis maculata Pitts, new species (Figs. 1-6) Male. — Length: Holotype 7 mm, para- type 8 mm. Color: Head, thorax, petiole, second tergum, coxae, trochanters and tar- si brownish-yellow. Femora and tibiae dark brown. Third through seventh ab- dominal segments reddish-brown. Two round black maculations on anterior fourth of T2 with deep punctation, ap- pearing raised above surrounding disk. Pubescence of head, pleural region, me- sosternum and meatsomal sternites pale. Pubescence of thoracic dorsum golden- brown to pale. Pubescence of abdominal tergites pale except black in areas of tergal maculations. Head: Head as wide as tho- rax, rounded behind eyes in dorsal view. Ocelli salient, ocellocular distance (Fig. 5) slightly greater than 2x width of lateral ocellus, interocellar distance slightly great- er than 2x width of lateral ocellus. Clyp- eus anteriorly forming a trapezoidal, trun- cated anterior lobe, slightly to moderately depressed below dorsal mandibular rim. Malar space 0.5 X maximum width of eye. Gena (Fig. 4) well developed, width ap- proximately equal to width of mandibular base. Mandible (Fig. 6) tridentate apically, ventrally with a slight excision not sub- tended by a distinct sub-basal tooth, with ventral carina ending before midlength and with complete dorsal carina, raised at midlength. Apical mandibular teeth with Is" tooth basal width 4x and length 6x the 3rd tooth; 2nd tooth basal width IX and length 2x the 3rd tooth. Antennal scrobe carinate above, with small tubercle. First flagellomere 1.6X length of pedicel; sec- ond flagellomere 1.3X length of first fla- gellomere. Ridges of hypostomal region unmodified. Punctation of vertex conflu- ent. Mesosoma: Pronotum, scutum and scu- tellum shallowly, coarsely, confluently punctate. Scutum with subcomplete no- tauli, absent on anterior third of scutum. Tegula glabrous. Propodeum coarsely punctate. Mesopleuron with oblique sul- cus indistinct; sculpture reticulate throughout. Mesosternum (Fig. 3) armed with pair of triangular, tapering processes, originating immediately anterior to me- socoxae, situated slightly medially from center of coxae, covered with simple pu- bescence; sinus broadly U-shaped. Meta- sternum tridentate. Mesocoxae approxi- mate and unarmed; metacoxa and tro- chanter unarmed. Wings brachypterous, reduced to 0.5 length of tegula. Metasoma: First segment (Fig. 4) petiolate, slender, not nodose, posteriorly moderately con- stricted dorsally and laterally, posterior width much less than base of second seg- ment. Tl sparsely punctate, punctations separated by at least 2x width. Anterior margin of T2 coarsely, confluently punc- tate becoming shallowly, sparsely punc- tate posteriorly; S2 moderately punctate, anterior fourth with median longitudinal carina; sternal felt line approximately 0.75 x length of tergal felt line. Posterior margin of Tl and T2 with plumose pubes- cence. Pygidium transverse, broader than long and subtruncate at apex. Hypo- pygium transverse, broader than long, lat- erally undefined by carinae. Paramere (Figs. 1, 2) arcuate, stout at base and little dorso-ventrally flattened, tapering, devoid of long setose pubescence. Cuspis elon- gate, reaching nearly to apex of paramere, outwardly curved, distal portion distinctly dilated and slightly spatulate, basal por- tion cylindrical; distal portion with dense, long simple pubescence, basal half sparse- ly and minutely pubescent. Digitus de- void of pubescence. Female. — Unknown Type material. — Holotype: "Brown Can- yon, Baboquivari Mountains, Arizona, September 6, 1958, Stange and Menke" (LACM). Paratype: Arizona, Santa Cruz County, Madera Canyon, Santa Rita Mountains, 17-18.VIII.1949, Lloyd Martin (LACM). Etymology. — From the Latin maculata "spotted," in reference to the pair of black DISCUSSION 32 Journal of Hymenoptera Research maculations on the anterior margin of the ous other diagnostic features. Morsyma is second tergite. apterous and is superficially similar to Comments. — The paratype closely re- Stctlwphotopsis because of its degree of bra- sembles the holotype in most features ex- chyptery; however, Morsyma, has a broad- cept that it is slightly larger. ly sessile abdomen, has smaller ovate eyes and lacks sternal processes, notali and a tooth on the antennal scrobe (Schuster StetlwpJiotopsis is a distinct genus in the 1958). Protophotopsis shares with Stethopho- Sphaeropthalmini (subtribe: Sphaerop- topsis a tridentate mandible and the lack thalmina). The unique sternal processes Gf a ventral mandibular tooth, but it has and the dilated, spatulate and elongate the anterior pronotal margin distinctly condition of the genitalic cuspis are ap- emarginate, notauli absent, mesosternum parently autapomorphic for the genus, unarmed and pubescence simple (Cambra Stethophotopsis will key to subfamily and Quintero 1997). Photomorphus and Sphaeropthalminae without difficulty in Stethophotopsis have well developed ster- existing keys by Brothers (1993, 1995). In nal felt lines and a tuberculate clypeal Schuster's (1958) key to the sphaeropthal- base; however, Photomorphus has dentate mine males of the North American South- ridges on the anterior margin Qf the me- west, Stethophotopsis fails to key beyond sosternum, has a ventral mandibular the first couplet, where it may be diag- tooth/ and either lacks or has only vesti. nosed by the autapomorphies listed giai pium0se pubescence (Schuster 1958). a ove" Some subgenera of Sphaeropthalma Blake Lelej and Nemkov (1997) presented a Sphaeropthalma, Physetapsis and Photopsioi- phylogeny for the subfamilies of Mutilli- , x , . c X , . , , . ,T. , *, J b . y . „ , des) and species of Odontophotopsis Viereck dae and synonymies for mutillid genera. , ., , , , „ i ■ , . , nave a similar hypopygium morphology Because this work remains controversial t1 , n, ,, , , • », ,, /r> .i innni c ii «- i , to that or Stethophotopsis. Also, some small (Brothers 1999), we follow Schuster s . ,. ., , , rn , r ,, , , „, /inco\ i c- L- £ x.i_ i ,11 individuals of Sphaeropthalma and Odon- (1958) classification of the sphaeropthal- , , , r ., f . , ™,, w ... . , tophotopsis are similar in having mcom- mine genera. The subfamilies recognized , , „, br here are those presented by Brothers plete notauli. The mesosternum of Sp/^r- (1975, 1999). Currently, there is no phylo- °Pthalma> however, is never modified with genetic hypothesis available for the sub- dentate nd§es or Processes. In S. (Sphaer- tribe Sphaeropthalmina. Although it is ap- °Pthalma) and S. (Photopsioides) the cuspis parent that Sphaeropthalma is paraphyletic also 1S dllated; however, it bears plumose (pers. obs.), the othei genera of Sphaer- pubescence and is not spatulate. For Odon- opthalmini may be monophyletic. tophotopsis, the cuspis is rod-like and the The new species described here cannot pararnere js much longer than the cuspis. be placed in any of the established genera Odontophotopsis (Odontophotopsis) has of Sphaeropthalmini because it differs smaU dentate processes situated on the from each one by characters considered to mesosternal midline far removed from the be of generic-level significance. In the dis- mesocoxae and O. (Periphotopsis) has swol- cussion that follows, we distinguish Steth- len> longitudinal processes running the ophotopsis from the related genera with length of the mesosternum. which it is most likely to be confused. Dilophotopsis Schuster and Ac r op] w topsis Morsyma Fox, Protophotopsis Schuster Schuster differ greatly from Stethophotopsis and Photomorphus Viereck share some in having a large dilated sub-basal tooth characters which are of taxonomic signif- on the mandibles, dorsoventrally flattened icance with Stethophotopsis, although they parameres and strongly depressed hypo- differ in genitalic morphologies and vari- pygia with distinctly carinate lateral mar- ItTrtittl Volume 9, Number 1, 2000 33 gins (Schuster 1958), all of which are lack- ing in Stethophotopsis. Stethophotopsis shows the greatest ap- parent affinity to the genus Acanthophotop- sis Schuster. These genera share the fol- lowing characters: (1) the clypeus is trun- cate and depressed below mandibular rim, (2) the genitalic cuspis is flattened and dilated and (3) the mandibles are not dentate below but have a small excision (Schuster 1958). Despite these similarities, Acanthophotopsis differs in a number of im- portant characters warranting separation, including the following: (1) loss of a me- sotibial spur, (2) flattened and arcuate me- sotibia, (3) cylindrical mesosternal pro- cesses arising anterior to the posterior margin of the mesosternum (whereas Stethophotopsis has triangulate ridges that arise from the posterior margin of the me- sosternum and appear to cup the anterior margin of the mesocoxae), (4) complete notauli, (5) anteriorly reduced gena and (6) sternal felt line absent. ACKNOWLEDGEMENTS We thank Denis Brothers (Peitermaritzburg, South Africa), Donald Manley (Clemson, SC) and Theresa Pitts-Singer (U.S. Forest Service) for critically review- ing the manuscript. Roy Snelling graciously loaned the specimens from the Natural History Museum of Los Angeles County, Los Angeles, California (LACM). LITERATURE CITED Brothers, D. J. 1975. Phylogeny and classification of the aculeate Hymenoptera, with special reference to Mutillidae. University of Kansas Science Bulletin 50: 483-648. Brothers, D. J. 1993. Family Mutillidae. Pp. 188-203. In: Goulet, H. and J.T. Huber (eds.), Hymenoptera of the World: An Identification Guide to the Families. Centre for Land and Biological Resources Re- search, Ottawa, Ontario. Research Branch, Agri- culture Canada, Publication 1894/E. 668 pp. Brothers, D. J. 1995. Mutillidae. Pp. 541-548, In: Han- son, P.E. and I.D. Gauld (eds.), The Hymenoptera of Costa Rica, Oxford University Press, Oxford, 893 pp. Brothers, D. J. 1999. Phylogeny and evolution of wasps, ants, and bees (Hymenoptera, Chrysidoi- dea, Vespoidea, and Apoidea). Zoologica Scripta 28: 233-249. Cambra T., R. A. and D. Quintero A. 1997. A Revision of Protophotopsis Schuster (Hymenoptera: Mutil- lidae). journal of Hymenoptera Research 6: 263-272. Lelej, A. S. and P. G. Nemkov. 1997. Phylogenv, evo- lution and classification of Mutillidae (Hyme- noptera). Far Eastern Entomologist 46: 1-24. Menke, A. S. 1993. Notauli and parapsidal lines: just what are they? Sphecos 24: 9-11. Schuster, R. M. 1958. A revision of the sphaeropthal- mine Mutillidae of America north of Mexico. II. Entomologica Americana 37: 1-130. J. HYM. RES. Vol. 9(1), 2000, pp. 34-40 New Genera of Angarosphecinae: Cretosphecium from Early Cretaceous of Mongolia and Eosphecinm from Early Eocene of Canada (Hymenoptera: Sphecidae) WOJCIECH J. PULAWSKI, ALEXANDR P. RASNITSYN, DENIS J. BROTHERS, AND S. Bruce Archibald (WJP) Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA; e-mail: wpulawski@calacademy.org; (APR) Paleontological Institute, Russian Academy of Sciences, Moscow 117647, Russia; e-mail: rasna@glasnet.ru; (DJB) School of Botany and Zoology, University of Natal, Pietermaritzburg, Private Bag X01, Scottsville 3209, South Africa; e-mail: brothers@zoology.unp.ac.za; (SBA) Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada* Abstract. — The new genus Cretosphecium Pulawski and Rasnitsyn is described for two Early Cretaceous species from Mongolia: C. lobatum Pulawski and Rasnitsyn (type species) and C. triste Pulawski and Rasnitsyn. Eosphecium Pulawski and Rasnitsyn is described to include only the new species E. naumanni Brothers and Archibald from the Early Eocene of Canada. Eosphe- cium, tentatively assigned to Angarosphecinae Rasnitsyn, 1975, may represent the first Cenozoic record for the subfamily, known heretofore only from Early Cretaceous deposits. Species of Angarosphecinae Rasnitsyn, 1975 (= Baissodinae Rasnitsyn, 1975) are early sphecid wasps that lack synapomor- phies of the extant subfamilies. Only a small fraction of the available material has been described so far (Rasnitsyn 1975; Rasnitsyn et al. 1998, 1999), and three ad- ditional fossils that we attribute to this subfamily are described here. These spec- imens are characterized by a unique wing venation, found in no other sphecids, ei- ther fossil or extant; and one shows pro- notal lobes, a unique synapomorphy of Apoidea. Two are from the Early Creta- ceous, and the third is from the Early Eo- cene, the first record from the Cenozoic Era for the subfamily. Until now, Anga- rosphecinae (including much undescribed material housed at the Paleontological In- * Currrnt address: Museum of ( omparative Zoolog) . 26 Oxford Street, Cambridge, Massachusetts 021 18; e-mail: barchiba Id"" oeb.liarv.ird.edu stitute, Russian Academy of Sciences) were known exclusively from the Early Cretaceous. Morphological terminology used here deviates from that of Brothers (1975) and Bohart and Menke (1976) in several ways and is as in Rasnitsyn et al. (1999). The fol- lowing structures, variously termed in the literature, are here defined or redefined for the sake of clarity and convenience: - adlateral line: parapsidal line of Bohart and Menke (1976); we prefer the term coined by Budrys and Kazenas (1992) as it is more informative (self-explanatory and analogous to the admedian line) and also because the term parapsidal line is used by many authors to desig- nate the structure that Bohart and Men- ke and most other hymenopterists call notaulus; interpostgenal suture: line of fusion of the postgenae, extending from the oral cavity to the occipital foramen; Volume 9, Number 1, 2000 35 - metasoma: abdomen excluding the pro- podeum (gaster of Bohart and Menke 1976); - postgena: area between hypostomal ca- rina and occipital foramen; - spiracular lobe (Rasnitsyn 1988): pro- notal lobe of Bohart and Menke (1976); - costal space: cell C of Brothers (1975), costal cell of Bohart and Menke (1976); - cell 1 + 2r: cell SC + R of Brothers (1975), submarginal cell I of Bohart and Menke (1976); - cell 3r: cell R of Brothers (1975), margin- al cell of Bohart and Menke (1976); - cells 2rm and 3rm: cells IS and 2S of Brothers (1975), submarginal cells II and III of Bohart and Menke (1976); - cells lmcu and 2mcu: cells S + M and 1M of Brothers (1975), discoidal cells I and II of Bohart and Menke (1976); - cell 2cua: lCu of Brothers (1975), sub- discoidal cell of Bohart and Menke (1976); - veins 2r-m and 3r-m: distal margins of cells 2rm and 3rm, respectively (as in Richards 1977, and Gauld and Bolton 1988), corresponding to crossveins ls-m and 2s-m of Brothers (1975), and lr-m and 2r-m of Bohart and Menke (1976); we consider the true lr-m to be present only in primitive Symphyta such as the xyelid genus Pleuroneura Konow or Xye- la lata D. Smith. - veinlet lr-rs: partial vein within cell 1 + 2r, originally separating cells lr and 2r; called lr by Bohart and Menke (1976). - crossvein 2r-rs: vein separating cells 1 + 2r and 3r, called r-s by Brothers (1975) and 2r by Bohart and Menke (1976). Figs. 1 and 2 are by APR and Fig. 3 is by DJB. Cretosphecium Pulawski and Rasnitsyn, new genus (Figs. 1, 2) Derivation of name. — Cretosphecium, from the Latin word creta (chalk) and the Greek word sphekion (little wasp); with reference to its occurrence in the Cretaceous Period. Gender neuter. Diagnosis. — Cretosphecium has an unusu- ally long cell 2rm whose posterior margin is longer than that of cell 1 -I- 2r, veins lm- cu and 2m-cu are received by cells 2rm and 3rm, respectively, and cell 3r is trun- cate apically. A similar venation is found in Eosphecium. In Cretosphecium, however, vein 2r-rs is less than the height of ptero- stigma, cell 3r receives vein 3r-m next to the apical truncation, vein lm-cu joins cell 2rm at the cell's midlength, cell 2rm is less than 3 times as long as high, and cells 2cua, 2mcu, and lmcu are about equal in height. In Eosphecium, vein 2r-rs is about equal to the height of pterostigma, vein 3r- m is markedly distant from the apex of cell 3r, vein lm-cu joins cell 2rm before the cell's midlength, cell 2rm is almost 5 times as long as high, and cells 2cua and 2mcu are about twice as high as cell lmcu. Description. — The following comple- ments the above characteristics. Pronotal hindmargin angulate dorsolaterally adja- cent to spiracular lobe. Scutum: notaulus complete, extending from hindmargin to foremargin; adlateral line extending from hindmargin almost to foremargin. Fore- wing: vein 2r-m oblique, sinuous or straight, vein 3r-m oblique and sinuous (in C. lobatum, not preserved in C. triste), its anterior end insignificantly closer to wing margin than to apex of cell 3r; vein cu-a postfurcal. Hindwing (preserved only in C. lobatum): cu-a postfurcal, angled with first abscissa of Cu. Metasoma sessile. Taxonomic position. — The presence of a spiracular lobe (visible in C. lobatum) dem- onstrates that Cretosphecium is a member of Apoidea. The elongate notaulus that reaches the mesoscutal hindmargin is an ancestral character not found in any of the extant apoid genera but is typical of the subfamily Angarosphecinae. Type species. — Cretosphecium lobatum Pu- lawski and Rasnitsyn, new species. Composition. — Two species from the Ear- ly Cretaceous of Mongolia. 36 Journal of Hymenoptera Research 5 mm Fig. 1. Cretosphecium lobatum Pulawski and Rasnitsyn, new species. Upper left and right: thoracic dorsum in two aspects; al: adlateral line; aS: apical metasomal sternum; cr: costal; ex: coxa; f: femur; hyc: hypostomal carina; md: mandible; msl: median mesoscutal line; n: notaulus; N,: pronotum; N,: metanotum; occ: occipital carina; of: occipital foramen; pge: postgena; ph, pl3: meso- and metathoracic venters; pN3: metapostnotum ( = propodeal enclosure); pp: propodeum; S,: metasomal sternum I; sc: scape; scl: scutellum; si: spiracular lobe; sme: submentum; sti: stipes; ta: tarsus; tg: tegula; ti: tibia; tr: trochanter. Labels for veins and cells are con- ventional or as explained in text. Relationship of included species. — We treat the two specimens described below as con- generic because of their similar wing ve- nation and in spite of their morphological differences (presence or absence of the in- terpostgenal bridge, form of the metatho- racic venter). These differences could be re- garded as generic, but we prefer to base our classification solely on the wing vena- tion, admittedly perhaps creating artificial assemblages of unrelated taxa. This, how- ever, appears to be the most realistic ap- proach, given that the body structures are unknown in the majority of fossils. Cretosphecium lobatum Pulawski and Rasnitsyn, new species (Fig- 1) Derivation of name. — Lobatum, a Latin neuter adjective meaning lobed; with ref- erence to the spiracular lobes which can be clearly identified in the holotype. Volume 9, Number 1, 2000 37 5 mm Fig. 2. Cretosphecium triste Pulawski and Rasnitsyn, new species; d: discrimen (interpleural suture); ipgs: interpostgenal suture; oc: eye; ps: propodeal spiracle. Other abbreviations as in Fig. 1. Diagnosis. — Larger than C. triste (e.g., length of forewing about 9.0 mm rather than about 6.8 mm), vein 2r-rs emerging at about two-thirds pterostigmal length, vein 2r-m sinuous, and foremargin of cell 3rm shorter than height of cell. Description. — Sex unknown. No coarse sculpture recognizable except pronotum transversely ridged. Head nearly circular, occipital carina complete, postgenae wide- ly separated, gap between them obscured by bases of maxillae and labium. Meso- scutum with rudimentary median line. Metathoracic venter forming a subtrian- gular elevation between hindcoxae. Fore- wing: vein 2r-rs emerging at about two thirds of pterostigmal length; vein 2r-m sinuous, its fore end closer to 3r-m than to 2r-rs; vein 3r-m oblique, sinuous; maxi- mum width of cell 3rm markedly less than its basal height. Length of midtrochanter more than twice width. Midbasitarsus lon- ger than midtarsomeres II-IV combined, midtarsomere IV longer than wide. Meta- somal segment I subtriangular, wider than long, hindmargin of sternum I shallowly emarginate. Sternum II longer and wider than I, its anterior margin straight. Apical sternum subtriangular, carinate laterally. Body length as preserved 16.2 mm, fore- wing about 9.0 mm long, head width 3.0 mm. Body dark (color unknown for miss- ing parts: antenna! flagellum, hindtibiae and tarsi), but the following are distinctly paler: midtarsus (except basitarsus base) and wing veins including pterostigma. 38 Journal of Hymenoptera Research 5 mm Fig. 3. Eosphecium naumanni Brothers and Archibald, new species. Type material. — Holotype: Mongolia, Bayanhongor Aymag: Bon Tsagan, 5-8 km north of Bon Tsagan Nuur (= Bon Tsagan Lake); Early Cretaceous, impressed in marl of the Khurilt rock unit, Bon Tsagan Series (Sinitza 1993), possibly of Aptian age (Ponomarenko 1990). Deposited in the Paleontological institute, Russian Acade- my of Sciences, Moscow, Russia (catalog number 3559/4533). Cretosphecium triste Pulawski and Rasnitsyn, new species (Fig. 2) Derivation of name. — Triste, Latin neuter adjective meaning sad; with reference to the specimen's poor preservation. Diagnosis. — Smaller than C. lobatum (e.g., length of forewing about 6.8 mm rather than about 9.0 mm), vein 2r-rs emerging at pterostigmal midlength, vein 2r-m straight, and foremargin of cell 3rm longer than basal height of the cell. Description. — Sex unknown. Head near- ly circular, postgenae broadly contacting mesally, occipital carina complete and meeting hypostomal carina near midline of head. Scutum with no median line. Apex of metapleural venter with triangu- lar emargination. Propodeal spiracle long, narrow. Forewing: vein 2r-rs emerging at pterostigmal midlength; vein 2r-m straight, its fore end equidistant from 2r- rs and 3r-m. Legs appearing thicker than in C. lobatum. Original color probably not preserved (entire specimen uniformly light). Type material. — Holotype: Mongolia: Bayanhongor Aymag: Bon-Tsagan (other data as under Cretosphecium lobatum above). Deposited in the Paleontological Institute, Russian Academy of Sciences, Volume 9, Number 1, 2000 39 Moscow, Russia (catalog number 3559/ 694). Eosphecium Pulawski and Rasnitsyn, new genus (Fig. 3) Derivation of name. — Eosphecium, from the Greek words eos (dawn, morning, ear- ly) and sphekion (little wasp); with refer- ence to the Eocene age of the specimen. Gender neuter. Type species. — Eosphecium naumanni Brothers and Archibald, new species. Diagnosis. — Like Creiosphecium, Eosphe- cium is characterized by an elongate cell 2rm (the length of its posterior margin is approximately equal to that of cell 1 + 2r), veins lm-cu and 2m-cu are received by cells 2rm and 3rm, respectively, and cell 3r is truncate apically. Unlike in that ge- nus, the length of vein 2r-rs in Eosphecium about equals the width of the pterostigma; the anterior end of vein 3r-m is removed from the apex of cell 3r by more than the height of cell 2rm; veins RS + M and M (= posterior margins of cells 1 + 2r, 2rm, and 3rm) form an almost straight line (al- though Creiosphecium lobatum approaches this condition); vein lm-cu joins cell 2rm before the cell's midlength; cell 2rm is al- most 5 times as long as high (less than 3 times in Cretosphecium); and cells 2cua and 2mcu are about twice as high as cell lmcu (about equal in Cretosphecium). Taxonomic position. — In the absence of other evidence, we assign Eosphecium to Angarosphecinae based solely on the fore- wing venation pattern that resembles that of Cretosphecium. We also consider the above differences and the major difference in stratigraphic age sufficient to warrant its description a separate genus. Eosphecium naumanni Brothers and Archibald, new species (Fig. 3) Derivation of name. — Named after Dr. Ken Naumann who collected the speci- men. Description. — The following characters are complementary to the generic diag- nosis above: vein 2r-rs emerging at ptero- stigmal midlength; vein 2r-m straight, its fore end about equidistant from 2r-rs and 3-rm; vein 3r-m markedly angled near midlength; vein cu-a slightly postfurcal; vein Cu2 sinuous. Costal space densely se- tose towards apex and deeply pigmented (evident in counterpart), remainder of wing pale; veins dark. Forewing length about 14.2 mm (thus considerably larger than both species of Cretosphecium). Type material. — Holotype (forewing only): Canada, British Columbia: Quil- chena, 50°07'40.3"N, 120°30'34.7"W; Early Eocene: Coldwater beds of the Kamloops Group, 52-54 mya (Mathewes and Ville- neuve in prep.). Deposited in the Depart- ment of Biology, Simon Fraser University, Burnaby, British Columbia, Canada, cata- log number Q-0423a (part) and Q-0423b (counterpart). ACKNOWLEDGMENTS We sincerely thank Dr. Ken Naumann (Langara College, Vancouver) who donated the specimen of Eosphecium naumanni to Simon Fraser University and made it available for study- Financial assistance to DJB from the University of Natal Research Commit- tee is acknowledged with gratitude. Comments by an anonymous reviewer helped improve the manu- script LITERATURE CITED Bohart, R. M., and A. S. Menke. 1976. Sphecid Wasps of the World. A generic revision. Universitv of Cal- ifornia Press, Berkeley, 1 os Angeles, 1 ondon. 1 color plate, IX + 695 pp. Brothers, I). J. 1975. Phylogenv and classification ot the aculeate Hvmenoptera, with special reference to Mutillidae. 77k University of Kansas Science Bul- letin 50: 483-64N. Budrys, E. R., and V. L. Kazenas. L992. Novye vidy royushchikh os rod a Piodontus (Hvmenoptera, Sphecidae) — New species of the genus Diodontus ( I [ymenoptera, Sphe< idae I. Zoologic heskiy Zhurnal 71 (8): 24-31. English translation: 1993 New species of digger wasps of the genus Dio- dontus. Entomological Review 72: 114 121. Gauld, [., and B. Bolton. L988. The Hymenoptera. Brit- ish Museum (Natural History) and Oxford I ni- versit) Press, Oxford, 332 pp. 40 Journal of Hymenoptera Research Ponomarenko, A. G. 1990. Nasekomyie i stratigrafiya nizhnego mela Mongolii (= Insects and the Low- er Cretaceous stratigraphy of Mongolia), p. 103- 108 in Krassilov V.A. (editor). Kontinental'nyi mel SSSR. Materiyaly Soveshchaniya Sovetskoy Rabochey Gruppy Proekta N" 245 "Korrelatsiya nemorsksogo mela" Mezhdunarodnoy Programmy Geologicheskoy Korrelatsyi (= Non-marine Cretaceous of the USSR. The Submission from the Conference of the Soviet Working Group of the 1GCP Project 245), Vladivos- tok, 1988. Dal'nevostochnyi Otdel Akademii Nauk SSSR, Vladivostok. 226 pp. Rasnitsyn, A. P. 1975. Vysshye pereponchatokrylye mezozoya — Hymenoptera Apocrita of Mesozoic. Akademiya Nauk SSSR. Trudy Paleontologicheskogo Instiiuta (= Academy of Sciences of the USSR. Transactions of the Palaeontological Institute) 147: 1-133, I-VIII. Rasnitsyn, A. P. 1988. An outline of evolution of the hymenopterous insects (order Vespida). Oriental Insects 22: 115-145. Rasnitsyn, A. P., E. A. Jarzembowski, and A. J. Ross. 1998. Wasps (Insecta: Vespida = Hymenoptera) from the Purbeck and Wealden Lower Creta- ceous) of Southern England and their biostrati- graphical and paleoenvironmental siginficance. Cretaceous Research 19: 329-391. Rasnitsyn, A. P., W. J. Pulawski, and X. Martinez- Delclos. 1999. Cretaceous digger wasps of the new genus Bestiola Pulawski and Rasnitsyn (Hy- menoptera: Sphecidae, Angarosphecinae). Jour- nal of Hymenoptera Research 8: 23-34. Richards, O. W. 1977. Hymenoptera. Introduction and keys to families, second edition. Handbooks for the identification of British Insects, Vol. VI, Part 1. Roy- al Entomological Society of London, London. 104 PP- Sinitza, S. M. 1993. Yura i nizhniy mel Tsentral'noy Mongolii (= Jurassic and Lower Cretaceous of Central Mongolia). Trudy Sovmestnoy Sovetsko- Mongol'skoy Paleontologicheskoy Ekspeditsii [ = Transactions of the Joint Soviet-Mongolian Paleon- tological Expedition). No. 42: 1-239 pp. J. HYM. RES. Vol. 9(1), 2000, pp. 41-47 A New Species of Nitela (Hymenoptera: Sphecidae: Larrinae) from Australia with Notes on the Nests and Prey of Two Species Robert W. Matthews Department of Entomology, University of Georgia, Athens, GA 30602, USA Abstract. — Nitela elegans Matthews, a new species from Australia, is described and illustrated. The first biological data for N. elegans and N. australiensis Shultz, both nesting in pithy stems, are presented. Both prey on Psocoptera, and appear to progressively provision their cells. Prey of N. elegans were nymphs of sp. B (Psocidae) and nymphs of Heterocaecilius sp. (Pseudocaeciliidae). Prey of N. australiensis were nymphs of Aaroniella razvlingsi Smithers (Philotarsidae). There appear to be at least two generations per year in Canberra, Australia. The pteromalid chalcid Eupelmo- photismus pulcher (Girault) was reared from pupae of both species. A clutch of 14 Ceraphronidae {lAphanogmus sp.), possibly a hyperparasite of E. pulcher, was found inside a cocoon of N. aus- traliensis. Although Nitela Latreille is found Nitela elegans Matthews, sp. nov. world-wide, with 43 species listed by Bo- (Figs. 1-8) hart and Menke (1975), only three species Types.— Holotype female, 35.19S., are described from Australia, and noth- 148;08Ev Deakin, ACT, 4 April 1999, R. W. ing has been published on the nesting be- Matthews, deposited in ANIC. Paratypes: havior of any Australian species. Smith- 4 females, same locality and collector data ers (1990) recorded three species of Pso- as holotype bearing dates 20.i.l999 (Bio coptera as prey of an unidentified Aus- Note 185)/ 23.i.l999 (one with label Bio tralian species listed only as Nitela sp., Note 186)/ and 3.^.1999, all deposited in but provided no nest details. Studies of ANIC other species of Nitela (Iwata 1939; Jan- Female.— Head: Globular, broader than vier 1962; Valkeila 1955), indicate that high. Eyes stongly convergent dorsally, members of the genus nest in pre-existing distance between eyes at level of the lat- cavities in stems, galls, and beetle bur- eraj oceHj about half distance between rows. Cells are separated from one anoth- eyes measured just above the toruli. Frons er with bits of woody debris piled loosely (Fig. 1) evenly convex, rugulose, with lon- in the burrow. gitudinal rugae more prominent. Vertex Material in the Australian National In- (Fig. 4) finely punctate, except space be- sect Collection (ANIC) in Canberra sug- tween lateral ocellus and orbit smooth; lat- gests that there are several undescribed eral ocelli separated by just less than their species of Nitela in Australia, but most are diameter, but narrowly separated from represented by only one or two speci- eye margin by about 0.25 their diameter, mens. In order to identify the two species Gena (Fig. 2) finely rugulose at mandible discussed here, the types of each of the bases, becoming faintly reticulate dorsally. three named species were studied. Vouch- Occipital carina entire, weakly costulate er specimens of the wasps, parasites, prey, along anterior margin. Clypeus (Fig. 3) and nests are deposited in the ANIC. smooth, the apical margin rounded, very 42 Journal of Hymenoptera Research Figs. 1-8. Nitela elegans Matthews, sp. nov., paratype female: l, head, frontal view. 2, head and prothorax, lateral view. 3, clypeus, labrum, mandibles, frontal view. \, head and pronotum, dorsal view. 5, mesoscutum and scutellum. 6, propodeum, dorsal view. 7, mesosoma, lateral view, 8, Tl dorsal view. Scale lines 0.1 mm. Volume 9, Number 1, 2000 43 slightly pointed medially, with prominent Wings hyaline, veins brown. Body Length: median longitudinal raised carina, evenly 5.0 mm. rounded in profile, not quite reaching Male: Unknown. clypeal margin. Labrum (Fig. 3) short, Notes. — In Turner's (1916) key this spe- smooth broadly emarginate apically. An- cies runs to N. kurandae Turner. It differs tennal scrobes (Fig. 1) faintly transversely in that the frons sculpture (Figs. 1, 4) is microreticulate. Scape about twice as long much more rugose, and the scapes and as maximum breadth, length slightly less basal half of the flagellum are entirely than length of pedicel plus first flagellar black. segment. Second flagellar segment 1.3X as Biology. — Two active nests of this spe- long as first. Mandibles (Fig. 3) bidentate, cies were found in slender (ca. 5 mm di- the inner tooth smaller, blunter, and short- ameter) pithy stems of an unknown dead er than apical tooth. Mesosoma: Transverse plant (possibly Lantana sp.) on 20 and 23 pronotal sulcus (Fig. 5) crenulate, slightly January 1999 in a suburban yard of Deak- broader laterally, discontinuous medially in, ACT. One nest was newly initiated and where it is broken by a posteriorly pro- the other nearly complete. The newly ini- jecting "V"; lateral margins of pronotum tiated nest was in a burrow 102 mm long weakly angulate. Mesoscutum (Fig. 5) and 2.0 mm in diameter. This nest con- convex,uniformly punctate, except lateral tained a single half grown larva about margins crenulate. Scutellum uniformly midway along the burrow and 8 Psocop- punctate, separated from mesoscutum by tera nymphs (sp. B, Psocidae). Two prey narrow costulate furrow. Mesopleuron were adhering to the larva's body, and the (Fig. 7) subalar area coriaceous; signum others were scattered along the burrow, deep, area below it becoming coarsely Those not yet fed upon were only lightly punctate; episternaulus a distinct narrow paralyzed, able to kick their legs and crenulate furrow; hypersternaulus dis- move their antennae, but lacked coordi- tinct, broader, crenulate to rugulose, fad- nation. The female was resting near the ing posteriorly; area anterior to epister- entrance, head facing out. No nest struc- naulus rugose; propodeum lateral face ture was evident; there was no prelimi- longitudinally strigose with weak rugu- nary plug or cell closure, lose interspaces (Fig. 7); propodeal dorsal The second nest's burrow was 136 mm face rugulose with longitudinal rugae long and 2.0 mm in diameter. It contained more prominent; propodeal hind face (Fig. three completed cells and a fourth partial- 6) less strongly rugulose, with the trans- ly provisioned. The stem appeared previ- verse rugae more prominent. Metasoma: Tl ously to have been used by another wasp (Fig. 8) more or less smooth and shining, as the basal 36 mm of the burrow was very faintly coriaceous dorsally, with faint tightly packed with pith fragments and transverse microreticulation towards api- old insect parts. Cell 1 was 12 mm long cal margin. T2-T6 with faint transverse mi- and contained a Nitcla cocoon snug croreticulation. Forewing: Length 3.0 mm. against the packed matter in the inner end Marginal cell distally truncate, weakly ap- of the burrow. The cylindrical tan-colored pendiculate. lr-m vein straight, interstitial cocoon was 5 mm long and 1.8 mm in di- with recurrent vein, and interrupted at ameter. It was later found to contain a ful- about 0.25 of its length. Color. Head, me- ly formed dead adult chalcidoid parasit- sosoma, metasoma non-metallic black, oid, Eupelmophotismus pulcher (Girault) Antennae black. Mandibles black basally, (Pteromalidae: Cleonyminae). Cell 2 was 8 lighter apically. Legs orange, except coxae mm long and contained fragments of an black, femora suffused with brown, and old cocoon. Cell 3 was 13 mm long and distal three tarsomeres brown to black, contained a mature larva spinning a ma- 44 Journal of Hymenoptera Research trix of silk. Several faecal pellets adhered to the larva's body. Cell 4 was incomplete and contained 2 prey, both nymphs of Heterocaeciluis sp. (Pseudocaeciliidae). No egg was present and both prey were light- ly paralyzed and able readily to move their appendages. Beyond these prey the female was resting facing out. The cells were separated by partitions consisting of numerous bits and pieces of organic debris, mostly not identifiable, but appearing to be small bits of bark, pith, insect exoskeleton fragments, seeds, husks, caterpillar faeces, etc. loosely packed along the burrow. The lengths of the partitions closing the three cells were 7 mm, 33 mm, and 2 mm long respective- ly. That cells 1 and 2 may have belonged to an older, prior nest is suggested by the fact that the cocoon in cell 2 was old and empty, the cocoon in cell one contained a dead parasite, and the closing plug of cell 2 was very long. Both nests contained incomplete cells apparently being actively provisioned by the respective females. The first contained a partly grown larva in the cell and the second did not yet have an egg. Taken to- gether, these facts suggest that either de- layed mass provisioning or progressive provisioning is practiced in this species. Regardless, it appears that cells are not closed until the larva is essentially full grown. The two prey species are typically found either on bark, branches, and twigs, or on the undersides of green leaves (C. N. Smithers, in litt.). Nine prey found in another nest (presumed to be N. elegans) were also identified as psocid sp. B. The parasitoid genus Eupelmophotismus with about eight known species (Nau- mann, unpublished) is endemic to Austra- lia and New Guinea. Previous hosts re- corded for E. pulcher are bees, including Hylaeus sp., Amphylaeus morosus (Smith), (both Colletidae) and Neoceratina australen- sis Perkins (Anthophoridae) (Boucek 1988), and the sphecid wasp, Psenulus /'//- terstitialis Cameron (Matthews 2000). All of these hosts are twig nesters like N. ele- gans. Presumably £. pulcher oviposits through the stem wall and attacks the pu- pal stages of its host, although it is possi- ble in the case of Nitela that it burrows through the loose cell partitions to reach its host. Nitela australiensis Schulz This entirely black species of Nitela is widespread throughout Australia, al- though it has not yet been recorded from the Northern Territory. Biology. — Turner (1916) speculated that it nested in old beetle burrows in dead eu- calypts. I encountered it nesting in pithy stems of various unidentified plants in Deakin, a suburb of Canberra, ACT. Con- tents of three nests discovered from 31 January to 4 March 1999 are reported here. Architecturally the nests were indistin- guishable from those of N. elegans, being in burrows which had been excavated in slender (ca. 5-7 mm diameter) pithy stems, ranging from 27 to 96 mm long and 2 mm in diameter. From one to five cells were separated by loose aggregations of bits of organic debris, and were indistin- guishable from those of N. elegans. Two nests were complete when collect- ed on 31 January 1999. One contained 5 cells, each with a typical tan-colored Nitela cocoon. On 27 February a single chalcid parasite (Eupelmophotismus pulcher) and three N. australiensis females were found to have emerged. One additional emerged wasp escaped. The second completed nest apparently consisted of two cells. Cell 1 contained a tan-colored Nitela cocoon 5.2 X 1.5 mm at the base of a short 27 mm long burrow. When later checked this co- coon was found to be moldy. One dead female adult was found among the closure debris which more or less filled the outer 20 mm of the burrow. It is likely that she had recently emerged from a second cell which had been destroyed in the process Volume 9, Number 1, 2000 45 Figs. 9-11. Nitela australiensis Schulz: 9, Portion of nest showing two cocoons, a mature larva, M~\d prey separated bv partitions of various lengths (scale marks are mm); 10, closeup of the 20 Aaroniella rawlingsi pre) packed tightly in the cell; 11, cocoon, closeup, and the silk parchment-like inner lining on the partition at right, plus bits ot the particles that separated the cells. Scale in 10 & 11 is the same, the cocoon is 4 mm long. since there were Nitela cocoon fragments was found resting in the burrow. This mixed among the closure particles. nest, in a burrow 74 mm long and 2 mm The third nest (Fig. 9) collected on 4 in diameter, contained four completed March 1999 was incomplete and a female cells and a fifth cell (Fig. 10) containing 20 DISCUSSION 46 Journal of Hymenoptera Research moribund psocid nymphs packed tightly coarse sawdust particles, bits of charcoal, together, one with an egg attached. The bits of a blackish resinous substance, and egg was on one of the innermost prey af- small bits of frass. From this nest three fe- fixed obliquely across the venter of the males and two males had been reared prey's thorax. It measured 1.2 by 0.3 mm. (E.A. Sugden, personal communication). Although laid on one of the first prey It appears that there are at least two items, the egg had not yet hatched. How- generations per year in the Canberra area, ever, the larva in the preceding cell was Evidence is circumstantial, based on the already spinning, having consumed all the fact that progeny from the nest collected prey. This suggests that new cells are not in late January emerged by late February, provisioned until the larva in the previous whereas progeny of the nest collected in cell is nearly full grown. early March had entered diapause. Prey were all nymphs of Aaroniella raw- lingsi Smithers (Philotarsidae), a bark dwelling species. The delicate tan-colored The identity of the Nitela species men- cocoons (Fig. 11) in cells 1-3 all contained tioned by Smithers (1990) remains un- diapausing prepupae and were 4.0-4.5 known. Associated specimens were not mm long. The mature spinning larva in found in the collection of the Australian cell four was later found dead. Each com- Museum. However, except for a single fe- pleted cell had a thin silk or parchment- male of Peripsocus milleri (Tillyard) (Per- like inner closing partition to which pieces ipsocidae), the prey used by this species of the particles separating the cells were were all nymphs belonging to the Elipso- attached. These partitions appear to have cidae and Caeciliidae. Thus the Australian been constructed by the mature larva dur- species of Nitela, like their congeners else- ing the process of forming its cocoon. Such where in the world, specialize on Psocop- a partition, which was always constructed tera (but see Zuijlen (1994) who notes a in the base of the cell, may help the larva possible record for Zoraptera for N. bifida to know the proper orientation for its co- Menke from Costa Rica), primarily those coon. groups that live on the surface of bark, Another nest, apparently belonging to with a strong preference for nymphs. Six this species, was collected on 31 January families of psocids (Caeciliidae, Elipsoci- 1999. It had been usurped by another cav- dae, Peripsocidae, Philotarsidae, Pseudo- ity-nesting wasp, Arpactophilus sp. (Sphe- caecilidae, and Psocidae) have now been cidae: Pemphredoninae). The basal 33 mm recorded as prey of Australian Nitela. of the burrow contained 4 cells with co- The provisioning data also suggest that coons, from three of which emergence had parental investment by the female is rath- taken place. The fourth cocoon, still intact, er extensive. Probably either progressive was opened to reveal a clutch of 14 adult provisioning or delayed mass provision- ceraphronid parasites (? Aphanogmus sp.), ing is normal in both species. At the very probably hyperparasites of E. pulcher. Also least, the female appears to wait to begin present in the nest debris was a single a new cell until the larva in the previous dead N. australiensis female. cell is nearly full grown. Interestingly, the In the Australian National Insect Collec- two parasitoids reared from these species tion is a series of N. australiensis reared on were both found inside the cocoon. Be- 21 Feb. 1986 from a trap nest from Nadgee cause parental care apparently ceases by Nature Reserve, New South Wales by E. the time the larva is full grown, perhaps A. Sugden. This artificial burrow was 60 it is not surprising that the cocoon stage mm long and had a bore diameter of 4.5 would be the most vulnerable to parasit- mm. It contained six cells separated by ism. Volume 9, Number 1, 2000 47 The only record of a nest of Nitela from southern Africa concerns an undescribed entirely black species, 4.6 mm in length (Fred and Sarah Gess, unpublished). It was constructed in a trap nest placed ver- tically among dry inflorescence stems of BerkJiei/n (Asteraceae) on a stream bank in the Goegap Nature Reserve, Springbok, Namaqualand in low karroid scrub. Col- lected on 21 October 1987, the nest burrow was 288 mm long and 6.5 mm diameter. The nest burrow was closed at both ends, with a crescent-shaped entrance at mid- length. The nest had been in the field for six days when collected, and was found to have two completed cells provisioned with unidentified Psocoptera and closed with dry plant detritus and seeds. In nei- ther cell were all the prey consumed be- fore the larva spun a creamy-white cocoon with dense brittle walls 0.06 mm thick and rounded at both ends. A female later emerged from a cocoon that was 5.2 mm long and 2.0 mm diameter. The other co- coon (not measured) produced a male. ACKNOWLEDGEMENTS I thank Dr. C. N. Smithers (Sydney Museum) for identifying the Psocoptera prey, and Dr. Gary A. Gib- son (Canadian National Collection, Ottawa, Canada) for identifying the chalcid parasite. Dr. Ian D. Nau- mann (CSIRO Entomology) provided valuable advice and assisted in identifying the Nitela species and the lAphanogmus sp. Fred and Sarah Gess kindly shar- eded unpublished notes on a South African species of Nitela, and their review of the manuscript greatly improved it. Holotypes of Nitela species were kindlv loaned from the British Museum of Natural History (London) and the Museum fur Naturkunde, Hum- boldt Universitat (Berlin) by Ms. Christine Thompson and Dr. F. Koch, respectively. Eric Hines (CSIRO) helped with the SEM photographs and the nest and prey photos were taken by David McClenaghan (CSI- RO). Financial support from the University of Geor- gia and a McMaster Fellowship from CSIRO are gratefully acknowledged. LITERATURE CITED Bohart, R. M. and A. S. Menke. 1976. Sphecid Wasps of the World. A Generic Revision. University of Cal- ifornia Press, Berkeley. Boucek, Z. 1988. Australasia)! Chalcidoidea (Hymenop- tera). A Biosystematic Revision of Genera of Fourteen Families, with a Reclassification of Species. CAB In- ternational, Wallingford. Iwata, K. 1939. Habits of Rhinonitela domestica Wil- liams, a minute book-louse hunter in Formosa. Mushi 12: 13-16. Janvier, H. 1962. Recherches sur les Hymenopteres nidifiants aphidivores. Annates des Sciences Na- turelles Zoologique (12) 4: 489-516. Matthews, R. W. 2000. Nesting biology of the Austra- lian stem-nesting wasp Psenulus interstitialis Cameron (Hymenoptera: Crabronidae: Pemphre- doninae) on Magnetic Island. Australian journal of Entomology 39: 25-28. Smithers, C.N. 1990. First record of Psocoptera as prey of Australian Sphecidae (Hymenoptera). Australian Entomological Magazine 17: 42. Turner, R. E. 1916. Notes on fossorial Hymenoptera. xxiii. On some Australian genera. Annals and Magazine of Natural History (8) 18: 277-288. Valkeila, E. 1955. Observations on the biology and development of Nitela spinolae Dhlb. Annates En- tomologica Fcnnici 21: 54-57. Zuijlen, J. VV. A. van 1994. The amazonica species group of the genus Nitela Latreille (Hymenop- tera: Sphecidae: Crabroninae). Zoologisches Meddelin Leiden 68: 249-269. J. HYM. RES. Vol. 9(1), 2000, pp. 48-52 New Synonyms in Central and South Asian Sphecidae (Hymenoptera) Vladimir L. Kazenas Zoological Institute of MS-AS, Akademgorodok, Almaty, 480060, Kazakhstan; e-mail: common@zool2.academ.alma-ata.su Abstract. — A study of type material at the Natural History Museum, London, England, revealed 9 new synonyms for species described from Central Asia and former British India. These names are: Ammophila bolanica Nurse, 1903 = Podalonia hirsuta mervensis (Radoszkowski, 1887); Cerceris nursei Turner, 1912 = Cerceris antennata F. Morawitz, 1890; Cerceris supposita Kohl, 1916 = Cerceris rothneyi Cameron, 1890; Cerceris compta Turner, 1912 = Cerceris turkestanica Radosz- kowski, 1893 (= Cerceris rufonodis Radoszkowski, 1877); Cerceris barrei Radoszkowski, 1893 = Cerceris tetradonta Cameron, 1890; Cerceris rhytichophora Turner, 1912 = Cerceris unidentata F. Morawitz, 1890; Laphyragogus turanicus Gussakovskij, 1952 = Laphyragogus kohlii (Bingham, 1896) (described in Lianthrena); Palarus nursei Turner, 1911 = Palarusfunerarius F. Morawitz, 1890; and Philanthus marikovskii Kazenas, 1977 = Philanthus elegantissimus Dalla Torre, 1897 (= Phi- lanthus elegans F. Smith, 1873). Lectotypes are designated for Ammophila bolanica, Cerceris comp- ta, Cerceris nursei, Cerceris rhytichophora, Laphyragogus kohlii, Laphyragogus turanicus, and Palarus nursei. One major problem facing students of Central Asian insects is their relation to the biotas of Pakistan and northwestern India. For a variety of reasons, Russian au- thors who studied the sphecid fauna of Central Asia over the last 140 years (Ev- ersmann, Radoszkowski, F. Morawitz, Shestakov, Gussakovskij, Marshakov, my- self and others) have not considered the species described from former British In- dia (now India and Pakistan) by Cameron, Bingham, Nurse, F. Smith, and Turner. These latter authors, conversely, showed little interest in the work of Russian au- thors. The two areas, however, are not only adjacent geographically, but they closely resemble each other in their habitat types and ecological conditions (ranging from lowland hot deserts to high moun- tains with glaciers). With hundreds of spe- cies described on each side, it is inevitable that the ratio of synonyms may be high. A number of synonyms in Sphecidae were established by Pulawski (1975, 1979, 1995), Marshakov (1977), Budrys (unpublished), and Antropov (unpublished). For more than 25 years I have been studying sphecid wasps of Kazakhstan and adjacent republics of Central Asia. I previously studied almost all the types of the species described by earlier Russian authors during my many visits to the Zoo- logical Institute of Russian Academy of Sciences in St. Petersburg, and to the Zoo- logical Museum of Moscow State Univer- sity (Moscow, Russia). It was therefore im- portant to compare these types with types of the species described by British authors. The Museum of Comparative Zoology at the Harvard University (Cambridge, Massachusetts, U.S.A.) awarded me an Ernst Mayr grant for the travel to the Nat- ural History Museum in London and to the University Museum of Natural Histo- ry in Oxford to study these types. I worked there for 6 weeks in August and September 1997. I have studied nearly 200 types and found nine new synonyms. Volume 9, Number 1, 2000 49 The following abbreviations are used in Cerceris supposita Kohl, 1916:122, 6. Syntypes: the text to designate institutions where the Turkmenistan: Serakhs (KRAK6W, exam- type material is housed: ined). New synonym. KRAKOW: Instytut Systematyki i Ewol- ucji Zwierzat, Polska Akade- mia Nauk, Krakow, Poland. NHML: The Natural History Museum, London, Great Britain. OXUM: University Museum of Natu- ral History, Oxford Universi- ty, Great Britain. ZIN: Zoological Institute, Russian Academy of Sciences, St. Pe- tersburg, Russia. ZMMU: Zoological Museum, Moscow State University, Moscow, Russia. LIST OF SPECIES (arranged alphabetically by their valid names) Cerceris antennata F. Morawitz Cerceris antennata F. Morawitz, 1890:598, 6. Ho- lotype: 6, Turkmenistan: Kiiren-Dagh (ZIN, examined). Cerceris nursei Turner, 1912:512, 9, 6. Lecto- type: 6, Pakistan: Quetta (NHML, exam- ined), present designation, in order to ensure that the name is properly and consistently applied. New synonym. The lectotype of C. nursei bears the fol- lowing labels: 1. Quetta 5.03, 2. Cerceris nursei Turn., type 6, and 3. Nurse Coll.: 1915-34. This species belongs to the specularis group. It differs from closely related spe- cies by the markedly concave clypeus, un- usually short pygidial plate, and presence of flat brushes of dense setae in postero- lateral corners of male sternum VI (Kazen- as 1984:201-203). Cerceris rothneyi Cameron Cerceris rothneyi Cameron, 1890:251, 9 (as Roth- neyi, incorrect original capitalization). Lecto- type: 9, India: West Bengal: Barrackpore (OXUM), designated bv Empev, 1984:7^ (Ar- ticle 74.6). The lectotype of Cerceris rothneyi should be in Oxford (Empey, 1984), but I was un- able to locate it. However, 3 specimens (2 9, 1 6) in NHML agree with the original description. The first specimen has the fol- lowing labels: 1. N. Kanara, and 2. Cerceris rothneyi Cam.; the second one: 1. N. Ka- nara, and 2. Bombay Presidency, pres. by E. Comber 1910-255; the third one: 1. T.R. Bell, Karachi, 2. Cerceris rothneyi Cam. 6\ and 3. 1911-276. I consider these specimens to be con- specific with C. supposita Kohl, a member of the bupresticida group. It differs from all closely related species by the combination of the clypeal structure in the female, form of the vertical lamella on female sternum V, sculpture of propodeal enclosure, color, and other features (Kazenas 1984:79-81). In particular, the clypeal free margin of the female has 4 teeth; the vertical lamella of sternum V has a roundly prominent margin; the propodeal enclosure is fullv unsculprured, shiny; the female pygidial plate is not narrowing posterad; male ter- gum VI and sternum VI each has a pos- terolateral tooth. Also, tergum I is partly red (also tergum II in some specimens), whereas female tergum IV has an uninter- rupted pale yellow fascia apically. Cerceris tetradonta Cameron Cerceris tetradonta Cameron, 1890:261, 9, 6. Syntypes: N. India: Poona (depository un- known). Cerceris barrel Radoszkowski, 1893:68, 9, 6". Syntypes: Turkmenistan: Serax (KRAKOW, not examined). New synonym. There are six specimens of C. tetradonta Cameron in NHML; one of them was col- lected in Pakistan (Karachi), three came from India (Abu, Deesa, and Khandala), and two from Sri Lanka. They are conspe- cific with specimens of Cerceris barrel Ra- doszkowski from Turkmenistan in Zl\ 50 Journal of Hymenoptera Research determined by Gussakovskij and Shesta- kov. C. tetradonta belongs to the albofascinta group and closely resembles C. albofascinta Rossi, but differs by the form of the clyp- eal free margin in the female and details of the body sculpture and color (Kazenas 1984:185-186). In particular, the clypeal free margin is conspicuously dentate, the mesopleuron and propodeal side have yellow spots, and the gastral sterna large- ly and the legs are yellow. Cerceris turkestanica Radoszkowski Cerceris rufonodis Radoszkowski, 1877:56, 9,6. Syntypes: Uzbekistan: Djisak, Tashkent; and Kyrghyzstan: Osh in Fergana Valley (ZMMU, examined). Preoccupied by Cerceris rufinodis F. Smith, 1856 (Article 58.12, use of different connecting vowel). Cerceris turkestanica Radoszkowski, 1893:66. Re- placement name for Cerceris rufonodis Ra- doszkowski, 1877 (proposed to replace Cer- ceris rufinoda Cresson, 1865). Cerceris compta Turner, 1912:803, 9 . Lectotype 9: Pakistan: Karachi (NHML, examined), present designation, in order to ensure that the name is properly and consistently ap- plied. New synonym. The lectotype female in NHML has the following labels: 1. Type, 2. E. Comber, Karachi, Oct. 09, 3. Cerceris compta Turn. Type, 4. Bombay Presidency, pres. by E. Comber 1910-255, and 5. B.M. Type Hym. 21.1, 362. The type of C. compta is identical with C. turkestanica. This species, a member of the rybyensis group, is characterized by the form of the female clypeus whose free margin is slightly sinuous on each side of the small, median incision. Also, sternum II has a prominent basal plate, the pro- podeal enclosure is almost entirely smooth, and the gastral color pattern is distinctive (see Kazenas, 1984:35); gastral segment I may be red or black. The pro- podeal side has a large yellow spot, and the legs are yellow except the femora are black ventrally. Cerceris unidentata F. Morawitz Cerceris unidentata F. Morawitz, 1890:601, 9. Holotype: 9, Turkmenistan: Kopet-Dagh near Chuli (ZIN, examined). Cerceris rhynchophora Turner, 1912:510, 9, 8. Lectotype: 9, Pakistan: Quetta (NHML, ex- amined), present designation, in order to en- sure that the name is properly and consis- tently applied. New synonym. The lectotype female of C. rhynchophora in NHML has the following labels: 1. Type H.T., 2. Quetta 5.03, 3. Cerceris rhynchopho- ra Turn., Type, 4. 9, and 5. B.M. Type Hym. 21.1.426. The specimens of C. unidentata F. Mor- awitz from Turkmenistan in ZIN are con- specific with the lectotype female of C. rhynchophora Turner. The species differs from other Cerceris by the following: pro- podeal enclosure closely punctate and with fine, transverse ridges; jugal lobe of hindwing 7-9 times shorter than anal cell; middle clypeal lobe in female with char- acteristic, overhanging, roof-like projec- tion, in male with narrow, longitudinal ca- rina and tridentate free margin (see Ka- zenas, 1984:178-180). Laphyragogns kohlii (Bingham) Lianthrena kohlii Bingham, 1896:213, 9,8. Lec- totype 8: "N. India", may be Pakistan: Pun- jab: no specific locality (NHML, examined), present designation, in order to ensure that the name is properly and consistently ap- plied. Laphyragogus turanicus Gussakovskij, 1952:227. Lectotype: 9, Tajikistan: Ayvadj at Kafirni- gan River (ZIN, examined), present desig- nation, in order to ensure that the name is properly and consistently applied. New syn- onym. There are 3 specimens in NHML. Of them, 1 9 and 1 8 were collected in Deesa and 1 o* (lectotype) is simply labeled North India. The last specimen has the fol- lowing labels: 1. Type, 2. N. Ind., 3. Lian- threna kohlii Bingh. 9 Type, and 4. B.M.Type 21.88. It is actually a male. De Beaumont (1959) and Gussakovskij ; IttlNUUfH Volume 9, Number 1, 2000 51 (1952) discussed color differences between kolilii and turanicus. The specimens I stud- ied do not differ morphologically and are very similar in color, so I consider them conspecific. The species differs from its congeners by the form of the first meta- tarsal article in the female and the struc- ture of the flagellum and sternum VII in the male (Gussakovskij 1952). Palarns funerarius F. Morawitz Palarus funerarius F. Morawitz, 1890:136, 9. Ho- lotype: 9, Mongolia: Zagan-Buryuk (ZIN, examined). Palarus nursei Turner, 1911:481, 9, 6. Lecto- type: 6: Pakistan: Quetta (NHML, exam- ined), present designation, in order to ensure that the name is properly and consistently applied. New synonym. There are 3 specimens (2 9 , 1 8 ) of P. nursei in NHML. The lectotype male has the following labels: 1. Type H.T., 2. Quet- ta 6.03, 3. 6,4. Palarus nursei Turner Type, 5. Col. C.G. Nurse Collection 1920-72, and 6. B.M. Type Hym. 21.77. The specimens of P. nursei from Quetta, Pakistan and of P. funerarius from many localities in Central Asia are very close morphologically and to my mind conspe- cific. The differences in color are not con- spicuous. P. funerarius is similar to P. bi- signatus F. Morawitz, but differs in color and in structure of the male flagellum (F. Morawitz 1890b:136-139). Also, male ster- num I of P. funerarius has a pair of tuber- cles (none in P. bisignatus) and the apical prominence of sternum II has 2 transverse carinae (one in P. bisignatus, evanescent in some specimens). The gaster of P. funer- arius has no red, and femora have a large black spot each. Philanthus elegantissimus Dalla Torre Philanthus elegans F. Smith, 1873:415, 9. Holo- type or syntypes: "N. India", may be Paki- stan: no specific locality (depository un- known). Preoccupied by Philanthus elegans F. Smith, 1856, now in Trachypus. Philanthus elegantissimus Dalla Torre, 1897:485. Replacement name for Philanthus elegans F. Smith, 1873. Philanthus marikovskii Kazenas, 1978:662, 9,6. Holotype 9: Kazakhstan: 15 km E Ayak-Kal- kan (ZIN, examined). New synonym. I was unable to locate the original spec- imens of F. Smith either in London or in Oxford, but 3 specimens in NHML (2 9 and 1 6) from Deesa probably collected by C.G. Nurse agree with the original de- scription. I consider them to be conspecific with Philanthus marikovskii. The species is morphologically close to Ph. venustus (Rossi) and Ph. rubriventris Kazenas, but differs in having extensive pale yellow coloration (Kazenas 1978: 662-664). Podalonia hirsidta mervensis (Radoszkowski) Ammophila mervensis Radoszkowski, 1887:89, 9, 6. Syntypes: Turkmenistan: Samsaul; Cau- casus; and Corsica (KRAKOW, not exam- ined). Ammophila bolanica Nurse, 1903:8, 9. Lectotype: 9, Pakistan: Quetta (NHML, examined), pre- sent designation, in order to ensure that the name is properly and consistently applied. New synonym. There are 3 9 of A. bolanica from Quetta in NHML. The lectotype has 6 labels: 1. Type, 2. Quetta, 3. 9, 4. Type, 5. Coll. C.G. Nurse Collection 1920-72, and 6. B.M. Type Hym. 21730. These specimens are conspecific with specimens of A. mervensis (Radoszkowski) from Transcaspia in ZIN and ZMMU. R.M. Bohart and A.S. Menke (1976) con- sider A. mervensis to be a subspecies of Po- dalonia hirsuta (Scopoli). It differs from the nominotypical subspecies in having an all black gaster. ACKNOWLEDGMENTS For thf support of mv work I thank Wojciech J. Pulawski (California Academy of Sciences, San Fran- cisco, California), Christopher O'Toole (University Museum of Natural History, Oxford), the staff of the Fintomology Department o\ the \atural History Mu- seum, London, and the Ernst Mayr Grants Committee (Museum of Comparative Zoology, Harvard Univer- 52 Journal of Hymenoptera Research sity, Cambridge, Massachusetts). I also thank Arnold S. Menke for his valuable comments on the manu- script. LITERATURE CITED Beaumont, J. de. 1959. Le genre Laphyragogus Kohl (Hym. Sphecid.). Revue Suisse de Zoologie 66, 4: 723-734. Bingham, C. T. 1896. On some exotic fossorial Hy- menoptera in the collection of the British Muse- um with descriptions of new species and of a new genus of the Pompilidae. The Journal of the Linnean Society. Zoology 25: 422-445, pi. XIX. Bohart, R. M., and A. S. Menke. 1976. Sphecid Wasps of the World. A generic revision. University of Cal- ifornia Press, Berkeley, Los Angeles, London. 1 color plate, IX + 695 pp. Cameron, P. 1890. Hymenoptera Orientalis, or con- tributions to a knowledge of the Hymenoptera of the Oriental Zoological Region. Part. II Memoirs and Proceedings of the Manchester Literary & Phil- osophical Society 4, 3: 239-284, pi. IX-X. Dalla Torre, K. W. von, 1897. Catalogus Hymenoptero- nun hucusque descriptorum systematicus et synon- ymicus. Vol. VIII: Fossores (Sphegidae). Lipsiae: Guilelmi Engelmann. 749 pp. Empey, H. N. 1984. Notes on the Cerceris types of Cameron from the Oriental Region (Hymenop- tera: Sphecidae: Philanthinae). journal of the En- tomological Society of Southern Africa 47: 75-82. Gussakovskij, V. V. 1952. Novye i maloizvestnye vidy Psammocharidae i Sphecidae (Hymenoptera) za- padnogo Tadzhikistana. Trudy Zoologicheskogo Instituta Akademii Nauk SSSR^ 10: 199-288. (In Russian) Kazenas, V. L. 1978. Novye vidy royushych os (Hy- menoptera, Sphecidae) iz yugo-vostochnogo Ka- zachstana. Entomologicheskoye Obozrenie 57, 3: 661-665. (In Russian) Kazenas, V. L. 1984. Royushyie osy Cercerisy Sredney Azii i Kazakhstana. Alma-Ata: Nauka Kaz. SSR. 232 pp. (In Russian) Kohl, F. F. 1916 (1915). Beitrag zur Kenntnis der Gat- tung Cerceris auf Grundlage der O. Radoszkov- skyschen Sammlung. Archiv fin Naturgeschichte, Abteilung A 88, 7: 107-125. Marshakov, V. G. 1977. Obzor ruyushehikh os triby Crabronini (Hymenoptera, Sphecidae) fauny SSSR. Rod Cnibro I abricius, 1775. Entomologiches koe Obozreniye 56: 854-872. (In Russian) Morawitz, F. 1889. Insecta, a cl. G. N. Potanin in Chi- na d in Mongolia novissime lecta. IV. I [ymenop- tera Aculeata. Horae Societatis Entomologicae Ros- sicac 23: 112-168. Morawitz, F. 1890. Hymenoptera fossoria transcas- pica nova. Horae Societatis Entomologicae Rossicae 24: 570-645. Nurse, C. G. 1903a. New species of Indian Aculeate Hymenoptera. Annals and Magazi>ie of Natural History (Series 7) 11: 511-526. Nurse, C. G. 1903b. New species of Indian Hymenop- tera. The journal of the Bombay Natural History So- ciety 15: 1-18. Pulawski, W. J. 1975 (1974). Synonymical notes on Larrinae and Astatinae (Hymenoptera: Spheci- dae). journal of the Washington Academy of Sciences 64: 308-323. Pulawski, W. J. 1979. A revision of the World Proso- pigastra Costa (Hymenoptera, Sphecidae). Polskie Pismo Entomologiczne 49: 3-134. Pulawski, W. ]. 1995. The wasp genus Gastrosericus Spinola, 1839 (Hymenoptera: Sphecidae). Mem- oirs of the California Academy of Sciences 18: 1-174. Radoszkowski, O 1877. Sphegidae. In: Voyage au Turkestan d'A. P. Fedchenko, fasc. 14, tome 2, partie 5. Izvestiya Imperatorskogo Obshchcstva Lyu- biteley Estestvoznaniya, Antropologii i Etnografii pri Imperatorskom Moskovskom Universitete 26: 1-87, pl. I-VIII. (In Russian) Radoszkowski, O. 1887. Faune Hymenopterologique Transcaspienne (suite). Horae Societatis Entomolo- gicae Rossicae 21: 88-101, pl. IV-V. Radoszkowski, O 1893. Faune hymenopterologique Transcaspienne. Supplement. Horae Societatis En- tomologicae Rossicae 27: 38-81,490-493. pl. IV-V. Smith, F. 1856. Catalogue of hymenopterous insects in the collection of the British Museum. Part IV. Sphegidae. Larridac, and Crabronidae. London, p. 207-497, pl. DC-XL Smith, F. 1873. Descriptions of new species of fosso- rial Hymenoptera in the collection of the British Museum and of a species of the rare genus Iswara belonging to the family Dorylidae. Annals and Magazine of Natural History (Series 4) 12: 49-59, 99-108, 253-260, 291-300, 402-415. Turner, R. E. 1911. Notes on fossorial Hymenoptera. IV. Remarks on the Genus Palarus. Annals and Magazine of Natural History (Series S) 7: 479-485. Turner, R. E. 1912a. A monograph of the wasps of the genus Cerceris inhabiting British India. With Notes on other Asiatic Species. The journal of the Bombay Natural History Society 21, 2: 476-516; 21,3: 794-819. Turner, R. E. 1912b. Notes on Fossorial Hymenop- tera.— X. On new species from the Oriental and Ethiopian Regions. Annals and Magazine of Natu ral History (Series S) 10: 361-377. J. HYM. RES. Vol. 9(1), 2000, pp. 53-61 The Biology of Thrincohalictus prognathus (Perez) (Hymenoptera: Halictidae: Halictini) Laurence Packer Department of Biology, York University, 4700 Keele St., Toronto, ON, M3J 1P3, Canada; E-mail: bugsrus@yorku.ca Abstract. — The Halictine bee Thrincohalictus prognathus (Perez) was studied in Israel in May and the beginning of June, 1998. Additional information was obtained from museum specimens. The species appears to be both common and widespread in northern Israel, particularly in the Galilee and Golan Heights where it occurs between altitudes of several hundred metres up to 1650m on Mount Hermon. Despite having an unusually elongate head, the species visits a wide range of flowers which generally do not possess a long corolla. Like most temperate halictines, only mated females overwinter and become active in the spring, as early as mid March. Males are found no earlier than mid May. The apparent age of females increases from March to May with newly eclosed, unworn individuals appearing in late May /early June. All worn early summer individ- uals were mated and had well developed ovaries. Thus, ovarian development and phenological data are inconsistent with the species being eusocial but are consistent with it being univoltine. The behaviour exhibited by paired females in artificial observation arenas indicates that T. prog- nathus possesses the repertoire of agonistic and cooperative behaviours usually found in halictines but that aggressive interactions predominate. Comparisons with other species indicate that the relative frequency of passing behaviour is very low and inconsistent with that expected for a communal species. I conclude that this species is probably solitary. Phylogenetic methods are useful not only in determining polarity of evolution- ary changes among character states, but also for predicting which species are par- ticularly deserving of study (Packer 1997). Based upon phylogenetic data, the mono- typic genus Thrincohalictus is thought to be particularly worthy of study because it appears to be the sister taxon to the well studied and behaviourally diverse genus Halictus (Eickwort et al. 1996; Packer 1997; Danforth et al. 1999). Thus, this species is particularly important for assessing the pattern of evolution of social behaviour in the bee tribe Halictini. Thrincohalictus prognathus (Perez) is a relatively large, non-metallic halictine with apical bands of tomentum on the ab- dominal terga and, as the specific epithet suggests, a long face. It is known, from comparatively few specimens, to occur in Turkey and northern Israel (Bluethgen 1955; Ebmer personal communication) at least as far south as Jerusalem where one male specimen is known from Mt. Scopus. In May and June 1998 I studied this spe- cies in Israel. Although detailed sociobio- logical analysis requires observation over several months (if not years, Yanega 1988; Sakagami and Packer 1994), some sugges- tive information can be obtained from samples of dissected bees (Dunn et al. 1999). Additionally, behaviours of bees caught from flowers but forced to interact in artificial arenas (circle tubes) may also be suggestive of social or solitary behav- ioural ancestry (McConnell-Garner and Kukuk 1997; Wcislo 1997; Paxton et al. 1999; Packer, unpublished observations). In this paper I present data on i) the dis- tribution of T. prognathus in Israel, ii) a list of the flowers that it visits, iii) data from 54 Journal of Hymenoptera Research dissected and /or measured bees from all available samples and iv) the results of cir- cle tube experiments. The last two sets of data are potentially useful in elucidating the type of social organisation T. progna- tlius might possess. METHODS Sampling and phenological assessment. — Most females of T. prognathus were col- lected with a hand net from flowers al- though some of the pinned specimens in- cluded in some analyses were swept from vegetation. Most males were collected while they flew rapidly over bushes or around small trees, presumably in search of females. Some females were pinned but most were preserved in buffered formalin (Pabalan 1998) for subsequent dissection, one was preserved in ethanol for DNA se- quencing (Danforth et al. 1999). At two lo- calities, a parking lot and adjacent road- side near a garbage dump near Hazor, and at the peak of Mount Meiron, both in the Upper Galilee, large numbers of fe- males were observed, not all of them were collected. All males were pinned. Flower visitation data were obtained from field samples and, when given, also from pinned specimens. Bees preserved in formalin were dis- sected to establish their degree of ovarian development and whether or not they had mated. The fraction of a fully developed oocyte found upon microscopic examina- tion of each of the six ovarioles was esti- mated by eye. Whether the terminal oo- cyte was developing or being resorbed was noted. Resorbing oocytes have a fuzzy outline and are often misshapen (Goukon et al. 1988; Pabalan 1998). The spermathecae of inseminated females are opaque whereas those of unmated females are clear and appear like glass bowls. Head widths of all females (museum specimens and those collected by the au- thor) were measured to ascertain whether there was any seasonal size variation sim- ilar to that found in eusocial species in which summer workers are generally smaller than spring gynes (Breed 1975; Packer and Knerer 1985). The number of nicks in one forewing margin was counted to estimate the relative age of specimens. The largest number of nicks counted was 24, consequently, bees with the entire mar- gin slightly eroded were coded as 25 and those with more extensive abrasion coded as 30. Although in reality wing wear mea- sures comparative levels of activity which will not always be directly related to age, it is of some utility. For example, com- pletely unworn individuals will either have recently eclosed or be overwintered females at the early stages of nest initia- tion and heavily worn individuals will have been active foragers probably for several weeks or more. Behavioural data. — Despite several days searching in one locality where the species was particularly common (in the vicinity of Hazor in the Upper Galilee), no nest sites were discovered. Behavioural obser- vations were made upon pairs of bees in- teracting in a "circle tube" — a 20cm long piece of clear plastic with the ends joined together such that moving bees are forced to repeatedly interact with one another (Breed et al. 1978). Methods generally followed those of earlier authors (Breed et al. 1978; Mc- Connell-Garner and Kukuk 1997; Wcislo 1997b; Paxton et al. 1999; Pabalan et al. submitted) with the exception that bees were placed into circle tubes within ten minutes of capture. This was because it has been discovered that, for some halic- tines at least, ovarian resorption begins within 30 minutes of captivity (Pabalan 1998 and unpublished observations on La- sio^lossum zephyrum), suggesting physio- logical changes are occurring which may be mirrored by altered behavioural inter- actions among females. The range of behaviours noted within circle tubes were similar to those often found in other studies (see references above). Bees would sometimes nudge or ■ Iffllmt Volume 9, Number 1, 2000 55 Table 1. List of flower records for Thrincohalictus prognathus. Locality # bees Females Mt. Hermon Neve Ativ Qirvat Shemona Ein Fit Hazor El Rom Mt. Meiron Males Katzrin Snir Ramot Naftali 8 2 3 1 1 1 1 dozens dozens 3 dozens 1 1 1 Xcpcta cilicica Labiatae Lonicera sp. Caprifoliaceae Phlomis chrisophyla Labiatae Cerasu> prostrata Rosaceac Papaver Papaveraceae Papaver Papaveraceae Onopordum blanche Certtaurea iberica Asteraceae Phlomis viscosa Labiatae Papaver Papaveraceae Silybum marianum Asteraceae Echinops Asteraceae Vitex anguscastus Verbenaceae Ballota saxatilis Labiatae lunge at one another, back away from one another either with or without reversing or pass one another, venter to venter, a manoeuvre requiring coordination be- tween individuals. A more aggressive be- haviour, the C-posture was also observed. In this stance an individual bends its ab- domen forward, beneath the thorax in an apparent attempt at stinging the individ- ual in front of it. It has also been suggest- ed that secretions from the Dufours gland may be released during this posture (Smith and Weller 1989). The occurrence of these behaviours was recorded contin- uously for 20 (one pair) or 30 (the remain- ing four pairs) minutes. RESULTS Flower records. — Table 1 shows the flow- er record data for those specimens for which this information was recorded. As can be seen, most of the flowers do not need a long tongue or face for their nectar to be accessible to the bees. However, a few individuals have been collected from long corolla flowers such as the labiate Ne- peta cilicica and for none of the observa- tions was it definitively recorded whether the bees were collecting pollen or nectar. Distribution. — Localities where T. prog- nathus was found and also those for which museum records are available are shown in Figure 1. The species appears to be common in northern Israel, particularly in the Upper Galilee and Golan Heights where it occurs at a wide varietv of alti- tudes, from just a few hundred metres above sea level as at Katzrin to over 1,000m at the summit of Mt. Meiron. It has also been collected at 1650m on Mt. Her- mon. There is a single male from Jerusa- lem, collected in the 1940's. Phenology and dissection data. — Figures 2 and 3 plot head width and wing wear against date of capture for all females available for study. The data are consis- tent with overwintered females becoming active in March, becoming increasingly worn as they forage through to mid May and with the first individuals of the next generation flying from mid May until June for mating and pre-diapause feeding. Midseason females (worn individuals col- lected in May) are not smaller than those found earlier in the year (mean head width early bees = 2.31mm, SD = 0.05, n = 18; midseason bees = 2.33mm, SD = 0.06, n = 37; t = 1.80, p > 0.05). In fact, in this instance the direction of the size dif- ference is in the opposite direction to that 56 Journal of Hymenoptera Research Fig. 1. Map of Israel showing location of sites where T. prognathic has been collected. H refers to Hazor and M to Mount Meiron, the two localities where most of the data dealt with in the text were obtained. expected if midseason bees were workers and early season ones gynes. Similarly, worn late females are no smaller than the unworn pre-overwintering females (Mann-Whitney U test, U = 87, t = 0.39, p > 0.5); a size difference would be ex- pected if the worn bees were workers and the unworn ones overwintering future foundresses. Thus, there is no size varia- tion suggestive of the caste differences found in annual eusocial species. A total of 20 bees from Hazor collected between May 7"' and 10"1 (an additional 10 were pinned) and 17 from Mt. Meiron col- lected on June 2nd (an additional 8 were pinned) were dissected. All Hazor bees were well worn with an average wing wear of 10.2 (SD = 4.5, n = 30). All form- aldehyde preserved bees were mated and had well developed ovaries and 15 of them (75%) had a chorionated oocyte and at least one more ovariole with either a healthy oocyte developing or a sizeable re- sorbing oocyte. The remaining five bees each contained at least one oocyte three quarters the size of a fully developed one. The average size of the 20 dissected bees was 2.33mm (SD = 0.07) these results change almost imperceptibly when pinned specimens are added to the sam- ple, mean size = 2.32 mm (SD = 0.06, n = 30). No T. prognathus were found at Ha- zor between May 28th and 30th, almost no flowers remained in bloom at this time. These data are consistent with the May Hazor samples being composed of forag- ing bees which had been active for at least several weeks and which were reproduc- tive individuals. That no males were found suggests that the sample dates pre- ceded the emergence of the next genera- tion. In contrast, the sample from Mt. Meiron at the beginning of June contained mostly unworn bees (22/25 or 88%) or bees with a total index of wear of one (2/25). Only one bee was well worn and it had 24 nicks in the wing margin. All bees in this sam- ple were mated and all but the heavily worn individual had no ovarian develop- ment, large amounts of fat in their abdo- mens and a crop full of nectar. The excep- tion had two % developed oocytes that were resorbing, its abdomen contained no obvious large fat deposits and its crop was empty. The average size of the Mt. Meiron sample is identical to that of the one from Hazor (mean size = 2.33mm, SD = 0.06, n = 25). These data are consistent with the sampling period having occurred during the emergence phase of the offspring gen- eration with only one individual of the pa- rental generation captured. None of the bees dissected showed any evidence of oophagy; they did not contain pasty-white material in their guts. 1 PhVre... Volume 9, Number 1, 2000 57 E S .c ■g T3 TO 0) I z.z> - 2.45- DD O 2.4 - o □ □ m o 000 2 35 - O oooo i mi i 000 ODD □ 2 3 - OOODOO □ o inn i OLD 00O o 2.25 - o o □ □ 00 ODD 2.2 - □ 2.15 - □ 2.1 - o 2 05 - 15-Mar 25-Mar 4-Apr 14-Apr 24-Apr 4-May 14-May 24-May 3-Jun 13-Jun 23-Jun Date O unworn Dworn Fig. 2. Head width against collection date for all females, circles represent unworn bees, squares worn ones. 30 n 25 20 <0 o c 15 10 04-0- 15-Mar O COO 00 000 00 o 000 oooo 00 oooo 000 oo o oo 30-Mar 14-Apr 29-Apr Date 14-May 29-May OO 13-Jun Fig. 3. Wing wear against collection date for all females. 58 Journal of Hymenoptera Research Table 2. Comparison of the frequencies of passing and agonistic interactions among paired individuals of various halictine species. Pass/FE is the frequency of successful passes per frontal encounter, A/FE is the frequency of aggressive interaction per frontal encounter. raxon pss/FE A/FE Reference Eusocial Species L. zephyrum queens L. zephyrum foragers L. zephyrum guards 0.57 0.21 0.00 1.28 0.56 0.61 Breed et al. 1978 Breed et al. 1978 Breed et al. 1978 Communal Species L. hemichalceum 0.81 0.02 Kukuk 1992 Solitary Species L. figueresi 0.14 0.41 Wcislo 1997b L. platycephalum L. (Ctenonomia) sp. 0.30 0.33 0.1 1 0.22 McConnell-Garner and Kukuk 1997 Species With Unknown Behaviour r. prognathus 0.03 0.29 this paper Bees collected and pinned from other samples also have the same size profile as those from samples discussed above, 34 bees from a variety of localities had an av- erage head width of 2.32mm (SD = 0.06). The first male was found at Ramot Naf- tali on May 12th, 7 more were found there on June lsl and one was collected at Mt. Meiron on June 2nd. All of these males were comparatively unworn. Few other records are available, but it seems that the first Ramot Naftali specimen is the earliest record and that males may be found until early July (Bluethgen 1955). Behavioural observations. — A total of 218 encounters between paired bees in five cir- cle tube experiments were observed. Thir- ty-four of these were C postures, 28 were nudges or lunges and 2 were pushes. Few- er cooperative behaviours were observed: 7 examples of back and follow and 6 suc- cessful passes, most of which were pre- ceded by pass attempts; a total of 7 en- counters resulted in pass attempts which failed and led to one or both of the bees backing off. Thus, 29.4% of all interactions were classified as antagonistic (varying between 10 and 35% among the pairs) and only 3% were passes (varying between 0 and 7%). The remainder involved one or both bees turning away and may thus be considered as avoidance behaviours. The relative frequency of agonistic and coop- erative behaviours did not vary much among the pairs, with the former always exceeding the latter. Comparisons of successful passes and aggressive acts per encounter are shown in table 2 for those halictines for which such data are available. The behaviour of T. prognathus shows the lowest rate of passing (except for guard:guard pairs in L. zephyrum) and a level of aggressive inter- action that falls within the range for soli- tary species and outside that for either communal or eusocial taxa. DISCUSSION Thrincohalictus prognathus has been con- sidered to be a rare species likely to have a preference for flowers with a long co- rolla. However, observations in northern Israel confirm that it is widespread (Figure 1) and often common and that it does not have a preference for long tubular flowers (Table 1 ). In solitary and communal halictine bees, all individuals in a sample collected from flowers may be expected to be of similar age (except during a period of overlap of Volume 9, Number 1, 2000 59 generations) and show similar evidence of reproductive activity. Early season sam- ples should be of mostly unworn bees with average wear increasing monotoni- cally until late in the season when the young offspring generation individuals fly to mate and to forage to fill their crops with nectar. At this time, bimodalitv in wear may be expected if some of the pa- rental generation individuals have sur- vived and are active. In samples of a eu- social species, early samples may be ex- pected to show evidence of reproduction and comparatively little heavy wear, mid- season samples should show varying pro- portions of reproductively inactive indi- viduals and a wide range of indices of wear and the latest samples should com- prise mostly unworn pre-diapause fe- males and some heavily worn ageing workers. Additionally, bees in samples of eusocial species taken in mid season should average smaller than those collect- ed in spring as workers are usually small- er than nest foundresses (Breed 1975) (al- though the size difference varies from re- quiring large sample sizes to achieve sta- tistical significance (Eickwort 1985) to being non-overlapping such that individ- uals can be classified as queens or workers on size alone (Knerer 1992)). The samples of T. prognathus show no evidence of eusociality. The earliest spec- imen found was collected in mid March. Most samples are from early May and bees at this time are well worn but show substantial and largely equivalent levels of reproductive development. No unmat- ed or ovarially undeveloped individuals have been detected at this time, which, be- ing just prior to the apparent emergence of the overwintering brood, should have consisted entirely, or almost entirely, of workers if the species were eusocial. Al- though data on recently emerged brood are mostly from a separate and higher al- titude (though nearby) locality, they indi- cate that foraging by females of this spe- cies has ceased by early June and that males are actively searching for overwin- tering females at this time. In most of the region under discussion here, there are al- most no flowers available for bee foraging after early June. Indeed, there was an ex- tremely marked deterioration of forage availability at Hazor between early May and the end of the month such that great bee abundance had changed to an almost complete absence, several hours of search- ing between May 28th and 30th failed to re- sult in any T. prognathus being found whereas earlier in the month several doz- ens could be observed in an hour. Thus, based upon phenology, dissection and wear data, I conclude that T. progna- thus is not an eusocial species in Israel. I now turn to the behavioural interactions among individuals observed in circle tubes. Behavioural profiles of bees in circle tubes seem to reflect the differences be- tween the conflictive relationships in eu- social and semisocial societies and har- monious interactions in communal ones (Kukuk and Crozier 1990; McConnell-Gar- ner and Kukuk 1997; Paxton et al. 1999). In eusocial societies, where competition among individuals over oviposition or the sex ratio of brood occurs, agonistic acts such as C-postures, lunges and pushing are common (McConnell-Garner and Ku- kuk 1997; Pabalan et al. submitted). Con- versely, communal species are much more tolerant, pass one another readily (Paxton et al. 1999) and unrelated individuals have even been observed performing trophal- laxis (Kukuk and Crozier 1990). Solitary species show intermediate levels of both aggressive and cooperative behaviours (Table 2). As noted above, the phenological and dissection data suggest that T. prognathus is not eusocial or semisocial. Similarly, communal behaviour would seem to be unlikely as the pass per encounter rates of bees with this type of colony organisation are consistently high (McConell-Garner and Kukuk 1997; Packer unpublished ob- 60 Journal of Hymenoptera Research servations) yet in T. prognathus they are the lowest recorded except for pairs of guards which are expected not to pass one another. Thus, I conclude that the data are most consistent with the hypothesis that this species is solitary. Clearly more de- tailed information is required and knowl- edge of the nest architecture of T. pirogna- tluis would also be of interest. In conclusion, T. prognathus would seem to be a locally abundant halictine in north- ern Israel, it does not seem to specialise on pollen or nectar from long corolla plants and its phenology and behavioural pat- terns in experimental arenas are consistent with it being a solitary species, although forms of quasisocial behaviour cannot be definitively disproven. ACKNOWLEDGMENTS I thank Drs. Amnon Freidberg, Abraham Hefetz and Avi Shmida for facilitating my studies in Israel, and, in particular, Rani Kasher and his family for ac- commodation during the study reported here and to Rani for access to his collection and for identifying the plants visited by T. prognathus. Jessica Janjic helped with the figures and Dr. J. Shore provided phytosystematic information. Financial support came from the Natural Sciences and Engineering Research Council of Canada. LITERATURE CITED Bluethgen, P. 1955. The halictinae (Hymen., Apoidea) of Israel I. Genus Halictus (Subgenera Halictus s.str. and Thrincohalictus). Bulletin of the Research Council of Israel 5: 5-23. Breed, M. D. 1975. Sociality and seasonal size varia- tion in halictine bees. Insectes Sociaux 22: 375-380. Breed, M. D., J. M.Silverman and W. J. Bell. 1978. Ag- onistic behavior, social interactions and behav- ioral specialization in a primitively eusocial bee. Insectes Sociaux 25: 351-364. Danforth, B. N., 1 1. Sauquet and L. Packer. Phylogeny of the bee genus Halictus (Hymenoptera: Halic- tidae) based on parsimony and likelihood anal- yses of EF-lalpha sequences. Molecular Phyloge netics and Evolution 13: 605-618. Dunn, M., P. L. Mitchell and L. Packer. 1999. Phe- nology and social biology of two sibling species of Halictus in an area of sympatry. Canadian lour ual of Zoology 76: I ickwort, G. C. 1986. first steps into eusociality: the sweat bee Dialictus lineatulus. Florida Entomologist 69: 742-754. Eickwort, G. C, K. R. Eickwort, J. M. Eickwort, J. Gor- don and A. Eickwort. 1996. Solitary behavior in a high-altitude population of the social sweat bee Halictus rubicundus (Hymenoptera: Halictidae). Behavioral Ecology and Sociobiology 38: 227-233. Goukon, K., Y. Maeta and S. F. Sakagami. 1988. Sea- sonal changes in ovarian state in a eusocial hal- ictine bee, Lasioglossum duplex, based on stages of the oldest oocytes in each ovariole (Hymenop- tera: Halictidae). Research in Population Ecology 29: 255-269. Knerer, G. 1992. The biology and social behaviour of Evylaeus malachurus (K.) (Hymenoptera; Halicti- dae) in different climatic regions of Europe. Zool- ogische Jahrbucher Abteilung fur Systematik, Oeko- logie und Geographie der Tiere 119: 261-290. Kukuk, P. F. 1992. Social interactions and familiarity in a communal halictine bee Lasioglossum (Chilal- ictus) hemichalceum. Ethology 91: 291-300. Kukuk, P. F. and R. H. Crozier. 1990. Trophallaxis in a communal halictine bee Lasioglossum (Chilalic- tus) erythrurum. Proceedings of the National Acad- emy of Sciences U.S.A. 87: 5401-5404. McConnell-Garner, J. and P. F. Kukuk. 1997. Behav- ioral interactions of two solitary halictine bees with comparisons among solitary, communal and eusocial species. Ethology 103: 19-32. Pabalan, N. A. 1998. Aspects of the interaction between reproductive morphology and social behaviour of a primitively eusocial sweat bee, Halictus ligatus Say (Hymenoptera; Halictidae). Unpublished Ph.D. the- sis, York University, Toronto. 149pp. Pabalan, N., K. G. Davey and L. Packer. Escalation of aggressive interactions during staged encounters in Halictus ligatus (Hymenoptera: Halictidae). Journal of Insect Behavior. Submitted. Packer, L. 1997. The relevance of phylogenetic sys- tematics to biology: examples from medicine and behavioral ecology, pi 1-29. In P. Grandcolas (ed). The Origin of Biodiversity in Insects: Phyloge- netic Tests of Evolutionary Scenarios. Memoires du Museum National d'Histoire Naturcllc vol. 173. 360pp. Packer, L. and G. Knerer. 1985. Social evolution and its correlates in bees of the subgenus Evylaeus (Hymenoptera: I lalictidae). Behavioral Ecology and Sociobiology 18: 363-375. Paxton, K. ]., P. F. Kukuk and J. Tengo. 1999. Effects ol familiarity and nestmate number on social in- teractions in two communal bees, Andrena SCOtica and Panurgus calcaratus (Hymenoptera, Andren- idae). Insectes Sociaux 46: 109-118. Sakagami, S. F. and L. Packer. 1W4. Delayed euso- ciality in halictine bees. p. 86 In: Lenoir, A., G. Volume 9, Number 1, 2000 61 Arnold and M. Lepage (eds). Les Insectes Sociaux. and social interactions in a solitary bee Lasioglos- Paris, Universite Paris Nord. 583pp. sum (Dialictus) figueresi (Hymenoptera: Halicti- Smith, B. H. and C. Weller. 1989. Social competition dae). Insectes Sociaux 44: 199-208. among gynes in halictine bees: the influence of Yanega, D. 1988. Social plasticity and early-diapaus- bee size and pheromones on behavior, journal of ing females in a primitively eusocial bee. Pro- Insect Behavior 2: 397-41 1. ceedings of the National Academy of Sciences U.S.A. Wcislo, W. T. 1997. Elements of dominance behavior 85: 4374-4377. J. HYM. RES. Vol. 9(1), 2000, pp. 62-70 Sperm Development and Ultrastructure of Mature Spermatozoa of Megalyra (Hymenoptera: Megalyroidea) Terence M. Newman and Donald L. J. Quicke (TMM) Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, UK; (DLJQ) Unit of Parasitoid Systematic^ CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, UK and Department of Entomology, The Natural History Museum, London SW7 5BD, UK Abstract. — Sperm ultrastructure and aspects of spermiogenesis are described for the first time for a member of the hymenopteran superfamily Megalyroidea, the parasitic wasp, Megalyra fas- ciipennis Westwood. The Megalyroidea are a poorly-known, putatively ancient group of the Apo- crita (wasp-waisted wasps), and they are the first member of the Evaniomorpha group of super- families (sensu Rasnitsyn 1988) whose sperm have been investigated in detail. Therefore their spermatology might provide clues as to the groundplan for the higher Hymenoptera. The Megalyroidea are among the more (Shaw 1990; Austin et al. 1994). Introduc- elusive and least well known of the para- tions of one Australian species, Megalyra sitic wasps, currently placed in the infor- fasciatipennis Westwood, into South Africa mal 'Evaniomorpha' group of superfami- in an attempt at biological control of Phor- lies as defined by Rasnitsyn (1988) which acantha has resulted in a population of the comprises the Ceraphronoidea, Evanoidea parasitoid that in some years becomes nu- s.l., Megalyroidea, Slephanoidea and Tri- merous enough to permit guaranteed and gonaloidea. However, few characters are relatively straightforward collecting, and known that unite this morphologically it was in one of those years that material and biologically heterogeneous group, was obtained for spermatological investi- and its monophyly is only weakly sup- gation. This study provides the first sper- ported (Ronquist et al. 1999). How the Me- matological information on the Megaly- galyroidea are related to the other apocri- roidea, and is also the first detailed ultra- tan (wasp-waisted) Hymenoptera is there- structural investigation of sperm for any fore not well understood, and several fea- of the 'Evaniomorpha' superfamilies. The tures suggest that they may be one of the work was carried out to provide basic ul- most basal groups, forming a transition trastructural information on sperm devel- between the parasitic sawfly family Orus- opment in this rarely studied group, and sidae and the other parasitic wasps (see to see if comparison with other Apocrita Gibson 1985; Shaw 1990; Heraty et al. can provide phylogenetic information. 1994; Dowton & Austin 1994; Vilhelmsen That sperm ultrastructure might pro- 1997). Few species are known and most vide new character systems for resolving are known from only one or a few speci- relationships within the Hymeoptera was mens in collections (Shaw 1990). They are muted by Jamieson (1987), it was not until idiobiont ectoparasitoids of concealed the preliminary comparative survey of hosts, probably the majority attacking the non-aculeates by Quicke et al. (1992) that wood-boring larvae of large beetles in- some of the wealth of characters they pos- cluding the pest cerambycid, Phoracantha sess became apparent. But, while a grow- ' Bttlnftuu Volume 9, Number 1, 2000 63 ing number of superfamilies of Hymenop- was not much wider than the tail at its tera have had at least one included species widest and could be seen to taper from the examined in detail for sperm ultrastruc- posterior of the nucleus to the tip of the hire (see for example, Thompson & Blum acrosome. 1967; Lensky et al. 1979; Lingmei & Dunsu The mature sperm of Megalyra (Fig. 1), 1987; Wheeler et al. 1990; Newman & illustrate all of the organelles so far de- Quicke 1998, 1999a,b) there are a number scribed in the sperm of other parasitic of important and phylogenetically signifi- wasps, i.e. axoneme, mitochondrial deriv- cant groups for which nothing is known, atives, deltoid bodies, acrosome and nu- for example, Orussoidea, Stephanoidea, cleus. In the testes cysts are found with Trigonaloidea and even the Evanoidea, sperm at many different stages of devel- Ceraphronoidea, Cynipoidea and Platy- opment. Primary and secondary sper- gastroidea, the last four of which are com- matocytes are found with the latter in a mon and easy to obtain alive. It is hoped syncytium formed by incomplete cytogen- therefore that the present work will help esis following the earlier mitotic division encourage others to investigate the sper- (Fig. 2). The structural features are as fol- matology of these taxa in order to add to lows: the body of phylogenetic information for Axoneme. — The tail portion of the ma- resolving higher relationships within the ture sperm contains a single axoneme order. with a 9 + 9 + 2 arrangement of micro- tubules for most of its length (Fie. 3). MATERIAL AND METHODS There are we„ developed acc°SSOry ° ruc'. Recently eclosed adult male Megalyrn tures, particularly linking the central pair fasciatipennis Westwood were collected in of microtubules (Fig. 3, arrowed). South Africa in September 1998 and trans- The axoneme develops from an electron ported by air to the U.K., where testes and dense structure which becomes positioned vas deferens were dissected under 2% glu- adjacent to the nucleus defining the pos- taraldehyde in phosphate buffered saline terior pole of the nucleus (Fig. 4). This (pH 7.2), fixed for two hours, then trans- structure, often referred to as a ring cen- ferred to 2% osmium tetroxide in cacodyl- triole (Fig. 5), contains the basal body of ate buffer (pH 7.2) for a further 2 hr fixa- the axoneme, from which the axoneme tion. After two buffer washes, tissue was elongates. However, during spermiogen- dehydrated to 50% ethanol and then con- esis, many sectioned cells are found with trasted with a saturated solution of uranyl between two and four axoneme profiles acetate in 50% ethanol prior to complete (Fig. 6). Usually one or more of the pro- dehydration, embedding in Epon resin files exhibits a loss of structural integrity and polymerisation overnight. Large sil- (arrowed Fig. 6) often appearing with less ver sections were picked up on high res- than the full complement of microtubules, olution grids and contrasted with uranyl It is not exactly clear what these represent acetate and lead citrate. and the possibility that they are degener- ating cells cannot be ruled out, though their prevalence and the presence in each Mature sperm of Megalyra fasciipennis of at least one apparently perfect axoneme collected from the vas deferens were not profile suggest that they are a normal de- formed into spermatodesmata. The indi- velopmental stage. No mature sperm with vidual sperm ranged in length from 160 to multiple tails have been found. 200 (Jim, though most were close to 180fxm Mitochondrial derivatives and centriolar ad- of which the head (acrosome plus nucleus) junct. — The mature sperm has two differ- constituted approximately 17%. The head ently sized crystalloid mitochondrial de- 64 Journal of Hymenoptera Research rivatives (Figs. 1 & 3), which run for the length of the tail. The larger of the two mitochondrial derivatives also runs into the head of the sperm, parallel to the nu- cleus (Fig. 7). In transverse section (Fig. 8) it can be seen that the mitochondrial de- rivative can occupy a greater proportion of the shaft area than the nucleus itself (Fig. 8, arroived). The mitochondrial derivatives are de- rived from a Nebenkern, the product of the fusion of large numbers of small mi- tochondria in the early spermatid. Unlike other sperm so far reported, the Neben- kern goes through a stage where the mi- tochondrial material forms a tube, before producing the asymmetrically-sized mito- chondrial derivatives (Fig. 9, arrowed). The association between the larger mitochon- drial derivative and the nucleus is evident from the earliest stages of nuclear shape change (Fig. 10). Mitochondrial material appears to be present at the locus of this change. The centriolar adjunct in the adult sperm is positioned between the smaller mitochondrial derivative and the nucleus (Fig. 1). During development there is a close association between the centriolar adjunct and the developing axoneme (Fig. 11). In the cells where with multiple axo- nemes additional centriolar adjuncts are also present (Fig. 12) indicative of the close association between the two organ- elles. Nucleus. — Axoneme development corn- menses before nuclear elongation which follows a similar pattern to that described for the braconid Aleiodes, with the forma- tion of lateral plates (Fig. 10) opposite ag- gregations of dense chromatin strands (cf Newman & Quicke 1998). Acrosome. — The acrosome forms from fusion of small vesicles at the posterior Figs. 1-2. Mature sperm and early spermiogenesis of Megalyra: 1, transversely sectioned vas deferens with mature sperm with axonemes (A), nucleus (N), asymmetrically sized mitochondrial derivatives (M) — includ- ing some with only one mitochondrial derivative (arrowed) — and centriolar adjunct (C) in some sections where it occupies the position of the smaller mitochondrial derivative just posterior to the nucleus (scale bar = 100 nm); 2, secondary spermatocytes forming a syncitium (S) because of incomplete cytokinesis following the earlier mitotic division, the thickened membrane (arrowed) indicating intercellular bridges (scale bar = 0.5 ixm). Figs. 3-6. Various stages in spermiogenesis of Megalyra: 3, transverse section of the tail portion of mature sperm illustrating the 9 + 9 + 2 arrangement of microtubules, with accessory filaments evident, particularly linking the central pair of microtubules (arrow), and with two deltoid bodies (D) in association with the mitochondrial derivatives (scale bar == 100 nm); 4, early spermatid illustrating the growth of the axoneme (arrowed) from a position at the posterior pole of the nucleus (N) (scale bar = 500 nm); 5, early spermatid showing ring centriole (R) surrounding the basal body of the developing axoneme (scale bar = 250 nm); 6, during development many cell sections show multiple axoneme profiles (A), closer than would be expected if they represented sections through a single convoluted structure — note disruption of one of the profiles (arrowed) (scale bar = 100 nm). Figs. 7-10. Various stages in spermiogenesis of Megalyra: 7, longitudinal section of the sperm head piece with the larger mitochondrial derivative (M) extending from the tail portion, where it is found adjacent to the axoneme, and then running parallel to the nucleus (arrowed) (scale bar - 0.5 |xm); 8, transverse section through mature spermatozoan illustrating differences in the size of the nucleus (N) and of the large mito- ( hondrial derivative (M) in the head portion of the sperm in the region of overlap between the two (scale bar LOO nm); 9, transverse sec lion ot early spermatid illustrating a stage during formation of the mitochondrial derivatives when the Nebenkern becomes tubular (arrowed) (si ale bar I |im); 10, one mitochondrial deriv- ative (arrrowed) lies partly in a groove along the nucleus during earlier stages ol development note also the lateral plates (between arrowheads) which appear to be anchor silt's for condensed chromatin during nuclear shape change (scale bar = 0.5 (xm). ftb. Volume 9, Number 1, 2000 65 m- *:...< ■- ■ fc^gj^i *-vT? *■'* v* ;-,f'u ■ r ■ v - . • . ■ v " *' ' '• 66 Journal of Hymenoptera Research Volume 9, Number 1, 2000 67 68 Journal of Hymenoftera Research pist j s ■ i?i > *t* &i$i «w i r i # ■ ry ft? »> j**Ak» ■ ■ ■ -Wj| Figs. 11-14. Various stages in spermiogenesis of Megalyra: 11, transverse section oi spermatid illustrating how the centriolar adjunct (C) is closely associated with the axoneme (A) during development (scale bar = LOO nm); 12, developing cyst ol spermatids showing that cells with multiple axonemes have accompanying centriolar adjuncts (scale bar = 1 pun); 13, during development the acrosome (A) is formed from the fusion of small vi'su Irs and becomes positioned .it the anterior pole of the nucleus (scale bar = 0.5 (xm); 14, acrosome in lougitidin.il section (A) covering the perforatorium (arrowed) (scale bar 100 nm). Volume 9, Number 1, 2000 69 pole of the nucleus (Fig. 13). Although it develops a perforatorium (Fig. 14, arrow), the acrosome is small and ill-developed in comparison with the rest of the sperm and was very hard to find despite searching hundreds of transverse sections. DISCUSSION The presence of two markedly differ- ently sized mitochondrial derivatives has been reported in other parasitic wasps' spermatozoa, e.g. Leptopilina, which be- longs to the only distantly related super- family Cynipoidea (Newman and Quicke 1999b). We have also found a similar ar- rangement in the sperm of the xyelid saw- fly, Xyela julii (Newman and Quicke 1999a). Whether this feature could be con- sidered 'primitive' is doubtful given that other sawflies (Cephnlcia of the Pamphili- idae and Tremex of the Siricidae) have equally sized mitochondrial derivatives. The large length of overlap between one of the mitochondrial derivatives and the nucleus has only been found in sperm of one other parasitic wasp, the distantly re- lated chalcidoid wasp genus Trichogramma (Lingmei & Dunsu 1987). The cynipoid, Leptopilina, which also has asymmetric sized mitochondrial derivatives, also has an overlap of nucleus and mitochondrial derivative, but over a much smaller dis- tance (Newman & Quicke 1999b). This intimate relationship of mitochon- drial material and nucleus exists from an early stage of spermiogenesis. The mito- chondrial material is found at the focus of the nuclear shape change which occurs during cell elongation, where the nucleus curves into a horse-shoe shape, a shape change similar to that found in Aleiodes (Newman & Quicke 1998). One aspect of development which has not been previ- ously reported is the occurrence of tubular elements in the Nebenkern after fusion of the small mitochondria of the early sper- matid. This is probably an apomorphic character state and it may be of potential phylogenetic significance within the Eva- niomorpha. ACKNOWLEDGEMENTS Ms Judy Moore most kindly collected the Megalyra in South Africa and went to considerable trouble to get them couriered to the U.K. alive for studv. The authors gratefully acknowledge the assistance ot \K Anne Dewar (EM Unit, Royal Brompton Hospital). This research was supported by the NERC (Natural Environment Research Council), Initiative in Taxon- omy and a grant from the Leverhulme Trust to DLJQ and Mike Fitton (The Natural History Museum). LITERATURE CITED Austin, A., D. L. J. Quicke and P. M. Marsh. 1994. Systematics of the hymenopteran parasitoids of the eucalypt longicorns, Phoracantha spp. (Cole- optera: Cerambycidae) in Australia, with a de- scription of a new doryctine genus (Hymenop- tera; Braconidae). Bulletin of Entomological Re- search 84: 145-174. Dowton, M. and A. D. Austin. 1994. Molecular phy- togeny of the insect order Hymenoptera: Apocri- tan relationships. Proceedings of the National Acad- emy of Sciences USA, 91: 9911-9915. Gibson, G. A. P. 1985. Some pro- and mesothoracic structures important for phylogenetic analysis of Hymenoptera, with a review of terms used for the structures. The Canadian Entomologist 117: 1395-1443. Heraty, J. M., J. B. Woolley and D. C. Darling. 1994. Phylogenetic implications of the mesofurca and mesopostnotum in Hymenoptera. Journal of Hy- menoptera Research 3: 241-277. Jamieson, B. G. M. 1987. The Ultrastructure ami Pint- logon/ of Insect Spermatozoa. Cambridge Univer- sity Press, Cambridge, 320pp. I ensky, Y, E. Ben-David, and H. Schindler. 1979. Ul- trastructure of the spermato/oan of the mature drone honeybee. Journal of Apiculture Researt h Is 264-271. Lingmei, D. and W. Dunsu. 1987. Ultrastructural ob- servations on the spermatozoa of Trichogramma. Acta Zoologica Sinica 33: 262-266. Newman, T. M. and D. L. J. Quicke. 1998. Sperm de- velopment in the imaginal testes of Aleiodes cox- alis (Hymenoptera: Braconidae: Rogadinae). Jour- nal of Hymenoptera Research 7: 25-37. Newman, T. M. and D. L. |. Quicke. 1999a. Ultrastruc- ture oi imaginal spermatozoa ol sawflies ill\ menoptera: Symphyta). Journal oj Hymenoptera Research 8: 35-47. Newman, T. M. and D. L. J. Quicke. L999b. Ultra- structure oi spermatozoa in leptopilina (Hyme- noptera: Cynipoidea: Eucoilidae). Journal of Hy menoptera Research 8: 197-203. 70 Journal of Hymenoptera Research Quicke, D. L. J., S. N. Ingram, H. S. Baillie and P. V. Gaitens. 1992. Sperm structure and ultrastructure in the Hymenoptera (Insecta). Zoologica Scripta 21: 381-402. Rasnitsyn, A. P. 1988. An outline of evolution of the hvmenopterous insects (Order Vespida). Oriental Insects 22: 115-145. Ronquist, F., A. P. Rasnitsyn, A. Roy, K. Eriksson, and M. Lindgren. 1999 Phytogeny of the Hymenop- tera: a cladistic reanalysis of Rasnitsyn's (1988) data. Zoologica Scripta 28: 13-50. Shaw, S. R. 1990. Phylogeny and biogeography of the parasitoid wasp family Megalyridae (Hymenop- tera). Journal of Biogeography 17: 569-581. Thompson, T. E. and M. S. Blum. 1967. Structure and behaviour of spermatozoa of the fire ant Solenop- sis saevissima (Hymenoptera: Formicidae). Annals of the Entomological Society of America 60: 632-642. Vilhelmsen, L. 1997. The phylogeny of lower Hyme- noptera (Insects), with a summary of the early evolutionary history of the order, journal of Zoo- logical Systematics and Evolutionary Research 35: 49-70. Wheeler, D. E., E. G. Crichton and P. H. Krutzsch. 1990. Comparative ultrastructure of ant sperma- tozoa (Formicidae: Hymenoptera). journal of Mor- phology 206: 343-350. J. HYM. RES. Vol. 9(1), 2000, pp. 71-84 Afrotropical Ants (Hymenoptera: Formicidae): Taxonomic Progress and Estimation of Species Richness Hamish G. Robertson Life Sciences Division, South African Museum, P.O. Box 61, Cape Town 8000, South Africa; e-mail: hroberts@samuseum.ac.za Abstract. — Forty-three of the 82 Afrotropical ant genera (52%) have been revised to modern standards resulting in a 50% increase in number of species. There are currently 1705 species of ants known from the Afrotropical region, a figure that could increase to over 2136 species if all ant genera receive a modern revision. To incorporate all undescribed species, total Afrotropical ant species richness was calculated by extrapolating from data on the proportion of undescribed species collected at Mkomazi Game Reserve in Tanzania and the Cape of Good Hope section of the Cape Peninsula National Park in South Africa. On this basis there are an estimated 3105 species of ants in the Afrotropical region, with 45% undescribed or currently occupying an infraspecific taxonomic rank. This extrapolation assumes that the average range sizes of described and unde- scribed species are similar, which in reality is unlikely because widely distributed species are more likely to have been collected and described. I present a method that distinguishes between widespread and localised species to correct for this problem, which extrapolates 4093 Afrotropical ants species, with 58% of species estimated to be undescribed or currently recognised only at infraspecific rank. It would take a highly productive systematist at least 21 years to complete revisions of the unrevised ant genera. A strategy is presented for improving specimen collection and taxonomy of Afrotropical ants. Until recently, working on the taxono- my of ants and identifying them to genus was hindered by poor, out-dated cata- logues and inadequate keys. Ant system- atists, however, now have three books that provide a synthesis of our current knowl- edge: Bolton has produced keys to ant genera of the world (Bolton 1994) and a catalogue of world ants (Bolton 1995b) and Ward et al. (1996) have provided a thoroughly researched bibliography of ant taxonomic literature. In addition, the book by Shattuck (1999) makes identification of Australian ant genera much easier than before. Despite the relative ease with which ants can now be identified to genus, spe- cies identification is much more difficult because many ant genera have not been recently revised. Species-level identifica- tions in recently revised genera can also be problematic when such revisions are based on collections with limited geo- graphic coverage that omit species and population variation. Older taxonomic works (mainly pre-1965) are difficult to use for identifying species because they are often burdened with poorly applied quadrinomials (genus, species, subspecies, variety) that do not correspond with evo- lutionary relationships. With the catalogue by Bolton (1995b), the bibliography by Ward ct al. (1996), and a wealth of new material collected using modern survey methods, the tools are available to tackle the revision of neglected ant genera. How- ever, at present there is no strategy in place for accomplishing this remaining work and basic information on approxi- mately how much work remains is nec- essary for establishing goals and priorities. Although Bolton (1995a) has provided a 72 Journal of Hymenoptera Research taxonomic and zoogeographical census of alternative approach, which involves dis- documented ant diversity, he did not as- tinguishing between widespread and lo- sess the current taxonomic health of all calised species, is presented here to ad- ants, nor does he address the issues of the dress the problem of range size differences proportion of ant genera that lack modern between described and undescribed spe- taxonomic treatment, and the effort that cies. such treatment will require. The Afrotropical region is defined here In addition to our inability to estimate according to Bolton (1994) as Africa south the effort that will be required to revise all of the Sahara and the southern half of the ant genera to modern taxonomic stan- Saudi Arabian Peninsula. Madagascar and dards, we are also unable to reliably esti- its nearby islands are excluded. Our mate how many ant species exist and how knowledge of the ant fauna of this region many are undescribed. On a world level, is the product mainly of the following tax- Gauld & Bolton (1988) estimated 15,000, onomists (see Ward et al. 1996 for publi- and Holldobler and Wilson (1990) esti- cation details): F. Smith (1851-1879); Mayr mated 20,000 species in total, these esti- (1853-1908); Forel (1869-1920's); Emery mates being largely intuitive. By the end (1869-1926); Santschi (1906-1941); Arnold of 1993, 9538 described valid ant species (1905-1962, including "A monograph of in the world were recognised (Bolton the Formicidae of South Africa", pub- 1995a), so according to the above two es- lished 1915-1926); Brown (1943-1995); and timates, 36-52% of ant species either re- Bolton (published 1969 to present), main undescribed or currently occupy an The present paper is aimed at establish- infraspecific rank. ing the current level of taxonomic knowl- Estimates of total diversity are often edge of Afrotropical ant species by assess- made using data from particular localities ing the proportion of ant genera that re- where it has been possible to estimate the main to be revised to modern taxonomic proportion of species that are unde- standards and the effort required to un- scribed, and to use this proportion to ex- dertake these revisions. In addition, I es- trapolate to a wider level (e.g. Hodkinson timate total ant species richness in the Af- and Casson 1991, Hodkinson 1992 for He- rotropical region using data on the pro- miptera). This approach implies that the portion of undescribed species collected at average range sizes of described and two widely separated localities in Africa, undescribed species are similar (Hodkin- namely Mkomazi Game Reserve in Tan- son and Hodkinson 1993; Hammond zania and the Cape of Good Hope section 1995). In practice, this is highly unlikely as of the Cape Peninsula National Park in widely distributed species are more likely South Africa. At both sites, ants have been to be captured and described than species intensively collected using a variety of with more localised distributions. Hodkin- methods, thereby increasing the probabil- son and Hodkinson (1993) examined this ity that the observed ratio of undescribed problem by comparing data from two to described species is a good estimate of sites and providing a statistical test of the ratio for those species that have not whether the probability of capture of de- been collected. I conclude by suggesting a scribed species was different from the strategy for improving both the collecting probability of capture of undescribed spe- and the taxonomic treatment of Afrotrop- cies and then using the ratio of these two ical ants, probabilities to adjust the final estimate of species diversity. The outcome of such a METHODS comparison depends largely on the simi- Estimation of species number increment larity and proximity of the two sites. An from revision of genera. — I divided the Af- Yfti«*i Volume 9, Number 1, 2000 73 rotropical ant genera between those that have received a modern revision (defined here as from 1965 onwards) and those that have not; revisions prior to 1965 are either incomplete or use the old quadrinomial system. Genera represented only by intro- duced species (Linepithema humile (Mayr), Wdsmannia auropunctata (Roger)) have been excluded. Genera that have received a modern revision are henceforth referred to as 'revised genera' and the remaining genera as 'unrevised genera'. For each re- vised genus I calculated the species incre- ment coefficient, i.e. the number of valid species divided by the number prior to re- vision. Descriptions of new species and the raising of subspecies to species in- creased the increment coefficient whereas synonymising of previously valid species decreased it. Lowering of rank from spe- cies to subspecies was not encountered in any of the revisions. In genera which have received more than one revision since 1970 (e.g. Ocymyrmex), I calculated the number of valid species before the first re- vision and the number by completion of the last revision. Subsequent papers de- scribing additional new species in revised genera were also included in the analysis. An increment coefficient for all revised genera was calculated based on the total number of valid species before and after for all revisions and new descriptions. For unrevised genera, I counted the number of valid species and multiplied this by the increment coefficient for re- vised genera to arrive at an estimate of the total number of species after revision. The total species for the revised genera plus this estimated value for the unrevised genera gives the 'Total estimated revised species'. Estimation of total species richness. — An estimate of total ant species richness in the Afrotropical region was obtained by de- termination of the proportions of de- scribed and undescribed species in Mkom- azi Game Reserve, Tanzania, and in the Cape of Good Hope section of the Cape Peninsula National Park, South Africa. Ants were collected in Mkomazi Game Reserve using pitfall traps, malaise traps, Winkler bag leaf litter extractions, soil sieving, light traps, sweeping, and collect- ing by hand (Robertson 1999). Mkomazi Game Reserve has a wide range of vege- tation types such as grassland, open and closed woodland, and hilltop forest. It lies in a region that does not have a long his- tory of ant collecting although the vege- tation types it contains have been sampled in other regions such as Kenya and Zim- babwe. The Cape of Good Hope section of the Cape Peninsula National Park consists of mesic mountain fynbos, west coast strand- veld and a few relict small patches of in- digenous evergreen forest. Because of its position near the southern tip of Africa, many naturalists have collected in the Cape Peninsula and one might thus expect the ant fauna to be well known. However, many cryptic species went unnoticed until the recent systematic use of collecting methods such as pitfall trapping and Winkler bag leaf litter extraction. For each of these two localities, I deter- mined the ratio of undescribed species that belong to revised genera: Therefore U/K = u/k U = u(K/k) where U is the number of undescribed species in the Afrotropical region, u is the number of undescribed species at the lo- cality, K is the number of known (i.e. val- id) species in the Afrotropical region and k is the number of known species at the locality. As already discussed, this ap- proach assumes that the average range size of known and undescribed species is similar (Hodkinson and Hodkinson 1993; Hammond 1995) which in reality is un- likely because widely distributed species would be more likely to have been col- 74 Journal of Hymenoptera Research lected and described. In order to reduce years from 1972 to 1987 he worked single- the bias, I arbitrarily categorised all mindedly at revising various ant genera, known Afrotropical species in revised For each of his taxonomic publications genera between 'widespread species' that during this period, irrespective of whether have recorded distributions over three or they concerned Afrotropical ants, I record- more countries and 'localised species' that ed the number of valid species before the have been recorded from only one or two revision, the number of new species de- countries (based on information obtained scribed and the number of valid species from the revisions as well as from records resulting from the revision. The number of in the South African Museum ant com- valid species in an unrevised genus divid- puter database). The average range size of ed by the average number of initial valid the localised species is likely to be more species processed per year by Bolton, similar to the average range size of the un- gives the number of years it would take described species than that of all known to revise the genus (at maximum produc- species together. By reducing the bias in tivity). this wav, the ratio of undescribed to lo- calised species at an Afrotropical level should therefore be similar to the ratio of Estimation of species number increment undescribed to localised species at a local from revision of genera. — Modern revision level and therefore of 43 (52%) of the Afrotropical ant genera ... has resulted in an overall 50% increase in number of species, so that the number of where, for recently revised genera, L is the species at the completion of the revisions number of localised known species in the was 1.5 times greater than the initial num- Afrotropical region and 1 is the number of ber (Table 1). At the one extreme are gen- localised known species at a particular lo- era such as Anochetus, Psalidomyrmex, Pris- cality. tomyrmex and Platythyrea that decreased in Accumulation of described species as afunc- number of species as a result of species tion of collecting date. — As an alternative being synonymised in the revision. At the way of assessing the proportion of ant other extreme are genera such as Axinidris, species that are still undescribed, for a Cyphoidris, Ocymyrmex, Paedalgus, Pyrami- sample of revised genera I recorded the ca and Strumigenys that more than dou- year the type was collected for each valid bled in size mainly as a result of the de- species and plotted the accumulation of scription of new species, species as a function of collection date. There are 862 valid species in the 39 The collection date was not recorded for (48%) unrevised Afrotropical genera (Ta- many of the earlier types and for these I ble 2) and as the initial number of species used the publication date instead. The ear- in the revised genera amounted to 561 ly ant taxonomists such as Emery and For- (Table 1 ), based on the relative number of el usually provided descriptions for new species we are therefore about 39% of the species within one or two years subse- way through revision of the genera to quent to their collection. modern standards. Revision of the unre- Time taken to revise genera. — In order to vised genera over a similar time period as estimate the time that it would take to re- the revised genera would have swelled vise the unrevised ant genera, I analysed the number of species from 862 species to the productivity of the world's most pro- about 1293 (Table 2). Together with the ductive ant systematist, Barry Bolton, who 843 species in the revised genera (Table 1) has also done most of the revisionary there is a total of 2136 estimated revised work on Afrotropical ants. During the 16 species, an increase of 20% over the 1705 Ihnaut Volume 9, Number 1, 2000 75 Table 1. List of the Afrotropical ant genera that have received one or more modern taxonomic revisions (1965 onwards). The initial number of valid species before the first modern revision, the final number of valid species known at present, and the increment coefficient (used in Table 2) are shown. Increment coefficient Modem revisions and subsequent lniti.il species Final species (Final /Initial) publications of new specie- Afroxyidris 0 1 Belshaw and Bolton (1994) Agraulomyrmex 0 2 Prins (1983) Arikylomyrma 0 1 Bolton (1973b, 1981b) Anochetus 24 18 0.75 Brown (1978) Aphomothyrmex 2 1 0.50 Snelling (1979b) Apomyrma 0 1 Brown, Gotwald and Lev- ieux (1971) Atopomyrmex 2 3 1.50 Bolton (1981b); Snelling (1992) Axinidris 3 13 4.33 Shattuck (1991) Baracidris 0 2 Bolton (1981b) Bondroitia 3 2 0.67 Bolton (1987) Calyptomyrmex 13 16 1.23 Bolton (1981a) Camponotus (fulvopilosus-group) 2 4 2.00 Robertson (1990); Robertson and Zachariades (1997) Cardiocondyla 11 9 0.82 Bolton (1982) Cataulacus 38 39 1.03 Bolton (1974a, 1982); Snel- ling (1979a) Concoctio 0 1 Brown (1974a,b) Cyphoidris 1 4 4.00 Bolton (1981b) Decamorium 2 2 1.00 Bolton (1976) Dicroaspis 2 2 1.00 Bolton (1981a) Diplomorium 1 1 1.00 Bolton (1987) Dolioponera 0 1 Brown (1974c,d) Leptogenys 32 56 1.75 Bolton (1975a) Leptothorax 11 11 1.00 Bolton (1982) Melissotarsus 6 3 0.50 Bolton (1982) Meranoplus 9 8 0.89 Bolton (1981a) Messor 14 14 1 .00 Bolton (1982); Collingwood (1993) Micfadaceton 3 2 0.67 Bolton (1983) Monomorium 90 149 1.66 Bolton (1987) Ocymyrmex 12 37 3.08 Bolton (1981b); Bolton and Marsh (1989) Odontomachus 1 2 2.00 Brown (1976) Paedalgns 3 9 3.00 Bolton and Belshaw (1993) Petalomyrmex 0 1 Smiling (1979b) Platythyrea 15 14 0.93 Brown (1975) Plectroctena 13 17 1.31 Bolton (1974b); Bolton, Got- wald and 1 eroux (1979) Polyrhachis 43 47 1.09 Bolton (1973a) Pristomyrmex 6 5 0.83 Bolton (1981b) Prdbolomyrmex 3 3 1.00 Taylor (1965); Brown (1975) Pyramica 24 63 2.63 Bolton (1983), Bolton (1999) Psalidomyrmex 8 6 0.75 Bolton (1975b) Rhoptromyrmex 3 5 1.67 Bolton (1976, 1986) Simopone 7 9 1.29 Brown (1975); Kutter (1976, 1977) Sphinctomyrmex 1 2 2.00 Brown (1975) Strumigenys 17 42 2.47 Bolton (1983), Bolton (1999) Terataner 5 6 1.20 Bolton (1981b) Tetramorium 131 209 1.60 Bolton (1976, 1980, 1985) Total 561 843 1.50 76 Journal of Hymenoptera Research Table 2. Afrotropical ant genera that have not received a modern taxonomic revision. The number of estimated species after revision was calculated by multiplying the number of valid species currently known by the Increment coefficient (1.5) in Table 1. The minimum years to revise a genus is calculated on the basis that B. Bolton processed species (i.e. the number of species before a revision) at a rate of 42/year. Acropyga Aenictogiton Aenictus Amblyopone Anoplolepis Asphinctopone Camponotus (excl. fulvopilosus-gioup) Carebara Cataglyphis Centromyrmex Cerapachys Crematogaster Cryptopone Discothyrea Don/Ins Ecphorella Hypoponera Lepisiota Leptanilla Myrmicaria Mystrium Oecophylla Oligomyrmex Pachycondyla Paratrechina Phasmomyrmex Pheidole Pheidologeton Phrynoponera Plagiolepis Prionopelta Proceratium Pseudolasius Santschiella Solenopsis Streblognathus Tapinoma hi hnomynnex Tetraponera Total Number of Estimated species Years (minimum) valid species after revision to revise genus 2 3 0.05 7 11 0.17 34 51 0.81 3 5 0.07 19 29 0.45 3 5 0.07 156 234 3.71 11 17 0.26 1 2 0.02 5 8 0.12 24 36 0.57 129 194 3.07 1 2 0.02 7 11 0.17 57 86 1.36 1 2 0.02 36 54 0.86 45 68 1.07 3 5 0.07 22 33 0.52 1 2 0.02 1 2 0.02 33 50 0.79 53 80 1.26 13 20 0.31 4 6 0.10 66 99 1.57 7 11 0.17 5 8 0.12 18 27 0.43 3 5 0.07 5 8 0.12 5 8 0.12 1 2 0.02 10 15 0.24 1 2 0.02 13 20 0.31 25 38 0.60 32 48 0.76 862 1293 20.52 valid species currently known from the Afrotropical region. Estimation of total species richness. — The percentage of undescribed species within recently revised genera is 30.9% for ants in Mkomazi Game Reserve and 32.3% for ants in Cape of Good Hope Nature Re- serve (Table 3), remarkably similar values considering the distance between the two localities, the differences in their habitat complements and differing histories of ant collecting in the two regions. If these per- centages are extrapolated to the unrevised genera, and the 50% increase from revis- SttW Willi II III Volume 9, Number 1, 2000 77 Table 3. Ant species diversity and composition in Mkomazi Game Reserve in Tanzania, Cape of Good Hope section of the Cape Peninsula National Park (CGH) in South Africa, and in the Afrotropical region as a whole. Mkomazi ecu rotal Afrotropical All general: Total recorded species 232 72 303* 1705 Revised genera: No. widespread species 54 2 56 249 No. localized species 11 19 30 594 Total known species 65 21 86 843 % localized species 16.9 90.5 34.9 70.5 No. undescribed species 29 10 39 7 Total species 94 31 125 ? % undescribed species 30.9 32.3 31.2 ? Only one species (Technomyrmex albipes (F. Smith)) shared between the two localities. ing genera is taken into account, then the percentage of undescribed species for all genera is 44% for Mkomazi Game Reserve and 45% for Cape of Good Hope. Further collecting and analysis of ants at these lo- calities is still taking place so the percent- ages above could change. There are con- siderable differences between the two lo- calities in the proportion of known species in revised genera that have localised dis- tributions covering only one or two coun- tries (Table 3). Whereas only 16.9% of known species in Mkomazi Game Reserve have localised distributions, 90.5% of spe- cies in Cape of Good Hope Nature Re- serve are localised. Many of the Mkomazi species are widely distributed in African savannahs, often ranging from the north- ern regions of South Africa through to Ethiopia or with a Sahel distribution from West Africa to Ethiopia and down into East Africa (Robertson 1999). The forest dwelling species also often have distribu- tions extending into central Africa and other countries in East Africa. Conversely, many of the species in the Cape of Good Hope section of the Cape Peninsula Na- tional Park are limited to the Cape fynbos, or have distributions that extend only as far as Namaqualand or KwaZulu-Natal. In the Afrotropical region as a whole, 70.5% of species are localised (Table 3). Based on the combined data from both Mkomazi Game Reserve and Cape of Good Hope, simple extrapolation of the proportion of undescribed species (using the formula U = uK/k) produces an esti- mated total of 3105 species for the Afro- tropical region, whereas exclusion of the widespread species in the calculation of new species (using the formula U = uL/ 1) increases the total by 32% to produce an estimated total of 4093 species (Table 4). The estimated diversity using the latter formula on only the Mkomazi data is 6104 species, twice as high as the estimate based on only Cape of Good Hope, caused by the large difference between them in the ratio of widespread to localised spe- cies. Based on the combined data from both localities, approximately 45-58% of ant species in the Afrotropical region are un- described or are currently ranked as sub- species when they should be ranked as species (Table 4). Accumulation of described species as a func- tion of collecting date. — Analysis of the ac- cumulation of described, valid species as a function of collecting date (Fig. 1 ) shows that undescribed species are being discov- ered at an undiminishing rate, with the species accumulation curve showing no signs of plateauing. The curve shows that 78 Journal of Hymenoptera Research Table 4. Estimates of total ant species diversity in the Afrotropical region, based on the proportion of undescribed species in Mkomazi Game Reserve in Tanzania and in Cape of Good Hope section of the Cape Peninsula National Park (CGH) in South Africa (Table 3). The first estimate uses the ratio of total known species at a regional and local level and the second the ratio of localized known species (defined here as species that have been recorded from only one or two countries) at a regional and local level (see Methods for explanation of formulae). Estimated using data from: Mkomazi CGH Both sites 1. Revised genera, based on total known species No. undescribed species (=uK/k) 376 Total species 1219 Ratio Total /Known 1.4462 2. Revised genera, based on localized species No. undescribed species ( = uL/l) 1566 Total species 2409 Ratio Total /Known 2.8577 Estimated total species in all genera1 Based on (1) above 3089 Based on (2) above 6104 % new species or species currently at infraspecific rank2 Based on (1) above 44.8 Based on (2) above 72.1 401 382 1244 1225 1.4762 1.4535 313 772 1156 1615 1.3709 1.9160 3153 3105 2928 4093 45.9 45.1 41.8 58.3 = (Total estimated revised species) X (Total /Known). Total estimated revised species = 843 + 1293 2136 (See Tables 1 & 2). 2 = ((Estimated - Known)/ Estimated) X 100. The number of known species = 1705. 600 n 1750 1775 1800 1825 1850 1875 1900 Year collected 1925 1950 1975 2000 Fig. 1. Accumulation of new Afrotropical species in selected genera in relation to collection date of the type (publication date used where collection date absent). Information from Bolton (1974a, 1980, 1981a, 1981b, 1982, 1983, 1987). Only revisions from 1980 onwards included except lor Cataulacus (Bolton L974a) which was updated in the 1982 publication. linWr: Volume 9, Number 1, 2000 79 Table 5. Number of ant species revised and de- scribed by B. Bolton over a 16 year period (includes an extra year to account for the preparation time for the 1973 publications). Initial species refers to the number of valid species before the revision began. Average number initial species processed per year = 672/16 = 42. Year published Initial species No. new species No. species after revision 1973 43 10 48 1974 29 7 33 1975 67 29 88 1976 63 14 72 1977 52 37 86 1978 0 0 0 1979 23 17 38 1980 102 63 176 1981 55 23 74 1982 87 16 81 1983 44 65 107 1984 0 1 1 1985 0 0 0 1986 0 0 0 1987 107 46 167 Total 672 328 971 there was a burst of collecting from 1910 to 1925, overlapping with the period when George Arnold was mainly active. Follow- ing this burst, species accumulation in- creased at a slower rate until the 1970's when a new generation of ant collectors started using collecting techniques such as pitfall trapping, tree fogging and Winkler bag leaf litter extractions that produced species that had previously gone unde- tected. These techniques are still yielding many new species and additional tech- niques such as soil sifting should start tap- ping into the subterranean species that are still poorly known. Time taken to revise genera. — Over a 16- year period, Barry Bolton revised genera that initially contained a total of 672 spe- cies, amounting to processing 42 initial species per year (Table 5). The total of 971 species after revision amounts to describ- ing and characterising a species every four working days. Over such a long period and taking into account other concerns and responsibilities, this is a formidable rate of species processing that is unlikely to be equalled or bettered by anyone working at an equivalent level of thor- oughness. Based on Bolton's level of productivity, it would take about 21 years for one per- son to revise the remaining unrevised gen- era (Table 2). The genera Camponotus, Cre- tnatogaster, Dorylus, Lepisiota, Pachycondyla and Pheidole would each take more than a year to revise. DISCUSSION Afrotropical ant species richness. — The method presented here, that of excluding widespread species in calculating the pro- portion of new species, has not been pre- sented before and is an attempt (following on Hodkinson and Hodkinson 1993) to address the problem of differences in the average geographical distribution of known and undescribed species. The es- timate of 4093 Afrotropical ant species produced by using this method is 32% greater than the 3105 species estimated us- ing the conventional ratio of undescribed to all known species. These estimates could be improved if more sites are in- cluded. In total, therefore, 45-58% of species are undescribed or currently incorrectly placed at subspecific rank. This range of values compares favourably with the 52% unknown species, calculated from H611- dobler and Wilson's (1990) estimate of 20000 species world-wide and the actual number (at the end of 1993) of 9538 spe- cies determined by Bolton (1995a). Shat- tuck (1999) states that the Australian ant species diversity might well be double that currently known, which also matches the estimates presented here for the Af- rotropical region. At a local level, Watt ei al. (1997) estimated that 40% of the ants they captured in Mbalmayo Forest Re- serve in southern Cameroon were unde- scribed which compares favourably with the 44-45% undescribed species for all 80 Journal of Hymenoptera Research genera recorded at the two localities in the present study. On the basis that 18% of the world's de- scribed ant species are found in the Afro- tropical region (calculated from Bolton 1995a), the species diversity estimates pre- sented here can be extrapolated to a world level to give an estimated world diversity of between 17250 and 22739 species. How- ever, as the Nearctic and Palaearctic ant faunas are much better known than those from other regions, the 18% Afro tropical ant species is likely to be an underesti- mate. Progress with sampling of ants in Africa. — Awareness about the threats to biodiver- sity have increased funding for inventory- based conservation research and as ants are a favoured indicator taxon (Andersen 1997), there has been a consequent tre- mendous recent growth in ant collections. Intensive sampling projects in the Afro- tropical region include: Mbalmayo Forest Reserve in southern Cameroon (Watt et al. 1997); coastal and interior forest in Gabon (Fisher, in prep.); Mkomazi Game Reserve in Tanzania (Robertson 1999); Cape of Good Hope, Robben Island, Brenton-On- Sea, Fairfield Farm near Napier, Kogel- berg Biosphere Reserve, and other sites in the Cape fynbos, South Africa (Robertson and co-workers, in prep.); Cape indige- nous evergreen forests (Fisher, in prep.); Mondi Estate in Kwazulu-Natal (Fisher, in prep.); and widespread pitfall trapping by E. Marais in Namibia. However, there are still enormous gaps in our coverage. Countries such as Angola, Mozambique, Malawi, Zambia, Central African Repub- lic, Ethiopia and Sudan have yet to be sampled using modern inventory tech- niques. Even the best sampled countries such as South Africa and Zimbabwe re- main patchily sampled and there is not one country in the Afrotropical region where ants have been sampled adequately in all major vegetation regions. Future strategies for collecting. — The best approach to adequately sampling the ants of a large area such as the Afrotropical re- gion is through intensive inventory-based sampling of particular localities by general collecting and a combination of replicated pitfall trapping, Winkler bag leaf litter ex- tractions, beating or sweeping of vegeta- tion, chemical knockdown of arboreal fau- na and soil sampling. The use of replicable sampling methods makes it possible to statistically compare sites using tech- niques described in Colwell & Codding- ton (1994) and Chazdon et al. (1998) and in this way to make scientifically based as- sessments of alpha, beta and gamma di- versity. Fisher (1996, 1998, 1999) has pio- neered this approach in Madagascar al- though only for leaf litter and ground fau- na. Recent studies, still unpublished (e.g. Fisher and Robertson in prep, for a site near Ambositra in Madagascar), have used a wider range of replicated sampling techniques. Superficial general collecting of many localities is of more limited value than the inventory approach although it is useful for providing distributional data. As ants are dominant and ecologically im- portant organisms in terrestrial ecosys- tems, growth of collections will also con- tinue due to the submission of specimens by ecologists and agricultural researchers for identification by ant systematists. Inventory-based assessments of areas for conservation using ants will ensure continued funding of scientifically-based ant collecting in the Afrotropical region, provided there remains backup by ant systematists. The areas to be sampled will be largely dictated by the conservation funding bodies and by the political stabil- ity of the areas that need assessment. Not- withstanding the political issues, the ne- glected countries such as those listed above, need attention. With this increased ant collecting, the need for more taxonom- ic work on ants will become all the more apparent. Current progress with Afrotropical ant tax- onomy.— Based on relative proportion o\ species, we are about 39% of the way ^( «!»-"■ Volume 9, Number 1, 2000 si through revision of the Afrotropical ant genera to modern standards and to revise the remaining genera would take one per- son 21 years to complete at 'Bolton speed'. On a world level, there are about 41 people currently working on the taxono- my of ants. Although this seems a large number, the productivity of most of these taxonomists is much less than that of B. Bolton and their work is often limited to regional faunas. In addition, a number of our key established 'global view' system- atists have either recently retired or are about to retire. We could end up with a situation similar to that in termite taxon- omy (Eggieton 1999) although we are like- ly to remain stronger in terms of number of systematists. The low taxonomic productivity of most ant systematists can be attributed largely to their occupation with other endeavours: ecological and biological research on their study organisms, administration, contract identification work, computer program- ming, and teaching duties. Revising ant genera at the rate that B. Bolton has achieved is therefore rarely attained and for most systematists one would need to settle for a speed of revision at best half or even quarter of Bolton's rate. However, there is room for improvement and I feel that ant systematists need to prioritise al- pha taxonomy and not let it take a back seat which seems to be increasingly the case. Unlike the situation in North America, Europe, South America and Asia, there is only one resident ant systematist in Africa and hence progress with documenting Af- rotropical ants will depend largely on the involvement of outside 'global-view' sys- tematists working on taxa that are repre- sented in the Afrotropical region. Strategies for advancing Afrotropical ant taxonomy. — The two goals of a strategy to improve ant taxonomy are firstly, to en- sure that the number oi ant systematists does not dwindle but remains stable or grows and secondly, to improve the effec- tiveness and productivity of current ant systematists. Regarding the first aspect, it is vital that the museums holding impor- tant ant collections are committed to em- ploying ant systematists. It is remarkable that the most important ant collection in the world at the Harvard Museum of Comparative Zoology has no full-time ant systematist committed to alpha taxonomy and has a curator that can work only part- time on this vital collection. The Natural History Museum in London, with the sec- ond largest ant collection in the world, should continue its support of a position in ant systematics once the present incum- bent retires. The South African Museum holds the largest ant collection in Africa and should also remain committed to sup- porting ant systematics, especially as it is important to maintain an ant identification service for applied entomologists in Afri- ca. Training is an important aspect of safe- guarding the future body of ant systema- tists because filling of positions in system- atics is usually controlled more by the quality of the candidate than by the group he/she works on. Hence, the contribution to training by ant systematists at univer- sities is essential to the future growth of ant systematics. Regarding the second component in the strategy, there are five ways in which the effectiveness and productivity of current ant systematists could be improved: (1) In order to cope with the conflict between projects geared to collection growth as op- posed to taxonomic projects, we need to make the latter a priority and plan time to spend on them. For instance, university lecturers often find it easiest to plan time for taxonomic work over the King vaca- tions. (2) Dedicated funding of ant taxo- nomic revisions along the same lines as the Australian Biological Resources fund- ing for catalogues, in which money is al- located in proportion to the size of the tax- on, would be ideal for improving goal-set- ting and productivity. In reality, this type 82 Journal of Hymenoptera Research of funding is rare because a taxonomic re- vision does not answer applied problems directly. (3) A more realistic approach to obtaining funding for ant taxonomy would be to link it to more easily obtained funding for applied field-based projects. Funding from these projects can be used for employing and training parataxono- mists for time-consuming sorting, mount- ing and curation of ants. Funding bodies should commit themselves to permitting a direct taxonomic component in the project so that there are funds to employ people to measure specimens and funds to visit overseas ant collections to examine types. Funding should also be built into these projects for storage and curation of the specimens. (4) There is a great need for training of, and exchange of ideas be- tween, established ant systematists, espe- cially the large number residing outside North America and Europe. Better com- munication via e-mail would help, but the funding of one or more training and plan- ning meetings would be ideal. (5) As there is still so much work to be done in revis- ing all Afrotropical ant genera (at least 21 man-years), it is important to prioritise groups for revision. In the Afrotropical re- gion, the unrevised groups encountered most frequently when identifying ants are Pheidole, Crematogaster and Camponotus. These groups also happen to be among the most diverse of the unrevised genera (Table 2) and are also among the most dif- ficult taxonomically, either because of worker polymorphism (Pheidole and Cam- ponotus) or because of a paucity of external morphological species-discriminating characters (Crematogaster). Not surprising- ly therefore, these groups have been avoided and to get them done quickly it would be best to develop a funded strat- egy- Ants are an economically and ecologi- cally important group in terrestrial eco- systems in the Afrotropical region and im- proving their taxonomy would in turn im- prove the networking of ecological, agri- cultural and behavioural ant research. The present study provides the information for planning a funded strategy to document the Afrotropical ant fauna. The challenge is to create a synergy between the differ- ent role players (systematists, ecologists, funding bodies) so that individual efforts are not swamped by the immensity of the job at hand. ACKNOWLEDGMENTS I thank Brian Fisher for his comments on the man- uscript and the following organisations and individ- uals for helping with field work and processing of samples: Royal Geographical Society, Tanzanian De- partment of Wildlife, Simon van Noort, Tony Russell- Smith, Daniel Mafunde, Omari Mohamed, Tandi Russell, Dawn Larsen, Cape Peninsula National Park (and prior to its formation, the management of the Cape of Good Hope Nature Reserve) and Miemsie Troskie. LITERATURE CITED Andersen, A. 1997. Ants as indicators of ecosystem restoration following mining: a functional group approach. Pp. 319-325 in P. Hale and D. Lamb (eds) Conservation Outside Nature Reserves. Bris- bane: University of Queensland. Arnold, G. 1915-1924. A monograph of the Formici- dae of South Africa. Annals of the South African Museum 14: 1-766. Arnold, G. 1926. A monograph of the Formicidae of South Africa. Appendix. Annals of the South Af- rican Museum 23: 191-295. Belshaw, R. and B. Bolton 1994. A new myrmicine ant genus from cocoa leaf litter in Ghana (Hy- menoptera: Formicidae). journal of Natural Histo- ry 28: 631-634. Bolton, B. 1973a. The ant genus Poh/rhachis F. Smith in the Ethiopian region (Hymenoptera: Formici- dae). Bulletin of the British Museum (Natural His- tory). Entomology 28: 283-369. Bolton, B. 1973b. A remarkable new arboreal ant ge- nus (Hym. Formicidae) from West Africa. Ento- mologist's Monthly Magazine 108: 234-237. Bolton, B. 1974a. A revision of the palaeotropical ar- boreal ant genus Cataulacus F. Smith (Hymenop- tera: Formicidae). Bulletin of the British Museum (Natural History). Entomology 30: 1-105. Bolton, B. 1974b. A revision of the ponerine ant genus Plectroctena F. Smith (Hymenoptera: Formicidae). Bulletin of the British Museum (Natural History). Entomology 30: 309-338. Bolton, B. 1975a. A revision of the ant genus l.epto- gem/s Roger (Hymenoptera: Formicidae) in the Ethiopian region with a review of the Malagasy Volume 9, Number 1, 2000 83 species. Bulletin of the British Museum (Natural History). Entomology 31: 235-305. Bolton, B. 1975b. A revision of the African ponerine ant genus Psalidomyrmex Andre (Hymenoptera: Formicidae). Bulletin of the British Museum (Nat- ural History). Entomology 32: 1-16. Bolton, B. 1976. The ant tribe Tetramoriini (Hyme- noptera: Formicidae). Constituent genera, review of smaller genera and revision of Triglyphothrix Forel. Bulletin of the British Museum (Natural His- tory). Entomology 34: 281-379. Bolton, B. 1980. The ant tribe Tetramoriini (Hyme- noptera: Formicidae). The genus Tetramorium Mayr in the Ethiopian zoogeographical region. Bulletin of the British Museum (Natural History). Entomology 40: 193-384. Bolton, B. 1981a. A revision of the ant genera Mer- anoplus F. Smith, Dicroaspis Emery and Calypto- myrntex Emery (Hymenoptera: Formicidae) in the Ethiopian zoogeographical region. Bulletin of the British Museum (Natural History). Entomology 42: 43-81. Bolton, B. 1981b. A revision of six minor genera of Myrmicinae (Hymenoptera: Formicidae) in the Ethiopian zoogeographical region. Bulletin of the British Museum (Natural History). Entomology 43: 245-307. Bolton, B. 1982. Afrotropical species of the myrmicine ant genera Cardiocondyla, Leptothorax, Melissotar- sus, Messor and Cataulacus (Formicidae). Bulletin of the British Museum (Natural History). Entomol- ogy 45: 307-370. Bolton, B. 1983. The Afrotropical dacetine ants (For- micidae). Bulletin of the British Museum (Natural History). Entomology 46: 267-416. Bolton, B. 1985. The ant genus Triglyphothrix Forel a svnonvm of Tetramorium Mayr (Hymenoptera: Formicidae). journal of Natural History 19: 243- 248. Bolton, B. 1986. A taxonomic and biological review of the tetramoriine ant genus Rhoptromyrmex (Hymenoptera: Formicidae). Systematic Entomol- ogy 11: 1-17. Bolton, B. 1987. A review of the Solenopsis genus- group and revision of Afrotropical Monomortum Mayr (Hymenoptera: Formicidae). Bulletin of the British Museum (Natural History). Entomology 54: 263-452. Bolton, B. 1994. Identification Guide to the Ant Genera of the World. Cambridge, Massachusetts: Harvard University Press, 222 pp. Bolton, B. 1995a. A taxonomic and zoogeographical census of the extant ant taxa (Hymenoptera: For- micidae). journal of Natural History 29: 1037-1056. Bolton, B. 1495b. A New General Catalogue of the Ants of the World. Cambridge, Massachusetts: Harvard University Press, 504 pp. Bolton, B. 1999. Ant genera of the tube Pacetonini (Hymenoptera: Formicidae). journal of Natural History 33: 1639-1689. Bolton, B. and R. Belshaw 1993. Taxonomy and bi- ology of the supposedly lestobiotic ant genus Paedalgus (Hym.: Formicidae). Systematic Ento- mology 18: 181-189. Bolton, B.^ W. H. Gotwald Jr. and J. M. Leroux 1979 ("1976"). A new West African ant of the genus Plectroctena with ecological notes (Hymenoptera: Formicidae). Annates de VUniversite Abidjan. Sene E (Ecologie) 9: 371-381. Bolton, B. and A. C. Marsh 1989. The Afrotropical thermophilic ant genus Ocymyrmex (Hymenop- tera: Formicidae). journal of Natural History 23: 1267-1308. Brown, W. L. Jr. 1974a. Concoctio genus nov. Pilot Reg- ister of Zoology Card No. 29. Brown, W. L. Jr. 1974b. Concoctio concenta. Pilot Reg- ister of Zoology Card No. 30. Brown, W. L. Jr. 1974c. Dolioponera genus nov. Pilot Register of Zoology Card No. 31. Brown, W. L. Jr. 1974d. Dolioponera fustigera species nov. Pilot Register of Zoology Card No. 32. Brown, W. L. Jr. 1975. Contributions toward a reclas- sification of the Formicidae. V. Ponerinae, tribes Platythyreini, Cerapachvini, Cviindromvrmecini, Acanthostichini, and Aenictogitini. Search. Agri- culture (Ithaca, N. Y.) 5(1): 1-115. Brown, W. L. Jr. 1976. Contributions toward a reclas- sification of the Formicidae. Part VI. Ponerinae, tribe Ponerini, subtribe Odontomachiti. Section A. Introduction, subtribal characters. Genus Odontomachus. Stadia Entomologica 19: 67-171. Brown, W. L. Jr. 1978. Contributions toward a reclas- sification of the Formicidae. Part VI. Ponerinae, tribe Ponerini, subtribe Odontomachiti. Section B. Genus Anochetus and bibliographv. Studia En- tomologica 20: 549-638. Brown, W. L. Jr., W. H. Gotwald Jr. and J. Levieux 1971 ("1970"). A new genus of ponerine ants from West Africa (Hvmenoptera: Formicidae) with ecological notes. Psyche (Cambridge) 77: 259- 275. Chazdon, R. L., R. K. Colwell, J. S. Penslow and M. R. Guariguata 1998. Statistical methods for esti- mating spei ies richness of woodv regeneration in primary and secondary rain forests of northeast- ern Costa Rica. Pp. 285-309 in F. Pallmeier and J. A. Comiskey (eds) Forest Biodiversity Research, Monitoring and Modeling: Conceptual Background and Old World Case Studies. Paris, France: Parthe- non. Collingwood, C. A. 1993. Description of the host ant of Cossyphodinus bremeri Ferrer sp. nov. (Hyme- noptera, Formicidae). Pp. 22t» 231 in 1 Ferrer and C. A. Collingwood, New species oi Cossyphodinus from Africa (Coleoptera, Tenebrionidae) and de- scription of the host ant Messoi ferreri sp. nov. 84 Journal of Hymenoptera Research (Hymenoptera, Formicidae). Entomofauna 14: 221-232. Colwell, R. K. and J. A. Coddington 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society of London. B. Biological Sciences 345: 101-118. Eggleton, P. 1999. Termite species description rates and the state of termite taxonomy. Insectes So- ciaux 46: 1-5. Fisher, B. L. 1996. Ant diversity patterns along an ele- vational gradient in the Reserve Naturelle Inte- grale d'Andringitra, Madagascar. Fieldiana: Zool- ogy 85: 93-108. Fisher, B. L. 1998. Ant diversity patterns along an ele- vational gradient in the Reserve Speciale d'Anjanaharibe-Sud and on the Western Masoala Peninsula, Madagascar. Fieldiana: Zoology 90: 39- 67. Fisher, B. L. 1999. Improving inventory efficiency: a case study of leaf-litter ant diversity in Mada- gascar. Ecological Applications 9: 714-731. Gauld, I. and B. Bolton (Eds) 1988. The Hymenoptera. Oxford: Oxford University Press, xii + 322 pp. Hammond, P. M. 1995. Described and estimated spe- cies numbers: an objective assessment of current knowledge. Pp. 29-71 in D. Allsopp, R. R. Col- well and D. L. Hawksworth (eds) Microbial Di- versity and Ecosystem Function. Wallingford, UK: CAB International. Hodkinson, I. D. 1992. Global insect diversity revis- ited, journal of Tropical Ecology 8: 505-508. Hodkinson, I. D. and D. Casson 1991. A lesser pre- dilection for bugs: Hemiptera (Insecta) diversity in tropical rain forests. Biological journal of the Liu- nean Society 43: 101-109. Hodkinson, I. D. and E. Hodkinson 1993. Pondering the imponderable: a probability-based approach to estimating insect diversity from repeat faunal samples. Ecological Entomology 18: 91-92. Holldobler, B. and E. O. Wilson 1990. The Ants. Cam- bridge, Massachusetts: Harvard University Press, xii + 732 pp. Kutter, H. 1976. Beitrag zur kenntnis der gattung Sim- opone Forel. Mitteilungeu iter Schweizerischeu Eu- tomologischen Gesellschaft 49: 273-276. Kutter, H. 1977. Zweiter beitrag zur kenntnis der gat- tung Simopone Forel. Mitteilungen der Schweizer- ischen Entomologischen Gesellschaft 50: 173-176. I'nns, A. | 1983. A new ant genus from southern Af- rica (Hymenoptera, Formicidae). Annals of the South African Museum 94: 1-11. Robertson, H. G. 1990. Unravelling the Campouotus fulvopilosus species complex (Hymenoptera: For- micidae). Pp. 327-328 in G. K. Veeresh, B. Mallik and C. A. Viraktamath (eds) Social Insects and the Environment. Proceedings of the 11th International Congress of ILISSI, 1990. New Delhi: Oxford & IBH Publishing Co. Robertson, H. G. 1999. Ants (Hymenoptera: Formici- dae) of Mkomazi. Pp. 321-336 in M. Coe, N. McWilliam, G. Stone and M. J. Packer (eds) Mkomazi: the Ecology, Biodiversity and Conservation of a Tanzanian Savanna. London: Royal Geograph- ical Society (with the Institute of British Geog- raphers). Robertson, H. G. and C. Zachariades 1997. Revision of the Campouotus fulvopilosus (De Geer) species- group (Hymenoptera: Formicidae). African Ento- mology 5: 1-18. Shattuck, S. O. 1991. Revision of the dolichoderine ant genus Axinidris (Hymenoptera: Formicidae). Sys- tematic Entomology 16: 105-120. Shattuck, S. O. 1999. Australian Ants: their Biology and Identification. Collingwood, Victoria: CSIRO Press, 226 pp. Snelling, R. R. 1979a. Three new species of the Pa- laeotropical arboreal ant genus Cataulacus (Hy- menoptera: Formicidae). Contributions in Science (Los Angeles) 315: 1-8. Snelling, R. R. 1979b. Aphomomyrmex and a related new genus of arboreal African ants (Hymenop- tera: Formicidae). Contributions in Science (Los Angeles) 316: 1-8. Snelling, R. R. 1992. Two unusual new myrmicine ants from Cameroon (Hymenoptera: Formici- dae). Psyche (Cambridge) 99: 95-101. Taylor, R. W. 1965. A monographic revision of the rare tropicopolitan ant genus Probolomyrmex Mayr (Hymenoptera: Formicidae). Transactions of the Royal Entomological Society of London 117: 345- 365. Ward, P. S., B. Bolton, S. O. Shattuck and W. L. Brown, jr. 1996. A Bibliography of Ant Systematica. Berkeley: University of California Press, 417pp. Watt, A. dJ N. E. Stork, P. Eggleton, D. Srivastava, B. Bolton, T. B. Larsen, M. J. D. Brendell and D. E. Bignell 1997. Impact of forest loss and regen- eration on insect abundance and diversity. Pp. 274-286 in A. D. Watt, N. E. Stork and M. D. Hunter (eds) Forests and Insects. London: Chap- man ^nd I lall. J. HYM. RES. Vol. 9(1), 2000, pp. 85-98 On the Identity of Pheidole vaslitii Pergande (Hymenoptera: Formicidae), a Neglected Ant from Baja California Philip S. Ward Department of Entomology, University of California, Davis, California 95616, USA; Fax: 530-752-1537; e-mail: psward@ucdavis.edu Abstract. — The ant Pheidole vaslitii Pergande has remained a taxonomic enigma since its descrip- tion over a hundred years ago from a series of workers collected in the Sierra San Lazaro, Baja California Sur, Mexico. A reexamination of the apparent type series in the USNM collection reveals that some of the specimens — including a major worker designated as "lectotype" by Creighton (1958) — are not true types and are not conspecific with P. vaslitii. Selection of a new lectotype of P. vaslitii secures its identity as a form closely related to P. hyatti. It differs from "typical" P. hyatti by the presence of more extensive sculpture on the head of the major and minor workers and by its shorter body appendages, but intermediate phenotypes occur in parts of Baja California and California. Based on current evidence P. hyatti is reasonably interpreted as a single, polytypic species, with both P. vaslitii and an infraspecific name — P. hyatti solitanea Wheeler, described from Point Loma, California — as newly recognized junior synonyms. In 1896 Pergande described a number of new ant taxa from northwestern Mexico, based on collections made by Eisen and Vaslit during a California Academy of Sci- ences expedition to the region (Pergande 1896). Most of the new ants were from Te- pic in the state of Nayarit, but a few spe- cies had been collected in the Cape Region of Baja California. The taxonomic status of most of these named forms has long since been resolved, but one of the species from Baja California — Pheidole vaslitii — has suf- fered an ongoing identity crisis. Despite pronouncements on P. imslitii by Forel (1901), Wheeler (1914), and Creighton (1958), its taxonomic status — whether it represents a distinct species or is a junior synonym — has never been satisfactorily determined. Creighton (1958) actually complicated the situation by designating as the "lectotype" of P. vaslitii a major worker that was not part of the series on which Pergande's original description was based. In preparing a checklist of the ants of Baja California (Johnson and Ward, in prep.) it became necessary to determine the relationship of P. vaslitii to other named taxa, and to deal with the conse- quences of Creighton's (1958) actions. It will be argued that Creighton's lectotype designation was invalid and that it inap- propriately tied the name "Pheidole vasli- tii" to a mainland Mexican species not found in Baja California. True P. z^aslitii appears to be a geographical variant of the widespread species Pheidole hyatti Emery. MATERIALS AND METHODS Specimens belonging to the original type series of Pheidole vaslitii were exam- ined from holdings of the California Acad- emy of Sciences, San Francisco (CASC) and the Smithsonian Institution, Washing- ton, DC (USNM). Other relevant material in the genus Pheidole was studied in those two collections and in the following: Nat- ural History Museum of Los Angeles County (LACM), California Department of Food and Agriculture, Sacramento (CDAE), Bohart Museum of Entomology, University of California at Davis (UCDC), and Robert A. Johnson collection, Tempe, 86 Journal of Hymenoptera Research Arizona (RAJC). Other collection abbrevi- ations cited in the text are: AMNH (Amer- ican Museum of Natural History, New York), MCSN (Museo Civico di Storia Na- turale, Genoa) and MCZC (Museum of Comparative Zoology, Harvard Universi- ty). All measurements were taken at 50 X using a Wild M5A microscope and a Ni- kon stage micrometer, and are presented here in millimeters to two decimal places. The following measurements and indices were used: HW HL EL SL BSW PrW I III CI SI REL Head width: maximum width of the head, as seen in full-face (fron- tal) view, excluding the eyes. Head length: length of the head, measured in full-face view, from the anterior clypeal margin to the midpoint of a line drawn across the posterior margin. Eye length: length of the eye, mea- sured with the head in full-face view. Scape length: chord length of the scape, from the base (excluding the neck) to the apex; this measure- ment was taken by positioning the scape so that both ends lay in the same focal plane. Basal scape width: maximum mea- surable width of the basal third of the scape (measured in major workers only). Pronotum width: maximum width of the pronotum, measured in dor- sal view. Length of the metatibia (hind tib- ia): length of the metatibia mea- sured in lateral view from the dis- tal extremity to the proximal end, excluding the medial lobe of the articulation with the femur (see Ward 1989, fig. 5). Cephalic index: HW/HL Scape index: SL/HW Relative eye length: EL/HL REL2 Relative eye length, using HW: EL/HW HTI Metatibial index: LHT/HW The following index of pilosity was em- ployed: HTC Metatibial setal count: number of standing hairs, i.e., those forming an angle of 45° or more with the cuticular surface (Wilson 1955), visible in outline on the outer (ex- tensor) surface of the metatibia. This count was taken with the line of view orthogonal to the plane of tibial flexion. TAXONOMIC HISTORY OF PHEIDOLE VASLITII Pergande (1896: 883) described Pheidole vaslitii from specimens collected by Eisen and Vaslit in the Sierra San Lazaro, Baja California Sur, in September 1894. The original series of twenty-two specimens was said to comprise nine major workers (or "soldiers") and thirteen minor workers ("workers"). In the same paper Pergande described several other Pheidole taxa, in- cluding one which he thought was related to P. vaslitii and which he named Pheidole obtusospinosa (Pergande 1896: 889). This latter species was from Tepic, Nayarit, however, and the description was based on a large series ("many specimens") of major workers only. Pheidole obtusospinosa was synonymized with P. vaslitii by Forel (1901: 430). Wheeler (1914) established, however, that P. obtusospinosa was the same as P. subdentata Pergande (1896), also described from Tepic, but based on minor workers only. By this time it was also clear that, unlike most Pheidole species, the ma- jor workers of P. subdentata are quite var- iable in size. Wheeler treated P. subdentata as a subspecies of P. vaslitii, a situation that continued until 1958 when Creighton assigned it species rank. Until recently this species has gone by the name P. subdentata Pergande, with P. obtusospinosa as a junior synonym, but Bolton (1995) pointed out tiiiLuitii Volume 9, Number 1, 2000 that Pergande's subdentata is preoccupied (it is a secondary junior homonym of Oec- ophthora subdentata Mayr 1853). P. obtusos- pinosa is the first available replacement name for the species described from Tepic, Nayarit. In the meantime the problem of the identity of Pheidole vaslitii was addressed by Creighton (1958) who examined Per- gande's type series in the USNM. He con- cluded that type series comprised more than one species, with most of the major workers — and all of those that matched Pergande's (1896) description of the P. vas- litii major — being Pheidole cockerelli Wheel- er (1908). The minor workers were said to be a mix of Pheidole crassicornis tctra Creighton (1950) and P. hyatti Emery (1895). Concerned about the replacement of P. cockerelli or P. crassicornis tetra by a more obscure senior synonym, Creighton chose as a lectotype of P. vaslitii a major worker that did not correspond to Pergan- de's (1896) original description. In fact, the identity and labeling of the specimen des- ignated as "lectotype" indicate that it was not part of the original type series of P. vaslitii (see below). Moreover, Creighton (1958) admitted that he could not say what species, if any, the "lectotype" rep- resented, i.e., he could not determine whether it belonged to any previously de- scribed species of Pheidole, except that it was not the species whose major worker was described by Pergande (1896) as P. vaslitii. This was the last action taken on P. vaslitii whose identity has thus re- mained in limbo for the last 40 years. Other Pheidole names associated at var- ious times with P. vaslitii are (1) P. hirtula Forel (1899), originally described as a va- riety of P. vaslitii, but later raised to spe- cies by Creighton (1958); (2) P. arizonica Santschi (1911), described as such, but treated as a variety of P. vaslitii subdentata by Wheeler (1914) and as a subspecies of P. vaslitii by Creighton (1950), before being synonymized under P. subdentata Pergan- de (Creighton 1958); and (3) P. acolhua 87 Wheeler (1914), originally described as a variety of P. vaslitii, but later synonymized under P. hirtula (Creighton 1958). REEXAMINATION OF THE TYPE SERIES Syntypes true and false The solution to the Pheidole vaslitii prob- lem lies in a careful analysis of the type specimens. In the USNM there are 14 spec- imens that appear, at first glance, to be part of the original type series. These spec- imens can be divided into two subsets (see also Creighton 1958: 208). Subset A comprises 11 specimens (4 major workers, 7 minor workers) each on their own pin and bearing two sets of la- bels (Fig. 1): (1) Pergande's handwritten locality label "Sierra /San Lazaro/Cape re- gion /L. Cal. Mex.", and (2) a printed label "Collection/T Pergande". Two specimens (one major and one minor worker) addi- tionally bear a third, black-bordered label with the following notation in Pergande's hand: "Pheidole/ vaslitii/n. sp./Type Perg.". On the label of the minor worker "n. sp." is placed in parentheses. As Creighton (1958) noted, the ink on the handwritten labels has faded to brown and the paper has yellowed. The major workers and most of the minor workers agree closely with Pergande's (1896) orig- inal description of P. vaslitii. Subset B consists of two major workers and one minor worker whose labeling is rather different (Fig. 2). The locality labels are handwritten, evidently by Pergande, in a black ink which has not, to this day, faded, and the label paper has not yel- lowed in color like that of subset A. The locality for the two majors is given as "Si- erra/S. Lazaro/Mex." and for the minor "Tepic/Mex.". All three specimens bear a red USNM type label as follows: "Type/ No. 4488/U.S.N.M.", with the number handwritten and the remaining text print- ed. One of the major workers also has a third label in Pergande's writing: "Phei- 88 Journal of Hymenoptera Research m Collection TPergancle h *")• fly. »£ m . X £ AA Figs. 1-2. 1, Sample of labels from specimens in "subset A" of the apparent type series of Pheidole vaslitii. The type label is from one of the major workers. See text for further details. 2, Sample of labels from specimens in "subset B" of the apparent type series of P. vaslitii. The identification label is from one of the major workers. All three specimens of "subset B" also bear a red USNM type label with the number "4488". See text for further details. dole/ vaslitii /Perg. Perg.". It is this speci- men that Creighton (1958) designated as the lectotype of P. vaslitii. None of these three "type" specimens matches the orig- inal description of P. vaslitii. In discussing the different appearance of the labels Creighton (1958: 209) sur- mised that Pergande had rewritten those of subset B when designating the speci- mens as types: "Dr. Smith informs me that Pergande noted in the type book of the National Museum that he had marked three specimens of vasliti as types. It would appear that when he did so he al- tered the locality labels of these three spec- imens, probably because he realized that the original labels were not holding up as well as might have been wished". Creighton (1958) did not mention — and perhaps did not examine — the minor worker from Tepic, but he freely admitted that the two major workers were not part of the series on which Pergande based his description of P. vaslitii: "They may be medias of a polymorphic species related to subdentata or the majors of a dimorphic one related to hyatti. But there is no doubt about one thing: neither of them contrib- uted anything to Pergande's description of the major of vasliti" (Creighton 1958: 210). As mentioned above, Creighton's choice of a lectotype that did not correspond to the species described by Pergande was motivated by a desire to maintain nomen- clatural stability: Creighton concluded that the species represented in subset A (i.e., the probable true syntypes of P. vas- litii) included P. cockerelli and P. crassicor- nis tetra and he did not want to see one of these younger (and better known) names relegated to synonymy. Insofar as the designated specimen was not a syntype of P. vaslitii, Creighton's (1958) choice of lectotype is invalid (ICZN, Article 74.2). Further evidence that his "lectotype" was not part of the original type series of P. iwslitii comes from con- sideration of the identity of the three spec- imens in subset B (Figs. 3-6): all three are Pheidole obtusospinosa Pergande, a species kBK Volume 9, Number 1, 2000 89 3 * «J 1 40 1 30 1 20 f E 1.10 _i (/) 1 00 0.90 hyatti Aobtusospinosa xvaslitn. subset A x "vaslitti" subset B 0 80 • A A a aA A A A A A A A A • • ** A* a * ♦ ♦ • ♦ ♦ ♦ ♦♦ X ' . i ■ • i 1 00 1 20 1 40 1 60 HW(mm) 1 80 2 00 /[ ♦hyatti Aobtusospinosa xvaslitn. subset A x "vashtii" subset B 0 90 085 - „4 ♦**♦ 0 80 - 5 0 75 in gg 0 70 - 0 65 - 0 60 0 50 #»♦« ♦♦♦ ♦ ♦ ♦ ♦ AA ^A £ *A A 060 070 080 PrW(mm) 0 90 1.00 ♦ hyatti Aobtusospinosa xvaslitn. subset A x "vaslitu" subset B w 1 40 1 30 1.20 ? - 1 10 i- i _i 1 00 090 0 80 A A ^ A AAA X . A^ -A A*1 A F a& A A ♦ • ♦ ♦ ♦ ♦ ♦ ♦*> J ♦♦ • ♦ ♦v 1 00 1 20 1 .40 1 60 HW(mm) 1 80 2 00 ♦ hyatti Aobtusospinosa xvaslitn subset A x"vaslitn". subset B 0 60 0 50 0 60 0 70 0 80 PrW (mm) 0 90 1 00 Figs. 3-6. Bivariate plots of various measurements and indices in major workers of Pheidole obtusospinosa and P. hyatti. Note that the P. vaslitii types from "subset A" fall within the values for P. hyatti, while those of "subset B" correspond to P. obtusospinosa. The plots exclude "supermajors" of P, obtusospinosa (HW > 2.20). which is widespread in adjacent mainland Mexico but which is not known to occur in Baja California. In the course of study- ing the ant fauna of Baja California I have examined many hundreds of specimens of Pheidole in various collections (CASC, CDAE, LACM, RAJC, UCDC, USNM). None of the Baja specimens belonged to P. obtusospinosa, although I encountered ex- amples of this distinctive species from So- nora, Sinaloa, Nayarit, Jalisco, and Arizo- na. In fact, the two major workers of sub- set B agree very closely with a series of seventeen P. obtusopsinosa majors from Te- pic, Nayarit (collected by Eisen and Vas- lit), which had been placed under P. vas- litii in the USNM collection. Given that this arrangement of specimens was due to Pergande, it indicates that he confused the two species. I conclude that Pergande's marking of the three specimens in subset B as "types" of P. vaslitii occurred after the original de- scription of that species, and that it in- volved the mislabeling of P. obtusospinosa specimens collected at Tepic, Nayarit. That Pergande was less than careful in these matters is indicated by the fact that in his original description of Pheidole gran- ulata he cites the type locality as Tepic (Pergande 1896: 891), although the type specimens were actually from San Jose del Cabo, Baja California Sur (Gregg 1969: 101). Removing the P. obtusospinosa speci- mens from consideration as valid types of P. vaslitii, we can refocus our attention on 90 Journal of Hymenoptera Research subset A which, it seems clear, contains the true syntypes. A study of the USNM specimens in sub- set A indicates that the following two spe- cies are involved: 1. A Pheidole species conspecific with, or closely related to, P. hyatti. This is rep- resented by four major workers and four minor workers. The majors were misidentified by Creighton (1958) as P. cockerelli, while he identified the minors as P. hyatti. In the collection of the Cal- ifornia Academy of Sciences (CASC) there are two additional P. vaslitii syn- types (one major worker, one minor) that belong to this species. They both bear faded labels "Sierra /San Lazaro" and "Pergande/Type", in Pergande's handwriting. 2. A second species of Pheidole, related to P. crassicornis Emery, represented by three minor workers. These were iden- tified by Creighton (1958) as P. crassi- cornis tetra. Designation of a new lectotype Given that subset A (in the USNM) and the two CASC specimens are part of the actual type series of P. vaslitii it is proper that the lectotype be chosen from among them. Of the two species present in the type series, only one is represented by ma- jor workers, so I have chosen as the lec- totype of P. vaslitii one of the major work- ers in the USNM series. This particular specimen (with HW 1.32, HL 1.40, SL 0.98, LHT 0.92) also bears the old faded "Phei- dole/vaslitii/n. sp./Type Perg." label. Formal lectotype designation is indicated below under "Taxonomic summary". The paralectotypes in USNM and CASC are then as follows: (1) four major workers (three in USNM, one in CASC) and five minor workers (four in USNM, one in CASC) conspecific with the lectotype, and (2) three minor workers (in USNM) not conspecific with the lectotype; these are discussed further below in the section en- titled "Identity of the second species in the type series". Specifically excluded from the paralectotype series are the three workers of P. obtusospinosa (subset B). With this action taken, the identity of P. vaslitii becomes linked with the question of the magnitude and nature of geograph- ical variation in Pheidole hyatti, a species widespread in the southwestern United States and northern Mexico. Relationship of Pheidole vaslitii to P. hyatti The major workers of P. vaslitii agree well with P. hyatti majors from other parts of Baja California and southwestern Unit- ed States. They have the base of the scape strongly bent and flattened (BSW/SL 0.12-0.14); the ventrolateral hypostomal teeth are well developed and spine-like, and displaced laterally about 0.8 times the distance from the midline to the dorsolat- eral hypostomal teeth; the posterior mar- gin of the head is deeply impressed me- dially, and this impression continues on the dorsum of the head as a narrow, trans- versely rugulose, median furrow which changes to a broader smooth, shiny strip above the frontal triangle; the mesonotum is angular in lateral profile; the propodeal spines are short, stout, and erect; the pet- iolar node is narrowly transverse; and the postpetiole is about 1.6 times broader than long, with obtuse, blunt lateral angles (as seen in dorsal view). Body sculpture is like that of P. hyatti, with the mesosoma largely foveolate and subopaque, the sculpture weakening medially and on the side of the pronotum; irregular transverse carinulae occur on the anterior portion of the pronotum. The head dimensions (HW 1.17-1.32, HL 1.25-1.40, CI 0.93-0.95), and relative lengths of the scapes (SL/PrW 1.36-1.50) and legs (LHT/PrW 1.35-1.43) fall within the values of P. hyatti and, in the case of the last two indices, largely outside those of P. cockerelli (see also Figs. 7-8). The body pilosity is relatively long and fine-tipped, not blunt-tipped as in P. Volume 9, Number I, 2000 91 1.15 1.10 1 05 £ 100 E 5J 095 |- 0.90 085 h 080 ♦ hyatti ■ cockerelli x vaslitn types «♦ ♦ ♦ t\ °a x>< X 052 056 0.60 0.64 0.68 0 72 0 76 0 80 PrW(mm) 078 ♦ hyatti ■ cockerelli x vaslitil types 0 52 0 56 0.60 0.64 0 68 PrW(mm) 072 076 080 9 1 04 1.00 0.96 f 0.92 E »j 0.88 084 080 076 ♦ hyatti ■cockerelli x vaslitii types *% - ♦ ♦ ♦ ♦ ♦ • X ♦ x XX ■ ♦ D ■ ■ ■ ■ ■ ■ - ♦ * ■ ■ ■ _ ■ ♦ - ♦ ♦ ■ 1 . 1 0 32 0 36 0 40 0 44 0 48 PrW(mm) 052 0.56 060 ♦ hyatti ■ cockerelli x vaslitn types 032 036 0 40 0 44 PrW(mm) 048 052 056 Figs. 7-10. Bivariate plots of various measurements and indices in major workers (Figs. 7-8) and minor workers (Figs. 9-10) of Pheidole hyatti and P. cockerelli. The P. vaslitii types (subset A) lie closer to the cloud of points representing P. hyatti. cockerelli (compare Figs. 12 and 13), and the tibiae are conspicuously hirsute (HTC 19-22, compared with HTC 1-17 in P. cockerelli). It may be that Creighton (1958) mistook these P. vaslitii majors for P. cockerelli be- cause the head sculpture is denser than is typical for P. hyatti. Fine reticulate-foveo- late sculpture extends to the posterolateral corners of the head so that the occipital lobes (normally shiny in P. hyatti) are dulled. Rugoreticulum extends more than half the distance from the upper margin of the compound eye to the occipital lobes. But the rugoreticulum is not as well de- veloped as in the majors of P. cockerelli, where it essentially covers the occipital lobes. Differences between P. cockerelli and P. vaslitii in pilosity, scape length, and leg length are also evident, and with respect to these characters the P. vaslitii types fall within the orbit of P. hyatti (Figs. 7-8, 13- 14, 17-18). The conspecific minor workers (4w in USNM, lw in CASC) agree with P. hyatti minors from other localities, although they tend to exhibit more extensive fove- olate sculpture on the head. This character varies widely, however, and when consid- ering P. hyatti minors from throughout the range of this species one finds all degrees of variation from an almost entirely smooth, shiny head (except between the compound eye and antennal insertions) to one dulled by extensive foveolate sculp- ture on all regions except medially. The P. 92 Journal of Hymenoptera Research 15 16 \x 17 1.0 mm 18 -I I Figs. 11-18. Pheidole major workers, lateral views of the mesosoma (11 14) and anterior views ot the null tibia (15-18). Pilosity shown in outline only. II, IS, /'. obtusospinosa (Tepic, Nayarit); 12, 16, /'. cockerelli (Yavapai Co., Arizona); 13, 17, /'. hyatti (/'. vaslitii paralectorype from Sierra San Lazaro, BCS); 14, 18, /'. In/nlti (Riverside Co., California). Volume 9, Number 1, 2000 93 vaslitii workers lack the weak rugoreticu- lum that typically occurs on the posterior portions of the head in minors of P. cock- erelli. The pilosity on these P. vaslitii minor workers is also more slender and flexuous than that of P. cockerelli. The tibial pilosity is conspicuous (HTC 14-19), like that of P. hyatti. Measurements and indices are well within the limits of P. hyatti: HW 0.63- 0.66, CI 0.82-0.84, SI 1.44-1.49, SL/PrW 2.01-2.08, REL 0.23-0.24, HTI 1.19-1.22, LHT 0.76-0.80, LHT/PrW 1.66-1.72 (n = 4 for the first six sets of values, because one of the USNM specimens is headless) (see also Figs. 9-10). Thus, Pheidole vaslitii falls within the range of variability that occurs in the widespread species, P. hyatti. In addition to denser head sculpture, the P. vaslitti types also tend to have shorter scapes and legs than P. hyatti from other regions (Figs. 3-10). The differences are not diagnostic, however, and individuals intermediate in morphology between P. vaslitii and more "typical" P. hyatti occur widely, especially on the Baja California peninsula and in California. It is possible that there is more than one biological species in this com- plex, but I can find no consistent pheno- typic discontinuities that would justify recognition of more than one taxon. With- out a more detailed (preferably genetic) analysis of the situation, it seems advis- able to treat P. hyatti as a single polytypic species, with P. vaslitii as its junior syno- nym. This is formally indicated below (un- der "Taxonomic summary"), along with the overdue synonymy of P. hyatti solita- nea Wheeler under P. hyatti. While it may be coincidental, it seems worth noting that other Pheidole species in Baja California show some evidence of north-south clines in the intensity of head sculpture. This is seen most strikingly in Pheidole yaqui Creighton and Gregg: pop- ulations of this species from California and northern Baja California have the pos- terior portion of the head of the major worker largely smooth and shining, while populations from farther south on the peninsula show increasing development of transverse rugulation on the posterior margin of the head (Ward, pers. obs.). Identity of the second species in the type series Three minor workers from the syntype series of P. vaslitii belong to a distinctly different species. The head is broader (HW 0.66-0.68, CI 0.88-0.90), and the scapes and legs are shorter (SI 1.18-1.21, HTI 1.00-1.02). The head is sublucid, with weak reticulate-foveolate sculpture over- lain by irregular longitudinal carinulae; the sculpture is less developed medially where there are more extensive shiny in- terspaces. The standing pilosity is short, sparse and blunt, with about 7-9 standing hairs visible in profile on the mesosoma dorsum, and none on the extensor surface of the tibiae. Creighton (1958) identified these work- ers as Pheidole crassicornis tetra Creighton. He actually cited the author as W. M. Wheeler (1908), but the name tetra did not become available until 1950 when Creigh- ton raised Wheeler's infrasubspecific name to subspecies rank (Bolton 1995). In the literature tetra has continued to be treated as a subspecies of P. crassicornis. Naves (1985) opined that it was a distinct species, similar to P. dwersipilosa Wheeler (1908), whereas Creighton (1950) empha- sized the existence of intermediates and considered all three names to refer to a single species. The three minor workers from Baja Cal- ifornia evidently belong to what could be called the P. erassieoniis complex. In com- parison with minor workers from Arizona and Texas, provisionally identified as P. dwersipilosa and P. erassieoniis tetra respec- tively, the Baja minors agree closely in overall habitus, pilosity and body mea- surements, but differ by having more con- spicuous longitudinal carinulae on the head, greater encroachment of reticulate- foveolate sculpture on the center of the 94 Journal of Hymenoptera Research pronotum, and slightly longer legs (LHT 0.66-0.70, HTI 1.00-1.02; compared with LHT 0.53-0.62 and HTI 0.89-0.97 in the Arizona and Texas material). With respect to pilosity they are most similar to the Ar- izona specimens, both having slightly shorter hairs than the Texas workers. In the Baja workers the length of the longest seta on the petiole is 0.090-0.104 mm, on the postpetiole 0.094-0.105 mm, and on abdominal tergite IV 0.081-0.096 mm. In the absence of any associated major work- ers— and given continued uncertainty about the relationship of crassicornis, tetra and diversipilosa to one another — it does not seem feasible to identify the minor workers any further at this time. TAXONOMIC SUMMARY Pheidole hyatti Emery 1895 (Figs. 13, 14, 17, 18) Pheidole hyatti Emery 1895. Syntype workers (majors, minors), San Jacinto, California (E. Hyatt) (AMNH, MCSN). Pheidole vaslitii Pergande 1896. Syntype work- ers, Sierra San Lazaro, Baja California Sur, Mexico (Eisen & Vaslit) (CASC: 1 major worker, 1 minor worker; USNM: 4 major workers, 4 minor workers); one major work- er in USNM here designated lectotype to clarify application of the name P. vaslilli (see previous discussion on page 90). Syn. nov. Pheidole hyatti var. ecitonodora Wheeler 1908. Syntype workers (majors, minors), Texas, New Mexico, Colorado (AMNH, LACM, MCZC). Synonymy by Creighton (1950: 180). Pheidole hyatti subsp. solitanca Wheeler 1915. Syntype workers (majors, minors), queens, Point Loma, San Diego, California (W. M. Wheeler; P. Leonard) (AMNH, MCZC). Syn. nov. Synonymy previously listed in a report by Snelling and George (1979: 104), not con- sidered a formal publication by Bolton (1995). P. coekerelli; Creighton (1958), misidentification of P. xmslitii syntypes (part) Diagnosis, major worker. — Medium-sized (HW 1.05-1.57; HL 1.14-1.63, LHT 0.81- 1 .09; n = 47); scape conspicuously flattened and bent basal ly (BSW/SL 0.09-0.15), of moderate length (SI 0.63-0.89, SL/PrW 1.28-1.79), not exceeding the posterior mar- gin of the head when laid back against the head; ventral hypostomal teeth well-devel- oped, spine-like, much closer to the dor- solateral hypostomal teeth than to the mid- line; posterior margin of head with deep median impression; eyes of moderate size, REL 0.15-0.19; pronotal humeri not prom- inent; mesonotum distinctly angular in profile; propodeal spines short, stout and directed posterodorsally; postpetiole with blunt, obtuse lateral angles (dorsal view); legs relatively long, HTI 0.64-0.81, LHT/ PrW 1.30-1.66. Anterolateral regions of head with rugoreticulate and reticulate-fo- veolate sculpture which variably invades the medial and posterior portions of head (a smooth, shiny area generally persists in the medial impression above the frontal tri- angle); posterior margin of head finely sculptured or smooth, lacking conspicuous rugoreticulum; mesosoma mostly foveolate and (sub)opaque, becoming sublucid on the side of the pronotum and on the pron- otal dorsum, where there are usually trans- verse carinulae. Body pilosity abundant, long, slender, fine-tipped; conspicuous on the tibiae (HTC 15-27). Color variable, from light orange-brown to dark reddish- brown, gaster often darker than the rest of body. Comments. — P. hyatti is found from Tex- as, Oklahoma and Colorado west to Ne- vada and California and south into north- ern Mexico (Kempf 1972; Smith 1979). As befits its wide distribution, the species oc- cupies a broad range of habitats including Chihuahuan, Sonoran and Great Basin de- serts, short-grass prairie, piny on-juni per woodland, oak woodland, riparian wood- land, chaparral, and coastal sage scrub (Droual 1983; Gregg 1963; Snelling and George 1979; Suarez et ai 1998; Wheeler and Wheeler 1973, 1986). With respect to foraging behavior, P. hyatti appears to be a generalist omnivore rather than a seed-har- vesting specialist (Wheeler 1908; Snelling and George 1979). Colonies of P. hyatti are Volume 9, Number 1, 2000 95 frequently subject to raids of the army ant, Neivumyrmex nigrescens (Droual and Topoff 1981; Mirenda et al. 1980; Ward 1999). Pheidole obtusospinosa Pergande 1896 (Figs. 11, 15) Pheidole obtusospinosa Pergande 1896. Syntype workers (majors), Tepic, Nayarit (Eisen & Vaslit) (LACM, USNM). Pheidole subdentata Pergande 1896. Syntype workers (minors), Tepic, Nayarit (Eisen & Vaslit) (LACM, USNM). Note: Wheeler (1914) synonymized P. obtusospinosa under P. subdentata, but Pergande's P. subdentata is preoccupied (Bolton 1995), so P. obtusospinosa is the first available replacement name. Pheidole vaslitii; Creighton (1958), misidentifi- cation and invalid "lectotype" designation. Diagnosis, major worker. — Large species, variable in size (HW 1.36-2.90, HL 1.39- 2.55, LHT 1.13-1.50; n = 28); in medium- sized workers (HW < 2.20) scape relative- ly long (SI 0.63-0.88, SL/PrW 1.32-1.66) and bent basally but usually not conspic- uously broadened (BSW/SL 0.07-0.10); in supermajors (HW > 2.20) scape relatively short (SI 0.44-0.48, SL/PrW 1.00-1.13) and basal portion notably broadened (BSW/SL ~ 0.14); ventral hypostomal teeth present, generally spine-like, much closer to the dorsolateral hypostomal teeth than to the midline; posterior margin of head with shallow, obtuse V-shaped impression, be- coming more deeply notched in super- majors (HW > 2.20); eyes relatively small, REL 0.11-0.16; pronotal humeri not strongly protuberant; mesonotum bluntly angular in profile; propodeal spines short, stout and directed posterodorsally; post- petiole with blunt, obtuse lateral angles (dorsal view), more prominent in super- majors; legs long, HTI 0.52-0.86, LHT/ PrW 1.19-1.62 (in all but supermajors HTI 0.69-0.86, LHT /PrW 1.45-1.62). Upper surface of mandibles smooth and shiny, except for weak basal striae. Dorsum of head largely covered with rugoreticulate and reticulate-foveolate sculpture, densest (and the longitudinal orientation of the ru- gulae least evident) in supermajors; me- sosoma mostly foveolate and (sub)opaque, becoming sublucid on the side of the pronotum and on the pronotal dorsum, where there are transverse carinulae. Body pilosity very abundant, shorter and more blunt-tipped than in P. hyatti, conspicuous on the tibiae (HTC 25-36). Color varying from orange-brown to dark reddish- brown, the gaster sometimes darker than the rest of the body. Comments. — P. obtusospinosa is known from Arizona, Sonora, Sinaloa, Nayarit and Jalisco. Creighton (1958) recorded the species, under the name P. subdentata, from "elevations up to 6300 feet in many of the mountains of southern Arizona" and noted that "while the number of sub- dentata nests in an area is often quite large, this species never seems to exclude other ants from such areas". In Arizona I have encountered P. obtusospinosa in oak-pine- juniper woodland and oak-juniper wood- land, at elevations ranging from 1670m to 2100m. Colonies were found nesting un- der stones, and workers (including sol- diers) foraged nocturnally on the ground, and came to tuna fish baits. Groups of mi- nor and major workers were also seen, without brood, under a rotting log and in a dead branch of Quercus grisea. There are recent records of P. obtusospinosa workers (mostly minors) visiting extrafloral nectar- ies of Ferocactus, Opuntia and Pachycereus at several sites near Punta Chueca, Sonora (leg. Kevin Walker). Pheidole sp. (crassicornis complex) Pheidole vaslitii Pergande 1896 (part); 3 minor workers, Sierra San Lazaro, Baja California Sur, Mexico (Eisen & Vaslit) (USNM); not conspecific with the above newly-designated lectotype of P. vaslitii. Pheidole crassicornis tetra; Creighton (1958). Comments. — The identity of the three minor workers is discussed above. Based on material from southern United States the major workers of this complex may be recognized by the following features (1) 96 Journal of Hymenoptera Research ventrolateral hypostomal teeth well devel- PrW < 1.16, (4) legs very short, HTI < oped, spine-like, displaced laterally about 0.62, LHT/PrW < 1.18, (5) standing pilos- 0.75 times the distance from the midline ity blunt-tipped, sparse on appendages, to the dorsolateral teeth, (2) base of scape HTC < 10, (6) posterior third of head flattened and broadened (BSW/SL largely smooth and shining, lacking retic- 0.14), (3) scapes very short, SI < 0.60, SL/ ulate-foveolate sculpture. KEY TO SPECIES ASSOCIATED WITH THE NAME PHEIDOLE VASUTII The following key is not intended to be comprehensive. It is concerned only with those Pheidole species that have been confused with Pheidole vaslitii, and is presented as a summary of the differences between them. Most of these species belong to the P. fallax group, as defined by E. O. Wilson in his forthcoming monograph on the New World Pheidole (Wilson, in prep.). Wilson recognizes a separate P. crassicornis group that seems likely to be nested phylogenetically within the P. fallax group, insofar as it possesses the basic features of the latter group but manifested as more derived states. The term "P. crassicornis complex", as used below, refers to three taxa in the P. crassicornis group (P. crassicornis, P. crassicornis tetra, and P. diversipilosa) whose relationships to one another need further clarification. Pheidole obtusospinosa and P. hirtula are very closely related. The distinctions cited in the key to majors are taken from Creighton (1958), and are admittedly slight. Future study may well show that these two taxa are simply geographical variants of a single species. I have been unable to find diagnostic differences between the minor workers of P. obtusospinosa and P. hirtula. Major workers 1 Legs relatively long, LHT 1.11-1.50, and eyes small (EL/LHT 0.16-0.21); major workers variable in size (HW 1.31-3.07), and "supermajors" (HW > 2.20) occur; body pilosity conspicuous, relatively short and dense (Figs. 11, 15), HTC 25^8 2 T Legs shorter, LHT < 1.10, and eyes relatively large (EL/LHT 0.22-0.29); major workers less variable in size (HW 1.05-1.57), supermajors lacking; body pilosity variable (Figs. 12- 14, 16-18). generally less abundant, HTC 0-27 3 2 In largest individuals (HW > 2.50) head strongly cordate, conspicuously narrowed to- wards the mandibular insertions (Creighton 1958, fig. 1); rugulate sculpture tending to be less developed on posterior half of head, which may be sublucid (northeastern Mexico, west to Chihuahua, Durango and Jalisco) hirtula Forel 2' In largest individuals (HW > 2.50), head less strongly cordate, less strikingly narrowed anteriorly (Creighton 1958, fig. 2); posterior half of head with rugulate and reticulate- foveolate sculpture tending to be more strongly developed (Arizona, Sonora, Sinaloa, Nay- arit, Jalisco) obtusospinosa Pergande 3 Scapes and legs very short, SL/PrW < 1.16, LHT/PrW < 1.18; standing pilosity sparse, HTC < 10 (southern United States, northern Mexico) crassicornis complex 3' Scapes and legs longer, SL/PrW 1.21-1.79, LHT/PrW 1.23-1.66; pilosity variable 4 4 Pilosity long, fine-tipped, and abundant (Figs. 13-14, 17-18), HTC 15-27; occipital lobes varying from smooth and shiny to reticulate-foveolate and opaque, but lacking conspic- uous rugoreticulum; scapes and legs longer, SL/PrW 1.28-1.79, LHT/PrW 1.30-1.66 (n 47) (see also Figs. 7-8), (southwestern United States, northern Mexico) hyatti Emery 4' Pilosity shorter, blunt-tipped and less dense (Figs. 12, 16), HTC 1-17; occipital lobes ru- goreticulate and subopaque; appendages shorter, on average; SL/PrW 1.21-1.33, LHT/ Volume 9, Number 1, 2000 97 PrW 1.23-1.36 (n = 13) (see also Figs. 7-8) (southwestern United States, northern Mexico) cockerelli Wheeler Minor workers 1 Legs long, LHT 0.85-0.95, and eyes small (EL/LHT 0.19-0.23); body pilosity abundant and moderately long, HTC 14-28 hirtula Forel and obtusospinosa Pergande V Legs shorter, LHT 0.52-0.85, and eyes relatively large (EL/LHT 0.22-0.29); body pilosity variable, often less abundant, HTC 0-24 3 2 Scapes and legs very short, SI 1.11-1.22, HTI 0.89-1.02 (n = 11); eyes smaller, REL2 0.23- 0.27; pilosity sparse and short, HTC 0-7 crassicornis complex 2' Scapes and legs longer, SI 1.24-1.71, HTI 1.07-1.38 (n = 41); eyes larger, REL2 0.27-0.34; pilosity longer and more abundant, HTC 7-24 3 3 Pilosity on body and appendages long, fine-tipped, and abundant, HTC 11-24; upper third of head smooth and shiny or partially invaded by foveolate sculpture and subopaque, but lacking rugulae; head more elongate (CI 0.77-0.85), scapes and legs longer, SL/PrW 1.95- 2.50, LHT/PrW 1.64-2.01 (n = 28) hyatti Emery 3' Pilosity less common, blunt-tipped, HTC 7-18; upper third of head largely opaque, covered with foveolate sculpture and overlain by weak rugoreticulum or longitudinal rugulation; head broader (CI 0.82-0.90) and appendages shorter, on average, SL/PrW 1.74-1.99, LHT/ PrW 1.49-1.67 (n = 13) cockerelli Wheeler CONCLUDING REMARKS The "Plieidole vaslitii problem" exempli- fies two difficulties that have often arisen in ant taxonomy: insufficient access to, or analysis of, type specimens on the one hand, and yet (paradoxically) a subtle ty- pological bias on the other hand, which has- sometimes led investigators to over- rate the significance of differences among populations of the same species. It is iron- ical that Creighton's (1958) dubious choice for the lectotype of P. vaslitii would not have occurred if he had recognized the major workers in the type series as vari- ants of the older-named Plieidole In/atti, in which the "typically" smooth shiny occip- ital lobes had become clouded by sculp- ture. To be fair, it could be argued that Creighton did not possess a sufficiently large and geographically extensive series of specimens of P. In/atti, especially from Baja California. From the vantage point of the more extensive data now available Plieidole In/atti appears to be a classic ex- ample of a polytypic species (Mayr 1982), i.e., one that consists of a series of geo- graphically dispersed and morphological- ly disparate populations, linked together by intermediate populations and showing evidence of recent or ongoing gene flow. At the same time, the possibility cannot be excluded that one or more cryptic species lurks within this complex of populations, especially in view of the broad range of habitats occupied. ACKNOWLEDGMENTS I thank the following curators for facilitating the study of museum specimens: Wojciech Pulawski (CASC), John Sorensen (CDAE), Roy Snelling (I. ACM), and Ted Schultz (USNM). Additional useful material from Baja California and adjacent regions was provided by Rolf Aalbu. April Boulton, led Case, Mark Trepanier, Andy Snare/, Kevin Walker and, especially, Robert fohnson. I am grateful to Sean Brady, Robert fohnson, |ohn I attke, Eric Grissell and anonymous reviewer tor comments on the manu- script, and to Ed Wilson for making available parts of his forthcoming revision of Pheidole. Support ol the University of California and the National Science Foundation is appreciatively acknowledged 98 Journal of Hymenoptera Research LITERATURE CITED Bolton, B. 1995. A new general catalogue of the ants of the world. Cambridge, Mass.: Harvard University Press, 504 pp. Creighton, W. S. 1950. The ants of North America. Bulletin of the Museum of Comparative Zoology 104: 1-585. Creighton, W. S. 1958 ("1957"). A revisionary study of Pheidole vasliti Pergande (Hymenoptera: For- micidae). Journal of the New York Entomological So- ciety 65: 203-212. Droual, R. 1983. The organization of nest evacuation in Pheidole desertorum Wheeler and P. hyatti Em- ery (Hymenoptera: Formicidae). Behavioral Ecol- ogy and Sociobiology 12: 203-208. Droual, R. and H. Topoff. 1981. The emigration be- havior of two species of the genus Pheidole (For- micidae: Myrmicinae). Psyche (Cambridge) 88: 135-150. Emery, C. 1895. Beitrage zur Kenntniss der norda- merikanischen Ameisenfauna. (Schluss). Zoolo- gische Jahrbiicher. Ahteilung fur Systematik, Geogra- phic und Biologic der Tierc 8: 257-360. Forel, A. 1899. Formicidae. [part]. Biologia Centrali- Americana. Hymenoptera 3: 57-80. Forel, A. 1901. I. Fourmis mexicaines recoltees par M. le professeur W.-M. Wheeler. II. A propos de la classification des fourmis. Annates de la Socictc Entomologiquc de Belgique 45: 123-141. Gregg, R. E. 1963. The ants of Colorado, with reference to their ecology, taxonomy, and geographic distribu- tion. Boulder: University of Colorado Press, xvi + 792 pp. Gregg, R. E. 1969. New species of Pheidole from Pa- cific Coast islands (Hymenoptera: Formicidae). Entomological News 80: 93-101. Kempf, W. W. 1972. Catalogo abreviado das formigas da regiao Neotropical. Studia Entomologica 15: 3- 344. Mayr, E. 1982. The growth of biological thought. Diver- sity, evolution and inheritance. Cambridge, Mass.: Harvard University Press, xii 4- 974 pp. Mayr, G. 1853. Einige neue Ameisen. Verhaudlungen der Zoologisch-Botanischen Vereins in Wien 2: 143- 150. Mirenda, J. T., D. G. Eakins, K. Gravelle and H. To- poff. 1C). Volume 9, Number 1, 2000 105 Figs. 1-6. Environment chamber scanning electron micrographs of Aneurobracon harleyi gen. et sp. now, male holotype: 1, habitus. 2, metapleuron. 3, scutellum to propodeum. 4, fore wing. 5, metasomal tergites 1-3. 6 lateral view of 5th metasomal tergite showing posterolateral emargination. Type species. — Aneuradesha harleyi Quicke Aneuradesha Harleyi Quicke sp. nov. sp. n. by monotypy and original designa- (Figs 1-6) tion. Remarks.— The complete lack of fore TWe material.— Male holotype: INDIA: wing vein CUlb and the reduced propo- Muzzafarnagar, 8.vi.l998, Atar Singh, ex deal carina are both probably autapomor- Asmangulk cuspidata, HE 23897 (BMNH). phies of Aneuradesha with respect to the Male paratype (BMNH): same data as ho- other genera of Adeshini. lotype. 106 Journal of Hymenoptera Research Antenna with 34 flagellomeres. Height of clypeus (excluding hypoclypeus): inter- tentorial distance:tentorio-ocular distance = 1.0:3.7:2.0. Height of eye: width of head across eyes: width of face = 1.0:2.6:1.4. Transverse diameter of posterior ocellus: distance between posterior ocelli:shortest distance between posterior ocellus and eye = 1.0:1.2:2.2. Mesosoma 1.2 times lon- ger than maximum height. Mesoscutum (Fig. 1) smooth except for punctures at ba- ses of setae; moderately densely setose ex- cept the mid-longitudinal part of the mid- dle lobe. Scutellar sulcus finely crenulate. Scutellum with a weak but distinct pit. Propodeum (Fig. 3) crenulate posteriorly. Posterior margin of hind wing weakly emarginate. Lengths of veins 2RS:3RSa:r- m = 1.3:1.28:1.0. Lengths of veins r:3RSa: 3RSb = 1:1.45:5.75. Lengths of veins (Rs + M)b:2M = 1.0:1.55. Fore wing vein lcu-a marginally postfurcal. Vein 2-1A tubular for approximately half length of 1st sub- discal cell. Base of hind wing evenly se- tose. Apex of vein C + SC + R with one especially thickened bristle (basal hamu- lus). Lengths of hind femur (excluding tro- chantellus): tibia:basitarsus = 2.0:3.3:1.0. First and 2nd metasomal tergites rugose with overlying granulose sculpture; more posterior tergites granulose superimposed on weak foveate sculpture. Posterior mar- gin of 5th tergite with a distinct, trans- verse subposterior groove (Fig. 6). Uniformly honey-yellow except for the flagellomeres, an ill-defined mark on the raised median area of the 1st metasomal tergite and a mark medio-basally on the 2nd metasomal tergite, which are blackish. Biologies of Adesha Albolineata and Aneuradesha Harleyi Aneuradesha harleyi is so far known only from the hispine chrysomelid beetle As- mangulia cuspidata Maulik, a leafminer pest of sugarcane and rice in India and In- donesia. Anwar (1944) records a "Micro- bracon" sp. as an ectoparasitoid of A. cus- pidata larvae causing up to 38% parasitisa- tion. Since, to our knowledge, no other braconid has ever been recorded from A. cuspidata, it is conceivable that "Microbra- con" represents a misidentification of the species treated here as A. harleyi. The authors recently had the opportu- nity to examine material of another para- sitoid of a hispine, Promecotheca cumingi Baly, a pest of coconut palm in south and southeast Asia (Gallego et al. 1983; Mathur et al. 1984). A series of 12 specimens, sent to the second author for identification, turned out to belong to the type species of the type genus of Adeshini, Adesha alboli- neata Cameron (see van Achterberg 1983, for a redescription of Adesha albolineata, and Quicke, 1986, for a key to species of Adesha). Previously, Adesha had been re- corded from Borneo and the Malay Pen- ninsula, and A. albolineata was known only from two specimens. According to the col- lector, A. albolineata is an ectoparasitoid of larvae of P. cumingi. There appears, there- fore, to be a very reasonable basis for con- sidering Adeshini as specialist ectoparasi- toids of hispine larvae. Adesha albolineata: Material Examined: 5 956* MALAYSIA: Sarawak, Buntal ex Promecotheca cumingi on Nipah (Cocos nu- cifera) 20.V.1998 (Drahman coll.); 1 9 same data except 17.ii.1998, HE 23809/ S132 (all specimens in The Natural History Muse- um, London). ACKNOWLEDGMENTS We thank Atar Singh, Sugarcane Research Station, Muzzafarnagar, Uttar Pradesh, India, and Megir Gumbek, Pusat Penyelidikan Pertanian, Sarawak, Malaysia, for providing specimens of A. harleyi and A. albolineata, respectively. Hasan Basibuyuk kindly made the SEMs. LITERATURE CITED Achterberg, C. van. 1983. Six new genera of Bracon- inae from the Afrotropical Region (Hymenop- tera, Braconidae). Tijdschrifl voor Entomologie 122: 175-202. Anwar, M. S. 1944. The natural control of Asmangulia cuspidata Maulik, the sugarcane leaf-miner, by parasites. Indian journal of Entomology 5: 248-249. Gallego, V. C, Baltazar, C. R., Cadapan, E. P. and Volume 9, Number 1, 2000 107 Abad, R. G. 1983. Some ecological studies on the coconut leafminer Promecotheca cumingi (Coleop- tera: Hispidae) and its hymenopterous parasit- oids in the Philippines. Philippine Entomologist. 6: 471-493. Mathur, P. N., Samathanam, G. J. and Singh, A. 1984. Coconut leaf miner beetle (Promecotheca cumingi Baly). Plant Protection Bulletin, India 36: 91-93. Quicke, D. L. J. 1986. A revision of the Adeshini van Achterberg with descriptions of three new gen- era from the Palaeotropics (Insecta, Hymenop- tera, Braconidae). Zoologica Scripta 15: 265-274. Quicke, D. L. j. 1987. The Old World genera of bra- conine wasps (Hymenoptera: Braconidae). Jour- nal of Naural History 21: 43-157. Quicke, D. L. J. 1988. A new genus and species of Adeshini (Hym., Braconidae, Braconinae) from Thailand. Entomologists' Monthly Magazine 124: 203-205. Quicke, D. L. J. and S. N. Ingram. 1993. Braconine wasps of Australia. Memoirs of the Queensland. Museum 31: 299-336. Wharton, R. A., P. M. Marsh and M. J. Sharkey (Eds). Manual of the New World genera of the family Bra- conidae (Hymenoptera). Special Publication of the International Society of Hymenopterists, No. 1. J. HYM. RES. Vol. 9(1), 2000, pp. 108-110 Microgastrine (Hymenoptera: Braconidae) Parasitoids of Colias lesbia (Fabricius) (Lepidoptera: Pieridae) Michael J. Sharkey, Kathie Finnell, Jason Leathers, and Jorge Frana (MJS, KF, JL) University of Kentucky, Department of Entomology, S-225 Ag. Science Center North, Lexington, KY 40546-0091, USA; (JF) INTA, Estacion Experimental Agropecuaria Rafaela, CC 22 2300-Rafaela, Santa Fe, Argentina Abstract. — The microgastrine (Braconidae) parasitoids of Colias lesbia (Fabricius), a lepidopteran pest of alfalfa in southern temperate regions of South America, are reviewed. Three species, Cotesia ayerza (Brethes), C. glomerata (Linnaeus) and C. lesbiae (Blanchard) are recognized and the mor- phological differences between them are presented. A lectotype for C. ayerza is designated. Colias lesbia (Fabricius) is a lepidopteran pest of alfalfa, and is widespread in the southern temperate regions of South America. Throughout its range, it is at- tacked by three microgastrine (Braconi- dae) parasitoids. The taxonomy and iden- tification of these parasitoids, which are all members of the genus Cotesia, have been problematic, with between two and five species being recognized in the com- plex. The purpose of this paper is to re- solve these problems and to present mor- phological characteristics to distinguish between the species of Cotesia that attack Colias lesbia. The subfamily Microgastrinae is cos- mopolitan and contains thousands of spe- cies (Shenefelt 1972). They are internal parasitoids of larval Lepidoptera and are important in the natural and manipulated control of many pest species. The genus Cotesia contains hundreds of species that have been formally described, but even more that have not yet been described. The genus is also cosmopolitan and may be distinguished from all other Microgas- trinae by the sculpture and shape of the first metasomal tergum and by the short ovipositor of females (cf. Mason 1981). Shenefelt (1972, 1980) lists three species of Microgastrinae as parasitoids of Colias lesbia, viz., Cotesia glomerata (Linnaeus), C. ayerza (Brethes), and C. lesbiae (Blanchard) (all as Apanteles). There are other micro- gastrine names recorded as parasitoids of Colias lesbia; however, owing to well-re- searched synonymies (cf. Shenefelt 1972), only these three remain current in the lit- erature. Syntypes of C. ayerza were borrowed from the Museo Argentino de Ciencias Naturales and specimens of C. lesbiae de- termined by both Blanchard and Brethes were borrowed from the Museo de La Pla- ta. The type of C. glomerata is no longer extant but the description by Nixon (1974) and reliably identified specimens from Eu- rope and North America (determined by G.E.J. Nixon and W.R.M. Mason respec- tively) were used for comparative purpos- es. C. glomerata may be distinguished from the other two species by the color of the hind femur which is yellow with a small melanic region in the apical 1 / 6. The hind femora of C. ayerza and C. lesbiae are uni- formly brown. C. ayerza and C. lesbiae may be distinguished from each other by the dimensions of the medial tergite of the first metasomal segment. That of C. ayerza is about as wide as long whereas that of C. lesbiae is more than 1.5 times longer It... Volume 9, Number 1, 2000 109 Table 1. Comparison of Cotesia species parasitizing Colias lesbia. Taxa Characters and st.iu a = C. ayzera g = C. glomerata 1 = C. lesbiae 1. Hind femur color. a) Entirely brown (melanic) b) Yellow except for brown in apical 1/6 2. Hypopygium of female. a) tapering to a sharp point apically b) with a deep semicircular emargination apically [c.f. Fig. 46 in Nixon (1974)] 3. First metsomal median tergite. a) more than 1.5 times longer than wide b) about as long as wide 4. Sculpture of posteromedial region of mesoscutum, directly anterad mesoscutellar sulcus. a) with longitudinal rugosities (Fig. 1) b) lacking longitudinal rugosities (Fig. 2) a, 1 g a, 1 g gJ a than wide. Other morphological differenc- es between the three species are summa- rized in Table 1. There may be a behavioral difference between C. lesbiae and the other two spe- cies. C. ayerza and C. glomerata, like most species of Cotesia, are gregarious (Mason 1981, Nixon 1974) with many eggs laid in each host, whereas C. lesbiae appears to be a solitary parasitoid (Hamity 1978). Brethes did not designate a holotype for Apanteles ayerza and we take this oppor- tunity to designate a lectotype and two paralectotypes. All three specimens are on Figs. 1-2. Dorsal aspects of mesosomata. 1 (left), Cotesia glomerata. The arrow on the mesoscutum indicates longitudinal rugosities; 2 (right), Cotesia lesbiae. no Journal of Hymenoptera Research the same pin, mounted on one, small, quadrate, piece of paper. The lectotype is a female and is the specimen in the middle of the piece of paper flanked by the two paralectotypes which both appear to be males, but owing to the poor condition of the specimens, sex is difficult to deter- mine. The lectotype is missing both left wings and the right hind wing but it is intact otherwise. All specimens appear to be conspecific. There are four original, hand-written labels on the pin. The upper- most reads "La Pampa 111.1920 J. Wil- liamson". The second reads, "parasite de Colias lesbia". The third label is small and has a few marks that are not decipherable; and the fourth reads "Apanteles Ayerza Brethes". We have also added a red label with the following: "Lectotype Apanteles ayerza Brethes designated by Sharkey, Fi- nell, and Leathers". ACKNOWLEDGMENTS We thank the UK Graduate School and the KYSS (Kentucky Young Scientists Summer) program for giving the second author the opportunity to work with the senior author. Thanks to Pete Southgate for photomicrographs, and to Eric Grissell, Jim Whitfield, Dan Potter, and an anonymous person for reviewing the manuscript. This research was facilitated by the Kentucky Experiment Station Project Number 99-08- 20. LITERATURE CITED Hamity, M. G. A. de. 1978. Aspectos biologicos de Apanteles lesbiae Blanchard (Hymenoptera: Bra- conidae) parasito de la isoca de la alfalfa Colias lesbia (F) (Lepidoptera: Pieridae). IDIA NE 337- 342. Jan-Jul. Pp: 39-45. Mason, W. R. M. 1981. The polyphyletic nature of Apanteles Foerster (Hymenoptera: Braconidae): a phylogeny and reclassification of Microgastrinae. Memoirs of the Entomological Society of Canada 115: 1-147. Nixon, G. E. J. 1974. A revision of the North-western European species of the glomeratus-group of Apanteles Forester (Hymenoptera: Braconidae). Bulletin of Entomoloical Research 64: 453-524. Shenefelt, R. D. 1972. Braconidae 4. Microgastrinae, Apanteles. In: Vecht, J. van des and Shenefelt, R. D. (Eds). Hymenopterorum Catalogus (novo editio). Pars 7. Pp. 429-668. Dr. W. Junk, The Hague. Shenefelt, R. D. 1980. Braconidae 11, Introduction, Guide to host names, Index to braconid names. In: Achterberg, C. van, and Shenefelt, R. D. (Eds) Hymenopterorum Catalogus (novo editio). Pars 16. Pp. 1-384. Dr. W. Junk, The Hague. J. HYM. RES. Vol. 9(1), 2000, pp. 111-141 Seven New Genera of the Subfamily Doryctinae (Hymenoptera: Braconidae) from the Old World Sergei A. Belokobylskij and Donald L. J. Quicke (SAB) Zoological Institute, Russian Academy, of Sciences, Universitetskaya nab. 1, St. Petersburg 199034, Russia; (DLJQ) Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berks SL5 7PY, U.K. and Department of Entomology, The Natural History Museum, London SW7 5BD, U.K. Abstract. — Seven new genera of the subfamily Doryctinae from the tropical and subtropical regions of the Old World are described and illustrated: Cryptodoryctes gen. nov. (type species Cr. hirneri sp. nov. from South Africa), Chelonodoryctes gen. nov. (type species Ch. inopinatus sp. nov. from Brunei), Synspilus gen. nov. (type species S. nitidus sp. nov. from Sarawak), Bra- codoryctes gen. nov. (type species B. tergalis sp. nov. from New Guinea), Afrospathius gen. nov. (type species Af. dispar sp. nov. from Africa), Hemispathius gen. nov. (type species H. polysten- oides sp. nov. from Uganda), and Atitidoryctes gen. nov. (type species An. pronotalis sp. nov. from Australia). Three further species of Bracodoryctes gen. nov. are also described: B. longitarsus sp. nov., B. curvinervis sp. nov., B. nigriceps sp. nov. The affinities of the new genera are dis- cussed. The Doryctinae are one of the most in- teresting and diversified subfamilies of Braconidae. More than one hundred and twenty valid genera have been described, which, for the most part, are restricted to tropical and subtropical regions (Shenefelt & Marsh 1976; Belokobylskij 1992; Marsh 1993). The generic richness of the Doryc- tinae is highest in the Neotropical region but this region has also been the most thoroughly investigated, especially by Paul Marsh, who has described many new genera from there (e.g. Marsh 1993), and has recently provided a key to the genera occurring in the New World (Marsh 1997). In comparison, the generic composition of the Old World fauna is really incomplete but, even so, it would appear that the Old World tropics are home to fewer genera than the New World. This pattern of ge- neric level diversity is, interestingly, the opposite of that found in the closely-relat- ed and apparently biologically similar Braconinae which have far more genera in the Old World tropics (Quicke 1987, Quicke 1997). Phylogenetic relationships between the genera of Doryctinae have until recently been based largely on the relatively im- poverished Nearctic and Palaearctic fau- nas (Fischer 1981, Belokobylskij 1992). Further, the characters used have princi- pally concerned external adult morpholo- gy such as the presence and absence of various carinae, sutures, and wing veins, and the shape of the metasoma. The dis- covery of many new internal characters involving male genitalia (Belokobylskij 1987), the venom apparatus (Quicke ct al. 1992a) and most recently, the internal sculpture of the ovipositor egg canal (Rah- man ct al. 1998), has greatly increased the possibility of obtaining a meaningful phy- logenetic tree. Seven new genera from the Old World tropics and subtropics are de- scribed here to make their names available for future publications on the phytogeny of the subfamily (Belokobylskij, Marsh & Quicke in preparation). Of the new taxa described below, Chc- lonodoryctes gen. nov., whose affinities are 112 Journal of Hymenoptera Research uncertain, is the first doryctine genus in which the 2nd and 3rd metasomal tergites are greatly enlarged, covering the follow- ing segments of the metasoma in the same way as is known for various genera of Ro- gadinae and Lysiterminae. Loss of both the occipital and prepectal carinae is one of the main characteristics of the subfam- ily Braconinae and is a character used in many older keys for separating the Bra- coninae from the Doryctinae. The new ge- nus Bracodoryctes gen. nov., of which four new species are described below, is the second known doryctine genus without these carinae, the first being Sirngra Cam- eron which was correspondingly separat- ed from other doryctines in the tribe Sir- agrini (Belokobylskij 1994). Transforma- tions of the scape are known in several do- ryctine genera (e.g. Syngaster Brulle, Pseudodoryctes Szepligeti, Siragra Camer- on, Binarea Brulle, Pseudorhoptrocentrus Granger, Jarra Marsh & Austin) as they are among the Braconinae (Quicke 1987). However, the new genus Synspilus gen. nov. is the first doryctine in which the scape not only has an apical lobe demar- cated by a preapical, transverse carina, but also has a strong basal constriction as in the Atanycolus Forster group of genera in the Braconinae (Quicke 1987) which are all parasitoids of bark-boring or subcortical beetles in dead standing or fallen wood. In addition, this genus is included in a group with an apically open 1st subdiscal cell and a strongly reduced vein 2RS (tribe Heterospilini: Belokobylskij 1992). Crypto- doryctes gen. nov. is an African genus that appears to be related to Priosphys Ender- lein and Odontodoryctes Granger, with which it shares the following putative syn- apomorphies: absense of the basoventral tubercle of the hind coxa and the short subbasal cell of the hind wing. It lacks an occipital carina but has a distinct pronope on the anterodorsal part of the neck. Two new genera of the tribe Spathiini are de- scribed in this paper. Afrospathius gen. nov. is especially interesting because it is the first genus of the subtribe Psenobolina to be found in the Old World. The male of the type species, A. dispar sp. nov., is characterised by the loss of vein r-m of the fore wing (present in the female) and by the presence of a large stigma-like enlarge- ment in the hind wing. Hemispathius gen. nov. is related to Spathiomorpha Tobias from the Palaearctic and Oriental Regions, and also to the Neotropical genus Notios- pathius Matthews & Marsh, making it par- ticularly interesting biogeographically. TERMINOLOGY AND COLLECTIONS Terminology follows that of Wharton et al. (1997). Because this is inconsistent with the useage by the senior author in many papers on the Doryctinae, a table is pro- vided giving the equivalent new terms as defined by Tobias (1976) (Table 1). The fol- lowing abbreviations are used: POL — pos- tocellar line, OOL — ocular-ocellar line, Od — maximum diameter of lateral ocel- lus. Specimens are held in The Natural History Museum, London, England (BMNH), the Bishop Museum, Honolulu (BPBM) and the Zoological Institute, Rus- sian Academy of Sciences, St. Petersburg, Russia (ZIP). DESCRIPTIONS OF NEW TAXA Cryptodoryctes gen. nov. Diagnosis. — This new genus is related to two other African genera, Priosphys En- derlein and to Odontodoryctes Granger. Cryptodoryctes differs from the other two by the hind coxa being without dorsal teeth, lack of an occipital carina, the 2nd tergite having lateral depressions, and the presence of a pronope. Cryptodoryctes gen. nov. differs from the Neotropical genus Megaloproctus Schulz in that the subbasal cell of the hind wing is very short, the oc- cipital carina lost, a pronope is present, and the 2nd metasomal suture has strong lateral breaks. Description. — Head: subcubical (Fig. 2), 1.4-1.5 times wider than long medially. Volume 9, Number 1, 2000 113 Table 1. Correspondence between wing venation terms employed here (see also Wharton et al. 1997) and those used in many papers on Doryctinae by Belokobylskij following Tobias (1976). Wharton ef al. (1997) and present paper lohi.is T * O ¥ f04- c 0 ° ositor Le o 0 o ° o o O n o 4 c m £ 0.3 (A O Q. > o 0.2 .■<*,'.*. ;>**•--• ■j 0.40 0.08 0.18 _0.16 t0. 14 £0.12 i 0.10 ^ 0.10 0.12 014 016 Clava Length (CLV) 0 18 -d 0.08 006 '? ♦ '»' . °4K 0.1 -1— — i— — r— — i— — r— — <— — r— —I 0.04 0 06 0.08 0.10 0.12 0.14 0.16 018 0.20 0.22 0.06 0.08 0 10 0.12 0.14 0 16 0.18 0 20 0.22 Funicle Length (FUN) Funicle Length (FUN) 0.6 U. IO 0.16 W _i • » < ' V > to. 14 ■ s .c » T a> <5 0.12 - ra TO g 0.10 - TO > -o 0.08 - 1c 0.06 - n 04 1 E 01 0.2 0.3 0.4 Ovipositor Length (OVL) 0.5 0.6 0.10 0.15 0.20 0.25 Midtibial Length (MTL) 030 035 Fig. 2. Two-variable scatterplots tor the five species of Encarsia. All measurements in mm. cala and £. strenua are clearly separated from each other on all of the bivariate plots (Fig. 2), although for ratio compari- sons, they are distinct, but overlapping, only for FWL/FWW (Table 4). In all cases, the five specimens of £. strenua reared from Parabemisia in California were small- est in size for most characteristics, usually forming a distinct cluster isolated from the majority of the southeast asian E. strenua (Fig. 2). Bivariate plots of the variables which best discriminate the species of Encarsia (Fig. 2), in general, demonstrate a strong positive correlation based on size. Fore wing length and width (Fig. 2A) have the most direct linear association. Size varia- tion of the host whiteflies does appear to have an effect on parasitoid size: Bemisia consistently yielded the smallest parasit- oids (£. bimaculata and £. protransvena), Parabemisia yielded moderately-sized wasps Volume 9, Number 1, 2000 149 (E. protransvena and £. stroma), and Or- chamoplatus yielded intermediate sized wasps (neocala). Dialeurodes, however, yielded almost the entire range of sizes found in £. citri, E. protransvena and £. strenua. Within Dialeurodes, D. citri and D. kirkaldii yielded the complete range of size variants for both £. protransvena and £. strenua; E. protransvena reared from D. ci- trifolii had a similar range of size variants, but only within the bounds of that species. Principal Components Analysis. — Individ- ual specimens of the five species are pro- jected on the first two principal compo- nents, which accounted for 92.6% of the overall variance (Fig. 3). The first principal component (PCI) accounts for 84.4% of the variance. Even with log-transformed data, size has an obvious impact on the distri- bution of points along PCI with the small- est species, £. bimaculata, having strong negative values along PCI and the larger species, E. citri and £. strenua, having strong positive values along PCI. Eigen- values and weights for the first two com- ponents are presented in Table 5. Margin- al fringe length (MFL) had almost no con- tribution to PCI. This absence of any cor- relation was reflected in a bivariate plot of MFL against FWL, which showed almost no correlation within species; a rather sur- prising result considering the importance of the relative length of the MFL in Encar- sia taxonomy. Along PCI, the other vari- ables had approximately equal contribu- tions, although measurements of the ovi- positor (OVL, TVL) had the largest influ- ence. Along PCII, the marginal fringe (MFL), ovipositor (OVL, TVL) and anten- na (FUN, F1L, F2L) had the greatest influ- ence. Five distinct clusters, representing the five species, occur within the first two principal components of the log-trans- formed data (Fig. 3). Encarsia protransvena had the greatest scatter of points and over- laps minimally with £. neocala, £. bimacu- lata and £. citri, but not at all with £. stren- ua. Considering that there are no a priori assumptions of group membership, we re- gard the clustering of points as a strong indicator of their group membership. Each overlapping group was clearly distin- guished on examination of the third prin- cipal component. A principal components analysis of only £. protransvena, using either host or local- ity as a priori groups, failed to separate out any meaningful clusters along the first three principal components. Canonical Variates Analysis. — Individual specimens are projected along the first two canonical variates (CVI & CVII) of the log- transformed data, which account for 89.3% of the original variance (Fig. 4). The five species are clearly discriminated, however no species can be completely separated along the first canonical variate and only E. neocala can be clearly separat- ed along the second. Although £. citri ap- pears to overlap with E. bimaculata and E. protransvena, it is very clearly discriminat- ed on a projection of the second and third canonical variate. The points for E. citri are in a plane clearly behind (class mean of -4.32 along CVIII; Table 6) those of E. bi- maculata (x = 0.88) and £. protransvena (x = —0.37). The single specimen of £. bi- maculata that appears to overlap the £. citri cluster along CVI and CVII is part of a long series reared from Bemisia in India. This specimen has a longer ovipositor (0.233 mm) than other E. bimaculata but, although it is an outlier on all of the bi- variate plots, it is not distinct from the main cluster and is never included in the E. citri group in the univariate compari- sons (Fig. 2). Also, the score for the spec- imen along CVIII (1.34) is about average from other £. bimaculata and very distinct from £. citri (-4.98 to -3.52). Encarsia pro- transvena and £. citri, which are difficult to separate on univariate characteristics, very marginally overlap on CVI & CVII; the clusters for each are distinct along CVII & CVIII, but with an overlap of scores along each variate. Notably, £. strenua and E. neocala are clearly separated on both CVI 150 Journal of Hymenoptera Research CM 00 0.3 0.2 = 0.1 0.0 c 0) c o CL E o Q- o f -0.2 ■0.3 • E. bimaculata o E. protransvena ▼ E. strenua V E. neocala -0.8. -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 Principal Component I (84.4 %) Fig. 3. Plot of the first two principal components from the principal components analysis of the log-trans- formed data set. The first principal component contains 84.4% of the sample variance, the second principal component contains 8.2%. Artificial boundaries define the limits of a priori groups. & CVII (Fig. 4) and CVII & CVIII, with the majority of separation in both cases occur- ring along CVII (Table 6). A reclassifica- tion of individuals using the discriminant Table 5. Eigenvalues and weights for the first two principal components, computed from the covariance matrix of the log-transformed data. Variable PCI PCI I Eigenv£ ilue 0.073 0.007 Proportion of Variance 0.844 0.082 FWI. 0.28 0.14 FWW 0.32 0.09 MFL 0.06 0.46 OVL 0.45 0.34 TVL 0.44 -0.55 MTL 0.32 0.11 CLV 0.14 0.16 FUN 0.30 0.29 F1L 0.33 0.30 F2L 0.31 0.37 functions resulted in 100% allocation to the correct a priori grouping. Stepwise dis- criminant analysis failed to identify any variables that could be excluded and still provide accurate classification of all spec- imens. The standardized and raw coefficients for the log-transformed data are presented in Table 6. The standardized coefficients represent the amount that the canonical variate will change for each change in the original variable by one standard devia- tion (Woolley & Browning 1987). Larger coefficients are generally better characters for discriminating points along that partic- ular analysis. Characteristics of the ovi- positor (OVL) and fore wing (FWL, FWW) had the strongest contribution along the first variate (CVI). These were also ranked as the best variables for discrimination in Volume 9, Number 1, 2000 151 CD c\i C\J CD -4-> CO 1 CO o 'c o c CO o 10 8 6 4 2 0 -2 -4 -6 -8 -8 o ▼ V E. bimaculata E. protransvena E. strenua E. neocala E. citri -6 -4 -2 0 8 10 Canonical Variate I (66.4 %) Fig. 4. Plot of the first two canonical variates from the canonical variates analysis of the log-transformed data set. The first canonical variate contains 66.4% of the sample variance, the second canonical variate con- tains 22.9%. Artificial boundaries define the limits of (7 priori groups. the stepwise analysis. Of the two most substantial features, £. strenua and E. neo- cala have a proportionally longer oviposi- tor, as already noticed, and £. bimaculata has a proportionally shorter fore wing. Length of the metatibia (MTL) and, again, the fore wing (FWL, FWW) were strong contributors to the second canonical vari- ate. A canonical variates analysis of only £. protransvena, using either host or locality as reference criteria, failed to separate out any meaningful clusters in different ordi- nations of the first three canonical vari- ates. The only segregation using host as a class criterion was along the first canonical variate, with individuals reared from Be- misia mostly negative (canonical scores be- low 0) and individuals from all of the oth- er whitefly species mostly positive (scores greater than -1). Bemisia are smaller than the other whiteflies, and this likely repre- sents segregation by size. Using geograph- ical locality as the class criterion, there was no meaningful segregation of clusters by locality along any axes; the only seg- regation of clusters was again host related, corresponding to species reared from Be- misia versus other whiteflies along the first canonical variate. The error rate for reclas- sifying the observations was 23.9% for all observations and 56.4% for specimens from Florida (n = 39). DISCUSSION Our initial grouping of species is sup- ported by the distinct clustering of groups found using either principal components or canonical variates analysis. Most of the specimens were segregated along the first 152 Journal of Hymenoptera Research Table 6. Standardized and raw coefficients and class means for the canonical variates analysis. The rows have been sorted by the elements of the vector of standardized coefficients for the first canonical variable. Standardized coefficients are the amount that the canonical structure will change for a change in the original variable of one standard deviation. Standardized coefficients Raw coefficients Variable cvi CV2 CV3 cvi CV2 CV3 OVL 5.91 -1.88 0.36 46.81 -14.85 2.86 FWW 2.88 6.02 -1.53 31.64 66.04 -16.83 TVL 1.11 1.85 1.08 8.55 14.16 8.26 F1L 0.74 0.10 -0.62 0.77 1.09 -6.42 MFL -0.01 -0.18 0.12 -0.21 -3.24 2.15 F2L -0.28 0.10 -1.84 -3.03 1.12 -19.99 CLV -0.57 0.68 1.19 -11.18 13.46 23.54 FUN -0.75 -0.80 1.52 -8.79 -9.44 17.86 MTL -1.40 -3.75 3.06 -15.59 -41.62 33.98 FWL -3.31 -2.65 -3.42 -42.56 -34.06 -44.04 Species Class means bimacult ta -5.08 1.49 0.88 protrtuis vena 0.45 -2.43 -0.37 citri -0.57 1.20 -4.32 strenua 7.98 0.02 1.79 neocala 5.84 6.39 -0.96 two axes in both analyses, and any area of overlap was resolved along the third axis. The fact that all specimens could be cor- rectly reclassified to their respective a priori groups is another indication of their distinctness. The complexity of the sepa- ration is, however, reflected by the inabil- ity to remove any of the variables to achieve complete reclassification. Encarsia bimaculata is the most distinc- tive species in both color and morphology, and none of the analyses would support these as mere color variants of £. protrans- vena. Both £. bimaculata and £. protransvena are parasites of Bemisia tabaci/argentifolii in the same geographical region, and yet both maintain their morphometric integ- rity. Encarsia citri can be separated from £. protransvena by the structure of the third valvula and setae of the midlobe. These characters are subtle and may not be trust- worthy. However, £. citri formed a mor- phometric grouping distinct from £. pro- transvena in all analyses. Again, both spe- cies use D. citri as a host but, even when using the same host, £. citri were consis- tently larger in size for almost all features except ovipositor length. The separation of £. citri from £. protransvena in the mor- phometric analyses was along all three axes, suggesting that both size and shape contributed to the difference. Encarsia protransvena were most variable for both size and host range. Hosts, which vary in size, have an obvious impact on size of the parasitoids. Specimens of £. protransvena reared from Bemisia were gen- erally smaller for most features than those reared from other hosts. In the canonical variates analysis by host, the specimens reared from Bemisia were separated along the first axis, with highest weights applied to the length and width of the fore wing (shorter and broader in specimens reared from Bemisia), although this was barely noticeable in the bivariate plots and is here considered as inconsequential. The full range of size, from smallest to largest were found in specimens reared from Di- aleurodes, and among these, D. citri yielded the largest size range. Whether the size range from D. citri is a result of parasitoids Volume 9, Number 1, 2000 153 emerging from different sized host or dif- ferent instars is unknown. Encarsia bimaculata, E. citri, E. neocala and E. strenua all have a southeast Asian origin. Encarsia bimaculata was purpose- fully introduced into Florida from India (Nguyen & Bennett 1995). Encarsia strenua was probably introduced into California from original material collected in India. Although material of £. strenua from India was reared in quarantine in Riverside, CA, in 1969, there is no record of a purposeful release of this species. The californian £. strenua was collected only between 1980 and 1982 at Tustin and the Irvine Ranch, both in Orange County, California. The or- igin of £. protransvena in the New World is more puzzling, especially as only two of the 40 species that we place in the stren- ua group are apparently endemic in the New World. Encarsia protransvena has been recorded primarily from the gulf coast states (Florida through Texas) and Puerto Rico, and from single rearings in Grand Cayman, Colombia (D. citrifolii), California (D. citri), and Honduras (B. ta- baci) (Polaszek et al. 1992). The records from Central and South America and Grand Cayman all appear to be valid rear- ings from field collected material. The sin- gle record from California is probably from release efforts being undertaken at that time (Bellows, pers. comm.). For all of the bivariate and multivariate analyses, these odd rearings of £. protransvena clus- tered within the main group of specimens and are indistinguishable by any set of measurements from other £. protransvena. Bemisia and both species of Dialeurodcs have been known from Florida since the early 1900's (Mound 1978; Nguyen et al. 1993), and yet no members of the strenua group were reared from whiteflies in any of these countries prior to 1984 (collection date of type material), even though there are several earlier collections of other spe- cies of Encarsia from whiteflies in Florida (Nguyen et al. 1993) and despite a re- newed research program focusing on the parasitoids of Dialeurodes in the late 1970's (Nguyen and Sailer 1979; Sailer et al. 1984). A collection of three specimens reared from Parabemisia in Valencia, Spain (in- cluded in the morphemetric analyses), three specimens from China and Taiwan reared from Dialeurodes and Aleurotrache- lus (Huang & Polaszek 1998), and one fe- male from Egypt reared from Dialeurodes (Polaszek et al. 1999) are the only collec- tions of £. protransvena from the Old World. Although from a unique host, the two specimens from Spain were certainly not unique morphologically and were about average (clustering centrally) for all bivariate and multivariate comparisons. The Egyptian specimen agrees in all char- acters with £. protransvena. The specimens from China and Taiwan were not part of this analysis, but exhibit diagnostic differ- ences (see comments after description of £. protransvena) from the New World and Spain material that might suggest they could be a different species close to £. pro- transvena or £. citri. If £. protransvena did have an eastern Palearctic or southeast Asian origin, we expect that it would be more widespread as it is in the New World. However, the distribution and suc- cess of £. protransvena cannot be used to demonstrate its origin, and the evidence remains equivocal over the origin of £. protransvena in the southeastern United States as to whether it was accidentally in- troduced from the Old World (western Pa- learctic or southeast Asia) and became abundant in its new habitat, or if it spread northward from an origin in South Amer- ica or the Carribean. It might be interest- ing to postulate that the appearance of £. protransvena coincides with the apparent replacement of Bemisia tabaci with B. argen- tifolii, but the parasitoid also was found to be common on other genera of whiteflies in the same geographical range. A discriminant function may be the best means of separating species such as £. pro- transvena and £. citri. However, in all cas- es, a majority of specimens can be sue- 154 Journal of Hymenoptera Research cessfully keyed using relatively straight- all of the specimens. Although still creat- forward measurements having a minimal ing complex couplets, we feel that this amount of overlap, and any specimens would be the best means to painlessly sep- within the zone of overlap can be segre- arate species and is the method used in gated using other characters which may the accompanying identification key. apply only to these specimens but not to KEY TO FEMALES OF COMMONLY ENCOUNTERED SPECIES OF THE STRENUA GROUP This identification key refers only to the species dealt with in this paper, a small representation of the 40 species that we have recently examined. It will, however, work for the species most commonly encountered in economically important crops. Members of the strenua group are rec- ognized by having a combination of 2-3 marginal setae along the dorsal margin of the costal cell at the apex, a bare area just above the stigmal vein, and scutellar sensillae closely placed or touching. 1. Ovipositor almost as long as gaster, more than 1.56 times as long as middle tibia, if between 1.5 and 1.65 times then fore wing less than 2.6 times as long as broad. Ovipositor robust, tip often bent at an angle in slide-mounted specimens 2 - Ovipositor clearly shorter than length of gaster, less than 1.5 times as long as middle tibia, some E. protransvena 1.5-1.6 times [deformed holotype 1.74 times], but then fore wing more than 2.67 times as long as broad [2.9 times in holotype of E. protransvena]. Ovipositor slender, tip always straight 3 2. Third valvula entirely yellow. Fore wing less than 2.44 times as long as broad and ovi- positor less than 2.6 times as long as clava E. neocala Heraty and Polaszek, n. sp. - Third valvula dark brown at the extreme tip, otherwise yellow (Fig. 16). Fore wing more than 2.44 times as long as broad and ovipositor usually more than 2.7 times as long as clava, if shorter then fore wing more than 2.5 times as long as broad . . . £. strenua (Silvestri) 3. Metasomal tergite 7 (tergite with spiracles) with 4 setae, only 2 long setae medial to cerci (Fig. 29); ocellar triangle irregularly aciculate E. sophia (Girault) (= transvena) - Mt7 with 6 setae, 4 long setae medial to cerci (Fig. 27); ocellar triangle usually reticulate, if aciculate then with a different pattern [not all species included in this keyl 4 4. Body with extensive dark pigmentation (at least a large part of the mesoscutum, or two or more gastral tergites, dark) (Fig 5). Ovipositor less than 1.8 times as long as clava E. bimaculata Heraty and Polaszek n. sp. - Body, including antennae, almost entirely yellow. Ovipositor more than 2.0 times as long as clava 5 5. Basal seta of third valvula long, exceeding base of subapical seta; subapical seta located half way between basal seta of third valvula and apex (Fig. 14). Midlobe of mesosoma usually with 5 pairs of setae, preapical pair overlapping or exceeding base of apical pair. Fore wing usually less than 2.6 times as long as broad (83.3% of specimens examined), if between 2.6 and 2.7 times, then clava more than 0.14 mm in length E. citri Ishii Basal seta of third valvula not reaching base of subapical seta; subapical seta located be- yond half way (0.65) between basal seta of third valvula and apex (Fig. 15). Midlobe of mesosoma usually with 4 pairs of setae, preapical pair not reaching base of apical pair. Fore wing usually more than 2.7 times as long as broad (92.8% of specimens examined), if between 2.6 and 2.7 times, then clava less than 0.14 mm in dorsal length E. protransvena Viggiani Volume 9, Number 1, 2000 155 ^^T^- Figs. 5-12. Encarsia bimaculata: 5-11, female; 5, body; 6, mid tibia; 7, antenna; 8, mid tarsus; 9, mouthparts; 10, base of fore wing; 1 1, tore wing; 12, antenna of male. Arrow points to apical two costal cell setae diagnostic of the strenua group. Encarsia bimaculata Heraty and Polaszek, new species (Figs. 5-12) Female. — Antenna with 6 flagellomeres, clava 3-segmented; Fl 2.0-2.2 X as long as broad and as long as F3; antenna yellow basally, slightly darkened apically. Head confused transverse colliculate (as in fin- gerprint), ocellar triangle similar but sculpture somewhat areolate; yellow ex- cept for a pale transverse band of brown across back of head; dorsal setae slight. Maxillary palpus 1 -segmented. Mandibles 3/3 dentate, marginal teeth acute. Meso- soma mostly yellow except following which are brown: pronotum, midlobe of mesoscutum anteriorly and medially, te- gula, axilla and propodeum submediallv. Mesosoma with light hexagonally areolate sculpture dorsally; midlobe with 4 pairs of 156 Journal of Hymenoptera Research Figs. 13-16. 13-14, Encarsia citri: 13, antenna of male; 14, third valvulae. 15, Encarsia protransvena, third valvulae; 16, Encarsia strenua, third valvulae. setae, setae delicate and about equal in size, side lobe with 3 pairs, axilla with 2 pairs (lateral pair minute) and scutellum with 2 pairs. Scutellar sensillae ovoid and separated by less than their own maxi- mum diameter; median groove usually narrow and distinct. Apical spur of mid tibia 0.6-0.8 X as long as basitarsus; basi- tarsomeres of midtarsus without pegs. Tarsal formula 5-5-5. Fore wing 2.58- 2.99 X as long as broad, marginal fringe 0.25-0.36 X width of fore wing; disc uni- formily setose; costal cell with row of 9- 10 minute setae and with 1 long marginal setae apically; submarginal vein with 2 large setae, basal area with 5-6 setae pos- terior to submarginal vein; frenal fold with several prominent thornlike spines; wing mostly hyaline, weakly infuscate at base of submarginal vein and base of fre- nal fold. Metasoma mostly yellow, tergites I + II mostly brown, sometimes with faint medial spot on tergites V + VI; laterally with weak cellulate reticulation; dorsal se- tal formula from tergite III: 2-2-2-6-6-6. Ovipositor 1. 09-1. 38 X as long as mid tib- ia, 1.33-1. 78 X as long as clava, 0.6 X as long as gaster (base to tip of third valvu- la); third valvula stout, 1. 6-2.2 X as long as broad, slightly extruded beyond epy- gium, entirely yellow. Male. — Coloration similar to female but darker with entire metasoma brown, head with a transverse medial band of brown infuscation, and wings may be weakly in- fuscate in the basal half. Antenna with 6 flagellomeres, apical two flagellomeres fused, with segments distinctly separated (Fig. 12). Comments. — This species is similar to E. dialeurodis Hayat for all features except for the darker coloration of the axillae, pron- otum and mesoscutum (medially) (versus all yellow in E. dialeurodis) and Fl about equal in length to F2 (versus distinctly subequal). It is readily distinguished from Volume 9, Number 1, 2000 157 other species of the stremia group by the dark mesosomal coloration of females and the shorter ovipositor to clava ratio. The only noticable difference in female color- ation was a more extensive dark pigmen- tation of the metasoma in the specimens from Hong Kong, with gastral tergites V & VII almost completely dark; otherwise these did not differ from other specimens. Two smashed specimens from Thailand, one reared from Bemisia t abaci but with no other data, and the other from Bemisia ta- baci on Gossypium hirsutum (Ban Ton Tea, 20.U992, F." Bennett 1224 [= Encarsia sp. in Nguyen & Bennett 1995]; both 9 USNM), are identical for all characters of this species, except the ocellar triangle is transversely aciculate, the mandibles are less acutely toothed and the mid tibial spur is only 0.44X as long as the basitar- sus. It is of interest that, assuming a ran- dom sampling, the original shipment of material of E. bimaculata from India (#1248) was only slightly biased for fe- males (6:4), whereas, except for the Hon- duras lab culture (1:17), only two males have been discovered in all of the subse- quent rearings. Another undescribed spe- cies in Florida (Gainesville; USNM) is very close, but it has a proportionally longer ovipositor (>2.0X as long as the clava), the female gaster is entirely dark, and the legs are also brown. This other species has also been reared from Bemisia and Trialeurodes abutilonea. Encarsia birnaculata is the Encar- sia I [India], G [Guatemala] and S [Sudan] species released in Florida against Bemisia argentifolia (Nguyen & Bennet 1995). Host. — Reared from Bemisia argentifolia and B. tabaci (Aleyrodidae). Distribution. — India, Philippines, Thai- land and USA (Florida, Texas?), and pos- sibly Sudan, Israel and Mexico [possibly culture contaminations]. Material Examined. — Holotype, V, India: labar- bhani, 19.vii.1994, culture in Gainesville, Florida, R. Nguyen, autoparasitoid, M9201S. Deposited in USNM. Paratypes (80 females, 29 males): Honduras: lab culture [Florida] vi.1992, quarantine 1261 (1 •■ 176 USNM). Hong Kong: kou loon Park, 14.vii.1992, F. Bennett Y944, Bemisia tabaci on Chamaescyel hirta (29 USNM). India: same data as holotype (4 9, USNM, 79 Id BMNH, 109 Id UCRC); quarantine [Florida], 30.V.1991, R. Nguyen 903, 1065, 1248, Bemisia tabaci (69 USNM); Tabarhani, 1 4. i v. 1992, R. Nguyen, ex Be- misia tabaci on tlibiscits (3 9 USNM); lab culture [ Flor- ida] quarantine, original shipment 124S, 1992 (69 4o" USNM). Phillipines: Benguet, 8.iii.l994, Legaspi, Carruthers, Poprawski, ex Bemisia tabaci on white po- tatoes, autoparasitoid, M94014 (119 UCRC, 59 BMNH); Quezon Dolores, 15.xii.1993, C. Moomaw, M93069, on eggplant (2 9 TAMU). Israel: [lab culture in Florida], R. Nguyen 1250, on Bemisia tabaci (79 16" USNM). USA: Florida: Alachua Co., Gainesville, 7.ix.l992, F. Bennett Y935, Bemisia tabaci on Euphorbia heterophylla (29 16 USNM); Alachua Co., Gainesville, 14. ix. 1992, F. Bennett Y958, Bemisia tabaci on Sesamum (29 USNM); Alachua Co., Gainesville, 27.viii.1992, F. Bennett Y936, Bemisia tabaci on Sesamum indicum (29 26 USNM); Alachua Co., Gainesville, 6.ix.l992, F. Bennett Y934, Bemisia tabaci on Chamaesyce hirta (39 Id USNM); Hamilton Co., Jasper, 8.iv.l992, F. Ben- nett Y693, Magnolia sp. (Id USNM); Texas: Mission biological control laboratory (in culture from India), 16.xii.1993, M92018 (5 9 TAMU). Mexico: Taxco, 1992, P. Stansley 1292, Bemisia tabaci on Chamaesyce hyssopifolia (1 9 USNM). Sudan: [lab culture in Flori- da], 30.iv.1992, R. Nguyen 1249, on Bemisia tabaci (39 USNM). Thailand: xi.1977, G. Yonimoton [?], cotton (19 USNM). Encarsia citri (Ishii) (Figs. 13, 14) Prospaltella citri Ishii, 1938: 29-30. Type data: Ja- pan: Nagasaki. Syntypes, female. Type de- pository: National Institute of Agroenviron- mental Sciences, Tsukuba, Japan. Described: female. Reared from Dialeurodes citri. Encarsia stroma; Polaszek ct al. 1992: 338. Incor- rect synonymy. Encarsia citri, Huang & Polaszek 1998: 352. Re- vised status. Female. — Antenna with 6 flagellomeres, clava 3-segmented; Fl 2.4-3.0X as long as broad and 0.9 X as long as F3; antenna very pale brown, contrasting in color to rest of the body which is yellow. Vertex weakly areolate to colliculate, ocellar tri- angle weakly areolate; dorsal setae stout. Maxillary palpus 1 -segmented. Mandibles chisel-shaped or very weakly 3/3 dentate, apical margin nearly flat. Mesosoma with weak hexagonally areolate sculpture dor- 158 Journal of Hymenoptera Research sally; midlobe with 5 pairs of setae (rarely 6 pairs), posterior and lateral setae only slightly stouter than medial setae, side lobe with 3 pairs, axilla with 2 pairs (lat- eral pair minute), and scutellum 2 pairs, medial pair lateral to sensillae. Scutellar sensillae ovoid and separated by less than than their own maximum diameter (rarely by a full diameter); median groove dis- tinct. Apical spur of mid tibia 0.7-0.8 X as long as basitarsus; basitarsomere of mid- tarsus with 2-3 strong pegs, tarsomeres 3- 5 each with 1 apical peg. Tarsal formula 5-5-5. Fore wing 2.39-2.68 X as long as broad, marginal fringe 0.1 8-0.25 X width of fore wing; disc uniformly setose; costal cell with row of 15-19 small setae and 2- 3 long marginal setae apically; submargin- al vein with 2 (rarely 3 on one side) large setae, basal area with 7-9 setae posterior to submarginal vein; frenal fold with sev- eral minute thornlike spines; wing hya- line. Metasoma yellow; laterally with weak cellulate reticulation; dorsal setal formula from tergite III: 2-2-2-6-6-6. Ovi- positor 1. 22-1. 34 X as long as mid tibia, 2.00-2.34 x as long as clava, 0.5-0.6 x as long as metasoma (base to tip of third val- vula); third valvula stout (Fig. 14), 1.26- 1.95X as long as broad, barely, if at all, extruded beyond epygium, entirely yel- low. Male. — Overall coloration pale brown, darker brown pattern on head and meso- soma similar to £. bimaculata female, gas- ter entirely brown. Setation pattern of me- sosoma and metasoma as in female; some males with fewer setae in basal area (13) or fore wing and costal cell (4). Basitarso- mere of middle leg with only 2 pegs. An- tenna with 5 flagellomeres, apical two fla- gellomeres (5&6) fused, with segments distinguished only by a break in the pat- tern of linearia (Fig. 13). Comments. — The shape and setation of the third valvula are distinct among all members of the strenua group. This spe- cies is most easily confused with £. pro- transvena but can be separated by the fea- tures outlined in the key and usually the presence of 5 pairs of setae on the midlobe of the mesoscutum (versus 4 pairs). Hosts. — Aleyrodidae: Dialeurodes citri (Ashmead) (Ishii 1938). Distribution. — Palaearctic: Japan (Ishii 1938). Material Examined (6 males, 12 fe- males).— Japan: Kyushu: Kagoshima City, 2.v, 15.vii.1970, S. Ohga, R70-31, R72-36, Dialeurodes citri on Citrus spp. (12$ 4d 1 pupa UCRC); Kagoshima, Taremian, 18- 19.viii.1972, S. Ohga, R72-50 (26 UCRC [with 26 of E. transvena on same slide]); Fukuoka, Tsuyazaki, 4.viii.l995, H. Kajita, ex Dialeurodes citri on Citrus (2? BMNH). Encarsia neocala Heraty and Polaszek, new species (Figs. 17-24) Female. — Antenna with 6 flagellomeres, clava 3-segmented; Fl 1.6-1.9X as long as broad and 0.8-1. Ox as long as F3; antenna yellow basally, darkened beyond or in- cluding pedicel. Vertex confused trans- verse colliculate, ocellar triangle similar but sculpture somewhat areolate; head yellow except for weak infuscation on in- ner margins of ocelli; dorsal setae slight. Maxillary palpus 1-segmented. Mandibles 2/2 or 3/3 dentate, ventral teeth acute, dorsal tooth blunt or rounded and often not apparent. Mesosoma mostly yellow except pronotum brown and axillae and propodeum laterally pale brown. Meso- soma with light hexagonally areolate sculpture dorsally; midlobe with 4-5 pairs of setae, posterior and lateral setae only slightly stouter than medial setae, side lobe with 3 pairs, axilla with 2 pairs (lat- eral pair minute), and scutellum 2 pairs, medial pair lateral to sensillae. Scutellar sensillae ovoid and separated by less than than their own maximum diameter; me- dian groove narrow and distinct. Apical spur of mid tibia 0.8-1 .OX as long as bas- itarsus; basitarsomere of midtarsus with 2-4 strong pegs, tarsomeres 4 and 5 each with 1 apical peg. Tarsal formula 5-5-5. Volume 9, Number 1, 2000 159 A ft i^Mi fL / i i i / -/- , •' / / '! l r /i / Figs. 17-24. Encarsia neocala, female: 17, antenna; 18, midtarsi; 19, third valvulae; 20, body; 21, mid tibia; 22, mandible; 23, base of fore wing; 24, fore wing. Fore wing 2.32-2.44X as long as broad, marginal fringe 0.29-0.38 X width of fore wing; disc uniformly setose; costal cell with row of 12-16 small setae and 2-4 long marginal setae apically; submarginal vein with 2 large setae, basal area with 4- 5 setae posterior to submarginal vein; fre- nal fold with several minute thornlike spines; wing mostly hyaline, weakly in- fuscate at base of submarginal vein. Me- tasoma mostly yellow, tergites I + II usu- ally brown, sometimes almost completely yellow; laterally with weak cellulate retic- ulation; dorsal setal formula from tergite III: 2-2-2-6-6-6. Ovipositor 1.60-1. 79 X as long as mid tibia, 2.16-2.56X as long as clava, 0.7X as long as metasoma (base to tip of third valvula); third valvula elon- gate (Fig. 19), 2. 6-3. Ox as long as broad, extruded beyond epygium, entirely yellow. 160 Journal of Hymenoptera Research Male. — Unknown. Comments. — The longer third valvulae, longer ovipositor relative to gaster length, and stouter ovipositor (indicated by a ten- dency for the tip to be bent in slide mounts) place this species as similar to £. stremia. However, in E. neocala the third valvula is entirely yellow (versus tipped brown in E. strenua), the midlobe of the mesoscutum is entirely yellow with the axillae slightly darker (versus entirely yel- low or the midlobe slightly darkened an- teriorly in £. strenua), fewer setae in the costal cell, and the ocellar triangle more weakly sculptured. Host. — Reared from Orchamoplatus cale- donicus (Dumbleton) (Aleyrodidae) [reads Orchamnus neocaledonicus on label] on Citrus. Material Examined. — Holotype, 9, New Caledonia, Noumea, 10.ix.1970, G. Fadres, ex [Orchamnus neocaledonicus]. Deposit- ed in USNM. Paratypes (37 females): same data, (5? USNM, 5$ F3MNH, 27? UCRC). Encarsia protransvena Viggiani (Figs. 15, 25-28) Encarsia protransvena Viggiani, 1985a: 89-90. Type data: USA: FL, Broward Co., Fort Lau- derdale. Holotype female, by original desig- nation. Type depository: IEUN. Described: female. Ulust. Reared from ~Dialeurod.es kirkal- dii. Placed in strenua group by Hayat (1989). Encarsia strenua; Polaszek et al. 1992: 388. De- scribed: female. Illust. (in part). Encarsia stroma; Schauff et al. 1996: 29. De- scribed: female. Illust. Misidentification. Encarsia protransvena; Schauff et al. 1996: 27. De- scribed: female. Illust.; Huang & Polaszek 1998: 1941; Polaszek et al. 1999: 158. Female. — Antenna with 6 flagellomeres, clava 3-segmented; Fl 2.1-3.2X as long as broad and 0.7-1.1 X as long as F3; antenna and entire body yellow except small spots of brown along inner margins of ocelli. Vertex weakly areolate, ocellar triangle areolate; dorsal setae stout. Maxillary pal- pus 1-segmented. Mandibles 3/3 dentate, teeth sharp or blunt. Mesosoma with weak hexagonally areolate sculpture dorsally; midlobe usually with 4 pairs of setae (rarely with as few as 3 setae per side and as many as 5 pairs total), posterior and lat- eral setae much stouter than medial setae (almost twice as broad), side lobe with 3 pairs, axilla with 2 pairs (lateral pair mi- nute), and scutellum 2 pairs, medial pair lateral to sensillae. Scutellar sensillae ovoid and separated by less than than their own maximum diameter (rarely by a full diameter); median groove distinct. Apical spur of mid tibia 0.8-0.9 X as long as basitarsus; basitarsomere of midtarsus with 2-3 strong pegs, tarsomeres 3-5 each with 1 apical peg. Tarsal formula 5-5-5. Fore wing 2.62-3.1 IX as long as broad, marginal fringe 0.20-0.40 X width of fore wing; disc uniformly setose; costal cell with row of 12-18 small setae and 1-3 long marginal setae apically; submarginal vein with 2 large setae, basal area with 7- 9 setae posterior to submarginal vein; fre- nal fold with several minute thornlike spines; wing hyaline. Metasoma laterally with weak cellulate reticulation; dorsal se- tal formula from tergite III: 2-2-2-6-6-6. Ovipositor 1.29-1. 74 X as long as mid tib- ia, 2.0-3. 37X as long as clava, 0.6-0.7X as long as metasoma (base to tip of third val- vula); third valvula stout (Fig. 15), 1.7- 2.8 X as long as broad, barely, if at all, ex- truded beyond epygium, entirely yellow. Male. — Overall coloration pale brown, darker brown pattern on head and meso- soma similar to £. bimaculata female, gas- ter entirely brown. Setation pattern of me- sosoma and metasoma as in female, but difference in size of setae on midlobe of mesoscutum much less. Costal cell with 9- 10 small setae and basal area with 5-7 se- tae. Basitarsomere of middle leg with only 2 pegs. Antenna with 5 flagellomeres, api- cal two flagellomeres (5&6) fused with li- nearia overlapping (Fig. 26). Comments. — This species can be separat- ed from other members of the strenua group by the shorter ovipositor and third valvulae, more delicate antenna (clava Volume 9, Number 1, 2000 161 Figs. 25-30. 25-28, Encarsia protransvena: 25, antenna of female; 26, antenna of male; 27, apex of gaster, dorsal view; 28, base of fore wing. 29, Encarsia sophia, apex of gaster. 30, Encarsia strenua, antenna oi male. only slightly broader than the funicle), they are almost always arranged in 5 pairs and longer fore wing. The setae of the in £. citri, and the arrangement of setae midlobe of the mesosoma are usually ar- (shorter and more apical) on the third val- ranged in 4 pairs (rarely 3 or 5), whereas vulae is distinct from £. citri. Of the hun- 162 Journal of Hymenoptera Research dreds of specimens examined, only the two males, collected at the same locality as other E. protransvena females, can be at- tributed to this species. This is probably the parasite recorded as Encarsia stremta attacking Bemisia argentifolia in South Car- olina (Simmons 1998). The specimens of E. protransvena de- scribed from China (1 female) and Taiwan (2 females) by Huang & Polaszek (1998) are nearly identical to those described in this paper. However, the taiwanese spec- imens have a band of 14 setae in the basal area of the fore wing (their fig. 273) as compared to a single row of only 7-9 setae in both E. protransvena and E. citri. They also describe 5 pairs of setae on the mid- lobe of the mesosoma, which is rarely en- countered in specimens from the New World (usually 4 pairs). The Chinese spec- imen agrees for all characters with £. pro- transvena, but it has a strongly and densely reticulate vertex, as compared to the very weak and more broadly spaced sculpture of all other £. protransena examined, and the basal segment of the clava is more dis- tinctly separated from the following seg- ment. More material will need to be ex- amined before these can be included or excluded from £. protransvena with confi- dence. Hosts. — Aleyrodidae: Aleurotrachelus rubi Takahashi (Huang & Polaszek 1998), Bemisia argentifolia Bellows & Perring, Be- misia tabaci (Gennadius), Dialeurodes citri (Ashmead), Dialeurodes citrifolii (Morgan), Dialeurodes kirkaldii (Kotinsky) (Viggiani 1985a), Parabemisia myricae (Kuwana), Tri- aleurodes abutiloneus (Haldeman), Trialeu- rodes packardi (Morrill) (Huang & Polaszek 1998), Trialeurodes variabilis (Quaintance). Diaspididae: Aspidiotus sp.?, Parlatoria zi- ziphil. [diaspidid hosts almost certainly are incorrect associations] Distribution. — Nearctic: United States of America: California (Polaszek et al. 1992), Florida (Viggiani 985a), Hawaii (Nguyen & Hamon 1989 [not examined]). Neotrop- ical: Colombia; Cayman Islands; Hondu- ras (Polaszek et al. 1992); Puerto Rico (Po- laszek et al. 1992). Palaearctic: Spain; Egypt (Polaszek et al. 1999). Oriental: Peo- ple's Republic of China: Guangdong; Tai- wan (Huang & Polaszek 1998). Material Examined (390 females, 2 males). Colom- bia: Valle: Dagua, 28.V.1991, R. Caballero, 108, Di- aleurodes citrifolii on Citrus aurantifolia (4 9 USNM). Grand Cayman: Cayman, 13. vi. 1986, F.D. Bennett, Y729, whitefly on Citrus (19 USNM); Savannah, 17.X.1987, F.D. Bennett, 164, Dialeurodes citrifolii on Citrus (2 9 USNM). Puerto Rico: Rio Piedras, 5.V.1990, 4.vi.l990, 18.ix.1988, 18.xi.1988, 24.xi.l988„ F.D. Bennett, Y501, Y560, Y1042, Y1043, Y1051, 128, 828, 1585, Dialeurodes citrifolii on Citrus, Dialeurodes citri on Citrus, Dialeurodes kirkaldyi on Jasminum, Gym- naspis on bromeliad (15 9 USNM); Corozal, 27.xi.1988, l.xii.1989, F.D. Bennett, 159, 161, Dialeu- rodes citrifolii on Citrus (5 9 USNM); Adjuntas, 15.xi. 1988, F.D. Bennett, Y719, Y1045, Y1046, 163, Di- aleurodes citrifolii on Citrus (89 USNM); Mayaquez, 16.xi.1988, F.D. Bennett, Y1047, 160, Dialeurodes citri- folii on Citrus, Parlatoria ziziphi on Citrus (5 9 USNM); Fortuna, 3.xii.l987, F.D. Bennett, 151, Dialeurodes on Citrus (8 9 USNM). Spain: Valencia, xi.1994, F.S. Mari, Parabemisia myricae (4 9 USNM). USA: Florida: Dade County: North Miami "Little Haiti", 28.V.1987, 22.vi.1987, R. Prange & F.D. Bennett, RP-1, Dialeurodes citrifolii on Citrus (209 lcT TAMU); 17.X.1987, F.D. Bennett & H. Glenn, FDB-3, Dialeurodes citrifolii on Citrus, Dialeurodes citrifolii on Dioscorea alata (109 TAMU); 19.U986, F.D. Bennett, J.H. Frank & R. Nguyen, Dialeurodes citrifolii on Citrus (19 USNM); 12.X.1987, F.D. Bennett, 1331, Aspidiotus destructor on Dioscorea alata, Dialeurodes citrifolii on Dioscorea alata (39 USNM); Miami, 12.x. 1987, F.D. Bennett & H. Glenn, 87, Dialeurodes citrifolii on Dioscorea alata (69 lcT USNM); 29.X.1990, F.D. Bennett, 746, Bemisia tabaci on Ricinus communis (19 USNM). Alachua County: Gainesville, records for every month of the year, 1989 to 1993, collectors include F.D. Bennett, G.A. Evans, J. Marenco and L. Nong, host records include Bemisia tabaci on soybean, Bemisia tabaci on Emilia sonchifolia. Trialeurodes abutilonea on Hibiscus mutabilis, Trialeu- rodes packardi on Cercis canadensis, Dialeurodes citri on Ligustrum siuense, Dialeurodes citri on Viburnum, Di- aleurodes citrifolii on Citrus, Bemisia tabaci on Sesamum indicum, Bemisia tabaci on Chamaesyce hyssopifolia, Be misia tabaci on Cassia obtusifolia, Bemisia tabaci on Eu phorbia heterophylla, Bemisia tabaci on Brassica oleracea var. acepbala, Bemisia tabaci on Hibiscus mutabilis, Be misia tabaci on Desmodium tortuosum (1709 USNM); Micanopy, 3.iv.l988, 28.vii.1991, ll.ix.1988, 21. ix. 1989, 21. x. 1989, F.D. Bennett, Y287, 1, 95, 127, 154, 771, Di- aleurodes citri on Citrus, Dialeurodes citri on Viburnum. Dialeurodes citri on Melia azadarach, Trialeurodes pack- ardi on Cercis canadensis (189 USNM); Alachua, Volume 9, Number 1, 2000 L63 xi.1992, H. McAuslane, 1339, Bemisia tabaci on peanut (39 USNM); near Alachua, 12.ix.1991, F.D. Bennett, Y399, Y400, Bemisia tabaci on soybean 5? (USNM). Monroe County: Isla Morada, 1. v. 1993, F.D. Bennett, Y1060, Dialeurodes citrifolii on Citrus aurantifolia (39 USNM). Broward County: Pompano Beach, 22.iii.1992, 23.vii.1991, 23.viii.1991, 4.xi.l988, F.D. Bennett, Y274, Y437, Y490, Y664, Bemisia talma on I m- ilia sonchifolia, Dialeurodes citrifolii on Citrus, Dialeu- rodes kirkaldyi on Jasminum (109 USNM). Orange County: Apopka, 29.xi.1989, K. Hoelmer, 380, Trialeu rodes variabilis on Carica papaya (29 USNM). Indian River County: Beach, 10. iv. 1987, D. Mooney, Creni- dorsum new species on Coccoloba uvifera (2 9 USNM). Jackson County: Mariana, 18.viii.1987, F.D. Bennett, 946, Dialeurodes citri on Melia (259 USNM). Bradford County: Starke, 26.iv.1989, W.A.A. Klerks, Pseudau- lacaspis cockerelli (1 9 USNM). Gadsden County: Quin- cy IFAS, 13. ix. 1991, F.D. Bennett, Y396, Y401, Bemisia tabaci on soybean, Bemisia tabaci on tomato (49 USNM). Palm Beach County: West Palm Beach, 6.ii.l991, l.iv.1991, 13.vi.1992, 19.x. 1991, F.D. Bennett, F104, Y428, Y450, Y838, 1183, Dialeurodes kirkaldyi on Jasminum, Bemisia tabaci on Chamaesyce hyssopifolia, Be- misia tabaci on Emilia sonchifolia (89 USNM). Hills- borough County: Ruskin, 29. vi. 1990, Eclipta (19 USNM). Osceola County: Orange Creek, 18.iii.1990, F.D. Bennett, 407, Trialeurodes abutilouea on Solatium americanum (19 USNM); Canoe Creek, 22.xi.1990, F.D. Bennett, 1091, Dialeurodes or Aleurothrixus floc- cosus on Citrus (1 9 USNM). County?: Brywood, 4.vi.l992, F.D. Bennett, Y901, black whitetly on Uq- uidambar styraciflua (19 USNM). South Carolina: Charleston County: Charleston, 31.viii.1993, 7.ix.l993, 29. ix. 1993, A. Simmons, 1610, 1614, 1615, Bemisia ta- baci on sweetpotato (29 USNM). Mississippi: Simp- son County: Magee, 10. ix. 1994, D.H. Headrick, DHH96-0423, DHH96-0424, DHH96-0438, Bemisia ar gentifolii on Cleome hasslerana (3 9 UCRC). Forrest County: Hattiesburg, 5.ix.l994, D.H. Headrick, DHH94-0361, Bemisia argentifolii on okra (1 9 UCRC). Georgia: Bryan County: Savannah, 15. xi. 1991, F.D. Bennett, 1 161, Bemisia tabaci on Chamaesyce hyssopifolia (69 USNM). Tift County: Tifton, 20.x. 1992, J. Cham- berlain, 1355, 1395, Bemisia tabaci on Gossypium hir- sutum (119 USNM); 29. iv. 1992, 15. ix. 1992,' W. Hud- son, Y714, Y715, Y716, Y717, 1346, Bemisia tabaci on Gossypium liirsutum, Dialeurodes citri on Jasminum (219 USNM). Encarsia sophia (Girault & Dodd) (Fig. 28) Coccophagus sophia Girault & Dodd, 1915: 49, 56. Type data: Australia: QLD, Cairns. Syntypes, female. Type depository: Brisbane: Queens- land Museum, Queensland, Australia; type no. Hy.2926. Prospaltella transvena Timberlake, 1926: 312-315. Type data: USA: Hawaii, Oahu. Holotype fe- male. Type depository: Honolulu: Bernice P. Bishop Museum, Dept. Ent. Coll., HI, USA; type no. 5690. Described: both sexes. Illust. Reared from Trialeurodes [as Aleyrodes] vapor- ariorum on tomato. Placed in lahorensis group by Hayat (1989) and in strenua group by Po- laszek et al. (1992) and Hayat (1998). Gerling (1985) states that E. sublutea is known in Ha- waii as E. transvena. New Synonymy. Prospaltella sophia; Compere 1931: 11. Change of combination. Prospaltella sublutea Silvestri, 1931: 20-22. Type data: Somalia: Duca [?]. Syntypes, female. Type depository: IEUN. Described: female. Illust. Synonymy with transvena by Gerling & Rivnay in Viggiani 1985: 90. Prospaltella betnisiae Ishii, 1938: 30. Type data: Japan: Ikawa-cho, Mei-Ken. Syntypes, fe- male. Type depository: National Institute of Agroenvironmental Sciences,Tsukuba, Japan. Described: female. Synonymy with transvena by Polaszek et al. 1992: 388-389. Reared from Parabemisia [as Bemisia] myricae Kuwana. Prospaltella flava Shafee, 1973: 254-255. Type data: India: Uttar Pradesh, Aligarh. Holotype female, by original designation. Type depos- itory: Aligarh: Aligarh Muslim University, Deptartment of Zoology, India. Described: female. Illust. Synonymy by Hayat 1989: 72. Preoccupied by flavus Compere 1936: 300. Questionably treated as a junior synonym of transvena by Viggiani (1985). The type mate- rial was reared from a coccid which, if cor- rect, represents a significant departure in host range. Encarsia sophia; Viggiani 1985b: 249. Described: female. Illust. Change of combination. Encarsia transvena; Gerling & Rivnay in Viggiani 1985a: 90-92. Described: both sexes. Illust. Change of combination. Encarsia shafeei Hayat, 1986: 163. Replacement name for E. flava Shafee. Encarsia transvena; Hayat 1989: 71-73; Polaszek et al. 1992: 388-389; Schauff et al. 1996: 31-33; Hayat 1998: 205-207; Huang & Polaszek 1998: 1954-1956. Described: both sexes. Illust. Continents. — This is the most distinctive species in the stroma group and can be recognized by the transversely striate ocel- lar triangle, a patch of longer setae in the 164 Journal of Hymenoptera Research posterior half of the wing disc, and pres- been some accidental movement, especial- ence of only 4 setae on Mt7 (Polaszek et ly to Hawaii (type locality of E. transvena), ah 1992; Schauff et al. 1997). All of the fe- determining the origin and movements male specimens examined were complete- through introductions of each geographi- ly yellow except for a single female from cal population will be essential to the cor- Thailand (ex Bemisia tabaci on Lantana ca- rect assignment of names, if this is indeed marara [USNM]), which has the same dark a species complex. pattern of colour as £. bimaculata. Especially for the purposes of biological The type specimen of £. sopJiia, like control and the associated need for accu- most of the Girault material, is in poor rate identification of museum and field condition; however, all of the diagnostic material, what concept of a species is most features shared with E. transvena, and oth- useful? Among the many definitions pro- er species names currently synonymized posed, there are only two basic and, in with E. tranvena, are clearly visible. Since some ways opposable, concepts. The Bio- it was described as different species in logical Species Concept (Mayr 1963) re- Australia (Girault & Dodd 1915), Hawaii quires that populations of the same spe- (Timberlake 1926), Somalia (Silvestri cies, whether they are in contact or not, 1931), and Japan (1938), we assume that have the potential to interbreed. The de- this group was widespread in the Old gree and conditions under which inter- World, prior to recent movements of spe- breeding will take place are problematic cies for use in the biological control of (Donoghue 1985), especially when they whiteflies. Both authors of this paper have are used to assess allopatric populations examined several hundred E. transvena that have varying degrees of reproductive from around the world and found no di- incompatibility (cf. Rosen & DeBach 1979). agnostic characters that would separate The ability to interbreed in Aphelinidae any population as distinct from the others, and other Chalcidoidea can be further Recent studies, however, have shown complicated by other factors, such as en- mating incompatibilities and morphomet- dosymbiotic bacteria that can induce re- ric (shape) differences between popula- productive incompatibility within or be- tions of £. transvena from Spain and Pak- tween populations (O'Neill et al. 1997; istan (G. Viggiani, pers. comm.) that al- Luck et al. in press). These incompatible lude to the potential for this to be a cryptic populations can even be cured through or sibling species complex. Should we antibiotics or heat treatments and inter- proceed with the synonymy or wait for breeding reestablished (Stouthamer & further evidence of potential species Luck 1993; Luck et al. in press), which boundary characteristics? The synonymy brings into question the criterion of in- is justified on both philosophical and prac- compatibility for separating otherwise in- tical grounds. For practical purposes, the distinguishable populations that have assignment of an appropriate specific ep- been treated previously as either cryptic, ithet to any of the geographically isolated sibling or semi species. In contrast, under populations is complicated by their taxo- a Phylogenetic Species Concept (Dono- nomic history. Five names were proposed ghue 1985), at least one diagnostic char- within this complex, of which £. sophia is acter is required for each species to denote the oldest, and £. transvena (described an evolutionary lineage. The simple no- from Hawaii, where it may have been ac- tion of reproductive incompatibility, the cidentally introduced from Japan, south- foundation of the Biological Species Con- east Asia or Australia) is probably the cept, is not sufficient. A discrete morpho- least applicable to the populations from logical character found in all individuals Pakistan or Spain under study. If there has is usually taken as the best criterion for Volume 9, Number 1, 2000 165 separating species; species are both diag- nosable and unique. Whether both molec- ular and morphometric (shape) character- istics are considered diagnostic under a Phylogenetic Species Concept is debat- able, especially since they may be describ- ing only population level differences that do not contribute to the "potential" for a species to interbreed (Avise & Wollenberg 1997). In any case, differences must be demonstrated over the range of a species to assure that they are not merely repre- sentative of clinal variation in a series of populations belonging to a single species. Currently, £. sophia and E. transz^ena, and all of the names synonymized of £. trans- vena, cannot be distinguished using diag- nostic morphological characters, and syn- onymy under the oldest valid senior name, E. sophia, is justified. Hosts. — Aleyrodidae: Acaudalei/rodes rhachipora (Singh) (Hayat 1989), Aleurocy- botus indicus David & Subramaniam (Po- laszek et al. 1992), Aleurodicus dispersus Russell (Polaszek et al. 1992), Aleurolobus sp. near niloticus (Hayat 1989a), Bemisia? (= Aleyrodes) hibisci (USNM 29067), Bemi- sia tabaci (Gennadius) (Polaszek et al. 1992, Hayat 1989, Gerling 1985), Parabemisia myricae (Kuwana) (Polaszek et al. 1992), Pealius hibisci (Kotinsky) (Timberlake 1926), Trialeurodes ricini (Misra) (Hayat 1998), Trialeurodes zmporariorum Westwood (Polaszek et al. 1992, Timberlake 1926, Gerling 1985). Aphididae: Aphis sacchari Zehntner? (Timberlake 1926). Coccidae:? (Shafee 1973). Psyllidae: Diaphorina citri Kuwayama (Polaszek et al. 1992). [The psyllid host is correct, although possibly E. sophia is a hyperparasitoid on Tamarixia radiata (Waterston) [Huang & Polaszek 1998], so the aphid and coccid associations also may be "primary" hosts of hyperpar- asitic males.] Distribution. — Cosmopolitan in the Old World, introduced in the New World. Af- rotropical: Burundi; Cape Verde (Hayat 1998); Ivory Coast; Morocco; Niger (Hayat 1998); Sierra Leone; Somalia (Silvestri 1931). Oriental: Hawaiian Islands (Tim- berlake 1926); Hong Kong; India (Hayat 1989); Indonesia; Sri Lanka (Hayat 1998); Pakistan (Hayat 1989); People's Republic of China (Huang & Polaszek 1998); Tai- wan; Thailand. Palaearctic: Japan (Ishii 1938); Spain. Material Examined (289 females, 71 males). — Burun- di: Bujumbura, vii.1987, J. Yaninek, whitefly on Cas- sava (1? USNM). Hong Kong: Kovvloon Park, 14.vii.1992, F.D. Bennett, Y944, Bemisia tabaci on Cha- maesyce (?) hirta (1 9 USNM). India: Rajkot, 9.ii.l958, G.W. Angolet, whitefly on Ricinu^ communis (49 USNM); India (no other locality), 19.xii.1990, G. But- ler, Bemisia tabaci (19 USNM). Indonesia: Java: Ban- doeng, x.1929, C. P. Clausen, 2420, Asterochiton (109 UCRC). Ivory Coast: 20 km west of Abidjan, 21.iv.1988, L.D.C. Fishppol, 125, Bemisia tabaci on Ma- nihot esculentum (2 9 USNM). Japan: Shikoku: Kochi, 23.vii.1980, P. DeBach, R.80.28, R.80.29, R.80.34, Par- abemisia myricae on Moras, Parabemisia myricae on Cit- rus (43 9 lei UCRC); 20-25.ix.1979, M. Rose, R79-67- 1 & 2, Parabemisia myricae on Morns (139 56 UCRC); Kyushu: Kagoshima City, 8.viii.l972, S. Ogo, R72-42, "Dialeu.rod.es citri on Citrus (39 UCRC); Honshu: Mie Prefecture, near Kuana City, 11, 15, 16, 18, 20, 21.viii.1981, M. Rose, R81-34, R81-35, R81-36, R81-37, R81-39, R81-40, R81-41, R81-42, Parabemisia myricae on Morns (60 9 46 UCRC); Shizuoka, L7.vii.1981, 5.viii.l982, K. Furuhashi, R82-30, on Moms (49 UCRC); 22-24.viii.1981, M. Rose, R81-43, Parabemisia myricae on Moras (19 UCRC). Morocco: Agadir, vi.1992, Parabemisia myricae (19 USNM). Niger: 4.xi.l987, Hansen, CIE A19340, Alcurocybotus indicus on rice (19 USNM). Pakistan: Sialkot, 25.ix.1969, R. Ahmad, R69-110, Dialeurodes citri on Citrus (16 UCRC); Pakistan (no other locality), L5.V.1987, L. Os- borne, 803, Bemisia tabaci (19 USNM). People's Re- public of China: Sichuan Province: Bei-Pei District, 19.viii.1980, P. DeBach, C3, whitefly on Moms (79 Id UCRC); Guangdong Province, Guangzhou, Ll.vii.1992, F.D. Bennett, Y894, whitefly on lactuca (16 USNM). Sierra Leone: viii.1960, Bemisia (79 USNM). Spain: Murcia, L9.vii.1994, EBCL— A. Kirk ,v L. Lacey, M93002, Bemisia tabaci on Lantana (19 9 UCRC). Taiwan: Tao-vuan, 3.xii.l993, C. Moomaw, Mission Biological Control Lab culture (Texas) M93054, Bemisia on Pomsetlia (39 TAMU). Thailand: Chiang Mai, 14.iii.1994, L. Lacey & A. Kirk M94041, Bemisia tabaci on Poinsettia (9 9 UCRC, 19 \6 TAMU); Chiang Mai University, L5.L1993, F.D. Ben- nett, Y1009, whitefly on Hibiscus mutabilis (1. USNM); Pang Hang, I5.iii.1994, I . Lacey & A. Kirk. M94049, Bemisia tabaci on Xanthium; W^n Ion Lea, 20.L1992, F.D. Bennett, 1224, Bernini tabaci on Gossy pium hirsutum (69 USNM); near Bangkok, iii.1992, F.D. Bennett, 1245, Bemisia tabaci on Solatium melon- 166 Journal of Hymenoptera Research gena (6 9 USNM). USA: Arizona: Cochise County: Guadalupe Canyon, 31 miles east of Douglas, 14.ix.1978, J.B. Woolley, 78035, whiteflies on mesquite (19 TAMU). Florida: Dade County: Miami, 14.vi.1992, 21.vii. 1991, 22.vii.1991, 31.viii.1991, 19.X.1991, F.D. Bennett, Y276, Y277, Y358, Y459, Y834, Bemisia tabaci on Chamaesyce hyssopifolia, Bemisia tabaci on Euphorbia(119 196 USNM); Homestead, 4.V.1992, F.D. Bennett, Y757, Bemisia talma on Emilia sonchifolia (16 USNM); Snapper Creek, 13. vi. 1992, 14.vi.1992, 19.X.1991, F.D. Bennett, Y457, Y830, Y837, Bemisia ta- baci on Emilia sonchifolia, Bemisia tabaci on Chamaesyce hyssopifolia (49 46 USNM). Alachua County: Ala- chua, xi.1992, H. McAuslane, 1376, Bemisia on peanut (3 9 USNM); Gainesville, 19.iii.1968, M. Kosztarab 874, Diaspis on Opuntia (2 9 USNM); 3.V.1992, G.A. Evans, 1251, on Ilex (16 USNM); 27.viii.1992, F.D. Bennett, Y936, Bemisia tabaci on Sesamum indicum (19 USNM); Micanopy, 20.U990, F.D. Bennett, Y487, whitefly on Citrus (19 USNM). Saint Lucie County: Fort Pierce, 9.viii.l992, F.D. Bennett, Y904, Bemisia ta- baci on Emilia sonchifolia (29 Id USNM). Okeechobee: Fort Drum, 17.x. 1991, F.D. Bennett, Y440, Bemisia ta- baci on Emilia (3 9 6c? USNM). Orange County: Apop- ka, 15.viii.1988, L. Osborne, Bemisia tabaci on Euphor- bia (179 46 TAMU); 10.ix.1987, Rose & Osborne, T87032, Bemisia tabaci (?) on Euphorbia (46 TAMU); vi-vii.1989, 14.x. 1989, 25.x. 1989, 29.xi.1989, K. Hoel- mer, Bemisia tabaci on Lantana, Trialeurodes variabilis on papaya (19 3d USNM). Broward County: Pom- pano Beach, 22.iii.1992, v.1993, 10.viii.1992, 19.x. 1991, 19.xi.1991, F.D. Bennett, Y448, Y541, Y665, Y908, Y1056, Bemisia tabaci on Chamaesyce hyssopifolia, Be- misia tabaci on Emilia sonchifolia (13 9 76 USNM). Palm Beach County: West Palm Beach, ll.iv.1992, 10.viii.1992, ll.ix.1992, F.D. Bennett, Y907, Y933, 1559, Bemisia tabaci on Emilia sonchifolia, Bemisia tabaci on Chamaesyce hyssopifolia (39 46* USNM). Monroe County: Isla Morada, 30.iv.1993, 18.x. 1991, F.D. Ben- nett, Y453, Y1108, Bemisia tabaci on Chamaesyce hys- sopifolia, Bemisia tabaci on Chamaesyce hirto (29 16 USNM); Key Largo, 18.x. 1991, F.D. Bennett, Y444, Be- misia tabaci on Desmodium tortuosum (39 36 USNM). Manatee County: Bradenton, 13. v. 1993, E. Vasquez, Y1053, Bemisia tabaci (49 26 USNM). Texas: Hidalgo County, Mission Biological Control Lab, 16.xii.1993, M93002 (MBCL culture voucher specimen), Bemisia tabaci (1 9 TAMU). California: Orange County: U.C. South Coast Field Station, 1 5-16. xii. 1982, S. Key, Par abemisia myricae on lemon (99 16 UCRC); Los An- geles County: San Gabriel, 22. iv. 1982, Rose & Ferren- tino, Parabemisia myricae on orange (19 UCRC). Ha- waii: Oahu: Moiliili, 15.iii.1984, B. Kumashiro, Bemisia tabaci on eggplant (7 9 TAMU). Encarsia strcnua (Silvestri) (Figs. 16, 30) Prospaltclla strcnua Silvestri, 1928: 34-36. Type data: China: Macao. Holotype female, by monotypy. Type depository: IEUN. De- scribed: female. Illust. Reared from Bemisia gijfardii on Citrus. Placed in strcnua group by Viggiani & Mazzone (1979) and Hayat (1989). Encarsia strenua, Viggiani & Mazzone, 1979: 46. Change of combination. Encarsia strenua; Polaszek et ai. 1992: 388; Schauff et al. 1996: 29. Described: female. Il- lust. Broadly defined to include what is now recognized as E. protransvena and E. citri. Encarsia strenua; Polaszek & Huang 1998: 1951- 53. Described: female. Illust. Based on Chi- nese and Taiwan material. Female. — Antenna with 6 flagellomeres, clava 3-segmented; Fl 2.6-3.6 X as long as broad and 0.9-1. OX as long as F3; antenna yellow basally, darkened beyond or in- cluding pedicel. Vertex and ocellar trian- gle strongly areolate; head yellow except for weak infuscation on inner margins of ocelli; dorsal setae robust. Maxillary pal- pus 1-segmented. Mandibles 2/2 or 2/3 dentate, teeth blunt to hardly recogniz- able, dorsal tooth, if visible, minute. Me- sosoma mostly yellow except pronotum pale brown medially and anterior margin of midlobe of mesoscutum sometimes pale brown, rarely axillae and metasoma more extensively pale brown. Mesosoma with hexagonally areolate sculpture dorsally; midlobe with 4-5 pairs of setae, posterior and lateral setae noticably stouter than medial setae, side lobe with 3 pairs (rarely with additional seta), axilla with 2 pairs (lateral pair minute), and scutellum 2 pairs, medial pair lateral to sensillae. Scu- tellar sensillae ovoid and separated by less than than their own maximum diameter; median groove narrow and distinct. Api- cal spur of mid tibia 0.7-0.9 X as long as basitarsus; basitarsomere of midtarsus with 4-5 large pegs, tarsomeres 3-5 each with 1 robust apical peg, rarely a single peg on tarsomere 2. Tarsal formula 5-5-5. Fore wing 2.44-2.67X as long as broad, marginal fringe 0.13-0.29X width of fore wing; disc uniformly setose; costal cell with row of 13-23 small setae and 1-3 Volume 9, Number 1, 2000 167 long marginal setae apically; submarginal vein with 2 large setae, basal area with 8- 12 setae posterior to submarginal vein; f re- nal fold without spines; wing hyaline. Me- tasoma mostly yellow, tergites I + II sometimes with faint infuscation, usually yellow; laterally with weak cellulate retic- ulation; dorsal setal formula from tergite III: 2-2-2-6-6-6. Ovipositor 1.56-1.97X as long as mid tibia, 2.43-3.26X as long as clava, 0.8-0.9 X as long as metasoma (base to tip of third valvula); third valvula elon- gate, 2.9-4.7X as long as broad, extruded beyond epygium, yellow except for ex- treme tip dark brown. Male. — Overall coloration pale brown, darker brown pattern on head and meso- soma similar to E. bimaculata female, gas- ter entirely brown. Setation pattern of me- sosoma and metasoma as in female but midlobe of mesoscutum with 4-5 pairs of setae. Midbasitarsomere with only 2 pegs. Antenna with 5 flagellomeres, apical two flagellomeres (5&6) fused, with segments distinguished only by a break in the pat- tern of linearia (Fig. 30). Comments. — The Californian population differs from the Asian forms by having darkened axillae and gaster (yellow in asian forms) and the third valvulae less elongate, being less than 3.5 X as long as broad. All females of E. strenua have at least 4 large pegs on the midbasitarso- mere, a long ovipositor and third valvula, 8-12 setae in the basal area of the fore wing, and the tip of the third valvula brown. The males from India have a slightly more elongate antenna (ca. 1.2X as long as head width), whereas the single male has a slightly more compact antenna (ca. 1.1 X as long as head width). The dif- ference could be due to different mount- ing techniques, and for other characters the males were identical. Hosts. — Aleyrodidae: Aleurobus subro- tundus Silvestri (Clausen 1934), Aleuropla- tus (Clausen 1934), Asterochilton (Clausen 1934), Bemisia gijfardii (Kotinsky) (Silvestri 1928), Dialeurodes citri (Ashmead) (Polasz- ek et al. 1992), Dialeurodes kirkaldii (Kotin- sky) (Polaszek et al. 1992), Parabemisia myr- icae (Kuwana) and Siphonius phillyreae (Haliday). The host records from Clausen (1934) have not been verified. The host range presented here is more restricted than that of Polaszek et al. (1992) after ex- clusion of much of the New World mate- rial now placed as E. protransvena. Distribution. — Nearctic: USA (Califor- nia). Oriental: China (Fujian, Guangdong) (Huang & Polaszek 1998); Hong Kong (Polaszek et al. 1992); India; Macau (Sil- vestri 1928). Material Examined (60 females, 40 males). — Hong Kong: Kowloon, 2.xii.l986, F. Bennett, ex Jasmine whitefly on Jasminium (19 16 USNM, 19 BMNH). India: Uttar Pradesh: Ranikhet, 4.viii, 7.viii, 18- 19.viii, 11. ix, lO.ix, 7.x. 1969, G. Chamora, R69-81, R69- 85, R69-89, R69-101, R69-103, D. citri on Citrus (8? 246 UCRC); West Bengal: Kalimpang, 2.ix, 12.ix, 18.ix.1969, Kurup (CIBC), R69-98, R69-104, 69-106, R69-116, Dialeurodes citri on Citrus (439 216 UCRC); Kayala: Bakarkhola [locality not verified], 21.vii.1969, G. Chondra, R69-77 (39 UCRC); [locality?] R69-93 (37 d UCRC). P.R. China: Guangdong: Guangzhou, ll.vii.1992, F. Bennett Y941, Dialeurodes on Cirrus (1 9 USNM). Israel? [no locality, just Tel-Aviv University header]: viii.1990, D. Gerling, on leaf with Siphonius phyllireae (29 USNM). ACKNOWLEDGMENTS We would like to thank Lai Shan Mui for the illus- trations and Dave Hawks for assistance in gathering measurement data. Mike Rose, Jim Woolley and Mike Schauff provided loans of material, and tremendous hospitality was offered by Gennaro Viggiani during visits to the Silvestri collection in Portici. This work was supported in part by grants from the National Biological Control Institute (to M. Schauff) .\nd the California Department of Food and Agriculture (JMH). LITERATURE CITED Avise, J. C. and K. Wollenberg. 1997. Phylogenetics and the origin of speed's. Proceedings oj the Na tional Academy of Sciences l,4: 7748-7755. Bellows, T. S., T. D. Paine, J. R. Gould, L. Bezark, J. C. Ball, W. Bentley, R. Coviello, J. Downer. P. I lam, D. Flaherty, P. Gouveia, K. Koehler, R. Molinar, N. O'Connell, E. Perry and G. Vogel. 1992. Biological control of ash whitefly: a success in progress. California Agriculture 46: 24, 27-28. Clausen, C. P. 1934. The natural enemies oi Aleyrod- 168 Journal of Hymenoptera Research idae in tropical Asia. Philippine Journal of Science 53: 253-265. Clausen, C. P. 1978. hit roil need parasites and predators of arthropod pests and weeds: A world review. Ag- ricultural Handbook 480 (USDA-ARS). 545 pp. Compere, H. 1931. A revision of the species of Coc- cophagus a genus of hymenopterous coccid-in- habiting parasites. Proceedings of the United States National Museum 78:1-132. Donoghue, M. J. 1985. A critique of the Biological Species Concept and recommendations for a phvlogenetic alternative. The Bryologist 88: 172- 181. Gerling, D. 1985. Parasitoids attacking Beiuisia tabaci (Horn.: Aleyrodidae) in Eastern Africa. Entomo- phaga 30: 163-165. Girault A. A. and A. P. Dodd. 1915. The cane grubs of Australia. Bulletin of the Bureau of Sugar Exper- iment Stations, Queensland Division of Entomology 2: 1-60. Hayat, M. 1986. Family Aphelinidae. Pages 143-171 in: Subba Rao, B. R. & Hayat, M, The Chalci- doidea of India and the adjacent countries. Part 2. A catalogue. Oriental Insects 20: 1-430. Hayat, M. 1989. A revision of the species of Encarsia Foerster (Hymenoptera: Aphelinidae) from India and the adjacent countries. Oriental Insects 23: 1- 131. Hayat, M. 1998. Aphelinidae of India (Hymenoptera: Chalcidoidea): A taxonomic revision. Memoirs of the American Entomological Society 13: 416 pp. Huang, J. 1994. Systematic Studies on Aphelinidae of China. Chongquing Publishing House, Chongqu- ing. 348 pp. Huang, J. and A. Polaszek. 1998. A revision of the Chinese species of Encarsia Forster (Hymenop- tera: Aphelinidae): parasitoids of whiteflies, scale insects and aphids (Hemiptera: Aleyrodidae, Diaspididae, Aphidoidea). journal of Natural His- tory 32: 1825-1966. Ishii, T. 1938. Descriptions of six new species of Aphelinidae with a revised table of genera. Kon- tyii 12: 27-32. Kuwana, I. 1922. Notes on a newly imported parasite of the spiny whitefly attacking citrus in japan. 5th Pacific Science Congress Proceedings 5: 3521- 3523. Luck, R. F., L. Nunney and K. Stouthamer. In press Factors affecting sex ratio and quality in the cul- turing of parasitii Hymenoptera: a genetic ami evolutionary perspective (Chapter 7). In Biologi- cal Control, T. Fischer el id. (ed). University of C al- ifornia Press. M.ivr, E. 1963. Animal Species and I volution. Harvard University Press. Meachum, C. A. and T. Duncan. 1987. Morphosys, ver sion 1.29. University of California, Berkeley. It ware) Mound, L. A. and S. H. Halsey. 1978. Whitefly of the 'world: A systematic catalogue of the Aleyrodidae (Ho- moptera) with host plant and natural enemy data. John Wiley & Sons: Chichester. Nguyen, R. and F. D. Bennett. 1995. Importation and field release of parasites against silverleaf white- fly, Bemisia argentifolii (Bellows and Perring) in Florida from 1990-1994. Proceedings of the Florida State Horticultural Society 108: 43-47. Nguyen, R., R. I. Sailer and A. B. Hamon. 1993. Cat- alog of Aleyrodidae on citrus and their natural enemies (Homoptera: Aleyrodidae). Occasional Papers of the Florida State Collection of Arthropods 8: 1-57.' Nguyen, R. and R. I. Sailer. 1987. Facultative hyper- parasitism and sex determination of Encarsia smi- thi (Silvestri) (Hymenoptera: Aphelinidae). An- nals of the Entomological Society of America 80: 713- 719. O'Neill, S. L., A. A. Hoffmann and J. H. Werren. 1997. Influential Passengers: Inherited Microorganisms and Arthropod Reproduction. Oxford University Press. Polaszek, A. 1991. Egg parasitism in Aphelinidae (Hymenoptera: Chalcidoidea) with special refer- ence to Centrodora and Encarsia species. Bulletin of Entomological Research 81: 97-106. Polaszek, A., S. Abd-Rabou and J. Huang. 1999. The egyptian species of Encarsia (Hymenoptera: Aphelinidae): a preliminary review. Zoologische Medelingeu Leiden 73: 131-163. Polaszek, A., G. A. Evans and F. D. Bennett. 1992. Encarsia parasitoids of Bemisia tabaci (Hymenop- tera: Aphelinidae, Homoptera: Aleyrodidae): a preliminary guide to identification. Bulletin of En- tomological Research 82: 375-392. Rosen, D. and P. DeBach 1979. Species of Apln/tis of the World (Hymenoptera: Aphelinidae). Series Entomologica 17: 801 pp. Sailer, R. I., R. E. Brown, B. Munir and J. C. E. Nick- erson. 1984. Dissemination of the citrus whitefly (Homoptera: Aleyrodidae) parasite Encarsia la- horensis (Howard) (Hymenoptera: Aphelinidae) and its effectiveness as a control agent in Florida. Bulletin of the Entomological Society of America 30: 36-39. Schauff, M. E., G. A. Evans and J. M. Heratv 1997. A pictorial guide to the species of Encarsia (Hvm: Aphelinidae) parasitic on whiteflies (1 lomoptera: Aleyrodidae) in North America. Proceedings of the Entomological Society of Washington c)8: 1-35. Shafee, S. A. N73. Indian species of the genus Pros- paltella Ashmead (Hym.: Aphelinidae). Entomo phaga 18: 251 -258. Silvestri, F. 1928 (1927). Contribuzione alia conoscen- za degli Aleurodidae (Insecta: Hymenoptera) viventi su Citrus in estremo Oriente e die loro parassiti. II Descrizione e notizie biologiche dei parassiti di Aleurodidi vivienti si Citrus. Bollet Volume 9, Number 1, 2000 169 tino del Laboratorio di Zoologia Generate e Agraria, Portici 21: 20-60. Silvestri, F. 1931. Descrizione di una nouva specie di Prospaltella della Somalia. Bollettino della Societa Entomologica Italiana 63: 20-22. Simmons, A. M. 1998. Survey of parasitoids of Bemisia argentifolii (Homoptera: Aleyrodidae) in coastal South Carolina using yellow sticky traps. Journal of Entomological Science 33: 7-14. Stouthamer, R. and R. F. Luck. 1991. Influence of mi- crobe-associated parthenogenesis on the fecun- dity of Trichogramma deion and T. pretiosum. En- tomologia Experimentia el Applicata 67: 183-192. Timberlake, P. H. 1926. New species of Hawaiian chalcid-flies (Hymenoptera). Proceedings of the Hawaiian Entomological Society 6: 305-320. Viggiani, G. 1985a. Notes on a few Aphelinidae, with descriptions of five new species of Encarsia Foers- ter (Hymenoptera, Chalcidoidea). Bollettino del Laboratorio di Entomologia Agraria 'Filippo Silvestri' di Portici 42: 81-94. Viggiani, G. 1985b. Additional notes and illustrations on some species of aphelinids described by A. A. Girault and A. P. Dodd in the genera Coccophagus Westwood, Encarsia Foerst. and Prospaltella Ashm. (Hym.: Chalcidoidea). Bollettino del Labor- atario di Entomologia Agraria 'Filippo Silvestri' ill Portici 42: 233-255. Viggiani, G. and P. Mazzone. 1979. Contributi alia conoscenza morfo-biologica delle specie del com- presso Encarsia Foerster -Prospaltella Ashmead (Hym. Aphelinidae). Bollettino del Laboratario di Entomologia Agraria 'Filippo Silvestri' di Portici 36: 42-50. Woollev, J. B. and H. W. Browning. 1997. Morpho- metry analysis of uniparental Aphytis reared from chaff scale, Parlatoria pergandii Comstock, on Texas citrus (Hymenoptera: Aphelinidae; Ho- moptera: Diaspididae). Proceedings of the Ento- mological Society of Washington 89: 77-94. J. HYM. RES. Vol. 9(1), 2000, pp. 170-175 A Peculiar New Genus and Species of Entedoninae (Chalcidoidea: Eulophidae) from Southeast Asia Rosichon Ubaidillah, John LaSalle, and Donald L. J. Quicke Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berks, SL5 7PY, UK and Department of Entomology, The Natural History Museum, Cromwell Road, London SW7 5BD, UK; and (RU) Museum Zoologicum Bogoriense, LIPI, Gedung Widyasatwaloka, Jin Raya km 46, Cibinong, Bogor 16911, Indonesia Abstract. — Ambocybe petiolata Ubaidillah and LaSalle gen. and sp. n. from Peninsular Malay- sia, Sulawesi and Papua New Guinea is described and illustrated and placed in the Entedoninae (Hymenoptera: Eulophidae). In addition to having several unique features, Ambocybe lacks im- portant characters that have been used previously as defining characters of the Entedoninae. Possible relationships of the new genus to other entedonine genera are discussed. Sorting of mass-collected material from Southeast Asia revealed a peculiar genus of Entedoninae (Eulophidae). This genus differs from other entedonines in having a strong ridge surrounding the frons on the front of the head, and a similar one sur- rounding the entire occipital region on the back of the head. In addition, it differs from most entedonines in having several pairs of setae on the scutellum, having a single dorsal seta on the submarginal vein, and lacking a frontal sulcus. We are describing this genus as part of an interest among the authors to, (1) de- scribe the eulophid fauna of Southeast Asia, and (2) provide a necessary frame- work for understanding variation within the Eulophidae, including exceptions to characters used for subfamily definition. A single diagnosis and description are offered for this new genus and new spe- cies. Without additional species it is im- possible to distinguish between species level and genus level characters. Acronyms used in the text are as fol- lows. Collections: ANIC, Australian Na- tional Insect Collection, CSIRO, Canberra, Australia; BMNH, The Natural History Museum, London, UK; MZB, Museum Zoologicum Bogoriense, Bogor, Indonesia; UCR, University of California, Riverside, California, USA; USNM, United Stated National Museum of Natural History, Washington, D.C., USA. Terminology: MV, marginal vein; OOL, oculo-ocellar length, the distance between the lateral ocellus and eye margin; POL, postero- ocellar length, the distance between the lateral ocelli; PMV, postmarginal vein; SMV, submarginal vein; and SV, stigmal vein. Ambocybe Ubaidillah and LaSalle, new genus (Figs. 1-7) Type species: Ambocybe petiolata Ubaidillah and LaSalle Diagnosis. — Head with a strong, invert- ed U-shaped ridge surrounding the frons (Fig. 4), and a similar one surrounding en- tire occipital region (Fig. 5). Scutellum with 5-6 pairs of setae (Fig. 6). Submar- ginal vein with single dorsal seta (Fig. 2). Forewing with speculum absent. Fore bas- Volume 9, Number 1, 2000 171 Figs. 1-3. Amboq/be petiolata, female. 1. Habitus. 2. Forewing. 3. Antenna. 172 Journal of Hymenoptera Research Figs. 4-7. Ambocybe petiolata, female: 4. Head, frontal view. 5. Head and mesosoma. 6. Scutellum and pro- podeum. 7. Propodeum and petiole. itarsus elongate, slender, about twice as long as second segment (Fig. 1). Pronotum small, not visible in dorsal view (Fig. 5). Propodeum with distinct plicae; with two subparallel median carinae; area between median carinae slightly depressed and sculptured; plicae converging along pos- terior margin of propodeum and ending at arcute adpetiolar carinae (Fig. 6). Petiole long and slender, reticulate dorsally (Fig. 7). Ambocybe petiolata Ubaidillah and LaSalle, new species (Figs. 1-7) Female. — Length of body 1.0-1.3 mm; length of forewing 0.75-0.95 mm. Head and body dark brown, lower face and me- tasoma paler. Antenna yellow-brown. Legs pale yellow except coxae yellow- brown. Sculpture on mesosoma reticulate, shiny. Volume 9, Number 1, 2000 173 Head (Figs. 4, 5) broad with shiny, weak coriaceous to imbricate sculpture and in- conspicuous pilosity. Compound eyes convex and large. Frons defined by an in- verted U-shaped ridge extending from an- terior ocellus along eye margin to lower face. Lower face produced medially, con- cealing clypeus, which is reduced and in- flected. Strong longitudinal median carina present between toruli. Back of head with strong ridge forming a large curve from vertex to gena and defining a large, smooth occiput. Temple narrow. Vertex shiny, although finely wrinkled, with fine transverse carina between lateral ocellus and compound eyes. POL/OOL = 1.6. Antenna (Fig. 3) with scape long and slen- der, about 5 times as long as wide, reach- ing above vertex. Pedicel short, slender, about 2 times as long as wide. Two trans- verse anelli. Funicle with three segments, all longer than wide. Club with two seg- ments. Mesosoma (Figs. 5-6) except propo- deum reticulate. Pronotum reduced and not visible in dorsal view. Mesoscutum transverse; midlobe with 2-3 pairs of se- tae; notaulus incomplete, but indicated posteriorly by wide shallow depression. Scutellum longer than broad, with 5-6 pairs of setae; slightly produced posteri- orly" so dorsellum concealed in dorsal view. Propodeum medially 0.40-0.50 as long as length of scutellum, with plicae and paired, subparallel median carinae; median carinae and plicae joined by trans- verse carina at posterior margin of scutel- lum. Legs long and slender; fore basitar- sus elongate, slender, about twice as long as second segment (Fig. 1). Forewing (Fig. 2) with SMV tapering at apex and joining parastigma anteriorly to base of parastig- ma. Speculum absent. SMV with 1 dorsal seta. PMV very reduced, almost absent. MV/SMV 1.5-1.75; MV/SV 4.45-5.4. Metasoma. Petiole (Fig. 7) unusually long, about three times as long as wide; broader distally; with 2 setae on each lat- eral margin; reticulate dorsally, with lon- gitudinal dorso-lateral carina. Gaster smooth, elongate elliptical. Ovipositor short, apex not visible in dorsal view. Hy- popygium extending 0.75-0.90 length of gaster. Male. — Unknown. Biology and host. — Unknown. Distribution. — Currently known from Peninsular Malaysia, Sulawesi (Indonesia) and Papua New Guinea. The somewhat disjunct distribution spans both sides of Wallace Line, and suggests that this spe- cies may be more widespread through Southeast Asia and Australasia. Material examined. — Holotype female: INDONESIA, Sulawesi, Dumoga Bone N.P., Toraut, 16-23.V.1985, Malaise trap, J.S. Noyes (MZB) (card mounted). 21 fe- male paratypes (all card mounted): IN- DONESIA: same data as holotype (5 fe- males MZB; 5 females BMNH; 3 females USNM; 3 females ANIC); same data as ho- lotype but v.1985 (2 females BMNH). MA- LAYSIA, Selangor, Serdang, UPM Cam- pus, 25.viii-3.lx.1992, Malaise Trap, J. LaSalle (1 female BMNH). PAPUA NEW GUINEA: Central Province, -45 km NW Port Moresby, 5km NW Brown River Bridge (Hiritano Hwy).29.xii.l985, G. Gordh, rainforest (1 female UCR); Central Province, 20 km SE Port Moresby, l.i.1986, G. Gordh, forest edge (1 female UCR). Etymology. — Ambocybe is formed from the Greek arnbon, for ridge or crest, and kybe for head. Gender feminine. The spe- cific name, petiolata, reflects the presence of a long petiole. Discussion. — Ambocybe is placed in the Entedoninae, although it differs from most other entedonines in several important characters. The Entedoninae is one of the best defined subfamilies of the Eulophidae (Boucek 1988, Schauff 1991). It is easily recognised by a variety of characters which include: scutellum with a single pair of setae (as opposed to two or more pairs); submarginal vein with two dorsal setae; mesoscutal midlobe with two pairs of setae; male scape with sensory pores re- 174 Journal of Hymenoptera Research stricted to the ventral edge; face with fron- tal sulcus distinctly separated from the an- terior ocellus; propodeum with subspira- cular tubercles; marginal vein relatively long; stigmal vein relatively short (Boucek 1988, Schauff 1991). The most important of these characters are the single pair of setae on the scutellum, the presence of only two strong setae on the submarginal vein, and the position of the frontal sulcus. None of these characters are present in Ambocybe, however all of these characters do show variation within the Entedoninae. A single pair of scutellar setae is one of the best characters for defining the Ente- doninae (Schauff 1991), and is found in al- most all members of the subfamily. Am- bocybe has 5-6 pairs of setae on the scutel- lum. There are a few other entedonines which have more than a single pair of se- tae on the scutellum. These include Para- Jwrismenus Girault (Boucek 1988), two spe- cies of Pediobius Walker (Kerrich 1973, Boucek 1977), and all members of the En- tedononecremnus genus group in the Eu- deromphalini (Entedononecremnus Girault, Aleuroctonus LaSalle and Schauff, Dasyom- phale LaSalle and Schauff; see LaSalle and Schauff 1994). Another important character for defin- ing the Entedoninae is the submarginal vein with two usually strong dorsal setae (Schauff 1991). Ambocybe only has a single seta. This character is also known in Myr- mokata Boucek (Boucek 1972) and two gen- era in the Euderomphalini, Pomphale Hu- sain, Rauf and Kudeshia and Baeoentedon Girault (LaSalle and Schauff 1994). The frontal sulcus in Entedoninae is generally distinctly separated from the an- terior ocellus (Schauff 1991). This sulcus is absent in Ambocybe. The frontal sulcus is absent from several other Entedoninae, in- cluding many species of Entedon Dalman and some Paracrias Ashmead (Schauff 1991), as well as members of the Entedon- onecremnus genus group (see above). The mesoscutal midlobe in Entedoninae typically has two pairs of setae (Schauff 1991), but this character is more homo- plastic than the preceding three charac- ters, and there are numerous exceptions. Ambocybe has 2-3 pairs of setae on the me- soscutal midlobe. Despite the numerous characters by which Ambocybe differs from other Ente- doninae, we still feel that it is best placed in this subfamily. Characters to support this are the incomplete notauli, the struc- ture of the carinae and plicae on the pro- podeum which are similar to those in Pe- diobius Walker, the presence of a small subspiracular propodeal tubercle, the rel- atively long marginal vein, and the short stigmal vein. The relationships of Ambocybe remain unknown. The presence of distinct plicae on the propodeum would suggest an af- filiation with other genera which also have this putatively synapomorphic character (such as Pediobius Walker, Pediobomyia Gi- rault, Rhynchentedon Girault, Apleurotropis Girault, Pleurotroppopsis Girault, ParaJwr- ismenus Girault, Zaommomentedon Girault, Schizocharis Kerrich, Platocharis Kerrich, Kratoysma Boucek, Horismenus Walker, Paracrias Ashmead). Within this group of genera, it could be closest to those genera which posses some form of paired median carinae on the propodeum, such as Pediob- ius, Pediobomyia, and Rhynchentedon. How- ever, it is not at all clear that Ambocybe is related to these genera, because all of the above genera possess a distinct transverse carina on the pronotum, and the prono- tum is distinct and clearly visible in dorsal view. Ambocybe lacks a transverse carina on the pronotum, and the pronotum is short and not visible in dorsal view. ACKNOWLEDGMENTS Wo thank the GEF-Indonesian Biodiversity and Collection Project tor funding KU and John Noyes (The Natural History Museum, London) tor provid- ing some specimens used in this study. Chris Burwell and Christer Hansson made many useful suggestions on the manuscript. Volume 9, Number 1, 2000 175 LITERATURE CITED Boucek, Z. 1972. Descriptions of new eulophid para- sites (Hym., Chalcidoidea) from Africa and the Canary Islands. Bulletin of Entomological Research 62: 199-205. Boucek, Z. 1977. Taxonomic studies on some Eulo- phidae (Hym.) of economic interest mainly from Africa. Entomophaga 21: 40H14. Boucek, Z. 1988. Australasian Chalcidoidea (Hymenop- tera). A biosystematic revision of genera of fourteen families, with a reclassification of species. CAB In- ternational, Wallingford. 832 pp. Kerrich, G. J. 1973. A revision of the tropical and sub- tropical species of the eulophid genus Pediobius Walker (Hymenoptera: Chalcidoidea). Bulletin of the British Museum (Natural History), Entomology 29: 115-200. LaSalle, J. and M. E. Schauff. 1994. Systematics of the tribe Euderomphalini (Hymenoptera: Eulophi- dae): parasitoids of whiteflies (Homoptera: Al- eyrodidae). Systematic Entomology 19: 235-258. Schauff, M. E. 1991. The Holarctic genera of Entedon- inae (Hymenoptera: Eulophidae). Contributions of the American Entomological Institute 26: 1-109. J. HYM. RES. Vol. 9(1), 2000, pp. 176-181 An Introduced Species of Epichrysocharis (Hymenoptera: Eulophidae) Producing Galls on Eucalyptus in California with Notes on the Described Species and Placement of the Genus M. E. SCHAUFF AND R. GARRISON (MES) Systematic Entomology Laboratory, USD A, PSI, Agricultural Research Service, c/o National Museum of Natural History, Washington, D.C. 20560-0168, USA; (RG) Department of Agricultural Commissioner, Weights and Measures, County of Los Angeles, 3400 La Madera Ave., El Monte, CA 91732, USA Abstract. — Epichrysocharis burzvelli Schauff, new species (Hymenoptera: Eulophidae) is de- scribed from specimens collected in southern California. Epichrysocharis burzvelli forms small blis- ter-like galls on the leaves of Eucalyptus citriodora. The previously described species of Epichryso- charis are reviewed and separated from E. burwelli. Evidence suggests that this species was acci- dentally introduced into the United States from Australia. In early 1999, specimens of a small chal- had been unknown, although it was stated cidoid wasp were submitted to the Sys- that they had been "associated with small tematic Entomology Laboratory, USDA, galls on Eucalypt leaves" (Boucek 1988). by the California Department of Food and The specimens recorded from California Agriculture (CDFA) for identification, were reared from small blister-like galls These tiny wasps were found emerging on the leaves of Eucalyptus citriodora and from galls on the leaves of Eucalyptus ci- unemerged specimens have been dissect- triodora in the Los Angeles area. Recently, ed from inside galls on the leaves. We can their occurrence has been noted in several find no evidence that E. burwelli is para- nurseries, and they are becoming wide- sitizing some other insect in or associated spread in the Los Angeles area. Study of with the galls. It can now be confirmed these specimens and subsequent rearings that this species is a gall former. The galls revealed them to belong to the genus Ep- themselves appear as small reddish or ichrysocharis (Hymenoptera: Eulophidae). brownish blisters on the surface of the leaf This genus was only known to occur in (Figs. 5, 6). The galls are expressed on both Australia and includes 3 species (Boucek surfaces of the leaves (that is, a single gall 1988). After study of the types of those produces a "blister" on both surfaces) and species it was determined that the speci- the wasps do not seem to have a prefer- mens reared in California represented a ence for one side or the other as emer- species unknown to science. We take this gence holes can be seen on both sides of a opportunity to describe this species and single leaf. The emergence holes are round present information on the known species and tend to be in the center of the gall of the genus to facilitate their identifica- (Fig. 6). The galls can be quite numerous tion. Given the known distribution of this and we have counted in excess of 40 in a genus, it is highly likely that this species single square centimeter of leaf surface, was accidentally introduced into Califor- Whether or not this gall formation causes nia from Australia. significant damage to the plant has not yet The biology of Epichrysocharis species been assessed. However, the appearance Volume 9, Number 1, 2000 177 Figs. 1-6. 1-4. Scanning electron micrographs: 1, Female antenna, dorsolateral view. 2, Female clava, dor- soapical view. 3, Mesosoma, dorsal view. 4, Propodeum. 5-6. Eucalyptus citriodora leaf: 5, Surface with galls. 6, Closeup of gall showing emergence hole. of these galls on nursery stock would most likely reduce the attractiveness of the plant and lower its market value. The wasps appear to be spreading and are now found in seven localities within Los Angeles County. During the editing of this paper an ad- ditional series of three specimens of an- other small tetrastichine was reared from Eucalyptus leaves in Santa Barbara, Cali- fornia. This species is readily distin- guished from Epichrysocharis burwelli, but because of the condition and limited num- ber of specimens I have been unable to de- finitively assign this species to a genus. Along with a third species of tetrastichine described by Headrick et al. (1995) from California, and introduced from Australia 178 Journal of Hymenoptera Research on Chamelaucium unciatum (Myrtaceae), it is apparent that phytophagous species are increasingly becoming established in the U.S. Terminology for morphology follows Gibson (1997) and LaSalle (1994). Acronyms for museums are: (BM) Bo- hart Museum, University of California, Davis, CA, USA; (CNC) Canadian Nation- al Collection, Ottawa, Ontario, Canada; (QM) Queensland Museum, Brisbane, Australia; (BMNH) The Natural History Museum, London, UK; (USNM) National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA. Epichrysocharis Girault Epichrysocharis Girault 1913b:36. Type species, Epichrysocharis fusca Girault. Original desig- nation. Brachychrysocharopsis Girault, 1922:103. Type species, Brachychrysocharopsis aligherini Gi- rault. By monotypy. Synonymy by Boucek (1988: 690). Diagnosis. — Length generally 0.5mm or less; antenna (Figs. 1, 2) with 2 anelli, 3 funicles, all quadrate or wider than long, Fl wider than F2 or F3, F2 smaller than F3, Fl closely appressed to F2 and par- tially extended over and covering F2 such that in some views F2 is hidden giving the impression that the funicle is 2-segment- ed, clava longer than funicle; clava 3-seg- mented, 3rd segment with suture oblique- ly angled from 2nd (Fig. 2); malar suture complete; pronotum very narrow medial- ly, overhung by anterior margin of mesos- cutum and not visible from above; mesos- cutum large, convex, without median lon- gitudinal groove, adnotaular setae restrict- ed to area near notaulus in one or two irregular lines and often with only 2-3 se- tae on each side; scutellum with subme- dian grooves and two pair of setae; pro- podeum very short medially, generally equal to or shorter than metanotum, with no median or subspiracular carinae; spi- racle round; metasoma sessile, broadly at- tached to mesosoma and with no obvious petiole, phragma generally projecting into gaster. Submarginal vein with single dor- sal seta, marginal vein subequal to sub- marginal, stigmal vein well developed and about 1/2 length of marginal. EpidwysocJiaris closely resembles a num- ber of small Australian tetrastichines which also have very short antennae but are currently placed in the large genus Aprostocetus. In addition, there are other genera generally considered closely relat- ed to Aprostocetus, which have shortened funicles (all 3 funiculars quadrate to slightly wider than long, about the same width, and funicle about equal in length to or slightly longer than clava [see, for example, A. (Epomphaloides) flavus (Boucek 1988, fig. 1191)]). While we have not stud- ied these species closely, and many are probably undescribed, one of the defining characters of Aprostocetus is that the raised lobe of the callus partially overhangs the outer rim of the propodeal spiracle (Gra- ham 1987; LaSalle 1994). We have not ob- served this condition in the species which we place in Epichrysocharis. Boucek (1988) synonymized Epentasti- chus Girault under Aprostocetus. The senior author has examined the type species, Epentastichus nugatorius Girault (QM). Un- fortunately, the body of the type has been lost and only two heads, mounted on a slide, remain. These heads are in poor con- dition, but the antennae of these two par- tial specimens are very similar to Epichry- socharis. However, given the fragmentary nature of the specimens, and in the ab- sence of more detailed revisionary work, we do not feel that enough evidence exists at this time to make any further nomen- clatural changes regarding the placement of £. nugatorius. Boucek (1988) also syn- onymized Epomphaloides with Aprostocetus. Several species of other genera from the Australian fauna have also been reared from or associated with galls on Eucalyp- tus. Although they were studied at the ge- neric level by Boucek (1988), the included species have not been critically revised. Volume 9, Number 1, 2000 179 What is apparent, is that this whole com- plex of generic names and species is much more diverse than indicated by the few species described at this time, and that much work on the taxonomy will need to be done to increase our understanding of the limits of this group of genera. Species of Epichn/socharis Epichn/socharis aligherini (Girault) Brachychrysocharopsis aligherini Girault 1922: 103. Transferred to Epichn/socharis by Boucek 1988. Diagnosis. — Mesosoma and metasoma black; forewing with apical fringe equal to about 1/3 wing width; ovipositor about 2/3 to 3/4 length of metasoma, slightly exserted past tip of gaster. This species is distinguished by the uniform dark color- ation and the elongate ovipositor which extends for well over half the length of the metasoma. Type. — The single type specimen of this species is slide mounted. It is badly crushed and disarticulated with the head, mesosoma, metasoma, legs, and antennae scattered over the slide. The head is frag- mented and the antennae mostly disartic- ulated and shriveled making it nearly im- possible to make out the relative sizes and numbers of segments. A single forewing remains attached to the mesosoma. The other wings cannot be found. The speci- men is not cleared, but a few characters can be discerned, and based on what can be seen, it appears that this species is ap- propriately placed in Epichrysocharis. There is no register number evident on the slide. Deposited in QM. Epichrysocharis fasca (Girault) Quadrastichus fusca Girault 1913b: 234. Trans- ferred to Epichrysocharis by Girault 1913b. Isotypic with Q. fusca. Diagnosis. — Body light brown or yellow, head yellow; antenna with Fl, F2, and F3 all wider than long and all subequal in length. Fl and F2 each with a single lon- gitudinal sensillum. Ovipositor about 1/3 length of metasoma. This species is dis- tinctive because it is the only one with a longitudinal sensillum on F2. Type. — The type (QM register no. Hy- 1847) of this species is a female, slide mounted with the body minus the head under one cover slip and the head under the second cover slip. The head is badly crushed, but one antenna is clearly visible. The body is not cleared. Deposited in QM. Epichrysocharis nigriventris (Girault) Epentastichus nigriventrus Girault 1913a: 242. Transferred to Epichn/socharis by Boucek 1988. Diagnosis. — Head yellow, mesosoma yellow orange, metasoma brown to dark brown, hind femur brown, hind tibia and tarsus yellow; mesoscutum and scutellum striate reticulate; mesocutum about 2x as long as scutellum, metasoma ovate and slightly shorter than mesosoma (30:25); antenna with Fl as long as wide, 3x as long as F2, clava about 4x as long as F2 and 2X as long as wide; only Fl and clava with multiporous plate sensillae; hypo- pygium reaching nearly to end of meta- soma. Types. — Two specimens (QM register no. Hy-1849) of this species are point mounted on a single pin. Girault noted that the species was described from two females, but did not designate a holotype. We hereby designate as lectotype the spec- imen nearest the point of the card mount and the other specimen as paralectotype. The lectotype is partially submerged in glue, but not as completely as the paralec- totype. The heads of both specimens have been slide mounted, and there is no way of associating any one head with one of the point mounted bodies although it seems that both specimens do belong to the same species. Deposited in QM. Epichrysocharis burwelli Schauff, new species (Figs. 1-4) Diagnosis. — Body brown, with face ver- tex, upper margin of eyes, and scrobes yel- 180 Journal of Hymenoptera Research low; Fl with longitudinal sensillum (Fig. 1); ovipositor about 1/2 length of meta- soma, with hypopygium reaching half the length; forewing fringe about 1/4 wing width. The distinctly brown and yellow head with a uniformly brown mesosoma and metasoma and relatively short ovi- positor set this species apart from the oth- ers in the genus. In addition, E. nigriventris has the hypopygium reaching almost to the end of the metasoma and £. aligherini has a longer ovipositor which is slightly exserted past the tip of the metasoma. Description. — Female. Length 0.5- 0.6mm. Color. Brown or light brown ex- cept following yellow: a small strip below each torulus, scrobes, vertex, dorsal occi- put, a small thin stripe bordering the eye from about the line of the toruli up and around the back of the eye for about half its length, antennae, and apices of tibiae and usually first three tarsomeres. Head: Face alutaceous to strigose, scobal basin nearly smooth, vertex and occiput rugose to finely alutaceous. Posterior ocel- li widely separated, anterior ocellus only about 1 diameter in front of posterior ocel- li and with POL about 5 X OOL. Mandible with 3 distinct teeth. Toruli inserted even with bottom of eye. Antenna (Figs. 1, 2) with scape 3X as long as wide, second anellus with 2 dorsal setae, Fl slightly lon- ger than wide (10:12) on ventral surface, as long as wide on dorsal surface, with single longitudinal sensillum, F2 2x as wide as long, loosely appressed to Fl and partially covered dorsally by Fl, F3 1.5X as wide as long, clava 2X as long as wide with several longitudinal sensillae some of which extend past tip of clava. Mesosoma: Pronotum reduced, not visi- ble medially from above. Mesoscutum, scutellum, dorsellum, and lateral propo- deum finely alutaceous or coriaceous (Fig. 3). Mesoscutal midlobe with 2 or 3 minute, inconspicuous setae at notaular margin. Dorsellum lightly alutaceous. Propodeum smooth to very lightly alutaceous laterally Fig. 4), with two minute setae laterad of spiracle. Forewing 2x as long as wide, with longest marginal fringe seta 1/4 width of wing. Ratio of costal cell : paras- tigma : submarginal vein : marginal vein : stigmal vein 20:15:25:28:15. Metasoma: Slightly shorter than meso- soma, ovate and slightly longer than wide, rounded posteriorly. First tergum nearly smooth medially, rest of mesosoma uni- formly, densely rugose reticulate. Ovipos- itor about 1/2 length of metasoma. Hy- popygium reaching about 1/2 length of metasoma. Male. — Similar to female. Types. — Holotype female on slide with data: California, Los Angeles Co., Monte- rey Park, Wilcox & 60 Fwy (s. side), 11 May, 1999. Reared Eucalyptus citriodora. Coll. D. Humphreys and M. Suim. Depos- ited in USNM. 52 female and 1 male par- atypes with same data. Paratypes depos- ited in BMNH, BM, CNC, QM. Etymology. — The species epithet honors Dr. C. Burwell, Queensland Museum without whose help this study could not have been completed. ACKNOWLEDGMENTS We are grateful to John Sorenson, California De- partment of Agriculture for sending the initial lot of specimens that led to this study. Chris Burwell, Queensland Museum, Brisbane, Australia for the loan of types and specimens. John LaSalle, CABI Biosci- ence, UK, F. C. Thompson, and D. R. Smith, System- atic Entomology Lab, USDA made valuable com- ments on the draft manuscript. Tami Carlow and Nit Malikul, Systematic Entomology Lab, ARS, USDA provided technical support. LITERATURE CITED Boucek, Z. 1988. Australasian Chalcidoidea (Hymenop tera): A biosystematic revision of genera of fourteen families, with a reclassification of speeies. CAB In- ternational. Wallingford, UK. 832pp. Gibson, G. A. P. 1997. Chapter 2, Morphology and terminology, pp. 16—44. In. Gibson, G. A. P. et al., eds. Annotated keys to the genera of North Amer- ican Chalcidoidea (Hymenoptera). NRC Research Press, Ottawa. Girault, A. A. 1913a. Australian Hymenoptera: Chal- cidoidea. IV. The family Eulophidae with do- Volume 9, Number 1, 2000 181 scriptions of new genera and species. Memoirs of phidae), with a revision of certain genera. Bulle- tin' Queenland Museum 2: 140-296. tin of the British Museum (Natural History). Into Girault, A. A. 1913b. A few new chalcidoid Hyme- mology Series 55: 1-392. noptera from Queensland, Australia. Bulletin of Headrick, D. H., J. LaSalle, and R. A. Redak. 1995. A the Wisconsin Natural History Society 11: 35-48. new genus of Australian Tetrastichinae (Hyme- Girault, A. A. 1922. New chalcid flies from eastern noptera: Eulophidae): an introduced pest of G?- Australia — II. (Hymenoptera: Chalcididae). Inse- melaucium unciatum (Mytaceae) in California. cutor Inscitiae Menstruus 10: 100-108. Journal of Natural History 29: 1029-1036. Graham, M. W. R. de v. 1987. A reclassification of the LaSalle, J. 1994. North American genera of Tetrasti- European Tetrastichinae (Hymenoptera: Eulo- chinae. journal of Natural History 28: 109-236. J. HYM. RES. Vol. 9(1), 2000, pp. 182-208 A Revision of the Panurgine Bee Genus Arhysosage (Hymenoptera: Andrenidae) Michael S. Engel Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024-5192, USA Abstract. — The South American bee genus Arhysosage Brethes (Panurginae: Calliopsini) is re- vised. In addition to the previously recognized Arhysosage flava Moure, A. cactorum Moure, A. ochracea (Friese), and A. bifasciata (Friese) (new combination), two new species are described: A. atrolnnata Engel and A. zamicra Engel. Arhysosage germana Moure is newly synonymized with A. ochracea and A. melanothricha Moure is synonymized with A. cactorum, while A. bifasciata is resurrected from synonymy under A. ochracea with Camptopoeum opuntiarum Torgensen as a junior subjective synonym (new synonymies). The genus is newly diagnosed and a key to the currently recognized species is presented. The name appearing in the literature as A. xanihina is a nomen nudum. The phylogenetic position of the genus among other calliopsine bees is briefly summarized as are relationships among the species. A cladistic analysis of Arhysosage produces a single tree with the following hierarchy: A. cactorum (A. flava, A. zamicra (A. ochracea (A. atrolunata, A. bifas- ciata))). Biological information on Arhysosage is summarized. The genus is presently known from Argentina, southern Brazil, Bolivia, and Paraguay and is apparently a specialist on Cactaceae (presently recorded from Echinocactus, Echinopsis, Gymnocalycium, Notocactus, Opuntia, and Tricho- cereus). Bees of the genus Arliysosage Brethes (1922) are among the most distinctive of the South American panurgines, being characterized by the fairly robust size of most individuals, large heads of males, and mostly yellow body coloration (Fig. 1). Individuals can be readily captured at flowers of various cactus genera (e.g., Gymnocalycium, Notocactus, Opuntia, &c.) upon which they are presumably oligolec- tic (Schlindwein 1992, Schlindwein and Wittmann 1995). Outside of their affinity for cactus flowers, however, the biology of Arhysosage species has not been the focus of any published study, although Schlind- wein and Wittmann (1995) give a few de- tails of mating behavior in Arhysosage. Their observations indicate that mating is initiated at cactus flowers. Males search flowers for females, sometimes staying motionless in inflorescences for up to sev- en minutes. Once a female appears, the male grabs her waist with his long man- dibles and initiates copulation. The couple frequently continues mating during flight and may visit several flowers throughout the encounter, with the female continuing to forage the whole time. Such mating be- havior is reminiscent in some respects to that described for Perdita (Macrotera) tex- ana (Cresson 1878). Like Arhysosage, this group is oligolectic on Cactaceae (Snelling and Danforth 1992, Neff and Danforth 1992) and also demonstrates a dramatic head-size polymorphism in males that, among other uses, allows males to grasp females during copulation (Danforth and Neff 1992). Future field work on Arhyso- sage species should explore possible etho- logical-morphological associations in males as has been done for Perdita. The genus was established by Brethes (1922) for an enigmatic bee species in northern Argentina, but its systematic po- Volume 9, Number 1, 2000 183 Fig. 1. Dorsal habitus of male Arhysosage ochracea (Friese). sitiori in the Panurginae remained unrec- ognized until the late 1950s when Moure and Michener were able to examine the type series (Moure 1958). Brethes was un- aware that Friese (1908) had already de- scribed the species in Camptopoeum Spi- nola (1843). Cockerell in 1940 described the species for a third time. Timberlake (1952a) was the first to place the genus in the Calliopsini but under the name Riti- ziella as he was unaware of, or not able to interpret, Brethes' work. Moure (1958) cor- rectly established the name of the genus, described two new species, and briefly discussed its placement in the Panurginae, although he disagreed with Timberlake's assignment of the genus to Calliopsini. Following a phylogenetic study of the Panurginae based on the external mor- phology of adults, Ruz (1986, 1991) placed Arfn/sosage in the tribe Calliopsini together with four other genera. Arln/sosage was placed as the sister to a clade consisting of Callonychium Brethes (1922) and Spinoliella Ashmead (1899) (Fig. 2). The three genera were grouped on the shared presence of yellow metasomal markings, the weak or absent transverse ridge on the male la- brum, absence of an inflection at the fe- male labral apex, position of the male an- tenna] sockets on the lower third or fourth of the face, broken pattern of keirotrichia on the inner surface of the female metati- bia, and absence of a volsella. Arhysosage was excluded from the Callonychium + Spinoliella clade by the primitive presence 184 Journal of Hymenoftera Research ■PERDITINI i — PROTOMELITURGINI Acamptopoeum Calliopsis c c I — Arhysosage Spinoliella Callonychium Fig. 2. Phylogeny of tribe Calliopsini (after Ruz 1991) indicating the position of Arhysosage and relat- ed genera. Perditini and Protomeliturgini are out- group tribes. of a slightly convex lower paraocular area, presence of keirotrichia on most of the in- ner surface of the male metatibia, pattern- ing of the keirotrichia on the inner surface of the female metatibia, composition of the metatibial scopa of only moderately abun- dant hairs, absence of a lateral ridge on the female S6, gonostylus being one-half to one-third the length of the gonocoxa, and absence of internal sclerotization in the aedeagus. While Arhysosage is known only in Argentina, southern-most Brazil, Paraguay, and southeastern Bolivia, both Callonychium and Spinoliella are somewhat more widely distributed. Callonychium oc- curs in Argentina, Brazil, Chile, Paraguay, and Peru (Ruz 1991, Toro and Herrera 1980) while Spinoliella is distributed in Chile (Ruz 1991, Toro 1995, Toro and Ruz 1972), Argentina, and Peru (Engel unpubl. data). Given the present collection locali- ties for Arhysosage it seems likely that the genus will someday be discovered in Uru- guay. Herein I present a revision of Arhysosage including a generic description modified from that provided by Ruz (1991), incor- porating the changes that result from ad- dition of new species to the genus. A key to calliopsine genera distinguishing Arhy- sosage has been presented by Ruz (1991) and Michener (in press). MATERIALS AND METHODS Morphological terminology generally follows Michener (1944) with additions for mandibular structure provided by Miche- ner and Fraser (1978); also, I use anal vein in place of vannal. The abbreviations F, S, and T are employed for flagellomere, me- tasomal sternum, and metasomal tergum, respectively. The most common species, A. ochracea, is described in detail and all other descriptions are referenced to this one so as to avoid repetition. A total of 479 specimens (215 9 9, 264 6 8 ) were examined during the course of this study. Specimens were provided by the following institutions: AMNH, Amer- ican Museum of Natural History, New York, New York, J. G. Rozen, Jr., and M. G. Rightmyer; BMNH, The Natural His- tory Museum (British Museum), London, United Kingdom, G. Else and C. Taylor; CAS, California Academy of Sciences, San Francisco, California, W. J. Pulawski and R. L. Zuparko; CTMI, Central Texas Mel- ittological Institute, Austin, Texas, J. L. Neff; CUIC, Cornell University Insect Col- lection, Ithaca, New York, J. K. Liebherr and E. R. Hoebeke; LACM, Natural His- tory Museum of Los Angeles County, Los Angeles, California, R. R. Snelling; MACN, Museo Argentino de Ciencias Na- turales, Buenos Aires, Argentina, A. Roig- Alsina; MLPA, Museo de La Plata, Univ- ersidad Nacional de La Plata, La Plata, Ar- gentina, J. A. Schnack; NHRS, Naturhis- toriska Riksmuseet, Stockholm, Sweden, T. Pape; PCIA, Personal Collection of Isa- bel Alves dos Santos, Sao Paulo, Brazil; SEMC, Snow Entomological Collection, Natural History Museum, University of Kansas, Lawrence, Kansas, R. W. Brooks and C. D. Michener; USNM, United States National Museum of Natural History, Smithsonian Institution, Washington, D.C., R. J. McGinley, M. Mello, and D. G. Furth; ZMHB, Zoologisches Museum an der Humboldt-Universitat, Berlin, Ger- many, F. Koch. Volume 9, Number 1, 2000 185 Table 1. Character matrix and character descrip- tions used in cladistic analysis of the internal phytog- eny of Arhysosage. Callonychium and Spinoliella are used as outgroups. Refer to Material and Methods for details of analysis. 1. Inner hind tibial spur: (0) straight (1) curved. 2. Compound eyes below: (0) parallel, (1) diverging. 3. Metasoma: (0) much broader than head, (1) as broad as head. 4. Male clypeal apex: (0) straight between lateral corners, (1) gently convex between lateral corners. 5. Male pygidial plate emar- gination: (0) absent, (1) present. 6. Male 57 lateral pro- cesses: (0) broad, expanded towards base, (1) narrow, not expanding towards base. 7. Aedeagus: (0) shorter than penis valves, (1) as long as or longer than penis valves. Taxa Character A. atrolunata A. bifasciata A. flava A. ochracea A. cactorum A. zamicra Spinoliella Calloin/chnun 1234567 1111111 1111111 1110101 1111101 1110000 1110101 0000000 0000710 A cladistic analysis of Arhysosage spe- cies was undertaken. Species of Callony- chium and Spinoliella were used as out- groups. Seven characters were identified and coded for the taxa employed. The sin- gle interrogative mark presented in the data matrix (Table 1) is a polymorphism for this character in Callonychium and not missing information. Character descrip- tions are given with the data matrix (Table 1 ). The data matrix was constructed in DADA (Nixon 1995) and analyzed using the ie* command in HENNIG86 (Farris 1988). Trees were visualized and printed using CLADOS (Nixon 1993). SYSTEMATICS Genus Arhysosage Brethes Arhysosage Brethes 1922: 121. Type species: Ar- hysosage johnsoni Brethes 1922 (= Campto- poeum ochraceum Friese 1908), monobasic. Ruz 1991: 238. Ruiziella Timberlake 1952a: 105. Type species: Camptopoeum ochraceum Friese 1908, original designation. Preoccupied ()iec Cortes 1951). Ruziapis Timberlake 1952b: 528. Type species: Camptopoeum ochraceum Friese 1908, autobas- ic. Replacement name for Ruiziella Timber- lake 1952a (nee Cortes 1951) and lapsus calami for Ruizapis Timberlake 1953 (justified emen- dation). Ruizapis Timberlake 1953: 598. Type species: Camptopoeum ochraceum Friese 1908, autobas- ic. Justified emendation of Ruziapis Timber- lake 1952b (lapsus calami). Diagnosis. — Arhysosage differs from oth- er South American panurgines by the ven- tral divergence of the compound eyes (Figs. 10, 12, 14-15, 18, 20, 22, 24), the broad heads (same figures as just men- tioned), the curved inner metatibial spur (Fig. 6), and the mostly yellow body col- oration (e.g., Fig. 1). Description. — Head broader than long in frontal view, broader than thorax (Fig. 1). Glossa longer than prementum, slender; paraglossa shorter than suspensorium; first segment of labial palp less than twice as long as combined lengths of segments 2-4. Maxillary blade longer than prepalpal part of galea; galeal comb absent. Labrum less than twice as broad as long, partially or entirely setose. Apical margin of clyp- eus with variably developed projection just outside lateral labral margin. Episto- mal sulcus forming an obtuse angle (Figs. 9-10). Inner subantennal suture angulate (Figs. 9-10); subantennal area wider than length of inner suture and than antennal socket; anterior tentorial pit near middle of outer subantennal suture (Fig. 16). An- tennal socket far below middle of face (Fig. 15). Lower median paraocular area slightly convex. Facial fovea strongly im- pressed, narrow (Fig. 10). Median ocellus set below upper tangent of compound eyes. Vertex convex. Mesepisternum with flattened anterior-facing surface reduced; preepisternal groove distinguishable only above scrobal level, continued downward as black line (difficult to see on dark in- tegument). Forewing with cu-a as long as or longer than second abscissa of M + Cu; lm-cu well distad lr-m; 2m-cu basad 2r- 186 Journal of Hymenoptera Research m; pterostigma longer than and slightly wider than prestigma, border within mar- ginal cell straight; apex of marginal cell obliquely and broadly truncate, longer than distance from apex to wing tip; first submarginal cell as long as or longer than second submarginal cell (Fig. 1). Hind wing with cu-a slightly less than one-half to one-third as long as second abscissa M + Cu; 10 distal hamuli arranged in a sin- gle series. Protarsomeres 2-A unmodified; malus of antenna cleaner pectinate. Me- sotibial spur half as long as or longer than mesobasitarsus, apex distinctly curved, finely serrate. Mesobasitarsus about as long as probasitarsus and shorter than metabasitarsus, tarsomeres 2-A unmodi- fied. Metatibial spurs curved at apices, outer spur as long as or longer than inner spur; teeth small. Metatarsus unmodified. Claws deeply cleft. Basal area of propo- deum slightly longer than metanotum, with exceedingly fine striae along apical margin, depressed medially. Metasoma with yellow bands very extensive to in- complete or absent; terga without setal bands; lateral fovea of T2 slightly de- pressed. Pubescence generally short and sparse; appressed hairs on most of dor- sum of mesosoma and metasoma. Punc- tures generally very fine and dense, nearly contiguous on mesoscutum. Male: Most of head and thorax yellow (Figs. 1, 10, 12, 14, 18, 22, 24). Labrum flat, with slight trans- verse ridge. Mandible arcuate, longer than compound eye (except in A. zamicra), up- per margin with prebasal projection (Fig. 18: arrow) and subapical tooth (Fig. 10). Length of clypeus more than four times width, gently protuberant. Flagellum un- modified, much shorter than head; Fl as long as or slightly longer than F2, about as long as broad. Inner orbits of com- pound eyes markedly divergent below. Pronotum with dorsal preapical ridge rounded, strong. Length of probasitarsus five times width. Metatibia with keirotri- chia on most of inner surface but sparser ventrally. Basitibial plate with borders well-defined. Rami of claws subequal in length. Metasoma wider than thorax (Fig. 1); T2-5 with gradulus posterolaterally long (surpassing middle of each tergum), carinate, and with postgradular depres- sion narrow, shallow; posterior marginal areas of Tl-5 minutely setose; pygidial plate well developed, abruptly elevated and carinate laterally towards apex; hem- itergum hexagonal; S4-5 with apical mar- gins slightly and broadly concave medi- ally; S6 distally bilobed, with small me- dian V-shaped emargination; S7 with two short, finger-like apicolateral projections, proximal arms long and forming a U (Figs. 26-31); S8 with long, clavate apico- median projection, abruptly separated from basal part which has weak median ridge dorsally (Figs. 32-37). Gonocoxal apodeme not inflexed; gonocoxa short, squared, completely fused both dorsally and ventrally; gonostylus short, finger- like, fused to gonocoxa; volsella apparent- ly absent or indistinguishably fused to gonocoxa; penis valve long, tapered to- ward apex, dorsally fused together by small, narrow bridge; aedeagus proximal- ly wider and fused to valve, distal half well sclerotized ventrally (Figs. 38-45). Fe- male: Yellow areas of head variable in size. Labrum nearly flat. Length of clypeus three times width, distinctly protuberant. Inner orbits of compound eyes only slight- ly divergent below. Pronotum with dorsal preapical ridge rounded, weak. Length of probasitarsus four times width. Metatibia longer than metabasitarsus; inner surface with keirotrichia in patch at base and apex, sparse or absent toward dorsal mar- gin, absent ventrally. Metatibial scopa of moderately dense and apparently simple, but minutely branched setae. Inner ramus of claws shorter than outer. Posterior mar- ginal areas of Tl-4 minutely setose; T7 not expanded dorsally but with conspicuous ventral proximal projection; Sl-5 minutely setose, as in male, but setae somewhat longer and denser; S6 with basal spine- like sclerotization, lateral margin with Volume 9, Number 1, 2000 187 \Spinoliella Callonychium cactorum zamicra ochracea m—m atrolunata bifasciata Fig. 3. Phytogeny of Arhysosage species (Length 8, C.I. 87, R.I. 90). Callonychium and Spinoliella are outgroup genera. Black bars indicate unreversed forward transitions while white bars indicate homoplastic character^ strong curved ridge, apical margin con- cave medially, apically with a well-de- fined, curved and dense band of curved setae. Sting short, not reaching stylus apex. Biological notes. — Species of Arhysosage are oligolectic on Cactaceae. The long, curved mandibles of males seem to be a modification for grasping the female dur- ing mating, while the mostly yellow body coloration appears to be an adaptation for minimizing visibility in flowers, which are generally yellow or off-white. Immature stages of A. flava have been discovered by Jerome G. Rozen, Jr. (AMNH) and will be treated in a forthcoming paper concerning the immature stages of Calliopsini (Rozen and Engel in prep.). Phytogeny. — The result of a cladistic analysis for Arhysosage (see Material and Methods) is presented in figure 3. A single tree resulted from analysis of the data ma- trix (Table 1); the topology had a length of 8, a C.I. of 87, and an R.I. of 90. This anal- ysis places A. cactorum as the sister to the remainder of Arhysosage. Two species, A. flava and A. zamicra, were unresolved in a polytomy (Fig. 3). These species are ex- ceedingly similar with A. zamicra possess- ing a number of autapomorphic features which allow for its recognition but fail to confidently group it with any other spe- cies of Arhysosage [i.e., whether sister to the remainder of Arhysosage (exclusive of A. cactorum), to A. flava, or to Arhysosage exclusive of A. flava and A. cactorum]. Arhysosage ochracea (Friese) (Figs. 1, 5, 8, 17-21, 26, 32, 38, 49) Camptopoeum ochraceum Friese 1908: 29. Exam- ined (ZMHB). Arhysosage johnsoni Brethes 1922: 122. Camptopoeum castellam Cockerell 1940: 1. Ex- amined (AMNH). Ruiziella ochracea (Friese); Timberlake 1952a: 105. Ruiziella castellani (Cockerell); Timberlake 1952a: 105. Arhysosage ochracea (Friese); Moure 1958: 44. Arhysosage genitalia Moure 1958: 47. New syn- onymy. Examined (SEMC). Diagnosis. — The species can be most readily separated from other Arhysosage species by the strong banding of the me- tasoma (Fig. 1). Description. — Male: Total body length 8.0-12.6 mm; forewing length 5.8-7.2 mm. Head width 2.3^.0 mm, length 1.6-3.0 188 Journal of Hymenoptera Research 8 Figs. 4-8. Characteristics of Arhysosage. 4, Pygidial plate of male 17 of Arhysosage cactorum Moure. 5, Pygidial plate of male 17 of A. ochracea (Friese). 6, Inner hind tibial spur of A. cactorum. 7, Basitibial plate of A. cactorum. 8, Basitibial plate of A. ochracea. mm. Mandible longer than compound eye; inner tooth well-defined, not partic- ularly strong, somewhat rounded (Figs. 18, 20). Upper interorbital distance 1.3-2.3 mm, lower interorbital distance 1.5-3.1 mm. Intertegular distance 1.4-2.5 mm. Basitibial plate apex broadly rounded (Fig. 8). Apex of pygidial plate slightly emar- ginate (Fig. 5). Apex of penis valve bend- ing ventrally (Figs. 45-49); aedeagus reaching, or very near to, apex of penis valve (Figs. 38^42, 45-49); terminalia oth- erwise as depicted in figures 26, 32, 38, and 49. Mandible mostly smooth with a few faint punctures on dorsal surface running from base in a narrow band to point of inner tooth; a few faint punctures in outer interspace, disappearing by point where outer ridge and condylar ridge meet; ven- tral surface with similar punctures as those of dorsal surface and disappearing by about the same point. Clypeus with faint, coarse punctures scattered over sur- face, integument otherwise smooth. Su- bantennal areas smooth and impunctate. Supraclypeal area below antennal sockets and between inner subantennal sutures as on clypeus; between antennal sockets punctures well-defined, smaller, and near- ly contiguous. Scape punctured as on su- praclypeal area between antennal sockets. Face lateral to outer subantennal sutures and below level of antennal sockets as on clypeus; at level of antennal sockets punc- tures become smaller, well-defined, and gradually more closely spaced until nearly contiguous by level just above antennal Volume 9, Number 1, 2000 189 sockets; remainder of face and vertex with such fine, well-defined punctures, nearly contiguous. Gena as on vertex except punctures becoming faint on lower half and separated by 1-2 times puncture width, integument otherwise smooth, punctures also become fainter posteriorly near preoccipital area. Postgena impunc- tate and smooth. Pronotum with minute, well-defined punctures on dorsal surface along border with mesoscutum, medially and anteriorly on collar integument im- punctate and imbricate; lateral surfaces smooth and impunctate except on prono- tal lobe which has a few minute punc- tures. Mesoscutum and scutellum with small, well-defined, nearly contiguous punctures. Tegula similar to mesoscutum except punctures quite faint. Metanotum with scattered faint, coarse, punctures, in- tegument between faintly imbricate. Pree- pisternal area as on mesoscutum except punctures becoming exceedingly faint and more widely spaced ventrally; mesepister- num with faint, coarse punctures separat- ed by less than puncture width, integu- ment between smooth, punctures becom- ing more widely spaced along posterior border and fainter ventrally; metepister- num with faint, minute punctures sepa- rated by width or less, integument be- tween smooth. Basitibial plate with mi- nute punctures separated by less than a puncture width. Propodeal lateral surface with minute, well-defined punctures sep- arated by 1-3 times puncture width, integ- ument between smooth; posterior surface with minute punctures separated by width or less, integument imbricate. An- terior surface of Tl faintly imbricate, re- mainder of surface minutely punctured, punctures nearly contiguous except apical margin imbricate and impunctate; T2-6 minutely punctured, punctures nearly contiguous except apical margins imbri- cate and impunctate; T7 imbricate; sterna imbricate with scattered, faint punctures. Head mostly yellow except facial fo- veae black and two spots on clypeus dark brown to black. Inner tooth, subapical tooth, and mandibular apex reddish brown to black. Proboscis light brown; hypostomal fossa as well as bordering ar- eas of postgena and preoccipital area dark brown to black. Labrum yellow. Fla- gellum light brown. Pronotum yellow. Mesoscutum, tegula, scutellum, and me- tanotum yellow. Preepisternum, mesepis- ternum, and metepisternum yellow ex- cept on ventral-facing surface dark brown to black; propleuron dark brown to black except posterolateral corner yellow. Inner halves of procoxa dark brown to black, remainder yellow; protrochanter yellow except ventral border brown; profemur yellow with brown spot on inner and ventral surface at base; remainder of fore- leg yellow; inner halves of mesocoxa dark brown to black; ventral and inner borders of mesotrochanter and mesofemur brown; ventral border of mesotibia light brown; remainder of midleg yellow; me- tacoxa and metatrochanter mostly dark brown or black except yellow on dorsal borders; metafemur with inner and outer borders brown, remainder yellow; inner border of metatibia and metabasitarsus light brown, remainder yellow; claws reddish brown at apices; mesotibial spine and metatibial spurs amber. Wing mem- brane hyaline; veins amber to dark brown. Propodeum yellow except basally bordering metanotum dark brown to black with mediolongitudinal, narrow line of black running from the basal area onto the posterior surface and ending me- dially at marginal area of propodeum. Mediolongitudinal line of Tl anterior sur- face amber, remainder yellow except api- cal margin amber; T2-6 yellow except graduli, areas lateral to graduli, and api- cal margins amber, although yellow areas on central disc become gradually and progressively narrower on T3-6 until mostly obscured on T6 by overhang of preceding tergum; 17 amber; sterna am- ber with dark brown on central discs ex- cept medial amber interruptions on S3-5. 190 Journal of Hymenoptera Research Pubescence generally sparse, golden, Face colored as in figures 17 and 19. moderately long, and simple. Particularly Gena yellow. Proboscis light brown; hy- dense areas of long setae along apicolater- postomal fossa, postgena, and preoccipital al margins of clypeus, just above and lat- area dark brown to black. Labrum brown, eral to antennal sockets, on postgena, and Antenna light brown. Pronotum black ex- on ventral borders of preoccipital area, cept pronotal lobe and posterior median Pronotum generally without hairs except border yellow. Mesoscutum black except at pronotal lobe; metanotum with mat of two longitudinal stripes and border with shorter, more dense hairs intermixed with tegula yellow. Scutellum yellow except moderately long, spare hairs. Mesotibia anterior border and median transverse and tarsus with short, stiff, amber setae band black, small longitudinal median along outer borders; metafemur with sim- black band connecting these two black ar- ilar setae on outer apex; metatibia and tar- eas; axilla black. Tegula and metanotum sus with longer, stiff, amber setae on outer yellow. Pleura black except metepister- surfaces. Terga with sparse hairs except num yellow. Propodeum as in male except lateral to pygidial plate where they are lateral surface bordering metepisternum long, dense, frequently branched, and am- dark brown to black. ber to fuscous; sterna similar except patch- Pubescence generally sparse, golden, es of long, amber to fuscous hairs on ei- moderately long, and simple. Particularly ther side of apical cleft of S6. dense areas of long setae along apicolater- Female: As described for the male ex- al margins of clypeus, just above and lat- cept as follows: Total body length 7.0-10.9 eral to antennal sockets, on postgena, and mm; forewing length 3.9-6.3 mm. Head at base of stipes. Terga with sparse hairs width 1.9-2.9 mm, length 1.5-2.5 mm. Up- except apex of T5 and lateral to pygidial per interorbital distance 1.2-1.9 mm, low- plate with long, dense, frequently er interorbital distance 1.3-2.2 mm. Inter- branched, amber hairs; sterna similar ex- tegular distance 1.3-2.0 mm. Pygidial cept borders of long, amber hairs on apical plate in profile straight or gently curved sterna. ventrally towards apex; dorsally gently ta- Variation.— Areas described in the male pering to narrowly rounded apex. as dark brown to black can vary to light Clypeus with faint, coarse punctures brown or even yellow (except facial foveae scattered over surface, most faint central- and clypeal spots). Similarly the areas de- ly, integument otherwise smooth. Supra- scribed as black in the female can some- clypeal area as on clypeus. Scape punc- times be lighter and appear as dark tured as on upper half of face. Face out- brown. The relative widths of the yellow side of outer subantennal sutures and be- areas on tne terga can vary dramatically low level of antennal sockets as on clypeus as well Tnere is, however, always some although punctures slightly smaller and yellow banding present on Tl-3. In fe- more faint; at level of antennal sockets maleS/ color variation is more dramatic as punctures becoming smaller, well-defined, is demonstrated by the facial patterns de- and gradually more closely spaced until picted in figures 17 and 19 and by the fact separated by puncture width or less by that the amber bands of the metasoma can level just above antennal sockets; remain- be quite broad and variable from light der of face and vertex with such fine, well- reddish brown to nearly black, defined punctures. Gena with scattered Holotype.— ARGENTINA: Mendoza: 9, faint punctures, integument otherwise 24 November 1906, Jensen (ZMHB). smooth, punctures become exceedingly Additional material— ARGENTINA: Catamarca: faint near preoccipital area and postgena. joyango-Colpes Site, int. Biol. Program, 24 October 197?, Tegula imbricate. J. L. NTeff, on Opuntia sulphurea (l 9 Id CAS). Andalgala, Volume 9, Number 1, 2000 191 IBP Program, Desert Scrub Site, ]. L. Neff, on Opimtia sulphured, various dates: 20, 24, 31 October 1972, 27, 31 January 1973 (49 936 6CAS). Andalgala Desert Site, IBP, J. L. Neff (193, Labrum and lower half of face; note the sharply curved inner subantennal suture and the position ot the anterior tentorial pit nearly at the midpoint of the outer subantennal suture. 198 Journal of Hymenoptera Research Pubescence along apicolateral margins of clypeus sparse and simple. Terga with sparse hairs except lateral to pygidial plate where they are long, dense, frequently branched, and amber; sterna similar with scattered short, amber hairs. Variation. — The areas of dark brown to black on the venter of the female can sometimes be light brown or with various regions being entirely yellow. Similarly, Tl in the female can sometimes have a small brown spot centrally by the bend separating the anterior-facing and dorsal- facing surfaces. On the face, females some- times have small black patches at the up- per border of the compound eyes. These patches can sometimes connect the black crescent of the upper face with the com- pound eye margins. Holotype.— ARGENTINA: Formosa: S, Ing. Juarez, December 1950, F. H. Walz (SEMC). Additional material. — ARGENTINA: Cata- marca: Recreo, December 1951, F. H. Walz (19USNM, lcJAMNH, 193ddSEMC). Cordoba: Jesus Maria, 3 December 1973, J. L. Neff, on Opun- tia sp. (1 96c? cJCTMI). Arguello, J. A. de Carlo and M. J. Viana (1 9 SEMC). Formosa: Ing. G. N. Juarez, 30 November 1949, F. Monros (191<5MACN). Ing. Juarez, December 1950, F. H. Walz (1<5AMNH, 1<3BMNH, l93dc?SEMC). Gran Guardia, 15 No- vember 1952, J. Foerster (191.5SEMC). San Luis: A. Stevenin (3d JMACN). Santiago del Estero: Rio Salado, Wagner (192ddMACN). [no date or col- lector's name] (29 93c? dMACN). El Pinto, Novem- ber 1956 (259 925 6 6 SEMC). Choya, January 1958 (49 9 176 cJSEMC). M. Gomez (1 9 1 <3SEMC). Dpto. Matara, Desvio 511, 24 October 1928, M. Gomez {26 6SEMC). Loreto, December 1992, M. Fritz (19AMNH). Salta: Cruz Quemada, 40 km S Gen- eral Guemes, 20 November 1989, J. G. Rozen and A. Roig-Alsina (29 921 6 6 AMNH). Same as pre- vious collection data except in copula on flowers of Opuntia sp. (19ldAMNH). Same as previous [9 don same pin] (191 6" AMNH). Cruz Quemada, 31 km S General Guemes, 10 November 1989, J. G. Rozen and A. Roig-Alsina, on Opuntia sp. (39 97ddAMNH). Cruz Quemada, 9 November 1993, J. G. and B. L. Rozen, on Opuntia sp. (16* AMNH). 20 km W-NW Hickmann, 12-14 No- vember 1989, J. G. Rozen and A. Roig-Alsina (1 92b" cJ AMNH). BOLIVIA: Santa Cruz: San Isidro (16* AMNH). PARAGUAY: Chaco, Loma Plata, Ar- riagado, February 1993 (lei AMNH). Floral records. — This species has at pres- ent been found only on flowers of an un- identified Opuntia (new record). Phenology. — Arhysosage flava has been captured from early November into early February. Comments. — This species resembles to some degree A. ochracea but differs most notably in the absence of metasomal bands, the elongate punctures of the clyp- eus (Figs. 15-16), the absence of dense se- tae on the lateral borders of the clypeus, and the patterning of black marks on the female face (Fig. 13). Arhysosage flava is most similar to the poorly known A. zam- icra but differs in the male mandible being longer than the compound eye (Fig. 14), the absence of black markings on the me- sepisterna, the presence of fine striae on the mandibular outer interspace (Fig. 15), and the broadly rounded basitibial plate apex (Fig. 8). The above specimens of this species in Bolivia and Paraguay are the first records for this genus in both countries. Arhysosage zamicra Engel, new species (Figs. 22, 27, 33, 39, 46) Diagnosis. — This is presently is distin- guished by the combination of the absence of banding on the metasoma, the elongate punctures of the clypeus, the mandible be- ing slightly shorter in length than the length of the compound eye, the absence of mandibular striae, the presence of ven- tral-facing black spots on the mesepister- na, and the pointed apex of the basitibial plate (Fig. 7). Description. — As for A. ochracea (see above) with the following modifications: Male: Total body length 6.8 mm; forewing length 4.2 mm. Head width 2 mm, length 1.3 mm. Mandible shorter than compound eye; inner tooth strong and pointed (Fig. 22). Upper interorbital distance 1.2 mm, lower interorbital distance 1.4 mm. Inter- tegular distance 1.3 mm. Basitibial plate apex pointed (similar to that depicted for Volume 9, Number 1, 2000 199 A. cactorum: Fig. 7). Terminalia as depicted in figures 27, 33, 39, and 46. Outer interspace of mandible with faint, coarse punctures, integument between faintly imbricate, without striae. Clypeus with faint, coarse punctures, punctures nearly contiguous and longitudinally ex- tended making surface appear roughened (as in A. flava). Supraclypeal area with coarse, nearly contiguous punctures. Face outside of outer subantennal sutures and below level of antennal sockets with punc- tures separated by 1-2 times puncture width, integument otherwise smooth; at level of antennal sockets punctures be- coming smaller, well-defined, and gradu- ally more closely spaced until nearly con- tiguous just above antennal sockets. Gena as on vertex except punctures becoming faint on lower half and separated by 2-3 times puncture width, integument other- wise smooth, punctures become faint near preoccipital area. Postgena faintly imbri- cate. Head mostly yellow except facial foveae black and two spots on clypeus dark brown. Inner tooth and mandibular apex reddish brown to black. Proboscis light brown; hypostomal fossa as well as bor- dering areas of postgena and preoccipital area yellow. Antenna yellow. Mesosoma yellow except posterior third of axilla black and spot of dark brown on ventral- facing surface of mesepisternum. Metaso- ma yellow except small dark brown spot laterally outside of gradulus on T2. Pubescence along apicolateral margins of clypeus sparse and simple. Terga with sparse hairs except lateral to pygidial plate where they are long, dense, frequently branched, and golden; sterna similar with- out patches of long, golden hairs on either side of apical cleft of S6, hairs golden, short, and not clustered into patches. Female: Unknown. Holotypc— ARGENTINA: Santiago del Estero: 6, El Pinto, November 1956 (SEMC). Etymology. — The specific epithet is a combination of the Greek words za (very) and mikros (small). Phenology. — This species has so far only been captured in November. Comments. — This is presently the small- est known species of Arhysosage. It is sim- ilar to smaller specimens of A. ochracea but differs most notably in the absence of the banding pattern on the metasoma, in this respect resembling A. flava to which it is perhaps a close relative. Like A. flava, A. zamicra has the elongate punctures of the clypeus but differs from this species in the mandible being slightly shorter than the compound eye, the absence of mandibular striae, the presence of ventral-facing black spots on the mesepisterna, and the point- ed apex of the basitibial plate (Fig. 7). Arhysosage cactorum Moure (Figs. 4, 6-7, 23-25, 31, 37, 43-44) Arhysosage melanotricha Schlindwein and Witt- mann 1995: 32. Lapsus calami and nomen nu- dum. Arhysosage cactorum Moure 1999: 241 Arhysosage melanothricha Moure 1999: 245. New synonymy. Diagnosis. — The male of this species is recognized by the upper half of the face being black (Figs. 24-25), the absence of an emargination at the pygidial plate apex (Fig. 4), the apices of penis valves not bending ventrally (Fig. 44), and the aedea- gus not extending apically near to the api- ces of the penis valves (Figs. 43-44). The female can be distinguished by the mostly yellow labrum, the absence of yellow markings of any sort on the metasoma, and the pygidial plate strongly curved ventrally towards apex in profile and dor- sally quickly tapering to a well-defined point. Description. — As for A. ochracea with the following modifications and additions: Male: Total body length 8.8 mm; fore wing length 5.7 mm. Head width 3 mm, length 2 mm. Mandible longer than compound eye; inner tooth strong and pointed (Figs. 24-25). Upper interorbital distance 1.8 200 Journal of Hymenoptera Research Figs. 17-19. Arhysosage ochracea (Friese), faces, pubescence omitted. 17, Female, most common color pattern. 18, Male, arrow indicates inner tooth. 19, Female, second facial pattern. Stippling indicates black areas, re- mainder yellow. mm, lower interorbital distance 2 mm. In- tertegular distance 1.8 mm. Basitibial plate apex pointed (Fig. 7). Apex of pygidial plate not emarginate (Fig. 4). Apex of pe- nis valve not bending ventrally (Fig. 44); aedeagus not extend apically near to apex of penis valve (Figs. 43-44); terminalia otherwise as depicted in figures 31, 37, and 43-44. Integument of mandible in outer inter- space roughened, becoming smooth by point where outer ridge and condylar ridge meet. Clypeus with faint, coarse punctures, punctures nearly contiguous and longitudinally extended making sur- face appear roughened. Subantennal areas faintly imbricate. Supraclypeal area below antennal sockets and between inner su- bantennal sutures minutely roughened; between antennal sockets punctures well- defined, smaller, and nearly contiguous. Face outside of outer subantennal sutures and below level of antennal sockets coarsely punctured, punctures separated by puncture width or less, integument be- tween smooth; at level of antennal sockets punctures become smaller, well-defined, and gradually more closely spaced until Volume 9, Number 1, 2000 201 1 '"'"•-,. : -.N w */*■■ sit ^ r - 1 .1 f^ ■- . -c_ \ .^JtLjy ^ : s- x ,..--^'"^1 IT* h '■ f .x 1 f vV - \ ^ / ll* i Ik \ 1 \ * ^^-*^-""" Hk».\ ■TF1 fl « 1 Jv r^ * ^1 IF1I Ell r \ /^^ - ^ *^l w I'M 1 ^a $p BP^ .•-" ' >^ ^>^^pSp , 20 Figs. 20-21. Scanning electron micrographs ot .-b/n/sc'SKyc oihracea (Friese), male head. 20, lull face. 21, La- bium, lower paraocular area, and supracivpeal area. 202 Journal of Hymenoptera Research Fig. 22. Arhysosage zamicra n. sp., male face, pubes- cence omitted. Stippling indicates black areas, re- mainder yellow. nearly contiguous by level just above an- tennal sockets; remainder of face and ver- tex with such fine, well-defined punctures, nearly contiguous. Gena as on vertex ex- cept punctures becoming smaller and faint. Pronotal lateral surfaces imbricate and impunctate. Tegula imbricate. Metan- otum with contiguous faint, coarse, punc- tures, integument between imbricate. Preepisternal area as on mesoscutum ex- cept punctures becoming faint and slight- ly more widely spaced ventrally; mesepis- ternum with faint, coarse punctures sep- arated by less than puncture width, integ- ument between faintly imbricate, punctures become fainter ventrally; mete- pisternum with faint, minute punctures separated by width or less, integument be- tween imbricate. Propodeal lateral surface with minute, well-defined punctures sep- arated by puncture width or less, integu- ment between smooth; posterior surface as on lateral surface except punctures faint. Head coloration as in figures 24-25. Pro- boscis brown; hypostomal fossa, postgena, and preoccipital area black. Labrum yel- low. Scape with inner surface black, outer surface yellow; remainder of antenna brown. Pronotum black except pronotal lobe, medioapical border, and lateral spot yellow. Mesoscutum black except border with tegula and two very small spots bor- dering median line yellow. Scutellum yel- low except mediobasal border black. Te- gula and metanotum yellow. Pleura black I igs. 23-24. Arhysosage cactorum Moure, faces, pubescence omitted. 23, Female. 24, Male. Stippling indicates black areas, remainder yellow. Volume 9, Number 1, 2000 203 Fig. 25. Arhysosage cactorum Moure, lateral view of male head. except upper quarter of preepisternal area and medial third of metepisternum yel- low. Coxae and trochanters black; basal border and inner surfaces of femora black, remainder yellow; remainder of legs yel- low. Propodeum yellow except basally bordering metanotum black with a medi- olongitudinal, narrow line of black run- ning from the basal area onto the posterior surface and ending medially at marginal area of propodeum; basal half of lateral surface black. Metasoma uniformly am- ber. Pubescence along apicolateral margins of clypeus sparse and simple. Terga with sparse hairs except lateral to pygidial plate where they are long, dense, frequently branched, and amber; sternal hairs amber, short, and not clustered into patches. Female: As described for the male ex- cept as indicated: Total body length 8.3 mm; forewing length 5.9 mm. Head width 2.9 mm, length 1.9 mm. Upper interorbital distance 1.7 mm, lower interorbital dis- tance 1.8 mm. Intertegular distance 1.8 mm. Pygidial plate in profile strongly Figs. 26-31. Male S7 of Arhysosage species, pubescence on right halves only. 26, Arhysosage ochracea (Friese). 27, A. zamicra n. sp. 28, A. flam Moure. 29, A. atrolunata n. sp. 30, A. bifasciata (Friese). 31, A. cactorum Moure. 204 Journal of Hymenoptera Research 36 Figs. 32-37. Male S8 of Arhysosage species, pubes- cence on right halves only. 32, Arhysosage ochracea (Friese). 33, A. zamicra n. sp. 34, A. flava Moure. 35, A. atrolunata n. sp. 36, A. bifasciata (Friese). 37, A. cac- toruiii Moure. curved ventrally towards apex, quickly ta- pering to well-defined point in dorsal view. Facial coloration as in figure 23. Gena yellow. Proboscis dark brown; hyposto- mal fossa, postgena, and preoccipital area dark brown. Labrum yellow except apical margin brown. Pronotum black except pronotal lobe, posterior median border, and anterior median border yellow. Me- soscutum black except border with tegula yellow. Tegula yellow. Axilla and anterior half of scutellum black, remainder yellow. Metanotum yellow except anterior border black. Pleura black. Legs dark brown ex- cept apices of pro- and meso-femora, out- er surface of protibia, inner surface of me- sotibia, and protarsus yellow. Basal area of propodeum yellow except basal margin and mediolongitudinal line black; lateral and posterior surfaces black except two yellow spots on either side of propodeal pit yellow. Terga amber except anterior- facing surface of Tl dark brown, median band of dark brown on T2, and apical half of T6; sterna light brown. Topotype. — BRAZIL: Rio Grande do Sul: 6* Lavras do sul ("Rincao do Infer- no") 11 January 1991, C. Schlindwein. I was unable to examine the holotype but have seen a male and female of Moure's species (identified by Padre Moure) in the collection of Isabel Alves dos Santos. The male I examined was collected at the same time and place as the holotype. Additional material— ARGENTINA: Salta: El Carril, 11 November 1989, J. G. Rozen and A. Roig-Alsina, on Opuntia sp. (6 AMNH). N. El Carril, 13 November 1993, J. G. and B. L. Rozen, on Opuntia sp. (9 AMNH). Sumalao, November 1994, M. Fritz (49 9 3c5 6 AMNH). BRAZIL: Rio Grande do Sul: Lavras do sul ("Rincao do Inferno") 11 January 1991, C. Schlindwein. (cJPCIA). Cacapava do Sul, 11 Novem- ber 1990, C. Schlindwein (9PCIA). Floral records. — Captured at flowers of an unidentified Opuntia. Phenology. — This species has been col- lected in November and January. KEY TO SPECIES OF ARHYSOSAG1 (Unknown and not included: Females of A. zamicra) 1. Males 2 - Females 7 2. Face predominantly yellow (Figs. 10, 12, 14, 18, 22); scape yellow or at most with small brown patches on inner surface; pygidial plate apex emarginate, sometimes weakly so (Fig. 5); apex of penis valve bent ventrally (Figs. 45-49) 3 Volume 9, Number 1, 2000 205 - Upper half of face entirely black (Fig. 24); scape yellow on outer surface, black on inner surface; pygidial plate apex not emarginate (Fig. 4); apex of penis valve not bent ventrally (Fig. 44) A. cactorum Moure 3. Face with black markings restricted to facial fovea (Figs. 12, 14, 18, 22) 4 - Face with black markings on facial fovea and with a black crescent-like area that connects foveae just above ocelli (Fig. 10) A. atrohmata n. sp. 4. Clypeus with coarse, elongate punctures (Fig. 16); clypeus with dense pubescence at ap- icolateral margins 5 - Clypeus with coarse, rounded punctures (Fig. 21); clypeus with sparse pubescence at apicolateral margins 6 5. Mandible longer than compound eye (Fig. 14), with fine striae on outer interspace (Fig. 15); mesepisterna yellow, without black markings; basitibial plate apex broadly rounded (Fig. 8) A. flava Moure - Mandible slightly shorter than compound eye (Fig. 22), outer interspace without striae, instead imbricate with coarse punctures; mesepisterna with paired black spots; basitibial plate apex pointed (Fig. 7) A. zamicra n. sp. 6. Metasoma banded, yellow with transverse amber bands (Fig. 1) A. ochracea (Friese) - Metasoma uniformly amber A. bifasciata (Friese) 7. Labrum black or brown, infrequently with some small yellow spots or bands; pygidial plate straight or weakly curved in profile, dorsally gently tapering to narrowly rounded apex 8 - Labrum mostly yellow except apical border brown; pygidial plate strongly curved in profile, dorsally quickly tapering to a well-defined point at apex A. cactorum Moure 8. Clypeal integument with coarse, rounded punctures (Fig. 21), clypeus with some black markings aside from paired spots of brown (Figs. 9, 11, 17, 19) 9 - Clypeal integument with coarse, elongate punctures (Fig. 16), clypeus yellow without black markings aside from paired spots of brown (Fig. 13) A. flava Moure 9. Propodeum entirely black, infrequently with small paired spots or transverse bands of yellow along posterior border of basal area; facial color pattern as in figure 11; metasoma mostly black with small yellow spots or bands A. bifasciata (Friese) - Propodeum yellow with black along anterior border of basal area and in a mediolongi- tudinal band running from border with metanotum to metasoma; facial color pattern as in figure 9, 17, or 19; metasoma frequently mostly yellow or amber, sometimes mostly dark but with complete transverse yellow bands on most segments 10 10. Metasoma uniformly amber, without yellow banding or spots; facial color pattern as in figure 9 A. atrohmata n. sp. - Metasoma banded; facial color pattern as in figure 17 or 19 A. ochracea (Friese) NOMINA NUDA IN ARHYSOSAGE Arhysosage xanthina Moure, notnen nudum Arhysosage xanthina Moure /;/ Schlindwein and Wittmann 1995: 32. Comments. — Schlindwein and Wittmann (1995) presented a study on the pollina- tion of the cactus genera Notocactus and Gymnocalycium in which they mention several species of Arhysosage visiting these flowers. Both in the text and in the ac- knowledgments they attribute several Ar- hysosage identifications and names to Pa- dre Moure. The information presented in Schlind- wein and Wittmann (1995) extends the range of the genus into southern-most Brazil. These authors record the locality at which they observed bees on cactus flow- ers as follows: Serra do Sudeste, southeast of Rio Grande do Sul (30°-32° S, 51°-54° W), Brazil, at approximately 500 m eleva- tion. The area is described as subtropical to temperate being humid most of the year 206 Journal of Hymenoptera Research Figs. 38-43. Male genitalia of Arhysosage species; left halves are ventral views, right halves are dorsal views. Aedeagus stippled in figures. 38, Arhysosage ochracea (Friese). 39, A. zamicra n. sp. 40, A. flava Moure. 41, A. atrolunata n. sp. 42, A. bifasciata (Friese). 43, cactorum Moure. Volume 9, Number 1, 2000 207 Figs. 44-49. Male genitalia of Arkysosage species in lateral view; outlines of penis valve apodemes omitted except for a two species so as to contrast their shapes and demonstate variation in the genus. Aedeagus stippled in figures. 44, Arhysosagc cactorum Moure. 45, Arh\/>o>n^e ativlitnatn n. sp. 46, A. zamicra n. sp. 47, A. bifasciata (Friese). 48, A. flava Moure. 49, A. ochracea (Friese). although with water deficiency from De- cember through February. These authors also record bees visiting several cactus flowers: Natacactus pah/acanthus, N. succi- neus, N. sellowii, Gymiwcalycium denudatum (Schlindwein and Wittmann 1995) and Fraileu phaeodisca, F. pygmaea, N. neohorstii, N. attains, Opuntia brunneogemmia, O. viri- dirubra (Schlindwein 1995, Schlindwein and Wittmann 1997). On average 95% of the pollen in loads of individual females came from a single cactus species (Schlindwein and Wittmann op. cit.). ACKNOWLEDGMENTS I am grateful to Isabel Alves dos Santos tor a loan of A. cactorum specimens and to Padre Jesus S. Moure for communications concerning Arhysosage. I am thankful to the persons mentioned in the Materials and Methods who graciously loaned material under their stewardship. Molly G. Rightmyer kindly assist- ed with the scanning electron microscopy and prep- aration of the final images. I thank John L. \ett who provided information on floral associations tor spe< - imens ot ArhysOSdge he collected during expeditions in 1972-1973 and 1986. John I Neff, Molly G. Right- myer, Jerome ti. Ro/en, Jr., and two anonymous re- viewers read early versions ot the manuscript .md provided valuable suggestions tor its improvement. Support tor my studies has been generously provided by Robert ci. Goelet, chairman Emeritus ot the \\l\l I board ot I rusUvs. LITERATURE CITED Ashmead, \\ . II L899. ( lassification ot the bees, or tlir superfamily Apoidea. Transactions of the American I ntomological Society 26: 4l> inn 208 Journal of Hymenoptera Research Brethes, J. 1922. Himenopteros y Dfpteros de varias procedencias. Anales de la Sociedad Cientifica Ar- gentina 93: 119-146. Cockerell, T. D. A. 1940. A new bee from the argen- tine republic. American Museum Novitates 1080: 1. Cortes, R. 1951. Nuevos generos de Tachininae chi- lenos con cerdas facio-orbitales (Diptera Tachin- idae). Revista Chilena Entomologia 1: 249-262. Cresson, E. T. 1878. Descriptions of new North Amer- ican Hymenoptera in the collection of the Amer- ican Entomological Society. Transactions of the American Entomological Society 7: 61-136. Danforth, B. N., and J. L. Neff. 1992. Male polymor- phism and polyethism in Perdita texana (Hyme- noptera: Andrenidae). Annals of the Entomological Society of America 85: 616-626. Farris, J. S. 1988. HENNIG86, v. 1.5, Program and Documentation. Port Jefferson, New York. Friese, H. 1908. Die Apidae (Blumenwespen) von Argen- tina uach den Reisenergebnissen der Herrn A. C. Jen- sen-Haarup und P. Jorgensen in den jahren 1904— 1907. Verlag "Flora og Fauna", Silkeborg, Den- mark, 111 + [4] pp. Jensen-Haarup, A. C. 1908. Biological researches amongst the argentine bees with special refer- ence to flowers they visit. Pages 95-107, In H. Friese Die Apidae (Blumenwespen) von Argen- tina nach den Reisenergebnissen der Herrn A. C. Jensen-Haarup und P. Jorgensen in den Jahren 1904-1907. Verlag "Flora og Fauna", Silkeborg, Denmark, 111 + [4] pp. Jorgensen, P. 1912. Revision der Apiden der Provinz Mendoza, Republica Argentina (Hym.). Zoologis- che Jahrbiicher, Abteilung fur Systematik, Geogra- phic, und Biologic der Tiere 32: 89-162. Michener, C. D. 1944. Comparative external mor- phology, phylogeny, and a classification of the bees. Bulletin of the American Museum of Natural History 82: 151-326. Michener, C. D. In press. The Bees of the World. Johns Hopkins University Press, Baltimore, United States. Michener, C. D., and A. Fraser. 1978. A comparative anatomical study of mandibular structure in bees. University of Kansas Science Bulletin 51: 463- 482. Moure, J. S. 1958. On the genus Arln/sosage Brethes from Argentina (Hymen., Apoidea, Panurginae). Entomological News 69: 43-48. Moure, J. S. 1999. Duas especies novas de Arln/sosage Brethes do Brasil (Rio Grande do Sul) e da Ar- gentina (Salta) (Hymenoptera, Apoidea, Panur- ginae). Revista Brasileira tie Zoologia, supl, .?, 16: 241-247. Neff, J. I.., and B. N. Danforth. 1992. The nesting and foraging behavior of Perdita texana (Cresson) (Hymenoptera: Andrenidae). Journal of the Kansas Entomological Sot levy 64: 394-405. Nixon, K. C. 1993. CLADOS, v. 1.4.88, Program and Documentation. Trumansburg, New York. Nixon, K. C. 1995. DADA, v. 0.96.1, Program and Documentation. Trumansburg, New York. Ruz, L. 1986. Classification and Phylogenetic Relation- ships of the Panurgine Bees (Hymenoptera- Andreni- dae). Ph.D. Dissertation, University of Kansas, Lawrence, Kansas, 312 pp. Ruz, L. 1991. Classification and phylogenetic relation- ships of the panurgine bees: The Calliopsini and allies (Hymenoptera: Andrenidae). University of Kansas Science Bulletin 54: 209-256. Schlindwein, C. 1992. Arln/sosage sp. (Hymenoptera, Apoidea, Andrenidae) an oligolectic pollinator of cactus flowers. Resumes do 12th Congresso Latino- Americano de Zoologia e 19th Congresso Brasileiro de Zoologia 1992: 61. Schlindwein, C. 1995. Melittophilous plants, their pol- len and flower visiting bees in southern Brazil. 2. Cactaceae. Biociencias, Porto Alegre 3: 35-71. Schlindwein, C, and D. Wittmann. 1995. Specialized solitary bees as effective pollinators of South Bra- zilian species of Notocactus and Gymnocalycium (Cactaceae). Bradleya 13: 25-34. Schlindwein, C, and D. Wittmann. 1997. Stamen movements in flowers of Opuntia (Cactaceae) fa- vour oligolectic pollinators. Plant Systematics and Evolution 204: 179-193. Spinola, M. 1843. Sur quelques Hymenopteres peu connus, recueillis en Espagne, pendant 1'annee 1842, par M. Victor Ghiliani, voyageur-natural- iste. Annalcs de la Societe Entomologique de France 1: 111-144. Snelling, R. R., and B. H. [sic: N.], Danforth. 1992. A review of Perdita, subgenus Macrotera (Hyme- noptera: Andrenidae). Contributions in Science 436: 1-12. Timberlake, P. H. 1952a. Descriptions of new species of Nomadopsis from California and Texas, and a new allied genus from South America. Annals of the Entomological Society of America 45: 104-118. Timberlake, P. H. 1952b. A new name for the bee genus Ruiziella. Annals of the Entomological Society of America 45: 528. Timberlake, P. H. 1953. Erratum. Annals of the Ento- mological Society of America 46: 598. Toro, H. 1995. Una nueva especie de Spinoliella (Apo- idea: Andrenidae) del norte de Chile. Acta Ento mologica Chilena L9: 105-107. Toro, H., and M. Herrera. 1980. Las especies chilenas del genero Callonychium (Andrenidae-Apoidea) v description de un nuevo genero [sic: subgenera]. Anales del Museo de Historia Natural de Valparaiso 13: 213-225. Toro, H., and L. Ruz. 1972. Revision del genero Spi- noliella (Andrenidae-Apoidea). Anales del Museo de Historia Natural de Valparaiso 5: 137-171, 293- 295. J. HYM. RES. Vol. 9(1), 2000, p. 209 NOTE Cretobestiola, a replacement name for Bestiola Pulawski and Rasnitsyn, 1999 (Hymenoptera: Sphecidae) Wojciech J. Pulawski a\d Alexandr P. Rasmtsyn (WJP) Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA; e-mail: wpulawski@calacademv.org; (APR) Paleontological Institute, Russian Academy of Sciences, Moscow 117647, Russia; e-mail: rasna@glasnet.ru We recently described the sphecid wasp genus Bestiola (in Rasnitsyn et al. 1999) to accommodate four species from Lower Cretaceous of Spain, eastern Russia, and Mongolia. This generic name, however, is preoccupied by Bestiola Nikol'skaya (1963), an aphelinid, as pointed out to us by Mr. John K. Page (Zoological Records, York, Great Britain) and also by Signor Guido Pagliano (Torino, Italy). We there- fore propose the name Cretobestiola to re- place it. The name is derived from the Lat- in creta (chalk), with reference to Creta- ceous geological period, and bestiola (little beast). LITERATURE CITED Nikol'skaya, M. N. 1963. Dva novykh roda afelinid (Hvmenoptera, Chalcidoidea), vyvedennykh iz bambukovoi shchitovki Odonapis secreta na Kav- kazke (Ckll.) [Two new genera of Aphelinidae (Hymenoptera, Chalcidoidea) bred from Odon- apis secreta (Ckll.) in Caucasus.] Entomologiches- koye Obozrenye 42: 186-189. Rasnitsyn, A. P., W. J. Pulawski, and X. Martinez- Delclos. 1999. Cretaeous digger wasps of the new genus Bestiola Pulawski and Rasnitsyn (Hvme- noptera: Sphecidae: Angarosphecinae). journal of Hymenoptera Research 8: 23-34. J. HYM. RES. Vol. 9(1), 2000, pp. 210-212 NOTE Two New Records of Pimpline Ichneumonids Attacking Battus pliilenor (Linnaeus) (Lepidoptera: Papilionidae) Karen R. Sime Department of Ecology & Evolutionary Biology, and Department of Entomology, Cornell University, Ithaca, New York 14850 A series of seven ichneumonids redis- covered in the American Museum of Nat- ural History collection offers two new host associations. All seven parasitoids were reared from pupae of Battus philenor (Lin- naeus); one of the specimens is Theronia atalantae (Poda), and the other six are Apechthis annulicornis (Cresson). The host pupae were all collected on October 3, 1960, at a single site ("North Carolina: above Crabtree to Betsey's Gap; 3956'; Haywood Co.; coll. Gertsch, Ivie"), and the wasps emerged in November 1960. The pupal remains were kept with the wasps, permitting confirmation of the hosts' identity. These associations are particularly inter- esting because Battus philenor is consid- ered relatively immune to attack by par- asitoids. It may be protected by toxins ac- quired from its food plants, which are the various North American species of Aris- tolochia (Aristolochiaceae). The adults have long been recognized as the unpalatable models for a Batesian mimicry ring (Brow- er 1958); like many other species in the tribe Troidini Nishida 1995), B. pliilenor se- questers substantial quantities of aristo- lochic acids to the adult stage (Sime, in prep.). Experimental evidence indicates that these compounds are unusually toxic, at least to non-adapted lepidopteran lar- vae (e.g. Miller and Feeny 1989). Haase (1893) was the first to link Aristolochi- aceae-feeding with low parasitism rates, observing that B pliilenor pupae never yielded parasitoids while the pupae of the palatable mimics Papilio troilus Linnaeus and P. glaucus Linnaeus frequently pro- duced ichneumonids. Immature B. philenor are commonly col- lected and studied, and the considerable rearing data available in collections and in both published and unpublished studies suggest that larval and larval-pupal para- sitoids are nearly nonexistent and pupal parasitoids rather rare. Among reports in the literature involving larvae, there is just a single record, lacking ecological data, for the tachinid Compsilura concinnata (Mei- gen) (Schaffner and Griswold 1934), which when in butterflies is a larval-pupal par- asitoid (Ford and Shaw 1991). I have reared some 90 B. philenor (found as lar- vae) from the Blue Ridge Mountains of Virginia (a habitat similar to that for the new records), but obtained only butter- flies. Rausher (1981), in an exhaustive study of B. philenor ecology in Texas, not- ed that in four field seasons no parasitoids emerged from several hundred field-col- lected larvae reared to adulthood. Other authors have collected pupae: West and Hazel (1982) reported no parasitoids in a Virginia study, though Sims and Shapiro (1983) found that Brachymeria ovata (Say) (Chalcididae) occasionally inflicts high mortality in some California populations of B. philenor. Gambrus amoenus (Graven- horst) (= nuncius (Say)) (Ichneumonidae) has reportedly been reared from B. phile- nor, but the genus is thought only to attack Volume 9, Number 1, 2000 211 cocooned hosts, and Townes and Townes (1962) and Gupta (1983) consider this spe- cies a specialist on Callosamia and a few other Saturniidae. Theronia atalantae is common and wide- spread; it is a polyphagous parasitoid of pupae apparently capable of both primary and secondary parasitism, though almost invariably found to be a secondary para- sitoid when possible to investigate (Townes 1940). Almost always associated with Lepidoptera, it has been reared from at least 16 families, most frequently Ly- mantriidae and Lasiocampidae. (The ap- parent bias towards these two families may be an artifact of the attention that pests such as gypsy moth and forest tent caterpillars have received.) Whether the new record represents primary or second- ary parasitism (perhaps on A. annulicornis) is not known. As a secondary parasitoid, however, its presence in 8. philenor would not necessarily be remarkable: if the pri- mary parasitoid detoxifies plant poisons in the lepidopteran pupa, it might itself become a non-toxic host for T. atalantae. ApechtJiis annulicornis is a pupal parasit- oid of Lepidoptera; it is reared less often than is T. atalantae, but a number of re- ports indicate that it has a broad host range, attacking Neophasia menapia (Felder & Felder) (Pieridae) and various species of Choristoneura (Tortricidae) and Orgyia (Ly- mantriidae) (Carlson 1979 and refs. there- in). This record, together with those of 8. ovata and C. concinnata (each of which at- tacks many families and over 100 species of Lepidoptera (Arnaud 1978; Halstead 1988)), indicates that 8. philenor is the oc- casional host of several relatively polyph- agous parasitoids that are reared much more often from other, less toxic Lepidop- tera. This phenomenon may reflect a broad constitutive tolerance of plant alle- lochemicals on the part of the generalists: that larval and larval-pupal hymenopter- an parasitoids are, in contrast, entirely lacking suggests that the unusual toxicity of aristolochic acids has prevented the evolution of specialist, koinobiont parasit- oids of 8. philenor. ACKNOWLEDGMENTS I thank Michael Engel and Eric Quinter (AMNH) for arranging the loan of specimens, and Rick Hoe- beke for assistance with the identification of ichneu- monid species. Paul Feeny, Cheryl Heinz, Mike McDonald and Mark Shaw provided helpful com- ments on the manuscript. LITERATURE CITED Arnaud, P. H. 1978. A host-parasite catalog of North American Tachinidae (Diptera). U.S. Department of Agriculture Miscellaneous Publications No. 1319: 1-860. Brower, J. V. Z. 1958. Experimental studies of mim- icry in some North American butterflies. II. Bat- tus philenor and Papilio iroilus, P. polyxenes, and P. glaucus. Evolution 12:123-136. Carlson, R. W. 1979. Family Ichneumonidae. pp. 315- 379 in: A catalog of Hymenaptera in America north of Mexico, volume 1 (eds K. V. Krombein, P. D. Hurd, D. R. Smith, and B. D. Burks). Smithsonian Institution Press, Washington D.C. Ford, T. H. and M. R. Shaw. 1991. Host records of some West Palearctic Tachinidae (Diptera). En- tomologist's Record and journal of Variation 103:23- 38. Gupta, V. 1983. The ichneumonid parasites associated with the gypsy moth (Lymantria dispar). Contri- butions to the American Entomological Institute 19: 1-168. Haase, E. 1893. Untersuchungen tiber die Mimicry auf Grundlage ernes natiirlichen Systems der Papilioni- den (Zwciter Theil). E. Nagele, Stuttgart. Halstead, J. A. 1988. First records of Platychalcis in North America and new host records of Ceratos- micra spp. (Hymenoptera: Chalcididae). Entomo- logical News 99:193-198. Miller, J. S. and Feeny, P. P. 1989. Interspecific differ- ences among swallowtail larvae (Lepidoptera: Papilionidae) in susceptibility to aristolochic ac- ids and berberine. Ecological Entomology 14:287- 296. Nishida, R. 1995. Sequestration of plant secondary compounds bv butterflies and moths. Chemoecol- ogy 5/6:127-138. Rausher, M. 1981. Host plant selection by Battus phi- lenor butterflies: the roles of predation, nutrition, and plant chemistry. Ecological Monograph'- 51(1): 1-20. Schaffner, 1. V. and Criswold, C. L. 1934. Macrolepi- doptera and their parasites reared from field col- lections in the northeastern part of the United States. U.S. Department of Agriculture Miscella neous Publications No. 188:l-lw). 212 Journal of Hymenoptera Research Sims, S. R. and Shapiro, A. M. 1983. Pupal color di- Part I. Memoirs of the American Entomological Iu- morphism in California Battus philenor L. (Papi- stitute 11:1-300. lionidae): mortality factors and selective advan- Townes, H. K., and Townes, M. 1962. Ichneumon-flies tage. Journal of the Lepidopterists' Society 37:236- of America north of Mexico: 3. Subfamily Geli- 243. nae, Tribe Mesostenini. United States National Townes, H. K. 1940. A revision of the Pimplini of Museum Bulletin 216 (3): 1-602. eastern North America. Annals of the Entomologi- West, D. A., and Hazel, W. N. 1982. An experimental cal Society of America 33:283-323. test of natural selection for pupation site in swal- Townes, H. K. 1969. The genera of Ichneumonidae, lowtail butterflies. Evolution 36:152-159. INSTRUCTIONS FOR AUTHORS General Policy. The Journal of Hymenoptera Research invites papers of high scientific quality reporting comprehensive research on all aspects of Hymenoptera, including biology behavior, ecology, systematics, taxonomy, genetics, and morphology Taxonomic papers describing single species are unlikely to be accepted unless a strong case is evident, such as importance in economic entomology or with concurrent biology or ecology. Manuscript length generally should not exceed 50 typed pages; however, no upper limit on length has been set for papers of exceptional quality and importance, including taxonomic monographs at generic or higher level. All papers will be reviewed by at least two referees. The referees will be chosen by the appropriate sub- ject editor. However, it would be helpful if authors would submit the names of two persons who are compe- tent to review the manuscript. The language of publication is English. Summaries in other languages are acceptable. The deadline for receipt of manuscripts is 1 October (for the April issue) and 1 April (for the October issue). Format and Preparation. Three copies of each manuscript, including copies of illustrations, should be submitted on letter size or A4 paper, double spaced, with at least 25 mm margins on all sides. On the upper left of the title page give name, address and telephone and fax numbers of the author to whom all corre- spondence is to be sent. The paper should have a concise and informative title, followed by the names and addresses of all authors. The sequence of material should be: title, author(s), abstract, text, acknowledgments, literature cited, appen- dix, figure legends, figure copies (each numbered and identified), tables (each numbered and with heading). Each of the following should start a new page: (1) title page, (2) abstract, (3) text, (4) literature cited, (5) figure legends, (6) footnotes. Following acceptance of the manuscript, the author should provide the editor with one copy of the manuscript accompanied by a copy on diskette using DD, double sided computer diskettes — IBM compati- ble MS DOS 5.25 inch or IBM and Macintosh 3.5 inch diskettes. (Authors who do not have access to a com- puter should submit three copies of the manuscript.) The paper may be submitted in most PC and Mac word processor programs such as Microsoft Word, FullWrite Professional, WordPerfect, WriteNow, Nisus, Mac- Write, or Mac Write II. If possible, all words that must be italicized should be done so, not underscored. Tables may be formatted in a spread sheet program such as MS Works or MS Excel. Text should be double- spaced typing, with 25 mm left and right margins. Tables should be put in a separate file. Diskettes should be accompanied by the name of the software program used (e.g., WordPerfect, Microsoft Word). Authors should keep backup copies of all material sent to the Editor. The Society cannot be responsible for diskettes or text mislaid or destroyed in transit or during editing. Illustrations should be planned for reduction to the dimension of the printed page (14 X 20.5 cm, column width 6.7 cm) and allow room for legends at the top and bottom. Do not make plates larger than 14 X 18 in. (35.5 X 46 cm). Individual figures should be mounted on a suitable drawing board or similar heavy stock. Photographs should be trimmed, grouped together and abutted when mounted. Figure numbers should be on the plate, and it is strongly recommended that names be included after the numbers (e.g., Fig. 2, texanus). Include title, author(s) and address(es), and illustration numbers on back of each plate. Original figures need not be sent until requested by the editor, usually after the manuscript has been accepted. Reference to figures /tables in the text should be in the style "(Fig. 1)" "(Table 1)". Measurements should be in the metric system. All papers must conform to the International Code of Zoological Nomenclature. The first mention of a plant or animal should include the full scientific name including the authority Genus names should not be abbre- viated at the beginning of a sentence. In taxonomic papers type specimens must be clearly designated, type depositories must be clearly indicated, and new taxa must be clearly differentiated from existing taxa by means of keys or differential diagnoses. Authors are required to deposit all type material in internationally recognized institutions (not private collections). Voucher specimens should be designated for specimens used in behavioral or autecological studies, and they should be deposited similarly Acceptance of taxonomic papers will not require use of cladistic methods; however, authors using them will be expected to specify the phylogenetic program used (if any), including discussion of program options used. A data matrix should be provided if the subject is complex. Cladograms must be hung with characters and these should include descriptors (not numbers alone) when feasible. The number of parsimonious cladograms generated should be stated and reasons given for the one adopted. Lengths and consistency indices should be provided. Adequate discussions should be given for characters, plesiomorphic conditions, and distributions of characters among outgroups when problematical. References in the text should be (Smith 1999), without a comma, or Smith (1999). Two articles by a single author should be (Smith 1999a, 1999b) or Smith (1999a, 1999b). For multiple authors, use the word "and," not the symbol "&" (Smith and Jones 1999). For papers in press, use "in press, not the expected publication date. The Literature Cited section should include all papers referred to in the paper. Journal names should be spelled out completely and in italic s. Charges. Publication charges are $10.00 per printed page. At least one author oi the paper must be a member of the International Society of Hymenopterists. Reprints are charged to the author and must be ordered when returning the proofs; there are no free reprints. Author's corrections and changes in proof .ire also charged to the author. Color plates will be billed at full cost to the author. All manuscripts and correspondence should be sent to: Dr. E. Eric Grissell Systematic Entomology Laboratory, USDA ' '< National Museum of Natural History Washington, DC 20560-0168 Phone: (202) 382-1781 Fax: (202) 786-9422 E-mail: egrissel@sel.barc.usda.gov CONTENTS (Continued from front cover) PULAWSKI, W. J., A. P. RASMTSYN, D. J. BROTHERS, and S. B. ARCHIBALD. New genera of Angarosphecinae: Cretospheciwn from Early Cretaceous of Mongolia and Eosphecium from Early Eocene of Canada (Hymenoptera: Sphecidae) 34 QUICKE, D. L. J. and A. POLASZEK. A new genus, and first host records, for the Adeshini: parasitoids of hispine beetles (Braconidae: Braconinae; Coleoptera: Chrysomelidae) 104 ROBERTSON, H. G. Afrotropical ants (Hymenoptera: Formicidae): taxonomic progress and estimation of species richness 71 SCHAUFF, M. E. and R. GARRISON. An introduced species of Epichrysocharis (Hymenop- tera: Eulophidae) producing galls on Eucalyptus in California with notes on the described species and placement of the genus 176 SHARKEY, M. J., K. FTNNELL, J. LEATHERS, and J. FRANA. Microgastrine (Hymenoptera: Braconidae) parasitoids of Colias lesbia (Fabricius) (Lepidoptera: Pieridae) 108 UBAIDILLAH, R., J. LaSALLE, and D. L. J. QUICKE. A peculiar new genus and species of Entedoninae (Chalcidoidea: Eulophidae) from Southeast Asia 170 WARD, P. S. On the identity of Pheidole vaslitii Pergande (Hymenoptera: Formicidae), a neglected ant from Baja California 85 NOTES PULAWSKI, W. J. and A. P. RASNITSYN. Cretobestiola, a replacement name for Bestiola Pulawski and Rasnitsyn, 1999 (Hymenoptera: Sphecidae) 209 SIME, K. R. Two new records of pimpline ichneumonids attacking Battus philenor (Linnaeus) (Lepidoptera: Papilionidae) 210 itfhoClETy Journal of Hymenoptera Research IBRAR^S Volume 9, Number 2 ^ October 2000 ISSN #1070-9428 CONTENTS AZEVEDO, C. O, J. C. SILVA-Jr, and L. A. de O. CAMPOS. Description of a new species of Emersonella (Hymenoptera: Eulophidae) from Brazil, with prelimary observations on its biology 298 BAPTISTE, L. A. and L. S. KJMSEY. New Caledonian Tiphiidae: revision of the genus Eirone (Hymenoptera: Thynninae) 395 CONTI, E., P. F. ROVERSI, and F. BIN. Morphofunctional adaptations of parasitoids attacking concealed eggs of two arboreal mirids in Italy 385 DELVARE, G. and J. LaSALLE. Trisecodes gen. n., (Hymenoptera: Eulophidae: Entedoninae), the first eulophid with three tarsal segments 305 DIAZ; F. A. The Venezuelan species of Pimpla (Hymenoptera: Ichneumonidae) 246 EBERHARD, W. G. The natural history and behavior of Hymenoepimecis argyraphaga (Hymenoptera: Ichneumonidae) a parasitoid of Plesiometa argyra (Araneae: Tetragnathidae) 220 EDGAR, P. K. and J. R. COELHO. Load-lifting constraints on provisioning and nest build- ing in the carpenter wasp, Monobia quadridens L. (Hymenoptera: Eumenidae) .... 370 FORTIER, J. C. Description of a new gregarious species of Aleiodes Wesmael (Hymenoptera: Braconidae: Rogadinae) 288 GAULD, I. D. The re-definition of pimpline genus Hymenoepimecis (Hymenoptera: Ich- neumonidae) with a description of a plesiomorphic new Costa Rican species .... 213 HANSSON, C. Description of a new genus of Entedoninae (Hymenoptera: Eulophidae) from the Neotropical Region, including three new species 313 JOHNSON, R. A. Reproductive biology of the seed-harvester ants Messor julianus (Per- gande) and Messor pergandei (Mayr) (Hymenoptera: Formicidae) in Baja California, Mexico 377 (Continued on back cover) INTERNATIONAL SOCIETY OF HYMENOPTERISTS Organized 1982; Incorporated 1991 OFFICERS FOR 2000 Andrew D. Austin, President John LaSalle, President-Elect James B. Woolley, Secretary John T. Huber, Treasurer E. Eric Grissell, Editor Subject Editors Symphyta and Parasitica Aculeata Biology: Mark Shaw Biology: Sydney Cameron Systematics: Donald Quicke Systematics: Wojciech Pulawski All correspondence concerning Society business should be mailed to the appropriate officer at the following addresses: President, Department of Crop Protection, University of Adelaide, Glen Os- mond, Australia 5064; Secretary, Department of Entomology, Texas A&M University, College Station, Texas 77843; Treasurer, Eastern Cereal & Oilseed Research Centre, Agriculture Canada, K. W. Neatby Building, Ottawa, Ontario, Canada Kl A 0C6; Editor, Systematic Entomology Labo- ratory, USD A, % National Museum of Natural History, Washington, D.C 20560-0168. Membership. Members shall be persons who have demonstrated interest in the science of ento- mology Annual dues for members are US$40.00 per year (US$35.00 if paid before 1 February), payable to The International Society of Hymenopterists. Requests for membership should be sent to the Treasurer (address above). Information on membership and other details of the Society may be found on the World Wide Web at http://IRIS.biosci.ohio-state.edu/ish. Journal. The Journal of Hymenoptera Research is published twice a year by the International Society of Hymenopterists, % Department of Entomology, Smithsonian Institution, Washington, D.C. 20560-0168, U.S.A. Members in good standing receive the Journal. Nonmember subscriptions are $60.00 (U.S. currency) per year. The Society does not exchange its publications for those of other societies. Please see inside back cover of this issue for information regarding preparation of manuscripts. Statement of Ownership Title of Publication: Journal of Hymenoptera Research. Frequency of Issue: Twice a year. Location of Office of Publication, Business Office of Publisher and Owner: International Society of Hymenopterists, % Department of Entomology, Smithsonian Institution, 10th and Constitution NW, Washington, D.C. 20560-0168, U.S.A. Editor: E. Eric Grissell, Systematic Entomology Laboratory, USDA, % National Museum of Nat- ural History, Washington, D.C. 20560-0168. Managing Editor and Known Bondholders or other Security Holders: none. This issue was mailed 27 October 2000 J. HYM. RES. Vol. 9(2), 2000, pp. 213-219 The Re-definition of Pimpline Genus Hymenoepimecis (Hymenoptera: Ichneumonidae) with a Description of a Plesiomorphic New Costa Rican Species Ian D. Gauld The Natural History Museum, Cromwell Road, London SW7 5BD, UK Abstract. — An aberrant new species of the Neotropical genus Hymenoepimecis, H. argyraphaga Gauld n. sp. is described from Costa Rica, and the genus is redefined to accommodate this new taxon. A key is provided to identify the Costa Rican species. It is hypothesised that H. argyraphaga occupies a basal position in the genus, and is the sister group to all other species. The systematic position of Hymenoepimecis within the Polysphincta genus-group is discussed and a sister-group relationship with Acrotaphus, another New World genus, is demonstrated. It is suggested that this clade has arisen from within a paraphyletic "Polysphincta" complex, a cosmopolitan group that has yet to be resolved as a series of monophyletic taxa. Host records for the clade are summarised, and it is shown that the group are specialist parasitoids of orb-web spiders of the families Ara- neidae and Tetragnathidae. Hymenoepimecis is an exclusively neo- tropical genus of ichneumonid wasps that belongs to the Polysphincta complex of genera, the "Polysphinctini" seiisu Townes (1969), a monophyletic clade that has aris- en from within the pimpline tribe Ephial- tini (Wahl and Gauld, 1998). This clade is biologically unique within the Ichneu- monidae because all members are koino- biont ectoparasitoids of spiders (Fitton et al. 1988; Gauld 1991; Gauld et al. 1998). Elsewhere in the Hymenoptera such an as- sociation is only known in a few Pompi- lidae (Wasbauer 1995). The female ichneu- monid temporarily paralyses a spider by stinging it, and then attaches an egg either to the cephalothorax (the Schizopyga/ Dreisbachia subgroup) or to the abdomen (the Polysphincta and Zatypota subgroups). The ichneumonid larva develops as an ec- tophagous parasitoid on the active spider. The Hymenoepimecis of southern Mesoa- merica are reasonably well-known (Gauld 1991; Gauld et al. 1998), but recently, a new species of has been found in Costa Rica which possesses certain plesiomorph- ic features that necessitate redefining the genus. The purpose of this paper is to do this, and to describe and characterise this new species, in order to provide a taxo- nomic background for the following paper (Eberhard, 2000) which describes the bi- ology of these fascinating insects. Genus Hymenoepimecis Viereck Epimecis Brulle 1846: 112. Type-species: Epimecis bicolor Brulle, designated by Ashmead 1900: 54. Qunior homonym of Epimecis Hiibner.] Hymenoepimecis Viereck 1912: 149. [Replace- ment name for Epimecis Brulle.] Diagnosis. — Medium to large insects (fore wing length 6-14 mm) which are generally orange with black marking, with the wings from more or less hyaline to completely black, occasionally black and yellow patterned. Head somewhat glo- bose, though abruptly declivous posteri- orly; clypeus simple, not transversely di- vided, flat, apically truncate or slightlv concave; mandible slender, strongly ta- pered with upper tooth distinctly the lon- ger; palp formula 5:4; occipital carina very strong, dorsallv convex, flange-like, con- tinuous to base of mandible, with part just 214 Journal of Hymenoptera Research below level of foramen magnum expand- ed to approach its counterpart below neck in some species; eyes large; antennae long and slender. Pronotum in profile from moderately to exceptionally long, with an- terior margin reflexed, projecting below the occipital flange, with part immediately behind this modified into a anteriorly opening "pocket-like" structure; epomia entirely absent. Mesoscutum smooth and polished, more or less glabrous; notauli weakly to moderately impressed; meso- pleuron polished, with epicnemial carina from completely absent to present ven- trally, but laterally not reaching above lev- el of lower corner of pronotum; metapleu- ron polished, with submetapleural carina usually absent, sometimes present anteri- orly; propodeum quite short and evenly rounded posteriorly, without discernible carina, except for vestiges peripherally, but never with any enclosed areae; pro- podeal spiracle more or less circular. Legs slender, but with fore legs variously de- veloped, sometimes of similar size to mid- dle legs, but often enlarged and with the fore femur conspicuously larger than mid- dle femur; claws of female with large bas- al lobe, which in many species is high and short, but in one large South American species group is long and low, almost tooth-like; claws of male simple, with a small internal membranous vesicle. Fore wing with 3rs-m entirely absent, but al- ways with 2rs-m quite long; hind wing with first abscissa of M+Cul straight or weakly angled proximal to its centre; dis- tal abscissa of Cu\ present, joining cu-a from slightly to conspicuously closer to M than to 1A Metasoma slender, depressed, polished and more or less impunctate; ter- gite II with weak to strong oblique im- pressions anterolaterally, tergites III-IV with weak lateromedian convexities; ovi- positor with a distally angulate basal swelling ventrally, with shaft from more or less straight to slightly up-curved, 1.0- 1.4 times the length of the hind tibia, weakly swollen centrally, apically elon- gately tapered to a fine sharp point. Remarks. — Hymenoepimecis is a Central and South American genus, comprising eight described species (Yu and Horst- mann 1997). Approximately ten unde- scribed species are known mainly from lowland or mid-altitude South America, occurring between sea-level and 1800 me- tres (in collections of American Entomo- logical Institute, Gainesville, and The Nat- ural History Museum, London). The geo- graphical range of the genus extends from tropical Mexico and Cuba south to sub- tropical Southern Brazil (c. 29°S). In earlier systematic works (e.g., Townes 1969), the genus has been described as lacking both the epicnemial and submetapleural cari- nae, but recently an exception has been found in Costa Rica. This somewhat ab- errant species has a discernible epicnemial carina and a more or less fully developed submetapleural carina. However, its ge- neric placement is attested by the posses- sion of two autapomorphies of Hymenoe- pimecis. First, the pronotum is mediodor- sally modified to have a unique anteriorly opening "pocket-like" flange just behind the reflexed anterior margin. Second, the fore legs are enlarged, with the femora slightly larger than the middle femur. These apomorphies are unique within the Pimplinae, and strongly suggest that the slightly expanded Hymenoepimecis is a monophyletic group. Systematic position. — Hymenoepimecis be- longs to the Polysphincta genus-group ( = Polysphinctini sensu Townes 1969), a clearly definable monophyletic clade of Ephialtini (Wahl and Gauld 1998). It is pu- tatively the sister group of another pri- marily Neotropical genus, Acrotaphus, a relationship that is supported by three au- tapomorphies. First, the occipital carina is strongly raised, flange-like, and projects posteriorly to surround the anterior re- flexed end of the pronotum. Second, the head is rounded (more or less "door-knob shaped") with the genae strongly nar- Volume 9, Number 2, 2000 215 other ■' Polysphincta" spp Polysphincta dizardi sp-grp Acrolaphus Hymenoepimecis argyraphaga other Hvmenoepimecis spp Fig. 1. Cladogram summarizing hypotheses of relationship of Hymenoepimecis argyraphaga. The derived fea- tures supporting these clades are: (1) long straight ovipositor with an angulate basal swelling; metasomal tergites II-III biconvex: (2) absence of epomia; enlarged ocelli and eyes: (3) occipital carina is strongly raised, flange-like, projecting backwards to surround the anterior reflexed end of the pronotum; head rounded with genae strongly narrowed from eyes to occipital flange; pronotum unusually elongate, with a long horizontal part mediodorsally: (4) pronotum mediodorsally with a forwardly directed "pocket-like" flange; fore legs enlarged, with the femora of similar size to or larger than the mid femur: (5) presence of a horizontal pronotal "shelf": (6) cocoon without a caudal orifice: (7) wings uniformly blackish: (8) loss of epicnemial carina; loss of submetapleural carina. All characters are polarized with reference to the condition in Tromatobia. rowed from the eyes to the occipital flange. Third, the pronotum is unusually- elongate, with a long horizontal part me- diodorsally. The Hymenoepimecis/ Acrotaphus clade is, in its turn, the sister group of yet another Neotropical group, the Polysphincta dizardi species-group (Gauld 1991). All three of these taxa lack any trace of an epomia, and are slender, highly polished insects with enlarged eyes and ocelli. This entire lineage is part of the Polys- phincta subgroup, a large cosmopolitan group of species that is characterized by having a uniquely basally swollen, ovi- positor, and possessing biconvex metaso- mal tergites II-IV. The detailed phylogeny of this clade has yet to be fully resolved, but the genus Polysphincta (sensu Townes, 1969 and all subsequent authors) is appar- ently a paraphyletic assemblage, although most of the species-groups within it are demonstrably monophyletic (Gauld 1991; Gauld et al. 1998) (Fig. 1). All recorded hosts of the Polysphincta subgroup (i.e., "Polysphincta" s.l., Acrota- phus and Hymenoepimecis) are orb- web spi- ders of the families Araneidae and Tetrag- nathidae (Nielsen 1923; Townes and Townes, 1960; Fitton et al. 1988; Fincke et al. 1990; Gauld 1991). All records of "Po- lysphincta" are from Araneidae (e.g. Townes and Townes 1960; Fitton et al. 1988; unpublished records in Natural His- tory Museum, London), whereas various species of Acrotaphus and Hymenoepimecis have been reared either from Araneidae or Tetragnathidae (Shannon 1913; Gauld et al. 1998; Eberhard 2000). Costa Rican species. — Despite a very in- tensive country-wide sampling pro- gramme (Hanson and Gauld, 1995), spe- 216 Journal of Hymenoptera Research cies of Hytnenoepimecis are rather seldom collected in Costa Rica. Three species have been recorded (Gauld et al. 1998), but re- cently a fourth, undescribed species has been found on the Pacific coastal plain. Unusually for parasitoids, the majority of the Costa Rican Hytnenoepimecis specimens in collections (25 out of 36) have been reared, rather than field-collected. Hosts are known for all of the four Costa Rican species. Hytnenoepimecis tedfordi Gauld has not uncommonly been reared by W.G. Eberhard as a parasitoid of Leucauge mar- iana (Keyserling) (Tetragnathidae) (Gauld 1991), several specimens of H. robertsae Gauld have been found in Panama para- sitizing Nephila clavipes (L.) (Tetragnathi- dae) (Fincke et al. 1990) and a single indi- vidual of the apparently rare H. heidyae Gauld has also been reared by W.G. Eber- hard at La Selva, from Cyrtophora nympha (Simon) (Araneidae). The new species has been reared from by W.G. Eberhard from Plesiometa argyra (Walckenaer) (Tetrag- nathidae). As mentioned above, the new species differs strikingly from all previously de- scribed Hytnenoepimecis in possessing a number of plesiomorphic features. It is de- scribed below, after a key which will fa- cilitate its separation from other, sympat- ric species. KEY TO SPECIES OF HYMENOEPIMECIS PRESENT IN COSTA RICA 1. Fore wing uniformly blackish; submetapleural carina discernible, reaching at least 0.5 times length of metapleuron (Fig. 2); epicnemial carina present ventrally, although not laterally extended far onto mesopleuron argyraphaga Gauld, sp. n. - Fore wing more or less hyaline, at most slightly infumate distally; submetapleural carina more or less absent (Fig. 3), at most discernible as a small denticle on extreme anterior margin of metapleuron; epicnemial carina only discernible as weak swellings ventrally either side of mid-line 2 2. Metasoma with tergites all entirely black; sternite I with a low, rounded swelling poste- riorly; pronotum long, so that distance from tegula to head is about 0.6 times distance from tegula to hind margin of propodeum tedfordi Gauld - Metasoma with tergites predominantly orange; sternite 1 with a large acute or nasute protuberance near posterior margin (Fig. 3); pronotum exceptionally long, so that distance from tegula to head is greater than 0.7 times distance from tegula to hind margin of propodeum 3 3. Hind coxa and femur orange; sternite I with an acute, thorn-like ventral projection; female with ovipositor 1.0-1.2 times as long as hind tibia; hind leg slender, with tibia and tarsus combined, more than 0.9 times fore wing length robertsae Gauld - Hind coxa and femur extensively black; sternite I with a high laterally compressed nasute ventral protuberance; female with ovipositor 1.3-1.4 times as long as hind tibia; hind leg fairly stout, with tibia and tarsus combined about 0.6 times fore wing length heidyae Gauld Hytnenoepimecis argyraphaga Gauld, head in dorsal view with gena long, sp. n. strongly but evenly narrowed behind eyes; ocelli of moderate size, the lateral Female.— lower face elongate 0.8 times one separated from eye by about 0.7 times as broad as high (from clypeofacial suture jts own maximum diameter; lower end of to base of antenna), flat, centrally smooth occipital carina only very weakly raised, and impunctate, laterally with fine setifer- not produced mesally to approach its ous punctures bearing long fine hairs; counterpart on the midline. Pronotum Volume 9, Number 2, 2000 217 Fig. 2-3. 2, Hymenoepimecis spp. 2, //. argyraphaga, head, and mesosoma, lateral. Note the comparatively plesiomorphic body shape, in comparison with Fig. 3. 3, H. robertsac, head, and mesosoma, lateral, showing typical body form of members of this genus. long so that distance from tegula to head present ventrally, weak but extending is about 0.6 times the distance from tegula across the ventral portion of the mesoso- to hind margin of pronotum; scutellum in ma so that it is just visible in profile; me- profile strongly convex; mesopleuron tapleuron quite convex, smooth and pol- smooth and polished; epicnemial carina ished, glabrous; submetapleural carina ex- 218 Journal of Hymenoptera Research tending at least 0.5 of length of pleuron, i.1996 (Eberhard) (Natural History Muse- usually more or less complete; propo- urn, London). Paratypes: 3 9, 3 6, same deum smooth, laterally with very fine se- locality as holotype (Eberhard) (American tiferous punctures. Fore wing length 8.0- Entomological Institute, Natural History 8.5 mm; cu-a slightly distal to base of Museum, London and INBio, Santo Do- Rs&M; Irs-m about 0.6 times as long as mingo, Costa Rica). abscissa of M between 2rs-m and 2m-cu; Remarks. — Hymenoepimecis argyraphaga hind wing with abscissa of Cul between may easily be recognized in Costa Rica by M and cu-a 0.3-0.4 times as long as cu-a. its black wings. I have seen no other spe- Hind leg moderately long and slender cies, described or undescribed with uni- with tibia plus tarsus 0.6 times the fore formly black wings, although several wing length; hind tarsal claws short, with South American species, including H. het- a deep basal lobe (this condition is com- eropus (Kriechbaumer) have black and yel- mon to all Costa Rican species, but in low patterned wings. Hymenoepimecis ar- South America a large number have the gyraphaga is also the only species I have claw unusually long, and the lobe low, seen in the genus with discernible epic- sometimes tooth-like). Metasoma moder- nemial and submetapleural carinae, and ately slender, tergite I 1.4-1.5 times as unlike other species it does not have the long as posteriorly broad, centrally evenly lower end of the occipital carina produced convex, with lateral carinae present only mesally more or less to meet its counter- anteriorly bordering the anterior concavi- part medioventrally and partially close the ty; sternite I with a low rounded medio- oral fossa. The possession of these pie- ventral prominence; tergite II 1.1-1.2 times siomorphic features strongly suggests H. as long as posteriorly broad, with weak argyraphaga is one of the more basal spe- oblique grooves anterolaterally; tergite III cies in the genus. This is supported by oth- about 1.2 times as long as posteriorly er features. H. argyraphaga has short, deep broad with a median anterior swelling, hind tarsal claws, like Acrotaphus, whereas centrally glabrous and with scattered hairs most South American species have highly around the periphery of the tergite; ter- modified long, low claws with a tooth-like gites IV-V similar in sculpture and pilosi- basal lobe. Additionally sternite I of this ty; ovipositor 1.0-1.1 times as long as hind species is not modified but it generally has tibia. a thorn-like protuberance in other species. Head black with mouthparts yellowish Furthermore, although enlarged, the fore brown; antenna blackish; mesosoma or- legs of H. argyraphaga are not as massive ange-brown; metasoma with anterior two as many of the apparently more derived or three tergites orange anteriorly and species in the genus. broadly infuscate posteriorly, the area of These preliminary suggestions about infuscation increasing in extent and inten- both the phylogenetic position of H. argyr- sity on each tergite progressively towards aphaga within Hymenoepimecis and of this the hind end, which has the tergites black; genus with respect to others in the genus- ovipositor sheath black. Anterior two group, have important implications for pairs of legs orange-brown, the hind legs understanding the evolution of biological blackish, with bases of coxae brownish, traits within this uniquely adapted group Wings blackish infumate, pterostigma and of ichneumonids. veins black. Male.— similar to female in structure ACKNOWLEDGMENTS and colour; claspers black. . ., . D.„ „ , , , ,, . , , . . r 1 thank bill Eberhard for allowing me to study nis Material examined.— Holotype 9, Costa reared specimens. David x. Wahl and l are collabo- Rica, Puntarenas Province, Parrita, 20 m, rating closely in an attempt to unravel the phyloge- Volume 9, Number 2, 2000 219 netic interrelationships of the Pimplinae, and the dis- cussion above has benefited from his insight. I thank David Wahl and Andy Bennett for their comments on the manuscript. LITERATURE CITED Ashmead, W. H. 1900. Classification of the ichneu- mon flies, or the superfamily Ichneumonoidea. Proceedings of the United States National Museum 23: 1-220. Brulle, A. 1846. Hymenopteres, 4. In: Lepeletier de Saint-Fargeau, A. (ed.) Histoirc Naturelle des In- sectes. Paris. 680pp. Eberhard, W. G. 2000. The natural history and behav- ior of Hymenoepimecis argyraphdga (Hymenoptera: Ichneumonidae) a parasitoid of Plesiometa argyra (Araneae: Tetragnathidae). Journal of Hymenoptera Research 9: 220-240. Fincke, O. M., Higgins, L. and Rojas, E. 1990. Para- sitism of Nephila chwipes (Araneae, Tetragnathi- dae) by an ichneumonid (Hymenoptera, Polys- phinctini) in Panama. Journal of Arachnology 18: 321-329. Fitton, M. G., Shaw, M. R. and Gauld, I. D. 1988. Pimpline ichneumon-flies (Hymenoptera, Ich- neumonidae, Pimplinae). Handbook for Identifica- tion of British Insects 7(1): 1-110. Gauld, I. D. 1991. The Ichneumonidae of Costa Rica, 1. Memoirs of the American Entotnological Institute 47: 1-589. Gauld, I. D., Ugalde-Gomez, J. A. and Hanson, P. E. 1998. Gui'a de los Pimplinae de Costa Rica. Re- vista Biologia Tropical 46 (Supplement 1): 1-189. Hanson, P. E. and Gauld, I. D. 1995. The Hymenoptera of Costa Rica. Oxford University Press, Oxford. 893 pp. Nielsen, E. 1923. Contributions to the life history of the pimpline spider parasites (Polysphincta, Zag- lyptus, Tromatobia) (Hym. Ichneum.). Entomologis- ke Meddelelser 14: 137-205. Shannon, R. C. 1913. Epimecis wiltii Cresson and its host. Proceedings of the Entomological Society of Washi)igton 15: 162. Townes, H. 1969. Genera of Ichneumonidae 1. Mem- oirs of the American Entomological Institute 11: 1- 300. Townes, H. and Townes, M. 1960. Ichneumon-Flies of America North of Mexico: 2. Subfamilies Ephialtinae, Xoridinae, Acaenitinae. United States National Museum Bulletin 216 (2): 1-676. Viereck, H. L. 1912. Descriptions of five new genera and twenty-six new species of Ichneumon-flies. Proceedings of the United States National Museum 42: 139-153.' Wahl, D. and Gauld I. D. 1998. The cladistics and higher classification of the Pimpliformes (Hy- menoptera: Ichneumonidae). Systematic Entomol- ogy 23: 265-298. Wasbauer, M. S. 1995. Pompilidae, pp. 522-539. In: Hanson, P.E. and Gauld, I.D. The Hymenoptera of Costa Rica. Oxford University Press, Oxford. 893 pp. Yu, D. and Horstmann, K. 1997. Catalogue of world Ichneumonidae (Hymenoptera). Memoirs of the American Entomological Institute 58: 1-1558. J. HYM. RES. Vol. 9(2), 2000, pp. 220-240 The Natural History and Behavior of Hymenoepimecis argyraphaga (Hymenoptera: Ichneumonidae) a Parasitoid of Plesiometa argyra (Araneae: Tetragnathidae) William G. Eberhard Smithsonian Tropical Research Institute, and Escuela de Biologia, Universidad de Costa Rica, Ciudad Universitaria, Costa Rica Abstract. — Larvae of the koinobiont ectoparasitoid Hymenoepimecis argyraphaga Gauld used a series of different adaptations as they matured to hold onto the abdomen of their host spider, Plesiometa argyra, its web, and the larval cocoon: the first instar did not leave the egg chorion, which was glued to the spider by the female wasp when she oviposited; the second instar used two pairs of ventral abdominal protuberances to help hold onto both the first larva's molted cuticle and to what was probably a sheet of coagulated spider hemolymph that adhered to the larva and to wounds that it made on the spider's abdomen; the early final instar grasped the shed second instar cuticle that remained attached to the coagulated hemolymph with the ventral surface of its abdomen; and the late final instar used a row of mobile dorsal protuberances with sharply curved spines that grasped lines of a unique web that the larva induced the spider to spin just before killing it, and then the larva's own cocoon silk. The pupa used a pair of toothed protuberances at the tip of the abdomen to stay at the upper end of its cocoon. Other aspects of the wasp's biology that are described include infanticide by adult females; aculeate-like lack of use of the ovipositor to oviposit; manipulation of host web-spinning behavior, apparently by means of a fast-acting larval secretion with long-term effects; manipulation of host bleeding; alternative tactics in attacks on spiders; use of pheromones by females to attract males; cocoon spinning behavior; and a bias to parasitize female rather than male spiders. Although Ichneumonidae is undoubt- species resembles that of some European edly one of the largest of all animal fam- polysphinctines. It is an external koino- ilies, remarkably little is known about the biont on a spider, the tetragnathid Nephila behavior of the larvae. Excluding studies clavipes (L.). The female temporarily par- of foraging behavior, adult behavior is alyzes the host by stinging it in the ceph- also poorly studied (e.g., Hanson and alothorax, and then glues an egg on its ab- Gauld 1995). The neotropical polysphinc- domen. Spiders with a wasp egg or a tine pimplines are no exception. The more young larva are active, and build appar- derived polysphinctines are known to be ently normal prey capture webs and feed koinobiont ectoparasitoids of spiders while the larva feeds by sucking the spi- (Gauld 1995, Wahl and Gauld 1998), and der's hemolymph and gradually matures, several European species were observed The spider's webs become more irregular in careful detail by G. C. Bignell (1898) and reduced one to two days before the and E. Nielsen (1923, 1928, 1929, 1935). larva kills it and constructs its pupal co- There is apparently only a single study of coon, which is attached to the spider's a neotropical species, that of Fincke et al. web. A second species, H. tedfordi Gauld, (1990) on Hymenoepimecis robertsae Gauld parasitizes another tetragnathid spider, (for probable identification see Gauld etal. Leucauge marinae Keyserling (Gauld et al. 1998). The general natural history of this 1998), but nothing more is known about Volume 9, Number 2, 2000 221 Table 1. Degree of development of eggs in the ovaries (6-7 ovarioles/ ovary) of female wasps of different ages (N = number of wasps; W = white eggs apparently ready to be laid, Y = yellow, still-immature eggs; uric acid determined on arbitrary scale of 0-3). Age of wasp outside cocoon (hrs) Mean length and coll >r - Fig. 1. Diagrammatic illustration of a female wasp hanging on paralyzed spider at the hub of its web and ovipositing at the point indicated (drawn from field notes; it is not certain whether the wasp's tarsi held the spider, as in the drawing, or the web just above it). terior and dorsal surface of the spider's abdomen. These probing movements probably served to locate, perhaps to sting, and to dislodge the eggs or larvae of previous wasps that had attacked this spider. In one case, repeated probing movements of the ovipositor in the vicin- ity of a first instar or early second instar larva may have included one or more in- sertions of the ovipositor into the larva, and they finally resulted in the larva being levered off of the spider's abdomen and onto the middle portion of the wasp's ovi- positor. The wasp then knocked off the larva to the ground with cleaning move- ments, and soon afterward laid an egg of her own. In two other cases an egg (iden- tity confirmed by subsequent collection) was moved onto the middle of the ovi- positor and then fell to the ground. Of 16 parasitized spiders collected in 1999 with eggs or first instar larvae on their abdomens, four gave further evi- dence that larvae are sometimes removed: there were one or more groups of larval feeding scars on portions of the abdomen that were inaccessible to the current larva in two cases, and there were larval feeding scars but no larvae and only an egg in two others. In contrast, only one spider was doubly parasitized, carrying both an egg and a second instar larva. Similar data from 2000 gave even more dramatic evi- dence of infanticide. Of 55 mature female spiders, nearly half (26) had at least one patch of feeding and bleeding scars on her abdomen (the total of additional patches was 43; the maximum on one spider was four). Only three spiders were doubly par- asitized. It should be noted that these data undoubtedly underestimate the frequency of infanticide, because removal of eggs from hosts cannot be detected using feed- ing scars. Finally the wasp oviposited. Holding her ovipositor sheaths elevated dorsally and her ovipositor pressed against the an- terior surface of the spider's abdomen (Fig. 1), she pressed the tip of her abdo- men near the surface of the spider's ab- domen briefly. The egg emerged from the tip of her abdomen (and not from the tip of her ovipositor), and adhered to the spi- der's abdomen. Within about 30 sec after ovipositing the wasp flew away. In no case did a wasp give any sign of attempt- ing to feed on the spider. Although eggs were generally placed on the anterior dor- sal surface of the spider's abdomen, the exact sites varied widely (Fig. 3). The oval egg was glued tightly on its ventral side to the spider's abdominal cuticle. Two eggs which were observed being laid and then inspected periodically hatched be- tween 48 and 72 hours later. An incomplete observation of one inter- action indicated that the wasps have an alternative hunting strategy that depends on deceiving the spider. When they were first encountered, the spider was resting at the edge of its orb, and the wasp was hanging immobile from radii in the free zone near the hub, facing downward with 224 Journal of Hymenoptera Research Fig. 2. Evidence of two infanticides. The anterior surface of the abdomen of a P. argyra spider bears a wasp egg, and two arrays of feeding and bleeding scars (dark spots) that were presumably produced by wasp larvae that had hatched from eggs laid previ- ously, but that were then removed when subsequent female wasps attacked this spider. most or all of her legs extended stiffly. The wasp appeared to be either dead or para- lyzed, and a gentle nudge of the wasp with my finger confirmed that she was completely immobile. She remained mo- tionless until the spider returned to the hub about 5 min later. As the spider ar- rived at the hub, however, there was a sudden tangle of legs and it quickly be- came clear that the wasp's ovipositor was inserted near the spider's mouth. Soon the Lateral Dorsal Fig. 3. Schematic representation of 125 sites of the anterior ends of egg and the bleeding scars of the larva on the spider's abdomen in lateral and dorsal views. spider became immobile, and the wasp, which had fooled both the spider and me by playing dead, proceeded to oviposit. Wasps also showed a certain flexibility in oviposition behavior. When one spider began to move after the wasp had spent several minutes attempting to remove an egg, the wasp moved to its anterior end and apparently stung it again, then re- sumed egg removal. Failed wasp attacks illustrate possible kinds of selection on spiders to avoid at- tacks. In one case the spider's orb was in- clined so that much of its surface was be- tween the wasp and the spider as the wasp hovered above and to the side of the spider; when the wasp finally struck, it hit and was arrested by the orb before reach- ing the spider, the spider moved away, and the wasp flew on. In two other cases (one first seen after the interaction had al- ready begun), the spider hung from its dragline about 10-15 cm below the hub, and when the wasp struck from above it hit and was arrested by the orb, and thus Volume 9, Number 2, 2000 225 did not reach the spider. The wasp then hovered nearby, and the spider twice climbed very rapidly to the hub but im- mediately dropped again. Finally the spi- der dashed to the edge of the web, and the wasp flew on. The presence of web threads between the spider and the wasp that were not right next to the spider thus appeared to reduce the likelihood that a wasp's attack would succeed. A final fail- ure occurred when a gentle wind appar- ently made it more difficult for the wasp to hover steadily near the spider, and she eventually crashed into the web and then flew away. Windier sites may thus be saf- er for spiders. Larvae First instar. — The first instar larva ap- parently burst open one end of the egg, but only its anterior end emerged from the chorion. Its posterior end remained lodged inside the chorion, and thus at- tached to the spider. The first instar larvae possessed neither the ventral nor the dor- sal structures used by later instars to hold onto the spider and its web. A small brown spot or feeding scar (see below) ap- peared on the spider's abdomen just be- yond the edge of the chorion of recently emerged larvae. The subsequent gradual accumulation of feeding scars on the spi- der's abdomen, which were always re- stricted to the vicinity of the larva's head, indicated that the larva fed at small holes it made in the spider's abdomen (Figs. 2, 4). As the larva grew, it gradually pro- truded more and more from the egg cho- rion. The head, the entire thorax, and the first two to three abdominal segments were free by the time the larva was ready to molt to the second instar, and by then there were 8-10 feeding scars on the spi- der's abdomen. The first instar lasted be- tween 58 and 69 hours (N = 2). First instar larvae were able to remain attached when the spider molted, as evi- denced by two pale, soft newly-molted adult spiders each having a first instar lar- molted cuticle feeding scar larval tap chorion feeding scar bleeding scar Fig. 4. Pattern of brown scars on the abdomen of a spider earning a second instar larva in lateral view (above) and ventral view, looking outward through the spider's cuticle (below). The sites of insertion of the larva's taps were not visible, and were deter- mined by subsequent dissection. va attached to its abdomen. In both cases (and in two other spiders) a portion of the cuticle of the previous instar (usually the dorsal portion of the cephalothorax plus a wrinkled portion of the abdomen) was at- tached to the spider's abdomen, apparent- ly at a bleeding scar (Fig. 4). Second instar. — The transformation from first to second instar was not witnessed, and the probable series of events was re- constructed from preserved specimens. The newly molted second instar larva was completely outside the collapsed egg cho- rion. The larva's ventral surface rested on the flattened, shed first instar larval cuti- cle, and this cuticle in turn rested on a large, stiff sheet of brown amorphous ma- terial (Figs. 4, 5) (the "saddle" of Nielsen 226 Journal of Hymenoptera Research Fig. 5. Micrographs of apparent feeding scars (A and B) and saddle material (C) of a second instar larva on the corrugated abdominal cuticle of a spider. Many feeding scars were produced into more or less conical protuberances of different shapes (A). A closeup view (B) shows that the scar material (presumably coagulated hemolymph) flowed over the corrugated abdominal cuticle around the perforation before hardening. The "saddle" material is amorphous, and is not a shed larval skin. 1923). The empty egg chorion was on the inner surface of the saddle that contacted the surface of the spider's abdomen. The saddle was probably composed of coagu- lated hemolymph, and had a different form in each parasitized spider. It was tightly attached to the spider's abdomen near the open end of the egg. The anterior end of the shed larval cuticle was near the anterior edge of this sheet, indicating that molting probably involved a split of the first instar's cuticle along the dorsal mid- line of most of the larva's body, rather than a rearward sloughing of cuticle as oc- curred at pupation (see below). The saddle adhered to both the ventral surface of the larva on its outer side, and to the spider's abdominal cuticle on its in- ner side, and could be peeled away intact from both except at the central attachment area. Here it was attached tightly to the spider's abdomen at several brown spots that were similar to the feeding scars men- tioned above but larger (Fig. 4). In some cases projections of the saddle extended into the larva's intersegmental grooves (Fig. 4), indicating that the saddle had been a liquid at some time after the larva molted. Two pairs of protuberances ("taps" of Nielsen 1923) on the ventral surface of the larva's segments 8 and 9 were inserted deeply into the saddle (and perhaps also the shed skin) (Fig. 4). They adhered so tightly to the saddle that it was Volume 9, Number 2, 2000 227 difficult to separate the larva from it with- sue after the specimen was preserved in out damaging the taps. alcohol, were smooth and dark, similar to The second instar larva apparently con- those of feeding scars, tinued feeding as before, as brown feeding The number of second instar larvae pri- scars gradually accumulated on the side of or to their final day (below) in collections the spider's abdomen on which the larva's made in 2000 was approximately equal to head rested. Many feeding scars were pro- the number of eggs and of first instar lar- duced externally into more or less conical vae (62 second instars, 73 eggs, 57 first in- shapes whose tips pointed anteriorly on stars). Assuming that the egg stage lasts the abdomen (toward the larva's head) about 2.5 days and that the relative num- (Fig. 5). The spider's pattern of coloration, bers of the different immature stages were which was due to soft tissues under the fairly constant over the space of a few transparent abdominal cuticle, was often weeks (supported by the similarity in (though not always) intact in areas with numbers in the surveys made on 28-30 feeding scars. Thus larvae probably usu- Jan. and 9-10 Feb, 2000 — -Chi2 = 2.7, df = ally consumed hemolymph, rather than 3, p > 0.4), this suggests that the second digesting other internal tissues. instar normally lasts two to three days in When the spider's abdominal tissue was nature. It can also last much longer, how- dissected away, there were no inward ex- ever. The second instar lasted 46 days in tensions of either the feeding scars or the one case in which the spider was kept cap- attachment spots; all were relatively five with only infrequent feeding. Spiders smooth on their inner surfaces. Smaller carrying first and second instar larvae oc- feeding scars were concentrated nearer the curred on apparently normal orbs in the egg (where the larva presumably fed field (Eberhard in prep.), when it was smaller — Fig. 4). The feeding The true number of larval instars is not scars had a remarkably regular distribu- certain. Fitton et al. (1988) speculated that tion, with larger spaces between larger all pimplines may have five larval instars, scars (Figs 2, 4). on the grounds that Pimpla does, with "the Saddle-like puddles of hemolymph did middle three being very similar and hard not form at experimental wounds (ap- to distinguish". If so, then the stages des- proximately 0.1 mm diameter) made by ignated here as first and second may ac- puricturing the abdominal cuticle with a tually represent three or four stages that I fine pin, despite the fact that these holes was unable to distinguish. The distribu- were larger than feeding scars (about 40 tion of the widths of the head capsules of |xm in dia — see Fig. 5). In nearly all cases 81 larvae (Fig. 6) did not clarify this. There the hole was immediately sealed by a were two peaks within the range of sizes small plug when the pin was withdrawn, classified here as first instars, while sev- This plug, which was little more than the eral final instar larvae (which can be rec- diameter of the hole, was initially liquid ognized inequivocably by the dorsal tu- when touched with the pin, but hardened bercles covered with curved spines) were to a solid within 10-15 s, and darkened to substantially smaller than several others a light brown color. In two cases in which which were clearly in the previous instar a small sheet (up to about 0.3 mm in di- (the curved spines were visible, but were ameter) of hemolymph emerged from the covered with a transparent layer of cuti- wound before a plug formed, the sheet cle). It may be that head capsule width is did not turn dark brown, but instead ac- not constant within an instar, as some quired a nearly transparent golden color, sclerites may be connected by elastic The inner surfaces of these wounds, re- membranes, vealed by dissecting away underlying tis- Final ("third") instar. — There were sev- 228 Journal of Hymenoptera Research CD _Q 10 5 "first instar" "second instar" ^ ' mature second instar confirmed third instar I 0.2 0.4 0.6 Width head capsule (mm) Fig. 6. Distribution of head capsule widths of 81 larvae, and approximate limits of classifications of larval instars used in this study. "Confirmed" final (supposed third) instar larvae had hooked spines on dorsal tubercles; "mature" second instar larvae had hooks that were clearly visible under the larva's dorsal cuticle. eral differences between second and third (final) instar larvae. The dorsal surfaces of eight segments (3-10 posterior to the head) of the final instar, which had been smooth in previous instars, each had a re- tractable two-lobed tubercle whose tips were covered with curved spines (Fig. 7). There was a pair of taps on the ventral surface of segments 8 and 9 as in the sec- ond instar, but they were free (Fig. 8), and not embedded in the saddle. The shed cu- ticle of the second instar formed a com- pact sheet bent into a cup. It adhered tightly to the saddle, apparently where the taps of the cuticle of the second instar were inserted. The saddle, in turn, still ad- hered tightly to the spider's abdomen. The larva grasped the posterior edge of the sheet of second instar cuticle between its final (13th) segment and the bulging ven- tral margin of its penultimate (12th) seg- ment (Fig. 8C). In addition, the larva's head capsule was substantially different (Fig. 9). The final instar was relatively brief. All larvae raised to final instars or collected in the field as swollen second instars or as final instars killed the spider the following Fig. 7. Hooks on a dorsal tubercle of a final instar larva that have snagged tangled spider silk. Volume 9, Number 2, 2000 229 coagulated hemolymph head capsule cuticle of 2nd instar head capsule coagulated hemolymph spider abdomen Fig. 8. (A) Attachment of the final instar larva to the spider's abdomen, seen in lateral view. (B) Dorsal view of the cup-shaped mass of larval cuticle. (C) Lateral view of the posterior tip of a larva removed from its attachment to the spider, showing the protruding ventral surface of segment 12 that gripped the edge of the cup of larval cuticle. night. An estimate from field collections of swollen last day second instar larvae and early final instars on spiders in 2000 was in accord with the estimate of a duration Fig. 9. Antero-lateral view of head capsule of final instar larva. of only one day or slightly less (only 18 individuals, as compared with 73 eggs and 57 first instars). The final day in the spider's life was eventful. At least some spiders built an orb of apparently normally design and size in the morning, but during the day the second instar larva grew to an esti- mated one quarter to one third of the vol- ume of the spider's abdomen. At about 23: 00-01:00 the spider built a modified "co- coon web" of a few highly reinforced ra- dial lines that was especially appropriate to support the cocoon the larva would build the next evening. In two cases the larva repeatedly extended its body nearly straight while the spider built the cocoon web. The cocoon web and the behavior em- ployed to build it are described elsewhere 230 Journal of Hymenoptera Research (Eberhard in press, in prep.). Briefly, the spider used one portion of one subroutine of frame line construction over and over to build a small number (mean 3.8 + 1.4, N = 39) of radial lines, each of which is a cable composed of many individual lines. This web did not resemble any prey cap- ture, resting, molting, or egg sacs webs normally built by P. argyra. Experimental removal of larvae showed there to be a complex, long-term effect on the spider's behavior that is probably mediated chem- ically. Typical cocoon web construction followed in three cases when a swollen second instar larva was removed from a spider that had been kept in a confined space until about midnight, when it would probably have begun cocoon web construction. These spiders were still alive the next day, and the next evening they each built a second typical cocoon web. Thirteen other spiders from which the larvae were removed between 22:00 and 02:00 built structures that were neither normal orbs nor cocoon webs. Three of these spiders were observed building. They placed radial lines from the hub to the edge using the behavior used to con- struct typical cocoon webs (Eberhard in prep.), but also broke and reeled up these lines while moving back toward the hub. The final products were sparse networks of more or less radial lines in which there were large accumulations of reeled up silk (fluff) near the hub. All of these spiders were also alive and active the next morn- ing. Soon after it finished the cocoon web, the parasitized spider became immobile. All of nine spiders were dead or complete- ly immobile by 03:00 (in one checked un- der the microscope, the heart had stopped beating), and the larva had lifted the an- terior portion of its body above that of the spider to grasp the lines of the web with its dorsal protuberances (see below). The posterior portion of its body remained at- tached to the saddle. The larva proceeded to suck the spider dry over the course of Extend Retract Fig. 10. Diagrammatic representation of the move- ments of the curved spines at the tip of a turret that caused lines to be snagged when the tubercle was everted (above), and release of lines when it was re- tracted (below). the morning, feeding first on the spider's abdomen and then on its cephalothorax. The mechanism by which the larva grasped and released lines with its hook- covered dorsal tubercles was revealed by observations under a dissecting micro- scope. The tubercles were extremely mo- bile, and could be extended so that the distance of their tips from the dorsal sur- face of the larva's body was up to about one third of the diameter of the body. They could also be retracted rapidly so that the entire tubercle and all of its hooks disappeared completely into a pocket on the dorsal surface of the larva's body. Be- cause the spines near the tip of each tu- bercle were sharply curved, eversion of the tubercle resulted in a grasping effect, as lines were snagged by the curved spines (Figs. 7, 10). The spines released their holds on lines when the tubercle was retracted into the larva's body. Volume 9, Number 2, 2000 231 Coordination of tubercle movements was complex. When the larva moved its anterior end, the anterior-most three tu- bercles often contacted the web only spo- radically. When the tubercles were out of contact they were often everted and re- tracted simultaneously. In contrast, the more posterior tubercles, which usually held onto threads and supported the lar- va's weight, moved less often and sequen- tially. When the larva moved its entire body forward or backward, each tubercle that was holding silk released its hold by retracting, everted toward the next tuber- cle where it seized silk, and the next tu- bercle then released its hold and was everted toward the next, and so on. These stepping movements swept along the lar- va's body rapidly, and it sometimes stepped with several tubercles in a second. Feeding by a final instar larva on a dead spider was observed under a dissecting microscope. The first stage of a feeding bout involved apparent searching and hooking of the mouthparts against the surface of the abdominal cuticle. I was not able to discern any pattern to these move- ments, nor any responses to the brown feeding scars already present. Finally, sometimes after minutes of such search- ing, the larva's head came to rest at a par- ticular site where it apparently began to produce a hole. After a minute or so, the larva began to suck. The rhythmic movements of its head were reminiscent of those of a nursing hu- man infant (Fig. 11). Approximately once every second the larva pulled its head slightly away from the abdomen without breaking contact with its mouthparts, then sprang slightly toward it again. In one case it was possible to see the flow of the spider's tissues through the transparent cuticle of the abdomen as the larva sucked. Clumps of abdominal tissue flowed steadily into the larva's mouth. Two timed feeding bouts lasted about 30 min. They ended when the larva pulled its head away and rested immobile for sev- spider abdomen Fig. 11. Diagrammatic representation of the head of a feeding larva; the stippled areas pulsed as the larva sucked. eral minutes. When the larva withdrew its head after feeding, there was a small amount of clear liquid on the surface of the spider which quickly dried up; no hole was visible, nor did a brown spot form. On three occasions it was clear that the larva interrupted sucking and that a clear liquid flowed from its mouth into the spi- der for several seconds. This liquid formed a small pool around the point where the larva's head contacted the ab- domen, and it also flowed under the cu- ticle, as deduced from brief movements of spider tissue away from the larva's mouthparts. Presumably the clear liquid contained digestive enzymes. When feeding ended, the larva freed it- self from the now more or less empty but intact cuticle of the spider, and it dropped to the ground below. This process was not observed directly, and it was not clear whether the larva actively unhooked the spider's tarsal claws from the web. The saddle was still attached to the discarded carcass of the spider, and bore the imprint of the larva's last two abdominal segments (Fig. 8). In two cases the discarded carcass of the spider represented about 50% of the larva's weight: larvae which weighed 15.7 and 26.2 mg discarded carcasses that weighed, respectively, 6.1 and 10.4 mg. The now somewhat greenish larva hung motionless, curled ventrally as it held the 232 Journal of Hymenoptera Research web with its dorsal tubercles, for the rest in snagging the line on the tubercle. The of the daylight hours. Larvae in the field larva then extended the rear portion of its were remarkably coordinated in killing body downward, thus probably pulling and consuming their hosts. All but one of more silk, and it moved its head back up 13 final instar larvae found before 13:00 to contact the spider web above and make were still feeding on the dead host, while another attachment there. The usual du- all of 11 found between 13:00 and 17:00 ration of one complete cycle was about 5- had dropped the spider and were resting 10 s. immobile at the hub of the cocoon web. By repeating this sequence of spinning Cocoon construction began soon after movements over and over, the larva grad- darkness fell (about 18:40). The process of Ually produced a bundle of lines that ran first attaching the cocoon's suspension line from the attachments to the spider web to the spider's web, then extending the down to the rear end of its own body and line below the web, and then forming the back up to the web again. This bundle walls of the cocoon was accomplished as would form the suspension line of the co- the larva slowly inched backward over a coon The suspension line was lengthened period of up to an hour or more. The larva when me larva grasped the bundle with maintained a hold on silk lines with its the dorsal tubercles, moving rearward dorsal tubercles at all times. At first it held along the bundi6/ and men resuming spin- onto the silk of the spider's cocoon web, ning moVements. Eventually the upper at- then later onto the suspension line of its tacnments of spinning movements were to cocoon. fae suspension line itself rather than to the The larva produced a silk line (or lines?) SDider's web from its head by pulling away from a ' The final ' q( CQCOOn construction point where the line was attached, and algo involved a si le ttern repeated then attached this line to others by tap- , /T:. 10T3N TU , , , , , . . J _/ over and over (Fig. 12B). The larva moved pine or rubbing its head against them. The .. , , . , . ■. , £ ., , „ 5. ,. ° , , ° „ , its head to touch the edge of the bundle first lines were attached repeatedly to the , ,. , , .. i j .. , , , , , . j , , i of lines already spun near its lower end, lines at the hub of the spider s web, and , „ , i . j- , j ,, i j- j ^ j then pulled away a short distance and were often somewhat dispersed. Gradu- , r , , , i i , ,, ,, , , , ... then moved back to touch the edge again ally they condensed into a single multi- ,. , . , , , , , . , . j j v ii. i jui slightly farther up, and repeated this be- stranded line as the larva moved back- & J , , i i ward a few millimeters. The suspension ^avior until [t reached the suspension line, line of the cocoon was produced by a sim- then bent downward again to begin the pie sequence of movements repeated over next series of attachments near the bottom and over (Fig. 12A). First the larva at- ed8e of the bundle- The larva thus Srad" tached its line to the spider's web, and ually extended the bundle laterally to drew out a line by moving its head down- form a bag-like sheet that enclosed the ward. Usually it paused immobile for sev- posterior portion of its body. Later the bag eral seconds, and then moved its head far- was gradually extended upward to en- ther, toward the posterior end of its body, close the upper portion of its body also, which was bent anteriorly. The head usu- The bag was closed by addition of lines to ally dabbed or scraped repeatedly against its inner surface, about 4-5 hrs after co- the posterior portion of the body, gradu- coon construction began. Once the bag ally touching points more and more to- was closed it was more difficult to observe ward its dorsal surface. During the last the larva's behavior, but it was clear that few scrapes it often dragged its mouth- it sometimes turned 180° to face down- parts across its rearmost dorsal tubercle, ward and add more lines to the bottom of Probably these movements often resulted the cocoon. Cocoon construction contin- Volume 9, Number 2, 2000 A cocoon web 233 B / cocoon web Is suspension line i Fig. 12. The sequence of movements as a larva built the suspension line of the cocoon (A ) and when it began the cocoon itself (B). 234 Journal of Hymenoptera Research ued through the night and the following morning. Cocoons spun in captivity were all very light yellow in color, and gradually dark- ened over the next day or so. Those kept in closed containers out of the light were especially pale. Some of those found in na- ture were dark yellow, while most were bright orange. About 36-65 hrs. after co- coon construction began, the larva ejected its meconium, which generally fell through the circular hole at the bottom of the cocoon (Gauld et al. 1998). One larva molted to a pupa between 4 and 5 days after killing its host. Pupa Pupal behavior could not be observed directly, except for the short series of rap- id dorso-ventral contractions of the abdo- men usually made when the pupa was disturbed. Indirect evidence suggests that the pair of toothed prominences near the tip of the pupal abdomen, which were im- mobile, served to engage the tip of the ab- domen against the inner wall of the co- coon, and thus hold the pupa at the upper end of its cocoon. All of seven living pu- pae checked by cutting a slit in the side of the cocoon were wedged at the upper end of the cocoon (the pupa occupied only an estimated average 83 ± 6% of the length of the cocoon). All pupae observed in co- coons were oriented with the head up- ward, and the circular emergence slit made by the adult was always at the up- per end of the cocoon. Presumably this position served to keep the pupa's poste- rior end away from the meconium or the shed larval skin that was sometimes pre- sent at the bottom tip of the cocoon. Four adults emerged 10-12 days after cocoon construction, so the pupal stage lasted about a week. A total of 26 males and 38 females were raised from field-collected cocoons (not significantly different from 50:50 with Chi2 Test). Other aspects of natural history Rates of parasitism. — Mature female spi- ders were approximately ten times more heavily parasitized than were mature males (42.9% of 203 females compared with 6.9% of 72 males in 1999; 66.4% of 125 females vs. 3.2% of 62 males in 2000). The rate of parasitism of mature females earlier in 1999 (27-30 Jan) was lower than that later (2-5 Feb.) the same year (26.5% of 83 females compared with 50% of 120 females — p<0.001 with Chi2). In contrast, the first survey in 2000 showed a higher rate for mature females than for the sec- ond survey that year (84.2% of 57 females on 28-30 Jan., vs. 51.5% of 68 on 9-10 Feb., p < 0.001 with Chi2). The rates of parasit- ism of mature males did not differ signif- icantly for the two surveys in either year (7.1% of 28 compared with 6.9% of 44 in 1999, 3.1% of 32 vs. 3.3% of 30 in 2000). Mature females were also more heavily parasitized than were immatures. In the first survey of 2000, 84.2% of mature fe- males, but only 50.8% of 63 penultimate females, 39.5% of 43 penultimate males, and 25.7% of ante-penultimate nymphs were paralyzed (all rates were lower than that for mature females, p < 0.001 with Chi2; ante-penultimates were less parasit- ized than penultimates, p = 0.033 with Chi2). Similarly, penultimate spiders showed less evidence of infanticide (25 cases in 73 spiders, compared with 43 cas- es among 55 mature females, p < 0.001 with Chi2). Enemies of the wasp. — Mortality in the co- coon was relatively low. Seven Conura of two different species, one in the immacu- lata group of the subgenus Ceratosmicra, and the other in the vau group of the sub- genus Conura (Chalcididae) were raised from 105 inhabited cocoons collected in the field in 1999, and 4 of 85 empty co- coons had Conura sp. pupal skins, giving a total rate of 5.8%. An eighth female Con- ura was captured after being first seen resting on the side of a cocoon in the field. Volume 9, Number 2, 2000 235 Two days later this cocoon contained a young pupa of H. argyraphaga, suggesting that the late larval or early pupal stage of the host was parasitized. Two chalcidids that pupated in captivity each had a dead H. argyraphaga pupa in the same cocoon, and three empty cocoons collected in the field that had a pupal cuticle of Conura also had the remains of a H. argyraphaga pupa. Other field mortality of stages in co- coons, presumably due to predators, was noted as removal of the entire cocoon from the cocoon web (8 cases), or com- plete removal of the cocoon's contents (as- sociated with a large hole in the side in two cases, and with multiple small ragged holes in two others). Two mature female spiders in the field carried dead first instar larvae, but the cause of death was not clear. Two adult H. argyraphaga, one male and one female, were found dead at the hubs of orb webs of the araneid Gasteracantha cancriformis (L.), despite the fact that the wasps seem to be chemically defended. Even recently emerged adults less than four hours old released a pungent odor when grasped between the fingers (neither large second instar larvae nor their host spiders had any perceptible smell or taste). One tug of war between a larva and a salticid spider at the hub of a new co- coon web resulted in the larva's loss of the dead spider, and nearly resulted in pre- dation on the larva. One further predatory event may have been an artifact, but it illustrates another potential danger to the wasp. A penulti- mate instar male spider with a swollen second instar larva was placed in a plastic bag in a more or less cramped position. An hour or so later, the spider was feed- ing on the anterior end of the larva, which was still attached posteriorly to the sad- dle. In no other case did a parasitized spi- der exhibit any behavior directed toward the larva on its abdomen. DISCUSSION Chemical manipulation of the host. — Lar- vae of H. argyraphaga manipulate both the behavior and the physiology of their hosts. The changes in the spider's behav- ior which resulted in the production of the otherwise unique, strong "cocoon" web that is particularly well designed to sus- tain the wasp's cocoon, involved induc- tion of the first steps of one subprogram of orb web construction that were repeat- ed over and over to the exclusion of others (Eberhard in press, in prep.). Experimental removal of larvae showed that these be- havioral changes occurred as a result of a fast-acting substance or substances with long-term effects introduced into the spi- der just before the larva molted to the final instar. It appears that the larva's induction of one type of building behavior and re- pression of others may represent separate effects, as their manifestation was partially uncoupled by early removal of the larva (Eberhard in prep.). Induction by the final instar during cocoon web construction seems unlikely, due to both the softness of the final instar's pale head capsule soon after molting, and the complete cocoon webs obtained after the mature second in- star was removed. In contrast, the death of the spider soon after the cocoon web was finished probably resulted from ma- terial injected by the final instar larva just after the cocoon web was finished, when it began to feed; spiders did not die when the larva was removed just prior to this molt. Manipulation of spider bleeding oc- curred when the larva molted from the first to second instar, and resulted in the production of the saddle. The form and the position of the saddle suggest that liq- uid hemolymph emerged in unusually large amounts from the large holes in the spider's abdomen and did not immediate- ly coagulate as usual (one saddle was about 0.9 X 1.5 mm, while the largest pud- dle of hemolymph produced by wound- 236 Journal of Hymenoptera Research ing with a minuten pin was only about were very small when they were found, 0.27 mm in diameter). These holes were apparently soon after the spider's molt, probably made just before the first instar Perhaps these larvae had not been hatched larva emerged from the egg chorion to long enough to inhibit the molt, molt, because the newly molted second in- Comparisons with related wasps. — There star larva would have had a soft head cap- are many points of similarity between the sule, presumably incapable of biting behavior and natural history of H. argyr- through the spider's cuticle. Judging by aphaga and H. robertsae (Fincke et al. 1990), the apparent difficulty that the larger final and with polysphinctines of other genera instar larva had in perforating the spider's (Bignell 1898, Nielsen 1923, Fitton et al. relatively tough cuticle, it probably took 1988, Gauld et al. 1998). Females of H. rob- the first instar larva many minutes to ertsae may also attack spiders at the hubs make these large holes. The spider's he- of their webs, sting the spider in the ceph- molymph must have coagulated only alothorax to produce a temporary paraly- slowly, and was evidently still liquid after sis (Fincke and colleagues witnessed only ecdysis occurred, since saddle material what were apparently aftermaths of at- sometimes flowed part way up the side of tacks, however), and lay an egg on the an- the second instar larva's cuticle (Fig. 4). terior surface of the abdomen. The female Liquid hemolymph may have helped the wasp also moves the ovipositor back and larva adhere to the spider during the del- forth over the spider's abdomen (for up to icate period after it had abandoned its egg five min) prior to ovipositing, and thus but was still a first instar and thus lacked may also remove previously deposited grasping structures. The extensive flow of eggs or larvae. Fincke et al. (1990) found hemolymph and the long delay before it four doubly parasitized spiders, however, coagulated contrast with the small plugs Thus H. argyraphaga is the only species yet of rapidly coagulated hemolymph at found in which it is certain that females puncture wounds made with a fine pin. kill the offspring from previous attacks on Presumably the larva added something to the host. The selective advantage of infan- the spider's blood which retarded coagu- ticide seems obvious. Only one of proba- lation. bly several hundred cocoon webs seen in One other possible manipulation was the field had two cocoons, and only one the inhibition of molting by host spiders, of these two produced a wasp. A second The evidence is only indirect, however, doubly parasitized spider was killed and Despite the fact that wasp larvae were consumed in captivity by the larger larva, able to remain attached when their host while the smaller larva fell to the ground molted (as also occurs in other polys- (still alive) with the discarded cadaver of phinctines — Nielsen 1923), and that the the spider. rate of parasitism of penultimate male spi- Similar use of the egg chorion to hold ders was not significantly different from onto the spider occurs in Acrodactyla mad- that of penultimate females (39.5% of 45 ida (Haliday) (= Polysphincta clypeata), penultimate males versus 50.8% of 63 pen- though Nielsen's (1923) drawings indicate ultimate females in 2000), the rate of par- that the egg of this species is also used by asitism of mature males was only about a larger larvae, rather than only the first in- tenth of that of mature females the same star as in H. argyraphaga. The dorsal tu- year. Of 32 parasitized penultimate and bercles ("warts") of the final instar are ante-penultimate spiders reared for two similar in form and placement to those of weeks, not a single spider molted. The Zatypota albicoxa (Walker) (= Polysphincta only evidence that spiders molted after be- eximia) and Polysphincta tuberosa Graven- ing parasitized involved wasp larvae that horst (Nielsen 1923). The two pairs of ven- Volume 9, Number 2, 2000 237 tral taps on segments 8 and 9 of the final instar larva of H. argyraphaga also resem- bled those of most of the species studied by Nielsen (1923), except that he recorded three pairs (on segments 7, 8 and 9) in P. tuber osa. The use of the sharply hooked setae on the tubercles to seize silk lines was appar- ently identical in all other species ob- served. Although Fincke et al. (1990) made no direct observations of larval behavior of H. robertsae, their Fig. 3 clearly shows dorsal tubercles on a large larva, leading to the supposition that they also have curved spines and are used to hold onto the spider's web. Bignell (1898) seems to have been correct in stating that lines were released by retracting these tubercles rath- er than by extending them, as also occurs in a pimpline ichneumonid that is proba- bly fairly closely related to Polysphinctini, Tromatobia oculatoria (Fabricius) (Nielsen 1923) (Nielsen 1923 may have been mis- taken in describing the opposite process of releasing by extending in Z. albicoxa). Only some details of how larvae of H. argyraphaga hold onto the host spider re- semble descriptions of other species. In- sertion of the paired ventral taps into a mass of material (the saddle) that adheres to the spider's abdomen, as described for second instar H. argyraphaga, has been seen in several other species (Nielsen 1923, 1935), though no difference was noted be- tween second and final instar larvae. Niel- sen (1923) stated that the cuticle of the first instar larva of Z. albicoxa was "glued to the host", but this was not true for H. ar- gyraphaga. In conjunction with this idea, Nielsen supposed that the saddle consist- ed of larval exuviae (Nielsen 1923, 1935), and this has been reiterated by later au- thors (e.g., Fitton et al. 1987, 1988). It seems likely, however, that the portion of the second instar's saddle that adhered to the host was coagulated spider hemo- lymph as in H. argyraphaga (Figs. 4, 5), rather than first instar larval cuticle. In fact Nielsen (1923) mentioned that wounds might be involved in allowing the larva to adhere to the spider. None of the descriptions of other spe- cies mentioned the final instar larva's change to hold onto the shed cuticle of the second instar with its terminal segments instead of its taps, as seen in H. argyrapha- ga. The mature larva's ability to release the spider in all of the species in order to pu- pate (presumably by relaxing the muscles that squeezed the shed cuticle), and the es- pecially active movements of the posterior tip of the larva during the only molting process that has ever been observed di- rectly (Nielsen 1923: 148-149 on Z. albi- coxa; ". . . the repeated attempts at fixing made by the hind end . . ."), suggest that similar changes may occur in other spe- cies. Both Nielsen and Bignell also noted that the larva of respectively Zatypota and an undetermined polysphinctine utilized the posterior part of the abdomen to pull out silk lines during cocoon construction, al- though their descriptions differ in details. It is not clear whether these differences were due to differences between species or, as supposed by Nielsen (1923), to dif- ferences in the precision of observations. The cocoons of H. robertsae were similar in form and color to those of H. argyraphaga. Increases in the duration of the larval stage when the host is feeding poorly, as in H. argyraphaga, probably also occurs in other species (Nielsen 1923). Females of H. robertsae were also larger than males, as is common among pimp- lines (Gauld et al. 1998), implying that ovi- positing females fertilize or refrain from fertilizing the egg on the basis of the size of the prey. If molting by the host is in- hibited by H. argyraphaga, then the size of the spider when it is attacked will corre- late with the size of the resulting wasp. Both species avoided parasitizing mature males of their hosts, probably for different reason. Mature male N. clavipes are prob- ably too small to produce an adult H. rob- ertsae (Fincke et al. 1990). Mature males of 238 Journal of Hymenoptera Research P. argyra were, in contrast, not too small argyraphaga was similar to that observed to produce adult wasps. For instance one (in captivity) in Schizopyga podagrica (Niel- moderately small mature male spider sen 1935). weighed 14.0 mg, somewhat more than One possible difference between H. ar- the 13.7 mg of a parasitized penultimate gyraphaga and H. robertsae is that the latter male and 12.1 mg of a parasitized ante- apparently does not induce the spider to penultimate male. Mature male P. argyra spin a highly modified cocoon web. Per- were only parasitized about a tenth as of- haps induction of behavioral changes has ten as mature females, however (totals of been lost, as the larger size of Nephila cla- 5.1% of 136 males and 51.8% of 328 fe- vipes and the corresponding greater dura- males in the two years). It was clear that bility of the mesh lines near its orbs both of the mature males parasitized in (which often remain more or less intact for 2000 had been attacked when they were in several days without repair by the spi- the penultimate instar, because part of the der — W. Eberhard unpub.) may make a cuticle from the previous instar adhered to modified web to support the cocoon un- the male's abdomen at the feeding scar be- necessary. Another possibility is that mod- low the larva in both cases. ification of spider behavior is a relatively The reason for lower parasitism of ma- recently derived character in H. argyrapha- ture males was presumably either because ga, but the probable plesiomorphic status female H. argyraphaga rejected mature of this species within Hymenoepimecis male spiders, or were less able to find and (Gauld 2000) and the ability of H. tedfordi attack them. Active rejection seems likely, to modify the behavior of L. mariana (W. because mature males of P. argyra often Eberhard, in prep.) argue against this idea, chase off smaller individuals and use their Neither Bignell (1898) nor Nielsen orbs to capture prey (10 of 11 males (1923) mention modified spider webs in checked for this detail were at the hubs of most of the polysphinctine larvae they an orb). Thus mature males are probably studied, but Shaw (1994:125) states that often exposed to hunting female wasps, "many spiders about to succumb to po- Active avoidance of males may be advan- lysphinctines seek a concealed site into tageous to the wasps because at least which they spin themselves". Nielsen sometimes mature males fail to construct (1923) noted that the last web that the host a cocoon web (Eberhard in prep.), thereby Cyclosa conica (Pallas) made before pupa- probably making the wasp's cocoon more tion by Polysphincta nielseni Roman was vulnerable to enemies. unusually small, and that such small orbs The mating system of H. argyraphaga ap- were especially resistant to damage. In parently differs from that of at least some contrast, he noted an apparently normal other ichneumonids in which males are at- web of a "certain TJieridia" with the co- tracted to sites where females are emerg- coon of a Acrodactyla degener (Haliday) ing. Males of H. argyraphaga were appar- (Nielsen 1923). The web of the theridiid ently not attracted to emerging or recently "Theridion" with the cocoon of a Zatypota emerged females as they rested on their albicoxa that he figured (Nielsen 1923) also cocoons, but quickly approached females seems normal, while Jimenez (1987) states after they flew to nearby vegetation. Fe- that Zatypota sp., which parasitizes Theri- males probably actively release a long- dion contreras, attaches its cocoon to the range attractant pheromone. Males ap- substrate rather than suspending it in the peared to concentrate their searching be- web. I have seen the cocoon of an uniden- havior at the tips of leaves of prominent tified species of Zatypota in an apparently plants, suggesting that they also use visual unmodified web of its theridiid host, Ane- stimuli. The very short copulation of H. losimus sp. These sparse data thus suggest Volume 9, Number 2, 2000 239 the preliminary conclusion that the non- der. The larger larvae won out over small- orb webs in which these wasps pupate are er individuals in two cases in which two not modified. A possible variation is Po- larvae grew on the same host. lysphincta tuberosa Gravenhorst, which Selection on spiders. — The success of at- parasitizes the orb weaver Araneus quadra- tacks by H. argyraphaga depended on the tus; but, judging from the figure of the co- wasp grasping the spider through the web coon and an accompanying web (Nielsen as it rested at the hub of its orb. Both the 1923), this species may pupate in the silk spider's observed defensive behavior retreat made by the host (Jones 1983) rath- (dropping quickly on a drag line below er than on the orb. Much more work re- the orb in response to the wasp's ap- mains to clarify the evolution of the ability proach from above), and the circumstanc- of these larvae to manipulate host web es in which some wasp attacks failed spinning behavior. (when the wasp hit the orb too far from The failure of female H. argyraphaga to the spider to grasp it) support the old idea use the ovipositor for oviposition resem- that meshes of lines associated with orb bles oviposition in aculeate wasps. Simi- webs function to defend the spider from lar, presumably convergent oviposition enemies (summaries in Lubin et al. 1982, direct from the genital opening has been Eberhard 1990). The strategy of attacking seen in an unspecified adelognathine ich- the spider from above functions well with neumonid, and is suspected in the braco- the more or less horizontal orbs of P. ar- nid Histeromerus (Shaw 1995). The ovipos- g}/ra, but leaves the wasp unable to follow itor of H. argyraphaga also injected para- the spider down its drag line when it lyzing venom into the spider host, and the drops, as Bignell (1898) observed an un- attacking wasp probably stabs the spider identified polysphinctine to do by walking with her unsheathed ovipositor during the down the line. instant she is landing and seizing it. In ad- This limitation on wasp attacks makes dition, the ovipositor was used to pry eggs it difficult to explain why late instar and and larvae of previous females from the mature female P. argyra seldom spin a spider's abdomen, and may also be used mesh above or below the orb. Such mesh- to sting these larvae. Live larvae are dif- es often occur in the web of earlier instars ficult to pry from the spider. One of the of this species. Among 31 webs of as many distinguishing traits of Polysphinctini in adult females, 90% lacked any mesh above general is a very sharply pointed ovipos- the orb, while the corresponding frequen- itor (Fitton et al. 1988). It is tempting to cy for 52 penultimate nymphs was 65% (p suppose that the sharp point is an adap- = 0.05 with Chi:). Unpublished data from tation to aid rapid penetration and im- student projects strongly suggest that spi- mobilization of spider hosts, which are ders build such meshes even more fre- potentially dangerous hosts. quently in earlier instars. The phylogeny Most European polysphinctines appear proposed for Plesiometa, Leucauge, and re- to oviposit at rather consistent sites on lated genera (Hormiga et al. 1995) sug- their hosts (Shaw 1998), but extensive de- gests that mesh construction is a derived scriptions of intra-specific variation in ovi- trait in these two genera, position sites on their host (e.g., Fig. 3) are not available for other polysphinctines, so it is not clear whether the substantial var- • tl"'ank Ian Gauld for help with literature and for iation in H. argyraphaga is unusual. It is kindlv identifying assorted ichneumonids 1 have ., , ,i , .. . . , brought him over the years. I am also eratetul to Im- possible that selection to escape infanti- , x, , , , >i , m r i dro I riacon tor locating a cruaaJ cocoon in the field. cide by subsequent females favors varia- Giibert Barrantes tor helping to End spiders, Paul tion in where eggs are placed on the spi- Hanson tor identifying the chalcidids, Maribelle Var- ACKNOWLEDGMENTS 240 Journal of Hymenoptera Research gas for producing the SEM images, and R. R. Askew, M. G. Fitton, Ian Gauld, Paul Hanson, and Mark Shaw for commenting on previous drafts. The Smith- sonian Tropical Research Institute and the Vicerrec- toria de Investigacion of the Universidad de Costa Rica provided financial support. LITERATURE CITED Bignell, G. C. 1898. The Ichneumonidae (parasitic flies) of the South of Devon. Report and Transac- tions—Devonshire Association for the Advancement of Science, Literature and Art (1898):458-504. Eberhard, W. G. 1990. Function and phylogeny of spi- der webs. Annual Revieiv of Ecology and System- atics 21:341-372. Eberhard. W. G. 2000. Spider web manipulation by a wasp larva. Nature 406:255-256. Eberhard, W. G. in prep. Under the influence: web construction behavior by Plesiometa argyra (Ara- neae: Tetragnathidae) parasitized by the ichneu- monid wasp Hymenoepitnecis argyraphaga (Hy- menoptera: Ichneumonidae). Fincke, O.M., L. Higgins and E. Rojas. 1990. Parasit- ism of Nephila clavipes (Araneae, Tetragnathidae) by an ichneumonid (Hymenoptera, Polysphinc- tini) in Panama. Journal of Arachnology 18:321- 329. Fitton, M. G., M. R. Shaw and A. D. Austin. 1987. The Hymenoptera associated with spiders in Europe. Zoological Journal of the Linnean Society 90:65-93. Fitton, M. G., M. R. Shaw and I. D. Gauld. 1988. Pimpline ichneumon-flies Hymenoptera, Ichneu- monidae (Pimplinae). Handbooks for the Identifi- cation of British Insects (Royal Entomological So- ciety of London) 7(1):1-110. Gauld, I. D. 1995. Ichneumonidae. pp. 390-431 In P. Hanson and I. D. Gauld (eds) Hymenoptera of Cos- ta Rica. Oxford University Press: Oxford. 904pp. Gauld, I. D. 2000. The re-definition of pimpline genus Hymenoepitnecis (Hymenoptera: Ichneumonidae) with a description of a plesiomorphic new Costa Rican species. Journal of Hymenoptera Research 9: 213-219. Gauld, I. D., J. A. Ugalde G. and P. Hanson. 1998. Guia de los Pimplinae de Costa Rica (Hymenop- tera: Ichneumonidae). Revista de Biologia Tropical 46 (Suppl. 1):1-189. Hanson, P. and I. Gauld. 1995. The Hymenoptera of Costa Rica. Oxford University Press: Oxford. Hormiga, G., W. G. Eberhard and J. A. Coddington 1995. Web construction behavior in Australian Phonognatha and the phylogeny of nephiline and tetragnathid spiders (Araneae, Tetragnathidae). Australian Journal of Zoology 43:313-364. Jimenez, M. L. 1987. Relaciones biologicas entre ar- anas y avispas. Folia Entomologica Mexicana 73: 173-183. Jones, D. 1983. The Larousse guide to spiders. New York: Larousse and Co. 320 pp. Lubin, Y. D., B. Opell, W. G. Eberhard and H. W. Levi. 1982. The "orb plus cone" webs in Ulobor- idae (Araneae) with a description of a new genus and four new species. Psyche 89:29-64. Nielsen, E. 1923. Contributions to the life history of the pimpline spider parasites (Polysphincta, Zag- lyptus, Tromatobia). Entomologiske Meddelelser 14: 137-205. Nielsen, E. 1928. A supplementary note upon the life histories of the Polysphinctas (Hym. Ichneum.). Entomologiske Meddelelser 16:152-155. Nielsen, E. 1929. A second supplementary note upon the life histories of the Polysphinctas (Hym. Ich- neum. ).Entomologiske Meddelelser 16:366-368. Nielsen, E. 1935. A third supplementary note upon the life histories of the Polysphinctas (Hym. Ich- neum.). Entomologiske Meddelelser 19:193-215. Shaw, M. R. 1994. Parasitoid host ranges. In Hawkins, B. A. and W. Sheehan eds. Parasitoid Community Ecology. New York: Oxford Univ. Press; 111-144. Shaw, M. R. 1995. Observations on the adult behav- iour and biology of Histeromerus mystactnus Wes- mael (Hymenoptera: Braconidae). The Entomolo- gist 114:1-13. Shaw, M. R. 1998. Hymenopteres europeens en par- ticulier les Polysphinctini dont les larves se nour- rissent d'araignees. Connaissance des Invertebres, Serie Arachnidcs 3:14-40. Wahl, D. B. and I. D. Gauld. 1998. The cladistics and higher classification of the Pimpliformes (Hy- menoptera: Ichneumonidae). Systematic Entomol- ogy 23:265-298. j. HYM. RES. Vol. 9(2), 2000, pp. 241-245 Molecular Confirmation of Host Records for Ichneumonoid Parasitoids of Wood-boring Beetle Larvae Nina M. Laurenne, Robert Belshaw, Gavin Broad and Donald L. J. Quicke (NML) Finnish Museum of Natural History, Zoological Museum, Entomological Division, P.O. Box 17 (P. Rautatiekatu 13), FIN-00014 University of Helsinki, Finland. (RB, GB, DLJQ) Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, U.K., (GB, DLJQ) Centre for Population Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, U.K., and (DLJQ) Department of Entomology, The Natural History Museum, London SW7 5BD, U.K. Abstract. — Field observations in Sabah of a large braconine wasp, Shelfordia sp., investigating and ovipositor-probing at a frass hole in an exposed tree root suggested the possible location of a host. Investigation of the substrate revealed a paralysed and partly consumed larva of an an- thribid beetle that had a 1st instar parasitic wasp larva on it. Both adult and larval wasps were sequenced, in separate laboratories, for the D2-D3 expansion region of the nuclear 28S rDNA gene. The sequences were identical and consideration of their unique features compared with many other sequences from ichneumonoids leads to confirmation that the hosts of Shelfordia spp. include Anthribidae living in tree roots. The implications of DNA sequencing for constructing quantitative food webs are discussed. Host records for parasitic Hymenoptera are notoriously unreliable as has been well documented by Shaw (1994) and Noyes (1994). Errors in the literature derive from several sources including misidentifica- tion of the parasitoid, misidentification of the host, misidentification of both, and through the wrong association of a para- sitoid with a putative host because of con- tamination of the rearing system. These problems are particularly true for con- cealed hosts such as wood-borers, gall makers, etc., because these substrates can contain many potential host species apart from the one that may be the focus of at- tention. Simply rearing a parasitoid and a potential host from a given piece of sub- strate has consequently often led to incor- rect conclusions about what is parasitising what. Although it is sometimes possible carefully to dissect substrate and to iden- tify any host and parasitoid remains, or to isolate and rear the host and its parasitoid in isolation, these are both difficult pro- cedures requiring much skill and at least some luck. The development of molecular techniques, in particular DNA sequencing, has opened new and in some respects eas- ier ways to solve these problems. The braconine genus Shelfordia Cameron is relatively common from India, through the Indo-Australian Region, to N.E. Aus- tralia, often being found in local collec- tions, no doubt because its species are rather large by parasitic wasp standards. Its identification has not been without problems though, and since its original description (Cameron 1902), nothing had been published on it under that name un- til it was discovered to be a senior syno- nym of Sigalphogastra Cameron (Quicke 1982). The genus name Sigalphogastra had, however, been very inconsistently ap- plied, and few of the older publications citing this name (Shenefelt 1978) actually refer to taxa congeneric with the type spe- cies. In 1984, Quicke described a new ge- nus Rostraulax, based on a species similar 242 Journal of Hymenoptera Research to Shelfordia sensn stricto, but which had an be searching. However, after several hours elongate labio-maxillary complex and of observation, the wasps were never seen partly fused apical flagellomeres (Quicke near this tree, but only near the path. The 1984a). Although Rostraulax is recognisa- wasps were very cautious; they remained ble by these autapomorphies, Shelfordia is, stationary on the low vegetation for long as far as we can tell at present, left para- periods, and were easily disturbed by tour- phyletic with respect to Rostraulax, and the ists passing by. After some while, however, letter was formally synonymised with we observed one and then another fly to Shelfordia by van Achterberg (1993). Prior land on a small (approximately 5cm diam- to now, there have been no available host eter) tree root that was partially exposed by data for the genus Shelfordia, although the the path. Again the wasps remained station- especially long ovipositor of one Indian ary for about 20 minutes when one of them species, S. longicaudata van Achterberg, approached a boring from which fresh- suggests that it is potentially a parasitoid looking wood particles were exuded. After of a host living deeply within wood (van antennating the site for several minutes, she Achterberg 1993). Since its original de- raised her metasoma and probed into the scription, some 35 species have been re- frass hole with her ovipositor and sheathes, classified into it (or its synonym, Rostrau- After a few moments she was observed lax) (Quicke 1983, 1984b, 1985, 1988, 1991, making marked twisting movements with Quicke and van Achterberg 1990, Quicke the apex of her metasoma and she was then and Koch 1990, van Achterberg 1993, van collected. Achterberg and O'Toole 1993), of which 11 are from the island of Borneo, all from METHODS Sarawak. No species-level keys are avail- Collection. — The apparently-ovipositing able for Shelfordia, and proper taxonomic adult female Shelfordia was collected with revision is required before the species re- a net and placed into a clean vial contain- ferred to here can be properly identified. ing 95% ethanol. Other adult wasps re- ferred to in this paper were collected in FIELD OBSERVATIONS me same way or in Maiaise traps contain- In August 1999, DLJQ and NML had the ing 95% ethanol. The tree root with frass opportunity to collaborate with the group of hole was dissected the same day with a Dr Maryati Mohamed of the Tropical Bio- saw, hammer and chisel, and a boring lo- diversity and Conservation Unit, Universiti cated below the frass hole which con- Malaysia, Sabah, and to observe some large tained a paralysed and partly deflated braconid wasps in lowland tropical rain for- beetle larva upon which was a single est at Poring (Kinabalu National Park), small (t/ the XV International Congress of Zoology. International Trust for Zoological No- menclature, London. 176 pp. International Commission on Zoological Nomencla- ture. 1985. International Code of Zoological Somen cloture, third edition. Charlesworth & Co., Hud- dersfield. 338 pp. International Commission on Zoological Nomencla- ture. 1987. Opinion 1424. Gnamptodon Halidaj 1833 (Insecta: Hymenoptera): Bracon pumilio Nees, 1834 designated as type species. Bulletin of Zoological Nomenclature 44: 55-56. International Commission on Zoological Nomencla- ture. 1999. International Code oj Zoological Nomen- 266 Journal of Hymenoptera Research cloture, fourth edition.Tipografia La Garangola, Padova. 306 pp. Latreille, P. A. 1804. Table methodique des insectes, pp. 129-200 In: Nouveau Dictionnaire d'Historie Naturelle..., first edition, volume 24. Deterville, Paris. 258 pp. Leach, W. E. 1815. Entomology, pp. 57-172. In: Brew- ster, D. (editor). Edinburgh Encyclopaedia, volume 9. Mackauer, M. 1961. Die Gattungen der Familie Aphi- diidae und ihre verwandtschaftliche Zuordnung (Hymenoptera: Ichneumonoidea). Beitraege zur Entomologie 11: 792-803. Mackauer, M. 1968. Pars 3 Aphidiidae. In: Ferriere, Ch. and J. van der Vecht (editors). Hymenoptero- rum Catalogus, nova editio. W. Junk, The Hague. 103 pp. Mackauer, M. and P. Stary. 1967. The Aphidiidae of the World. Index of Entomophagous Insects. Le Fran- cois, Paris. 195 pp. Marshall, T. A. 1872. Braconidae, pp. 96-132. A Cat- alog of the British Hymenoptera. Part 4: Chrysidi- dae, lchneumonidae , Braconidae, and Evaniidae. Napier, Entomological Society of London, Lon- don. 137 pp. Marshall, T. A. 1885. Monograph of British Braconi- dae. Part I. Transactions of the Royal Entomological Society of London 1885: 1-280. Marshall, T. A. 1887. Monograph of British Braconi- dae. Part II. Transactions of the Royal Entomological Society of London 1887: 51-131. Marshall, T. A. 1888. Les Braconides. In: Andre, E. Species des Hymenopteres D'Europe & D'Algerie, volume 4. Andre, Beaune (Cote D'Or) 609 pp. [Shenefelt (1965) indicates publication in 1890 but this is incorrect.] Marshall, T. A. 1889. A Monograph of British Bracon- idae. Part III. Transactions of the Royal Entomolog- ical Society of London 1889: 149-210. Marshall, T. A. 1891. A monograph of British Bracon- idae. Part IV. Transactions of the Entomological So- ciety of London 1891: 7-61. Marshall, T. A. 1898. Braconidae. In: Dalla Torre, C. G. de Catalogus Hymenopterorum, IV, volume 8. Engelmann, Leipzig. 323 pp. Mason, W. R. M. 1969. Muesebeckiini, a new tribe of Braconidae (Hymenoptera). Proceedings of the En- tomological Society of Washington 71: 263-278. Mason, W. R. M. 1974. A generic synopsis of Brachistini (Hymenoptera: Braconidae) and recognition of the name Charmon Haliday. Proceedings of the Entomo- logical Society of Washingto)i 76: 235-246. Mason, W. R. M. 1978. A new genus, species and fam- ily of Hymenoptera (Ichneumonoidea) from Chile. Proceedings of the Entomological Society of Washington 80: 606-610. Mason, W. R. M. 1981a. The polyphyletic nature of Apanteles Foerster (Hymenoptera: Braconidae): a phylogeny and reclassification of Microgastrinae. Memoirs of the Entomological Society of Canada 115: 1-147. Mason, W. R. M. 1981b. Paxylommatidae: The correct family-group name for Hybrizon Fallen (Hyme- noptera: Ichneumonoidea), with figures of un- usual antennal sensilla. Canadian Entomologist 113: 433^39. Mason, W.R.M. 1983. A new South African subfamily related to Cardiochilinae (Hymenoptera: Bracon- idae). Contributions of the American Entomological Institute 20: 49-62. Menke, A. S. 1997. Family-group names in Sphecidae (Hymenoptera: Apoidea). Journal of Hymenoptera Research 6: 243-255. Michener, C. D. 1986. Family-group names among bees. Journal of the Kaiisas Entomological Society 59: 219-234. Muesebeck, C. F. W. 1928. Family Braconidae, pp. 895-920. In: Leonard, M. D. A list of the insects of New York with a list of the spiders and certain allied groups. Memoirs of the Cornell University Agricultural Experiment Station, volume 101. [Cover dated August, 1926, but date of issue was January, 1928.] Muesebeck, C. F. W. and L. M. Walkley 1951. Family Braconidae, pp. 90-184. In: Muesebeck, C. F. W., K. V. Krombein, and H. K. Townes Hymenoptera of America, north of Mexico Synoptic Catalog. U.S. Government Printing Office, Washington, D. C. 1420 pp. Nees von Esenbeck, C. G. 1812. Ichneumonides ad- sciti in genera et familias divisi a Dre. Nees ab Esenbeck. Magazin Gesellschaft Naturforschender Freunde zu Berlin 5: 3-37. [Volume dated 1811, published 1812.] Nees von Esenbeck, C. G. 1814. Ichneumonides ad- sciti in genera et familias divisi a Dre. Nees ab Esenbeck. Continuatio. Magazin Gesellschaft Na- turforschender Freunde zu Berlin 6: 183-221. [Vol- ume dated 1812, published 1814.] Nees von Esenbeck, C. G. 1816. Ichneumonides ad- sciti in genera et familias divisi a Dre. Nees ab Esenbeck. Continuatio. Magazin Gesellschaft Na- turforschender Freunde zu Berlin 7: 243-277. [Vol- ume dated 1813, published 1816.] Nees von Esenbeck, C. G. 1819. Appendix ad J. L. C. Gravenhorst conspectum generum et familiarum ichneumonidum. Genera et familias ichneumon- idum adscitorum exhibens. lsis 9: 299-310. [Vol- ume dated 1818, published 1819.] Nees von Esenbeck, C. G. 1834. Hymenopterorum Ich- neumonibus affinium monographiae, genera Europaea ei species illustrantes, volume 1. Cotta, Stuttgart and Tuebingen. 320 pp. Oltra Moscardo, M. and M. Michelena Saval, 1988. Contribution al conocimiento de los Microgastri- nae en la Peninsula Iberica (Hym., Braconidae): Volume 9, Number 2, 2000 267 III. Cotesini. Eos Rei'ista Espanola de Entomologia 64: 165-171. Papp, J. 1994. Maxfischeria tricolor gen. n. et sp. n. from Australia (Insecta: Hymenoptera: Braconidae). Annalen des Naturhistorischen Museum in Wien 96: 143-147. Parfitt, E. 1881. The fauna of Devon. Order Hym. Fam. Ich. Section Pupivora. Reports and Transac- tions of the Devonshire Association for the Advance- ment of Science, Literature, and Art 13: 240-292. Quicke, D. L. J. 1987a. A new subfamily of Braconi- dae, the Vaepellinae, based on a new genus and species from Ghana (Insecta, Hymenoptera). Zoologica Scripta 16: 73-77. Quicke, D. L. J. 1987b. The Old World genera of bra- conine wasps (Hymenoptera: Braconidae). jour- nal of Natural History 21: 43-157. Quicke, D.LJ. and V.I. Tobias, 1990. New genera of braconid wasps of the subfamily Braconinae (Hymenoptera, Braconidae) from Australia. En- tomologischeskoye Obozrenie 69: 164-180. Quicke, D. L. J. and C. van Achterberg. 1990. Phylog- eny of the subfamilies of the family Braconidae (Hymenoptera: Ichneumonoidea). Zoologische Verhandelingen Leiden 258: 1-95. Quilis Perez, M. 1931. Especies nuevas de Aphidiidae espanoles (Hym. Brae). Eos Rei'ista Espanola de Entomologia 7: 25-84. Regies Internationales de la Nomenclature Zoologique Adoptees par les Congres Internationaux de Zoologie. 1905. F. R. de Rudeval, Paris. Say, T. 1836. Descriptions of new species of North American Hymenoptera, and observations on some already described. Boston journal of Natural History 1: 209-305. Schmiedeknecht, O. 1907. Die Hymenopteren Mitteleu- ropas. Fischer, Jena. 804 pp. Schulz, W. A. 1911. Zweihundert alte Hymenopteren. Zoologische Annalen 4: 1-220. Sharkey, M. J. 1992. Cladistics and tribal classification of the Agathidinae (Hymenoptera: Braconidae). journal of Natural History 26: 425^148. Sharkey, M. J. 1993. Family Braconidae, pp. 362-395. In: Goulet, H. and J. T. Huber (editors). Hyme- noptera of the World: An identification guide to fam- ilies. Research Branch, Agriculture Canada Pub- lication 1894/E, Ottawa. 668 pp. Sharma, V. 1984. A new subfamily Excultinae with description of type genus; from Jammu (India) (Hymenoptera, Braconidae). Reichenbachia 22: 75- 77. Shaw, S. R. 1985. A phylogenetic study of the subfam- ilies Meteorinae and Euphorinae (Hymenoptera: Braconidae. Entomography 3: 277-370. Shaw, M. R. and T. Huddleston. 1991. Classification and biology of braconid wasps. Handbooks for the Identification of British Insects 7(11): 1-126. Shenefelt, R. D. 1965. A contribution towards knowl- edge of the world literature regarding Braconi- dae. Beitraege zur Entomologie 15: 243-500. Shenefelt, R. D. 1969. Pars 4 Braconidae 1 Hybrizon- tinae, Euphorinae, Cosmophorinae, Neoneuri- nae, Macrocentrinae. In: Ferriere, Ch. and J. van der Vecht (editors). Hymenopterorum Catalogus, nova editio. Junk, The Hague. 176 pp. Shenefelt, R. D. 1980. Pars 16 Braconidae 11: Intro- duction, Guide to host names, Index to braconid names. In: Achterberg, C. van and R. D. Shenefelt (editors). Hymenopterorum Catalogus, nova editio. Junk, The Hague. 384 pp. Shestakov, A. 1932. Zur Kenntnis der asiatischen Bra- coniden. Zoologischer Anzeiger 99: 255-263. [Sometimes cited as 1930 or (1930) 1932, but date submitted = 24 March 1932]. Spinola, M. 1808. Insectorum Liguriae Species Novae aut Rariores, quas in Agro Ligustico nuper Detexit, Des- cripsit, et iconibus Illustravit, volume 2. Gravier, Genua. 263 pp. Stephens, J. F. 1829. The Nomenclature of British In- sects. In: Systematic Catalogue of British Insects. Baldwin and Cradock, London. 68 pp. Szepligeti, G. 1901.Tropische Cenocoelioniden und Braconiden aus der Sammlung des Ungarischen National-Museums. Termeszetrajzi Fuzetek 24: 353-402. Szepligeti, G. 1904a. Ueber Gnathobracon A. Costa. Annuario del Museo Zoologico della R. Universita di Napoli 1: 2. Szepligeti, G. V. 1904b. Hymenoptera. Fam. Braconi- dae. In: Wytsman, P. Genera Insectorum, volume 22. Verteneuil and Desmet, Bruxelles. 253 pp. Szepligeti, G. 1908. Braconiden aus der Sammlung des Ungarischen National-Museums. II. Theil. Annates Musei Nationalis Hungarici 6: 397—127. Telenga, N. A. 1952. Origin and evolution of parasit- ism in Hymenoptera Parasitica and development of their fauna in the USSR. Academy of Sciences of the Ukrainian SSR. 112 pp. [In Russian, English translation published in 1969] Telenga, N. A. 1955. Braconidae, subfamily Micro- gastrinae, subfamily Agathinae. Fauna USSR. Hy- menoptera 5(4): 1-311. [In Russian, English trans- lation published in 1964] Tobias, V. I. 1957. On morphology, taxonomy and phytogeny of the supertribus Braconina Tel. (1 1\ • menoptera, Braconidae). Zoologicheskii Zhurnal 36: 1338-1354. [In Russian] Tobias, V. I. 1962. A new subfamily ol Braconidae from Middle Asia. Transactions of the Zoological Institute Academy of Sciences, USSR 30: 268-270. [In Russian] Tobias, V. I. 1967. A review of the classification phy- togeny and evolution of the family Braconidae (Hymenoptera). Entomologicheskoye Obozrenie 46: M5-669. [In Russian] Tobias. V. I. 1968. Problems ot the classification and 268 Journal of Hymenoptera Research phylogeny of the family Braconidae (Hymenop- tera). Entomological Society, USSR Academy of Sci- ences Special Issue Dedicated to the Memory ofN. A. Kholodkovski: 3-43. [In Russian] Tobias, V. I. 1976a. Gnaptogaster mongolica gen. and sp. n. from a new tribe of braconids (Hymenop- tera, Braconidae, Opiinae) and some problems of the phylogeny and evolution of this family. In- sects of Mongolia 4: 315-321. [In Russian] Tobias, V. I. 1976b. Braconidae of the Caucasus (Hy- menoptera, Braconidae). Akademia Nauk, Len- ingrad. 110: 1-286. [In Russian] Tobias, V. I. 1979. Two new and little-known subfam- ilies of Braconidae (Hymenoptera) from Austra- lia. Entomologicheskoye Obozrenie 58: 128-142. [In Russian] Tobias, V. I. 1986. Keys to the insects of the European part of the USSR. HI. Hymenoptera, 4: 1-501. Akademia Nauk, Leningrad. [In Russian, English translation published in 1995] Tobias, V. I. 1987. New taxa of braconids (Hymenop- tera, Braconidae) from Baltic amber. Entomologi- cheskoye Obozrenie 66: 845-859. [In Russian] Tobias, V. 1.1988. A new subfamily of braconids (Hy- menoptera, Braconidae) from Australia. Entomo- logicheskoye Obozrenie 67: 644-648. [In Russian] Tobias, V. I. 1990. In: Quicke, D.L.J, and V.I. Tobias. New genera of braconid wasps of the subfamily Braconinae (Hymenoptera, Braconidae) from Aus- tralia. Entomologischeskoye Obozrenie 69: 164-180. Viereck, H. L. 1910. Braconidae, pp. 605-615, Alysi- dae, pp. 615-616, Capitoniidae, p. 616. In: Smith, J. Annual Report of the Neiv Jersey State Museum Including a Report of the Insects of New Jersey, 1909. MacCrellish and Quigley, Trenton. Viereck, H. L. 1914. Type species of the genera of ichneumon flies. Bulletin of the United States Na- tional Museum 83: 1-186. [Date of publication Jan- uary 13, 1914.] Viereck, H. L. 1918. A list of families and subfamilies of ichneumon-flies or the Superfamily Ichneu- monoidea (Hymenoptera). Proceedings of the Bio- logical Society of Washington 31: 69-74. Viereck, H. L. 1919a. Additions and corrections to "a list of families and subfamilies of ichneumon- flies or the Super Family Ichneumonoidea (Hy- menoptera)." Proceedings of the Biological Society of Washington 32: 48. Viereck, H. L. 1919b. Additions to "a list of families and subfamilies of ichneumon-flies or the Super- family Ichneumonoidea (Hymenoptera)." Pro ceedings of the Biological Society of Washington 32: 198. Viereck, H. L. 1921. First supplement to "type species of the genera of ichneumon-flies." Proceedings of the United States National Museum 59: 129-150. Wesmael, C. 1835. Monographie des Braconides de Belgique. Nouveaux memoires de V academic royale des sciences et belles-lettres de Bruxelles 9: 1-252. Wharton, R. A. 1978. Exodontiellini, a new tribe of Opiinae with exodont mandibles (Hymenoptera: Braconidae). Pan-Pacific Entomologist 53: 297-303. [Volume dated October, 1977, but issued 14 Feb- ruary 1978.] Wharton, R. A. 1993. Review of the Hormiini (Hy- menoptera: Braconidae) with a description of new taxa. Journal of Natural History 27: 107-171. Wharton, R. A., P. M. Marsh and M. j/sharkey. 1997. Identification manual to the New World Genera of the family Braconidae (Hymenoptera). Inter- national Society of Hymenopterists Special Pub- lication No. 1. 439 pp. Wharton, R. A., S. R. Shaw, M. J. Sharkey, D. B. Wahl, J. B. Woolley, J. B. Whitfield, P. M. Marsh and W. Johnson. 1991. Phylogeny of the subfamilies of the family Braconidae (Hymenoptera: Ichneu- monoidea): A reassessment. Cladistics 8: 199-235. Whitfield, J. B. 1992. The polyphyletic origin of en- doparasitism in the cyclostome lineages of Bra- conidae (Hymenoptera). Systematic Entomology 17: 273-286. Whitfield, J. B. and W. R. M. Mason 1994. Mendesel- linae, a new subfamily of braconid wasps (Hy- menoptera, Braconidae) with a review of rela- tionships within the microgastroid assemblage. Systematic Entomology 19: 61-76. Whitfield, J. B. and P. D. Dangerfield. 1997. Cardi- ochilinae, pp. 177-183. In: Wharton, R. A., P. M. Marsh and M. J. Sharkey (editors). Identification manual to the New World Genera of the family Braconidae (Hymenoptera). International Society of Hymenopterists Special Publication No. 1. Zettel, H. 1990. Eine revision der Gattungen der Che- loninae (Hymenoptera, Braconidae) mit Beschrei- bungen neuer Gattungen und Arten. Annalen des Naturhistorischen Museum in Wien 91: 147-196 APPENDIX 1 The following is an alphabetical list of family- group names in the Braconidae, with indication of their subfamily placement. Though classifications vary from author to author, and are presently contro- versial, the subfamily arrangement used here is based largely on Quicke and van Achterberg (1990) and van Achterberg (1993, 1995). Tribes and subtribes largely follow Mackauer (1968), Mason (1981), van Achter- berg (1987, 1988, 1991b, 1992, 1994, 1995), Shaw (1985), Quicke (1987), Zettel (1990), Belokobylskij (1992), Sharkey (1992), and van Achterberg and Aus- tin (1992). Acampsini van Achterberg and Austin, 1992 (Sigal- phinae) Acampsohelconini Tobias, 1987 (Cenocoeliinae) Acanthobraconini Fahringer, 1928 (Braconinae) Acanthodoryctina Belokobylskij, 1992 (Doryctinae) Volume 9, Number 2, 2000 269 Acanthohormiini Belokobylskij, 1993 (Lysiterminae) Aclitina Mackauer, 1961 (Aphidiinae) Acrisidini Hellen, 1957 (Rhyssalinae) Adeliinae Viereck, 1918 (Adeliinae) Ademonini Fischer, 1970 (Opiinae) Adeshini van Achterberg, 1983 (Braconinae) Afrocampsini van Achterberg and Austin, 1992 (Si- galphinae) Agathidinae (-ini) Haliday, 1833 (Agathidinae) Aleiodini Muesebeck, 1928 (Rogadinae) Alloeini Ashmead, 1900 (Alysiinae) Alysiinae (-ini) Leach, 1815 (Alysiinae) Amicrocentrinae van Achterberg, 1979 (Amicrocen- trinae) Aneurobraconini Fahringer, 1936 (Agathidinae) Antestrigini van Achterberg, 1987 (Orgilinae) Apantelini Viereck, 1918 (Microgastrinae) Aphidiinae (-ini) Haliday, 1833 (Aphidiinae) Aphrastobraconini Ashmead, 1900 (Braconinae) Apozyginae Mason, 1978 (Apozyginae) Archaphidina Mackauer, 1968 (Aphidiinae) Argamaniini van Achterberg, 1991 (Braconinae) Aspidobraconina van Achterberg, 1984 (Braconinae) Atanycolina Fahringer, 1928 (Braconinae) Austrohormiini Belokobylskij, 1993 (Hormiinae) Avgini Belokobylskij, 1993 (Hormiinae) Baeocentrini Fahringer, 1928 (Opiinae) Bassinae Nees, 1812 (invalid) Bassinae Viereck, 1918 (Agathidinae) Bathvaulacini Quicke, 1987 (Braconinae) Betylobraconinae (-ini) Tobias, 1979 (Betylobraconi- nae) Binareini Fischer, 1981 (Doryctinae) Biosterini Fischer, 1970 (Opiinae) Bitomina Fischer, 1981 (Opiinae) Blacinae (-ini) Foerster, 1862 (Blacinae) Blacozonini van Achterberg, 1988 (Blacinae) Brachistini Foerster, 1862 (Helconinae) Braconidae (-inae) Nees, 1812 Brulleiini van Achterberg, 1983 (Helconinae) Caenophanina Belokobylskij, 1992 (Doryctinae) Calyptini Marshall, 1872 (Helconinae) Canberriini Belokobylskij, 1993 (Hormiinae) Capitoniini Viereck, 1910 (Cenocoeliinae) Cardiochilinae Ashmead, 1900 (1887) (Cardiochilinae) Cedriina Belokobylskij, 1993 (Pambolinae) Cenocoeliinae (-ini) Szepligeti, 1901 (Cenocoeliinae) Centistini Capek, 1970 (Euphorinae) Cercobarconini Tobias, 1979 (Trachypetinae) Chalaropini van Achterberg, 1988 (Blacinae) Charmontinae van Achterberg, 1979 (Charmontinae) Cheloninae (-ini) Foerster, 1862 (Cheloninae) Chelonohelconini Tobias, 1987 (Helconinae) Chiviniini Shestakov, 1932 (Braconinae) Chremylini Hellen, 1957 (Pambolinae) Chremylomorphina Belokobylskij, 1993 (Pambolinae) Clinocentrini van Achterberg, 1991 (Rogadinae) Coeliniina Viereck, 1919 (Alysiinae) Coeloidini Tobias, 1957 (Braconinae) Coleopiina Fischer, 1970 (Opiinae) Cosmophorini Capek, 1958 (Euphorinae) Cotesiini Mason, 1981 (Microgastrinae) Cremnoptini Sharkey, 1992 (Agathidinae) Cryptoxilonini Tobias, 1986 (Euphorinae) Dacnusini Foerster, 1862 (Alysiinae) Dendrosotina Fischer, 1981 (Doryctinae) Desmiostomatini Fischer, 1972 (Opiinae) Dimerina Belokobylskij, 1993 (Pambolinae) Dinocampini Shaw, 1985 (Euphorinae) Diospilitinae Tobias, 1987 (Aphidiinae) Diospilini Foerster, 1862 (Helconinae) Dirrhopinae van Achterberg, 1984 (Dirrhopinae) Disophrini Sharkey, 1992 (Agathidinae) Doryctinae (-ini) Foerster, 1862 (Doryctinae) Doryctomorphina Belokobylskij, 1992 (Rhyssalinae) Dyscoletini van Achterberg, 1988 (Blacinae) Earinini Sharkey, 1992 (Agathidinae) Ecnomiinae van Achterberg, 1985 (Ecnomiinae) Ecphylini Hellen, 1957 (Doryctinae) Elasmosomini Viereck, 1918 (Neoneurinae) Embobraconina van Achterberg, 1995 (Doryctinae) Ephedrini Mackauer, 1961 (Aphidiinae) Euphorinae (-ini) Foerster, 1862 (Euphorinae) Euurobraconini Ashmead, 1900 (Braconinae) Evaniodini Fischer, 1981 (Doryctinae) Excultinae Sharma, 1984 (Chrysididae) Exodontiellini Wharton, 1978 (Opiinae) Exothecinae Foerster, 1862 (Exothecinae) Facitorini van Achterberg, 1995 (Betylobraconinae) Forniciini Mason, 1981 (Microgastrinae) Gastrothecini Fahringer, 1928 (invalid) Glyptomorphina Tobias, 1957 (Braconinae) Gnamptodontinae Fischer, 1970 (Gnamptodontinae) Gnaptogastrini Tobias, 1976 (Gnamptodontinae) Gnathobraconini Szepligeti, 1904 (Braconinae) Gynocryptina Quilis Perez, 1931 (Aphidiinae) Habrobraconini Fahringer, 1928 (Braconinae) Hecabolini Foerster, 1862 (Doryctinae) Helconinae (-ini) Foerster, 1862 (Helconinae) Helorimorphini Schmiedeknecht, 1907 (Euphorinae) Heterospilini Fischer, 1981 (Doryctinae) Histeromerinae Fahringer, 1930 (Histeromerinae) Holcobraconini Cameron, 1905 (Doryctinae) Homolobinae (-ini) van Achterberg, 1979 (Homolo- binae) Hormiinae (-ini) 1-oer^ter, 18h2 (Hormiinae) Hydrangeocolini Whitfield, 1442 (Hormiinae) Ichneutinae (-ini) Foerster, lSh2 (Ichneutinae) Incubinae Brues and Melander, 1932 (Aphidiinae) Iphiaulacini Fahringer, 1928 (Braconinae) Isomecina Tobias, 1990 (Braconinae) Iseurini Hedqvist, 1959 (Cenocoeliinae) [vondroviina Belokobylskij, 1992 (Doryctinae) Khoikhoiinae Mason, 1983 (Khoikhoiinae) Labaniini Belokobylskij, 1992 (Doryctinae) Leiophronini Foerster, 1862 (Euphorinae) 270 Journal of Hymenoptera Research Leptorhaconotini Belokobylskij, 1992 (Doryctinae) Leuriniini Belokobylskij, 1993 (Hormiinae) Lissogastrini Oltra and Michelena, 1988 (Microgastrinae) Loxocephalini Shaw, 1985 (invalid) Lysiphlebina Mackauer, 1961 (Aphidiinae) Lysiterminae (-ini) Tobias, 1968 (Lysiterminae) Macrocentrinae Foerster, 1862 (Macrocentrinae) Mannokeraiini van Achterberg, 1995 (Masoninae) Masoninae (-ini) van Achterberg, 1995 (Masoninae) Maxfischeriini Papp, 1994 (Helconinae) Mendesellinae Whitfield and Mason, 1994 (Mende- sellinae) Mesocoelini Viereck, 1918 (Agathidinae) Mesostoinae van Achterberg, 1975 (Mesostoinae) Meteorideinae (-ini) Capek, 1965 (Meteorideinae) Meteorini Cresson, 1887 (Euphorinae) Microbraconini Bridwell, 1920 (Braconinae) Microctonini Shaw, 1985 (Euphorinae) Microdini Foerster, 1862 (Agathidinae) Microgastrinae (-ini) Foerster, 1862 (Microgastrinae) Microplitini Mason, 1981 (Microgastrinae) Microtypinae Szepligeti, 1908 (Microtypinae) Mimagathidini Enderlein, 1905 (Orgilinae) Minangini De Saeger, 1948 (Sigalphinae) Miracinae Viereck, 1918 (Miracinae) Monitoriellini Belokobylskij, 1992 (Rhyssalinae) Monoctonina Mackauer, 1961 (Aphidiinae) Mononeurina Fischer, 1981 (Doryctinae) Muesebeckiini Mason, 1969 (Ichneutinae) Myiocephalini Chen and van Achterberg, 1997 (Eu- phorinae) Neobraconinae Hellen, 1957 (nomen nudum) Neoclinocentrina Fischer, 1981 (Doryctinae) Neoneurinae (-ini) Bengtsson, 1918 (Neoneurinae) Odontobraconina Granger, 1949 (Doryctinae) Odontosphaeropygini Zettel, 1990 (Cheloninae) Oncometeorini Tobias, 1987 (Euphorinae) Oncophanina Belokobylskij, 1993 (Rhyssalinae) Opiinae (-ini) Blanchard, 1845 (Opiinae) Orgilinae (-ini) Ashmead, 1900 (Orgilinae) Pambolideina Belokobylskij, 1992 (Doryctinae) Pambolinae (-ini) Marshall, 1885 (Pambolinae) Parahormiini Belokobylskij, 1993 (Hormiinae) Paralipsina Mackauer, 1961 (Aphidiinae) Pedinotina Fischer, 1981 (Doryctinae) Pelecystomini Viereck, 1918 (Rogadinae) Pentatermini Belokobylskij, 1990 (Lysiterminae) Percnobraconini Belokobylskij, 1992 (Doryctinae) Percnobraconoidina Belokobylskij, 1992 (Doryctinae) Perilitini Foerster, 1862 (Euphorinae) Phanerotomellina Zettel, 1990 (Cheloninae) Phanerotomini Baker, 1926 (Cheloninae) Phanomerini Fahringer, 1928 (Exothecinae) Physaraiini van Achterberg, 1984 (Braconinae) Planitorini van Achterberg, 1995 (Betylobraconinae) Pokomandyina Fischer, 1970 (Opiinae) Praini Mackauer, 1961 (Aphidiinae) Praonopterinae Tobias, 1988 (Mesostoinae) Pronkiini van Achterberg, 1990 (Meteorideinae) Prosyntretina Tobias, 1987 (Euphorinae) Protaphidiina Mackauer, 1961 (Aphidiinae) Proteropini van Achterberg, 1976 (Ichneutinae) Pselaphaninae van Achterberg, 1985 (Pselaphaninae) Psenobolina Enderlein, 1912 (Doryctinae) Pseudodicrogeniini Fahringer, 1936 (Braconinae) Pseudohelconina van Achterberg, 1990 (Helconinae) Pseudophanerotomina Zettel, 1990 (Cheloninae) Pseudospathiini Enderlein, 1912 (Doryctinae) Pseudostephaniscini Fahringer, 1928 (Doryctinae) Ptesimogastrina Belokobylskij, 1992 (Doryctinae) Rhaconotina Fahringer, 1928 (Doryctinae) Rhamnurini Fahringer, 1929 (Braconinae) Rhysipolinae Belokobylskij, 1984 (Rhysipolinae) Rhyssalinae (-ini) Foerster, 1862 (Rhyssalinae) Rogadinae (-ini) Foerster, 1862 (Rogadinae) Semionini Tobias, 1987 (Microgastrinae) Sericobraconini Belokobylskij, 1992 (Doryctinae) Sigalphinae (-ini) Haliday, 1833 (Sigalphinae) Siragrini Belokobylskij, 1994 (Doryctinae) Sisupalina Belokobylskij, 1992 (Doryctinae) Spathiini Foerster, 1862 (Doryctinae) Spathioplitina Belokobylskij, 1992 (Doryctinae) Spathiostenina Belokobylskij, 1992 (Doryctinae) Stantonini Viereck, 1919 (Orgilinae) Stegnocellini van Achterberg, 1988 (Blacinae) Stenocorsina Fischer, 1981 (Doryctinae) Stephaniscini Enderlein, 1912 (invalid) Stiropiini van Achterberg, 1993 (Rogadinae) Syngastrini van Achterberg, 1993 (nomen nudum) Syntretini Shaw, 1985 (Euphorinae) Telengaiinae Tobias, 1962 (Telengaiinae) Tetratermini van Achterberg, 1996 (Lysiterminae) Townesilitina Shaw, 1985 (Euphorinae) Toxoneurinae Cresson, 1887 (invalid) Trachypetinae (-ini) Schulz, 1911 (Trachypetinae) Triaspina Viereck, 1918 (Helconinae) Trigonophasmina Belokobylskij, 1992 (Doryctinae) Trioxina Ashmead, 1900 (Aphidiinae) Triterminae Belokobylskij, 1993 (Lysiterminae) Ussurohelconini van Achterberg, 1994 (Cenocoeliinae) Vaepellinae Quicke, 1987 (Vaepellinae) Vervoorrihelconina van Achterberg, 1998 (Helconinae) Victoroviellina Tobias, 1986 (Braconinae) Vipiina Gahan, 1917 (Braconinae) Westwoodiellini van Achterberg, 1992 (Homolobinae) Xiphozelinae van Achterberg, 1979 (Xiphozelinae) Xyeloblacini van Achterberg and Altenhofer, 1997 (Blacinae) Yeliconini van Achterberg, 1991 (Rogadinae) Ypsistocerini Cushman, 1923 (Doryctinae) Zelini Ashmead, 1900 (Euphorinae) Zemiotini \,m Achterberg, 1976 (Euphorinae) J. HYM. RES. Vol. 9(2), 2000, pp. 271-276 Taxonomic Notes on Costa Rican Mendesellinae (Ichneumonoidea: Braconidae), with Description of a New Central American Species of Mendesella A. A. Valerio1 and J. B. Whitfield (AAV) Central American Institute of Biological Research and Conservation (CIBRC), P.O. Box 2398-2050 San Pedro de Montes de Oca, San Jose, Costa Rica, e-mail: avalerio@cibrc.org; (JBW) Department of Entomology, University of Arkansas, Fayetteville, Arkansas 72701, USA, e-mail: jvvhitfie@comp.uark.edu. Abstract. — One new species of mendeselline braconid wasp is described, Mendesella orianae Valerio and Whitfield sp. nov., and the male of Epsilogaster tico Whitfield and Mason is described for the first time. Mendesella orianae from Costa Rica represents the northernmost record of Men- desella in the Americas, and the first recorded species of this genus in Costa Rica. The male genitalia of E. tico is also described, providing the first record of male genitalic characters for the subfamily Mendesellinae. The braconid subfamily Mendesellinae was described by Whitfield & Mason (1994), who included two genera and nine species, all new. So far as is known, the species are endoparasitoids of Lepidop- tera feeding within plant tissue, but bio- logical records are sparse, and specimens of this subfamily are rarely collected in general. Since the original description there- has been no further review of the distribution and biology of the known species, nor any further species described. Recently, new specimens of both genera in the subfamily have been discovered by the senior author in the collections of the In- stituto Nacional de Biodiversidad (INBio). The new records add significantly to the known geographical distribution of men- deselline genera, as well as the morphol- ogy of male Mendesellinae. One of the species is new to science and it is describe below. ' Current Address: Department of Entomology, Uni- versity of Arkansas, Fayetteville, AR 72701 USA, e- mail: avaleri@comp.uark.edu. MATERIAL AND METHODS The morphological terminology used in the species descriptions is that of Huber and Sharkey (1993), and Schuh (1989); ex- cept for the morphology of the propo- deum, which is used sensu Townes (1969, Fig. E). The cuticular sculpturing termi- nology is that of Harris (1979), while the terminology for the wing venation is a variation of the Comstock-Needham sys- tem used by Sharkey and Wharton (1997, Fig. 15). The metasoma of one specimen of Ep- silogaster tico was detached, placed in warm 10% KOH overnight, and run through ethanol and xylene baths into Eu- paral mounting medium for slide-mount- ing. The male genitalia were pulled away from the remainder of the metasoma and illustrated using a microprojector. Epsilogaster tico Whitfield and Mason 1994 (Figs, la, 2, 3) Male. — Body color: Mainlv light yellow; scape and pedicel light brown as face, ver- tex, frons, hind tarsomeres, middle telo- 272 Journal of Hymenoptera Research a b Fig. 1. Wing venation for male Epsilogaster tico Whitfield & Mason (a), and for female Mendesella orianae Valerio & Whitfield sp. nov. (b). tarsus and metasomal terga 4-6; metano- aline (Fig. la); wing veins with light yel- tum with darker yellow than remainder low coloration. Body length = 2.61-2.75 mesosoma; compound eyes silver; palpus mm; Fore wing length = 2.44-2.47 mm. whitish-yellow. Fore and hind wings hy- Head: head height /compound eye height Volume 9, Number 2, 2000 273 2 3 4 Figs. 2-A. 2,3. Epsilogaster tico, male. 2, Face, showing enlarged eyes. 3, Genital capsule. 4, Propodeum and anterior metasomal tergites of Mendesella orianae, sp. now = 1.17-1.18; head height /compound eye length = 1.53-1.65; length of first antennal flagellomere = 0.18-0.22 mm; length of first antennal flagellomere /width of first antennal flagellomere = 4.40-5.20; length of first antennal flagellomere /length of second antennal flagellomere = 1.29-1.44; length of first antennal flagellomere/ length of third antennal flagellomere = 1.38-1.44; distal flagellomere length/ width of distal flagellomere = 4.67-5.0; in- tertentorial pit distance = 0.13-0.16 mm; ocell-ocular distance = 0.03-0.04 mm; dis- tance between toruli and tentorial pits = 0.25-0.27 mm; face wide at dorsal edge of clypeus = 0.28 mm. Antenna with 27 fla- 274 Journal of Hymenoptera Research gellomeres; face nitid, without punctate lose sculpturing, medial area raised form- sculpturing, but with few scattered ob- ing two semicircular lower areas that con- scure rugulose sculpturing features; vertex tain 5 costulae each; terga 2-6 smooth and and gena with coarsely-punctate sculptur- unsculptured, also terga 2-6 much more ing as on posterior-ventral area of head; strongly desclerotized that first metasomal occiput nitid; malar suture present; malar tergum. space very short; ocelli forming an equi- Material examined. — Described from two lateral triangle. Mesosoma: mesosomal males: Costa Rica, Alajuela, RNVS Cafio length /mesosomal width = 1.43-1.55, me- Negro, Playuelas, 20 m. 1-18 /ii/ 1994, Col. sosomal height = 0.60-0.61 mm; hind tibia K. Martinez. length /hind tibia maximum width = Comments. — The studied males have a 3.60-3.89; propleural distal edge with a more elongate metasoma and strongly en- ridge; pronotum with mid longitudinal larged compound eyes than the known fe- area with scrobiculate sculpturing; meson- male (Fig. 2). Also, E. tico male eyes are otum evenly setose throughout; notaulus strongly enlarged in comparison with conspicuously foveate throughout; area at those described for £. bicolor, the only oth- notaulus union without a depressed area er males known for the genus, and some punctate sculpturing present by The male genitalia (Fig. 3) are the first setae; union of notaulus near transscutal to be described for a mendeselline braco- articulation; transscutal articulation with a nid. As with most Cardiochilinae and Mi- smooth carina that is absent laterally; scu- crogastrinae (Maeto 1996), the cuspis and tellar sulcus wide, with 3 deep pits pre- digitus freely articulate, and there are sent; scutellum subpentagonal, without about 10 very small teeth distally on the any posterior apico-medial pits or sculp- digitus, as in some Cardiochilinae. The turing; antero-medial area of metanotum basal ring is longer than in most micro- with two subcircular pits, remainder of gastrines (i.e., forms a broad transverse medial area nitid; axillary troughs of me- band), but shorter than in some derived sonotum with spaced scrobiculate sculp- microgastrine genera. Thus for the most turing; metapleuron dorsally nitid, re- part, mendeselline male genital capsules mainder with areolate-rugulose sculptur- exhibit relatively plesiomorphic features, ing present; propodeum, parallel mid-Ion- as would be expected from their proposed gitudinal carinae united by complete phylogenetic relationships (Whitfield and transversal carinae between them, remain- Mason 1994). der with areolate-rugulose sculpturing present, latero-longitudinal carinae cris- Mendesella orianae Valerio & Whitfield tate as mid-loneitudinal carinae (no other SP' nov- carinae as cristate), areas between carinae vrigs- iD' ^) nitid or with fine obscure colliculate sculp- Male. — Body color: Mainly honey yel- turing present. Metasoma: distal width of low, with antennal flagellomeres as dark tergum 1/ basal width of tergum 1 = brown as inter-ocellar space; scape and 0.38-0.39; length of tergum 1 / distal width pedicel light brown, metasomal terga 2-7 of tergum 1 = 4.0-4.40; first metasomal (except median area of tergum 2 and 3 tergum: dorso-lateral carinae with well- whitish-yellow), hind telo and basitarsus, defined and cristate dorsal carinae curving hind tibia distal 1/5 and fore as middle close to one another apically and extend- leg telotarsus. First metasomal tergum ing 0.70 length of tergum, conspicuously with lateral areas yellow. Head, mesono- cristate basally; rugulose sculpturing turn and hind tibia with a darker tone of densely present dasally on tergum, re- yellow than mesopleuron coloration; com- mainder with more widely spaced rugu- pound eyes silver; palpus whitish-yellow. Volume 9, Number 2, 2000 275 Fore and hind wing slightly infuscate throughout; wing veins light brownish- yellow, except hind RS and 2M nebulose (with a short basal area tubular) with a dark brown coloration. Body length = 3.80 mm; Fore wing length = 3.19 mm. Head: head height /compound eye height = 1.25; head height /compound eye length = 1.67; length of first antennal flagellom- ere = 0.26 mm; length of first antennal fla- gellomere/ width of first antennal flagel- lomere = 3.2; length of first antennal fla- gellomere/ length of second antennal fla- gellomere = 1.33; length of first antennal flagellomere/ length of third antennal fla- gellomere = 1.45; distal flagellomere length /width of distal flagellomere = 3.0; intertentorial pit distance = 0.18 mm; ocell-ocular distance = 0.05 mm; distance between toruli and tentorial pits = 0.28 mm; face wide at dorsal edge of clypeus = 0.36 mm. Antenna with 38 flagellom- eres, distal flagellomere with a long and thick spine; face, frons, vertex, gena (ex- cept ventral area) densely punctate; clyp- eus mainly nitid, with scattered punctate sculpturing present; malar suture present; malar space very short; ocelli forming an equilateral triangle. Mesosoma: mesoso- mal length /mesosomal width = 1.64, me- sosomal height = 0.82 mm; hindtibial length /hindtibial width = 3.5; Pronotal lateral areas with few obscure longitudi- nal lineate sculpturing; mesonotum even- ly punctate throughout as setose; notaulus foveate anteriorly, remainder with ob- scure foveate sculpturing; area at notaulus union with rugulo-punctate sculpturing in a depressed area; union of notaulus near transscutal articulation; mesopleuron nit- id, setose ventrally and remainder scat- tered setose; sternaulus obscurely im- pressed and nitid; transscutal articulation with a smooth carina; scutellar sulcus deep and wide, with 9 conspicuous pits present; scutellum subpentagonal, with two subrectangular pits close to one an- other at posterior apico-median area; an- tero-medial area of metanotum with two subcircular pits near scutellar subrectan- gular pits, distal edge punctate through- out; axillary troughs of mesonotum with coarsely-rugose sculpturing; propodeum with parallel mid-longitudinal carinae united by transversal carinae between them, remainder with areolate-rugulose sculpturing present, latero-longitudinal carinae as cristate as mid-longitudinal ca- rinae (no other carinae as strongly cris- tate). Metasoma: Distal width of tergum 1/ Basal width of tergum 1 = 1.10; Length of tergum 1/ Distal width of tergum 1 = 2.11; first metasomal tergum, dorso-lateral carinae strong, dorsal carinae curving close to another apically and not conspic- uously cristate, rugulose sculpturing throughout tergum, medial area raised forming two semicircular lower areas that contain 5 costulae each; terga 2-6 smooth and unsculptured; tergum 2 with a "E" shaped form, the mid-dorsal part as strongly sclerotized as lateral areas of it, but not reaching half of the length of the tergum. Holoti/pe. — Costa Rica, Guanacaste, Parque Nacional Guanacaste, Estac. Los Almendros, 300 m. 23/iii-28/iv/1994, Col. E. Lopez. Deposited in INBio. Comments. — This species is similar to Mendesella magna and M. braziliensis, both described by Whitfield & Mason (1994), but can be separated from them by the ab- sence of two elongated mid-lateral pits on the first metasomal tergum, by metasomal tergum 2 having an "E" shaped form with the mid-dorsal area as strongly sclerotized as the lateral areas, and the presence of rugulose sculpturing throughout the first metasomal tergum. Also, M. orianae can be separated from any other Mendesella spe- cies by the fore wing pattern of very weak infuscation, the mesopleuron and prono- tum not being darkened, the smaller bodv size, and the darker metasomal terga 3-6. Etymology. — Gender: feminine. The pre- sent species is named in honor of Oriana Valerio Contreras; live longer and pros- per! 276 Journal of Hymenoptera Research DISCUSSION At present, the appearance of Mendesella orianae in Costa Rica expands the northern limit of known distribution for the genus Mendesella. Until now, the genus was re- ported only from Brazil, Bolivia and Ec- uador (Whitfield & Mason 1994) with no observed specimens of the genus in the Caribbean zone of America. In contrast, the genus Epsilogaster has a reported northern limit of the southern U.S. (in- cluding the Caribbean area), with a south- ern limit in Brasil. The present observed sparse distribution is likely to be the result of the lack of specimens collected in other areas, in combination with the difficulty of identifying braconid wasps. ACKNOWLEDGMENTS We would like to thank The Instituto Nacional de Biodiversidad (INBio) for the loan of their mendesel- line material. Thanks also to Axel Retana for his ac- curate comments, to Andy Deans for his help in the wing drawing technique, to "la canalla biologica de la calle de la perdision" in Costa Rica for their con- stant support, and of course to "Cukha" for her un- conditional presence. LITERATURE CITED Harris, R. A. 1979. A glossary of surface sculpturing. Occasional Papers of the Bureau of Entomology of the California Department of Agriculture. No. 28, p 32. Huber, J. T. and M. J. Sharkey. 1993. Structure, Chap- ter 3. Pp 13-59. In: Goulet, H. and J.T. Huber (eds.), Hymenoptera of the world: An identification guide of families, Agriculture Canada Publication 18894/E, Ottawa. Maeto, K. 1996. Inter-generic variation in the external male genitalia of the subfamily Microgastrinae (Hymenoptera, Braconidae), with a reassessment of Mason's tribal system. Journal of Hymenoptera Research 5: 38-52. Schuh, R. T. (ed.). 1989. The Torre-Bueno Glossary of Entomology (revised edition). The New York En- tomological Society, New York, 849 pp. Sharkey, M. J. and R. A. Wharton. 1997. Morphology and terminology, pp. 19-37 in: Wharton, R. A., P. M. Marsh and M. J. Sharkey, eds., Manual of the New World Genera of the Family Braconidae (Hy- menoptera), Special Publication of the Internation- al Society of Hymenopterists, No.l. 439 pp. Townes, H. 1969. The genera of Ichneumonidae, Part 1; Ephialtinae to Agriotypinae. Memoirs of the Amer- ican Entomological Institute 11: 1-300. Whitfield, J. B. 1997. Mendesellinae, pp. 320-323 in: Wharton, R. A., P. M. Marsh and M. J. Sharkey, eds., Manual of the New World Genera of the Family Braconidae (Hymenoptera), Special Publication of the International Society of Hymenopterists, No.l. 439 pp. Whitfield, J. B. & W. R. M. Mason. 1994. Mendesel- linae, a new subfamily of the braconid wasp (Hy- menoptera, Braconidae) with a review of rela- tionships within the microgastroid assemblage. Systematic Entomology 19: 61-76. J. HYM. RES. Vol. 9(2), 2000, pp. 277-287 Revision of the Enigmatic Genus Marshiella Shaw in the New World with the Description of Three New Species (Hymenoptera: Braconidae: Euphorinae) Scott R. Shaw and Paul M. Marsh (SRS) U.W. Insect Museum, c/o Department of Renewable Resources, University of Wyoming, Laramie, Wyoming 82071-3354 USA; (PMM) Cooperating Scientist, USDA Systematic Entomology Laboratory, c/o U.S. National Museum of Natural History, MRC-168, Washington, D.C. 20560, USA (correspondence address: P.O. Box 384, North Newton, Kansas 67117 USA) Abstract. — The genus Marshiella Shaw is revised for the New World region. Included are two previously described species, Marshiella plumicornis (Ruthe) and M. pulvillicornis (Walley and MacKay), and three newly described species, M. bobella Shaw, M. lettermani Shaw, and M. marshi Marsh. A key to New World species is provided, along with species descriptions, diag- nostic characters, distribution, antennal micromorphology, and phylogenetic patterns. The euphorine braconid genus Marshiel- la was erected by Shaw (1985) to include two enigmatic species with unusually modified antennae that had previously been placed in other genera. Marshiella plumicornis (Ruthe) was formerly placed in Microctonus Wesmael (Shenefelt 1969), while M. pulvillicornis (Walley and MacKay) was previously placed in Streblo- cera Westwood (Walley and MacKay 1963),. but Shaw (1985) demonstrated that these species were closely related based on the uniquely modified, densely setose basal flagellomeres (Figs. 1-2) and must be assigned to a new genus. Previously the genus was known only from the Hol- arctic region, extending as far south as Mexico in the New World (Shaw 1985). More recently the genus has been record- ed as far south as Costa Rica (Shaw 1997) and two new species have been recentlv described from China (Chen and van Ach- terberg 1997) extending the known distri- bution to the Oriental region. In this paper the New World species are reviewed and three new species are described including material from Arizona, Texas, Mexico, Costa Rica, and Brazil. A full generic diagnosis for the genus Marshiella has been published by Shaw (1985) and more recently by Chen and van Achterberg (1997), so it is not necessary to repeat that information here. Recognition of the genus is quite easy as it is the only braconid with flagellomeres 1-4 or 1-5 flattened and densely setose ventrally (as in Figs. 1-2, 5-6, 10-16, 18-19). Specimens can be keyed to genus using the keys pro- vided by Shaw (1985), Chen and van Ach- terberg (1997), or Shaw (1997). Very little is known about the biology of Marshiella species, but they are pre- sumed to be koniobiont endoparasitoids of adult Coleoptera as are most other bas- al clades of the Euphorinae (Shaw 1985, 1988; Shaw and Huddleston 1991). Only one species, M. plumicornis (Ruthe) has been reared from a host, an anthicid bee- tle, Notoxus monoceros L. (Smith 1953; Gor- nitz 1937). The same species has been col- lected at cantharadin-baited traps in Mich- igan, indicating that Marshiella species probably orient to their hosts using chem- ical cues (Shaw 1985; Dettner 1997). Most- ly Marshiella species are quite rare, with 278 Journal of Hymenoptera Research the Canadian species M. pulvillicornis (Walley and MacKay) and the two Chi- nese species, M. binarius Chen and van Achterberg and M. sinensis Chen and van Achterberg, being known only from the holotypes. However, two of the new spe- cies treated in this paper have been col- lected in series from Malaise traps in Cos- ta Rica, and occasionally are attracted to lights. Males are totally unknown. The function of the modified flagellum in Mar- shiella females has not been observed, but its form suggests the possibility that it is an adaptation for grasping the host beetle during oviposition since female euphori- nes oviposit by swinging the metasoma ventrally and anteriorly and exserting the ovipositor forward between the legs and in front of the face of the advancing wasp. Authorship for new species is by either Shaw or Marsh, as indicated for each spe- cies. KEY TO THE FEMALES OF NEW WORLD SPECIES OF MARSHIELLA SHAW 1. Dorso-lateral areas of propodeum entirely rugulose (Fig. 3); flagellomeres 2-A in dorsal view only slightly broader than flagellomeres 8-10 Marshiella plumicornis (Ruthe) - Dorso-lateral areas of propodeum entirely smooth and highly polished (Fig. 4); flagellom- eres 2-A in dorsal view strongly widened and heart-shaped (Figs. 5-6), about 2X broader than flagellomeres 8-10 2 2. Flagellomere 5 in dorsal view nearly cylindrical, and only slighter wider than flagellomere 6, not strongly flattened 3 - Flagellomere 5 in dorsal view strongly flattened, distinctly heart-shaped and nearly 2X broader than flagellomere 6 4 3. Body size very small, less than 2 mm long; antenna short, with only 17 flagellomeres; mesosoma reddish brown; known only from Canada Marshiella pulvillicornis (Walley and MacKay) - Body size larger, more than 2 mm long; antenna longer, with 19-21 flagellomeres; meso- soma jet black; known only from Costa Rica .... Marshiella lettermani Shaw, new species 4. Metasomal tergum 1 entirely smooth on posterior half, beyond spiracles; ocellar-ocular space pale yellowish white; known only from Brazil Marshiella tnarshi Marsh, new species - Metasomal tergum 1 finely rugulose on posterior half, beyond spiracles; ocellar-ocular space entirely or partly dark chocolate brown to black; know from Arizona, Texas, Mexico, and Costa Rica Marshiella bobella Shaw, new species Marshiella bobella Shaw, new species (Figs. 1-2, 4, 6, 8, 12-13) Description of holoti/pe female. — Body length 2.3 mm; forewing length 2.3 mm; ovipositor length 1.1 mm. Color: frons, vertex, temple, and entire mesosoma black; flagellomeres 1-5 dorsally, remain- der of flagellum, wing venation, entire metasoma and ovipositor sheath dark chocolate brown; scape, pedicel, flagel- lomeres 1-5 ventrally, remainder of head, legs entirely, and ovipositor light yellow- ish brown; eye silvery gray; wing mem- brane hyaline; setae, especially on flagel- lomeres 1-5 ventrally, pale silvery white. Head: scape 2.3 X longer than wide, apical ventral margin protruding and carinate longitudinally; flagellomeres 1-5 modi- fied, pulvilliform, heart-shaped in dorsal view, ventrally densely setose with long wispy setae (Figs. 12-13); flagellomere 1 2x longer than wide; flagellomere 2 about as long as wide; flagellomeres 3-4 slightly wider than long; flagellomere 5 smaller, about as long as wide; remainder of fla- gellum comprising 15 flagellomeres of Volume 9, Number 2, 2000 279 Figs. 1-2. Marshiella bobella Shaw, anterior view. 1, head and antennae. 2, basal section of antenna show- ing modified flagellomeres 1-5. normal form; face somewhat protruding below antennal insertions; eyes conver- gent ventrally, closest near mid-point of face; ocelli small, ocellar-ocular distance 5x width of lateral ocellus. Mesosoma: mesoscutum smooth and highly polished except notauli finely foveate, lateral lobes mostly devoid of setae; scutellar furrow 2- foveate; scutellar disc and dorso-lateral faces of propodeum smooth and highly polished; mesopleuron smooth and highly polished except smooth and finely-foveate sternaulus; length of marginal cell 0.6X length of pterostigma; smooth dorso-lat- eral surfaces of propodeum margined pos- teriorly by V-shaped carinae; posterior surface of propodeum slightly depressed medially, margined laterally by carinae; median line of propodeum, posterior, and lateral surfaces rugulose, but polished and Figs. 3-4. Propodeum, dorso-lateral view, 220X. 3, Marshiella plumicornis (Ruthe). 4, Marshiella bobella Shaw. shining. Metasoma: petiolate tergum 1 narrow basally, then gradually wider, posterior width 3x wider than extreme basal width; petiole 7x longer than wide at extreme base; dorsal surface of petiole finely longitudinally rugose; remainder of metasoma smooth and highly polished; ovipositor length 2X length of tergum 1, basally emerging from longitudinal ven- tral slit about at mid-point of metasoma below tergum 3, widely separated from base of ovipositor sheath. Variation. — paratype females appearing similar to holotype except apical 'normal' section of flagellum with 14 or 15 flagel- lomeres. The ovipositor is extremely flex- ible and varies in appearance depending on its position at death. Ovipositor shape varies from curved along basal Vi to nearly straight. Depending on the degree to which the tip of the metasoma is flexed ventrally and anteriorly the ovipositor base may appear to emerge apically or 280 Journal of Hymenoptera Research ventrally near the mid-point of the meta- flagellomere 5 modified along with flagel- soma. The specimens from Texas have less lomeres 1-4, larger body size, and longer dark brown color on the top of the head, flagellum. The only other Marshiella spe- but otherwise agree with the diagnosis of cies with flagellomere 5 modified is Mar- this species. shiella marshi from Brazil, which can be Material examined. — Holotype female: distinguished by its smooth metasomal Costa Rica, San Jose Province: Zurqui de tergum 1 and lighter head color. Moravia, 1600 m, April 1992, P. Hanson, Etymology. — The species name is an ar- Malaise trap, deposited at University of bitrary combination of letters to form a eu- Wyoming. Paratype females: 1 same data phonious combination, as holotype; 1 same data except July 1990; 6 same data except November-December Marshiella lettermani Shaw, new 1990; 1 same data except May 1992; 3 same species data except June 1992; 1 same data except (Fl§s- 14_16) July 1992; 1 same data except February Description of holotype female. — Body 1996. Cartago Province: 1, La Cangreja, length 2.9 mm; forewing length 2.8 mm; 1950m, July 1991, P. Hanson, Malaise trap; ovipositor length 1.3 mm. Color: frons me- 1 same data except June-July 1992. Guan- dially, ocellar triangle, and entire meso- acaste Province: 1, P. N. Guanacaste, 9 km soma except prosternum black; flagellom- S Santa Cecilia, Estacion Pitilla, 700 m, 2- eres 1-4 dorsally, remainder of flagellum, 19 March 1992, P. Rios, INBio barcode prosternum, wing venation, entire meta- CR1 000^420531; 1 same data except 31 soma and ovipositor sheath dark choco- March-15 April 1992, INBio barcode late brown; scape, pedicel, flagellomeres CR1000-771575; 1, Tierras Morenas, 700m, 1-4 ventrally, remainder of head, legs en- December 1992, G. Rodriguez, INBio bar- tirely, and ovipositor very pale yellowish code CR1001-288145; 1, Santa Rosa Na- brown; eye silvery gray; wing membrane tional Park, 300 m, 14 August-6 Septem- hyaline; setae, especially on flagellomeres ber 1986, I. D. Gauld and D. Janzen, Mai- 1-5 ventrally, pale silvery white. Head: aise trap, Bosque Humedo, mature ever- scape 2.3 X longer than wide, apical ven- green dry forest, fully shaded. Mexico: 1, tral margin protruding and carinate lon- Chis., L. Montebello National Park, 5000 gitudinally; flagellomeres 1-4 modified, ft., 30 May 1969, Malaise trap. U.S.A.: 1, pulvilliform, heart-shaped in dorsal view, Arizona, 5 mi. W Portal, 7 July 1956, O. L. ventrally densely setose with long wispy Cartwright, light trap; 2, Texas, Sabine setae on flagellomeres 1 and 4, shorter Co., 9 mi. E. Hemphill, 23 June to 2 July bent Velcro-like setae on flagellomeres 2- 1989, Anderson and Morris, flight inter- 3 (Figs. 14-16); flagellomere 1 2X longer cept trap in beech /magnolia forest. Para- than wide; flagellomere 2 about as long as types deposited at University of Wyo- wide; flagellomere 3 slightly wider than ming, Universidad de Costa Rica, Instituto long; flagellomere 4 smaller, about as long Nacional de Biodiversidad, Texas A&M as wide; remainder of flagellum compris- University, and U.S. National Museum of ing 17 flagellomeres of normal form; face Natural History. somewhat protruding below antennal in- Comments. — Specimens of this species sertions; eyes convergent ventrally, closest from Arizona and Mexico had been pre- near mid-point of face; ocelli small, ocel- viously treated as variations of M. pulvil- lar-ocular distance 3.5 X width of lateral licornis by Shaw (1985), but the substantial ocellus. Mesosoma: mesoscutum smooth series of specimens now available from and highly polished except notauli finely Costa Rica show this to be a distinct spe- foveate, lateral lobes mostly devoid of se- cies most easily separated by its having tae; scutellar furrow 2-foveate; scutellar Volume 9, Number 2, 2000 281 Figs. 5-6. Basal section of antennal flagellum, dorso- lateral view. 5, Marshiella pulvillicornis (Walley and MacKay), 230 X. 6, Marshiella bobella Shaw, 235 X. disc and dorso-lateral faces of propodeum smooth and highly polished; mesopleuron smooth and highly polished except smooth and coarsely-foveate sternaulus; length of marginal cell 0.9 x length of pter- ostigma; smooth dorso-lateral surfaces of propodeum margined posteriorly by V- shaped carinae; posterior surface of pro- podeum slightly depressed medially, mar- gined laterally by carinae; median line of propodeum, posterior, and lateral surfaces rugulose, but polished and shining. Me- tasoma: petiolate tergum 1 narrow basal- ly, then gradually wider, posterior width 2.8 X wider than extreme basal width; pet- iole 6X longer than wide at extreme base; dorsal surface of petiole finely longitudi- nally rugose on basal 3/4; posterior 1/4 of petiole and remainder of metasoma smooth and highly polished; ovipositor length 1.9X length of tergum 1, basally emerging from longitudinal ventral slit about at mid-point of metasoma below tergum 3, widely separated from base of ovipositor sheath. Figs. 7-9. Metasomal tergum 1, dorso-lateral view. 7, Marshiella plumicornis (Ruthe), 220x. 8, Marshiella bobella Shaw, 195X; 9, Marshiella marshi Marsh, 220x. 282 Journal of Hymenoptera Research Figs. 10-11. Marshiella plumicornis (Ruthe), basal section of antennal flagellum, ventral view. 10, flagellomeres 1-5, 295 X. 11, detail of flagellomeres 1-3 showing spatulate setae, 700x. Variation. — paratype females appearing similar to holotype except apical 'normal' section of flagellum with 15 to 17 flagel- lomeres, body size 2.2 to 2.9 mm, and dark brown covering most of frons and vertex. Material examined. — Holotype female: Costa Rica, Puntarenas Province: San Vito, Estacion Biological Las Alturas, 1500 m, December 1991, P. Hanson, Malaise trap, deposited at University of Wyoming, Lar- amie. Paratype females: 1 same data as ho- lotype; 1 same data except November 1991; 2 same data except June 1992. Para- types deposited at University of Wyoming and Universidad de Costa Rica. Comments. — Marshiella lettermani is most similar to M. pulvillicornis (Walley and MacKay) in that both of these species have flagellomeres 1-4 greatly widended and densely setose, but flagellomere 5 is not so modified. The Costa Rican species M. let- termani can be distinguished from M. pul- villicornis by its larger body size (nearly 3 mm), black mesosoma, short bent Velcro- like setae on flagellomeres 2-3 (Fig. 16), and longer flagellum with 19-21 flagel- lomeres. M. pulvillicornis is much smaller (less than 2 mm), has a reddish brown me- sosoma, shorter flagellum with only 17 fla- gellomeres, entirely long wispy flagellar setae, and is known only from Canada. Etymology. — The species name is a pa- tronym for David Letterman, host of the Late Show, in appreciation for his out- standing contributions to late night enter- tainment. Many a dull day has been im- Volume 9, Number 2, 2000 283 Figs. 12-13. Marshiella bobella Shaw, basal section of antennal flagellum, ventral view. 12, flagellomeres 1- 5, 215x. 13, detail of flagellomere 5 showing long wispy setae, 905 X. proved by his humor. It somehow seems appropriate that a really weird insect should be named in his honor. Marshiella marshi Marsh, new species (Figs. 9, 17-19) Description of holotype female. — Body length 2.9 mm; forewing length 2.8 mm; ovipositor length 1.2 mm. Color: frons, vertex, and temple medially light brown; margins of ocelli and notauli black; flagel- lomeres 1-5 dorsally, remainder of flagel- lum, wing venation, mesosoma except prosternum, metasoma except petiole ba- sally, and ovipositor sheath dark reddish brown; scape, pedicel, flagellomeres 1-5 ventrally, remainder of head, legs entirely, and ovipositor very pale yellowish brown to yellowish white; eye silvery gray; wing membrane hyaline; setae, especially on flagellomeres 1-5 ventrally, pale silvery white. Head: scape 3.0 x longer than wide, apical ventral margin protruding (Fig. 17) and carinate longitudinally; flagellomeres 1-5 modified, pulvilliform, heart-shaped in dorsal view, ventrally densely setose with long wispy setae, some of which are intertwined or braided (Figs. 18-19); fla- gellomere 1 1.5X longer than wide; flagel- lomere 2 about as long as wide; flagellom- eres 3^4 slightly wider than long; flagel- lomere 5 smaller, about as long as wide; remainder of flagellum comprising 15 fla- gellomeres of normal form; face somewhat protruding below antennal insertions; eyes convergent ventrally, closest near mid-point of face; ocelli small, ocellar-oc- ular distance 4X width of lateral ocellus. Mesosoma: mesoscutum smooth and highly polished except notauli finely fo- veate, lateral lobes mostly devoid of setae medially; scutellar furrow 2-foveate; scu- tellar disc and dorso-lateral faces of pro- podeum smooth and highly polished; me- sopleuron smooth and highly polished ex- cept smooth and finely-foveate sternaulus; length of marginal cell 0.7x length of pterostigma; smooth dorso-lateral surfaces of propodeum margined posteriorly by V- shaped carinae; posterior surface of pro- podeum slightly depressed medially, mar- gined laterally by carinae; median line of propodeum, posterior, and lateral surfaces rugulose, but polished and shining. Me- tasoma: petiolate tergum 1 narrow basal- ly, then gradually wider, posterior width 3x wider than extreme basal width; peti- ole 7x longer than wide at extreme base; dorsal surface of basal 1 /2 of petiole finely longitudinally rugose; posterior 1/2 of petiole and remainder of metasoma smooth and highly polished; ovipositor length 2x length of tergum 1, basally emerging from longitudinal ventral slit about at mid-point of metasoma below tergum 3, widely separated from base of ovipositor sheath. Material examined, — Holotype female: Brazil, Rondonia, Vilhena, 21 degrees 40 minutes S, 60 degrees 08 minutes W, Oc- tober 1973, M. Alvarenga, deposited at Ca- nadian National Collection, Ottawa. 284 Journal of Hymenoptera Research Figs. 14-16. Marshiella lettermani Shaw, basal section of antennal flagellum, ventral view. 14, flagellomeres 1-5, 215X. 15, flagellomeres 3-4, 745X, showing variation of seta form and density. 16, flagellomere 3, 1450X, showing detail of short, bent, Velcro-like setae. Comments. — Marshiella marshi can be separated from most other species by its having flagellomere 5 modified along with flagellomeres 1-4, larger body size, and longer flagellum. The antennal scape is slighly more protruberant and setose (Fig. 17) than other species. Some of the long wispy setae on flagellomeres 2-3 are in- tertwined or braided (Fig. 18-19), but the sample size is too limited to determine if this is natural or a post-mortem effect of preservation methods. The only other Marshiella species with flagellomere 5 modified is M. bobella from Arizona, Costa Rica, and Mexico, which can be distin- guished from M. marshi by its rugulose metasomal tergum 1 and darker head col- or. Etymology. — The species is named in honor of Jon Marsh, son of the junior au- thor. Marshiella plumicornis (Ruthe) (Figs. 3, 7, 10-11) Microctonus plumicornis Ruthe, 1856. Re- classified by Shaw, 1985. Description of female based on North Amer- ican material. — Body length 1.8-1.9 mm; forewing length 1.7-1.8 mm; ovipositor length 0.6-0.7 mm. Color: frons, vertex, temple, gena, mesosoma except pronotum and prosternum, flagellomeres 4-16, wing venation, entire metasoma and ovipositor sheath dark chocolate brown to black (fad- ing to yellowish brown in old specimens); scape, pedicel, flagellomeres 1-3, remain- der of head, legs entirely, and ovipositor light yellowish brown; eye silvery gray; wing membrane hyaline; setae, especially on flagellomeres 1—4 ventrally, pale sil- very white. Head: scape 2.2 X longer than wide, apical ventral margin not protrud- ing or carinate longitudinally; flagellom- eres 1-4 modified, pulvilliform, narrowly heart-shaped in dorsal view, ventrally densely setose with setae expanded and flattened apically into spatulate tips (Figs. 10-11); flagellomere 1 3x longer than wide; flagellomeres 2-3 about 2X as long as wide; flagellomere 4 1.5X wider than long; remainder of flagellum comprising 12 flagellomeres of normal form; face not protruding below antennal insertions; eyes convergent ventrally, closest near ventral margin of face; ocelli small, ocel- lar-ocular distance 5X width of lateral Volume 9, Number 2, 2000 285 Figs. 17-19. Marshiella marshi Marsh, basal section of antenna. 17, scape and pedicel, lateral view, 380x. 18, flagellomeres 1-5, ventral view showing long wispy setae, 250x. 19, flagellomere 2, ventral view showing detail of intertwined, braided setae, 885x. ocellus. Mesosoma: mesoscutum smooth and highly polished except notauli fove- ate, median and lateral lobes mostly de- void of setae; scutellar furrow 2-foveate; scutellar disc smooth and highly polished; mesopleuron smooth and highly polished except coarsely-foveate sternaulus; length of marginal cell 0.4 X length of pterostig- ma; rugulose dorso-lateral surfaces of pro- podeum margined posteriorly by V- shaped carinae; posterior surface of pro- podeum slightly depressed medially, mar- gined laterally by carinae; median line of propodeum, dorso-lateral, posterior, and lateral surfaces rugulose, but somewhat polished and shining. Metasoma: petiolate tergum 1 narrow basally, then gradually wider, posterior width 3x wider than ex- treme basal width; petiole 5X longer than wide at extreme base; dorsal surface of petiole finely longitudinally rugose; re- mainder of metasoma smooth and highly polished; ovipositor length 1.6X length of tergum 1, basally emerging from longitu- dinal ventral slit apically to near mid- point of metasoma below tergum 3, wide- ly separated from base of ovipositor sheath. Material examined. — U.S.A., Arizona: 1 female, Safford, 4 November 1955, G. D. Butler, swept from alfalfa. California: 1 fe- male with cocoon, Chino, July 1932, A. J. Basinger, ex. peaches infested by A. linea- tella. Florida: 1 female, St. Lucie Co., 31 March-5 April 1930, J. R. Barass, Florida fruit fly trap survey; 1 female, Indian Riv- er, 10 March 1930, J. R. Barass, Florida fruit fly trap survey. Michigan: 2 females, Ingham Co., Dewitt Township, 10-17 June 1982, D. K. Young, taken at cantharadin bait. 1 female, Kalamazoo Co., Harrison Lake, T3S, R12W, sec. 34, 16-26 June 1982, J. K. Young, taken at cantharadin bait. South Carolina: 1 female, Clemson, no date, G. G. Ainslie; 1 female, Clio, 22 July 1936, on cotton, lot 36-30265. Deposited at U.S. National Museum of Natural History, Washington, D.C. Holotype female from Germany examined by Shaw (1985), de- posited at the Natural History Museum, London. Comments. — This species has the widest distribution of any Marshiella species, be- ing recorded from both Europe (Chen and van Achterberg 1997) and the United States (Shaw 1985). M. plumicornis can be easily distinguished from all other New World species of Marshiella by the dorso- 286 Journal of Hymenoptera Research lateral areas of the propodeum being en- discovered, therefore there is no need to tirely rugulose and flagellomeres 2-4, in redescribe the species here. In addition to dorsal view, being only slightly wider other characters mentioned in the key, this than flagellomeres 8-10. species can be distinguished from other The distinctive flagellar setae micromor- Marshiella by its small body size (less than phology (Figs. 10-11), with the ventral se- 2mm), short flagellum with 17 flagellom- tae of flagellomeres 1-4 having flattened, eres, and reddish brown mesosoma. spatulate tips, was previously document- Discussion of PJiylogenetic Consider- ed by Shaw (1985). It was previously pre- ations. — Our understanding of variation sumed that this setal micromorphology for characters of possible phylogenetic sig- was a characteristic of the genus Marshiella nificance in Marshiella species is no doubt (Shaw 1985; Chen and van Achterberg limited by the scarcity of material for sev- 1997), however, the more complete survey eral species, and also the lack of biological presented here indicates that setal micro- data for all but one species. Nevertheless, morphology varies among Marshiella spe- it is tempting to speculate on the possible cies, and the spatulate form has so far only phylogenetic interpretation of several been documented in M. plumicornis. characters, especially the unique antennal Biology. — M. plumicornis has been reared modifications which are presumptive syn- from the anthicid Notoxus monoceros L. in apomorphies by out-group comparison Europe (Gornitz 1937; Smith 1953), but it with the presumed sister-group, Toumesi- has not yet been reared in North America, lit us (Shaw 1985). It has been attracted to cantharadin bait Four of the New World species appear (Shaw 1985), suggesting possible chemo- to form a distinctive monophyletic cluster sensory location of coleopteran hosts including M. bobella, M. lettermani, M. mar- (Dettner 1997). Dan Young (pers. comm.) shi, and M. pulvillicornis. Synapomorphies reports having seen braconids attracted to supporting this lineage include greatly cantharadin in Michigan (presumably this widened and strongly flattened basal fla- species) attacking anthicids near the trap, gellomeres (Fig. 18), modified scape with One specimen was reared from peaches the apical ventral margin protruding (Fig. infested with the gelechiid Anarsia linea- 17), dorsum of propodeum with large tella Zeller (the peach twig borer), but it smooth areas devoid of sculpture (Fig. 4), seems unlikely that this was the actual and face with pale coloration. Although host (more likely it was a beetle in the the two Chinese species were not available same substrate). for examination, the published descrip- tions agree more closely with the condi- Marshiella pulvillicornis (Walley and tk)ns seen in the more basal M plumicornis JVlacKay) which has more cylindrical and less flat- ^ §• ' tened flagellomeres, shorter and less pro- Streblocera pulvillicornis Walley and truding scape, rugose propodeum, and MacKay, 1963. Reclassified by Shaw, 1985. brown face. The significance of a derived Material examined. — Holotype female, cluster including M. bobella, M. lettermani, Canada, Quebec, Gatineau Park, Meach M. marshi, and M. pulvillicornis is two-fold. Lake, 9 June 1961, G.S. Walley, sweeping, First, this indicates that all the more basal No. 8223, deposited in Canadian National species of Marshiella have holarctic or pa- Collection, Ottawa. learctic distributions, while all the species Comments. — This species was described with Neotropical distributions belong to a by Walley and MacKay (1963) based on a derived strictly New World cluster. This single female specimen from Quebec. No is consistent with an hypothesis of one in- new material of the species has since been vasion of South America, from the north Volume 9, Number 2, 2000 in- temperate zone. Second, it shows that the one species with known biology is basal, suggesting the possibility that the highly modified antennae of the more southern New World species may also be modifi- cations for locating hosts via cantharadin or other semiochemicals. ACKNOWLEDGMENTS Special thanks to Prof. Paul Hanson, of the Univ- ersidad de Costa Rica, for running the Malaise Trap network over many years, which yielded most of the Costa Rican material treated here. Specimens were also provided bv the Canadian National Collection, Ottawa, the Instituto Nacional de Biodiversidad (IN- Bio), Heredia, and the U.S. National Museum of Nat- ural History, Washington, D.C. Thanks to Prof. Dan Young, of the University of Wisconsin, for sharing his observations on braconids attracted to cantharadin. LITERATURE CITED Chen, X. And C. van Achterberg. 1997. Revision of the subfamily Euphorinae (excluding the tribe Meteorini Cresson) (Hvmenoptera: Braconidae) from China. Zoologische Verhandelingen 313: 1- 217. Dettner, K. 1997. Chapter 8. Inter- and intraspecific transfer of toxic insect compound cantharadin. Pp. 115-145, In: Dettner, K. et al. (eds.), Vertical food web interactions, Ecological Studies, Vol. 130, Springer- Verlag, Berlin, Heidelberg. Gornitz, K. 1937. Cantharadin als gift und anlock- ungsmittel fur insekten. Arbeit Physiologische An- gewandten Entomologie Berlin-dahlem 4: 116-157. Shaw, M. R. and T. Huddleston. 1991. Classification and biology of braconid wasps (Hvmenoptera: Braconidae). Handbooks for the identification of Brit- ish insects 7: 1-126. Shaw, S. R. 1985. A phylogenetic study of the subfam- ilies Meteorinae and Euphorinae (Hymenoptera: Braconidae). Entomography 3: 277-370. Shaw, S. R. 1988. Euphorine phvlogeny: the evolution of diversity in host-utilization by parasitoid wasps (Hymenoptera: Braconidae). Ecological En- tomology 13: 323-335. Shaw, S.R. i997. Subfamily Euphorinae. Pp. 234-234, In: Wharton, R. A., P. M. Marsh and M. J. Shar- key, eds. Manual of the New World genera of the family Braconidae. Special publication of the In- ternational Society ofHymenopterists. No. 1, 438 pp. Shenefelt, R. D. 1969. Hymenopterorum Catalogus, Braconidae 1, Euphorinae. W. Junk, The Hague, 176 pp. Smith, O. J. 1953. Species, distribution, and host re- cords of the braconid genera Microctonus and Perilitus (Hvmenoptera: Braconidae). Ohio Journal of Science 53: 173-178. Walley, S. G. and M. R. MacKay. 1963. The discovery of Streblocera in Canada (Hymenoptera: Braconi- dae). The Canadian Entomologist 95: 999-1001. J. HYM. RES. Vol. 9(2), 2000, pp. 288-291 Description of a New Gregarious Species of Aleiodes Wesmael (Hymenoptera: Braconidae: Rogadinae) Joseph C. Fortier Gonzaga University, Spokane, Washington 99258, USA Abstract. — Aleiodes leptocarina Fortier, a new species from Costa Rica, is described and illustrated. Specimens were reared from a large lepidopteran caterpillar, making this species one of only three presently known gregarious Aleiodes species. Morphological comparisons are made with A. stig- mator, the other New World gregarious species. The rogadine braconid genus Aleiodes clusively on non-catocaline noctiiids. They Wesmael is worldwide in distribution suggest an evolutionary pattern for koino- (Shaw et ah 1997). Fortier and Shaw (1999) biont endoparasitoids in which more de- list 208 Aleiodes species worldwide. Evi- rived species tend to have narrower host dence (Fortier, unpublished data) suggests ranges than less derived species (Fortier that it may be far more species-rich in the and Shaw 1999). Neotropics than had heretofore been Evidence (Fortier and Shaw 1999) sug- known. gests a coevolutionary trend between Al- Species of tribe Rogadini, to which Al- eiodes and their lepidopteran hosts. Basal eiodes belongs, are koinobiont endoparasi- host families are more likely to be at- toids of more or less exposed lepidopteran tacked by basal Aleiodes species while the larvae (Shaw 1983, 1994; Shaw and Hud- most derived host family, Noctuidae, is dleston 1991; Shaw 1995, Fortier and Shaw more likely to be attacked by derived Al- 1999). A characteristic of Rogadini is that eiodes species. pupation takes place inside the dead So far as known previous to this study, host's larval skin, which hardens and only two Aleiodes species, a Palaearctic darkens to become a 'mummy' (Shaw and species and the Nearctic species A. stig- Huddleston 1991). rnator (Say), are gregarious (Shaw and Aleiodes host mummies can usually be Huddleston 1991, Shaw 1997). distinguished from those of its putative sister group Rogas (Whitfield, 1992) in that METHODS 1) a slit is cut in the ventral area of the This species can be identified as a mem- host's thoracic region, through which a ber of the subfamily Rogadinae by using sticky substance is emitted, which often the keys of Shaw (1995), Shaw and Hud- functions to glue the mummy to a sub- dleston (1991), or Wharton et ah (1997). strate, and 2) the emergence hole cut by My definition of Aleiodes follows that of the emerging adult parasitoid is normally van Achterberg (1991), Fortier and Shaw less jagged and more circular as compared (1999), Shaw (1993), and Shaw et ah (1997). with Rogas (Shaw 1995, 1997; see Fig. 8 Specimens can be determined as Aleiodes this paper). using the key of Shaw (1997). Fortier and Shaw have argued that bas- Terminology follows that used by Shaw al Aleiodes species tend to be less host spe- et ah (1997). Microsculpture terminology cific while derived species tend to feed ex- follows that of Harris (1979). Wing vein Volume 9, Number 2, 2000 289 Figs. 1-4. Aleiodes leptocarina. 1, Head. 2, Mesopleuron. 3, Flagellomeres near middle of antenna. 4, Wings. terminology agrees with that adopted by Wharton et al. (1997) and by Goulet and Huber (1993). Aleiodes leptocarina Fortier, new species (Fig. 1-7) Female. — Body color: honey yellow, legs and mandibles honey yellow except man- dibular teeth and ocellar triangle black, clypeus occasionally with black; anten- nomeres dark honey yellow, wings hya- line, veins, stigma dark honey yellow. Body length: 6-7 mm; fore wing length, 5.5-6.0 mm. Head: 44-46 flagellomeres, first flagellomere length about 1.6 times width, 20th flagellomere length about 1.5 times width (Fig. 3); oral opening height slightly greater than or equal to width, clypeus protruding with distinct apical edge (Fig. 1); occipital carina interrupted at apex and not meeting hypostomal ca- rina; ocelli moderately large, slightly greater than ocell-ocular distance; face finely colliculate; longitudinal ridge be- tween antennae occasionally extending down face up to 0.35 length of antenno- clypeal space; vertex colliculate; temples colliculate. Mesosoma: pronotum collicu- late; mesonotum and scutellum collicula- te; notauli without carinae; mesopleuron finely colliculate with a dull shine, subalar groove without carinae, sternaulus absent (Fig. 2); propodeum rugose or rugulose posteriad, faintly rugulose over colliculate surface anteriad, or without rugae (Fig. 5). Legs: tarsal claws of all legs completely pectinate (Fig. 6); hind coxae evenly colli- culate dorsally and laterally. Wings (Figs. 4): hyaline; front wing vein r about half the length of 3RSa and about 1.3 times length of rm; vein lCU-a about 0.8 length of lCU-b; rear wing vein m-cu absent; RS slightly recurved, marginal cell narrowest at midpoint; vein r-m shorter than vein 1M; vein 1A meeting and terminating at 290 Journal of Hymenoptera Research Figs. 5-8. Aleiodes leptocarina. 5, Propodeum. 6, Metatarsal claw. 7, Second metasomal tergite. 8, Dorsal view of host caterpillar. apex of vein cu-a; vein M+CU over 1.5 times as long as vein 1M. Metasoma: me- tasomal tergite weakly rugulose, raised transverse carinae of first metasomal ter- gite at base joining medially to form me- dian carina, median carina becoming fainter apically, first metasomal tergite length slightly less than or equal to apical width, basal width slightly greater than 0.5 apical width; second metasomal tergite weakly rugulose, median carina weak, oc- casionally absent in apical (Fig. 7); third metasomal tergite mostly shiny, finely col- liculate with faint rugulation antero-me- dially, median carina usually absent ba- sally, never present apically; tergites api- cad of third metasomal tergite smooth; ovipositor sheath length about 0.35 times basitarsis length. Male. — Essentially as in female, except 41-43 flagellomeres, 20th flagellomere length over twice the width, forewing length 5.0-5.5 mm., first metasomal tergite length slightly greater than apical width, third metasomal tergite weakly rugulose in basal half, weakly rugulose or shiny- colliculate in apical half, median carina oc- casionally weakly present in basal half. Holotype.— Female: COSTA RICA, Car- tago, P. N. Tapanti, 1150 m., LS194000, 559800, IX-5-1995, G. Mora, collector. De- posited in INBio. Paratypes.— COSTA RICA: 76 females, 15 males, Cartago, P. N. Tapanti, 1150 m., LSI 94000, 559800, IX-5-1995, G. Mora, col- lector. Paratypes deposited in INBio, Wheeling Jesuit University Insect Collec- tion, Texas A&M Department of Entomol- ogy Insect Collection, and Rocky Moun- tain Systematic Entomology Laboratory. Distribution. — Known only from Costa Rica. Biology. — Holotype and paratypes all reared from a single lepidopteran host Volume 9, Number 2, 2000 291 (Donald Davis, personal communication) (Fig. 8) about 7 cm. long. Comments. — This species and A. stigma- tor (Say) are the only described gregarious New World species. The tarsal claw is closely similar to that of A. stigmator in ar- rangement of the pectin and in overall shape. This species differs from A. stig- mator in having more than 34 antennom- eres, clypeus protruding with distinct api- cal edge, shiny mesopleuron, finely retic- ulate rugulation on first and second me- tasomal tergites, and females with little or no rugulation on third metasomal tergite. Morphological features including short pronotum place it in the gastritor species- group. A. stigmator is also in this species- group (Fortier and Shaw 1999, Shaw et al. 1997). The clypeus protruding and edged apically is a derived character state found elsewhere in the gastritor species-group. Etymology. — From the Greek leptos meaning "fine, delicate," and carina mean- ing "ridge," in reference to the finely re- ticulate rugulation on the first two meta- somal tergites. ACKNOWLEDGMENTS I thank Ugalde Gomez Jesus, curator of the Hy- menoptera collection at INBio, Instituto Nacional de Biodiversidad, Santo Domingo de Heredia, Costa Rica, who provided specimens for this study. Thanks are due Scott R. Shaw, curator of the Rocky Mountain Systematic Entomology Laboratory, University of Wyoming, Laramie, Wyoming, who made those spec- imens accessible to me. I also thank Ms. Diane Barry at NIOSH (National Institute for Occupational Safety and Health) in Morgantown, West Virginia, for her assistance with the Scanning Electron Microscope and photography- I would also like to thank Mark Shaw and Don Quicke for reviewing the manuscript and offering constructive comments, and Don Davis for identifying the host. LITERATURE CITED Achterberg, C. van. 1991. Revision of the genera of the Afrotropical and W. Palaearctic Rogadinae Foerster (Hymenoptera: Braconidae). Zoologische Verhandelingen 273: 1-102. Fortier, J. and S. R. Shaw. 1999. Cladistics of the Al- eiodes lineage of the subfamily Rogadinae (Hy- menoptera: Braconidae). journal of Hymenoptera Research 8:204-237. Goulet, H. and J. T. Huber. 1993. Hymenoptera of the World: An identification guide to families. Agricul- ture Canada Publication 1894/ E, Ottawa. Harris, R. A. 1979. A Glossary of Surface Sculpturing. Occasional Papers in Entomology, California Depart- ment of Agriculture 28: 1-33. Shaw, M. R. 1983. On[e] evolution of endoparasitism: the biology of some genera of Rogadinae (Bra- conidae). Contributions of the American Entomolog- ical Institute 20:307-328. Shaw, M. R. 1994. Chapter 7, Parasitoid host ranges. Pp. 112-144, In: Hawkins, B. A. and W. Sheehan [eds.], Parasitoid Community Ecology, Oxford Uni- versity Press, Oxford. Shaw, M. R. and T. Huddleston. 1991. Classification and biology of braconid wasps. Handbooks for the Identification of British Insects 7: 1-126. Shaw, S. R. 1993. Systematic status of Eucystomastax Brues and characterization of the Neotropical species (Hymenoptera: Braconidae: Rogadinae). journal of Hymenoptera Research 2:1-11. Shaw, S. R. 1995. Chapter 12.2, Braconidae. Pp. 431- 463, In: Hanson, P. E. ajd I. D. Gauld [eds.], The Hymenoptera of Costa Rica, Oxford University' Press, Oxford. Shaw, S. R. 1997. Subfamily Rogadinae s.s. Pp. 403- 412, In: Wharton, R. A., P. M. Marsh, and M. J. Sharkey [eds.], Manual of New World genera of the family Braconidae. Special Publication of the Inter- national Society of Hymenopterists. No.l, 438 pp. Shaw, S. R., P. M. Marsh, and J. C. Fortier. 1997. Re- vision of North American Aleiodes Wesmael (Part 1): the pulchripes Wesmael species-group in the New World (Hymenoptera: Braconidae, Roga- dinae). journal of Hymenoptera Research 6(1): 10 35. Wharton, R. A., P. M. Marsh, and M. J. Sharkey, eds. 1997. Manual of New World genera of the family Braconidae. Special Publication of the International Society of Hymenopterists. No.l, 438 pp. Whitfield, J. B. 1992. The polyphyletic origin ot en- doparasitism in the cyclostome lineages oi Bra- conidae (Hymenoptera). Systematic Entomology 17: 273-286. J. HYM. RES. Vol. 9(2), 2000, pp. 292-297 Descriptions and Biological Notes on Two New Phytophagous Species of the Genus Allorhogas from Brasil (Hymenoptera: Braconidae: Doryctinae) Paul M. Marsh, Margarete Valverde de Macedo and Marina C. P. Pimental (PMM) P. O. Box 384, North Newton, Kansas 67117, USA (cooperating scientist, Systematic Entomology Laboratory, U. S. Department of Agriculture, Washington, DC); (MVM, MCPP) Laboratorio de Ecologia de Insetos, Departamento de Ecologia, CP 68020, IB, Universidade Federal do Rio de Janeiro, Ilha do Fundao, Rio de Janeiro, Brasil, CEP 21941-590 Abstract. — Two new species of the genus Allorhogas are described from the Poco das Antas Biological Reserve, Rio de Janeiro State, Brasil: A. spermaphagus Marsh, reared from seed pods of Stryphnodendron polyphyllum (Leguminosae) and A. brasiliensis Marsh, reared from seed pods of Pithecellobium pedicellare (Leguminosae). Biological notes are provided and a key is presented to the species reared from plant seeds or galls in Brasil. For centuries, species in the family Bra- For many years, there have been records conidae were thought to be always para- of species in the genus Allorhogas being sitoids of other insects. Macedo and Mon- reared from plant galls but no firm biolog- teiro (1989) presented the first document- ical data was available. For example, A. ed case of phytophagy in the family Bra- galicola Gahan was reared from oak galls conidae for a species of the braconid (Gahan 1912) and A. heringeri (Guimaraes) genus Allorhogas (subsequently described and A. muesebecki (Guimaraes) were as A. dyspistus by Marsh 1991) which at- reared from plant galls (Guimaraes 1957). tacks seeds of the legume Pithecellobium No species have ever been definitely tortum Martius in Brasil. Since then, more reared as parasitoids of other insects in records of phytophagy in the Braconidae galls or seed pods As noted above and have been noted. Infante, et al. (1995) from the bioiogicai information presented showed that species of the genus Momto- here/ it ig now firmly established that spe. riella formed galls on Philodendron in Cen- ^ ^ the g attack geed dg and are tral and South America. Ramirez and ., c .u i ^™^v i-ii r i not parasitic on any other insect, rurther- Marsh (1996) described two species of the , \ , ., , c „ ' , , „ r„. , . , more, several undescnbed species from genus Psenobolus from Costa Rica which _ _. , , , / , ... ,■ ti ..i Costa Rica have been reared from various develop as inquilines in figs. Recently, , . „ , , . ., , , . , . . Austin and Dangerfield (1998) showed leaf §al s' althou§h ™ detailed biological that a species of the genus Mesostoa forms studies have yet been made- , galls on Banksia in Australia and Macedo The following new species of the genus et al. (1998) provided further information Allorhogas are described in order to pro- on the biology of A. dyspistus in Brasil. vide names for further biological studies One of us (PMM) has seen a new species being done by two of us (MVM, MCPP). of Allorhogas from Costa Rica that has These species were reared from seeds of been reared from another species of Pithe- Stryphnodendron polyphyllum Martius and cellobium. Thus, it is now well established Pithecellobium pedicellare (DC.) Benth. in that phytophagy does occur in the family Brasil and preliminary observations are Braconidae. given under each species description be- Volume 9, Number 2, 2000 293 low. The genus is in need of study for the Neotropical Region. There are an estimat- ed 50 species from Brasil, nearly all un- described and mostly without biological information. A study is in progress on the genus from Costa Rica with an estimate of 25 undescribed species. The names of the two new species are to be attributed only to the senior author; all biological observations were made by the junior authors. The genus Allorhogas can be identified by keys presented in Marsh (1997). Morphological terminology is based on Wharton, et al. (1997). KEY TO SPECIES OF ALLORHOGAS REARED FROM PLANT SEEDS OR GALLS IN BRASIL (Based on females only) 1. Ovipositor barely visible, shorter than first metasomal tergum 2 - Ovipositor at least half as long as metasoma 3 2(1). Malar space 2/5 eye height dyspistus Marsh - Malar space nearly equal to eye height muesebecki (Guimaraes) 3(1). Fore wing vein m-cu arising distad of vein 2RS heringeri (Guimaraes) - Fore wing vein m-cu interstitial with or arising basad of vein 2RS 4 4(3). Fore wing vein m-cu interstitial with 2RS (Fig. 2); body length 3.50-3.75 mm; 29-30 antennomeres, flagellum light brown spermaphagns Marsh, new species Fore wing vein m-cu arising basad of 2RS, thus a short section of (RS+M)b present (Fig. 1); body length 4.0-4.5 mm; 34-35 antennomeres, flagellum honey yellow on basal 1/2, brown on apical 1/2 brasiliensis Marsh, new species Allorhogas spermaphagns Marsh, new species (Figs. 2-6) Female. — Body color: head, mesosoma and metasoma honey yellow; legs yellow; flagellum and pedicel light brown; wings hyaline, fore wing vein C+Sc + R and stig- ma yellow, rest of veins light brown, hind wing veins yellow. Body length: 3.50-3.7 mm. Head: 29-30 antennomeres; face ru- gulose-coriaceous medially, strongly ru- gose along inner eye margins; frons exca- vated, coriaceous; vertex (Fig. 3) coria- ceous, weakly rugulose behind ocelli; tem- ple coriaceous; malar space 1/3 eye height; oral opening small, circular, di- ameter equal to basal width of mandible; ocell-ocular distance twice diameter of lat- eral ocellus; occipital carina distinct, meet- ing hypostomal carina. Mesosoma (Figs. 4-5): pronotum weakly rugulose-coria- ceous laterally with deep median scrobic- ulate groove bordered ventrally by carina; mesonotum sharply declivous anteriorly, mesonotum and scutellum coriaceous, middle lobe with median scrobiculate groove extending to anterior edge of me- sonotum; notauli distinctly scrobiculate, meeting in large rugose area before scu- tellum; scutellar furrow with three cross carinae; mesopleuron coriaceous, sternau- lus deep and smooth or weakly scrobicu- late; propodeum rugose, with two baso- lateral semicircular rugulose-coriaceous areas enclosed by carina, apical area above metasoma insertion usually smooth, en- closed by carina. Legs: fore tibia with scat- tered row of 15-20 stout spines on anterior edge; hind coxa with distinct baso-lateral tubercle. Wings (Fig. 2): fore wing vein r nearly as long as 3RSa, vein m-cu intersti- tial with 2RS, first subdiscal cell open, vein 2cu-a absent; hind wing vein M + CU about equal in length to 1M, vein m-cu curved slightly toward wing apex. Meta- soma (Fig. 6): first tergum longitudinally costate-rugulose, wider at apex than long, medially with two strong longitudinal ca- rinae setting off raised median area and 294 Journal of Hymenoptera Research Figs. 1-2. Wings of Allorhogas species. 1, brasiliensis n. sp. 2, spermaphagus n. sp. connected at base by distinct cross carina; tergum 2 + 3 longitudinally costate on bas- al 2/3, remainder coriaceous, groove be- tween terga 2 and 3 weak or absent; ter- gum 4 costate on basal 1/5, remainder co- riaceous; remainder of terga coriaceous; ovipositor about 2/3 length of metasoma. Male. — Essentially as in female; femora swollen, hind femur with length about twice width. Holotype Female. — BRASIL: Poco das Antas Biological Reserve, Silva Jardim County, Rio de Janeiro State, September 1995, M. V. Macedo, reared from seeds of Stryphnodendron sp. Deposited in Depar- tamento de Ecologia-IB-CCS, Universidad Federal do Rio de Janeiro, Brasil. Paratypes. — BRASIL: 7 females, 4 males, same data as holotype. Deposited in: Universidade Federal do Rio de Janiero, Brasil; Universidade Federal de Sao Car- los, Brasil; National Museum of Natural History, Washington, DC. Comments. — This species is very similar to brasiliensis, but the most consistent dis- tinguishing characters are: smaller size (3.5-3.75 mm) than in brasiliensis (4.0-4.5); antenna with 29-30 antennomeres (34-35 in brasiliensis); entirely brown flagellum (yellow at base, brown at apex in brasilien- Volume 9, Number 2, 2000 295 Figs. 3-6. Allorhogas spermaphagus n. sp. 3, vertex. 4, mesosoma, lateral view. 5, mesosoma, dorsal view. 6, metasoma, dorsal view. sis); fore wing vein m-cu interstitial with vein 2RS (basad of 2RS in brasiliensis); fore wing vein r about equal to vein 3RSa (3/ 4 length of 3RSa in brasiliensis). In some specimens of spermaphagus the face is often more distinctly rugose near the eyes than in spermaphagus; also the vertex of sper- maphagus is often rugulose behind the ocelli but usually only coriaceous in bras- iliensis. Biology. — This species was reared from seeds of Stryphnodendron polyphyllum Mar- tius (Leguminosae) which is native to the Atlantic forest region of Brasil. The biolo- gy is similar to that described for A. dys- pistus by Macedo and Monteiro (1989) and Macedo et al. (1998). Pods of Stryphnoden- dron contain about 10 seeds arranged side by side. The Allorhogas female oviposits directly into immature seeds when abun- dant endosperm and a small embryo are still present. Oviposition is directly through the pod wall and the egg is placed inside the seed. After oviposition by the braconid, the seed divides internal- ly and externally (see Macedo et al. (1998)), resulting in an intact region joined to the funicle where the seed embryo is 296 Journal of Hymenoptera Research usually found. In many cases this region groove bordered ventrally by carina; me- of the seed continues growing even after sonotum sharply declivous anteriorly, me- the adult Allorhogas has emerged. In most sonotum and scutellum coriaceous, mid- of the attacked seeds that were observed die lobe with median scrobiculate groove (n = 34) more than one Allorhogas was not extending to anterior edge of meson- found in a single seed — 29% were ob- otum; notauli distinctly scrobiculate, served to have two braconids per seed meeting in large rugose area before scu- and 62% had three per seed. In these cas- tellum; scutellar furrow with three cross es, more than one division of the seed oc- carinae; mesopleuron coriaceous, sternau- curs but a single intact region is still found lus deep, scrobiculate; propodeum rugose, with the seed embryo. with two baso-lateral semicircular rugu- Two species of chalcid wasps were lose-coriaceous areas enclosed by carina, reared from the same pods: Lycrus sp. apical area above metasoma insertion (Pteromalidae) and Eurytoma sp. (Eury- strongly rugose. Legs: fore tibia with scat- tomidae). Although we did not observe tered row of 15-20 stout spines on anterior these wasps emerging directly from Allor- edge; hind coxa with distinct baso-lateral hogas larvae, one was observed feeding on tubercle. Wings (Fig. 1): fore wing vein r an Allorhogas larva. Because the chalcid about 3/4 length of 3RSa, vein m-cu meet- pupae were dissected from seeds exhibit- ing RS+Ma slightly before 2RS, thus a ing the same damage caused by the brae- short segment of (RS+M)b visible, first onids, we are assuming the chalcids were subdiscal cell open, vein 2cu-a absent; attacking the Allorhogas. hind wing vein M+CU about 3/4 length Etymology. — The specific name is from of 1M, vein m-cu curved slightly toward the Greek sperma meaning seed and the wing apex. Metasoma: first tergum Ion- Greek phagein meaning to eat in reference gitudinally costate, wider at apex than to the biology of this seed-eating braconid. long, medially with two strong longitudi- nal carinae setting off raised median area Allorhogas brasiliensis Marsh, new and connected at base by distinct cross ca_ species rina; tergum 2+3 longitudinally costate on ^ 1§' ' basal 2/3, remainder coriaceous, groove Female. — Body color: head, mesosoma between terga 2 and 3 weak or absent; ter- and metasoma dark honey yellow, pro- gum 4 costate on basal 1/5, remainder co- podeum and mesonotum often light riaceous; remainder of terga coriaceous; brown; legs yellow; flagellum honey yel- ovipositor slightly more than 1/2 length low on basal 1/2, turning to brown on of metasoma. apical 1/2; wings hyaline, fore wing vein Male. — Essentially as in female; femora C+Sc + R yellow, stigma brown, rest of swollen, hind femur with length about veins light brown, hind wing veins yel- twice width; propodeum and first meta- low. Body length: 4.0-4.5 mm. Head: 34- somal tergum dark brown. 35 antennomeres; face rugulose-coria- Holotype Female. — BRASIL: Pocp das ceous; frons excavated, coriaceous; vertex Antas Biological Reserve, Silva Jardim and temple coriaceous; malar space 1/3 County, Rio de Janeiro State, October eye height; oral opening small, circular, 1995, M. C. Pimentel, reared from seeds of diameter equal to basal width of mandi- Pithecellobium sp. Deposited in Departa- ble; ocell-ocular distance twice diameter of mento de Ecologia-IB-CCS, Universidad lateral ocellus; occipital carina distinct, Federal do Rio de Janeiro, Brasil. meeting hypostomal carina. Mesosoma: Paratypes. — BRASIL: 6 females, 6 males, pronotum weakly rugulose-coriaceous lat- same data as holotype. Deposited in Univ- erally with deep median scrobiculate ersidade Federal do Rio de Janiero, Brasil, Volume 9, Number 2, 2000 297 Universidade Federal de Sao Carlos, Bras- il, and National Museum of Natural His- tory, Washington, DC. Comments. — See comments under sper- maphagus for distinguishing characters of the two species. Biology. — This species was reared from seeds of Pithecellobium pedicellare (DC.) Benth. Preliminary observations show that the biology is similar to that of A. sper- maphagus. The seed damage caused by both A. spermaphagus and A. brasiliensis is very similar to that observed for A. dys- pistus. Etymology. — The specific name is in ref- erence to the locality of this species. ACKNOWLEDGMENTS We thank Eric Grissell, Systematic Entomology Laboratory, USDA, Washington, DC, for identifica- tion of the chalcids and the staff of the Poco das An- tas Biological Reserve for providing field facilities. LITERATURE CITED Austin, A. D. and P. C. Dangerfield. 1998. Biology of Mesostoa kerri Austin and Wharton (Insecta: Hy- menoptera: Braconidae: Mesostoinae), an endem- ic Australian wasp that causes stem galls on Banksia marginata Cav. Australian journal of Botany 46:559-569." Gahan, A. B. 1912. Descriptions of two new genera and six new species of parasitic Hymenoptera. Proceedings of the Entomological Society of Washing ton 14:2-6. Guimaraes, J. A. 1957. Cintribuicao ao estudo de Ce- cidilogia Brasiliana. Thesis, Escuela National Agronotma, Universidad Federal Rura do, Rio de Ja- neiro, p. 25. Infante, F., P. Hanson and R. Wharton. 1995. Phyto- phagy in the genus Monitoriella (Hymenoptera: Braconidae) with description of a new species. Annals of the Entomological Society of America 88: 406-415. Macedo, M. V. de and R. T. Monteiro. 1989. Seed pre- option by a braconid wasp, Allorhogas sp. (Hy- menoptera). journal of the New York Entomological Society 97:359-362. Macedo, M. V. de, M C. P. Pimentel and R. C. Vieira. 1998. Response of Pithecellobium tortum Martius (Leguminosae) seeds to the attack of the phy- tophagous braconid Allorhogas dyspistus Marsh (Hymenoptera: Braconidae). journal of Hymenop- tera Research 7(2):274-279. Marsh, P. M. 1991. Description of a phytophagous doryctine braconid from Brasil (Hymenoptera: Braconidae). Proceedings of the Entomological So- ciety of Washington 93:92-95. Marsh, P. M. 1997. Subfamily Doryctinae, pp. 207- 233. In: Wharton, R. A., P. M. Marsh and M. J. Sharkey, eds., Manual of the New World genera of the family Braconidae (Hymenoptera). Special Publication of the International Society of Hymen opterists. No. 1, 439 pp. Ramirez, W. B. and P. M. Marsh. 1996. A review of the genus Psenobolus (Hymenoptera: Braconidae) from Costa Rica, an inquiline fig wasp with bra- chypterous males, with description of two new species, journal of Hymenoptera Research 5:64-72. Wharton, R. W., P. M. Marsh and M. J. Sharkey, eds. 1997. Manual of the New World genera of the family Braconidae (Hymenoptera). Special Publi- cation of the International Society of Hymenopterists. No. 1, 439 pp. J. HYM. RES. Vol. 9(2), 2000, pp. 298-304 Description of a New Species of Emersonella (Hymenoptera: Eulophidae) from Brazil, with Preliminary Observations on its Biology Celso O. Azevedo, Juvenal C. Silva-Jr, and Lucio A. de O. Campos (COA) Universidade Federal do Espfrito Santo, Departamento de Biologia, Av. Marechal Campos 1468, Maruipe, 29040-090 Vitoria, ES, Brazil; (JCSJ) Universidade Estadual do Sudoeste da Bahia, Departamento de Ciencias Biologicas, Rua Jose Moreira Sobrinho s/n, 45200-000 Jequie, BA, Brazil; (LAOC) Universidade Federal de Vicosa, Departamento de Biologia Geral, 36571-000 Vicosa, MG, Brazil Abstract. — Emersonella trimacnlata Azevedo and Silva, new species, an egg parasitoid of chrysomelids, from northeastern Brazil is described and illustrated. Host age influences percent parasitism, which decreases as age increases, but it does not influence the sex ratio. The egg phase has an average duration of 1.05 ± 0.04 days, larval phase of 5.04 ± 0.19 days, pupal phase of 6.74 ± 1.16 days. The total time of development is 13.29 ± 0.53 days for males and 13.96 ± 0.48 days for females. Emersonella Girault is a small genus of For the biological studies, mated and Entedoninae (Hymenoptera, Eulophidae) nulliparous females were maintained sep- restricted to the New World (Boucek arately in flasks of 50ml containing a drop 1977). All species with known biology are of a 1:1 solution of honey and water stuck idiobiont endoparasitoids of eggs of Chry- to the wall of the glass as food, covered somelidae (Cox 1994), mainly Cassidinae. with cotton and kept at 28 ± 1°C The par- De Santis (1983) revised the genus and de- asitoid and its host Zatrephina meticulosa scribed four species from Brazil, Argenti- (Spaeth) (Coleoptera, Chrysomelidae) was na and Uruguay. In this paper, a new spe- obtained from field collections from Sao cies of Emersonella from the State of Mar- LuiZ/ state of Maranhao, northeastern Bra- anhao in northeastern Brazil, is described zil This beetle occurs natUrally on the and illustrated. Some preliminary biolog- leaves of Ipomoea pes-caprae L. (Convolvu- ical studies are also included. Influence of laceae) in coast sand plain the host age on the capacity of parasitism TwQ experiments were carried out. Ex_ and sex ratio of the parasitoid is verified, iment x was to verif the influence of the egg morphology is characterized, and ,u u . .u . •.■ , , , bt> r * i i i me nost aSe on me percent parasitism and the longevity of the egg, larval and pupal .. c . , ., T ,u. i ° j . , r r sex ratio of the parasitoid. In this experi- phases are determined. , nn r„ ... , ,.c r ment, 80 egg masses of Z. meticulosa of dir- MATERIAL AND METHODS ferent ages, varying from 1 to 8 days, were Morphological terminology for the de- offered to females, separately in a flask. scription generally follows Gibson (1997) The females were kept with the host egg and sculpture follows Harris (1979). The masses for 10 days before being removed material examined was provided by the from the flasks. Experiment 2 was to ver- Entomological Collection of Universidade ify the duration of the egg, larval and pu- Federal do Espirito Santo (UFES) and pal phases, and morphological character- Universidade Federal de Vicosa (UFVB). istics of the egg. Host egg masses at 48 Volume 9, Number 2, 2000 299 Figures 1-6. Emersonella trimaculata, female, dorsal view. 1, head; 2, antenna; 3, thorax; 4, propodeal disc, 5, metasoma; 6, ovipositor sheathes. hours old were offered to groups of 30 to 40 nulliparous, recently mated females for 3 hours. To verify the phase of parasitoid development, samples of 15 host eggs were dissected in physiologic solution, at intervals of 24 hours. To verify the para- sitoid egg phase, 10 host eggs were ob- served at one hour intervals. Observations were made from the 15"1 until the 29"1 hour to determine the type, morphology and size of the egg. Emersonella trimaculata Azevedo and Silva, new species (Figs. 1-12) Female.— Length 0.85-1. 01mm. Head and body black, except: head, scutellum and propodeum with yellowish green me- 300 Journal of Hymenoptera Research ■V / 1 ?" AM 3 - ■■ ^^1 aA -^H ggv\\v1B ^V| ■M ■ri^^^^IZ?/ ^^ Figures 7-12. Emersonella trimaculata, male, 7, head, dorsal view; 8, mandibles, lateral view; 9, antenna, lateral view; 10, antenna sensillae, lateral view; mesosoma, dorsal view; 12, metasoma, dorsal view. tallic reflections; pronotum and scutum with blue metallic reflections; mandible testaceous-brown; scape yellowish white, pedicel brown, flagellomeres black; legs slightly yellowish, distal tarsomeres and anterior face of fore femur darker, coxae black with weak blue metallic reflection; bristles on body pale yellow; wings hya- line and veins light brown. Head (Fig. 1): 1.22-1.42 X as wide as long, 1.06-1.18 X wider than thorax in dorsal view. Frons and gena smooth, face nearly so, vertex imbricate. Distance between lateral ocellus 2.25-3.0 X as long as distance from lateral ocelli to eye. Transverse fronto-facial su- ture complete and conspicuous. Scrobal depression as long as scape. Malar space about 0.43-0.53 X the height of eye. To- Volume 9, Number 2, 2000 301 T3 -J .Js 'ffi es ss pa, o a r. 'J r. P 01/ Oil 1.2 l.o • 0,8 0.6 0.4 0.2 0.0 -0.2 1 2 3 4 5 6 7 X Eggs masses age (Days) Figure 13. Percentage of egg mass posture not parasited by Emersonella trimaculata, in different ages of the host eggs. The bars represent the standard deviation. rulus slightly closer to eye margin than clypeus. Vertex with ocular-ocellar suture. Eye with short hairs. Mandible bidentate, the lower larger, and the upper with up- per margin serrated (Fig. 8). Antenna (Fig. 2): scape 3.75^4.3 X longer than wide; ped- icel almost twice as long as wide; three anelli, the first slightly larger, the others subequal, funicular segments subquadra- te, slightly longer than wide, club uniseg- mented, apex extended into terminal spine, 2-3 X longer than wide and 1.33- 1.5 X longer than funicular segments. Sen- sillae capitate and elongate with apex slightly directed upward (Fig. 10). Meso- soma (Figs. 3-4): subquadrate in dorsal view, 1.16-1.25 X longer than wide, arched in lateral view. Pronotum not vis- ible in full dorsal view, with a row of six setae along the posterior margin. Meso- scutum foveolate, with two rows of two adnotaular setae on each side, notaulus very weak, missing medially. Scutellum imbricate, enlarged, 0.75-0.89 X longer than mesosoma, 0.55-058 X as wide as me- sosoma, about as long as wide, lateral sides distinctly convex. Dorsellum smooth. Propodeum smooth, with a pair of sub- triangular depressions touching the ante- rior margin, with weak longitudinal striae medially, separated from each other by 1.2-1.5 X their width, without median ca- rina and plica, with a sublateral carinae strongly arched, spiracle rounded, sepa- rated from the anterior margin of propo- deum by about 1.0 X their diameter. Cal- lus with two setae. Mesopleuron with a small anterior central pit. Forewing with marginal vein 2.3-2.4 X longer than sub- marginal vein. Postmarginal vein about as long as stigmal vein. Metasoma (Fig. 5): stout, subsessile, with few setae, 1.03 X as long as mesosoma; first gastral tergite large, about 0.5 X length of gaster, lateral margin evenly convex in dorsal view. Ovi- positor sheath short (Fig. 6), anterior half concealed, ovipositor stylus 0.87 X as long as gaster. Male. — Length 0.9-1.06 mm. Same color as female, except by: fore femur, fore tar- sus and the other distal tarsomeres darker; gaster with three yellowish white spots, a pair of spots at anterior corner of first gas- tral segment, straight anteriorly and 302 Journal of Hymenoptera Research rounded behind, separated from each oth- front of the large one, and the femora and er by 2.2-2.5 X their diameter, the third tibiae are yellowish white rather than spot very large, occupying nearly the en- black as in the two former species. This tire width of the posterior half of the first species is also similar to £. rotunda (Ash- gastral tergite. Head (Fig. 7): Distance be- mead), but in the last species the mid and tween lateral ocelli about 4-6 X as long as hindcoxae are white and funicular seg- distance from lateral ocellus to eye. Anten- ments are slightly longer. Emersonella tri- na (Fig. 9): funicular segments slender, in maculata displays the same pattern of sex- ratio of about 2:2:2.6:2.3 X as long as wide; ual dimorphism as other species in the ge- club 3-4 X as long as wide. Mesosoma nus. The male has funicular segments lon- (Fig. 11): 1.29-1.66 X longer than wide, ger than those of the female and the gaster scutellum 0.6-0.66 X wider than mesoso- has three light spots on the first gastral ma. Forewing with marginal vein 2.1-2.4 tergite, while in the female the gaster is X longer than submarginal vein. Metaso- evenly black. ma (Fig. 12): petiolate, petiole larger be- Biology. — A total of 80 egg masses were hind, first gastral tergite 1.4-2 X longer analyzed, with 57 parasitized (71.3%) and than the rest of gaster, with anterior mar- 23 (28.7%) not parasitized. The results in- gin straight medially and angulate at cor- dicate that host age influences parasitism ner laterally. Genitalia: paramere devel- by E. trimaculata. Parasitism decreased oped inward ventrally, with an apical se- when older egg masses were offered to the tae directed outward; digitus wide, with females (logistic regression x2 = 12.7; g.l. two conspicuous spines directed outward = 1 and p <0.01; Fig. 13). The variation apically, and with a small outer tooth; ae- observed in parasitism was from only 11% deagus with two lobes rounded apically; not parasitized in the one-day-old egg phallobase little developed in ventral side; masses up to 100% not parasitized in the aedeagus apodeme extending beyond the eight-day-old egg masses, basal margin of phallobase only slightly. According to the logistic regression, the Material examined. — 9 holotype, 11 9 9 expected value for eight-day-old egg mas- and 15 6 6 para types BRAZIL, Maranhao, ses was approximately 65% not parasit- Sao Luiz, coast sand plain vegetation, ized, although the value was 100%. Only 26. i. 1998, J. C. Silva Jr. col. (UFES); 298 9 9 three egg masses were observed in the and 298 6 6 BRAZIL, reared in laboratory samples on the 8th day, while for the other (AMNH, BMNH, CASC, CNCI, CUIC, days the number of egg masses was never DCBU, DZUP, EMUS, FSCA, IGBE, INPA, smaller than nine. This difference might LACM, MCZH, MEPG, MZSP, OSUC, explain the deviation in relation to the PMAE, UCDC, UCRC, UFES, UFVB, model. USNM). This same pattern was observed when Remarks. — This species runs to Emerso- the number of individual eggs parasitized nella niveipes Girault in the key presented was verified in each egg mass. Parasitism by De Santis (1983), but here the mandible varied on the average from 81%, for one- is bidentate and the male has three white day-old egg masses, up to 12%, for eight- spots in the gaster, while E. niveipes has day-old egg masses. Thus, as age of host the mandible with six teeth and the male increases, a reduction in parasitism occurs has two transverse stripes just beyond the in both the number of egg masses and the middle of the gaster. E. ooecia De Santis number of eggs parasitized with each and E. lecitophaga De Santis are two spe- mass. cies with a sub-basal white spot in the gas- Host age does not influence sex ratio, ter of males as in E. trimaculata, but here The difference in sex ratio produced by fe- there are two additional small spots in males of £. trimaculata in egg masses of Volume 9, Number 2, 2000 303 § 0,96 0.92 0.88 0.84 o 0.80 C 0.76 0.72 • 0.68 ■ 0.64 I I I I 12 3 4 5 6 7 Eggs age (Days) Figure 14. Relationship between the age of the egg and the sex ratio of Emersonella trimaculata. The bars represent the standard deviation. different ages is not significant (x2 = 0.2; g.l. = 1 and p = 0.7; Fig. 14). £. trimaculata eggs are 25.6 ±1.4 \xm in length with a maximum width of 6.8 ±1.0 (xm (Fig. 15). The eggs are simple, hymen- opteriform, oblong or ovoid, slightly arched and with both poles smoothly round, with chorion delicate and without ornamentation as usually found in Hy- menoptera (Clausen 1940). The egg phase had an average duration of 1.04 ± 0.04 days. The micropyle of the eggs was not observed, probably due to the transpar- ency of the eggs (Fig. 15). However, there is a differentiated area in the anterior area, which might indicate the presence of the micropyle, which is usually located in the anterior region of the egg. In some species, however, it has been observed in the pos- terior area (Quicke 1997). The larval phase has a duration of 5.0 ± 0.2 days, while the pupal phase lasts 6.7 ±1.2 days. The pigmentation process be- gins in the first day of the pupal phase. Males of this species emerge before the fe- males (x: = 99.13; p <0.01), with almost all males emerging by the end of the sixth day of the pupal stage (91.56%), but only a s a X Figure 15. Eggs of Emersonella trimaculata; a, 2 hours of development; b, 20 hours of development. (Scale Kir = 5 |j.m). 304 Journal of Hymenoptera Research 65.47% of the females. The total time of development was 13.3 ± 0.5 for males and 14.0 ± 0.5 for females. In E. trimaculata, a small variation is seen in the duration of the egg and larval phases and a larger variation in the pupal phase. This can indicate the existence of mechanisms that synchronize the phases in this species. The results seem to indicate an abbreviation of the pupal phase of the males without loss of absorption of nutri- ents, since size differences do not exist be- tween males and females. ACKNOWLEDGMENT We thank J. LaSalle (IEE CABI Bioscience) who compared this new species to E. niveipes and carefully revised the manuscript. LITERATURE CITED Boucek, Z. 1977. Descriptions of two new species of Neotropical Eulophidae (Hymenoptera) of eco- nomic interest, with taxonomic notes on related species and genera. Bulletin of Entomological Re- search 67:1-15. Clausen, C. P. 1940. Entomophagous Insects. New York, McGraw-Hill. Cox, M. L. 1994. The Hymenoptera and Diptera par- asitoids of Chrysomelidae, p. 419-467. In: Jolivet, P.H., M. L. Cox and E. Petitpierre (eds.). Novel aspects of the biology of Chrysomelidae. Nether- lands, Kluwer Academic Publishers. De Santis, L. 1983. Las especies argentinas, uruguayas y brasilenas del genero Emersonella Girault, 1916 (Insecta, Hymenoptera, Eulophidae). Anais da So- ciedade Entomologica do Brasill2(2):249-259 . Gibson, G. A. P. 1997. Morphology and Terminology, p. 16^4. In: Gibson, G. A. P., J. T. Huber & J. B. Wooley. Annotated keys to the genera of Nearctic Chalcidoidea (Hymenoptera). Ottawa, NRC Re- search Press. Harris, R. A. 1979. A glossary of surface sculpturing. Occasional Papers in Entomologi/, California Depart- ment of Agriculture 28:1-31. Quicke, D. L. J. 1997. Parasitic zvasps. London, Chap- man & Hall. J. HYM. RES. Vol. 9(2), 2000, pp. 305-312 Trisecodes gen. n., (Hymenoptera: Eulophidae: Entedoninae), the First Eulophid with Three Tarsal Segments Gerard Delvare and John LaSalle (GD) CIRAD, TA 40/02, Avenue d'Agropolis, 34398 Montpellier Cedex 5, France, e-mail: gerard.delvare@cirad.fr ; (JL) Unit of Parasitoids Systematics, CABI Biosciences UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berks, SL5 7PY, UK, e-mail: j.lasalle@ic.ac.uk Abstract. — The neotropical genus Trisecodes, and its type species T. agromyzae, are described in the eulophid subfamily Entedoninae. Its placement within the Chalcidoidea and the Eulophidae is discussed. Its hosts, all belonging to Agromyzidae (Diptera) are listed. This is the first Eulo- phidae, as well as the first member of the Chalcidoidea outside of the Trichogrammatidae, where both sexes have three tarsal segments. Resume. Les auteurs decrivent le nouveau genre neotropical Trisecodes, ainsi que son espece type, T. agromyzae, dans la sous-famille des Entedoninae (Eulophidae). lis discutent ensuite de sa position systematique a l'interieur des Chalcidoidea puis des Eulophidae. lis citent les hotes de cette espece, qui appartiennent tous a la famille des Agromyzidae (Diptera). II s'agit du premier Eulophidae, mais aussi du premier chalcidien n'appartenant pas aux Trichogrammatidae, dont les deux sexes presentent des tarses trimeres. Understanding the evolution of the su- perfamily Chalcidoidea, and deriving ro- bust phylogenetic hypotheses, is a major challenge. The difficulty comes from the biodiversity of the group, one of the most speciose superfamilies of the Hymenop- tera (Noyes 1998), and the plasticity of both morphological and biological fea- tures exhibited by these wasps (Gibson 1990; Gibson et al. 1999). The genus de- scribed below is an example of such plas- ticity. Up to now, the only Hymenoptera known to have 3-segmented tarsi were members of the family Trichogrammati- dae, and some highly derived and dimor- phic male fig wasps (Agaonidae). The presence of three tarsal segments in both sexes has been considered a synapomor- phy for the Trichogrammatidae, many of whose members also have lines of setae on the forewing. The wasp described in this paper, Trisecodes agromyzae, shares these features with the Trichogrammatidae but does not belong to this family. Specimens of the newly described spe- cies have been deposited in the following institutions: Natural History Museum, London, UK (BMNH), Centre de Cooper- ation Internationale en Recherche Agron- omique pour le Developpement, Montpel- lier, France (CIRAD), United States Na- tional Museum, Washington D. C, USA (USNM), Canadian National Collection of Insects and other Arthropods, Ottawa, Canada (CNC). Trisecodes Delvare and LaSalle, new genus (Figs. 1-12) Name derivation. — A combination of tri- suggesting the tarsal formula and -secodes from Asecodes, a genus of Entedoninae. Some members of Asecodes have lines of setae on the forewing which are similar to those seen in Trisecodes. Gender neuter. Type species. — Trisecodes agromyzae Del- vare and LaSalle, new species. Head (Fig. 1): Frons with distinct scrobal 306 Journal of Hymenoptera Research Figs. 1-6. Trisecodes agromyzae, new genus, new species. 1, head in frontal view; 2, head and basal pari of mesosoma in dorsal view; 3, mesonotum; 4, apical part of mesosoma; 5, propodeum in posterior view; 6, mesosoma in lateral view. Volume 9, Number 2, 2000 307 Figs. 7-12. Trisecodes agromyzae. 7, fore tibia and tarsus; 8, female metasoma in dorsal view; 9, forewing; 10, detail of the base of the forewing; 11, female antenna; 12, male flagellum. sutures which extend dorsally to the fron- tal sutures and define a median strip, the strip slightly raised over the surface of the frons and slightly overlapping it laterally- Frontal sutures V-shaped, clearly separat- ed from the median ocellus. Malar sulcus groove like. Clypeus not delimited from lower face. Mandibles bidendate, the teeth acute, of same length. Maxillary palpi bis- egmented, labial palpi unisegmented. La- brum strongly bilobed ventrally. Antenna (Figs. 11-12): Flagellum with one anellus, a 3-segmented funicle and a 3-segmented clava. Funicular segments transverse, bearing multiporous plate sensilla of two types: usual ones, as found in other Chal- cidoidea (elongate sensilla longitudinally oriented with raised and sharp apex) and half-ring ones, embracing each of the seg- ments as well as the apex of the clava; lat- ter sensilla are in oblique or even tran- verse orientation (except on third clava] segment). Multiporous plate sensilla of the usual type are replaced in males by 308 Journal of Hymenoftera Research whorls of long setae. Mesosoma (Figs. 2- with slight bluish reflections. Tarsi whit- 6): Pronotum and mesonotum with dis- ish. Wings hyaline. Veins dark brown, tinct raised reticulation, same short and Head (Fig. 1): Relative measurements very sparse setae, the former narrowly vis- (made from a slide mounted specimen): ible dorsally, with its posterior margin Height 70, width 87, fronto-vertex in front deeply emarginate. Mesoscutum and scu- of the median ocellus 48, POL 23, OOL 9, tellum with distinct median furrow. No- lateral ocelli diameter 7, eyes height 43, tauli complete, straight, of same appear- width of eye (frontal view) 16, malar space ance throughout. Axillae distinctly pro- 20, oral fossa 21, antennal toruli-eye dis- truding anteriorly. Scutellum without sub- tance 14, distance between toruli 10, toruli lateral furrows or lines, distinctly convex diameter 7, toruli-lower margin of clypeus both dorsally and laterally. Propodeum length 10. Inner margin of eyes parallel, mostly reticulate, with spiracular groove Malar groove nearly straight as well as ge- hardly impressed, without subspiracular nae. Labrum with a row of 4 hairs (one tubercle, supracoxal flange and nucha, specimen examined). Lower edge of an- Spiracle formed through incision of the tennal toruli at lower eye margin. One anterior margin of the propodeum, not r0w of 7-8 hairs along inner margin of eye completely closed. Phragma not reaching plus one pore. Lower face very faintly the posterior margin of the mesosoma. squamose, frons reticulate. Upper frons, Prepectus large, reticulate, triangular in above frontal sutures, coriaceous ( = en- lateral view. Mesopleuron convex anteri- graved reticulate), more or less imbricate, orly, femoral scrobe reticulate, upper me- vertex with elongate cells which are trans- sepimeron very large, separated from low- versely oriented on its anterior part then er mesepimeron by a row of punctures, oblique. Antenna (Fig. 11): Scape nearly 5 the latter very small. Metapleuron small times as iong as wide (28:6). Pedicel slight- but visible, triangular, delimited from the jy longer than wide (in lateral view 9:7). propodeum by a broad, shallow groove. pirst funicular segment very transverse (7: Legs (Fig. 7): All tarsi 3-segmented, seg- n^ following segments progressively less ments about of same length. Apical spur so First claval segment slightly transverse of foretibia slightly curved, bifid, the furca (10:12), second one subquadrate, last one at mid length of the spur, the inner branch eiongate (9:5). Each of the flagellar seg- hardly visible and much shorter than the ments bearing 3 ring-like sensilla, in near- outer one. Wings (Figs 9-10). Submarginal ly transverse orientation on Fl, more vein with one dorsal seta. No break be- oblique in the following segments. Two tween submarginal vein and parastigma, dongate sensma at apex of M3 are pincer. but a hyaline break between parastigma Hke Mesosoma (Figs. 2-6): Relative mea- and marginal vein. Forewing with stigmal surements: Lengm 95/ width 66/ iength of and postmarginal veins very short. Disc of mesoscutum 37/ of scutellum 45, width of wing sparsely setose. Hairlines (Rsl and r- ^^ 4Q Pronotum beari 3^ irs of m) originating from stigmal vein present. ^ Qn terior in Mid iobe of me_ Cubital vein setose and distinct. Metaso- r , _ „° . , , „«*,.,. _. ^x ^ . , soscutum bearing 7-8 pairs of short setae ma (Fie. 8): Caster on a transverse petiole, , ,, ,, ,. °, . .„„ . , „& . T . f .. .. laterallv. Median furrow on posterior two with 7 tergites. Last tergite not subdivid- Li . , ,1 , c . n -.u , „ ,^ . f? Ll , thirds, smooth on bottom. Scutellum with ed. Cereal hairs on small tubercles. , . ., , c r.£.u furrow visible on anterior tour fifths. Trisecodes agromyzae Delvare and Three pairs of short hairs respectively LaSalle, new species slightly before and behind mid length of (Figs. 1-12) scutellum, last pair, longer, near apex. Female. — Length 0.70-0.85 mm. Color: Dorsellum as a raised reticulate plate. Lat- Body, legs except tarsi and antennae dark eral part of metanotum also reticulate but Volume 9, Number 2, 2000 309 sculpturing finer than on dorsellum. Pro- Material examined. — Holotype 9 : BE- podeal callus with one short seta. Lower LIZE: Cayo Province, Las Cuevas, 550 m, mesepimeron ventrally punctate. Meta- Chiquibul Forest (Lewis O. T.), ex Haplo- pleuron smooth. Fore basitarsus with a myza sp. on Senna cobanensis, 27 VIII 1998 row of 5 bristly hairs ventrally. Relative (NQ 5649) (Lewis O. T.)(in BMNH). Para- measurements on mid leg. Tibia 51, tarsus types (in BMNH, CIRAD, USNM, CNC). 40, basitarsus 15 (on dorsal outline), apical Same locality and collector, ex Calycomy- spur of tibia 14; latter simple and straight, zal cassiae on Senna cobanensis, 21 IX 1997 Measurements of hind leg. Tibia 53, basi- (1 6,4 9 9) (NQ 16.040, 16.042, 16.043 & tarsus 14, apical spur of tibia 7. Forewing 16.046), 26 IX 1997 (1 9 ) (NQ 389), 5 XI (Figs 9-10): Relative measurements: 1997 (1 6) (NQ 1543), 26 VIII 1998 (1 6, 1 Length 164, width 82, costal cell 45, mar- 9) (NQ 5573), 29 VIII 1998 (1 9) (NQ 5716) ginal vein 40, stigmal 5, postmarginal 3, and 31 VIII 1998 (1 9 ) (NQ 5914), 4 IX 1998 fringe 9. Wing very distinctly widened at (19) (N9 6023); 1 6, same locality and apex and regularly rounded. Costal cell collector, 5 II 1998 (N° 1719); same locality with 3-4 dorsal hairs at apex, ventrally and collector, on Stizophyllum riparium, 6 bare. Parastigma bearing 2 dorsal hairs, VI 1998 (1 6 ) (NQ 3300), 30 VII 1998 (1 6 ) marginal vein with 9-10 dorsal hairs of (Ne 4895); 1 8, same locality and collector, about twice the width of the vein. A line ex Nesomyza sp. on Amphilophium panicu- of 7-8 setae present near anterior wing latum (N- 4805); 1 8, same locality and margin, originating near stigmal vein and collector, ex Calycomyza sp. on Triwnfetta extending apically. Basal vein with 2-3 bogotensis, 6 VII 1998 (N° 4305); 1 8 , same hairs; another one closely beside it. Anal locality and collector, on Sida rhombilolia, vein (subcubital) represented by a row of 11 VIII 1998 (N° 5342); same locality and 5-6 sparse hairs in front of the retinacu- collector, ex Calycomyza sidae on Sida acuta, lum and more basally by 2-3 isolated 10 X 1997 (1 6,1 9) (NQ 905 & 598), 5 II hairs behind the basal vein. Basal cell bare. 1998 (1 6,4 9 ) (NQ 1700, 1701, 1702, 1703 Hind wing: Relative measurements: & 1704); 5 IX 1998 (1 9) (N2 6125). COSTA Length 112, width 33. Wing bearing 3 RICA: Guanacaste, Santa Rosa N. P., 300 hamuli, narrowly rounded at apex. Me- m, 25 IV / 16 V 1987 (Janzen D. H. & tasoma (Fig. 8): petiole with transverse Gauld I. D.) H-l-O (H)& H-2-C (1 6). lamina on anterior margin. Gaster about GUADELOUPE:, Bouillante, Pigeon, 7 XI 1.35 times as long as wide (95:69) on slide- 1995, reared from a leaf of Daphnopsis mounted specimens. Gastral tergite 1 de- americana mined by Liriomyza sclnnidti, pressed behind petiole. Tergite 2 with a (Etienne J.)(l 9) (Ne 13783/GP 1195). row of 5 hairs laterally (Fig. 8). Each of the Distribution. — Belize, Costa Rica and following tergites with one dorsal line of Guadeloupe (Neotropical Region), hairs. Ovipositor valvula 70, gonostyli 12. Hosts. — Agromyzidae (Diptera) mining Anterior end of the ovipositor near the plants belonging to various families (Big- base of the gaster. Cereal hairs short, on noniaceae: Stizophyllum riparium and Am- small tubercles. philophium paniculatum; Leguminosae: Sida Male (Fig. 12). — Length 0.60-0.80 mm. rhombilolia; Malvaceae: Sida acuta and S. All characters identical but antennal fla- rhombifolia; Thymeleaceae: Daphnopsis gellum devoid of elongate sensilla, latter americana; Tiliaceae: Triwnfetta bogotensis) replaced by whorls of long hairs. Only 1, and belonging to several different genera: rarely 2, ring-like sensilla on each of the Liriomyza, Calycomyza and Nesomyza. flagellar segments. Gaster length 80, width According to O. T. Lewis (pers. comm.), 45. One long cereal hair. Two long hairs T. agromyzae is a larval parasitoid. on apex of epipygium. Systettiatic placement. — The combination 310 Journal of Hymenoftera Research of characters exhibited by T. agromyzae (1997) showed that the tibial spur is more does not match any of the currently rec- variable than formerly thought. However ognized families of chalcids. Trichogram- recent molecular studies (Gauthier e t al. matidae is the only family of Hymenop- 2000) demonstrate that Eulophidae when tera having 3-segmented tarsi in both sex- including Elasmus does actually represent es, and it has long been considered to be a monophyletic group, one of the most easily recognizable and Many of the derived states found in In- definable families of Chalcidoidea because secodes are found in some eulophids (see of this character (Gibson et al. 1999). Ad- Table 1). For example the shape and place- ditionally, many trichogrammatids have ment of the frontal sutures, the reduced lines of setae on the forewing. On the sur- number of tarsal segments, the reduced face, it would thus appear that Trisecodes number of setae on the submarginal vein, should be placed in this family. However, the short stigmal vein, the very short post- the number of tarsal segments is a reduc- marginal vein, and the presence of line of tion character, and a loss of a tarsal seg- hairs on the forewing. The fore tibial spur ment (5 to 4) is seen many times in the of Trisecodes is also similar to what was Chalcidoidea in the families Eulophidae, recently illustrated for the Eulophidae Encyrtidae, Aphelinidae, Mymaridae, Te- (LaSalle et al. 1997). Eulophidae include fi- tracampidae, and even Pteromalidae nally many larval or /and pupal parasit- (LaSalle et al. 1997). It is not inconceivable oids of agromyzid flies, that the loss of an additional tarsal seg- If we consider the following suite of ment has occured in one of these other lin- characters 1 ) reduced number in antennal eages. The presence of lines of setae on the segments; 2) special structure of the fore forewing radiating from the stigma is also tibial spur; and 3) petiolate gaster with known in several chalcid families: Tricho- phragma restricted to mesosoma, not en- grammatidae, Eulophidae (Euderinae, En- tering metasoma, then the unique possi- tedoninae), Torymidae and Pteromalidae bility of familial placement for Trisecodes (Colotrechninae, Ormocerinae: Systasini), within a currently recognized family of and does not restrict a genus to any one Chalcidoidea is Eulophidae. family. Moreover, in trichogrammatids, Table 1 lists the characters shared by the funicle is at most 2-segmented, the Trisecodes with species of Eulophidae ac- gaster is broadly attached to the mesoso- cording to their subfamilial placement, ma and the phragma goes into the meta- Some are left unpolarized, others are de- soma, most often deeply so; the frontal su- rived. The table shows that Trisecodes tures are also different from Trisecodes, the shares the largest number of derived sim- mesosoma never has raised reticulation, ilarites with the Entedoninae. and the fore tibial spur is simple instead It must however be mentioned that the of bifid. Finally all the trichogrammatids placement of Trisecodes within this sub- whose biology is known are egg parasit- family poses problems as it differs from oids of various insects. most other entedonines in several impor- Despite only having three tarsal seg- tant characters. Entedoninae is one of the ments, we are therefore placing Trisecodes best defined subfamilies of the Eulophidae in the Eulophidae. Unfortunately, there is (Boucek 1988; Schauff 1991). Support for little convincing morphological evidence its monoplyly includes: scutellum with a for the monophyly of this family. The single pair of setae; submarginal vein with combination of 4 tarsal segments and a two dorsal setae; mesoscutal midlobe with short, simple fore tibial spur have been two pairs of setae; face with frontal su- considered to be the best characters sup- tures distinctly separated from the anteri- porting their monophyly, but LaSalle et al. or ocellus; male scape with sensory pores Volume 9, Number 2, 2000 311 Table 1. Morphological characters of Trisecodes and their distribution in the subfamilies of Eulophidae. Subfamily Eulophinae Euderinae Tetrastichinae Entedoninae many species many species Character Unporalized characters notauii: complete and deep scutellum general habitus mesonotum: raised reticulation mesopleuron Derived states antennal segments: all species reduced number mandibular formula frontal sutures anelli: reduced number mesoscutum: median groove scutellum: median groove hairs on propodeal callus: reduced number hairs on submarginal vein: reduced number fore tibial spur bifid (LaSalle et al. 1997) short stigmal vein short postmarginal vein lines of hairs figuring Rsl, & r-m all species all species most species all species some species (at least) most species most species most species all species few species very few species most species some species: (see Schauff 1991 fig. 81 p. 101) all species hosts: Agromyzidae several species a few species many species some species some species (at least) most species a few species all species some species most species many species some species some species many species all species some species (at least) all species most species some species several species restricted to the ventral edge; propodeum with a subspiracular tubercle; marginal vein relatively long; stigmal vein relative- ly short (Boucek 1988; Schauff 1991). Unfortunately, many of these characters are absent in Trisecodes. There are 3 pairs of very small setae on the scutellum, onlv a single seta on the dorsal surface of the submarginal vein, and no subspiracular spiracle. Also the notauii are complete, an- other character which is unusual but not unknown in the Entedoninae. However, exceptions are known to all of the char- acters listed above (see Ubaidillah et al. 2000 for some examples), and Trisecodes does possess one very strong character to support it as an entedonine, the shape and placement of the frontal sutures. Addi- tionally, one genus of Entedoninae, Ase- codes, which includes species previously 312 Journal of Hymenoptera Research placed in Teleopterus (Hansson 1996), may have the forewing with almost identical lines of setae radiating from the stigmal vein. However, it is doubtful that there is a close relationship between Trisecodes and Asecodes based on other characters, such as the strength of the sculpturing in Triseco- des. ACKNOWLEDGMENTS We are grateful to Dr. O. T. Lewis who provided most of the material used for the description of the Trisecodes and for the information on its biology. We also thank Dr. C. Hansson and Dr. J. Pinto for dis- cussion about the placement of Trisecodes agromyzae and useful comments on the manuscript. LITERATURE CITED Boucek, Z. 1988. Australasian Clialcidoidea (Hymenop- tera): A biosystematic revision of genera of fourteen families, with a reclassification of species. Walling- ford, UK, CAB International, 832 pp. Gauthier, N., J. LaSalle, D. L. J. Quicke and H. C. J. Godfray. 2000. Phylogeny of Eulophidae (Hy- menoptera, Chalcidoidea), with a reclassification of Eulophinae and the recognition that Elasmi- dae are derived eulophids. Systematic Entomology (in press). Gibson, G. A. P. 1990. A word on chalcidoid classi- fication. Chalcid Forum, 13: 7-9. Gibson, G. A. P., J. M. Heraty and J. B. Woolley. 1999. Phylogenetics and classification of Chalcidoidea and Mymarommatoidea a review of current con- cepts (Hymenoptera, Apocrita). Zoologica Scripta, 28(1-2): 87-124. Hansson, C. 1996. The status of the genera Asecodes Forster, Ionympha Graham and Teleopterus Silves- tri (Hymenoptera: Eulophidae), with a review of Nearctic species. Entomologica Scandinavica, 27(2): 159-167. LaSalle, J., A. Polaszek, J. S. Noyes and G. Zolnerow- ich. 1997. A new whitefly parasitoid (Hymenop- tera: Pteromalidae: Eunotinae), with comments on its placement, and implications for classifica- tion of Chalcidoidea with particular reference to Eriaporinae (Hymenoptera: Aphelinidae). Sys- tematic Entomology, 22(2): 131-150. Noyes, J. S. 1998. Catalogue of the Chalcidoidea of the World. CD-Rom. Amsterdam, Nederland, Expert Center for Taxonomic Information. Schauff, M.E. 1991. The Holarctic genera of Entedon- inae (Hymenoptera: Entedoninae). Contributions of the American Entomological Institute, 26(4): 1- 108. Ubaidillah, R., J. LaSalle, D. J. L. Quicke. 2000. A pe- culiar new genus and species of Entedoninae (Chalcidoidea: Eulophidae) from Southeast Asia. Journal of Hymenoptera Research, 9: 170-175. J. HYM. RES. Vol. 9(2), 2000, pp. 313-319 Description of a New Genus of Entedoninae (Hymenoptera: Eulophidae) from the Neotropical Region, Including Three New Species Christer Hansson Department of Zoology, Lund University, Helgonavagen 3, S-223 62 Lund, Sweden, e-mail: christer.hansson@zool.lu.se Abstract. — Acanthala gen.n. including three new species, albiclava, plaumanni, pubipennis, of the subfamily Entedoninae (Hymenoptera: Eulophidae) is described from the Neotropical Region (Belize, Brazil and Costa Rica). Acanthala is unique among Eulophidae by having the forewing with a row of strong setae on the dorsal surface of the marginal vein. The three species are known only from the female sex and nothing is known about their biology. The Eulophidae has a worldwide distri- bution and is one of the largest families of Chalcidoidea. The knowledge of the group is unevenly distributed, with a strong displacement towards the northern hemisphere, although the group is expect- ed to be more species rich in tropical ar- eas. Estimates of the eulophid fauna in the New World tropics corroborates this (e.g. Gaston et al. 1996, LaSalle and Schauff 1995), but very little is known about the eulophid fauna of this region. To. increase our knowledge of eulophids in general, and of the eulophid fauna in the Neotropical region in particular, a new ge- nus with unique morphological features is described below. The descriptions of three new species belonging to the new genus are also included. Unfortunately information regarding the biology is not yet known. Acronyms of museums used in the text are as follows: BMNH: The Natural His- tory Museum, London; CNC: Canadian National Collections of Insects and Arach- nids, Ottawa; INBio: Instituto Nacional de Biodiversidad, Santo Domingo, Costa Rica; LUZM: Lund University Zoology Museum, Sweden; MIUCR: Museo de In- sectos, Universidad de Costa Rica; USNM: United States Museum of Natural History, Washington, D.C. Acanthala Hansson, new genus Type species. — Acanthala pubipennis Hansson, new species. Diagnosis. — Dorsal surface of marginal vein with a row of strong setae (Figs. 1, 8); eyes hairy (Figs. 1, 6, 7); mandibles with a single tooth at apex (Fig. 7); frontal cross-groove incomplete, not reaching eyes (Fig. 6), or missing (Fig. 7); pedicel conspicuously hairy on dorsal surface (Figs. 2-A); mesoscutum and scutellum with small-meshed and strong reticulation (Fig. 1), hence dull. Description. — Flagellum with sensilla ampullacea short and symmetric, present on all segments. Antenna with 2-3 anelli. Mandibles with a single tooth at apex. Clypeus weakly delimited laterally, but not delimited dorsally. Antennal scrobes join on frontal cross-groove (or cross- groove missing). Frontal cross-groove V- shaped, not reaching eyes, or missing. Eyes hairy. Occiput with a weak median groove in upper part, close to occipital margin (plaumanni, pubipennis), or median groove missing (albiclava). Pronotum well developed and clearly visible in dorsal view, without transverse carina. Midlobe of mesoscutum with two pair of strong se- tae; notauli not visible (plaumanni, pubipcn- 314 Journal of Hymenoptera Research Fig. 1. Acanthala pubipennis habitus. nis), or indicated in anterior 1/2 (albiclava). Scutellum with one pair of strong setae, situated at equal distance from anterior and posterior margins of scutellum (pubi- pennis) or closer to anterior margin (albi- clava, plaumanni). Transepimeral sulcus (i.e. the sulcus separating upper and lower mesepimeron) almost straight (pubipennis, albiclava) or curved (plaumanni). Dorsellum visible in dorsal view. Propodeal callus with two setae. Forewing rounded; costal cell narrow; postmarginal vein 0.8-1.6 X as long as stigmal vein; speculum open (al- biclava, plaumanni) or closed (pubipennis) below; radial cell bare, without stigmal hairlines. Petiole short, hardly visible in dorsal view, and transverse. Biology. — Not known. Distribution. — Neotropical region (Be- lize, Brazil, Costa Rica). Etymology. — Named after the row of strong spinelike setae on dorsal surface of marginal vein: acanth-ala = spiny wing. The gender is regarded as feminine. Volume 9, Number 2, 2000 315 Figs. 2-8. Acanthala spp. females. 2-4, antenna in lateral view: 2, A. pubipennis; 3, A. plaumanni; 4, .-\. albiclava. 5, thoracic dorsum in lateral view of A. pubipennis. 6-7, head in frontal view: 6, A. pubipennis; 7, A. albiclava. 8, forewing of A. albiclava. Discussion. — The dorsal row of setae on the marginal vein is a unique character state for Acanthala within the Eulophidae, and hence a strong apomorphy. The man- dibles with a single apical tooth is also unique to Acanthala. A similar character state is present in some species of Paracrias Ashmead, but different from Acanthala since in Paracrias there is a single large tooth and a small second tooth dorsally (Fig. 14 in Schauff 1985). Another apo- morphy present in Acanthala is the incom- plete or missing frontal cross-groove. However, this apomorphy is present in several entedonine genera, e.g. in some species of Chn/socharis Forster, in most species of Entedon Dalman, in Entcdono- necremnus Girault and Eprhopalotus Gi- rault. The occurence among genera not otherwise shown to be closely related (e.g. Schauff 1991, LaSalle & Schauff 1994) does not indicate a high information value with regard to relationship. The dorsal row of strong setae on mar- ginal vein and the dense and strong but still fine reticulation on vertex and thorac- ic dorsum makes Acanthala easily recog- nizable, habituallv not resembling any 316 Journal of Hymenoptera Research other entedonine genus from the Neotrop- ical region. In Boucek (1988) Acanthala runs either to Chri/socharis (couplet 149), or to Chrysono- tomyia (couplet 153). In Schauff et al. (1997) Acanthala runs either to couplet 126 (Chrysocharis or Grahamia) or (with some difficulties, due to the fact that Acanthala does not possess the complete combina- tion of characters presented in the cou- plets) to either Asecodes (couplet 132) or to Neochrysocharis (couplet 134). However, the row of strong setae on dorsal surface of marginal vein makes Acanthala easy to separate from above mentioned genera, and from any other entedonine genera. Note: the first character used under cou- plet 134 in Schauff et al. (1997), the shape of the transepimeral sulcus has been con- fused: Closterocerus has a strongly arched sulcus while Neochrysocharis has a weakly curved or straight sulcus! (Hansson 1995). KEY TO FEMALES OF ACANTHALA 1. Predominantly yellowish-brown nonmetallic species; antenna with a distinct antennal clava (Fig. 4), clava white and remaining flagellum brown; frontal cross-groove missing (Fig. 7) albiclava new species - Predominantly dark and ± metallic species; antenna without distinct clava (Figs. 2, 3), flagellum completely pale brown; frontal cross-groove present medially (Fig. 6) 2 2. Forewing (Fig. 1) with comparatively dense setation, speculum closed below; flagellum narrower, e.g. flagellomeres II and III 2x as long as wide, and with more distinct constric- tions between flagellomeres (Fig. 2) pubipennis new species - Forewing with comparatively sparse setation (as in albiclava (Fig. 8)), speculum open below; flagellum stouter, e.g. flagellomeres II 1.7X and III 1.4 x as long as wide, and with less distinct constrictions between flagellomeres (Fig. 3) plaumanni new species Acanthala albiclava Hansson, new species (Figs. 4, 7, 8) Diagnosis. — Predominantly yellowish- brown nonmetallic, with only major part of vertex and upper 1/2 of occiput metal- lic (bluish-purple); antenna (Fig. 4): scape comparatively wide, 3.8 X as long as me- dian width, yellowish-brown; pedicel and flagellomeres 1-3 brown, flagellomeres 4- 5 white with ventral surface densely se- tose, flagellum with a distinct clava; with- out frontal cross-groove (Fig. 7); setae on vertex and thoracic dorsum comparatively thin, as thick as setae on marginal vein; compared to pubipennis (Fig. 1), forewing in albiclava (Fig. 8) with sparse setation (wing surface distad of speculum with same setation, Fig. 8 only shows setation below base of marginal vein and on sur- face just distad of postmarginal and stig- mal veins), speculum open below (i.e. cu- bital hairline missing below speculum), hind margin of forewing strongly curved upwards just below base of marginal vein; anteromedian part of propodeum strongly raised into a peak; propodeum with a complete median carina that splits in two carinae in posterior part; propodeal sur- face reticulate. Female. — Length of body = 0.9-1.0 mm. Colour: Scape yellowish-brown; pedicel and flagellomeres 1-3 brown, flagello- meres 4-5 white. Frons yellowish-brown, with a white stripe from eye to eye along upper border (Fig. 7). Vertex yellowish- brown in front of anterior ocellus, remain- ing vertex metallic bluish-purple. Upper 1 /2 of occiput metallic bluish-purple, low- er 1/2 yellowish-brown. Mesosoma, in- cluding legs, yellowish-brown. Forewing with a weak infuscate stripe below stigmal vein, stripe reaches hind margin of wing. Volume 9, Number 2, 2000 317 Gaster yellowish-brown. Head: Antenna as in Fig. 4; with three discoid anelli. Ra- tios height of eye /malar space /width of mouth: 2.1/1.0/1.4. Frons with rather strong small-meshed reticulation, meshes isodiametric. Vertex dull, with strong small-meshed reticulation. Ratios distanc- es between posterior ocelli /one posterior ocellus and eye /posterior ocelli and occip- ital margin: 2.0/1.0/1.0. Occiput without a weak median groove in upper part; oc- cipital margin rounded. Ratio width of head /width of thorax (measured across mesoscutum, just in front of base of fore- wing) = 1.2. Mesosoma: Mesoscutum and scutellum dull, with rather strong small- meshed reticulation, meshes on mesoscu- tum isodiametric, on scutellum slightly elongate. Dorsellum concave and strongly reticulate with small meshes. Forewing speculum open below; ratio length of postmarginal vein/ length of stigmal vein = 0.8; ratios length of wing (measured from base of marginal vein to the point along outer margin of forewing farthest away from base of marginal vein) /length of marginal vein /height of wing: 1.9/1.0/ 1.0. Anteromedian part of propodeum strongly raised into a peak, with a com- plete median carina that splits in two ca- rinae in posterior part; propodeal surface reticulate with small meshes. Metasoma: Gaster ovate; ratio length of mesosoma/ length of gaster = 0.7-0.8. Type material. — Holotype female: BRA- ZIL: Bahia Itabuna, ll-14.ii.1984, F. Ben- ton (deposited in BMNH). Paratypes: Two females with same label data as holotype (1 female in BMNH, 1 female in LUZM); from same locality as holotype but col- lected iv.1983 (1 female, in BMNH), 2- 6.V.1983 (1 female, in USNM), viii.1983 (1 female in BMNH, 1 female in LUZM). Etymology. — Named after white anten- nal clava: albi-clava = white club. Acanthala plaumanni Hansson, new species (Fig. 3) Diagnosis. — Predominantly dark and metallic species; entire antenna brown, scape comparatively narrow, 5.7X as long as median width (Fig. 3), flagellum with- out distinct clava, flagellomeres stout and with less distinct constrictions between them; head shrivelled in type series, but frontal cross-groove visible at least medi- ally; setae on vertex and thoracic dorsum comparatively strong, about twice as thick as setae on marginal vein (as in pubipennis (Fig. 5)); hind margin of forewing not strongly curved upwards below base of marginal vein (as in pubipennis (Fig. 1)); forewing with comparatively (compared to pubipennis (Fig. 1)) sparse setation (as in albiclava (Fig. 8)), speculum open below (as in albiclava (Fig. 8)); propodeum with weak reticulation, smooth and shiny in some places. Female. — Length of body = 0.8 mm (in both type-specimens). Colour: Antenna pale brown. Frons golden-green. Vertex metallic bluish-purple. Occiput golden. Mesoscutum golden-green. Scutellum me- tallic purple in median 1/2, golden-green in lateral 1/4 in holotype; paratype with entire scutellum golden-green. Propo- deum golden-green. Fore and hind coxae dark and metallic, mid coxa infuscate; femora infuscate; tibiae and tarsi pale. Forewing weakly infuscate below margin- al vein, infuscation reaching to hind mar- gin of wing. Gaster golden-purple. Head: Antenna as in Fig. 3; with one discoid and one slightly larger anellus. Ratios height of eye /malar space /width of mouth: 1.9/ 1.1/1.0. Frons with rather weak small- meshed reticulation, meshes ± isodiamet- ric. Vertex with weak reticulation, shiny. Ratios distances between posterior ocelli/ one posterior ocellus and eye /posterior ocelli and occipital margin: 1.7/1.0/1.0. Occiput with a weak median groove in upper part, close to occipital margin; oc- cipital margin rounded. Ratio width of head /width of thorax (measured across mesoscutum, just in front of base of fore- wing) = 1.0. Mesosoma: Mesoscutum and scutellum dull, with rather strong small- meshed reticulation, meshes isodiametric. 318 Journal of Hymenoptera Research Dorsellum concave and reticulate. Fore- coxa pale; femora dark; tibiae and tarsi in- wing speculum open below; ratio length fuscate. Forewing weakly infuscate below of postmarginal vein /length of stigmal marginal vein, infuscation reaching to vein = 1.4; ratios length of wing (mea- hind margin of wing. Gaster golden-pur- sured from base of marginal vein to the plish. Head: Antenna as in Fig. 2; with point along outer margin of forewing far- two discoid anelli. Ratios height of eye/ thest away from base of marginal vein)/ malar space /width of mouth: 1.6/1.0/1.0. length of marginal vein /height of wing: Frons with rather weak small-meshed re- 2.0/1.2/1.0. Propodeum with weak retic- ticulation, meshes transverse. Vertex dull, ulation, smooth and shiny in some places, with strong small-meshed reticulation. Ra- Metasoma: Gaster ovate; ratio length of tios distances between posterior ocelli/ mesosoma /length of gaster = 0.9-1.0. one posterior ocellus and eye /posterior Type material. — Holotype female: BRA- ocelli and occipital margin: 2.0/1.7/1.0. ZIL: Santa Catarina, Nova Teutonia, Occiput with a weak median groove in xi.1949, F. Plaumann (deposited in upper part, close to occipital margin; oc- BMNH). Paratype: One female from same cipital margin rounded. Ratio width of locality as holotype but collected ix.1943 head /width of thorax (measured across (in BMNH). mesoscutum, just in front of base of fore- Etymology. — Named after F. Plaumann, wing) = 0.9. Mesosoma: Mesoscutum and collector of type series. scutellum dull, with rather strong small- meshed reticulation, meshes isodiametric. Acanthala pubipennis Hansson, new Dorsellum convex and reticulate with species small meshes. Forewing speculum small (rigs, i, z, d, t>; ancj cioseci below, cubital hairline curved Diagnosis. — Predominantly dark and upwards towards base of marginal vein; metallic species; entire antenna brown, ratio length of postmarginal vein /length scape comparatively narrow (Fig. 2), 4.7X of stigmal vein = 1.6; ratios length of wing as long as median width, flagellum with- (measured from base of marginal vein to out distinct clava, flagellomeres slender the point along outer margin of forewing and with distinct constrictions between farthest away from base of marginal them; setae on vertex and thoracic dorsum vein) /length of marginal vein /height of comparatively strong, about twice as thick wing: 1.8/1.1/1.0. Propodeum smooth as setae on marginal vein (Fig. 5); frontal and shiny; propodeal callus with small- cross-groove present, missing only close meshed reticulation. Metasoma: Gaster to eyes (Fig. 6); forewing (Fig. 1) with ovate; ratio length of mesosoma /length of comparatively dense setation, speculum gaster = 0.8-1.0. small and closed below, cubital hairline Type material. — Holotype female: BE- curved upwards towards base of marginal LIZE: Las Cuevas, ix.1995, T. King & A. vein; hind margin of forewing not strong- Howe (deposited in BMNH). Paratypes: ly curved upwards below base of margin- Following from same locality as holotype al vein; propodeum smooth and shiny but collected iv.1995 (1 female, in LUZM), (Fig. 1). vi.1995 (1 female, in BMNH); 1 female Female. — Length of body = 0.7-1.0 mm. COSTA RICA: Alajuela, Perias Blancas, Colour: Antenna pale brown. Frons dark 700m, ii.1987, E. Cruz (in CNC); 1 female with weak golden tinges. Vertex dark with COSTA RICA: Guanacaste, P.N. Santa weak metallic tinges. Occiput dark with Rosa, 300m, 20. xii. 1986-10. i. 1987, D.H. metallic tinges. Mesoscutum, scutellum Janzen & I.D. Gauld (in BMNH); follow- and propodeum dark with golden-purple ing from same locality as previous but col- tinges. Fore and hind coxae dark, mid lected 31. i-21 .ii.1987 (1 female, in USNM), Volume 9, Number 2, 2000 319 16.v-6.vi.1987 (1 female, in INBio); 1 fe- male COSTA RICA: Puntarenas, Golfo Duke, 3km SW Rincon, 10m, vii.1991, P. Hanson (in MIUCR). Etymology. — Named after densely setose forewing: pubi-pennis = hairy wing. ACKNOWLEDGMENTS I am indebted to the following persons for their aid in loan of material: C. Godoy and J. A. Ugalde (IN- Bio), P. Hanson (MIUCR), J.T. Huber (CNC) and J.S. Noyes (BMNH). The TMR Programme (of the Euro- pean Union) is hereby acknowledged for funding a stay at the NHM in 1999. I also would like to thank the Board of Trustees of the Natural History Museum in London for permitting access to NHM collections. LITERATURE CITED Boucek, Z. 1988. Australasian Chalcidoidea (Hymenop- tera), a biosystematic revision of genera of fourteen fatuities, with a reclassification of species. CAB In- ternational. Wallingford. 832 pp. Gaston, K.J.; Gauld, I.D. and Hanson, P. 1996. The size and composition of the hymenopteran fauna of Costa Rica, journal of Biogeography 23:105-113. Hansson, C. 1995. Revision of the Nearctic species of Neochrysocharis Kurdjumov (Hymenoptera: Eu- lophidae). Entomologica Samdinavica 26:27-46. LaSalle, J. & Schauff, M.E. 1994. Systematics of the tribe Euderomphalini (Hymenoptera: Eulophi- dae): parasitoids of whiteflies (Homoptera: Al- eyrodidae). Systematic Entomology 19: 235-258. LaSalle, J. and Schauff, M.E. 1995. Eulophidae, in Hanson, P.E. and Gauld, I.D. The Hymenoptera of Costa Rica. Oxford University Press. 893 pp. Schauff, M.E. 1985. The New World genus Paracrias Ashmead (Hymenoptera: Eulophidae). Proceed- ings of the Entomological Society of Washington 87: 98-109. Schauff, M.E. 1991. The Holarctic genera of Entedon- inae (Hymenoptera: Eulophidae). Contributions of the American Entomological Institute 26(4): 1-109. Schauff, M.E.; LaSalle, J. and Coote, L.D. 1997. Eulo- phidae, in Gibson, G.A.P.; Huber, J.T. and Wool- ley, J.B. Annotated Keys to the getiera of Nearctic Chalcidoidea (Hymenoptera). NRC Research Press. Ottawa. 794 pp. J. HYM. RES. Vol. 9(2), 2000, pp. 320-323 The Identity of Pteroptrix imitatrix (Fullaway) (Hymenoptera: Aphelinidae) Andrew Polaszek and Gerhard L. Prinsloo (AP) Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park Asoct, Berks SL5 7PY, UK; (GLP) Plant Protection Research Institute, Private Bag X134, Pretoria 0001, South Africa Abstract. — Taxonomic notes are provided on two aphelinid wasps, Pteroptrix imitatrix (Fulla- way), described from Hawaii, and Pteroptrix albifemur (Girault), described from Australia. The former is shown to be a synonym of the latter. New taxonomic, distributional and biological data are provided. Among several Aphelinidae collected in type designation and distributional infor- the Galapagos Islands by Dr John Heraty mation are published to clear up some of (UCR) were three specimens of a Pteroptrix the many taxonomic problems that still species belonging to the maritima-group exist in this genus. sensu Viggiani and Garonna (1993). This Terminology follows Hayat (1983) ex- species-group formerly comprised part of cept that the terms mesosoma and meta- the genus Archenomus, and using the key soma replace thorax plus propodeum, and to world species of Archenomus by Prins- gaster, respectively. Abbreviations of de- loo and Neser (1990) the specimens were positories can be found under "Acknowl- identified by the first author as Pteroptrix edgments". ( = Archenomus) albifemur (Girault). Con- sulting the description of P. imitatrix (Ful- Pteroptrix albifemur (Girault 1915) laway) and non-type material of that spe- (Fi8s- l-3) cies at the United States National Muse- Aptewptrix albifemur Girault 1915: 65. urn, it became apparent that the speci- Archenomus albifemur (Girault): Prinsloo and mens from the Galapagos were also very Neser 1990: 23. close, if not identical, to that species. Pteroptrix albifemur (Girault): Viggiani and Gar- Prinsloo and Neser did not treat P. imita- onna 1993: 61; Hayat 1998:245. trix as the type was not located by them. Pseudopteroptrix imitatrix Fullaway 1918: 464. P. imitatrix was described from Hawaii, Syn. nov. and after consultation with Mr G. Nishida Archenomus imitatrix (Fullaway): Prinsloo and of the Bishop Museum the type material Neser 1990: 23. was finally located by Mr B. Kumashiro at Pteroptrix imitatrix (Fullaway): Viggiani and the Department of Agriculture. We have Garonna 1993: 60. compared the types of P. albifemur (Gi- Female.— Colour: Antenna with flagel- rault), P. imitatrix (Fullaway) and the Gal- lUm pale brown, the scape with its outer apagos (Ecuador) material, and find them edge darker brown, radicle dark brown. to be conspecific. Subsequently, material Head with the lower occiput, genae and from Florida, Puerto Rico and India was stemmaticum dark brown, the eye mar- also found to belong to P. albifemur. The gins darkest, frons and upper occiput pal- following synonymy, redescription, lecto- er brown. Mesosoma brown, the scutel- Volume 9, Number 2, 2000 321 Figs. 1-3. Pteroptrix albifemur, female. 1, Antenna. 2, Fore wing. 3, Mesonotum, showing sculpture of mesos- cutal mid lobe (specimen from Santa Cruz, Galapagos). lum strikingly white, side lobes and bor- der of mesoscutum paler brown. Legs white, the hind coxa and leading edge of the hind femur slightly darkened. Wings hyaline, slightly darkened below the mar- ginal vein. Metasoma brown. Morpholo- gy: Mandibles with two teeth and a trun- cation. Antennal formula 1,1,3,3 (fig. 1); scape slightly more than 2 X pedicel length. Funicle segments all longer than wide, subequal in length. F2 slightly the shortest; funicle about 0.5 x length of club. Flagellum with the following num- bers of longitudinal sensilla: Fl: 0; F2: 0; F3: 0-1; F4: 3-4; F5: 3-4; F6: 3. Mid lobe of mesoscutum (fig. 3) with 6-7 setae, each lateral lobe with 1, each axilla with 1, and scutellum with 4. Fore wing (fig. 2) with 1 seta on submarginal vein, 2-4 setae in bas- al cell. Anterior margin of marginal vein with 5-6 setae, and one large seta at the junction of the submarginal vein and par- astigma. Maximum width of wing 2.6-2.9 X longest seta on marginal fringe. Length of second valvifers 3.3-3.6 x third valvu- lae. Second valvifer and third valvula combined 1.1 X length mid tibia. Meta- soma oval, longer than wide and about 1.5 X length mesosoma. Terga II-VII with 0, 1 + 1, 1 + 1, 1 + 1, 1+2+1 and 1+2 + 1 setae respectively. Male. — Unknown Variation. — Insignificant in the material examined. 322 Journal of Hymenoptera Research Material examined. — Type material: Ho- mur. Fullaway (1918) mentioned a slide- lotype female Apteroptrix albifemur Girault mounted specimen reared from Hemiber- [AUSTRALIA: Cairns, Gordonvale] Type lesia (as Aspidiotus) rapax (Comstock). This HY/2962 (3894) (QM). Lectotype female specimen has not been seen by us. (here designated) Pseudopteroptrix imitatrix Distribution. — Australia, Ecuador, Ha- Fullaway [HAWAII:] Honolulu. 17.i.[19]18 waii (and presumably widespread in the ex Howardia biclavis (HDA, slide-mount- Pacific), India, Puerto Rico, USA (Florida), ed); paralectotypes, 5 females: 1 female Discussion. — Pteropterix albifemur be- [HAWAII:] Honolulu, Oahu, 12.i.l8 (D.T. longs to the maritima-group of Pteroptrix, Fullaway) Hoivardia biclavis (BMNH, slide- where it was correctly placed by Viggiani mounted). [HAWAII:] Honolulu, Oahu, and Garonna (1993). This group is char- 4.ii.l8 (D.T. Fullaway) Howardia biclavis (1 acterized by lacking the complete sulcus female HDA, card-point; 1 female USNM above the occipital foramen and antennal slide-mounted). [HAWAII:] Tantalus, el. formula of female 1.1.3.3. Viggiani and 1300 ft (J. Kotinsky) H. biclavis (2female, Garonna (1993) were mistaken in suggest- HDA). Additional material: ECUADOR: ing that P. imitatrix belongs to the bicolor- Galapagos, Santa Cruz, Darwin Sta. 20m group. Prinsloo and Neser (1990) placed PAN 14-18.V.91 (J. Heraty) arid zone (lfe- albifemur in their peratus-group, and sug- male, BMNH). ECUADOR: Galapagos, Is- gested imitatrix could belong to their in- abela, C. Azul, 3 Km W Cal. Iguana 200 colus-group. These last-mentioned species m 25.V.91 (J. Heraty) deciduous forest groups were combined into the maritima- H91/061 (1 female, BMNH). ECUADOR: group, following the redefinition of spe- Galapagos, Isabela, Alcedo 7 Km SW NE cies groups by Viggiani and Garonna Playa 600 m 25.vi.91 (J. Heraty) arid forest (1993), a step that was necessitated by the H91/118 (1 K, BMNH). ECUADOR: Fer- incorporation of Archenomus into Pterop- nandina, 5 Km NE Cabo Hammon 110 m trix. Pteroptrix albifemur is morphologically 4-10.V.91 (J. Heraty) pan. Palo Santo forest close to the following species in the mari- H91/031 (1 female, USNM). INDIA: Kar- tima-group: P. opaca Erdos, P. patriciae nataka, 25 Km W. of Mudigere 28.x. — (Prinsloo and Neser) and P. abnormis 3.xi.l979 J.S. Noyes (1 female, BMNH; (Prinsloo and Neser). We have not exam- Hayat det.). PUERTO RICO: Indiera 9- ined type material of P. opaca (described 10.iii.1936 H.L. Dozier "ex Howardia bi- from Hungary), which appears to be lost clavis on sapotaceous tree, Lucuma sp." (7 (J. Papp, personal communication). We female, USNM). PUERTO RICO: Maya- have, however, examined specimens from guez 18.x. 1935 H.L. Dozier "sweeping Ro- Hungary (though not from the type local- ble and roadside vegetation at 1000 ft" (1 ity) and from Italy, which agree in all re- female, USNM). PUERTO RICO: Rio Pie- spects with the original description. Pter- dras 3.ix.l912 T.H. Jones "from twig of optrix opaca differs from P. albifemur most achiote, Bixa orellana, on which Hoivardia strikingly in the colour of the legs (tibiae biclavis was present" (1 female, USNM). and femora dark in P. opaca, very largely USA: Florida, Oneco J.W. Collins 2.ix.l922 pale in P. albifemur). The wing of P. opaca (1 female, USNM). is much more deeply infuscated below the Host. — Diaspididae: Howardia biclavis marginal vein than in P. albifemur, as well (Comstock). A pantropical, polyphagous as being noticeably more densely setose, species (Williams and Watson, 1988) oc- Pteroptrix abnormis also differs from P. al- casionally recorded as a pest, for example bifemur in the colour and setation of the of citrus (Grillo et al., 1983). The "eulo- fore wings, although the legs and anten- phid" parasite of H. biclavis recorded by nae are very similar in colour and propor- the latter authors could well be P. albife- tions to those of P. albifemur. The sculpture Volume 9, Number 2, 2000 323 of the face and mesoscutum of P. abnormis is, however, much less pronounced than in P. albifemur. Differences between P. pa- triciae and P. opaca are slight, being re- stricted largely to the colour of the legs (although the Fl and F3 of the female an- tenna in P. patriciae are slightly longer than those of P. opaca). In this respect, P. patriciae is somewhat intermediate be- tween P. opaca and P. albifemur. We sug- gest that a thorough review of the species comprising the maritima-group of Pterop- trix be carried out before synonymising any of these species based on the limited material available during this study. ACKNOWLEDGMENTS For the loan or donation of material, we are grate- ful to the following: J. Heraty, University of Califor- nia, Riverside, (UCR); M. Schauff, United States Na- tional Museum, Washington, D.C. (USNM); G. Mon- teith, Queensland Museum, Brisbane, Australia (QM); W. Nagamine, Hawaii Department of Agricul- ture, Honolulu (HDA); G. Viggiani, University of Na- ples, Portici, Italy; J. Papp, Hungarian National Nat- ural History Museum, Budapest. The authors also thank John LaSalle for his help and advice. This study was facilitated by a Smithsonian Institution visitors' grant to the first author, which is gratefully acknowl- edged. Additional space and facilities during this study were kindlv provided by the Department of Entomology, The Natural History Museum, London (BMNH). LITERATURE CITED Fullaway, D.T. (1918) A new genus of pteroptricine Aphelinidae (Hymenoptera). Proceedings of the Hawaiian Entomological Society 3: 463^164. Girault, A. A. (1915) Australian Hymenoptera Chal- cidoidea — VII: Encyrtidae. Memoirs of the Queens- land Museum 4: 1-184. Grillo, M., N. del Valle and M. Alvarez (1983) Ho- wardia biclavis (Comstock) (Homoptera: Cocci- dae) en citricos. Centro Agricola 10: 47-53. Hayat, M. (1983) The genera of Aphelinidae (Hyme- noptera) of the World. Systematic I ntomology 8: 63-102. Hayat, M. (1998) Aphelinidae of India (Hymenoptera: Chalcidoidea): a taxonomic revision. Monographs on Entomology International 13: 1-416. Prinsloo, G.L. and O.C. Neser (1990) The southern African species of Archenomus Howard (Hyme- noptera: Aphelinidae) with a key to the species of the world. Entomology Memoir, Department of Agricultural Development, Republic of South Africa 79: 1-26. Viggiani, G. and A. P. Garonna (1993) Le specie itali- ane del complesso Archenomus Howard, Archen- omiscus Nikolskaya, Hispaniella Mercet e Pterop- trix Westwood, con nuove combinazioni generi- che (Hymenoptera: AphelinidaeJ Bollettino del La- boratory di Entomologia agraria "F Hippo Silvestri" Portici 48: 57-88. Williams, D.J. and Watson, G.W. (1988) The scale in- sects of the tropical South Pacific region Part 1 The armoured scales (Diaspididae) CAB International, Wallingford, U.K. 290 pp. J. HYM. RES. Vol. 9(2), 2000, pp. 324-346 A Review of the Sphex flavipennis Species Group (Hymenoptera: Apoidea: Sphecidae: Sphecini) A. S. Menke1 and W. J. Pulawski (ASM) Systematic Entomology Laboratory, PSI, USD A, National Museum of Natural History, Washington D.C. 20560-0168, USA; (WJP) Dept. of Entomology, California Academy of Sciences, San Francisco, California 94118-4599, USA, e-mail: wpulawski@calacademy.org Abstract. — The Sphex flavipennis species group, a Palearctic assemblage formerly called the max- illosus species group, is characterized, its species are diagnosed, keyed, their distributions sum- marized, and male antennae illustrated. The large Eurasian wasp formerly known as maxillosns Fabricius or rufocinctus Brulle must now be called fnnerarius Gussakovskij. Similarly, the species formerly known as afer Lepeletier must now be called leuconotus Brulle. The following species are included (new synonyms are listed in parentheses): atropilosns Kohl, 1885; flavipennis Fabricius 1793 (rufocinctus Brulle 1833); fnnerarius Gussakovskij 1934 (maxillosus Fabricius 1793, a junior homonym of Sphex maxillosus Poiret 1787; obscurus Fischer de Waldheim 1843; and mavromous- takisi de Beaumont 1947); leuconotus Brulle 1833 (triangulum Brulle 1833, a junior homonym of Sphex triangulum Villers 1789; afer Lepeletier 1845; sordidus Dahlbom 1845; tristis Kohl 1885; pluniipes Radoszkowski 1886, a junior homonym of Sphex plumipes Drury 1773; and pachysoma Kohl 1890); libycus de Beaumont 1956; melas Gussakovskij 1930; and oxianus Gussakovskij 1928 (nubilis de Beaumont 1968). A lectotype is designated for Sphex funerarius Gussakovskij 1930, and a neotype is designated for Sphex leuconotus Brulle 1833. Descriptive notes are provided for the type material of Sphex atropilosns Kohl, fitnerarius Gussakovskij, leuconotus Brulle, rufocinctus Brulle, and triangulum Brulle. It is rather ironic, after all these years, that the proper scientific names have not been established for Sphex maxillosus Fa- bricius 1793 and afer Lepeletier 1845, since these represent two of the largest Palearc- tic sphecid wasps. This problem is cor- rected here. Initially our study was prompted by Menke's examination in 1964 of Brulle's type specimens at the Mu- seum National d'Histoire Naturelle, Paris, which suggested that two taxa recognized by Kohl (1890) as synonyms of maxillosus (rufocinctus Brulle and triangulum Brulle) were not conspecific with that species. This was corroborated by Pulawski in 1975, who re-examined the same material. 'Mailing address: Ammophila Research Institute, 1429 Franklin St., Bisbee, AZ 85603-6211, USA, e-mail: asm@theriver.com In 1994 both of us restudied the types and confirmed our prior assessments of them. These were important findings because van der Vecht (1959) noted that maxillosus Fabricius was a junior homonym and the species needed a replacement name. Van der Vecht believed that leuconotus Brulle was the oldest available replacement for maxillosus. However, we are certain that leuconotus is a senior synonym of afer Le- peletier. In this paper we establish that fu- nerarius Gussakovskij is the proper name for maxillosus Fabricius and leuconotus Brulle the proper name for afer Lepeletier. Our study has enabled us to construct an identification key to the Palearctic spe- cies of the flavipennis group. This key should be regarded as provisional because these wasps are taxonomically difficult, and we have not made an exhaustive Volume 9, Number 2, 2000 325 study of the species. Too little is known about geographic variation in the flavipen- nis group, particularly the color of the wings, setation, legs, and gaster. These may prove to be variable within some spe- cies, but a very large specimen database would be required to resolve species lim- its and the significance of geographic var- iation. The apparent variation in male placoid distribution on the antenna also needs to be carefully analyzed, as well as possible variation in the male genitalia. Currently the male genitalia seem identi- cal or nearly so in the taxa we discuss here. The family group names used in the ti- tle are based on Melo (1999). SOURCES OF MATERIAL Abbreviations used to indicate location of specimens are listed below with corre- sponding institutions. CAS: California Academy of Sciences, San Francisco, California, U.S.A. Copenhagen: Zoologisk Museum, Copen- hagen, Denmark Dresden: Staatliches Museum fur Tierkun- de, Dresden, Germany Genova: Museo Civico di Storia Naturale "Giacomo Doria", Genova, Italy Krakow: Instytut Systematyki i Ewolucji Zwierzat, Polska Akademia Nauk, Kra- kow, Poland Lausanne: Musee Cantonal de Zoologie, Lausanne, Switzerland Lund: University of Lund, Lund, Sweden Menke: A. S. Menke Collection, Ammo- phila Research Institute, Bisbee, Arizo- na, U.S.A. Moscow: Zoological Museum, Moscow State University, Moscow, Russia Palermo: Istituto di Zoologia, Universita di Palermo, Palermo, Italy Paris: Museum National d'Histoire Natu- relle, Paris, France Stockholm: Naturhistoriska Riksmuseet, Stockholm, Sweden St. Petersburg: Zoological Institute, Rus- sian Academy of Sciences, St. Peters- burg, Russia USNM: United States National Museum of Natural History, Washington D.C., U.S.A. Wien: Naturhistorisches Museum, Wien, Austria Zurich: Entomologisches Institut, Eidge- nossische Technische Hochschule, Zur- ich, Switzerland METHODS Morphological terms used here follow Bohart and Menke (1976) except that we follow Salmon (1929) for the name of the curved, cord-like "tendon" at the base of the petiole. Salmon called this the funicle, and we adopt his term because it is not a tendon in the true sense of the word. Measurements of the abdominal petiole are made as follows: the width is mea- sured at the base of tergum I, the length is measured from the base of the funicle (on the petiole) to the base of tergum I. When describing color we use the term "red" to indicate non-black areas of the legs. In reality, the true color more closely approximates amber, or reddish brown in many instances. Setal color is described as pale or white vs. dark brown or black. In some cases, pale or white setae are really silver. For the uncommonly collected species we have listed locality records known to us. The fourth edition of the International Code of Zoological Nomenclature (ICZN 1999) stipulates in Article 74.7.3 that "to be valid, a lectotype designation made af- ter 1999 must contain an express state- ment of the taxonomic purpose of the des- ignation". These statements must accom- pany each designation (Article 74.3). Tra- ditionally lectotype designations have been indicated by the words "present des- ignation" and the purpose is clear to all, namely to fix the identify of the name in- volved. To add a statement after each des- ignation seems redundant and repetitive 326 Journal of Hymenoptera Research to us, but to satisfy the Code, we have fol- gellomeres are more elongate than in lowed "present designation" with the fol- those two species. The female mid and lowing statement "in order to ensure the hindlegs are black, but the foretibia and name's proper and consistent use". tarsus are reddish. Species discrimination in the flavipennis The flavipennis Species Group grQup ig ^^ espedally in females The species discussed here belong to a Previous authors have used wing color, lineage that de Beaumont (1960) called the setal color, presence or absence of long se- maxillosus group. Because maxillosus is a tae on the femora, and proportions of the junior homonym, we have renamed it the petiole and /or comparisons of its length flavipennis group, using the oldest valid witn the length of one of the hindtarso- species name. meres to separate females of some species. Females of the flavipennis group share However, the separation of females of /la- the following defining characters: clypeal vipennis, /unerarius, and oxianus is partic- disk conspicuously convex, but abruptly ularly vexing, and association with males depressed near free margin, delimiting a may often be the only reliable way to narrow, flat rim; labral apex with inverted identify these species. The most useful Y- or V-shaped carina whose arms project male character is the number and width as small lobes; base of Y or V often ex- of placoids on the flagellum (although tending as a median carina toward labral they are variable in /umrarius especially; base. In addition to these apparently de- see that spedes bdow for details). other rived characters, the species lack certain male features that have been used by pre. specializations of other Sphex: the metan- vious WQrkers are ^ ]ength rf tRe lateml otum is simple (not bituberculate); the ^ bmshes of sternum VII, the form of propodeal dorsum lacks coarse, widely .11 1 c • j 1 u. r r J the clypeal tree margin, and color pattern. spaced, transverse carinae or wrinkles; ^ n i_ ,.i_ u r , / ,,,,, . , Generally, however, these characters are and the propodeal side lacks a vertical , „ ,. , , ,, , .. ,. , , ., „. . . , . , not wholly reliable. Male genitalia lack the swelling anterior to the spiracular groove. 1,1 . . e j • . , , .1 1 i r . t ■ unusual elaborations found in some sec- Males apparently lack features of species . , , . ..... . r .,. J r tions of the genus and appear identical in group significance. . . , ° _ . . rr _- Ti .u r 11 n 1 species of the flavivennis group. The penis the group contains the following Pale- r, , , J r t° \ .,r , ... . 1 r, 1, 100c a valve head is arcuate in lateral profile and arctic species: atropilosus Kohl 1885, /lain- r pennis Fabricius 1793, /unerarius Gussa- is aymed ventrallY with a row of teeth. kovskij 1934 (= maxillosus Fabricius 1793, The lowing authors provide keys, re- nec Poiret 1787; ru/ocinctus of authors after cords' and other valuable information on 1975), leuconotus Brulle 1833 (= a/er Lepe- sPecies in the flavipennis group: Kohl letier 1845), libycus de Beaumont 1956, me- (1890)' worldwide revision of genus; Dus- las Gussakovskij 1930, and oxianus Gus- met and Mercet (1906), Spanish species; sakovskij 1928. Roth (1925), North African species; Ber- In addition to the seven species listed land and Bernard (1949), French species; above we have studied three males and de Beaumont (1951), Moroccan species, one female from Morocco collected at Ait- and (1960), placoids of male antenna; Sco- Saouin (between Ouarzazate and Agdz) biola (1960), Romanian species; Kazenas southwest of the Jbel Sarhro Mts. (CAS) (1978), Kazakh species; Pulawski (1978), that may be a new species or an extreme species of the European part of former form of one of the currently known spe- USSR; Mingo and Gayubo (1984), Spanish cies. They resemble flavipennis and funer- species; Hamon, Fonfria, and Tussac arius, but the male antenna has broad pla- (1991), French species; and Bitsch et al. coids on flagellomeres IV- VI, and the fla- (1997), western European species. Volume 9, Number 2, 2000 327 KEY TO SPECIES OF THE SPHEX FLAVIPENNIS GROUP Females 1. Erect setae of head and thorax dark brown or black 2 - Erect setae of head and thorax white or pale yellow 4 2. Gaster red and black (at least tergum II partly red); foretibia(at least partly) and tarsus amber, midtibia partly amber; wings yellow tinted with dark apical band, veins brown; Iberian Peninsula to Greece, Slovakia, Hungary, Romania, Kazakhstan; Algeria(?) atropilosus Kohl - Body all black; wings darkly infumate, with no apical band 3 3. Appressed setae of face brown; hindfemur asetose; Iran, Turkmenistan melas Gussakovskij - Appressed setae of face silver; hindfemur with erect setae; northeastern Libya, northwest- ern Egypt libycus de Beaumont 4. Petiole wider than long (as measured dorsally from base of funicle to base of tergum I) or at least not longer than wide; Mediterranean region to central Asia leuconotus Brulle - Petiole longer than wide 5 5. Legs black (including tarsi); gaster usually all red (sometimes bicolored or all black); east- ern Mediterranean region to Afghanistan and Tajikistan oxianus Gussakovskij - Some or all legs partly to entirely red (at least foretibia with some red apically); gaster bicolored (all black in some areas of Kazakhstan, Siberia, China); widespread in Palearctic Region 6 6. Pronotal collar and scutum with appressed white setae (best seen from in front); length 24-32 mm flavipennis Fabricius - Pronotal collar and scutum without appressed white setae (scutum sometimes with nar- row, median stripe of white setae or traces of white setae posterolaterally); length 16-26 mm funerarius Gussakovskij Males 1. Flagellomere III with placoid 2 - Flagellomere III without placoid 4 2. Erect setae of head and thorax dark brown or black; flagellomeres III-VII with placoids; at least tergum II partly red; Iberian Peninsula to Greece, Slovakia, Romania, Hungary, Ka- zakhstan; Algeria(?) atropilosus Kohl - Erect setae of head and thorax white or pale yellow; placoid distribution and gaster color variable 3 3. Gaster red basally, black distally (western Palearctic; Sichuan, China), or all black (e. Ka- zakhstan, Siberia; Gansu, China); placoids usually present on flagellomeres III-VIII (Fig. 2), but sometimes on II-VIII, II-IX, III-VI (Fig. 1), III-IX (or IV- VI in Corsica) funerarius Gussakovskij - Gaster all black; placoids present on flagellomeres III-VI (Fig. 3); eastern Mediterranean region to Tajikistan and Afghanistan oxianus Gussakovskij 4. Wings nearly uniformly darkly infumate, with no apical dark band; remaining body black; erect setae of head and thorax dark brown or black 5 - Wings nearly hyaline or yellowish to slightly infumate, with darker apical band; gaster at least partly red; legs black or partly red; erect setae of head and thorax dark brown to pale yellow or white 6 5. Iran, Turkmenistan melas Gussakovskij - Northeastern Libya, northwestern Egypt libycus de Beaumont 6. Flagellomeres V-VI (Figs. 6-7) or only VI with placoids; gaster color variable flavipennis Fabricius - Flagellomeres IV- VI with placoids (Figs. 4-5) (Corsican funerarius will key here, but gaster is red and black); gaster black leuconotus Brulle 328 Journal of Hymenoptera Research Volume 9, Number 2, 2000 329 Figs. 4-9. Scanning electron photographs of male antenna showing placoids on flagellomeres (= F). 4-5, Sphex leuamotus, 4 is specimen from Cherkes, Cvprus with placoids on F IY-YI; 5 is specimen from Zaragoza, Spain with placoids on F rV-VI. 6-7, Sphex flauipennis, 6 is specimen from Carpentras, France with placoids on F Y-VI; 7 is specimen from Zaragoza, Spain with placoids on F V-VL 8, Sphex melas from Repetek, Turk- menistan with placoids on F V-VI. 9, Sphex libycus from Marsa Matruh, Egypt with placoids on F V-YI. DISCUSSION OF SPECIES Sphex atropilosus Kohl Sphex maxUlosus var. atropilosus Kohl 1885:202. Holotype: female, "Tultscha" [= Tulcea, Ro- mania] (VVien), examined. Sphex. atrohirtus Kohl 1890:437 (lapsus for and redescription of atropilosus, raised to species.) Subsequent records as atropilosus: Berland 1952: 88 (France); Leclercq 1955:19 (Africa); Le- clercq 1956:324 (Greece); Bajari 1957:79 (Hun- gary); de Beaumont 1962:19 (Spain); de Beau- mont 1965:14 (Greece); Pulawski 1978:183 (s. Russia, Caucasus); Mingo and Gayubo 1984: 145 (Spain); lozan 1986:367 (Hungary); Ga- yubo 1987:106 (Spain); Hamon, Fonfria, and Tussac 1991:131 (France); Bitsch et al., 1997: 69 (s. France); Shkuratov 1998:97 (Rostov Ob- last', Russia). Subsequent records as atrohirtus: Roth 1925:397 (Africa); Chaudoir 1947:142 (France); Zavalil and Snoflak 1948:168 (Czechoslovakia); de Andrade 1949:8 (Portugal); Berland and Ber- nard 1949:4 (France); Hamon 1950:29 (France); Benedek 1968:70 (Hungary); Bal- thasar 1972:421 (Czechoslovakia); Kazenas 1978:40 (Kazakhstan); Dollfuss 1989:12, 15 (type material); Padr (in Sedivy) 1989:166 (Slovakia). Recognition. — The dark erect setae of the head and thorax distinguish atropilosus from other species in the fUwipennis group with red and black legs and gaster. The short female petiole (at most minimallv longer than wide) is similar to that of leu- amotus, but the erect setae are pale in that species. The female mid- and hindfemora have erect setae but thev are shorter on the dorsum than in leuamotus. The broad Figs. 1-3. Scanning electron photographs of male antenna showing placoids on flagellomeres ( F). 1-2, Sphex funerarius, 1 is specimen from Tanger, Morocco with placoids on F III-Y1 2 is specimen from Italy with placoids on F III-YII1. 3. Sphex oxianus from Kondara Canyon, Tadjikistan with placoids on h III-YI 330 Journal of Hymenoptera Research placoids on male flagellomeres III- VII dif- ferentiate atropilosus from leuconotus which has narrower ones on IV- VI (Figs. 4-5). The size range of female atropilosus is 18- 27 mm which is somewhat less than the common parameters for females of leucon- otus (22-33 mm). Material examined. — Kohl's holotype bears his handwritten label "atrohirtus Kohl Type", and he obviously intended the species name to be atrohirtus, but it was published as atropilosus. A second fe- male from "Transcauc." also has Kohl's label "atrohirtus Kohl type" (Vienna), but it was not mentioned in the original de- scription and cannot be a type. It was list- ed by Kohl (1890) who also had material from Sarepta (= Volgograd, Russia). Other material examined: SPAIN: Vil- larina, Salamanca, July 21, 1995 (female, Menke), Valdepenas, June 21, 1983 (male, Menke), Madrid (female, CAS); FRANCE: Canet, June 14, 1948 (female, Menke), St. Nazaire, June 14, 1948 (male, Menke); GREECE: Kalamata (one of a series stud- ied by de Beaumont 1965) (male, CAS); RUSSIA, Kalmyk Republic: 10 km NW of Chernozemel'sk (female, CAS); and AL- GERIA: "Oran, 1895", collected by Schmiedeknecht (three females, one male, Vienna) and identified by Kohl as "atro- hirtus" (i.e., atropilosus). However, these Algerian specimens may have incorrect provenance labels (see Distribution be- low). Holoh/pe features. — Kohl's holotype has yellow tinted wings. The legs are largely black but the following are reddish: ante- rior face of forefemur, and all of foretibia and tarsus; distal fourth of anterior face of midfemur, and anterior face of midtibia. Tergum I is red but there is a large, cir- cular black spot on the anterior face. Ter- gum II is entirely red, and III is red lat- erally. Sterna I-III are red (except petiole is black). The specimen is 21 mm long. Variation. — Petiole length and color of the gaster and legs vary in atropilosus. Fe- male petiole length varies from 0.9 X to 1.1 X its apical width, but it is usually slightly more than the distal width (ex- ample ratios are 18:17, 19.5:17, 21:20). Sometimes the dimensions are equal. In the largest female studied, a specimen 27 mm long labeled Oran, the petiole is slightly shorter than its width (ratio 21:24). In smaller specimens from Oran the peti- ole is as long as wide. Increasing body size thus may be correlated with a shortening of the petiole. The male petiole is longer than wide (20.5:17.5, Oran; 20:14, Greece). Gaster color in the female from Cher- nozemel'sk is similar to the holotype, but the red is limited to tergum I apically and laterally and tergum III basolaterally. Red is reduced in the Spanish females: narrow strip along distal margin of tergum I, ter- gum II laterally, and most of sternum II. The Spanish male is similar, but addition- ally has red laterally at the extreme base of tergum III (M. Ohl in litt. to Menke says tergum III is all black in Spanish males he has studied). A female and male from France are similar, but the red is more ex- tensive in the female: tergum II all red, and III red basolaterally. In the Greek male red covers the sides of tergum I, all of II, tergum III basolaterally, all of ster- num II, and the basal half or so of sternum III. Legs are bicolored in females, but the amount of black and red varies. The fe- male from Russia is similar to the holo- type, but the posterior surface of the for- etibia is black. The midleg of one female from Oran is black except for a small, cir- cular amber spot at the femoral apex, and the hind surfaces of the foreleg are dark. In the Spanish females red is limited to the apex of the fore- and mid femora anteri- orly, and to the anterior surfaces of the fore- and midtibiae. The legs are all black in the French and Greek males. Females are 18-27 mm, and males are 17-22 mm long. Distribution. — Northern Mediterranean region (except Italy) eastward to Kazakhs- Volume 9, Number 2, 2000 331 tan. The species is uncommonly collected, worn away. In such cases association with although locally abundant at times. males may offer the only reliable means of Specimens mentioned above labeled identification (note however, that if long, Oran, Algeria, may have been mislabeled erect setae are present but appressed setae since the species has not been collected in are absent, it is likely that a specimen is that country by modern workers (e.g., de not flavipennis). Females of flavipennis tend Beaumont, Guichard, Roth); nor did col- to be larger than those oifunerarius (24—32 lectors of the 1890's find it (Saunders 1910, mm long versus 16-26 mm). Some females Morice 1911). The only other record from of Sphex leuconotus have appressed white Algeria is by Roth (1925) who saw a fe- setae on the thoracic dorsum but the pet- male from Orleansville (now El Asnama) iole is shorter than wide or at most as long dated 1867 in the Sichel Collection (Paris), as wide. Females of flavipennis have a pet- iole that is longer than wide. Sphex flavipennis Fabricius The appressed pale setae are less devel_ (rigs. 6-/) oped in males, but placoids are found Sphex flavipennis Fabricius 1793:201. Lectotype: only on flagellomeres V-VI (rarely only on female, "Italia" (Copenhagen), designated by VI), and they are narrow (Figs. 6-7). Infu- van der Vecht 1961:31, not examined. nerarius, placoids are broad (Figs. 1-2) and Sphex rufocinctus Brulle 1833:367. Holotype (or usually present on flagellomeres III-VIII syntype): male, "Petalidi, Moree" [= Korone (but see that species for placoid variation). or Koroni, Peloponnesus, Greece] (Paris), ex- Males of leuconotus have placoids on fla- amined. New synonym. gellomeres IV- VI (Figs. 4-5). Sphex bicolor Dahlbom 1845:437. Holotype: Females are 24-32 mm, and males are male, "Dalmatia" [= coastal Croatia and 17_26 mm lone Monteneerol (Berlin?), (junior primary horn- r> • i u j a. u ,~ ,J . . , ' v, . . K,„„r., - Previous workers have used other char- onvm of Sphex bicolor Fabricius 1775), not ex- ■■,.<■ n ■ i i , c . ix ui -loo-. ->r> i. acters to identifv flavipennis, but we have amined. Svnonvmv bv Kohl 1881:39 who saw , , , Dahlbom's material (with maxillosus), and found them unreliable: yellow-tinted Kohl 1890:236 (with flavipennis). wings and golden erect setae on the face. Sphex cinereorufocinctus Dahlbom 1845:437. Svn- The erect facial setae are sometimes silver types: male, "Rhodus" [= Rhodes, Greece] in flavipennis, and nearly always this color (Lund), not examined. Synonymy by de in funerarius. The yellow tinted wings, Beaumont 1949:127 who saw Dahlbom's ma- from which the species derives it name, is terial. not a reliable recognition character for fla- Sphex sellae Gribodo 1873:86. Holotype: female, vipennis because some funerarius also have "Sicilia" [= Sicily, Italy] (Genova?), not ex- yellowish wings, although typically they amined. Synonymy by Kohl 1890:236. are hghtlv brown stained. Several authors Sphex flavipennis var. rufodorsatus De-Stefani have usecj comparative lengths of the pet- 1887:88, pi. 2, fig. 8. Holotype: female, "Sici- iole and hindtarsomere I or III to distin- lia" [= Sicily, Italy] (Palermo?, Geneva? de- jsh females of flavi ms from finwrar. stroyed?), not examined. Svnonvmv by Kohl • T. • i i .• i i '.i 1890-716 lus' 1S unc^ear h°w petiole length was measured by these workers, but we have Recognition. — The presence in most fe- been unable to find any useful differences, males of appressed white setae on the Comparisons of the length and width of pronotal collar and scutum identifies fla- the female petiole itself show nearlv the vipennis and separates the species from the same parameters in both species (15 spec- similar funerarius which lacks such pubes- imens of each species measured): flavipen- cence. Unfortunately the appressed setae nis — petiole length is 1.1 to 1.4 X width; fu- are poorly developed in some popula- nerarius — petiole length is 1.1 to 1.6X tions, and in older material they are often length. 332 Journal of Hymenoptera Research Hamon, Fonfria, and Tussac (1991) il- lustrated two male characters that earlier authors had used to separate flavipennis and fun erarius: the form of the clypeal free margin and the length of the lateral setal brushes of sternum VII. These characters are useful, but they are not always reli- able. The clypeal free margin in flavipennis is usually a simple arc, but sometimes there is a shallow emargination that may be broad or narrow. In funerarius the clyp- eal free margin usually has a pronounced emargination (see Fig. 10 in Hamon et al. 1991). The lateral setal brushes of male sternum VII are shorter in flavipennis in comparison to funerarius (see figs. 11-12 in Hamon et al. 1991), but the reliability of this character remains to be proven. Type material. — Brulle's holotype (or sole surviving syntype) of rufocinctus, a male, has been studied by each of us on two occasions. The specimen is poorly preserved, very dirty, and badly worn. The mandibles are truncated from consid- erable use, and the setation of the clypeal disk is worn away. The type has lost most of its antennae; the scapes and pedicels re- main, and the right antenna still has fla- gellomeres I-II. Although the pronotum is dirty, some silver appressed setae (one of the characteristics of flavipennis) are faintly visible. Traces of appressed silver setae are also visible in the scutal furrows and on the hindmargin of the scutellum. The clypeal margin is not emarginate although it is straight at the middle. The lateral setal brushes on sternum VII are dirty, but the setae are as short as in flavipennis (shorter than in funerarius). The body length of 25 mm is also typical of male flavipennis. The gaster is mostly black, but tergum II is red except for a narrow, transverse dark band along the distal margin, and tergum III is red laterally. This color pattern is typical of flavipennis. These characters convince us that the type of rufocinctus is not conspe- cific with funerarius (maxillosus of authors), but instead is flavipennis Fabricius. The bi- colored gaster indicates that rufocinctus is not conspecific with leuconotus. We have been unsuccessful in locating the type material of bicolor Dahlbom, sellae Gribodo, and rufodorsatus De-Stefani, and have relied on Kohl (1890) for their syn- onymy. Variation. — Occasionally the thoracic dorsum of female flavipennis has reddish areas, a trait that immediately identifies Mediterranean specimens as this species (similar reddish areas are found on some central Asian specimens of funerarius). De Beaumont and Bytinski-Salz (1955) noted that occasional Israeli females have an en- tirely red thorax, but they occur with nor- mally colored specimens. The thorax of some females in Iran is also extensively red (de Beaumont 1957). De Beaumont (1960) noted a male from Cyprus with a placoid only on flagellomere VI. In a male of flavipennis from the Mashad area, Iran (CAS), the lateral setal brushes of SVII are longer than a midocellus di- ameter. Four other males from Mashad have typically short brushes. Distribution. — Mediterranean region in- cluding islands of Mallorca, Sardinia, Si- cily, Crete, Rhodes and Cyprus; eastward to Hungary, Bulgaria, and south-central Asia (Kazakhstan, Uzbekistan, Tajikistan, Turkmenistan); Arabia: United Arab Emir- ates; Iran, Afghanistan. Sphex funerarius Gussakovskij (= Sphex maxillosus of authors before 1976, or rufocinctus of authors after 1975) (Figs. 1-2) Sphex maxillosus Fabricius 1793:208. Lectotype: female, "Barbaria" [= northwest Africa] (Co- penhagen), designated by van der Vecht 1961:30, (junior primary homonym of Sphex maxillosus Poiret 1787). Sphex obscurus Fischer de Waldheim 1843:122. Syntypes: male, "in Rossia australi" (Mos- cow, St. Petersburg, Dresden?), (junior pri- mary homonym of Sphex obscurus Schrank 1802, and Sphex obscurus Fabricius 1804). Syn- onymy with maxillosus by Kohl 1895:69. New synonym. Volume 9, Number 2, 2000 333 Sphex maxillosus var. pedibus nigris Zanon 1925: 90. Holotype: female, Fueihat, Libya (Geno- va?) . Polynomial, not available, see Art. 11.4 of the Code. Sphex funerarius Gussakovskij 1934:3. Lectotype: male, [Bei-lung-shui,] S. Kansu [= Gansu], China (Stockholm), here designated in order to ensure the name's proper and consistent application, examined. New synonym. Sphex maxillosus ssp. mavromoustakisi de Beau- mont 1947:383. Holotype: female, Polemidia Hills, Cyprus (Lausanne), examined. New synonym. Sphex rufocinctus (misinterpretations since 1975): Lomholdt 1975:68 (Gotland I., Swe- den); Bohart and Menke 1976:116 (listed, no- menclature problems); Guichard 1978:270 (Greece); Richards 1979:400 (British Channel Is.); Pagliano 1980:110 (Italy); Dollfuss 1983:2 (Austria); Mingo and Gayubo 1984:146 (Spain); Schmidt and Westrich 1983:120 (Greece); Gayubo 1984:356 (Portugal); Gayu- bo and Tormos 1984:8 (Spain); Pagliano 1984: 367 (Italy); Chevin and Chevin 1985:38 (France); Eiroa and Novoa 1985:23 (Spain); Jozan 1985:55 (Hungary), 76 (floral records), 83 (ecological and zoogeographic character- istics); Pagliano 1985:12 (Italy); Tormos and Jimenez 1985:32 (Spain); Westrich and Schmidt 1985b:112 (Germany: endangered in Baden-Wurttemberg); Gayubo 1986a:35 (Spain), 1986b:30 (Spain); Gayubo and Heras 1986:26 (Spain); Gayubo and Sanza 1986:27 (Spain); Gayubo and Tormos 1986a:8 (Spain), 1986b:4 (Spain); Islamov 1986:515 (Uzbeki- stan); Asis and Jimenez 1987:23 (Spain); Ga- yubo 1987:106 (Spain); Tormos and Jimenez 1987a:122 (Spain), 1987b:316 (Spain); Anders- son et al. 1987:72 (endangered in Sweden); Dollfuss 1987:18 (latest Austrian specimens collected in 1952 and 1953); Schmidt and Westrich 1987:358 (Germany); Chevin 1988: 14 (France); Dollfuss 1988:20 (Austria); Jan- zon 1988:1 (Sweden); Karsai 1988:99 (Hun- gary); Islamov 1989:49 (Uzbekistan: nest and prey); Jacobs 1989:3 (Germany); Jozan 1989: 100 (Hungary); Asis, Gayubo, and Tormos 1990:240 (mature larva); Gayubo, Asis, and Tormos 1990:9 (Spain); Jacobs and Oehlke 1990:122, 132 (German Democratic Republic: not collected after 1960); Pagliano 1990:60 (in catalogue of Italian Sphecidae); Day 1991 :xix (summary of European Endangered Hyme- noptera Lists); Dollfuss 1991:27 (in revision of Austrian Sphecidae); Gayubo, Borsato, and Osella 1991:392 (Italy); Gayubo and Torres 1991:Table I and p. 81 (Spain: effects of urban pressure); Hamon, Fonfria, and Tussac 1991: 128, 131 (in key to French Sphecini), 132 (France); Jozan 1991:602 (Hungary); Kazenas and Nasyrova 1991:38 (Kazakhstan); Negri- solo 1991:316 (Italy); Schembri 1991:177 (pre- vious records from Malta); Gayubo, Borsato, and Osella 1992:275 (Greece, Turkey); Jozan 1992:171 (Hungary); Kazenas and Tobias 1992:29 (sleeping aggregations); Gayubo, Tormos, and Asis 1993:308 (teratological specimen); Torregrosa, Gayubo, Tormos, and Asis 1993:11 (Spain); Luchetti 1993:10 (Italy: Sardegna); Dollfuss 1994:98 (endangered in Austria); Gayubo and Borsato 1994:199 (Ita- ly); Tormos, Asis, and Gayubo 1994:187 (Spain); Jozan 1995:104 (Hungary); Krasno- bayev et al. 1995:139 (Russia: Samara Ob- last'); Negrisolo 1995:18, 22 (floral records); Pagliano and Pesarini 1995:83 (Italy); Pagli- ano and Scaramozzino 1995:731 (Italy); Schmid-Egger, Risch and Niehuis 1995:208 (Germany); Vernier 1995:176 (in key); Gus- enleitner 1996a:809 (Austria), 1996b:818 (Cro- atia); Minoranskiy and Shkuratov 1996:81 (Russia: Rostov Oblast'); Schmid-Egger 1996: 19 (Germany); Schmid-Egger, Schmidt, and Doczkal 1996:374, 378 (Germany: endan- gered); Voblenko, Gorobchishin and Nester- ov 1996:14 (Ukraine); Bitsch et al. 1997:71 (in sphecid fauna of western Europe); Schmidt and Schmid-Egger 1997:27 (in checklist of German Sphecidae); Dollfuss, Gusenleitner, and Bregant 1998:509 (Austria); Gonzalez, Gayubo, and Torres 1998:72, 73 (Spain); Zehnder and Zettel 1999:13 (Switzerland); Gonzalez, Gayubo, and Torres 1999:334. Recognition. — Sphex funerarius varies over its extensive range, both in female and male color and number of male an- tennal placoids. Some females especiallv can be difficult to separate from oxianus and flavipennis. Males are easiest to iden- tify because most specimens have broad placoids on flagellomeres III-VIII (Fig. 2) (males from Corsica are among the more notable exceptions — see Variation below). Only two other species have a placoid on 334 Journal of Hymenoptera Research III: atropilosus and oxianus. However, the erect setae on the head and thorax are white or pale yellow in male fiinerarius, while they are dark brown or black in atro- pilosus. Most commonly males of fiinerar- ius have a bicolored gaster in contrast to the all black gaster of oxianus. Males of fii- nerarius from China can have a black gas- ter, but the presence of placoids on flagel- lomeres III- VIII will separate them from oxianus which has placoids only on III-VI (Fig. 3) and is unknown from China. The following are variable in the species, but may aid in identification. The male clypeal free margin usually has a pro- nounced broad emargination in fiinerarius, and is arcuate or shallowly emarginate in flavipennis. However, exceptions in both species weaken the usefulness of this dif- ference. The lateral setal brushess on male sternum VII are longer than a midocellus diameter in fiinerarius, while in other spe- cies the these are usually shorter than, or about equal to a midocellus diameter. However, there are also exceptions to this character (see Variation under flavipennis). The absence of appressed pale setae on the pronotal collar and scutum separates females of fiinerarius from flavipennis. Fe- males of flavipennis typically have ap- pressed silver setae on the thorax, but this pubescence is often sparse or worn away. Thus, association with males is often the best way to identify females of both spe- cies. As noted under flavipennis, female petiole length is essentially identical be- tween the two species. Females of fiiner- arius usually have bicolored fore- and midlegs, and the hindleg, except for the tarsus, is usually black. However, red is limited to the foretibial apex in some me- lanic Chinese specimens. A similar leg pattern is present in one Iranian female seen by us, but, unlike the black Chinese specimens, the gaster is bicolored. Females of oxianus have entirely black legs, includ- ing the tarsi. Females are 16-26 mm, and males are 17-20 mm long. Nomenclatural history. — This relatively common Palearctic species was known as Sphex maxillosus Fabricius 1793, for more than 170 years. Then van der Vecht (1959: 214) pointed out that Fabricius' name was a junior homonym of Sphex maxillosus Poiret 1787, currently assigned to the ge- nus Chlorion. He noted that unless Poiret's name was suppressed, maxillosus Fabricius would have to be replaced by the next available synonym. Van der Vecht be- lieved that to be leuconotus Brulle based upon Dalla Torre's (1897) catalog, but that is not true. Bohart and Menke (1976:116) called the species rufocinctus Brulle 1833, adopting the first of several synonyms of maxillosus recognized by Kohl (1890:433), even though Menke pointed out (footnote 23) that the type of rufocinctus appeared to be a synonym of a different species, flavi- pennis Fabricius. Menke urged European workers to study the problem and clarify the identity of Brulle's species. That action has not been forthcoming, and, unfortu- nately, nearly all contemporary authors since 1976 have used rufocinctus as the proper name for the invalid maxillosus Fa- bricius. Our studies of the Brulle types es- tablishes that none of his names, rufocinc- tus, leuconotus, or triangulum, are available for maxillosus Fabricius. Instead, the oldest available name is fiinerarius Gussakovskij 1934. Type material. — Gussakovskij (1934) list- ed 15 males and 3 females of fiinerarius from Bei-lung-shi, S. Kansu, China, 15- VI- 30, and 2 males from N. O. Szechuan, Chi- na, 20-V-30, all collected by Dr. Hummel, and also 3 males and a female from south- ern Altai (Karasengir) [= Siberia], and one male from Saissan [= Lake Zaisan, Ka- zakhstan]. We have examined three male and six female syntypes (Stockholm). All but one male are labeled Kina [= China], S. Kansu, Sven Hedins Exp. Ctr. Asien, Dr. Hummel, and one male and one female from Kansu have Gussakovskij's species labels with the word "typus". One male is labeled Kina, N. O. Szechuan, etc., and Volume 9, Number 2, 2000 335 it also has a Gussakovskij label. We have ical color pattern of the species. Interme- selected and labeled as lectotype the male diate leg color patterns are seen in four from S. Kansu with Gussakovskij's "ty- females from Iran (CAS): the foretibia is pus" label. We have also put paralectoty- black and red, and the midleg is all black pe labels on the male and female that have except for an inner distal red spot on the Gussakovskij labels. The three males have femur, and the tarsi are sometimes brown- placoids on flagellomeres III-VIII and are ish (Elburz Mts.); or the foretibia is black entirely black. The females are black ex- and red and the midtibia is reddish on its cept the apical half of the foretibia which anterior surface only, and the midtarsus is is brownish red. In one female the whole brownish (Khorasan). Another Iranian fe- outer side of the foretibia is reddish. In male from Tilabad (USNM) has entirely two females the inner apex of the forefe- black legs except for red on the anterior mur is also reddish. surface of the foretibia. Variation in placoid distribution. — In a se- In females of most populations of funer- ries of papers de Beaumont (1960:229; arius, the gaster is red with a black apex 1961a:272; 1961b:45; 1967:276; 1970b:4) (segments IV-VI or V-VI typically). How- studied variation in the number of pla- ever, the gaster may be all black in females coids in western Palearctic males. He re- from eastern Kazakhstan, southern Sibe- corded their distribution as follows: on ria, and the provinces of Sichuan and Gan- flagellomeres III-VIII (most European su in China (sternum II pale laterally in specimens, Fig. 2), but occasionally II-VIII some specimens from Gansu). The gaster or III-VII; II-VIII or II-IX (Turkey); II-VIII is also all black in males from Siberia and (Crete, Iran); II-IX (Afghanistan); II-VIII or Gansu, China. In females from Cyprus, II-IX (Cyprus); and III-VI (Fig. 1), III-VII, the gaster (except the black petiole), fem- or III-VIII (north Africa). Males from the ora and tibiae are red. This insular popu- island of Corsica are exceptional in lacking lation was described as maxillosus mavrotn- placoids on flagellomeres II and III (de oustakisi by de Beaumont (1947). We have Beaumont 1960:229, also one male in examined the type series of mavromoustak- CAS). On this island placoids are only isi (2 females, 3 males, from Cyprus), and found on flagellomeres IV-VI, and we sus- do not feel that recognizing the subspecies pect this to be the typical condition on that is warranted. The Corsican population, for island. example, is as distinct based upon male Color variation and subspecies. — Female antennal placoid distribution, and yet we legs are bicolored throughout the range of also feel it needs no name. funerarius, although the hindleg is often The thorax is typically all black in fu- largely black. The following are common- nerarius, but it is partly red in a female ly red: tibiae and tarsi of the fore- and from the Kopet-Dagh Mountains south of midlegs and the hindtarsi (the midtibia Askhabad, Turkmenistan (CAS). The fol- varies from all red to partially or wholly lowing are red in this specimen: pronotum black). Sometimes the fore- and mid-fern- apicomesally and laterally, most of scu- ora have red areas distally. However, the turn, scutellum, postscutellum and part of legs are almost wholly black in females propodeal dorsum. Additionally only the from Gansu, China (type series of funer- last gastral segment (VI) is black in this arius) except for some reddish brown on specimen, and the inner side of the hind- the forefemoral apex and inner side of the tibia is reddish. foretibia. In one specimen of this series red Larva. — The mature larva of funerarius occurs only on the inner apical half of the was described by Asis, Gayubo, and Tor- foretibia. Specimens that we have seen mos (1990). from other parts of China display the typ- Conservation. — Sphex funerarius was list- 336 Journal of Hymenoptera Research ed as endangered in Austria, Germany, and Sweden (summary in Day 1991, also Andersson et al. 1987, and Janzon 1988). Dollfuss (1987) reported that the last Aus- trian specimens were collected in 1952 and 1953, and Jacobs and Oehlke (1990) noted that no specimens were collected in the former German Democratic Republic after 1960. One female, however, was found in Austria in 1996 (Gusenleitner 1996a). Schmid-Egger, Risch, and Niehuis (1995) reported that the species was relatively common in the upper Rhine area of Ger- many in the 1950's and early 1960's, but disappeared subsequently. However, sightings of fiinerarius increased in the states of Hessen, Baden- Wurttemberg, and Rheinland-Palatinate. M. Ohl in lift, to Menke says the species has also been found recently in Bavaria. Janzon (1988) noted that on the Swedish island of Got- land funerarius occurs in open sandy areas, often along the shore. Habitat overgrowth is the main danger to populations of fu- nerarius on this Baltic island. Distribution. — Mediterranean Basin in- cluding North Africa from Morocco to Egypt. The Balearic Islands (Compte Sart 1959), as well as the islands of Corsica, Sardinia, Sicily, Malta, Crete, and Cyprus. Recorded from Jersey in the Channel Is. (Richards 1979), but the species does not occur on the British Isles proper. Most of Europe north to Gotland I. and Faro I. in the Baltic Sea but not in Sweden proper (Andersson et al. 1987) and Poland; east- ward to Hungary, Romania, Bulgaria, s. Russia, Siberia (Krasnoyarsk), and south- central Asia: Iran, Afghanistan, Uzbeki- stan, Kazakhstan, Turkmenistan, Tajikis- tan; n. China (Gansu, Sichuan, Nei Mon- gol, Shandong, Liaoning). Sphex leuconotus Brulle, new status (Sphex afer of authors) (Figs. 4-5) Sphex leuconotus Brulle 1833:366. Holotype: fe- male, "Petalidi, Moree" [= Korone or Ko- roni, Peloponnesus}, Greece, (originally Par- is, now lost). Neotype: holotype of Sphex triangulum Brulle, present designation in or- der to ensure the name's proper and consis- tent use (Paris), examined. Sphex triangulum Brulle 1833:365, pi. 50, fig. 6. Holotype: female, "Petalidi, Moree" [= Ko- rone or Koroni, Peloponnesus], "a la fin de Mai", Greece (Paris), (junior primary hom- onym of Sphex triangulum Villers 1789), ex- amined. New synonym. Sphex afer Lepeletier 1845:350. Lectotype: fe- male, Oran, Algeria, designated by Menke in Bohart and Menke 1976:114 (Paris). New syn- onym. Sphex sordidus Dahlbom 1845:436. Syntypes: sex unknown, "Rhodus" [= Rhodes, Greece] (Stockholm?). Regarded as subspecies of afer by de Beaumont 1953:195. New synonym. Sphex tristis Kohl 1885:200. Syntypes: male, Spain (Wien). Synonymy with afer ssp. sordi- dus by de Beaumont 1953:105. New syno- nym. Sphex plumipes Radoszkowski 1886:25, figs. 18a- i. Holotype: male, Askhabad [Turkmenistan] (Krakow), (junior primary homonym of Sphex plumipes Drury 1773), examined. New synonym. Sphex pachysoma Kohl 1890:436. Syntypes: fe- male, "Kilasi" and "Kuba Breku" [= Kilyazi and Kuba, Azerbaijan?]; Cyprus; "Syra" [ = Syros I., Greece] (Wien). Synonymy with tris- tis by Dusmet and Mercet 1906:516, with afer by Schulz 1911:68 and with afer ssp. sordidus by de Beaumont 1953:195. New synonym. Recognition. — The female and most males of leuconotus can be recognized by an unusually short petiole whose distal width is usually greater than the length. In occasional specimens the measure- ments are equal. In 12 females measured, the petiole width varied from 1.0X to 1.5X its length, with the average being 1.2 X. Some females have appressed white setae on top of the pronotal collar. Traces of ap- pressed white setae can also sometimes be seen on the scutum. In the male, narrow placoids are present on flagellomeres IV- VI (Figs. 4-5). Females are 25-34 mm, and males are 17-23 mm long. Type material and synonymy. — Kohl Volume 9, Number 2, 2000 337 (1890:433) synonymized triangulum Brulle dicate that his leuconotus was similar to Sa- with Sphex maxillosus Fabricius, and Bo- vigny's figure. In any case, his description hart and Menke (1976) listed it as a syno- of leuconotus agrees with afer sordidus. nym of rufocinctus Brulle. The holotype (or Since the differences between leuconotus the only surviving syntype) is actually and triangulum appear to represent age identical with Sphex afer sordidus Dahlbom. differences of the specimens, and Brulle's The specimen has the typical short petiole own feeling was that he would otherwise of afer (ratio 24:19). The mid and hindfe- regard them as identical, we are certain mora have long, erect setae on the upper, that leuconotus is conspecific with triangul- outer, and lower surfaces. Tergum I of the um. Accordingly we have made the type type is mostly black (apex red), and terga of triangulum the neotype of leuconotus IV- VI are also black. Terga II-III are red. and have so labeled it. The legs are black and the wings are yel- A result of the foregoing is that Sphex lowish. The name triangulum Brulle is a ju- afer sordidus must now be called leuconotus nior homonym. Brulle. According to de Beaumont (1953: The next available name for this species 195), tristis Kohl and pachysoma Kohl are is leuconotus Brulle which was collected at synonyms of sordidus. Thus they are new the same locality as triangulum. We were synonyms of leuconotus Brulle s.s. unable to find a type of leuconotus in Paris, We have also studied the male type of and apparently it has been destroyed, plumipes Radoszkowski. It is labeled "plu- Brulle stated that leuconotus and triangul- mosus" in Radoszkowski's handwriting, um were very similar, differing primarily and agrees with his description, but the in the form of the mandible and the seta- type lacks its antennae and the gastral tion of the clypeus. His description of the apex, presumably lost when he extracted mandible of triangulum suggests that he the genitalia (which are glued to a piece had a female with mandibular wear com- of card mounted on the pin with the spec- mon to old specimens, the result of much imen). Sphex plumipes was tentatively syn- nest excavation, and this is confirmed by onymized with pachysoma (i.e., leuconotus) examination of the holotype. The mandi- by Kohl (1890:436), whose interpretation is bles are badly worn. Concurrent with such most likely correct. The type is all black wear is the loss of clypeal setation, and with some appressed white setae on the Brulle described the clypeus of triangulum collar, almost silvery scutal setae, and as being largely devoid of setae. Brulle's slightly yellowish wings with a darker specimen of leuconotus, on the other hand, apical band. This color combination is must have been a fresh one, judging from found only in leuconotus and oxianus, but his description. The mandible was long, the holotype's relatively short gastral pet- and the clypeus was mostly covered with iole (length equal to 1.25X apical width) setae. In his description of leuconotus, Brul- suggests leuconotus. In any case, Radosz- le refered to plate 14, fig. 1, in Savigny's kowski's name is a junior homonym. (1809-1829) Description de I'Egypte. The fig- Color variation and subspecies. — Gaster ures on the plates from this work are re- color varies in females. Sometimes it is en- markable for their time, but they were tirely red (Cyprus, Uzbekistan), or tergum based on material from Egypt, and Sphex I may be black basally and IV-VI entirely afer is not known from that country. The so (Cyprus, Spain, n.w. Africa). In Roma- wasp shown in figure 1 has a short petiole nia the gaster may be all black (Scobiola although it is longer than wide. The ac- 1960), and de Beaumont (1960:227) stud- companying figure of the mandible shows ied a single female with a black gaster an unworn one with a long, acuminate from eastern Libya and four females with apex. Perhaps Brulle simply wanted to in- bicolored gasters from western Libya. 338 Journal of Hymenoptera Research Males of leuconotus from northwest Af- Taxonomic history. — Spkex libycus was rica have darker erect body setae than first described in detail by Roth (in Schul- specimens from the northern Mediterra- thess 1926:210) who studied one female nean and eastward, and the wings are from Agedabia in Cyrenaica, Libya, call- smoky. The wings are clear in the eastern ing it "Sphex maxillosus var. tota nigra, alis Libyan female mentioned above, and valde infumatis." He noted the all black those from western Libya are less smoky body (except for partly dark red mandi- than females from northwest Africa, bles), the dark body setation, and intense- Those who would like to treat the north- ly infumate wings. De Beaumont (1956) west African population as a subspecies recognized that Roth's wasp was a differ- (an action we do not endorse) will have to ent species and gave it the name libycus. call it leuconotus ssp. afer. Material examined. — We have seen three Distribution. — Mediterranean region (ex- females and four males collected 64 km W cept Egypt and known only from the Pyr- of Marsa Matruh, EGYPT, by Pulawski on enees-Orientales Departement in France) 28-29 May 1993 (CAS, USNM). including islands of Sardinia, Rhodes, and Distribution. — Known only from the Cyprus; eastward to Bulgaria, Greece, northeast coast of Africa: northeastern Turkey, Israel, Iraq, Iran, Afghanistan and Libya and northwestern Egypt, the central Asian republics of Kazakhstan, Turkmenistan and Uzbekistan. sPhex melas Gussakovskij (Fig. 8) Sphex libycus de Beaumont Sphex melas Gussakovskij 1930:207. Syntypes: (Fig. 9) male, female, Repetek, Turkmenistan (St. Pe- tersburg), not examined. Sphex libycus de Beaumont 1956:182. Holotype: „ ... n , , . , r, , „ r, j- tl /^- ■\^ i Recognition. — Sphex melas is nearly female, Porto Bardia, Libya (Zurich), not ex- ..,„.. • , • ,, amined. Subsequent record: de Beaumont unique in the /7ni;/^nms group in being all 1960:227 (Libya, Egypt). black' including the wings, legs, and all erect setae. Appressed silver setae are Recognition and status. — This species is found only on the face of the male. Sphex wholly black except for appressed silver libycus is the only similar species, but the facial setae in both sexes. The wings are brown appressed facial setae of female me- darkly infumate. The female hindfemur las separate it from libycus which has silver has erect setae, and the male has placoids facial setae. Males of the two species can- on flagellomeres V-VI (Fig. 9). Sphex liby- not presently be separated morphological- cus is almost identical to melas Gussakov- ly. Other characters of melas include: no skij, another all black species, and certain- silvery setae on pronotum, male with ly the males cannot presently be separat- broad placoids on flagellomeres V-VI (Fig. ed. The only difference in the females is 8, placoids narrower in flavipennis Figs. 6- the color of the appressed facial setae: sil- 7), and lateral setal brushes of sternum VII very in libycus and brown in melas. Erect markedly shorter than in funerarius. hindfemoral setae are present on the pos- Females are 22-28 mm, and males are teroventral edge in libycus, but in melas 15.5-17 mm long. they may be present (Iranian female) or Some male specimens of Sphex funerar- absent (Turkmen female). Much more ma- ius from the eastern part of its range (e. terial of both species is needed to deter- Kazakhstan, s. Siberia, China) are all mine if libycus is a valid species or merely black, as are males of oxianus and some a geographic color form of melas. males of leuconotus. However, the erect Females are 24-28 mm, and males are body setae are pale in these species, and 20-22 mm long. the wings are only slightly infumate. Volume 9, Number 2, 2000 339 Material examined. — We have examined topotypic material of melas (2 females, 4 males) determined by Gussakovskij (Mos- cow, CAS, USNM), and also a female from Hamadan, IRAN (CAS) and a female from Imam Baba, TURKMENISTAN (USNM). Distribution. — Turkmenistan and Iran. Sphex oxianns Gussakovskij (Fig- 3) Sphex oxianus Gussakovskij 1928:3. Syntypes: male, "Kara-tau mountains, right shore of Amu-Darya, below Khiva" [Uzbekistan] (St. Petersburg), not examined. Subseqent re- cords: Gussakovskij 1930:208 (description of female; Turkmenistan, Uzbekistan); Gussa- kovskij 1933:273 (Iran); Gussakovskij 1935: 413 (Tadjikistan); de Beaumont 1960:170 (Af- ghanistan); de Beaumont 1967:276 (Turkey); de Beaumont 1968:156 (redescription); de Beaumont 1969:81 (Turkey); de Beaumont 1970a:393 (Afghanistan); de Beaumont 1970b: 4 (Iran). Sphex oxianus form nubilus de Beaumont 1968: 156. Holotype: female, Ein Gedi, Israel (W. Schlaffle Collection, Kaiseraugst, Switzer- land), not examined. Bohart and Menke (1976:116) listed nubilus as a subspecies of ox- ianus thus validating de Beaumont's name — Article 45.6.4.1 of the Code, 4th edition (ICZN, 1999). New synonym. Recognition. — The entirely black legs, in- cluding the tarsi, separates oxianus from other species in the flavipennis group ex- cept melas and libycus which are similarly colored. However, oxianus has pale erect setae on the head and thorax, unlike the dark setation of melas and libycus. The fe- male petiole is longer (length 1.6-2.0 X width) than in the other species of the group. For example: 0.6-l.Ox apical width in leuconotus, 0.9-1.1 X apical width in atro- pilosus, 1.1-1.4X apical width in flavipen- nis, and 1.1-1.6X apical width in funerar- ius. In the male, broad placoids are present on flagellomeres III-VI (Fig. 3), the gaster is all black, and the lateral setal brushes of sternum VII are inconspicuous (shorter than in funerarius). Females of oxianus from the eastern Mediterranean with an all black gaster can be distinguished from the similarly col- ored melas by the pale erect body setae. Most females of oxianus have virtually no erect setae on the mid- and hindfemora, or setation is very sparse. In contrast, fe- males of flavipennis and funerarius usuallv have some erect setae on the mid- and hindfemora, especially on the lower sur- face. The femora of female leuconotus have considerable erect setae. Diagnostic characters given by Gussa- kovskij (1928, 1930) are of little value be- cause he compared the species to Sphex pruinosus Germar rather than to a species in the flavipennis group. De Beaumont (1968) thought that oxianus differed from flavipennis and funerarius (as maxillosus) in being slenderer, in having a slightly lon- ger female petiole, finer body sculpture, and a female clypeus that is smooth and shiny along the median line and next to the clypeal lip. However, only the female petiole length is useful. Females are 19-29 mm and males are 14.5-22 mm long. Material examined. — We have seen a male and female identified by Gussakov- skij as oxianus from TADJIKISTAN: Chan- ghir and Kabadian (Menke, USNM), and one female identified by Gussakovskij from TURKMENISTAN: Krasnovodsk (Menke). The following have also been studied, all in the CAS: TURKEY: Urfa (male, female); TURMENISTAN: Bahar- den (3 females); TADJIKISTAN: Ramit, Dushanbe, Khodzha, and Kondara Can- yon (3 males, 4 females). Variation. — Both Gussakovskij and de Beaumont thought the female was char- acterized by an entirely red gaster (exclud- ing the petiole), although de Beaumont saw an all black specimen from Israel (his nubilus). De Beaumont (1969:81) noted a female from Turkey in which gastral seg- ments V-VI were black, the remainder red. We have seen similar females from Urfa, Turkey, and Baharden, Turkmenistan (all CAS). The gaster and petiole are all red in 340 Journal of Hymenoptera Research a female from Dushanbe, Tajikistan (CAS). From this we conclude that the female gaster varies from all red to all black in oxianus. Distribution. — Israel, Turkey, Iran, Af- ghanistan, Uzbekistan, Turkmenistan, and Tajikistan. ACKNOWLEDGMENTS We thank the following people for lending speci- mens for our study: Michel Sartori, Musee Cantonal de Zoologie, Lausanne, Switzerland, who lent the type series of Sphex maxillosus mavromoustakisi de Beaumont and helped us locate the holotype of nu- bilis; Max Fischer, Naturhistorisches Museum, Wien, Austria, who lent the type material of Sphex atrohirtus Kohl; Lars-Ake Janzon, Naturhistoriska Riksmuseet, Stockholm, Sweden, who lent syntypes of Sphex fu- nerarius Gussakovskij and translated an essential part of his 1988 paper from Swedish into English; Wal- demar Celary, Instytut Systematyki i Ewolucji Zwierzat, Krakow, Poland, who lent the type of Sphex plumipes Radoszkowski; and Janine Casevitz-Weu- lersse, Museum National d'Histoire Naturelle, Paris, France, who lent the type material of Sphex triangul- um Brulle and rufocintus Brulle. Guido Pagliano, Mu- seo Regionale di Scienze Naturali, Torino, Italy, at- tempted to locate the type material of Sphex sellae Gri- bodo and Sphex flavipennis var. rufodorsatus De-Ste- fani. He was unsuccessful, but we thank him for his efforts. Alexander Antropov, Zoological Museum of the Moscow State University, Moscow, Russia, provided us with material of Sphex melas Gussakovskij. Sever- iano Gayubo, Universidad de Salamanca, Spain, Maximilian Schwarz, Ansfelden, Austria, and Jacques Hamon, Gaillard, France sent material of SpJiex atro- pilosus Kohl. Darrell Ubick, California Academy of Sciences, San Francisco, California, assisted with scanning electron photographs of the antennae. Vincent Lee California Academy of Sciences, San Francisco, California; Mi- chael Prentice, Albany, California; and Michael Ohl, Zoologisches Museum der Humbolt Universitat, Ber- lin, Germany, reviewed the manuscript for which we extend our deepest appreciation. LITERATURE CITED (Dates in parentheses represent incorrect dates print- ed on original publications) Andersson, H., C.-C. Coulianos, B. Ehnstrom, O. Hammarstedt, L. Imby, L.-A. Janzon, A. Linde- low, and H.W. Walden, 1987. Hotade evertebra- ter in Sverige — Threatened invertebrates in Swe- den. Entomologisk Tidskrift 108:65-75. Asis, J. D., S. F. Gayubo, and J. Tormos, 1990 (1989). Preimaginal stages of Iberian Sphecidae V. Sphex rufocinctus Brulle and Ammophila laevicollis Ed. Andre (Hymenoptera, Sphecidae). Annates His- torico-Naturales Musei Nationalis Hungarici 81: 239-244. Asis, J. D., and R. Jimenez, 1987. Contribucion al con- ocimiento de la esfecidofauna de la provincia de Castellon. I. Sphecinae, Pemphredoninae, Asta- tinae y Crabroninae (Hym. Sphecidae). Boleti'n de la Asociacion Espanola de Entomologia 11:19-29. Bajari, E., 1957. Magyarorszagi uj kaparodarzsak (Hym. Sphecidae). Rovartani Kbzlemenyek 10:133- 144. Balthasar, V., 1972. Grabwespen Sphecoidea. Fauna CSSR 20, 471 p. Benedek, P., 1968. A hazai kaparodarazsak (Hym., Sphecoidea) elohelyi viszonyai es viraglatogatasa — The biotopes and the flower-visiting of the Hun- garian Sphecoidea. Rovartani Kdzlentinyek (New Series) 21:65-88. Berland, L., 1952. Sur quelques Hymenopteres des Pyrenees-Orientales dont un Sphex nouveau pour la Fance. Vie et Milieu 3:87-89. Berland, L. and F. Bernard, 1949 (1947). Les Sphex de France. Annates de la Societe Entomologique de France 116:1-16. Bitsch, J., Y. Barbier, S. F. Gayubo, K. Schmidt, and M. Ohl, 1997. Hymenopteres Sphecidae d'Europe Occidentale, Volume 2. Faune de France 82:1^29. Bohart, R. M., and A. S. Menke, 1976. Sphecid Wasps of the World. A generic revision. University of Cal- ifornia Press, Berkeley, Los Angeles, London. 1 color plate, ix + 695 p. Brulle, A., 1832-1833. Expedition scientifique de Moree. Section des sciences physiques. Tome III. Ire partie. Zoologie. Deuxieme section. — Des animaux articules. F.G. Levrault, Paris. 400 p. (1-288 = 1832; 289- 400 = 1833). Dating after Sherborn and Wood- ward (1901:336). Chaudoir, G., 1947. Captures d'Hymenopteres. L'Entomologiste 3:142. Chevin, H., 1988. Les Hymenopteres Sphecides du department de la Manche. Memoires de la Societe Nationale des Sciences Naturelles et Mathematiques de Cherbourg 59 (1981-1985):13-19. Chevin, H., and S. Chevin, 1985. Les Hymenopteres Sphecides et Pompilides du Massif de la Clape (Aude). Bulletin de la Societe Versaillaise des Sci- ences Naturelles (4) 2:34-43. Compte Sart, A., 1959. Sobre los esfecidos de Mallor- ca. Graellsia 17:129-140. Dahlbom, A. G., 1843-1845. Hymenoptera Europaea praecipue borealia; formis typicis nonnullis Sped erum Generumve Exoticorum nut Extraneorum prop- ter nexum systematica-^ associatis; per Familias, Gen era, Species et Varietates disposita atque descripta. Tomus: Sphex in scnsu Linneano. Officina Lund- Volume 9, Number 2, 2000 341 bergiana, Lund, xliv + 528 p. (fasc. 1:1-172, 1843; fasc. 2:173-352, 1844, fasc. 3:353-528, i-xliv, ta- bles, 1845). Dating after Menke (1974). Dalla Torre, C. G. de, 1897. Catalogus Hymenopterorum hucusquc descriptorum systematica et synonymicus, Volumen VIII: Fossores (Sphegidae). Guilelmi En- gelmann, Lipsiae. 749 p. Day, M. C, 1991. Towards the conservation of acu- leate Hymenoptera in Europe. Convention on the Conservation of European Wildlife and Natural Hab- itats, Nature and Environment Series 51:1-44, i- xxxiii. de Andrade, N. F., 1949. Esfecidos de Portugal (Hym. Sphecidae). Memdrias e Estudos do Museu Zoolo- gico da Universidade de Coimbra (194):l-27. de Beaumont, J., 1947. Sphecidae (Hym.) de 1'ile de Chypre. Mitteilungen der Schweizerischen Entomo- logischen Geselhchaft 20:381-402. de Beaumont, J., 1949. Synonymie de quelques espe- ces de Sphecidae (Hym.). Mitteilungen der Schweizerischen Entomologischen Gesellschaft 22: 127-128. de Beaumont, J., 1951 (1949). Hymenopteres recoltes par une mission Suisse au Maroc (1947) Spheci- dae 1. Bulletin de la Societe des Sciences Naturelles du Maroc 29:259-284. de Beaumont, J., 1953. Notes sur quelques types de Sphecidae decrits par A. G. Dahlbom (Hym.). Opuscula Entomologica 18:193-198. de Beaumont, J., 1956. Sphecidae (Hym.) recoltes en Libye et au Tibesti par M. Kenneth M. Guichard. Bulletin of the Britsh Museum (Natural History) En- tomology 4(5):167-215. de Beaumont, J., 1957. Sphecidae du nord de l'lran (Hym.). Mitteilungen der Schweizerischen Entomo- logischen Geselhchaft 30:127-139. de Beaumont, J., 1960. Sphecidae (Hym.) recoltes en Tripolitaine et en Cyrenai'que par M. Kenneth M. Guichard. Bulletin of the British Museum (Natural History) Entomology 9:219-251. de Beaumont, ]., 1961a. Ergebnisse der Deutschen Af- ghanistan-Expedition 1956 der Landessammlun- gen fur Naturkunde Karlsruhe. Sphecidae (Hy- menoptera). Beitrdge zur Naturkundlichen For- schung in Siidwestdeutschland 19:271-276. de Beaumont, J., 1961b. Sphecidae de 1'ile de Crete (Hym.). Mitteilungen der Schweizerischen Entomo logischen Gesellschaft 34:43-52. de Beaumont, J., 1962. Contribution a l'etude des Sphecidae d'Espagne (Hymenoptera). Eos, Revis ta Espaiiola de Entomologia 38:17-40. de Beaumont, J., 1965. Les Sphecidae de la Grece (Hym.). Mitteilungen der Schweizerischen 1 ntomo logischen Gesellschaft 38:1-65. de Beaumont, J., 1967. Hymenoptera from Turkey. Sphecidae, I. With appendix. Sphex Linne, sub- genus Palmodes Kohl par P. Roth. Bulletin of the Britsh Museum (Natural History) Entomology 19(5): 253-382. de Beaumont, J., 1968. Sphecidae palearctiques nou- veaux ou peu connus (Hym.). Mitteilungen der Schweizerischen Entomologischen Gesellschaft 41: 145-168. de Beaumont, J., 1969. Sphecidae de Turquie (Hym.). Mitteilungen der Schweizerischen Entomologischen Gesellschaft 42:79-95. de Beaumont, J., 1970a (1969). Beitrage zur Kenntnis der Fauna Afghanistans. (Sammelergebnisse von O. Jakes 1963-1964, D. Povolny 1965, D. Povolny & Fr. Tenora 1966, J. Simek 1965-66, D. Povolny, J. Gaisler, Z. Sebek & Fr. Tenora 1967). Spheci- dae, Hym. Acta Musei Moraviae 54 (Supplemen- tum):385-406. de Beaumont, J., 1970b. Sphecidae de l'lran (Hym.). Stuttgarter Beitrage zur Naturkunde aus dem Staa- tlichen Museum fur Naturkunde in Stuttgart (220): 1-18. de Beaumont, J., and H. Bytinski-Salz, 1955. The Sphecidae (Hymen.) of Erez Israel. I. (Subfam.: Sphecinae, Nyssoninae; Trib.: Bembecini, Sti/mi). The Bulletin of the Research Council of Israel, Section B: Zoology 5:32-60. De-Stefani, T., 1887. Un nuovo genere di Crabronidi et altri imenotteri nuovi o poco cogniti raccolti in Sicilia. // Naturalista Siciliano 6:59-62, 85-90, 110-114, 143-147, pi. II [incorrectly referred to as pi. I in text). Dollfuss, H., 1983. Catalogus Faunae Austriae. Ein sys- tematischer Verzeichnis oiler auf bsterreichischem Ge- biete festgestellten Tierarten. Teil XVII: Fam.: Sphe- cidae. Verlag der Osterreichischen Akademie der Wissenschaften, Wien. 32 p. Dollfuss, H., 1987. Neue und bemerkenswerte Funde von Grabwespen (Hymenoptera, Sphecidae) in Osterreich. Linzer Biologische Beitrdge 19:17-25. Dollfuss, H., 1988. Faunistische Untersuchungen uber die Brauchbarkeit von Grabwespen (Hymenop- tera, Sphecidae) als Umweltindikatoren durch Vergleich neuer und alterer Aufnahmen von aus- gewashlten Lokalfaunen im ostlichen Osterreich. Linzer Biologische Beitrage 20:3-36. Dollfuss, H., 1989. Verzeichnis der Grabwespentypen am Naturhistorischen Museum in Wien (Hyme- noptera/ Sphecidae). Kataloge der wissenschaftlich- en Sammlungen des Naturhistorischen Museums in Wien, Entomologie 7(4): 1-26. Dollfuss, H., 1WI. Bestimmungsschlussel der Grab- wespen Nord- und Zentraleuropa^ (llvmenop- tera, Sphecidae) mil spe/iellen Angaben /ur Grabwespenfauna Osterreichs. Stapfia 24:1-247. Dollfuss, H., 1994. Rote Liste gefardeter Grabwespen (Hymenoptera, Sphecidae), p. 95-104 in J. Gepp (editor): Rote Listen gefahrdeter Tiere Osterreichs. Gnine Reihe des Bundesministeriums tur I m- 342 Journal of Hymenoptera Research welt, Jugend und Familie, volume 2. Styria Me- dien Service, Graz. 355 p. Dollfuss, H., J. Gusenleitner, and E. Bregant, 1998. Grabwespen in Burgenland (Hymenoptera: Sphecidae). Stapfia 55:507-552. Dusmet, J. M., and R. G. Mercet, 1906. Los Sphex de Espana. Boletin de la Real Sociedad Espanola de His- toria Natural 6:500-517. Eiroa, M. E., and F. Novoa, 1985. La entomofauna de las dunas de la playa de Barra (Cangas, Ponte- vedra), I: Hymenoptera, Aculeata. Boletin de la Asociacion Espanola de Entomologia 9:15-30. Fabricius, J. C, 1793. Entomologia systematiea emendata et aucta. Secundum classes, ordines, genera, species adjectis synonymis, locis, observationibus, descrip- tionibus. Vol. 2. Christ. Gottl. Proft, Hafniae. VIII + 519 p. Fischer de Waldheim, G., 1843. Observata quaedam de Hymenopteris Rossicis. Magazin de Zoologie (2) 5:1-4, pi. 122. Gayubo, S. F., 1984. Aportacion al conocimiento de los esfecidos de El Algarve (Portugal) (Hymv Sphecidae). Boletim da Sociedad Portuguesa de En- tomologia 2:349-387. Gayubo, S. F., 1986a. Notas sobre la esfecidofauna an- daluza (Hymenoptera: Sphecidae). Trabajos y Monografias del Departamento de Zoologia, Univer- sidad de Granada (N.S.) 8:31-44. Gayubo, S. F., 1986b. Fauna esfecidologica de la Prov- incia de Zamora. I. Sphecinae, Pemphredoninae, Astatinae y Larrinae (Hymenoptera: Sphecidae). Anales de Biologia, Universidad de Murcia 7:27-35. Gayubo, S. F., 1987 (1986). Fauna esfecidologica de la Provincia de Ciudad Real. I. Sphecinae, Pem- phredoninae, Astatinae y Larrinae. Graellsia 42: 103-119. Gayubo, S. F., J. D. Asis, and J. Tormos, 1990 (1989). Nuevos datos sobre la esfecidofauna iberica II (Hymenoptera: Sphecidae). Eos, Revista Espanola de Entomologia 65:7-21. Gayubo, S. F., and W. Borsato, 1994 (1991). Contri- bucion al conocimiento de la esfecidofauna ital- iana (Hymenoptera: Sphecidae). Bollettino del Museo Civico di Storia Naturale di Verona 18:195— 225. Gayubo, S. F., V. Borsato, and G. Osella, 1991 (1988). Esfecidos Italianos de la Coleccion del Museo Civico de Historia Natural de Verona (Italia). Bol- lettino del Museo Civico di Storia Naturale di Verona 15:383^24. Gayubo, S. F., V. Borsato, and G. Osella, 19^2 (1989). Esfecidos palearticos de la collection del Museo Civico de Historia Natural de Verona (Italia) (Hymenoptera: Sphecidae). Bollettino del Museo Civico di Storia Naturale di Verona 16:269-294. Gayubo, S. F., and C. Heras, 1986. Esfecidofauna de las areas de Cuellar y Segovia (Provincias de Se- govia y Valladolid). Hymenoptera: Sphecidae. Acta Salamanticensia 78:1-105. Gayubo, S. F., and F. Sanza, 1986. Esfecidofauna de la margen derecha de la cuenca alta del Duero (Hymenoptera: Sphecidae). Acta Salamanticensia, Serie Vana 74:1-115. Gayubo, S. F., and J. Tormos, 1984. Nuevas aportaciones al conocimiento de la esfecidofauna valenciana (Hytn., Sphecidae). Publica Fundacion Entomologica "Juan de Torres Sala", Serie Hymenoptera, Cuaderno 1. Valencia, title page, 35 unnumbered p. Gayubo, S. F., and J. Tormos, 1986a. Notas sobre la esfecidofauna de Castellan de la Plana (Hytn., Sphe- cidae). Publica Fundacion Entomologica "Juan de Torres Sala", Valencia, Serie Hymenoptera, Cuaderno 3. Valencia. 22 p. Gayubo, S. F., and J. Tormos, 1986b. Nuevas aporta- ciones al conociemiento de la esfecidofauna valenciana {ID (Hym., Sphecidae). Publica Fundacion Ento- mologica "Juan de Torres Sala", Serie Hymenop- tera, Cuaderno 2. Valencia. 9 p. Gayubo, S. F., J. Tormos, and J. D. Asis, 1993. Tera- tologias in esfecidos ibericos (Hymenoptera, Sphecidae). Fragmenta Entomologica 24:307-311. Gayubo, S. F., and Torres, F., 1991. Efecto de la pre- sion urbana sobre abejas y avispas (Hymenop- tera, Aculeata). IV. Sphecidae. Boletin de la Socie- dad Espanola de Historia Natural (Seccion Biologia) 87:73-92. Gonzalez, J. A., S. F. Gayubo, and F. Torres, 1998. Diversidad y abundancia de esfecidos (Hyme- noptera, Sphecidae) en un sector arenoso de la Submeseta Norte (Espana). Boletin de la Real So- ciedad Espanola de Historia Natural 94:71-85. Gonzalez, J. A., S. F. Gayubo, and F. Torres, 1999. Estudio comparativo de la biodiversidad de es- fecidos colectados mediante trampa Malaise en un sector arenoso de la cuenca de Duero (Es- pana) (Hymenoptera: Sphecidae). Nouvelle Revue d'Entomologie (Nouvelle Serie) 15:351-370. Gribodo, G., 1873. Contribuzioni alia fauna imenot- terologica italiana. Bollettino delta Societa Entomo- logica Italiana 5:73-87. Guichard, K. M., 1978. Additions to the Greek wasp fauna (Hymenoptera: Sphecidae). Annates Musei Goulandris 4:267-272. Gusenleitner, J., 1996a. Hymenopterologische Noti- zen aus Osterreich - 6 (Insecta: Hymenoptera aculeata). Linzer Biologische Beitrage 28:809-816. Gusenleitner, J., 1996b. Kurzbericht iiber Sphecinae in Istrien (Croatia) (Hymenoptera, Sphecidae). Lin- zer Biologische Beitrage 28:817-819. Gussakovskij, V. V., 1928. Novyie vidy Sphecidae iz Zakaspiya i Khivy— Sphecidarum species novae. Izvestii/a kursov Prikladnoy Zoologii I Fitopatologii (= Bulletin de I'Institut de Zoologie Appliquee et de Phytopathologie) 4:3-19. Volume 9, Number 2, 2000 343 Gussakovskij, V. V., 1930. Novyie i malo izvestnyie vidy rodov Ammophila Kby. I Sphex L. (Hyme- noptera, Sphecidae) — Species novae vel parum cognitae generum Ammophila Kby. et Sphex L. (Hymenoptera, Sphecidae). Russkoye Entomologi- cheskoye Obozreniye 24:199-211. Gussakovskij, V. V., 1933 (1932). Sphecidae i Psam- mocharidae (Hymenoptera), sobrannyie N. Za- rudnym v vostochnoy Persii — Sphecidae et Psammocharidae (Hymenoptera), a cl. N. Zarud- nyi in Persia orientali collectae. Trudy Zoologis- cheskogo Instituta Akademii Nauk SSSR (= Travaux de I'Institut Zoologique de I'Academie des Sciences de I'URSS) 1:369-404 (misprinted as 269-304). Gussakovskij, V. V., 1934. Schwedisch-Chinesische Wissenschaftliche Expedition nach den nordwes- tlichen Provinzen Chinas unter Leitung von Dr. Sven Hedin und Prof. Sii Ping-Chang. Insekten gesammelt vom Schwedischen Arzt der Expedi- tion Dr. David Hummel 1927-1930. 41. Hyme- noptera. 6. Sphegidae. Arkiv for Zoologi 27A(21): 1-15. Gussakovskij, V. V., 1935. K faune os (Hymenoptera, Sphecodea et Vespodea) Tadzhikistana — Sphe- codea und Vespodea von Tadjikistan. Trudy Tad- zhikskoy Bazy Akademii Nauk SSSR (= Travaux de la Filiate de I'Academie des Sciences de I'URSS au Tadjikistan) 5:409-467. Hamon, J., 1950. Quelques Hymenopteres vespifor- mes rares pris dans le Pyrenees-Orientales. L'Entomologiste 6:28-29. Hamon, J., R. Fonfria, and M. Tussac, 1991. Les Sphe- cini de France continentale et de Corse (Hyme- noptera: Sphecidae). Especes presentes et cles d'identification. Bulletin de la Societe Entomolo- gique de France 96:125-134. Internationa] Commission on Zoological Nomencla- ture, 1999. International Code of Zoological Nomen- clature. Fourth Edition adopted by the International Union of Biological Sciences. The International Trust for Zoological Nomenclature, London. XXIX + 306 p. Islamov, Sh. D., 1986. Royushchiye osy (Hymenop- tera, Sphecidae) gornykh rayonov Uzbekistana — Digger-wasps (Hymenoptera, Sphecidae) in the mountainous regions of Uzbekistan. Entomologi cheskoye Obozreniye 65:513-534. [English transla- tion: 1987. Digger-wasps (Hymenoptera, Spheci- dae) of the mountainous regions of Uzbekistan. Entomological Review 66:35-59.) Islamov, Sh. D., 1989. Royushchiye osy (Hymenop- tera, Sphecidae) okhotyashchyesya za pryamok- rylymi nasekomymi v zapadnom Tian-Shane. Uzbekskiy Biologicheskiy Zhurnal 5:49-50. Jacobs, H. J., 1989. Ein Verzeichnis der bisher in Mecklenburg nachgewiesenen Grabwespenarten (Hymenoptera, Sphecidae). Natur und Natur- schutz in Mecklenburg-Vorpommem 28:2-36. Jacobs, H. J., and J. Oehlke, 1990. Beitrage zur Insek- tenfauna der DDR: Hymenoptera: Sphecidae. 1. Nachtrag. Beitrage zur Entomologie 40:121-229. Janzon, L.-A., 1988. Status och skydd av rovstekeln Sphex rufocinctus. Inventeringar 1987. Lansstyrel- sen i Gotlands Lan, Naturvdrdsfunktionen 1988:1-6. Jozan, Z., 1985. Del-Dunatul kaparodarazs (Hyme- noptera, Sphecoidea) faunajanak alapvetese — Grundriss der Sphecoidea-Fauna (Hymenoptera) des Siid-Transdanubiens, Ungarn. A Janus Pan nonius Mtizeum Evkonyve 29:53-86. Jozan, Z., 1986. The Scolioidea and Sphecoidea Fauna of the Kiskunzag National Park. The Fauna of the Kiskunsdg National Park, p. 365-381. Jozan, Z., 1989. A Tihanyi Tajvedelmi Korzet fullan- kos faunaja (Hymenoptera, Aculeata), I — Aculea- ta-Fauna (Hymenoptera) des Landschaftschutz- gebietes von Tihany, Ungarn, I. Folia Musei His- torico-Naturalis Bakonyiensis 8:79-110. Jozan, Z., 1991. The Scolioidea, Sphecoidea and Apoidea fauna of the Bdtorliget Nature Reserves (Hymenop- tera: Aculeata), p. 601-621 in S. Mahunka (editor): The Batorliget Nature Reserves — after forty years. Vol. 2. Magyar Termeszettudomanyi Mu- zeum, Budapest, p. 499-848 Jozan, Z., 1992. A Boronka-melleki Tajvedelmi Korzet fullankos hartyasszarnyu (Hymenoptera, Acu- leata) faunajanak alapvetese — Grundriss der Aculeaten-Fauna (Hymenoptera) des Land- schaftschutzgebietes Boronka-melleki, Ungarn. Dundntuli Dolgozatok Termeszettudomanyi Sorozat 7:163-210. Jozan, Z., 1995. Adatok a tervezett Duna-Drava Ne- mzeti Park fullankos hartyasszarnyu (Hymenop- tera, Aculeata) faunajanak ismeretehez — Beitrage zur Kenntnis der Aculeatenfauna (Hymenoptera) des Nationalparkes Duna-Drava (Ungarn, Siid- Transdanubien). Dundntuli Dolgozatok Termeszet- tudomanyi Sorozat 8:99-115. Karsai, I., 1988. Ecofaunistical investigations of sphe- roid fauna on a sandy grassland. Acta Biologica (Szeged) 34:97-106. Kazenas, V. L., 1978. Royushchiye osy Kazakhstana i Sred ney Azii (Hymenoptera, Sphecidae). Opredelitel'—The digger zoasps of Kazakhstan and Middle Asia (Hyme- noptera, Sphecidae). The determinant. Izdatektvo Nauka Kazakhskoy SSR, Alma Ata. 172 p. Kazenas, V. L., and S. R. Nasyrova, 1991. Royushchi- ye osy (Hymenoptera, Sphecidae) — vragi pry- amokrylykh (Orthoptera) v pustvnnov zone Ka- zakhstana [ Digger wasps (Hymenoptera, Sphecidae) as predators of Orthoptera in the de- sert zone of Kazakhstan]. Izvestiya Akademii Nauk Kazakhskoy SSR. Seriya Biologicheskaya 6:37-40. Kazenas, V. I .. and V. I robias, L992 Nochevochnyie skopleniya royushchikh os (Hymenoptera, Sphe- cidae) v yugo-vostochnom Kazakhstane — Night sleeping aggregations oi digger wasps (Hvme 344 Journal of Hymenoptera Research noptera, Sphecidae) in southeastern Kazakhstan. Entomologicheskoye Obozreniye 71:28-31. Kohl, F. F., 1881. Sphegidologische Studien. Ueber ei- nige Dahlbom'sche Sphexarten. Entomologische Nachrichten 7:27-31, 37-40, 53-56, 90-94, 239-244. Kohl, F. F., 1885. Die Gattungen der Sphecinen und die palaarktische S/?/;ex-Arten. Termeszetrajzi Fiiz- etek 9:154-207, pi. VII-VIII. Kohl, F. F., 1890. Die Hymenopterengruppe der Sphe- cinen. I. Monographie der naturlichen Gattung Sphex Linne (sens. lat.). Annalen des k. k. Natur- historischen Hofmuseums 5:77-194, 317-462., pi. VIII-XII, Kohl, F. F., 1895. Zur Monographie der naturlichen Gattung Sphex Linne. Annalen des k. k. Naturhis- torischen Ho/museums 10:42-74, pi. IV-V. Krasnobayev, Y., P., A. V. Antropov, I. V. Lyubvina, and S. I. Zabelin, 1995. Fauna bespozvonochnykh Zhiguley [= Invertebrates of Zhiguli Hills, Eu- ropean Russia]. V. Otryad Hymenoptera (Insec- ta). Samarskaya Luka 6:123-144. Leclercq, J., 1955. Hymenoptera Sphecoidea (Sphecidae I. Subfam. Sphecinae). Exploration du Pare National de I'Upemba, fasc. 34. 137 p. Leclercq, J., 1956. Mission E. Janssens et R. Toilet en Grece (Juillet-aout 1953). 14e Note. Hymenop- tera— Sphecidae et Vespidae. Bulletin & Annates de la Societe Entomologique de Belgique 92:324-326. Lepeletier de Saint Fargeau, A., 1845. Histoire naturelle des Insectes. Hymenopteres, Vol. 3. Librairie Ency- clopedique de Robert, Paris. 646 p., pi. 25-36. Lomholdt, O., 1975. The Sphecidae (Hymenoptera) of Fennoscandia and Denmark. Faufia Entomologica Scandinavica 4(l):l-224. Luchetti, D., 1993. Lista preliminaria degli Imenotteri dell'arcipelago della Maddalena (Sardegna). Bol- lettino della Societa Entomologica Italiana 125:103- 108. Melo, G. A. R., 1999. Phylogenetic relationships and classification of the major lineages of Apoidea (Hymenoptera), with emphasis on the crabronid wasps. Scientific Papers, Natural History Museum, The University of Kansas (14):l-55. Menke, A. S., 1974. The dates of publication of A.G. Dahlbom's Hymenoptera Europaea, vol. 1. Pol- skie Pismo Entomologiczne 44:315-317. Mingo, E., and S. F. Gayubo, 1984 (1983). Sphecidae de Espaha. I. Ampulicinae y Sphecinae (Hym.). Eos, Revista Espafwla de Entomologia 59:137-164. Minoranskiy, V. A., and A. V. Shkuratov, 1996. K fau- ne royushshikh os (Hymenoptera, Sphecidae) Rostovskoy oblasti. [= On the sphecid fauna of Rostov Oblast', Russia]. Izvestiya Vysshikh Uch- ebnykh Zavedeniy. Severo-Kavkazskiy Regyon. Estestvennyie nauki 1996(4):80-83. Morice, F. D., 1911. Hymenoptera Aculeata collected in Algeria. The Sphegidae. (Being Part V of the work commenced by the late Edward Saunders, F. R. S., in Trans. Ent. Soc. 1901, p. 515.). Trans- actions of the Entomological Society of London 1911: 62-135. Negrisolo, E., 1991 (1990). Hymenoptera Sphecidae e Scoliidae raccolti nel Friuli-Venezia Giulia e nel Veneto orientale. Gortania, Atti del Museo Friulano di Storia Naturale 12:313-327. Negrisolo, E., 1995. The aculeate communities (Hy- menoptera Aculeata) of two coastal areas of the Veneto region (northeastern Italy). Lavori, cieta Veneziana di Scienze Naturali 20:15-25. Padr, Z., 1989. Sphecoidea, p. 165-171 in J. Sedivy (editor): Enumeratio Insectorum Bohemoslovak- iae. Acta Faunistica Entomologica Musei Natiorudis Pragae. 194 p. Pagliano, G., 1980. Fauna imenotterologica delle Langhe. (Nota faunistica II — Sphecidae). Rivista Piemontese di Storia Naturale 1:105-135. Pagliano, G., 1984. Sphecinae italiani (Hymenoptera: Sphecidae). Parte I. Bollettino, Museo Regionale di Scienze Naturali, Torino 2:359-370. Pagliano, G., 1985. Sphecinae italiani (Hymenoptera: Sphecidae). Parte II. Bollettino, Museo Regionale di Scienze Naturali, Torino 3:5-27. Pagliano, G., 1990. Catalogo degli Imenotteri italiani. II. — Sphecidae. Bollettino, Museo Regionale di Scienze Naturali, Torino 8 (1): 53-141. Pagliano, G., and F. Pesarini, 1995. Nota preliminare sugli imetotteri sfecidi della provincia di Ferrara (Hymenoptera Sphecidae). Quaderno della Stazio- ne Ecologica del Museo Civico de Storia Naturale di Ferrara 8:81-95. Pagliano, G., and P. Scaramozzino, 1995. Hymenop- tera Gasteruptiidae, Ichneumonidae e Aculeata (esclusi Chrysidoidea, Mutillidae e Formicidae), p. 723-738 in B. Massa (editor): Arthropoda di Lampedusa, Linosa e Pantelleria (Canale di Si- cilia, Mar Mediterraneo). // Naturalista Siciliano 19 (Supplement). Pulawski, W. }., 1978. Nadsem. Sphecoidea, p. 173-279 in G. S. Medvedev (editor): Opredelitel' naseko- mykh evropeyskoy chasti SSSR, Tom III. Perepon- chatokrylyye, Pervaya chast' [= Keys to the iden- tification of insects of the European USSR, Vol. 3, part 1]. Nauka, Leningrad. 584 p. Radoszkowski, O., 1886. Faune Hymenopterologique Transcaspienne. Horae Societatis Entomologicae Rossicae 20:3-56, pi. I-XI. Richards, O. W., 1979 (1978). The Hymenoptera Acu- leata of the Channel Islands. Report ami Transac- tions, La Societe Guernesiaise 1978:389^124. Roth, P., 1925. Les Sphex del'Afrique du Nord. An- nates de la Societe Lutomologique de France 94:365- 404. Salmon, K. A., 1929. The external morphology of Pep- sis elegans Lepeletier (Hymenoptera: Psammo- charidae). Transactions of The American Entomo- logical Society 55:119-153. Volume 9, Number 2, 2000 345 Saunders, E., 1910. Hymenoptera Aculeata collected in Algeria by the Rev. Alfred Edwin Eaton, M. A., F. E. S., and the Rev. Francis David Morice, M. A., F. E. S. Part IV. Descriptions of new Sphegidae. Transactions of the Entomological Soci- ety of London 1910:517-531. Savigny, M. J. C. L. de, 1809-1829. Description de I'Egypte, on recueil des observations et des recherches qui ont ete fades en Egypte pendant I'expedition de Varmee francaise, public par les ordres de sa Majeste I 'Empereur Napoleon le Grand. Vol. 1, Part 4, Ex- plication sommaire des Planches. Insectes [by J- V. Audouin], p. 187-202 [1826]. See Tollitt (1986) for details about this work. Schembri, S., 1991. Sphecids [sic] wasps of the Mal- tese Islands. A review (Hymenoptera Sphecidae). Bollettino della Societa Entomologica Italiana 122: 175-184. Schmid-Egger, C, 1996. Neue oder bemerkenswerte siidwestdeutsche Stechimmenfunde. BeinbiX 7: 18-21. Schmid-Egger, C, S. Risch, and O. Niehuis, 1995. Die Wildbienen und Wespen in Rheinland-Pfalz (Hy- menoptera, Aculeata). Verbreitung, Okologie und Gefahrdungsituation. Fauna und Flora in Rheinland-Pfalz Beiheft 16:1-296. Schmid-Egger, C, K. Schmidt, and D. Doczkal, 1996. Rote Liste der Grabwespen Baden-Wiirttembergs (Hymenoptera: Sphecidae). Natur und Laudschaft 71:371-380. Schmidt, K., and Ch. Schmidt-Egger, 1997. Kritisches Verzeichnis der deutschen Grabwespen (Hyme- noptera, Sphecidae). Mitteilungen der Arbeitsge- meinschaft Ostivestfdlisch-Lippischer Entomologen 13, Beiheft 3:1-35. Schmidt, K., and P. Westrich, 1983. Beitrag zur Kennt- nis der Grabwespenfauna Nord-Griechenlands (Hymenoptera, Sphecidae). Nachrichtenblatt der Bayerischen Entomologen 32:118-126. Schmidt, K., and P. Westrich, 1987. Stechimmen (Hy- menoptera: Aculeata) des Mainzer Sandes und des Gonsenheimer Waldes. Mainzer Naturwissen schaftliches Archiv 25:351-107. Schulthess, A., 1926. Contribution a la connaissance de la faune des Hymenopteres de l'Afrique du Nord. Deuxieme Partie — Fossores (en collabora- tion avec P. Roth). Bulletin de la Sociite d'Histoire Nat ii relic de l'Afrique du Nord 17:206-220. Schulz, W. A., 1911. Grabwespen-Tvpen Tourniers, Brulles, Lepeletiers und Schencks. Societu> I ///<> mologica 26:57-59, 63-64, 67-68. Scobiola, X., 1960. Les especes de Sphex l.inne et Am- mophila Kirby (Hymenoptera, Sphecidae) de la faune de la Republique Populaire Roumaine. Travuiix du Museum d'Histoire Naturelle Grigore Antipa 2:229-239. Sherborn, C. D., and B. B. Woodward, 1901. Dates of publication of the zoological and botanical por- tions of some French voyages. — Part II. Ferret and Galinier's 'Voyage en Abyssinie'; 'Lefebvre's 'Voyage en Abyssinie'; 'Exploration scientifique de l'Algerie'; Castelnau's 'Amerique du Sud'; Dumont d'Urville's 'Voyage de 1' Astrolabe'; La- place's 'Voyage sur la Favorite'; Jacquemont's 'Voyage dans I'lnde'; Trehouart's 'Commission scientifique d'lslande'; Cailliaud, 'Voyage a Mer- oe'; 'Expedition scientifique de Moree'; Fabre, 'Commission scientifique du Nord'; Du IVtit Thouars, 'Voyage de la Venus'; and on the dates of the 'Faune Francaise.' Annals and Magazine of Natural History (7) 8:161-164, 333-336, 491-194. Shkuratov, A. V., 1998. Dopolneniya k faune royush- chikh os (Hymenoptera, Sphecidae) Rostovskoy oblasti. Izvestiya Vysshikh Uchebnykh Zavedeniy. Sevew-Kavkazskiy Regyon. Estestvennyie Nauki 3: 97-99. Tollitt, M. E., 1986. Dates and authorship of the text volumes of the Histoire Naturelle section of Sa- vigny's Description de I'Egypte. A. N.(S.) 2515. The Bulletin of Zoological Nomenclature 43:107-111. Tormos, J., J. D. Asfs, and S. F. Gayubo, 1994. Espe- cidofauna [sic] de la Provincia de Albacete (Hy- menoptera: Sphecidae). Al-Basit. Revista de E^tu dios Albacetenses 20:183-246. Tormos, J., and R. Jimenez, 1985. Contribucion al con- ocimiento de la esfecidofauna de la Provincia de Alicante (Hym. Sphecidae). Mediterranea (Serie de Estudios Biologicos) 8:29-15. Tormos, J., and R. Jimenez, 1987a. Esfecidos de la Provincia de Valencia (Hym., Sphecidae). Gracll- sia 42:121-130. Tormos, J., and R. Jimenez, 1987b. Estudio compara- tive de la esfecidofauna de la Dehesa de El Saler (Hymenoptera, Sphecoidea). Eos, Revista Espanola de Entomologia 62:315-318. Torregrosa, S. G., S. F. Gayubo, J. Tormos, and J. D. Asfs, 1993. Esfecidos de la provincia de Alicante (Hymenoptera: Sphecidae). Boletin de la Asocia- cion Espanola de Entomologia 17:9-25. van der Vecht, J., 1959. Notes on aculeate Hymenop- tera described in the period 1758-1810. Part III. Entomologische Berichten 19:211-215. van der Vecht, J., 1961. Hymenoptera Sphecoidea Fa- briciana. Zoologische Verhandelingen 4S:l-85. Vernier, R., 1995. Isodontia mcxicana (Sauss.), un Sphe- cini americain naturalise en Suisse (Hymenop- tera, Sphecidae). Mitteilungen do Schweizerischen Entomologischen Gesellschafi 68 169 177. Voblenko, A. S., V. A. Gorobchishin, and M. A. Nes- terov, 1996. Digger wasps (Hymenoptera, Sphe- cidae) of Ukrainian Polesye. Sphecos 30:14-15. Westrich, P., and K. Schmidt, 1985. Role Liste der Ste- chimmen Baden-Wiirttembergs (I [ymenoptera Aculeata ausser Chrysididae) (Stand 1. 1. 1985). Verqffentlichungen fur Naturschutz und land schaftspflege in Baden-Wurttemberg 59-60:93- 120. 346 Journal of Hymenoptera Research Zanon, V., 1925. Contribute) alia conescenza della fau- Zehnder, G., and J. Zettel, 1999. Auensukzession un na entomologicha di Bengasi. Memorie della Pout. Zonation im Rottensand (Pfynwald, Kt. VS). I. Accademia delle Science Nuovi Lincei (2) 8:61-100. Wiederbesiedlung einer Uberschwemmungsfla- Zavadil, V., and ]. Snoflak, 1948. Kutilky (Sphecidae) che durch Grabwespen (Hymenoptera, Spheci- Ceskoslovenske Republiky. Entomolgicke Prirucky dae). Mitteihtiigen der Schweizerichen Entomolo- Entomologickych Listit (13):1-179. gischen Gesellschaft 72:123-137. J. HYM. RES. Vol. 9(2), 2000, pp. 347-351 Didineis massaica, New Species, the First Afrotropical Member of the Genus, and Redescription of Didineis nigricans Morice, 1911 (Hymenoptera: Apoidea: Crabronidae: Bembicinae) WOJCIECH J. PULAWSKI Department of Entomology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA, e-mail: wpulawski@calacademy.org Abstract. — Didineis massaica Pulawski, new species, is the first member of the genus known to occur in subsaharan Africa. It can be recognized by its all black gaster and proportions of the flagellomeres. The type series was collected 27 km SSE of Naivasha in the Rift Valley Province of Kenya. Didineis nigricans Morice, 1911, from Algeria is redescribed. Didineis is a genus of 26 currently known species (Bohart and Menke 1976, Guichard 1990). Of them, 17 are Palearctic (ranging from Great Britain to Japan, and from Denmark to Algeria, Turkey, and Turkmenistan), 7 are Nearctic (Pennsyl- vania to Florida, Washington to Baja Cal- ifornia, extending south to Nuevo Leon State in Mexico), 1 is Neotropical (Cuba only), and 1 Oriental (Bengal). During a recent expedition to Kenya, Jere S. Schwei- kert and I collected a series of specimens of a previously unknown species, the first to be found in subsaharan Africa. Didineis nigricans Morice, a little known species from Algeria, is redescribed as a byprod- uct of this study. The family group names used in the ti- tle are based on Menke (1997) and Melo (1999). The morphological terminology follows Bohart and Menke (1976). Didineis massaica Pulawski, new species Derivation of name. — Massaica is a Neo- latin feminine adjective derived from the Massai people of Kenya in whose tribal area this species was collected. Diagnosis. — Didineis massaica differs from most of its congeners in having an all black gaster (rather than red at least basally). The only other Didineis with an all black gaster are: barbieri (de Beaumont), bucharica Gussakovskij, the male of latro (de Beaumont), nigricans Morice (terga somewhat reddish apically), orientalis Cameron, sibirica Gussakovskij, and tur- anica Gussakovskij. Of these species, I have seen only the holotype of nigricans and one male of latro, but they can be dis- tinguished from massaica by the characters detailed below following the original de- scriptions of Cameron (1897), de Beau- mont (1967, 1968), and Gussakovskij (1937), Tsuneki's (1968) description of Di- dineis sibirica nipponica, and the key by Ne- mkov (1995): Didineis barbieri from Algeria (female unknown) has flagellomere I and IX about 1.3 x and less than 1.0 X as long as wide, respectively. In massaica, these ratios are about 2.0 and 1.0 respectively. Didineis bucharica from Uzbekistan (female un- known) has flagellomeres I-III somewhat ex- panded laterally, and flagellomeres III and IV no longer than wide. In the male of massaica, the respective flagellomeres are cylindrical and longer than wide. Didineis latro from Turkey has a black scape, appressed tergal setae, and no tergal fasciae. In massaica, the scapal venter is yellow, ter- gum I bears several erect setae laterally, and at least tergum I has a broadly interrupted setal fascia. 348 Journal of Hymenoptera Research oo DO Figs. 1-6. Didineis species: 1-3, Didineis massaica: 1, male clypeus. 2, apical male flagellomeres. 3, propodeal enclosure. 4-6, Didineis nigricans: 4, male clypeus. 5, apical male flagellomeres. 6, propodeal enclosure. Didineis nigricans from Algeria (female un- known) has an obtusely tridentate clypeal free margin (Fig. 4) and flagellomere I about as long as wide. In the male of massaica, the clypeal free margin is evenly arcuate and fla- gellomere I is about 2.0 X as long as wide. Didineis orien talis from Bengal (female un- known) has yellow markings on the tegula and tibiae, whereas such markings lack in massaica. Didineis sibirica from Eastern Siberia and Japan has female flagellomeres abotit 2.0 x as long as wide, and male flagellomeres I and IX about 1.5 X as long as wide, respectively. In massaica, female flagellomeres II and IX are about 2.5 X and 1.5 X as long as wide, re- spectively, and male flagellomeres I and IX are about 2.0 and 1.0 X as long as wide, re- spectively. Didineis turanica (female unknown) has the me- dian and penultimate flagellomeres longer than wide, forefemur with ventral margin Volume 9, Number 2, 2000 349 somewhat flattened, and foretibia slightly broader apically than near the midlength. In the male of massaica, flagellomeres VIII and IX are about as long as wide, the forefemur is not flattened (ventral margin evenly arcu- ate), and the foretibia is not broadened api- cally. Description. — Frons finely, evenly punc- tate, punctures less than 1 diameter apart (averaging more than 1 diameter apart on interocellar area and between ocelli and orbits). Pronotum with no transverse ca- rina. Mesoscutum evenly punctate, punc- tures about 1 diameter apart. Mesopleu- ron dull, punctate (punctures less than 1 diameter apart except more posteroven- trally), without well-defined ridges poster- oventrally. Propodeal enclosure narrow, acutely angulate apically (Fig. 3); side ir- regularly ridged, apicolateral spine well defined, as in lunicornis Fabricius (the type species of the genus). Tergal punctures well defined. Mesopleural setae in female denser be- low scrobal sulcus than on remaining sur- face, almost forming a discrete patch. Ter- gum I with apical, broadly interrupted se- tal fascia (also II and III in one of the males examined). Tergum I laterally with erect setae (some setae on remaining terga also erect). Head, thorax, and gaster black (includ- ing flagellum and pronotal lobe), but man- dible yellowish red mesally, scapal venter pale yellow, and tegula dark brown in one of the males. Femora black in female ex- cept forefemur reddish brown ventrally and apically; in male forefemur reddish brown except dorsum black in basal half, midfemur reddish brown except dorsum black, and hindfemur all black or with brownish outer surface. Tibiae reddish brown, mid- and hindtibiae darkened api- cally. Tarsi dark brown. Wing membrane slightly infumate, with darker fascia that covers all marginal cell, distal portion of submarginal I, all of submarginal II, distal part of discoidal II, an area of varying width distad of discoidal II, and in female submarginal III. 9 — Clypeus uniformly, closely punctate except impunctate and shiny anteromesal- ly (impunctate area about as long mesally as basal punctate area); free margin tri- dentate. Length of flagellomere II about 2.5 x width, of flagellomere IX about 1.5 X width. Dorsoexternal hindtibial setae suberect, shorter setae interspersed with longer ones. 6 — Clypeus uniformly, closely punc- tate, free margin nearly straight mesally (Fig. 1). Scape not concave laterally. Fla- gellomeres I-X cylindrical (flagellomere X with usual apicoventral expansion); dorsal length of I about 2.0 X apical width, equal to that of II whose length is about 1.7 x width, following articles progressively shorter, VIII and IX about as long as wide; flagellomere XI markedly curved, with sharp apex (Fig. 2). Foretibia and foretar- sus not widened. Records. — Holotype: 6\ Kenya: Rift Val- ley Province, 27 km SSE Naivasha at 0°54.6'S 36°31.0'E, 3 June 1999, W.J. Pu- lawski and J.S. Schweikert (California Academy of Sciences). Paratypes: same data and depository (1 9,2 6). Didineis nigricans Morice (Figs. 4-6) Didineis nigricans Morice, 1911:111, 6. Holo- type: 6, Algeria: Biskra (Oxford University Museum). — Gussakovskij, 1937: 616 (original description copied, discussion of characters); R. Bohart and Menke, 1976:459 (listed). This distinctive species is known from a single specimen collected more than 100 years ago that has never been reexamined since the original description. The latter omits some important structures (e.g., the shape of the propodeal enclosure) and in- accurately depicts some structures (e.g., "clypeus evidenter tridentatus"). Diagnosis. — The male of nigricans can be immediately recognized by its obtusely tridentate clypeal free margin (Fig. 4) and by the proportions of the flagellomeres 350 Journal of Hymenoptera Research (flagellomere I about as long as wide, II about 1.5 X as long as wide, VIII and IX as long as wide). Description. — Frons finely, evenly punc- tate, punctures about 1 diameter apart (several diameter apart on interocellar area and between ocelli and orbits). Pron- otum without transverse carina. Mesos- cutum evenly punctate, punctures aver- aging about 2 diameters apart. Mesopleu- ron dull, somewhat irregularly punctate, with punctures no more than 1 diameter apart except episcrobal area shiny, with punctures more than 1 diameter apart, without well-defined ridges posteroven- trally. Propodeal enclosure somewhat broadened and not acutely angulate at apex (Fig. 6); side irregularly ridged, api- colateral spine slightly shorter than in {un- icornis. Tergal punctures well defined, but markedly smaller on tergum I than on re- maining terga. Gastral terga without setal fasciae. Ter- gum I laterally with erect setae. Head, thorax, and gaster black (includ- ing flagellum and pronotal lobe), but man- dible reddish mesally and scapal venter reddish brown, and apical depressions of terga reddish from certain angles. Femora reddish brown (hindfemur somewhat darkened basally), tibiae and tarsi reddish. Wings membrane slightly infumate, with marginal cell, submarginal cell II, and dor- soapical portion of discoidal cell II mini- mally darker. 6. — Clypeus uniformly, closely punc- tate, free margin obtusely tridentate mes- ally (Fig. 4). Scape not concave laterally. Flagellomeres I-X cylindrical (flagellomere X with usual apicoventral expansion); dor- sal length of I about equal to apical width and equal to 0.75 of II (whose dorsal length is 1.5 X apical width); III-VII be- coming gradually shorter, VIII and IX as long as wide; flagellomere XI markedly curved, with apex sharp (Fig. 5). Foretibia and foretarsus not widened. Collecting date.— -29 May 1898. Material examined. — Only the holotype was seen. ACKNOWLEDGMENTS I sincerely thank Jere S. Schweikert (California Academy of Sciences, San Francisco, California) for helping collect the specimens, Christopher O'Toole (Oxford University Museum, Great Britain), for lend- ing the holotype of Didineis nigricans, Virginia Kirsch (San Francisco), for generating the illustrations, and E. Eric Grissell (Systematic Entomology Laboratory, USD A, Washington, D.C.), Arnold S. Menke (Am- mophila Research Institute, Bisbee, Arizona) and Mi- chael A. Prentice (Albany, California) for reviewing an earlier version of the manuscript. LITERATURE CITED Bohart, R.M., and A.S. Menke. 1976. Sphecid Wasps of the World. A generic revision. University of Cali- fornia Press, Berkeley, Los Angeles, London. 1 color plate, IX + 695 pp. Cameron, P. 1897. Hymenoptera Orientalia, or con- tributions to a knowledge of the Hymenoptera of the Oriental Zoological Region. Part VI. Memoirs and Proceedings of the Manciiester Literary & Phil- osophical Society 41 (4):l-28, pi. 16. de Beaumont, J. 1967. Hymenoptera from Turkey. Sphecidae, I. With Appendix. Sphex Linne, Sub- genus Pahnodes Kohl par P. Roth. Bulletin of the British Museum (Natural History). Entomology 19: 253-382. de Beaumont, J. 1968. Sphecidae palearctiques nou- veaux ou peu connus (Hym.). Mitteilungen der Schweizerischen Entomologischen Gesellschaft 41: 145-168. Guichard, K.M. 1990. New Sphecidae from South Spain (Hymenoptera, Sphecidae). Entomofauna 11:273-280. Gussakovskij, V.V. 1937. Obzor palearkticheskikh vi- dov rodov Didineis Wesm., Pison Latr. i Psen Latr. (Hymenoptera Sphecodea) — Especes palearc- tiques des genres Didineis Wesm., Pison Latr. et Psen Latr. (Hymenoptera Sphecodea). Trudy Zoologicheskogo Instituta Akademii Nauk SSSR ( = Travaux de 1'lnstitut Zoologique de 1'Academie des Sciences de 1'USSR) 4:599-698, pi. 1. Melo, G.A.R. 1999. Phylogenetic relationships and classification of the major lineages of Apoidea (Hymenoptera), with emphasis on crabronid wasps. Scientific Papers. Natural History Museum. The University of Kansas 14:1-55. Menke, A.S. 1997. Family-group names in Sphecidae (Hymenoptera: Apoidea). journal of Hymenoptera Research 6:243-255. Morice, F.D. 1911. Hymenoptera aculeata collected in Algeria. The Sphegidae (Being I'art V of the work commenced by the late Edward Saunders, F.R.S., Volume 9, Number 2, 2000 351 in Trans. Entomol. Soc, 1901, p. 515). The Trans- Setchatokrylyie, skorpionnitsy, perqjonchatokrylyie. actions of the Entomological Society of London 1911: Chast' 1. Nauka, Sankt-Peterburg. 604 + 2 un- 62-135. numbered pp. Nemkov, P.G. 1995. Podsem. Nyssoninae, p. 445^166 Tsuneki, K. 1968. Descriptions and records of some in P.A. Lehr (editor). Opredelitel' nasekomykh fossorial wasps in Japan (Hym., Sphecidae). Eti- Dal'nego Vostoka Rossii i> shesti tomakh. Tom IV. zenia 27:1-8. J. HYM. RES. Vol. 9(2), 2000, pp. 352-362 A Revision of Entomosericus Dahlbom 1845 (Hymenoptera: Apoidea: "Sphecidae") with Description of a New Species Christian Schmid-Egger Flemingstr. 10, D-10557 Berlin, Germany, e-mail: schmideggC@aol.com Abstract. — Entomosericus Dahlbom, 1845, is a palearctic genus with three species: E. haiiseri Schmid-Egger, new species, from Israel, Jordan and southern Turkey, E. kaufinani Radoszkowski from southern-central Asia, and E. concinnus Dahlbom from eastern and southern Mediterranean area, Ukraine and Russia. E. concinnus rufescens de Beaumont is synonymized with E. concinnus (new synonym). A key to species identification is presented. A lectotype of E. kaufinani is desig- nated. Gastral color is determined unsuitable for distinguishing species. Entomosericus Dahlbom, 1845, is a pa- Handlirsch 1888), while females from the learctic genus of three species. Bohart and mainland Greece have red terga I and II. Menke (1976) and Menke (1997) placed it Later, he (1967) reported two females of the in its own subfamily because they could species from Turkey with red basal terga, not associate it with any other sphecid considering the presence of E. kaufinani in group. Kazenas and Alexander (1993) sug- the western Palaearctic to be doubtful, gested a relationship to the clade of Nys- My own studies confirmed the obser- soninae (now Bembecinae sensu Menke vations of de Beaumont (1965, 1967). Ad- 1997), Philanthinae and Astatinae based ditionally, a new species from the eastern on larval characters. Melo (1999) finally Mediterranean area was found. Sternal proposed a relationship to the Pemphre- and antennal characters allow males to be doninae. He considered Entomosericus as a reliably distinguished and the geographi- sister subtribe to the Psenini (sensu Bohart cal distribution of the species to be delin- and Menke 1976). eated. Females of the species remain dif- Two species of Entomosericus have been ficult to distinguish and require compari- described: £. concinnus Dahlbom, 1845, from son with determined material to be iden- the Greek island Rhodos and E. kaufinani tified. Radoszkowski, 1877, from central Asia. In Sources of material. — Specimens from the current keys, the two species have been pri- following institutions were examined (ab- marily distinghuished through color. breviations used are given): According to Balthasar (1972), all of the Arens coll. Werner Arens, Bad terga are black in E. concinnus while the Hersfeld, Germany first two or three terga are red in E. kaufi- CAS California Academy of Sci- mani. Pulawski (1978) also described the ences, San Francisco, Cali- respective form of antennal flagellomere fornia, USA XI as a distinguishing character. Hartmann coll. Peter Hartmann, Bay- De Beaumont (1965) noted a broad geo- reuth, Germany graphic variation in color and punctation in Hauser coll. Martin Hauser, Univ. these species. He noted that all males of £. Illinois, Urbana, USA concinnus from Rhodos, Corfu, mainland MCZL Musee Cantonal der Zoolo- Greece, and Croatia have black terga (also gie, Lausanne, Switzerland Volume 9, Number 2, 2000 353 NHMW Naturhistorisches Museum, Wien, Austria Niehuis coll. Oliver Niehuis, Mar- burg, Germany OLL Oberoesterreichisches Lan- desmuseum, Linz, Austria SE coll. Christian Schmid-Eg- ger, Berlin, Germany SMNS Staatliches Museum fur Na- turkunde, Stuttgart, Germany ZMHB Zoologisches Museum der Humbold-Universitat, Ber- lin, Germany ZMUM Zoological Museum of the Moscow State University ZSM Zoologische Staatssammlung, Muenchen, Germany Technical terms. — The morphological ter- minology employed in this paper follows Bohart and Menke (1976). KEY TO SPECIES OF ENTOMOSERICUS Males 1. Flagellomere XI shovel-like, flattened (ventral surface is concave), narrower than more based flagellomeres, curved downwards (Fig. 3). Middle flagellomeres with ill-defined lateral tyloids (Fig. 3). Sternum VIII nearly flat, without lateral thickening, with long setae (Fig. 7). Inner (transparent) appendix of gonostylus twice as wide as outer (opaque) ap- pendix, the latter with long setae (Fig. 6). Disk of sternum II with evanescent punctures that are 4-6 diameters apart, sternal surface shiny, unsculptured. Central Asia kaufmani Radoszkowski - Flagellomere XI excavated beneath, nearly as wide as more basal flagellomeres. Flagellom- eres with well-defined central tyloids (Fig. 1, 2). Sternum VIII uneven, with lateral thick- ening, asetose or with short setae (Fig. 10, 11). Inner (transparent) appendix of gonostylus half as wide as outer (opaque) appendix (Fig. 4), the latter asetose. Disk of sternum II with more densely arranged punctures that are 0.5-3 diameters apart. Eastern and southwest Mediterrean area 2 2. Flagellomeres VIII-XI reddish. Lateral thickening of sternum VIII with an oval concavity (Fig. 11). Sternum II unsculptured between punctures, with short setae. Tyloids on flagel- lomeres VII-VIII well-defined, flagellomere VI with pointed tyloid, flagellomere IX with ill-defined lateral tyloid (Fig. 1). At least tergum I red (southern Turkey to Israel, northwest Africa) or terga black (Europe to Turkey). France?, eastern Mediterranean area, southwest central Asia, northwest Africa concinnus Dahlbom - Flagellomeres black. Lateral thickening of sternum VIII evenly convex (Fig. 10). Sternum II microsculptured between punctures, nearly asetose. Tyloids on flagellomeres VII-VIII smaller, flagellomeres VI and IX both with a minute, ill-defined tyloid (Fig. 2). Terga black. Southern Turkey, Israel, Jordan hauseri Schmid-Egger, new species Females 1. Shiny area between lateral ocellus and eye extending to eye margin (Fig. 12). Clypeal free margin with five distinct teeth, median teeth prominent, distinctly larger than lateral teeth (Fig. 15). Sternum II with fine and scattered, similary sized punctures, that are many di- ameters apart. Terga I and II red. Central Asia kaufmani Radoszkowski - Shiny area between lateral ocellus and eye smaller, and not extended to eye margin, or absent (Fig. 13, 14). Medial teeth of the clypeal free margin less prominent, not distinctly larger than lateral teeth (figs 16, 17). Sternum II with uneven sized punctures that are unevenly distributed and about 1-3 diameters apart. Terga I and II red or black. Eastern and southwest Mediterrean area 2 2. Shiny area between lateral ocellus and eye essentially absent (Fig. 14). Gaster black. Punc- tation of sternum II denser than in concinnus, punctures 0.5-1.5 diameters apart. Surface 354 Journal of Hymenoptera Research of sterna more densely microsculprured, not shiny. Tergum II without a distinct step, or edge, at the base of the apical tergal depression. Apical tergal depression of tergum II markedly broader medially. Southern Turkey, Israel, Jordan hanseri Schmid-Egger, new species Shiny area between lateral ocellus and eye larger, about as large as lateral ocellus (Fig. 13). Gaster red basally (at least tergum I red in specimens from southern Turkey, Israel, Jordan) except all black in specimens from Rhodos Island (Greece) and western Turkey. Punctures of sternum II less dense arranged, separeted by 1-3 diameters, interspaces shiny. Tergum II usually with a distinct step, or edge, at the base of apical tergal depression. France?, eastern Mediterranean area, southwest central Asia, northwest Africa . . concinnus Dahlbom Entomosericus Dahlbom Entomosericus Dahlbom 1845: 486. Type species Entomericus (sic) concinnus Dahl- bom 1845, by monotypy. Spelled Entho- tnosericus by Dahlbom, 1845, on Tabula Examinationis Synoptica Generum Nys- sonidarum. Entomosericus hauseri Schmid-Egger, new species (Figs. 2, 4, 5, 9, 10, 14, 16, 18) Diagnosis and discussion. — The male of Entomosericus hauseri is easily distin- guished from concinnus by its different form of sternum VII. Also, the tyloids and setae of last sternum differ from those of concinnus. The female closely resembles concinnus, but punctures are denser than in that species. Also, both sexes are differ- ent in color from concinnus in the areas where they occur sympatrically. £. hauseri is black, while concinnus has red terga. Only in Turkey might black males of both species occur together. Male. — 9-11 mm. Body black, tarsal seg- ments and lateral parts of tergum I red- dish. Head densely punctate, including area on each side of lateral ocellus (cf. Fig. 14). Clypeal free margin with five small teeth, outer tooth minute, nearly rounded. Inner margin of eye and clypeus with long setal fascia. Face with long erect setae. An- tenna black, with small, oblong tyloids on venter of flagellomeres VII-VIII and point- form tyloids on flagellomere VI and IX (Fig. 2). Tyloids smaller and less distinct than in concinnus. Thorax and upper part of head with long silver setae. Thorax in- cluding scutellum and propodeum dense- ly punctate, interspaces shiny. Lower me- sopleuron inpunctate, shiny. Gaster and legs only with short setae. Terga I and II without microsculpture, punctures 0.1-1 diameters apart. Terga III-VII rugulose. Sterna microscopially punctate, interspac- es microsculprured, sterna II and III with dense, coarse punctation, punctures 0.1-1 diameters apart. Disk of sternum II of smaller specimens nearly unsculptured. Sternum VIII: Fig. 10. Setal fascia at distal margin of sterna III and IV small, setae very thin and short (half as long as in con- cinnus). Sterna VI and VIII (in profile) with very short, erect setae (Fig. 9). Genitalia: Figs. 4, 5. Inner translucent appendix of gonostylus half as wide as outer (opaque) appendage. Wing venation and stigma dark brown, costal venation yellowish, wing membrane slightly darkened, yel- lowish-brown. Female. — 9-10 mm. Body black, with apex of tergites, sides of tergum I and tar- sal segments slightly reddish. Punctures denser and larger than in concinnus. Head li^s. 1-6. Entomosericus spp., male. 1-3, antenna! flagellum; a: ventral view; b: lateral view. 1, E. concinnus. 2, E. hauseri. 3, E. kaufmani. 4-6, gonostylus, lateral view. 4, /.'. hauseri. 5, E. hauseri. 6, /.'. kaufmani. Volume 9, Number 2, 2000 355 1a 1b 2a 2b 3a 3b 6 356 Journal of Hymenoptera Research 12 IV U — -~^^ V\N^^W% „,/<'^ #*iD 10 11 15 16 17 Figs. 7-17. Entomocericus spp. 7-9, male sterna VI- VID, lateral view. 7, E. kaufmani. 8, £. concinnus. 9, E. hauseri. 10-11, sterna VII-VIII, ventral view. 10, E. hauseri. 11, E. concinnus. 12-14, punctation of female between eye (left) and lateral ocellus. 12, E. kaufmani. 13, E. concinnus. 14, E. hauseri. 15-17, clypeal free mar- gin of female. 15, E. kaufmani. 16, E. hauseri. 17, E. concinnus. densely punctate, including area on each side of lateral ocellus (Fig. 14). Clypeus basally minutely, densely punctate, api- cally with longitudinal ridges (Fig. 16). Scutellum smooth and shiny basally, coarsely, longitutionally ridged apically. Punctures of tergum 1 1-2 diameters apart, without punctures on disk. Punctures of terga II-V 2-3 diameters apart, punctures half as wide as in concinnus. Tergum II without a distinct step, or edge, at the base of the apical tergal depression. Apical ter- gal depression of tergum II markedly broader medially. Punctures of sternum II 2-3 diameters apart. Sterna II-VI finely and densely microsculptured, basolateral corner finely, densely punctate. Wings ve- nation dark brown, wing membrane slightly darkened. Habitat. — In northern Israel, hauseri was found in the Mediterranean climate area with limestone substrate. The specimens were flying near the ground in grass on small paths and were collected by sweep- ing plants. There were unused fields and cattle pasture nearby. Etymology. — The species is dedicated to Martin Hauser, a friend and specialist of Stratiomyidae (Diptera), who supported the expedition to Israel in 1996. Geographic distribution (Fig. 18). — Israel and Jordan to southern Turkey. Type material— HOLOTYPE male: IS- RAEL, 40 km NE Haifa, 1 km E Hurfeish, 33.01°N 35.21° E, 16.05.1996 leg. SE (ZSM). PAPvATYPES: ISRAEL: 40 km NE Haifa, 1 km E Hurfeish, 33.01°N 35.21° E, 16 May 1996 8 males (SE, Hauser); 15 km E Qiryat Shemona, Foothill of Hermon, 33.15° N 35.44° E, 17 May 1996 2 males (SE). JOR- DAN: Jarash 1 May 1996 1 male (OLL); North Shuna 30 April 1996 1 female (OLL). TURKEY: Mardin, 23 May 1988 2 males (SE)- Urfa, 3 June 1988 1 female (MCZL); 25 km E Golbasi 7 June 1998 6 males 11 females (OLL); 10 km NW Ga- ziantep 20 June 1997 1 male (OLL); N. of Akseki, 19 June 1998 7 males (OLL); Tuz- lagozu (Baykan) 4 June 1998 1 male (OLL); Kahraman Maras, 40 km SE, 10 June 1998 19 males 1 female (OLL); 30 km N Erdem- li, Aslani 17 June 1998 1 male 1 female (OLL)— Gaziantep, Nizep 27 May 1978 1 male (CAS); Kuzuzcebelen/Mersin 25 May 1998 1 male (OLL). Entomosericus concinnus Dahlbom 1845 (Figs. 1, 8, 11, 13, 17, 19) Entomosericus concinnus Dahlbom, 1845: 486, male, Holotype or syntypes: Greece, Rhodos Island (Lund) Entomosericus concinnus rufescens Beau- Volume 9, Number 2, 2000 357 ♦39 00 .37 00 |CFF 1.2 - 1996 .35 00 .33 00 ♦27 00 ♦31 00 ♦33 00 +35 00 ♦37 00 ♦41 00 ♦45 00 Fig. 18. Collecting localities of Entomosericus hauseri in the eastern Mediterranean. mont 1950: 403. Holotype: female Al- geria, Taouiala (MCZL). New Syno- nym. Diagnosis and discussion. — The male of this species is easily characterized by the form of sternum VIII and the flagellom- eres. Females may be distinguished from the other species by punctation and other characters (see key). Entomosericus concin- nus has been confused with kaufmani. However, concinnus occurs only in the western Palearctic, whereas kaufmani is a central asian species. The ranges of the species overlap in the Ural river area in northern Kazakhstan and Russia. £. con- cinnus has two color forms. In Europe and northern and central Turkey the male gas- ter is all black, while in southern Turkey, Syria, Israel and northwest Africa the gas- ter base is red. Females of the species have a red gaster base except in specimens from Rhodos (the type locality) and western Turkey, where the gaster is all black. The isolated color form in the Dahlbom's type material might have caused confusion for earlier authors (Handlirsch 1888). Females with an all black gaster have not been found in other areas. The subspecies con- cinnus rufescens Beaumont, which was de- scribed from northwest Africa, is only a light color form of the species. It differs slightly from typical concinnus (see below), which itself is variable in the extend of gaster coloration in the eastern Mediter- 358 Journal of Hymenoptera Research 9 concinnus A kaufmani +5300 +4800 +4300 +38.00 ♦33 00 +28 00 -200 +3 00 +8 00 +13 00 +18.00 +23.00 +28 00 +33 00 +38 00 +43 00 +48 00 +53.00 +58 00 +63 00 +68 00 +73.00 +78 00 +83 00 Fig. 19. Collecting localities of Entomosericus concinnus and E. kaufmani in southern Europe, western and central Asia and northern Africa. ranean area. Males of concinnus concinnus from Syria and Israel are similar to males of concinnus rufescens from north-west Af- rica, while females of concinnus rufescens have a more light red gaster than females of concinnus concinnus. Therefore, the sta- tus of rufescens as a subspecies of concinnus is not justified due to the observed vari- ability in concinnus. Male. — 9-12 mm. Body black, or with red basal terga. Tarsi red. Head densely punctate, shiny area between eye and lat- eral ocellus present (cf. Fig. 13). Clypeal free margin with five small teeth. Upper part of face with long erect setae. Dorsal side of flagellomeres VIII-XI, surface of flagellomere X-XI red. Flagellomeres VII- VIII with large and well-defined tyloids; flagellomere VI with small, pointed tyloid; tyloid at flagellomere IX indistinct, ventro- laterally situated (Fig. 1). Body sparsely punctate. Thorax with long, silver setae. Thorax surface and mesopleuron with shiny interspaces, punctures 1-2 diame- ters apart. Terga I-II coarsely punctate, punctures 0.2-1 diameter apart. Terga III- VI finely, densely punctate. Disk of ster- num II shiny, coarsely and densely punc- tate, punctues 1-3 diameters apart. Ster- num II laterally in specimens form Tur- key, Europe with fine, dense punctation; in specimens from southern Turkey, Isra- el, Morokko with the same punctation as disk. Sterna III-VI more or less micros- culptured, finely, densely punctate. Lat- eral thickening of sternum VII with an oval concavity (Fig. 11). Lateral edge of sternum III with long band of brown se- tae. Sternum VIII (in profile) with dense erect setae as long as fourth tarsal segment of hind leg, sterna V and VI with erect se- tae (Fig. 8). Genitalia similar to that of hau- seri spec. n. Wings venation and stigma dark brown, costal and basal venation yel- lowish, wing membrane slightly darkened yellowish brown. Female. — 10-11 mm. Body black, or with red basal terga. Tarsi partly red. Head densely punctate, area beside lateral ocel- lus smooth and shiny (Fig. 13). Clypeal free margin with five rounded teeth (Fig. 17), often worn down. Clypeus apically longitudinally ridged. Mesoscutum with smoth spaces between longitudinal punc- tures. Punctation of thorax variable, but punctures of mesoscutum at least many diameters apart. Anterior part of scutel- lum smooth, posterior part longitudinally wrinkled. Mesopleuron densely punctate in upper part, punctures in lower part Volume 9, Number 2, 2000 359 many diameters apart. Terga I-III distinct- ly and densely punctate, punctures 0.2-1 diameter apart. Base of terga III-IV with denser punctation than the apex. Apical margin of tergum II variable: in specimens from Europe and Turkey it is straight (or barely emarginated) and strongly stepped with keel basally (a step between disk of tergum II and apical tergal depression); in specimens from Israel and Jordan it is slightly rounded; in specimens from Mo- rocco and Algeria it is barely stepped, with a rounded basal edge. Sternum II with a few deep punctures of variable di- ameter (1-2 diameters apart), with fine mi- crosculptured between punctures. Wing venation dark brown except with yellow- ish basal venation in specimen not from southern Turkey. Variation. — Males and females from Mo- rocco differ slightly from Turkish and Greek specimens: Diameter of punctures of the terga and sterna are half as wide. The apical margin of tergum II is barely dented with a rounded flat basal edge. Color variaton. — Europe, northern and central Turkey, Russia and Ukraine: Males with all terga black; females with tergum I and II red. Rhodos island (Greece) and Western Turkey (Aegean coast): Males and females with all terga black. Syria, Is- rael, southern Turkey and Jordan: Males with tergum I red and tergum II black with red parts; females with terga I-II red. Northwest Africa: Males with tergum I and base of tergum II red. Females with terga I-III and base of tergum IV red, some specimens with tergum IV black. Doubtful specimen. — A single female from Jordan (20km S North Shuna, Tall al Arbatin 19 April 1998, coll. SE) could not assigned to species. It has a black gaster (normally gaster base is red in Jordan), a shiny area with only few punctures be- tween the lateral ocellus and the eye, the tergum II is as in average concinnus. A male of concinnus was collected on the same day at the same locality. Geographic distribution (Fig. 19). — East- ern mediterranean area to southern Russia and northwest Africa. Records.— ALGERIA: Taouiala [33'54'N 01 '51 'EL Holotype of E. concinnus rufescens (de Beaumont 1950, in British Museum London). Not seen, but other spec- imens of type series examined. ARMENIA: Urcadzor 14 June 1988 1 male (gaster black) (OLL). BULGARIA Melnik 23 May 1985 3 males 22 June 1997 2 males (OLL); Slancev Brjag 15 June 1997 1 male (OLL); San- danski 14 July 1979 2 females (CAS); Slancev Brjag July 1972 1 female 2 males (OLL) (males: gaster black, females: terga I and II red); Albena 17 July 1978 1 female (OLL); Ivanski near Schumen 15-30 July 1969 1 male (coll Burger).? FRANCE: Gallia mer., Ancev, 1897 1 male (NHMW, gaster black). Specimen prob- abely mislabeled. GREECE, MAINLAND: Alexandro- pulos 16 June 1994 1 female (NHMW); Alt Korinth 28 May 1996 1 male; 7 June 1997 2 males 1 female (Arens); Korinth 15 May 1970 (ZMHB); Attika 1 male (NHMW); Cephalonia 4 males 1 female (NHMW) (det Maidl as kaufmani); Graecia, Argostoli? 26 Mav 1908 1 male 1 female (NHMW) (female det Maidl als kaufmani); Attika 1 female (NHMW) (det Maidl als kaufmani); Olympia 10 June 1961 1 female (CAS); Olympia, Alfios-Tal 4 June 1995 1 male (Arens); Ka- lamata, Avia 14 May 1995 2 females; 10 June 1996 2 females (Arens); Peloponnes, Midea 19 June 1996 1 female (Arens); Peloponnes, Sparta, Amyklai 5 June 1996 1 female 1 male (Arens); Peloponnes, Sparta, Menelaion 26 May 1997 3 males; 21 May 1997 2 males 1 female; 4 June 1996 6 males 6 females (Arens); Pe- loponnes, Sykion 8 June 1997 2 females (Arens); Olympia 19 June 1961 1 female; 1 June 1963 1 female 1 male (MCZL); Pyrgos 11 June 1961 3 females (MCZL); Kalamata 15 Mav 1964 2 females 5 males (MCZL); Nea Kefissia 19 June 1957 1 female 2 males (MCZL) (males: gaster black, females: terga I and II red).; GREECE, RHODOS ISLAND: Kritinia, 2 May 1990 9 males 1 female (SMNS) ; Salakos, Kamiros 3 May 1990 2 males 2 females (SMNS) ; Rhodos, no specific locality, June 1939 3 males (MCZL); Rhodus, no specific locality, 1869 4 males 2 females (NHMW); W Apolakkia, N Monolithos 23 April 1998 1 male 1 female (Hartmann) (gaster black). HUNGARY: male, without date and exact locality, may be today' s Ro- mania (ZMHB, also referred by Handlirsch 1895: 850 gaster black); 'Ungarn coll. Hindlmayer' 1 male (ZSM). ISRAEL: Lehavim junction, 11 km N IVVr Sheva, 27 March 1991, 2 males (SE) (gastral base red ). JORDAN: Petra 14 May 1995 3 males (OLL) (gastral base red); North Shuna 29 April 1996 1 female (OLL); 20km S North Shuna, Tall al Arbatin 19 April 1998 I male (OLL) (gastral base black). MOROKKO: Agadir, 15 April 1947 1 male 1 female (MCZL, Paratype of / c. rufescens de Beaumont): Iftilt, 15 June 1962 1 male (MCZL, Paratype of / c. rufescens lie Beaumont); 50 km NE Taroundannt, 5 km E Kreuzung Tizn-Test/ Aoulouz 12 April 1996 4 females (Niehuis, SE); 10 km 360 Journal of Hymenoptera Research N Rich, 23 May 1995 1 female (OLL); 10 km S Bouarfa 20 May 1995 1 female (OLL); 40 km S Guercif 15 May 1995 1 female (OLL). ROMANIA: Dobrogea, Camar- aua Fetii 26 June 1993 3 males (OLL) (gaster black); Tultscha [= Tulcea] 1859 1 female (leg. Mann, NHMW, terga I and II red); Mehadia [leg] Mann 1859 2 females (NHMW, gastral base red). Scobiola-Palade (1966) mentions E. concinnus and E. kaufmani from Ro- mania, but the latter is probably the red form of E. concinnus. RUSSIA: Southern Ural, Kargala b. Oren- burg, 1915-1917 1 male (ZMHB, gaster black). SYRIA: Mezze near Damascus 21 May 1954 2 males 3 females (CAS, MCZL); Damascus, road to Kissoue 2-18 May 1960 2 males 1 female (MCZL); Marbi, 9 May 1996 1 male (OLL); 30 km s Suwayda, Dibbin 15-17 May 1996 1 male; 10 km SE Suwayda, Kafr 19 May 1996 1 male (OLL); Anata, 50 km SE of Suwayda 20-21 May 1996 1 female (OLL) (gastral base red). TURKEY: First terga of males black, tergum I and II of fe- males red: Horasan, 18 km E Delibaba 25 June 1993 2 males (OLL); 30 km N Kutahya, Porsuk Baraji 15 June 1997 2 males (OLL); Guriin 7 June 1970 1 female (CAS); Konya, Karaman 11 June 1979 1 male (CAS); Urfa 21 May 1972 1 female (CAS); Sille, Konya, 16 June 1968 1 male (MCZL); Amasya 1,400 ft, 9 June 1959 1 male (MCZL); Asia minor 11 July 1852 1 fe- male (ZSM); Osmaneli 14 June 1997 3 males 1 female (OLL); 40 km E Mut, Cornelek 29 May 1996 4 males 1 female (OLL); Capadocia, Urgup 13 June 1998 8 males 1 female (OLL); Capadocia, 10 km NW Urgup 15 June 1998 1 male 1 female (OLL); Tuzlagozu (Bay- kan) 4 June 1998 1 male (OLL); Bozkir 26 May 1998 1 male (OLL); Agri 27 June 1993 1 male (OLL); G6- reme 23 June 1993 1 male (OLL); Konya, 30 km S of Aksehir 24 June 1998 1 male (OLL); 20 km SE Hora- san, Delibaba 3 July 1997 1 female (OLL); Bolu, 17 km 5 Seben 17 June 1998 1 female (OLL); Manisa, 30 km E 20 June 1998 1 female (OLL); Hop Gecidi, Mardin 6 June 1998 1 female (OLL); Ankara, 40 km W Ayas 26 June 1998 1 female (OLL); Sivas, 45 km E Yarhisar 24 June 1993 1 female (OLL); Taskesigi/10 km E An- talya 1998 18 males (OLL); Gevas/Van Golii 29 June 1993 1 female (OLL); Terga I and II of males red: Elazig 7 June 1980 1 male (SE); Urfa 20 May 1967 1 male (MCZL) -Halfeti (Birecik) 31 May 1998 1 male (OLL); Tuzlagozu (Baykan) 4 June 1998 1 male (OLL); Halfeti 3-5 May 1994 1 male (OLL). Gaster of fe- males black: W-Turkey, SSO Milas, Camkoy lake 20 June 1998 1 female (Niehuis); W-Turkey, SO Milas, Akyoi 19 June 1998 1 female (Niehuis). UKRAINE: Falzfeinowo a. Dnipr 12 May-7 June 1914 1 male (gaster black) (ZMHB); Otuzysches Tal, auf Teucrium polium, leg. Wuczeticz, 4 July 1926 1 female (NHMW, det Maidl as kaufmani, tergum I and 2 red) (= prob- ably Otuzy Valley or Otuzskaya Dolina in Crimea, Ukraine); Umg. Tokluk, near Sudak, at Reseda lutea 4 June 1924 1 male Wuczeticz [leg] (NHMW, gaster black) (= probably Sudak in Crimea, Ukraine) Entomosericiis kaufmani Radoszkowski 1877 (Figs. 3, 6, 7, 12, 15, 19) Enthomosericus kaufmani Radoszkowski 1877: 46. Male, female. Misspelled 'kauf- manni' in most subsequent publications. Lectotype, male: Kasachstan, Kyzylkum [desert], 28 April 1871 (A.P.Fedchenko coll.) [appr. 43°13'N-71° 35'E]. Designed as lectotype by A. Antropov (ZMUM). Diagnosis. — The male of Entomosericus kaufmani is easily recognizable by its flat sternum VIII and its ventrally concave fla- gellomere XL Females have a large shiny area between the lateral ocellus and eye. The species ist smaller and more slender than concinnus. It occurs only in southern- central Asia from Kazakhstan to Turk- menistan (Kazenas and Alexander 1993). An isolated record comes from Uralsk in northern Kazakhstan. All examined 'kauf- mani' (det. Madl or Handlirsch, coll. NHMW) from Europe or Turkey are mis- identifications of the red colored form of concinnus. Male. — 8-9 mm, Body black, tergum I and parts of tergum II red. Legs red except tibiae. Face covered with a dense silver pubescence in lower part and erect setae in upper part. Area between eye and lat- eral ocellus only with few punctures near eye (cf. Fig. 12). Clypeal free margin with 5 teeth, median tooth and lateral teeth dis- tinctly smaller than mediolateral teeth. Flagellomere XI with a concave ventral surface witch is ventrally curved. Ventral surface of flagellomeres V-IX with indis- tinct lateral tyloids and with half moon shaped red spots. Ventral surface of fla- gellomeres X-XI completly red, flagellom- ere XI also partly red on dorsal surface (Fig. 3). Punctation of body less dense than in concinnus, spaces between punc- tures shiner than in concinnus. Punctures of terga I and II 0.5-2 diameters apart. Sterna II and III unsculptured with only few scattered punctures, that are separet- ed by many diameters. Disk of sterna IV Volume 9, Number 2, 2000 361 and V finely and densely punctate, later- ally only with a few coarse punctures. Sternum III with long band of setae (as in concinnus), medial setae of band half as long as lateral setae. Sterna VI and VIII (sometimes V) with long silver setae (Fig. 7). Sternum VIII nearly flat, without lateral thickening. Inner transparent appendix of gonostylus twice as wide as outer opaque appendage (Fig. 6). Wings venation and stigma dark brown except yellowish costal and basal venation. Wing membrane slightly darkened. Female. — 9-10 mm. Body, including legs, black, terga and sterna I-II red. Head finely and densely punctate (punctures 0.5-1 diameter apart). Clypeus apically in- distinctly longitudinal ridged, without punctation, basally finely punctate. Clyp- eal free margin with five well developed teeth. Shiny area between lateral ocellus and eye large, extending to eye, some- times with scattered punctures (Fig. 12). Thorax unsculptured, shiny, coarsely punctate. Upper part of mesopleuron densely, finely punctate, lower part only with few punctures. Punctures at disk of terga I-II 1-2 diameters apart, punctures of succeeding terga more densely arranged. Edge between disk of tergum II and apical tergal depression rounded, apical margin of tergum II slightly dented. Disk of sterna II-III nearly unsculptured, sternum II with only a few scattered punctures that are many diameters apart. Wings venation dark brown, wing membrane slightly darkened. Life history. — Kazenas and Alexander (1993) described the nest architecture and larva of kaufmani from southeastern Ka- zakhstan, Talas River. The females dig nearly vertical burrows in sandy soil and fill each cell with eight to 18 leafhoppers (Homoptera, Cicadellidae). Type material— PARALECTOTYPES (Lectotype see above): All designed as paralectotypes by A. Antropov. The type series is deposited in ZMUM and includes the following specimens: Zeravshan valley 9 May 1869 2 males (A.P. Fedchenko coll.) [appr. 39°33'N-63°40'E]; Zeravshan valley 23 May 1869 1 male (A.P. Fedchenko coll.); Chardara, 25 April 1871 2 males (A.P. Fed- chenko coll.) [41°15'N-67 °58'E]; Chardara 27 April 1871 1 female 1 male (A.P. Fed- chenko coll.); Kyzylkum [desert] 28 April 1871 1 female (A.P. Fedchenko coll.) [appr. 42°40'N-63°37'E]; Kyzylkum [desert] 1 May 1871 2 males (A.P. Fedchenko coll.); Syutkent, 3 May 1871 1 male (A.P. Fed- chenko coll.) [41°55'N-68°5'E]; Bayrakum [Baygakum] 4 May 1871 1 male (A.P. Fed- chenko coll.) [44°18'N-66°28'E]; Karak- skaya steppe 6 May 1871 1 female (A.P. Fedchenko coll.) [appr. 49°18'N-69°50'E]. Geographic distribution (Fig. 19). — South- ern-central Asia from Kazakhstan to Turk- menistan. Records.— KAZAKHSTAN: 2 females 1 male from coll Radoszkowski (ZMHB), [male without locality, females from Chardara and Kyzylkum (in russian let- ters), all labelled as 'Type', probably belonging to the syntypes serie, not designated as paralectotypes]; 10 km E Ddjambul 31 May 1994 6 males 3 females (OLL) -Darbaza 40 km N Tachkent 30 May 1994 4 males 8 females (OLL); Vanovka, 80 km E Djambul 1 male (OLL); Alma Ata 1 May 1994 1 female (OLL); 10 km N Chayan 1 male (CAS); vicinity of village Togusken on Talas River 1 female (CAS); vicinity of Uralsk 1 female (CAS). TADJIKISTAN: 3 km W Dusti, 130 km 5 Duchanbe 15-16 May 1991 1 female (OLL). TURK- MENIA: Sandikazi 3-13 May 1993 18 males 3 females (OLL); Askahbad 22 May 1964 1 male (CAS); Star. Nisa/Ashabad 28 April 1977 1 male (OLL). UZBEK- ISTAN: Samarkand 19-21 May 1994 ca. 200 males ca. 40 females (OLL); Czirczik 28 May 1994 14 males 5 females (OLL); 5 km W Ddjizak 23 May 1994 9 males 6 females (OLL); Djuma 1 male 1 female (CAS)- Sam- arkand: Chupan-Ata-Mountain 2 males 2 females (CAS)- Sary-Agach in Tashkent Distrikt 1 male (CAS). ACKNOWLEDGMENTS I greatly appreciate the help of Dr. Werner \rens, Bad Hersfeld, Germany, Frit/ Gusenleitner, Lin/, Austria, Dr. Frank Koch, Berlin /Germany, Dr. Peter Hartmann, Bayreuth, Germany, Dr. Stephan Schodl Wien, Austria, Erich Diller and Johannes Schubert, Miinchen, Germany, Martin Hauser, Urbana, Illinois, Oliver Niehuis, Albersweiler, Germany, Dr. Till Os- tein, Stuttgart, Germany, Dr. Wojdech Pulawski, San Francisco, California and Dr. Michael Sartori, Lau- sanne, Switzerland, tor the loan of specimens. Also, I extend my sincere thanks to Wojdech Pulawski tor 362 Journal of Hymenoptera Research his advice and suggestions, Alexander V. Antropov, Moscow, for his help, and Kevin Holsten, Urbana, Illinois for kindly checking the english. LITERATURE CITED Balthasar, V. 1972. Grabwespen; Sphecoidea. Fauna CSSR 20. 471 pp. Prag, Accidentia Verlag. Bohart, R. M. and A. S. Menke 1976. Sphecid wasps of the world. A generic revision. University of Cal- ifornia Press, Berkeley, ix + 695 pp. De Beaumont, J. 1950. Sphecidae Hymenoptera re- coltes en Algerie et au Maroc par M. Kenneth M.Guichard. Bulletin British Museum Natural His- tory Entomology, !'■ 391-727. De Beaumont, J. 1965. Les Sphecidae de la Grece Hym. Mitteilungen Schweizer Entomologische Ge- sellschaft 38: 1-56. De Beaumont, J. and P Roth 1967. Hymenoptera from Turkey. Sphecidae, I. Bulletin British Muse- um Natural History Entomology 19: 251-382. Dahlbom, A.G. 1843-1845. Hymenoptera Europaea praecipue borealia; formis typicis nonnullis Spe- cierum Generumve Exoticorum aut Extraneorum propter nexum systematicus associatis. Tomus Sphex in sensu Linneano. Offincina Lundbergiana, Lundberg fasc. 3: 353-528, 1845 dating after Men- ke 1974. Handlirsch, A. 1888. Monographie der mit Nysson und Bembex verwandten Grabwespen. II. Sit- zungsberichte der Kaiserlichen Akademie der Wissen- schaften in Wien, Mathemathisch-Naturwissenschaf- tliche Classe 96: 219-311. Handlirsch, A. 1995. Nachtrage und SchluGwort zur Monographie der mit Nysson und Bembex ver- wandten Grabwespen. Sitzungsberichte der Aka- demie fur Wissenschaften, Wien Abteilung 1, 104: 801-1079 + Tafeln. Kazenas, V.L. and B.A. Alexander 1993. The nest, prey, and larva of Entomosericus kaufmani Ra- doszkowski. Hymenoptera: Sphecidae. journal of Hymenoptera Research 2: 221-226. Menke, A. S. 1974. The dates of publications of A.G. Dahlbom's Hymenoptera Europaea, vol. 1, Pol- skie Pismo Entomologia 44: 315-317. Menke, A. S. 1997. Family-group names in Sphecidae Hymenoptera: Apoidea. Journal of Hyttienoptera Research 6: 243-255. Melo, G.A. 1999. Phylogenetic relationships and clas- sifications of the major lineages of Apoidea Hy- menoptera, with emphasis on the Crabronid wasps. Scientific Papers. Natural History Museum of the University of Kansas 14: 1-55. Pulawski, W. 1978. Nadsem. Sphecoidea, p. 173-279 in G.S. Medvedev Editor. Opredelitel' naseko- mykh yevropeyskoy chasti 5SR, Tom III. Pere- ponchatokrylyye, Pervaya chast' [Keys to the Identification of insects of European USSR, Vol. 3, part 1]. Nauka, Leningrad 584 pp. Radoszkowskii, O. 1877., ,Sphegidae„ in: Voyage au Turkestan d'A.P. Fedchenko, fasc. 14, tome 2, partie 5. Bulletin societe Imperialis Amis Sciences Naturalis 26: 1-87. Scobiola-Palade, X. 1966. Donnees nouvelles concer- nant les Hymenopteres du delta du Danube. Tra- veaux Musee Histoire naturalis Griorgi Antipa 6: 389-396. J. HYM. RES. Vol. 9(2), 2000, pp. 363-369 New Odour Glands in Xylocopa Males (Hymenoptera: Apoidea: Anthophoridae) B. O. SCHLUMPBERGER AND D. WlTTMANN Institut fur Landwirtschaftliche Zoologie und Bienenkunde, Universitat Bonn, Melbweg 42, 53127 Bonn, Germany Abstract. — Odour glands were found in all tergites in males of the neotropical Xylocopa bima- culata Friese and X. nigrocincta Smith as well as in the palaearctic X. violacea Linne. Furthermore odour glands were found in the fore, middle and hind legs of X. bimaculata and X. nigrocincta. Males of X. bimaculata and X. nigrocincta establish non-resource based territories to which they attract females by spreading secretions of the odour glands over their body. For this they brush with specialized hairs on their hind legs over the abdomen, which they extend in order to expose the pores of the odour glands. The function of glands in all three pairs of legs is discussed in comparison with similar findings of odour glands in other species of carpenter bees and leafcutter bees. Territorial male carpenter bees that ac- tively search for females establish two kinds of territories: at nesting sites or at flowering plants (Hurd and Linsley 1975, Alcock 1991). Such territories may cover an area of several square meters. Other males attract females with sex phero- mones to territories which contain neither nests nor food plants, so called non-re- source based territories. These territories may cover a space less than one cubic me- ter, in many cases they measure 10 to 15 cm'. In some species a single male estab- lishes its territory alone, for example in a tree. In others conspecific males establish their non-resource based territories to- gether in close vicinity, forming so called leks. In these cases the distance between single territories may be as short as one meter. Males in non-resource based territories have to cope with two problems: they have to leave their territories in order to find nectar as energy for their long lasting territorial flights. In Xylocopa nigrocincta males dehydrate the nectar that is fed to them by the mother in the nest. Thereby they get rid of excess water, and with a higher concentration of sugar they im- prove their energy budget. Of males with the same amount of sugar in their crop those fly longer in their territories which have diluted the sugar in small volumes of water (Wittmann and Scholz 1989). Furthermore their success in attracting a female depends on the amount of pher- omones secreted per time. This may be one of the reasons why males establish their territories close to each other in leks. The advantage of such leks is probably that together the males have better chanc- es to attract a female. Within the genus Xylocopa, odour glands have been described from the head (mandibular gland) and from the thorax (mesosomal gland) (Wheeler ct al. 1976, Vinson et al. 1986). Furthermore odour glands were found recently in the fore legs of some Old and New World Xylocopa males (Wittmann and Blochtein 1995). Males of X. hisutissima and X. sulcatipes have been reported marking their mating territories with the secretions of the man- dibular glands (Velthuis and Camargo 1975 a & b, Hefetz 1983). Territorial males can identify conspecific males as intruders of 364 Journal of Hymenoptera Research their territory by the secretions of the man- Compounds were identified by their mass dibular glands. In consequence this odour spectra and using Kovats indices, elicits defence behaviour in the owner of the territory (Velthuis and Camargo 1975 a & RESULTS b). Mesosomal glands are male specific. The Abdominal glands. — SEM analyses of the males frequently brush their legs over their sternites and tergites of X. bimaculata re- body thereby spreading the odour (Vinson vealed that pores of odour glands are pre- et al. 1986). The secretions of these glands sent in tergites I — VI. These pores are lo- are supposed to act as territorial phero- cated on the frontal part of the tergites mones in non-resource based territories which are hidden under the anterior ter- (Gerling et al. 1989). gites (Fig. la). They are the openings of chitinous ducts which lead into glandular MATERIAL AND METHODS cdls in the abdomen. After maceration All observations on behaviour were these ducts are visible on the inner surface made between September and November of the tergites. In tergites I-V we found the 1995 in the natural habitat in the forest highest amount of pores. On tergite I and reservation area Pro-Mata of the PUC- II the pores were scattered, while on ter- Uni versify Porto Alegre (Brazil). The area gites III, IV and V most of the pores ap- is located in the northern highlands of Rio peared clumped. Between 10 and 20 pores Grande do Sul, the Serra Geral (50°-51° W were grouped in round areas with a di- and 29°-30° S) in an elevation of ca. 900 ameter of 8-10 fjim (Fig. lb). The distances m, about 150 km north of Porto Alegre. between these areas ranged between 20 For identification we used the key to and 40 \xm. On the inner side of the tergite subgenera by Hurd and Moure (1963). these groups of pores correspond with a Furthermore the bees were compared with bundle of cuticular ducts of odour glands the collection of the Biological Research (Fig. 2a). The terminal part of each duct is Station of the University of Tubingen / covered with short lateral ducts. This end Germany at the PUC-University in Porto apparatus is normally inside the glandular Alegre, Brazil. Critical species were iden- cell that has been macerated (Fig. 2b). In tified by Dr. J. S. Moure. tergite VI only a few scattered pores were To analyse the territorial behaviour found, mostly on the lateral parts of the males were filmed during the territorial tergite. flight with a Panasonic F15. Description of hairs on the abdomen. — For SEM-analysis we used a Stereoscan Noteworthy are three different types of 250 Mk2 and a Hitachi S-800. The bees, or hairs on the tergites. On those parts of the parts of them, were macerated in 5 % tergites that are covered by the anterior KOH for 24 hours, after which they were tergites the hairs are plumose and about dehydrated in 50-100 % ethanol and dried 200 (xm long. The hairs on the posterior for 24 hours at 30°C. If glands with chitin- part of the tergite are unbranched and be- ized ducts are present, these structures re- tween 0.5 and 1 mm long. Most conspic- main after the mazeration. uous are the bunches of bristles on the To check the presence of volatile sub- sides of the last three tergites. stances in the area of glandular pores we Abdominal glands in males of other Xylo- washed the abdomen of a freshly killed ter- copa species. — We also found pores of ritorial male in pentane. The samples were odour glands in all tergites of X. nigro- analyzed with a Fisons MD-800 GC-MS on cincta and in X. violacea. In both species a fused silica column, DB-5 (15 m X 0.32 single pores have also a diameter of 2 (xm. mm), the temperature was programmed However, they are not arranged in groups from 80°C (for 2 min) to 200°C at 10°C/min. as in X. bimaculata. The chitinous ducts of Volume 9, Number 2, 2000 365 Figs. 1-2. 1, Outer surface of X. bimaculata tergite IV: la, The median of the tergite is on the right borderline, the pores are hidden under the dense coverage of short hairs; lb, Aggregation of 12 pores of odour glands. Each of the pores has a diameter of about 2 u.m. 2, Inner side of the tergite IV: 2a, Bundles of cuticular ducts of odour glands, diameter of each duct is ca. 2 jjum. 2b, Cuticular duct and the terminal apparatus. Scale bars: la) 1 mm, lb) 10 (xm, 2a) 20 (xm, 2b) 40 ^m. the odour glands have the same appear- ance in all three species (Fig.3a & b). Glands in legs. — On the fore, middle and hind legs of X. bimaculata we found pores of odour glands. On the fore legs the pores were on the dorsal and anterior side of the basitarsus and also on the ventral side of the other tarsalia of the fore leg. On the middle leg pores were found on the dorsal side of the basitarsus, while on the hind leg the pores were found on the ventral side of the basitarsus. In X. nigrocincta pores were also found on the basitarsus of the fore leg. Hairs on the fore legs. — A notable feature on the fore leg of X. bimaculata is a specific type of hair. When the male is viewed frontally the hairs of the anterior side of the tarsalia and parts of the tibia appear shining white. The front leg bears on the anterior dorsal edge of the tarsalia a fan of bristles which are about three times as long as the diameter of the basitarsus. Such shiny white hairs are also present on the tarsi of the middle legs. Hairs on the hind legs. — On the ventral side of femur and tibia of the hind legs we found a conspicuous type of hairs. They have smooth shafts and their tips are broadened and flattened forming a con- cave spatula (Fig. 4a). The concave side points to the surface of the cuticula. Be- sides these hairs the femur, tibia and the tarsi bear long pointed bristles with a 366 Journal of Hymenoptera Research Figs. 3-4. 3, Chitinous odour ducts (with a diameter of ca. 2 |xm). 3a, on the inner side of tergite I of X. nigrocincta; 3b, on the inner side of tergite II of X. violacea. 4, Hairs on the hind legs of male X. bimaculata. 4a, the widened and flattened tips form a concave spatula. 4b, hairs on the anterior side of the femur. 4c, hairs on the anterio-ventral side of the tibia. Scale bars: 3a) 40 |xm, 3b) 100 (xm, 4a) 40 |xm, 4b) 200 |xm, 4c) 200 pmi. rough surface. On the femur the spatula- long axis (Fig. 4c). The function of this po- like hairs are directed rectangular to the sition of the hairs on the hind leg becomes long axis of the femur (Fig. 4b) whereas clear when we look at the behaviour of X. on the tibia they are directed parallel to its bimaculata males in their territories. Volume 9, Number 2, 2000 367 Territories. — The mating period of X. hi- landed were observed to brush their legs maculata is from early October until late over their body. November. Due to the high altitude of the While on wing in their territory the study site there is only one mating period, males held their fore and middle legs close Males were recorded to fly in their terri- to the body, so that the shiny white hairs tories between 06:15 and 16:30, the mini- on the fore and middle leg basitarsalia di- mum temperature was 18°C. Individual rected downwind and forward. The hind territorial flights lasted up to 1.5 hours. legs were stretched out backwards and The territories of these males were al- away from the body. With high frequency ways found at the very margins of the ar- (up to 11 times per minute) the males aucaria forest in non-flowering trees or brushed with their hind legs from anterior shrubs. The males established their terri- to posterior over the dorsilateral parts of tories always on the sunny side of the for- the abdomen, bending the abdomen ests and shifted them according to the po- downwards. They then rubbed the hind sition of the sun. They flew in territories legs to each other and then to their middle with a radius of ca. 30 cm, sited between legs. Less frequently they rubbed the mid- twigs or close to a bough, always in the die legs to the fore legs and rarely the fore shadow, positioned between 0.5 and 4 me- legs to the head. Towards the end of the ters above the ground. The males kept territorial flight the males carried out their head downwind and changed their these brushing movements with an ever direction (not the position) according to decreasing frequency, the wind. Marked males were found to oc- During the observation of 167 territories cupy the same territory on different days, not a single female approached a male in However, this was not the rule. its territory. In 1995 we observed 167 territories, 74% Males of X. bimaculata regurgitated and of them as single territories, in 26% 2-A dehydrated nectar while hovering in the males established territories in close vicin- territory. ity in the same tree. In 1994 we found leks Aggressions between males. — Males in with up to 10 males. The number of males neighbouring territories have sometimes in a lek changed frequently often within a been observed to suddenly attack their few minutes. Sometimes the distance be- neighbour. Some incoming males attacked tweeri the territories measured only 50 cm. territorial males immediately or were at- Territorial behaviour. — Within their terri- tacked by the hovering male. Sometimes tories the males of X. bimaculata hovered the defending male left the territory to flv with loud buzzing mostly at one position, towards the intruder and hovered in front only shifting the direction from time to of him until one of them started to show time to stay with the abdomen in the up- antagonistic behaviour, including hits wind position or in search for approach- with the front legs, tumbling down while ing females. Sometimes a male moved to clinging to each other or chasing the op- establish the territory in another place ponent until both got out of sight of the without any recognizable reason. Males observer. that left their territory probably to feed re- Finally one male returned to the former turned to their former or to a different po- territorial position, or in some cases even sition. In about 5% of the observed terri- both males started to hover in close vicin- torial flights males landed on the substrate ity. Aggressive behaviour could be artifi- within the territories. However, we could daily initiated when we approached a never observe them to rub their mandi- dead male closer than 30 cm to a territorial bles, their abdomen or legs over the sub- male, strate. Some of these males which had Volatile substances in cuticular wash- 368 Journal of Hymenoptera Research ings. — In cuticular washings from the dor- Those substances may more likely serve as silateral parts of the abdomen we could short range signals, maybe for mate accep- determine a series of alkanes with a chain tance in a female choice system or to de- length of 18 and longer as well as alkenes tect the males at the margins of the forests, with the same chain length. Further studies on the chemical properties of the gland secretions and their function uio^uajivji\ during territorial and mating behaviour Males of X. bimaculata were found to es- are necessary, tablish non-resource based territories, ei- Unfortunately we could not observe ther alone or together with other males in copulations in X. bimaculata. Therefore, leks. We could not observe that they mark any further considerations on the function substrate in their territories in order to at- of the modified forelegs can only be hy- tract females. Instead males were seen to pothetical. brush their body with their legs. We sup- Anzenberger (1977) clearly observed pose that while the hind legs brush over that during copulation males of X. (Meso- the abdomen they take up secretions of trichia) torrida Westwood cover at least a the odour glands and spread them during part of the female's compound eyes with further movements over the plumose the fan of long bristles on their mid legs, hairs over the abdomen and over the hind Osten (1989) showed that in Xylocopa spe- and middle legs. Good evidence for this is cies from Africa and Sri Lanka such fans that the males bend their abdomen down- on fore legs also function as blind folds wards and stretch it so that the pores of during copulation. These blind folds may the odour glands are exposed. While the have the effect that females stop flying males move their hind legs over their ab- when grabbed in mid air or prevent them domen femur and tibia are held in a 90° to take off when mounted by a male on angle. The different exposition of the spe- substrate. zialized hairs on femur and tibia ensures The phenomenon that male Hymenop- that they brush straight over the pores of tera have blind folds and odour glands in glands. So the spatula-like tips can take up modified forelegs was so far found in the secretions from the plumose hairs sur- more than one hundred species of me- rounding these pores and spread them gachilid bees, in several species of neo- subsequently over the long bristles on tropical and Old World carpenter bees both sides of the abdomen and over the and furthermore in a sphecid wasp (Crabro other legs. These movements and the high cribrarius) (Wittmann 1992, Wittmann and frequency with which they are carried out Blochtein 1995, Blochtein 1995). strongly suggest that the males perfume Osten (1989) has described that during their body in order to attract conspecifics. copula position in X. perforator males hold While the male is emitting these secre- their basitarsi on the female's head. In tions he is facing downwind. This position these basitarsi Wittmann and Blochtein possibly ensures that the male might see (1995) found odour glands. Combining incoming females that follow an odour both findings we strongly suggest that trace. The white areas in his face and the males during copula bring the secretions of shiny white hairs on his middle and front the odour glands in close contact with the legs could then serve as a further signal antennae of the female. Further students of for approaching females to detect the male X. bimaculata mating behaviour may there- in its territory. fore check whether such a copula position The chain length of the alkanes and al- can also be found in these bees, kenes we found in the cuticular washings In contrary to male X. nigrocincta, which suggests that they are not highly volatile, concentrate the nectar in the mother's nest Volume 9, Number 2, 2000 369 before each territorial flight (Wittmann and Scholz 1989), the males of X. bimacu- lata dehydrated nectar while hovering in the territory. We suppose that thereby male X. bhnaculata improve their energy budget as has been shown for X. nigro- cincta. While the latter are fed by their mother in the nest, the males of X. bhna- culata collect the nectar by themselves and therefore have to evaporate the redundant water during the territorial flight. ACKNOWLEDGMENTS This study was carried out in co-operation with the PUC-University, Porto Alegre, Rio Grande do Sul, Brazil, the University of Tubingen and the University of Bonn. We thank Dr. J.S. Moure for identification of the bees, Horst Schoppmann for scillfull help at the scanning microscope and Andreas Jux for the GC- MS analysis. LITERATURE CITED Alcock, J. 1991. Mate-locating behaviour of Xylocopa California* arizonensis Cresson (Hymenoptera: An- thoporidae). Journal of the Kansas Entomological Society 64: 349-356. Anzenberger, G. 1977. Ethological study of African Carpenter bees of the genus Xylocopa (Hymenop- tera: Anthophoridae). Zeitschrift fur Tierpsycholo- gie 44: 337-374. Blochtein, B. 1995. Die Bedeutung spezialisierter Beinstrukturen und Duftdriisen der Mannchen in der Paarungsbiologie von Blattschneiderbienen (Hymenoptera; Megachilidae). Dissertation Univ- ersitdt Tubingen, 96 p. Gerling, D., Velthuis, H. H. W. and Hefetz, A. 1989. Bionomics of the large Carpenter bees of the ge- nus Xylocopa. Annual Reviews of Entomology 34: 163-190. Hefetz, A. 1983. Function of secretion of mandibular gland of male in territorial behavior of Xylocopa sulcatipes (Hymenoptera: Anthophoridae). Jour nal of Chemical Ecology 9: 923-931. Hurd, P. D. and Linsley, E. G. 1975. The principal Larrea bees of the southwestern United States (Hymenoptera: Apoidea) Smithsonian Contribu- tions to Zoology 193: 1—47. Hurd, P. D. & Moure, J. S. 1963. A classification of the large carpenter bees (Xylocopini) (Hymenop- tera, Apoidea). University of California, Berkeley Publications in Entomology, Vol. 29. 365 pp. Osten, T. 1989. Vergleichend-funktionsmorphologis- che Untersuchungen des Paarungsverhaltens von Platynopoda und Mesotrichia (Hymenoptera: Xylocopini). Stuttgarter Beitriige zur Naturkunde, Ser. A, 433, 18 pp. Velthuis, H. H. W. and Camargo, J. M. F. 1975a. Ob- servations on male territories in a carpenter bee, Xylocopa (Neoxylocopa) hirsutissima Maidl (Hy- menoptera, Anthophoridae). Zeitschrift fur Tierp- sychologie 38: 409^118. Velthuis, H. H. W. and Camargo, J. M. F. 1975b. Fur- ther observations on the function of male terri- tories in the carpenter bee Xylocopa (Neoxylocopa) hirsutissima Maidl (Hymenoptera, Anthophori- dae). Netherlands Journal of Zoology 25: 516-528. Vinson, S. B., Frankie, G. W. and Williams, H. J. 1986. Description of a new dorsal mesosomal gland in two Xylocopa species (Hymenoptera: Anthophor- idae) from Costa Rica. Journal of the Kansas En- tomological Society 59: 185-189. Wheeler, j' W., Evans, S. L., Blum, M. S., Velthuis, H. H. W., and Camargo, J. M. F. 1976. ds-2-Methyl- 5-hydroxyhexanoic acid lactone in the mandib- ular gland secretion of a carpenter bee. Tetrahe- dron Letters 45: 4029-4032. Wittmann, D. 1992. Funktionsmorphologie der Vor- derbeine von Megachiliden-Mannchen. 13. Ta- gung der deutschsprachigen lUSSI-Sektio)i, Blaubeu ren. 17 p. Wittmann, D. and Blochtein, B. 1995. Why males of leafcutter bees hold the females' antennae with their front legs during mating. Apidologie 2b: 181 195. Wittmann, D. and Scholz, E. 1989. Nectar dehydra- tion by male carpenter bees as preparation for mating flights. Behavioral Ecology 6 Sociobiology 25:387-391. J. HYM. RES. Vol. 9(2), 2000, pp. 370-376 Load-lifting Constraints on Provisioning and Nest Building in the Carpenter Wasp, Monobia quadridens L. (Hymenoptera: Eumenidae) Paula K. Edgar and Joseph R. Coelho (PKE) Department of Instructional Services, John Wood Community College, 150 South 48th Street, Quincy, Illinois 62301, USA; (JRC) Department of Biology, Culver-Stockton College, One College Hill, Canton, Missouri 63435, USA Abstract. — The foraging and mud-carrying capacity of the trap-nesting carpenter wasp, Monobia quadridens L., was examined in relation to load-lifting ability. The body mass of caterpillar prey collected increased over the course of the season. Consequently, the ability of the wasps to carry prey became compromised late in the season. Caterpillar mass was not correlated with wasp size, but the mass of mudballs used in nest construction was related to wasp size. Wasp foraging may be constrained by the size of pyralid caterpillars available at any particular time, which changes because of caterpillar growth. Mudballs are constructed by the wasps themselves; therefore, wasps may be able to optimize mudball size in accordance with their own size, but mudballs were much lighter than caterpillars and never approached the upper limit of the wasps' ability to carry them. Optimal foraging theory suggests that an animal will experience increased fitness as it becomes more efficient at obtaining food or energy. One obstacle encountered by flying insects that carry food loads is the need to generate sufficient lift force to re- main airborne. Prey selection may be lim- ited by the size of the prey the insect can successfully carry while in flight. Marden (1987) demonstrated that maximum still-air lift force in flying animals depends primar- ily on flight muscle mass (MhJ. A fixed minimum ratio of flight muscle mass to to- tal mass lifted, the marginal flight muscle ratio, is required for successful takeoff. An- imals with higher flight muscle ratios (FMR) have greater maneuverability, and should be better able to lift and carry loads, seize prey, avoid predators and vie for ter- ritories and mates (Marden 1987). Recent studies examining the relation- ship between maximum lift force and ac- tual load carriage in foraging and provi- sioning wasps have used ground-nesting species, including Vespula spp. (Coelho and Hoagland 1995), Sphecius speciosus Drury (Coelho 1997), and Sphex ichneu- moneus L. (Coelho and LaDage 1999). In the present study we investigate load car- riage during provisioning and mud-car- rying in Monobia quadridens L. (Hymenop- tera: Eumenidae), an aerial nester. Aerial nesting may apply additional constraints to load carriage. Two ground nesting species, the cicada killer (Sphecius speciosus) and the great golden digger wasp (Sphex ichneumoneus) are able to car- ry heavier loads than are theoretically pos- sible, primarily by climbing vegetation, then flying toward their burrows (Coelho 1997, Coelho and LaDage 1999). Although presumably using maximum power, such overloaded wasps can only descend. Ae- rial nesters lack this option, as the final flight to the nest requires a vertical take- off and ascent to the nest entrance. M. quadridens may not attempt to carry near- maximal loads if the load compromises flight maneuverability and nest entry. The carpenter wasp, M. quadridens, read- ily nests in old borings of carpenter bees, Xylocopa spp. (Tandy 1908, Rau 1935), and Volume 9, Number 2, 2000 371 is the largest vespoid wasp to use wood trap-nests (Krombein 1967). A mature fe- male removes debris inside the nest cavity and collects mudballs by moistening soil with water stored in the wasp's crop (Spradbery 1973) or with saliva (Evans and Eberhard 1970). Mudballs are used to con- struct the nest's interior plug, cellular par- titions and exterior plug. Prior to mass pro- visioning with paralyzed caterpillar prey, a female deposits an egg near the inner end of the cell (Krombein 1967, Spradbery 1973). A partition is constructed between cells. A vacant space, the vestibular cell, is made near the nest's opening and sealed with a thick exterior plug. Monobia quadri- dens takes four to seven days to provision a nest (Krombein 1967). The larva emerges 5 to 8 days after the egg is laid, feeds on the paralyzed cater- pillars, applies a varnish to the cell's in- terior and pupates. The elapsed time be- tween pupation and adult emergence av- erages 17 days for males and 18 days for females (Krombein 1967). Teneral adults remain inside their cells for 2 to 3 days while their integument and wings sclero- tize, then chew through the cell partition to escape from the nest (Krombein 1967, Cowan 1991). Monobia quadridens often provisions its nest with a single species of Lepidoptera; with pyralid caterpillars the most fre- quent. Stenomid and tortricid caterpillars were also used to provision nests (Krom- bein 1967). Female M. quadridens collecting long caterpillars, 10-18 mm, used fewer caterpillars per cell than those using prey that were only 6-13 mm in length (Krom- bein 1967). Theoretically, wasps carrying larger loads make fewer trips, thus con- serving time and energy (Reavey 1993). The demands of temporal and energetic efficiency therefore interact with the con- straints of load-lifting. This study examines how load-lifting limitations influence the provisioning and mud-carrying strategies of M. quadridens. A female wasp should carry caterpillars and mudballs that are near the maximum load-lifting capacity without exceeding it in order to save time and energy by re- ducing the number of trips. In other words, a female carrying prey or mudballs should have a FMR slightly above the marginal FMR for Hymenoptera, 0.179 (Marden 1987). MATERIALS AND METHODS Field research was conducted from June through September 1996, 1997 and 1998 at Alice L. Kibbe Life Science Station, Han- cock County, Illinois. Observations and data collection were limited to sunny days because M. quadridens was not very active on overcast or rainy days. M. quadridens females were nesting in abandoned carpenter bee holes in the wooden support beams beneath the upper level porch on the east side of the Frank House, a wooden frame building housing the field station's education center. To provide additional nesting sites for M. quadridens, artificial trap-nests were con- structed according to Krombein (1967). A 12.7-mm diameter hole was drilled along the central longitudinal axis of straight- grain pine boards (38.1 mm X 38.1 mm x 200 mm) to a depth of approximately 152 mm. 12.7 mm dia holes adequately accom- modate M. quadridens females' large size (Krombein 1967). Metal brackets held the traps in place on the faces of the support beams at two meters. Initially, female wasps without prey were captured with an insect net and co- axed individually into a 1.5 ml microcen- trifuge tube (ventilated by puncturing a hole in the lid) and placed in the refrig- erator for 30 minutes. Wasps were marked on the dorsal side of the thorax with one or two small dots of enamel hobby paint. Special care was taken to avoid getting paint on antennae, wings or spiracles. Body mass (Mb) for each wasp was deter- mined to the nearest 0.001 g on an OhausK electronic balance. The wasp was then placed outdoors near the nesting site and 372 Journal of Hymenoptera Research allowed to recover fully and fly away. In- dividuals recaptured over the course of several days had Mh measured for each of those days. Subsequent captures of marked females were made whenever they returned with a caterpillar or mudball. Wasp and cater- pillar were collected and placed into a ventilated tube and unventilated tube, re- spectively. The caterpillar's body mass (Mprey) was determined to the nearest 0.001 g. Wasps returning with mudballs were also captured. Wasp and mudball were collected and placed into a ventilated tube and unventilated tube, respectively. Gen- tle handling of the mudballs was em- ployed to prevent them from crumbling. The mudball's mass (Mmud) was deter- mined to the nearest 0.001 g. In both sit- uations, marked wasps were chilled, re- weighed, and released as previously de- scribed. Multiple caterpillars and /or mud- balls were often collected from a single individual. After a wasp arrived with its third (at most) caterpillar it was placed in an airtight tube and frozen. The number of mudballs collected before the wasp was taken was highly variable. For two indi- viduals, both caterpillar and mudball data were obtained. Female wasps were frozen and transported to the laboratory for ad- ditional measurements. An Ohaus analyt- ical balance accurate to ± 0.0001 g was used to determine Mb for each female wasp. The head, abdomen, legs and wings were cut away and thorax mass (Mlh) was measured. Because flight muscle compos- es 95% of thorax mass in Hymenoptera (Marden 1987), flight muscle ratio was cal- culated as 0.95Mth/Mb. Operational (load- ed) flight muscle ratio (FMR,,) was deter- mined as 0.95Mth/(Mb + mass of load car- ried). Voucher specimens were deposited in the Entomology Museum of Western Il- linois University. Average Mb and unloaded FMR were calculated for individual wasps that car- ried multiple caterpillars and /or mud- balls. The individual averages were then 0.35 ~ 0.3 0.05). The average Mb was 0.2184 ± 0.0059, n = 54. The smallest wasp (0.09 g) was observed hauling two caterpillars simultaneously with a total prey mass of 0.104 g. This load was almost identical to the average load carried by a female nearly three times the size of the smallest wasp. Caterpillar body mass increased over the course of the season and was linearly related to Julian day (Mprey = -0.299 + 0.001 9day, n = 74, R2 = 0.316, F = 33.3, p < 0.0001) (Figure 1). Late-season caterpil- lars were nearly double the Mb of those taken during June and July. Increasing caterpillar size affected the Volume 9, Number 2, 2000 373 cc U.4 0.35 -o 0.3 CO -2 0.25 >. *■ 0.2 + 0.15 180 £ -3. -marginal 200 220 240 Julian day 260 Fig. 2. The effect of time of year on operational (prey-loaded) flight muscle ratio in wasps carrying caterpillars. The marginal level indicates the mini- mum flight muscle ratio required for successful take- off. 20 18 5 12 E 8 I 6 4 2 0 Marginal FMR Ll-Dil HH. 0.14 0.18 0.22 0.26 0.3 0.34 0 38 0 42 Prey-loaded FMR Fig. 3. The distribution of operational flight muscle ratios among wasps carrying caterpillars. The mar- ginal level indicates the minimum flight muscle ratio required for successful take-off. prey loaded FMR„ (0.2633 ± 0.0054, n = 71), which decreased significantly over the course of the season (FMR,, = 0.544- 0.0131day, n = 71, R2 = 0.267, F = 25.1, p < 0.0001) (Figure 2). Late in the season, the increase in prey mass caused the load- ed FMRs to fall below the marginal FMR for Hymenoptera. In four of 71 (5.6%) foraging events, the wasp had an average loaded FMR below the marginal FMR (Figure 3). Each of these individuals carried large caterpillars weighing an average of 0.252 g. This mass was 233% greater than the overall mean Mprev (0.108 g). No early-season, prey-lad- en wasp approached the marginal FMR for Hymenoptera. As wasp size increased, the size of mud- balls (mean = 0.0513 ± 0.0032 g) used during nest construction increased. Signif- icant relationships demonstrated the effect of wasp mass on mudball mass: Mb versus Mmud (Mmud = 0.0051 + 0.1943Mb, n = 38, R2 = 0.270, F = 13.3, p = 0.0008) (Figure 4). A weaker, but significant effect of Mlh on Mmud was also present (Mmiu) = -0.0054 + 0.6072M,,,, n = 27, R2 = 0.205, F = 6.44, p = 0.018). The average mudball mass was 50% less than and significantly different from (t = 7.36, df = 99, p < 0.0001, t-test) the aver- age prey mass (0.1081 g ± 0.0069, n = 74). Consequently, the mean FMR for females loaded with mudballs, (0.3099 ± 0.0033, n = 27) was well above the marginal FMR. DISCUSSION Wasp body mass and thorax mass did not affect the size of prey provisioned. If females were actively selecting prey by size, they should take the largest caterpil- lars they can lift. Thus, larger wasps would be choosing larger prey, as occurs in S. ichneumoneus (Coelho and LaDage 1999) and Palinodes laeviventris Cresson (Gwynne and Dodson 1983). Females would spend less time foraging and en- ergy would be conserved. However, ob- servations of M. quadridens did not sup- port this hypothesis. As the season progressed, caterpillar 0.1 _0 08 a> lo.06 E I 0 04 T3 5 0.02 0 14 0.16 0 18 0 2 0.22 0.24 0.26 0.28 0.3 0 32 Wasp body mass (g) Fig. 4. The effect of wasp bodv mass on mass of mudballs carried bv Monobia quadridens. 374 Journal of Hymenoptera Research size increased, consequently decreasing time and energy resulting from failed for- the FMRo of foraging wasps. Early in the aging attempts on large caterpillars sug- season, females exhibited no difficulty in gests that either small caterpillars were in carrying prey. Prey mass was never large short supply, or that success was frequent enough to substantially decrease the enough to outweigh failure. Indeed, on oc- FMR,,; therefore, loaded females remained casion females successfully brought in well above the marginal FMR for Hyme- prey larger than themselves. In one case, noptera. In August, however, provision- a female was loaded with a caterpillar 1.5 ing females encountered load-lifting con- times greater than her own body mass, straints because of the increasingly large Similar effects of changing prey size be- caterpillars. As a result, the FMR of loaded cause of prey growth are known from oth- individuals dropped near or slightly be- er species. Seasonal variation in caterpillar low the marginal FMR. size also dramatically affects the provi- Considering the prey cues available to sioning style of the solitary digger wasp predatory wasps and their acute visual Ammophila sabulosa L. (Field 1992). Large abilities, it is doubtful that early in the sea- prey are taken more frequently during the son M. quadridens females would be inca- second part of the nesting period (entire pable of finding larger caterpillars if such nesting period runs from late June prey were present (Stamp and Wilkens through early September) and are carried 1993). It is more likely that females pro- on foot. Early in the season, when smaller visioned smaller prey at this time because prey are more common, provisioning they were the most readily available, if not wasps require a greater number of small the only suitable prey available. prey, which they carry in flight. Further- In early summer, females, regardless of more, smaller wasps multiply-provision individual size, primarily foraged on their nests (using smaller prey) more fre- small-bodied caterpillars. Caterpillars pre- quently than larger wasps (Field 1992). sumably grew as the summer progressed, Brockman and Grafen (1992) describe and late in the season M. quadridens fe- the effect of the growth of spiders on their males were not always successful in their predator, the mud-dauber, Trypoxylon pol- attempts to carry the larger caterpillars. Hum Say. At the start of the season, wasps On one occasion a female made no at- forage on a genus of spiders (Eustala) that tempt to recover a large caterpillar after overwinter as adults. The majority of the dropping it. Another low-flying wasp fell season, wasps provision with the genus to the ground without dropping the prey Neoscona, which overwinter as spiderlings. item, crawled 30 cm up a beam then flew As spider size gradually increases, wasps 183 cm (horizontal flight) and landed on a late in the season experience difficulty chair. Again, she tumbled to the ground hauling the larger spiders, frequently and dragged the caterpillar 147 cm dropping them. Additionally, wasps ex- through the grass prior to abandoning it. pend more energy and risk being attacked These two late season caterpillars had an by large adult spiders. Landes et al. (1987) average mass of 0.256 g, which was twice found that "wasps collected spiders in the size of the largest early season cater- numbers relative to their seasonal and rel- pillar. As a result, individual FMR,, fell to ative abundance, accessibility as prey, or between 0.239 and 0.165, and caterpillars size suitability." were dropped. In addition to carrying caterpillar prey, Typically, predators choose prey that female M. quadridens also carried mud- are large enough to make it worth their balls used in nest construction. Wasp time and energy, yet small enough to be body mass significantly influenced the easily carried (Reavey 1993). The loss of mass of mudballs carried. This effect sug- Volume 9, Number 2, 2000 375 gests that females constructed mudballs of a size proportional to their individual body size. These findings are consistent with Archer's (1977) study of Paravespula vulgaris (L.), in which forager body size was significantly correlated to the earthen load carried by wasps leaving the nest. In contrast to several of the prey-loaded FMRs that dropped below the marginal FMR, females never carried mudballs large enough to substantially decrease their FM^. The FMRmud values were far greater than the marginal FMR. M. quad- ridens may not optimize energy costs of mudball production and carriage because of the style used to carry mudballs. Wasps may be restricted to making small, round mudballs compact enough to be easily car- ried in their mandibles. The difference in carriage style between prey and mud per- haps best explains why female M. quadri- dens could haul heavier prey loads than mudballs. Caterpillars were grasped with all legs and held lengthwise against the fe- male's underside without altering the cen- ter of gravity (Evans 1962). In contrast, mudballs carried with the mandibles and forelegs placed additional weight toward the head, altering balance. To compensate, individuals may have been restricted to hauling mudballs that were much lighter than the prey. On average, M. quadridens' unloaded FMR (0.385), although higher than the mean for Hymenoptera (0.34 ± 0.013, n = 15; data from Marden 1987, Coelho 1991, 1997, Coelho & Hoagland 1995, Coelho and LaDage 1999), is similar to that of oth- er vespoids such as Vespula (Coelho and Hoagland 1995). M. quadridens' FMR was considerably lower than that of the ground-nesting sphecids Sphecius speciosus (0.416, Coelho 1997) and Sphex ichneumo- neus (0.462, Coelho and LaDage 1999). M. quadridens is therefore less maneuverable than the ground nesters when unladen. However, the mass allocation of M. quad- ridens should be matched to the maximum demands of load carriage, which occur when prey are carried. Only 5.6% of pro- visioning events caused M. quadridens to have an FMR,, below marginal. In contrast, Sphecius speciosus and Sphex ichneumoneus provision at levels below marginal FMR„ 90% (Coelho 1997) and 25% (Coelho and LaDage 1999) of the time, respectively. Therefore, M. quadridens is on average more maneuverable when provisioning than the ground nesters. Aerial nesting may, in fact, carry maneuverability restric- tions as predicted. ACKNOWLEDGMENTS Appreciation is given to Richard V. Anderson and Gregg Dieringer for their support and insightful sug- gestions. Gratitude is extended to Robert W. Longair for identifying Monobia quadridens and for his helpful advice. Sharon Bringer, Quoc Giang and Brian Kras- kiewicz provided additional assistance in the field. Matthew McClure helped identify the caterpillars. Andy Edgar constructed the trap nests. LITERATURE CITED Archer, M. E. 1977. The weights of forager loads of Paravespula vulgaris (Linn.) (Hymenoptera: Ves- pidae) and the relationship of load weight to for- ager size. Insectes Sociaux 24: 95-102. Brockman, H. ]. and A. Grafen. 1992. Sex ratios and life history patterns of a solitary wasp, Trypoxy- lon (Trypargilum) politum (Hvmenoptera: Spheci- dae). Behavioral Ecology and Sociobiology 30: 7-27. Coelho, J. R. 1991. The effect of thorax temperature on force production during tethered flight in honeybee (Apis mellifera) drones, workers, and queens. Physiological Zoology 64: 823-835. Coelho, J. R. 1997. Sexual dimorphism and flight be- havior in cicada killers, Sphecius speciosus. Oikos 79: 371-375. Coelho, J. R. and J. Hoagland. 1995. Load-lifting ca- pacities of three species of yellow-jackets (\ es pula) foraging on honev-bee corpses. Functional Ecology 9: 171-174. Coelho, J.R. and L.D. LaDage. 1999. Foraging capacity of the great golden digger wasp, Sphex ichneu- moneus L. Ecological Entomology 24: 4S0-4S3. Cowan, D. P. 1991. The solitary and presoda] vespi- dae. pp. 33-73. In: K. G. Ross and R. VV. Mat- thews (eds), The Social Biology <■' Wasps. Corn- stock Publishing Associates, Ithaca, 673 pp. Evans, H. E. 1962. The evolution ol prey-carrying mechanisms in wasps / volution 16: 468 483 Evans, H. E. and M. J. Eberhard. 1970. The Wasps. The University of Michigan Press, Ann Arbor 265 PP 376 Journal of Hymenoptera Research Field, J. 1992. Patterns of provisioning and parental investment in the solitary digger wasp Ammo- phila sabulosa. Ecological Entomology 17: 43-51. Gwynne, D. T. and G. N. Dodson. 1983. Nonrandom provisioning by the digger wasp, Palmodes laevi- ventris (Hymenoptera: Sphecidae). Annals of the Entomological Society of America 76: 434^436. Krombein, K. V. 1967. Trap-Nesting Wasps and Bees: Life Histories, Nests, and Associates. Smithsonian Press, Washington, D.C., 570 pp. Landes, D. A., M. S. Obin, A. B. Cady, and J. H. Hunt. 1987. Seasonal and latitudinal variation in spider prey of the mud dauber Chalybion californiatm (Hymenoptera: Spechidae). Journal of Arachnology 15: 249-256. Marden, ]. H. 1987. Maximum lift production during takeoff in flying animals. Journal of Experimental Biology 130: 235-358. Reavey, D. 1993. Why body size matters to caterpil- lars, pp. 248-279. In: N. E. Stamp and T. M. Cas- ey (eds), Caterpillars: Ecological and Evolutionary Constraints on Foraging. Chapman and Hill, New York, 587 pp. Rau, P. 1935. The courtship and mating of the wasp, Monobia quadridens (Hymen. :Vespidae). Entomo- logical News 16: 57-58. Spradbery, J. P. 1973. Wasps: an Account of the Biology and Natural History of Solitary and Social Wasps. University of Washington Press, Seattle, 408 pp. Stamp, N. E. and R. T. Wilkens. 1993. On the cryptic side of life: being unapparent to enemies and the consequences for foraging and growth of cater- pillars, pp. 283-330. In: N. E. Stamp and T. M. Casey (eds), Caterpillars: Ecological and Evolution- ary Constraints on Foraging. Chapman and Hill, New York, 587 pp. Tandy, M. 1908. The carpenter mud wasp (Mo)iobia quadridens). Entomological News 19: 231-232. Zar, J. 1996. Biostatistical A>ialysis. Prentice-Hall, New Jersey, 620 p. J. HYM. RES. Vol. 9(2), 2000, pp. 377-384 Reproductive Biology of the Seed-harvester Ants Messor julianus (Pergande) and Messor pergandei (Mayr) (Hymenoptera: Formicidae) in Baja California, Mexico Robert A. Johnson Department of Biology, Arizona State University, Tempe, Arizona 85287-1501, USA Abstract. — The seed-harvester ant Messor julianus (Pergande) exhibits a parapatric distribution pattern with the ecologically equivalent congener M. pergandei (Mayr) in the Baja California pen- insula of Mexico; M. pergandei replaces M. julianus in drier soil microhabitats within the contact zone between these two species. This paper describes the reproductive biology of M. julianus and M. pergandei to provide a first step in understanding factors involved in causing this replacement pattern. Mating flights of M. julianus were observed over a several week period from early Feb- ruary to early March, and thus appear similar to those of M. pergandei. Likewise, starting nests of both species contained one foundress. Moreover, the similar ecology and mating flights of M. julianus and M. pergandei suggest that the replacement pattern exhibited by these two species is associated with patterns of foundress survival. In regard to mating flights, both M. julianus and M. pergandei are postulated to have diverged from the putative ancestral condition of summer mating flights that occur in other Nearctic congeners. This seasonal difference in timing of the mating flight for these two species correlates with their being the only Nearctic species of Messor that are restricted to hot desert habitats. Alate females for both M. julianus and M. pergandei have poor tolerance to high temperature relative to desert ants in the genera Aphaenogaster and Pogon- omyrmex. The seed-harvesting ant genus Messor (Hymenoptera: Myrmicinae) is common throughout the southwestern deserts of the United States and northwestern Mex- ico. Four species of Messor, M. andrei (Mayr), M. julianus (Pergande), M. pergan- dei (Mayr), and M. stoddardi (Emery), oc- cur in the Baja California peninsula of Mexico; M. julianus is endemic to the pen- insula (Johnson 2000a; R. Johnson and P. Ward, unpubl. data) (Fig. 1). Two of these species, M. julianus and M. pergandei, are common in most low elevation habitats (< — 1000 m) with their combined geographic distributions encompassing all but the northwest portion of the peninsula. The other two species, M. andrei and M. stod- dardi, are largely restricted to coastal and adjacent inland areas along the Pacific Coast. In the Baja California peninsula, M. andrei is restricted to the relatively mesic California Floristic province in the north- west portion of the peninsula (R. Johnson and P. Ward, unpubl. data). The range of M. stoddardi extends to central portions of the peninsula, but this species rarely co- exists with M. julianus or M. pergandei (R. Snelling, unpubl. data; R. Johnson, pers. obs.) (Fig. 1). Messor julianus and M. pergandei are eco- logically similar species. Colonies of both species consist of many thousands of workers that forage in long columns (Johnson 2000a), and these two species are the only Nearctic Messor that are restricted to occurring in hot desert habitats (Wheel- er and Wheeler 1973). Geographically, M. julianus is mostly restricted to central and southern portions of the peninsula, while M. pergandei occurs in eastern portions of the peninsula to as far south as northeast- ern BCS (the state of Baja California Sur) 378 Journal of Hymenoptera Research 5 California Baja California Messor A andrei o julianus • pergandei a stoddardi Baja California Sur 70 0 70 140 Kilometers Fig. 1. Geographic distribution of species of Mt'ssor that occur in the Baja California peninsula, Mexico. The full geographic distribution of each species is given in Johnson (2000a). (Fig. 1). In northern and central BC (the state of Baja California), M. julianus inhab- its a narrow range along the cool Pacific Coast, while M. pergandei is restricted to xeric desert areas along the eastern coast. Moreover, these two species exhibit a par- apatric distribution pattern (herein de- fined as species that occupy separate but adjoining areas, such that only a small fraction of individuals in each encounters the other [Futuyma and Mayer 1980]) with ranges of the two species overlapping across a contact zone in the central pen- insula (Fig. 1). While the two species are often sympatric within the contact zone, the pattern across the contact zone is one in which the two species replace one an- other along local gradients that correlate Volume 9, Number 2, 2000 379 with abiotic habitat features (Johnson out the flight period each day, alates fly- 2000a). Across the contact zone, M. pergan- ing from nests were counted during se- dei inhabits the drier microhabitats, i.e., quential 2 minute visits to each nest. Am- those that are lower in elevation or in bient temperature was measured periodi- which the soils have a higher percentage cally about 5 cm above ground using a composition of sand (drier soils) (R. John- thermocouple thermometer, son, pers. obs.). The number of foundresses per starting Colony founding is the most vulnerable colony was determined by excavating stage in the life history of ants (Tschinkel founding nests of M. julianus and M. per- 1992, Herbers 1993, Johnson 1998), and it gandei. Data for M. julianus were collected is this stage that likely determines micro- 17-18 km west of La Purisima, BCS, and macro-distribution of adult colonies. (26°09'N, 112°13'W) in March 1992, and Thus, comparative data on reproductive near Punta San Hipolito, BCS, (27°00'N, biology of M. julianus and M. pergandei 114°00'W) in February 1998. These same provide a first step in understanding fac- data were collected for M. pergandei near tors involved in causing the replacement Highway 1 at 17.5 km west of Bahia de los pattern exhibited by these two species. Angeles, BC, (28°59'N, 113°44'W) and Mating flights of M. pergandei are well along Highway 1 at 6 km south of the pa- known and typically occur between late ved turnoff to Bahia de los Angeles, BC, January and mid-March (Pollock and Riss- (29°00'N, 114°10'W) in February 1995. ing 1985, Ryti 1988, Cahan et al. 1998), and I assessed relative tolerance to high tem- thus deviate seasonally from the summer perature by comparing survival for alate mating flights of other Nearctic Messor females of M. julianus and M. pergandei (Creighton 1953, Cole 1963, McCluskey with that of two species of Aphaenogaster 1963, Wheeler and Wheeler 1973, Snelling (A. albisetosa and A. cockerelli) and four and George 1979, Brown 1999, R. Snelling, species of Pogonomyrmex (P. barbatus, P. oc- unpubl. data, M. Bennett, pers. comm.). In cidentalis, P. rugosus, and P. salinus) (see contrast, the mating flights and the female Table 1 for collection data); mating flights sexuals of M. julianus are undescribed in of both species of Aphaenogaster and all the literature. Based on the fact that M. ju- four species of Pogonomyrmex are trig- lianus and M. pergandei are the only Ne- gered by summer rains (Johnson 2000a). arctic Messor that are restricted to hot de- Trials used test tubes that were partially sert habitats, I hypothesized that the sea- filled with water trapped by cotton plugs, sonal timing of mating flights was similar Alate females were placed into the tubes for these two species. I assessed potential and the openings were plugged with physiological constraints on mating flight moist cotton, thus providing ad libitum season by comparing high temperature water at both ends. Trials at each temper- tolerance for alate females of Messor, ature used one tube containing 25 individ- Aphaenogaster, and Pogonomyrmex. uals of one species that had been collected from at least four colonies. A separate set Mb 1 HODS Qf individuais was used at each tempera- I observed ten colonies of M. julianus ture. Each species was tested over 1° C in- over nine days in mid-February 1993, near crements that resulted in mortality rang- Highway 1 at 26 km northwest of Santa ing from 0-100%. The tubes were placed Rosalia, BCS (27°23'N, 112°28'W). All ten in a darkened incubator for 2 hours at the colonies had large nests and an active for- appropriate temperature; individuals un- aging column. The site was a sandy plain able to right themselves after that time dominated by Larrea tridentata, Opuntia were considered dead. Mortality data cholla, and Pachycereus pringlei. Through- were compared among congeners and 380 Journal of Hymenoptera Research Table 1. Collection data for alate females used in temperature tolerance tests. Localities are in the United States except as noted. Species Collection Locale Latitude Longitude Elevation (m) Collection Date Aphaenogaster A. albisetosa Mayr AZ: Pinal Co., 8 km NE Casa Grande 32°56'N 111°42'W 430 26 JUL 1995 A. cockerelli Andre Messor NM: Hildago Co., Jet. Hwys 9 & 80 31°56'N 109°02'W 1260 16 JUL 1994 M. julanus (Pergande) MEXICO: Baja Califor- nia, 47.5 km S Bahia de los Angeles 28°38'N 113°20'W 80 3 FEB 1995 M. pergandei (Mayr) MEXICO: Baja Califor- nia, 29 km S Bahia de los Angeles 28°41'N 113°26'W 105 4 FEB 1995 Pogonomyrmex P. barbatus (Smith) NM: Hildago Co., 3 km N Rodeo 31°52'N 109°02'W 1225 5 JUL 1993 P. occidentalis (Cresson) P. rugosus Emery P. salinus Olsen AZ: Yavapai Co., Chino Valley NM: Hildago Co., Jet. Hwys 9 & 80 NV: Clark Co., Dry Lake 34°46'N 31°56'N 35°54'N 112°27'W 109°02'W 114°56'W 1450 1260 520 30 JUL 1995 5 JUL 1993 30 AUG 1993 across genera using a contingency table analysis. For across genera comparisons, data sometimes were not available for temperatures below or above those caus- ing 0 and 100% mortality. To make these comparisons, I assumed that the upper and lower bounds were threshold temper- atures. Voucher specimens are deposited at the Museum of Comparative Zoology, Har- vard University, Cambridge, Massachu- setts, the Los Angeles County Museum of Natural History, Los Angeles, California, and the Robert A. Johnson collection, Tempe, Arizona. RESULTS I observed mating flights of M. julianus from early February to early March. At the site northwest of Santa Rosalia, BCS, alates flew from 6 of the 10 observation colonies, with few alates released each day. Flights occurred on 4 of 9 days and were relatively synchronous among colo- nies, i.e., on a given day, alates either flew from several colonies or none of the colo- nies (Figure 2). Mating flights typically oc- curred from 0830-1000 h (MST) at tem- peratures of 16-23° C. Flights occurred ir- respective of mild breezes or overcast skies, but were precluded by light rain or moderate breezes. Alates temporarily re- treated into nests following gusts of wind. After the last day, I excavated the ten ob- servation colonies and several adjacent colonies. In all colonies, alates were pre- sent in very low numbers or absent. Mating flights of M. julianus were also observed on 5 March, 1992, at 17 km west of La Purfsima, BCS, (26°09'N, 112°13'W) and on 5 February, 1995, at 48.7 km south of Bahia de los Angeles, BC (28°38'N, 113°20'W). At the latter site, ten marked colonies of M. julianus were observed on 31 January and 5 February. On 31 January, no alates were observed outside any of the ten colonies and no foundresses were lo- cated at the site. On 5 February, alates were observed outside 6 of the 10 colonies and individuals flew from one of these nests and from several unmarked nests. Mating flights of M. pergandei were ob- Volume 9, Number 2, 2000 381 10 5 i i i i 10 5 Z i i i i i CM K 10 LU a. 5 I LU CD l,o Z s 5 I i i i < 10 5 ! I 10 5 i i i 13 14 15 16 17 18 19 20 21 FEBRUARY Fig. 2 Flight activity for six colonies of Messor ju- lianus northwest of Santa Rosalia, Baja California Sur, Mexico, during February 1993. For each day, data are from the two minute observation period in which the highest number of alates were released from the nest. served on 26 February, 1991, at Punta Es- trella, BC, (30°55'N, 114°43'W) on 29 Feb- ruary, 1992, along Highway 1 at 16.0 km east of San Ignacio, BCS, (27°20'N, 112°46'W) and on 1-2 February, 1995, along Highway 1 at 17.5 km west of Bahia de los Angeles, BC (28°59'N, 113°44'W). The latter site had been visited several days earlier, but no foundresses of M. per- gandei were located. All of 44 and 127 starring nests for M. julianus and M. per- gandei, respectively, contained one foun- dress (Table 2). Tolerance to high temperature was first compared between congeners. Species within all three genera had similar toler- ance to high temperature (Chi-square, P > 0.10), so data for species within each ge- nus were pooled. In contrast, tolerance to high temperature varied significantly across genera (Chi-square = 231.9, 14 df, P < 0.001). Subsequent between-genus tests demonstrated that temperature tol- erance differed between Messor and both Aphaenogaster (Chi-square = 34.9, 5 df, P < 0.001) and Pogonomyrmex (Chi-square = 181.3, 7 df, P < 0.001). The primary con- tributors to Chi-square values were at low temperatures, where mortality was higher than expected for Messor and lower than expected for Aphaenogaster and Pogonomyr- mex (Table 3). Overall, most individuals survived to 42^13° C in Messor, 44—45° C in Aphaenogaster, and 46-47° C in Pogono- myrmex. DISCUSSION Reproductive biology of M. julianus ap- pears similar to that of M. pergandei. The asvnchronous late winter to early spring mating flights of both species appear cued by photoperiod (McCluskey 1963) and ex- tend for several weeks both within and among nests (Pollock and Rissing 1985). Mating flights of M. julianus were ob- served from early February7 through early March, and anecdotal observations sug- gest that most alates were released during this several week period. For example, at one site alates were observed outside of nests in earlv Februarv but not several days earlier, which suggests that this was near the beginning of the mating flight season. Similarly, few if any alates could be excavated from nests after late Febru- ary, suggesting the end of the mating flight season. Mating flights of M. pergandei occur from earlv to mid-morning under clear skies as air temperatures reach about 22 C. The lower temperatures at which A I. ju- lianus initiates mating flights are associat- 382 Journal of Hymenoptera Research Table 2. Number of foundresses in starting nests of Messor julianus and M. pergandei in the Baja California peninsula, Mexico. For location, BC = Baja Californa; BCS = Baja California Sur. Foundress Association Size Location Latitude Longitude Elevation (m) Species 1 >i Date M. julianus 40 0 17-18 km W La Purisima, BCS 26°09'N 112°13'W 100 4-5 MAR 1992 4 0 Punta San Hipoli- to, BCS 26°58'N 113°59'W 5 10 FEB 1998 Total 44 0 M. pergandei 23 0 Punta Estrella, BC 30°55'N 114°43'W 5 25-27 FEB 1991 51 0 Highway 1 at 17.5 km W of Bahia de los Angeles, BC 28°59'N 113°44'W 80 3 FEB 1995 53 0 Highway 1 at 6 km S of turnoff to Bahia de los Angeles, BC 29°00'N 114°10'W 365 3 FEB 1995 Total 127 0 ed with this species also foraging at much lower temperatures. During winter and spring, foraging columns of M. julianus form prior to dusk and foraging continues into the night until ground temperatures decrease to < 11° C (R. Johnson, unpubl. data). Conversely, M. pergandei forages di- urnally during this season, beginning after ground temperatures reach about 18° C (Bernstein 1974). While flights of M. per- gandei are often precluded by overcast weather or slight breezes (Pollock and Rissing 1985), those of M. julianus often proceeded, at least at low levels, under these conditions. Both M. pergandei and M. julianus are haplometrotic (one foundress per starting nest) in the Baja California peninsula. In M. pergandei, the number of foundresses varies geographically from haplometrosis in southern California to pleometrosis (multiple foundresses per starting nest) in southeastern California and Arizona (Pol- lock and Rissing 1985; Ryti 1988; Cahan et al. 1998; Rissing et al. 2000). However, ma- ture colonies of M. pergandei have a single Table 3. High temperature tolerance (°C) for alate females in the ant genera Aphaenogaster, Messor, and Pogonorm/rtnex. Values are per cent mortality for 25 individuals over 2 h with ad libitum moisture. Species I emperature ( C) 40 41 43 44 45 4h 48 A. albisetosa A. cockerelli M. julianus M. pergandei P. barbatus P. occidental is P. rugosus P. salinus 0.0 0.0 4.0 0.0 0.0 48.0 80.0 100.0 0.0 4.0 44.0 100.0 100.0 0.0 72.0 100.0 100.0 4.0 88.0 100.0 100.0 100.0 0.0 0.0 0.0 44.0 100.0 0.0 12.0 100.0 100.0 0.0 2.0 0.0 48.0 100.0 0.0 0.0 80.0 100.0 Volume 9, Number 2, 2000 383 queen regardless of the initial founding sumably having invaded North America strategy (Rissing and Pollock 1987; S. Riss- from Asia via Beringia (R. Snelling, pers. ing and J. Parker, unpubl. data). These comm.). Alternatively, some recent evi- data extend the distribution of haplome- dence suggests that the Aphaenogaster spe- trosis for M. pergandei from southeastern cies belonging to the former Novomessor California to its southern range limit in the (including A. albisetosa and A. cockerelli) Baja California peninsula such that both are the sister group to Nearctic Messor M. pergandei and M. julianus are haplo- (Bennett 2000). Moreover, evaluating the metrotic where the two species are sym- ancestral mating flight condition depends patric. Mating occurs in the air for M. per- upon determining the appropriate out- gandei (S. Rissing, pers. comm.) and M. an- group. The relationship between M. juli- drei (R. Johnson, pers. obs.), with in copulo anus and M. pergandei is also poorly re- pairs of both species sometimes falling to solved (Bennett 2000) because the clade the ground; mating in M. julianus is prob- containing these two species also contains ably similar. M. lariversi, which is a Great Basin species The replacement pattern exhibited by that has summer mating flights. M. julianus and M. pergandei is common One possible evolutionary scenario is among seed-harvester ants in western the late winter to early spring flights North America (Johnson 2000a). For ex- evolved one time and are common by de- ample, the ecologically equivalent species scent in M. pergandei and M. julianus. If Pogonomyrmex barbatus and P. rugosus seg- this is the case, the shift from a summer regate microhabitats along gradients of to a late winter to early spring flight sea- soil texture. Foundresses of P. rugosus son may have been necessitated by phys- have a higher tolerance to desiccating con- iological constraints related to tempera- ditions, which correlates with this species ture tolerance, and prerequisite to these occurring in drier soil microhabitats (John- two species invading hot desert habitats, son 2000b). The similar ecology and mat- Moreover, alate females of both species ing flights of M. julianus and M. pergandei have poor heat tolerance compared to spe- suggest that the micro- and macro-habitat cies of Aphaenogaster and Pogonomyrmex, differences exhibited by these two species suggesting Messor foundresses could not are also associated with patterns of foun- survive soil temperatures present during dress survival. Given that M. pergandei in- summer. A similar temperature constraint habits the more xeric micro- and macro- occurs in the desert leaf-cutter ant Acro- habitats, it is predicted that foundresses of myrmex versicolor, whose mating flights this species are more desiccation tolerant are triggered by late summer rains. Fe- than are foundresses of M. julianus (John- males of A. versicolor also have poor tol- son 2000a). erance to high temperature, and the foun- The late winter to early spring mating dresses survive by selectively initiating flights of M. julianus and M. pergandei are nests in shaded, cooler microhabitats correlated with these being the only two (Rissing et al. 1986). Nearctic Messor that are restricted to hot desert habitats (Wheeler and Wheeler ACKNOWLEDGMENTS 1973). However, it is difficult to determine Tnis manuscript benefited from comments and dis- the sequence of character-state change in cussions uith S.W. Rissing and R.R. Snelling, and by , .•, ,. n. , , .,, . xt .• comments from two anonymous reviewers. timing of the mating flight within Nearctic Messor because the phylogenetic relation- LITERATURE CITED ships of this species group are unclear. Bennett M 2000 Systematics ot tlu, Nearcti( l,,im. Messor is a predominantly Old World ge- vore ant genus Veromessor. PhD thesis, Univ. Ca- nus, with the Nearctic components pre- lit. 384 Journal of Hymenoptera Research Bernstein, R.A. 1974. Seasonal food abundance and foraging activity in some desert ants. American Naturalist 108:490-498. Brown, M.J.F. 1999. Semi-claustral founding and worker behaviour in gynes of Messor andrei. In- sectes Sociaux 46:194-195. Cahan, S., Helms, K.R., Rissing, S.W. 1998. An abrupt transition in colony founding behaviour in the ant Messor pergandei. Animal Behaviour 55:1583- 1594. Cole, A.C. 1963. A new species of Veromessor from the Nevada test site and notes on related species (Hymenoptera: Formicidae). Annals of the Ento- mological Society of America 56:678-682. Creighton, W.S. 1953. New data on the habits of the ants of the genus Veromessor. American Museum Novitates 1612:1-18. Futuyma, D.J., Mayer, G.C. 1980. Non-allopatric spe- ciation in animals. Systematic Zoology 29:254-271. Herbers, J.M. 1993. Ecological determinants of queen number in ants, pp. 262-293. In L. Keller (ed.), Queen Number and Sociality in Insects. Oxford University Press, New York. Johnson, R.A. 1998. Foundress survival and brood production in the desert seed-harvester ants Po- gonomyrmex rugosus and P. barbatus (Hymenop- tera, Formicidae). Insectes Sociaux 45:255-266. Johnson, R.A. 2000a. Seed-harvester ants (Hymenop- tera: Formicidae) of North America: an overview of ecology and biogeography. Sociobiology 36:83- 122. Johnson, R.A. 2000b. Habitat segregation based on soil texture and body size in the seed-harvester ants Pogonomyrmex rugosus and P. barbatus. Eco- logical Entomology, in press. McCluskey, E.S. 1963. Rhythms and clocks in har- vester and Argentine ants. Physiological Zoology 36:273-292. Pollock, G.B., Rissing, S.W. 1985. Mating season and colony foundation of the seed-harvester ant, Ver- omessor pergandei. Psyche 92:125-134. Rissing, S.W., Pollock, G.B. 1987. Queen aggression, pleometrotic advantage and brood raiding in the ant Veromessor pergandei (Hymenoptera: Formi- cidae). Animal Behaviour 35:975-981. Rissing, S.W., Johnson, R.A., Martin, J.W. 2000. Col- ony founding behavior of some desert ants: geo- graphic variation in metrosis. Psyche 103:95-101. Rissing, S.W., Johnson, R.A., Pollock, G.B. 1986. Natal nest distribution and pleometrosis in the desert leaf-cutter ant Acromyrmex versicolor (Hymenop- tera: Formicidae). Psyche 93:177-186. Ryti, R.T. 1988. Geographic variation in cooperative colony foundation in Veromessor pergandei (Hy- menoptera: Formicidae). Pan-Pacific Entomologist 64:255-257. Snelling, R.R., George, CD. 1979. The Taxonomy, Dis- tribution and Ecology of California Desert Ants (Hy- menoptera: Formicidae). Report to Bur. Land Man- age., US Dept. Int., Riverside, Calif. Tschinkel, W.R. 1992. Brood raiding and the popu- lation dynamics of founding and incipient colo- nies of the fire ant, Solenopsis invicta. Ecological Entomology 17:179-188. Wheeler, G.C., Wheeler, J. 1973. Ants of Deep Canyon. Univ. Calif. Press, Berkeley. J. HYM. RES. Vol. 9(2), 2000, pp. 385-394 Morphofunctional Adaptations of Parasitoids Attacking Concealed Eggs of Two Arboreal Mirids in Italy Eric Conti, Pio F. Roversi, and Ferdinando Bin (EC, FB) Department of Arboriculture and Plant Protection, University of Perugia, Borgo XX Giugno, Perugia, 06121 Italy; (PFR) Experimental Institute for Agricultural Zoology, Forest Entomology Section, Ministry for Agricultural Politics, Via Lanciola 12/ A, Firenze, 50125 Italy Abstract. — The morphofunctional adaptations of the egg parasitoids of two arboreal mirids (Het- eroptera: Miridae) for reaching the concealed eggs of their hosts, Calocoris quadripunctatus (Villers) and Calocoris trivialis (Costa), were studied in a natural ecosystem (oak forest) and in an agroe- cosystem (orange groves), respectively. Calocoris quadripunctatus is a predator of immature stages of Tortrix viridana (L.) (Lepidoptera: Tortricidae) in deciduous oak forests, Quercus spp., in Tus- cany. Eggs are laid in clusters and concealed among the scales of dead buds, where they are exploited by two parasitoid species. Chaetostricha walkeri (Forster) (Hymenoptera: Trichogram- matidae) has a long ovipositor which is inserted between the scales to reach the host eggs, and is therefore an "ovipositor prober". Instead, Telenomus sp. laricis Walker group (Hymenoptera: Scelionidae) has a depressed metasoma which is introduced between the scales, and is therefore a "metasomal prober". Calocoris trivialis is a phytophagous species that damages orange groves, Citrus sinensis (L.), and other Citrus spp. in Sicily. The eggs are concealed in the soft, decaying wood of old pruning wounds and are attacked by at least two parasitoids. Aprostocetus n. sp. near miridivorus (Domenichini) (Hymenoptera: Eulophidae) is probably both an "ovipositor prober" and an "ovipositor driller", as it can also drill through the wood. Telenomus lopicida Silvestri (Hymenoptera: Scelionidae) has a long ovipositor and a compressed metasoma, which is intro- duced into the host incision, and is therefore a "metasomal prober". Such morphological adap- tations appear to be linked to the host oviposition sites and explain some aspects of parasitoid exploitation efficiency; they may also help interpret other host-parasitoid associations that are unknown or questionable. Mirid bugs (Heteroptera: Miridae), the largest family of Heteroptera, are very common both in arboreal and herbaceous ecosystems, were they feed on plants (phytophagous species), other arthropods (zoophagous) or both (zoophytophagous) (Wagner and Weber 1964, Alomar and Wiedenmann 1996). Their eggs are elon- gated with a true operculum on their an- terior pole and are more or less deeply embedded in dead or living plant tissues, or concealed between plant organs (Kul- lenberg 1946, South wood 1956, Cobben 1968, Hinton 1981). In spite of these kinds of protection, mirid eggs can be attacked by parasitoids belonging to Eulophidae, Trichogrammatidae, Scelionidae and My- maridae (Hymenoptera) (Bin and Vinson unpublished). Some of these host-parasit- oid associations from a natural ecosystem, oak forest, in Tuscany (central Italy) and from an agroecosystem, orange groves, in Sicily (southern Italy), are described here, focusing on morphological features relat- ed to parasitoid strategies. Oaks in Tuscany are attacked by Tortrix viridana (L.) (Lepidoptera: Tortricidae), which is distributed over large portions of the Palaearctic region, from northern Eu- rope to the Mediterranean region, and during population outbreaks can seriously defoliate vast groves (Bogenschutz 1978). Eggs and possibly young larvae of this tortricid are preyed upon (Roversi unpub- 386 Journal of Hymenoptera Research lished) by the zoophytophagous mirid Calocoris quadripunctatus (Villers) (Wagner and Weber 1964), the population size of which was found to be directly dependent on T. viridana density (Roversi et al. in preparation). Calocoris quadripunctatus has one generation per year and overwinters in the egg stage. Eggs are concealed in dead buds, between external scales that are partially spaced out, where they are attacked by Chaetostricha walkeri Forster (Hymenoptera: Trichogrammatidae) and a Telenomus sp. (Hymenoptera: Scelionidae) (Conti et al. 1997, Roversi et al. 1998) which belongs to the T. laricis Walker species group (Huggert 1983, Johnson 1984); both are new host records. Parasitoid impact on the predator, pooled for the two species, averaged 17% in 1994 (Conti et al. 1997, Roversi et al. in preparation). Orange groves in Sicily are attacked by the phytophagous mirid Calocoris trivialis (Costa), which causes apical deformation of shoots, leaf necrosis and drop of flower buds (Barbagallo 1970). This mirid also has one generation per year and overwin- ters in the egg stage, but the eggs are em- bedded in the soft dead and decaying wood of old pruning wounds. These eggs are attacked by Aprostocetus miridivorus (Domenichini) (Barbagallo 1969, 1970, Graham 1987), Aprostocetus new species near miridivorus (Hymenoptera: Eulophi- dae) (Conti et al. 1991, 1997) and Telenomus lopicida Silvestri (Hymenoptera: Scelioni- dae) (Barbagallo 1970, Conti et al. 1991, 1997). Total parasitoid impact averaged 57-70% in the different years and loca- tions (Barbagallo 1969, 1970, Conti et al. 1991, 1997, Roversi et al. in preparation). Aprostocetus miridivorus and T. lopncida were also recorded from overwintering eggs of the mirid Capsodes lineolatus, deep- ly embedded inside incisions in herba- ceous plants (Silvestri 1932, 1939, Graham 1987). However, it is unknown whether the same parasitoids shift between two al- ternative hosts or if they are two different biotypes, with different habitat preferenc- es. Some aspects of these mirid — egg par- asitoid associations on oak and orange groves are described in this paper by com- bining observations on host oviposition sites and parasitoid morphological adap- tations, with the aim of defining oviposi- tion strategies. MATERIALS AND METHODS Sampling Procedures Oak groves. — Field research on egg par- asitoids associated with C. quadripunctatus was carried out in five permanent sam- pling areas with mixed stands of Quercus pubescens Will, and Q. cerris L., ranging from 350 m to 700 m above sea level, in central Tuscany (central Italy). Quercus pu- bescens is included in the overstorey, as the masts are more palatable to livestock, while Q. cerris is part of the understorey and is used mainly for timber. In this landscape of gently rolling hills, oak stands normally alternate with vineyards and olive groves. During the winters of 1994 and 1996, 36 branches, 1 m long, were collected in each sampling area (total 180 branches) and transferred to the laboratory. All dead buds were counted, collected and kept un- der a shelter at outdoor conditions. In spring, when nymphs and parasitoids had emerged, buds were dissected and exam- ined under the stereomicroscope. Orange groves. — Overwintering eggs of C. trivialis, embedded in the soft decayed wood of the pruning wounds, were col- lected on 9-10 January 1991 and 2-5 Feb- ruary 1992 in 6 orange, Citrus sinensis (L.), groves, the same both years, in 5 localities of the Provinces of Catania and Siracusa (Sicily, southern Italy). Pruning wounds of different sizes and suitable for oviposition by C. trivialis were sampled randomly over the whole area of the groves. Twenty (20) to 62 samples per grove were collected in 1991 (total 224 Volume 9, Number 2, 2000 387 samples), 30 per grove in 1992 (total 180 samples). The samples were then trans- ferred to the laboratory and kept under controlled conditions (25 ± 1 °C, 60%-95% RH; photoperiod L:D 14:10) in different kinds of screened containers (30 X 160 mm glass tubes, 140 X 25 mm Petri dishes, 185 X 140 X 290 mm plastic food contain- ers), depending on the size of the samples. All the material was examined under the stereomicroscope in summer, when emer- gence of nymphs and parasitoids was complete. Host Oviposition Sites and Parasitoid Structures For both species of Calocoris, eggs were classified in the following categories: — healthy (i.e., eclosed), — parasitized (i.e., containing the parasitoid or with the pres- ence of an emergence hole), — dead due to other agents (predators or unknown). Egg length was measured under a stereomi- croscope and the characteristics of healthy and parasitized eggs were described. In the case of C. quadripunctatus, the distri- bution of healthy and parasitized eggs in each cluster was mapped. The position of the parasitoid piercing point on the host egg was recorded when visible under the stereoscopic or com- pound microscope. However, on eggs of C. trivialis such parasitoid punctures are often hidden by remains of wood that are glued to the chorion, due to a sticky se- cretion from the ovipositing female. In an attempt to dissolve this secretion, eggs were sonicated at room temperature for at least 15 minutes in different solvents (wa- ter, ethanol, acetone, chloroform), or soaked in boiling chloroform (61.2 °C) in a beaker inside a water bath for 15 min- utes. Parasitoids' structures used for reaching the concealed host eggs were evaluated by measuring the total body length, metaso- ma size and ovipositor length of 5 or 10 females of each species. To measure ovi- positor length, the two scelionids were cleared with potassium hydroxide and mounted on slides. Measurements were made using a micrometer eyepiece mount- ed on a compound microscope, or under a Nikon E 600 microscope connected to a JVC TK-C1380 video camera, and using a computer program for visual image anal- ysis (Lucia 3.5). RESULTS AND DISCUSSION Host oviposition sites. — The eggs of C. quadripunctatus, concealed among the scales of dead buds on oak, are laid in clusters, most frequently with 3 to 5 eggs, and nothing appears externally to indicate their presence. In 1994 and 1996 buds with eggs contained on average 6 to 8 clusters each (Roversi et al. in preparation) (Figs. 1, 3 and 4). In contrast, the eggs of C. tri- vialis, concealed in the pruning wounds on orange trees, are laid singly and distrib- uted variably on the wound surface where the decaying wood is soft enough for ovi- position. The wounds sampled in 1991 and 1992 had a very variable diameter, from 13 to 98 mm, with an average of 45 mm. Wounds with eggs contained on av- erage 8.3 eggs each in 1991 and 5.1 in 1992. Normally these eggs were deeply embed- ded, with the operculum at 386 ± 37.2 |xm (Mean ± SEM, n = 10) under the substrate and not visible externally, although par- tially exposed eggs were also found in rare cases. The oviposition incisions, that indicate egg presence, may close partially or completely when rainwater swells the wood, thus becoming inconspicuous or to- tally invisible (Figs. 2, 11 and 12). Both the dead buds on oak and the pruning wounds on orange trees con- tained recently laid eggs and eggs that had been laid during previous seasons, thus indicating that they provide a suit- able oviposition substrate for several years. Also, because of wood erosion by atmospheric agents, old egg shells of C. trivialis on pruning wounds were often more or less exposed and directly visible 388 Journal of Hymenoptera Research Figs. 1-2. 1, Dead oak bud partially dissected to show egg clusters of Calocoris quadripunctatus concealed between scales. 2, Pruning wound with soft decaying wood on orange tree and, in the inserted photo, a cross section partially exposing an embedded egg of Calocoris trivialis. under the stereomicroscope or even to the naked eye. Parasitoids' morphological adaptations and strategies. — How do the parasitoids reach and attack such concealed and, therefore, protected eggs? The egg parasitoids of C. quadripuncta- tus in oak buds have developed two dif- ferent morphological adaptations and strategies (Figs. 5, 6, 8 and 9). The tricho- grammatid Chaetostricha walkeri has a very long ovipositor, 3.15 times the length of its metasoma. The proximal end reaches the prothoracic coxae, where it is articulated on a special structure, probably used as an extension device (Figs. 5 and 6, Table 1). In contrast, the scelionid Telenomus sp. lar- icis group has a depressed (dorso-ventral- ly flattened) metasoma, 69% of the length of the whole body, which facilitates its in- sertion between bud scales. The ovipositor is 86% of the length of the metasoma and is invaginated in it, from which it is ex- tended for parasitization (Figs. 8 and 9; Table 1). Similarly, lygaeid egg parasitoids in Telenomus and Eumicrosoma also have depressed metasoma (A. Polaszek, pers. comm.). In spite of these adaptations, both C. zvalkeri and Telenomus sp. laricis group can hardly reach the most internal eggs, as only the first (peripheral) and second (sub- peripheral) ones are normally attacked, al- though there are a few exceptions (Roversi et al. in preparation). In addition, both par- asitoids attack the peripheral egg more frequently than the sub-peripheral one, al- though C. walkeri can reach both much more often than Telenomus sp. (Fig. 19). Whether in such cases one species can dis- criminate between healthy eggs and eggs parasitized by the conspecific or other spe- cies is unknown. Similar differences of ef- fectiveness, due to metasoma morphology and ovipositor length, were also observed on species attacking egg masses of the rice Volume 9, Number 2, 2000 389 8 H 1mm Figs. 3-10. Calocoris quadripunctatus and its egg parasitoids. 3, Schematic section of a dead oak bud showing C. quadripunctatus egg clusters concealed between scales. 4, Detail of a scale with a duster of lour eggs. 5 and 6, Lateral and dorsal views of Chaetostriclia walkcri Forster showing the long ovipositor 7 I lost egg parasitized by C. walkeri showing the parasitoid piercing point on the chorion. 8 and l>, I ateraJ and dorsal views of Telenomus sp. laricis group showing its depressed metasoma with the ovipositor indicated (dotted line). 10, Host egg parasitized by Telenomus sp. laricis group showing the parasitoid piercing point on the chorion and the characteristic transversal band. 390 Journal of Hymenoptera Research 17 1mm 1mm Figs. 11-18. Calocoris trivialis and its egg parasitoids. 11, Example of a pruning wound on orange showing a distribution of C. trivialis eggs (circles). 12, Schematic section of a host oviposition incision to indicate egg position in relation to the substrate. 13 and 14, Lateral and dorsal views of Aprostocetus n. sp. near miridivorus. 15, Host egg parasitized by Aprostocetus n. sp. near miridivorus showing the parasitoid piercing point on the chorion. 16 and 17, Lateral and dorsal views of Telenomus lopicida showing its compressed metasoma with the long ovipositor indicated (dotted lines). 18, Host egg parasitized by T. lopicida showing the parasitoid piercing point on the chorion and the characteristic transversal band. Volume 9, Number 2, 2000 391 Table 1. Measurements (mean ± SEM; in mm) of female parasitoid body parts to show morphological adaptations for reaching the host egg concealed between dead bud scales (C. quadripunctatus) or in decaying wood (C. trivialis). Host Body length Metasoma dimensions Parasitoid Length Lateral Max. Min. Dorso-ventral (max.) Ovipositor length Chaetostricha walkeri" C. 4-ptmctatus 0.843 ± 0.022 0.479 ± 0.015 0.268 + 0.005 0.274 ± 0.010 1.511 ± 0.029 Telenomus sp. - C. 4-pitnctatus 0.796 ± 0.020 0.548 ± 0.010 0.141 + 0.002 — 0.053 ± 0.002 0.472 ± 0.004 Aprostocetus sp : C. trivialis 1.230 £ 0.053 0.652 ± 0.017 0.284 ± 0.008 — 0.330 ± 0.014 0.617 ± 0.025 Telenomus lopicidam C. trivialis 1.095 ± 0.018 0.635 ± 0.015 0.162 ± 0.004 0.027 ± 0.005 0.146 ± 0.007 0.832 ± 0.039 10; 5. stemborer Scirpophaga incertulas (Walker) (as Tryporyza incertulas Walker) (Lepidop- tera: Pyralidae) in Vietnam (Vu and Nguyen 1987) and on Telenomus busseolae Gahan attacking the maize stemborer Se- samia nonagrioides (Lefevre) (Lepidoptera: Noctuidae) in Greece (Alexandri and Tsit- sipis 1990). The egg parasitoids of Calocoris trivialis on orange have also developed two dif- ferent strategies (Figs. 13, 14, 16 and 17). The eulophid Aprostocetus n. sp. near mir- idivorus shows no apparent morphological adaptations. The ovipositor is 95% of the length of the metasoma and is probably introduced into the incision or inserted by drilling through the wooden substrate (Figs. 13 and 14; Table 1). Laboratory ob- servations on the oviposition behavior of A. miridivorus towards Calocoris norvegicus (Gmelin), indicate that the parasitoid in- tensely antennates oviposition incisions containing host eggs and, in most cases, inserts its ovipositor through the incision, although it also appears to drill through the dead wood (Conti and Bin unpub- lished). This is especially important in the open field where, because of wood swell- O250 9- O 200 - o S 150 O) £ 100 ■o o N (A (0 i- (0 Q. IT 50 Parasitized egg (black colored): MM peripheral sub-peripheral periph. + sub-periph. A3 '* VS///A Telenomus sp. C. walkeri Fig. 19. Frequency of Calocoris quadripunctatus egg clusters (4 eggs per cluster on average) parasitized by Telenomus sp. laricis group or Chaetostricha walkeri, related to the position of the parasitized egg/s in the ( luster. For the sake of clarity, the very few cases of more than 2 parasitized eggs per cluster have not been represented in the graph. 392 Journal of Hymenoptera Research ing caused by rain, the incisions can be that this area can be excluded as a piercing partially or completely closed. site by the parasitoid. Such parasitized The scelionid T. lopncida presents a much eggs of C. trivialis were also characterized more evident adaptation. The distal half by a light amber-colored chorion and ap- (53% of total length) of its metasoma is lat- peared darker close to eclosion because of erally compressed and its distal quarter the presence of the darkening pupa or the (23%) is only 27 fxm wide (Figs. 16 and 17; adult, visible within. Similar to C. quadri- Table 1). This strongly facilitates its intro- punctatus, eggs of C. trivialis parasitized by duction into the host oviposition incisions, T. lopncida show a dark transverse band at as has been observed with eggs of Capso- the level of the meta thorax (Fig. 18) while des lineolatus (Brulle) (Heteroptera: Miri- those parasitized by Aprostocetus sp. do dae) embedded in stems of Scabiosa col- not show any particular pattern (Barba- umbaria L. (Silvestri 1932, 1939). In addi- gallo 1969, Mineo and Sinacori 1978) (Fig. tion, the ovipositor of T. lopncida is 1.31 15). times the length of the metasoma and is Escaping from the host substrate is an- partially contained in a dorso-metasomal other critical situation for which the hump (Figs. 16 and 17; Table 1). emerging parasitoids, both males and fe- The parasitoid strategies described males, need a morphological adaptation above are also confirmed by the position (Quicke 1997). Parasitoids of C. quadri- of the parasitoid piercing point on the egg punctatus in most cases escape from the chorion (Figs. 7, 10, 15 and 18). In the case buds by crawling out between the scales, of C. quadripunctatus, this point is evi- although C. walkeri may also chew circular denced by the presence of a brown circu- holes in the scales. In contrast, both para- lar area localized on the exposed side of sitoids of C. trivialis have to chew a hole the egg (Figs. 7 and 10). Such eggs, when of variable length in the soft decaying parasitized by C. walkeri, are visibly swol- wood, depending on their distance from len and light amber colored. In contrast, the surface. While in most cases the emer- those attacked by Telenomus sp. laricis gence hole of both species is localized in group are slightly opaque, with a dark the apical third of the egg, including the transverse band evident from the pupal egg cap area, sometimes T. lopncida pu- stage of the parasitoid onwards, and ap- pates upside-down and therefore has to pear darker close to eclosion, due to the chew a much longer tunnel in the wood, presence of the pupa or the adult, visible This is obviously performed with mandi- within. bles well adapted to chew plant tissues. In the case of C. trivialis, the parasitoid No other adaptive cephalic feature is piercing point is less recognizable under presented by T. lopncida since its frons is the microscope because the darkened area smooth, whereas other species, also be- is often not appreciable and because part longing to the T. laricis group, exhibit a of the chorion is generally hidden by marked scale-like sculpture on the frons wood remains glued to it. These remains and vertex (Johnson 1984) probably useful could be removed only mechanically, for escaping in combination with mandi- when possible at all, as the glue is not sol- bles (Bin and Conti unpublished) and uble in any of the solvents tested. The have a pointed head, the distinctive fea- piercing points, when distinguishable, ture of the laricis group (A. Polaszek, pers. were localized on the apical third of the comm.). concave egg side, under the egg cap area (Figs. 15 and 18). No piercing points on CONCLUSIONS the egg cap were observed and, indeed, Ideally, host-parasitoid associations due to its thickness and hardness, it seems should be defined using a complete set of Volume 9, Number 2, 2000 393 characters, ranging from physical and chemical cues for habitat and host loca- tion, to physiological and biochemical in- teractions for host suitability. However, some pieces of such a complex mosaic can be provided by comparing the features of the microhabitat, selected by the host to escape adverse climatic events, with the ability of the parasitoid to overcome such physical barriers in order to reach the host and eventually emerge from it. It is clear that in some cases the mor- phology of the ovipositor (Austin 1983), the metasoma and, possibly, other body parts are evolutionarily linked to the ex- ploitation of particular hosts. Therefore, in the future these morphological adapta- tions may be used to predict the most like- ly host groups or oviposition sites (Austin 1983). The parasitoid species considered here appear to be well adapted for reaching their concealed host eggs, and the mor- phological adaptations especially involve the metasoma and /or the ovipositor sys- tem. Their oviposition strategies can be defined by comparison with similar strat- egies described in the literature (Gauld and Hanson 1995, Smith et al. 1993; Smith and Wiedenmann 1997), although more direct observations are needed to better understand their behavior. Chaetostricha ivalkeri inserts its long ovipositor between bud scales and is therefore an "ovipositor prober". Aprostocetus n. sp. near miridivo- rus is both an "ovipositor prober" and an "ovipositor driller", as it probably inserts its terebra either inside the host oviposi- tion wound or through the soft wooden substrate. In contrast, both Telenomus spe- cies are "metasomal probers". In fact, their metasoma is adapted to reach the host eggs by inserting at least part of it into the host oviposition sites, specifically between bud scales in the case of Teleno- mus sp. laricis group, or in the host ovi- position incision in the case of T. lopicida. We have shown that Chaetostricha ivalk- eri is an egg parasitoid of Calocoris quad- ripunctatus. This is a new host record and it suggests that previous ones were erro- neous. In fact, C. walkeri has been reported from eggs of Tortrix viridana (Kolubajiv 1959 according to Du Merle 1983, Marti- nek 1963), from the coccid Leucaspis pini Htg. (Nikol'skaya 1952), or from unknown hosts, supposedly xylophagous larvae (Silvestri 1917 citing Foster) or Heterop- tera eggs embedded in wood (Silvestri 1917). The eggs of C. quadripunctatus, the true host, were evidently overlooked be- cause they were not visible externally. Many of the available host records for other Telenomus species in the laricis group are similarly questionable, and a careful reassessment is needed before supposed host associations can be accepted as reli- able. ACKNOWLEDGMENTS We are grateful, for insect identifications, to Paride Dioli (Calocoris quadripunctatus), John LaSalle (Apros- tocetus n. sp. near miridivorus) and Gennaro Viggiani (Chaetostricha ivalkeri). We also thank the reviewers for their useful comments on the manuscript, Sebas- tiano Barbagallo and Santi Longo for collaboration and hospitality in Sicily, Beyene Ayano Ejigu for helping with data collection on C. trivialis in the lab- oratory, Mario Michelassi and Aldo Mommi for pho- tographs and technical assistance, and Fabrizio Pen- nacchio for parasitoid measurements using the visual image analysis program. This research was financial- ly supported by MURST 40%, MURST 60% and MiPA Research Project "Biological and integrated pest control in agricultural and forest plants", re- search group "Forest plants". LITERATURE CITED Alexandri, M. P. and J. A. Tsitsipis. 1990. Influence of the egg parasitoid Platytelenomus busseolae (1 Ivm : Scelionidae) on the population of Sesamia nona grioides (Lep.: Noctuidae) in central Greece. £;/- tomophaga 35(1): 61-70. Alomar, O. and R. N. Wiedenmann (eds.). 1996. Zoophytophagous Heteroptera: Implications for life luston/ and integrated pest management. The I nto- mological Society of America ["nomas Say Pub- lications in Entomology, Lanham, Maryland, 202 PP- Austin, A. D. 1983. Morphologv and mechanics of the ovipositor system of Ceratobaeus Ashmead (Hy- menoptera: Scelionidae) and related genera. In 394 Journal of Hymenoptera Research ternational Journal of Insect Morphology and Embry- ology 12(2/3): 139-155. Barbagallo, S. 1969. Appunti morfo-biologici su Te- trastichus miridivorus Domenichini (Hymenop- tera: Eulophidae) parassita oofago di Eterotteri Miridi. Bollettino di Zoologia Agraria e di Bachicol- tura (Ser. 11) 9: 115-122. Barbagallo, S. 1970. Contributo alia conoscenza del Calocoris (Closterotomus) trivialis (Costa) (Rhyn- chota-Heteroptera, Miridae). Morfologia dell'adulto e biologia. Entonwlogica 6: 1-104. Bogenschutz, H. 1978. Tortricinae. Pp. 55-89 In: W. Schwenke (ed.), Die Forstschddlinge Europas. 3 Band. Paul Parey, Hamburg, Berlin. Cobben, R. H. 1968. Evolutionary trends in Heteroptera. Part I. Eggs, architecture of the shell, gross embry- ology and eclosion. Centre for Agricultural Pub- lishing and Documentation, Wageningen, 475 pp. Conti, E., S. Colazza and F. Bin. 1991. Imenotteri par- assitoidi di uova endofitiche di Eterotteri: alcune associazioni nei Miridi e nei Nabidi. Atti XVI Congresso Nazionale Italiano di Entomologia, Bari— Martina Franca, 23-28 Sept. 1991: 929-931. Conti, E., P.F. Roversi and F. Bin. 1997. Egg parasit- oids associated with arboreal mirids Hym.: Par- asitica; Het.: Miridae). Proceedings from the 6'h European Workshop on Insect Parasitoids. Va- lencia, 1-^ March 1997. Boletin de la Asociacion Es- pafwla de Entomologia, Suplemento al Volumen 21: 94-95. Du Merle, P. 1983. Les facteurs de mortalite des oeufs de Tortrix viridana (L.) (Lep., Tortricidae). Agron- omic 3(5): 429^34. Gauld, I. D. and P. E. Hanson. 1995. The structure of Hymenoptera. Pp. 102-137 In: P. E. Hanson and I. D. Gauld (eds.) The Hymenoptera of Costa Rica. Oxford University Press, Oxford, 893 pp. Graham, M. W. R. de V. 1987. A reclassification of the European Tetrastichinae (Hymenoptera: Eulo- phidae), with a revision of certain genera. Bulle- tin of the British Museum (Natural History). Ento- mology series 55(1): 116-117. Hinton, H. E. 1981. Biology of insect eggs. Vol. I-III. Pergamon Press, Oxford, 1125 pp. Huggert, L. 1983. On Telenomus, mainly European; redescriptions, new taxa, synonymies and com- binations (Hymenoptera, Proctotrupoidea: Sce- lionidae). Entomologica Scandinavica 14: 145-167. Johnson, N. F. 1984. Systematics of Nearctic Teleno- mus: classification and revisions of the podisi and phymatae species groups (Hymenoptera: Scelion- idae). Bulletin of the Ohio Biological Survey 6(3): 1- 113. Kolubajiv, S. 1959. Orientacni pokusy s umelym chovem drobnenky rodu Trichogramma z hmy- zich lesnich skudcu. Shorn CSAZV-Lesnictvi, 32(12): 1057-1070. Kullenberg, B. 1946. Studien iiber die biologie der Capsiden. Zoologiska Bidrag 23: 1-522. Martinek, V. 1963. Nekteri vajecni cizopasnici lesnich skudcu a moznosti jejich pestovani. Prdce Vy- zkumnych Ustavu Lesnickych CSSR 26: 5^-8. Mineo, G. and A. Sinacori. 1978. Studi morfo-biolo- gici comparativi sugli stadi preimmaginali degli Scelionidi (Hym. Proctotrupoidea). IV. Nota su Telenomus lopicida Silv. Bollettino dell'Istituto di Entomologia Agraria e dell'Osservatorio di Fitopato- logia di Palermo 10: 105-112. Nikol'skaya M. N. 1952. The Chalcid Fauna of the U.S.S.R. (Chalcidoidea). Keys to the fauna of the U.S.S.R. No. 44. Zoological Institute of the Acad- emy of Sciences of the U.S.S.R., Moscow, Lenin- grad (Translated from Russian. Israel Program for Scientific Translations, Jerusalem 1963). Quicke, D. L. J. 1997. Parasitic wasps. Chapman & Hall, London, 470 pp. Roversi, P. F., E. Conti and F. Bin. 1998. Ooparassi- toidi di Miridi arboricoli di ambienti forestall e agrari. Atti XVIII Congresso Nazionale Italiano di Entomologia, Maratea, 21-26 giugno 1998: 225. Silvestri, F. 1917. Contribuzione alia conoscenza del Genere Centrobia Forster (Hymenoptera, Chalci- didae). Bollettino di Zoologia Generate e Agraria del- ta R. Scuola Superiore d'Agricoltura in Portici 12: 245-251. Silvestri, F. 1932. Contribuzione alia conoscenza del Lopus lineolatus (Brulle) e di un suo parassita. So- ciete Entomologique de France, Livre du Centenaire, pp. 561-565. Silvestri, F. 1939. Lopus lineolatus Brulle. Pp. 262-269 In: F. Silvestri, Compendio di Entomologia Applica- ta. Parte speciale Vol. 1. Tipografia Bellavista, Por- tici. Smith, J. W. Jr. and R. N. Wiedenmann. 1997. For- aging strategies of stemborer parasites and their application to biological control. Insect Science and its Application 17(1): 37-49. Smith, J. W. Jr., R. N. Wiedenmann and W. A. Ov- erholt. 1993. Parasites of lepidopteran stemborers of tropical gramineous plants. ICIPE Science Press, Nairobi, 89 pp. Southwood, T. R. E. 1956. The structure of the eggs of the terrestrial Heteroptera and its relationship to the classification of the group. Transactions of the Royal Enomological Society of London 108: 163- 221. Vu, Quang Con and Van San Nguyen. 1987. Effec- tiveness of egg parasites (Hymenoptera) depend- ing on the parasite abdomen structure and on the type of egg-batches in lepidopteran rice pests. Zoologichesky Zhurnal 66(1): 60-65. [In Russian]. Wagner, E. and H. H. Weber. 1964. Faune de France. Vol. 67. Heteropteres Miridae. Federation Francaise des Societes de Sciences Naturelles, Paris, 591 pp. J. HYM. RES. Vol. 9(2), 2000, pp. 395-415 New Caledonian Tiphiidae: Revision of the Genus Eirone (Hymenoptera: Thynninae) Lawrence A. Baptiste and Lynn S. Kimsey Bohart Museum of Entomology, Department of Entomology, University of California, Davis, California 95616 Abstract. — Ten new species of New Caledonian Eirone are described based on males, including anone Kimsey, invini Baptiste, koghisica Kimsey, laniensis Baptiste, maigretae Baptiste, nasalis Kimsey, nepouiensis Baptiste, paniensis Kimsey, rivierensis Baptiste, and zvebbi Kimsey. In ad- dition, females of invini, transversa Brown, nepouiensis, and xvebbi, are also described. Male genitalia are described and illustrated for both the new species and for the first time for colorata Brown, marginata Brown, fieocaledonica Turner, nigra Brown, obtusidens Turner, salteri Brown, subtuberculata Brown and transversa Brown. These previously described species are also rediagnosed. A key to males of the species of New Caledonian Eirone is also included. As recently as the early 1980's only nine species of Eirone Westwood were known from New Caledonia (Brown 1984). Sub- sequent malaise trap collecting by Michael Irwin, Evert and Marion Schlinger and Donald Webb have demonstrated that the thynnine fauna of New Caledonia is sub- stantially richer than previously thought. Their efforts have turned up ten undescri- bed species of Eirone. Based on their col- lecting efforts, Eirone appears to be the only thynnine genus present on New Cal- edonia. MATERIALS AND METHODS Specimens were obtained for this study from the following institutions and indi- viduals: Bernice P. Bishop Museum, Hon- olulu, Hawaii, G. Nishida (HONOLULU); Illinois Natural History Survey, Universi- ty of Illinois, Champaign-Urbana, M. E. Ir- win (URBANA), Canadian National Insect Collection, Ottawa, Ontario, L. Masner (OTTAWA), University of California, Da- vis, S. L. Heydon (DAVIS). Holotypes of the new species described below will all be deposited in the Museum Nationale d'Histoire Naturelle, Paris (PARIS). The type of Eirone impunctata Brown was to have been deposited in the collection of the Rydalmere Quarantine Station, NSW. This collection has been moved to the Ag- ricultural Scientific Collections Unit, NSW Agriculture, Orange Agricultural Institute, Orange, Australia (P. S. Gillespie). How- ever, the type cannot be located. Specimen and type repositories are indicated in the text by the city name in capital letters. The following abbreviations are used for the sake of brevity: F = flagellomere, MOD = midocellus diameter, PD = punc- ture diameter. SPECIES DESCRIPTIONS Eirone anone Kimsey, new species (Fig. 19) Male. — Body length 15-18 mm; fore- wing length 12-14 mm; face with tiny con- tiguous punctures between inner eye mar- gins and antennal sockets; frons with punctures 0.5-2 apart, except impunctate along medial sulcus and adjacent to anten- nal socket; clypeus with punctures larger than along inner eye margin, punctures contiguous; clypeal apex broadly triangu- lar, with single apicomedial lobe; F-I length 2.4-2.5 X breadth; F-II length 3.0- 396 Journal of Hymenoptera Research Pouebo Pourma Noumea Fig. 1. Distribution map of Eirone species in New Caledonia. 3.2 X breadth; pronotal punctures 1-3 PD apart; mesopleural punctures 1-2 PD apart separated by dense, fine shagreen- ing; mesopleural lamella posteriorly rounded; scutal punctures densest along notauli (nearly contiguous), medially sparsest (0.5-1 PD apart); scutellum nearly impunctate medially, punctures denser laterally, 0.5-1.0 PD apart; propodeum punctate medially and laterally, punctures 0.5-1.0 PD apart, impunctate sublaterally; metasomal sternum I sharply keeled and somewhat hooked posteriorly, abruptly declivous posteriorly; epipygium and hy- popygium flattened and apex narrowly subtruncate; genital capsule (Fig. 19): gon- ocoxa with medial angular lobe on inner surface; aedeagus apical column broad, and flattened apically, ending abruptly with rounded basal angle and narrow, hooked apical lobe, column extending more than half its length beyond penis valve bending dorsally; penis valve with short dorsal hook, hooked ventral lobe and elongate basal lobe; volsella with hooked bilobate dorsal surface and obso- lescent asetose ventral surface; color: or- ange, with extensive black markings on vertex, extending down along side anten- nal sockets, on occiput, pronotum anteri- orly and medially, scutum except large yellow medial spot, mesopleural margins, scutellum anteromedially, coxae dorsally, propodeum anterior margin black to en- tirely black, metasomal segments I and VI entirely black, segment VII anterior mar- gin black; gonocoxa orange; wing mem- brane yellow-tinted, veins black; vestiture long, erect and orange-colored. Type material. — Holotype 6; trail to Mt. Panie, 22 km NW Hienghene, 11-25 Nov. 1992, Webb & Schlinger, malaise trap (PARIS). Paratypes: 15 6 6, same data as holotype; 3 6 6, Mt. Mandjanie, 5.3 km WSW Pouebo, 500 m, 9-26 Nov. 1992, D. W. Webb, malaise trap (DAVIS, ILLI- NOIS). This species was collected in No- vember. Etymology. — The species name, anone, is a nonsense combination of letters, as- sumed to be feminine. Discussion. — This species closely resem- bles obtusidens in coloration, presence of a midcoxal spine, rounded mesopleural la- mella, and unilobate, densely setose clyp- eus. It can be immediately separated from obtusidens by the strongly keeled metaso- mal tergum I and apically acute aedeagus. The unilobate clypeus is unique to these two species. Eirone colorata Brown (Fig. 9, 22) Eirone colorata Brown 1984:254. Holotype 6: New Caledonia: Col de Ho (HONOLULU), examined. Male. — Body length 12-15 mm; fore- Volume 9, Number 2, 2000 397 wing length 9-12 mm; face with tiny con- tiguous punctures between inner eye mar- gins and antennal sockets; frons with punctures 1—4 PD apart; clypeus with punctures larger than along inner eye margin, punctures 0.5-1.0 PD apart; clyp- eal apex broadly truncate, 2.6-2.8 MOD wide; F-I length 2.2 X breadth; F-II length 2.8 X breadth; pronotal punctures \-A PD apart; mesopleural punctures 1-2 PD apart, finely, shagreened between; meso- pleural lamella rounded apically; scutum medially, with punctures 1.0-1.5 PD apart, punctures becoming finer and denser lat- erally, 0.5-1.0 PD apart; scutellar punc- tures 1-4 PD apart; propodeum punctate medially and laterally, punctures 0.5-1.0 PD apart, impunctate sublaterally, finely and densely shagreened between punc- tures; midcoxa with apical spine (Fig. 9); metasomal sternum I keeled medially, keel extending two-thirds of length, slop- ing obliquely to posterior margin; epipy- gium and hypopygium flattened and broadly rounded apically; genital capsule (Fig. 22): gonocoxa with low submedial angular projection on inner surface; aedea- gus unmodified, with slender apical col- umn extending less than half its length be- yond penis valve, and bending ventrally; penis valve saddle-shaped with short dor- sal hook; volsella with small bilobate dor- sal projection and ventral surface with 6- 8 long erect setae; color orange, becoming yellower on face, pronotum, tegula and subalar region of mesopleuron; scape or- ange, flagellum dark brown, black mark- ings on vertex, pronotal middle, scutum black except large medial mark and sides yellow, scutellum anteriorly black, pro- podeum anteriorly, mesopleuron along dorsal margin, coxae and femoral dorsal margin, metathorax laterally, apical ab- dominal segments darker; wing mem- brane yellow-tinted, veins black; pubes- cence silvery on head and thorax and black on metasoma. Material examined. — 21 6 6 (including the holotype); Col de Ho, 22 km NW Hienghene (trail to Mt. Panie), and Mt Panie, 9.7 km NW Sarramea. These spec- imens were collected in the months of Oc- tober, November, January and February. Discussion. — Eirone colorata most closely resembles rivierensis based on the emar- ginate mesopleural lamellae, midcoxal spine, metasomal sternum I obliquely sloping posteriorly, and apically truncate clypeus. The two species differ in the ves- titure of the abdominal terga, rivierensis has dense tufts of long black setae on T- VI and VII, which are lacking in colorata, and the aedeagus of colorata is small and unremarkable, whereas the apical aedea- gal column in rivierensis is broadly flat- tened and subtriangular. Finally, colorata specimens are at least partly reddish or- ange, whereas rivierensis specimens lack this reddish orange coloration. Eirone emarginata Brown (Fig. 23) Eirone emarginata Brown 1984:256. Holotype 8; New Caledonia: Col des Rousettes (HONO- LULU), examined. Male. — Body length 16-21 mm; fore- wing length 11-15 mm; face with tinv, contiguous punctures between inner eye margins and antennal sockets; frons with punctures irregular, 0.5-1.0 PD apart, ex- cept impunctate medial sulcus; clypeus convex medially, with punctures larger than along inner eye margin, punctures contiguous to 0.5 PD; antennal lobe with small subsidiary ventrally projecting lobe; clypeal apex medially emarginate, acutelv pointed on either side, 2.6 MOD wide; F- I length 2.5X breadth; F-II length 3.0X breadth; pronotal punctures 0.5-1.0 PD apart; mesopleural punctures 0.5 PD apart; mesopleural lamella short, posteri- orlv emarginate; scutal punctures medial- ly 0.5-1.0 PD apart, becoming finer and nearly contiguous laterally; scutellar punctures large and 0.5-1.0 PD apart near midline becoming smaller, denser and nearly contiguous laterally; propodeum 398 Journal of Hymenoptera Research 2. koghisica 3. laniensis 8. webbi 14. salteri 15. rivierensis 16. irwini Figs. 2-16. Eirone species. 2-6, Front view of male face, with right antenna removed. 7, Lateral view of male face. 8, 10, Ventral view of male meso- and metathorax. 9, Oblique ventral view of male meso- ^nd metathorax. 11, Ventral view of male mesothorax. 12-15, Lateral view of male metasomal segment 1. 16, Ventral view of male metasomal segment 1. Abbreviations: ct = clypeal truncation, MOD = midocellus diameter, td = tran- santennal distance. Volume 9, Number 2, 2000 399 17. neocaledonica 18. salteri 20. laniensis 21 . maigretae 19. anone 22. colorata 26. laniensis 27. maigretae 28. nasalis Figs. 17-28. Eirone species. 17-19, Dorsal view of male epipygium. 20-21, Lateral view oi male abdominal apex. 22-28, Interior surfaces of genital capsule. Abbreviations: a = aedeagus, g = gonocoxa, pv penis valve, v = volsella. 400 Journal of Hymenoptera Research 29. neocaledonica 30. nepouiensis 31. nigra 32. obtusidens 33. paniensis 34. rivierensis 35. sa/ten 36. subtuberculata 37. transversa 38. webb/ Figs. 29-38. Eirone species, Interior surfaces of genital capsule. Left side gonocoxa and volsella. Right side penis valve uppermost. Inner margins of both sides are the ventral surface of the genital capsule; outer margins are the dorsal surface. 35, Right side of genital capsule only. Abbreviations: a = aedeagus, g = gonocoxa, pv = penis valve, v = volsella. punctate medially and laterally, punctures extending half length, sloping obliquely to 1-2 PD apart, impunctate sublaterally; posterior margin; epipygium strongly con- midcoxa without apical spine; metasomal vex, posterior margin broadly rounded, sternum I with short, carinate medial keel with slight indentation medially; hypo- Volume 9, Number 2, 2000 401 pygium flat, broadly rounded apically; thorax and metasoma finely and densely shagreened between punctures; genital capsule (Fig. 23): gonocoxa narrowed me- dially, inner surface smooth without pro- jections, ridges or lobes; aedeagus apical column elongate, extending more than half its length beyond penis valve, bend- ing abruptly ventrally; penis valve leaf- like with small dorsal hook; volsella with large hoof-like dorsal lobe, ventral lobe with short spine-like setae; color black, with small, pale yellow spot on antennal lobe, genital capsule yellow to orange; wing membrane brown-tinted, veins black; pubescence silvery. Material examined. — 9 6 6 (including the holotype); 22 km NW Hienghene; Riviere Bleue Prov. Pk., km 25.8 Riviere Bleue Road; 7.5 km NW Sarramea; 17 km nne Noumea; Riviere Bleue Prov. Pk., trail to Valle de Pourina (DAVIS, HONOLULU, URBANA). Specimens were collected in October, November and February. Discussion. — As discussed under webbi, there are five species of New Caledonian Eirone that lack a midcoxal spine. Of these emarginata and maigretae are most similar. Both are large-bodied, with the epipygium broadly rounded apically and strongly convex in profile, and both have a sharply cornered clypeal truncation. Diagnostic features of emarginata include the medially emarginate clypeal truncation, black body and apically truncate hypopygium. Eirone impunctata Brown Eirone impunctata Brown 1984:256. Holotype 8: New Caledonia: Foret de la Thy (Reposito- ry?). Material examined. — No specimens of this species have been seen. The type is apparently unavailable. According to Brown (1984) the holotype was collected in April. Discussion. — See webbi. Eirone irwini Baptiste, new species (Figs. 16, 24) Male. — Body length 15-17 mm; fore- wing length 12-14 mm; face with small, nearly contiguous punctures between in- ner eye margins and antennal sockets; frons with punctures 0.5-1 PD apart, ex- cept nearly impunctate medial longitudi- nal band; clypeus with punctures much larger than along inner eye margin and nearly contiguous; clypeal apex weakly trilobate, with medial lobe extending fur- thest, distance between outer lobes 3.3 MOD; F-I length 2.4-2.5 X breadth; F-II length 3.4-3.5 X breadth; pronotal punc- tures 1-3 PD apart; mesopleural punctures 1-2 PD apart medially, finely shagreened between; mesopleural lamella apically rounded and bending slightly ventrally; scutal punctures medially 1-2 PD apart, becoming finer and denser laterally 0.5 PD apart; scutellar punctures 1-1.5 PD apart, except for medial longitudinal impunctate stripe; propodeum punctate medially and laterally (punctures 1-2 PD apart), im- punctate along sublateral band and above petiolar socket; metasomal sternum I sharply keeled medially, sloping obliquely to pointed apex, apex extending over base of metasomal sternum II (Fig. 16); epipy- gium narrowed and flattened apically, apical margin indented medially; hypo- pygium narrowly rounded apically; geni- tal capsule (Fig. 24): gonocoxa with pro- truding submedial longitudinal ridge; ae- deagus apical column arcuate, with apical flag, strongly bending dorsally; penis valve saddle-shaped, with long dorsal hook; volsella with large bilobate dorsal end, ventral part short, with brush of dense, stout setae. Body black, with yel- low or cream-colored markings on: lower half of inner eye margins, lower two- thirds of outer eye margins, spot on upper inner of eye margin, spot on antennal lobe, transverse band on anterior carina of pronotum, anterolateral spot on prono- tum, tegula basally, anterolateral edge of 402 Journal of Hymenoftera Research scutellum, submedial spots on metano- tum; wing membrane light brown, darkest near apex, with black veins; vestiture black on face and metasoma, silvery on rest of head and thorax. Female. — Body length 11 mm; head and thorax coarsely punctate, punctures con- tiguous to 0.5 PD apart; abdominal punc- tures slightly smaller and 0.5-1.0 PD apart; frons with medial longitudinal sul- cus, extending more than halfway to pos- terior head margin; clypeal margin cen- trally concave and pointed apically; F-I length 1.3X breadth, F-II length 1.6 X breadth; propodeum subrectangular in dorsal view, with transverse basal carina and lateral edge marked by sharp carina; T-I punctures longitudinally striatiform; metasomal sternum I with short medial longitudinal carina; body dark brown to black, with pale, silvery setae. Type material. — Holotype 6 (flagellom- eres affixed to point below specimen): NEW CALEDONIA, Sarramea, 24 Dec. 1991, M. E. Irwin (PARIS). Paratypes: 6 6 6, 3 6 6, same data as holotype; 1 6, 1 9 (in copula), Sarramea, 24-25 Dec. 1991, M. E. Irwin, D. W. Webb, malaise trap across forest opening; 1 6, trail to Mt. Panie, 22 km NW Hienghene, 11 Nov. 1992, D. W. Webb, E. & M. Schlinger, mal- aise trap in tropical rainforest, 457 m (DA- VIS, ILLINOIS). This species was collected in November and December. Etymology. — The species is named in honor of the collector, Michael E. Irwin. Discussion. — Males of both iriuini and ni- gra share the oddly keeled metasomal ster- num I, and both are black with pale mark- ings. However, irwini can be immediately distinguished by the presence of a mid- coxal spine and fully developed meso- pleural lamellae. The shape of the aedea- gus is also unique to irwini. Eirone koghisica Kimsey, new species (Fig. 2, 25) Male. — Body length 11-13 mm; fore- wing length 10-11 mm; face (Fig. 2), punc- tures widely separated on frons and ver- tex, 2-A PD apart, becoming smaller and denser toward inner eye margin, tiny and contiguous between inner eye margin and antennal socket; clypeal punctures contig- uous and somewhat striatiform medially; clypeal apex slightly flared anteriorly, slightly depressed medially when viewed in profile, apex weakly trilobate and 3 MOD wide between outer lobes; F-I length 2X breadth; F-II length 2.3 X breadth; pronotal and scutal punctures sparse and 2-A PD apart medially, becoming smaller and denser laterally, contiguous to 0.5 PD apart; mesopleural punctures 0.5-1.0 PD apart; scutellum impunctate basally, be- coming punctate posterolaterally, punc- tures 0.5-1.0 PD apart; propodeum finely shagreened, punctures 0.5-1.0 PD apart, with impunctate sublateral area; midcoxal spine present; abdominal segments finely shagreened, with scattered punctures 2-A PD apart; metasomal sternum I with me- dial keel extending three-fourths of total length, abruptly declivitous posteriorly; epipygium with narrowly rounded apex; hypopygium with narrowly rounded apex; genital capsule (Fig. 25): gonocoxa with submedial welt or swelling on inner surface; aedeagus apical column bending dorsally, with apex broadened; penis valve with large dorsal hook; volsella dor- sally broadened and subrectangular, ven- trally with brief row of dense short setae or spines. Body black, with pale yellow or cream-colored markings on lower three quarters of inner and outer eye margins, antennal lobes, transverse band on lower face from mandibular bases across clype- us, mandibular base, transverse band along anterior pronotal carina broken me- dially, arc on posterior lobe of pronotum near tegula, basally on tegula, scutum at posterolaterally, metanotum medially; wing membrane brown becoming darker in medial cell, with brown veins; vestiture silvery on head and thorax, black on me- tasoma. Female. — Unknown. Volume 9, Number 2, 2000 403 Type material. — Holotype 6: NEW CAL- EDONIA, Prov. Sud, Mt Koghis, 17 km nne Noumea, malaise across path in rain- forest, 425m, 8-10 Jan. 1996, M. E. Irwin, D. W. Webb, E. I. Schlinger, 22°10'34"S 166°30'17"E (PARIS). Paratypes: 6 6 6, 2 6 6 , Riviere Bleue Prov. Pk., trail to Upper Riviere Bleue, 19-28 Nov. 1992, 305 m, D. W. Webb; Riviere Bleue Prov. Pk., km 21.9 Riviere Bleue road, Nov. 1992, 290 m, M. E. Irwin, D. W. Webb; 2 6 6, same as pre- vious, except, 5-16 Nov. 1992, 290 m, D. W. Webb; 1 6, Riviere Bleue Prov. Pk., trail to Vallee de Pourina, 19-28 Nov. 1992, 850 m, D. W. Webb; 1 6, Mt Koghis, 17 km nne Noumea 5-15 Nov. 1992, D. W. Webb (DAVIS, ILLINOIS). This species was collected in November and early Jan- uary. Etymology. — This species is named after the collection site on Mt. Koghis. Discussion. — Eirone khogisica most close- ly resembles laniensis, based on the apical- ly convex clypeus, shallowly convex me- tasomal sternum I, presence of a midcoxal spine and well-developed mesopleural la- mella. Both species are basically black with yellow to whitish markings. How- ever, koghisica can be distinguished from laniensis by the trilobate clypeus, and api- cally rounded or truncate epipygium. Eirone laniensis Baptiste, new species (Figs. 3, 20, 26) Male. — Body length 10-16 mm; fore- wing length 9-13 mm; face (Fig. 3); punc- tation as in koghisica, except clypeal punc- tures 1-2 PD apart; clypeal apex gently convex medially and 2.5 MOD wide; F-I length 2.3X breadth; F-II length 3.6X breadth; midcoxal spine present; metaso- mal sternum I with medial longitudinal carina extending about halfway from an- terior edge; epipygium with narrowly truncate bilobate apex, medially de- pressed subapically, with long setae aris- ing from margins, strongly flattened in profile; hypopygium narrowly rounded apically and flattened in profile, with short, stout fringe of setae along apex (Fig. 20); genital capsule (Fig. 26): gonocoxa with subapical ridge; aedeagus with apical column slender and elongate, extending two-thirds length of gonocoxa, bending dorsally; penis valve appearing strongly trilobate, with long dorsal hook; volsella strongly bilobate dorsally, ventrally with densely setose surface. Head black, with yellow or cream-colored markings on an- tennal lobes, lower half of clypeus, inner and outer eye margins, spot behind eye, base of mandible; pronotal band on ante- rior carina, on pronotal lobe near tegula; mesopleuron black, with yellow spot be- low tegula; scutum black with yellow spot basally on tegula; scutellum and metano- tum with yellow markings medially and on anterolateral edge; propodeum with posterior yellow submedial spots; legs or- ange, except coxae black with yellow-or- ange ventral markings ventrally; metaso- ma orange, with black markings basally and apically on T-I, apically on metasomal sternum II and T-V, T-VI, metasomal ster- num I, S-V and S-VI entirely black; wing membrane light brown with dark brown veins; vestiture silvery on head and tho- rax, orange on metasomal except black in specimens from Mt. Koghis and Mt. Dzu- mac. Female. — Unknown. Type material. — Holotype 6: NEW CAL- EDONIA, Upper La Ni Valley, 2-17 Nov. 1992, D. W. Webb, 103 m, malaise trap across logging road (PARIS). Paratypes 42 6 6, 34 6 6 , same data as holotype; 2 6 6, Riviere Bleue Prov. Pk., trail to Vallee de Pourina, 19-28 Nov. 1992, 850 m, D. W. Webb, malaise trap across forest path; 1 6 , Riviere Bleue Prov. Pk., km 25.8, Riviere Bleue road, 30 Oct - 3 Nov. 1992, 213 m, M. E. Irwin, D. W. Webb; 2 6 6, Mt. Kogh- is, 500 m, 17 km nne Noumea, 5-15 Nov. 1992, D. W. Webb; 1 6, Mt. Koghis, 500 m, 17 km nne Noumea, 1-3 Nov. 1992, M. E. Irwin, D. W. Webb, malaise trap in tropical forest; 1 6, Mt. Koghis, 800 m, 1- 6 Sept. 1972, J. F. McAlpine; 1 6, Mt. Dzu- 404 Journal of Hymenoftera Research mac, 900 m, 166°28'E 22°1'45"S, 1-10 Nov. face; aedeagus apical column elongate, 1986, R. L. Brown, Malaise trap (DAVIS, sinuous, and slender, bending ventrally; ILLINOIS, OTTAWA). This species was penis valve with large, slender dorsal collected in November. hook and ventral lobe strongly expanded Etymology. — Eirone laniensis is named af- apically; volsella dorsally large and fist- ter the collection site in the La Ni Valley. like with small subsidiary lobe with 3-5 Discussion. — Eirone laniensis most closely long apical setae, ventrally with curved resembles koghisica as discussed under digitate lobe margined by erect, long, that species. Additional diagnostic fea- somewhat curly setae, apical margin with tures include the presence of a midcoxal short erect setae. Head yellow, with black spine, apically convex clypeal margin and markings, except orange on pedicel, ven- well-developed mesopleural lamella, tral surface of scape and apical rim of Specimens from Mt. Koghis and Mt. Dzu- clypeus, on frons back to occiput, diagonal mac tend to be darker, with more black bands extending from upper posterior eye coloration than the rest. However, struc- margins to occiput, flagellomeres and dor- turally they appear to be the same as the sal surface of scape, mandibular apex; rest of the paratypes. pronotum yellow, with black markings anteriorly, submedially behind anterior Eirone maigretae Baptiste, new species carina to lateml area posterior margin (rigs. 4, 1U, 11, I/) transparent and orange-tinted; mesopleu- Male. — Body length 18-21 mm; fore- ron orange, with black markings along wing length 14-16 mm; face (Fig. 4) punc- upper anterior edge to wing fossa; scutum tures tiny between antennal lobes and in- black, except anteromedially orange, with ner eye margins, contiguous to 0.5 PD posterior submedial yellow longitudinal apart, becoming slightly larger and less bands; scutellum black, with yellow sub- dense, 0.5-1.0 PD apart on frons and ver- medially; metanotum black with yellow tex; clypeus medially gently convex, api- band medially; propodeum orange with cal truncation 2.2-2.3 MOD wide, punc- black submedially, and medial longitudi- tures small and 0.5-1.0 PD apart medially; nal yellow band; legs orange, with black F-I 2.2 X; F-II length 3.2 X breadth; prono- dorsally on the coxae, trochanters and tal punctures tiny and 2-4 PD apart; scutal femora; metasoma orange, with black on punctures tiny and nearly contiguous lat- posterolateral area of T-I and on hypopy- erally, becoming 0.5-2 PD apart laterally; gium; wing membrane brown, darkest mesopleural punctures 0.5-2.0 PD apart; apically, with dark brown-black veins; mesopleural lamella apically rounded and vestiture silvery to pale yellow on head flat against body; scutellum punctate me- and thorax, orange on metasoma. dially and submedially, punctures 1-2 PD Female. — Unknown, apart; propodeum finely and densely sha- Type material. — Holotype 6: NEW CAL- greened, punctures 1-2 PD apart, impunc- EDONIA, trail to Mt. Panie, 22 km NW tate laterally; abdominal segments finely Hienghene, 11-25 Nov. 1992, D. W. Webb, shagreened, punctures tiny and 4-6 PD E. & M. Schlinger, malaise trap, in tropical apart; metasomal sternum I with medial forest, 457 m (PARIS). Paratypes 28 6 6: longitudinal keel extending about halfway 19 6 6, same data as holotype; 6 6, same from anterior edge, sloping obliquely to as holotype, except, 11 Nov. 1992, E. & M. posterior margin; epipygium strongly con- Schlinger, 238-548 m; 1 6, Mt. Koghis, 500 vex in profile, apex rounded (Fig. 21); hy- m, 17 km nne Noumea, 3-4 Nov. 1992, M. popygium with broadly rounded apex; E. Irwin, D. W. Webb, E. & M. Schlinger; genital capsule (Fig. 27): gonocoxa with at 1 6 , Mt. Koghis, 500 m, 17 km nne Nou- most an obsolescent ridge on inner sur- mea, 24-26 Dec. 1991, M. E. Irwin, D. W. Volume 9, Number 2, 2000 405 Webb; 1 6, Mt. Panie, 250-350 m, 30 Oct. 1986, R. L. Brown, sweeping; (DAVIS, IL- LINOIS, OTTAWA). This species was col- lected from late October into late Decem- ber. Etymology. — This species of Eirone is named after Carolyn S. Maigret. Discussion. — Eirone maigretae can be dis- tinguished from the other species lacking a midcoxal spine as discussed under emar- ginata and webbi. Eirone nasalis Kimsey, new species (Figs. 7, 12, 28) Male. — Body length 15-19 mm; fore- wing length 10-14 mm; face (Fig. 7) lower facial punctures tiny and nearly contigu- ous, except upper clypeal margin nearly impunctate; clypeus projecting subapical- ly, with well-developed, horizontally ori- ented ventral bevel, apex medially form- ing anteriorly projecting knob, with long pale setae, strongly nasiform in profile; frons punctures larger and 0.5-1.0 PD apart, except nearly impunctate medially below midocellus; vertex punctures 0.5- 1.0 PD apart; F-I length 2.7 X breadth; F-II length 2.8 X breadth; pronotal and scutal punctures 0.5-1.0 PD apart, becoming sparsest medially; mesopleural punctures 0.5 PD apart; scutellar punctures 1 PD apart laterally, becoming sparser medial- ly; midcoxal spine present; propodeum primarily punctate, punctures 0.5-1.0 PD apart, except anterolaterally impunctate; abdominal segments finely shagreened, punctures irregularly spaced, 2-6 PD apart; metasomal sternum I with medial keel along entire length and abruptly de- clivous posteriorly (Fig. 12); epipygium with medial depression, apex bilobate, flattened in profile; hypopygium subrect- angular, slightly narrowed and subtrun- cate apically; genital capsule (Fig. 28): gonocoxa with subapical ridge on inner surface and large subbasal lobe on dorsal margin connected to setose medial lobe on inner surface; aedeagus apical column elongate, extending more than two-thirds length of gonocoxa, sharply angled ven- trally, apex broadly elongate triangular; penis valve with slender and elongate dor- sal hook and ventral lobe; volsella broadly truncate dorsally, ventral surface with row of erect curved setae. Black, with cream- colored or light yellow band along inner and outer eye margins; wing membrane light brown with brown veins; vestiture silvery on head, thorax and metasomal basal segments, black on apical segments. Female. — Unknown. Type material. — Holotype 6: NEW CAL- EDONIA, trail to Mt. Panie, 22 km NW Hienghene, 11-25 Nov. 1992, D. W. Webb, E. & M. Schlinger, malaise trap in tropical forest, 600 m (PARIS). Paratypes 39 6 6, 25 8 8, same data as holotype, except 457 m; 6 6 6, same data as holotype, except 457 m and 11 Nov. 1992; 4 6 6, Mt. Mand- janie, 5.3 km wsw Pouebo, 9-26 Nov. 1992, D. W. Webb, 550 m, malaise trap in tropical forest (DAVIS, ILLINOIS). This species was collected in November. Etymology. — The species name refers to the nose-like projection of the clypeus. Discussion. — This species can be imme- diately recognized by the projecting nose- like clypeus. Additional diagnostic fea- tures include the large rounded meso- pleural lamella, presence of a midcoxal spine and black body color. Eirone neocaledonica Williams (Figs. 17, 29) Eirone neocaledonica Williams 1945:415. Holo- type 6: New Caledonia: Thi River Valley near St. Louis (WASHINGTON), examined. Male. — Body length 9-11 mm; forewing length 7-9 mm; face with tiny contiguous punctures between inner eye margins and antennal sockets; frons nearlv impunctate, with punctures widely separated, partic- ularly along medial sulcus; clypeus with punctures much larger than along inner eye margin, punctures 0.5-1.0 PD apart, but obscured by fine shagreening; clypeal apex narrowly truncate, truncation 2.2 406 Journal of Hymenoptera Research MOD wide; F-I length 2.2 X breadth; F-II length 3X breadth; pronotum almost im- punctate with few scattered lateral punc- tures; mesopleural punctures 0.5-1.5 PD apart, with dense, fine shagreening be- tween; mesopleural lamella rounded pos- teriorly; scutal punctures medially 3-5 PD apart, becoming finer and denser laterally, 1-2 PD apart; scutellum nearly impunctate medially, laterally punctures 1-2 PD apart, with dense, fine shagreening; pro- podeum laterally with large tubercle, nearly impunctate except small medial patch of punctures (2-5 PD apart) and lat- erally below tubercle punctures smaller and 0.5-1.0 PD apart; midcoxal spine pre- sent; metasomal sternum I strongly trian- gular in profile, sloping abruptly to pos- terior margin; epipygium convex with two large digitate apicolateral lobes; hypopy- gium flat and broadly angulate apically; genital capsule (Fig. 29): gonocoxa with medial swelling associated with tuft of long setae on inner surface; aedeagus api- cal column elongate, slender and sinuous, bending ventrally; penis valve with broad dorsal lobe and broad, wing-like apical lobe; volsella with broad dorsal lobe, se- tose medial lobe and asetose ventral sur- face; color: black; wing membrane untint- ed basally, brown-tinted apically, partic- ularly in marginal cell of forewing; pubes- cence on head and thorax silvery, on metasoma black. Material examined. — 3 males (including the holotype): 9.1 km NW Sarramea, Thi River Valley, and Mt. Koghis (17 km nne Noumea) (HONOLULU, URBANA, WASHINGTON). This species was col- lected in the months of December and Jan- uary. Discussion. — Eirone neocaledonica is one of the most unusually modified species and does not resemble any other in the genus. It can be immediately recognized in the male by the bidigitate epipygium, tuberculate propodeum and mesopleuron and truncate clypeal apex. Eirone nepouiensis Baptiste, new species (Figs. 10, 13, 30) Male. — Body length 10-14 mm; fore- wing length 9-11 mm; clypeal punctures discrete and nearly touching; clypeal apex bilobate in ventral view, apical bevel near- ly horizontal in profile, apical truncation 2.8 MOD wide; punctures along lower in- ner eye margin tiny and contiguous; frons punctures much larger and 0.5-2 PD apart; vertex punctures 0.5-1.0 PD apart; F-I length 2.0-2.1 x breadth; F-II length 2.4-2.5 X breadth; pronotal punctation 1-4 PD apart, except nearly impunctate me- dially; scutal punctures 0.5 PD apart an- teriorly and laterally, becoming 1-3 PD apart medially; scutellar punctures contig- uous laterally, 2-3 PD apart medially; me- sopleural punctation 1-3 PD apart, becom- ing much finer and denser near meta- pleural suture; propodeal punctures 1-2 PD apart, becoming impunctate sublater- ally; metasomal punctures 2-A PD apart; midcoxal spine present (Fig. 10); metaso- mal sternum I with medial carina extend- ing one-third to one-half distance from an- terior edge, obliquely sloping to posterior margin (Fig. 13); epipygium broadly rounded apically, medially depressed; hy- popygium with bluntly rounded apex; genital capsule (Fig. 30): gonocoxa broad, inner surface without distinct ridges or lobes; aedeagus apical column slender, short, extending only slightly beyond pe- nis valve; penis valve with broad, wing- like apical and basal lobes, dorsal lobe broadly rounded and hooked apically; volsella with bilobate and hooked dorsal surface, ventrally asetose and abbreviated; color black, with yellow markings on an- tennal lobes, clypeus except distal edge, inner and outer eye margins, basal two- thirds of mandibles, across anterior pron- otal carina (broken medially) and near dorsolateral apex, pronotal lobe, basal edge of tegula, band submedially on scu- tum, spot on scutellum anterolateral^ and posterolaterally, metanotum submedially Volume 9, Number 2, 2000 407 and at hind wing base attachment, meso- pleuron with spot below tegula, forefemur anteriorly; propodeum posterolaterally; T- I-III with lateral spots; wing membrane brown, with dark brown veins; vestiture brownish on head and thorax, black on metasoma. Female. — unknown. Type material. — Holotype 6 (right an- tennal flagellomeres 2-3 missing, flagel- lomeres IV-XI mounted on point below specimen): Presqu'ile de Pindai, 6 km sw Nepoui, 25 Dec. 1991, M. E. Irwin (PARIS). Paratypes, 2 6 6, Plage de Pindai, 6 km sw Nepoui, 7-13 Nov. 1992, D. W. Webb, malaise trap in coastal dunes area (DAVIS, ILLINOIS). Specimens were collected in November. Etymology. — This species is named after the collection site on Nepoui. Discussion. — The flattened metasomal sternum I, emarginate mesopleural lamel- la, midcoxal spine and black coloration are characteristics shared by nepouiensis and transversa. However, nepouiensis can be distinguished from transversa by the ven- trally bilobate clypeus and nearly horizon- tal clypeal bevel. Eirone nigra Brown (Figs. 11, 31) Eirone nigra Brown 1984:257. Holotype 6: New Caledonia: Hienghene (HONOLULU), ex- amined. Male. — Body length 13 mm; forewing length 11 mm; face with tiny contiguous punctures between inner eye margins and antennal sockets; frons impunctate around medial sulcus, laterally punctures 0.5-1.0 PD apart; clypeus with punctures larger than along inner eye margin, punctures 0.5-1.0 PD apart; clypeus convex medially, apex narrowly truncate, truncation 1.2 MOD; F-I length twice breadth; F-II length 2.8 X breadth; pronotal punctures 1-3 PD apart; mesopleural punctures 1-2 PD apart; mesopleural lamella short, apically emarginate; scutal punctures sparsest me- dially (1-3 PD apart), becoming finer and denser laterally (0.5-1.0 PD apart); scutel- lum impunctate medially, denser laterally, lateral punctures 0.5-1.0 PD apart; pro- podeum punctate in small medial patch (punctures 1-2 PD apart) and laterally (punctures 0.5-1.0 PD apart), impunctate sublaterally; midcoxa without apical spine; metasomal sternum I gently con- vex, with short medial carina extending about half length, sloping gently to pos- terior margin; epipygium strongly convex, apically subtruncate; hypopygium flat, broadly rounded apically; genital capsule (Fig. 31): gonocoxa with swelling above volsella on inner surface; aedeagus with slender apical column bending ventrally; penis valve with long dorsal hook, ventral lobe with long slender digitate apical lobe bending dorsally; volsella with large, broadly rounded dorsal lobe, ventrally with densely setose digitate lobe; color: black, wing membrane dark brown-tinted, veins black; pubescence on head and tho- rax silvery, metasoma black. Material examined. — Only the holotype has been seen. It was collected in Hiengh- ene in January. Discussion. — The lack of a midcoxal spine and short mesopleural lamella sug- gests a close relationship between nigra, and tfiaigretae and emarginata. However, the shorter F-I, clypeal truncation narrow- er than the transantennal distance and the black abdominal setae distinguish nigra from these two species. Eirone obtusidens Turner (Figs. 5, 32) Eirone obtusidens Turner 1919:236. Holotype 6: New Caledonia: Noumea (LONDON), ex- amined. Eirone obtusidens var. supcrstcs Cockerell 1929: 239. Holotype 8: New Caledonia: Bourail (WASHINGTON). Synonymized by Brown 1984. Male. — Body length 13-14 mm; fore- wing length 11-12 mm; face with tiny con- tiguous punctures between inner eye mar- 408 Journal of Hymenoptera Research gins and antennal sockets; frons with Eirone paniensis Kimsey, new species punctures 0.5-2 apart, except impunctate (Fig. 6, 33) along medial sulcus and adjacent to anten- . , , „ , - , _ nal socket; clypeus with punctures larger , Mf 7?°^ length 7-12 mm; forewing than along inner eye margin, punctures lf §th ^J° mm; faCe Potation; contiguous; clypeal apex broadly triangu- ^PeUS ^greened, most punctures 1-2 lar, with single apicomedial lobe; F-I PD apart, becoming sparser apically, inner length 2.4-2.5 X breadth; F-II length 3.0- e^e mar§insr with tir7' nearly contiguous 3.2X breadth; pronotal punctures 1-3 PD Punctures; fro™ with punctures 2-5 PD apart; mesopleural punctures 1-2 PD aPart' imPunctate medially; clypeus gent- apart separated by dense, fine shagreen- ^ convex' aPex broadly truncate, trunca- ing; mesopleural lamella posteriorly tlon 2 MOD Wlde; vertex Pictures 2-5 rounded; scutal punctures densest along PD aPart; W len§th 2X breadth; F-II notauli (nearly contiguous), medially length 3 X breadth; malar lobe subtending sparsest (0.5-1 PD apart); scutellum nearly mandibular articulation tooth-like; prono- impunctate medially, punctures denser tum impunctate and polished dorsally, laterally, 0.5-1.0 PD apart; propodeum Slde Wlth punctures 1-2 PD apart; meso- punctate medially and propodeum punc- Pleuron bulging and knob-like medially, tate medially and laterally, 0.5-1.0 PD punctures 0.5-2 PD apart, densest dorsal- apart; propodeum punctate medially and lY' mesopleural lamella broadly rounded laterally, punctures 0.5-1.0 PD apart, im- aPicallY and bending slightly ventrally; punctate sublaterally; metasomal sternum scutum strongly depressed posteriorly I sharply keeled, keel somewhat hooked alon§ notalices, finely shagreened, scutal posteriorly, abruptly declivous posterior- punctures 0.5-5 PD apart, becoming near- ly; epipygium and hypopygium flattened ty impunctate posteromedially; scutellar and apex narrowly subtruncate; genital punctures 1-5 PD apart, nearly impunc- capsule (Fig. 32): gonocoxa with low me- tate medially; propodeum finely shagree- dial angular lobe on inner surface; aedea- ned and impunctate dorsally except for gus elongate, apical column broad and punctate medial band, strongly rubercu- flattened, bending dorsally, apically trun- late sublaterally, becoming densely punc- cate, column extending more than half its tate laterally, punctures 0.5-2 PD apart; length beyond penis valve; penis valve midcoxa with acute spine; metasomal ster- with short dorsal hook, large, rounded num I broadly triangular in profile, slop- ventral lobe and elongate basal lobe; vol- ing obliquely toward posterior margin, sella with bilobate dorsal projection, apical not medially carinate; epipygium broadly lobe capitate and ventral surface slender rounded apically, apical margin with with row of long setae. slight medial indentation, convex in pro- Material examined. — 7 6 S (including the file; hypopygium broadly rounded and holotype of obtusidens), from Prov. Sud, spinose, with flattened apex; genital cap- 7.5 km NW and 1 km NW Sarramea, Ri- sule (Fig. 33): gonocoxal inner surface viere Bleue, Mt. Panie, Noumea and Bour- smooth; aedeagus apical column slender, ail. Specimens were collected in October, elongate and sinuous, extending more November and January. than two-thirds length of gonocoxa; penis Discussion. — The shape of the clypeus in valve with rounded dorsal and ventral obtusidens and anone is unique among the lobes; volsella with large truncate and se- New Caledonian species. Eirone obtusidens tose dorsal lobe, ventral surface with erect, can be distinguished from anone by the dense row of setae. Body black, with pale gently convex metasomal sternum I as dis- yellow or cream-colored markings on cussed under that species. mandibular base, lower half of clypeus, Volume 9, Number 2, 2000 409 along inner and outer eye margins, on an- mesopleural lamella rounded apically and tennal lobes, transverse band (broken me- bending ventrally; scutum punctate and dially and sublaterally) along anterior shagreened laterally, punctures 0.5-1 PD pronotal carina, spot on pronotal lobe, dis- apart, becoming sparsely punctate medi- tal forecoxal spot and subalar spot on me- ally with punctures 1-3 PD apart; scutel- sopleuron, spot on tegula and on scutel- lum laterally punctate, punctures 0.5-1 PD lum laterally; wing membrane brown, be- apart, medially polished and impunctate; coming darkest on apical third, with black propodeum finely shagreened, with punc- veins; vestiture silvery on head and tho- tures 1-2 PD apart, except laterally 0.5-1 rax, black on metasoma. PD apart, sublaterally impunctate; mid- Female. — Unknown. coxal spine present; metasomal sternum I Type material. — Holotype 6: trail to Mt. broadly triangular in lateral view, sloping Panie, 22 km NW Hienghene, 11-25 Nov. obliquely posteriorly (Fig. 15); epipygium 1992, 600 m, tropical forest, malaise trap, narrowly rounded apically, flattened in D. W. Webb and E. & M. Schlinger (PAR- profile; hypopygium with narrowly IS). Paratypes, 5 8 8, same data as holo- rounded apex, apical rim medially thick- type (DAVIS, ILLINOIS). All specimens ened and spinose; genital capsule (Fig. 34): were collected in November. gonocoxa with angulate ridge subapically Etymology. — Eirone paniensis is named on inner surface; aedeagus with apical col- after the collection site, Mt. Panie. umn Flattened and becoming broadly sub- Discussion. — Much like neocaledonica, triangular apically; penis valve small with paniensis has a prominent knob on either sharply hooked dorsal and ventral lobes, side of the propodeum and very similar occupying half or less the distance across aedeagus and penis valve. Both species genital capsule; volsella dorsally fist-like, are relatively small-bodied, 7-10 mm long, ventral half forming a long flat surface and black, with few pale markings. How- with row of dense, stout setae. Body black, ever, paniensis can be distinguished from with light yellow or cream-colored mark- neocaledonica by the strongly tuberculate ings on clypeus above apical margin, mesopleuron, gonocoxa with large interior along lower two-thirds of inner and outer brush of long setae, and tooth-like genal eye margins, antennal lobes, band along projection. In addition, paniensis has a pronotal transverse anterior carina (bro- midcoxal spine and well-developed me- ken medially), band anterolaterally and sopleural lamella posterior lobe of pronotum; wing mem- brane brown, with brown veins; vestiture Eirone rivierensis Baptiste, new species silvery on head and thorax metasomal (Figs. 15, 34) terga j_VII with long/ erect black setae Male. — Body length 8-14 mm; forewing along posterior margin that are densest on length 9-12 mm; facial punctures 1-3 PD VII; sternum II-VI with long, erect black apart, except area between inner eye mar- setae along posterior margin, gin and antennal bases with tiny, nearly Female. — Unknown, contiguous punctures, and narrow im- Type material. — Holotype 8 (left anten- punctate longitudinal band below mido- nal flagellomeres II-XIII missing): Riviere cellus; clypeus convex medially and api- Bleue Prov. Pk., 19-20 Nov. 1992, 213 m, cally truncate, apical truncation 2 MOD D. W. Webb & E. and M. Schlinger (PAR- wide; F-I length 2x breadth; F-II length IS). Paratypes 9 8 6, 1 6, Riviere Bleue 3X breadth; pronotal punctures 1-3 PD Prov. Pk., km 25.8 Riviere Bleue road, 5- apart; mesopleuron projecting medially, 16 Nov. 1992, 213 m, D. W. Webb, E. & M. integument shagreened, with punctures Schlinger; 1 8 6, ibid., except, 30 Oct.-3 1-2 PD apart, becoming denser dorsally; Nov., M. E. Irwin, D. W. Webb; 2 8 6, 410 Journal of Hymenoptera Research ibid., except, km 19.6, 18-20 Nov. 1992, D. apart, laterally 1-2 PD apart; propodeum W. Webb; 2 6 6, ibid., except, km 21.9, 20- with oblique sublateral swelling, punc- 28 Nov. 1992, M. E. Irwin, D. W. Webb; 1 tures medially 1-3 PD apart, laterally con- 6, 30 km NW Yate, 550 m, 27-28 Decern- tiguous to 0.5 PD apart, impunctate sub- ber 1991, M. E. Irwin, D. W. Webb; 1 6 , laterally; midcoxa with short apical tooth; Riviere Bleue Prov. Pk., trail to upper Ri- meso-, metathorax and propodeum finely viere Bleue, 5-16 Nov. 1992, 290 m, D. W. and densely shagreened between punc- Webb; 1 6\ Riviere Bleue Prov. Pk., trail tures; metasomal sternum I with sharp, to Vallee de Pourina, 19-28 Nov. 1992, 850 posteriorly hooked, medial keel, over- m, D. W. Webb; (DAVIS, ILLINOIS). This hanging metasomal sternum II; epipy- species was collected in November and gium strongly convex, apically narrowed December. and truncate; hypopygium flat, apically Etymology. — This species is named after narrowed and bilobate; genital capsule the collection site in Riviere Bleue Prov- (Fig. 35): gonocoxa inner surface smooth, ince. without distinct ridges or lobes; aedeagus Discussion. — This species can be recog- with apical column elongate, slender and nized by the primarily black coloration, sinuous, bending ventrally; penis valve without any reddish coloration, presence large, dorsal lobe with blunt hook, ventral of a midcoxal spine, rounded mesopleural lobe large and wing-like, extending api- lamellae and truncate clypeal margin. E. cally; volsella with long curly setae along rivierensis is most similar to colorata but apical surface, dorsally large and capitate, can be separated by the dense black tuft ventrally with large flat lobe tipped by of setae protruding from T-VI and VII. two long setae; color black, with whitish This tuft of setae is absent in colorata. markings on antennal lobes, inner and outer eye margins, mandible base, clype- Eirone salteri Brown us, transverse anterior pronotal margin, (Figs. 16, 18, 35) pronotum adjacent to tegula, mesopleuron Urone salteri Brown 1984:250. Holotype 6 : New bel°W te§ula' scutellum ™ th two lateral Caledonia: St. Louis "Val" (HONOLULU), sPots; ^etanorum medially; legs orange; examined. wing veins orange, except stigma black; wing membrane dark yellow-tinted, apex Male. — Body length 12-14 mm; fore- often browner; vestiture on head and mo- wing length 10-12 mm; face with tiny, rax silvery to yellowish, on metanotum contiguous punctures between inner eye brown to black. margins and antennal sockets; frons with Material examined. — 4 6 6 (including ho- punctures 0.5-1.0 apart, except nearly im- lotype): Sarramea, St. Louis Val, and 1 km punctate medially; clypeus subapically n and 9.1 km NW Sarramea. All speci- concave, apex narrowly produced into mens were collected in late December, two lobes, 1.2-1.4 MOD apart, surface January and February (DAVIS, HONO- finely shagreened, punctures much larger LULU, URBANA). than along inner eye margin, 0.5-1.0 PD Discussion. — Diagnostic features of sal- apart; F-I length 2.2-2.3 X breadth; F-II teri include the presence of a midcoxal length 3.2-3.4X breadth; pronotal punc- spine, mesopleural lamella short and tures 0.5-1.0 PD apart; mesopleural punc- sharply emarginate, keel-like metasomal tures 0.5-1.0 PD apart; mesopleural la- sternum I overhanging metasomal ster- mella short, apically truncate; scutal punc- num IT, apically bilobate clypeus, bilobate tation medially 1-2 PD apart, becoming epipygium and amber-colored wing mem- finer and denser laterally, 0.5-1.0 PD brane. Metasomal sternum I is the same apart; scutellar medial punctures 2-A PD odd shape as seen in irwini, but the other Volume 9, Number 2, 2000 411 features given above will immediately separate the two. Eirone subtuberculata Brown (Fig. 36) Eirone subtuberculata Brown 1984:253. Holotype 6: New Caledonia: Mts. de Koghis (HON- OLULU), examined. Mfl/e.--Body length 15-17 mm; fore- wing length 11-14 mm; face with tiny con- tiguous punctures between inner eye mar- gins and antennal sockets; frons with punctures contiguous to 0.5 PD apart, ex- cept impunctate along medial sulcus; clypeus produced into subapical knob, with punctures larger than along inner eye margin, punctures contiguous and shagreened; clypeal apex truncate, trun- cation 3.0-3.2 MOD across; F-I length 2.2- 2.3 X breadth; F-II length 3x breadth; pronotal punctures 1-3 PD apart; meso- pleural punctures 0.5-1.0 PD apart, sha- greened between punctures; mesopleural lamella posteriorly triangular to rounded; scutal punctures medially 0.5-1.0 PD apart, becoming finer and denser laterally, contiguous to 0.5 PD apart; scutellar punc- tures 2-3 PD apart, laterally 0.5-1.0 PD apart; propodeum punctures 1-2 PD apart except anterolaterally, impunctate and finely shagreened; midcoxa with short apical tooth; metasomal sternum I strong- ly produced into slightly hooked keel, sloping vertically to sternum II; epipy- gium flattened, apically broadly bilobate; hypopygium flat, broadly truncate apical- ly, with slight medial emargination; geni- tal capsule (Fig. 36): gonocoxa with large submedial swelling on inner surface and subbasal lobe on ventral margin; aedeagus apical column elongate, about two-thirds as long as gonocoxa, apically curved, ex- panded and lanceolate; penis valve with slender dorsal and ventral, apically hooked lobes, barely obscuring aedeagal base; volsella with broad dorsal lobe and slender obsolescent ventral surface, with short erect setae along apical margin; color black, with occasional small faint pale mark on inner and outer eye margin; wing membrane brown-tinted, veins black; ves- titure of head and thorax silvery, metaso- mal brown to black. Material studied.— 13 6 6: ML Koghis; 17 km nne Noumea (DAVIS, URBANA, HONOLULU). All specimens were col- lected in the months of December, January and February. Discussion. — Eirone subtuberculata ap- pears to be most similar to rivierensis and colorata based on the presence of a mid- coxal spine, emarginate mesopleural la- mella, apically truncate epipygium and thin clypeal apical margin. However, it is much larger than these two species, rang- ing from 12-16 mm long. The metasomal sternum I of subtuberculata differs as well resembling that of nasalis with an abruptly declivous posterior margin. Eirone transversa Brown (Fig. 37) Eirone transversa Brown 1984:254. Holotype 8: New Caledonia: Mts. de Koghis (HONO- LULU), examined. Male. — Body length 10-15 mm; fore- wing length 8-12 mm; face with tiny con- tiguous punctures between inner eye mar- gins and antennal sockets; frons with punctures 1-2 PD apart, except impunc- tate along medial sulcus; clypeus slightlv convex, with transverse subapical bevel, punctures larger than along inner eye margin, 0.5-1.0 PD apart, shagreened be- tween; clypeal apex truncate, truncation 2.0-2.2 MOD wide; F-I length 2.1-2.2X breadth; F-II length 2.5-2.6 X breadth; pronotal punctures tiny, 1^4 PD apart; mesopleural punctures 0.5-1.0 PD apart, finely shagreened between; mesopleural lamella short, apically emarginate; scutal punctures medially 1-2 PD apart, becom- ing finer and denser laterallv along notau- li, contiguous to 1 PD apart; scutellum nearly impunctate medially, lateral punc- tures 1-2 PD apart; propodeum punctate 412 Journal of Hymenoptera Research medially and laterally, punctures 1-A PD Riviere Bleue Road, km 21 and 26; Riviere apart, impunctate anterolaterally and fine- Bleue Prov. Pk, 30 and 36 km NW Yate; ly shagreened; midcoxa with apical spine; Riviere Bleue Prov. Pk., trail to Vallee de metasomal sternum I broadly convex, Pourina; Riviere Bleue Prov Pk., trail to with short basal carina, sloping gradually Upper Riviere Bleue (DAVIS, ILLINOIS), to posterior margin; epipygium strongly Specimens were collected in the months of convex, apex slightly bilobate or emargin- October through January, ate; hypopygium flat, with broadly round- Discussion. — Most similar to nepouiensis, ed apex; genital capsule (Fig. 37): gono- as discussed under that species, transversa coxa narrowed medially, inner surface rel- can be distinguished from nepouiensis by atively smooth without discrete ridges or the clypeal apex broadly rounded in ven- lobes; aedeagus with slender and elongate tral view and the apical clypeal bevel apical column, reaching to apical third of oblique when viewed in profile, gonocoxal, apical column medially arcu- ate, extending apically, not bending dor- Eirone webbi Kimsey, new species sally or ventrally; penis valve dorsal and U^gs- °> ^o) ventral lobes broadly rounded without Male. — Body length 8-13 mm; forewing apical hook; volsella dorsally bilobate length 7-11 mm; facial punctures tiny, with long erect setae along apical margin, contiguous to 1 PD apart across lower ventral surface with row of elongate erect face, frons and vertex essentially impunc- setae; color black, with yellowish mark- tate and shiny; clypeus gently convex, ings on antennal lobes, apical clypeal mar- apex narrowly truncate, truncation 2.1-2.3 gin, mandibular base, spot on inner and MOD wide; F-I length 2X breadth; F-II outer eye margin, pronotum with medi- length 3x breadth; pronotum highly pol- ally broken transverse anterior band, ished and nearly impunctate, except lat- pronotal angle adjacent to tegula, tegula, erally with scattered tiny punctures; me- small anterolateral scutellar spot; wing sopleural punctation consisting of small membrane brown-tinted, veins black; pu- punctures about 1 PD apart, becoming bescence of head and thorax silvery, me- nearly impunctate ventrally; mesopleural tasoma black. lamellae acute, strongly narrowed apically Female. — Body length 7-8 mm; body and bending somewhat ventrally (Fig. 8); coarsely punctate, with fine, dense longi- scutum highly polished, punctures 2-8 PD tudinal striae; facial punctures large and apart; scutellum with punctures 1-5 PD somewhat striatiform; frons with obscure apart, densest laterally; propodeum essen- medial longitudinal groove; F-I as long as tially impunctate, densely and finely sha- broad; F-II length 1.2X breadth; clypeal greened; metasomal sternum I with short apex truncate; pronotum rounded lateral- medial carina forming obtuse angle in ly; propodeum impunctate laterally, elon- profile; epipygium with narrowly round- gate with flattened dorsal surface and lat- ed apex; hypopygium broadly rounded, eral edges rounded, not carinate; T-I con- with narrow, thickened spine-rimmed api- tinuous with II, not constricted posterior- cal margin; genital capsule (Fig. 38): par- ly; metasomal sternum I with strongly amere with smooth inner surface; aedea- projection medial keel, appearing sharply gus with short apical column; penis valve triangular in profile. with rounded dorsal lobe, ventral lobe Material studied. — 91 6 6, 1 9: Mt. short and apically hooked; volsella with Koghis; 17 km nne Noumea; Riviere Bleue large truncate dorsal lobe, setose medial Prov. Pk., 9-10 km NW Sarramea; 13 km lobe and ventral surface asetose. Body se Kaala-Gomen; trail to Mt. Panie, 22 km black, with pale yellow or cream-colored NW Hienghene; Riviere Bleue Prov. Pk. markings on mandibular bases, clypeus, Volume 9, Number 2, 2000 413 inner and outer eye margins, antennal m, malaise trap in tropical forest; 2 6 6,1 lobe, pronotum with transverse band on 9, Riviere Bleue Prov. Pk., 30 km NW anterior carina and narrow band on pos- Yate, 270 m, 27 Dec. 1991, M. E. Irwin, D. terior lobe, mesopleuron with small sub- W. Webb; 1 6, ibid., except 36 km NW alar spot, tegula with small spot, scutel- Yate, 21 Dec. 1991, M. E. Irwin, D. W. lum with lateral spot, metanotum with Webb; 5 6 6, Riviere Bleue Prov. Pk., trail medial spot, propodeum with postero- to Vallee de Pourina, 19-28 Nov. 1992, 850 medial and small lateral spots (faint in ho- m, D. W. Webb; 19 6 6, 1 9, trail to Mt. lotype); wing membrane brown with Panie, 22 km NW Hienghene, 11-25 Nov. black veins; vestiture sparse, silvery to 1992, D. W. Webb, E. & M. Schlinger; 1 6, yellowish on head and thorax, brown to same as previous location, except 11 Nov. black on metasoma. 1992; 1 6 , Mt. Do, 14 km ne Bouloupari, Female.— Body length 5-6 mm; frons 31 Oct.^4 Nov. 1992, M. E. Irwin, D. W. with medial longitudinal irregular groove Webb, E. & M. Schlinger, 1000 m;l d, Mt. extending two- thirds head length; head, Panie, 250-350 ft., 30 Oct. 1986, R. L. thorax and abdomen covered with fine Brown; 1 6, Riviere Bleue, 166°39.55'E dense longitudinal striae; clypeal margin 22°6'S, 18 Oct. 1986, R. L. Brown (DAVIS, medially concave and broadly truncate; F- ILLINOIS). Specimens were collected in I length lx breadth; F-II length 1.3X the months of October and November, breadth; pronotum subquadrate, slightly Other material examined. — 1 9, Riviere narrowed anteriorly, dorsally impunctate Bleue Prov. Pk., trail to Vallee de Pourina, except for irregular row of punctures ex- 19-28 Nov. 1992, 850 m, D. W. Webb; 1 9, tending medially and continuing across same as above, except trail to Upper Riv. scutellum and propodeum; propodeum Bleue, 16-19 Nov. 1992, 290 m. long and flattened, lateral edge rounded, Etymologx/. — This species is named after not carinate, impunctate medially, with the collector, Donald W. Webb, punctures clustered along margins; T-I no- Discussion. — Most individuals of this dose, strongly constricted posteriorly; me- species are entirely black. However, some tasomal sternum I shallowly convex, with- are orange or have varying amounts of out distinct medial keel or carina; tergal black and orange coloration. Eirone webbi punctures sparse and somewhat striati- is one of the New Caledonian species lack- form; T-VI with elevated longitudinal me- ing a midcoxal spine, the others being im- dial impunctate welt; body dark brown punctata, emarginata, nigra and maigretae. with pale setae. Unlike these species, webbi can be distin- Type material. — Holotype 6: NEW CAL- guished by the strongly apically narrowed EDONIA, Riviere Bleue Prov. Pk., trail to and flattened epipygium and hypopy- Upper Riviere Bleue, 5-16 Nov 1992, 290 gium. The epipygium is slightly indented m, D. W. Webb, malaise trap across forest apicomedially and not broadly rounded as path (PARIS). Paratypes, 57 6 6, 11 6 6, in the other species. In webbi the clypeal same data as holotype; 2 6 6, ibid., except apex is broadly truncate, and the trunca- 3-5 Nov. 1992; 5 6 6, ibid., except 19-20 tion has blunt lateral corners. In maigretae Dec. 1992; 1 6, Riviere Bleue Prov. Pk., km and even more so in emarginata, the cor- 19.6 Riviere Bleue rd., 20-28 Nov. 1992, ners are acute, and in emarginata the clvp- 183 m, D. W. Webb; 1 6 , Riviere Bleue eus is apicomedially emarginate. The epi- Prov. Pk., km 25.8 Riviere Bleue rd., 17 pygium in maigretae, nigra and emarginata Nov. 1992, 213 m, D. W. Webb, E & M. is strongly convex in profile, not flattened Schlinger; 1 6 , ibid., except 19-20 Nov. as it is in webbi. Eirone maigretae and emar- 1992; 6 6 6, Mt. Mandjanie, 5.3 km wsw ginata are considerably larger than webbi, Pouebo, 9-26 Nov 1992, D. W. Webb, 550 ranging between 18 and 22 mm in length. 414 Journal of Hymenoptera Research Although color and body size vary consid- erably in webbi, this species can be distin- guished by the acute and ventrally point- ed mesopleural lamella. In the few indi- viduals where the lamellae have been bro- ken off the narrowed and flattened epipygium and hypopygium and config- uration of the clypeus will prove diagnos- tic. £. impunctata is another species of Eirone from New Caledonia that lacks a midcoxal spine. Although we have not been able to see the type of impunctata, webbi appears to be a different species based on Brown's description and illustrations of impunctata. A few specimens of webbi are small, rela- tively impunctate, and are the ferrugine- ous color mentioned by Brown in his de- scription of impunctata (1984). However, Brown does not mention the peculiar me- sopleural lamellae seen in webbi and his illustration of the face of impunctata indi- cates that the clypeal truncation is consid- erably narrower than that of webbi. KEY TO MALES OF THE NEW CALEDONIAN EIRONE SPECIES 1. 3. Midcoxa without apical spine or tooth on posterior angle (as in Figs. 8, 11) 2 Midcoxa with apical spine or tooth on posterior angle (as in Figs. 9, 10) 6 Mesopleural lamella elongate, apically acute (Fig. 8) and bending ventrally; epipygium strongly narrowed apically and flattened in profile (as in Fig. 20) webbi Kimsey Mesopleural lamella foreshortened, apically emarginate (as in Fig. 10), not bending ven- trally; epipygium broadly rounded or truncate apically and strongly convex in profile (as in Fig. 21) 3 Small, body length 8 mm or less impunctata Brown Larger, body longer than 8 mm 4 4. Flagellomere I barely twice as long as broad; metasomal setae blackish; clypeus strongly convex medially, apex narrowly truncate, truncation narrower than distance from the outer margin of one antennal socket to that of the other; wing membrane dark amber . . nigra Brown - Flagellomere I more than twice as long as broad; metasomal setae pale, silvery to yellow- ish; clypeus flattened medially, apex broadly truncate, truncation as wide or wider than distance from the outer margin of one antennal socket to that of the other; wing mem- brane yellowish to light brown tinted 5 5. Body black; clypeus with lateral angles of apical truncation acute or dentate; hypopygium apically truncate emarginata Brown - Body predominantly orange, with some black and yellow; clypeus with apical truncation broadly angulate laterally (Fig. 4); hypopygium apically broadly rounded maigretae Baptiste 6. Mesopleural lamella abbreviated and emarginate posteriorly 7 - Mesopleural lamella not abbreviated, rounded or acute posteriorly 9 7. Sternum I medially keeled, posterior margin pointed medially and extending over base of sternum II (Fig. 14); clypeal apex bilobate; hypopygium apically bilobate (Fig. 18) .... salteri Brown - Sternum I flattened posteriorly (as in Fig. 13); clypeus broadly rounded in profile, with broad subrectangular apical bevel 8 8. Clypeal apex in ventral view bilobate, apical bevel nearly horizontal in profile nepouiensis Baptiste Clypeal apex in ventral view broadly rounded, apical bevel oblique in profile transversa Brown 9. Clypeus subapically bulging, with polished medial knob and subtriangular apical bevel, nasiform in profile (Fig. 7) nasalis Kimsey Clypeus subapically flattened or broadly rounded, apex truncate or medially lobate (as in Figs. 2-6) 10 Volume 9, Number 2, 2000 415 10. Clypeal apex medially clearly convex or lobate (Figs. 2, 3, 5) 11 - Clypeal apex truncate, flat or concave (Figs. 4, 6) 15 11. Metasomal sternum I medially keeled, posterior margin pointed medially and extending over base of sternum II (Fig. 12 and as in Fig. 14); clypeus weakly trilobate invini Baptiste - Metasomal sternum 1 posteriorly flattened or arched, if arched then abruptly declivitous before base of sternum II, not overlapping II (as in Figs. 12, 13, 15); clypeus convex or trilobate (Figs. 2, 3), or unilobate (Fig. 5) 12 12. Clypeal apex medially unilobate (Fig. 5) 13 - Clypeal apex medially convex or trilobate 14 13. Metasomal sternum I strongly keeled and abruptly declivitous posteriorly (as in Fig. 12) anone Kimsey - Metasomal sternum I gently convex medially gently sloping posteriorly to sternum II . . . obtusidens Turner 14. Clypeal apex medially obtusely rounded and thin-edged, without bevel (Fig. 3), broadly convex in profile; epipygium apicomedially emarginate (Fig. 20) laniensis Baptiste - Clypeal apex medially broadly trilobate and projecting anteriorly, with well-developed horizontal bevel seen in ventral view (Fig. 2); epipygium apicomedially rounded or trun- cate koghisica Kimsey 15. Gena with tooth-like projection adjacent to mandibular condyle (Fig. 6); mesopleuron medially tuberculate; propodeum sublaterally tuberculate paniensis Kimsey - Gena rounded or with small angle adjacent to mandibular condyle (as in Figs. 2, 3, 5); mesopleuron and propodeum not tuberculate (except propodeum in neocaledonica) .... 16 16. Epipygium terminating in long digitate apicolateral lobes, flattened medially (Fig. 17); propodeum sublaterally tuberculate neocaledonica Williams - Epipygium apically truncate or rounded, without digitate lobes; propodeum evenly con- vex, not tuberculate 17 17. Sternum I strongly arched and abruptly declivitous before base of sternum II (as in Fig. 12); clypeus with polished subapical tubercle and polished and thickened apical margin snbtubercnlata Brown - Sternum I broadly rounded or elevated medially and extending obliquely to posterior margin (as in Fig. 15); clypeus without subapical tubercle, with thin apical margin .... 18 18. Terga VI and VII with dense tufts of long dense setae; body color black rivierensis Baptiste - Terga VI and VII without dense tufts of setae, setae long and scattered; color black to orange colorata Brown ACKNOWLEDGMENTS previously known New Caledonian species. In- ternational Journal of Entomology 26:49-2::>t, This study was made possible by the intensive col- Tumer R E. i907. A revision of the Thynnidae of lecting efforts of Michael Irwin, Evert and Marion Australia. Part I. Proceedings of the Unnaean Schlinger and Donald Webb. Thanks also to the col- ch/ (l, Vi r S(,////( u-;/(.s 32:206-290 lection managers and scientists of the bishop Muse- vVjHiamS/ F. X. 1945. The aculeate wasps of New Cal- um, Canadian National Collection and the Illinois edonia, with natural history notes. Proceedings of Natural History Survey. tlu. Hllu,lUlllll Entomological Society 12:07 437 LITERATURE CITED Brown, G. R. 1984. New species of Eironc from New Caledonia and Australia, with a review of the J. HYM. RES. Vol. 9(2), 2000, pp. 416^26 Review of the Species of Deutereulophus (Hymenoptera: Chalcidoidea: Eulophidae) of North America Michael E. Schauff Systematic Entomology Laboratory, ARS, USD A, National Museum of Natural History, Washington, D.C. 20560-0168, USA Abstract. — Although species of Deutereulophus (Hymenoptera: Chalcidoidea: Eulophidae) are known in America north of Mexico, none have been described. In this paper five new species are recognized: D. arizonensis, D. floridensis, D. occularis, D. pecki, and D. smithi, and a key is presented. Ashmead (1904) described the genus to recognize specimens in these collections Eulophopteryx and included two species, and organize them for study. It is difficult, not realizing that the generic name was if not impossible, for other researchers to preoccupied (Moschler 1878). The name sort material from collections, because of Deutereulophus was supplied by Schulz the lack of identification aids or illustra- (1906) as a replacement for Eulophopteryx. tions to guide them. I have not personally Girault (1913) described the genus Ente- been able to examine or sort through all donomorpha and Girault (1913, 1915, 1922, collections and therefore, I have not been 1938) and Yoshimoto and Ishii (1965) later able to obtain specimens from many parts added several species from Australia and of North America. This is especially true Guam to this genus which was synony- io\the ™est coast of th* U-S" ^Canada. j i.i_ r-> j. ii u t c ii a Although western collections have not mized with Deutereulophus by LaSalle and . , , r . <■ ^ , , Schauff (1992). Although Deutereulophus Y^ded specimens o( Deutereulophus to . - T ', . . /f, i cc L i date, it is likely that at least one or a few occurs in North America (Schauff et al. \ ., „_-_,, , , . , species occur along the west coast. 1997), no species from that region have ^ ^ j reyiew ^ g frQm been described and no keys or other in- North America and provide names and a formation are available. k for t\xem The majority of specimens used for this ' Acronyms for museums are: (CNC) Ca- study are from the eastern U.S. due to two nadian National Collection, Ottawa, On- factors: 1) extensive malaise trapping by tari0/ Canada; (USNM) National Museum Dr. D. R. Smith in Maryland and Virginia 0f Natural History, Smithsonian Institu- and an extensive collecting effort along tion, Washington, D. C., USA; (BMNH) the eastern seaboard in the mid to late The Natural History Museum, London, 1980s by a team of scientists from Canada, UK. Terminology for morphology follows and 2) the author is one of the few scien- Gibson (1997). A diagram showing mea- tists familar with the genus and was able surements of the head is given in Fig. 9. KEY TO SPECIES OF DEUTEREULOPHUS 1) Legs beyond coxae yellow; scutellum smooth (Figs. 5, 8) 2 Legs beyond coxae at least partly black or brown, scutellum with distinct alutaceous or reticulate sculpture (Figs. 1, 3) 4 Volume 9, Number 2, 2000 417 2) First funicular segment equal in length to second (Fig. 14) pecki, n. sp. - First funicular segment distinctly longer than second (about 1.5 x as long) (Fig. 13) 3 3) Thorax and head with distinct metallic green sheen, posterior margin of the gena on line with the posterior margin of eye; vertex rounded and area behind evenly sculptured floridensis, n. sp. - Thorax and head black, without metallic green coloration; posterior margin of eye pro- jecting behind posterior margin of gena (Fig. 7); vertex with transverse carina just behind ocelli, area below carina with distinct sculpture medially, smooth laterally . . occularis, n. sp. 4) Vertex with distinct transverse carina behind ocelli (Fig. 1), nearly smooth behind posterior ocelli smithi, n. sp - Vertex rounded and without distinct median carina, mostly reticulate behind posterior ocelli with v-shaped smooth area medially (Fig. 2) arizonensis, n. sp. Deutereulophus Schulz Eulophopteiyx Ashmead 1904: 341, 342, 374. Types species Eulophopteryx chapadae Ash- mead (original designation). Preoccupied by Eulophopteryx Moschler 1878: 684. Deutereulophus Schulz 1906: 146. Replacement name for Eulophopteryx Ashmead 1904 (not Eulophopteryx Moschler, 1878). Entedouomorpha Girault 1913: 261. Type species Entedonotnorpha tennysoni Girault (original designation). Synonymy by LaSalle and Schauff 1992: 17. Diagnosis. — Head concave behind (Fig. 1), often with a transverse carina on the vertex behind the ocelli and with the posterior eye margin contiguous with the back of the head such that there is no noticeable temple when viewed dorsally; eyes setose; clypeus delimited by sutures above and lateral to the mouth margin; pronotum semiglobose, and rounded anteriorly, without a trans- verse carina; notauli complete; scutellum with curving (sinuate) lateral grooves con- verging posteriorly and meeting medially (Figs. 3, 5); propodeum with a simple me- dian carina diverging posteriorly and bounding a large open area at the nucha (Figs. 4, 6), laterally with at least a partial plical carina and with a transverse carina below the spiracle that defines the dorsal edge of an area which lies nearly perpen- dicular to the spiracular surface and that usually contains a group of setae below and medial to the callus setae; petiole with a for- ward projecting flange on dorsal and lateral surface; hypopygium reaching about half length of metasoma; outer ovipositor plates visible and generally reticulate; female fu- nicle 3-segmented and with first funicular segment usually pedunculate (Fig. 13), clava 3-segmented; male funicle 4-segmented and usually with at least first 2 funicles pedun- culate (Fig. 14); stigmal vein well devel- oped; postmarginal vein equal to or slightly longer than stigmal vein. Discussion. — While no phylogenetic analysis of relationships within the Eulo- phinae has yet been published, several characters of the thorax (e.g. the lateral scutellar grooves, complete notauli, me- dian propodeal carina) strongly suggest that Deutereulophus is closely related to a group of genera that includes Hyssopus, Elachertus, and Diglyphomorphomyia and several others. The shape of the pronotum is very similar to that found in Hyssopus and is a feature not shared by most of the other eulophid genera. The pedunculate male antennae and the usually pedunculate first funicular seg- ment of the females, however, argue for a close relationship to Diglyphomorphomyia (an Australian genus). The presence of a group of setae below the spiracle, which is apparently an extension of the usual line of callus setae, may be unique to this genus. I have not been able to examine all the types of Australian Deutereulophus so it is not possible to make a definitive state- ment about that character at this time. In 418 Journal of Hymenoptera Research some species from the South Pacific re- ed, reticulate behind with inverted v- gion, the area below the callus contains shaped smooth area (Fig. 2); mesoscutum only one or a few setae. and scutellum alutaceous (Fig. 3). The forward projecting flange (Fig. 4) This species is similar to D. smithi but on the dorsal and lateral surface of the differs in that smithi has the vertex with a petiole is also a feature which I have not distinct transverse carina (Fig. 1), and the observed in other similar genera of eulo- area behind the carina is lightly sculp- phines and which seems to be constant hired, nearly smooth, and without a dis- within the species examined. Some species tinct v-shaped area. of Hoplocrepis have lateral flanges on the Description. — Female. Length 1.9 mm. petiole, but no dorsal flange. Color black except: scape yellow, flagel- The posterior bifurcation of the median lum brown; fore coxa brown to black, rest propodeal carina that forms a large open of leg yellow. Head: Face lightly reticulate areola at the posterior median margin of to nearly smooth, sculpture slightly heavi- the propodeum, and the presence of at er below toruli. In frontal view, head wid- least a short forward projecting lateral ca- er than high (40:35). Gena reticulate. Clyp- rina that originates from the anterior lat- eus set off by irregular lateral and dorsal eral margin of the areola (Figs. 4, 6) both suture line. Mandible with one large ven- seem to be unique to Deutereulophus. They tral tooth and 3 smaller dorsal teeth. Malar are present in all species and may be the suture complete, slightly curved. Ratio of most reliable defining characteristic of the malar space: eye height 10:32. Toruli in- group. I have seen species of Hoplocrepis serted level with lower margin of eye. Ra- which have a somewhat similar condition tio of width of face: width of eye 34:15. in which the median propodeal carina Occiput rounded, reticulate and with a ends posteriorly at the nucha but does not central inverted v-shaped smooth area, clearly bifurcate or enclose a cell or areola. Posterior margin of eye on same line as The presence of sinuate lateral grooves posterior margin of gena. Ocelli slightly on the scutellum has been used in keys removed from margin of occiput, POL 2X (e.g. Schauff et al. 1997) but this character, OQl. Antenna: Scape about 7X as long as while stable for all North America species wide Ratio 0f length of Fl:F2:F3:Clava 14: I have examined, is highly variable in oth- H:ll:19, width 6 at Fl to 7 at clava. Me- er parts of the world and has been ob- sosoma: (Fig. 3). Pronotum reticulate ex- served in a modified form in some Aus- cept at postenor margin, with numerous tralian species of eulophids which I would scattered setae. Mesoscutum smooth an- not place in Deutereulophus. Recent au- teriorly/ then becoming reticulate, with 2 thors seem to agree that this genus con- pairg of getae along notaular margin, side tains a natural group of related species, lobes reticulate laterally, then alutaceous and I believe that the characters cited with a of getae along anterior mar. above reinforce that conclusion. Exami- . Axi]la smooth Scutellum alutaceous nation of specimens borrowed from mu- tQ reticulate Metanotum anteriorly pro- seums in the Pacific basin indicate that mi- Qut frQm scutellum and bordered merous species exist in that region and a median dorsa] ed b a of that the center of diversity of Deutereu- ^^ gnd anQther band of fjner al_ ophus is probably the Australasian realm. ... J . . . d-,*~~j«,,«, r r j veoh along posterior margin. Propodeum Deutereulophus arizonensis Schauff, new medially smooth, with about 12 long species white setae laterad of spiracle and about (Figs. 2, 3, 4) 20-25 setae below the spiracle. Petiole Diagnosis. — Legs yellow except fore (Fig. 4) rugose dorsally, as long as wide, coxa black; funicles brown; vertex round- anterodorsal flange well developed, with Volume 9, Number 2, 2000 419 Figs. 1-6. Scanning electron micrographs of Deutereulophus. 1, D. smithi, thorax and head. 2, D. arizonensis, head, posterior view. 3. D. arizonensis thorax. 4, D. arizonensis, propodeum and petiole. 5, D. floridensis mesoma. 6, D. floridensis propodeum. lateral flanges well developed. Forewing. Hyaline, 2.3 X as long as wide. Submargin- al vein with 6-7 dorsal setae. Ratio of sub- marginal: marginal: stigmal: postmarginal vein = 33:33:12:15. Metasoma: Ovate, slightly longer than wide. Ovipositor sheaths reaching just past tip of gaster. Male. — Unknown. Distribution. — Known only from Arizo- na. Types. — Holotype female with data: Ar- izona, Patagonia, 27 June 1953. W. W. Wirth Collector. Deposited in USNM. Etymology. — This species is named for the locality of the type. Deittereulophus floridensis new species (Figs. 5, 6, 12) Diagnosis. — Head, lateral lobes of me- soscutum, and propodeum metallic green; in lateral view, posterior margin of eye not distinctly protruding behind hind margin 420 Journal of Hymenoptera Research of gena; vertex rounded, without trans- verse carina and area behind uniformly smooth or lightly sculptured; mesoscutal midlobe and axilla smooth; scutellum smooth or very lightly alutaceous (Fig. 5); metasoma mostly yellow; legs white to yellow. Male scape with a small sensory patch containing only about 2 sensilla (Fig. 12). This species is similar to D. occularis, which shares a smooth scutellum and yel- low legs. However, D. floridensis has the thorax and head distinctly metallic green whereas D. occularis has a black head. D. floridensis has the posterior margin of the eye in lateral view not projected behind the gena and the vertex is rounded and uniformly scuptured behind (occularis with median transverse carina on vertex and the posterior margin of the eye pro- jecting behind margin of gena (Fig. 7) and a distinctly sculptured area below occiput contrasting with the smooth lateral areas). Description. — Female. Length 1.6-2.0 mm. Color: face, mesoscutal midlobe, ax- illa, and propodeum, occasionally lateral margin of metasoma metallic green; back of head, pronotum, scutellum, lateral tho- rax, ventral thorax, petiole black; antenna yellow except clava sometimes light brown; legs yellow to white; metasoma yellow except with brown on lateral mar- gin and usually with an elongate trian- gular brown area medially on the poster- odorsal surface. Head: Face and vertex smooth. In frontal view, wider than high (57:38). Gena reticulate. Clypeus set off by irregular suture line. Mandible with one large ventral tooth and 3 smaller dorsal teeth. Malar groove complete, slightly curved. Ratio of malar space: eye height = 8:13. Toruli inserted level with lower mar- gin of eye. Ratio of width of face: width of eye = 31:15. Occiput rounded, area be- hind smooth to lightly alutaceous, shiny. Posterior margin of eye in lateral view on same line as posterior margin of gena. Ocelli contiguous with margin of occiput, POL 2X OOL. Antenna: Scape about 6X as long as wide. Ratio of length of F1:F2: F3:Clava = 15:11:12:22, width 6 at Fl to 7 at clava. Mesosoma: Pronotum weakly alutaceous to smooth, with numerous scattered setae (Fig. 5). Mesoscutum smooth anteriorly, then becoming aluta- ceous posteriorly, with 2 pairs of setae along notaular margin, side lobes smooth, with group of setae along anterior margin. Axilla smooth. Scutellum very weakly alu- taceous medially. Metanotum smooth, projecting out from scutellum anteriorly. Propodeum smooth medially, with about 10 long white setae lateral to spiracle and about 12-15 setae below the spiracle (Fig. 6). Petiole rugose dorsally, about as long as wide, anterodorsal flange small with lateral flanges well developed. Forewing. Hyaline, 2.3 X as long as wide. Submargin- al vein with 5-6 dorsal setae. Ratio sub- marginal: marginal: stigmal: postmarginal vein = 40:30:12:15. Metasoma: Ovate, slightly longer than wide. Ovipositor sheaths reaching just past tip of gaster. Male. — Similar to the female except: An- tenna with scape about 5X as long as wide. Flagellum with ratio of F1:F2:F3:F4: clava = 15:15:15:15:24. Scape with a very small sensory patch containing only 2 sen- silla just above midline (Fig. 12). Distribution. — Known only from Flori- da. Types. — Holotype female with data: Florida: Monroe Co., No Name Key, 23.11- 3.VI. 1986. S&J Peck, 86-13, hammock, FLT. Deposited in CNC. Paratypes: 1 fe- male and 2 males with same data; 2 males with same data except 3.VIII-18.XI.1985, S&J Peck, hammock forest, malaise & FLT; 1 female and 1 male with same as previ- ous except 4. V^4. VIII. 1985; 5 females and 3 males same as holotype except Fat Deer Key, 18.XI.1985-25.II.1986, S&J Peck, ham- mock forest, malaise & flight intercept trap; 1 male with same data as previous except 2.VIII-16.XI.1985; 3 females and 5 males same as holotype except Big Pine Key, SI, T67S, R29E, 30.VII-17-XI.1985, S&J Peck, Cactus Hammock, malaise & Volume 9, Number 2, 2000 421 Figs. 7-8. Scanning electron micrographs of Deutereulophus occularis. 7, Head and anterior mesosoma, side view. 8, mesosoma, dorsal view. flight intercept trap, forest; 1 female same as previous except 6.VIII-17.IX.1985; 1 fe- male same as holotype except Everglades National Park, 1.5 km NW Royal Palm, 3.III.-28.IV.1985; 1 female same as holo- type except N. Key Largo, Sec. 35, l.VIII- 16.XI.1985, S&J Peck, hammock forest, malaise & flight intercept trap; 1 female same as holotype except Sugar Loaf Key, Kitchings, 26.II-6.VI.1986, S&J Peck, 86-29, hammock forest FIT; 1 male same as pre- vious except SE Va, S23, 26.II-6.VI. 1986, 86- 31, hammock for., FLT, deposited in CNC; 1 female and 1 male same as previous de- posited in USNM. Etymology. — This species is named for the state locality of the type series, Florida. Deuternlophus occularis Schauff, new species (Figs. 7, 8, 10) Diagnosis. — Legs and antenna yellow; posterior margin of eye projecting behind posterior margin of gena (Fig. 7) in lateral view; vertex with transverse carina and area below distinctly sculptured medially contrasting with smooth lateral area; head and thorax black; mesocutum and axilla smooth. Male scape with sensory patch entending for most of top half of scape with about 13 sensillae (Fig. 10). This species is somewhat similar to Deu- tereulophus floridensis which also has the dorsal thorax mostly smooth and shiny. However, D. floridensis has a distinct me- tallic green sheen to the head (black in oc- cularis), mesoscutal side lobes and propo- deum, and the posterior margin of the gena is on line with the posterior margin of the eye. In addition, D. floridensis has the posterior margin of the eye ending about in line with the posterior margin of the gena when viewed laterally, and the vertex is rounded and not carinate. Description. — Female. Length 2.3 mm. Color: Head and thorax black, antenna and legs light yellow except base of fore coxa brown; metasoma yellow except lat- eral margin of all but first tergum brown and with a median triangular brown spot which covers the last two terga and the median portion of the previous two terga. Head: Face and vertex mostly smooth with faint reticulation near eyes and on vertex. In frontal view, wider than high (56:38). Gena smooth. Clypeus set off by irregular suture line. Mandible with one large ventral tooth and 3 smaller dorsal teeth. Malar groove complete, slightlv curved. Ratio of malar space: eye height = 10:25. Toruli inserted level with lower margin of eye. Ratio of width of face: width of eye = 28:15. Vertex with trans- verse carina, area behind carina smooth laterally and striate alutaceous to reticu- late medially. Posterior margin of eye pro- jecting behind posterior margin of gena (Fig. 7). Ocelli contiguous with margin of 422 Journal of Hymenoptera Research vertex, POL 2.5 X OOL. Antenna: Scape about 6.5 X as long as wide. Flagellum with ratio Fl:F2:F3:clava = 15:11:11:20, width 6 at Fl to 7 at clava. Scape with a sensory patch containing about 13 sensilla extending for almost entire length of top half (Fig. 10). Mesosoma: Pronotum weak- ly alutaceous to smooth, with numerous scattered setae (Fig. 8). Mesoscutum smooth anteriorly, then becoming aluta- ceous, with 2 pairs of setae along notaular margin, side lobes smooth, with a group of setae along anterior edge. Axillae smooth. Scutellum very weakly aluta- ceous medially. Metanotum projecting out from scutellum anteriorly, smooth. Pro- podeum smooth medially, with about 7 long white setae latera to spiracle and about 30 setae below spiracle. Petiole ru- gose dorsally, about as long as wide, dor- sal anterior flange large and tongue-like with lateral flanges well developed. Fore- wing. Hyaline, 2.5 X longer than wide. Submarginal vein with 6-7 dorsal setae. Ratio submarginal: marginal: stigmal: postmarginal veins: 45:50:20:25. Metaso- ma: Ovate, slightly longer than wide. Ovi- positor sheaths reaching just past tip of gaster. Male. — Similar to the female except: An- tenna with scape about 5X as long as wide, with sensory patch about V2 length of scape and containing 12-15 sensillae (Fig. 10). Flagellum with 4 funicular seg- ments and 2-segmented clava. Ratio of Fl: F2:F3:F4:clava = 15:15:15:15:24. Distribution. — Known only from the type locality in Florida. Types. — Holotype female with data: Florida: Monroe Co., NoName Key, 19-XI- 85-25-11-86. S. & J. Peck. Hammock forest. Malaise & FIT. Deposited in CNC. Para- types 1 female and 1 male with same data except female collected 3-VIII-18-XI-85 and male collected 4-III-29-IV-85. Female paratype deposited in USNM, male in CNC. Etymology. — This species epithet refers to the eyes. Deutereulophus pecki Schauff, new species (Fig. 14) Diagnosis. — Legs yellow; antennal funi- cle brown; first funicular segment equal in length to second; mesoscutum, axilla, and scutellum mostly smooth with only very light reticulation on anterior mesoscutum. This species is similar to D. occularis and D. floridensis in that the scutellum is smooth and the legs yellow. However, D. pecki has the first funicular segment of the female antenna equal in length to the sec- ond (distinctly longer in both D. occularis and D. floridensis) and the flagellum brown (yellow in D. occularis and D. flori- densis). Description. — Female Length 1.1 mm. Color: Head and thorax dark brown; scape and legs light yellow; funicle brown; me- tasoma yellow. Head: Face and vertex mostly smooth with faint reticulation near eyes and on vertex. In frontal view, head wider than high (56:38). Gena smooth. Clypeus set off by irregular suture line. Mandible with one large ventral tooth and 3 smaller dorsal teeth. Malar suture com- plete, straight. Ratio of malar space: eye height = 12:25. Toruli inserted level with lower margin of eye. Ratio of width of face: width of eye = 33:15. Vertex acute, but without transverse carina, striate alu- taceous to reticulate above and smooth be- low. Posterior margin of eye ending be- fore posterior margin of gena. Ocelli slightly removed from margin of occiput, POL 2.0 X OOL. Antenna: Scape about 6.5 X longer than wide. Ratio of length of Fl:F2:F3:clava = 10:10:11:24, width 6 at Fl to 7 at clava (Fig. 14). Mesosoma: Prono- tum weakly alutaceous to smooth, with numerous scattered setae. Mesoscutum smooth anteriorly to very weakly reticu- late at lateral margin, with 2 pairs of setae along notaular margin, side lobes smooth medially and weakly reticulate at lateral margin, with a single seta along anterior edge. Axilla smooth. Scutellum smooth Volume 9, Number 2, 2000 423 EH FW occularis smithi floridensis Figs. 9-14. Deutereulophus. 9, Generalized head, anterior view. OOL = ocellar ocular length. POL = posterior ocellar length. EH = eye height. EW = eye width. FW = face width. FH = face height. MS = malar space. HW = head width. 10-12, male scapes. Figs. 13-14. Antennae. 13, female. 14, male. with only a faint hint of reticulation me- dially. Metanotum smooth, projecting out from scutellum anteriorly. Propodeum smooth medially, with about 7 long white setae lateral to spiracle and about 6 setae below spiracle. Petiole rugose dorsallv, about as long as wide, and anterodorsal flange large and tongue-like with lateral flanges well developed. Forewing. Hya- line, 2.4X as long as wide. Submarginal vein with 4 dorsal setae. Ratio submargin- al: marginal: stigmal: postmarginal veins = 40:30:20:15. Metasoma: Ovate, slightly longer than wide. Ovipositor sheaths reaching just past tip of gaster. Male. — Unknown. Distribution. — Known only from the type locality in Florida. Types. — Holotype female with data: Florida: Monroe Co., Key Largo, Sec. 35, 424 Journal of Hymenoptera Research 1-VIII-16-XI-85. S&J Peck. Hammock for- est. Malaise & FLT. Deposited in CNC. Etymology. — This species is named for Stuart Peck who collected many of the specimens included in this study. Deutereulophus smithi Schauff, new species (Figs. 1, 11, 13) Diagnosis. — Legs except coxae white to yellow except fore coxae black; first funic- ular segment distinctly longer than sec- ond; occiput with distinct median carina, evenly, but very weakly sculptured, near- ly smooth behind posterior ocelli; mesos- cutal side lobes reticulate or alutaceous; scutellum alutaceous. Male scape with small sensory patch containing about 6 sensillae (Fig. 11). This species is similar to D. arizonensis, which also has mostly yellow legs (except for the fore coxae). However, D. arizonen- sis has a rounded vertex without a distinct transverse carina and occiput with a tri- angular smooth spot bordered laterally by distinct scupture (Fig. 2). Description. — Female. Length 2.0-2.2 mm. Color black except as follows: scape white, pedicel and first funicular segment yellow, remainder of flagellum brown; fore coxa brown or brown basally, becom- ing yellow in apical half, rest of legs yel- low to white with occasional brownish in- fuscation on fore femur. Head: Face weak- ly reticulate to nearly smooth, sculpture slightly stronger below toruli. Head in frontal view, wider than high (FH: FW = 60:43). Gena reticulate. Clypeus set off by irregular suture line. Mandible with one large ventral tooth and 3 smaller dorsal teeth. Malar suture complete, slightly curved. Ratio of malar space: eye height = 8:13. Toruli inserted level with lower mar- gin of eye. Ratio of width of face: width of eye = 38:15. Vertex with distinct trans- verse carina, occiput smooth to weakly alutaceous, shiny. Posterior margin of eye on same line as posterior margin of gena. Ocelli contiguous with margin of occiput, POL 2X OOL. Antenna: Scape about 7X as long as wide. Ratio of length of F1:F2: F3:clava = 20:14:15:25, width 6 at Fl to 7 at clava (Fig. 13). Mesosoma: Pronotum reticulate except at posterior margin, with numerous scattered setae (Fig. 1). Mesos- cutum smooth anteriorly, then becoming alutaceous to reticulate posteriorly, with 2 pairs of setae along notaular margin, side lobes reticulate laterally, then alutaceous with a group of setae along anterior mar- gin. Axilla smooth. Scutellum alutaceous to reticulate. Metanotum projecting out from scutellum anteriorly and bordered along median dorsal edge by a group of alveoli and by another band of finer al- veoli along posterior margin. Propodeum medially smooth, with about 12 long white setae laterad of spiracle and about 20-25 setae below spiracle. Petiole rugose dorsally, slightly wider than long, anter- odorsal flange small with lateral flanges well developed. Forewing. Hyaline or with very slight infuscation below mar- ginal vein, 2.1 X longer than wide. Sub- marginal vein with 6-7 dorsal setae. Ratio submarginal: marginal: stigmal: postmar- ginal vein = 35:35:12:15. Metasoma: Ovate, slightly longer than wide. Ovipos- itor sheaths reaching just past tip of gas- ter. Male. — Similar to female except as fol- lows: Fore coxae sometimes completely yellow; antenna with scape about 5X as long as wide and with a small sensory patch containing about 7 sensillae just above midline (Fig. 11). Flagellum with ra- tio of Fl:F2:F3:F4:clava = 20:17:17:17:24. Distribution. — Florida, Georgia, Louisi- ana, Maryland, Texas, and Virginia. Most records are from Maryland, Virginia, and Florida but this is no doubt due to exten- sive collecting in these areas. The range is almost certainly broader than indicated by the specimens available. Variation. — The forewing of females may have a noticeably infuscate brown area behind the marginal vein which ex- tends to the hind margin of the wing. Col- Volume 9, Number 2, 2000 42? oration of the metasoma ranges from near- ly entirely black to mostly yellow with some brown spots laterally. The fore coxa may be entirely yellow or have the basal half brown to black. In one specimen from Texas with a very darkly infuscated area on the forewing, the legs are also light brown. The flagellum varies from a dark honey yellow to brown but in most spec- imens the color is distinctly lighter on the first funicle. Types. — Holotype female with data: Vir- ginia: Essex Co., 1 mi. E. Dunnsville, 17- IX-10-X-1991, Malaise trap. D. R. Smith. Deposited in USNM. Paratypes: 3 females same as holotype; 3 females Virginia: Louisa Co., 4 mi. S Cuckoo, 12-27. V.1987, J. Kloke & D. R. Smith, malaise trap; 1 fe- male and 4 males with same as previous except 19.VIII-2.X.1987; 1 female and 1 male 16-31. VII. 1987; 1 male 25.VI- 5.VTI.1987; 2 males 3-24.IX.1987; 1 female VA: Page Co., Shenandoah Nat. Pk., 5- 22.V.1987, 1300m, CNC [BRD] Hym. team, malaise trap in meadow; 1 male VA: Fair- fax Co., near Annandale, 29.III-11.IV.1988, malaise trap, D.R. Smith; 1 male Mary- land: Howard Co., Clarksville, 3.VIII.1986; 1 female same as previous except 13.VII.1986; 2 females Florida: Monroe Co., Big Pine Key, Watson's Hammock, 3-VI-27.Vin.1986, S&J Peck, malaise trap deposited in USNM; 1 female Texas: Jim Wells Co., 8 mi. W Ben Bolt La Copita Re- search Station, 20. V.1987, 87/006, J.B. Woolley deposited in TAMU; following all deposited in CNC: 4 females Maryland: Calvert Co., 6 km W Prince Frederick, 18- 26. VIII. 1986, Sharkey; same as previous except 3 females and 5 males 25.VIII.1986; 2 females and 1 male 7 km S Prince Fred- erick, 24.IX-14.XI.1987, Malaise trap, hard- wood forest, CNC [BRD] Hym team; 3 males 4 mi. S Prince Frederick Co., 16. IV- 7.V.1987, L. Masner, Flight intercept trap; Calvert Co., 1 female Scientist's Cliffs, 7. VII. 1987, G. Gibson; 1 female Port Re- public, VIII-IX.1986, Sharkey & Monroe; 1 female Chesapeake Bay Beach, 13. VI. 1985, L. Masner, s.s., plants in forest; 2 males Seneca, Potomac River trail Mouth of Sen- eca Creek.), 16.VI.1986, L. Masner, SS, un- dergrowth along old canal; 1 male Prince George's Co., Paruxent Research Station 21-29.VI.1986. D. Wahl, Malaise trap; 1 fe- male Kentucky: Rowan Co., 24 km SW Morehead, Cave Run Lake, 14. V- 20.VIII.1983, M. Kaulbars; 1 female and 1 male Georgia: Clarke Co., Lake Herrick, Oconee Forest Park., 11-12.VII.1987, L. Dumouchel; 1 male Mcintosh Co., Sapelo Island, 9-21.IX.1987, live oak forest, CNC [BRD] team; 1 male 15.IX-16.XI.1987; Ath- ens, 1 female 14.IX.1987, L. Masner; 1 fe- male and 1 male Louisiana: Grant Parish, 28 km N Alexandria, Stuart Lake Camp- ground, 19-21. V. 1983, M. Kaulbars; 4 males Florida: Alachua Co., Gainesville, l.V-20. VIII. 1988, D. Wahl, flight intercept trap; 1 female same as previous except (AEI), 30.IV.1987, SS, L. Masner, 87/14; 1 male 10-20.11.1987, W. Mason; 1 female 8- 14.IV.1987; 1 female 15-22.111.1987, Mal- aise trap, hardwood forest; 1 male Amer- ican Entomological Institute, 9-17.IV.1986, G. Gibson, sweep; 3 females and 5 males Tallahassee, 18-23.V.1986, H. Howden, flight intercept trap; 3 females Monroe Co., Big Pine Key, Watson's Hammock, I.XI.84-3.III.1985, S&J Peck flight intercept trap; 1 female same as previous except 19.XI.1985-25.II.1986; 1 female 3.V- 3.VIII.1985; 23.II-3.VI.1986; 1 female 3.VL- 27. VIII. 1986, hammock forest, Malaise trap /flight intercept trap, 86-10; 2 females Dade Co., S Miami, 7900 Swth St., Old Cutler Hammock, 21.II-1.VI.1986, flight intercept trap, S&J Peck, hammock; 1 fe- male Chekika State Recreation Area, 50 km SW Miami, Grossman Hamrruu k 1.IX.1984-3.III.1986, S&J Peck, flight inter- cept trap; 1 male 3.III.-28.IV.1985; 2 fe- males and 3 males Fat Deer Key, 4. III- 28.IV.1985, S&J Peck, hammock forest, malaise & flight intercept trap; 4 females Monroe Co., Everglades National Park, Royal Palm Hammock; 1.XI-3.III.19 S&J Peck, malaise; 2 females and 3 males 426 Journal of Hymenoptera Research 1.5 km NW Royal Palm, 3.III.-28.IV.1985, hardwood hammock forest, malaise-flight intercept trap; 2 females N. Key Largo, Sec. 35, 4.III^.VIII.1985, S&J Peck, ham- mock forest, malaise & flight intercept trap; 4 females and 1 male same as pre- vious except 4.III-28.IV.1985; 1 male 4.VIII-16.XI.1985; 1 female St John's Co., Theodore Roosevelt Preserve, 13. X. 1980, Masner & Bowen, 8029; 1 female Liberty Co., Torreya St. Pk, 7.X.1980, 8022, Masner & Bowen; 4 females and 4 males Texas: San Jacinto Co., 5 km S Coldspring Double Lake Campground, 22-24.V.1983, M. Kaulbars; 1 male Brazos Co., College Sta- tion, 1982, R. Wharton, M. Hrncir, pan trap. Etymology. — This species is named in honor of David R. Smith, Systematic En- tomology Laboratory, USD A, who collect- ed part of the type series and whose col- lecting over the years has added greatly to the U.S. National Collection of Insects. ACKNOWLEDGMENTS I thank Dr. C. Burwell, Queensland Museum, Bris- bane; Dr. S. Heydon, Bohart Museum, University of California, Davis; Dr. J. LaSalle, CABI Bioscience, UK; and Dr. J. Huber, Canadian Forestry Service and Ca- nadian National Collection, Ottawa for the loan of specimens. Ms. Tami Carlow provided technical as- sistance and editorial support. Drs. S. Heydon, N. Vandenberg, and D. Smith provided useful com- ments on drafts of the manuscript. LITERATURE CITED Ashmead, W. H. 1904. Classification of the chalcid flies of the superfamily Chalcidoidea, with de- scriptions of new species in the Carnegie Muse- um, collected in South America by Herbert H. Smith. Memoirs of the Carnegie Museum 1: i-ix, 225-551. Gibson, G. A. P. 1997. Chapter 2, Morphology and terminology, pp. 16-44. In Gibson, G. A. P. et al., (eds.) Annotated keys to the genera of North Amer- ica)! Chalcidoidea (Hymenoptera). NRC Research Press, Ottawa. Girault, A. A. 1913. Australian Hymenoptera Chal- cidoidea— IV. The family Eulophidae with de- scriptions of new genera and species. Memoirs of the Queensland Museum 2:140-296. Girault, A. A. 1915. Australian Hymenotpera Chal- cidoidea IV. Supplement. Memoirs of the Queens- land Museum 3:180-299. Girault, A. A. 1922. New chalcid flies from eastern Australia (Hymenoptera, Chalcididae). II. lnse- cutor Inscitiae Menstruus. 10:100-108. Girault, A. A. 1938. Some new Australasian insects which are parasites (Hym., Chalcidoidea). Revista de Entomolgia, Rio de Janeiro 8:80-89. LaSalle, J. and M. E. Schauff. 1992. Preliminary stud- ies on neotropical Eulophidae (Hymenoptera: Chalcidoidea): Ashmead, Cameron, Howard and Walker species. Contributions of the American En- tomological Society 27: 1-47 pp. Moschler, H. B. 1878 [1877]. Beitrage zur Schmetter- lings-Fauna von Surinam. 11. Verhandlungen Zoologisch-botanischen Gesellschaft im Wien 27:629- 700. Schauff, M. E., J. LaSalle, and L. Coote. 1997. Family Eulophidae, p. 327-429. In Gibson, G. A. P. et al., (eds.) Annotated keys to the genera of North Amer- ican Chalcidoidea (Hymenoptera). NRC Research Press, Ottawa. Schulz, W. A. 1906. Spolia Hymenopterologica. 355pp. Paderborn. Yoshimoto, C. M. and T. Ishii. 1965. Insects of Micro- nesia: Hymenotpera Chalcidoidea: Eulophidae, Encyrtidae (part), Pteromalidae. Insects of Micro- nesia 19: 109-178. J. HYM. RES. Vol. 9(2), 2000, pp. 427-429 NOTE The Presence of Antero-lateral Abdominal Glands in Euderomphale (Hymenoptera: Chalcidoidea: Eulophidae) John LaSalle and Andrew Polaszek Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berks, SL5 7PY, UK Quicke et al. (1997) discussed the pres- ence of antero-lateral abdominal glands (ALAGs) in the braconid subfamily Bra- coninae, and provided detailed anatomi- cal descriptions of them. Virtually all bra- conines possess these glands, although they are not known in other braconid sub- families. They are eversible, sac-like glan- dular invaginations of the unsclerotized lateral cuticle between the terga and ster- na on the first and second metasomal (sec- ond and third abdominal) segments. AL- AGs are present in both sexes, and pro- duce an odoriferous secretion which is characteristic of braconine wasps. The exact function of ALAGs in bracon- ines is unknown. Since the glands are everted, and their product secreted, most notably when the wasps are disturbed, Quicke et al. (1977) suggested that the product might be a chemical that was dis- tasteful to predators. However, they also discussed several problems with this pos- sibility, including that braconines were not distasteful to vertebrates, and that ar- thropods avoided non-braconines (with- out ALAGs) as readily as they avoided braconines (with ALAGs). Similarly, the role of ALAGs in the production of sex pheremones was questioned as these glands are present in both sexes. Another suggestion was that the ALAGs play a more general role in intra-specific signal- ling, such as an aggregation or alarm pheremone. Quicke et al. (1997) recorded ALAGs only in Braconinae and Diprion similis (Diprionidae), although they con- cluded that it was unlikely that these are homologous structures. The present paper reports the presence of ALAGs in the eulophid genus Euder- omphale, species of which are parasitoids of whiteflies (see LaSalle and Schauff 1994 for a discussion of the systematic place- ment of this genus). This is the first report of the presence of these glands in any member of the Chalcidoidea. The ALAGs in Euderomphale seem sim- ilar to those of braconines in that they ap- pear to be eversible, sac-like invaginations of the unsclerotized cuticle between the terga and sterna of the second metasomal (third abdominal, first gastral) segment. There is only a single pair of the glands, and these are present in both sexes (Figs. 1-6). As with braconines, their function is unknown, but it is not clear that they are used in response to disturbance, as one of us (AP) has actually observed Euderom- phale held in plastic bags everting and re- tracting these ALAGs while there was nothing obvious to disturb them. It may be that in Euderomphale, as suggested by Quicke et al. (1997), the ALAGs play a more general role in intra-specific signal- ling, but what this role might be remains unknown. It is also worth noting that the shape of the anterior margin of the gastral tergites appears to provide useful characters for differentiation of species in Euderomphale. For example, in £. cortinae Graham (Figs. 428 Journal of Hymenoftera Research Volume 9, Number 2, 2000 42m 3-4) the second gastral tergite is deeply emarginate medially, and there are similar emarginations on tergites 3 and 4, al- though these are more prominent in males (Fig. 4) than females (Fig. 3). In £. flavi- media (Howard) (Figs. 5-6), the first gastral tergite is slightly produced medially with a very small incision, with very broad, shallow lateral emarginations; tergites 3 and 4 are entire or only very slightly emarginate. This condition is found in both males and females. Slide-mounted material is necessary to clearly see this character. LITERATURE CITED LaSalle, J. and M.E. Schauff. 1994. Systematics of the tribe Euderomphalini (Hvmenoptera: Eulophi- dae): parasitoids of whiteflies (Homoptera: Al- eyrodidae). Systematic Entomology 19: 235-2^ Quicke, D.L.J., R.A. Wharton and H. Sittertz-Bhatkar. 1997. Antero-lateral abdominal scent gland-- oi braconine wasps (Hymenoptera: Braconidae). Journal of Hymenoptera Research 6: 219-230. Figs. 1-6. Euderomphale species. 1-4. Euderomphale cortinae Graham. 1, male gastei showing ALAGs. 2, close up of ALAG. 3, female gaster. 4, male gaster. 5-6. Euderomphale flavimedia (Howard). 5, female gaster. ft, male gaster. J. HYM. RES. Vol. 9(2), 2000, pp. 430-431 NOTE First Possible Host Record for the Braconid Wasp Genus Diamblomera Enderlein (Hymenoptera: Braconinae) Donald L. J. Quicke, Gavin Broad, Nina M. Laurenne, and Jupailin Naiman (DLJQ, GB) Unit of Parasitoid Systematics, CABI Bioscience UK Centre (Ascot), Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, U.K., Centre for Population Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, U.K., (DLJQ) Department of Entomology, The Natural History Museum, London SW7 5BD, U.K.; (NML) Finnish Museum of Natural History, Zoological Museum, Entomological Division, P.O. Box 17 (P. Rautatiekatu 13), FIN-00014 University of Helsinki, Finland. (JN) Tropical Biodiversity and Conservation Unit, Universiti Malaysia Sabah, Sabah, Malaysia. Determining host associations for para- itor more or less between her hind legs sitic wasps is fraught with difficulties and and started 'drilling'. This activity was many published records are erroneous watched for approximately 15 minutes (Noyes 1994; Shaw 1994). These errors and as she then started moving her me- stem both from misidentifications of hosts tasoma around more noticeably, it was and parasitoids, and from wrongly assum- thought that she may have located a host ing associations. They are particularly or even oviposited. At this point she was problematic when it comes to concealed caught and the exact position where her hosts, especially those living in deep or ovipositor was 'drilling' was carefully not- potentially complex situations where there ed. We proceeded to cut into this piece of may be more than one species, and the dead vine. The stem of the vine is flat- identity of the true host or hosts in these tened and at this place was approximately situations is often ambiguous. 10 cm wide by 3.5 cm thick. Along a Field observations in Sabah of a large bra- length of vine of approximately 40 cm we conine wasp, Diamblomera sp., strongly in- discovered more than twenty, apparently dicate that it is a parasitoid of larvae of cer- conspecific cerambycid beetle larvae be- ambycid beetles belonging to the subfamily longing to the subfamily Lamiinae (iden- Lamiinae. Details of our observations and tified as such because of the complete ab- identifications are provided below. sense of legs, and the relatively elongate During a reconnaissance visit to Poring head with the cardines, submentum and Springs, a resort within lowland rainforest maxillary articulating areas fused). These of Kinabalu National Park, Sabah, NML were almost all of very similar size (ap- and DLJQ noticed a large female braconid proximately 20 mm long, range 12-23 wasp flying around a large dead (uniden- mm, and 2-3 mm wide) and apparently of tified) tree. However, the wasp subse- an appropriate size to be a host for the quently settled on a vine hanging from the wasp, Diamblomera sp., the females of tree and a second, apparently conspecific which were approximately 18 mm long female wasp was then noticed already to (excluding ovipositor), but a little narrow- be sitting on the same piece of vine about er than the beetle larvae. No other poten- 20 cm away. Both wasps intermittently tial hosts were present, i.e. no other even walked over that piece of vine, apparently remotely similarly sized insect larvae were searching for hosts, and after some time found. Unfortunately, no host was found (approximately 5 minutes) one female immediately below the point of ovipositor raised her metasoma, 'located' her ovipos- penetration but since it took almost twen- Volume 9, Number 2, 2000 431 ty minutes for us fully to cut out and dis- sect this piece of vine, it is quite possible that any beetle larva had simply moved away from that place, possibly due to the disturbance we caused. A third individual of the wasp was subsequently observed flying around the site but she did not land on the now damaged vine. The vine was identified as Agelaea borneensis (Hook. F.) Merr. (Connaraceae) by Mr. Sukup Akin. Agelaea is a common SE Asian genus of trees, vines and shrubs (Jarvie and Ermay- anti 1996 onwards). Given that two females of the same spe- cies of Diamblomera were intensively inves- tigating the same piece of host substrate with one starting to go through oviposition behaviour, and that the substrate contained many, apparently suitably sized larvae of a single species of cerambycid beetle, we feel confident that this is a valid host-parasitoid association. Unfortunately there are no identification keys to the species of Diam- blomera. Only two species were described under that generic name originally (Ender- lein 1920; Quicke and Achterberg 1990), and no further species have subsequently been transferred to it despite ongoing reclassifi- catory work; however, there probably exist other described species that are currently classified under different genera and full re- vision is needed. This is the nearest thing to a first host record for a member of the genus Diam- blomera, although it may be objected that none of the more than twenty putative hosts seen appeared to have been parasit- ised. The larger braconines belonging to the Aphrastobraconini (= Iphiaulacini) are often parasitic on concealed wood or stem boring hosts and this association with a vine-feeding cerambycid is therefore not exceptional, though records of parasitism of hosts in vines (lianas) are rare. Further, from what little is known about the hosts of the larger braconines (almost entirely from temperate taxa, particularly those as- sociated with forestry pests), it appears that many attacking subcortical beetles may be quite polvphagous — but whether all the host records that are listed in She- nefelt (1978) for members of such bracon- ine genera as Atanycolus Foerster, which may have similar biologies to Diamblom- era, are reliable is far from certain. That there is some degree of specialisation is apparent from the fact that, at the same locality, we observed several other species of large Braconinae — belonging to other genera — that were each showing interest in different sites or dead trees (Laurenne et al. 2000). Only the accumulation of ac- curate host records and other details of host ecology, such as substrate, plant or fungal associations, for these taxa will start to show what factors may be impor- tant in determining host ranges. The specimens of Diamblomera and the beetle larvae are deposited in The Natural History Museum, London. ACKNOWLEDGMENTS We would like to thank Professor Marvati and Dr. Homathevi Rahman (TBCU, UMS, Sabah) for their assistance with planning this visit, Mr. Sukup Akin who kindly managed to identify the vine from a rath- er poor specimen, and Dr. Mark Shaw for valuable improvements to the MS. LITERATURE CITED Enderlein, G. 1920. Zur Kenntnis aussereuropaischer Braconiden. Ardriv fiir Naturgesddchte 84iAk 51- 224. Jarvie, J.K. and Ermayanti (1996 onwards). Tree Gen- era of Borneo — Descriptions and Illustrations. http://django.harTMtrd.edu/users/jjartrie Borneo.htm. Laurenne, NI. M., R. Belshaw, G. Broad and D. L. J. Quicke. 2000. Molecular confirmation of host re- cords for ichneumonoid parasitoids of wood-bor- ing beetle larvae, journal of Hymenoptera Research 9:241-24- Noyes, J. S. 1994. The reliability of published host-par- asitoid records: A taxonomist's view. Norwegian Journal of Agricultural Science, Supplement 16: 59-69. Quicke, D. L. J. and C. van Achterberg. 1990. The type specimens of Enderlein's Braconinae (Hymenop- tera: Braconidae) housed in Warsaw. Tijdschrifl voor Entomologie 133: 251-264. Shaw, M. R. 1994. Parasitoid host ranges. In: B. A. Hawkins and \\ Sheehan 'I ds). Parasitoid Com- munity Ecology. Oxford University Press, Oxford, pp. lil-144 Shenefelt, R.D. 197S. Hymenopterorum Catalog) Editio). Braconidae 10. lunk, The Hague, pp. 142- 1872 J. HYM. RES. Vol. 9(2), 2000, pp. 432 CORRECTION P. S. Ward. On the identity of Pheidole vaslitii Pergande (Hymenoptera: Formicidae), a neglected ant from Baja California. Journal of Hxjmenoptera Research 9(1): 85-98. Page 94, left column, line 30. The name proof stage by the editor, who regrets hav- "vaslilli" is incorrect and should be spelled ing added a further complication to the vaslitii. This error was created at the nomenclature surrounding this name. INSTRUCTIONS FOR AUTHORS General Policy. The Journal of Hymenoptera Research invites papers of high scientific quality reporting comprehensive research on all aspects of Hymenoptera, including biology, behavior, ecology, systematics, taxonomy, genetics, and morphology. Taxonomic papers describing single species are unlikely to be accepted unless a strong case is evident, such as importance in economic entomology or with concurrent biology or ecology. Manuscript length generally should not exceed 50 typed pages; however, no upper limit on length has been set for papers of exceptional quality and importance, including taxonomic monographs at generic or higher level. All papers will be reviewed by at least two referees. The referees will be chosen by the appropriate sub- ject editor. However, it would be helpful if authors would submit the names of two persons who are compe- tent to review the manuscript. The language of publication is English. Summaries in other languages are acceptable. The deadline for receipt of manuscripts is 1 October (for the April issue) and 1 April (for the October issue). Format and Preparation. Three copies of each manuscript, including copies of illustrations, should be submitted on letter size or A4 paper, double spaced, with at least 25 mm margins on all sides. On the upper left of the title page give name, address and telephone and fax numbers of the author to whom all corre- spondence is to be sent. The paper should have a concise and informative title, followed by the names and addresses of all authors. The sequence of material should be: title, author(s), abstract, text, acknowledgments, literature cited, appen- dix, figure legends, figure copies (each numbered and identified), tables (each numbered and with heading). Bach of the following should start a new page: (1) title page, (2) abstract, (3) text, (4) literature cited, (5) figure legends, (6) footnotes. Following acceptance of the manuscript, the author should provide the editor with one copy of the manuscript accompanied by a copy on diskette using DD, double sided computer diskettes — IBM compati- ble MS DOS 5.25 inch or IBM and Macintosh 3.5 inch diskettes. (Authors who do not have access to a com- puter should submit three copies of the manuscript.) The paper may be submitted in most PC and Mac word processor programs such as Microsoft Word, FullWrite Professional, WordPerfect, WriteNow, Nisus, Mac- Write, or MacWrite II. If possible, all words that must be italicized should be done so, not underscored. Tables may be formatted in a spread sheet program such as MS Works or MS Excel. Text should be double- spaced typing, with 25 mm left and right margins. Tables should be put in a separate file. Diskettes should be accompanied by the name of the software program used (e.g., WordPerfect, Microsoft Word). Authors should keep backup copies of all material sent to the Editor. The Society cannot be responsible for diskettes or text mislaid or destroyed in transit or during editing. Illustrations should be planned for reduction to the dimension of the printed page (14 X 20.5 cm, column width 6.7 cm) and allow room for legends at the top and bottom. Do not make plates larger than 14 X 18 in. (35.5 X 46 cm). Individual figures should be mounted on a suitable drawing board or similar heavy stock. Photographs should be trimmed, grouped together and abutted when mounted. Figure numbers should be on the plate, and it is strongly recommended that names be included after the numbers (e.g., Fig. 2, texanus). Include title, author(s) and address(es), and illustration numbers on back of each plate. Original figures need not be sent until requested by the editor, usually after the manuscript has been accepted. Reference to figures/tables in the text should be in the style "(Fig. 1)" "(Table 1)". Measurements should be in the metric system. All papers must conform to the International Code of Zoological Nomenclature. The first mention of a plant or animal should include the full scientific name including the authority. Genus names should not be abbre- viated at the beginning of a sentence. In taxonomic papers type specimens must be clearly designated, type depositories must be clearly indicated, and new taxa must be clearly differentiated from existing taxa bj means of keys or differential diagnoses. Authors are required to deposit all type material in internationally recognized institutions (not private collections). Voucher specimens should be designated for specimens used in behavioral or autecological studies, and they should be deposited similarly. Acceptance of taxonomic papers will not require use of cladistic methods; however, authors using them will be expected to specify the phylogenetic program used (if any), including disc ussnm of program options used. A data matrix should be provided if the subject is complex. Cladograms must be hung with chara< ters and these should include descriptors (not numbers alone) when feasible. The number oi parsimonious cladograms generated should be stated and reasons given for the one adopted. I engths 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 b) a single author should be (Smith 1999a, 1999b) or Smith (1999a, 1999b). lor 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, lournal names should be spelled out completely and in italics. Charges. Publication charges are $10.00 per printed page. At least one author of the paper must be a member of the International Society of HymenopteristS. Reprints are charged to the author and must be ordered when returning the proofs; there an' no free reprints. .Author's corrections and i hanges in proof are also charged to the author. Color plates will be billed at full cost to the author All manuscripts and correspondence should be sent to: Dr. E. Fric C.rissell Systematic Entomology Laboratory, USDA r, National Museum ot Natural History Washington, DC 20560-0168 Phone: (202) 382-1781 Fax: (202) 7Xh-W422 E-mail: ci;rissel(