Vr ay EL TLTe ee im ¥ “UA Mrmr ane ‘i coats y i War : : : eee: : Ea Coda martes . . : sot ewes EAM Aon? alinsinnindie a) usu" Sy eet Sindits Ay nttaay, . cua ; 3 \ tf Ta iene wckokahye daa :, : : ; tet 4 ain Perera o : : ‘ ' ‘ 7 4 i x * WradNerrdviningetnoa, mea tat, Behe al Sesto tm lagal Man acnsasucsanintic irom mr cant ie HerteAtqndagie ts aay aTh wali i RPL eran saFaardan dade atyeh At 2. anes A eWenia st a * ae vay Se ae Frew kliay UiKe san oe ey x Tava ata jebadteada'y mda hain EE SNE Wt Sai? Pannen tan mate tin Fave aadanh) ; rina Sue Stare? LGL Saeituintare In a tbe tyes Nese nth SS VR aay Nk Masa ty egy £ FTN EA a HUAN A SD hia fe teae Shir PEE Wwe Heres, SIEERS on AS! o7aeataeee Deiinttacen ett es cists Meena ers arity petri f wrvene sais Wein EG Pte re aa rrertn rie ere Rtrane? yee Taggers Pitan ve Bete ae Macey Hae TG errors Partie Nyt dandy cy Slee Re PCR Sts yg a Pea are UIP Gana uae are en Ae ite Be PAT CY rf Detgse ene Piatra et uiaprinny weceoerd arte weyt Wein “ a wate siles 28y sane Hy SPE ss wT ate ae Seg ayer yoy Abb dng 1, Y ant ist t nar aes ras errr aan Oley tie iededeg Ly rp ea " wesiye Tau et pat igs ae ini ‘re pide nine or WEEE piateestar ety Semler etry Witzrety wpb Pe wat eee te FE eee rodeo x Wey ate cease enon ner oie Dxote) eee PATO ACM ear eT SUE ICEL py ee ee tied Fer ean bn eeu eee et tae Oy sews Ueber pier yy art ye ove Pe agalieihy cepa tater Dia seosda Fat na yy ern enn viene Tee ray lenny A Min sah hen Pie at Dime et esrb fy i, sein Saree re emt aries 2d An iH ee a ee. ee ne a AN act heh poled et bdr PALS I oe te, eee ea ey Oh, wot GUvuqaie at tt oe SMITHSONIAN LG hd ; NVINOSHLIWS NVINOSHLIWS: SMITHSONIAN NVINOSHLIWS bs SMITHSONIAN INSTITUTION NOILNLIL NVINOSHLINS S3IYVYAIT LIBRARIES LIBRARIES SMITHSONIAN NOILNLILSNI LIBRARIES NOILALILSNI NOILALILSNI LIBRARIES -IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SAIYVYSIT LIBRAR S Ee = =e z = ; = ee ro) E 2 5 2s = 2 ae > = NS, * =) > =? a = a R i ) = 2 iS eu S ES Z ° Z ? 2 ° NOILALILSNI NVINOSHLINS S31YVUGIT_LIBRARIES | SMITHSONIAN INSTITUTION. NOILMAL. & S ES | ae ee 4 7 BQ 3 z Iw g g ZR 8 z by = z E INS Zz S Zz LY i 2 a We = 2 @ n” LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHLINS _LIBRAR 2) > > : Sg tes” macs! 4 gWS = = “PY - SRS EF 5 = “fy = z a SY 5 = ae we : eae a : J! NOLLNLILSNITNVINOSHLINS | S318VYGI7_ LIBRARIES, SMITHSONIAN” INSTITUTION | NOLLNLIJ = ae fe) = S) = fe) ies) — wo — “Ly ioe] — > We 5 2 —« bey > 5 > Ss = > = fe Yh. > = 2 WY = 2 = OF KF? = = w = aa SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSHLIWS, S3lYvualy LIBRAR = = z Es : = Ss NXg : : | nv 7) ~~ SA n 7) \ SS e) DS : AW re) e) Wes a E WN Z, =i SS Ps aS = = xs NOILNLILSNI SMITHSONIAN INSTITUTION NOILNL11 NVINOSHLINS S3ZIY¥YVYd!I1 LIBRARIES \ LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYE!IT LIBRAR LIBRARIES SMITHSONIAN LIBRARIES SMITHSONIAN LIBRARIES NOILALILSNI NOILNLILSNI NOILNLILSNI INSTITUTION Ssaluvugiq saluyvyaly INSTITUTION INSTITUTION NOILMNLILSNI NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLI- % WS S31uv¥ai7 LIBRAR ae Ps aa NVINOSHLINS S3IYVYSIT SMITHSONIAN NYINOSHLIWS NVINOSHLIWS SMITHSONIAN G NOILNLILSNI NVINOSHLIWS SMITHSONIAN _ INSTITUTION S SMITHSONIAN S N S en a a ae a Oe 1 Paes * Ww Ve ome = hades ) ) 2» LY fy 2 > \ 2 > = if | Z nh eee 7 z @ VINOSHLINS S31YVUaIT_LIBRARIES | SMITHSONIAN INSTITUTION |, NOILNLILSNI NVINOSH “e+ rad a = = SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLIWS S31YVYS17_ LIBRARIES SMITHSO! | = = “ = n | & = % K& # 3 = Oneee = fod = AX. No « =a ce 4 7 = < = \ = <3 Yb 5 cc G WMO «= 5 a Wife fs) = 5 = 5 a “je z = z a =) z =J S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSH is cm SB im z ss 5 2 - Gy > = cs D Lily, P pe) = 5 2 Ji f= = : e a E bl i = E a = m a” m n m — (2) = on = wn INSTITUTION NOILNLILSNI NVINOSHLINS S31uVYGgIT_ LIBRARIES Zz “s 72) z n ae ” ae = < = =< ES x = = = : 4H z Na = 3 E 5 Ne i a ~~ S 7) salt?) o a YK MQY 2 r ° WE Oo zc fo) fs og A AY (e) E WN Zz, = = i NS zZ = NY >" = > = Ss > a re = ” = n = NVINOSHLINS S31YVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSH 7p) = o a - = | = a = 4 or = & EN | c = Ps yy, = = AN j . a4 =a j = > = > | = z 2) oy hea 7) 2 7) | ae INSTITUTION NOILNLILSNI_NVINOSHLINS S31YVY8IT_LIBRARIES SMITHSO = = 7) = n } G6 KS a wu ey G& All numbers are mean + standard error. ‘Only one 7. nigrescens plant used; 3 eggs laid in a single stem. 12 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON } 4 Fig. 2a-f. Egg of L. mozardi. a, External appearance of the oviposition site. b, Rim surrounding the oviposition opening internally. c, Tapered end of the egg attached to the plant tissue. d, L. mozardi egg. e, Micropylar area. f, Chorion surface. E = egg; Ex = egg extension; M = micropyle; R = nm. ventral surfaces of the coxites (Fig. 3e). These basiconica sensilla are 5.5 wm (4-6 um) in length and are situated on a cuticular depression (Fig. 3e). There are pores (ca. | um wide) on both surfaces which are inter- spersed among the basiconic sensilla (Fig. 3e). The sensilla trichodea and basiconica are stained with the silver nitrate. In the behavioral experiments, none of the 18 treated females excavated any ovi- position sites on the stems or laid any eggs whereas the untreated females made ovi- position excavations and deposited eggs. Many of the treated females left the stems after 20 to 30 minutes. DISCUSSION Oviposition behavior of L. mozardi in crimson clover is consistent with that re- ported by Folsom (1909) in red clover and by Wildermuth and Gates (1920) in alfalfa. Eggs are usually laid singly but occasionally 2 or more are deposited in a single puncture. Normally one, and infrequently 2, punc- ture(s) are found per stem. The distribution of oviposition sites may be an important consideration in sampling for clover stem borer egg populations in crimson clover. Most oviposition (54/63 sites) occurs in the stem interval between the 2nd and 4th internodes. Sampling time for egg-infested crimson clover stems could be reduced by limiting samples to stem sec- tions including only the 2nd through 4th internodes. Previously reported hosts of L. mozardi in the genus 7rifolium include only T. pra- tense (Lintner 1881, Folsom 1909) and 7. vesiculosum (Knight et al. 1976). Since ovi- position by L. mozardi occurs in 7 addi- tional Trifolium species (Table 1) these clo- vers also should be considered potential hosts for this insect pending confirmation of their suitability as hosts for larval de- velopment. The excavation for oviposition is similar to that described for L. mozardi by Girault (1907) but he does not mention the rim sur- rounding the opening internally. The egg shape of L. mozardi differs from Languria VOLUME 91, NUMBER | Fig. 3a-e. Ovipositor of L. mozardi. a, Dorsal surface. b, Ventral surface. c, Sensilla at the base of the stylus. d, Sensilla on the apex of the stylus. e, Short sensilla basiconica on the dorsal surface. erythrocephalus Blatchley and Acropteroxys gracilis (Crotch) in that the end opposite the micropyle is more tapered in L. mozardi as compared to a blunt end in the other 2 lan- guriid species (Piper 1978, Genung et al. 1980). The extension from the tapered end of the egg is used to attach the egg to the plant tissue and this structure was not ob- served in other languriid species but is sim- ilar in structure and function to the one found on the egg of Lyctus brunneus (Ste- phens) (Iwata and Nishimoto 1982). Sensilla are situated on the apex of the styli and near the base of the styli in other beetles such as Acanthoscelides obtectus Say (Bruchidae) (Szentesi 1976), Altica lythri Aube (Chrysomelidae) (Phillips 1978), and Thanasimus dubius (F.) (Cleridae) (Baker and Nebeker 1986). There are.differences in the types and numbers of sensilla on the ovipositor of the three previously men- tioned species as compared to those on L. mozardi. Sensilla chaetica, trichoidea, and basiconica are situated on the apex of the styli and near the base of the styli in A. obtectus (Szentesi 1976) and T. dubius (Bak- er and Nebeker 1986) but L. mozardi has no sensilla basiconica in this region. The 14 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON number of sensilla on the apex of the style of L. mozardi (9-10) is similar to A. /ythri (8-11) (Phillips 1978) but differs from A. obtectus (22-24) and T. dubius (23-25) (Szentesi 1976, Baker and Nebeker 1986). The short basiconic sensilla which are sit- uated on the dorsal and ventral surfaces of the ovipositor of L. mozardi are lacking on the ovipositors of the other three beetle species. The behavioral experiment indicates that the sensilla on the ovipositor of L. mozardi are needed for the deposition of an egg in the plant substrate. Sensilla on the OVvipos- itor of other insects are known to respond to chemicals such as salts, water, and amino acids (Rice 1976, 1977, Chadha and Roome 1980, Hood-Henderson 1982, Liscia et al. 1982). The sensilla trichoidea and basiconi- ca on the ovipositor of L. mozardi may re- spond to chemicals in the clover plant which are released during the excavation of an ovi- position site. ACKNOWLEDGMENTS The assistance of Mr. D. G. McMinn, Bi- ological Technician, and Vincent McDaniel in plant production and insect collection is gratefully acknowledged. Contribution of Mississippi Agricultural and Forestry Ex- periment Station in cooperation with USDA, ARS. Journal Article No. 64/3 of the Mississippi Agricultural and Forestry Experiment Station. Research contributes to the Southern Regional Project No. S-201, Role of Legume Cover Crops in Conser- vation Tillage Production Systems. Men- tion of trademark or proprietary products does not constitute a guarantee or warranty of the products by the U.S. Department of Agriculture and does not imply their ap- proval of the exclusion of other products. LITERATURE CITED Baker, G. T. and T. E. Nebeker. 1986. Sensory re- ceptors on the ovipositor of Thanasimus dubius (F.) (Coleoptera: Cleridae). Ann. Soc. Entomol. Fr. 22: 49-52. Chadha, G. and R. Roome. 1980. Oviposition be- havior and the sensilla of the ovipositor of Chilo patellus and Spodoptera littoralis (Lepidoptera). J. Zool., Lond. 192: 169-178. Ellsbury, M. M. and F. M. Davis. 1982. Front- mounted motorcycle net for mass collection of clover insects., J. Econ. Entomol. 75: 251-253. Folsom, J. W. 1909. Insect pests of clover and alfalfa. Ill. Agric. Exp. Sta. Bull. No. 134. 197. Genung, W. G., R. Woodruff, and E. Grissell. 1980. Languria erythrocephalus: Host plants, immature stages, parasites, and habits (Coleoptera: Langu- riidae). Florida Entomol. 63: 206-210. Girault, A. A. 1907. Oviposition of Languria moz- ardi Latreille. Entomol. News 18: 366-367. Hood-Henderson, D. E. 1982. Fine structure and neurophysiology of a gustatory sensillum on the ovipositors of Metasyrphus venablesi and Eu- peodes volucris (Diptera: Syrphidae). Can. J. Zool. 60: 3187-3195. Iwata, R. and K. Nishimoto. 1982. Observations on the external morphology and the surface structure of Lyctus brunneus (Stephens) (Coleoptera, Lyc- tidae) by scanning electron microscopy. Kontyu 50: 10-22. Knight, W. E., O. W. Barnett, L. L. Singleton, and C. M. Smith. 1976. Potential disease and insect problems in arrowleaf clover. Am. Soc. Agron. Southern Branch (Abstract) 3: 7. Mobile, AL, Feb. 1-4, 1976. Lintner, J. A. 1881. The insects of the clover plant. Trans. N.Y. State Agric. Soc. 32: 187-207. Liscia, A., R. Crnjar, A. Angioy, P. Pietra, and J. Stof- folano. 1982. I chemosensilla dell ovopositore in Tabanus nigrovittatus Macq.), Chrysops fuligt- nosus (Wied.) e Rhagoletis pomonella (Walsh.). Boll. Soc. Ital. Biol. Spmt. 58: 1325-1329. Phillips, W. M. 1978. Sensilla types from the OvI- positor of the flea beetle Altica /ythri (Coleoptera: Chrysomelidae). Entomol. Exp. Appl. 24: 399- 400. Piper, G. L. 1978. Life history of Acropteroxys gra- cilis (Coleoptera: Languriidae) on common rag- weed in northeastern Ohio. Ohio J. Sci. 78: 304— 309. Rice, M. J. 1976. Contact chemoreceptors on the ovipositor of Lucilia cuprina (Wied.), the Austra- lian sheep blowfly. Aust. J. Zool. 24: 353-360. 1977. Blowfly ovipositor receptor neurone sensitive to nonovalent cation concentration. Na- ture 268: 747-749. Szentesi, A. 1976. The effect of the amputation of head appendages on the oviposition of the bean weevil, Acanthoscelides obtectus Say (Coleoptera: Bruchidae). Symp. Biol. Hung. 16: 275-281. Wildermuth, V. L. and F. H. Gates. 1920. Clover stem-borer as an alfalfa pest. USDA Bull. No. 889. 25. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 15-21 MORPHOLOGY OF THE MOUTH PARTS AND ANTENNA OF THE LARVA OF THE CLOVER STEM BORER, LANGURIA MOZARDI LATREILLE (COLEOPTERA: LANGURIIDAE) GERALD T. BAKER AND MICHAEL M. ELLSBURY (GTB) Department of Entomology, P.O. Drawer EM, Mississippi State, Mississippi 39762; (MME) USDA-ARS, Crop Science Research Laboratory, Forage Research Unit, P.O. Box 5367, Mississippi State, Mississippi 39762. Abstract.—The morphology of the mouth parts and antennae, including the sensilla on these structures, is described. The mouth parts have several types of sensilla (basiconica, digitiformia, chaetica, trichoidea, campaniformia, and coronal pegs). The apex of each labial (10 sensilla) and maxillary (11 sensilla) palpus has the highest number of sensilla. The striated region on each mandible is composed of several striae and each of these has several rows of blunt, stout pegs. The antennae possess trichoid, basiconic, campaniform, and placoid sensilla and an antennal sensory appendix which is situated on segment 2. Key Words: lLanguriidae, Languria mozardi, larva, mouth parts, antennae, sensilla The clover stem borer, Languria mozardi Latreille, is nearly ubiquitous in North America on a wide range of cultivated and wild host plants (Wildermuth and Gates 1920). Adults occur frequently in the annual clovers Trifolium incarnatum L. (crimson clover), 7. vesiculosum Savi (arrowleaf clo- ver), and 7. alexandrinum L. (berseem clo- ver), grown for forage, soil stabilization, and as cover crops in the Southeast. The clover stem borer causes lodging, re- duced seed set and poor forage quality as a result of removal of stem tissue, loss of fo- liage under stress, and increased fiber con- tent (Wildermuth and Gates 1920). The im- pact of L. mozardi on forage production has probably been underestimated because of the seclusive nature of the adults and the cryptic habits of the larvae. The insect is likely to be of economic importance where clovers are grown for hay or seed produc- tion. Wildermuth and Gates (1920) recom- mended cultural control through crop ro- tation, early cutting for hay production, and sanitation of field margins to reduce over- wintering populations. Selection and breed- ing of clovers for stem characteristics to re- duce establishment and survival of the larva would be an alternative control method. The present studies were initiated to describe the morphology, distribution, and number of sensory receptors on larval mouth parts and antennae of L. mozardi that may mediate feeding or tunneling activity of the larva through contact with the host plant. MATERIALS AND METHODS Specimens of final (Sth) instar larvae of L. mozardi were reared on artificial diet (Rose and McCabe 1973). Use of the diet permitted verification of molting to ensure that Sth instar larvae were examined. The techniques for SEM and staining the spec- imens are given in Baker et al. (1986). All SEM observations are based on 8 last (Sth) instar larvae. 16 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 1-5. to the mandibles. 2. several rows of stout, conical setae. C = campaniform sensillum; L = labrum; M = mandible; m = mola; P = > prostheca; Sc = sensillum chaeticum; St RESULTS The broad, rounded labrum has 24 sen- silla chaetica that are mostly situated on the posterior edge of the labrum where they come in contact with the mandibles (Fig. 1). Just above these sensilla, there are 2 cam- paniform sensilla that are situated near the mid-line of the labrum. The mandibles have a tridentate apex, a distinct mola, a small and pointed prosthe- ca and a striated region near the mola (Figs. Labrum and mandibles of L. mozardi. |. Distal row of sensilla chaetica on the labrum that extend Mandible, apical view. 3. Mandible, ectal surface. 4. Molar region. 5. Stria composed of = Stria; 1, 2, 3 = mandibular dentes. 2, 3, 4). The 2 apical teeth are similar in size while the basal tooth is smaller. A pros- theca is distad of the mola which is deeply grooved and the grooves increase 1n density from the basal to the distal region of the mola. Laterad of the mola is a striated re- gion (at low magnifications) but each stria (at high magnifications) is composed of sev- eral rows of blunt, stout spines (Figs. 3, 5). Two sensilla chaetica and several campan- iform sensilla are situated on the ectal sur- face of the mandibles (Fig. 3). VOLUME 91, NUMBER 1 17 Figs. 6-12. Labial and maxillary regions of the mouth parts. 6. Labium, maxillae, and mandibles. 7. Labial palpus, apex. 8. Maxilla, inner surface. 9. Maxilla, outer surface. 10. Sensilla trichodea at base of maxillary dentes. 11. Digitiform sensillum. 12. Sensilla on apex of maxillary palpus. Cp = coronal peg; DS = digitiform sensillum; La = labium; M = mandible; Mx = maxilla; Sb = sensillum basiconicum (blunt); Sb, = sensillum basiconicum (tapered); Ss = sensillum styloconicum; Str = sensillum trichodeum. 18 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON There is a total of 8 trichoid sensilla on the ligula, mentum, and submentum of the labium (Fig. 6). The labial palpi are 2-seg- mented and the 2nd segment is longer. At the apex of segment 2 there are 10 sensilla (Fig. 7). The central sensory receptor is a styloconic sensillum which has a short base with a long peg (Fig. 7). There are 8 sensilla basiconica, 7 that are blunt and one that has a tapered tip (Fig. 7). These sensilla are lo- cated on the periphery of the terminal palpal segment. There is also one dome shaped sensillum that is situated on a cuticular pro- trusion (Fig. 7). This sensillum is known as a coronal peg (Doane and Klingler 1978). The maxilla is well developed with a 3-segmented palpus and a mala which is long and curved with 2 large dentes at the apex (Figs. 8, 9). Just below the dentes there is a row of large sensilla chaetica on the inner surface of the mala (Fig. 8), and on the outer surface, 4 sensilla trichoidea are located at the base of the dentes (Figs. 9, 10). The maxillary palpi are 3-segmented and the 3rd is the longest. There are 2 sensilla chaetica on each palpal segment and | 1 sen- silla are situated on the apex of segment 3 (Figs. 8, 9, 12). These sensilla are similar to those on the apex of the labial palpus but there are more tapered basiconic sensilla on the maxillary palpus (Fig. 12). A single dig- itiform sensillum 1s located on the inner sur- face of the last palpal segment (Fig. 11). The stipe and cardo are asperate (Fig. 8). There are 2 pouch-like structures that are situated between the submentum of the labium and the cardo of each maxilla (Fig. 6). These structures are also asperate. There are 3 antennal segments and seg- ment 2 is the longest (Fig. 13). Segment 1 is devoid of any hair-like sensory structures but there are 4 campaniform sensilla (Figs. 13, 14). On the apical periphery of segment 2 there are 5 blunt trichoid sensilla (Figs. 14, 15). A single, long sensillum chaeticum is situated on the outer, lateral surface of subsegment 2 (Fig. 13). The antennal sen- sory appendix is on the ventral side of seg- ment 2 near the intersegmental membrane between segments 2 and 3 (Fig. 15). Laterad of this sensillum there are 2 sensilla basi- conica (Fig. 15). Segment 3 has a raised area at the apex on which are situated 3 sensilla, a campan- iform sensillum, and a trichoid and basi- conic peg (Fig. 16). Just proximal of these receptors are 3 trichoid sensilla (Fig. 16). A placoid sensillum is situated on the ventral surface of segment 3 near the antennal sen- sory appendix (Fig. 15). Only the sensilla chaetica and campaniformia that are situ- ated on the mouth parts and antennae do not stain with crystal violet or reduced sil- ver, thus indicating their nonporosity. All other sensilla that are mentioned pick up these stains indicating that these sensilla have a porous cuticular peg. DISCUSSION The close proximity of the labral sensilla chaetica to the underlying mandibles indi- cates that these sensilla probably act in monitoring the position of the labrum and mandibles during feeding. Campaniform sensilla are known to act as stress receptors and the 2 sensilla situated on the middle portion of the labrum may function as cu- ticular stress receptors. In previous descriptions of larval langu- riid mandibles (Comstock and Comstock 1916, Peterson 1951) there is no mention of the structure of the molar surface and the striae that occur laterad the mola. The deep- ly grooved molar surface aids in grinding plant material that is to be ingested. Several rows of blunt spines comprise each of the striae that are present on the mandibles. They are also present on the same region of Tribolium larvae (Kvenborg 1977). The function of the striae is unknown, but they could be involved in stridulation or groom- ing. The 2 nonporous sensilla chaetica and several campaniform sensilla on the ectal surface of each mandible probably function as those located on the labrum, to monitor VOLUME 91, NUMBER | Figs. 13-16. cuticular stress and to relay information on the position of the mouth parts in relation to one another. The general structure of the labium is similar to the descriptions for other langu- riid species (Wildermuth and Gates 1920, Piper 1978). L. mozardi differs from Acro- pteroxys gracilis (Newman) in the number and types of sensilla situated on the apex of the labial palpi. A. gracilis has 11-12 basi- conic sensilla on the apex of the palpus (Pip- er 1978), while L. mozardi has a total of 9- 10 sensilla but there are 2 types of basiconic Larval antenna of L. mozardi. 13. Antenna, dorsal surface. 14. Antenna, ventral surface. 15. Intersegmental region between antennal segments 2 and 3. 16. Antenna, apex. Asa = antennal sensory appendix; C = campaniform sensillum; Ps = placoid sensillum; Sb = sensillum basiconicum; Str = sensillum trichodeum. sensilla, a sensillum styloconicum and a co- ronal peg. A similar situation exists when one com- pares the number and types of sensilla on the maxillary palpal apex of L. mozardi and A. gracilis. L. mozardi has 10-11 sensilla, 1 styloconic and coronal sensillum and 2 types of basiconic sensilla while A. gracilis has 11-12 basiconic sensilla. The total number of apical sensilla on the maxillary (11) and labial (10) palpi of L. mozardi is similar to what 1s found on 7ribolium lar- vae (13 & 12) (Ryan and Behan 1973) and 20 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Hypera larvae (12 & 11) (Bland, 1983). But the various types of apical sensilla on these structures of L. mozardi resemble the apical sensilla on wireworm larvae (Doane and Klingler 1978). The maxillary and labial apical basiconic and styloconic sensilla stain with crystal vi- olet and reduced silver indicating that these sensilla have a porous cuticle. Also, behav- ioral and electrophysiological studies on other coleopterous larvae have shown that similar receptors respond to CO, and other chemicals and therefore may be considered to be chemoreceptors (Klingler 1966, White et al. 1974, Doane and Klingler 1978, Mitchell 1978). The external structure of the digitiform sensillum is similar to what is found on oth- er coleopterous larvae (Zacharuk et al. 1977, Doane and Klingler 1978, Guse and Hono- michl 1980, Honomichl 1980). Ultrastruc- tural and electrophysiological evidence in- dicate that this sensillum is a type of mechanoreceptor that is involved in mon- itoring the larva’s tunneling activity (Za- charuck et al. 1977). The total number of antennal sensilla on L. mozardi (22) is similar to Tenebrio mol- itor L. (24 sensilla) (Pierantoni 1969) and Tribolium larvae (21 sensilla) (Behan and Ryan 1978). The morphology of the basi- conic and porous trichoid sensilla resembles that found on other coleopterous larvae, and ultrastructural studies indicate that these sensilla are chemoreceptors (Behan and Ryan 1978, Bloom et al. 1982a, b). The antennal sensory appendix of L. mozardi is similar in morphology and position on the antenna as that on elaterid larvae (Zacharuk 1962). The fine structure of the sensory ap- pendix suggests that it functions as an ol- factory receptor (Scott and Zacharuk 1971). The ultrastructure of the placoid sensillum on tenebrionid larvae indicates a chemo- sensory function (Behan and Ryan 1978, Bloom et al. 1982), but electrophysiological data is lacking. ACKNOWLEDGMENTS Assistance of Mr. D. G. McMinn, Bio- logical Technician, and Vincent O. Mc- Daniel during these studies is gratefully ac- knowledged. Contribution of Mississippi Agricultural and Forestry Experiment Sta- tion in operation with USDA, ARS. Journal Article No. 64/4 of the Mississippi Agri- cultural and Forestry Experiment Station. Research contributes to Southern Regional Project No. S-201, Role of Legume Cover Crops in Conservation Tillage Production Systems. Mention of trademark or propri- etary products does not constitute a guar- antee or warranty of the products by the U.S. Department of Agriculture and does not imply approval to the exclusion of other products. LITERATURE CITED Baker, G. T., W. L. Parrott, and J. N. Jenkins. 1986. Sensory receptors on the larval maxillae and labia of Heliothis zea and Heliothis virescens (Lepidop- tera: Noctuidae). Int. J. Insect Morphol. Embryol. 15: 227-232. Behan, M. and M. F. Ryan. 1978. Ultrastructure of antennal sensory receptors of Tribolium larvae (Coleoptera: Tenebrionidae). Int. J. Insect Mor- phol. Embryol. 7: 221-236. Bland, R. G. 1983. Sensilla on the antennae, mouth parts, and body of the larva of the alfalfa weevil, Hypera postica (Gyllenhal) (Coleoptera: Curcu- lionidae). Int. J. Insect Morphol. Embryol. 12: 261— 212: Bloom, J. W., R. Y. Zacharuk, and A. E. Holodniuk. 1982a. Ultrastructure of the larval antenna of Te- nebrio molitor L.(Coleoptera: Tenebrionidae): Structure of the trichoid and uniporous peg sen- silla. Can. J. Zool. 60: 1528-1544. 1982b. Ultrastructure of the larval antenna of Tenebrio molitor L. (Coleoptera: Tenebrioni- dae): Structure of the blunt-tipped peg and paillate sensilla. Can. J. Zool. 60: 1545-1556. Comstock, J. H. and A. B. Comstock. 1916. Manual for the Study of Insects. Comstock Publ. Co., Ith- aca, NY, 701 p. Doane, J. F. and J. Klingler. 1978. Location of CO,- receptive sensilla on larvae of the wireworms Agriotes lineatus-obscurus and Limonius califor- nicus. Ann. Entomol. Soc. Am. 71: 357-363. Guse, G. W. and K. Honomichl. 1980. Die digiti- VOLUME 91, NUMBER 1 formen Sensillen auf dem Maxillarpalpus von Co- leoptera II. Feinstrucktur bei Agabus bipustulatus (L.) and Hydrobius fuscipes (L.). Protoplasma 103: 55-68. Honomichl, K. 1980. Die digitiformen Sensillen auf dem Maxillarpalpus von Coleoptera I. Verglei- chendtopographische Untersuchungdes kuticular- en Apparates. Zool. Anz. 204: 1-12. Klingler, J. 1966. Uber den Sitz der CO,-Receptoren bei der Larve von Otiorhynchus sulcatus. Ento- mol. Exp. Appl. 9: 271-277. Kvenborg, J. E. 1977. Scanning electron microscopic study of adult stored product beetle mandibles. J. Assoc. Off. Analyt. Chem. 60: 1185-1209. Mitchell, B. K. 1978. Some aspects of gustation in the larval red turnip beetle, Entomoscelis ameri- cana, related to feeding and host plant selection. Entomol. Exp. Appl. 24: 340-349. Peterson, A. 1951. Larvae of Insects, Part II. Edwards Bros. Inc., Ann Arbor, MI, 416 p. Pierantoni, R. L. 1969. Mechano- and chemo-recep- tors in the antenna of Tenebrio molitor in the larval stage. Proc. Elect. Micros. Soc. Am. 2: 246-247. Piper, G. L. 1978. Life history of Acropteroxys gra- culis (Coleoptera: Languriidae) on common rag- weed in northeastern Ohio. Ohio J. Sci. 78: 304— 309. 21 Rose, R. I. and J. M. McCabe. 1973. Laboratory rearing techniques for the southern corn root- worm. J. Econ. Entomol. 66: 398-400. Ryan, M. F. and M. Behan. 1973. The sensory re- ceptors of Tribolium larvae. Physiol. Zool. 46: 238- 244. Scott, D. A.and R. Y. Zacharuk. 1971. Fine structure of the antennal sensory appendix in the larvae of Ctenicera destructor (Brown) (Elateridae: Coleop- tera). Can. J. Zool. 49: 199-210. White, R. A., U. Paim, and W. D. Seabrook. 1974. Maxillary and labial sites of CO,-sensitive recep- tors of larval Orthosoma brunneum (Forster) (Co- leoptera: Cerambycidae). J. Comp. Physiol. 88: 235-246. Wildermuth, H. L. and F. H. Gates. 1920. Clover stem borer as an alfalfa pest. USDA Bull. No. 889. 25 pp. Zacharuk, R. Y. 1962. Sense organs of the head of larvae of some Elateridae (Coleoptera): their dis- tribution, structure and innervation. J. Morphol. 111: 1-34. Zacharuk, R. Y., P. J. Albert, and F. W. Bellamy. 1977. Ultrastructure and function of digitiform sensilla on the labial palp ofa larval elaterid (Coleoptera). Can. J. Zool. 55: 569-578. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 22-25 NORTH AMERICAN MOTHS DESCRIBED BY L. A. G. BOSC D’ANTIC (LEPIDOPTERA: NOCTUIDAE, PYRALIDAE) Scott E. MILLER AND V1TOR O. BECKER (SEM) Bishop Museum, Box 19000-A, Honolulu, Hawaii 96817; (VOB) Centro de Pesquisa Agropecuaria dos Cerrados, Caixa postal 70-0023, 73300-Planaltina, DF, Brasil. Abstract.—Three species of moths were described from ‘Caroline’ by Bosc [1800]. Crambus adspergillus Bosc was recognized by Franclemont and Todd (1983) as Renia adspergillus (Noctuidae), but the other two names have remained unplaced. We recognize Alucita cereella Bosc as conspecific with Nola sorghiella Riley, n. syn. (Noctuidae) and Pyralis sacculana Bosc as conspecific with C/ydonopteron tecomae Riley, n. syn. (Pyral- idae). Key Words: The frenchman Louis Auguste Guillaume Bosc d’Antic (1759-1828) lived in the United States during the French Revolution and collected insects, especially in South Carolina (Blake 1952; Zimsen 1964: 16-17). His collection, which contained many type specimens including several of those de- scribed by J. C. Fabricius, went to the Mu- s¢um National d’Histoire Naturelle, Paris (MNHN). At least two of the three species described by Bosc [1 800] from North Amer- ica were present in the MNHN around the turn of the 18th century, because they were illustrated by Coquebert (1801). However, no Lepidopteran types from the Bosc col- lection are now extant in the Paris museum (Zimsen 1964, P. Viette pers. comm.). Bosc [1800] described three species of North American moths, but the paper was overlooked for over a century. The part of the journal in which Bosc’s paper appears does not beara date, but is headed **Prairial, an 8 de la République,” meaning it was pub- lished in 1800, the eighth year of the French Revolution (an earlier part bears the head- ing “Germinel, an 6 de la République (Mars 1798)”). It appears that the first use of any nomenclature, Nearctic, moths, Noctuidae, Pyralidae of Bosc’s names is that of Kaye and Lamont (1927), who applied one to a Trinidad species of Chrysauginae (Pyralidae). Of the three species described by Bosc [1800], only one, Crambus adspergillus, has been in- cluded in the North American literature (Franclemont and Todd 1983). Although the types are not available for study, the descriptions and the figures pre- sented by Bosc allow identification of the species. We have also checked the relevant bibliographies for homonymy and found that the three names are nomenclatorally available. Two of the species, Crambus adspergillus and Pyralis sacculana, were illustrated in color by Coquebert de Montbret (1801), but no mention of A/ucita cereella was included. NOcTUIDAE Renia adspergillus (Bosc) Figs. 1, 4 Crambus adspergillus Bosc [1800]: 114, fig. 2 [Type locality: United States, “Caro- line”’].—Coquebert 1801: 72, pl. XVII, fig. 1OA-B. VOLUME 91, NUMBER | Figs. 1-3. Alucita cereella, Fig. 3: Pyralis sacculana. Figs. 4-6. cereella (Bosc); Fig. 6: Clydonopteron sacculana (Bosc). Figs. 4-6 approximately twice natural size. Renia larvalis Grote 1872: 26 [Type local- ity: United States, Pennsylvania and Tex- as] [Synonymy by Franclemont and Todd 1983] Renia restrictalis Grote 1872: 26 [Type lo- cality: United States, Pennsylvania] [Syn- onymy by Franclemont and Todd 1983] Renia adspergillus (Bosc).—Franclemont and Todd 1983. Of the thirteen species of Renia recog- nized in North America, the figure resem- bles two species, previously described as fla- vipunctalis (Geyer) and /arvalis (Grote). We follow Franclemont and Todd (1983: 121), Reproductions of illustrations from Bosc ({1800]: pl. VII), Fig. 1: Crambus adspergillus, Fig. 2: Figures of specimens matching Bosc’s illustrations: Fig. 4: Renia adspergillus (Bosc); Fig. 5: Nola who know the group well and were the first revisers, in recognizing adspergillus as con- specific with /arvalis. Nola cereella (Bosc), New ComMBINATION Figs. 2, 5 Alucita cereella Bosc [1800]: 115, fig. 4 [Type locality: United States, “Caroline’’] Nola sorghiella Riley 1882: 187, pl. XI, fig. 1 [Type locality: United States, Alabama and Florida], n. syn. Nola portoricensis Moeschler 1890: [Type locality: Puerto Rico], n. syn. 118 24 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Because of the small size of the figure and the quality of printing of the plate, this was the most difficult of the three Bosc species to identify. The pattern shown in the plate resembles species belonging to various fam- ilies, and the plate shows two protruding appendages in front of the head that could be interpreted as either palpi or front legs. The shape and pattern resemble those of some species of Donacaula Meyrick (Pyr- alidae), especially the species belonging to the melinella group. The figure also resem- bles the pattern of some species of Ethmia Huebner (Oecophoridae), especially F. tri- furcella (Chambers). We have two main reasons for consid- ering cereella as conspecific with sorghiella. First, Bosc stated that it is a serious pest of grain. Secondly, although his description of the damage seems to be confused with that caused by the Hessian fly (Mayetiola de- structor (Say), Diptera: Cecidomyiidae), and Sitotroga cerealella (Olivier) (Lepidoptera: Gelechiidae), no other known grain pest re- sembles the figure of cereella. Bosc’s original description follows: “Elle est cendrée. Les bords des ailes sont brans et paroissent renflés. On voit sur le disque de chaque aile supérieure deux points bruns. Les ailes posterieures sont d’un gris brillant. “Cette Alucite est la teigne qui, sous le nom d’Hessian fly, a fait, il ya environ douze ans, de grandes ravages dans les bleds d’Ameérique, et a menace de les étendre sur ceux d’Europe, ce qui a oblige le parlement d’Angleterre de proscrire les bleds qui veno- ient de ce pays. Sa larve a seize pattes; elle est d’un blanc verdatre. Elle mange d’abord le germe du grain, puis la farine, ne laissant que l’écorce. I] n’y a qu’une larve dans cha- que grain: elle est trés-feconde et les gén- érations se succédent rapidement. Au défaut de bled, elle attaque le mais, et se multiplie considérablement dans les greniers ot l’on conserve ce grain. “Le meilleur moyen pour détruire la larva de cette Alucite, c’est de faire passer les grains dans une étuve trés-chaude.” PYRALIDAE Clydonopteron sacculana (Bosc), New ComMBINATION Figs. 3, 6 Pyralis sacculana Bosc [1800]: 115, fig. 3A, 3B [Type locality: United States, ““Caro- line”’].—Coquebert 1801: 71, pl. X VII, fig. TA-C. Clydonopteron tecomae Riley 1880: 286, figs. 152-153 [Type locality: United States, probably Missouri], n. syn. There is no doubt that sacculana is the same species described by Riley (1880) as tecomae. The peculiar shape of the wings and palpi, combined with the male struc- tures on the costa of the forewing makes this species readily recognized. Kaye and Lamont (1927) and Amsel (1956) identified specimens of Chrysauginae from Trinidad and Venezuela as sacculana and placed it in the genus Salobrena Walker. Cashatt (1969) and Munroe (1983: 79), however, regard Salobrena and Clydonopteron Riley as distinct genera. Following their arrange- ment, we decided to place sacculana in Cly- donopteron. Specimens that are similar to the North American sacculana have been collected throughout the neotropics to southern Brazil, and have been identified as tecomae (Hampson 1897: 649, Lima 1950: 20). However, a full revision of the group is needed to establish the number of species involved, their distribution, synonymy, and generic assignments. Also, we are not the first to recognize the synonymy of sacculana with tecomae. In the USNM collection there isa label, added by Carl Heinrich, proposing this synonymy, based on Coquebert (1801), but Heinrich never published his discovery. ACKNOWLEDGMENTS We thank the Museum of Comparative Zoology Library, Harvard University, for providing copies of Bosc’s illustrations, and the lepidopterists of the National Museum of Natural History and Systematic Ento- mology Laboratory, U.S. Department of VOLUME 91, NUMBER 1 Agriculture (especially R. W. Hodges and D. C. Ferguson), for their assistance in ver- ifying the identities of Bosc’s species. R. Stanger translated Bosc’s paper for us. The photographs of specimens were taken by V. Krantz of the Smithsonian Institution. M. J. Scoble helped verify the date of Bosc [1800]. Research for this paper was con- ducted while both authors had Smithsonian Institution fellowships. N. L. Evenhuis, D. C. Ferguson, and R. W. Hodges reviewed the manuscript. LITERATURE CITED Amsel, H. G. 1956. Microlepidoptera Venezolana. Bol. Entomol. Venez. 10: 1-336, pl. I-CX. Blake, D. H. 1952. American Chrysomelidae in the Bosc collection (Coleoptera). Proc. Entomol. Soc. Wash. 54: 57-68. Bosc d’Antic, L. A. G. [1800]. Description de trois espécies de Lépidoptéres de la Caroline. Bulletin des Sciences, par la Societe Philomathique, Paris [series 1] 2(39): 114-115, pl. VII. [German trans- lation in Archiv fiir zoologie und zootomie 3(1): 188-190, 1802] Cashatt, E. D. 1969. Revision of the Chrysauginae of North America (Lepidoptera: Pyralidae) [ab- stract only]. Diss. Abstr. (B) 29: 4696. Coquebert de Montbert, A. J. 1801. Illustratio icono- graphica insectorum quae in Musaeis Parisinis ob- servavit et in lucem edidit Joh. Christ. Fabricius, praemissis ejusdem descriptionibus; accedunt 25 species plurimae, vel minus aut nondum cognitae. Volume II, pp. [45]-90, pl. XI-XX. Franclemont, J. G. and E. L. Todd. 1983. Noctuidae, pp. 120-159. In Hodges, R. W., ed., Check List of the Lepidoptera of America north of Mexico. London, E. W. Classey Ltd. xxiv + 284 pp. Grote, A. R. 1872. Descriptions of North American Noctuidae.—No. |. Trans. Amer. Entomol. Soc. 4: 20-28. Hampson, G. F. 1897. On the classification of the Chrysauginae, a subfamily of moths of the family Pyralidae. Proc. Zool. Soc. London 1897: 633- 692, 74 figs. Kaye, W. J. and N. Lamont. 1927. A catalogue of the Trinidad Lepidoptera Heterocera (moths). Mem. Dept. Agr., Trinidad and Tobago 3: vii + 144 + xv pp. Lima, A. M. da Costa. 1950. Insetos do Brasil. 6.° tomo. Lepidopteros. 2a parte. Rio de Janeiro, Es- cola Nacional de Agronomia. 420 pp., 331 figs. Moeschler, H. B. 1890. Die Lepidopteren-Fauna der Insel Portorico. Abh. Senkenb. Naturforsch. Ges. 16: 77-360, 1 pl. Munroe, E.G. 1983. Pyralidae, pp. 67-85. Jn Hodges, R. W., ed., Check List of the Lepidoptera of Amer- ica North of Mexico. London, E. W. Classey Ltd. xxiv + 284 pp. Riley, C. V. 1880. Ona new pyralid infesting the seed pods of the trumpet vine. C/ydonopteron tecomae, nov. gen., n. sp. Amer. Entomol. 3: 286-288. 1882. Report of the entomologist, in Report of the Commissioner of Agriculture for the years 1881 and 1882. Washington, U.S. Government Printing Office. Zimsen, E. 1964. The type material of I.C. Fabricius. Copenhagen, Munksgaard. 656 pp. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 26-28 THE INS AND OUTS OF A TROPICAL SOCIAL WASP NEST CHRISTOPHER K. STARR Department of Entomology, National Museum of Natural History, Smithsonian Insti- tution, Washington, D.C. 20560 (Present address: Department of Horticulture, University of Georgia, Athens, Georgia 30602). Abstract. —An extraordinarily large nest of the social wasp Synoeca septentrionalis is recorded from Costa Rica. The wasps had formed two holes through which they could enter or leave the nest, unlike the usual single-hole condition in the genus. A statistical test shows that one nest-hole was preferentially used for entering and the other for leaving, which presumably increased the colony’s traffic efficiency. It is postulated that this ten- dency was a statistical result of small individual behavior differences, so that it need not require any organizing mechanism at the colony level. Key Words: Vespine and swarm-founding polistine wasps (Hymenoptera: Vespidae), except for the small genus Apoica, characteristically nest either in secure cavities or, More com- monly, construct an envelope around the brood-combs (Jeanne 1975, Wenzel, in press). This allows them to restrict access to the interior of the nest. There have been occasional observations of nests with two or more entrance holes (Chopra 1925, Maidl 1934, Richards and Richards 1951, R. S. Jacobson, pers. comm.; pers. obs.), and Réaumur (1722) intimated that it is usual for Vespula sp. (probably V. germanica and/ or vulgaris) nests to have two holes. It now seems certain, though, that all social wasps and bees with regular envelopes character- istically have one entrance hole. This is ap- parently taken for granted, although the nests of ants and termites often have many such holes. Synoeca_ septentrionalis Richards is a widespread, conspicuous, swarm-founding polistine of Central and South America (Richards 1978). Its nest begins as a single comb flat against a tree trunk or limb or nest, Synoeca, Vespidae, wasp other substantial surface, surrounded by a domed envelope with prominent ridges run- ning across it and typically with a round hole at the high end. As the colony grows, it may add a new lobe to the nest at the high end, obscuring the old hole and replacing it with a new one on the new lobe. I have seen occasional nests with two or three lobes in Costa Rica; Rau (1933) mentioned a five- lobed nest in Panama, and Buysson (1906) figured one from Mexico with six lobes, ap- parently the largest nest reported for the ge- nus. In early 1979, on a large fig tree (Ficus sp.) near the headquarters of the Santa Rosa National Park of Costa Rica, I found an active S. septentrionalis nest with two re- markable features. First, it consisted of nine lobes with a total length of about 3 meters (Fig. 1). Second, it had paired holes, each of the usual form and in the usual position (Fig. 2), with a combined perimeter of about 10 to 12 cm. Réaumur (1722) reported that Vespula workers consistently enter through one of the two nest-holes and leave through the VOLUME 91, NUMBER | Fig. 1. Nine-lobed nest of Synoeca septentrionalis in Santa Rosa National Park, Costa Rica. It is on the underside of a large branch at an angle of about 60° from horizontal. Fig. 2. The uppermost lobes of the nest, showing the two entrance holes. Wasps on the envelope are about 20 mm long. 27 other. Although the quantitative study of behavior was nearly unknown in his time, we know that Réaumur made original ob- servations on Vespula colonies, and it is rea- sonable to suppose that he found at least one nest with two holes and watched it long enough to gain an impression of directed traffic. Indeed, if a colony has considerable trafhe through a bottleneck, it makes bio- logical sense that any separation of the in- ward and outward streams will increase the efficiency of passage. Accordingly, I pre- dicted that the very large Santa Rosa Sy- noeca colony would preferentially treat one of its two nest-holes as an entrance and the other as an exit. Traffic at the nest was usually so heavy that I could not simultaneously monitor the direction of movement through both holes. In each of eight observations periods, I se- quentially recorded a) the hole-choice of 50 wasps without regard to whether they en- tered or left the nest, b) direction of passage of 25 wasps at one hole, and c) direction of 25 wasps at the other hole. The observation periods were at various times when the wasps were active over the course of 10 days and totaled about two hours. The hole-choice results (a) show greater use of the right-hand hole, which was the site of 280 (70%) of the 400 movements. The direction-of-passage data (b and c) show apparently more wasps leaving than enter- ing the nest (Table 1). If the hole-choice bias is entered as a correction factor (e.g. by mul- tiplying the left-hole figures by ¥ or the right- hole figures by % in Table 1), though, the overall inward and outward traffic during the observation time is found to be almost Table 1. Cumulative numbers of wasps leaving and entering the nest out of 25 moving through each hole during each of eight observations periods. Left Hole Right Hole Total Entering 64 114 178 Leaving 136 86 222 Total 200 200 28 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 2. Percentage breakdown of total traffic dur- ing the observation time. Based on Table | and the hole-choice results, as explained in text. Left Hole Right Hole Total Entering 9.6 39.9 49.5 Leaving 20.4 30.1 50.5 Total 30.0 70.0 100.0 identical. Table 2 thus gives the percentage breakdown of the traffic which would pre- sumably have been recorded if I had been able to monitor all passages at once. The bias-corrected results show a signif- icant difference between the two holes in the directional distribution of traffic (Chi- square, P < 0.01). As seen in Table 2, it is equivalent to what would be recorded if 20% of the wasps consistently leave the nest by the left hole and return by the right hole, while the other 80% each tend to leave and return by the same hole. There is some indication in Synoeca spp. ofan unusual flexibility in nest construction (Vecht 1967, Overal 1982), which may ac- count for this nest’s two-holed condition. It makes little difference here whether this originated and was maintained as an adap- tive response to extraordinarily large col- ony-size or through ordinary building errors. On the other hand, the origin of the ten- dency toward a plausibly more efficient flow of traffic is pertinent. There is no evidence for a higher organizing mechanism in Sy- noeca which could account for this, and we need not postulate any. The most parsi- monious hypothesis is that the small be- havioral differences which unavoidably arise between individuals have been summed to produce a meaningful phenomenon at the colony level. ACKNOWLEDGMENTS My stay in Costa Rica was supported by a graduate assistantship from the Depart- ment of Entomology, University of Geor- gia. A grant-in-aid from Sigma Xi provided research supplies. I am grateful to Jorge Morales and other staffof the National Parks Service of Costa Rica for making it con- venient and pleasant to work in Santa Rosa. Criticism from Bob Jeanne, John Wenzel and the journal’s reviewers led to substan- tial improvements in the paper. LITERATURE CITED Buysson, R. Du. 1906. Monographie des vespides appartenant aux genres Apoica et Synoeca. Ann. Soc. Entomol. France 75: 333-362. Chopra, B. 1925. Notes on a nest of the common Indian hornet, Vespa cincta Fabr. J. Bombay Nat. Hist. Soc. 30: 858-860. Jeanne, R. L. 1975. The adaptiveness of social wasp nest architecture. Quart. Rev. Biol. 50: 267-287. Maidl, F. 1934. Die Lebensgewohnheiten und In- stinkte der staatenbildenden Insekten. Vienna: Fritz Wagner 823 pp. Overal, W. L. 1982. Acoustical behavior and variable nest architecture in Synoeca virginea (Hymenop- tera, Vespidae). J. Georgia Entomol. Soc. 17: 1-4. Rau, P. 1933. Jungle Bees and Wasps of Barro Col- orado Island. Kirkwood, Missouri, Publ. by au- thor 324 pp. Reaumur, R. A. F. de. 1722. Histoire des guépes. Mem. Acad. R. Sci. Paris 21: 302-364. Richards, O. W. 1978. The Social Wasps of the Americas, Excluding the Vespinae. London, Brit- ish Museum (Natural History) 580 pp. Richards, O. W. and M. J. Richards. 1951. Obser- vations of the social wasps of South America (Hy- menoptera, Vespidae). Trans. Entomol. Soc. Lon- don 102: 1-170. Vecht, J. van der. 1967. Bouwproblemen van sociale wespen. Verh. K. Ned. Akad. Wetensch. (Afd. Na- tuurkunde) 76: 59-68. Wenzel, J. W. In press. Evolution of nest architecture in social vespids. /n K. G. Ross, and R. W. Mat- thews, eds., The Social Biology of Wasps. Ithaca, Cornell Univ. Press. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 29-34 NOTES ON THE GENUS HYBRIZON IN NORTH AMERICA (HYMENOPTERA: PAXYLOMMATIDAE) PAUL M. MARSH Systematic Entomology Laboratory, U.S. Department of Agriculture, Agricultural Re- search Service, % U.S. National Museum of Natural History, NHB 168, Washington, D.C. 20560. Abstract.—The two North American species of the unusual genus Hybrizon Fallén are redescribed from a large collection of specimens made in Virginia. Brief comments are made on the taxonomic placement of the genus and on the observed sex ratio of the collected material. Key Words: The genus Hybhrizon Fallen is one of the most peculiar and taxonomically confusing groups in the Ichneumonoidae. It contains seven species (five Palearctic, two Nearctic) and is the only member in the family Paxy- lommatidae except for one undescribed ge- nus from Japan. Because it lacks a second recurrent vein in the fore wing, the genus has often been classified as a subfamily of the Braconidae (Wesmael 1835, Curtis 1837, Haliday 1840, Muesebeck and Walkley 1951, Marsh 1963, Shenefelt 1969, van Achterberg 1976, Watanabe 1984). How- ever, it also has been classified as a subfam- ily of the Ichneumonidae (Rasnitsyn 1980, Gauld 1984), or in a distinct family (Wa- tanabe 1946, Tobias 1968, Marsh 1971, 1979, Mason 1981, van Achterberg 1984, Marsh et al. 1987). Mason (1981) argued convincingly that Hybrizon should be ex- cluded from the Braconidae because it lacks a critical synapomorphy of the family, namely, the fusion of abdominal terga 2 and 3. Furthermore, van Achterberg (1984) gave two synapomorphies of wing venation that show the Paxylommatidae are more closely related to the Ichneumonidae than to the Braconidae. The same conclusion was taxonomy, Ichneumonoidea, ant-parasites reached by Sharkey and Wahl (1987), who suggested that Hybrizon might be placed within the Ichneumonidae. This action had already been proposed by Rasnitsyn (1980) who classified Hybrizon as a subfamily of the Ichneumonidae. However, Mason (1981) argued against this in favor of a sep- arate family classification, the Paxylom- matidae, and I have followed his classifi- cation in this paper. During the summers of 1986 and 1987, my colleague, David R. Smith, operated several Malaise traps in two locations in Virginia, at his home in Annandale (a sub- urb of Washington, D.C.) and near Cuckoo in Louisa County. Approximately 200 spec- imens of Hybrizon were collected during these two years representing two species. Prior to this the U.S. National Museum contained only about 50 specimens of the genus. Approximately 7 of the specimens collected by Smith are ri/eyvi (Ashmead); the other '/ are a distinct species which I thought was undescribed but now have identified as the previously unknown female of /lavo- cinctus (Ashmead). I have provided descrip- tions and a key to separate the species be- low. Additional specimens were borrowed 30 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON from the Canadian National Collection, Ot- tawa, Canada (M. Sharkey), the American Entomological Institute, Gainesville, Flor- ida (H. Townes), the Museum of Compar- ative Zoology, Harvard University, Cam- bridge, Massachusetts (S. Shaw), and the Riksmuseum van Natuurlike Histoire, Leiden, The Netherlands (C. van Achter- berg). The biology of these unusual wasps has not been satisfactorily established. They are associated with ant nests and are likely to be endoparasitoids of ant larvae. Donis- thorpe and Wilkinson (1930) give the most extensive review of the biology. Of interest is the high ratio of females to males of the North American species in the National Collection, 241:14 in ri/eyi and 137:5 in flavocinctus. Female biased sex ra- tios are predicted by the local mate com- petition (LMC) model of Hamilton (1967), . where females place offspring in dis- crete patches of the resource (in this case, ant colonies), and those offspring mate ran- domly in their patch before female offspring disperse to colonize new patches” (Waage 1985). On the other hand, the observed fe- male biased sex ratio could merely be an artifact of collecting techniques, assuming that mating occurs in or near ant colonies and females searching for new colonies are the main dispersers. Thus, random sweep- ing or flight intercept traps would produce mostly females, whereas collections made in ant colonies might yield a more balanced sex ratio. Family Paxylommatidae Pachylommatoidae Foerster, 1862: 247. Oldest family-group name (see Mason 1981 for discussion). Hybrizon Fallen Hybrizon Fallén, 1813, p. 19. No species. Type-species: Hybrizon latebricola Nees, 1834. Monotypic, first included species by Nees (1834:28). Paxylomma de Brébisson, 1825: 23. Type- species: Paxylomma buccatum de Bréb- bison. Monotypic. Synonymy by Wes- mael, 1835. Plancus Curtis, 1833: 188. Type-species: Plancus apicalis Curtis. Monotypic. Syn- onymy by Stephens, 1835. Eurypterna Foerster, 1862: 247. Type- species: Paxylomma cremieri Romand. Monotypic. Synonymy by Marshall, 1891. Eupachylomma Ashmead, 1894: 58. Type- species: Wesmaelia rileyi Ashmead. Orig- inal designation. Synonymy by Watana- be, 1935. Ogkosoma Haupt, 1913: 52. Type-species: Ogkosoma schwarzi Haupt. Monotypic. Synonymy by Strand, 1914. The names Paxyloma (Stephens 1835), Paxylomme (Wesmael 1835), Paxyllomma (Curtis 1837), Paxylloma (Blanchard 1840), and Pachylomma (Ratzeburg 1848) are all to be considered emendations of Paxylom- ma (see Shenefelt 1969 and Mason 1981). Because of the small size of these wasps and the lack of a second recurrent vein in the fore wing, Hybrizon will key to Braconi- dae in most general textbooks with keys to Hymenoptera families. In view of this, 7y- brizon was included by Marsh et al. (1987) in their identification manual for North American genera of Braconidae. The genus can be diagnosed by reference to couplet | of that key and the associated figures. Adult Hybrizon have a distinctive habitus (Fig. 3): narrow head with bulging eyes and deep an- terior tentorial pits (Figs. 6, 7), strongly arched thorax, long spindly legs, and long thin abdomen. The two North American species of Hy- brizon can be separated by the following key. Ocelli small, ocell-ocular distance at least equal to diameter of lateral ocellus, often greater (Fig. 9); first segment of radius in fore wing shorter than first segment of discoideus and about ' length of recurrent vein, branchial cell not as tall as disco- cubital cell (Fig. 1); head, thorax and abdomen usually entirely black rileyi (Ashmead) Ocelli larger, ocell-ocular distance less than diam- eter of lateral ocellus, often less than half (Fig. 8); VOLUME 91, NUMBER | disco-cubital Figs. 1, 2. first segment of radius equal to or longer than dis- coideus and about *% length of recurrent vein; bra- chial cell equal in height to disco-cubital cell (Fig. 2); head black, at least pronotum, mesopleuron and base of abdominal terga 3 and 4 honey yellow, sometimes thorax and abdomen extensively marked with honey yellow . . flavocinctus (Ashmead) Hybrizon rileyi (Ashmead) Figs; 3,4:.7; 9 Wesmaelia rileyi Ashmead, 1899: 641. Ho- lotype female in U.S. National Museum, Washington, D.C. Female. Length of body, 2-3 mm. Color: head black, clypeus and mouthparts white; antennal scape and pedicel yellow, flagellum black; thorax black or dark brown, rarely deep honey yellow; legs yellow with hind femur, tibia, and coxa often light brown; tegula yellow; abdomen black or dark brown, rarely basal segments dark honey yellow. Head: very weakly reticulate, smooth and shining; ocellar-ocular distance equal to or greater than diameter of lateral ocellus (Fig. 9): clypeus lengthened, apical margin well below level of lower eye margin, malar space slanted (Fig. 7); antenna with 11 flagello- meres. Thorax: pro and mesothorax smooth and shining; propodeum irregularly rugose, without any indication of median longitu- dinal carina. Abdomen: terga smooth and shining, terga 1 and 2 sometimes weakly striate at base (Fig. 4). Wings (Fig. 1): first segment of radius shorter than first segment 31 Wings of Hybrizon species. |, H. rileyi(Ashmead). 2, H. flavocinctus (Ashmead) (scale = 0.5 mm). of discoideus and about '2 length of recur- rent vein, brachial cell not as tall as disco- cubital cell. Male. Essentially similar to female. Type locality. UNITED STATES: Ox- ford, Indiana. Material examined. 241 99, 14 4¢ from the following states and provinces: District of Columbia, Georgia, Indiana, Iowa, Kan- sas, Maine, Maryland, Michigan, New Hampshire, New Jersey, New York, North Carolina, Nova Scotia, Ontario, Pennsyl- vania, Quebec, South Carolina, Virginia, West Virginia, Wisconsin. Biology. The type material is recorded as being reared from Toxopotera (= Schiza- phis) graminum, but this is probably not correct. Three specimens from New Hamp- shire are labelled “Attracted to disturbed nest of Lasius alienus.” This species is easily distinguished from flavocinctus by its darker color, smaller ocelli, and wing venation. Hybrizon flavocinctus (Ashmead) Figs. 2, 5, 6, 8 Eupachylomma flavocincta Ashmead, 1894: 59. Holotype female in U.S. National Museum, Washington, D.C. Female. Length of body, 3.5—-4 mm. Col- or: head black, clypeus and mouth parts light yellow; antennal scape and pedicel yellow, flagellum black; prothorax honey yellow; 32 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Mj a tt Hybrizon species. 3, H. rileyi (Ashmead), habitus (scale = 2 mm). 4, H. rileyi, abdominal terga 1-2 (scale = 500 u). 5, H. flavocinctus (Ashmead), abdominal terga 1-2 (scale = 500 yu). 6, H. flavocinctus, face Figs. 3-9. (scale = 100 uw). 7, H. rileyi, face (scale (scale = 200 x). mesonotum dark brown or black, some- times with yellow longitudinal lines; scu- tellum yellow with brown spot at base: mesopleuron varying from entirely brown to yellow; propodeum dark brown; tegula yellow; legs yellow, hind femur, tibia, and coxa light brown; abdomen brown, terga 3 and 4 yellow at base. Head: reticulate and 200 uw). 8, H. flavocinctus, vertex (scale = 200 yu). 9, H. rileyi, vertex dull; ocell-ocular distance less than diam- eter of lateral ocellus, inner edge of each ocellus margined by a scrobiculate groove (Fig. 8); clypeus short, apical margin only slightly below level of lower eye margin, malar space nearly horizontal, eyes bulging below (Fig. 6); antenna with 11 flagello- meres. Thorax: pro- and mesothorax smooth VOLUME 91, NUMBER 1 and shining; propodeum irregularly rugose, often with a short median carina. Abdomen: first and second terga usually distinctly striate (Fig. 5), rest of terga smooth and shining. Wings (Fig. 2): first segment of ra- dius equal to or longer than discoideus and about *% length of recurrent vein, brachial cell about as tall as discocubital cell. Male. Essentially as in female, occasion- ally body mostly honey yellow. Type locality. UNITED STATES: Wash- ington, D.C. Material examined. 137 99, 5 64 from the following states and provinces: District of Columbia, Maryland, Michigan, New York, Ontario, Virginia, Wisconsin. Biology. Unknown. Prior to this study, the only authentically determined specimen of flavocinctus was the male holotype. The large number of female specimens collected in Virginia were gen- erally much darker in color than the holo- type and I had thought them to be an un- described species. After closer examination, they agree morphologically with the holo- type and I now consider them to be the undescribed female of flavocinctus. This species differs from ri/ey/ in its larger size, larger ocelli, generally lighter body col- or, and wing venation. It is also very similar to the European buccatus (de Brébisson) which is distinguished by its darker body color, by having stronger sculpturing on the head which is almost punctate, and by hav- ing a few punctures on the mesonotum along where the notauli would be. ACKNOWLEDGMENTS Scott Shaw, Museum of Comparative Zo- ology, Harvard University, read the manu- script and offered many helpful suggestions for improvement, especially the sections on classification and biology, for which I am grateful. David Wahl, American Entomo- logical Institute, Gainesville, Florida and Robert Smiley, Systematic Entomology Laboratory, Beltsville, Maryland also of- fered useful comments on the manuscript. 33 Thanks also go to David Smith for oper- ating the Malaise traps that produced the specimens which motivated this study. LITERATURE CITED van Achterberg, C. 1976. Hybrizontinae or Hybri- zontidae? (Hymenoptera, Ichneumonoidea). Entomol. Berich. 36: 61-64. 1984. Essay on the phylogeny of Braconidae (Hymenoptera: Ichneumonoidea). Entomol. Tidskr. 105: 41-58. Ashmead, W. H. 1889 (1888). Descriptions of new Braconidae in the collection of the U.S. National Museum. Proc. U.S. Natl. Mus. 11: 611-671. 1894. Notes on the family Pachylommato- idae. Proc. Entomol. Soc. Wash. 3: 55-60. Blanchard, C. E. 1840. Histoire nataurelle des ani- maux articulés. 3, Hyménoptéres. Paris, P. Du- meénil. 672 pp. de Brébisson, L. A. 1825. Paxylomme, Paxylomma. p. 23 in Lepeletier and Serville, Encycl. Meth. Paris, Pankouke. Ins. vol. 10, 833 pp. Curtis, J. 1833. Characters of some undescribed gen- era and species, indicated in the “Guide to an Arrangement of British Insects.” Entomol. Mag. 1: 186-199. 1837. A guide to an arrangement of British insects. 2nd Edition. London, Westley. vi. + 294 pp. Donisthorpe, H. St. J. K. and D. S. Wilkinson. 1930. Notes on the genus Paxylomma (Hym. Brac.), with the description of a new species taken in Britain. Trans. Entomol. Soc. London 78: 87-93. Fallen, C. F. 1813. Spec. Nov. Hym. Disp. Meth., p. 19. Foerster, A. 1862. Synopsis der Familien und Gat- tungen der Braconiden. Verh. Naturh. Ver. Preuss. Rheinl. 19: 225-288. Gauld, I. D. 1984. An introduction to the Ichneu- monidae of Australia. Brit. Mus. (Nat. Hist.) Pub. no. 895, 413 pp. Haliday, A. H. 1840. Braconidae, pp. 61-65. Jn Westwood, J O., ed., An Introduction to the Modern Classification of Insects. Synopsis of the Genera of British Insects. London, Longman. 158 pp. Hamilton, W. D. 1967. Extraordinary sex ratios. Sci- ence 156: 477-488. Haupt, H. 1913. Beitrage zur Hymenopteran-Fauna von Halle a. S. und Ungengend. Mitt. Entomol. Ges. Halle 5-7: 47-62. Marsh, P.M. 1963. A key to the Nearctic subfamilies of the family Braconidae (Hymenoptera). Ann. Entomol. Soc. Amer. 56: 522-527. 1971. Keys to the genera of the families Bra- conidae, Aphidiidae, and Hybrizontidae (Hyme- noptera). Ann. Entomol. Soc. Amer. 64: 841-850. 34 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 1979. Family Braconidae, pp. 144-295. In Krombein, K. V. et al., eds., Catalog of Hyme- nopterain America North of Mexico. Washington, D.C., Smithsonian Institution Press. Marsh, P. M., S. R. Shaw, and R. A. Wharton. 1987. An identification manual for the North American genera of the family Braconidae (Hymenoptera). Mem. Entomol. Soc. Wash. 13, 98 pp. Marshall, T. A. 1891. Les Braconides. Jn Andre, E., Species de Hyménoptéres d’Europe et d’Algerie, vol. 5, 635 pp. Mason, W. R. M. 1981. Paxylommatidae: The cor- rect family-group name for Hybrizon Fallén (Hy- menoptera: Ichneumonoidea), with figures of un- usual antennal sensilla. Can. Entomol. 113: 433- 439. Muesebeck, C. F. W. and L. M. Walkley. 1951. Fam- ily Braconidae, pp. 90-184. Jn Muesebeck, C. F. W. et al., eds., Hymenoptera of American North of Mexico, Synoptic Catalog. U.S. Dept. Agr., Agr. Monog. 2. Nees ab Esenbeck, C. G. 1834. Hymenopterorum Ichneumonibus afinum monographiae, genera Europaea et species illustrantes. Vol. 1, 320 pp. Stuttgart, Tubingen, Cotta. Rasnitsyn, A. 1980. The origin and evolution of Hy- menoptera [in Russian]. Trudy Paleontol. Inst. 174, 190 pp. Ratzeburg, J. T. C. 1848. Die Ichneumonen der For- stinsecten in forstlicher und entomologischer Be- ziehung. Berlin, Nicolai. 2, 238 pp. Sharkey, M. J. and D. Wahl. 1987. Phylogeny of the Ichneumonoidea. Ichnews, no. 10, pp. 4-10. Shenefelt, R. D. 1969. Braconidae 1, Hybrizoninae, Euphorinae, Cosmophorinae, Neoneurinae, Mac- rocentrinae, pp. 1-176. /n Ferriére, Ch. and J. van der Vecht, eds., Hymenopterorum Catalogus (new edition), part 4. The Hague, Junk. Stephens, J. F. 1835. Illustrations of British Ento- mology. Mandibulata 7, 310 pp. London, Bladwin and Cradock. Strand, E. 1914. Bemerkungen tiber Paxylommati- nae. Entomol. Mitt. 3: 27-31. Tobias, V. I. 1968. On the classification and phylog- eny of the family Braconidae, pp. 3-43. Acad. Sci. USSR (Special issue dedicated to memory of H. A. Kholodkovskogo). [In Russian.] Waage, J. K. 1985. Family planning in parasitoids: adaptive patterns of progeny and sex allocation, pp. 63-95. In Waage and Greathead, eds., Insect Parasitoids. London, Academic Press. Watanabe, C. 1935. On two hymenopterous guests of ants in Japan. Insecta Mats. 9: 90-94. 1946. Taxonomic revision of the family Paxylommatidae (Hymenoptera). Matsumushi 1: 24-31. [In Japanese with English summary.] . 1984. Notes on Paxylommatinae with review of Japanese species (Hymenoptera: Braconidae). Kontyu 52: 553-556. Wesmael, C. 1835. Monographie des Braconides de Belgique. Nouv. Mém. Acad. Sci. R. Bruxelles 9, 252 pp. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 35-50 THE GENERA ATAENOGERA AND PHYCUS IN THE NEW WORLD (DIPTERA: THEREVIDAE: PHYCINAE) DONALD W. WEBB AND MICHAEL E. IRWIN Illinois Natural History Survey, 607 East Peabody Drive, Champaign, Illinois 61820. Abstract. —Two of the three species names available for the New World genus Ataenog- era Krober are synonymized in this paper. The genus Phycus Walker, previously unknown from the New World, is represented here by two new species. A key to the New World species of these two genera is provided along with species descriptions and distributions. Descriptions of the pupal stage of Ataenogera abdominalis and of the larva and pupa of Phycus frommeri n. sp. are included. Key Words: distribution In a revisionary monograph on the Nearctic genera of Therevidae, Irwin and Lyneborg (1981) described several new gen- era, redescribed others, and assigned an un- described therevid species from northwest- ern Mexico and southern California to the genus Phycus. They separated the genera Phycus Walker and Ataenogera Krober on the basis of several morphological charac- teristics and suggested that species of Atae- nogera have no close relatives in the Old World other than Phycus. Subsequently, Lyneborg (1983) characterized and critical- ly evaluated the Old World species of Phy- cus. Because Phycus was unknown from the New World prior to Irwin and Lyneborg’s study, we undertook to compare that genus to the closely related Ataenogera. The species in these genera are similar in gross structure and are placed together within the subfam- ily Phycinae. Because Ataenogera and Phy- cus are Clearly separated as a new clade from such New World phycine genera as Heni- comyia Coquillett on the one hand, and Pherocera Cole, Schlingeria Irwin, and Par- apherocera Irwin on the other (Irwin and Diptera, Therevidae, Phycus, Ataenogera, New World, species descriptions, Lyneborg 1981), we elected to describe and diagnose their New World species here in a single publication. METHODS The morphological terminology used for the male terminalia was originally defined and described by Lyneborg (1968) and sub- sequently modified by Lyneborg (1972, 1976, 1978) and Irwin (1977a, b); that for the female terminalia was defined and de- scribed by Irwin (1976); that for the larva and pupa follows Irwin (1972). Other mor- phological features are described in termi- nology generally accepted in the literature on Diptera. The range for each measure- ment and ratio is followed by the average; the range for each setal count is followed by the mode. Each specimen was assigned a THER- EVIDAE/M. E. IRWIN/SPECIMEN num- ber attached to the specimen. This number is used to associate the ecological and label data with a given specimen and is printed in italics in this paper. The data are incor- porated into an automated data manage- 36 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ment system originally designed by Rauch (1970). To conserve space and include as much information as possible about each speci- men, a layout adopted from Irwin (1983) is used in the “Specimens Examined” section of each species. Many of the terms used there are explained by Stuckenberg and Ir- win (1973). The layout follows these typo- graphical conventions: 1) Full capitals: LARGEST POLITICAL UNIT (country, or state within the United States). 2) Boldface type: intermediate political unit (state or province outside the United States or county within the United States) and elevation expressed 1n m above sea level. 3) Roman or normal print: smallest po- litical unit (city or town) and modifier of that unit (distances in km, direction, and subunits of that unit). 4) Collector names. Acronyms were used for the following collectors: M. E. Irwin, MEI; R. M. Worley, RMW;; S. I. Frommer, SIF. 5) Numbers of specimens is followed by the sex designation M for male or F for female. 6) A semicolon terminates one series of specimens and signals the beginning of the next. Data not repeated in a subsequent se- ries are the same as those of preceding se- ries. Depositories.—Paratypes of Phycus frommeri n. sp. have been deposited in the following museums: AMNH, American Museum of Natural History, New York; ANSP, Academy of Natural Sciences, Philadelphia; ASU, Arizona State Univer- sity, Tempe; AMS, Australian Museum, Sydney; BMH, Bernice P. Bishop Museum, Honolulu, Hawaii; BMNH, British Mu- seum of Natural History, London; BYU, Brigham Young University, Provo, Utah; CAS, California Academy of Sciences, San Francisco; CIS, California Insect Survey, University of California, Berkeley; CMNH, Field Museum of Natural History, Chicago; CMP, Carnegie Museum, Pittsburgh; CNC, Canadian National Collection, Ottawa; CSDA, California State Department of Ag- riculture, Sacramento; CSIRO, Common- wealth Scientific Industrial Research Or- ganization, Canberra, Australia; CSU, Colorado State University, Fort Collins; CU, Cornell University, Ithaca; DEI, Deutsches Entomologische Institut, Berlin, East Ger- many; DSIR, Department of Scientific and Industrial Research, Nelson, New Zealand; DZSA, Departamento de Zoologia Agricul- tura, Sao Paulo, Brazil; EEA, Estacion Ex- perimental Agronomica, Universidad de Chile, Maipu; FSCA, Florida State Collec- tion of Arthropods, Gainesville; IAS, Insti- tute of Agricultural Sciences, Tokyo; IE, In- stituto di Entomologia, Bologna, Italy; IML, Instituto Miguel Lillo, Tucuman, Argen- tina; INHS, Illinois Natural History Survey, Champaign; INIA, Instituto Nacional de Investigaciones Agricolas, Chapingo, Mex- ico; IOC, Instituto Oswaldo Cruz, Rio de Janeiro, Brazil; IRSN, Institut Royal des Sciences Naturelle de Belgique, Brussels; ISU, Iowa State University, Ames; ITM, Instituto Technologico y de Estudios Super- iores, Monterrey, Mexico; KSU, Kansas State University, Manhattan; KUF, Kyu- shu University, Fukuoka, Japan; LACM, Natural History Museum of Los Angeles County, Los Angeles; Museum of Compar- ative Zoology, Harvard University, Cam- bridge, Massachusetts; MEI, M. E. Irwin Collection, MHN, Museo de Historia Nat- ural Javier Prado, Lima, Peru; MMB, Mo- ravske Museum, Brno, Czechoslovakia; MNH, Musei Nationalis Hungarici, Buda- pest, Hungary; MNHN, Museum National d’Histoire Naturelle, Paris, France; MSU, Michigan State University, East Lansing; NCSU, North Carolina State University, Raleigh; NMB, Naturhistorisches Museum, Basel, Switzerland; NMP, Natal Museum, Pietermaritzburg, South Africa; NMSU, New Mexico State University, Las Cruces; Nevada State Department of Agriculture, Reno; OSM, Ohio State Museum, Colum- VOLUME 91, NUMBER | bus; OSU, Oregon State University, Cor- vallis,; PAS, Polish Academy of Sciences, Warsaw, Poland; RNHL, Rijkmuseum van Natuurlijke Historie, Leiden, The Nether- lands; SDAP, State Department of Agri- culture, Harrisburg, Pennsylvania; SDCM, San Diego County Museum, San Diego; SJSC, San Jose State University, San Jose, California; SMN, Staatlichen Museum fiir Naturkunde, Stuttgart, West Germany; SWRS, Southwestern Research Station (AMNH), Portal, Arizona; TAM, Texas Ag- ricultural and Mechanical University, Col- lege Station, UA, University of Arizona, Tucson; UAC, University of Alberta, Ed- monton, Alberta, Canada; UBC, University of British Columbia, Vancouver, Canada; UCD, University of California, Davis; UCM, University of Colorado Museum, Boulder; UCR, University of California Riverside; UCVM, Universidad Central de Venezuela, Maracay, Venezuela; UG, Uni- versity of Georgia, Athens; UI, University of Idaho, Moscow; UK, University of Kan- sas, Lawrence; UM, University of Minne- sota, Saint Paul; UMA, University of Mich- igan, Ann Arbor; UNLP, Museo de Ciecias Naturales, Universidad Nacional de La Pla- ta, La Plata, Argentina; USI, University of Southern Illinois, Carbondale; USNM, United States National Museum, Washing- ton, D.C.; USSR, Zoological Institute USSR, Leningrad; USU, Utah State University, Logan; UTA, University of Texas, Austin; UTI, University of Tel Aviv, Israel; UU, University of Utah, Salt Lake City; UW, University of Wisconsin, Madison; UZM, Universitetets Zoologiske Museum, Copen- hagen, Denmark; VNM, Naturhistorisches Museum Vienna, Austria; WSU, Washing- ton State University, Pullman; ZIB, Zool- ogisches Institut, Berlin, West Germany; ZSI, Zoological Survey of India Collection, Calcutta, India. The paratypes of Phycus frontalis n. sp. are deposited in the following museums: AMNH, American Museum of Natural History, New York; INHS, Illinois Natural History Survey, Champaign; 37 USNM, United States National Museum, Washington, D.C. Key TO NEw WoRrLD SPECIES OF ATAENOGERA AND PHYCUS 1. Maxillary palpus one-segmented (Fig. 3); eye margins distinctly divergent from level of ocel- lar tubercle toward genae (Fig. 2); discal cell pointed basally (Fig. 4); sternites 5-8 in males, 5-7 in females (Fig. 10) with numerous lan- ceolate setae ventrally; male genitalia with large, subtriangular hypandnum (Fig. 8) oe sult Sichags eater ans a Ataenogera abdominalis Krober - Maxillary palpus two-segmented (Fig. 18); eye margins divergent slightly from level of ocellar tubercle toward genae (Fig. 17); discal cell trun- cate basally (Fig. 19); sternites lacking lanceo- late setae; male genitalia lacking hypandrium or much reduced. (Phycus Walker) .......... 2 2. Eyes separated dorsally by distance greater than width of ocellar tubercle (Fig. 17); width of frons at level of lateral ocelli greater than 0.5 times width at level of antennal bases; pos- terolateral extensions of gonocoxites short, thick in ventral view (Fig. 23) : . Phycus frommeri n. sp. - Eyes sepamited dorsally only by width of ocellar tubercle (Fig. 35); width of frons at level of lateral ocelli less than 0.5 times width at level of antennal bases; posterolateral extensions of gonocoxites narrow, attenuate in ventral view (Fig. 39) Phycus frontalis n. sp. Superficially, the genus Afaenogera 1s morphologically similar to Phycus, the two genera are however quite distinct. The best distinguishing characteristics are: (a) max- illary palpus two-segmented in Phycus, one- segmented in Ataenogera; (b) fore coxae have 1-2 apical setae in Phycus, 4-6 apical setae in Ataenogera; (c) ventral epandrial sclerite fused anteriorly with epandrium in Phycus, fused laterally with epandrium in Alaenog- era; (d) hypandrium extremely small or ab- sent in Phycus, large in Ataenogera; (e) ven- tral apodeme of aedeagus composed of two long, narrow rods in Phycus, very reduced, appearing absent in Afaenogera. Genus Ataenogera Krober Ataenogera Krober (1914: 31): (1932: 255). Type species: Krober, by monotypy. Malloch A. abdominalis Figs. 1-15. Ataenogera abdominalis.—1. Antenna (5747). 2. Head of male, frontal view (5747). 3. Maxillary palpus, lateral view (5747). 4. Wing, dorsal view (5747). 5. Male tergite 8, epandrium, cerci, and ventral epandrial sclerite, dorsal view (576/). 6. Male ventral epandrial sclerite, ventral view (576/). 7. Male gonocoxites and gonostylus, dorsal view (576/). 8. Male gonocoxites and gonostylus, ventral view (576/). 9. Male aedeagus, lateral view (5761). 10. Female abdomen, terminal segments, lateral view with enlargement of ventral lanceolate setae (5752). 11. Female terminalia, dorsal view (5752). 12. Female terminalia, ventral view (5752). 13. Pupal exuvium lateral view (5764). 14. Pupa exuvium, ventral view (left), dorsal view (right) (5764). 15. Frontal plate (5764). Abbreviations: (AS) antennal sheath; (C) cercus; (DA) dorsal apodeme; (Dp) distiphallus; (DSp) dorsal spiracle; (EA) ejaculatory apodeme; (Epa) epandrium; (F) furca; (Gs) gonostylus; (Gx) gonocoxites; (Hpa) hy- pandrium; (S,) sternite 8; (T,) tergite 8; (T.) tergite 9; (VES) ventral epandrial sclerite. Scale = 0.1 mm, unless otherwise indicated. 38 VOLUME 91, NUMBER | Leptocera Kroéber (1928: 117). Type species: L. gracilis Kréber by monotypy. Preoc- cupied by Olivier (1813: 489). Ziehenia Krober (1929: January: 434). New name for Leptocera Krober. Epileptocera Richards (1929: August: 171). Unjustified new name for Leptocera Krober. Moderate-sized, slender flies. Head.—Frons (Fig. 2) of both sexes at its narrowest as wide as ocellar tubercle, to- mentum brownish gray; setae sparse, dor- solateral to antennal bases and in males in two lateral rows ventral to ocellar tubercle; male eyes dichoptic, eye margin diverges distinctly from level of ocellar tubercle to- wards gena, facets equal, setae absent; para- facial lacking setae; head slightly protrudes at antennal level; facial and genal callus ab- sent; length of antenna |.3-2.0 times head depth; antenna long (Fig. 1); scape cylin- drical; pedicel subtriangular; flagellum elongate, tapered apically, setae absent; style terminal, two-segmented, with minute ter- minal spine; clypeus concave, setae absent; maxillary palpus one-segmented (Fig. 3). Thorax.—np 2-3, sa 1, pa 1, dc 0-1, sc 1. Vittae absent; mesonotal setae moder- ately long, subappressed; prosternum bare in and around central depression; pleural setae elongate, scattered on propleuron and dorsal fourth of anepisternum. Wing (Fig. 4).—Ground color hyaline to pale brown, clouding apparent near apex; veins brown; pterostigma absent; setulae dorsal on R,; subcostal cell and cell r, elongate, open; veins R, and R, subequal; cell r, large, encloses apex of wing, length 1.5—2.6 times width at apex; veins M,, M,, and M, originate sep- arately from apical margin of discal cell; cell m, generally closed, petiolate; discal cell acute basally; posterior cells 5; posterior cubital cell closed with short petiole; m-cu/ r-m subequal. Legs.— Fore coxae moderate- ly long, sparsely haired, bare on posterior surface, with 4-6 stiff setae on anterior sur- face; femora lacking stiff setae; tibiae with several stiff apical setae. 39 Abdomen.—Slender, especially in male, longer than wide, laterally compressed; dor- sum convex, shining fuscous to black; setae sparse, short, sternites 5-8 in males, 5-7 in females (Fig. 10) with numerous lanceolate setae ventrally. Male terminalia (Figs. 5- 9).—Tergite 8 (T;) wide, deeply constricted medially. Sternite 8 large, simple. Epan- drium (Epa), longer in midline than maxi- mum width, posterior margin broadly emarginate; cerci (C) extending posteriorly beyond epandrium and ventral epandrial sclerite; ventral epandrial sclerite, ventral view (VES) setose, attached laterally to epandrium with no membranous attach- ment to aedeagus, posterior margin broadly emarginate. Gonocoxites in ventral view not united ventrally, attached anteriorly by nar- row membrane; hypandrium (Hpa) broad, subtriangular, large. Aedeagus with dorsal apodeme laterally attached by a strongly sclerotized bridge to basal section of para- mere; distiphallus narrow, downwardly di- rected; ventral apodeme lacking; ejacula- tory apodeme (EA) large, apical fourth expanded. Female terminalia (Figs. 11- 12).—Tergite 8 (T;) subrectangular, wider than long, posterior margin broadly con- cave. Tergite 9 (T,) suboval, separated me- dially, with several strong setae along pos- terior margin. Cerci (C) subtriangular, membranous, with scattered setae. Sternite 8 (S,) large, as long as wide, with short me- dian emargination on posterior margin. Sternite 9 greatly modified, invaginated be- neath sternite 8 to form internal furca (F). Immature stages. — Characteristics of pu- pal exuvium (Figs. 13-15) in description of species. Ataenogera abdominalis Krober Ataenogera abdominalis Kroéber (1914: 31): Malloch (1932: 255). Leptocera gracilis Kréver (1928: 118). New synonym. Henicomyia brevicornis Bromley (1934: 361). New synonym. Derivation of name. — abdomen (Latin) = belly; a/is (Latin) = pertaining to. 40 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Diagnosis.—Ataenogera abdominalis is the only recognized species in this genus. The characteristics given in the key and de- scription of the genus separate it from Phy- cus. Examination of the type specimens of Ataenogera abdominalis, Leptocera gracilis, and Henicomyia brevicornis revealed no distinctive characteristics that would justify retaining them as separate species. Some variation was noted in the color pattern of the fore and middle femora but this appears to represent clinal variation. Description of male lectotype (5759).— Body length 8.8 mm. Head.— Head depth 0.84 mm. Ocellar tu- bercle fuscous, subshiny to lightly tomen- tose; ocellar setae black, moderately long, scattered. Eyes fuscous. Frons fuscous, to- mentum light gray, variable, with silver pile lateral to antennal bases, setae (Fig. 2) white, moderately long, dorsolateral to antennal bases and dark brown, short, in lateral row ventral to ocellar tubercle. Antenna (Fig. 1) brown, pruinosity light gray, length of an- tenna 1.9 times head depth; length of scape 0.40 mm, 2.9 times width, 4.0 times length of pedicel, setae dark brown, short, with several stiff setae ventrally; length of pedicel 0.10 mm, 0.7 times width, setae dark brown, in apical band; length of flagellum 1.02 mm, 5.7 times width, 2.6 times length of scape; length of basal stylomere 0.02 mm, length of apical stylomere 0.06 mm. Parafacial narrow, tomentum dense silver. Genal setae white, elongate, scattered. Maxillary palpus (Fig. 3) dark brown, pruinosity light gray, length 0.52 mm, 5.2 times width, setae white to pale yellow, elongate, scattered. Label- lum black, pruinosity light gray; setae fus- cous, short, scattered. Postocular setae white, elongate, abundant ventrally becoming black, short, appressed dorsally. Thorax.—np 2, sa 1, pa 1, dc 0, sc 1. Dark brown in ground color, tomentum dark gray; setae white, moderately long, subappressed. Postpronotal lobe concolorous with thorax; setae white, moderately long, scattered. Pleuron fuscous, tomentum dense gray, anepisternum with ventral three-fourths glossy. Pleural setae white, elongate, scat- tered on propleuron, scattered on dorsal fourth of anepisternum, absent on remain- ing pleural sclerites. Scutellum fuscous, to- mentum dark gray. Postnotum and latero- tergite dark brown, subshiny to pruinosity light gray; setae on laterotergite white, elon- gate, abundant. Wing (Fig. 4).—Length 5.7 mm, 3.8 times width. Membrane hyaline. Pterostigma dark brown, narrow. Halter dark brown. Legs.—Dark brown, tomen- tum dark gray, concolorous. Anterior tu- bercle on hind coxa round, fuscous, apical half pale yellow. Abdomen.—Dark brown, subshiny, ter- gites 2-4 with narrow, white posterior mar- gin; setae black, short, appressed on dark areas, white, moderately long on white pos- terior margin, sternites 5-8 with dark brown lanceolate setae (Fig. 10). Male termi- nalia.—Characteristics given in description of genus. Variation in males (N = 10).— Body length 6.3-8.8, 7.9 mm. Head depth 0.60--0.84, 0.70 mm. The frons varies in having the tomentum entire to the ventral third glossy. Length of antenna |.8—2.2, 2.0 times head depth; length of scape 0.32-0.46, 0.40 mm, 2.9-4.2, 3.6 times width, 3.2-5.3, 4.4 times length of pedicel; length of pedicel 0.08- 0.10, 0.09 mm, 0.7-0.8, 0.8 times width; length of flagellum 0.74—-1.02, 0.84 mm, 4.1- 5.7, 4.9 times width, 1.6-2.6, 2.1 times length of scape; length of basal stylomere 0.02 mm, length of apical stylomere 0.06- 0.10, 0.08 mm. Length of maxillary palpus 0.36-0.52, 0.44 mm, 5.2-7.3, 5.9 times width. Notopleural setae 2-3, 2. Dorsocen- tral setae 0-1, 0. Wing length 4.5-5.7, 5.0 mm, 3.5-3.8, 3.7 times width. A cline exists in the color pattern of the fore and middle femora from dark brown in specimens from Argentina, Paraguay, and southern Brazil to dark yellow in specimens from Mexico. Female.—Similar to male with following exceptions (N = 10). Body length 7.5-8.6, 7.9 mm. Head depth 0.84-0.98, 0.92 mm. VOLUME 91, NUMBER 1 The frons shows variation similar to male. Length of antenna |.3-1.5, 1.4 times head depth; length of scape 0.44—-0.50, 0.47 mm, 3.7-5.0, 4.2 times width, 4.8-6.0, 5.4 times length of pedicel; length of pedicel 0.08- 0.10, 0.08 mm, 0.67-0.83, 0.75 times width; length of flagellum 0.70-0.72, 0.71 mm, 3.9- 4.5, 4.3 times width, 1.4-1.6, 1.5 times length of scape; length of basal stylomere 0.02-0.04, 0.03 mm, length of apical sty- lomere 0.06-0.12, 0.08 mm. Fore and mid- dle femora show variation similar to males. Characteristics of terminalia (Figs. 1 1-12) given in description of genus. Seasonal activity.—In the specimens ex- amined, adults were taken throughout the year with no difference in the collecting pe- riod between Central America and southern South America. Females (25) were collected 2.8 times more often than males (9). Pupa (Figs. 13-15).—Length 8.0 mm (N = 1), width 1.5 mm. Alar process not spi- nose. Labial sheath broad (Fig. 15), truncate apically, not bisecting proboscial sheath. Length of antennal sheath 0.56 mm, length of subapical spine 0.08 mm. Thoracic spi- racle tapered apically, apex truncate, length 0.20 mm, 2.0 times width. Posterior spines elongate, not divergent apically, length 0.44 mm. Dorsal spines on abdominal segments I-VII in transverse row on posterior third of segment. Abdominal spiracles on seg- ments I-VII. Type material.—A syntypic series of 3 males of Ataenogera abdominalis Kréber is in the Museum fiir Naturkunde der Humboldt- Universitat zu Berlin. A male of this series, herein designated the lectotype, was col- lected at San Bernardino, Paraguay, in IV, by K. Fiebrig S. V. The holotype female (not male as stated in the description by Bromley 1934) of Henicomyia brevicornis Bromley (American Museum of Natural History) was collected at Kartabo, British Guiana, on May 2, 1924. A syntypic series of one male and one female of Leptocera gracilis Krober is in the Zoologische Institut, Halle. The male of this series, herein designated the lecto- 41 type, was collected at Parana, Brazil, in De- cember. Distribution.—The range of Ataenogera abdominalis, a widespread Neotropical species, extends from southern Mexico to Argentina and Uruguay. Specimens examined (37).—ARGEN- TINA: Tucuman: Amaicha, |6-XI-1966, L. A. Stange, 1M. Mendoza: Potrerillos, 6-I- 1927, F. and M. Edwards, | F. La Rioja: La Rioja, 1928, 1F. Catamarca: 17 km NW Chumbicha, 1143 m, 25-XII-1971, C. A. Pearson, |F. BOLIVIA: Beni: Rio Itenez opposite Costa Marques, Brazil, 1—3-IX- 1964, J. K. Bouseman and J. Lussenkop, 1F; 4-6-IX-1964, 1M; 30-31-VIII-1964, 1F; Rio Itenez, Pampa de Meio, | 1-13-IX-1964, J. K. Bouseman and J. Lussenkop, IF. BRAZIL: Parana: 1912, 1M 1F; Para, Bak- er, 1F; Santa Catarina: Nova Teutonia, 27°8', 52°23’, 16-XII-1947, F. Plaumann, 1M; I-1945, 1F; 27°11’, 52°23’, 18-XII- 1959, F. Plaumann, IF. BRITISH GUIANA: Kartabo, 20-V-1924, 1F. COS- TA RICA: Guanacaste: La Pacifica, 4 km NW Canas [Las Canas], 1-9-IV-1974, P. A. Opler, 1M 4F; 29-III-1974, 1F. EL SAL- VADOR: Quezaltepeque, 2-XI-1977, MEI, 1F. MEXICO: Morelos: 11.7 km S Yaute- pec, 17-VIII-1962, N. L. Marston, 2F; Puebla: 4.8 km NW Petlalcingo, 3-1V-1962, F. D. Parker, 1F. PANAMA: Canal Zone, Ancon, 4-V-1926, C. T. Greene, 1F; 9-IV- 1926, 1F. PARAGUAY: San Bernardino, IV, K. Fiebrig, 1M; 5-IV, 1M; -/-/-, 1M. TRINIDAD: St. Augustine, V-1959, F. D. Barrett, 1F. URUGUAY: Minas: Arassu- ahy, X-1929, Thieman, 1F. VENEZUELA: Barinitas: 10 km SE Barinitas, 1-III-1986, R. B. Miller, 1M. Genus Phycus Walker Phycus Walker (1850: 1): Irwin and Lyne- borg (1981: 260); Lyneborg (1978: 212). Type species: Xylophagus canescens Walker (1848: 129) by monotypy (= X)- lophagus brunneus Wiedemann, 1824: 19). 42 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Caenophanes Loew (1874: 415): Lyneborg (1978: 212). Type species: C. insignis Loew (1874: 415) by monotypy. Preoc- cupied by Caenophanes Foerster (1862: 236). Caenophanomyia Bezzi (1902: 191): Lyne- borg (1978: 212). New name for Caeno- phanes Loew (1874: 415). Paraphycus Becker (1923: 62): Lyneborg (1978: 212). Type species: Phycus nitidus Wulp (1897: 137) by original designation. Caenophaniella Séguy (1941: 112): Lyne- borg (1978: 212). Type species: C. nigra Séguy (1941: 112) by original designation. Moderate-sized, slender flies. Head.—Frons of both sexes at its narrow- est from as wide as to more than twice as wide as ocellar tubercle (Figs. 17, 35), wider in female than in male of same species, with shining black areas and tomentose areas, setae sparse, scanty, or absent; male eyes dichoptic, eye margins only slightly diver- gent from level of ocellar tubercle toward genae (Figs. 17, 35), facets equal, setae ab- sent; parafacial lacking setae; head slightly to distinctly protruding at antennal level; facial and genal calli absent; antenna (Fig. 16) elongate, length 1.4-1.9 times head depth; scape cylindrical; pedicel subtrian- gular; flagellum elongate, tapered apically, setae absent, ratio of lengths of flagellum to scape differs greatly among species; flagellar style terminal, two-segmented, with minute terminal spine; clypeus with surface con- cave, setae absent; maxillary palpus two- segmented (Fig. 18), basal segment cylin- drical, apical segment oval or lobate with large, apical sensory pit. Thorax.—np 1-2, sa 1, pa 1, de 0-1, sc 1; vittae absent; mesonotal setae short, uni- form, erect; prosternum bare in and around central depression; pleural setae very short, sparse. Wing (Fig. 19).—Pterostigma ab- sent; setulae dorsal on R,; subcostal cell and cell r, elongate, open; veins R, and R, sub- equal; cell r, large, enclosed apex of wing, m, usually closed, petiolate; discal cell trunctae basally; m-cu/r-m subequal. Legs. — Fore coxae elongate, sparsely setose, with 3 stiff apical macrosetae; middle coxae bare on posterior surface; femora without setae; tibiae with short setae. Abdomen.—Slender, especially in male, laterally compressed toward apex; dorsum convex, shining black or reddish brown; se- tae sparse and short; male and female lack lanceolate setae ventrally. Male terminalia (Figs. 20-24).—Tergite 8 (Ts) comparative- ly wide and deeply to moderately constrict- ed medially. Sternite 8 large and simple. Epandrium (Epa) simple, without posterior incision, may be shorter, equal, or longer in midline than maximum width; cerci (C) free, well sclerotized, extending posteriorly be- yond epandrium and ventral epandrial sclerite. Ventral epandrial sclerite, ventral view (VES) large, well sclerotized, setose, extending to anterior margin of epandrium with membranous attachment to dorsal apodeme of aedeagus. Hypandrium absent in New World species. Gonocoxites (Gx) broad, heavily sclerotized; separated ven- tromedially. Gonostylus (Gs) large, thick, extending slightly beyond apex of gonocox- ite. Aedeagus laterally attached by a strong- ly sclerotized bridge to midsection of par- amere; distiphallus (Dp) forms a fine, downwardly directed tube; dorsal apodeme (DA) broad; ventral apodeme (VA) shaped as two elongate extensions; ejaculatory apo- deme (EA) large; paramere composed of a distal process, which curves in inward and downward approaching base of gonostylus, and a stout basal apodeme. Female termi- nalia (Figs. 25, 26).—Tergite 8 (T,) broad, broadly concave along posterior margin. Tergite 9 (T,) subtriangular, separated me- dially with several strong setae along pos- terior margin. Tergite 10 bilobed, membra- nous. Cerci (C) membranous, simple, with scattered setae. Sternite 8 (S,) large, as long as wide, with short median incision on pos- VOLUME 91, NUMBER 1 43 21 Figs. 16-26. Phycus frommeri.—16. Antenna (5446). 17. Head of male, frontal view (5/69). 18. Maxillary palpus, lateral view (5446). 19. Wing, dorsal view (5/69). 20. Male tergite 8, epandrium, cerci, and ventral epandrial sclerite, dorsal view (5446). 21. Male ventral epandrial sclerite, ventral view (5446). 22. Male gono- coxites and gonostylus, dorsal view (5446). 23. Male gonocoxites and gonostylus, ventral view (5446). 24. Male aedeagus, lateral view (5446). 25. Female terminalia, dorsal view (5730). 26. Female terminalia, ventral view (5730). Abbreviations: (ApSg) apical segment; (BaSg) basal segment; (C) cercus; (DA) dorsal apodeme; (Dp) distiphallus; (EA) ejaculatory apodeme; (Epa) epandrium; (F) furca; (Gs) gonostylus; (Gx) gonocoxites; (S,) sternite 8; (Stp) stipes; (T,) tergite 8; (T,) tergite 9; (T,,) tergite 10; (VA) ventral apodeme; (VES) ventral epandrial sclerite. Scale = 0.1 mm, unless otherwise indicated. terior margin. Sternite 9 greatly modified, Washingtonia palms or other vegetation in invaginated beneath sternite 8 to form in- desert canyon bottoms (Irwin, personal ob- ternal furca (F). servation). Several Old World species of Phycus have been observed walking along dead and fal- len tree trunks 1n riverine habitats. The New World species from western North America Derivation of name: this species 1s named have been observed walking on rocks under in honor of Saul I. Frommer, University of Phycus frommeri Webb and Irwin, NEw SPECIES 44 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON California, Riverside, who collected most of the specimens. Diagnosis.—P. frommeri can be distin- guished from P. frontalis by the following combination of characteristics: eyes (Fig. 17) separated by a distance greater than the width of the ocellar tubercle; width of frons at level of lateral ocelli greater than 0.5 times width at level of antennal bases; postero- lateral extensions of gonocoxites in ventral view (Fig. 23) short, thick. Description of male holotype (5/69).— Length (excluding antenna) 6.7 mm. Head.—Ocellar tubercle black, tomen- tum light gray; ocellar setae black, short, appressed. Eyes dark brown, separated dor- sally by distance greater than width of ocel- lar tubercle (as in Fig. 17). Frons black, to- mentum light gray, converging slightly at vertex (as in Fig. 17), setae black, short, erect, in mediolateral row. Antenna (as in Fig. 16) dark brown, tomentum light gray; length of scape 0.32 mm, 2.3 times width, 2.3 times length of pedicel, setae black, scat- tered, erect, with several larger macrosetae ventrally; length of pedicel 0.14 mm, 1.0 times width, setae black, short, stiff} length of flagellum 0.78 mm, 4.9 times width, 2.4 times length of scape; length of basal sty- lomere 0.02 mm, length of apical stylomere 0.10 mm. Parafacial narrow, dark brown, tomentum silver. Genal setae white, elon- gate. Clypeus brown, tomentum silver. Maxillary palpus (as in Fig. 18) dark brown, tomentum silver; length of basal segment 0.34 mm, 5.4 times width, setae black, elon- gate; length of apical segment 0.10 mm, 1.7 times width, 0.29 times length of basal seg- ment, setae shorter than those of basal seg- ment. Labellum dark brown, tomentum light gray; setae along ventral margin brown, moderately long. Postocular setae white, fine, abundant ventrally becoming black, stiff, erect toward vertex. Thorax.— Black, tomentum light silver, setae white, fine, scattered; np 2, sa 1, pa 1, de 0, sc 1. Postpronotal lobe concolorous with thorax; setae white, fine, scattered. Propleuron, anepisternum, katepisternum, meron dark brown, tomentum silver; ane- pimeron fuscous, glossy. Pleural setae white to pale yellow, abundant on propleuron, and scattered over anepisternum; absent on re- maining pleural sclerites. Scutellum black, tomentum silver; setae white to pale yellow, scattered. Postnotum and laterotergite dark brown to black, tomentum silver; setae on laterotergite white to silver, elongate, abun- dant. Wing (as in Fig. 19).—Length 5.2 mm, width 1.8 mm, length 2.9 times width. Wing membrane hyaline with apical third pale smoky brown. Halter dark brown, tomen- tum light gray. Legs.—Coxae dark brown, tomentum silver; femora, tibiae and tarsi dark brown. Coxae with 3 stiff, brown apical setae. Abdomen. — Dark brown, subshiny; setae dark brown, short, appressed with pale yel- low, elongate setae laterally on tergite 1. Male terminalia (as in Figs. 20—24).—Tergite 8 (T,) as wide as epandrium, deeply concave medially. Epandrium (Epa) as long as wide medially; cerci broad (C), rounded apically; ventral epandrial sclerite (VES), ventral view, deeply incised medially on apical margin. Gonocoxites (Gx) in dorsal view broad, heavily sclerotized; in ventral view separated medially. Gonostylus (Gs) large, thick, extending slightly beyond apex of gonocoxites. Variation in males (N = 10).— Body length (excluding antenna) 5.7—7.3, 6.6 mm. Head depth 0.76-0.84, 0.81 mm. Length of an- tenna 1.5-1.9, 1.7 times head depth; length of scape 0.30-0.40, 0.34 mm, 2.3-3.0, 2.5 times width, 2.3— 2.9, 2.5 length of pedicel; length of pedicel 0.12—-0.14, 0.13 mm, 0.8- 1.2, 1.0 times width; length of flagellum 0.76-0.92, 0.83 mm, 4.7-5.8, 5.0 times width, 1.7-2.5, 2.0 times length of scape; length of basal stylomere 0.02 mm, length of apical stylomere 0.10 mm. Length of bas- al segment of maxillary palpus 0.34—0.40, 0.38 mm, 4.5—5.7, 5.0 times width; length of apical segment 0.10-0.14, 0.12 m, 1.2- 1.8, 1.5 times width, 0.26-0.39, 0.31 times VOLUME 91, NUMBER 1 length of basal segment. Wing length 4.5— 6.8, 4.9 mm, 2.8-3.0, 2.9 times width. Vari- ation in the color pattern was found in one specimen (5329) where the anepisternum, anepimeron, meron, metepleuron, middle and hind coxae, and entire abdomen were pale brown and subshiny. Female.—Similar to male with frontal se- tae scattered. Female terminalia (Figs. 25, 26).—Char- acteristics given in description of genus. Variation in females (N = 10).—Body length (excluding antenna) 8.2—9.6, 8.8 mm. Head depth 0.94—0.96, 0.95 mm. Length of antenna 1|.4—-1.6, 1.5 head depth; length of scape 0.40-0.46, 0.43 mm, 2.5-2.9, 2.7 times width, 2.0-2.9, 2.4 times length of pedicel; length of pedicel 0.16-0.20, 0.18 mm, |.0-1.3, 1.2 times width; length of fla- gellum 0.68-0.80, 0.74 mm, 4.3-5.0, 4.6 times width, 1.1-1.3, 1.2 times length of scape; length of basal stylomere 0.02 mm, length of apical stylomere 0.10 mm. Length of basal segment of maxillary palpus 0.42- 0.46, 0.45 mm, 5.3-5.8, 5.7 times width; length of apical segment 0.14 mm, 1.8 times width, 3.0—3.3, 3.2 times length of basal seg- ment. Wing length 6.2—6.5, 6.4 mm, length 2.0-2.2, 2.0 times width. Ecology.—The majority of specimens were collected in Malaise traps situated in canyon washes. Individual specimens were collected on rocks and under Washingtonia palms, with a single male collected on Er- iogonum fasciculatum polifolium (5602). In the collections examined, a distinct differ- ence was noted in the pattern between tim- ing of adults collected in Mexico and in Cal- ifornia. In Mexico, 92% of the adults were collected between 12 March and 5 May, while 8% were collected between 23 August and 8 November; 77% of the adults were collected during April. In California, adults were collected between 25 April and 29 July; the majority of specimens were collected between early May and mid June. Overall, males (405) were collected 1.7 times more frequently than females (243). Figure 27 45 No. of Adults 1970 Adults No, of go 15 31 15 30 15 APRIL MAY JUNE JULY Fig. 27. Phycus frommert.—Flight periodicity in Deep Canyon, Riverside County, California, in 1969 and 1970 as measured in a single Malaise trap. shows the variation in flight periodicity at Deep Canyon, Riverside County, Califor- nia, during 1969 and 1970. Two female lar- vae of Phycus frommeri were collected by R. B. Miller and L. A. Stange, one (5620) from under small conglomerates of fine soil at the base of a tree in a shaded gully and the other (56/9) in loose soil under boul- ders, both in the state of Colima, Mexico. These were reared to adults in our labora- tory. Larva.— Head (Figs. 28-31) length 0.42 mm; metacephalic rod (MR) elongate, pos- terior third clavate, length 0.86 mm, 2.0 times length of head. Prothoracic spiracle (Fig. 30). Posterior spiracle (Fig. 31). Pupa (Figs. 32-34).— Length 7.7-9.1 mm (N = 2), width 1.7-2.0 mm. Alar process not spinose. Labial sheath broad, truncate apically, not bisecting proboscial sheath. Length of antennal sheath 0.54 mm, length of subapical spine 0.14 mm. Length of tho- racic spiracle 0.38 mm, tapered apically, apex truncate. Posterior spines elongate, not divergent apically, length 0.50 mm. Dorsal spines on abdominal segments I-VII and 46 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ye eae MR ae - KE MP i: s 28 — Sa ANT a a iI A , Sus ) AS G / EN vee / K Sa © ] t / XS} Clay SEN Qe |. ‘ \) 33 \\ —— 10 a ) Bert t\ i ° , \\ wy — ry é s by | 3 \I % 29 | ( ca lh \{ LAY | : KT N 4 K 4h KK \ 32 j De pw \ a) \ i \ fp’ us Ne y MR—+ = \ \\ h [! yl S il ; } 34 \ ? x pi | Xx ull WF. NM 4% | Re ad ka [ ? « Nelo | A 31 \ \V/ Figs. 28-34. Phycus frommeri larva and pupa exuvia (56/9)—28. Larval head capsule, lateral view. 29. Larval head capsule, dorsal view (right), ventral view (left). 30. Prothoracic spiracle of larva, lateral view. 31. Posterior spiracle of larva, posterior view. 32. Pupal exuvium, lateral view. 33. Pupal exuvium, dorsal view (right), ventral view (left). 34. Frontal plate of pupal exuvium. Abbreviations: (ANT) antenna; (AS) antennal sheath; (DSp) dorsal spiracle; (M) mandible; (MP) maxillary palpus; (MR) metacephalic rod. Scale = 0.1 mm, unless otherwise stated. ventral spines on segments I-VII fine, not fused basally. Abdominal spiracles on seg- ments I-VII moderately thick, tubular. Distribution. — The range of Phycus from- meri extends from southern California to the southern tip of Baja California Sur and along the northwestern border of Sonora, Mexico southward to the state of Colima. Specimens examined (661).— Holotype: male, Irwin specimen number 5/69, CAS type no. 15741 (on permanent loan to the California Academy of Sciences from the University of California, Riverside); Cali- fornia, Riverside County, P. L. Boyd Desert Research Center, 3.5 mi S Palm Desert, marker #57, 18-—23-V-1970, S. I. Frommer, in a Malaise trap. Paratypes as follows: UNITED STATES—CALIFORNIA: Riy- erside Co.: P. L. Boyd Desert Deep Canyon Research Center, 5.6 km S Palm Desert, 6- 13-VI-1969, SIF, 14 M 13 F; 6-8-V-1970, SIF, 1 M; 15-18-V-1970, SIF, 23 M 5 F; 18-23-V-1970, SIF, 50 M 20 F; 27-V-1-VI- 1970, SIF, 35 M 15 F; 13-18-VI-1969, SIF RMW, 13 M 10 F; 18-19-VI-1969, SIF RMW, 2M 2F; 19-20-VI-1969, SIFRMW, 1M 1 F; 20-24-VI-1969, SIF RMW, 4 M 1 F; 8-VII-1969, SIF RMW, 1 F; 4-6-V- 1970, SIF RMW, | F; 26-IV-3-V-1970, SIF RMW, | M; 8-12-V-1970, SIF RMW, 7 M 2 F; 11-13-V-1970, SIF RMW, 5 M; 13- 15-V-1970, SIF RMW, 8 M 2 F; 23-25-V- 1970, SIF RMW, 16 M 18 F; 25-26-V-1970, SIF RMW, 25 M 17 F,; 1-4-VI-1970, SIF RMW, 8 M4 F; 21-29-V-1973, A. B. Tabet, 2 M; 5-13-VI-1973, A. B. Tabet, 7 M 4 F; 24-V-1969, MEI SIF, 44 M 12 F; 8-VI-1965, MEI, | M 1 F; 16-V-1969, MEI, 1 M 1 F; 18-V-1969, MEI, 8 M3 F; 21-V-1969, MEI, 11 M 2 F; 13-20-VI-1973, A. B. Tabet, 5 VOLUME 91, NUMBER 1 M; 21-VI-1962, EIS, 1 F; 19-V-1964, MEI, 1 M; 11-VI-1965, MEI, | F; 9-16-V-1973, A. B. Tabet, 1 M; 5 mi W Sage, 3-VII-1963, P. D. Hurd, | F; 2-VII-1963, E. I. Schlinger, 1 F; Idyllwild, 27-VI-1956, M.S. Wasbauer, 1 M; Carrizo Creek, 30-VI-1964, E. I. Schlinger, | F; Massacre Canyon, 300 yds up from Highway 79, 29-VII-1964, MEI, 1 F. San Diego Co.: Culp Canyon, | 2-VI-1958, E. I. Schlinger, 12 F; Borego, 1-V-1946, J. S. Perry, | M; 9.7 km E San Diego, 26-VI- 1963, H. L. Griffin, | F, R. L. Langston, | F; San Vicente Res., 229 m, 16-VI-1965, MEI, 1 F. MEXICO—Baja California Sur: 100 km NW La Paz, Arroyo Guadalupe, 107 m, 20-IV-1968, MEI, 2 M; 14.5 km S Loreto, 17-[V-1968, MEI, 12 M 1 F; 4.8 km E San Ignacio, 171 m, 14-IV-1968, MEI, 1 F; 3 km E La Burrera, 515 m, 2-3-IX- 1977, J. L. Fisher and R. L. Westcott, | M; 2.5 km E La Burrera, 549 m, E. M. and J. L. Fisher, | M: 6.4 km NW Don Pancho, 1-VII-1964, MEI, 1 F; Las Barracas, ca. 30 km E Santiago, 25-3 1-III-1982, P. DeBach, 1 M; 1-6-IV-1982, P. DeBach, 2 M IF; 13- 18-IV-1982, P. DeBach, 1 F; 19-24-IV- 1982, P. DeBach, | M; 25-30-IV-1982, P. DeBach, | M; 7-12-V-1982, P. DeBach, 4 M 2 F. Sinaloa: 86.9 km S Culiacan, 164.6 m, 23-I1V-1969, MEI, | M; “Las Escondi- das,” 106.2 km N Mazatlan, 137 m, 22-IV- 1968, MEI, 16 M 25 F. Colima: Los Tem- panes, 12-III-1985, R. B. Miller and L. A. Stange, | F reared from larva; Rio Salado, 7 km S Colima, 14-III-1985, | F reared from larva. Phycus frontalis Webb and Irwin, NEw SPECIES Derivation of name: front (Latin) = brow; alis (Latin) = pertaining to. Diagnosis.—P. frontalis can be distin- guished from P. frommeri by the following combination of characteristics: eyes sepa- rated dorsally only by width of ocellar tu- bercle (Fig. 35); frons distinctly convergent towards vertex (Fig. 35), width at level of lateral ocelli less than 0.5 times width at 47 level of antennal bases; posterolateral ex- tensions of gonocoxites in ventral view nar- row, attenuate (Fig. 39). Description of male holotype (5/79).— Length (excluding antenna) 6.8 mm. Head.—Ocellar tubercle fuscous, tomen- tum light gray; ocellar setae fuscous, mod- erately long, scattered. Eyes fuscous, sepa- rated medially by distance equal to width of ocellar tubercle (Fig. 35). Frons fuscous, tomentum light gray, dense silver around antennal bases; width at level of ocellar tu- bercle 0.40 times width at level of antennal bases (Fig. 35); setae pale yellow, moder- ately long, in mediolateral row; callus ab- sent. Antenna as in P. frommeri (Fig. 16), fuscous, tomentum light gray; scape, length 0.34 mm, 2.8 times width, 2.8 times length of pedicel, setae fuscous, short, subap- pressed with several large, stiff macrosetae ventrally; pedicel, length 0.12 mm, 1.0 times width, setae fuscous, short, suberect; flagel- lum, length 0.82 mm, 4.6 times width, 2.4 times length of scape; basal stylomere, length 0.04 mm, apical stylomere, length 0.10 mm, apical spine minute. Parafacial fuscous, to- mentum dense silver. Genal setae white to silver, elongate, abundant. Clypeus fuscous, tomentum dense silver. Maxillary palpus as in P. frommeri (Fig. 18), dark brown, to- mentum light gray; length of basal segment 0.34 mm, 4.3 times width, setae dark brown, moderately long, erect, on ventral surface; length of apical segment 0.14 mm, 0.41 times length of basal segment, setae sparse. Labellum dark brown, tomentum light gray; setae brown, moderately long, scattered. Postocular setae white to silver, elongate, abundant on ventral half, with numerous black, stiff setae dorsally. Thorax.—Fuscous, tomentum light gray; setae white to silver, moderately long, scat- tered; np 2, sa 1, pa 1, dc 0, sc 1. Postprono- tal lobe concolorous with thorax; setae white to silver, moderately long, scattered. Pro- pleuron, anepisternum, katepisternum, meron fuscous, tomentum silver, anepi- meron glossy. Pleural setae white to silver, 48 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON \ 37 \ Figs. 35-40. Phycus frontalis (5180). 35. Head of male, frontal view. 36. Male epandrium, cerci, and ventral epandrial plate, dorsal view. 37. Male ventral epandrial plate, ventral view. 38. Male gonocoxites and gonostylus, dorsal view. 39. Male gonocoxites and gonostylus, ventral view. 40. Male aedeagus, lateral view. Scale = 0.1 mm, unless otherwise stated. elongate, abundant on propleuron, scat- tered on anepisternum, on ventral third of katepisternum, absent on anepimeron and meron. Scutellum fuscous, tomentum light gray; setae pale yellow, moderately long, scattered. Postnotum and laterotergite fus- cous, tomentum light gray; setae on later- otergite white to silver, elongate, abundant. Wing as in P. frommeri (Fig. 19).—Length 5.0 mm, width 1.8 mm, length 2.8 times width. Halter fuscous, tomentum light gray. Legs.— Dark brown, coxae with tomentum silver; anterior tubercle on hind coxa dark brown, apical half pale. Abdomen.—Fuscous, subshiny; setae fuscous, short, appressed, mixed with pale yellow, elongate, suberect setae. Male Ter- minalia (as in Figs. 36-40).—Tergite 8 sim- ilar to P. frommeri. Epandrium rectangular, 1.25 times longer than wide, posterior mar- gin truncate; cerci lobate; ventral epandrial sclerite, ventral view, with dark brown se- tae. Gonocoxites in ventral view with pos- terolateral extension narrow, attenuate. Variation in males (N = 2).— Body length (excluding antenna) 6.8-6.9, 6.8 mm. Head depth 0.78-0.92, 0.85 mm. Antenna, length 1.1-1.5, 1.3 times head depth; scape, length 0.32-0.34, 0.33 mm, 2.3-2.8, 2.6 times width, 2.8—3.2, 3.0 times length of pedicel; pedicel, length 0.10-0.12, 0.11 mm. Max- illary palpus, basal segment length 0.30- 0.34, 0.32 mm, 3.8—-4.3, 4.1 times width; apical segment length 0.10-0.14, 0.12 mm, 1.3-1.8, 1.6 times width, 0.33-0.41, 0.37 times length of basal segment. Wing length 5.0-5.5, 5.3 mm, 2.8-3.1, 3.0 times width. Female.—Similar to male. Female Ter- minalia.—As in P. frommeri (Figs. 25, 26). Variation in females (N = 2).—Body length (excluding antenna) 7.4 mm. Head depth 1.10-1.18, 1.14 mm. Antenna length VOLUME 91, NUMBER | 1.4 times head depth; scape, length 0.41- 0.44, 0.43 mm, 2.6-2.8, 2.7 times width, 2.43.0, 2.7 times length of pedicel; pedicel, length 0.14-0.18, 0.16 mm, 0.8-1.0, 0.9 times width; flagellum, length 0.84 mm, 4.7 times width, 2.0 times length of scape; basal stylomere length 0.04 mm, apical stylomere length 0.10 mm. Maxillary palpus, basal segment length 0.40-0.52, 0.46 mm, 5.0- 5.2, 5.1 times width; apical segment length 0.16-0.20, 0.18 mm, 1.4-2.0, 1.6 times width; 0.38—0.40, 0.39 times length of basal segment. Wing length 6.7—7.7, 7.2 mm, 3.0- 3.1, 3.15 times width. Distribution. — The range of Phycus fron- talis extends from Yucatan, Mexico to Cos- ta Rica. Specimens examined (4).— Holotype: male, Irwin specimen number 5/79 (AMNB), Mexico, Yucatan, Chichén Itza, VI-29. Paratypes as follows: MEXICO— Yucatan: Chichén Itza, VI-1929, | M 1 F. COSTA RICA—La Suiza, 1924, P. Schild, ] F. ACKNOWLEDGMENTS We thank Drs. G. L. Godfrey, L. M. Page, and W. E. LaBerge for reviewing this manu- script and Audrey Hodgins for editorial comments. We also thank the following in- stitutions and their curators or former cu- rators for the loan of material relevant to this study: American Museum of Natural History, P. Wygodzinsky; British Museum (Natural History), K. V. G. Smith; Califor- nia Department of Food and Agriculture, M. S. Wasbauer; California Insect Survey, F. R. Cole; Canadian National Collection, H. J. Teskey; Cornell University, L. L. Pe- chuman; Instituto Miguel Lillo, Tucuman, Argentina, M. L. de Grosso; Kansas State University, D. H. Blocker; Naturhisto- risches Museum, Wien, R. Lichtenberg; E. I. Schlinger Collection; U.S. National Mu- seum of Natural History, W. W. Wirth; C. W. O’Brien Collection; University of Cali- fornia, Davis, R. O. Schuster; University of California, Riverside, S. I. Frommer; Zool- 49 ogisches Institut, Halle, Professor Hiising; Zoologisches Museum aus der Humboldt- Universitat zu Berlin, H. Schumann. Support for this study was supplied in part by the Illinois Natural History Survey and the University of Illinois at Urbana- Champaign. Additional support was pro- vided to M. E. Irwin by the National De- fense Education Act Fellowship, the Uni- versity of California Regents Patent Fund, the University of California Dry Lands In- stitute, and the University of California Deep Canyon Research Fund. LITERATURE CITED Becker, T. 1923. Wissenschaftliche Ergebnisse der mit Unterstiitzung der Academie der in Wien aus der Erbschaft Treitl von F. Werner unternommen zoologischen Expedition nach dem angloagyp- tuschen Sudan (Kordofan) 1914. VI. Diptera. Denkschr. Akad. Wiss., Wien 98: 57-82. Bezzi, M. 1902. Neue Namen fir einige Dipteren- Gattungen. Z. Syst. Hymenopt. Dipterol. 2: 190- 192. Bromley, S. W. 1934. Therevidae, pp. 360-361. /n Curran, C. H., ed., The Diptera of Kartabo, Bartica District, British Guiana, with descriptions from other British Guiana localities. Bull. Am. Mus. Nat. Hist. 66: 287-532. Foerster, A. 1862. Synopsis der Familien und Gat- tungen der Braconen. Verh. Naturh. Ver. Preuss. Rheinl. 19: 225-288. Irwin, M.E. 1972. Diagnoses and habitat preferences of the immature stages of three South African species of the Yestomyza-group (Diptera: Ther- evidae). Ann. Natal Mus. 21: 377-389. 1976. Morphology of the terminalia and known ovipositing behaviour of female Therevi- dae (Diptera: Asiloidea), with an account of cor- related adaptations and comments on phyloge- netic relationships. Ann. Natal Mus. 22: 913-935. . 1977a. Two new genera and four new species of the Pherocera-group from western North Amer- ica, with observations on habitats and behavior (Diptera: Therevidae: Phycinae). Proc. Entomol. Soc. Wash. 79: 422-451. 1977b. A new genus and species of stiletto- flies from southwestern North America with close affinities to Chilean and Australian genera (Dip- tera: Therevidae: Therevinae). Pan-Pac. Entomol. 58: 287-296. 1983. The boharti species group of the genus Pherocera (Diptera: Therevidae; Phycinae). Pan- Pac. Entomol. 59: 113-139. 50 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Irwin, M.E.and L. Lyneborg. 1981 [1980]. The gen- era of Nearctic Therevidae. Bull. Ill. Nat. Hist. Surv. 32: 193-277. Kréber, O. 1914. Beitrage zur Kenntnis der Therevi- den und Omphraliden. Jahrb. Hamb. Wiss. Anst. 31: 29-74. 1928. Neue und wenig bekannte Dipteren aus den Familien Omphralidae, Conopidae und Therevidae. Konowia 7: 116-134. . 1929. Neue Beitrage zur Kenntnis der There- viden und Tabaniden (Dipt.). Dtsch. Entomol. Ztschr. 1928: 417-434. Loew, H. 1874. Diptera nova a Hug. Theod. Chnis- topho collecta. Z. Ges. Naturw. Berlin. N. F. 9: 413-420. Lyneborg, L. 1968. A comparative description of the male terminalia in Thereva Latr., Dialineura Rond., and Psilocephala Zett. (Diptera, Therevidae). Entomol. Meddel. 36: 546-559. 1972. A revision of the Yestomyza-group of Therevidae (Diptera). Ann. Natal Mus. 21: 297- 376. 1976. A revision of the therevine stiletto- flies (Diptera: Therevidae) of the Ethiopian region. Bull. Br. Mus. Nat. Hist. (Entomol.) 33: 189-346. 1978. The Afrotropical species of Phycus Walker (Diptera: Therevidae). Entomol. Scand. 9: 212-233. 1983. A review of the Palaearctic genera of Phycinae (Insecta, Diptera, Therevidae). Steen- strupia 9: 181-205. Malloch, J. R. 1932. Rhagionidae (Leptidae), Ther- evidae, Scenopinidae, Mydaidae, Asilidae, Lon- chopteridae, pp. 199-283. Jn British Museum (Natural History), Diptera of Patagonia and South Chile. Part V, Fascicle 3. Olivier, A. G. 1813. Premiér mémoire sur quelques insectes qui attaquent les céréales. Mem. Soc. Agric. Seine 16: 477-495. Rauch, P. A. 1970. Electronic data processing for entomological museums, an economical approach to an expensive problem. Ph.D. dissertation in entomology, University of California, Riverside. 78 pp. Richards, O. W. 1929. Systematic notes on the Bor- boridae (Diptera), with descriptions ofa new species of Leptocera (Limosina). Entomol. Month. Mag. 65: 171-176. Seguy, E. 1941. Diptéres recueillis par M. L. Chopard d@’Alger a la Céte dIvoire. Ann. Soc. Entomol. Fr. 109(1940): 109-130. Stuckenberg, B. R. and M. E. Irwin. 1973. Standards for entomological labels. Bull. Entomol. Soc. Am. 19: 164-168. Walker, F. 1848. List of the specimens of dipterous insects in the collection of the British Museum. 1: 1-229. London. 1850. Diptera, 1: 1-76. Jn Saunders, W. W., ed., [1856] Insecta Saundersiana: Or characters of undescribed insects in the collection of William Wilson Saunders. London. Wiedemann, C. R. W. 1824. Munus rectoris in Aca- demia Christiana Albertina aditurus analecta en- tomologica ex Museo Regio Havniensi. 60 pp., | pl. Kiliae [Kiel]. Wulp, F. M. vander. 1897. Zur Dipteren-Fauna von Ceylon. Természetr. Fiizet. 20: 136-144. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 51-54 A MID-SUMMER COMPARISON OF SIZES AND GROWTH RATES AMONG NYMPHS OF THREE SYMPATRIC MANTIDS (MANTODEA: MANTIDAE) IN TWO OLD-FIELD HABITATS L. E. HURD AND R. M. EISENBERG Ecology Program, School of Life Sciences, University of Delaware, Newark, Delaware 19716. Abstract.—We compared three species of mantid nymphs in two field habitats (sites CHRY and AG), in two censuses during mid-summer 1986. Tenodera sinensis (Saussure) nymphs exhibited no difference between sites, either in size of nymphs, or in rate of maturation. Both 7. angustipennis and M. religiosa were significantly larger in site CHRY than in site AG in August, although no such difference had been evident in July. More individuals of these two species had also matured in CHRY than in AG by the August census. These data suggest that 7. angustipennis and M. religiosa were more food limited at site AG than at site CHRY during the time just prior to maturation. These results are consistent with the hypothesis that 7enodera sinensis is more of a habitat generalist than the other two species. Key Words: ators Mantids are generalist predators subject to food limitation which varies in impor- tance during a growing season, as well as between habitats and years (Matsura et al. 1975, Hurd et al. 1978, Hurd and Eisenberg 1984, Eisenberg et al. 1981). Food limita- tions on newly hatched nymphs can retard development and decrease size of the im- ago, which in turn reduces fecundity (E1- senberg et al. 1981, Matsura and Marooka 1983, Hurd and Eisenberg 1984, Hurd and Rathet 1986). Therefore, comparing sizes of mantids collected at the same time from different habitats could provide a relative measure of the differences between habitats in terms of food limitation and fitness for a given species. We can then ask the same question of any number of mantid species found together in the same habitat to dis- cern whether the resource level in a given Mantodea, Mantidae, Tenodera, Mantis, food limitation, sympatry, pred- field is qualitatively the same for each species. Most ecological work on mantids has concentrated on a single species in a single habitat during either the first month of life or adulthood. Rathet and Hurd (1983) stud- ied growth rates and habitat placement within a single field site, of nymphs of three species which commonly occur together in old fields in northern Delaware: Tenodera sinensis (Saussure), 7. angustipennis (Saus- sure), Mantis religiosa (Linnaeus). These three morphologically similar species hatch at different times, differ in body size (7. sinensis > T. angustipennis > M. religiosa), and/or inhabit different levels of foliage within a field. These niche differences sug- gest to us that there may be differences in resource utilization among these species. In an exploratory study designed to detect both 52 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON MANTID LENGTH (mm) hes Ten. aA a a 50 0 50 ee (%) Fig. |. Sizes of mantid nymphs in two habitats on 19 July 1986. Open bars represent frequency of nymphs of various sizes found at site AG; shaded bars are for site CHRY. Ts = Tenodera sinensis, Ta = T. angus- tipennis, Mr = Mantis religiosa. One-way ANOVA: F = 24.945: df = 5, 72; P < 0.001. LSD comparisons indicate 7. sinensis is significantly larger than the other two species; no differences between sites. site and interspecific differences, we com- pared sizes and growth rates of these three mantid species in two old fields during the last nymphal stages. MATERIALS AND METHODS The two old-field habitats in our study are in Newark, New Castle County, Dela- ware. They are approximately 3 km apart and have different vegetation. One field, site AG, is located on the experimental farm of the School of Agriculture, University of Delaware, and is dominated by a dense ground cover of timothy (Ph/eum pratense) and Canada bluegrass (Poa compressa) with patches of goldenrod (Solidago spp.) and thistle (Cirsium spp.). The other field, site CHRY, is adjacent to the Chrysler plant in Newark, and is dominated by goldenrod with a ground cover of grasses, chiefly tim- othy. We censused mantids in both habitats on 100 5 Ts 904 i Eaey MANTID LENGTH (mm) = 40 S| 30 Ttolahalalstetapalctetelfqtetalislalml alata heliatn lite lam 50 0 50 0 50 0 50 FREQUENCY (%) Fig. 2. Sizes of mantid nymphs in two habitats on 13 August 1986. Open bars represent frequency of nymphs of various sizes found at site AG; shaded bars are for site CHRY. Ts = Tenodera sinensis, Ta = T. angustipennis, Mr = Mantis religiosa. One-way AN- OVA: F = 25.714; df = 5, 84; P < 0.001. LSD com- parisons indicate 7. sinensis is significantly larger than the other two species; site differences indicated by as- terisks. 19 July and 13 August 1986, which brack- eted the last month of nymphal life for these species. Individuals were hand caught, iden- tified, measured for length (front of head to tip of abdomen), and then released at the point of capture. Sex was recorded for those mantids which had matured by the second census. Lengths were compared among species and habitats on each census date with one-way ANOVA and least signifcant difference post hoc comparison using Stat- graphics (STSC, version 2.1). RESULTS AND DISCUSSION Tenodera sinensis nymphs were larger than the other two species at both sites in July (one-way ANOVA, F = 24.945; df = 5, 72; P < 0.001), and M. religiosa was not significantly different in size from 7. an- gustipennis (Fig. 1). This interspecific dif- ference was expected from earlier results VOLUME 91, NUMBER 1 Table 1. 53 Numbers of individuals, % adult and adult sex ratio for three mantid species in two habitats (AG and CHRY) in July and August of 1986. Mr = Mantis religiosa, Ts = Tenodera sinensis, Ta = T. angustipennis. Species Site July N August V % Adult M:F AG 12 1 75 Ee MECH ETOSe CHRY 20 20 100 10:10 T. angustipennis AG ay Y - CHRY 23 8 25 2:0 T. sinensis AG 2 2 oy CHRY 26 22 59 7:6 (Rathet and Hurd 1983) and published size differences for adults (Gurney 1950, Helfer 1963). There were no differences in size be- tween sites for any species at this time. Ap- parently, then, the sites did not differ in food limitation from egg hatch until mid-July. There were no adults of any species present in either field on 19 July. The August census revealed a number of differences which were not apparent in July (Fig. 2). Tenodera sinensis was stil signifi- cantly larger than the other two species at both sites (one-way ANOVA, F = 25.714; df = 5, 84; P < 0.001), and did not exhibit a difference between sites in size or per- centage of adults (Table 1). However, both M. religiosa and T. angustipennis were sig- nificantly larger at site CHRY than at site AG (Fig. 2), and more had matured at site CHRY (Table 1). The size differences could not be attributed solely to a greater number of adults at CHRY because even the largest adults at AG were smaller than the largest adults at CHRY for both species. These data suggest that there was greater food limita- tion during the intervening month at site AG for these two species than at site CHRY. Some adult M. religiosaat CHRY were con- siderably larger (75-80 mm) than previ- ously published records for this species (about 65 mm) (Gurney 1950, Helfer 1963). The sizes of the other two species (Fig. 2) were consistent with the literature. Eisenberg et al. (1981) hypothesized that mantids could rely on insects which forage on late summer flowers (e.g. goldenrod) for a major portion of their prey when insect biomass declines in the rest of the field. Hurd (1989) found that adult female 7. sinensis on flowers gained significantly more weight and deposited oothecae with more eggs (i.e. had greater fitness) than those on plants not in flower. We have found mostly 7. sinensis on these flowers, which may explain why they apparently were not as food limited at site AG as the other two species. Thus, while all three species might rely on the same prey early in the season, M. religiosa and T. an- gustipennis could be better indicators of habitat quality in terms of resident arthro- pod prey availability than 7. sinensis late in the season. A detailed examination of the diets of these species in the field is needed to test these ideas. ACKNOWLEDGMENTS We thank Cynthia Kendall for help in the field, and two anonymous referees for very thorough reviews of the manuscript. This work was supported by NSF Grant BSR 8506181, and is contribution #1 11 from the Ecology Program, University of Delaware. LITERATURE CITED Eisenberg, R. M., L. E. Hurd, and J. A. Bartley. 1981. Ecological consequences of food limitation for adult mantids (Tenodera aridifolia sinensis Saussure). Am. Midl. Nat. 106: 209-218. Gurney, A. B. 1950. Praying mantids of the United States, native and introduced. Smithsonian Inst. Annu. Rep. 1950: 339-362. Helfer, J. R. 1963. How to Know the Grasshoppers, Cockroaches, and Their Allies. Dubuque, Iowa, W. C. Brown. 353 pp. Hurd, L. E. 1989. On the importance of late season 54 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON flowers to the fitness of an insect predator, Ten- odera sinensis Saussure (Orthoptera: Mantidae), in an old field community. Entomologist (U.K.). Hurd, L. E.and R. M. Eisenberg. 1984. Experimental density manipulations of the predator Tenodera sinensis (Orthoptera: Mantidae) in an old-field community. I. Mortality, development and dis- persal of juvenile mantids. J. Anim. Ecol. 53: 269- 281. Hurd, L. E., R. M. Eisenberg, and J. O. Washburn. 1978. Effects of experimentally manipulated den- sity on field populations of the Chinese mantis (Tenodera aridifolia sinensis Saussure). Am. Midl. Nat. 99: 58-64. Hurd, L. E. and |. H. Rathet. 1986. Functional re- sponse and success in juvenile mantids. Ecology 67: 163-167. Matsura, T., T. Inoue, and Y. Hosomi. 1975. Eco- logical studies of a mantid, Paratenodera angus- tipennis de Saussure I. Evaluation of the feeding condition in natural habitats. Res. Pop. Ecol. (To- kyo) 17: 64-76. Matsura, T. and K. Marooka. 1983. Influence of prey density on fecundity in a mantis, Paratenodera angustipennis (S.). Oecologia 56: 306-312. Rathet, 1. H. and L. E. Hurd. 1983. Ecological rela- tionships among three co-occurring mantids, Tenodera sinensis (Saussure), T. angustipennis (Saussure), and Mantis religiosa (Linnaeus). Am. Midl. Nat. 110: 240-248. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, p. 54 NOTE Ixodes downsi (Acari: Ixodes downsi Kohls was originally de- scribed from Aripo Cave, Trinidad, on the basis of a male, female, and three nymphs found on the wall of the cave, and one larva found on a bat, Anoura g. geoffroyi Gray in the cave (Kohls 1957. Proc. Entomol. Soc. Wash. 59: 257-264). On the basis of this scant information, Kohls (op. cit.) specu- lated that it might be a bat tick although he noted that oil birds, Steatornis caripensis Humboldt, nested in the cave and a large rat was also seen (in the cave). J. downsi has not been recorded since the original descrip- tion. We report a second locality for the species at a considerable distance from the type lo- cality. An engorged female was collected from the throat of a young, fully feathered S. caripensis found on the floor of a cave near Tingo Maria, Department Huanuco, Peru, on 11 February 1975 by Baker. Bats were heard in the cave but none were seen and several unidentified parrots flew from the cave. This locality is approximately 2600 km southwest of the type locality, on the Ixodidae) from Peru opposite side of South America. The finding of a specimen on an oil bird only confuses the issue as to whether /. downsi is a bat tick or a bird tick. There are no known species of /xodes commonly found on both birds and bats, and additional collecting from hosts will be necessary before the true host relationships can be ascertained. The length and width measurements of the engorged female from Peru, compared with those of the unengorged allotype fe- male in parentheses, were 7.22 (2.53) mm and 4.07 (1.87) mm, respectively. In all oth- er respects the Peruvian specimen agrees with the original description. The specimen will be deposited in the Florida State Collection of Arthropods, Florida Department of Agriculture and Consumer Services, Gainesville, Florida. Nixon Wilson, Department of Biology, University of Northern Iowa, Cedar Falls, Iowa 50614; W. Wilson Baker, Tall Tim- bers Research Station, Rt. 1, Box 678, Tal- lahassee, Florida 32312. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 55-58 TEMPORAL DISTRIBUTION OF HATCHING TIMES IN THREE SYMPATRIC MANTIDS (MANTODEA: MANTIDAE) WITH IMPLICATIONS FOR NICHE SEPARATION AND COEXISTENCE L. E. HurpD AND R. M. EISENBERG Ecology Program, School of Life Sciences, University of Delaware, Newark, Delaware 19716. Abstract. —Mantids are members of a guild of generalist predators in arthropod com- munities. Three species commonly coexist in old fields: Tenodera sinensis Saussure, T. angustipennis Saussure, and Mantis religiosa Linnaeus. Egg hatch occurs early in the spring, producing high predator density when prey are in short supply. First instar nymphs of the two Tenodera species, which are most similar in size and occupy the same vegetational stratum in the field, are most divergent in timing of egg hatch. Nymphs of 7. sinensis hatch first, and enter the second instar as 7. angustipennis are beginning to hatch. The timing of egg hatch for Mantis religiosa, which is much smaller than the other two and occupies a different portion of the vegetation, is intermediate. Therefore nymphs of the three species are not the same body size at the same time; body size determines size of prey which can be captured most efficiently. We suggest that the temporal disparity in egg hatch may be one mechanism for coexistence among these species by reducing niche overlap when resources are critically limiting. Key Words: patry Praying mantids (Mantodea: Mantidae) are members of a guild (sensu Root 1967) of generalist predators which inhabit ar- thropod communities. This guild includes other taxa, notably spiders, and all members have in common a bitrophic niche; they oc- cupy both the third and fourth trophic levels by virtue of feeding on other predators as well as on herbivores. Thus, the relationship among guild members is complicated by the potential for both interspecific competition and mutual predation. Added to this is the propensity for cannibalism in some groups. Therefore, interspecific coexistence in this guild, especially among very similar species, may depend upon factors which mitigate a variety of antagonistic interactions. Three similar species of mantids com- Mantodea, Mantidae, Tenodera, Mantis, bitrophic niche, predators, sym- monly co-occur in northern Delaware. Te- nodera sinensis Saussure, the largest of the three, was originally introduced from the orient near Philadelphia, Pennsylvania, in 1896 (Laurent 1898), and has occurred in Delaware at least since the 1930s (Jones 1933). Its congener, 7. angustipennis Saus- sure, was found in Maryland in 1926 (Gur- ney 1950), and had become established in Delaware by 1930 (Jones 1933). Mantis re- ligiosa Linnaeus, the smallest of the three, was introduced into the northeastern United States in 1899 (Gurney 1950); its duration of residence in Delaware is uncertain (Rath- et and Hurd 1983). All three species are sympatric in Asia (Jing-jin et al. 1981). We have been studying various popula- tions of mantids, especially 7. sinensis, in 56 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON northern Delaware over the past decade (E1- senberg and Hurd 1977, Hurd et al. 1978, Eisenberg et al. 1981, Rathet and Hurd 1983, Hurd and Eisenberg 1984a, b, Hurd 1985, Hurd and Rathet 1986, Hurd 1988). We have observed that all three species can be found in the same general habitat: succes- sional old fields. Emergence from oothecae takes place in early spring when arthropod biomass is very low. In fact, mantid bio- mass at egg hatch is apt to be much greater than biomass of all other arthropods com- bined, so nymphs are likely to be severely food limited (Hurd and Eisenberg 1984a, b). It is therefore reasonable to ask: how do the niches of these similar predators differ to allow them to coexist in a limited envi- ronment? Rathet and Hurd (1983) found that both Tenodera species occupy the same vegetational stratum (30-60 cm_ foliage height), whereas M. religiosa is generally found lower in the vegetation (5-25 cm). Therefore, M. religiosa may escape preda- tion from the two larger species, although this does not rule out interspecific compe- tition. Rathet and Hurd (1983) also found that appearance of first instar nymphs in the field was not simultaneous among the three species, and that as a consequence there was a consistent interspecific size difference throughout the season. Mantid nymphs of different sizes have different optimal prey sizes based upon capture efficiency (Bartley 1983). In the present study we ask whether there is an intrinsic difference in timing of egg hatch which could be another mecha- nism which reduces niche overlap in this guild of generalist predators. MATERIALS AND METHODS We collected oothecae of 7. sinensis, T. angustipennis, and M. religiosa from two old fields in Newark, New Castle County, Delaware, on 10 February 1986. The col- lection sites are approximately 3 km apart, and represent somewhat different vegeta- tional assemblages. One field is located on the experimental farm property of the School of Agriculture, University of Delaware (= site AG), and dominated by a dense ground cover of timothy (Phleum pratense) and blue-stemmed grass (Poa compressa) with patches of goldenrod (Solidago spp.) and thistle (Cirsium canadensis). The other site is adjacent to the Chrysler plant in Newark (= site CHRY), and is dominated by gold- enrod with a sparse ground cover of grasses, chiefly timothy. Twelve oothecae of each mantid species from each of the two field sites were re- turned to the laboratory and immediately placed in separate containers and incubated at 25°C and ambient light. When eggs hatched, newly emerged nymphs were re- moved from the containers and killed by freezing, after which they were counted. A random sample of 20-30 nymphs from each group was measured for body length (tip of abdomen to front of head). RESULTS Tenodera sinensis began hatching after 18 days of incubation, and hatching continued for 11 days from oothecae collected from both sites, with a few nymphs emerging from AG oothecae as late as day 33 (Fig. 1). The pattern of hatching was sporadic with no clear difference between field sites. In only two cases did nymphs emerge on two or more consecutive days from the same ootheca, which is consistent with results from Eisenberg and Hurd (1977). Mean size of nymphs from site CHRY (9.7 mm, SD = (0.37) was somewhat larger than from site AG (9.4 mm, SD = 0.40) (t = —2.84, df= 48, P < 0.01). Tenodera angustipennis did not start hatching until 7. s7nensis had finished, and duration of hatch was six days (Fig. 1). The pattern of hatching was nearly identical for eggs from both sites. Only one ootheca ex- hibited hatching on two consecutive days. Mean body length of nymphs (8.9 mm, SD = (0.20) was not different between sites. Mantis religiosa was the only one of the three species to exhibit a marked site dif- VOLUME 91, NUMBER 1 ference in hatching pattern (Fig. |). Hatch- ing time was intermediate, and overlapped with both Tenodera species. Unlike Teno- dera spp., all but one ootheca produced nymphs on consecutive days, as many as five days in a row. As with 7. sinensis, nymphs from site CHRY were larger (6.1 mm, SD = 0.31) than nymphs from site AG (mean = 5.8 mm, SD = 0.20) (t = —3.78, df = 58, P < 0.001). DISCUSSION We found that under identical laboratory conditions of a “common garden” experi- ment, eggs of 7. sinensis hatch before those of the other two species. This is in agree- ment with field observations of first ap- pearance of nymphs (Rathet and Hurd 1983), and indicates an intrinsic difference between species. Matsura (pers. comm.) has also observed that 7. sinensis hatches before T. angustipennis in Japanese fields. A potential significance of this phenolog- ical difference is a reduction in overlap among species with regard to optimal prey size (Bartley 1983), which could reduce competition for limited food early in the growing season: by the time 7. angustipen- nis and M. religiosa begin to hatch, the ear- liest 7. sinensis nymphs would be entering the second instar (Hurd and Eisenberg 1984a, Hurd and Rathet 1986). This is like- ly to be more important for the two Teno- dera species, since they are similar in size at emergence and occupy the same vegeta- tional stratum, than for M. religiosa which occupies a different portion of the vegeta- tion (Rathet and Hurd 1983) and is much smaller. The fact that the timing of egg hatch in M. religiosa overlaps the other two species may be further indication that interaction between these genera is less important than between species of Tenodera. Of course the question remains as to whether the tem- poral differences in hatching arose from his- toric interspecific competition, or represent a fortuitous preadaptation to living in the same habitat. 57 Mantis religiosa 40 Tenodera angustipennis PERCENT TOTAL EGGS HATCHING no Tenodera sinenis 18 20 25 30 35 40 DAYS OF INCUBATION Fig. 1. Timing and pattern of egg hatch, as a percent of total eggs hatching for each species from oothecae collected at two field sites: AG = shaded bars, CHRY = open bars. In instances where food is severely lim- iting, the size difference between second in- star nymphs of 7. sinensis and first instar T. angustipennis would confer an additional benefit on the former species, that of being able to use its congener (and later-hatching conspecifics) as a food source. We have ob- served in the laboratory that intra-instar predation between and within these con- generic nymphs is relatively uncommon, as is the case with M. religiosa according to Roeder (1936). However, we have also ob- served that the difference in size between successive instars is sufficiently great to pro- voke much cannibalism and interspecific predation (Hurd 1988). This may help to explain why we have never found 7. an- gustipennis to be abundant in fields wherein T. sinensis was well-established. We found two indications of differences between field sites: the difference in hatch- ing patterns for M. religiosa, and the larger nymphs at site CHRY for both this species and 7. sinensis. The difference in nymph 58 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON size may well be an indication of differences in feeding conditions for adult mantids dur- ing the previous season. Eisenberg et al. (1981) found that food limitation among adults plays an important role in biomass of oothecae; egg size may also be affected. This would indicate that adults of these two species were more food limited at site AG than at site CHRY. We are currently testing this. Another possibility is that the size differ- ences represent ecotypes for one or both species. This remains to be tested, as does the question of what is the gain in fitness by hatching at a larger size? One possibility is a decline in time of development, al- though 7. sinensis, at least, is flexible in this regard (Hurd and Rathet 1986). We cannot as yet offer a satisfactory explanation for the difference in hatching patterns for /. reli- giosa (unless ecotypic); however, roughly the same pattern difference was observed the following year (unpublished data). The question of the significance of multiple con- secutive-day hatching of eggs in this species also remains to be answered. ACKNOWLEDGMENTS This work was supported by NSF Grant BSR 8506181. This is contribution #110 from the Ecology Program, University of Delaware. LITERATURE CITED Bartley, J. A. 1983. Prey selection and capture by the Chinese mantid (7enodera sinensis Saussure). Ph.D. dissertation, University of Delaware, New- ark, DE. Eisenberg, R. M. and L. E. Hurd. 1977. An ecological study of the emergence characteristics for egg cases ofthe Chinese mantis (Tenodera aridifolia sinensis Saussure). Am. Midl. Nat. 97: 478-482. Eisenberg, R. M., L. E. Hurd, and J. A. Bartley. 1981. Ecological consequences of food limitation for adult mantids (7enodera aridifolia sinensis Saussure). Am. Midl. Nat. 106: 209-218. Gurney, A. B. 1950. Praying mantids of the United States, native and introduced. Smithson. Inst. Annu. Rep., 339-362. Hurd, L. E. 1985. Ecological considerations of man- tids as biocontrol agents. Antenna, Bull. Royal Entomol. Soc. London 9: 19-22. 1988. Consequences of divergent egg phe- nology to predation and coexistence in two sym- patric, congeneric mantids (Orthoptera: Manti- dae). Oecologia 76: 547-550. Hurd, L. E. and R. M. Eisenberg. 1984a. Experi- mental density manipulations of the predator Tenodera sinensis (Orthoptera: Mantidae) in an old-field community. I. Mortality, development and dispersal of juvenile mantids. J. Anim. Ecol. 53: 269-281. 1984b. Experimental density manipulations of the predator 7enodera sinensis (Orthoptera: Mantidae) in an old-field community. II. The in- fluence of mantids on arthropod community struc- ture. J. Anim. Ecol. 53: 955-967. Hurd, L. E., R. M. Eisenberg, and J. O. Washburn. 1978. Effects of experimentally manipulated den- sity on field populations of the Chinese mantis (Tenodera aridifolia sinensis Saussure). Am. Mid. Nat. 99: 58-64. Hurd, L. E. and I. H. Rathet. 1986. Functional re- sponse and success in juvenile mantids. Ecology 67: 163-167. Jing-jin, Y., X. Chong-hua, Y. De-fu, and L. Ying-mei. 1981. Studies on the bionomics of six mantids from China. Coll. Res. Papers, Chinese Acad. For- estry 2: 67-74. Jones, F. M. 1933. Another oriental mantis well es- tablished in the U.S. (Tenodera angustipennis Saussure). Entomol. News 44: 1-3. Laurent, P. 1898. A species of Orthoptera. Entomol. News 9: 144-145. Rathet, 1. H. and L. E. Hurd. 1983. Ecological rela- tionships among three co-occurring mantids, Tenodera sinensis Saussure), 7. angustipennis (Saussure), and Mantis religiosa (Linnaeus). Am. Midl. Nat. 110: 240-248. Roeder, K. D. 1936. Raising the praying mantis for experimental purposes. Science 83: 582-583. Root, R. B. 1967. The niche exploitation pattern of the blue-gray gnatcatcher. Ecol. Monogr. 37: 95- 124. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 59-65 NEW WORLD SPECIES OF HOLCOPELTE AND IONYMPHA (HYMENOPTERA: EULOPHIDAE), WITH DESCRIPTIONS OF TWO NEW SPECIES CHRISTER HANSSON Department of Systematic Zoology, Lund University, Helgonavagen 3, S-223 62 Lund, Sweden. Abstract.—The genera Holcopelte Forster and Jonympha Graham (Hymenoptera: Eu- lophidae) are for the first time recorded from the Americas. The American Holcopelte comprises two new species: H. americana from Canada and the United States, and H. huggerti from the United States, Ecuador and Peru. Females and males are described for both species. Jonympha is represented by two species in the Americas: J. carne (Walker) from Canada, Ecuador and Peru, and /. ochus (Walker) from Canada. The hosts are unknown for all four species. Key Words: Forster (1856) described Holcopelte to in- clude Elachistus obscurus (Forster, 1841). Later, he described H/. fulvipes (1861), which was later synonymized with H. obscura by Boucek & Askew (1968). Thomson (1878) included Holcopelte as a species-group in the genus Derostenus; he also described D. sulciscuta which was later transferred to Holcopelte (Graham, 1959). Ashmead (1894) described six species of Holcopelte from the Island of St. Vincent. Later (1904), he synonymized Holcopelte with Horisme- nus Walker. Erdés (1958) described Horis- menus lenticeps from Hungary, which was later transferred to Holcopelte (Boucek & Askew, 1968). Graham (1959) resurrected Holcopelte and presented a key to the two British species. Boucek (1969) described Holcopelte stelteri from Germany and gave a key to the four European species. The six species described from St. Vin- cent by Ashmead were Horismenus, and prior to this paper no records of Holcopelte from the Americas have been published. Since only the females are known in the Eulophidae, Holcopelte, Ionvmpha, New World, taxonomy European species, this is the first time male Holcopelte are described. Hosts are only known for two of the European species (ob- scura, Stelteri), both have been reared from gallmidges (Diptera, Cecidomyiidae) (Bou- éek & Askew 1968, Boucek 1969). Ionympha was described by Graham (1959) to include Entedon carne and E. ochus; both species were originally de- scribed by Walker (1839). No additional species have been described. The genus was only known from Europe. Hosts are un- known for both species. Abbreviations used in the text are: HE = height of an eye; MO = width of mouth opening; MS = malar space; OOL = dis- tance between one posterior ocellus and eye; POL = distance between posterior ocelli; POO = distance between posterior ocelli and occipital margin; WH = width of head (dor- sal view); WT = width of thorax across shoulders. Abbreviations of museums and private collections are as follows: BMNH = British Museum (Natural History), London, England; CH = collection of the author; 60 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON CNC = Canadian National Collections, Ot- tawa, Canada; LUZM = Lund University Zoological Museum, Lund, Sweden; USNM = National Museum of Natural History, Washington, D.C., USA. Genus Holcopelte Forster Holcopelte Forster, 1856: 78. Type-species: Elachistus obscurus Forster, 1841: 40, by original designation. Diagnosis.— Head lenticular; face, frons and occiput smooth and polished; interan- tennal elevation high and clearly delimited (Figs. 3, 5, 7, 9); mandibles four-dentate (Fig. 7); antenna with two small, discoid anelli; notaular depressions distinct, long and narrow (Fig. 1); thorax with a median furrow extending from posterior mesoscu- tum to anterior scutellum (Fig. 1); post- marginal vein shorter than stigmal vein; male four basal flagellar segments with a single whorl of long hairs at the base of each segment, fifth segment with a basal whorl and with scattered hairs in front of whorl (Figs. 4, 8). Remarks.—The monophyly of Holco- pelte is shown through the following syn- apomorphies: 1) face, frons and occiput smooth and polished; 2) notaular depres- sions distinct, long and narrow, and clearly delimited from remaining mesoscutum; 3) thorax with a median furrow extending from posterior mesoscutum to anterior scutel- lum. The American species differ from Euro- pean species by having a pale petiole and female forecoxa brown (European species with petiole dark and all coxae with same color in female). KEY TO THE NEw WORLD SPECIES OF HOLCOPELTE l= Females\. =. 2 — Males Antennal scrobes joining before reaching ver- tical line of frontal fork (Fig. 3); scape 5.0 x as long as wide with 6-7 setae along ventral edge (Fig. 2); reticulation on median thoracic dor- sum engraved (slightly raised on the sides and i) along hind part of scutellum); cubital hair-line of forewing almost straight (Fig. 1); propodeum sculptured and with 4 plicae (Fig. 1); petiole conical without a dorsal shield La hee shad siviscere katie H. americana new species - Antennal scrobes never meet (Fig. 7); scape 8.0 x as long as wide with 3-5 setae along ven- tral edge (Fig. 6); thoracic dorsum with raised reticulation; cubital hair-line of forewing strongly sinuate below speculum (Fig. 12); pro- podeum smooth to almost smooth with 2 pli- cae (Fig. 11); petiole quadrangular (shape transverse to slightly elongate) with a dorsal shield that covers petiolar foramen (Figs. 10, 0 |) ee ee Ae . H. huggerti new species 3. Petiole small and conical, without a dorsal shield (like Fig. 1); cubital hair-line of forewing almost straight (like Fig. 1); scape apically ex- panded (Fig. 4); flagellum infuscate, segments 210=2S: x asslongas wide s,s saeco . H. americana new species - Petiole distinctly elongate (1. 5-3.0 x as long as wide) with a dorsal shield (like Figs. 10, 11); cubital hair-line of forewing strongly sinuate below speculum (like Fig. 12); scape distinctly narrowing at apex (Fig. 8); flagellum pale, seg- ments 3.5 (apical segment 5.0) as long as wide and slightly narrowed medially . H. huggerti new species Holcopelte americana, NEw SPECIES Figs. 1-5 Diagnosis.—Petiole small and conical, without a dorsal shield; cubital hair-line of forewing almost straight; reticulation on median thoracic dorsum engraved; propo- deum with four plicae; ratio width/length of dorsellum = 2.4/1.0. Female: antennal scrobes joining before reaching vertical line of frontal fork; scape 5.0 as long as wide with 6-7 setae along ventral edge. Male: scape apically expanded; flagellum infus- cate; flagellar segments 2.0-2.5 x as long as wide; head transverse in frontal view. Description.— Female: Scape yellowish- white, more or less infuscate in apical part, remaining antenna brown. Face, frons and occiput golden-green, occasionally bluish. Thorax varying from blackish to brownish, usually also with a slight golden-green tinge VOLUME 91, NUMBER | 61 er a wi PG hc % yes Sess : ‘ Figs. 1-9. Holcopelte spp. 1, ° H. americana new species, habitus. 2, Same, lateral view of head. 3, Same, frontal view of head. 4, ¢ H. americana, lateral view of head. 5, Same, frontal view of head. 6, 2? H. huggerti new species, lateral view of head. 7, Same, frontal view of head. 8, ¢ H. huggerti, lateral view of head. 9, Same, frontal view of head. Scale representing 0.50 mm. 62 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 10-14. 10-12, 2 Holcopelte huggerti new species. 10, Side view of petiole. 11, Dorsal view of propodeum and petiole. 12, Base of forewing. 13-14, 2 Jonympha carne (Walker). 13, Frontal view of head. 14, Left mandible. Scale to the left representing 0.25 mm (Fig. 10), and 0.50 mm (Figs. 11-13); scale to the nght representing 0.10 mm (Fig. 14). (lacking in some specimens). Forecoxa brown, mid- and hindcoxae pale (usually whitish, but occasionally yellowish), re- maining parts of legs varying from yellowish to predominantly infuscate. Wings hyaline or weakly infuscate, veins pale. Petiole yel- lowish-white. Gaster with same color as thorax. Length of body: 0.9-1.3 mm. Fla- gellum with all five segments free, segment I 1.5-2 x, Il about 2 x, If] and IV 2.3-2.5 x, and V about 3 x as long as wide. Ratios HE/ MS/MO = 4.2/1.0/3.0. Malar space 1.5 x as wide as width of scape in widest part. Frontal fork V-shaped. Inner orbit of eye with one row of setae. Ratios POL/OOL/ POO = 8.4/3.6/1.0. Entire occipital margin with a fine and sharp edge, extending from eye to eye. Ratio WH/WT = 1.3/1.0. Pro- notal collar without transverse carina. An- terior part of mesoscutum with reticulation slightly raised above the surface, with trans- verse meshes. Median and posterior meso- scutum with finer and engraved reticula- tion, meshes isodiametric or slightly elongate. Notaular depressions clearly de- limited from remaining mesoscutum, nar- row and extending along posterior half of mesoscutum. Midlobe of mesoscutum with a fine furrow, usually extending along pos- terior-median half of mesoscutum, but oc- casionally shorter. Scutellum with fine and engraved reticulation, slightly raised along sides and hind part. Meshes slightly to very elognate in median part, isodiametric along sides and hind part. Furrow on median mesoscutum continuing on scutellum, ex- tension varying from anterior to entire scu- tellum. Dorsellum 2.4 as wide as long, with irregular sculpture and usually divided into three concave areas. Forewing round- ed, hindwing with apex varying from round- ed to sharply pointed. The shape of apex of hindwing is linked to the size of the speci- men: small specimens have a pointed apex,while large specimens have a rounded apex. Speculum developed and closed be- low, cubital hair-line almost straight. Ratios length of marginal/postmarginal/stigmal veins = 7.1/1.0/1.2. Anterior part of pro- podeum with a relatively wide transverse furrow, extending between stigmata. With two pair of plicae, outer pair separating pro- podeal callus from propodeum, inner pair situated half way between outer plica and median propodeum (these are occasionally bifurcate in posterior part), with or without a median carina. With a pair of carinae as- cending from upper corners of petiolar fo- ramen, reaching half way up on propodeum. Propodeal surface usually with weak reticu- lation and/or other weak sculpture. Pro- podeal callus with two setae. Petiolar fora- men triangular to rounded in shape. Petiole conical, at most as long as wide, usually VOLUME 91, NUMBER 1 slightly transverse. Gaster ovate and mod- erately acuminate posteriorly, about 1.2 x as long as thorax + propodeum. Male: Color like female, except all coxae brown and head with brighter color. Length of body: | mm. Scape apically expanded, flagellar segments I-III about 2 x, [V and V about 2.5 x as long as wide. Malar space as wide as width of scape in widest part. Ratios HE/MS/MO = 3.4/1.0/2.6, POL/OOL/ POO = 14.0/6.0/1.0, WH/WT = 1.5/1.0. Gaster slightly longer than thorax + pro- podeum. Otherwise as in female. Type material.—Holotype 2 labelled: “USA: West Virginia, Greenbrier Co., 10 miles E. Richwood, Summit Lake, 29.vu1.1983, leg. L. Huggert,” in LUZM. Paratypes: 2 2 with same label as holotype; 1 9 “USA: West Virginia, Greenbrier Co., Richwood, Summit Lake, 27.vu1.1983, leg. L. Huggert’’; 2 2, 1 6 “USA: West Virginia, Pocahontas Co., Falls of Hills Creek, 22.vii.1983, leg. L. Huggert’; 3 2 “USA: West Virginia, Pocahontas Co., Falls of Hills Creek, Monongahela Forest, 26.vii.1983, leg. L. Huggert”’; | 6 “USA: West Virginia, Pocahontas Co., Dogway Road, Monon- gahela Forest, 26.vi1.1983, leg. L. Huggert”’; 1 2 **Canada: Quebec, Bouchette, Lac Rod- dick, 12.1x.1982, leg. L. Huggert.”” These paratypes deposited: 3 2, 1 6 in CH, 5 9, 1 6 in LUZM, 1 2 in USNM; 1 ¢ **N.S.C.B.H.N.P., Skyline Trail, 25.vii. 1983,” “Birch PG 635802, B.R.I. Survey”; 1 2 “Prince Edward I., Sand Banks Prov. Pks., 25.vii.1982, L. Masner”’; 1 2 “St. Law- rence Is. Nat. Park, Ontario, Thwartway Is.,” “19 vii.1976, L. Masner, Code 4212-7”; fi- nal three paratypes in CNC. Distribution.— Canada (Nova Scotia, Ontario, Prince Edward Island, Quebec) and the United States (West Virginia). Holcopelte huggerti, New SPECIES Figs. 6-12 Diagnosis. —Petiole quadrate (2), or 1.5— 3.0 as long as wide (4) with a thin dorsal shield that covers petiolar foramen; cubital 63 hair-line strongly sinuate below speculum; reticulation on thoracic dorsum raised; midlobe of mesoscutum separated from scutellum by a narrow transverse furrow; propodeum with two plicae; ratio width/ length of dorsellum = 5.9/1.0. Female: an- tennal scrobes never meet; scape 8.0 as long as wide, with 3-5 setae along ventral edge. Male: scape narrowed apically; flagel- lum pale; flagellar segments 3.5 x (apical segment 5.0 x) as long as wide and slightly narrowed in median part; head about as high as wide in frontal view. Description.— Female: Scape yellowish- white, remaining antenna infuscate. Frons and occiput dark brown with a weak golden or golden-green tinge. Thorax dark brown with a golden tinge. Forecoxa brown, mid- and hindcoxae pale (usually white, but oc- casionally yellowish), remaining parts of fore- and midlegs usually more or less in- fuscate (especially femora), hindleg usually predominantly pale. Wings hyaline with pale veins. Petiole yellowish-white. Gaster with same color as thorax. Length of body: 0.9- 1.3 mm. Scape narrow with 3-5 setae along ventral edge. All five flagellar segments free, segment I about 2.8 x, HH, II and V about 3.8, and IV about 3.4 as long as wide. Ratios HE/MS/MO = 4.2/1.0/2.4. Malar space about 2.5 x as wide as width of scape in widest part. Frontal fork V-shaped. Inner orbit of eye with one row of setae. Ratios POL/OOL/POO = 13.0/10.0/1.0. Entire occipital margin with a fine and sharp edge, extending from eye to eye. Ratio WH/WT = 1.5/1.0. Pronotal collar without trans- verse carina. Mesoscutum and scutellum with fine but distinctly raised reticulation, stronger along sides and hind part of scu- tellum. Meshes elongate in median part of both mesoscutum and scutellum, otherwise isodiametric. Median furrow on_ thorax weak, extending from posterior ' of meso- scutum to anterior 3 of scutellum, some- times absent on mesoscutum. Notaular depressions like americana. Midlobe of mesoscutum separated from scutellum by a narrow furrow. Dorsellum short, 5.9 as 64 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON wide as long, with surface convex. Shape of wings like americana. Speculum present and closed below, cubital hair-line strongly sin- uate below speculum. Ratios length of mar- ginal/postmarginal/stigmal veins = 7.4/1.0/ 1.3. Anterior part of propodeum with a rela- tively wide transverse furrow, extending be- tween stigmata. Only one pair of plicae pres- ent, situated in level with stigmata. Also with a pair of carinae ascending from upper corners of petiolar foramen, reaching about half way up on propodeum. Propodeal sur- face otherwise smooth. Propodeal callus with two setae. Petiolar foramen quadrate, upper margin rounded. Petiole quadrate, with a thin dorsal shield anteriorly, shield covering petiolar foramen. Gaster ovate, more acuminate than americana posteriorly, about 1.3 as long as thorax + propodeum. Male: Color like female, except entire an- tenna pale and all coxae infuscate. Length of body: 1.1-1.4 mm. Scape narrowed api- cally. Flagellar segments I-IV about 3.5 x, and V about 5.0 as long as wide. Malar space 1.2x as wide as width of scape in widest part. Ratios HE/MS/MO = 2.9/1.0/ 2.0, POL/OOL/POO = 15.0/7.2/1.0, WH/ WT = 1.3/1.0. Median furrow on thorax extending along entire scutellum in some specimens. Petiole 1.5—3.0 x as long as wide. Gaster slightly longer than thorax + propo- deum. Otherwise as in female. Type material.—Holotype @ labelled: “USA: West Virginia, Greenbrier Co., 10 miles E. Richwood, Summit Lake, 29.vu1. 1983, leg. L. Huggert” in LUZM. Paratypes: 2 2 with same label as holotype; | ° “USA: West Virginia, Pocahontas Co., Falls of Hills Creek, Monongahela Forest, 26.vu.1983, leg. L. Huggert”; 1 @ “Peru: Huanucu, To- cache, 2.11.1984, leg. L. Huggert”’; 1 2 “‘Peru: Cusco, Machu Pichu, 21.xii.1983, leg. L. Huggert”; 2 2 6 6 “Peru: Cusco, Aqua Cal- iente, 28.x1i.1983, leg. L. Huggert’’; 1 2 “Ec- uador: Napo, Lumbaqui, 10-1 1.111.1983, leg. L. Huggert”; 1 9 “Ecuador: Pichin, Rio Pal- enque, 4.ii.1983, leg. L. Huggert”; 2 2 “Ec- uador: Pichin, Puerto Quito, 2.111.1983, leg. L. Huggert’’; 4 2 2 6 in CH, 623 6in LUZM, 1?1¢6in USNM. Distribution.—The United States (West Virginia), Ecuador and Peru. Genus Jonympha Graham Tonympha Graham, 1959: 199. Type- species: Entedon ochus Walker, 1839: 21, by original designation. Diagnosis.— Mandibles multidentate, long and narrow (Fig. 14); antennal scrobes never meet (Fig. 13); frons below fork with fine reticulation at least in some places; an- tenna with two small and discoid anelli; male flagellar segments with hairs placed evenly (not only a basal whorl as in Holcopelte), notaular depressions shallow and not de- limited from remaining mesoscutum; pro- podeum smooth and shiny without plicae; male gaster with a pale subbasal spot. Remarks.-The monophyly of Jonympha is shown through the following synapo- morphies: 1) mandibles long and narrow; 2) male gaster with a pale subbasal spot. For identification and description of the species see Graham (1959: 199-200). The key in Graham can be supplemented with the following character: female /. carne with 2-5 setae on propodeal callus, 7-9 setae in ochus. Tonympha carne (Walker) Figs. 13-14 Entedon carne Walker, 1839: 123. Ionympha carne (Walker), Graham, 1959: 200. Material. —CANADA: | 2 Ontario, Ron- deau Prov. Park, 26.vili. 1982. ECUADOR: 1 ° Pichin, Tinalandia, 800 m, 7.11.1983. PERU: 1 2 Cusco, Ollantaytambo, 19.xi. 1983: 1 2 Cusco, Machu Pichu, 21.xii.1983. Allleg. L. Huggert (1 2 in CH, 3 ?in LUZM). Type material of E. carne in BMNH (not seen). Remarks. — Most European specimens of this species have dark coxae and remaining parts of legs predominantly infuscate. Three VOLUME 91, NUMBER 1 of the females from the Americas have mid- and hindcoxae and remaining hindleg pale. However, there are some European speci- mens with hindcoxa pale, and one of the females from Peru has all legs entirely in- fuscate. The infuscation of the legs frequent- ly shows a high degree of intraspecific vari- ation in many species of Eulophidae and I do not regard this difference in color be- tween European and American specimens as a species character. American specimens of I. carne have, on the average, a weaker reticulation on the thoracic dorsum than European specimens. The midlobe of meso- scutum is usually partly, and the scutellum is predominantly, smooth and polished in American specimens. However, in some species of Eulophidae with a Holarctic dis- tribution, e.g. Chrysocharis prodice (Walk- er), American specimens tend to have a weaker reticulation on the thoracic dorsum (Hansson 1987). This tendency becomes even more apparent in specimens from the southern Nearctic region. Distribution.—West Palearctic (Boucek & Askew 1968), Canada (Ontario), Ecuador and Peru. Jonympha carne was previously not recorded from the Americas. Tonympha ochus (Walker) Entedon ochus Walker, 1839: 21. Ionympha ochus (Walker), Graham, 1959: 200. Material CANADA: 1 @ British Co- lumbia, N. Vancouver, 31.viii.1960, S. M. Clark (CNC). This specimen agrees well with European specimens. Type material of /. ochus in BMNH (not seen). Distribution.—West Palearctic (Boucek & Askew 1968) and Canada (British Colum- 65 bia). Jonympha ochus was previously not recorded from the Nearctic Region. ACKNOWLEDGMENTS I am particularly grateful to Lars Huggert (Lund, Sweden) who collected the majority of the material forming the base of this pa- per. My thanks also to R. Danielsson (LUZM), M. E. Schauff (USNM) and C. M. Yoshimoto (CNC) for loan of material. LITERATURE CITED Ashmead, W. H. 1894. Report on the parasitic Cy- nipidae, part of the Braconidae, the Ichneumon- idae, the Proctotrypidae, and part of the Chalcidi- dae. Part II. J. Linn. Soc. (Zool.) 25: 108-188. 1904. Classification of the chalcid flies or the superfamily Chalcidoidea, with descriptions of new species in the Carnegie Museum, collected in South America by Herbert H. Smith. Mem. Carneg. Mus. 1(-IX): 225-551. Bouéek, Z. 1969. Descriptive and taxonomic notes on ten, mainly new, species of West Palearctic Eulophidae (Hymenoptera). Acta Entomol. Mus. Natl. Pragae 38: 525-543. Boucek, Z. and R. R. Askew. 1968. Palearctic Eu- lophidae (excl. Tetrastichinae). Index of Ento- mophagous Insects. Le Francois, Paris. 254 pp. Erdos, J. 1958. Eulophidae in Hungaria recenter de- tectae. Acta Zool. Hung. 3: 205-223. Forster, A. 1841. Beitrage zur Monographie der Pteromalinen Nees. Aachen. 46 pp. 1856. Hymenopterologische Studien. II. Chalcidiae und Proctotrupii. Aachen. 152 pp. . 1861. Ein Tagin den Hochalpen. Progr. Real- schule Aachen fur 1860/61. 44 pp. Graham, M. W. R. de V. 1959. Keys to the British genera and species of Elachertinae, Eulophinae, Entedontinae and Euderinae (Hym., Chalcidoi- dea). Trans. Soc. Br. Entomol. 13: 169-204. Hansson, C. 1987. Revision of the New World species of Chrysocharis Forster (Hymenoptera: Eulophi- dae). Entomol. Scand. Suppl. 31: 1-87. Thomson, C. G. 1878. Hymenoptera Scandinaviae. Vol. 5. Lund. 307 pp. Walker, F. 1839. Monographia Chalciditum. I. Lon- don. 333 pp. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 66-70 NATURALLY OCCURRING HOST SITES FOR XYLOPHILIC CECIDOMYIIDAE (DIPTERA) Emity A. Rock AND DALE JACKSON (EAR) Wayne General and Technical College, 10470 Smucker Road, Orrville, Ohio 44667; (DJ) Department of Biology, The University of Akron, Ohio 44325. Abstract. — Existing data on host sites for xylophilic Cecidomyiidae larval development are derived from observations of sawn trees and logs. Field studies of hardwood trees at three sites in northeast Ohio show that strong winds and animal activity break live branches of suitable size and thus expose vessels of =75 u which are used for larval development. Such damaged branches can support large populations of midges. Larvae develop in both the proximal and distal sides of the break. Branches remain suitable for oviposition and larval development for one to several months, depending on branch diameter and weather conditions. Key Words: Xylophilic Cecidomyiidae use freshly ex- posed vessels of hardwoods as a larval hab- itat. Relationship between host selection and vessel diameter has been previously re- ported (Rock and Jackson 1985, 1986). As with that of other investigators (Kieffer 1900, Brues 1922), our initial encounter with these cecidomyiids began when we observed fe- males swarming on the cut surfaces of logs and stumps of trees exposed in logging op- erations. Although we also used cut logs during earlier studies, we speculated on the availability of such niches under natural conditions. Our current investigation ex- amines 1) the role of strong winds and an- imal activity in exposing larval niches, 2) utilization of vessels in live branches that remain on the tree, 3) the number of larvae that one branch can support and 4) the length of time after exposure that a branch remains suitable for use. METHODS Storm and squirrel damage.—Two sites in northeast Ohio were monitored for two wood loving, tree damage, squirrel activity years (1985 and 1986) for live, broken branches. One site was a small island of approximately 7500 square meters located near a lake shore with five mature oak (Quercus alba L.) and one mature ash (Frax- inus americana) trees (known xylophilic cecidomyiid hosts (Rock and Jackson 1986)). The other was part of a large sub- urban garden with a study area of 5625 square meters with five mature oak and three mature ash trees. The areas below the trees were mowed regularly to facilitate collection of fallen twigs and branches. We recorded the diameter of only the live branches be- cause previous studies have shown that dead wood is not a suitable larval habitat. Some larger branches were kept at the field site to monitor their use by cecidomyids. The suitability of the branches for larval devel- opment was verified by subsequent collec- tion of larvae from the branch ends. Squirrels were active at the field sites and numerous live host tree branches were col- lected that squirrels had severed by chew- ing. Branches broken by squirrels were VOLUME 91, NUMBER 1 67 YO —-———— ——— ay i 70 60 f 50 ie ; 4 40 ; a 30 8 fs B 20 10 SB, 0 rae ieee oe cele — es rr ae Fer ee rt Fe a eee 5 [a east 0.0 0.5 15 2.0 2.5 3.0 15°50 55 60 65 70 7.5 8.0 85 9.0 9.5 100 BRANCH DIAMETER (inm) Fig. 1. Number and size of live branches broken from hardwood hosts at test sites. Branches <2 mm d not included in collection. readily identified by teeth marks. Branch diameters were recorded. Suitability of proximal section of broken branches.—In earlier studies, we had for convenience used only the severed branch sections. The current study investigates the use of the intact basal parts as their use would double the number of potential larval de- velopment sites. We cut a total of fifty-nine branches rang- ing from 13 mm d to 30 mm d from ash, oak and elm trees at field sites in Akron and Orrville, Ohio. The severed branches were placed at the base of the tree to confirm the presence of females. Branch ends on the trees were left exposed for ten days to allow fe- males to oviposit. After this time, the ends were covered by a plastic cup; the branch and cup were then covered with a nylon bag to hold the cup in place. Rain wetting the branches stimulated the emergence of lar- vae. We recorded the number of larvae that fell into the cups without identifying them to species. The branches were uncovered for three days to allow for possible further ovi- position; the cups were then replaced. This procedure was repeated for up to seven weeks. Potential larval yield from small branch- es.—We had previously determined that host branches of 6 mm or greater in di- ameter usually contain xylem vessels of =>75 u d, which are suitable for larval develop- ment (Rock and Jackson 1986), but never recorded the number of midges that emerged from indvidual twigs. Twenty sassafras (Sassafras albidum (Nutt.) Nees) branches with diameters varying from 3 mm to 20 mm were offered to a population of 300 caged Xylodiplosis longistylus Gagné fe- males for one week. The branches were held at room temperature in plastic bags for 14 68 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON NUMBER OF BRANCHES COLLECTED 40 45 5.0 5.5 60 65 70 75 80 85 9.0 BRANCH DIAMETER (mm) Fig. 2. days. Each was then soaked in a separate container of water, and the number of larvae emerging from both exposed ends was re- corded. Period of suitability of cut branches for larval development.—To establish if there was a connection between length of use of a severed branch and branch diameter, we monitored larval emergence in branch sam- ples of various diameters from the field sites. Unidentified xylophilic larvae emerged from vessels in approximately 14 days after ovi- position if the branch was soaked in water (Rock and Jackson 1985). When a wood sample ceased to yield larvae, we assumed it had lost its attractiveness to females ap- proximately two weeks earlier. We then compared duration of attractiveness to sample diameter. RESULTS AND DISCUSSION Strong winds create a source of suitable Oviposition sites. Although the majority of Number and size of live branches broken from host trees by squirrel activity. the live branches broken from host trees during an entire summer are <5 mm d (Fig. 1), high winds break off some branches of 6 mm d or greater, which contain suitably sized vessels. Branches with diameters rang- ing from 15 mm to 35 mm were occasion- ally broken throughout the summer. At oth- er locations we observed major damage due to lightning and heavy snow that also cre- ated sources of exposed vessels. As dis- cussed below, the availability of larger branches is very important as they can sup- port several generations of gall midges each summer. Squirrels expose the ends of live branches when they feed and build their nests (Short- en 1954). Although most of the branches are less than 6 mm d (Fig. 2), some larger branches are severed and can serve as suit- able sites for midge development. Gray squirrels (Sciurus nigra) are known to build nests in many of the hardwoods that are also hosts to xylophilic midges. Uhlig (1955), in VOLUME 91, NUMBER 1 NUMBER OF LARVAE COLLECTED (Thousands) 69 20.0 9.0 10.0 BRANCH DIAMETER (mm) Fig. 3. an extensive study of the gray squirrel in West Virginia, noted the use of oak, elm, and hickory, and less frequently willow, walnut, and sassafras, in nest building. Wal- nut and oak trees at our field sites contained squirrel nests. Leaf nests built of twigs and leaves from the tree in which they are lo- cated have been noted from early spring throughout the summer months. Juvenile squirrels are the primary leaf nest builders, and the actual number of nests is directly proportional to the rearing success of the spring and summer litters (Uhlig 1955). One juvenile may build more than one nest throughout the summer and early fall. Thus, with several juvenile squirrels in an area, a supply of exposed branches exists for sev- eral months. Fig. 3 summarizes a laboratory study, us- ing sassafras, which shows a direct correla- tion between branch diameter and the num- ber of larvae collected. We previously determined that larvae require vessels of Relationship of number of larvae collected to branch diameter: Sassafras albidum. =75 ud for development and that it is pos- sible to predict the number of potential lar- val habitats based on the number of vessels/ mm? in the sapwood. The 6 to 8 mm d branches are suitable for larval develop- ment but each yielded only a few larvae. However, the 20 mm d sassafras sample had approximately 1850 vessels =75 » d/mm? in each end, and it yielded over 1700 larvae. Preliminary field data indicate that actual utilization of vessels is much less than in laboratory samples. A 12 mm dash branch, for example, yielded 6 larvae, and a 50 mm d branch yielded 100 larvae during 12 days of field collecting. In natural conditions, small, broken branches are available to the flies throughout the summer and can sup- port small overlapping populations. Al- though large branches are available less fre- quently, they permit rapid population increases and provide larval niches for ex- tended periods of time. The live basal sections of severed branch- 70 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 1. Relationship of host sample diameter to length of attractiveness. Pengtn c8) Attrac- Diameter tiveness Host Plant (cm)* (Weeks) Juglans nigra L. 35 15 Fraxinus americana 22 10 Carya ovata (Mill.) K. Koch 21 12 Quercus alba L. 14 14 Ulmus americana L. 12 14 Populus deltoides Bartr. 12 10 U. americana 11 1 Salix babylonica Marsh 9 8 Sassafras albidum (Nutt.) Nees 7 8 F. americana 5 8 F. americana 2to4 8 F. americana 1 to 2 6 F. americana 0.6 to | 4 * All samples 25 to 35 cm long. es are attractive to females and suitable for larval development. Larvae emerged from most of the attached broken branches, the largest number collected at one time being 30 from a 30 mm ash branch. No larvae emerged after the sixth week of exposure, and we assumed branches had ceased to at- tract female midges. The time period during which midges uti- lize a cut branch (Table 1) is directly pro- portional to its diameter. As the life cycle of most xylophilic species is approximately 4 weeks, during the summer in northeast Ohio it is possible for one 35 cm d log to support three generations of gall midges. Major factors that limit the period of at- tractiveness are speed of decay and rainfall. General observations over several summers indicate that abnormally wet periods pro- mote fungal growth on exposed branch ends and so reduce the length of time that a log is suitable. Weather records, however, 1n- dicate that summers with increased shower activity are associated with more frequent episodes of high winds which lead to an increase in the number of severed branches (Robert Thompson, personal communica- tion, National Weather Service, North Can- ton, Ohio 1988). LITERATURE CITED Brues, C. T. 1922. Some hymenopterous parasites of lignicolous Itonididae. Proc. Am. Acad. of Arts and Sci. 57: 263-287. Kieffer, J. J. 1900. Monographie des Cecidomyides d’Europe et d’Algérie. Ann. Soc. Entomol. Fr. 69: 181-472 and pls. 15-44. Rock, E. and D. Jackson. 1985. The biology of xy- lophilic Cecidomyiidae (Diptera). Proc. Entomol. Soc. Wash. 87: 135-141. 1986. Host selection in xylophilic Cecido- myiidae (Diptera). Proc. Entomol. Soc. Wash. 88: 316-319. Shorten, M. 1954. Squirrels. Collins, London. 212 pp. Thompson, R. 1988. Personal communication. Na- tional Weather Service, North Canton, Ohio. Uhlig, H.G. 1955. The gray squirrel: Its life history, ecology, and population characteristics in West Virginia. Pittman-Robertson Project 31-R. Con- servation Commission of West Virginia. 175 pp. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 71-80 TWO NEW TERRESTRIAL ISOPODA (ONISCIDEA) FROM CORALLINE CAYS OF VENEZUELA’S CARIBBEAN COAST Maurizio G. PAOLETTI AND BENJAMIN R. STINNER (MGP) Department of Biology, University of Padova, 35100-Padova, Italy, Visiting Professor, Department of Entomology, The Ohio State University; (BRS) Department of Entomology, The Ohio State University, Wooster, Ohio 44691. Abstract.—Two terrestrial, halophilous isopods, Metastenoniscus neotropicalis n. sp. (Stenoniscidae) and Armadilloniscus caraibicus n. sp. (Scyphacidae) are described from coralline cays of Venezuela’s Caribbean coast. Buchnerillo litoralis Verh. and Stenoniscus pleonalis Aubert and Dollfus are reported from the Caribbean region. Key Words: Isopoda, Oniscidea, Neotropical, Caribbean, Metastenoniscus neotropicalis n. sp., Armadilloniscus caraibicus n. sp., Buchnerillo litoralis, Stenoniscus pleonalis Knowledge of the Venezuelan terrestrial isopoda fauna is reported in a few post- worldwar papers: Brian 1957, Vandel 1952, 1968, 1972, Andersson 1960, Mulaik 1960, Strinati 1971, Schultz 1971, 1983, 1984, following the classic studies of Van Name 1936, 1940, 1942. We have followed in gen- eral the terminology proposed by Holdich 1984, Holdich et al. 1984. Our work is mostly based on SEM (Scanning Electron Microscopy). The two new species that we describe be- long to the halophilous seacoast fauna. The tergites covered by longitudinal ribs suggest that these species belong to the eco-mor- phological category of creepers (Schmalfuss 1984, Paoletti 1987). ONISCOIDEA Stenoniscidae Metastenoniscus neotropicalis n. sp. Type locality. Coralline key (Cayo) of Pla- yuela, Parque Morrocoy, estado Falcon, Venezuela. 28 females and 12 males were collected in soil litter of Coccolobis uvifera (L) Jacq. (Polygonaceae), by M. G. Paoletti on January 1, 1986 (Paoletti 1988). Male holotype, allotype and paratypes are located in the M. G. Paoletti collection; 3 female paratypes are deposited in the Mu- seo Zoologico of the Padova University; 3 female paratypes are located in the general- collection of Instituto Museo de Zoologia Agricola, Universitad Central de Venezue- la, Maracay, Aragua, Venezuela. Diagnosis. This new species is similar to Metastenoniscus osellai Taiti and Ferrara, 1981. The smaller body is more cylindrical and holds less enlarged epimera. The telson is shorter and three-lobed. Dimensions: length males: 1.4-1.86 mm, mean 1.57 mm; width males: 0.42-0.56 mm, mean 0.50; length females: 1.48-—2.42 mm, mean 2.14 mm; width females: 0.48-0.84 mm, mean 0.71 mm (Figs. 1, 2). Body ornamentation: consists mainly of subcircular plaques (Figs. 4A, C, D); body shape subparallel, costulated, light violet colored; pereonites I-IV without medial rib (Figs. 3A, C). 72 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 0.57274 0.532 4 E E 0492 B= mo) = a 0.452 4 y -13.8493+01428-x r= 0.2107 p-01942 a 0412 T T T if T 136 150 1.64 178 192 Length (mm) Fig. 1. Regression for male dimensions of Metas- tenoniscus neotropicalis n. sp. Head: with three posterior main tubercles and two median main tubercles (Figs. 3A, B, C); eye with 34 ommatidia; antenna with flagellum consisting of two articles; the sec- ond article bears on its foreword-facing sur- face three aesthetascs (Figs. 3E, F). Anten- nule two-articulated with three apical aesthetascs, the exterior occasionally bro- ken at the base (fig. 4E). Telson: short and distinctly three-lobed (Figs. 3C, D; 4B, D). Uropods: the basis is longer than wide (Fig. 4B), the exopod bears distally a spike of 4—5 aesthetascs (Figs. 4D, F); the endo- pod much longer than exopod (Figs. 4C, D), thickened in the middle, end with a spike of three elements (Fig. 4C). Pleopods: male endopod of first and sec- ond pair are little differentiated (Fig. 5). Affinities. M. neotropicalis n. sp. is dis- tinct from Metastenoniscus osellai Taiti and Ferrara, 1981, described from Bali by the following features: 1. females and males are smaller; 2. epimeral appendages less developed and body smaller; 3. head with a different arrangement of tu- bercles especially in the posterior part: five on M. osellai and three on M. neo- tropicalis n. sp.; 4. pereonites I and II with only a hind ves- tige of medial rib; in M. osellai the per- 0.86 4 y --0.1586 + 03349-x r= 0.8204 p< 0.0001 a 48 0.76 4 0.66 5 Width (mm 0.56 5 046 5 ‘lt Tig 144 1.70 196 2.22 248 Length (mm) Fig. 2. Regression on female dimensions of Me- tastenoniscus neotropicalis n. sp. eonite from | to 4 has medial costa well developed (Taiti and Ferrara 1981 Figs. lGand 2©@): . telson shorter and distinctly three-lobed; 6. uropod exopods shorter and with wider basis. Nn Habitat. Found under Coccolobis uvifera (L.) Jacq. (Polygonaceae) (uva de plaja) lit- ter, eating decayed litter tissue (Paoletti 1987 Figs. 7D, G, H). This isopod was not yet found in the intertidal zone but strictly in the interior part of the coralline cays and above the upper tidal level. Distribution. Discovery of M. neotropi- calis n. sp. considerably enlarges the known geographic range of the family Stenonisci- dae. In fact Metastenoniscus is now repre- sented not only in the Oriental region (Bali and the Andaman Islands) but also in a Pan- tropical belt including the Caribbean region. That they only now have been discovered in the Caribbean region is probably due to their small size. Discussion. Stenoniscus pleonalis Aubert and Dollfus, 1890 (sensu Vandel 1962) is easily distinguishable from Metastenonis- cus oSellai (Taiti and Ferrara, 1981) and from M. neotropicalis n. sp. It was reported only rarely beyond the Mediterranean region and in the Neotropical region only twice (Vandel 1968, Schultz 1972). We collected Stenon- iscus pleonalis Aubert and Dollfus (sensu VOLUME 91, NUMBER 1 Fig. 3. ee ee Metastenoniscus neotropicalis n. sp. Female: A dorsal view, B head frontal view, C lateral view, D 73 100umS —e 2 pleon and telson, E and F second antennal flagellum article from dorsal and ventral views. Vandel 1962) in Florida Cays, Tavernier Creek, on rocky beach: in the soil, under Posidonia and on coconut drift in the beach, M. G. Paoletti, October 26, 1987. This lat- ter observation is interesting. Under the Posidonia drift in the same location was col- lected one female of Buchnerillo litoralis Verhoeff, 1943, which represents a first re- cord for the Neotropical region (Schultz and Johnson 1984), and Vandeloscia culebre (Moore, 1901), Ty/los niveus Budde-Lund. and Armadilloniscus ellipticus (Harger, 1878). ONISCOIDEA Scyphacidae Armadilloniscus caraibicus n. sp. Type locality. Coralline cays (Cayos) of Parque Morrocoy, estado Falcon, Venezue- la: Cayo Sombrero, one gravid female 31.X11.1985; Cayo Cico, 8 males, 16 fe- males, 7 gravid females, 3.1.1986 were col- lected by M. G. Paoletti. Isopods were col- lected in the intertidal zone under coralline rocks lying on the sandy beach at Cayo Cico; at Cayo Sombrero the gravid female was 74 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON NW Aho >, aoe | TRS: ORION Fig. 4. Metastenoniscus neotropicalis n. sp. Female: A I and II pereonites, B ventral view of hind pereon, pleon and uropods, C uropodal endopods, D below view of pleon and uropods, E second article of antennulae with aesthetascs, F uropodal exopod. found among small woody debris on the coarse sandy beach (Paoletti 1988). Male holotype, allotype and paratypes are located in M. G. Paoletti collection; 3 fe- male paratypes are deposited in the Museo Zoologico of the Padova University; 3 fe- male paratypes are deposited in the general- collection of Instituto Museo de Zoologia Agricola, Universitad Central de Venezue- la, Maracay, Aragua, Venezuela. Diagnosis. Body elliptical, dull brown, covered with prominent ridges of tubercles on the head, forming ribs on the pereon and pleon (Figs. 10, 11). Ornamentation con- sists of circular plaques (Figs. 11F, H) and digitiform trichomes. VOLUME 91, NUMBER | Fig: 5: Metastenoniscus neotropicalis n. sp. Male, A and B pleopod I endopods, C pleopods I exopod, D pleopods II endopod. Armadilloniscus caraibicus n. sp. Male, E pleopods I, F pleopods II, G VII male pereopods. Dimensions: length males: |.92—2.46 mm, mean 2.27; width males: 0.84-1.26 mm, mean 1.05 (Fig. 6); length non gravid fe- males: 1.25—3.9 mm, mean 2.34; width non gravid females: 0.75-1.92 mm, mean 1.095 (Fig. 7); length gravid females: 2.85—3.36 1.30 4 y - 0.0437 + 0.4449-x = 0.37 ASI p= 0.45 4 E E = 1064 2 Ss 0944 a 0.824 T a af = a5 | T. st 189 2.04 219 2.34 249 Length (mm) Fig.6. Armadilloniscus caraibicus n. sp. Regression of male dimensions. mm, mean 3.0; width gravid females: 1.35— 1.56 mm, mean 1.44 (Fig. 8); 6-9 pulli per gravid female were counted (Fig. 9). Head: with lateral lobes truncated at tip and median lobe pointed, with three main tubercles in the posterior and two in the 23-4 y = 0.0302 *0.4546:x ol 0.84 1 a9) P<0.000 ‘ E E = 154 Sl = 11 a 07-4 a =| aE T a i 14 19 27 35 43 Length (mm) Fig. 7. Armadilloniscus caraibicus n. sp. Regression of non gravid female dimensions. 76 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 1584 : 3.38 4 > - a a y = 2.1322+01178-x r= 0.564 a 152 < 3244 p-00912 D 2 2 E ® E 2 - 4.464 ® 310-4 = c 3 5 ‘ 0.7228 +0,2364 = 140 4 y & 296-4 r?- 0.205 5 ee p= 0.397 a a 1.344 2824 T T T 7 T 2.82 296 3.10 3.24 3.38 Length (mm) Pulli per female Fig. 8. Armadilloniscus caraibicus n. sp. Regression Fig.9. Armadilloniscus caraibicus n. sp. Regression of gravid female dimensions. of gravid female length and pulli per female. BD he - 100 pm Fig. 10. Armadilloniscus caraibicus n. sp. Female: A and B pleon; C and D head and pereon. VOLUME 91, NUMBER | Fig. 11. Armadilloniscus caraibicus n. sp. Female: A-D body surface in different perspectives; E head; F pereonite ornamentation; G and H pleon and peculiar tubercle ornamentation. anterior part (Figs. 1OC, D, 11A, C, D, E). (Figs. 12C, D); antennule characteristically Eye with 4—5 visible ommatidia; antenna, three-articulated and apically bifurcated; it without enlargements, holds four flagellar bears aesthetascs (5 on specimens exam- articles, the second with three and the third ined) (Figs. 12A, B). with two aesthetascs on the inferior surface Body ornamentation: pereon with four 77 78 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON : c ; » Fs, Fig. 12. Armadilloniscus caraibicus n. sp. Female A-F. Male G and H. A and B antennules; C and D antennal four-article flagellum in dorsal and ventral view; E forelegs; F maxillipeds; G and H male pleopods I. VOLUME 91, NUMBER 1 main medial ridges and four shallow lateral ridges of tubercles (Figs. 10C, 1 1 A-E); pleon supporting two main ridges of tubercles (Figs. 1OA, B; 11B, D, G). Uropods: endopod somewhat longer than basis, holding a distal spike of 3-4 aesthe- tascs (Figs. 10B, 11G). Male pereopods I and VII slightly modified (Fig. SG). Pleo- pods: male pleopods vary little from other members of the genus (Figs. SE-F). Affinities. The development of tubercle ribs on the head, pereon and pleon are the features of Armadilloniscus caraibicus n. sp. by which it can easily be differentiated from other species (Arcangeli 1957, Schultz 1972, 1977, Garthwaite et al. 1985). From the more tuberculated A. coronacapitalis Men- zies, this new species is distinguishable for its smaller size, less tuberculation, differing sculpture on the head, the shape of pleopod I endopod, antennae, and antennules. It can be separated from Armadilloniscus quadri- cornis Vandel, 1971, 1973 by different head structure and ornamentation and by its smaller size. Habitat. The specimens were found un- der coralline rocks on the sandy beach in the intertidal zone. Only one specimen was found between woody debris on the beach. The intertidal habitat seems to be the pre- ferred habitat of the genus 4rmadilloniscus. Distribution and Discussion. Armadil- loniscus caraibicus n. sp. was found on cor- alline cays (cayos) of Parque Morrocoy, Venezuela which extends southward in the Neotropical region the distribution of Ar- madilloniscus. A revision of the described species and of Scyphacidae genera 1s needed to have a better taxonomic and zoogeo- graphic understanding of the group. At pres- ent little is known about the phylogenetic relationships of the Scyphacidae and other neotropical isopods. ACKNOWLEDGMENTS Weare indebted to Carlos Bordon for field collection and discussion. A. L. Dreon, C. Furlan, U. Arezzini, C. Friso assisted with 79 the figures and photographs, C. Britton and F. Ferrara with bibliographical research. F. Purrington revised the text. The Ministero Italiano della Pubblica Istruzione provided financial support. LITERATURE CITED Andersson, A. 1960. South American isopods in the collection of the Swedish State Museum of Natural History. Arkiv for Zool. 12: 537-569. Arcangeli, A. 1957. Il genere Armadilloniscus Ulj. € gli Scyphacidae. Atti Acc. Sc. Torino 91: 1-30. Brian, A. 1957. Descrizione di Neosanfilippia vene- zuelana n. gen., n. sp. di Isopodo terrestre trog- lobio. Ann. Mus. Civ. St. Nat. Genova 69: 352- 360. Ferrara, F. and S. Taiti. 1981. Isopodi terrestri delle isole Adamane. Boll. Mus. Civ. St. Nat., Verona 8: 459-492. Garthwaite, R. L., F. G., Hochberg, and C. Sassaman. 1985. The occurrence and distribution of terres- trial isopods on Santa Cruz Island with prelimi- nary data for the other California islands. Bull. Southern California Acad. Sci. 84: 23-37. Holdich, D. M. 1984. The cuticular surface of wood- lice: A search for receptors. The biology of terres- trial isopods, S. L. Sutton and D. M. Holdich, eds., Zool. Soc. London Symp. 53: 9-48. Holdich, D. M., R. J., Lincoln, and J. P., Ellis. 1984. The biology of terrestrial isopods: Terminology and classification. The biology of terrestrial iso- pods, S. L. Sutton and D. M. Holdich, eds., Zool. Soc. London, Symp. 53: 1-6. Mulaik, S. B. 1960. Contribucion al conocimiento de los Isopodos terrestres de Mexico. Revista de la Soc. Mexicana de Hist. Natural 21: 79-220. Paoletti, M. G. 1988. Life strategies of isopods and “soil invertebrates” in neotropical Venezuela. Monit. Zool. It., Mon. Ser. (In press.) Schmalfuss, H. 1984. Eco-morphological strategies in terrestrial isopods. The biology of terrestrial isopods, L. S. Sutton and D. M. Holdich, eds., Zool. Soc. London Symp. 53: 49-63. Schultz,G. A. 1971. A review of species of the family Scyphacidae in the New World. Proc. Biol. Soc. Wash. 84: 477-488. 1972. Ecology and systematics of terrestrial isopod crustaceans from Bermuda. Crustaceana, Supp. 3, pp. 79-99. . 1977. Terrestrial isopod crustaceans from St. Catherines Island, Georgia. Georgia J. Sci. 35: 151- 158. 1983. Disposition of three species of Onis- coidea from Western Atlantic seashores. Proc. Biol. Soc. Wash. 96: 440-451. 80 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Oniscoidea from Belize, Central America. J. Nat. Hist. 19: 3-14. Schultz, G. A. and C. Johnson. 1984. Terrestrial iso- pod crustaceans from Florida. J. Crust. Biol. 4: 154-171. Strinati, P. 1971. Recherches biospeleologiques en Amerique du Sud. Ann. de Speleologie 26: 439- 450. Taiti, S. and F. Ferrara. 1981. Metastenoniscus osellai genere e nuova specie di isopodo terrestre dell’ isola di Bali. Boll. Mus. Civ. St. Nat., Verona 8: 443- 452. Vandel, A. 1952. Etude des Isopodes terrestres re- coltes au Venezuela par le dr. G. Marcuzzi. Mem. Mus. Civ. St. Nat. Verona 3: 59-203. 1962. Faune de France Isopodes terrestres. Lechevalier, Paris. 1968. I. Isopodes terrestres. Mission zoolo- 1984. Three new and five other species of gique belge aux iles Galapagos et en Ecuador (N. et J. Leleup, 1964-1965). Resultats scientifiques. Premiere partie 1: 35-168. . 1971. Les Isopodes terrestres des iles Rennell et Bellona. The natural history of the Rennell Is- land, T. Wolff, ed., 6: 139-153. . 1972. Les Isopodes terrestres de la Colombie. Studies on the Neotropical Fauna 7: 147-172. . 1973. Les Isopodes terrestres de la Melanesie. Zool. verhandelingen, Leiden 125: 1-160. Van Name, W. G. 1936. The American land and fresh-water isopod Crustacea. Bull. Amer. Mus. Nat. Hist. 71: 1-535. 1940. A supplement to the American land and fresh-water isopod Crustacea. Bull. Amer. Mus. Nat. Hist. 77: 109-142. . 1942. A second supplement to the American land and fresh-water isopod Crustacea. Bull. Amer. Mus. Nat. Hist. 80: 299-329. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 81-87 ANTS ASSOCIATED WITH A COLEOPTEROUS LEAF-BUD GALL ON XYLOPIA AROMATICA (ANNONACEAE) G. WILSON FERNANDES, WILLIAM J. BOECKLEN, ROGERIO P. MARTINS, AND ALEXANDRE G. CASTRO (GWF, WJB) Box 5640, Department of Biological Sciences, Northern Arizona Uni- versity, Flagstaff, Arizona 86011; RPM, AGC Departamento de Biologia Geral, Caixa Postal 2486, ICB/Universidade Federal de Minas Gerais, 30.000-Belo Horizonte-MG, Brazil. Abstract.—We describe the ant fauna (11 species divided into four genera) associated with a coleopterous leaf-bud gall on Xylopia aromatica (Annonaceae) in Minas Gerais, Brazil. Ant occupied galls are significantly larger than are unoccupied galls, and the vari- ance of gall diameters for occupied galls is significantly smaller than is the variance for all galls. There is a significant positive relationship between ant colony size and gall diameter. The six most common species of ants may partition galls according to gall size, as there are significant differences in the diameters of occupied galls among species. At the community level, patterns of co-occurrence of ants are indistinguishable from those expected under a random assortment model. Individual ant species do exhibit non-random patterns of co-occurrence. Key Words: Annonaceae, ants, Brazil, coleopterous gall, community ecology, habitat selection, insect galls, Minas Gerais, resource partitioning, Xy/opia aro- matica Plant galls represent an important re- source for many species other than the gall formers. Owing to their localized concen- tration of nutritive tissues and their marked succulence, galls provide favourable breed- ing sites for a variety of species (Brandhorst 1962, Mani 1964, Shorthouse 1973, Yu- kawa 1983). Galls protect their inhabitants from inclement weather (Felt 1940, Uhler 1951, Sandlant 1979; but see Baust et al. 1979) and natural enemies (Askew 1961, 1980, but see Price et al. 1986, 1987). Use of galls ranges from species that open galls simply to prey upon gall formers and in- quilines to those that depend exclusively on gall tissues for food and shelter. Beauvisage (1883, cited in Mani 1964) applied the term “‘locatari” to species, other than the gall formers, associated with insect galls. Mani (1964) divides the locatari into 33 categories according to their ecological niches. The locatari often represent a vast fauna with the most numerous categories consisting of parasites and predators. For example, Stegagno (1904) reported 177 species associated with cynipid galls on Quercus in Italy; of these, 138 are parasites and predators. Species that inhabit galls after emergence of gall formers and inquilines are the ‘“‘suc- cessori’’ (Mani 1964, Yukawa 1983). Most of these are plant-nesting ants and myr- mecophilous insects, such as aphids and coccids. Others include mites, spiders, thrips, collembola, bees, and wasps (re- viewed by Mani 1964). The most common ant genera associated with old galls are: Camponotus, Cataulacus, Crematogaster, 82 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON GERAIS Sao Paulo we = 2h omine Rio de Janeiro Fig. 1. Colobopsis, Lasius, Leptothorax, and Olop- sis (Walsh 1864, Patton 1879, Wheeler 1910, Ping 1920, Sturtevant 1925, Mani 1964, Torossian 1971a, b). Despite their impor- tance and dominance within the successori, little is known about the ecologies of these ants other than compendiums of ant species associated with particular species of gall for- mers (e.g. Brandhorst 1962, Espadaler and Nieves 1983). Here, we describe the ant fauna associ- ated with a coleopterous leaf-bud gall on Xylopia aromatica (Annonaceae). We de- scribe patterns in distribution and abun- dance of the ant fauna and document ele- ments of habitat selection and resource partitioning. MATERIAL AND METHODS One of us (G.W.F.) collected galls from the Ecological Station of Parapitinga, Trés Marias, Minas Gerais, Brazil during one sample period in mid-May, 1984. The sta- tion is a continental island located in a man- Location of the Ecological Station of Parapitinga in Trés Marias, Minas Gerais, Brasil. made lake (Trés Marias Lake) between lat- itude 18°-19° south and longitude 45°-46° west (Fig. 1). The host tree, Xylopia aromatica, oc- curred in a gallery forest along the northwest border of the island. Only three individuals of approximately six meters tall bearing galls were observed in the area surveyed. The galls are induced by an unidentified species of curculionid (Coleoptera). Gall formers held as vouchers were kept in the author’s collection. The galls are spherical and gla- brous and occur on leaf-buds (Fig. 2). They are green when occupied by the gall former, but turn brown after its emergence. All galls which were on trees (n = 114) were collected and subsequently measured and dissected in the laboratory. Galls that fall on the forest floor are also utilized by ants; however they are not included here because of insufficient sample size. In this paper, we use the word “colony” as any group of ants, composed of workers and larvae (eggs, queen, and pupae if pres- VOLUME 91, NUMBER I e." Fig. 2. 21.5 mm. ent), occupying a single gall. In addition, we defined those galls in which we found only workers as “groups of workers.” STATISTICAL ANALYSES We compared the mean diameter of galls that were occupied by ants with the mean diameter of unoccupied galls using one-way 83 Coleopterous leaf-bud galls found on Xylopia aromatica. Gall diameter ranged from 5.0 to analysis of variance. We also compared the variance of occupied galls with the variance of all galls (including occupied galls) with a Chi-square test (Sokal and Rohlf 1969, page 175). We assumed that our collection con- sisted of the entire population of galls on the island and that the set of occupied galls was a sample from that population. 84 Table |. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Ant species associated with coleopterous leaf-bud galls on Xylopia aromatica (Annonaceae). (2) Individuals Gall Diameter Ant Species Number of Colonies Adult Larvae/Pupae (2) mm Azteca bicolor l 1.0 - 14.5 Azteca sp. 17 28.1 8.9 16.4 Leptothorax wilda 13 5.4 4.6 14.7 Leptothorax sp. | 45.0 37.0 17.0 Pseudomyrmex flavidus 4 0.8 5:5 16.4 Pseudomyrmex sp. A 1 1.0 = 12:5 Pseudomyrmex sp. B 1 1.0 — 15.0 Zacryptocerus pallens 7 2.0 - 12.9 Zacryptocerus pusillus 8 51-3 12.5 15.9 Zacryptocerus sp. A 11 S27, DD) 14.9 Zacryptocerus sp. B l 4.0 4.0 10.0 We compared the mean diameters of oc- cupied galls among the six most common species of ants by one-way analysis of vari- ance. We estimated the relationship between the logarithm of colony size (number of indi- viduals, of all castes, in a gall) and gall di- ameter through linear regression. Finally, we examined patterns of occur- rence within galls. We compared the num- ber of galls that contained 0, 1, 2, and 3 species of ants with those expected under a Poisson distribution. We estimated A, the Poisson parameter, from the sample. We compared the observed distribution to the expected distribution with a Chi-square goodness-of-fit test. RESULTS We collected 114 galls of which 52 (45.6%) were occupied by ants. Eleven species of ants were represented in the sample; the most common were Azteca sp. which oc- cupied 17 galls, Leptothorax wilda which occupied 13 galls, and Zacryptocerus sp. “A” which occupied 11 galls (Table 1). Five species were represented by a single indi- vidual. “Colony sizes” ranged from three individuals to a colony of Zacryptocerus pusillus which contained 129 adults and lar- vae. The internal structure of the galls was highly modified in some cases, unmodified in others (Fig. 3). Modification is defined as any internal architectural departure, such as tunnel and gall wall holes, from the usual spheroid larval chamber. In particular, galls inhabited by Azteca sp. and Zacryptocerus pusillus exhibited extensive modifications when compared to unmodified galls. Galls inhabited by Pseudomyrmex flavidus were unmodified. Modified galls typically con- tained larger colonies than did unmodified galls. Ants typically occupied large galls (Fig. 4). The mean diameter of occupied galls, 15.6 mm, was significantly larger than the mean diameter of unoccupied galls, 13.6 mm (Fi.112 = 12.8, P < 0.001). The variance of diameters of occupied galls was significantly smaller than would be expected if ants were selecting galls at random (x?,,; = 72.1, P < 0.005). There was a significantly positive linear relationship between the logarithm of col- ony size (number of individuals) and gall diameter (Fig. 5). The six more common species of ants fur- ther partitioned the subset of occupied galls according to gall size; mean gall diameters of occupied galls were significantly different among species (F; 5, = 2.74, P < 0.05). Az- teca sp. typically occupied the largest galls (% = 16.4 mm), Zacryptocerus pallens the smallest (¥ = 12.9 mm). We collected 65 colonies distributed VOLUME 91, NUMBER 1 Fig. 3. cm Cross-sections of galls showing internal modifications by ants. Gall (A) contained Azteca sp., (B) Leptothorax sp., (C) Zacryptocerus pallens, (D) Zacryptocerus pusillus, and (E) Pseudomyrmex flavidus. among | 14 galls. We estimated X, the Pois- son parameter, as 65/114 = 0.57. The ob- served distribution of colonies among galls closely resembled the expected under a Poisson distribution (Table 2). The good- ness-of-fit test indicated no significant dif- ALL DIAME TEF MM Fig. 4. Relative frequency histograms of gall di- ameter for the entire set of galls (foreground) and for the subset of galls that were occupied by ants (back- ground). ference between the observed and expected distributions (x7, = 2.51, P > 0.10). Individual species did exhibit frequencies of co-occurrence that appeared different from those expected by chance alone (Table 3). For example, Zacryptocerus pusillus $ aoe e a ae ne at ce e e = ee, ry - < (©) ee enn 0 e oO a T 2 Ts 12.0 15C 18.0 PAK GALL DIAMETER Fig. 5. Relationship between the logarithm of col- ony size (number of individuals) and gall diameter. The relationship is Y = —0.929 + 0.122X (F, ,; = 20.1, P < 0.001; R? = 0.24). 86 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 2. Distribution of colonies among galls. Ex- pected values are based on a Poisson distribution with dA = 0.57. Colonies/Gall Observed Expected 0 62 64.5 1 42 36.8 2 7 10.5 3 3 2.0 never jointly occupied a gall, while approx- imately 4 such co-occurrences would be ex- pected at random. Azteca sp. and Lepto- thorax wilda were roughly half as likely to co-inhabit a gall as chance alone would pre- dict. On the other hand, Zacryptocerus pal- /ens exhibited a slight tendency to share galls. DISCUSSION Galls represent an important resource for this ant community, and ants exploit this resource in a non-random fashion. Ants se- lect significantly larger galls, over a narrow- er range of sizes, than chance alone would predict. Both the difference in means and the difference in variance indicate habitat selection. There may be strong selective pressures for this habitat selection as indi- cated by the significant positive relationship between colony size and gall size. Selection of larger galls by ants may be due to several reasons, among them increase in queen fit- ness. However, more work is called for to observe the selection and the use of larger versus smaller galls. It is not clear whether galls are a limiting resource to these ants. On the other hand, the most common species of ants exhibited resource partitioning. This is a necessary condition for stable coexistence of multi- species assemblages exploiting similar, lim- iting resources predicted by Lotka-Volterra based analyses of community dynamics (May 1973, Schoener 1974). Of course, nei- ther observation by itself is conclusive. Un- occupied galls may be the result of interfer- ence competition or priority effects (sensu Torres 1984). Significant differences among Table 3. Patterns of co-occurrences among ant species. Joint Occurrences Species Observed Expected Azteca bicolor l 0.58 Azteca sp. 4 Tad Leptothorax wilda 3 5.93 Leptothorax sp. 0 0.58 Pseudomyrmex flavidus 3 2B) Pseudomyrmex sp. A 1 0.58 Pseudomyrmex sp. B 0 0.58 Zacryptocerus pallens 5 3.54 Zacryptocerus pusillus 0 3.99 Zacryptocerus sp. A 5 5.19 Zacryptocerus sp. B 0 0.58 species in the diameters of occupied galls may not be the result of competitively in- duced resource partitioning, but rather, species’ idiosyncratic responses to their en- vironment (James et al. 1984). Correlative data provide notoriously weak inferences regarding mechanisms (Brady 1979); ex- periments are required to establish causa- tion. Patterns of species co-occurrence at the community level are indistinguishable from those predicted from a random assortment model. However at the constituent level, individual species deviate from expecta- tion. In particular, Zacryptocerus pusillus never shares a gall, although approximately four joint occurrences are predicted. Fowler et al. (1985) provide evidence that Z. pusil- /us is interspecifically territorial. Whether deviations from chance for the other species are biologically significant is unclear. Un- fortunately, little is known about the ecol- ogies of these ants. Gall-inhabiting ants may provide an ideal system to examine the determinants of community organization in arboreal ants. Galls are a discrete resource, amenable to experimental manipulation in time and space. ACKNOWLEDGMENTS We thank Hélcio R. Pimenta and Ivo das Chagas for field assistance. We also thank VOLUME 91, NUMBER 1 A. Mabelis and W. W. Benson for infor- mation on ants, H. Castanheira and L. M. Araujo for identification of ants, L. Ki- noshita for identification of Xy/opia aro- matica, and W. E. Clark for identification of the gall former insect. In addition, we thank the comments of H. Larew and an anonymous reviewer on an early draft of this manuscript. The Departamento de Biologia Geral, ICB, Universidade Federal de Minas Gerais and the staff of Codevasf- Trés Marias, MG provided logistical sup- port. This work was supported by a spe- cialization fellowship from SEMA/CNPg (n° 10.6193/84) to G. W. Fernandes. LITERATURE CITED Askew, R.R. 1961. On the biology of the inhabitants of oak galls of Cynipidae (Hymenoptera) in Brit- ain. Transactions of the Society of British Ento- mology 14: 237-268. 1980. The diversity of insect communities in leaf mines and plant galls. Journal of Animal Ecology 49: 817-829. Baust, J. G., R. Grandee, G. Condon, and R. E. Mor- rissey. 1979. The diversity of overwintering strategies utilized by separate populations of gall insects. Physiological Zoology 52: 572-580. Brady, R. H. 1979. Natural selection and the criteria by which a theory is judged. Systematic Zoology 28: 600-621. Brandhorst, C. T. 1962. The microcommunity as- sociated with the gall of Wa/shia amorphella (Lep- idoptera: Cosmopterygidae) on Amorpha_ fruti- cosa. Annals of the Entomological Society of America 55: 476-479. Espadaler, X. and J. L. Nieves. 1983. Hormigas (Hy- menoptera, Formicidae) probladoras de agallas abandonadas de cinipideos (Hymenoptera, Cyni- pidae) sobre Quercus sp. en la peninsula Iberica. Boletin de la Estacion Central de Ecologia 12: 89- 93. Felt, E. P. 1940. Plant Galls and Gall Makers. Com- stock, Ithaca. New York. Fowler, H. G., M. M. Costa, and J. Justi. 1985. Re- lacgao e reconhecimento individual em Paracryp- tocerus pusillus (Klug). XII Congresso Brasileiro de Zoologia (abstract), Campinas, Brasil, Page 130. James, F. C., R. F. Johnston, N. O. Wamer, J. Niemi, and W. J. Boecklen. 1984. The Grinnellian niche of the wood thrush. American Naturalist 124: 17- 30. Mani, M.S. 1964. Ecology of Plant Galls. W. Junk, The Hague. 87 May, R. M. 1973. Stability and complexity in model ecosystems. Princeton University Press, Prince- ton. Patton, W. H. 1879. A gall-inhabiting ant. American Naturalist 13: 126-127. Ping, C. 1920. Some inhabitants of the round gall of golden-rod. Journal of Entomological Zoology 7: 161-177. Price, P. W., G. W. Fernandes, and G. L. Waring. 1987. Adaptive nature of insect galls. Environ- mental Entomology 16: 15-24. Price, P. W., G. L. Waring, and G. W. Fernandes. 1986. Hypothesis on the adaptive nature of galls. Proceedings of the Entomological Society of Washington 88: 361-363. Sandlant, G. R. 1979. Arthropod successori inhab- iting willow galls during autumn in Christchurch, New Zealand. Mauri Ora 7: 83-93. Schoener, T. W. 1974. Resource partitioning in eco- logical communities. Science 185: 27-39. Shorthouse, J. D. 1973. The insect community as- sociated with rose galls of Diplolepis polita (Cy- nipidae, Hymenoptera). Quaestones Entomologi- cae 9: 55-98. Sokal, R. R. and F. J. Rohlf. man, San Francisco. Stegagno, G. 1904. I locatari dei cecidozoi sin qui noti in Italia. Marcellia 3: 18-53. Sturtevant, A. H. 1925. Notes on the ant fauna of oak galls in Woods Hole region. Psyche 32: 313- 314. Torossian, C. 197la. Faune secondaire des galles de Cynipidae: I. Etude systématique des fourmis et des principaux arthropods récoltés dans les galles. Insectes Sociaux 18: 135-154. 1971b. Etude biologique des fourmis forest- ieres peuplant les galles de Cynipidae des chénes. Insectes Sociaux 18: 193-202. Torres, J. A. 1984. Niches and coexistence of ant communities in Puerto Rico: Repeated patterns. Biotropica 16: 284-295. Uhler,C. D. 1951. Biology and Ecology of the golden- rod gall fly, Eurosta solidaginis (Fitch). Cornell University Agricultural Station Memoirs 300. Walsh, B. D. 1864. On insects, coleopterous, hy- menopterous, and dipterous, inhabiting the galls of certain species of willows. Proceedings of the Entomological Society of Philadelphia 3: 543-641. Wheeler, W. M. 1910. Ants. Columbia University Press, New York. Yukawa, J. 1983. Arthropod community centered upon the neolitsea leaf gall midge, Pseudoasphon- dylia neolitseae Y ukawa (Diptera: Cecidomyiidae) and its host plant, Neolitsea sericea (Blume) Koldz. (Lauraceae). Memoirs of the Faculty of Agricul- ture 19: 89-96. 1969. Biometry. Free- PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 88-111 REVISION OF THE ORNATUS SPECIES GROUP OF THE GENUS ANTHONOMUS GERMAR (COLEOPTERA: CURCULIONIDAE) WAYNE E. CLARK AND Horace R. BURKE (WEC) Department of Entomology and Alabama Agricultural Experiment Station, Au- burn University, Alabama 36849; (HRB) Department of Entomology, Texas A&M Uni- versity, College Station, Texas 77843. Abstract.—The seven species in the Anthonomus ornatus species group are A. ornatus Blanchard, A. signatipennis Blanchard, A. chilicola Clark, new species, A. kuscheli Clark, new species, 4. araucanus Clark, new species, A. blanchardi Clark, new species, and A. berberidis Clark, new species. These are hypothesized to constitute a monophyletic group on the basis of morphological characters of the adult weevils, their common association with plants in the genus Berberis (Berberidaceae) and their restricted distributions in southern Chile and Argentina. Other characters determined to be apomorphic by com- parisons with other anthonomines were used to produce a phylogeny of the species in the A. ornatus group. Characters diagnostic of the group and of each of the species are pre- sented, along with illustrations and a key to adults of the species. The larval and pupal stages of A. ornatus and A. kuscheli are also described and illustrated. Key Words: The Anthonomus ornatus group includes A, ornatus Blanchard, A. signatipennis Blan- chard, and five heretofore undescribed species. The species occur in southern Chile and Argentina at least as far south as Isla Navarino (55°S) in the Chilean province of Magallanes northward to the province of Coquimbo (30°S). Known hosts of the species are plants in the genus Berberis (Ber- beridaceae). The members of the A. ornatus group appear to be the only representatives of the subfamily Anthonominae in Chile. This fact is noteworthy because the antho- nomines are rich and diverse in the re- mainder of the Neotropical Region (O’Brien and Wibmer 1982, Wibmer and O’Brien 1986). It seems likely that the sister group of the 4. ornatus group will be found among the Nearctic or Neotropical species, but hy- potheses of such relationships cannot be formulated until these species are more Anthonomus, Anthonominae, PAUP, phylogeny, Berberis, Chile, Argentina completely known taxonomically. Mean- while, the descriptions, illustrations and keys presented in this paper should facilitate identification of the species in the 4. ornatus group. MATERIALS AND METHODS Specimens of 785 adults, including the types of the previously described species, were examined. These were from the col- lections of the following individuals and in- stitutions (letter codens identify the collec- tions in the text): AMNH The American Museum of Natural History, New York City, New York, USA (L. H. Herman, Jr.); CACA_ Collection of Allan C. Ashworth, Fargo, North Dakota, USA; CCBM_ Collection of C. Bordon, Maracay, Venezuela; VOLUME 91, NUMBER 1 CWOB Collection of C. W. O’Brien, Tal- lahassee, Florida, USA; Collection of H. and A. Howden, Ottawa, Ontario, Canada; Museum of Comparative Zoology, Cambridge, Massachusetts, USA (S. R. Shaw); Muséum National d’Histoire Nat- urelle, Paris, France (H. Perrin); Museo Nacional de Historia Nat- ural, Santiago, Chile (M. Elgueta 1D Museu de Zoologia, Universidade de Sao Paulo, Sao Paulo, Brazil (U. R. Martins); Division of Scientific and Indus- trial Research, Auckland, New Zealand (G. Kuschel); Texas A&M University, College Station, Texas, USA (H. R. Burke); National Museum of Natural His- tory, Washington, D.C., USA (D. R. Whitehead). HAHC MCZC MNHN MNNC MZSP NZAC TAMU USNM Measurements of adult specimens were made with an ocular micrometer in a dis- secting microscope as follows: total length from anterior margin of eye to elytral apex in lateral view; width across elytra at widest point; length of pronotum, dorsally, from anterior to posterior margins; length of ros- trum from anteroventral margin of eye to apex, across arc, in lateral view; length of distal portion of rostrum from antennal in- sertion to apex in lateral view; width of frons at narrowest point between eyes; width of base of rostrum just distad of eyes in dorsal view; and width of pro- and metafemora, in anterior view, excluding the ventral teeth. The range and, in parentheses, the mean and sample size of each measurement are given for each species. Descriptions of larvae are based on full- grown specimens. The terminology follows Anderson (1947) and Thomas (1957). Where differences in terminology exist between the two systems, the Thomas terms are placed in parentheses following those of Anderson. 89 Terminology of pupal characters follows Burke (1968). THE ANTHONOMUS ORNATUS GROUP Adults of the seven species in the A. or- natus group are about average in size for the genus (specimens examined range from 1.88 to 4.00 mm in length), somewhat elongate and slender in body form, without signifi- cant elytral prominences, and with relative- ly slender profemora armed with a mod- erately large ventral tooth and a smaller, distal tooth. They are distinguished by their contrasting patterns of white, pallid to dark ferruginous and fuscous scales that cover corresponding areas of pallid or dark integ- ument (Figs. 1-12). No other anthonomines examined have exactly this same pattern. Adults of the species also have an elongate, flat, distal endophallic sclerite (Figs. 19-30). They are further characterized as follows: Head: vertex with narrow ferruginous scales, frons with broader, more pallid scales; eyes strongly convex, slightly to promi- nently elevated, separated by distance 0.6- 0.9 x width of rostrum at base; antennal funiculus with 7 articles. Rostrum: slender; proximal portion with well-developed ca- rinae and sulci; finely setose. Prothorax: pronotum densely, coarsely punctate, with narrow, attenuate, pallid to dark ferrugi- nous scales and broader, more pallid scales in narrow middorsal vitta; pleuron with small anteromedian patch of pallid, broad scales that is feebly developed in small spec- imens. E/ytra: humeri not prominent; sides subparallel. Pygidium: not impressed or sul- cate. Legs: pro- and mesocoxae with dense, broad, pallid scales that are weakly devel- oped in small specimens; profemur slender, ca. 1.1-1.2 x stouter than metafemur, with small, conical, acute ventral tooth and smaller distal tooth; metafemur with small ventral tooth and minute distal tooth; pro- tibial uncus slender, slightly curved. Geni- talia: male tegmen with long parameres. Distribution (Figs. 42-45). The species in the 4. ornatus group are known mainly from 90 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 1-6. Habitus of adults (lines represent | mm): 1) A. signatipennis, male, Quinta Pittet, Magallanes, Chile, lateral view; 2) the same, dorsal view; 3) A. ornatus, male, Rio El Ganso, Seno de Otway, Magallanes, Chile, lateral view; 4) the same, dorsal view; 5) 4. kuscheli, female, Chepu, Chiloé, Chile, lateral view; 6) the same, dorsal view. the Valdivian and Magellanic forest regions which, according to Kuschel (1960), extend south of the 36th parallel beyond the south- ernmost portion of the continent to Isla Navarino at the 55th parallel. Blanchard (1851: 387) stated that both A. signatipennis and A. ornatus were found in “la provincia de Coquimbo.” These records, along with records of A. signatipennis from the Chilean provinces of Valparaiso and Santiago (Figs. 42, 43), extend the limits of the group sig- nificantly north of the Valdivian Forest into more arid regions. A possible explanation for this extension is found in the observa- tion by Davis (1986) that in parts of the Coquimban Desert Province, moisture from coastal fog supports remnants of the Val- divian forest. Davis listed several southern VOLUME 91, NUMBER | 91 Figs. 7-12. lateral view; 8) the same, dorsal view; 9) A. chilicola, male, 20 km. E Manzanar, Malleco, Chile, lateral view; 10) the same, dorsal view; 11) 4. berberidis, male, Frutillar, Llanquihue, Chile, lateral view; 12) the same, dorsal view. temperate plant species (not including Ber- beris, the known hosts of the members of the A. ornatus group), as occurring in these moist refugia. Plant associations. Six of the 7 species of the 4. ornatus group are known to be as- sociated with one or more of the following Berberidaceae: Berberis buxifolia Lamarck; Habitus of adults (lines represent | mm): 7) 4. araucanus, male, Laguna Malleco, Malleco, Chile, B. chilensis Hort. ex C. Koch; B. darwini Hooker; and B. ilicifolia Forster in Comm. The two species for which the immatures are known, A. ornatus and A. kuscheli, de- velop in flower buds. The site of develop- ment of the remaining species is not known. Label data indicate that adults of A. sig- natipennis and A. ornatus have been col- 92 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 18 17 Figs. 13-16. Pygidium, adult female, dorsal view (line represents 0.25 mm): 13) A. signatipennis, Quinta Pittet, Magallanes, Chile; 14) 4. ornatus, Lago Frio, Aisén, Chile; 15) A. kuscheli, Panquipulli, Valdivia, Chile; 16) A. araucanus, Laguna Laja, Bio-Bio, Chile. Figs. 17-18. Abdomen, adult female, ventral view (line represents 0.25 mm): 17) A. ornatus, Lago Frio, Aisén, Chile; 18) 4. blanchardi, El Coigo, Curic6, Chile. lected together, on the same date, at several velop in flower buds, but the developmental localities. One of these records indicates that site of A. signatipennis is unknown. Adults specimens of the two species were taken on of the two species have also been taken at the same day on B. buxifolia. Larvae ofone the same locality on B. ilicifolia, but one of the species, 4. ornatus, are known to de- day apart. Adult 4. blanchardi have been = 29 30 Figs. 19-30. Aedeagus, adult male (line represents 0.5 mm): 19) 4. signatipennis, Estancia Fenton, Magal- lanes, Chile, dorsal view; 20) the same, lateral view; 21) 4. ornatus, Trapatrapa, Concepcion, dorsal view; 22) the same, lateral view; 23) A. ornatus, Lago Frio, Aisén, dorsal view; 24) the same, lateral view; 25) 4. blanchardi, El Coigo, Curico, Chile, dorsal view; 26) the same, lateral view; 27) A. kuscheli, Pichinahuel, Malleco, Chile, dorsal view; 28) 4. araucanus, 20 km. E Manzanar, Malleco, Chile, dorsal view; 29) 4. chilicola, 20 km. E Manzanar, Malleco, Chile, dorsal view; 30) 4. berberidis, Frutillar, Llanquihue, Chile, dorsal view. 93 94 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 31-37. Figs. 38-41. Third instar larva of Anthonomus kuscheli. 31) lateral view; 32) head capsule, frontal view; 33) labium and right maxilla, ventral view; 34) epipharynx; 35) clypeus and labrum; 36) abdominal segments 3 and 4, dorsal view; 37) mandible. Line accompanying Fig. 31 = 2 mm; Fig. 32 = 0.5 mm; Fig. 33 = 0.25 mm; other figures greatly enlarged. Pupa of Anthonomus ornatus. 38) head, rostrum and ventral view of prothorax; 39) prothorax, dorsal view; 40) terminal abdominal segments, dorsal view; 41) 9th abdominal segment; lateral view. Line accompanying Fig. 38 = | mm; Fig. 39 = 1 mm; Fig. 40 = 0.5 mm. collected on B. buxifolia, but not at the same locality as A. signatipennis and A. ornatus. Adults of another pair of species, 4. kuscheli and A. berberidis, have been collected on B. darwini at the same time and locality. Discussion. No observed characters of adults of the members of the A. ornatus group can be cited with confidence as evi- dence of a sister group relationship with any anthonomine group (Clark 1987a, b, c, 1988, Clark and Burke 1985, 1986a, b, c, Clark and Martins 1987). The larvae of two of the species, A. ornatus and A. kuscheli, are known. These larvae trace to A. pomorum L. and A. nebulosus LeConte in a key to larvae of members of the subfamily (Ahmad and Burke 1972), but there is no other in- dication that these species are very closely related. Larvae of A. kuscheli are distinct in possessing three rather than four epipharyn- Fig. 42. Map of southern South America showing the distribution of 4. signatipennis Fig. 43. Map of southern South America showing the distribution of 4. ornatus Fig. 44. Map of southern South America showing the distributions of 4. kuscheli (circles) and A. chilicola (square). Fig. 45. Map of southern South America showing the distributions of 4. blanchardi (circles), A. araucanus (triangles) and A. berberidis (squares). 95 96 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON geal sensory pores. Larvae of Coccotorus scutellaris (LeConte) are the only other an- thonomines known to have only three epi- pharyngeal sensory pores, but available in- formation does not indicate that this species is closely related to the 4. ornatus group. The pupae of members of the 4. ornatus group have a single process on the 9th ab- dominal segment. This structure, however, is also shared with several other, apparently unrelated anthonomines, including Tachy- pterellus quadrigibbus Say, Pseudanthono- mus validus Dietz, Anthonomus albopilosus Dietz, A. stupulosus Champion, A. griseis- quamis Champion, A. unipustulatus Cham- pion and 4. nubiloides Fall (Burke 1968). The association of the species in the 4. ornatus group with plants in the genus Ber- beris is biogeographically significant. Guil- lermo Kuschel (pers. comm.), speaking from years of experience collecting weevils and other insects in Chile and neighboring coun- tries, noted that the 4. ornatus group mem- bers were found exclusively on Berberis. He also insisted that he “frequently and quite thoroughly” checked Berberis plants all over northern Chile, Peru and Bolivia, but found no anthonomines on those plants there. The +500 species of Berberis are distributed throughout the north temperate zones and southward, mainly at higher altitudes, in tropical regions of Africa, Asia and South America (Cronquist 1981: 130, Good 1964: 80-81). As noted by Cabrera and Willink (1973: 98), species of Berberis are abundant in the Valdivian and Magellanic forests of southern Chile and adjacent portions of Ar- gentina. Raven (1963: 155) listed Berberis among several genera that used the moun- tains to extend their distribution from the northern hemisphere to the south. No other anthonomines are known to have hosts in the family Berberidaceae (Burke [1976] list- ed plant families known to contain anthonomine hosts). It is fairly certain that the members of the A. ornatus group are the only anthonomines to be found in Chile. No others are listed by Wibmer and O’Brien (1986) in their checklist of South American Curculionidae. Schenkling and Marshall (1934) did list two other Chilean Curculionidae, Anthonomus variabilis Philipi and A. australis Philipi, as anthonomines, but Kuschel (1950: 17) de- termined that these are actually members of the genus Rhopalomerus Blanchard in the subfamily Eugnominae. Key TO ADULTS OF THE SPECIES OF ANTHONOMUS IN THE A. ORNATUS GROUP 1. Elytra (Figs. 1, 2) with oblique, alternating light and dark fasciae, without posterolateral mac- ulae; elytral interstria 2 with discrete postero- median pallid line bounded anteriorly and posteriorly by short, dark portions (Figs. 1, 2) Baer? signatipennis 1’. Elytra with large, posterolateral maculae (Figs. 3-6, 9, 10) (feebly developed in some speci- mens, Figs. 7, 8, 11, 12); elytral interstria 2 without discrete posteromedian pallid seg- ment bounded anteriorly and posteriorly by Short;:dark portions’; sacs. scene oe ieeee 2 Sutural elytral interstriae with acute, overlap- ping apical extensions; metatibia of male with ventral margin strongly concave in distal 7, with enlarged, excavated apical mucro; aedea- gus asymmetrical (Figs. 29, 30) ‘. Sutural elytral interstriae without apical ex- tensions; metatibia of male with ventral mar- gin not or only slightly concave in distal *, apical mucro small; aedeagus symmetrical (Figs. 19-28) 3. Sternum 5 of male with large ‘posteromedian prominence (Fig. 11); pallid annuli around posterolateral elytral maculae weakly devel- oped (Fig. 12) . a eee TESS eee berberidis 3’. Sternum 5 of male einen posteromedian prominence; pallid annuli around postero- lateral maculae distinct (Fig. 10) ...... chilicola 4. Elytra with posterolateral maculae completely surrounded dorsally by pallid annuli (Figs. 5, 6); pygidium of female with subapical prom- inences (Fig. 15); body slender, somewhat flat- tened dorsally (Fig. 5) ................ kuscheli 4’. Elytra with posterolateral maculae incom- pletely surrounded dorsally by pallid annuli (Figs. 3, 4), or macula feebly developed (Figs. 7, 8); pygidium of female with subapical prom- inences weakly developed (Fig. 14) or absent; body less slender, less flattened (Figs. 3,7) .. 5 5. Elytral pattern distinct (Figs. 3, 4); sternum 5 of female with shallow emarginations (Fig. 18) or deep incisions (Fig. 17) on each side of me- th i) VOLUME 91, NUMBER | dian prominence; aedeagus with apicolateral prominences (Figs. 21, 23, 25); pygidium of female without apicomedian prominence (Fig. wn oy = pss = = uo} S 5 = oO p a Leo} a oO < fa) ° b> a (3s ie oOo. Se Gus ia} Oo. > oS: <4 prominent midbasal macula (Figs. 7, 8); ster- num 5 of female without emarginations or in- cisions; apex of aedeagus narrowed to apex, without apicolateral projections (Fig. 28); py- gidium of female with small apicomedian prominence (Fig. 16) . . araucanus 6. Sternum 5 of female with deep posteromedian incisions (Fig. 17); aedeagus (Figs. 21, 23) about as broad at broadly rounded apex as at base, with acutely pointed apicolateral prominences low emarginations on each side of small, blunt, posteromarginal prominence; aedeagus (Fig. 25) narrower at apex than at base, apex slightly distended, with blunt, weakly developed api- colateral prominences . ... blanchardi Anthonomus signatipennis Blanchard igs. 15.2, 13, 19; 20).42 Anthonomus signatipennis Blanchard 1851: 387. Holotype. Chile, “provincia de Co- quimbo,” male, labelled with a green disc and [67/ 7] [TYPE] [MUSEUM PARIS/ CHILI/ CL. GAY 1849] [Anthonomus/ signatipennis/ Type Blanchard/ H. Perrin det. 19] (MNHN). Blackwelder 1947: 839. Schenkling and Marshall 1934: 59. Wib- mer and O’Brien 1986: 204. Recognition (Figs. 1, 2). Adults of A. sig- natipennis are distinguished by the follow- ing combination of characters: Elytra (Figs. 1, 2) without posterolateral maculae; interstria 2 with posteromedian pallid portion bounded anteriorly and posteriorly with short dark portions; in- terstriae 3, 5, 7 and 9 with long postero- median dark portion contrasting with short dark portions on interstriae 4, 6 and 8; aedeagus (Figs. 19, 20) with broad, blunt apicomedian prominence. They are easily distinguished from adults of the other members of the A. ornatus group by the distinct elytral pattern. They lack the 97 large posterolateral elytral maculae char- acteristic of adults of the other members of the group. Distinctive are the prominent posthumeral patch of pallid scales, the oblique discal fascia of pallid scales extend- ing across interstriae 1-4, and the oblique, alternating dark and pallid elytral fasciae. They are most likely to be confused with adult 4. ornatus but, in addition to the dif- ferent elytral pattern (cf. Figs. 1—4), the fe- male sternum 5 lacks posteromarginal emarginations and the male aedeagus lacks acute subapicolateral prominences (cf. Figs. 19S 23): Adult male. Length: 3.28-3.64 mm (x = 3.40, n = 10). Width: 1.46-1.60 mm (x = 1.53,n = 10). Rostrum: length 1.30-1.51 x (x = 1.40, n = 10) pronotal length; distal portion finely punctate, length 18-30% (x = 23,n = 10) of total rostral length. Prothorax: pronotum with pallid scales in broad sub- apical fascia and in posterolateral vittae. Elytra: interstriae 3 and 5 slightly elevated at base and on disc; integument and scales dark on humeri, on basal, elevated portions of interstriae 3 and 5, on discal portions of interstriae 1-3, in oblique, irregular, an- teromedian and posteromedian fasciae, and on apices of interstriae 4-6; pallid integu- ment and scales predominant in small oblique patches extending across interstriae 1-3, in large, rectangular patch behind hu- meri, in narrow, oblique, anteromedian fas- cia that extends posteriorly from interstria 5 to suture, in broader, oblique postero- median fascia, and on declivity. Abdomen: posteromedian portion of sternum 5 slightly depressed behind transverse impression. Legs: protibia with slight ventral marginal prominence at proximal 3; metatibia nar- row, ventral margin slightly, broadly con- vex in middle *3; metatibial mucro short, straight, acute, not excavated. Adult female. Length: 3.40-3.72 mm (x = 3.52, n = 10). Width: 1.56-1.66 mm (x = 1.61,n = 10). Rostrum: length 1.25-1.47 x (X¥ = 1.36, n = 10) pronotal length; distal portion subcylindrical, shallowly punctate, 98 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON glabrous, length 22-29% (* = 25, n = 10) of total rostral length. Pygidium (Fig. 13): apical margin rounded, without subapical ridge. Abdomen: sternum 5 without median emargination. Legs: metatibial mucro mi- nute. Distribution (Fig. 42). The 336 adult specimens of A. signatipennis examined are from the following localities. ARGEN- TINA. Neuquén: San Martin de los Andes. Rio Negro: Norquinc6é. Santa Cruz: Cana- don Leon; Lago Argentina. CHILE. Aisén: Chile Chico; Coyhaique; 10 km. N Puerto Ibanez. Bio-Bio: Los Barros, Laguna Laja; Volcan Copahue. Cautin: Cherquenco. Chiloé: Castro; Chepu; Dalcahue. Concep- cidn: Salto de Laja. Llanquihue: Lepihue, W of Puerto Montt; Puerto Varas; Los Muer- mos; Maullin; Puerto Montt. Magallanes: Estancia Cameron; Estancia Fenton; La- guna Amarga; Puerto Williams; Punta Are- nas; Quinta Pittet; Rio Santa Maria. Mal- leco: Lonquimay; Licura; Laguna Malleco; 19 km. E Manzanar; 20 km. E Manzanar; Pichinahuel, Cordillera Nahuelbuta; 38 km. SE Victoria. Nuble: Las Cabras, Cordillera Chillan; Shangrila, 75 km. E Chillan. Osor- no: Laguna la Copa, Parque Nacional de Puyehue. Santiago: Melocoton; 4 km. W Portillo; Quinta San Ramon. Talca: 5 km. W Molino. Valparaiso: Algarrobo. The specimens examined were collected during the months of October through February and in April. Plant associations. Adults of A. signati- pennis were collected on Berberis buxifolia at Chepu, Chiloé, Quinta Pittet and Puerto Williams, Magallanes; on B. chilensis at Sal- to de Laja, Concepcion and Algarrobo, Val- paraiso; on B. darwinii at Laguna la Copa, Osorno, and on B. ilicifolia at Chile Chico, Aisén, Chile. Anthonomus ornatus Blanchard Figs. 3, 4, 14, 17, 21-24, 38-41, 43 Anthonomus ornatus Blanchard 1851: 387. Holotype. Chile, female, “provincia de Coquimbo,” labelled with green disc and [15/ 43] [illegible handwritten label] [An- thonomus/ ornatus.] [TYPE] [MUSEUM PARIS/ CHILE/ Gay 1843] (MNHN). Kuschel 1950: 17 (= biplagiatus Fair- maire, not Redtenbacher [redtenbacheri Blackwelder]). Ashworth and Hoganson 1987. Schenkling and Marshall 1934: 58. Wibmer and O’Brien 1986: 204 (= bipla- giatus Fairmaire, not Redtenbacher [red- tenbacheri Blackwelder)). Anthonomus biplagiatus Fairmaire 1884: 503-504. Holotype. Chile: ‘“Punta-Are- na,” female [223] [TYPE] [MUSEUM PARIS/ SANTA-CRUZ/ LEBRUN 1883] [anthonomus/ biplagiatus/ Fairm] (MNHN). Kuschel 1950: 17. Wibmer and O’Brien 1986: 204. Anthonomus redtenbacheri Blackwelder 1947: 839. Replacement name for A. bi- plagiatus Fairmaire (1884) not Redten- bacher (1867). Recognition (Figs. 3, 4). Adults of A. or- natus are distinguished by the following combination of characters: Elytra (Figs. 3, 4) with large posterolateral maculae; pallid dorsal annuli around maculae broken dorsomedially by dark posteromedian portion on interstria 2; sutural elytral interstriae without apical extensions; sternum 5 of female (Fig. 17) with deep apicomedian emarginations on each side of slender median projection; aedeagus (Figs. 21-24) symmetrical, with rounded apex and acute lateral promi- nences; pygidium of female (Fig. 14) with weakly developed subapical promi- nences. They are distinguished from adults of A. signatipennis by the characters discussed under the latter. The elytral patterns are dis- tinct in all specimens of A. ornatus exam- ined but there is considerable variation in the ratio of dark and pallid portions. In some, the pallid portions are extensive and clothed wtih predominantly white scales, with dark fuscous portions correspondingly VOLUME 91, NUMBER 1 limited primarily to the elevated basal and discal portions of interstria 3 and to the edges of the posterolateral maculae. In oth- ers, the pallid portions are more extensive, the pallid scales are mostly ferruginous, and dark integument and fuscous scales are ex- tensive on the basal portions of interstriae 2-5, on the median portion of the disc, on the posterolateral maculae and on the de- clivity. The specimen illustrated (Figs. 3, 4) is somewhat intermediate between the ex- tremes. There is also variation in the shape of the aedeagus. At one extreme the sides, seen in dorsal view (Fig. 21), are subparallel behind the apicolateral projections and the lateral plates are only slightly distended dor- sally, as seen in lateral view (Fig. 22). At the other extreme, the sides are broadly con- stricted in the distal *4 in dorsal view (Fig. 23) and the lateral plates are correspond- ingly distended dorsally in lateral view (Fig. 24). This variation 1s not geographical; both extremes are exhibited in specimens from Trapatrapa, Concepcion, which do not dif- fer otherwise. Adult male. Length: 2.56-3.32 mm (x = 2.90, n = 10). Width: 1.16-1.60 mm (x = 1.39, n = 10). Rostrum: length 1.40-1.70 x (x = 1.60, n = 10) pronotal length; distal portion finely punctate, length 23-32% (x = 28, n= 10) of total rostral length. Prothorax: pronotum without distinct subapical fascia and dorsolateral vittae of pallid scales. E/y- tra: interstria 3 with slightly elevated basal and discal portions; dark integument and scales predominant on humeri, on basal el- evated portions of interstriae 3 and 5 and on discal portions of interstriae 1-3, form- ing irregular edges of posterolateral mac- ulae, and on apices of interstriae 4—6; pallid integument and scales predominant on posthumeral portions, in broad annuli around edges of posterolateral maculae, in median portion of that macula, and on de- clivity. Abdomen: posteromedian portion of sternum 5 slightly depressed behind trans- verse impression. Legs: protibia with slight midventral marginal prominence; metatib- eye) la narrow, straight, with slight ventral mar- ginal prominence; metatibial mucro large, curved, excavated. Adult female. Length: 2.48-3.32 mm (x = 2.98, n = 10). Width: 1.12-1.58 mm (x = 1.41, n = 10). Rostrum: length 1.59-1.76 x (X = 1.68, n = 10) pronotal length; distal portion slender, shining, glabrous, length 29- 37% (X = 35, n = 10) of total rostral length. Legs: metatibial mucro minute. Larva (2 specimens from Frutillar, Llan- quihue, Chile, collected November 2, 1983, by G. Kuschel, ex flower buds Berberis buxi- folia). The third instar larva of A. ornatus resembles that of 4. kuscheli in size and other characters except as follows: it is more strongly curved; the ninth abdominal seg- ment is not as long; and there are four epi- pharyngeal sensory pores instead of three as in A. kuscheli. The significance of these ap- parent differences will have to await the availability of additional specimens. Pupa (5 specimens from Frutillar, Llan- quihue, Chile, collected November 2, 1983, by G. Kuschel ex flower buds of Berberis buxifolia). Body: length 3.8-4.2 mm (n = 5). Head (Fig. 38): frontal setae straight to slightly curved; each borne on summit of small, rounded tubercle; pair separated by distance greater than length of a seta. Su- praorbital setae absent. One pair of fine ba- sirostral (interocular) setae; each about '2 length of frontal seta. One pair of minute sessile distirostral setae. Pronotum (Fig. 39): pronotal setae straight to feebly curved; se- tae on anterior margin slightly longer and stouter than posterior setae. Anteromedian setae each borne subapically on anterior face of conical, sharply pointed tubercle; tuber- cles separated by distance equal to ca. width of tubercle at base. Three pairs of antero- lateral setae; each borne on summit of slight to distinct tubercle; tubercle of anterolateral 3 often taller than others; setae on each side of pronotum in straight to feebly curved line. Posteromedian setae each borne at or near apex of subconical to acutely pointed tubercle; tubercles separated by distance 100 slightly greater than length of a seta. Three pairs of posterolateral setae; arranged in curved line on each side of pronotum; each borne on summit of small, rounded tuber- cle. Mesonotum: three pairs of straight mesonotal setae; each borne on summit of rounded tubercle or seta, innermost oc- casionally subapical on small, acutely point- ed tubercle. Metanotum: three pairs of straight to curved metanotal setae; more widely spaced than mesonotals; each borne on summit of rounded tubercle; metanotal 2 usually located closer to 3 than to 1. Ab- domen (Fig. 40): three pairs of discotergal setae and occasionally with additional se- taless tubercles on some terminal segments; setae each borne on summit of low tubercle on first two abdominal terga, remaining ter- ga with seta located at base of sharply point- ed tubercles that become progressively larg- er posteriorly. Laterotergal setae 1 and 2 present on each of first 8 terga; seta 1 mi- nute, borne at base of small, sharply pointed tubercle on all terga; seta 2 borne subapi- cally on sharply pointed tubercle, slightly curved, longer than discotergal setae. An- teronotal setae absent. Spiracles well de- veloped on first 5 abdominal segments, fee- bly developed on segment 6, absent on others. Laterosternal and sub-laterosternal setae absent. Segment 9 bearing a single sharply pointed process which in side view (Fig. 41) is turned slightly upward apically; minute seta borne on each side of segment ca. midway between apex and base of seg- ment. Distribution (Fig. 43). The 235 adult specimens of A. ornatus examined are from the following localities: ARGENTINA. Neuquén: San Martin de los Andes. Rio Ne- gro: Colonia Catedral; Norquinco; Rio Foy- el. Santa Cruz: Canadon Leon. Tierra del Fuego: Bahia San Sebastian, Punta de Are- nas. CHILE. Aisén: Chile Chico; Coy- haique; Lago Escondido; Lago Frio; 10 km. N Puerto Ibanez; Quellon. Bio-Bio: Los Barros, Laguna Laja. Chiloé: Chepu. Con- cepcion: Trapatrapa. Linares: Parral. Llan- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON quihue: Frutillar; Rio Penon, Maullin. Ma- gallanes: Cerro Castillo; Dos Lagunas; Estancia Cameron; Estancia Canelo; Estan- cia Fenton; Isla Navarino; Laguna Azul and Ultima Esperanza, Parque Nacional Torres del Paine; Laguna Los Robles; Quinta Pit- tet; Puerto Williams; Punta Arenas; Rusfin; Rio Tres Pasos; Rio el Ganso, Seno de Ot- way. Malleco: Icalma; La Fusta; 20 km. E Manzanar; Marimenuco; Cordillera Na- huelbuta: Villa Portales. Nuble: Cordillera Chillan, Las Cabras, Las Trancas, 70 km. E Chillan; El Marchant; Recinto; 4.5 km. SE Recinto. Osorno: Antillanca; 2 km. SW Vn. Casablanca, Parque Nacional de Puyehue. The specimens examined were collected during the months of October through Feb- ruary and in April and August. Only one of the specimens on which the reference to A. ornatus by Ashworth and Hoganson (1987: 887) is based is that species. The others are A. signatipennis, A. berberidis and A. kus- cheli. Plant associations. Specimens of A. or- natus were collected on Berberis buxifolia at Chepu, Chiloé, Frutillar, Llanquihue, and Quinta Pittet and Puerto Williams, Maga- llanes; on Berberis ilicifolia at Chile Chico, Aisén, and Puerto Williams, Magallanes; and on Berberis sp. at Trapatrapa, Concep- cion, and Laguna Azul and Ultima Espe- ranza, Magallanes, Chile. Anthonomus blanchardi Clark, NEw SPECIES Figs. 18, 25, 26, 45 Type specimens. Holotype: Chile, male [CHILE/ El Coigo/ Cord. Curico/ Oct-Nov 1959] [G. Kuschel/ Collection] [Entomol- ogy/ Division/ D.S.I.R./ New Zealand] (NZAC). Paratypes: Chile, 1 male, 2 females [CHILE/ El Coigo/ Cord. Curico/ Nov—Dec 1959] [G. Kuschel/ Collection] [Entomol- ogy/ Division/ D.S.I.R./ New Zealand]; | female [Santiago/ Farellones/ 2200 m/ 10 Dec 1950] [G. Kuschel/ Berberis/ buxifolia] [Coll./ Kuschel] [Entomology/ Division/ D.S.I.R./ New Zealand]; 1 female [Chile: VOLUME 91, NUMBER 1 8600'/ Farellones/ P. Santiago/ XII-25- 1968] [under dung L &/ C. W. O’Brien]; | female [Chile Santiago/ Penalolén/ 25-I- 1975/ Coll: M. Beeche]; 1 female [Altos de Vilches/ Talca A280 mts./ 25—26-XI-1970/ J. Solvericens]. Total paratypes, 7 (CWOB, MNNC, NZAC). Recognition. Adults of A. blanchardi are distinguished by the following combination of characters: Elytra (cf. Figs. 3, 4) with large postero- lateral maculae; pallid dorsal annuli around macula broken medially by dark posteromedian portion on interstria 2; sutural elytral interstriae without apical extensions; sternum 5 of female (Fig. 18) with shallow apicomedian emarginations on each side of short median prominence; aedeagus (Fig. 25, 26) symmetrical, nar- rowed in distal '2, with rounded apex and short, acute apicolateral prominences; pygidium of female (cf. Fig. 14) with weakly developed subapical promi- nences. They are distinguished from adults of the closely allied 4. ornatus by characters of the male genitalia and of sternum 5 of the fe- male. The aedeagus of A. blanchardi (Figs. 25, 26) is constricted and narrowed in the distal 7 to the apex. The apex itself is slight- ly distended, rather than being rounded as in A. ornatus (cf. Figs. 21, 23, 25). It is also narrower than the apex in that species and has less well-developed apicolateral prom- inences. The median posteromarginal prominence on sternum 5 of the female is bounded on each side by shallow emargi- nations rather than being deeply incised as in A. ornatus (cf. Figs. 17, 18). Adult male. Length: 2.48-2.68 mm (xX = 2.58, n = 2). Width: 1.30-1.34 mm (Xx = 1.32, n = 2). Rostrum: length 1.43-1.55 x (x = 1.49, n = 2) pronotal length; distal portion finely punctate, length 32-35% (x = 34, n = 2) of total rostral length. Prothorax: pronotum without distinct subapical fascia and dorsolateral vittae of pallid scales. E/y- 101 tra: interstria 3 with slightly elevated basal and discal portions; dark integument and scales predominant on humeri, on basal el- evated portions of interstriae 3 and 5 and on discal portions of interstriae 1-3, form- ing irregular edges of posterolateral mac- ulae, and on apices of interstriae 4—6; pallid integument and scales predominant on posthumeral portions, in broad annul around edges of posterolateral maculae, in median portion of macula, and on declivity. Abdomen: posteromedian portion of ster- num 5 slightly depressed behind transverse impression. Legs: protibia with slight mid- ventral marginal prominence; metatibia narrow, straight, with slight ventral margin- al prominence; metatibial mucro large, curved, excavated. Adult female. Length: 2.48-3.16 mm (x = 2.81, n = 6). Width: 1.10-1.50 mm (x = 1.34, n = 6). Rostrum: length 1.54-1.69 x (X = 1.62, n = 6) pronotal length; distal portion slender, shining, glabrous, length 30- 40% (X = 35, n = 6) of total rostral length. Legs: metatibial mucro minute. Distribution (Fig. 45). Anthonomus blan- chardi is known from the type series from the following localities. CHILE. Curico: El Coigo, Cordillera Curico. Santiago: Farel- lones; Penalolén. Ta/ca: Altos de Vilches. Plant associations. The paratype of 4. blanchardi from Farellones, Santiago, Chile, was collected on Berberis buxifolia. Etymology. This species is named for Charles Emile Blanchard (1819-1900) in honor of his contribution to the study of the A. ornatus group. Anthonomus kuscheli Clark, NEw SPECIES Figs. 5, 6, 15, 27, 31-37, 44 Type specimens. Holotype: Chile, male (CHILE/ Pichinahuel/ Cord. Nahuelbuta/ 1-31 Jan 1959] [G. Kuschel/ Collection] [Entomology/ Division/ D.S.I.R./ New Zealand] (NZAC). Paratypes: Argentina, | female [Argentina/ Rio Negro/ L. Nahuel- huapi/ Llao Llao] [24 Nov 1950/ W. Witt- 102 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON mer] [Coll. Kuschel] [Entomology/ Divi- sion/ D.S.I.R./ New Zealand]; 1 male {Argentina/ R. Negro/ L. Nahuelhuapi] [1 Mar 1949/ W. Wittmer] [Coll. Kuschel] [Entomology/ Division/ D.S.I.R./ New Zealand]. Chile, 1 male [CHILE] [Ento- mology/ Division/ D.S.I.R./ New Zealand]; 3 males [Aysen/ Rio Correntoso/ 22-I-1971/ F. Silva G.] [Ex. Berberis/ buxifolia]; 1 male, 2 females [CHILE: P./ Aisen, 7 km./ W. Coyhaique/ I-21-1968] [at night L. &/ C. W. O’Brien]; 1 female [CHILE: P./ Aisen, 15 km./ W. Coyhaique/ I-21-1968] [Collec- tors: L &/ C. W. O’Brien]; | male [Bio-Bio/ Pemehue/ R. Renaico] [12 Jan 1946/ G. Kuschel] [Coll./ Kuschel] [Entomology/ Di- vision/ D.S.I.R./ New Zealand]; 1 male [male symbol] [Cautin/ Vn. Llaima/ 17 Sep 1951/ M. Codoceo] [Coll./ Kuschel] [Ento- mology/ Division/ D.S.I.R./ New Zealand]; 1 male, | female |CHILE/ Chiloé/ Chepu/ 22.10.58] [s/ Berberis/ darwini] [Coll./ Kus- chel] [Entomology/ Division/ D.S.I.R./ New Zealand]; | male, | female [CHILE/ Chiloé/ Chepu/ 23.9.54] [Coll./ Kuschel] [Ento- mology/ Division/ D.S.I.R./ New Zealand]; 6 males, 7 females [CHILE, Llanquihue/ Frutillar/ 31 October 1983/ G. Kuschel] [Berberis/ darwini/ em. 9 Nov 1983]; 1 fe- male [CHILE/ Frutillar/ 27 Feb 1950/ G. Kuschel] [G. Kuschel/ Collection] [Ento- mology/ Division/ D.S.1.R./ New Zealand]; 1 male [CHILE/ Pichinahuel/ Cord. Na- huelbuta/ 1-31 Jan 1959] [G. Kuschel/ Col- lection] [Entomology/ Division/ D.S.I.R./ New Zealand]; 2 males [CHILI/ Pichina- huel 1200 m/ 14-18 Feb 1956] [G. Kuschel/ Collection] [Entomology/ Division/ D.S.I.R./ New Zealand]; 12 males, 17 females (CHILE: 19 km./ E. Manzanar/ P. Malleco/ XI-3-1967] [Collectors: L &/ C. W. O’Brien]; 1 female [Chile: 20 km./ SE Victoria/ P. Malleco/ XI-3-1967] [Collectors: L &/ C. W. O’Brien] [compared/ with type] [An- thonomus/ ornatus/ Blanch/ det COB 1971]; 2 males, 2 females [Chile: 20 km./ SE. Vic- toria/ P. Malleco/ XI-3-1967] [Collectors: L&/C.W. O’Brien]; | male, | female [Chile: 38 km./ SE. Victoria/ P. Malleco/ XI-3- 1967] [Collectors: L &/ C. W. O’Brien]; 1 female [CHILE: Laguna de Copa/ P. N. de Puyehue, Osorno Pro./ Site 20, El. 520 m. 7-XII-77/ Valdivian Rain Forest/ Ash- worth, Hoganson, Mooers] [on Berberis darwinii] [Anthonomus/ ornatus/ Blan- chard]; 14 males, 8 females [CHILE: La- guna Espejo/ P. N. de Puyehue, Osorno Pro./ Site 17, El. 520 m. 21-XII-77/ Valdivian Rain Forest/ Ashworth, Hoganson, Mooers] {on Berberis darwinii] [Anthonomus/ or- natus/ Blanchard]; | male, 2 females [CHI- LI/ Panquipulli/ Valdivia/ 12 Jan 1944] [G. Kuschel/ Collection] [Entomology/ Divi- sion / D.S.I.R./ New Zealand]. Total para- types, 95; (CACA, CWOB, MNNC, NZAC, TAMU). Recognition (Figs. 5, 6). Adults of 4. kus- cheli are distinguished by the following combination of characters: Elytra (Figs. 5, 6) with posterolateral mac- ulae; pallid dorsal annuli around maculae uninterrupted by dark portion on inter- stria 2; sutural elytral interstriae without apical extensions; aedeagus (Fig. 27) sym- metrical, narrowed apically, with slight subapicolateral prominences; pygidium of female (Fig. 15) with subapical prom- inences; sternum 5 of female with small apicomedian prominence; endophallus (Fig. 27) with two dentate median scler- ites in addition to distal sclerite. They resemble A. ornatus but are more elongate in body form and have the discal area of the elytra more flattened so that the basal and discal elevations of interstria 3 are not so distinct. In addition, the pallid annuli around the dorsal portions of the posterolateral elytral maculae are complete in A. kuscheli (Figs. 5, 6), not interrupted by a dark portion of interstria 2 as in 4. ornatus (Figs. 3, 4). The anterior margin of the macula is also less irregular in A. kus- cheli. Additional differences are found in the male genitalia and in the female abdomen and pygidium. The aedeagus of 4. kuscheli VOLUME 91, NUMBER | is strongly narrowed toward the apex and the apicolateral prominences are weakly de- veloped (Fig. 27). The deep posteromedian incisions of sternum 5 of the female A. or- natus are replaced in A. kuscheli by a small apicomedian prominence. The female py- gidium has large subapical prominences in A kuscheli (Fig. 15), whereas the promi- nences are weakly developed in 4. ornatus (Fig. 14). Adult male. Length: 2.72-3.00 mm ( 2.86, n = 10). Width: 1.22-1. 1.26, n = 10). Rostrum: length 1.31-1.59 x (x = 1.46, n = 10) pronotal length; length of distal portion 24-33% (X = 28, n = 10) of total rostral length. Prothorax: pronotum without distinct subapical fascia and dor- solateral vittae of pallid scales. E/ytra: in- terstria 3 with slightly elevated basal por- tion, discal elevation feebly developed; dark integument and scales predominant on bas- al and discal elevations of interstria 3 and around edges of large, posteromedian mac- ulae; pallid integument and scales predom- inant on sides behind humeri and in com- plete annuli around dorsal portions of posterolateral maculae. Abdomen: postero- median portion of sternum 5 slightly de- pressed behind transverse impression. Legs: protibia with slight ventral marginal prom- inence at proximal '3; metatibia narrow, straight, with slight ventral marginal prom- inence at proximal 4; metatibial mucro large, curved, excavated. Adult female. Length: 2.60-3.12 mm (x = 2.88, n = 10). Width: 1.12-1.40 mm (x = 1.26,n = 10). Rostrum: length 1.38-1.61 x (¥ = 1.49, n = 10) pronotal length; distal portion subcylindrical, shallowly punctate, glabrous, length 29-38% (x = 33, n = 10) total rostral length. Abdomen: sternum 5 with small apicomedian prominence. Legs: metatibial mucro short, curved. Larva (8 specimens from Frutillar, Llan- quihue, Chile, collected October 31, 1983, by G. Kuschel, ex flower buds of Berberis darwint). Body (Fig. 31): curved; length 3.1— 5.4 mm (n = 8). Asperities rounded to sub- 103 conical, fairly evenly distributed over body. Color dingy white, except for brownish pig- mented pronotal area. Head (Fig. 32): dark brown; width of head capsule 0.51-0.54 mm (n = 8). Basal article of antenna bearing elongate-oval accessory appendage and three short, stout setae, one of which is longer than others. Endocarina (endocarinal line) distinct, slightly longer than '2 length of frons. Epicranial suture (coronal suture) less than '2 length of head capsule. Four pairs of frontal setae; setae | and 3 short, seta 4 ca. 2x length of 3, seta 5 distinctly longer than 4; seta 3 located dorsolaterad of 4. Two pairs of frontal sensilla, one pair in front of frontal setae | and other pair located dor- solaterad of setae 3. Five pairs of dorsal epicranial setae; setae 1, 4 and 5 slender, long, ca. equal in length; setae 2 and 3 nearly equal in length, distinctly shorter than other dorsal epicranial setae; seta 3 remote from frontal suture; seta 4 closer to frontal suture than seta 1. Four pairs of minute, peglike posterior epicranial setae that tend to be arranged in a slightly curved line; the lower seta of series located directly above or slightly mesad of dorsal epicranial seta 2. Three pairs of posterior epicranial sensilla; one pair near vertex; one pair dorsad and mesad of dorsal epicranial seta 1; one pair about halfway between dorsal epicranial se- tae 4 and 5. Lateral epicranial seta | short, about 4 length of seta 2. Ventral epicranial setae 1 and 2 ca. equal in length. Clypeus (Fig. 35) with an oblique narrow, nonpig- mented area on each side. Clypeal setae slender; seta | slightly longer and some- times located closer to anterior margin of frons than 2. One pair of clypeal sensilla that are located closer to seta | than to 2. Three pairs of labral setae: setae | and 2 ca. same length; seta 3 shorter. Epipharynx (Fig. 34) (epipharyngeal lining) bearing two pairs of anteromedian setae of ca. same length. Three pairs of stout anterolateral setae pres- ent. Labral rods (tormae) stout, converging posteriorly. Three epipharyngeal sensory pores arranged in single cluster. Two pairs 104 of stout median epipharyngeal spines pres- ent between labral rods (tormae). Mandible (Fig. 37) with two well-defined teeth; oc- casionally outermost tooth considerably re- duced; two slender setae ca. same length; one sensillum. Maxillary palps (Fig. 33) with apical article shghtly longer than basal ar- ticle, bearing several minute papillae at apex and sensillum; basal article with short seta and two sensilla. Stipial setae 1, 3 and 4 long, seta | slightly longer than other two; seta 2 much shorter than others. Mala (la- cinial lobe) bearing | 1 stout setae, 6 ventral (ventral lacinial) and 5 dorsal (dorsal lacin- ial). Labial palpus consists of one segment that bears several small papillae at apex and sensillum. Premental sclerite with long pos- terior process. One pair of long premental (prelabial) setae present. Three pairs of glos- sal setae of ca. equal size. Postmental setae (postlabial) 1 and 3 each shorter than 2. Thorax (Fig. 31) (setae described on one side of body only): pronotum bearing 8 long setae and two short ones. Five postdorsal setae present; setae 2 and 5 shorter than others. Two spiracular (alar) setae present; one several times longer than other. Epi- pleural (dorsopleural) lobe bearing one long seta. Two long pleural (ventropleural) setae on prothorax; one seta on each of meso- and metathorax. Three long pedal (laterosternal) setae. Sternal (mediosternal) seta minute. Abdomen (Fig. 31): posterior margins of most or all of first seven abdominal seg- ments emarginate dorsally (Fig. 36). Ab- dominal segments |—7 bearing three distinct dorsal folds. Prodorsum (fold I) of segments 1-8 with short setae. Five postdorsal (setae of abdominal segment, fold III) setae pres- ent; setae 1, 3 and 5 longer than 2 and 4. Two spiracular (alar) setae; seta 2 ca. 3 x longer than |. Epipleurum (dorsopleurum) bearing one long and one short seta. One short pleural (ventropleural) seta and pedal setae ca. same length. Two minute pedal setae present. Anus subterminal; surround- ed by four lobes, each bearing minute seta. Seven pairs of minute setae borne on seg- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ment 9. Spiracles bicameral; air tubes with 6 annul. Pupa (5 specimens from Frutillar, Llan- quihue, collected October 31, 1983, by G. Kuschel, ex flower buds of Berberis dar- wini). The pupa of A. kuscheli is similar in form and size to A. ornatus except the pos- teromedian pronotal tubercles tend to be larger and the setae are borne apically rather than subapically as usual with the latter species. The best distinguishing character appears to be the discotergal setae on terga 6, 7 and 8 that are borne on the summits of rounded tubercles rather than subapically on acutely pointed tubercles as in 4. orna- tus. Distribution (Fig. 44). The type series of A. kuscheli consists of adult specimens from the following localities. ARGENTINA. Rio Negro: Llao Llao, Lago Nahuel Huapi. CHILE. Aisén: 7 km. W Coyhaique, Rio Correntoso. Bio-Bio: Pemehue. Cautin: Volcan Llaima. Chiloé: Chepu. Llanquihue: Frutillar. Malleco: 19 km. E Manzanar, Cordillera Nahuelbuta, Pichinahuel, 20 and 38 km. SE Victoria. Valdivia: Panquipulli. Plant associations. Specimens of A. kus- cheli were collected on Berberis darwini at Chepu, Chiloe, and Frutillar, Llanquihue; and on B. buxifolia at Rio Correntoso, Aisén, Chile. Larvae and pupae were taken from flower buds of B. darwini. Etymology. This species is named in hon- or of Guillermo Kuschel to commemorate his contributions to the understanding of the weevil fauna of Chile and as a token of appreciation for help extended in the prep- aration of this paper. Anthonomus araucanus Clark, New SPECIES Figs. 7, 8, 16, 28, 45 Type specimens. Holotype: Chile, male {(CHILE/ Laguna Laja/ Los Barros 1500 m/ 14 Jan 1948] [on/ Berberis] [G. Kuschel/ Collection] [Entomology/ Division/ D.S.I.R./ New Zealand] (NZAC). Paratypes: Chile, 2 females [CHILE/ Laguna Laja/ Los Barros VOLUME 91, NUMBER 1 1500 m/ 14 Jan 1948] [on/ Berberis] [G. Kuschel/ Collection] [Entomology/ Divi- sion/ D.S.I.R./ New Zealand]; 2 males [Bio- Bio/ Cord. Pemehue/ 1500 m] [16 Jan 1944/ G. Kuschel] [Coll./ Kuschel] [Entomology/ Division/ D.S.I.R./ New Zealand]; | female [Bio-Bio/ La. Laja/ Los Barros/ 1500 m] [13 Jan 1945/ G. Kuschel] [Coll./ Kuschel] [Entomology/ Division/ D.S.I.R./ New Zealand]; | male, 4 females [Chile: 19 km/ E. Manzanar/ P. Malleco/ XI-3-1967] [Col- lectors L &/ C. W. O’Brien]; 2 males [CHILE, Malleco Prov./ 20 km E Manza- nar/ 1100 m 19-21.xii./ 1976 H. F. Howden] [beating]; 4 males [Malleco/ L. Malleco/ 1100 m] [22 Jan 1946/ G. Kuschel] [Coll./ Kuschel] [Entomology/ Division/ D.S.1.R./ New Zealand]; 1 female [Chile. 12 km./ W. Curicautin (sic)/ P. Malleco/ XI-3-1967] [Collectors L &/ C. W. O’Brien]; 3 males, 3 females [Chile: 20 km./ SE. Victoria/ P. Malleco/ XI-3-1967]. Total paratypes, 23; (CWOB, HAHC, NZAC, TAMU). Recognition. Adults of A. araucanus are distinguished by the following combination of characters: Elytra (Figs. 7, 8) with prominent mid- basal macula; sutural elytral interstriae without apical extensions; posterolateral elytral maculae weakly developed (Figs. 7, 8), pallid annuli around maculae bro- ken by dark portion on posteromedian portion of interstria 2; rostrum of female short, relatively stout; pygidium of female (Fig. 16) with small apicomedian prom- inence; aedeagus (Fig. 28) symmetrical, narrowed to bluntly rounded apex; en- dophallus (Fig. 28) with two dentate me- dian sclerites in addition to distal sclerite. They are similar in appearance to adults of A. berberidis. They are relatively small. In addition, the elytral patterns, with the ex- ception of the prominent, dark midbasal macula, are feebly developed in both species (cf. Figs. 7, 8, 11, 12). The two are easily distinguished; 4. araucanus lacks the apical extensions of the sutural elytral interstriae 105 and the large prominence on sternum 5 of the male. In addition, adults of A. araucanus have the aedeagus symmetrical rather than asymmetrical as in adult 4. berberidis (cf. Figs. 28, 30). Adult male. Length: 1.88-2.52 mm (x = 2.36, n = 10). Width: 0.84—-1.22 mm (x = 1.12,n = 10). Rostrum: length 1.33-1.64 x (x = 1.44, n = 10) pronotal length; length of distal portion 23-32% (X = 27, n = 10) of total rostral length. Prothorax: pronotum without subapical fascia and dorsolateral vittae of pallid scales, middorsal vitta in- distinct. E/ytra: interstria 3 with slight basal and discal elevations; dark integument and scales predominant in large, midbasal mac- ula and forming edges of weakly developed posterolateral maculae. 4bdomen: sternum 5 with slight apicomedian prominence. Legs: protibia with slight ventral marginal prom- inence at proximal '4; metatibia narrow, ventral margin slightly curved in distal 74, with slight prominence at proximal '4; me- tatibial mucro large, curved, excavated. Adult female. Length: 2.00-2.44 mm (x = 2.30, n = 10). Width: 1.02-1.22 mm (x = 1.08, n = 10). Rostrum: length 1.00-1.52 x (X = 1.37, n = 10) pronotal length; distal portion subcylindrical, shallowly punctate, glabrous, length 25-34% (x = 30, n = 10) total rostral length. 4bdomen: sternum 5 with posterior margin straight. Legs: meta- tibia straight, narrow at apex; metatibial mucro short, straight. Distribution (Fig. 45). The type series of A, araucanus consists of adult specimens from the following localities. CHILE. Bio- Bio: Laguna Laja, Cordillera Pemehue. Malleco: 12 km. W Curacautin; 20 km. E Manzanar; Laguna Malleco; 20 km. SE Vic- toria. Plant associations. The holotype and two paratypes were collected on Berberis sp. at Los Barros and Laguna Laja, Bio-Bio, Chile. Etymology. The name of this weevil is derived from that of the Araucanians and emphasizes the fact that it occurs in the heart of their country and that of the Araucaria. 106 Anthonomus chilicola Clark, NEw SPECIES Figs. 9, 10, 29, 44 Type specimens. Holotype: Chile, male [CHILE, Malleco Prov./ 20 km E Manza- nar/ 1100 m 19-21.x11./ 1976 H. F. How- den] [beating] (HAHC). Paratypes: Chile, 3 males [CHILE, Malleco Prov./ 20 km E Manzanar/ 1100 m 19-21.xi1./ 1976 H. F. Howden] [beating] (HAHC). Recognition (Figs. 9, 10). Adults of A. chilicola are distinguished by the following combination of characters: Sutural elytral interstriae with acute, overlapping apical extensions; with pos- terolateral elytral maculae (Figs. 9, 10); pallid annuli around dorsal portions of maculae extending across part of short posteromedian dark portion on interstria 2; metatibia of male with ventral margin strongly concave in distal 7, apical mucro enlarged, excavated; aedeagus (Fig. 29) asymmetrical, constricted medially, with slight subapicolateral prominences. Like adults of A. kuscheli, adult A. chilicola are more slender than those of A. ornatus (cf. Figs. 4, 10). The elytral pattern in A. chilicola is essentially the same as in A. or- natus, but the pallid annuli around the pos- terolateral maculae are partially interrupted by enroachment of dark integument and scales on interstria 2 (Fig. 10). Adult 4. chilicola are further distinguished from adults of both of these species, however, by the apical extensions of the sutural elytral interstriae, by the male metatibia which is strongly, broadly concave on the ventral margin and has a much larger, more strong- ly excavated metatibial mucro, and by the asymmetrical aedeagus. These are all char- acters adult 4. chilicola share with adult A. berberidis, which are distinguished by the less distinct elytral pattern (cf. Figs. 9-12), the large posterolateral prominence on ster- num 5 of the male, and by differences in the aedeagus (cf. Figs. 29, 30). PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Adult male. Length; 3.40-4.00 mm (x = 3.75, n = 4). Width: 1.40-1.64 mm (x = 1.57, n = 4). Rostrum: length 1.47-1.67 x (x = 1.58, n = 4) pronotal length; length of distal portion 21-31% (X = 25, n = 4) of total rostral length. Prothorax: pronotum without distinct subapical fascia and dor- solateral vittae of pallid scales. E/ytra: in- terstria 3 with slightly elevated basal and discal portions; dark integument and scales predominant on humeri and on basal and discal elevations of interstria 3, less dense on basal portions of interstriae 2, 4 and 5 and on discal portion of 2, forming broad, irregular edges of posteromedian maculae; pallid integument and scales predominant behind humeri and in oblique fascia ex- tending posteriorly to posteromedian por- tion of interstria 2 then curving posteriorly to apices of interstriae 4-6. Abdomen: pos- teromedian portion of sternum 5 slightly depressed. Legs: procoxae with dense, broad, pallid scales; protibia with slight ventral marginal prominence at proximal '3; meta- tibia narrow, slightly curved, with slight ventral marginal prominence at proximal Ys; metatibial mucro large, curved, exca- vated. Adult female. Unknown. Distribution (Fig. 44). The type series of A. chilicola consists of adult specimens from the following locality. CHILE. Malleco: 20 km. E Manzanar. Plant associations. Unknown. Etymology. The name of this species is derived from that of its native country, em- phasizing the unique austral distribution of the group to which it belongs. Anthonomus berberidis Clark, NEw SPECIES Figs.) lil 125 30745 Type specimens. Holotype: Chile, male [CHILE/ Frutillar/ 14 Dec 1943/ G. Kus- chel] [G. Kuschel/ Collection] [Entomolo- gy/ Division/ D.S.I.R./ New Zealand] (NZAC). Paratypes: ARGENTINA, | male, 1 female [ARGENTINA/ El Bolson, Rio/ VOLUME 91, NUMBER 1 Negro/ [X-20-1963/ A. Kovacs]. Chile, | male [Chili] [Deyr.]; 2 males, 4 females [CHILE: 19 km. S. of Ancud,/ Is. Chiloe/ Site C2. El. 120 m. 14-II-79/ Valdivian Rain Forest/ Ashworth, Hoganson, Gordon] [on Berberis darwin] [Anthonomus/ ornatus/ Blanchard]; 5 males, 6 females [CHILE/ Chiloé/ Chepu/ 22.10.58] [s/ Berberis/ dar- wini] [Coll./ Kuschel] [Entomology/ Divi- sion/ D.S.I.R./ New Zealand]; 3 males, 2 females [Chile: 9 km./ E. Chepu/ Chiloe Is/ II-4-1968] [at night L. &/ C. W. O’Brien]; 2 females [Chile: 13/ km. S. Castro/ Is Chi- loe/ II-5-1968] [Collectors: L &/ C. W. O’Brien]; 1 female [Dalcahue/ Chiloe, CHILE]; | female [CHILE/ Dalcahue/ Chi- loe/ 10-20 Nov 1957] [G. Kuschel/ Collec- tion] [Entomology/ Division/ D.S.I.R./ New Zealand]; 1 male [Quellon-Chi/ loe I-1955/ J.S. Vargas]; | female [CHILE: Chiloé/ Dal- cahue/ 17-31.1.1962/ Luis Pena] [H. & A. Howden/ Collection]; | male, | female [Pto Montt/ 15.1.71]; 1 female [Chile: 12 km. N./ Puerto Montt/ P. Llanquihue/ II-8- 1968] [at night L. &/ C. W. O’Brien]; 3 males, 5 females [CHILE/ Llanquihue/ Frutillar/ 13.9.54] [Coll./ Kuschel] [Entomology/ Di- vision/ D.S.I.R./ New Zealand]; | male, 3 females [CHILE, 2 km./ S. Frutillar/ P. Llanquihue/ 280’ II-3-1968] [Collectors: L &/C.W. O’Brien]; | male, | female [CHILE/ Llanquihue/ Frutillar/ 23.9.54] [Coll./ Kus- chel] [Entomology/ Division/ D.S.I.R./ New Zealand]; 3 males, 2 females [CHILE/ Llan- quihue/ Frutillar/ 24.2.50] [Coll./ Kuschel] [Entomology/ Division/ D.S.I.R./ New Zealand]; 1 male, 3 females [CHILE/ Fru- tillar/ 14 Dec 1943/ G. Kuschel] [G. Kus- chel/ Collection] [Entomology/ Division/ D.S.1.R./ New Zealand]; 1 male [CHILE/ Frutillar/ 28 Dec 1944/ G. Kuschel] [G. Kuschel/ Collection] [Entomology/ Divi- sion/ D.S.1I.R./ New Zealand]; | male, 7 fe- males [CHILE/ Frutillar/ 27 Feb 1950/ G. Kuschel] [G. Kuschel/ Collection] [Ento- mology/ Division/ D.S.I.R./ New Zealand]; 2 males, 1 female [MAULLIN a/ Las Que- mas/ Llanquihue/ 17,-Dic.1972/ Coll: L. E. 107 Pena]; | male [CHILE/ Volcan Osorno/ 4.3.50/ Kuschel leg.] [Coll./ Kuschel] [Ento- mology/ Division/ D.S.I.R./ New Zealand]. 1 male [CHILE: Laguna de Copa/ P. N. de Puyehue, Osorno Pro./ Site 20, El. 520 m. 7-XII-77/ Valdivian Rain Forest/ Ash- worth, Hoganson, Mooers] [on Berberis darwinii] [Anthonomus/ ornatus/ Blan- chard]; 3 males, 2 females [CHILE: Laguna Espejo/ P. N. de Puyehue, Osorno Pro./ Site 17, El. 520 m. 21-XII-77/ Valdivian Rain Forest/ Ashworth, Hoganson, Mooers] [on Berberis darwinii] [Anthonomus/ ornatus/ Blanchard]; 3 males, 8 females [CHILE: La- guna Espejo/ P. N. de Puyehue, Osorno Pro./ Site 17A. El. 520 m. 10-I-79/ Valdivian Rain Forest/ A. C. Ashworth, J. W. Hoganson] {on Berberis sp.] [Anthonomus/ ornatus/ Blanchard]; Total paratypes, 87; (AMNH, CACA, HAHC, MNNC, MCZC, NZAC, TAMU). Recognition (Figs. 11, 12). Adults of 4. berberidis are distinguished by the following combination of characters: Sutural elytral interstriae with acute, overlapping apical extensions; postero- lateral elytral maculae feebly developed (Figs. 11, 12), surrounded dorsally by complete pallid annuli; sternum 5 of male with large posteromedian prominence; metatibia of male with ventral margin strongly concave in distal 74, with en- larged, excavated, apical mucro; pygid- ium of female with small apicomedian prominence; aedeagus (Fig. 30) asym- metrical, constricted medially, with blunt subapicolateral prominences. These, and adults of the closely related A. chilicola have the aedeagus similarly asym- metrical (cf. Figs. 29, 30), similarly modi- fied male metatibiae with the ventral mar- ginal concavity, enlarged, strongly excavated apical mucrones, and similar extension of the extreme apices of the sutural elytral in- terstriae. This latter feature is even more pronounced in adult 4. berberidis where the 108 extension of the right elytron overlaps that of the left one. The elytral pattern of light and dark integument and scales is also sim- ilar in adult 4. chilicola and A. berberidis, but is weakly developed in adult 4. berber- idis, which have a prominent midbasal macula (cf. Figs. 9, 12). The large apico- median prominence on sternum 5 of the male is unique to adults of A. berberidis. Adult male. Length: 2.52-3.04 mm (x = 2.83, n = 10). Width: 1.12-1.44 mm (% = 1.28, n = 10). Rostrum: length 1.02-1.48 x (X = 1.32, n = 10) pronotal length; length of distal portion 22-31% (¥ = 25, n = 10) of total rostral length. Prothorax: pronotum without subapical fascia and dorsolateral vittae of pallid scales; middorsal vitta in- distinct. E/ytra: interstria 3 without distinct basal and discal elevations; dark integu- ment and scales predominant in large, mid- basal macula and forming edges of weakly developed posterolateral maculae. Legs: protibia with distinct ventral marginal prominence at proximal '/3; metatibia nar- row, ventral margin strongly curved in dis- tal 74, with distinct prominence at proximal Y4; metatibial mucro large, curved, exca- vated. Adult female. Length: 2.16-3.16 mm (x = 2.87, n = 10). Width: 0.92-1.36 mm (x = 1.26, n = 10). Rostrum: length 0.87-1.40 x (X = 1.28, n = 10) pronotal length; distal portion subcylindrical, shallowly punctate, glabrous, length 26-33% (xX = 29, n = 10) total rostral length. Abdomen: sternum 5 with posterior margin straight. Legs: meta- tibia straight, narrow at apex, apical mucro short, straight. Distribution (Fig. 45). The type series of A. berberidis consists of adult specimens from the following localities. ARGEN- TINA. Rio Negro: El Bolson. CHILE. Chi- loé: 19 km. S Ancud; 13 km. S Castro; Che- pu; Dalcahue; Puerto Montt; 12 km. N Puerto Montt; Quellon. Lianquihue: Frutil- lar; 2 km. S Frutillar; Maullin. Osorno: La- guna la Copa and Laguna Espejo, Parque Nacional de Puyehue; Volcan Osorno. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Plant associations. Specimens of A. ber- beridis were collected on Berberis darwini 19 km. S Ancud and at Chepu, Chiloé, and at Laguna la Copa and Laguna Espejo, Par- que Nacional de Puyehue, Osorno, Chile. Label data indicate that adults have been collected in January, February, March, Sep- tember, October, November and Decem- ber. Etymology. The name of this species is derived from that of the plant genus with which it and several other 4. ornatus group members are associated. PHYLOGENY The classification of the subfamily An- thonominae has been called “chaotic” (Burke 1976). Some 500, mostly New World, named species are assigned to the subfam- ily, but recent revisionary studies (Clark 1987a, b, c, 1988, Clark and Burke 1985, 1986a, b, c) indicate that the actual number of species may be twice that. Within this unwieldy mass of species, some apparently natural groupings do present themselves, but it is not often possible to identify the sister groups of these. This is the case with the A. ornatus group. The group seemed by pre- liminary inspection to be natural, but iden- tification of the sister group remains for fu- ture studies. Thus, in order to apply the outgroup method of comparison to identify apomorphic characters in the 4. ornatus group, it was necessary to attempt to iden- tify characters unique to the group by ex- amining a broad, though incomplete, array of anthonomine taxa. Further studies may reveal that some of these characters are more widely distributed. The most detailed comparisons included those of characters of the 4. ornatus group with those of anthonomines examined in connection with ongoing revisionary stud- ies. These include the species in the An- thonomus albolineatus, furcatus, grandis, gularis, unipustulatus, and venustus groups, the species of the Anthonomus subgenera Anthomorphus, Anthonomocyllus and An- VOLUME 91, NUMBER | 109 107,12 a signatipennis ornatus ace blanchardi 2,4,7,8- kuscheli ee | 3,5,9-——] Fig. 46. thonomorphus, the genera Atractomerus, Loncophorus, Omogonus and Pseudantho- nomus, and the types of most of the named Neotropical Anthonominae. Numerous Nearctic and Old World species have also been examined, but not in as much detail. There is no indication that any of the species examined should be placed in the 4. ornatus group, nor are there indications that any of them is the sister group of the 4. ornatus group. As a result of the comparisons made, 13 characters (listed below) were determined to be unique to adults of the 4. ornatus group. The distribution of these characters among the seven species in the group is de- picted in Table 1. Presence 1s indicated by a score of “1,” absence by a score of “0,” “missing,” because the female of 4. chili- araucanus Phylogenetic tree depicting hypothesized relationships of the species of the A. ornatus group. cola is unknown, by a score of 9. These characters were analyzed using the PAUP computer programs developed by Swofford (1985), on an IBP Personal Computer to determine the most parsimonious branch- ing pattern consistent with this distribution of characters. Analysis of the 13 characters, all given equal weight, with the ALLTREES option which searches all possible trees, produced the phylogenetic tree depicted in Fig. 46. The characters, with comments on their distributions under the constraints of the accepted solution, are listed below. 1) Elytral integument with markedly con- trasting patterns of pallid and dark por- tions clothed with correspondingly pal- lid and dark scales ranging from white through pallid to dark ferruginous to Table 1. Data matrix for the phylogenetic tree in Fig. 46. Characters 0 0 0 0) 0 0 0 0 0 1 ] 1 ] | 2 3 4 5 6 7 8 9 0 I 2 3 Outgroup 0 0 0 0 0 0 0 0 0 0 0 0 (0) signatipennis 1 0 0 0 0 l 0 0 0 0 0 0 0 ornatus l 1 0 1 0 I 0 i 0) l l ] 0 blanchardi 1 1 0 1 0 1 0 1 0 1 ] 1 (0) kuscheli 1 l 0 1 0 1 1 l 0 l ] 0 0 araucanus l 1 0 0 0 1 1 1 0) 0 0 0 l chilicola 1 1 1 1 l l 0 l 1 9 9 9 9 ] 1 1 1 1 0 0 1 berberidis 110 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON fuscous (Figs. 1-12). Although con- trasting patterns of light and dark in- tegument and light and dark scales are not unique to the A. ornatus group, the patterns are similar in the species in the group and do not seem to be exactly duplicated in any of the other anthon- omines. 2) Elytra with large, posterolateral mac- ulae (Figs. 3-12). These are weakly de- veloped in A. araucanus and A. ber- beridis (Figs. 7, 8, 11, 12). 3) Sutural elytral interstriae with acute, overlapping apical extensions. 4) Aedeagus with apicolateral promi- nences (Figs. 21, 23, 25, 27, 29). These are assumed to have been lost in 4. araucanus (Fig. 28) and A. berberidis (Fig. 30). 5) Aedeagus asymmetrical (Figs. 29, 30). 6) Endophallus with an elongate, flat, dis- tal sclerite (Figs. 19-30). 7) Endophallus with two dentate median sclerites (Figs. 27, 28). These are as- sumed to have been secondarily lost in the ancestor of A. ornatus and A. blan- chardi and in that of A. chilicola and A. berberidis. 8) Parameres of tegmen of male genitalia close together or fused at base. 9) Metatibia of male with ventral margin strongly concave in distal 74, with en- larged, excavated, apical mucro. 10) Pygidium of female with subapical prominences (Figs. 14, 15). These are weakly developed and assumed to be obsolescent in 4. ornatus (Fig. 14) and A. blanchardi. 11) Sternum 5 of female with apicome- dian prominence. This is a simple prominence in 4. kuscheli, a promi- nence bounded by shallow emargina- tions in A. blanchardi (Fig. 18), a long prominence bounded on each side by deep incisions in 4. ornatus (Fig. 17). 12) Sternum 5 of female incised or emar- ginate (Figs. 17, 18). 13) Pygidium of female with apicomedian prominence (Fig. 16). ACKNOWLEDGMENTS Thanks are extended to the individuals and institutions listed in the Materials and Methods section for arranging loans of the specimens used in this study. Guillermo Kuschel initially suggested the group as suit- able for revision, provided information and other valuable assistance, along with a crit- ical review of the manuscript. Charles W. O’Brien and Guillermo Wibmer also re- viewed the manuscript. This paper is pub- lished as Alabama Agricultural Experiment Station Journal Series No. 17-881496P and Texas Agricultural Experiment Station T.A. 23182. LITERATURE CITED Ahmad, M. and H. R. Burke. 1972. Larvae of the weevil tribe Anthonomini (Coleoptera: Curculion- idae). Misc. Publ. Entomol. Soc. Amer. 8: 31-81. Anderson, W. H. 1947. A terminology for the ana- tomical characters useful in the taxonomy of wee- vil larvae. Proc. Entomol. Soc. Wash. 49: 123- 132. Ashworth, A. C. and J. W. Hoganson. 1987. Co- leoptera bioassociations along an elevational gra- dient in the Lake Region of southern Chile, and comments on the postglacial development of the fauna. Ann. Entomol. Soc. Amer. 80(6): 865-895. Blackwelder, R. E. 1947. Checklist of the coleopte- rous insects of Mexico, Central America, the West Indies and South America, Part 5. Bull. U.S. Natl. Mus. (185): I-IV, 765-925. Blanchard, C. E. 1851. Fauna chilena. Insectos. Co- ledpteros, pp. 286-429. In Gay, C., ed., Historia fisica y politica de Chile, vol. 5, Zool., pp. 285- 563. Burke, H. R. 1968. Pupae of the weevil tribe An- thonomini (Coleoptera: Curculionidae). Tech. Monog. 5, Texas Agric. Exp. Sta., 92 pp. 1976. Bionomics of the anthonomine wee- vils. Ann. Rev. Entomol. 21: 283-303. Cabrera, A. L. and A. Willink. 1973. Biogeografia de América Latina. Organizacion de los Estados Americanos, Washington, D.C., vi + 120 pp. Clark, W. E. 1987a. Revision of the unipustulatus group of the weevil genus Anthonomus Germar (Coleoptera: Curculionidae). Coleopts Bull. 41(1): 73-88. 1987b. Revision of the Nearctic species of Pseudanthonomus Dietz (Coleoptera: Curculion- idae). Coleopts Bull. 41(3): 263-285. 1987c. Revision of the Anthonomus subge- nus Anthomorphus Weise (Coleoptera: Curculion- idae). Quaest. Entomol. 23: 317-364. VOLUME 91, NUMBER 1 1988. The species of Anthonomus in the al- bolineatus group (Coleoptera: Curculionidae). Trans. Amer. Entomol. Soc. 113: 309-359. Clark, W. E. and H. R. Burke. 1985. Revision of the venustus species group of the weevil genus An- thonomus Germar (Coleoptera: Curculionidae). Trans. Amer. Entomol. Soc. 111: 103-170. 1986a. Revision of the gu/aris species group of the genus Anthonomus Germar (Coleoptera: Curculionidae). Coleopts Bull. 40(1): 1-26. 1986b. Phylogeny of the species of the An- thonomus subgenus Anthonomorphus Dietz, with discussion of relationships with Anthonomus grandis Boheman (Coleoptera: Curculionidae). J. Kans. Entomol. Soc. 59(3): 508-516. 1986c. A new neotropical species of Anthon- omus Germar (Coleoptera: Curculionidae) asso- ciated with Bombacopsis quinata (Bombacaceae). Proc. Entomol. Soc. Wash. 88(2): 320-327. Clark, W. E. and R. P. Martins. 1987. Anthonomus biplagiatus Redtenbacher (Coleoptera: Curculion- idae), a Brazilian weevil associated with Kielmey- era (Guttiferae). Coleopts Bull. 41(2): 157-164. Cronquist, A. 1981. An Integrated System of Clas- sification of Flowering Plants. Columbia Univer- sity Press, New York. xviii + 1262 pp. Davis, D. R. 1986. A new family of Monotrysian moths from Austral South America (Lepidoptera: Palaephatidae), with a phylogenetic review of the Monotrysia. Smiths. Contr. Zool., 434. Smithson- ian Institution Press, Washington, D.C., iv + 202 pp. 11 Fairmaire, L. 1884. Note sur quelques Coléoptéres de Magellan et de Santa-Cruz. Ann. Soc. Entomol. France, 1883 (1884), ser. 6, 3: 483-506. Good, R. 1964. The Geography of Flowering Plants. Longmans, Green and Col, Ltd., London. Kuschel, G. 1950. Nuevas sinonimias, revalida- ciones y combinaciones (9° aporte a Col. Curcu- lionidae). Agric. Téc., Chile 10(1): 10-21. 1960. Terrestrial zoology in southern Chile. Proc. Royal Soc. London, series B. 152: 540-550. O’Brien, C. W. and G. J. Wibmer. 1982. Annotated checklist of the weevils (Curculionoidea senus lato) of North America, Central America, and the West Indies (Coleoptera: Curculionidae). Mem. Amer. Entomol. Inst. (34): i-ix, 1-382. Raven, P. H. 1963. Amphitropical relationships in the floras of North and South America. Quart. Rev. Biol. 38(2): 151-177. Schenkling, S. and G. A. K. Marshall. 1934. Coleop- terorum Catalogus, Pars 139, Curculionidae: An- thonominae, pp. 3-82; Laemosaccinae, pp. 1-8. (vol. 29). Swofford, D. L. 1985. PAUP: Phylogenetic Analysis Using Parsimony. Users Manual, Illinois Natural History Survey, Champaign, Ill. Thomas, J. B. 1957. The use of larval anatomy in the study of bark beetles (Coleoptera: Scolytidae). Can. Entomol. 89, Suppl. 5, 45 pp. Wibmer, G. J. and C. W. O’Brien. 1986. Annotated checklist of the weevils (Curculionidae sensu lato) of South America (Coleoptera: Curculionoidea). Mem. Amer. Entomol. Inst. (39): i-xvi, 1-563. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 112-123 A REVIEW OF THE WORLD SPECIES OF NOTOGLYPTUS MASI (HYMENOPTERA: PTEROMALIDAE) Steve L. HEYDON Department of Entomology, NHB; Mail Stop 165, Smithsonian Institution, Washing- ton, D.C. 20560. Abstract. — Until the present review, Notog/yptus Masi was a monotypic genus containing the single species N. virescens Masi, which was known from throughout the Old World. This species is redescribed, and its known distribution is extended to the Nearctic region. Four new species of Nofoglyptus from the New World are described: N. bidentatus, N. luteicrus, N. nesiotes and N. tzeltales. Notoglyptus is redescribed, its relationships to other genera of the Miscogasterinae are discussed, and a key is given for distinguishing the known world species. The host(s) of species in this genus are still unknown. Key Words: taxonomy Masi (1917) described the genus Noto- glyptus and included two species, N. vires- cens Masi from the Seychelles Islands, and N. niger Masi from Italy. He designated the latter species as the type-species of the ge- nus. Boucek (1976) synonymized N. niger with N. virescens. This cosmopolitan species was by then known to occur throughout southern Europe, across to India, and down into southern Africa (Graham 1969, Boucek 1976). I have seen specimens from Japan and a single male from North America. During the course of sorting collections for a continuing study of the Nearctic misco- gasterine Pteromalidae, I came across spec- imens belonging to four undescribed species of Notoglyptus from the New World—N. bi- dentatus n. sp. from South America, N. tzel- tales n. sp. from Central America and north- ern South America, N. /uteicrus n. sp. from the continental Nearctic region south to Venezuela, and N. nesiotes n. sp. from the Caribbean and coastal southeastern U.S. Graham (1969) placed Notog/yptus in the Sphegigasterini (Miscogasterinae), probably Hymenoptera, Pteromalidae, Notoglyptus, Nearctic, Palearctic, Neotropics, because the elongate T1 is found in other genera he placed there such as Crypto- prymna Forster and Novitzkyanus Boucek. Genera with an elongate T1 occur sporad- ically throughout the Pteromalidae, and I feel its common occurrence in these three genera is due to convergence (Heydon 1988). The closest relatives of Notoglyptus are among genera related to Halticoptera Spi- nola and Thinodytes Graham. These genera all share a similar unique propodeal struc- ture—the median panels of the propodeum are reticulately sculptured or smooth and have a distinct median carina and plicae connected posteriorly by a W-shaped carina (compare Figs. 12 and 13). This carina is most distinctly developed in N. bidentatus and N. fzeltales, the most morphologically generalized Notoglyptus species. The ter- minal two segments of the maxillary palps of male Halticoptera are lamellately ex- panded, and this character state also occurs in Notoglyptus luteicrus (Fig. 11). However, using this character state as direct evidence for a close relationship between Halticop- VOLUME 91, NUMBER 1 teraand Notoglyptus is complicated because N. luteicrus is one of the more morpholog- ically derived members of the genus; this character state is not present in the more primitive extant species of Notog/yptus. The characters defining Notoglyptus are reviewed in the Discussion section follow- ing the generic description below. Noto- glyptus may have evolved in South America since that is where the most primitive species (N. bidentatus and N. tzeltales) are found. MATERIALS AND METHODS This study is based on examination of 67 specimens from the museums whose acro- nyms are given 1n the Acknowledgment sec- tion below. Type depositions are given in parentheses in the appropriate section un- der each species description. Terminology in this paper generally follows that of Gra- ham (1969), except that club is used instead of clava and the gastral terga are numbered 1-7 starting with the basal tergite of the gas- ter. The following abbreviations are used: the multiporous plate sensillae are MPP sensillae, the lower ocular line is LOcL, the antennal funicular segments are F1, F2,.. , F6, and the gastral terga are T1, T2,.. ,T7. The units of measurement given in the de- scriptions can be converted to millimeters by multiplying by 0.02. Notoglyptus Masi Notoglyptus Masi, 1917: 181. Type species N. niger Masi (original designation). Ga- han and Fagan, 1923: 98. Peck, Boucek, and Hoffer, 1964: 36 (key). Graham, 1969: 124 (key), p. 140. Boucek, 1976: 15. Dzhanokmen, 1978: 77 (key), p. 80. Fa- rooqi and Subba Rao, 1985: 259, 310F, 310G; 1986: 295. Description. Color: Body ranges from black to metallic green or blue; legs metallic or entirely nonmetallic. Structure: Head ovate (Fig. 7) or triangular (Fig. 1) in an- terior view; clypeus with anterior margin straight or produced (Fig. 7), bidentate in 113 N. bidentatus (Fig. 1); genal concavities well developed, extending 2 to 5 malar length. Antenna inserted near or below LOcL; with 2 anelli, 6 funicular segments, and 3-seg- mented club; MPP sensillae in 1 row on each funicular segment (Figs. 3, 5); club often with micropilosity on terminal segment(s) (Figs. 3, 5), apical spine not present. Meso- soma with pronotum reduced, much nar- rower and depressed relative to mesoscu- tum (Figs. 8, 10, 12), horizontal collar developed in N. bidentatus (Fig. 8), N. tzel- tales, and N. virescens but not present in N. luteicrus (Fig. 10) and N. nesiotes (Fig. 12), this collar bordered anteriorly by weak to strong anterior transverse carina; mesoscu- tum with notauli complete, groovelike (Figs. 8, 10, 12); upper epimeron smooth; scutel- lum with scutoscutellar sulcus foveolate medially in N. nesiotes (Fig. 12) and N. vi- rescens, frenal sulcus distinct (Figs. 8, 10, 12), frenum weakly coriaceous or smooth; propodeum with median panels alveolate or smooth, plicae and median carina distinct, connected posteriorly by W-shaped carina (Fig. 9). Wing hyaline; relative lengths of veins as submarginal > marginal > post- marginal > stigmal; basal cell and vein se- tate (Figs. 2, 4) except in N. tzeltales (Fig. 6); fore wing sometimes with distinct ad- marginal setae. Petiole quadrate to elongate, granulate to alveolate; median carina some- times present; basal flange present. Gaster ovate; Tl nearly concealing succeeding ter- ga, hind margin straight; hypopygium ex- tending nearly to tip of gaster. Male max- illary palps pale in all species except N. bidentatus, terminal two segments lamel- lately expanded in N. /uteicrus (Fig. 11). Discussion. The following combination of character states will reliably distinguish Notoglyptus: the presence of distinct genal concavities; a 13-segmented antenna; com- plete, groovelike notauli (Figs. 8, 10, 12); the propodeum with the median carina and plicae distinct, and connected posteriorly by W-shaped carina (Fig. 9); the petiole about as long as wide, with a basal ventral flange; 114 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 1-7. (ww (exe) 1-2, Notoglyptus bidentatus n. sp., male. 1, Anterior view of head. 2, Fore wing. 3, Notoglyptus luteicrus n. sp., female. 3. Antenna. 4—5, Notoglyptus nesiotes n. sp., female. 4, Fore wing. 5, Antenna. 6-7, Notoglyptus tzeltales n. sp., female. 6, Fore wing. 7, Anterior view of head. Tl almost completely covering the entire length of gaster, with its hind margin entire; and the hypopygium extending to the tip of the gaster. An autapomorphy defining No- toglyptus is the reduction in size of the pronotum relative to the rest of the meso- soma. This character 1s easily seen in com- parison with other closely related misco- gasterine genera such as Halticoptera (compare Figs. 8, 10, 12 with 13). The unique fovea in the center of the scutellum sO prominent in the type species is clearly ofno more than specific value when looking at this genus on a worldwide basis. The species of Notoglyptus can be divided into two distinct species-groups. Notoglyp- tus bidentatus and N. tzeltales comprise the first group which is characterized by a large body size, the body and legs with extensive bright metallic coloration, the anterior mar- gin of the clypeus produced (Figs. 1, 7), the toruli located below the LOcL (Figs. 1, 7), the female antennal club with no micropi- losity, the fore wing lacking any distinct ad- marginal setae, and the petiole without a median carina. Except possibly for the low insertion of the antennae, these characters states are all plesiomorphic compared with those defining the other species-group. No- toglyptus luteicrus, N. nesiotes, and N. vi- rescens form the second species-group. These species are relatively small, have the VOLUME 91, NUMBER 1 Figs. 8-13. 8-9, Notoglyptus bidentatus n. sp., male. 8, Dorsolateral view of mesosoma. 9, Frenum, dorsellum, and propodeum. 10-11, Notoglyptus luteicrus n. sp., male. 10, Dorsolateral view of mesosoma. 11, Maxillary palps. 12, Notoglyptus nesiotes n. sp., female. 12, Dorsolateral view of mesosoma. 13, Halticoptera sp., female. 13, Dorsolateral view of mesosoma. Scale bar = 0.1 mm. body color very dark, the legs (including the antennal club with micropilosity, the fore coxae) yellow and nonmetallic, the anterior wing with distinct admarginal setae, and the margin of the clypeus nearly straight, the petiole with a median carina. toruli located above the LOcL, the female Despite these obvious morphological dif- 116 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ferences, the species I have placed together in Notoglyptus belong together as a mono- phyletic unit within the Miscogasterinae on the basis on the reduced size of the prono- tum, and with respect to closely related gen- era such as Halticoptera, because of the elongate first gastral tergite and hypopyg- ium. Division of the two species-groups into separate genera would result in the species- group containing Notoglyptus bidentatus and N. tzeltales being paraphyletic. Further, the important character of the presence or ab- sence of the horizontal pronotal collar cuts across the species-groups as I have outlined them here. Notoglyptus virescens has a hor- izontal collar as do N. bidentatus and N. tzeltales; yet, in the characters separating the species-groups, N. nigrescens belongs with N. /uteicrus and N. nesiotes. Key TO WORLD SPECIES OF NoOTOGLYPTUS Masi 1. Pronotum lacking horizontal collar, sloping downward immediately from anterior margin of mesoscutum (Figs. 10, 12) ; — Pronotum with short horizontal collar sepa- rated from declivitous neck by a weak to strongly developed transverse carina (Fig. 8) .. 3. 2. Scutellum with a pair of diverging foveae at base (Fig. 12). Dorsellum cariniform. Female antenna with funicular segments 5—6 quadrate to elongate (Fig. 5). (West Indies and coastal southeastern United States) nesiotes Heydon — Scutellum with scutoscutellar sulcus a contin- uous deep furrow, not foveolate (Fig. 10). Dor- sellum bandlike, length nearly half that of fre- num. Female antenna with funicular segments 5-6 transverse (Fig. 3). (Continental North America south to Venezuela) luteicrus Heydon 3. Scutellum with discal fovea. Torulus above lower ocular line. Coxae and femora yellow, nonmetallic .. virescens Masi — Scutellum smoothly convex (Fig. 8). Torulus below lower ocular line (Figs. 1, 7). Coxae and femora blue or green, metallic. (Neotropical) 4. 4. Head triangular in anterior view; anterior mar- gin of clypeus bidentate (Fig. 1). Basal cell of fore wing setate (Fig. 2). Body length 2 mm or more. (South America) bidentatus Heydon — Head ovate in anterior view; anterior margin of clypeus truncate (Fig. 7). Basal cell of fore wing bare (Fig. 6). Body length 1.8 mm or less. (Central America and northern South Ameri- ca) tzeltales Heydon io) Notoglyptus bidentatus, New SPECIES Figs. 1-2, 8-9 Holotype, female. Description. Color: Head dark green; mesosoma mostly dark green with strong coppery reflections on dorsum and upper epimeron; pronotum, propodeum, petiole, gaster bluish black. Antenna with scape, pedicel dark green; fla- gellum black. Maxillary palp brown. Legs with coxae dark blue; fore, middle trochan- ters orange-yellow, hind trochanter brown; femora blue-green, orange-yellow basally and apically; tibiae orange-yellow basally, remainder orange-brown with weak metal- lic reflections medially; tarsi dark brown. Wing veins brown. Sculpture: Clypeus, gena finely coria- ceous; head finely alveolate otherwise. Mesosoma (Fig. 8) with middle lobe of mesoscutum alveolate; side lobes, scutel- lum, frenum (weakly) finely coriaceous; dorsellum smooth; median panels of pro- podeum alveolate (Fig. 8); petiole finely al- veolate, alveoli 2x as long as wide; gastral terga smooth. Structure: Body length 2.4 mm. Head (Fig. 1) triangular in anterior view, width 1.4 x height (46:34), 2.2 x length (46:21); clypeus separated from face by deep furrowlike ca- rina, anterior margin produced and biden- tate; genal concavity shallow, extending 1 malar distance; frons regularly concave be- tween eyes; eye height 1.4 length (16.5: 12.0), 1.1 malar length (16.5:15.0), eye length 1.5 temple length (12:8); ratio of MOD, OOL, POL, LOL as 3:11:8:4. An- tenna with torulus one inside diameter be- neath LOcL; combined length of pedicel and flagellum 0.85 head width (39:46); rela- tive lengths of scape, pedicel, anelli, Fl-6, club as 22.0:7.0:3.0:4.0:4.5:4.0:4.0:4.0:3.5: 7.0; widths of Fl, F6, club as 3:5:5; second anellus 2x as large as first; club without micropilosity. Mesosoma (Fig. 8) length 1.6 width (59:36); collar with weak an- terior transverse carina; scutellum regularly rounded, scutoscutellar sulcus continuous VOLUME 91, NUMBER 1 furrow; dorsellum bandlike, length equal to frenal length; spiracles ovate, <1 x own di- ameter from anterior margin of propodeum;: nucha bordered by carina anteriorly. Fore wing (Fig. 2) length 2.3 width (145:63); ratio of submarginal, marginal, postmargin- al, stigmal vein lengths as 50.0:30.0:24.0: 13.5; costal cell densely setate; basal cell setate to base along anterior margin; spe- culum closed posteriorly; no distinct ad- marginal setae present. Petiole broadening apically, length 1.8 x apical width (15.0:8.5); lacking median carina. Gaster length 1.4 x width (42:30); deep, height 1 x width (30: 30). Allotype, male. Color: Similar to holo- type female except front of head blue-green with violet reflections, anelli and funicular segments tan ventrally. Structure: Body length 2.4 mm. Antenna with combined length of pedicel and flagellum 0.93 x head width (40:43); lengths of scape, pedicel, anelli, Fl-6, club as 20:5:3:4:4:4:4:4:4:9; widths of Fl, F6, club as 3.5:4.0:4.0; funic- ular segments tend to be hemispherical. Maxillary palps slender. Petiole length 2.2 x width (15.5:7.0). Gaster length 1.3 x width (67-28): Variation. The body length of the para- type males ranges between 2.1 and 2.8 mm. Though all the paratype males were col- lected from approximately the same area, they show considerable variation in color. Most are basically green like the allotype, but one is blue, and several of the others have greater or smaller areas of the head and mesosoma purple. One male has a pet- iole only 1.4 times as long as wide; the pet- iole of the other males ranges between 1.8 and 2.3 as long as wide. The basal cell varies from totally setate, as in the holotype, to two specimens which have only a couple of rows of setae along the anterior margin of the basal cell. Discussion. Unique features of this species are the triangular head (Fig. 1), the broad and deep antennal scrobes, the long malar distance, the bidentate clypeus (Fig. 1), the 117 deep sulcus around the clypeus (Fig. 1), and the dark maxillary palps. Notoglyptus bi- dentatus exhibits a few primitive character states not found in congeners. It has a dis- tinct pronotal collar (Fig. 8) (a character state shared with N. tze/tales and N. virescens), a bidentate clypeus, and dark maxillary palps. The extensively setate wings (Fig. 2) may also be primitive. However, setate wings are common in miscogasterine species living at high elevations or latitudes, so it is difficult to say in this case whether the setate wings of N. bidentatus are primitive or an adap- tation to the mountainous habitats where this species occurs. Etymology. The specific name is from the Latin word bidentatus, meaning two- toothed, and refers to the unique bidentate state of the anterior margin of the clypeus in this species. Biology. Nothing is known of the host(s) of this species. Type material. The holotype, allotype, and 5 male paratypes (all CNC) were collected in Ecuador in the elfin forest at 3800 meters along the Quito-Baeza Road on | March 1979 by W. Mason. Four additional para- types (CNC, INHS, USNM) were collected as follows: Ecuador. Napo (4100 m, Quito- Baeza Road), 24:-II- 1983, 1 4, Paruma (4200 m, Quito-Baeza Road), 14-17-II- 1982, 3 4. Notoglyptus luteicrus, New SPECIES Figs. 3, 10-11 Holotype, female. Description. Color: Head, mesosoma, petiole black, except the following blue: clypeus; pairs of diffuse spots extending anteriorly from lateral ocelli; ver- tex along orbits; lateral parts of pronotum and mesoscutum, frenum, dorsellum, pro- podeum; pleural regions posteriodorsally. Gaster brown, Tl with bluish reflections. Antenna with scape yellow-brown, darker in apical 4: pedicel brown; flagellum dark brown. Mandibles yellow, teeth reddish yel- low. Maxillary palps yellow. Legs yellow, pretarsi dark brown. Wing veins pale brown. 118 Sculpture: Clypeus smooth; remainder of head, mesoscutum (Fig. 10), scutellum del- icately alveolate; frenum coriaceous (Fig. 10); dorsellum smooth; median panels of propodeum alveolate, pattern very weak in center of each panel: petiole granulate; gas- tral terga smooth. Structure: Body length 1.4 mm. Head ovate in anterior view, width 1.3 height (27:21), 2.2 x length (27:12); clypeus set off from face by obscure sulcus, anterior mar- gin straight; genal concavity extending half- way to eye; antennal scrobes shallow; eye height 2.4 length (13.0:5.5), 2.2 malar length (13:6), length 4.9 x temple length (9.5: 2.0); ratio of MOD, OOL, POL, LOL as 1.5: 5:6.5:3. Antenna (Fig. 3) with torulus one outside diameter above LOcL; combined length of pedicel and flagellum 0.93 x head width (25:27); relative lengths of scape, ped- icel, anelli, Fl-6, club as 13.0:4.0:1.0:2.0: 2.5:2.0:2.0:2.0:1.5:8.0; widths of Fl, F6, club as 2.0:3.5:4.0; anelli subequal in size; micropilosity completely covering ventral side of terminal segment. Mesosoma length 1.4 width (28:20); collar undeveloped: scutellum regularly rounded, scutoscutellar sulcus continuous furrow; dorsellum length ‘> frenal length; spiracles round, | x own diameter from anterior margin of propo- deum; nucha acarinate anteriorly. Fore wing length 3.1 width (65:21); ratio of sub- marginal, marginal, postmarginal, stigmal vein lengths as 20:11:9:6; costal cell with single complete row of setae; basal vein with row of 4 setae; basal cell with | seta on left wing; speculum open posteriorly; | row of distinct admarginal setae present. Petiole length 1.1 x width (4.5:4.0); median carina present. Gaster ovate, length 1.4 width (29:21); height 0.86 x width (18:21). Allotype, male. Description. Color: Sim- ilar to holotype except reflections of frenum, dorsellum, propodeum green. Structure: body length 1.4 mm. Antenna with com- bined length of pedicel and flagellum 1.2 x head width (31.0:26.5); relative lengths of scape, pedicel, anelli, F1-6, club as 14:3:1: PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 3:3:3:3:3:3:9; widths of Fl, F6, club as 2:3: 3. Maxillary palps (Fig. 11) with 2 apical segments lamellately expanded. Petiole length 1.2 x width (6:5). Gaster length 1.4 x width (22:16). Variation. Body size in the females ranges from 1.0 to 1.5 mm; in the males, 1.2 to 1.4 mm. The patches of metallic coloration on the head and mesosoma vary among specimens in extent, intensity, and color. The color varies from blue, as in the ho- lotype, to dark green. The specimens from the eastern U.S. and Canada usually have the scape brown only at the apex, but those from Mexico and the western U.S. and Can- ada have the scape mostly brown. The setal patterns of the wings are variable, but the wings generally resemble that in Fig. 4. Discussion. The unique feature of this species is the lamellately expanded male maxillary palps (Fig. 11). Within its species- group, Notoglyptus luteicrus resembles N. nesiotes in lacking any horizontal collar on the pronotum (Fig. 10) and a patch of mi- cropilosity on the apical segment of the fe- male club (Fig. 3). Notoglyptus luteicrus has spots of metallic coloration located in sim- ilar places to those in N. nesiotes but the boundaries of the spots on the head are dif- fuse in N. /uteicrus and sharp in N. nesiotes. Etymology. The specific name comes from the Latin words /uteus, meaning yellow, and crus, meaning leg, and refers to the yellow legs of this species. Biology. Nothing is known of the host(s) of this species, but specimens have been collected from alfalfa, soybeans, Baccharis L. (Compositae), and the crucifers Sisym- brium altissimum L. and Descurainia so- phia (L.) P. B. Webb. Type material. The holotype (INHS) was collected from the railroad siding at the end of Gerty Drive on the South Farms of the University of Illinois, Champaign County, Illinois, on 25 June 1982, by S. L. Heydon. The allotype (INHS) came from the same locality, but was collected 21 August 1981. Eighteen additional paratypes were collect- VOLUME 91, NUMBER | ed as follows (AMNH, CNC, INHS, SEC, USNM): Canada. ALBERTA: Lethbridge, 5-VII-1956, (swept from brome) 2 2; Lost River, Onefour (10 mi. WNW Wild Horse), 3-VI-1956, 1 2. ONTARIO: 13 mi. N Belle- ville, 27-V-1970, 1 °. United States. CAL- IFORNIA: Rancho Santa Fe, 14-1-1959, (alfalfa field) 1 9. COLORADO: Fort Col- lins, V- 1894, (alfalfa) 1 6. FLORIDA: Col- lier Seminole State Park (Collier Co.), 25- 26:V-1978, 1 2; Fort Ogden, 8-IV-1952, 1 2. IDAHO: Hansen, 29: V- 1930, (Sisymbri- um altissimum and Descurainia sophia) | 9. INDIANA: Hovey Lake (Posey Co.), 3- VI-1981, 1 6. ILLINOIS: Dixon Springs Ag- ricultural Research Station (Pope Co.), 29- 31-VII-1980, 1 6. LOUISIANA: Cameron Parish, 4:-VIII-1969, (soybeans) | @. MARYLAND: Patuxent Research Center (Prince Georges Co.), 1- VII: 1982, 12. NEW JERSEY: Ramsey, 31-VII-1918, 1 9. NEW MEXICO: Elmendorf, 21-VII-1936, 1 2. UTAH: Richfield, 18-V-1954, 1 9; Utah Lake (Utah Co.), 1 2. Mexico. VERACRUZ: Jalapa, HI-IV-1965, 1 2. Venezuela. POR- TUGUESA: 10 km N Biscucuy, 9: VI-1981 (sweeping Baccharis), 1 6. Country? San Ra- fael Jicoltepec, 1 2, 3 6. Notoglyptus nesiotes, NEw SPECIES Figs. 4-5, 12 Holotype, female. Description. Color: Head, mesosoma, petiole black except pro- podeum metallic coppery; inner orbit along vertex, area between median and lateral ocelli, pronotum and mesoscutum laterally, posteriodorsal pleural region metallic green. Gaster dark reddish brown with coppery re- flections dorsally. Antenna with scape, anel- li yellow; pedicel, flagellum brown. Man- dibles yellow, teeth reddish yellow. Maxillary palps yellow. Legs yellow, pre- tarsi black. Wing veins yellow, parastigma darker. Sculpture: Body sculpture weak, face co- riaceous, mesoscutum (Fig. 12) weakly al- 119 veolate mesally, petiole granulate, body nearly smooth otherwise. Structure: Body length 1.4 mm. Head ovate in anterior view, width 1.4 height (22:16), 2.2 x length (22:10); clypeus sepa- rated from face by obscure sulcus, anterior margin straight; antennal scrobes shallow; genal concavity weakly developed, extend- ing only '4 of way to eye; eye height 1.2 x length (9.0:7.5), 2.0 x malar length (9.0:4.5), eye length 3.8 x temple length (7.5:2.0); ra- tio of MOD, OOL, POL, LOL as 1.5:5:5.5: 3. Antenna (Fig. 5) with torulus | own outside diameter above LOcL; combined length of flagellum and pedicel 1.0 head width (23:22): relative lengths of scape, ped- icel, anelli, Fl-6, club as 10.5:3.5:1.0:2.0: 2.0:2.0:2.0:2.5:2.0:6.0; widths of Fl, F6, club as 1.5:2.0:2.0; anelli subequal in size; club with patch of micropilosity down ven- tral side of apical segment. Mesosoma (Fig. 12) length 1.7 width (26:15); collar not developed; scutellum uniformly convex, scutoscutellar sulcus with pair of contiguous diverging foveae mesally; dorsellum cari- nalike; spiracles 1.5 x own diameter from anterior margin of propodeum, nucha bor- dered by carina anteriorly. Fore wing (Fig. 4) length 2.7 x width (48:18); submarginal, marginal, postmarginal, stigmal vein lengths as 17.0:11.5:7.0:5.0; costal cell with one complete row of setae plus a few others dis- tally; basal vein setate; speculum open pos- teriorly; distinct row of admarginal setae present. Petiole length 0.88 width (3.5: 4.0); median carina present. Gaster ovate, length 1.4 width (25:18); height 1.0 x width (18:18); T2-7 protruding from be- neath T1. Male unknown. Variation. The specimens in the type se- ries from Isla Mona are all of a rather uni- form size, coloration, and morphology. The number of setae along the basal vein varies from just a couple of setae to a row extend- ing down the length of the basal cell and curving basally down the cubital vein. About half the specimens have a few setae distally 120 in the basal cell. The paratype female from Sapelo Island, Georgia, lacks the metallic patches on the head and is more distinctly sculptured than the series from Isla Mona. Its body sculpturing resembles that given above for N. /uteicrus. It also resembles N. luteicrus in lacking the distinct carina along the anterior margin of the nucha. However, it has long terminal funicular segments of the antenna, a foveolate scutoscutellar sul- cus (Fig. 12), and a cariniform dorsellum, features which are all diagnostic for N. /u- teicrus. Collection of specimens from a wid- er geographic range may help make sense of the morphological divergence between the populations from Puerto Rico and those of the continental U.S. Discussion. Unique features of this species are the weak body sculpture (Fig. 12) and the cariniform dorsellum (Fig. 12). Noto- glyptus nesiotes and N. virescens both have terminal funicular segments of the antenna quadrate or elongate and a foveolate scu- toscutellar sulcus, but N. nesiotes lacks the fovea on the disc of the scutellum charac- teristic of N. virescens. Etymology. The species name is derived from the Greek word nesiotes, meaning in- sular, and refers to the island distribution of this species. Biology. The host(s) of this species are unknown. The paratype female from Sapelo Island, Georgia, was collected on Spartina Schreber (Gramineae). Type material. Holotype (USNM) and 8 paratype females (CNC, USNM) were col- lected on Isla Mona, Puerto Rico, in August 1944, by H. A. Beatty. One additional para- type female was collected as follows: United States. GEORGIA: Sapelo Island (Mcin- tosh Co.), 10-I1X-1963 (on Spartina). Notoglyptus tzeltales, New SPECIES Figs. 6-7 Holotype, female. Description. Color: Front of head dull dark green; vertex, dor- sum of mesosoma blue-green; propodeum, PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON pleural regions, coxae green; petiole bluish black; gaster dark reddish brown with strong bluish reflections. Antenna with scape blue- green; remainder brown, pedicel with weak metallic reflections. Mandibles brownish yellow; teeth pale brownish red. Maxillary palps cream-colored. Legs with most of femora brown with weak metallic reflec- tions; tibiae brownish yellow, slightly dark- er mesally; basal tarsal segment brownish yellow, rest of tarsi darkening distally tll pretarsus black. Wing veins reddish brown. Sculpture: Clypeus granulate; head, mesosoma, scutellum regularly alveolate; frenum, dorsellum, propodeum smooth; petiole finely strigulate dorsally; gaster smooth. Structure: Body length 1.6 mm. Head (Fig. 7) broadly ovate in anterior view, width 1.3x height (32:24), 2.1 length (32:15); clypeus separated from face by distinct sul- cus, anterior margin produced but truncate; antennal scrobes shallow; genal concavity extending 2 way to eye; eye height 1.3 x length (14:11), 1.8 malar length (14:8), length 3.7 temple length (11:3); ratio of MOD, OOL, POL, LOL as 2.5:6:6:3. An- tenna with torulus located just below LOcL; combined length of pedicel and flagellum 0.72 head width (23:32); relative lengths of scape, pedicel, anelli, Fl—-6, club as 14.0: 4.5:1.5:2.0:2.0:2.0:2.0:2.0:2.0:5.0; relative widths of Fl, F6, club as 2.0:3.0:3.5, all funicular segments except Fl transverse; anelli subequal in size; club lacking ventral patch of micropilosity. Mesosoma length 1.4 width (34:24); collar developed, an- terior edge rounded; scutellum regularly convex, scutoscutellar sulcus continuous furrow; dorsellum length about ' frenal length; spiracles ovate, <1 x own diameter from anterior margin of propodeum; carina bordering nucha anteriorly. Fore wing (Fig. 6) length 2.2 width (71:33); ratio of sub- marginal, marginal, postmarginal, stigmal vein lengths as 28:18:8.5:6; costal cell with one complete row of setae and couple of others distally; basal cell and vein bare; VOLUME 91, NUMBER | speculum open posteriorly; no distinct ad- marginal setae. Petiole length 1.6 width (8:5); lacking median carina. Gaster ovate, length 1.7 x width (30:18); deep, height | x maximum width (18:18). Allotype, male. Color: Pattern similar to holotype but antenna beyond basal half of pedicel nonmetallic, brownish yellow; club slightly darker on outer side. Body length 1.5 mm (critical point-dried). Antenna with combined length of pedicel and flagellum 0.73 x head width (20.5:28.0); relative lengths of scape, pedicel, anelli, F1-6, club as 12:3:1:2:2:2:2:2:2:5: relative widths of F1, F6, club as 2:3:3; funicular segments cylin- drical. Maxillary palps slender. Petiole length 2.0 x width. Gaster length 1.6 x width (31:20). Variation. The color of N. tzeltales varies greatly over its geographic range. The fe- male from Huetamo de Nunez is almost black except for the dorsum of the meso- soma which is dark green, and the antennal flagellum is yellow beyond the basal half of the pedicel. The male has the face violet, the remainder of the head and mesosoma green, the pedicel to F6 yellow, and the club brown. Females of the series from Vene- zuela have the body greenish black and the antennal flagellum brown. In the males, one paratype has antennal coloration like that of the allotype, while the other has the fla- gellum brown on the outer face and pale on the inner face. The males from Venezuela also have the tip of the middle tibia dark- ened. The Venezuela series was collected into alcohol and then critical point-dried. This process can sometimes alter the colors of specimens. Whether the color differences in N. tzeltales are caused by environmental influences or genetic differences needs study. Discussion. Unique features of N. tzel- tales are the bare basal cell and vein (Fig. 6). This species most closely resembles N. bidentatus in having metallic legs and the anterior margin of the clypeus produced; although in N. tze/tales (Fig. 7), the clypeus does not have the median emargination 121 present in N. bidentatus (Fig. 1). The two species are reliably distinguished by the characters given in the key. Etymology. The specific name of this species comes from the Tzeltales Indians who live in the area of Mexico around San Cristobal de las Casas. Biology. The host(s) of this species is un- known. Type material. The holotype (CNC) is from San Cristobal de las Casas, Mexico, and was collected 1-3 June 1969. The al- lotype male and 8 female and 2 male para- types (USNM) were collected in Venezuela, 10 km north of Biscucuy, on 9 June 1981, by E. E. Grissell, while sweeping Baccharis (Compositae). Ten female and | male para- types were collected as follows (CNC, USNM): Costa Rica. Cartago (1500 m), VIII-1980, 1 9; Heredia, 10-VIII-1975, 2 2. Mexico. CHIAPAS: San Cristobal de las Casas, 27-1V-1969, 2 2, 12-VI-1969, 1 9. MICHOACAN: Huetamo de Nunez, 7: III: 1972, 1 2, 1 6. QUERETARO: 10 mi. E San Juan Del Rio, 30-VII-1954, 2 ¢. Guate- mala. GUATEMALA: Guatemala City, IX- LOS 9S lS: Notoglyptus virescens Masi Notoglyptus virescens Masi, 1917: 181-183. Gahan and Fagan, 1923: 98. Boucek, 1961: 67. Graham, 1969: 140. Boucek, 1976: 15 (synonymy). Boucek, Subba Rao, and Farooqi, 1978: 448. Farooqi and Subba Rao, 1986: 295. Holotype, 8, BMNH Hym. 5.873 (examined). Notoglyptus niger Masi, 1917: 181. Erdés, 1948: 38. Boucek, 1961: 67. Peck et al., 1964: 36. Boucek, 1976: 15 (synonymy). Farooqi and Menon, 1973: 55. Boucek, 1977: 46. Dzhanokmen, 1978: 80. Bou- cek et al., 1978: 448. The following description of N. virescens is based on a pair of specimens on loan to me from the BMNH. Female. Description. Color: Head, meso- soma, petiole dark green; gaster brown. An- tenna with scape yellow, slightly more brown 122 apically; remainder brown, pedicel pale ventrally. Mandible yellow, teeth reddish brown. Maxillary palps yellow. Legs yellow. Wing veins yellowish brown. Sculpture: Clypeus smooth, remainder of head delicately alveolate; mesoscutum roughly alveolate medially, becoming co- riaceous laterally; scutellum coriaceous; frenum, dorsellum smooth; median panels of propodeum alveolate, sculpturing weak in center of panels; petiole alveolate; gaster smooth. Structure: Body length 1.3 mm. Head ovate in anterior view, width 1.3 x height (23.5:18.0), 2.0 length (23.5:11.5); clyp- eus separated from face by obscure sulcus, anterior margin nearly straight, slightly re- flexed; antennal scrobes shallow; genal con- cavity extending ' way to eye; eye height 1.4 length (11:8), 2.8 malar length (11: 4), length 2.7 x temple length (8:3); ratio of MOD, OOL, POL, LOL as 2.0:4.5:6.0:3.0. Antenna with torulus 1x own diameter above LOcL; combined length of pedicel and flagellum 1.1 x head width (25.0:23.5): relative lengths of scape, pedicel, anelli, Fl— 6, sclubvas -11:0:3.0; 10:210:2'522-5:2:5:225: 2.5:7.0; relative widths of Fl, F6, club as 1.75:2.0:2.0; anelli subequal in size; micro- pilosity in line down ventral side of terminal two segments of club. Mesosoma length 1.5 width (28:19); horizontal collar de- veloped, anterior edge carinate; scutellum with distinct discal fovea, scutoscutellar sul- cus foveolate medially; dorsellum length '/ frenal length; propodeum with spiracles cir- cular, 1.5 own diameter from anterior margin of propodeum; nucha not bordered by carina anteriorly. Fore wing length 2.5 x width (57:23); ratio of submarginal, mar- ginal, postmarginal, stigmal vein lengths as 20:12:9:5; costal cell with | complete setal row basally and a couple others apically; basal cell with apical quarter setate; spe- culum closed posteriorly; distinct admar- ginal setae present. Petiole length 1.2 x width (5:4); with weak median carina. Gaster length 1.3 width (21:16); height 1.1 x width (17:16). PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Male. Color: Similar to female except face with face bluish, dorsum of mesosoma greenish. Structure: Body length 1.0 mm. Antenna with combined length of pedicel and flagellum 1.5 x head width (31:21); rel- ative lengths of scape, pedicel, anelli, F1—6, club as 10.0:2.5:1.0:3.0:3.0:3.0:3.0:3.5:3.0: 8.0; relative widths of Fl, F6, club as 2:2: 2. Maxillary palps slender. Gaster length 1.5 width (27:18). Variation. The male from Ontario is a little larger (body length 1.3 mm), its scape is mostly brown, its legs are more amber, and its petiole lacks the median carina; but agrees otherwise with the above description. Discussion. Unique features possessed by this species are the fovea on the disc of the scutellum and the club having a line of mi- cropilosity down the ventral side. This species 1s intermediate in morphology with- in the genus since it has the developed hor- izontal pronotal collar like N. bidentatus and N. tzeltales, but has the straight clypeal mar- gin, the toruli located above the LOcL, the antennal club with micropilosity, yellow legs, admarginal setae, and a median carina on the petiole like N. /uteicrus and N. nesiotes. Distribution. This species is found from southern Europe to India, Japan, and south- ern Africa (Boucek 1976, 1977). There is a male in the Canadian National Collection from Mount Pakenham, Ontario. Material Examined (BMNH, CNC). In- dia. Delhi, IMRI, 1-IV-1977, (on grass) 1 2. Zimbabwe. Salisbury, | ¢. Japan. KYO- TO: Kibune, 6-VIII-1980, 1 ¢@. Canada. ONTARIO: Mount Pakenham, 11-VI- 1969, 1 6. ACKNOWLEDGMENTS I wish to thank the following persons for the loan of material: M. Favreau, American Museum of Natural History (AMNH), New York, NY; Dr. J. S. Noyes, British Museum of Natural History (BMNH), London; Dr. G. P. Gibson, Canadian National Collec- tion (CNC), Ottawa, ON; Dr. W. E. La- Berge, Illinois Natural History Survey, Champaign, IL; Dr. G. W. Byers, Snow En- VOLUME 91, NUMBER 1 tomological Collection (SEC), Lawrence, KS; and Dr. E. E. Grissell, United States Na- tional Museum (USNM), Washington, D.C. I would like to express my appreciation to Shirely McClellan and Drs. W. E. LaBerge and George Godfrey (INHS), and two anon- ymous readers for reviewing this paper. I would like to thank Drs. W. E. LaBerge and E. E. Grissell for help and encouragement, J. Sherrod (INHS) for assistance with the illustrations, the staff at the Center of Elec- tron Microscopy at the University of Illinois at Urbana-Champaign for use of the facil- ities, and Molly Scott (INHS) for preparing the plates. LITERATURE CITED Boucek, Z. 1961. Beitrage zur Kenntnis der Ptero- maliden-fauna von Mitteleuropa, mit Beschrei- bungen neuer arten und Gattungen (Hymenop- tera). Acta Entomol. Mus. Natn. Pragae 34: 55- 95. 1976. African Pteromalidae (Hymenoptera): New taxa, synonymies and combinations. J. Ento- mol. Soc. South. Afr. 39: 9-31. 1977. A faunistic review of the Yugoslavian Chalcidoidea (Parasitic Hymenoptera). Acta Entomol. Jugosl. 13 (Suppl.): 1-145. Boucek, Z., B. R. Subba Rao, and S. I. Farooqi. 1978. A preliminary review of Pteromalidae (Hymenop- tera) of India and adjacent countries. Orient. In- sects 12: 433-468. Dzhanokmen, K. A. 1978. [Identification of the 1n- sects of the European part of the USSR. Vol. 3. Hymenoptera. Second part. Pteromalidae.] Opre- deliteli Faune USSR, No. 120: 57-228. Erdos, J. 1948. Additamenta ad cognitionem faunae Chalcidoidarum (Hymenoptera) in alveo Carpa- thorum. IV. 11. Pteromalidae. Frag. Faun. Hung. 11: 36-51. Farooqi, S. l.and M.G. R. Menon. 1973. New com- bination proposed for a pteromalid species Oxy- sychus sauteri (Masi) comb. n. (= Dinarmus sau- teri Masi). Entomol. Newslet. 3: 55. Farooqi, S. I. and B. R. Subba Rao. 1985. Family Pteromalidae, pp. 254-263. Jn Subba Rao, B. R. and Hayat, M., eds., The Chalcidoidea (Insecta: Hymenoptera) of India and the adjacent countries. Part I. Reviews of families and keys to families and genera. Orient. Insects 19: 163-310. . 1986. Family Pteromalidae, pp. 279-306. Jn Subba Rao, B. R. and Hayat, M., eds., The Chal- cidoidea (Insecta: Hymenoptera) of India and the adjacent countries. Part II. A catalogue of Chal- cidoidea of India and the adjacent countries. Ori- ent. Insects 20: 1-430. Gahan, A. B. and M. Fagan. 1923. The type species of the genera of Chalcidoidea or chalcid-flies. U.S. Natn. Mus. Bull. 124: 1-173. Graham, M. W.R. de V. 1969. The Pteromalidae of northwestern Europe (Hymenoptera, Chalcidoi- dea). Bull. Brit. Mus. (Nat. Hist.) Entomol. Suppl. 16: 1-908. Heydon, S. L. 1988. The Sphegigasterini: A cladistic analysis and generic classification with reviews of selected genera (Hymenoptera: Pteromalidae). Ph.D. Thesis, Univ. of Illinois at Urbana—Cham- paign, Urbana. 167 pp. Masi, L. 1917. Chalcididae of the Seychelles Islands. Novit. Zool. 24: 121-230. Peck, O., Z. Boucek, and A. Hoffer. 1964. Keys to the Chalcidoidea of Czechoslovakia. Mem. Ento- mol. Soc. Canada 34: 1-120. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, p. 124 Book REVIEW Bees of the New Genus Ctenoceratina i” Africa South of the Sahara (Hymenoptera: Apoidea). By Howell V. Daly. University of California Publications in Entomology 108: 1-69. February 1988. University of California Press. Once again, Howell Daly has prepared an excellent taxonomic revision. This mono- graph of the new genus Ctenoceratina Daly and Moure (tribe Ceratinini) encompasses 13 species of small carpenter bees indige- nous to subsaharan Africa. Included are a new subgenus, Simioceratina Daly and Moure, and three new species, Ctenocera- tina malindiae Daly, C. paulyi Daly, and C. rossi Daly. A key, synonymies, descriptions, distributions (precise, but unfortunately without locality maps), and some biological data including parasites are provided. The publication’s format provides a clear, easily followed explanation of how and why various taxonomic decisions were made. Thus. the text will prove useful to both ex- perienced and novice hymenopterists. The high quality illustrations include SEM pho- tographs (Figs. 1-8) and line drawings (Figs. 9-50). One minor complaint is that figure numbering (particularly of facing plates pp. 64-65) is not consistently from left to right or from top to bottom. The left to right convention seems most easily followed. Dr. Daly’s thorough approach to this study is evidenced by his efforts to assess vari- ability. Through rearing, careful observa- tion of live material, and collection at select localities, he documented striking allome- tric differences among nest mates, and among, individuals within local popula- tions. (Such differences undoubtedly led earlier taxonomists to distinguish some 55 species.) When large series were available, an effort was also made to assess variation between populations. I appreciated Dr. Da- ly’s treatment of species displaying complex geographic variation. Rather than designat- ing formal subspecies, informal “forms” were discussed for these polytypic species. I also appreciated the clarity provided by his “similar species” discussions contained within the species descriptions. A critical test of the key was never made since our African material was still on loan to Dr. Daly. However, the key was easily read and the illustrations made interpreta- tion appear straight forward. One point of potential confusion that should be noted is his use of the term “scale.” The genus name, Ctenoceratina, was selected because of the comb-like rows of enlarged setae (‘‘scales’’) found on the apical edges of terga and sterna 2-4 and frequently 5. While not incorrect, this usage might momentarily baffle the un- suspecting bee worker who dives into the key without previously reading the intro- duction. Reviewing this work whetted my appetite for future investigations of Ctenoceratina. We are indebted to Howell for providing yet another valuable contribution. Beth B. Norden, Department of Ento- mology, Smithsonian Institution, National Museum of Natural History—105, Wash- ington, D.C. 20560. — PROC. ENTOMOL. SOC. WASH. 91(1), 1989, p. 125 Book REVIEW Control of Arthropod Pests of Livestock: A Review of Technology. By Roger O. Drummond, John E. George, and Sidney E. Kunz. 1988. CRC Press, Inc., Boca Ra- ton. Florida, 245 pp. Price US $125.00. This book reviews a large number of pa- pers published in the U.S. and Canada con- cerning the materials and methods which have been used to control arthropod pests of livestock. It is principally a summary of the history of chemical treatments. It is not a manual of pest control techniques, nor is it a treatise on the evolution and/or failures of chemical insecticides or acaracides which have been used against pests of livestock. The book covers the entire spectrum of ar- thropod pests of livestock. The material re- viewed on the biology and significance of each pest ranges from one to two pages and the review of control technologies used against each pest ranges from three to six- teen pages. The first 15 chapters are arranged by com- modity, i.e. cattle, horses, sheep and goats, swine and poultry, and each chapter is sub- divided into sections on individual pests. The biology and importance of each pest is discussed briefly, followed by a two-to-sev- en page review of control efforts, and each chapter is summed up by an Overview and Current Technology section. The overview sections are well done and provide a concise and valuable statement of current control methods for each pest group. Chapters 16 through 18 review the con- trol efforts which have been used against pests that breed in poultry manure, against ticks and chiggers and against flies in live- stock manure and around structures. These chapters are well arranged but the five-to- nine pages devoted to each subject neces- sarily limit the number of papers and tech- nologies reviewed. Chapter 19 provides a 4-page review of pest resistance to insecticides and is prob- ably the most disappointing chapter in the book, particularly in view of the emphasis that the authors place on the historical usage of chemicals as insecticides and acaracides. Chapter 20 provides a nice review of the major sterile insect release programs (6 pages) and chapter 21 reviews the status of biocontrol technology (5 pages). Chapter 22 provides an intriguing pre- view of the possibilities of developing the resistance of livestock to arthropod pests (5 pages). There is a 4-page appendix which lists the trade and chemical names of many of the insecticides and acaracides mentioned in the book, but no addresses are listed for the manufacturers of the chemicals. Although a large number of references are listed for each pest, the list is, of necessity, incomplete and even some bibliographies (1.e. West, West and Peters) are not men- tioned. There are no lists of the published bibliographies or reviews of literature per- taining to an individual pest or commodity except for the sterile insect technique for screwworms. There are no charts, chronological lists of control methods (except for cattle grubs), or analyses of the histories of the development of resistance of pests to chemicals. In summary, this is an ambitious book which meets the stated objectives of the au- thors to provide a history of control tech- nologies. It should be of interest to anyone who is interested in the history of livestock pest control methods and should be valu- able as a background source to anyone who is just entering the livestock pest control field. In my opinion, however, it is grossly overpriced at U.S. $125.00. Lawrence G. Pickens, Livestock Insects Laboratory, Livestock and Poultry Sciences Institute, USDA-ARS, Beltsville, Maryland 20705. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 126-127 Book REVIEW Ecological Methods with Particular Ref- erence to the Study of Insect Populations. T. R. E. Southwood. 1978. Second re- vised edition. Chapman and Hall, Lon- don, New York; xxiv + 524 pp. $32.50 Softcover. This second edition, now available in pa- perback, is a thoroughly updated version of the first edition from 1966. Some chapters have been completely or largely rewritten and substantial new information has been added. Most impressively, and vividly il- lustrating the rapid growth of available eco- logical information, close to a thousand new references have been included. The book presents a synthesis of ““methods that are peculiar to the ecologist,” and especially those aimed at the study of insect popula- tions. The usage, however, is much broader as techniques generally applied to the study of earthworms, mollusks, nematodes, planktonic organisms, and several non-in- sect arthropods have been included. The ap- proach is mainly syn-ecological and only little attention has been paid to the habitat although the importance of spatial and tem- poral heterogeneity as well as predictability is stressed. The first chapter is introductory and discusses classification of population estimates, errors and confidence, and chap- ter 2 deals with the sampling program and the description of population dispersion. Chapters 3-8 are mainly aimed at the ques- tion of how to obtain an absolute population estimate and contain marking techniques, capture-recapture methods, and sampling procedures. Chapters 9-11 deal with de- mographical aspects. In chapter 12 systems analysis and modelling are briefly discussed, and the realm of community ecology is reached in chapter 13 which deals with di- versity, species packing and habitat. The last chapter discusses productivity and energy budgets. ‘Ecological Methods’ spans the gap be- tween suction traps made from homestyle vacuum cleaners and, for example, the “‘lin- ear dioristical [systems] model with alge- braically defined transfer functions.” To at- tain this, a lot of information has to be presented in a fairly condensed state, but Southwood manages to do this in a very coherent and readable style. Sampling techniques are discussed thor- oughly and their strengths and weaknesses with regard to their ease, the samples they produce and the financial resources they de- mand are stressed. Much attention is paid to possible sources of errors, whether stem- ming from the equipment proper or from the heterogeneity and unpredictability of living organisms or their habitats. Basic mathematical and statistical theory is not discussed at length but is still presented in a very clear form and with ample references for those interested in its derivation. It is, perhaps, somewhat ironical that a book devoted to the study of populations does not provide an explicit discussion on how to define a population as a spatiotem- poral entity. Largely left unanswered are questions like: How are geographical limits defined?, which constituing features emerge?, and, what are the temporal di- mensions? A decade has passed since the 1978 edi- tion, and methods and theory in insect ecol- ogy have kept evolving. Emerging fields as tropical rainforest canopy ecology have de- veloped largely new techniques, and much new information is available on patterns of population growth, e.g. how to predict in- sect pest outbreaks using integrated trap- ping with various combinations of phero- mones, kairomones, colors, and host-mimic lures. The need for precision and critical cost/benefit considerations, however, re- mains important, and it is still imperative to be able to evaluate the limitations and VOLUME 91, NUMBER | possible biases of the techniques involved in any study. Therefore, the book still stands out as a must for anyone teaching the prac- tical aspects of insect (or animal) ecology, and as an extremely valuable manual for all who need to investigate animal populations, 127 whether as a research ecologist, a conser- vationist or an applied entomologist. Thomas Pape, Zoological Museum, Universitetsparken 15, DK-2100 Copenha- gen, Denmark. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, p. 127 NOTE Rediscovery of the Ant Gnamptogenys hartmani (Hymenoptera: Formicidae) in Eastern Texas Wheeler (1915. Bull. Amer. Mus. Nat. Hist. 34: 390) described Gnamptogenys hartmani from a single worker collected in Huntsville, Walker Co., eastern Texas. This record seemed anomalous for a genus oth- erwise confined to tropical environments (Brown. 1961. Psyche 68: 69). Brown (op cit) suggested that the Texas record was either a locality error or an adventive spec- imen taken from bananas (it also occurs in Honduras). There are no ecological data as- sociated with the specimen. Brown (op cit) concluded that the presence of Gnamptoge- nys in Texas remained to be convincingly demonstrated. We have collected a second worker spec- imen from Texas, Brazos Co., 10 km N of Kurten, about 60 km west of the type lo- cality. The specimen was collected in a pit- fall trap in an open grassy area located 30 meters from dense post oak (Quercus spp.) woods on 5 Aug. 1987 (trap was in field for 24 h). Although we did extensive pitfall trapping, baiting, general collecting and berlese funnel sampling of the litter in the study area, we have not collected any ad- ditional specimens. We have not collected this species in several other similar areas in eastern Texas, using similar methods. This specimen was collected in a disturbed area (mowed and occasionally grazed by cattle), but the ant was definitely not associated with any introduced products. These data, in ad- dition to the collection of this species in Louisiana (Echols. 1964. Ann. Entomol. Soc. Amer. 57: 137) clearly demonstrate that this species is a rarely collected member of the fauna of the United States. This species preys on the brood of the ant Trachymyrmex sep- tentrionalis (McCook) (Echols op cit), which is common in the area. The ant, Afta texana (Buckley), 1s also common and may serve as a prey species. It is gratifying that this interesting ele- ment of our native ant fauna has withstood the onslaught of the spread of the imported fire ant (Solenopsis invicta Buren), which eliminates many native ant species. The ant was collected in an area of sandy soil where the density of the imported fire ant is rel- atively low (115 mounds/ha). The specimen is in the collection of WPM, field number 9676. Alex Mintzer suggested the study area, O. L. Tate gave us permission to use the area, Mark Strain, Shelley Stonecipher and Cecil Pinder assisted with the field work. The re- search was supported by the Texas Depart- ment of Agriculture. This report is ap- proved as Number TA 23642 of the Texas Agricultural Experiment Station. Wiliam P. MacKay and S. Bradleigh Vinson, Department of Entomology, Texas A&M University, College Station, Texas 77843. PROC. ENTOMOL. SOC. WASH. 91(1), 1989, pp. 128-132 PROCEEDINGS of the ENTOMOLOGICAL SOCIETY of WASHINGTON Volume 90 OFFICERS FOR THE YEAR 1988 President F. Eugene Wood President-Elect F. Christian Thompson Recording Secretary Richard G. Robbins Corresponding Secretary John M. Kingsolver Treasurer Norman E. Woodley Editor Hiram G. Larew Custodian Anne M. Wieber Program Chairman Warren E. Steiner Membership Chairman Geoffrey B. White Delegate to The Washington Academy of Sciences Manya B. Stoetzel Published by the Society WASHINGTON, D.C. 1988 TABLE OF CONTENTS, VOLUME 90 ARTICLES BAKER, G. T.—See CHAN, W. P. BAO, N.—See ROBINSON, W. H. BARNES, J. K.—Notes on the biology and immature stages of Poecilographa decora (Loew) (Diptera: Sclomyzidae) ... ie eas SR, HE aoe TEP eee te 474 BENNETT, F. D.—See PALMER, W. A. BIN, F.—See JOHNSON, N. F. BOLDT, P. E., W. WOODS, and T. O. ROBBINS—Phytophagous insect fauna of Baccharis sarothroides Gray (Asteraceae) in Arizona and New Mexico ..............-..00+++e000- 207 BURGER, J. F.—A new genus and two new species of Pangoniini (Diptera: Tabanidae) of zoogeographic interest from Sabah, Malaysia ............... ooh abi ratictey Necteratay Sher sabe eR CARLSON, R. W.—See DREA, J. J. CARPENTER, J. M. and M. C. DAY—Nomenclatural notes on Polistinae (Hymenoptera: VOLUME 91, NUMBER 1 CAVE, R. D.—See NORTON, R. A. CHAN, W. P., G. T. BAKER, and M. M. ELLSBURY —Sensilla on the larvae of four Hypera Species:(Coleopteras Curculionidae)! Ma. as.2s as see sa aaiandendltea) pee e eee eene enous CHEMSAK, J. A. and C. FELLER—New species of Gamniivatac fom Twin Cays, Belize (Coleoptera) tee create Sete ice re tte ee ne Ne nnn ccs este Gt ee Ores CHESSER, R. K.—See VAN DEN BUSSCHE, R. A. CLEMENT, S. L. and T. MIMMOCCHI— Occurrence of selected flower head insects of Cen- taurea solstitialisin Italyand Greece... 02.0 en ees cae eea seas wees snnecass CLEMENT, S. L., T. MIMMOCCHI, R. SOBHIAN, and P. DUNN—Host Seahah) ofadule Eustenopus hirtus (Waltl) (Coleoptera: Curculionidae), a potential biological control agent of yellow starthistle, Centaurea solstitialis L. (Asteraceae, Cardueae) ............. DAVIS, D. R. and E. G. MILSTREY — Description and biology of Acrolophus phalcen ives idoptera: Tineidae), a new moth commensal from gopher tortoise burrows in Florida . . DAY, M. C.—See CARPENTER, J. M. DIATLOFF, G. and W. A. PALMER—The host specificity and biology of Aristotelia ivae Busch (Gelechiidae) and Lorita baccharivora Pogue (Tortricidae), two microlepidoptera se- lected as biological control agents for Baccharis halimfolia (Asteraceae) in Australia ...... DREA, J. J. and R. W. CARLSON — Establishment of Cybocephalus sp. (Coleoptera: Nitidu- lidae) from Korea on Unaspis euonymi (Homoptera: Diaspididae) in the eastern United SS TALES Me ehyn saiegs ys ere ca oy ase een cys ehastoreentearaxeGter oc is tieer nich orca ia Nes atlas eraaiara te eee Mapa Stee DROOZ, A. T. and H. H. NEUNZIG—Notes on the biology of two Phycitines (Lepidoptera: Pyralidae) associated with Toumeyella pini (Homoptera: Coccidae) on pine .............. DUNN, P.—See CLEMENT, S. L. ELLSBURY, M. M.—See CHAN, W. P. EMERSON, K. C. and R. D. PRICE—A new species of Haematomyzus (Mallophaga: Hae- matomyzidae) off the bushpig, Potamochoerus porcus, from Ethiopia, with comments on lice FOUN GON GpIGS ears ee teleaie ces oer ite cae atone stern tk RE Sereno ete tas ain FELLER, C.—See CHEMSAK, J. A. GITTINS, A. R.—See TRIPLETT, D. C. GOEDEN, R. D.—Gall formation by the capitulum-infesting fruit fly, Tephritis stigmatica (Di pterassle pi mit dae) peer ceere anys 2 eee ee Me ae Se ee See ere ke eee GOTT, K. M.—See NEAL, J. W., Jr. GRISSELL, E. E.—See HEYDON, S. L. HALSTEAD, J. A.—Belaspidia longicauda, new species, the first Nearctic Belaspidia (Hy- menoptera: Chalcididae) ............. bens Se SORE ae HANNA, R. B.—See LANE, M. A. HANSON, P. E. and J. C. MILLER—Notes on the biology of Caenocephus aldrichi Bradley (ivimenopteras Gephidae)i mes: sacrsce sian = ce ateniere ite cic caectace eins cuits Gas qgrmicuets HANSSON, C.—A revision of the genus Me. ot oeRAIS anid a review of det genus ‘Grain (Eivmenopteras lop hidae)) cen emis estes ici cic nreiaeiier n= one ree crete oa HARMON, J. D. and M. H. ROSS— Effects of aalathion and diazinon exposure on feniale German cockroaches (Dictyoptera: Blattellidae) and their oothecae ............ HEYDON, S. L.—A review of the Nearctic species of Cryptoprymna Forster, with the decane tion of a new genus, Polstonia (Hymenoptera: Pteromalidae) ........................... HEYDON, S. L. and E. E. GRISSELL—A review of Nearctic Mes rismus Walker and Toxeuma Walker (Hymenoptera: Chalcidoidea: Pteromalidae) .......... Cer igioat SuoaeActacr aero eee HOEBEKE, E. R.—See WHEELER, A. G., Jr. HUANG, Y. M.—Aedes (Stegomyia) josiahae, a new species of the simpsoni subgroup (Diptera: Culicidae) . RA Tere JOHNSON, N. F. ec. of Australian ‘Telenominae (Hiymenopter: Scelionidae) of A. Pp. Dodd and A. A. Girault Ne a Rie eC ee eee rere JOHNSON, N. F. and F. BIN— Tele. OTRAS egies EVER BTETe: Sedionidae) associnied with the eggs of Zygaenidae (Lepidoptera) 269 179 47 501 164 458 307 44 338 7 87 130 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON JOHNSON, W. T.—See SPECKER, D. R. KONDRATIEFF, B. C.—See STARK, B. P. KURCZEWSKI, E. J.—See KURCZEWSKI, F. E. KURCZEWSKI, F. E., E. J. KURCZEWSKI, and M. G. SPOFFORD—Nesting behavior of Aporinellus wheeleri Bequaert and A. taeniolatus (Dalla Torre) (Hymenoptera: Pompilidae) 294 KURCZEWSKI, F. E.—See LANE, M. A. LANE, M. A., F. E. KURCZEWSKI, and R. B. HANNA—Antennal sensilla and setae of Evagetes parvus: (Hymenoptera; Pompilidae)) trae a ccie « cassieisteisiec cle acess edwin 428 MacDONALD, J. F.—New synonyms pertaining to Chelifera and generic key for North Amer- ican Hemerodromiunae!(DipterasEmpididae)i- mere. ss: > 2 ee ceeeie eee celia haters 98 MALDONADO-CAPRILES, J.—See SANTIAGO-BLAY, J. A. MARSHALL, S. A. and D. J.S. MONTAGNES— Ceroptera longicauda, a second North Amer- ican species in the kleptoparasitic genus Ceroptera Macquart (Diptera: Sphaeroceridae) ... 189 MATHIS, W. N.—First record of the genus Procanace Hendel from North America, with the descriptioniofianew:species:(DipteraxCanacidae) wath. certerct-clatsm cee ose «te ceils cies clea 329 MATHIS, W. N.—First record of the shore-fly genus Placopsidella Kertész from North America (DipterazEphydridae)) «<:.5 c 5. % 45 ccls, wire om teat oe toesteteretete eed ercekel ries cic tanche aie Clea nee eeer a epee 126 PICKENS, L. G.— Control of Arthropod Pests of Livestock: A Review of Technology ......... 125 TABLE OF GONTENTIS,, VOLUMES SON Ness ete ctctsta vi oteiathebtnt ito kele sircte lel can atekcnrel < h ainiat etal tata 128 IL. 91 APRIL 1989 NO. 2 (ISSN 0013-8797) PROCEEDINGS of the ENTOMOLOGICAL SOCIETY of WASHINGTON PUBLISHED UAR--ER OaTHSONE, CONTENTS | ‘#2 Qeseee APPEL, A. G. and R. C. SPONSLER—Water and temperature glatiog $ofshe-s iri tive lophagous cockroach Cryptocercus punctulatus Scudder (Dictyopteraefryptocercidaey ... 153 ASQUITH, A. and D. ADAMSKI—Description of the predaceous larva of Pseudogaurax sig- MAES eGew (I pteras GHIOLOpldae) pam. se daie,2 cin ae te a eis hates yokes clays aes S 2) le, ees le oho ears 185 DIETRICH, C. H.—Surface sculpturing of the abdominal integument of Megobracidae and other AMICHEN OM MUNCH A ELOMODIEKA) I azrrir a arte gcpeyecctare) sfere stn spots ox eee ers MP ers peel oe Gaahia 143 GAGNE, R. J. and P. E. BOLDT—A new species of Neolasioptera (Diptera: Cecidomyiidae) from Baccharis (Asteraceae) in southern United States and the Dominican Republic .... 169 GOEDEN, R. D.—Host plants of Neaspilota in California (Diptera: Tephritidae) ........... 164 HENRY, C. S.—The unique purring song of Chrysoperla comanche (Banks), a green lacewing of the rufilabris species group (Neuroptera: Chrysopidae) ......................2220... 133 KELLEY, R. W.—New species of micro-caddisflies (Trichoptera: Hydroptilidae) from New ealedOniass Vata CURanGUni ine op eee een ees neh Me AT anole aaah lero hae desea 190 KNISLEY, C. B., D. L. REEVES, and G. T. STEPHENS—Behavior and development of the wasp Pterombrus rufiventris hyalinatus Krombein (Hymenoptera: Tiphiidae), a parasite of larvaletiper beeties (Coleoptera: Cicindelidae)) 0.0 estore wet tne wan aicvers fies) sles Soares ct ayers 179 LASALLE, J.— Notes on the genus Zagrammosoma (Hymenoptera: Eulophidae) with description DAMIEN DECICS Ae met: Parity si Malia take meio Sauer gi elcnis “Saft lavolefergrot a Sahat ots sual oe eke yh Sas 230 MACKAY, W. P. and S. B. VINSON—Two new ants of the genus Solenopsis (Diplorhoptrum) fromeastern Texas (Hymenoptera: Formicidae) .. 2... 0.5.0 cde cca e ect e ese s st enenee 175 MILLER, T. D.—First Nearctic record of the genus Nordlanderia (Hymenoptera: Eucoilidae), MILA eSCMOLONS OLEWO NEW SPECIES! ase... Rut b-cfe ire Sae) tues re ZONpe Zils G22 925uee 24S on 260 noe TEMPERATURE, in °C Fig. 3. Effect of temperature on volley duration for all calls of all males and females of Chrysoperla co- manche. The filled circle at each temperature is the mean of at least 15 volleys delivered at that temper- ature, and the error bars bracket one standard devia- uon of the mean. The linear regression equation, its squared variance, and its sample base are entered above the line plot. creased, leveled off, and then decreased dur- ing the course of each volley. The mean of this modulation frequency was about 21 Hz, but started at approximately 14 Hz and stopped at 17.5 Hz (Table 3). It was clear from watching receptive lacewings at low temperatures that this modulating oscilla- tion corresponded to gross vertical move- ments of the insect’s abdomen, while the higher, primary frequency was produced by a low-amplitude abdominal “buzzing” su- perimposed on the low-frequency carrier (Fig. 2). At high temperatures, primary and modulating frequencies were both high enough and of sufficiently low amplitude to render the motions of the abdomen nearly invisible. Males and females sang identically, in every respect (Tables 1, 2, 3). In 27 of 36 observed heterosexual duets, the female an- swered a spontaneous call produced by the male well before the male completed his song. When the female finished, a few sec- onds passed before the male initiated another volley, and then the female again began to sing partway through the male’s song. Thus C. comanche duets were not “polite,” in the sense of nonoverlapping. In fact, the beats that were produced from the acoustical interaction between two su- VOLUME 91, NUMBER 2 137 Table 1. Principal characteristics of the songs of individual males of Chrysoperla comanche, normalized to 27°C. Means and standard deviations are tabulated, with the number of measured volleys entered parenthetically. Averages and standard deviations of the means are shown in the bottom rows. The letters A to D refer to the parts of the song specified in Figure 1. Males Abdominal Vibration Frequency (strokes/second) Modulation Frequency (maxima/second) a Sone uration Indiv Initial (A) Mid! (B) Mid2 (C) End (D) Initial (A) Mid (C) End (D) Kec) 1 — 91.63 90.82 88.14 12.84 22.79 20.76 18.16 = + 11.35 = 1037 + 2.20 - + 0.43 + 1.47 + 5.65 = (7) (7) (7) (1) (7) (5) (7) 2 = 90.38 87.15 85.01 15.34 22 e712. 21.20 21.60 ~ C13 +'0:05 + 0.50 - + 0.45 + 0.00 + 4.19 = (4) (4) (4) (1) (4) (4) (4) 3 77.72 88.11 86.88 81.86 12.87 19.70 16.09 LS: + 4.16 + 2.41 + 3.00 lil + 1.64 + 1.02 + 1.44 +417 (4) (8) (9) (9) (4) (8) (6) (9) 4 77.56 86.64 85.55 81.10 8.35 20.85 17.09 17.99 + 3.13 = 2.33 e121 = 2.69 - + O75 se SAT + 3.58 (12) (16) (16) (11) (1) (16) (2) (15) 5 70.16 89.56 87.48 81.03 _ 21.74 18.56 18.01 — a= 0)54/) + 0.47 + 4.32 ad 10:37 + 1.61 + 5.43 (1) (6) (6) (4) = (6) (3) (4) 6 727 90.86 87.79 77.01 = 20.54 = 16.29 + 5.91 2:19 E52 ED. _ + 0.94 - + 6.00 (4) (5) (5) (4) = (5) = (5) a 75.92 86.18 83.98 80.47 16.20 21:93 20.13 17.14 + 0.00 + 3.40 + 1.78 + 0.76 + 4.24 + 0:39 + 3.02 #11052 (2) (7) (7) (7) (5) (7) (6) (7) 8 68.28 85.17 84.99 72.51 14.51 20.13 16.11 13.56 #275512 + 1.08 + 1.24 + 2.01 an PB Xo) + 0.68 += 1253 = 3.29 (11) (11) (11) (1) (5) (11) (11) (11) 9 713.92 89.71 85.63 81.09 — 22.25 18.33 18.06 + 2.45 + 2.28 + 0.33 3.20) — + 0.43 + 1.42 + 4.84 (5) (6) (6) (6) = (6) (6) (6) Avg: 73.68 88.69 86.70 80.91 13.35 21.41 18.53 16.88 SD: 3.66 2.27 1.99 4.42 2519 1.14 2.02 3.00 N: 7.00 9.00 9.00 9.00 6.00 9.00 8.00 9.00 perimposed tones of similar but not iden- tical frequency were very audible over head- phones during the overlapping phases of the songs. One to ten of such sequences always occurred before copulation. Just prior to copulation, the distinctions between male and female calls became unclear: in five ob- served copulations, each partner sang with- out pause during the final 15 seconds of courtship, producing a cacophony of sub- strate-borne noise. The last phase of court- ship was also characterized by a five to ten second period of wing bumping and mutual abdominal tapping, as in C. rufilabris and most other lacewing species. Insects were tested and measured at a number of different temperatures. Data for three of the eight principal song features are graphed against temperature in Figs. 3 and 4, with the linear regression equations placed 138 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 2. Principal characteristics of the songs of individual females of Chrysoperla comanche, normalized to 27°C. Means and standard deviations are tabulated, with the number of measured volleys entered paren- thetically. Averages and standard deviations of the means are shown in the bottom rows. The letters A to D refer to the parts of the song specified in Figure 1. Females Abdominal Vibration Frequency (strokes/second) Modulation Frequency (maxima/second) Bee uration Indiv Initial (A) Mid! (B) Mid2 (C) End (D) Initial (A) Mid (C) End (D) (seconds) l = 89.40 88.50 85.75 = 22.03 = 17.01 _ + 1.96 213 + 1.40 - + 0.58 = 216.15 = (3) (3) (2) = (3) = (3) 2 78.59 86.90 85.69 82.04 13.14 21.49 16.64 7.69 + 2.80 + 1°56 = (0195 ae SiS) De. #10533 seu D EDIT + 4.60 (5) (6) (6) (6) (3) (6) (6) (6) 3} 79.72 88.17 86.75 83.29 13.34 20.75 19.12 13.56 + 3.46 + 1.60 + 1.02 = 2:96 + 3.54 + 1.34 + 2.10 = 21677 (4) (9) (9) (6) (2) (9) (4) (9) 4 76.49 86.50 84.89 76.10 14.23 D1R25 16.01 11.91 + 2.94 ae) lal + 0.89 3533 + 2.87 + 0.35 + 1.63 223.85) (16) (22) (22) (22 (6) (22) (22) (20) 5 72.58 86.85 85.31 78.46 15.34 21.27 18.36 16.48 + 4,3] = 1k69 + 1.14 2513:35) +, 1.28 + 0.48 + 1.45 sey eDil, (9) (18) (18) (18) (3) (18) (17) (17) 6 73.66 88.68 84.90 79.53 - 21.41 17.09 12.07 - + 175 + 1.60 12911 — + 0.41 + 0.42 + 6.05 () (4) (4) (4) = (4) (4) (4) 7 70.61 90.09 88.19 79.31 _ 19.76 16.75 16.39 + 4.10 412225 = 133) 2.51 = + 0.32 + JT 2.611) (5) (10) (10) (9) = (10) (9) (10) 8 81.00 87.42 84.52 78.10 12.56 21.56 17.33 11.39 + 1.65 + 2.66 + 1.36 22199 + 1.81 + 0.74 "1216 + 6.62 (10) (17) (17) (15) (5) (15) (12) (14) Avg 76.09 88.00 86.09 80.32 13.72 21.19 1733) 13.31 SD: 3.91 1.31 1:55 3.14 1.09 0.68 1.07 3:21 N 7.00 8.00 8.00 8.00 5.00 8.00 7.00 8.00 beside each graph. Volley/SRU duration (Fig. 3) varied inversely with temperature, but the correlation was not very good: the regression coefficient, R*, was a low 0.11. All primary and modulating frequency components (Fig. 4) showed a direct, linear temperature relationship, and those in the middle portions of the call were tightly cor- related with temperature changes (R? = 0.84 to 0.85). The Q,, values calculated from the regression equations (Table 4) were statis- tically equivalent for all frequency compo- nents of the calls. Compared with C. plora- bunda and Chrysopiella minora Banks, however, pitch alterations of the songs of C. comanche were less pronounced for a given temperature change. The temperature regression equations al- lowed conversion of all measured values to their equivalents at 27°C. Individual lace- wings varied little in their normalized av- erage values for most of the principal song features (Tables 1 and 2). When different populations of insects were compared (Ta- ble 3), males were insignificantly different from females, and Isabella Lake individuals VOLUME 91, NUMBER 2 90+ @ = MID SONG, PRIMARY FREQ. 852 te a 80 i N t s = we 754 : Y = 2.13x + 28.48, r2=0.84, N=160 lu 4 se 7s } e. er > 244 o ; # = MID SONG, MODULATION FREQ a 21-4 Theat ad a so | 4 Ww 18 i 1 a Y = 1.13x — 9.43, r2=0.85, N=157 15 qa rr ae 20 «21 22) 8235) G24 25) = 26) 27) 28 TEMPERATURE, in OC Fig. 4. Effect of temperature on the primary and modulation frequencies of the mid-volley portion of the songs of all males and females of C. comanche. The filled circle or triangle at each temperature is the mean of at least 15 volleys delivered at that temperature, and the error bars bracket one standard deviation of the mean. The linear regression equation, its squared vari- ance, and its sample base are entered below each line plot. were essentially identical to those from Mentone. Only one call feature showed sig- nificant differences between compared sam- ples: Mid2 primary frequency in the Isabella and Mentone populations differed by 2 Hz. Song duration was the most variable of the measured song parameters among individ- uals (Tables 1 and 2). DISCUSSION The song of Chrysoperla comanche was found to be unique among those that have been described for eleven other species of its genus. First, it is characterized by the longest unbroken volley of any North American lacewing, averaging nearly 15 seconds in duration, but occasionally ex- ceeding 30 to 35 seconds. The species that most nearly matches C. comanche in volley length is its North American sister species, C. rufilabris, with volleys of four to eight seconds (Henry 1980a). Another species with volleys of comparable length is the un- described P3 sibling of C. plorabunda. Its volleys, produced several per SRU, can be as long as eight seconds in individuals from the Kofa Mountains of Arizona (unpub- 139 / i) _-100 200 300 400 z soo Ps MILLISECONDS AMPLITUDE 0.00 1.00 2.00 3.00 4.00 5.00 6.00 TIME, in SECONDS Fig. 5. Digitized oscillograph of the shortest re- peated unit (SRU or volley) of the male “long call” of Chrysoperla rufilabris. Inset is a half-second section taken from the middle of the volley, which shows elev- en individual oscillations of the abdomen (black smears). Note the absence of a clear frequency struc- ture; the abdomen strikes the substrate and generates high-frequency noise. lished data). The only lacewing species that seems to exceed C. comanche in volley length is an undescribed relative of C. car- nea from near Ticino in the southern Alps of Europe. A volley in this species may last for more than two minutes. However, this case may not be strictly comparable, be- cause each long bout of abdominal vibra- tion in the Ticino form actually consists of numerous, independent short volleys re- peated about eight times per second (Henry 1983a). A second remarkable feature of the C. comanche call is the modulation of the pri- mary abdominal vibration frequency by regular, intense oscillations of lower fre- quency. Some green lacewings that produce a series of short volleys in rapid succession approach but do not attain the condition seen in C. comanche. For example, a sibling species of C. downesi from the redwood for- ests of California (Founder’s Grove) repeats its short volleys at nearly 10 per second (un- published data), and two sibling species of C. carnea from different parts of the Alps of Switzerland may attain volley repetition rates of about eight per second (Henry 1983a, 140 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 3. Principal characteristics of the songs of Chrysoperla comanche, normalized to 27°C and averaged for all measured volleys produced by males versus females (upper section of table) or produced by individuals from Isabella Lake versus Mentone, California (lower section). The corresponding features of the songs of Chrysoperla rufilabris are entered at the bottom of the table for comparison. Mean values and standard deviations are tabulated, with the number of measured individuals (column 1) or volleys (all other columns) entered parenthetically. Significant differences (between rows) at the 1% level are indicated by asterisks. N/a = not applicable. Abdominal Vibration Frequency (strokes/second) Modulation Frequency (maxima/second) Toone rat Initial (A) Mid! (B) Mid2 (C) End (D) Initial (A) Mid (C) End (D) (Geconds) All Males: 73.66 88.06 86.41 80.40 14.21 21.19 18.21 16.36 (N = 9) SEEDEDO) =. 2494 + 2.35 + 5.24 S23" EZ SG NG (39) (70) (71) (63) (17) (70) (43) (67) All Females: 76.51 87.54 85.64 78.70 13.70 21.14 Les 13.36 (N = 8) + 4.57 + 2.14 Ea 1!6)/ + 3.68 22339, 101825) EO nA OF (50) (89) (89) (82) (19) (87) (74) (83) All Isabella: 75.67 88.06 86.42 *** 80.19 13.59 20.45 16.47 14.90 (N = 13) + 4.38 + 2.38 + 2.04 + 4.48 25 0H SE 122" 22:37 75108 (61) (118) (119) (106) (21) (118) (82) (112) All Mentone: 74.37 86.94 84.75" 77.41 14.42 20.95 16.99 14.13 (N = 4) + 6:57 + 2.78 Si + 3.87 9317 10196 = 27) G83 (28) (41) (41) (39) (15) (39) (35) (38) Totals: 75.26 87.77 85.98 79.44 13.94 21.16 93 14.70 (N = 17) tS 17) 9253 + 2.03 + 4.49 £2579" VEO], 2314 E558 (89) (159) (160) (145) (36) (SH) (UA), (150) Chrysoperla n/a n/a n/a n/a — 18.20 = BHII6) rufilabris: =e — +1.23 (N = 10, 28 C) _ (116) - (183) 1985a). In contrast, C. comanche’s signals consist of long, continuous volleys, with no pauses in primary abdominal vibration ac- tivity between the amplitude peaks (Fig. 2). Also, the rate or frequency of modulation, about 21 Hz, is higher by a factor of two or more than in species with discrete, rapidly repeated volleys. The interaction of the pri- mary and modulation frequencies at two temperature extremes is illustrated in Fig. 2. The universal pattern of singing within the carnea group of Chrysoperla is the pro- duction of identical SRU’s alternately by each individual of a courting pair. Whether these are single volleys or complex series of volleys, those of one partner do not overlap or acoustically interfere with those of the other. In contrast, duets in C. comanche are not “polite”; the female typically answers the male halfway through his one-volley call, and then both partners often sing simulta- neously for several minutes before copulat- ing. Similarly, in C. comanche’s closest North American relative, C. rufilabris, the male produces a song consisting of a single, long volley of abdominal vibration, and the receptive female also answers the male in the middle of his volley (Henry 1980a). Other basic song features are shared be- tween C. comanche and C. rufilabris (com- pare Figs. | and 5). In both species, the SRU is an exceptionally long volley of abdominal vibration that increases steadily in ampli- tude to a crescendo near its end. Neither species seems to repeat its long volleys at any regular interval: the volley, SRU, and song are all one and the same. And finally, both species modulate the intensity of their calls with very low frequency, high ampli- tude oscillations of the abdomen, at ap- VOLUME 91, NUMBER 2 141 Table 4. Q,, factors over the range of 20 to 30°C for each of the principal features of the songs of all individuals of Chrysoperla comanche. Corresponding values for Chrysoperla plorabunda (Henry 1983b) and for Chrysopiella minora (Henry and Johnson in press) are also tabulated. N/a = not applicable. Abdominal Vibration Frequency (strokes/second) Species Initial (A) Mid! (B) Mid2 (C) Chrysoperla comanche 1.19 132 1.30 (N = 17) Chrysoperla plorabunda 1.61 1.47 1.47 (N = 10) Chrysopiella minora 1.80 1.81 1.81 (N = 15) Modulation Frequency (maxima/second) Song Duration End (D) Initial (A) Mid (C) End (D) (seconds) 1.28 1.18 1.86 1.67 1.78 125 n/a n/a n/a 1.83 1.71 n/a n/a n/a Dali) proximately the same rate: 21 Hz in C. co- manche, and 17 Hz in C. rufilabris. These shared characteristics probably reflect the close phylogenetic relationship between the two species, because such traits are not found in the carnea group of Chrysoperla. However, important differences also exist between C. comanche and C. rufilabris. First, C. rufilabris has sexually dimorphic calls, which were not found in C. comanche. Male C. rufilabris usually produce long volleys, but females answer with much shorter bursts of abdominal vibration and have not been found to produce anything resembling the male SRU (Fig. 5; see also fig. 1F and G in Henry 1980a). Second, the song of C. rufi- labris is devoid of any primary frequency structure analogous to that of C. comanche; only lower frequency oscillations, corre- sponding to C. comanche’s modulation fre- quency, are apparent (Fig. 5, inset). It is possible that C. rufilabris, unlike other lace- wings that tremulate, strikes the substrate with its abdomen, because considerable 500 Hz noise is carried on the 17 Hz amplitude peaks (Fig. 5, inset). Whatever the cause, the songs of the two species are distinctive and “sound” different from one another through headphones, despite their several shared characteristics. Whether or not C. comanche and C. rufi- labris can interbreed is unknown. The two species overlap geographically, thus provid- ing the potential for hybridization: C. rufi- labris occurs commonly in fields and mead- ows throughout most of North and Central America, including the American Southwest and California (Bram and Bickley 1963), and C. comanche has been collected at many sites in California, Arizona, New Mexico and Texas (Bickley and MacLeod 1956). Both species reproduce continually under long daylight conditions, suggesting pro- longed overlap in their breeding seasons in areas of sympatry. Although the relevant breeding experiments remain to be done, the pronounced differences between the songs of the two sister species probably pre- clude successful interspecific courtship and mating. The phylogenetic position of another common member of the rufi/abris group, C. harrisii, is uncertain. It, too, sings during courtship, but its song is a peculiar, pro- longed session of untidy, almost random vibrational bursts, and has not yet been characterized in detail. Its dark green color, conifer-associated habits, and unique male genitalia (Bram and Bickley 1963) make it easy to distinguish from C. rufilabris or C. comanche, and its song has not been found to exhibit any of the features shared by C. rufilabris and C. comanche. This indicates that the latter two species are more closely related to each other than either is to C. harrisii, and that C. harrisii may be the sister group of the rufilabris-comanche lineage. The reproductive behavior of the fourth rel- ative of these species, C. externa, has not been investigated. 142 My studies of C. comanche and its rela- tives suggest that singing behavior has taken a different evolutionary direction in the ru- filabris group of Chrysoperla than it has in the carnea group. It also provides evidence for separating Chrysoperla into two genera, to avoid the confusion of species groups. In both groups, singing 1s essential to court- ship, and is complex in most species, per- haps because of the reinforcement of be- havioral isolating “‘mechanisms’’ in genetically similar, potentially interbreed- ing taxa (Butlin 1987). However, polite duets between partners have developed princi- pally in the carnea group, while the three members of the rufilabris group have ac- quired long, overlapping calls that may in- volve drumming on the substrate rather than tremulation. ACKNOWLEDGMENTS This work was supported by National Sci- ence Foundation award BSR-8508080, Charles S. Henry Principal Investigator. I thank my research assistant and wife, Julie J. Henry, for help in maintaining living lace- wings. Phillip Adams (California State Uni- versity, Fullerton) helped greatly in the col- lection and identification of specimens, but, more importantly, provided me with the conceptual background in chrysopid tax- onomy and phylogeny that made this study possible. Special thanks go to an anony- mous reviewer, who worked hard to im- prove the clarity of the manuscript. LITERATURE CITED Agnew, C. W., W. L. Sterling, and D. A. Dean. 1981. Notes on the Chrysopidae and Hemerobiidae of eastern Texas with keys for their identification. Southwestern Entomol., suppl. 4: 1-20. Bickley, W. E. and E. G. MacLeod. 1956. A synopsis of the nearctic Chrysopidae with a key to the gen- era (Neuroptera). Proc. Entomol. Soc. Wash. 58: 177-202. Bram, R. A. and W. E. Bickley. 1963. The green lacewings of the genus Chrysopa in Maryland PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON (Neuroptera: Chrysopidae). Univ. Maryland Agric. Exp. Stn., Bull. A-124: 1-18. Butlin, R. K. 1987. Species, speciation, and rein- forcement. Am. Nat. 130: 461-464. Hagen, K. S. and R. L. Tassan. 1966. The influence of protein hydrolysates of yeast and chemically defined diet upon the fecundity of Chrysopa car- nea Steph. Vest. csl. Spol. Zool. 30: 219-227. Henry, C.S. 1979. Acoustical communication during courtship and mating in the green lacewing Chry- sopa carnea (Neuroptera: Chrysopidae). Ann. Entomol. Soc. Am. 72: 68-79. 1980a. Acoustical communication in Chry- sopa rufilabris (Neuroptera: Chrysopidae), a green lacewing with two distinct calls. Proc. Entomol. Soc. Wash. 82: 1-8. 1980b. The courtship call of Chrysopa dow- nesi Banks (Neuroptera: Chrysopidae): Its evolu- tionary significance. Psyche 86: 291-297. 1980c. The importance of low-frequency, substrate-borne sounds in lacewing communica- tion (Neuroptera: Chrysopidae). Ann. Entomol. Soc. Am. 73: 617-621. . 1983a. Acoustic recognition of sibling species within the holarctic lacewing Chrysoperla carnea (Neuroptera: Chrysopidae). Syst. Entomol. 8: 293- 301. 1983b. Temperature-induced changes in the calls of the green lacewing, Chrysoperla plorabun- da (Neuroptera: Chrysopidae). Psyche 90: 343- 360. 1985a. Sibling species, call differences, and speciation in green lacewings (Neuroptera: Chry- sopidae: Chrysoperla). Evolution 39: 965-984. 1985b. The proliferation of cryptic species in Chrysoperla green lacewings through song di- vergence. Fla. Entomol. 68: 18-38. 1986. Good vibrations. Natural History 95: 46-53. Henry, C. S. and J. B. Johnson. In Press. Sexual sing- ing in a non-chrysoperlan green lacewing, Chry- sopiella minora Banks. Can. J. Zool. Séméria, Y. 1977. Discussion de la validité taxono- mique du sous-genre Chrysoperla Steinmann (Planipennia, Chrysopidae). Nouv. Rev. Entomol. 7: 235-238. Sheldon, J. K. and E.G. MacLeod. 1974. Studies on the biology of the Chrysopidae. 5. The develop- mental and reproductive maturation rates of Chry- sopa carnea (Neuroptera, Chrysopidae). Entomol. News 85: 159-169. Tauber, C. A. and M. J. Tauber. 1987. Inheritance of seasonal cycles in Chrysoperla (Insecta: Neu- roptera). Genet. Res., Camb. 49: 215-223. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 143-152 SURFACE SCULPTURING OF THE ABDOMINAL INTEGUMENT OF MEMBRACIDAE AND OTHER AUCHENORRHYNCHA (HOMOPTERA) CHRISTOPHER H. DIETRICH Department of Entomology, Box 7613, North Carolina State University, Raleigh, North Carolina 27695-7613. Abstract. —Scanning electron microscopic studies of the abdominal integument of au- chenorrhynchous Homoptera indicate the presence of a variety of features including sensilla, acanthae, and microtrichia, the variability of which cannot be resolved by the dissecting microscope. Three classes of structures are recognized based on their celluar components: |) multicellular processes (sensilla); 2) unicellular processes (acanthae); and 3) subcellular processes (microtrichia). Four kinds of sensilla were found: sensilla coelo- conica, sensilla papillosa (previously undescribed), sensilla trichodea, and sensilla pla- codea. In a survey of 46 genera representing 12 families (Eurymelidae, Cicadellidae, Membracidae, Biturritiidae, Aetalionidae, Cercopidae, Aphrophoridae, Tibicinidae, Cixi- idae, Delphacidae, Fulgoridae, and Flatidae), sensilla and non-sensory protuberances were found on the non-genital abdominal terga of members of all auchenorrhynchous super- families except the Fulgoroidea. Key Words: ing, sensilla Examination of the abdominal integu- ment of many auchenorrhynchous Homop- tera by scanning electron microscopy re- veals several fine-structural features of potential interest to systematists. Previ- ously, these features, which include various types of sensilla and non-sensory protuber- ances, have been neglected or viewed su- perficially as textures in revisionary works. The objectives of this paper are to describe the common fine-structural features of in- tegument sculpturing in the Membracidae and related groups, to propose a nomencla- ture for such features, and to discuss their diversity and value for hypothesizing rela- tionships among taxa. Harris (1979) summarized the descrip- tive terminology of insect integumental sculpturing, proposing the use of relative rather than absolute measures of size of the Homoptera, Auchenorrhyncha, morphology, integument, surface sculptur- individual elements of sculpturing (e.g. punctate vs. puctulate). He further proposed the prefixing of terms for sculpturing visible only at magnifications greater than 100 x with “micro-” (e.g. micropunctate). While such terminology is useful at the descriptive level, 1t can be misleading with regards to homology if the elements of the sculpturing have fine-structure themselves. For exam- ple, the terms “punctate” and “foveate” could refer to inornate depressions in the integument as well as sensillar pores. There- fore, some nomenclatural means for differ- entiating integument textures based on the fine structure of the individual elements is needed if homologies among such structures are to be taken into account. The terminology used herein is derived as much as possible from the hypothesized functional and developmental aspects of the 144 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Fig. 1. brochosomes (b). features described following Snodgrass’ (1935) classification of the cuticular features of insects as updated by Richards and Rich- ards (1979) and Zacharuk (1985). Most detailed observations of the integ- umental fine-structure of Auchenorrhyncha have been restricted to the mouthparts (e.g. Backus and McLean 1983), antennae (e.g. Lewis and Marshall 1970, Bourgoin 1985), and genitalia (e.g. Restrepo-Mejia 1980). Other observations include Wood and Mor- ris’ (1974) survey of 100 species of Mem- bracidae for the presence of articulated hairs (sensilla trichodea) on the pronotum, Wood’s (1975) observations on the ultra- structure of the membracid pronotal integ- ument, and Kitching and Filshie’s (1974) study of the anal apparatus of membracid nymphs. Smith and Littau (1960) and Giin- thart (1977) documented the presence of minute spherical excretory granules called brochosomes covering the bodies of some leafhoppers. Presently, such observations Detail of abdominal integument of a cicadellid, Draeculacephala sp. showing microtrichia (m) and have not been applied to systematic studies and the integument of body parts other than those mentioned above has been virtually ignored. MATERIALS AND METHODS Most of the materials examined were dried museum specimens (see Table | for a list of taxa examined). All are deposited in the North Carolina State University Insect Col- lection (Department of Entomology, North Carolina State University, Raleigh). The ab- domen of each was removed, glued to an aluminum stub with conductive graphite paint, coated with gold or gold/palladium in a sputter coater and observed and pho- tographed in a JEOL T200 or Philips SOST scanning electron microscope at 10-20 kV. To determine whether air-drying signifi- cantly distorted the fine-structural features being examined, some specimens preserved in 70% ethanol were dehydrated in a graded VOLUME 91, NUMBER 2 145 Figs. 2-10. Abdominal sensilla of auchenorrhynchous Homoptera. 2-6, Sensilla coeloconica. 2, Enchenopa sp. 3, Cymbomorpha amazona. 4, Acutalis tartarea. 5, Oncometopia orbona. 6, Philaenus sp. 7-9, Sensilla papillosa. 7, Cymbomorpha amazona. 8, Aconophora cultellata Walker. 9, Hypsoprora coronata. 10, Sensillum placodeum, Centrodontus atlas. series of ethanol (80, 95, and 100%), dried in a Tousimis Samdri-PVT-3B critical point drier and coated and observed as described above. In this comparison, air-drying did not appear to significantly distort the fine- structural features observed. Many of the cicadellids examined had the integument covered with brochosomes (Fig. 1; for a description, see Smith and Littau 1960) which obscured other features of the integument. According to Giinthart (1977), these can be removed by soaking the spec- imens in hexane, diethyl ether, or methanol. (Interestingly, the presence of brochosomes may itself be of taxonomic significance. They were found on several cicadellids, an eu- rymelid, and an aetalionid, but not on any of the other specimens examined—see Ta- blew) To determine the cellular basis for the individual elements of surface sculpturing, the abdomens of two specimens preserved in 70% ethanol, one pharate and one fully- sclerotized adult, were embedded in Spurr medium (Hayat 1986), sectioned with an ultramicrotome, slide mounted, stained with 146 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Table 1. Summary of observations on the abdom- inal integument of selected auchenorrynchous Ho- moptera. (Arrangement of higher taxa follows that of Evans (1977) except where otherwise noted.) Other Taxon Sensilla Features* Cicadelloidea Eurymelidae Eurymela sp. tac m, b Cicadellidae Ledra sp. t;c d Evacanthus sp. lie m, b Idiocerus sp. m, b Oncometopia sp. tc m Draeculacephala sp. m, b Membracoidea Membracidae (sensu Deitz 1975). Centrotinae Platycentrini Tylocentrus sp. t a Nessorhininae Nessorhinini Goniolomus sp. tac d Membracinae Aconophorini Aconophora spp. ticepay sd Talipedini Talipes appendiculata tc d (da Fonseca) Hoplophorionini Alchisme sp. t,c,pa m Hoplophorion sp. t,c,pa m Ochropepla sp. t,c,pa a Platycotis vittata (Fabricius) t,c,pa =m Potnia sp. t,c,pa d Stalotypa sp. t,c, pa m Umbonia crassicornis (Amyot & Serville) t,c,pa m Membracin1 Campylenchia latipes (Say) tc d Enchenopa sp. me d Hypsoprorini Hypsoprora coronata (Fabricius) t, pa a Centrodontini Centrodontus atlas Goding t,¢,pl =m Darninae Cymbomorphini Cymbomorpha amazona Stal t'¢;;pay ‘mm Procyrtini Procyrta sp. t,.c m Table 1. Continued. Other Taxon Sensilla Features* Darnini Stictopelta sp. tC m Hyphinoini Hyphinoe sp. tc m Hemikypthini Proterpia sp. t,c m Smiliinae Acutalini Acutalis tartarea (Say) in d Ceresini Spissistilus festinus (Say) tic m Cyphonia sp. tC m Amastrini Vanduzeea arquata (Say) tic m Smiliini Atymna querci (Fitch) tc m Archasia auriculata (Fitch) tC m Tragopini Tragopa sp. tic m Polyglyptini Entylia carinata (Forster) t,c,pa m Polyrhyssa sp. t,c,pa m Stegaspidinae Microcentrini Microcentrus caryae (Fitch) tc m Heteronotinae Heteronotini Heteronotus sp. LEXe m Biturritiidae (= Lampropteridae sensu Evans 1948) Tropidaspis sp. tic d Biturritia sp. tc m Aetalionidae Aetalion reticulatum (Linnaeus) tie d,b Darthula hardwickii (Gray) t d Cercopoidea Cercopidae Prosapia bicincta (Say) - = Aphrophoridae Philaenus sp. tac m Cicadoidea Tibicinidae Tibicen sp. t,¢; pl’ “m Fulgoroidea Cixiidae Pintalia sp. VOLUME 91, NUMBER 2 Table 1. Continued. Other Taxon Sensilla Features* Delphacidae Liburniella ornata (Stal) — _ Fulgoridae Alphina glauca (Metcalf) ~~ - Flatidae Metcalfa pruinosa (Say) - a * KEY: a = simple acanthae; b = brochosomes; ¢ = sensilla coeloconica; d = toothed acanthae; m = mi- crotrichia; pa = sensilla papillosa; pl = sensilla pla- codea; t = sensilla trichodea; — = absence of above features. methylene blue, and examined under a compound microscope. RESULTS Sensilla and non-sensory protuberances were found on the non-genital abdominal terga of adult representatives of all the au- chenorrhynchous superfamilies except Ful- goroidea (i.e. Membracoidea, Cicadello- idea, Cercopoidea, and Cicadoidea: see Table 1). Richards and Richards (1979) classified the cuticular protuberances of insects into four major groups based on their cellular structure: 1) multicellular undifferentiated (spines); 2) multicellular differentiated (sen- silla); 3) unicellular (acanthae); and 4) sub- cellular (microtrichia). Three of these (2, 3, and 4) were found on the abdominal integ- ument of membracids. They are described as follows: A. Sensilla. Four kinds of sen- silla-like structures were found and are named (with one addition) according to the classifications of Snodgrass (1935) and Za- charuk (1985) based on external morphol- ogy. The following designations are neces- sarily tentative until the functions of these features are elucidated through ultrastruc- tural studies. Sensilla trichodea are socketed, hair-like structures— setae — present in varying num- bers and sizes on the abdomen of many Au- chenorrhyncha (e.g. Fig. 24). Sensilla coeloconica are pegs or cones set 147 in pits approximately 1-20 um in diameter. They vary in size, abundance, and fine- structure. When present they appear as punctations or foveae or are not visible at low magnifications (Figs. 2-6). Sensilla placodea consist of a plate of sen- sory cuticle surrounded by a membranous ring 1-5 wm in diameter (Fig. 10). They were found on only two of the specimens ex- amined (see Table |) but eventually may be found to be common in some groups (e.g. Cicadoidea). Sensilla papillosa are previously unde- scribed structures, presumed to be sensilla, consisting of groups of papillae 2—5 um in diameter that may be associated with a pore or membrane. They are less abundant than sensilla coeloconica and are generally not visible at low magnifications (Figs. 7-9). Further study of the ultrastructure of these features is needed to determine whether they can be assigned to any of the other classes of sensilla. B. Acanthae, microtrichia, and deriva- tives. The unicellular structures of Mem- bracidae vary from single tooth-like or pec- tinate processes (acanthae) to divided groups of subcellular projections (microtrichia). They also vary in size and relative abun- dance, and give the integument a grainy to finely pubescent appearance at low magni- fication. Differentiation among acanthae and microtrichia presents some difficulty as it requires knowledge of the developmental origins of these structures (Richards and Richards 1979). Thin-sections of a pharate adult Archasia auriculata (Fitch) (Fig. 12, cf. Fig. 11) indicate that the small hair-like projections on the integument surface are relatively numerous compared to the un- derlying epidermal cells, demonstrating their subcellular nature. It seems likely that each of the arcuate groups of these microtrichia seen in Fig. | 1 corresponds to an individual epidermal cell and is therefore derived from an acanthus (sensu Richards and Richards 1979). There is often a wide range of variation in the relative amount of development of Figs. 11-15. acanthae and microtrichia on individual specimens. The general trend in Membrac- idae is from maximal to minimal devel- opment from anterior to posterior ends of the individual terga (Fig. 13). Thus, when comparing taxa, it may be desirable to con- sider homologous sites on the specimens, or to compare only the maximal or minimal conditions of these structures. Patterns of subcellular sculpturing, other than microtrichia, that are uniformly dis- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 11-12. Cellular structure of the cuticular protuberances of a membracid, Archasia auriculata. 11, Distribution of microtrichia (m) and sensilla trichodea (t) on abdominal tergum IV. 12, Thin-section through a portion of abdominal tergum integument showing positions of epidermal cells (e) relative to microtrichia. 13. A portion of abdominal tergum IV of Enchenopa sp. (Membracidae) illustrating variation in the integument sculpturing from anterior (left) to posterior (right). 14-15. Microtextures of membracid integument. 14, Micro- granulate, Acutalis tartarea. 15, Microrugose, Atymna guerci. tributed over the integument surface may be referred to as “microtexture.”’ Thus far, three conditions have been observed and are named according to the convention of Harris (1979): microglabrous (smooth, e.g. Fig. 27); microgranulate (grainy, Fig. 14); and microrugose (wrinkled, Fig. 15). DISCUSSION The observations documented here dem- onstrate the need to study the fine-structure VOLUME 91, NUMBER 2 Figs. 16-21. Comparisons of foveae on the abdominal terga of three membracids (lateral views). 16-17, Goniolomus tricorniger. 16, Terga II] and IV. 17, Detail, tergum IV. 18-19, Talipes appendiculata. 18, Terga II-V. 19, Detail, tergum IV. 20-21, Hypsoprora coronata. 20. Terga II and IV. 21, Detail, tergum III. of integumental sculpturing if such features are to be used in systematics. Simply char- acterizing integument features as textures may be misleading with regards to homol- ogies among such characters at the level of the individual fine-structural elements. Fur- thermore, examination of the integument fine-structure yields a wealth of morpho- logical information that is potentially ap- plicable to phylogenetic studies. These 150 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 22-27. Variation of non-sensory protuberances on abdominal tergum IV in two membracid tribes. 22- 24, Hoplophorionini. 22, Ochropepla sp. 23, Potnia sp. 24, Alchisme sp. 25-27, Aconophorini. 25, Aconophora cultellata Walker (t, sensillum trichodeum). 26, Aconophora grisescens. 27, Aconophora sp. points are best illustrated with examples Hypsoprora coronata (Fabricius) all have from the membracid taxa surveyed. large pits on the abdominal terga bearing a The species Goniolomus tricorniger Stal, superficial resemblance to one another (Figs. Talipes appendiculata (da Fonseca), and 16, 18, and 20). However, the fine-structure VOLUME 91, NUMBER 2 of these pits differs greatly: G. tricorniger (Fig. 17) has simple inornate pits, 7. ap- pendiculata (Fig. 19) has a single coeloconic sensillum in each pit, and H. coronata (Fig. 21) has a single papillose sensillum and three to five non-sensillar depressions per pit. A trend in the evolution of subcellular protuberances is evident when representa- tives of the tribe Hoplophorionini (sensu Deitz 1975) are compared. In the genus Ochropepla (Fig. 22), for example, the pro- tuberances are maximally developed as small acanthae, each bearing one or two acute projections. In the genus Potnia (Fig. 23), many of these acanthae are further di- vided into groups of three or more projec- tions. The groups of microtrichia found in the genus Alchisme (Fig. 24) are each ap- parently homologous to a single acanthus of Ochropepla. A similar trend occurs among the species of the genus 4conophora. The non-sensory protuberances of Aconophora cultellata Walker (Fig. 25) are maximally developed as single to tridentate acanthae. The acan- thae of Aconophora grisescens (Germar) (Fig. 26) are multidentate. In a third (unde- scribed) species (Fig. 27), the acanthae each have several blunt crenulations, suggesting a secondary reduction of the pointed den- ticulae found in other species of the genus. A more extensive survey of the integu- mental fine-structure of Auchenorrhyncha is needed to determine whether this previ- ously overlooked complex of characters supports current taxonomic classifications of the group. Furthermore, elucidation of the functions of the sensilla described herein may provide valuable information concern- ing the evolution and ecology of the taxa in which they occur. ACKNOWLEDGMENTS I am indebted to N. A. Leidy for provid- ing the initial micrographs that inspired this work, to C. B. Moore for invaluable tech- nical assistance, and to L. L. Deitz, M. H. Farrier, S. H. McKamey, H. H. Neunzig, 151 and an anonymous reviewer for many use- ful comments on the manuscript. This work was funded, in part, through a mini-grant from the Center for Electron Microscopy, North Carolina State University. Paper No. 11529 of the Journal Series of the North Carolina Agricultural Research Service, Ra- leigh, NC 27695. LITERATURE CITED Backus, E. A. and D. L. McLean. 1983. Sensory sys- tems and feeding behavior of leafhoppers. II. A comparison of sensillar morphologies of several species (Homoptera: Cicadellidae). J. Morphology 176: 3-14. Bourgoin, T. 1985. Morphologie antennaire des Tet- tigometridae (Hemiptera, Fulgoromorpha). Nouv. Revue Entomol. (N. S.) 2(1): 11-20. Deitz, L. L. 1975. Classification of the higher cate- gories of the New World treehoppers (Homoptera: Membracidae) North Carolina Agric. Exp. Sta. Tech. Bull. 225. 177 pp. Evans, J. W. 1948. Some observations on the clas- sification of the Membracidae and on the ancestry, phylogeny, and distribution of the Jassoidea. Trans. Royal Entomol. Soc. London 99(15): 497-515. 1977. The leafhoppers and froghoppers of Australia and New Zealand (Homoptera: Cica- delloidea and Cercopoidea). Part 2. Rec. Austra- lian Mus. 31(3): 83-129. Giinthart, H. 1977. Einfluss des insektenalters auf bestimmungsmerkmale. Biotaxonomische und rasterelektronmicroscopische untersuchungen bei kKleinzikaden (Hom. Auchenorrhyncha, Cicadel- lidae). Mitt. Schweiz. Entomol. Ges. 50: 189-201. Harris, R. 1979. A glossary of surface sculpturing. Occas. Pap. Bur. Entomol. California Dep. Agric. 28: 1-31. Hayat, M.A. 1986. Basic techniques for transmission electron microscopy. Academic Press, Orlando, Florida, 411 pp. Kitching, R. L. and B. K. Filshie. 1974. The mor- phology and mode of action of the anal apparatus of membracid nymphs with special reference to Sextius virescens (Fairmaire) (Homoptera). J. Entomol. (sec. A)49(1): 81-88. Lewis, C. T. and A. T. Marshall. 1970. The ultra- structure of the sensory plaque organs of the an- tennae of the Chinese lantern fly, Pyrops cande- laria L. (Homoptera, Fulgoridae). Tissue & Cell 2(3): 375-385. Restrepo-Mejia, R. 1980. Membracidos de Colom- bia-I. Revision parcial de las especies del género Alchisme Kirkaldy (Homoptera: Membracidae: 152 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Hoplophorioninae). Caldasia (Bogota) 13(61): 103- 164. Richards, A. G. and P. A. Richards. 1979. The cu- ticular protuberances of insects. Int. J. Insect Mor- phol. Embryol. 8: 143-157. Smith, D. S. and V. C. Littau. 1960. Cellular spe- cialization in the excretory epithelia of an insect, Macrosteles fascifrons Stal (Homoptera). J. Bio- phys. Biochem. Cytol. 8: 103-133. Snodgrass, R. E. 1935. Principles of Insect Mor- phology. McGraw-Hill, New York, 667 pp. Wood, T. K. 1975, Studies on the function of the membracid pronotum (Homoptera). II. Histology. Proc. Entomol. Soc. Washington 77(1): 78-82. Wood, T. K. and G. K. Morris. 1974. Studies on the function of the membracid pronotum (Homop- tera). I. Occurrence and distribution of articulated hairs. Can. Entomol. 106: 143-148. Zacharuk, R. Y. 1985. Antennae and sensilla, pp. 1- 69. In Kerkut, G. A. and L. I. Gilbert, eds., Com- prehensive Insect Physiology, Biochemistry, and Pharmacology. Vol. 6. Nervous System: Sensory. Pergamon Press, Oxford. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 153-157 WATER AND TEMPERATURE RELATIONS OF THE PRIMITIVE XYLOPHAGOUS COCKROACH CR YPTOCERCUS PUNCTULATUS SCUDDER (DICTYOPTERA: CRYPTOCERCIDAE) A. G. APPEL AND R. C. SPONSLER Department of Entomology and Alabama Agricultural Experiment Station, Auburn University, Alabama 36849-5413. Abstract. — Percent total body water content, cuticular permeability (CP), rate of water loss, and critical thermal maxima and minima were determined for the xylophagous cockroach Cryptocercus punctulatus. There was no difference between female and male C. punctulatus for any of the measured variables. Mean percent total body water was 75% and cuticular permeability was 44.1 weg cm~* h~' mm Hg '!. Critical thermal temperatures were 40.3°C and —7.8°C for maxima and minima, respectively. The CP of C. punctulatus was similar to that of closely related blattid cockroaches and two sympatric termite species. Critical thermal temperatures were not similar to those of related taxa. Key Words: lations, desiccation The primitive xylophagous cockroach, Cryptocercus punctulatus Scudder (Dic- tyoptera: Cryptocercidae), inhabits moist decaying hard and soft wood in the southern and northwestern United States. This species is considered an example of the close phy- logenetic relationship between the cock- roaches and termites (Cornwell 1968). Both groups have cellulose digesting gut protozoa and similar proventricular and genital mor- phology (McKittrick 1964). Within the Blattaria, C. punctulatus is most closely re- lated to the Blattidae, the family that con- tains the pest genera Blatta and Periplaneta. Not only do C. punctulatus consume wood, but they live in small social groups within wood. Colony galleries consist of networks of horizontal shafts, smaller trans- verse shafts, and enlarged arena-like areas (Nalepa 1984). The logs containing colonies Dictyoptera, Cryptocercus punctulatus, temperature sensitivity, water re- are quite damp, containing up to 80% mois- ture by weight (Appel pers. comm.). These cockroaches are primarily found in their galleries, but occasionally occur in the soil and leaf litter near decaying wood. Thus, C. punctulatus lives in a moist environment, buffered from rapid humidity and temper- ature changes. The purpose of this study was to examine aspects of the water relations and temper- ature tolerance of C. punctulatus and to compare these parameters with those of closely related cockroaches and termites. MATERIALS AND METHODS Specimens of C. punctulatus were ob- tained 7 March 1987 and 12 January 1988 from decaying logs in Cleburne Co., Ala- bama, off U.S. 431, 2.8 km SE of the Cle- burne-Calhoun Co. line (G. W. Folkerts, 154 Department of Zoology and Wildlife Sci- ence, Auburn University, pers. comm.). Adults, nymphs, and wood were returned to the laboratory for study. Water relations experiments. — The water content, cuticular permeability (CP), and percent of total body water (% TBW) lost over time was determined gravimetrically with a digital balance (0.01 mg sensitivity) (Appel et al. 1983, Mack and Appel 1986). Percent TBW was calculated as the differ- ence between the initial mass of live spec- imens and their mass after drying, divided by the initial live mass multiplied by 100. Dry mass was determined after the speci- mens were dried at ca. 50°C for at least three days or until two successive weighings did not differ >0.01 mg. CP was calculated as the ug of water lost per unit surface area (cm”’) per unit time (h) per unit saturation deficit (mm Hg) ina des- iccator chamber. Individual cockroaches were placed in uncovered 30-ml plastic cups that were put into an 1|1-liter desiccator jar maintained at 30 + 0.1°C and 0-2% RH. The conditions in the desiccator were mon- itored with a Cole-Parmer LCD Digital Hy- grometer (Model 3309-50). The hygrometer sensor was positioned at the same height in the desiccator as the specimens. Surface area was estimated for each specimen by Meeh’s formula: S = 12-M”, where S = surface area (cm?) and M = initial mass (g) (Edney and McFarlane 1974). Mass losses were mea- sured at 2, 4, 6, 8, 10, and 24 h, but cuticular permeability was only calculated from 2 h-data. Preliminary experiments showed no significant difference in the amount of water lost by live and HCN-killed cockroaches at 2 h. Therefore, 2-h respiratory water was less than the sensitivity of the balance (0.01 mg) and cuticular permeability could be cal- culated from live cockroaches without com- plications of respiratory mass losses. Hourly mass loss and % TBW was used to calculate % TBW lost over time. Temperature relations experiments. — Critical thermal maxima (CTMax) and PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON minima (CTMin) were defined as the upper or lower temperatures, respectively, that in- duced reversible knock-down when tem- peratures were rapidly increased (CTMax) or decreased (CTMin) at 1°C/min and 100% RH. Adult female and male C. punctulatus were used for CTMax and CTMin experi- ments. To determine CTMax, a 130-ml glass jar containing about 120 ml of water was sealed with a metal lid and placed into a 0.45-liter glass jar. A 5.5-cm-diameter plas- tic petri dish bottom containing a mois- tened disk of Whatman No. 2 filter paper was placed on top of the inner jar and 65 ml of water added to the space between the two jars. One cockroach was placed in the petri dish and the 0.45-liter jar was sealed with a metal lid and band. A small hole was cut in the center of the lid and a copper- constantan thermocouple (0.74-mm diam- eter) connected to a digital thermometer (Sensortek BAT-12) was threaded through the hole to contact the bottom of the petri dish. The thermocouple wire was attached to a 15-cm wooden applicator to ease ma- nipulation. Body temperature and response to probing were determined by gently touching the thermocouple to the mem- brane between the cockroaches’ meso- and metathoracic coxae. The jar-probe appa- ratus was placed on a hot plate that was adjusted to increase the temperature inside the apparatus at 1°C/min. Cockroaches were judged as knocked down when they lost co- ordination and the ability to right them- selves. Recovery was assessed after 1 min at 22.8°C. CTMin were measured in a covered 5.5- cm-diameter plastic petri dish containing a moistened disk of filter paper. A small hole was cut in the petri dish cover to allow ac- cess of the same thermocouple and probe as used for the CTMax determinations. A chill table was used as the cooling device and a constant rate of 1°C/min decline in temperature was maintained by varying the height of the petri dish above the chill table. Temperatures were measured as above, but VOLUME 91, NUMBER 2 knock down was defined as the lack of leg movement when probed. Recovery was as- sessed after 15 min at 22.3°C. Leg move- ment was used in CTMin experiments be- cause chilled cockroaches were not always immobilized on their dorsum as with CTMax. Statistical analysis.—A completely ran- domized design and a Student’s f-test (SAS Institute 1982) were used to determine sig- nificant differences in initial mass, % TBW, and cuticular permeability between sexes. Regression was used to determine if cuti- cular permeability was related to initial live mass or % TBW, and how % TBW loss was related to time. Data are expressed as means + SE. Completely randomized designs with sex as the factors were used for CTMax and CTMin. Student’s t-tests (P < 0.05) were used to determine significance. Data are ex- pressed as means + SE. RESULTS AND DISCUSSION Masses of C. punctulatus nymphs and adults ranged from 133.9 to 876.31 mg for second instars and adults, respectively. There was no difference between the masses of adult females (829.76 + 15.78 mg) and males (781.83 + 17.93 mg). All stages con- tained about 75% TBW and there was no difference between the % TBW of adult fe- males and males (Table 1). Edney (1977) summarized the % TBW ofa variety of ar- thropods (45-92% TBW) and Appel et al. (1983) reported a range of 62.7 to 71.8% TBW for ten cockroach species. The more primitive blattid species generally had greater % TBW than blattellid or blaberid species. Although the significance of % TBW is unknown, the close phylogenetic rela- tionship of C. punctulatus with the blattids may relate to the similarity of % TBW among these species. In addition, the % TBW of two sympatric termite species, the For- mosan subterranean termite, Coptotermes formosanus Shiraki and the eastern subter- ranean termite, Reticulitermes flavipes Kol- ler (both Rhinotermitidae), ranged from 62.3 155 50 — @ @ y = 1.78x + 2.73, 2 = 0.99 =e 40 } O Y = 1.85X + 1.51, r2 = 0.99 Percent Total Body Water Lost Time (h) Fig. 1. Cumulative % TBW lost by female (O) and male (@) Cryptocercus punctulatus during desiccation at 30°C and 0-2% RH. to 75.9 (Sponsler and Appel unpublished data). Percent TBW loss of female (F = 32,502.1, P = 0.0001) and male (F = 2340.1, P = 0.0001) C. punctulatus increased linearly with desiccation time (Fig. 1). There was a significant difference between the rate of % TBW loss (slope +2 SE) of female and male C. punctulatus. This difference in rate prob- ably resulted in the significant difference in 24-h mortality (Table 1). The greater mean mass of females, approximately 120 mg or 90 mg of water, accounts for the observed difference in desiccation-induced mortality. Linear rates of % TBW loss have been re- ported for cockroaches (Appel et al. 1983), termites (Sponsler and Appel unpublished data), and other insects (Mack and Appel 1986, Mack et al. 1988). Cuticular permeability determines the rate of water loss. There was no relationship be- tween the CP of female or male C. punc- tulatus and initial mass (F = 0.12; df = 1; P = 0.74 and F = 0.00; df = 1; P = 0.99, respectively). Similarly, there was no rela- tionship between CP and % TBW for males (F = 1-92; df= 1; P = 0:20), The CP of females, however, declined linearly (F = 6.47; df = 1; P = 0.03) with % TBW (CP = —0.91 % TBW + 111.69; r? = 0.45). Appel et al. (1986) found that water loss of adult male Periplaneta fuliginosa (Serville) was 156 Table 1. and male C. punctulatus [x + SE; n = 20]. Sex Initial Mass (mg) Female 596.69 + 78.20 Male 476.11 + 67.70 «yg cm? h"' mm Hg". negatively related to body lipid content. The negative relationship observed with female C. punctulatus may also be related to lipid content and factors such as age and repro- ductive status that affect lipid content. There was no difference between the CP of female and male C. punctulatus (Table 1). The CP of C. punctulatus (44.1 ug cm~? h~' mm Hg _') was similar to those of blat- tid cockroaches such as Periplaneta amer- icana (L.), Periplaneta australasiae (Fab.), Blatta orientalis (L.), and Periplaneta brun- nea Burmeister with CP values of 53.7, 43.1, 43.4, and 41.7 ug cm-* h-'! mm Hg "|, re- spectively (Appel et al. 1983). The CPs of worker C. formosanus and R. flavipes were also similar to C. punctulatus (37.5 and 27.8 ugcm~*h~'mm Hg |, respectively) (Spon- sler and Appel unpublished data). Arthro- pods with CPs of approximately 40 gener- ally inhabit hygric environments such as tropical forests, moist wood, and sewers (Edney 1977). Even though the CPs of female and male C. punctulatus were not different, the rates of % TBW loss (see above) were signifi- cantly different. Differences in initial mass explain this apparent discrepancy. Al- though both sexes had the same CP (44.1 ug cm~* h~' mm Hg~!), there was signifi- cantly more water in females than in males. Therefore, females lose a lower percentage of their % TBW than males at any given time. — Table 2. Sex CTMax Female 39.6 + 0.28 (39.0-40.5) Male 40.9 + 0.21 (39.7-40.9) % TBW 74.87 76.01 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Initial masses, percent total body water (% TBW), cuticula permeability, and mortality of female Cuticular Permeability* Mortality at 24 h 1.39 0.66 43.52 + 1.89 44.73 + 3.25 30% 90% tame There was no difference between the crit- ical thermal temperatures of female and male C. punctulatus. Critical thermal tem- peratures ranged from 40.9 to —8.5°C for CTMax and CTMin, respectively (Table 2). The CTMax of C. punctulatus is the lowest recorded for cockroaches. Previously re- ported cockroach CT Max ranged from 47.6 to 51.4°C (Appel et al. 1983); the CTMax of C. punctulatus was nearly 7.5°C lower than that of the next lowest cockroach species. Positive correlations between CTMax and habitat temperatures have been established for desert cockroaches (Cohen and Pinto 1977, Cohen and Cohen 1981), desert termites (Rust et al. 1979), and mel- oid beetles (Cohen and Pinto 1977). Appel et al. (1983) found that the CTMax of ten cockroach species, representing three fam- ilies, could be divided into four statistically distinct groups: <48.1°C, 48.7-49.1°C, 49.3-49.5°C, and >49.6°C. There was no apparent phylogenetic relationship in the ability to tolerate high temperatures, and our results with C. punctulatus support this conclusion. Few studies have reported the CTMin of insects. Sponsler and Appel (unpublished data) determined the CTMin for soldier and worker C. formosanus and R. flavipes. The mean CTMin for these termites was 13.2°C and there was no difference between stages or species. Although collected at the same time of year as these termites, the CTMin Critical thermal temperatures (°C) of adult C. punctulatus [x + SE (minimum—maximum); n = 10]. CTMin —§8.5 + 0.52 (—10.0——7.2) =7al 0:55; (—9:0-——7,.1) VOLUME 91, NUMBER 2 of C. punctulatus ranged from —10.0 to —7.1°C. In addition, C. punctulatus remain active at —5.0 to 0°C (Appel unpublished data). Preliminary studies with P. ameri- cana and B. orientalis indicated that the CTMin of these cockroaches was much higher (ca. 10°C) than that of C. punctulatus. Differences in temperature tolerance be- tween C. punctulatus and sympatric ter- mites are likely due to differences in micro- habitats. Cryptocercus punctulatus live in moist, fallen, partially buried logs in dense forests (Seelinger and Seelinger 1983, Na- lepa 1984). These logs provide a habitat in- sulated by moisture. With leaf littered soil as a heat-sink, these logs maintain lower than ambient temperatures (Appel unpub- lished data). During hot spells, evaporation cools the logs. It is also possible that the cockroaches move to preferred cooler lo- cations within their extensive galleries. Rhinotermitid termites, however, usually maintain soil contact even when they forage on surface debris. Since most colonies are located well below the soil surface (and frost line) and are composed of thousands of met- abolically active individuals, these termites are probably exposed to more constant, warm temperatures than are log inhabiting C. punctulatus. Similar CPs among C. punctulatus, blat- tid cockroaches, and termites may reflect their similar high-humidity habitats or close phylogenetic relationships. The interstitial spaces in wood or soil containing >17% moisture are at saturated humidity and do not contribute to significant body water loss. Since all of these groups inhabit moist areas, it 1s not surprising that they all have similar CPs. Our results concur with the conclu- sions of Appel et al. (1983) that, in cock- roaches, CP is related to habitat moisture and temperature sensitivity is related to habitat temperature. ACKNOWLEDGMENTS We thank George W. Folkerts, Depart- ment of Zoology and Wildlife Science, Au- 157 burn University, for collecting the C. punc- tulatus and for reviewing the manuscript. Alabama Agricultural Experiment Station Journal Series No. 17-881834P. LITERATURE CITED Appel, A. G., D. A. Reierson, and M. K. Rust. 1983. Comparative water relations and temperature sen- sitivity of cockroaches. Comp. Biochem. Physiol. 74A: 357-361. 1986. Water relations of the smokybrown cockroach, Periplaneta fuliginosa. J. Insect. Phys- iol. 32: 623-628. Cohen, A. C. and J. L. Cohen. 1981. Microclimate temperature and water relations of two species of desert cockroach. Comp. Biochem. Physiol. 69A: 165-167. Cohen, A. C. and J. D. Pinto. 1977. An evaluation of xeric adaptiveness of several species of blister beetles (Meloidae). Ann. Entomol. Soc. Am. 70: 741-749. Cornwell, P. B. 1968. The Cockroach, Volume I. Hutchinson & Co. LTD, London. 391 pp. Edney, E. B. 1977. Water Balance in Land Arthro- pods. Springer, New York. 282 pp. Edney, E. B. and J. McFarlane. 1974. The effects of temperature on transpiration in the desert cock- roach, Arenivaga investigata and in Periplaneta americana. Physiol. Zool. 47: 1-12. Mack, T. P. and A. G. Appel. 1986. Water relations of immature and adult lesser cornstalk borers, Elasmopalpus lignosellus (Lepidoptera: Pyrali- dae). Ann. Entomol. Soc. Am. 79: 579-582. Mack, T. P., A. G. Appel, C. B. Backman, and P. J. Trichilo. 1988. Water relations of several ar- thropod predators in the peanut agroecosystem. Environ. Entomol. 17: 778-781. McKittrick, F.A. 1964. Evolutionary studies of cock- roaches. Cornell Univ. Agric. Exp. Sta. Mem. 389: 1-197. Nalepa, C. A. 1984. Colony composition, protozoan transfer and some life history characteristics of the woodroach Cryptocercus punctulatus Scudder (Dictyoptera: Cryptocercidae). Behav. Ecol. So- ciobiol. 14: 273-279. Rust, M. K., D. A. Reierson, and R. H. Scheffrahn. 1979. Comparative habits, host utilization and xeric adaptations of the southwest drywood ter- mites, Jncisitermes fruticavus Rust and Jncisi- termes minor (Hagen) (Isoptera: Kalotermitidae). Sociobiology 4: 239-255. SAS Institute. 1982. SAS user’s guide: Statistics. SAS Institute, Cary, N.C. Seelinger, G. and U. Seelinger. 1983. On the social organisation, alarm and fighting in the primitive cockroach Cryptocercus punctulatus Scudder. Z. Tierpsychol. 61: 315-333. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 158-163 FIRST NEARCTIC RECORD OF THE GENUS NORDLANDERIA (HYMENOPTERA: EUCOILIDAE), WITH DESCRIPTIONS OF TWO NEW SPECIES TERRY D. MILLER Department of Entomology, Washington State University, Pullman, Washington 99164. Abstract. —The genus Nordlanderia (Hymenoptera: Cynipoidea, Eucoilidae) is reported from the Nearctic region for the first time and two new species are described. They differ from their African counterparts by the absence of coriaceous sculpturing below the meso- pleural carina and the presence of apical punctulations on the abdominal tergites. This genus is apparently restricted in distribution to arid areas of western North America and southern Africa. Key Words: coilidae Quinlan (1986) described the genus Nord- landeria and three new species from ma- terial collected in southern Africa. The ge- nus is distinguished by the presence of triangular projections on the supraclypeal area and the anterior region of the face. This is consistent with the general tendency to- ward facial projections in the Gronotoma group of genera. This genus, previously known only from Africa, can be added to the known Nearctic fauna with the discov- ery of the two new species described. METHODS AND TERMS Descriptions and measurements are largely based on scanning electron micro- graphs due to the small size of the specimens (about | mm total length) and the predom- inately reflective integument. Terms for sur- face sculpture follow Harms (1979) and morphological terms are after Richards (1977), Quinlan (1978) and Nordlander (1982). Type material will be deposited in the following institutions: British Museum Distribution, southern Africa, western North America, Cynipoidea, Eu- (Natural History); California Academy of Sciences; and the University of Idaho. Nordlanderia merickeli Miller, NEw SPECIES Figs. 1-6 Female.—1.15 mm long. Head and tho- rax black, antennae dark brown, gaster dark red-brown, legs yellow-brown and wings hyaline. Head subcircular in frontal view. Com- pound eyes normal size, weakly converging ventrally and with several very short scat- tered hairs between facets. Vertex smooth, with few hairs; ocelli of moderate size; pos- terior ocellar line longer than lateral ocellar line which is longer than ocular ocellar line. Occiput smooth, with few setiferous punc- tures. Face smooth, with scattered hairs be- low toruli and line of hairs along the inner orbital margins. Frons slightly raised. Sub- ocular sulcus distinct, with a few poorly de- fined striae above and below (Fig. 1). Cheeks below sulcus and supraclypeal area strongly VOLUME 91, NUMBER 2 protruding, forming three spine-like keels (Figs. 1, 2, 4). Mandibles bidentate; inner tooth two-thirds length of outer tooth (Fig. 1). Antennae 13-segmented, clavate, with very weakly defined 8-segmented club, seg- ment 3 equal in length to segment 4. Thorax convex in lateral view. Pronotal plate not protruding; median bridge width to plate width ratio, 13:35; lateral margins parallel; foveae on lateral margins open (Fig. 3). Propleuron smooth, with scattered long hairs except on anterior margin where ves- titure is both dense and long (Fig. 2). Meso- scutum smooth, with line of hairs replacing the notali. Scutellum with lateral bars smooth; scutellar plate longer than wide, with large posterior fovea and margins of plate bordered by setiferous punctures (Fig. 5); dorsal surface of scutellum above margin carinate-rugose, with circle of inwardly bent hairs around plate; scutellum below margin with widely spaced and regularly radiating striae (Fig. 6). Mesopleural carina complete; pleuron above and below suture smooth; area anterior to subalar pit depressed (Fig. 2). Metanotal plate indented posteriorly, as wide as distance between the propodeal ca- rinae (Fig. 5). Metapleuron smooth; meta- pleural groove absent; pleuron depressed dorsal to obsolete ridge | (Fig. 2); lobe over propodeal spiracle wing-like, nearly 2x as long as wide (Fig. 6). Propodeal carinae na- ked, parallel anteriorly and converging to- wards nucha posteriorly; lateral carinal ex- tensions continuing to near spiracle (Figs. 5, 6); propodeal surface smooth, with few 159 hairs between the carinae, and densely pu- bescent laterally (Fig. 5). Legs normal and moderately pubescent. Gaster with segment | obscured by seg- ment 2 when viewed laterally; segment | forming narrow ring; tergite 2 with a basal ring of hairs, lacking any felt-like pubes- cence; tergite 2 the largest, occupying whole lateral surface of gaster; posterior margins of segment 2—4 punctulate; punctulations of tergite 2 faint. Wing surface pubescent, with marginal ciliation of moderate length; second radial abscissa noticeably longer than first radial abscissa; radial cell closed on wing margin; submarginal vein distinct; median, discoi- dal and subdiscoidal veins barely discern- ible in some specimens. Male.—1.18 mm long. Antennae 15-seg- mented, filiform; segment 3 equal in length to segment 4, very slightly curved, not swol- len distally. Habitus similar to female, dif- fering in that metapleural lobe over pro- podeal spiracle nearly as wide as long; medial area of propodeum densely pubescent; fa- cial protuberances not generally as well de- veloped. Material examined.— U.S.A. Idaho: Nez Perce Co., Hells Gate State Park, T. D. Mil- ler collector; 13-V-83 (1 male), 5-VI-83 (1 male, | female), 20-VI-83 (1 male), 15-VII- 83 (1 male), 20-VII-83 (3 females), 23-VII- 83 (1 male, | female). Washington: Asotin Co., 9 mi. W. Clarkston, T. D. Miller col- lector; 20-V-84 (1 male). Types.— Holotype female, allotype male, — Figs. 1-6. Nordlanderia merickeli. 1, Frontal view of female head (scale line 75 um). 2, Lateral view of female head and thorax (scale line 80 um). 3, Pronotal plate of male (scale line 30 um). 4, Dorsal view of female facial projections (scale line 30 um). 5, Posterior view of female thorax (scale line 150 um). 6, Lateral view of female scutellum (scale line 35 um). (mb = median bridge; Ic = lateral carinae; mpl = metapleural lobe.) Figs. 7-12. Nordlanderia navajoae. 7, Frontal view of male head (scale line 75 um). 8, Anterior view of male head and thorax (scale line 100 um). 9, Dorsal view of female head (scale line 105 um). 10, Pronotal plate of male (scale line 87 um). 11, Lateral view of female thorax (scale line 135 wm). 12, Posterior view of female thorax (scale line 150 um). (mb = median bridge; rl = metapleural ridge 1; st = scutellar tubercle; mnp = metanotal plate.) PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON iN a4 ea —Q = =) Z a isa] = =) — 2) > 162 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON and 2 female paratypes to be deposited in the California Academy of Sciences; | male and | female paratype to be deposited in the British Museum (Natural History); 1 male and | female paratype gold coated on scanning electron microscope (S.E.M.) stubs and 2 uncoated males and | female paratype deposited in the University of Idaho. Etymology.—This species is named in honor of Frank W. Merickel of the Univer- sity of Idaho for his valued friendship and support over the years. Remarks.—This species is distinguished from the other new Nearctic species, na- vajoae, by the broader median bridge of the pronotal plate, size of the metanotal plate, complete lateral extensions of the propodeal carinae, lack of sculpture on the mesopleu- ron, ridge | absent on the metapleuron, and the well developed lobe over the propodeal spiracle. Nordlanderia nayajoae Miller New SPECIES Figs. 7-12 Female.—1.1 mm long. Antennae, head, thorax and gaster black, legs yellow-brown and wings hyaline. Head subcircular in frontal view. Com- pound eyes normal size, not protruding, nearly parallel. Vertex smooth, with scat- tered setiferous punctures, ocelli of mod- erate size; posterior ocellar line longer than lateral ocellar line which is longer than ocu- lar ocellar line (Fig. 9). Occiput smooth, with few hairs. Face smooth, with scant vestiture below toruli and irregular line of hairs along inner ocular margin. Frons slightly protrud- ing. Subocular sulcus distinct, with well de- fined striae above and below, cheeks com- pletely striate (Fig. 7). Cheeks below subocular sulcus and supraclypeal area moderately protruding to form 3 tooth-like projections (Figs. 7, 8, 9). Mandibles biden- tate; inner tooth three-fourths length of out- er tooth (Fig. 7). Antennae 13-segmented, clavate, with weakly defined 10-segmented club; segment 4 greater in length than seg- ment 3. Thorax convex in lateral view. Pronotal plate protruding slightly; median bridge width to plate width ratio, 9:55; lateral mar- gins of plate curved, not parallel; foveae on lateral margins open (Fig. 10). Propleuron smooth, with scattered long hairs except an- terior margin which is covered with short, dense hairs (Figs. 8, 11). Mesoscutum smooth; notali obsolete, replaced by line of hairs in some specimens. Scutellum with lateral bars smooth; scutellar plate longer than wide, with large posterior fovea and lateral margins with single puncture at mid- length; anterior of lateral puncture is small, setiferous tubercle (Fig. 12), setae visible only in perfect specimens; dorsal surface of scutellum above margin areolate-rugose, with few hairs, scutellum below margin car- inate-rugose (Figs. 11, 12). Mesopleural ca- rina complete; area below suture smooth, light striations present above suture ante- riorly and just below subalar pit (Figs. 8, 11). Metanotal plate indented posteriorly, not as wide as distance between the pro- podeal carinae (Fig. 12). Metapleuron smooth; metapleural groove absent, ridge 1 distinct (Fig. 11); lobe over propodeal spi- racle short, wider than long. Propodeal ca- rinae parallel anteriorly and converging to- wards nucha posteriorly; lateral carinal extensions poorly developed, not reaching spiracles (Fig. 12); propodeal surface smooth, moderately pubescent, except for naked carinae. Legs normal, moderately pu- bescent. Gaster with segment | obscured by seg- ment 2 when viewed laterally; segment | forming narrow ring; tergite 2 with basal ring of hairs, lacking any felt-like pubes- cence; tergite 2 the largest, occupying whole lateral surface of gaster; posterior margins of tergites 3 and 4 punctulate. Wing surface pubescent, with short mar- ginal ciliation; forewing broad, apically rounded; second radial abscissa slightly VOLUME 91, NUMBER 2 longer than first radial abscissa; radial cell closed on wing margin. Male.—1.1 mm long. Antennae 15-seg- mented, filiform; segment 3 curved and swollen distally, greater in length than seg- ment 4. Habitus similar to female, differing in that sculpturing of dorsal scutellar surface is less well defined. Material examined.—U.S.A. Arizona. Apache Co., | mi. south of Ganado, T. D. Miller and F. W. Merickel collectors; 20- VI/ 1-VII-85 (2 females, 1 male). Types. — Female holotype to be deposited in the California Academy of Sciences; | male and | female paratype gold coated on S.E.M. stubs deposited in the University of Idaho. Etymology.— This species was named for the Navajo Indians upon whose tribal lands this species was discovered. Remarks.—This species is distinguished from merickeli by the distinct striations on the cheeks and mesopleuron, the setiferous tubercles on the scutellar plate, metapleural ridge | present, the narrow median bridge of the pronotal plate, absence of lateral pro- podeal carinae, and the differences in an- tennal characters. DISCUSSION The distribution of Nordlanderia in North America is apparently restricted to arid areas of the western United States. In addition to the two new species described in this paper several undescribed species await further study. All Nearctic Nordlanderia species ex- amined so far, including the undescribed 163 ones, differ from their African counterparts due to the absence of coriaceous sculpturing below the mesopleural carina and the pres- ence of apical punctulations on the abdom- inal tergites. ACKNOWLEDGMENTS I am grateful for the help of Mr. John Quinlan, British Museum (Natural Histo- ry), for helpful comments and comparisons with type material in their collection and to Dr. Goran Nordlander, Swedish University of Agricultural Sciences, for the many very helpful comments on the manuscript. Ap- preciation 1s also extended to Dr. William J. Turner, Washington State University and to Dr. James B. Johnson and Mr. Frank W. Merickel of the University of Idaho for their review of the manuscript. LITERATURE CITED Harris, R. A. 1979. A glossary of surface sculpturing. Occas. Pap. Ent., State Calif., Dep. Food Agric. 28. 31 pp. Nordlander, G. 1982. Systematics and phylogeny of an interrelated group of genera within the family Eucoilidae (Insecta: Hymenoptera, Cynipoidea). Dissertation, Univ. of Stockholm, Dept. of Zo- ology, Stockholm. 32 pp. Quinlan, J. 1978. Hymenoptera: Cynipoidea; Eu- coilidae. Handbk. Ident. Br. Insects, Royal Ent. Soc. London, 8(1b): 1-58. 1986. A key to the Afrotropical genera of Eucoilidae (Hymenoptera), with a revision of cer- tain genera. Bull. Br. Mus. (Nat. Hist.) Ent. Ser. 52(4): 243-366. Richards, O. W. 1977. Hymenoptera. Introduction and Keys to Families, 2nd ed. Handbk. Ident. Br. Insects, Royal Ent. Soc. London, London, 6(1): 1- 100. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 164-168 HOST PLANTS OF NEASPILOTA IN CALIFORNIA (DIPTERA: TEPHRITIDAE) RICHARD D. GOEDEN Department of Entomology, University of California, Riverside, California 92521. Abstract. —New host-plant records from California are reported for the tephritids, Nea- spilota achilleae Johnson, N. aenigma Freidberg and Mathis, N. albiseta Freidberg and Mathis, N. appendiculata Freidberg and Mathis, N. caillistigma Freidberg and Mathis, N. stecki Freidberg and Mathis, and N. viridescens Quisenberry. Neaspilota stecki is initially reported from California. The host plants of these tephritids are mainly in the tribe Astereae of Asteraceae, although N. achilleae was reared for the first time from the western United States and from capitula of Stephanomeria virgata Bentham in the tribe Cichoreae. Besides new records from 13 species in six genera of Astereae for N. viridescens, another record for this tephritid from Lepidospartum squamatum (Gray) Gray is the first from the tribe Senecioneae for any Neaspilota. Key Words: Recent revision of the genus Neaspilota (Diptera: Tephritidae) by Freidberg and Mathis (1986) facilitated determination of specimens that I had reared from capitula of California Asteraceae (= Compositae) and enabled this report on the host-plant rela- tions of this taxon. I hereby respond to their concern that “... Confirmed rearing rec- ords are especially scarce from western United States, where half of the known species of Neaspilota occur.” MATERIALS AND METHODS Host-plant information on Neaspilota was obtained since 1980 mainly in southern California, an ecologically diverse region that is defined and treated botanically by Munz (1974) and encompasses roughly the southern third of the State. The materials and methods used in sampling mature ca- pitula and rearing Tephritidae from sam- ples were described by Goeden (1985). Sweep-net collections supplemented rear- ings. Flies were identified with keys by Diptera, Tephritidae, host plants, Neaspilota, subgenus Neorellia, Asteraceae Freidberg and Mathis (1986). Selected spec- imens were confirmed by F. L. Blanc, Cal- ifornia Department of Food and Agriculture Sacramento, and by A. Freidberg, Tel-Aviv University, Israel. All host-plant identifi- cations were confirmed by A. C. Sanders, Curator of the Herbarium of the University of California, Riverside. Plant names follow Munz and Keck (1959) and Munz (1974). RESULTS AND DISCUSSION Freidberg and Mathis (1986) reviewed the sparse information available on the natural history and host plants of Neaspilota. All known species develop in flower or seed heads of Asteraceae without forming galls. I offer the following information on the host plants of Neaspilota species in California. Neaspilota achilleae Johnson.—This species was reared from three separate sam- ples of capitula of Stephanomeria virgata Bentham collected in southern San Diego County as follows: 2 é and 2 2, SE of Barrett Junction, 8-X-1987; 7 46 and 6 9, Bratton VOLUME 91, NUMBER 2 Valley, 20-X-1987; 2 6, Deerhorn Valley, 20-X-1987. This is a new host-plant genus for N. achilleae and the first rearing record for this fly from the western United States. Freidberg and Mathis (1986) list four species of Aster, three of Chrysopsis, three of Erigeron, four of Hieraceum, and one species each of Prenanthes, Sericocarpus, and Trilisa as hosts of N. achilleae from eastern United States. I have reared other genera and species, but not this tephritid, from one to five samples each of 10 species of Aster, three of Chrysopsis, nine of Erig- eron, and two of Hieraceum from Califor- nia. In addition, I have not reared N. achil- leae from one sample of Stephanomeria cichoriacea Gray, three samples of Steph- anomeria exigua Nuttall, nor six samples of Stephanomeria pauciflora (Torrey) Nut- tall. Hieraceum, Stephanomeria, and Trili- sa belong to the tribe Cichorieae (= Lac- tuceae); Aster, Chrysopsis, and Erigeron, to the tribe Astereae. Lack of rearing records and host-plant tribal affinities suggest that the “peculiar,” apparently discontinuous, eastern and western distributions of N. achilleae by Freidberg and Mathis (1986) may represent allopatric populations with different host-plant specificities. They also reported that specimens examined from California and Arizona were smaller on the average than their eastern counterparts, which may be added evidence of allopatry. The record for N. achilleae “‘taken ... on Heracleum sp.” (Umbelliferae) in Phillips (1946) probably represents adults collected from a non-host. Sweep records from As- teraceae may be poor indicators of the host- plant affinities of nonfrugivorous, nongal- licolous Tephritidae (Goeden 1987, and unpublished data). The western populations of N. achilleae appear more stenophagous than the eastern populations, possibly mo- nophagous, and, therefore, could have been derived from eastern ancestral stock via Ci- choreae, and could be a separate species. Neaspilota aenigma Freidberg and Math- is.—Only five males have previously been 165 found in California, including the holotype and three allotypes swept in Inyo County (Freidberg and Mathis 1986). I have since reared the following specimens from capit- ula samples: Acamptopappus shockleyi Gray, 7 6 and 6 2, SE end of Kingston (Mountain) Range, NE San Bernardino Co., 27-V-1982; Lessingia lemmonii Gray, 1 °, Lockwood Valley, Los Padres Nat. Forest, Ventura Co., at ca. 1700 m elev., 31-VIII-1986; 1 8, Mo- jave River Forks, San Bernardino Nat. For- est, SW San Bernardino Co., 13-VIII-1987; Machaeranthera tortifolia (Gray) Cronquist and Keck, 2 4, along Death Valley Road, Last Chance Mountain Range, at 1130 m elev., Inyo Co., 9-VI-1987. Freidberg and Mathis (1986) reported its host plants as Haplopappus gooddingii (A. Nelson) Munz and Johnston, and possibly H. hartwegi (Gray) Blake. I have not yet sampled H. gooddingii or reared N. aenigma from 17 other California species of Hap- lopappus sampled. Haplopappus hartwegi does not occur in California (Munz and Keck 1959, Munz 1974), and may be an invalid species name (A. C. Sanders in litt. 1988). Still, all four host genera mentioned above belong to the tribe Astereae (Munz and Keck 1959), and, thus, constitute a definable grouping of host Asteraceae as reported for other oligophagous Tephritidae, e.g. certain California species of Trupanea (Goeden 1985) and Urophora (Goeden 1987). Freid- berg and Mathis (1986) noted that the ma- jority of known hosts of Neaspilota spp. in the subgenus Neore/lia, with which the pres- ent report is concerned, belong to the As- tereae. Neaspilota albiseta Freidberg and Math- is.—Freidberg and Mathis (1986) listed the host plants of this southwestern species as unknown. The following is the first host- plant rearing record for N. albiseta: 8 males and 2 females reared from the same sample of mature heads of Machaeranthera torti- folia from along Death Valley Road in the Last Chance Mountains of Inyo County in 1987 that also yielded N. aenigma. Freid- 166 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON berg and Mathis (1986) noted that some plant species are hosts to two or three species of Neaspilota. My rearing record confirms this co-occurrence in capitula from one lo- cal host-plant species population. Synphagy also is acommon mode of resource sharing in other genera of capitula-feeding Tephriti- dae, e.g. Trupanea and Urophora spp. (Goe- den 1985, 1987). Neaspilota appendiculata Freidberg and Mathis. —Freidberg and Mathis (1986) also listed the host plants of this species as un- known. The following are the first rearing records for N. appendiculata: Corethrogyne filaginifolia (Hooker and Arnott) Nuttall, 1 2, N of Thomas Mountain, San Bernardino Nat. Forest, Riverside Co., 29-IX-1982, 3 é and 5 9, Big Sandy Bluff, Sierra Nat. For- est, Fresno Co., 235000 23-VIII-1988: Ma- caeranthera canescens (Pursh) Gray, 2 6 and 10 2, Onyx Peak, San Gorgonio Mountains, San Bernardino Nat. Forest, SE San Ber- nardino Co., 22-V-1987. Both hosts are As- tereae. Neaspilota brunneostigma Doane.— Neaspilota brunneostigma, as revised by Freidberg and Mathis (1986), has not been found in California. Therefore, all of the host-plant records for this species listed in Wasbauer (1972) remain unconfirmed. Neaspilota callistigma Freidberg and Mathis.—The only host recorded for this species is Haplopappus venetus (Humboldt, Bonpland, Kunth) Blake var. vernonioides (Nuttall) Haller (Foote and Blanc 1963, Freidberg and Mathis 1986), which I con- firmed when I reared one female from ca- pitula collected near the Stanton Ranch Air- field on Santa Cruz Island, Santa Barbara Co., 13-X-1983. This female was identified as Neaspilota sp. in Goeden (1986). I sub- sequently reared two males and 16 females from capitula of H. venetus collected in Marrow Valley, SE of Dulzura, San Diego Co., 20-X-1987. My other rearing records are from H. acradenius (Greene) Blake: 1 4 and | 9, Coyote Wells, E of Ocotillo, 100-m elev., SW Imperial Co., 6-XI-1986; 3 4, Sentenac Canyon, San Diego Co., 7-X-1987; 10 6 and 5 2, Mountain Springs Pass, SE San Diego Co., 7-X-1987. I have reared other tephritid genera and species, but not N. callistigma from one to five sam- ples each of 15 other species of Haplopap- pus, therefore, this tephritid may be nearly monophagous. Neaspilota stecki Freidberg and Math- is.—Heretofore known only from 6 males swept in New Mexico (Freidberg and Math- is 1986). One male and 2 females were reared by me from capitula of Acamptopappus sphaerocephalus (Harvey and Gray) Gray collected at Snow Creek, Riverside Co., 7-V-1981. This is a new state record and the first host record for N. stecki. Neaspilota viridescens Quisenberry.—As revised by Freidberg and Mathis (1986), N. viridescens is the most common and wide- spread species in this genus in California. Formerly reported only from Aster spinosus Bentham (Freidberg and Mathis 1986), this tephritid was additionally reared from ca- pitula of: 4. eatonii (Gray) Howell, 6 4 and 13 2, Perazo Meadow, NW of Truckee, Ta- hoe Nat. Forest, 2200-m elev., Plumas Nat. Forest, Plumas Co., 10-IX-1986; A. scro- pulorum Gray, 46 é and 42 2, Cowhorn Val- ley, Inyo Nat. Forest, NE Inyo Co., 30-VI- 1982; Chrysothamnus teretifolius Durand and Hilgard, 53 6 and 53 9, Mountain Springs, SW Imperial Co., 6-XI-1986:; Evrig- eron argentatus Gray, 3 6 and | 2, White Mountain, 1630-m elev., Inyo Nat. Forest, Inyo Co., 18-VI-1986; E. divergens Torrey and Gray, 7 é and 11 2, Seven Oaks, along Santa Ana River, San Bernardino Nat. For- est, SW San Bernardino Co., 5-VII-1983; Gutierrezia sarothrae (Pursh) Britton and Rusby, | ¢, Mountain Springs, SW Imperial Co., 6-XI-1986; Haplopappus cooperi (Gray) Hall, 284 46 and 280 2, Walker Well, Walker Pass, Kern Co., 21-V-1986; H. ericoides (Lessing) Hooker and Arnott subsp. blakei C. B. Wolf, 1 ¢ and 3 2, Orcutt, Santa Bar- bara Co., 12-XI-1980; H. /aricifolius Gray, 13 d6and 12 2, NW of Kessler Peak at S end VOLUME 91, NUMBER 2 of Ivanpah Mountains, NE San Bernardino Co., 21-X-1982; H. linearifolius de- Candolle, 33 6 and 20 2, N Meadow Creek, Sequoia Nat. Forest, Tulare Co., 21-V-1986; H. palmeri Gray, 3 4 and 4 2, N of Barrett Junction, San Diego Co., 20-X-1987; H. propinquus Blake, 27 4 and 35 2, along Kitchen Creek, Cleveland Nat. Forest, San Diego Co., 14-X-1981; Lepidospartum squamatum (Gray) Gray, 11 éand 10°, NW of Stepladder Mountains, SE San Bernar- dino Co., 22-X-1982; Machaeranthera ca- nescens (Pursh) Gray, | 6 and 4 2, Keystone Canyon, Inyo Nat. Forest, Inyo Co., 1-IX- 1982; M. gracilis (Nuttall) Gray, 2 6 and 2 9, Cedar Canyon, New York Mountains, NE San Bernardino Co., 20-IX-1983. In addition, I confirmed the host record for Aster spinosus with flies reared as fol- lows: 9 6 and 13 2, Obsidian Butte, Imperial Co., 7-XII-1983; 4 3, Calipatria, Imperial Co., 28-XI-1984; 17 6 and 28 2, NW of Blythe, Riverside Co., 2-XTI-1987. The hosts of N. viridescens include plant species from two tribes of Asteraceae: As- tereae (Aster, Chrysothamnus, Erigeron, Gutierrezia, Haplopappus, Macaeranthera) and Senecioneae (Lepidospartum). Appar- ently, this is the first rearing record of a Neaspilota from the Senecioneae (Freidberg and Mathis 1986). I have confirmed Lepi- dospartum squamatum as a host as follows: 7 6 and 7 2, SE of Essex at Danby Cross- roads, SE San Bernardino Co., 20-X-1982. The coexistence in southern California of an apparently generalist feeder like N. vir- idescens with oligophagous and monopha- gous congeners has been shown for several generalist 7rupanea species, including T. jo- nesi Curran, T. nigricornis (Coquillett), and T. radifera (Goeden 1985). However, another interpretation of these rearing rec- ords is that N. viridescens, as presently de- fined, is composed of several oligophagous and monophagous species not yet separable morphologically. Neaspilota wilsoni Blanc and Foote.—In stating that host plants of N. wi/soni are un- 167 known, Freidberg and Mathis (1986) missed my rearing record for this tephritid from Haplopappus squarrosus Hooker and Ar- nott subsp. grindeloides (deCandolle) Keck (Goeden 1983). They listed Coreopsis cal- liosidea deCandolle in the tribe Heliantheae as an unconfirmed, but suspected host. My rearing record from a member of the As- tereae casts doubt on the validity of this unconfirmed record for this possible mono- phage. Also, I sampled, but was unable to obtain specimens from up to five samples of four other species of Coreopsis or 16 other species of Haplopappus, many of which yielded other Tephritidae. CONCLUSION The genus Neaspilota, as now known in California, is composed of species with a full range of host-plant specificities, includ- ing apparently strictly monophagous N. al- biseta, N. stecki and possibly N. wilsoni; nearly monophagous N. callistigma; oligo- phagous N. aenigma and N. appendiculata as well as general feeders, N. viridescens and possibly N. achilleae. Other species record- ed from California (Freidberg and Mathis 1986), but as yet not reared by me, for which host data are unknown or too scant to as- sess, include N. footei Freidberg and Mathis and N. pubescens Freidberg and Mathis. My host-plant records for California species of Neaspilota generally support Freidberg and Mathis’ (1986) ideas about the monophyletic lineages of the two sub- genera, Neaspilota sensu stricto, with hosts mostly in the tribe Vernoniaeae, and Neo- rellia, with hosts mostly in the Astereae. My records associate N. aenigma, N. albiseta, N. appendiculata, N. callistigma, N. stecki, N. viridescens, and N. wilsoni with hosts in the tribe Astereae. Hosts for Neaspilota (Neorellia) species other than Astereae, were confirmed by my rearing records from Ci- choreae and Senecioneae. These departures may be evidence of active speciation in the genus Neaspilota involving changes in host plants (Bush 1974, 1975). 168 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ACKNOWLEDGMENTS My thanks to F. L. Blanc, A. Freidberg, D. H. Headrick, and W. N. Mathis for their helpful comments on earlier drafts of this manuscript. The technical support of D. H. Headrick and D. W. Ricker is gratefully ac- knowledged. LITERATURE CITED Bush, G. L. 1974. Mechanisms of sympatric host race formation of the true fruit flies (Tephritidae), pp. 3-23. In White, M. J. D., ed., Genetic Mecha- nisms of Speciation in Insects. Australia and New Zealand Book Co., Sydney, Australia. 1975. Modes of animal speciation. Annu. Rev. Ecol. Syst. 6: 339-394. Foote, R. H. and F. L. Blanc. 1963. The fruit flies or Tephritidae of California. Bull. Calif. Insect Surv. Te St pp: Freidberg, A. and W. N. Mathis. 1986. Studies of Terelliinae (Diptera: Tephritidae): A revision of the genus Neaspilota Osten Sacken. Smithson. Contrib. Zool. 434: 1-75. Goeden, R. D. 1983. Initial host-plant records for five species of fruit flies from southern California (Diptera: Tephritidae). Proc. Entomol. Soc. Wash. 85: 399-400. 1985. Host-plant relations of Trupanea spp. (Diptera: Tephritidae) in southern California. Proc. Entomol. Soc. Wash. 87: 564-571. 1986. New records of Tephritidae (Diptera) from Santa Cruz Island, California. Pan-Pac. Entomol. 62: 326-328. 1987. Host-plant relations of native Uro- phora spp. (Diptera: Tephritidae) in southern Cal- ifornia. Proc. Entomol. Soc. Wash. 89: 269-274. Munz, P. A. 1974. A Flora of Southern California. Univ. Calif. Press, Berkeley. 1086 pp. Munz, P. A. and D. D. Keck. 1959. A California Flora. Univ. Calif. Press, Berkeley. 1681 pp. Phillips, V. T. 1946. The biology and identification of trypetid larvae (Diptera: Trypetidae). Mem. Amer. Entomol. Soc. 12: 1-161. Wasbauer, M. W. 1972. An annotated host catalog of the fruit flies of America north of Mexico (Dip- tera: Tephritidae). Calif. Dep. Agric. Bur. Ento- mol. Occas. Pap. 19. 172 pp. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 169-174 A NEW SPECIES OF NEOLASIOPTERA (DIPTERA: CECIDOMYIIDAE) FROM BACCHARIS (ASTERACEAE) IN SOUTHERN UNITED STATES AND THE DOMINICAN REPUBLIC RAYMOND J. GAGNE AND PAUL E. BOLDT (RJG) Systematic Entomology Laboratory, PSI, Agricultural Research Service, USDA, % U.S. National Museum NHB 168, Washington, D.C. 20560; (PEB) Grassland, Soil and Water Research Laboratory, Agricultural Research Service, USDA, 808 East Blackland Road, Temple, Texas 76502. Abstract.— A new species, Neolasioptera rostrata Gagné (Diptera: Cecidomyiidae), which galls flower receptacles of several species of Baccharis (Asteraceae), is described and illustrated. It is known from Maryland, Florida, Texas, and New Mexico in the USA, and from the Dominican Republic. A field study in Texas showed that N. rostrata has two generations per year and that overwintering larvae diapause from November to the fol- lowing September. Key Words: A new species of gall midge, Neolasiop- tera rostrata, 1s described that forms a gall in the receptacle of male and female flowers of several species of Baccharis in Maryland, Florida, Texas, and New Mexico, and the Dominican Republic. The life history of N. rostrata was studied on saltwillow, Bac- charis halimifolia L. (Asteraceae: Astereae), in Texas. Saltwillow is a woody, perennial, dioecious shrub (2 to 3 m ht), that grows from Texas to Florida and north to New York (Tarver et al. 1979). This shrub in- vades pastures, rangeland, and fallow fields (Hardin 1959), but may be toxic for cattle (Kingsbury 1964, Manley et al. 1982). It is currently being studied in Texas as a poten- tial target for biological control (DeLoach et al. 1986). MATERIALS AND METHODS Anatomical terminology in the species description follows the Manual of Nearctic Diptera (McAlpine et al. 1981). Some spec- imens were permanently mounted in Can- gall midges, Baccharis, saltwillow ada balsam on slides. The specimens used for the description have been deposited in the National Museum of Natural History (USNM), Washington, D.C. The field study of N. rostrata in Texas was part of a general survey of the phy- tophagous insect fauna of B. halimifolia. Observations were made on three estab- lished plots, one to two ha each, in aban- doned pastures at Waller, Waller Co., and Port Lavaca, Calhoun Co., and along three kilometers of roadside ditch at Indianola, Calhoun Co., Texas. Each plot contained 50 to 100 plants of B. halimifolia. Flowers were sampled at about two week intervals from September 14 to November 10, 1987. At each plot, two to four stems were removed from 10 male and !0 female plants and bagged. Sample stems grew one to two m above the ground and bore 50 to 100 flow- ers. In the laboratory, each field sample was divided into four groups, and 25 flowers from each group were removed without bias for dissection. The number of galled flowers 170 and the presence of larvae, pupae, and par- asites were recorded. Galls collected on No- vember 10 at Port Lavaca were measured in cross-section at 100 with a calibrated ocular micrometer in a stereo microscope. The plots were inspected monthly from De- cember, 1981 to February, 1988, but few flowers were collected because they did not remain on the plants. In addition to the foregoing collections, large numbers of infested flowers of B. hal- imifolia were collected between October and December, 1986 at Conroe, Montgomery Co.; Liberty, Liberty Co.; and Monroe City, Chambers Co., Texas. Some flowers were dissected; others were held for emergence of adults. On July 10, 1986, a collection was made of infested flowers of Baccharis ne- glecta Britt. at Fort Leaton, Presidio Co., Texas. Neolasioptera rostrata Gagne, New SPECIES Adult.—Scale color pattern: frons white; posterior surface of head and all of scutum brown; legs white ventrally, brown dorsally; leading edge of wing brown except for white spot at juncture of RS and costa; anterior half of tergites 1-7 brown, posterior half silvery-white. Antenna with 11 flagello- meres in 6 (n = 10) (Fig. 5), 12 in 2 (n = 10). Mouthparts (Figs. 4, 5): labrum long- attenuate; labellum elongate, broadly rounded at apex, second segment more than 3 times as long as first; palpus 4-segmented. Thoracic vestiture: scutum with dorsocen- tral and lateral rows of setae, covered else- where with scales; anepisternum with scales on dorsal half to two-thirds; katepisternum bare; anepimeron with 9-12 setae and 0-2 scales. Wing length: 4, 1.2-1.6 mm (n = 10); Figs. 1-10. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 9, 1.4-1.6 mm (n = 10). Length RS to re- mainder of wing: 6, 0.55—0.60; 2, 0.54—-0.57. Male abdomen (Figs. 7-10): Tergites 1-7 short, with single row of setae along entire posterior margin, covered elsewhere with scales; tergite 8 short, unsclerotized and without scales on mesal third, sclerotized and covered with scales on lateral third. Sternites 2-6 with setae inside periphery, scales elsewhere; sternites 7-8 short, with setae posteriorly only, scales elsewhere. Genitalia as in Figs. 8-10, setulae on gono- stylus extending to midlength on venter. Female abdomen (Fig. 6): Tergites 1-6 longer than in 4, vestiture as in 4; tergite 7 less than half width of 6, double row of setae present along posterior margin, scales pres- ent on posterior half; tergite 8 approxi- mately twice length of tergite 7, longitudi- nally divided except near anterior end. Sternites 2-7 similar to 1-6 of 4, sternite 8 slightly shorter than the preceding, its setae scattered across sclerite; tergite 6 is 0.21- 0.23 length of distal half of ovipositor (n = 10). Last instar.—Length, 1.5-1.7 mm (n = 10). Integument pebbled. Spatula (Fig. 2) anteriorly with 2 triangular lobes. Papillae: three laterals on each side of spatula, two with short setae, one without; inner pleurals without setae on prothorax, with setae on remaining segments; four terminals on anal segment (Fig. 1), each with short seta; re- mainder characteristic of genus but setae short. Specimens examined.—Holotype: 4, FLORIDA, Dodge I., Miami, emerged XI- 1970 from seedheads of Baccharis glome- ruliflora, collected XI-1970, C. E. Steg- maier, Jr. Paratypes: FLORIDA: 1 4, same data as holotype. MARYLAND: 6 4, 6 2 Neolasioptera rostrata (1-2, 4-10) and N. lathami (3). 1, Eighth and anal larval segments. 2, Spatula and associated papillae. 3-5, Heads, 3, 5 in frontal view, 4 in side view. 6, Female postabdomen. 7, Male abdominal segments 5-8. 8, Male genitalia (one gonopod shown). 9, Male genitalia, lateral view. 10, Gonostylus. Scale line for Figs. 1-5, 8, 9 = 0.10 mm; 6, 7 = 0.05 mm; 10 = 0.01 mm. 171 VOLUME 91, NUMBER 2 172 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs JH. Baccharis halimifolia. Left, flowering branchlet and leaf, 1 x. Center, flower heads, one in sagittal section to show receptacle swelling, 2. Right, flower in sagittal section to show swollen receptacle and curled larva, 6x. Talbot Co., emerged VIII 1985 from flower heads of Baccharis halimifolia, collected X-1984, V. Krischik. NEW MEXICO: 2 last-instar larvae, 4 mi. e. Loving, Eddy Co., from flower heads of Baccharis salicina, T. O. Robbins. TEXAS: 3 2, 3 2, Ft. Leaton St. Pk., Presidio Co., reared from flower re- ceptacle galls on Baccharis neglecta, col- lected VI-10-1986, T. O. Robbins; 10 last- instar larvae, Indianola, Calhoun Co., from flower receptacle galls on Baccharis hal- imifolia, X1-10-1987, P. E. Boldt; 1 4, Lib- erty, Liberty Co., Hwy U.S. 90, from flower receptacle galls on Baccharis halimifolia, X-10-1986, P. E. Boldt; 1 2°, Port Lavaca, Calhoun Co., from flower receptacle galls on Baccharis halimifolia, X-28-1987, P. E. Table 1. halimifolia at three locations in Texas, 1987. Boldt; 1 4, 1 9, Waller, Waller Co., reared X-11-1986 from flower receptacle galls on Baccharis halimifolia, collected XI-1987, P. E. Boldt. DOMINICAN REPUBLIC: 5 4, 4 °, road from Constanza to San Jose de Ocoa, Prov. La Vega, emerged VII-1987 from flowers of Baccharis myrsinites, col- lected VII-17-1987, A. L. Norrbom. Distribution.— Maryland on Baccharis halimifolia, Florida on B. glomeruliflora; Texas on B. halimifolia and B. neglecta; New Mexico on B. salicina; and Dominican Re- public on B. myrsinites. Remarks. — Adults of N. rostrata are sim- ilar to those of Neolasioptera lathami Gagné, which causes stem galls on Baccharis spp. from New York to Texas (Gagné 1971). The Mean percentage (+SD) of galls of Neo/asioptera rostrata in male and female flowers of Baccharis Waller Port Lavaca Indianola Date Male Female Male Female Male Female September 24 0 0 8.0 + 3.2 0 8.0 + 11.2 0 October 14 1:0. .2:0 0 5.0+3.6 16.0 + 8.6 3/0¥a) 5.2 2.0 + 2.4 October 28 35:0 = 11.4 3010 78:3 20.0 + 7.3 6.0 + 7.6 51.0 + 8.4 12.0 + 8.9 November 10* a 11.0 + 6.0 _ 10.0 + 9.4 - 32.0 + 10.8 * No male flowers remaining on plant. VOLUME 91, NUMBER 2 principal difference between adults of these species 1s the length of the mouthparts. The labrum and labella of N. rostrata are very long, reaching almost to the distal end of the outstretched third palpal segment (Figs. 4, 5). The labrum and labella of N. lathami are similar to those of other neolasiopteras and reach only to the distal end of the sec- ond palpal segment (Fig. 3). Larvae of both N. rostrata and N. lathami have three lateral thoracic papillae, two with setae and one without on each side of the spatula. The inner pleural prothoracic pa- pilla, slightly lateral to the group of lateral papillae, has no seta in N. rostrata (Fig. 2). Further, N. rostrata has four terminal pa- pillae (Fig. 1) instead of the six found in N. lathami. BIOLOGICAL NOTES Neolasioptera rostrata produces an en- largement of the receptacle of either male or female flowers of B. halimifolia (Fig. 11). The gall is usually single-celled with one single larva. Two of the 227 galled flowers dissected during this work contained two larvae, each in its own cell and separated from the other by a thin wall. The full-grown larva is curled in a circle and fills the cham- ber. The gall in the receptacle is not appar- ent externally until the mature larva pushes the top of the gall into a conical shape and forms an exit hole at the apex. The hole is then covered with a silky membrane pre- sumably produced by the larva. The larva then pupates or diapauses. Fifteen galls measured 0.69 + 0.12 mm wide by 1.06 + 0.20 mm high (< + SD). The development of the midge is closely related to the bud and flower development of its host. In Maryland, adults emerged in August and September, 1984 from flowers collected in October of the previous year. Emergence coincided with normal flowering of B. halimifolia in Maryland. In 1987, galls were first observed in Texas on September 24, in male flower buds at two of three lo- cations (Table 1). Male flowers were prob- 3 ably attacked first because they developed sooner and were larger than female flowers. Pupae were present in both male and female flowers on October 14 at Port Lavaca. Emp- ty galls, indicating previous adult emer- gence, were found on October 28, the only sample date when N. rostrata was present in both male and female flowers at all plots. Of 600 senescent or blooming flowers sampled, 151 contained galls (25.2% infes- tation). The actual infestation rate was probably smaller because most of the se- nescent flowers had already dropped from the plant by that date and were uninfested. Of the 151 galls, adult midges had emerged from 37. Eight others contained pupae, nine contained small larvae, and 97 contained medium or large larvae. The presence of galls in the receptacles did not appear to affect pollen or seed production. Adult emergence in late October indicated the probability ofa second generation. The small larvae present at that time were probably the progeny of the last adults of the first generation to oviposit. On November 10, all male flowers had dropped, but some senescent female flowers remained on the shrubs. Of the 300 flowers sampled, 53 were galled (17.7% infestation) and contained medium to large quiescent larvae. Only a few empty galls or galls con- taining larvae were found in each of the following months through February, 1988 when sampling was discontinued. One species of Tetrastichus sp. (Eulophi- dae: Hymenoptera) was found feeding ex- ternally on a larva collected October 28, 1987 at Port Lavaca. Another was reared from a pupa and from galls collected Oc- tober 10, 1986 at Liberty. Platygaster sp. (Platygasteridae) was also reared from galls collected October 10, 1986, at Liberty, Tex- as. ACKNOWLEDGMENTS We thank the following individuals for their help with this publication: Deborah Leather Roney drew Fig. 11 and inked Figs. 174 1-10; V. Krischik, A. L. Norrbom, T. O. Robbins, and C. E. Stegmaier, Jr. collected specimens of the new species; M. E. Schauff identified the Hymenoptera; and D. R. Fer- guson, K. M. Harris, V. Krischik, W. A. Palmer, and R. V. Peterson reviewed a draft of the ms. LITERATURE CITED DeLoach, C., P. E. Boldt, H. Cordo, H. Johnson, and J. Cuda. 1986. Weeds common to Mexican and U.S. rangelands: Proposals for biological control and ecological studies, pp. 49-68. Jn Patton, D., V. Gonzales, C. Medina, L. Segura, and R. Hamre, eds., Management and Utilization of And Land Plants: Symposium Proceedings. Saltillo, Mexico. USDA For. Serv. Gen. Tech. Rep. RM-135. Gagné, R. J. 1971. Two new species of North Amer- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ican Neolasioptera from Baccharis (Diptera: Ceci- domyiidae—Compositae). Proc. Entomol. Soc. Wash. 73: 153-157. Hardin, J. 1959. Some notes on weeds in North Car- olina. J. South. Appalachian Bot. Club 24: 22-23. Kingsbury, J. M. 1964. Poisonous Plants of the United States and Canada. Prentice-Hall Inc., Englewood Cliffs, New Jersey. Manley, G. D., G. T. Edds, and S. F. Sundlof. 1982. Cattle deaths from poisonous plants. Fla. Vet. J. P1220: McAlpine, J. F., B. V. Peterson, G. E. Shewell, H. J. Teskey, J. R. Vockeroth, and D. M. Wood, eds. 1981. Manual of Nearctic Diptera. Vol. 1. Re- search Branch, Agriculture Canada. Monograph No. 27. vi + 674 pp. Tarver, D., J. Rodgers, M. Mahler, and L. Lazor. 1979. Aquatic and wetland plants of Florida. Fla. Dept. Nat. Resources. 72 pp. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 175-178 TWO NEW ANTS OF THE GENUS SOLENOPSIS (DIPLORHOPTRUM) FROM EASTERN TEXAS (HYMENOPTERA: FORMICIDAE) WILLIAM P. MAcKAy AND S. BRADLEIGH VINSON Department of Entomology, Texas A&M University, College Station, Texas 77843. Abstract. — Solenopsis subterranea, n. sp. and S. puncticeps, n. sp. are described from 10 K N of Kurten, Brazos Co., Texas, USA. Solenopsis subterranea also occurs in Louisiana. Both species are members of the subgenus Dip/orhoptrum as it 1s presently defined. These two species have potential importance as natural enemies of founding queens of the imported fire ant, So/enopsis invicta Buren. Key Words: Diplorhoptrum Ants of Solenopsis (Diplorhoptrum) are difficult to identify as the workers are very small (often less that 2 mm long) and the species demonstrate considerable conver- gence in the worker caste (Creighton 1950). They are primarily hypogaeic or “‘geo- bionts” (Kempf 1961), and most species are seldom collected unless special techniques are used. Their nests are often found in close proximity to the nests of other ant species from which they presumably steal brood or food. Ants of this subgenus are important as predators of founding queens of the im- ported fire ant, Solenopsis invicta Buren (Lammers 1987). There are undoubtedly many undescribed species in the subgenus, and it is in great need of revision. However, we are describing these two species at this time because of the need to place names on natural enemies which may be important in controlling populations of the imported fire ant. Solenopsis (Diplorhoptrum) subterranea MacKay and Vinson, New SPECIES Figs. 1-4 Description (worker).—Head length (HL—anterior median border of clypeus to Formicidae, red imported fire ant, Solenopsis invicta predation, biocontrol, median occipital margin) 0.34-0.38 mm; head width (HW—maximum, at eye level) 0.24-0.25 mm; maximum eye length (EL) 0.02-0.03 mm: scape length (SL—excluding basal condyle) 0.19-0.21 mm; Weber’s length (WL—anterior pronotal declivity to metasternal lobe) 0.30-0.35 mm; maxi- mum petiolar width (PW) 0.08-0.09 mm; maximum postpetiolar width (PPW) 0.09- 0.10 mm; cephalic index (CI = HW x 100/ HL) 67-74; ocular index (OI = EL x 100/ HL) 6-7; scape index (SI = SL x 100/HL) 52-63. Mandible with 4 well defined teeth (Fig. 1), smooth and shining with scattered punctures; clypeus with 2 well developed carinae which end in long, well developed teeth on both sides of median lobe, area between carinae depressed, smooth, shin- ing; dorsum of head smooth, polished, with numerous scattered coarse punctures which are much greater in diameter than the hairs which arise from them (Fig. 1); occiput somewhat concave in full face view; anten- nal scape smooth and shining with punc- tures which are much smaller in diameter than those on dorsum of head, with nu- merous erect and suberect hairs: eye small with a total of about 3 ommatidia. Mesosoma smooth and shining, with 176 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Figs. 1-4 Holotype of S. subterranea: |. Full face view showing representative section of punctures; 2. Lateral view of mesosoma; 3. Lateral view of petiole and postpetiole; 4. Dorsal view of petiole and postpetiole. Figs. 5-8. Holotype of S. puncticeps: 5. Full face view showing representative section of punctures; 6. Lateral view of mesosoma; 7. Lateral view of petiole and postpetiole; 8. Dorsal view of petiole and postpetiole. All illustrations are drawn to same scale. scattered punctures not much greater in di- _ petiole (Fig. 4); gaster smooth and shining ameter than the hairs which arise from them _ with scattered punctures not much greater (Fig. 2); petiole and postpetiole rounded (Fig. in diameter than hairs which arise from 3), width of postpetiole greater than that of them. VOLUME 91, NUMBER 2 Entire dorsa of head and mesosoma cov- ered by short (0.01-0.04 mm), erect and suberect hairs (Fig. 2). Light yellowish- brown with gaster and legs slightly darker, tips of mandibular teeth and eyes dark brown or black. Female and male: Unknown. Etymology.—From Latin, subter—be- neath, and terra—soil, as we have found this species only below the soil surface. Type material.—Holotype and 20 para- types to be deposited in the Museum of Comparative Zoology, Harvard University; additional paratypes (180) to be deposited in the United States National Museum, American Museum of Natural History, Cal- ifornia Academy of Sciences, Los Angeles County Museum of Natural History, British Museum of Natural History, Museo de His- toria Natural (México), the Field Museum of Natural History, the Florida State Col- lection of Arthropods, Universidade de Sao Paulo, Museo Argentino de Ciéncias Nat- urales, Universidad Nacional de Colombia, the Insect Collections of Texas A&M Uni- versity and Mississippi State University, and the collection of WPM. Distribution.— Known from the type lo- cality, 10 K N Kurten, Brazos Co. TX (east- ern Texas, on N side of Ferrill Creek Rd., 3.94 K E of turnoff from Farm Road 2038 North) and southwestern Louisiana. A sin- gle worker was captured in a subterranean trap baited with a mealworm in Louisiana (not designated as a paratype). It was col- lected in Calcasieu Parish, Sam Houston Jones State Park on 17 Aug. 1987, collection number 9724-1. We have sampled inten- sively throughout the southeastern region of the United States with subterranean baits and have not collected this species in any other locality. Biology.—Captured in subterranean pit- fall traps at 10 cm depth, baited with a cooked mixture of eggs, hamburger and honey, as well as a trap baited with live mealworms. We have not captured this species on the soil surface at the type lo- 177 cality, despite intensive searches of the area by five different individuals. Solenopsis (Diplorhoptrum) puncticeps MacKay and Vinson, New SPECIES Figs. 5-8 Description (worker).—HL 0.50-0.51 mm; HW 0.48-0.49 mm; EL 0.04 mm; SL 0.35-0.36 mm; WL 0.59-0.61 mm; PW 0.14-0.15 mm; PPW 0.16-0.18 mm; CI 95; OI 7-8: SI 68-73. Mandible with 4 well defined teeth (Fig. 5), smooth and shining with scattered punctures; clypeus with 2 well defined teeth, depressed, smooth and shin- ing between the teeth; dorsum of head smooth and shining, but with numerous large, coarse punctures (Fig. 5); occipital border slightly concave; scape smooth and shining with much smaller punctures than on the surface of head; mesosoma smooth and shining with a few scattered punctures; spiracle on propodeum very large and round in shape (Fig. 6); ventral peduncular tooth of petiole well developed, flattened and moderately pointed (Fig. 7); petiolar node not as wide as postpetiolar node (Fig. 8). Entire body surface with erect hairs (Fig. 6): those on head short (0.02—0.05 mm) and subequal in length, those on dorsum of mesosoma longer (up to 0.8 mm) and un- even in length (Fig. 6), hairs on petiole, post- petiole and gaster similar to those on me- sosoma. Color medium yellowish-brown with mandibular teeth and eyes somewhat darker. Female and male: Unknown. Etymology. — From Latin, puncta—punc- ture and ceps—derivation of caput for head. Type material.—Holotype and 10 para- types will be deposited in the Museum of Comparative Zoology, 110 paratypes will be distributed to the institutions mentioned previously. Distribution. — Known only from the type locality, 10 K N Kurten, Brazos Co. TX (same type locality as S. suwbterranea). Biology. — Captured in a subterranean trap 178 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON baited with the cooked mixture of eggs, hamburger and honey. Discussion.—Both of these species are members of the group of thief ants in which the dorsum of the head is covered with coarse punctures which are much larger in diameter than the hairs which arise from them (couplets 10-13 of Creighton’s 1950 key to species). Solenopsis subterranea is ap- parently most closely related to, and could be confused with S. tennesseensis Smith (S. longiceps in Creighton’s key). It differs in that the postpetiole is not circular as seen from above, and the head and mesosoma are covered by short erect and suberect hairs (0.01-0.04 mm long and essentially equal in length in S. subterranea, uneven in length with a range of 0.04-0.08 mm in S. fen- nesseensis). Solenopsis tennesseensis 1S somewhat larger (HL 0.41, HW 0.30, WL 0.44), but the CI(73), OI(7) and SI(59) are within the range of S. subterranea. This new species differs from S. krockowi Wheeler as it is much smaller (8S. Arockowi: HL 0.51- 0.54, WL 0.53-0.57), and has much smaller eyes (diameter in S. krockowi = 0.05, OI 9- 10). The clypeal teeth are very small in S. krockowi and the erect hairs on the head and mesosoma of S. krockowi are uneven in length and range from 0.03-0.13 mm. It differs from S. salina Wheeler in that the tooth on the anterior peduncle of the petiole of the new species is blunt and compressed (it is usually sharply pointed in S. salina). Hairs on the body surface are longer in S. salina (0.04-0.08 mm) and are greatly un- equal in length. Solenopsis puncticeps is distinctive in that the punctures on the dorsum of the head are very large and coarse. It could be con- fused with the closely related S. pergandei Forel which has similar punctures, but the postpetiole from above is not round in shape, as it is in S. pergandei. It can be distin- guished from S. krockowi by the coarser punctures on the dorsum of the head, and the diarneter of the propodeal spiracle which is about twice that of the spiracle of S. krockowi. It is easily separated from S. sub- terranea by the hairs on the propodeum which are much longer and unequal in length. We suspect these species are important natural enemies of the founding queens of the imported fire ant, as are others in the subgenus Dip/orhoptrum (Lammers, 1987). The fire ant density at the type locality of the two new species 1s much lower than in the surrounding area, possibly due in part to the presence of these species as well as an abundance of other species of the same subgenus. ACKNOWLEDGMENTS Dr. David Smith of the Systematic Ento- mology Laboratory, Agriculture Research Service, Washington, DC loaned us para- types of S. tennesseensis Smith. Dr. Alex Mintzer suggested the study area, assisted in collecting ants in the area and critically read the manuscript. Dr. Awinash Bhatkar and an anonymous reviewer critically read the manuscript. Mark Strain, Shelley Stone- cipher and Cecil Pinder helped search the area for colonies. The research was partially supported by Texas State Funds #2000- 1925. Approved as TA-23930 by the Di- rector of the Texas Agriculture Experiment Station, College Station. LITERATURE CITED Creighton, W. S. 1950. The ants of North America. Bull. Mus. Comp. Zool. 104: 1-585 + 57 plates. Kempf, W. W. 1961. A survey of the ants of the soil fauna in Surinam (Hymenoptera: Formicidae). Studia Entomol. 4: 481-524. Lammers, J. 1987. Mortality factors associated with the founding queens of Solenopsis invicta Buren, the red imported fire ant: A study of the native ant community in Central Texas. Unpublished MS Thesis, Texas A&M University, 206 pp. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 179-184 BEHAVIOR AND DEVELOPMENT OF THE WASP PTEROMBRUS RUFIVENTRIS HYALINATUS KROMBEIN (HYMENOPTERA: TIPHIIDAE), A PARASITE OF LARVAL TIGER BEETLES (COLEOPTERA: CICINDELIDAE) C. BARRY KNISLEY, DARREN L. REEVES, AND GREGORY T. STEPHENS Department of Biology, Randolph-Macon College, Ashland, Virginia 23005. Abstract. —Pterombrus rufiventris hyalinatus was found attacking third instar larvae of three Cicindela species in Arizona. Its parasitic behavior is generally similar to that described for tropical species of Prerombrus, but different in having a more elaborate burrow plugging behavior and one generation per year. Female wasps enter open burrows or occasionally dig into plugged burrows, sting the host larvae several times under the head or thorax, and oviposit on the second abdominal sternum. Burrows are closed with a primary plug of compacted soil immediately above the larvae and a secondary plug loosely packing the remainder of the burrow. Mean development time is 2.93 days for the egg stage and 8.70 days for larvae. When mature, wasp larvae detach from the host, spin a cocoon in the larval burrow, and emerge the following July with the onset of the ““monsoon”’ rains. Key Words: Tiphudae, wasp Two genera of tiphiid wasps of the subfamily Methochinae, Methocha and Pterombrus, are parasites of tiger beetle lar- vae. Some of the natural history and be- havior of several species of Methocha has been described (Adlerz 1906, Williams 1916, 1919, Iwata 1934, 1936, Burdick and Wasbauer 1959, Wilson and Farish 1973). Pterombrus is primarily tropical in distri- bution and the limited knowledge about it comes from studies of P. cicindelicidus and P. theringi in Brazil (Williams 1928) and P. piceus in Costa Rica (Palmer 1976). These studies included field observations and notes of parasitic behavior and development times. Pterombrus rufiventris is the only known Cicindela, larvae, parasite, parasitic behavior, Pterombrus, tiger beetle, United States species of the genus. Krom- bein (1949) distinguished two subspecies, P. r. rufiventris from eastern Texas, Louisiana, North Carolina, and Virginia, and P. r. hy- alinatus from western Texas, Arizona, and California. Nothing has been published on the biology of this species. During ecological studies of tiger beetles in southeastern Ar- izona, one of us (CBK) discovered P. r. hy- alinatus parasitizing larvae of Cicindela ob- soleta, C. pulchra, and C. marutha. Because of its possible importance in limiting these tiger beetle populations, field and labora- tory studies of this wasp were initiated. We present here observations on parasitic be- havior and development. Other studies present rates of parasitism of Cicindela 180 species (Knisley 1987, Knisley and Juliano, 1988) and factors that influence rates of parasitism (Knisley, in prep.). METHODS Field studies. — Field studies were done in the Sulphur Springs Valley, near Willcox, Cochise Co., AZ, during July and August from 1983-87. Eight sites representing four habitat types (playa, saltbush flat, grassland, sand ridge), each with one or more species of adult or larval Cicindela (Knisley 1987) were checked for wasps one to two times per week. During these checks all Prerom- brus were counted to determine relative abundance and seasonal activity. Wasp be- havior was observed and recorded. Laboratory studies.—The behavior of Pterombrus was studied under laboratory conditions of summer photoperiod (13L: 11D) at 24—28°C in an observation cham- ber. The chamber was made from a trans- parent plastic “shoe” box (17 x 32 x 9 cm) and supported by 20 cm wooden legs placed under each corner. An acrylic tube (2.5 x 18 cm) filled with soil and containing an active third instar of C. pulchra (Knisley and Pearson 1984) was inserted through a hole in the bottom of the box so that the top of the tube was flush with the surface of a 5—6 cm layer of soil in the box. The top of the tube was wrapped with tape to ensure a tight fit of the tube into the hole. For clear observation of wasp behavior, only tubes in which the larva had dug a burrow along the side wall of the tube were used for the trials. To begin an observation trial a female wasp was placed into the covered plastic box. We observed the complete sequence of parasitic behavior for three wasps and partial se- quences for three others. Development time of Pterombrus was de- termined from host preference studies (Knisley, in prep.) in trials in which a tiger beetle larva in a rearing tube was confined with a wasp. Each parasitized larva was carefully dug from the tubes and transferred onto a disc of moistened paper towel in a PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON plastic petri dish (10 cm dia). Develop- mental progress and body length of larval stages were recorded daily. RESULTS AND DISCUSSION Field studies.— Distribution and behav- ior of wasps in the field were based on over 70 sightings of adult wasps. Only four males were observed and all were flying near fe- males and appeared to be attempting to mate. No mating was seen. Adult wasps were found at six sites. Only one wasp was ob- served each at two of these sites, a ditch edge and a pond edge. No wasps were ob- served in the playa or playa edge habitats. Wasps were relatively common (two to five observed during each survey) at two sand ridge sites where there were large popula- tions (over 500) of C. marutha larvae and ata nearby (400 m) saltbush flat where there were several hundred C. pulchra. Pterom- brus was also common at a higher elevation grassland site near Chiricahua National Monument which had a population of over 300 C. obsoleta larvae. Adult wasps were first observed at field sites within one or two days of the first sig- nificant July rainfall (>8-10 mm). The ac- tual dates of first sightings were July 6 in 1986, July 10 in 1985 and July 16 in 1987. Wasp emergence may be triggered by the onset of the summer “monsoon” rains, as is tiger beetle adult emergence and larval activity. Numbers of wasp sightings de- creased from July through late August at all sites, but the pattern of abundance at the grassland site was about two weeks later than at the other sites. In the field, we observed P. r. hyalinatus parasitizing only third in- stars of C. pulchra, C. obsoleta, and C. ma- rutha, but in the laboratory some second instars of these species were attacked. Palm- er (1976) found that P. piceus parasitizes primarily second instars of Pseudoxychila tarsalis Bates. Searching behavior.—We observed the searching behavior of 15 wasps in the field. Female wasps walked rapidly over the VOLUME 91, NUMBER 2 ground, frequently turning and darting, in what appears to be a random pattern. The wasps stopped to examine or probe with their antennae holes, cracks or surface depressions. Often they passed within sev- eral centimeters of active larval burrows then moved away. Wasps initiated an attack only when they came in contact with the burrow edge. Pterombrus piceus searches in a sim- ilar manner and seems to prefer to attack larvae that move down the burrow when they approach (Palmer 1976). We observed four instances of wasps dig- ging out plugged burrows of third instar C. marutha, a species which typically plugs its burrows during most of the day (Knisley 1987). In each case, the wasp used its man- dibles to grasp bits of sand from the plugged area, moved back 4 to 6 centimeters and dropped the sand, eventually forming a small arc or circle of sand around the burrow mouth. Removal of the 2 to 4 cm plug took 19 to 33 minutes, after which the wasp en- tered the burrow to attack the larva. Meth- ocha (Padgen 1926) and P. itheringi (Wil- liams 1928) have also been observed digging out plugged burrows. In addition, Methocha reportedly digs in soil where tiger beetle lar- vae had previously occurred (Williams 1919). These observations suggest that wasps may detect the odor of host larvae or their burrows once they come in close proximity to them. In the laboratory chambers, Pre- rombrus initiated attacks on larvae in 4 to 38 minutes (x = 16.2). In the field, wasps were observed to search for 15-30 minutes without finding a larva. Pterombrus did not seem to fly when searching for burrows, and seldom flew un- less disturbed. Short flights (7-20 m), ob- served when wasps searched unsuccessfully in an area for 15 to 20 minutes, could in- dicate dispersal to new areas of potential larval habitat. Attack, stinging and oviposition.—At- tacks of host larvae in the laboratory were initiated by female P. r. hyalinatus moving quickly into the burrow head first, causing 181 the larva to retreat. The wasp continued down the burrow until her head touched the larva’s head. Sometimes the wasp used its mandibles to grasp the head of the larva before quickly curving her abdomen for- ward and stinging the larva under the head or thorax. This caused the larva to move further down the burrow. The wasp re- mained inactive for 2-5 minutes, then cau- tiously approached the larva and probed the larva’s head with its antenna. In two of the trials with large third instars of C. obsoleta, the larvae responded with slight movement and were again stung once or twice. The wasp used one or more of three meth- ods to slightly raise the larva in the burrow: grasping the larva’s head with its mandibles and pulling from above, pushing up on the caudal portion of the larva’s abdomen, or grasping the larva’s forelegs with its man- dibles and pushing while positioned beside the larva. These movements often wedged the larva’s cephalothoracic shield in the burrow. In two cases another sting was then applied to the larva’s venter. We never ob- served larvae flipping out of the burrow. This may be a defense reaction of the larva and is reported in studies of Pterombrus (Palmer 1976) and Methocha (Mury Meyer 1983). We also did not observe wasps being seized by tiger beetle larvae. Methocha has been observed to allow the larva to seize it with its mandibles before quickly stinging the larva. Palmer (1976) reported one in- stance of this behavior in P. piceus. After positioning the larva or sometimes after oviposition the wasp used the tip of its abdomen to pack the bottom of the bur- row. Pterombrus theringi shows similar packing behavior after using its mandibles to loosen the soil (Williams 1928). We ob- served one wasp grasp at the larva’s second abdominal sternum. The wasp may have been chewing, feeding or preparing an ovi- position site. Burdick and Wasbauer (1959) reported that M. californica commonly feeds on fluids exuding from sting punctures and suggested that this may provide nutrients 182 that are necessary for successful oviposi- tion. To oviposit, the wasp positioned itself venter to venter with the larva, moved for- ward using the tip of its abdomen to sweep back and forth over the larva’s abdominal sternites, then stopped and deposited an egg on the midline of the second abdominal sternum. In two cases the wasp next stung the beetle larva one or more times near the egg. These stings or the chewings near the Oviposition site may prepare an access for the newly hatched wasp larva to enter the host to feed. The number of stings given the beetle lar- va by both Pterombrus and Methocha is variable. Any movement by the larva seems to be sufficient to trigger a stinging response in Methocha (Burdick and Wasbauer 1959). Our observations suggested this may be true for Pterombrus. Stings also seemed to be given after certain behavioral events, such as positioning the larva and oviposition. Host larvae were apparently completely paralyzed by the stings because they did not move in their burrows, and exhibited only slight movement when probed. The effect of the stings of Methocha species has been correlated with the site of oviposition. Host larvae stung by species that oviposit on the host’s abdomen are permanently paralyzed by the host; host larvae stung by species that Oviposit on the venter of the metathorax recover and move actively in their burrows (Williams 1919, Iwata 1936). Burrow preparation and plugging. — After Oviposition the wasp moved above the larva and constructed two plugs, a compact pri- mary plug immediately above the larva and a loosely packed secondary plug which filled the rest of the burrow. To form the primary plug, the wasp positioned itself 24 cm above the paralyzed larva, curved its abdomen forward across the burrow shaft and used its mandibles to dislodge soil particles from the sides of the burrow. Soil fell onto the underside of the abdomen, was compacted there with the hind legs and then pressed PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON onto the sides of the burrow with the ab- domen. As the wasp moved in a circle, the activity eventually formed a circular shelf of soil. To complete the primary plug, the wasp moved to near the top of the burrow, used its mandibles to dislodge soil onto the shelf, then moved down and appeared to use its hind legs to cover the hole in the shelf left by its abdomen. The primary plug was 1.5-3.0 cm thick and required 7-15 minutes to complete. Next, the wasp left the burrow and searched the area within 10-20 cm of the burrow entrance for bits of soil, stones, twigs and other materials for the secondary plug. The wasp carried each piece in its mandibles and usually dropped it into the burrow from without. Filling of the burrow and comple- tion of the secondary plug required from 21 to over 140 trips and 13-41 minutes (x = 18, n = 12). Types of materials available around the burrow and depth of the burrow seemed to account for the variation in num- ber of trips and time. For example, filling of burrows at the saltbush flat site required more time because plugging materials were scarce. At a sand ridge site and in two lab- oratory chambers where the soil was nearly pure sand, wasps formed most of the sec- ondary plug by pushing sand into the bur- row with the hind legs. Plugging was then completed by filling in with larger materials. In general, wasps seemed to be more selec- tive of materials during the final phase of burrow plugging. The surface appearance of the plugged burrows was distinctive at each of the sites. The secondary plugs at the grassland site consisted of small pebbles. Those at the salt- bush flat had small twigs, bits of plant ma- terial or flakes of soil. Sand ridge burrows were filled with sand and often eroded around the edge. Methocha plugged bur- rows with a variety of materials (Bouwman 1909, Champion and Champion 1914), possibly selecting them on the basis of avail- ability and ease of transport (Burdick and Wasbauer 1959). Complete closure of the VOLUME 91, NUMBER 2 Table 1. at 24-28°C. 183 Duration and size (length in mm) of developmental stages of P. r. hyalinatus reared in the laboratory Egg length (mm) Duration egg stage (days) Larval length (mm) on day after hatch: iB ANDO Hh WwW 8 Duration of larval stage (to detachment from host) (days) Total development time (days) burrow may be necessary to protect the de- veloping wasp against natural enemies. We noticed that several incompletely plugged burrows were dug into by ants and the host larvae were eaten. Plugging burrows with two plugs by P. r. hyalinatus involves more elaborate behav- ior than the single plugging by other species of Pterombrus. This may be an adaptation to the dry environment of P. r. hyalinatus where infrequent rains are less likely to nat- urally close the burrow. The single plug con- structed by P. piceus (Palmer 1976) resem- bles the secondary plug described here. The single plug of P. iheringi (Williams 1928) is similar to the primary plug of P. r. hyali- natus. Observations of Methocha indicate plugging involves a complete filling of the burrow (Alderz 1906, Bouwman 1909, Bur- dick and Wasbauer 1959). Development.— Upon hatching the wasp larva typically remained at the site of ovi- position on the second abdominal sternum and began extracting the liquid contents of the host larva. Growth was rapid with lar- vae increasing by about 1-1.5 mm per day, from a mean length of 2.8 mm at hatching to 14.8 mm when fully grown after eight days. Detachment from the host and initi- ation of cocoon spinning began when the host was completely consumed and shriv- eled. The cephalothoracic shield of the host N Mean SD Range 15 2.39 0.14 2.10-2.52 14 2:93 0.59 2.30-3.90 9 2.86 0.38 2.52-3.57 8 3.61 0.34 2.45-4.34 9 4.70 0.58 3.90-6.09 7 6.56 0.95 5.18-7.42 11 5.76 1.01 4.64-6.24 6 9.12 1.35 5.76-10.88 8 9.04 2.26 8.00-11.20 8 14.88 1.29 13.12-16.16 [5 8.70 1.35 7.80-10.50 25: 12.33 2.65 10.70-15.30 was typically attached to the top of the pupal cocoon and apparently served as a starting point for cocoon spinning. The pupal co- coon was similar to P. cicindelidicus (Wil- liams 1928). Adult Pterombrus collected at the saltbush flat and grassland sites which had larger host tiger beetle larvae (C. pul- chra and C. obsoleta, respectively) had a mean length of 13.5 mm and those collected at the sand ridge sites with the smaller C. marutha had a mean length of 10.5 mm. In laboratory host preference studies pupal size was highly correlated with host larva size (Knisley, in prep.). Mean time from oviposition to comple- tion of the pupal cocoon was 13.2 days (SD = 1.6, range 10.2-17.3) (Table 1). Mean du- ration of stages was 2.93 days for the egg, 8.70 days for the larva, and 1.60 days from larval detachment to completion of the co- coon. Larval development is about 12 days for P. piceus (Palmer 1976) and 10-12 days for several Methocha species. The behavior we observed for P. r. hy- alinatus was similar in most aspects to that of other species of the genus. The more elab- orate plugging behavior and the one gen- eration per year may be adaptations to a different environment. Our observations also indicate similarity between the behav- ior of Pterombrus and Methocha and sug- gest that these behaviors have evolved in 184 response to the highly specialized life of lar- val tiger beetles. The apparent consistent difference between these genera is the initial attack behavior in which Methocha allows itself to be seized by the larva. Studies on searching, host selection, and abundance of these two genera of parasitic wasps could provide important additional information for determining how they limit tiger beetle populations. ACKNOWLEDGMENTS Mark R. Carter, David R. Beatty, and Frank W. Ellis assisted in the field work. Karl V. Krombein identified P. r. hyalina- tus. This research was supported in part by grants from the National Science Founda- tion (DEB8119256 and BSR8415949). LITERATURE CITED Adlerz, G. 1906. Methoca ichneumonides Latr., dess lefnadsattoch utvecklingsstadier. Arkiv for Zoo- logie, Bd. 3, No. 4, p. 1-48, 1 pl. Uppsala and Stockholm. Bouwman, B. E. 1909. Uber die Lebenweise von Methoca ichneumonides Latr. Tijdschr. voor Ent. 52: 284-294. Burdick, D. J. and M. S. Wasbauer. 1959. Biology of Methocha californica Westwood (Hymenop- tera: Tiphiidae). Wasmann Jour. Biol. 17: 75-88. Champion, H. G. and R. J. Champion. 1914. Ob- servations on the life-history of Methoca ichneu- monides Latr. Ent. Monthly Mag., Ser. 2, 25: 266- 270. Iwata, Kunio. 1934. Comparative studies on the hab- its of solitary wasps. Tenthredo 4(1-2): 1-146. 1936. Biology of two Japanese species of Methoca with the description ofa new species (Hy- menoptera, Thynnidae). Kontyu 10: 67-89. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Knisley, C. B. 1987. Habitats, food resources, and natural enemies of a community of larval Cicin- dela in southeastern Arizona (Coleoptera: Cicin- delidae). Can. J. Zool. 65: 1191-2000. Knisley, C. B. and S. A. Juliano. 1988. Survival, development, and size of larval tiger beetles: Ef- fects of food and water. Ecology 69: 1983-1992. Knisley, C. B. and D. L. Pearson. 1984. Biosystem- atics of larval tiger beetles of the Sulphur Springs Valley, Arizona. Descriptions of new species and a review of characters for Cicindela (Coleoptera: Cicindelidae). Trans. Amer. Entomol. Soc. 110: 465-551. Krombein, K. V. 1949. A new subspecies of Prerom- brus rufiventris (Cresson). Pan-Pacific Entomol. 25: 88-89. Mury Meyer, E. J. 1983. An analysis of survivorship and foraging methods in larvae of three sympatric species of tiger beetles occurring in central Penn- sylvania. Ph.D. dissertation. Pennsylvania State Univ., University Park, PA. Padgen, H. T. 1926. Observations on the habits and parthenogenesis of Methoca ichneumonides Latr. Trans. Entomol. Soc. London Ser. 4, 26: 591-597. Palmer, M. K. 1976. Notes on the biology of Pterom- brus piceus Krombein (Hymenoptera: Tiphiidae). Proc. Entomol. Soc. Wash. 78: 369-375. Williams, F. X. 1916. Notes on the life-history of Methoca stygia Say. Psyche 23: 121-125. . 1919. Phillipine wasp studies II. Descriptions of new species and life history studies. Bull. Exp. Stn. Hawaiian Sugar Planters Assoc., Entomol. Ser. 14: 19-180. . 1928. Pterombrus, a wasp-enemy of the larva of tiger beetles. /n Studies in Tropical Wasps— Their Hosts and Associates (with Descriptions of New Species). Bull. Exp. Stn. Hawaiian Sugar Planters Assoc., Entomol. Ser. 19: 144-151. Wilson, E. O. and D. J. Farish. 1973. Predatory be- haviour in the ant-like wasp Methocha stygia (Say) (Hymenoptera: Tiphiidae). Anim. Behav. 21: 292- 295. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 185-189 DESCRIPTION OF THE PREDACEOUS LARVA OF PSEUDOGAURAX SIGNATUS (LOEW) (DIPTERA: CHLOROPIDAE) ADAM ASQUITH AND DAviID ADAMSKI (AA) Systematic Entomology Laboratory, Entomology Department, Oregon State Uni- versity, Corvallis, Oregon 97331; (DA) Department of Entomology, Drawer EM, Missis- sippi State University, Mississippi State, Mississippi 39762. Abstract.—A description of the predaceous larva of Pseudogaurax signatus (Loew) (Diptera: Chloropidae) includes an illustration of the cephalopharyngeal skeleton, spinule pattern and posterior spiracles and scanning electron micrographs of the cuticular ar- mature, facial mask and the anterior and posterior spiracles. Key Words: Pseudogaurax Malloch is a large, mostly tropical group (Sabrosky 1966) whose lar- vae are predators of the immature stages of several species of Araneae and Insecta. Pseudogaurax signatus (Loew) was origi- nally placed in the genus Gaurax Loew. Sa- brosky (1945, 1966) has provided the only taxonomic treatment on this group since Hall (1937) transferred the species to the genus Pseudogaurax. P. signatus has been reared from egg sacs of the spiders Latro- dectus mactans (Fabr.), Argiope riparia (Lu- cas) and Araneus sp., oothecae of Mantidae (Davidson 1896, Hall 1937, Breland 1941), and the pupae of Hyalophora cecropia (L.) (Saturniidae) and Euproctis chrysorrhoea (L.) (Lymantriidae), (Sabrosky 1945). Chloropid larvae display diverse feeding habits, including phytophagy, scatophagy, necrophagy, and predation. The larval stages of most chloropid species are unknown, and most available descriptions are for phy- tophagous species. The only known larval descriptions of predaceous chloropids are for Steleocerellus latiseta (Lamb) (Kirk- Spriggs 1986) and Pseudogaurax anchora (Loew) (Howard 1916); in the latter how- ever, only the anterior spiracles and pupar- larval morphology, ultrastructure, spider-parasite ium were described. Roberts (1971) and Teskey (1981b) have emphasized that de- tailed larval descriptions of all immature stages together with descriptions of adults will undoubtedly be useful in elucidating phylogenetic relationships and the evolu- tion of trophic diversity in this group of flies. We thus describe the third-instar larva of Pseudogaurax signatus (Loew). MATERIALS AND METHODS One Argiope sp. egg sac was collected in the field and taken to the laboratory for dis- section. A portion of the total number of third-instar larvae of P. signatus found within the egg sac were extracted, killed in KAAD, and preserved in 70% ethanol. The incision made in the egg sac was closed and the egg sac placed in a rearing chamber at room temperature until adult flies emerged. For SEM examination, the larvae were dehydrated in ethanol and degreased in pen- tane. The specimens were then rehydrated, washed in 0.1 M phosphate buffer (pH 7.3), and postfixed in 2% osmium tetroxide in the same buffer. After dehydration in eth- anol, specimens were critical point dried, 186 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON vs vuwuyvyy Yee YEE Yuriy vy On vuwy OV UUY YY Pseudogaurax signatus, larva. 1, Lateral view. Line scale = 0.25 mm. 2, Ventral view. Line scale Figs. 1-3a. = 0.25 mm. 3, Spinules on anteroventral portion of Irst abdominal segment (enlargement of area indicated by arrow in Fig. 1). Line scale = 10 um. 3a, Spinule pattern on 4th abdominal segment (ventral view). Line scale = 0.05 mm. mounted on stubs with silver paint and S-2R scanning electron microscope at an coated with gold-palladium in a Polaron accelerating voltage of 20 kV. e5100 sputter coater. The ultrastructure of For study of the cephalopharyngeal skel- the larva was studied with an Hitachi HH-_ eton, larvae were cleared in 20% hot potas- PastmB Hyphar 4 — ——— cephalopharyngeal skeleton (lateral view). D Corn = Dorsal Cornu, Hyphar = Hypopharyngeal Sclerite, Md = Mandible, Pastm B = Parastomal Bar, V Corn = Ventral Cornu. Line scale = 0.12 mm. Fig. 4. Pseudogaurax signatus, larva, VOLUME 91, NUMBER 2 Figs. 5-11. Pseudogaurax signatus, larva. 5, Thorax, and Irst abdominal segment (dorsal view). 6, Head, facial mask (anterolateral view). 7, Head and prothorax (lateral view). 8, Head (ventrolateral view). 9, Maxillary palpus (ventrolateral view). 10, Anal plate (posterior view). Line scale for Figs. 5-7 = 100 um. Line scale for Figs. 8-10 = 10 um. A = Antenna, ASp = Anterior Spiracle, FR = Frontal Rami, MxP = Maxillary Palpus, MxR = Maxillary Ring, PSp = Posterior Spiracle. 11, Posterior spiracle (posterior view). Line scale = 0.05 mm. 188 11 sium hydroxide, mounted on microscope slides in glycerine, and examined with an Olympus compound microscope. DESCRIPTION Larva vermiform, tapering anterior of third abdominal segment, length 4.0-5.0 mm, greatest width 0.96 mm, creamy white; anterior margin of thoracic and first abdom- inal segments with several irregular rows of spinules on creeping welts (Figs. 1-3, 5). Spinules arranged in 5 regular but broken rows on ventral surface of abdominal seg- ments 2-10; spinules coarse, oriented pos- teriorly, variable in size with those on 4th row larger (Fig. 3a); all abdominal segments with two annulations, thoracic segments without annulations. Facial mask on cephalic segment bilobed, hooded by first thoracic segment (Figs. 6- 7); frontal rami consisting of two oral ridges, serrate and overlapping (Fig. 8); antenna prominent, one-segmented, directed an- teroventrally (Figs. 6-8); maxillary ring in- complete, forming a semicircle, opening dorsally; maxillary palpus in two sensillar groups (Figs. 8-9); a ventral triad is delim- ited by fleshy folds above the maxillary ring, abutting a dorsally located pair. Mandibles of cephalopharyngeal skeleton separate, strongly sclerotized, basal portion deepest, narrowing distally, curved apically (Fig. 4); hypopharyngeal sclerite nearly same length as mandibles, similar in degree of PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON melanization and sclerotization; parasto- mal sclerites slender, appearing as narrow bands above hypopharyngeal sclerite in lat- eral view, fused basally with tentorophar- yngeal sclerite, but much less sclerotized; labial sclerite present, small, located be- tween anterior ends of hypopharyngeal sclerites, not visible in lateral view; ventral cornu of tentoropharyngeal sclerite broader and longer than dorsal cornu, both pro- cesses heavily sclerotized along inner mar- gins. Anterior spiracles palmate, with seven short papillae (Figs. 6-7); posterior spiracles 0.06 mm wide, convergent with protuber- ant bases 0.18 mm apart, three spiracular openings, oval, longitudinal axis of middle opening directed dorsally, lateral openings angled, longitudinal axes converging to- wards middle opening dorsally (Fig. 1 1); anal plate oval and fleshy (Fig. 10). Specimens examined.—Larvae (n = 28): Label data Miss(issippi), Oktibbeha Co., Dorman Lake; 21 Jan(uary) 1986; Coll. Adam Asquith. Collected from Argiope sp. egg sac. Adults: (6 males, 7 females): Label data same as above except for second label: Reared from Argiope sp. egg sac. All vouch- er specimens are deposited in the Missis- sippi Entomological Museum at Mississippi State University. DISCUSSION The larva of Pseudogaurax signatus 1s comparable to that of P. anchora. Both lack the frond-like scales found on the posterior spiracles of other chloropid species such as Meromyza laeta Meigen (Fedoseyeva 1966), Polyodaspis ruficornis Macquart (Kiauka 1974), and Ectocephala capillata (Coquil- let), (Deeming 1977, 1985). The antennae of P. signatus appear to be longer and more prominent than those of P. anchora, but being a contractile organ (Yamada et al. 1981) the variation in length may be a result of differential extension during fixation. We note that the terms antenna and maxillary palpi are adopted here because of their use in the taxonomic literature (Teskey 198 1a), VOLUME 91, NUMBER 2 but these structures also represent the dorsal and terminal organs respectively of Bolwig (1946), that are commonly used in descrip- tive morphology (Chu-Wang and Axtell 1971, 1972, Yamada et al. 1981). The spinules found on the body of P. sig- natus are typical for many described chlo- ropid larvae. However, in Meromyza they are absent (Fedoseyeva 1966) and in Poly- odaspis ruficornis Macquart the spinules are restricted to the ventral surface of the ab- domen (Kiauka 1974). The mandibles of P. signatus are similar to those of the necroph- agous chloropid Conioscinella hinkleyi (Malloch) (Norrbom 1983), and a preda- ceous species Steleocerellus latiseta (Lamb) (Kirk-Spriggs 1986), in being slightly more elongate and pointed than those of phy- tophagous larvae. Unfortunately, no other characters seem to indicate the predatory habits of this larva. ACKNOWLEDGMENTS We thank Curt Sabrosky, Systematic Entomology Laboratory, USDA, U.S. Na- tional Museum, for providing us with larvae of P. anchora for examination. We thank Anna Asquith for assistance with the illus- trations. We also thank John D. Lattin, De- partment of Entomology, Oregon State Uni- versity, Gerald T. Baker and Larry D. Corpus, Department of Entomology, Mis- sissippi State University, for review of the manuscript; and Greta E. Tyson, Head, Electron Microscope Center, Mississippi State University, for providing us with the use of the electron microscope and for the preparation of photographic plates. LITERATURE CITED Bolwig, N. 1946. Senses and sense organs of the an- terior end of the house fly larva. Vid. Medd. dansk nat.-hist. Foren. 109: 81-217. Breland, O. P. 1941. Podagrion mantis Ashmead and other parasites of praying mantid egg cases (Hym.: Chalcidoidea; Dipt.: Chloropidae). Ann. Ent. Soc. Am. 34: 99-113. Chu-Wang, I. W. and R. C. Axtell. 1971. Fine struc- ture of the dorsal organ of the house fly larva, Musca domestica L. Z. Zellforsch. 117: 17-34. 189 . 1972. Fine structure of the terminal organ of the house fly larva, Musca domestica L. Z. Zell- forsch. 127: 287-305. Davidson, A. 1896. Parasites of spider eggs. Entomol. News 7: 319-320. Deeming, J. C. 1977. The immature stages of Ela- chiptereicus abessynicus Becker and Pachylophus lugens Loew (Diptera: Chloropidae). Bull. Ento- mol. Res. 67: 325-328. 1985. A description of the adult and im- mature stages of Ectecephala capillata (Coquillet) (Diptera: Chloropidae). Proc. Entomol. Soc. Wash. 87: 402-407. Fedoseyeva, L. I. 1966. Larval diagnostics of grain- grass flies of the genus Meromyza Mg. (Diptera, Chloropidae). Entomol. Rev. 45: 108-113. Hall, D. G. 1937. The North and Central American spider parasites of the genus Pseudogaurax (Dip- tera: Chloropidae). J. Wash. Acad. Sci. 27: 255- 261. Howard, L. O. 1916. The Insect Book. Doubleday, Page and Company, New York. 429 pp. Kiauka, G. F. 1974. The larvae of Polyodaspis rufi- cornis Mcq. (Diptera, Chloropidae), living in wal- nuts. Entomol. Rev. 53: 154-156. Kirk-Spriggs, A. H. 1986. The puparia of some West African species of Mepachymerus and Steleoce- rellus (Diptera: Chloropidae). J. Nat. Hist. 20: 767— 776. Norrbom, A. L. 1983. Four acalyptrate Diptera reared from dead horseshoe crabs. Entomol. News 9: 1 17— PAL Roberts, M. J. 1971. The structure of the mouthparts of some calypterate dipteran larvae in relation to their feeding habits. Acta. zool. Stockh. 52: 171- 188. Sabrosky, C. W. 1945. Three new African Chloropi- dae with a discussion of the status of Pseudogaurax (Diptera). Proc. Zool. Soc. London 114: 456-461. 1966. Three new Brazilian species of Pseu- dogaurax with a synopsis of the genus in the west- ern hemisphere (Diptera, Chloropidae). Papeis Avulsos de Dept. Zool. 19: 117-127. Teskey, H. J. 1981a. Morphology and terminology— Larvae, pp. 65-88. Jn McAlpine, J. F., et al., eds., Manual of Nearctic Diptera. Vol. 1, Monograph 27, Research Branch Agriculture Canada, vi + 674 pp. 1981b. Key to families—Larvae, pp. 125- 147. In McAlpine, J. F., et al., eds., Manual of Nearctic Diptera. Vol. 1, Monograph 27, Research Branch Agriculture Canada, vi + 674 pp. Yamada, Y., Y. Ishikawa, T. Ikeshoji, and Y. Mat- sumoto. 1981. Cephalic sensory organs of the onion fly larva, Hylemya antiqua Meigen (Dip- tera: Anthomyiidae) responsible for host-plant finding. Appl. Entomol. Zool. 16: 121-128. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 190-202 NEW SPECIES OF MICRO-CADDISFLIES (TRICHOPTERA: HYDROPTILIDAE) FROM NEW CALEDONIA, VANUATU AND FIJI ROBERT W. KELLEY Biologist, Enwright Laboratories, 25 Woods Lake Rd., Greenville, South Carolina 29607. Abstract. —New species of micro-caddisfly genera Oxyethira Eaton, Paroxyethira Mose- ly, Acritoptila Wells and Caledonotrichia Sykora are described from the South Pacific islands of New Caledonia, Vanuatu (Efate) and Fiji. Range extensions of genera Parox- yethira and Acritoptila are included and a new subgenus, endemic to New Caledonia, is added to the genus Oxyethira. Key Words: Numerous species of micro-caddisflies have been reported from the Australasian biogeographic region. Wells has noted a to- tal of 12 micro-caddisfly genera from Aus- tralia: Hydroptila Dalman, Oxyethira Eaton, Orthotrichia Eaton, Orphninotrichia Mosely, Maydenoptila Neboiss, Xuthotri- chia Mosely, Hellyethira Neboiss, Austra- trichia Wells, and Tricholeiochiton (Kloet & Hincks). Only two hydroptilid genera, Oxyethira and Paroxyethira, are known from New Zealand and one genus, Cale- donotrichia Sykora, from New Caledonia. Genera of the tribe Hydroptilini Stephens seem to be the predominant micro-caddis- flies in the region. An examination of ma- terial from the Bishop Museum in Hawaii revealed previously unpublished species of four established hydroptilid genera on the Melanesian islands of New Caledonia, Efate (Vanuatu) and Fiji. Described from New Caledonia are five new species of Oxvethira, two new species of Caledonotrichia and six new species of Acritoptila. This brings to 17 the number of species, distributed among four genera, of micro-caddisflies known from the island. caddisfly, new species, New Caledonia, Hydroptilidae One new species of Oxyethira is described from Vanuatu and one from Fiji. These dis- coveries extend the range of Oxyethira to include Melanesia, of Paroxyethira to in- clude New Caledonia as well as New Zea- land, and of Acritoptila which was previ- ously considered endemic to Australia. New species of Oxyethira from Melanesia 1n- clude species in the subgenus Trichoglene Neboiss and a new subgenus, Pacifica. The collections also showed a high incidence of endemism at the species level in New Cal- edonia. The length and number of antennal segments were unavailable for several species. All type specimens are deposited in the Bishop Museum in Hawaii. Genus Acritoptila Wells, 1982 Type species: Acritoptila globosa Wells, 1982, by original designation. Six New Caledonian species of the genus Acritoptila are described below. This genus in the Hellyethira complex of genera, was previously known only from Australia, with two species in Western Australia and one species in Queensland. The males of these VOLUME 91, NUMBER 2 crinita 7 glossocercus Figs. 1-8. 191 disjuncta 4 ; ap planichela Male terminalia of Acritoptila Wells, lateral and dorsal views. A. amphapsis: 1, lateral. A. chiasma: 2, lateral; 3, tergum X (dorsal). 4. crinita: 4, lateral. A. disjuncta: 5, lateral; 6, tergum X (dorsal). 4. glossocercus: 7, lateral. A. planichela: 8, lateral. ap, apico-mesal process; bp, bilobed process; 1a, inferior appendages; sg, subgenital processes; tX, tergum X; vp, ventrolateral process. new species share with the Australian species of the genus, fused inferior appendages (Fig. 17); a complex tergum X with spines and projections (Fig. 3) and long rod-like pro- cesses (Fig. 7), described as parameres by Wells (1982), projecting from the lateral margin of tergum X. In most of the new species the fusion of the inferior appendages is more complete than in the Australian species. Females, although not described here, share with the Hellyethira generic complex shortened apodemes on segments VIII and IX. The spur formula, wing ve- nation and other characters of the new species are consistent with the generic de- scription by Wells (1982). Acritoptila amphapsis, NEw SPECIES Figs. 1, 9, 10 Male.—Length 2.7 mm. Segment VII: ap- ico-mesal process of venter elongate. Seg- ment IX: pleuron produced into blunt pos- terior lobe; venter excised anteriorly and posteriorly. Segment X: dorsum split near base into sclerotized processes which con- verge at apices; ventrolateral processes reaching base of subgenital processes. In- ferior appendages: truncate, joined by nar- row bridge. Subgenital processes: conver- gent, arch-shaped in lateral view. Aedeagus (Fig. 10): elongate, bifurcate apically. 192 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON WwWwS'0 ap q 15. disjuncta ali glossocercus at 1 3 crinita 11° chiasma 20 ap 1 9 planichela Figs. 9-20. Male terminalia of Acritoptila Wells, aedeagus and ventral views. 4. amphapsis: 9, ventral; 10, aedeagus. 4. chiasma: 11, ventral; 12, aedeagus. A. crinita: 13, ventral; 14, aedeagus. 4. disjuncta: 15, ventral with inset of caudal view of inferior appendages; 16, aedeagus. 4. glossocercus: 17, ventral; 18, aedeagus. A. planichela: 19, ventral; 20, aedeagus. ap, apico-mesal process; bp, bilobed process; ej, ejaculatory duct; ia, inferior appendages; IAe, inferior appendage extensor muscle; sg, subgenital processes; vp, ventrolateral processes. Female.— Unknown. Etymology.— Named for the arch-shaped subgenital processes. Holotype.—Male. New Caledonia: Ho- nailu River, 26 Oct. 1958, C. R. Joyce. Paratypes.— None. Diagnosis.—The semi-fused inferior ap- pendages (Fig. 9) and relatively short ven- trolateral lobes of tergum X (Fig. 1) are more similar to male genitalia of the Australian Acritoptila than to the genitalia of other New Caledonian species. Acritoptila chiasma, New SPECIES Figss2,.3 sll 2 Male.—Antennae 34 segmented. Seg- ment VIII: ventral apico-mesal process elongate. Segment IX: pleuron with narrow sclerotized process on lateral margin. Ter- gum X: elaborately sculptured with mesal, blackened tips and elongate ventrolateral rod-like processes (Fig. 3). Inferior append- ages: completely fused; membranous caudal lobe with 2 short setae. Subgenital process- es: mesal margin bilobed, with inner lobe auriculate; single lateral seta; bilobed pro- cess with long lobes. Aedeagus: apex largely membranous with small apical point (Fig. 12)s Female. — Unknown. Etymology.—Named for the membra- nous mesal area dividing tergum X into right and left halves. VOLUME 91, NUMBER 2 Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same as holotype, 26 4. Pla- teau du Dogny, 20 Nov. 1958, | 2. Diagnosis.—This species is most closely related to the New Caledonian 4. glosso- cercus and A. crinita. Like those species, it bears a lateral process on pleuron IX. But, the tenth tergum is quite distinctive. Acritoptila crinita, NEw SPECIES Figs. 4, 13, 14 Male.—Segment VII: ventral apico-mesal process elongate (AP). Segment VIII: dor- sum with lateral fringe of elongate setae (not in illustration). Segment IX: lateral process short, broad, setose. Tergum X: largely membranous except for sclerotized lateral margin and caudolateral process. Inferior appendages: completely fused into rounded structure (ventral view); membranous cau- dal lobe with two short setae. Subgenital processes: mesally bifid and blackened; two lateral setae; rod-like sclerite connecting lat- eral lobes of subgenital processes to base of tergum X; bilobed process with short lobes covered with minute setae. Inferior ap- pendages: extensor muscle inserted on lat- eral lobe of subgenital processes. Aedeagus: apex with several sclerotized processes (Fig. 14). Female.— Unknown. Etymology.— Named for long fringe of se- tae on dorsum VIII. Holotype.— Male. New Caledonia: head- waters of Honailu River, 26 Oct. 1958, C. R. Joyce. Paratypes.— None. Diagnosis. — This species is a sister species of A. glossocercus, with nearly identical ven- trolateral rod-shaped processes on tergum X (Figs. 4, 7). It may be distinguished from A. glossocercus by the rounded fused infe- rior appendages (Fig. 13), the setose bilobed process (Fig. 13) and the non-dilated cau- dolateral processes of tergum X (Fig. 4). 193 Acritoptila disjuncta, NEw SPECIES Figs. 5, 6, 15, 16 Male.—Antennae 35 segmented. Seg- ment IX: pleuron acutely pointed; venter broadly excised. Tergum X: ventral pro- cesses broadened distally with narrow pointed lobe, asymmetrical; remainder of tergum membranous. Inferior appendages: fused and broadly oval in ventral view; membranous mesal lobe with 3 setae at tip and one short seta laterad of base of lobe on each side. Subgenital processes: mesally connected dorsad of inferior appendages and with convergent distal processes; paired caudolateral setae; apices directed ventrally; bilobed process short. Aedeagus: simple, with distal sinuate process. Female.— Unknown. Etymology.—Named for the appearance that tergum X 1s not joined to segment IX. Holotype.—Male. New Caledonia: mountain stream up Boulari River, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 5 2. New Caledonia, Plaine des Lacs area, 3 Nov. 1958, C. R. Joyce. Diagnosis.— This species is grouped with chiasma, crinita and glossocercus because the males bear an elongate ventrolateral process on tergum X. It differs because its processes are thicker and asymmetrical, and the inferior appendages are broad. Acritoptila glossocercus, NEw SPECIES Figs. 7, 17, 18 Male.—Segment VII: ventral apico-mesal process elongate. Segment IX: lateral pro- cess of pleuron attenuate, moderate in length. Tergum X: largely membranous, with curved lateral process; elongate ven- trolateral rod-shaped processes widened subapically. Inferior appendages: complete- ly fused; tongue-shaped in caudal view; membranous caudal lobe with two short se- tae. Subgenital processes: mesally heavily 194 sclerotized and bilobed; one lateral seta; bi- lobed process with short lobes. Aedeagus: single apical sclerotized process (Fig. 18). Female. — Unknown. Etymology.—Named for the tongue- shaped inferior appendages. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes. — None. Diagnosis. —A sister species of 4. crinita, with similar ventrolateral rod-shaped pro- cesses on tergum X (Figs. 4, 7), this species may be most easily recognized by the di- lated condition of these processes (Fig. 7) and the tongue-shaped inferior appendages (caudal view—Fig. 17) Acritoptila planichela, New SPECIES Figs. 8, 19, 20 Male.—Segment VII. Apico-mesal pro- cess of venter short, acutely pointed. Seg- ment IX: pleuron truncate; venter anterior- ly truncate. Tergum X: distally flared with ventral flat, heavily sclerotized, claw-shaped processes. Inferior appendages: fused into broad plate with pair of ventral protuber- ances and single spine at each lateral mar- gin. Sclerotized subgenital structure ventrad of aedeagus, with narrow sclerotized band connecting it to tergum X. Aedeagus: sim- ple, elongate, lacking titillator (Fig. 20). Female.— Unknown. Etymology.—Named for the flat claw of tergum X. Holotype.—Male. New Caledonia: mountain stream up Boulari River, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 2 é. Diagnosis. — This 1s the most divergent of the New Caledonian species of Acritoptila. The subgenital structures are difficult to homologize with the subgenital processes of other species of Acritoptila. It is the only species in the genus with a short apico-me- sal process of venter VII. Relationships with PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON other New Caledonian Acritoptila are un- clear. Genus Caledonotrichia Sykora, 1967 Type species: Caledonotrichia iiliesi Sykora, 1967. As noted by Wells (1983), males of Ca/e- donotrichia closely resemble those of May- denoptila Neboiss. Indeed, if only genitalic characters are analyzed, the two genera could be synonymous. Adult males of Caledono- trichia may be distinguished by the paired subgenital processes. In Maydenoptila, the subgenital processes are fused or absent. Caledonotrichia species have a transverse suture on the mesoscutellum, as do Maydenoptila species. In the generic de- scription given by Marshall (1979), the su- ture was described as absent in Caledono- trichia. Caledonotrichia charadra, New SPECIES Figs. 21-23 Male.—Segment IX: anterior and caudal margins of venter truncate; pleuron sinuate; dorsum a broad band. Subgenital processes: parallel, contiguous, caudally spatulate in ventral view and anteriorly divergent. Membranous sac within segment IX. Infe- rior appendages: bilobed and attached to dorsolateral margin of segment IX; upper lobe sickle-shaped with blackened, ventral- ly directed tooth on mesal margin; ventral lobe triangular in shape. Aedeagus: largely membranous with elongate sclerotized pro- cess originating at base of aedeagus (Fig. 23). Female.— Unknown. Etymology.— Named habitat of holotype in mountain stream. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 31 4. Diagnosis.—This species is closely relat- ed to C. minor Sykora, particularly in the shape of the inferior appendages and ae- VOLUME 91, NUMBER 2 Figs. 21-32. 22, ventral; 23, aedeagus. C. extensa: 24, lateral; 25, ventral; 26, aedeagus. P. dumagnes: 27, lateral; 28, ventral; 29, aedeagus. P nigrispina: 30, lateral; 31, ventral; 32, aedeagus. ap, apio-mesal process; as, aedeagal sheath; ej, ejaculatory duct; ia, inferior appendages; iae, inferior appendage extensor muscle; ial, inferior appendage lower lobe; iau, inferior appendage upper lobe; ms, membranous sac; sg, subgenital process; ti, titillator; vl, ventral lobe. deagus. It may be distinguished by the shape of the subgenital processes. Caledonotrichia extensa, NEw SPECIES Figs. 24-26 Male.—Antennae 27 segmented. Seg- ment IX: venter protruded anteriorly within segment VIII; caudal margin broadly ex- cised. Subgenital processes: elongate and narrow. Tergum X: elongate, tongue-shaped. Inferior appendages: bilobed; attached to dorsolateral margin of segment IX; dorsal lobe long, with ventrally directed tooth and + re] P. nigrispina Male terminalia of Caledonotrichia Sykora and Paroxyethira Mosely. C. charadra: 21, lateral; peg-like setae at base; ventral lobe broadly oval. Aedeagus: elongate; dorsal crease along most of length; membranous lobe extends distally beyond tip of ejaculatory duct. Female.— Unknown. Etymology.—Named for divergent con- dition of inferior appendages. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, | 4. Diagnosis. —C. extensa is distinctive, with the lobes of the inferior appendages longer than wide. It is likely a sister to the distinct 196 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON group composed of the other species in the genus. Genus Oxyethira Eaton, 1873 Subgenus Trichoglene (Neboiss, 1977) Type species: Oxyethira albiceps (Mac- Lachlan, 1862). Two species of Oxyethira from New Cal- edonia are placed in subgenus 7richoglene Neboiss along with the New Zealand O. al- biceps and the Australian brevis Wells, co- lumba (Neboiss), mienica Wells, triangu- lata Wells and retracta Wells. Males of the two new species share with these other species a plesiomorphic character; a com- plete, non-excised segment VIII, and two apomorphic characters; aedeagus with re- curved sub-distal spinous process and subgenital processes widely separated and partly fused with each pleuron of segment IX. Males of O. caledoniensis n. sp. are sim- ilar to O. brevis in having a short titillator, but seem to be more primitive than all Aus- tralian and New Zealand species in the re- tention of a sclerotized mesal connection between the subgenital processes. O. insu- laris n. sp. males are aberrant, with an en- larged forked structure, apparently the modified inferior appendages, on elongate venter IX. The forked structure is like that found in males of the subgenus Dactylotri- chia Kelley but is probably not homologous. The type species of 7richoglene was 1n- correctly identified as O. columba (Neboiss) in Kelley (1984). It should be O. albiceps (MacLachlan). A redescription of the sub- genus was given in Kelley (1984). Oxyethira caledoniensis, New SPECIES Figs. 33, 42, 56 Male.—Segment VIII: cylindrical, with- out excisions. Segment IX: dorsum narrow in lateral view; venter reaching anterior end of segment VIII. Inferior appendages: re- duced to small, widely separated bilobed areas of sclerotization. Subgenital process- es: widely separated, narrow, connected by thin mesal sclerotized strip; dorsolaterally fused to pleuron IX; bilobed processes short. Aedeagus: titillator short; recurved subdis- tal process. Female.— Unknown. Etymology.—Named for island where holotype was collected. Holotype. — Male. New Caledonia: Plum, 20-60 m, malaise trap, 23-25 Mar. 1968, J. L. Gressitt & T. C. Maa. Paratypes.— None. Oxyethira insularis, New SPECIES Figs. 34, 43, 57 Male.—Length 1.7 mm. Segment VIII: cylindrical, without excisions. Segment IX: dorsum broad; venter elongate, reaching an- terior end of segment VII; venter with large forked process, fused inferior appendages, subtending membranous, serrately tipped plate. Subgenital processes: acutely pointed, connected mesally by curved sclerotized band and connected dorsolaterally to pleu- ron IX; bilobed processes short. Aedeagus: ttillator lacking; sclerotized, recurved pro- cesses at apex. Female.— Unknown. Etymology.— Named for its island habi- tat. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, | é. Genus Oxyethira Eaton, 1873 Subgenus Pacificotrichia, n. subgen. Seven new species of Oxyethira from New Caledonia, Vanuatu and Fiji are arranged here in a new subgenus endemic to the South Pacific islands. Relationships with other subgenera are unclear, although Pacificotri- chia is most similar to Dampfitrichia Mose- ly. Similarities between the males include the shallowly excised venter VIII (Fig. 62), the deeply excised dorsum VIII (Fig. 37), fused R, and R, forewing veins, configura- VOLUME 91, NUMBER 2 eye) Vill caledoniensis 36 indorsennus 39 scutica ap Figs. 33-41. 197 a7 Se 35 insularis dorsennus 38 oropedion 41 fijiensis Male terminalia of Oxyethira Eaton, lateral views. 33, O. caledoniensis. 34, O. insularis. 35, O. dorsennus. 36, O. indorsennus. 37, O. melasma. 38, O. oropedion. 39, O. scutica. 40, O. efatensis. 41, O. fijiensis. ap, apico-mesal process; bp, bilobed process; ia, inferior appendages; sg, subgenital processes; sl, tapered seta. tion of the subgenital processes, and reduc- tion or loss of the pre-apical spur on the meso-tibia. The relatively simple male gen- italia, with a reduced ninth segment, are like the genitalia of the minima group of sub- genus Dampfitrichia, the only other known Australasian group of Oxyethira besides the endemic subgenus Trichoglene. But, there are important differences: unlike males of the minima group, the subgenital processes are distally fused (Fig. 56), do not bear distal setae (Fig. 63) and retain the bilobed pro- cess. The lack of significant shared apo- morphies prohibit the placement of these species in subgenus Dampfitrichia. Two distinct groups can be identified in subgenus Pacificotrichia. The oropedion group includes species O. dorsennus n. sp., indorsennus n. sp., melasma n. sp., orope- dion n. sp. and scutica n. sp. In males of these species, venter IX does not reach the anterior end of segment VIII (Fig. 35) and the fused subgenital processes bear a pro- nounced mesodistal projection (Fig. 58). Females are characterized by elongate apo- demes (Fig. 50) and cerci (Fig. 53), as well as by a sclerotized lateroventral projection on tergum VIII. The efatensis group in- cludes O. efatensis n. sp. and fijiensis n. sp. Males of this group are distinguished by venter IX which extends into segment VII (Fig. 40), subgenital processes which are not connected anteromesally (Fig. 63) and an aedeagus with a tapered dorsal sclerotized 198 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON 42 43 44 45 46 47 48 ® dorsennus SCcutica 22 scutica scutica Figs. 42-55. as o q S \ 50 T= 2 , oropedion Si = r 3 x c) a =) | 3 : ° eh 2 ® Ss a c o ® | = i=) o E 3 ® =} oO ® c = 8 al oO S {\ 2 \ L \ BY sacl oil oropedion sp efatensis Aedeagi and female terminalia of Oxyethira Eaton. Figs. 42-49, aedeagi. 42, O. caledoniensis. 43, O. insularis. 44, O. dorsennus. 45, O. melasma. 46, O. oropedion. 47, O. scutica. 48, O efatensis. 49, O fyiensis. Figs. 50-53, female terminalia. O. oropedion: 50, lateral; 51, ventral. O. scutica: 52, lateral; 53, ventral. O. efaensis: 54, lateral; 55, ventral. ap, apodeme IX; e}, ejaculatory duct; hl, horizontal lamella; sp, spermathecal process; ss, spermathecal sclerite; ti, titullator; tX, tergum X; vl, ventral lobe. process and a membranous ventral tube (Fig. 49). Females bear short apodemes and cerci (Fig. 54). Subgenus Pacificotrichia, n. subgen. Type species: Oxyethira oropedion, n. sp. Spur formula: 0-2-4 or 0-3-4. Forewings: R, and R, fused. Male.—Length 2.1-2.4 mm. Antennae 24-30 segmented; sparse placoid sensilla. Segment VIII: venter shallowly excised (Fig. 63); dorsum deeply excised. Segment IX: dorsum narrow (Fig. 40); posterolateral pro- cesses lacking; venter reaching to anterior end of segment VIII or posterior end of seg- ment VII (Figs. 37, 40). Inferior append- ages: small or lacking (Fig. 40); setal lobes usually indistinct. Subgenital processes: fused distally (Fig. 63); bilobed process short (Fig. 62). Aedeagus: titillator absent (Fig. 48). Female.—There is too much variability among females to describe subgeneric char- acters. Range. — New Caledonia; Vanuatu; Fi1. VOLUME 91, NUMBER 2 caledoniensis t 0.5mm 4 61 scutica oropedion Figs. 56-63. 57 insularis 199 Vill- J =; —bp 58 dorsennus 59 melasma 62 efatensis fijiensis Male terminalia of Oxyethira Eaton, ventral views. 56, O. caledoniensis. 57, O. insularis. 58, O. dorsennus. 59, O. melasma. 60, O. oropedion. 61, O. scutica. 62, O. efatensis. 63, O. fyiensis. bp, bilobed process; ia, inferior appendages: mr, mesal ridge; sg, subgenital processes; sl, setal lobe. oropedion Group Oxyethira dorsennus, New SPECIES Figs. 35, 44, 58 Spur formula: 0-2-4. Male.—Length 2.1 mm. Antennae 25 seg- mented. Segmented VII: ventral apico-me- sal process present. Segment VIII: dorsum with deep excision in lateral view. Inferior appendages: darkened, nearly contiguous mesally. Subgenital processes: triangular fused apex. Aedeagus: lacking distal pro- cesses. Female.— Unknown. Etymology.—Named for the hump on dorsum VIII. Holotype.— Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, | 2. New Caledonia, Plum, 20-60 m, 23-25 Mar. 1958, 1 3. Oxyethira indorsennus, NEw SPECIES Figs. 36, 44, 58 Spur formula: 0-3-4. Male.—Length 2.1 mm. Antennae: 25 segmented. Segment VII: ventral apico- mesal process present. Segment VIII: dor- sum gradually excised in lateral view. In- ferior appendages: darkened, nearly contig- uous mesally. Subgenital processes: triangular fused apex. Aedeagus: lacking distal processes. 200 Female. — Unknown. Etymology.— Named for absence of dor- sal hump on dorsum VIII. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 5 2. Diagnosis.—This species differs from O. dorsennus, n. sp. only in the shape of dor- sum VIII (Figs. 35, 36) and the spur for- mula. Oxyethira melasma, NEw SPECIES Figs. 37, 45, 59 Spur formula: 0-2-4. Male.—Antennae: 23-30 segmented. Segment VII: ventral apico-mesal process large. Segment VIII: venter with shallow, acutely pointed excision; setae stout with tapered curved apices. Segment IX: venter with mesal ridge, caudal end tapered to acutely pointed tip; dorsum membranous. Inferior appendages: indistinct; setal lobes lacking. Subgenital processes: with ventral black spot at fused apex; elongate bilobed process with mesally contiguous lobes. Ae- deagus: apical sclerotized process. Female.— Unknown. Etymology.—Named for the black spot on the subgenital processes. Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, | 4. New Caledonia, Honailu River headwaters, 26 Oct. 1958, 1 6. Oxyethira oropedion, New SPECIES Figs. 38, 46, 50, 51, 60 Spur formula: 0-3-4. Male.—Antennae: 26 segmented. Seg- ment VII: ventral apico-mesal process pres- ent. Segment VIII: venter not excised; dor- sum deeply excised. Inferior appendages: widely separated; setal lobes forming setose PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON membranous bridge between inferior ap- pendages. Aedeagus: without apical pro- cesses. Female.—Segment VIII: tergum and apo- demes elongate; sternum with ventral black spot. Spermathecal sclerite indistinct. Etymology.—Greek for plateau, the ho- lotype habitat. Holotype.— Male. New Caledonia: Pla- teau de Dogny, light trap, 20 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, | é. Diagnosis.—In ventral aspect, the males are distinguished by the combination of widely separated inferior appendages and broad subgenital processes. The females are easily distinguished by the black spot on sternum VIII. Oxyethira scutica, New SPECIES Figs. 39, 47, 52, 53, 61 Spur formula; 0-2-4. Male.—Antennae: 26 segmented. Seg- ment VII: ventral apico-mesal process pres- ent. Segment VIII: pleuron roundly trun- cate; venter moderately excised. Inferior appendages: minute contiguous lobes at caudal tip of venter IX; setal lobes lacking. Subgenital processes: with paied black spots bordering caudomesal protrusion; bilobed process with elongate lobes. Aedeagus: nar- row elongate process extending from apex. Female.— Antennae: 20 segmented. Seg- ment VIII: tergum and apodemes elongate. Spermathecal process distinct. Sclerotized sac present caudad of spermathecal process. Etymology.—Named for the elongate whip-like process of the aedeagus (Fig. 47). Holotype.—Male. New Caledonia: mountain stream up Boulari River, light trap, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 3 2. Same locality as holotype, 17 Nov. 1958, 17 6. Diagnosis. — Males of this species may be recognized by the minute inferior append- VOLUME 91, NUMBER 2 ages and triangular (ventral view) subgenital processes, in addition to the shape of the aedeagal process. Oxyethira scutica is the only species of subgenus Pacificotrichia with paired black teeth along the posterior fused margin of the subgenital processes. This character 1s common among species of other neotropical subgenera of Oxyethira. efatensis Group Oxyethira efatensis, NEw SPECIES Figs. 40, 48, 54, 55, 62 Spur formula: 0-3-4. Male. —Length 2.4 mm. Antennae 26 seg- mented. Segment VII: ventral apico-mesal process present. Segment VIII: venter and pleuron slightly excised. Inferior append- ages: triangular in shape, discrete. Aede- agus: with dorsal, sclerotized, attenuate band subtended by membranous tube (Fig. 48). Female.—Length 2.4 mm. Antennae 21 segmented. Segment VIII: tergum short with anterior margin heavily sclerotized. Seg- ment IX: lacking. Spermathecal process teardrop-shaped; horizontal lamella dis- tinctly sclerotized. Etymology.—Named for locality of col- lection. Holotype.— Male. Vanuatu: Efate (NW), Maat, Ambryn Village, 3M., 18 Aug. 1957, light trap, J. L. Cressitt. Paratypes.—Same data as holotype, 4 4. Same locality as holotype, 19 Aug. 1957, 5 6. Diagnosis.—This species may be distin- guished from O. fijiensis n. sp. by the excised pleuron VIII and the median ventral pro- jection of the fused subgenital processes. Oxyethira fijiensis, New SPECIES Figs. 41, 49, 63 Spur formula: 0-3-4. Male. —Segment VII: ventral apico-mesal process lacking. Segment VIII: venter with small mesal excision; pleuron truncate. Seg- 201 ment IX: pleuron with dorsolateral point on anterior margin; venter pointed anteriorly. Inferior appendages: small, indistinct, widely separated. Subgenital processes: fused cau- dal margin with mesal tooth. Aedeagus: one sclerotized blade-like process and one membranous process enclosing ejaculatory duct (Fig. 49). Female.— Unknown. Etymology.— Named for collection local- ity of holotype. Holotype.— Male. Fiji: Levu, Nandari- vatu, Jan. 1955, N. L. H. Krauss. Paratypes.— None. Diagnosis.—This species may be distin- guished by the distomesal tooth on the subgenital processes. Genus Paroxyethira Mosely, 1924 The two new species of Paroxyethira Mosely described here are the first known from this genus outside New Zealand. Both species are similar to previously described species of Paroxyethira in the morphology of the terminal abdominal segments as well as wing venation, spur formula and other non-genitalic characters. The genus is char- acterized by the elongate ventral process of segment VIII (Fig. 30) and the asymmetrical aedeagal sheath in the males (Fig. 30). A description of the genus was given by Mar- shall (1979). Paroxyethira dumagnes NEw SPECIES Figs. 27, 29, 31 Male.—Antennae 25 segmented. Seg- ment VII: ventral apico-mesal process pres- ent. Segment VIII: length short; venter with elongate spatulate process; pleuron with elongate setae. Segment IX: dorsum asym- metrical and tilted anteriorly; pleuron pro- truded caudally. Inferior appendages: elon- gate with numerous black setae on inner margin; extensor muscles originate on an- terior pleuron IX and insert on anterolateral corners of inferior appendages. Aedeagus: 202 sheath short, asymmetrical, with several short setae and a long anterior apodeme; one muscle originates caudally on venter IX and inserts on apex of apodeme, another inserts on caudal end of sheath; aedeagus elongate, with titillator and two distal lobes. Female. — Unknown. Etymology.— Named for the inferior ap- pendages, which have black setae which ap- pear like magnetic filings. Holotype.— Male. New Caledonia: Bou- lari River, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 2 2. Same locality as holotype, 17 Nov. 1958, 165 3. Diagnosis.—Paroxyethira dumagnes males have a short aedeagal sheath with an anterior apodeme (Fig. 27) as do males of P. tillyardi Mosely. However the apex of the aedeagal sheath is different from all other known species of the genus. Paroxyethira nigrispina, New SPECIES Figs. 30-32 Male.—Antennae 22 segmented. Seg- ment VII: short ventral apico-mesal process present. Segment VIII: length short; venter with elongate, flattened, gradually tapered process. Segment IX: dorsum excised to an- terior margin, asymmetrical; pleuron pro- truded caudally. Inferior appendages: short- er than segment IX. Aedeagus: sheath elongate, with small spine at midlength and thick spine at base of apical process; apical process sinuate and black at tip: aedeagus elongate with titillator and two distal lobes. Female. — Unknown. Etymology.— Named for black tip of ae- deagal sheath. Holotype.— Male. New Caledonia: Bou- lari River, 3 Nov. 1958, C. R. Joyce. Paratypes.—Same data as holotype, 86 ¢. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Same locality as holotype, 17 Nov. 1958, 9 8. Diagnosis.—Paroxyethira nigrispina males have an elongate aedeagal sheath sim- ilar to males of P. kimminsi Leader. The apex of the sheath however is distinct. ACKNOWLEDGMENTS I am grateful to the Bishop Museum in Hawaii and Alice Wells of the University of Adelaide for the loan of specimens. John Morse of Clemson University and Enwright Laboratories of Greenville, South Carolina were also supportive. This publication is Technical Contribution #2806 of the Clem- son University Agricultural Experiment Station. LITERATURE CITED Eaton, A. E. 1873. On the Hydroptilidae, a family of Trichoptera. Trans. Entomol. Soc. London 1873: 125-150. Kelley, R. W. 1984. Phylogeny, morphology and clas- sification of the micro-caddisfly genus Oxyethira Eaton (Trichoptera: Hydroptilidae). Trans. Amer. Entomol. Soc. 110: 435-463. Mac Lachlan, R. 1862. Characters of new species of exotic Trichoptera. Trans. Entomol. Soc. London (3)1: 301-311. Marshall, J. E. 1979. A review of the genera of the Hydroptilidae (Trichoptera). Bull. Br. Mus. (Nat. Hist.) Entomol. 39: 135-239. Mosely, M.E. 1924. New Zealand Hydroptilidae (Or- der Trichoptera). Trans. R. Soc. N.Z. 55: 670-673. Neboiss, A. 1977. A taxonomic and zoogeographic study of Tasmanian caddis-flies (Insecta: Trichop- tera). Mem. Natl. Mus. Victoria 38: 1-208. Sykora, J. 1967. Trichoptera collected by Prof. J. Illies in New Guinea and New Caledonia. Pac. Insects 9: 585-595. Wells, A. 1982. Tricholeiochiton Kloet & Hincks and new genera in the Australian Hydroptilidae (Tr- choptera). Aust. J. Zool. 30: 251-270. 1983. New species in the Australian Hydrop- tilidae (Trichoptera), with observations on rela- tionships and distributions. Aust. J. Zool. 31: 629- 649. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 203-205 WASP PARASITES OF THE BURDOCK SEED MOTH, METZNERIA LAPPELLA ZELLER (GELECHIIDAE): NEW HOST RECORD FOR BRACON MELLITOR SAY (HYMENOPTERA: BRACONIDAE) FOSTER FORBES PURRINGTON AND DEBORAH H. STINNER Department of Entomology, The Ohio State University, The Ohio Agricultural Research and Development Center, Wooster, Ohio 44691. Abstract. —Bracon mellitor parasitizes the burdock seed moth, Metzneria lappella. This is anew host record for this wasp, best known as a parasite of the boll weevil, Anthonomus grandis. Agathis malvacearum, another wasp parasite of this moth, is newly reported from the Pacific coast of Washington, a westward range extension of 2000 km. We also report a transcontinental North American distribution for the host—parasite association of M. lappella with Hyssopus thymus, the paramount Nearctic wasp parasite of the European pine shoot moth, Rhyacionia buoliana. Key Words: The burdock seed moth, Metzneria lap- pella Zeller (Lepidoptera: Gelechiidae) eats the seeds of burdock (the Palearctic genus, Arctium L. (Compositae)) by excavating in the indehiscent flower head, or bur. Natu- ralized in North America, this univoltine moth is found wherever common burdock, A. minus (Hill) Bernhardi, is established (Purrington 1970). Field collections of bur- dock burs made in the United States have yielded several wasp species that parasitize M. lappella \arvae (Juhala 1967, Purrington 1970, 1979, Purrington and Uleman 1972). In this paper, we report a new primary wasp parasite of this moth and record new dis- tributions for two other parasites. We obtained about 100 burs in Matta- poisett, MA (12-II-86), which yielded 250 M. lappella larvae and 7 overwintering pre- pupal larvae of Bracon mellitor Say (Hy- menoptera: Braconidae). This was the first record of B. mellitor as a parasite of M. lappella. Wasp larvae were enclosed in pa- Braconidae, burdock, distribution, Gelechiidae pery silken cocoons inside host excavations within the burs. Bracon mellitor is a solitary univoltine ectoparasite of several larval weevils and moths in the Nearctic region (Adams etal. 1969, Cross and Chesnut 1971, Krombein et al. 1979, Cuda and Burke 1983). Itis an important parasite of the boll weevil, Anthonomus grandis Boheman (Co- leoptera: Curculionidae) (Adams et al. 1969, Sturm and Sterling 1986, Pencoe and Phil- lips 1987). Agathis malvacearum Latreille (= A. metzneriae Muesebeck) (Hymenoptera: Braconidae) is an Old World solitary uni- voltine endoparasitic wasp introduced into the Western Hemisphere, probably with burdock containing M. lappella. In the Pale- arctic, 4. malvacearum is also found on the moths, /. carlinella Stainton (Gelechiidae), Evetria resinella L. (Gelechiidae), and Co- leophora otitae Zeller (Coleophoridae) (Shenefelt 1970). North American distri- bution records are Quebec, New England, 204 Long Island, and west to North Dakota (Ju- hala 1967, Krombein et al. 1979). We found A. malvacearum on M. lappella in burdock collected at Portland, OR (Reed College, 25- XII-85, col. C. B. Purrington), Bellfountain, OR (10-I-86), and Seattle, WA (Pike Place, 6-1-86). The new records extend the wasp’s known range by 2000 km westward to the Pacific coast. Evidently burdock has been present in western Washington since at least the 1930's (Gunther 1973). It arrived in the New World with European settlers in the same way that it has become ubiquitous: its burs are armed with hooked spiny bracts that engage fur and cloth. Although Fyles (1899) speculates that M. lappella arrived in Quebec from Eu- rope only late in the 19th century, burdock infested with M. lappella and its braconid parasite, A. malvacearum, likely became es- tablished in North America soon after the first settlements were made along the At- lantic coast. Gross et al. (1980) review the introduction and early distribution of bur- dock in North America. We recovered Hyssopus thymus Girault (Hymenoptera: Eulophidae), a multivol- tine, gregarious, larval ectoparasite from M. lappella in burrs obtained at Mattapoisett, MA (12-II-86), Wooster, OH (10-IV-86), and Kent, WA (2-I-86). These new collec- tion records establish a coast-to-coast North American distribution for the M. /appella— H. thymus host-parasite association, a re- lationship first reported from North Dakota (Purrington 1970). According to Syme (1974), H. thymus is the most common, widespread, and effective Nearctic parasite of the European pine shoot moth, Rhy- acionia buoliana (Schiffermiiller) (Tortrici- dae). We anticipate that wherever burdock be- comes established it will have arrived with some members of a small integrated insect community. We predict this community will include the Old World co-immigrants, !. PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON lappella and A. malvacearum, as well as eclectic elements from native fauna. ACKNOWLEDGMENTS We thank P. M. Marsh (USDA-Beltsville) for identifying A. malvacearum and B. mel- litor, and M. E. Schauff (USDA-Beltsville) for identifying H. thymus. C. B. Purrington supplied burdock burs from Portland, OR; G. Smith and S. Blue provided support fa- cilities in Seattle, WA. Salaries and research support provided by State and Federal Funds appropriated to The Ohio Agricul- tural Research and Development Center, The Ohio State University. Manuscript number / 10-88. LITERATURE CITED Adams, C. H., W. H. Cross, and H.C. Mitchell. 1969. Biology of Bracon mellitor, a parasite of the boll weevil. J. Econ. Entomol. 62: 889-896. Cross, W. H. and T. L. Chesnut. 1971. Arthropod parasites of the boll weevil, Anthonomus grandis: 1. An annotated list. Ann. Entomol. Soc. Am. 64: 516-526. Cuda, J. P. and H. R. Burke. 1983. Trichobaris brid- welli, a new host for Bracon mellitor. Southwest. Entomol. 8: 65-66. Fyles, T. W. 1899. Notes on the season of 1899. Ann. Rept. Entomol. Soc. Ontario 19: 105-106. Gross, R. S., P. A. Werner, and W. R. Hawthorn. 1980. The biology of Canadian weeds. 38. Arctium mi- nus (Hill) Bernh. and A. /appa L. Can. J. Plant Sci. 60: 621-634. Gunther, E. 1973. Ethnobotany of western Washing- ton. Univ. Washington Press, Seattle, 71 pp. Juhala, C. 1967. Notes on parasitic Hymenoptera associated with a gelechiid moth, Metzneria lap- pella, in the common burdock, and description of a new species of Agathis (Braconidae). Ann. Ento- mol. Soc. Am. 60: 95-97. Krombein, K. V., P. D. Hurd, Jr., D. R. Smith, and B. D. Burks. 1979. Catalog of Hymenoptera in America north of Mexico. Smithsonian Institution Press, Washington, DC, 2209 pp. Pencoe. N. L. and J. R. Phillips. 1987. The cotton boll weevil: Legend, myth, reality. J. Entomol. Sci. Suppl. 1: 30-51. Purrington, F. F. 1970 Ecology of Metzneria lappella (Lepidoptera: Gelechiidae) and its hymenopterous VOLUME 91, NUMBER 2 parasites in eastern North Dakota. Ann. Entomol. Soc. Am. 63: 942-945. 1979. Biology of the hyperparasitic wasp Perilampus similis (Hymenoptera: Perilampidae). Great Lakes Entomol. 12: 63-66. Purrington, F. F.and J.S. Uleman. 1972. Brood size of the parasitic wasp Hyssopus thymus (Hyme- noptera: Eulophidae): Functional correlation with the mass of a cryptic host. Ann. Entomol. Soc. Am. 65: 280-281. 205 Shenefelt, R. D. 1970. Braconidae 3: Agathidinae, pp. 342-343. /n Ferriere, C. and J. van der Vecht, eds., Hymenopterorum Catalogus, W. Junk, Dor- drecht. Sturm, M. M. and W. L. Sterling. 1986. Boll weevil mortality factors within flower buds of cotton. Bull. Entomol. Soc. Am. 32: 239-247. Syme, P. D. 1974. Observations of the fecundity of Hyssopus thymus (Hymenoptera: Eulophidae). Can. Entomol. 106: 1327-1332. PROC. ENTOMOL. SOC. WASH. 91(2), 1989, pp. 206-229 THE COCKROACH GENUS MARGATTEA SHELFORD, WITH A NEW SPECIES FROM THE KRAKATAU ISLANDS, AND REDESCRIPTIONS OF SEVERAL SPECIES FROM THE INDO-PACIFIC REGION (DICTYOPTERA: BLATTARIA: BLATTELLIDAE) Louis M. RoTH Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138. (Correspondence: Box 540, Sherborn, Massachusetts 01770.) Abstract. — Diagnostic characters for the cockroach genus Margattea Shelford are given. Theganosilpha Kumar and Princis 1s synonymized with Margattea. Margattea paracey- lanica, n. sp. from the Krakatau Islands is described. The following Wargattea species are redescribed: ceylanica (Saussure), nimbata (Shelford) (previously considered a synonym of ceylanica), perspicillaris (Karny), anceps (Krauss), contingens (Walker) [= humeralis (Walker)], and /ongealata (Brunner). A key to the above males is provided. Symploce obtusifrons (Walker) (sp. incertae sedis) is transferred to Margattea. The following species of Margattea are transferred to the genus Ba/ta Tepper: aurea Hanitsch, setifera (Hanitsch), parvula (Bolivar), crassivenosa (Bolivar), and /ongicercata (Bolivar). The position of the male’s genital hook and rotation of the ootheca in subfamily placement (Pseudophyllo- dromiinae or Blattellinae) is discussed. Key Words: Princis (1969: 862, 1971: 1162) listed 44 species of Margattea Shelford. It 1s a widely distributed genus with species occurring in Asia, Indonesia, Malaysia, South Pacific, and Africa. This paper is a result of my attempt to identify a few specimens of Mar- gattea collected in the Krakatau Islands and Australia. After reading the many species descriptions it was evident that the genus needed revision and that the various taxa should be redescribed with particular em- phasis on male genitalia. In the original de- scriptions of known species, 12 were based on both sexes, 17 on males, and 15 on fe- males only. Even when males were de- scribed, a number of characters which are known to be important today (e.g. type of front femur, whether or not the tarsal claws are toothed, whether or not the male has a Cockroaches, Blattellidae, taxonomy, Krakatau Islands tergal gland, and the shapes of the male gen- ital phallomeres were not mentioned). Iden- tification of females, when only that sex is known, is difficult unless the specimens are topotypic and the species have distinctive color patterns. Based on descriptions a number of species of Margattea belong in another genus. For example, I have exam- ined the types of the following species and they all belong in the genus Ba/ta Tepper: Margattea aurea Hanitsch (HDEO), M. se- tifera Hanitsch (NRSS), M. parvula Bolivar (BMNH), 7. crassivenosa Bolivar (BMNH), and M. longicercata Bolivar (BMNH). The Krakatau specimens were collected by personnel from institutions in Australia, Indonesia, and the United Kingdom during 1984 and 1985 expeditions (Thornton and Rosengren, in press). Thirteen species of VOLUME 91, NUMBER 2 cockroaches were taken and I will report on these elsewhere. Only one species of Mar- gattea was collected and it did not agree with the four species of the genus previously reported from these islands; I am describing it as new. I will present the diagnostic char- acters of the genus and redescribe the fol- lowing taxa that have been recorded from the Krakataus by others: M. ceylanica (Saussure), Mf. contingens (Walker) [= M. humeralis (Walker)], M. anceps (Krauss), and M. nimbata (Shelford). Kuchinga He- bard has been considered a junior synonym of Margattea, and I have concluded that Theganosilpha Kumar and Princis also is a junior synonym and therefore will rede- scribe their type species [Kuchinga longealata (Brunner) and Theganosilpha perspicillaris (Karny)]. The following museums and their cura- tors or collection managers loaned me spec- imens: (ANIC)—Australian National Insect Collection, CSIRO, Canberra, A:G@-r.. Australia: Dr. D.C.F. Rentz. (ANSP)—Academy of Natural Sciences of Philadelphia, PA, U.S.A.; Mr. Donald Azuma. (BMNH)-—British Museum (Natural His- tory), London, England; Mrs. Judith Marshall. (BPBM)—Bernice P. Bishop Museum, Honolulu, HI, U.S.A.; Mr. Gordon M. Nishida. (DEIG)—Deutsches Entomologisches Institut, Eberswald, Germany; Dr. G. Petersen. (HDEO)—Hope Department of Ento- mology, University Museum, Oxford, England; Dr. George C. McGavin. (MCZH)— Museum of Comparative Zo- ology, Harvard University, Cambridge, MA, U.S.A. (MNHG)— Museum d’ Histoire Naturelle, 207 Geneva, Switzerland; Dr. Bernd Hauser. (NRSS)—Naturhistoriska Riksmuseet, Stockholm, Sweden; Per Inge Persson. (RNHL)—Riksmuseum van Natuurlijke Historie, Leiden, The Neth- erlands; Dr. Jan van Tol. (TUVA)—La Trobe University, Bun- doora, Victoria, Australia; Mr. Patrick J. Vaughan and Prof. I.W.B. Thornton. (ZILS)—Zoological Institute, Lund, Sweden; Dr. R. Danielsson. Genus Margattea Shelford Margattea Shelford 1911:155. Type species, by monotypy: Blatta ceylanica Saussure, Rehn 1931: 302; Bey-Bienko 1938: 121; 1950: 145; Princis 1969: 862. Kuchinga Hebard 1929:39, 41. Type species, by selection: Phyllodromia longealata Brunner; Hanitsch 1931b: 392 (synony- mized with Margattea), Bey-Bienko 1938: 121 (synonymized with Margattea). Theganosilpha Kumar and Princis 1978: 33: Asahina 1979: 114, 119. Type species by monotypy: Theganopteryx perspicillaris Karny. New Synonymy. Diagnosis.—The following characteriza- tion is based on the species mentioned ear- lier as well as a study of more than 25 de- scribed and undescribed species which I plan to present elsewhere: Third and fourth max- illary palpal segments usually longer than the fifth (Fig. 40). Tegmina and wings fully developed or variably reduced, hind wings rarely absent; if reduced, hind wings are gen- erally shorter than the tegmina and may be vestigial: if fully developed, hind wings have the discoidal and median veins simple, sub- costa and most costal veins clubbed or thickened distad, cubitus vein essentially straight or weakly curved, with 1-4 com- plete (rarely O branches in reduced wings) and no incomplete branches, apical triangle small (Figs. 6, 13). Abdominal terga unspe- 208 cialized, or with a group of setae near the posteromedial margin of eighth tergum (Figs. 24, 28); this specialization may be hidden under the seventh tergum. Antero- ventral margin of front femur Type B, or B, (with | to 4 proximal spines) (Fig. 34) (both types may occur in the same species, dif- fering between the sexes), rarely Type C,; all 4 proximal tarsomeres with pulvilli, tar- sal claws symmetrical with inner margins generally denticulate (Figs. 7, 20, 32; the serrations often subobsolete and sometimes only seen at high magnification of a com- pound microscope), arolia present. Supra- anal plate usually relatively short, trans- verse, right and left male paraprocts similar, relatively simple plates (Figs. 8, 16). Male subgenital plate symmetrical or subsym- metrical styles simple, cylindrical, generally similar (Figs. 1, 29, 44); some species with a stylelike process at the base of each style giving the appearance of having 4 rather than 2 styles (Figs. 51, 69, 72). Male geni- talia with hooklike phallomere on the right side; median sclerite may have an associ- ated sclerite sometimes with setal brushes (Figs. 1, 26, 29, 69). Comments.—The principal diagnostic characters used by Kumar and Princis to distinguish Theganosilpha from Thegan- opteryx were hind wings reduced in both sexes, eighth tergum of male weakly spe- cialized, anteroventral margin of front fe- mur Type B, (some large proximal spines followed by a row of piliform spinules and terminating in 3 distal spines), inner mar- gins of tarsal claws finely denticulate. Asa- hina (1979: 119) accepted Theganosilpha as a valid genus and correctly pointed out that it was closer to Wargattea than to Thegan- opteryx. Rehn (1931) stated that the front femur of M. cevlanica was Type B, and that the tarsal claws were equal with internal mar- gins unspecialized (i.e. not denticulate). Bey- Bienko (1938: 121) believed that the tarsal claws of MW. cevlanica and M. longealata are equal and unspecialized, but later (Bey- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON Bienko 1958: 587) correctly stated that the former has barely visible notched tarsal claws. The claws of !. /ongealata also have subobsolete serrations. Asahina (1979: 119) pointed out that since some species of Var- gattea have toothed tarsal claws, this char- acter cannot be used to distinguish it from Theganosilpha. The size of the denticles may be so small that they cannot be seen at the magnifications usually used in examining pinned specimens with a dissecting micro- scope. If slides are prepared and the claws are examined at high magnification with a compound microscope, the minute teeth, or their remnants (Figs. 7, 20) can be seen. The male abdominal terga of 7. ceylan- icaand M. longealata are unspecialized. The male of Theganosilpha perspicillaris has a setal tuft on the eighth tergum. There are many examples of genera in which the males of a genus have or lack abdominal tergal glands (used in courtship behavior), and in those species that have them the specializa- tions may be found on one or more seg- ments. For example, species of Blattella may have glands on T7 only, or on T7 and T8, and the morphology of the glands may differ considerably between different species (Roth 1985a). In species of Parcoblatta Hebard, the males have glands on T1 only (6 spp.), Tl and T2 (4 spp.), or they are absent (2 spp.) (Hebard 1917). Asahina (1979: 119) separated Thegano- silpha from Margattea as follows: In The- ganosilpha the front femur is Type B,, the tegmina are narrowed apically, somewhat nail shaped, rather coriaceous, and the hind wings are rudimentary but sometimes one half the length of the tegmina. In Vargattea the front femur is Type B,, the tegmina are not narrowed apically or nail shaped, and it and the hind wing are fully developed. Hebard (1929: 4) pointed out that wing re- duction often is an unreliable generic char- acter (unless correlated with other more sta- ble features), and the armament on the front femur also is subject to individual variation. Kumar and Princis (1978: 33) described the VOLUME 91, NUMBER 2 hind wings in both sexes of 7. perspicillaris as reduced to small narrow lappets about half the length of the tegmina. The male lectotype and two male paralectotypes of this species actually have vestigial wings distinctly less than half the length of the tegmina. The hind wings of a male 7. per- spicillaris (fig. 25 in Asahina 1979) are more than half the length of the tegmina and their veins are well developed. It is possible for a genus to have more than one type of front femur. Species of Balta may have Type B or C (Hebard 1943: 39). Bey-Bienko (1938: 121) claimed that Margattea inermis Bey-Bienko is aberrant in having a Type C, rather than Type B femur. Margattea hemiptera Bey-Bienko (Bey-Bienko 1958) and Margattea elongata Kumar (Kumar 1975: 105) have reduced tegmina and wings and Type B, front fe- murs. I have examined specimens of The- ganosilpha ogatai Asahina and Thegano- silpha satsumana Asahina (identified by Asahina) and in both species the males have Type B, and the females Type B, femurs. There is no correlation between Type B, femurs and wing reduction, and Type B, femurs with fully developed organs of flight. The difference of 1 distal spine on the anteroventral margin of the front femur (Type B, vs. Type B,), the variable reduc- tion of tegmina and wings, and the presence or absence of a male tergal gland cannot be used to distinguish between Margattea and Theganosilpha, and Iam synonymizing the latter genus. The diagnostic characters for the genus Margattina Bey-Bienko are so similar to those of Margattea (Bey-Bienko 1958: 587) that I believe the former genus (known only from the type species Margat- tina trispina Bey-Bienko, a single male from China) should be reevaluated. SUBFAMILY PLACEMENT OF MARGATTEA The genitalia of male cockroaches are strongly asymmetrical and one of the more conspicuous structures is a hooklike phal- lomere (Fig. 29) although in some species 209 (Blaberidae) it may be greatly reduced. Ac- cording to McKittrick (1964) in the Blat- tellidae the hook is on the left side in the Blattellinae, Ectobiinae, and Nyctiborinae, or on the right in the Plectopterinae (= Pseu- dophyllodromuinae in Vickery and Kevan’s 1983, classification). McKittrick also claimed that the females of Blattellinae, Ec- tobiinae, and Nyctiborinae rotate their oothecae during oviposition, whereas the Plectopterinae do not. This concept holds for a large number of species but Brown (1975) found that Ectobius panzeri Ste- phens has the male genital hook on the right side although the female rotates its ootheca (Brown 1973). Bohn (1987) found that the male of Phyllodromica (Lobolampra) mon- tana Chopard and several other species of the ““montana-group” of Phyllodromica (Ectobiinae) show a similar reversal of the right-left asymmetry in male genitalia and have the hook on the right whereas some other species of the genus have the hook on the left. Although Bohn did not mention oviposition behavior of P. montana in his paper, he (personal communication) was “convinced” that it rotates its ootheca and that it and Ectobius panzeri are true Ecto- biinae, all of which rotate their oothecae. In Margattea the male’s genital hook is on the right side which in McKittrick’s sys- tem places it in the Plectopterinae. How- ever, 2 females of Margattea nimbata were carrying oothecae in the rotated position (Fig. 25) a behavior which is characteristic of the Blattellinae. As Bohn (1987: 303) pointed out, it is not sufficient to use the position of the genital hook in placing blat- tellids in their respective subfamilies. He also claimed that symmetry reversal has oc- curred several times independently during the evolution of the Blattaria and that it is possible there are species of Blattellinae with reversed symmetry as those found in the Ectobiinae. Bohn concluded that other characteristics such as oothecal rotation, and especially structure of the phallomeric scle- rites be considered. 210 Since the position of the male’s genital hook cannot always be used to distinguish Plectopterinae from Blattellinae, oothecal rotation or its absence appears to be the best discriminating character, and I used it to place a number of Blattellidae into these 2 subfamilies (Roth 1968a, 1971: 134). Thus M. nimbata, which rotates its ootheca, be- longs in the Blattellinae even though the genital hook is on the right side. But the problem here is more complex. The ovi- position behavior of the species other than nimbata is not known. Three females of Margattea longealata (Brunner) were car- rying oothecae whose keels were directed dorsad (i.e. non-rotated) (Fig. 62). It is pos- sible that these specimens were collected be- fore they could rotate their oothecae prior to depositing them. However, the ootheca is distinctly different in morphology from that of nimbata. It is particularly unusual in being wider than high and strongly re- sembles the oothecae of Lophoblatta brevis Rehn and Lophoblatta arlei Albuquerque from South America (see figs. 1-6 77 Roth 1968b). These are the only known members of the Plectopterinae that carry their oothe- cae with their keels dorsad (i.e. non-rotated) until the eggs hatch; other species in this subfamily drop the non-rotated oothecae shortly after it is formed. One ootheca of M. longealata (KOH treated) lacked cal- cium oxalate crystals. The oothecae of L. brevis and L. arlei have relatively few cal- cium oxalate crystals, but egg cases 1n other species in the Plectopterinae have large amounts of calcium oxalate crystals. The amount of calcium oxalate in cock- roach oothecae decreased during the evo- lution of ovoviviparity and viviparity in the Blaberoidea. Oothecae of the Blattellinae usually have large amounts of calclum ox- alate crystals but the amount of this com- pound varies from dense to sparse to absent in species of Blattella, all of which carry their oothecae externally in the rotated po- sition until the eggs hatch (Roth 1968a). Species of the African blattellid genus Stay- PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON ella Roth (Roth 1984) have an ootheca that is typical of species of Blattella, but it is retracted and carried internally until the eggs hatch (Roth 1982). The egg cases of Stayvella lack calcium oxalate as do species of Bla- beridae, all of which incubate their eggs in- ternally. The oothecae ot /. /ongealata are some- what transparent in the specimens studied and their eggs are undeveloped (the eggs from one of the oothecae were removed and cleared and showed no embryonic devel- opment). It is possible that (7. longealata carries its ootheca in the vertical position for the entire gestation period, as in Lo- phoblatta, but the 3 females reported here were collected shortly after their oothecae were formed and therefore their eggs show no development. It is probable that Mar- gattea contingens (Walker) has an ootheca similar to that of /ongealata because the male and female genitalia of these 2 taxa are so much alike (Figs. 51, 57-59, 61, 63, 68-70). The male genitalia (particularly the left and right phallomeres) of M. /ongealata and M. contingens differ considerably from those of other species in the genus. Perhaps these two species should be placed in a separate genus; if this is done Hebard’s Kuchinga (type species, /ongealata) would be resur- rected. For the present I am placing them in a species group of Margattea. SPECIES GROUPS OF .WARGATTEA I have examined about 25 described species of Margattea and about 10 taxa that apparently are new. Thus far my study in- dicates that the genus can be arranged in seven species groups. The seven species dis- cussed in this paper belong to 3 groups as follows (other species in the groups will be presented elsewhere): Ceylanica species group.—Eighth ab- dominal tergum unspecialized. Accessory stylelike structures near the base of the styles absent (Fig. 1). Tegmina and wings fully de- veloped. Front femur Type B;. Species: cey- lanica. VOLUME 91, NUMBER 2 Perspicillaris species group.—Eighth ab- dominal tergum specialized (Figs. 9, 24, 28, 43). Interstylar region may be simply a con- tinuation of the lateral margins of the plate (Figs. 23, 44), or there is a small plate some- times bearing small dark spines (Fig. 29). Accessory stylelike structures absent (Figs. 18, 29). Tegmina and wings fully developed or variably reduced, the wings sometimes vestigial. Front femur Type B, or B,. Species: perspicillaris, nimbata, paraceylanica, an- ceps. This is the largest group and contains at least an additional twelve species. Contingens species group.—Eighth ab- dominal tergum unspecialized. Lateral cor- ners of subgenital plate produced having the appearance of styles (accessory stylelike structures; Figs. 55, 56). Male genital phal- lomeres: hook portion of right phallomere, slender, elongate, strongly curved; median phallomere needlelike, accessory median phallomere absent; left phallomere roundly bulbous basally, relatively simple (Figs. 51, 57-59) (cp. the differences in the taxa belonging to the other 2 species groups). Tegmina and wings fully developed. Front femur Type B,. Species: contingens, lon- gealata. Key TO SOME MALES OF INDo-PAcIFIC Ma4RGATTEA* |. Eighth abdominal segment without a tergal gland 2 — Eighth abdominal segment with a tergal gland 2. Subgenital plate with lateral corners produced giving the appearance of accessory stylelike structures (Figs. 51, 55, 56, 69) 3} — Lateral corners of subgenital plate rounded, not stylelike (Fig. 1) ceylanica 3. Head brown with a yellowish orange band be- tween antennal sockets (Fig. 65). Pronotal disk with a broad dark brown macula (Fig. 64) longealata * Includes only males discussed in this paper. I have seen specimens only of nimbata and paraceylanica from the Krakataus. The earlier records of ceylanica, con- tingens, and anceps from these islands are question- able. — Head with an orangish macula on vertex be- tween the antennal sockets blending into the rest of the lighter face (Fig. 47). Pronotal disk light brown without distinct markings (Fig. 53) Fsthcky