Table Of Content

Papers Scientific Museum Natural History The University of Kansas 11 July 2002 Number 26:1-47 The Bee Genus Chilicola in the Tropical Andes, with Observations on Nesting Biology and a Phylogenetic Analysis of the Subgenera (Hymenoptera: Colletidae, Xeromelissinae)^ MCZ LIBRARY By JUL 2 3 2008 Charles D. Michener HARVARD Division ofEntomologi/, Nniural Histon/Museum, and Department ofEcologyand Evolutionan) Bi'Slbg}/, The Uiiiivrsiti/ofKansas, Lawrence, Kansas 66045 CONTENTS ABSTRACT 2 RESUMEN 2 INTRODUCTION 3 Acknowledgments 3 ABBREVIATIONS, TERMS, AND METHODS 4 SUBGENERAOFCH/L/COL/l 5 Key totheSubgenera of Ch/l;col4 Found in theAndesNorth ofChile 8 THE GROUPOF CHILICOLA {ANOEDISCELIS) ASHMEADI 8 KeytoMalesofthe Chilicola (Anoedicelis)ashmeadiGroup 11 Chilicola (Anoediscelis) ashmeadiCrawford 11 Ch/l;col4 (Anoediscelis) colombiana newspecies 13 Chilicola (Anoediscelis)misticanewspecies 13 Chilicola (Anoediscelis) venezuelana newspecies 13 Ch/l;col4 (Anoediscelis) wygodzinskyinewspecies 14 Chilicola (Anoediscelis) xanthostoma newspecies 14 Chilicola (Anoediscelis) xanthognatha newspecies 16 Chilicola (Anoediscelis) cooi'eri newspecies 16 Chilicola (Anoediscelis) pedunculata new species 16 'Contribution Number3230 from the Division ofEntomology, Natural History Museum,The University ofKansas. Lawrence, Kansas66045, USA ©NaturalHistoryMuseum,TheUniversityofKansas ISSN No. 1094-0782 twL»^,.„,^ Ernst Wu^|H*^. .t r*- - ,.; L'''- 1 ,^:- Papers Scientific Museum Natural History The University of Kansas 11 July 2002 Number 26:1-47 The Bee Genus Chilicola in the Tropical Andes, with Observations on Nesting Biology and a Phylogenetic Analysis of the Subgenera (Hymenoptera: Colletidae, Xeromelissinae)^ By JUL 2 3 2008 Charles D. Michener HARVARD Division ofEuiotnolo^y, Natural History Museum, and Department ofEcologyand EvolutioiumiBiolbgn/, ' Tlie University ofKansas, Laivrence, Kansas 66045 CONTENTS ABSTRACT 2 RESUMEN 2 INTRODUCTION 3 Acknowledgments 3 ABBREVIATIONS, TERMS,AND METHODS 4 SUBGENERAOFCH/L/COM 5 Key tothe Subgenera of Chilicola Found in theAndesNorth ofChile 8 THE GROUPOF CHILICOLA (ANOEDISCELIS) ASHMEADI 8 KeytoMalesofthe Chilicola (Anoedicelis)ashmeadiGroup 11 Chilicola (Anoediscelis) ashmeadiCrawford 11 Chilicola (Anoediscelis) colombiana new species 13 Chilicola (Anoediscelis)mistica newspecies 13 Chilicola (Anoediscelis) venezuelana new species 13 Chilicola (Anoediscelis) wycodzinskyinewspecies 14 Chilicola (Anoediscelis) xanthostoma newspecies 14 Ch;ucol4 (Anoediscelis) xanthognatha newspecies 16 Chilicola (Anoediscelis) cooperinew species 16 Chilicola (Anoediscelis) pedunculata new species 16 'Contribution Number 3230 from the Division ofEntomology. Natural Historv Museum. The University ofKansas, Laurence. Kan.sas66045. USA ©Natural HistoryMuseum,TheUniversityofKan.sas ISSNNo. 1094-0782 2 Scientific Papers, Natural History Museum, The University of Kansas SUBGENUSHYLAEOSOMA ASHMEAD 17 KeytotheAndean Speciesoe Hylaesoma 18 Chilicola {Hylaesoma) aequatoriensis Benoist 18 Chiucola (Hylaesoma) involuta newspecies 20 Chilicola (Hylaesoma) umbonata newspecies 21 Chilicola (Hylaesoma) caneinewspecies 21 Chilicola (Hylaesoma) bellinewspecies 23 Chilicola (Hylaesoma) smithpardoi newspecies 24 Noteon ExtralimitalHylaesoma 26 OREOD/SCEL/S NEW SUBGENUS 26 Illustrationsof Oreodiscelis 28 Keytothe Speciesofthe Subgenus Oreod/scel/s 28 Chilicola (Oreodiscelis) espeleticola newspecies 29 NestsofChilicola espeleticola 31 Chilicola (Oreodiscelis) brooksinewspecies 32 CH/L/COL4 (Oreodiscelis) benoistiana newspecies 32 Chilicola (Oreodiscelis) quitensis Benoist 35 CH/i.;coL4 (Oreodiscelis) cuzcoens/s newspecies 36 Ch;l;col4 (Oreodiscelis) bigibbosa new species 36 Chilicola (Oreodiscelis) maculipes newspecies 39 Chilicola (Oreodiscelis) styliventris (Friese) 40 Nestsof Chilicola styliventris 41 Chilicola (Oreodiscelis) transversaria newspecies 41 Chilicola (Oreodiscelis) gibbosa newspecies 42 Chilicola (Oreodiscelis) brzoskai newspecies 43 Chilicola (Oreodiscelis) simplexnewspecies 45 PhytogenyoftheSpeciesof the Subgenus Oreodiscelis 45 LITERATURE CITED 47 ABSTRACT This is a revision ofthe species ofthegenus Chilicola found in theSouthAmericanAndes, Peru toVenezuela,aboveelevationsof1000m. Inadditiontothefourpreviouslydescribed species known from this area,23newspeciesare recognized and described. They fall into threegroups: (1)TheC. asliinciidigroup ofthe subgenus Anoediscelis, for C. ashmcadi (Crawford) and the following new species: C. cokvnbiana, mistica, veneziielana, un/godziiiskyi, xantliostonia, xaiitho^)wtha, coopcri, pediinctilntn; (2) Subgenus Hylncosoiiia for C. aequatoriensis Benoistand the following new species: C. involuta, umbonata, canei, belli, siuithpanioi; (3) The new subgenus Oroediscelis (type species, Chilicola styliventris) for C. styliventris Friese, quitensis Benoist, and the following new species: C. espeleticola, brooksi, benoistiana, ciizcoensis, bigibbosa, maculipes, transversaria, gibbosa, brzoskai, simplex. Phylogenetic analyses are presented for the subgenera of Chilicola and for the species of the subgenus Oroediscelis. Aspects of nesting biology are presented for C. espeleticola and C. styliventris. Key Words: Chilicola, Anoediscelis, Hylaeosoina, Oreodiscelis, Colletidae, nests. South America, Andes. RESUMEN El presente trabajo es una revision de las especies del genero Chilicola que se encuentran en los AndesSuramericanos desde Peru hasta Venezuela porencima de los 1000 msnm. Adicionalmentea la especie deMesoamerica y lasquatroespeciesconocidaspara elarea,seestablecen y describen 23especiesnuevas. Las especies se distribuyen en tres grupos: (1) el grupo ashmeadi del subgenero Anoediscelis, para C. ashmeadi (Crawford) y las siguientes especies nuevas: C. colombiana, mistica, venezuelana, wygodzinskyi, xantliostoma, xanthognatha,cooperi,pedunculata;(2)subgeneroHylaeosomaparaC.aequatoriensisBenoistylassiguientesespecies nuevas: C. involuta, umbonata,canei, belli,smithpardoi; (3) el nuevo subgenero Oroediscelis (especie tipo, Chilicola styliventris)paraC.styliventrisFriese,C.quitensisBenoist,y lassiguientesespeciesnuevas:C.espeleticola,brooksi, benoistiana, cuzcoensis, bigibbosa, maculipes, transversaria, gibbosa, brzoskai, simplex. Se presentan analisis filogeneticospara lossubgenerosdeChilicolaypara lasespeciesdelsubgeneroOroediscelis,ademasdealgunos aspectos de los habitos de nidificacion para C. espeleticola y C. styliventris. Palabras claves: Chilicola, Anoediscelis, Hylaeosoma, Oroediscelis, Colletidae, nidos, Suramerica, Andes. Bee Genus Chiucola in the TropicalAndes INTRODUCTION The genus Chilicola consists of small, slender, largely blackbees (Fig. 1) ranging from Santa Cruz Province, Ar- gentina,andAsien,Chile,northtothestatesofTamaulipas andJaliscoinMexicoandSt. Vincentin theLesserAntilles. In much of this wide range, species of Chilicola are few andspecimensrarelycollected. Theliteratureshowsthem tobeabundantanddiverseonlyinChile,whichhasafauna of32knownspecies,placed insevensubgenera (Toroand Moldenke, 1979;Toro, 1986). Michener(1995)reduced the number ofChilean subgenera to four, but the diversity is clearly substantial. Onlythreespeciesfrom theAndesnorthofChilehave been described (Benoist, 1942, two species from Ecuador; Friese,1908,onespeciesfromPeru);afourthspeciesfound inColombia wasdescribed fromCentralAmerica. Itnow appears, however, that there is a rich fauna in the Andes ofPeru,Ecuador,Colombia,andVenezuela,i.e.,inthePima Fig. 1. Oiilicohi(Orocdiscclis)espdcticolaMichener,male(photo and Paramo Nordandino and adjacent provinces recog- by M. S. Engel). nized by Morrone (2001). The presentpaperconcerns the 27 species known from these mountains. Because these Anocdiscelis in Chile; the second ranges from the Andean beesaresmall, because there are relatively few beecollec- region and Brazil and Bolivia, north to Mexico; the third tors, and above all because Andean weather is so often group, Owediscelis, is limited to the Andean region north notfavorableforflightbyinsectslikebeesthatareusually ofChile. Thus, exceptpossibly for the C. ashmeadigroup, in their nests when the weather is cold, windy, cloudy, theAndean fauna does notconsist ofa northern montane foggy, and rainy, collecting with a net from flowers is of- extension of the rich, temperate zoneChilean andArgen- tennota successfulwayofsampling theAndean Chilicola tine fauna. At least the Andean Hylaeosoma presumably fauna. Many more species will no doubt be found, for were derived from lowland tropical antecedents. several ofthose described herein are known from a single Hi/lacosouia exists in the lowland tropics and was present locality and one or a few specimens. Benoist (1942) ap- intheDominicanRepublicanddoubtlesselsewhereinthe pears tohavebeen the firstperson to realize that they can tropics at least as early as Miocene times; Chilicola be obtained from pithy stems in Ecuador (regardless of {Hylaeosoiua) gracilis Michener and Poinar (1997) and C. the weather), and Robert W. Brooks of the University of (H.) elcctwdoniinica Engel (1999) were based on fossils in Kansas recognized the same thing in the Venezuelan Dominican amber. The subgenus does not occur in tem- Andes, where he collected many specimens used in this perate austral South America; its southernmost known study. This collecting method should greatly increase localitiesarerecordedbelowinanoteonextralimital(non- knowledgeofthisgenusinthefuture,although somespe- Andean) species at the end of the account of species of cies may nest principally orexclusively in burrows made Hi/laeosoiiia. The origin of the Andean endemic subgenus by beetles in wood or woody stems, rather than in pithy Owediscelis is notknown. stems. Traps thatcan function whenever there isa period Acknowledgements ofsunny weather also would be worth trying. Allspecimensusedinthepresentpaperwereobtained I am especially indebted toDr RobertW. Brookswho at or above the elevation of 1,000 m. The only lowland collected many of the specimens ofAndean Chilicola and species known from Peru to Venezuela are Chilicola who assembled much oftheborrowed material forstudy. (Hylaeosoma)stciwcephalaBrooksand MichenerfromAma- Healsoprovided thenestsand informationonnestingbi- zonian Colombia and an undescribed speciesprobably of ology oftwospecies. Itwasouroriginal intenttopublish the C. ashiiicadi group (males not known) from Talara in this study jointly, but pressure of other responsibilities northern coastal Peru. made his furthercontributions impossible. Andean Chilicola can be easily divided into three Dr. Michael S. Engel made the photographic illustra- groups: (1) the C. ashiiieadi group usually included in the tions and the analyses of my character matrices with subgenus Anocdiscelis, (2) the subgenus Hylacosoina, and NONA. (3) the new subgenus Owediscelis. The firstgroup ranges The cooperation of many museum curators and col- from the Andean region to Mexico, with other species of lectorswho loaned specimensis muchappreciated. They Scientific Papers. Natural History Museum, The Univershv oi Kansas arelistedinthesectiononAbbre\iations.Termsand Meth- Finally, Dr. J. S. Ashe, Curator-in-charge of the Ento- ods. Roy S. Snelling and Bryan N. Danforth were espe- mology Division, University of Kansas Natural History cially helpful in arranging for the disposition of impor- Museum and BiodiversityResearchCenter, kindl\-permit- tant material. ted my use of space and facilities. ABBREVIATIONS, TERMS, AND METHODS — T. tergum, followed by the number of a metasomal Davis = Department of Entomology, University of Cali- tergum—; thus Tl is the first tergum of the metasoma. fornia, Davis, California 95616, U.S.A. [L. Kimsey]. S. sternu—m (asexplained forT). Ithaca = Department of Entomology, Cornell University, Alveolus . antennal socket. — Ithaca, New York 14853, U.S.A. [E. R. Hoebeke]. Distal stigmal perpendicular. an imaginary line LaMolina = MuseodeEntomologia, UniversidadNacional acrossthefrontwing(Fig.3)tliroughtheapexofthestigma Agraria La Molina,nearLima,Peru [ClorindaVergara, and at a right angle to thecostal margin ofthe wing. The C. Rasmussen]. descriptionsbelow record where this line leaves the mar- Lawrence =Entomology Division and Snow Entomology ginalcell,inrelationtothesecondsubmarginalcell,i.e.,in Collection,Unix'ersityofKansasNaturalHistoryMu- relation to the anterior end of the first or of the second seum, Lawrence, Kansas 66045, U.S.A. submarginal crossvein. Lima = Museode Historia Natural, UniversidadNacional Measurements and illustrations of the hind tibia of Mayor de San Marcos, Lima, Peru [G. Lamas, C. males of the subgenus Oroediscelis appear repeatedly in Rasmussen]. the following pages. The measurements were made on Logan = Bee Biology and Systematics Laboratory, Utah the upper outer view with the preapical brush of hairs State University, Logan, Utah 84322, U.S.A. [T. L. characteristicofmostspeciesofthissubgenusinfullview, Griswold]. the crests on the under surface being out of sight. For il- London=TheNaturalHistoryMuseum, LondonSW75BD, lustrations, however, the tibia, is tilted to show the sum- England, U.K. ]C. Taylor]. mit of the innercrest visiblebehind the outer crest. Los Angeles = Natural History Museum of Los Angeles ForadditionalcommentsontheillustrationsofOroediscelis, County, Los Angeles, California 90007, U.S.A. [R. R. see the accouiitofthat subgenus. Snelling]. Facial proportions were obtained by dividing the Lyme Regis = M. Cooper collection, L\-me Regis, Dorset, length(apexofclypeus tosummitofvertex in facial view) England, U.K. by the width (greatest distance between apparent outer Maracay = Museo del Instituto de Zoologia Agricola, limitsofeyes in facial view), usinganeye-piece microme- Universidad Central de Venezuela, Maracay 2101A, ter in a dissecting microscope. Venezuela ]]. L. Garcia, L. J. jolly]. Forewing length was measured with thesameequip- Medellin=MuseoEntomologico"FranciscoLuisGallego," ment from the base of the enlarged base of vein R to the Unixersidad NacionaldeColombia,Medellin,Colom- wing apex. bia jGilberto Morales-Soto]. For all illustrations divided by a vertical line, the left New York = American Museum of Natural History, New hand side is dorsal, the right hand side ventral. York, New York 10024, U.S.A. G. Rozen]. [J. Following is a list of collections and museums con- Ottawa =CanadianNationalCollectioninCentreforLand taining material that I used, the cities by which they are and Biological Resources,Ottawa,Ontario,Canada JL. identified (in italics) in thetext,and inbrackets, thenames Dumochel]. of the persons who facilitated my use of the specimens. Paris= Mu.seum National d'HistoireNaturelle,75005Paris, Aarhus = Claus Rasmussen collection, to be placed in France [J. Casevitz-Weulersse]. Aarhus, Denmark. PCAM-Austin = Programa Cooperati\'o sohre la Berkeley = Essig Entomological Museum, University of Apifaima Mexicana, specimens at Central Texas California, Berkeley, California 94720, U.S.A. [Cheryl Melittological Institute, .Austin, Texas78731, U.S.A. Barr]. [J. L. Neff]. Berlin = Museum fiir Naturkunde der Humboldt- Sao Paulo = Museu de Zoologia, Unixersidaele de Sao Universilat, Berlin, Germany [F. Koch]. Paulo, Sao Paulo, Brazil [R. F. Brandao]. Bogota Humboldt = Instituto Alexander von Humboldt, San Francisco = California Academy of Sciences, Golden Santa Fe de Bogota 2 DC, Colombia [F. Fernandez]. GatePark,San Francisco94118,U.S.A. [VV.J. Pulawski]. Bogota Univ. = Departamento de Biologia, Universidad Washington = National Museum of Natural History, Nacional deColombia, Santa Fede Bogota,Colombia Smithsonian Institution, Washington, D. C. 20560, (G. Nates Parra, and V.H. Gonzalez-B.j. U.S.A. [R.J. McGinley]. . Bee Genus Chiucola in the TropicalAndes In listing collecting localities, I have given provinces important, because of occasional duplication of place ordepartments in italics. When these were not on labels, names within a country. Bracketed names are my inter- thesenamesareinbrackets. Thisissometimespotentially pretations, and therefore possibly incorrect. SUBGENERAOFCHILICOLA Michener(1995,2000)divitiedtheXeromelissinaeinto Table2. Charactersand theirstatesforsubgenericanalysis. two tribes, Xeromelissini and Chilicolini, the last for two genera, Chilicola and Xeiiochilicola. Michener and Rozen 1 innerorbits: nearlystraight(0);slightlyemarginateatupper third or fourth (1). (1999), however, described the genus Geodiscelis, which 2. Anteriortentorialpit: notextended(0);extendeddownward combines characters of the two tribes. They therefore did asdeepshininggroovealongepistomalsuturenearlytoapex not recognize tribes in the subfamily. ofclypeus (1). Toshed lighton thegenus Chilicola asa whole, and to 3. Head ofmale: not more than 1.2 times as long as wide (0); show relationships and justification for the new Andean more than 1.5 timesas long aswide (1). subgenus, a preliminary phylogenetic study was made, 45.. FMaacliaalrfsopvaecea:oflfienmeaarle(:0);abbsreonatde(0r);twhealnl-ldoenfgin(1e)d; (l1o).nger than using representatives of all genus-group taxa including broad (2). those that were synonymized by Michener (1995, 2000). 6. Clypeusofmale: dark (0);with yellow (1). The species used in this study of Chilicola, with current 7. Clypeusoffemale: dark (0);withyellow (1). generic and subgeneric names, are listed in Table 1. The 98.. FFaarraaooccuullaarraarreeaassooffmfeamlael:e:dadrakrk(0)(;0)w;iwtihthyeylellolwow(1)(.1). outgroups were Xenochilicoln, Xeromelissa, and Geodiscelis, 10. Faraocular lobe: absent (0); extending downward into species 18 to 20 in Table 1. clypeus (1). The characters, characterstates, and their codes used 11. Face: withoutdepressionextendingdorsolaterallv froman- in the study are listed in Table 2. Table 3 is a matrix that tennal socket (0); with suchdepression (1). shows the distribution of the character states among the 12. Labrum: broaderthanlong (0);aboutas longasbroad (1). 13. Last antennal segment of male: normal (0); much reduced taxa. (1);a merenub so thatantenna appears tobe 12-segmented AnalysiswasbyperformedbyMichaelS. Engel.Char- (2). acterstatesweretreatedasnonadditive(notordered).The 14. Pronotum: with dorsal surface small (short) and at same datamatrixwasconstructedinWinclada(Nixon,1991)and level as scutum (0); very short and below level of scutum, otfheNaoniamly(sGioslwobaosffm,a1d9e93)u.siTnwgothmeirme/i;*maanldlmeangxt*hctorememsawnedrse 15. emExpetidseitnedurimnnagplabgrertlodooevwcel:sicvreioxtbtoauelsndg(ir1)no.govteol(1o)w.erpartofthorax(0);not produced through Winclada (Length 84, Consistency In- 16. Basal area ofpropodeum: about as long as metanotum (0); dex 52, Retention Index 64), using unambiguous optimi- longerthanmetanotum(1);abouttwicelengthofmetanotum zations only. Figure 2 is a strict consensus tree based on (2); about three times lengthofmetanotum (3); shorter than metanotum these two. (4). 17. Stigma with marginsbasal tovein r: divergingapically (0); Table 1. Species used in phvlogeneticanalysisofthesubgenera paralell (1). ofChilicola. 18. Stigma with margin within marginal cell: convex (0);nearly straight (1). 1. C. (Anoediscelis)ashmcadi (Crawford) 19. Distal stigmal perpendicular: crossing near middle of sec- 2. C. (Anoediscelis)herbsti(Friese) ond submarginal cell (0); crossing near base ofsecond sub- 3. C. (Stenoediscehs) inermis (Friese) marginal cell (1); crossing near apex ofsecond submarginal 4. C. {Chilicolas.str.) mbriventrisSpinola cell;crossingbasal to second submarginal cell (3). 5. C. {Chilioediscelis)patagonicaToreand Moldenke 20. Hindtibialspurs: slenderandalmoststraight(0);strongand 6. C. (Hylaeosomn) mcxicannToreand Michener curved (1). 7. C. {Hylaeosomn)aeqnatoricnsis Benoist 21. Claws of female: bifurcate (0); with inner ramus rudimen- 8. C. {Oediscelis) ivnialis (Philipps) tary (1). 9. C. {hiloprosopii) solerviceiisiToroand Moldenke 22. Hind basitarsus of male: simple (0); with ventral swelling 10. C. {Oediscelisca) dalmcidai Moure basally or medially (1). 11. C. {Heteroediscelis) mavidaToroand Moldenke 23. Hindfemurofmale: notoralittleswollen(0);stronglyswol- 12. C. {Prosopoides)firosopoidcs (Ducke) len (1). 13. C. {Pseiidoscelis) rostrata (Friese) 24. Hind tibia of male: longer than femur, when folded reach- 14. C. {Owediscelis)espeleticola Michener ingbase oftrochanter unless swollen so thatitcannotdo so 15. C. {Owediscelis) brzoskaiMichener (0);shorter than toas longas femur (1). 16. C. (unassigned) hahiiiHerbst 25. SI ofmale:withlargetubercleorprojection(0);withoutsuch 17. C. (unassigned)^v;i/f(['m'r; Moure a projection (1). 18. XcnochUicola mamignaToro & Moldenke 26. Hind tibia of male: unmodified (0); strongly swollen dis- 19. Xeromelissa wilmattaeCockerell tally(l). 20. Geodiscelis iriegacephala Michenerand Rozen 27. Hind trochanter of male: simple (0); with ventral angle or protuberance (1). Scientific Papers, Natural History Museum, The University of Kansas Table2. Continued 28. S4tifmale: simple (0); with twotuberclesorprojections(1). 34. Malegonoforceps: with ventral, mesial angleor lobedistal 29. S8 ofmale, distal process: slender, deeply bitid (0); greatly tovolsella (0); withoutsuchanangleor lobe (1). broadenedmedially,bifid(1);truncateorcmarginate,ifbroad- 35. Thoraciclength(measuredfromposteriormarginofpronotal ened,broadestatapex (2);slenderand pointed apically (3). lobe to metasomal articulation): little more than thoracic 30. S7ofmale, apical lobes: distinct from body ofsternum (0); height (lateral view) (0); nearly 1.5 times thoracic height, or unrecognizably fused tobodyofsternum (1). more (1). 31. S7ofmale: withfourapical lobes(0);withonepairoflobes 36. Topofhead: notgreatlydeveloped abovesummitsofeyes, reduced sothatonly twoareeasily recognizable (1). margin in facial view extending mesiad from slightlyabove 32. Apexofpenisvalve: withtwomembranousappendages(0); summits of eyes and upper ocular tangent usually passing with one such appendage (1); without such appendages (or throughoraboveanteriorocellus(0);stronglyde\'elopedabove they areminute) (2). summitsofeyes,margininfacial viewcontinued upward be- 33. Malegonostylus: recognizablydistinctfromgonocoxite(0); yondeyesbeforebendingmesiadandupperoculartangentin indistinguishably united withgonocoxite(1). astrictlyfacial view usuallybelowanteriorocellus(1). 5 9 16 17 18 19 23 26 35 010113110 Xeromelissa wilmattae 1 12 21 29 30 31 33 113 110 Geodiscelis megacephala Prosopoidesprosopoides 2 3 1 Pseudoscelis rostrata 3 5 18 24 6 22 26 28 29 32 12 11 Oroediscelis espeleticola r-C>-»0 • • O- — c 1 1 1 1 Oroediscelis brzoskai Stenoediscelis inermis Anoediscelis ashmeadi Anoediscelis herbsti — 3 Hylaeosoma mexicana Hylaeosoma aequatoriensis Xenochilicola mamigna Oediscelisca dalmeidai UNASSIGNED gutierrezi Heteroediscelis mavida - UNASSIGNED hahni 6 8 13 Idioprosopis solervicensi 1 iJ 16 Oediscelis vernalis 2 1 — 20 Chilicola rubriventris 15 21 2i 25 26 110 10 Chilioediscelispatagonica Fig. 2. Preliminary phylogeneticstudyofthesubgenera ofChilicola; Xeromelissaand Geodiscelisareoutgroups. This is a strict consensus tree based on two minimum-length trees found by NONA (Length 84, CI 52, RI 64). Black dots represent unique changes,whitedots(circles)representhomoplasiouschangesthatoccurelsewherein thistree. Numbersabovedotsarecharacter numbers; numbersbelow dotsshow thestatearisingat that point. Bee Genus Chilicola in the TropicalAndes Table3. Characterstatesofrepresentative taxa used in thesubgenericanalysis. Characters 1-18 Taxon 10 11 12 13 14 15 16 17 18 C. (Anoediscelis)nshiiieiuii 8 Scientific Papers, Natural History Museum, The University of Kansas I regard this analysisaspreliminarybecausesomany 2. Head above antennal alveolus w^ith depression (some- cladesaresupportedbyonlyoneortwocharacters.Anew times very weak) extending up toward ocellocular re- analysisbased on moretaxaand morecharactersisneces- gion; S8 ofmale with apical process deeply bifid (as in mm sar\'before the resultsshould be used asabasis forclassi- Figs. 14c, 15c);body length4.0 to5.5 butas littleas mm fication. Inparticular,Ihesitatetoplace/4H0t'(f;srt'//swithin 3.0 in C. sniithpardoi Hylaeosoma Hylaeosoma. However, recognition of the new subgenus Head without depression above antennal alveolus; S8 Oroediscelis is well supported. ofmale with apical process truncate (as in Figs. 5b, 6c); mm KeytotheSubgener.a of Chilicola Known From theAndes body length 3.0 to 3.75 Anoediscelis North ofChile TheGroupof Chiucola {Anoediscelis) ashmeadi 1. Distal stigmal perpendicular crossing or near to first This group consists of nine species and ranges from submarginal crossvein or at leastbefore middle ofsec- central Mexico (states of Jalisco and Puebla) to Peru. It ond submarginal cell (Fig. 3a);malarspaceonethirdas agrees with the subgeneric characters as indicated by the long as broad or more (Fig. 4d); S4 ofmale with pairof keysanddescriptionsofToroand Moldenke (1979),using tubercles or projections; hind tibia and usually Anoediscelisinanarrowsense,andofMichener(1994, 1995, basitarsus ofmale swollen and modified... Oroediscelis 2000), recognizing the subgenus in a broader sense, ex- Distal stigmal perpendicular crossing near middle of cept that the stigma is sometimes as long as the costal second submarginal cell or more distally (Fig. 3b-d); margin of the marginal cell. The following are characters malar space linear (= absent) (Fig. 4a-c), rarely about of the species group: one third as long asbroad; S4 ofmale simple or nearly Body length 3.0 to3.75 mm; forewing length2.4 to2.9 so; hind tibia and basitarsus ofmale slender, notmodi- mm. Coloration: Black, clypeus of male usually largely fied 2 yelloworwith yellow area, faceotherwiseblack (orrarely a Fig.3. ForewingsofC/«7ico/rtwithdistalstigmal perpcndicuLirsindiciitodIn broken lines, a,C. {Oroniisceli$)espclctiivlaparatype; b, C. (Hi/laeosoma) nearcniici (female); c, C. {Hylaeosonia) ncquntoriciniis; d, C. (Aiiocdixclis) xoiitlio^iintlw paratype. (I'hotus by M. S. Engel.)