m
tjr
f ret <
ii:!^
^
ICCt
S?»!i!!o
^ 1% •'
vKv
•i • ^ • • t
li f i »
11 fit
C i 111
r i .% 1 1 '^
B^ ^'rf ^^■#' ^^^ ^^^.tt"
Mi
(
vk
Scientific Papers
Natural History Museum 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 of Entomologi/, Nniural Histon/ Museum, and Department of Ecology and Evolutionan) Bi'Slbg}/,
The Uiiiivrsiti/ of Kansas, Lawrence, Kansas 66045
CONTENTS
ABSTRACT 2
RESUMEN 2
INTRODUCTION 3
Acknowledgments 3
ABBREVIATIONS, TERMS, AND METHODS 4
SUBGENERA OF CH/L/COL/l 5
Key to the Subgenera of Ch/l;col4 Found in the Andes North of Chile 8
THE GROUP OF CHILICOLA {ANOEDISCELIS) ASHMEADI 8
Key to Males of the Chilicola (Anoedicelis) ashmeadi Group 11
Chilicola (Anoediscelis) ashmeadi Crawford 11
Ch/l;col4 (Anoediscelis) colombiana new species 13
Chilicola (Anoediscelis) mistica new species 13
Chilicola (Anoediscelis) venezuelana new species 13
Ch/l;col4 (Anoediscelis) wygodzinskyi new species 14
Chilicola (Anoediscelis) xanthostoma new species 14
Chilicola (Anoediscelis) xanthognatha new species 16
Chilicola (Anoediscelis) cooi'eri new species 16
Chilicola (Anoediscelis) pedunculata new species 16
'Contribution Number 3230 from the Division of Entomology, Natural History Museum, The University of Kansas. Lawrence, Kansas 66045, USA
© Natural History Museum, The University of Kansas ISSN No. 1094-0782
Ernst twL»^,.„,^
Wu^
|H*^.
.t r* - -
1 ,^:-
,.;■ L'''-
Scientific Papers
Natural History Museum 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 of Ecology and EvolutioiumiBiolbgn/, '
Tlie University of Kansas, Laivrence, Kansas 66045
CONTENTS
ABSTRACT 2
RESUMEN 2
INTRODUCTION 3
Acknowledgments 3
ABBREVIATIONS, TERMS, AND METHODS 4
SUBGENERA OF CH/L/COM 5
Key to the Subgenera of Chilicola Found in the Andes North of Chile 8
THE GROUP OF CHILICOLA (ANOEDISCELIS) ASHMEADI 8
Key to Males of the Chilicola (Anoedicelis) ashmeadi Group 11
Chilicola (Anoediscelis) ashmeadi Crawford 11
Chilicola (Anoediscelis) colombiana new species 13
Chilicola (Anoediscelis) mistica new species 13
Chilicola (Anoediscelis) venezuelana new species 13
Chilicola (Anoediscelis) wycodzinskyi new species 14
Chilicola (Anoediscelis) xanthostoma new species 14
Ch;ucol4 (Anoediscelis) xanthognatha new species 16
Chilicola (Anoediscelis) cooperi new species 16
Chilicola (Anoediscelis) pedunculata new species 16
'Contribution Number 3230 from the Division of Entomology. Natural Historv Museum. The University of Kansas, Laurence. Kan.sas 66045. USA
© Natural History Museum, The University of Kan.sas ISSN No. 1094-0782
2 Scientific Papers, Natural History Museum, The University of Kansas
SUBGENUS HYLAEOSOMA ASHMEAD 17
Key to the Andean Species oe Hylaesoma 18
Chilicola {Hylaesoma) aequatoriensis Benoist 18
Chiucola (Hylaesoma) involuta new species 20
Chilicola (Hylaesoma) umbonata new species 21
Chilicola (Hylaesoma) canei new species 21
Chilicola (Hylaesoma) belli new species 23
Chilicola (Hylaesoma) smithpardoi new species 24
Note on Extralimital Hylaesoma 26
OREOD/SCEL/S NEW SUBGENUS 26
Illustrations of Oreodiscelis 28
Key to the Species of the Subgenus Oreod/scel/s 28
Chilicola (Oreodiscelis) espeleticola new species 29
Nests of Chilicola espeleticola 31
Chilicola (Oreodiscelis) brooksi new species 32
CH/L/COL4 (Oreodiscelis) benoistiana new species 32
Chilicola (Oreodiscelis) quitensis Benoist 35
CH/i.;coL4 (Oreodiscelis) cuzcoens/s new species 36
Ch;l;col4 (Oreodiscelis) bigibbosa new species 36
Chilicola (Oreodiscelis) maculipes new species 39
Chilicola (Oreodiscelis) styliventris (Friese) 40
Nests of Chilicola styliventris 41
Chilicola (Oreodiscelis) transversaria new species 41
Chilicola (Oreodiscelis) gibbosa new species 42
Chilicola (Oreodiscelis) brzoskai new species 43
Chilicola (Oreodiscelis) simplex new species 45
Phytogeny of the Species of the Subgenus Oreodiscelis 45
LITERATURE CITED 47
ABSTRACT This is a revision of the species of the genus Chilicola found in the South American Andes, Peru to Venezuela, above elevations of 1000 m. In addition to the four previously described species known from this area, 23 new species are recognized and described. They fall into three groups: (1) The C. asliinciidi group of the 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 Benoist and 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 Andes Suramericanos desde Peru hasta Venezuela por encima de los 1000 msnm. Adicionalmente a la especie de Mesoamerica y las quatro especies conocidas para el area, se establecen y describen 23 especies nuevas. 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) subgenero Hylaeosoma para C. aequatoriensis Benoist y las siguientes especies nuevas: C. involuta, umbonata, canei, belli, smithpardoi; (3) el nuevo subgenero Oroediscelis (especie tipo, Chilicola styliventris) para C. styliventris Friese, C. quitensis Benoist, y las siguientes especies nuevas: C. espeleticola, brooksi, benoistiana, cuzcoensis, bigibbosa, maculipes, transversaria, gibbosa, brzoskai, simplex. Se presentan analisis filogeneticos para los subgeneros de Chilicola y para las especies del subgenero Oroediscelis, ademas de algunos 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 Tropical Andes
INTRODUCTION
The genus Chilicola consists of small, slender, largely black bees (Fig. 1) ranging from Santa Cruz Province, Ar- gentina, and Asien, Chile, north to the states of Tamaulipas and Jalisco in Mexico and St. Vincent in the Lesser Antilles. In much of this wide range, species of Chilicola are few and specimens rarely collected. The literature shows them to be abundant and diverse only in Chile, which has a fauna of 32 known species, placed in seven subgenera (Toro and Moldenke, 1979; Toro, 1986). Michener (1995) reduced the number of Chilean subgenera to four, but the diversity is clearly substantial.
Only three species from the Andes north of Chile have been described (Benoist, 1942, two species from Ecuador; Friese, 1908, one species from Peru); a fourth species found in Colombia was described from Central America. It now appears, however, that there is a rich fauna in the Andes of Peru, Ecuador, Colombia, and Venezuela, i.e., in the Pima and Paramo Nordandino and adjacent provinces recog- nized by Morrone (2001). The present paper concerns the 27 species known from these mountains. Because these bees are small, because there are relatively few bee collec- tors, and above all because Andean weather is so often not favorable for flight by insects like bees that are usually in their nests when the weather is cold, windy, cloudy, foggy, and rainy, collecting with a net from flowers is of- ten not a successful way of sampling the Andean Chilicola fauna. Many more species will no doubt be found, for several of those described herein are known from a single locality and one or a few specimens. Benoist (1942) ap- pears to have been the first person to realize that they can be obtained from pithy stems in Ecuador (regardless of the weather), and Robert W. Brooks of the University of Kansas recognized the same thing in the Venezuelan Andes, where he collected many specimens used in this study. This collecting method should greatly increase knowledge of this genus in the future, although some spe- cies may nest principally or exclusively in burrows made by beetles in wood or woody stems, rather than in pithy stems. Traps that can function whenever there is a period of sunny weather also would be worth trying.
All specimens used in the present paper were obtained at or above the elevation of 1,000 m. The only lowland species known from Peru to Venezuela are Chilicola (Hylaeosoma) stciwcephala Brooks and Michener from Ama- zonian Colombia and an undescribed species probably of the C. ashiiicadi group (males not known) from Talara in northern coastal Peru.
Andean Chilicola can be easily divided into three groups: (1) the C. ashiiieadi group usually included in the subgenus Anocdiscelis, (2) the subgenus Hylacosoina, and (3) the new subgenus Owediscelis. The first group ranges from the Andean region to Mexico, with other species of
Fig. 1. Oiilicohi (Orocdiscclis) espdcticola Michener, male (photo by M. S. Engel).
Anocdiscelis in Chile; the second ranges from the Andean region and Brazil and Bolivia, north to Mexico; the third group, Owediscelis, is limited to the Andean region north of Chile. Thus, except possibly for the C. ashmeadi group, the Andean fauna does not consist of a northern montane extension of the rich, temperate zone Chilean and Argen- tine fauna. At least the Andean Hylaeosoma presumably were derived from lowland tropical antecedents. Hi/lacosouia exists in the lowland tropics and was present in the Dominican Republic and doubtless elsewhere in the tropics at least as early as Miocene times; Chilicola {Hylaeosoiua) gracilis Michener and Poinar (1997) and C. (H.) elcctwdoniinica Engel (1999) were based on fossils in Dominican amber. The subgenus does not occur in tem- perate austral South America; its southernmost known localities are recorded below in a note on extralimital (non- Andean) species at the end of the account of species of Hi/laeosoiiia. The origin of the Andean endemic subgenus Owediscelis is not known.
Acknowledgements
I am especially indebted to Dr Robert W. Brooks who collected many of the specimens of Andean Chilicola and who assembled much of the borrowed material for study. He also provided the nests and information on nesting bi- ology of two species. It was our original intent to publish this study jointly, but pressure of other responsibilities made his further contributions impossible.
Dr. Michael S. Engel made the photographic illustra- tions and the analyses of my character matrices with NONA.
The cooperation of many museum curators and col- lectors who loaned specimens is much appreciated. They
Scientific Papers. Natural History Museum, The Universh v oi Kansas
are listed in the section on Abbre\ iations. Terms and Meth- ods. Roy S. Snelling and Bryan N. Danforth were espe- cially helpful in arranging for the disposition of impor- tant material.
Finally, Dr. J. S. Ashe, Curator-in-charge of the Ento- mology Division, University of Kansas Natural History Museum and Biodiversity Research Center, kindl\- permit- ted my use of space and facilities.
ABBREVIATIONS, TERMS, AND METHODS
T. — tergum, followed by the number of a metasomal tergum; thus Tl is the first tergum of the metasoma.
S. — sternum (as explained for T).
Alveolus . — antennal socket.
Distal stigmal perpendicular. — an imaginary line across the front wing (Fig. 3) tlirough the apex of the stigma and at a right angle to the costal margin of the wing. The descriptions below record where this line leaves the mar- ginal cell, in relation to the second submarginal cell, i.e., in relation to the anterior end of the first or of the second submarginal crossvein.
Measurements and illustrations of the hind tibia of males of the subgenus Oroediscelis appear repeatedly in the following pages. The measurements were made on the upper outer view with the preapical brush of hairs characteristic of most species of this subgenus in full view, the crests on the under surface being out of sight. For il- lustrations, however, the tibia, is tilted to show the sum- mit of the inner crest visible behind the outer crest. For additional comments on the illustrations of Oroediscelis, see the accouiit of that subgenus.
Facial proportions were obtained by dividing the length (apex of clypeus to summit of vertex in facial view) by the width (greatest distance between apparent outer limits of eyes in facial view), using an eye-piece microme- ter in a dissecting microscope.
Forewing length was measured with the same equip- ment from the base of the enlarged base of vein R to the wing apex.
For all illustrations divided by a vertical line, the left hand side is dorsal, the right hand side ventral.
Following is a list of collections and museums con- taining material that I used, the cities by which they are identified (in italics) in the text, and in brackets, the names of the persons who facilitated my use of the specimens.
Aarhus = Claus Rasmussen collection, to be placed in
Aarhus, Denmark. Berkeley = Essig Entomological Museum, University of
California, Berkeley, California 94720, U.S.A. [Cheryl
Barr]. Berlin = Museum fiir Naturkunde der Humboldt-
Universilat, Berlin, Germany [F. Koch]. Bogota Humboldt = Instituto Alexander von Humboldt,
Santa Fe de Bogota 2 DC, Colombia [F. Fernandez]. Bogota Univ. = Departamento de Biologia, Universidad
Nacional de Colombia, Santa Fe de Bogota, Colombia
(G. Nates Parra, and V.H. Gonzalez-B.j.
Davis = Department of Entomology, University of Cali- fornia, Davis, California 95616, U.S.A. [L. Kimsey]. Ithaca = Department of Entomology, Cornell University,
Ithaca, New York 14853, U.S.A. [E. R. Hoebeke]. La Molina = Museo de Entomologia, Universidad Nacional
Agraria La Molina, near Lima, Peru [Clorinda Vergara,
C. Rasmussen]. Lawrence =Entomology Division and Snow Entomology
Collection, Unix'ersity of Kansas Natural History Mu- seum, Lawrence, Kansas 66045, U.S.A. Lima = Museo de Historia Natural, Universidad Nacional
Mayor de San Marcos, Lima, Peru [G. Lamas, C.
Rasmussen]. Logan = Bee Biology and Systematics Laboratory, Utah
State University, Logan, Utah 84322, U.S.A. [T. L.
Griswold]. London =The Natural History Museum, London SW7 5BD,
England, U.K. ]C. Taylor]. Los Angeles = Natural History Museum of Los Angeles
County, Los Angeles, California 90007, U.S.A. [R. R.
Snelling]. Lyme Regis = M. Cooper collection, L\-me Regis, Dorset,
England, U.K. Maracay = Museo del Instituto de Zoologia Agricola,
Universidad Central de Venezuela, Maracay 2101A,
Venezuela ]]. L. Garcia, L. J. jolly]. Medellin = Museo Entomologico "Francisco Luis Gallego,"
Unixersidad Nacional de Colombia, Medellin, Colom- bia jGilberto Morales-Soto]. New York = American Museum of Natural History, New
York, New York 10024, U.S.A. [J. G. Rozen]. Ottawa = Canadian National Collection in Centre for Land
and Biological Resources, Ottawa, Ontario, Canada JL.
Dumochel]. Paris = Mu.seum National d'Histoire Naturelle, 75005 Paris,
France [J. Casevitz-Weulersse]. PCAM-Austin = Programa Cooperati\'o sohre la
Apifaima Mexicana, specimens at Central Texas
Melittological Institute, .Austin, Texas 78731, U.S.A.
[J. L. Neff]. Sao Paulo = Museu de Zoologia, Unixersidaele de Sao
Paulo, Sao Paulo, Brazil [R. F. Brandao]. San Francisco = California Academy of Sciences, Golden
Gate Park, San Francisco 94 118, U.S.A. [VV. J. Pulawski]. Washington = National Museum of Natural History,
Smithsonian Institution, Washington, D. C. 20560,
U.S.A. [R.J. McGinley].
Bee Genus Chiucola in the Tropical Andes
In listing collecting localities, I have given provinces or departments in italics. When these were not on labels, these names are in brackets. This is sometimes potentially
important, because of occasional duplication of place names within a country. Bracketed names are my inter- pretations, and therefore possibly incorrect.
SUBGENERA OF CHILICOLA
Michener (1995, 2000) divitied the Xeromelissinae into two tribes, Xeromelissini and Chilicolini, the last for two genera, Chilicola and Xeiiochilicola. Michener and Rozen (1999), however, described the genus Geodiscelis, which combines characters of the two tribes. They therefore did not recognize tribes in the subfamily.
To shed light on the genus Chilicola as a whole, and to show relationships and justification for the new Andean subgenus, a preliminary phylogenetic study was made, using representatives of all genus-group taxa including those that were synonymized by Michener (1995, 2000). The species used in this study of Chilicola, with current generic and subgeneric names, are listed in Table 1. The outgroups were Xenochilicoln , Xeromelissa, and Geodiscelis, species 18 to 20 in Table 1.
The characters, character states, and their codes used in the study are listed in Table 2. Table 3 is a matrix that shows the distribution of the character states among the taxa.
Analysis was by performed by Michael S. Engel. Char- acter states were treated as nonadditive (not ordered). The data matrix was constructed in Winclada (Nixon, 1991) and the analysis was made using the re/;* and max* commands of Noim (Goloboff, 1993). Two mimimal length trees were produced through Winclada (Length 84, Consistency In- dex 52, Retention Index 64), using unambiguous optimi- zations only. Figure 2 is a strict consensus tree based on these two.
Table 1. Species used in phvlogenetic analysis of the subgenera of Chilicola.
1. C. (Anoediscelis) ashmcadi (Crawford)
2. C. (Anoediscelis) herbsti (Friese)
3. C. (Stenoediscehs) inermis (Friese)
4. C. {Chilicola s. str.) mbriventris Spinola
5. C. {Chilioediscelis) patagonica Tore and Moldenke
6. C. (Hylaeosomn) mcxicann Tore and Michener
7. C. {Hylaeosomn) aeqnatoricnsis Benoist
8. C. {Oediscelis) ivnialis (Philipps)
9. C. {hiloprosopii) solerviceiisi Toro and Moldenke
10. C. {Oediscelisca) dalmcidai Moure
11. C. {Heteroediscelis) mavida Toro and Moldenke
12. C. {Prosopoides) firosopoidcs (Ducke)
13. C. {Pseiidoscelis) rostrata (Friese)
14. C. {Owediscelis) espeleticola Michener
15. C. {Owediscelis) brzoskai Michener
16. C. (unassigned) hahiii Herbst
17. C. (unassigned) ^v;i/f(['m'r; Moure
18. XcnochUicola mamigna Toro & Moldenke
19. Xeromelissa wilmattae Cockerell
20. Geodiscelis iriegacephala Michener and Rozen
Table 2. Characters and their states for subgeneric analysis.
1 . inner orbits: nearly straight (0); slightly emarginate at upper third or fourth (1).
2. Anterior tentorial pit: not extended (0); extended downward as deep shining groove along epistomal suture nearly to apex of clypeus (1).
3. Head of male: not more than 1.2 times as long as wide (0); more than 1.5 times as long as wide (1).
4. Facial fovea of female: absent (0); well-defined (1).
5. Malar space: linear (0); broader than long (1); longer than broad (2).
6. Clypeus of male: dark (0); with yellow (1).
7. Clypeus of female: dark (0); with yellow (1).
8. Faraocular areas of male: dark (0); with yellow (1).
9. Faraocular areas of female: dark (0); with yellow (1).
10. Faraocular lobe: absent (0); extending downward into clypeus (1).
11. Face: without depression extending dorsolaterallv from an- tennal socket (0); with such depression (1).
12. Labrum: broader than long (0); about as long as broad (1).
13. Last antennal segment of male: normal (0); much reduced (1); a mere nub so that antenna appears to be 12-segmented (2).
14. Pronotum: with dorsal surface small (short) and at same level as scutum (0); very short and below level of scutum, medium part declivitous (1).
15. Episternal groove: extending to lower part of thorax (0); not extending below scrobal groove (1).
16. Basal area of propodeum: about as long as metanotum (0); longer than metanotum (1); about twice length of metanotum (2); about three times length of metanotum (3); shorter than metanotum (4).
17. Stigma with margins basal to vein r: diverging apically (0); paralell (1).
18. Stigma with margin within marginal cell: convex (0); nearly straight (1).
19. Distal stigmal perpendicular: crossing near middle of sec- ond submarginal cell (0); crossing near base of second sub- marginal cell (1); crossing near apex of second submarginal cell; crossing basal to second submarginal cell (3).
20. Hind tibial spurs: slender and almost straight (0); strong and curved (1).
21. Claws of female: bifurcate (0); with inner ramus rudimen- tary (1).
22. Hind basitarsus of male: simple (0); with ventral swelling basally or medially (1).
23. Hind femur of male: not or a little swollen (0); strongly swol- len (1).
24. Hind tibia of male: longer than femur, when folded reach- ing base of trochanter unless swollen so that it cannot do so (0); shorter than to as long as femur (1).
25. SI of male: with large tubercle or projection (0); without such a projection (1).
26. Hind tibia of male: unmodified (0); strongly swollen dis- tally(l).
27. Hind trochanter of male: simple (0); with ventral angle or protuberance (1).
Scientific Papers, Natural History Museum, The University of Kansas
Table 2. Continued
28. 29.
30.
31.
32.
33.
S4 tif male: simple (0); with two tubercles or projections (1). S8 of male, distal process: slender, deeply bitid (0); greatly broadened medially, bifid (1); truncate or cmarginate, if broad- ened, broadest at apex (2); slender and pointed apically (3). S7 of male, apical lobes: distinct from body of sternum (0); unrecognizably fused to body of sternum (1 ). S7 of male: with four apical lobes (0); with one pair of lobes reduced so that only two are easily recognizable (1). Apex of penis valve: with two membranous appendages (0); with one such appendage (1); without such appendages (or they are minute) (2).
Male gonostylus: recognizably distinct from gonocoxite (0); indistinguishably united with gonocoxite (1).
34.
35.
36.
Male gonoforceps: with ventral, mesial angle or lobe distal to volsella (0); without such an angle or lobe (1). Thoracic length (measured from posterior margin of pronotal lobe to metasomal articulation): little more than thoracic height (lateral view) (0); nearly 1.5 times thoracic height, or more (1).
Top of head: not greatly developed above summits of eyes, margin in facial view extending mesiad from slightly above summits of eyes and upper ocular tangent usually passing through or above anterior ocellus (0); strongly de\'eloped above summits of eyes, margin in facial view continued upward be- yond eyes before bending mesiad and upper ocular tangent in a strictly facial view usually below anterior ocellus (1 ).
5 9 16 17 18 19 23 26 35
010113110 1 12 21 29 30 31 33
0 113 110
2 3 0 1
Xeromelissa wilmattae Geodiscelis megacephala Prosopoides prosopoides Pseudoscelis rostrata
3 5 18 24
6 22 26 28 29 32
r-C>-»0 • • O-
c 1 1 1 1 0
Oroediscelis espeleticola
12 11
— Oroediscelis brzoskai Stenoediscelis inermis Anoediscelis ashmeadi Anoediscelis herbsti
3
— Hylaeosoma mexicana Hylaeosoma aequatoriensis Xenochilicola mamigna Oediscelisca dalmeidai UN ASSIGNED gutierrezi
Heteroediscelis mavida - UN ASSIGNED hahni
6 8 13
0 0 1
iJ 16
2 1 20
Idioprosopis solervicensi Oediscelis vernalis
— Chilicola rubriventris
15 21 2i 25 26
110 10
Chilioediscelis patagonica
Fig. 2. Preliminary phylogenetic study of the subgenera of Chilicola; Xeromelissa and Geodiscelis are outgroups. 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, white dots (circles) represent homoplasious changes that occur elsewhere in this tree. Numbers above dots are character numbers; numbers below dots show the state arising at that point.
Bee Genus Chilicola in the Tropical Andes
Table 3. Character states of representative taxa used in the subgeneric analysis.
Taxon
Characters 1-18
10 11 12 13 14 15 16 17 18
|
C. (Anoediscelis) nshiiieiuii |
1 0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. (Anoeiiiscelis) IwrbstI |
1 0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. {Stenoediscelis) inermis |
1 0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
|
C. (Chilicola) rubriventris |
1 0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
C. {Chilioediscelis) patagonica |
1 0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
|
C. (Hi/laeosonm) mexiama |
1 0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. (Hylncosouin) nequatorietisis |
1 0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. (Oediscelis) vernalis |
1 0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
1 |
0 |
0 |
|
C. (Idioprosopis) solervicensi |
1 0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
|
C. (Oediscelisca) dalmeidai |
1 0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. (Heteroediscelis) mavida |
1 0 |
1 |
0 |
1 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. {Prosopoides) pwsopoides |
1 1 |
-> |
1 |
1 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
3 |
0 |
0 |
|
C. (Pseiidoscelis) rostrata |
1 1 |
-> |
0 |
2 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
3 |
0 |
0 |
|
C. (Oroediscelis) espeleticola |
1 0 |
1 |
0 |
2 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
1 |
|
C. (Oroediscelis) brzoskai |
1 0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
0 |
|
C. (unassigned) liahiii |
1 0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
C. (unassigned) giitierrezi |
1 0 |
0 |
0 |
0 |
1 |
7 |
1 |
7 |
0 |
0 |
0 |
0 |
0 |
0 |
7 |
0 |
0 |
|
C. (Xeiiochilicola) mamigna |
D 0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
4 |
0 |
1 |
|
Xeromelissa zvilmattae |
1 1 |
0 |
0 |
0 |
1 |
1 |
1 |
1 |
1 |
0 |
0 |
0 |
1 |
0 |
0 |
1 |
1 |
|
Geodiscelis iiwgncephala |
3 1 |
0 |
0 |
1 |
1 |
1 |
0 |
0 |
1 |
0 |
1 |
0 |
1 |
0 |
2 |
0 |
0 |
|
Characters 19- |
-36 |
||||||||||||||||
|
Taxon ] |
9 20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
31 |
32 |
33 |
34 |
35 |
36 |
|
C. (A}iocdiscelis) aslwiendi |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
1 |
|
C. (Anoediscelis) herbsti |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
2 |
|
C. (Stenoediscelis) inermis |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
0 |
|
C. (Chilicola) rubriventris |
1 |
1 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
2 |
0 |
0 |
0 |
|
C. (Chilioediscelis) patagonica |
7 |
1 |
1 |
0 |
1 |
0 |
1 |
0 |
1 |
0 |
2 |
0 |
0 |
0 |
|
C. (Hylaeosoma) mexicana |
2 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
|
C. (Hylaeosoma) aeqiiatoriensis |
2 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
2 |
|
C. (Oediscelis) vernalis |
1 |
0 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
2 |
0 |
0 |
0 |
|
C. (Idioprosopis) solervicensi |
1 |
0 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
2 |
0 |
0 |
0 |
|
C. (Oediscelisca) dalmeidai |
0 |
0 |
0 |
0 |
1 |
1 |
1 |
1 |
1 |
0 |
2 |
0 |
1 |
2 |
|
C. (Heteroediscelis) mavida |
1 |
0 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
2 |
0 |
0 |
2 |
|
C. (Prosopoides) prosopoides |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
2 |
|
C. (Pseiidoscelis) rostrata |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
2 |
|
C. (Oroediscelis) espeleticola |
1 |
0 |
0 |
1 |
0 |
1 |
1 |
1 |
0 |
1 |
1 |
0 |
0 |
0 |
|
C. (Oroediscelis) brzoskai |
1 |
0 |
0 |
1 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
0 |
0 |
|
C. (unassigned) hahni |
1 |
0 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
2 |
0 |
0 |
7 |
|
C. (unassigned) giitierrezi |
7 |
0 |
7 |
0 |
1 |
1 |
1 |
1 |
0 |
0 |
2 |
0 |
1 |
7 |
|
C. (Xenochilicola) mamigna |
1 |
0 |
0 |
0 |
0 |
1 |
1 |
0 |
0 |
0 |
3 |
0 |
1 |
2 |
|
Xeromelissa wilmattae |
3 |
0 |
0 |
0 |
1 |
0 |
1 |
1 |
0 |
0 |
2 |
0 |
0 |
2 |
|
Geodiscelis megaceplmla |
1 |
0 |
1 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
3 |
1 |
1 |
2 |
Figure 2 shows Xeromelissa and Geodiscelis as outgroups, but the other intended outgroup genus Xenochilicola, placed close to Chilicola by Michener (1995, 2000), appears within Chilicola and should be regarded as a subgenus unless Chilicola is broken up. Anoediscelis (in- cluding Stenoediscelis according to Michener, 1995) is paraphyletic relative to Hi/laeosonia. For the time being I
do not accept this conclusion, which would place Anoediscelis as a junior synonym of Hylaeosoma. As to the clade containing the unassigned species, all the taxa ex- cept Chilicola sensu stricto and Chilioediscelis were united under Oediscelis by Michener (1995, 2000); the last is here shown as paraphyletic relati\'e to Chilicola sensu stricto and Chilioediscelis.
8
Scientific Papers, Natural History Museum, The University of Kansas
I regard this analysis as preliminary because so many clades are supported by only one or two characters. A new analysis based on more taxa and more characters is neces- sar\' before the results should be used as a basis for classi- fication. In particular, I hesitate to place /4H0t'(f;srt'//s within Hylaeosoma. However, recognition of the new subgenus Oroediscelis is well supported.
Key to the Subgener.a of Chilicola Known From the Andes North of Chile
1. Distal stigmal perpendicular crossing or near to first submarginal crossvein or at least before middle of sec- ond submarginal cell (Fig. 3a); malar space one third as long as broad or more (Fig. 4d); S4 of male with pair of tubercles or projections; hind tibia and usually basitarsus of male swollen and modified ... Oroediscelis
Distal stigmal perpendicular crossing near middle of second submarginal cell or more distally (Fig. 3b-d); malar space linear (= absent) (Fig. 4a-c), rarely about one third as long as broad; S4 of male simple or nearly so; hind tibia and basitarsus of male slender, not modi- fied 2
2. Head above antennal alveolus w^ith depression (some- times very weak) extending up toward ocellocular re- gion; S8 of male with apical process deeply bifid (as in Figs. 14c, 15c); body length 4.0 to 5.5 mm but as little as 3.0 mm in C. sniithpardoi Hylaeosoma
Head without depression above antennal alveolus; S8 of male with apical process truncate (as in Figs. 5b, 6c); body length 3.0 to 3.75 mm Anoediscelis
The Group of Chiucola {Anoediscelis) ashmeadi
This group consists of nine species and ranges from central Mexico (states of Jalisco and Puebla) to Peru. It agrees with the subgeneric characters as indicated by the keys and descriptions of Toro and Moldenke (1979), using Anoediscelis in a narrow sense, and of Michener (1994, 1995, 2000), recognizing the subgenus in a broader sense, ex- cept that the stigma is sometimes as long as the costal margin of the marginal cell. The following are characters of the species group:
Body length 3.0 to 3.75 mm; forewing length 2.4 to 2.9 mm. Coloration: Black, clypeus of male usually largely yellow or with yellow area, face otherwise black (or rarely
a
Fig. 3. Forewingsof C/«7ico/rt with distal stigmal perpcndicuLirs indiciitod In broken lines, a, C. {Oroniisceli$)espclctiivla paratype; b, C. (Hi/laeosoma) near cniici (female); c, C. {Hylaeosonia) ncquntoriciniis; d, C. (Aiiocdixclis) xoiitlio^iintlw paratype. (I'hotus by M. S. Engel.)
Bhh Geni's Chilicola in the Tropical Andes
a
d
Fig. 4. Faces of Chilicoln, females at the left, males at the right, a [Hijiacosoinn) (lequntorleusis; d, C. {Oroediscelis) espeleticola.
C. (Aiwcdiscclis) coloiiibia)hr, b, C. {Hi/lneosovia) umbonnta; c, C.
10
Scientific Papers, Natural History Museum, The University of Kansas
face all black); in Chilicola cooperi lower paraocular areas in addition to clypeus yellow; mandible in males of C. xanthognatha and C. xanthostoma largely yellow (females unknown); labrum also with yellow in C. coopcri and C. xanthostoma; flagellum yellow-brown to dark brown be- neath, dark brown to black above; legs with following parts yellow: front tibia (largely), front and middle tarsi (often dusky apically, mid tarsus sometimes mostly dark), api- ces of femora (rarely reduced to small light brown areas), bases and apices of mid and hind tibiae (rarely reduced to light brown areas or even blackish), sometimes part of hind tarsus, especially basal part. Sculpturing: Light, surface almost everywhere dulled by lineolation or other microsculpturing, well-separated small punctures on much of head and thorax. Pubescence: Short, sparse, dull whit- ish or yellowish, rarely dusky, not forming metasomal fas- ciae laterally; scopa of female white, weakly developed on hind femur and tibia, well developed and hairs with many branches on SI to S3, longest on S2. Head: Face approximately as wide as long in both sexes; upper or- bital tangent near or below lower margin of median ocel- lus; vertex seen from front a uniform curve between sum- mits of eyes; frons convex, without depression or fossa receiving scape; frontal line absent or weakly indicated; frontal tubercle weakly indicated, between antennal bases; clypeus and supraclypeal area not protuberant, in profile only weakly convex except distinctly so at lower end of clypeus; eyes strongly converging below so that lower interocular distance is about half of upper interocular tiis- tance; emargination of inner eye margin extremely weak, adjacent to emargination strip of impunctate, smooth cu- ticle above which is sometimes shiny area of well-sepa- rated punctures; antennal sockets near middle of face, interalveolar distance twice alveolocular distance; lower margin of clypeus only slightly below lower ocular tan- gent; upper clypeal width, between subantennal sutures, shorter than distance from lower end of subantennal su- ture to eye margin; lower clypeal margin laterally curved back at side of labrum, not or little elevated to form tooth; labrum of male about three times as wide as long, of fe- male over twice as wide as long; maxillary palpus less than half as long as prementum, with first two segments ro- bust, less than twice as long as broad, third and sixth seg- ments longest, fourth and fifth somewhat shorter, sixth broadest preapically (as in other subgenera) with apical third or fourth narrower and looking like a small apical segment; labial palpus about as long as glossa, second seg- ment shortest, less than twice as long as broad, other seg- ments subequal (proboscis not fully examined for some species, but uniform in structure for all that could be seen); subantennal sutures about twice as long as diameter of alveolus, converging below; anterior tentorial pit not elon- gate (Michener, 1995, Fig 1); malar space linear. Thorax:
With pronotal collar narrow, elevated part less than half as wide as maximum width of flagellum. Stigma two-thirds to three-fourths as long, or rarely almost as long as, length of marginal cell on costa; first recurrent vein meeting or slightly basal to first submarginal crossvein (Fig. 3d). Dor- sal surface of propodeum as long as or longer than scutel- lum. Posterior tibia of male slender, about fi\e times as long as wide, widest in apical third (Michener, 1994, Fig. 1 ). Metasoma: First tergum about 1 .5 times as long as broad in males, as long as broad in females. Terga shining, discs transversely lineolate, punctures minute and scattered or reduced to minute, scattered hair bases; marginal zones smooth, impunctate, hairless, but with hair bases laterally on T4 to T6 of males, less smooth and with more hair bases on T4 to T5 of females. Exposed sterna unmodified. S8 with apical process expanded and truncate distally (as in Fig. 5b). Penis valve without dorsoapical membranous process except single large process in C. ashmeadi.
I have found no reliable specific characters to distin- guish females of this group. Differences in punctation and fine sculpturing of the scutum and face are apparent among individuals but do not seem to differentiate these species. Likewise the yellow areas of the legs \'ary among indi- viduals but seem to offer no specific characters.
A\'ailable series of males of most species are small and one could reasonably wonder if the species recognized herein might be based on individual variants. However, C. ashmeadi is available in large numbers from Central America; 1 have dissected two Colombian specimens and eight from localities in Costa Rica without finding signifi- cant variation in the specific characters.
Three female specimens from Talara, Piura, Peru (Lo- gan) are probably members of the Chilicola ashmeadi group. In the absence of males, and because Talara is coastal, not Andean, they remain unnamed. They differ from all other females of the C. ashmeadi group in the largely yellow la- brum, mandibles, legs including trochanters, tegulae (translucent), and axillary sclerites.
The species of Anoediscelis other than the Chilicola ashmeadi group are quite diverse in contrast to the similar species in the C. ashmeadi group; thus in the cladogram presented above, the subgenus is paraphyletic. Except for the C. ashmeadi group, thev are all from Chile; additional species of the subgenus may exist in Argentina. The Chil- ean species differ from the C. ashmeadi group in the char- acters indicated below for each numbered group. (1) C. inermis (Friese) and C. mailen Toro and Moldenke (the Stenoediscelis of Toro and Moldenke, 1979) have the apical lobes of S7 of the male directed apically rather than later- ally, and each penis valve has two preapical membranous projections. The hind tibiae of the males are somewhat thickened, about or less than four times as long as broad. (2) C. olmiie Toro and Moldenke and C. oropliila Toro and
Bee Genus Chilicola in the Tropical Andes
11
Moldenke (included in Hcteroediscelis by Toro and Moldenke) also have two preapical membranous projec- tions on each penis valve, and long hairs on both lobes of S7. Their tentorial pits are somewhat more elongate and the lower part of the clypeus more protuberant than those of the C. aslimendi group. Also included in Heteroediscelis by Toro and Moldenke is C. minor (Philippi) which differs from members of the C. aslvnendi group by the strong tooth on the hind trochanter of the male as well as in the form of the lobes of S7, with the lower apicolateral lobe small and pointed laterodistally. (3) The three species placed in Anoediscelis by Toro and Moldenke (1979) are so diverse that each could be in its own group. Chilicola plebeia SpLnola has thickened hind tibiae in the male, as in group (1) above; C. plebeia and i/ali Toro and Moldenke have the apical pro- cess of S8 of the male deeply bilobed and the lateral lobes developed so as to nearly meet the apical process. Chilicola herbsti (Friese) has long hairs on both upper and lower apicolateral lobes of S7 of the male. At least one species in each numbered group above has yellow lower paraocular areas in the male, a feature known in the C. ashmeadi group only in C. coopcri. The Chilean species most similar to the ashmeadi group are C. herbsti (Friese) and C. minor (Philippi); in some males of the latter the lower paraocular area is marked with yellow. Of the species of the C. ashmeadi group, the most distinctive, especially in the form of S7, is C. pediiiiciilata new species (Fig. 12c).
hi the accounts of the species unique features found in oi\ly one species are italicized.
Key to males of the Chilicola (Anoediscelis) ashmeadi group
1. Mandible yellow or largely so 2
Mandible dark brown to black 4
2. Lower paraocular area with yellow; S6 with submar- ginal bristles, without small brush of dense setae (Fig. 13c) C. cooperi
Lower paraocular area blackish; S6 with midapical brush of dense setae (Fig. lie) 3
3. S7 with coarse setae of upper apicolateral lobe re- stricted to distal half of lobe, lower lobe produced to
attenuate apex anterior to upper lobe (Fig. 10)
C. xanthostoma
S7 with coarse setae of upper apicolateral lobe extend- ing nearly to base of lobe, lower lobe with apex less
attenuate and overlapping upper lobe (Fig 11a)
C. xaiithof^natha
4. S7 with upper apicolateral lobe pedunculate with api- cal fringe of large, long hairs (Fig. 12d); clypeus black C. pediiiiciilata
S7 with upper apicolateral lobe broadest at base, ta- pering, without large, long hairs; clypeus at least partly
yellow except in some specimens of C. colomhiana ... 5
5. S7 with lower apicolateral lobe large, deeply bifid, with small setae on one apex (Fig. 8a); midapical tuft or brush of S6 small (Fig. 8b) C. venezuelana
S7 with lower apicolateral lobe small, hairless; midapical tuft or brush of S6 well developed or ab- sent 6
6. S6 without midapical tuft or brush (Fig. 9c)
C. wygodzinskyi
S6 with distinct midapical tuft or brush (Figs. 5d, e; 6c; 8b) 7
7. Penis valves each with diverging preapical membra- nous projection extending beyond level of apices of gonoforceps (Fig. 5a) C. ashmeadi
Penis valves without divergiiig membranous process, not attaining level of apices of gonoforceps (as in Fig. 9a) 8
8. S6 with erect midapical tuft or biiish arising near trun- cate sternal margin and relatively broad (Fig. 6c)
C. colombiana
S6 with erect midapical ttift or brush near convex ster- nal margin and relatively narrow (Fig. 5d, e)
C. mistica
Chilicola {Anoediscelis) ashmeadi (Crawford) (Figure 5)
Hi/laeosorna ashmeadi Crawford, 1906:161. Cockerell, 1919:185.
Type in Washington. Oediscelis costaricensis Friese, 1916:302. Type in Berlin, not seen;
"typus" in Neiv York. Prosopis homirdiella Cockerell, 1918:423. Type in Washi7igton. Chilicola ashmeadi: Eickwort, 1967:42. Snelling, 1971:52. Chilicola (Anoediscelis) ashmeadi: Michener, 1994:91. Ayala,
Griswold, and Yanega, 1996:444.
Characters of this species were illustrated by Michener (1994:79, 90, 92). Distinguishing characters of males are the following:
Yellow area of clypeus minutely sculptured like adja- cent black areas except smooth (but with scattered punc- tures) lower part, yellow area with margin irregular, of- ten asymmetrical, not reaching upper margin of clypeus, sometimes reduced to area less than half that of clypeus; mandible black, apex reddish; under surface of tlagellum tan, distal segments not quite as wide as upper width of clypeus; middle flagellar segments about as long as broad, somewhat variable depending on curvature of antenna. Tl longer than broad. S6 with midapical tuft of hairs that diverge so that tuft is often divided medially, sternal mar- gin convex; S7 and genitalia as Ln Figure 5, penis valves each with rounded, diverging preapical membranous process extending beyond level of apices of gonoforceps (Fig. 5a).
12
Scientific Papers, Natirai. History Museum, The University of Kansas
Figs. 5-6. 5: CMicoln {Anoediscelh) nshmciuii, male, a. Genitalia (A, apex of penis valve; M preapical membraneous process of penis valve); b, S8; c, S7; d, e. Apical part of S6 of two individuals. 6: C. (A.) colombiana, male, a, S7; b, S8; c. Apical part of S6. The genitalia are similar to those of C. wygodzinskyi.
This species is well known from central Mexico (states of Puebla to Nayarit) south to Costa Rica. The type locali- ties and many collecting localities were listed by Michener (1994). The following are new locality and floral records for Mesoamerica: MEXICO: Oaxaca: Km 2(11 and 202 on highway from Tuxtepec to Oaxaca, 2000 m, 26 Dec. 1990, 19 Feb. 1991 and 19 May 1991, all on flowers of Wi^andia urens (L. Godinez; I mo re nee); Monte Alban, 24 March 1991, on Pera barbellata (L. Godinez; Lawrenee); Chiapas: Parc:|ue Nacional Montebello, 50 km east of La Trinitaria, 1460 m, 14 Apr. 1993, on Baeeharis trinervis (J. L. Neff, PCAM, deposited at Austin). On three dif- ferent dates many specimens were taken in Oaxaca on
Wi;^andia (Hydroph\llaceae) which may bo a fa\cired food plant.
The Andean and first South American records for this species are the following: COLOMBIA: Valle: three males, two females, Cali, probably 1000 m, 8 Jan. 1972 (C. D. Michener, Laivrence); one male, Pichinde, 1700 m, Aug. 28, 1967 (P and B. Wygodzinskv, Wc.r York). The speci- mens from Cali were taken in a dry disturbed area at the edge of the city. At slightly higher elevations and in moister habitats a few kilometers west of Cali, at Pichinde, one male of C. ashmeaiii, one male of C. zvi/i^odzinski/i, and several males of C. colombiana, as well as several uniden- tified females of this group, were also taken.
Bee Genus Chiucula in the Tropical Andes
13
Chilicola (Aiioediscelis) colombiana new species
(Figures 4a; 6)
This species agrees with the characters listed for the group of Chilicola nslimeadi, and the specific characters in- dicated in the key to the species of that group. The fol- lowing characters are those of the male.
Yellow area of clypeus minutely sculptured like adja- cent black areas, yellow area with margin irregular, often asymmetrical, not reaching upper margin of clypeus or (as in holotype) reaching that margin only at median point (or yellow absent in some specimens from Ecuador); man- dible blackish, apex reddish; under surface of flagellum tan, distal segments as wide as upper width of clypeus; middle flagellar segments about as long as broad, slightly longer in some specimens probably depending on curva- ture of flagellum. Tl longer than broad. S6 with mid api- cal fringe, longest hairs at sides of fringe, which ends abruptly laterally, bases of hairs in gently recurved series near weakly concave middle part of sternal margin (Fig. 6c). S7 and S8 as in Figure 6a, b; S7 with lower apicolateral lobe inconspicuous, small triangular (Fig. 6b), upper apicolateral lobe slender, with hairs short; genitalia simi- lar to those of Chilicola unjgodzinskyi.
Holotype male and three male paratypes: COLOM- BIA: Valle: Pichinde, 1630 m, 26 Sept. 1976 (Bell, Breed, and Michener, Lawrence). Additional paratypes, all from Valle: one male and one female, summit west of Cali, 2000 m altitude, 17 Sept. 1976 (Bell, Breed, and Michener, Lawrence); one male, same locality, 18 Sept. 1976 (C. D. Michener, Lawrence); one female (not a paratype), same locality, 2 Feb. 1977 (M. Breed, C. D. Michener, Lawrence); four males and three females, 10 km west of Cali, 1400 m, 19 Sept. 1976 (Bell, Breed, and Michener, Lawrence). Fe- males collected with males have been considered paratypes even though I do not know how to distinguish them from related species.
Additional material: COLOMBIA: Antioquia: one male, three females. La Selva, Rionegro, 2120 m, Oct., 1994, on Riibiis (A. Correa, A Vasquez, Medellin); two females, same locality, on soil, June, 1975 (J. Cano, R. Aiiez, Medellin). A female without a metasoma from Valle: 6 miles [9.7 km] west of Cali, 1630 m, 20 Mar 1955 (E. I. Schlinger, E. S. Ross, San Fra)icisco), may be this species. A female that is likely to be C. colondnaiia is from Antioquia: Giiarne, 3 Oct. 1976 (C. D. Michener, Lawrence). A male from ECUADOR: Pichincha, Calderon, 8500 ft. [2972 m], 1 May 1958 (R. W. Hodges, Ithaca) agrees with C. coloiidnana morphologically but lacks the yellow clypeal area. Two males and three fe- males are from Cumbaya, 7 km east of Quito, 2600 m, 17 June 1987 (M. Cooper, Lyme Re;^is). One of these males has the clypeus entirely black while the other has a small, ill- defined triangular yellowish spot on the lower third of clypeus.
Specimens have been taken in Fehruary. March. June. Au- gust. September, and October
Etymology: An adjectival form based on Colombia.
Chilicola (Anoediscclis) mistica new species (Figure 7)
This species exhibits a mixture of the characters of Chilicola ashnieadi and C. colombiana. Thus the hair tuft on the apex of S6 is as in C. ashmeadi while the penis valves and the lower apicolateral lobes of S7 are more nearly as in C. colombiana. The following are characters of the male.
Yellow of clypeus minutely sculptured like adjacent areas except smooth (but with scattered punctures) be- low, yellow area with irregular and somewhat asymmetri- cal margins; mandible black, apex reddish; under surface of flagellum tan (dusky on last three segments), distal seg- ments nearly as wide as upper width of civpeus; middle flagellar segments about as long as broad to slightly longer than broad, seemingly slightly longer than in C. ashmeadi and colombiana (this statement difficult to document be- cause of variation with curvature of flagellum); Tl longer than broad. S6 as in C. ashmeadi with midapical tuft of hairs that diverge so that tuft is divided medially, apical sternal margin convex; S7 as in Figure 7, with lower apicolateral lobe small, triangular; S8 as in C. ashmeadi; genitalia similar to those of C. loi/godzinski/i.
Holotype male: VENEZUELA: [Merida]: Merida, 1950 m, 13 Sept. 1973 (R. M. Bohart, Logan). One male paratype with apex of metasoma missing, same data (Lawrence).
Etymology: misticus, Latin, mixed, with reference to the mixture of the characters of other species possessed by this species.
Chilicola (Anoediscelis) venezuelatta new species (Figure 8)
This species from near the eastern end of the Venezu- elan montane spur agrees with the characters listed for the group of Chilicola ashmeadi, and the specific characters in- dicated in the key to species of that group. The following characters are those of the male.
Yellow area of clypeus, like rest of face, dull with fine microsculpturing, yellow area with margin somewhat ir- regular, not reaching upper margin of clypeus; mandible black; middle flagellar segments longer than broad; un- der surface of flagellum brown, distal segments about four- fifths as wide as upper width of clypeus. Tl longer than broad. S6 with weakly developed mid apical tuft of di- verging hairs, suggestive of that of Chilicola ashmeadi but less well developed; margin of sternum convex (Fig. 8b). 57 as in Figure 8a with ventral apicolateral lobe large and deeply emarginate, unlike that of any other species of this group; upper apicolateral lobe slender, hairs short and restricted
14
Scientific Papers, Naturae History Museum, The University of Kansas
Figs. 7-9. 7: Chilicola (Anoediscelis) misticn, male. Apical part of S7. S6 and S8 are as in C. ashmeadi; genitalia as in C. un/godzinskyi. 8: C. (A.) venezuelana, male. &, Apical part of S7; b. Apical part of S6. SB about as in C. nshmcadi but minute apical setae even smaller; genitalia about as in C. wygodzj'ws/a/!. 9: C. (A.) un/godzi)tskyi, male, a. Genitalia, b, Apical part of S7;c, Apical part of S6. SB is similar to that of C. ashmeadi.
to apex; S8 similar to that of C. ashmeadi but with hairs at apical comers of apical process even smaller; genitalia simi- lar to those of C. wygodzinskyi.
Holotype male: VENEZUELA: Distrito Federal: Parque Nacional El Avila, El Eden, 15 May 1975 (E. E. Dietz, Marncai/).
Additional material: A female possibly of this species is from each of the following localities: VENEZUELA: Aragiia: Rancho Grande Biological Station, Portachuelo Pass, 1100 m [10° 21' 0" N, 67° 41' 0" W], 4 June 1998 (J. Ashe, R. Brooks, R. Hanley, Laiorence); La Mora, near Colonia Tovar, 21 km north of La Victoria on Route 4, 28 May 1988 (A. L. Norrbom, G. J. Steck, Washington).
Etymology: An adjectival form based on Venezuela, the country of origin.
Chilicola (Anoediscelis) wygodzinskyi new species (Figure 9)
In a series of two males and five females of the Chilicola ashmeadi group from Pichinde, Colombia {New York), one male was found on dissection to be C. ashmeadi, the t)ther a well defined new species, here described. This species agrees with the characters listed for the group of C. ashmeadi, and the specific characters indicated in the key to species. The following characters are based on the male.
Yellow area of clypeus shiny, especially below, as in Chilicola xanthognatha, but not reaching upper margin of clypeus, lateral margin of yellow area concave, not irregu- lar; mandible black; middle flagellar segments longer than
broad; under surface of flagellum brown, distal segments about four-fifths as wide as upper width of clypeus. Tl conspicuously longer than broad; apex of S6 broadly roimded with sparse hairs (Fig. 9c), as in C. cooperi, no tuft or dense fringe as in other species of the C. ashmeadi group. S7 and genitalia as m Figures 9a, b; S7 suggestive of C. ashmeadi but upper lateroapical lobe with coarser, better separated hairs. S8 similar to that of C. ashmeadi. Penis valves not attaining apices of gonoforceps, lacking apical membranous appendages.
Holotype male: COLOMBIA: Valle: Pichinde, 1700 m, 28 Aug. 1967 (P and B. Wygodzinsky, New York).
Females bearing the same collection data may belong to this species, although a male C. ashmeadi was in the same series of specimens and Chilicola colomhiana has been taken at the same locality.
Etymology: This species is named for one of its collectors, the late Peter Wygod/insky, in recognition of his extensive work on neotropical insects.
Cliilicola [Anoediscelis) xanthostoma new species (Figure 10)
This species agrees with the characters listed for the group of Chilicola ashmeadi and indicated in the key to species. It is the only species with the labrum largely yel- low. The following are characters of the male:
Yellow area of clypeus dull, minutelv sculptured like adjacent black areas, yellow area extending full width of clypeus below, but not reaching upper margin of clypeus;
Bee Genus Chiucol\ in the Tropical Andes
15
Figs. 10-13. 10: CInlicoIa {Ai}oediscclis) xanthoshviia, male. Apical part of S7. S8 as in C. ashmcadi but with posterior margins of lateral lobes even more strongly convex. Genitalia as in C. wygodziiifh/i. S6 is missing in the preparation of the paratype of C. X(inthof.toi)in but in the undissecteci holotype S6 looks like that of S. xauthof^iwthn or possibly C. nshiiiendi. 11: C. {A.) xauthogimtba, male, a. Apical part of S7; b, S8; c. Apical part of S6. The genitalia are similar to those of C. un/godzinski/i. 12: C. (A.) pedunculata, male. a,Genitalia;b, Apical part of S6;cS8;d,S7. 13: C. {A.)cooperi,ma\e. a. Apical part of S7;b,S8;c, Apical part of S6. The genitalia are similar to those of C. m/godziiiskyi.
punctures of lower part of clypeus not sharply defined; mandible yellow, base narrowly black and apex reddish; labrum with large basal and discal area yellow; middle and penultimate flagellar segments about as long as broad;
under surface of flagellum tan, distal segments about as wide as upper width of clypeus. Legs with more than usual amount of yellow, bases and apices of mid and hind tibiae broadly yellow, front and mid tarsi yellow, hind
16
Scientific Papers, Natlkal Hisiury Museum, The Univershy oi Kansas
tcirsiis dusky with base of basitarsus vellcnv. Tl longer than broad. S6 much as in Chilicohi a^lumadi, with midapical fringe abruptly ending laterally, lateral hairs not longest, sternal margin convex. S7 as in Figure 10, with lower iipicoliitcnil lobe extended and pointed laterally, upper apicolateral lobe broad basally, tapering and pointed laterally, with lon^, coarse hairs on posterior margin. S8 as in C. asluneadi but posterior margins of lateral lobes even more strongly convex. Genitalia similar to those of C. wygodzinskyi.
Holotype male: PERU: Cuzco: Limatambo, 7 Feb. 1978 (P. M. Marsh, Washington). One male paratype, same data (Lawrence). Only the paratype carries a label indicating the collector.
Etymology: Greek .xanthos, yellow, plus stoma, mouth, with reference to the yellow clypeus, labrum, and man- dibles.
Chilicola (Anoediscelis) xanthognatha new species (Figures 3d, 11)
This species agrees with the characters listecl for the group of Chilicola ashiueadi, and the specific characters in- dicated in the key to species. The following characters are those of males; females of this species are unknown.
Yellow area of clypeus (at least lower part) smooth except for a few punctures, shining, triangular, extending from lower to upper clypeal margins, lateral margins of yellow area slightly irregular. Upper margin of clypeus, between subantennal sutures, usually shorter than dis- tance from lower end of subantennal suture to eye mar- gin (usually equal in other species of Chilicola ashineadi group). Mandible yellowish except for black base and red- dish apex. Middle flagellar segments longer than broad; under surface of flagellum tan, distal segments about as wide as upper width of clypeus. Tl little longer than broad (more conspicuously longer in other species of the group). S6 with apical fringe about 0.2 as wide as width of ster- num, smaller than fringe of C. cokvnbiana, bases of hairs along a straight median part of sternal margin. S7 as i)i Figures lla; with lower apicolateral lobes larger tluvi in C. colombiana and upper lobes with long blade-like hairs, longest hairs basally; SB as in Figure lla; genitalia similar to those of C. wygodzinskyi.
Holotype male and three male parat\'pes: PERU: \junin\: La Merced, Rio Chanchamayo, 17 June 1920 (Cornell Univ. Exp., Lawrence, except two paratypes at Ithaca). Two paratypes lack metasomas.
Additional material: PERU: \Junin\: one female, Chanchamayo, 1800 m, 1 Apr. 1940 (W. K. W|eyrauchl, Lawrence) may belong to this species. It appears to be in- distinguishable from females of Chilicola ashineadi, colombiana, etc.
Etymology: .xanthos, Greek, yellow, p\v]sgnalhos, Greek, jaw, with reference to the yellow mandibles.
Chilicola (Anoediscelis) cooperi new species (Figure 13)
Except for yellow on the lower paraocular areas, this species agrees with the characters listed for the group of Chilicola ashmeadi; see the specific characters indicated in the key to species of that group. The following are charac- ters of the male.
Clypeus entirely yellow (or in paratype with small dusky area at each side near upper margin), minutely roughened, with widely scattered punctures. Upper mar- gin of clypeus, between subantennal sutures, subec]ual to distance from lower end of subantennal suture to eye mar- gin. Lozuer fwraocular area with yellow extending up almost to leivl ofsuiinnit of clypeus, or in paratype limited to area below tentorial pit. Mandible (except for reddish apex) and labrum yellow. Middle flagellar segments about as long as broad, varying with antennal curvature; under surface of flagel- lum tan, distal segments about as broad as upper width of clypeus. Tl longer than broad. S6 with scattered marginal hairs (as in Chilicola -wygodzinskyi), without a ttift or fringe (Fig. 13c); S7 (Fig. 13a) similar to that of C. xanthognatha but with marginal bladelike hairs somewhat smaller. SB as in Figure 13b. Genitalia similar to that of C. zvygodzinskyi. Genitalia and distal sterna of both specimens were dis- sected and found to be alike in spite of the difference in lower paraocular markings.
Holotype male: PERU: Amazonas: Valley of Rio Umbamba below Chachapoyas, 1690 m, 10 May 19B2 (M. Cooper, Loiuion). Paratype male: Rodriguez de Mendoza, 1400 m, 23 May 1984 (M. Cooper, Lyme Regis). The collec- tor requested that the holotype be placed in London.
Etymology: This species is named for its collector, Mr. M. Cooper, of Lyme Regis, England, in recognition of his extensive collections of minute Hymenoptera in South America.
Chilicola (Anoediscelis) pcdwiciilata new species (Figure 12)
Except for its black face, this species agrees with the characters listed for the group of Chilicola ashmeadi; see also the specific characters indicated in the key to the species oi that group. The following characters are those of the male.
Face and mandible entireh' black, clypeus minutely sculptured like adjacent areas; middle flagellar segments longer than broad; under surface of flagellum brown, dis- tal segmeTits narrower than upper width ot chpeus. Tl about as broad as long. Apex of S6 with midapical fringe similar to that of Chilicola colombiana (Fig. 12b), bases of hairs tt)rming gently recurved band along straight middle part of sternal margin. S7 with ventral apicolateral lobe small, Iriiniguliv (Fig. 1 2d): S7 with dorsal apicolateral lobe peduncu-
Bee Genus Chilicola in the Tropical Andes
17
late, strongly expanded apically, thus completely different from other species of the C. ashmeadi group, with apical fringe of long, bladelike hairs; S8 as in Figure 12c; penis valves much shorter than expanded apices ofgonoforccps, thus genitalia (Fig 12a) also differing from other species of the group.
Holotype male: PERU: [/»»/'»]: La Merced, Rio Chanchamayo, 17 June 1920 (Cornell Univ. Exped., lot 569; Lawrence). One male paratype without metasoma, same
data as holotype but (Ithaca). The metasoma of the holo- type has been glued to the side of the micropin mount, the apex has been removed to a genitalia vial on the pin.
This species was evidently collected with specimens of Chilicola xanthognatha.
Etymology: pedunculata, Latin, pedunculate, footlike, with reference to the shape of the dorsal apicolateral lobe ofS7.
SUBGENUS HYLAEOSOMA ASHMEAD
Hylaeosoma Ashmead, 1898:284. Type species: Hylaeosomn longiceps Ashmead, 1898, by original designation. See Michener (1994:81) for note on date of publication of this specific name.
TTiis subgenus was defined by Michener (1994, 1995, 2000); it ranges from northeastern Mexico (state of Tamaulipas) to Bolivia and the state of Sao Paulo, Brazil. Michener (1995) mentioned Colombian species whose wing venation does not agree with that of other species of Hylaeosoma known at that time. Also in Andean species, Tl of females is only about as long as broad (somewhat longer in Chilicola smithpardoi), and in some species the head is only about as long as broad, in contrast to previ- ously recognized species of Hylaeosoma. For such reasons recognition of some of the Andean species as Hylaeosoma was not possible based on earlier literature, and indeed such species may support the union of the paraphyletic Anoediscelis with Hylaeosoma as suggested by Figure 2. However, these Andean species fall clearly within Hylaeosoma in its usual sense. The best character for its rec- ognition is the elongate, deeply bifid S8 of the male. (This sternum is also bifid in Oroediscelis but is relatively short; see below.) The description below is based on the Andean species (with some annotations about other species).
Body length 4.0 to 5.5 mm; forewing length 3.50 to 3.75 mm (thus smaller than species of Hylaeosoma from else- where), even smaller in Chilicola smithpardoi, body 3.5 mm, forewing 2.8 mm. Coloration: Black without yellow facial markings, or with ill-defined yellowish area in some males of C. smithpardoi and belli; antennal flagellum dark brown beneath, light brown in C. smithpardoi; tegula black to brownish black; legs black except that C. belli and smithpardoi have extreme apices of femora and bases of tibiae, and more extensive areas of tibiae and tarsi yellow- ish brown to yellow; posterior pronotal lobe in C. belli and smithpardoi yellowish brown posteriorly; wings somewhat dusky, veins and stigma black; posterior margins of metasomal segments inconspicuously pallid, preceded by brownish or in C. canei and in aequatoriensis merely dark brownish. Sculpturing: Variable, see descriptions of spe- cies. Pubescence: Sparse, dull whitish, yellowish dusky to dusky on head and thoracic dorsum; long scattered hairs of lower part of clypeus shorter than scape except as long as scape in C. aequatoriensis and canei in which lower part
of clypeus is quite bristly; weak lateral apical fasciae of white hairs on Tl to T3 or even more weakly on T4, such fasciae absent on C. aequatoriensis and almost absent on C. canei; T2 and T3 of females, T2 to T4 of males, with broad basal zones of very short, erect hairs arising from finely densely punctate, depressed zones (such basal zones not evident in C. smithpardoi); hind femoral and sternal scopae of females dense, rather long, white. Head: Extending upward well above summits of eyes so that in facial view, lateral margins of head extend upward for distance about equal to length of scape before curving strongly mesad, except in C. involuta in which lateral margins bend mesad little abo\'e summits of eyes; upper orbital tangent well below level of lower margin of median ocellus except at that level in C. involuta; frons convex; depression or fossa receiviiig antemial scape extending up and toward sum- mit of eye from each alveolus, except in C. aequatoriensis in which this depression is weak and scarcely if at all recog- nizable; mesal wall of fossa often with limited smooth ar- eas between pimctures except in C. aequatoriensis; frontal line absent or feebly suggested; frontal tubercle absent except variably developed between antennal bases in fe- males of C. aequatoriensis; clypeus and supraclypeal areas rather flat, lower part of clypeus distinctly convex, espe- cially in C. aequatoriensis, upper part of clypeus distinctly convex in profile in C. aequatoriensis and less conspicuously so in male of C. canei; eyes converging below, lower interocular distance more than half of upper interocular distance; irregular shining area along inner ocular mar- gin, especially in or near emargination; above emargin- ation extension of shining ridges often form a slender fovea-like shining depression and in C. involuta female, a well formed small fovea, less well developed in at least one female of C. caiiei; antennal sockets near middle of face, interalveolar distance twice alveolocular distance except in C. aequatoriensis and involuta in which it is less than twice alveolocular distance; upper clypeal width, between subantennal sutures, shorter than distance from lower end of subantennal suture to eye margin; lower margin of clypeus below lower ocular tangent by distance more than width of scape, this clypeal margin laterall)' elevated and curved backward at side of labrimi forming projection that in facial view looks like a strong tooth at each side of la-
18
Scientific Papers, Natl ral Histor-*- Museum. The University of Kansas
brum; labrum about 2.5 times as broad as long; maxillary palpus over half as long as prementum (in C. aeqiiatoriensis nearly as long as prementum), first two segments robust, about twice as long as broad, remaining segments longer, subequal or last segment longest (thus maxillary palpus ordinary compared to some lowland species of Hi/laeosornn in which they are greatly elongated); labial palpus exceed- ing glossa, segments subequal or second slightly shorter than others; subanterinal sutures converging below, little longer than alveolar diameter (C. aeqiiatoriensis), markedly longer, or (in C. belli, smithpanioi and umhonala) about twice as long as alveolar diameter; anterior tentorial pit not elon- gate (C. aeqiiatoriensis) or usually with shiny groove on epistomal suture extending about half way down from main site of pit toward mandibular articulation; malar space linear (i.e., absent) except about one-fourth as long as wide in C. invohita. Antennal scape about or nearly three times as long as broad, broadest in distal half; pedicel and first flagellar segment both variable in both sexes from slightly broader than long to about twice as long as broad; middle flagellar segments usually longer than broad but in female of C. aeqiiatoriensis broader than long although flagellar segments 8 to 10 each longer than broad. Thorax: Pronotal collar narrower than maximum width of flagel- lum (more than twice width of flagellum in non-Andean megalostigma group). Wing venation (Fig. 3b, c) variable, not always agreeing with subgeneric characters indicated by Michener (1994, 1995). Dorsal surface of propodeum longer than scutellum and than posterior surface of propodeum. Legs slender, basal half of hind tibia espe- cially slender. Metasoma: First segment longer than wide in males, in females about as long as wide or longer than wide in C. smithpanioi. Terga shining with well separated punctures, 11 and T3 with depressed basal zones with fine punctures, lineolations and short hairs, except in C. smithpardoi; Tl to T5 with broad impunctate, but shining lineolate marginal zones, that of T5 of female narrower. Hidden sterna and genitalia as illustrated, dorsal and ven- tral apidateral lobes of S7 closely appressed or fused, so that there may be only one rather than two such lobes on each side; S8 about twice as long as wide, deeply bifid apical ly.
In the descriptions below, most of the specific char- acters mentioned in the subgeneric description are omitted.
Key to the Andean species of the subgenus Hylaeosoma
1. Head somewhat broader than length measured from lower clypeal margin to lower margin of median ocel- lus (Fig. 4c); distal stigmal perpendicular (= line through apex of stigma perpendicular to wing margin) passing through or slightlv distal to apex of second submarginal cell (Fig. 3c); fossa for scape above anten-
nal alveolus almost unrecognizable (Fig. 4c)
C. aeqiiatoriensis
Head appearing more elongate, about as broad as long measured as above; distal stigmal perpendicular pass- ing through second submarginal cell (Fig. 3b) or through or beyond its apex in C. belli and smithpardoi; fossa for scape above antennal alveolus distinct (Fig. 4b) 2
2. Hypostomal area dull, densely finely strigose or coarsely lineolate; apical processes of SB of male broad, less than three times as long as basal widths. (Fig. 15c)
C. invohita
Hypostomal area shiny, punctures or grooves widely separated; apical processes of SB of male slender, four or more times as long as basal width. (Figs. 16b, 17c, 18b) 3
3. Propodeal triangle coarsely reticulate or with five to seven irregular longitudinal striae; smooth paired, preapical bosses of S2 to S5 of male weak, not notice- able in lateral view of metasoma (Fig. 20a); apices of male gonoforceps angulate 4
Propodeal triangle with many fine striae diverging posteriorly; smooth paired, preapical bosses of S2 to S5 of male strong, noticeable in lateral view (Fig. 20b, c); apices of male gonoforceps rounded 5
4. Middle of mesepistemum with large puntures sepa- rated by about a puncture width; S7 of male as in Fig- ure 17b C. belli
Middle of mesepistemum shining, with large punctures separated by more than a puncture width; S7 of male as in figure 19b C. smithpardoi
5. Punctures of scutum mostly separated by about two puncture widths of dull, strcingly microsculptured ground C. canei
Punctures of scutum separated by less than a puncture width of shinv although microsculptured ground (fe- male unknown, but would presumably be here)
C. iimbonata
Chilicola (Hi/laeosonia) ncqiiatorieusis Benoist
(Figures 3c, 4c, 14) Chilicola aeqiiatoriensis Benoist, 1942:81.
The depressions abo\e the antennal alveoli of this spe- cies are weaker than in any other known species of Hi/laeosoma (see description of subgenus, above) and the large stigma, extending as far as or beyond the apex of the distal submarginal cell, is also unique in the subgenus. The head is round if one ignores the great development above the eyes; other Hi/laeosoma have more elongate heads. Other specific characters are indicated below and in the descrip- tion of the subgenus.
Bee Genus Chiucola in the Tropical Andes
19
Figs. 14-15 1 5: Chilicola {Hi/lneosowa) acquatoriensis, male, a, b, Genitalia; c, S8; d, S7. 15: C. (H.) involuta, male, a, b. Genitalia; c, S8; d, S7.
20
Scientific Papers, Natural History Museum, The University of Kansas
Male: Face 1.05 to 1.08 times longer than broad, thus almost round. Head surface dull, very closely and strongly punctate above le\el of antennae, below antennae becom- ing granular, lower genal area grading into the strongly longitudinally microstriate and slightly more shiny hypostomal area. Preoccipital ridge sharply rounded but not carinate, distant from lateral ocellus by about or more than 1.5 times diameter of that ocellus. Antennal flagel- lum less than twice as thick distally as basally, first flagel- lar segment as broad as long or broader, second and third equal, much broader than long; middle flagellar segments longer than broad. Scutum, scutellum, and side of thorax dull, microsculptured, with scattered small punctures sepa- rated by several puncture widths or closer on lateral parts of scutimi; erect hairs on scutum little longer than ocellar diameter or twice as long as ocellar diameter in holotype from Quito, Ecuador; propodeal triangle with fine, close, longitudinal radiating striae not reaching apex, striae dis- tinct in holotype, often irregular producing a granular sur- face in specimens from other localities, including those in Ecuador. Wings with stigma as long as or slightly longer than length of marginal cell on costal wing margin; distal stigmal perpendicular slightly distal to or crossing second submarginal crossvein, thus at or beyond apex of second submarginal cell; first recurrent vein joining first submar- ginal cell about one-fifth or one-sixth of length of cell from its apex. Legs black. Exposed metasomal sterna simple, S2 to S5 with preapical sublateral smooth, slightly convex areas comparable to bosses or tubercles of some other spe- cies, but not noticeable in lateral view. Genitalia and hid- den sterna as in Figure 14; ventral side of ventral apicalateral lobe of T7 with strong transverse ridge bear- ing strong hairs, surface distal to ridge sloping away and hairy.
Female: In addition to usual sexual characters, differs from description of male as follows: Face about 1.14 to 1.20 times longer than broad. Frons with surface largely stri- gose; hypostomal area more shiny and anterior part only weakly microstriate. Preoccipital ridge distant from lateral ocellus by about twice diameter of ocellus. Antennal fla- gellum almost twice as thick apically as basally, middle flagellar segments broader than long.
Holotype male: ECUADOR: [Pidinicha]: Quito, 2850 m according to Benoist (1942), 27 Oct. 1930 (collected by Benoist in Riibu^ stem, Paris). I have examined the type; it seems identical to specimens recorded below except for its longer hair (see above), but 1 did not examine the geni- talia or hidden sterna of the unique type.
Additional material: PERU; Auiazomis: Chachapoyas, 2300 m, 17 Mar. 1984 (M. Cooper, L\/nic Regis). ECUADOR: Tiingurahua: one male, Baiios, 2100 m, 14 July 1981 (M. Cooper, Lyme Regis); one male, Baiios, 2400 m, 21 July 1981 (M. Cooper, Lyme Regis); one female, Baiios, 2100 m, 26
Jan. 1982 (M. Cooper, Lyme Regis); one female, trail south of Banos, 27 Jan. 1976 (Spangler et al., Washington); one female, 39 km east of Bahos, 25 Jan. 1976 at black light (Spangler et al., Washington). COLOMBIA: Anlioquia: one male, two females, Rionegro, 2120 m. May, 1975, on Frambuesa [probablv Rubiis] (J. Cano, Medelli)i); two males, two females, same locality, Sept. 1974, on Rubus except one male on Clibadium (A. Correa, Medellin). Cundinamarca: one female, 12 mi SE of Bogota, 2930 m, 13 Mar. 1955 (E. I. Schlinger, E. S. Ross, San Francisco); one female, Sueva, 2500-2700 m, 27 Apr. 1987 (R. Ospina, Bogota Univ.); one male, Usme, 2700 m, 28 Sept. 1991 (C. Sarmiento, Bogota Univ.); one male, Vereda de Cetine near Cota, 40" 49' N, 74" 6' W, 2800 m, 24 July 1983 (A. Lievano, Bogota Univ.). Valle: two females, near Pichinde, 1900 m, 3 Feb. 1977 (M. D. Breed, C. D. Michener, Lawrence). VENEZUELA: Merida: about 130 specimens of both sexes, near La Carbonera, 42.4 km NW Merida, 8" 37' 38 "N, 72" 21' 10" W, 2360 m, 25 May 1998 on flowers of Rubus (J. Ashe, R. Brooks, R. Hanley, Lawrence); one male, one female, Merida, 26 Jan. 1986 (M. Cooper, Li/me Regis); one female, two males, Merida, 17 Nov 1972 (G. E. Bohart, Logan).
Specimens have been collected in every month of the year except June, August, and December.
Chilicohi (Hylaeoscviia) invohita new species (Figure 15)
The dull, minutely striate hypostomal area is unique among Andean Hylaeosoma. The broad, tapering apical processes of S8 of this species are also distinctive. For ad- ditional characters, note the subgeneric description, espe- cially the facial foveae of the female and the presence of a small malar space in both sexes.
Male: Face about 1.23 times longer than broad. Head surface dull, very closely punctate throughout, becoming finely strigose on upper genal area and closely lineolate or finely striate below and on entire hypostomal area. Preoc- cipital ridge sharply roundcti but not carinate, distant from lateral ocellus by about 1.5 times diameter of that ocellus. Antemial flagellum less than twice as thick distally as ba- sally, first flagellar segment about 1.6 limes as long as broad, longer than pedicel; second slightly longer than broad; third about as broad as long; middle flagellar seg- ments longer than broad. Thoracic punctation as described for Chilicola aecjuatoriensis; erect hairs on scutum variable, longest about three times as long as ocellar diameter; propodeal triangle with fine longitudinal radiating striae. Wings with stigma shorter than length of marginal cell on costal wing margin; distal stigmal perpendicular crossing middle of second submarginal cell; first recurrent vein al- most meeting first submarginal cross\ein. Legs black. Ex- posed metasomal sterna as in C. aequatoriensis. Genitalia and
Bee Genus Chiucoia in the Tropical Andes
21
hidden sterna as in Figure 15. T7 with ventral apicolateral lobe thickened or folded with two posterior margins.
Female: In addition to usual sexual characters, differs from description of male as follows: Face about 1 .27 times longer than broad. Head with punctation slightly less dense than in male, genal and hypostomal areas less lineolate but dull and microsculptured. Antennal flagel- lum almost twice as thick apically as basally; second and third flagellar segments each about as long as broad; middle segments much broader than long. Erect hairs on scutum mostly about as long as ocellar diameter
Holotype male: ECUADOR: Azuny: near Mollecuro, 45 km west northwest of Cuen[c]a, 3200 m, 29 Dec. 1992 (M. Wasbaeux, Davis); one paratype female, Rio Mazan Reserve, 10 km W of Cuenca, 3150 m, 2 Ian. 1992 (C. Carlton, R. Leschen, Lawrence). The collector of the holo- type was probably M. Wasbauer. In spite of the different lengths of scutal hairs, the two specimens (male and fe- male) appear to be the same species.
Etymology: involiita, Latin, intricate, with reference to the complex folding of the lobes of S7 of the male.
Oiilicola {Hylaeosoina) innbonata new species (Figures 16, 20b)
This species, like Oiilicola canei, has strong sternal tu- bercles in the male (Fig. 20b). Among Andean Hi/laeosoiun it is unique in having the middle flagellar segments of the male broader than long, the antenna suggesting that of a female although 13-segmented. The constriction near the middle of S8 of the male is distinctive (Fig. 16b). For addi- tional characters, note the subgeneric description as well as the specific description below:
Male: Face about 1.25 times longer than broad. Head rather uniformly strigose-punctate, limited areas between punctures minutely roughened but shining, lower gena more shining with punctures more widely separated, sur- face grading into hypostomal area which is anteriorly with- out punctures, highly shining, even though at certain angles weakly lineolate. Preoccipital ridge sharp, almost carinate, with depressed zone across head anterior to ridge, ridge distant from lateral ocellus by about 1.5 times diam- eter of that ocellus. Antennal tlagellum like that of females, about twice as thick apically as at base, first flagellar seg- ment 1.75 times as long as broad, as long as pedicel; sec- ond and third segments subequal, each about as long as broad; middle flagellar segments broader than long. Scutum and scutellum with punctures in some areas of scutum separated by a puncture width, elsewhere by less, ground between punctures shining but minutely rough- ened; sides of thorax duller with minute sculpturing, punc- tures scattered and inconspicuous; longest erect hairs of scutum over 1 .5 times length of ocellar diameter; propodeal
triangle with fine, close, longitudinal radiating striae. Legs brownish-black, "knees" brownish, basal fifth of hind tibia testaceous (but see "Additional material"). Wings with stigma shorter than length of marginal cell on costal wing margin; distal stigmal perpendicular passing through sec- ond submarginal cell; first recurrent vein almost meeting first submarginal crossvein. S2 to S5 each with preapical, sublateral smooth, shining elevated area or tubercle, clearly visible in lateral view (Fig. 20b), these areas of S5 fused to form transverse ridge. Hidden sterna as in Figures 16b,c; genitalia (damaged in paratype) similar to those oi Oiilicola iitzKiliita but lacking clumps of large hairs laterally on gonoforceps, and volsella shaped as in Figure 16a.
Holotype male and one male paratype: COLOMBIA: Valle: Summit west of Cali, 2000 m, 31 Jan. 1977 (M. Breed, C. D. Michener, Lawrence). One male paratype: Same lo- cality, 17 Sept. 1976 (Bell, Breed and Michener, Lawrence).
Additional material: One male from ECUADOR: Loja: La Argelia, 4" 2.023' S, 79" 11.765' W, 2350 m, Feb. 20, 2001 (C. Rasmussen, Lima) differs from Colombian material in the more finely punctate scutum, slender tlagellum, yel- lowish mandible, and on the legs the following parts yel- low: apices of femora, bases of tibiae (about one-third of hind tibia), apices of tibiae, and tarsi except blackish distal segments. One female from a nearby locality, Malacatos, Feb. 26, 2001 (C. Rasmussen, Lima) agrees in these features with the male except that the front tarsus is almost all yel- low, the hind tibia has the basal yellow zone only one- fourth as long as the tibia and the hind tarsus is mostly blackish. These specimens probably represent another spe- cies but the genitalia and sterna of the male are so similar to those of Colombian specimens that recognition of an- other species on the basis of the male seems premattire. VENEZUELA: Aragua: one female, Rancho Grande, 1100 m, 17-20 Jan. 1978, black light, cloud forest (J. B. Heppner, Washington). It differs from Oiilicola innbonata males in the larger stigma, the distal stigmal perpendicular falling im- mediately distal to the second submarginal crossvein. This character usually does not differ between sexes, and I sup- pose this female to represent a new species.
Etymology: umhonata, Latin, knobed or bossed, with reference to the prominent sternal bosses of the males.
Oiilicola {Hylaeosoina) canei new species
(Figures 3b, 18, 20c)
Like Oiilicola innbonata, the male of this species has strong tubercles on the metasomal sterna. C. canei differs from C. nmbonata in the scutal and scutellar punctures which are widely separated by dull ground as in C. aeqiiatoriensis. Some specific characters are indicated in the subgeneric description, as well as in the specific descrip- tion below.
22
Scientific Papers, Natural History Museum, The University of Kansas
Figs. 16-18. 16: ChiIicok{Hi/laesoma)umbonatn, male, a, Volsella; b, S8; c, S7. 17; C. {H.) belli, male, a. Distal part of gonoforceps, ventral view, genitalia otherwise similar to those of C. acijuntoricnsis; b, S7; c, S8. 18; C. (H.) canci, male, a. Genitalia; b, S8; c, S7.
Bee Genus Chiijcol\ in the Tropical Andi;s
23
Male: Face about 1.12 times longer than broad. Head surface dull, densely strigose punctate, below level of an- tennae punctures scarcely recognizable, surface being stri- gose granular; upper genal area with punctures separated by shiny ground, this grading into shiny hypostomal area with few, irregular punctures and only feeble lineolation, mostly posteriorly. Preoccipital ridge sharp, almost cari- nate, distant from lateral ocellus by about ocellar diam- eter. Antennal tlagellum about twice as thick apically as basally, first flagellar segment over 1.5 times as long as wide, second segment slightly longer than broad, third as long as broaci; micidle segments slightly longer than broad (perhaps as long as broad in straightened flagellum). Tho- racic punctation as described for C. aequatoriensis; erect hairs on scutum almost as long as two ocellar diameters; propcideal triangle with numerous fine radiating striae diverging from finely granulostriate base, and becoming weak on rather shiny marginal area; one longitudinal me- dian stria particularly strong. Wiiigs with stigma shorter than length of marginal cell on costal wing margin; distal stigmal perpendicular cutting through second submarginal cell; first recurrent vein joining first submargiiial cell about one-eighth of length of cell from its apex. Legs black. S2 to S5 each with preapical, sublateral, smooth, shining elevated area or tubercle clearly visible in lateral view, elevations of S5 fused to form transverse ridge. Genitalia and hidden sterna as in Figure 18.
Female: In addition to usual sexual characters, differs from description of male as follows: Face about 1.23 times longer than broad. Frons in front of ocelli with some shin- ing ground between punctures visible on convexities; clypeus with distinct, often round punctures on minutely roughened ground (upper clypeal surface, seen in profile nearly flat); hypostomal area shining but more conspicu- ously lineolate than in male. Preoccipital ridge separated from lateral ocellus bv abc^ut 1.3 ocellar diameters. Third flagellar segment broader than long; middle segments broader than long. Propodeal triangle without strong me- dian stria. First recurrent vein almost meeting first sub- marginal crossvein.
Holotype male: COLOMBIA: Aiitioqiiia: 1 km north northwest of Sonscin, 21 Jan. 1984 (James H. Cane, Linvroice).
Additional material: COLOMBIA; Aiitioqiiia: four fe- males, 15 km west of Medelh'n, 18 Jan. 1984 (James H. Cane, Lawrence). These females might be a different (new) spe- cies since they are from another locality and differ from the male in the longer head, flatter clypeus, clypeal punc- tation, wmg venation, and other characters as indicated above. They are not paratypes.
In the holotype, T7 was only slightly exposed and the apical metasomal sclerites were retracted so that the metasoma seemed blunt (Fig. 20c). These features disap-
peared with dissection of the terminalia and are probably not specific characters but rather depend on the condition of the specimen when it died.
Etymology: This species is named after its collector and authority on bee biology, James H. Cane of the USDA Bee Biology and Systematics Laboratory at Utah State University, Logan.
Chilicoln {Hi/laeosoma) belli new species (Figures 17, 20a)
Among Andean Hylneosonia, this species resembles Cliilicohi siiiitlipnnioi in the coarsely reticulate or coarsely striate propodeal triangle with only about five or seven irregular striae plus about two shorter striae on each side. The pallid "knees" and apices of the tibiae are also dis- tinctive among Andean Hylaeosoma although in C. siiiitlipnnioi these areas are more extensive and yellow. In addition to the characters described below, note the other specific characters incorporated into the description of the subgenus.
Male: Mandible and labrum largely pale yellow. The following parts brownish yellow: posterior part of pronotal lobe, extreme apices of femora, bases and apices of tibiae (on front leg extendiiig along outer side of tibia, on hind leg basal fourth of tibia pale yellow), tarsi (dusky except pale base of hind tarsus). Face about 1.16 times longer than broad. In addition to small lateral tergal fasciae of white hairs (see subgeneric description), band of short white hairs in groove between pronotum and scutum. Head surface dull, very closely punctate throughout, only on lower ge- nal area grading into shining hypostomal area with a few weak punctures and weak lineolation. Preoccipital ridge sharp, almost carinate, with depressed zone across head anterior to ridge, ridge distant from lateral ocellus by about 1.5 times diameter of that ocellus. Antennal flagellum al- most twice as wide apically as basally (segment 2), first flagellar segment 1.5 times as long as broad, second and third subequal, about as long as broad; middle flagellar segments longer than broad. Scutum, scutellum, and side of thorax with strong punctures mostly separated by less than a puncture width of shining but minutely sculptured ground; erect hairs on scutum little longer than ocellar di- ameter; propodeal triangle coarsely reticulate, without numerous fine striae found in other species, but reticula- tion reducible to about five irregular longitudinal striae with one or two shorter ones on each side. Wings with stigma shorter than length of marginal cell on costal wing margin; distal stigmal perpendicular slightly basal to sec- ond submarginal crossvein, thus within second submar- ginal cell; first recurrent vein joining first submarginal cell about one-sixth of length of cell from its apex. Exposed metasomal sterna almost simple, S2 to S4 with distinct
24
Scientific Papers, Natural History Museum, The University of Kansas
Fig. 19. Chilicola (Hi/laeosomn) sinithpnnioi, male, a. Genitalia; b, S7; c, S8.
preapical sublateral smooth convex areas resembling bosses or tubercles of some other species but so low as to be scarcely recognizable in lateral view (Fig. 20a); S5 with similar but scarcely elevated areas. Genitalia and hidden sterna as in Figure 17a to c, S7 with ventral lobe having transverse ridge as in Chilicola. aeqiiatoriciisis; genitalia simi- lar to those of C. aequatoriensis but gonoforceps with apex pointed and ventobasal lobe truncate.
Female: In addition to usual sexual characters, differs from description of male as follows: Mandible and labrum black. Other pale areas of male less evident, more brown- ish testaceous, hind tarsus entirely dusky. Face about 1.14 times longer than broad. Punctures of face and vertex somewhat less close, mostly somewhat elongate, with a little shining ground bet\veen them. Preoccipital ridge dis- tant from lateral ocellus by about diameter of that ocellus. Antennal flagellum more than twice as wide apically as basally; third flagellar segment broader than long; middle flagellar segments broader than long. Propodeal triangle with five striae (about seven in specimen from Rionegro) much less irregular than in male, longitudinal rather than diverging as in other species.
Holotype male and one female paratype: COLOMBIA: Valle: 10 km west of Call, 1400 m, 19 Sept. 1976 (boll, lireed and Michener, Laiurence).
Additional material: COLOMBIA: Autioqiiia: one fe- male. La Selva, Rionegro, 2120 m, October 1974, on flow- ers of Borrcria (A. Vasquez, Medellin). VENEZUELA: Merida: one female, Merida, 17 Nov 1972 (G. E. Bohart, Logan); Tnijillo: two males and one female. La Mesa, 11 Nov.
1973 (R. M. Bohart, Logan). The specimen from Merida lacks the pale hair in the grocn'e between the pronotum and the scutum. The specimens from La Mesa ha\ e more pale color on the legs than in the types, the tarsi being brownish to almost testaceous and the trochanters of one of the males being yellowish; in this last indi\'idual there is also a yel- low area on the lower part of the clypeus. In all the addi- tional specimens the striae of the propodeal triangle are rather straight, as in the paratype, not so irregular as in the holotype.
Ftymologv: This species is named in honor of the late William J. Bell of the Department of Entomology, Univer- sity of Kansas, one of the collectors, in recognition of his intense interest in tropical insects and insect behavior, and his successful career as a now much missed professor of Entomology at the University of Kansas.
Chilicola (Hylaeosoma) smithpardoi new species (Figure 19) This is the smallest known species of Hylaeosoma (see description of the subgenus) and the only one other than Chilicola belli with a yellowish area on the clypeus of some males. It is similar in various ways to C. belli as indicated at various points in the description of the subgenus. It dif- fers from C. belli by characters indicated in the kev to the species of Hylaeosoma, as well as by the more extensive yellow markings on the legs, the less pointed apex of the male gonoforceps (Fig. 19a), etc. In addition to the charac- ters listed below, note the other specific characters incoporatcd into the subgeneric description.
Bee Genus Chilicula in the Tropical Andes
25
Fig. 20. Lateral views of metasomas of males, a, Clulicoln {H\/lacosoiiin} belli; b, C. (H. ) innhmata; c, C. (H.) canci. Drawings by Robert W. Brooks.
Male; Lower part of the clypeus with ill-defined trian- gular yellow area in holotype, absent in some paratypes, median point (upper part brownish) extending to upper third of clypeus in holotype; the following parts yellow: labrum, mandible (except dark apex), posterior part of pronotal lobe, apices of femora, bases and apices of tibiae (rest of front tibia brownish yellow, basal third to almost half of hind tibia yellow), tarsi (distal segments reddish yellow on mid and hind legs). Face about 1.21 times longer than broad. Hairs of head and thorax mostly dusky, only about half as long as ocellar diameter, about as long as ocellar diameter and pale on upper gena, preoccipital ridge, mesepisternum, and on metasoma where thev form weak, whitish lateral tergal fasciae (see subgeneric description). Head surface closely punctate and therefore dull except for small shiny impunctate area lateral to lateral ocellas and shiny interspaces (mostly less than a puncture width) on lower genal and hypostomal areas. Preoccipital ridge
sharp, almost carinate, strong depression across head an- terior to ridge, which is distant from lateral ocellus by about 1 .25 ocellar diameters. Antennal flagellum about 1.5 times as wide apically as basally (segment 2), first and third flagellar segments about as long as broad, second broader than long, middle and preapical flagellar segments about 1.5 times as long as broad. Scutum, scutellum, and side of thorax with strong punctures mostly separated by half a puncture width (more on mesepisternum) of shining ground; erect hairs on scutum inconspicuous, about half as long as ocellar diameter; propodeal triangle coarsely reticulate, with about five longitudinal striae or only me- dian stria recognizable. Wings with stigma slightly shorter than length of marginal cell on costal wing margin; distal stigmal perpendicular meeting or slightly distal to second submarginal crossvein; first recurrent vein joining first submarginal cell about one-fifth or one-sixth of length of cell from its apex. Exposed metasomal sterna almost simple, S2 to S4 as described for Chilicola belli, smooth el- evated areas scarcely evident on S4 and S5. Genitalia and hidden sterna as in Figure 19; note angulate (but not acute as in C. belli) apex of gonostylus and blunt, peglike setae along anterior margin of apicolateral lobe of S7.
Female: In addition to the usual sexual characters, dif- fers from description of male as follows: Clypeus and la- brum black, mandible reddish yellow. Middle part of front tibia dusky; hind tibia with basal fourth to half yellow; tarsi yellow, distitiarsi of mid and hind legs dusky. Face about 1.17 times longer than broad. Head surface not quite so closely punctate, especially in ocellar area punctures well separated by shining but minutely roughened ground. Preoccipital ridge distant from lateral ocellus by 1.5 ocel- lar diameters. Antennal flagellum twice as wide apically as basally (segment 2), first segment slightly longer than broad, second about as broad as long, segments 3 to 7 much broader than long, 4 to 6 twice as broad as long, segments 8 and 9 about as broad as long or slightly broader than long. Propodeal triangle dull with median stria.
Holotype male: COLOMBIA: Antioquia: Municipio de Amalf, Canon del Rio Force, Santa Lucia, 6° 46' 20.4" N, 75" 6' 8.4" W, 1125 m, in rastrojo alto (secondary forest), collected with a net (Jama) 10-12 am Aug. 28, 1997 (Allan Smith-Pardo, Mcdelliii). One male and two female paratypes: Same data as holotype. Four female paratypes (one headless): Same data as holotype but 2-4 pm (one, Lawrence). Two male and one female paratypes: Same data as holotype but 2-4 pm, Sept. 17, 1997 (one of each sex, Lawrence). One female paratype: Same data as holotype but collected on decomposed fish bait, 2-4 pm {Nezv York.) (Except as otherwise indicated, all specimens are at Meiiellin.)
Et\'mologv: Named for Allan Smith-Pardo who col- lected all the material and helped to arrange for the loan.
26
SciENTinc Papers, Natural History Museum, The University oi- Kansas
Note on extraijmitai.. Hylafosoma
While on the subject of South American Chilicoln, it is worth recording major range extensions for the subgenus Hylaeosoma.
In an unpublished thesis (Wilms, 1995, p. 183), Chilicola (Hylaeosoma) megalostigma (Ducke) is recorded from the Boraceia reser\'e in the Serra do Mar of Sao Paulo, Brazil. Wilms' specimens {Sao Paulo) were compared by me with specimens from the type locality, Serra de Baturite, Ceara, collected by Ducke (Siio Paulo), and appeared idenrical. This is some 2300 km south of the Ceara locality (northeastern
Brazil), which is the previously known southern record for the subgenus. Specimens from Boraceia were dated January 20 and 31, 1994, and were taken by Wilms on flow- ers of a cactus, Rhipsalis capilliformis.
Secondly, in the collection of M. Cooper {Lijme Regis) there are specimens of the Chilicola megalostigma group from Puerto Bermudez, Pasco, Peru, 800 m, and from Rurrenabaque, La Paz, Bolivia, 270 m. Thus, Hylaeosoma ranges southward not only in the Serra do Mar of Brazil but also in the margin of the Andean uplift, at elevations lower than those where tlie distinctive Ajideaii fauna occurs.
OREODISCELIS NEW SUBGENUS
Type species: Oediscelis sti/livcutris Friese, 1908.
In the key to subgenera of Chilicola by Michener (1995), species of this group run to Anoediscelis with some diffi- culty. The epistomal suture below the tentorial pit is strong and could be interpreted (incorrectly, 1 believe) as an ex- tension of the pit. Avoiding this difficulty, one reaches cou- plet 6. In C. styliventris the male hind tibia is clearly longer than the femur so that when folded it reaches or nearly reaches the base of the trochanter. This species was there- fore placed in Anoediscelis by Michener (1995), although with the comment that the hind tibia of the male is more modified than in other species of Anoediscelis. In some spe- cies described below, however, the hind tibia is even more enlarged and modified than in C. styliventris. It is longer than the femur, but partly because of its enlargement, it can only be folded to reach the middle of the trochanter, thus for this measurement more nearly resembling the species of the subgenus Oediscelis. The greatly enlarged hind femur of male Oediscelis (sensu Michener, 1995) dif- fers from both Anoediscelis and Oroediscelis.
In the key to subgenera of Chilicola by Toro and Moldenke (1979), one encounters problems with species of Oroediscelis at couplet 3. Females of Oroediscelis have the first flagellar segment longer than the pedicel (or as long as the pedicel in C. simplex and C. transversaria) , thus lead- ing to Idioprosopis (included in Oediscelis by Michener, 1995), while males of some species of Oroediscelis have the first flagellar segment shorter than the second and run to Anoediscelis.
Some Chilean species of the subgenus Anoediscelis [Chilicola inermis (Friese), mailen Toro and Moldenke, plebeia Spinola] have the distal half of the hind tibia of the male somewhat enlarged (see Toro and Moldenke, 1979), sug- gesting the more elaborate condition of Oroediscelis. In Anoediscelis, however, the tibia is not thickened to support irmer and outer crests; there is a single ventral ridge or crest. Moreover, the male basitarsi of Anoediscelis are simple, unlike those of Oroediscelis. The presence of a dis- tinct malar space in Oroediscelis easily differentiates it from Anoediscelis.
No species of Oroediscelis is recorded from Chile; thus characteristics of this group were not considered by Toro and Moldenke (1979) in their seminal work on the Chil- ean Xeromelisstnae. The subgenus is known only in the Andes from Bolivia and Peru to Venezuela. The principal characters of the subgenus are enumerated below.
Body length 5.5 to 8.5 mm; forewing length 3.5 to 5.7 mm. Coloration: Black, face of male sometimes black but frequently with transverse apical yellow clypeal mark and with yellow on lower paraocular area; base of mandible of male rarely with yellow spot; flagellum sometimes dark brown on under side; tarsi rarely partly dark brown; ante- rior tibia with anterior surface yellow in Chilicola bigibbosa and maculipes, and sometimes with small yellow streak in C. brzoskai; posterior margins of metasomal terga translu- cent or sometimes brownish, sterna partly dark brown, such brown coloration especially evident in C. bigibbosa; wings somewhat dusky, veins and stigma black or dark brown in C. simplex. Sculpturing: Variable, see specific descriptions. Pubescence: Sparse, whitish, dusky in some lights on vertex and dorsum of thorax, longest, densest, and whitest on lower genal area, venter of thorax, and else- where as noted in specific descriptions; posterior margins of Tl to T4 or T5 usually with weak lateral apical fasciae of pale hairs. Scopa sparse and weak on hind femora, lim- ited to a few long plumose hairs on SI; S2 with areas of long, plumose hairs laterally, these hairs converging over impunctate and hairless area medially, perhaps forming a corbicula; S3 without scopal hairs, its vestiture similar to that of S4. Head: Not greatly developed above eyes as in Hylaeosoma; upper orbital tangent at or above level of lower margin of anterior ocellus; preoccipital ridge behind ver- tex sharp, not carinate, overhanging, distant from lateral ocellus by about 1.5 ocellar diameters (or near 1.75 ocellar diameters in C. cuzcoensis); frons weakly convex; fossa for scape above alveolus absent; frontal line absent or feebly indicated; frontal tubercle absent or small, between anten- na! bases; clypeus and supraclypeal area rather flat in pro- file, lower part of clypeus somewhat convex, upper part of supraclypeal area somewhat protuberant; eyes converg-
Bee Genus Chiucola in the Tropical Andes
27
ing below but lower interocular distance much more than half of upper interocular distance; emargination of inner eye margin rather broad and shallow; surface of face at and slightly above ocular emargination in female with punctures separated by shiny ground, forming undefined shiny area; antennal sockets near middle of face; interalveolar ciistance equal to or slightly less than alveolocular distance; upper clypeal width, between subantennal sutures, less than to about equal to distance from lower end of subantennal suture to eye margin; lower margin of clypeus below lower ocular tangent by distance considerably more than width of scape; labrum about 2.5 times as wide as long; maxillary palpus over half as long as prementum, first two segments shortest but second sometimes over twice as long as broad, remaining seg- ments longer, last sometimes longer than others; labial palpus exceeding glossa, second segment shorter than oth- ers; subantemial sutures converging below, about 1.5 times (or in the female of C. bigibhosa nearly twice, and interme- diate in C. gibbosn and brzoskai) as long as diameter of al- veolus; anterior tentorial pit not noticeably elongate (see introductory remarks on this subgenus); malar space about 0.4 to 1.2 times as long as broad (Fig. 4d). Antemial scape widest at apex, about four times as long as broad; pedicel in male about as long as broad, in female longer than broad; male with flagellum scarcely thickened distally, almost parallel sided, with a tendency to be weakly crenulate; first flagellar segment at least two-thirds as long as second, sometimes as long as second; second as long as third, seg- ments of basal half all much longer than broad, distal half with segments shorter but usually at least a little longer than broad; female with tlagellum less than twice as thick distally as basally, first flagellar segment 1.5 to nearly 2.0 times as long as broad, shorter in C. transversaria, longer than second and third segments which are broader than long, as are all subsequent segments except the last. Tho- rax: Pronotal collar about as wide as flagellum. Forewing (Fig. 3a) with stigma shorter than length of marginal cell on wiiig margin, distal stigmal perpendicular basal or dis- tal to level of first submarginal crossvein; first recurrent vein nearly meeting first submarginal crossvein. Dorsal surface of propodeum as long as scutellum or slightly less, shorter than posterior surface of propodeum, propodeal triangle extending onto posterior surface. Middle femur of male with basal concavity on under side (Fig. 29a), some-
Fig. 21 . Diagrams of processes of S4 of males of Chilicoln {Oroeiiiscclis). At left, lateral views with apex of sternum indi- cated diagrammaticalv by a black dot; at right, ventral views, a, C. espclcticola; b, C. brooksi; c, C. benoistiniia: d, C. qiiiteiisif: e, C. cuzcocusis; f, C. bigibbosa: g, C. maculipcs; h, C. sti/liventris: i, C. transivrsnria (ventral view as in C. gibbosn); j, C. gibbosa; k, C. brzoshii (ventral view as in C. gibbosn); 1, C. simplex.
~k^
\J
\J
28
Scientific Papers, Natural History Museum, The University of Kansas
times accentuated by basal projection, especially in C. bigibbosn (Fig. 28a). Posterior leg of male with femur mod- erately swollen, tibia and basitarsus modified, tibia longer than femur and when flexed reaching to middle or basal part of trochanter; tibia with upper outer surface broadest subapically where it bears a preapical brush of rather long hairs often largely hiding slender distal part of tibia, this brush almost absent in C. bigibhosa, and iiuiciilipL's; under surface of tibia hairless, shining, distal half or third of tibia with a concavity between outer and inner parallel crests except outer crest low and concavity shallow or absent in C. bigibbosa, brzoskai, maculipes and simplex. Metasoma: First segment about as broad as long. Terga with depressed apical shining or microlineolate zones. S4 of male with a pair of preapical processes or large tubercles (Fig. 21). Geni- talia and hidden sterna of male as shown in figures; S7 with ventral apicolateral lobes large and complex, dorsal lobes with distal parts small, slender, heavily sclerotized; S8 medially constricted, apex broadly bifid.
Etymology: Owediscelis from oreo- Greek, mountain, plus Oediscelis, a major genus-group name in the Xeromelissinae, with reference to the restriction of this subgenus to the Andes.
Illustrations of Oreodiscelis
In both genitalic and leg illustrations, certain points indicated by letters are the same in the different views il- lustrated. For the legs, A marks the distal end of the inner crest of the hind tibia; B marks the distal end of the outer crest, C and D are locations, often convexities, on the outer and inner tibial crests respectively. X marks the projections or angle on the outer surface of the hind basitarius while Y marks the projections or angle on the under surface. Hair tufts on the leg are diagrammatic; numerous other hairs are omitted. A dotted line on the tibia represents a rounded rather than a sharp or carinate ridge. Dotted areas on the legs are testaceous to yellow. The position of the leg is im- portant in comparing specimens with drawings. A few degrees of rotation makes a difference in visible crests. A, B, and C on the drawings of genitalia mark membranous lobes of the penis valves, while P marks a penis valve. Dotted areas on genitalic figures are membranous pro- cesses of the penis valves.
Key to the Species of the Subgenus Oroediscelis
Males:
L Processes of S4 relatively long so that emargination formed by them is deeper than wide (Fig. 21a-e, h) .. 2
Processes or tubercles of S4 shorter so that emargin- ation formed by them is wider than deep (Fig. 21g, i-1) ' 6
2. S4 with projections converging to preapical region be-
yond which they diverge (Fig 21 e), divergent parts strongly flattened; middorsal parts of Tl to T4 with punctures minute, inconspicuous, separated by several puncture widths; clypeus with yellow subapical area C. cuzcoensis
S4 with projections nearly parallel, not strongly flat- tened apically except in Chilicola qiiitcnsis (Fig. 21a-c, h); middorsal parts of Tl to T4 with punctures coarser, conspicuous, in at least some areas with punctures sepa- rated by one puncture width; clypeus black or fre- quently with yellow subapical area in C. styliventris. 3
3. Processes of S4 in lateral view with apices strongly bidentate, the teeth of about equal size (Fig. 21 h); first flagellar segment on upper surface slightly shorter than second (distinctly shorter on lower surface)
C. styliventris
Processes of S4 in lateral view not toothed or with small upper preapical tooth; first flagellar segment on upper surface as long as second (commonly shorter on lower surface) 4
4. Hind basitarsus with thickening on outer surface stron- gest near middle, forming rounded convexity as seen in ventral view (Fig. 25b) C. quitensis
Hind basitarsus with thickening strongest on outer surface basal to middle, where it is abruptly cut off to form premedian angular tooth as seen in ventral view (Figs. 23f and 24c) 5
5. Inner crest on under surface of hind tibia ending in an apical right angle, and with another angle at midlength of tibia (A and D in Fig. 24d) C. beuoistiana
Inner crest on under surface of hind tibia ending in erect, acute, projecting angle and without angle at midlength of tibia (A in Fig. 23g) C. brooksi
6. Hind basitarsus with two strong projections or lobes on lower margin (Figs. 27e and 30b) (malar space at least nearly as long as broad) 7
Hind basitarsus with basal iialf thickened and some- times biconvex or with one strong median projection on lower region but without two strong projections or lobes 8
7. Base of frt)nt femur with \ entral projecting lobe (Fig. 28b), of middle femur with basal tooth followea by emargination (Fig. 28a); S2 with sublateral tubercles not connected by strong ridge, smaller than tubercles of S4 in lateral view; S4 with tubercles reaching sternal mar- gin (Fig 21f) C. bigibbosn
Bases of femora unmodified except for concavity at base of middle femur (Fig. 29); S2 with sublateral tubercles
Bee Genus Chilicola in the Tropical Andes
29
connected by strong ridge, these tubercles larger than those of S4 as seen in lateral view; S4 with tubercles arising from sternal surface such that neither expanded
bases nor apices reach sternal margin (Fig 21g) C.
uuKuUpes
8. Malar space nearly as long as wide; S4 with projections slender, pointed in lateral view (Fig. 21a); lower paraocular areas with yellow C. espeleiicola
Malar space half to two thirds as long as wide; S4 with projections robust, like strong tubercles; face black or rarely lower paraocular areas with yellow 9
9. Hind basitarsis with lobe (Fig. 32b) projecting near middle of lower margin; projections of S4 long enough to reach sternal margin if they could be pressed down (Fig. 21i, j) ; 10
Hind basitarsus with or without weak thickening of middle of lower margin (Fig. 33b, 34f); projections of S4 mere tubercles, not long enough to reach sternal margin (Fig. 21k, 1) 11
10. S2 with sublateral tubercle with crest obliquely trans- verse, appearing acute or right angular in lateral view
C. transvenaria
S2 with sublateral tubercle with crest longitudinal, broadly rounded in lateral view C. gihbosa
11. Hind basitarsus almost unmodified; hind tibia with apex of inner crest strc>ngly acute (Fig. 34e, f)
C. simplex
Hind basitarsus with median convexity on outer sur- face (Fig. 33a); hind tibia with apex of iiiner crest usu- ally rounded C. brzoskai
Females (unknown for Cliilicola ciizcocnsis and C. iimcidipi's):
1. Malar area about as long as broad 2
Malar area about half as long as broad 3
2. Long hairs of scutum nearly as long as scape
C. espeleticoln
Long hairs of scutum about half as long as scape
C. bigibbosa
3. Body length 6.5 to 7.0 mm C. beiioistiniia, quitoisis
Body length 5.0 to 6.0 mm 4
4. Upper two-thirds of clypeus with fine microsculpturing but rather shiny and with conspicuous punctures sepa- rated by a puncture width or more C. simplex
Upper two-thirds of clypeus dull with microsculpturing, with punctures weak, not well de- fined, and widely separated
C. brzoskni, gibbosa, st\/liventris, traiisversaria
Chilicola {Owediscelis) espeleticola new species (Figures 1, 3a, 4d, 21a, 22, 29)
TTiis is one of the largest species of the genus, equal in size to Chilicola brooksi and slightly larger than C. qiiitensis. It differs from other species of Oroediscelis except C. bigibbosa and maculipes in the large malar space and the more produced clypeus. Those species can be distin- guished from C. espeleticoln by the modification of the hind basitarsus of the male (Fig. 22e-g) and the slightly longer malar space.
Male: Length 6.5 to 8.5 mm; forewing length 5.5 to 5.7 mm. Face about 1.1 times longer than broad. Clypeus black; lower paraocular area beside clypeus with yellow mark. Head surface dull, closely anci finely punctate above antennae, lower face dull with microsculpture, with well separated inconspicuous punctures, except lower third of clypeus with large punctures; malar space nearly as long as wide; lower ocular tangent near middle of clypeus; ge- nal area finely strigose punctate above, grading into coarsely lineolate hypostomal area which becomes smoother anteriorly, finally nearly smooth and shiny in anterior one-fourth. First flagellar segment as long as sec- ond, slightly shorter on lower (or anterior, i.e., shortest) side, each nearly twice as long as broad; last three or four flagellar segments subequal, each slightly longer than broad or segments 9 and 10 sometimes as long as broad. Thorax dull, strongly microsculptured, dorsally with small punctures separated by a puncture width or more, on mesepisternum punctures coarser. Dorsal surface of propodeum shorter than scutellum, finely granular with many fine longitudinal striae, often rather irregular. Long- est erect hairs on scutum and scutellum as long as scape, those on side of thorax and lower genal area even longer. Stigma over half as long as marginal cell on wing margin, distal stigmal perpendicular basal to first submarginal crossvein or meeting that crossvein. Hind tibia (Figs. 22e- g) nearly three times as long as greatest width, at preapicaj brush; bare area of outer surface shining but with minute microsculpturing; inner crest with strong marginal carina ending in a broad rounded area (A) folded on outer side; hind basitarsus (Fig. 22e-g) nearly as long as remaining tarsal segments together, widest slightly before middle because of low broad rounded expansion of lower mar- gin, this expansion in ventral view extending little beyond outline of basitarsus; outer surface of hind basitarsus dis- tally concave, shining but with fine microsculpturing. Metasoma dull microlineolate, punctures small, separated by a puncture width or more on Tl to T3, denser on more posterior terga; posterior marginal zones of terga not punc- tate but lineolate, brownish, laterally partly covered by pale hairs forming weak lateral fasciae. S2 and S3 each with two preapical sublateral shining bosses, S2 with bare area
30 Scientific Papers, Natural History Museum, Thh Univi;rsitv oi- Kansas
A
Bee Genus Chilicol\ in the Tropical Andes
31
medially and long hairs laterally, suggesting female scopa; S4 with two slightly tapering projections (Fig. 21a), flat- tened at tips and sharply pointed in lateral view and apically curved ventrally, emargination between them much broader than deep, depressed area between them with short, erect hairs, lateral to them long, incurved hairs; S3 with apical fringe represented only by short, inconspicu- ous hairs, lateral hairs long, cur\'ed, erect; S6 with surface mostly shining and hairless, with broad median depres- sion defined laterally by rounded ridge, lateral to which is deep depression, at extreme side dense tuft of yellow hairs, apical margin with fringe of well spaced hairs surpassed by transparent rounded lamella. Genitalia and hidden sterna as in Figure 22a-d; S7 with posterior margin of apicolateral lobe membranous and sometimes irregular (right side. Fig. 22c), midlateral process of same lobe form- ing rounded dark plate that may hang down (Fig. 22c) or project more laterally (Fig. 22d); S8 similar to that of Chilkola bivoksi; genitalia differing from those of C. brooksi, especially in the two long membranous processes from the apex of each penis valve and the less hairv gonoforceps (Fig. 22a, b).
Female: Agrees with description of male except for usual sexual features and the following: Length 6.0 to 8.0 mm; forewing length 4.6 to 5.4 mm. Face without yellow markings. First flagellar segment over 1.5 times as long as broad (see description of female antenna in account of subgenus). Mesepisternal punctures similar to those of thoracic dorsum. Posterior marginal zones of metasomal terga with punctures and strong lineolation extending nearer to tergal margins.
Holotype male and 42 male and 93 female paratypes: VENEZUELA: Mcridn: Paso Pico Aguila, Paramo de Mucuchies, 8° 51' 5" N, 70" 48' 34" W, 3710 m, 26 May 1998, in dead dry flowering stems of Espeletia sclniltzii (J. Ashe, R. Brooks, R. Hanlev, Lawrence, with paratypes distributed to Dnvis, Oiiawa, Los Angeles, Londo}i, New York, Paris, Sivi Francisco, Washington). One additional male paratype, same data but 21 May 1998 and not from Espeletia stems but caught on flowers (R. Broc~)ks, pers. comm.) (Lawrence). Additional material: VENEZUELA: Merida: two fe- males. Paramo Las Cruces, 3900-4300 m, 11 Sept. 1982 (P Berry, Ithaca); three males, four females. Paramo de Mucuchies, Alto de Timotes, 8" 51' 24" N, 70" 49' 30" W,
Fig. 22. CliiUcola (Oroediscclis) espeleticola, male, a, b. Genita- lia, dorsal (gonoforceps omitted), ventral and lateral views; c, S7; d, Apicolateral part of S7 of another individual, dorsal view, with laterobasal lobe more horizontal instead of directed ven- trally as in c (Posterior margins of the apicolateral processes are membranous, delicate, asymmetrical or damaged); e-g, Posterior tibia and basitarsus in \entral, inner and outer views. For fur- ther explanation, see account of the subgenus.
4020 m, 26 May 1998, in dead dry flowering stems of Esphietia timotensis (J. Ashe, R. Brooks, R. Hanley, Lawrence); one male, Parque Nacional Sierra Nevada, trail between L. Mucubaji and L. Negra, 3500 m, 4 June 1988 (A. L. Norrbom, G. J. Steck, Washington).
Two female specimens differ from Chilicola espeleticola in that the malar space is about three-fourths as long as broad, the clypeus is cut below the middle by the lower ocular tangent, and the clypeus has a longitudinal median depressed line (rarely well developed in C. espeleticola). Thus they resemble C. quitensis and its close relatives, which, however, have an even shorter malar space and less produced clypeus. Probably these females represent a new species, the male of which remains unknown. They are from PERU: Lima: MaKicana, 2400 m, one 23 June 1955 (C. D. Michener, Lawrence), the other 12 May 1920 (Cornell Exped. Lot 569, /f/!(7cn).
Etymology: Tlie specific name is based on Espeletia (Asteraceae), the well known paramo plant in whose dead flowering stalks the type series was taken.
Nests of Chilicola espeleticola
Tlie paratype series of 42 males and 93 females was extracted at Paso Pico Aguila, 3710 m altitude, from about 35 nests found in dead, dry flowering stems on living plants of Espeletea schiiltzii on May 26, 1998. The following is based on observations by R. W. Brooks. Occupied stems sloped upward (vertical to about 20" above horizontal) to the bro- ken apices that allowed access to the pith. Hundreds of similar horizontal to drooping stems were examined with- out finding nests of Chilicola. About a kilometer away in the same paramo, at Alto de Timotes, 4020 m, in stems of Espeletia timotensis, tinly three nests were found among hundreds of appropriate looking, dead, dry, broken, old flowering stems. At both sites, no other kinds of suitable pithy stems were available.
The bees (or possibly initially other insects) burrowed into pith in stems 7 to 9 mm in diameter, forming nests 5 to 25 cm deep and 3 to 4.5 mm in diameter. This variablility in burrow diameter suggests that some were excavated by some other insects before arrival of a bee.
The cells had been largely destroyed by the emerged adult bees which were found in cold weather lined up head to tail in burrows, but Brooks' impression is that few adults had left the nests; only one (a male) was found on a flower during an hour of sunshine (collecting by two people at the Paso Pico Aguila site). The female-biased sex ratio, however, possibly indicates departure of some males. Two to about 12 adults occupied each nest and probably indi- cated the number of cells that had been in each nest. Since all were unworn, the bees that made and provisioned the nests must have disappeared and their progenies were ready to leave the nests.
32
Scientific Papers, Natural History Museum, The University of Kansas
The cells appear to have been similar to those of Chilicola styliventris described elsewhere in this paper. Measurements between remnants of intercellular walls indicated that cells were 10 mm long. Two measurable cell diameters were each 3 mm. The burrows, as indicated above, arc often larger than 3 mm in diameter. The cell mem- brane in such cases were separated from the burrow walls by spaces containing scattered threads extending from the burrow wall to, and supporting, the cell membrane.
Nearly all the larval feces, bits of cell membrane, and other debris appeared to have been removed from the nests by the bees. Perhaps one female was preparing to re-use each burrow.
Chilicola {Oroediscelis) brooksi new^ species (Figures 21b, 23)
From the other similarity large species, Chilicola espeleticoln, C. brooksi differs by the short malar space and clypeus, as well as other characters indicated below. The right angular to acute tooth on the outer side of the hind basitarsus is distinctive for C. brooksi. It differs from the description of C. espeleticola as follows:
Male: Length 7.5 to 8.5 mm; forewing length 5.4 to 5.5 mm. Face about 1.08 times longer than broad. Clypeus, malar space, lower ocular tangent, and hypostomal area as described for Chilicola quitensis. Last three or four flagel- lar segments each longer than broad. Scutum and scutel- lum with punctures almost as dense as they can be. Long- est hairs of scutum and scutellum about three-fourths as long as scape, those of lower side of thorax and lower ge- nal area as long as scape. Hind tibia about three times as long as greatest width, inner crest ending in sharp erect tooth, basal to which crest is gradually reduced (Fig. 23g). Hind basitarsus shorter than remaining tarsal segments taken together, only about as long as 2.5 following seg- ments; hind basitarsus widest near base because of expan- sion of lower margin, outer expansion forming right-an- gular to acute tooth about one-third of length of basitarsus from base (Fig. 23f). Metasoma more shiny although interspaces finely lineolate, punctures not denser on pos- terior than on anterior terga; posterior marginal zones of terga with well separated fine punctures extending near to margins; lateral fasciae of pale hairs absent; S4 and S5 as described for C. quitensis; S6 with apical fringe of long, well spaced hairs surpassing apical transparent lamella. Genitalia and hidden sterna as in Figures 23a-e; differences from C. espeleticola are indicated under that species; S7 has anterolateral margin of the ventral lobes wrinkled and ir- regular and the basal part covered with minute spicules (Fig. 23c, right half).
Female: Agrees with description of male except for usual sexual features. Length 7 mm; forewing length 5.4 mm. Face about 1.1 times longer than broad.
Holotype male and three male and one female paratypes: VENEZUELA: Merida: 42.4 km northwest of Merida, near La Carbonera, 8" 37' 38" N, 74" 21* 10" W, 2360 m, 25 May 1998, from Riibus flowers 0- Ashe, R. Brooks, R. Hanley, Lawrence).
Additional material: VENEZUELA: Trujillo: one male, 7.2 km from Bocono, old Trujillo road, 9" 20' 42" N, 70° 17' 47" W, 2200-2300 m, 20 May 1998 (J. Ashe, R. Brooks, R. Hanley, Laxvrence). COLOMBIA: Antioquia: one male. La Selva, Rionegro, 2120 m, August, 1975, on Polygonum segetum (J. Cano, Medellin); one female, same locality, July 1975, in stem (R. Afiez, Medellin).
The female paratype has the thoracic hair badly worn so that few long hairs are present and one cannot verify this character. Both sexes lack the pale lateral hair fasciae on the metasomal terga but in the long series of Chilicola quitensis this feature is variable, perhaps due to wear.
Etymology: The name is a patronymic for Dr. R. W. Brooks, one of the collectors of the specimens as well as a contributor toward completion of this paper.
Chilicola (Oroediscelis) benoistiana new species (Figures 21c, 24)
This species is slightly smaller than Chilicola quitensis but is morphologically more similar to the larger C. brooksi, from which it differs in the hind tibial structure of the male, as indicated in the key to species. As noted in the descrip- tion of the subgenus, its lower subantennal sutures are more strongly convergent below than in other species of the subgenus. It agrees with the description of C. espeleticola except as follows:
Male: Length 6.0 to 7.5 mm; forewing length 4.7 mm. Malar space about half as long as broad; subantennal su- tures strongly converging below; lower ocular tangent below middle of clypeus. Last four flagellar segments each slightly longer than broad. Mesoscutum with punctures separated by much less than a puncture width. Dorsal sur- face of propodeum about as long as scutellum. Distal stig- mal perpendicular meeting first submarginal crossvein. Hind tibia over three times as long as breadth at preapical brush; inner crest ending in right angular apical tooth (Fig. 24d, A) and with another tooth at midlength of tibia (Fig. 24d, D), outer crest very broad, broadest well before apex of tibia; hind basitarsus shorter than remaining tarsal seg- ments together, widest near base because of rounded ex- pansion of lower margin (Fig. 24c), in ventral \'iew form- ing strong tooth slightly basal to middle of basitarsus on outer surface (rounded in specimen from Route de Calderon). Metasomal terga with punctures separated by less than one puncture width except for impunctate mar- ginal zones; S2 and especially S3 with bosses higher than in C. espeleticola; S4 with projections strong, emargination between them deeper than broad, surface between them
Bee Genus Chiucola in the Tropical, Andes
33
23d
23e
Fig. 23. Chilicolii {Oivcdifcclis) bivoksi, male, a, b, Genitalia, dorsal, ventral and lateral \-ievvs; c, S7; d,e, S8, dorsal, ventral and outer views; f, g, Posterior tibia and basitarsus in ventral and outer vieu's. For further explanation, see account of the subgenus.
with few hairs (none posteriorly), projections not flattened, apices minutely bilobed (Fig. 21c) and in side view show- ing small preapical tooth on upper surface (this tooth ab- sent in specimen from Route de Calderon); hairs lateral to projections of S4 long but not strongly incurved; S6 with ridges lateral to broad median concavity sharp, almost carinate, converging posteriorly; longest hairs of fringe of S6 surpassing apical translucent lamella. Genitalia and hidden sterna much as in C. stylivoitris, anterolateral mar- gins of ventral lobes of S7 more irregular, posterior parts differently shaped (Fig. 24b); mesal margins of gonoforceps
with longer hairs, penis valves with apical membranous appendages longer (Fig. 24a); S8 as in Chilicoln ciizcociisis or with apical hairs nearly all marginal.
Female: Agrees with description of male except for the usual sexual features and the following; Length 6 mm; forewing length 4.4 mm. Mesepisternum with punctures similar to those of dorsum of thorax but separated by about three puncture widths. Metasomal terga with punctures fine and separated by several puncture widths.
Holotype male, one male and three female paratypes: ECUADOR; [Pichiiichn]: Sambiza (no other data) (R.
34
Scientific Papers, Natural History Museum, The University of Kansas
Figs. 24-25. 24: Chilicola (Orocdiscclis) benoistiana, male, a. Genitalia; b, S7; c, d. Posterior tibia and basitarsus, ventral and outer views. S8 is as in C. cuzcoensis or with hairs all marginal. 25: C. (O.) quiteiisis, male, a, S7; b, c. Posterior tibia and basitarsus, ventral and outer views. Genitalia and S8 are as in C. brzoskni. For further explanation, see accoimt of the subgenus.
Benoist, Paris); one male and one female paratype: Catacallao, 22 Feb. 1931 (R. Benoist, Pari^).
Additional material: ECUADOR: [Picliinchti]: one male. Route de Calderon, 10 Oct. 1930 (R. Benoist, Paris) differs from the type material by characters indicated in the above
description; it may represent another species. In the key it runs best to Chilicola quiteiisis from which it differs in the form of the processes of S4 and of the crests of the hind tibia. One female from Cumbayo, 19 June 1931 (R. Benoist, Paris) may belong to this species.
Bee Genus Chiucola in the Tropical Andes
35
Etymology: The name benoistiann is an adjectival pat- ronymic in honor of Raymond Benoist, the first student of Andean Chilicoln and the collector of the specimens here described.
Chilicola (Owediscelis) quitensis Benoist (Figures 21 d, 25)
Chilicola quitensis Benoist, 1942:79.
Chilicoln quitensis var. immaculata Benoist, 1942:80. New Syn- onymy.
This species agrees with the description of Chilicola espjeleticola except as indicated below. As in the case of C. brooksi, the shorter malar space and less produced clypeus distinguish it easily from C. espeleticola.
Male: Length 6.5 to 7.5 mm; forewing length 5.2 to 5.6 mm. Face about 1.05 times longer than broad. Lower paraocular yellow areas as in Chilicoln espeleticola or re- duced, sometimes to mere line, or rarely entirely absent (var. immaculata). Clypeus with scattered large punctures, those on upper part often little smaller than those of lower third, depressed median longitudinal line of clypeus com- monly evident; malar space about half as long as wide; lower ocular tangent below middle of clypeus; smooth anterior extremity of hypostomal area short. Thorax with punctures separated by less than a puncture width. Dor- sal surface of propodeum slightly shorter than scutellum, striae quite irregular. Longest hairs of scutum about two- thirds as long as scape, those of scutellum about three- fourths as long as scape, those of sides of thorax and lower genal area nearly as long as scape. Hind tibia about three times as long as greatest width, imter crest strongly cari- nate, ending in sharp apical spine (Fig. 25c) and with weak angle near middle; hind basitarsus slightly shorter than remaining tarsal segments together, widest near base (Fig. 25c) because of broad rounded lobe of lower margin, this expansion continuing to beyond middle of basitarsus as small secondary lobe, projections on outer surface rounded (Fig. 25b). Metasoma more shiny although interspaces finely lineolate. S4 with projections long, often slightly enlarged apically, emargination between them deeper than wide (Fig. 21d), area between their bases with a few rather long hairs, without short hairs; S5 without apical fringe; S6 with lateral tuft not very yellow, marginal fringe weak, broken medially, much surpassed by transparent lamella. Genitalia and S8 as in C. brooksi; S7 with ventral apicolateral lobes more slender (Fig. 25a), basal part with minute spi- cules as in C. brooksi.
Female: Agrees with description of male except for usual sexual features. Forewing length 4.5 to 5.6 mm. Face about LOS times longer than broad. First flagellar segment about L5 times as long as broad.
Lectotype male (of Chilicola quitojsis) here designated: ECUADOR: [Pichincha]: Quito [2850 m elevation, accord-
ing to Benoist], 24 Oct. 1931 (R. Benoist, Paris). This speci- men also bears a red label, "allotype" and a white printed label "Museum Paris, Coll. R. Benoist 1924." These labels were presumably attached bv the Museum after receipt of the Benoist collection; see comments below. Lectotype male (of C. quitensis var. immaculata) here designated: [Piclmcha]: Quito, 16 Oct. 1930, "Rubus" (R. Benoist, Paris). This speci- men also bears an olive green printed label "Museum Paris/Equateur/Coll. R. Benoist 1924." Such labels, on all of Benoist's specimens, must have been added after the collection arrived at the Museum although the specimens were collected years later than the 1924 date on the labels. Lectotypes are designiated to verify the meaning of the name quitensis; in Benoist's series were also specimens of C. benoistiana.
When he described this species, Benoist considered the female first and the male characters subsequently. Accord- ingly, in the Paris Museum a female has a red holotype label and a male, a red allotype label. Such labeling has not been published. In any case, Benoist did not label any specimens as types of any kind; he did write the name "Chilicola quitensis" in red ink on labels of two females, of which one was at the head of the series of specimens, and it is this female that has addiHonally been labeled holo- type by the Museum. It is not a holotype since it was not published as such by Benoist. Since in this group the prin- cipal specific characters are in the male, I here designate and label as lectotype of Chilicola quitensis the male that had been previously labeled as the allotype.
Of the 19 males in Benoist's series, only two completely lack yellow facial markings. One of these bears Benoist's label in red ink, "var. immaculata." 1 designate and label that specimen as lectotype of Chilicola quitensis var. immaculata Benoist.
Additional material: Of the 35 specimens in the Paris Museum, most are from Quito. Nearby localities for this species listed by Benoist (1942) are as follows: Rumipamba, Cotocollao, route de Calderon, Palmira, valle de Lloa, Paramos de Aloag and Otavalo. Of these localities, only Palmira is verifiable by male specimens. Specimens from Cotocallao and route de Calderon are Chilicola benoistiana or a related species, while those from Otavalo are Hylaeus benoisti Michener, a species that superficially resembles C. quitensis. Dr. Luis Coloma of the Universidad Catolica de Quito has verified that these localities are near Quito in the Province of Pichincha, with the possible exception of Palmira, which is the name of a town in the pro\'ince of Chimborazo. Almost certainly Benoist's Palmira was a dif- ferent place, near Quito. One additional male specimen is from ECUADOR: Napo: Misahualli, 1-2 April, 1986 (S. McKamey, Berkeley).
Benoist (1942) collected Chilicola quitensis throughout the year. Although some of his specimens are labeled "Ru-
36
Scientific Papers, Natural History Museum. The University of Kansas
bus," no doubt meaning taken from nests in dead pithy stems of Rubiis, others are not so labeled and some were taken on flowers. The implication is clear that the bees are active throughout the year. Benoist indicated that the spe- cies visits various flowers but is especially abundant on flowers of Baccharif pol\/aiitlm (Asteraceae).
Chilicoln (Omediscelis) cuzcoensis new species (Figures 21e, 26)
This species resembles Cliilicola espeleficola, from wh ich it differs in the shorter malar space and clypeus, thus re- sembling C. qidtcnsis, bivokfii, etc. It differs from those spe- cies in the subapical yellow clypeal mark and the very fine, sparse punctation of the metasomal terga. C. cuzcoensis differs from the description of C. espclcticoln as follows;
Male: Length 7.5 mm; forewing length 5.6 mm. Face about 1.05 times longer than broad. Clypeus with subapi- cal transverse yellow lunule (in addition to yellow of paraocular area). Lower face (i.e., below antennae) with well separated strong punctures; malar space about two- thirds as long as wide. First flagellar segment as long as second, even on lower side; last three or four flagellar seg- ments each distinctly longer than broad. Thoracic puncta- tion strong, particularly on posterior part of scutum and on scutellum where punctures are separated by more than one puncture width of shining ground; punctures of ante- rior two-thirds of scutum separated a puncture width or less of finely sculptured, dull ground. Dorsal surface of propodeum as long as scutellum. Thoracic and cephalic hairs about as described for Chilicoln bwoksi. Hind tibia about three times as long as greatest width, similar to that of C. bwoksi. Hind basitarsus as long as tarsal segments 2 to 4 taken together, expansion of outer surface (ventral view) forming rounded, obtuse tooth on outer side about two-fifths of distance from base to apex, thus differing from right angular or acute tooth of C. bivoksi. Metasomal terga with punctures minute, widely separated. S2 with hairs of lateral area not long; S4 with projections long, widest preapically, apices diverging, emargination between them deeper than wide, area between their bases with ordinary hairs (Fig. 21e); S5 with sparse apical fringe, not erect; S6 with rounded ridges diverging basad, api- ces each produced into small projection near margin of sternum, areas lateral to ridges not deeply depressed, apical margin with fringe of well spaced hairs nearly reaching apex of transparent lamella. Genitalia and hid- den sterna as in Figures 26a-d; broad midlateral pro- cess of apicolateral lobe of S7 resembles that of C. espeleticoln.
Holotype male: PERU: [Ctizco]: Cuzco (=Cusco)[?], 31 Jan. 1968 (A. Garcia, C. Porter, Lawrence). The printed label originally read Pisac, Cuzco. The word Pisac has been crossed out, leaving Cuzco, probably meaning the
vicinity of the city of Cuzco rather than the department. Pisac is a town near Cuzco.
Etymology: The name is based on that of the city and department of Cuzco, Peru, where the species was col- lected.
Chilicola (Oroediscelis) bigibbosa new species (Figures 21 f, 27, 28)
Of known species of Oroediscelis, this is the one with the longest malar area, the other species with the malar area nearly as long being Chilicola espeleticoln and C. inaculipes. C. bigibbosa resembles C. mnculipes in the pres- ence of two large lobes on the ventral margin of the hind basitarsus of the male, but is unique in the femoral charac- ters of the male described below. This species agrees with the description of C. espeleficola except as follows (see also the subgeneric description):
Male: Length 7.0 mm; forewing length 5.1 mm. Face about 1.19 times longer than broad. Clypeus with trans- verse subapical yellow lunule (in addition to vellow on paraocular area); clypeus with strong longitudinal median groove; malar area as long as wide. Antennal flagellum as described for Chilicola cuzcooisis. Thorax shining although microsculptured between punctures, which are strong and mostly separated by less than a punctvire width. Longest hairs of scutum about half as long as scape, those of scutel- lum and sides and venter of thorax about two-thirds as long as scape, those of lower genal area longer than scape, dense. Distal stigmal perpendicular slightly beyond first submarginal crossvein. Front femur with strong, basal flat keel-like projection (Fig. 28b); mid femur with sharp pro- truding angle in same position, followed by emargination (Fig. 28a). Hind tibia over twice as long as greatest width, preapical brush almost absent, margin of inner crest and its apical tooth yellow, folded toward outer side as in C. iiinciilipes and C. espeleticoln (Fig. 27d); hind basitarsus about as long as tarsal segments two to four together, lower mar- gin expanded to form two lobes, one near base, the other (larger) near micfdle of basitarsus (Fig. 27e); basal parts of these lobes (especially the middle basitarsal lobe) expanded a little beyond outline of basitarsus in ventral view, form- ing obtuse angle at midlength of basitarsus on it's outer surface. Metasoma shining but microlineolate, with rather large punctuies separated by a puncture widtii c^r more;
Figs. 26-29. 26: Chilicoln (Oroediscelis) cuzcoensis, male, a, b. Genitalia, dorsal, ventral and lateral views; c, S8; d, S7. 27: C. (O.) bigibbosa, male, a. Genitalia; b. Apex of S8; c, Laterodistal part of S7, dorsal view; d, e. Posterior tibia and basitarsus in ven- tral and outer views. For further explanation, see account of sub- genus. 28: C. (O.) bi;^ibbosn, male, anterior views of middle (a) and front (b) femora. 29: C. (O.) espeleticoln, male, anterior views of middle (a) and front (b) femora.
Bee Genus Chilicoia in the Tropical Andes
37
38
Scientific Papers, Natural History Museljm. The University of Kansas
Bee Genus Chiucola in the Tropical Andes
39
S2 with subapical lateral bosses (as found in C. espclcticola) high, hairy, brown, and elongate longitudinally; S4 with pair of large preapical tapering tubercles (Fig. 21 f), area between them bare, lineolate, area lateral to them with long, sparse hairs; posterior margin of S4 with apical fringe of long hairs; S5 with apical fringe of well spaced long hairs, much shorter medially, long lateral hairs of S5 sparse; S6 nearly flat, dull with microsculpturing, except for lateral elevation bearing dense tuft of yellowish hairs, apical margin with dense fringe of short, robust hairs that do not reach apex of narrow transparent lamella. Genitalia and hidden sterna similar to those of C. simplex but note more truncate margin of midlateral part of ventral apicolateral lobe of S7 (Fig. 27c), longer hairs and deeper emargination of S8 (Fig. 27b) and different distribution of ventral hairs on gonoforceps (Fig. 27a).
Female: Forewing length 4.7 to 5.0 mm. Agrees with description of male (as modified from that of Chilicola I'spt'lcticola) except for usual sexual features and the fol- lowing: Face about 1.28 times as long as broad, without yellow. Malar area slightly longer than broad, lower ocu- lar tangent above middle of clypeus. First flagellar seg- ment about 1.5 times as king as broad. Punctation of tho- rax as described for male of C. espclcticola. Legs without yellow. Hairs of thorax worn but apparently similar tti those of male. PunctaHon of metasomal terga as described for male of C. espeleiicola.
Holotype male: PERU: \]\inin\. Acolla (near Jauja), 3460 m, 24 Feb. 1955 (F Blancas, Lnivrcncc). One male paratype, same data as holotype but 26 Feb. 1955 (F. Blancas, Lima). Five female paratypes, same data but May 1953 (three specimens), Apr. 1950, and Apr. 1952 (all F. Blancas, Lima) Two female paratypes: same data but April 1950 (Lawrence). The label on the holotvpe is handwritten; the labels on the paratypes are machine printed, and each of the specimens in Lawrence has an additional label, "Cole(;ao Campos Seabra." Nonetheless the locality, eleva- tion, and collector are the same as those of the holotype. These specimens must have gone by exchange or purchase to the Seabra collection in Rio de Janeiro, thence to the University of Kansas.
Figs. 30-32. 30: Chilicola {Owediscelis) macuHpes, male, a, b. Posterior tibia and basitarsus in ventral and outer views. The genitaiic and sternal structures are similar to those of C. simplex and C.bigibbosa. 31: C. (O.) styliivntris,ma\e. a. Genitalia; b, S7; c, d, Posterior tibia and basitarsus, ventral and outer views. S8 is similar to that of C. faroofoi. 32: C. (O.)^/W'0S(!, male, a, b. Poste- rior tibia and basitarsus, ventral and outer views, specimen from Banos, Ecuador c, Outer view of hind basitarsus of specimen from Rionegro, Colombia. The genitalia and sternal structures are similar to those of C. simplex. For further explanation, see account of the subgenus.
Additional material: One female, without associated males, is from PERU: [//(///»]: Huancayo, 3217 m, 26 Mar. 1962 (R. Garcia, Lima). Four other females without associ- ated males may belong to this species although they are smaller (length 5.5 mm). The one from Peru has the malar space slightly broader than long, those from Bolivia have this space about as long as broad. Probably, when males are known, these specimens will be recognized as one or two new species. The data are as follows: PERU: [Huamico]: one female, 10 mi (16 km) southwest of Las Palmas, 1000 m, 26 Sept. 1954 (E. S. Ross, E. 1. Schlinger, San Francisco). BOLIVIA: La Paz: two females, Coroico to Chulomani, 7 Mar 1968 (Garcia and Porter, Los Angeles); one female, Sorata, 2700 m, 7 Apr. 1982 (M. Cooper,'Li/»!t' Regis). These are the only Bolivian records for the subgenus Oroediscelis.
Etymology: From gibbosa, Latin, protuberant, plus bi-, two, with reference to the two projections on the hind basitarsus of the male.
Chilicola (Owediscelis) mactilipes new species (Figures 21g, 30)
The two large lobes on the lower margin of the hind basitarsus of the male are not as large as in Chilicola bigibbosa and the basal lobe is much broader than the distal one, unlike C. bigibbosa. Nonetheless C. maciilipes and bigibbosa are close relatives as indicated by the long malar space and the yellow on the imier hind tibial crest. This species agrees with the description of C. espeleiicola except as described below (see also subgeneric description).
Male: Length 6 to 7 mm (estimated from specimens with apices of metasomas removed); forewing length 4.7 mm. Face about 1.16 times longer than broad. Clypeus with transverse apical yellow band (in addition to yellow on paraocular area); malar area as long as wide. Flagellar seg- ments 9 and 10 about as long as broad, 11 distinctly longer. Thoracic punctation and hairs about as described for Chilicola bigibbosa (hairs of paratype worn and matted). Dorsal surface of propodeum as long as scutellum. Wing venation as in C. bigibbosa. Hind tibia (Fig. 30a, b) about three times as long as broad, preapical brush absent, outer crest of under surface scarcely recognizable, inner crest strong, apical part folded toward outer side as in C. bigibbosa, ending in narrowly rounded tooth, this crest and tooth, and in paratype base of area between crests, yellow; hind basitarsus with two lobes on lower margin (Fig. 30b), smaller one in middle of basitarsus, basal one broadly con- vex and bare (with short hairs in C. bigibbosa), in ventral view basal half (actually more) of basitarsus thickened on outer side, distal half slender, much as in C. bigibbosa, outer side of lobes with large yellow area (Fig. 30a). Metasoma with punctation and microsculpture much as in C. espeleiicola but somewhat more shining; S2 with subapical lateral elevations rounded in lateral view and higher than
40
Scientific Papers, Natural History Museum, The University of Kansas
processes of S4, these projections on S2 longitudinally elon- gate and connected by strong, bare, trans\erse ridge; S3 with lateral bosses distinct; S4 with pair of preapical tu- bercles (Fig. 21 g) not reaching posterior margin of ster- num, area between tubercles bare; S4 and S5 with apical fringes, interrupted medially, of well spaced long hairs and without patches of long hairs laterally; S6 nearly flat with lateral tuft of yellowish dusky hairs, few short hairs apically not attaining apex of transparent rounded lamella. Genitalia and hidden sterna similar to those of C. simplex.
Molotype male, PERU: [Apiirimac]: Curahuasi, 10 Feb. 1978 (P. M. Marsh, Lazvrence). One male paratype, Cttzco: Limatambo, 21 Feb. 1978 (P. M. Marsh, Wasliin;^- ton).
Etymology: uiaciiln, Latin, spot, plus pes, Latin, leg or foot, with references to the yellow areas on the hind tibiae and basitarsi. Such areas on the hind basitarsi are not found in any other species of the subgenus.
Chilicola (Oroediscelis) styliventris (Friese) (Figures 21h, 31)
Oediscelis styliventris Friese, 1908:340, 424 or in reprint, 9, 93. Chilicola styliventris: Cockerell, 1926:218.
The male of this species is easily distinguished from all others by the long processes of S4 with slightly enlarged bilobed apices that reach the apex of the metasoma when the latter is retracted. Females are not reliably distinguished from those of Chilicola transversaria, gibbosa, and brzoskai , but have slightly longer malar areas and longer antennae with the middle flagellar segments about as long as wide. The dorsum of the thorax is more coarsely punctate in brzoskai . All of these characters seem to be variable and assignment of females not accompanied by males to spe- cies is dubious. C. styliventris differs from the description of C. espeleticola as follows:
Male: Length 5.5 to 6.5 mm; forewing length 4.2 to 4.7 mm. Face about 1 .09 times longer than broad. Clypeus with transverse yellow mark near lower margin, sometimes re- duced to small yellow spot; lower paraocular area beside clypeus with yellow mark. Head less dull than in Chilicola espeleticola but sculpturing as described for that species. Malar space about half as long as broad; lower ocular tan- gent at lower third of clypeus. First flagellar segment about two-thirds as long as second, the latter nearly twice as long as broad; distal flagellar segments all distinctly longer than broad. Thorax with punctures mostly separated by less than a puncture width. Longest hairs of scutum about two- thirds as long as scape, those of scutellum as long as scape, those of sides of thorax and lower genal areas longer than scape. (Hairs evidently easily worn off, since some speci- mens lack all long hairs.) Distal stigmal perpendicular en- tering base of second submarginal cell. Hind tibia nearly three times as long as greatest width at preapical brush;
inner crest with elevation beyond middle of tibia (Fig. 31d) and apical acute or rounded tooth, separated by broad concavity; hind basitarsus little longer than next three tar- sal segments together, basal half thickened on outer sur- face, forming in ventral view a right angular tooth at midlength of basitarsus (Fig. 31c), outer surface not con- cave. Metasoma shining, apical terga and basal parts of first three terga lineolate, punctures strong, mostly sepa- rated by about a puncture width, posterior margins of terga smooth, covered by weak pale hair fasciae at extreme sides. S2 with posterior half bare medially, moderately long hairs basally and laterally; S4 with projections long, apices ex- panded and briefly bilobed in lateral view (Fig. 21 h), emar- gination between them deeper than wide, area between them with a few moderately long hairs, sparse very long hairs lateral to projections; S5 with apical fringe absent, very long hairs on lateral part of S5; S6 with lateral de- pressions less deep than in C. espeleticola. Genitalia and hidden sterna as in C. brooksi but ventral lateroapical lobes of S7 differently shaped (Fig. 31b), membranous posteri- orly in both species; S8 with apical hairs commonly longer than in C. brooksi; apical membranous appendages of pe- nis valves shorter than in C. brooksi (Fig. 31a).
Female: Agrees with description of male except for usual sexual features, and the following: Length 5 mm; forewing length 4.2 to 4.7 mm. Face about 1.17 times longer than broad. First flagellar segment about 1.5 times as long as broad. Longest hairs of scutum about half as long as scape, those of scutellum, sides of thorax, and lower genal area two- thirds or three-fourths as long as scape (all avail- able females badly worn and these statements about hair lengths are not reliable). Metasomal terga lineolate, punc- tures finer than in male, mostly separated by more than puncture width; posterior marginal zones less strongly lineolate, covered by fasciae of white hairs at sides.
Lectotype female here designated: PERU, 1900, with orange label "Typus" and Friese's identification label dated 1907 (New York). Friese (1908) described this species in a paper on Argentine bees, and gave only the following lo- cality information: "2 males and 2 females, von Peru durch Rolle erhalten"; this corresponds with the specimen desig- nated as lectotype in New York. Material in Berlin consists of a female and a male, each with a handwritten label, Arequipa, Peru, and the notation XI (probably November). The label of the Berlin male also is marked "2500" (prob- ably meters altitude). Each specimen also bears Friese's identification label, that for the female and a Washington male printed 1909, that for the Berlin male, 1907. The fe- male has an orange "Typus" label; the Berlin male, a printed label "Coll. Friese," and the Washington male, a red label "cotype no. 13206 USNM." Asearch in June 1999, kindly made by Michael S. Engel, did not reveal additional material in Berlin. As the species was described in 1908,
Bee Genus Chiucola in the Tropical Andes
41
the specimens with 1909 identification labels are probably not part of the type series, in spite of orange and red la- bels. Moreover, Friese presumably would have given the locality, Arequipa, if these specimens had been before him when he described the species. Because specific charac- ters of females are almost lacking, it would have been de- sirable that the lectotype be a male, but unfortunately the female from New York labelled only "Peru, 1900" is the only specimen that appears to be part of the type series.
Additional material: PERU: [Arequiim]: one male, Arequipa, 2500 m, November (Berlin); five females, same locality, 2300 m, 2 July 1979 (C. Porter, A. Cerbone, Laiorence); two females, same locality, 23 Sept. 1963 (M. Alvarenga, Lazorence); one male, same locality, 2300 m, 28 Feb. 1984 (M. Cooper, Lyme Regis); five males, one female, same locality, Apr. 1922 (E. Escomel, San Franciseo and Lo- gan). [Ai/aciicho]: two males, Puquio (Cap. Lucanas), 3400 m, Apr. 1950 (F. Blancas, Lima and Laiurence). Ancasli: one male, no exact locality, 5 Jan. 1969 (H. Picho, La Molina). Cuzco: Pisac, one male, 3 Feb. 1986 (A. Garcia, C. Porter, Los Angeles); six males, two females, Pisaq (=Pisac), 3042 m, 1 May 1977 (R. Garcia, Lima); one male, Machu Picchu, 1900 m, 4-19 Sept. 1964 (C. C. Porter, Los Angeles); two males, Cuzco, 3800 m, 20 Sept. - 2 Oct. 1964 (c' C. Porter, Los Angeles). Junin: three males, one female, Tarma, 3000 m, 1 Dec. 1942 (Lawrence); thirteen males, four females, Tarma, 3500 m, 11" 25' 30" S, 75" 45' 24" W, 22 Oct. 1998 (R.W. Brooks, Laiorence); five females. La Victoria, Mantaro, 3252 m, 16 June 2000 (C. Rasmussen, Aarhus); one female, Huancayo, La Victoria, 10 Nov. 2000 (C. Rasmussen, Aarlnis). Unidentified locality: one male, Yura, Peru (New York). ECUADOR: [Tungurahua]: one male, one female, Oriente to Ambato, 2600 m, Oct. 1956 (Lazorence).
An unassociated female from "Tingo," Peru (Washing- ton), was recorded as Cliilicola styliventris by Cockerell (1926); without associated males its identification is in doubt. Other females, probably of C. sti/liventris but with- out associated males, are those from La Victoria (two lo- calities), cited above.
Nests of Chiucola styliventris
Tlie specimens from Tarma collected by R. W. Brooks were found in nests in dead stems of Acln/rocline alnta (Kunth) DC [Asteraceae]. The stems were more or less erect, the tops broken off, allowing access to the pith in which the bees had burrowed.
Stem diameters were 5 to 6 mm; the burrows were 2.5 to 3.0 mm in diameter. Nest No. 1, 10 cm deep, contained 9 male bees and three closed cells that produced males; thus the total progeny was 12 males. Nest No. 2, 11 cm deep, contained three (unworn) females and one male bee. Most of the cells were in bad condition from the emergence of the bees and perhaps from opening of the nest and the
number of cells was not clear, but two closed cells each contained a large dead larva (possibly living when col- lected) with feces, and one contained a dead adult female with wings not expanded. The broken cells included one nearly filled with larval feces, probably crowded into it by emerging adults, and two with old, uneaten provisions. The probable number of cells was nine; one produced a male, and four females. Obviously the bees that made and provisioned Nests 1 and 2 had disappeared, and the prog- eny had mostly matured but had not left the natal nests.
Nest No. 3 (depth unknown but seemingly less than Nos. 1 and 2) contained a single female and no cells. Pre- sumably it was a young bee that had left its natal nest and was preparing a new nest.
Cells were 7.5 to 8 mm long, 2.5 mm in diameter, cy- lindrical with transverse ends, and made of very thin, deli- cate, transparent, cellophane-like membrane, in which some slender, unbranched fibers can be detected. When the membrane is torn, these fibers sometimes project free from the matrix of the membrane.
The membrane between cells is at least sometimes double, consisting of the closure of one cell and the base of the next, applied directly on the former. Thus both are transverse, hard to separate. The membrane of the base of a cell is not cup-shaped as in Cliilicola aslnneadi (Crawford) (Eickwort, 1967), except that the lowermost cell, based on the concave surface of pith at the bottom of the burrow, has a cup-shaped base. The lateral walls of a cell are in contact with the pith walls of the burrow, but do not ad- here firmly.
The feces of the larvae are short cylindrical, about 0.50 nim long and 0.22 n\m in thickness, with rounded ends, and are packed into the base of the cell forming a yellow to dark brown, somewhat irregular plug about 1 mm thick, concave on its upper surface. The feces must be rather soft when fresh for while recognizable in the plug, they were obviously to some degree shaped by being pressed to- gether. Occasionaly two or three separate pellets were found at the upper end of a cell.
Cliilicola (Oroediscelis) transversaria new species (Figure 21-i)
Males of this species and Cliilicola gibbosa differ from all other Oroediscelis in the single, large, median projection from the lower margin of the hind basitarsus. C. transz'crsaria differs from C. gibbosa in the pair of tubercles of S2 of the male which are flattened to form short trans- verse (actually somewhat oblique) ridges, appearing as sharp tubercles in profile. The sternal surface between and near these tubercles is hairy; it is bare in C. gibbosa.
Male: Agrees with description of Cliilicola espeleticola except as follows: Length 6.0 to 6.5 mm; forewing length 4.8 mm. Face about 1.05 times longer than broad, all black
42
Scii:n Tiiic Papkrs, Natural History Museum, The University of Kansas
except one paratype (collected by G. E. Bohart) with yel- low mark on lower paraociilar area between lower end of eye and clypeus. Head sculpturing and structure as de- scribed for C. brzoskai but first flagellar segment about two- thirds as long as second on lower surface. Scutum and scutellum with punctures mostly separated by less than puncture width or much less laterally on scutum, ground dull with microsculpturing to rather smooth in holotype. Hair lengths of thorax as described for C. brzoskai; lower genal area with longest hairs only two-thirds as long as scape (shortest of any OwedifCflis). Stigma slightly less than two-thirds as long as marginal cell on wing margin, distal stigmal perpendicular entering base of second submarginal cell. Middle femur with well defined ventral concavity occupying basal third. Hind tibia about three times as long as greatest width at preapical brush, inner crest ending in broad, flattened, rounded projection whose plane is par- allel to imaginary plane through bases of tibial spurs (at right angles to plane of equivalent flattened apical tooth of C. brzoskai); hind basitarsus with strong rounded pro- jection slightly basal to middle of lower margin, projec- tion of outer surface absent. Metasomal terga as described for C. brzoskai. S2 with bosses elevated to form short, ob- liquely transverse, shining ridges converging anteriorly, appearing as acute or right angular tubercles in lateral view; surface between and behind these ridges with some small punctures and short hairs. Projections of S4 small, apices blunt (Fig. 21i); otherwise S4, S5, S6 as in C. brzoskai , fringes of well spaced hairs on S4 and S5 conspicuous. Hidden sterna and genitalia similar to those of C. simplex, S8 with apical hairs longer and more numerous, about 14 on each side; ventral mesal lobe of gonoforceps with hairs longer and denser.
Female: Agrees with description of male as modified from that of Chilicola espelctkola except for usual sexual features and the following: Length 5.5 to 6.0 mm; forew- ing length 4.4 to 4.8 mm. Face about 1 .17 times longer than broad. First flagellar segment slightly longer than broad. Mesepisternal punctures weak, not larger than those of scutum, surface dull. Longest hairs of scutum about half as long as scape (largely worn off of some specimens); hairs of lower genal area shorter than scape, sparse. Metasoma with punctures smaller than in male and widely spaced, broad impunctate areas especially on Tl; posterior marginal areas of terga lineolate but shin- ing, with \'ery weak fasciae of pale hairs at extreme sides.
Holotype male: VENEZUELA: Merida: 42.4 km north- west of Merida, near La Carbonera, 8" 37' 38" N, 74" 21 ' 10" W, 2360 m, 25 May 1998, on flowers of Rubus (J. Ashe, R. Brooks, R. Hanley, Lawrence). Four male and three female paratypes: Merida, 13 Sept. 1978 (B. Villegas, R. M. Bohart, Logan except one male, Lawrence); two male paratypes.
same locality, 17 Nov. 1972 (G. E. Bohart, Logan)
Additional material: VENEZUELA: Merida: one fe- male, Libertador, 3 July 1979 (R. W. Brooks, A. A. Grigarick, J. McLaughlin, R. O. Schuster, Davis); the locality supports the tentative identification of this specimen, which could also be C. gibbosa so far as size and morphology are con- cerned. The short first flagellar segment suggests that the following specimen may also be C. transversaria: COLOM- BIA: Norte de Santander: one female, Oroque, 10 June 1965 (J. and B. Bechyne, Mnracay).
Etymology: transversaria, Latin, transverse, with ref- erence to the tubercles of 52 of the male.
Chilicola (Oroediscelis) gibbosa new species (Figures 21j, 32)
This species is very much like Chilicola transversaria; the characters of S2 of the male are the only ones that clearly separate this species from C. transversaria. It agrees with the description of that species except as follows:
Male: Length 6.5 mm; forewing length 4.6 mm. Ground between pimctures on scutimi shiny but minutely lineolate, on scutellum smooth. Hair of lower genal area longer than scape. Hind basitarsus with projection on outer side (X of Fig. 32a) only slightly de\'eloped in holotype but in speci- mens listed below from Ecuador it is distinct as in Figure 32a. S2 with bosses modified to form short, sharp, longi- tudinal, parallel ridges, appearing as gentlv rounded pro- jections in lateral view; surface between and behind these ridges smooth, without punctures or hairs. Hidden sterna and genitalia as in C. simplex, apical hairs of S8 sometimes more numerous and longer, as in C. brzoskai.
Female: Forewing length 4.8 mm. Face about 1.13 rimes longer than broad. First flagellar segment about 1.5 times as long as broad. Lower genal hairs as long as scape.
Holotype male and one female paratvpe: COLOMBIA: Caldas: Pensilvania, 8000 feet (2427 m), 26 June 1978 (J. H. Cane, Lawrence). One female from the type locality, 24 June 1978 (J. H. Cane, Lnzvrence) is larger than the female paratype but probably the same species.
Additional material: COLOMBIA: Antioquia: two males and two females. La Selva, Rionegro, June, Julv, and Sept. 1975, females on Bidens pilosa and unidentified Asteraceae, males in dead stems (R. Aiiez, Medellin). One of the two males has the hind basitarsal ventral projection reduced as in Figure 32c (compare with Fig. 32b); other characters are the same. ECUADOR: Tungurahua: one male, Banos, 1820 m, 11 Nov. 1955 (E. I. Schlinger, E. S. Ross, San Francisco); one male, Banos, 2300 m, 4 Feb. 1984 (M. Coo- per, Lyme Regis). Manabi: one male, Cojimies, 11 May 1949 (W. Clarke-Macintvre, Washington). Manabi province is on the coast, and this last specimen is doubtfully Andean. It is larger than the holotype, like the large female from the type localitv, and the hind basitarsal projection as well as
Bee Genus Chilicola in the Tropical Andes
43
the projections on S2 are stronger than those of the holo- type. Lojiv. Loja, La Argelia, 4" 2.023' S, 79" 11.765' W, 20 Feb. 2001 (C. Rasmussen, Aarhus) likewise has the basitarsal and S2 projections stronger than in the holotype.
The following females from different areas are likely to be this species; the first flagellar segment seems slightly longer than in Cliilicola transversaria. COLOMBIA: Stvitandcr: one female, Gambita La Raima El Calvario, Alto de la Vieja; 2200 m, 1 Apr. 1994 (E. E. Pelacio, Bogota Humboldt). Risaralda: one female, Pereira, Parque Regional Natural Ucumari, 1800-2000 m. Mar. 1991 (L. Schneider, Bogota Humboldt); one female, Estacion La Suiza, Municipio La Florida, Parque Regional Natural Ucumari, 2000 m, Dec. 1991 (L. Sclineider, Bogota Uiuv.). PERU: Cuzco: one female, Cuzco, 20 Dec. 1967 (G. E. Bohart, Logan); in absense of associated males, this could well be another species. VEN- EZUELA: Apmrei: one male, 6 km west of Mantecal, 17 Mar. 1982 (G. F. and J. F. Hevel, Wasltiugton); this male differs from others in the strong transverse ridge between the projections of S2, as in C. maculipcs. Possibly it represents another species.
Etymology: gibbofa, Latin, protuberant, with reference to the strong protuberance on the lower margin of the hind basitarsus of the male.
Chilicola {Orocdiscelis) brzoskai new species (Figures 21k, 33)
This is one of the smallest species of the subgenus. The projections on S4 of the male are tapering tubercles, no more prominent than the longitudinal, keel-like projections on S2. Chilicola gibbosa and C. simplex are the only other species with these characters of S2; C. gibbosa differs from C. brzoskai in the large median projection on the hind basitarsus of the male, whereas C. simplex ciiffers from C. brzoskai in the almost simple hind basitarsus. C. brzoskai differs from C. espeleticola as follows:
Male: Length 5.0 to 5.5 mm; forewing length 3.6 to 4.0 mm. Face about 1.08 times longer than broad, entirely black. Head surface below antemiae with well separated strong punch-ires, largest, closest, and interspaces shining and smooth on lower two-fifths of clypeus; clypeus with longitudinal median depressed line; malar space half as long as broad; lower ocular tangent at lower quarter of clypeus. First flagellar segment about three-fourths as long as second on lower surface, about as long as second on upper surface, seconci segment little over 1.5 times as long as broad; last three segments each somewhat longer than broad. Scutum and scutellum with strong punctures mostly separated by half a puncture width of shining although minutely sculptured ground; mesepisternum similar but lower part with punctures larger and more widely sepa- rated; side of propodeum, in contrast, dull and minuteh', densely strigose. Dorsal surface of propodeum almost as
long as scutellum. Longest hairs of scutum about two- thirds as long as scape, those of scutellum and sides of thorax about as long as scape, those of lower genal area longer than scape but not especially dense. Stigma about two-thirds as long as marginal cell on wing margin, distal stigmal perpendicular entering base of second submarginal cell. Middle femur with ventral concavity occupying basal two-fifths of femur. Hind tibia about three times longer than broad, imier crest ending in broad, flat, rounded api- cal tooth as in Chilicola gibbosa; hind basitarsus with con- vexity on lower margin near middle (Fig. 33b), convexity on outer surface weak. Metasoma shining, punctures strong, separated by two or more puncture widths of smooth or feebly lineolate ground dorsally, laterally sur- face dulled by lineolation; posterior marginal zones of terga smooth, lateral hair fasciae absent; S2 without bare me- dian area and long lateral hairs, boss elevated to form strong longitudinal ridge, rounded in lateral view; projec- tions of S4 strongly tapering but rounded tubercles (Fig. 21k), emargination between them about four times as wide as deep, area between them bare; S4 and S5 without abun- dant long lateral hairs, each with apical fringe of long, well spaced, suberect hairs, that of S4 less conspicuous and re- duced medially; S6 with most of surface nearly flat, shin- ing, minutely lineolate, but laterally with elevation bear- ing tuft of dense, slightly coppery hair. Genitalia and hid- den sterna as in C. simplex; penis valves (Fig. 33c) differ from those shown for C. simplex primarily or entirely be- cause they are undamaged.
Female: Agrees with description of male except for usual sexual features and the following: Length 5 mm; forewing length 4.0 mm. Face about 1.2 times longer than broad. First flagellar segment about 1.5 times as long as broad. Punctures of scutellum and median area of scutum separated by one or two puncture widths. Longest hairs of scutum half as long as scape, those of lower genal area about as long as scape. Posterior marginal zones of terga minutely lineolate but shining, laterally covered by weak fasciae of pale hairs.
Holotype male, one male and one female paratype: COLOMBIA: Valle: 10 km west of Call, 1400 m, 19 Sept. 1976 (Bell, Breed and Michener, Lawrence). Two male paratypes: Summit west of Call, 2000 m, 17 Sept. 1976 (Bell, Breed and Michener, Lawrence); one male paratype (head- less), same locality, 20 Oct. 1976 (M. Breed, Lawrence). Three male paratypes: Pichinde, 1700 m, 28 Aug. 1967 (P & B. Wygodzinsky, New York).
Additional material: One female perhaps of this spe- cies is from ECUADOR: [Chimborazo]: Moya, east of Alausi, 2500 m, Nov 1970 (L. E. Peha, Sao Paulo). The localit)' sug- gests that this might be Chilicola simplex but the clypeus is not as shiny as in the specimens of that species. One male, lacking the metasoma, may represent a new related spe-
44
Scientific Papers, Natural History Museum, The Universh y ot Kansas
Figs. 33-34. 33: Chilicola (Oreodiscclis) brzoskni, iii.ik'. <i, b, l\xstori»r basitarsus in ventral and oiitfr \ie\vs. The hind tibia is similar to that of C. ^ibbofci. c, Dorsal and ventral views of penis valves; genitalia and hidden sterna as in C. simplex. 34: C. (O.) simplex, male, a, b, Dorsoventral and lateral views of genitalia; c, S8; d, S7; e, f. Posterior tibia and basitarsus, ventral and outer views. For further explanation, see account of the subgenus.
cies; it is a little l.irge for C. brzoskni anci the ape.x of the inner crest of the hind tibia is pointed rather than rounded. It is from PERU: [Hiudiuco]: east side of Carpish Moun- tains, 74 km southwest of Tingo Maria, 2800 m, 17 Oct. 1954 (E. I. Schlinger, F. S. Ross, San Francisco).
Three other males from Peru (one without the metasoma) differ from the above description in the fol- lowing parhculars: Thorax less shining, scutum and scutel- lum with slightly smaller punctures, mostly separated by less than a puncture width of ground distinctly dulled by microsculpturing. Inner crest of hind tibia ending in flat, rounded projection that is narrower than that of Colom- bian Cliilicola brzoskni. Bosses or protuberances of S2 bluntly right angular in lateral view, height over half of basal wicith
(in C\)lombian specimens, they are rounded, height less than half of basal width). Ventral apicolateral lobe of S7 with basal piece slightly broader than in Colombian C. brzoskni. Both the pimctation and the inner hind tibial crest vary among Colombian specimens, although not reach- ing the states described above. 1 tentatively consider these specimens to be variants of C. brzoskai . Tlie localitv data are as follows: PERU: [jiiniii]: two males, Huacapistana, Rio I'arma, 1-2 hine 1920 (Cornell Uniw Exped., Itlinca). I'asco: t)ne male, six females, Yanachaga-Chemillen Na- tional Park, 15,=S0 m, 10" 16' 24" S, 75'36' 48" VV, 20 Oct. 1999. attracti'd to wet sand (R. Brooks, Lawrence, specimen num- bers (male) 145611, (females) 146603, 145683, 145727, 145751, 145835, and 145829.
Bee Genius Chiucoia in the Tkoi'icm. Andes
45
Etymology: This species is named for Dr. Da\'id VV. Brzoska, enthusiastic bee collector and tiger beetle special- ist who accompanied R. W. Brooks on the field trip on which the series from Pasco was collected.
Chllicoln (OroL'disci'lis) simplex new species (Figures 21-1, 34)
This species resembles Chilicola brzoskai in the small, tuberculate (tapering to rounded apices) projections of S4 of the male, but differs from that species as well as from other species of Oroedisci'lis in the almost unmodified hind basitarsi. The shiny clypeus of the two known specimens (both sexes) also distinguishes this species from its close relatives. C. simplex differs from C. cspcleticoln as follows:
Male: Length 5 mm (estimated from badly broken specimen); forewing length 3.5 mm. Face about 1.12 times longer than broad, entirely black. Head surface below an- tennae, especially clypeus, shining although upper half of clypeus faintly lineolate; head otherwise as described for Chilicola brzoskai except first flagellar segment perhaps shorter, about two-thirds as long as second on lower sur- face. Scutum with punctures mostly closer, separated by less than a puncture width; lower mesepisternum shining with smooth interspaces. Dorsal surface of propodeum as long as scutellum, shining. Longest hairs of scutum about half as long as scape, those of scutellum two-thirds as long as scape, those of lower side of thorax as long as scape, and those of lower genal area longer than scape but sparse. Wing as described for C. brzoskai. Hind tibia about three times longer than broad, distal end of inner crest right an- gular with slight acute tip; hind basitarsus slightly longer than remaining tarsal segments together, almost simple but basal half slightly thickened near middle of outer side (Fig. 34e). Metasomal terga as described for C. brzoskai: S2 with two preapical tubercles ("bosses") slightly longitudinally elongate, so reduced as to be invisible in lateral view, me- dian area of sternum bare; subsequent sterna as in C. brzoskai except projections of S4 even more reduced, emar- gination between them five or six times as wide as deep, a few hairs on space between them, and fringes of both S4 and S5 reduced medially. Hidden sterna and genitalia as illustrated in Figure 34a-d.
Female: Agrees with description of male as modified from that of Chilicola espelcticola except for usual sexual features and the following: Forewing length 3.9 mm. Face about 1.2 times longer than broad. First flagellar segment 1.5 times longer than broad. Lower mesepistemal area dull, punctures weak and widely separated. Posterior marginal zones of terga minutely lineolate but shining, laterally cov- ered by weak fasciae of pale hairs.
Holotype male: ECUADOR: ICIiimborazo]: Alausi, 9450 feet [2907 m], 18 June 1914 (H. S. Parish, Lrnvrcnce). Paratype female, same data (Ithaca).
Etymology: simplex, Latin, simple, with reference to the almost unmodified hind basitarsi of the male.
Phylogeny of the Species of the Subc;enus Oroediscelis
The Andes are inhabited by only some of the species of Anoediscelis and Hi/laeosoma; phylogenetic analysis of Andean members of these subgenera therefore seems in- appropriate at this time. All species of Oroediscelis, how- ever, are Andean and an analysis seems worth while. The characters selected and their states arc listed in Table 4. The methods were the same as indicated for the phyloge- netic analysis of subgenera. Outgroups were ChUicola (Amwdiscclis) i)icrmis (Friese) and C. (Prosopoidcs) pnosopoides (Friese). The matrix of character states is shown in Table 5. Asterisks indicate variation among individuals of a spe- cies. Multiple state characters were treated as nonaddi- tive. Analysis resulted in one tree (Fig. 35) with a length of 42, consistency index of 69, and retention index of 76.
The groupings seem reasonable. I am doubtful that the Chilicola macidipes-higibbosa group deserves such an iso- lateci position as is indicated by the few characters used in the analysis. A problem is finding additional characters with discrete states.
Table 4. Characters and character states for analysis of the males of the species of Oroediscelis.
1. Clypeus with yellow area (0); without yellow (1).
2. Paraocular areas with yellow (0); without veilow ( 1 ).
3. Anterior tibia with anterior surface yellow (0); without yel- low (1).
4. hiner crest of hind tibia with yellow (0); without yellow (1).
5. First flagellar segment shorter than second, at least on ante- rior surface (0); as long as second (1).
6. Hind basitarsus with one lateral and one ventral gibbosity, almost simple in C. simplex (0); with one lateral and two ven- tral gibbosities (1); simple (2).
7. Brush of hind tibia well de\-eloped (0); almost absent (1).
8. Outer crest on under side of hind tibia weak or absent dis- tally, concavity of tibia short and little dexeloped (0); outer crest well developed and concavity well developed (1); crest absent (2).
9. Inner crest on under side of hind tibia regularly reduced to- ward base of tibia (0); inner crest elevated near middle of tibia or in C. simplex, scarcely recognizable (1); imier crest absent (2).
10. Distal angle of inner crest on under side of hind tibia obtuse, little developed (0); angle projecting, about right angular (1); angle projecting acute (2); angle developed as broad mesal flap (3); strongly projecting but rounded (4); angle absent (5).
1 1 . Hairs of apical fringe of S6 not attaining apex of translucent apical lamella of S6 (0); exceeding apex (1).
12. Processes of S4 long, emargination between them deeper than wide (0); processes of S4 short, emargination between them wider than deep (1 ); processes absent (2).
13. Malar space about half as long as broad, or in C. euzcoensis, two-thirds as long as broad (0); malar space nearly as long as broad (1); malar space linear, i.e. absent (2).
46
Scientific Papers, Natural History Museum, The University of Kansas
Table 4. Continued.
14. Penis valves each with two, subapical membranous append- ages (0); penis valves v^'ithout or with only one such append- age (1).
15. S7 with apex of ventral laterodistal lobe a slender process (0); lobe broad, partly membranous (1); lobe a hairy process (2).
16. S7 with broad basal part of ventral laterobasal lobe: relatively flat (0); part seemingly curled (1); part projecting as a broad, more or less erect lobe (2); part rather short and simple, hairy (3); part seemingly absent (4).
|
Table 5. Character states used in the analysis of species of the subgenus Oroediscelis. |
|||||||
|
Character |
|||||||
|
Taxon I 23456789 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
|
C. brzoskai |
1 |
1 |
|
C. nmculipes |
0 |
0 |
|
C. bigibhosa |
0 |
0 |
|
C. simplex |
1 |
1 |
|
C. cuzcoensis |
0 |
0 |
|
C. quitensis |
1 |
7 |
|
C. espeleticoln |
1 |
0 |
|
C. benoistiana |
1 |
0 |
|
C. brooksi |
1 |
0 |
|
C. styliventris |
0 |
0 |
|
C. gibbosa |
1 |
1 |
|
C. transversaria |
1 |
7 |
|
C. prosopokies |
0 |
1 |
|
C. inennis |
0 |
0 |
|
0 |
0 |
0 |
4 |
0 |
1 |
0 |
0 |
0 |
0 |
||
|
0 |
1 |
1 |
3 |
0 |
1 |
1 |
0 |
0 |
0 |
||
|
1 |
1 |
1 |
3 |
0 |
1 |
1 |
0 |
0 |
0 |
||
|
() |
0 |
0 |
2 |
0 |
1 |
0 |
0 |
0 |
|||
|
1 |
0 |
0 |
0 |
2 |
0 |
0 |
0 |
0 |
2 |
||
|
1 |
0 |
0 |
2 |
0 |
0 |
0 |
1 |
1 |
|||
|
1 |
0 |
0 |
3 |
0 |
1 |
1 |
0 |
2 |
|||
|
0 |
0 |
0 |
1 |
1 |
0 |
0 |
1 |
1 |
|||
|
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
1 |
1 |
||
|
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
1 |
|||
|
0 |
0 |
0 |
4 |
0 |
i |
0 |
0 |
0 |
0 |
||
|
0 |
0 |
0 |
4 |
0 |
1 |
0 |
0 |
0 |
0 |
||
|
1 |
2 |
1 |
2 |
2 |
5 |
1 |
2 |
1 |
1 |
2 |
3 |
|
0 |
2 |
1 |
2 |
2 |
5 |
1 |
2 |
2 |
0 |
2 |
4 |
2 li
-OO — C. (P. ) prosopoides
13
-• — C. (S.) inermis
4
nmculipes
— bigibbosa
gibbosa transversaria brzoskai simplex
cuzcoensis
styliventris
10
benoistiana brooksi
Fig. 35. Phylogenetic study of the species of the subgenus Oivediscelis. Chilicola iiwrnji^ (Friese) of the subgenus Aiwediscelis or Stenoediscelis and C. prosopoides (Duckc) of the subgenus Prosopoides are outgroups. All the other species belong to the subgenus Oroediscelis. The single resultant tree had a length of 42, CI 69, Rl 76. Explanation as for Figure 2.
Bee Genus Chiucola in the Tropical Andes
47
LITERATURE CITED
Ashmead W. H. 1898. Some new genera of bees. Psyche 8:282-
285. Ayala, R., T. L. Griswold, and D. Yanega. 1996. Apoidea (Hy-
menoptera), pp. 423-464, in ]. Llorente B., A. N. Garcia A.,
and E. Gonzalez S. (eds.), Biodiversidad,Ttaxonomia i/
Biogeografia dc Artropodos de Mexico. Mexico: Universidad
Nacional Autiinoina de Mexico. Benoist, R. 1942. Les hymenopteres qui liabitent les tiges de
ronce aux environs de Quito (Equateur). Annales de la
Societe Entomologique de France 111:75-90. Cockerell, T. D. A. 1918. Descriptions and records of bees LXXXl.
Annals and Magazine of Natural History (9)2:418-425. Cockerell, T. D. A. 1919. Bees in the collection of the United
States National Museum-3. Proceedings of the U. S. National
Museum 55:167-221. Cockerell, T. D. A. 1926. Descriptions and records of bees CVIII.
Annals and Magazine of Natural History (9)17:214-226. Crawford, J. C. 1906. Some Costa Rican bees. Transactions of
the American Entomological Society 32:157-163. Eickwort, G. C. 1967. Aspects of the biology of Chilicola ashmeadi
in Costa Rica. Journal of the Kansas Entomological Society
40:42-73. Engel, M.S. 1999. A new xeromelissine bee in Tertiary amber of
the Dominican Republic, Entomologia Scandinavia 30:453-
458. Friese, H. 1908. Die Apidae(Blumenwespen) von Argentina nach
den Reisenergebnissen der Herren A. C. Jensen-Haarup und
P. Jorgensen in den Jahren 1904-1907. Flora og Fauna
10:331—425 (reprint, pp. 1-94).
Friese, H. 1916. Zur Bienenfauna von Costa Rica. Stettiner Entomologische Zeitgung, 77:287-348.
Goloboff, PA. 1993. NoName (NONA), version 1.5.1 Program and documentation, Tucuman, Argentina: Fundacion e Instituto Miguel Lillo.
Michener, C. D. 1994. Mexican and Central American species of Clulicola. Folia Entomologica Mexicana no. 85(]992):77-93.
Michener, C. D. 1995. A classification of the bees of the subfam- ily Xeromelissinae. Journal of the Kansas Entomological Society 68:332-345.
Michener, C. D. and G. Poinar. 1997. The known bee fauna of the Dominican Amber. Journal of the Kansas Entomologi- cal Society 69(suppl. 1996):353-361.
Morrone, J. J. 2001. Biogeografia de America Latina i/ el Caribe, Zaragoza, Spain: M and T Manuales y Tesis, 148 pp.
Nixon, K.C. 1999. Winclada, version 0.9.99 tuc. 13, beta. Pro- gram and Documentation, Ithaca, NY: Cornell University.
Snelling, R. R. 1971. A new synonymy in the genus Chilicola. Bulletin of Southern California Academy of Science 70:52.
Toro, H. 1986. Lista preliminar de los apidos chilenos. Acta Entomologica Chilena 13:121-132.
Toro H. and A. Moldenke. 1979. Revision de los Xeromelissinae Chilenos. Anales del Museo de Historia Natural, Valparaiso 12:95-182.
Wilms, W. 1995. Die Beinenfauna im Kiistenregenioald Brasiliens und Hue Bezicliungeii zii Bliitenpflanzen: Fallstudie Boraceia, Sao Paulo, Tubingen: Dissertation Eberhard-Karls- Universitatat, [6J+219 pp.
«>*044 072 228
752
PUBLICATIONS OF THE NATURAL HISTORY MUSEUM, THE UNIVERSITY OF KANSAS
The Uni\'orsity of Kansas Publications, Museum of Natural History, beginning with Volume 1 in 1946, was discontinued with Volume 20 in 1971. Shorter research papers formerly published in the above series were published as The University of Kansas Natural History Museum Occasional Papers until Number 180 in Decem- ber 1996. The Miscellaneous Publications of The Unixersitv of Kansas Natural His- tory Museum began with Number 1 in 1946 and ended with Number 68 in Febru- ary 1996. Monographs of The University of Kansas Natural History Museum were initiated in 1970 and discontinued with Number 8 in 1992. The University of Kan- sas Science Bulletin, beginning with Volume 1 in 1902, was discontinued with Vol- ume 55 in 1996. The foregoing publication series are now combined in a new series entitled Scientific Papers, Natural History Museum, The University of Kansas, be- gun with Number 1 in 1997. Special Publications began in 1976 and continue as an outlet for longer contributions and are available by purchase only. All manuscripts are subject to critical review by intra- and extramural specialists; final acceptance is at the discretion of the editor.
The publication is printed on acid-free paper. Publications are composed us- ing Microsoft Word" and Adobe PageMaker' on a Macintosh computer and are printed by The University of Kansas Printing Services.
Institutional libraries interested in exchanging publications may obtain the Sci- entific Papers, Natural History Museum, The University of Kansas, by addressing the Exchange Librarian, Tlie University of Kansas Libraries, Lawrence, Kansas 66045- 2800, USA. Available back issues of The University of Kansas Science Bulletin may be purchased from the Library Sales Section, Retrieval Services Department, The University of Kansas Libraries, Lawrence, Kansas 66045-2800, USA. Available is- sues of former publication series. Scientific Papers, and Special Publications of the Natural History Museum can be purchased from the Office of Publications, Natu- ral History Museum, The University of Kansas, Lawrence, Kansas 66045-2454, USA. Purchasing information can be obtained by calling (785) 864-4450, fax (785) 864- 5335, or e-mail (kunhm@ukans.edu). VISA and MasterCard accepted; include ex- piration dati'.
Sfrii:s Editor: William E. Duellman
Printkd bv
Tiir UNivtitsm oi- Kansas Pkintinc. Sn<victs
L,\vvRF.\'CF., Kansas
§.
'I C» r ■*
irk 9
it!
i ♦ i
'■ • i •
■^ ii i
Wi
I iTfTi
• r *
rt ij
i t i
m iTa
WW'i
» I • «I*JJ.
mN
KKMMt
lfe]«.
m
titit.
mi
€
•
iCociL
»*I*«C«*C»
c*
ID'
i"^
*
•j
■ A « -E Ife ' # ini « i » €^
■ m ii 15 in* B • It 1^
t)Vi 1 ct 1 1 *■ i^^^^^
^
:^*3"
Pi 1 • • JTl Tl I '11 ■ li H # ii n
o
Ojl.
O.
i • 11
I % 1 #^ .'ii^ M^ a 1
lA
^
d
\\y
\M r
1 ■■ ',
! I
■i^«K^ ■* M'T ■pr •s^w ~ jji .
wm
.iK
^osiai
iW'jk^M^W/Wjgrff^^WJL^it^SiK'^4